From ddf75a86aba2cfb7ec4497e8692b60c8c8fe0ee7 Mon Sep 17 00:00:00 2001 From: Jianglei Nie Date: Tue, 29 Nov 2022 16:15:42 +0800 Subject: [PATCH 001/765] auxdisplay: hd44780: Fix potential memory leak in hd44780_remove() hd44780_probe() allocates a memory chunk for hd with kzalloc() and makes "lcd->drvdata->hd44780" point to it. When we call hd44780_remove(), we should release all relevant memory and resource. But "lcd->drvdata ->hd44780" is not released, which will lead to a memory leak. We should release the "lcd->drvdata->hd44780" in hd44780_remove() to fix the memory leak bug. Fixes: 718e05ed92ec ("auxdisplay: Introduce hd44780_common.[ch]") Reviewed-by: Andy Shevchenko Reported-by: kernel test robot Signed-off-by: Jianglei Nie Signed-off-by: Miguel Ojeda --- drivers/auxdisplay/hd44780.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/auxdisplay/hd44780.c b/drivers/auxdisplay/hd44780.c index 8b2a0eb3f32a4..d56a5d508ccd7 100644 --- a/drivers/auxdisplay/hd44780.c +++ b/drivers/auxdisplay/hd44780.c @@ -322,8 +322,10 @@ static int hd44780_probe(struct platform_device *pdev) static int hd44780_remove(struct platform_device *pdev) { struct charlcd *lcd = platform_get_drvdata(pdev); + struct hd44780_common *hdc = lcd->drvdata; charlcd_unregister(lcd); + kfree(hdc->hd44780); kfree(lcd->drvdata); kfree(lcd); From bab550371cb13fe7adc4105a57d0530565740fc6 Mon Sep 17 00:00:00 2001 From: Deepak R Varma Date: Mon, 26 Dec 2022 17:10:48 +0530 Subject: [PATCH 002/765] coccinelle: api/atomic_as_refcounter: include message type in output A common practice is to grep for "WARNING" or "ERROR" text in the report output from a Coccinelle semantic patch script. So, include the text "WARNING: " in the report output generated by the semantic patch for desired filtering of the output. Also improves the readability of the output. Here is an example of the old and new outputs reported: xyz_file.c:131:39-40: atomic_add_unless xyz_file.c:131:39-40: WARNING: atomic_add_unless xyz_file.c:196:6-25: atomic_dec_and_test variation before object free at line 208. xyz_file.c:196:6-25: WARNING: atomic_dec_and_test variation before object free at line 208. Signed-off-by: Deepak R Varma Acked-by: Julia Lawall --- scripts/coccinelle/api/atomic_as_refcounter.cocci | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/scripts/coccinelle/api/atomic_as_refcounter.cocci b/scripts/coccinelle/api/atomic_as_refcounter.cocci index e63d52408b86f..bbe5b29329335 100644 --- a/scripts/coccinelle/api/atomic_as_refcounter.cocci +++ b/scripts/coccinelle/api/atomic_as_refcounter.cocci @@ -55,7 +55,7 @@ identifier fname6 =~ ".*call_rcu.*"; p1 << r1.p1; p2 << r1.p2; @@ -msg = "atomic_dec_and_test variation before object free at line %s." +msg = "WARNING: atomic_dec_and_test variation before object free at line %s." coccilib.report.print_report(p1[0], msg % (p2[0].line)) @r4 exists@ @@ -88,7 +88,7 @@ fname@p2(y, ...); p1 << r4.p1; p2 << r4.p2; @@ -msg = "atomic_dec_and_test variation before object free at line %s." +msg = "WARNING: atomic_dec_and_test variation before object free at line %s." coccilib.report.print_report(p1[0], msg % (p2[0].line)) @r2 exists@ @@ -107,7 +107,7 @@ atomic64_add_unless(&(a)->x,-1,1)@p1 @script:python depends on report@ p1 << r2.p1; @@ -msg = "atomic_add_unless" +msg = "WARNING: atomic_add_unless" coccilib.report.print_report(p1[0], msg) @r3 exists@ @@ -126,5 +126,5 @@ x = atomic64_add_return@p1(-1, ...); @script:python depends on report@ p1 << r3.p1; @@ -msg = "x = atomic_add_return(-1, ...)" +msg = "WARNING: x = atomic_add_return(-1, ...)" coccilib.report.print_report(p1[0], msg) From fad376fce0af58deebc5075b8539dc05bf639af3 Mon Sep 17 00:00:00 2001 From: Liu Shixin via Jfs-discussion Date: Thu, 3 Nov 2022 11:01:59 +0800 Subject: [PATCH 003/765] fs/jfs: fix shift exponent db_agl2size negative As a shift exponent, db_agl2size can not be less than 0. Add the missing check to fix the shift-out-of-bounds bug reported by syzkaller: UBSAN: shift-out-of-bounds in fs/jfs/jfs_dmap.c:2227:15 shift exponent -744642816 is negative Reported-by: syzbot+0be96567042453c0c820@syzkaller.appspotmail.com Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Liu Shixin Signed-off-by: Dave Kleikamp --- fs/jfs/jfs_dmap.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/jfs/jfs_dmap.c b/fs/jfs/jfs_dmap.c index 765838578a722..a3eb1e8269477 100644 --- a/fs/jfs/jfs_dmap.c +++ b/fs/jfs/jfs_dmap.c @@ -193,7 +193,8 @@ int dbMount(struct inode *ipbmap) bmp->db_agwidth = le32_to_cpu(dbmp_le->dn_agwidth); bmp->db_agstart = le32_to_cpu(dbmp_le->dn_agstart); bmp->db_agl2size = le32_to_cpu(dbmp_le->dn_agl2size); - if (bmp->db_agl2size > L2MAXL2SIZE - L2MAXAG) { + if (bmp->db_agl2size > L2MAXL2SIZE - L2MAXAG || + bmp->db_agl2size < 0) { err = -EINVAL; goto err_release_metapage; } From 39bee2e6acc2522d88feb324b18178b34cb7b75a Mon Sep 17 00:00:00 2001 From: Sergey Shtylyov Date: Tue, 20 Dec 2022 20:17:32 +0300 Subject: [PATCH 004/765] f2fs: file: drop useless initializer in expand_inode_data() In expand_inode_data(), the 'new_size' local variable is initialized to the result of i_size_read(), however this value isn't ever used, so we can drop this initializer... Found by Linux Verification Center (linuxtesting.org) with the SVACE static analysis tool. Signed-off-by: Sergey Shtylyov Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index ecbc8c135b494..6426b50b70b85 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -1697,7 +1697,7 @@ static int expand_inode_data(struct inode *inode, loff_t offset, .err_gc_skipped = true, .nr_free_secs = 0 }; pgoff_t pg_start, pg_end; - loff_t new_size = i_size_read(inode); + loff_t new_size; loff_t off_end; block_t expanded = 0; int err; From f35474ec0059c318f9d1aff1d492a5494beb6293 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Wed, 21 Dec 2022 20:14:45 +0800 Subject: [PATCH 005/765] f2fs: fix to support .migrate_folio for compressed inode Add missed .migrate_folio for compressed inode, in order to support migration of compressed inode's page. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/compress.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c index 2532f369cb10f..719b0a0184b07 100644 --- a/fs/f2fs/compress.c +++ b/fs/f2fs/compress.c @@ -1813,6 +1813,7 @@ unsigned int f2fs_cluster_blocks_are_contiguous(struct dnode_of_data *dn) const struct address_space_operations f2fs_compress_aops = { .release_folio = f2fs_release_folio, .invalidate_folio = f2fs_invalidate_folio, + .migrate_folio = filemap_migrate_folio, }; struct address_space *COMPRESS_MAPPING(struct f2fs_sb_info *sbi) From b3107b3854c93ea380ac373c0032fcf15f31178a Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Tue, 20 Dec 2022 19:56:02 +0800 Subject: [PATCH 006/765] f2fs: remove unused PAGE_PRIVATE_ATOMIC_WRITE Commit 3db1de0e582c ("f2fs: change the current atomic write way") has removed all users of PAGE_PRIVATE_ATOMIC_WRITE, remove its definition and related functions. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/f2fs.h | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-) diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index e8953c3dc81ab..9c7df2f1df73d 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -1396,19 +1396,17 @@ static inline void f2fs_clear_bit(unsigned int nr, char *addr); * Layout A: lowest bit should be 1 * | bit0 = 1 | bit1 | bit2 | ... | bit MAX | private data .... | * bit 0 PAGE_PRIVATE_NOT_POINTER - * bit 1 PAGE_PRIVATE_ATOMIC_WRITE - * bit 2 PAGE_PRIVATE_DUMMY_WRITE - * bit 3 PAGE_PRIVATE_ONGOING_MIGRATION - * bit 4 PAGE_PRIVATE_INLINE_INODE - * bit 5 PAGE_PRIVATE_REF_RESOURCE - * bit 6- f2fs private data + * bit 1 PAGE_PRIVATE_DUMMY_WRITE + * bit 2 PAGE_PRIVATE_ONGOING_MIGRATION + * bit 3 PAGE_PRIVATE_INLINE_INODE + * bit 4 PAGE_PRIVATE_REF_RESOURCE + * bit 5- f2fs private data * * Layout B: lowest bit should be 0 * page.private is a wrapped pointer. */ enum { PAGE_PRIVATE_NOT_POINTER, /* private contains non-pointer data */ - PAGE_PRIVATE_ATOMIC_WRITE, /* data page from atomic write path */ PAGE_PRIVATE_DUMMY_WRITE, /* data page for padding aligned IO */ PAGE_PRIVATE_ONGOING_MIGRATION, /* data page which is on-going migrating */ PAGE_PRIVATE_INLINE_INODE, /* inode page contains inline data */ @@ -1453,19 +1451,16 @@ PAGE_PRIVATE_GET_FUNC(nonpointer, NOT_POINTER); PAGE_PRIVATE_GET_FUNC(reference, REF_RESOURCE); PAGE_PRIVATE_GET_FUNC(inline, INLINE_INODE); PAGE_PRIVATE_GET_FUNC(gcing, ONGOING_MIGRATION); -PAGE_PRIVATE_GET_FUNC(atomic, ATOMIC_WRITE); PAGE_PRIVATE_GET_FUNC(dummy, DUMMY_WRITE); PAGE_PRIVATE_SET_FUNC(reference, REF_RESOURCE); PAGE_PRIVATE_SET_FUNC(inline, INLINE_INODE); PAGE_PRIVATE_SET_FUNC(gcing, ONGOING_MIGRATION); -PAGE_PRIVATE_SET_FUNC(atomic, ATOMIC_WRITE); PAGE_PRIVATE_SET_FUNC(dummy, DUMMY_WRITE); PAGE_PRIVATE_CLEAR_FUNC(reference, REF_RESOURCE); PAGE_PRIVATE_CLEAR_FUNC(inline, INLINE_INODE); PAGE_PRIVATE_CLEAR_FUNC(gcing, ONGOING_MIGRATION); -PAGE_PRIVATE_CLEAR_FUNC(atomic, ATOMIC_WRITE); PAGE_PRIVATE_CLEAR_FUNC(dummy, DUMMY_WRITE); static inline unsigned long get_page_private_data(struct page *page) From 6779b5db90c5b925293f7ccc5ed5336c5b24ed50 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Wed, 21 Dec 2022 20:13:45 +0800 Subject: [PATCH 007/765] f2fs: fix to call clear_page_private_reference in .{release,invalid}_folio b763f3bedc2d ("f2fs: restructure f2fs page.private layout") missed to call clear_page_private_reference() in .{release,invalid}_folio, fix it, though it's not a big deal since folio_detach_private() was called to clear all privae info and reference count in the page. BTW, remove page_private_reference() definition as it never be used. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 2 ++ fs/f2fs/f2fs.h | 1 - 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 97e816590cd95..017c9746a705b 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -3698,6 +3698,7 @@ void f2fs_invalidate_folio(struct folio *folio, size_t offset, size_t length) } } + clear_page_private_reference(&folio->page); clear_page_private_gcing(&folio->page); if (test_opt(sbi, COMPRESS_CACHE) && @@ -3723,6 +3724,7 @@ bool f2fs_release_folio(struct folio *folio, gfp_t wait) clear_page_private_data(&folio->page); } + clear_page_private_reference(&folio->page); clear_page_private_gcing(&folio->page); folio_detach_private(folio); diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 9c7df2f1df73d..5fdefb82509d4 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -1448,7 +1448,6 @@ static inline void clear_page_private_##name(struct page *page) \ } PAGE_PRIVATE_GET_FUNC(nonpointer, NOT_POINTER); -PAGE_PRIVATE_GET_FUNC(reference, REF_RESOURCE); PAGE_PRIVATE_GET_FUNC(inline, INLINE_INODE); PAGE_PRIVATE_GET_FUNC(gcing, ONGOING_MIGRATION); PAGE_PRIVATE_GET_FUNC(dummy, DUMMY_WRITE); From 8d3c1fa3fa5eacfd14f5b018eddb6c1a91c57783 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Nov 2022 10:15:09 +0100 Subject: [PATCH 008/765] f2fs: don't rely on F2FS_MAP_* in f2fs_iomap_begin When testing with a mixed zoned / convention device combination, there are regular but not 100% reproducible failures in xfstests generic/113 where the __is_valid_data_blkaddr assert hits due to finding a hole. This seems to be because f2fs_map_blocks can set this flag on a hole when it was found in the extent cache. Rework f2fs_iomap_begin to just check the special block numbers directly. This has the added benefits of the WARN_ON showing which invalid block address we found, and being properly error out on delalloc blocks that are confusingly called unwritten but not actually suitable for direct I/O. Fixes: 1517c1a7a445 ("f2fs: implement iomap operations") Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 24 ++++++++++++++---------- 1 file changed, 14 insertions(+), 10 deletions(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 017c9746a705b..86585c0922f2d 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -4153,20 +4153,24 @@ static int f2fs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, */ map.m_len = fscrypt_limit_io_blocks(inode, map.m_lblk, map.m_len); - if (map.m_flags & (F2FS_MAP_MAPPED | F2FS_MAP_UNWRITTEN)) { - iomap->length = blks_to_bytes(inode, map.m_len); - if (map.m_flags & F2FS_MAP_MAPPED) { - iomap->type = IOMAP_MAPPED; - iomap->flags |= IOMAP_F_MERGED; - } else { - iomap->type = IOMAP_UNWRITTEN; - } - if (WARN_ON_ONCE(!__is_valid_data_blkaddr(map.m_pblk))) - return -EINVAL; + /* + * We should never see delalloc or compressed extents here based on + * prior flushing and checks. + */ + if (WARN_ON_ONCE(map.m_pblk == NEW_ADDR)) + return -EINVAL; + if (WARN_ON_ONCE(map.m_pblk == COMPRESS_ADDR)) + return -EINVAL; + if (map.m_pblk != NULL_ADDR) { + iomap->length = blks_to_bytes(inode, map.m_len); + iomap->type = IOMAP_MAPPED; + iomap->flags |= IOMAP_F_MERGED; iomap->bdev = map.m_bdev; iomap->addr = blks_to_bytes(inode, map.m_pblk); } else { + if (flags & IOMAP_WRITE) + return -ENOTBLK; iomap->length = blks_to_bytes(inode, next_pgofs) - iomap->offset; iomap->type = IOMAP_HOLE; From 62a134bd8941198eee8b23584e35be6a9ab835d1 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Nov 2022 10:15:10 +0100 Subject: [PATCH 009/765] f2fs: decouple F2FS_MAP_ from buffer head flags m_flags is never interchanged with the buffer_heads b_flags directly, so use separate codepoints from that. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/f2fs.h | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 5fdefb82509d4..28828a8c959fd 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -692,13 +692,11 @@ struct extent_tree_info { }; /* - * This structure is taken from ext4_map_blocks. - * - * Note that, however, f2fs uses NEW and MAPPED flags for f2fs_map_blocks(). + * State of block returned by f2fs_map_blocks. */ -#define F2FS_MAP_NEW (1 << BH_New) -#define F2FS_MAP_MAPPED (1 << BH_Mapped) -#define F2FS_MAP_UNWRITTEN (1 << BH_Unwritten) +#define F2FS_MAP_NEW (1U << 0) +#define F2FS_MAP_MAPPED (1U << 1) +#define F2FS_MAP_UNWRITTEN (1U << 2) #define F2FS_MAP_FLAGS (F2FS_MAP_NEW | F2FS_MAP_MAPPED |\ F2FS_MAP_UNWRITTEN) From da8c7fecc9c7ba91b6d5ff5726189f269686a40c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Nov 2022 10:15:11 +0100 Subject: [PATCH 010/765] f2fs: rename F2FS_MAP_UNWRITTEN to F2FS_MAP_DELALLOC NEW_ADDR blocks are purely in-memory preallocated blocks, and thus equivalent to what the core FS code calls delayed allocations, and not unwritten extents which do have on-disk blocks allocated from which reads always return zeroes until they are converted to written status. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 8 ++++---- fs/f2fs/f2fs.h | 4 ++-- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 86585c0922f2d..93625d9900e8b 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -1660,9 +1660,9 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, bidx = f2fs_target_device_index(sbi, blkaddr); if (map->m_len == 0) { - /* preallocated unwritten block should be mapped for fiemap. */ + /* reserved delalloc block should be mapped for fiemap. */ if (blkaddr == NEW_ADDR) - map->m_flags |= F2FS_MAP_UNWRITTEN; + map->m_flags |= F2FS_MAP_DELALLOC; map->m_flags |= F2FS_MAP_MAPPED; map->m_pblk = blkaddr; @@ -1984,7 +1984,7 @@ int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, compr_appended = false; /* In a case of compressed cluster, append this to the last extent */ - if (compr_cluster && ((map.m_flags & F2FS_MAP_UNWRITTEN) || + if (compr_cluster && ((map.m_flags & F2FS_MAP_DELALLOC) || !(map.m_flags & F2FS_MAP_FLAGS))) { compr_appended = true; goto skip_fill; @@ -2030,7 +2030,7 @@ int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, compr_cluster = false; size += blks_to_bytes(inode, 1); } - } else if (map.m_flags & F2FS_MAP_UNWRITTEN) { + } else if (map.m_flags & F2FS_MAP_DELALLOC) { flags = FIEMAP_EXTENT_UNWRITTEN; } diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 28828a8c959fd..fea2735167e53 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -696,9 +696,9 @@ struct extent_tree_info { */ #define F2FS_MAP_NEW (1U << 0) #define F2FS_MAP_MAPPED (1U << 1) -#define F2FS_MAP_UNWRITTEN (1U << 2) +#define F2FS_MAP_DELALLOC (1U << 2) #define F2FS_MAP_FLAGS (F2FS_MAP_NEW | F2FS_MAP_MAPPED |\ - F2FS_MAP_UNWRITTEN) + F2FS_MAP_DELALLOC) struct f2fs_map_blocks { struct block_device *m_bdev; /* for multi-device dio */ From bc29835a9d4860df93a663d659e07dfdd8b4f629 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Nov 2022 10:15:12 +0100 Subject: [PATCH 011/765] f2fs: split __submit_bio Split __submit_bio into one function each for reads and writes, and a helper for aligning writes. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/compress.c | 2 +- fs/f2fs/data.c | 111 +++++++++++++++++++++++---------------------- fs/f2fs/f2fs.h | 4 +- 3 files changed, 61 insertions(+), 56 deletions(-) diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c index 719b0a0184b07..e4db3b7932a37 100644 --- a/fs/f2fs/compress.c +++ b/fs/f2fs/compress.c @@ -1070,7 +1070,7 @@ static int prepare_compress_overwrite(struct compress_ctx *cc, if (ret) goto out; if (bio) - f2fs_submit_bio(sbi, bio, DATA); + f2fs_submit_read_bio(sbi, bio, DATA); ret = f2fs_init_compress_ctx(cc); if (ret) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 93625d9900e8b..f5ef7d07191b6 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -507,65 +507,66 @@ static bool f2fs_crypt_mergeable_bio(struct bio *bio, const struct inode *inode, return fscrypt_mergeable_bio(bio, inode, next_idx); } -static inline void __submit_bio(struct f2fs_sb_info *sbi, - struct bio *bio, enum page_type type) +void f2fs_submit_read_bio(struct f2fs_sb_info *sbi, struct bio *bio, + enum page_type type) { - if (!is_read_io(bio_op(bio))) { - unsigned int start; + WARN_ON_ONCE(!is_read_io(bio_op(bio))); + trace_f2fs_submit_read_bio(sbi->sb, type, bio); - if (type != DATA && type != NODE) - goto submit_io; + iostat_update_submit_ctx(bio, type); + submit_bio(bio); +} - if (f2fs_lfs_mode(sbi) && current->plug) - blk_finish_plug(current->plug); +static void f2fs_align_write_bio(struct f2fs_sb_info *sbi, struct bio *bio) +{ + unsigned int start = + (bio->bi_iter.bi_size >> F2FS_BLKSIZE_BITS) % F2FS_IO_SIZE(sbi); - if (!F2FS_IO_ALIGNED(sbi)) - goto submit_io; + if (start == 0) + return; - start = bio->bi_iter.bi_size >> F2FS_BLKSIZE_BITS; - start %= F2FS_IO_SIZE(sbi); + /* fill dummy pages */ + for (; start < F2FS_IO_SIZE(sbi); start++) { + struct page *page = + mempool_alloc(sbi->write_io_dummy, + GFP_NOIO | __GFP_NOFAIL); + f2fs_bug_on(sbi, !page); - if (start == 0) - goto submit_io; + lock_page(page); - /* fill dummy pages */ - for (; start < F2FS_IO_SIZE(sbi); start++) { - struct page *page = - mempool_alloc(sbi->write_io_dummy, - GFP_NOIO | __GFP_NOFAIL); - f2fs_bug_on(sbi, !page); + zero_user_segment(page, 0, PAGE_SIZE); + set_page_private_dummy(page); - lock_page(page); + if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) + f2fs_bug_on(sbi, 1); + } +} - zero_user_segment(page, 0, PAGE_SIZE); - set_page_private_dummy(page); +static void f2fs_submit_write_bio(struct f2fs_sb_info *sbi, struct bio *bio, + enum page_type type) +{ + WARN_ON_ONCE(is_read_io(bio_op(bio))); + + if (type == DATA || type == NODE) { + if (f2fs_lfs_mode(sbi) && current->plug) + blk_finish_plug(current->plug); - if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) - f2fs_bug_on(sbi, 1); + if (F2FS_IO_ALIGNED(sbi)) { + f2fs_align_write_bio(sbi, bio); + /* + * In the NODE case, we lose next block address chain. + * So, we need to do checkpoint in f2fs_sync_file. + */ + if (type == NODE) + set_sbi_flag(sbi, SBI_NEED_CP); } - /* - * In the NODE case, we lose next block address chain. So, we - * need to do checkpoint in f2fs_sync_file. - */ - if (type == NODE) - set_sbi_flag(sbi, SBI_NEED_CP); } -submit_io: - if (is_read_io(bio_op(bio))) - trace_f2fs_submit_read_bio(sbi->sb, type, bio); - else - trace_f2fs_submit_write_bio(sbi->sb, type, bio); + trace_f2fs_submit_write_bio(sbi->sb, type, bio); iostat_update_submit_ctx(bio, type); submit_bio(bio); } -void f2fs_submit_bio(struct f2fs_sb_info *sbi, - struct bio *bio, enum page_type type) -{ - __submit_bio(sbi, bio, type); -} - static void __submit_merged_bio(struct f2fs_bio_info *io) { struct f2fs_io_info *fio = &io->fio; @@ -573,12 +574,13 @@ static void __submit_merged_bio(struct f2fs_bio_info *io) if (!io->bio) return; - if (is_read_io(fio->op)) + if (is_read_io(fio->op)) { trace_f2fs_prepare_read_bio(io->sbi->sb, fio->type, io->bio); - else + f2fs_submit_read_bio(io->sbi, io->bio, fio->type); + } else { trace_f2fs_prepare_write_bio(io->sbi->sb, fio->type, io->bio); - - __submit_bio(io->sbi, io->bio, fio->type); + f2fs_submit_write_bio(io->sbi, io->bio, fio->type); + } io->bio = NULL; } @@ -746,7 +748,10 @@ int f2fs_submit_page_bio(struct f2fs_io_info *fio) inc_page_count(fio->sbi, is_read_io(fio->op) ? __read_io_type(page) : WB_DATA_TYPE(fio->page)); - __submit_bio(fio->sbi, bio, fio->type); + if (is_read_io(bio_op(bio))) + f2fs_submit_read_bio(fio->sbi, bio, fio->type); + else + f2fs_submit_write_bio(fio->sbi, bio, fio->type); return 0; } @@ -848,7 +853,7 @@ static int add_ipu_page(struct f2fs_io_info *fio, struct bio **bio, /* page can't be merged into bio; submit the bio */ del_bio_entry(be); - __submit_bio(sbi, *bio, DATA); + f2fs_submit_write_bio(sbi, *bio, DATA); break; } f2fs_up_write(&io->bio_list_lock); @@ -911,7 +916,7 @@ void f2fs_submit_merged_ipu_write(struct f2fs_sb_info *sbi, } if (found) - __submit_bio(sbi, target, DATA); + f2fs_submit_write_bio(sbi, target, DATA); if (bio && *bio) { bio_put(*bio); *bio = NULL; @@ -1107,7 +1112,7 @@ static int f2fs_submit_page_read(struct inode *inode, struct page *page, } inc_page_count(sbi, F2FS_RD_DATA); f2fs_update_iostat(sbi, NULL, FS_DATA_READ_IO, F2FS_BLKSIZE); - __submit_bio(sbi, bio, DATA); + f2fs_submit_read_bio(sbi, bio, DATA); return 0; } @@ -2137,7 +2142,7 @@ static int f2fs_read_single_page(struct inode *inode, struct page *page, *last_block_in_bio, block_nr) || !f2fs_crypt_mergeable_bio(bio, inode, page->index, NULL))) { submit_and_realloc: - __submit_bio(F2FS_I_SB(inode), bio, DATA); + f2fs_submit_read_bio(F2FS_I_SB(inode), bio, DATA); bio = NULL; } if (bio == NULL) { @@ -2284,7 +2289,7 @@ int f2fs_read_multi_pages(struct compress_ctx *cc, struct bio **bio_ret, *last_block_in_bio, blkaddr) || !f2fs_crypt_mergeable_bio(bio, inode, page->index, NULL))) { submit_and_realloc: - __submit_bio(sbi, bio, DATA); + f2fs_submit_read_bio(sbi, bio, DATA); bio = NULL; } @@ -2445,7 +2450,7 @@ static int f2fs_mpage_readpages(struct inode *inode, #endif } if (bio) - __submit_bio(F2FS_I_SB(inode), bio, DATA); + f2fs_submit_read_bio(F2FS_I_SB(inode), bio, DATA); return ret; } diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index fea2735167e53..0e240ac13ca1c 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -3780,8 +3780,8 @@ int __init f2fs_init_bioset(void); void f2fs_destroy_bioset(void); int f2fs_init_bio_entry_cache(void); void f2fs_destroy_bio_entry_cache(void); -void f2fs_submit_bio(struct f2fs_sb_info *sbi, - struct bio *bio, enum page_type type); +void f2fs_submit_read_bio(struct f2fs_sb_info *sbi, struct bio *bio, + enum page_type type); int f2fs_init_write_merge_io(struct f2fs_sb_info *sbi); void f2fs_submit_merged_write(struct f2fs_sb_info *sbi, enum page_type type); void f2fs_submit_merged_write_cond(struct f2fs_sb_info *sbi, From 04a91ab016847440c8c937dda628079070f38c7a Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Nov 2022 10:15:14 +0100 Subject: [PATCH 012/765] f2fs: add a f2fs_lookup_extent_cache_block helper All but three callers of f2fs_lookup_extent_cache just want the block address. Add a small helper to simplify them. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 51 +++++++++++++++--------------------------- fs/f2fs/extent_cache.c | 11 +++++++++ fs/f2fs/f2fs.h | 2 ++ fs/f2fs/gc.c | 5 ++--- 4 files changed, 33 insertions(+), 36 deletions(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index f5ef7d07191b6..e80d7d250993b 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -1214,14 +1214,9 @@ int f2fs_reserve_block(struct dnode_of_data *dn, pgoff_t index) int f2fs_get_block(struct dnode_of_data *dn, pgoff_t index) { - struct extent_info ei = {0, }; - struct inode *inode = dn->inode; - - if (f2fs_lookup_read_extent_cache(inode, index, &ei)) { - dn->data_blkaddr = ei.blk + index - ei.fofs; + if (f2fs_lookup_read_extent_cache_block(dn->inode, index, + &dn->data_blkaddr)) return 0; - } - return f2fs_reserve_block(dn, index); } @@ -1232,15 +1227,14 @@ struct page *f2fs_get_read_data_page(struct inode *inode, pgoff_t index, struct address_space *mapping = inode->i_mapping; struct dnode_of_data dn; struct page *page; - struct extent_info ei = {0, }; int err; page = f2fs_grab_cache_page(mapping, index, for_write); if (!page) return ERR_PTR(-ENOMEM); - if (f2fs_lookup_read_extent_cache(inode, index, &ei)) { - dn.data_blkaddr = ei.blk + index - ei.fofs; + if (f2fs_lookup_read_extent_cache_block(inode, index, + &dn.data_blkaddr)) { if (!f2fs_is_valid_blkaddr(F2FS_I_SB(inode), dn.data_blkaddr, DATA_GENERIC_ENHANCE_READ)) { err = -EFSCORRUPTED; @@ -2641,7 +2635,6 @@ int f2fs_do_write_data_page(struct f2fs_io_info *fio) struct page *page = fio->page; struct inode *inode = page->mapping->host; struct dnode_of_data dn; - struct extent_info ei = {0, }; struct node_info ni; bool ipu_force = false; int err = 0; @@ -2653,9 +2646,8 @@ int f2fs_do_write_data_page(struct f2fs_io_info *fio) set_new_dnode(&dn, inode, NULL, NULL, 0); if (need_inplace_update(fio) && - f2fs_lookup_read_extent_cache(inode, page->index, &ei)) { - fio->old_blkaddr = ei.blk + page->index - ei.fofs; - + f2fs_lookup_read_extent_cache_block(inode, page->index, + &fio->old_blkaddr)) { if (!f2fs_is_valid_blkaddr(fio->sbi, fio->old_blkaddr, DATA_GENERIC_ENHANCE)) { f2fs_handle_error(fio->sbi, @@ -3328,7 +3320,6 @@ static int prepare_write_begin(struct f2fs_sb_info *sbi, struct dnode_of_data dn; struct page *ipage; bool locked = false; - struct extent_info ei = {0, }; int err = 0; int flag; @@ -3376,20 +3367,16 @@ static int prepare_write_begin(struct f2fs_sb_info *sbi, } } else if (locked) { err = f2fs_get_block(&dn, index); - } else { - if (f2fs_lookup_read_extent_cache(inode, index, &ei)) { - dn.data_blkaddr = ei.blk + index - ei.fofs; - } else { - /* hole case */ - err = f2fs_get_dnode_of_data(&dn, index, LOOKUP_NODE); - if (err || dn.data_blkaddr == NULL_ADDR) { - f2fs_put_dnode(&dn); - f2fs_do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, - true); - WARN_ON(flag != F2FS_GET_BLOCK_PRE_AIO); - locked = true; - goto restart; - } + } else if (!f2fs_lookup_read_extent_cache_block(inode, index, + &dn.data_blkaddr)) { + /* hole case */ + err = f2fs_get_dnode_of_data(&dn, index, LOOKUP_NODE); + if (err || dn.data_blkaddr == NULL_ADDR) { + f2fs_put_dnode(&dn); + f2fs_do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, true); + WARN_ON(flag != F2FS_GET_BLOCK_PRE_AIO); + locked = true; + goto restart; } } @@ -3409,7 +3396,6 @@ static int __find_data_block(struct inode *inode, pgoff_t index, { struct dnode_of_data dn; struct page *ipage; - struct extent_info ei = {0, }; int err = 0; ipage = f2fs_get_node_page(F2FS_I_SB(inode), inode->i_ino); @@ -3418,9 +3404,8 @@ static int __find_data_block(struct inode *inode, pgoff_t index, set_new_dnode(&dn, inode, ipage, ipage, 0); - if (f2fs_lookup_read_extent_cache(inode, index, &ei)) { - dn.data_blkaddr = ei.blk + index - ei.fofs; - } else { + if (!f2fs_lookup_read_extent_cache_block(inode, index, + &dn.data_blkaddr)) { /* hole case */ err = f2fs_get_dnode_of_data(&dn, index, LOOKUP_NODE); if (err) { diff --git a/fs/f2fs/extent_cache.c b/fs/f2fs/extent_cache.c index 342af24b2f8cf..1daf8c88c09bf 100644 --- a/fs/f2fs/extent_cache.c +++ b/fs/f2fs/extent_cache.c @@ -1047,6 +1047,17 @@ bool f2fs_lookup_read_extent_cache(struct inode *inode, pgoff_t pgofs, return __lookup_extent_tree(inode, pgofs, ei, EX_READ); } +bool f2fs_lookup_read_extent_cache_block(struct inode *inode, pgoff_t index, + block_t *blkaddr) +{ + struct extent_info ei = {}; + + if (!f2fs_lookup_read_extent_cache(inode, index, &ei)) + return false; + *blkaddr = ei.blk + index - ei.fofs; + return true; +} + void f2fs_update_read_extent_cache(struct dnode_of_data *dn) { return __update_extent_cache(dn, EX_READ); diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 0e240ac13ca1c..cc737f2a9e827 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -4182,6 +4182,8 @@ void f2fs_destroy_extent_cache(void); void f2fs_init_read_extent_tree(struct inode *inode, struct page *ipage); bool f2fs_lookup_read_extent_cache(struct inode *inode, pgoff_t pgofs, struct extent_info *ei); +bool f2fs_lookup_read_extent_cache_block(struct inode *inode, pgoff_t index, + block_t *blkaddr); void f2fs_update_read_extent_cache(struct dnode_of_data *dn); void f2fs_update_read_extent_cache_range(struct dnode_of_data *dn, pgoff_t fofs, block_t blkaddr, unsigned int len); diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index 6e2cae3d2e717..83e68ec7763de 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -1150,7 +1150,6 @@ static int ra_data_block(struct inode *inode, pgoff_t index) struct address_space *mapping = inode->i_mapping; struct dnode_of_data dn; struct page *page; - struct extent_info ei = {0, }; struct f2fs_io_info fio = { .sbi = sbi, .ino = inode->i_ino, @@ -1168,8 +1167,8 @@ static int ra_data_block(struct inode *inode, pgoff_t index) if (!page) return -ENOMEM; - if (f2fs_lookup_read_extent_cache(inode, index, &ei)) { - dn.data_blkaddr = ei.blk + index - ei.fofs; + if (f2fs_lookup_read_extent_cache_block(inode, index, + &dn.data_blkaddr)) { if (unlikely(!f2fs_is_valid_blkaddr(sbi, dn.data_blkaddr, DATA_GENERIC_ENHANCE_READ))) { err = -EFSCORRUPTED; From cf342d3beda000b4c60990755ca7800de5038785 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Nov 2022 10:15:15 +0100 Subject: [PATCH 013/765] f2fs: add a f2fs_get_block_locked helper This allows to keep the f2fs_do_map_lock based locking scheme private to data.c. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 16 ++++++++++++++-- fs/f2fs/f2fs.h | 3 +-- fs/f2fs/file.c | 4 +--- 3 files changed, 16 insertions(+), 7 deletions(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index e80d7d250993b..904dfd1f16d0d 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -1212,7 +1212,7 @@ int f2fs_reserve_block(struct dnode_of_data *dn, pgoff_t index) return err; } -int f2fs_get_block(struct dnode_of_data *dn, pgoff_t index) +static int f2fs_get_block(struct dnode_of_data *dn, pgoff_t index) { if (f2fs_lookup_read_extent_cache_block(dn->inode, index, &dn->data_blkaddr)) @@ -1451,7 +1451,7 @@ static int __allocate_data_block(struct dnode_of_data *dn, int seg_type) return 0; } -void f2fs_do_map_lock(struct f2fs_sb_info *sbi, int flag, bool lock) +static void f2fs_do_map_lock(struct f2fs_sb_info *sbi, int flag, bool lock) { if (flag == F2FS_GET_BLOCK_PRE_AIO) { if (lock) @@ -1466,6 +1466,18 @@ void f2fs_do_map_lock(struct f2fs_sb_info *sbi, int flag, bool lock) } } +int f2fs_get_block_locked(struct dnode_of_data *dn, pgoff_t index) +{ + struct f2fs_sb_info *sbi = F2FS_I_SB(dn->inode); + int err; + + f2fs_do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, true); + err = f2fs_get_block(dn, index); + f2fs_do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, false); + + return err; +} + /* * f2fs_map_blocks() tries to find or build mapping relationship which * maps continuous logical blocks to physical blocks, and return such diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index cc737f2a9e827..744cb442085f6 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -3800,7 +3800,7 @@ void f2fs_set_data_blkaddr(struct dnode_of_data *dn); void f2fs_update_data_blkaddr(struct dnode_of_data *dn, block_t blkaddr); int f2fs_reserve_new_blocks(struct dnode_of_data *dn, blkcnt_t count); int f2fs_reserve_new_block(struct dnode_of_data *dn); -int f2fs_get_block(struct dnode_of_data *dn, pgoff_t index); +int f2fs_get_block_locked(struct dnode_of_data *dn, pgoff_t index); int f2fs_reserve_block(struct dnode_of_data *dn, pgoff_t index); struct page *f2fs_get_read_data_page(struct inode *inode, pgoff_t index, blk_opf_t op_flags, bool for_write, pgoff_t *next_pgofs); @@ -3811,7 +3811,6 @@ struct page *f2fs_get_lock_data_page(struct inode *inode, pgoff_t index, struct page *f2fs_get_new_data_page(struct inode *inode, struct page *ipage, pgoff_t index, bool new_i_size); int f2fs_do_write_data_page(struct f2fs_io_info *fio); -void f2fs_do_map_lock(struct f2fs_sb_info *sbi, int flag, bool lock); int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, int create, int flag); int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 6426b50b70b85..705a7eb4df99b 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -113,10 +113,8 @@ static vm_fault_t f2fs_vm_page_mkwrite(struct vm_fault *vmf) if (need_alloc) { /* block allocation */ - f2fs_do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, true); set_new_dnode(&dn, inode, NULL, NULL, 0); - err = f2fs_get_block(&dn, page->index); - f2fs_do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, false); + err = f2fs_get_block_locked(&dn, page->index); } #ifdef CONFIG_F2FS_FS_COMPRESSION From 2f51ade9524c609fcc4b05f230ecda356cd10b85 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Nov 2022 10:15:16 +0100 Subject: [PATCH 014/765] f2fs: f2fs_do_map_lock Split f2fs_do_map_lock into a lock and unlock helper to make the code using it easier to read. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 45 +++++++++++++++++++++++---------------------- 1 file changed, 23 insertions(+), 22 deletions(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 904dfd1f16d0d..212f2797e379e 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -1451,19 +1451,20 @@ static int __allocate_data_block(struct dnode_of_data *dn, int seg_type) return 0; } -static void f2fs_do_map_lock(struct f2fs_sb_info *sbi, int flag, bool lock) +static void f2fs_map_lock(struct f2fs_sb_info *sbi, int flag) { - if (flag == F2FS_GET_BLOCK_PRE_AIO) { - if (lock) - f2fs_down_read(&sbi->node_change); - else - f2fs_up_read(&sbi->node_change); - } else { - if (lock) - f2fs_lock_op(sbi); - else - f2fs_unlock_op(sbi); - } + if (flag == F2FS_GET_BLOCK_PRE_AIO) + f2fs_down_read(&sbi->node_change); + else + f2fs_lock_op(sbi); +} + +static void f2fs_map_unlock(struct f2fs_sb_info *sbi, int flag) +{ + if (flag == F2FS_GET_BLOCK_PRE_AIO) + f2fs_up_read(&sbi->node_change); + else + f2fs_unlock_op(sbi); } int f2fs_get_block_locked(struct dnode_of_data *dn, pgoff_t index) @@ -1471,9 +1472,9 @@ int f2fs_get_block_locked(struct dnode_of_data *dn, pgoff_t index) struct f2fs_sb_info *sbi = F2FS_I_SB(dn->inode); int err; - f2fs_do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, true); + f2fs_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO); err = f2fs_get_block(dn, index); - f2fs_do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, false); + f2fs_map_unlock(sbi, F2FS_GET_BLOCK_PRE_AIO); return err; } @@ -1548,7 +1549,7 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, next_dnode: if (map->m_may_create) - f2fs_do_map_lock(sbi, flag, true); + f2fs_map_lock(sbi, flag); /* When reading holes, we need its node page */ set_new_dnode(&dn, inode, NULL, NULL, 0); @@ -1732,7 +1733,7 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, f2fs_put_dnode(&dn); if (map->m_may_create) { - f2fs_do_map_lock(sbi, flag, false); + f2fs_map_unlock(sbi, flag); f2fs_balance_fs(sbi, dn.node_changed); } goto next_dnode; @@ -1778,7 +1779,7 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, f2fs_put_dnode(&dn); unlock_out: if (map->m_may_create) { - f2fs_do_map_lock(sbi, flag, false); + f2fs_map_unlock(sbi, flag); f2fs_balance_fs(sbi, dn.node_changed); } out: @@ -3350,7 +3351,7 @@ static int prepare_write_begin(struct f2fs_sb_info *sbi, if (f2fs_has_inline_data(inode) || (pos & PAGE_MASK) >= i_size_read(inode)) { - f2fs_do_map_lock(sbi, flag, true); + f2fs_map_lock(sbi, flag); locked = true; } @@ -3385,7 +3386,7 @@ static int prepare_write_begin(struct f2fs_sb_info *sbi, err = f2fs_get_dnode_of_data(&dn, index, LOOKUP_NODE); if (err || dn.data_blkaddr == NULL_ADDR) { f2fs_put_dnode(&dn); - f2fs_do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, true); + f2fs_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO); WARN_ON(flag != F2FS_GET_BLOCK_PRE_AIO); locked = true; goto restart; @@ -3399,7 +3400,7 @@ static int prepare_write_begin(struct f2fs_sb_info *sbi, f2fs_put_dnode(&dn); unlock_out: if (locked) - f2fs_do_map_lock(sbi, flag, false); + f2fs_map_unlock(sbi, flag); return err; } @@ -3438,7 +3439,7 @@ static int __reserve_data_block(struct inode *inode, pgoff_t index, struct page *ipage; int err = 0; - f2fs_do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, true); + f2fs_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO); ipage = f2fs_get_node_page(sbi, inode->i_ino); if (IS_ERR(ipage)) { @@ -3454,7 +3455,7 @@ static int __reserve_data_block(struct inode *inode, pgoff_t index, f2fs_put_dnode(&dn); unlock_out: - f2fs_do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, false); + f2fs_map_unlock(sbi, F2FS_GET_BLOCK_PRE_AIO); return err; } From 44b0dfebbd829c64c242c0c6ee10f8a88ecfa8b3 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Nov 2022 10:15:17 +0100 Subject: [PATCH 015/765] f2fs: reflow prepare_write_begin Reflow prepare_write_begin so that it reads more straight forward, and so that there is one place that does an extent cache lookup instead of three, two of which are hidden in f2fs_get_block calls. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 61 ++++++++++++++++++++++++++------------------------ 1 file changed, 32 insertions(+), 29 deletions(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 212f2797e379e..efd3ebaa4fc82 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -3333,8 +3333,8 @@ static int prepare_write_begin(struct f2fs_sb_info *sbi, struct dnode_of_data dn; struct page *ipage; bool locked = false; + int flag = F2FS_GET_BLOCK_PRE_AIO; int err = 0; - int flag; /* * If a whole page is being written and we already preallocated all the @@ -3344,13 +3344,12 @@ static int prepare_write_begin(struct f2fs_sb_info *sbi, return 0; /* f2fs_lock_op avoids race between write CP and convert_inline_page */ - if (f2fs_has_inline_data(inode) && pos + len > MAX_INLINE_DATA(inode)) - flag = F2FS_GET_BLOCK_DEFAULT; - else - flag = F2FS_GET_BLOCK_PRE_AIO; - - if (f2fs_has_inline_data(inode) || - (pos & PAGE_MASK) >= i_size_read(inode)) { + if (f2fs_has_inline_data(inode)) { + if (pos + len > MAX_INLINE_DATA(inode)) + flag = F2FS_GET_BLOCK_DEFAULT; + f2fs_map_lock(sbi, flag); + locked = true; + } else if ((pos & PAGE_MASK) >= i_size_read(inode)) { f2fs_map_lock(sbi, flag); locked = true; } @@ -3371,32 +3370,36 @@ static int prepare_write_begin(struct f2fs_sb_info *sbi, set_inode_flag(inode, FI_DATA_EXIST); if (inode->i_nlink) set_page_private_inline(ipage); - } else { - err = f2fs_convert_inline_page(&dn, page); - if (err) - goto out; - if (dn.data_blkaddr == NULL_ADDR) - err = f2fs_get_block(&dn, index); + goto out; } - } else if (locked) { - err = f2fs_get_block(&dn, index); - } else if (!f2fs_lookup_read_extent_cache_block(inode, index, - &dn.data_blkaddr)) { + err = f2fs_convert_inline_page(&dn, page); + if (err || dn.data_blkaddr != NULL_ADDR) + goto out; + } + + if (!f2fs_lookup_read_extent_cache_block(inode, index, + &dn.data_blkaddr)) { + if (locked) { + err = f2fs_reserve_block(&dn, index); + goto out; + } + /* hole case */ err = f2fs_get_dnode_of_data(&dn, index, LOOKUP_NODE); - if (err || dn.data_blkaddr == NULL_ADDR) { - f2fs_put_dnode(&dn); - f2fs_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO); - WARN_ON(flag != F2FS_GET_BLOCK_PRE_AIO); - locked = true; - goto restart; - } + if (!err && dn.data_blkaddr != NULL_ADDR) + goto out; + f2fs_put_dnode(&dn); + f2fs_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO); + WARN_ON(flag != F2FS_GET_BLOCK_PRE_AIO); + locked = true; + goto restart; } - - /* convert_inline_page can make node_changed */ - *blk_addr = dn.data_blkaddr; - *node_changed = dn.node_changed; out: + if (!err) { + /* convert_inline_page can make node_changed */ + *blk_addr = dn.data_blkaddr; + *node_changed = dn.node_changed; + } f2fs_put_dnode(&dn); unlock_out: if (locked) From 3cf684f2f8e0229714fb6d051508b42d3320e78f Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Nov 2022 10:15:18 +0100 Subject: [PATCH 016/765] f2fs: simplify __allocate_data_block Just use a simple if block for the conditional call to inc_valid_block_count. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index efd3ebaa4fc82..697020cef8200 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -1431,13 +1431,12 @@ static int __allocate_data_block(struct dnode_of_data *dn, int seg_type) return err; dn->data_blkaddr = f2fs_data_blkaddr(dn); - if (dn->data_blkaddr != NULL_ADDR) - goto alloc; - - if (unlikely((err = inc_valid_block_count(sbi, dn->inode, &count)))) - return err; + if (dn->data_blkaddr == NULL_ADDR) { + err = inc_valid_block_count(sbi, dn->inode, &count); + if (unlikely(err)) + return err; + } -alloc: set_summary(&sum, dn->nid, dn->ofs_in_node, ni.version); old_blkaddr = dn->data_blkaddr; f2fs_allocate_data_block(sbi, NULL, old_blkaddr, &dn->data_blkaddr, From a9e292f2aeb737b263b1c14281be239405e1bb19 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Tue, 13 Dec 2022 23:08:37 +0800 Subject: [PATCH 017/765] docs: f2fs: fix html doc error There is a problem with the html converted from the previous commit 6047de5482c3 ("f2fs: add barrier mount option") code submission. Probably something like this: barrier If this option is set, cache_flush commands are allowed to be Let's fix it. Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- Documentation/filesystems/f2fs.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/filesystems/f2fs.rst b/Documentation/filesystems/f2fs.rst index 220f3e0d3f559..2055e72871fe6 100644 --- a/Documentation/filesystems/f2fs.rst +++ b/Documentation/filesystems/f2fs.rst @@ -158,7 +158,7 @@ nobarrier This option can be used if underlying storage guarantees If this option is set, no cache_flush commands are issued but f2fs still guarantees the write ordering of all the data writes. -barrier If this option is set, cache_flush commands are allowed to be +barrier If this option is set, cache_flush commands are allowed to be issued. fastboot This option is used when a system wants to reduce mount time as much as possible, even though normal performance From ffdeab71d5cf0bcd45467b3badb3a1ef8f19a565 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Nov 2022 10:15:19 +0100 Subject: [PATCH 018/765] f2fs: remove f2fs_get_block Fold f2fs_get_block into the two remaining callers to simplify the call chain a bit. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 18 +++++++----------- 1 file changed, 7 insertions(+), 11 deletions(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 697020cef8200..e6f94591717b8 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -1212,14 +1212,6 @@ int f2fs_reserve_block(struct dnode_of_data *dn, pgoff_t index) return err; } -static int f2fs_get_block(struct dnode_of_data *dn, pgoff_t index) -{ - if (f2fs_lookup_read_extent_cache_block(dn->inode, index, - &dn->data_blkaddr)) - return 0; - return f2fs_reserve_block(dn, index); -} - struct page *f2fs_get_read_data_page(struct inode *inode, pgoff_t index, blk_opf_t op_flags, bool for_write, pgoff_t *next_pgofs) @@ -1469,10 +1461,12 @@ static void f2fs_map_unlock(struct f2fs_sb_info *sbi, int flag) int f2fs_get_block_locked(struct dnode_of_data *dn, pgoff_t index) { struct f2fs_sb_info *sbi = F2FS_I_SB(dn->inode); - int err; + int err = 0; f2fs_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO); - err = f2fs_get_block(dn, index); + if (!f2fs_lookup_read_extent_cache_block(dn->inode, index, + &dn->data_blkaddr)) + err = f2fs_reserve_block(dn, index); f2fs_map_unlock(sbi, F2FS_GET_BLOCK_PRE_AIO); return err; @@ -3450,7 +3444,9 @@ static int __reserve_data_block(struct inode *inode, pgoff_t index, } set_new_dnode(&dn, inode, ipage, ipage, 0); - err = f2fs_get_block(&dn, index); + if (!f2fs_lookup_read_extent_cache_block(dn.inode, index, + &dn.data_blkaddr)) + err = f2fs_reserve_block(&dn, index); *blk_addr = dn.data_blkaddr; *node_changed = dn.node_changed; From cd8fc5226bef3a1fda13a0e61794a039ca46744a Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Nov 2022 10:15:20 +0100 Subject: [PATCH 019/765] f2fs: remove the create argument to f2fs_map_blocks The create argument is always identicaly to map->m_may_create, so use that consistently. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 63 ++++++++++++++++--------------------- fs/f2fs/f2fs.h | 3 +- fs/f2fs/file.c | 12 +++---- include/trace/events/f2fs.h | 11 +++---- 4 files changed, 38 insertions(+), 51 deletions(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index e6f94591717b8..1cc5c6cb31468 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -1477,8 +1477,7 @@ int f2fs_get_block_locked(struct dnode_of_data *dn, pgoff_t index) * maps continuous logical blocks to physical blocks, and return such * info via f2fs_map_blocks structure. */ -int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, - int create, int flag) +int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, int flag) { unsigned int maxblocks = map->m_len; struct dnode_of_data dn; @@ -1507,38 +1506,31 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, pgofs = (pgoff_t)map->m_lblk; end = pgofs + maxblocks; - if (!create && f2fs_lookup_read_extent_cache(inode, pgofs, &ei)) { - if (f2fs_lfs_mode(sbi) && flag == F2FS_GET_BLOCK_DIO && - map->m_may_create) - goto next_dnode; + if (map->m_may_create || + !f2fs_lookup_read_extent_cache(inode, pgofs, &ei)) + goto next_dnode; - map->m_pblk = ei.blk + pgofs - ei.fofs; - map->m_len = min((pgoff_t)maxblocks, ei.fofs + ei.len - pgofs); - map->m_flags = F2FS_MAP_MAPPED; - if (map->m_next_extent) - *map->m_next_extent = pgofs + map->m_len; + /* Found the map in read extent cache */ + map->m_pblk = ei.blk + pgofs - ei.fofs; + map->m_len = min((pgoff_t)maxblocks, ei.fofs + ei.len - pgofs); + map->m_flags = F2FS_MAP_MAPPED; + if (map->m_next_extent) + *map->m_next_extent = pgofs + map->m_len; - /* for hardware encryption, but to avoid potential issue in future */ - if (flag == F2FS_GET_BLOCK_DIO) - f2fs_wait_on_block_writeback_range(inode, + /* for hardware encryption, but to avoid potential issue in future */ + if (flag == F2FS_GET_BLOCK_DIO) + f2fs_wait_on_block_writeback_range(inode, map->m_pblk, map->m_len); - if (map->m_multidev_dio) { - block_t blk_addr = map->m_pblk; - - bidx = f2fs_target_device_index(sbi, map->m_pblk); + if (map->m_multidev_dio) { + bidx = f2fs_target_device_index(sbi, map->m_pblk); - map->m_bdev = FDEV(bidx).bdev; - map->m_pblk -= FDEV(bidx).start_blk; - map->m_len = min(map->m_len, + map->m_bdev = FDEV(bidx).bdev; + map->m_pblk -= FDEV(bidx).start_blk; + map->m_len = min(map->m_len, FDEV(bidx).end_blk + 1 - map->m_pblk); - - if (map->m_may_create) - f2fs_update_device_state(sbi, inode->i_ino, - blk_addr, map->m_len); - } - goto out; } + goto out; next_dnode: if (map->m_may_create) @@ -1602,7 +1594,7 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, set_inode_flag(inode, FI_APPEND_WRITE); } } else { - if (create) { + if (map->m_may_create) { if (unlikely(f2fs_cp_error(sbi))) { err = -EIO; goto sync_out; @@ -1776,7 +1768,7 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, f2fs_balance_fs(sbi, dn.node_changed); } out: - trace_f2fs_map_blocks(inode, map, create, flag, err); + trace_f2fs_map_blocks(inode, map, flag, err); return err; } @@ -1798,7 +1790,7 @@ bool f2fs_overwrite_io(struct inode *inode, loff_t pos, size_t len) while (map.m_lblk < last_lblk) { map.m_len = last_lblk - map.m_lblk; - err = f2fs_map_blocks(inode, &map, 0, F2FS_GET_BLOCK_DEFAULT); + err = f2fs_map_blocks(inode, &map, F2FS_GET_BLOCK_DEFAULT); if (err || map.m_len == 0) return false; map.m_lblk += map.m_len; @@ -1972,7 +1964,7 @@ int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, map.m_len = cluster_size - count_in_cluster; } - ret = f2fs_map_blocks(inode, &map, 0, F2FS_GET_BLOCK_FIEMAP); + ret = f2fs_map_blocks(inode, &map, F2FS_GET_BLOCK_FIEMAP); if (ret) goto out; @@ -2105,7 +2097,7 @@ static int f2fs_read_single_page(struct inode *inode, struct page *page, map->m_lblk = block_in_file; map->m_len = last_block - block_in_file; - ret = f2fs_map_blocks(inode, map, 0, F2FS_GET_BLOCK_DEFAULT); + ret = f2fs_map_blocks(inode, map, F2FS_GET_BLOCK_DEFAULT); if (ret) goto out; got_it: @@ -3807,7 +3799,7 @@ static sector_t f2fs_bmap(struct address_space *mapping, sector_t block) map.m_next_pgofs = NULL; map.m_seg_type = NO_CHECK_TYPE; - if (!f2fs_map_blocks(inode, &map, 0, F2FS_GET_BLOCK_BMAP)) + if (!f2fs_map_blocks(inode, &map, F2FS_GET_BLOCK_BMAP)) blknr = map.m_pblk; } out: @@ -3915,7 +3907,7 @@ static int check_swap_activate(struct swap_info_struct *sis, map.m_seg_type = NO_CHECK_TYPE; map.m_may_create = false; - ret = f2fs_map_blocks(inode, &map, 0, F2FS_GET_BLOCK_FIEMAP); + ret = f2fs_map_blocks(inode, &map, F2FS_GET_BLOCK_FIEMAP); if (ret) goto out; @@ -4140,8 +4132,7 @@ static int f2fs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, if (flags & IOMAP_WRITE) map.m_may_create = true; - err = f2fs_map_blocks(inode, &map, flags & IOMAP_WRITE, - F2FS_GET_BLOCK_DIO); + err = f2fs_map_blocks(inode, &map, F2FS_GET_BLOCK_DIO); if (err) return err; diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 744cb442085f6..60f421fe14eee 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -3811,8 +3811,7 @@ struct page *f2fs_get_lock_data_page(struct inode *inode, pgoff_t index, struct page *f2fs_get_new_data_page(struct inode *inode, struct page *ipage, pgoff_t index, bool new_i_size); int f2fs_do_write_data_page(struct f2fs_io_info *fio); -int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, - int create, int flag); +int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, int flag); int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, u64 start, u64 len); int f2fs_encrypt_one_page(struct f2fs_io_info *fio); diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 705a7eb4df99b..939c7247c3678 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -1743,7 +1743,7 @@ static int expand_inode_data(struct inode *inode, loff_t offset, f2fs_unlock_op(sbi); map.m_seg_type = CURSEG_COLD_DATA_PINNED; - err = f2fs_map_blocks(inode, &map, 1, F2FS_GET_BLOCK_PRE_DIO); + err = f2fs_map_blocks(inode, &map, F2FS_GET_BLOCK_PRE_DIO); file_dont_truncate(inode); f2fs_up_write(&sbi->pin_sem); @@ -1756,7 +1756,7 @@ static int expand_inode_data(struct inode *inode, loff_t offset, map.m_len = expanded; } else { - err = f2fs_map_blocks(inode, &map, 1, F2FS_GET_BLOCK_PRE_AIO); + err = f2fs_map_blocks(inode, &map, F2FS_GET_BLOCK_PRE_AIO); expanded = map.m_len; } out_err: @@ -2604,7 +2604,7 @@ static int f2fs_defragment_range(struct f2fs_sb_info *sbi, */ while (map.m_lblk < pg_end) { map.m_len = pg_end - map.m_lblk; - err = f2fs_map_blocks(inode, &map, 0, F2FS_GET_BLOCK_DEFAULT); + err = f2fs_map_blocks(inode, &map, F2FS_GET_BLOCK_DEFAULT); if (err) goto out; @@ -2651,7 +2651,7 @@ static int f2fs_defragment_range(struct f2fs_sb_info *sbi, do_map: map.m_len = pg_end - map.m_lblk; - err = f2fs_map_blocks(inode, &map, 0, F2FS_GET_BLOCK_DEFAULT); + err = f2fs_map_blocks(inode, &map, F2FS_GET_BLOCK_DEFAULT); if (err) goto clear_out; @@ -3225,7 +3225,7 @@ int f2fs_precache_extents(struct inode *inode) map.m_len = end - map.m_lblk; f2fs_down_write(&fi->i_gc_rwsem[WRITE]); - err = f2fs_map_blocks(inode, &map, 0, F2FS_GET_BLOCK_PRECACHE); + err = f2fs_map_blocks(inode, &map, F2FS_GET_BLOCK_PRECACHE); f2fs_up_write(&fi->i_gc_rwsem[WRITE]); if (err) return err; @@ -4464,7 +4464,7 @@ static int f2fs_preallocate_blocks(struct kiocb *iocb, struct iov_iter *iter, flag = F2FS_GET_BLOCK_PRE_AIO; } - ret = f2fs_map_blocks(inode, &map, 1, flag); + ret = f2fs_map_blocks(inode, &map, flag); /* -ENOSPC|-EDQUOT are fine to report the number of allocated blocks. */ if (ret < 0 && !((ret == -ENOSPC || ret == -EDQUOT) && map.m_len > 0)) return ret; diff --git a/include/trace/events/f2fs.h b/include/trace/events/f2fs.h index 31d994e6b4ca9..9183a0a11e26f 100644 --- a/include/trace/events/f2fs.h +++ b/include/trace/events/f2fs.h @@ -569,10 +569,10 @@ TRACE_EVENT(f2fs_file_write_iter, ); TRACE_EVENT(f2fs_map_blocks, - TP_PROTO(struct inode *inode, struct f2fs_map_blocks *map, - int create, int flag, int ret), + TP_PROTO(struct inode *inode, struct f2fs_map_blocks *map, int flag, + int ret), - TP_ARGS(inode, map, create, flag, ret), + TP_ARGS(inode, map, flag, ret), TP_STRUCT__entry( __field(dev_t, dev) @@ -584,7 +584,6 @@ TRACE_EVENT(f2fs_map_blocks, __field(int, m_seg_type) __field(bool, m_may_create) __field(bool, m_multidev_dio) - __field(int, create) __field(int, flag) __field(int, ret) ), @@ -599,7 +598,6 @@ TRACE_EVENT(f2fs_map_blocks, __entry->m_seg_type = map->m_seg_type; __entry->m_may_create = map->m_may_create; __entry->m_multidev_dio = map->m_multidev_dio; - __entry->create = create; __entry->flag = flag; __entry->ret = ret; ), @@ -607,7 +605,7 @@ TRACE_EVENT(f2fs_map_blocks, TP_printk("dev = (%d,%d), ino = %lu, file offset = %llu, " "start blkaddr = 0x%llx, len = 0x%llx, flags = %u, " "seg_type = %d, may_create = %d, multidevice = %d, " - "create = %d, flag = %d, err = %d", + "flag = %d, err = %d", show_dev_ino(__entry), (unsigned long long)__entry->m_lblk, (unsigned long long)__entry->m_pblk, @@ -616,7 +614,6 @@ TRACE_EVENT(f2fs_map_blocks, __entry->m_seg_type, __entry->m_may_create, __entry->m_multidev_dio, - __entry->create, __entry->flag, __entry->ret) ); From 0094e98bd1477a6b7d97c25b47b19a7317c35279 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Nov 2022 10:15:21 +0100 Subject: [PATCH 020/765] f2fs: factor a f2fs_map_blocks_cached helper Add a helper to deal with everything needed to return a f2fs_map_blocks structure based on a lookup in the extent cache. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 65 +++++++++++++++++++++++++++++--------------------- 1 file changed, 38 insertions(+), 27 deletions(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 1cc5c6cb31468..cb77f64e521ec 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -1472,6 +1472,41 @@ int f2fs_get_block_locked(struct dnode_of_data *dn, pgoff_t index) return err; } +static bool f2fs_map_blocks_cached(struct inode *inode, + struct f2fs_map_blocks *map, int flag) +{ + struct f2fs_sb_info *sbi = F2FS_I_SB(inode); + unsigned int maxblocks = map->m_len; + pgoff_t pgoff = (pgoff_t)map->m_lblk; + struct extent_info ei = {}; + + if (!f2fs_lookup_read_extent_cache(inode, pgoff, &ei)) + return false; + + map->m_pblk = ei.blk + pgoff - ei.fofs; + map->m_len = min((pgoff_t)maxblocks, ei.fofs + ei.len - pgoff); + map->m_flags = F2FS_MAP_MAPPED; + if (map->m_next_extent) + *map->m_next_extent = pgoff + map->m_len; + + /* for hardware encryption, but to avoid potential issue in future */ + if (flag == F2FS_GET_BLOCK_DIO) + f2fs_wait_on_block_writeback_range(inode, + map->m_pblk, map->m_len); + + if (f2fs_allow_multi_device_dio(sbi, flag)) { + int bidx = f2fs_target_device_index(sbi, map->m_pblk); + struct f2fs_dev_info *dev = &sbi->devs[bidx]; + + map->m_bdev = dev->bdev; + map->m_pblk -= dev->start_blk; + map->m_len = min(map->m_len, dev->end_blk + 1 - map->m_pblk); + } else { + map->m_bdev = inode->i_sb->s_bdev; + } + return true; +} + /* * f2fs_map_blocks() tries to find or build mapping relationship which * maps continuous logical blocks to physical blocks, and return such @@ -1487,7 +1522,6 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, int flag) int err = 0, ofs = 1; unsigned int ofs_in_node, last_ofs_in_node; blkcnt_t prealloc; - struct extent_info ei = {0, }; block_t blkaddr; unsigned int start_pgofs; int bidx = 0; @@ -1495,6 +1529,9 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, int flag) if (!maxblocks) return 0; + if (!map->m_may_create && f2fs_map_blocks_cached(inode, map, flag)) + goto out; + map->m_bdev = inode->i_sb->s_bdev; map->m_multidev_dio = f2fs_allow_multi_device_dio(F2FS_I_SB(inode), flag); @@ -1506,32 +1543,6 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, int flag) pgofs = (pgoff_t)map->m_lblk; end = pgofs + maxblocks; - if (map->m_may_create || - !f2fs_lookup_read_extent_cache(inode, pgofs, &ei)) - goto next_dnode; - - /* Found the map in read extent cache */ - map->m_pblk = ei.blk + pgofs - ei.fofs; - map->m_len = min((pgoff_t)maxblocks, ei.fofs + ei.len - pgofs); - map->m_flags = F2FS_MAP_MAPPED; - if (map->m_next_extent) - *map->m_next_extent = pgofs + map->m_len; - - /* for hardware encryption, but to avoid potential issue in future */ - if (flag == F2FS_GET_BLOCK_DIO) - f2fs_wait_on_block_writeback_range(inode, - map->m_pblk, map->m_len); - - if (map->m_multidev_dio) { - bidx = f2fs_target_device_index(sbi, map->m_pblk); - - map->m_bdev = FDEV(bidx).bdev; - map->m_pblk -= FDEV(bidx).start_blk; - map->m_len = min(map->m_len, - FDEV(bidx).end_blk + 1 - map->m_pblk); - } - goto out; - next_dnode: if (map->m_may_create) f2fs_map_lock(sbi, flag); From 817c968b79d02588370e3d1dc5e5961cfd57d2b1 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Nov 2022 10:15:22 +0100 Subject: [PATCH 021/765] f2fs: factor out a f2fs_map_no_dnode Factor out a helper to return a hole when no dnode was found. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 47 ++++++++++++++++++++++++----------------------- 1 file changed, 24 insertions(+), 23 deletions(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index cb77f64e521ec..2f05d6f750835 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -1472,6 +1472,28 @@ int f2fs_get_block_locked(struct dnode_of_data *dn, pgoff_t index) return err; } +static int f2fs_map_no_dnode(struct inode *inode, + struct f2fs_map_blocks *map, struct dnode_of_data *dn, + pgoff_t pgoff) +{ + struct f2fs_sb_info *sbi = F2FS_I_SB(inode); + + /* + * There is one exceptional case that read_node_page() may return + * -ENOENT due to filesystem has been shutdown or cp_error, return + * -EIO in that case. + */ + if (map->m_may_create && + (is_sbi_flag_set(sbi, SBI_IS_SHUTDOWN) || f2fs_cp_error(sbi))) + return -EIO; + + if (map->m_next_pgofs) + *map->m_next_pgofs = f2fs_get_next_page_offset(dn, pgoff); + if (map->m_next_extent) + *map->m_next_extent = f2fs_get_next_page_offset(dn, pgoff); + return 0; +} + static bool f2fs_map_blocks_cached(struct inode *inode, struct f2fs_map_blocks *map, int flag) { @@ -1553,29 +1575,8 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, int flag) if (err) { if (flag == F2FS_GET_BLOCK_BMAP) map->m_pblk = 0; - - if (err == -ENOENT) { - /* - * There is one exceptional case that read_node_page() - * may return -ENOENT due to filesystem has been - * shutdown or cp_error, so force to convert error - * number to EIO for such case. - */ - if (map->m_may_create && - (is_sbi_flag_set(sbi, SBI_IS_SHUTDOWN) || - f2fs_cp_error(sbi))) { - err = -EIO; - goto unlock_out; - } - - err = 0; - if (map->m_next_pgofs) - *map->m_next_pgofs = - f2fs_get_next_page_offset(&dn, pgofs); - if (map->m_next_extent) - *map->m_next_extent = - f2fs_get_next_page_offset(&dn, pgofs); - } + if (err == -ENOENT) + err = f2fs_map_no_dnode(inode, map, &dn, pgofs); goto unlock_out; } From fdbf69a7f5be2896af3a9a6213fb0ab8e194b190 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Nov 2022 10:15:23 +0100 Subject: [PATCH 022/765] f2fs: refactor the hole reporting and allocation logic in f2fs_map_blocks Add a is_hole local variable to figure out if the block number might need allocation, and untangle to logic to report the hole or fill it with a block allocation. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 113 ++++++++++++++++++++++++------------------------- 1 file changed, 56 insertions(+), 57 deletions(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 2f05d6f750835..c25ae041c0a54 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -1547,6 +1547,7 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, int flag) block_t blkaddr; unsigned int start_pgofs; int bidx = 0; + bool is_hole; if (!maxblocks) return 0; @@ -1587,78 +1588,76 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, int flag) next_block: blkaddr = f2fs_data_blkaddr(&dn); - - if (__is_valid_data_blkaddr(blkaddr) && - !f2fs_is_valid_blkaddr(sbi, blkaddr, DATA_GENERIC_ENHANCE)) { + is_hole = !__is_valid_data_blkaddr(blkaddr); + if (!is_hole && + !f2fs_is_valid_blkaddr(sbi, blkaddr, DATA_GENERIC_ENHANCE)) { err = -EFSCORRUPTED; f2fs_handle_error(sbi, ERROR_INVALID_BLKADDR); goto sync_out; } - if (__is_valid_data_blkaddr(blkaddr)) { - /* use out-place-update for driect IO under LFS mode */ - if (f2fs_lfs_mode(sbi) && flag == F2FS_GET_BLOCK_DIO && - map->m_may_create) { + /* use out-place-update for direct IO under LFS mode */ + if (map->m_may_create && + (is_hole || (f2fs_lfs_mode(sbi) && flag == F2FS_GET_BLOCK_DIO))) { + if (unlikely(f2fs_cp_error(sbi))) { + err = -EIO; + goto sync_out; + } + + switch (flag) { + case F2FS_GET_BLOCK_PRE_AIO: + if (blkaddr == NULL_ADDR) { + prealloc++; + last_ofs_in_node = dn.ofs_in_node; + } + break; + case F2FS_GET_BLOCK_PRE_DIO: + case F2FS_GET_BLOCK_DIO: err = __allocate_data_block(&dn, map->m_seg_type); if (err) goto sync_out; - blkaddr = dn.data_blkaddr; + if (flag == F2FS_GET_BLOCK_PRE_DIO) + file_need_truncate(inode); set_inode_flag(inode, FI_APPEND_WRITE); + break; + default: + WARN_ON_ONCE(1); + err = -EIO; + goto sync_out; } - } else { - if (map->m_may_create) { - if (unlikely(f2fs_cp_error(sbi))) { - err = -EIO; - goto sync_out; - } - if (flag == F2FS_GET_BLOCK_PRE_AIO) { - if (blkaddr == NULL_ADDR) { - prealloc++; - last_ofs_in_node = dn.ofs_in_node; - } - } else { - WARN_ON(flag != F2FS_GET_BLOCK_PRE_DIO && - flag != F2FS_GET_BLOCK_DIO); - err = __allocate_data_block(&dn, - map->m_seg_type); - if (!err) { - if (flag == F2FS_GET_BLOCK_PRE_DIO) - file_need_truncate(inode); - set_inode_flag(inode, FI_APPEND_WRITE); - } - } - if (err) - goto sync_out; + + blkaddr = dn.data_blkaddr; + if (is_hole) map->m_flags |= F2FS_MAP_NEW; - blkaddr = dn.data_blkaddr; - } else { - if (f2fs_compressed_file(inode) && - f2fs_sanity_check_cluster(&dn) && - (flag != F2FS_GET_BLOCK_FIEMAP || - IS_ENABLED(CONFIG_F2FS_CHECK_FS))) { - err = -EFSCORRUPTED; - f2fs_handle_error(sbi, - ERROR_CORRUPTED_CLUSTER); - goto sync_out; - } - if (flag == F2FS_GET_BLOCK_BMAP) { - map->m_pblk = 0; - goto sync_out; - } - if (flag == F2FS_GET_BLOCK_PRECACHE) - goto sync_out; - if (flag == F2FS_GET_BLOCK_FIEMAP && - blkaddr == NULL_ADDR) { - if (map->m_next_pgofs) - *map->m_next_pgofs = pgofs + 1; - goto sync_out; - } - if (flag != F2FS_GET_BLOCK_FIEMAP) { - /* for defragment case */ + } else if (is_hole) { + if (f2fs_compressed_file(inode) && + f2fs_sanity_check_cluster(&dn) && + (flag != F2FS_GET_BLOCK_FIEMAP || + IS_ENABLED(CONFIG_F2FS_CHECK_FS))) { + err = -EFSCORRUPTED; + f2fs_handle_error(sbi, + ERROR_CORRUPTED_CLUSTER); + goto sync_out; + } + + switch (flag) { + case F2FS_GET_BLOCK_PRECACHE: + goto sync_out; + case F2FS_GET_BLOCK_BMAP: + map->m_pblk = 0; + goto sync_out; + case F2FS_GET_BLOCK_FIEMAP: + if (blkaddr == NULL_ADDR) { if (map->m_next_pgofs) *map->m_next_pgofs = pgofs + 1; goto sync_out; } + break; + default: + /* for defragment case */ + if (map->m_next_pgofs) + *map->m_next_pgofs = pgofs + 1; + goto sync_out; } } From fdb7ccc3f9cb316c399b072c7a75a106678eb421 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Sat, 19 Nov 2022 03:18:39 +0800 Subject: [PATCH 023/765] f2fs: introduce IS_F2FS_IPU_* macro IS_F2FS_IPU_* macro can be used to identify whether f2fs ipu related policies are enabled. BTW, convert to use BIT() instead of open code. Signed-off-by: Yangtao Li Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 25 ++++++++++--------------- fs/f2fs/segment.c | 4 ++-- fs/f2fs/segment.h | 15 +++++++++++++++ fs/f2fs/super.c | 4 ++-- 4 files changed, 29 insertions(+), 19 deletions(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index c25ae041c0a54..60fb44a200c1a 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -2539,34 +2539,29 @@ static inline bool check_inplace_update_policy(struct inode *inode, struct f2fs_io_info *fio) { struct f2fs_sb_info *sbi = F2FS_I_SB(inode); - unsigned int policy = SM_I(sbi)->ipu_policy; - if (policy & (0x1 << F2FS_IPU_HONOR_OPU_WRITE) && - is_inode_flag_set(inode, FI_OPU_WRITE)) + if (IS_F2FS_IPU_HONOR_OPU_WRITE(sbi) && + is_inode_flag_set(inode, FI_OPU_WRITE)) return false; - if (policy & (0x1 << F2FS_IPU_FORCE)) + if (IS_F2FS_IPU_FORCE(sbi)) return true; - if (policy & (0x1 << F2FS_IPU_SSR) && f2fs_need_SSR(sbi)) + if (IS_F2FS_IPU_SSR(sbi) && f2fs_need_SSR(sbi)) return true; - if (policy & (0x1 << F2FS_IPU_UTIL) && - utilization(sbi) > SM_I(sbi)->min_ipu_util) + if (IS_F2FS_IPU_UTIL(sbi) && utilization(sbi) > SM_I(sbi)->min_ipu_util) return true; - if (policy & (0x1 << F2FS_IPU_SSR_UTIL) && f2fs_need_SSR(sbi) && - utilization(sbi) > SM_I(sbi)->min_ipu_util) + if (IS_F2FS_IPU_SSR_UTIL(sbi) && f2fs_need_SSR(sbi) && + utilization(sbi) > SM_I(sbi)->min_ipu_util) return true; /* * IPU for rewrite async pages */ - if (policy & (0x1 << F2FS_IPU_ASYNC) && - fio && fio->op == REQ_OP_WRITE && - !(fio->op_flags & REQ_SYNC) && - !IS_ENCRYPTED(inode)) + if (IS_F2FS_IPU_ASYNC(sbi) && fio && fio->op == REQ_OP_WRITE && + !(fio->op_flags & REQ_SYNC) && !IS_ENCRYPTED(inode)) return true; /* this is only set during fdatasync */ - if (policy & (0x1 << F2FS_IPU_FSYNC) && - is_inode_flag_set(inode, FI_NEED_IPU)) + if (IS_F2FS_IPU_FSYNC(sbi) && is_inode_flag_set(inode, FI_NEED_IPU)) return true; if (unlikely(fio && is_sbi_flag_set(sbi, SBI_CP_DISABLED) && diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index ae3c4e5474efa..37117cb530aed 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -3487,7 +3487,7 @@ int f2fs_inplace_write_data(struct f2fs_io_info *fio) stat_inc_inplace_blocks(fio->sbi); - if (fio->bio && !(SM_I(sbi)->ipu_policy & (1 << F2FS_IPU_NOCACHE))) + if (fio->bio && !IS_F2FS_IPU_NOCACHE(sbi)) err = f2fs_merge_page_bio(fio); else err = f2fs_submit_page_bio(fio); @@ -5126,7 +5126,7 @@ int f2fs_build_segment_manager(struct f2fs_sb_info *sbi) sm_info->rec_prefree_segments = DEF_MAX_RECLAIM_PREFREE_SEGMENTS; if (!f2fs_lfs_mode(sbi)) - sm_info->ipu_policy = 1 << F2FS_IPU_FSYNC; + sm_info->ipu_policy = BIT(F2FS_IPU_FSYNC); sm_info->min_ipu_util = DEF_MIN_IPU_UTIL; sm_info->min_fsync_blocks = DEF_MIN_FSYNC_BLOCKS; sm_info->min_seq_blocks = sbi->blocks_per_seg; diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h index 3ad1b7b6fa946..e77518c49f388 100644 --- a/fs/f2fs/segment.h +++ b/fs/f2fs/segment.h @@ -681,6 +681,21 @@ enum { F2FS_IPU_HONOR_OPU_WRITE, }; +#define F2FS_IPU_POLICY(name) \ +static inline int IS_##name(struct f2fs_sb_info *sbi) \ +{ \ + return SM_I(sbi)->ipu_policy & BIT(name); \ +} + +F2FS_IPU_POLICY(F2FS_IPU_FORCE); +F2FS_IPU_POLICY(F2FS_IPU_SSR); +F2FS_IPU_POLICY(F2FS_IPU_UTIL); +F2FS_IPU_POLICY(F2FS_IPU_SSR_UTIL); +F2FS_IPU_POLICY(F2FS_IPU_FSYNC); +F2FS_IPU_POLICY(F2FS_IPU_ASYNC); +F2FS_IPU_POLICY(F2FS_IPU_NOCACHE); +F2FS_IPU_POLICY(F2FS_IPU_HONOR_OPU_WRITE); + static inline unsigned int curseg_segno(struct f2fs_sb_info *sbi, int type) { diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 1f812b9ce985b..87d56a9883e65 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -4089,8 +4089,8 @@ static void f2fs_tuning_parameters(struct f2fs_sb_info *sbi) if (f2fs_block_unit_discard(sbi)) SM_I(sbi)->dcc_info->discard_granularity = MIN_DISCARD_GRANULARITY; - SM_I(sbi)->ipu_policy = 1 << F2FS_IPU_FORCE | - 1 << F2FS_IPU_HONOR_OPU_WRITE; + SM_I(sbi)->ipu_policy = BIT(F2FS_IPU_FORCE) | + BIT(F2FS_IPU_HONOR_OPU_WRITE); } sbi->readdir_ra = true; From 5eaac835f27f2de6b73412d7c24e755733b49de0 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Fri, 16 Dec 2022 23:50:00 +0800 Subject: [PATCH 024/765] f2fs: fix to avoid potential deadlock There is a potential deadlock reported by syzbot as below: F2FS-fs (loop2): invalid crc value F2FS-fs (loop2): Found nat_bits in checkpoint F2FS-fs (loop2): Mounted with checkpoint version = 48b305e4 ====================================================== WARNING: possible circular locking dependency detected 6.1.0-rc8-syzkaller-33330-ga5541c0811a0 #0 Not tainted ------------------------------------------------------ syz-executor.2/32123 is trying to acquire lock: ffff0000c0e1a608 (&mm->mmap_lock){++++}-{3:3}, at: __might_fault+0x54/0xb4 mm/memory.c:5644 but task is already holding lock: ffff0001317c6088 (&sbi->sb_lock){++++}-{3:3}, at: f2fs_down_write fs/f2fs/f2fs.h:2205 [inline] ffff0001317c6088 (&sbi->sb_lock){++++}-{3:3}, at: f2fs_ioc_get_encryption_pwsalt fs/f2fs/file.c:2334 [inline] ffff0001317c6088 (&sbi->sb_lock){++++}-{3:3}, at: __f2fs_ioctl+0x1370/0x3318 fs/f2fs/file.c:4151 which lock already depends on the new lock. Chain exists of: &mm->mmap_lock --> &nm_i->nat_tree_lock --> &sbi->sb_lock Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&sbi->sb_lock); lock(&nm_i->nat_tree_lock); lock(&sbi->sb_lock); lock(&mm->mmap_lock); Let's try to avoid above deadlock condition by moving __might_fault() out of sbi->sb_lock coverage. Fixes: 95fa90c9e5a7 ("f2fs: support recording errors into superblock") Link: https://lore.kernel.org/linux-f2fs-devel/000000000000cd5fe305ef617fe2@google.com/T/#u Reported-by: syzbot+4793f6096d174c90b4f7@syzkaller.appspotmail.com Signed-off-by: Chao Yu Reviewed-by: Eric Biggers Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 939c7247c3678..bef0ee57e88d0 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -2336,6 +2336,7 @@ static int f2fs_ioc_get_encryption_pwsalt(struct file *filp, unsigned long arg) { struct inode *inode = file_inode(filp); struct f2fs_sb_info *sbi = F2FS_I_SB(inode); + u8 encrypt_pw_salt[16]; int err; if (!f2fs_sb_has_encrypt(sbi)) @@ -2360,12 +2361,14 @@ static int f2fs_ioc_get_encryption_pwsalt(struct file *filp, unsigned long arg) goto out_err; } got_it: - if (copy_to_user((__u8 __user *)arg, sbi->raw_super->encrypt_pw_salt, - 16)) - err = -EFAULT; + memcpy(encrypt_pw_salt, sbi->raw_super->encrypt_pw_salt, 16); out_err: f2fs_up_write(&sbi->sb_lock); mnt_drop_write_file(filp); + + if (!err && copy_to_user((__u8 __user *)arg, encrypt_pw_salt, 16)) + err = -EFAULT; + return err; } From cec32b00faa940f4ed91a939d3beb71410334039 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Thu, 15 Dec 2022 23:03:30 +0800 Subject: [PATCH 025/765] f2fs: add missing doc for fault injection sysfs We supported configuring fault injection parameter via sysfs w/ below commits, however, we forgot to add doc entry, fix it. commit 087968974fcd ("f2fs: add fault injection to sysfs") /sys/fs/f2fs/fault_injection/fault_* commit 1ecc0c5c50ce ("f2fs: support configuring fault injection per superblock") /sys/fs/f2fs//fault_* Cc: Sheng Yong Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- Documentation/ABI/testing/sysfs-fs-f2fs | 39 +++++++++++++++++++++++++ 1 file changed, 39 insertions(+) diff --git a/Documentation/ABI/testing/sysfs-fs-f2fs b/Documentation/ABI/testing/sysfs-fs-f2fs index 9e3756625a819..aaa379bb8a8fc 100644 --- a/Documentation/ABI/testing/sysfs-fs-f2fs +++ b/Documentation/ABI/testing/sysfs-fs-f2fs @@ -669,3 +669,42 @@ Contact: "Ping Xiong" Description: When DATA SEPARATION is on, it controls the age threshold to indicate the data blocks as warm. By default it was initialized as 2621440 blocks (equals to 10GB). + +What: /sys/fs/f2fs//fault_rate +Date: May 2016 +Contact: "Sheng Yong" +Contact: "Chao Yu" +Description: Enable fault injection in all supported types with + specified injection rate. + +What: /sys/fs/f2fs//fault_type +Date: May 2016 +Contact: "Sheng Yong" +Contact: "Chao Yu" +Description: Support configuring fault injection type, should be + enabled with fault_injection option, fault type value + is shown below, it supports single or combined type. + + =================== =========== + Type_Name Type_Value + =================== =========== + FAULT_KMALLOC 0x000000001 + FAULT_KVMALLOC 0x000000002 + FAULT_PAGE_ALLOC 0x000000004 + FAULT_PAGE_GET 0x000000008 + FAULT_ALLOC_BIO 0x000000010 (obsolete) + FAULT_ALLOC_NID 0x000000020 + FAULT_ORPHAN 0x000000040 + FAULT_BLOCK 0x000000080 + FAULT_DIR_DEPTH 0x000000100 + FAULT_EVICT_INODE 0x000000200 + FAULT_TRUNCATE 0x000000400 + FAULT_READ_IO 0x000000800 + FAULT_CHECKPOINT 0x000001000 + FAULT_DISCARD 0x000002000 + FAULT_WRITE_IO 0x000004000 + FAULT_SLAB_ALLOC 0x000008000 + FAULT_DQUOT_INIT 0x000010000 + FAULT_LOCK_OP 0x000020000 + FAULT_BLKADDR 0x000040000 + =================== =========== From 8358014d6be8f3cb507d247d6a623e5961f848d0 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Thu, 15 Dec 2022 14:05:06 +0800 Subject: [PATCH 026/765] f2fs: avoid to check PG_error flag After below changes: commit 14db0b3c7b83 ("fscrypt: stop using PG_error to track error status") commit 98dc08bae678 ("fsverity: stop using PG_error to track error status") There is no place in f2fs we will set PG_error flag in page, let's remove other PG_error usage in f2fs, as a step towards freeing the PG_error flag for other uses. Cc: Eric Biggers Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 2 -- fs/f2fs/gc.c | 1 - fs/f2fs/inline.c | 1 - fs/f2fs/node.c | 3 --- fs/f2fs/segment.c | 1 - 5 files changed, 8 deletions(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 60fb44a200c1a..af906ed6d53d3 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -2701,7 +2701,6 @@ int f2fs_do_write_data_page(struct f2fs_io_info *fio) goto out_writepage; set_page_writeback(page); - ClearPageError(page); f2fs_put_dnode(&dn); if (fio->need_lock == LOCK_REQ) f2fs_unlock_op(fio->sbi); @@ -2737,7 +2736,6 @@ int f2fs_do_write_data_page(struct f2fs_io_info *fio) goto out_writepage; set_page_writeback(page); - ClearPageError(page); if (fio->compr_blocks && fio->old_blkaddr == COMPRESS_ADDR) f2fs_i_compr_blocks_update(inode, fio->compr_blocks - 1, false); diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index 83e68ec7763de..7444c392eab16 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -1364,7 +1364,6 @@ static int move_data_block(struct inode *inode, block_t bidx, dec_page_count(fio.sbi, F2FS_DIRTY_META); set_page_writeback(fio.encrypted_page); - ClearPageError(page); fio.op = REQ_OP_WRITE; fio.op_flags = REQ_SYNC; diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c index 21a495234ffd7..08e302d32118d 100644 --- a/fs/f2fs/inline.c +++ b/fs/f2fs/inline.c @@ -174,7 +174,6 @@ int f2fs_convert_inline_page(struct dnode_of_data *dn, struct page *page) /* write data page to try to make data consistent */ set_page_writeback(page); - ClearPageError(page); fio.old_blkaddr = dn->data_blkaddr; set_inode_flag(dn->inode, FI_HOT_DATA); f2fs_outplace_write_data(dn, &fio); diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c index dde4c04587049..558b318f73167 100644 --- a/fs/f2fs/node.c +++ b/fs/f2fs/node.c @@ -1650,7 +1650,6 @@ static int __write_node_page(struct page *page, bool atomic, bool *submitted, } set_page_writeback(page); - ClearPageError(page); fio.old_blkaddr = ni.blk_addr; f2fs_do_write_node_page(nid, &fio); @@ -2079,8 +2078,6 @@ int f2fs_wait_on_node_pages_writeback(struct f2fs_sb_info *sbi, spin_unlock_irqrestore(&sbi->fsync_node_lock, flags); f2fs_wait_on_page_writeback(page, NODE, true, false); - if (TestClearPageError(page)) - ret = -EIO; put_page(page); diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 37117cb530aed..a27ffc6272645 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -3422,7 +3422,6 @@ void f2fs_do_write_meta_page(struct f2fs_sb_info *sbi, struct page *page, fio.op_flags &= ~REQ_META; set_page_writeback(page); - ClearPageError(page); f2fs_submit_page_write(&fio); stat_inc_meta_count(sbi, page->index); From 185a453bf1b5688f8c77f2646b0b6f3b1cbdddca Mon Sep 17 00:00:00 2001 From: Yuwei Guan Date: Tue, 13 Dec 2022 17:34:19 +0800 Subject: [PATCH 027/765] f2fs: deliver the accumulated 'issued' to __issue_discard_cmd_orderly() Any of the following scenarios will send more than the number of max_requests at a time, which will not meet the design of the max_requests limit. - Set max_ordered_discard larger than discard_granularity from userspace. - It is a small size device, discard_granularity can be tuned to 1 in f2fs_tuning_parameters(). We need to deliver the accumulated @issued to __issue_discard_cmd_orderly() to meet the max_requests limit. BTW, convert the parameter type of @issued in __submit_discard_cmd(). Signed-off-by: Yuwei Guan Cc: Bagas Sanjaya Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 26 ++++++++++++-------------- 1 file changed, 12 insertions(+), 14 deletions(-) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index a27ffc6272645..e2f95f46d2989 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -1095,9 +1095,8 @@ static void __update_discard_tree_range(struct f2fs_sb_info *sbi, block_t start, block_t len); /* this function is copied from blkdev_issue_discard from block/blk-lib.c */ static int __submit_discard_cmd(struct f2fs_sb_info *sbi, - struct discard_policy *dpolicy, - struct discard_cmd *dc, - unsigned int *issued) + struct discard_policy *dpolicy, + struct discard_cmd *dc, int *issued) { struct block_device *bdev = dc->bdev; unsigned int max_discard_blocks = @@ -1378,8 +1377,8 @@ static void __queue_discard_cmd(struct f2fs_sb_info *sbi, mutex_unlock(&SM_I(sbi)->dcc_info->cmd_lock); } -static unsigned int __issue_discard_cmd_orderly(struct f2fs_sb_info *sbi, - struct discard_policy *dpolicy) +static void __issue_discard_cmd_orderly(struct f2fs_sb_info *sbi, + struct discard_policy *dpolicy, int *issued) { struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info; struct discard_cmd *prev_dc = NULL, *next_dc = NULL; @@ -1387,7 +1386,6 @@ static unsigned int __issue_discard_cmd_orderly(struct f2fs_sb_info *sbi, struct discard_cmd *dc; struct blk_plug plug; unsigned int pos = dcc->next_pos; - unsigned int issued = 0; bool io_interrupted = false; mutex_lock(&dcc->cmd_lock); @@ -1414,9 +1412,9 @@ static unsigned int __issue_discard_cmd_orderly(struct f2fs_sb_info *sbi, } dcc->next_pos = dc->lstart + dc->len; - err = __submit_discard_cmd(sbi, dpolicy, dc, &issued); + err = __submit_discard_cmd(sbi, dpolicy, dc, issued); - if (issued >= dpolicy->max_requests) + if (*issued >= dpolicy->max_requests) break; next: node = rb_next(&dc->rb_node); @@ -1432,10 +1430,8 @@ static unsigned int __issue_discard_cmd_orderly(struct f2fs_sb_info *sbi, mutex_unlock(&dcc->cmd_lock); - if (!issued && io_interrupted) - issued = -1; - - return issued; + if (!(*issued) && io_interrupted) + *issued = -1; } static unsigned int __wait_all_discard_cmd(struct f2fs_sb_info *sbi, struct discard_policy *dpolicy); @@ -1463,8 +1459,10 @@ static int __issue_discard_cmd(struct f2fs_sb_info *sbi, if (i + 1 < dpolicy->granularity) break; - if (i + 1 < dcc->max_ordered_discard && dpolicy->ordered) - return __issue_discard_cmd_orderly(sbi, dpolicy); + if (i + 1 < dcc->max_ordered_discard && dpolicy->ordered) { + __issue_discard_cmd_orderly(sbi, dpolicy, &issued); + return issued; + } pend_list = &dcc->pend_list[i]; From 3da73f102309fe29150e5c35acd20dd82063ff67 Mon Sep 17 00:00:00 2001 From: Miaoqian Lin Date: Mon, 5 Dec 2022 12:06:42 +0400 Subject: [PATCH 028/765] objtool: Fix memory leak in create_static_call_sections() strdup() allocates memory for key_name. We need to release the memory in the following error paths. Add free() to avoid memory leak. Fixes: 1e7e47883830 ("x86/static_call: Add inline static call implementation for x86-64") Signed-off-by: Miaoqian Lin Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20221205080642.558583-1-linmq006@gmail.com Cc: Josh Poimboeuf Cc: Peter Zijlstra --- tools/objtool/check.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tools/objtool/check.c b/tools/objtool/check.c index 4350be739f4fa..cab1a162781c4 100644 --- a/tools/objtool/check.c +++ b/tools/objtool/check.c @@ -679,6 +679,7 @@ static int create_static_call_sections(struct objtool_file *file) if (strncmp(key_name, STATIC_CALL_TRAMP_PREFIX_STR, STATIC_CALL_TRAMP_PREFIX_LEN)) { WARN("static_call: trampoline name malformed: %s", key_name); + free(key_name); return -1; } tmp = key_name + STATIC_CALL_TRAMP_PREFIX_LEN - STATIC_CALL_KEY_PREFIX_LEN; @@ -688,6 +689,7 @@ static int create_static_call_sections(struct objtool_file *file) if (!key_sym) { if (!opts.module) { WARN("static_call: can't find static_call_key symbol: %s", tmp); + free(key_name); return -1; } From 4cd0ca1fe9a79d81a001ff14f14035531773fe43 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Fri, 18 Nov 2022 23:44:54 +0100 Subject: [PATCH 029/765] rtc: ds1307: Convert to i2c's .probe_new() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit .probe_new() doesn't get the i2c_device_id * parameter, so determine that explicitly in the probe function. Signed-off-by: Uwe Kleine-König Link: https://lore.kernel.org/r/20221118224540.619276-561-uwe@kleine-koenig.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-ds1307.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/rtc/rtc-ds1307.c b/drivers/rtc/rtc-ds1307.c index def9b7f9d9577..e86ba84df6cbe 100644 --- a/drivers/rtc/rtc-ds1307.c +++ b/drivers/rtc/rtc-ds1307.c @@ -1712,9 +1712,9 @@ static const struct regmap_config regmap_config = { .val_bits = 8, }; -static int ds1307_probe(struct i2c_client *client, - const struct i2c_device_id *id) +static int ds1307_probe(struct i2c_client *client) { + const struct i2c_device_id *id = i2c_client_get_device_id(client); struct ds1307 *ds1307; const void *match; int err = -ENODEV; @@ -2011,7 +2011,7 @@ static struct i2c_driver ds1307_driver = { .name = "rtc-ds1307", .of_match_table = ds1307_of_match, }, - .probe = ds1307_probe, + .probe_new = ds1307_probe, .id_table = ds1307_id, }; From 275dd5dc67f2a4d7a81c22869ff594701519dc9b Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Fri, 16 Dec 2022 10:45:06 +0800 Subject: [PATCH 030/765] MAINTAINERS: Add f2fs's patchwork >From now on, f2fs also has its own patchwork link, thanks to Jaegeuk for starting this tool! Let's update it to f2fs entry. Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index 7f86d02cb427a..d0485b58b9d90 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -7922,6 +7922,7 @@ M: Chao Yu L: linux-f2fs-devel@lists.sourceforge.net S: Maintained W: https://f2fs.wiki.kernel.org/ +Q: https://patchwork.kernel.org/project/f2fs/list/ B: https://bugzilla.kernel.org/enter_bug.cgi?product=File%20System&component=f2fs T: git git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs.git F: Documentation/ABI/testing/sysfs-fs-f2fs From ebaaec351e4fef86afe38002b6a8c15178fd3610 Mon Sep 17 00:00:00 2001 From: Zhang Qilong Date: Tue, 6 Sep 2022 22:53:47 +0800 Subject: [PATCH 031/765] f2fs: start freeing cluster pages from the unused number We can start freeing cluster page(s) from which compression is not used. It will get better performance. Signed-off-by: Zhang Qilong Signed-off-by: Jaegeuk Kim --- fs/f2fs/compress.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c index e4db3b7932a37..dd1caba46eed7 100644 --- a/fs/f2fs/compress.c +++ b/fs/f2fs/compress.c @@ -690,9 +690,7 @@ static int f2fs_compress_pages(struct compress_ctx *cc) vm_unmap_ram(cc->cbuf, cc->nr_cpages); vm_unmap_ram(cc->rbuf, cc->cluster_size); - for (i = 0; i < cc->nr_cpages; i++) { - if (i < new_nr_cpages) - continue; + for (i = new_nr_cpages; i < cc->nr_cpages; i++) { f2fs_compress_free_page(cc->cpages[i]); cc->cpages[i] = NULL; } From b5a711acab305e04278c136c841ba37c589c16a1 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Tue, 29 Nov 2022 20:29:28 +0800 Subject: [PATCH 032/765] f2fs: judge whether discard_unit is section only when have CONFIG_BLK_DEV_ZONED The current logic, regardless of whether CONFIG_BLK_DEV_ZONED is enabled or not, will judge whether discard_unit is SECTION, when f2fs_sb_has_blkzoned. In fact, when CONFIG_BLK_DEV_ZONED is not enabled, this judgment is a path that will never be accessed. At this time, -EINVAL will be returned in the parse_options function, accompanied by the message "Zoned block device support is not enabled". Let's wrap this discard_unit judgment with CONFIG_BLK_DEV_ZONED. Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/super.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 87d56a9883e65..1d057a4c66424 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -1294,19 +1294,18 @@ static int parse_options(struct super_block *sb, char *options, bool is_remount) * zone alignment optimization. This is optional for host-aware * devices, but mandatory for host-managed zoned block devices. */ -#ifndef CONFIG_BLK_DEV_ZONED - if (f2fs_sb_has_blkzoned(sbi)) { - f2fs_err(sbi, "Zoned block device support is not enabled"); - return -EINVAL; - } -#endif if (f2fs_sb_has_blkzoned(sbi)) { +#ifdef CONFIG_BLK_DEV_ZONED if (F2FS_OPTION(sbi).discard_unit != DISCARD_UNIT_SECTION) { f2fs_info(sbi, "Zoned block device doesn't need small discard, set discard_unit=section by default"); F2FS_OPTION(sbi).discard_unit = DISCARD_UNIT_SECTION; } +#else + f2fs_err(sbi, "Zoned block device support is not enabled"); + return -EINVAL; +#endif } #ifdef CONFIG_F2FS_FS_COMPRESSION From a1357a91ec9ee72cf7683ac17b5f26dd5b4e6e5f Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Thu, 29 Dec 2022 21:18:28 +0800 Subject: [PATCH 033/765] f2fs: mark f2fs_init_compress_mempool w/ __init f2fs_init_compress_mempool() only initializes the memory pool during the f2fs module init phase. Let's mark it as __init like any other function. Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/compress.c | 2 +- fs/f2fs/f2fs.h | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c index dd1caba46eed7..b196b881f3bbc 100644 --- a/fs/f2fs/compress.c +++ b/fs/f2fs/compress.c @@ -564,7 +564,7 @@ module_param(num_compress_pages, uint, 0444); MODULE_PARM_DESC(num_compress_pages, "Number of intermediate compress pages to preallocate"); -int f2fs_init_compress_mempool(void) +int __init f2fs_init_compress_mempool(void) { compress_page_pool = mempool_create_page_pool(num_compress_pages, 0); return compress_page_pool ? 0 : -ENOMEM; diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 60f421fe14eee..378734b0d99a0 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -4251,7 +4251,7 @@ bool f2fs_compress_write_end(struct inode *inode, void *fsdata, int f2fs_truncate_partial_cluster(struct inode *inode, u64 from, bool lock); void f2fs_compress_write_end_io(struct bio *bio, struct page *page); bool f2fs_is_compress_backend_ready(struct inode *inode); -int f2fs_init_compress_mempool(void); +int __init f2fs_init_compress_mempool(void); void f2fs_destroy_compress_mempool(void); void f2fs_decompress_cluster(struct decompress_io_ctx *dic, bool in_task); void f2fs_end_read_compressed_page(struct page *page, bool failed, @@ -4320,7 +4320,7 @@ static inline struct page *f2fs_compress_control_page(struct page *page) WARN_ON_ONCE(1); return ERR_PTR(-EINVAL); } -static inline int f2fs_init_compress_mempool(void) { return 0; } +static inline int __init f2fs_init_compress_mempool(void) { return 0; } static inline void f2fs_destroy_compress_mempool(void) { } static inline void f2fs_decompress_cluster(struct decompress_io_ctx *dic, bool in_task) { } From 390d0b99212e42c6f9116562a8442888f6a0141d Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Thu, 22 Dec 2022 03:26:34 +0800 Subject: [PATCH 034/765] f2fs: remove unnecessary blank lines Just cleanup. Signed-off-by: Yangtao Li Signed-off-by: Jaegeuk Kim --- fs/f2fs/f2fs.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 378734b0d99a0..d1fcaa973a53e 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -1382,8 +1382,6 @@ enum { MEMORY_MODE_LOW, /* memory mode for low memry devices */ }; - - static inline int f2fs_test_bit(unsigned int nr, char *addr); static inline void f2fs_set_bit(unsigned int nr, char *addr); static inline void f2fs_clear_bit(unsigned int nr, char *addr); From 1cd7565449de6e838ff5a3fa75fbd2ce58c6d955 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Wed, 21 Dec 2022 03:12:12 +0800 Subject: [PATCH 035/765] f2fs: add a f2fs_ prefix to punch_hole() and expand_inode_data() For example, f2fs_collapse_range(), f2fs_collapse_range(), f2fs_insert_range(), the functions used in f2fs_fallocate() are all prefixed with f2fs_, so let's keep the name consistent. Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index bef0ee57e88d0..58b4200f7fdf0 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -1110,7 +1110,7 @@ int f2fs_truncate_hole(struct inode *inode, pgoff_t pg_start, pgoff_t pg_end) return 0; } -static int punch_hole(struct inode *inode, loff_t offset, loff_t len) +static int f2fs_punch_hole(struct inode *inode, loff_t offset, loff_t len) { pgoff_t pg_start, pg_end; loff_t off_start, off_end; @@ -1682,7 +1682,7 @@ static int f2fs_insert_range(struct inode *inode, loff_t offset, loff_t len) return ret; } -static int expand_inode_data(struct inode *inode, loff_t offset, +static int f2fs_expand_inode_data(struct inode *inode, loff_t offset, loff_t len, int mode) { struct f2fs_sb_info *sbi = F2FS_I_SB(inode); @@ -1830,7 +1830,7 @@ static long f2fs_fallocate(struct file *file, int mode, if (offset >= inode->i_size) goto out; - ret = punch_hole(inode, offset, len); + ret = f2fs_punch_hole(inode, offset, len); } else if (mode & FALLOC_FL_COLLAPSE_RANGE) { ret = f2fs_collapse_range(inode, offset, len); } else if (mode & FALLOC_FL_ZERO_RANGE) { @@ -1838,7 +1838,7 @@ static long f2fs_fallocate(struct file *file, int mode, } else if (mode & FALLOC_FL_INSERT_RANGE) { ret = f2fs_insert_range(inode, offset, len); } else { - ret = expand_inode_data(inode, offset, len, mode); + ret = f2fs_expand_inode_data(inode, offset, len, mode); } if (!ret) { From c40e15a9a59f79e79d9500f1fd019321ec35b959 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Wed, 21 Dec 2022 02:39:04 +0800 Subject: [PATCH 036/765] f2fs: merge f2fs_show_injection_info() into time_to_inject() There is no need to additionally use f2fs_show_injection_info() to output information. Concatenate time_to_inject() and __time_to_inject() via a macro. In the new __time_to_inject() function, pass in the caller function name and parent function. In this way, we no longer need the f2fs_show_injection_info() function, and let's remove it. Suggested-by: Chao Yu Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/checkpoint.c | 5 +---- fs/f2fs/data.c | 8 ++------ fs/f2fs/dir.c | 4 +--- fs/f2fs/f2fs.h | 44 ++++++++++++++------------------------------ fs/f2fs/file.c | 4 +--- fs/f2fs/gc.c | 4 +--- fs/f2fs/inode.c | 4 +--- fs/f2fs/node.c | 4 +--- fs/f2fs/segment.c | 5 +---- fs/f2fs/super.c | 8 ++------ 10 files changed, 25 insertions(+), 65 deletions(-) diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c index 56f7d0d6a8b2f..d68b3c991888d 100644 --- a/fs/f2fs/checkpoint.c +++ b/fs/f2fs/checkpoint.c @@ -171,10 +171,8 @@ static bool __is_bitmap_valid(struct f2fs_sb_info *sbi, block_t blkaddr, bool f2fs_is_valid_blkaddr(struct f2fs_sb_info *sbi, block_t blkaddr, int type) { - if (time_to_inject(sbi, FAULT_BLKADDR)) { - f2fs_show_injection_info(sbi, FAULT_BLKADDR); + if (time_to_inject(sbi, FAULT_BLKADDR)) return false; - } switch (type) { case META_NAT: @@ -622,7 +620,6 @@ int f2fs_acquire_orphan_inode(struct f2fs_sb_info *sbi) if (time_to_inject(sbi, FAULT_ORPHAN)) { spin_unlock(&im->ino_lock); - f2fs_show_injection_info(sbi, FAULT_ORPHAN); return -ENOSPC; } diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index af906ed6d53d3..c940da1c540f5 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -295,10 +295,8 @@ static void f2fs_read_end_io(struct bio *bio) iostat_update_and_unbind_ctx(bio, 0); ctx = bio->bi_private; - if (time_to_inject(sbi, FAULT_READ_IO)) { - f2fs_show_injection_info(sbi, FAULT_READ_IO); + if (time_to_inject(sbi, FAULT_READ_IO)) bio->bi_status = BLK_STS_IOERR; - } if (bio->bi_status) { f2fs_finish_read_bio(bio, intask); @@ -335,10 +333,8 @@ static void f2fs_write_end_io(struct bio *bio) iostat_update_and_unbind_ctx(bio, 1); sbi = bio->bi_private; - if (time_to_inject(sbi, FAULT_WRITE_IO)) { - f2fs_show_injection_info(sbi, FAULT_WRITE_IO); + if (time_to_inject(sbi, FAULT_WRITE_IO)) bio->bi_status = BLK_STS_IOERR; - } bio_for_each_segment_all(bvec, bio, iter_all) { struct page *page = bvec->bv_page; diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c index 8e025157f35c9..9ccdbe120425e 100644 --- a/fs/f2fs/dir.c +++ b/fs/f2fs/dir.c @@ -732,10 +732,8 @@ int f2fs_add_regular_entry(struct inode *dir, const struct f2fs_filename *fname, } start: - if (time_to_inject(F2FS_I_SB(dir), FAULT_DIR_DEPTH)) { - f2fs_show_injection_info(F2FS_I_SB(dir), FAULT_DIR_DEPTH); + if (time_to_inject(F2FS_I_SB(dir), FAULT_DIR_DEPTH)) return -ENOSPC; - } if (unlikely(current_depth == MAX_DIR_HASH_DEPTH)) return -ENOSPC; diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index d1fcaa973a53e..53d7ca96df637 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -1867,12 +1867,10 @@ struct f2fs_sb_info { }; #ifdef CONFIG_F2FS_FAULT_INJECTION -#define f2fs_show_injection_info(sbi, type) \ - printk_ratelimited("%sF2FS-fs (%s) : inject %s in %s of %pS\n", \ - KERN_INFO, sbi->sb->s_id, \ - f2fs_fault_name[type], \ - __func__, __builtin_return_address(0)) -static inline bool time_to_inject(struct f2fs_sb_info *sbi, int type) +#define time_to_inject(sbi, type) __time_to_inject(sbi, type, __func__, \ + __builtin_return_address(0)) +static inline bool __time_to_inject(struct f2fs_sb_info *sbi, int type, + const char *func, const char *parent_func) { struct f2fs_fault_info *ffi = &F2FS_OPTION(sbi).fault_info; @@ -1885,12 +1883,14 @@ static inline bool time_to_inject(struct f2fs_sb_info *sbi, int type) atomic_inc(&ffi->inject_ops); if (atomic_read(&ffi->inject_ops) >= ffi->inject_rate) { atomic_set(&ffi->inject_ops, 0); + printk_ratelimited("%sF2FS-fs (%s) : inject %s in %s of %pS\n", + KERN_INFO, sbi->sb->s_id, f2fs_fault_name[type], + func, parent_func); return true; } return false; } #else -#define f2fs_show_injection_info(sbi, type) do { } while (0) static inline bool time_to_inject(struct f2fs_sb_info *sbi, int type) { return false; @@ -2223,10 +2223,8 @@ static inline void f2fs_lock_op(struct f2fs_sb_info *sbi) static inline int f2fs_trylock_op(struct f2fs_sb_info *sbi) { - if (time_to_inject(sbi, FAULT_LOCK_OP)) { - f2fs_show_injection_info(sbi, FAULT_LOCK_OP); + if (time_to_inject(sbi, FAULT_LOCK_OP)) return 0; - } return f2fs_down_read_trylock(&sbi->cp_rwsem); } @@ -2314,7 +2312,6 @@ static inline int inc_valid_block_count(struct f2fs_sb_info *sbi, return ret; if (time_to_inject(sbi, FAULT_BLOCK)) { - f2fs_show_injection_info(sbi, FAULT_BLOCK); release = *count; goto release_quota; } @@ -2594,10 +2591,8 @@ static inline int inc_valid_node_count(struct f2fs_sb_info *sbi, return err; } - if (time_to_inject(sbi, FAULT_BLOCK)) { - f2fs_show_injection_info(sbi, FAULT_BLOCK); + if (time_to_inject(sbi, FAULT_BLOCK)) goto enospc; - } spin_lock(&sbi->stat_lock); @@ -2721,11 +2716,8 @@ static inline struct page *f2fs_grab_cache_page(struct address_space *mapping, if (page) return page; - if (time_to_inject(F2FS_M_SB(mapping), FAULT_PAGE_ALLOC)) { - f2fs_show_injection_info(F2FS_M_SB(mapping), - FAULT_PAGE_ALLOC); + if (time_to_inject(F2FS_M_SB(mapping), FAULT_PAGE_ALLOC)) return NULL; - } } if (!for_write) @@ -2742,10 +2734,8 @@ static inline struct page *f2fs_pagecache_get_page( struct address_space *mapping, pgoff_t index, int fgp_flags, gfp_t gfp_mask) { - if (time_to_inject(F2FS_M_SB(mapping), FAULT_PAGE_GET)) { - f2fs_show_injection_info(F2FS_M_SB(mapping), FAULT_PAGE_GET); + if (time_to_inject(F2FS_M_SB(mapping), FAULT_PAGE_GET)) return NULL; - } return pagecache_get_page(mapping, index, fgp_flags, gfp_mask); } @@ -2795,10 +2785,8 @@ static inline void *f2fs_kmem_cache_alloc(struct kmem_cache *cachep, if (nofail) return f2fs_kmem_cache_alloc_nofail(cachep, flags); - if (time_to_inject(sbi, FAULT_SLAB_ALLOC)) { - f2fs_show_injection_info(sbi, FAULT_SLAB_ALLOC); + if (time_to_inject(sbi, FAULT_SLAB_ALLOC)) return NULL; - } return kmem_cache_alloc(cachep, flags); } @@ -3372,10 +3360,8 @@ static inline bool is_dot_dotdot(const u8 *name, size_t len) static inline void *f2fs_kmalloc(struct f2fs_sb_info *sbi, size_t size, gfp_t flags) { - if (time_to_inject(sbi, FAULT_KMALLOC)) { - f2fs_show_injection_info(sbi, FAULT_KMALLOC); + if (time_to_inject(sbi, FAULT_KMALLOC)) return NULL; - } return kmalloc(size, flags); } @@ -3389,10 +3375,8 @@ static inline void *f2fs_kzalloc(struct f2fs_sb_info *sbi, static inline void *f2fs_kvmalloc(struct f2fs_sb_info *sbi, size_t size, gfp_t flags) { - if (time_to_inject(sbi, FAULT_KVMALLOC)) { - f2fs_show_injection_info(sbi, FAULT_KVMALLOC); + if (time_to_inject(sbi, FAULT_KVMALLOC)) return NULL; - } return kvmalloc(size, flags); } diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 58b4200f7fdf0..f5c1b78149540 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -782,10 +782,8 @@ int f2fs_truncate(struct inode *inode) trace_f2fs_truncate(inode); - if (time_to_inject(F2FS_I_SB(inode), FAULT_TRUNCATE)) { - f2fs_show_injection_info(F2FS_I_SB(inode), FAULT_TRUNCATE); + if (time_to_inject(F2FS_I_SB(inode), FAULT_TRUNCATE)) return -EIO; - } err = f2fs_dquot_initialize(inode); if (err) diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index 7444c392eab16..e59c006c10f77 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -72,11 +72,9 @@ static int gc_thread_func(void *data) continue; } - if (time_to_inject(sbi, FAULT_CHECKPOINT)) { - f2fs_show_injection_info(sbi, FAULT_CHECKPOINT); + if (time_to_inject(sbi, FAULT_CHECKPOINT)) f2fs_stop_checkpoint(sbi, false, STOP_CP_REASON_FAULT_INJECT); - } if (!sb_start_write_trylock(sbi->sb)) { stat_other_skip_bggc_count(sbi); diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c index ff6cf66ed46b2..01b9e6f85f6be 100644 --- a/fs/f2fs/inode.c +++ b/fs/f2fs/inode.c @@ -809,10 +809,8 @@ void f2fs_evict_inode(struct inode *inode) if (F2FS_HAS_BLOCKS(inode)) err = f2fs_truncate(inode); - if (time_to_inject(sbi, FAULT_EVICT_INODE)) { - f2fs_show_injection_info(sbi, FAULT_EVICT_INODE); + if (time_to_inject(sbi, FAULT_EVICT_INODE)) err = -EIO; - } if (!err) { f2fs_lock_op(sbi); diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c index 558b318f73167..fbd1d25fecc22 100644 --- a/fs/f2fs/node.c +++ b/fs/f2fs/node.c @@ -2541,10 +2541,8 @@ bool f2fs_alloc_nid(struct f2fs_sb_info *sbi, nid_t *nid) struct f2fs_nm_info *nm_i = NM_I(sbi); struct free_nid *i = NULL; retry: - if (time_to_inject(sbi, FAULT_ALLOC_NID)) { - f2fs_show_injection_info(sbi, FAULT_ALLOC_NID); + if (time_to_inject(sbi, FAULT_ALLOC_NID)) return false; - } spin_lock(&nm_i->nid_list_lock); diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index e2f95f46d2989..ba8331434703c 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -384,10 +384,8 @@ int f2fs_commit_atomic_write(struct inode *inode) */ void f2fs_balance_fs(struct f2fs_sb_info *sbi, bool need) { - if (time_to_inject(sbi, FAULT_CHECKPOINT)) { - f2fs_show_injection_info(sbi, FAULT_CHECKPOINT); + if (time_to_inject(sbi, FAULT_CHECKPOINT)) f2fs_stop_checkpoint(sbi, false, STOP_CP_REASON_FAULT_INJECT); - } /* balance_fs_bg is able to be pending */ if (need && excess_cached_nats(sbi)) @@ -1140,7 +1138,6 @@ static int __submit_discard_cmd(struct f2fs_sb_info *sbi, dc->len += len; if (time_to_inject(sbi, FAULT_DISCARD)) { - f2fs_show_injection_info(sbi, FAULT_DISCARD); err = -EIO; } else { err = __blkdev_issue_discard(bdev, diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 1d057a4c66424..5fc83771042d3 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -1371,10 +1371,8 @@ static struct inode *f2fs_alloc_inode(struct super_block *sb) { struct f2fs_inode_info *fi; - if (time_to_inject(F2FS_SB(sb), FAULT_SLAB_ALLOC)) { - f2fs_show_injection_info(F2FS_SB(sb), FAULT_SLAB_ALLOC); + if (time_to_inject(F2FS_SB(sb), FAULT_SLAB_ALLOC)) return NULL; - } fi = alloc_inode_sb(sb, f2fs_inode_cachep, GFP_F2FS_ZERO); if (!fi) @@ -2594,10 +2592,8 @@ static ssize_t f2fs_quota_write(struct super_block *sb, int type, int f2fs_dquot_initialize(struct inode *inode) { - if (time_to_inject(F2FS_I_SB(inode), FAULT_DQUOT_INIT)) { - f2fs_show_injection_info(F2FS_I_SB(inode), FAULT_DQUOT_INIT); + if (time_to_inject(F2FS_I_SB(inode), FAULT_DQUOT_INIT)) return -ESRCH; - } return dquot_initialize(inode); } From f08142bc3a60f5af632ba48dc2fef8797043086d Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Sat, 17 Dec 2022 13:24:48 +0800 Subject: [PATCH 037/765] f2fs: convert to use MIN_DISCARD_GRANULARITY macro Commit 1cd2e6d54435 ("f2fs: define MIN_DISCARD_GRANULARITY macro") introduce it, let's convert to use MIN_DISCARD_GRANULARITY macro. Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index ba8331434703c..f8e0ccfa0f522 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -1604,9 +1604,9 @@ static unsigned int __wait_all_discard_cmd(struct f2fs_sb_info *sbi, return __wait_discard_cmd_range(sbi, dpolicy, 0, UINT_MAX); /* wait all */ - __init_discard_policy(sbi, &dp, DPOLICY_FSTRIM, 1); + __init_discard_policy(sbi, &dp, DPOLICY_FSTRIM, MIN_DISCARD_GRANULARITY); discard_blks = __wait_discard_cmd_range(sbi, &dp, 0, UINT_MAX); - __init_discard_policy(sbi, &dp, DPOLICY_UMOUNT, 1); + __init_discard_policy(sbi, &dp, DPOLICY_UMOUNT, MIN_DISCARD_GRANULARITY); discard_blks += __wait_discard_cmd_range(sbi, &dp, 0, UINT_MAX); return discard_blks; @@ -1689,7 +1689,8 @@ static int issue_discard_thread(void *data) if (sbi->gc_mode == GC_URGENT_HIGH || !f2fs_available_free_memory(sbi, DISCARD_CACHE)) - __init_discard_policy(sbi, &dpolicy, DPOLICY_FORCE, 1); + __init_discard_policy(sbi, &dpolicy, DPOLICY_FORCE, + MIN_DISCARD_GRANULARITY); else __init_discard_policy(sbi, &dpolicy, DPOLICY_BG, dcc->discard_granularity); From 45c98f5a58f36c35ecf5a149cbf69cf5fd022120 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Mon, 12 Dec 2022 21:36:44 +0800 Subject: [PATCH 038/765] f2fs: convert discard_wake and gc_wake to bool type discard_wake and gc_wake have only two values, 0 or 1. So there is no need to use int type to store them. BTW, move discard_wake to the end of the discard_cmd_control structure. Before: - sizeof(struct discard_cmd_control): 8392 After move: - sizeof(struct discard_cmd_control): 8384 Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/f2fs.h | 2 +- fs/f2fs/gc.c | 4 ++-- fs/f2fs/gc.h | 2 +- fs/f2fs/segment.c | 2 +- fs/f2fs/segment.h | 2 +- fs/f2fs/sysfs.c | 4 ++-- 6 files changed, 8 insertions(+), 8 deletions(-) diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 53d7ca96df637..c9c6ae966ba84 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -402,7 +402,6 @@ struct discard_cmd_control { struct list_head wait_list; /* store on-flushing entries */ struct list_head fstrim_list; /* in-flight discard from fstrim */ wait_queue_head_t discard_wait_queue; /* waiting queue for wake-up */ - unsigned int discard_wake; /* to wake up discard thread */ struct mutex cmd_lock; unsigned int nr_discards; /* # of discards in the list */ unsigned int max_discards; /* max. discards to be issued */ @@ -420,6 +419,7 @@ struct discard_cmd_control { atomic_t discard_cmd_cnt; /* # of cached cmd count */ struct rb_root_cached root; /* root of discard rb-tree */ bool rbtree_check; /* config for consistence check */ + bool discard_wake; /* to wake up discard thread */ }; /* for the list of fsync inodes, used only during recovery */ diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index e59c006c10f77..97e846263c7c0 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -57,7 +57,7 @@ static int gc_thread_func(void *data) /* give it a try one time */ if (gc_th->gc_wake) - gc_th->gc_wake = 0; + gc_th->gc_wake = false; if (try_to_freeze()) { stat_other_skip_bggc_count(sbi); @@ -183,7 +183,7 @@ int f2fs_start_gc_thread(struct f2fs_sb_info *sbi) gc_th->max_sleep_time = DEF_GC_THREAD_MAX_SLEEP_TIME; gc_th->no_gc_sleep_time = DEF_GC_THREAD_NOGC_SLEEP_TIME; - gc_th->gc_wake = 0; + gc_th->gc_wake = false; sbi->gc_thread = gc_th; init_waitqueue_head(&sbi->gc_thread->gc_wait_queue_head); diff --git a/fs/f2fs/gc.h b/fs/f2fs/gc.h index 19b956c2d697a..15bd1d680f677 100644 --- a/fs/f2fs/gc.h +++ b/fs/f2fs/gc.h @@ -41,7 +41,7 @@ struct f2fs_gc_kthread { unsigned int no_gc_sleep_time; /* for changing gc mode */ - unsigned int gc_wake; + bool gc_wake; /* for GC_MERGE mount option */ wait_queue_head_t fggc_wq; /* diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index f8e0ccfa0f522..8aafa1f32eccf 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -1696,7 +1696,7 @@ static int issue_discard_thread(void *data) dcc->discard_granularity); if (dcc->discard_wake) - dcc->discard_wake = 0; + dcc->discard_wake = false; /* clean up pending candidates before going to sleep */ if (atomic_read(&dcc->queued_discard)) diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h index e77518c49f388..ad6a9c19f46a4 100644 --- a/fs/f2fs/segment.h +++ b/fs/f2fs/segment.h @@ -939,6 +939,6 @@ static inline void wake_up_discard_thread(struct f2fs_sb_info *sbi, bool force) if (!wakeup || !is_idle(sbi, DISCARD_TIME)) return; wake_up: - dcc->discard_wake = 1; + dcc->discard_wake = true; wake_up_interruptible_all(&dcc->discard_wait_queue); } diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c index 83a366f3ee80e..805b632a3af07 100644 --- a/fs/f2fs/sysfs.c +++ b/fs/f2fs/sysfs.c @@ -511,7 +511,7 @@ static ssize_t __sbi_store(struct f2fs_attr *a, } else if (t == 1) { sbi->gc_mode = GC_URGENT_HIGH; if (sbi->gc_thread) { - sbi->gc_thread->gc_wake = 1; + sbi->gc_thread->gc_wake = true; wake_up_interruptible_all( &sbi->gc_thread->gc_wait_queue_head); wake_up_discard_thread(sbi, true); @@ -521,7 +521,7 @@ static ssize_t __sbi_store(struct f2fs_attr *a, } else if (t == 3) { sbi->gc_mode = GC_URGENT_MID; if (sbi->gc_thread) { - sbi->gc_thread->gc_wake = 1; + sbi->gc_thread->gc_wake = true; wake_up_interruptible_all( &sbi->gc_thread->gc_wait_queue_head); } From 7a2b15cfa8dbbd54beb4e2ce7b2f42eb0ad00425 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Thu, 22 Dec 2022 03:19:32 +0800 Subject: [PATCH 039/765] f2fs: support accounting iostat count and avg_bytes Previously, we supported to account iostat io_bytes, in this patch, it adds to account iostat count and avg_bytes: time: 1671648667 io_bytes count avg_bytes [WRITE] app buffered data: 31 2 15 Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/f2fs.h | 7 +- fs/f2fs/iostat.c | 129 ++++++++++++++++++------------------ fs/f2fs/segment.c | 2 +- include/trace/events/f2fs.h | 2 +- 4 files changed, 69 insertions(+), 71 deletions(-) diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index c9c6ae966ba84..d4729f8af2478 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -1191,7 +1191,7 @@ enum iostat_type { FS_META_READ_IO, /* meta read IOs */ /* other */ - FS_DISCARD, /* discard */ + FS_DISCARD_IO, /* discard */ NR_IO_TYPE, }; @@ -1854,8 +1854,9 @@ struct f2fs_sb_info { #ifdef CONFIG_F2FS_IOSTAT /* For app/fs IO statistics */ spinlock_t iostat_lock; - unsigned long long rw_iostat[NR_IO_TYPE]; - unsigned long long prev_rw_iostat[NR_IO_TYPE]; + unsigned long long iostat_count[NR_IO_TYPE]; + unsigned long long iostat_bytes[NR_IO_TYPE]; + unsigned long long prev_iostat_bytes[NR_IO_TYPE]; bool iostat_enable; unsigned long iostat_next_period; unsigned int iostat_period_ms; diff --git a/fs/f2fs/iostat.c b/fs/f2fs/iostat.c index 3166a8939ed4f..acf834c772912 100644 --- a/fs/f2fs/iostat.c +++ b/fs/f2fs/iostat.c @@ -18,79 +18,68 @@ static struct kmem_cache *bio_iostat_ctx_cache; static mempool_t *bio_iostat_ctx_pool; +static inline unsigned long long iostat_get_avg_bytes(struct f2fs_sb_info *sbi, + enum iostat_type type) +{ + return sbi->iostat_count[type] ? div64_u64(sbi->iostat_bytes[type], + sbi->iostat_count[type]) : 0; +} + +#define IOSTAT_INFO_SHOW(name, type) \ + seq_printf(seq, "%-23s %-16llu %-16llu %-16llu\n", \ + name":", sbi->iostat_bytes[type], \ + sbi->iostat_count[type], \ + iostat_get_avg_bytes(sbi, type)) + int __maybe_unused iostat_info_seq_show(struct seq_file *seq, void *offset) { struct super_block *sb = seq->private; struct f2fs_sb_info *sbi = F2FS_SB(sb); - time64_t now = ktime_get_real_seconds(); if (!sbi->iostat_enable) return 0; - seq_printf(seq, "time: %-16llu\n", now); + seq_printf(seq, "time: %-16llu\n", ktime_get_real_seconds()); + seq_printf(seq, "\t\t\t%-16s %-16s %-16s\n", + "io_bytes", "count", "avg_bytes"); /* print app write IOs */ seq_puts(seq, "[WRITE]\n"); - seq_printf(seq, "app buffered data: %-16llu\n", - sbi->rw_iostat[APP_BUFFERED_IO]); - seq_printf(seq, "app direct data: %-16llu\n", - sbi->rw_iostat[APP_DIRECT_IO]); - seq_printf(seq, "app mapped data: %-16llu\n", - sbi->rw_iostat[APP_MAPPED_IO]); - seq_printf(seq, "app buffered cdata: %-16llu\n", - sbi->rw_iostat[APP_BUFFERED_CDATA_IO]); - seq_printf(seq, "app mapped cdata: %-16llu\n", - sbi->rw_iostat[APP_MAPPED_CDATA_IO]); + IOSTAT_INFO_SHOW("app buffered data", APP_BUFFERED_IO); + IOSTAT_INFO_SHOW("app direct data", APP_DIRECT_IO); + IOSTAT_INFO_SHOW("app mapped data", APP_MAPPED_IO); + IOSTAT_INFO_SHOW("app buffered cdata", APP_BUFFERED_CDATA_IO); + IOSTAT_INFO_SHOW("app mapped cdata", APP_MAPPED_CDATA_IO); /* print fs write IOs */ - seq_printf(seq, "fs data: %-16llu\n", - sbi->rw_iostat[FS_DATA_IO]); - seq_printf(seq, "fs cdata: %-16llu\n", - sbi->rw_iostat[FS_CDATA_IO]); - seq_printf(seq, "fs node: %-16llu\n", - sbi->rw_iostat[FS_NODE_IO]); - seq_printf(seq, "fs meta: %-16llu\n", - sbi->rw_iostat[FS_META_IO]); - seq_printf(seq, "fs gc data: %-16llu\n", - sbi->rw_iostat[FS_GC_DATA_IO]); - seq_printf(seq, "fs gc node: %-16llu\n", - sbi->rw_iostat[FS_GC_NODE_IO]); - seq_printf(seq, "fs cp data: %-16llu\n", - sbi->rw_iostat[FS_CP_DATA_IO]); - seq_printf(seq, "fs cp node: %-16llu\n", - sbi->rw_iostat[FS_CP_NODE_IO]); - seq_printf(seq, "fs cp meta: %-16llu\n", - sbi->rw_iostat[FS_CP_META_IO]); + IOSTAT_INFO_SHOW("fs data", FS_DATA_IO); + IOSTAT_INFO_SHOW("fs cdata", FS_CDATA_IO); + IOSTAT_INFO_SHOW("fs node", FS_NODE_IO); + IOSTAT_INFO_SHOW("fs meta", FS_META_IO); + IOSTAT_INFO_SHOW("fs gc data", FS_GC_DATA_IO); + IOSTAT_INFO_SHOW("fs gc node", FS_GC_NODE_IO); + IOSTAT_INFO_SHOW("fs cp data", FS_CP_DATA_IO); + IOSTAT_INFO_SHOW("fs cp node", FS_CP_NODE_IO); + IOSTAT_INFO_SHOW("fs cp meta", FS_CP_META_IO); /* print app read IOs */ seq_puts(seq, "[READ]\n"); - seq_printf(seq, "app buffered data: %-16llu\n", - sbi->rw_iostat[APP_BUFFERED_READ_IO]); - seq_printf(seq, "app direct data: %-16llu\n", - sbi->rw_iostat[APP_DIRECT_READ_IO]); - seq_printf(seq, "app mapped data: %-16llu\n", - sbi->rw_iostat[APP_MAPPED_READ_IO]); - seq_printf(seq, "app buffered cdata: %-16llu\n", - sbi->rw_iostat[APP_BUFFERED_CDATA_READ_IO]); - seq_printf(seq, "app mapped cdata: %-16llu\n", - sbi->rw_iostat[APP_MAPPED_CDATA_READ_IO]); + IOSTAT_INFO_SHOW("app buffered data", APP_BUFFERED_READ_IO); + IOSTAT_INFO_SHOW("app direct data", APP_DIRECT_READ_IO); + IOSTAT_INFO_SHOW("app mapped data", APP_MAPPED_READ_IO); + IOSTAT_INFO_SHOW("app buffered cdata", APP_BUFFERED_CDATA_READ_IO); + IOSTAT_INFO_SHOW("app mapped cdata", APP_MAPPED_CDATA_READ_IO); /* print fs read IOs */ - seq_printf(seq, "fs data: %-16llu\n", - sbi->rw_iostat[FS_DATA_READ_IO]); - seq_printf(seq, "fs gc data: %-16llu\n", - sbi->rw_iostat[FS_GDATA_READ_IO]); - seq_printf(seq, "fs cdata: %-16llu\n", - sbi->rw_iostat[FS_CDATA_READ_IO]); - seq_printf(seq, "fs node: %-16llu\n", - sbi->rw_iostat[FS_NODE_READ_IO]); - seq_printf(seq, "fs meta: %-16llu\n", - sbi->rw_iostat[FS_META_READ_IO]); + IOSTAT_INFO_SHOW("fs data", FS_DATA_READ_IO); + IOSTAT_INFO_SHOW("fs gc data", FS_GDATA_READ_IO); + IOSTAT_INFO_SHOW("fs cdata", FS_CDATA_READ_IO); + IOSTAT_INFO_SHOW("fs node", FS_NODE_READ_IO); + IOSTAT_INFO_SHOW("fs meta", FS_META_READ_IO); /* print other IOs */ seq_puts(seq, "[OTHER]\n"); - seq_printf(seq, "fs discard: %-16llu\n", - sbi->rw_iostat[FS_DISCARD]); + IOSTAT_INFO_SHOW("fs discard", FS_DISCARD_IO); return 0; } @@ -141,9 +130,9 @@ static inline void f2fs_record_iostat(struct f2fs_sb_info *sbi) msecs_to_jiffies(sbi->iostat_period_ms); for (i = 0; i < NR_IO_TYPE; i++) { - iostat_diff[i] = sbi->rw_iostat[i] - - sbi->prev_rw_iostat[i]; - sbi->prev_rw_iostat[i] = sbi->rw_iostat[i]; + iostat_diff[i] = sbi->iostat_bytes[i] - + sbi->prev_iostat_bytes[i]; + sbi->prev_iostat_bytes[i] = sbi->iostat_bytes[i]; } spin_unlock_irqrestore(&sbi->iostat_lock, flags); @@ -159,8 +148,9 @@ void f2fs_reset_iostat(struct f2fs_sb_info *sbi) spin_lock_irq(&sbi->iostat_lock); for (i = 0; i < NR_IO_TYPE; i++) { - sbi->rw_iostat[i] = 0; - sbi->prev_rw_iostat[i] = 0; + sbi->iostat_count[i] = 0; + sbi->iostat_bytes[i] = 0; + sbi->prev_iostat_bytes[i] = 0; } spin_unlock_irq(&sbi->iostat_lock); @@ -169,6 +159,13 @@ void f2fs_reset_iostat(struct f2fs_sb_info *sbi) spin_unlock_irq(&sbi->iostat_lat_lock); } +static inline void __f2fs_update_iostat(struct f2fs_sb_info *sbi, + enum iostat_type type, unsigned long long io_bytes) +{ + sbi->iostat_bytes[type] += io_bytes; + sbi->iostat_count[type]++; +} + void f2fs_update_iostat(struct f2fs_sb_info *sbi, struct inode *inode, enum iostat_type type, unsigned long long io_bytes) { @@ -178,33 +175,33 @@ void f2fs_update_iostat(struct f2fs_sb_info *sbi, struct inode *inode, return; spin_lock_irqsave(&sbi->iostat_lock, flags); - sbi->rw_iostat[type] += io_bytes; + __f2fs_update_iostat(sbi, type, io_bytes); if (type == APP_BUFFERED_IO || type == APP_DIRECT_IO) - sbi->rw_iostat[APP_WRITE_IO] += io_bytes; + __f2fs_update_iostat(sbi, APP_WRITE_IO, io_bytes); if (type == APP_BUFFERED_READ_IO || type == APP_DIRECT_READ_IO) - sbi->rw_iostat[APP_READ_IO] += io_bytes; + __f2fs_update_iostat(sbi, APP_READ_IO, io_bytes); #ifdef CONFIG_F2FS_FS_COMPRESSION if (inode && f2fs_compressed_file(inode)) { if (type == APP_BUFFERED_IO) - sbi->rw_iostat[APP_BUFFERED_CDATA_IO] += io_bytes; + __f2fs_update_iostat(sbi, APP_BUFFERED_CDATA_IO, io_bytes); if (type == APP_BUFFERED_READ_IO) - sbi->rw_iostat[APP_BUFFERED_CDATA_READ_IO] += io_bytes; + __f2fs_update_iostat(sbi, APP_BUFFERED_CDATA_READ_IO, io_bytes); if (type == APP_MAPPED_READ_IO) - sbi->rw_iostat[APP_MAPPED_CDATA_READ_IO] += io_bytes; + __f2fs_update_iostat(sbi, APP_MAPPED_CDATA_READ_IO, io_bytes); if (type == APP_MAPPED_IO) - sbi->rw_iostat[APP_MAPPED_CDATA_IO] += io_bytes; + __f2fs_update_iostat(sbi, APP_MAPPED_CDATA_IO, io_bytes); if (type == FS_DATA_READ_IO) - sbi->rw_iostat[FS_CDATA_READ_IO] += io_bytes; + __f2fs_update_iostat(sbi, FS_CDATA_READ_IO, io_bytes); if (type == FS_DATA_IO) - sbi->rw_iostat[FS_CDATA_IO] += io_bytes; + __f2fs_update_iostat(sbi, FS_CDATA_IO, io_bytes); } #endif diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 8aafa1f32eccf..311243dda4cef 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -1182,7 +1182,7 @@ static int __submit_discard_cmd(struct f2fs_sb_info *sbi, atomic_inc(&dcc->issued_discard); - f2fs_update_iostat(sbi, NULL, FS_DISCARD, len * F2FS_BLKSIZE); + f2fs_update_iostat(sbi, NULL, FS_DISCARD_IO, len * F2FS_BLKSIZE); lstart += len; start += len; diff --git a/include/trace/events/f2fs.h b/include/trace/events/f2fs.h index 9183a0a11e26f..3852085198fb5 100644 --- a/include/trace/events/f2fs.h +++ b/include/trace/events/f2fs.h @@ -1972,7 +1972,7 @@ TRACE_EVENT(f2fs_iostat, __entry->fs_cdrio = iostat[FS_CDATA_READ_IO]; __entry->fs_nrio = iostat[FS_NODE_READ_IO]; __entry->fs_mrio = iostat[FS_META_READ_IO]; - __entry->fs_discard = iostat[FS_DISCARD]; + __entry->fs_discard = iostat[FS_DISCARD_IO]; ), TP_printk("dev = (%d,%d), " From 03d7a1053cf72372be22b43faada5bca12ff183d Mon Sep 17 00:00:00 2001 From: Michal Kubecek Date: Wed, 18 Jan 2023 11:52:15 +0100 Subject: [PATCH 040/765] objtool: Check that module init/exit function is an indirect call target Some out-of-tree modules still do not use module_init() / module_exit() macros and simply create functions with magic names init_module() and cleanup_module() instead. As a result, these functions are not recognized as indirect call targets by objtool and such module fails to load into an IBT enabled kernel. This old way is not even documented any more but it is cleaner to issue a warning than to let the module fail on load without obvious reason. Signed-off-by: Michal Kubecek Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20230118105215.B9DA960514@lion.mk-sys.cz --- tools/objtool/Documentation/objtool.txt | 8 ++++++++ tools/objtool/check.c | 7 +++++++ 2 files changed, 15 insertions(+) diff --git a/tools/objtool/Documentation/objtool.txt b/tools/objtool/Documentation/objtool.txt index 8a671902a1875..8e53fc6735ef2 100644 --- a/tools/objtool/Documentation/objtool.txt +++ b/tools/objtool/Documentation/objtool.txt @@ -410,6 +410,14 @@ the objtool maintainers. can remove this warning by putting the ANNOTATE_INTRA_FUNCTION_CALL directive right before the call. +12. file.o: warning: func(): not an indirect call target + + This means that objtool is running with --ibt and a function expected + to be an indirect call target is not. In particular, this happens for + init_module() or cleanup_module() if a module relies on these special + names and does not use module_init() / module_exit() macros to create + them. + If the error doesn't seem to make sense, it could be a bug in objtool. Feel free to ask the objtool maintainer for help. diff --git a/tools/objtool/check.c b/tools/objtool/check.c index cab1a162781c4..7c40bd51c75a5 100644 --- a/tools/objtool/check.c +++ b/tools/objtool/check.c @@ -847,8 +847,15 @@ static int create_ibt_endbr_seal_sections(struct objtool_file *file) list_for_each_entry(insn, &file->endbr_list, call_node) { int *site = (int *)sec->data->d_buf + idx; + struct symbol *sym = insn->sym; *site = 0; + if (opts.module && sym && sym->type == STT_FUNC && + insn->offset == sym->offset && + (!strcmp(sym->name, "init_module") || + !strcmp(sym->name, "cleanup_module"))) + WARN("%s(): not an indirect call target", sym->name); + if (elf_add_reloc_to_insn(file->elf, sec, idx * sizeof(int), R_X86_64_PC32, From 90226f6b17a3edcb0bddaf2f16991861c99d6a15 Mon Sep 17 00:00:00 2001 From: Doug Berger Date: Fri, 20 Jan 2023 11:01:42 -0800 Subject: [PATCH 041/765] rtc: brcmstb-waketimer: introduce WKTMR_ALARM_EVENT flag This commit defines bit 0 as the bit of interest within the BRCMSTB_WKTMR_EVENT register to make the implementation more readable. Signed-off-by: Doug Berger Acked-by: Florian Fainelli Link: https://lore.kernel.org/r/20230120190147.718976-2-opendmb@gmail.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-brcmstb-waketimer.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/drivers/rtc/rtc-brcmstb-waketimer.c b/drivers/rtc/rtc-brcmstb-waketimer.c index c74130e8f496d..fbeb8be6664b5 100644 --- a/drivers/rtc/rtc-brcmstb-waketimer.c +++ b/drivers/rtc/rtc-brcmstb-waketimer.c @@ -34,6 +34,7 @@ struct brcmstb_waketmr { }; #define BRCMSTB_WKTMR_EVENT 0x00 +#define WKTMR_ALARM_EVENT BIT(0) #define BRCMSTB_WKTMR_COUNTER 0x04 #define BRCMSTB_WKTMR_ALARM 0x08 #define BRCMSTB_WKTMR_PRESCALER 0x0C @@ -41,9 +42,17 @@ struct brcmstb_waketmr { #define BRCMSTB_WKTMR_DEFAULT_FREQ 27000000 +static inline bool brcmstb_waketmr_is_pending(struct brcmstb_waketmr *timer) +{ + u32 reg; + + reg = readl_relaxed(timer->base + BRCMSTB_WKTMR_EVENT); + return !!(reg & WKTMR_ALARM_EVENT); +} + static inline void brcmstb_waketmr_clear_alarm(struct brcmstb_waketmr *timer) { - writel_relaxed(1, timer->base + BRCMSTB_WKTMR_EVENT); + writel_relaxed(WKTMR_ALARM_EVENT, timer->base + BRCMSTB_WKTMR_EVENT); (void)readl_relaxed(timer->base + BRCMSTB_WKTMR_EVENT); } @@ -147,7 +156,6 @@ static int brcmstb_waketmr_getalarm(struct device *dev, { struct brcmstb_waketmr *timer = dev_get_drvdata(dev); time64_t sec; - u32 reg; sec = readl_relaxed(timer->base + BRCMSTB_WKTMR_ALARM); if (sec != 0) { @@ -156,8 +164,7 @@ static int brcmstb_waketmr_getalarm(struct device *dev, rtc_time64_to_tm(sec, &alarm->time); } - reg = readl_relaxed(timer->base + BRCMSTB_WKTMR_EVENT); - alarm->pending = !!(reg & 1); + alarm->pending = brcmstb_waketmr_is_pending(timer); return 0; } From 2cd98b22c1443d1f2921a371baee658da184868e Mon Sep 17 00:00:00 2001 From: Doug Berger Date: Fri, 20 Jan 2023 11:01:43 -0800 Subject: [PATCH 042/765] rtc: brcmstb-waketimer: non-functional code changes These changes are not intended to affect functionality, but simplify the source code. They are performed here to simplify review and reduce confusion with other changes in this set. Since set_alarm includes the alarm_irq_enable functionality call it directly from that function for simplicity (even though it does nothing at the moment). The order of the declarations is changed to prevent the need for a prototype. The function device_init_wakeup() is used to replace the functions device_set_wakeup_capable() and device_wakeup_enable() since it is equivalent. Signed-off-by: Doug Berger Acked-by: Florian Fainelli Link: https://lore.kernel.org/r/20230120190147.718976-3-opendmb@gmail.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-brcmstb-waketimer.c | 25 ++++++++++++------------- 1 file changed, 12 insertions(+), 13 deletions(-) diff --git a/drivers/rtc/rtc-brcmstb-waketimer.c b/drivers/rtc/rtc-brcmstb-waketimer.c index fbeb8be6664b5..582c23793550d 100644 --- a/drivers/rtc/rtc-brcmstb-waketimer.c +++ b/drivers/rtc/rtc-brcmstb-waketimer.c @@ -169,6 +169,16 @@ static int brcmstb_waketmr_getalarm(struct device *dev, return 0; } +/* + * Does not do much but keep the RTC class happy. We always support + * alarms. + */ +static int brcmstb_waketmr_alarm_enable(struct device *dev, + unsigned int enabled) +{ + return 0; +} + static int brcmstb_waketmr_setalarm(struct device *dev, struct rtc_wkalrm *alarm) { @@ -182,17 +192,7 @@ static int brcmstb_waketmr_setalarm(struct device *dev, brcmstb_waketmr_set_alarm(timer, sec); - return 0; -} - -/* - * Does not do much but keep the RTC class happy. We always support - * alarms. - */ -static int brcmstb_waketmr_alarm_enable(struct device *dev, - unsigned int enabled) -{ - return 0; + return brcmstb_waketmr_alarm_enable(dev, alarm->enabled); } static const struct rtc_class_ops brcmstb_waketmr_ops = { @@ -228,8 +228,7 @@ static int brcmstb_waketmr_probe(struct platform_device *pdev) * Set wakeup capability before requesting wakeup interrupt, so we can * process boot-time "wakeups" (e.g., from S5 soft-off) */ - device_set_wakeup_capable(dev, true); - device_wakeup_enable(dev); + device_init_wakeup(dev, true); timer->irq = platform_get_irq(pdev, 0); if (timer->irq < 0) From 516ae02c38ff3ae867f9b19fa050f78157e2bdae Mon Sep 17 00:00:00 2001 From: Doug Berger Date: Fri, 20 Jan 2023 11:01:44 -0800 Subject: [PATCH 043/765] rtc: brcmstb-waketimer: compensate for lack of wktmr disable Since the WKTMR hardware block cannot be disabled it is necessary for the driver to accommodate for associated timing hazards. This commit targets the following possibilities: A possible race between clearing a wktmr event and the alarm expiring is made one-sided by setting the alarm to its maximum value before clearing the event. Programming alarm values close to the current time may not trigger events if the counter advances while the alarm is being programmed. After programming an alarm, a check is made to ensure that it is either in the future or an expiration event is pending. Signed-off-by: Doug Berger Acked-by: Florian Fainelli Link: https://lore.kernel.org/r/20230120190147.718976-4-opendmb@gmail.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-brcmstb-waketimer.c | 52 +++++++++++++++++++---------- 1 file changed, 34 insertions(+), 18 deletions(-) diff --git a/drivers/rtc/rtc-brcmstb-waketimer.c b/drivers/rtc/rtc-brcmstb-waketimer.c index 582c23793550d..c791e92532b8e 100644 --- a/drivers/rtc/rtc-brcmstb-waketimer.c +++ b/drivers/rtc/rtc-brcmstb-waketimer.c @@ -31,6 +31,8 @@ struct brcmstb_waketmr { struct notifier_block reboot_notifier; struct clk *clk; u32 rate; + unsigned long rtc_alarm; + bool alarm_en; }; #define BRCMSTB_WKTMR_EVENT 0x00 @@ -52,6 +54,11 @@ static inline bool brcmstb_waketmr_is_pending(struct brcmstb_waketmr *timer) static inline void brcmstb_waketmr_clear_alarm(struct brcmstb_waketmr *timer) { + u32 reg; + + timer->alarm_en = false; + reg = readl_relaxed(timer->base + BRCMSTB_WKTMR_COUNTER); + writel_relaxed(reg - 1, timer->base + BRCMSTB_WKTMR_ALARM); writel_relaxed(WKTMR_ALARM_EVENT, timer->base + BRCMSTB_WKTMR_EVENT); (void)readl_relaxed(timer->base + BRCMSTB_WKTMR_EVENT); } @@ -59,12 +66,22 @@ static inline void brcmstb_waketmr_clear_alarm(struct brcmstb_waketmr *timer) static void brcmstb_waketmr_set_alarm(struct brcmstb_waketmr *timer, unsigned int secs) { + unsigned int now; + brcmstb_waketmr_clear_alarm(timer); /* Make sure we are actually counting in seconds */ writel_relaxed(timer->rate, timer->base + BRCMSTB_WKTMR_PRESCALER); - writel_relaxed(secs + 1, timer->base + BRCMSTB_WKTMR_ALARM); + writel_relaxed(secs, timer->base + BRCMSTB_WKTMR_ALARM); + now = readl_relaxed(timer->base + BRCMSTB_WKTMR_COUNTER); + + while ((int)(secs - now) <= 0 && + !brcmstb_waketmr_is_pending(timer)) { + secs = now + 1; + writel_relaxed(secs, timer->base + BRCMSTB_WKTMR_ALARM); + now = readl_relaxed(timer->base + BRCMSTB_WKTMR_COUNTER); + } } static irqreturn_t brcmstb_waketmr_irq(int irq, void *data) @@ -155,27 +172,30 @@ static int brcmstb_waketmr_getalarm(struct device *dev, struct rtc_wkalrm *alarm) { struct brcmstb_waketmr *timer = dev_get_drvdata(dev); - time64_t sec; - sec = readl_relaxed(timer->base + BRCMSTB_WKTMR_ALARM); - if (sec != 0) { - /* Alarm is enabled */ - alarm->enabled = 1; - rtc_time64_to_tm(sec, &alarm->time); - } + alarm->enabled = timer->alarm_en; + rtc_time64_to_tm(timer->rtc_alarm, &alarm->time); alarm->pending = brcmstb_waketmr_is_pending(timer); return 0; } -/* - * Does not do much but keep the RTC class happy. We always support - * alarms. - */ static int brcmstb_waketmr_alarm_enable(struct device *dev, unsigned int enabled) { + struct brcmstb_waketmr *timer = dev_get_drvdata(dev); + + if (enabled && !timer->alarm_en) { + if ((int)(readl_relaxed(timer->base + BRCMSTB_WKTMR_COUNTER) - + readl_relaxed(timer->base + BRCMSTB_WKTMR_ALARM)) >= 0 && + !brcmstb_waketmr_is_pending(timer)) + return -EINVAL; + timer->alarm_en = true; + } else if (!enabled && timer->alarm_en) { + timer->alarm_en = false; + } + return 0; } @@ -183,14 +203,10 @@ static int brcmstb_waketmr_setalarm(struct device *dev, struct rtc_wkalrm *alarm) { struct brcmstb_waketmr *timer = dev_get_drvdata(dev); - time64_t sec; - if (alarm->enabled) - sec = rtc_tm_to_time64(&alarm->time); - else - sec = 0; + timer->rtc_alarm = rtc_tm_to_time64(&alarm->time); - brcmstb_waketmr_set_alarm(timer, sec); + brcmstb_waketmr_set_alarm(timer, timer->rtc_alarm); return brcmstb_waketmr_alarm_enable(dev, alarm->enabled); } From eae258edcb8705932c9e5c61a99f91d8235f688b Mon Sep 17 00:00:00 2001 From: Doug Berger Date: Fri, 20 Jan 2023 11:01:45 -0800 Subject: [PATCH 044/765] rtc: brcmstb-waketimer: rename irq to wake_irq In preparation for adding a second interrupt to service RTC interrupts, the existing interrupt is renamed from the generic 'irq' to 'wake_irq' to more clearly convey its role. It is also converted to an unsigned int. Finally, the driver message that outputs the IRQ number when registered is removed since devm_rtc_register_device() already provides a report of registration and the interrupts can be found in /proc/interrupts. Signed-off-by: Doug Berger Acked-by: Florian Fainelli Link: https://lore.kernel.org/r/20230120190147.718976-5-opendmb@gmail.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-brcmstb-waketimer.c | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) diff --git a/drivers/rtc/rtc-brcmstb-waketimer.c b/drivers/rtc/rtc-brcmstb-waketimer.c index c791e92532b8e..e25f9fcd6ed1c 100644 --- a/drivers/rtc/rtc-brcmstb-waketimer.c +++ b/drivers/rtc/rtc-brcmstb-waketimer.c @@ -27,7 +27,7 @@ struct brcmstb_waketmr { struct rtc_device *rtc; struct device *dev; void __iomem *base; - int irq; + unsigned int wake_irq; struct notifier_block reboot_notifier; struct clk *clk; u32 rate; @@ -117,7 +117,7 @@ static int brcmstb_waketmr_prepare_suspend(struct brcmstb_waketmr *timer) int ret = 0; if (device_may_wakeup(dev)) { - ret = enable_irq_wake(timer->irq); + ret = enable_irq_wake(timer->wake_irq); if (ret) { dev_err(dev, "failed to enable wake-up interrupt\n"); return ret; @@ -246,9 +246,10 @@ static int brcmstb_waketmr_probe(struct platform_device *pdev) */ device_init_wakeup(dev, true); - timer->irq = platform_get_irq(pdev, 0); - if (timer->irq < 0) + ret = platform_get_irq(pdev, 0); + if (ret < 0) return -ENODEV; + timer->wake_irq = (unsigned int)ret; timer->clk = devm_clk_get(dev, NULL); if (!IS_ERR(timer->clk)) { @@ -263,7 +264,7 @@ static int brcmstb_waketmr_probe(struct platform_device *pdev) timer->clk = NULL; } - ret = devm_request_irq(dev, timer->irq, brcmstb_waketmr_irq, 0, + ret = devm_request_irq(dev, timer->wake_irq, brcmstb_waketmr_irq, 0, "brcmstb-waketimer", timer); if (ret < 0) goto err_clk; @@ -278,8 +279,6 @@ static int brcmstb_waketmr_probe(struct platform_device *pdev) if (ret) goto err_notifier; - dev_info(dev, "registered, with irq %d\n", timer->irq); - return 0; err_notifier: @@ -317,7 +316,7 @@ static int brcmstb_waketmr_resume(struct device *dev) if (!device_may_wakeup(dev)) return 0; - ret = disable_irq_wake(timer->irq); + ret = disable_irq_wake(timer->wake_irq); brcmstb_waketmr_clear_alarm(timer); From 473a8ce756fdbd6e9fc0ffbf3ceb88394058763e Mon Sep 17 00:00:00 2001 From: WANG Xuerui Date: Mon, 9 Jan 2023 09:35:11 +0800 Subject: [PATCH 045/765] dt-bindings: rtc: Add Loongson LS2X RTC support Document the binding for the LS2X RTC block found on the Loongson-2K SoC and the LS7A bridge, originally appearing on the Loongson-2H. Signed-off-by: WANG Xuerui Signed-off-by: Binbin Zhou Acked-by: Rob Herring Link: https://lore.kernel.org/r/0288efeb4209e4a49af07de6399045aaa00a970c.1673227292.git.zhoubinbin@loongson.cn Signed-off-by: Alexandre Belloni --- Documentation/devicetree/bindings/rtc/trivial-rtc.yaml | 2 ++ 1 file changed, 2 insertions(+) diff --git a/Documentation/devicetree/bindings/rtc/trivial-rtc.yaml b/Documentation/devicetree/bindings/rtc/trivial-rtc.yaml index d9fc120c61cc3..0ef1ffd54655c 100644 --- a/Documentation/devicetree/bindings/rtc/trivial-rtc.yaml +++ b/Documentation/devicetree/bindings/rtc/trivial-rtc.yaml @@ -47,6 +47,8 @@ properties: - isil,isl1218 # Intersil ISL12022 Real-time Clock - isil,isl12022 + # Loongson-2K Socs/LS7A bridge Real-time Clock + - loongson,ls2x-rtc # Real Time Clock Module with I2C-Bus - microcrystal,rv3028 # Real Time Clock Module with I2C-Bus From c690048ed59b5f7df91899507f210ba6591c0d8f Mon Sep 17 00:00:00 2001 From: Wadim Egorov Date: Wed, 11 Jan 2023 17:34:04 +0100 Subject: [PATCH 046/765] dt-bindings: rtc: Move rv3028 from trivial-rtc.yaml into own schema file Move RV3028 RTC bindings from trivial-rtc.yaml into microcrystal,rv3032.yaml. Signed-off-by: Wadim Egorov Reviewed-by: Rob Herring Link: https://lore.kernel.org/r/20230111163404.3526248-2-w.egorov@phytec.de Signed-off-by: Alexandre Belloni --- .../bindings/rtc/microcrystal,rv3028.yaml | 54 +++++++++++++++++++ .../devicetree/bindings/rtc/trivial-rtc.yaml | 2 - 2 files changed, 54 insertions(+), 2 deletions(-) create mode 100644 Documentation/devicetree/bindings/rtc/microcrystal,rv3028.yaml diff --git a/Documentation/devicetree/bindings/rtc/microcrystal,rv3028.yaml b/Documentation/devicetree/bindings/rtc/microcrystal,rv3028.yaml new file mode 100644 index 0000000000000..5ade5dfad048a --- /dev/null +++ b/Documentation/devicetree/bindings/rtc/microcrystal,rv3028.yaml @@ -0,0 +1,54 @@ +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/rtc/microcrystal,rv3028.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Microchip RV-3028 RTC + +allOf: + - $ref: rtc.yaml# + +maintainers: + - Alexandre Belloni + +properties: + compatible: + const: microcrystal,rv3028 + + reg: + maxItems: 1 + + interrupts: + maxItems: 1 + + trickle-resistor-ohms: + enum: + - 3000 + - 5000 + - 9000 + - 15000 + +required: + - compatible + - reg + +unevaluatedProperties: false + +examples: + - | + #include + i2c { + #address-cells = <1>; + #size-cells = <0>; + + rtc@51 { + compatible = "microcrystal,rv3028"; + reg = <0x51>; + pinctrl-0 = <&rtc_nint_pins>; + interrupts-extended = <&gpio1 16 IRQ_TYPE_LEVEL_HIGH>; + trickle-resistor-ohms = <3000>; + }; + }; + +... diff --git a/Documentation/devicetree/bindings/rtc/trivial-rtc.yaml b/Documentation/devicetree/bindings/rtc/trivial-rtc.yaml index 0ef1ffd54655c..38ac5f6b2f6c4 100644 --- a/Documentation/devicetree/bindings/rtc/trivial-rtc.yaml +++ b/Documentation/devicetree/bindings/rtc/trivial-rtc.yaml @@ -50,8 +50,6 @@ properties: # Loongson-2K Socs/LS7A bridge Real-time Clock - loongson,ls2x-rtc # Real Time Clock Module with I2C-Bus - - microcrystal,rv3028 - # Real Time Clock Module with I2C-Bus - microcrystal,rv3029 # Real Time Clock - microcrystal,rv8523 From 344f4030f6c50a9db2d03021884c4bf36191b53a Mon Sep 17 00:00:00 2001 From: Samuel Holland Date: Thu, 29 Dec 2022 15:53:19 -0600 Subject: [PATCH 047/765] rtc: sun6i: Always export the internal oscillator On all variants of the hardware, the internal oscillator is one possible parent for the AR100 clock. It needs to be exported so we can model that relationship correctly in the devicetree. Fixes: c56afc1844d6 ("rtc: sun6i: Expose internal oscillator through device tree") Signed-off-by: Samuel Holland Acked-by: Jernej Skrabec Link: https://lore.kernel.org/r/20221229215319.14145-1-samuel@sholland.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-sun6i.c | 16 ++++------------ 1 file changed, 4 insertions(+), 12 deletions(-) diff --git a/drivers/rtc/rtc-sun6i.c b/drivers/rtc/rtc-sun6i.c index ed5516089e9a0..7038f47d77ff4 100644 --- a/drivers/rtc/rtc-sun6i.c +++ b/drivers/rtc/rtc-sun6i.c @@ -136,7 +136,6 @@ struct sun6i_rtc_clk_data { unsigned int fixed_prescaler : 16; unsigned int has_prescaler : 1; unsigned int has_out_clk : 1; - unsigned int export_iosc : 1; unsigned int has_losc_en : 1; unsigned int has_auto_swt : 1; }; @@ -271,10 +270,8 @@ static void __init sun6i_rtc_clk_init(struct device_node *node, /* Yes, I know, this is ugly. */ sun6i_rtc = rtc; - /* Only read IOSC name from device tree if it is exported */ - if (rtc->data->export_iosc) - of_property_read_string_index(node, "clock-output-names", 2, - &iosc_name); + of_property_read_string_index(node, "clock-output-names", 2, + &iosc_name); rtc->int_osc = clk_hw_register_fixed_rate_with_accuracy(NULL, iosc_name, @@ -315,13 +312,10 @@ static void __init sun6i_rtc_clk_init(struct device_node *node, goto err_register; } - clk_data->num = 2; + clk_data->num = 3; clk_data->hws[0] = &rtc->hw; clk_data->hws[1] = __clk_get_hw(rtc->ext_losc); - if (rtc->data->export_iosc) { - clk_data->hws[2] = rtc->int_osc; - clk_data->num = 3; - } + clk_data->hws[2] = rtc->int_osc; of_clk_add_hw_provider(node, of_clk_hw_onecell_get, clk_data); return; @@ -361,7 +355,6 @@ static const struct sun6i_rtc_clk_data sun8i_h3_rtc_data = { .fixed_prescaler = 32, .has_prescaler = 1, .has_out_clk = 1, - .export_iosc = 1, }; static void __init sun8i_h3_rtc_clk_init(struct device_node *node) @@ -379,7 +372,6 @@ static const struct sun6i_rtc_clk_data sun50i_h6_rtc_data = { .fixed_prescaler = 32, .has_prescaler = 1, .has_out_clk = 1, - .export_iosc = 1, .has_losc_en = 1, .has_auto_swt = 1, }; From 6cc7a8262b57db0a2ff6c141bb163c1395b7c4ef Mon Sep 17 00:00:00 2001 From: Doug Berger Date: Tue, 24 Jan 2023 12:14:29 -0800 Subject: [PATCH 048/765] dt-bindings: rtc: brcm,brcmstb-waketimer: add alarm interrupt A second interrupt can optionally be specified for this device to be used for generating RTC alarm interrupts. Signed-off-by: Doug Berger Acked-by: Florian Fainelli Reviewed-by: Krzysztof Kozlowski Link: https://lore.kernel.org/r/20230124201430.2502371-2-opendmb@gmail.com Signed-off-by: Alexandre Belloni --- .../bindings/rtc/brcm,brcmstb-waketimer.yaml | 21 ++++++++++++------- 1 file changed, 14 insertions(+), 7 deletions(-) diff --git a/Documentation/devicetree/bindings/rtc/brcm,brcmstb-waketimer.yaml b/Documentation/devicetree/bindings/rtc/brcm,brcmstb-waketimer.yaml index 9fe079917a986..c6c57636c729e 100644 --- a/Documentation/devicetree/bindings/rtc/brcm,brcmstb-waketimer.yaml +++ b/Documentation/devicetree/bindings/rtc/brcm,brcmstb-waketimer.yaml @@ -11,7 +11,8 @@ maintainers: description: The Broadcom STB wake-up timer provides a 27Mhz resolution timer, with the - ability to wake up the system from low-power suspend/standby modes. + ability to wake up the system from low-power suspend/standby modes and + optionally generate RTC alarm interrupts. allOf: - $ref: "rtc.yaml#" @@ -24,8 +25,14 @@ properties: maxItems: 1 interrupts: - description: the TIMER interrupt - maxItems: 1 + minItems: 1 + items: + - description: the TIMER interrupt + - description: the ALARM interrupt + description: + The TIMER interrupt wakes the system from low-power suspend/standby modes. + An ALARM interrupt may be specified to interrupt the CPU when an RTC alarm + is enabled. clocks: description: clock reference in the 27MHz domain @@ -35,10 +42,10 @@ additionalProperties: false examples: - | - rtc@f0411580 { + rtc@f041a080 { compatible = "brcm,brcmstb-waketimer"; - reg = <0xf0411580 0x14>; - interrupts = <0x3>; - interrupt-parent = <&aon_pm_l2_intc>; + reg = <0xf041a080 0x14>; + interrupts-extended = <&aon_pm_l2_intc 0x04>, + <&upg_aux_aon_intr2_intc 0x08>; clocks = <&upg_fixed>; }; From 24304a87158aab01b4ccb9b2638c2c623a9a7bd4 Mon Sep 17 00:00:00 2001 From: Doug Berger Date: Tue, 24 Jan 2023 12:14:30 -0800 Subject: [PATCH 049/765] rtc: brcmstb-waketimer: allow use as non-wake alarm The wake interrupt only fires when the system is in a suspend state. Fortunately we have another interrupt that fires in a non-suspend state at the L2 controller UPG_AUX_AON. Add support for this interrupt line so we can use the alarm in a non-wake context. Signed-off-by: Doug Berger Acked-by: Florian Fainelli Link: https://lore.kernel.org/r/20230124201430.2502371-3-opendmb@gmail.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-brcmstb-waketimer.c | 55 +++++++++++++++++++++++++++-- 1 file changed, 52 insertions(+), 3 deletions(-) diff --git a/drivers/rtc/rtc-brcmstb-waketimer.c b/drivers/rtc/rtc-brcmstb-waketimer.c index e25f9fcd6ed1c..1efa81cecc273 100644 --- a/drivers/rtc/rtc-brcmstb-waketimer.c +++ b/drivers/rtc/rtc-brcmstb-waketimer.c @@ -28,6 +28,7 @@ struct brcmstb_waketmr { struct device *dev; void __iomem *base; unsigned int wake_irq; + unsigned int alarm_irq; struct notifier_block reboot_notifier; struct clk *clk; u32 rate; @@ -56,6 +57,8 @@ static inline void brcmstb_waketmr_clear_alarm(struct brcmstb_waketmr *timer) { u32 reg; + if (timer->alarm_en && timer->alarm_irq) + disable_irq(timer->alarm_irq); timer->alarm_en = false; reg = readl_relaxed(timer->base + BRCMSTB_WKTMR_COUNTER); writel_relaxed(reg - 1, timer->base + BRCMSTB_WKTMR_ALARM); @@ -88,7 +91,25 @@ static irqreturn_t brcmstb_waketmr_irq(int irq, void *data) { struct brcmstb_waketmr *timer = data; - pm_wakeup_event(timer->dev, 0); + if (!timer->alarm_irq) + pm_wakeup_event(timer->dev, 0); + return IRQ_HANDLED; +} + +static irqreturn_t brcmstb_alarm_irq(int irq, void *data) +{ + struct brcmstb_waketmr *timer = data; + + /* Ignore spurious interrupts */ + if (!brcmstb_waketmr_is_pending(timer)) + return IRQ_HANDLED; + + if (timer->alarm_en) { + if (!device_may_wakeup(timer->dev)) + writel_relaxed(WKTMR_ALARM_EVENT, + timer->base + BRCMSTB_WKTMR_EVENT); + rtc_update_irq(timer->rtc, 1, RTC_IRQF | RTC_AF); + } return IRQ_HANDLED; } @@ -114,7 +135,7 @@ static void wktmr_read(struct brcmstb_waketmr *timer, static int brcmstb_waketmr_prepare_suspend(struct brcmstb_waketmr *timer) { struct device *dev = timer->dev; - int ret = 0; + int ret; if (device_may_wakeup(dev)) { ret = enable_irq_wake(timer->wake_irq); @@ -122,9 +143,17 @@ static int brcmstb_waketmr_prepare_suspend(struct brcmstb_waketmr *timer) dev_err(dev, "failed to enable wake-up interrupt\n"); return ret; } + if (timer->alarm_en && timer->alarm_irq) { + ret = enable_irq_wake(timer->alarm_irq); + if (ret) { + dev_err(dev, "failed to enable rtc interrupt\n"); + disable_irq_wake(timer->wake_irq); + return ret; + } + } } - return ret; + return 0; } /* If enabled as a wakeup-source, arm the timer when powering off */ @@ -192,7 +221,11 @@ static int brcmstb_waketmr_alarm_enable(struct device *dev, !brcmstb_waketmr_is_pending(timer)) return -EINVAL; timer->alarm_en = true; + if (timer->alarm_irq) + enable_irq(timer->alarm_irq); } else if (!enabled && timer->alarm_en) { + if (timer->alarm_irq) + disable_irq(timer->alarm_irq); timer->alarm_en = false; } @@ -269,6 +302,19 @@ static int brcmstb_waketmr_probe(struct platform_device *pdev) if (ret < 0) goto err_clk; + brcmstb_waketmr_clear_alarm(timer); + + /* Attempt to initialize non-wake irq */ + ret = platform_get_irq(pdev, 1); + if (ret > 0) { + timer->alarm_irq = (unsigned int)ret; + ret = devm_request_irq(dev, timer->alarm_irq, brcmstb_alarm_irq, + IRQF_NO_AUTOEN, "brcmstb-waketimer-rtc", + timer); + if (ret < 0) + timer->alarm_irq = 0; + } + timer->reboot_notifier.notifier_call = brcmstb_waketmr_reboot; register_reboot_notifier(&timer->reboot_notifier); @@ -317,6 +363,8 @@ static int brcmstb_waketmr_resume(struct device *dev) return 0; ret = disable_irq_wake(timer->wake_irq); + if (timer->alarm_en && timer->alarm_irq) + disable_irq_wake(timer->alarm_irq); brcmstb_waketmr_clear_alarm(timer); @@ -346,4 +394,5 @@ module_platform_driver(brcmstb_waketmr_driver); MODULE_LICENSE("GPL v2"); MODULE_AUTHOR("Brian Norris"); MODULE_AUTHOR("Markus Mayer"); +MODULE_AUTHOR("Doug Berger"); MODULE_DESCRIPTION("Wake-up timer driver for STB chips"); From 947e8876c0426971c34c422569c29ec02d711bb3 Mon Sep 17 00:00:00 2001 From: Hugo Villeneuve Date: Tue, 20 Dec 2022 10:22:37 -0500 Subject: [PATCH 050/765] dt-bindings: rtc: pcf2127: remove pca/pcf2129 from trivial RTC devices list pca/pcf2129 devices can also have the 'reset-source' property, so remove them from the trivial RTC devices list. Signed-off-by: Hugo Villeneuve Reviewed-by: Bruno Thomsen Reviewed-by: Rob Herring Acked-by: Krzysztof Kozlowski Link: https://lore.kernel.org/r/20221220152237.1125178-1-hugo@hugovil.com Signed-off-by: Alexandre Belloni --- Documentation/devicetree/bindings/rtc/nxp,pcf2127.yaml | 5 ++++- Documentation/devicetree/bindings/rtc/trivial-rtc.yaml | 2 -- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/Documentation/devicetree/bindings/rtc/nxp,pcf2127.yaml b/Documentation/devicetree/bindings/rtc/nxp,pcf2127.yaml index cde7b1675ead4..a1148eb22c245 100644 --- a/Documentation/devicetree/bindings/rtc/nxp,pcf2127.yaml +++ b/Documentation/devicetree/bindings/rtc/nxp,pcf2127.yaml @@ -14,7 +14,10 @@ maintainers: properties: compatible: - const: nxp,pcf2127 + enum: + - nxp,pca2129 + - nxp,pcf2127 + - nxp,pcf2129 reg: maxItems: 1 diff --git a/Documentation/devicetree/bindings/rtc/trivial-rtc.yaml b/Documentation/devicetree/bindings/rtc/trivial-rtc.yaml index 38ac5f6b2f6c4..eb75861c28c32 100644 --- a/Documentation/devicetree/bindings/rtc/trivial-rtc.yaml +++ b/Documentation/devicetree/bindings/rtc/trivial-rtc.yaml @@ -53,8 +53,6 @@ properties: - microcrystal,rv3029 # Real Time Clock - microcrystal,rv8523 - - nxp,pca2129 - - nxp,pcf2129 # Real-time Clock Module - pericom,pt7c4338 # I2C bus SERIAL INTERFACE REAL-TIME CLOCK IC From 334c7b13d38321e47d1a51dba0bef9f4c403ec75 Mon Sep 17 00:00:00 2001 From: Emil Renner Berthing Date: Wed, 9 Nov 2022 12:37:24 +0100 Subject: [PATCH 051/765] pwm: sifive: Always let the first pwm_apply_state succeed Commit 2cfe9bbec56ea579135cdd92409fff371841904f added support for the RGB and green PWM controlled LEDs on the HiFive Unmatched board managed by the leds-pwm-multicolor and leds-pwm drivers respectively. All three colours of the RGB LED and the green LED run from different lines of the same PWM, but with the same period so this works fine when the LED drivers are loaded one after the other. Unfortunately it does expose a race in the PWM driver when both LED drivers are loaded at roughly the same time. Here is an example: | Thread A | Thread B | | led_pwm_mc_probe | led_pwm_probe | | devm_fwnode_pwm_get | | | pwm_sifive_request | | | ddata->user_count++ | | | | devm_fwnode_pwm_get | | | pwm_sifive_request | | | ddata->user_count++ | | ... | ... | | pwm_state_apply | pwm_state_apply | | pwm_sifive_apply | pwm_sifive_apply | Now both calls to pwm_sifive_apply will see that ddata->approx_period, initially 0, is different from the requested period and the clock needs to be updated. But since ddata->user_count >= 2 both calls will fail with -EBUSY, which will then cause both LED drivers to fail to probe. Fix it by letting the first call to pwm_sifive_apply update the clock even when ddata->user_count != 1. Fixes: 9e37a53eb051 ("pwm: sifive: Add a driver for SiFive SoC PWM") Signed-off-by: Emil Renner Berthing Signed-off-by: Thierry Reding --- drivers/pwm/pwm-sifive.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/pwm/pwm-sifive.c b/drivers/pwm/pwm-sifive.c index 62b6acc6373db..393a4b97fc19e 100644 --- a/drivers/pwm/pwm-sifive.c +++ b/drivers/pwm/pwm-sifive.c @@ -161,7 +161,13 @@ static int pwm_sifive_apply(struct pwm_chip *chip, struct pwm_device *pwm, mutex_lock(&ddata->lock); if (state->period != ddata->approx_period) { - if (ddata->user_count != 1) { + /* + * Don't let a 2nd user change the period underneath the 1st user. + * However if ddate->approx_period == 0 this is the first time we set + * any period, so let whoever gets here first set the period so other + * users who agree on the period won't fail. + */ + if (ddata->user_count != 1 && ddata->approx_period) { mutex_unlock(&ddata->lock); return -EBUSY; } From b3c650ad9bb88ecf36b9aeacf9e7eb7478258da7 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 21 Nov 2022 17:15:41 +0100 Subject: [PATCH 052/765] pwm: Move pwm_capture() dummy to restore order MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Move the dummy pwm_capture(), to make the declaration order of all dummies to match the declaration order of the real functions. Signed-off-by: Geert Uytterhoeven Acked-by: Uwe Kleine-König Signed-off-by: Thierry Reding --- include/linux/pwm.h | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/include/linux/pwm.h b/include/linux/pwm.h index 161e91167b9c0..7b7b93b6fb81a 100644 --- a/include/linux/pwm.h +++ b/include/linux/pwm.h @@ -440,13 +440,6 @@ static inline int pwm_config(struct pwm_device *pwm, int duty_ns, return -EINVAL; } -static inline int pwm_capture(struct pwm_device *pwm, - struct pwm_capture *result, - unsigned long timeout) -{ - return -EINVAL; -} - static inline int pwm_enable(struct pwm_device *pwm) { might_sleep(); @@ -458,6 +451,13 @@ static inline void pwm_disable(struct pwm_device *pwm) might_sleep(); } +static inline int pwm_capture(struct pwm_device *pwm, + struct pwm_capture *result, + unsigned long timeout) +{ + return -EINVAL; +} + static inline int pwm_set_chip_data(struct pwm_device *pwm, void *data) { return -EINVAL; From 3066bc2d58be31275afb51a589668f265e419c37 Mon Sep 17 00:00:00 2001 From: Fabrice Gasnier Date: Wed, 23 Nov 2022 14:36:52 +0100 Subject: [PATCH 053/765] pwm: stm32-lp: fix the check on arr and cmp registers update MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The ARR (auto reload register) and CMP (compare) registers are successively written. The status bits to check the update of these registers are polled together with regmap_read_poll_timeout(). The condition to end the loop may become true, even if one of the register isn't correctly updated. So ensure both status bits are set before clearing them. Fixes: e70a540b4e02 ("pwm: Add STM32 LPTimer PWM driver") Signed-off-by: Fabrice Gasnier Acked-by: Uwe Kleine-König Signed-off-by: Thierry Reding --- drivers/pwm/pwm-stm32-lp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pwm/pwm-stm32-lp.c b/drivers/pwm/pwm-stm32-lp.c index 514ff58a4471d..f315fa106be87 100644 --- a/drivers/pwm/pwm-stm32-lp.c +++ b/drivers/pwm/pwm-stm32-lp.c @@ -127,7 +127,7 @@ static int stm32_pwm_lp_apply(struct pwm_chip *chip, struct pwm_device *pwm, /* ensure CMP & ARR registers are properly written */ ret = regmap_read_poll_timeout(priv->regmap, STM32_LPTIM_ISR, val, - (val & STM32_LPTIM_CMPOK_ARROK), + (val & STM32_LPTIM_CMPOK_ARROK) == STM32_LPTIM_CMPOK_ARROK, 100, 1000); if (ret) { dev_err(priv->chip.dev, "ARR/CMP registers write issue\n"); From 3e98855ca0cf823330b27f51be41e92fdbaa9057 Mon Sep 17 00:00:00 2001 From: AngeloGioacchino Del Regno Date: Mon, 28 Nov 2022 12:20:28 +0100 Subject: [PATCH 054/765] dt-bindings: pwm: mediatek: Convert pwm-mediatek to DT schema This converts pwm-mediatek.txt to mediatek,mt2712-pwm.yaml schema; while at it, the clock names were clarified as previously they were documented as "pwmX-Y", but valid names are "pwmN" only. Also, the example was changed to use "mediatek,mt2712-pwm" instead for consistency with the schema filename. Signed-off-by: AngeloGioacchino Del Regno Reviewed-by: Krzysztof Kozlowski Signed-off-by: Thierry Reding --- .../bindings/pwm/mediatek,mt2712-pwm.yaml | 93 +++++++++++++++++++ .../devicetree/bindings/pwm/pwm-mediatek.txt | 52 ----------- 2 files changed, 93 insertions(+), 52 deletions(-) create mode 100644 Documentation/devicetree/bindings/pwm/mediatek,mt2712-pwm.yaml delete mode 100644 Documentation/devicetree/bindings/pwm/pwm-mediatek.txt diff --git a/Documentation/devicetree/bindings/pwm/mediatek,mt2712-pwm.yaml b/Documentation/devicetree/bindings/pwm/mediatek,mt2712-pwm.yaml new file mode 100644 index 0000000000000..dbc974bff9e91 --- /dev/null +++ b/Documentation/devicetree/bindings/pwm/mediatek,mt2712-pwm.yaml @@ -0,0 +1,93 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/pwm/mediatek,mt2712-pwm.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: MediaTek PWM Controller + +maintainers: + - John Crispin + +allOf: + - $ref: pwm.yaml# + +properties: + compatible: + oneOf: + - enum: + - mediatek,mt2712-pwm + - mediatek,mt6795-pwm + - mediatek,mt7622-pwm + - mediatek,mt7623-pwm + - mediatek,mt7628-pwm + - mediatek,mt7629-pwm + - mediatek,mt8183-pwm + - mediatek,mt8365-pwm + - mediatek,mt8516-pwm + - items: + - enum: + - mediatek,mt8195-pwm + - const: mediatek,mt8183-pwm + + reg: + maxItems: 1 + + "#pwm-cells": + const: 2 + + interrupts: + maxItems: 1 + + clocks: + minItems: 2 + maxItems: 10 + + clock-names: + description: + This controller needs two input clocks for its core and one + clock for each PWM output. + minItems: 2 + items: + - const: top + - const: main + - const: pwm1 + - const: pwm2 + - const: pwm3 + - const: pwm4 + - const: pwm5 + - const: pwm6 + - const: pwm7 + - const: pwm8 + +required: + - compatible + - reg + - "#pwm-cells" + - clocks + - clock-names + +additionalProperties: false + +examples: + - | + #include + #include + #include + + pwm0: pwm@11006000 { + compatible = "mediatek,mt2712-pwm"; + reg = <0x11006000 0x1000>; + #pwm-cells = <2>; + interrupts = ; + clocks = <&topckgen CLK_TOP_PWM_SEL>, <&pericfg CLK_PERI_PWM>, + <&pericfg CLK_PERI_PWM0>, <&pericfg CLK_PERI_PWM1>, + <&pericfg CLK_PERI_PWM2>, <&pericfg CLK_PERI_PWM3>, + <&pericfg CLK_PERI_PWM4>, <&pericfg CLK_PERI_PWM5>, + <&pericfg CLK_PERI_PWM6>, <&pericfg CLK_PERI_PWM7>; + clock-names = "top", "main", + "pwm1", "pwm2", + "pwm3", "pwm4", + "pwm5", "pwm6", + "pwm7", "pwm8"; + }; diff --git a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt b/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt deleted file mode 100644 index 554c96b6d0c3e..0000000000000 --- a/Documentation/devicetree/bindings/pwm/pwm-mediatek.txt +++ /dev/null @@ -1,52 +0,0 @@ -MediaTek PWM controller - -Required properties: - - compatible: should be "mediatek,-pwm": - - "mediatek,mt2712-pwm": found on mt2712 SoC. - - "mediatek,mt6795-pwm": found on mt6795 SoC. - - "mediatek,mt7622-pwm": found on mt7622 SoC. - - "mediatek,mt7623-pwm": found on mt7623 SoC. - - "mediatek,mt7628-pwm": found on mt7628 SoC. - - "mediatek,mt7629-pwm": found on mt7629 SoC. - - "mediatek,mt8183-pwm": found on mt8183 SoC. - - "mediatek,mt8195-pwm", "mediatek,mt8183-pwm": found on mt8195 SoC. - - "mediatek,mt8365-pwm": found on mt8365 SoC. - - "mediatek,mt8516-pwm": found on mt8516 SoC. - - reg: physical base address and length of the controller's registers. - - #pwm-cells: must be 2. See pwm.yaml in this directory for a description of - the cell format. - - clocks: phandle and clock specifier of the PWM reference clock. - - clock-names: must contain the following, except for MT7628 which - has no clocks - - "top": the top clock generator - - "main": clock used by the PWM core - - "pwm1-3": the three per PWM clocks for mt8365 - - "pwm1-8": the eight per PWM clocks for mt2712 - - "pwm1-6": the six per PWM clocks for mt7622 - - "pwm1-5": the five per PWM clocks for mt7623 - - "pwm1" : the PWM1 clock for mt7629 - - pinctrl-names: Must contain a "default" entry. - - pinctrl-0: One property must exist for each entry in pinctrl-names. - See pinctrl/pinctrl-bindings.txt for details of the property values. - -Optional properties: -- assigned-clocks: Reference to the PWM clock entries. -- assigned-clock-parents: The phandle of the parent clock of PWM clock. - -Example: - pwm0: pwm@11006000 { - compatible = "mediatek,mt7623-pwm"; - reg = <0 0x11006000 0 0x1000>; - #pwm-cells = <2>; - clocks = <&topckgen CLK_TOP_PWM_SEL>, - <&pericfg CLK_PERI_PWM>, - <&pericfg CLK_PERI_PWM1>, - <&pericfg CLK_PERI_PWM2>, - <&pericfg CLK_PERI_PWM3>, - <&pericfg CLK_PERI_PWM4>, - <&pericfg CLK_PERI_PWM5>; - clock-names = "top", "main", "pwm1", "pwm2", - "pwm3", "pwm4", "pwm5"; - pinctrl-names = "default"; - pinctrl-0 = <&pwm0_pins>; - }; From 2781f8e9203685208c9f3717593601d4b4674372 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Mon, 19 Dec 2022 09:10:08 +0100 Subject: [PATCH 055/765] pwm: lp3943: Drop unused i2c include MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit lp3943 is a platform driver that doesn't use any symbol provided in . So drop the include. Signed-off-by: Uwe Kleine-König Signed-off-by: Thierry Reding --- drivers/pwm/pwm-lp3943.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/pwm/pwm-lp3943.c b/drivers/pwm/pwm-lp3943.c index 215ef90691144..35675e4058c6b 100644 --- a/drivers/pwm/pwm-lp3943.c +++ b/drivers/pwm/pwm-lp3943.c @@ -8,7 +8,6 @@ */ #include -#include #include #include #include From 193a639fed92793ad10e327cfb2be7175be01425 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Thu, 22 Dec 2022 03:20:01 +0800 Subject: [PATCH 056/765] f2fs: add iostat support for flush In this patch, it adds to account flush count. Signed-off-by: Yangtao Li Signed-off-by: Jaegeuk Kim --- fs/f2fs/f2fs.h | 1 + fs/f2fs/iostat.c | 1 + fs/f2fs/segment.c | 2 ++ 3 files changed, 4 insertions(+) diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index d4729f8af2478..331c330ea31dc 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -1192,6 +1192,7 @@ enum iostat_type { /* other */ FS_DISCARD_IO, /* discard */ + FS_FLUSH_IO, /* flush */ NR_IO_TYPE, }; diff --git a/fs/f2fs/iostat.c b/fs/f2fs/iostat.c index acf834c772912..91b384bea5ac5 100644 --- a/fs/f2fs/iostat.c +++ b/fs/f2fs/iostat.c @@ -80,6 +80,7 @@ int __maybe_unused iostat_info_seq_show(struct seq_file *seq, void *offset) /* print other IOs */ seq_puts(seq, "[OTHER]\n"); IOSTAT_INFO_SHOW("fs discard", FS_DISCARD_IO); + IOSTAT_INFO_SHOW("fs flush", FS_FLUSH_IO); return 0; } diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 311243dda4cef..976316218bd3f 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -506,6 +506,8 @@ static int __submit_flush_wait(struct f2fs_sb_info *sbi, trace_f2fs_issue_flush(bdev, test_opt(sbi, NOBARRIER), test_opt(sbi, FLUSH_MERGE), ret); + if (!ret) + f2fs_update_iostat(sbi, NULL, FS_FLUSH_IO, 0); return ret; } From c5f9db2548d0ac988df81aa34cc63f3b2ed374f9 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Thu, 22 Dec 2022 16:18:55 +0800 Subject: [PATCH 057/765] f2fs: drop useless initializer and unneeded local variable No need to initialize idx twice. BTW, remove the unnecessary cnt variable. Signed-off-by: Yangtao Li Signed-off-by: Jaegeuk Kim --- fs/f2fs/iostat.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/fs/f2fs/iostat.c b/fs/f2fs/iostat.c index 91b384bea5ac5..ed8176939aa57 100644 --- a/fs/f2fs/iostat.c +++ b/fs/f2fs/iostat.c @@ -87,8 +87,7 @@ int __maybe_unused iostat_info_seq_show(struct seq_file *seq, void *offset) static inline void __record_iostat_latency(struct f2fs_sb_info *sbi) { - int io, idx = 0; - unsigned int cnt; + int io, idx; struct f2fs_iostat_latency iostat_lat[MAX_IO_TYPE][NR_PAGE_TYPE]; struct iostat_lat_info *io_lat = sbi->iostat_io_lat; unsigned long flags; @@ -96,12 +95,11 @@ static inline void __record_iostat_latency(struct f2fs_sb_info *sbi) spin_lock_irqsave(&sbi->iostat_lat_lock, flags); for (idx = 0; idx < MAX_IO_TYPE; idx++) { for (io = 0; io < NR_PAGE_TYPE; io++) { - cnt = io_lat->bio_cnt[idx][io]; iostat_lat[idx][io].peak_lat = jiffies_to_msecs(io_lat->peak_lat[idx][io]); - iostat_lat[idx][io].cnt = cnt; - iostat_lat[idx][io].avg_lat = cnt ? - jiffies_to_msecs(io_lat->sum_lat[idx][io]) / cnt : 0; + iostat_lat[idx][io].cnt = io_lat->bio_cnt[idx][io]; + iostat_lat[idx][io].avg_lat = iostat_lat[idx][io].cnt ? + jiffies_to_msecs(io_lat->sum_lat[idx][io]) / iostat_lat[idx][io].cnt : 0; io_lat->sum_lat[idx][io] = 0; io_lat->peak_lat[idx][io] = 0; io_lat->bio_cnt[idx][io] = 0; From 120e0ea12d90ad15127ad50944975fc8c81666b7 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Wed, 4 Jan 2023 19:40:29 +0800 Subject: [PATCH 058/765] f2fs: introduce discard_io_aware_gran sysfs node The current discard_io_aware_gran is a fixed value, change it to be configurable through the sys node. Signed-off-by: Yangtao Li Signed-off-by: Jaegeuk Kim --- Documentation/ABI/testing/sysfs-fs-f2fs | 9 +++++++++ fs/f2fs/f2fs.h | 1 + fs/f2fs/segment.c | 3 ++- fs/f2fs/sysfs.c | 13 +++++++++++++ 4 files changed, 25 insertions(+), 1 deletion(-) diff --git a/Documentation/ABI/testing/sysfs-fs-f2fs b/Documentation/ABI/testing/sysfs-fs-f2fs index aaa379bb8a8fc..75420c242cc44 100644 --- a/Documentation/ABI/testing/sysfs-fs-f2fs +++ b/Documentation/ABI/testing/sysfs-fs-f2fs @@ -708,3 +708,12 @@ Description: Support configuring fault injection type, should be FAULT_LOCK_OP 0x000020000 FAULT_BLKADDR 0x000040000 =================== =========== + +What: /sys/fs/f2fs//discard_io_aware_gran +Date: January 2023 +Contact: "Yangtao Li" +Description: Controls background discard granularity of inner discard thread + when is not in idle. Inner thread will not issue discards with size that + is smaller than granularity. The unit size is one block(4KB), now only + support configuring in range of [0, 512]. + Default: 512 diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 331c330ea31dc..f3c5f7740c1a0 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -409,6 +409,7 @@ struct discard_cmd_control { unsigned int min_discard_issue_time; /* min. interval between discard issue */ unsigned int mid_discard_issue_time; /* mid. interval between discard issue */ unsigned int max_discard_issue_time; /* max. interval between discard issue */ + unsigned int discard_io_aware_gran; /* minimum discard granularity not be aware of I/O */ unsigned int discard_urgent_util; /* utilization which issue discard proactively */ unsigned int discard_granularity; /* discard granularity */ unsigned int max_ordered_discard; /* maximum discard granularity issued by lba order */ diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 976316218bd3f..bd1cd98fa6eb5 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -1059,7 +1059,7 @@ static void __init_discard_policy(struct f2fs_sb_info *sbi, dpolicy->granularity = granularity; dpolicy->max_requests = dcc->max_discard_request; - dpolicy->io_aware_gran = MAX_PLIST_NUM; + dpolicy->io_aware_gran = dcc->discard_io_aware_gran; dpolicy->timeout = false; if (discard_type == DPOLICY_BG) { @@ -2063,6 +2063,7 @@ static int create_discard_cmd_control(struct f2fs_sb_info *sbi) if (!dcc) return -ENOMEM; + dcc->discard_io_aware_gran = MAX_PLIST_NUM; dcc->discard_granularity = DEFAULT_DISCARD_GRANULARITY; dcc->max_ordered_discard = DEFAULT_MAX_ORDERED_DISCARD_GRANULARITY; if (F2FS_OPTION(sbi).discard_unit == DISCARD_UNIT_SEGMENT) diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c index 805b632a3af07..e396851a6dd10 100644 --- a/fs/f2fs/sysfs.c +++ b/fs/f2fs/sysfs.c @@ -473,6 +473,17 @@ static ssize_t __sbi_store(struct f2fs_attr *a, return count; } + if (!strcmp(a->attr.name, "discard_io_aware_gran")) { + if (t > MAX_PLIST_NUM) + return -EINVAL; + if (!f2fs_block_unit_discard(sbi)) + return -EINVAL; + if (t == *ui) + return count; + *ui = t; + return count; + } + if (!strcmp(a->attr.name, "discard_granularity")) { if (t == 0 || t > MAX_PLIST_NUM) return -EINVAL; @@ -825,6 +836,7 @@ F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, max_discard_request, max_discard_req F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, min_discard_issue_time, min_discard_issue_time); F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, mid_discard_issue_time, mid_discard_issue_time); F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, max_discard_issue_time, max_discard_issue_time); +F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, discard_io_aware_gran, discard_io_aware_gran); F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, discard_urgent_util, discard_urgent_util); F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, discard_granularity, discard_granularity); F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, max_ordered_discard, max_ordered_discard); @@ -960,6 +972,7 @@ static struct attribute *f2fs_attrs[] = { ATTR_LIST(min_discard_issue_time), ATTR_LIST(mid_discard_issue_time), ATTR_LIST(max_discard_issue_time), + ATTR_LIST(discard_io_aware_gran), ATTR_LIST(discard_urgent_util), ATTR_LIST(discard_granularity), ATTR_LIST(max_ordered_discard), From 2f3a9ae990a7881c9a57a073bb52ebe34fdc3160 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Mon, 9 Jan 2023 11:44:49 +0800 Subject: [PATCH 059/765] f2fs: introduce trace_f2fs_replace_atomic_write_block Commit 3db1de0e582c ("f2fs: change the current atomic write way") removed old tracepoints, but it missed to add new one, this patch fixes to introduce trace_f2fs_replace_atomic_write_block to trace atomic_write commit flow. Fixes: 3db1de0e582c ("f2fs: change the current atomic write way") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 3 +++ include/trace/events/f2fs.h | 37 +++++++++++++++++++++++++++++++++++++ 2 files changed, 40 insertions(+) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index bd1cd98fa6eb5..ebfa759527ac5 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -255,6 +255,9 @@ static int __replace_atomic_write_block(struct inode *inode, pgoff_t index, } f2fs_put_dnode(&dn); + + trace_f2fs_replace_atomic_write_block(inode, F2FS_I(inode)->cow_inode, + index, *old_addr, new_addr, recover); return 0; } diff --git a/include/trace/events/f2fs.h b/include/trace/events/f2fs.h index 3852085198fb5..fe6bcf5f917dc 100644 --- a/include/trace/events/f2fs.h +++ b/include/trace/events/f2fs.h @@ -1290,6 +1290,43 @@ DEFINE_EVENT(f2fs__page, f2fs_vm_page_mkwrite, TP_ARGS(page, type) ); +TRACE_EVENT(f2fs_replace_atomic_write_block, + + TP_PROTO(struct inode *inode, struct inode *cow_inode, pgoff_t index, + block_t old_addr, block_t new_addr, bool recovery), + + TP_ARGS(inode, cow_inode, index, old_addr, new_addr, recovery), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(ino_t, ino) + __field(ino_t, cow_ino) + __field(pgoff_t, index) + __field(block_t, old_addr) + __field(block_t, new_addr) + __field(bool, recovery) + ), + + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->ino = inode->i_ino; + __entry->cow_ino = cow_inode->i_ino; + __entry->index = index; + __entry->old_addr = old_addr; + __entry->new_addr = new_addr; + __entry->recovery = recovery; + ), + + TP_printk("dev = (%d,%d), ino = %lu, cow_ino = %lu, index = %lu, " + "old_addr = 0x%llx, new_addr = 0x%llx, recovery = %d", + show_dev_ino(__entry), + __entry->cow_ino, + (unsigned long)__entry->index, + (unsigned long long)__entry->old_addr, + (unsigned long long)__entry->new_addr, + __entry->recovery) +); + TRACE_EVENT(f2fs_filemap_fault, TP_PROTO(struct inode *inode, pgoff_t index, unsigned long ret), From 0e8d040bfa4c476d7d2a23119527c744c7de13cd Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Mon, 9 Jan 2023 11:44:50 +0800 Subject: [PATCH 060/765] f2fs: clear atomic_write_task in f2fs_abort_atomic_write() Otherwise, last .atomic_write_task will be remained in structure f2fs_inode_info, resulting in aborting atomic_write accidentally in race case. Meanwhile, clear original_i_size as well. Fixes: 7a10f0177e11 ("f2fs: don't give partially written atomic data from process crash") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index ebfa759527ac5..5e603b808487d 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -201,9 +201,12 @@ void f2fs_abort_atomic_write(struct inode *inode, bool clean) clear_inode_flag(inode, FI_ATOMIC_FILE); stat_dec_atomic_inode(inode); + F2FS_I(inode)->atomic_write_task = NULL; + if (clean) { truncate_inode_pages_final(inode->i_mapping); f2fs_i_size_write(inode, fi->original_i_size); + fi->original_i_size = 0; } } From bdb8bf7d56afd1d22c12c61455d732d3baff2bde Mon Sep 17 00:00:00 2001 From: Ian Rogers Date: Thu, 26 Jan 2023 11:06:04 -0800 Subject: [PATCH 061/765] objtool: Install libsubcmd in build Including from tools/lib can create inadvertent dependencies. Install libsubcmd in the objtool build and then include the headers from there. Signed-off-by: Ian Rogers Link: https://lore.kernel.org/r/20230126190606.40739-2-irogers@google.com Signed-off-by: Josh Poimboeuf --- tools/objtool/.gitignore | 1 + tools/objtool/Build | 2 -- tools/objtool/Makefile | 31 +++++++++++++++++++++++-------- 3 files changed, 24 insertions(+), 10 deletions(-) diff --git a/tools/objtool/.gitignore b/tools/objtool/.gitignore index 14236db3677f6..4faa4dd72f350 100644 --- a/tools/objtool/.gitignore +++ b/tools/objtool/.gitignore @@ -2,3 +2,4 @@ arch/x86/lib/inat-tables.c /objtool fixdep +libsubcmd/ diff --git a/tools/objtool/Build b/tools/objtool/Build index 33f2ee5a46d3b..a3cdf8af6635a 100644 --- a/tools/objtool/Build +++ b/tools/objtool/Build @@ -16,8 +16,6 @@ objtool-y += libctype.o objtool-y += str_error_r.o objtool-y += librbtree.o -CFLAGS += -I$(srctree)/tools/lib - $(OUTPUT)libstring.o: ../lib/string.c FORCE $(call rule_mkdir) $(call if_changed_dep,cc_o_c) diff --git a/tools/objtool/Makefile b/tools/objtool/Makefile index a3a9cc24e0e37..3505ae4b0e364 100644 --- a/tools/objtool/Makefile +++ b/tools/objtool/Makefile @@ -12,9 +12,13 @@ srctree := $(patsubst %/,%,$(dir $(CURDIR))) srctree := $(patsubst %/,%,$(dir $(srctree))) endif -SUBCMD_SRCDIR = $(srctree)/tools/lib/subcmd/ -LIBSUBCMD_OUTPUT = $(or $(OUTPUT),$(CURDIR)/) -LIBSUBCMD = $(LIBSUBCMD_OUTPUT)libsubcmd.a +LIBSUBCMD_DIR = $(srctree)/tools/lib/subcmd/ +ifneq ($(OUTPUT),) + LIBSUBCMD_OUTPUT = $(abspath $(OUTPUT))/libsubcmd +else + LIBSUBCMD_OUTPUT = $(CURDIR)/libsubcmd +endif +LIBSUBCMD = $(LIBSUBCMD_OUTPUT)/libsubcmd.a OBJTOOL := $(OUTPUT)objtool OBJTOOL_IN := $(OBJTOOL)-in.o @@ -28,7 +32,8 @@ INCLUDES := -I$(srctree)/tools/include \ -I$(srctree)/tools/arch/$(HOSTARCH)/include/uapi \ -I$(srctree)/tools/arch/$(SRCARCH)/include \ -I$(srctree)/tools/objtool/include \ - -I$(srctree)/tools/objtool/arch/$(SRCARCH)/include + -I$(srctree)/tools/objtool/arch/$(SRCARCH)/include \ + -I$(LIBSUBCMD_OUTPUT)/include WARNINGS := $(EXTRA_WARNINGS) -Wno-switch-default -Wno-switch-enum -Wno-packed -Wno-nested-externs CFLAGS := -Werror $(WARNINGS) $(KBUILD_HOSTCFLAGS) -g $(INCLUDES) $(LIBELF_FLAGS) LDFLAGS += $(LIBELF_LIBS) $(LIBSUBCMD) $(KBUILD_HOSTLDFLAGS) @@ -38,6 +43,7 @@ elfshdr := $(shell echo '$(pound)include ' | $(CC) $(CFLAGS) -x c -E - CFLAGS += $(if $(elfshdr),,-DLIBELF_USE_DEPRECATED) AWK = awk +MKDIR = mkdir BUILD_ORC := n @@ -57,13 +63,22 @@ $(OBJTOOL): $(LIBSUBCMD) $(OBJTOOL_IN) $(QUIET_LINK)$(CC) $(OBJTOOL_IN) $(LDFLAGS) -o $@ -$(LIBSUBCMD): fixdep FORCE - $(Q)$(MAKE) -C $(SUBCMD_SRCDIR) OUTPUT=$(LIBSUBCMD_OUTPUT) +$(LIBSUBCMD_OUTPUT): + @$(MKDIR) -p $@ + +$(LIBSUBCMD): fixdep $(LIBSUBCMD_OUTPUT) FORCE + @$(MAKE) -C $(LIBSUBCMD_DIR) O=$(LIBSUBCMD_OUTPUT) \ + DESTDIR=$(LIBSUBCMD_OUTPUT) prefix= subdir= \ + $@ install_headers + +$(LIBSUBCMD)-clean: + $(call QUIET_CLEAN, libsubcmd) + $(Q)$(RM) -r -- $(LIBSUBCMD_OUTPUT) -clean: +clean: $(LIBSUBCMD)-clean $(call QUIET_CLEAN, objtool) $(RM) $(OBJTOOL) $(Q)find $(OUTPUT) -name '*.o' -delete -o -name '\.*.cmd' -delete -o -name '\.*.d' -delete - $(Q)$(RM) $(OUTPUT)arch/x86/lib/inat-tables.c $(OUTPUT)fixdep $(LIBSUBCMD) + $(Q)$(RM) $(OUTPUT)arch/x86/lib/inat-tables.c $(OUTPUT)fixdep FORCE: From 8c4526ca6a45e7ff915c2b33b54db6b773291fac Mon Sep 17 00:00:00 2001 From: Ian Rogers Date: Thu, 26 Jan 2023 11:06:05 -0800 Subject: [PATCH 062/765] objtool: Properly support make V=1 The Q variable was being used but never correctly set up. Add the setting up and use in place of @. Signed-off-by: Ian Rogers Link: https://lore.kernel.org/r/20230126190606.40739-3-irogers@google.com Signed-off-by: Josh Poimboeuf --- tools/objtool/Makefile | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/tools/objtool/Makefile b/tools/objtool/Makefile index 3505ae4b0e364..d54b66986627c 100644 --- a/tools/objtool/Makefile +++ b/tools/objtool/Makefile @@ -45,6 +45,12 @@ CFLAGS += $(if $(elfshdr),,-DLIBELF_USE_DEPRECATED) AWK = awk MKDIR = mkdir +ifeq ($(V),1) + Q = +else + Q = @ +endif + BUILD_ORC := n ifeq ($(SRCARCH),x86) @@ -56,18 +62,18 @@ export srctree OUTPUT CFLAGS SRCARCH AWK include $(srctree)/tools/build/Makefile.include $(OBJTOOL_IN): fixdep FORCE - @$(CONFIG_SHELL) ./sync-check.sh - @$(MAKE) $(build)=objtool + $(Q)$(CONFIG_SHELL) ./sync-check.sh + $(Q)$(MAKE) $(build)=objtool $(OBJTOOL): $(LIBSUBCMD) $(OBJTOOL_IN) $(QUIET_LINK)$(CC) $(OBJTOOL_IN) $(LDFLAGS) -o $@ $(LIBSUBCMD_OUTPUT): - @$(MKDIR) -p $@ + $(Q)$(MKDIR) -p $@ $(LIBSUBCMD): fixdep $(LIBSUBCMD_OUTPUT) FORCE - @$(MAKE) -C $(LIBSUBCMD_DIR) O=$(LIBSUBCMD_OUTPUT) \ + $(Q)$(MAKE) -C $(LIBSUBCMD_DIR) O=$(LIBSUBCMD_OUTPUT) \ DESTDIR=$(LIBSUBCMD_OUTPUT) prefix= subdir= \ $@ install_headers From 71a298ca208e48bd69e1bf61a22a762624885093 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Mon, 9 Jan 2023 11:47:34 +0800 Subject: [PATCH 063/765] f2fs: remove unneeded f2fs_cp_error() in f2fs_create_whiteout() f2fs_rename() has checked CP_ERROR_FLAG, so remove redundant check in f2fs_create_whiteout(). Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/namei.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c index 6032589099ce4..82923273f4bb8 100644 --- a/fs/f2fs/namei.c +++ b/fs/f2fs/namei.c @@ -926,9 +926,6 @@ static int f2fs_tmpfile(struct user_namespace *mnt_userns, struct inode *dir, static int f2fs_create_whiteout(struct user_namespace *mnt_userns, struct inode *dir, struct inode **whiteout) { - if (unlikely(f2fs_cp_error(F2FS_I_SB(dir)))) - return -EIO; - return __f2fs_tmpfile(mnt_userns, dir, NULL, S_IFCHR | WHITEOUT_MODE, true, whiteout); } From d48a7b3a72f121655d95b5157c32c7d555e44c05 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Mon, 9 Jan 2023 11:49:20 +0800 Subject: [PATCH 064/765] f2fs: fix to do sanity check on extent cache correctly In do_read_inode(), sanity_check_inode() should be called after f2fs_init_read_extent_tree(), fix it. Fixes: 72840cccc0a1 ("f2fs: allocate the extent_cache by default") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/inode.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c index 01b9e6f85f6be..fd2ee55549858 100644 --- a/fs/f2fs/inode.c +++ b/fs/f2fs/inode.c @@ -413,12 +413,6 @@ static int do_read_inode(struct inode *inode) fi->i_inline_xattr_size = 0; } - if (!sanity_check_inode(inode, node_page)) { - f2fs_put_page(node_page, 1); - f2fs_handle_error(sbi, ERROR_CORRUPTED_INODE); - return -EFSCORRUPTED; - } - /* check data exist */ if (f2fs_has_inline_data(inode) && !f2fs_exist_data(inode)) __recover_inline_status(inode, node_page); @@ -482,6 +476,12 @@ static int do_read_inode(struct inode *inode) f2fs_init_read_extent_tree(inode, node_page); f2fs_init_age_extent_tree(inode); + if (!sanity_check_inode(inode, node_page)) { + f2fs_put_page(node_page, 1); + f2fs_handle_error(sbi, ERROR_CORRUPTED_INODE); + return -EFSCORRUPTED; + } + f2fs_put_page(node_page, 1); stat_inc_inline_xattr(inode); From 2381a68556c0d2cb027f9dbfabc654a899782c62 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Mon, 16 Jan 2023 22:12:28 +0800 Subject: [PATCH 065/765] f2fs: fix to show discard_unit mount opt Convert to show discard_unit only when has DISCARD opt. Signed-off-by: Yangtao Li Signed-off-by: Jaegeuk Kim --- fs/f2fs/super.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 5fc83771042d3..ab8a77ffc1f43 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -1908,10 +1908,17 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root) seq_puts(seq, ",disable_roll_forward"); if (test_opt(sbi, NORECOVERY)) seq_puts(seq, ",norecovery"); - if (test_opt(sbi, DISCARD)) + if (test_opt(sbi, DISCARD)) { seq_puts(seq, ",discard"); - else + if (F2FS_OPTION(sbi).discard_unit == DISCARD_UNIT_BLOCK) + seq_printf(seq, ",discard_unit=%s", "block"); + else if (F2FS_OPTION(sbi).discard_unit == DISCARD_UNIT_SEGMENT) + seq_printf(seq, ",discard_unit=%s", "segment"); + else if (F2FS_OPTION(sbi).discard_unit == DISCARD_UNIT_SECTION) + seq_printf(seq, ",discard_unit=%s", "section"); + } else { seq_puts(seq, ",nodiscard"); + } if (test_opt(sbi, NOHEAP)) seq_puts(seq, ",no_heap"); else @@ -2035,13 +2042,6 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root) if (test_opt(sbi, ATGC)) seq_puts(seq, ",atgc"); - if (F2FS_OPTION(sbi).discard_unit == DISCARD_UNIT_BLOCK) - seq_printf(seq, ",discard_unit=%s", "block"); - else if (F2FS_OPTION(sbi).discard_unit == DISCARD_UNIT_SEGMENT) - seq_printf(seq, ",discard_unit=%s", "segment"); - else if (F2FS_OPTION(sbi).discard_unit == DISCARD_UNIT_SECTION) - seq_printf(seq, ",discard_unit=%s", "section"); - if (F2FS_OPTION(sbi).memory_mode == MEMORY_MODE_NORMAL) seq_printf(seq, ",memory=%s", "normal"); else if (F2FS_OPTION(sbi).memory_mode == MEMORY_MODE_LOW) From c1085957dece02bda586cbffc1328f3bca09325f Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Thu, 12 Jan 2023 21:34:43 +0800 Subject: [PATCH 066/765] f2fs: clarify compress level bit offset commit 3fde13f817e2 ("f2fs: compress: support compress level") introduce compress level, which macro(COMPRESS_LEVEL_OFFSET) is 8, But use wrong comment about compress level. Let's fix it. Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- include/linux/f2fs_fs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/f2fs_fs.h b/include/linux/f2fs_fs.h index ee0d75d9a302d..1701f25117ea4 100644 --- a/include/linux/f2fs_fs.h +++ b/include/linux/f2fs_fs.h @@ -315,7 +315,7 @@ struct f2fs_inode { __u8 i_log_cluster_size; /* log of cluster size */ __le16 i_compress_flag; /* compress flag */ /* 0 bit: chksum flag - * [10,15] bits: compress level + * [8,15] bits: compress level */ __le32 i_extra_end[0]; /* for attribute size calculation */ } __packed; From b1c5ef26e4e80277641afcfedffe598c3023c095 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Fri, 13 Jan 2023 03:14:04 +0800 Subject: [PATCH 067/765] f2fs: return true if all cmd were issued or no cmd need to be issued for f2fs_issue_discard_timeout() f2fs_issue_discard_timeout() returns whether discard cmds are dropped, which does not match the meaning of the function. Let's change it to return whether all discard cmd are issued. After commit 4d67490498ac ("f2fs: Don't create discard thread when device doesn't support realtime discard"), f2fs_issue_discard_timeout() is alse called by f2fs_remount(). Since the comments of f2fs_issue_discard_timeout() doesn't make much sense, let's update it. Signed-off-by: Yangtao Li Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 13 ++++++++++--- fs/f2fs/super.c | 7 +++---- 2 files changed, 13 insertions(+), 7 deletions(-) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 5e603b808487d..95a5e6ef7e808 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -1656,7 +1656,14 @@ void f2fs_stop_discard_thread(struct f2fs_sb_info *sbi) } } -/* This comes from f2fs_put_super */ +/** + * f2fs_issue_discard_timeout() - Issue all discard cmd within UMOUNT_DISCARD_TIMEOUT + * @sbi: the f2fs_sb_info data for discard cmd to issue + * + * When UMOUNT_DISCARD_TIMEOUT is exceeded, all remaining discard commands will be dropped + * + * Return true if issued all discard cmd or no discard cmd need issue, otherwise return false. + */ bool f2fs_issue_discard_timeout(struct f2fs_sb_info *sbi) { struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info; @@ -1664,7 +1671,7 @@ bool f2fs_issue_discard_timeout(struct f2fs_sb_info *sbi) bool dropped; if (!atomic_read(&dcc->discard_cmd_cnt)) - return false; + return true; __init_discard_policy(sbi, &dpolicy, DPOLICY_UMOUNT, dcc->discard_granularity); @@ -1675,7 +1682,7 @@ bool f2fs_issue_discard_timeout(struct f2fs_sb_info *sbi) __wait_all_discard_cmd(sbi, NULL); f2fs_bug_on(sbi, atomic_read(&dcc->discard_cmd_cnt)); - return dropped; + return !dropped; } static int issue_discard_thread(void *data) diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index ab8a77ffc1f43..fddff5deaed2f 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -1546,7 +1546,7 @@ static void f2fs_put_super(struct super_block *sb) { struct f2fs_sb_info *sbi = F2FS_SB(sb); int i; - bool dropped; + bool done; /* unregister procfs/sysfs entries in advance to avoid race case */ f2fs_unregister_sysfs(sbi); @@ -1576,9 +1576,8 @@ static void f2fs_put_super(struct super_block *sb) } /* be sure to wait for any on-going discard commands */ - dropped = f2fs_issue_discard_timeout(sbi); - - if (f2fs_realtime_discard_enable(sbi) && !sbi->discard_blks && !dropped) { + done = f2fs_issue_discard_timeout(sbi); + if (f2fs_realtime_discard_enable(sbi) && !sbi->discard_blks && done) { struct cp_control cpc = { .reason = CP_UMOUNT | CP_TRIMMED, }; From 9b13a8662ea61430005d3de3fc1d73e383b1ce5a Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Tue, 17 Jan 2023 21:24:42 +0800 Subject: [PATCH 068/765] f2fs: fix to check warm_data_age_threshold hot_data_age_threshold is a non-zero positive number, and condition 2 includes condition 1, so there is no need to additionally judge whether t is 0. And let's remove it. Signed-off-by: Yangtao Li Signed-off-by: Jaegeuk Kim --- fs/f2fs/sysfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c index e396851a6dd10..3c6425f9ed0ad 100644 --- a/fs/f2fs/sysfs.c +++ b/fs/f2fs/sysfs.c @@ -689,7 +689,7 @@ static ssize_t __sbi_store(struct f2fs_attr *a, } if (!strcmp(a->attr.name, "warm_data_age_threshold")) { - if (t == 0 || t <= sbi->hot_data_age_threshold) + if (t <= sbi->hot_data_age_threshold) return -EINVAL; if (t == *ui) return count; From b1b9896718bc1a212dc288ad66a5fa2fef11353d Mon Sep 17 00:00:00 2001 From: Alexander Potapenko Date: Mon, 21 Nov 2022 12:21:32 +0100 Subject: [PATCH 069/765] fs: f2fs: initialize fsdata in pagecache_write() When aops->write_begin() does not initialize fsdata, KMSAN may report an error passing the latter to aops->write_end(). Fix this by unconditionally initializing fsdata. Suggested-by: Eric Biggers Fixes: 95ae251fe828 ("f2fs: add fs-verity support") Signed-off-by: Alexander Potapenko Reviewed-by: Eric Biggers Signed-off-by: Jaegeuk Kim --- fs/f2fs/verity.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/f2fs/verity.c b/fs/f2fs/verity.c index c352fff88a5e6..3f4f3295f1c66 100644 --- a/fs/f2fs/verity.c +++ b/fs/f2fs/verity.c @@ -81,7 +81,7 @@ static int pagecache_write(struct inode *inode, const void *buf, size_t count, size_t n = min_t(size_t, count, PAGE_SIZE - offset_in_page(pos)); struct page *page; - void *fsdata; + void *fsdata = NULL; int res; res = aops->write_begin(NULL, mapping, pos, n, &page, &fsdata); From 9a5571cff4ffcfc24847df9fd545cc5799ac0ee5 Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Sun, 22 Jan 2023 23:04:14 -0800 Subject: [PATCH 070/765] f2fs: fix information leak in f2fs_move_inline_dirents() When converting an inline directory to a regular one, f2fs is leaking uninitialized memory to disk because it doesn't initialize the entire directory block. Fix this by zero-initializing the block. This bug was introduced by commit 4ec17d688d74 ("f2fs: avoid unneeded initializing when converting inline dentry"), which didn't consider the security implications of leaking uninitialized memory to disk. This was found by running xfstest generic/435 on a KMSAN-enabled kernel. Fixes: 4ec17d688d74 ("f2fs: avoid unneeded initializing when converting inline dentry") Cc: # v4.3+ Signed-off-by: Eric Biggers Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/inline.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c index 08e302d32118d..72269e7efd260 100644 --- a/fs/f2fs/inline.c +++ b/fs/f2fs/inline.c @@ -421,18 +421,17 @@ static int f2fs_move_inline_dirents(struct inode *dir, struct page *ipage, dentry_blk = page_address(page); + /* + * Start by zeroing the full block, to ensure that all unused space is + * zeroed and no uninitialized memory is leaked to disk. + */ + memset(dentry_blk, 0, F2FS_BLKSIZE); + make_dentry_ptr_inline(dir, &src, inline_dentry); make_dentry_ptr_block(dir, &dst, dentry_blk); /* copy data from inline dentry block to new dentry block */ memcpy(dst.bitmap, src.bitmap, src.nr_bitmap); - memset(dst.bitmap + src.nr_bitmap, 0, dst.nr_bitmap - src.nr_bitmap); - /* - * we do not need to zero out remainder part of dentry and filename - * field, since we have used bitmap for marking the usage status of - * them, besides, we can also ignore copying/zeroing reserved space - * of dentry block, because them haven't been used so far. - */ memcpy(dst.dentry, src.dentry, SIZE_OF_DIR_ENTRY * src.max); memcpy(dst.filename, src.filename, src.max * F2FS_SLOT_LEN); From e6261beb0c629403dc58997294dd521bd23664af Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Mon, 23 Jan 2023 17:46:01 +0800 Subject: [PATCH 071/765] f2fs: allow set compression option of files without blocks Files created by truncate have a size but no blocks, so they can be allowed to set compression option. Fixes: e1e8debec656 ("f2fs: add F2FS_IOC_SET_COMPRESS_OPTION ioctl") Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index f5c1b78149540..391da47761e32 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -3941,7 +3941,7 @@ static int f2fs_ioc_set_compress_option(struct file *filp, unsigned long arg) goto out; } - if (inode->i_size != 0) { + if (F2FS_HAS_BLOCKS(inode)) { ret = -EFBIG; goto out; } From ae267fc1cfe9f941afcb90dc963ee6448dae73cf Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Mon, 9 Jan 2023 11:44:51 +0800 Subject: [PATCH 072/765] f2fs: fix to abort atomic write only during do_exist() Commit 7a10f0177e11 ("f2fs: don't give partially written atomic data from process crash") attempted to drop atomic write data after process crash, however, f2fs_abort_atomic_write() may be called from noncrash case, fix it by adding missed PF_EXITING check condition f2fs_file_flush(). - application crashs - do_exit - exit_signals -- sets PF_EXITING - exit_files - put_files_struct - close_files - filp_close - flush (f2fs_file_flush) - check atomic_write_task && PF_EXITING - f2fs_abort_atomic_write Fixes: 7a10f0177e11 ("f2fs: don't give partially written atomic data from process crash") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 391da47761e32..eaae2ad3957b0 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -1876,7 +1876,8 @@ static int f2fs_file_flush(struct file *file, fl_owner_t id) * until all the writers close its file. Since this should be done * before dropping file lock, it needs to do in ->flush. */ - if (F2FS_I(inode)->atomic_write_task == current) + if (F2FS_I(inode)->atomic_write_task == current && + (current->flags & PF_EXITING)) f2fs_abort_atomic_write(inode, true); return 0; } From 2163a691c5f3d30326ff09193c97aec47cec9eba Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 19 Jan 2023 07:36:18 +0100 Subject: [PATCH 073/765] f2fs: remove __add_sum_entry This function just assigns a summary entry. This can be done entirely typesafe with an open code struct assignment that relies on array indexing. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 23 ++--------------------- 1 file changed, 2 insertions(+), 21 deletions(-) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 95a5e6ef7e808..39071003c6d73 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -2339,19 +2339,6 @@ bool f2fs_is_checkpointed_data(struct f2fs_sb_info *sbi, block_t blkaddr) return is_cp; } -/* - * This function should be resided under the curseg_mutex lock - */ -static void __add_sum_entry(struct f2fs_sb_info *sbi, int type, - struct f2fs_summary *sum) -{ - struct curseg_info *curseg = CURSEG_I(sbi, type); - void *addr = curseg->sum_blk; - - addr += curseg->next_blkoff * sizeof(struct f2fs_summary); - memcpy(addr, sum, sizeof(struct f2fs_summary)); -} - /* * Calculate the number of current summary pages for writing */ @@ -3278,13 +3265,7 @@ void f2fs_allocate_data_block(struct f2fs_sb_info *sbi, struct page *page, f2fs_wait_discard_bio(sbi, *new_blkaddr); - /* - * __add_sum_entry should be resided under the curseg_mutex - * because, this function updates a summary entry in the - * current summary block. - */ - __add_sum_entry(sbi, type, sum); - + curseg->sum_blk->entries[curseg->next_blkoff] = *sum; __refresh_next_blkoff(sbi, curseg); stat_inc_block_count(sbi, curseg); @@ -3587,7 +3568,7 @@ void f2fs_do_replace_block(struct f2fs_sb_info *sbi, struct f2fs_summary *sum, } curseg->next_blkoff = GET_BLKOFF_FROM_SEG0(sbi, new_blkaddr); - __add_sum_entry(sbi, type, sum); + curseg->sum_blk->entries[curseg->next_blkoff] = *sum; if (!recover_curseg || recover_newaddr) { if (!from_gc) From 37abc36ed2d39e57ac0fad163d061366b4a015cb Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Mon, 23 Jan 2023 21:02:07 +0100 Subject: [PATCH 074/765] rtc: ab-eoz9: use IRQ flags obtained from fwnode Allow the IRQ type to be passed from the device tree if available as there may be components changing the trigger type of the interrupt between the RTC and the IRQ controller. Link: https://lore.kernel.org/r/20230123200217.1236011-2-alexandre.belloni@bootlin.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-ab-eoz9.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/rtc/rtc-ab-eoz9.c b/drivers/rtc/rtc-ab-eoz9.c index 2f8deb8c4cd3e..34611f6dedcba 100644 --- a/drivers/rtc/rtc-ab-eoz9.c +++ b/drivers/rtc/rtc-ab-eoz9.c @@ -536,9 +536,14 @@ static int abeoz9_probe(struct i2c_client *client) clear_bit(RTC_FEATURE_ALARM, data->rtc->features); if (client->irq > 0) { + unsigned long irqflags = IRQF_TRIGGER_LOW; + + if (dev_fwnode(&client->dev)) + irqflags = 0; + ret = devm_request_threaded_irq(dev, client->irq, NULL, abeoz9_rtc_irq, - IRQF_TRIGGER_LOW | IRQF_ONESHOT, + irqflags | IRQF_ONESHOT, dev_name(dev), dev); if (ret) { dev_err(dev, "failed to request alarm irq\n"); From badba1e5b111c6f32038953efc8078cba3390326 Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Mon, 23 Jan 2023 21:02:08 +0100 Subject: [PATCH 075/765] rtc: hym8563: use IRQ flags obtained from fwnode Allow the IRQ type to be passed from the device tree if available as there may be components changing the trigger type of the interrupt between the RTC and the IRQ controller. Link: https://lore.kernel.org/r/20230123200217.1236011-3-alexandre.belloni@bootlin.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-hym8563.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/rtc/rtc-hym8563.c b/drivers/rtc/rtc-hym8563.c index cc710d682121b..7d5a298a9a3bc 100644 --- a/drivers/rtc/rtc-hym8563.c +++ b/drivers/rtc/rtc-hym8563.c @@ -518,9 +518,14 @@ static int hym8563_probe(struct i2c_client *client) } if (client->irq > 0) { + unsigned long irqflags = IRQF_TRIGGER_LOW; + + if (dev_fwnode(&client->dev)) + irqflags = 0; + ret = devm_request_threaded_irq(&client->dev, client->irq, NULL, hym8563_irq, - IRQF_TRIGGER_LOW | IRQF_ONESHOT, + irqflags | IRQF_ONESHOT, client->name, hym8563); if (ret < 0) { dev_err(&client->dev, "irq %d request failed, %d\n", From f181987ef4772c2d9412b43bc3d0e34fe368cddb Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Mon, 23 Jan 2023 21:02:09 +0100 Subject: [PATCH 076/765] rtc: m41t80: use IRQ flags obtained from fwnode Allow the IRQ type to be passed from the device tree if available as there may be components changing the trigger type of the interrupt between the RTC and the IRQ controller. Link: https://lore.kernel.org/r/20230123200217.1236011-4-alexandre.belloni@bootlin.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-m41t80.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/rtc/rtc-m41t80.c b/drivers/rtc/rtc-m41t80.c index 494052dbd39ff..c1963f7c424d7 100644 --- a/drivers/rtc/rtc-m41t80.c +++ b/drivers/rtc/rtc-m41t80.c @@ -914,9 +914,14 @@ static int m41t80_probe(struct i2c_client *client) "wakeup-source"); #endif if (client->irq > 0) { + unsigned long irqflags = IRQF_TRIGGER_LOW; + + if (dev_fwnode(&client->dev)) + irqflags = 0; + rc = devm_request_threaded_irq(&client->dev, client->irq, NULL, m41t80_handle_irq, - IRQF_TRIGGER_LOW | IRQF_ONESHOT, + irqflags | IRQF_ONESHOT, "m41t80", client); if (rc) { dev_warn(&client->dev, "unable to request IRQ, alarms disabled\n"); From 5434a4e472c71448bbdc061f21ebf1b0a2d23753 Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Mon, 23 Jan 2023 21:02:10 +0100 Subject: [PATCH 077/765] rtc: pcf2123: use IRQ flags obtained from fwnode Allow the IRQ type to be passed from the device tree if available as there may be components changing the trigger type of the interrupt between the RTC and the IRQ controller. Link: https://lore.kernel.org/r/20230123200217.1236011-5-alexandre.belloni@bootlin.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pcf2123.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/rtc/rtc-pcf2123.c b/drivers/rtc/rtc-pcf2123.c index e13b5e695d06a..e714661e61a91 100644 --- a/drivers/rtc/rtc-pcf2123.c +++ b/drivers/rtc/rtc-pcf2123.c @@ -413,9 +413,14 @@ static int pcf2123_probe(struct spi_device *spi) /* Register alarm irq */ if (spi->irq > 0) { + unsigned long irqflags = IRQF_TRIGGER_LOW; + + if (dev_fwnode(&spi->dev)) + irqflags = 0; + ret = devm_request_threaded_irq(&spi->dev, spi->irq, NULL, pcf2123_rtc_irq, - IRQF_TRIGGER_LOW | IRQF_ONESHOT, + irqflags | IRQF_ONESHOT, pcf2123_driver.driver.name, &spi->dev); if (!ret) device_init_wakeup(&spi->dev, true); From 7e815272c8d0922ad1d0def11496ac63fa71fa78 Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Mon, 23 Jan 2023 21:02:11 +0100 Subject: [PATCH 078/765] rtc: pcf85063: use IRQ flags obtained from fwnode Allow the IRQ type to be passed from the device tree if available as there may be components changing the trigger type of the interrupt between the RTC and the IRQ controller. Link: https://lore.kernel.org/r/20230123200217.1236011-6-alexandre.belloni@bootlin.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pcf85063.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/rtc/rtc-pcf85063.c b/drivers/rtc/rtc-pcf85063.c index 754e03984f986..71a4563559819 100644 --- a/drivers/rtc/rtc-pcf85063.c +++ b/drivers/rtc/rtc-pcf85063.c @@ -621,9 +621,14 @@ static int pcf85063_probe(struct i2c_client *client) clear_bit(RTC_FEATURE_ALARM, pcf85063->rtc->features); if (config->has_alarms && client->irq > 0) { + unsigned long irqflags = IRQF_TRIGGER_LOW; + + if (dev_fwnode(&client->dev)) + irqflags = 0; + err = devm_request_threaded_irq(&client->dev, client->irq, NULL, pcf85063_rtc_handle_irq, - IRQF_TRIGGER_LOW | IRQF_ONESHOT, + irqflags | IRQF_ONESHOT, "pcf85063", pcf85063); if (err) { dev_warn(&pcf85063->rtc->dev, From 3542db1d1fd62f75eb69079d219d88170626444f Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Mon, 23 Jan 2023 21:02:12 +0100 Subject: [PATCH 079/765] rtc: pcf8523: use IRQ flags obtained from fwnode Allow the IRQ type to be passed from the device tree if available as there may be components changing the trigger type of the interrupt between the RTC and the IRQ controller. Link: https://lore.kernel.org/r/20230123200217.1236011-7-alexandre.belloni@bootlin.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pcf8523.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/rtc/rtc-pcf8523.c b/drivers/rtc/rtc-pcf8523.c index 92de99f11a7a5..2e111cdb94f76 100644 --- a/drivers/rtc/rtc-pcf8523.c +++ b/drivers/rtc/rtc-pcf8523.c @@ -445,13 +445,18 @@ static int pcf8523_probe(struct i2c_client *client) clear_bit(RTC_FEATURE_UPDATE_INTERRUPT, rtc->features); if (client->irq > 0) { + unsigned long irqflags = IRQF_TRIGGER_LOW; + + if (dev_fwnode(&client->dev)) + irqflags = 0; + err = regmap_write(pcf8523->regmap, PCF8523_TMR_CLKOUT_CTRL, 0x38); if (err < 0) return err; err = devm_request_threaded_irq(&client->dev, client->irq, NULL, pcf8523_irq, - IRQF_SHARED | IRQF_ONESHOT | IRQF_TRIGGER_LOW, + IRQF_SHARED | IRQF_ONESHOT | irqflags, dev_name(&rtc->dev), pcf8523); if (err) return err; From dd7166c8ba6e27568ba1c3fa8874316150ce694c Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Mon, 23 Jan 2023 21:02:13 +0100 Subject: [PATCH 080/765] rtc: pcf85363: use IRQ flags obtained fromfwnode Allow the IRQ type to be passed from the device tree if available as there may be components changing the trigger type of the interrupt between the RTC and the IRQ controller. Link: https://lore.kernel.org/r/20230123200217.1236011-8-alexandre.belloni@bootlin.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pcf85363.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/rtc/rtc-pcf85363.c b/drivers/rtc/rtc-pcf85363.c index c05b722f00605..5de323acd1786 100644 --- a/drivers/rtc/rtc-pcf85363.c +++ b/drivers/rtc/rtc-pcf85363.c @@ -400,12 +400,17 @@ static int pcf85363_probe(struct i2c_client *client) clear_bit(RTC_FEATURE_ALARM, pcf85363->rtc->features); if (client->irq > 0) { + unsigned long irqflags = IRQF_TRIGGER_LOW; + + if (dev_fwnode(&client->dev)) + irqflags = 0; + regmap_write(pcf85363->regmap, CTRL_FLAGS, 0); regmap_update_bits(pcf85363->regmap, CTRL_PIN_IO, PIN_IO_INTA_OUT, PIN_IO_INTAPM); ret = devm_request_threaded_irq(&client->dev, client->irq, NULL, pcf85363_rtc_handle_irq, - IRQF_TRIGGER_LOW | IRQF_ONESHOT, + irqflags | IRQF_ONESHOT, "pcf85363", client); if (ret) dev_warn(&client->dev, "unable to request IRQ, alarms disabled\n"); From 827009a8d3f80d81c6da634f4ab1a33e2d539aa4 Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Mon, 23 Jan 2023 21:02:14 +0100 Subject: [PATCH 081/765] rtc: pcf8563: use IRQ flags obtained from fwnode Allow the IRQ type to be passed from the device tree if available as there may be components changing the trigger type of the interrupt between the RTC and the IRQ controller. Link: https://lore.kernel.org/r/20230123200217.1236011-9-alexandre.belloni@bootlin.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pcf8563.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/rtc/rtc-pcf8563.c b/drivers/rtc/rtc-pcf8563.c index 0a7fd94784651..7e720472213c7 100644 --- a/drivers/rtc/rtc-pcf8563.c +++ b/drivers/rtc/rtc-pcf8563.c @@ -558,9 +558,14 @@ static int pcf8563_probe(struct i2c_client *client) pcf8563->rtc->set_start_time = true; if (client->irq > 0) { + unsigned long irqflags = IRQF_TRIGGER_LOW; + + if (dev_fwnode(&client->dev)) + irqflags = 0; + err = devm_request_threaded_irq(&client->dev, client->irq, NULL, pcf8563_irq, - IRQF_SHARED | IRQF_ONESHOT | IRQF_TRIGGER_LOW, + IRQF_SHARED | IRQF_ONESHOT | irqflags, pcf8563_driver.driver.name, client); if (err) { dev_err(&client->dev, "unable to request IRQ %d\n", From bfff849f1deff5607c8fb1031be319cd88e0ecd7 Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Mon, 23 Jan 2023 21:02:15 +0100 Subject: [PATCH 082/765] rtc: rv3029c2: use IRQ flags obtained from fwnode Allow the IRQ type to be passed from the device tree if available as there may be components changing the trigger type of the interrupt between the RTC and the IRQ controller. Link: https://lore.kernel.org/r/20230123200217.1236011-10-alexandre.belloni@bootlin.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-rv3029c2.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/rtc/rtc-rv3029c2.c b/drivers/rtc/rtc-rv3029c2.c index e4fdd47ae066c..0852f6709a859 100644 --- a/drivers/rtc/rtc-rv3029c2.c +++ b/drivers/rtc/rtc-rv3029c2.c @@ -735,9 +735,14 @@ static int rv3029_probe(struct device *dev, struct regmap *regmap, int irq, return PTR_ERR(rv3029->rtc); if (rv3029->irq > 0) { + unsigned long irqflags = IRQF_TRIGGER_LOW; + + if (dev_fwnode(dev)) + irqflags = 0; + rc = devm_request_threaded_irq(dev, rv3029->irq, NULL, rv3029_handle_irq, - IRQF_TRIGGER_LOW | IRQF_ONESHOT, + irqflags | IRQF_ONESHOT, "rv3029", dev); if (rc) { dev_warn(dev, "unable to request IRQ, alarms disabled\n"); From c4b12f89f5aefe559522e4e41528a6c719fcdcdb Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Mon, 23 Jan 2023 21:02:16 +0100 Subject: [PATCH 083/765] rtc: rv3032: use IRQ flags obtained from fwnode Allow the IRQ type to be passed from the device tree if available as there may be components changing the trigger type of the interrupt between the RTC and the IRQ controller. Link: https://lore.kernel.org/r/20230123200217.1236011-11-alexandre.belloni@bootlin.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-rv3032.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/rtc/rtc-rv3032.c b/drivers/rtc/rtc-rv3032.c index c3bee305eacc6..bf6954ec5943e 100644 --- a/drivers/rtc/rtc-rv3032.c +++ b/drivers/rtc/rtc-rv3032.c @@ -930,9 +930,14 @@ static int rv3032_probe(struct i2c_client *client) return PTR_ERR(rv3032->rtc); if (client->irq > 0) { + unsigned long irqflags = IRQF_TRIGGER_LOW; + + if (dev_fwnode(&client->dev)) + irqflags = 0; + ret = devm_request_threaded_irq(&client->dev, client->irq, NULL, rv3032_handle_irq, - IRQF_TRIGGER_LOW | IRQF_ONESHOT, + irqflags | IRQF_ONESHOT, "rv3032", rv3032); if (ret) { dev_warn(&client->dev, "unable to request IRQ, alarms disabled\n"); From 11bfd6fcebfa25a77e601e68f5d4ba897e7b3047 Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Mon, 23 Jan 2023 21:02:17 +0100 Subject: [PATCH 084/765] rtc: rv8803: use IRQ flags obtained from fwnode Allow the IRQ type to be passed from the device tree if available as there may be components changing the trigger type of the interrupt between the RTC and the IRQ controller. Link: https://lore.kernel.org/r/20230123200217.1236011-12-alexandre.belloni@bootlin.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-rv8803.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/rtc/rtc-rv8803.c b/drivers/rtc/rtc-rv8803.c index b581b6d5ad731..53d1de01b7197 100644 --- a/drivers/rtc/rtc-rv8803.c +++ b/drivers/rtc/rtc-rv8803.c @@ -641,9 +641,14 @@ static int rv8803_probe(struct i2c_client *client) return PTR_ERR(rv8803->rtc); if (client->irq > 0) { + unsigned long irqflags = IRQF_TRIGGER_LOW; + + if (dev_fwnode(&client->dev)) + irqflags = 0; + err = devm_request_threaded_irq(&client->dev, client->irq, NULL, rv8803_handle_irq, - IRQF_TRIGGER_LOW | IRQF_ONESHOT, + irqflags | IRQF_ONESHOT, "rv8803", client); if (err) { dev_warn(&client->dev, "unable to request IRQ, alarms disabled\n"); From 4bbdced5db096bded4e2d10259e9a5e0070a4cb3 Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Mon, 23 Jan 2023 21:02:06 +0100 Subject: [PATCH 085/765] rtc: rx8010: use IRQ flags obtained from fwnode Allow the IRQ type to be passed from the device tree if available as there may be components changing the trigger type of the interrupt between the RTC and the IRQ controller. Link: https://lore.kernel.org/r/20230123200217.1236011-1-alexandre.belloni@bootlin.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-rx8010.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/rtc/rtc-rx8010.c b/drivers/rtc/rtc-rx8010.c index d090565707390..b9c8dad262085 100644 --- a/drivers/rtc/rtc-rx8010.c +++ b/drivers/rtc/rtc-rx8010.c @@ -394,10 +394,14 @@ static int rx8010_probe(struct i2c_client *client) return PTR_ERR(rx8010->rtc); if (client->irq > 0) { - dev_info(dev, "IRQ %d supplied\n", client->irq); + unsigned long irqflags = IRQF_TRIGGER_LOW; + + if (dev_fwnode(&client->dev)) + irqflags = 0; + err = devm_request_threaded_irq(dev, client->irq, NULL, rx8010_irq_1_handler, - IRQF_TRIGGER_LOW | IRQF_ONESHOT, + irqflags | IRQF_ONESHOT, "rx8010", client); if (err) { dev_err(dev, "unable to request IRQ\n"); From 04596d4b3e0d20697193341694e8c99961702259 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Mon, 19 Dec 2022 09:14:40 +0100 Subject: [PATCH 086/765] rtc: max8907: Drop unused i2c include MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit rtc-max8907 is a platform driver that doesn't use any symbol provided in . So drop the include. Signed-off-by: Uwe Kleine-König Link: https://lore.kernel.org/r/20221219081440.1399791-1-u.kleine-koenig@pengutronix.de Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-max8907.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/rtc/rtc-max8907.c b/drivers/rtc/rtc-max8907.c index db3495d102747..af97140dd00a5 100644 --- a/drivers/rtc/rtc-max8907.c +++ b/drivers/rtc/rtc-max8907.c @@ -9,7 +9,6 @@ */ #include -#include #include #include #include From cd955bdd6aa5ec54cdef622a142f8899a64b5446 Mon Sep 17 00:00:00 2001 From: Ian Rogers Date: Thu, 26 Jan 2023 11:06:06 -0800 Subject: [PATCH 087/765] objtool: Fix HOSTCC flag usage HOSTCC is always wanted when building objtool. Setting CC to HOSTCC happens after tools/scripts/Makefile.include is included, meaning flags (like CFLAGS) are set assuming say CC is gcc, but then it can be later set to HOSTCC which may be clang. tools/scripts/Makefile.include is needed for host set up and common macros in objtool's Makefile. Rather than override the CC variable to HOSTCC, just pass CC as HOSTCC to the sub-makes of Makefile.build, the libsubcmd builds and also to the linkage step. Signed-off-by: Ian Rogers Link: https://lore.kernel.org/r/20230126190606.40739-4-irogers@google.com Signed-off-by: Josh Poimboeuf --- tools/objtool/Makefile | 27 +++++++++++++++------------ 1 file changed, 15 insertions(+), 12 deletions(-) diff --git a/tools/objtool/Makefile b/tools/objtool/Makefile index d54b66986627c..83b100c1e7f68 100644 --- a/tools/objtool/Makefile +++ b/tools/objtool/Makefile @@ -2,11 +2,6 @@ include ../scripts/Makefile.include include ../scripts/Makefile.arch -# always use the host compiler -AR = $(HOSTAR) -CC = $(HOSTCC) -LD = $(HOSTLD) - ifeq ($(srctree),) srctree := $(patsubst %/,%,$(dir $(CURDIR))) srctree := $(patsubst %/,%,$(dir $(srctree))) @@ -34,13 +29,18 @@ INCLUDES := -I$(srctree)/tools/include \ -I$(srctree)/tools/objtool/include \ -I$(srctree)/tools/objtool/arch/$(SRCARCH)/include \ -I$(LIBSUBCMD_OUTPUT)/include +# Note, EXTRA_WARNINGS here was determined for CC and not HOSTCC, it +# is passed here to match a legacy behavior. WARNINGS := $(EXTRA_WARNINGS) -Wno-switch-default -Wno-switch-enum -Wno-packed -Wno-nested-externs -CFLAGS := -Werror $(WARNINGS) $(KBUILD_HOSTCFLAGS) -g $(INCLUDES) $(LIBELF_FLAGS) -LDFLAGS += $(LIBELF_LIBS) $(LIBSUBCMD) $(KBUILD_HOSTLDFLAGS) +OBJTOOL_CFLAGS := -Werror $(WARNINGS) $(KBUILD_HOSTCFLAGS) -g $(INCLUDES) $(LIBELF_FLAGS) +OBJTOOL_LDFLAGS := $(LIBELF_LIBS) $(LIBSUBCMD) $(KBUILD_HOSTLDFLAGS) # Allow old libelf to be used: -elfshdr := $(shell echo '$(pound)include ' | $(CC) $(CFLAGS) -x c -E - | grep elf_getshdr) -CFLAGS += $(if $(elfshdr),,-DLIBELF_USE_DEPRECATED) +elfshdr := $(shell echo '$(pound)include ' | $(HOSTCC) $(OBJTOOL_CFLAGS) -x c -E - | grep elf_getshdr) +OBJTOOL_CFLAGS += $(if $(elfshdr),,-DLIBELF_USE_DEPRECATED) + +# Always want host compilation. +HOST_OVERRIDES := CC="$(HOSTCC)" LD="$(HOSTLD)" AR="$(HOSTAR)" AWK = awk MKDIR = mkdir @@ -61,12 +61,14 @@ export BUILD_ORC export srctree OUTPUT CFLAGS SRCARCH AWK include $(srctree)/tools/build/Makefile.include -$(OBJTOOL_IN): fixdep FORCE +$(OBJTOOL_IN): fixdep $(LIBSUBCMD) FORCE $(Q)$(CONFIG_SHELL) ./sync-check.sh - $(Q)$(MAKE) $(build)=objtool + $(Q)$(MAKE) $(build)=objtool $(HOST_OVERRIDES) CFLAGS="$(OBJTOOL_CFLAGS)" \ + LDFLAGS="$(OBJTOOL_LDFLAGS)" + $(OBJTOOL): $(LIBSUBCMD) $(OBJTOOL_IN) - $(QUIET_LINK)$(CC) $(OBJTOOL_IN) $(LDFLAGS) -o $@ + $(QUIET_LINK)$(HOSTCC) $(OBJTOOL_IN) $(OBJTOOL_LDFLAGS) -o $@ $(LIBSUBCMD_OUTPUT): @@ -75,6 +77,7 @@ $(LIBSUBCMD_OUTPUT): $(LIBSUBCMD): fixdep $(LIBSUBCMD_OUTPUT) FORCE $(Q)$(MAKE) -C $(LIBSUBCMD_DIR) O=$(LIBSUBCMD_OUTPUT) \ DESTDIR=$(LIBSUBCMD_OUTPUT) prefix= subdir= \ + $(HOST_OVERRIDES) EXTRA_CFLAGS="$(OBJTOOL_CFLAGS)" \ $@ install_headers $(LIBSUBCMD)-clean: From d93ee0553cf2e83c1696a18423bcf05b94b85e1d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Tue, 27 Dec 2022 16:00:57 +0000 Subject: [PATCH 088/765] objtool: Make struct entries[] static and const MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This data is not modified and not used outside of special.c. Also adapt its users to the constness. Signed-off-by: Thomas Weißschuh Link: https://lore.kernel.org/r/20221216-objtool-memory-v2-1-17968f85a464@weissschuh.net Signed-off-by: Josh Poimboeuf --- tools/objtool/special.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tools/objtool/special.c b/tools/objtool/special.c index 9c8d827f69afb..baa85c31526b3 100644 --- a/tools/objtool/special.c +++ b/tools/objtool/special.c @@ -26,7 +26,7 @@ struct special_entry { unsigned char key; /* jump_label key */ }; -struct special_entry entries[] = { +static const struct special_entry entries[] = { { .sec = ".altinstructions", .group = true, @@ -65,7 +65,7 @@ static void reloc_to_sec_off(struct reloc *reloc, struct section **sec, *off = reloc->sym->offset + reloc->addend; } -static int get_alt_entry(struct elf *elf, struct special_entry *entry, +static int get_alt_entry(struct elf *elf, const struct special_entry *entry, struct section *sec, int idx, struct special_alt *alt) { @@ -139,7 +139,7 @@ static int get_alt_entry(struct elf *elf, struct special_entry *entry, */ int special_get_alts(struct elf *elf, struct list_head *alts) { - struct special_entry *entry; + const struct special_entry *entry; struct section *sec; unsigned int nr_entries; struct special_alt *alt; From cfd66e81799f4a2fdc6447fa99bdb1871f45ff08 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Tue, 27 Dec 2022 16:00:58 +0000 Subject: [PATCH 089/765] objtool: Make struct check_options static MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit It is not used outside of builtin-check.c. Also remove the unused declaration from builtin.h . Signed-off-by: Thomas Weißschuh Link: https://lore.kernel.org/r/20221216-objtool-memory-v2-2-17968f85a464@weissschuh.net Signed-off-by: Josh Poimboeuf --- tools/objtool/builtin-check.c | 2 +- tools/objtool/include/objtool/builtin.h | 2 -- 2 files changed, 1 insertion(+), 3 deletions(-) diff --git a/tools/objtool/builtin-check.c b/tools/objtool/builtin-check.c index a4f39407bf59a..7c175198d09fc 100644 --- a/tools/objtool/builtin-check.c +++ b/tools/objtool/builtin-check.c @@ -65,7 +65,7 @@ static int parse_hacks(const struct option *opt, const char *str, int unset) return found ? 0 : -1; } -const struct option check_options[] = { +static const struct option check_options[] = { OPT_GROUP("Actions:"), OPT_CALLBACK_OPTARG('h', "hacks", NULL, NULL, "jump_label,noinstr,skylake", "patch toolchain bugs/limitations", parse_hacks), OPT_BOOLEAN('i', "ibt", &opts.ibt, "validate and annotate IBT"), diff --git a/tools/objtool/include/objtool/builtin.h b/tools/objtool/include/objtool/builtin.h index fa45044e38630..2a108e648b7a6 100644 --- a/tools/objtool/include/objtool/builtin.h +++ b/tools/objtool/include/objtool/builtin.h @@ -7,8 +7,6 @@ #include -extern const struct option check_options[]; - struct opts { /* actions: */ bool dump_orc; From 8045b8f0b17edf375849f83c80dd05194850b6ed Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Tue, 27 Dec 2022 16:00:59 +0000 Subject: [PATCH 090/765] objtool: Allocate multiple structures with calloc() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit By using calloc() instead of malloc() in a loop, libc does not have to keep around bookkeeping information for each single structure. This reduces maximum memory usage while processing vmlinux.o from 3153325 KB to 3035668 KB (-3.7%) on my notebooks "localmodconfig". Note this introduces memory leaks, because some additional structs get added to the lists later after reading the symbols and sections from the original object. Luckily we don't really care about memory leaks in objtool. Signed-off-by: Thomas Weißschuh Link: https://lore.kernel.org/r/20221216-objtool-memory-v2-3-17968f85a464@weissschuh.net Signed-off-by: Josh Poimboeuf --- tools/objtool/elf.c | 42 ++++++++++++++--------------- tools/objtool/include/objtool/elf.h | 4 +++ 2 files changed, 25 insertions(+), 21 deletions(-) diff --git a/tools/objtool/elf.c b/tools/objtool/elf.c index 64443a7f4bbf9..6806ce01d9334 100644 --- a/tools/objtool/elf.c +++ b/tools/objtool/elf.c @@ -284,13 +284,13 @@ static int read_sections(struct elf *elf) !elf_alloc_hash(section_name, sections_nr)) return -1; + elf->section_data = calloc(sections_nr, sizeof(*sec)); + if (!elf->section_data) { + perror("calloc"); + return -1; + } for (i = 0; i < sections_nr; i++) { - sec = malloc(sizeof(*sec)); - if (!sec) { - perror("malloc"); - return -1; - } - memset(sec, 0, sizeof(*sec)); + sec = &elf->section_data[i]; INIT_LIST_HEAD(&sec->symbol_list); INIT_LIST_HEAD(&sec->reloc_list); @@ -422,13 +422,13 @@ static int read_symbols(struct elf *elf) !elf_alloc_hash(symbol_name, symbols_nr)) return -1; + elf->symbol_data = calloc(symbols_nr, sizeof(*sym)); + if (!elf->symbol_data) { + perror("calloc"); + return -1; + } for (i = 0; i < symbols_nr; i++) { - sym = malloc(sizeof(*sym)); - if (!sym) { - perror("malloc"); - return -1; - } - memset(sym, 0, sizeof(*sym)); + sym = &elf->symbol_data[i]; sym->idx = i; @@ -918,13 +918,13 @@ static int read_relocs(struct elf *elf) sec->base->reloc = sec; nr_reloc = 0; + sec->reloc_data = calloc(sec->sh.sh_size / sec->sh.sh_entsize, sizeof(*reloc)); + if (!sec->reloc_data) { + perror("calloc"); + return -1; + } for (i = 0; i < sec->sh.sh_size / sec->sh.sh_entsize; i++) { - reloc = malloc(sizeof(*reloc)); - if (!reloc) { - perror("malloc"); - return -1; - } - memset(reloc, 0, sizeof(*reloc)); + reloc = &sec->reloc_data[i]; switch (sec->sh.sh_type) { case SHT_REL: if (read_rel_reloc(sec, i, reloc, &symndx)) @@ -1453,16 +1453,16 @@ void elf_close(struct elf *elf) list_for_each_entry_safe(sym, tmpsym, &sec->symbol_list, list) { list_del(&sym->list); hash_del(&sym->hash); - free(sym); } list_for_each_entry_safe(reloc, tmpreloc, &sec->reloc_list, list) { list_del(&reloc->list); hash_del(&reloc->hash); - free(reloc); } list_del(&sec->list); - free(sec); + free(sec->reloc_data); } + free(elf->symbol_data); + free(elf->section_data); free(elf); } diff --git a/tools/objtool/include/objtool/elf.h b/tools/objtool/include/objtool/elf.h index bb60fd42b46f4..1c90f0ac0d531 100644 --- a/tools/objtool/include/objtool/elf.h +++ b/tools/objtool/include/objtool/elf.h @@ -39,6 +39,7 @@ struct section { char *name; int idx; bool changed, text, rodata, noinstr, init, truncate; + struct reloc *reloc_data; }; struct symbol { @@ -104,6 +105,9 @@ struct elf { struct hlist_head *section_hash; struct hlist_head *section_name_hash; struct hlist_head *reloc_hash; + + struct section *section_data; + struct symbol *symbol_data; }; #define OFFSET_STRIDE_BITS 4 From 21a899f9fc78be6b289ee4627bccadf560930eb5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Tue, 27 Dec 2022 16:01:02 +0000 Subject: [PATCH 091/765] objtool: Optimize layout of struct symbol MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Reduce the size of struct symbol on x86_64 from 208 to 200 bytes. This structure is allocated a lot and never freed. This reduces maximum memory usage while processing vmlinux.o from 2919716 KB to 2917988 KB (-0.5%) on my notebooks "localmodconfig". Signed-off-by: Thomas Weißschuh Link: https://lore.kernel.org/r/20221216-objtool-memory-v2-6-17968f85a464@weissschuh.net Signed-off-by: Josh Poimboeuf --- tools/objtool/include/objtool/elf.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/tools/objtool/include/objtool/elf.h b/tools/objtool/include/objtool/elf.h index 1c90f0ac0d531..ad0024da262b0 100644 --- a/tools/objtool/include/objtool/elf.h +++ b/tools/objtool/include/objtool/elf.h @@ -50,12 +50,11 @@ struct symbol { GElf_Sym sym; struct section *sec; char *name; - unsigned int idx; - unsigned char bind, type; + unsigned int idx, len; unsigned long offset; - unsigned int len; unsigned long __subtree_last; struct symbol *pfunc, *cfunc, *alias; + unsigned char bind, type; u8 uaccess_safe : 1; u8 static_call_tramp : 1; u8 retpoline_thunk : 1; From a20717aca33b1ff133f513721050fe6c3d7f97b5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Tue, 27 Dec 2022 16:01:03 +0000 Subject: [PATCH 092/765] objtool: Optimize layout of struct special_alt MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Reduce the size of struct special_alt from 72 to 64 bytes. Signed-off-by: Thomas Weißschuh Link: https://lore.kernel.org/r/20221216-objtool-memory-v2-7-17968f85a464@weissschuh.net Signed-off-by: Josh Poimboeuf --- tools/objtool/include/objtool/special.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/objtool/include/objtool/special.h b/tools/objtool/include/objtool/special.h index dc4721e190023..86d4af9c5aa9d 100644 --- a/tools/objtool/include/objtool/special.h +++ b/tools/objtool/include/objtool/special.h @@ -19,6 +19,7 @@ struct special_alt { bool skip_orig; bool skip_alt; bool jump_or_nop; + u8 key_addend; struct section *orig_sec; unsigned long orig_off; @@ -27,7 +28,6 @@ struct special_alt { unsigned long new_off; unsigned int orig_len, new_len; /* group only */ - u8 key_addend; }; int special_get_alts(struct elf *elf, struct list_head *alts); From a0a9ad95ddccaefa0f743d0af427cba936b9daac Mon Sep 17 00:00:00 2001 From: ye xingchen Date: Tue, 20 Sep 2022 06:30:53 +0000 Subject: [PATCH 093/765] um: Remove the unneeded result variable Return the value epoll_ctl() directly instead of storing it in another redundant variable. Reported-by: Zeal Robot Signed-off-by: ye xingchen Signed-off-by: Richard Weinberger --- arch/um/os-Linux/irq.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/arch/um/os-Linux/irq.c b/arch/um/os-Linux/irq.c index 98ea910ef87cd..cf7e49c08b210 100644 --- a/arch/um/os-Linux/irq.c +++ b/arch/um/os-Linux/irq.c @@ -127,12 +127,10 @@ int os_mod_epoll_fd(int events, int fd, void *data) int os_del_epoll_fd(int fd) { struct epoll_event event; - int result; /* This is quiet as we use this as IO ON/OFF - so it is often * invoked on a non-existent fd */ - result = epoll_ctl(epollfd, EPOLL_CTL_DEL, fd, &event); - return result; + return epoll_ctl(epollfd, EPOLL_CTL_DEL, fd, &event); } void os_set_ioignore(void) From 28b2bb06a381816c2c111afba4984f8ea2c17e0f Mon Sep 17 00:00:00 2001 From: Yang Li Date: Wed, 21 Sep 2022 09:06:09 +0800 Subject: [PATCH 094/765] um: remove unneeded semicolon while(){}, semicolon do not need to be appended. Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=2237 Reported-by: Abaci Robot Signed-off-by: Yang Li Signed-off-by: Richard Weinberger --- arch/um/drivers/virtio_uml.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/um/drivers/virtio_uml.c b/arch/um/drivers/virtio_uml.c index 588930a0ced17..198aaed81ce60 100644 --- a/arch/um/drivers/virtio_uml.c +++ b/arch/um/drivers/virtio_uml.c @@ -412,7 +412,7 @@ static irqreturn_t vu_req_read_message(struct virtio_uml_device *vu_dev, if (msg.msg.header.flags & VHOST_USER_FLAG_NEED_REPLY) vhost_user_reply(vu_dev, &msg.msg, response); irq_rc = IRQ_HANDLED; - }; + } /* mask EAGAIN as we try non-blocking read until socket is empty */ vu_dev->recv_rc = (rc == -EAGAIN) ? 0 : rc; return irq_rc; From e0820368d0101452a9a35c903a549b7f32a6edff Mon Sep 17 00:00:00 2001 From: "Fabio M. De Francesco" Date: Thu, 13 Oct 2022 18:17:00 +0200 Subject: [PATCH 095/765] hostfs: Replace kmap() with kmap_local_page() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The use of kmap() is being deprecated in favor of kmap_local_page(). There are two main problems with kmap(): (1) It comes with an overhead as the mapping space is restricted and protected by a global lock for synchronization and (2) it also requires global TLB invalidation when the kmap’s pool wraps and it might block when the mapping space is fully utilized until a slot becomes available. With kmap_local_page() the mappings are per thread, CPU local, can take page faults, and can be called from any context (including interrupts). It is faster than kmap() in kernels with HIGHMEM enabled. Furthermore, the tasks can be preempted and, when they are scheduled to run again, the kernel virtual addresses are restored and still valid. Therefore, replace kmap() with kmap_local_page() in hostfs_kern.c, it being the only file with kmap() call sites currently left in fs/hostfs. Cc: "Venkataramanan, Anirudh" Suggested-by: Ira Weiny Signed-off-by: Fabio M. De Francesco Signed-off-by: Richard Weinberger --- fs/hostfs/hostfs_kern.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index 277468783feee..609f692246430 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -412,7 +412,7 @@ static int hostfs_writepage(struct page *page, struct writeback_control *wbc) if (page->index >= end_index) count = inode->i_size & (PAGE_SIZE-1); - buffer = kmap(page); + buffer = kmap_local_page(page); err = write_file(HOSTFS_I(inode)->fd, &base, buffer, count); if (err != count) { @@ -428,9 +428,9 @@ static int hostfs_writepage(struct page *page, struct writeback_control *wbc) err = 0; out: - kunmap(page); - + kunmap_local(buffer); unlock_page(page); + return err; } @@ -441,7 +441,7 @@ static int hostfs_read_folio(struct file *file, struct folio *folio) loff_t start = page_offset(page); int bytes_read, ret = 0; - buffer = kmap(page); + buffer = kmap_local_page(page); bytes_read = read_file(FILE_HOSTFS_I(file)->fd, &start, buffer, PAGE_SIZE); if (bytes_read < 0) { @@ -458,8 +458,9 @@ static int hostfs_read_folio(struct file *file, struct folio *folio) out: flush_dcache_page(page); - kunmap(page); + kunmap_local(buffer); unlock_page(page); + return ret; } @@ -484,9 +485,9 @@ static int hostfs_write_end(struct file *file, struct address_space *mapping, unsigned from = pos & (PAGE_SIZE - 1); int err; - buffer = kmap(page); + buffer = kmap_local_page(page); err = write_file(FILE_HOSTFS_I(file)->fd, &pos, buffer + from, copied); - kunmap(page); + kunmap_local(buffer); if (!PageUptodate(page) && err == PAGE_SIZE) SetPageUptodate(page); From 3271e27bba90f676fa530592ba8ec9fe51938abf Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 18 Oct 2022 12:49:49 +0200 Subject: [PATCH 096/765] um: protect VMA iteration Due to changes in the iteration, there are now lockdep checks indicating that we're missing locking here. Add the missing locking where it's needed. Signed-off-by: Johannes Berg Acked-By: Anton Ivanov Signed-off-by: Richard Weinberger --- arch/um/kernel/tlb.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/um/kernel/tlb.c b/arch/um/kernel/tlb.c index ad449173a1a1c..fa43bcd9ba0b6 100644 --- a/arch/um/kernel/tlb.c +++ b/arch/um/kernel/tlb.c @@ -597,6 +597,8 @@ void force_flush_all(void) struct vm_area_struct *vma; VMA_ITERATOR(vmi, mm, 0); + mmap_read_lock(mm); for_each_vma(vmi, vma) fix_range(mm, vma->vm_start, vma->vm_end, 1); + mmap_read_unlock(mm); } From 8f88c73afe481f93d40801596927e8c0047b6d96 Mon Sep 17 00:00:00 2001 From: Xiang Yang Date: Tue, 15 Nov 2022 15:32:25 +0800 Subject: [PATCH 097/765] um: vector: Fix memory leak in vector_config If the return value of the uml_parse_vector_ifspec function is NULL, we should call kfree(params) to prevent memory leak. Fixes: 49da7e64f33e ("High Performance UML Vector Network Driver") Signed-off-by: Xiang Yang Acked-By: Anton Ivanov Signed-off-by: Richard Weinberger --- arch/um/drivers/vector_kern.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/um/drivers/vector_kern.c b/arch/um/drivers/vector_kern.c index ded7c47d2fbe5..131b7cb295767 100644 --- a/arch/um/drivers/vector_kern.c +++ b/arch/um/drivers/vector_kern.c @@ -767,6 +767,7 @@ static int vector_config(char *str, char **error_out) if (parsed == NULL) { *error_out = "vector_config failed to parse parameters"; + kfree(params); return -EINVAL; } From d119595f873e4cf4c7a03aa9ac00d960772487e1 Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Tue, 22 Nov 2022 11:07:32 +0100 Subject: [PATCH 098/765] um: Switch printk calls to adhere to correct coding style This means having the string literal in one line and using __func__ where appropriate. Signed-off-by: Benjamin Berg Signed-off-by: Richard Weinberger --- arch/um/kernel/exec.c | 4 +- arch/um/os-Linux/skas/mem.c | 19 +++-- arch/um/os-Linux/skas/process.c | 121 ++++++++++++++++---------------- 3 files changed, 69 insertions(+), 75 deletions(-) diff --git a/arch/um/kernel/exec.c b/arch/um/kernel/exec.c index 58938d75871af..827a0d3fa5890 100644 --- a/arch/um/kernel/exec.c +++ b/arch/um/kernel/exec.c @@ -29,8 +29,8 @@ void flush_thread(void) ret = unmap(¤t->mm->context.id, 0, TASK_SIZE, 1, &data); if (ret) { - printk(KERN_ERR "flush_thread - clearing address space failed, " - "err = %d\n", ret); + printk(KERN_ERR "%s - clearing address space failed, err = %d\n", + __func__, ret); force_sig(SIGKILL); } get_safe_registers(current_pt_regs()->regs.gp, diff --git a/arch/um/os-Linux/skas/mem.c b/arch/um/os-Linux/skas/mem.c index 3b4975ee67e2d..953fb10f3f931 100644 --- a/arch/um/os-Linux/skas/mem.c +++ b/arch/um/os-Linux/skas/mem.c @@ -60,8 +60,8 @@ static inline long do_syscall_stub(struct mm_id * mm_idp, void **addr) printk(UM_KERN_ERR "Registers - \n"); for (i = 0; i < MAX_REG_NR; i++) printk(UM_KERN_ERR "\t%d\t0x%lx\n", i, syscall_regs[i]); - panic("do_syscall_stub : PTRACE_SETREGS failed, errno = %d\n", - -n); + panic("%s : PTRACE_SETREGS failed, errno = %d\n", + __func__, -n); } err = ptrace(PTRACE_CONT, pid, 0, 0); @@ -81,20 +81,17 @@ static inline long do_syscall_stub(struct mm_id * mm_idp, void **addr) offset = *((unsigned long *) mm_idp->stack + 1); if (offset) { data = (unsigned long *)(mm_idp->stack + offset - STUB_DATA); - printk(UM_KERN_ERR "do_syscall_stub : ret = %ld, offset = %ld, " - "data = %p\n", ret, offset, data); + printk(UM_KERN_ERR "%s : ret = %ld, offset = %ld, data = %p\n", + __func__, ret, offset, data); syscall = (unsigned long *)((unsigned long)data + data[0]); - printk(UM_KERN_ERR "do_syscall_stub: syscall %ld failed, " - "return value = 0x%lx, expected return value = 0x%lx\n", - syscall[0], ret, syscall[7]); - printk(UM_KERN_ERR " syscall parameters: " - "0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx\n", + printk(UM_KERN_ERR "%s: syscall %ld failed, return value = 0x%lx, expected return value = 0x%lx\n", + __func__, syscall[0], ret, syscall[7]); + printk(UM_KERN_ERR " syscall parameters: 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx\n", syscall[1], syscall[2], syscall[3], syscall[4], syscall[5], syscall[6]); for (n = 1; n < data[0]/sizeof(long); n++) { if (n == 1) - printk(UM_KERN_ERR " additional syscall " - "data:"); + printk(UM_KERN_ERR " additional syscall data:"); if (n % 4 == 1) printk("\n" UM_KERN_ERR " "); printk(" 0x%lx", data[n]); diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c index b24db6017ded6..b1ea53285af1b 100644 --- a/arch/um/os-Linux/skas/process.c +++ b/arch/um/os-Linux/skas/process.c @@ -118,8 +118,8 @@ void wait_stub_done(int pid) err = ptrace(PTRACE_CONT, pid, 0, 0); if (err) { - printk(UM_KERN_ERR "wait_stub_done : continue failed, " - "errno = %d\n", errno); + printk(UM_KERN_ERR "%s : continue failed, errno = %d\n", + __func__, errno); fatal_sigsegv(); } } @@ -130,11 +130,10 @@ void wait_stub_done(int pid) bad_wait: err = ptrace_dump_regs(pid); if (err) - printk(UM_KERN_ERR "Failed to get registers from stub, " - "errno = %d\n", -err); - printk(UM_KERN_ERR "wait_stub_done : failed to wait for SIGTRAP, " - "pid = %d, n = %d, errno = %d, status = 0x%x\n", pid, n, errno, - status); + printk(UM_KERN_ERR "Failed to get registers from stub, errno = %d\n", + -err); + printk(UM_KERN_ERR "%s : failed to wait for SIGTRAP, pid = %d, n = %d, errno = %d, status = 0x%x\n", + __func__, pid, n, errno, status); fatal_sigsegv(); } @@ -195,15 +194,15 @@ static void handle_trap(int pid, struct uml_pt_regs *regs, err = ptrace(PTRACE_POKEUSER, pid, PT_SYSCALL_NR_OFFSET, __NR_getpid); if (err < 0) { - printk(UM_KERN_ERR "handle_trap - nullifying syscall " - "failed, errno = %d\n", errno); + printk(UM_KERN_ERR "%s - nullifying syscall failed, errno = %d\n", + __func__, errno); fatal_sigsegv(); } err = ptrace(PTRACE_SYSCALL, pid, 0, 0); if (err < 0) { - printk(UM_KERN_ERR "handle_trap - continuing to end of " - "syscall failed, errno = %d\n", errno); + printk(UM_KERN_ERR "%s - continuing to end of syscall failed, errno = %d\n", + __func__, errno); fatal_sigsegv(); } @@ -212,11 +211,10 @@ static void handle_trap(int pid, struct uml_pt_regs *regs, (WSTOPSIG(status) != SIGTRAP + 0x80)) { err = ptrace_dump_regs(pid); if (err) - printk(UM_KERN_ERR "Failed to get registers " - "from process, errno = %d\n", -err); - printk(UM_KERN_ERR "handle_trap - failed to wait at " - "end of syscall, errno = %d, status = %d\n", - errno, status); + printk(UM_KERN_ERR "Failed to get registers from process, errno = %d\n", + -err); + printk(UM_KERN_ERR "%s - failed to wait at end of syscall, errno = %d, status = %d\n", + __func__, errno, status); fatal_sigsegv(); } } @@ -256,8 +254,8 @@ static int userspace_tramp(void *stack) addr = mmap64((void *) STUB_CODE, UM_KERN_PAGE_SIZE, PROT_EXEC, MAP_FIXED | MAP_PRIVATE, fd, offset); if (addr == MAP_FAILED) { - printk(UM_KERN_ERR "mapping mmap stub at 0x%lx failed, " - "errno = %d\n", STUB_CODE, errno); + printk(UM_KERN_ERR "mapping mmap stub at 0x%lx failed, errno = %d\n", + STUB_CODE, errno); exit(1); } @@ -267,8 +265,7 @@ static int userspace_tramp(void *stack) UM_KERN_PAGE_SIZE, PROT_READ | PROT_WRITE, MAP_FIXED | MAP_SHARED, fd, offset); if (addr == MAP_FAILED) { - printk(UM_KERN_ERR "mapping segfault stack " - "at 0x%lx failed, errno = %d\n", + printk(UM_KERN_ERR "mapping segfault stack at 0x%lx failed, errno = %d\n", STUB_DATA, errno); exit(1); } @@ -286,8 +283,8 @@ static int userspace_tramp(void *stack) sa.sa_sigaction = (void *) v; sa.sa_restorer = NULL; if (sigaction(SIGSEGV, &sa, NULL) < 0) { - printk(UM_KERN_ERR "userspace_tramp - setting SIGSEGV " - "handler failed - errno = %d\n", errno); + printk(UM_KERN_ERR "%s - setting SIGSEGV handler failed - errno = %d\n", + __func__, errno); exit(1); } } @@ -322,8 +319,8 @@ int start_userspace(unsigned long stub_stack) MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); if (stack == MAP_FAILED) { err = -errno; - printk(UM_KERN_ERR "start_userspace : mmap failed, " - "errno = %d\n", errno); + printk(UM_KERN_ERR "%s : mmap failed, errno = %d\n", + __func__, errno); return err; } @@ -336,8 +333,8 @@ int start_userspace(unsigned long stub_stack) pid = clone(userspace_tramp, (void *) sp, flags, (void *) stub_stack); if (pid < 0) { err = -errno; - printk(UM_KERN_ERR "start_userspace : clone failed, " - "errno = %d\n", errno); + printk(UM_KERN_ERR "%s : clone failed, errno = %d\n", + __func__, errno); return err; } @@ -345,31 +342,31 @@ int start_userspace(unsigned long stub_stack) CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED | __WALL)); if (n < 0) { err = -errno; - printk(UM_KERN_ERR "start_userspace : wait failed, " - "errno = %d\n", errno); + printk(UM_KERN_ERR "%s : wait failed, errno = %d\n", + __func__, errno); goto out_kill; } } while (WIFSTOPPED(status) && (WSTOPSIG(status) == SIGALRM)); if (!WIFSTOPPED(status) || (WSTOPSIG(status) != SIGSTOP)) { err = -EINVAL; - printk(UM_KERN_ERR "start_userspace : expected SIGSTOP, got " - "status = %d\n", status); + printk(UM_KERN_ERR "%s : expected SIGSTOP, got status = %d\n", + __func__, status); goto out_kill; } if (ptrace(PTRACE_OLDSETOPTIONS, pid, NULL, (void *) PTRACE_O_TRACESYSGOOD) < 0) { err = -errno; - printk(UM_KERN_ERR "start_userspace : PTRACE_OLDSETOPTIONS " - "failed, errno = %d\n", errno); + printk(UM_KERN_ERR "%s : PTRACE_OLDSETOPTIONS failed, errno = %d\n", + __func__, errno); goto out_kill; } if (munmap(stack, UM_KERN_PAGE_SIZE) < 0) { err = -errno; - printk(UM_KERN_ERR "start_userspace : munmap failed, " - "errno = %d\n", errno); + printk(UM_KERN_ERR "%s : munmap failed, errno = %d\n", + __func__, errno); goto out_kill; } @@ -403,14 +400,14 @@ void userspace(struct uml_pt_regs *regs, unsigned long *aux_fp_regs) * just kill the process. */ if (ptrace(PTRACE_SETREGS, pid, 0, regs->gp)) { - printk(UM_KERN_ERR "userspace - ptrace set regs " - "failed, errno = %d\n", errno); + printk(UM_KERN_ERR "%s - ptrace set regs failed, errno = %d\n", + __func__, errno); fatal_sigsegv(); } if (put_fp_registers(pid, regs->fp)) { - printk(UM_KERN_ERR "userspace - ptrace set fp regs " - "failed, errno = %d\n", errno); + printk(UM_KERN_ERR "%s - ptrace set fp regs failed, errno = %d\n", + __func__, errno); fatal_sigsegv(); } @@ -421,28 +418,28 @@ void userspace(struct uml_pt_regs *regs, unsigned long *aux_fp_regs) singlestepping(NULL)); if (ptrace(op, pid, 0, 0)) { - printk(UM_KERN_ERR "userspace - ptrace continue " - "failed, op = %d, errno = %d\n", op, errno); + printk(UM_KERN_ERR "%s - ptrace continue failed, op = %d, errno = %d\n", + __func__, op, errno); fatal_sigsegv(); } CATCH_EINTR(err = waitpid(pid, &status, WUNTRACED | __WALL)); if (err < 0) { - printk(UM_KERN_ERR "userspace - wait failed, " - "errno = %d\n", errno); + printk(UM_KERN_ERR "%s - wait failed, errno = %d\n", + __func__, errno); fatal_sigsegv(); } regs->is_user = 1; if (ptrace(PTRACE_GETREGS, pid, 0, regs->gp)) { - printk(UM_KERN_ERR "userspace - PTRACE_GETREGS failed, " - "errno = %d\n", errno); + printk(UM_KERN_ERR "%s - PTRACE_GETREGS failed, errno = %d\n", + __func__, errno); fatal_sigsegv(); } if (get_fp_registers(pid, regs->fp)) { - printk(UM_KERN_ERR "userspace - get_fp_registers failed, " - "errno = %d\n", errno); + printk(UM_KERN_ERR "%s - get_fp_registers failed, errno = %d\n", + __func__, errno); fatal_sigsegv(); } @@ -494,8 +491,8 @@ void userspace(struct uml_pt_regs *regs, unsigned long *aux_fp_regs) unblock_signals_trace(); break; default: - printk(UM_KERN_ERR "userspace - child stopped " - "with signal %d\n", sig); + printk(UM_KERN_ERR "%s - child stopped with signal %d\n", + __func__, sig); fatal_sigsegv(); } pid = userspace_pid[0]; @@ -555,15 +552,15 @@ int copy_context_skas0(unsigned long new_stack, int pid) err = ptrace_setregs(pid, thread_regs); if (err < 0) { err = -errno; - printk(UM_KERN_ERR "copy_context_skas0 : PTRACE_SETREGS " - "failed, pid = %d, errno = %d\n", pid, -err); + printk(UM_KERN_ERR "%s : PTRACE_SETREGS failed, pid = %d, errno = %d\n", + __func__, pid, -err); return err; } err = put_fp_registers(pid, thread_fp_regs); if (err < 0) { - printk(UM_KERN_ERR "copy_context_skas0 : put_fp_registers " - "failed, pid = %d, err = %d\n", pid, err); + printk(UM_KERN_ERR "%s : put_fp_registers failed, pid = %d, err = %d\n", + __func__, pid, err); return err; } @@ -574,8 +571,8 @@ int copy_context_skas0(unsigned long new_stack, int pid) err = ptrace(PTRACE_CONT, pid, 0, 0); if (err) { err = -errno; - printk(UM_KERN_ERR "Failed to continue new process, pid = %d, " - "errno = %d\n", pid, errno); + printk(UM_KERN_ERR "Failed to continue new process, pid = %d, errno = %d\n", + pid, errno); return err; } @@ -583,8 +580,8 @@ int copy_context_skas0(unsigned long new_stack, int pid) pid = data->parent_err; if (pid < 0) { - printk(UM_KERN_ERR "copy_context_skas0 - stub-parent reports " - "error %d\n", -pid); + printk(UM_KERN_ERR "%s - stub-parent reports error %d\n", + __func__, -pid); return pid; } @@ -594,8 +591,8 @@ int copy_context_skas0(unsigned long new_stack, int pid) */ wait_stub_done(pid); if (child_data->child_err != STUB_DATA) { - printk(UM_KERN_ERR "copy_context_skas0 - stub-child %d reports " - "error %ld\n", pid, data->child_err); + printk(UM_KERN_ERR "%s - stub-child %d reports error %ld\n", + __func__, pid, data->child_err); err = data->child_err; goto out_kill; } @@ -603,8 +600,8 @@ int copy_context_skas0(unsigned long new_stack, int pid) if (ptrace(PTRACE_OLDSETOPTIONS, pid, NULL, (void *)PTRACE_O_TRACESYSGOOD) < 0) { err = -errno; - printk(UM_KERN_ERR "copy_context_skas0 : PTRACE_OLDSETOPTIONS " - "failed, errno = %d\n", errno); + printk(UM_KERN_ERR "%s : PTRACE_OLDSETOPTIONS failed, errno = %d\n", + __func__, errno); goto out_kill; } @@ -672,8 +669,8 @@ int start_idle_thread(void *stack, jmp_buf *switch_buf) kmalloc_ok = 0; return 1; default: - printk(UM_KERN_ERR "Bad sigsetjmp return in " - "start_idle_thread - %d\n", n); + printk(UM_KERN_ERR "Bad sigsetjmp return in %s - %d\n", + __func__, n); fatal_sigsegv(); } longjmp(*switch_buf, 1); From d5dbcfe7ee31ccea85a8cbf7ff84a2808d6c8f76 Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Tue, 22 Nov 2022 11:07:33 +0100 Subject: [PATCH 099/765] um: Declare fix_range_common as a static function It is only used within the same file. Signed-off-by: Benjamin Berg Signed-off-by: Richard Weinberger --- arch/um/kernel/tlb.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/um/kernel/tlb.c b/arch/um/kernel/tlb.c index fa43bcd9ba0b6..7d050ab0f78af 100644 --- a/arch/um/kernel/tlb.c +++ b/arch/um/kernel/tlb.c @@ -314,8 +314,8 @@ static inline int update_p4d_range(pgd_t *pgd, unsigned long addr, return ret; } -void fix_range_common(struct mm_struct *mm, unsigned long start_addr, - unsigned long end_addr, int force) +static void fix_range_common(struct mm_struct *mm, unsigned long start_addr, + unsigned long end_addr, int force) { pgd_t *pgd; struct host_vm_change hvc; From 23892d383bee15b64f5463bd7195615734bb2415 Mon Sep 17 00:00:00 2001 From: Yifei Liu Date: Wed, 3 Aug 2022 15:53:12 +0000 Subject: [PATCH 100/765] jffs2: correct logic when creating a hole in jffs2_write_begin Bug description and fix: 1. Write data to a file, say all 1s from offset 0 to 16. 2. Truncate the file to a smaller size, say 8 bytes. 3. Write new bytes (say 2s) from an offset past the original size of the file, say at offset 20, for 4 bytes. This is supposed to create a "hole" in the file, meaning that the bytes from offset 8 (where it was truncated above) up to the new write at offset 20, should all be 0s (zeros). 4. Flush all caches using "echo 3 > /proc/sys/vm/drop_caches" (or unmount and remount) the f/s. 5. Check the content of the file. It is wrong. The 1s that used to be between bytes 9 and 16, before the truncation, have REAPPEARED (they should be 0s). We wrote a script and helper C program to reproduce the bug (reproduce_jffs2_write_begin_issue.sh, write_file.c, and Makefile). We can make them available to anyone. The above example is shown when writing a small file within the same first page. But the bug happens for larger files, as long as steps 1, 2, and 3 above all happen within the same page. The problem was traced to the jffs2_write_begin code, where it goes into an 'if' statement intended to handle writes past the current EOF (i.e., writes that may create a hole). The code computes a 'pageofs' that is the floor of the write position (pos), aligned to the page size boundary. In other words, 'pageofs' will never be larger than 'pos'. The code then sets the internal jffs2_raw_inode->isize to the size of max(current inode size, pageofs) but that is wrong: the new file size should be the 'pos', which is larger than both the current inode size and pageofs. Similarly, the code incorrectly sets the internal jffs2_raw_inode->dsize to the difference between the pageofs minus current inode size; instead it should be the current pos minus the current inode size. Finally, inode->i_size was also set incorrectly. The patch below fixes this bug. The bug was discovered using a new tool for finding f/s bugs using model checking, called MCFS (Model Checking File Systems). Signed-off-by: Yifei Liu Signed-off-by: Erez Zadok Signed-off-by: Manish Adkar Signed-off-by: Richard Weinberger --- fs/jffs2/file.c | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c index 3cf71befa4754..96b0275ce9574 100644 --- a/fs/jffs2/file.c +++ b/fs/jffs2/file.c @@ -137,19 +137,18 @@ static int jffs2_write_begin(struct file *filp, struct address_space *mapping, struct jffs2_inode_info *f = JFFS2_INODE_INFO(inode); struct jffs2_sb_info *c = JFFS2_SB_INFO(inode->i_sb); pgoff_t index = pos >> PAGE_SHIFT; - uint32_t pageofs = index << PAGE_SHIFT; int ret = 0; jffs2_dbg(1, "%s()\n", __func__); - if (pageofs > inode->i_size) { - /* Make new hole frag from old EOF to new page */ + if (pos > inode->i_size) { + /* Make new hole frag from old EOF to new position */ struct jffs2_raw_inode ri; struct jffs2_full_dnode *fn; uint32_t alloc_len; - jffs2_dbg(1, "Writing new hole frag 0x%x-0x%x between current EOF and new page\n", - (unsigned int)inode->i_size, pageofs); + jffs2_dbg(1, "Writing new hole frag 0x%x-0x%x between current EOF and new position\n", + (unsigned int)inode->i_size, (uint32_t)pos); ret = jffs2_reserve_space(c, sizeof(ri), &alloc_len, ALLOC_NORMAL, JFFS2_SUMMARY_INODE_SIZE); @@ -169,10 +168,10 @@ static int jffs2_write_begin(struct file *filp, struct address_space *mapping, ri.mode = cpu_to_jemode(inode->i_mode); ri.uid = cpu_to_je16(i_uid_read(inode)); ri.gid = cpu_to_je16(i_gid_read(inode)); - ri.isize = cpu_to_je32(max((uint32_t)inode->i_size, pageofs)); + ri.isize = cpu_to_je32((uint32_t)pos); ri.atime = ri.ctime = ri.mtime = cpu_to_je32(JFFS2_NOW()); ri.offset = cpu_to_je32(inode->i_size); - ri.dsize = cpu_to_je32(pageofs - inode->i_size); + ri.dsize = cpu_to_je32((uint32_t)pos - inode->i_size); ri.csize = cpu_to_je32(0); ri.compr = JFFS2_COMPR_ZERO; ri.node_crc = cpu_to_je32(crc32(0, &ri, sizeof(ri)-8)); @@ -202,7 +201,7 @@ static int jffs2_write_begin(struct file *filp, struct address_space *mapping, goto out_err; } jffs2_complete_reservation(c); - inode->i_size = pageofs; + inode->i_size = pos; mutex_unlock(&f->sem); } From 1b42b1a36fc946f0d7088425b90d491b4257ca3e Mon Sep 17 00:00:00 2001 From: George Kennedy Date: Tue, 15 Nov 2022 10:14:44 -0500 Subject: [PATCH 101/765] ubi: ensure that VID header offset + VID header size <= alloc, size Ensure that the VID header offset + VID header size does not exceed the allocated area to avoid slab OOB. BUG: KASAN: slab-out-of-bounds in crc32_body lib/crc32.c:111 [inline] BUG: KASAN: slab-out-of-bounds in crc32_le_generic lib/crc32.c:179 [inline] BUG: KASAN: slab-out-of-bounds in crc32_le_base+0x58c/0x626 lib/crc32.c:197 Read of size 4 at addr ffff88802bb36f00 by task syz-executor136/1555 CPU: 2 PID: 1555 Comm: syz-executor136 Tainted: G W 6.0.0-1868 #1 Hardware name: Red Hat KVM, BIOS 1.13.0-2.module+el8.3.0+7860+a7792d29 04/01/2014 Call Trace: __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0x85/0xad lib/dump_stack.c:106 print_address_description mm/kasan/report.c:317 [inline] print_report.cold.13+0xb6/0x6bb mm/kasan/report.c:433 kasan_report+0xa7/0x11b mm/kasan/report.c:495 crc32_body lib/crc32.c:111 [inline] crc32_le_generic lib/crc32.c:179 [inline] crc32_le_base+0x58c/0x626 lib/crc32.c:197 ubi_io_write_vid_hdr+0x1b7/0x472 drivers/mtd/ubi/io.c:1067 create_vtbl+0x4d5/0x9c4 drivers/mtd/ubi/vtbl.c:317 create_empty_lvol drivers/mtd/ubi/vtbl.c:500 [inline] ubi_read_volume_table+0x67b/0x288a drivers/mtd/ubi/vtbl.c:812 ubi_attach+0xf34/0x1603 drivers/mtd/ubi/attach.c:1601 ubi_attach_mtd_dev+0x6f3/0x185e drivers/mtd/ubi/build.c:965 ctrl_cdev_ioctl+0x2db/0x347 drivers/mtd/ubi/cdev.c:1043 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:870 [inline] __se_sys_ioctl fs/ioctl.c:856 [inline] __x64_sys_ioctl+0x193/0x213 fs/ioctl.c:856 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x3e/0x86 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0x0 RIP: 0033:0x7f96d5cf753d Code: RSP: 002b:00007fffd72206f8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f96d5cf753d RDX: 0000000020000080 RSI: 0000000040186f40 RDI: 0000000000000003 RBP: 0000000000400cd0 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000400be0 R13: 00007fffd72207e0 R14: 0000000000000000 R15: 0000000000000000 Allocated by task 1555: kasan_save_stack+0x20/0x3d mm/kasan/common.c:38 kasan_set_track mm/kasan/common.c:45 [inline] set_alloc_info mm/kasan/common.c:437 [inline] ____kasan_kmalloc mm/kasan/common.c:516 [inline] __kasan_kmalloc+0x88/0xa3 mm/kasan/common.c:525 kasan_kmalloc include/linux/kasan.h:234 [inline] __kmalloc+0x138/0x257 mm/slub.c:4429 kmalloc include/linux/slab.h:605 [inline] ubi_alloc_vid_buf drivers/mtd/ubi/ubi.h:1093 [inline] create_vtbl+0xcc/0x9c4 drivers/mtd/ubi/vtbl.c:295 create_empty_lvol drivers/mtd/ubi/vtbl.c:500 [inline] ubi_read_volume_table+0x67b/0x288a drivers/mtd/ubi/vtbl.c:812 ubi_attach+0xf34/0x1603 drivers/mtd/ubi/attach.c:1601 ubi_attach_mtd_dev+0x6f3/0x185e drivers/mtd/ubi/build.c:965 ctrl_cdev_ioctl+0x2db/0x347 drivers/mtd/ubi/cdev.c:1043 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:870 [inline] __se_sys_ioctl fs/ioctl.c:856 [inline] __x64_sys_ioctl+0x193/0x213 fs/ioctl.c:856 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x3e/0x86 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0x0 The buggy address belongs to the object at ffff88802bb36e00 which belongs to the cache kmalloc-256 of size 256 The buggy address is located 0 bytes to the right of 256-byte region [ffff88802bb36e00, ffff88802bb36f00) The buggy address belongs to the physical page: page:00000000ea4d1263 refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x2bb36 head:00000000ea4d1263 order:1 compound_mapcount:0 compound_pincount:0 flags: 0xfffffc0010200(slab|head|node=0|zone=1|lastcpupid=0x1fffff) raw: 000fffffc0010200 ffffea000066c300 dead000000000003 ffff888100042b40 raw: 0000000000000000 0000000000100010 00000001ffffffff 0000000000000000 page dumped because: kasan: bad access detected Memory state around the buggy address: ffff88802bb36e00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ffff88802bb36e80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 >ffff88802bb36f00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc ^ ffff88802bb36f80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc ffff88802bb37000: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ================================================================== Fixes: 801c135ce73d ("UBI: Unsorted Block Images") Reported-by: syzkaller Signed-off-by: George Kennedy Signed-off-by: Richard Weinberger --- drivers/mtd/ubi/build.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index a901f8edfa41d..2178eb4115b36 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -663,6 +663,12 @@ static int io_init(struct ubi_device *ubi, int max_beb_per1024) ubi->ec_hdr_alsize = ALIGN(UBI_EC_HDR_SIZE, ubi->hdrs_min_io_size); ubi->vid_hdr_alsize = ALIGN(UBI_VID_HDR_SIZE, ubi->hdrs_min_io_size); + if (ubi->vid_hdr_offset && ((ubi->vid_hdr_offset + UBI_VID_HDR_SIZE) > + ubi->vid_hdr_alsize)) { + ubi_err(ubi, "VID header offset %d too large.", ubi->vid_hdr_offset); + return -EINVAL; + } + dbg_gen("min_io_size %d", ubi->min_io_size); dbg_gen("max_write_size %d", ubi->max_write_size); dbg_gen("hdrs_min_io_size %d", ubi->hdrs_min_io_size); From aa6d148e6d6270274e3d5a529b71c54cd329d17f Mon Sep 17 00:00:00 2001 From: Li Hua Date: Mon, 21 Nov 2022 19:18:47 +0800 Subject: [PATCH 102/765] ubifs: Fix build errors as symbol undefined With CONFIG_UBIFS_FS_AUTHENTICATION not set, the compiler can assume that ubifs_node_check_hash() is never true and drops the call to ubifs_bad_hash(). Is CONFIG_CC_OPTIMIZE_FOR_SIZE enabled this optimization does not happen anymore. So When CONFIG_UBIFS_FS and CONFIG_CC_OPTIMIZE_FOR_SIZE is enabled but CONFIG_UBIFS_FS_AUTHENTICATION is not set, the build errors is as followd: ERROR: modpost: "ubifs_bad_hash" [fs/ubifs/ubifs.ko] undefined! Fix it by add no-op ubifs_bad_hash() for the CONFIG_UBIFS_FS_AUTHENTICATION=n case. Fixes: 16a26b20d2af ("ubifs: authentication: Add hashes to index nodes") Signed-off-by: Li Hua Reviewed-by: Sascha Hauer Signed-off-by: Richard Weinberger --- fs/ubifs/ubifs.h | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/fs/ubifs/ubifs.h b/fs/ubifs/ubifs.h index 478bbbb5382f8..2f1f315810949 100644 --- a/fs/ubifs/ubifs.h +++ b/fs/ubifs/ubifs.h @@ -1623,8 +1623,13 @@ static inline int ubifs_check_hmac(const struct ubifs_info *c, return crypto_memneq(expected, got, c->hmac_desc_len); } +#ifdef CONFIG_UBIFS_FS_AUTHENTICATION void ubifs_bad_hash(const struct ubifs_info *c, const void *node, const u8 *hash, int lnum, int offs); +#else +static inline void ubifs_bad_hash(const struct ubifs_info *c, const void *node, + const u8 *hash, int lnum, int offs) {}; +#endif int __ubifs_node_check_hash(const struct ubifs_info *c, const void *buf, const u8 *expected); From 203a55f04f66eea1a1ca7e5a302a7f5c99c62327 Mon Sep 17 00:00:00 2001 From: Liu Shixin Date: Thu, 20 Oct 2022 12:30:31 +0800 Subject: [PATCH 103/765] ubifs: Fix memory leak in ubifs_sysfs_init() When insmod ubifs.ko, a kmemleak reported as below: unreferenced object 0xffff88817fb1a780 (size 8): comm "insmod", pid 25265, jiffies 4295239702 (age 100.130s) hex dump (first 8 bytes): 75 62 69 66 73 00 ff ff ubifs... backtrace: [] slab_post_alloc_hook+0x9c/0x3c0 [] __kmalloc_track_caller+0x183/0x410 [] kstrdup+0x3a/0x80 [] kstrdup_const+0x66/0x80 [] kvasprintf_const+0x155/0x190 [] kobject_set_name_vargs+0x5b/0x150 [] kobject_set_name+0xbb/0xf0 [] do_one_initcall+0x14c/0x5a0 [] do_init_module+0x1f0/0x660 [] load_module+0x6d7e/0x7590 [] __do_sys_finit_module+0x19f/0x230 [] __x64_sys_finit_module+0x73/0xb0 [] do_syscall_64+0x35/0x80 [] entry_SYSCALL_64_after_hwframe+0x63/0xcd When kset_register() failed, we should call kset_put to cleanup it. Fixes: 2e3cbf425804 ("ubifs: Export filesystem error counters") Signed-off-by: Liu Shixin Signed-off-by: Richard Weinberger --- fs/ubifs/sysfs.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/ubifs/sysfs.c b/fs/ubifs/sysfs.c index 06ad8fa1fcfb0..54270ad36321e 100644 --- a/fs/ubifs/sysfs.c +++ b/fs/ubifs/sysfs.c @@ -144,6 +144,8 @@ int __init ubifs_sysfs_init(void) kobject_set_name(&ubifs_kset.kobj, "ubifs"); ubifs_kset.kobj.parent = fs_kobj; ret = kset_register(&ubifs_kset); + if (ret) + kset_put(&ubifs_kset); return ret; } From 6addbe91fcccd446036d5acc8e54a9b4395ac4bb Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?M=C3=A5rten=20Lindahl?= Date: Thu, 13 Oct 2022 14:02:49 +0200 Subject: [PATCH 104/765] ubi: block: Reduce warning print to info for static volumes MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If volume size is not multiple of the sector size 512 a warning is printed saying that the last non-sector aligned bytes will be ignored. This should be valid for resizable volumes, but when creating static volumes which are read only this will always be printed even if the unaligned data is deliberate. The message is still valid but the severity should be lowered for static volumes. Signed-off-by: Mårten Lindahl Signed-off-by: Richard Weinberger --- drivers/mtd/ubi/block.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/mtd/ubi/block.c b/drivers/mtd/ubi/block.c index 75eaecc8639f0..d725215ae66e5 100644 --- a/drivers/mtd/ubi/block.c +++ b/drivers/mtd/ubi/block.c @@ -354,9 +354,12 @@ static int calc_disk_capacity(struct ubi_volume_info *vi, u64 *disk_capacity) u64 size = vi->used_bytes >> 9; if (vi->used_bytes % 512) { - pr_warn("UBI: block: volume size is not a multiple of 512, " - "last %llu bytes are ignored!\n", - vi->used_bytes - (size << 9)); + if (vi->vol_type == UBI_DYNAMIC_VOLUME) + pr_warn("UBI: block: volume size is not a multiple of 512, last %llu bytes are ignored!\n", + vi->used_bytes - (size << 9)); + else + pr_info("UBI: block: volume size is not a multiple of 512, last %llu bytes are ignored!\n", + vi->used_bytes - (size << 9)); } if ((sector_t)size != size) From c2c36cc6ca23e614f9e4238d0ecf48549ee9002a Mon Sep 17 00:00:00 2001 From: Zhihao Cheng Date: Tue, 11 Oct 2022 11:47:27 +0800 Subject: [PATCH 105/765] ubifs: Rectify space budget for ubifs_symlink() if symlink is encrypted Fix bad space budget when symlink file is encrypted. Bad space budget may let make_reservation() return with -ENOSPC, which could turn ubifs to read-only mode in do_writepage() process. Fetch a reproducer in [Link]. Link: https://bugzilla.kernel.org/show_bug.cgi?id=216490 Fixes: ca7f85be8d6cf9 ("ubifs: Add support for encrypted symlinks") Signed-off-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- fs/ubifs/dir.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c index 0f29cf2011361..aaff3f3f0aa3b 100644 --- a/fs/ubifs/dir.c +++ b/fs/ubifs/dir.c @@ -1151,7 +1151,6 @@ static int ubifs_symlink(struct user_namespace *mnt_userns, struct inode *dir, int err, sz_change, len = strlen(symname); struct fscrypt_str disk_link; struct ubifs_budget_req req = { .new_ino = 1, .new_dent = 1, - .new_ino_d = ALIGN(len, 8), .dirtied_ino = 1 }; struct fscrypt_name nm; @@ -1167,6 +1166,7 @@ static int ubifs_symlink(struct user_namespace *mnt_userns, struct inode *dir, * Budget request settings: new inode, new direntry and changing parent * directory inode. */ + req.new_ino_d = ALIGN(disk_link.len - 1, 8); err = ubifs_budget_space(c, &req); if (err) return err; From 1b2ba09060e41adb356b9ae58ef94a7390928004 Mon Sep 17 00:00:00 2001 From: Zhihao Cheng Date: Tue, 11 Oct 2022 11:47:28 +0800 Subject: [PATCH 106/765] ubifs: Rectify space budget for ubifs_xrename() There is no space budget for ubifs_xrename(). It may let make_reservation() return with -ENOSPC, which could turn ubifs to read-only mode in do_writepage() process. Fix it by adding space budget for ubifs_xrename(). Fetch a reproducer in [Link]. Link: https://bugzilla.kernel.org/show_bug.cgi?id=216569 Fixes: 9ec64962afb170 ("ubifs: Implement RENAME_EXCHANGE") Signed-off-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- fs/ubifs/dir.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c index aaff3f3f0aa3b..94634b872e382 100644 --- a/fs/ubifs/dir.c +++ b/fs/ubifs/dir.c @@ -1576,6 +1576,10 @@ static int ubifs_xrename(struct inode *old_dir, struct dentry *old_dentry, return err; } + err = ubifs_budget_space(c, &req); + if (err) + goto out; + lock_4_inodes(old_dir, new_dir, NULL, NULL); time = current_time(old_dir); @@ -1601,6 +1605,7 @@ static int ubifs_xrename(struct inode *old_dir, struct dentry *old_dentry, unlock_4_inodes(old_dir, new_dir, NULL, NULL); ubifs_release_budget(c, &req); +out: fscrypt_free_filename(&fst_nm); fscrypt_free_filename(&snd_nm); return err; From c04cc68da856eac2b59e2a421c3f81a2afc560db Mon Sep 17 00:00:00 2001 From: Zhihao Cheng Date: Tue, 11 Oct 2022 11:47:29 +0800 Subject: [PATCH 107/765] ubifs: Add comments and debug info for ubifs_xrename() Just like other operations (eg. ubifs_create, do_rename), add comments and debug information for ubifs_xrename(). Signed-off-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- fs/ubifs/dir.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c index 94634b872e382..7a1dcd14e387d 100644 --- a/fs/ubifs/dir.c +++ b/fs/ubifs/dir.c @@ -1566,6 +1566,15 @@ static int ubifs_xrename(struct inode *old_dir, struct dentry *old_dentry, ubifs_assert(c, fst_inode && snd_inode); + /* + * Budget request settings: changing two direntries, changing the two + * parent directory inodes. + */ + + dbg_gen("dent '%pd' ino %lu in dir ino %lu exchange dent '%pd' ino %lu in dir ino %lu", + old_dentry, fst_inode->i_ino, old_dir->i_ino, + new_dentry, snd_inode->i_ino, new_dir->i_ino); + err = fscrypt_setup_filename(old_dir, &old_dentry->d_name, 0, &fst_nm); if (err) return err; From b248eaf049d9cdc5eb76b59399e4d3de233f02ac Mon Sep 17 00:00:00 2001 From: Zhihao Cheng Date: Tue, 11 Oct 2022 11:47:30 +0800 Subject: [PATCH 108/765] ubifs: Fix wrong dirty space budget for dirty inode Each dirty inode should reserve 'c->bi.inode_budget' bytes in space budget calculation. Currently, space budget for dirty inode reports more space than what UBIFS actually needs to write. Fixes: 1e51764a3c2ac0 ("UBIFS: add new flash file system") Signed-off-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- fs/ubifs/budget.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/ubifs/budget.c b/fs/ubifs/budget.c index e8b9b756f0aca..986e6e4081c76 100644 --- a/fs/ubifs/budget.c +++ b/fs/ubifs/budget.c @@ -400,7 +400,7 @@ static int calc_dd_growth(const struct ubifs_info *c, dd_growth = req->dirtied_page ? c->bi.page_budget : 0; if (req->dirtied_ino) - dd_growth += c->bi.inode_budget << (req->dirtied_ino - 1); + dd_growth += c->bi.inode_budget * req->dirtied_ino; if (req->mod_dent) dd_growth += c->bi.dent_budget; dd_growth += req->dirtied_ino_d; From 25fce616a61fc2f1821e4a9ce212d0e064707093 Mon Sep 17 00:00:00 2001 From: Zhihao Cheng Date: Tue, 11 Oct 2022 11:47:31 +0800 Subject: [PATCH 109/765] ubifs: do_rename: Fix wrong space budget when target inode's nlink > 1 If target inode is a special file (eg. block/char device) with nlink count greater than 1, the inode with ui->data will be re-written on disk. However, UBIFS losts target inode's data_len while doing space budget. Bad space budget may let make_reservation() return with -ENOSPC, which could turn ubifs to read-only mode in do_writepage() process. Fetch a reproducer in [Link]. Link: https://bugzilla.kernel.org/show_bug.cgi?id=216494 Fixes: 1e51764a3c2ac0 ("UBIFS: add new flash file system") Signed-off-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- fs/ubifs/dir.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c index 7a1dcd14e387d..4509d9fa9e6e3 100644 --- a/fs/ubifs/dir.c +++ b/fs/ubifs/dir.c @@ -1324,6 +1324,8 @@ static int do_rename(struct inode *old_dir, struct dentry *old_dentry, if (unlink) { ubifs_assert(c, inode_is_locked(new_inode)); + /* Budget for old inode's data when its nlink > 1. */ + req.dirtied_ino_d = ALIGN(ubifs_inode(new_inode)->data_len, 8); err = ubifs_purge_xattrs(new_inode); if (err) return err; From e874dcde1cbf82c786c0e7f2899811c02630cc52 Mon Sep 17 00:00:00 2001 From: Zhihao Cheng Date: Tue, 11 Oct 2022 11:47:32 +0800 Subject: [PATCH 110/765] ubifs: Reserve one leb for each journal head while doing budget UBIFS calculates available space by c->main_bytes - c->lst.total_used (which means non-index lebs' free and dirty space is accounted into total available), then index lebs and four lebs (one for gc_lnum, one for deletions, two for journal heads) are deducted. In following situation, ubifs may get -ENOSPC from make_reservation(): LEB 84: DATAHD free 122880 used 1920 dirty 2176 dark 6144 LEB 110:DELETION free 126976 used 0 dirty 0 dark 6144 (empty) LEB 201:gc_lnum free 126976 used 0 dirty 0 dark 6144 LEB 272:GCHD free 77824 used 47672 dirty 1480 dark 6144 LEB 356:BASEHD free 0 used 39776 dirty 87200 dark 6144 OTHERS: index lebs, zero-available non-index lebs UBIFS calculates the available bytes is 6888 (How to calculate it: 126976 * 5[remain main bytes] - 1920[used] - 47672[used] - 39776[used] - 126976 * 1[deletions] - 126976 * 1[gc_lnum] - 126976 * 2[journal heads] - 6144 * 5[dark] = 6888) after doing budget, however UBIFS cannot use BASEHD's dirty space(87200), because UBIFS cannot find next BASEHD to reclaim current BASEHD. (c->bi.min_idx_lebs equals to c->lst.idx_lebs, the empty leb won't be found by ubifs_find_free_space(), and dirty index lebs won't be picked as gced lebs. All non-index lebs has dirty space less then c->dead_wm, non-index lebs won't be picked as gced lebs either. So new free lebs won't be produced.). See more details in Link. To fix it, reserve one leb for each journal head while doing budget. Link: https://bugzilla.kernel.org/show_bug.cgi?id=216562 Fixes: 1e51764a3c2ac0 ("UBIFS: add new flash file system") Signed-off-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- fs/ubifs/budget.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/fs/ubifs/budget.c b/fs/ubifs/budget.c index 986e6e4081c76..d76eb7b39f564 100644 --- a/fs/ubifs/budget.c +++ b/fs/ubifs/budget.c @@ -209,11 +209,10 @@ long long ubifs_calc_available(const struct ubifs_info *c, int min_idx_lebs) subtract_lebs += 1; /* - * The GC journal head LEB is not really accessible. And since - * different write types go to different heads, we may count only on - * one head's space. + * Since different write types go to different heads, we should + * reserve one leb for each head. */ - subtract_lebs += c->jhead_cnt - 1; + subtract_lebs += c->jhead_cnt; /* We also reserve one LEB for deletions, which bypass budgeting */ subtract_lebs += 1; From 9af31d6ec1a4be4caab2550096c6bd2ba8fba472 Mon Sep 17 00:00:00 2001 From: Li Zetao Date: Fri, 21 Oct 2022 18:21:56 +0800 Subject: [PATCH 111/765] ubi: Fix use-after-free when volume resizing failed There is an use-after-free problem reported by KASAN: ================================================================== BUG: KASAN: use-after-free in ubi_eba_copy_table+0x11f/0x1c0 [ubi] Read of size 8 at addr ffff888101eec008 by task ubirsvol/4735 CPU: 2 PID: 4735 Comm: ubirsvol Not tainted 6.1.0-rc1-00003-g84fa3304a7fc-dirty #14 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.14.0-1.fc33 04/01/2014 Call Trace: dump_stack_lvl+0x34/0x44 print_report+0x171/0x472 kasan_report+0xad/0x130 ubi_eba_copy_table+0x11f/0x1c0 [ubi] ubi_resize_volume+0x4f9/0xbc0 [ubi] ubi_cdev_ioctl+0x701/0x1850 [ubi] __x64_sys_ioctl+0x11d/0x170 do_syscall_64+0x35/0x80 entry_SYSCALL_64_after_hwframe+0x46/0xb0 When ubi_change_vtbl_record() returns an error in ubi_resize_volume(), "new_eba_tbl" will be freed on error handing path, but it is holded by "vol->eba_tbl" in ubi_eba_replace_table(). It means that the liftcycle of "vol->eba_tbl" and "vol" are different, so when resizing volume in next time, it causing an use-after-free fault. Fix it by not freeing "new_eba_tbl" after it replaced in ubi_eba_replace_table(), while will be freed in next volume resizing. Fixes: 801c135ce73d ("UBI: Unsorted Block Images") Signed-off-by: Li Zetao Reviewed-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- drivers/mtd/ubi/vmt.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c index 8fcc0bdf06358..74637575346e8 100644 --- a/drivers/mtd/ubi/vmt.c +++ b/drivers/mtd/ubi/vmt.c @@ -464,7 +464,7 @@ int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs) for (i = 0; i < -pebs; i++) { err = ubi_eba_unmap_leb(ubi, vol, reserved_pebs + i); if (err) - goto out_acc; + goto out_free; } spin_lock(&ubi->volumes_lock); ubi->rsvd_pebs += pebs; @@ -512,6 +512,8 @@ int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs) ubi->avail_pebs += pebs; spin_unlock(&ubi->volumes_lock); } + return err; + out_free: kfree(new_eba_tbl); return err; From 1e591ea072df7211f64542a09482b5f81cb3ad27 Mon Sep 17 00:00:00 2001 From: Li Zetao Date: Fri, 21 Oct 2022 18:21:57 +0800 Subject: [PATCH 112/765] ubi: Fix unreferenced object reported by kmemleak in ubi_resize_volume() There is a memory leaks problem reported by kmemleak: unreferenced object 0xffff888102007a00 (size 128): comm "ubirsvol", pid 32090, jiffies 4298464136 (age 2361.231s) hex dump (first 32 bytes): ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ backtrace: [] __kmalloc+0x4d/0x150 [] ubi_eba_create_table+0x76/0x170 [ubi] [] ubi_resize_volume+0x1be/0xbc0 [ubi] [] ubi_cdev_ioctl+0x701/0x1850 [ubi] [] __x64_sys_ioctl+0x11d/0x170 [] do_syscall_64+0x35/0x80 [] entry_SYSCALL_64_after_hwframe+0x46/0xb0 This is due to a mismatch between create and destroy interfaces, and in detail that "new_eba_tbl" created by ubi_eba_create_table() but destroyed by kfree(), while will causing "new_eba_tbl->entries" not freed. Fix it by replacing kfree(new_eba_tbl) with ubi_eba_destroy_table(new_eba_tbl) Fixes: 799dca34ac54 ("UBI: hide EBA internals") Signed-off-by: Li Zetao Reviewed-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- drivers/mtd/ubi/vmt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c index 74637575346e8..9fbc64b997cef 100644 --- a/drivers/mtd/ubi/vmt.c +++ b/drivers/mtd/ubi/vmt.c @@ -515,7 +515,7 @@ int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs) return err; out_free: - kfree(new_eba_tbl); + ubi_eba_destroy_table(new_eba_tbl); return err; } From 4a1ff3c5d04b9079b4f768d9a71b51c4af578dd2 Mon Sep 17 00:00:00 2001 From: Li Zetao Date: Sat, 22 Oct 2022 19:52:11 +0800 Subject: [PATCH 113/765] ubifs: Fix memory leak in alloc_wbufs() kmemleak reported a sequence of memory leaks, and show them as following: unreferenced object 0xffff8881575f8400 (size 1024): comm "mount", pid 19625, jiffies 4297119604 (age 20.383s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [] __kmalloc+0x4d/0x150 [] ubifs_mount+0x307b/0x7170 [ubifs] [] legacy_get_tree+0xed/0x1d0 [] vfs_get_tree+0x7d/0x230 [] path_mount+0xdd4/0x17b0 [] __x64_sys_mount+0x1fa/0x270 [] do_syscall_64+0x35/0x80 [] entry_SYSCALL_64_after_hwframe+0x46/0xb0 unreferenced object 0xffff8881798a6e00 (size 512): comm "mount", pid 19677, jiffies 4297121912 (age 37.816s) hex dump (first 32 bytes): 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk backtrace: [] __kmalloc+0x4d/0x150 [] ubifs_wbuf_init+0x52/0x480 [ubifs] [] ubifs_mount+0x31f5/0x7170 [ubifs] [] legacy_get_tree+0xed/0x1d0 [] vfs_get_tree+0x7d/0x230 [] path_mount+0xdd4/0x17b0 [] __x64_sys_mount+0x1fa/0x270 [] do_syscall_64+0x35/0x80 [] entry_SYSCALL_64_after_hwframe+0x46/0xb0 The problem is that the ubifs_wbuf_init() returns an error in the loop which in the alloc_wbufs(), then the wbuf->buf and wbuf->inodes that were successfully alloced before are not freed. Fix it by adding error hanging path in alloc_wbufs() which frees the memory alloced before when ubifs_wbuf_init() returns an error. Fixes: 1e51764a3c2a ("UBIFS: add new flash file system") Signed-off-by: Li Zetao Reviewed-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- fs/ubifs/super.c | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index d0c9a09988bc7..32cb147597960 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -833,7 +833,7 @@ static int alloc_wbufs(struct ubifs_info *c) INIT_LIST_HEAD(&c->jheads[i].buds_list); err = ubifs_wbuf_init(c, &c->jheads[i].wbuf); if (err) - return err; + goto out_wbuf; c->jheads[i].wbuf.sync_callback = &bud_wbuf_callback; c->jheads[i].wbuf.jhead = i; @@ -841,7 +841,7 @@ static int alloc_wbufs(struct ubifs_info *c) c->jheads[i].log_hash = ubifs_hash_get_desc(c); if (IS_ERR(c->jheads[i].log_hash)) { err = PTR_ERR(c->jheads[i].log_hash); - goto out; + goto out_log_hash; } } @@ -854,9 +854,18 @@ static int alloc_wbufs(struct ubifs_info *c) return 0; -out: - while (i--) +out_log_hash: + kfree(c->jheads[i].wbuf.buf); + kfree(c->jheads[i].wbuf.inodes); + +out_wbuf: + while (i--) { + kfree(c->jheads[i].wbuf.buf); + kfree(c->jheads[i].wbuf.inodes); kfree(c->jheads[i].log_hash); + } + kfree(c->jheads); + c->jheads = NULL; return err; } From 7af73882dd1cea5d32ac0cc9eda263635295de50 Mon Sep 17 00:00:00 2001 From: ZhaoLong Wang Date: Tue, 25 Oct 2022 16:52:49 +0800 Subject: [PATCH 114/765] ubi: fastmap: Add fastmap control support for module parameter The UBI driver can use the IOCTL to disable the fastmap after the mainline 669d204469c4 ("ubi: fastmap: Add fastmap control support for 'UBI_IOCATT' ioctl"). To destroy the fastmap on a old image, we need to reattach the device in user space. However, if the UBI driver build in kernel and the UBI volume is the root partition, the UBI device cannot be reattached in user space. To disable fastmap in this case, the UBI must provide the kernel cmdline parameters to disable fastmap during attach. This patch add 'enable_fm' as 5th module init parameter of mtd=xx to control fastmap enable or not. When the value is 0, fastmap will not create and existed fastmap will destroyed for the given ubi device. Default value is 0. To enable or disable fastmap during module loading, fm_autoconvert must be set to non-zero. +-----------------+---------------+---------------------------+ | \ | enable_fm=0 | enable_fm=1 | +-----------------+---------------+---------------------------+ |fm_autoconvert=Y | disable fm | enable fm | +---------------------------------+---------------------------+ |fm_autoconvert=N | disable fm | Enable fastmap if fastmap | | | | exists on the old image | +-------------------------------------------------------------+ Example: # - Attach mtd1 to ubi1, disable fastmap, mtd2 to ubi2, enable fastmap. # modprobe ubi mtd=1,0,0,1,0 mtd=2,0,0,2,1 fm_autoconvert=1 # - If 5th parameter is not specified, the value is 0, fastmap is disable # modprobe ubi mtd=1 fm_autoconvert=1 Link: https://bugzilla.kernel.org/show_bug.cgi?id=216623 Signed-off-by: ZhaoLong Wang Signed-off-by: Richard Weinberger --- drivers/mtd/ubi/build.c | 24 ++++++++++++++++++++---- 1 file changed, 20 insertions(+), 4 deletions(-) diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index 2178eb4115b36..eaf3c0c86b2b7 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -35,7 +35,7 @@ #define MTD_PARAM_LEN_MAX 64 /* Maximum number of comma-separated items in the 'mtd=' parameter */ -#define MTD_PARAM_MAX_COUNT 4 +#define MTD_PARAM_MAX_COUNT 5 /* Maximum value for the number of bad PEBs per 1024 PEBs */ #define MAX_MTD_UBI_BEB_LIMIT 768 @@ -53,12 +53,14 @@ * @ubi_num: UBI number * @vid_hdr_offs: VID header offset * @max_beb_per1024: maximum expected number of bad PEBs per 1024 PEBs + * @enable_fm: enable fastmap when value is non-zero */ struct mtd_dev_param { char name[MTD_PARAM_LEN_MAX]; int ubi_num; int vid_hdr_offs; int max_beb_per1024; + int enable_fm; }; /* Numbers of elements set in the @mtd_dev_param array */ @@ -1254,7 +1256,7 @@ static int __init ubi_init(void) mutex_lock(&ubi_devices_mutex); err = ubi_attach_mtd_dev(mtd, p->ubi_num, p->vid_hdr_offs, p->max_beb_per1024, - false); + p->enable_fm == 0 ? true : false); mutex_unlock(&ubi_devices_mutex); if (err < 0) { pr_err("UBI error: cannot attach mtd%d\n", @@ -1433,7 +1435,7 @@ static int ubi_mtd_param_parse(const char *val, const struct kernel_param *kp) int err = kstrtoint(token, 10, &p->max_beb_per1024); if (err) { - pr_err("UBI error: bad value for max_beb_per1024 parameter: %s", + pr_err("UBI error: bad value for max_beb_per1024 parameter: %s\n", token); return -EINVAL; } @@ -1444,13 +1446,25 @@ static int ubi_mtd_param_parse(const char *val, const struct kernel_param *kp) int err = kstrtoint(token, 10, &p->ubi_num); if (err) { - pr_err("UBI error: bad value for ubi_num parameter: %s", + pr_err("UBI error: bad value for ubi_num parameter: %s\n", token); return -EINVAL; } } else p->ubi_num = UBI_DEV_NUM_AUTO; + token = tokens[4]; + if (token) { + int err = kstrtoint(token, 10, &p->enable_fm); + + if (err) { + pr_err("UBI error: bad value for enable_fm parameter: %s\n", + token); + return -EINVAL; + } + } else + p->enable_fm = 0; + mtd_devs += 1; return 0; } @@ -1463,11 +1477,13 @@ MODULE_PARM_DESC(mtd, "MTD devices to attach. Parameter format: mtd= Date: Mon, 14 Nov 2022 18:26:24 +0800 Subject: [PATCH 115/765] ubi: Fix possible null-ptr-deref in ubi_free_volume() It willl cause null-ptr-deref in the following case: uif_init() ubi_add_volume() cdev_add() -> if it fails, call kill_volumes() device_register() kill_volumes() -> if ubi_add_volume() fails call this function ubi_free_volume() cdev_del() device_unregister() -> trying to delete a not added device, it causes null-ptr-deref So in ubi_free_volume(), it delete devices whether they are added or not, it will causes null-ptr-deref. Handle the error case whlie calling ubi_add_volume() to fix this problem. If add volume fails, set the corresponding vol to null, so it can not be accessed in kill_volumes() and release the resource in ubi_add_volume() error path. Fixes: 801c135ce73d ("UBI: Unsorted Block Images") Suggested-by: Zhihao Cheng Signed-off-by: Yang Yingliang Reviewed-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- drivers/mtd/ubi/build.c | 1 + drivers/mtd/ubi/vmt.c | 12 ++++++------ 2 files changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index eaf3c0c86b2b7..5e90a4423b699 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -470,6 +470,7 @@ static int uif_init(struct ubi_device *ubi) err = ubi_add_volume(ubi, ubi->volumes[i]); if (err) { ubi_err(ubi, "cannot add volume %d", i); + ubi->volumes[i] = NULL; goto out_volumes; } } diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c index 9fbc64b997cef..2c867d16f89f7 100644 --- a/drivers/mtd/ubi/vmt.c +++ b/drivers/mtd/ubi/vmt.c @@ -582,6 +582,7 @@ int ubi_add_volume(struct ubi_device *ubi, struct ubi_volume *vol) if (err) { ubi_err(ubi, "cannot add character device for volume %d, error %d", vol_id, err); + vol_release(&vol->dev); return err; } @@ -592,15 +593,14 @@ int ubi_add_volume(struct ubi_device *ubi, struct ubi_volume *vol) vol->dev.groups = volume_dev_groups; dev_set_name(&vol->dev, "%s_%d", ubi->ubi_name, vol->vol_id); err = device_register(&vol->dev); - if (err) - goto out_cdev; + if (err) { + cdev_del(&vol->cdev); + put_device(&vol->dev); + return err; + } self_check_volumes(ubi); return err; - -out_cdev: - cdev_del(&vol->cdev); - return err; } /** From 2de203d8ab51657664a60f2d4cd89cb7d75b7eac Mon Sep 17 00:00:00 2001 From: ZhaoLong Wang Date: Mon, 14 Nov 2022 22:46:46 +0800 Subject: [PATCH 116/765] ubi: Fix permission display of the debugfs files Some interface files in debugfs support the read method dfs_file_read(), but their rwx permissions is shown as unreadable. In the user mode, the following problem can be clearly seen: # ls -l /sys/kernel/debug/ubi/ubi0/ total 0 --w------- 1 root root 0 Oct 22 16:26 chk_fastmap --w------- 1 root root 0 Oct 22 16:26 chk_gen --w------- 1 root root 0 Oct 22 16:26 chk_io -r-------- 1 root root 0 Oct 22 16:26 detailed_erase_block_info --w------- 1 root root 0 Oct 22 16:26 tst_disable_bgt --w------- 1 root root 0 Oct 22 16:26 tst_emulate_bitflips --w------- 1 root root 0 Oct 22 16:26 tst_emulate_io_failures --w------- 1 root root 0 Oct 22 16:26 tst_emulate_power_cut --w------- 1 root root 0 Oct 22 16:26 tst_emulate_power_cut_max --w------- 1 root root 0 Oct 22 16:26 tst_emulate_power_cut_min It shows that these files do not have read permission 'r', but we can actually read their contents. # echo 1 > /sys/kernel/debug/ubi/ubi0/chk_io # cat /sys/kernel/debug/ubi/ubi0/chk_io 1 User's permission access is determined by capabilities. Of course, the root user is not restricted from reading these files. When reading a debugfs file, the process is as follows: ksys_read() vfs_read() if (file->f_op->read) file->f_op->read() full_proxy_open() real_fops->read() dfs_file_read() -- Read method of debugfs file. else if (file->f_op->read_iter) new_sync_read() else ret = -EINVAL -- Return -EINVAL if no read method. This indicates that the debugfs file can be read as long as the read method of the debugfs file is registered. This patch adds the read permission display for file that support the read method. Signed-off-by: ZhaoLong Wang Signed-off-by: Richard Weinberger --- drivers/mtd/ubi/debug.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) diff --git a/drivers/mtd/ubi/debug.c b/drivers/mtd/ubi/debug.c index fcca6942dbdd0..27168f511d6d4 100644 --- a/drivers/mtd/ubi/debug.c +++ b/drivers/mtd/ubi/debug.c @@ -504,6 +504,7 @@ int ubi_debugfs_init_dev(struct ubi_device *ubi) { unsigned long ubi_num = ubi->ubi_num; struct ubi_debug_info *d = &ubi->dbg; + umode_t mode = S_IRUSR | S_IWUSR; int n; if (!IS_ENABLED(CONFIG_DEBUG_FS)) @@ -518,41 +519,41 @@ int ubi_debugfs_init_dev(struct ubi_device *ubi) d->dfs_dir = debugfs_create_dir(d->dfs_dir_name, dfs_rootdir); - d->dfs_chk_gen = debugfs_create_file("chk_gen", S_IWUSR, d->dfs_dir, + d->dfs_chk_gen = debugfs_create_file("chk_gen", mode, d->dfs_dir, (void *)ubi_num, &dfs_fops); - d->dfs_chk_io = debugfs_create_file("chk_io", S_IWUSR, d->dfs_dir, + d->dfs_chk_io = debugfs_create_file("chk_io", mode, d->dfs_dir, (void *)ubi_num, &dfs_fops); - d->dfs_chk_fastmap = debugfs_create_file("chk_fastmap", S_IWUSR, + d->dfs_chk_fastmap = debugfs_create_file("chk_fastmap", mode, d->dfs_dir, (void *)ubi_num, &dfs_fops); - d->dfs_disable_bgt = debugfs_create_file("tst_disable_bgt", S_IWUSR, + d->dfs_disable_bgt = debugfs_create_file("tst_disable_bgt", mode, d->dfs_dir, (void *)ubi_num, &dfs_fops); d->dfs_emulate_bitflips = debugfs_create_file("tst_emulate_bitflips", - S_IWUSR, d->dfs_dir, + mode, d->dfs_dir, (void *)ubi_num, &dfs_fops); d->dfs_emulate_io_failures = debugfs_create_file("tst_emulate_io_failures", - S_IWUSR, d->dfs_dir, + mode, d->dfs_dir, (void *)ubi_num, &dfs_fops); d->dfs_emulate_power_cut = debugfs_create_file("tst_emulate_power_cut", - S_IWUSR, d->dfs_dir, + mode, d->dfs_dir, (void *)ubi_num, &dfs_fops); d->dfs_power_cut_min = debugfs_create_file("tst_emulate_power_cut_min", - S_IWUSR, d->dfs_dir, + mode, d->dfs_dir, (void *)ubi_num, &dfs_fops); d->dfs_power_cut_max = debugfs_create_file("tst_emulate_power_cut_max", - S_IWUSR, d->dfs_dir, + mode, d->dfs_dir, (void *)ubi_num, &dfs_fops); debugfs_create_file("detailed_erase_block_info", S_IRUSR, d->dfs_dir, From 944e096aa24071d3fe22822f6249d3ae309e39ea Mon Sep 17 00:00:00 2001 From: Zhihao Cheng Date: Fri, 18 Nov 2022 17:02:35 +0800 Subject: [PATCH 117/765] ubifs: Re-statistic cleaned znode count if commit failed Dirty znodes will be written on flash in committing process with following states: process A | znode state ------------------------------------------------------ do_commit | DIRTY_ZNODE ubifs_tnc_start_commit | DIRTY_ZNODE get_znodes_to_commit | DIRTY_ZNODE | COW_ZNODE layout_commit | DIRTY_ZNODE | COW_ZNODE fill_gap | 0 write master | 0 or OBSOLETE_ZNODE process B | znode state ------------------------------------------------------ do_commit | DIRTY_ZNODE[1] ubifs_tnc_start_commit | DIRTY_ZNODE get_znodes_to_commit | DIRTY_ZNODE | COW_ZNODE ubifs_tnc_end_commit | DIRTY_ZNODE | COW_ZNODE write_index | 0 write master | 0 or OBSOLETE_ZNODE[2] or | DIRTY_ZNODE[3] [1] znode is dirtied without concurrent committing process [2] znode is copied up (re-dirtied by other process) before cleaned up in committing process [3] znode is re-dirtied after cleaned up in committing process Currently, the clean znode count is updated in free_obsolete_znodes(), which is called only in normal path. If do_commit failed, clean znode count won't be updated, which triggers a failure ubifs assertion[4] in ubifs_tnc_close(): ubifs_assert_failed [ubifs]: UBIFS assert failed: freed == n [4] Commit 380347e9ca7682 ("UBIFS: Add an assertion for clean_zn_cnt"). Fix it by re-statisticing cleaned znode count in tnc_destroy_cnext(). Fetch a reproducer in [Link]. Link: https://bugzilla.kernel.org/show_bug.cgi?id=216704 Fixes: 1e51764a3c2a ("UBIFS: add new flash file system") Signed-off-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- fs/ubifs/tnc.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/fs/ubifs/tnc.c b/fs/ubifs/tnc.c index 488f3da7a6c6c..2df56bbc68657 100644 --- a/fs/ubifs/tnc.c +++ b/fs/ubifs/tnc.c @@ -3053,6 +3053,21 @@ static void tnc_destroy_cnext(struct ubifs_info *c) cnext = cnext->cnext; if (ubifs_zn_obsolete(znode)) kfree(znode); + else if (!ubifs_zn_cow(znode)) { + /* + * Don't forget to update clean znode count after + * committing failed, because ubifs will check this + * count while closing tnc. Non-obsolete znode could + * be re-dirtied during committing process, so dirty + * flag is untrustable. The flag 'COW_ZNODE' is set + * for each dirty znode before committing, and it is + * cleared as long as the znode become clean, so we + * can statistic clean znode count according to this + * flag. + */ + atomic_long_inc(&c->clean_zn_cnt); + atomic_long_inc(&ubifs_clean_zn_cnt); + } } while (cnext && cnext != c->cnext); } From 122deabfe1428bffe95e2bf364ff8a5059bdf089 Mon Sep 17 00:00:00 2001 From: Zhihao Cheng Date: Fri, 18 Nov 2022 17:02:36 +0800 Subject: [PATCH 118/765] ubifs: dirty_cow_znode: Fix memleak in error handling path Following process will cause a memleak for copied up znode: dirty_cow_znode zn = copy_znode(c, znode); err = insert_old_idx(c, zbr->lnum, zbr->offs); if (unlikely(err)) return ERR_PTR(err); // No one refers to zn. Fix it by adding copied znode back to tnc, then it will be freed by ubifs_destroy_tnc_subtree() while closing tnc. Fetch a reproducer in [Link]. Link: https://bugzilla.kernel.org/show_bug.cgi?id=216705 Fixes: 1e51764a3c2a ("UBIFS: add new flash file system") Signed-off-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- fs/ubifs/tnc.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/fs/ubifs/tnc.c b/fs/ubifs/tnc.c index 2df56bbc68657..2469f72eeaabb 100644 --- a/fs/ubifs/tnc.c +++ b/fs/ubifs/tnc.c @@ -267,11 +267,18 @@ static struct ubifs_znode *dirty_cow_znode(struct ubifs_info *c, if (zbr->len) { err = insert_old_idx(c, zbr->lnum, zbr->offs); if (unlikely(err)) - return ERR_PTR(err); + /* + * Obsolete znodes will be freed by tnc_destroy_cnext() + * or free_obsolete_znodes(), copied up znodes should + * be added back to tnc and freed by + * ubifs_destroy_tnc_subtree(). + */ + goto out; err = add_idx_dirt(c, zbr->lnum, zbr->len); } else err = 0; +out: zbr->znode = zn; zbr->lnum = 0; zbr->offs = 0; From fb8bc4c74ae4526d9489362ab2793a936d072b84 Mon Sep 17 00:00:00 2001 From: Zhihao Cheng Date: Wed, 1 Jun 2022 10:59:59 +0800 Subject: [PATCH 119/765] ubifs: ubifs_writepage: Mark page dirty after writing inode failed There are two states for ubifs writing pages: 1. Dirty, Private 2. Not Dirty, Not Private There is a third possibility which maybe related to [1] that page is private but not dirty caused by following process: PA lock(page) ubifs_write_end attach_page_private // set Private __set_page_dirty_nobuffers // set Dirty unlock(page) write_cache_pages lock(page) clear_page_dirty_for_io(page) // clear Dirty ubifs_writepage write_inode // fail, goto out, following codes are not executed // do_writepage // set_page_writeback // set Writeback // detach_page_private // clear Private // end_page_writeback // clear Writeback out: unlock(page) // Private, Not Dirty PB ksys_fadvise64_64 generic_fadvise invalidate_inode_page // page is neither Dirty nor Writeback invalidate_complete_page // page_has_private is true try_to_release_page ubifs_releasepage ubifs_assert(c, 0) !!! Then we may get following assertion failed: UBIFS error (ubi0:0 pid 1492): ubifs_assert_failed [ubifs]: UBIFS assert failed: 0, in fs/ubifs/file.c:1499 UBIFS warning (ubi0:0 pid 1492): ubifs_ro_mode [ubifs]: switched to read-only mode, error -22 CPU: 2 PID: 1492 Comm: aa Not tainted 5.16.0-rc2-00012-g7bb767dee0ba-dirty Call Trace: dump_stack+0x13/0x1b ubifs_ro_mode+0x54/0x60 [ubifs] ubifs_assert_failed+0x4b/0x80 [ubifs] ubifs_releasepage+0x7e/0x1e0 [ubifs] try_to_release_page+0x57/0xe0 invalidate_inode_page+0xfb/0x130 invalidate_mapping_pagevec+0x12/0x20 generic_fadvise+0x303/0x3c0 vfs_fadvise+0x35/0x40 ksys_fadvise64_64+0x4c/0xb0 Jump [2] to find a reproducer. [1] https://linux-mtd.infradead.narkive.com/NQoBeT1u/patch-rfc-ubifs-fix-assert-failed-in-ubifs-set-page-dirty [2] https://bugzilla.kernel.org/show_bug.cgi?id=215357 Fixes: 1e51764a3c2ac0 ("UBIFS: add new flash file system") Signed-off-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- fs/ubifs/file.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c index f2353dd676ef0..1f429260a85fc 100644 --- a/fs/ubifs/file.c +++ b/fs/ubifs/file.c @@ -1032,7 +1032,7 @@ static int ubifs_writepage(struct page *page, struct writeback_control *wbc) if (page->index >= synced_i_size >> PAGE_SHIFT) { err = inode->i_sb->s_op->write_inode(inode, NULL); if (err) - goto out_unlock; + goto out_redirty; /* * The inode has been written, but the write-buffer has * not been synchronized, so in case of an unclean @@ -1060,11 +1060,17 @@ static int ubifs_writepage(struct page *page, struct writeback_control *wbc) if (i_size > synced_i_size) { err = inode->i_sb->s_op->write_inode(inode, NULL); if (err) - goto out_unlock; + goto out_redirty; } return do_writepage(page, len); - +out_redirty: + /* + * redirty_page_for_writepage() won't call ubifs_dirty_inode() because + * it passes I_DIRTY_PAGES flag while calling __mark_inode_dirty(), so + * there is no need to do space budget for dirty inode. + */ + redirty_page_for_writepage(wbc, page); out_unlock: unlock_page(page); return err; From 66f4742e93523ab2f062d9d9828b3e590bc61536 Mon Sep 17 00:00:00 2001 From: Zhihao Cheng Date: Wed, 1 Jun 2022 11:00:00 +0800 Subject: [PATCH 120/765] ubifs: ubifs_releasepage: Remove ubifs_assert(0) to valid this process There are two states for ubifs writing pages: 1. Dirty, Private 2. Not Dirty, Not Private The normal process cannot go to ubifs_releasepage() which means there exists pages being private but not dirty. Reproducer[1] shows that it could occur (which maybe related to [2]) with following process: PA PB PC lock(page)[PA] ubifs_write_end attach_page_private // set Private __set_page_dirty_nobuffers // set Dirty unlock(page) write_cache_pages[PA] lock(page) clear_page_dirty_for_io(page) // clear Dirty ubifs_writepage do_truncation[PB] truncate_setsize i_size_write(inode, newsize) // newsize = 0 i_size = i_size_read(inode) // i_size = 0 end_index = i_size >> PAGE_SHIFT if (page->index > end_index) goto out // jump out: unlock(page) // Private, Not Dirty generic_fadvise[PC] lock(page) invalidate_inode_page try_to_release_page ubifs_releasepage ubifs_assert(c, 0) // bad assertion! unlock(page) truncate_pagecache[PB] Then we may get following assertion failed: UBIFS error (ubi0:0 pid 1683): ubifs_assert_failed [ubifs]: UBIFS assert failed: 0, in fs/ubifs/file.c:1513 UBIFS warning (ubi0:0 pid 1683): ubifs_ro_mode [ubifs]: switched to read-only mode, error -22 CPU: 2 PID: 1683 Comm: aa Not tainted 5.16.0-rc5-00184-g0bca5994cacc-dirty #308 Call Trace: dump_stack+0x13/0x1b ubifs_ro_mode+0x54/0x60 [ubifs] ubifs_assert_failed+0x4b/0x80 [ubifs] ubifs_releasepage+0x67/0x1d0 [ubifs] try_to_release_page+0x57/0xe0 invalidate_inode_page+0xfb/0x130 __invalidate_mapping_pages+0xb9/0x280 invalidate_mapping_pagevec+0x12/0x20 generic_fadvise+0x303/0x3c0 ksys_fadvise64_64+0x4c/0xb0 [1] https://bugzilla.kernel.org/show_bug.cgi?id=215373 [2] https://linux-mtd.infradead.narkive.com/NQoBeT1u/patch-rfc-ubifs-fix-assert-failed-in-ubifs-set-page-dirty Fixes: 1e51764a3c2ac0 ("UBIFS: add new flash file system") Signed-off-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- fs/ubifs/file.c | 19 ++++++++++++++----- 1 file changed, 14 insertions(+), 5 deletions(-) diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c index 1f429260a85fc..10c1779af9c51 100644 --- a/fs/ubifs/file.c +++ b/fs/ubifs/file.c @@ -1472,14 +1472,23 @@ static bool ubifs_release_folio(struct folio *folio, gfp_t unused_gfp_flags) struct inode *inode = folio->mapping->host; struct ubifs_info *c = inode->i_sb->s_fs_info; - /* - * An attempt to release a dirty page without budgeting for it - should - * not happen. - */ if (folio_test_writeback(folio)) return false; + + /* + * Page is private but not dirty, weird? There is one condition + * making it happened. ubifs_writepage skipped the page because + * page index beyonds isize (for example. truncated by other + * process named A), then the page is invalidated by fadvise64 + * syscall before being truncated by process A. + */ ubifs_assert(c, folio_test_private(folio)); - ubifs_assert(c, 0); + if (folio_test_checked(folio)) + release_new_page_budget(c); + else + release_existing_page_budget(c); + + atomic_long_dec(&c->dirty_pg_cnt); folio_detach_private(folio); folio_clear_checked(folio); return true; From 76f9476ece445a07aeb72df9d896cd563fb5b50f Mon Sep 17 00:00:00 2001 From: Zhihao Cheng Date: Tue, 9 Aug 2022 15:06:19 +0800 Subject: [PATCH 121/765] ubi: fastmap: Fix missed fm_anchor PEB in wear-leveling after disabling fastmap After disabling fastmap(ubi->fm_disabled = 1), fastmap won't be updated, fm_anchor PEB is missed being scheduled for erasing. Besides, fm_anchor PEB may have smallest erase count, it doesn't participate wear-leveling. The difference of erase count between fm_anchor PEB and other PEBs will be larger and larger later on. In which situation fastmap can be disabled? Initially, we have an UBI image with fastmap. Then the image will be atttached without module parameter 'fm_autoconvert', ubi turns to full scanning mode in one random attaching process(eg. bad fastmap caused by powercut), ubi fastmap is disabled since then. Fix it by not getting fm_anchor if fastmap is disabled in ubi_refill_pools(). Fetch a reproducer in [Link]. Link: https://bugzilla.kernel.org/show_bug.cgi?id=216341 Fixes: 4b68bf9a69d22d ("ubi: Select fastmap anchor PEBs considering ...") Signed-off-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- drivers/mtd/ubi/fastmap-wl.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/mtd/ubi/fastmap-wl.c b/drivers/mtd/ubi/fastmap-wl.c index 0ee452275578d..863f571f1adb5 100644 --- a/drivers/mtd/ubi/fastmap-wl.c +++ b/drivers/mtd/ubi/fastmap-wl.c @@ -146,13 +146,15 @@ void ubi_refill_pools(struct ubi_device *ubi) if (ubi->fm_anchor) { wl_tree_add(ubi->fm_anchor, &ubi->free); ubi->free_count++; + ubi->fm_anchor = NULL; } - /* - * All available PEBs are in ubi->free, now is the time to get - * the best anchor PEBs. - */ - ubi->fm_anchor = ubi_wl_get_fm_peb(ubi, 1); + if (!ubi->fm_disabled) + /* + * All available PEBs are in ubi->free, now is the time to get + * the best anchor PEBs. + */ + ubi->fm_anchor = ubi_wl_get_fm_peb(ubi, 1); for (;;) { enough = 0; From a240bc5c43130c6aa50831d7caaa02a1d84e1bce Mon Sep 17 00:00:00 2001 From: Zhihao Cheng Date: Sat, 30 Jul 2022 19:28:37 +0800 Subject: [PATCH 122/765] ubi: Fix UAF wear-leveling entry in eraseblk_count_seq_show() Wear-leveling entry could be freed in error path, which may be accessed again in eraseblk_count_seq_show(), for example: __erase_worker eraseblk_count_seq_show wl = ubi->lookuptbl[*block_number] if (wl) wl_entry_destroy ubi->lookuptbl[e->pnum] = NULL kmem_cache_free(ubi_wl_entry_slab, e) erase_count = wl->ec // UAF! Wear-leveling entry updating/accessing in ubi->lookuptbl should be protected by ubi->wl_lock, fix it by adding ubi->wl_lock to serialize wl entry accessing between wl_entry_destroy() and eraseblk_count_seq_show(). Fetch a reproducer in [Link]. Link: https://bugzilla.kernel.org/show_bug.cgi?id=216305 Fixes: 7bccd12d27b7e3 ("ubi: Add debugfs file for tracking PEB state") Fixes: 801c135ce73d5d ("UBI: Unsorted Block Images") Signed-off-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- drivers/mtd/ubi/wl.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c index 68eb0f21b3fe2..f45df3b773739 100644 --- a/drivers/mtd/ubi/wl.c +++ b/drivers/mtd/ubi/wl.c @@ -890,8 +890,11 @@ static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk, err = do_sync_erase(ubi, e1, vol_id, lnum, 0); if (err) { - if (e2) + if (e2) { + spin_lock(&ubi->wl_lock); wl_entry_destroy(ubi, e2); + spin_unlock(&ubi->wl_lock); + } goto out_ro; } @@ -1130,14 +1133,18 @@ static int __erase_worker(struct ubi_device *ubi, struct ubi_work *wl_wrk) /* Re-schedule the LEB for erasure */ err1 = schedule_erase(ubi, e, vol_id, lnum, 0, false); if (err1) { + spin_lock(&ubi->wl_lock); wl_entry_destroy(ubi, e); + spin_unlock(&ubi->wl_lock); err = err1; goto out_ro; } return err; } + spin_lock(&ubi->wl_lock); wl_entry_destroy(ubi, e); + spin_unlock(&ubi->wl_lock); if (err != -EIO) /* * If this is not %-EIO, we have no idea what to do. Scheduling From 4d57a7333e26040f2b583983e1970d9d460e56b0 Mon Sep 17 00:00:00 2001 From: Zhihao Cheng Date: Mon, 13 Jun 2022 14:59:04 +0800 Subject: [PATCH 123/765] ubi: ubi_wl_put_peb: Fix infinite loop when wear-leveling work failed Following process will trigger an infinite loop in ubi_wl_put_peb(): ubifs_bgt ubi_bgt ubifs_leb_unmap ubi_leb_unmap ubi_eba_unmap_leb ubi_wl_put_peb wear_leveling_worker e1 = rb_entry(rb_first(&ubi->used) e2 = get_peb_for_wl(ubi) ubi_io_read_vid_hdr // return err (flash fault) out_error: ubi->move_from = ubi->move_to = NULL wl_entry_destroy(ubi, e1) ubi->lookuptbl[e->pnum] = NULL retry: e = ubi->lookuptbl[pnum]; // return NULL if (e == ubi->move_from) { // NULL == NULL gets true goto retry; // infinite loop !!! $ top PID USER PR NI VIRT RES SHR S %CPU %MEM COMMAND 7676 root 20 0 0 0 0 R 100.0 0.0 ubifs_bgt0_0 Fix it by: 1) Letting ubi_wl_put_peb() returns directly if wearl leveling entry has been removed from 'ubi->lookuptbl'. 2) Using 'ubi->wl_lock' protecting wl entry deletion to preventing an use-after-free problem for wl entry in ubi_wl_put_peb(). Fetch a reproducer in [Link]. Fixes: 43f9b25a9cdd7b1 ("UBI: bugfix: protect from volume removal") Fixes: ee59ba8b064f692 ("UBI: Fix stale pointers in ubi->lookuptbl") Link: https://bugzilla.kernel.org/show_bug.cgi?id=216111 Signed-off-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- drivers/mtd/ubi/wl.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c index f45df3b773739..9e14319225c97 100644 --- a/drivers/mtd/ubi/wl.c +++ b/drivers/mtd/ubi/wl.c @@ -976,11 +976,11 @@ static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk, spin_lock(&ubi->wl_lock); ubi->move_from = ubi->move_to = NULL; ubi->move_to_put = ubi->wl_scheduled = 0; + wl_entry_destroy(ubi, e1); + wl_entry_destroy(ubi, e2); spin_unlock(&ubi->wl_lock); ubi_free_vid_buf(vidb); - wl_entry_destroy(ubi, e1); - wl_entry_destroy(ubi, e2); out_ro: ubi_ro_mode(ubi); @@ -1260,6 +1260,18 @@ int ubi_wl_put_peb(struct ubi_device *ubi, int vol_id, int lnum, retry: spin_lock(&ubi->wl_lock); e = ubi->lookuptbl[pnum]; + if (!e) { + /* + * This wl entry has been removed for some errors by other + * process (eg. wear leveling worker), corresponding process + * (except __erase_worker, which cannot concurrent with + * ubi_wl_put_peb) will set ubi ro_mode at the same time, + * just ignore this wl entry. + */ + spin_unlock(&ubi->wl_lock); + up_read(&ubi->fm_protect); + return 0; + } if (e == ubi->move_from) { /* * User is putting the physical eraseblock which was selected to From b5dd034f8f4a8405713842fb0a6f35e5062d95b7 Mon Sep 17 00:00:00 2001 From: Jiapeng Chong Date: Tue, 27 Sep 2022 11:57:58 +0800 Subject: [PATCH 124/765] UBI: Fastmap: Fix kernel-doc drivers/mtd/ubi/fastmap.c:104: warning: expecting prototype for new_fm_vhdr(). Prototype was for new_fm_vbuf() instead. Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=2289 Reported-by: Abaci Robot Signed-off-by: Jiapeng Chong Signed-off-by: Richard Weinberger --- drivers/mtd/ubi/fastmap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c index ca2d9efe62c3c..28c8151a0725d 100644 --- a/drivers/mtd/ubi/fastmap.c +++ b/drivers/mtd/ubi/fastmap.c @@ -93,7 +93,7 @@ size_t ubi_calc_fm_size(struct ubi_device *ubi) /** - * new_fm_vhdr - allocate a new volume header for fastmap usage. + * new_fm_vbuf() - allocate a new volume header for fastmap usage. * @ubi: UBI device description object * @vol_id: the VID of the new header * From 415c94532ebb2838dc9e2e5a5e609a6140caa5c5 Mon Sep 17 00:00:00 2001 From: Yang Li Date: Thu, 2 Jun 2022 14:55:56 +0800 Subject: [PATCH 125/765] ubifs: Fix some kernel-doc comments Remove warnings found by running scripts/kernel-doc, which is caused by using 'make W=1'. fs/ubifs/journal.c:1221: warning: Function parameter or member 'old_inode' not described in 'ubifs_jnl_rename' fs/ubifs/journal.c:1221: warning: Function parameter or member 'old_nm' not described in 'ubifs_jnl_rename' fs/ubifs/journal.c:1221: warning: Function parameter or member 'new_inode' not described in 'ubifs_jnl_rename' fs/ubifs/journal.c:1221: warning: Function parameter or member 'new_nm' not described in 'ubifs_jnl_rename' fs/ubifs/journal.c:1221: warning: Function parameter or member 'whiteout' not described in 'ubifs_jnl_rename' fs/ubifs/journal.c:1221: warning: Excess function parameter 'old_dentry' description in 'ubifs_jnl_rename' fs/ubifs/journal.c:1221: warning: Excess function parameter 'new_dentry' description in 'ubifs_jnl_rename' Reported-by: Abaci Robot Signed-off-by: Yang Li Reviewed-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- fs/ubifs/journal.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/fs/ubifs/journal.c b/fs/ubifs/journal.c index d02509920bafa..dc52ac0f4a345 100644 --- a/fs/ubifs/journal.c +++ b/fs/ubifs/journal.c @@ -1201,9 +1201,13 @@ int ubifs_jnl_xrename(struct ubifs_info *c, const struct inode *fst_dir, * ubifs_jnl_rename - rename a directory entry. * @c: UBIFS file-system description object * @old_dir: parent inode of directory entry to rename - * @old_dentry: directory entry to rename + * @old_inode: directory entry's inode to rename + * @old_nm: name of the old directory entry to rename * @new_dir: parent inode of directory entry to rename - * @new_dentry: new directory entry (or directory entry to replace) + * @new_inode: new directory entry's inode (or directory entry's inode to + * replace) + * @new_nm: new name of the new directory entry + * @whiteout: whiteout inode * @sync: non-zero if the write-buffer has to be synchronized * * This function implements the re-name operation which may involve writing up From 422125232f6200f0490b7d7fb97722723eb8876d Mon Sep 17 00:00:00 2001 From: Yang Li Date: Wed, 17 Nov 2021 15:03:04 +0800 Subject: [PATCH 126/765] ubifs: Fix kernel-doc Fix function name in fs/ubifs/io.c kernel-doc comment to remove some warnings found by clang(make W=1 LLVM=1). fs/ubifs/io.c:497: warning: expecting prototype for wbuf_timer_callback(). Prototype was for wbuf_timer_callback_nolock() instead fs/ubifs/io.c:513: warning: expecting prototype for new_wbuf_timer(). Prototype was for new_wbuf_timer_nolock() instead fs/ubifs/io.c:538: warning: expecting prototype for cancel_wbuf_timer(). Prototype was for cancel_wbuf_timer_nolock() instead Reported-by: Abaci Robot Signed-off-by: Yang Li Signed-off-by: Richard Weinberger --- fs/ubifs/io.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/fs/ubifs/io.c b/fs/ubifs/io.c index 1607a3c76681a..01d8eb1703820 100644 --- a/fs/ubifs/io.c +++ b/fs/ubifs/io.c @@ -488,7 +488,7 @@ void ubifs_prep_grp_node(struct ubifs_info *c, void *node, int len, int last) } /** - * wbuf_timer_callback - write-buffer timer callback function. + * wbuf_timer_callback_nolock - write-buffer timer callback function. * @timer: timer data (write-buffer descriptor) * * This function is called when the write-buffer timer expires. @@ -505,7 +505,7 @@ static enum hrtimer_restart wbuf_timer_callback_nolock(struct hrtimer *timer) } /** - * new_wbuf_timer - start new write-buffer timer. + * new_wbuf_timer_nolock - start new write-buffer timer. * @c: UBIFS file-system description object * @wbuf: write-buffer descriptor */ @@ -531,7 +531,7 @@ static void new_wbuf_timer_nolock(struct ubifs_info *c, struct ubifs_wbuf *wbuf) } /** - * cancel_wbuf_timer - cancel write-buffer timer. + * cancel_wbuf_timer_nolock - cancel write-buffer timer. * @wbuf: write-buffer descriptor */ static void cancel_wbuf_timer_nolock(struct ubifs_wbuf *wbuf) From 7198c9c00338287fe364d76bba35b3b10feec3c5 Mon Sep 17 00:00:00 2001 From: Yu Zhe Date: Fri, 18 Nov 2022 16:51:04 +0800 Subject: [PATCH 127/765] jffs2: fix spelling mistake "neccecary"->"necessary" There is a spelling mistake in comment. Fix it. Signed-off-by: Yu Zhe Signed-off-by: Richard Weinberger --- fs/jffs2/fs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c index 66af51c416195..e952b0bd548b6 100644 --- a/fs/jffs2/fs.c +++ b/fs/jffs2/fs.c @@ -403,7 +403,7 @@ int jffs2_do_remount_fs(struct super_block *sb, struct fs_context *fc) /* We stop if it was running, then restart if it needs to. This also catches the case where it was stopped and this is just a remount to restart it. - Flush the writebuffer, if neccecary, else we loose it */ + Flush the writebuffer, if necessary, else we loose it */ if (!sb_rdonly(sb)) { jffs2_stop_garbage_collect_thread(c); mutex_lock(&c->alloc_sem); From d5711ae52d5a548d16e5c177c92c15338ec63624 Mon Sep 17 00:00:00 2001 From: Zhang Xiaoxu Date: Thu, 27 Oct 2022 20:49:05 +0800 Subject: [PATCH 128/765] jffs2: Use function instead of macro when initialize compressors The initialized compressors should be released if one of them initialize fail, this is the pre-patch for fix the problem, use function instead of the macro in jffs2_compressors_init() to simplify the codes, no functional change intended. Signed-off-by: Zhang Xiaoxu Signed-off-by: Richard Weinberger --- fs/jffs2/compr.c | 17 +---------------- fs/jffs2/compr.h | 26 ++++++++++++++++++++------ 2 files changed, 21 insertions(+), 22 deletions(-) diff --git a/fs/jffs2/compr.c b/fs/jffs2/compr.c index 4849a4c9a0e24..afe74c65f1e40 100644 --- a/fs/jffs2/compr.c +++ b/fs/jffs2/compr.c @@ -365,19 +365,12 @@ void jffs2_free_comprbuf(unsigned char *comprbuf, unsigned char *orig) int __init jffs2_compressors_init(void) { /* Registering compressors */ -#ifdef CONFIG_JFFS2_ZLIB jffs2_zlib_init(); -#endif -#ifdef CONFIG_JFFS2_RTIME jffs2_rtime_init(); -#endif -#ifdef CONFIG_JFFS2_RUBIN jffs2_rubinmips_init(); jffs2_dynrubin_init(); -#endif -#ifdef CONFIG_JFFS2_LZO jffs2_lzo_init(); -#endif + /* Setting default compression mode */ #ifdef CONFIG_JFFS2_CMODE_NONE jffs2_compression_mode = JFFS2_COMPR_MODE_NONE; @@ -401,18 +394,10 @@ int __init jffs2_compressors_init(void) int jffs2_compressors_exit(void) { /* Unregistering compressors */ -#ifdef CONFIG_JFFS2_LZO jffs2_lzo_exit(); -#endif -#ifdef CONFIG_JFFS2_RUBIN jffs2_dynrubin_exit(); jffs2_rubinmips_exit(); -#endif -#ifdef CONFIG_JFFS2_RTIME jffs2_rtime_exit(); -#endif -#ifdef CONFIG_JFFS2_ZLIB jffs2_zlib_exit(); -#endif return 0; } diff --git a/fs/jffs2/compr.h b/fs/jffs2/compr.h index 5e91d578f4ed8..3716b6b7924c0 100644 --- a/fs/jffs2/compr.h +++ b/fs/jffs2/compr.h @@ -88,18 +88,32 @@ int jffs2_rubinmips_init(void); void jffs2_rubinmips_exit(void); int jffs2_dynrubin_init(void); void jffs2_dynrubin_exit(void); +#else +static inline int jffs2_rubinmips_init(void) { return 0; } +static inline void jffs2_rubinmips_exit(void) {} +static inline int jffs2_dynrubin_init(void) { return 0; } +static inline void jffs2_dynrubin_exit(void) {} #endif #ifdef CONFIG_JFFS2_RTIME -int jffs2_rtime_init(void); -void jffs2_rtime_exit(void); +extern int jffs2_rtime_init(void); +extern void jffs2_rtime_exit(void); +#else +static inline int jffs2_rtime_init(void) { return 0; } +static inline void jffs2_rtime_exit(void) {} #endif #ifdef CONFIG_JFFS2_ZLIB -int jffs2_zlib_init(void); -void jffs2_zlib_exit(void); +extern int jffs2_zlib_init(void); +extern void jffs2_zlib_exit(void); +#else +static inline int jffs2_zlib_init(void) { return 0; } +static inline void jffs2_zlib_exit(void) {} #endif #ifdef CONFIG_JFFS2_LZO -int jffs2_lzo_init(void); -void jffs2_lzo_exit(void); +extern int jffs2_lzo_init(void); +extern void jffs2_lzo_exit(void); +#else +static inline int jffs2_lzo_init(void) { return 0; } +static inline void jffs2_lzo_exit(void) {} #endif #endif /* __JFFS2_COMPR_H__ */ From 3432e57493c278c1cb48abc2d69c6c055a19723e Mon Sep 17 00:00:00 2001 From: Zhang Xiaoxu Date: Thu, 27 Oct 2022 20:49:06 +0800 Subject: [PATCH 129/765] jffs2: Fix list_del corruption if compressors initialized failed There is a list_del corruption when remove the jffs2 module: list_del corruption, ffffffffa0623e60->next is NULL WARNING: CPU: 6 PID: 6332 at lib/list_debug.c:49 __list_del_entry_valid+0x98/0x130 Modules linked in: jffs2(-) ] CPU: 6 PID: 6332 Comm: rmmod Tainted: G W 6.1.0-rc2+ #5 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.14.0-1.fc33 04/01/2014 RIP: 0010:__list_del_entry_valid+0x98/0x130 ... Call Trace: jffs2_unregister_compressor+0x3e/0xe0 [jffs2] jffs2_zlib_exit+0x11/0x30 [jffs2] jffs2_compressors_exit+0x1e/0x30 [jffs2] exit_jffs2_fs+0x16/0x44f [jffs2] __do_sys_delete_module.constprop.0+0x244/0x370 do_syscall_64+0x35/0x80 entry_SYSCALL_64_after_hwframe+0x46/0xb0 If one of the compressor initialize failed, the module always insert success since jffs2_compressors_init() always return success, then something bad may happen during remove the module. For this scenario, let's insmod failed. Signed-off-by: Zhang Xiaoxu Signed-off-by: Richard Weinberger --- fs/jffs2/compr.c | 33 ++++++++++++++++++++++++++++----- 1 file changed, 28 insertions(+), 5 deletions(-) diff --git a/fs/jffs2/compr.c b/fs/jffs2/compr.c index afe74c65f1e40..764f19dec3f0f 100644 --- a/fs/jffs2/compr.c +++ b/fs/jffs2/compr.c @@ -364,12 +364,24 @@ void jffs2_free_comprbuf(unsigned char *comprbuf, unsigned char *orig) int __init jffs2_compressors_init(void) { + int ret = 0; /* Registering compressors */ - jffs2_zlib_init(); - jffs2_rtime_init(); - jffs2_rubinmips_init(); - jffs2_dynrubin_init(); - jffs2_lzo_init(); + ret = jffs2_zlib_init(); + if (ret) + goto exit; + ret = jffs2_rtime_init(); + if (ret) + goto exit_zlib; + ret = jffs2_rubinmips_init(); + if (ret) + goto exit_rtime; + ret = jffs2_dynrubin_init(); + if (ret) + goto exit_runinmips; + ret = jffs2_lzo_init(); + if (ret) + goto exit_dynrubin; + /* Setting default compression mode */ #ifdef CONFIG_JFFS2_CMODE_NONE @@ -389,6 +401,17 @@ int __init jffs2_compressors_init(void) #endif #endif return 0; + +exit_dynrubin: + jffs2_dynrubin_exit(); +exit_runinmips: + jffs2_rubinmips_exit(); +exit_rtime: + jffs2_rtime_exit(); +exit_zlib: + jffs2_zlib_exit(); +exit: + return ret; } int jffs2_compressors_exit(void) From 5a4fed7cd97ac3d9708982d3801e132e2a36c126 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 19 Jan 2023 07:36:19 +0100 Subject: [PATCH 130/765] f2fs: simplify do_checkpoint For each loop add a local curseg_info pointer insted of looking it up for each of the three fields. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/checkpoint.c | 22 ++++++++++------------ 1 file changed, 10 insertions(+), 12 deletions(-) diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c index d68b3c991888d..7e2b44db15875 100644 --- a/fs/f2fs/checkpoint.c +++ b/fs/f2fs/checkpoint.c @@ -1470,20 +1470,18 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc) ckpt->elapsed_time = cpu_to_le64(get_mtime(sbi, true)); ckpt->free_segment_count = cpu_to_le32(free_segments(sbi)); for (i = 0; i < NR_CURSEG_NODE_TYPE; i++) { - ckpt->cur_node_segno[i] = - cpu_to_le32(curseg_segno(sbi, i + CURSEG_HOT_NODE)); - ckpt->cur_node_blkoff[i] = - cpu_to_le16(curseg_blkoff(sbi, i + CURSEG_HOT_NODE)); - ckpt->alloc_type[i + CURSEG_HOT_NODE] = - curseg_alloc_type(sbi, i + CURSEG_HOT_NODE); + struct curseg_info *curseg = CURSEG_I(sbi, i + CURSEG_HOT_NODE); + + ckpt->cur_node_segno[i] = cpu_to_le32(curseg->segno); + ckpt->cur_node_blkoff[i] = cpu_to_le16(curseg->next_blkoff); + ckpt->alloc_type[i + CURSEG_HOT_NODE] = curseg->alloc_type; } for (i = 0; i < NR_CURSEG_DATA_TYPE; i++) { - ckpt->cur_data_segno[i] = - cpu_to_le32(curseg_segno(sbi, i + CURSEG_HOT_DATA)); - ckpt->cur_data_blkoff[i] = - cpu_to_le16(curseg_blkoff(sbi, i + CURSEG_HOT_DATA)); - ckpt->alloc_type[i + CURSEG_HOT_DATA] = - curseg_alloc_type(sbi, i + CURSEG_HOT_DATA); + struct curseg_info *curseg = CURSEG_I(sbi, i + CURSEG_HOT_DATA); + + ckpt->cur_data_segno[i] = cpu_to_le32(curseg->segno); + ckpt->cur_data_blkoff[i] = cpu_to_le16(curseg->next_blkoff); + ckpt->alloc_type[i + CURSEG_HOT_DATA] = curseg->alloc_type; } /* 2 cp + n data seg summary + orphan inode blocks */ From 6392e9ff8bba228746e37b78b960de6de855fc9d Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 19 Jan 2023 07:36:20 +0100 Subject: [PATCH 131/765] f2fs: add a f2fs_curseg_valid_blocks helper Add a helper to return the valid blocks on log and SSR segments, and replace the last two uses of curseg_blkoff with it. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 32 +++++++++++++++----------------- fs/f2fs/segment.h | 6 ------ 2 files changed, 15 insertions(+), 23 deletions(-) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 39071003c6d73..50af5fcb88e93 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -2339,6 +2339,15 @@ bool f2fs_is_checkpointed_data(struct f2fs_sb_info *sbi, block_t blkaddr) return is_cp; } +static unsigned short f2fs_curseg_valid_blocks(struct f2fs_sb_info *sbi, int type) +{ + struct curseg_info *curseg = CURSEG_I(sbi, type); + + if (sbi->ckpt->alloc_type[type] == SSR) + return sbi->blocks_per_seg; + return curseg->next_blkoff; +} + /* * Calculate the number of current summary pages for writing */ @@ -2348,15 +2357,11 @@ int f2fs_npages_for_summary_flush(struct f2fs_sb_info *sbi, bool for_ra) int i, sum_in_page; for (i = CURSEG_HOT_DATA; i <= CURSEG_COLD_DATA; i++) { - if (sbi->ckpt->alloc_type[i] == SSR) - valid_sum_count += sbi->blocks_per_seg; - else { - if (for_ra) - valid_sum_count += le16_to_cpu( - F2FS_CKPT(sbi)->cur_data_blkoff[i]); - else - valid_sum_count += curseg_blkoff(sbi, i); - } + if (sbi->ckpt->alloc_type[i] != SSR && for_ra) + valid_sum_count += + le16_to_cpu(F2FS_CKPT(sbi)->cur_data_blkoff[i]); + else + valid_sum_count += f2fs_curseg_valid_blocks(sbi, i); } sum_in_page = (PAGE_SIZE - 2 * SUM_JOURNAL_SIZE - @@ -3877,15 +3882,8 @@ static void write_compacted_summaries(struct f2fs_sb_info *sbi, block_t blkaddr) /* Step 3: write summary entries */ for (i = CURSEG_HOT_DATA; i <= CURSEG_COLD_DATA; i++) { - unsigned short blkoff; - seg_i = CURSEG_I(sbi, i); - if (sbi->ckpt->alloc_type[i] == SSR) - blkoff = sbi->blocks_per_seg; - else - blkoff = curseg_blkoff(sbi, i); - - for (j = 0; j < blkoff; j++) { + for (j = 0; j < f2fs_curseg_valid_blocks(sbi, i); j++) { if (!page) { page = f2fs_grab_meta_page(sbi, blkaddr++); kaddr = (unsigned char *)page_address(page); diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h index ad6a9c19f46a4..0f3f05cb8c29f 100644 --- a/fs/f2fs/segment.h +++ b/fs/f2fs/segment.h @@ -710,12 +710,6 @@ static inline unsigned char curseg_alloc_type(struct f2fs_sb_info *sbi, return curseg->alloc_type; } -static inline unsigned short curseg_blkoff(struct f2fs_sb_info *sbi, int type) -{ - struct curseg_info *curseg = CURSEG_I(sbi, type); - return curseg->next_blkoff; -} - static inline void check_seg_range(struct f2fs_sb_info *sbi, unsigned int segno) { f2fs_bug_on(sbi, segno > TOTAL_SEGS(sbi) - 1); From 2df79573ef0215431dc4be6b0d4b084204a0820a Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 19 Jan 2023 07:36:22 +0100 Subject: [PATCH 132/765] f2fs: refactor __allocate_new_segment Simplify the check whether to allocate a new segment or reuse an open one. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 50af5fcb88e93..3712752d395af 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -2913,16 +2913,12 @@ static void __allocate_new_segment(struct f2fs_sb_info *sbi, int type, struct curseg_info *curseg = CURSEG_I(sbi, type); unsigned int old_segno; - if (!curseg->inited) - goto alloc; - - if (force || curseg->next_blkoff || - get_valid_blocks(sbi, curseg->segno, new_sec)) - goto alloc; - - if (!get_ckpt_valid_blocks(sbi, curseg->segno, new_sec)) + if (!force && curseg->inited && + !curseg->next_blkoff && + !get_valid_blocks(sbi, curseg->segno, new_sec) && + !get_ckpt_valid_blocks(sbi, curseg->segno, new_sec)) return; -alloc: + old_segno = curseg->segno; new_curseg(sbi, type, true); stat_inc_seg_type(sbi, curseg); From dede3525edbff6e6244668f81d66a48105d3e43b Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 19 Jan 2023 07:36:23 +0100 Subject: [PATCH 133/765] f2fs: remove __allocate_new_section Just fold this trivial wrapper into the only caller. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 3712752d395af..55720251755ea 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -2925,17 +2925,11 @@ static void __allocate_new_segment(struct f2fs_sb_info *sbi, int type, locate_dirty_segment(sbi, old_segno); } -static void __allocate_new_section(struct f2fs_sb_info *sbi, - int type, bool force) -{ - __allocate_new_segment(sbi, type, true, force); -} - void f2fs_allocate_new_section(struct f2fs_sb_info *sbi, int type, bool force) { f2fs_down_read(&SM_I(sbi)->curseg_lock); down_write(&SIT_I(sbi)->sentry_lock); - __allocate_new_section(sbi, type, force); + __allocate_new_segment(sbi, type, true, force); up_write(&SIT_I(sbi)->sentry_lock); f2fs_up_read(&SM_I(sbi)->curseg_lock); } From 4a2095887340e75dfb07575950fcdb7fbcbec64b Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 19 Jan 2023 07:36:24 +0100 Subject: [PATCH 134/765] f2fs: refactor next blk selection Remove __refresh_next_blkoff by opencoding the SSR vs LFS segment check in the only caller, and then add helpers for SSR block selection and blkoff randomization instead. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 48 +++++++++++++++++++++++------------------------ 1 file changed, 23 insertions(+), 25 deletions(-) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 55720251755ea..14a9651e4f317 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -2632,30 +2632,10 @@ static int __next_free_blkoff(struct f2fs_sb_info *sbi, return __find_rev_next_zero_bit(target_map, sbi->blocks_per_seg, start); } -/* - * If a segment is written by LFS manner, next block offset is just obtained - * by increasing the current block offset. However, if a segment is written by - * SSR manner, next block offset obtained by calling __next_free_blkoff - */ -static void __refresh_next_blkoff(struct f2fs_sb_info *sbi, - struct curseg_info *seg) +static int f2fs_find_next_ssr_block(struct f2fs_sb_info *sbi, + struct curseg_info *seg) { - if (seg->alloc_type == SSR) { - seg->next_blkoff = - __next_free_blkoff(sbi, seg->segno, - seg->next_blkoff + 1); - } else { - seg->next_blkoff++; - if (F2FS_OPTION(sbi).fs_mode == FS_MODE_FRAGMENT_BLK) { - /* To allocate block chunks in different sizes, use random number */ - if (--seg->fragment_remained_chunk <= 0) { - seg->fragment_remained_chunk = - get_random_u32_inclusive(1, sbi->max_fragment_chunk); - seg->next_blkoff += - get_random_u32_inclusive(1, sbi->max_fragment_hole); - } - } - } + return __next_free_blkoff(sbi, seg->segno, seg->next_blkoff + 1); } bool f2fs_segment_has_free_slot(struct f2fs_sb_info *sbi, int segno) @@ -3232,6 +3212,19 @@ static int __get_segment_type(struct f2fs_io_info *fio) return type; } +static void f2fs_randomize_chunk(struct f2fs_sb_info *sbi, + struct curseg_info *seg) +{ + /* To allocate block chunks in different sizes, use random number */ + if (--seg->fragment_remained_chunk > 0) + return; + + seg->fragment_remained_chunk = + get_random_u32_inclusive(1, sbi->max_fragment_chunk); + seg->next_blkoff += + get_random_u32_inclusive(1, sbi->max_fragment_hole); +} + void f2fs_allocate_data_block(struct f2fs_sb_info *sbi, struct page *page, block_t old_blkaddr, block_t *new_blkaddr, struct f2fs_summary *sum, int type, @@ -3261,8 +3254,13 @@ void f2fs_allocate_data_block(struct f2fs_sb_info *sbi, struct page *page, f2fs_wait_discard_bio(sbi, *new_blkaddr); curseg->sum_blk->entries[curseg->next_blkoff] = *sum; - __refresh_next_blkoff(sbi, curseg); - + if (curseg->alloc_type == SSR) { + curseg->next_blkoff = f2fs_find_next_ssr_block(sbi, curseg); + } else { + curseg->next_blkoff++; + if (F2FS_OPTION(sbi).fs_mode == FS_MODE_FRAGMENT_BLK) + f2fs_randomize_chunk(sbi, curseg); + } stat_inc_block_count(sbi, curseg); if (from_gc) { From 88c9edfd3c4cf129d6259085c4cc899051fa1fdc Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 19 Jan 2023 07:36:25 +0100 Subject: [PATCH 135/765] f2fs: remove __has_curseg_space Just open code the logic in the only caller, where it is more obvious. Signed-off-by: Christoph Hellwig Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 19 ++++++++----------- 1 file changed, 8 insertions(+), 11 deletions(-) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 14a9651e4f317..f46a8c0daeb36 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -3087,13 +3087,6 @@ int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range) return err; } -static bool __has_curseg_space(struct f2fs_sb_info *sbi, - struct curseg_info *curseg) -{ - return curseg->next_blkoff < f2fs_usable_blks_in_seg(sbi, - curseg->segno); -} - int f2fs_rw_hint_to_seg_type(enum rw_hint hint) { switch (hint) { @@ -3235,6 +3228,7 @@ void f2fs_allocate_data_block(struct f2fs_sb_info *sbi, struct page *page, unsigned long long old_mtime; bool from_gc = (type == CURSEG_ALL_DATA_ATGC); struct seg_entry *se = NULL; + bool segment_full = false; f2fs_down_read(&SM_I(sbi)->curseg_lock); @@ -3261,6 +3255,8 @@ void f2fs_allocate_data_block(struct f2fs_sb_info *sbi, struct page *page, if (F2FS_OPTION(sbi).fs_mode == FS_MODE_FRAGMENT_BLK) f2fs_randomize_chunk(sbi, curseg); } + if (curseg->next_blkoff >= f2fs_usable_blks_in_seg(sbi, curseg->segno)) + segment_full = true; stat_inc_block_count(sbi, curseg); if (from_gc) { @@ -3279,10 +3275,11 @@ void f2fs_allocate_data_block(struct f2fs_sb_info *sbi, struct page *page, if (GET_SEGNO(sbi, old_blkaddr) != NULL_SEGNO) update_sit_entry(sbi, old_blkaddr, -1); - if (!__has_curseg_space(sbi, curseg)) { - /* - * Flush out current segment and replace it with new segment. - */ + /* + * If the current segment is full, flush it out and replace it with a + * new segment. + */ + if (segment_full) { if (from_gc) { get_atssr_segment(sbi, type, se->type, AT_SSR, se->mtime); From a28bca0f47feb5cdfc22be0e563bd4da2aed74f7 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 19 Jan 2023 07:36:21 +0100 Subject: [PATCH 136/765] f2fs: factor the read/write tracing logic into a helper Factor the logic to log a path for reads and writs into a helper shared between the read_iter and write_iter methods. Signed-off-by: Christoph Hellwig Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 61 +++++++++++++++++++++----------------------------- 1 file changed, 26 insertions(+), 35 deletions(-) diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index eaae2ad3957b0..65199574cd7db 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -4341,6 +4341,27 @@ static ssize_t f2fs_dio_read_iter(struct kiocb *iocb, struct iov_iter *to) return ret; } +static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) +{ + struct inode *inode = file_inode(iocb->ki_filp); + char *buf, *path; + + buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); + if (!buf) + return; + path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); + if (IS_ERR(path)) + goto free_buf; + if (rw == WRITE) + trace_f2fs_datawrite_start(inode, iocb->ki_pos, count, + current->pid, path, current->comm); + else + trace_f2fs_dataread_start(inode, iocb->ki_pos, count, + current->pid, path, current->comm); +free_buf: + kfree(buf); +} + static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) { struct inode *inode = file_inode(iocb->ki_filp); @@ -4350,24 +4371,9 @@ static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) if (!f2fs_is_compress_backend_ready(inode)) return -EOPNOTSUPP; - if (trace_f2fs_dataread_start_enabled()) { - char *p = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); - char *path; - - if (!p) - goto skip_read_trace; + if (trace_f2fs_dataread_start_enabled()) + f2fs_trace_rw_file_path(iocb, iov_iter_count(to), READ); - path = dentry_path_raw(file_dentry(iocb->ki_filp), p, PATH_MAX); - if (IS_ERR(path)) { - kfree(p); - goto skip_read_trace; - } - - trace_f2fs_dataread_start(inode, pos, iov_iter_count(to), - current->pid, path, current->comm); - kfree(p); - } -skip_read_trace: if (f2fs_should_use_dio(inode, iocb, to)) { ret = f2fs_dio_read_iter(iocb, to); } else { @@ -4673,24 +4679,9 @@ static ssize_t f2fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) if (preallocated < 0) { ret = preallocated; } else { - if (trace_f2fs_datawrite_start_enabled()) { - char *p = f2fs_kmalloc(F2FS_I_SB(inode), - PATH_MAX, GFP_KERNEL); - char *path; - - if (!p) - goto skip_write_trace; - path = dentry_path_raw(file_dentry(iocb->ki_filp), - p, PATH_MAX); - if (IS_ERR(path)) { - kfree(p); - goto skip_write_trace; - } - trace_f2fs_datawrite_start(inode, orig_pos, orig_count, - current->pid, path, current->comm); - kfree(p); - } -skip_write_trace: + if (trace_f2fs_datawrite_start_enabled()) + f2fs_trace_rw_file_path(iocb, orig_count, WRITE); + /* Do the actual write. */ ret = dio ? f2fs_dio_write_iter(iocb, from, &may_need_sync) : From 2eae077e6e46f9046d383631145750e043820dce Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Thu, 2 Feb 2023 15:04:56 +0800 Subject: [PATCH 137/765] f2fs: reduce stack memory cost by using bitfield in struct f2fs_io_info This patch tries to use bitfield in struct f2fs_io_info to improve memory usage. struct f2fs_io_info { ... unsigned int need_lock:8; /* indicate we need to lock cp_rwsem */ unsigned int version:8; /* version of the node */ unsigned int submitted:1; /* indicate IO submission */ unsigned int in_list:1; /* indicate fio is in io_list */ unsigned int is_por:1; /* indicate IO is from recovery or not */ unsigned int retry:1; /* need to reallocate block address */ unsigned int encrypted:1; /* indicate file is encrypted */ unsigned int post_read:1; /* require post read */ ... }; After this patch, size of struct f2fs_io_info reduces from 136 to 120. [Nathan: fix a compile warning (single-bit-bitfield-constant-conversion)] Signed-off-by: Nathan Chancellor Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/checkpoint.c | 6 +++--- fs/f2fs/compress.c | 5 +++-- fs/f2fs/data.c | 10 +++++----- fs/f2fs/f2fs.h | 18 +++++++++--------- fs/f2fs/gc.c | 8 ++++---- fs/f2fs/node.c | 2 +- fs/f2fs/segment.c | 6 +++--- 7 files changed, 28 insertions(+), 27 deletions(-) diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c index 7e2b44db15875..89ce08b0ff7c6 100644 --- a/fs/f2fs/checkpoint.c +++ b/fs/f2fs/checkpoint.c @@ -70,7 +70,7 @@ static struct page *__get_meta_page(struct f2fs_sb_info *sbi, pgoff_t index, .old_blkaddr = index, .new_blkaddr = index, .encrypted_page = NULL, - .is_por = !is_meta, + .is_por = !is_meta ? 1 : 0, }; int err; @@ -237,8 +237,8 @@ int f2fs_ra_meta_pages(struct f2fs_sb_info *sbi, block_t start, int nrpages, .op = REQ_OP_READ, .op_flags = sync ? (REQ_META | REQ_PRIO) : REQ_RAHEAD, .encrypted_page = NULL, - .in_list = false, - .is_por = (type == META_POR), + .in_list = 0, + .is_por = (type == META_POR) ? 1 : 0, }; struct blk_plug plug; int err; diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c index b196b881f3bbc..a469cdde6bace 100644 --- a/fs/f2fs/compress.c +++ b/fs/f2fs/compress.c @@ -1213,10 +1213,11 @@ static int f2fs_write_compressed_pages(struct compress_ctx *cc, .page = NULL, .encrypted_page = NULL, .compressed_page = NULL, - .submitted = false, + .submitted = 0, .io_type = io_type, .io_wbc = wbc, - .encrypted = fscrypt_inode_uses_fs_layer_crypto(cc->inode), + .encrypted = fscrypt_inode_uses_fs_layer_crypto(cc->inode) ? + 1 : 0, }; struct dnode_of_data dn; struct node_info ni; diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index c940da1c540f5..754841bce389f 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -992,7 +992,7 @@ void f2fs_submit_page_write(struct f2fs_io_info *fio) bio_page = fio->page; /* set submitted = true as a return value */ - fio->submitted = true; + fio->submitted = 1; inc_page_count(sbi, WB_DATA_TYPE(bio_page)); @@ -1008,7 +1008,7 @@ void f2fs_submit_page_write(struct f2fs_io_info *fio) (fio->type == DATA || fio->type == NODE) && fio->new_blkaddr & F2FS_IO_SIZE_MASK(sbi)) { dec_page_count(sbi, WB_DATA_TYPE(bio_page)); - fio->retry = true; + fio->retry = 1; goto skip; } io->bio = __bio_alloc(fio, BIO_MAX_VECS); @@ -2776,10 +2776,10 @@ int f2fs_write_single_data_page(struct page *page, int *submitted, .old_blkaddr = NULL_ADDR, .page = page, .encrypted_page = NULL, - .submitted = false, + .submitted = 0, .compr_blocks = compr_blocks, .need_lock = LOCK_RETRY, - .post_read = f2fs_post_read_required(inode), + .post_read = f2fs_post_read_required(inode) ? 1 : 0, .io_type = io_type, .io_wbc = wbc, .bio = bio, @@ -2900,7 +2900,7 @@ int f2fs_write_single_data_page(struct page *page, int *submitted, } if (submitted) - *submitted = fio.submitted ? 1 : 0; + *submitted = fio.submitted; return 0; diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index f3c5f7740c1a0..5449c82773398 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -1210,19 +1210,19 @@ struct f2fs_io_info { struct page *encrypted_page; /* encrypted page */ struct page *compressed_page; /* compressed page */ struct list_head list; /* serialize IOs */ - bool submitted; /* indicate IO submission */ - int need_lock; /* indicate we need to lock cp_rwsem */ - bool in_list; /* indicate fio is in io_list */ - bool is_por; /* indicate IO is from recovery or not */ - bool retry; /* need to reallocate block address */ - int compr_blocks; /* # of compressed block addresses */ - bool encrypted; /* indicate file is encrypted */ - bool post_read; /* require post read */ + unsigned int compr_blocks; /* # of compressed block addresses */ + unsigned int need_lock:8; /* indicate we need to lock cp_rwsem */ + unsigned int version:8; /* version of the node */ + unsigned int submitted:1; /* indicate IO submission */ + unsigned int in_list:1; /* indicate fio is in io_list */ + unsigned int is_por:1; /* indicate IO is from recovery or not */ + unsigned int retry:1; /* need to reallocate block address */ + unsigned int encrypted:1; /* indicate file is encrypted */ + unsigned int post_read:1; /* require post read */ enum iostat_type io_type; /* io type */ struct writeback_control *io_wbc; /* writeback control */ struct bio **bio; /* bio for ipu */ sector_t *last_block; /* last block number in bio */ - unsigned char version; /* version of the node */ }; struct bio_entry { diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index 97e846263c7c0..0a9dfa4598606 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -1156,8 +1156,8 @@ static int ra_data_block(struct inode *inode, pgoff_t index) .op = REQ_OP_READ, .op_flags = 0, .encrypted_page = NULL, - .in_list = false, - .retry = false, + .in_list = 0, + .retry = 0, }; int err; @@ -1245,8 +1245,8 @@ static int move_data_block(struct inode *inode, block_t bidx, .op = REQ_OP_READ, .op_flags = 0, .encrypted_page = NULL, - .in_list = false, - .retry = false, + .in_list = 0, + .retry = 0, }; struct dnode_of_data dn; struct f2fs_summary sum; diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c index fbd1d25fecc22..19a1fee88a363 100644 --- a/fs/f2fs/node.c +++ b/fs/f2fs/node.c @@ -1586,7 +1586,7 @@ static int __write_node_page(struct page *page, bool atomic, bool *submitted, .op_flags = wbc_to_write_flags(wbc), .page = page, .encrypted_page = NULL, - .submitted = false, + .submitted = 0, .io_type = io_type, .io_wbc = wbc, }; diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index f46a8c0daeb36..69b01b5c0ce0a 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -3314,10 +3314,10 @@ void f2fs_allocate_data_block(struct f2fs_sb_info *sbi, struct page *page, struct f2fs_bio_info *io; if (F2FS_IO_ALIGNED(sbi)) - fio->retry = false; + fio->retry = 0; INIT_LIST_HEAD(&fio->list); - fio->in_list = true; + fio->in_list = 1; io = sbi->write_io[fio->type] + fio->temp; spin_lock(&io->io_lock); list_add_tail(&fio->list, &io->io_list); @@ -3398,7 +3398,7 @@ void f2fs_do_write_meta_page(struct f2fs_sb_info *sbi, struct page *page, .new_blkaddr = page->index, .page = page, .encrypted_page = NULL, - .in_list = false, + .in_list = 0, }; if (unlikely(page->index >= MAIN_BLKADDR(sbi))) From b90e5086df6bf5ba819216d5ecf0667370bd565f Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Sat, 28 Jan 2023 18:30:11 +0800 Subject: [PATCH 138/765] f2fs: clean up i_compress_flag and i_compress_level usage .i_compress_level was introduced by commit 3fde13f817e2 ("f2fs: compress: support compress level"), but never be used. This patch updates as below: - load high 8-bits of on-disk .i_compress_flag to in-memory .i_compress_level - load low 8-bits of on-disk .i_compress_flag to in-memory .i_compress_flag - change type of in-memory .i_compress_flag from unsigned short to unsigned char. w/ above changes, we can avoid unneeded bit shift whenever during .init_compress_ctx(), and shrink size of struct f2fs_inode_info. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/compress.c | 8 +++----- fs/f2fs/f2fs.h | 7 +++---- fs/f2fs/inode.c | 16 +++++++++++++--- 3 files changed, 19 insertions(+), 12 deletions(-) diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c index a469cdde6bace..e4851f7a43d8d 100644 --- a/fs/f2fs/compress.c +++ b/fs/f2fs/compress.c @@ -241,7 +241,7 @@ static int lz4_init_compress_ctx(struct compress_ctx *cc) unsigned int size = LZ4_MEM_COMPRESS; #ifdef CONFIG_F2FS_FS_LZ4HC - if (F2FS_I(cc->inode)->i_compress_flag >> COMPRESS_LEVEL_OFFSET) + if (F2FS_I(cc->inode)->i_compress_level) size = LZ4HC_MEM_COMPRESS; #endif @@ -267,8 +267,7 @@ static void lz4_destroy_compress_ctx(struct compress_ctx *cc) #ifdef CONFIG_F2FS_FS_LZ4HC static int lz4hc_compress_pages(struct compress_ctx *cc) { - unsigned char level = F2FS_I(cc->inode)->i_compress_flag >> - COMPRESS_LEVEL_OFFSET; + unsigned char level = F2FS_I(cc->inode)->i_compress_level; int len; if (level) @@ -340,8 +339,7 @@ static int zstd_init_compress_ctx(struct compress_ctx *cc) zstd_cstream *stream; void *workspace; unsigned int workspace_size; - unsigned char level = F2FS_I(cc->inode)->i_compress_flag >> - COMPRESS_LEVEL_OFFSET; + unsigned char level = F2FS_I(cc->inode)->i_compress_level; if (!level) level = F2FS_ZSTD_DEFAULT_CLEVEL; diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 5449c82773398..e80144384acbe 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -869,7 +869,7 @@ struct f2fs_inode_info { unsigned char i_compress_algorithm; /* algorithm type */ unsigned char i_log_cluster_size; /* log of cluster size */ unsigned char i_compress_level; /* compress level (lz4hc,zstd) */ - unsigned short i_compress_flag; /* compress flag */ + unsigned char i_compress_flag; /* compress flag */ unsigned int i_cluster_size; /* cluster size */ unsigned int atomic_write_cnt; @@ -4358,9 +4358,8 @@ static inline int set_compress_context(struct inode *inode) if ((F2FS_I(inode)->i_compress_algorithm == COMPRESS_LZ4 || F2FS_I(inode)->i_compress_algorithm == COMPRESS_ZSTD) && F2FS_OPTION(sbi).compress_level) - F2FS_I(inode)->i_compress_flag |= - F2FS_OPTION(sbi).compress_level << - COMPRESS_LEVEL_OFFSET; + F2FS_I(inode)->i_compress_level = + F2FS_OPTION(sbi).compress_level; F2FS_I(inode)->i_flags |= F2FS_COMPR_FL; set_inode_flag(inode, FI_COMPRESSED_FILE); stat_inc_compr_inode(inode); diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c index fd2ee55549858..61a991b490ee7 100644 --- a/fs/f2fs/inode.c +++ b/fs/f2fs/inode.c @@ -460,11 +460,17 @@ static int do_read_inode(struct inode *inode) (fi->i_flags & F2FS_COMPR_FL)) { if (F2FS_FITS_IN_INODE(ri, fi->i_extra_isize, i_log_cluster_size)) { + unsigned short compress_flag; + atomic_set(&fi->i_compr_blocks, le64_to_cpu(ri->i_compr_blocks)); fi->i_compress_algorithm = ri->i_compress_algorithm; fi->i_log_cluster_size = ri->i_log_cluster_size; - fi->i_compress_flag = le16_to_cpu(ri->i_compress_flag); + compress_flag = le16_to_cpu(ri->i_compress_flag); + fi->i_compress_level = compress_flag >> + COMPRESS_LEVEL_OFFSET; + fi->i_compress_flag = compress_flag & + (BIT(COMPRESS_LEVEL_OFFSET) - 1); fi->i_cluster_size = 1 << fi->i_log_cluster_size; set_inode_flag(inode, FI_COMPRESSED_FILE); } @@ -686,13 +692,17 @@ void f2fs_update_inode(struct inode *inode, struct page *node_page) if (f2fs_sb_has_compression(F2FS_I_SB(inode)) && F2FS_FITS_IN_INODE(ri, F2FS_I(inode)->i_extra_isize, i_log_cluster_size)) { + unsigned short compress_flag; + ri->i_compr_blocks = cpu_to_le64(atomic_read( &F2FS_I(inode)->i_compr_blocks)); ri->i_compress_algorithm = F2FS_I(inode)->i_compress_algorithm; - ri->i_compress_flag = - cpu_to_le16(F2FS_I(inode)->i_compress_flag); + compress_flag = F2FS_I(inode)->i_compress_flag | + F2FS_I(inode)->i_compress_level << + COMPRESS_LEVEL_OFFSET; + ri->i_compress_flag = cpu_to_le16(compress_flag); ri->i_log_cluster_size = F2FS_I(inode)->i_log_cluster_size; } From 933141e4eb49d8b48721e2377835063a1e8fb823 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Sat, 28 Jan 2023 18:32:26 +0800 Subject: [PATCH 139/765] f2fs: fix to handle F2FS_IOC_START_ATOMIC_REPLACE in f2fs_compat_ioctl() Otherwise, 32-bits binary call ioctl(F2FS_IOC_START_ATOMIC_REPLACE) will fail in 64-bits kernel. Fixes: 41e8f85a75fc ("f2fs: introduce F2FS_IOC_START_ATOMIC_REPLACE") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 65199574cd7db..f05148a9dc99a 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -4814,6 +4814,7 @@ long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) case F2FS_IOC32_MOVE_RANGE: return f2fs_compat_ioc_move_range(file, arg); case F2FS_IOC_START_ATOMIC_WRITE: + case F2FS_IOC_START_ATOMIC_REPLACE: case F2FS_IOC_COMMIT_ATOMIC_WRITE: case F2FS_IOC_START_VOLATILE_WRITE: case F2FS_IOC_RELEASE_VOLATILE_WRITE: From 3aa51c61cb4a4dcb40df51ac61171e9ac5a35321 Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Mon, 30 Jan 2023 15:20:09 -0800 Subject: [PATCH 140/765] f2fs: retry to update the inode page given data corruption If the storage gives a corrupted node block due to short power failure and reset, f2fs stops the entire operations by setting the checkpoint failure flag. Let's give more chances to live by re-issuing IOs for a while in such critical path. Cc: stable@vger.kernel.org Suggested-by: Randall Huang Suggested-by: Chao Yu Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/inode.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c index 61a991b490ee7..28c9c72dda2a7 100644 --- a/fs/f2fs/inode.c +++ b/fs/f2fs/inode.c @@ -724,18 +724,19 @@ void f2fs_update_inode_page(struct inode *inode) { struct f2fs_sb_info *sbi = F2FS_I_SB(inode); struct page *node_page; + int count = 0; retry: node_page = f2fs_get_node_page(sbi, inode->i_ino); if (IS_ERR(node_page)) { int err = PTR_ERR(node_page); - if (err == -ENOMEM) { - cond_resched(); + /* The node block was truncated. */ + if (err == -ENOENT) + return; + + if (err == -ENOMEM || ++count <= DEFAULT_RETRY_IO_COUNT) goto retry; - } else if (err != -ENOENT) { - f2fs_stop_checkpoint(sbi, false, - STOP_CP_REASON_UPDATE_INODE); - } + f2fs_stop_checkpoint(sbi, false, STOP_CP_REASON_UPDATE_INODE); return; } f2fs_update_inode(inode, node_page); From 0dbbf0fb38d5ec5d4138d1aeaeb43d9217b9a592 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Sat, 21 Jan 2023 00:16:55 +0800 Subject: [PATCH 141/765] f2fs: fix to avoid potential memory corruption in __update_iostat_latency() Add iotype sanity check to avoid potential memory corruption. This is to fix the compile error below: fs/f2fs/iostat.c:231 __update_iostat_latency() error: buffer overflow 'io_lat->peak_lat[type]' 3 <= 3 vim +228 fs/f2fs/iostat.c 211 static inline void __update_iostat_latency(struct bio_iostat_ctx *iostat_ctx, 212 enum iostat_lat_type type) 213 { 214 unsigned long ts_diff; 215 unsigned int page_type = iostat_ctx->type; 216 struct f2fs_sb_info *sbi = iostat_ctx->sbi; 217 struct iostat_lat_info *io_lat = sbi->iostat_io_lat; 218 unsigned long flags; 219 220 if (!sbi->iostat_enable) 221 return; 222 223 ts_diff = jiffies - iostat_ctx->submit_ts; 224 if (page_type >= META_FLUSH) ^^^^^^^^^^ 225 page_type = META; 226 227 spin_lock_irqsave(&sbi->iostat_lat_lock, flags); @228 io_lat->sum_lat[type][page_type] += ts_diff; ^^^^^^^^^ Mixup between META_FLUSH and NR_PAGE_TYPE leads to memory corruption. Fixes: a4b6817625e7 ("f2fs: introduce periodic iostat io latency traces") Reported-by: kernel test robot Reported-by: Dan Carpenter Suggested-by: Chao Yu Suggested-by: Jaegeuk Kim Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/iostat.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/fs/f2fs/iostat.c b/fs/f2fs/iostat.c index ed8176939aa57..96637756eae8d 100644 --- a/fs/f2fs/iostat.c +++ b/fs/f2fs/iostat.c @@ -223,8 +223,12 @@ static inline void __update_iostat_latency(struct bio_iostat_ctx *iostat_ctx, return; ts_diff = jiffies - iostat_ctx->submit_ts; - if (iotype >= META_FLUSH) + if (iotype == META_FLUSH) { iotype = META; + } else if (iotype >= NR_PAGE_TYPE) { + f2fs_warn(sbi, "%s: %d over NR_PAGE_TYPE", __func__, iotype); + return; + } if (rw == 0) { idx = READ_IO; From 8c0ed062ce27f6b7f0a568cb241e2b4dd2d9e6a6 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Tue, 31 Jan 2023 22:47:00 +0800 Subject: [PATCH 142/765] f2fs: fix to update age extent correctly during truncation nr_free may be less than len, we should update age extent cache w/ range [fofs, len] rather than [fofs, nr_free]. Fixes: 71644dff4811 ("f2fs: add block_age-based extent cache") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index f05148a9dc99a..a83117325b873 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -617,7 +617,7 @@ void f2fs_truncate_data_blocks_range(struct dnode_of_data *dn, int count) fofs = f2fs_start_bidx_of_node(ofs_of_node(dn->node_page), dn->inode) + ofs; f2fs_update_read_extent_cache_range(dn, fofs, 0, len); - f2fs_update_age_extent_cache_range(dn, fofs, nr_free); + f2fs_update_age_extent_cache_range(dn, fofs, len); dec_valid_block_count(sbi, dn->inode, nr_free); } dn->ofs_in_node = ofs; From a84153f939808102dfa10904aa0f743e734a3e1d Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Tue, 31 Jan 2023 22:47:01 +0800 Subject: [PATCH 143/765] f2fs: fix to update age extent in f2fs_do_zero_range() We should update age extent in f2fs_do_zero_range() like we did in f2fs_truncate_data_blocks_range(). Fixes: 71644dff4811 ("f2fs: add block_age-based extent cache") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index a83117325b873..a17aca50c18cc 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -1494,6 +1494,7 @@ static int f2fs_do_zero_range(struct dnode_of_data *dn, pgoff_t start, } f2fs_update_read_extent_cache_range(dn, start, 0, index - start); + f2fs_update_age_extent_cache_range(dn, start, index - start); return ret; } From 91cc8fbcc8c705310fc6e39899e6e6685531935b Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 12 Jan 2023 17:15:40 +0100 Subject: [PATCH 144/765] ubi: block: set BLK_MQ_F_BLOCKING Set BLK_MQ_F_BLOCKING so that the block layer always calls ->queue_rq from process context and drop the driver internal workqueue. Signed-off-by: Christoph Hellwig Signed-off-by: Richard Weinberger --- drivers/mtd/ubi/block.c | 97 ++++++++++++----------------------------- 1 file changed, 28 insertions(+), 69 deletions(-) diff --git a/drivers/mtd/ubi/block.c b/drivers/mtd/ubi/block.c index d725215ae66e5..f5d036203fe7f 100644 --- a/drivers/mtd/ubi/block.c +++ b/drivers/mtd/ubi/block.c @@ -35,7 +35,6 @@ #include #include #include -#include #include #include #include @@ -62,7 +61,6 @@ struct ubiblock_param { }; struct ubiblock_pdu { - struct work_struct work; struct ubi_sgl usgl; }; @@ -82,8 +80,6 @@ struct ubiblock { struct gendisk *gd; struct request_queue *rq; - struct workqueue_struct *wq; - struct mutex dev_mutex; struct list_head list; struct blk_mq_tag_set tag_set; @@ -181,20 +177,29 @@ static struct ubiblock *find_dev_nolock(int ubi_num, int vol_id) return NULL; } -static int ubiblock_read(struct ubiblock_pdu *pdu) +static blk_status_t ubiblock_read(struct request *req) { - int ret, leb, offset, bytes_left, to_read; - u64 pos; - struct request *req = blk_mq_rq_from_pdu(pdu); + struct ubiblock_pdu *pdu = blk_mq_rq_to_pdu(req); struct ubiblock *dev = req->q->queuedata; + u64 pos = blk_rq_pos(req) << 9; + int to_read = blk_rq_bytes(req); + int bytes_left = to_read; + /* Get LEB:offset address to read from */ + int offset = do_div(pos, dev->leb_size); + int leb = pos; + struct req_iterator iter; + struct bio_vec bvec; + int ret; - to_read = blk_rq_bytes(req); - pos = blk_rq_pos(req) << 9; + blk_mq_start_request(req); - /* Get LEB:offset address to read from */ - offset = do_div(pos, dev->leb_size); - leb = pos; - bytes_left = to_read; + /* + * It is safe to ignore the return value of blk_rq_map_sg() because + * the number of sg entries is limited to UBI_MAX_SG_COUNT + * and ubi_read_sg() will check that limit. + */ + ubi_sgl_init(&pdu->usgl); + blk_rq_map_sg(req->q, req, pdu->usgl.sg); while (bytes_left) { /* @@ -206,14 +211,17 @@ static int ubiblock_read(struct ubiblock_pdu *pdu) ret = ubi_read_sg(dev->desc, leb, &pdu->usgl, offset, to_read); if (ret < 0) - return ret; + break; bytes_left -= to_read; to_read = bytes_left; leb += 1; offset = 0; } - return 0; + + rq_for_each_segment(bvec, req, iter) + flush_dcache_page(bvec.bv_page); + return errno_to_blk_status(ret); } static int ubiblock_open(struct block_device *bdev, fmode_t mode) @@ -289,47 +297,15 @@ static const struct block_device_operations ubiblock_ops = { .getgeo = ubiblock_getgeo, }; -static void ubiblock_do_work(struct work_struct *work) -{ - int ret; - struct ubiblock_pdu *pdu = container_of(work, struct ubiblock_pdu, work); - struct request *req = blk_mq_rq_from_pdu(pdu); - struct req_iterator iter; - struct bio_vec bvec; - - blk_mq_start_request(req); - - /* - * It is safe to ignore the return value of blk_rq_map_sg() because - * the number of sg entries is limited to UBI_MAX_SG_COUNT - * and ubi_read_sg() will check that limit. - */ - blk_rq_map_sg(req->q, req, pdu->usgl.sg); - - ret = ubiblock_read(pdu); - - rq_for_each_segment(bvec, req, iter) - flush_dcache_page(bvec.bv_page); - - blk_mq_end_request(req, errno_to_blk_status(ret)); -} - static blk_status_t ubiblock_queue_rq(struct blk_mq_hw_ctx *hctx, const struct blk_mq_queue_data *bd) { - struct request *req = bd->rq; - struct ubiblock *dev = hctx->queue->queuedata; - struct ubiblock_pdu *pdu = blk_mq_rq_to_pdu(req); - - switch (req_op(req)) { + switch (req_op(bd->rq)) { case REQ_OP_READ: - ubi_sgl_init(&pdu->usgl); - queue_work(dev->wq, &pdu->work); - return BLK_STS_OK; + return ubiblock_read(bd->rq); default: return BLK_STS_IOERR; } - } static int ubiblock_init_request(struct blk_mq_tag_set *set, @@ -339,8 +315,6 @@ static int ubiblock_init_request(struct blk_mq_tag_set *set, struct ubiblock_pdu *pdu = blk_mq_rq_to_pdu(req); sg_init_table(pdu->usgl.sg, UBI_MAX_SG_COUNT); - INIT_WORK(&pdu->work, ubiblock_do_work); - return 0; } @@ -404,7 +378,7 @@ int ubiblock_create(struct ubi_volume_info *vi) dev->tag_set.ops = &ubiblock_mq_ops; dev->tag_set.queue_depth = 64; dev->tag_set.numa_node = NUMA_NO_NODE; - dev->tag_set.flags = BLK_MQ_F_SHOULD_MERGE; + dev->tag_set.flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_BLOCKING; dev->tag_set.cmd_size = sizeof(struct ubiblock_pdu); dev->tag_set.driver_data = dev; dev->tag_set.nr_hw_queues = 1; @@ -442,31 +416,18 @@ int ubiblock_create(struct ubi_volume_info *vi) dev->rq = gd->queue; blk_queue_max_segments(dev->rq, UBI_MAX_SG_COUNT); - /* - * Create one workqueue per volume (per registered block device). - * Remember workqueues are cheap, they're not threads. - */ - dev->wq = alloc_workqueue("%s", 0, 0, gd->disk_name); - if (!dev->wq) { - ret = -ENOMEM; - goto out_remove_minor; - } - list_add_tail(&dev->list, &ubiblock_devices); /* Must be the last step: anyone can call file ops from now on */ ret = add_disk(dev->gd); if (ret) - goto out_destroy_wq; + goto out_remove_minor; dev_info(disk_to_dev(dev->gd), "created from ubi%d:%d(%s)", dev->ubi_num, dev->vol_id, vi->name); mutex_unlock(&devices_mutex); return 0; -out_destroy_wq: - list_del(&dev->list); - destroy_workqueue(dev->wq); out_remove_minor: idr_remove(&ubiblock_minor_idr, gd->first_minor); out_cleanup_disk: @@ -485,8 +446,6 @@ static void ubiblock_cleanup(struct ubiblock *dev) { /* Stop new requests to arrive */ del_gendisk(dev->gd); - /* Flush pending work */ - destroy_workqueue(dev->wq); /* Finally destroy the blk queue */ dev_info(disk_to_dev(dev->gd), "released"); put_disk(dev->gd); From 0b3bc49c936c1e6f399fcd2028ce24c3de9e7e59 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 12 Jan 2023 22:40:12 -0800 Subject: [PATCH 145/765] ubi: use correct names in function kernel-doc comments Fix kernel-doc warnings by using the correct function names in their kernel-doc notation: drivers/mtd/ubi/eba.c:72: warning: expecting prototype for next_sqnum(). Prototype was for ubi_next_sqnum() instead drivers/mtd/ubi/wl.c:176: warning: expecting prototype for wl_tree_destroy(). Prototype was for wl_entry_destroy() instead drivers/mtd/ubi/misc.c:24: warning: expecting prototype for calc_data_len(). Prototype was for ubi_calc_data_len() instead Signed-off-by: Randy Dunlap Cc: Richard Weinberger Cc: Miquel Raynal Cc: Vignesh Raghavendra Cc: linux-mtd@lists.infradead.org Signed-off-by: Richard Weinberger --- drivers/mtd/ubi/eba.c | 2 +- drivers/mtd/ubi/misc.c | 2 +- drivers/mtd/ubi/wl.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c index 09c408c45a621..403b79d6efd5a 100644 --- a/drivers/mtd/ubi/eba.c +++ b/drivers/mtd/ubi/eba.c @@ -61,7 +61,7 @@ struct ubi_eba_table { }; /** - * next_sqnum - get next sequence number. + * ubi_next_sqnum - get next sequence number. * @ubi: UBI device description object * * This function returns next sequence number to use, which is just the current diff --git a/drivers/mtd/ubi/misc.c b/drivers/mtd/ubi/misc.c index 7b30c8ee3e82d..1794d66b6eb72 100644 --- a/drivers/mtd/ubi/misc.c +++ b/drivers/mtd/ubi/misc.c @@ -10,7 +10,7 @@ #include "ubi.h" /** - * calc_data_len - calculate how much real data is stored in a buffer. + * ubi_calc_data_len - calculate how much real data is stored in a buffer. * @ubi: UBI device description object * @buf: a buffer with the contents of the physical eraseblock * @length: the buffer length diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c index 9e14319225c97..40f39e5d6dfcc 100644 --- a/drivers/mtd/ubi/wl.c +++ b/drivers/mtd/ubi/wl.c @@ -165,7 +165,7 @@ static void wl_tree_add(struct ubi_wl_entry *e, struct rb_root *root) } /** - * wl_tree_destroy - destroy a wear-leveling entry. + * wl_entry_destroy - destroy a wear-leveling entry. * @ubi: UBI device description object * @e: the wear-leveling entry to add * From b03a41a495df35f8e8d25220878bd6b8472d9396 Mon Sep 17 00:00:00 2001 From: qixiaoyu1 Date: Thu, 2 Feb 2023 16:20:27 +0800 Subject: [PATCH 146/765] f2fs: fix wrong calculation of block age Currently we wrongly calculate the new block age to old * LAST_AGE_WEIGHT / 100. Fix it to new * (100 - LAST_AGE_WEIGHT) / 100 + old * LAST_AGE_WEIGHT / 100. Signed-off-by: qixiaoyu1 Signed-off-by: xiongping1 Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/extent_cache.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/fs/f2fs/extent_cache.c b/fs/f2fs/extent_cache.c index 1daf8c88c09bf..c8efc957c230b 100644 --- a/fs/f2fs/extent_cache.c +++ b/fs/f2fs/extent_cache.c @@ -874,11 +874,18 @@ void f2fs_update_read_extent_tree_range_compressed(struct inode *inode, static unsigned long long __calculate_block_age(unsigned long long new, unsigned long long old) { - unsigned long long diff; + unsigned int rem_old, rem_new; + unsigned long long res; - diff = (new >= old) ? new - (new - old) : new + (old - new); + res = div_u64_rem(new, 100, &rem_new) * (100 - LAST_AGE_WEIGHT) + + div_u64_rem(old, 100, &rem_old) * LAST_AGE_WEIGHT; - return div_u64(diff * LAST_AGE_WEIGHT, 100); + if (rem_new) + res += rem_new * (100 - LAST_AGE_WEIGHT) / 100; + if (rem_old) + res += rem_old * LAST_AGE_WEIGHT / 100; + + return res; } /* This returns a new age and allocated blocks in ei */ From 844545c51a5b2a524b22a2fe9d0b353b827d24b4 Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Thu, 2 Feb 2023 17:02:39 -0800 Subject: [PATCH 147/765] f2fs: fix cgroup writeback accounting with fs-layer encryption When writing a page from an encrypted file that is using filesystem-layer encryption (not inline encryption), f2fs encrypts the pagecache page into a bounce page, then writes the bounce page. It also passes the bounce page to wbc_account_cgroup_owner(). That's incorrect, because the bounce page is a newly allocated temporary page that doesn't have the memory cgroup of the original pagecache page. This makes wbc_account_cgroup_owner() not account the I/O to the owner of the pagecache page as it should. Fix this by always passing the pagecache page to wbc_account_cgroup_owner(). Fixes: 578c647879f7 ("f2fs: implement cgroup writeback support") Cc: stable@vger.kernel.org Reported-by: Matthew Wilcox (Oracle) Signed-off-by: Eric Biggers Acked-by: Tejun Heo Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 754841bce389f..8a636500db0ef 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -739,7 +739,7 @@ int f2fs_submit_page_bio(struct f2fs_io_info *fio) } if (fio->io_wbc && !is_read_io(fio->op)) - wbc_account_cgroup_owner(fio->io_wbc, page, PAGE_SIZE); + wbc_account_cgroup_owner(fio->io_wbc, fio->page, PAGE_SIZE); inc_page_count(fio->sbi, is_read_io(fio->op) ? __read_io_type(page) : WB_DATA_TYPE(fio->page)); @@ -949,7 +949,7 @@ int f2fs_merge_page_bio(struct f2fs_io_info *fio) } if (fio->io_wbc) - wbc_account_cgroup_owner(fio->io_wbc, page, PAGE_SIZE); + wbc_account_cgroup_owner(fio->io_wbc, fio->page, PAGE_SIZE); inc_page_count(fio->sbi, WB_DATA_TYPE(page)); @@ -1023,7 +1023,7 @@ void f2fs_submit_page_write(struct f2fs_io_info *fio) } if (fio->io_wbc) - wbc_account_cgroup_owner(fio->io_wbc, bio_page, PAGE_SIZE); + wbc_account_cgroup_owner(fio->io_wbc, fio->page, PAGE_SIZE); io->last_block_in_bio = fio->new_blkaddr; From 04d7a7ae43fc4eed800efadbb4a18059172afb19 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Thu, 2 Feb 2023 17:41:23 +0800 Subject: [PATCH 148/765] f2fs: fix f2fs_show_options to show nogc_merge mount option Commit 5911d2d1d1a3 ("f2fs: introduce gc_merge mount option") forgot to show nogc_merge option, let's fix it. Signed-off-by: Yangtao Li Signed-off-by: Jaegeuk Kim --- fs/f2fs/super.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index fddff5deaed2f..4ec2cbbc47ebe 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -1902,6 +1902,8 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root) if (test_opt(sbi, GC_MERGE)) seq_puts(seq, ",gc_merge"); + else + seq_puts(seq, ",nogc_merge"); if (test_opt(sbi, DISABLE_ROLL_FORWARD)) seq_puts(seq, ",disable_roll_forward"); From d23be468eada21c828058e0e8d60409eaec373ab Mon Sep 17 00:00:00 2001 From: qixiaoyu1 Date: Sat, 4 Feb 2023 17:43:45 +0800 Subject: [PATCH 149/765] f2fs: add sysfs nodes to set last_age_weight Signed-off-by: qixiaoyu1 Signed-off-by: xiongping1 Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- Documentation/ABI/testing/sysfs-fs-f2fs | 5 +++++ fs/f2fs/extent_cache.c | 15 +++++++++------ fs/f2fs/f2fs.h | 1 + fs/f2fs/sysfs.c | 11 +++++++++++ 4 files changed, 26 insertions(+), 6 deletions(-) diff --git a/Documentation/ABI/testing/sysfs-fs-f2fs b/Documentation/ABI/testing/sysfs-fs-f2fs index 75420c242cc44..0f17adc804885 100644 --- a/Documentation/ABI/testing/sysfs-fs-f2fs +++ b/Documentation/ABI/testing/sysfs-fs-f2fs @@ -717,3 +717,8 @@ Description: Controls background discard granularity of inner discard thread is smaller than granularity. The unit size is one block(4KB), now only support configuring in range of [0, 512]. Default: 512 + +What: /sys/fs/f2fs//last_age_weight +Date: January 2023 +Contact: "Ping Xiong" +Description: When DATA SEPARATION is on, it controls the weight of last data block age. diff --git a/fs/f2fs/extent_cache.c b/fs/f2fs/extent_cache.c index c8efc957c230b..aef308e871ab4 100644 --- a/fs/f2fs/extent_cache.c +++ b/fs/f2fs/extent_cache.c @@ -871,19 +871,21 @@ void f2fs_update_read_extent_tree_range_compressed(struct inode *inode, } #endif -static unsigned long long __calculate_block_age(unsigned long long new, +static unsigned long long __calculate_block_age(struct f2fs_sb_info *sbi, + unsigned long long new, unsigned long long old) { unsigned int rem_old, rem_new; unsigned long long res; + unsigned int weight = sbi->last_age_weight; - res = div_u64_rem(new, 100, &rem_new) * (100 - LAST_AGE_WEIGHT) - + div_u64_rem(old, 100, &rem_old) * LAST_AGE_WEIGHT; + res = div_u64_rem(new, 100, &rem_new) * (100 - weight) + + div_u64_rem(old, 100, &rem_old) * weight; if (rem_new) - res += rem_new * (100 - LAST_AGE_WEIGHT) / 100; + res += rem_new * (100 - weight) / 100; if (rem_old) - res += rem_old * LAST_AGE_WEIGHT / 100; + res += rem_old * weight / 100; return res; } @@ -917,7 +919,7 @@ static int __get_new_block_age(struct inode *inode, struct extent_info *ei, cur_age = ULLONG_MAX - tei.last_blocks + cur_blocks; if (tei.age) - ei->age = __calculate_block_age(cur_age, tei.age); + ei->age = __calculate_block_age(sbi, cur_age, tei.age); else ei->age = cur_age; ei->last_blocks = cur_blocks; @@ -1244,6 +1246,7 @@ void f2fs_init_extent_cache_info(struct f2fs_sb_info *sbi) atomic64_set(&sbi->allocated_data_blocks, 0); sbi->hot_data_age_threshold = DEF_HOT_DATA_AGE_THRESHOLD; sbi->warm_data_age_threshold = DEF_WARM_DATA_AGE_THRESHOLD; + sbi->last_age_weight = LAST_AGE_WEIGHT; } int __init f2fs_create_extent_cache(void) diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index e80144384acbe..44f2d76525bfa 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -1671,6 +1671,7 @@ struct f2fs_sb_info { /* The threshold used for hot and warm data seperation*/ unsigned int hot_data_age_threshold; unsigned int warm_data_age_threshold; + unsigned int last_age_weight; /* basic filesystem units */ unsigned int log_sectors_per_block; /* log2 sectors per block */ diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c index 3c6425f9ed0ad..8789b3b53fb68 100644 --- a/fs/f2fs/sysfs.c +++ b/fs/f2fs/sysfs.c @@ -697,6 +697,15 @@ static ssize_t __sbi_store(struct f2fs_attr *a, return count; } + if (!strcmp(a->attr.name, "last_age_weight")) { + if (t > 100) + return -EINVAL; + if (t == *ui) + return count; + *ui = (unsigned int)t; + return count; + } + *ui = (unsigned int)t; return count; @@ -956,6 +965,7 @@ F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, revoked_atomic_block, revoked_atomic_block) /* For block age extent cache */ F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, hot_data_age_threshold, hot_data_age_threshold); F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, warm_data_age_threshold, warm_data_age_threshold); +F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, last_age_weight, last_age_weight); #define ATTR_LIST(name) (&f2fs_attr_##name.attr) static struct attribute *f2fs_attrs[] = { @@ -1055,6 +1065,7 @@ static struct attribute *f2fs_attrs[] = { ATTR_LIST(revoked_atomic_block), ATTR_LIST(hot_data_age_threshold), ATTR_LIST(warm_data_age_threshold), + ATTR_LIST(last_age_weight), NULL, }; ATTRIBUTE_GROUPS(f2fs); From d9bac032ac0de8f510b44904b6eb7272679883d1 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Wed, 1 Feb 2023 18:47:02 +0800 Subject: [PATCH 150/765] f2fs: use iostat_lat_type directly as a parameter in the iostat_update_and_unbind_ctx() Convert to use iostat_lat_type as parameter instead of raw number. BTW, move NUM_PREALLOC_IOSTAT_CTXS to the header file, adjust iostat_lat[{0,1,2}] to iostat_lat[{READ_IO,WRITE_SYNC_IO,WRITE_ASYNC_IO}] in tracepoint function, and rename iotype to page_type to match the definition. Reported-by: kernel test robot Reported-by: Dan Carpenter Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 4 +-- fs/f2fs/iostat.c | 48 +++++++++++++++------------------ fs/f2fs/iostat.h | 19 ++++++------- include/trace/events/f2fs.h | 54 ++++++++++++++++++------------------- 4 files changed, 60 insertions(+), 65 deletions(-) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 8a636500db0ef..890644cc2e3d7 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -292,7 +292,7 @@ static void f2fs_read_end_io(struct bio *bio) struct bio_post_read_ctx *ctx; bool intask = in_task(); - iostat_update_and_unbind_ctx(bio, 0); + iostat_update_and_unbind_ctx(bio); ctx = bio->bi_private; if (time_to_inject(sbi, FAULT_READ_IO)) @@ -330,7 +330,7 @@ static void f2fs_write_end_io(struct bio *bio) struct bio_vec *bvec; struct bvec_iter_all iter_all; - iostat_update_and_unbind_ctx(bio, 1); + iostat_update_and_unbind_ctx(bio); sbi = bio->bi_private; if (time_to_inject(sbi, FAULT_WRITE_IO)) diff --git a/fs/f2fs/iostat.c b/fs/f2fs/iostat.c index 96637756eae8d..3d5bfb1ad585d 100644 --- a/fs/f2fs/iostat.c +++ b/fs/f2fs/iostat.c @@ -14,7 +14,6 @@ #include "iostat.h" #include -#define NUM_PREALLOC_IOSTAT_CTXS 128 static struct kmem_cache *bio_iostat_ctx_cache; static mempool_t *bio_iostat_ctx_pool; @@ -210,53 +209,48 @@ void f2fs_update_iostat(struct f2fs_sb_info *sbi, struct inode *inode, } static inline void __update_iostat_latency(struct bio_iostat_ctx *iostat_ctx, - int rw, bool is_sync) + enum iostat_lat_type lat_type) { unsigned long ts_diff; - unsigned int iotype = iostat_ctx->type; + unsigned int page_type = iostat_ctx->type; struct f2fs_sb_info *sbi = iostat_ctx->sbi; struct iostat_lat_info *io_lat = sbi->iostat_io_lat; - int idx; unsigned long flags; if (!sbi->iostat_enable) return; ts_diff = jiffies - iostat_ctx->submit_ts; - if (iotype == META_FLUSH) { - iotype = META; - } else if (iotype >= NR_PAGE_TYPE) { - f2fs_warn(sbi, "%s: %d over NR_PAGE_TYPE", __func__, iotype); + if (page_type == META_FLUSH) { + page_type = META; + } else if (page_type >= NR_PAGE_TYPE) { + f2fs_warn(sbi, "%s: %d over NR_PAGE_TYPE", __func__, page_type); return; } - if (rw == 0) { - idx = READ_IO; - } else { - if (is_sync) - idx = WRITE_SYNC_IO; - else - idx = WRITE_ASYNC_IO; - } - spin_lock_irqsave(&sbi->iostat_lat_lock, flags); - io_lat->sum_lat[idx][iotype] += ts_diff; - io_lat->bio_cnt[idx][iotype]++; - if (ts_diff > io_lat->peak_lat[idx][iotype]) - io_lat->peak_lat[idx][iotype] = ts_diff; + io_lat->sum_lat[lat_type][page_type] += ts_diff; + io_lat->bio_cnt[lat_type][page_type]++; + if (ts_diff > io_lat->peak_lat[lat_type][page_type]) + io_lat->peak_lat[lat_type][page_type] = ts_diff; spin_unlock_irqrestore(&sbi->iostat_lat_lock, flags); } -void iostat_update_and_unbind_ctx(struct bio *bio, int rw) +void iostat_update_and_unbind_ctx(struct bio *bio) { struct bio_iostat_ctx *iostat_ctx = bio->bi_private; - bool is_sync = bio->bi_opf & REQ_SYNC; + enum iostat_lat_type lat_type; - if (rw == 0) - bio->bi_private = iostat_ctx->post_read_ctx; - else + if (op_is_write(bio_op(bio))) { + lat_type = bio->bi_opf & REQ_SYNC ? + WRITE_SYNC_IO : WRITE_ASYNC_IO; bio->bi_private = iostat_ctx->sbi; - __update_iostat_latency(iostat_ctx, rw, is_sync); + } else { + lat_type = READ_IO; + bio->bi_private = iostat_ctx->post_read_ctx; + } + + __update_iostat_latency(iostat_ctx, lat_type); mempool_free(iostat_ctx, bio_iostat_ctx_pool); } diff --git a/fs/f2fs/iostat.h b/fs/f2fs/iostat.h index 2c048307b6e0b..eb99d05cf2727 100644 --- a/fs/f2fs/iostat.h +++ b/fs/f2fs/iostat.h @@ -8,20 +8,21 @@ struct bio_post_read_ctx; +enum iostat_lat_type { + READ_IO = 0, + WRITE_SYNC_IO, + WRITE_ASYNC_IO, + MAX_IO_TYPE, +}; + #ifdef CONFIG_F2FS_IOSTAT +#define NUM_PREALLOC_IOSTAT_CTXS 128 #define DEFAULT_IOSTAT_PERIOD_MS 3000 #define MIN_IOSTAT_PERIOD_MS 100 /* maximum period of iostat tracing is 1 day */ #define MAX_IOSTAT_PERIOD_MS 8640000 -enum { - READ_IO, - WRITE_SYNC_IO, - WRITE_ASYNC_IO, - MAX_IO_TYPE, -}; - struct iostat_lat_info { unsigned long sum_lat[MAX_IO_TYPE][NR_PAGE_TYPE]; /* sum of io latencies */ unsigned long peak_lat[MAX_IO_TYPE][NR_PAGE_TYPE]; /* peak io latency */ @@ -57,7 +58,7 @@ static inline struct bio_post_read_ctx *get_post_read_ctx(struct bio *bio) return iostat_ctx->post_read_ctx; } -extern void iostat_update_and_unbind_ctx(struct bio *bio, int rw); +extern void iostat_update_and_unbind_ctx(struct bio *bio); extern void iostat_alloc_and_bind_ctx(struct f2fs_sb_info *sbi, struct bio *bio, struct bio_post_read_ctx *ctx); extern int f2fs_init_iostat_processing(void); @@ -67,7 +68,7 @@ extern void f2fs_destroy_iostat(struct f2fs_sb_info *sbi); #else static inline void f2fs_update_iostat(struct f2fs_sb_info *sbi, struct inode *inode, enum iostat_type type, unsigned long long io_bytes) {} -static inline void iostat_update_and_unbind_ctx(struct bio *bio, int rw) {} +static inline void iostat_update_and_unbind_ctx(struct bio *bio) {} static inline void iostat_alloc_and_bind_ctx(struct f2fs_sb_info *sbi, struct bio *bio, struct bio_post_read_ctx *ctx) {} static inline void iostat_update_submit_ctx(struct bio *bio, diff --git a/include/trace/events/f2fs.h b/include/trace/events/f2fs.h index fe6bcf5f917dc..1322d34a5dfc1 100644 --- a/include/trace/events/f2fs.h +++ b/include/trace/events/f2fs.h @@ -2082,33 +2082,33 @@ TRACE_EVENT(f2fs_iostat_latency, TP_fast_assign( __entry->dev = sbi->sb->s_dev; - __entry->d_rd_peak = iostat_lat[0][DATA].peak_lat; - __entry->d_rd_avg = iostat_lat[0][DATA].avg_lat; - __entry->d_rd_cnt = iostat_lat[0][DATA].cnt; - __entry->n_rd_peak = iostat_lat[0][NODE].peak_lat; - __entry->n_rd_avg = iostat_lat[0][NODE].avg_lat; - __entry->n_rd_cnt = iostat_lat[0][NODE].cnt; - __entry->m_rd_peak = iostat_lat[0][META].peak_lat; - __entry->m_rd_avg = iostat_lat[0][META].avg_lat; - __entry->m_rd_cnt = iostat_lat[0][META].cnt; - __entry->d_wr_s_peak = iostat_lat[1][DATA].peak_lat; - __entry->d_wr_s_avg = iostat_lat[1][DATA].avg_lat; - __entry->d_wr_s_cnt = iostat_lat[1][DATA].cnt; - __entry->n_wr_s_peak = iostat_lat[1][NODE].peak_lat; - __entry->n_wr_s_avg = iostat_lat[1][NODE].avg_lat; - __entry->n_wr_s_cnt = iostat_lat[1][NODE].cnt; - __entry->m_wr_s_peak = iostat_lat[1][META].peak_lat; - __entry->m_wr_s_avg = iostat_lat[1][META].avg_lat; - __entry->m_wr_s_cnt = iostat_lat[1][META].cnt; - __entry->d_wr_as_peak = iostat_lat[2][DATA].peak_lat; - __entry->d_wr_as_avg = iostat_lat[2][DATA].avg_lat; - __entry->d_wr_as_cnt = iostat_lat[2][DATA].cnt; - __entry->n_wr_as_peak = iostat_lat[2][NODE].peak_lat; - __entry->n_wr_as_avg = iostat_lat[2][NODE].avg_lat; - __entry->n_wr_as_cnt = iostat_lat[2][NODE].cnt; - __entry->m_wr_as_peak = iostat_lat[2][META].peak_lat; - __entry->m_wr_as_avg = iostat_lat[2][META].avg_lat; - __entry->m_wr_as_cnt = iostat_lat[2][META].cnt; + __entry->d_rd_peak = iostat_lat[READ_IO][DATA].peak_lat; + __entry->d_rd_avg = iostat_lat[READ_IO][DATA].avg_lat; + __entry->d_rd_cnt = iostat_lat[READ_IO][DATA].cnt; + __entry->n_rd_peak = iostat_lat[READ_IO][NODE].peak_lat; + __entry->n_rd_avg = iostat_lat[READ_IO][NODE].avg_lat; + __entry->n_rd_cnt = iostat_lat[READ_IO][NODE].cnt; + __entry->m_rd_peak = iostat_lat[READ_IO][META].peak_lat; + __entry->m_rd_avg = iostat_lat[READ_IO][META].avg_lat; + __entry->m_rd_cnt = iostat_lat[READ_IO][META].cnt; + __entry->d_wr_s_peak = iostat_lat[WRITE_SYNC_IO][DATA].peak_lat; + __entry->d_wr_s_avg = iostat_lat[WRITE_SYNC_IO][DATA].avg_lat; + __entry->d_wr_s_cnt = iostat_lat[WRITE_SYNC_IO][DATA].cnt; + __entry->n_wr_s_peak = iostat_lat[WRITE_SYNC_IO][NODE].peak_lat; + __entry->n_wr_s_avg = iostat_lat[WRITE_SYNC_IO][NODE].avg_lat; + __entry->n_wr_s_cnt = iostat_lat[WRITE_SYNC_IO][NODE].cnt; + __entry->m_wr_s_peak = iostat_lat[WRITE_SYNC_IO][META].peak_lat; + __entry->m_wr_s_avg = iostat_lat[WRITE_SYNC_IO][META].avg_lat; + __entry->m_wr_s_cnt = iostat_lat[WRITE_SYNC_IO][META].cnt; + __entry->d_wr_as_peak = iostat_lat[WRITE_ASYNC_IO][DATA].peak_lat; + __entry->d_wr_as_avg = iostat_lat[WRITE_ASYNC_IO][DATA].avg_lat; + __entry->d_wr_as_cnt = iostat_lat[WRITE_ASYNC_IO][DATA].cnt; + __entry->n_wr_as_peak = iostat_lat[WRITE_ASYNC_IO][NODE].peak_lat; + __entry->n_wr_as_avg = iostat_lat[WRITE_ASYNC_IO][NODE].avg_lat; + __entry->n_wr_as_cnt = iostat_lat[WRITE_ASYNC_IO][NODE].cnt; + __entry->m_wr_as_peak = iostat_lat[WRITE_ASYNC_IO][META].peak_lat; + __entry->m_wr_as_avg = iostat_lat[WRITE_ASYNC_IO][META].avg_lat; + __entry->m_wr_as_cnt = iostat_lat[WRITE_ASYNC_IO][META].cnt; ), TP_printk("dev = (%d,%d), " From 267c159f9c7bcb7009dae16889b880c5ed8759a8 Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Sun, 5 Feb 2023 19:13:39 -0800 Subject: [PATCH 151/765] f2fs: fix kernel crash due to null io->bio We should return when io->bio is null before doing anything. Otherwise, panic. BUG: kernel NULL pointer dereference, address: 0000000000000010 RIP: 0010:__submit_merged_write_cond+0x164/0x240 [f2fs] Call Trace: f2fs_submit_merged_write+0x1d/0x30 [f2fs] commit_checkpoint+0x110/0x1e0 [f2fs] f2fs_write_checkpoint+0x9f7/0xf00 [f2fs] ? __pfx_issue_checkpoint_thread+0x10/0x10 [f2fs] __checkpoint_and_complete_reqs+0x84/0x190 [f2fs] ? preempt_count_add+0x82/0xc0 ? __pfx_issue_checkpoint_thread+0x10/0x10 [f2fs] issue_checkpoint_thread+0x4c/0xf0 [f2fs] ? __pfx_autoremove_wake_function+0x10/0x10 kthread+0xff/0x130 ? __pfx_kthread+0x10/0x10 ret_from_fork+0x2c/0x50 Cc: stable@vger.kernel.org # v5.18+ Fixes: 64bf0eef0171 ("f2fs: pass the bio operation to bio_alloc_bioset") Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 890644cc2e3d7..d04220cc2d260 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -653,6 +653,9 @@ static void __f2fs_submit_merged_write(struct f2fs_sb_info *sbi, f2fs_down_write(&io->io_rwsem); + if (!io->bio) + goto unlock_out; + /* change META to META_FLUSH in the checkpoint procedure */ if (type >= META_FLUSH) { io->fio.type = META_FLUSH; @@ -661,6 +664,7 @@ static void __f2fs_submit_merged_write(struct f2fs_sb_info *sbi, io->bio->bi_opf |= REQ_PREFLUSH | REQ_FUA; } __submit_merged_bio(io); +unlock_out: f2fs_up_write(&io->io_rwsem); } From 146949defda868378992171b9e42318b06fcd482 Mon Sep 17 00:00:00 2001 From: Jinyoung CHOI Date: Mon, 6 Feb 2023 20:56:00 +0900 Subject: [PATCH 152/765] f2fs: fix typos in comments This patch is to fix typos in f2fs files. Signed-off-by: Jinyoung Choi Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/checkpoint.c | 4 ++-- fs/f2fs/compress.c | 2 +- fs/f2fs/data.c | 8 ++++---- fs/f2fs/extent_cache.c | 4 ++-- fs/f2fs/file.c | 6 +++--- fs/f2fs/namei.c | 2 +- fs/f2fs/segment.c | 2 +- 7 files changed, 14 insertions(+), 14 deletions(-) diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c index 89ce08b0ff7c6..1369ec892a2ca 100644 --- a/fs/f2fs/checkpoint.c +++ b/fs/f2fs/checkpoint.c @@ -792,7 +792,7 @@ static void write_orphan_inodes(struct f2fs_sb_info *sbi, block_t start_blk) */ head = &im->ino_list; - /* loop for each orphan inode entry and write them in Jornal block */ + /* loop for each orphan inode entry and write them in journal block */ list_for_each_entry(orphan, head, list) { if (!page) { page = f2fs_grab_meta_page(sbi, start_blk++); @@ -1122,7 +1122,7 @@ int f2fs_sync_dirty_inodes(struct f2fs_sb_info *sbi, enum inode_type type, } else { /* * We should submit bio, since it exists several - * wribacking dentry pages in the freeing inode. + * writebacking dentry pages in the freeing inode. */ f2fs_submit_merged_write(sbi, DATA); cond_resched(); diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c index e4851f7a43d8d..b40dec3d7f799 100644 --- a/fs/f2fs/compress.c +++ b/fs/f2fs/compress.c @@ -1225,7 +1225,7 @@ static int f2fs_write_compressed_pages(struct compress_ctx *cc, loff_t psize; int i, err; - /* we should bypass data pages to proceed the kworkder jobs */ + /* we should bypass data pages to proceed the kworker jobs */ if (unlikely(f2fs_cp_error(sbi))) { mapping_set_error(cc->rpages[0]->mapping, -EIO); goto out_free; diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index d04220cc2d260..28e09682b056e 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -2386,7 +2386,7 @@ static int f2fs_mpage_readpages(struct inode *inode, #ifdef CONFIG_F2FS_FS_COMPRESSION if (f2fs_compressed_file(inode)) { - /* there are remained comressed pages, submit them */ + /* there are remained compressed pages, submit them */ if (!f2fs_cluster_can_merge_page(&cc, page->index)) { ret = f2fs_read_multi_pages(&cc, &bio, max_nr_pages, @@ -2792,7 +2792,7 @@ int f2fs_write_single_data_page(struct page *page, int *submitted, trace_f2fs_writepage(page, DATA); - /* we should bypass data pages to proceed the kworkder jobs */ + /* we should bypass data pages to proceed the kworker jobs */ if (unlikely(f2fs_cp_error(sbi))) { mapping_set_error(page->mapping, -EIO); /* @@ -2911,7 +2911,7 @@ int f2fs_write_single_data_page(struct page *page, int *submitted, redirty_out: redirty_page_for_writepage(wbc, page); /* - * pageout() in MM traslates EAGAIN, so calls handle_write_error() + * pageout() in MM translates EAGAIN, so calls handle_write_error() * -> mapping_set_error() -> set_bit(AS_EIO, ...). * file_write_and_wait_range() will see EIO error, which is critical * to return value of fsync() followed by atomic_write failure to user. @@ -2945,7 +2945,7 @@ static int f2fs_write_data_page(struct page *page, } /* - * This function was copied from write_cche_pages from mm/page-writeback.c. + * This function was copied from write_cache_pages from mm/page-writeback.c. * The major change is making write step of cold data page separately from * warm/hot data page. */ diff --git a/fs/f2fs/extent_cache.c b/fs/f2fs/extent_cache.c index aef308e871ab4..8d922c592dae3 100644 --- a/fs/f2fs/extent_cache.c +++ b/fs/f2fs/extent_cache.c @@ -233,7 +233,7 @@ struct rb_node **f2fs_lookup_rb_tree_for_insert(struct f2fs_sb_info *sbi, * @prev_ex: extent before ofs * @next_ex: extent after ofs * @insert_p: insert point for new extent at ofs - * in order to simpfy the insertion after. + * in order to simplify the insertion after. * tree must stay unchanged between lookup and insertion. */ struct rb_entry *f2fs_lookup_rb_tree_ret(struct rb_root_cached *root, @@ -718,7 +718,7 @@ static void __update_extent_tree_range(struct inode *inode, if (!en) en = next_en; - /* 2. invlidate all extent nodes in range [fofs, fofs + len - 1] */ + /* 2. invalidate all extent nodes in range [fofs, fofs + len - 1] */ while (en && en->ei.fofs < end) { unsigned int org_end; int parts = 0; /* # of parts current extent split into */ diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index a17aca50c18cc..300eae8b54154 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -303,7 +303,7 @@ static int f2fs_do_sync_file(struct file *file, loff_t start, loff_t end, * for OPU case, during fsync(), node can be persisted before * data when lower device doesn't support write barrier, result * in data corruption after SPO. - * So for strict fsync mode, force to use atomic write sematics + * So for strict fsync mode, force to use atomic write semantics * to keep write order in between data/node and last node to * avoid potential data corruption. */ @@ -1806,7 +1806,7 @@ static long f2fs_fallocate(struct file *file, int mode, return -EOPNOTSUPP; /* - * Pinned file should not support partial trucation since the block + * Pinned file should not support partial truncation since the block * can be used by applications. */ if ((f2fs_compressed_file(inode) || f2fs_is_pinned_file(inode)) && @@ -1856,7 +1856,7 @@ static long f2fs_fallocate(struct file *file, int mode, static int f2fs_release_file(struct inode *inode, struct file *filp) { /* - * f2fs_relase_file is called at every close calls. So we should + * f2fs_release_file is called at every close calls. So we should * not drop any inmemory pages by close called by other process. */ if (!(filp->f_mode & FMODE_WRITE) || diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c index 82923273f4bb8..f9aafb7ac44df 100644 --- a/fs/f2fs/namei.c +++ b/fs/f2fs/namei.c @@ -963,7 +963,7 @@ static int f2fs_rename(struct user_namespace *mnt_userns, struct inode *old_dir, /* * If new_inode is null, the below renaming flow will - * add a link in old_dir which can conver inline_dir. + * add a link in old_dir which can convert inline_dir. * After then, if we failed to get the entry due to other * reasons like ENOMEM, we had to remove the new entry. * Instead of adding such the error handling routine, let's diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 69b01b5c0ce0a..ead3f35f501d7 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -3616,7 +3616,7 @@ void f2fs_wait_on_page_writeback(struct page *page, /* submit cached LFS IO */ f2fs_submit_merged_write_cond(sbi, NULL, page, 0, type); - /* sbumit cached IPU IO */ + /* submit cached IPU IO */ f2fs_submit_merged_ipu_write(sbi, NULL, page); if (ordered) { wait_on_page_writeback(page); From c5bf83483382600988d7db5ffe9fcd1936b491fd Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Mon, 6 Feb 2023 22:43:08 +0800 Subject: [PATCH 153/765] f2fs: fix to set ipu policy For LFS mode, it should update outplace and no need inplace update. When using LFS mode for small-volume devices, IPU will not be used, and the OPU writing method is actually used, but F2FS_IPU_FORCE can be read from the ipu_policy node, which is different from the actual situation. And remount to lfs mode should be disallowed when f2fs ipu is enabled, let's fix it. Fixes: 84b89e5d943d ("f2fs: add auto tuning for small devices") Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.h | 10 +++++++++- fs/f2fs/super.c | 15 +++++++++++---- fs/f2fs/sysfs.c | 9 +++++++++ 3 files changed, 29 insertions(+), 5 deletions(-) diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h index 0f3f05cb8c29f..8ee5e5db9287c 100644 --- a/fs/f2fs/segment.h +++ b/fs/f2fs/segment.h @@ -670,6 +670,8 @@ static inline int utilization(struct f2fs_sb_info *sbi) #define SMALL_VOLUME_SEGMENTS (16 * 512) /* 16GB */ +#define F2FS_IPU_DISABLE 0 + enum { F2FS_IPU_FORCE, F2FS_IPU_SSR, @@ -679,10 +681,16 @@ enum { F2FS_IPU_ASYNC, F2FS_IPU_NOCACHE, F2FS_IPU_HONOR_OPU_WRITE, + F2FS_IPU_MAX, }; +static inline bool IS_F2FS_IPU_DISABLE(struct f2fs_sb_info *sbi) +{ + return SM_I(sbi)->ipu_policy == F2FS_IPU_DISABLE; +} + #define F2FS_IPU_POLICY(name) \ -static inline int IS_##name(struct f2fs_sb_info *sbi) \ +static inline bool IS_##name(struct f2fs_sb_info *sbi) \ { \ return SM_I(sbi)->ipu_policy & BIT(name); \ } diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 4ec2cbbc47ebe..c11a161ba5bec 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -1346,12 +1346,12 @@ static int parse_options(struct super_block *sb, char *options, bool is_remount) } if (test_opt(sbi, DISABLE_CHECKPOINT) && f2fs_lfs_mode(sbi)) { - f2fs_err(sbi, "LFS not compatible with checkpoint=disable"); + f2fs_err(sbi, "LFS is not compatible with checkpoint=disable"); return -EINVAL; } if (test_opt(sbi, ATGC) && f2fs_lfs_mode(sbi)) { - f2fs_err(sbi, "LFS not compatible with ATGC"); + f2fs_err(sbi, "LFS is not compatible with ATGC"); return -EINVAL; } @@ -2304,6 +2304,12 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data) } } #endif + if (f2fs_lfs_mode(sbi) && !IS_F2FS_IPU_DISABLE(sbi)) { + err = -EINVAL; + f2fs_warn(sbi, "LFS is not compatible with IPU"); + goto restore_opts; + } + /* disallow enable atgc dynamically */ if (no_atgc == !!test_opt(sbi, ATGC)) { err = -EINVAL; @@ -4085,8 +4091,9 @@ static void f2fs_tuning_parameters(struct f2fs_sb_info *sbi) if (f2fs_block_unit_discard(sbi)) SM_I(sbi)->dcc_info->discard_granularity = MIN_DISCARD_GRANULARITY; - SM_I(sbi)->ipu_policy = BIT(F2FS_IPU_FORCE) | - BIT(F2FS_IPU_HONOR_OPU_WRITE); + if (!f2fs_lfs_mode(sbi)) + SM_I(sbi)->ipu_policy = BIT(F2FS_IPU_FORCE) | + BIT(F2FS_IPU_HONOR_OPU_WRITE); } sbi->readdir_ra = true; diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c index 8789b3b53fb68..6082e132257a3 100644 --- a/fs/f2fs/sysfs.c +++ b/fs/f2fs/sysfs.c @@ -706,6 +706,15 @@ static ssize_t __sbi_store(struct f2fs_attr *a, return count; } + if (!strcmp(a->attr.name, "ipu_policy")) { + if (t >= BIT(F2FS_IPU_MAX)) + return -EINVAL; + if (t && f2fs_lfs_mode(sbi)) + return -EINVAL; + SM_I(sbi)->ipu_policy = (unsigned int)t; + return count; + } + *ui = (unsigned int)t; return count; From 9e615dbba41e2a56bb4ae50c56f4c10294fa16b8 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Mon, 6 Feb 2023 22:43:09 +0800 Subject: [PATCH 154/765] f2fs: add missing description for ipu_policy node IPU policy can be disabled, let's add description for it and other policy. Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- Documentation/ABI/testing/sysfs-fs-f2fs | 27 ++++++++++++++++--------- 1 file changed, 17 insertions(+), 10 deletions(-) diff --git a/Documentation/ABI/testing/sysfs-fs-f2fs b/Documentation/ABI/testing/sysfs-fs-f2fs index 0f17adc804885..94132745ecbe7 100644 --- a/Documentation/ABI/testing/sysfs-fs-f2fs +++ b/Documentation/ABI/testing/sysfs-fs-f2fs @@ -49,16 +49,23 @@ Contact: "Jaegeuk Kim" Description: Controls the in-place-update policy. updates in f2fs. User can set: - ==== ================= - 0x01 F2FS_IPU_FORCE - 0x02 F2FS_IPU_SSR - 0x04 F2FS_IPU_UTIL - 0x08 F2FS_IPU_SSR_UTIL - 0x10 F2FS_IPU_FSYNC - 0x20 F2FS_IPU_ASYNC - 0x40 F2FS_IPU_NOCACHE - 0x80 F2FS_IPU_HONOR_OPU_WRITE - ==== ================= + ===== =============== =================================================== + value policy description + 0x00 DISABLE disable IPU(=default option in LFS mode) + 0x01 FORCE all the time + 0x02 SSR if SSR mode is activated + 0x04 UTIL if FS utilization is over threashold + 0x08 SSR_UTIL if SSR mode is activated and FS utilization is over + threashold + 0x10 FSYNC activated in fsync path only for high performance + flash storages. IPU will be triggered only if the + # of dirty pages over min_fsync_blocks. + (=default option) + 0x20 ASYNC do IPU given by asynchronous write requests + 0x40 NOCACHE disable IPU bio cache + 0x80 HONOR_OPU_WRITE use OPU write prior to IPU write if inode has + FI_OPU_WRITE flag + ===== =============== =================================================== Refer segment.h for details. From 3478c83cf26bbffd026ae6a56bcb1fe544f0834e Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Wed, 14 Dec 2022 15:08:18 -0500 Subject: [PATCH 155/765] ext4: improve xattr consistency checking and error reporting Refactor the in-inode and xattr block consistency checking, and report more fine-grained reports of the consistency problems. Also add more consistency checks for ea_inode number. Reviewed-by: Andreas Dilger Signed-off-by: Theodore Ts'o Link: https://lore.kernel.org/r/20221214200818.870087-1-tytso@mit.edu Signed-off-by: Theodore Ts'o --- fs/ext4/xattr.c | 126 ++++++++++++++++++++++++++++++------------------ 1 file changed, 80 insertions(+), 46 deletions(-) diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 69a1b8c6a2eca..e51052a247cc6 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -184,27 +184,73 @@ ext4_xattr_handler(int name_index) } static int -ext4_xattr_check_entries(struct ext4_xattr_entry *entry, void *end, - void *value_start) +check_xattrs(struct inode *inode, struct buffer_head *bh, + struct ext4_xattr_entry *entry, void *end, void *value_start, + const char *function, unsigned int line) { struct ext4_xattr_entry *e = entry; + int err = -EFSCORRUPTED; + char *err_str; + + if (bh) { + if (BHDR(bh)->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC) || + BHDR(bh)->h_blocks != cpu_to_le32(1)) { + err_str = "invalid header"; + goto errout; + } + if (buffer_verified(bh)) + return 0; + if (!ext4_xattr_block_csum_verify(inode, bh)) { + err = -EFSBADCRC; + err_str = "invalid checksum"; + goto errout; + } + } else { + struct ext4_xattr_ibody_header *header = value_start; + + header -= 1; + if (end - (void *)header < sizeof(*header) + sizeof(u32)) { + err_str = "in-inode xattr block too small"; + goto errout; + } + if (header->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC)) { + err_str = "bad magic number in in-inode xattr"; + goto errout; + } + } /* Find the end of the names list */ while (!IS_LAST_ENTRY(e)) { struct ext4_xattr_entry *next = EXT4_XATTR_NEXT(e); - if ((void *)next >= end) - return -EFSCORRUPTED; - if (strnlen(e->e_name, e->e_name_len) != e->e_name_len) - return -EFSCORRUPTED; + if ((void *)next >= end) { + err_str = "e_name out of bounds"; + goto errout; + } + if (strnlen(e->e_name, e->e_name_len) != e->e_name_len) { + err_str = "bad e_name length"; + goto errout; + } e = next; } /* Check the values */ while (!IS_LAST_ENTRY(entry)) { u32 size = le32_to_cpu(entry->e_value_size); + unsigned long ea_ino = le32_to_cpu(entry->e_value_inum); - if (size > EXT4_XATTR_SIZE_MAX) - return -EFSCORRUPTED; + if (!ext4_has_feature_ea_inode(inode->i_sb) && ea_ino) { + err_str = "ea_inode specified without ea_inode feature enabled"; + goto errout; + } + if (ea_ino && ((ea_ino == EXT4_ROOT_INO) || + !ext4_valid_inum(inode->i_sb, ea_ino))) { + err_str = "invalid ea_ino"; + goto errout; + } + if (size > EXT4_XATTR_SIZE_MAX) { + err_str = "e_value size too large"; + goto errout; + } if (size != 0 && entry->e_value_inum == 0) { u16 offs = le16_to_cpu(entry->e_value_offs); @@ -216,66 +262,54 @@ ext4_xattr_check_entries(struct ext4_xattr_entry *entry, void *end, * the padded and unpadded sizes, since the size may * overflow to 0 when adding padding. */ - if (offs > end - value_start) - return -EFSCORRUPTED; + if (offs > end - value_start) { + err_str = "e_value out of bounds"; + goto errout; + } value = value_start + offs; if (value < (void *)e + sizeof(u32) || size > end - value || - EXT4_XATTR_SIZE(size) > end - value) - return -EFSCORRUPTED; + EXT4_XATTR_SIZE(size) > end - value) { + err_str = "overlapping e_value "; + goto errout; + } } entry = EXT4_XATTR_NEXT(entry); } - + if (bh) + set_buffer_verified(bh); return 0; + +errout: + if (bh) + __ext4_error_inode(inode, function, line, 0, -err, + "corrupted xattr block %llu: %s", + (unsigned long long) bh->b_blocknr, + err_str); + else + __ext4_error_inode(inode, function, line, 0, -err, + "corrupted in-inode xattr: %s", err_str); + return err; } static inline int __ext4_xattr_check_block(struct inode *inode, struct buffer_head *bh, const char *function, unsigned int line) { - int error = -EFSCORRUPTED; - - if (BHDR(bh)->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC) || - BHDR(bh)->h_blocks != cpu_to_le32(1)) - goto errout; - if (buffer_verified(bh)) - return 0; - - error = -EFSBADCRC; - if (!ext4_xattr_block_csum_verify(inode, bh)) - goto errout; - error = ext4_xattr_check_entries(BFIRST(bh), bh->b_data + bh->b_size, - bh->b_data); -errout: - if (error) - __ext4_error_inode(inode, function, line, 0, -error, - "corrupted xattr block %llu", - (unsigned long long) bh->b_blocknr); - else - set_buffer_verified(bh); - return error; + return check_xattrs(inode, bh, BFIRST(bh), bh->b_data + bh->b_size, + bh->b_data, function, line); } #define ext4_xattr_check_block(inode, bh) \ __ext4_xattr_check_block((inode), (bh), __func__, __LINE__) -static int +static inline int __xattr_check_inode(struct inode *inode, struct ext4_xattr_ibody_header *header, void *end, const char *function, unsigned int line) { - int error = -EFSCORRUPTED; - - if (end - (void *)header < sizeof(*header) + sizeof(u32) || - (header->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC))) - goto errout; - error = ext4_xattr_check_entries(IFIRST(header), end, IFIRST(header)); -errout: - if (error) - __ext4_error_inode(inode, function, line, 0, -error, - "corrupted in-inode xattr"); - return error; + return check_xattrs(inode, NULL, IFIRST(header), end, IFIRST(header), + function, line); } #define xattr_check_inode(inode, header, end) \ From 11768cfd98136dd8399480c60b7a5d3d3c7b109b Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Fri, 16 Dec 2022 21:02:12 -0800 Subject: [PATCH 156/765] ext4: use ext4_fc_tl_mem in fast-commit replay path To avoid 'sparse' warnings about missing endianness conversions, don't store native endianness values into struct ext4_fc_tl. Instead, use a separate struct type, ext4_fc_tl_mem. Fixes: dcc5827484d6 ("ext4: factor out ext4_fc_get_tl()") Cc: Ye Bin Signed-off-by: Eric Biggers Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20221217050212.150665-1-ebiggers@kernel.org Signed-off-by: Theodore Ts'o --- fs/ext4/fast_commit.c | 44 +++++++++++++++++++++++++------------------ 1 file changed, 26 insertions(+), 18 deletions(-) diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c index 4594b62f147bb..b06de728b3b6c 100644 --- a/fs/ext4/fast_commit.c +++ b/fs/ext4/fast_commit.c @@ -1332,8 +1332,14 @@ struct dentry_info_args { char *dname; }; +/* Same as struct ext4_fc_tl, but uses native endianness fields */ +struct ext4_fc_tl_mem { + u16 fc_tag; + u16 fc_len; +}; + static inline void tl_to_darg(struct dentry_info_args *darg, - struct ext4_fc_tl *tl, u8 *val) + struct ext4_fc_tl_mem *tl, u8 *val) { struct ext4_fc_dentry_info fcd; @@ -1345,16 +1351,18 @@ static inline void tl_to_darg(struct dentry_info_args *darg, darg->dname_len = tl->fc_len - sizeof(struct ext4_fc_dentry_info); } -static inline void ext4_fc_get_tl(struct ext4_fc_tl *tl, u8 *val) +static inline void ext4_fc_get_tl(struct ext4_fc_tl_mem *tl, u8 *val) { - memcpy(tl, val, EXT4_FC_TAG_BASE_LEN); - tl->fc_len = le16_to_cpu(tl->fc_len); - tl->fc_tag = le16_to_cpu(tl->fc_tag); + struct ext4_fc_tl tl_disk; + + memcpy(&tl_disk, val, EXT4_FC_TAG_BASE_LEN); + tl->fc_len = le16_to_cpu(tl_disk.fc_len); + tl->fc_tag = le16_to_cpu(tl_disk.fc_tag); } /* Unlink replay function */ -static int ext4_fc_replay_unlink(struct super_block *sb, struct ext4_fc_tl *tl, - u8 *val) +static int ext4_fc_replay_unlink(struct super_block *sb, + struct ext4_fc_tl_mem *tl, u8 *val) { struct inode *inode, *old_parent; struct qstr entry; @@ -1451,8 +1459,8 @@ static int ext4_fc_replay_link_internal(struct super_block *sb, } /* Link replay function */ -static int ext4_fc_replay_link(struct super_block *sb, struct ext4_fc_tl *tl, - u8 *val) +static int ext4_fc_replay_link(struct super_block *sb, + struct ext4_fc_tl_mem *tl, u8 *val) { struct inode *inode; struct dentry_info_args darg; @@ -1506,8 +1514,8 @@ static int ext4_fc_record_modified_inode(struct super_block *sb, int ino) /* * Inode replay function */ -static int ext4_fc_replay_inode(struct super_block *sb, struct ext4_fc_tl *tl, - u8 *val) +static int ext4_fc_replay_inode(struct super_block *sb, + struct ext4_fc_tl_mem *tl, u8 *val) { struct ext4_fc_inode fc_inode; struct ext4_inode *raw_inode; @@ -1609,8 +1617,8 @@ static int ext4_fc_replay_inode(struct super_block *sb, struct ext4_fc_tl *tl, * inode for which we are trying to create a dentry here, should already have * been replayed before we start here. */ -static int ext4_fc_replay_create(struct super_block *sb, struct ext4_fc_tl *tl, - u8 *val) +static int ext4_fc_replay_create(struct super_block *sb, + struct ext4_fc_tl_mem *tl, u8 *val) { int ret = 0; struct inode *inode = NULL; @@ -1708,7 +1716,7 @@ int ext4_fc_record_regions(struct super_block *sb, int ino, /* Replay add range tag */ static int ext4_fc_replay_add_range(struct super_block *sb, - struct ext4_fc_tl *tl, u8 *val) + struct ext4_fc_tl_mem *tl, u8 *val) { struct ext4_fc_add_range fc_add_ex; struct ext4_extent newex, *ex; @@ -1828,8 +1836,8 @@ static int ext4_fc_replay_add_range(struct super_block *sb, /* Replay DEL_RANGE tag */ static int -ext4_fc_replay_del_range(struct super_block *sb, struct ext4_fc_tl *tl, - u8 *val) +ext4_fc_replay_del_range(struct super_block *sb, + struct ext4_fc_tl_mem *tl, u8 *val) { struct inode *inode; struct ext4_fc_del_range lrange; @@ -2025,7 +2033,7 @@ static int ext4_fc_replay_scan(journal_t *journal, struct ext4_fc_replay_state *state; int ret = JBD2_FC_REPLAY_CONTINUE; struct ext4_fc_add_range ext; - struct ext4_fc_tl tl; + struct ext4_fc_tl_mem tl; struct ext4_fc_tail tail; __u8 *start, *end, *cur, *val; struct ext4_fc_head head; @@ -2144,7 +2152,7 @@ static int ext4_fc_replay(journal_t *journal, struct buffer_head *bh, { struct super_block *sb = journal->j_private; struct ext4_sb_info *sbi = EXT4_SB(sb); - struct ext4_fc_tl tl; + struct ext4_fc_tl_mem tl; __u8 *start, *end, *cur, *val; int ret = JBD2_FC_REPLAY_CONTINUE; struct ext4_fc_replay_state *state = &sbi->s_fc_replay_state; From 934b0de1e9fdea93c4c7f2e18915c54fae67bdc6 Mon Sep 17 00:00:00 2001 From: Wang Jianjian Date: Mon, 19 Dec 2022 09:51:28 +0800 Subject: [PATCH 157/765] ext4: don't show commit interval if it is zero If commit interval is 0, it means using default value. Fixes: 6e47a3cc68fc ("ext4: get rid of super block and sbi from handle_mount_ops()") Signed-off-by: Wang Jianjian Link: https://lore.kernel.org/r/20221219015128.876717-1-wangjianjian3@huawei.com Signed-off-by: Theodore Ts'o --- fs/ext4/super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 260c1b3e3ef2c..3b9e30e1afd91 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -2146,7 +2146,7 @@ static int ext4_parse_param(struct fs_context *fc, struct fs_parameter *param) return 0; case Opt_commit: if (result.uint_32 == 0) - ctx->s_commit_interval = JBD2_DEFAULT_MAX_COMMIT_AGE; + result.uint_32 = JBD2_DEFAULT_MAX_COMMIT_AGE; else if (result.uint_32 > INT_MAX / HZ) { ext4_msg(NULL, KERN_ERR, "Invalid commit interval %d, " From 269d119481008cd725ce32553332593c0ecfc91c Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Tue, 7 Feb 2023 21:48:08 +0800 Subject: [PATCH 158/765] f2fs: fix to do sanity check on extent cache correctly In do_read_inode(), sanity check for extent cache should be called after f2fs_init_read_extent_tree(), fix it. Fixes: 72840cccc0a1 ("f2fs: allocate the extent_cache by default") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/extent_cache.c | 25 +++++++++++++++++++++++++ fs/f2fs/f2fs.h | 1 + fs/f2fs/inode.c | 22 ++++++---------------- 3 files changed, 32 insertions(+), 16 deletions(-) diff --git a/fs/f2fs/extent_cache.c b/fs/f2fs/extent_cache.c index 8d922c592dae3..28b12553f2b34 100644 --- a/fs/f2fs/extent_cache.c +++ b/fs/f2fs/extent_cache.c @@ -19,6 +19,31 @@ #include "node.h" #include +bool sanity_check_extent_cache(struct inode *inode) +{ + struct f2fs_sb_info *sbi = F2FS_I_SB(inode); + struct f2fs_inode_info *fi = F2FS_I(inode); + struct extent_info *ei; + + if (!fi->extent_tree[EX_READ]) + return true; + + ei = &fi->extent_tree[EX_READ]->largest; + + if (ei->len && + (!f2fs_is_valid_blkaddr(sbi, ei->blk, + DATA_GENERIC_ENHANCE) || + !f2fs_is_valid_blkaddr(sbi, ei->blk + ei->len - 1, + DATA_GENERIC_ENHANCE))) { + set_sbi_flag(sbi, SBI_NEED_FSCK); + f2fs_warn(sbi, "%s: inode (ino=%lx) extent info [%u, %u, %u] is incorrect, run fsck to fix", + __func__, inode->i_ino, + ei->blk, ei->fofs, ei->len); + return false; + } + return true; +} + static void __set_extent_info(struct extent_info *ei, unsigned int fofs, unsigned int len, block_t blk, bool keep_clen, diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 44f2d76525bfa..21596e0266ba5 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -4137,6 +4137,7 @@ void f2fs_leave_shrinker(struct f2fs_sb_info *sbi); /* * extent_cache.c */ +bool sanity_check_extent_cache(struct inode *inode); struct rb_entry *f2fs_lookup_rb_tree(struct rb_root_cached *root, struct rb_entry *cached_re, unsigned int ofs); struct rb_node **f2fs_lookup_rb_tree_ext(struct f2fs_sb_info *sbi, diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c index 28c9c72dda2a7..2c250120d5da1 100644 --- a/fs/f2fs/inode.c +++ b/fs/f2fs/inode.c @@ -262,22 +262,6 @@ static bool sanity_check_inode(struct inode *inode, struct page *node_page) return false; } - if (fi->extent_tree[EX_READ]) { - struct extent_info *ei = &fi->extent_tree[EX_READ]->largest; - - if (ei->len && - (!f2fs_is_valid_blkaddr(sbi, ei->blk, - DATA_GENERIC_ENHANCE) || - !f2fs_is_valid_blkaddr(sbi, ei->blk + ei->len - 1, - DATA_GENERIC_ENHANCE))) { - set_sbi_flag(sbi, SBI_NEED_FSCK); - f2fs_warn(sbi, "%s: inode (ino=%lx) extent info [%u, %u, %u] is incorrect, run fsck to fix", - __func__, inode->i_ino, - ei->blk, ei->fofs, ei->len); - return false; - } - } - if (f2fs_sanity_check_inline_data(inode)) { set_sbi_flag(sbi, SBI_NEED_FSCK); f2fs_warn(sbi, "%s: inode (ino=%lx, mode=%u) should not have inline_data, run fsck to fix", @@ -488,6 +472,12 @@ static int do_read_inode(struct inode *inode) return -EFSCORRUPTED; } + if (!sanity_check_extent_cache(inode)) { + f2fs_put_page(node_page, 1); + f2fs_handle_error(sbi, ERROR_CORRUPTED_INODE); + return -EFSCORRUPTED; + } + f2fs_put_page(node_page, 1); stat_inc_inline_xattr(inode); From 19409796578c879a41e88ddbdbce50c19457658d Mon Sep 17 00:00:00 2001 From: Ahmad Fatoum Date: Wed, 23 Nov 2022 10:55:26 +0100 Subject: [PATCH 159/765] include/linux/bcd.h: provide bcd_is_valid() helper bcd2bin(0x0A) happily returns 10, despite this being an invalid BCD value. RTC drivers converting possibly corrupted BCD timestamps might want to validate their input before calling bcd2bin(). Provide a macro to do so. Unlike bcd2bin and bin2bcd, out-of-line versions are not implemented. Should the macro experience enough use, this can be retrofitted. Signed-off-by: Ahmad Fatoum Signed-off-by: Sascha Hauer Link: https://lore.kernel.org/r/20221123095527.2771434-2-s.hauer@pengutronix.de Signed-off-by: Alexandre Belloni --- include/linux/bcd.h | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/include/linux/bcd.h b/include/linux/bcd.h index 118bea36d7d49..abbc8149178e6 100644 --- a/include/linux/bcd.h +++ b/include/linux/bcd.h @@ -14,8 +14,12 @@ const_bin2bcd(x) : \ _bin2bcd(x)) +#define bcd_is_valid(x) \ + const_bcd_is_valid(x) + #define const_bcd2bin(x) (((x) & 0x0f) + ((x) >> 4) * 10) #define const_bin2bcd(x) ((((x) / 10) << 4) + (x) % 10) +#define const_bcd_is_valid(x) (((x) & 0x0f) < 10 && ((x) >> 4) < 10) unsigned _bcd2bin(unsigned char val) __attribute_const__; unsigned char _bin2bcd(unsigned val) __attribute_const__; From e5c594233fcf1a55a439dec103aa815cdbf392a7 Mon Sep 17 00:00:00 2001 From: Sascha Hauer Date: Wed, 23 Nov 2022 10:55:27 +0100 Subject: [PATCH 160/765] rtc: rv8803: invalidate date/time if alarm time is invalid RTC core never calls rv8803_set_alarm with an invalid alarm time, so if an invalid alarm time > 0 is set, external factors must have corrupted the RTC's alarm time and possibly other registers. Play it safe by marking the date/time invalid, so all registers are reinitialized on a ->set_time. This may cause existing setups to lose time if they so far set only date/time, but ignored that the alarm registers had an invalid date value, e.g.: rtc rtc0: invalid alarm value: 2020-3-27 7:82:0 These systems will have their ->get_time return -EINVAL till ->set_time initializes the alarm value (and sets a new time). Signed-off-by: Ahmad Fatoum Signed-off-by: Sascha Hauer Link: https://lore.kernel.org/r/20221123095527.2771434-3-s.hauer@pengutronix.de Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-rv8803.c | 45 +++++++++++++++++++++++++++++++++------- 1 file changed, 38 insertions(+), 7 deletions(-) diff --git a/drivers/rtc/rtc-rv8803.c b/drivers/rtc/rtc-rv8803.c index 53d1de01b7197..25c3b9e4f515a 100644 --- a/drivers/rtc/rtc-rv8803.c +++ b/drivers/rtc/rtc-rv8803.c @@ -70,6 +70,7 @@ struct rv8803_data { struct mutex flags_lock; u8 ctrl; u8 backup; + u8 alarm_invalid:1; enum rv8803_type type; }; @@ -165,13 +166,13 @@ static int rv8803_regs_init(struct rv8803_data *rv8803) static int rv8803_regs_configure(struct rv8803_data *rv8803); -static int rv8803_regs_reset(struct rv8803_data *rv8803) +static int rv8803_regs_reset(struct rv8803_data *rv8803, bool full) { /* * The RV-8803 resets all registers to POR defaults after voltage-loss, * the Epson RTCs don't, so we manually reset the remainder here. */ - if (rv8803->type == rx_8803 || rv8803->type == rx_8900) { + if (full || rv8803->type == rx_8803 || rv8803->type == rx_8900) { int ret = rv8803_regs_init(rv8803); if (ret) return ret; @@ -238,6 +239,11 @@ static int rv8803_get_time(struct device *dev, struct rtc_time *tm) u8 *date = date1; int ret, flags; + if (rv8803->alarm_invalid) { + dev_warn(dev, "Corruption detected, data may be invalid.\n"); + return -EINVAL; + } + flags = rv8803_read_reg(rv8803->client, RV8803_FLAG); if (flags < 0) return flags; @@ -313,12 +319,19 @@ static int rv8803_set_time(struct device *dev, struct rtc_time *tm) return flags; } - if (flags & RV8803_FLAG_V2F) { - ret = rv8803_regs_reset(rv8803); + if ((flags & RV8803_FLAG_V2F) || rv8803->alarm_invalid) { + /* + * If we sense corruption in the alarm registers, but see no + * voltage loss flag, we can't rely on other registers having + * sensible values. Reset them fully. + */ + ret = rv8803_regs_reset(rv8803, rv8803->alarm_invalid); if (ret) { mutex_unlock(&rv8803->flags_lock); return ret; } + + rv8803->alarm_invalid = false; } ret = rv8803_write_reg(rv8803->client, RV8803_FLAG, @@ -344,15 +357,33 @@ static int rv8803_get_alarm(struct device *dev, struct rtc_wkalrm *alrm) if (flags < 0) return flags; + alarmvals[0] &= 0x7f; + alarmvals[1] &= 0x3f; + alarmvals[2] &= 0x3f; + + if (!bcd_is_valid(alarmvals[0]) || + !bcd_is_valid(alarmvals[1]) || + !bcd_is_valid(alarmvals[2])) + goto err_invalid; + alrm->time.tm_sec = 0; - alrm->time.tm_min = bcd2bin(alarmvals[0] & 0x7f); - alrm->time.tm_hour = bcd2bin(alarmvals[1] & 0x3f); - alrm->time.tm_mday = bcd2bin(alarmvals[2] & 0x3f); + alrm->time.tm_min = bcd2bin(alarmvals[0]); + alrm->time.tm_hour = bcd2bin(alarmvals[1]); + alrm->time.tm_mday = bcd2bin(alarmvals[2]); alrm->enabled = !!(rv8803->ctrl & RV8803_CTRL_AIE); alrm->pending = (flags & RV8803_FLAG_AF) && alrm->enabled; + if ((unsigned int)alrm->time.tm_mday > 31 || + (unsigned int)alrm->time.tm_hour >= 24 || + (unsigned int)alrm->time.tm_min >= 60) + goto err_invalid; + return 0; + +err_invalid: + rv8803->alarm_invalid = true; + return -EINVAL; } static int rv8803_set_alarm(struct device *dev, struct rtc_wkalrm *alrm) From 6a22a5dbc69473c2c46d1a13fe6cbc7bdaacb2a6 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Wed, 25 Jan 2023 23:24:24 +0100 Subject: [PATCH 161/765] dt-bindings: rtc: Convert Amlogic Meson vrtc controller binding Convert Amlogic Meson vrtc controller binding to yaml. Signed-off-by: Heiner Kallweit Reviewed-by: Rob Herring Link: https://lore.kernel.org/r/d820d54b-d082-589a-621f-2795d885696a@gmail.com Signed-off-by: Alexandre Belloni --- .../bindings/rtc/amlogic,meson-vrtc.yaml | 44 +++++++++++++++++++ .../bindings/rtc/rtc-meson-vrtc.txt | 22 ---------- 2 files changed, 44 insertions(+), 22 deletions(-) create mode 100644 Documentation/devicetree/bindings/rtc/amlogic,meson-vrtc.yaml delete mode 100644 Documentation/devicetree/bindings/rtc/rtc-meson-vrtc.txt diff --git a/Documentation/devicetree/bindings/rtc/amlogic,meson-vrtc.yaml b/Documentation/devicetree/bindings/rtc/amlogic,meson-vrtc.yaml new file mode 100644 index 0000000000000..a89865fa676a6 --- /dev/null +++ b/Documentation/devicetree/bindings/rtc/amlogic,meson-vrtc.yaml @@ -0,0 +1,44 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/rtc/amlogic,meson-vrtc.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Amlogic Virtual RTC (VRTC) + +maintainers: + - Neil Armstrong + +description: | + This is a Linux interface to an RTC managed by firmware, hence it's + virtual from a Linux perspective. The interface is 1 register where + an alarm time (in seconds) is to be written. + The alarm register is a simple scratch register shared between the + application processors (AP) and the secure co-processor (SCP.) When + the AP suspends, the SCP will use the value of this register to + program an always-on timer before going sleep. When the timer expires, + the SCP will wake up and will then wake the AP. + +allOf: + - $ref: rtc.yaml# + +properties: + compatible: + enum: + - amlogic,meson-vrtc + + reg: + maxItems: 1 + +required: + - compatible + - reg + +additionalProperties: false + +examples: + - | + rtc@a8 { + compatible = "amlogic,meson-vrtc"; + reg = <0x000a8 0x4>; + }; diff --git a/Documentation/devicetree/bindings/rtc/rtc-meson-vrtc.txt b/Documentation/devicetree/bindings/rtc/rtc-meson-vrtc.txt deleted file mode 100644 index c014f54a98530..0000000000000 --- a/Documentation/devicetree/bindings/rtc/rtc-meson-vrtc.txt +++ /dev/null @@ -1,22 +0,0 @@ -* Amlogic Virtual RTC (VRTC) - -This is a Linux interface to an RTC managed by firmware, hence it's -virtual from a Linux perspective. The interface is 1 register where -an alarm time (in seconds) is to be written. - -Required properties: -- compatible: should be "amlogic,meson-vrtc" -- reg: physical address for the alarm register - -The alarm register is a simple scratch register shared between the -application processors (AP) and the secure co-processor (SCP.) When -the AP suspends, the SCP will use the value of this register to -program an always-on timer before going sleep. When the timer expires, -the SCP will wake up and will then wake the AP. - -Example: - - vrtc: rtc@0a8 { - compatible = "amlogic,meson-vrtc"; - reg = <0x0 0x000a8 0x0 0x4>; - }; From c88db0eff9722fc2b6c4d172a50471d20e08ecc6 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:27 +0100 Subject: [PATCH 162/765] rtc: pm8xxx: fix set-alarm race Make sure to disable the alarm before updating the four alarm time registers to avoid spurious alarms during the update. Note that the disable needs to be done outside of the ctrl_reg_lock section to prevent a racing alarm interrupt from disabling the newly set alarm when the lock is released. Fixes: 9a9a54ad7aa2 ("drivers/rtc: add support for Qualcomm PMIC8xxx RTC") Cc: stable@vger.kernel.org # 3.1 Signed-off-by: Johan Hovold Reviewed-by: David Collins Link: https://lore.kernel.org/r/20230202155448.6715-2-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pm8xxx.c | 24 ++++++++++-------------- 1 file changed, 10 insertions(+), 14 deletions(-) diff --git a/drivers/rtc/rtc-pm8xxx.c b/drivers/rtc/rtc-pm8xxx.c index 716e5d9ad74d1..d114f0da537d2 100644 --- a/drivers/rtc/rtc-pm8xxx.c +++ b/drivers/rtc/rtc-pm8xxx.c @@ -221,7 +221,6 @@ static int pm8xxx_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alarm) { int rc, i; u8 value[NUM_8_BIT_RTC_REGS]; - unsigned int ctrl_reg; unsigned long secs, irq_flags; struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; @@ -233,6 +232,11 @@ static int pm8xxx_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alarm) secs >>= 8; } + rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl, + regs->alarm_en, 0); + if (rc) + return rc; + spin_lock_irqsave(&rtc_dd->ctrl_reg_lock, irq_flags); rc = regmap_bulk_write(rtc_dd->regmap, regs->alarm_rw, value, @@ -242,19 +246,11 @@ static int pm8xxx_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alarm) goto rtc_rw_fail; } - rc = regmap_read(rtc_dd->regmap, regs->alarm_ctrl, &ctrl_reg); - if (rc) - goto rtc_rw_fail; - - if (alarm->enabled) - ctrl_reg |= regs->alarm_en; - else - ctrl_reg &= ~regs->alarm_en; - - rc = regmap_write(rtc_dd->regmap, regs->alarm_ctrl, ctrl_reg); - if (rc) { - dev_err(dev, "Write to RTC alarm control register failed\n"); - goto rtc_rw_fail; + if (alarm->enabled) { + rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl, + regs->alarm_en, regs->alarm_en); + if (rc) + goto rtc_rw_fail; } dev_dbg(dev, "Alarm Set for h:m:s=%ptRt, y-m-d=%ptRdr\n", From eb245631836b4843199d7176d1597759dda4ee9e Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:28 +0100 Subject: [PATCH 163/765] rtc: pm8xxx: drop spmi error messages Drop the unnecessary error messages after every spmi regmap access, which are not expected to fail. Signed-off-by: Johan Hovold Link: https://lore.kernel.org/r/20230202155448.6715-3-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pm8xxx.c | 71 ++++++++++------------------------------ 1 file changed, 17 insertions(+), 54 deletions(-) diff --git a/drivers/rtc/rtc-pm8xxx.c b/drivers/rtc/rtc-pm8xxx.c index d114f0da537d2..f49bda999e024 100644 --- a/drivers/rtc/rtc-pm8xxx.c +++ b/drivers/rtc/rtc-pm8xxx.c @@ -105,10 +105,8 @@ static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) alarm_enabled = 1; ctrl_reg &= ~regs->alarm_en; rc = regmap_write(rtc_dd->regmap, regs->alarm_ctrl, ctrl_reg); - if (rc) { - dev_err(dev, "Write to RTC Alarm control register failed\n"); + if (rc) goto rtc_rw_fail; - } } /* Disable RTC H/w before writing on RTC register */ @@ -120,51 +118,39 @@ static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) rtc_disabled = 1; rtc_ctrl_reg &= ~PM8xxx_RTC_ENABLE; rc = regmap_write(rtc_dd->regmap, regs->ctrl, rtc_ctrl_reg); - if (rc) { - dev_err(dev, "Write to RTC control register failed\n"); + if (rc) goto rtc_rw_fail; - } } /* Write 0 to Byte[0] */ rc = regmap_write(rtc_dd->regmap, regs->write, 0); - if (rc) { - dev_err(dev, "Write to RTC write data register failed\n"); + if (rc) goto rtc_rw_fail; - } /* Write Byte[1], Byte[2], Byte[3] */ rc = regmap_bulk_write(rtc_dd->regmap, regs->write + 1, &value[1], sizeof(value) - 1); - if (rc) { - dev_err(dev, "Write to RTC write data register failed\n"); + if (rc) goto rtc_rw_fail; - } /* Write Byte[0] */ rc = regmap_write(rtc_dd->regmap, regs->write, value[0]); - if (rc) { - dev_err(dev, "Write to RTC write data register failed\n"); + if (rc) goto rtc_rw_fail; - } /* Enable RTC H/w after writing on RTC register */ if (rtc_disabled) { rtc_ctrl_reg |= PM8xxx_RTC_ENABLE; rc = regmap_write(rtc_dd->regmap, regs->ctrl, rtc_ctrl_reg); - if (rc) { - dev_err(dev, "Write to RTC control register failed\n"); + if (rc) goto rtc_rw_fail; - } } if (alarm_enabled) { ctrl_reg |= regs->alarm_en; rc = regmap_write(rtc_dd->regmap, regs->alarm_ctrl, ctrl_reg); - if (rc) { - dev_err(dev, "Write to RTC Alarm control register failed\n"); + if (rc) goto rtc_rw_fail; - } } rtc_rw_fail: @@ -183,28 +169,22 @@ static int pm8xxx_rtc_read_time(struct device *dev, struct rtc_time *tm) const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; rc = regmap_bulk_read(rtc_dd->regmap, regs->read, value, sizeof(value)); - if (rc) { - dev_err(dev, "RTC read data register failed\n"); + if (rc) return rc; - } /* * Read the LSB again and check if there has been a carry over. * If there is, redo the read operation. */ rc = regmap_read(rtc_dd->regmap, regs->read, ®); - if (rc < 0) { - dev_err(dev, "RTC read data register failed\n"); + if (rc < 0) return rc; - } if (unlikely(reg < value[0])) { rc = regmap_bulk_read(rtc_dd->regmap, regs->read, value, sizeof(value)); - if (rc) { - dev_err(dev, "RTC read data register failed\n"); + if (rc) return rc; - } } secs = value[0] | (value[1] << 8) | (value[2] << 16) | @@ -241,10 +221,8 @@ static int pm8xxx_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alarm) rc = regmap_bulk_write(rtc_dd->regmap, regs->alarm_rw, value, sizeof(value)); - if (rc) { - dev_err(dev, "Write to RTC ALARM register failed\n"); + if (rc) goto rtc_rw_fail; - } if (alarm->enabled) { rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl, @@ -271,10 +249,8 @@ static int pm8xxx_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alarm) rc = regmap_bulk_read(rtc_dd->regmap, regs->alarm_rw, value, sizeof(value)); - if (rc) { - dev_err(dev, "RTC alarm time read failed\n"); + if (rc) return rc; - } secs = value[0] | (value[1] << 8) | (value[2] << 16) | ((unsigned long)value[3] << 24); @@ -282,10 +258,9 @@ static int pm8xxx_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alarm) rtc_time64_to_tm(secs, &alarm->time); rc = regmap_read(rtc_dd->regmap, regs->alarm_ctrl, &ctrl_reg); - if (rc) { - dev_err(dev, "Read from RTC alarm control register failed\n"); + if (rc) return rc; - } + alarm->enabled = !!(ctrl_reg & PM8xxx_RTC_ALARM_ENABLE); dev_dbg(dev, "Alarm set for - h:m:s=%ptRt, y-m-d=%ptRdr\n", @@ -315,19 +290,15 @@ static int pm8xxx_rtc_alarm_irq_enable(struct device *dev, unsigned int enable) ctrl_reg &= ~regs->alarm_en; rc = regmap_write(rtc_dd->regmap, regs->alarm_ctrl, ctrl_reg); - if (rc) { - dev_err(dev, "Write to RTC control register failed\n"); + if (rc) goto rtc_rw_fail; - } /* Clear Alarm register */ if (!enable) { rc = regmap_bulk_write(rtc_dd->regmap, regs->alarm_rw, value, sizeof(value)); - if (rc) { - dev_err(dev, "Clear RTC ALARM register failed\n"); + if (rc) goto rtc_rw_fail; - } } rtc_rw_fail: @@ -366,8 +337,6 @@ static irqreturn_t pm8xxx_alarm_trigger(int irq, void *dev_id) rc = regmap_write(rtc_dd->regmap, regs->alarm_ctrl, ctrl_reg); if (rc) { spin_unlock(&rtc_dd->ctrl_reg_lock); - dev_err(rtc_dd->rtc_dev, - "Write to alarm control register failed\n"); goto rtc_alarm_handled; } @@ -375,17 +344,11 @@ static irqreturn_t pm8xxx_alarm_trigger(int irq, void *dev_id) /* Clear RTC alarm register */ rc = regmap_read(rtc_dd->regmap, regs->alarm_ctrl2, &ctrl_reg); - if (rc) { - dev_err(rtc_dd->rtc_dev, - "RTC Alarm control2 register read failed\n"); + if (rc) goto rtc_alarm_handled; - } ctrl_reg |= PM8xxx_RTC_ALARM_CLEAR; rc = regmap_write(rtc_dd->regmap, regs->alarm_ctrl2, ctrl_reg); - if (rc) - dev_err(rtc_dd->rtc_dev, - "Write to RTC Alarm control2 register failed\n"); rtc_alarm_handled: return IRQ_HANDLED; From 182c23bbfea3713206b0da3fbbb7350e197a92dd Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:29 +0100 Subject: [PATCH 164/765] rtc: pm8xxx: use regmap_update_bits() Switch to using regmap_update_bits() instead of open coding read-modify-write accesses. Signed-off-by: Johan Hovold Reviewed-by: David Collins Link: https://lore.kernel.org/r/20230202155448.6715-4-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pm8xxx.c | 87 ++++++++++------------------------------ 1 file changed, 22 insertions(+), 65 deletions(-) diff --git a/drivers/rtc/rtc-pm8xxx.c b/drivers/rtc/rtc-pm8xxx.c index f49bda999e024..8c2847ac64f4c 100644 --- a/drivers/rtc/rtc-pm8xxx.c +++ b/drivers/rtc/rtc-pm8xxx.c @@ -78,10 +78,10 @@ static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) { int rc, i; unsigned long secs, irq_flags; - u8 value[NUM_8_BIT_RTC_REGS], alarm_enabled = 0, rtc_disabled = 0; - unsigned int ctrl_reg, rtc_ctrl_reg; struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; + u8 value[NUM_8_BIT_RTC_REGS]; + bool alarm_enabled; if (!rtc_dd->allow_set_time) return -ENODEV; @@ -97,31 +97,16 @@ static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) spin_lock_irqsave(&rtc_dd->ctrl_reg_lock, irq_flags); - rc = regmap_read(rtc_dd->regmap, regs->alarm_ctrl, &ctrl_reg); + rc = regmap_update_bits_check(rtc_dd->regmap, regs->alarm_ctrl, + regs->alarm_en, 0, &alarm_enabled); if (rc) goto rtc_rw_fail; - if (ctrl_reg & regs->alarm_en) { - alarm_enabled = 1; - ctrl_reg &= ~regs->alarm_en; - rc = regmap_write(rtc_dd->regmap, regs->alarm_ctrl, ctrl_reg); - if (rc) - goto rtc_rw_fail; - } - /* Disable RTC H/w before writing on RTC register */ - rc = regmap_read(rtc_dd->regmap, regs->ctrl, &rtc_ctrl_reg); + rc = regmap_update_bits(rtc_dd->regmap, regs->ctrl, PM8xxx_RTC_ENABLE, 0); if (rc) goto rtc_rw_fail; - if (rtc_ctrl_reg & PM8xxx_RTC_ENABLE) { - rtc_disabled = 1; - rtc_ctrl_reg &= ~PM8xxx_RTC_ENABLE; - rc = regmap_write(rtc_dd->regmap, regs->ctrl, rtc_ctrl_reg); - if (rc) - goto rtc_rw_fail; - } - /* Write 0 to Byte[0] */ rc = regmap_write(rtc_dd->regmap, regs->write, 0); if (rc) @@ -139,16 +124,14 @@ static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) goto rtc_rw_fail; /* Enable RTC H/w after writing on RTC register */ - if (rtc_disabled) { - rtc_ctrl_reg |= PM8xxx_RTC_ENABLE; - rc = regmap_write(rtc_dd->regmap, regs->ctrl, rtc_ctrl_reg); - if (rc) - goto rtc_rw_fail; - } + rc = regmap_update_bits(rtc_dd->regmap, regs->ctrl, PM8xxx_RTC_ENABLE, + PM8xxx_RTC_ENABLE); + if (rc) + goto rtc_rw_fail; if (alarm_enabled) { - ctrl_reg |= regs->alarm_en; - rc = regmap_write(rtc_dd->regmap, regs->alarm_ctrl, ctrl_reg); + rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl, + regs->alarm_en, regs->alarm_en); if (rc) goto rtc_rw_fail; } @@ -275,21 +258,18 @@ static int pm8xxx_rtc_alarm_irq_enable(struct device *dev, unsigned int enable) unsigned long irq_flags; struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; - unsigned int ctrl_reg; u8 value[NUM_8_BIT_RTC_REGS] = {0}; + unsigned int val; spin_lock_irqsave(&rtc_dd->ctrl_reg_lock, irq_flags); - rc = regmap_read(rtc_dd->regmap, regs->alarm_ctrl, &ctrl_reg); - if (rc) - goto rtc_rw_fail; - if (enable) - ctrl_reg |= regs->alarm_en; + val = regs->alarm_en; else - ctrl_reg &= ~regs->alarm_en; + val = 0; - rc = regmap_write(rtc_dd->regmap, regs->alarm_ctrl, ctrl_reg); + rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl, + regs->alarm_en, val); if (rc) goto rtc_rw_fail; @@ -318,7 +298,6 @@ static irqreturn_t pm8xxx_alarm_trigger(int irq, void *dev_id) { struct pm8xxx_rtc *rtc_dd = dev_id; const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; - unsigned int ctrl_reg; int rc; rtc_update_irq(rtc_dd->rtc, 1, RTC_IRQF | RTC_AF); @@ -326,15 +305,8 @@ static irqreturn_t pm8xxx_alarm_trigger(int irq, void *dev_id) spin_lock(&rtc_dd->ctrl_reg_lock); /* Clear the alarm enable bit */ - rc = regmap_read(rtc_dd->regmap, regs->alarm_ctrl, &ctrl_reg); - if (rc) { - spin_unlock(&rtc_dd->ctrl_reg_lock); - goto rtc_alarm_handled; - } - - ctrl_reg &= ~regs->alarm_en; - - rc = regmap_write(rtc_dd->regmap, regs->alarm_ctrl, ctrl_reg); + rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl, + regs->alarm_en, 0); if (rc) { spin_unlock(&rtc_dd->ctrl_reg_lock); goto rtc_alarm_handled; @@ -343,13 +315,11 @@ static irqreturn_t pm8xxx_alarm_trigger(int irq, void *dev_id) spin_unlock(&rtc_dd->ctrl_reg_lock); /* Clear RTC alarm register */ - rc = regmap_read(rtc_dd->regmap, regs->alarm_ctrl2, &ctrl_reg); + rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl2, + PM8xxx_RTC_ALARM_CLEAR, 0); if (rc) goto rtc_alarm_handled; - ctrl_reg |= PM8xxx_RTC_ALARM_CLEAR; - rc = regmap_write(rtc_dd->regmap, regs->alarm_ctrl2, ctrl_reg); - rtc_alarm_handled: return IRQ_HANDLED; } @@ -357,22 +327,9 @@ static irqreturn_t pm8xxx_alarm_trigger(int irq, void *dev_id) static int pm8xxx_rtc_enable(struct pm8xxx_rtc *rtc_dd) { const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; - unsigned int ctrl_reg; - int rc; - - /* Check if the RTC is on, else turn it on */ - rc = regmap_read(rtc_dd->regmap, regs->ctrl, &ctrl_reg); - if (rc) - return rc; - if (!(ctrl_reg & PM8xxx_RTC_ENABLE)) { - ctrl_reg |= PM8xxx_RTC_ENABLE; - rc = regmap_write(rtc_dd->regmap, regs->ctrl, ctrl_reg); - if (rc) - return rc; - } - - return 0; + return regmap_update_bits(rtc_dd->regmap, regs->ctrl, PM8xxx_RTC_ENABLE, + PM8xxx_RTC_ENABLE); } static const struct pm8xxx_rtc_regs pm8921_regs = { From 8d273f33fd090a2c270c67b6ac7fa03f5a7eee3f Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:30 +0100 Subject: [PATCH 165/765] rtc: pm8xxx: drop bogus locking Since commit c8d523a4b053 ("drivers/rtc/rtc-pm8xxx.c: rework to support pm8941 rtc") which removed the shadow control register there is no need for a driver lock. Specifically, the rtc ops are serialised by rtc core and the interrupt handler only unconditionally disables the alarm using the alarm_ctrl register. Note that since only the alarm enable bit of alarm_ctrl is used after enabling the RTC at probe, the locking was not needed when doing open coded read-modify-write cycles either. Signed-off-by: Johan Hovold Reviewed-by: David Collins Link: https://lore.kernel.org/r/20230202155448.6715-5-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pm8xxx.c | 67 +++++++++++++--------------------------- 1 file changed, 21 insertions(+), 46 deletions(-) diff --git a/drivers/rtc/rtc-pm8xxx.c b/drivers/rtc/rtc-pm8xxx.c index 8c2847ac64f4c..053a04f74a918 100644 --- a/drivers/rtc/rtc-pm8xxx.c +++ b/drivers/rtc/rtc-pm8xxx.c @@ -53,7 +53,6 @@ struct pm8xxx_rtc_regs { * @rtc_alarm_irq: rtc alarm irq number. * @regs: rtc registers description. * @rtc_dev: device structure. - * @ctrl_reg_lock: spinlock protecting access to ctrl_reg. */ struct pm8xxx_rtc { struct rtc_device *rtc; @@ -62,7 +61,6 @@ struct pm8xxx_rtc { int rtc_alarm_irq; const struct pm8xxx_rtc_regs *regs; struct device *rtc_dev; - spinlock_t ctrl_reg_lock; }; /* @@ -77,11 +75,11 @@ struct pm8xxx_rtc { static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) { int rc, i; - unsigned long secs, irq_flags; struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; u8 value[NUM_8_BIT_RTC_REGS]; bool alarm_enabled; + unsigned long secs; if (!rtc_dd->allow_set_time) return -ENODEV; @@ -95,51 +93,46 @@ static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) secs >>= 8; } - spin_lock_irqsave(&rtc_dd->ctrl_reg_lock, irq_flags); - rc = regmap_update_bits_check(rtc_dd->regmap, regs->alarm_ctrl, regs->alarm_en, 0, &alarm_enabled); if (rc) - goto rtc_rw_fail; + return rc; /* Disable RTC H/w before writing on RTC register */ rc = regmap_update_bits(rtc_dd->regmap, regs->ctrl, PM8xxx_RTC_ENABLE, 0); if (rc) - goto rtc_rw_fail; + return rc; /* Write 0 to Byte[0] */ rc = regmap_write(rtc_dd->regmap, regs->write, 0); if (rc) - goto rtc_rw_fail; + return rc; /* Write Byte[1], Byte[2], Byte[3] */ rc = regmap_bulk_write(rtc_dd->regmap, regs->write + 1, &value[1], sizeof(value) - 1); if (rc) - goto rtc_rw_fail; + return rc; /* Write Byte[0] */ rc = regmap_write(rtc_dd->regmap, regs->write, value[0]); if (rc) - goto rtc_rw_fail; + return rc; /* Enable RTC H/w after writing on RTC register */ rc = regmap_update_bits(rtc_dd->regmap, regs->ctrl, PM8xxx_RTC_ENABLE, PM8xxx_RTC_ENABLE); if (rc) - goto rtc_rw_fail; + return rc; if (alarm_enabled) { rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl, regs->alarm_en, regs->alarm_en); if (rc) - goto rtc_rw_fail; + return rc; } -rtc_rw_fail: - spin_unlock_irqrestore(&rtc_dd->ctrl_reg_lock, irq_flags); - - return rc; + return 0; } static int pm8xxx_rtc_read_time(struct device *dev, struct rtc_time *tm) @@ -184,9 +177,9 @@ static int pm8xxx_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alarm) { int rc, i; u8 value[NUM_8_BIT_RTC_REGS]; - unsigned long secs, irq_flags; struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; + unsigned long secs; secs = rtc_tm_to_time64(&alarm->time); @@ -200,25 +193,22 @@ static int pm8xxx_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alarm) if (rc) return rc; - spin_lock_irqsave(&rtc_dd->ctrl_reg_lock, irq_flags); - rc = regmap_bulk_write(rtc_dd->regmap, regs->alarm_rw, value, sizeof(value)); if (rc) - goto rtc_rw_fail; + return rc; if (alarm->enabled) { rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl, regs->alarm_en, regs->alarm_en); if (rc) - goto rtc_rw_fail; + return rc; } dev_dbg(dev, "Alarm Set for h:m:s=%ptRt, y-m-d=%ptRdr\n", &alarm->time, &alarm->time); -rtc_rw_fail: - spin_unlock_irqrestore(&rtc_dd->ctrl_reg_lock, irq_flags); - return rc; + + return 0; } static int pm8xxx_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alarm) @@ -255,14 +245,11 @@ static int pm8xxx_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alarm) static int pm8xxx_rtc_alarm_irq_enable(struct device *dev, unsigned int enable) { int rc; - unsigned long irq_flags; struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; u8 value[NUM_8_BIT_RTC_REGS] = {0}; unsigned int val; - spin_lock_irqsave(&rtc_dd->ctrl_reg_lock, irq_flags); - if (enable) val = regs->alarm_en; else @@ -271,19 +258,17 @@ static int pm8xxx_rtc_alarm_irq_enable(struct device *dev, unsigned int enable) rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl, regs->alarm_en, val); if (rc) - goto rtc_rw_fail; + return rc; /* Clear Alarm register */ if (!enable) { rc = regmap_bulk_write(rtc_dd->regmap, regs->alarm_rw, value, sizeof(value)); if (rc) - goto rtc_rw_fail; + return rc; } -rtc_rw_fail: - spin_unlock_irqrestore(&rtc_dd->ctrl_reg_lock, irq_flags); - return rc; + return 0; } static const struct rtc_class_ops pm8xxx_rtc_ops = { @@ -302,25 +287,18 @@ static irqreturn_t pm8xxx_alarm_trigger(int irq, void *dev_id) rtc_update_irq(rtc_dd->rtc, 1, RTC_IRQF | RTC_AF); - spin_lock(&rtc_dd->ctrl_reg_lock); - /* Clear the alarm enable bit */ rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl, regs->alarm_en, 0); - if (rc) { - spin_unlock(&rtc_dd->ctrl_reg_lock); - goto rtc_alarm_handled; - } - - spin_unlock(&rtc_dd->ctrl_reg_lock); + if (rc) + goto out; /* Clear RTC alarm register */ rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl2, PM8xxx_RTC_ALARM_CLEAR, 0); if (rc) - goto rtc_alarm_handled; - -rtc_alarm_handled: + goto out; +out: return IRQ_HANDLED; } @@ -398,9 +376,6 @@ static int pm8xxx_rtc_probe(struct platform_device *pdev) if (rtc_dd == NULL) return -ENOMEM; - /* Initialise spinlock to protect RTC control register */ - spin_lock_init(&rtc_dd->ctrl_reg_lock); - rtc_dd->regmap = dev_get_regmap(pdev->dev.parent, NULL); if (!rtc_dd->regmap) { dev_err(&pdev->dev, "Parent regmap unavailable.\n"); From cb9bb7b2364bb5f4f51226ce1f9ec6ffda618f0a Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:31 +0100 Subject: [PATCH 166/765] rtc: pm8xxx: return IRQ_NONE on errors In the unlikely event that disabling the alarm and clearing the status ever fails, return IRQ_NONE instead of IRQ_HANDLED. Signed-off-by: Johan Hovold Link: https://lore.kernel.org/r/20230202155448.6715-6-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pm8xxx.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/rtc/rtc-pm8xxx.c b/drivers/rtc/rtc-pm8xxx.c index 053a04f74a918..dc7e659cbb2aa 100644 --- a/drivers/rtc/rtc-pm8xxx.c +++ b/drivers/rtc/rtc-pm8xxx.c @@ -291,14 +291,14 @@ static irqreturn_t pm8xxx_alarm_trigger(int irq, void *dev_id) rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl, regs->alarm_en, 0); if (rc) - goto out; + return IRQ_NONE; /* Clear RTC alarm register */ rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl2, PM8xxx_RTC_ALARM_CLEAR, 0); if (rc) - goto out; -out: + return IRQ_NONE; + return IRQ_HANDLED; } From f081b74c1c748a7da972c782c2f974f239a9b51f Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:32 +0100 Subject: [PATCH 167/765] rtc: pm8xxx: drop unused register defines Drop the original register defines which have been unused since commit c8d523a4b053 ("drivers/rtc/rtc-pm8xxx.c: rework to support pm8941 rtc"). Signed-off-by: Johan Hovold Reviewed-by: David Collins Link: https://lore.kernel.org/r/20230202155448.6715-7-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pm8xxx.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/drivers/rtc/rtc-pm8xxx.c b/drivers/rtc/rtc-pm8xxx.c index dc7e659cbb2aa..90027a7cfb128 100644 --- a/drivers/rtc/rtc-pm8xxx.c +++ b/drivers/rtc/rtc-pm8xxx.c @@ -12,12 +12,6 @@ #include #include -/* RTC Register offsets from RTC CTRL REG */ -#define PM8XXX_ALARM_CTRL_OFFSET 0x01 -#define PM8XXX_RTC_WRITE_OFFSET 0x02 -#define PM8XXX_RTC_READ_OFFSET 0x06 -#define PM8XXX_ALARM_RW_OFFSET 0x0A - /* RTC_CTRL register bit fields */ #define PM8xxx_RTC_ENABLE BIT(7) #define PM8xxx_RTC_ALARM_CLEAR BIT(0) From 79dd75661e4284169768859012a4bf6898cef758 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:33 +0100 Subject: [PATCH 168/765] rtc: pm8xxx: use unaligned le32 helpers Use the unaligned le32 helpers instead of open coding when accessing the time and alarm registers. Signed-off-by: Johan Hovold Reviewed-by: David Collins Link: https://lore.kernel.org/r/20230202155448.6715-8-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pm8xxx.c | 26 ++++++++------------------ 1 file changed, 8 insertions(+), 18 deletions(-) diff --git a/drivers/rtc/rtc-pm8xxx.c b/drivers/rtc/rtc-pm8xxx.c index 90027a7cfb128..5ff6898bcacec 100644 --- a/drivers/rtc/rtc-pm8xxx.c +++ b/drivers/rtc/rtc-pm8xxx.c @@ -12,6 +12,8 @@ #include #include +#include + /* RTC_CTRL register bit fields */ #define PM8xxx_RTC_ENABLE BIT(7) #define PM8xxx_RTC_ALARM_CLEAR BIT(0) @@ -68,25 +70,21 @@ struct pm8xxx_rtc { */ static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) { - int rc, i; struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; u8 value[NUM_8_BIT_RTC_REGS]; bool alarm_enabled; unsigned long secs; + int rc; if (!rtc_dd->allow_set_time) return -ENODEV; secs = rtc_tm_to_time64(tm); + put_unaligned_le32(secs, value); dev_dbg(dev, "Seconds value to be written to RTC = %lu\n", secs); - for (i = 0; i < NUM_8_BIT_RTC_REGS; i++) { - value[i] = secs & 0xFF; - secs >>= 8; - } - rc = regmap_update_bits_check(rtc_dd->regmap, regs->alarm_ctrl, regs->alarm_en, 0, &alarm_enabled); if (rc) @@ -157,9 +155,7 @@ static int pm8xxx_rtc_read_time(struct device *dev, struct rtc_time *tm) return rc; } - secs = value[0] | (value[1] << 8) | (value[2] << 16) | - ((unsigned long)value[3] << 24); - + secs = get_unaligned_le32(value); rtc_time64_to_tm(secs, tm); dev_dbg(dev, "secs = %lu, h:m:s == %ptRt, y-m-d = %ptRdr\n", secs, tm, tm); @@ -169,18 +165,14 @@ static int pm8xxx_rtc_read_time(struct device *dev, struct rtc_time *tm) static int pm8xxx_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alarm) { - int rc, i; u8 value[NUM_8_BIT_RTC_REGS]; struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; unsigned long secs; + int rc; secs = rtc_tm_to_time64(&alarm->time); - - for (i = 0; i < NUM_8_BIT_RTC_REGS; i++) { - value[i] = secs & 0xFF; - secs >>= 8; - } + put_unaligned_le32(secs, value); rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl, regs->alarm_en, 0); @@ -219,9 +211,7 @@ static int pm8xxx_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alarm) if (rc) return rc; - secs = value[0] | (value[1] << 8) | (value[2] << 16) | - ((unsigned long)value[3] << 24); - + secs = get_unaligned_le32(value); rtc_time64_to_tm(secs, &alarm->time); rc = regmap_read(rtc_dd->regmap, regs->alarm_ctrl, &ctrl_reg); From c996956fcc5b7756eb04615cc36618acaa85caa9 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:34 +0100 Subject: [PATCH 169/765] rtc: pm8xxx: clean up time and alarm debugging Clean up the time and alarm callback debugging by using a consistent and succinct human-readable (i.e. non-raw) format. Signed-off-by: Johan Hovold Reviewed-by: David Collins Link: https://lore.kernel.org/r/20230202155448.6715-9-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pm8xxx.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/drivers/rtc/rtc-pm8xxx.c b/drivers/rtc/rtc-pm8xxx.c index 5ff6898bcacec..8c364e5d3b574 100644 --- a/drivers/rtc/rtc-pm8xxx.c +++ b/drivers/rtc/rtc-pm8xxx.c @@ -83,7 +83,7 @@ static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) secs = rtc_tm_to_time64(tm); put_unaligned_le32(secs, value); - dev_dbg(dev, "Seconds value to be written to RTC = %lu\n", secs); + dev_dbg(dev, "set time: %ptRd %ptRt (%lu)\n", tm, tm, secs); rc = regmap_update_bits_check(rtc_dd->regmap, regs->alarm_ctrl, regs->alarm_en, 0, &alarm_enabled); @@ -158,7 +158,7 @@ static int pm8xxx_rtc_read_time(struct device *dev, struct rtc_time *tm) secs = get_unaligned_le32(value); rtc_time64_to_tm(secs, tm); - dev_dbg(dev, "secs = %lu, h:m:s == %ptRt, y-m-d = %ptRdr\n", secs, tm, tm); + dev_dbg(dev, "read time: %ptRd %ptRt (%lu)\n", tm, tm, secs); return 0; } @@ -191,8 +191,7 @@ static int pm8xxx_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alarm) return rc; } - dev_dbg(dev, "Alarm Set for h:m:s=%ptRt, y-m-d=%ptRdr\n", - &alarm->time, &alarm->time); + dev_dbg(dev, "set alarm: %ptRd %ptRt\n", &alarm->time, &alarm->time); return 0; } @@ -220,8 +219,7 @@ static int pm8xxx_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alarm) alarm->enabled = !!(ctrl_reg & PM8xxx_RTC_ALARM_ENABLE); - dev_dbg(dev, "Alarm set for - h:m:s=%ptRt, y-m-d=%ptRdr\n", - &alarm->time, &alarm->time); + dev_dbg(dev, "read alarm: %ptRd %ptRt\n", &alarm->time, &alarm->time); return 0; } From a375510efeda0dfbad205cd1de8b57f63d0779c9 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:35 +0100 Subject: [PATCH 170/765] rtc: pm8xxx: rename struct device pointer Rename the driver-data struct device pointer by dropping the "rtc" prefix which is both redundant and misleading (as this is a pointer to the platform device and not the rtc class device). Signed-off-by: Johan Hovold Reviewed-by: David Collins Link: https://lore.kernel.org/r/20230202155448.6715-10-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pm8xxx.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/rtc/rtc-pm8xxx.c b/drivers/rtc/rtc-pm8xxx.c index 8c364e5d3b574..0fdbd233b10e4 100644 --- a/drivers/rtc/rtc-pm8xxx.c +++ b/drivers/rtc/rtc-pm8xxx.c @@ -48,7 +48,7 @@ struct pm8xxx_rtc_regs { * @allow_set_time: indicates whether writing to the RTC is allowed * @rtc_alarm_irq: rtc alarm irq number. * @regs: rtc registers description. - * @rtc_dev: device structure. + * @dev: device structure */ struct pm8xxx_rtc { struct rtc_device *rtc; @@ -56,7 +56,7 @@ struct pm8xxx_rtc { bool allow_set_time; int rtc_alarm_irq; const struct pm8xxx_rtc_regs *regs; - struct device *rtc_dev; + struct device *dev; }; /* @@ -372,7 +372,7 @@ static int pm8xxx_rtc_probe(struct platform_device *pdev) "allow-set-time"); rtc_dd->regs = match->data; - rtc_dd->rtc_dev = &pdev->dev; + rtc_dd->dev = &pdev->dev; rc = pm8xxx_rtc_enable(rtc_dd); if (rc) From 4727b58fc84daf6d7097ac3528a6517456a5e110 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:36 +0100 Subject: [PATCH 171/765] rtc: pm8xxx: rename alarm irq variable Clean up the driver somewhat by renaming the driver-data alarm irq variable by dropping the redundant "rtc" prefix. Signed-off-by: Johan Hovold Reviewed-by: David Collins Link: https://lore.kernel.org/r/20230202155448.6715-11-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pm8xxx.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/rtc/rtc-pm8xxx.c b/drivers/rtc/rtc-pm8xxx.c index 0fdbd233b10e4..ea867b20573a6 100644 --- a/drivers/rtc/rtc-pm8xxx.c +++ b/drivers/rtc/rtc-pm8xxx.c @@ -46,7 +46,7 @@ struct pm8xxx_rtc_regs { * @rtc: rtc device for this driver. * @regmap: regmap used to access RTC registers * @allow_set_time: indicates whether writing to the RTC is allowed - * @rtc_alarm_irq: rtc alarm irq number. + * @alarm_irq: alarm irq number * @regs: rtc registers description. * @dev: device structure */ @@ -54,7 +54,7 @@ struct pm8xxx_rtc { struct rtc_device *rtc; struct regmap *regmap; bool allow_set_time; - int rtc_alarm_irq; + int alarm_irq; const struct pm8xxx_rtc_regs *regs; struct device *dev; }; @@ -364,8 +364,8 @@ static int pm8xxx_rtc_probe(struct platform_device *pdev) return -ENXIO; } - rtc_dd->rtc_alarm_irq = platform_get_irq(pdev, 0); - if (rtc_dd->rtc_alarm_irq < 0) + rtc_dd->alarm_irq = platform_get_irq(pdev, 0); + if (rtc_dd->alarm_irq < 0) return -ENXIO; rtc_dd->allow_set_time = of_property_read_bool(pdev->dev.of_node, @@ -391,7 +391,7 @@ static int pm8xxx_rtc_probe(struct platform_device *pdev) rtc_dd->rtc->range_max = U32_MAX; /* Request the alarm IRQ */ - rc = devm_request_any_context_irq(&pdev->dev, rtc_dd->rtc_alarm_irq, + rc = devm_request_any_context_irq(&pdev->dev, rtc_dd->alarm_irq, pm8xxx_alarm_trigger, IRQF_TRIGGER_RISING, "pm8xxx_rtc_alarm", rtc_dd); @@ -404,7 +404,7 @@ static int pm8xxx_rtc_probe(struct platform_device *pdev) if (rc) return rc; - rc = dev_pm_set_wake_irq(&pdev->dev, rtc_dd->rtc_alarm_irq); + rc = dev_pm_set_wake_irq(&pdev->dev, rtc_dd->alarm_irq); if (rc) return rc; From 3c3326394ba420608d0665aef846b2268c9c9629 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:37 +0100 Subject: [PATCH 172/765] rtc: pm8xxx: clean up comments Clean up the driver comments somewhat and remove obsolete, incorrect or redundant ones. Signed-off-by: Johan Hovold Reviewed-by: David Collins Link: https://lore.kernel.org/r/20230202155448.6715-12-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pm8xxx.c | 39 +++++++++++++++++---------------------- 1 file changed, 17 insertions(+), 22 deletions(-) diff --git a/drivers/rtc/rtc-pm8xxx.c b/drivers/rtc/rtc-pm8xxx.c index ea867b20573a6..8a94a19e0d145 100644 --- a/drivers/rtc/rtc-pm8xxx.c +++ b/drivers/rtc/rtc-pm8xxx.c @@ -23,13 +23,13 @@ /** * struct pm8xxx_rtc_regs - describe RTC registers per PMIC versions - * @ctrl: base address of control register - * @write: base address of write register - * @read: base address of read register - * @alarm_ctrl: base address of alarm control register - * @alarm_ctrl2: base address of alarm control2 register - * @alarm_rw: base address of alarm read-write register - * @alarm_en: alarm enable mask + * @ctrl: address of control register + * @write: base address of write registers + * @read: base address of read registers + * @alarm_ctrl: address of alarm control register + * @alarm_ctrl2: address of alarm control2 register + * @alarm_rw: base address of alarm read-write registers + * @alarm_en: alarm enable mask */ struct pm8xxx_rtc_regs { unsigned int ctrl; @@ -42,12 +42,12 @@ struct pm8xxx_rtc_regs { }; /** - * struct pm8xxx_rtc - rtc driver internal structure - * @rtc: rtc device for this driver. - * @regmap: regmap used to access RTC registers - * @allow_set_time: indicates whether writing to the RTC is allowed + * struct pm8xxx_rtc - RTC driver internal structure + * @rtc: RTC device + * @regmap: regmap used to access registers + * @allow_set_time: whether the time can be set * @alarm_irq: alarm irq number - * @regs: rtc registers description. + * @regs: register description * @dev: device structure */ struct pm8xxx_rtc { @@ -90,7 +90,7 @@ static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) if (rc) return rc; - /* Disable RTC H/w before writing on RTC register */ + /* Disable RTC */ rc = regmap_update_bits(rtc_dd->regmap, regs->ctrl, PM8xxx_RTC_ENABLE, 0); if (rc) return rc; @@ -111,7 +111,7 @@ static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) if (rc) return rc; - /* Enable RTC H/w after writing on RTC register */ + /* Enable RTC */ rc = regmap_update_bits(rtc_dd->regmap, regs->ctrl, PM8xxx_RTC_ENABLE, PM8xxx_RTC_ENABLE); if (rc) @@ -242,7 +242,7 @@ static int pm8xxx_rtc_alarm_irq_enable(struct device *dev, unsigned int enable) if (rc) return rc; - /* Clear Alarm register */ + /* Clear alarm register */ if (!enable) { rc = regmap_bulk_write(rtc_dd->regmap, regs->alarm_rw, value, sizeof(value)); @@ -269,13 +269,13 @@ static irqreturn_t pm8xxx_alarm_trigger(int irq, void *dev_id) rtc_update_irq(rtc_dd->rtc, 1, RTC_IRQF | RTC_AF); - /* Clear the alarm enable bit */ + /* Disable alarm */ rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl, regs->alarm_en, 0); if (rc) return IRQ_NONE; - /* Clear RTC alarm register */ + /* Clear alarm status */ rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl2, PM8xxx_RTC_ALARM_CLEAR, 0); if (rc) @@ -332,9 +332,6 @@ static const struct pm8xxx_rtc_regs pmk8350_regs = { .alarm_en = BIT(7), }; -/* - * Hardcoded RTC bases until IORESOURCE_REG mapping is figured out - */ static const struct of_device_id pm8xxx_id_table[] = { { .compatible = "qcom,pm8921-rtc", .data = &pm8921_regs }, { .compatible = "qcom,pm8058-rtc", .data = &pm8058_regs }, @@ -382,7 +379,6 @@ static int pm8xxx_rtc_probe(struct platform_device *pdev) device_init_wakeup(&pdev->dev, 1); - /* Register the RTC device */ rtc_dd->rtc = devm_rtc_allocate_device(&pdev->dev); if (IS_ERR(rtc_dd->rtc)) return PTR_ERR(rtc_dd->rtc); @@ -390,7 +386,6 @@ static int pm8xxx_rtc_probe(struct platform_device *pdev) rtc_dd->rtc->ops = &pm8xxx_rtc_ops; rtc_dd->rtc->range_max = U32_MAX; - /* Request the alarm IRQ */ rc = devm_request_any_context_irq(&pdev->dev, rtc_dd->alarm_irq, pm8xxx_alarm_trigger, IRQF_TRIGGER_RISING, From 35d9c472925748a1cb1f5b6cc8ae71cf8138e30f Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:38 +0100 Subject: [PATCH 173/765] rtc: pm8xxx: use u32 for timestamps The PMIC RTC registers are 32-bit so explicitly use u32 rather than unsigned long for timestamps to reflect the hardware. This will also help avoid unintentional range extensions when adding support for managing an external offset. Signed-off-by: Johan Hovold Link: https://lore.kernel.org/r/20230202155448.6715-13-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pm8xxx.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/rtc/rtc-pm8xxx.c b/drivers/rtc/rtc-pm8xxx.c index 8a94a19e0d145..b1ce246c501a1 100644 --- a/drivers/rtc/rtc-pm8xxx.c +++ b/drivers/rtc/rtc-pm8xxx.c @@ -74,7 +74,7 @@ static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; u8 value[NUM_8_BIT_RTC_REGS]; bool alarm_enabled; - unsigned long secs; + u32 secs; int rc; if (!rtc_dd->allow_set_time) @@ -83,7 +83,7 @@ static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) secs = rtc_tm_to_time64(tm); put_unaligned_le32(secs, value); - dev_dbg(dev, "set time: %ptRd %ptRt (%lu)\n", tm, tm, secs); + dev_dbg(dev, "set time: %ptRd %ptRt (%u)\n", tm, tm, secs); rc = regmap_update_bits_check(rtc_dd->regmap, regs->alarm_ctrl, regs->alarm_en, 0, &alarm_enabled); @@ -131,10 +131,10 @@ static int pm8xxx_rtc_read_time(struct device *dev, struct rtc_time *tm) { int rc; u8 value[NUM_8_BIT_RTC_REGS]; - unsigned long secs; unsigned int reg; struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; + u32 secs; rc = regmap_bulk_read(rtc_dd->regmap, regs->read, value, sizeof(value)); if (rc) @@ -158,7 +158,7 @@ static int pm8xxx_rtc_read_time(struct device *dev, struct rtc_time *tm) secs = get_unaligned_le32(value); rtc_time64_to_tm(secs, tm); - dev_dbg(dev, "read time: %ptRd %ptRt (%lu)\n", tm, tm, secs); + dev_dbg(dev, "read time: %ptRd %ptRt (%u)\n", tm, tm, secs); return 0; } @@ -168,7 +168,7 @@ static int pm8xxx_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alarm) u8 value[NUM_8_BIT_RTC_REGS]; struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; - unsigned long secs; + u32 secs; int rc; secs = rtc_tm_to_time64(&alarm->time); @@ -201,9 +201,9 @@ static int pm8xxx_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alarm) int rc; unsigned int ctrl_reg; u8 value[NUM_8_BIT_RTC_REGS]; - unsigned long secs; struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; + u32 secs; rc = regmap_bulk_read(rtc_dd->regmap, regs->alarm_rw, value, sizeof(value)); From da862c3df6add928e2f51d6cadec128a9a1940f3 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:39 +0100 Subject: [PATCH 174/765] rtc: pm8xxx: refactor read_time() In preparation for adding support for setting the time by means of an externally stored offset, refactor read_time() by adding a new helper that can be used to retrieve the raw time as stored in the RTC. Signed-off-by: Johan Hovold Link: https://lore.kernel.org/r/20230202155448.6715-14-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pm8xxx.c | 54 ++++++++++++++++++++++++---------------- 1 file changed, 33 insertions(+), 21 deletions(-) diff --git a/drivers/rtc/rtc-pm8xxx.c b/drivers/rtc/rtc-pm8xxx.c index b1ce246c501a1..2f96a178595c9 100644 --- a/drivers/rtc/rtc-pm8xxx.c +++ b/drivers/rtc/rtc-pm8xxx.c @@ -59,6 +59,37 @@ struct pm8xxx_rtc { struct device *dev; }; +static int pm8xxx_rtc_read_raw(struct pm8xxx_rtc *rtc_dd, u32 *secs) +{ + const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; + u8 value[NUM_8_BIT_RTC_REGS]; + unsigned int reg; + int rc; + + rc = regmap_bulk_read(rtc_dd->regmap, regs->read, value, sizeof(value)); + if (rc) + return rc; + + /* + * Read the LSB again and check if there has been a carry over. + * If there has, redo the read operation. + */ + rc = regmap_read(rtc_dd->regmap, regs->read, ®); + if (rc < 0) + return rc; + + if (reg < value[0]) { + rc = regmap_bulk_read(rtc_dd->regmap, regs->read, value, + sizeof(value)); + if (rc) + return rc; + } + + *secs = get_unaligned_le32(value); + + return 0; +} + /* * Steps to write the RTC registers. * 1. Disable alarm if enabled. @@ -129,33 +160,14 @@ static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) static int pm8xxx_rtc_read_time(struct device *dev, struct rtc_time *tm) { - int rc; - u8 value[NUM_8_BIT_RTC_REGS]; - unsigned int reg; struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); - const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; u32 secs; + int rc; - rc = regmap_bulk_read(rtc_dd->regmap, regs->read, value, sizeof(value)); + rc = pm8xxx_rtc_read_raw(rtc_dd, &secs); if (rc) return rc; - /* - * Read the LSB again and check if there has been a carry over. - * If there is, redo the read operation. - */ - rc = regmap_read(rtc_dd->regmap, regs->read, ®); - if (rc < 0) - return rc; - - if (unlikely(reg < value[0])) { - rc = regmap_bulk_read(rtc_dd->regmap, regs->read, - value, sizeof(value)); - if (rc) - return rc; - } - - secs = get_unaligned_le32(value); rtc_time64_to_tm(secs, tm); dev_dbg(dev, "read time: %ptRd %ptRt (%u)\n", tm, tm, secs); From 9e5a799138042ac8276e6744c548b0411f371600 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:40 +0100 Subject: [PATCH 175/765] rtc: pm8xxx: clean up local declarations Clean up local declarations somewhat by using the reverse xmas style consistently throughout. Signed-off-by: Johan Hovold Reviewed-by: David Collins Link: https://lore.kernel.org/r/20230202155448.6715-15-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pm8xxx.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/rtc/rtc-pm8xxx.c b/drivers/rtc/rtc-pm8xxx.c index 2f96a178595c9..922aef0f02416 100644 --- a/drivers/rtc/rtc-pm8xxx.c +++ b/drivers/rtc/rtc-pm8xxx.c @@ -177,9 +177,9 @@ static int pm8xxx_rtc_read_time(struct device *dev, struct rtc_time *tm) static int pm8xxx_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alarm) { - u8 value[NUM_8_BIT_RTC_REGS]; struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; + u8 value[NUM_8_BIT_RTC_REGS]; u32 secs; int rc; @@ -210,12 +210,12 @@ static int pm8xxx_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alarm) static int pm8xxx_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alarm) { - int rc; - unsigned int ctrl_reg; - u8 value[NUM_8_BIT_RTC_REGS]; struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; + u8 value[NUM_8_BIT_RTC_REGS]; + unsigned int ctrl_reg; u32 secs; + int rc; rc = regmap_bulk_read(rtc_dd->regmap, regs->alarm_rw, value, sizeof(value)); @@ -238,11 +238,11 @@ static int pm8xxx_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alarm) static int pm8xxx_rtc_alarm_irq_enable(struct device *dev, unsigned int enable) { - int rc; struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; u8 value[NUM_8_BIT_RTC_REGS] = {0}; unsigned int val; + int rc; if (enable) val = regs->alarm_en; @@ -355,9 +355,9 @@ MODULE_DEVICE_TABLE(of, pm8xxx_id_table); static int pm8xxx_rtc_probe(struct platform_device *pdev) { - int rc; - struct pm8xxx_rtc *rtc_dd; const struct of_device_id *match; + struct pm8xxx_rtc *rtc_dd; + int rc; match = of_match_node(pm8xxx_id_table, pdev->dev.of_node); if (!match) From c94fb939e65155bc889e62396f83ef4317d643ac Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:41 +0100 Subject: [PATCH 176/765] rtc: pm8xxx: drop error messages For consistency with the rest of the driver, drop the last two error messages for conditions that should only occur during development, if ever. Signed-off-by: Johan Hovold Link: https://lore.kernel.org/r/20230202155448.6715-16-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pm8xxx.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/drivers/rtc/rtc-pm8xxx.c b/drivers/rtc/rtc-pm8xxx.c index 922aef0f02416..eff2782beeed6 100644 --- a/drivers/rtc/rtc-pm8xxx.c +++ b/drivers/rtc/rtc-pm8xxx.c @@ -368,10 +368,8 @@ static int pm8xxx_rtc_probe(struct platform_device *pdev) return -ENOMEM; rtc_dd->regmap = dev_get_regmap(pdev->dev.parent, NULL); - if (!rtc_dd->regmap) { - dev_err(&pdev->dev, "Parent regmap unavailable.\n"); + if (!rtc_dd->regmap) return -ENXIO; - } rtc_dd->alarm_irq = platform_get_irq(pdev, 0); if (rtc_dd->alarm_irq < 0) @@ -402,10 +400,8 @@ static int pm8xxx_rtc_probe(struct platform_device *pdev) pm8xxx_alarm_trigger, IRQF_TRIGGER_RISING, "pm8xxx_rtc_alarm", rtc_dd); - if (rc < 0) { - dev_err(&pdev->dev, "Request IRQ failed (%d)\n", rc); + if (rc < 0) return rc; - } rc = devm_rtc_register_device(rtc_dd->rtc); if (rc) From 2985cda83b8107213e108f95c3cb8caa0da74918 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Tue, 31 Jan 2023 21:48:13 -0800 Subject: [PATCH 177/765] rtc: moxart: switch to using gpiod API Switch the driver from legacy gpio API that is deprecated to the newer gpiod API that respects line polarities described in ACPI/DT. This makes driver use standard property name for its gpios ("rtc-*-gpios" vs "gpios-rtc-*"), however there is a quirk in gpiolib to also recognize legacy names and keep compatibility with older DTSes: eaf1a29665cd ("gpiolib: of: add a quirk for legacy names in MOXA ART RTC"). Signed-off-by: Dmitry Torokhov Link: https://lore.kernel.org/r/20230201054815.4112632-1-dmitry.torokhov@gmail.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-moxart.c | 89 ++++++++++++++++------------------------ 1 file changed, 36 insertions(+), 53 deletions(-) diff --git a/drivers/rtc/rtc-moxart.c b/drivers/rtc/rtc-moxart.c index 6b24ac9e1cfa8..2247dd39ee4b6 100644 --- a/drivers/rtc/rtc-moxart.c +++ b/drivers/rtc/rtc-moxart.c @@ -10,14 +10,15 @@ * Moxa Technology Co., Ltd. */ +#include #include #include #include #include #include #include -#include -#include +#include +#include #define GPIO_RTC_RESERVED 0x0C #define GPIO_RTC_DATA_SET 0x10 @@ -55,7 +56,9 @@ struct moxart_rtc { struct rtc_device *rtc; spinlock_t rtc_lock; - int gpio_data, gpio_sclk, gpio_reset; + struct gpio_desc *gpio_data; + struct gpio_desc *gpio_sclk; + struct gpio_desc *gpio_reset; }; static int day_of_year[12] = { 0, 31, 59, 90, 120, 151, 181, @@ -67,10 +70,10 @@ static void moxart_rtc_write_byte(struct device *dev, u8 data) int i; for (i = 0; i < 8; i++, data >>= 1) { - gpio_set_value(moxart_rtc->gpio_sclk, 0); - gpio_set_value(moxart_rtc->gpio_data, ((data & 1) == 1)); + gpiod_set_value(moxart_rtc->gpio_sclk, 0); + gpiod_set_value(moxart_rtc->gpio_data, ((data & 1) == 1)); udelay(GPIO_RTC_DELAY_TIME); - gpio_set_value(moxart_rtc->gpio_sclk, 1); + gpiod_set_value(moxart_rtc->gpio_sclk, 1); udelay(GPIO_RTC_DELAY_TIME); } } @@ -82,11 +85,11 @@ static u8 moxart_rtc_read_byte(struct device *dev) u8 data = 0; for (i = 0; i < 8; i++) { - gpio_set_value(moxart_rtc->gpio_sclk, 0); + gpiod_set_value(moxart_rtc->gpio_sclk, 0); udelay(GPIO_RTC_DELAY_TIME); - gpio_set_value(moxart_rtc->gpio_sclk, 1); + gpiod_set_value(moxart_rtc->gpio_sclk, 1); udelay(GPIO_RTC_DELAY_TIME); - if (gpio_get_value(moxart_rtc->gpio_data)) + if (gpiod_get_value(moxart_rtc->gpio_data)) data |= (1 << i); udelay(GPIO_RTC_DELAY_TIME); } @@ -101,15 +104,15 @@ static u8 moxart_rtc_read_register(struct device *dev, u8 cmd) local_irq_save(flags); - gpio_direction_output(moxart_rtc->gpio_data, 0); - gpio_set_value(moxart_rtc->gpio_reset, 1); + gpiod_direction_output(moxart_rtc->gpio_data, 0); + gpiod_set_value(moxart_rtc->gpio_reset, 1); udelay(GPIO_RTC_DELAY_TIME); moxart_rtc_write_byte(dev, cmd); - gpio_direction_input(moxart_rtc->gpio_data); + gpiod_direction_input(moxart_rtc->gpio_data); udelay(GPIO_RTC_DELAY_TIME); data = moxart_rtc_read_byte(dev); - gpio_set_value(moxart_rtc->gpio_sclk, 0); - gpio_set_value(moxart_rtc->gpio_reset, 0); + gpiod_set_value(moxart_rtc->gpio_sclk, 0); + gpiod_set_value(moxart_rtc->gpio_reset, 0); udelay(GPIO_RTC_DELAY_TIME); local_irq_restore(flags); @@ -124,13 +127,13 @@ static void moxart_rtc_write_register(struct device *dev, u8 cmd, u8 data) local_irq_save(flags); - gpio_direction_output(moxart_rtc->gpio_data, 0); - gpio_set_value(moxart_rtc->gpio_reset, 1); + gpiod_direction_output(moxart_rtc->gpio_data, 0); + gpiod_set_value(moxart_rtc->gpio_reset, 1); udelay(GPIO_RTC_DELAY_TIME); moxart_rtc_write_byte(dev, cmd); moxart_rtc_write_byte(dev, data); - gpio_set_value(moxart_rtc->gpio_sclk, 0); - gpio_set_value(moxart_rtc->gpio_reset, 0); + gpiod_set_value(moxart_rtc->gpio_sclk, 0); + gpiod_set_value(moxart_rtc->gpio_reset, 0); udelay(GPIO_RTC_DELAY_TIME); local_irq_restore(flags); @@ -247,53 +250,33 @@ static int moxart_rtc_probe(struct platform_device *pdev) if (!moxart_rtc) return -ENOMEM; - moxart_rtc->gpio_data = of_get_named_gpio(pdev->dev.of_node, - "gpio-rtc-data", 0); - if (!gpio_is_valid(moxart_rtc->gpio_data)) { - dev_err(&pdev->dev, "invalid gpio (data): %d\n", - moxart_rtc->gpio_data); - return moxart_rtc->gpio_data; - } - - moxart_rtc->gpio_sclk = of_get_named_gpio(pdev->dev.of_node, - "gpio-rtc-sclk", 0); - if (!gpio_is_valid(moxart_rtc->gpio_sclk)) { - dev_err(&pdev->dev, "invalid gpio (sclk): %d\n", - moxart_rtc->gpio_sclk); - return moxart_rtc->gpio_sclk; - } - - moxart_rtc->gpio_reset = of_get_named_gpio(pdev->dev.of_node, - "gpio-rtc-reset", 0); - if (!gpio_is_valid(moxart_rtc->gpio_reset)) { - dev_err(&pdev->dev, "invalid gpio (reset): %d\n", - moxart_rtc->gpio_reset); - return moxart_rtc->gpio_reset; - } - - spin_lock_init(&moxart_rtc->rtc_lock); - platform_set_drvdata(pdev, moxart_rtc); - - ret = devm_gpio_request(&pdev->dev, moxart_rtc->gpio_data, "rtc_data"); + moxart_rtc->gpio_data = devm_gpiod_get(&pdev->dev, "rtc-data", + GPIOD_IN); + ret = PTR_ERR_OR_ZERO(moxart_rtc->gpio_data); if (ret) { - dev_err(&pdev->dev, "can't get rtc_data gpio\n"); + dev_err(&pdev->dev, "can't get rtc data gpio: %d\n", ret); return ret; } - ret = devm_gpio_request_one(&pdev->dev, moxart_rtc->gpio_sclk, - GPIOF_DIR_OUT, "rtc_sclk"); + moxart_rtc->gpio_sclk = devm_gpiod_get(&pdev->dev, "rtc-sclk", + GPIOD_ASIS); + ret = PTR_ERR_OR_ZERO(moxart_rtc->gpio_sclk); if (ret) { - dev_err(&pdev->dev, "can't get rtc_sclk gpio\n"); + dev_err(&pdev->dev, "can't get rtc sclk gpio: %d\n", ret); return ret; } - ret = devm_gpio_request_one(&pdev->dev, moxart_rtc->gpio_reset, - GPIOF_DIR_OUT, "rtc_reset"); + moxart_rtc->gpio_reset = devm_gpiod_get(&pdev->dev, "rtc-reset", + GPIOD_ASIS); + ret = PTR_ERR_OR_ZERO(moxart_rtc->gpio_reset); if (ret) { - dev_err(&pdev->dev, "can't get rtc_reset gpio\n"); + dev_err(&pdev->dev, "can't get rtc reset gpio: %d\n", ret); return ret; } + spin_lock_init(&moxart_rtc->rtc_lock); + platform_set_drvdata(pdev, moxart_rtc); + moxart_rtc->rtc = devm_rtc_device_register(&pdev->dev, pdev->name, &moxart_rtc_ops, THIS_MODULE); From e8c9efd5d52f1da1c9e012466b9df94a38cc1c00 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Tue, 31 Jan 2023 21:48:14 -0800 Subject: [PATCH 178/765] dt-bindings: rtc: moxart: use proper names for gpio properties MOXA ART RTC driver has been switched to gpiod API and is now using properly named properties for its gpios (with gpiolib implementing a quirk to recognize legacy names). Change binding document to use proper names as well. Signed-off-by: Dmitry Torokhov Link: https://lore.kernel.org/r/20230201054815.4112632-2-dmitry.torokhov@gmail.com Signed-off-by: Alexandre Belloni --- .../devicetree/bindings/rtc/moxa,moxart-rtc.txt | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/Documentation/devicetree/bindings/rtc/moxa,moxart-rtc.txt b/Documentation/devicetree/bindings/rtc/moxa,moxart-rtc.txt index c9d3ac1477feb..1374df7bf9d69 100644 --- a/Documentation/devicetree/bindings/rtc/moxa,moxart-rtc.txt +++ b/Documentation/devicetree/bindings/rtc/moxa,moxart-rtc.txt @@ -3,15 +3,15 @@ MOXA ART real-time clock Required properties: - compatible : Should be "moxa,moxart-rtc" -- gpio-rtc-sclk : RTC sclk gpio, with zero flags -- gpio-rtc-data : RTC data gpio, with zero flags -- gpio-rtc-reset : RTC reset gpio, with zero flags +- rtc-sclk-gpios : RTC sclk gpio, with zero flags +- rtc-data-gpios : RTC data gpio, with zero flags +- rtc-reset-gpios : RTC reset gpio, with zero flags Example: rtc: rtc { compatible = "moxa,moxart-rtc"; - gpio-rtc-sclk = <&gpio 5 0>; - gpio-rtc-data = <&gpio 6 0>; - gpio-rtc-reset = <&gpio 7 0>; + rtc-sclk-gpios = <&gpio 5 0>; + rtc-data-gpios = <&gpio 6 0>; + rtc-reset-gpios = <&gpio 7 0>; }; From 4737a703528c769c4fde6b68462f656f91f4ad99 Mon Sep 17 00:00:00 2001 From: Paul Cercueil Date: Sun, 29 Jan 2023 12:04:39 +0000 Subject: [PATCH 179/765] dt-bindings: rtc: Add #clock-cells property The RTC in the JZ4770 is compatible with the JZ4760, but has an extra register that permits to configure the behaviour of the CLK32K pin. The same goes for the RTC in the JZ4780. With this change, the RTC node is now also a clock provider on these SoCs, so a #clock-cells property is added. Signed-off-by: Paul Cercueil Reviewed-by: Rob Herring Link: https://lore.kernel.org/r/20230129120442.22858-2-paul@crapouillou.net Signed-off-by: Alexandre Belloni --- .../devicetree/bindings/rtc/ingenic,rtc.yaml | 29 +++++++++++++++++++ 1 file changed, 29 insertions(+) diff --git a/Documentation/devicetree/bindings/rtc/ingenic,rtc.yaml b/Documentation/devicetree/bindings/rtc/ingenic,rtc.yaml index af78b67b3da4d..de9879bdb3175 100644 --- a/Documentation/devicetree/bindings/rtc/ingenic,rtc.yaml +++ b/Documentation/devicetree/bindings/rtc/ingenic,rtc.yaml @@ -11,6 +11,17 @@ maintainers: allOf: - $ref: rtc.yaml# + - if: + not: + properties: + compatible: + contains: + enum: + - ingenic,jz4770-rtc + - ingenic,jz4780-rtc + then: + properties: + "#clock-cells": false properties: compatible: @@ -39,6 +50,9 @@ properties: clock-names: const: rtc + "#clock-cells": + const: 0 + system-power-controller: description: | Indicates that the RTC is responsible for powering OFF @@ -83,3 +97,18 @@ examples: clocks = <&cgu JZ4740_CLK_RTC>; clock-names = "rtc"; }; + + - | + #include + rtc: rtc@10003000 { + compatible = "ingenic,jz4780-rtc", "ingenic,jz4760-rtc"; + reg = <0x10003000 0x4c>; + + interrupt-parent = <&intc>; + interrupts = <32>; + + clocks = <&cgu JZ4780_CLK_RTCLK>; + clock-names = "rtc"; + + #clock-cells = <0>; + }; From d644b133f78d6d8efd36f7b1703bebca09036f0b Mon Sep 17 00:00:00 2001 From: Paul Cercueil Date: Sun, 29 Jan 2023 12:04:40 +0000 Subject: [PATCH 180/765] rtc: jz4740: Use readl_poll_timeout Use readl_poll_timeout() from instead of using custom poll loops. The timeout settings are different, but that shouldn't be much of a problem. Instead of polling 10000 times in a close loop, it polls for one millisecond. Signed-off-by: Paul Cercueil Link: https://lore.kernel.org/r/20230129120442.22858-3-paul@crapouillou.net Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-jz4740.c | 18 ++++++------------ 1 file changed, 6 insertions(+), 12 deletions(-) diff --git a/drivers/rtc/rtc-jz4740.c b/drivers/rtc/rtc-jz4740.c index c383719292c7d..4ee6e5ee09b17 100644 --- a/drivers/rtc/rtc-jz4740.c +++ b/drivers/rtc/rtc-jz4740.c @@ -7,6 +7,7 @@ #include #include +#include #include #include #include @@ -69,19 +70,15 @@ static inline uint32_t jz4740_rtc_reg_read(struct jz4740_rtc *rtc, size_t reg) static int jz4740_rtc_wait_write_ready(struct jz4740_rtc *rtc) { uint32_t ctrl; - int timeout = 10000; - do { - ctrl = jz4740_rtc_reg_read(rtc, JZ_REG_RTC_CTRL); - } while (!(ctrl & JZ_RTC_CTRL_WRDY) && --timeout); - - return timeout ? 0 : -EIO; + return readl_poll_timeout(rtc->base + JZ_REG_RTC_CTRL, ctrl, + ctrl & JZ_RTC_CTRL_WRDY, 0, 1000); } static inline int jz4780_rtc_enable_write(struct jz4740_rtc *rtc) { uint32_t ctrl; - int ret, timeout = 10000; + int ret; ret = jz4740_rtc_wait_write_ready(rtc); if (ret != 0) @@ -89,11 +86,8 @@ static inline int jz4780_rtc_enable_write(struct jz4740_rtc *rtc) writel(JZ_RTC_WENR_MAGIC, rtc->base + JZ_REG_RTC_WENR); - do { - ctrl = readl(rtc->base + JZ_REG_RTC_WENR); - } while (!(ctrl & JZ_RTC_WENR_WEN) && --timeout); - - return timeout ? 0 : -EIO; + return readl_poll_timeout(rtc->base + JZ_REG_RTC_WENR, ctrl, + ctrl & JZ_RTC_WENR_WEN, 0, 1000); } static inline int jz4740_rtc_reg_write(struct jz4740_rtc *rtc, size_t reg, From ff6fd3770e9687d7b849a0e826a32563bfcb98da Mon Sep 17 00:00:00 2001 From: Paul Cercueil Date: Sun, 29 Jan 2023 12:04:41 +0000 Subject: [PATCH 181/765] rtc: jz4740: Use dev_err_probe() Use dev_err_probe() where it makes sense to simplify a bit the code. Signed-off-by: Paul Cercueil Link: https://lore.kernel.org/r/20230129120442.22858-4-paul@crapouillou.net Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-jz4740.c | 20 +++++++------------- 1 file changed, 7 insertions(+), 13 deletions(-) diff --git a/drivers/rtc/rtc-jz4740.c b/drivers/rtc/rtc-jz4740.c index 4ee6e5ee09b17..9ffa764aa71eb 100644 --- a/drivers/rtc/rtc-jz4740.c +++ b/drivers/rtc/rtc-jz4740.c @@ -329,17 +329,13 @@ static int jz4740_rtc_probe(struct platform_device *pdev) device_init_wakeup(dev, 1); ret = dev_pm_set_wake_irq(dev, irq); - if (ret) { - dev_err(dev, "Failed to set wake irq: %d\n", ret); - return ret; - } + if (ret) + return dev_err_probe(dev, ret, "Failed to set wake irq\n"); rtc->rtc = devm_rtc_allocate_device(dev); - if (IS_ERR(rtc->rtc)) { - ret = PTR_ERR(rtc->rtc); - dev_err(dev, "Failed to allocate rtc device: %d\n", ret); - return ret; - } + if (IS_ERR(rtc->rtc)) + return dev_err_probe(dev, PTR_ERR(rtc->rtc), + "Failed to allocate rtc device\n"); rtc->rtc->ops = &jz4740_rtc_ops; rtc->rtc->range_max = U32_MAX; @@ -356,10 +352,8 @@ static int jz4740_rtc_probe(struct platform_device *pdev) ret = devm_request_irq(dev, irq, jz4740_rtc_irq, 0, pdev->name, rtc); - if (ret) { - dev_err(dev, "Failed to request rtc irq: %d\n", ret); - return ret; - } + if (ret) + return dev_err_probe(dev, ret, "Failed to request rtc irq\n"); if (of_device_is_system_power_controller(np)) { dev_for_power_off = dev; From 5ddfa148de8cf5491fd1c89522c7cad859db8c88 Mon Sep 17 00:00:00 2001 From: Paul Cercueil Date: Sun, 29 Jan 2023 12:04:42 +0000 Subject: [PATCH 182/765] rtc: jz4740: Register clock provider for the CLK32K pin On JZ4770 and JZ4780, the CLK32K pin is configurable. By default, it is configured as a GPIO in input mode, and its value can be read through GPIO PD14. With this change, clients can now request the 32 kHz clock on the CLK32K pin, through Device Tree. This clock is simply a pass-through of the input oscillator's clock with enable/disable operations. This will permit the WiFi/Bluetooth chip to work on the MIPS CI20 board, which does source one of its clocks from the CLK32K pin. Signed-off-by: Paul Cercueil Link: https://lore.kernel.org/r/20230129120442.22858-5-paul@crapouillou.net Signed-off-by: Alexandre Belloni --- drivers/rtc/Kconfig | 2 +- drivers/rtc/rtc-jz4740.c | 56 ++++++++++++++++++++++++++++++++++++++++ 2 files changed, 57 insertions(+), 1 deletion(-) diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig index 677d2601d3057..d2b6d20a67454 100644 --- a/drivers/rtc/Kconfig +++ b/drivers/rtc/Kconfig @@ -1690,7 +1690,7 @@ config RTC_DRV_MPC5121 config RTC_DRV_JZ4740 tristate "Ingenic JZ4740 SoC" depends on MIPS || COMPILE_TEST - depends on OF + depends on OF && COMMON_CLK help If you say yes here you get support for the Ingenic JZ47xx SoCs RTC controllers. diff --git a/drivers/rtc/rtc-jz4740.c b/drivers/rtc/rtc-jz4740.c index 9ffa764aa71eb..59d279e3e6f5b 100644 --- a/drivers/rtc/rtc-jz4740.c +++ b/drivers/rtc/rtc-jz4740.c @@ -6,6 +6,7 @@ */ #include +#include #include #include #include @@ -13,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -26,6 +28,7 @@ #define JZ_REG_RTC_WAKEUP_FILTER 0x24 #define JZ_REG_RTC_RESET_COUNTER 0x28 #define JZ_REG_RTC_SCRATCHPAD 0x34 +#define JZ_REG_RTC_CKPCR 0x40 /* The following are present on the jz4780 */ #define JZ_REG_RTC_WENR 0x3C @@ -45,6 +48,9 @@ #define JZ_RTC_WAKEUP_FILTER_MASK 0x0000FFE0 #define JZ_RTC_RESET_COUNTER_MASK 0x00000FE0 +#define JZ_RTC_CKPCR_CK32PULL_DIS BIT(4) +#define JZ_RTC_CKPCR_CK32CTL_EN (BIT(2) | BIT(1)) + enum jz4740_rtc_type { ID_JZ4740, ID_JZ4760, @@ -57,6 +63,8 @@ struct jz4740_rtc { struct rtc_device *rtc; + struct clk_hw clk32k; + spinlock_t lock; }; @@ -254,6 +262,7 @@ static void jz4740_rtc_power_off(void) static const struct of_device_id jz4740_rtc_of_match[] = { { .compatible = "ingenic,jz4740-rtc", .data = (void *)ID_JZ4740 }, { .compatible = "ingenic,jz4760-rtc", .data = (void *)ID_JZ4760 }, + { .compatible = "ingenic,jz4770-rtc", .data = (void *)ID_JZ4780 }, { .compatible = "ingenic,jz4780-rtc", .data = (void *)ID_JZ4780 }, {}, }; @@ -295,6 +304,38 @@ static void jz4740_rtc_set_wakeup_params(struct jz4740_rtc *rtc, jz4740_rtc_reg_write(rtc, JZ_REG_RTC_RESET_COUNTER, reset_ticks); } +static int jz4740_rtc_clk32k_enable(struct clk_hw *hw) +{ + struct jz4740_rtc *rtc = container_of(hw, struct jz4740_rtc, clk32k); + + return jz4740_rtc_reg_write(rtc, JZ_REG_RTC_CKPCR, + JZ_RTC_CKPCR_CK32PULL_DIS | + JZ_RTC_CKPCR_CK32CTL_EN); +} + +static void jz4740_rtc_clk32k_disable(struct clk_hw *hw) +{ + struct jz4740_rtc *rtc = container_of(hw, struct jz4740_rtc, clk32k); + + jz4740_rtc_reg_write(rtc, JZ_REG_RTC_CKPCR, 0); +} + +static int jz4740_rtc_clk32k_is_enabled(struct clk_hw *hw) +{ + struct jz4740_rtc *rtc = container_of(hw, struct jz4740_rtc, clk32k); + u32 ckpcr; + + ckpcr = jz4740_rtc_reg_read(rtc, JZ_REG_RTC_CKPCR); + + return !!(ckpcr & JZ_RTC_CKPCR_CK32CTL_EN); +} + +static const struct clk_ops jz4740_rtc_clk32k_ops = { + .enable = jz4740_rtc_clk32k_enable, + .disable = jz4740_rtc_clk32k_disable, + .is_enabled = jz4740_rtc_clk32k_is_enabled, +}; + static int jz4740_rtc_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; @@ -364,6 +405,21 @@ static int jz4740_rtc_probe(struct platform_device *pdev) dev_warn(dev, "Poweroff handler already present!\n"); } + if (device_property_present(dev, "#clock-cells")) { + rtc->clk32k.init = CLK_HW_INIT_HW("clk32k", __clk_get_hw(clk), + &jz4740_rtc_clk32k_ops, 0); + + ret = devm_clk_hw_register(dev, &rtc->clk32k); + if (ret) + return dev_err_probe(dev, ret, + "Unable to register clk32k clock\n"); + + ret = of_clk_add_hw_provider(np, of_clk_hw_simple_get, &rtc->clk32k); + if (ret) + return dev_err_probe(dev, ret, + "Unable to register clk32k clock provider\n"); + } + return 0; } From 1dd0510f6d4b85616a36aabb9be38389467122d9 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Sat, 11 Feb 2023 04:07:06 +1100 Subject: [PATCH 183/765] xfs: fix low space alloc deadlock I've recently encountered an ABBA deadlock with g/476. The upcoming changes seem to make this much easier to hit, but the underlying problem is a pre-existing one. Essentially, if we select an AG for allocation, then lock the AGF and then fail to allocate for some reason (e.g. minimum length requirements cannot be satisfied), then we drop out of the allocation with the AGF still locked. The caller then modifies the allocation constraints - usually loosening them up - and tries again. This can result in trying to access AGFs that are lower than the AGF we already have locked from the failed attempt. e.g. the failed attempt skipped several AGs before failing, so we have locks an AG higher than the start AG. Retrying the allocation from the start AG then causes us to violate AGF lock ordering and this can lead to deadlocks. The deadlock exists even if allocation succeeds - we can do a followup allocations in the same transaction for BMBT blocks that aren't guaranteed to be in the same AG as the original, and can move into higher AGs. Hence we really need to move the tp->t_firstblock tracking down into xfs_alloc_vextent() where it can be set when we exit with a locked AG. xfs_alloc_vextent() can also check there if the requested allocation falls within the allow range of AGs set by tp->t_firstblock. If we can't allocate within the range set, we have to fail the allocation. If we are allowed to to non-blocking AGF locking, we can ignore the AG locking order limitations as we can use try-locks for the first iteration over requested AG range. This invalidates a set of post allocation asserts that check that the allocation is always above tp->t_firstblock if it is set. Because we can use try-locks to avoid the deadlock in some circumstances, having a pre-existing locked AGF doesn't always prevent allocation from lower order AGFs. Hence those ASSERTs need to be removed. Signed-off-by: Dave Chinner Reviewed-by: Allison Henderson Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_alloc.c | 69 ++++++++++++++++++++++++++++++++------- fs/xfs/libxfs/xfs_bmap.c | 14 -------- fs/xfs/xfs_trace.h | 1 + 3 files changed, 58 insertions(+), 26 deletions(-) diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index f8ff81c3de761..ffe6345bfafcb 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -3164,10 +3164,13 @@ xfs_alloc_vextent( xfs_alloctype_t type; /* input allocation type */ int bump_rotor = 0; xfs_agnumber_t rotorstep = xfs_rotorstep; /* inode32 agf stepper */ + xfs_agnumber_t minimum_agno = 0; mp = args->mp; type = args->otype = args->type; args->agbno = NULLAGBLOCK; + if (args->tp->t_firstblock != NULLFSBLOCK) + minimum_agno = XFS_FSB_TO_AGNO(mp, args->tp->t_firstblock); /* * Just fix this up, for the case where the last a.g. is shorter * (or there's only one a.g.) and the caller couldn't easily figure @@ -3201,6 +3204,13 @@ xfs_alloc_vextent( */ args->agno = XFS_FSB_TO_AGNO(mp, args->fsbno); args->pag = xfs_perag_get(mp, args->agno); + + if (minimum_agno > args->agno) { + trace_xfs_alloc_vextent_skip_deadlock(args); + error = 0; + break; + } + error = xfs_alloc_fix_freelist(args, 0); if (error) { trace_xfs_alloc_vextent_nofix(args); @@ -3232,6 +3242,8 @@ xfs_alloc_vextent( case XFS_ALLOCTYPE_FIRST_AG: /* * Rotate through the allocation groups looking for a winner. + * If we are blocking, we must obey minimum_agno contraints for + * avoiding ABBA deadlocks on AGF locking. */ if (type == XFS_ALLOCTYPE_FIRST_AG) { /* @@ -3239,7 +3251,7 @@ xfs_alloc_vextent( */ args->agno = XFS_FSB_TO_AGNO(mp, args->fsbno); args->type = XFS_ALLOCTYPE_THIS_AG; - sagno = 0; + sagno = minimum_agno; flags = 0; } else { /* @@ -3248,6 +3260,7 @@ xfs_alloc_vextent( args->agno = sagno = XFS_FSB_TO_AGNO(mp, args->fsbno); flags = XFS_ALLOC_FLAG_TRYLOCK; } + /* * Loop over allocation groups twice; first time with * trylock set, second time without. @@ -3276,19 +3289,21 @@ xfs_alloc_vextent( if (args->agno == sagno && type == XFS_ALLOCTYPE_START_BNO) args->type = XFS_ALLOCTYPE_THIS_AG; + /* - * For the first allocation, we can try any AG to get - * space. However, if we already have allocated a - * block, we don't want to try AGs whose number is below - * sagno. Otherwise, we may end up with out-of-order - * locking of AGF, which might cause deadlock. - */ + * If we are try-locking, we can't deadlock on AGF + * locks, so we can wrap all the way back to the first + * AG. Otherwise, wrap back to the start AG so we can't + * deadlock, and let the end of scan handler decide what + * to do next. + */ if (++(args->agno) == mp->m_sb.sb_agcount) { - if (args->tp->t_firstblock != NULLFSBLOCK) - args->agno = sagno; - else + if (flags & XFS_ALLOC_FLAG_TRYLOCK) args->agno = 0; + else + args->agno = sagno; } + /* * Reached the starting a.g., must either be done * or switch to non-trylock mode. @@ -3300,7 +3315,14 @@ xfs_alloc_vextent( break; } + /* + * Blocking pass next, so we must obey minimum + * agno constraints to avoid ABBA AGF deadlocks. + */ flags = 0; + if (minimum_agno > sagno) + sagno = minimum_agno; + if (type == XFS_ALLOCTYPE_START_BNO) { args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); @@ -3322,9 +3344,9 @@ xfs_alloc_vextent( ASSERT(0); /* NOTREACHED */ } - if (args->agbno == NULLAGBLOCK) + if (args->agbno == NULLAGBLOCK) { args->fsbno = NULLFSBLOCK; - else { + } else { args->fsbno = XFS_AGB_TO_FSB(mp, args->agno, args->agbno); #ifdef DEBUG ASSERT(args->len >= args->minlen); @@ -3335,6 +3357,29 @@ xfs_alloc_vextent( #endif } + + /* + * We end up here with a locked AGF. If we failed, the caller is likely + * going to try to allocate again with different parameters, and that + * can widen the AGs that are searched for free space. If we have to do + * BMBT block allocation, we have to do a new allocation. + * + * Hence leaving this function with the AGF locked opens up potential + * ABBA AGF deadlocks because a future allocation attempt in this + * transaction may attempt to lock a lower number AGF. + * + * We can't release the AGF until the transaction is commited, so at + * this point we must update the "firstblock" tracker to point at this + * AG if the tracker is empty or points to a lower AG. This allows the + * next allocation attempt to be modified appropriately to avoid + * deadlocks. + */ + if (args->agbp && + (args->tp->t_firstblock == NULLFSBLOCK || + args->pag->pag_agno > minimum_agno)) { + args->tp->t_firstblock = XFS_AGB_TO_FSB(mp, + args->pag->pag_agno, 0); + } xfs_perag_put(args->pag); return 0; error0: diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index c8c65387136c9..de6d585c00f1e 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -3413,21 +3413,7 @@ xfs_bmap_process_allocated_extent( xfs_fileoff_t orig_offset, xfs_extlen_t orig_length) { - int nullfb; - - nullfb = ap->tp->t_firstblock == NULLFSBLOCK; - - /* - * check the allocation happened at the same or higher AG than - * the first block that was allocated. - */ - ASSERT(nullfb || - XFS_FSB_TO_AGNO(args->mp, ap->tp->t_firstblock) <= - XFS_FSB_TO_AGNO(args->mp, args->fsbno)); - ap->blkno = args->fsbno; - if (nullfb) - ap->tp->t_firstblock = args->fsbno; ap->length = args->len; /* * If the extent size hint is active, we tried to round the diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 6b0e9ae7c513c..adddaeb44a563 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -1877,6 +1877,7 @@ DEFINE_ALLOC_EVENT(xfs_alloc_small_notenough); DEFINE_ALLOC_EVENT(xfs_alloc_small_done); DEFINE_ALLOC_EVENT(xfs_alloc_small_error); DEFINE_ALLOC_EVENT(xfs_alloc_vextent_badargs); +DEFINE_ALLOC_EVENT(xfs_alloc_vextent_skip_deadlock); DEFINE_ALLOC_EVENT(xfs_alloc_vextent_nofix); DEFINE_ALLOC_EVENT(xfs_alloc_vextent_noagbp); DEFINE_ALLOC_EVENT(xfs_alloc_vextent_loopfailed); From f08f984c63e9980614ae3a0a574b31eaaef284b2 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Sat, 11 Feb 2023 04:08:06 +1100 Subject: [PATCH 184/765] xfs: prefer free inodes at ENOSPC over chunk allocation When an XFS filesystem has free inodes in chunks already allocated on disk, it will still allocate new inode chunks if the target AG has no free inodes in it. Normally, this is a good idea as it preserves locality of all the inodes in a given directory. However, at ENOSPC this can lead to using the last few remaining free filesystem blocks to allocate a new chunk when there are many, many free inodes that could be allocated without consuming free space. This results in speeding up the consumption of the last few blocks and inode create operations then returning ENOSPC when there free inodes available because we don't have enough block left in the filesystem for directory creation reservations to proceed. Hence when we are near ENOSPC, we should be attempting to preserve the remaining blocks for directory block allocation rather than using them for unnecessary inode chunk creation. This particular behaviour is exposed by xfs/294, when it drives to ENOSPC on empty file creation whilst there are still thousands of free inodes available for allocation in other AGs in the filesystem. Hence, when we are within 1% of ENOSPC, change the inode allocation behaviour to prefer to use existing free inodes over allocating new inode chunks, even though it results is poorer locality of the data set. It is more important for the allocations to be space efficient near ENOSPC than to have optimal locality for performance, so lets modify the inode AG selection code to reflect that fact. This allows generic/294 to not only pass with this allocator rework patchset, but to increase the number of post-ENOSPC empty inode allocations to from ~600 to ~9080 before we hit ENOSPC on the directory create transaction reservation. Signed-off-by: Dave Chinner Reviewed-by: Allison Henderson Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_ialloc.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index 5118dedf9267b..e8068422aa21f 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -1737,6 +1737,7 @@ xfs_dialloc( struct xfs_perag *pag; struct xfs_ino_geometry *igeo = M_IGEO(mp); bool ok_alloc = true; + bool low_space = false; int flags; xfs_ino_t ino; @@ -1767,6 +1768,20 @@ xfs_dialloc( ok_alloc = false; } + /* + * If we are near to ENOSPC, we want to prefer allocation from AGs that + * have free inodes in them rather than use up free space allocating new + * inode chunks. Hence we turn off allocation for the first non-blocking + * pass through the AGs if we are near ENOSPC to consume free inodes + * that we can immediately allocate, but then we allow allocation on the + * second pass if we fail to find an AG with free inodes in it. + */ + if (percpu_counter_read_positive(&mp->m_fdblocks) < + mp->m_low_space[XFS_LOWSP_1_PCNT]) { + ok_alloc = false; + low_space = true; + } + /* * Loop until we find an allocation group that either has free inodes * or in which we can allocate some inodes. Iterate through the @@ -1795,6 +1810,8 @@ xfs_dialloc( break; } flags = 0; + if (low_space) + ok_alloc = true; } xfs_perag_put(pag); } From d5753847b216db0e553e8065aa825cfe497ad143 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Sat, 11 Feb 2023 04:09:06 +1100 Subject: [PATCH 185/765] xfs: block reservation too large for minleft allocation When we enter xfs_bmbt_alloc_block() without having first allocated a data extent (i.e. tp->t_firstblock == NULLFSBLOCK) because we are doing something like unwritten extent conversion, the transaction block reservation is used as the minleft value. This works for operations like unwritten extent conversion, but it assumes that the block reservation is only for a BMBT split. THis is not always true, and sometimes results in larger than necessary minleft values being set. We only actually need enough space for a btree split, something we already handle correctly in xfs_bmapi_write() via the xfs_bmapi_minleft() calculation. We should use xfs_bmapi_minleft() in xfs_bmbt_alloc_block() to calculate the number of blocks a BMBT split on this inode is going to require, not use the transaction block reservation that contains the maximum number of blocks this transaction may consume in it... Signed-off-by: Dave Chinner Reviewed-by: Allison Henderson Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_bmap.c | 2 +- fs/xfs/libxfs/xfs_bmap.h | 2 ++ fs/xfs/libxfs/xfs_bmap_btree.c | 19 +++++++++---------- 3 files changed, 12 insertions(+), 11 deletions(-) diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index de6d585c00f1e..0930c441159dc 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4242,7 +4242,7 @@ xfs_bmapi_convert_unwritten( return 0; } -static inline xfs_extlen_t +xfs_extlen_t xfs_bmapi_minleft( struct xfs_trans *tp, struct xfs_inode *ip, diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h index 01c2df35c3e35..524912f276f87 100644 --- a/fs/xfs/libxfs/xfs_bmap.h +++ b/fs/xfs/libxfs/xfs_bmap.h @@ -220,6 +220,8 @@ int xfs_bmap_add_extent_unwritten_real(struct xfs_trans *tp, struct xfs_inode *ip, int whichfork, struct xfs_iext_cursor *icur, struct xfs_btree_cur **curp, struct xfs_bmbt_irec *new, int *logflagsp); +xfs_extlen_t xfs_bmapi_minleft(struct xfs_trans *tp, struct xfs_inode *ip, + int fork); enum xfs_bmap_intent_type { XFS_BMAP_MAP = 1, diff --git a/fs/xfs/libxfs/xfs_bmap_btree.c b/fs/xfs/libxfs/xfs_bmap_btree.c index cfa052d40105d..18de4fbfef4e9 100644 --- a/fs/xfs/libxfs/xfs_bmap_btree.c +++ b/fs/xfs/libxfs/xfs_bmap_btree.c @@ -213,18 +213,16 @@ xfs_bmbt_alloc_block( if (args.fsbno == NULLFSBLOCK) { args.fsbno = be64_to_cpu(start->l); args.type = XFS_ALLOCTYPE_START_BNO; + /* - * Make sure there is sufficient room left in the AG to - * complete a full tree split for an extent insert. If - * we are converting the middle part of an extent then - * we may need space for two tree splits. - * - * We are relying on the caller to make the correct block - * reservation for this operation to succeed. If the - * reservation amount is insufficient then we may fail a - * block allocation here and corrupt the filesystem. + * If we are coming here from something like unwritten extent + * conversion, there has been no data extent allocation already + * done, so we have to ensure that we attempt to locate the + * entire set of bmbt allocations in the same AG, as + * xfs_bmapi_write() would have reserved. */ - args.minleft = args.tp->t_blk_res; + args.minleft = xfs_bmapi_minleft(cur->bc_tp, cur->bc_ino.ip, + cur->bc_ino.whichfork); } else if (cur->bc_tp->t_flags & XFS_TRANS_LOWMODE) { args.type = XFS_ALLOCTYPE_START_BNO; } else { @@ -248,6 +246,7 @@ xfs_bmbt_alloc_block( * successful activate the lowspace algorithm. */ args.fsbno = 0; + args.minleft = 0; args.type = XFS_ALLOCTYPE_FIRST_AG; error = xfs_alloc_vextent(&args); if (error) From 36b6ad2d9cb81b0d52ae1598286ca5809cd39003 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Sat, 11 Feb 2023 04:10:06 +1100 Subject: [PATCH 186/765] xfs: drop firstblock constraints from allocation setup Now that xfs_alloc_vextent() does all the AGF deadlock prevention filtering for multiple allocations in a single transaction, we no longer need the allocation setup code to care about what AGs we might already have locked. Hence we can remove all the "nullfb" conditional logic in places like xfs_bmap_btalloc() and instead have them focus simply on setting up locality constraints. If the allocation fails due to AGF lock filtering in xfs_alloc_vextent, then we just fall back as we normally do to more relaxed allocation constraints. As a result, any allocation that allows AG scanning (i.e. not confined to a single AG) and does not force a worst case full filesystem scan will now be able to attempt allocation from AGs lower than that defined by tp->t_firstblock. This is because xfs_alloc_vextent() allows try-locking of the AGFs and hence enables low space algorithms to at least -try- to get space from AGs lower than the one that we have currently locked and allocated from. This is a significant improvement in the low space allocation algorithm. Signed-off-by: Dave Chinner Reviewed-by: Allison Henderson Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_bmap.c | 168 +++++++++++---------------------- fs/xfs/libxfs/xfs_bmap_btree.c | 30 +++--- 2 files changed, 66 insertions(+), 132 deletions(-) diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 0930c441159dc..31a0738d017e0 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -645,16 +645,9 @@ xfs_bmap_extents_to_btree( args.tp = tp; args.mp = mp; xfs_rmap_ino_bmbt_owner(&args.oinfo, ip->i_ino, whichfork); - if (tp->t_firstblock == NULLFSBLOCK) { - args.type = XFS_ALLOCTYPE_START_BNO; - args.fsbno = XFS_INO_TO_FSB(mp, ip->i_ino); - } else if (tp->t_flags & XFS_TRANS_LOWMODE) { - args.type = XFS_ALLOCTYPE_START_BNO; - args.fsbno = tp->t_firstblock; - } else { - args.type = XFS_ALLOCTYPE_NEAR_BNO; - args.fsbno = tp->t_firstblock; - } + + args.type = XFS_ALLOCTYPE_START_BNO; + args.fsbno = XFS_INO_TO_FSB(mp, ip->i_ino); args.minlen = args.maxlen = args.prod = 1; args.wasdel = wasdel; *logflagsp = 0; @@ -662,17 +655,14 @@ xfs_bmap_extents_to_btree( if (error) goto out_root_realloc; + /* + * Allocation can't fail, the space was reserved. + */ if (WARN_ON_ONCE(args.fsbno == NULLFSBLOCK)) { error = -ENOSPC; goto out_root_realloc; } - /* - * Allocation can't fail, the space was reserved. - */ - ASSERT(tp->t_firstblock == NULLFSBLOCK || - args.agno >= XFS_FSB_TO_AGNO(mp, tp->t_firstblock)); - tp->t_firstblock = args.fsbno; cur->bc_ino.allocated++; ip->i_nblocks++; xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_BCOUNT, 1L); @@ -804,13 +794,8 @@ xfs_bmap_local_to_extents( * Allocate a block. We know we need only one, since the * file currently fits in an inode. */ - if (tp->t_firstblock == NULLFSBLOCK) { - args.fsbno = XFS_INO_TO_FSB(args.mp, ip->i_ino); - args.type = XFS_ALLOCTYPE_START_BNO; - } else { - args.fsbno = tp->t_firstblock; - args.type = XFS_ALLOCTYPE_NEAR_BNO; - } + args.fsbno = XFS_INO_TO_FSB(args.mp, ip->i_ino); + args.type = XFS_ALLOCTYPE_START_BNO; args.total = total; args.minlen = args.maxlen = args.prod = 1; error = xfs_alloc_vextent(&args); @@ -820,7 +805,6 @@ xfs_bmap_local_to_extents( /* Can't fail, the space was reserved. */ ASSERT(args.fsbno != NULLFSBLOCK); ASSERT(args.len == 1); - tp->t_firstblock = args.fsbno; error = xfs_trans_get_buf(tp, args.mp->m_ddev_targp, XFS_FSB_TO_DADDR(args.mp, args.fsbno), args.mp->m_bsize, 0, &bp); @@ -854,8 +838,7 @@ xfs_bmap_local_to_extents( ifp->if_nextents = 1; ip->i_nblocks = 1; - xfs_trans_mod_dquot_byino(tp, ip, - XFS_TRANS_DQ_BCOUNT, 1L); + xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_BCOUNT, 1L); flags |= xfs_ilog_fext(whichfork); done: @@ -3025,9 +3008,7 @@ xfs_bmap_adjacent( struct xfs_bmalloca *ap) /* bmap alloc argument struct */ { xfs_fsblock_t adjust; /* adjustment to block numbers */ - xfs_agnumber_t fb_agno; /* ag number of ap->firstblock */ xfs_mount_t *mp; /* mount point structure */ - int nullfb; /* true if ap->firstblock isn't set */ int rt; /* true if inode is realtime */ #define ISVALID(x,y) \ @@ -3038,11 +3019,8 @@ xfs_bmap_adjacent( XFS_FSB_TO_AGBNO(mp, x) < mp->m_sb.sb_agblocks) mp = ap->ip->i_mount; - nullfb = ap->tp->t_firstblock == NULLFSBLOCK; rt = XFS_IS_REALTIME_INODE(ap->ip) && (ap->datatype & XFS_ALLOC_USERDATA); - fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp, - ap->tp->t_firstblock); /* * If allocating at eof, and there's a previous real block, * try to use its last block as our starting point. @@ -3101,13 +3079,6 @@ xfs_bmap_adjacent( prevbno += adjust; else prevdiff += adjust; - /* - * If the firstblock forbids it, can't use it, - * must use default. - */ - if (!rt && !nullfb && - XFS_FSB_TO_AGNO(mp, prevbno) != fb_agno) - prevbno = NULLFSBLOCK; } /* * No previous block or can't follow it, just default. @@ -3143,13 +3114,6 @@ xfs_bmap_adjacent( gotdiff += adjust - ap->length; } else gotdiff += adjust; - /* - * If the firstblock forbids it, can't use it, - * must use default. - */ - if (!rt && !nullfb && - XFS_FSB_TO_AGNO(mp, gotbno) != fb_agno) - gotbno = NULLFSBLOCK; } /* * No next block, just default. @@ -3236,7 +3200,7 @@ xfs_bmap_select_minlen( } STATIC int -xfs_bmap_btalloc_nullfb( +xfs_bmap_btalloc_select_lengths( struct xfs_bmalloca *ap, struct xfs_alloc_arg *args, xfs_extlen_t *blen) @@ -3247,8 +3211,13 @@ xfs_bmap_btalloc_nullfb( int error; args->type = XFS_ALLOCTYPE_START_BNO; - args->total = ap->total; + if (ap->tp->t_flags & XFS_TRANS_LOWMODE) { + args->total = ap->minlen; + args->minlen = ap->minlen; + return 0; + } + args->total = ap->total; startag = ag = XFS_FSB_TO_AGNO(mp, args->fsbno); if (startag == NULLAGNUMBER) startag = ag = 0; @@ -3280,6 +3249,13 @@ xfs_bmap_btalloc_filestreams( int notinit = 0; int error; + if (ap->tp->t_flags & XFS_TRANS_LOWMODE) { + args->type = XFS_ALLOCTYPE_FIRST_AG; + args->total = ap->minlen; + args->minlen = ap->minlen; + return 0; + } + args->type = XFS_ALLOCTYPE_NEAR_BNO; args->total = ap->total; @@ -3460,19 +3436,15 @@ xfs_bmap_exact_minlen_extent_alloc( xfs_bmap_compute_alignments(ap, &args); - if (ap->tp->t_firstblock == NULLFSBLOCK) { - /* - * Unlike the longest extent available in an AG, we don't track - * the length of an AG's shortest extent. - * XFS_ERRTAG_BMAP_ALLOC_MINLEN_EXTENT is a debug only knob and - * hence we can afford to start traversing from the 0th AG since - * we need not be concerned about a drop in performance in - * "debug only" code paths. - */ - ap->blkno = XFS_AGB_TO_FSB(mp, 0, 0); - } else { - ap->blkno = ap->tp->t_firstblock; - } + /* + * Unlike the longest extent available in an AG, we don't track + * the length of an AG's shortest extent. + * XFS_ERRTAG_BMAP_ALLOC_MINLEN_EXTENT is a debug only knob and + * hence we can afford to start traversing from the 0th AG since + * we need not be concerned about a drop in performance in + * "debug only" code paths. + */ + ap->blkno = XFS_AGB_TO_FSB(mp, 0, 0); args.fsbno = ap->blkno; args.oinfo = XFS_RMAP_OINFO_SKIP_UPDATE; @@ -3515,13 +3487,11 @@ xfs_bmap_btalloc( struct xfs_mount *mp = ap->ip->i_mount; struct xfs_alloc_arg args = { .tp = ap->tp, .mp = mp }; xfs_alloctype_t atype = 0; - xfs_agnumber_t fb_agno; /* ag number of ap->firstblock */ xfs_agnumber_t ag; xfs_fileoff_t orig_offset; xfs_extlen_t orig_length; xfs_extlen_t blen; xfs_extlen_t nextminlen = 0; - int nullfb; /* true if ap->firstblock isn't set */ int isaligned; int tryagain; int error; @@ -3533,34 +3503,17 @@ xfs_bmap_btalloc( stripe_align = xfs_bmap_compute_alignments(ap, &args); - nullfb = ap->tp->t_firstblock == NULLFSBLOCK; - fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp, - ap->tp->t_firstblock); - if (nullfb) { - if ((ap->datatype & XFS_ALLOC_USERDATA) && - xfs_inode_is_filestream(ap->ip)) { - ag = xfs_filestream_lookup_ag(ap->ip); - ag = (ag != NULLAGNUMBER) ? ag : 0; - ap->blkno = XFS_AGB_TO_FSB(mp, ag, 0); - } else { - ap->blkno = XFS_INO_TO_FSB(mp, ap->ip->i_ino); - } - } else - ap->blkno = ap->tp->t_firstblock; + if ((ap->datatype & XFS_ALLOC_USERDATA) && + xfs_inode_is_filestream(ap->ip)) { + ag = xfs_filestream_lookup_ag(ap->ip); + ag = (ag != NULLAGNUMBER) ? ag : 0; + ap->blkno = XFS_AGB_TO_FSB(mp, ag, 0); + } else { + ap->blkno = XFS_INO_TO_FSB(mp, ap->ip->i_ino); + } xfs_bmap_adjacent(ap); - /* - * If allowed, use ap->blkno; otherwise must use firstblock since - * it's in the right allocation group. - */ - if (nullfb || XFS_FSB_TO_AGNO(mp, ap->blkno) == fb_agno) - ; - else - ap->blkno = ap->tp->t_firstblock; - /* - * Normal allocation, done through xfs_alloc_vextent. - */ tryagain = isaligned = 0; args.fsbno = ap->blkno; args.oinfo = XFS_RMAP_OINFO_SKIP_UPDATE; @@ -3568,30 +3521,19 @@ xfs_bmap_btalloc( /* Trim the allocation back to the maximum an AG can fit. */ args.maxlen = min(ap->length, mp->m_ag_max_usable); blen = 0; - if (nullfb) { - /* - * Search for an allocation group with a single extent large - * enough for the request. If one isn't found, then adjust - * the minimum allocation size to the largest space found. - */ - if ((ap->datatype & XFS_ALLOC_USERDATA) && - xfs_inode_is_filestream(ap->ip)) - error = xfs_bmap_btalloc_filestreams(ap, &args, &blen); - else - error = xfs_bmap_btalloc_nullfb(ap, &args, &blen); - if (error) - return error; - } else if (ap->tp->t_flags & XFS_TRANS_LOWMODE) { - if (xfs_inode_is_filestream(ap->ip)) - args.type = XFS_ALLOCTYPE_FIRST_AG; - else - args.type = XFS_ALLOCTYPE_START_BNO; - args.total = args.minlen = ap->minlen; - } else { - args.type = XFS_ALLOCTYPE_NEAR_BNO; - args.total = ap->total; - args.minlen = ap->minlen; - } + + /* + * Search for an allocation group with a single extent large + * enough for the request. If one isn't found, then adjust + * the minimum allocation size to the largest space found. + */ + if ((ap->datatype & XFS_ALLOC_USERDATA) && + xfs_inode_is_filestream(ap->ip)) + error = xfs_bmap_btalloc_filestreams(ap, &args, &blen); + else + error = xfs_bmap_btalloc_select_lengths(ap, &args, &blen); + if (error) + return error; /* * If we are not low on available data blocks, and the underlying @@ -3678,7 +3620,7 @@ xfs_bmap_btalloc( if ((error = xfs_alloc_vextent(&args))) return error; } - if (args.fsbno == NULLFSBLOCK && nullfb && + if (args.fsbno == NULLFSBLOCK && args.minlen > ap->minlen) { args.minlen = ap->minlen; args.type = XFS_ALLOCTYPE_START_BNO; @@ -3686,7 +3628,7 @@ xfs_bmap_btalloc( if ((error = xfs_alloc_vextent(&args))) return error; } - if (args.fsbno == NULLFSBLOCK && nullfb) { + if (args.fsbno == NULLFSBLOCK) { args.fsbno = 0; args.type = XFS_ALLOCTYPE_FIRST_AG; args.total = ap->minlen; diff --git a/fs/xfs/libxfs/xfs_bmap_btree.c b/fs/xfs/libxfs/xfs_bmap_btree.c index 18de4fbfef4e9..76a0f0d260a47 100644 --- a/fs/xfs/libxfs/xfs_bmap_btree.c +++ b/fs/xfs/libxfs/xfs_bmap_btree.c @@ -206,28 +206,21 @@ xfs_bmbt_alloc_block( memset(&args, 0, sizeof(args)); args.tp = cur->bc_tp; args.mp = cur->bc_mp; - args.fsbno = cur->bc_tp->t_firstblock; xfs_rmap_ino_bmbt_owner(&args.oinfo, cur->bc_ino.ip->i_ino, cur->bc_ino.whichfork); - if (args.fsbno == NULLFSBLOCK) { - args.fsbno = be64_to_cpu(start->l); - args.type = XFS_ALLOCTYPE_START_BNO; + args.fsbno = be64_to_cpu(start->l); + args.type = XFS_ALLOCTYPE_START_BNO; - /* - * If we are coming here from something like unwritten extent - * conversion, there has been no data extent allocation already - * done, so we have to ensure that we attempt to locate the - * entire set of bmbt allocations in the same AG, as - * xfs_bmapi_write() would have reserved. - */ + /* + * If we are coming here from something like unwritten extent + * conversion, there has been no data extent allocation already done, so + * we have to ensure that we attempt to locate the entire set of bmbt + * allocations in the same AG, as xfs_bmapi_write() would have reserved. + */ + if (cur->bc_tp->t_firstblock == NULLFSBLOCK) args.minleft = xfs_bmapi_minleft(cur->bc_tp, cur->bc_ino.ip, - cur->bc_ino.whichfork); - } else if (cur->bc_tp->t_flags & XFS_TRANS_LOWMODE) { - args.type = XFS_ALLOCTYPE_START_BNO; - } else { - args.type = XFS_ALLOCTYPE_NEAR_BNO; - } + cur->bc_ino.whichfork); args.minlen = args.maxlen = args.prod = 1; args.wasdel = cur->bc_ino.flags & XFS_BTCUR_BMBT_WASDEL; @@ -247,7 +240,7 @@ xfs_bmbt_alloc_block( */ args.fsbno = 0; args.minleft = 0; - args.type = XFS_ALLOCTYPE_FIRST_AG; + args.type = XFS_ALLOCTYPE_START_BNO; error = xfs_alloc_vextent(&args); if (error) goto error0; @@ -259,7 +252,6 @@ xfs_bmbt_alloc_block( } ASSERT(args.len == 1); - cur->bc_tp->t_firstblock = args.fsbno; cur->bc_ino.allocated++; cur->bc_ino.ip->i_nblocks++; xfs_trans_log_inode(args.tp, cur->bc_ino.ip, XFS_ILOG_CORE); From 692b6cddeb65a5170c1e63d25b1ffb7822e80f7d Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Sat, 11 Feb 2023 04:11:06 +1100 Subject: [PATCH 187/765] xfs: t_firstblock is tracking AGs not blocks The tp->t_firstblock field is now raelly tracking the highest AG we have locked, not the block number of the highest allocation we've made. It's purpose is to prevent AGF locking deadlocks, so rename it to "highest AG" and simplify the implementation to just track the agno rather than a fsbno. Signed-off-by: Dave Chinner Reviewed-by: Allison Henderson Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_alloc.c | 12 +++++------- fs/xfs/libxfs/xfs_bmap.c | 4 ++-- fs/xfs/libxfs/xfs_bmap_btree.c | 6 +++--- fs/xfs/libxfs/xfs_btree.c | 2 +- fs/xfs/xfs_bmap_util.c | 2 +- fs/xfs/xfs_inode.c | 2 +- fs/xfs/xfs_reflink.c | 2 +- fs/xfs/xfs_trace.h | 8 ++++---- fs/xfs/xfs_trans.c | 4 ++-- fs/xfs/xfs_trans.h | 2 +- 10 files changed, 21 insertions(+), 23 deletions(-) diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index ffe6345bfafcb..974b23abca60e 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -3169,8 +3169,8 @@ xfs_alloc_vextent( mp = args->mp; type = args->otype = args->type; args->agbno = NULLAGBLOCK; - if (args->tp->t_firstblock != NULLFSBLOCK) - minimum_agno = XFS_FSB_TO_AGNO(mp, args->tp->t_firstblock); + if (args->tp->t_highest_agno != NULLAGNUMBER) + minimum_agno = args->tp->t_highest_agno; /* * Just fix this up, for the case where the last a.g. is shorter * (or there's only one a.g.) and the caller couldn't easily figure @@ -3375,11 +3375,9 @@ xfs_alloc_vextent( * deadlocks. */ if (args->agbp && - (args->tp->t_firstblock == NULLFSBLOCK || - args->pag->pag_agno > minimum_agno)) { - args->tp->t_firstblock = XFS_AGB_TO_FSB(mp, - args->pag->pag_agno, 0); - } + (args->tp->t_highest_agno == NULLAGNUMBER || + args->pag->pag_agno > minimum_agno)) + args->tp->t_highest_agno = args->pag->pag_agno; xfs_perag_put(args->pag); return 0; error0: diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 31a0738d017e0..504fd69fe2c60 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4192,7 +4192,7 @@ xfs_bmapi_minleft( { struct xfs_ifork *ifp = xfs_ifork_ptr(ip, fork); - if (tp && tp->t_firstblock != NULLFSBLOCK) + if (tp && tp->t_highest_agno != NULLAGNUMBER) return 0; if (ifp->if_format != XFS_DINODE_FMT_BTREE) return 1; @@ -6079,7 +6079,7 @@ xfs_bmap_finish_one( struct xfs_bmbt_irec *bmap = &bi->bi_bmap; int error = 0; - ASSERT(tp->t_firstblock == NULLFSBLOCK); + ASSERT(tp->t_highest_agno == NULLAGNUMBER); trace_xfs_bmap_deferred(tp->t_mountp, XFS_FSB_TO_AGNO(tp->t_mountp, bmap->br_startblock), diff --git a/fs/xfs/libxfs/xfs_bmap_btree.c b/fs/xfs/libxfs/xfs_bmap_btree.c index 76a0f0d260a47..afd9b2d962a31 100644 --- a/fs/xfs/libxfs/xfs_bmap_btree.c +++ b/fs/xfs/libxfs/xfs_bmap_btree.c @@ -184,11 +184,11 @@ xfs_bmbt_update_cursor( struct xfs_btree_cur *src, struct xfs_btree_cur *dst) { - ASSERT((dst->bc_tp->t_firstblock != NULLFSBLOCK) || + ASSERT((dst->bc_tp->t_highest_agno != NULLAGNUMBER) || (dst->bc_ino.ip->i_diflags & XFS_DIFLAG_REALTIME)); dst->bc_ino.allocated += src->bc_ino.allocated; - dst->bc_tp->t_firstblock = src->bc_tp->t_firstblock; + dst->bc_tp->t_highest_agno = src->bc_tp->t_highest_agno; src->bc_ino.allocated = 0; } @@ -218,7 +218,7 @@ xfs_bmbt_alloc_block( * we have to ensure that we attempt to locate the entire set of bmbt * allocations in the same AG, as xfs_bmapi_write() would have reserved. */ - if (cur->bc_tp->t_firstblock == NULLFSBLOCK) + if (cur->bc_tp->t_highest_agno == NULLAGNUMBER) args.minleft = xfs_bmapi_minleft(cur->bc_tp, cur->bc_ino.ip, cur->bc_ino.whichfork); diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c index da8c769887fdd..c4649cc624e1b 100644 --- a/fs/xfs/libxfs/xfs_btree.c +++ b/fs/xfs/libxfs/xfs_btree.c @@ -2943,7 +2943,7 @@ xfs_btree_split( DECLARE_COMPLETION_ONSTACK(done); if (cur->bc_btnum != XFS_BTNUM_BMAP || - cur->bc_tp->t_firstblock == NULLFSBLOCK) + cur->bc_tp->t_highest_agno == NULLAGNUMBER) return __xfs_btree_split(cur, level, ptrp, key, curp, stat); args.cur = cur; diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c index 867645b74d889..a09dd26064792 100644 --- a/fs/xfs/xfs_bmap_util.c +++ b/fs/xfs/xfs_bmap_util.c @@ -1410,7 +1410,7 @@ xfs_swap_extent_rmap( /* Unmap the old blocks in the source file. */ while (tirec.br_blockcount) { - ASSERT(tp->t_firstblock == NULLFSBLOCK); + ASSERT(tp->t_highest_agno == NULLAGNUMBER); trace_xfs_swap_extent_rmap_remap_piece(tip, &tirec); /* Read extent from the source file */ diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index d354ea2b74f96..dbe274b8065dd 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1367,7 +1367,7 @@ xfs_itruncate_extents_flags( unmap_len = XFS_MAX_FILEOFF - first_unmap_block + 1; while (unmap_len > 0) { - ASSERT(tp->t_firstblock == NULLFSBLOCK); + ASSERT(tp->t_highest_agno == NULLAGNUMBER); error = __xfs_bunmapi(tp, ip, first_unmap_block, &unmap_len, flags, XFS_ITRUNC_MAX_EXTENTS); if (error) diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c index 5535778a98f92..57bf59ff48543 100644 --- a/fs/xfs/xfs_reflink.c +++ b/fs/xfs/xfs_reflink.c @@ -610,7 +610,7 @@ xfs_reflink_cancel_cow_blocks( if (error) break; } else if (del.br_state == XFS_EXT_UNWRITTEN || cancel_real) { - ASSERT((*tpp)->t_firstblock == NULLFSBLOCK); + ASSERT((*tpp)->t_highest_agno == NULLAGNUMBER); /* Free the CoW orphan record. */ xfs_refcount_free_cow_extent(*tpp, del.br_startblock, diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index adddaeb44a563..2ac98d8ddbfde 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -1801,7 +1801,7 @@ DECLARE_EVENT_CLASS(xfs_alloc_class, __field(char, wasfromfl) __field(int, resv) __field(int, datatype) - __field(xfs_fsblock_t, firstblock) + __field(xfs_agnumber_t, highest_agno) ), TP_fast_assign( __entry->dev = args->mp->m_super->s_dev; @@ -1822,12 +1822,12 @@ DECLARE_EVENT_CLASS(xfs_alloc_class, __entry->wasfromfl = args->wasfromfl; __entry->resv = args->resv; __entry->datatype = args->datatype; - __entry->firstblock = args->tp->t_firstblock; + __entry->highest_agno = args->tp->t_highest_agno; ), TP_printk("dev %d:%d agno 0x%x agbno 0x%x minlen %u maxlen %u mod %u " "prod %u minleft %u total %u alignment %u minalignslop %u " "len %u type %s otype %s wasdel %d wasfromfl %d resv %d " - "datatype 0x%x firstblock 0x%llx", + "datatype 0x%x highest_agno 0x%x", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->agno, __entry->agbno, @@ -1846,7 +1846,7 @@ DECLARE_EVENT_CLASS(xfs_alloc_class, __entry->wasfromfl, __entry->resv, __entry->datatype, - (unsigned long long)__entry->firstblock) + __entry->highest_agno) ) #define DEFINE_ALLOC_EVENT(name) \ diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index 7bd16fbff5341..53ab544e4c2cb 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -102,7 +102,7 @@ xfs_trans_dup( INIT_LIST_HEAD(&ntp->t_items); INIT_LIST_HEAD(&ntp->t_busy); INIT_LIST_HEAD(&ntp->t_dfops); - ntp->t_firstblock = NULLFSBLOCK; + ntp->t_highest_agno = NULLAGNUMBER; ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES); ASSERT(tp->t_ticket != NULL); @@ -278,7 +278,7 @@ xfs_trans_alloc( INIT_LIST_HEAD(&tp->t_items); INIT_LIST_HEAD(&tp->t_busy); INIT_LIST_HEAD(&tp->t_dfops); - tp->t_firstblock = NULLFSBLOCK; + tp->t_highest_agno = NULLAGNUMBER; error = xfs_trans_reserve(tp, resp, blocks, rtextents); if (error == -ENOSPC && want_retry) { diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h index 55819785941cc..6e3646d524ceb 100644 --- a/fs/xfs/xfs_trans.h +++ b/fs/xfs/xfs_trans.h @@ -132,7 +132,7 @@ typedef struct xfs_trans { unsigned int t_rtx_res; /* # of rt extents resvd */ unsigned int t_rtx_res_used; /* # of resvd rt extents used */ unsigned int t_flags; /* misc flags */ - xfs_fsblock_t t_firstblock; /* first block allocated */ + xfs_agnumber_t t_highest_agno; /* highest AGF locked */ struct xlog_ticket *t_ticket; /* log mgr ticket */ struct xfs_mount *t_mountp; /* ptr to fs mount struct */ struct xfs_dquot_acct *t_dqinfo; /* acctg info for dquots */ From 55d5c3a386d74d3f374023c8fa386f524a9192e8 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Sat, 11 Feb 2023 04:12:06 +1100 Subject: [PATCH 188/765] xfs: don't assert fail on transaction cancel with deferred ops We can error out of an allocation transaction when updating BMBT blocks when things go wrong. This can be a btree corruption, and unexpected ENOSPC, etc. In these cases, we already have deferred ops queued for the first allocation that has been done, and we just want to cancel out the transaction and shut down the filesystem on error. In fact, we do just that for production systems - the assert that we can't have a transaction with defer ops attached unless we are already shut down is bogus and gets in the way of debugging whatever issue is actually causing the transaction to be cancelled. Remove the assert because it is causing spurious test failures to hang test machines. Signed-off-by: Dave Chinner Reviewed-by: Allison Henderson Reviewed-by: Darrick J. Wong --- fs/xfs/xfs_trans.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index 53ab544e4c2cb..8afc0c080861d 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -1078,10 +1078,10 @@ xfs_trans_cancel( /* * It's never valid to cancel a transaction with deferred ops attached, * because the transaction is effectively dirty. Complain about this - * loudly before freeing the in-memory defer items. + * loudly before freeing the in-memory defer items and shutting down the + * filesystem. */ if (!list_empty(&tp->t_dfops)) { - ASSERT(xfs_is_shutdown(mp) || list_empty(&tp->t_dfops)); ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES); dirty = true; xfs_defer_cancel(tp); From 905a77077573056d7af508f35373f66ed8b4a39e Mon Sep 17 00:00:00 2001 From: David Gow Date: Sat, 17 Dec 2022 12:44:34 +0800 Subject: [PATCH 189/765] rust: arch/um: Use 'pie' relocation mode under UML UML expects a position independent executable for some reason, so tell rustc to generate pie objects. Otherwise we get a bunch of relocations we can't deal with in libcore. Signed-off-by: David Gow Signed-off-by: Richard Weinberger --- arch/um/Makefile | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/um/Makefile b/arch/um/Makefile index f1d4d67157be0..ae321282dc6f6 100644 --- a/arch/um/Makefile +++ b/arch/um/Makefile @@ -68,6 +68,8 @@ KBUILD_CFLAGS += $(CFLAGS) $(CFLAGS-y) -D__arch_um__ \ -Din6addr_loopback=kernel_in6addr_loopback \ -Din6addr_any=kernel_in6addr_any -Dstrrchr=kernel_strrchr +KBUILD_RUSTFLAGS += -Crelocation-model=pie + KBUILD_AFLAGS += $(ARCH_INCLUDE) USER_CFLAGS = $(patsubst $(KERNEL_DEFINES),,$(patsubst -I%,,$(KBUILD_CFLAGS))) \ From 8849818679478933dd1d9718741f4daa3f4e8b86 Mon Sep 17 00:00:00 2001 From: David Gow Date: Sat, 17 Dec 2022 12:44:35 +0800 Subject: [PATCH 190/765] rust: arch/um: Disable FP/SIMD instruction to match x86 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The kernel disables all SSE and similar FP/SIMD instructions on x86-based architectures (partly because we shouldn't be using floats in the kernel, and partly to avoid the need for stack alignment, see: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=53383 ) UML does not do the same thing, which isn't in itself a problem, but does add to the list of differences between UML and "normal" x86 builds. In addition, there was a crash bug with LLVM < 15 / rustc < 1.65 when building with SSE, so disabling it fixes rust builds with earlier compiler versions, see: https://github.com/Rust-for-Linux/linux/pull/881 Signed-off-by: David Gow Reviewed-by: Sergio González Collado Signed-off-by: Richard Weinberger --- arch/x86/Makefile.um | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/arch/x86/Makefile.um b/arch/x86/Makefile.um index b3c1ae084180d..d2e95d1d4db77 100644 --- a/arch/x86/Makefile.um +++ b/arch/x86/Makefile.um @@ -1,6 +1,12 @@ # SPDX-License-Identifier: GPL-2.0 core-y += arch/x86/crypto/ +# +# Disable SSE and other FP/SIMD instructions to match normal x86 +# +KBUILD_CFLAGS += -mno-sse -mno-mmx -mno-sse2 -mno-3dnow -mno-avx +KBUILD_RUSTFLAGS += -Ctarget-feature=-sse,-sse2,-sse3,-ssse3,-sse4.1,-sse4.2,-avx,-avx2 + ifeq ($(CONFIG_X86_32),y) START := 0x8048000 From 0438aadfa69a345136f5ba4f582e0f769450ee0d Mon Sep 17 00:00:00 2001 From: David Gow Date: Sat, 17 Dec 2022 12:44:36 +0800 Subject: [PATCH 191/765] rust: arch/um: Add support for CONFIG_RUST under x86_64 UML MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit CONFIG_RUST currently supports x86_64, but does not support it under UML. With the previous patches applied, adding support is trivial: add CONFIG_HAVE_RUST to UML if X86_64 is set. The scripts/generate_rust_target.rs file already checks for CONFIG_X86_64, not CONFIG_X86, so is prepared for UML support. The Rust support does not currently support X86_32. Also, update the Rust architecture support documentation to not that this is being maintained: I intend to look after this as best I can. Signed-off-by: David Gow Reviewed-by: Sergio González Collado Tested-by: Sergio González Collado Signed-off-by: Richard Weinberger --- Documentation/rust/arch-support.rst | 2 ++ arch/um/Kconfig | 1 + 2 files changed, 3 insertions(+) diff --git a/Documentation/rust/arch-support.rst b/Documentation/rust/arch-support.rst index 6982b63775da5..a526ca1c688be 100644 --- a/Documentation/rust/arch-support.rst +++ b/Documentation/rust/arch-support.rst @@ -17,3 +17,5 @@ Architecture Level of support Constraints ============ ================ ============================================== ``x86`` Maintained ``x86_64`` only. ============ ================ ============================================== +``um`` Maintained ``x86_64`` only. +============ ================ ============================================== diff --git a/arch/um/Kconfig b/arch/um/Kconfig index ad4ff3b0e91e5..4db186f019ae4 100644 --- a/arch/um/Kconfig +++ b/arch/um/Kconfig @@ -28,6 +28,7 @@ config UML select TRACE_IRQFLAGS_SUPPORT select TTY # Needed for line.c select HAVE_ARCH_VMAP_STACK + select HAVE_RUST if X86_64 config MMU bool From 5541992e512de8c9133110809f767bd1b54ee10d Mon Sep 17 00:00:00 2001 From: Ammar Faizi Date: Sat, 24 Dec 2022 00:23:38 +0700 Subject: [PATCH 192/765] x86: um: vdso: Add '%rcx' and '%r11' to the syscall clobber list The 'syscall' instruction clobbers '%rcx' and '%r11', but they are not listed in the inline Assembly that performs the syscall instruction. No real bug is found. It wasn't buggy by luck because '%rcx' and '%r11' are caller-saved registers, and not used in the functions, and the functions are never inlined. Add them to the clobber list for code correctness. Fixes: f1c2bb8b9964ed31de988910f8b1cfb586d30091 ("um: implement a x86_64 vDSO") Signed-off-by: Ammar Faizi Signed-off-by: Richard Weinberger --- arch/x86/um/vdso/um_vdso.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/arch/x86/um/vdso/um_vdso.c b/arch/x86/um/vdso/um_vdso.c index 2112b8d146688..ff0f3b4b6c45e 100644 --- a/arch/x86/um/vdso/um_vdso.c +++ b/arch/x86/um/vdso/um_vdso.c @@ -17,8 +17,10 @@ int __vdso_clock_gettime(clockid_t clock, struct __kernel_old_timespec *ts) { long ret; - asm("syscall" : "=a" (ret) : - "0" (__NR_clock_gettime), "D" (clock), "S" (ts) : "memory"); + asm("syscall" + : "=a" (ret) + : "0" (__NR_clock_gettime), "D" (clock), "S" (ts) + : "rcx", "r11", "memory"); return ret; } @@ -29,8 +31,10 @@ int __vdso_gettimeofday(struct __kernel_old_timeval *tv, struct timezone *tz) { long ret; - asm("syscall" : "=a" (ret) : - "0" (__NR_gettimeofday), "D" (tv), "S" (tz) : "memory"); + asm("syscall" + : "=a" (ret) + : "0" (__NR_gettimeofday), "D" (tv), "S" (tz) + : "rcx", "r11", "memory"); return ret; } From 2f2be5102480b1058182fa6c4b1e5c1732d6760c Mon Sep 17 00:00:00 2001 From: Peter Foley Date: Thu, 12 Jan 2023 23:49:07 -0500 Subject: [PATCH 193/765] um: Make the definition of cpu_data more compatible MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Match the x86 implementation to improve build errors. Noticed when building allyesconfig. e.g. ../arch/um/include/asm/processor-generic.h:94:19: error: called object is not a function or function pointer 94 | #define cpu_data (&boot_cpu_data) | ~^~~~~~~~~~~~~~~ ../drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_topology.c:2157:16: note: in expansion of macro ‘cpu_data’ 2157 | return cpu_data(first_cpu_of_numa_node).apicid; Signed-off-by: Peter Foley Signed-off-by: Richard Weinberger --- arch/um/include/asm/processor-generic.h | 2 +- arch/um/kernel/um_arch.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/um/include/asm/processor-generic.h b/arch/um/include/asm/processor-generic.h index bb5f06480da95..7414154b8e9ae 100644 --- a/arch/um/include/asm/processor-generic.h +++ b/arch/um/include/asm/processor-generic.h @@ -91,7 +91,7 @@ struct cpuinfo_um { extern struct cpuinfo_um boot_cpu_data; -#define cpu_data (&boot_cpu_data) +#define cpu_data(cpu) boot_cpu_data #define current_cpu_data boot_cpu_data #define cache_line_size() (boot_cpu_data.cache_alignment) diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c index 786b44dc20c98..8dcda617b8bf6 100644 --- a/arch/um/kernel/um_arch.c +++ b/arch/um/kernel/um_arch.c @@ -96,7 +96,7 @@ static int show_cpuinfo(struct seq_file *m, void *v) static void *c_start(struct seq_file *m, loff_t *pos) { - return *pos < nr_cpu_ids ? cpu_data + *pos : NULL; + return *pos < nr_cpu_ids ? &boot_cpu_data + *pos : NULL; } static void *c_next(struct seq_file *m, void *v, loff_t *pos) From ffb1b4a41016295e298409c9dbcacd55680bd6d4 Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Fri, 10 Feb 2023 14:42:01 -0800 Subject: [PATCH 194/765] x86/unwind/orc: Add 'signal' field to ORC metadata Add a 'signal' field which allows unwind hints to specify whether the instruction pointer should be taken literally (like for most interrupts and exceptions) rather than decremented (like for call stack return addresses) when used to find the next ORC entry. Signed-off-by: Josh Poimboeuf Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/d2c5ec4d83a45b513d8fd72fab59f1a8cfa46871.1676068346.git.jpoimboe@kernel.org --- arch/x86/include/asm/orc_types.h | 4 +++- arch/x86/include/asm/unwind_hints.h | 10 +++++----- arch/x86/kernel/unwind_orc.c | 5 ++--- include/linux/objtool.h | 11 +++++++---- tools/arch/x86/include/asm/orc_types.h | 4 +++- tools/include/linux/objtool.h | 11 +++++++---- tools/objtool/orc_dump.c | 4 ++-- 7 files changed, 29 insertions(+), 20 deletions(-) diff --git a/arch/x86/include/asm/orc_types.h b/arch/x86/include/asm/orc_types.h index 5a2baf28a1dcd..1343a62106de9 100644 --- a/arch/x86/include/asm/orc_types.h +++ b/arch/x86/include/asm/orc_types.h @@ -57,12 +57,14 @@ struct orc_entry { unsigned sp_reg:4; unsigned bp_reg:4; unsigned type:2; + unsigned signal:1; unsigned end:1; #elif defined(__BIG_ENDIAN_BITFIELD) unsigned bp_reg:4; unsigned sp_reg:4; - unsigned unused:5; + unsigned unused:4; unsigned end:1; + unsigned signal:1; unsigned type:2; #endif } __packed; diff --git a/arch/x86/include/asm/unwind_hints.h b/arch/x86/include/asm/unwind_hints.h index f66fbe6537dd7..e7c71750b3093 100644 --- a/arch/x86/include/asm/unwind_hints.h +++ b/arch/x86/include/asm/unwind_hints.h @@ -15,7 +15,7 @@ UNWIND_HINT type=UNWIND_HINT_TYPE_ENTRY end=1 .endm -.macro UNWIND_HINT_REGS base=%rsp offset=0 indirect=0 extra=1 partial=0 +.macro UNWIND_HINT_REGS base=%rsp offset=0 indirect=0 extra=1 partial=0 signal=1 .if \base == %rsp .if \indirect .set sp_reg, ORC_REG_SP_INDIRECT @@ -45,11 +45,11 @@ .set type, UNWIND_HINT_TYPE_REGS .endif - UNWIND_HINT sp_reg=sp_reg sp_offset=sp_offset type=type + UNWIND_HINT sp_reg=sp_reg sp_offset=sp_offset type=type signal=\signal .endm -.macro UNWIND_HINT_IRET_REGS base=%rsp offset=0 - UNWIND_HINT_REGS base=\base offset=\offset partial=1 +.macro UNWIND_HINT_IRET_REGS base=%rsp offset=0 signal=1 + UNWIND_HINT_REGS base=\base offset=\offset partial=1 signal=\signal .endm .macro UNWIND_HINT_FUNC @@ -67,7 +67,7 @@ #else #define UNWIND_HINT_FUNC \ - UNWIND_HINT(ORC_REG_SP, 8, UNWIND_HINT_TYPE_FUNC, 0) + UNWIND_HINT(ORC_REG_SP, 8, UNWIND_HINT_TYPE_FUNC, 0, 0) #endif /* __ASSEMBLY__ */ diff --git a/arch/x86/kernel/unwind_orc.c b/arch/x86/kernel/unwind_orc.c index cdf6c60601700..37307b40f8daf 100644 --- a/arch/x86/kernel/unwind_orc.c +++ b/arch/x86/kernel/unwind_orc.c @@ -484,6 +484,8 @@ bool unwind_next_frame(struct unwind_state *state) goto the_end; } + state->signal = orc->signal; + /* Find the previous frame's stack: */ switch (orc->sp_reg) { case ORC_REG_SP: @@ -563,7 +565,6 @@ bool unwind_next_frame(struct unwind_state *state) state->sp = sp; state->regs = NULL; state->prev_regs = NULL; - state->signal = false; break; case UNWIND_HINT_TYPE_REGS: @@ -587,7 +588,6 @@ bool unwind_next_frame(struct unwind_state *state) state->regs = (struct pt_regs *)sp; state->prev_regs = NULL; state->full_regs = true; - state->signal = true; break; case UNWIND_HINT_TYPE_REGS_PARTIAL: @@ -604,7 +604,6 @@ bool unwind_next_frame(struct unwind_state *state) state->prev_regs = state->regs; state->regs = (void *)sp - IRET_FRAME_OFFSET; state->full_regs = false; - state->signal = true; break; default: diff --git a/include/linux/objtool.h b/include/linux/objtool.h index 62c54ffbeeaac..9ac3df3fccf01 100644 --- a/include/linux/objtool.h +++ b/include/linux/objtool.h @@ -15,6 +15,7 @@ struct unwind_hint { s16 sp_offset; u8 sp_reg; u8 type; + u8 signal; u8 end; }; #endif @@ -49,7 +50,7 @@ struct unwind_hint { #ifndef __ASSEMBLY__ -#define UNWIND_HINT(sp_reg, sp_offset, type, end) \ +#define UNWIND_HINT(sp_reg, sp_offset, type, signal, end) \ "987: \n\t" \ ".pushsection .discard.unwind_hints\n\t" \ /* struct unwind_hint */ \ @@ -57,6 +58,7 @@ struct unwind_hint { ".short " __stringify(sp_offset) "\n\t" \ ".byte " __stringify(sp_reg) "\n\t" \ ".byte " __stringify(type) "\n\t" \ + ".byte " __stringify(signal) "\n\t" \ ".byte " __stringify(end) "\n\t" \ ".balign 4 \n\t" \ ".popsection\n\t" @@ -129,7 +131,7 @@ struct unwind_hint { * the debuginfo as necessary. It will also warn if it sees any * inconsistencies. */ -.macro UNWIND_HINT type:req sp_reg=0 sp_offset=0 end=0 +.macro UNWIND_HINT type:req sp_reg=0 sp_offset=0 signal=0 end=0 .Lunwind_hint_ip_\@: .pushsection .discard.unwind_hints /* struct unwind_hint */ @@ -137,6 +139,7 @@ struct unwind_hint { .short \sp_offset .byte \sp_reg .byte \type + .byte \signal .byte \end .balign 4 .popsection @@ -174,7 +177,7 @@ struct unwind_hint { #ifndef __ASSEMBLY__ -#define UNWIND_HINT(sp_reg, sp_offset, type, end) \ +#define UNWIND_HINT(sp_reg, sp_offset, type, signal, end) \ "\n\t" #define STACK_FRAME_NON_STANDARD(func) #define STACK_FRAME_NON_STANDARD_FP(func) @@ -182,7 +185,7 @@ struct unwind_hint { #define ASM_REACHABLE #else #define ANNOTATE_INTRA_FUNCTION_CALL -.macro UNWIND_HINT type:req sp_reg=0 sp_offset=0 end=0 +.macro UNWIND_HINT type:req sp_reg=0 sp_offset=0 signal=0 end=0 .endm .macro STACK_FRAME_NON_STANDARD func:req .endm diff --git a/tools/arch/x86/include/asm/orc_types.h b/tools/arch/x86/include/asm/orc_types.h index 5a2baf28a1dcd..1343a62106de9 100644 --- a/tools/arch/x86/include/asm/orc_types.h +++ b/tools/arch/x86/include/asm/orc_types.h @@ -57,12 +57,14 @@ struct orc_entry { unsigned sp_reg:4; unsigned bp_reg:4; unsigned type:2; + unsigned signal:1; unsigned end:1; #elif defined(__BIG_ENDIAN_BITFIELD) unsigned bp_reg:4; unsigned sp_reg:4; - unsigned unused:5; + unsigned unused:4; unsigned end:1; + unsigned signal:1; unsigned type:2; #endif } __packed; diff --git a/tools/include/linux/objtool.h b/tools/include/linux/objtool.h index 62c54ffbeeaac..9ac3df3fccf01 100644 --- a/tools/include/linux/objtool.h +++ b/tools/include/linux/objtool.h @@ -15,6 +15,7 @@ struct unwind_hint { s16 sp_offset; u8 sp_reg; u8 type; + u8 signal; u8 end; }; #endif @@ -49,7 +50,7 @@ struct unwind_hint { #ifndef __ASSEMBLY__ -#define UNWIND_HINT(sp_reg, sp_offset, type, end) \ +#define UNWIND_HINT(sp_reg, sp_offset, type, signal, end) \ "987: \n\t" \ ".pushsection .discard.unwind_hints\n\t" \ /* struct unwind_hint */ \ @@ -57,6 +58,7 @@ struct unwind_hint { ".short " __stringify(sp_offset) "\n\t" \ ".byte " __stringify(sp_reg) "\n\t" \ ".byte " __stringify(type) "\n\t" \ + ".byte " __stringify(signal) "\n\t" \ ".byte " __stringify(end) "\n\t" \ ".balign 4 \n\t" \ ".popsection\n\t" @@ -129,7 +131,7 @@ struct unwind_hint { * the debuginfo as necessary. It will also warn if it sees any * inconsistencies. */ -.macro UNWIND_HINT type:req sp_reg=0 sp_offset=0 end=0 +.macro UNWIND_HINT type:req sp_reg=0 sp_offset=0 signal=0 end=0 .Lunwind_hint_ip_\@: .pushsection .discard.unwind_hints /* struct unwind_hint */ @@ -137,6 +139,7 @@ struct unwind_hint { .short \sp_offset .byte \sp_reg .byte \type + .byte \signal .byte \end .balign 4 .popsection @@ -174,7 +177,7 @@ struct unwind_hint { #ifndef __ASSEMBLY__ -#define UNWIND_HINT(sp_reg, sp_offset, type, end) \ +#define UNWIND_HINT(sp_reg, sp_offset, type, signal, end) \ "\n\t" #define STACK_FRAME_NON_STANDARD(func) #define STACK_FRAME_NON_STANDARD_FP(func) @@ -182,7 +185,7 @@ struct unwind_hint { #define ASM_REACHABLE #else #define ANNOTATE_INTRA_FUNCTION_CALL -.macro UNWIND_HINT type:req sp_reg=0 sp_offset=0 end=0 +.macro UNWIND_HINT type:req sp_reg=0 sp_offset=0 signal=0 end=0 .endm .macro STACK_FRAME_NON_STANDARD func:req .endm diff --git a/tools/objtool/orc_dump.c b/tools/objtool/orc_dump.c index 4f1211fec82ce..2d8ebdcd1db3c 100644 --- a/tools/objtool/orc_dump.c +++ b/tools/objtool/orc_dump.c @@ -211,8 +211,8 @@ int orc_dump(const char *_objname) print_reg(orc[i].bp_reg, bswap_if_needed(&dummy_elf, orc[i].bp_offset)); - printf(" type:%s end:%d\n", - orc_type_name(orc[i].type), orc[i].end); + printf(" type:%s signal:%d end:%d\n", + orc_type_name(orc[i].type), orc[i].signal, orc[i].end); } elf_end(elf); From 37064583f63eca93c98a9cdf2360485ea05f617a Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Fri, 10 Feb 2023 14:42:02 -0800 Subject: [PATCH 195/765] x86/entry: Fix unwinding from kprobe on PUSH/POP instruction If a kprobe (INT3) is set on a stack-modifying single-byte instruction, like a single-byte PUSH/POP or a LEAVE, ORC fails to unwind past it: Call Trace: dump_stack_lvl+0x57/0x90 handler_pre+0x33/0x40 [kprobe_example] aggr_pre_handler+0x49/0x90 kprobe_int3_handler+0xe3/0x180 do_int3+0x3a/0x80 exc_int3+0x7d/0xc0 asm_exc_int3+0x35/0x40 RIP: 0010:kernel_clone+0xe/0x3a0 Code: cc e8 16 b2 bf 00 66 0f 1f 44 00 00 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 0f 1f 44 00 00 41 57 41 56 41 55 41 54 cc <53> 48 89 fb 48 83 ec 68 4c 8b 27 65 48 8b 04 25 28 00 00 00 48 89 RSP: 0018:ffffc9000074fda0 EFLAGS: 00000206 RAX: 0000000000808100 RBX: ffff888109de9d80 RCX: 0000000000000000 RDX: 0000000000000011 RSI: ffff888109de9d80 RDI: ffffc9000074fdc8 RBP: ffff8881019543c0 R08: ffffffff81127e30 R09: 00000000e71742a5 R10: ffff888104764a18 R11: 0000000071742a5e R12: ffff888100078800 R13: ffff888100126000 R14: 0000000000000000 R15: ffff888100126005 ? __pfx_call_usermodehelper_exec_async+0x10/0x10 ? kernel_clone+0xe/0x3a0 ? user_mode_thread+0x5b/0x80 ? __pfx_call_usermodehelper_exec_async+0x10/0x10 ? call_usermodehelper_exec_work+0x77/0xb0 ? process_one_work+0x299/0x5f0 ? worker_thread+0x4f/0x3a0 ? __pfx_worker_thread+0x10/0x10 ? kthread+0xf2/0x120 ? __pfx_kthread+0x10/0x10 ? ret_from_fork+0x29/0x50 The problem is that #BP saves the pointer to the instruction immediately *after* the INT3, rather than to the INT3 itself. The instruction replaced by the INT3 hasn't actually run, but ORC assumes otherwise and expects the wrong stack layout. Fix it by annotating the #BP exception as a non-signal stack frame, which tells the ORC unwinder to decrement the instruction pointer before looking up the corresponding ORC entry. Reported-by: Chen Zhongjin Signed-off-by: Josh Poimboeuf Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/baafcd3cc1abb14cb757fe081fa696012a5265ee.1676068346.git.jpoimboe@kernel.org --- arch/x86/entry/entry_64.S | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S index 15739a2c09833..8d21881adf86c 100644 --- a/arch/x86/entry/entry_64.S +++ b/arch/x86/entry/entry_64.S @@ -385,7 +385,14 @@ SYM_CODE_END(xen_error_entry) */ .macro idtentry vector asmsym cfunc has_error_code:req SYM_CODE_START(\asmsym) - UNWIND_HINT_IRET_REGS offset=\has_error_code*8 + + .if \vector == X86_TRAP_BP + /* #BP advances %rip to the next instruction */ + UNWIND_HINT_IRET_REGS offset=\has_error_code*8 signal=0 + .else + UNWIND_HINT_IRET_REGS offset=\has_error_code*8 + .endif + ENDBR ASM_CLAC cld From 9b30eac1b35a448081165cd906daf939d8750bfb Mon Sep 17 00:00:00 2001 From: Allen-KH Cheng Date: Tue, 17 Jan 2023 09:40:22 +0800 Subject: [PATCH 196/765] dt-bindings: watchdog: mtk-wdt: Add reset-by-toprgu support In some applications, the mtk-wdt requires the TOPRGU (Top Reset Generation Unit) to reset timer after system resets. Add optional mediatek,reset-by-toprgu property to enable it. Signed-off-by: Allen-KH Cheng Reviewed-by: AngeloGioacchino Del Regno Acked-by: Krzysztof Kozlowski Link: https://lore.kernel.org/r/20230117014023.2993-2-allen-kh.cheng@mediatek.com Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- .../devicetree/bindings/watchdog/mediatek,mtk-wdt.yaml | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/Documentation/devicetree/bindings/watchdog/mediatek,mtk-wdt.yaml b/Documentation/devicetree/bindings/watchdog/mediatek,mtk-wdt.yaml index b3605608410c6..55b34461df1bb 100644 --- a/Documentation/devicetree/bindings/watchdog/mediatek,mtk-wdt.yaml +++ b/Documentation/devicetree/bindings/watchdog/mediatek,mtk-wdt.yaml @@ -52,6 +52,12 @@ properties: description: Disable sending output reset signal type: boolean + mediatek,reset-by-toprgu: + description: The Top Reset Generation Unit (TOPRGU) generates reset signals + and distributes them to each IP. If present, the watchdog timer will be + reset by TOPRGU once system resets. + type: boolean + '#reset-cells': const: 1 From a224764f97d28f1132cab02b79ae4c08a6de471e Mon Sep 17 00:00:00 2001 From: Allen-KH Cheng Date: Tue, 17 Jan 2023 09:40:23 +0800 Subject: [PATCH 197/765] watchdog: mtk_wdt: Add reset_by_toprgu support In some cases, the MediaTek watchdog requires the TOPRGU to reset timer after system resets. Provide a reset_by_toprgu parameter for configuration. Signed-off-by: Allen-KH Cheng Reviewed-by: AngeloGioacchino Del Regno Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20230117014023.2993-3-allen-kh.cheng@mediatek.com Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/mtk_wdt.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/watchdog/mtk_wdt.c b/drivers/watchdog/mtk_wdt.c index 3e6212591e697..a9c437598e7eb 100644 --- a/drivers/watchdog/mtk_wdt.c +++ b/drivers/watchdog/mtk_wdt.c @@ -50,6 +50,7 @@ #define WDT_MODE_IRQ_EN (1 << 3) #define WDT_MODE_AUTO_START (1 << 4) #define WDT_MODE_DUAL_EN (1 << 6) +#define WDT_MODE_CNT_SEL (1 << 8) #define WDT_MODE_KEY 0x22000000 #define WDT_SWRST 0x14 @@ -70,6 +71,7 @@ struct mtk_wdt_dev { spinlock_t lock; /* protects WDT_SWSYSRST reg */ struct reset_controller_dev rcdev; bool disable_wdt_extrst; + bool reset_by_toprgu; }; struct mtk_wdt_data { @@ -279,6 +281,8 @@ static int mtk_wdt_start(struct watchdog_device *wdt_dev) reg &= ~(WDT_MODE_IRQ_EN | WDT_MODE_DUAL_EN); if (mtk_wdt->disable_wdt_extrst) reg &= ~WDT_MODE_EXRST_EN; + if (mtk_wdt->reset_by_toprgu) + reg |= WDT_MODE_CNT_SEL; reg |= (WDT_MODE_EN | WDT_MODE_KEY); iowrite32(reg, wdt_base + WDT_MODE); @@ -408,6 +412,9 @@ static int mtk_wdt_probe(struct platform_device *pdev) mtk_wdt->disable_wdt_extrst = of_property_read_bool(dev->of_node, "mediatek,disable-extrst"); + mtk_wdt->reset_by_toprgu = + of_property_read_bool(dev->of_node, "mediatek,reset-by-toprgu"); + return 0; } From 3c22939eb345952b5c814df9cc6f9b2d56982453 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 31 Dec 2022 16:02:17 +0100 Subject: [PATCH 198/765] watchdog: realtek_otto: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Acked-by: Sander Vanheule Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/5e4255782fbb43d1b4b5cd03bd12d7a184497134.1672498920.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/realtek_otto_wdt.c | 17 ++--------------- 1 file changed, 2 insertions(+), 15 deletions(-) diff --git a/drivers/watchdog/realtek_otto_wdt.c b/drivers/watchdog/realtek_otto_wdt.c index 2a5298c5e8e4d..2c30ddd574c59 100644 --- a/drivers/watchdog/realtek_otto_wdt.c +++ b/drivers/watchdog/realtek_otto_wdt.c @@ -235,27 +235,14 @@ static const struct watchdog_info otto_wdt_info = { WDIOF_PRETIMEOUT, }; -static void otto_wdt_clock_action(void *data) -{ - clk_disable_unprepare(data); -} - static int otto_wdt_probe_clk(struct otto_wdt_ctrl *ctrl) { - struct clk *clk = devm_clk_get(ctrl->dev, NULL); - int ret; + struct clk *clk; + clk = devm_clk_get_enabled(ctrl->dev, NULL); if (IS_ERR(clk)) return dev_err_probe(ctrl->dev, PTR_ERR(clk), "Failed to get clock\n"); - ret = clk_prepare_enable(clk); - if (ret) - return dev_err_probe(ctrl->dev, ret, "Failed to enable clock\n"); - - ret = devm_add_action_or_reset(ctrl->dev, otto_wdt_clock_action, clk); - if (ret) - return ret; - ctrl->clk_rate_khz = clk_get_rate(clk) / 1000; if (ctrl->clk_rate_khz == 0) return dev_err_probe(ctrl->dev, -ENXIO, "Failed to get clock rate\n"); From a8a9b98057458cfcbf145af647916ca9f1a63ccb Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 31 Dec 2022 15:25:31 +0100 Subject: [PATCH 199/765] watchdog: pnx4008: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/61f4e39db4c88408ee0149580e9aa925b784bc93.1672496714.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/pnx4008_wdt.c | 15 +-------------- 1 file changed, 1 insertion(+), 14 deletions(-) diff --git a/drivers/watchdog/pnx4008_wdt.c b/drivers/watchdog/pnx4008_wdt.c index e0ea133c1690e..87a44a5675a14 100644 --- a/drivers/watchdog/pnx4008_wdt.c +++ b/drivers/watchdog/pnx4008_wdt.c @@ -179,11 +179,6 @@ static struct watchdog_device pnx4008_wdd = { .max_timeout = MAX_HEARTBEAT, }; -static void pnx4008_clk_disable_unprepare(void *data) -{ - clk_disable_unprepare(data); -} - static int pnx4008_wdt_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; @@ -195,18 +190,10 @@ static int pnx4008_wdt_probe(struct platform_device *pdev) if (IS_ERR(wdt_base)) return PTR_ERR(wdt_base); - wdt_clk = devm_clk_get(dev, NULL); + wdt_clk = devm_clk_get_enabled(dev, NULL); if (IS_ERR(wdt_clk)) return PTR_ERR(wdt_clk); - ret = clk_prepare_enable(wdt_clk); - if (ret) - return ret; - ret = devm_add_action_or_reset(dev, pnx4008_clk_disable_unprepare, - wdt_clk); - if (ret) - return ret; - pnx4008_wdd.bootstatus = (readl(WDTIM_RES(wdt_base)) & WDOG_RESET) ? WDIOF_CARDRESET : 0; pnx4008_wdd.parent = dev; From 7f7f8ad0725cf31af68aaccccee6d24b8c886679 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 31 Dec 2022 15:22:57 +0100 Subject: [PATCH 200/765] watchdog: pic32-wdt: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/4335b4201b535ebc749a98bad0b99e3cb5317c39.1672496563.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/pic32-wdt.c | 17 +---------------- 1 file changed, 1 insertion(+), 16 deletions(-) diff --git a/drivers/watchdog/pic32-wdt.c b/drivers/watchdog/pic32-wdt.c index 41715d68d9e97..6d1a00222991f 100644 --- a/drivers/watchdog/pic32-wdt.c +++ b/drivers/watchdog/pic32-wdt.c @@ -162,11 +162,6 @@ static const struct of_device_id pic32_wdt_dt_ids[] = { }; MODULE_DEVICE_TABLE(of, pic32_wdt_dt_ids); -static void pic32_clk_disable_unprepare(void *data) -{ - clk_disable_unprepare(data); -} - static int pic32_wdt_drv_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; @@ -186,22 +181,12 @@ static int pic32_wdt_drv_probe(struct platform_device *pdev) if (!wdt->rst_base) return -ENOMEM; - wdt->clk = devm_clk_get(dev, NULL); + wdt->clk = devm_clk_get_enabled(dev, NULL); if (IS_ERR(wdt->clk)) { dev_err(dev, "clk not found\n"); return PTR_ERR(wdt->clk); } - ret = clk_prepare_enable(wdt->clk); - if (ret) { - dev_err(dev, "clk enable failed\n"); - return ret; - } - ret = devm_add_action_or_reset(dev, pic32_clk_disable_unprepare, - wdt->clk); - if (ret) - return ret; - if (pic32_wdt_is_win_enabled(wdt)) { dev_err(dev, "windowed-clear mode is not supported.\n"); return -ENODEV; From e0912ea8714deb598c7763ec60c9ba83b157acf7 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 31 Dec 2022 15:20:29 +0100 Subject: [PATCH 201/765] watchdog: pic32-dmt: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/f9a4dcfc6d31bd9c1417e2d97a40cc2c1dbc6f30.1672496405.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/pic32-dmt.c | 15 +-------------- 1 file changed, 1 insertion(+), 14 deletions(-) diff --git a/drivers/watchdog/pic32-dmt.c b/drivers/watchdog/pic32-dmt.c index f43062b3c4c81..bc4ccddc75a3f 100644 --- a/drivers/watchdog/pic32-dmt.c +++ b/drivers/watchdog/pic32-dmt.c @@ -164,11 +164,6 @@ static struct watchdog_device pic32_dmt_wdd = { .ops = &pic32_dmt_fops, }; -static void pic32_clk_disable_unprepare(void *data) -{ - clk_disable_unprepare(data); -} - static int pic32_dmt_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; @@ -184,20 +179,12 @@ static int pic32_dmt_probe(struct platform_device *pdev) if (IS_ERR(dmt->regs)) return PTR_ERR(dmt->regs); - dmt->clk = devm_clk_get(dev, NULL); + dmt->clk = devm_clk_get_enabled(dev, NULL); if (IS_ERR(dmt->clk)) { dev_err(dev, "clk not found\n"); return PTR_ERR(dmt->clk); } - ret = clk_prepare_enable(dmt->clk); - if (ret) - return ret; - ret = devm_add_action_or_reset(dev, pic32_clk_disable_unprepare, - dmt->clk); - if (ret) - return ret; - wdd->timeout = pic32_dmt_get_timeout_secs(dmt); if (!wdd->timeout) { dev_err(dev, "failed to read watchdog register timeout\n"); From 5eb3fb95eab28a6dbcd2eab6898e03af6658ee0e Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 31 Dec 2022 13:26:09 +0100 Subject: [PATCH 202/765] watchdog: lpc18xx: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/d4c675190d3ddfbba5c354edb4274757f9117304.1672489554.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/lpc18xx_wdt.c | 30 ++---------------------------- 1 file changed, 2 insertions(+), 28 deletions(-) diff --git a/drivers/watchdog/lpc18xx_wdt.c b/drivers/watchdog/lpc18xx_wdt.c index 60b6d74f267dd..1b9b5f21a0df5 100644 --- a/drivers/watchdog/lpc18xx_wdt.c +++ b/drivers/watchdog/lpc18xx_wdt.c @@ -197,16 +197,10 @@ static const struct watchdog_ops lpc18xx_wdt_ops = { .restart = lpc18xx_wdt_restart, }; -static void lpc18xx_clk_disable_unprepare(void *data) -{ - clk_disable_unprepare(data); -} - static int lpc18xx_wdt_probe(struct platform_device *pdev) { struct lpc18xx_wdt_dev *lpc18xx_wdt; struct device *dev = &pdev->dev; - int ret; lpc18xx_wdt = devm_kzalloc(dev, sizeof(*lpc18xx_wdt), GFP_KERNEL); if (!lpc18xx_wdt) @@ -216,38 +210,18 @@ static int lpc18xx_wdt_probe(struct platform_device *pdev) if (IS_ERR(lpc18xx_wdt->base)) return PTR_ERR(lpc18xx_wdt->base); - lpc18xx_wdt->reg_clk = devm_clk_get(dev, "reg"); + lpc18xx_wdt->reg_clk = devm_clk_get_enabled(dev, "reg"); if (IS_ERR(lpc18xx_wdt->reg_clk)) { dev_err(dev, "failed to get the reg clock\n"); return PTR_ERR(lpc18xx_wdt->reg_clk); } - lpc18xx_wdt->wdt_clk = devm_clk_get(dev, "wdtclk"); + lpc18xx_wdt->wdt_clk = devm_clk_get_enabled(dev, "wdtclk"); if (IS_ERR(lpc18xx_wdt->wdt_clk)) { dev_err(dev, "failed to get the wdt clock\n"); return PTR_ERR(lpc18xx_wdt->wdt_clk); } - ret = clk_prepare_enable(lpc18xx_wdt->reg_clk); - if (ret) { - dev_err(dev, "could not prepare or enable sys clock\n"); - return ret; - } - ret = devm_add_action_or_reset(dev, lpc18xx_clk_disable_unprepare, - lpc18xx_wdt->reg_clk); - if (ret) - return ret; - - ret = clk_prepare_enable(lpc18xx_wdt->wdt_clk); - if (ret) { - dev_err(dev, "could not prepare or enable wdt clock\n"); - return ret; - } - ret = devm_add_action_or_reset(dev, lpc18xx_clk_disable_unprepare, - lpc18xx_wdt->wdt_clk); - if (ret) - return ret; - /* We use the clock rate to calculate timeouts */ lpc18xx_wdt->clk_rate = clk_get_rate(lpc18xx_wdt->wdt_clk); if (lpc18xx_wdt->clk_rate == 0) { From 4de0224c6fbebfa4018a640ec34e5c160dffb959 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 31 Dec 2022 13:57:49 +0100 Subject: [PATCH 203/765] watchdog: of_xilinx_wdt: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Acked-by: Michal Simek Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/2b041dc8230a4ed255051bb2d323da8a51a8d0be.1672491445.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/of_xilinx_wdt.c | 16 +--------------- 1 file changed, 1 insertion(+), 15 deletions(-) diff --git a/drivers/watchdog/of_xilinx_wdt.c b/drivers/watchdog/of_xilinx_wdt.c index 3318544366b89..2a079ca04aa3b 100644 --- a/drivers/watchdog/of_xilinx_wdt.c +++ b/drivers/watchdog/of_xilinx_wdt.c @@ -154,11 +154,6 @@ static u32 xwdt_selftest(struct xwdt_device *xdev) return XWT_TIMER_FAILED; } -static void xwdt_clk_disable_unprepare(void *data) -{ - clk_disable_unprepare(data); -} - static int xwdt_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; @@ -193,7 +188,7 @@ static int xwdt_probe(struct platform_device *pdev) watchdog_set_nowayout(xilinx_wdt_wdd, enable_once); - xdev->clk = devm_clk_get(dev, NULL); + xdev->clk = devm_clk_get_enabled(dev, NULL); if (IS_ERR(xdev->clk)) { if (PTR_ERR(xdev->clk) != -ENOENT) return PTR_ERR(xdev->clk); @@ -211,15 +206,6 @@ static int xwdt_probe(struct platform_device *pdev) "The watchdog clock freq cannot be obtained\n"); } else { pfreq = clk_get_rate(xdev->clk); - rc = clk_prepare_enable(xdev->clk); - if (rc) { - dev_err(dev, "unable to enable clock\n"); - return rc; - } - rc = devm_add_action_or_reset(dev, xwdt_clk_disable_unprepare, - xdev->clk); - if (rc) - return rc; } /* From c4b8e92bc1a8d860ca53a41c25cd84fa03efaf50 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 31 Dec 2022 11:53:34 +0100 Subject: [PATCH 204/765] watchdog: imgpdc: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/1f8d1ce1e6a63c507a291aea624b1337326cc563.1672483996.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/imgpdc_wdt.c | 31 +++---------------------------- 1 file changed, 3 insertions(+), 28 deletions(-) diff --git a/drivers/watchdog/imgpdc_wdt.c b/drivers/watchdog/imgpdc_wdt.c index b57ff3787052d..a55f801895d48 100644 --- a/drivers/watchdog/imgpdc_wdt.c +++ b/drivers/watchdog/imgpdc_wdt.c @@ -175,16 +175,11 @@ static const struct watchdog_ops pdc_wdt_ops = { .restart = pdc_wdt_restart, }; -static void pdc_clk_disable_unprepare(void *data) -{ - clk_disable_unprepare(data); -} - static int pdc_wdt_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; u64 div; - int ret, val; + int val; unsigned long clk_rate; struct pdc_wdt_dev *pdc_wdt; @@ -196,38 +191,18 @@ static int pdc_wdt_probe(struct platform_device *pdev) if (IS_ERR(pdc_wdt->base)) return PTR_ERR(pdc_wdt->base); - pdc_wdt->sys_clk = devm_clk_get(dev, "sys"); + pdc_wdt->sys_clk = devm_clk_get_enabled(dev, "sys"); if (IS_ERR(pdc_wdt->sys_clk)) { dev_err(dev, "failed to get the sys clock\n"); return PTR_ERR(pdc_wdt->sys_clk); } - pdc_wdt->wdt_clk = devm_clk_get(dev, "wdt"); + pdc_wdt->wdt_clk = devm_clk_get_enabled(dev, "wdt"); if (IS_ERR(pdc_wdt->wdt_clk)) { dev_err(dev, "failed to get the wdt clock\n"); return PTR_ERR(pdc_wdt->wdt_clk); } - ret = clk_prepare_enable(pdc_wdt->sys_clk); - if (ret) { - dev_err(dev, "could not prepare or enable sys clock\n"); - return ret; - } - ret = devm_add_action_or_reset(dev, pdc_clk_disable_unprepare, - pdc_wdt->sys_clk); - if (ret) - return ret; - - ret = clk_prepare_enable(pdc_wdt->wdt_clk); - if (ret) { - dev_err(dev, "could not prepare or enable wdt clock\n"); - return ret; - } - ret = devm_add_action_or_reset(dev, pdc_clk_disable_unprepare, - pdc_wdt->wdt_clk); - if (ret) - return ret; - /* We use the clock rate to calculate the max timeout */ clk_rate = clk_get_rate(pdc_wdt->wdt_clk); if (clk_rate == 0) { From 616a2fe3cda63d4e6eb3a29f925f14b3766c633e Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 31 Dec 2022 09:12:28 +0100 Subject: [PATCH 205/765] watchdog: cadence: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/615c6c3c46c3ee8e3136725af0ab0b51e1298091.1672474336.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/cadence_wdt.c | 17 +---------------- 1 file changed, 1 insertion(+), 16 deletions(-) diff --git a/drivers/watchdog/cadence_wdt.c b/drivers/watchdog/cadence_wdt.c index bc99e91649306..23d41043863f6 100644 --- a/drivers/watchdog/cadence_wdt.c +++ b/drivers/watchdog/cadence_wdt.c @@ -274,11 +274,6 @@ static const struct watchdog_ops cdns_wdt_ops = { .set_timeout = cdns_wdt_settimeout, }; -static void cdns_clk_disable_unprepare(void *data) -{ - clk_disable_unprepare(data); -} - /************************Platform Operations*****************************/ /** * cdns_wdt_probe - Probe call for the device. @@ -333,21 +328,11 @@ static int cdns_wdt_probe(struct platform_device *pdev) watchdog_stop_on_reboot(cdns_wdt_device); watchdog_set_drvdata(cdns_wdt_device, wdt); - wdt->clk = devm_clk_get(dev, NULL); + wdt->clk = devm_clk_get_enabled(dev, NULL); if (IS_ERR(wdt->clk)) return dev_err_probe(dev, PTR_ERR(wdt->clk), "input clock not found\n"); - ret = clk_prepare_enable(wdt->clk); - if (ret) { - dev_err(dev, "unable to enable clock\n"); - return ret; - } - ret = devm_add_action_or_reset(dev, cdns_clk_disable_unprepare, - wdt->clk); - if (ret) - return ret; - clock_f = clk_get_rate(wdt->clk); if (clock_f <= CDNS_WDT_CLK_75MHZ) { wdt->prescaler = CDNS_WDT_PRESCALE_512; From 2cf46c636f4010baf696bb6afa43bebb2aa30a7a Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 31 Dec 2022 13:34:44 +0100 Subject: [PATCH 206/765] watchdog: meson_gxbb: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Reviewed-by: Guenter Roeck Reviewed-by: Martin Blumenstingl Link: https://lore.kernel.org/r/6c5948373d309408095c1a098b7b4c491c5265f7.1672490071.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/meson_gxbb_wdt.c | 16 +--------------- 1 file changed, 1 insertion(+), 15 deletions(-) diff --git a/drivers/watchdog/meson_gxbb_wdt.c b/drivers/watchdog/meson_gxbb_wdt.c index 981a2f7c3bec2..35d80cb39856c 100644 --- a/drivers/watchdog/meson_gxbb_wdt.c +++ b/drivers/watchdog/meson_gxbb_wdt.c @@ -146,16 +146,10 @@ static const struct of_device_id meson_gxbb_wdt_dt_ids[] = { }; MODULE_DEVICE_TABLE(of, meson_gxbb_wdt_dt_ids); -static void meson_clk_disable_unprepare(void *data) -{ - clk_disable_unprepare(data); -} - static int meson_gxbb_wdt_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct meson_gxbb_wdt *data; - int ret; u32 ctrl_reg; data = devm_kzalloc(dev, sizeof(*data), GFP_KERNEL); @@ -166,18 +160,10 @@ static int meson_gxbb_wdt_probe(struct platform_device *pdev) if (IS_ERR(data->reg_base)) return PTR_ERR(data->reg_base); - data->clk = devm_clk_get(dev, NULL); + data->clk = devm_clk_get_enabled(dev, NULL); if (IS_ERR(data->clk)) return PTR_ERR(data->clk); - ret = clk_prepare_enable(data->clk); - if (ret) - return ret; - ret = devm_add_action_or_reset(dev, meson_clk_disable_unprepare, - data->clk); - if (ret) - return ret; - platform_set_drvdata(pdev, data); data->wdt_dev.parent = dev; From 0a03207205f95abd90c8e1abb00c158066560ed2 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 31 Dec 2022 11:37:39 +0100 Subject: [PATCH 207/765] watchdog: davinci: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/6a4cf8e8b9d8f555c77395ba2ecadc205553774d.1672483046.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/davinci_wdt.c | 18 +----------------- 1 file changed, 1 insertion(+), 17 deletions(-) diff --git a/drivers/watchdog/davinci_wdt.c b/drivers/watchdog/davinci_wdt.c index 584a56893b81c..5f2184bda7b27 100644 --- a/drivers/watchdog/davinci_wdt.c +++ b/drivers/watchdog/davinci_wdt.c @@ -189,14 +189,8 @@ static const struct watchdog_ops davinci_wdt_ops = { .restart = davinci_wdt_restart, }; -static void davinci_clk_disable_unprepare(void *data) -{ - clk_disable_unprepare(data); -} - static int davinci_wdt_probe(struct platform_device *pdev) { - int ret = 0; struct device *dev = &pdev->dev; struct watchdog_device *wdd; struct davinci_wdt_device *davinci_wdt; @@ -205,21 +199,11 @@ static int davinci_wdt_probe(struct platform_device *pdev) if (!davinci_wdt) return -ENOMEM; - davinci_wdt->clk = devm_clk_get(dev, NULL); + davinci_wdt->clk = devm_clk_get_enabled(dev, NULL); if (IS_ERR(davinci_wdt->clk)) return dev_err_probe(dev, PTR_ERR(davinci_wdt->clk), "failed to get clock node\n"); - ret = clk_prepare_enable(davinci_wdt->clk); - if (ret) { - dev_err(dev, "failed to prepare clock\n"); - return ret; - } - ret = devm_add_action_or_reset(dev, davinci_clk_disable_unprepare, - davinci_wdt->clk); - if (ret) - return ret; - platform_set_drvdata(pdev, davinci_wdt); wdd = &davinci_wdt->wdd; From f3a0dd648961f09936278752a8e9299a55f82bc5 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 31 Dec 2022 11:59:57 +0100 Subject: [PATCH 208/765] watchdog: imx7ulp: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/f23a2cf84958adca255b82fd688e7cee0461760f.1672484376.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/imx7ulp_wdt.c | 15 +-------------- 1 file changed, 1 insertion(+), 14 deletions(-) diff --git a/drivers/watchdog/imx7ulp_wdt.c b/drivers/watchdog/imx7ulp_wdt.c index 2897902090b39..7ca486794ba7f 100644 --- a/drivers/watchdog/imx7ulp_wdt.c +++ b/drivers/watchdog/imx7ulp_wdt.c @@ -299,11 +299,6 @@ static int imx7ulp_wdt_init(struct imx7ulp_wdt_device *wdt, unsigned int timeout return ret; } -static void imx7ulp_wdt_action(void *data) -{ - clk_disable_unprepare(data); -} - static int imx7ulp_wdt_probe(struct platform_device *pdev) { struct imx7ulp_wdt_device *imx7ulp_wdt; @@ -321,7 +316,7 @@ static int imx7ulp_wdt_probe(struct platform_device *pdev) if (IS_ERR(imx7ulp_wdt->base)) return PTR_ERR(imx7ulp_wdt->base); - imx7ulp_wdt->clk = devm_clk_get(dev, NULL); + imx7ulp_wdt->clk = devm_clk_get_enabled(dev, NULL); if (IS_ERR(imx7ulp_wdt->clk)) { dev_err(dev, "Failed to get watchdog clock\n"); return PTR_ERR(imx7ulp_wdt->clk); @@ -336,14 +331,6 @@ static int imx7ulp_wdt_probe(struct platform_device *pdev) dev_info(dev, "imx7ulp wdt probe\n"); } - ret = clk_prepare_enable(imx7ulp_wdt->clk); - if (ret) - return ret; - - ret = devm_add_action_or_reset(dev, imx7ulp_wdt_action, imx7ulp_wdt->clk); - if (ret) - return ret; - wdog = &imx7ulp_wdt->wdd; wdog->info = &imx7ulp_wdt_info; wdog->ops = &imx7ulp_wdt_ops; From 6cc0768ba6f5293a4435612164a2cfbf2edbd025 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Fri, 30 Dec 2022 17:41:31 +0100 Subject: [PATCH 209/765] watchdog: rtd119x: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/14b521b821279bc5111dc80b55d0936c5767c737.1672418470.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/rtd119x_wdt.c | 16 +--------------- 1 file changed, 1 insertion(+), 15 deletions(-) diff --git a/drivers/watchdog/rtd119x_wdt.c b/drivers/watchdog/rtd119x_wdt.c index 834b94ff3f903..95c8d7abce42e 100644 --- a/drivers/watchdog/rtd119x_wdt.c +++ b/drivers/watchdog/rtd119x_wdt.c @@ -94,16 +94,10 @@ static const struct of_device_id rtd119x_wdt_dt_ids[] = { { } }; -static void rtd119x_clk_disable_unprepare(void *data) -{ - clk_disable_unprepare(data); -} - static int rtd119x_wdt_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct rtd119x_watchdog_device *data; - int ret; data = devm_kzalloc(dev, sizeof(*data), GFP_KERNEL); if (!data) @@ -113,18 +107,10 @@ static int rtd119x_wdt_probe(struct platform_device *pdev) if (IS_ERR(data->base)) return PTR_ERR(data->base); - data->clk = devm_clk_get(dev, NULL); + data->clk = devm_clk_get_enabled(dev, NULL); if (IS_ERR(data->clk)) return PTR_ERR(data->clk); - ret = clk_prepare_enable(data->clk); - if (ret) - return ret; - ret = devm_add_action_or_reset(dev, rtd119x_clk_disable_unprepare, - data->clk); - if (ret) - return ret; - data->wdt_dev.info = &rtd119x_wdt_info; data->wdt_dev.ops = &rtd119x_wdt_ops; data->wdt_dev.timeout = 120; From be80ae3d799dab1b46c661158b2b5d6695258178 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 31 Dec 2022 09:10:17 +0100 Subject: [PATCH 210/765] watchdog: bcm7038: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Acked-by: Florian Fainelli Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/9c055911e9f557b7239000c8e6cfa0cc393a19e9.1672474203.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/bcm7038_wdt.c | 15 +-------------- 1 file changed, 1 insertion(+), 14 deletions(-) diff --git a/drivers/watchdog/bcm7038_wdt.c b/drivers/watchdog/bcm7038_wdt.c index 9388838899aca..e038dd66b8198 100644 --- a/drivers/watchdog/bcm7038_wdt.c +++ b/drivers/watchdog/bcm7038_wdt.c @@ -127,11 +127,6 @@ static const struct watchdog_ops bcm7038_wdt_ops = { .get_timeleft = bcm7038_wdt_get_timeleft, }; -static void bcm7038_clk_disable_unprepare(void *data) -{ - clk_disable_unprepare(data); -} - static int bcm7038_wdt_probe(struct platform_device *pdev) { struct bcm7038_wdt_platform_data *pdata = pdev->dev.platform_data; @@ -153,17 +148,9 @@ static int bcm7038_wdt_probe(struct platform_device *pdev) if (pdata && pdata->clk_name) clk_name = pdata->clk_name; - wdt->clk = devm_clk_get(dev, clk_name); + wdt->clk = devm_clk_get_enabled(dev, clk_name); /* If unable to get clock, use default frequency */ if (!IS_ERR(wdt->clk)) { - err = clk_prepare_enable(wdt->clk); - if (err) - return err; - err = devm_add_action_or_reset(dev, - bcm7038_clk_disable_unprepare, - wdt->clk); - if (err) - return err; wdt->rate = clk_get_rate(wdt->clk); /* Prevent divide-by-zero exception */ if (!wdt->rate) From cd9e5d04dc47264e1101ed134e383020ab45a272 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 31 Dec 2022 09:00:31 +0100 Subject: [PATCH 211/765] watchdog: armada_37xx: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/ccb096879a1309b9918ae956d6bdb9668c69bcda.1672473617.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/armada_37xx_wdt.c | 15 +-------------- 1 file changed, 1 insertion(+), 14 deletions(-) diff --git a/drivers/watchdog/armada_37xx_wdt.c b/drivers/watchdog/armada_37xx_wdt.c index ac9fed1ef681b..e58652939f8a9 100644 --- a/drivers/watchdog/armada_37xx_wdt.c +++ b/drivers/watchdog/armada_37xx_wdt.c @@ -246,11 +246,6 @@ static const struct watchdog_ops armada_37xx_wdt_ops = { .get_timeleft = armada_37xx_wdt_get_timeleft, }; -static void armada_clk_disable_unprepare(void *data) -{ - clk_disable_unprepare(data); -} - static int armada_37xx_wdt_probe(struct platform_device *pdev) { struct armada_37xx_watchdog *dev; @@ -280,18 +275,10 @@ static int armada_37xx_wdt_probe(struct platform_device *pdev) return -ENOMEM; /* init clock */ - dev->clk = devm_clk_get(&pdev->dev, NULL); + dev->clk = devm_clk_get_enabled(&pdev->dev, NULL); if (IS_ERR(dev->clk)) return PTR_ERR(dev->clk); - ret = clk_prepare_enable(dev->clk); - if (ret) - return ret; - ret = devm_add_action_or_reset(&pdev->dev, - armada_clk_disable_unprepare, dev->clk); - if (ret) - return ret; - dev->clk_rate = clk_get_rate(dev->clk); if (!dev->clk_rate) return -EINVAL; From 1a8d192bccb2fbbdcfcd13e89e86ab8a0945ea75 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Fri, 30 Dec 2022 17:49:47 +0100 Subject: [PATCH 212/765] watchdog: qcom: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/7c2d5f3815949faf6d3a0237a7b5f272f00a7ae9.1672418969.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/qcom-wdt.c | 16 +--------------- 1 file changed, 1 insertion(+), 15 deletions(-) diff --git a/drivers/watchdog/qcom-wdt.c b/drivers/watchdog/qcom-wdt.c index 0d2209c5eaca7..d776474dcdf34 100644 --- a/drivers/watchdog/qcom-wdt.c +++ b/drivers/watchdog/qcom-wdt.c @@ -175,11 +175,6 @@ static const struct watchdog_info qcom_wdt_pt_info = { .identity = KBUILD_MODNAME, }; -static void qcom_clk_disable_unprepare(void *data) -{ - clk_disable_unprepare(data); -} - static const struct qcom_wdt_match_data match_data_apcs_tmr = { .offset = reg_offset_data_apcs_tmr, .pretimeout = false, @@ -226,21 +221,12 @@ static int qcom_wdt_probe(struct platform_device *pdev) if (IS_ERR(wdt->base)) return PTR_ERR(wdt->base); - clk = devm_clk_get(dev, NULL); + clk = devm_clk_get_enabled(dev, NULL); if (IS_ERR(clk)) { dev_err(dev, "failed to get input clock\n"); return PTR_ERR(clk); } - ret = clk_prepare_enable(clk); - if (ret) { - dev_err(dev, "failed to setup clock\n"); - return ret; - } - ret = devm_add_action_or_reset(dev, qcom_clk_disable_unprepare, clk); - if (ret) - return ret; - /* * We use the clock rate to calculate the max timeout, so ensure it's * not zero to avoid a divide-by-zero exception. From a42912ac400fdc05cffb6bc1e263eff466bbf8b2 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Fri, 30 Dec 2022 17:36:20 +0100 Subject: [PATCH 213/765] watchdog: rzn1: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/b1f8b5453791035ad534bd5ed36b49798ff4d9b2.1672418166.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/rzn1_wdt.c | 18 +----------------- 1 file changed, 1 insertion(+), 17 deletions(-) diff --git a/drivers/watchdog/rzn1_wdt.c b/drivers/watchdog/rzn1_wdt.c index 55ab384b99657..980c1717adb5d 100644 --- a/drivers/watchdog/rzn1_wdt.c +++ b/drivers/watchdog/rzn1_wdt.c @@ -98,11 +98,6 @@ static const struct watchdog_ops rzn1_wdt_ops = { .ping = rzn1_wdt_ping, }; -static void rzn1_wdt_clk_disable_unprepare(void *data) -{ - clk_disable_unprepare(data); -} - static int rzn1_wdt_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; @@ -132,23 +127,12 @@ static int rzn1_wdt_probe(struct platform_device *pdev) return ret; } - clk = devm_clk_get(dev, NULL); + clk = devm_clk_get_enabled(dev, NULL); if (IS_ERR(clk)) { dev_err(dev, "failed to get the clock\n"); return PTR_ERR(clk); } - ret = clk_prepare_enable(clk); - if (ret) { - dev_err(dev, "failed to prepare/enable the clock\n"); - return ret; - } - - ret = devm_add_action_or_reset(dev, rzn1_wdt_clk_disable_unprepare, - clk); - if (ret) - return ret; - clk_rate = clk_get_rate(clk); if (!clk_rate) { dev_err(dev, "failed to get the clock rate\n"); From 8c5210dbdf7e924b435c892d979472191d9cec95 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Fri, 30 Dec 2022 17:32:42 +0100 Subject: [PATCH 214/765] watchdog: visconti: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/13e8cdf17556da111d1d98a8fe0b1dc1c78007e2.1672417940.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/visconti_wdt.c | 17 +---------------- 1 file changed, 1 insertion(+), 16 deletions(-) diff --git a/drivers/watchdog/visconti_wdt.c b/drivers/watchdog/visconti_wdt.c index 83ef55e66ca86..cef0794708e7e 100644 --- a/drivers/watchdog/visconti_wdt.c +++ b/drivers/watchdog/visconti_wdt.c @@ -112,11 +112,6 @@ static const struct watchdog_ops visconti_wdt_ops = { .set_timeout = visconti_wdt_set_timeout, }; -static void visconti_clk_disable_unprepare(void *data) -{ - clk_disable_unprepare(data); -} - static int visconti_wdt_probe(struct platform_device *pdev) { struct watchdog_device *wdev; @@ -134,20 +129,10 @@ static int visconti_wdt_probe(struct platform_device *pdev) if (IS_ERR(priv->base)) return PTR_ERR(priv->base); - clk = devm_clk_get(dev, NULL); + clk = devm_clk_get_enabled(dev, NULL); if (IS_ERR(clk)) return dev_err_probe(dev, PTR_ERR(clk), "Could not get clock\n"); - ret = clk_prepare_enable(clk); - if (ret) { - dev_err(dev, "Could not enable clock\n"); - return ret; - } - - ret = devm_add_action_or_reset(dev, visconti_clk_disable_unprepare, clk); - if (ret) - return ret; - clk_freq = clk_get_rate(clk); if (!clk_freq) return -EINVAL; From 98b7a1613011d9d5ec76841a2a5a689c3010d1f3 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 31 Dec 2022 08:57:22 +0100 Subject: [PATCH 215/765] watchdog: apple: Use devm_clk_get_enabled() helper The devm_clk_get_enabled() helper: - calls devm_clk_get() - calls clk_prepare_enable() and registers what is needed in order to call clk_disable_unprepare() when needed, as a managed resource. This simplifies the code and avoids the need of a dedicated function used with devm_add_action_or_reset(). Signed-off-by: Christophe JAILLET Reviewed-by: Guenter Roeck Reviewed-by: Sven Peter Link: https://lore.kernel.org/r/6f312af6160d1e10b616c9adbd1fd8f822db964d.1672473415.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/apple_wdt.c | 18 +----------------- 1 file changed, 1 insertion(+), 17 deletions(-) diff --git a/drivers/watchdog/apple_wdt.c b/drivers/watchdog/apple_wdt.c index 16aca21f13d6a..eddeb0fede896 100644 --- a/drivers/watchdog/apple_wdt.c +++ b/drivers/watchdog/apple_wdt.c @@ -136,11 +136,6 @@ static int apple_wdt_restart(struct watchdog_device *wdd, unsigned long mode, return 0; } -static void apple_wdt_clk_disable_unprepare(void *data) -{ - clk_disable_unprepare(data); -} - static struct watchdog_ops apple_wdt_ops = { .owner = THIS_MODULE, .start = apple_wdt_start, @@ -162,7 +157,6 @@ static int apple_wdt_probe(struct platform_device *pdev) struct apple_wdt *wdt; struct clk *clk; u32 wdt_ctrl; - int ret; wdt = devm_kzalloc(dev, sizeof(*wdt), GFP_KERNEL); if (!wdt) @@ -172,19 +166,9 @@ static int apple_wdt_probe(struct platform_device *pdev) if (IS_ERR(wdt->regs)) return PTR_ERR(wdt->regs); - clk = devm_clk_get(dev, NULL); + clk = devm_clk_get_enabled(dev, NULL); if (IS_ERR(clk)) return PTR_ERR(clk); - - ret = clk_prepare_enable(clk); - if (ret) - return ret; - - ret = devm_add_action_or_reset(dev, apple_wdt_clk_disable_unprepare, - clk); - if (ret) - return ret; - wdt->clk_rate = clk_get_rate(clk); if (!wdt->clk_rate) return -EINVAL; From 00bdbc9a860152cafc0bcfa4fc70f38401101ebc Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Thu, 22 Dec 2022 23:43:31 +0000 Subject: [PATCH 216/765] watchdog: wdat_wdt: Avoid unimplemented get_timeleft MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit As per the specification the action QUERY_COUNTDOWN_PERIOD is optional. If the action is not implemented by the physical device the driver would always report "0" from get_timeleft(). Avoid confusing userspace by only providing get_timeleft() when implemented by the hardware. Signed-off-by: Thomas Weißschuh Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221221-wdat_wdt-timeleft-v1-1-8e8a314c36cc@weissschuh.net Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/wdat_wdt.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/watchdog/wdat_wdt.c b/drivers/watchdog/wdat_wdt.c index ce7a4a9e4b03c..0ba99bed59fc4 100644 --- a/drivers/watchdog/wdat_wdt.c +++ b/drivers/watchdog/wdat_wdt.c @@ -301,13 +301,12 @@ static const struct watchdog_info wdat_wdt_info = { .identity = "wdat_wdt", }; -static const struct watchdog_ops wdat_wdt_ops = { +static struct watchdog_ops wdat_wdt_ops = { .owner = THIS_MODULE, .start = wdat_wdt_start, .stop = wdat_wdt_stop, .ping = wdat_wdt_ping, .set_timeout = wdat_wdt_set_timeout, - .get_timeleft = wdat_wdt_get_timeleft, }; static int wdat_wdt_probe(struct platform_device *pdev) @@ -436,6 +435,9 @@ static int wdat_wdt_probe(struct platform_device *pdev) list_add_tail(&instr->node, instructions); } + if (wdat->instructions[ACPI_WDAT_GET_CURRENT_COUNTDOWN]) + wdat_wdt_ops.get_timeleft = wdat_wdt_get_timeleft; + wdat_wdt_boot_status(wdat); wdat_wdt_set_running(wdat); From c4d5660afbdcd3f0fa3bbf563e059511fba8445f Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:42 +1100 Subject: [PATCH 217/765] xfs: active perag reference counting We need to be able to dynamically remove instantiated AGs from memory safely, either for shrinking the filesystem or paging AG state in and out of memory (e.g. supporting millions of AGs). This means we need to be able to safely exclude operations from accessing perags while dynamic removal is in progress. To do this, introduce the concept of active and passive references. Active references are required for high level operations that make use of an AG for a given operation (e.g. allocation) and pin the perag in memory for the duration of the operation that is operating on the perag (e.g. transaction scope). This means we can fail to get an active reference to an AG, hence callers of the new active reference API must be able to handle lookup failure gracefully. Passive references are used in low level code, where we might need to access the perag structure for the purposes of completing high level operations. For example, buffers need to use passive references because: - we need to be able to do metadata IO during operations like grow and shrink transactions where high level active references to the AG have already been blocked - buffers need to pin the perag until they are reclaimed from memory, something that high level code has no direct control over. - unused cached buffers should not prevent a shrink from being started. Hence we have active references that will form exclusion barriers for operations to be performed on an AG, and passive references that will prevent reclaim of the perag until all objects with passive references have been reclaimed themselves. This patch introduce xfs_perag_grab()/xfs_perag_rele() as the API for active AG reference functionality. We also need to convert the for_each_perag*() iterators to use active references, which will start the process of converting high level code over to using active references. Conversion of non-iterator based code to active references will be done in followup patches. Note that the implementation using reference counting is really just a development vehicle for the API to ensure we don't have any leaks in the callers. Once we need to remove perag structures from memory dyanmically, we will need a much more robust per-ag state transition mechanism for preventing new references from being taken while we wait for existing references to drain before removal from memory can occur.... Signed-off-by: Dave Chinner Reviewed-by: Allison Henderson Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_ag.c | 70 +++++++++++++++++++++++++++++++++++++++ fs/xfs/libxfs/xfs_ag.h | 31 ++++++++++++----- fs/xfs/scrub/bmap.c | 2 +- fs/xfs/scrub/fscounters.c | 4 +-- fs/xfs/xfs_fsmap.c | 4 +-- fs/xfs/xfs_icache.c | 2 +- fs/xfs/xfs_iwalk.c | 6 ++-- fs/xfs/xfs_reflink.c | 2 +- fs/xfs/xfs_trace.h | 3 ++ 9 files changed, 105 insertions(+), 19 deletions(-) diff --git a/fs/xfs/libxfs/xfs_ag.c b/fs/xfs/libxfs/xfs_ag.c index bb0c700afe3cb..46e25c682bf41 100644 --- a/fs/xfs/libxfs/xfs_ag.c +++ b/fs/xfs/libxfs/xfs_ag.c @@ -94,6 +94,68 @@ xfs_perag_put( trace_xfs_perag_put(pag->pag_mount, pag->pag_agno, ref, _RET_IP_); } +/* + * Active references for perag structures. This is for short term access to the + * per ag structures for walking trees or accessing state. If an AG is being + * shrunk or is offline, then this will fail to find that AG and return NULL + * instead. + */ +struct xfs_perag * +xfs_perag_grab( + struct xfs_mount *mp, + xfs_agnumber_t agno) +{ + struct xfs_perag *pag; + + rcu_read_lock(); + pag = radix_tree_lookup(&mp->m_perag_tree, agno); + if (pag) { + trace_xfs_perag_grab(mp, pag->pag_agno, + atomic_read(&pag->pag_active_ref), _RET_IP_); + if (!atomic_inc_not_zero(&pag->pag_active_ref)) + pag = NULL; + } + rcu_read_unlock(); + return pag; +} + +/* + * search from @first to find the next perag with the given tag set. + */ +struct xfs_perag * +xfs_perag_grab_tag( + struct xfs_mount *mp, + xfs_agnumber_t first, + int tag) +{ + struct xfs_perag *pag; + int found; + + rcu_read_lock(); + found = radix_tree_gang_lookup_tag(&mp->m_perag_tree, + (void **)&pag, first, 1, tag); + if (found <= 0) { + rcu_read_unlock(); + return NULL; + } + trace_xfs_perag_grab_tag(mp, pag->pag_agno, + atomic_read(&pag->pag_active_ref), _RET_IP_); + if (!atomic_inc_not_zero(&pag->pag_active_ref)) + pag = NULL; + rcu_read_unlock(); + return pag; +} + +void +xfs_perag_rele( + struct xfs_perag *pag) +{ + trace_xfs_perag_rele(pag->pag_mount, pag->pag_agno, + atomic_read(&pag->pag_active_ref), _RET_IP_); + if (atomic_dec_and_test(&pag->pag_active_ref)) + wake_up(&pag->pag_active_wq); +} + /* * xfs_initialize_perag_data * @@ -196,6 +258,10 @@ xfs_free_perag( cancel_delayed_work_sync(&pag->pag_blockgc_work); xfs_buf_hash_destroy(pag); + /* drop the mount's active reference */ + xfs_perag_rele(pag); + XFS_IS_CORRUPT(pag->pag_mount, + atomic_read(&pag->pag_active_ref) != 0); call_rcu(&pag->rcu_head, __xfs_free_perag); } } @@ -314,6 +380,7 @@ xfs_initialize_perag( INIT_DELAYED_WORK(&pag->pag_blockgc_work, xfs_blockgc_worker); INIT_RADIX_TREE(&pag->pag_ici_root, GFP_ATOMIC); init_waitqueue_head(&pag->pagb_wait); + init_waitqueue_head(&pag->pag_active_wq); pag->pagb_count = 0; pag->pagb_tree = RB_ROOT; #endif /* __KERNEL__ */ @@ -322,6 +389,9 @@ xfs_initialize_perag( if (error) goto out_remove_pag; + /* Active ref owned by mount indicates AG is online. */ + atomic_set(&pag->pag_active_ref, 1); + /* first new pag is fully initialized */ if (first_initialised == NULLAGNUMBER) first_initialised = index; diff --git a/fs/xfs/libxfs/xfs_ag.h b/fs/xfs/libxfs/xfs_ag.h index 191b22b9a35bf..aeb21c8df201e 100644 --- a/fs/xfs/libxfs/xfs_ag.h +++ b/fs/xfs/libxfs/xfs_ag.h @@ -32,7 +32,9 @@ struct xfs_ag_resv { struct xfs_perag { struct xfs_mount *pag_mount; /* owner filesystem */ xfs_agnumber_t pag_agno; /* AG this structure belongs to */ - atomic_t pag_ref; /* perag reference count */ + atomic_t pag_ref; /* passive reference count */ + atomic_t pag_active_ref; /* active reference count */ + wait_queue_head_t pag_active_wq;/* woken active_ref falls to zero */ char pagf_init; /* this agf's entry is initialized */ char pagi_init; /* this agi's entry is initialized */ char pagf_metadata; /* the agf is preferred to be metadata */ @@ -111,11 +113,18 @@ int xfs_initialize_perag(struct xfs_mount *mp, xfs_agnumber_t agcount, int xfs_initialize_perag_data(struct xfs_mount *mp, xfs_agnumber_t agno); void xfs_free_perag(struct xfs_mount *mp); +/* Passive AG references */ struct xfs_perag *xfs_perag_get(struct xfs_mount *mp, xfs_agnumber_t agno); struct xfs_perag *xfs_perag_get_tag(struct xfs_mount *mp, xfs_agnumber_t agno, unsigned int tag); void xfs_perag_put(struct xfs_perag *pag); +/* Active AG references */ +struct xfs_perag *xfs_perag_grab(struct xfs_mount *, xfs_agnumber_t); +struct xfs_perag *xfs_perag_grab_tag(struct xfs_mount *, xfs_agnumber_t, + int tag); +void xfs_perag_rele(struct xfs_perag *pag); + /* * Per-ag geometry infomation and validation */ @@ -193,14 +202,18 @@ xfs_perag_next( struct xfs_mount *mp = pag->pag_mount; *agno = pag->pag_agno + 1; - xfs_perag_put(pag); - if (*agno > end_agno) - return NULL; - return xfs_perag_get(mp, *agno); + xfs_perag_rele(pag); + while (*agno <= end_agno) { + pag = xfs_perag_grab(mp, *agno); + if (pag) + return pag; + (*agno)++; + } + return NULL; } #define for_each_perag_range(mp, agno, end_agno, pag) \ - for ((pag) = xfs_perag_get((mp), (agno)); \ + for ((pag) = xfs_perag_grab((mp), (agno)); \ (pag) != NULL; \ (pag) = xfs_perag_next((pag), &(agno), (end_agno))) @@ -213,11 +226,11 @@ xfs_perag_next( for_each_perag_from((mp), (agno), (pag)) #define for_each_perag_tag(mp, agno, pag, tag) \ - for ((agno) = 0, (pag) = xfs_perag_get_tag((mp), 0, (tag)); \ + for ((agno) = 0, (pag) = xfs_perag_grab_tag((mp), 0, (tag)); \ (pag) != NULL; \ (agno) = (pag)->pag_agno + 1, \ - xfs_perag_put(pag), \ - (pag) = xfs_perag_get_tag((mp), (agno), (tag))) + xfs_perag_rele(pag), \ + (pag) = xfs_perag_grab_tag((mp), (agno), (tag))) struct aghdr_init_data { /* per ag data */ diff --git a/fs/xfs/scrub/bmap.c b/fs/xfs/scrub/bmap.c index d50d0eab196ab..dbbc7037074c4 100644 --- a/fs/xfs/scrub/bmap.c +++ b/fs/xfs/scrub/bmap.c @@ -662,7 +662,7 @@ xchk_bmap_check_rmaps( error = xchk_bmap_check_ag_rmaps(sc, whichfork, pag); if (error || (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)) { - xfs_perag_put(pag); + xfs_perag_rele(pag); return error; } } diff --git a/fs/xfs/scrub/fscounters.c b/fs/xfs/scrub/fscounters.c index 4777e7b89fdc6..ef97670970c31 100644 --- a/fs/xfs/scrub/fscounters.c +++ b/fs/xfs/scrub/fscounters.c @@ -117,7 +117,7 @@ xchk_fscount_warmup( if (agi_bp) xfs_buf_relse(agi_bp); if (pag) - xfs_perag_put(pag); + xfs_perag_rele(pag); return error; } @@ -249,7 +249,7 @@ xchk_fscount_aggregate_agcounts( } if (pag) - xfs_perag_put(pag); + xfs_perag_rele(pag); if (error) { xchk_set_incomplete(sc); return error; diff --git a/fs/xfs/xfs_fsmap.c b/fs/xfs/xfs_fsmap.c index 88a88506ffffe..120d284a03fe4 100644 --- a/fs/xfs/xfs_fsmap.c +++ b/fs/xfs/xfs_fsmap.c @@ -688,11 +688,11 @@ __xfs_getfsmap_datadev( info->agf_bp = NULL; } if (info->pag) { - xfs_perag_put(info->pag); + xfs_perag_rele(info->pag); info->pag = NULL; } else if (pag) { /* loop termination case */ - xfs_perag_put(pag); + xfs_perag_rele(pag); } return error; diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c index ddeaccc04aec9..0f4a014dded3d 100644 --- a/fs/xfs/xfs_icache.c +++ b/fs/xfs/xfs_icache.c @@ -1767,7 +1767,7 @@ xfs_icwalk( if (error) { last_error = error; if (error == -EFSCORRUPTED) { - xfs_perag_put(pag); + xfs_perag_rele(pag); break; } } diff --git a/fs/xfs/xfs_iwalk.c b/fs/xfs/xfs_iwalk.c index 7558486f49371..c31857d903a45 100644 --- a/fs/xfs/xfs_iwalk.c +++ b/fs/xfs/xfs_iwalk.c @@ -591,7 +591,7 @@ xfs_iwalk( } if (iwag.pag) - xfs_perag_put(pag); + xfs_perag_rele(pag); xfs_iwalk_free(&iwag); return error; } @@ -683,7 +683,7 @@ xfs_iwalk_threaded( break; } if (pag) - xfs_perag_put(pag); + xfs_perag_rele(pag); if (polled) xfs_pwork_poll(&pctl); return xfs_pwork_destroy(&pctl); @@ -776,7 +776,7 @@ xfs_inobt_walk( } if (iwag.pag) - xfs_perag_put(pag); + xfs_perag_rele(pag); xfs_iwalk_free(&iwag); return error; } diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c index 57bf59ff48543..f5dc46ce98031 100644 --- a/fs/xfs/xfs_reflink.c +++ b/fs/xfs/xfs_reflink.c @@ -927,7 +927,7 @@ xfs_reflink_recover_cow( for_each_perag(mp, agno, pag) { error = xfs_refcount_recover_cow_leftovers(mp, pag); if (error) { - xfs_perag_put(pag); + xfs_perag_rele(pag); break; } } diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 2ac98d8ddbfde..eb5e49d44f13e 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -189,6 +189,9 @@ DEFINE_EVENT(xfs_perag_class, name, \ DEFINE_PERAG_REF_EVENT(xfs_perag_get); DEFINE_PERAG_REF_EVENT(xfs_perag_get_tag); DEFINE_PERAG_REF_EVENT(xfs_perag_put); +DEFINE_PERAG_REF_EVENT(xfs_perag_grab); +DEFINE_PERAG_REF_EVENT(xfs_perag_grab_tag); +DEFINE_PERAG_REF_EVENT(xfs_perag_rele); DEFINE_PERAG_REF_EVENT(xfs_perag_set_inode_tag); DEFINE_PERAG_REF_EVENT(xfs_perag_clear_inode_tag); From 368e2d09b41caa5b44a61bb518c362f46d6d615c Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:52 +1100 Subject: [PATCH 218/765] xfs: rework the perag trace points to be perag centric So that they all output the same information in the traces to make debugging refcount issues easier. This means that all the lookup/drop functions no longer need to use the full memory barrier atomic operations (atomic*_return()) so will have less overhead when tracing is off. The set/clear tag tracepoints no longer abuse the reference count to pass the tag - the tag being cleared is obvious from the _RET_IP_ that is recorded in the trace point. Signed-off-by: Dave Chinner Reviewed-by: Allison Henderson Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_ag.c | 25 +++++++++---------------- fs/xfs/xfs_icache.c | 4 ++-- fs/xfs/xfs_trace.h | 21 +++++++++++---------- 3 files changed, 22 insertions(+), 28 deletions(-) diff --git a/fs/xfs/libxfs/xfs_ag.c b/fs/xfs/libxfs/xfs_ag.c index 46e25c682bf41..7cff618753402 100644 --- a/fs/xfs/libxfs/xfs_ag.c +++ b/fs/xfs/libxfs/xfs_ag.c @@ -44,16 +44,15 @@ xfs_perag_get( xfs_agnumber_t agno) { struct xfs_perag *pag; - int ref = 0; rcu_read_lock(); pag = radix_tree_lookup(&mp->m_perag_tree, agno); if (pag) { + trace_xfs_perag_get(pag, _RET_IP_); ASSERT(atomic_read(&pag->pag_ref) >= 0); - ref = atomic_inc_return(&pag->pag_ref); + atomic_inc(&pag->pag_ref); } rcu_read_unlock(); - trace_xfs_perag_get(mp, agno, ref, _RET_IP_); return pag; } @@ -68,7 +67,6 @@ xfs_perag_get_tag( { struct xfs_perag *pag; int found; - int ref; rcu_read_lock(); found = radix_tree_gang_lookup_tag(&mp->m_perag_tree, @@ -77,9 +75,9 @@ xfs_perag_get_tag( rcu_read_unlock(); return NULL; } - ref = atomic_inc_return(&pag->pag_ref); + trace_xfs_perag_get_tag(pag, _RET_IP_); + atomic_inc(&pag->pag_ref); rcu_read_unlock(); - trace_xfs_perag_get_tag(mp, pag->pag_agno, ref, _RET_IP_); return pag; } @@ -87,11 +85,9 @@ void xfs_perag_put( struct xfs_perag *pag) { - int ref; - + trace_xfs_perag_put(pag, _RET_IP_); ASSERT(atomic_read(&pag->pag_ref) > 0); - ref = atomic_dec_return(&pag->pag_ref); - trace_xfs_perag_put(pag->pag_mount, pag->pag_agno, ref, _RET_IP_); + atomic_dec(&pag->pag_ref); } /* @@ -110,8 +106,7 @@ xfs_perag_grab( rcu_read_lock(); pag = radix_tree_lookup(&mp->m_perag_tree, agno); if (pag) { - trace_xfs_perag_grab(mp, pag->pag_agno, - atomic_read(&pag->pag_active_ref), _RET_IP_); + trace_xfs_perag_grab(pag, _RET_IP_); if (!atomic_inc_not_zero(&pag->pag_active_ref)) pag = NULL; } @@ -138,8 +133,7 @@ xfs_perag_grab_tag( rcu_read_unlock(); return NULL; } - trace_xfs_perag_grab_tag(mp, pag->pag_agno, - atomic_read(&pag->pag_active_ref), _RET_IP_); + trace_xfs_perag_grab_tag(pag, _RET_IP_); if (!atomic_inc_not_zero(&pag->pag_active_ref)) pag = NULL; rcu_read_unlock(); @@ -150,8 +144,7 @@ void xfs_perag_rele( struct xfs_perag *pag) { - trace_xfs_perag_rele(pag->pag_mount, pag->pag_agno, - atomic_read(&pag->pag_active_ref), _RET_IP_); + trace_xfs_perag_rele(pag, _RET_IP_); if (atomic_dec_and_test(&pag->pag_active_ref)) wake_up(&pag->pag_active_wq); } diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c index 0f4a014dded3d..8b2823d85a68c 100644 --- a/fs/xfs/xfs_icache.c +++ b/fs/xfs/xfs_icache.c @@ -255,7 +255,7 @@ xfs_perag_set_inode_tag( break; } - trace_xfs_perag_set_inode_tag(mp, pag->pag_agno, tag, _RET_IP_); + trace_xfs_perag_set_inode_tag(pag, _RET_IP_); } /* Clear a tag on both the AG incore inode tree and the AG radix tree. */ @@ -289,7 +289,7 @@ xfs_perag_clear_inode_tag( radix_tree_tag_clear(&mp->m_perag_tree, pag->pag_agno, tag); spin_unlock(&mp->m_perag_lock); - trace_xfs_perag_clear_inode_tag(mp, pag->pag_agno, tag, _RET_IP_); + trace_xfs_perag_clear_inode_tag(pag, _RET_IP_); } /* diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index eb5e49d44f13e..1d99c30f66f00 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -159,33 +159,34 @@ TRACE_EVENT(xlog_intent_recovery_failed, ); DECLARE_EVENT_CLASS(xfs_perag_class, - TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, int refcount, - unsigned long caller_ip), - TP_ARGS(mp, agno, refcount, caller_ip), + TP_PROTO(struct xfs_perag *pag, unsigned long caller_ip), + TP_ARGS(pag, caller_ip), TP_STRUCT__entry( __field(dev_t, dev) __field(xfs_agnumber_t, agno) __field(int, refcount) + __field(int, active_refcount) __field(unsigned long, caller_ip) ), TP_fast_assign( - __entry->dev = mp->m_super->s_dev; - __entry->agno = agno; - __entry->refcount = refcount; + __entry->dev = pag->pag_mount->m_super->s_dev; + __entry->agno = pag->pag_agno; + __entry->refcount = atomic_read(&pag->pag_ref); + __entry->active_refcount = atomic_read(&pag->pag_active_ref); __entry->caller_ip = caller_ip; ), - TP_printk("dev %d:%d agno 0x%x refcount %d caller %pS", + TP_printk("dev %d:%d agno 0x%x passive refs %d active refs %d caller %pS", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->agno, __entry->refcount, + __entry->active_refcount, (char *)__entry->caller_ip) ); #define DEFINE_PERAG_REF_EVENT(name) \ DEFINE_EVENT(xfs_perag_class, name, \ - TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, int refcount, \ - unsigned long caller_ip), \ - TP_ARGS(mp, agno, refcount, caller_ip)) + TP_PROTO(struct xfs_perag *pag, unsigned long caller_ip), \ + TP_ARGS(pag, caller_ip)) DEFINE_PERAG_REF_EVENT(xfs_perag_get); DEFINE_PERAG_REF_EVENT(xfs_perag_get_tag); DEFINE_PERAG_REF_EVENT(xfs_perag_put); From 498f0adbcdb6a68403bfb9645a7555b789a7fee4 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:52 +1100 Subject: [PATCH 219/765] xfs: convert xfs_imap() to take a perag Callers have referenced perags but they don't pass it into xfs_imap() so it takes it's own reference. Fix that so we can change inode allocation over to using active references. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_ialloc.c | 45 ++++++++++++++------------------------ fs/xfs/libxfs/xfs_ialloc.h | 3 ++- fs/xfs/scrub/common.c | 13 +++++++---- fs/xfs/xfs_icache.c | 2 +- 4 files changed, 28 insertions(+), 35 deletions(-) diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index e8068422aa21f..f120a48813cc9 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -2217,15 +2217,15 @@ xfs_difree( STATIC int xfs_imap_lookup( - struct xfs_mount *mp, - struct xfs_trans *tp, struct xfs_perag *pag, + struct xfs_trans *tp, xfs_agino_t agino, xfs_agblock_t agbno, xfs_agblock_t *chunk_agbno, xfs_agblock_t *offset_agbno, int flags) { + struct xfs_mount *mp = pag->pag_mount; struct xfs_inobt_rec_incore rec; struct xfs_btree_cur *cur; struct xfs_buf *agbp; @@ -2280,12 +2280,13 @@ xfs_imap_lookup( */ int xfs_imap( - struct xfs_mount *mp, /* file system mount structure */ - struct xfs_trans *tp, /* transaction pointer */ + struct xfs_perag *pag, + struct xfs_trans *tp, xfs_ino_t ino, /* inode to locate */ struct xfs_imap *imap, /* location map structure */ uint flags) /* flags for inode btree lookup */ { + struct xfs_mount *mp = pag->pag_mount; xfs_agblock_t agbno; /* block number of inode in the alloc group */ xfs_agino_t agino; /* inode number within alloc group */ xfs_agblock_t chunk_agbno; /* first block in inode chunk */ @@ -2293,17 +2294,15 @@ xfs_imap( int error; /* error code */ int offset; /* index of inode in its buffer */ xfs_agblock_t offset_agbno; /* blks from chunk start to inode */ - struct xfs_perag *pag; ASSERT(ino != NULLFSINO); /* * Split up the inode number into its parts. */ - pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ino)); agino = XFS_INO_TO_AGINO(mp, ino); agbno = XFS_AGINO_TO_AGBNO(mp, agino); - if (!pag || agbno >= mp->m_sb.sb_agblocks || + if (agbno >= mp->m_sb.sb_agblocks || ino != XFS_AGINO_TO_INO(mp, pag->pag_agno, agino)) { error = -EINVAL; #ifdef DEBUG @@ -2312,20 +2311,14 @@ xfs_imap( * as they can be invalid without implying corruption. */ if (flags & XFS_IGET_UNTRUSTED) - goto out_drop; - if (!pag) { - xfs_alert(mp, - "%s: agno (%d) >= mp->m_sb.sb_agcount (%d)", - __func__, XFS_INO_TO_AGNO(mp, ino), - mp->m_sb.sb_agcount); - } + return error; if (agbno >= mp->m_sb.sb_agblocks) { xfs_alert(mp, "%s: agbno (0x%llx) >= mp->m_sb.sb_agblocks (0x%lx)", __func__, (unsigned long long)agbno, (unsigned long)mp->m_sb.sb_agblocks); } - if (pag && ino != XFS_AGINO_TO_INO(mp, pag->pag_agno, agino)) { + if (ino != XFS_AGINO_TO_INO(mp, pag->pag_agno, agino)) { xfs_alert(mp, "%s: ino (0x%llx) != XFS_AGINO_TO_INO() (0x%llx)", __func__, ino, @@ -2333,7 +2326,7 @@ xfs_imap( } xfs_stack_trace(); #endif /* DEBUG */ - goto out_drop; + return error; } /* @@ -2344,10 +2337,10 @@ xfs_imap( * in all cases where an untrusted inode number is passed. */ if (flags & XFS_IGET_UNTRUSTED) { - error = xfs_imap_lookup(mp, tp, pag, agino, agbno, + error = xfs_imap_lookup(pag, tp, agino, agbno, &chunk_agbno, &offset_agbno, flags); if (error) - goto out_drop; + return error; goto out_map; } @@ -2363,8 +2356,7 @@ xfs_imap( imap->im_len = XFS_FSB_TO_BB(mp, 1); imap->im_boffset = (unsigned short)(offset << mp->m_sb.sb_inodelog); - error = 0; - goto out_drop; + return 0; } /* @@ -2376,10 +2368,10 @@ xfs_imap( offset_agbno = agbno & M_IGEO(mp)->inoalign_mask; chunk_agbno = agbno - offset_agbno; } else { - error = xfs_imap_lookup(mp, tp, pag, agino, agbno, + error = xfs_imap_lookup(pag, tp, agino, agbno, &chunk_agbno, &offset_agbno, flags); if (error) - goto out_drop; + return error; } out_map: @@ -2407,14 +2399,9 @@ xfs_imap( __func__, (unsigned long long) imap->im_blkno, (unsigned long long) imap->im_len, XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks)); - error = -EINVAL; - goto out_drop; + return -EINVAL; } - error = 0; -out_drop: - if (pag) - xfs_perag_put(pag); - return error; + return 0; } /* diff --git a/fs/xfs/libxfs/xfs_ialloc.h b/fs/xfs/libxfs/xfs_ialloc.h index 9bbbca6ac4edf..4cfce2eebe7e5 100644 --- a/fs/xfs/libxfs/xfs_ialloc.h +++ b/fs/xfs/libxfs/xfs_ialloc.h @@ -12,6 +12,7 @@ struct xfs_imap; struct xfs_mount; struct xfs_trans; struct xfs_btree_cur; +struct xfs_perag; /* Move inodes in clusters of this size */ #define XFS_INODE_BIG_CLUSTER_SIZE 8192 @@ -47,7 +48,7 @@ int xfs_difree(struct xfs_trans *tp, struct xfs_perag *pag, */ int xfs_imap( - struct xfs_mount *mp, /* file system mount structure */ + struct xfs_perag *pag, struct xfs_trans *tp, /* transaction pointer */ xfs_ino_t ino, /* inode to locate */ struct xfs_imap *imap, /* location map structure */ diff --git a/fs/xfs/scrub/common.c b/fs/xfs/scrub/common.c index 613260b04a3dc..033bf6730ece7 100644 --- a/fs/xfs/scrub/common.c +++ b/fs/xfs/scrub/common.c @@ -636,6 +636,7 @@ xchk_get_inode( { struct xfs_imap imap; struct xfs_mount *mp = sc->mp; + struct xfs_perag *pag; struct xfs_inode *ip_in = XFS_I(file_inode(sc->file)); struct xfs_inode *ip = NULL; int error; @@ -671,10 +672,14 @@ xchk_get_inode( * Otherwise, we really couldn't find it so tell userspace * that it no longer exists. */ - error = xfs_imap(sc->mp, sc->tp, sc->sm->sm_ino, &imap, - XFS_IGET_UNTRUSTED | XFS_IGET_DONTCACHE); - if (error) - return -ENOENT; + pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, sc->sm->sm_ino)); + if (pag) { + error = xfs_imap(pag, sc->tp, sc->sm->sm_ino, &imap, + XFS_IGET_UNTRUSTED | XFS_IGET_DONTCACHE); + xfs_perag_put(pag); + if (error) + return -ENOENT; + } error = -EFSCORRUPTED; fallthrough; default: diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c index 8b2823d85a68c..c9a7e270a4286 100644 --- a/fs/xfs/xfs_icache.c +++ b/fs/xfs/xfs_icache.c @@ -586,7 +586,7 @@ xfs_iget_cache_miss( if (!ip) return -ENOMEM; - error = xfs_imap(mp, tp, ip->i_ino, &ip->i_imap, flags); + error = xfs_imap(pag, tp, ip->i_ino, &ip->i_imap, flags); if (error) goto out_destroy; From dedab3e4379d298ed60b6c52a15168807b48d57a Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:52 +1100 Subject: [PATCH 220/765] xfs: use active perag references for inode allocation Convert the inode allocation routines to use active perag references or references held by callers rather than grab their own. Also drive the perag further inwards to replace xfs_mounts when doing operations on a specific AG. Signed-off-by: Dave Chinner Reviewed-by: Allison Henderson Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_ag.c | 3 +- fs/xfs/libxfs/xfs_ialloc.c | 63 +++++++++++++++++++------------------- fs/xfs/libxfs/xfs_ialloc.h | 2 +- 3 files changed, 33 insertions(+), 35 deletions(-) diff --git a/fs/xfs/libxfs/xfs_ag.c b/fs/xfs/libxfs/xfs_ag.c index 7cff618753402..a3bdcde958459 100644 --- a/fs/xfs/libxfs/xfs_ag.c +++ b/fs/xfs/libxfs/xfs_ag.c @@ -925,8 +925,7 @@ xfs_ag_shrink_space( * Make sure that the last inode cluster cannot overlap with the new * end of the AG, even if it's sparse. */ - error = xfs_ialloc_check_shrink(*tpp, pag->pag_agno, agibp, - aglen - delta); + error = xfs_ialloc_check_shrink(pag, *tpp, agibp, aglen - delta); if (error) return error; diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index f120a48813cc9..53d6b7766609c 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -169,14 +169,14 @@ xfs_inobt_insert_rec( */ STATIC int xfs_inobt_insert( - struct xfs_mount *mp, + struct xfs_perag *pag, struct xfs_trans *tp, struct xfs_buf *agbp, - struct xfs_perag *pag, xfs_agino_t newino, xfs_agino_t newlen, xfs_btnum_t btnum) { + struct xfs_mount *mp = pag->pag_mount; struct xfs_btree_cur *cur; xfs_agino_t thisino; int i; @@ -514,14 +514,14 @@ __xfs_inobt_rec_merge( */ STATIC int xfs_inobt_insert_sprec( - struct xfs_mount *mp, + struct xfs_perag *pag, struct xfs_trans *tp, struct xfs_buf *agbp, - struct xfs_perag *pag, int btnum, struct xfs_inobt_rec_incore *nrec, /* in/out: new/merged rec. */ bool merge) /* merge or replace */ { + struct xfs_mount *mp = pag->pag_mount; struct xfs_btree_cur *cur; int error; int i; @@ -609,9 +609,9 @@ xfs_inobt_insert_sprec( */ STATIC int xfs_ialloc_ag_alloc( + struct xfs_perag *pag, struct xfs_trans *tp, - struct xfs_buf *agbp, - struct xfs_perag *pag) + struct xfs_buf *agbp) { struct xfs_agi *agi; struct xfs_alloc_arg args; @@ -831,7 +831,7 @@ xfs_ialloc_ag_alloc( * if necessary. If a merge does occur, rec is updated to the * merged record. */ - error = xfs_inobt_insert_sprec(args.mp, tp, agbp, pag, + error = xfs_inobt_insert_sprec(pag, tp, agbp, XFS_BTNUM_INO, &rec, true); if (error == -EFSCORRUPTED) { xfs_alert(args.mp, @@ -856,20 +856,20 @@ xfs_ialloc_ag_alloc( * existing record with this one. */ if (xfs_has_finobt(args.mp)) { - error = xfs_inobt_insert_sprec(args.mp, tp, agbp, pag, + error = xfs_inobt_insert_sprec(pag, tp, agbp, XFS_BTNUM_FINO, &rec, false); if (error) return error; } } else { /* full chunk - insert new records to both btrees */ - error = xfs_inobt_insert(args.mp, tp, agbp, pag, newino, newlen, + error = xfs_inobt_insert(pag, tp, agbp, newino, newlen, XFS_BTNUM_INO); if (error) return error; if (xfs_has_finobt(args.mp)) { - error = xfs_inobt_insert(args.mp, tp, agbp, pag, newino, + error = xfs_inobt_insert(pag, tp, agbp, newino, newlen, XFS_BTNUM_FINO); if (error) return error; @@ -981,9 +981,9 @@ xfs_inobt_first_free_inode( */ STATIC int xfs_dialloc_ag_inobt( + struct xfs_perag *pag, struct xfs_trans *tp, struct xfs_buf *agbp, - struct xfs_perag *pag, xfs_ino_t parent, xfs_ino_t *inop) { @@ -1429,9 +1429,9 @@ xfs_dialloc_ag_update_inobt( */ static int xfs_dialloc_ag( + struct xfs_perag *pag, struct xfs_trans *tp, struct xfs_buf *agbp, - struct xfs_perag *pag, xfs_ino_t parent, xfs_ino_t *inop) { @@ -1448,7 +1448,7 @@ xfs_dialloc_ag( int i; if (!xfs_has_finobt(mp)) - return xfs_dialloc_ag_inobt(tp, agbp, pag, parent, inop); + return xfs_dialloc_ag_inobt(pag, tp, agbp, parent, inop); /* * If pagino is 0 (this is the root inode allocation) use newino. @@ -1594,8 +1594,8 @@ xfs_ialloc_next_ag( static bool xfs_dialloc_good_ag( - struct xfs_trans *tp, struct xfs_perag *pag, + struct xfs_trans *tp, umode_t mode, int flags, bool ok_alloc) @@ -1606,6 +1606,8 @@ xfs_dialloc_good_ag( int needspace; int error; + if (!pag) + return false; if (!pag->pagi_inodeok) return false; @@ -1665,8 +1667,8 @@ xfs_dialloc_good_ag( static int xfs_dialloc_try_ag( - struct xfs_trans **tpp, struct xfs_perag *pag, + struct xfs_trans **tpp, xfs_ino_t parent, xfs_ino_t *new_ino, bool ok_alloc) @@ -1689,7 +1691,7 @@ xfs_dialloc_try_ag( goto out_release; } - error = xfs_ialloc_ag_alloc(*tpp, agbp, pag); + error = xfs_ialloc_ag_alloc(pag, *tpp, agbp); if (error < 0) goto out_release; @@ -1705,7 +1707,7 @@ xfs_dialloc_try_ag( } /* Allocate an inode in the found AG */ - error = xfs_dialloc_ag(*tpp, agbp, pag, parent, &ino); + error = xfs_dialloc_ag(pag, *tpp, agbp, parent, &ino); if (!error) *new_ino = ino; return error; @@ -1790,9 +1792,9 @@ xfs_dialloc( agno = start_agno; flags = XFS_ALLOC_FLAG_TRYLOCK; for (;;) { - pag = xfs_perag_get(mp, agno); - if (xfs_dialloc_good_ag(*tpp, pag, mode, flags, ok_alloc)) { - error = xfs_dialloc_try_ag(tpp, pag, parent, + pag = xfs_perag_grab(mp, agno); + if (xfs_dialloc_good_ag(pag, *tpp, mode, flags, ok_alloc)) { + error = xfs_dialloc_try_ag(pag, tpp, parent, &ino, ok_alloc); if (error != -EAGAIN) break; @@ -1813,12 +1815,12 @@ xfs_dialloc( if (low_space) ok_alloc = true; } - xfs_perag_put(pag); + xfs_perag_rele(pag); } if (!error) *new_ino = ino; - xfs_perag_put(pag); + xfs_perag_rele(pag); return error; } @@ -1902,14 +1904,14 @@ xfs_difree_inode_chunk( STATIC int xfs_difree_inobt( - struct xfs_mount *mp, + struct xfs_perag *pag, struct xfs_trans *tp, struct xfs_buf *agbp, - struct xfs_perag *pag, xfs_agino_t agino, struct xfs_icluster *xic, struct xfs_inobt_rec_incore *orec) { + struct xfs_mount *mp = pag->pag_mount; struct xfs_agi *agi = agbp->b_addr; struct xfs_btree_cur *cur; struct xfs_inobt_rec_incore rec; @@ -2036,13 +2038,13 @@ xfs_difree_inobt( */ STATIC int xfs_difree_finobt( - struct xfs_mount *mp, + struct xfs_perag *pag, struct xfs_trans *tp, struct xfs_buf *agbp, - struct xfs_perag *pag, xfs_agino_t agino, struct xfs_inobt_rec_incore *ibtrec) /* inobt record */ { + struct xfs_mount *mp = pag->pag_mount; struct xfs_btree_cur *cur; struct xfs_inobt_rec_incore rec; int offset = agino - ibtrec->ir_startino; @@ -2196,7 +2198,7 @@ xfs_difree( /* * Fix up the inode allocation btree. */ - error = xfs_difree_inobt(mp, tp, agbp, pag, agino, xic, &rec); + error = xfs_difree_inobt(pag, tp, agbp, agino, xic, &rec); if (error) goto error0; @@ -2204,7 +2206,7 @@ xfs_difree( * Fix up the free inode btree. */ if (xfs_has_finobt(mp)) { - error = xfs_difree_finobt(mp, tp, agbp, pag, agino, &rec); + error = xfs_difree_finobt(pag, tp, agbp, agino, &rec); if (error) goto error0; } @@ -2928,15 +2930,14 @@ xfs_ialloc_calc_rootino( */ int xfs_ialloc_check_shrink( + struct xfs_perag *pag, struct xfs_trans *tp, - xfs_agnumber_t agno, struct xfs_buf *agibp, xfs_agblock_t new_length) { struct xfs_inobt_rec_incore rec; struct xfs_btree_cur *cur; struct xfs_mount *mp = tp->t_mountp; - struct xfs_perag *pag; xfs_agino_t agino = XFS_AGB_TO_AGINO(mp, new_length); int has; int error; @@ -2944,7 +2945,6 @@ xfs_ialloc_check_shrink( if (!xfs_has_sparseinodes(mp)) return 0; - pag = xfs_perag_get(mp, agno); cur = xfs_inobt_init_cursor(mp, tp, agibp, pag, XFS_BTNUM_INO); /* Look up the inobt record that would correspond to the new EOFS. */ @@ -2968,6 +2968,5 @@ xfs_ialloc_check_shrink( } out: xfs_btree_del_cursor(cur, error); - xfs_perag_put(pag); return error; } diff --git a/fs/xfs/libxfs/xfs_ialloc.h b/fs/xfs/libxfs/xfs_ialloc.h index 4cfce2eebe7e5..ab8c30b4ec22c 100644 --- a/fs/xfs/libxfs/xfs_ialloc.h +++ b/fs/xfs/libxfs/xfs_ialloc.h @@ -107,7 +107,7 @@ int xfs_ialloc_cluster_alignment(struct xfs_mount *mp); void xfs_ialloc_setup_geometry(struct xfs_mount *mp); xfs_ino_t xfs_ialloc_calc_rootino(struct xfs_mount *mp, int sunit); -int xfs_ialloc_check_shrink(struct xfs_trans *tp, xfs_agnumber_t agno, +int xfs_ialloc_check_shrink(struct xfs_perag *pag, struct xfs_trans *tp, struct xfs_buf *agibp, xfs_agblock_t new_length); #endif /* __XFS_IALLOC_H__ */ From bab8b795185bf37801a4f7ee5c321eee288c2f10 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:52 +1100 Subject: [PATCH 221/765] xfs: inobt can use perags in many more places than it does Lots of code in the inobt infrastructure is passed both xfs_mount and perags. We only need perags for the per-ag inode allocation code, so reduce the duplication by passing only the perags as the primary object. This ends up reducing the code size by a bit: text data bss dec hex filename orig 1138878 323979 548 1463405 16546d (TOTALS) patched 1138709 323979 548 1463236 1653c4 (TOTALS) Signed-off-by: Dave Chinner Reviewed-by: Allison Henderson Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_ag_resv.c | 2 +- fs/xfs/libxfs/xfs_ialloc.c | 25 +++++++++++---------- fs/xfs/libxfs/xfs_ialloc_btree.c | 37 ++++++++++++++------------------ fs/xfs/libxfs/xfs_ialloc_btree.h | 20 ++++++++--------- fs/xfs/scrub/agheader_repair.c | 7 +++--- fs/xfs/scrub/common.c | 8 +++---- fs/xfs/xfs_iwalk.c | 4 ++-- 7 files changed, 47 insertions(+), 56 deletions(-) diff --git a/fs/xfs/libxfs/xfs_ag_resv.c b/fs/xfs/libxfs/xfs_ag_resv.c index 5af123d13a637..7fd1fea95552f 100644 --- a/fs/xfs/libxfs/xfs_ag_resv.c +++ b/fs/xfs/libxfs/xfs_ag_resv.c @@ -264,7 +264,7 @@ xfs_ag_resv_init( if (error) goto out; - error = xfs_finobt_calc_reserves(mp, tp, pag, &ask, &used); + error = xfs_finobt_calc_reserves(pag, tp, &ask, &used); if (error) goto out; diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index 53d6b7766609c..c8b31c06a95de 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -176,13 +176,12 @@ xfs_inobt_insert( xfs_agino_t newlen, xfs_btnum_t btnum) { - struct xfs_mount *mp = pag->pag_mount; struct xfs_btree_cur *cur; xfs_agino_t thisino; int i; int error; - cur = xfs_inobt_init_cursor(mp, tp, agbp, pag, btnum); + cur = xfs_inobt_init_cursor(pag, tp, agbp, btnum); for (thisino = newino; thisino < newino + newlen; @@ -527,7 +526,7 @@ xfs_inobt_insert_sprec( int i; struct xfs_inobt_rec_incore rec; - cur = xfs_inobt_init_cursor(mp, tp, agbp, pag, btnum); + cur = xfs_inobt_init_cursor(pag, tp, agbp, btnum); /* the new record is pre-aligned so we know where to look */ error = xfs_inobt_lookup(cur, nrec->ir_startino, XFS_LOOKUP_EQ, &i); @@ -1004,7 +1003,7 @@ xfs_dialloc_ag_inobt( ASSERT(pag->pagi_freecount > 0); restart_pagno: - cur = xfs_inobt_init_cursor(mp, tp, agbp, pag, XFS_BTNUM_INO); + cur = xfs_inobt_init_cursor(pag, tp, agbp, XFS_BTNUM_INO); /* * If pagino is 0 (this is the root inode allocation) use newino. * This must work because we've just allocated some. @@ -1457,7 +1456,7 @@ xfs_dialloc_ag( if (!pagino) pagino = be32_to_cpu(agi->agi_newino); - cur = xfs_inobt_init_cursor(mp, tp, agbp, pag, XFS_BTNUM_FINO); + cur = xfs_inobt_init_cursor(pag, tp, agbp, XFS_BTNUM_FINO); error = xfs_check_agi_freecount(cur); if (error) @@ -1500,7 +1499,7 @@ xfs_dialloc_ag( * the original freecount. If all is well, make the equivalent update to * the inobt using the finobt record and offset information. */ - icur = xfs_inobt_init_cursor(mp, tp, agbp, pag, XFS_BTNUM_INO); + icur = xfs_inobt_init_cursor(pag, tp, agbp, XFS_BTNUM_INO); error = xfs_check_agi_freecount(icur); if (error) @@ -1926,7 +1925,7 @@ xfs_difree_inobt( /* * Initialize the cursor. */ - cur = xfs_inobt_init_cursor(mp, tp, agbp, pag, XFS_BTNUM_INO); + cur = xfs_inobt_init_cursor(pag, tp, agbp, XFS_BTNUM_INO); error = xfs_check_agi_freecount(cur); if (error) @@ -2051,7 +2050,7 @@ xfs_difree_finobt( int error; int i; - cur = xfs_inobt_init_cursor(mp, tp, agbp, pag, XFS_BTNUM_FINO); + cur = xfs_inobt_init_cursor(pag, tp, agbp, XFS_BTNUM_FINO); error = xfs_inobt_lookup(cur, ibtrec->ir_startino, XFS_LOOKUP_EQ, &i); if (error) @@ -2248,7 +2247,7 @@ xfs_imap_lookup( * we have a record, we need to ensure it contains the inode number * we are looking up. */ - cur = xfs_inobt_init_cursor(mp, tp, agbp, pag, XFS_BTNUM_INO); + cur = xfs_inobt_init_cursor(pag, tp, agbp, XFS_BTNUM_INO); error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE, &i); if (!error) { if (i) @@ -2937,17 +2936,17 @@ xfs_ialloc_check_shrink( { struct xfs_inobt_rec_incore rec; struct xfs_btree_cur *cur; - struct xfs_mount *mp = tp->t_mountp; - xfs_agino_t agino = XFS_AGB_TO_AGINO(mp, new_length); + xfs_agino_t agino; int has; int error; - if (!xfs_has_sparseinodes(mp)) + if (!xfs_has_sparseinodes(pag->pag_mount)) return 0; - cur = xfs_inobt_init_cursor(mp, tp, agibp, pag, XFS_BTNUM_INO); + cur = xfs_inobt_init_cursor(pag, tp, agibp, XFS_BTNUM_INO); /* Look up the inobt record that would correspond to the new EOFS. */ + agino = XFS_AGB_TO_AGINO(pag->pag_mount, new_length); error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE, &has); if (error || !has) goto out; diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c index 8c83e265770c1..d657af2ec3503 100644 --- a/fs/xfs/libxfs/xfs_ialloc_btree.c +++ b/fs/xfs/libxfs/xfs_ialloc_btree.c @@ -36,8 +36,8 @@ STATIC struct xfs_btree_cur * xfs_inobt_dup_cursor( struct xfs_btree_cur *cur) { - return xfs_inobt_init_cursor(cur->bc_mp, cur->bc_tp, - cur->bc_ag.agbp, cur->bc_ag.pag, cur->bc_btnum); + return xfs_inobt_init_cursor(cur->bc_ag.pag, cur->bc_tp, + cur->bc_ag.agbp, cur->bc_btnum); } STATIC void @@ -427,11 +427,11 @@ static const struct xfs_btree_ops xfs_finobt_ops = { */ static struct xfs_btree_cur * xfs_inobt_init_common( - struct xfs_mount *mp, /* file system mount point */ - struct xfs_trans *tp, /* transaction pointer */ struct xfs_perag *pag, + struct xfs_trans *tp, /* transaction pointer */ xfs_btnum_t btnum) /* ialloc or free ino btree */ { + struct xfs_mount *mp = pag->pag_mount; struct xfs_btree_cur *cur; cur = xfs_btree_alloc_cursor(mp, tp, btnum, @@ -456,16 +456,15 @@ xfs_inobt_init_common( /* Create an inode btree cursor. */ struct xfs_btree_cur * xfs_inobt_init_cursor( - struct xfs_mount *mp, + struct xfs_perag *pag, struct xfs_trans *tp, struct xfs_buf *agbp, - struct xfs_perag *pag, xfs_btnum_t btnum) { struct xfs_btree_cur *cur; struct xfs_agi *agi = agbp->b_addr; - cur = xfs_inobt_init_common(mp, tp, pag, btnum); + cur = xfs_inobt_init_common(pag, tp, btnum); if (btnum == XFS_BTNUM_INO) cur->bc_nlevels = be32_to_cpu(agi->agi_level); else @@ -477,14 +476,13 @@ xfs_inobt_init_cursor( /* Create an inode btree cursor with a fake root for staging. */ struct xfs_btree_cur * xfs_inobt_stage_cursor( - struct xfs_mount *mp, - struct xbtree_afakeroot *afake, struct xfs_perag *pag, + struct xbtree_afakeroot *afake, xfs_btnum_t btnum) { struct xfs_btree_cur *cur; - cur = xfs_inobt_init_common(mp, NULL, pag, btnum); + cur = xfs_inobt_init_common(pag, NULL, btnum); xfs_btree_stage_afakeroot(cur, afake); return cur; } @@ -708,9 +706,8 @@ xfs_inobt_max_size( /* Read AGI and create inobt cursor. */ int xfs_inobt_cur( - struct xfs_mount *mp, - struct xfs_trans *tp, struct xfs_perag *pag, + struct xfs_trans *tp, xfs_btnum_t which, struct xfs_btree_cur **curpp, struct xfs_buf **agi_bpp) @@ -725,16 +722,15 @@ xfs_inobt_cur( if (error) return error; - cur = xfs_inobt_init_cursor(mp, tp, *agi_bpp, pag, which); + cur = xfs_inobt_init_cursor(pag, tp, *agi_bpp, which); *curpp = cur; return 0; } static int xfs_inobt_count_blocks( - struct xfs_mount *mp, - struct xfs_trans *tp, struct xfs_perag *pag, + struct xfs_trans *tp, xfs_btnum_t btnum, xfs_extlen_t *tree_blocks) { @@ -742,7 +738,7 @@ xfs_inobt_count_blocks( struct xfs_btree_cur *cur = NULL; int error; - error = xfs_inobt_cur(mp, tp, pag, btnum, &cur, &agbp); + error = xfs_inobt_cur(pag, tp, btnum, &cur, &agbp); if (error) return error; @@ -779,22 +775,21 @@ xfs_finobt_read_blocks( */ int xfs_finobt_calc_reserves( - struct xfs_mount *mp, - struct xfs_trans *tp, struct xfs_perag *pag, + struct xfs_trans *tp, xfs_extlen_t *ask, xfs_extlen_t *used) { xfs_extlen_t tree_len = 0; int error; - if (!xfs_has_finobt(mp)) + if (!xfs_has_finobt(pag->pag_mount)) return 0; - if (xfs_has_inobtcounts(mp)) + if (xfs_has_inobtcounts(pag->pag_mount)) error = xfs_finobt_read_blocks(pag, tp, &tree_len); else - error = xfs_inobt_count_blocks(mp, tp, pag, XFS_BTNUM_FINO, + error = xfs_inobt_count_blocks(pag, tp, XFS_BTNUM_FINO, &tree_len); if (error) return error; diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.h b/fs/xfs/libxfs/xfs_ialloc_btree.h index 26451cb76b98b..e859a6e052309 100644 --- a/fs/xfs/libxfs/xfs_ialloc_btree.h +++ b/fs/xfs/libxfs/xfs_ialloc_btree.h @@ -46,12 +46,10 @@ struct xfs_perag; (maxrecs) * sizeof(xfs_inobt_key_t) + \ ((index) - 1) * sizeof(xfs_inobt_ptr_t))) -extern struct xfs_btree_cur *xfs_inobt_init_cursor(struct xfs_mount *mp, - struct xfs_trans *tp, struct xfs_buf *agbp, - struct xfs_perag *pag, xfs_btnum_t btnum); -struct xfs_btree_cur *xfs_inobt_stage_cursor(struct xfs_mount *mp, - struct xbtree_afakeroot *afake, struct xfs_perag *pag, - xfs_btnum_t btnum); +extern struct xfs_btree_cur *xfs_inobt_init_cursor(struct xfs_perag *pag, + struct xfs_trans *tp, struct xfs_buf *agbp, xfs_btnum_t btnum); +struct xfs_btree_cur *xfs_inobt_stage_cursor(struct xfs_perag *pag, + struct xbtree_afakeroot *afake, xfs_btnum_t btnum); extern int xfs_inobt_maxrecs(struct xfs_mount *, int, int); /* ir_holemask to inode allocation bitmap conversion */ @@ -64,13 +62,13 @@ int xfs_inobt_rec_check_count(struct xfs_mount *, #define xfs_inobt_rec_check_count(mp, rec) 0 #endif /* DEBUG */ -int xfs_finobt_calc_reserves(struct xfs_mount *mp, struct xfs_trans *tp, - struct xfs_perag *pag, xfs_extlen_t *ask, xfs_extlen_t *used); +int xfs_finobt_calc_reserves(struct xfs_perag *perag, struct xfs_trans *tp, + xfs_extlen_t *ask, xfs_extlen_t *used); extern xfs_extlen_t xfs_iallocbt_calc_size(struct xfs_mount *mp, unsigned long long len); -int xfs_inobt_cur(struct xfs_mount *mp, struct xfs_trans *tp, - struct xfs_perag *pag, xfs_btnum_t btnum, - struct xfs_btree_cur **curpp, struct xfs_buf **agi_bpp); +int xfs_inobt_cur(struct xfs_perag *pag, struct xfs_trans *tp, + xfs_btnum_t btnum, struct xfs_btree_cur **curpp, + struct xfs_buf **agi_bpp); void xfs_inobt_commit_staged_btree(struct xfs_btree_cur *cur, struct xfs_trans *tp, struct xfs_buf *agbp); diff --git a/fs/xfs/scrub/agheader_repair.c b/fs/xfs/scrub/agheader_repair.c index d75d82151eeba..b80b9111e781a 100644 --- a/fs/xfs/scrub/agheader_repair.c +++ b/fs/xfs/scrub/agheader_repair.c @@ -873,8 +873,7 @@ xrep_agi_calc_from_btrees( xfs_agino_t freecount; int error; - cur = xfs_inobt_init_cursor(mp, sc->tp, agi_bp, - sc->sa.pag, XFS_BTNUM_INO); + cur = xfs_inobt_init_cursor(sc->sa.pag, sc->tp, agi_bp, XFS_BTNUM_INO); error = xfs_ialloc_count_inodes(cur, &count, &freecount); if (error) goto err; @@ -894,8 +893,8 @@ xrep_agi_calc_from_btrees( if (xfs_has_finobt(mp) && xfs_has_inobtcounts(mp)) { xfs_agblock_t blocks; - cur = xfs_inobt_init_cursor(mp, sc->tp, agi_bp, - sc->sa.pag, XFS_BTNUM_FINO); + cur = xfs_inobt_init_cursor(sc->sa.pag, sc->tp, agi_bp, + XFS_BTNUM_FINO); error = xfs_btree_count_blocks(cur, &blocks); if (error) goto err; diff --git a/fs/xfs/scrub/common.c b/fs/xfs/scrub/common.c index 033bf6730ece7..848a8e32e56f0 100644 --- a/fs/xfs/scrub/common.c +++ b/fs/xfs/scrub/common.c @@ -478,15 +478,15 @@ xchk_ag_btcur_init( /* Set up a inobt cursor for cross-referencing. */ if (sa->agi_bp && xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_INO)) { - sa->ino_cur = xfs_inobt_init_cursor(mp, sc->tp, sa->agi_bp, - sa->pag, XFS_BTNUM_INO); + sa->ino_cur = xfs_inobt_init_cursor(sa->pag, sc->tp, sa->agi_bp, + XFS_BTNUM_INO); } /* Set up a finobt cursor for cross-referencing. */ if (sa->agi_bp && xfs_has_finobt(mp) && xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_FINO)) { - sa->fino_cur = xfs_inobt_init_cursor(mp, sc->tp, sa->agi_bp, - sa->pag, XFS_BTNUM_FINO); + sa->fino_cur = xfs_inobt_init_cursor(sa->pag, sc->tp, sa->agi_bp, + XFS_BTNUM_FINO); } /* Set up a rmapbt cursor for cross-referencing. */ diff --git a/fs/xfs/xfs_iwalk.c b/fs/xfs/xfs_iwalk.c index c31857d903a45..21be93bf006dd 100644 --- a/fs/xfs/xfs_iwalk.c +++ b/fs/xfs/xfs_iwalk.c @@ -275,7 +275,7 @@ xfs_iwalk_ag_start( /* Set up a fresh cursor and empty the inobt cache. */ iwag->nr_recs = 0; - error = xfs_inobt_cur(mp, tp, pag, XFS_BTNUM_INO, curpp, agi_bpp); + error = xfs_inobt_cur(pag, tp, XFS_BTNUM_INO, curpp, agi_bpp); if (error) return error; @@ -390,7 +390,7 @@ xfs_iwalk_run_callbacks( } /* ...and recreate the cursor just past where we left off. */ - error = xfs_inobt_cur(mp, iwag->tp, iwag->pag, XFS_BTNUM_INO, curpp, + error = xfs_inobt_cur(iwag->pag, iwag->tp, XFS_BTNUM_INO, curpp, agi_bpp); if (error) return error; From 20a5eab49d354a2837e0af3f07f92a104de52804 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:52 +1100 Subject: [PATCH 222/765] xfs: convert xfs_ialloc_next_ag() to an atomic This is currently a spinlock lock protected rotor which can be implemented with a single atomic operation. Change it to be more efficient and get rid of the m_agirotor_lock. Noticed while converting the inode allocation AG selection loop to active perag references. Signed-off-by: Dave Chinner Reviewed-by: Allison Henderson Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_ialloc.c | 17 +---------------- fs/xfs/libxfs/xfs_sb.c | 3 ++- fs/xfs/xfs_mount.h | 3 +-- fs/xfs/xfs_super.c | 1 - 4 files changed, 4 insertions(+), 20 deletions(-) diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index c8b31c06a95de..86b7adc355e6e 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -1576,21 +1576,6 @@ xfs_dialloc_roll( return error; } -static xfs_agnumber_t -xfs_ialloc_next_ag( - xfs_mount_t *mp) -{ - xfs_agnumber_t agno; - - spin_lock(&mp->m_agirotor_lock); - agno = mp->m_agirotor; - if (++mp->m_agirotor >= mp->m_maxagi) - mp->m_agirotor = 0; - spin_unlock(&mp->m_agirotor_lock); - - return agno; -} - static bool xfs_dialloc_good_ag( struct xfs_perag *pag, @@ -1748,7 +1733,7 @@ xfs_dialloc( * an AG has enough space for file creation. */ if (S_ISDIR(mode)) - start_agno = xfs_ialloc_next_ag(mp); + start_agno = atomic_inc_return(&mp->m_agirotor) % mp->m_maxagi; else { start_agno = XFS_INO_TO_AGNO(mp, parent); if (start_agno >= mp->m_maxagi) diff --git a/fs/xfs/libxfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c index 1eeecf2eb2a77..99cc03a298e21 100644 --- a/fs/xfs/libxfs/xfs_sb.c +++ b/fs/xfs/libxfs/xfs_sb.c @@ -909,7 +909,8 @@ xfs_sb_mount_common( struct xfs_mount *mp, struct xfs_sb *sbp) { - mp->m_agfrotor = mp->m_agirotor = 0; + mp->m_agfrotor = 0; + atomic_set(&mp->m_agirotor, 0); mp->m_maxagi = mp->m_sb.sb_agcount; mp->m_blkbit_log = sbp->sb_blocklog + XFS_NBBYLOG; mp->m_blkbb_log = sbp->sb_blocklog - BBSHIFT; diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 8aca2cc173ac1..f3269c0626f05 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -210,8 +210,7 @@ typedef struct xfs_mount { struct xfs_error_cfg m_error_cfg[XFS_ERR_CLASS_MAX][XFS_ERR_ERRNO_MAX]; struct xstats m_stats; /* per-fs stats */ xfs_agnumber_t m_agfrotor; /* last ag where space found */ - xfs_agnumber_t m_agirotor; /* last ag dir inode alloced */ - spinlock_t m_agirotor_lock;/* .. and lock protecting it */ + atomic_t m_agirotor; /* last ag dir inode alloced */ /* Memory shrinker to throttle and reprioritize inodegc */ struct shrinker m_inodegc_shrinker; diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 0c4b73e9b29d2..96375b5622fd3 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -1922,7 +1922,6 @@ static int xfs_init_fs_context( return -ENOMEM; spin_lock_init(&mp->m_sb_lock); - spin_lock_init(&mp->m_agirotor_lock); INIT_RADIX_TREE(&mp->m_perag_tree, GFP_ATOMIC); spin_lock_init(&mp->m_perag_lock); mutex_init(&mp->m_growlock); From 7ac2ff8bb3713c7cb43564c04384af2ee7cc1f8d Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:52 +1100 Subject: [PATCH 223/765] xfs: perags need atomic operational state We currently don't have any flags or operational state in the xfs_perag except for the pagf_init and pagi_init flags. And the agflreset flag. Oh, there's also the pagf_metadata and pagi_inodeok flags, too. For controlling per-ag operations, we are going to need some atomic state flags. Hence add an opstate field similar to what we already have in the mount and log, and convert all these state flags across to atomic bit operations. Signed-off-by: Dave Chinner Reviewed-by: Allison Henderson Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_ag.h | 27 ++++++++++++++---- fs/xfs/libxfs/xfs_alloc.c | 23 ++++++++------- fs/xfs/libxfs/xfs_alloc_btree.c | 2 +- fs/xfs/libxfs/xfs_bmap.c | 2 +- fs/xfs/libxfs/xfs_ialloc.c | 14 ++++----- fs/xfs/libxfs/xfs_ialloc_btree.c | 4 +-- fs/xfs/libxfs/xfs_refcount_btree.c | 2 +- fs/xfs/libxfs/xfs_rmap_btree.c | 2 +- fs/xfs/scrub/agheader_repair.c | 28 +++++++++--------- fs/xfs/scrub/fscounters.c | 9 ++++-- fs/xfs/scrub/repair.c | 2 +- fs/xfs/xfs_filestream.c | 5 ++-- fs/xfs/xfs_super.c | 46 ++++++++++++++++++------------ 13 files changed, 101 insertions(+), 65 deletions(-) diff --git a/fs/xfs/libxfs/xfs_ag.h b/fs/xfs/libxfs/xfs_ag.h index aeb21c8df201e..187d30d9bb139 100644 --- a/fs/xfs/libxfs/xfs_ag.h +++ b/fs/xfs/libxfs/xfs_ag.h @@ -35,13 +35,9 @@ struct xfs_perag { atomic_t pag_ref; /* passive reference count */ atomic_t pag_active_ref; /* active reference count */ wait_queue_head_t pag_active_wq;/* woken active_ref falls to zero */ - char pagf_init; /* this agf's entry is initialized */ - char pagi_init; /* this agi's entry is initialized */ - char pagf_metadata; /* the agf is preferred to be metadata */ - char pagi_inodeok; /* The agi is ok for inodes */ + unsigned long pag_opstate; uint8_t pagf_levels[XFS_BTNUM_AGF]; /* # of levels in bno & cnt btree */ - bool pagf_agflreset; /* agfl requires reset before use */ uint32_t pagf_flcount; /* count of blocks in freelist */ xfs_extlen_t pagf_freeblks; /* total free blocks */ xfs_extlen_t pagf_longest; /* longest free space */ @@ -108,6 +104,27 @@ struct xfs_perag { #endif /* __KERNEL__ */ }; +/* + * Per-AG operational state. These are atomic flag bits. + */ +#define XFS_AGSTATE_AGF_INIT 0 +#define XFS_AGSTATE_AGI_INIT 1 +#define XFS_AGSTATE_PREFERS_METADATA 2 +#define XFS_AGSTATE_ALLOWS_INODES 3 +#define XFS_AGSTATE_AGFL_NEEDS_RESET 4 + +#define __XFS_AG_OPSTATE(name, NAME) \ +static inline bool xfs_perag_ ## name (struct xfs_perag *pag) \ +{ \ + return test_bit(XFS_AGSTATE_ ## NAME, &pag->pag_opstate); \ +} + +__XFS_AG_OPSTATE(initialised_agf, AGF_INIT) +__XFS_AG_OPSTATE(initialised_agi, AGI_INIT) +__XFS_AG_OPSTATE(prefers_metadata, PREFERS_METADATA) +__XFS_AG_OPSTATE(allows_inodes, ALLOWS_INODES) +__XFS_AG_OPSTATE(agfl_needs_reset, AGFL_NEEDS_RESET) + int xfs_initialize_perag(struct xfs_mount *mp, xfs_agnumber_t agcount, xfs_rfsblock_t dcount, xfs_agnumber_t *maxagi); int xfs_initialize_perag_data(struct xfs_mount *mp, xfs_agnumber_t agno); diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index 974b23abca60e..4508862239a88 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -2435,7 +2435,7 @@ xfs_agfl_reset( struct xfs_mount *mp = tp->t_mountp; struct xfs_agf *agf = agbp->b_addr; - ASSERT(pag->pagf_agflreset); + ASSERT(xfs_perag_agfl_needs_reset(pag)); trace_xfs_agfl_reset(mp, agf, 0, _RET_IP_); xfs_warn(mp, @@ -2450,7 +2450,7 @@ xfs_agfl_reset( XFS_AGF_FLCOUNT); pag->pagf_flcount = 0; - pag->pagf_agflreset = false; + clear_bit(XFS_AGSTATE_AGFL_NEEDS_RESET, &pag->pag_opstate); } /* @@ -2605,7 +2605,7 @@ xfs_alloc_fix_freelist( /* deferred ops (AGFL block frees) require permanent transactions */ ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES); - if (!pag->pagf_init) { + if (!xfs_perag_initialised_agf(pag)) { error = xfs_alloc_read_agf(pag, tp, flags, &agbp); if (error) { /* Couldn't lock the AGF so skip this AG. */ @@ -2620,7 +2620,8 @@ xfs_alloc_fix_freelist( * somewhere else if we are not being asked to try harder at this * point */ - if (pag->pagf_metadata && (args->datatype & XFS_ALLOC_USERDATA) && + if (xfs_perag_prefers_metadata(pag) && + (args->datatype & XFS_ALLOC_USERDATA) && (flags & XFS_ALLOC_FLAG_TRYLOCK)) { ASSERT(!(flags & XFS_ALLOC_FLAG_FREEING)); goto out_agbp_relse; @@ -2646,7 +2647,7 @@ xfs_alloc_fix_freelist( } /* reset a padding mismatched agfl before final free space check */ - if (pag->pagf_agflreset) + if (xfs_perag_agfl_needs_reset(pag)) xfs_agfl_reset(tp, agbp, pag); /* If there isn't enough total space or single-extent, reject it. */ @@ -2803,7 +2804,7 @@ xfs_alloc_get_freelist( if (be32_to_cpu(agf->agf_flfirst) == xfs_agfl_size(mp)) agf->agf_flfirst = 0; - ASSERT(!pag->pagf_agflreset); + ASSERT(!xfs_perag_agfl_needs_reset(pag)); be32_add_cpu(&agf->agf_flcount, -1); pag->pagf_flcount--; @@ -2892,7 +2893,7 @@ xfs_alloc_put_freelist( if (be32_to_cpu(agf->agf_fllast) == xfs_agfl_size(mp)) agf->agf_fllast = 0; - ASSERT(!pag->pagf_agflreset); + ASSERT(!xfs_perag_agfl_needs_reset(pag)); be32_add_cpu(&agf->agf_flcount, 1); pag->pagf_flcount++; @@ -3099,7 +3100,7 @@ xfs_alloc_read_agf( return error; agf = agfbp->b_addr; - if (!pag->pagf_init) { + if (!xfs_perag_initialised_agf(pag)) { pag->pagf_freeblks = be32_to_cpu(agf->agf_freeblks); pag->pagf_btreeblks = be32_to_cpu(agf->agf_btreeblks); pag->pagf_flcount = be32_to_cpu(agf->agf_flcount); @@ -3111,8 +3112,8 @@ xfs_alloc_read_agf( pag->pagf_levels[XFS_BTNUM_RMAPi] = be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAPi]); pag->pagf_refcount_level = be32_to_cpu(agf->agf_refcount_level); - pag->pagf_init = 1; - pag->pagf_agflreset = xfs_agfl_needs_reset(pag->pag_mount, agf); + if (xfs_agfl_needs_reset(pag->pag_mount, agf)) + set_bit(XFS_AGSTATE_AGFL_NEEDS_RESET, &pag->pag_opstate); /* * Update the in-core allocbt counter. Filter out the rmapbt @@ -3127,6 +3128,8 @@ xfs_alloc_read_agf( if (allocbt_blks > 0) atomic64_add(allocbt_blks, &pag->pag_mount->m_allocbt_blks); + + set_bit(XFS_AGSTATE_AGF_INIT, &pag->pag_opstate); } #ifdef DEBUG else if (!xfs_is_shutdown(pag->pag_mount)) { diff --git a/fs/xfs/libxfs/xfs_alloc_btree.c b/fs/xfs/libxfs/xfs_alloc_btree.c index 549a3cba0234d..0f29c7b1b39f3 100644 --- a/fs/xfs/libxfs/xfs_alloc_btree.c +++ b/fs/xfs/libxfs/xfs_alloc_btree.c @@ -315,7 +315,7 @@ xfs_allocbt_verify( level = be16_to_cpu(block->bb_level); if (bp->b_ops->magic[0] == cpu_to_be32(XFS_ABTC_MAGIC)) btnum = XFS_BTNUM_CNTi; - if (pag && pag->pagf_init) { + if (pag && xfs_perag_initialised_agf(pag)) { if (level >= pag->pagf_levels[btnum]) return __this_address; } else if (level >= mp->m_alloc_maxlevels) diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 504fd69fe2c60..a2282655a0f63 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -3147,7 +3147,7 @@ xfs_bmap_longest_free_extent( int error = 0; pag = xfs_perag_get(mp, ag); - if (!pag->pagf_init) { + if (!xfs_perag_initialised_agf(pag)) { error = xfs_alloc_read_agf(pag, tp, XFS_ALLOC_FLAG_TRYLOCK, NULL); if (error) { diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index 86b7adc355e6e..943972a085752 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -998,8 +998,8 @@ xfs_dialloc_ag_inobt( int i, j; int searchdistance = 10; - ASSERT(pag->pagi_init); - ASSERT(pag->pagi_inodeok); + ASSERT(xfs_perag_initialised_agi(pag)); + ASSERT(xfs_perag_allows_inodes(pag)); ASSERT(pag->pagi_freecount > 0); restart_pagno: @@ -1592,10 +1592,10 @@ xfs_dialloc_good_ag( if (!pag) return false; - if (!pag->pagi_inodeok) + if (!xfs_perag_allows_inodes(pag)) return false; - if (!pag->pagi_init) { + if (!xfs_perag_initialised_agi(pag)) { error = xfs_ialloc_read_agi(pag, tp, NULL); if (error) return false; @@ -1606,7 +1606,7 @@ xfs_dialloc_good_ag( if (!ok_alloc) return false; - if (!pag->pagf_init) { + if (!xfs_perag_initialised_agf(pag)) { error = xfs_alloc_read_agf(pag, tp, flags, NULL); if (error) return false; @@ -2603,10 +2603,10 @@ xfs_ialloc_read_agi( return error; agi = agibp->b_addr; - if (!pag->pagi_init) { + if (!xfs_perag_initialised_agi(pag)) { pag->pagi_freecount = be32_to_cpu(agi->agi_freecount); pag->pagi_count = be32_to_cpu(agi->agi_count); - pag->pagi_init = 1; + set_bit(XFS_AGSTATE_AGI_INIT, &pag->pag_opstate); } /* diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c index d657af2ec3503..3675a0d293103 100644 --- a/fs/xfs/libxfs/xfs_ialloc_btree.c +++ b/fs/xfs/libxfs/xfs_ialloc_btree.c @@ -291,8 +291,8 @@ xfs_inobt_verify( * Similarly, during log recovery we will have a perag structure * attached, but the agi information will not yet have been initialised * from the on disk AGI. We don't currently use any of this information, - * but beware of the landmine (i.e. need to check pag->pagi_init) if we - * ever do. + * but beware of the landmine (i.e. need to check + * xfs_perag_initialised_agi(pag)) if we ever do. */ if (xfs_has_crc(mp)) { fa = xfs_btree_sblock_v5hdr_verify(bp); diff --git a/fs/xfs/libxfs/xfs_refcount_btree.c b/fs/xfs/libxfs/xfs_refcount_btree.c index e1f7898666831..d20abf0390fc8 100644 --- a/fs/xfs/libxfs/xfs_refcount_btree.c +++ b/fs/xfs/libxfs/xfs_refcount_btree.c @@ -227,7 +227,7 @@ xfs_refcountbt_verify( return fa; level = be16_to_cpu(block->bb_level); - if (pag && pag->pagf_init) { + if (pag && xfs_perag_initialised_agf(pag)) { if (level >= pag->pagf_refcount_level) return __this_address; } else if (level >= mp->m_refc_maxlevels) diff --git a/fs/xfs/libxfs/xfs_rmap_btree.c b/fs/xfs/libxfs/xfs_rmap_btree.c index 7f83f62e51e0b..d3285684bb5eb 100644 --- a/fs/xfs/libxfs/xfs_rmap_btree.c +++ b/fs/xfs/libxfs/xfs_rmap_btree.c @@ -313,7 +313,7 @@ xfs_rmapbt_verify( return fa; level = be16_to_cpu(block->bb_level); - if (pag && pag->pagf_init) { + if (pag && xfs_perag_initialised_agf(pag)) { if (level >= pag->pagf_levels[XFS_BTNUM_RMAPi]) return __this_address; } else if (level >= mp->m_rmap_maxlevels) diff --git a/fs/xfs/scrub/agheader_repair.c b/fs/xfs/scrub/agheader_repair.c index b80b9111e781a..c37e6d72760b9 100644 --- a/fs/xfs/scrub/agheader_repair.c +++ b/fs/xfs/scrub/agheader_repair.c @@ -191,14 +191,15 @@ xrep_agf_init_header( struct xfs_agf *old_agf) { struct xfs_mount *mp = sc->mp; + struct xfs_perag *pag = sc->sa.pag; struct xfs_agf *agf = agf_bp->b_addr; memcpy(old_agf, agf, sizeof(*old_agf)); memset(agf, 0, BBTOB(agf_bp->b_length)); agf->agf_magicnum = cpu_to_be32(XFS_AGF_MAGIC); agf->agf_versionnum = cpu_to_be32(XFS_AGF_VERSION); - agf->agf_seqno = cpu_to_be32(sc->sa.pag->pag_agno); - agf->agf_length = cpu_to_be32(sc->sa.pag->block_count); + agf->agf_seqno = cpu_to_be32(pag->pag_agno); + agf->agf_length = cpu_to_be32(pag->block_count); agf->agf_flfirst = old_agf->agf_flfirst; agf->agf_fllast = old_agf->agf_fllast; agf->agf_flcount = old_agf->agf_flcount; @@ -206,8 +207,8 @@ xrep_agf_init_header( uuid_copy(&agf->agf_uuid, &mp->m_sb.sb_meta_uuid); /* Mark the incore AGF data stale until we're done fixing things. */ - ASSERT(sc->sa.pag->pagf_init); - sc->sa.pag->pagf_init = 0; + ASSERT(xfs_perag_initialised_agf(pag)); + clear_bit(XFS_AGSTATE_AGF_INIT, &pag->pag_opstate); } /* Set btree root information in an AGF. */ @@ -333,7 +334,7 @@ xrep_agf_commit_new( pag->pagf_levels[XFS_BTNUM_RMAPi] = be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAPi]); pag->pagf_refcount_level = be32_to_cpu(agf->agf_refcount_level); - pag->pagf_init = 1; + set_bit(XFS_AGSTATE_AGF_INIT, &pag->pag_opstate); return 0; } @@ -434,7 +435,7 @@ xrep_agf( out_revert: /* Mark the incore AGF state stale and revert the AGF. */ - sc->sa.pag->pagf_init = 0; + clear_bit(XFS_AGSTATE_AGF_INIT, &sc->sa.pag->pag_opstate); memcpy(agf, &old_agf, sizeof(old_agf)); return error; } @@ -618,7 +619,7 @@ xrep_agfl_update_agf( xfs_force_summary_recalc(sc->mp); /* Update the AGF counters. */ - if (sc->sa.pag->pagf_init) + if (xfs_perag_initialised_agf(sc->sa.pag)) sc->sa.pag->pagf_flcount = flcount; agf->agf_flfirst = cpu_to_be32(0); agf->agf_flcount = cpu_to_be32(flcount); @@ -822,14 +823,15 @@ xrep_agi_init_header( struct xfs_agi *old_agi) { struct xfs_agi *agi = agi_bp->b_addr; + struct xfs_perag *pag = sc->sa.pag; struct xfs_mount *mp = sc->mp; memcpy(old_agi, agi, sizeof(*old_agi)); memset(agi, 0, BBTOB(agi_bp->b_length)); agi->agi_magicnum = cpu_to_be32(XFS_AGI_MAGIC); agi->agi_versionnum = cpu_to_be32(XFS_AGI_VERSION); - agi->agi_seqno = cpu_to_be32(sc->sa.pag->pag_agno); - agi->agi_length = cpu_to_be32(sc->sa.pag->block_count); + agi->agi_seqno = cpu_to_be32(pag->pag_agno); + agi->agi_length = cpu_to_be32(pag->block_count); agi->agi_newino = cpu_to_be32(NULLAGINO); agi->agi_dirino = cpu_to_be32(NULLAGINO); if (xfs_has_crc(mp)) @@ -840,8 +842,8 @@ xrep_agi_init_header( sizeof(agi->agi_unlinked)); /* Mark the incore AGF data stale until we're done fixing things. */ - ASSERT(sc->sa.pag->pagi_init); - sc->sa.pag->pagi_init = 0; + ASSERT(xfs_perag_initialised_agi(pag)); + clear_bit(XFS_AGSTATE_AGI_INIT, &pag->pag_opstate); } /* Set btree root information in an AGI. */ @@ -928,7 +930,7 @@ xrep_agi_commit_new( pag = sc->sa.pag; pag->pagi_count = be32_to_cpu(agi->agi_count); pag->pagi_freecount = be32_to_cpu(agi->agi_freecount); - pag->pagi_init = 1; + set_bit(XFS_AGSTATE_AGI_INIT, &pag->pag_opstate); return 0; } @@ -993,7 +995,7 @@ xrep_agi( out_revert: /* Mark the incore AGI state stale and revert the AGI. */ - sc->sa.pag->pagi_init = 0; + clear_bit(XFS_AGSTATE_AGI_INIT, &sc->sa.pag->pag_opstate); memcpy(agi, &old_agi, sizeof(old_agi)); return error; } diff --git a/fs/xfs/scrub/fscounters.c b/fs/xfs/scrub/fscounters.c index ef97670970c31..f0c7f41897b90 100644 --- a/fs/xfs/scrub/fscounters.c +++ b/fs/xfs/scrub/fscounters.c @@ -86,7 +86,8 @@ xchk_fscount_warmup( for_each_perag(mp, agno, pag) { if (xchk_should_terminate(sc, &error)) break; - if (pag->pagi_init && pag->pagf_init) + if (xfs_perag_initialised_agi(pag) && + xfs_perag_initialised_agf(pag)) continue; /* Lock both AG headers. */ @@ -101,7 +102,8 @@ xchk_fscount_warmup( * These are supposed to be initialized by the header read * function. */ - if (!pag->pagi_init || !pag->pagf_init) { + if (!xfs_perag_initialised_agi(pag) || + !xfs_perag_initialised_agf(pag)) { error = -EFSCORRUPTED; break; } @@ -220,7 +222,8 @@ xchk_fscount_aggregate_agcounts( break; /* This somehow got unset since the warmup? */ - if (!pag->pagi_init || !pag->pagf_init) { + if (!xfs_perag_initialised_agi(pag) || + !xfs_perag_initialised_agf(pag)) { error = -EFSCORRUPTED; break; } diff --git a/fs/xfs/scrub/repair.c b/fs/xfs/scrub/repair.c index 4b92f9253ccd2..d0b1644efb896 100644 --- a/fs/xfs/scrub/repair.c +++ b/fs/xfs/scrub/repair.c @@ -206,7 +206,7 @@ xrep_calc_ag_resblks( return 0; pag = xfs_perag_get(mp, sm->sm_agno); - if (pag->pagi_init) { + if (xfs_perag_initialised_agi(pag)) { /* Use in-core icount if possible. */ icount = pag->pagi_count; } else { diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 34b21a29c39bc..7e8b25ab6c463 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -125,7 +125,7 @@ xfs_filestream_pick_ag( pag = xfs_perag_get(mp, ag); - if (!pag->pagf_init) { + if (!xfs_perag_initialised_agf(pag)) { err = xfs_alloc_read_agf(pag, NULL, trylock, NULL); if (err) { if (err != -EAGAIN) { @@ -159,7 +159,8 @@ xfs_filestream_pick_ag( xfs_ag_resv_needed(pag, XFS_AG_RESV_NONE)); if (((minlen && longest >= minlen) || (!minlen && pag->pagf_freeblks >= minfree)) && - (!pag->pagf_metadata || !(flags & XFS_PICK_USERDATA) || + (!xfs_perag_prefers_metadata(pag) || + !(flags & XFS_PICK_USERDATA) || (flags & XFS_PICK_LOWSPACE))) { /* Break out, retaining the reference on the AG. */ diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 96375b5622fd3..2479b5cbd75ec 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -247,6 +247,32 @@ xfs_fs_show_options( return 0; } +static bool +xfs_set_inode_alloc_perag( + struct xfs_perag *pag, + xfs_ino_t ino, + xfs_agnumber_t max_metadata) +{ + if (!xfs_is_inode32(pag->pag_mount)) { + set_bit(XFS_AGSTATE_ALLOWS_INODES, &pag->pag_opstate); + clear_bit(XFS_AGSTATE_PREFERS_METADATA, &pag->pag_opstate); + return false; + } + + if (ino > XFS_MAXINUMBER_32) { + clear_bit(XFS_AGSTATE_ALLOWS_INODES, &pag->pag_opstate); + clear_bit(XFS_AGSTATE_PREFERS_METADATA, &pag->pag_opstate); + return false; + } + + set_bit(XFS_AGSTATE_ALLOWS_INODES, &pag->pag_opstate); + if (pag->pag_agno < max_metadata) + set_bit(XFS_AGSTATE_PREFERS_METADATA, &pag->pag_opstate); + else + clear_bit(XFS_AGSTATE_PREFERS_METADATA, &pag->pag_opstate); + return true; +} + /* * Set parameters for inode allocation heuristics, taking into account * filesystem size and inode32/inode64 mount options; i.e. specifically @@ -310,24 +336,8 @@ xfs_set_inode_alloc( ino = XFS_AGINO_TO_INO(mp, index, agino); pag = xfs_perag_get(mp, index); - - if (xfs_is_inode32(mp)) { - if (ino > XFS_MAXINUMBER_32) { - pag->pagi_inodeok = 0; - pag->pagf_metadata = 0; - } else { - pag->pagi_inodeok = 1; - maxagi++; - if (index < max_metadata) - pag->pagf_metadata = 1; - else - pag->pagf_metadata = 0; - } - } else { - pag->pagi_inodeok = 1; - pag->pagf_metadata = 0; - } - + if (xfs_set_inode_alloc_perag(pag, ino, max_metadata)) + maxagi++; xfs_perag_put(pag); } From 76257a15873ccce817e0c4441f6bb66fb8f8201c Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:53 +1100 Subject: [PATCH 224/765] xfs: introduce xfs_for_each_perag_wrap() In several places we iterate every AG from a specific start agno and wrap back to the first AG when we reach the end of the filesystem to continue searching. We don't have a primitive for this iteration yet, so add one for conversion of these algorithms to per-ag based iteration. The filestream AG select code is a mess, and this initially makes it worse. The per-ag selection needs to be driven completely into the filestream code to clean this up and it will be done in a future patch that makes the filestream allocator use active per-ag references correctly. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_ag.h | 45 +++++++++++++++++++++- fs/xfs/libxfs/xfs_bmap.c | 78 ++++++++++++++++++++++---------------- fs/xfs/libxfs/xfs_ialloc.c | 32 ++++++++-------- 3 files changed, 105 insertions(+), 50 deletions(-) diff --git a/fs/xfs/libxfs/xfs_ag.h b/fs/xfs/libxfs/xfs_ag.h index 187d30d9bb139..8f43b91d4cf35 100644 --- a/fs/xfs/libxfs/xfs_ag.h +++ b/fs/xfs/libxfs/xfs_ag.h @@ -237,7 +237,6 @@ xfs_perag_next( #define for_each_perag_from(mp, agno, pag) \ for_each_perag_range((mp), (agno), (mp)->m_sb.sb_agcount - 1, (pag)) - #define for_each_perag(mp, agno, pag) \ (agno) = 0; \ for_each_perag_from((mp), (agno), (pag)) @@ -249,6 +248,50 @@ xfs_perag_next( xfs_perag_rele(pag), \ (pag) = xfs_perag_grab_tag((mp), (agno), (tag))) +static inline struct xfs_perag * +xfs_perag_next_wrap( + struct xfs_perag *pag, + xfs_agnumber_t *agno, + xfs_agnumber_t stop_agno, + xfs_agnumber_t wrap_agno) +{ + struct xfs_mount *mp = pag->pag_mount; + + *agno = pag->pag_agno + 1; + xfs_perag_rele(pag); + while (*agno != stop_agno) { + if (*agno >= wrap_agno) + *agno = 0; + if (*agno == stop_agno) + break; + + pag = xfs_perag_grab(mp, *agno); + if (pag) + return pag; + (*agno)++; + } + return NULL; +} + +/* + * Iterate all AGs from start_agno through wrap_agno, then 0 through + * (start_agno - 1). + */ +#define for_each_perag_wrap_at(mp, start_agno, wrap_agno, agno, pag) \ + for ((agno) = (start_agno), (pag) = xfs_perag_grab((mp), (agno)); \ + (pag) != NULL; \ + (pag) = xfs_perag_next_wrap((pag), &(agno), (start_agno), \ + (wrap_agno))) + +/* + * Iterate all AGs from start_agno through to the end of the filesystem, then 0 + * through (start_agno - 1). + */ +#define for_each_perag_wrap(mp, start_agno, agno, pag) \ + for_each_perag_wrap_at((mp), (start_agno), (mp)->m_sb.sb_agcount, \ + (agno), (pag)) + + struct aghdr_init_data { /* per ag data */ xfs_agblock_t agno; /* ag to init */ diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index a2282655a0f63..87ad56e034c03 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -3136,17 +3136,14 @@ xfs_bmap_adjacent( static int xfs_bmap_longest_free_extent( + struct xfs_perag *pag, struct xfs_trans *tp, - xfs_agnumber_t ag, xfs_extlen_t *blen, int *notinit) { - struct xfs_mount *mp = tp->t_mountp; - struct xfs_perag *pag; xfs_extlen_t longest; int error = 0; - pag = xfs_perag_get(mp, ag); if (!xfs_perag_initialised_agf(pag)) { error = xfs_alloc_read_agf(pag, tp, XFS_ALLOC_FLAG_TRYLOCK, NULL); @@ -3156,19 +3153,17 @@ xfs_bmap_longest_free_extent( *notinit = 1; error = 0; } - goto out; + return error; } } longest = xfs_alloc_longest_free_extent(pag, - xfs_alloc_min_freelist(mp, pag), + xfs_alloc_min_freelist(pag->pag_mount, pag), xfs_ag_resv_needed(pag, XFS_AG_RESV_NONE)); if (*blen < longest) *blen = longest; -out: - xfs_perag_put(pag); - return error; + return 0; } static void @@ -3206,9 +3201,10 @@ xfs_bmap_btalloc_select_lengths( xfs_extlen_t *blen) { struct xfs_mount *mp = ap->ip->i_mount; - xfs_agnumber_t ag, startag; + struct xfs_perag *pag; + xfs_agnumber_t agno, startag; int notinit = 0; - int error; + int error = 0; args->type = XFS_ALLOCTYPE_START_BNO; if (ap->tp->t_flags & XFS_TRANS_LOWMODE) { @@ -3218,24 +3214,24 @@ xfs_bmap_btalloc_select_lengths( } args->total = ap->total; - startag = ag = XFS_FSB_TO_AGNO(mp, args->fsbno); + startag = XFS_FSB_TO_AGNO(mp, args->fsbno); if (startag == NULLAGNUMBER) - startag = ag = 0; + startag = 0; - while (*blen < args->maxlen) { - error = xfs_bmap_longest_free_extent(args->tp, ag, blen, + *blen = 0; + for_each_perag_wrap(mp, startag, agno, pag) { + error = xfs_bmap_longest_free_extent(pag, args->tp, blen, ¬init); if (error) - return error; - - if (++ag == mp->m_sb.sb_agcount) - ag = 0; - if (ag == startag) + break; + if (*blen >= args->maxlen) break; } + if (pag) + xfs_perag_rele(pag); xfs_bmap_select_minlen(ap, args, blen, notinit); - return 0; + return error; } STATIC int @@ -3245,7 +3241,8 @@ xfs_bmap_btalloc_filestreams( xfs_extlen_t *blen) { struct xfs_mount *mp = ap->ip->i_mount; - xfs_agnumber_t ag; + struct xfs_perag *pag; + xfs_agnumber_t start_agno; int notinit = 0; int error; @@ -3259,33 +3256,50 @@ xfs_bmap_btalloc_filestreams( args->type = XFS_ALLOCTYPE_NEAR_BNO; args->total = ap->total; - ag = XFS_FSB_TO_AGNO(mp, args->fsbno); - if (ag == NULLAGNUMBER) - ag = 0; + start_agno = XFS_FSB_TO_AGNO(mp, args->fsbno); + if (start_agno == NULLAGNUMBER) + start_agno = 0; - error = xfs_bmap_longest_free_extent(args->tp, ag, blen, ¬init); - if (error) - return error; + pag = xfs_perag_grab(mp, start_agno); + if (pag) { + error = xfs_bmap_longest_free_extent(pag, args->tp, blen, + ¬init); + xfs_perag_rele(pag); + if (error) + return error; + } if (*blen < args->maxlen) { - error = xfs_filestream_new_ag(ap, &ag); + xfs_agnumber_t agno = start_agno; + + error = xfs_filestream_new_ag(ap, &agno); if (error) return error; + if (agno == NULLAGNUMBER) + goto out_select; - error = xfs_bmap_longest_free_extent(args->tp, ag, blen, - ¬init); + pag = xfs_perag_grab(mp, agno); + if (!pag) + goto out_select; + + error = xfs_bmap_longest_free_extent(pag, args->tp, + blen, ¬init); + xfs_perag_rele(pag); if (error) return error; + start_agno = agno; + } +out_select: xfs_bmap_select_minlen(ap, args, blen, notinit); /* * Set the failure fallback case to look in the selected AG as stream * may have moved. */ - ap->blkno = args->fsbno = XFS_AGB_TO_FSB(mp, ag, 0); + ap->blkno = args->fsbno = XFS_AGB_TO_FSB(mp, start_agno, 0); return 0; } diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index 943972a085752..20d2365524a48 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -1725,7 +1725,7 @@ xfs_dialloc( bool ok_alloc = true; bool low_space = false; int flags; - xfs_ino_t ino; + xfs_ino_t ino = NULLFSINO; /* * Directories, symlinks, and regular files frequently allocate at least @@ -1773,39 +1773,37 @@ xfs_dialloc( * or in which we can allocate some inodes. Iterate through the * allocation groups upward, wrapping at the end. */ - agno = start_agno; flags = XFS_ALLOC_FLAG_TRYLOCK; - for (;;) { - pag = xfs_perag_grab(mp, agno); +retry: + for_each_perag_wrap_at(mp, start_agno, mp->m_maxagi, agno, pag) { if (xfs_dialloc_good_ag(pag, *tpp, mode, flags, ok_alloc)) { error = xfs_dialloc_try_ag(pag, tpp, parent, &ino, ok_alloc); if (error != -EAGAIN) break; + error = 0; } if (xfs_is_shutdown(mp)) { error = -EFSCORRUPTED; break; } - if (++agno == mp->m_maxagi) - agno = 0; - if (agno == start_agno) { - if (!flags) { - error = -ENOSPC; - break; - } + } + if (pag) + xfs_perag_rele(pag); + if (error) + return error; + if (ino == NULLFSINO) { + if (flags) { flags = 0; if (low_space) ok_alloc = true; + goto retry; } - xfs_perag_rele(pag); + return -ENOSPC; } - - if (!error) - *new_ino = ino; - xfs_perag_rele(pag); - return error; + *new_ino = ino; + return 0; } /* From ecd788a92460eef44c5444290757bfd0f38d91b4 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:53 +1100 Subject: [PATCH 225/765] xfs: rework xfs_alloc_vextent() It's a multiplexing mess that can be greatly simplified, and really needs to be simplified to allow active per-ag references to propagate from initial AG selection code the the bmapi code. This splits the code out into separate a parameter checking function, an iterator function, and allocation completion functions and then implements the individual policies using these functions. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_alloc.c | 464 +++++++++++++++++++++++--------------- 1 file changed, 285 insertions(+), 179 deletions(-) diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index 4508862239a88..5afda109aaefc 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -3151,29 +3151,20 @@ xfs_alloc_read_agf( } /* - * Allocate an extent (variable-size). - * Depending on the allocation type, we either look in a single allocation - * group or loop over the allocation groups to find the result. + * Pre-proces allocation arguments to set initial state that we don't require + * callers to set up correctly, as well as bounds check the allocation args + * that are set up. */ -int /* error */ -xfs_alloc_vextent( - struct xfs_alloc_arg *args) /* allocation argument structure */ +static int +xfs_alloc_vextent_check_args( + struct xfs_alloc_arg *args) { - xfs_agblock_t agsize; /* allocation group size */ - int error; - int flags; /* XFS_ALLOC_FLAG_... locking flags */ - struct xfs_mount *mp; /* mount structure pointer */ - xfs_agnumber_t sagno; /* starting allocation group number */ - xfs_alloctype_t type; /* input allocation type */ - int bump_rotor = 0; - xfs_agnumber_t rotorstep = xfs_rotorstep; /* inode32 agf stepper */ - xfs_agnumber_t minimum_agno = 0; + struct xfs_mount *mp = args->mp; + xfs_agblock_t agsize; - mp = args->mp; - type = args->otype = args->type; + args->otype = args->type; args->agbno = NULLAGBLOCK; - if (args->tp->t_highest_agno != NULLAGNUMBER) - minimum_agno = args->tp->t_highest_agno; + /* * Just fix this up, for the case where the last a.g. is shorter * (or there's only one a.g.) and the caller couldn't easily figure @@ -3195,199 +3186,314 @@ xfs_alloc_vextent( args->mod >= args->prod) { args->fsbno = NULLFSBLOCK; trace_xfs_alloc_vextent_badargs(args); + return -ENOSPC; + } + return 0; +} + +/* + * Post-process allocation results to set the allocated block number correctly + * for the caller. + * + * XXX: xfs_alloc_vextent() should really be returning ENOSPC for ENOSPC, not + * hiding it behind a "successful" NULLFSBLOCK allocation. + */ +static void +xfs_alloc_vextent_set_fsbno( + struct xfs_alloc_arg *args, + xfs_agnumber_t minimum_agno) +{ + struct xfs_mount *mp = args->mp; + + /* + * We can end up here with a locked AGF. If we failed, the caller is + * likely going to try to allocate again with different parameters, and + * that can widen the AGs that are searched for free space. If we have + * to do BMBT block allocation, we have to do a new allocation. + * + * Hence leaving this function with the AGF locked opens up potential + * ABBA AGF deadlocks because a future allocation attempt in this + * transaction may attempt to lock a lower number AGF. + * + * We can't release the AGF until the transaction is commited, so at + * this point we must update the "first allocation" tracker to point at + * this AG if the tracker is empty or points to a lower AG. This allows + * the next allocation attempt to be modified appropriately to avoid + * deadlocks. + */ + if (args->agbp && + (args->tp->t_highest_agno == NULLAGNUMBER || + args->agno > minimum_agno)) + args->tp->t_highest_agno = args->agno; + + /* Allocation failed with ENOSPC if NULLAGBLOCK was returned. */ + if (args->agbno == NULLAGBLOCK) { + args->fsbno = NULLFSBLOCK; + return; + } + + args->fsbno = XFS_AGB_TO_FSB(mp, args->agno, args->agbno); +#ifdef DEBUG + ASSERT(args->len >= args->minlen); + ASSERT(args->len <= args->maxlen); + ASSERT(args->agbno % args->alignment == 0); + XFS_AG_CHECK_DADDR(mp, XFS_FSB_TO_DADDR(mp, args->fsbno), args->len); +#endif +} + +/* + * Allocate within a single AG only. + */ +static int +xfs_alloc_vextent_this_ag( + struct xfs_alloc_arg *args, + xfs_agnumber_t minimum_agno) +{ + struct xfs_mount *mp = args->mp; + int error; + + error = xfs_alloc_vextent_check_args(args); + if (error) { + if (error == -ENOSPC) + return 0; + return error; + } + + args->agno = XFS_FSB_TO_AGNO(mp, args->fsbno); + if (minimum_agno > args->agno) { + trace_xfs_alloc_vextent_skip_deadlock(args); + args->fsbno = NULLFSBLOCK; return 0; } - switch (type) { - case XFS_ALLOCTYPE_THIS_AG: - case XFS_ALLOCTYPE_NEAR_BNO: - case XFS_ALLOCTYPE_THIS_BNO: - /* - * These three force us into a single a.g. - */ - args->agno = XFS_FSB_TO_AGNO(mp, args->fsbno); - args->pag = xfs_perag_get(mp, args->agno); + args->pag = xfs_perag_get(mp, args->agno); + error = xfs_alloc_fix_freelist(args, 0); + if (error) { + trace_xfs_alloc_vextent_nofix(args); + goto out_error; + } + if (!args->agbp) { + trace_xfs_alloc_vextent_noagbp(args); + args->fsbno = NULLFSBLOCK; + goto out_error; + } + args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); + error = xfs_alloc_ag_vextent(args); - if (minimum_agno > args->agno) { - trace_xfs_alloc_vextent_skip_deadlock(args); - error = 0; - break; - } + xfs_alloc_vextent_set_fsbno(args, minimum_agno); +out_error: + xfs_perag_put(args->pag); + return error; +} + +/* + * Iterate all AGs trying to allocate an extent starting from @start_ag. + * + * If the incoming allocation type is XFS_ALLOCTYPE_NEAR_BNO, it means the + * allocation attempts in @start_agno have locality information. If we fail to + * allocate in that AG, then we revert to anywhere-in-AG for all the other AGs + * we attempt to allocation in as there is no locality optimisation possible for + * those allocations. + * + * When we wrap the AG iteration at the end of the filesystem, we have to be + * careful not to wrap into AGs below ones we already have locked in the + * transaction if we are doing a blocking iteration. This will result in an + * out-of-order locking of AGFs and hence can cause deadlocks. + */ +static int +xfs_alloc_vextent_iterate_ags( + struct xfs_alloc_arg *args, + xfs_agnumber_t minimum_agno, + xfs_agnumber_t start_agno, + uint32_t flags) +{ + struct xfs_mount *mp = args->mp; + int error = 0; - error = xfs_alloc_fix_freelist(args, 0); + ASSERT(start_agno >= minimum_agno); + + /* + * Loop over allocation groups twice; first time with + * trylock set, second time without. + */ + args->agno = start_agno; + for (;;) { + args->pag = xfs_perag_get(mp, args->agno); + error = xfs_alloc_fix_freelist(args, flags); if (error) { trace_xfs_alloc_vextent_nofix(args); - goto error0; - } - if (!args->agbp) { - trace_xfs_alloc_vextent_noagbp(args); break; } - args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); - if ((error = xfs_alloc_ag_vextent(args))) - goto error0; - break; - case XFS_ALLOCTYPE_START_BNO: /* - * Try near allocation first, then anywhere-in-ag after - * the first a.g. fails. + * If we get a buffer back then the allocation will fly. */ - if ((args->datatype & XFS_ALLOC_INITIAL_USER_DATA) && - xfs_is_inode32(mp)) { - args->fsbno = XFS_AGB_TO_FSB(mp, - ((mp->m_agfrotor / rotorstep) % - mp->m_sb.sb_agcount), 0); - bump_rotor = 1; + if (args->agbp) { + error = xfs_alloc_ag_vextent(args); + break; } - args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); - args->type = XFS_ALLOCTYPE_NEAR_BNO; - fallthrough; - case XFS_ALLOCTYPE_FIRST_AG: + + trace_xfs_alloc_vextent_loopfailed(args); + /* - * Rotate through the allocation groups looking for a winner. - * If we are blocking, we must obey minimum_agno contraints for - * avoiding ABBA deadlocks on AGF locking. + * Didn't work, figure out the next iteration. */ - if (type == XFS_ALLOCTYPE_FIRST_AG) { - /* - * Start with allocation group given by bno. - */ - args->agno = XFS_FSB_TO_AGNO(mp, args->fsbno); + if (args->agno == start_agno && + args->otype == XFS_ALLOCTYPE_START_BNO) args->type = XFS_ALLOCTYPE_THIS_AG; - sagno = minimum_agno; - flags = 0; - } else { - /* - * Start with the given allocation group. - */ - args->agno = sagno = XFS_FSB_TO_AGNO(mp, args->fsbno); - flags = XFS_ALLOC_FLAG_TRYLOCK; + + /* + * If we are try-locking, we can't deadlock on AGF locks so we + * can wrap all the way back to the first AG. Otherwise, wrap + * back to the start AG so we can't deadlock and let the end of + * scan handler decide what to do next. + */ + if (++(args->agno) == mp->m_sb.sb_agcount) { + if (flags & XFS_ALLOC_FLAG_TRYLOCK) + args->agno = 0; + else + args->agno = minimum_agno; } /* - * Loop over allocation groups twice; first time with - * trylock set, second time without. + * Reached the starting a.g., must either be done + * or switch to non-trylock mode. */ - for (;;) { - args->pag = xfs_perag_get(mp, args->agno); - error = xfs_alloc_fix_freelist(args, flags); - if (error) { - trace_xfs_alloc_vextent_nofix(args); - goto error0; - } - /* - * If we get a buffer back then the allocation will fly. - */ - if (args->agbp) { - if ((error = xfs_alloc_ag_vextent(args))) - goto error0; + if (args->agno == start_agno) { + if (flags == 0) { + args->agbno = NULLAGBLOCK; + trace_xfs_alloc_vextent_allfailed(args); break; } - trace_xfs_alloc_vextent_loopfailed(args); + flags = 0; + if (args->otype == XFS_ALLOCTYPE_START_BNO) { + args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); + args->type = XFS_ALLOCTYPE_NEAR_BNO; + } + } + xfs_perag_put(args->pag); + args->pag = NULL; + } + if (args->pag) { + xfs_perag_put(args->pag); + args->pag = NULL; + } + return error; +} - /* - * Didn't work, figure out the next iteration. - */ - if (args->agno == sagno && - type == XFS_ALLOCTYPE_START_BNO) - args->type = XFS_ALLOCTYPE_THIS_AG; +/* + * Iterate from the AGs from the start AG to the end of the filesystem, trying + * to allocate blocks. It starts with a near allocation attempt in the initial + * AG, then falls back to anywhere-in-ag after the first AG fails. It will wrap + * back to zero if allowed by previous allocations in this transaction, + * otherwise will wrap back to the start AG and run a second blocking pass to + * the end of the filesystem. + */ +static int +xfs_alloc_vextent_start_ag( + struct xfs_alloc_arg *args, + xfs_agnumber_t minimum_agno) +{ + struct xfs_mount *mp = args->mp; + xfs_agnumber_t start_agno; + xfs_agnumber_t rotorstep = xfs_rotorstep; + bool bump_rotor = false; + int error; - /* - * If we are try-locking, we can't deadlock on AGF - * locks, so we can wrap all the way back to the first - * AG. Otherwise, wrap back to the start AG so we can't - * deadlock, and let the end of scan handler decide what - * to do next. - */ - if (++(args->agno) == mp->m_sb.sb_agcount) { - if (flags & XFS_ALLOC_FLAG_TRYLOCK) - args->agno = 0; - else - args->agno = sagno; - } + error = xfs_alloc_vextent_check_args(args); + if (error) { + if (error == -ENOSPC) + return 0; + return error; + } - /* - * Reached the starting a.g., must either be done - * or switch to non-trylock mode. - */ - if (args->agno == sagno) { - if (flags == 0) { - args->agbno = NULLAGBLOCK; - trace_xfs_alloc_vextent_allfailed(args); - break; - } + if ((args->datatype & XFS_ALLOC_INITIAL_USER_DATA) && + xfs_is_inode32(mp)) { + args->fsbno = XFS_AGB_TO_FSB(mp, + ((mp->m_agfrotor / rotorstep) % + mp->m_sb.sb_agcount), 0); + bump_rotor = 1; + } + start_agno = max(minimum_agno, XFS_FSB_TO_AGNO(mp, args->fsbno)); + args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); + args->type = XFS_ALLOCTYPE_NEAR_BNO; - /* - * Blocking pass next, so we must obey minimum - * agno constraints to avoid ABBA AGF deadlocks. - */ - flags = 0; - if (minimum_agno > sagno) - sagno = minimum_agno; - - if (type == XFS_ALLOCTYPE_START_BNO) { - args->agbno = XFS_FSB_TO_AGBNO(mp, - args->fsbno); - args->type = XFS_ALLOCTYPE_NEAR_BNO; - } - } - xfs_perag_put(args->pag); - } - if (bump_rotor) { - if (args->agno == sagno) - mp->m_agfrotor = (mp->m_agfrotor + 1) % - (mp->m_sb.sb_agcount * rotorstep); - else - mp->m_agfrotor = (args->agno * rotorstep + 1) % - (mp->m_sb.sb_agcount * rotorstep); - } - break; - default: - ASSERT(0); - /* NOTREACHED */ + error = xfs_alloc_vextent_iterate_ags(args, minimum_agno, start_agno, + XFS_ALLOC_FLAG_TRYLOCK); + if (bump_rotor) { + if (args->agno == start_agno) + mp->m_agfrotor = (mp->m_agfrotor + 1) % + (mp->m_sb.sb_agcount * rotorstep); + else + mp->m_agfrotor = (args->agno * rotorstep + 1) % + (mp->m_sb.sb_agcount * rotorstep); } - if (args->agbno == NULLAGBLOCK) { - args->fsbno = NULLFSBLOCK; - } else { - args->fsbno = XFS_AGB_TO_FSB(mp, args->agno, args->agbno); -#ifdef DEBUG - ASSERT(args->len >= args->minlen); - ASSERT(args->len <= args->maxlen); - ASSERT(args->agbno % args->alignment == 0); - XFS_AG_CHECK_DADDR(mp, XFS_FSB_TO_DADDR(mp, args->fsbno), - args->len); -#endif + xfs_alloc_vextent_set_fsbno(args, minimum_agno); + return error; +} + +/* + * Iterate from the agno indicated from args->fsbno through to the end of the + * filesystem attempting blocking allocation. This does not wrap or try a second + * pass, so will not recurse into AGs lower than indicated by fsbno. + */ +static int +xfs_alloc_vextent_first_ag( + struct xfs_alloc_arg *args, + xfs_agnumber_t minimum_agno) +{ + struct xfs_mount *mp = args->mp; + xfs_agnumber_t start_agno; + int error; + + error = xfs_alloc_vextent_check_args(args); + if (error) { + if (error == -ENOSPC) + return 0; + return error; } - /* - * We end up here with a locked AGF. If we failed, the caller is likely - * going to try to allocate again with different parameters, and that - * can widen the AGs that are searched for free space. If we have to do - * BMBT block allocation, we have to do a new allocation. - * - * Hence leaving this function with the AGF locked opens up potential - * ABBA AGF deadlocks because a future allocation attempt in this - * transaction may attempt to lock a lower number AGF. - * - * We can't release the AGF until the transaction is commited, so at - * this point we must update the "firstblock" tracker to point at this - * AG if the tracker is empty or points to a lower AG. This allows the - * next allocation attempt to be modified appropriately to avoid - * deadlocks. - */ - if (args->agbp && - (args->tp->t_highest_agno == NULLAGNUMBER || - args->pag->pag_agno > minimum_agno)) - args->tp->t_highest_agno = args->pag->pag_agno; - xfs_perag_put(args->pag); - return 0; -error0: - xfs_perag_put(args->pag); + start_agno = max(minimum_agno, XFS_FSB_TO_AGNO(mp, args->fsbno)); + + args->type = XFS_ALLOCTYPE_THIS_AG; + error = xfs_alloc_vextent_iterate_ags(args, minimum_agno, + start_agno, 0); + xfs_alloc_vextent_set_fsbno(args, minimum_agno); return error; } +/* + * Allocate an extent (variable-size). + * Depending on the allocation type, we either look in a single allocation + * group or loop over the allocation groups to find the result. + */ +int +xfs_alloc_vextent( + struct xfs_alloc_arg *args) +{ + xfs_agnumber_t minimum_agno = 0; + + if (args->tp->t_highest_agno != NULLAGNUMBER) + minimum_agno = args->tp->t_highest_agno; + + switch (args->type) { + case XFS_ALLOCTYPE_THIS_AG: + case XFS_ALLOCTYPE_NEAR_BNO: + case XFS_ALLOCTYPE_THIS_BNO: + return xfs_alloc_vextent_this_ag(args, minimum_agno); + case XFS_ALLOCTYPE_START_BNO: + return xfs_alloc_vextent_start_ag(args, minimum_agno); + case XFS_ALLOCTYPE_FIRST_AG: + return xfs_alloc_vextent_first_ag(args, minimum_agno); + default: + ASSERT(0); + /* NOTREACHED */ + } + /* Should never get here */ + return -EFSCORRUPTED; +} + /* Ensure that the freelist is at full capacity. */ int xfs_free_extent_fix_freelist( From 2edf06a50f5bbe664283f3c55c480fc013221d70 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:53 +1100 Subject: [PATCH 226/765] xfs: factor xfs_alloc_vextent_this_ag() for _iterate_ags() The core of the per-ag iteration is effectively doing a "this ag" allocation on one AG at a time. Use the same code to implement the core "this ag" allocation in both xfs_alloc_vextent_this_ag() and xfs_alloc_vextent_iterate_ags(). This means we only call xfs_alloc_ag_vextent() from one place so we can easily collapse the call stack in future patches. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_alloc.c | 50 ++++++++++++++++++++------------------- 1 file changed, 26 insertions(+), 24 deletions(-) diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index 5afda109aaefc..98defd19e09e6 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -3244,6 +3244,28 @@ xfs_alloc_vextent_set_fsbno( /* * Allocate within a single AG only. */ +static int +__xfs_alloc_vextent_this_ag( + struct xfs_alloc_arg *args) +{ + struct xfs_mount *mp = args->mp; + int error; + + error = xfs_alloc_fix_freelist(args, 0); + if (error) { + trace_xfs_alloc_vextent_nofix(args); + return error; + } + if (!args->agbp) { + /* cannot allocate in this AG at all */ + trace_xfs_alloc_vextent_noagbp(args); + args->agbno = NULLAGBLOCK; + return 0; + } + args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); + return xfs_alloc_ag_vextent(args); +} + static int xfs_alloc_vextent_this_ag( struct xfs_alloc_arg *args, @@ -3267,21 +3289,9 @@ xfs_alloc_vextent_this_ag( } args->pag = xfs_perag_get(mp, args->agno); - error = xfs_alloc_fix_freelist(args, 0); - if (error) { - trace_xfs_alloc_vextent_nofix(args); - goto out_error; - } - if (!args->agbp) { - trace_xfs_alloc_vextent_noagbp(args); - args->fsbno = NULLFSBLOCK; - goto out_error; - } - args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); - error = xfs_alloc_ag_vextent(args); + error = __xfs_alloc_vextent_this_ag(args); xfs_alloc_vextent_set_fsbno(args, minimum_agno); -out_error: xfs_perag_put(args->pag); return error; } @@ -3319,24 +3329,16 @@ xfs_alloc_vextent_iterate_ags( args->agno = start_agno; for (;;) { args->pag = xfs_perag_get(mp, args->agno); - error = xfs_alloc_fix_freelist(args, flags); + error = __xfs_alloc_vextent_this_ag(args); if (error) { - trace_xfs_alloc_vextent_nofix(args); + args->agbno = NULLAGBLOCK; break; } - /* - * If we get a buffer back then the allocation will fly. - */ - if (args->agbp) { - error = xfs_alloc_ag_vextent(args); + if (args->agbp) break; - } trace_xfs_alloc_vextent_loopfailed(args); - /* - * Didn't work, figure out the next iteration. - */ if (args->agno == start_agno && args->otype == XFS_ALLOCTYPE_START_BNO) args->type = XFS_ALLOCTYPE_THIS_AG; From 4811c933ea1ab7de86507dc9f7c9d3d9d71cafb5 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:53 +1100 Subject: [PATCH 227/765] xfs: combine __xfs_alloc_vextent_this_ag and xfs_alloc_ag_vextent There's a bit of a recursive conundrum around xfs_alloc_ag_vextent(). We can't first call xfs_alloc_ag_vextent() without preparing the AGFL for the allocation, and preparing the AGFL calls xfs_alloc_ag_vextent() to prepare the AGFL for the allocation. This "double allocation" requirement is not really clear from the current xfs_alloc_fix_freelist() calls that are sprinkled through the allocation code. It's not helped that xfs_alloc_ag_vextent() can actually allocate from the AGFL itself, but there's special code to prevent AGFL prep allocations from allocating from the free list it's trying to prep. The naming is also not consistent: args->wasfromfl is true when we allocated _from_ the free list, but the indication that we are allocating _for_ the free list is via checking that (args->resv == XFS_AG_RESV_AGFL). So, lets make this "allocation required for allocation" situation clear by moving it all inside xfs_alloc_ag_vextent(). The freelist allocation is a specific XFS_ALLOCTYPE_THIS_AG allocation, which translated directly to xfs_alloc_ag_vextent_size() allocation. This enables us to replace __xfs_alloc_vextent_this_ag() with a call to xfs_alloc_ag_vextent(), and we drive the freelist fixing further into the per-ag allocation algorithm. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_alloc.c | 65 +++++++++++++++++++++------------------ 1 file changed, 35 insertions(+), 30 deletions(-) diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index 98defd19e09e6..edcbf8fd10ea0 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -1140,22 +1140,38 @@ xfs_alloc_ag_vextent_small( * and of the form k * prod + mod unless there's nothing that large. * Return the starting a.g. block, or NULLAGBLOCK if we can't do it. */ -STATIC int /* error */ +static int xfs_alloc_ag_vextent( - xfs_alloc_arg_t *args) /* argument structure for allocation */ + struct xfs_alloc_arg *args) { - int error=0; + struct xfs_mount *mp = args->mp; + int error = 0; ASSERT(args->minlen > 0); ASSERT(args->maxlen > 0); ASSERT(args->minlen <= args->maxlen); ASSERT(args->mod < args->prod); ASSERT(args->alignment > 0); + ASSERT(args->resv != XFS_AG_RESV_AGFL); + + + error = xfs_alloc_fix_freelist(args, 0); + if (error) { + trace_xfs_alloc_vextent_nofix(args); + return error; + } + if (!args->agbp) { + /* cannot allocate in this AG at all */ + trace_xfs_alloc_vextent_noagbp(args); + args->agbno = NULLAGBLOCK; + return 0; + } + args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); + args->wasfromfl = 0; /* * Branch to correct routine based on the type. */ - args->wasfromfl = 0; switch (args->type) { case XFS_ALLOCTYPE_THIS_AG: error = xfs_alloc_ag_vextent_size(args); @@ -1176,7 +1192,6 @@ xfs_alloc_ag_vextent( ASSERT(args->len >= args->minlen); ASSERT(args->len <= args->maxlen); - ASSERT(!args->wasfromfl || args->resv != XFS_AG_RESV_AGFL); ASSERT(args->agbno % args->alignment == 0); /* if not file data, insert new block into the reverse map btree */ @@ -2721,7 +2736,7 @@ xfs_alloc_fix_freelist( targs.resv = XFS_AG_RESV_AGFL; /* Allocate as many blocks as possible at once. */ - error = xfs_alloc_ag_vextent(&targs); + error = xfs_alloc_ag_vextent_size(&targs); if (error) goto out_agflbp_relse; @@ -2735,6 +2750,18 @@ xfs_alloc_fix_freelist( break; goto out_agflbp_relse; } + + if (!xfs_rmap_should_skip_owner_update(&targs.oinfo)) { + error = xfs_rmap_alloc(tp, agbp, pag, + targs.agbno, targs.len, &targs.oinfo); + if (error) + goto out_agflbp_relse; + } + error = xfs_alloc_update_counters(tp, agbp, + -((long)(targs.len))); + if (error) + goto out_agflbp_relse; + /* * Put each allocated block on the list. */ @@ -3244,28 +3271,6 @@ xfs_alloc_vextent_set_fsbno( /* * Allocate within a single AG only. */ -static int -__xfs_alloc_vextent_this_ag( - struct xfs_alloc_arg *args) -{ - struct xfs_mount *mp = args->mp; - int error; - - error = xfs_alloc_fix_freelist(args, 0); - if (error) { - trace_xfs_alloc_vextent_nofix(args); - return error; - } - if (!args->agbp) { - /* cannot allocate in this AG at all */ - trace_xfs_alloc_vextent_noagbp(args); - args->agbno = NULLAGBLOCK; - return 0; - } - args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); - return xfs_alloc_ag_vextent(args); -} - static int xfs_alloc_vextent_this_ag( struct xfs_alloc_arg *args, @@ -3289,7 +3294,7 @@ xfs_alloc_vextent_this_ag( } args->pag = xfs_perag_get(mp, args->agno); - error = __xfs_alloc_vextent_this_ag(args); + error = xfs_alloc_ag_vextent(args); xfs_alloc_vextent_set_fsbno(args, minimum_agno); xfs_perag_put(args->pag); @@ -3329,7 +3334,7 @@ xfs_alloc_vextent_iterate_ags( args->agno = start_agno; for (;;) { args->pag = xfs_perag_get(mp, args->agno); - error = __xfs_alloc_vextent_this_ag(args); + error = xfs_alloc_ag_vextent(args); if (error) { args->agbno = NULLAGBLOCK; break; From 74c36a8689d3d8ca9d9e96759c9bbf337e049097 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:53 +1100 Subject: [PATCH 228/765] xfs: use xfs_alloc_vextent_this_ag() where appropriate Change obvious callers of single AG allocation to use xfs_alloc_vextent_this_ag(). Drive the per-ag grabbing out to the callers, too, so that callers with active references don't need to do new lookups just for an allocation in a context that already has a perag reference. The only remaining caller that does single AG allocation through xfs_alloc_vextent() is xfs_bmap_btalloc() with XFS_ALLOCTYPE_NEAR_BNO. That is going to need more untangling before it can be converted cleanly. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_ag.c | 3 +- fs/xfs/libxfs/xfs_alloc.c | 26 ++++++++------- fs/xfs/libxfs/xfs_alloc.h | 6 ++++ fs/xfs/libxfs/xfs_bmap.c | 52 +++++++++++++++++------------- fs/xfs/libxfs/xfs_bmap_btree.c | 22 ++++++------- fs/xfs/libxfs/xfs_ialloc.c | 9 ++++-- fs/xfs/libxfs/xfs_ialloc_btree.c | 3 +- fs/xfs/libxfs/xfs_refcount_btree.c | 3 +- fs/xfs/scrub/repair.c | 3 +- 9 files changed, 74 insertions(+), 53 deletions(-) diff --git a/fs/xfs/libxfs/xfs_ag.c b/fs/xfs/libxfs/xfs_ag.c index a3bdcde958459..053d77a283f74 100644 --- a/fs/xfs/libxfs/xfs_ag.c +++ b/fs/xfs/libxfs/xfs_ag.c @@ -887,6 +887,7 @@ xfs_ag_shrink_space( struct xfs_alloc_arg args = { .tp = *tpp, .mp = mp, + .pag = pag, .type = XFS_ALLOCTYPE_THIS_BNO, .minlen = delta, .maxlen = delta, @@ -938,7 +939,7 @@ xfs_ag_shrink_space( return error; /* internal log shouldn't also show up in the free space btrees */ - error = xfs_alloc_vextent(&args); + error = xfs_alloc_vextent_this_ag(&args); if (!error && args.agbno == NULLAGBLOCK) error = -ENOSPC; diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index edcbf8fd10ea0..7e4452c11421c 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -2723,7 +2723,6 @@ xfs_alloc_fix_freelist( targs.agbp = agbp; targs.agno = args->agno; targs.alignment = targs.minlen = targs.prod = 1; - targs.type = XFS_ALLOCTYPE_THIS_AG; targs.pag = pag; error = xfs_alloc_read_agfl(pag, tp, &agflbp); if (error) @@ -3271,14 +3270,17 @@ xfs_alloc_vextent_set_fsbno( /* * Allocate within a single AG only. */ -static int +int xfs_alloc_vextent_this_ag( - struct xfs_alloc_arg *args, - xfs_agnumber_t minimum_agno) + struct xfs_alloc_arg *args) { struct xfs_mount *mp = args->mp; + xfs_agnumber_t minimum_agno = 0; int error; + if (args->tp->t_highest_agno != NULLAGNUMBER) + minimum_agno = args->tp->t_highest_agno; + error = xfs_alloc_vextent_check_args(args); if (error) { if (error == -ENOSPC) @@ -3293,11 +3295,8 @@ xfs_alloc_vextent_this_ag( return 0; } - args->pag = xfs_perag_get(mp, args->agno); error = xfs_alloc_ag_vextent(args); - xfs_alloc_vextent_set_fsbno(args, minimum_agno); - xfs_perag_put(args->pag); return error; } @@ -3480,6 +3479,7 @@ xfs_alloc_vextent( struct xfs_alloc_arg *args) { xfs_agnumber_t minimum_agno = 0; + int error; if (args->tp->t_highest_agno != NULLAGNUMBER) minimum_agno = args->tp->t_highest_agno; @@ -3488,17 +3488,21 @@ xfs_alloc_vextent( case XFS_ALLOCTYPE_THIS_AG: case XFS_ALLOCTYPE_NEAR_BNO: case XFS_ALLOCTYPE_THIS_BNO: - return xfs_alloc_vextent_this_ag(args, minimum_agno); + args->pag = xfs_perag_get(args->mp, + XFS_FSB_TO_AGNO(args->mp, args->fsbno)); + error = xfs_alloc_vextent_this_ag(args); + xfs_perag_put(args->pag); + break; case XFS_ALLOCTYPE_START_BNO: return xfs_alloc_vextent_start_ag(args, minimum_agno); case XFS_ALLOCTYPE_FIRST_AG: return xfs_alloc_vextent_first_ag(args, minimum_agno); default: + error = -EFSCORRUPTED; ASSERT(0); - /* NOTREACHED */ + break; } - /* Should never get here */ - return -EFSCORRUPTED; + return error; } /* Ensure that the freelist is at full capacity. */ diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h index 2c3f762dfb581..0a9ad6cd18e2e 100644 --- a/fs/xfs/libxfs/xfs_alloc.h +++ b/fs/xfs/libxfs/xfs_alloc.h @@ -124,6 +124,12 @@ int /* error */ xfs_alloc_vextent( xfs_alloc_arg_t *args); /* allocation argument structure */ +/* + * Allocate an extent in the specific AG defined by args->fsbno. If there is no + * space in that AG, then the allocation will fail. + */ +int xfs_alloc_vextent_this_ag(struct xfs_alloc_arg *args); + /* * Free an extent. */ diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 87ad56e034c03..baf11bc4d091a 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -789,6 +789,8 @@ xfs_bmap_local_to_extents( memset(&args, 0, sizeof(args)); args.tp = tp; args.mp = ip->i_mount; + args.total = total; + args.minlen = args.maxlen = args.prod = 1; xfs_rmap_ino_owner(&args.oinfo, ip->i_ino, whichfork, 0); /* * Allocate a block. We know we need only one, since the @@ -3506,8 +3508,7 @@ xfs_bmap_btalloc( xfs_extlen_t orig_length; xfs_extlen_t blen; xfs_extlen_t nextminlen = 0; - int isaligned; - int tryagain; + int isaligned = 0; int error; int stripe_align; @@ -3528,7 +3529,6 @@ xfs_bmap_btalloc( xfs_bmap_adjacent(ap); - tryagain = isaligned = 0; args.fsbno = ap->blkno; args.oinfo = XFS_RMAP_OINFO_SKIP_UPDATE; @@ -3576,9 +3576,9 @@ xfs_bmap_btalloc( * allocation with alignment turned on. */ atype = args.type; - tryagain = 1; args.type = XFS_ALLOCTYPE_THIS_BNO; args.alignment = 1; + /* * Compute the minlen+alignment for the * next case. Set slop so that the value @@ -3595,34 +3595,37 @@ xfs_bmap_btalloc( args.minlen - 1; else args.minalignslop = 0; + + args.pag = xfs_perag_get(mp, + XFS_FSB_TO_AGNO(mp, args.fsbno)); + error = xfs_alloc_vextent_this_ag(&args); + xfs_perag_put(args.pag); + if (error) + return error; + + if (args.fsbno != NULLFSBLOCK) + goto out_success; + /* + * Exact allocation failed. Now try with alignment + * turned on. + */ + args.pag = NULL; + args.type = atype; + args.fsbno = ap->blkno; + args.alignment = stripe_align; + args.minlen = nextminlen; + args.minalignslop = 0; + isaligned = 1; } } else { args.alignment = 1; args.minalignslop = 0; } - args.minleft = ap->minleft; - args.wasdel = ap->wasdel; - args.resv = XFS_AG_RESV_NONE; - args.datatype = ap->datatype; error = xfs_alloc_vextent(&args); if (error) return error; - if (tryagain && args.fsbno == NULLFSBLOCK) { - /* - * Exact allocation failed. Now try with alignment - * turned on. - */ - args.type = atype; - args.fsbno = ap->blkno; - args.alignment = stripe_align; - args.minlen = nextminlen; - args.minalignslop = 0; - isaligned = 1; - if ((error = xfs_alloc_vextent(&args))) - return error; - } if (isaligned && args.fsbno == NULLFSBLOCK) { /* * allocation failed, so turn off alignment and @@ -3650,8 +3653,13 @@ xfs_bmap_btalloc( return error; ap->tp->t_flags |= XFS_TRANS_LOWMODE; } + args.minleft = ap->minleft; + args.wasdel = ap->wasdel; + args.resv = XFS_AG_RESV_NONE; + args.datatype = ap->datatype; if (args.fsbno != NULLFSBLOCK) { +out_success: xfs_bmap_process_allocated_extent(ap, &args, orig_offset, orig_length); } else { diff --git a/fs/xfs/libxfs/xfs_bmap_btree.c b/fs/xfs/libxfs/xfs_bmap_btree.c index afd9b2d962a31..d42c1a1da1fc0 100644 --- a/fs/xfs/libxfs/xfs_bmap_btree.c +++ b/fs/xfs/libxfs/xfs_bmap_btree.c @@ -21,6 +21,7 @@ #include "xfs_quota.h" #include "xfs_trace.h" #include "xfs_rmap.h" +#include "xfs_ag.h" static struct kmem_cache *xfs_bmbt_cur_cache; @@ -200,14 +201,18 @@ xfs_bmbt_alloc_block( union xfs_btree_ptr *new, int *stat) { - xfs_alloc_arg_t args; /* block allocation args */ - int error; /* error return value */ + struct xfs_alloc_arg args; + int error; memset(&args, 0, sizeof(args)); args.tp = cur->bc_tp; args.mp = cur->bc_mp; xfs_rmap_ino_bmbt_owner(&args.oinfo, cur->bc_ino.ip->i_ino, cur->bc_ino.whichfork); + args.minlen = args.maxlen = args.prod = 1; + args.wasdel = cur->bc_ino.flags & XFS_BTCUR_BMBT_WASDEL; + if (!args.wasdel && args.tp->t_blk_res == 0) + return -ENOSPC; args.fsbno = be64_to_cpu(start->l); args.type = XFS_ALLOCTYPE_START_BNO; @@ -222,15 +227,9 @@ xfs_bmbt_alloc_block( args.minleft = xfs_bmapi_minleft(cur->bc_tp, cur->bc_ino.ip, cur->bc_ino.whichfork); - args.minlen = args.maxlen = args.prod = 1; - args.wasdel = cur->bc_ino.flags & XFS_BTCUR_BMBT_WASDEL; - if (!args.wasdel && args.tp->t_blk_res == 0) { - error = -ENOSPC; - goto error0; - } error = xfs_alloc_vextent(&args); if (error) - goto error0; + return error; if (args.fsbno == NULLFSBLOCK && args.minleft) { /* @@ -243,7 +242,7 @@ xfs_bmbt_alloc_block( args.type = XFS_ALLOCTYPE_START_BNO; error = xfs_alloc_vextent(&args); if (error) - goto error0; + return error; cur->bc_tp->t_flags |= XFS_TRANS_LOWMODE; } if (WARN_ON_ONCE(args.fsbno == NULLFSBLOCK)) { @@ -262,9 +261,6 @@ xfs_bmbt_alloc_block( *stat = 1; return 0; - - error0: - return error; } STATIC int diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index 20d2365524a48..74ad88853a8c0 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -630,6 +630,7 @@ xfs_ialloc_ag_alloc( args.mp = tp->t_mountp; args.fsbno = NULLFSBLOCK; args.oinfo = XFS_RMAP_OINFO_INODES; + args.pag = pag; #ifdef DEBUG /* randomly do sparse inode allocations */ @@ -683,7 +684,8 @@ xfs_ialloc_ag_alloc( /* Allow space for the inode btree to split. */ args.minleft = igeo->inobt_maxlevels; - if ((error = xfs_alloc_vextent(&args))) + error = xfs_alloc_vextent_this_ag(&args); + if (error) return error; /* @@ -731,7 +733,8 @@ xfs_ialloc_ag_alloc( * Allow space for the inode btree to split. */ args.minleft = igeo->inobt_maxlevels; - if ((error = xfs_alloc_vextent(&args))) + error = xfs_alloc_vextent_this_ag(&args); + if (error) return error; } @@ -780,7 +783,7 @@ xfs_ialloc_ag_alloc( args.mp->m_sb.sb_inoalignmt) - igeo->ialloc_blks; - error = xfs_alloc_vextent(&args); + error = xfs_alloc_vextent_this_ag(&args); if (error) return error; diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c index 3675a0d293103..fa6cd25029708 100644 --- a/fs/xfs/libxfs/xfs_ialloc_btree.c +++ b/fs/xfs/libxfs/xfs_ialloc_btree.c @@ -103,6 +103,7 @@ __xfs_inobt_alloc_block( memset(&args, 0, sizeof(args)); args.tp = cur->bc_tp; args.mp = cur->bc_mp; + args.pag = cur->bc_ag.pag; args.oinfo = XFS_RMAP_OINFO_INOBT; args.fsbno = XFS_AGB_TO_FSB(args.mp, cur->bc_ag.pag->pag_agno, sbno); args.minlen = 1; @@ -111,7 +112,7 @@ __xfs_inobt_alloc_block( args.type = XFS_ALLOCTYPE_NEAR_BNO; args.resv = resv; - error = xfs_alloc_vextent(&args); + error = xfs_alloc_vextent_this_ag(&args); if (error) return error; diff --git a/fs/xfs/libxfs/xfs_refcount_btree.c b/fs/xfs/libxfs/xfs_refcount_btree.c index d20abf0390fc8..a980fb18bde2c 100644 --- a/fs/xfs/libxfs/xfs_refcount_btree.c +++ b/fs/xfs/libxfs/xfs_refcount_btree.c @@ -67,6 +67,7 @@ xfs_refcountbt_alloc_block( memset(&args, 0, sizeof(args)); args.tp = cur->bc_tp; args.mp = cur->bc_mp; + args.pag = cur->bc_ag.pag; args.type = XFS_ALLOCTYPE_NEAR_BNO; args.fsbno = XFS_AGB_TO_FSB(cur->bc_mp, cur->bc_ag.pag->pag_agno, xfs_refc_block(args.mp)); @@ -74,7 +75,7 @@ xfs_refcountbt_alloc_block( args.minlen = args.maxlen = args.prod = 1; args.resv = XFS_AG_RESV_METADATA; - error = xfs_alloc_vextent(&args); + error = xfs_alloc_vextent_this_ag(&args); if (error) goto out_error; trace_xfs_refcountbt_alloc_block(cur->bc_mp, cur->bc_ag.pag->pag_agno, diff --git a/fs/xfs/scrub/repair.c b/fs/xfs/scrub/repair.c index d0b1644efb896..5f4b50aac4bbe 100644 --- a/fs/xfs/scrub/repair.c +++ b/fs/xfs/scrub/repair.c @@ -326,6 +326,7 @@ xrep_alloc_ag_block( args.tp = sc->tp; args.mp = sc->mp; + args.pag = sc->sa.pag; args.oinfo = *oinfo; args.fsbno = XFS_AGB_TO_FSB(args.mp, sc->sa.pag->pag_agno, 0); args.minlen = 1; @@ -334,7 +335,7 @@ xrep_alloc_ag_block( args.type = XFS_ALLOCTYPE_THIS_AG; args.resv = resv; - error = xfs_alloc_vextent(&args); + error = xfs_alloc_vextent_this_ag(&args); if (error) return error; if (args.fsbno == NULLFSBLOCK) From 85843327094f9de9cf0129cd9a3a43128c6f5ac8 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:53 +1100 Subject: [PATCH 229/765] xfs: factor xfs_bmap_btalloc() There are several different contexts xfs_bmap_btalloc() handles, and large chunks of the code execute independent allocation contexts. Try to untangle this mess a bit. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_bmap.c | 333 +++++++++++++++++++++++---------------- 1 file changed, 196 insertions(+), 137 deletions(-) diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index baf11bc4d091a..a28d57b823962 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -3196,13 +3196,13 @@ xfs_bmap_select_minlen( } } -STATIC int +static int xfs_bmap_btalloc_select_lengths( struct xfs_bmalloca *ap, struct xfs_alloc_arg *args, xfs_extlen_t *blen) { - struct xfs_mount *mp = ap->ip->i_mount; + struct xfs_mount *mp = args->mp; struct xfs_perag *pag; xfs_agnumber_t agno, startag; int notinit = 0; @@ -3216,7 +3216,7 @@ xfs_bmap_btalloc_select_lengths( } args->total = ap->total; - startag = XFS_FSB_TO_AGNO(mp, args->fsbno); + startag = XFS_FSB_TO_AGNO(mp, ap->blkno); if (startag == NULLAGNUMBER) startag = 0; @@ -3258,7 +3258,7 @@ xfs_bmap_btalloc_filestreams( args->type = XFS_ALLOCTYPE_NEAR_BNO; args->total = ap->total; - start_agno = XFS_FSB_TO_AGNO(mp, args->fsbno); + start_agno = XFS_FSB_TO_AGNO(mp, ap->blkno); if (start_agno == NULLAGNUMBER) start_agno = 0; @@ -3496,170 +3496,229 @@ xfs_bmap_exact_minlen_extent_alloc( #endif -STATIC int -xfs_bmap_btalloc( - struct xfs_bmalloca *ap) +/* + * If we are not low on available data blocks and we are allocating at + * EOF, optimise allocation for contiguous file extension and/or stripe + * alignment of the new extent. + * + * NOTE: ap->aeof is only set if the allocation length is >= the + * stripe unit and the allocation offset is at the end of file. + */ +static int +xfs_bmap_btalloc_at_eof( + struct xfs_bmalloca *ap, + struct xfs_alloc_arg *args, + xfs_extlen_t blen, + int stripe_align) { - struct xfs_mount *mp = ap->ip->i_mount; - struct xfs_alloc_arg args = { .tp = ap->tp, .mp = mp }; - xfs_alloctype_t atype = 0; - xfs_agnumber_t ag; - xfs_fileoff_t orig_offset; - xfs_extlen_t orig_length; - xfs_extlen_t blen; - xfs_extlen_t nextminlen = 0; - int isaligned = 0; + struct xfs_mount *mp = args->mp; + xfs_alloctype_t atype; int error; - int stripe_align; - ASSERT(ap->length); - orig_offset = ap->offset; - orig_length = ap->length; + /* + * If there are already extents in the file, try an exact EOF block + * allocation to extend the file as a contiguous extent. If that fails, + * or it's the first allocation in a file, just try for a stripe aligned + * allocation. + */ + if (ap->offset) { + xfs_extlen_t nextminlen = 0; - stripe_align = xfs_bmap_compute_alignments(ap, &args); + atype = args->type; + args->type = XFS_ALLOCTYPE_THIS_BNO; + args->alignment = 1; + /* + * Compute the minlen+alignment for the next case. Set slop so + * that the value of minlen+alignment+slop doesn't go up between + * the calls. + */ + if (blen > stripe_align && blen <= args->maxlen) + nextminlen = blen - stripe_align; + else + nextminlen = args->minlen; + if (nextminlen + stripe_align > args->minlen + 1) + args->minalignslop = nextminlen + stripe_align - + args->minlen - 1; + else + args->minalignslop = 0; + + args->pag = xfs_perag_get(mp, XFS_FSB_TO_AGNO(mp, args->fsbno)); + error = xfs_alloc_vextent_this_ag(args); + xfs_perag_put(args->pag); + if (error) + return error; + + if (args->fsbno != NULLFSBLOCK) + return 0; + /* + * Exact allocation failed. Reset to try an aligned allocation + * according to the original allocation specification. + */ + args->pag = NULL; + args->type = atype; + args->fsbno = ap->blkno; + args->alignment = stripe_align; + args->minlen = nextminlen; + args->minalignslop = 0; + } else { + args->alignment = stripe_align; + atype = args->type; + /* + * Adjust minlen to try and preserve alignment if we + * can't guarantee an aligned maxlen extent. + */ + if (blen > args->alignment && + blen <= args->maxlen + args->alignment) + args->minlen = blen - args->alignment; + args->minalignslop = 0; + } + + error = xfs_alloc_vextent(args); + if (error) + return error; + + if (args->fsbno != NULLFSBLOCK) + return 0; + + /* + * Allocation failed, so turn return the allocation args to their + * original non-aligned state so the caller can proceed on allocation + * failure as if this function was never called. + */ + args->type = atype; + args->fsbno = ap->blkno; + args->alignment = 1; + return 0; +} + +static int +xfs_bmap_btalloc_best_length( + struct xfs_bmalloca *ap, + struct xfs_alloc_arg *args, + int stripe_align) +{ + struct xfs_mount *mp = args->mp; + xfs_extlen_t blen = 0; + int error; + + /* + * Determine the initial block number we will target for allocation. + */ if ((ap->datatype & XFS_ALLOC_USERDATA) && xfs_inode_is_filestream(ap->ip)) { - ag = xfs_filestream_lookup_ag(ap->ip); - ag = (ag != NULLAGNUMBER) ? ag : 0; - ap->blkno = XFS_AGB_TO_FSB(mp, ag, 0); + xfs_agnumber_t agno = xfs_filestream_lookup_ag(ap->ip); + if (agno == NULLAGNUMBER) + agno = 0; + ap->blkno = XFS_AGB_TO_FSB(mp, agno, 0); } else { ap->blkno = XFS_INO_TO_FSB(mp, ap->ip->i_ino); } - xfs_bmap_adjacent(ap); - - args.fsbno = ap->blkno; - args.oinfo = XFS_RMAP_OINFO_SKIP_UPDATE; - - /* Trim the allocation back to the maximum an AG can fit. */ - args.maxlen = min(ap->length, mp->m_ag_max_usable); - blen = 0; + args->fsbno = ap->blkno; /* - * Search for an allocation group with a single extent large - * enough for the request. If one isn't found, then adjust - * the minimum allocation size to the largest space found. + * Search for an allocation group with a single extent large enough for + * the request. If one isn't found, then adjust the minimum allocation + * size to the largest space found. */ if ((ap->datatype & XFS_ALLOC_USERDATA) && xfs_inode_is_filestream(ap->ip)) - error = xfs_bmap_btalloc_filestreams(ap, &args, &blen); + error = xfs_bmap_btalloc_filestreams(ap, args, &blen); else - error = xfs_bmap_btalloc_select_lengths(ap, &args, &blen); + error = xfs_bmap_btalloc_select_lengths(ap, args, &blen); if (error) return error; /* - * If we are not low on available data blocks, and the underlying - * logical volume manager is a stripe, and the file offset is zero then - * try to allocate data blocks on stripe unit boundary. NOTE: ap->aeof - * is only set if the allocation length is >= the stripe unit and the - * allocation offset is at the end of file. + * Don't attempt optimal EOF allocation if previous allocations barely + * succeeded due to being near ENOSPC. It is highly unlikely we'll get + * optimal or even aligned allocations in this case, so don't waste time + * trying. */ - if (!(ap->tp->t_flags & XFS_TRANS_LOWMODE) && ap->aeof) { - if (!ap->offset) { - args.alignment = stripe_align; - atype = args.type; - isaligned = 1; - /* - * Adjust minlen to try and preserve alignment if we - * can't guarantee an aligned maxlen extent. - */ - if (blen > args.alignment && - blen <= args.maxlen + args.alignment) - args.minlen = blen - args.alignment; - args.minalignslop = 0; - } else { - /* - * First try an exact bno allocation. - * If it fails then do a near or start bno - * allocation with alignment turned on. - */ - atype = args.type; - args.type = XFS_ALLOCTYPE_THIS_BNO; - args.alignment = 1; - - /* - * Compute the minlen+alignment for the - * next case. Set slop so that the value - * of minlen+alignment+slop doesn't go up - * between the calls. - */ - if (blen > stripe_align && blen <= args.maxlen) - nextminlen = blen - stripe_align; - else - nextminlen = args.minlen; - if (nextminlen + stripe_align > args.minlen + 1) - args.minalignslop = - nextminlen + stripe_align - - args.minlen - 1; - else - args.minalignslop = 0; - - args.pag = xfs_perag_get(mp, - XFS_FSB_TO_AGNO(mp, args.fsbno)); - error = xfs_alloc_vextent_this_ag(&args); - xfs_perag_put(args.pag); - if (error) - return error; - - if (args.fsbno != NULLFSBLOCK) - goto out_success; - /* - * Exact allocation failed. Now try with alignment - * turned on. - */ - args.pag = NULL; - args.type = atype; - args.fsbno = ap->blkno; - args.alignment = stripe_align; - args.minlen = nextminlen; - args.minalignslop = 0; - isaligned = 1; - } - } else { - args.alignment = 1; - args.minalignslop = 0; + if (ap->aeof && !(ap->tp->t_flags & XFS_TRANS_LOWMODE)) { + error = xfs_bmap_btalloc_at_eof(ap, args, blen, stripe_align); + if (error) + return error; + if (args->fsbno != NULLFSBLOCK) + return 0; } - error = xfs_alloc_vextent(&args); + error = xfs_alloc_vextent(args); if (error) return error; + if (args->fsbno != NULLFSBLOCK) + return 0; - if (isaligned && args.fsbno == NULLFSBLOCK) { - /* - * allocation failed, so turn off alignment and - * try again. - */ - args.type = atype; - args.fsbno = ap->blkno; - args.alignment = 0; - if ((error = xfs_alloc_vextent(&args))) - return error; - } - if (args.fsbno == NULLFSBLOCK && - args.minlen > ap->minlen) { - args.minlen = ap->minlen; - args.type = XFS_ALLOCTYPE_START_BNO; - args.fsbno = ap->blkno; - if ((error = xfs_alloc_vextent(&args))) - return error; - } - if (args.fsbno == NULLFSBLOCK) { - args.fsbno = 0; - args.type = XFS_ALLOCTYPE_FIRST_AG; - args.total = ap->minlen; - if ((error = xfs_alloc_vextent(&args))) + /* + * Try a locality first full filesystem minimum length allocation whilst + * still maintaining necessary total block reservation requirements. + */ + if (args->minlen > ap->minlen) { + args->minlen = ap->minlen; + args->type = XFS_ALLOCTYPE_START_BNO; + args->fsbno = ap->blkno; + error = xfs_alloc_vextent(args); + if (error) return error; - ap->tp->t_flags |= XFS_TRANS_LOWMODE; } - args.minleft = ap->minleft; - args.wasdel = ap->wasdel; - args.resv = XFS_AG_RESV_NONE; - args.datatype = ap->datatype; + if (args->fsbno != NULLFSBLOCK) + return 0; + + /* + * We are now critically low on space, so this is a last resort + * allocation attempt: no reserve, no locality, blocking, minimum + * length, full filesystem free space scan. We also indicate to future + * allocations in this transaction that we are critically low on space + * so they don't waste time on allocation modes that are unlikely to + * succeed. + */ + args->fsbno = 0; + args->type = XFS_ALLOCTYPE_FIRST_AG; + args->total = ap->minlen; + error = xfs_alloc_vextent(args); + if (error) + return error; + ap->tp->t_flags |= XFS_TRANS_LOWMODE; + return 0; +} + +static int +xfs_bmap_btalloc( + struct xfs_bmalloca *ap) +{ + struct xfs_mount *mp = ap->ip->i_mount; + struct xfs_alloc_arg args = { + .tp = ap->tp, + .mp = mp, + .fsbno = NULLFSBLOCK, + .oinfo = XFS_RMAP_OINFO_SKIP_UPDATE, + .minleft = ap->minleft, + .wasdel = ap->wasdel, + .resv = XFS_AG_RESV_NONE, + .datatype = ap->datatype, + .alignment = 1, + .minalignslop = 0, + }; + xfs_fileoff_t orig_offset; + xfs_extlen_t orig_length; + int error; + int stripe_align; + + ASSERT(ap->length); + orig_offset = ap->offset; + orig_length = ap->length; + + stripe_align = xfs_bmap_compute_alignments(ap, &args); + + /* Trim the allocation back to the maximum an AG can fit. */ + args.maxlen = min(ap->length, mp->m_ag_max_usable); + + error = xfs_bmap_btalloc_best_length(ap, &args, stripe_align); + if (error) + return error; if (args.fsbno != NULLFSBLOCK) { -out_success: xfs_bmap_process_allocated_extent(ap, &args, orig_offset, orig_length); } else { From 319c9e874ac8721acdb6583e3459ef595e5ed0a6 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:53 +1100 Subject: [PATCH 230/765] xfs: use xfs_alloc_vextent_first_ag() where appropriate Change obvious callers of single AG allocation to use xfs_alloc_vextent_first_ag(). This gets rid of XFS_ALLOCTYPE_FIRST_AG as the type used within xfs_alloc_vextent_first_ag() during iteration is _THIS_AG. Hence we can remove the setting of args->type from all the callers of _first_ag() and remove the alloctype. While doing this, pass the allocation target fsb as a parameter rather than encoding it in args->fsbno. This starts the process of making args->fsbno an output only variable rather than input/output. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_alloc.c | 33 ++++++++++++++++++--------------- fs/xfs/libxfs/xfs_alloc.h | 10 ++++++++-- fs/xfs/libxfs/xfs_bmap.c | 31 ++++++++++++++++--------------- 3 files changed, 42 insertions(+), 32 deletions(-) diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index 7e4452c11421c..bd77e645398b3 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -3183,7 +3183,8 @@ xfs_alloc_read_agf( */ static int xfs_alloc_vextent_check_args( - struct xfs_alloc_arg *args) + struct xfs_alloc_arg *args, + xfs_fsblock_t target) { struct xfs_mount *mp = args->mp; xfs_agblock_t agsize; @@ -3201,13 +3202,13 @@ xfs_alloc_vextent_check_args( args->maxlen = agsize; if (args->alignment == 0) args->alignment = 1; - ASSERT(XFS_FSB_TO_AGNO(mp, args->fsbno) < mp->m_sb.sb_agcount); - ASSERT(XFS_FSB_TO_AGBNO(mp, args->fsbno) < agsize); + ASSERT(XFS_FSB_TO_AGNO(mp, target) < mp->m_sb.sb_agcount); + ASSERT(XFS_FSB_TO_AGBNO(mp, target) < agsize); ASSERT(args->minlen <= args->maxlen); ASSERT(args->minlen <= agsize); ASSERT(args->mod < args->prod); - if (XFS_FSB_TO_AGNO(mp, args->fsbno) >= mp->m_sb.sb_agcount || - XFS_FSB_TO_AGBNO(mp, args->fsbno) >= agsize || + if (XFS_FSB_TO_AGNO(mp, target) >= mp->m_sb.sb_agcount || + XFS_FSB_TO_AGBNO(mp, target) >= agsize || args->minlen > args->maxlen || args->minlen > agsize || args->mod >= args->prod) { args->fsbno = NULLFSBLOCK; @@ -3281,7 +3282,7 @@ xfs_alloc_vextent_this_ag( if (args->tp->t_highest_agno != NULLAGNUMBER) minimum_agno = args->tp->t_highest_agno; - error = xfs_alloc_vextent_check_args(args); + error = xfs_alloc_vextent_check_args(args, args->fsbno); if (error) { if (error == -ENOSPC) return 0; @@ -3406,7 +3407,7 @@ xfs_alloc_vextent_start_ag( bool bump_rotor = false; int error; - error = xfs_alloc_vextent_check_args(args); + error = xfs_alloc_vextent_check_args(args, args->fsbno); if (error) { if (error == -ENOSPC) return 0; @@ -3444,25 +3445,29 @@ xfs_alloc_vextent_start_ag( * filesystem attempting blocking allocation. This does not wrap or try a second * pass, so will not recurse into AGs lower than indicated by fsbno. */ -static int +int xfs_alloc_vextent_first_ag( struct xfs_alloc_arg *args, - xfs_agnumber_t minimum_agno) -{ + xfs_fsblock_t target) + { struct xfs_mount *mp = args->mp; + xfs_agnumber_t minimum_agno = 0; xfs_agnumber_t start_agno; int error; - error = xfs_alloc_vextent_check_args(args); + if (args->tp->t_highest_agno != NULLAGNUMBER) + minimum_agno = args->tp->t_highest_agno; + + error = xfs_alloc_vextent_check_args(args, target); if (error) { if (error == -ENOSPC) return 0; return error; } - start_agno = max(minimum_agno, XFS_FSB_TO_AGNO(mp, args->fsbno)); - + start_agno = max(minimum_agno, XFS_FSB_TO_AGNO(mp, target)); args->type = XFS_ALLOCTYPE_THIS_AG; + args->fsbno = target; error = xfs_alloc_vextent_iterate_ags(args, minimum_agno, start_agno, 0); xfs_alloc_vextent_set_fsbno(args, minimum_agno); @@ -3495,8 +3500,6 @@ xfs_alloc_vextent( break; case XFS_ALLOCTYPE_START_BNO: return xfs_alloc_vextent_start_ag(args, minimum_agno); - case XFS_ALLOCTYPE_FIRST_AG: - return xfs_alloc_vextent_first_ag(args, minimum_agno); default: error = -EFSCORRUPTED; ASSERT(0); diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h index 0a9ad6cd18e2e..9be46f493340c 100644 --- a/fs/xfs/libxfs/xfs_alloc.h +++ b/fs/xfs/libxfs/xfs_alloc.h @@ -19,7 +19,6 @@ unsigned int xfs_agfl_size(struct xfs_mount *mp); /* * Freespace allocation types. Argument to xfs_alloc_[v]extent. */ -#define XFS_ALLOCTYPE_FIRST_AG 0x02 /* ... start at ag 0 */ #define XFS_ALLOCTYPE_THIS_AG 0x08 /* anywhere in this a.g. */ #define XFS_ALLOCTYPE_START_BNO 0x10 /* near this block else anywhere */ #define XFS_ALLOCTYPE_NEAR_BNO 0x20 /* in this a.g. and near this block */ @@ -29,7 +28,6 @@ unsigned int xfs_agfl_size(struct xfs_mount *mp); typedef unsigned int xfs_alloctype_t; #define XFS_ALLOC_TYPES \ - { XFS_ALLOCTYPE_FIRST_AG, "FIRST_AG" }, \ { XFS_ALLOCTYPE_THIS_AG, "THIS_AG" }, \ { XFS_ALLOCTYPE_START_BNO, "START_BNO" }, \ { XFS_ALLOCTYPE_NEAR_BNO, "NEAR_BNO" }, \ @@ -130,6 +128,14 @@ xfs_alloc_vextent( */ int xfs_alloc_vextent_this_ag(struct xfs_alloc_arg *args); +/* + * Iterate from the AG indicated from args->fsbno through to the end of the + * filesystem attempting blocking allocation. This is for use in last + * resort allocation attempts when everything else has failed. + */ +int xfs_alloc_vextent_first_ag(struct xfs_alloc_arg *args, + xfs_fsblock_t target); + /* * Free an extent. */ diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index a28d57b823962..efca40fea8f12 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -3248,13 +3248,6 @@ xfs_bmap_btalloc_filestreams( int notinit = 0; int error; - if (ap->tp->t_flags & XFS_TRANS_LOWMODE) { - args->type = XFS_ALLOCTYPE_FIRST_AG; - args->total = ap->minlen; - args->minlen = ap->minlen; - return 0; - } - args->type = XFS_ALLOCTYPE_NEAR_BNO; args->total = ap->total; @@ -3462,9 +3455,7 @@ xfs_bmap_exact_minlen_extent_alloc( */ ap->blkno = XFS_AGB_TO_FSB(mp, 0, 0); - args.fsbno = ap->blkno; args.oinfo = XFS_RMAP_OINFO_SKIP_UPDATE; - args.type = XFS_ALLOCTYPE_FIRST_AG; args.minlen = args.maxlen = ap->minlen; args.total = ap->total; @@ -3476,7 +3467,7 @@ xfs_bmap_exact_minlen_extent_alloc( args.resv = XFS_AG_RESV_NONE; args.datatype = ap->datatype; - error = xfs_alloc_vextent(&args); + error = xfs_alloc_vextent_first_ag(&args, ap->blkno); if (error) return error; @@ -3623,10 +3614,21 @@ xfs_bmap_btalloc_best_length( * size to the largest space found. */ if ((ap->datatype & XFS_ALLOC_USERDATA) && - xfs_inode_is_filestream(ap->ip)) + xfs_inode_is_filestream(ap->ip)) { + /* + * If there is very little free space before we start a + * filestreams allocation, we're almost guaranteed to fail to + * find an AG with enough contiguous free space to succeed, so + * just go straight to the low space algorithm. + */ + if (ap->tp->t_flags & XFS_TRANS_LOWMODE) { + args->minlen = ap->minlen; + goto critically_low_space; + } error = xfs_bmap_btalloc_filestreams(ap, args, &blen); - else + } else { error = xfs_bmap_btalloc_select_lengths(ap, args, &blen); + } if (error) return error; @@ -3673,10 +3675,9 @@ xfs_bmap_btalloc_best_length( * so they don't waste time on allocation modes that are unlikely to * succeed. */ - args->fsbno = 0; - args->type = XFS_ALLOCTYPE_FIRST_AG; +critically_low_space: args->total = ap->minlen; - error = xfs_alloc_vextent(args); + error = xfs_alloc_vextent_first_ag(args, 0); if (error) return error; ap->tp->t_flags |= XFS_TRANS_LOWMODE; From 2a7f6d41d8b72412228ede538bdf0e81bf9738f4 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:53 +1100 Subject: [PATCH 231/765] xfs: use xfs_alloc_vextent_start_bno() where appropriate Change obvious callers of single AG allocation to use xfs_alloc_vextent_start_bno(). Callers no long need to specify XFS_ALLOCTYPE_START_BNO, and so the type can be driven inward and removed. While doing this, also pass the allocation target fsb as a parameter rather than encoding it in args->fsbno. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_alloc.c | 24 ++++++++++--------- fs/xfs/libxfs/xfs_alloc.h | 13 ++++++++-- fs/xfs/libxfs/xfs_bmap.c | 43 ++++++++++++++++++++-------------- fs/xfs/libxfs/xfs_bmap_btree.c | 9 ++----- 4 files changed, 51 insertions(+), 38 deletions(-) diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index bd77e645398b3..008b3622b286d 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -3189,7 +3189,6 @@ xfs_alloc_vextent_check_args( struct xfs_mount *mp = args->mp; xfs_agblock_t agsize; - args->otype = args->type; args->agbno = NULLAGBLOCK; /* @@ -3345,7 +3344,7 @@ xfs_alloc_vextent_iterate_ags( trace_xfs_alloc_vextent_loopfailed(args); if (args->agno == start_agno && - args->otype == XFS_ALLOCTYPE_START_BNO) + args->otype == XFS_ALLOCTYPE_NEAR_BNO) args->type = XFS_ALLOCTYPE_THIS_AG; /* @@ -3373,7 +3372,7 @@ xfs_alloc_vextent_iterate_ags( } flags = 0; - if (args->otype == XFS_ALLOCTYPE_START_BNO) { + if (args->otype == XFS_ALLOCTYPE_NEAR_BNO) { args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); args->type = XFS_ALLOCTYPE_NEAR_BNO; } @@ -3396,18 +3395,22 @@ xfs_alloc_vextent_iterate_ags( * otherwise will wrap back to the start AG and run a second blocking pass to * the end of the filesystem. */ -static int +int xfs_alloc_vextent_start_ag( struct xfs_alloc_arg *args, - xfs_agnumber_t minimum_agno) + xfs_fsblock_t target) { struct xfs_mount *mp = args->mp; + xfs_agnumber_t minimum_agno = 0; xfs_agnumber_t start_agno; xfs_agnumber_t rotorstep = xfs_rotorstep; bool bump_rotor = false; int error; - error = xfs_alloc_vextent_check_args(args, args->fsbno); + if (args->tp->t_highest_agno != NULLAGNUMBER) + minimum_agno = args->tp->t_highest_agno; + + error = xfs_alloc_vextent_check_args(args, target); if (error) { if (error == -ENOSPC) return 0; @@ -3416,14 +3419,15 @@ xfs_alloc_vextent_start_ag( if ((args->datatype & XFS_ALLOC_INITIAL_USER_DATA) && xfs_is_inode32(mp)) { - args->fsbno = XFS_AGB_TO_FSB(mp, + target = XFS_AGB_TO_FSB(mp, ((mp->m_agfrotor / rotorstep) % mp->m_sb.sb_agcount), 0); bump_rotor = 1; } - start_agno = max(minimum_agno, XFS_FSB_TO_AGNO(mp, args->fsbno)); - args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); + start_agno = max(minimum_agno, XFS_FSB_TO_AGNO(mp, target)); + args->agbno = XFS_FSB_TO_AGBNO(mp, target); args->type = XFS_ALLOCTYPE_NEAR_BNO; + args->fsbno = target; error = xfs_alloc_vextent_iterate_ags(args, minimum_agno, start_agno, XFS_ALLOC_FLAG_TRYLOCK); @@ -3498,8 +3502,6 @@ xfs_alloc_vextent( error = xfs_alloc_vextent_this_ag(args); xfs_perag_put(args->pag); break; - case XFS_ALLOCTYPE_START_BNO: - return xfs_alloc_vextent_start_ag(args, minimum_agno); default: error = -EFSCORRUPTED; ASSERT(0); diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h index 9be46f493340c..80e2c16f4cde7 100644 --- a/fs/xfs/libxfs/xfs_alloc.h +++ b/fs/xfs/libxfs/xfs_alloc.h @@ -20,7 +20,6 @@ unsigned int xfs_agfl_size(struct xfs_mount *mp); * Freespace allocation types. Argument to xfs_alloc_[v]extent. */ #define XFS_ALLOCTYPE_THIS_AG 0x08 /* anywhere in this a.g. */ -#define XFS_ALLOCTYPE_START_BNO 0x10 /* near this block else anywhere */ #define XFS_ALLOCTYPE_NEAR_BNO 0x20 /* in this a.g. and near this block */ #define XFS_ALLOCTYPE_THIS_BNO 0x40 /* at exactly this block */ @@ -29,7 +28,6 @@ typedef unsigned int xfs_alloctype_t; #define XFS_ALLOC_TYPES \ { XFS_ALLOCTYPE_THIS_AG, "THIS_AG" }, \ - { XFS_ALLOCTYPE_START_BNO, "START_BNO" }, \ { XFS_ALLOCTYPE_NEAR_BNO, "NEAR_BNO" }, \ { XFS_ALLOCTYPE_THIS_BNO, "THIS_BNO" } @@ -128,6 +126,17 @@ xfs_alloc_vextent( */ int xfs_alloc_vextent_this_ag(struct xfs_alloc_arg *args); +/* + * Best effort full filesystem allocation scan. + * + * Locality aware allocation will be attempted in the initial AG, but on failure + * non-localised attempts will be made. The AGs are constrained by previous + * allocations in the current transaction. Two passes will be made - the first + * non-blocking, the second blocking. + */ +int xfs_alloc_vextent_start_ag(struct xfs_alloc_arg *args, + xfs_fsblock_t target); + /* * Iterate from the AG indicated from args->fsbno through to the end of the * filesystem attempting blocking allocation. This is for use in last diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index efca40fea8f12..94826f35fdaee 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -646,12 +646,11 @@ xfs_bmap_extents_to_btree( args.mp = mp; xfs_rmap_ino_bmbt_owner(&args.oinfo, ip->i_ino, whichfork); - args.type = XFS_ALLOCTYPE_START_BNO; - args.fsbno = XFS_INO_TO_FSB(mp, ip->i_ino); args.minlen = args.maxlen = args.prod = 1; args.wasdel = wasdel; *logflagsp = 0; - error = xfs_alloc_vextent(&args); + error = xfs_alloc_vextent_start_ag(&args, + XFS_INO_TO_FSB(mp, ip->i_ino)); if (error) goto out_root_realloc; @@ -792,15 +791,15 @@ xfs_bmap_local_to_extents( args.total = total; args.minlen = args.maxlen = args.prod = 1; xfs_rmap_ino_owner(&args.oinfo, ip->i_ino, whichfork, 0); + /* * Allocate a block. We know we need only one, since the * file currently fits in an inode. */ - args.fsbno = XFS_INO_TO_FSB(args.mp, ip->i_ino); - args.type = XFS_ALLOCTYPE_START_BNO; args.total = total; args.minlen = args.maxlen = args.prod = 1; - error = xfs_alloc_vextent(&args); + error = xfs_alloc_vextent_start_ag(&args, + XFS_INO_TO_FSB(args.mp, ip->i_ino)); if (error) goto done; @@ -3208,7 +3207,6 @@ xfs_bmap_btalloc_select_lengths( int notinit = 0; int error = 0; - args->type = XFS_ALLOCTYPE_START_BNO; if (ap->tp->t_flags & XFS_TRANS_LOWMODE) { args->total = ap->minlen; args->minlen = ap->minlen; @@ -3500,7 +3498,8 @@ xfs_bmap_btalloc_at_eof( struct xfs_bmalloca *ap, struct xfs_alloc_arg *args, xfs_extlen_t blen, - int stripe_align) + int stripe_align, + bool ag_only) { struct xfs_mount *mp = args->mp; xfs_alloctype_t atype; @@ -3565,7 +3564,10 @@ xfs_bmap_btalloc_at_eof( args->minalignslop = 0; } - error = xfs_alloc_vextent(args); + if (ag_only) + error = xfs_alloc_vextent(args); + else + error = xfs_alloc_vextent_start_ag(args, ap->blkno); if (error) return error; @@ -3591,13 +3593,17 @@ xfs_bmap_btalloc_best_length( { struct xfs_mount *mp = args->mp; xfs_extlen_t blen = 0; + bool is_filestream = false; int error; + if ((ap->datatype & XFS_ALLOC_USERDATA) && + xfs_inode_is_filestream(ap->ip)) + is_filestream = true; + /* * Determine the initial block number we will target for allocation. */ - if ((ap->datatype & XFS_ALLOC_USERDATA) && - xfs_inode_is_filestream(ap->ip)) { + if (is_filestream) { xfs_agnumber_t agno = xfs_filestream_lookup_ag(ap->ip); if (agno == NULLAGNUMBER) agno = 0; @@ -3613,8 +3619,7 @@ xfs_bmap_btalloc_best_length( * the request. If one isn't found, then adjust the minimum allocation * size to the largest space found. */ - if ((ap->datatype & XFS_ALLOC_USERDATA) && - xfs_inode_is_filestream(ap->ip)) { + if (is_filestream) { /* * If there is very little free space before we start a * filestreams allocation, we're almost guaranteed to fail to @@ -3639,14 +3644,18 @@ xfs_bmap_btalloc_best_length( * trying. */ if (ap->aeof && !(ap->tp->t_flags & XFS_TRANS_LOWMODE)) { - error = xfs_bmap_btalloc_at_eof(ap, args, blen, stripe_align); + error = xfs_bmap_btalloc_at_eof(ap, args, blen, stripe_align, + is_filestream); if (error) return error; if (args->fsbno != NULLFSBLOCK) return 0; } - error = xfs_alloc_vextent(args); + if (is_filestream) + error = xfs_alloc_vextent(args); + else + error = xfs_alloc_vextent_start_ag(args, ap->blkno); if (error) return error; if (args->fsbno != NULLFSBLOCK) @@ -3658,9 +3667,7 @@ xfs_bmap_btalloc_best_length( */ if (args->minlen > ap->minlen) { args->minlen = ap->minlen; - args->type = XFS_ALLOCTYPE_START_BNO; - args->fsbno = ap->blkno; - error = xfs_alloc_vextent(args); + error = xfs_alloc_vextent_start_ag(args, ap->blkno); if (error) return error; } diff --git a/fs/xfs/libxfs/xfs_bmap_btree.c b/fs/xfs/libxfs/xfs_bmap_btree.c index d42c1a1da1fc0..b8ad95050c9bd 100644 --- a/fs/xfs/libxfs/xfs_bmap_btree.c +++ b/fs/xfs/libxfs/xfs_bmap_btree.c @@ -214,9 +214,6 @@ xfs_bmbt_alloc_block( if (!args.wasdel && args.tp->t_blk_res == 0) return -ENOSPC; - args.fsbno = be64_to_cpu(start->l); - args.type = XFS_ALLOCTYPE_START_BNO; - /* * If we are coming here from something like unwritten extent * conversion, there has been no data extent allocation already done, so @@ -227,7 +224,7 @@ xfs_bmbt_alloc_block( args.minleft = xfs_bmapi_minleft(cur->bc_tp, cur->bc_ino.ip, cur->bc_ino.whichfork); - error = xfs_alloc_vextent(&args); + error = xfs_alloc_vextent_start_ag(&args, be64_to_cpu(start->l)); if (error) return error; @@ -237,10 +234,8 @@ xfs_bmbt_alloc_block( * a full btree split. Try again and if * successful activate the lowspace algorithm. */ - args.fsbno = 0; args.minleft = 0; - args.type = XFS_ALLOCTYPE_START_BNO; - error = xfs_alloc_vextent(&args); + error = xfs_alloc_vextent_start_ag(&args, 0); if (error) return error; cur->bc_tp->t_flags |= XFS_TRANS_LOWMODE; From db4710fd12248e5d4c3842520cd13f034136576b Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:54 +1100 Subject: [PATCH 232/765] xfs: introduce xfs_alloc_vextent_near_bno() The remaining callers of xfs_alloc_vextent() are all doing NEAR_BNO allocations. We can replace that function with a new xfs_alloc_vextent_near_bno() function that does this explicitly. We also multiplex NEAR_BNO allocations through xfs_alloc_vextent_this_ag via args->type. Replace all of these with direct calls to xfs_alloc_vextent_near_bno(), too. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_alloc.c | 50 ++++++++++++++++++------------ fs/xfs/libxfs/xfs_alloc.h | 14 ++++----- fs/xfs/libxfs/xfs_bmap.c | 6 ++-- fs/xfs/libxfs/xfs_ialloc.c | 27 ++++++---------- fs/xfs/libxfs/xfs_ialloc_btree.c | 5 ++- fs/xfs/libxfs/xfs_refcount_btree.c | 7 ++--- 6 files changed, 55 insertions(+), 54 deletions(-) diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index 008b3622b286d..85e3b65286ac3 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -3479,35 +3479,47 @@ xfs_alloc_vextent_first_ag( } /* - * Allocate an extent (variable-size). - * Depending on the allocation type, we either look in a single allocation - * group or loop over the allocation groups to find the result. + * Allocate an extent as close to the target as possible. If there are not + * viable candidates in the AG, then fail the allocation. */ int -xfs_alloc_vextent( - struct xfs_alloc_arg *args) +xfs_alloc_vextent_near_bno( + struct xfs_alloc_arg *args, + xfs_fsblock_t target) { + struct xfs_mount *mp = args->mp; + bool need_pag = !args->pag; xfs_agnumber_t minimum_agno = 0; int error; if (args->tp->t_highest_agno != NULLAGNUMBER) minimum_agno = args->tp->t_highest_agno; - switch (args->type) { - case XFS_ALLOCTYPE_THIS_AG: - case XFS_ALLOCTYPE_NEAR_BNO: - case XFS_ALLOCTYPE_THIS_BNO: - args->pag = xfs_perag_get(args->mp, - XFS_FSB_TO_AGNO(args->mp, args->fsbno)); - error = xfs_alloc_vextent_this_ag(args); - xfs_perag_put(args->pag); - break; - default: - error = -EFSCORRUPTED; - ASSERT(0); - break; + error = xfs_alloc_vextent_check_args(args, target); + if (error) { + if (error == -ENOSPC) + return 0; + return error; } - return error; + + args->agno = XFS_FSB_TO_AGNO(mp, target); + if (minimum_agno > args->agno) { + trace_xfs_alloc_vextent_skip_deadlock(args); + return 0; + } + + args->agbno = XFS_FSB_TO_AGBNO(mp, target); + args->type = XFS_ALLOCTYPE_NEAR_BNO; + if (need_pag) + args->pag = xfs_perag_get(args->mp, args->agno); + error = xfs_alloc_ag_vextent(args); + if (need_pag) + xfs_perag_put(args->pag); + if (error) + return error; + + xfs_alloc_vextent_set_fsbno(args, minimum_agno); + return 0; } /* Ensure that the freelist is at full capacity. */ diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h index 80e2c16f4cde7..45a428e770f07 100644 --- a/fs/xfs/libxfs/xfs_alloc.h +++ b/fs/xfs/libxfs/xfs_alloc.h @@ -113,19 +113,19 @@ xfs_alloc_log_agf( struct xfs_buf *bp, /* buffer for a.g. freelist header */ uint32_t fields);/* mask of fields to be logged (XFS_AGF_...) */ -/* - * Allocate an extent (variable-size). - */ -int /* error */ -xfs_alloc_vextent( - xfs_alloc_arg_t *args); /* allocation argument structure */ - /* * Allocate an extent in the specific AG defined by args->fsbno. If there is no * space in that AG, then the allocation will fail. */ int xfs_alloc_vextent_this_ag(struct xfs_alloc_arg *args); +/* + * Allocate an extent as close to the target as possible. If there are not + * viable candidates in the AG, then fail the allocation. + */ +int xfs_alloc_vextent_near_bno(struct xfs_alloc_arg *args, + xfs_fsblock_t target); + /* * Best effort full filesystem allocation scan. * diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 94826f35fdaee..da5809d3d004f 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -3246,7 +3246,6 @@ xfs_bmap_btalloc_filestreams( int notinit = 0; int error; - args->type = XFS_ALLOCTYPE_NEAR_BNO; args->total = ap->total; start_agno = XFS_FSB_TO_AGNO(mp, ap->blkno); @@ -3565,7 +3564,7 @@ xfs_bmap_btalloc_at_eof( } if (ag_only) - error = xfs_alloc_vextent(args); + error = xfs_alloc_vextent_near_bno(args, ap->blkno); else error = xfs_alloc_vextent_start_ag(args, ap->blkno); if (error) @@ -3612,7 +3611,6 @@ xfs_bmap_btalloc_best_length( ap->blkno = XFS_INO_TO_FSB(mp, ap->ip->i_ino); } xfs_bmap_adjacent(ap); - args->fsbno = ap->blkno; /* * Search for an allocation group with a single extent large enough for @@ -3653,7 +3651,7 @@ xfs_bmap_btalloc_best_length( } if (is_filestream) - error = xfs_alloc_vextent(args); + error = xfs_alloc_vextent_near_bno(args, ap->blkno); else error = xfs_alloc_vextent_start_ag(args, ap->blkno); if (error) diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index 74ad88853a8c0..b85f038de9e35 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -717,23 +717,17 @@ xfs_ialloc_ag_alloc( isaligned = 1; } else args.alignment = igeo->cluster_align; - /* - * Need to figure out where to allocate the inode blocks. - * Ideally they should be spaced out through the a.g. - * For now, just allocate blocks up front. - */ - args.agbno = be32_to_cpu(agi->agi_root); - args.fsbno = XFS_AGB_TO_FSB(args.mp, pag->pag_agno, args.agbno); /* * Allocate a fixed-size extent of inodes. */ - args.type = XFS_ALLOCTYPE_NEAR_BNO; args.prod = 1; /* * Allow space for the inode btree to split. */ args.minleft = igeo->inobt_maxlevels; - error = xfs_alloc_vextent_this_ag(&args); + error = xfs_alloc_vextent_near_bno(&args, + XFS_AGB_TO_FSB(args.mp, pag->pag_agno, + be32_to_cpu(agi->agi_root))); if (error) return error; } @@ -743,11 +737,11 @@ xfs_ialloc_ag_alloc( * alignment. */ if (isaligned && args.fsbno == NULLFSBLOCK) { - args.type = XFS_ALLOCTYPE_NEAR_BNO; - args.agbno = be32_to_cpu(agi->agi_root); - args.fsbno = XFS_AGB_TO_FSB(args.mp, pag->pag_agno, args.agbno); args.alignment = igeo->cluster_align; - if ((error = xfs_alloc_vextent(&args))) + error = xfs_alloc_vextent_near_bno(&args, + XFS_AGB_TO_FSB(args.mp, pag->pag_agno, + be32_to_cpu(agi->agi_root))); + if (error) return error; } @@ -759,9 +753,6 @@ xfs_ialloc_ag_alloc( igeo->ialloc_min_blks < igeo->ialloc_blks && args.fsbno == NULLFSBLOCK) { sparse_alloc: - args.type = XFS_ALLOCTYPE_NEAR_BNO; - args.agbno = be32_to_cpu(agi->agi_root); - args.fsbno = XFS_AGB_TO_FSB(args.mp, pag->pag_agno, args.agbno); args.alignment = args.mp->m_sb.sb_spino_align; args.prod = 1; @@ -783,7 +774,9 @@ xfs_ialloc_ag_alloc( args.mp->m_sb.sb_inoalignmt) - igeo->ialloc_blks; - error = xfs_alloc_vextent_this_ag(&args); + error = xfs_alloc_vextent_near_bno(&args, + XFS_AGB_TO_FSB(args.mp, pag->pag_agno, + be32_to_cpu(agi->agi_root))); if (error) return error; diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c index fa6cd25029708..9b28211d5a4c5 100644 --- a/fs/xfs/libxfs/xfs_ialloc_btree.c +++ b/fs/xfs/libxfs/xfs_ialloc_btree.c @@ -105,14 +105,13 @@ __xfs_inobt_alloc_block( args.mp = cur->bc_mp; args.pag = cur->bc_ag.pag; args.oinfo = XFS_RMAP_OINFO_INOBT; - args.fsbno = XFS_AGB_TO_FSB(args.mp, cur->bc_ag.pag->pag_agno, sbno); args.minlen = 1; args.maxlen = 1; args.prod = 1; - args.type = XFS_ALLOCTYPE_NEAR_BNO; args.resv = resv; - error = xfs_alloc_vextent_this_ag(&args); + error = xfs_alloc_vextent_near_bno(&args, + XFS_AGB_TO_FSB(args.mp, args.pag->pag_agno, sbno)); if (error) return error; diff --git a/fs/xfs/libxfs/xfs_refcount_btree.c b/fs/xfs/libxfs/xfs_refcount_btree.c index a980fb18bde2c..f3b860970b260 100644 --- a/fs/xfs/libxfs/xfs_refcount_btree.c +++ b/fs/xfs/libxfs/xfs_refcount_btree.c @@ -68,14 +68,13 @@ xfs_refcountbt_alloc_block( args.tp = cur->bc_tp; args.mp = cur->bc_mp; args.pag = cur->bc_ag.pag; - args.type = XFS_ALLOCTYPE_NEAR_BNO; - args.fsbno = XFS_AGB_TO_FSB(cur->bc_mp, cur->bc_ag.pag->pag_agno, - xfs_refc_block(args.mp)); args.oinfo = XFS_RMAP_OINFO_REFC; args.minlen = args.maxlen = args.prod = 1; args.resv = XFS_AG_RESV_METADATA; - error = xfs_alloc_vextent_this_ag(&args); + error = xfs_alloc_vextent_near_bno(&args, + XFS_AGB_TO_FSB(args.mp, args.pag->pag_agno, + xfs_refc_block(args.mp))); if (error) goto out_error; trace_xfs_refcountbt_alloc_block(cur->bc_mp, cur->bc_ag.pag->pag_agno, From 5f36b2ce79f254dd00cdc88374271df7ce843d56 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:54 +1100 Subject: [PATCH 233/765] xfs: introduce xfs_alloc_vextent_exact_bno() Two of the callers to xfs_alloc_vextent_this_ag() actually want exact block number allocation, not anywhere-in-ag allocation. Split this out from _this_ag() as a first class citizen so no external extent allocation code needs to care about args->type anymore. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_ag.c | 6 ++-- fs/xfs/libxfs/xfs_alloc.c | 63 ++++++++++++++++++++++++++++++++------ fs/xfs/libxfs/xfs_alloc.h | 13 ++++++-- fs/xfs/libxfs/xfs_bmap.c | 6 ++-- fs/xfs/libxfs/xfs_ialloc.c | 6 ++-- fs/xfs/scrub/repair.c | 4 +-- 6 files changed, 72 insertions(+), 26 deletions(-) diff --git a/fs/xfs/libxfs/xfs_ag.c b/fs/xfs/libxfs/xfs_ag.c index 053d77a283f74..86696a1c6891b 100644 --- a/fs/xfs/libxfs/xfs_ag.c +++ b/fs/xfs/libxfs/xfs_ag.c @@ -888,7 +888,6 @@ xfs_ag_shrink_space( .tp = *tpp, .mp = mp, .pag = pag, - .type = XFS_ALLOCTYPE_THIS_BNO, .minlen = delta, .maxlen = delta, .oinfo = XFS_RMAP_OINFO_SKIP_UPDATE, @@ -920,8 +919,6 @@ xfs_ag_shrink_space( if (delta >= aglen) return -EINVAL; - args.fsbno = XFS_AGB_TO_FSB(mp, pag->pag_agno, aglen - delta); - /* * Make sure that the last inode cluster cannot overlap with the new * end of the AG, even if it's sparse. @@ -939,7 +936,8 @@ xfs_ag_shrink_space( return error; /* internal log shouldn't also show up in the free space btrees */ - error = xfs_alloc_vextent_this_ag(&args); + error = xfs_alloc_vextent_exact_bno(&args, + XFS_AGB_TO_FSB(mp, pag->pag_agno, aglen - delta)); if (!error && args.agbno == NULLAGBLOCK) error = -ENOSPC; diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index 85e3b65286ac3..5093fa856f355 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -3272,28 +3272,34 @@ xfs_alloc_vextent_set_fsbno( */ int xfs_alloc_vextent_this_ag( - struct xfs_alloc_arg *args) + struct xfs_alloc_arg *args, + xfs_agnumber_t agno) { struct xfs_mount *mp = args->mp; xfs_agnumber_t minimum_agno = 0; + xfs_rfsblock_t target = XFS_AGB_TO_FSB(mp, agno, 0); int error; if (args->tp->t_highest_agno != NULLAGNUMBER) minimum_agno = args->tp->t_highest_agno; - error = xfs_alloc_vextent_check_args(args, args->fsbno); + if (minimum_agno > agno) { + trace_xfs_alloc_vextent_skip_deadlock(args); + args->fsbno = NULLFSBLOCK; + return 0; + } + + error = xfs_alloc_vextent_check_args(args, target); if (error) { if (error == -ENOSPC) return 0; return error; } - args->agno = XFS_FSB_TO_AGNO(mp, args->fsbno); - if (minimum_agno > args->agno) { - trace_xfs_alloc_vextent_skip_deadlock(args); - args->fsbno = NULLFSBLOCK; - return 0; - } + args->agno = agno; + args->agbno = 0; + args->fsbno = target; + args->type = XFS_ALLOCTYPE_THIS_AG; error = xfs_alloc_ag_vextent(args); xfs_alloc_vextent_set_fsbno(args, minimum_agno); @@ -3472,12 +3478,51 @@ xfs_alloc_vextent_first_ag( start_agno = max(minimum_agno, XFS_FSB_TO_AGNO(mp, target)); args->type = XFS_ALLOCTYPE_THIS_AG; args->fsbno = target; - error = xfs_alloc_vextent_iterate_ags(args, minimum_agno, + error = xfs_alloc_vextent_iterate_ags(args, minimum_agno, start_agno, 0); xfs_alloc_vextent_set_fsbno(args, minimum_agno); return error; } +/* + * Allocate within a single AG only. + */ +int +xfs_alloc_vextent_exact_bno( + struct xfs_alloc_arg *args, + xfs_fsblock_t target) +{ + struct xfs_mount *mp = args->mp; + xfs_agnumber_t minimum_agno = 0; + int error; + + if (args->tp->t_highest_agno != NULLAGNUMBER) + minimum_agno = args->tp->t_highest_agno; + + error = xfs_alloc_vextent_check_args(args, target); + if (error) { + if (error == -ENOSPC) + return 0; + return error; + } + + args->agno = XFS_FSB_TO_AGNO(mp, target); + if (minimum_agno > args->agno) { + trace_xfs_alloc_vextent_skip_deadlock(args); + return 0; + } + + args->agbno = XFS_FSB_TO_AGBNO(mp, target); + args->fsbno = target; + args->type = XFS_ALLOCTYPE_THIS_BNO; + error = xfs_alloc_ag_vextent(args); + if (error) + return error; + + xfs_alloc_vextent_set_fsbno(args, minimum_agno); + return 0; +} + /* * Allocate an extent as close to the target as possible. If there are not * viable candidates in the AG, then fail the allocation. diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h index 45a428e770f07..63d5ad4ed8a3c 100644 --- a/fs/xfs/libxfs/xfs_alloc.h +++ b/fs/xfs/libxfs/xfs_alloc.h @@ -114,10 +114,10 @@ xfs_alloc_log_agf( uint32_t fields);/* mask of fields to be logged (XFS_AGF_...) */ /* - * Allocate an extent in the specific AG defined by args->fsbno. If there is no - * space in that AG, then the allocation will fail. + * Allocate an extent anywhere in the specific AG given. If there is no + * space matching the requirements in that AG, then the allocation will fail. */ -int xfs_alloc_vextent_this_ag(struct xfs_alloc_arg *args); +int xfs_alloc_vextent_this_ag(struct xfs_alloc_arg *args, xfs_agnumber_t agno); /* * Allocate an extent as close to the target as possible. If there are not @@ -126,6 +126,13 @@ int xfs_alloc_vextent_this_ag(struct xfs_alloc_arg *args); int xfs_alloc_vextent_near_bno(struct xfs_alloc_arg *args, xfs_fsblock_t target); +/* + * Allocate an extent exactly at the target given. If this is not possible + * then the allocation fails. + */ +int xfs_alloc_vextent_exact_bno(struct xfs_alloc_arg *args, + xfs_fsblock_t target); + /* * Best effort full filesystem allocation scan. * diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index da5809d3d004f..c507645f30313 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -3514,7 +3514,6 @@ xfs_bmap_btalloc_at_eof( xfs_extlen_t nextminlen = 0; atype = args->type; - args->type = XFS_ALLOCTYPE_THIS_BNO; args->alignment = 1; /* @@ -3532,8 +3531,8 @@ xfs_bmap_btalloc_at_eof( else args->minalignslop = 0; - args->pag = xfs_perag_get(mp, XFS_FSB_TO_AGNO(mp, args->fsbno)); - error = xfs_alloc_vextent_this_ag(args); + args->pag = xfs_perag_get(mp, XFS_FSB_TO_AGNO(mp, ap->blkno)); + error = xfs_alloc_vextent_exact_bno(args, ap->blkno); xfs_perag_put(args->pag); if (error) return error; @@ -3546,7 +3545,6 @@ xfs_bmap_btalloc_at_eof( */ args->pag = NULL; args->type = atype; - args->fsbno = ap->blkno; args->alignment = stripe_align; args->minlen = nextminlen; args->minalignslop = 0; diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index b85f038de9e35..82ea6aa5af107 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -662,8 +662,6 @@ xfs_ialloc_ag_alloc( goto sparse_alloc; if (likely(newino != NULLAGINO && (args.agbno < be32_to_cpu(agi->agi_length)))) { - args.fsbno = XFS_AGB_TO_FSB(args.mp, pag->pag_agno, args.agbno); - args.type = XFS_ALLOCTYPE_THIS_BNO; args.prod = 1; /* @@ -684,7 +682,9 @@ xfs_ialloc_ag_alloc( /* Allow space for the inode btree to split. */ args.minleft = igeo->inobt_maxlevels; - error = xfs_alloc_vextent_this_ag(&args); + error = xfs_alloc_vextent_exact_bno(&args, + XFS_AGB_TO_FSB(args.mp, pag->pag_agno, + args.agbno)); if (error) return error; diff --git a/fs/xfs/scrub/repair.c b/fs/xfs/scrub/repair.c index 5f4b50aac4bbe..1b71174ec0d6d 100644 --- a/fs/xfs/scrub/repair.c +++ b/fs/xfs/scrub/repair.c @@ -328,14 +328,12 @@ xrep_alloc_ag_block( args.mp = sc->mp; args.pag = sc->sa.pag; args.oinfo = *oinfo; - args.fsbno = XFS_AGB_TO_FSB(args.mp, sc->sa.pag->pag_agno, 0); args.minlen = 1; args.maxlen = 1; args.prod = 1; - args.type = XFS_ALLOCTYPE_THIS_AG; args.resv = resv; - error = xfs_alloc_vextent_this_ag(&args); + error = xfs_alloc_vextent_this_ag(&args, sc->sa.pag->pag_agno); if (error) return error; if (args.fsbno == NULLFSBLOCK) From 74b9aa63193b288191d6f01b61c819cef2807138 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:54 +1100 Subject: [PATCH 234/765] xfs: introduce xfs_alloc_vextent_prepare() Now that we have wrapper functions for each type of allocation we can ask for, we can start unravelling xfs_alloc_ag_vextent(). That is essentially just a prepare stage, the allocation multiplexer and a post-allocation accounting step is the allocation proceeded. The current xfs_alloc_vextent*() wrappers all have a prepare stage, the allocation operation and a post-allocation accounting step. We can consolidate this by moving the AG alloc prep code into the wrapper functions, the accounting code in the wrapper accounting functions, and cut out the multiplexer layer entirely. This patch consolidates the AG preparation stage. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_alloc.c | 120 ++++++++++++++++++++++++-------------- 1 file changed, 76 insertions(+), 44 deletions(-) diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index 5093fa856f355..dc1b6c6e1de7c 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -1144,31 +1144,8 @@ static int xfs_alloc_ag_vextent( struct xfs_alloc_arg *args) { - struct xfs_mount *mp = args->mp; int error = 0; - ASSERT(args->minlen > 0); - ASSERT(args->maxlen > 0); - ASSERT(args->minlen <= args->maxlen); - ASSERT(args->mod < args->prod); - ASSERT(args->alignment > 0); - ASSERT(args->resv != XFS_AG_RESV_AGFL); - - - error = xfs_alloc_fix_freelist(args, 0); - if (error) { - trace_xfs_alloc_vextent_nofix(args); - return error; - } - if (!args->agbp) { - /* cannot allocate in this AG at all */ - trace_xfs_alloc_vextent_noagbp(args); - args->agbno = NULLAGBLOCK; - return 0; - } - args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); - args->wasfromfl = 0; - /* * Branch to correct routine based on the type. */ @@ -3201,11 +3178,18 @@ xfs_alloc_vextent_check_args( args->maxlen = agsize; if (args->alignment == 0) args->alignment = 1; + + ASSERT(args->minlen > 0); + ASSERT(args->maxlen > 0); + ASSERT(args->alignment > 0); + ASSERT(args->resv != XFS_AG_RESV_AGFL); + ASSERT(XFS_FSB_TO_AGNO(mp, target) < mp->m_sb.sb_agcount); ASSERT(XFS_FSB_TO_AGBNO(mp, target) < agsize); ASSERT(args->minlen <= args->maxlen); ASSERT(args->minlen <= agsize); ASSERT(args->mod < args->prod); + if (XFS_FSB_TO_AGNO(mp, target) >= mp->m_sb.sb_agcount || XFS_FSB_TO_AGBNO(mp, target) >= agsize || args->minlen > args->maxlen || args->minlen > agsize || @@ -3217,6 +3201,41 @@ xfs_alloc_vextent_check_args( return 0; } +/* + * Prepare an AG for allocation. If the AG is not prepared to accept the + * allocation, return failure. + * + * XXX(dgc): The complexity of "need_pag" will go away as all caller paths are + * modified to hold their own perag references. + */ +static int +xfs_alloc_vextent_prepare_ag( + struct xfs_alloc_arg *args) +{ + bool need_pag = !args->pag; + int error; + + if (need_pag) + args->pag = xfs_perag_get(args->mp, args->agno); + + error = xfs_alloc_fix_freelist(args, 0); + if (error) { + trace_xfs_alloc_vextent_nofix(args); + if (need_pag) + xfs_perag_put(args->pag); + args->agbno = NULLAGBLOCK; + return error; + } + if (!args->agbp) { + /* cannot allocate in this AG at all */ + trace_xfs_alloc_vextent_noagbp(args); + args->agbno = NULLAGBLOCK; + return 0; + } + args->wasfromfl = 0; + return 0; +} + /* * Post-process allocation results to set the allocated block number correctly * for the caller. @@ -3268,7 +3287,8 @@ xfs_alloc_vextent_set_fsbno( } /* - * Allocate within a single AG only. + * Allocate within a single AG only. Caller is expected to hold a + * perag reference in args->pag. */ int xfs_alloc_vextent_this_ag( @@ -3301,7 +3321,10 @@ xfs_alloc_vextent_this_ag( args->fsbno = target; args->type = XFS_ALLOCTYPE_THIS_AG; - error = xfs_alloc_ag_vextent(args); + error = xfs_alloc_vextent_prepare_ag(args); + if (!error && args->agbp) + error = xfs_alloc_ag_vextent(args); + xfs_alloc_vextent_set_fsbno(args, minimum_agno); return error; } @@ -3339,13 +3362,19 @@ xfs_alloc_vextent_iterate_ags( args->agno = start_agno; for (;;) { args->pag = xfs_perag_get(mp, args->agno); - error = xfs_alloc_ag_vextent(args); - if (error) { - args->agbno = NULLAGBLOCK; + args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); + error = xfs_alloc_vextent_prepare_ag(args); + if (error) break; - } - if (args->agbp) + + if (args->agbp) { + /* + * Allocation is supposed to succeed now, so break out + * of the loop regardless of whether we succeed or not. + */ + error = xfs_alloc_ag_vextent(args); break; + } trace_xfs_alloc_vextent_loopfailed(args); @@ -3378,10 +3407,8 @@ xfs_alloc_vextent_iterate_ags( } flags = 0; - if (args->otype == XFS_ALLOCTYPE_NEAR_BNO) { - args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); + if (args->otype == XFS_ALLOCTYPE_NEAR_BNO) args->type = XFS_ALLOCTYPE_NEAR_BNO; - } } xfs_perag_put(args->pag); args->pag = NULL; @@ -3485,7 +3512,8 @@ xfs_alloc_vextent_first_ag( } /* - * Allocate within a single AG only. + * Allocate at the exact block target or fail. Caller is expected to hold a + * perag reference in args->pag. */ int xfs_alloc_vextent_exact_bno( @@ -3515,9 +3543,10 @@ xfs_alloc_vextent_exact_bno( args->agbno = XFS_FSB_TO_AGBNO(mp, target); args->fsbno = target; args->type = XFS_ALLOCTYPE_THIS_BNO; - error = xfs_alloc_ag_vextent(args); - if (error) - return error; + + error = xfs_alloc_vextent_prepare_ag(args); + if (!error && args->agbp) + error = xfs_alloc_ag_vextent(args); xfs_alloc_vextent_set_fsbno(args, minimum_agno); return 0; @@ -3526,6 +3555,8 @@ xfs_alloc_vextent_exact_bno( /* * Allocate an extent as close to the target as possible. If there are not * viable candidates in the AG, then fail the allocation. + * + * Caller may or may not have a per-ag reference in args->pag. */ int xfs_alloc_vextent_near_bno( @@ -3550,21 +3581,22 @@ xfs_alloc_vextent_near_bno( args->agno = XFS_FSB_TO_AGNO(mp, target); if (minimum_agno > args->agno) { trace_xfs_alloc_vextent_skip_deadlock(args); + args->fsbno = NULLFSBLOCK; return 0; } args->agbno = XFS_FSB_TO_AGBNO(mp, target); args->type = XFS_ALLOCTYPE_NEAR_BNO; - if (need_pag) - args->pag = xfs_perag_get(args->mp, args->agno); - error = xfs_alloc_ag_vextent(args); + + error = xfs_alloc_vextent_prepare_ag(args); + if (!error && args->agbp) + error = xfs_alloc_ag_vextent(args); + + xfs_alloc_vextent_set_fsbno(args, minimum_agno); if (need_pag) xfs_perag_put(args->pag); - if (error) - return error; - xfs_alloc_vextent_set_fsbno(args, minimum_agno); - return 0; + return error; } /* Ensure that the freelist is at full capacity. */ From e4d174260779ff0e2dc5de951c2e235721b02b05 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:54 +1100 Subject: [PATCH 235/765] xfs: move allocation accounting to xfs_alloc_vextent_set_fsbno() Move it from xfs_alloc_ag_vextent() so we can get rid of that layer. Rename xfs_alloc_vextent_set_fsbno() to xfs_alloc_vextent_finish() to indicate that it's function is finishing off the allocation that we've run now that it contains much more functionality. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_alloc.c | 122 ++++++++++++++++++++------------------ 1 file changed, 63 insertions(+), 59 deletions(-) diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index dc1b6c6e1de7c..afc6bd48b0a54 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -1163,36 +1163,6 @@ xfs_alloc_ag_vextent( ASSERT(0); /* NOTREACHED */ } - - if (error || args->agbno == NULLAGBLOCK) - return error; - - ASSERT(args->len >= args->minlen); - ASSERT(args->len <= args->maxlen); - ASSERT(args->agbno % args->alignment == 0); - - /* if not file data, insert new block into the reverse map btree */ - if (!xfs_rmap_should_skip_owner_update(&args->oinfo)) { - error = xfs_rmap_alloc(args->tp, args->agbp, args->pag, - args->agbno, args->len, &args->oinfo); - if (error) - return error; - } - - if (!args->wasfromfl) { - error = xfs_alloc_update_counters(args->tp, args->agbp, - -((long)(args->len))); - if (error) - return error; - - ASSERT(!xfs_extent_busy_search(args->mp, args->pag, - args->agbno, args->len)); - } - - xfs_ag_resv_alloc_extent(args->pag, args->resv, args); - - XFS_STATS_INC(args->mp, xs_allocx); - XFS_STATS_ADD(args->mp, xs_allocb, args->len); return error; } @@ -3237,18 +3207,21 @@ xfs_alloc_vextent_prepare_ag( } /* - * Post-process allocation results to set the allocated block number correctly - * for the caller. + * Post-process allocation results to account for the allocation if it succeed + * and set the allocated block number correctly for the caller. * - * XXX: xfs_alloc_vextent() should really be returning ENOSPC for ENOSPC, not + * XXX: we should really be returning ENOSPC for ENOSPC, not * hiding it behind a "successful" NULLFSBLOCK allocation. */ -static void -xfs_alloc_vextent_set_fsbno( +static int +xfs_alloc_vextent_finish( struct xfs_alloc_arg *args, - xfs_agnumber_t minimum_agno) + xfs_agnumber_t minimum_agno, + int alloc_error, + bool drop_perag) { struct xfs_mount *mp = args->mp; + int error = 0; /* * We can end up here with a locked AGF. If we failed, the caller is @@ -3271,19 +3244,54 @@ xfs_alloc_vextent_set_fsbno( args->agno > minimum_agno)) args->tp->t_highest_agno = args->agno; - /* Allocation failed with ENOSPC if NULLAGBLOCK was returned. */ - if (args->agbno == NULLAGBLOCK) { + /* + * If the allocation failed with an error or we had an ENOSPC result, + * preserve the returned error whilst also marking the allocation result + * as "no extent allocated". This ensures that callers that fail to + * capture the error will still treat it as a failed allocation. + */ + if (alloc_error || args->agbno == NULLAGBLOCK) { args->fsbno = NULLFSBLOCK; - return; + error = alloc_error; + goto out_drop_perag; } args->fsbno = XFS_AGB_TO_FSB(mp, args->agno, args->agbno); -#ifdef DEBUG + ASSERT(args->len >= args->minlen); ASSERT(args->len <= args->maxlen); ASSERT(args->agbno % args->alignment == 0); XFS_AG_CHECK_DADDR(mp, XFS_FSB_TO_DADDR(mp, args->fsbno), args->len); -#endif + + /* if not file data, insert new block into the reverse map btree */ + if (!xfs_rmap_should_skip_owner_update(&args->oinfo)) { + error = xfs_rmap_alloc(args->tp, args->agbp, args->pag, + args->agbno, args->len, &args->oinfo); + if (error) + goto out_drop_perag; + } + + if (!args->wasfromfl) { + error = xfs_alloc_update_counters(args->tp, args->agbp, + -((long)(args->len))); + if (error) + goto out_drop_perag; + + ASSERT(!xfs_extent_busy_search(mp, args->pag, args->agbno, + args->len)); + } + + xfs_ag_resv_alloc_extent(args->pag, args->resv, args); + + XFS_STATS_INC(mp, xs_allocx); + XFS_STATS_ADD(mp, xs_allocb, args->len); + +out_drop_perag: + if (drop_perag) { + xfs_perag_put(args->pag); + args->pag = NULL; + } + return error; } /* @@ -3325,8 +3333,7 @@ xfs_alloc_vextent_this_ag( if (!error && args->agbp) error = xfs_alloc_ag_vextent(args); - xfs_alloc_vextent_set_fsbno(args, minimum_agno); - return error; + return xfs_alloc_vextent_finish(args, minimum_agno, error, false); } /* @@ -3413,10 +3420,10 @@ xfs_alloc_vextent_iterate_ags( xfs_perag_put(args->pag); args->pag = NULL; } - if (args->pag) { - xfs_perag_put(args->pag); - args->pag = NULL; - } + /* + * The perag is left referenced in args for the caller to clean + * up after they've finished the allocation. + */ return error; } @@ -3473,8 +3480,7 @@ xfs_alloc_vextent_start_ag( (mp->m_sb.sb_agcount * rotorstep); } - xfs_alloc_vextent_set_fsbno(args, minimum_agno); - return error; + return xfs_alloc_vextent_finish(args, minimum_agno, error, true); } /* @@ -3507,8 +3513,7 @@ xfs_alloc_vextent_first_ag( args->fsbno = target; error = xfs_alloc_vextent_iterate_ags(args, minimum_agno, start_agno, 0); - xfs_alloc_vextent_set_fsbno(args, minimum_agno); - return error; + return xfs_alloc_vextent_finish(args, minimum_agno, error, true); } /* @@ -3537,6 +3542,7 @@ xfs_alloc_vextent_exact_bno( args->agno = XFS_FSB_TO_AGNO(mp, target); if (minimum_agno > args->agno) { trace_xfs_alloc_vextent_skip_deadlock(args); + args->fsbno = NULLFSBLOCK; return 0; } @@ -3548,8 +3554,7 @@ xfs_alloc_vextent_exact_bno( if (!error && args->agbp) error = xfs_alloc_ag_vextent(args); - xfs_alloc_vextent_set_fsbno(args, minimum_agno); - return 0; + return xfs_alloc_vextent_finish(args, minimum_agno, error, false); } /* @@ -3564,8 +3569,8 @@ xfs_alloc_vextent_near_bno( xfs_fsblock_t target) { struct xfs_mount *mp = args->mp; - bool need_pag = !args->pag; xfs_agnumber_t minimum_agno = 0; + bool needs_perag = args->pag == NULL; int error; if (args->tp->t_highest_agno != NULLAGNUMBER) @@ -3585,6 +3590,9 @@ xfs_alloc_vextent_near_bno( return 0; } + if (needs_perag) + args->pag = xfs_perag_get(mp, args->agno); + args->agbno = XFS_FSB_TO_AGBNO(mp, target); args->type = XFS_ALLOCTYPE_NEAR_BNO; @@ -3592,11 +3600,7 @@ xfs_alloc_vextent_near_bno( if (!error && args->agbp) error = xfs_alloc_ag_vextent(args); - xfs_alloc_vextent_set_fsbno(args, minimum_agno); - if (need_pag) - xfs_perag_put(args->pag); - - return error; + return xfs_alloc_vextent_finish(args, minimum_agno, error, needs_perag); } /* Ensure that the freelist is at full capacity. */ From 230e8fe8462ffda0849ea40b61dcf9f233854076 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:54 +1100 Subject: [PATCH 236/765] xfs: fold xfs_alloc_ag_vextent() into callers We don't need the multiplexing xfs_alloc_ag_vextent() provided anymore - we can just call the exact/near/size variants directly. This allows us to remove args->type completely and stop using args->fsbno as an input to the allocator algorithms. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_alloc.c | 100 ++++++++++---------------------------- fs/xfs/libxfs/xfs_alloc.h | 17 ------- fs/xfs/libxfs/xfs_bmap.c | 10 +--- fs/xfs/xfs_trace.h | 8 +-- 4 files changed, 29 insertions(+), 106 deletions(-) diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index afc6bd48b0a54..50eb44851f234 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -36,10 +36,6 @@ struct workqueue_struct *xfs_alloc_wq; #define XFSA_FIXUP_BNO_OK 1 #define XFSA_FIXUP_CNT_OK 2 -STATIC int xfs_alloc_ag_vextent_exact(xfs_alloc_arg_t *); -STATIC int xfs_alloc_ag_vextent_near(xfs_alloc_arg_t *); -STATIC int xfs_alloc_ag_vextent_size(xfs_alloc_arg_t *); - /* * Size of the AGFL. For CRC-enabled filesystes we steal a couple of slots in * the beginning of the block for a proper header with the location information @@ -772,8 +768,6 @@ xfs_alloc_cur_setup( int error; int i; - ASSERT(args->alignment == 1 || args->type != XFS_ALLOCTYPE_THIS_BNO); - acur->cur_len = args->maxlen; acur->rec_bno = 0; acur->rec_len = 0; @@ -887,7 +881,6 @@ xfs_alloc_cur_check( * We have an aligned record that satisfies minlen and beats or matches * the candidate extent size. Compare locality for near allocation mode. */ - ASSERT(args->type == XFS_ALLOCTYPE_NEAR_BNO); diff = xfs_alloc_compute_diff(args->agbno, args->len, args->alignment, args->datatype, bnoa, lena, &bnew); @@ -1132,40 +1125,6 @@ xfs_alloc_ag_vextent_small( return error; } -/* - * Allocate a variable extent in the allocation group agno. - * Type and bno are used to determine where in the allocation group the - * extent will start. - * Extent's length (returned in *len) will be between minlen and maxlen, - * and of the form k * prod + mod unless there's nothing that large. - * Return the starting a.g. block, or NULLAGBLOCK if we can't do it. - */ -static int -xfs_alloc_ag_vextent( - struct xfs_alloc_arg *args) -{ - int error = 0; - - /* - * Branch to correct routine based on the type. - */ - switch (args->type) { - case XFS_ALLOCTYPE_THIS_AG: - error = xfs_alloc_ag_vextent_size(args); - break; - case XFS_ALLOCTYPE_NEAR_BNO: - error = xfs_alloc_ag_vextent_near(args); - break; - case XFS_ALLOCTYPE_THIS_BNO: - error = xfs_alloc_ag_vextent_exact(args); - break; - default: - ASSERT(0); - /* NOTREACHED */ - } - return error; -} - /* * Allocate a variable extent at exactly agno/bno. * Extent's length (returned in *len) will be between minlen and maxlen, @@ -1351,7 +1310,6 @@ xfs_alloc_ag_vextent_locality( bool fbinc; ASSERT(acur->len == 0); - ASSERT(args->type == XFS_ALLOCTYPE_NEAR_BNO); *stat = 0; @@ -3137,6 +3095,7 @@ xfs_alloc_vextent_check_args( xfs_agblock_t agsize; args->agbno = NULLAGBLOCK; + args->fsbno = NULLFSBLOCK; /* * Just fix this up, for the case where the last a.g. is shorter @@ -3295,8 +3254,11 @@ xfs_alloc_vextent_finish( } /* - * Allocate within a single AG only. Caller is expected to hold a - * perag reference in args->pag. + * Allocate within a single AG only. This uses a best-fit length algorithm so if + * you need an exact sized allocation without locality constraints, this is the + * fastest way to do it. + * + * Caller is expected to hold a perag reference in args->pag. */ int xfs_alloc_vextent_this_ag( @@ -3305,7 +3267,6 @@ xfs_alloc_vextent_this_ag( { struct xfs_mount *mp = args->mp; xfs_agnumber_t minimum_agno = 0; - xfs_rfsblock_t target = XFS_AGB_TO_FSB(mp, agno, 0); int error; if (args->tp->t_highest_agno != NULLAGNUMBER) @@ -3317,7 +3278,7 @@ xfs_alloc_vextent_this_ag( return 0; } - error = xfs_alloc_vextent_check_args(args, target); + error = xfs_alloc_vextent_check_args(args, XFS_AGB_TO_FSB(mp, agno, 0)); if (error) { if (error == -ENOSPC) return 0; @@ -3326,12 +3287,10 @@ xfs_alloc_vextent_this_ag( args->agno = agno; args->agbno = 0; - args->fsbno = target; - args->type = XFS_ALLOCTYPE_THIS_AG; error = xfs_alloc_vextent_prepare_ag(args); if (!error && args->agbp) - error = xfs_alloc_ag_vextent(args); + error = xfs_alloc_ag_vextent_size(args); return xfs_alloc_vextent_finish(args, minimum_agno, error, false); } @@ -3355,6 +3314,7 @@ xfs_alloc_vextent_iterate_ags( struct xfs_alloc_arg *args, xfs_agnumber_t minimum_agno, xfs_agnumber_t start_agno, + xfs_agblock_t target_agbno, uint32_t flags) { struct xfs_mount *mp = args->mp; @@ -3369,7 +3329,6 @@ xfs_alloc_vextent_iterate_ags( args->agno = start_agno; for (;;) { args->pag = xfs_perag_get(mp, args->agno); - args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); error = xfs_alloc_vextent_prepare_ag(args); if (error) break; @@ -3379,16 +3338,18 @@ xfs_alloc_vextent_iterate_ags( * Allocation is supposed to succeed now, so break out * of the loop regardless of whether we succeed or not. */ - error = xfs_alloc_ag_vextent(args); + if (args->agno == start_agno && target_agbno) { + args->agbno = target_agbno; + error = xfs_alloc_ag_vextent_near(args); + } else { + args->agbno = 0; + error = xfs_alloc_ag_vextent_size(args); + } break; } trace_xfs_alloc_vextent_loopfailed(args); - if (args->agno == start_agno && - args->otype == XFS_ALLOCTYPE_NEAR_BNO) - args->type = XFS_ALLOCTYPE_THIS_AG; - /* * If we are try-locking, we can't deadlock on AGF locks so we * can wrap all the way back to the first AG. Otherwise, wrap @@ -3412,10 +3373,8 @@ xfs_alloc_vextent_iterate_ags( trace_xfs_alloc_vextent_allfailed(args); break; } - + args->agbno = target_agbno; flags = 0; - if (args->otype == XFS_ALLOCTYPE_NEAR_BNO) - args->type = XFS_ALLOCTYPE_NEAR_BNO; } xfs_perag_put(args->pag); args->pag = NULL; @@ -3464,13 +3423,11 @@ xfs_alloc_vextent_start_ag( mp->m_sb.sb_agcount), 0); bump_rotor = 1; } - start_agno = max(minimum_agno, XFS_FSB_TO_AGNO(mp, target)); - args->agbno = XFS_FSB_TO_AGBNO(mp, target); - args->type = XFS_ALLOCTYPE_NEAR_BNO; - args->fsbno = target; + start_agno = max(minimum_agno, XFS_FSB_TO_AGNO(mp, target)); error = xfs_alloc_vextent_iterate_ags(args, minimum_agno, start_agno, - XFS_ALLOC_FLAG_TRYLOCK); + XFS_FSB_TO_AGBNO(mp, target), XFS_ALLOC_FLAG_TRYLOCK); + if (bump_rotor) { if (args->agno == start_agno) mp->m_agfrotor = (mp->m_agfrotor + 1) % @@ -3484,9 +3441,9 @@ xfs_alloc_vextent_start_ag( } /* - * Iterate from the agno indicated from args->fsbno through to the end of the + * Iterate from the agno indicated via @target through to the end of the * filesystem attempting blocking allocation. This does not wrap or try a second - * pass, so will not recurse into AGs lower than indicated by fsbno. + * pass, so will not recurse into AGs lower than indicated by the target. */ int xfs_alloc_vextent_first_ag( @@ -3509,10 +3466,8 @@ xfs_alloc_vextent_first_ag( } start_agno = max(minimum_agno, XFS_FSB_TO_AGNO(mp, target)); - args->type = XFS_ALLOCTYPE_THIS_AG; - args->fsbno = target; - error = xfs_alloc_vextent_iterate_ags(args, minimum_agno, - start_agno, 0); + error = xfs_alloc_vextent_iterate_ags(args, minimum_agno, start_agno, + XFS_FSB_TO_AGBNO(mp, target), 0); return xfs_alloc_vextent_finish(args, minimum_agno, error, true); } @@ -3547,12 +3502,10 @@ xfs_alloc_vextent_exact_bno( } args->agbno = XFS_FSB_TO_AGBNO(mp, target); - args->fsbno = target; - args->type = XFS_ALLOCTYPE_THIS_BNO; error = xfs_alloc_vextent_prepare_ag(args); if (!error && args->agbp) - error = xfs_alloc_ag_vextent(args); + error = xfs_alloc_ag_vextent_exact(args); return xfs_alloc_vextent_finish(args, minimum_agno, error, false); } @@ -3594,11 +3547,10 @@ xfs_alloc_vextent_near_bno( args->pag = xfs_perag_get(mp, args->agno); args->agbno = XFS_FSB_TO_AGBNO(mp, target); - args->type = XFS_ALLOCTYPE_NEAR_BNO; error = xfs_alloc_vextent_prepare_ag(args); if (!error && args->agbp) - error = xfs_alloc_ag_vextent(args); + error = xfs_alloc_ag_vextent_near(args); return xfs_alloc_vextent_finish(args, minimum_agno, error, needs_perag); } diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h index 63d5ad4ed8a3c..2b246d74c1890 100644 --- a/fs/xfs/libxfs/xfs_alloc.h +++ b/fs/xfs/libxfs/xfs_alloc.h @@ -16,21 +16,6 @@ extern struct workqueue_struct *xfs_alloc_wq; unsigned int xfs_agfl_size(struct xfs_mount *mp); -/* - * Freespace allocation types. Argument to xfs_alloc_[v]extent. - */ -#define XFS_ALLOCTYPE_THIS_AG 0x08 /* anywhere in this a.g. */ -#define XFS_ALLOCTYPE_NEAR_BNO 0x20 /* in this a.g. and near this block */ -#define XFS_ALLOCTYPE_THIS_BNO 0x40 /* at exactly this block */ - -/* this should become an enum again when the tracing code is fixed */ -typedef unsigned int xfs_alloctype_t; - -#define XFS_ALLOC_TYPES \ - { XFS_ALLOCTYPE_THIS_AG, "THIS_AG" }, \ - { XFS_ALLOCTYPE_NEAR_BNO, "NEAR_BNO" }, \ - { XFS_ALLOCTYPE_THIS_BNO, "THIS_BNO" } - /* * Flags for xfs_alloc_fix_freelist. */ @@ -64,8 +49,6 @@ typedef struct xfs_alloc_arg { xfs_agblock_t min_agbno; /* set an agbno range for NEAR allocs */ xfs_agblock_t max_agbno; /* ... */ xfs_extlen_t len; /* output: actual size of extent */ - xfs_alloctype_t type; /* allocation type XFS_ALLOCTYPE_... */ - xfs_alloctype_t otype; /* original allocation type */ int datatype; /* mask defining data type treatment */ char wasdel; /* set if allocation was prev delayed */ char wasfromfl; /* set if allocation is from freelist */ diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index c507645f30313..ee402ac8c551d 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -3501,7 +3501,6 @@ xfs_bmap_btalloc_at_eof( bool ag_only) { struct xfs_mount *mp = args->mp; - xfs_alloctype_t atype; int error; /* @@ -3513,14 +3512,12 @@ xfs_bmap_btalloc_at_eof( if (ap->offset) { xfs_extlen_t nextminlen = 0; - atype = args->type; - args->alignment = 1; - /* * Compute the minlen+alignment for the next case. Set slop so * that the value of minlen+alignment+slop doesn't go up between * the calls. */ + args->alignment = 1; if (blen > stripe_align && blen <= args->maxlen) nextminlen = blen - stripe_align; else @@ -3544,17 +3541,15 @@ xfs_bmap_btalloc_at_eof( * according to the original allocation specification. */ args->pag = NULL; - args->type = atype; args->alignment = stripe_align; args->minlen = nextminlen; args->minalignslop = 0; } else { - args->alignment = stripe_align; - atype = args->type; /* * Adjust minlen to try and preserve alignment if we * can't guarantee an aligned maxlen extent. */ + args->alignment = stripe_align; if (blen > args->alignment && blen <= args->maxlen + args->alignment) args->minlen = blen - args->alignment; @@ -3576,7 +3571,6 @@ xfs_bmap_btalloc_at_eof( * original non-aligned state so the caller can proceed on allocation * failure as if this function was never called. */ - args->type = atype; args->fsbno = ap->blkno; args->alignment = 1; return 0; diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 1d99c30f66f00..bb7ccb5feecab 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -1799,8 +1799,6 @@ DECLARE_EVENT_CLASS(xfs_alloc_class, __field(xfs_extlen_t, alignment) __field(xfs_extlen_t, minalignslop) __field(xfs_extlen_t, len) - __field(short, type) - __field(short, otype) __field(char, wasdel) __field(char, wasfromfl) __field(int, resv) @@ -1820,8 +1818,6 @@ DECLARE_EVENT_CLASS(xfs_alloc_class, __entry->alignment = args->alignment; __entry->minalignslop = args->minalignslop; __entry->len = args->len; - __entry->type = args->type; - __entry->otype = args->otype; __entry->wasdel = args->wasdel; __entry->wasfromfl = args->wasfromfl; __entry->resv = args->resv; @@ -1830,7 +1826,7 @@ DECLARE_EVENT_CLASS(xfs_alloc_class, ), TP_printk("dev %d:%d agno 0x%x agbno 0x%x minlen %u maxlen %u mod %u " "prod %u minleft %u total %u alignment %u minalignslop %u " - "len %u type %s otype %s wasdel %d wasfromfl %d resv %d " + "len %u wasdel %d wasfromfl %d resv %d " "datatype 0x%x highest_agno 0x%x", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->agno, @@ -1844,8 +1840,6 @@ DECLARE_EVENT_CLASS(xfs_alloc_class, __entry->alignment, __entry->minalignslop, __entry->len, - __print_symbolic(__entry->type, XFS_ALLOC_TYPES), - __print_symbolic(__entry->otype, XFS_ALLOC_TYPES), __entry->wasdel, __entry->wasfromfl, __entry->resv, From 8b81356825ffb96b3b167b4dabbbbdb506bb0e0b Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:54 +1100 Subject: [PATCH 237/765] xfs: move the minimum agno checks into xfs_alloc_vextent_check_args All of the allocation functions now extract the minimum allowed AG from the transaction and then use it in some way. The allocation functions that are restricted to a single AG all check if the AG requested can be allocated from and return an error if so. These all set args->agno appropriately. All the allocation functions that iterate AGs use it to calculate the scan start AG. args->agno is not set until the iterator starts walking AGs. Hence we can easily set up a conditional check against the minimum AG allowed in xfs_alloc_vextent_check_args() based on whether args->agno contains NULLAGNUMBER or not and move all the repeated setup code to xfs_alloc_vextent_check_args(), further simplifying the allocation functions. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_alloc.c | 88 +++++++++++++++------------------------ 1 file changed, 33 insertions(+), 55 deletions(-) diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index 50eb44851f234..94cea96caf5d9 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -3089,14 +3089,18 @@ xfs_alloc_read_agf( static int xfs_alloc_vextent_check_args( struct xfs_alloc_arg *args, - xfs_fsblock_t target) + xfs_fsblock_t target, + xfs_agnumber_t *minimum_agno) { struct xfs_mount *mp = args->mp; xfs_agblock_t agsize; - args->agbno = NULLAGBLOCK; args->fsbno = NULLFSBLOCK; + *minimum_agno = 0; + if (args->tp->t_highest_agno != NULLAGNUMBER) + *minimum_agno = args->tp->t_highest_agno; + /* * Just fix this up, for the case where the last a.g. is shorter * (or there's only one a.g.) and the caller couldn't easily figure @@ -3123,11 +3127,16 @@ xfs_alloc_vextent_check_args( XFS_FSB_TO_AGBNO(mp, target) >= agsize || args->minlen > args->maxlen || args->minlen > agsize || args->mod >= args->prod) { - args->fsbno = NULLFSBLOCK; trace_xfs_alloc_vextent_badargs(args); return -ENOSPC; } + + if (args->agno != NULLAGNUMBER && *minimum_agno > args->agno) { + trace_xfs_alloc_vextent_skip_deadlock(args); + return -ENOSPC; + } return 0; + } /* @@ -3266,28 +3275,19 @@ xfs_alloc_vextent_this_ag( xfs_agnumber_t agno) { struct xfs_mount *mp = args->mp; - xfs_agnumber_t minimum_agno = 0; + xfs_agnumber_t minimum_agno; int error; - if (args->tp->t_highest_agno != NULLAGNUMBER) - minimum_agno = args->tp->t_highest_agno; - - if (minimum_agno > agno) { - trace_xfs_alloc_vextent_skip_deadlock(args); - args->fsbno = NULLFSBLOCK; - return 0; - } - - error = xfs_alloc_vextent_check_args(args, XFS_AGB_TO_FSB(mp, agno, 0)); + args->agno = agno; + args->agbno = 0; + error = xfs_alloc_vextent_check_args(args, XFS_AGB_TO_FSB(mp, agno, 0), + &minimum_agno); if (error) { if (error == -ENOSPC) return 0; return error; } - args->agno = agno; - args->agbno = 0; - error = xfs_alloc_vextent_prepare_ag(args); if (!error && args->agbp) error = xfs_alloc_ag_vextent_size(args); @@ -3400,16 +3400,15 @@ xfs_alloc_vextent_start_ag( xfs_fsblock_t target) { struct xfs_mount *mp = args->mp; - xfs_agnumber_t minimum_agno = 0; + xfs_agnumber_t minimum_agno; xfs_agnumber_t start_agno; xfs_agnumber_t rotorstep = xfs_rotorstep; bool bump_rotor = false; int error; - if (args->tp->t_highest_agno != NULLAGNUMBER) - minimum_agno = args->tp->t_highest_agno; - - error = xfs_alloc_vextent_check_args(args, target); + args->agno = NULLAGNUMBER; + args->agbno = NULLAGBLOCK; + error = xfs_alloc_vextent_check_args(args, target, &minimum_agno); if (error) { if (error == -ENOSPC) return 0; @@ -3451,14 +3450,13 @@ xfs_alloc_vextent_first_ag( xfs_fsblock_t target) { struct xfs_mount *mp = args->mp; - xfs_agnumber_t minimum_agno = 0; + xfs_agnumber_t minimum_agno; xfs_agnumber_t start_agno; int error; - if (args->tp->t_highest_agno != NULLAGNUMBER) - minimum_agno = args->tp->t_highest_agno; - - error = xfs_alloc_vextent_check_args(args, target); + args->agno = NULLAGNUMBER; + args->agbno = NULLAGBLOCK; + error = xfs_alloc_vextent_check_args(args, target, &minimum_agno); if (error) { if (error == -ENOSPC) return 0; @@ -3481,28 +3479,18 @@ xfs_alloc_vextent_exact_bno( xfs_fsblock_t target) { struct xfs_mount *mp = args->mp; - xfs_agnumber_t minimum_agno = 0; + xfs_agnumber_t minimum_agno; int error; - if (args->tp->t_highest_agno != NULLAGNUMBER) - minimum_agno = args->tp->t_highest_agno; - - error = xfs_alloc_vextent_check_args(args, target); + args->agno = XFS_FSB_TO_AGNO(mp, target); + args->agbno = XFS_FSB_TO_AGBNO(mp, target); + error = xfs_alloc_vextent_check_args(args, target, &minimum_agno); if (error) { if (error == -ENOSPC) return 0; return error; } - args->agno = XFS_FSB_TO_AGNO(mp, target); - if (minimum_agno > args->agno) { - trace_xfs_alloc_vextent_skip_deadlock(args); - args->fsbno = NULLFSBLOCK; - return 0; - } - - args->agbno = XFS_FSB_TO_AGBNO(mp, target); - error = xfs_alloc_vextent_prepare_ag(args); if (!error && args->agbp) error = xfs_alloc_ag_vextent_exact(args); @@ -3522,32 +3510,22 @@ xfs_alloc_vextent_near_bno( xfs_fsblock_t target) { struct xfs_mount *mp = args->mp; - xfs_agnumber_t minimum_agno = 0; + xfs_agnumber_t minimum_agno; bool needs_perag = args->pag == NULL; int error; - if (args->tp->t_highest_agno != NULLAGNUMBER) - minimum_agno = args->tp->t_highest_agno; - - error = xfs_alloc_vextent_check_args(args, target); + args->agno = XFS_FSB_TO_AGNO(mp, target); + args->agbno = XFS_FSB_TO_AGBNO(mp, target); + error = xfs_alloc_vextent_check_args(args, target, &minimum_agno); if (error) { if (error == -ENOSPC) return 0; return error; } - args->agno = XFS_FSB_TO_AGNO(mp, target); - if (minimum_agno > args->agno) { - trace_xfs_alloc_vextent_skip_deadlock(args); - args->fsbno = NULLFSBLOCK; - return 0; - } - if (needs_perag) args->pag = xfs_perag_get(mp, args->agno); - args->agbno = XFS_FSB_TO_AGBNO(mp, target); - error = xfs_alloc_vextent_prepare_ag(args); if (!error && args->agbp) error = xfs_alloc_ag_vextent_near(args); From 3432ef6111997f39d2f708dd508159dfaca72942 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:54 +1100 Subject: [PATCH 238/765] xfs: convert xfs_alloc_vextent_iterate_ags() to use perag walker Now that the AG iteration code in the core allocation code has been cleaned up, we can easily convert it to use a for_each_perag..() variant to use active references and skip AGs that it can't get active references on. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_ag.h | 22 ++++++--- fs/xfs/libxfs/xfs_alloc.c | 96 +++++++++++++++++---------------------- 2 files changed, 57 insertions(+), 61 deletions(-) diff --git a/fs/xfs/libxfs/xfs_ag.h b/fs/xfs/libxfs/xfs_ag.h index 8f43b91d4cf35..5e18536dfdcec 100644 --- a/fs/xfs/libxfs/xfs_ag.h +++ b/fs/xfs/libxfs/xfs_ag.h @@ -253,6 +253,7 @@ xfs_perag_next_wrap( struct xfs_perag *pag, xfs_agnumber_t *agno, xfs_agnumber_t stop_agno, + xfs_agnumber_t restart_agno, xfs_agnumber_t wrap_agno) { struct xfs_mount *mp = pag->pag_mount; @@ -260,10 +261,11 @@ xfs_perag_next_wrap( *agno = pag->pag_agno + 1; xfs_perag_rele(pag); while (*agno != stop_agno) { - if (*agno >= wrap_agno) - *agno = 0; - if (*agno == stop_agno) - break; + if (*agno >= wrap_agno) { + if (restart_agno >= stop_agno) + break; + *agno = restart_agno; + } pag = xfs_perag_grab(mp, *agno); if (pag) @@ -274,14 +276,20 @@ xfs_perag_next_wrap( } /* - * Iterate all AGs from start_agno through wrap_agno, then 0 through + * Iterate all AGs from start_agno through wrap_agno, then restart_agno through * (start_agno - 1). */ -#define for_each_perag_wrap_at(mp, start_agno, wrap_agno, agno, pag) \ +#define for_each_perag_wrap_range(mp, start_agno, restart_agno, wrap_agno, agno, pag) \ for ((agno) = (start_agno), (pag) = xfs_perag_grab((mp), (agno)); \ (pag) != NULL; \ (pag) = xfs_perag_next_wrap((pag), &(agno), (start_agno), \ - (wrap_agno))) + (restart_agno), (wrap_agno))) +/* + * Iterate all AGs from start_agno through wrap_agno, then 0 through + * (start_agno - 1). + */ +#define for_each_perag_wrap_at(mp, start_agno, wrap_agno, agno, pag) \ + for_each_perag_wrap_range((mp), (start_agno), 0, (wrap_agno), (agno), (pag)) /* * Iterate all AGs from start_agno through to the end of the filesystem, then 0 diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index 94cea96caf5d9..6a037173d20d9 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -3156,6 +3156,7 @@ xfs_alloc_vextent_prepare_ag( if (need_pag) args->pag = xfs_perag_get(args->mp, args->agno); + args->agbp = NULL; error = xfs_alloc_fix_freelist(args, 0); if (error) { trace_xfs_alloc_vextent_nofix(args); @@ -3255,8 +3256,8 @@ xfs_alloc_vextent_finish( XFS_STATS_ADD(mp, xs_allocb, args->len); out_drop_perag: - if (drop_perag) { - xfs_perag_put(args->pag); + if (drop_perag && args->pag) { + xfs_perag_rele(args->pag); args->pag = NULL; } return error; @@ -3304,6 +3305,10 @@ xfs_alloc_vextent_this_ag( * we attempt to allocation in as there is no locality optimisation possible for * those allocations. * + * On return, args->pag may be left referenced if we finish before the "all + * failed" return point. The allocation finish still needs the perag, and + * so the caller will release it once they've finished the allocation. + * * When we wrap the AG iteration at the end of the filesystem, we have to be * careful not to wrap into AGs below ones we already have locked in the * transaction if we are doing a blocking iteration. This will result in an @@ -3318,72 +3323,55 @@ xfs_alloc_vextent_iterate_ags( uint32_t flags) { struct xfs_mount *mp = args->mp; + xfs_agnumber_t agno; int error = 0; - ASSERT(start_agno >= minimum_agno); - - /* - * Loop over allocation groups twice; first time with - * trylock set, second time without. - */ - args->agno = start_agno; - for (;;) { - args->pag = xfs_perag_get(mp, args->agno); +restart: + for_each_perag_wrap_range(mp, start_agno, minimum_agno, + mp->m_sb.sb_agcount, agno, args->pag) { + args->agno = agno; error = xfs_alloc_vextent_prepare_ag(args); if (error) break; - - if (args->agbp) { - /* - * Allocation is supposed to succeed now, so break out - * of the loop regardless of whether we succeed or not. - */ - if (args->agno == start_agno && target_agbno) { - args->agbno = target_agbno; - error = xfs_alloc_ag_vextent_near(args); - } else { - args->agbno = 0; - error = xfs_alloc_ag_vextent_size(args); - } - break; - } - - trace_xfs_alloc_vextent_loopfailed(args); - - /* - * If we are try-locking, we can't deadlock on AGF locks so we - * can wrap all the way back to the first AG. Otherwise, wrap - * back to the start AG so we can't deadlock and let the end of - * scan handler decide what to do next. - */ - if (++(args->agno) == mp->m_sb.sb_agcount) { - if (flags & XFS_ALLOC_FLAG_TRYLOCK) - args->agno = 0; - else - args->agno = minimum_agno; + if (!args->agbp) { + trace_xfs_alloc_vextent_loopfailed(args); + continue; } /* - * Reached the starting a.g., must either be done - * or switch to non-trylock mode. + * Allocation is supposed to succeed now, so break out of the + * loop regardless of whether we succeed or not. */ - if (args->agno == start_agno) { - if (flags == 0) { - args->agbno = NULLAGBLOCK; - trace_xfs_alloc_vextent_allfailed(args); - break; - } + if (args->agno == start_agno && target_agbno) { args->agbno = target_agbno; - flags = 0; + error = xfs_alloc_ag_vextent_near(args); + } else { + args->agbno = 0; + error = xfs_alloc_ag_vextent_size(args); } - xfs_perag_put(args->pag); + break; + } + if (error) { + xfs_perag_rele(args->pag); args->pag = NULL; + return error; } + if (args->agbp) + return 0; + /* - * The perag is left referenced in args for the caller to clean - * up after they've finished the allocation. + * We didn't find an AG we can alloation from. If we were given + * constraining flags by the caller, drop them and retry the allocation + * without any constraints being set. */ - return error; + if (flags) { + flags = 0; + goto restart; + } + + ASSERT(args->pag == NULL); + trace_xfs_alloc_vextent_allfailed(args); + return 0; } /* @@ -3524,7 +3512,7 @@ xfs_alloc_vextent_near_bno( } if (needs_perag) - args->pag = xfs_perag_get(mp, args->agno); + args->pag = xfs_perag_grab(mp, args->agno); error = xfs_alloc_vextent_prepare_ag(args); if (!error && args->agbp) From 35bf2b1abc9a753321db3b3787f5b5de6bd2b370 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:54 +1100 Subject: [PATCH 239/765] xfs: convert trim to use for_each_perag_range To convert it to using active perag references and hence make it shrink safe. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/xfs_discard.c | 50 ++++++++++++++++++++------------------------ 1 file changed, 23 insertions(+), 27 deletions(-) diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c index bfc829c07f035..afc4c78b9eed6 100644 --- a/fs/xfs/xfs_discard.c +++ b/fs/xfs/xfs_discard.c @@ -21,23 +21,20 @@ STATIC int xfs_trim_extents( - struct xfs_mount *mp, - xfs_agnumber_t agno, + struct xfs_perag *pag, xfs_daddr_t start, xfs_daddr_t end, xfs_daddr_t minlen, uint64_t *blocks_trimmed) { + struct xfs_mount *mp = pag->pag_mount; struct block_device *bdev = mp->m_ddev_targp->bt_bdev; struct xfs_btree_cur *cur; struct xfs_buf *agbp; struct xfs_agf *agf; - struct xfs_perag *pag; int error; int i; - pag = xfs_perag_get(mp, agno); - /* * Force out the log. This means any transactions that might have freed * space before we take the AGF buffer lock are now on disk, and the @@ -47,7 +44,7 @@ xfs_trim_extents( error = xfs_alloc_read_agf(pag, NULL, 0, &agbp); if (error) - goto out_put_perag; + return error; agf = agbp->b_addr; cur = xfs_allocbt_init_cursor(mp, NULL, agbp, pag, XFS_BTNUM_CNT); @@ -71,10 +68,10 @@ xfs_trim_extents( error = xfs_alloc_get_rec(cur, &fbno, &flen, &i); if (error) - goto out_del_cursor; + break; if (XFS_IS_CORRUPT(mp, i != 1)) { error = -EFSCORRUPTED; - goto out_del_cursor; + break; } ASSERT(flen <= be32_to_cpu(agf->agf_longest)); @@ -83,15 +80,15 @@ xfs_trim_extents( * the format the range/len variables are supplied in by * userspace. */ - dbno = XFS_AGB_TO_DADDR(mp, agno, fbno); + dbno = XFS_AGB_TO_DADDR(mp, pag->pag_agno, fbno); dlen = XFS_FSB_TO_BB(mp, flen); /* * Too small? Give up. */ if (dlen < minlen) { - trace_xfs_discard_toosmall(mp, agno, fbno, flen); - goto out_del_cursor; + trace_xfs_discard_toosmall(mp, pag->pag_agno, fbno, flen); + break; } /* @@ -100,7 +97,7 @@ xfs_trim_extents( * down partially overlapping ranges for now. */ if (dbno + dlen < start || dbno > end) { - trace_xfs_discard_exclude(mp, agno, fbno, flen); + trace_xfs_discard_exclude(mp, pag->pag_agno, fbno, flen); goto next_extent; } @@ -109,32 +106,30 @@ xfs_trim_extents( * discard and try again the next time. */ if (xfs_extent_busy_search(mp, pag, fbno, flen)) { - trace_xfs_discard_busy(mp, agno, fbno, flen); + trace_xfs_discard_busy(mp, pag->pag_agno, fbno, flen); goto next_extent; } - trace_xfs_discard_extent(mp, agno, fbno, flen); + trace_xfs_discard_extent(mp, pag->pag_agno, fbno, flen); error = blkdev_issue_discard(bdev, dbno, dlen, GFP_NOFS); if (error) - goto out_del_cursor; + break; *blocks_trimmed += flen; next_extent: error = xfs_btree_decrement(cur, 0, &i); if (error) - goto out_del_cursor; + break; if (fatal_signal_pending(current)) { error = -ERESTARTSYS; - goto out_del_cursor; + break; } } out_del_cursor: xfs_btree_del_cursor(cur, error); xfs_buf_relse(agbp); -out_put_perag: - xfs_perag_put(pag); return error; } @@ -152,11 +147,12 @@ xfs_ioc_trim( struct xfs_mount *mp, struct fstrim_range __user *urange) { + struct xfs_perag *pag; unsigned int granularity = bdev_discard_granularity(mp->m_ddev_targp->bt_bdev); struct fstrim_range range; xfs_daddr_t start, end, minlen; - xfs_agnumber_t start_agno, end_agno, agno; + xfs_agnumber_t agno; uint64_t blocks_trimmed = 0; int error, last_error = 0; @@ -193,18 +189,18 @@ xfs_ioc_trim( end = start + BTOBBT(range.len) - 1; if (end > XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks) - 1) - end = XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks)- 1; - - start_agno = xfs_daddr_to_agno(mp, start); - end_agno = xfs_daddr_to_agno(mp, end); + end = XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks) - 1; - for (agno = start_agno; agno <= end_agno; agno++) { - error = xfs_trim_extents(mp, agno, start, end, minlen, + agno = xfs_daddr_to_agno(mp, start); + for_each_perag_range(mp, agno, xfs_daddr_to_agno(mp, end), pag) { + error = xfs_trim_extents(pag, start, end, minlen, &blocks_trimmed); if (error) { last_error = error; - if (error == -ERESTARTSYS) + if (error == -ERESTARTSYS) { + xfs_perag_rele(pag); break; + } } } From 89563e7dc099343bf7792515452e1a24005d98a6 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:54 +1100 Subject: [PATCH 240/765] xfs: factor out filestreams from xfs_bmap_btalloc_nullfb There's many if (filestreams) {} else {} branches in this function. Split it out into a filestreams specific function so that we can then work directly on cleaning up the filestreams code without impacting the rest of the allocation algorithms. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_bmap.c | 167 ++++++++++++++++++++++----------------- 1 file changed, 96 insertions(+), 71 deletions(-) diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index ee402ac8c551d..187200488ac0a 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -3234,8 +3234,8 @@ xfs_bmap_btalloc_select_lengths( return error; } -STATIC int -xfs_bmap_btalloc_filestreams( +static int +xfs_bmap_btalloc_filestreams_select_lengths( struct xfs_bmalloca *ap, struct xfs_alloc_arg *args, xfs_extlen_t *blen) @@ -3576,54 +3576,109 @@ xfs_bmap_btalloc_at_eof( return 0; } +/* + * We have failed multiple allocation attempts so now are in a low space + * allocation situation. Try a locality first full filesystem minimum length + * allocation whilst still maintaining necessary total block reservation + * requirements. + * + * If that fails, we are now critically low on space, so perform a last resort + * allocation attempt: no reserve, no locality, blocking, minimum length, full + * filesystem free space scan. We also indicate to future allocations in this + * transaction that we are critically low on space so they don't waste time on + * allocation modes that are unlikely to succeed. + */ static int -xfs_bmap_btalloc_best_length( +xfs_bmap_btalloc_low_space( + struct xfs_bmalloca *ap, + struct xfs_alloc_arg *args) +{ + int error; + + if (args->minlen > ap->minlen) { + args->minlen = ap->minlen; + error = xfs_alloc_vextent_start_ag(args, ap->blkno); + if (error || args->fsbno != NULLFSBLOCK) + return error; + } + + /* Last ditch attempt before failure is declared. */ + args->total = ap->minlen; + error = xfs_alloc_vextent_first_ag(args, 0); + if (error) + return error; + ap->tp->t_flags |= XFS_TRANS_LOWMODE; + return 0; +} + +static int +xfs_bmap_btalloc_filestreams( struct xfs_bmalloca *ap, struct xfs_alloc_arg *args, int stripe_align) { - struct xfs_mount *mp = args->mp; + xfs_agnumber_t agno = xfs_filestream_lookup_ag(ap->ip); xfs_extlen_t blen = 0; - bool is_filestream = false; int error; - if ((ap->datatype & XFS_ALLOC_USERDATA) && - xfs_inode_is_filestream(ap->ip)) - is_filestream = true; + /* Determine the initial block number we will target for allocation. */ + if (agno == NULLAGNUMBER) + agno = 0; + ap->blkno = XFS_AGB_TO_FSB(args->mp, agno, 0); + xfs_bmap_adjacent(ap); /* - * Determine the initial block number we will target for allocation. + * If there is very little free space before we start a + * filestreams allocation, we're almost guaranteed to fail to + * find an AG with enough contiguous free space to succeed, so + * just go straight to the low space algorithm. */ - if (is_filestream) { - xfs_agnumber_t agno = xfs_filestream_lookup_ag(ap->ip); - if (agno == NULLAGNUMBER) - agno = 0; - ap->blkno = XFS_AGB_TO_FSB(mp, agno, 0); - } else { - ap->blkno = XFS_INO_TO_FSB(mp, ap->ip->i_ino); + if (ap->tp->t_flags & XFS_TRANS_LOWMODE) { + args->minlen = ap->minlen; + return xfs_bmap_btalloc_low_space(ap, args); } - xfs_bmap_adjacent(ap); /* * Search for an allocation group with a single extent large enough for * the request. If one isn't found, then adjust the minimum allocation * size to the largest space found. */ - if (is_filestream) { - /* - * If there is very little free space before we start a - * filestreams allocation, we're almost guaranteed to fail to - * find an AG with enough contiguous free space to succeed, so - * just go straight to the low space algorithm. - */ - if (ap->tp->t_flags & XFS_TRANS_LOWMODE) { - args->minlen = ap->minlen; - goto critically_low_space; - } - error = xfs_bmap_btalloc_filestreams(ap, args, &blen); - } else { - error = xfs_bmap_btalloc_select_lengths(ap, args, &blen); + error = xfs_bmap_btalloc_filestreams_select_lengths(ap, args, &blen); + if (error) + return error; + + if (ap->aeof) { + error = xfs_bmap_btalloc_at_eof(ap, args, blen, stripe_align, + true); + if (error || args->fsbno != NULLFSBLOCK) + return error; } + + error = xfs_alloc_vextent_near_bno(args, ap->blkno); + if (error || args->fsbno != NULLFSBLOCK) + return error; + + return xfs_bmap_btalloc_low_space(ap, args); +} + +static int +xfs_bmap_btalloc_best_length( + struct xfs_bmalloca *ap, + struct xfs_alloc_arg *args, + int stripe_align) +{ + xfs_extlen_t blen = 0; + int error; + + ap->blkno = XFS_INO_TO_FSB(args->mp, ap->ip->i_ino); + xfs_bmap_adjacent(ap); + + /* + * Search for an allocation group with a single extent large enough for + * the request. If one isn't found, then adjust the minimum allocation + * size to the largest space found. + */ + error = xfs_bmap_btalloc_select_lengths(ap, args, &blen); if (error) return error; @@ -3635,50 +3690,16 @@ xfs_bmap_btalloc_best_length( */ if (ap->aeof && !(ap->tp->t_flags & XFS_TRANS_LOWMODE)) { error = xfs_bmap_btalloc_at_eof(ap, args, blen, stripe_align, - is_filestream); - if (error) + false); + if (error || args->fsbno != NULLFSBLOCK) return error; - if (args->fsbno != NULLFSBLOCK) - return 0; } - if (is_filestream) - error = xfs_alloc_vextent_near_bno(args, ap->blkno); - else - error = xfs_alloc_vextent_start_ag(args, ap->blkno); - if (error) + error = xfs_alloc_vextent_start_ag(args, ap->blkno); + if (error || args->fsbno != NULLFSBLOCK) return error; - if (args->fsbno != NULLFSBLOCK) - return 0; - - /* - * Try a locality first full filesystem minimum length allocation whilst - * still maintaining necessary total block reservation requirements. - */ - if (args->minlen > ap->minlen) { - args->minlen = ap->minlen; - error = xfs_alloc_vextent_start_ag(args, ap->blkno); - if (error) - return error; - } - if (args->fsbno != NULLFSBLOCK) - return 0; - /* - * We are now critically low on space, so this is a last resort - * allocation attempt: no reserve, no locality, blocking, minimum - * length, full filesystem free space scan. We also indicate to future - * allocations in this transaction that we are critically low on space - * so they don't waste time on allocation modes that are unlikely to - * succeed. - */ -critically_low_space: - args->total = ap->minlen; - error = xfs_alloc_vextent_first_ag(args, 0); - if (error) - return error; - ap->tp->t_flags |= XFS_TRANS_LOWMODE; - return 0; + return xfs_bmap_btalloc_low_space(ap, args); } static int @@ -3712,7 +3733,11 @@ xfs_bmap_btalloc( /* Trim the allocation back to the maximum an AG can fit. */ args.maxlen = min(ap->length, mp->m_ag_max_usable); - error = xfs_bmap_btalloc_best_length(ap, &args, stripe_align); + if ((ap->datatype & XFS_ALLOC_USERDATA) && + xfs_inode_is_filestream(ap->ip)) + error = xfs_bmap_btalloc_filestreams(ap, &args, stripe_align); + else + error = xfs_bmap_btalloc_best_length(ap, &args, stripe_align); if (error) return error; From 6b637ad0c7be85ecb795697ea51051039b753da2 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:55 +1100 Subject: [PATCH 241/765] xfs: get rid of notinit from xfs_bmap_longest_free_extent It is only set if reading the AGF gets a EAGAIN error. Just return the EAGAIN error and handle that error in the callers. This means we can remove the not_init parameter from xfs_bmap_select_minlen(), too, because the use of not_init there is pessimistic. If we can't read the agf, it won't increase blen. The only time we actually care whether we checked all the AGFs for contiguous free space is when the best length is less than the minimum allocation length. If not_init is set, then we ignore blen and set the minimum alloc length to the absolute minimum, not the best length we know already is present. However, if blen is less than the minimum we're going to ignore it anyway, regardless of whether we scanned all the AGFs or not. Hence not_init can go away, because we only use if blen is good from the scanned AGs otherwise we ignore it altogether and use minlen. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_bmap.c | 85 +++++++++++++++++----------------------- 1 file changed, 37 insertions(+), 48 deletions(-) diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 187200488ac0a..89398172d8be1 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -3139,8 +3139,7 @@ static int xfs_bmap_longest_free_extent( struct xfs_perag *pag, struct xfs_trans *tp, - xfs_extlen_t *blen, - int *notinit) + xfs_extlen_t *blen) { xfs_extlen_t longest; int error = 0; @@ -3148,14 +3147,8 @@ xfs_bmap_longest_free_extent( if (!xfs_perag_initialised_agf(pag)) { error = xfs_alloc_read_agf(pag, tp, XFS_ALLOC_FLAG_TRYLOCK, NULL); - if (error) { - /* Couldn't lock the AGF, so skip this AG. */ - if (error == -EAGAIN) { - *notinit = 1; - error = 0; - } + if (error) return error; - } } longest = xfs_alloc_longest_free_extent(pag, @@ -3167,32 +3160,28 @@ xfs_bmap_longest_free_extent( return 0; } -static void +static xfs_extlen_t xfs_bmap_select_minlen( struct xfs_bmalloca *ap, struct xfs_alloc_arg *args, - xfs_extlen_t *blen, - int notinit) + xfs_extlen_t blen) { - if (notinit || *blen < ap->minlen) { - /* - * Since we did a BUF_TRYLOCK above, it is possible that - * there is space for this request. - */ - args->minlen = ap->minlen; - } else if (*blen < args->maxlen) { - /* - * If the best seen length is less than the request length, - * use the best as the minimum. - */ - args->minlen = *blen; - } else { - /* - * Otherwise we've seen an extent as big as maxlen, use that - * as the minimum. - */ - args->minlen = args->maxlen; - } + + /* + * Since we used XFS_ALLOC_FLAG_TRYLOCK in _longest_free_extent(), it is + * possible that there is enough contiguous free space for this request. + */ + if (blen < ap->minlen) + return ap->minlen; + + /* + * If the best seen length is less than the request length, + * use the best as the minimum, otherwise we've got the maxlen we + * were asked for. + */ + if (blen < args->maxlen) + return blen; + return args->maxlen; } static int @@ -3204,7 +3193,6 @@ xfs_bmap_btalloc_select_lengths( struct xfs_mount *mp = args->mp; struct xfs_perag *pag; xfs_agnumber_t agno, startag; - int notinit = 0; int error = 0; if (ap->tp->t_flags & XFS_TRANS_LOWMODE) { @@ -3220,17 +3208,17 @@ xfs_bmap_btalloc_select_lengths( *blen = 0; for_each_perag_wrap(mp, startag, agno, pag) { - error = xfs_bmap_longest_free_extent(pag, args->tp, blen, - ¬init); - if (error) + error = xfs_bmap_longest_free_extent(pag, args->tp, blen); + if (error && error != -EAGAIN) break; + error = 0; if (*blen >= args->maxlen) break; } if (pag) xfs_perag_rele(pag); - xfs_bmap_select_minlen(ap, args, blen, notinit); + args->minlen = xfs_bmap_select_minlen(ap, args, *blen); return error; } @@ -3243,7 +3231,6 @@ xfs_bmap_btalloc_filestreams_select_lengths( struct xfs_mount *mp = ap->ip->i_mount; struct xfs_perag *pag; xfs_agnumber_t start_agno; - int notinit = 0; int error; args->total = ap->total; @@ -3254,11 +3241,13 @@ xfs_bmap_btalloc_filestreams_select_lengths( pag = xfs_perag_grab(mp, start_agno); if (pag) { - error = xfs_bmap_longest_free_extent(pag, args->tp, blen, - ¬init); + error = xfs_bmap_longest_free_extent(pag, args->tp, blen); xfs_perag_rele(pag); - if (error) - return error; + if (error) { + if (error != -EAGAIN) + return error; + *blen = 0; + } } if (*blen < args->maxlen) { @@ -3274,18 +3263,18 @@ xfs_bmap_btalloc_filestreams_select_lengths( if (!pag) goto out_select; - error = xfs_bmap_longest_free_extent(pag, args->tp, - blen, ¬init); + error = xfs_bmap_longest_free_extent(pag, args->tp, blen); xfs_perag_rele(pag); - if (error) - return error; - + if (error) { + if (error != -EAGAIN) + return error; + *blen = 0; + } start_agno = agno; - } out_select: - xfs_bmap_select_minlen(ap, args, blen, notinit); + args->minlen = xfs_bmap_select_minlen(ap, args, *blen); /* * Set the failure fallback case to look in the selected AG as stream From 05cf492a8d01f48d4b8d8f0b93f2d75de7349f12 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:55 +1100 Subject: [PATCH 242/765] xfs: use xfs_bmap_longest_free_extent() in filestreams The code in xfs_bmap_longest_free_extent() is open coded in xfs_filestream_pick_ag(). Export xfs_bmap_longest_free_extent and call it from the filestreams code instead. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_bmap.c | 2 +- fs/xfs/libxfs/xfs_bmap.h | 2 ++ fs/xfs/xfs_filestream.c | 22 ++++++++-------------- 3 files changed, 11 insertions(+), 15 deletions(-) diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 89398172d8be1..16628fdbcd551 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -3135,7 +3135,7 @@ xfs_bmap_adjacent( #undef ISVALID } -static int +int xfs_bmap_longest_free_extent( struct xfs_perag *pag, struct xfs_trans *tp, diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h index 524912f276f87..b52cfdcb93206 100644 --- a/fs/xfs/libxfs/xfs_bmap.h +++ b/fs/xfs/libxfs/xfs_bmap.h @@ -168,6 +168,8 @@ static inline bool xfs_bmap_is_written_extent(struct xfs_bmbt_irec *irec) #define xfs_valid_startblock(ip, startblock) \ ((startblock) != 0 || XFS_IS_REALTIME_INODE(ip)) +int xfs_bmap_longest_free_extent(struct xfs_perag *pag, + struct xfs_trans *tp, xfs_extlen_t *blen); void xfs_trim_extent(struct xfs_bmbt_irec *irec, xfs_fileoff_t bno, xfs_filblks_t len); unsigned int xfs_bmap_compute_attr_offset(struct xfs_mount *mp); diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 7e8b25ab6c463..2eb702034d05d 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -124,17 +124,14 @@ xfs_filestream_pick_ag( trace_xfs_filestream_scan(mp, ip->i_ino, ag); pag = xfs_perag_get(mp, ag); - - if (!xfs_perag_initialised_agf(pag)) { - err = xfs_alloc_read_agf(pag, NULL, trylock, NULL); - if (err) { - if (err != -EAGAIN) { - xfs_perag_put(pag); - return err; - } - /* Couldn't lock the AGF, skip this AG. */ - goto next_ag; - } + longest = 0; + err = xfs_bmap_longest_free_extent(pag, NULL, &longest); + if (err) { + xfs_perag_put(pag); + if (err != -EAGAIN) + return err; + /* Couldn't lock the AGF, skip this AG. */ + goto next_ag; } /* Keep track of the AG with the most free blocks. */ @@ -154,9 +151,6 @@ xfs_filestream_pick_ag( goto next_ag; } - longest = xfs_alloc_longest_free_extent(pag, - xfs_alloc_min_freelist(mp, pag), - xfs_ag_resv_needed(pag, XFS_AG_RESV_NONE)); if (((minlen && longest >= minlen) || (!minlen && pag->pagf_freeblks >= minfree)) && (!xfs_perag_prefers_metadata(pag) || From 8f7747ad8c52cde585b9456f6dbd1984af7b97bc Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:55 +1100 Subject: [PATCH 243/765] xfs: move xfs_bmap_btalloc_filestreams() to xfs_filestreams.c xfs_bmap_btalloc_filestreams() calls two filestreams functions to select the AG to allocate from. Both those functions end up in the same selection function that iterates all AGs multiple times. Worst case, xfs_bmap_btalloc_filestreams() can iterate all AGs 4 times just to select the initial AG to allocate in. Move the AG selection to fs/xfs/xfs_filestreams.c as a single interface so that the inefficient AG interation is contained entirely within the filestreams code. This will allow the implementation to be simplified and made more efficient in future patches. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_bmap.c | 94 +++++------------------------------- fs/xfs/libxfs/xfs_bmap.h | 3 ++ fs/xfs/xfs_filestream.c | 100 ++++++++++++++++++++++++++++++++++++++- fs/xfs/xfs_filestream.h | 5 +- 4 files changed, 115 insertions(+), 87 deletions(-) diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 16628fdbcd551..11facb8a6b3a7 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -3222,68 +3222,6 @@ xfs_bmap_btalloc_select_lengths( return error; } -static int -xfs_bmap_btalloc_filestreams_select_lengths( - struct xfs_bmalloca *ap, - struct xfs_alloc_arg *args, - xfs_extlen_t *blen) -{ - struct xfs_mount *mp = ap->ip->i_mount; - struct xfs_perag *pag; - xfs_agnumber_t start_agno; - int error; - - args->total = ap->total; - - start_agno = XFS_FSB_TO_AGNO(mp, ap->blkno); - if (start_agno == NULLAGNUMBER) - start_agno = 0; - - pag = xfs_perag_grab(mp, start_agno); - if (pag) { - error = xfs_bmap_longest_free_extent(pag, args->tp, blen); - xfs_perag_rele(pag); - if (error) { - if (error != -EAGAIN) - return error; - *blen = 0; - } - } - - if (*blen < args->maxlen) { - xfs_agnumber_t agno = start_agno; - - error = xfs_filestream_new_ag(ap, &agno); - if (error) - return error; - if (agno == NULLAGNUMBER) - goto out_select; - - pag = xfs_perag_grab(mp, agno); - if (!pag) - goto out_select; - - error = xfs_bmap_longest_free_extent(pag, args->tp, blen); - xfs_perag_rele(pag); - if (error) { - if (error != -EAGAIN) - return error; - *blen = 0; - } - start_agno = agno; - } - -out_select: - args->minlen = xfs_bmap_select_minlen(ap, args, *blen); - - /* - * Set the failure fallback case to look in the selected AG as stream - * may have moved. - */ - ap->blkno = args->fsbno = XFS_AGB_TO_FSB(mp, start_agno, 0); - return 0; -} - /* Update all inode and quota accounting for the allocation we just did. */ static void xfs_bmap_btalloc_accounting( @@ -3577,7 +3515,7 @@ xfs_bmap_btalloc_at_eof( * transaction that we are critically low on space so they don't waste time on * allocation modes that are unlikely to succeed. */ -static int +int xfs_bmap_btalloc_low_space( struct xfs_bmalloca *ap, struct xfs_alloc_arg *args) @@ -3606,36 +3544,25 @@ xfs_bmap_btalloc_filestreams( struct xfs_alloc_arg *args, int stripe_align) { - xfs_agnumber_t agno = xfs_filestream_lookup_ag(ap->ip); xfs_extlen_t blen = 0; int error; - /* Determine the initial block number we will target for allocation. */ - if (agno == NULLAGNUMBER) - agno = 0; - ap->blkno = XFS_AGB_TO_FSB(args->mp, agno, 0); - xfs_bmap_adjacent(ap); + + error = xfs_filestream_select_ag(ap, args, &blen); + if (error) + return error; /* - * If there is very little free space before we start a - * filestreams allocation, we're almost guaranteed to fail to - * find an AG with enough contiguous free space to succeed, so - * just go straight to the low space algorithm. + * If we are in low space mode, then optimal allocation will fail so + * prepare for minimal allocation and jump to the low space algorithm + * immediately. */ if (ap->tp->t_flags & XFS_TRANS_LOWMODE) { args->minlen = ap->minlen; - return xfs_bmap_btalloc_low_space(ap, args); + goto out_low_space; } - /* - * Search for an allocation group with a single extent large enough for - * the request. If one isn't found, then adjust the minimum allocation - * size to the largest space found. - */ - error = xfs_bmap_btalloc_filestreams_select_lengths(ap, args, &blen); - if (error) - return error; - + args->minlen = xfs_bmap_select_minlen(ap, args, blen); if (ap->aeof) { error = xfs_bmap_btalloc_at_eof(ap, args, blen, stripe_align, true); @@ -3647,6 +3574,7 @@ xfs_bmap_btalloc_filestreams( if (error || args->fsbno != NULLFSBLOCK) return error; +out_low_space: return xfs_bmap_btalloc_low_space(ap, args); } diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h index b52cfdcb93206..dd08361ca5a69 100644 --- a/fs/xfs/libxfs/xfs_bmap.h +++ b/fs/xfs/libxfs/xfs_bmap.h @@ -12,6 +12,7 @@ struct xfs_ifork; struct xfs_inode; struct xfs_mount; struct xfs_trans; +struct xfs_alloc_arg; /* * Argument structure for xfs_bmap_alloc. @@ -224,6 +225,8 @@ int xfs_bmap_add_extent_unwritten_real(struct xfs_trans *tp, struct xfs_bmbt_irec *new, int *logflagsp); xfs_extlen_t xfs_bmapi_minleft(struct xfs_trans *tp, struct xfs_inode *ip, int fork); +int xfs_bmap_btalloc_low_space(struct xfs_bmalloca *ap, + struct xfs_alloc_arg *args); enum xfs_bmap_intent_type { XFS_BMAP_MAP = 1, diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 2eb702034d05d..a641404aa9a6c 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -12,6 +12,7 @@ #include "xfs_mount.h" #include "xfs_inode.h" #include "xfs_bmap.h" +#include "xfs_bmap_util.h" #include "xfs_alloc.h" #include "xfs_mru_cache.h" #include "xfs_trace.h" @@ -263,7 +264,7 @@ xfs_filestream_get_parent( * * Returns NULLAGNUMBER in case of an error. */ -xfs_agnumber_t +static xfs_agnumber_t xfs_filestream_lookup_ag( struct xfs_inode *ip) { @@ -312,7 +313,7 @@ xfs_filestream_lookup_ag( * This is called when the allocator can't find a suitable extent in the * current AG, and we have to move the stream into a new AG with more space. */ -int +static int xfs_filestream_new_ag( struct xfs_bmalloca *ap, xfs_agnumber_t *agp) @@ -358,6 +359,101 @@ xfs_filestream_new_ag( return err; } +static int +xfs_filestreams_select_lengths( + struct xfs_bmalloca *ap, + struct xfs_alloc_arg *args, + xfs_extlen_t *blen) +{ + struct xfs_mount *mp = ap->ip->i_mount; + struct xfs_perag *pag; + xfs_agnumber_t start_agno; + int error; + + args->total = ap->total; + + start_agno = XFS_FSB_TO_AGNO(mp, ap->blkno); + if (start_agno == NULLAGNUMBER) + start_agno = 0; + + pag = xfs_perag_grab(mp, start_agno); + if (pag) { + error = xfs_bmap_longest_free_extent(pag, args->tp, blen); + xfs_perag_rele(pag); + if (error) { + if (error != -EAGAIN) + return error; + *blen = 0; + } + } + + if (*blen < args->maxlen) { + xfs_agnumber_t agno = start_agno; + + error = xfs_filestream_new_ag(ap, &agno); + if (error) + return error; + if (agno == NULLAGNUMBER) + goto out_select; + + pag = xfs_perag_grab(mp, agno); + if (!pag) + goto out_select; + + error = xfs_bmap_longest_free_extent(pag, args->tp, blen); + xfs_perag_rele(pag); + if (error) { + if (error != -EAGAIN) + return error; + *blen = 0; + } + start_agno = agno; + } + +out_select: + /* + * Set the failure fallback case to look in the selected AG as stream + * may have moved. + */ + ap->blkno = args->fsbno = XFS_AGB_TO_FSB(mp, start_agno, 0); + return 0; +} + +/* + * Search for an allocation group with a single extent large enough for + * the request. If one isn't found, then the largest available free extent is + * returned as the best length possible. + */ +int +xfs_filestream_select_ag( + struct xfs_bmalloca *ap, + struct xfs_alloc_arg *args, + xfs_extlen_t *blen) +{ + xfs_agnumber_t start_agno = xfs_filestream_lookup_ag(ap->ip); + + /* Determine the initial block number we will target for allocation. */ + if (start_agno == NULLAGNUMBER) + start_agno = 0; + ap->blkno = XFS_AGB_TO_FSB(args->mp, start_agno, 0); + xfs_bmap_adjacent(ap); + + /* + * If there is very little free space before we start a filestreams + * allocation, we're almost guaranteed to fail to find a better AG with + * larger free space available so we don't even try. + */ + if (ap->tp->t_flags & XFS_TRANS_LOWMODE) + return 0; + + /* + * Search for an allocation group with a single extent large enough for + * the request. If one isn't found, then adjust the minimum allocation + * size to the largest space found. + */ + return xfs_filestreams_select_lengths(ap, args, blen); +} + void xfs_filestream_deassociate( struct xfs_inode *ip) diff --git a/fs/xfs/xfs_filestream.h b/fs/xfs/xfs_filestream.h index 403226ebb80bb..df9f7553e106d 100644 --- a/fs/xfs/xfs_filestream.h +++ b/fs/xfs/xfs_filestream.h @@ -9,13 +9,14 @@ struct xfs_mount; struct xfs_inode; struct xfs_bmalloca; +struct xfs_alloc_arg; int xfs_filestream_mount(struct xfs_mount *mp); void xfs_filestream_unmount(struct xfs_mount *mp); void xfs_filestream_deassociate(struct xfs_inode *ip); -xfs_agnumber_t xfs_filestream_lookup_ag(struct xfs_inode *ip); -int xfs_filestream_new_ag(struct xfs_bmalloca *ap, xfs_agnumber_t *agp); int xfs_filestream_peek_ag(struct xfs_mount *mp, xfs_agnumber_t agno); +int xfs_filestream_select_ag(struct xfs_bmalloca *ap, + struct xfs_alloc_arg *args, xfs_extlen_t *blen); static inline int xfs_inode_is_filestream( From a52dc2ad363088d0e0ab05a71f0496e2377e5cc9 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:55 +1100 Subject: [PATCH 244/765] xfs: merge filestream AG lookup into xfs_filestream_select_ag() The lookup currently either returns the cached filestream AG or it calls xfs_filestreams_select_lengths() to looks up a new AG. This has verify the AG that is selected, so we end up doing "select a new AG loop in a couple of places when only one really is needed. Merge the initial lookup functionality with the length selection so that we only need to do a single pick loop on lookup or verification failure. This undoes a lot of the factoring that enabled the selection to be moved over to the filestreams code. It makes xfs_filestream_select_ag() an awful messier, but it has to be made worse before it can get better in future patches... Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/xfs_filestream.c | 184 +++++++++++++++------------------------- 1 file changed, 70 insertions(+), 114 deletions(-) diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index a641404aa9a6c..23044dab2001c 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -258,55 +258,6 @@ xfs_filestream_get_parent( return dir ? XFS_I(dir) : NULL; } -/* - * Find the right allocation group for a file, either by finding an - * existing file stream or creating a new one. - * - * Returns NULLAGNUMBER in case of an error. - */ -static xfs_agnumber_t -xfs_filestream_lookup_ag( - struct xfs_inode *ip) -{ - struct xfs_mount *mp = ip->i_mount; - struct xfs_inode *pip = NULL; - xfs_agnumber_t startag, ag = NULLAGNUMBER; - struct xfs_mru_cache_elem *mru; - - ASSERT(S_ISREG(VFS_I(ip)->i_mode)); - - pip = xfs_filestream_get_parent(ip); - if (!pip) - return NULLAGNUMBER; - - mru = xfs_mru_cache_lookup(mp->m_filestream, pip->i_ino); - if (mru) { - ag = container_of(mru, struct xfs_fstrm_item, mru)->ag; - xfs_mru_cache_done(mp->m_filestream); - - trace_xfs_filestream_lookup(mp, ip->i_ino, ag); - goto out; - } - - /* - * Set the starting AG using the rotor for inode32, otherwise - * use the directory inode's AG. - */ - if (xfs_is_inode32(mp)) { - xfs_agnumber_t rotorstep = xfs_rotorstep; - startag = (mp->m_agfrotor / rotorstep) % mp->m_sb.sb_agcount; - mp->m_agfrotor = (mp->m_agfrotor + 1) % - (mp->m_sb.sb_agcount * rotorstep); - } else - startag = XFS_INO_TO_AGNO(mp, pip->i_ino); - - if (xfs_filestream_pick_ag(pip, startag, &ag, 0, 0)) - ag = NULLAGNUMBER; -out: - xfs_irele(pip); - return ag; -} - /* * Pick a new allocation group for the current file and its file stream. * @@ -359,83 +310,70 @@ xfs_filestream_new_ag( return err; } -static int -xfs_filestreams_select_lengths( +/* + * Search for an allocation group with a single extent large enough for + * the request. If one isn't found, then the largest available free extent is + * returned as the best length possible. + */ +int +xfs_filestream_select_ag( struct xfs_bmalloca *ap, struct xfs_alloc_arg *args, xfs_extlen_t *blen) { struct xfs_mount *mp = ap->ip->i_mount; struct xfs_perag *pag; - xfs_agnumber_t start_agno; + struct xfs_inode *pip = NULL; + xfs_agnumber_t agno = NULLAGNUMBER; + struct xfs_mru_cache_elem *mru; int error; args->total = ap->total; + *blen = 0; - start_agno = XFS_FSB_TO_AGNO(mp, ap->blkno); - if (start_agno == NULLAGNUMBER) - start_agno = 0; - - pag = xfs_perag_grab(mp, start_agno); - if (pag) { - error = xfs_bmap_longest_free_extent(pag, args->tp, blen); - xfs_perag_rele(pag); - if (error) { - if (error != -EAGAIN) - return error; - *blen = 0; - } + pip = xfs_filestream_get_parent(ap->ip); + if (!pip) { + agno = 0; + goto new_ag; } - if (*blen < args->maxlen) { - xfs_agnumber_t agno = start_agno; + mru = xfs_mru_cache_lookup(mp->m_filestream, pip->i_ino); + if (mru) { + agno = container_of(mru, struct xfs_fstrm_item, mru)->ag; + xfs_mru_cache_done(mp->m_filestream); - error = xfs_filestream_new_ag(ap, &agno); - if (error) - return error; - if (agno == NULLAGNUMBER) - goto out_select; + trace_xfs_filestream_lookup(mp, ap->ip->i_ino, agno); + xfs_irele(pip); - pag = xfs_perag_grab(mp, agno); - if (!pag) - goto out_select; + ap->blkno = XFS_AGB_TO_FSB(args->mp, agno, 0); + xfs_bmap_adjacent(ap); - error = xfs_bmap_longest_free_extent(pag, args->tp, blen); - xfs_perag_rele(pag); - if (error) { - if (error != -EAGAIN) - return error; - *blen = 0; + pag = xfs_perag_grab(mp, agno); + if (pag) { + error = xfs_bmap_longest_free_extent(pag, args->tp, blen); + xfs_perag_rele(pag); + if (error) { + if (error != -EAGAIN) + return error; + *blen = 0; + } } - start_agno = agno; + if (*blen >= args->maxlen) + goto out_select; + } else if (xfs_is_inode32(mp)) { + xfs_agnumber_t rotorstep = xfs_rotorstep; + agno = (mp->m_agfrotor / rotorstep) % + mp->m_sb.sb_agcount; + mp->m_agfrotor = (mp->m_agfrotor + 1) % + (mp->m_sb.sb_agcount * rotorstep); + xfs_irele(pip); + } else { + agno = XFS_INO_TO_AGNO(mp, pip->i_ino); + xfs_irele(pip); } -out_select: - /* - * Set the failure fallback case to look in the selected AG as stream - * may have moved. - */ - ap->blkno = args->fsbno = XFS_AGB_TO_FSB(mp, start_agno, 0); - return 0; -} - -/* - * Search for an allocation group with a single extent large enough for - * the request. If one isn't found, then the largest available free extent is - * returned as the best length possible. - */ -int -xfs_filestream_select_ag( - struct xfs_bmalloca *ap, - struct xfs_alloc_arg *args, - xfs_extlen_t *blen) -{ - xfs_agnumber_t start_agno = xfs_filestream_lookup_ag(ap->ip); - - /* Determine the initial block number we will target for allocation. */ - if (start_agno == NULLAGNUMBER) - start_agno = 0; - ap->blkno = XFS_AGB_TO_FSB(args->mp, start_agno, 0); +new_ag: + ap->blkno = XFS_AGB_TO_FSB(args->mp, agno, 0); xfs_bmap_adjacent(ap); /* @@ -446,14 +384,32 @@ xfs_filestream_select_ag( if (ap->tp->t_flags & XFS_TRANS_LOWMODE) return 0; - /* - * Search for an allocation group with a single extent large enough for - * the request. If one isn't found, then adjust the minimum allocation - * size to the largest space found. - */ - return xfs_filestreams_select_lengths(ap, args, blen); + error = xfs_filestream_new_ag(ap, &agno); + if (error) + return error; + if (agno == NULLAGNUMBER) { + agno = 0; + goto out_select; + } + + pag = xfs_perag_grab(mp, agno); + if (!pag) + goto out_select; + + error = xfs_bmap_longest_free_extent(pag, args->tp, blen); + xfs_perag_rele(pag); + if (error) { + if (error != -EAGAIN) + return error; + *blen = 0; + } + +out_select: + ap->blkno = XFS_AGB_TO_FSB(mp, agno, 0); + return 0; } + void xfs_filestream_deassociate( struct xfs_inode *ip) From ba34de8defe013e4062bdc2ed57d748d6807a96a Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:55 +1100 Subject: [PATCH 245/765] xfs: merge new filestream AG selection into xfs_filestream_select_ag() This is largely a wrapper around xfs_filestream_pick_ag() that repeats a lot of the lookups that we just merged back into xfs_filestream_select_ag() from the lookup code. Merge the xfs_filestream_new_ag() code back into _select_ag() to get rid of all the unnecessary logic. Indeed, this makes it obvious that if we have no parent inode, the filestreams allocator always selects AG 0 regardless of whether it is fit for purpose or not. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/xfs_filestream.c | 112 ++++++++++++++-------------------------- 1 file changed, 40 insertions(+), 72 deletions(-) diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 23044dab2001c..713766729dcf6 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -98,16 +98,18 @@ xfs_fstrm_free_func( static int xfs_filestream_pick_ag( struct xfs_inode *ip, - xfs_agnumber_t startag, xfs_agnumber_t *agp, int flags, - xfs_extlen_t minlen) + xfs_extlen_t *longest) { struct xfs_mount *mp = ip->i_mount; struct xfs_fstrm_item *item; struct xfs_perag *pag; - xfs_extlen_t longest, free = 0, minfree, maxfree = 0; - xfs_agnumber_t ag, max_ag = NULLAGNUMBER; + xfs_extlen_t minlen = *longest; + xfs_extlen_t free = 0, minfree, maxfree = 0; + xfs_agnumber_t startag = *agp; + xfs_agnumber_t ag = startag; + xfs_agnumber_t max_ag = NULLAGNUMBER; int err, trylock, nscan; ASSERT(S_ISDIR(VFS_I(ip)->i_mode)); @@ -115,7 +117,6 @@ xfs_filestream_pick_ag( /* 2% of an AG's blocks must be free for it to be chosen. */ minfree = mp->m_sb.sb_agblocks / 50; - ag = startag; *agp = NULLAGNUMBER; /* For the first pass, don't sleep trying to init the per-AG. */ @@ -125,8 +126,8 @@ xfs_filestream_pick_ag( trace_xfs_filestream_scan(mp, ip->i_ino, ag); pag = xfs_perag_get(mp, ag); - longest = 0; - err = xfs_bmap_longest_free_extent(pag, NULL, &longest); + *longest = 0; + err = xfs_bmap_longest_free_extent(pag, NULL, longest); if (err) { xfs_perag_put(pag); if (err != -EAGAIN) @@ -152,7 +153,7 @@ xfs_filestream_pick_ag( goto next_ag; } - if (((minlen && longest >= minlen) || + if (((minlen && *longest >= minlen) || (!minlen && pag->pagf_freeblks >= minfree)) && (!xfs_perag_prefers_metadata(pag) || !(flags & XFS_PICK_USERDATA) || @@ -258,58 +259,6 @@ xfs_filestream_get_parent( return dir ? XFS_I(dir) : NULL; } -/* - * Pick a new allocation group for the current file and its file stream. - * - * This is called when the allocator can't find a suitable extent in the - * current AG, and we have to move the stream into a new AG with more space. - */ -static int -xfs_filestream_new_ag( - struct xfs_bmalloca *ap, - xfs_agnumber_t *agp) -{ - struct xfs_inode *ip = ap->ip, *pip; - struct xfs_mount *mp = ip->i_mount; - xfs_extlen_t minlen = ap->length; - xfs_agnumber_t startag = 0; - int flags = 0; - int err = 0; - struct xfs_mru_cache_elem *mru; - - *agp = NULLAGNUMBER; - - pip = xfs_filestream_get_parent(ip); - if (!pip) - goto exit; - - mru = xfs_mru_cache_remove(mp->m_filestream, pip->i_ino); - if (mru) { - struct xfs_fstrm_item *item = - container_of(mru, struct xfs_fstrm_item, mru); - startag = (item->ag + 1) % mp->m_sb.sb_agcount; - } - - if (ap->datatype & XFS_ALLOC_USERDATA) - flags |= XFS_PICK_USERDATA; - if (ap->tp->t_flags & XFS_TRANS_LOWMODE) - flags |= XFS_PICK_LOWSPACE; - - err = xfs_filestream_pick_ag(pip, startag, agp, flags, minlen); - - /* - * Only free the item here so we skip over the old AG earlier. - */ - if (mru) - xfs_fstrm_free_func(mp, mru); - - xfs_irele(pip); -exit: - if (*agp == NULLAGNUMBER) - *agp = 0; - return err; -} - /* * Search for an allocation group with a single extent large enough for * the request. If one isn't found, then the largest available free extent is @@ -326,6 +275,7 @@ xfs_filestream_select_ag( struct xfs_inode *pip = NULL; xfs_agnumber_t agno = NULLAGNUMBER; struct xfs_mru_cache_elem *mru; + int flags = 0; int error; args->total = ap->total; @@ -334,13 +284,14 @@ xfs_filestream_select_ag( pip = xfs_filestream_get_parent(ap->ip); if (!pip) { agno = 0; - goto new_ag; + goto out_select; } mru = xfs_mru_cache_lookup(mp->m_filestream, pip->i_ino); if (mru) { agno = container_of(mru, struct xfs_fstrm_item, mru)->ag; xfs_mru_cache_done(mp->m_filestream); + mru = NULL; trace_xfs_filestream_lookup(mp, ap->ip->i_ino, agno); xfs_irele(pip); @@ -354,7 +305,7 @@ xfs_filestream_select_ag( xfs_perag_rele(pag); if (error) { if (error != -EAGAIN) - return error; + goto out_error; *blen = 0; } } @@ -366,13 +317,18 @@ xfs_filestream_select_ag( mp->m_sb.sb_agcount; mp->m_agfrotor = (mp->m_agfrotor + 1) % (mp->m_sb.sb_agcount * rotorstep); - xfs_irele(pip); } else { agno = XFS_INO_TO_AGNO(mp, pip->i_ino); - xfs_irele(pip); } -new_ag: + /* Changing parent AG association now, so remove the existing one. */ + mru = xfs_mru_cache_remove(mp->m_filestream, pip->i_ino); + if (mru) { + struct xfs_fstrm_item *item = + container_of(mru, struct xfs_fstrm_item, mru); + agno = (item->ag + 1) % mp->m_sb.sb_agcount; + xfs_fstrm_free_func(mp, mru); + } ap->blkno = XFS_AGB_TO_FSB(args->mp, agno, 0); xfs_bmap_adjacent(ap); @@ -382,33 +338,45 @@ xfs_filestream_select_ag( * larger free space available so we don't even try. */ if (ap->tp->t_flags & XFS_TRANS_LOWMODE) - return 0; + goto out_select; + + if (ap->datatype & XFS_ALLOC_USERDATA) + flags |= XFS_PICK_USERDATA; + if (ap->tp->t_flags & XFS_TRANS_LOWMODE) + flags |= XFS_PICK_LOWSPACE; - error = xfs_filestream_new_ag(ap, &agno); + *blen = ap->length; + error = xfs_filestream_pick_ag(pip, &agno, flags, blen); if (error) - return error; + goto out_error; if (agno == NULLAGNUMBER) { agno = 0; - goto out_select; + goto out_irele; } pag = xfs_perag_grab(mp, agno); if (!pag) - goto out_select; + goto out_irele; error = xfs_bmap_longest_free_extent(pag, args->tp, blen); xfs_perag_rele(pag); if (error) { if (error != -EAGAIN) - return error; + goto out_error; *blen = 0; } +out_irele: + xfs_irele(pip); out_select: ap->blkno = XFS_AGB_TO_FSB(mp, agno, 0); return 0; -} +out_error: + xfs_irele(pip); + return error; + +} void xfs_filestream_deassociate( From 3e43877a9dac13771ac722462c87bea0bdc50759 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:55 +1100 Subject: [PATCH 246/765] xfs: remove xfs_filestream_select_ag() longest extent check Picking a new AG checks the longest free extent in the AG is valid, so there's no need to repeat the check in xfs_filestream_select_ag(). Remove it. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/xfs_filestream.c | 18 +----------------- 1 file changed, 1 insertion(+), 17 deletions(-) diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 713766729dcf6..95e28aae35ab6 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -276,7 +276,7 @@ xfs_filestream_select_ag( xfs_agnumber_t agno = NULLAGNUMBER; struct xfs_mru_cache_elem *mru; int flags = 0; - int error; + int error = 0; args->total = ap->total; *blen = 0; @@ -351,27 +351,11 @@ xfs_filestream_select_ag( goto out_error; if (agno == NULLAGNUMBER) { agno = 0; - goto out_irele; - } - - pag = xfs_perag_grab(mp, agno); - if (!pag) - goto out_irele; - - error = xfs_bmap_longest_free_extent(pag, args->tp, blen); - xfs_perag_rele(pag); - if (error) { - if (error != -EAGAIN) - goto out_error; *blen = 0; } -out_irele: - xfs_irele(pip); out_select: ap->blkno = XFS_AGB_TO_FSB(mp, agno, 0); - return 0; - out_error: xfs_irele(pip); return error; From f38b46bbfa76a854c4c2a27b1617d66fefbb3f80 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:55 +1100 Subject: [PATCH 247/765] xfs: factor out MRU hit case in xfs_filestream_select_ag Because it now stands out like a sore thumb. Factoring out this case starts the process of simplifying xfs_filestream_select_ag() again. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/xfs_filestream.c | 133 +++++++++++++++++++++++++--------------- 1 file changed, 83 insertions(+), 50 deletions(-) diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 95e28aae35ab6..147296a1079e3 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -259,10 +259,85 @@ xfs_filestream_get_parent( return dir ? XFS_I(dir) : NULL; } +/* + * Lookup the mru cache for an existing association. If one exists and we can + * use it, return with the agno and blen indicating that the allocation will + * proceed with that association. + * + * If we have no association, or we cannot use the current one and have to + * destroy it, return with blen = 0 and agno pointing at the next agno to try. + */ +int +xfs_filestream_select_ag_mru( + struct xfs_bmalloca *ap, + struct xfs_alloc_arg *args, + struct xfs_inode *pip, + xfs_agnumber_t *agno, + xfs_extlen_t *blen) +{ + struct xfs_mount *mp = ap->ip->i_mount; + struct xfs_perag *pag; + struct xfs_mru_cache_elem *mru; + int error; + + mru = xfs_mru_cache_lookup(mp->m_filestream, pip->i_ino); + if (!mru) + goto out_default_agno; + + *agno = container_of(mru, struct xfs_fstrm_item, mru)->ag; + xfs_mru_cache_done(mp->m_filestream); + + trace_xfs_filestream_lookup(mp, ap->ip->i_ino, *agno); + + ap->blkno = XFS_AGB_TO_FSB(args->mp, *agno, 0); + xfs_bmap_adjacent(ap); + + pag = xfs_perag_grab(mp, *agno); + if (!pag) + goto out_default_agno; + + error = xfs_bmap_longest_free_extent(pag, args->tp, blen); + xfs_perag_rele(pag); + if (error) { + if (error != -EAGAIN) + return error; + *blen = 0; + } + + /* + * We are done if there's still enough contiguous free space to succeed. + */ + if (*blen >= args->maxlen) + return 0; + + /* Changing parent AG association now, so remove the existing one. */ + mru = xfs_mru_cache_remove(mp->m_filestream, pip->i_ino); + if (mru) { + struct xfs_fstrm_item *item = + container_of(mru, struct xfs_fstrm_item, mru); + *agno = (item->ag + 1) % mp->m_sb.sb_agcount; + xfs_fstrm_free_func(mp, mru); + return 0; + } + +out_default_agno: + if (xfs_is_inode32(mp)) { + xfs_agnumber_t rotorstep = xfs_rotorstep; + *agno = (mp->m_agfrotor / rotorstep) % + mp->m_sb.sb_agcount; + mp->m_agfrotor = (mp->m_agfrotor + 1) % + (mp->m_sb.sb_agcount * rotorstep); + return 0; + } + *agno = XFS_INO_TO_AGNO(mp, pip->i_ino); + return 0; + +} + /* * Search for an allocation group with a single extent large enough for - * the request. If one isn't found, then the largest available free extent is - * returned as the best length possible. + * the request. If one isn't found, then adjust the minimum allocation + * size to the largest space found. */ int xfs_filestream_select_ag( @@ -271,12 +346,10 @@ xfs_filestream_select_ag( xfs_extlen_t *blen) { struct xfs_mount *mp = ap->ip->i_mount; - struct xfs_perag *pag; struct xfs_inode *pip = NULL; - xfs_agnumber_t agno = NULLAGNUMBER; - struct xfs_mru_cache_elem *mru; + xfs_agnumber_t agno; int flags = 0; - int error = 0; + int error; args->total = ap->total; *blen = 0; @@ -287,48 +360,10 @@ xfs_filestream_select_ag( goto out_select; } - mru = xfs_mru_cache_lookup(mp->m_filestream, pip->i_ino); - if (mru) { - agno = container_of(mru, struct xfs_fstrm_item, mru)->ag; - xfs_mru_cache_done(mp->m_filestream); - mru = NULL; - - trace_xfs_filestream_lookup(mp, ap->ip->i_ino, agno); - xfs_irele(pip); - - ap->blkno = XFS_AGB_TO_FSB(args->mp, agno, 0); - xfs_bmap_adjacent(ap); - - pag = xfs_perag_grab(mp, agno); - if (pag) { - error = xfs_bmap_longest_free_extent(pag, args->tp, blen); - xfs_perag_rele(pag); - if (error) { - if (error != -EAGAIN) - goto out_error; - *blen = 0; - } - } - if (*blen >= args->maxlen) - goto out_select; - } else if (xfs_is_inode32(mp)) { - xfs_agnumber_t rotorstep = xfs_rotorstep; - agno = (mp->m_agfrotor / rotorstep) % - mp->m_sb.sb_agcount; - mp->m_agfrotor = (mp->m_agfrotor + 1) % - (mp->m_sb.sb_agcount * rotorstep); - } else { - agno = XFS_INO_TO_AGNO(mp, pip->i_ino); - } + error = xfs_filestream_select_ag_mru(ap, args, pip, &agno, blen); + if (error || *blen >= args->maxlen) + goto out_rele; - /* Changing parent AG association now, so remove the existing one. */ - mru = xfs_mru_cache_remove(mp->m_filestream, pip->i_ino); - if (mru) { - struct xfs_fstrm_item *item = - container_of(mru, struct xfs_fstrm_item, mru); - agno = (item->ag + 1) % mp->m_sb.sb_agcount; - xfs_fstrm_free_func(mp, mru); - } ap->blkno = XFS_AGB_TO_FSB(args->mp, agno, 0); xfs_bmap_adjacent(ap); @@ -347,8 +382,6 @@ xfs_filestream_select_ag( *blen = ap->length; error = xfs_filestream_pick_ag(pip, &agno, flags, blen); - if (error) - goto out_error; if (agno == NULLAGNUMBER) { agno = 0; *blen = 0; @@ -356,7 +389,7 @@ xfs_filestream_select_ag( out_select: ap->blkno = XFS_AGB_TO_FSB(mp, agno, 0); -out_error: +out_rele: xfs_irele(pip); return error; From 3054face139f9c77566a90a0524dd85c2f38c7f2 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:55 +1100 Subject: [PATCH 248/765] xfs: track an active perag reference in filestreams Rather than just track the agno of the reference, track a referenced perag pointer instead. This will allow active filestreams to prevent AGs from going away until the filestreams have been torn down. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/xfs_filestream.c | 100 +++++++++++++++++----------------------- 1 file changed, 43 insertions(+), 57 deletions(-) diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 147296a1079e3..c92429272ff7c 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -23,7 +23,7 @@ struct xfs_fstrm_item { struct xfs_mru_cache_elem mru; - xfs_agnumber_t ag; /* AG in use for this directory */ + struct xfs_perag *pag; /* AG in use for this directory */ }; enum xfs_fstrm_alloc { @@ -50,43 +50,18 @@ xfs_filestream_peek_ag( return ret; } -static int -xfs_filestream_get_ag( - xfs_mount_t *mp, - xfs_agnumber_t agno) -{ - struct xfs_perag *pag; - int ret; - - pag = xfs_perag_get(mp, agno); - ret = atomic_inc_return(&pag->pagf_fstrms); - xfs_perag_put(pag); - return ret; -} - -static void -xfs_filestream_put_ag( - xfs_mount_t *mp, - xfs_agnumber_t agno) -{ - struct xfs_perag *pag; - - pag = xfs_perag_get(mp, agno); - atomic_dec(&pag->pagf_fstrms); - xfs_perag_put(pag); -} - static void xfs_fstrm_free_func( void *data, struct xfs_mru_cache_elem *mru) { - struct xfs_mount *mp = data; struct xfs_fstrm_item *item = container_of(mru, struct xfs_fstrm_item, mru); + struct xfs_perag *pag = item->pag; - xfs_filestream_put_ag(mp, item->ag); - trace_xfs_filestream_free(mp, mru->key, item->ag); + trace_xfs_filestream_free(pag->pag_mount, mru->key, pag->pag_agno); + atomic_dec(&pag->pagf_fstrms); + xfs_perag_rele(pag); kmem_free(item); } @@ -105,11 +80,11 @@ xfs_filestream_pick_ag( struct xfs_mount *mp = ip->i_mount; struct xfs_fstrm_item *item; struct xfs_perag *pag; + struct xfs_perag *max_pag = NULL; xfs_extlen_t minlen = *longest; xfs_extlen_t free = 0, minfree, maxfree = 0; xfs_agnumber_t startag = *agp; xfs_agnumber_t ag = startag; - xfs_agnumber_t max_ag = NULLAGNUMBER; int err, trylock, nscan; ASSERT(S_ISDIR(VFS_I(ip)->i_mode)); @@ -125,13 +100,16 @@ xfs_filestream_pick_ag( for (nscan = 0; 1; nscan++) { trace_xfs_filestream_scan(mp, ip->i_ino, ag); - pag = xfs_perag_get(mp, ag); + err = 0; + pag = xfs_perag_grab(mp, ag); + if (!pag) + goto next_ag; *longest = 0; err = xfs_bmap_longest_free_extent(pag, NULL, longest); if (err) { - xfs_perag_put(pag); + xfs_perag_rele(pag); if (err != -EAGAIN) - return err; + break; /* Couldn't lock the AGF, skip this AG. */ goto next_ag; } @@ -139,7 +117,10 @@ xfs_filestream_pick_ag( /* Keep track of the AG with the most free blocks. */ if (pag->pagf_freeblks > maxfree) { maxfree = pag->pagf_freeblks; - max_ag = ag; + if (max_pag) + xfs_perag_rele(max_pag); + atomic_inc(&pag->pag_active_ref); + max_pag = pag; } /* @@ -148,8 +129,9 @@ xfs_filestream_pick_ag( * loop, and it guards against two filestreams being established * in the same AG as each other. */ - if (xfs_filestream_get_ag(mp, ag) > 1) { - xfs_filestream_put_ag(mp, ag); + if (atomic_inc_return(&pag->pagf_fstrms) > 1) { + atomic_dec(&pag->pagf_fstrms); + xfs_perag_rele(pag); goto next_ag; } @@ -161,15 +143,12 @@ xfs_filestream_pick_ag( /* Break out, retaining the reference on the AG. */ free = pag->pagf_freeblks; - xfs_perag_put(pag); - *agp = ag; break; } /* Drop the reference on this AG, it's not usable. */ - xfs_filestream_put_ag(mp, ag); + atomic_dec(&pag->pagf_fstrms); next_ag: - xfs_perag_put(pag); /* Move to the next AG, wrapping to AG 0 if necessary. */ if (++ag >= mp->m_sb.sb_agcount) ag = 0; @@ -194,10 +173,10 @@ xfs_filestream_pick_ag( * Take the AG with the most free space, regardless of whether * it's already in use by another filestream. */ - if (max_ag != NULLAGNUMBER) { - xfs_filestream_get_ag(mp, max_ag); + if (max_pag) { + pag = max_pag; + atomic_inc(&pag->pagf_fstrms); free = maxfree; - *agp = max_ag; break; } @@ -207,17 +186,26 @@ xfs_filestream_pick_ag( return 0; } - trace_xfs_filestream_pick(ip, *agp, free, nscan); + trace_xfs_filestream_pick(ip, pag ? pag->pag_agno : NULLAGNUMBER, + free, nscan); - if (*agp == NULLAGNUMBER) + if (max_pag) + xfs_perag_rele(max_pag); + + if (err) + return err; + + if (!pag) { + *agp = NULLAGNUMBER; return 0; + } err = -ENOMEM; item = kmem_alloc(sizeof(*item), KM_MAYFAIL); if (!item) goto out_put_ag; - item->ag = *agp; + item->pag = pag; err = xfs_mru_cache_insert(mp->m_filestream, ip->i_ino, &item->mru); if (err) { @@ -226,12 +214,14 @@ xfs_filestream_pick_ag( goto out_free_item; } + *agp = pag->pag_agno; return 0; out_free_item: kmem_free(item); out_put_ag: - xfs_filestream_put_ag(mp, *agp); + atomic_dec(&pag->pagf_fstrms); + xfs_perag_rele(pag); return err; } @@ -284,20 +274,15 @@ xfs_filestream_select_ag_mru( if (!mru) goto out_default_agno; - *agno = container_of(mru, struct xfs_fstrm_item, mru)->ag; + pag = container_of(mru, struct xfs_fstrm_item, mru)->pag; xfs_mru_cache_done(mp->m_filestream); - trace_xfs_filestream_lookup(mp, ap->ip->i_ino, *agno); + trace_xfs_filestream_lookup(mp, ap->ip->i_ino, pag->pag_agno); - ap->blkno = XFS_AGB_TO_FSB(args->mp, *agno, 0); + ap->blkno = XFS_AGB_TO_FSB(args->mp, pag->pag_agno, 0); xfs_bmap_adjacent(ap); - pag = xfs_perag_grab(mp, *agno); - if (!pag) - goto out_default_agno; - error = xfs_bmap_longest_free_extent(pag, args->tp, blen); - xfs_perag_rele(pag); if (error) { if (error != -EAGAIN) return error; @@ -307,6 +292,7 @@ xfs_filestream_select_ag_mru( /* * We are done if there's still enough contiguous free space to succeed. */ + *agno = pag->pag_agno; if (*blen >= args->maxlen) return 0; @@ -315,7 +301,7 @@ xfs_filestream_select_ag_mru( if (mru) { struct xfs_fstrm_item *item = container_of(mru, struct xfs_fstrm_item, mru); - *agno = (item->ag + 1) % mp->m_sb.sb_agcount; + *agno = (item->pag->pag_agno + 1) % mp->m_sb.sb_agcount; xfs_fstrm_free_func(mp, mru); return 0; } From eb70aa2d8ed9a6fc3525f305226c550524390cd2 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:55 +1100 Subject: [PATCH 249/765] xfs: use for_each_perag_wrap in xfs_filestream_pick_ag xfs_filestream_pick_ag() is now ready to rework to use for_each_perag_wrap() for iterating the perags during the AG selection scan. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/xfs_filestream.c | 101 ++++++++++++++++------------------------ 1 file changed, 41 insertions(+), 60 deletions(-) diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index c92429272ff7c..71fa44485a2f6 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -83,9 +83,9 @@ xfs_filestream_pick_ag( struct xfs_perag *max_pag = NULL; xfs_extlen_t minlen = *longest; xfs_extlen_t free = 0, minfree, maxfree = 0; - xfs_agnumber_t startag = *agp; - xfs_agnumber_t ag = startag; - int err, trylock, nscan; + xfs_agnumber_t start_agno = *agp; + xfs_agnumber_t agno; + int err, trylock; ASSERT(S_ISDIR(VFS_I(ip)->i_mode)); @@ -97,13 +97,9 @@ xfs_filestream_pick_ag( /* For the first pass, don't sleep trying to init the per-AG. */ trylock = XFS_ALLOC_FLAG_TRYLOCK; - for (nscan = 0; 1; nscan++) { - trace_xfs_filestream_scan(mp, ip->i_ino, ag); - - err = 0; - pag = xfs_perag_grab(mp, ag); - if (!pag) - goto next_ag; +restart: + for_each_perag_wrap(mp, start_agno, agno, pag) { + trace_xfs_filestream_scan(mp, ip->i_ino, agno); *longest = 0; err = xfs_bmap_longest_free_extent(pag, NULL, longest); if (err) { @@ -111,6 +107,7 @@ xfs_filestream_pick_ag( if (err != -EAGAIN) break; /* Couldn't lock the AGF, skip this AG. */ + err = 0; goto next_ag; } @@ -129,77 +126,61 @@ xfs_filestream_pick_ag( * loop, and it guards against two filestreams being established * in the same AG as each other. */ - if (atomic_inc_return(&pag->pagf_fstrms) > 1) { - atomic_dec(&pag->pagf_fstrms); - xfs_perag_rele(pag); - goto next_ag; - } - - if (((minlen && *longest >= minlen) || - (!minlen && pag->pagf_freeblks >= minfree)) && - (!xfs_perag_prefers_metadata(pag) || - !(flags & XFS_PICK_USERDATA) || - (flags & XFS_PICK_LOWSPACE))) { - - /* Break out, retaining the reference on the AG. */ - free = pag->pagf_freeblks; - break; + if (atomic_inc_return(&pag->pagf_fstrms) <= 1) { + if (((minlen && *longest >= minlen) || + (!minlen && pag->pagf_freeblks >= minfree)) && + (!xfs_perag_prefers_metadata(pag) || + !(flags & XFS_PICK_USERDATA) || + (flags & XFS_PICK_LOWSPACE))) { + /* Break out, retaining the reference on the AG. */ + free = pag->pagf_freeblks; + break; + } } /* Drop the reference on this AG, it's not usable. */ atomic_dec(&pag->pagf_fstrms); -next_ag: - /* Move to the next AG, wrapping to AG 0 if necessary. */ - if (++ag >= mp->m_sb.sb_agcount) - ag = 0; + } - /* If a full pass of the AGs hasn't been done yet, continue. */ - if (ag != startag) - continue; + if (err) { + xfs_perag_rele(pag); + if (max_pag) + xfs_perag_rele(max_pag); + return err; + } + if (!pag) { /* Allow sleeping in xfs_alloc_read_agf() on the 2nd pass. */ - if (trylock != 0) { + if (trylock) { trylock = 0; - continue; + goto restart; } /* Finally, if lowspace wasn't set, set it for the 3rd pass. */ if (!(flags & XFS_PICK_LOWSPACE)) { flags |= XFS_PICK_LOWSPACE; - continue; + goto restart; } /* - * Take the AG with the most free space, regardless of whether - * it's already in use by another filestream. + * No unassociated AGs are available, so select the AG with the + * most free space, regardless of whether it's already in use by + * another filestream. It none suit, return NULLAGNUMBER. */ - if (max_pag) { - pag = max_pag; - atomic_inc(&pag->pagf_fstrms); - free = maxfree; - break; + if (!max_pag) { + *agp = NULLAGNUMBER; + trace_xfs_filestream_pick(ip, *agp, free, 0); + return 0; } - - /* take AG 0 if none matched */ - trace_xfs_filestream_pick(ip, *agp, free, nscan); - *agp = 0; - return 0; - } - - trace_xfs_filestream_pick(ip, pag ? pag->pag_agno : NULLAGNUMBER, - free, nscan); - - if (max_pag) + pag = max_pag; + free = maxfree; + atomic_inc(&pag->pagf_fstrms); + } else if (max_pag) { xfs_perag_rele(max_pag); - - if (err) - return err; - - if (!pag) { - *agp = NULLAGNUMBER; - return 0; } + trace_xfs_filestream_pick(ip, pag->pag_agno, free, 0); + err = -ENOMEM; item = kmem_alloc(sizeof(*item), KM_MAYFAIL); if (!item) From 571e259282a43f58b1f70dcbf2add20d8c83a72b Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:56 +1100 Subject: [PATCH 250/765] xfs: pass perag to filestreams tracing Pass perags instead of raw ag numbers, avoiding the need for the special peek function for the tracing code. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/xfs_filestream.c | 29 +++++------------------------ fs/xfs/xfs_filestream.h | 1 - fs/xfs/xfs_trace.h | 37 ++++++++++++++++++++----------------- 3 files changed, 25 insertions(+), 42 deletions(-) diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 71fa44485a2f6..81aebe3e09bac 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -31,25 +31,6 @@ enum xfs_fstrm_alloc { XFS_PICK_LOWSPACE = 2, }; -/* - * Allocation group filestream associations are tracked with per-ag atomic - * counters. These counters allow xfs_filestream_pick_ag() to tell whether a - * particular AG already has active filestreams associated with it. - */ -int -xfs_filestream_peek_ag( - xfs_mount_t *mp, - xfs_agnumber_t agno) -{ - struct xfs_perag *pag; - int ret; - - pag = xfs_perag_get(mp, agno); - ret = atomic_read(&pag->pagf_fstrms); - xfs_perag_put(pag); - return ret; -} - static void xfs_fstrm_free_func( void *data, @@ -59,7 +40,7 @@ xfs_fstrm_free_func( container_of(mru, struct xfs_fstrm_item, mru); struct xfs_perag *pag = item->pag; - trace_xfs_filestream_free(pag->pag_mount, mru->key, pag->pag_agno); + trace_xfs_filestream_free(pag, mru->key); atomic_dec(&pag->pagf_fstrms); xfs_perag_rele(pag); @@ -99,7 +80,7 @@ xfs_filestream_pick_ag( restart: for_each_perag_wrap(mp, start_agno, agno, pag) { - trace_xfs_filestream_scan(mp, ip->i_ino, agno); + trace_xfs_filestream_scan(pag, ip->i_ino); *longest = 0; err = xfs_bmap_longest_free_extent(pag, NULL, longest); if (err) { @@ -169,7 +150,7 @@ xfs_filestream_pick_ag( */ if (!max_pag) { *agp = NULLAGNUMBER; - trace_xfs_filestream_pick(ip, *agp, free, 0); + trace_xfs_filestream_pick(ip, NULL, free); return 0; } pag = max_pag; @@ -179,7 +160,7 @@ xfs_filestream_pick_ag( xfs_perag_rele(max_pag); } - trace_xfs_filestream_pick(ip, pag->pag_agno, free, 0); + trace_xfs_filestream_pick(ip, pag, free); err = -ENOMEM; item = kmem_alloc(sizeof(*item), KM_MAYFAIL); @@ -258,7 +239,7 @@ xfs_filestream_select_ag_mru( pag = container_of(mru, struct xfs_fstrm_item, mru)->pag; xfs_mru_cache_done(mp->m_filestream); - trace_xfs_filestream_lookup(mp, ap->ip->i_ino, pag->pag_agno); + trace_xfs_filestream_lookup(pag, ap->ip->i_ino); ap->blkno = XFS_AGB_TO_FSB(args->mp, pag->pag_agno, 0); xfs_bmap_adjacent(ap); diff --git a/fs/xfs/xfs_filestream.h b/fs/xfs/xfs_filestream.h index df9f7553e106d..84149ed0e3402 100644 --- a/fs/xfs/xfs_filestream.h +++ b/fs/xfs/xfs_filestream.h @@ -14,7 +14,6 @@ struct xfs_alloc_arg; int xfs_filestream_mount(struct xfs_mount *mp); void xfs_filestream_unmount(struct xfs_mount *mp); void xfs_filestream_deassociate(struct xfs_inode *ip); -int xfs_filestream_peek_ag(struct xfs_mount *mp, xfs_agnumber_t agno); int xfs_filestream_select_ag(struct xfs_bmalloca *ap, struct xfs_alloc_arg *args, xfs_extlen_t *blen); diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index bb7ccb5feecab..107bc8692f23e 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -74,6 +74,7 @@ struct xfs_inobt_rec_incore; union xfs_btree_ptr; struct xfs_dqtrx; struct xfs_icwalk; +struct xfs_perag; #define XFS_ATTR_FILTER_FLAGS \ { XFS_ATTR_ROOT, "ROOT" }, \ @@ -638,8 +639,8 @@ DEFINE_BUF_ITEM_EVENT(xfs_trans_bhold_release); DEFINE_BUF_ITEM_EVENT(xfs_trans_binval); DECLARE_EVENT_CLASS(xfs_filestream_class, - TP_PROTO(struct xfs_mount *mp, xfs_ino_t ino, xfs_agnumber_t agno), - TP_ARGS(mp, ino, agno), + TP_PROTO(struct xfs_perag *pag, xfs_ino_t ino), + TP_ARGS(pag, ino), TP_STRUCT__entry( __field(dev_t, dev) __field(xfs_ino_t, ino) @@ -647,10 +648,10 @@ DECLARE_EVENT_CLASS(xfs_filestream_class, __field(int, streams) ), TP_fast_assign( - __entry->dev = mp->m_super->s_dev; + __entry->dev = pag->pag_mount->m_super->s_dev; __entry->ino = ino; - __entry->agno = agno; - __entry->streams = xfs_filestream_peek_ag(mp, agno); + __entry->agno = pag->pag_agno; + __entry->streams = atomic_read(&pag->pagf_fstrms); ), TP_printk("dev %d:%d ino 0x%llx agno 0x%x streams %d", MAJOR(__entry->dev), MINOR(__entry->dev), @@ -660,39 +661,41 @@ DECLARE_EVENT_CLASS(xfs_filestream_class, ) #define DEFINE_FILESTREAM_EVENT(name) \ DEFINE_EVENT(xfs_filestream_class, name, \ - TP_PROTO(struct xfs_mount *mp, xfs_ino_t ino, xfs_agnumber_t agno), \ - TP_ARGS(mp, ino, agno)) + TP_PROTO(struct xfs_perag *pag, xfs_ino_t ino), \ + TP_ARGS(pag, ino)) DEFINE_FILESTREAM_EVENT(xfs_filestream_free); DEFINE_FILESTREAM_EVENT(xfs_filestream_lookup); DEFINE_FILESTREAM_EVENT(xfs_filestream_scan); TRACE_EVENT(xfs_filestream_pick, - TP_PROTO(struct xfs_inode *ip, xfs_agnumber_t agno, - xfs_extlen_t free, int nscan), - TP_ARGS(ip, agno, free, nscan), + TP_PROTO(struct xfs_inode *ip, struct xfs_perag *pag, + xfs_extlen_t free), + TP_ARGS(ip, pag, free), TP_STRUCT__entry( __field(dev_t, dev) __field(xfs_ino_t, ino) __field(xfs_agnumber_t, agno) __field(int, streams) __field(xfs_extlen_t, free) - __field(int, nscan) ), TP_fast_assign( __entry->dev = VFS_I(ip)->i_sb->s_dev; __entry->ino = ip->i_ino; - __entry->agno = agno; - __entry->streams = xfs_filestream_peek_ag(ip->i_mount, agno); + if (pag) { + __entry->agno = pag->pag_agno; + __entry->streams = atomic_read(&pag->pagf_fstrms); + } else { + __entry->agno = NULLAGNUMBER; + __entry->streams = 0; + } __entry->free = free; - __entry->nscan = nscan; ), - TP_printk("dev %d:%d ino 0x%llx agno 0x%x streams %d free %d nscan %d", + TP_printk("dev %d:%d ino 0x%llx agno 0x%x streams %d free %d", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->ino, __entry->agno, __entry->streams, - __entry->free, - __entry->nscan) + __entry->free) ); DECLARE_EVENT_CLASS(xfs_lock_class, From f8f1ed1ab3babad46b25e2dbe8de43b33fe7aaa6 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:56 +1100 Subject: [PATCH 251/765] xfs: return a referenced perag from filestreams allocator Now that the filestreams AG selection tracks active perags, we need to return an active perag to the core allocator code. This is because the file allocation the filestreams code will run are AG specific allocations and so need to pin the AG until the allocations complete. We cannot rely on the filestreams item reference to do this - the filestreams association can be torn down at any time, hence we need to have a separate reference for the allocation process to pin the AG after it has been selected. This means there is some perag juggling in allocation failure fallback paths as they will do all AG scans in the case the AG specific allocation fails. Hence we need to track the perag reference that the filestream allocator returned to make sure we don't leak it on repeated allocation failure. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_bmap.c | 39 ++++++++++++----- fs/xfs/xfs_filestream.c | 93 ++++++++++++++++++++++++---------------- 2 files changed, 85 insertions(+), 47 deletions(-) diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 11facb8a6b3a7..34de6e6898c48 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -3428,6 +3428,7 @@ xfs_bmap_btalloc_at_eof( bool ag_only) { struct xfs_mount *mp = args->mp; + struct xfs_perag *caller_pag = args->pag; int error; /* @@ -3455,9 +3456,11 @@ xfs_bmap_btalloc_at_eof( else args->minalignslop = 0; - args->pag = xfs_perag_get(mp, XFS_FSB_TO_AGNO(mp, ap->blkno)); + if (!caller_pag) + args->pag = xfs_perag_get(mp, XFS_FSB_TO_AGNO(mp, ap->blkno)); error = xfs_alloc_vextent_exact_bno(args, ap->blkno); - xfs_perag_put(args->pag); + if (!caller_pag) + xfs_perag_put(args->pag); if (error) return error; @@ -3483,10 +3486,14 @@ xfs_bmap_btalloc_at_eof( args->minalignslop = 0; } - if (ag_only) + if (ag_only) { error = xfs_alloc_vextent_near_bno(args, ap->blkno); - else + } else { + args->pag = NULL; error = xfs_alloc_vextent_start_ag(args, ap->blkno); + ASSERT(args->pag == NULL); + args->pag = caller_pag; + } if (error) return error; @@ -3545,12 +3552,13 @@ xfs_bmap_btalloc_filestreams( int stripe_align) { xfs_extlen_t blen = 0; - int error; + int error = 0; error = xfs_filestream_select_ag(ap, args, &blen); if (error) return error; + ASSERT(args->pag); /* * If we are in low space mode, then optimal allocation will fail so @@ -3559,22 +3567,31 @@ xfs_bmap_btalloc_filestreams( */ if (ap->tp->t_flags & XFS_TRANS_LOWMODE) { args->minlen = ap->minlen; + ASSERT(args->fsbno == NULLFSBLOCK); goto out_low_space; } args->minlen = xfs_bmap_select_minlen(ap, args, blen); - if (ap->aeof) { + if (ap->aeof) error = xfs_bmap_btalloc_at_eof(ap, args, blen, stripe_align, true); - if (error || args->fsbno != NULLFSBLOCK) - return error; - } - error = xfs_alloc_vextent_near_bno(args, ap->blkno); + if (!error && args->fsbno == NULLFSBLOCK) + error = xfs_alloc_vextent_near_bno(args, ap->blkno); + +out_low_space: + /* + * We are now done with the perag reference for the filestreams + * association provided by xfs_filestream_select_ag(). Release it now as + * we've either succeeded, had a fatal error or we are out of space and + * need to do a full filesystem scan for free space which will take it's + * own references. + */ + xfs_perag_rele(args->pag); + args->pag = NULL; if (error || args->fsbno != NULLFSBLOCK) return error; -out_low_space: return xfs_bmap_btalloc_low_space(ap, args); } diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 81aebe3e09bac..523a3b8b5754b 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -53,8 +53,9 @@ xfs_fstrm_free_func( */ static int xfs_filestream_pick_ag( + struct xfs_alloc_arg *args, struct xfs_inode *ip, - xfs_agnumber_t *agp, + xfs_agnumber_t start_agno, int flags, xfs_extlen_t *longest) { @@ -64,7 +65,6 @@ xfs_filestream_pick_ag( struct xfs_perag *max_pag = NULL; xfs_extlen_t minlen = *longest; xfs_extlen_t free = 0, minfree, maxfree = 0; - xfs_agnumber_t start_agno = *agp; xfs_agnumber_t agno; int err, trylock; @@ -73,8 +73,6 @@ xfs_filestream_pick_ag( /* 2% of an AG's blocks must be free for it to be chosen. */ minfree = mp->m_sb.sb_agblocks / 50; - *agp = NULLAGNUMBER; - /* For the first pass, don't sleep trying to init the per-AG. */ trylock = XFS_ALLOC_FLAG_TRYLOCK; @@ -89,7 +87,7 @@ xfs_filestream_pick_ag( break; /* Couldn't lock the AGF, skip this AG. */ err = 0; - goto next_ag; + continue; } /* Keep track of the AG with the most free blocks. */ @@ -146,16 +144,19 @@ xfs_filestream_pick_ag( /* * No unassociated AGs are available, so select the AG with the * most free space, regardless of whether it's already in use by - * another filestream. It none suit, return NULLAGNUMBER. + * another filestream. It none suit, just use whatever AG we can + * grab. */ if (!max_pag) { - *agp = NULLAGNUMBER; - trace_xfs_filestream_pick(ip, NULL, free); - return 0; + for_each_perag_wrap(mp, start_agno, agno, pag) + break; + atomic_inc(&pag->pagf_fstrms); + *longest = 0; + } else { + pag = max_pag; + free = maxfree; + atomic_inc(&pag->pagf_fstrms); } - pag = max_pag; - free = maxfree; - atomic_inc(&pag->pagf_fstrms); } else if (max_pag) { xfs_perag_rele(max_pag); } @@ -167,16 +168,29 @@ xfs_filestream_pick_ag( if (!item) goto out_put_ag; + + /* + * We are going to use this perag now, so take another ref to it for the + * allocation context returned to the caller. If we raced to create and + * insert the filestreams item into the MRU (-EEXIST), then we still + * keep this reference but free the item reference we gained above. On + * any other failure, we have to drop both. + */ + atomic_inc(&pag->pag_active_ref); item->pag = pag; + args->pag = pag; err = xfs_mru_cache_insert(mp->m_filestream, ip->i_ino, &item->mru); if (err) { - if (err == -EEXIST) + if (err == -EEXIST) { err = 0; + } else { + xfs_perag_rele(args->pag); + args->pag = NULL; + } goto out_free_item; } - *agp = pag->pag_agno; return 0; out_free_item: @@ -236,7 +250,14 @@ xfs_filestream_select_ag_mru( if (!mru) goto out_default_agno; + /* + * Grab the pag and take an extra active reference for the caller whilst + * the mru item cannot go away. This means we'll pin the perag with + * the reference we get here even if the filestreams association is torn + * down immediately after we mark the lookup as done. + */ pag = container_of(mru, struct xfs_fstrm_item, mru)->pag; + atomic_inc(&pag->pag_active_ref); xfs_mru_cache_done(mp->m_filestream); trace_xfs_filestream_lookup(pag, ap->ip->i_ino); @@ -246,6 +267,8 @@ xfs_filestream_select_ag_mru( error = xfs_bmap_longest_free_extent(pag, args->tp, blen); if (error) { + /* We aren't going to use this perag */ + xfs_perag_rele(pag); if (error != -EAGAIN) return error; *blen = 0; @@ -253,12 +276,18 @@ xfs_filestream_select_ag_mru( /* * We are done if there's still enough contiguous free space to succeed. + * If there is very little free space before we start a filestreams + * allocation, we're almost guaranteed to fail to find a better AG with + * larger free space available so we don't even try. */ *agno = pag->pag_agno; - if (*blen >= args->maxlen) + if (*blen >= args->maxlen || (ap->tp->t_flags & XFS_TRANS_LOWMODE)) { + args->pag = pag; return 0; + } /* Changing parent AG association now, so remove the existing one. */ + xfs_perag_rele(pag); mru = xfs_mru_cache_remove(mp->m_filestream, pip->i_ino); if (mru) { struct xfs_fstrm_item *item = @@ -297,46 +326,38 @@ xfs_filestream_select_ag( struct xfs_inode *pip = NULL; xfs_agnumber_t agno; int flags = 0; - int error; + int error = 0; args->total = ap->total; *blen = 0; pip = xfs_filestream_get_parent(ap->ip); if (!pip) { - agno = 0; - goto out_select; + ap->blkno = XFS_AGB_TO_FSB(mp, 0, 0); + return 0; } error = xfs_filestream_select_ag_mru(ap, args, pip, &agno, blen); - if (error || *blen >= args->maxlen) + if (error) goto out_rele; - - ap->blkno = XFS_AGB_TO_FSB(args->mp, agno, 0); - xfs_bmap_adjacent(ap); - - /* - * If there is very little free space before we start a filestreams - * allocation, we're almost guaranteed to fail to find a better AG with - * larger free space available so we don't even try. - */ + if (*blen >= args->maxlen) + goto out_select; if (ap->tp->t_flags & XFS_TRANS_LOWMODE) goto out_select; + ap->blkno = XFS_AGB_TO_FSB(args->mp, agno, 0); + xfs_bmap_adjacent(ap); + *blen = ap->length; if (ap->datatype & XFS_ALLOC_USERDATA) flags |= XFS_PICK_USERDATA; if (ap->tp->t_flags & XFS_TRANS_LOWMODE) flags |= XFS_PICK_LOWSPACE; - *blen = ap->length; - error = xfs_filestream_pick_ag(pip, &agno, flags, blen); - if (agno == NULLAGNUMBER) { - agno = 0; - *blen = 0; - } - + error = xfs_filestream_pick_ag(args, pip, agno, flags, blen); + if (error) + goto out_rele; out_select: - ap->blkno = XFS_AGB_TO_FSB(mp, agno, 0); + ap->blkno = XFS_AGB_TO_FSB(mp, args->pag->pag_agno, 0); out_rele: xfs_irele(pip); return error; From bd4f5d09cc93c8ca51e4efea86ac90a4bb553d6e Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 13 Feb 2023 09:14:56 +1100 Subject: [PATCH 252/765] xfs: refactor the filestreams allocator pick functions Now that the filestreams allocator is largely rewritten, restructure the main entry point and pick function to seperate out the different operations cleanly. The MRU lookup function should not handle the start AG selection on MRU lookup failure, and nor should the pick function handle building the association that is inserted into the MRU. This leaves the filestreams allocator fairly clean and easy to understand, returning to the caller with an active perag reference and a target block to allocate at. Signed-off-by: Dave Chinner Reviewed-by: Darrick J. Wong --- fs/xfs/xfs_filestream.c | 268 +++++++++++++++++++++------------------- fs/xfs/xfs_trace.h | 9 +- 2 files changed, 145 insertions(+), 132 deletions(-) diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 523a3b8b5754b..22c13933c8f80 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -48,37 +48,33 @@ xfs_fstrm_free_func( } /* - * Scan the AGs starting at startag looking for an AG that isn't in use and has - * at least minlen blocks free. + * Scan the AGs starting at start_agno looking for an AG that isn't in use and + * has at least minlen blocks free. If no AG is found to match the allocation + * requirements, pick the AG with the most free space in it. */ static int xfs_filestream_pick_ag( struct xfs_alloc_arg *args, - struct xfs_inode *ip, + xfs_ino_t pino, xfs_agnumber_t start_agno, int flags, xfs_extlen_t *longest) { - struct xfs_mount *mp = ip->i_mount; - struct xfs_fstrm_item *item; + struct xfs_mount *mp = args->mp; struct xfs_perag *pag; struct xfs_perag *max_pag = NULL; xfs_extlen_t minlen = *longest; xfs_extlen_t free = 0, minfree, maxfree = 0; xfs_agnumber_t agno; - int err, trylock; - - ASSERT(S_ISDIR(VFS_I(ip)->i_mode)); + bool first_pass = true; + int err; /* 2% of an AG's blocks must be free for it to be chosen. */ minfree = mp->m_sb.sb_agblocks / 50; - /* For the first pass, don't sleep trying to init the per-AG. */ - trylock = XFS_ALLOC_FLAG_TRYLOCK; - restart: for_each_perag_wrap(mp, start_agno, agno, pag) { - trace_xfs_filestream_scan(pag, ip->i_ino); + trace_xfs_filestream_scan(pag, pino); *longest = 0; err = xfs_bmap_longest_free_extent(pag, NULL, longest); if (err) { @@ -129,13 +125,20 @@ xfs_filestream_pick_ag( } if (!pag) { - /* Allow sleeping in xfs_alloc_read_agf() on the 2nd pass. */ - if (trylock) { - trylock = 0; + /* + * Allow a second pass to give xfs_bmap_longest_free_extent() + * another attempt at locking AGFs that it might have skipped + * over before we fail. + */ + if (first_pass) { + first_pass = false; goto restart; } - /* Finally, if lowspace wasn't set, set it for the 3rd pass. */ + /* + * We must be low on data space, so run a final lowspace + * optimised selection pass if we haven't already. + */ if (!(flags & XFS_PICK_LOWSPACE)) { flags |= XFS_PICK_LOWSPACE; goto restart; @@ -148,9 +151,9 @@ xfs_filestream_pick_ag( * grab. */ if (!max_pag) { - for_each_perag_wrap(mp, start_agno, agno, pag) + for_each_perag_wrap(args->mp, 0, start_agno, args->pag) break; - atomic_inc(&pag->pagf_fstrms); + atomic_inc(&args->pag->pagf_fstrms); *longest = 0; } else { pag = max_pag; @@ -161,44 +164,10 @@ xfs_filestream_pick_ag( xfs_perag_rele(max_pag); } - trace_xfs_filestream_pick(ip, pag, free); - - err = -ENOMEM; - item = kmem_alloc(sizeof(*item), KM_MAYFAIL); - if (!item) - goto out_put_ag; - - - /* - * We are going to use this perag now, so take another ref to it for the - * allocation context returned to the caller. If we raced to create and - * insert the filestreams item into the MRU (-EEXIST), then we still - * keep this reference but free the item reference we gained above. On - * any other failure, we have to drop both. - */ - atomic_inc(&pag->pag_active_ref); - item->pag = pag; + trace_xfs_filestream_pick(pag, pino, free); args->pag = pag; - - err = xfs_mru_cache_insert(mp->m_filestream, ip->i_ino, &item->mru); - if (err) { - if (err == -EEXIST) { - err = 0; - } else { - xfs_perag_rele(args->pag); - args->pag = NULL; - } - goto out_free_item; - } - return 0; -out_free_item: - kmem_free(item); -out_put_ag: - atomic_dec(&pag->pagf_fstrms); - xfs_perag_rele(pag); - return err; } static struct xfs_inode * @@ -227,29 +196,29 @@ xfs_filestream_get_parent( /* * Lookup the mru cache for an existing association. If one exists and we can - * use it, return with the agno and blen indicating that the allocation will - * proceed with that association. + * use it, return with an active perag reference indicating that the allocation + * will proceed with that association. * * If we have no association, or we cannot use the current one and have to - * destroy it, return with blen = 0 and agno pointing at the next agno to try. + * destroy it, return with longest = 0 to tell the caller to create a new + * association. */ -int -xfs_filestream_select_ag_mru( +static int +xfs_filestream_lookup_association( struct xfs_bmalloca *ap, struct xfs_alloc_arg *args, - struct xfs_inode *pip, - xfs_agnumber_t *agno, - xfs_extlen_t *blen) + xfs_ino_t pino, + xfs_extlen_t *longest) { - struct xfs_mount *mp = ap->ip->i_mount; + struct xfs_mount *mp = args->mp; struct xfs_perag *pag; struct xfs_mru_cache_elem *mru; - int error; + int error = 0; - mru = xfs_mru_cache_lookup(mp->m_filestream, pip->i_ino); + *longest = 0; + mru = xfs_mru_cache_lookup(mp->m_filestream, pino); if (!mru) - goto out_default_agno; - + return 0; /* * Grab the pag and take an extra active reference for the caller whilst * the mru item cannot go away. This means we'll pin the perag with @@ -265,103 +234,148 @@ xfs_filestream_select_ag_mru( ap->blkno = XFS_AGB_TO_FSB(args->mp, pag->pag_agno, 0); xfs_bmap_adjacent(ap); - error = xfs_bmap_longest_free_extent(pag, args->tp, blen); - if (error) { - /* We aren't going to use this perag */ - xfs_perag_rele(pag); - if (error != -EAGAIN) - return error; - *blen = 0; - } - /* - * We are done if there's still enough contiguous free space to succeed. * If there is very little free space before we start a filestreams - * allocation, we're almost guaranteed to fail to find a better AG with - * larger free space available so we don't even try. + * allocation, we're almost guaranteed to fail to find a large enough + * free space available so just use the cached AG. */ - *agno = pag->pag_agno; - if (*blen >= args->maxlen || (ap->tp->t_flags & XFS_TRANS_LOWMODE)) { - args->pag = pag; - return 0; + if (ap->tp->t_flags & XFS_TRANS_LOWMODE) { + *longest = 1; + goto out_done; } + error = xfs_bmap_longest_free_extent(pag, args->tp, longest); + if (error == -EAGAIN) + error = 0; + if (error || *longest < args->maxlen) { + /* We aren't going to use this perag */ + *longest = 0; + xfs_perag_rele(pag); + return error; + } + +out_done: + args->pag = pag; + return 0; +} + +static int +xfs_filestream_create_association( + struct xfs_bmalloca *ap, + struct xfs_alloc_arg *args, + xfs_ino_t pino, + xfs_extlen_t *longest) +{ + struct xfs_mount *mp = args->mp; + struct xfs_mru_cache_elem *mru; + struct xfs_fstrm_item *item; + xfs_agnumber_t agno = XFS_INO_TO_AGNO(mp, pino); + int flags = 0; + int error; + /* Changing parent AG association now, so remove the existing one. */ - xfs_perag_rele(pag); - mru = xfs_mru_cache_remove(mp->m_filestream, pip->i_ino); + mru = xfs_mru_cache_remove(mp->m_filestream, pino); if (mru) { struct xfs_fstrm_item *item = container_of(mru, struct xfs_fstrm_item, mru); - *agno = (item->pag->pag_agno + 1) % mp->m_sb.sb_agcount; - xfs_fstrm_free_func(mp, mru); - return 0; - } -out_default_agno: - if (xfs_is_inode32(mp)) { + agno = (item->pag->pag_agno + 1) % mp->m_sb.sb_agcount; + xfs_fstrm_free_func(mp, mru); + } else if (xfs_is_inode32(mp)) { xfs_agnumber_t rotorstep = xfs_rotorstep; - *agno = (mp->m_agfrotor / rotorstep) % - mp->m_sb.sb_agcount; + + agno = (mp->m_agfrotor / rotorstep) % mp->m_sb.sb_agcount; mp->m_agfrotor = (mp->m_agfrotor + 1) % (mp->m_sb.sb_agcount * rotorstep); - return 0; } - *agno = XFS_INO_TO_AGNO(mp, pip->i_ino); + + ap->blkno = XFS_AGB_TO_FSB(args->mp, agno, 0); + xfs_bmap_adjacent(ap); + + if (ap->datatype & XFS_ALLOC_USERDATA) + flags |= XFS_PICK_USERDATA; + if (ap->tp->t_flags & XFS_TRANS_LOWMODE) + flags |= XFS_PICK_LOWSPACE; + + *longest = ap->length; + error = xfs_filestream_pick_ag(args, pino, agno, flags, longest); + if (error) + return error; + + /* + * We are going to use this perag now, so create an assoication for it. + * xfs_filestream_pick_ag() has already bumped the perag fstrms counter + * for us, so all we need to do here is take another active reference to + * the perag for the cached association. + * + * If we fail to store the association, we need to drop the fstrms + * counter as well as drop the perag reference we take here for the + * item. We do not need to return an error for this failure - as long as + * we return a referenced AG, the allocation can still go ahead just + * fine. + */ + item = kmem_alloc(sizeof(*item), KM_MAYFAIL); + if (!item) + goto out_put_fstrms; + + atomic_inc(&args->pag->pag_active_ref); + item->pag = args->pag; + error = xfs_mru_cache_insert(mp->m_filestream, pino, &item->mru); + if (error) + goto out_free_item; return 0; +out_free_item: + xfs_perag_rele(item->pag); + kmem_free(item); +out_put_fstrms: + atomic_dec(&args->pag->pagf_fstrms); + return 0; } /* * Search for an allocation group with a single extent large enough for - * the request. If one isn't found, then adjust the minimum allocation - * size to the largest space found. + * the request. First we look for an existing association and use that if it + * is found. Otherwise, we create a new association by selecting an AG that fits + * the allocation criteria. + * + * We return with a referenced perag in args->pag to indicate which AG we are + * allocating into or an error with no references held. */ int xfs_filestream_select_ag( struct xfs_bmalloca *ap, struct xfs_alloc_arg *args, - xfs_extlen_t *blen) + xfs_extlen_t *longest) { - struct xfs_mount *mp = ap->ip->i_mount; - struct xfs_inode *pip = NULL; - xfs_agnumber_t agno; - int flags = 0; + struct xfs_mount *mp = args->mp; + struct xfs_inode *pip; + xfs_ino_t ino = 0; int error = 0; + *longest = 0; args->total = ap->total; - *blen = 0; - pip = xfs_filestream_get_parent(ap->ip); - if (!pip) { - ap->blkno = XFS_AGB_TO_FSB(mp, 0, 0); - return 0; + if (pip) { + ino = pip->i_ino; + error = xfs_filestream_lookup_association(ap, args, ino, + longest); + xfs_irele(pip); + if (error) + return error; + if (*longest >= args->maxlen) + goto out_select; + if (ap->tp->t_flags & XFS_TRANS_LOWMODE) + goto out_select; } - error = xfs_filestream_select_ag_mru(ap, args, pip, &agno, blen); + error = xfs_filestream_create_association(ap, args, ino, longest); if (error) - goto out_rele; - if (*blen >= args->maxlen) - goto out_select; - if (ap->tp->t_flags & XFS_TRANS_LOWMODE) - goto out_select; + return error; - ap->blkno = XFS_AGB_TO_FSB(args->mp, agno, 0); - xfs_bmap_adjacent(ap); - *blen = ap->length; - if (ap->datatype & XFS_ALLOC_USERDATA) - flags |= XFS_PICK_USERDATA; - if (ap->tp->t_flags & XFS_TRANS_LOWMODE) - flags |= XFS_PICK_LOWSPACE; - - error = xfs_filestream_pick_ag(args, pip, agno, flags, blen); - if (error) - goto out_rele; out_select: ap->blkno = XFS_AGB_TO_FSB(mp, args->pag->pag_agno, 0); -out_rele: - xfs_irele(pip); - return error; - + return 0; } void diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 107bc8692f23e..7dc0fd6a65047 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -668,9 +668,8 @@ DEFINE_FILESTREAM_EVENT(xfs_filestream_lookup); DEFINE_FILESTREAM_EVENT(xfs_filestream_scan); TRACE_EVENT(xfs_filestream_pick, - TP_PROTO(struct xfs_inode *ip, struct xfs_perag *pag, - xfs_extlen_t free), - TP_ARGS(ip, pag, free), + TP_PROTO(struct xfs_perag *pag, xfs_ino_t ino, xfs_extlen_t free), + TP_ARGS(pag, ino, free), TP_STRUCT__entry( __field(dev_t, dev) __field(xfs_ino_t, ino) @@ -679,8 +678,8 @@ TRACE_EVENT(xfs_filestream_pick, __field(xfs_extlen_t, free) ), TP_fast_assign( - __entry->dev = VFS_I(ip)->i_sb->s_dev; - __entry->ino = ip->i_ino; + __entry->dev = pag->pag_mount->m_super->s_dev; + __entry->ino = ino; if (pag) { __entry->agno = pag->pag_agno; __entry->streams = atomic_read(&pag->pagf_fstrms); From 2c4d3841a82b88ae8a7b518dc6206f84f68e705a Mon Sep 17 00:00:00 2001 From: Peter Foley Date: Thu, 12 Jan 2023 23:49:08 -0500 Subject: [PATCH 253/765] um: Avoid pcap multiple definition errors Change the function name in pcap_kern to avoid conflicting with libpcap.a. e.g. ld: /usr/lib/gcc/x86_64-pc-linux-gnu/12/../../../../lib64/libpcap.a(pcap.o): in function `pcap_init': (.text+0x7f0): multiple definition of `pcap_init'; arch/um/drivers/pcap_kern.o:pcap_kern.c:(.text.unlikely+0x0): first defined here Signed-off-by: Peter Foley Signed-off-by: Richard Weinberger --- arch/um/drivers/pcap_kern.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/um/drivers/pcap_kern.c b/arch/um/drivers/pcap_kern.c index cfe4cb17694cc..25ee2c97ca21e 100644 --- a/arch/um/drivers/pcap_kern.c +++ b/arch/um/drivers/pcap_kern.c @@ -15,7 +15,7 @@ struct pcap_init { char *filter; }; -void pcap_init(struct net_device *dev, void *data) +void pcap_init_kern(struct net_device *dev, void *data) { struct uml_net_private *pri; struct pcap_data *ppri; @@ -44,7 +44,7 @@ static int pcap_write(int fd, struct sk_buff *skb, struct uml_net_private *lp) } static const struct net_kern_info pcap_kern_info = { - .init = pcap_init, + .init = pcap_init_kern, .protocol = eth_protocol, .read = pcap_read, .write = pcap_write, From 910dba41239224069c17bb6c30c9d43c1ba229f1 Mon Sep 17 00:00:00 2001 From: Peter Foley Date: Thu, 12 Jan 2023 23:49:09 -0500 Subject: [PATCH 254/765] um: Prevent building modules incompatible with MODVERSIONS The manual ld invocation in arch/um/drivers doesn't play nicely with genksyms. Given the problematic modules are deprecated anyway, just prevent building them when using MODVERSIONS. e.g. MODPOST Module.symvers arch/um/drivers/.pcap.o.cmd: No such file or directory Signed-off-by: Peter Foley Signed-off-by: Richard Weinberger --- arch/um/drivers/Kconfig | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/um/drivers/Kconfig b/arch/um/drivers/Kconfig index a4f0a19fbe14b..36911b1fddcf0 100644 --- a/arch/um/drivers/Kconfig +++ b/arch/um/drivers/Kconfig @@ -261,6 +261,7 @@ config UML_NET_VECTOR config UML_NET_VDE bool "VDE transport (obsolete)" depends on UML_NET + depends on !MODVERSIONS select MAY_HAVE_RUNTIME_DEPS help This User-Mode Linux network transport allows one or more running @@ -309,6 +310,7 @@ config UML_NET_MCAST config UML_NET_PCAP bool "pcap transport (obsolete)" depends on UML_NET + depends on !MODVERSIONS select MAY_HAVE_RUNTIME_DEPS help The pcap transport makes a pcap packet stream on the host look From 6aa56115c73b37270e53aa91984bdb8b60164ec7 Mon Sep 17 00:00:00 2001 From: Peter Foley Date: Thu, 12 Jan 2023 23:49:10 -0500 Subject: [PATCH 255/765] um: Use CFLAGS_vmlinux link-vmlinux.sh doesn't use LDFLAGS_vmlinux when linking the kernel for UML. Move the LDFLAGS_EXESTACK options into CFLAGS_vmlinux so they're actually respected. e.g. /usr/lib/gcc/x86_64-pc-linux-gnu/12/../../../../x86_64-pc-linux-gnu/bin/ld: warning: .tmp_vmlinux.kallsyms3.o: missing .note.GNU-stack section implies executable stack /usr/lib/gcc/x86_64-pc-linux-gnu/12/../../../../x86_64-pc-linux-gnu/bin/ld: NOTE: This behaviour is deprecated and will be removed in a future version of the linker /usr/lib/gcc/x86_64-pc-linux-gnu/12/../../../../x86_64-pc-linux-gnu/bin/ld: warning: vmlinux has a LOAD segment with RWX permissions Reviewed-by: David Gow Signed-off-by: Peter Foley Signed-off-by: Richard Weinberger --- arch/um/Makefile | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/um/Makefile b/arch/um/Makefile index ae321282dc6f6..d37cac5d9f430 100644 --- a/arch/um/Makefile +++ b/arch/um/Makefile @@ -141,11 +141,10 @@ ifeq ($(CONFIG_LD_IS_BFD),y) LDFLAGS_EXECSTACK += $(call ld-option,--no-warn-rwx-segments) endif -LD_FLAGS_CMDLINE = $(foreach opt,$(KBUILD_LDFLAGS),-Wl,$(opt)) +LD_FLAGS_CMDLINE = $(foreach opt,$(KBUILD_LDFLAGS) $(LDFLAGS_EXECSTACK),-Wl,$(opt)) # Used by link-vmlinux.sh which has special support for um link export CFLAGS_vmlinux := $(LINK-y) $(LINK_WRAPS) $(LD_FLAGS_CMDLINE) -export LDFLAGS_vmlinux := $(LDFLAGS_EXECSTACK) # When cleaning we don't include .config, so we don't include # TT or skas makefiles and don't clean skas_ptregs.h. From f09c3fcf67a3294c981b861981143e38895c5a59 Mon Sep 17 00:00:00 2001 From: Peter Foley Date: Fri, 13 Jan 2023 11:56:11 -0500 Subject: [PATCH 256/765] um: put power options in a menu Because having them all dumped at top-level is a bit messy. Signed-off-by: Peter Foley Signed-off-by: Richard Weinberger --- arch/um/Kconfig | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/um/Kconfig b/arch/um/Kconfig index 4db186f019ae4..dc36ea6e508f2 100644 --- a/arch/um/Kconfig +++ b/arch/um/Kconfig @@ -243,4 +243,8 @@ source "arch/um/drivers/Kconfig" config ARCH_SUSPEND_POSSIBLE def_bool y +menu "Power management options" + source "kernel/power/Kconfig" + +endmenu From 83e913f52aba69149261742aa9ea4ceea7bf182d Mon Sep 17 00:00:00 2001 From: Peter Foley Date: Fri, 13 Jan 2023 12:03:00 -0500 Subject: [PATCH 257/765] um: Support LTO Only a handful of changes are necessary to get it to work. Signed-off-by: Peter Foley Signed-off-by: Richard Weinberger --- arch/um/Kconfig | 2 ++ arch/um/Makefile | 2 +- arch/x86/um/vdso/Makefile | 2 +- 3 files changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/um/Kconfig b/arch/um/Kconfig index dc36ea6e508f2..541a9b18e3435 100644 --- a/arch/um/Kconfig +++ b/arch/um/Kconfig @@ -25,6 +25,8 @@ config UML select GENERIC_IRQ_SHOW select GENERIC_CPU_DEVICES select HAVE_GCC_PLUGINS + select ARCH_SUPPORTS_LTO_CLANG + select ARCH_SUPPORTS_LTO_CLANG_THIN select TRACE_IRQFLAGS_SUPPORT select TTY # Needed for line.c select HAVE_ARCH_VMAP_STACK diff --git a/arch/um/Makefile b/arch/um/Makefile index d37cac5d9f430..8186d4761bda6 100644 --- a/arch/um/Makefile +++ b/arch/um/Makefile @@ -144,7 +144,7 @@ endif LD_FLAGS_CMDLINE = $(foreach opt,$(KBUILD_LDFLAGS) $(LDFLAGS_EXECSTACK),-Wl,$(opt)) # Used by link-vmlinux.sh which has special support for um link -export CFLAGS_vmlinux := $(LINK-y) $(LINK_WRAPS) $(LD_FLAGS_CMDLINE) +export CFLAGS_vmlinux := $(LINK-y) $(LINK_WRAPS) $(LD_FLAGS_CMDLINE) $(CC_FLAGS_LTO) # When cleaning we don't include .config, so we don't include # TT or skas makefiles and don't clean skas_ptregs.h. diff --git a/arch/x86/um/vdso/Makefile b/arch/x86/um/vdso/Makefile index 6fbe97c52c991..6825e146a62ff 100644 --- a/arch/x86/um/vdso/Makefile +++ b/arch/x86/um/vdso/Makefile @@ -61,7 +61,7 @@ CFLAGS_REMOVE_um_vdso.o = -pg -fprofile-arcs -ftest-coverage # quiet_cmd_vdso = VDSO $@ cmd_vdso = $(CC) -nostdlib -o $@ \ - $(VDSO_LDFLAGS) $(VDSO_LDFLAGS_$(filter %.lds,$(^F))) \ + $(CC_FLAGS_LTO) $(VDSO_LDFLAGS) $(VDSO_LDFLAGS_$(filter %.lds,$(^F))) \ -Wl,-T,$(filter %.lds,$^) $(filter %.o,$^) && \ sh $(srctree)/$(src)/checkundef.sh '$(NM)' '$@' From 314a1408b79a844dafdcde867d90de5d509409b7 Mon Sep 17 00:00:00 2001 From: Vincent Whitchurch Date: Fri, 20 Jan 2023 09:02:32 +0100 Subject: [PATCH 258/765] um: virt-pci: implement pcibios_get_phb_of_node() Implement pcibios_get_phb_of_node() as x86 does in order to allow PCI busses to be associated with devicetree nodes. Signed-off-by: Vincent Whitchurch Acked-by: Johannes Berg Signed-off-by: Richard Weinberger --- arch/um/drivers/virt-pci.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/arch/um/drivers/virt-pci.c b/arch/um/drivers/virt-pci.c index 3ac220dafec4a..6884e1be38e46 100644 --- a/arch/um/drivers/virt-pci.c +++ b/arch/um/drivers/virt-pci.c @@ -533,6 +533,25 @@ static void um_pci_irq_vq_cb(struct virtqueue *vq) } } +/* Copied from arch/x86/kernel/devicetree.c */ +struct device_node *pcibios_get_phb_of_node(struct pci_bus *bus) +{ + struct device_node *np; + + for_each_node_by_type(np, "pci") { + const void *prop; + unsigned int bus_min; + + prop = of_get_property(np, "bus-range", NULL); + if (!prop) + continue; + bus_min = be32_to_cpup(prop); + if (bus->number == bus_min) + return np; + } + return NULL; +} + static int um_pci_init_vqs(struct um_pci_device *dev) { struct virtqueue *vqs[2]; From 935f8f7a0123ffa0a2dd7234713dc2ebeeb08955 Mon Sep 17 00:00:00 2001 From: Vincent Whitchurch Date: Thu, 26 Jan 2023 11:47:10 +0100 Subject: [PATCH 259/765] um-virt-pci: Make max delay configurable There is a hard coded maximum time for which the driver busy-loops waiting for a response from the virtio device. This default time may be too short for some systems, so make it a module parameter so that it can be set via the kernel command line. Signed-off-by: Vincent Whitchurch Signed-off-by: Richard Weinberger --- arch/um/drivers/virt-pci.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/um/drivers/virt-pci.c b/arch/um/drivers/virt-pci.c index 6884e1be38e46..0aa71c08210de 100644 --- a/arch/um/drivers/virt-pci.c +++ b/arch/um/drivers/virt-pci.c @@ -54,7 +54,8 @@ static struct irq_domain *um_pci_inner_domain; static struct irq_domain *um_pci_msi_domain; static unsigned long um_pci_msi_used[BITS_TO_LONGS(MAX_MSI_VECTORS)]; -#define UM_VIRT_PCI_MAXDELAY 40000 +static unsigned int um_pci_max_delay_us = 40000; +module_param_named(max_delay_us, um_pci_max_delay_us, uint, 0644); struct um_pci_message_buffer { struct virtio_pcidev_msg hdr; @@ -155,7 +156,7 @@ static int um_pci_send_cmd(struct um_pci_device *dev, kfree(completed); if (WARN_ONCE(virtqueue_is_broken(dev->cmd_vq) || - ++delay_count > UM_VIRT_PCI_MAXDELAY, + ++delay_count > um_pci_max_delay_us, "um virt-pci delay: %d", delay_count)) { ret = -EIO; break; From 522c532c4fe730258118fb146577ec9bedf6b807 Mon Sep 17 00:00:00 2001 From: Vincent Whitchurch Date: Fri, 27 Jan 2023 15:30:27 +0100 Subject: [PATCH 260/765] virt-pci: add platform bus support This driver registers PCI busses, but the underlying virtio protocol could just as easily be used to provide a platform bus instead. If the virtio device node in the devicetree indicates that it's compatible with simple-bus, register platform devices instead of handling it as a PCI bus. Only one platform bus is allowed and the logic MMIO region for the platform bus is placed at an arbitrarily-chosen address away from the PCI region. Signed-off-by: Vincent Whitchurch Signed-off-by: Richard Weinberger --- arch/um/drivers/virt-pci.c | 91 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 91 insertions(+) diff --git a/arch/um/drivers/virt-pci.c b/arch/um/drivers/virt-pci.c index 0aa71c08210de..2ba89347053ac 100644 --- a/arch/um/drivers/virt-pci.c +++ b/arch/um/drivers/virt-pci.c @@ -8,6 +8,7 @@ #include #include #include +#include #include #include #include @@ -39,6 +40,8 @@ struct um_pci_device { unsigned long status; int irq; + + bool platform; }; struct um_pci_device_reg { @@ -48,6 +51,7 @@ struct um_pci_device_reg { static struct pci_host_bridge *bridge; static DEFINE_MUTEX(um_pci_mtx); +static struct um_pci_device *um_pci_platform_device; static struct um_pci_device_reg um_pci_devices[MAX_DEVICES]; static struct fwnode_handle *um_pci_fwnode; static struct irq_domain *um_pci_inner_domain; @@ -481,6 +485,9 @@ static void um_pci_handle_irq_message(struct virtqueue *vq, struct virtio_device *vdev = vq->vdev; struct um_pci_device *dev = vdev->priv; + if (!dev->irq) + return; + /* we should properly chain interrupts, but on ARCH=um we don't care */ switch (msg->op) { @@ -581,6 +588,55 @@ static int um_pci_init_vqs(struct um_pci_device *dev) return 0; } +static void __um_pci_virtio_platform_remove(struct virtio_device *vdev, + struct um_pci_device *dev) +{ + virtio_reset_device(vdev); + vdev->config->del_vqs(vdev); + + mutex_lock(&um_pci_mtx); + um_pci_platform_device = NULL; + mutex_unlock(&um_pci_mtx); + + kfree(dev); +} + +static int um_pci_virtio_platform_probe(struct virtio_device *vdev, + struct um_pci_device *dev) +{ + int ret; + + dev->platform = true; + + mutex_lock(&um_pci_mtx); + + if (um_pci_platform_device) { + mutex_unlock(&um_pci_mtx); + ret = -EBUSY; + goto out_free; + } + + ret = um_pci_init_vqs(dev); + if (ret) { + mutex_unlock(&um_pci_mtx); + goto out_free; + } + + um_pci_platform_device = dev; + + mutex_unlock(&um_pci_mtx); + + ret = of_platform_default_populate(vdev->dev.of_node, NULL, &vdev->dev); + if (ret) + __um_pci_virtio_platform_remove(vdev, dev); + + return ret; + +out_free: + kfree(dev); + return ret; +} + static int um_pci_virtio_probe(struct virtio_device *vdev) { struct um_pci_device *dev; @@ -594,6 +650,9 @@ static int um_pci_virtio_probe(struct virtio_device *vdev) dev->vdev = vdev; vdev->priv = dev; + if (of_device_is_compatible(vdev->dev.of_node, "simple-bus")) + return um_pci_virtio_platform_probe(vdev, dev); + mutex_lock(&um_pci_mtx); for (i = 0; i < MAX_DEVICES; i++) { if (um_pci_devices[i].dev) @@ -643,6 +702,12 @@ static void um_pci_virtio_remove(struct virtio_device *vdev) struct um_pci_device *dev = vdev->priv; int i; + if (dev->platform) { + of_platform_depopulate(&vdev->dev); + __um_pci_virtio_platform_remove(vdev, dev); + return; + } + /* Stop all virtqueues */ virtio_reset_device(vdev); vdev->config->del_vqs(vdev); @@ -880,6 +945,30 @@ void *pci_root_bus_fwnode(struct pci_bus *bus) return um_pci_fwnode; } +static long um_pci_map_platform(unsigned long offset, size_t size, + const struct logic_iomem_ops **ops, + void **priv) +{ + if (!um_pci_platform_device) + return -ENOENT; + + *ops = &um_pci_device_bar_ops; + *priv = &um_pci_platform_device->resptr[0]; + + return 0; +} + +static const struct logic_iomem_region_ops um_pci_platform_ops = { + .map = um_pci_map_platform, +}; + +static struct resource virt_platform_resource = { + .name = "platform", + .start = 0x10000000, + .end = 0x1fffffff, + .flags = IORESOURCE_MEM, +}; + static int __init um_pci_init(void) { int err, i; @@ -888,6 +977,8 @@ static int __init um_pci_init(void) &um_pci_cfgspace_ops)); WARN_ON(logic_iomem_add_region(&virt_iomem_resource, &um_pci_iomem_ops)); + WARN_ON(logic_iomem_add_region(&virt_platform_resource, + &um_pci_platform_ops)); if (WARN(CONFIG_UML_PCI_OVER_VIRTIO_DEVICE_ID < 0, "No virtio device ID configured for PCI - no PCI support\n")) From b99ddbe8336ee680257c8ab479f75051eaa49dcf Mon Sep 17 00:00:00 2001 From: Masahiro Yamada Date: Wed, 8 Feb 2023 01:41:56 +0900 Subject: [PATCH 261/765] UML: define RUNTIME_DISCARD_EXIT With CONFIG_VIRTIO_UML=y, GNU ld < 2.36 fails to link UML vmlinux (w/wo CONFIG_LD_SCRIPT_STATIC). `.exit.text' referenced in section `.uml.exitcall.exit' of arch/um/drivers/virtio_uml.o: defined in discarded section `.exit.text' of arch/um/drivers/virtio_uml.o collect2: error: ld returned 1 exit status This fix is similar to the following commits: - 4b9880dbf3bd ("powerpc/vmlinux.lds: Define RUNTIME_DISCARD_EXIT") - a494398bde27 ("s390: define RUNTIME_DISCARD_EXIT to fix link error with GNU ld < 2.36") - c1c551bebf92 ("sh: define RUNTIME_DISCARD_EXIT") Fixes: 99cb0d917ffa ("arch: fix broken BuildID for arm64 and riscv") Reported-by: SeongJae Park Signed-off-by: Masahiro Yamada Tested-by: SeongJae Park Signed-off-by: Richard Weinberger --- arch/um/kernel/vmlinux.lds.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/um/kernel/vmlinux.lds.S b/arch/um/kernel/vmlinux.lds.S index 16e49bfa2b426..53d719c04ba94 100644 --- a/arch/um/kernel/vmlinux.lds.S +++ b/arch/um/kernel/vmlinux.lds.S @@ -1,4 +1,4 @@ - +#define RUNTIME_DISCARD_EXIT KERNEL_STACK_SIZE = 4096 * (1 << CONFIG_KERNEL_STACK_ORDER); #ifdef CONFIG_LD_SCRIPT_STATIC From 8a6ca543646f2940832665dbf4e04105262505e2 Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Thu, 9 Feb 2023 10:00:02 +0100 Subject: [PATCH 262/765] um: virtio_uml: free command if adding to virtqueue failed If adding the command fails (i.e. the virtqueue is broken) then free it again if the function allocated a new buffer for it. Fixes: 68f5d3f3b654 ("um: add PCI over virtio emulation driver") Signed-off-by: Benjamin Berg Signed-off-by: Richard Weinberger --- arch/um/drivers/virt-pci.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/um/drivers/virt-pci.c b/arch/um/drivers/virt-pci.c index 2ba89347053ac..035c97aadb380 100644 --- a/arch/um/drivers/virt-pci.c +++ b/arch/um/drivers/virt-pci.c @@ -137,8 +137,11 @@ static int um_pci_send_cmd(struct um_pci_device *dev, out ? 1 : 0, posted ? cmd : HANDLE_NO_FREE(cmd), GFP_ATOMIC); - if (ret) + if (ret) { + if (posted) + kfree(cmd); goto out; + } if (posted) { virtqueue_kick(dev->cmd_vq); From 8e9cd85139a2149d5a7c121b05e0cdb8287311f9 Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Thu, 9 Feb 2023 10:00:03 +0100 Subject: [PATCH 263/765] um: virtio_uml: mark device as unregistered when breaking it Mark the device as not registered anymore when scheduling the work to remove it. Otherwise we could end up scheduling the work multiple times in a row, including scheduling it while it is already running. Fixes: af9fb41ed315 ("um: virtio_uml: Fix broken device handling in time-travel") Signed-off-by: Benjamin Berg Signed-off-by: Richard Weinberger --- arch/um/drivers/virtio_uml.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/um/drivers/virtio_uml.c b/arch/um/drivers/virtio_uml.c index 198aaed81ce60..ecfaae7096e85 100644 --- a/arch/um/drivers/virtio_uml.c +++ b/arch/um/drivers/virtio_uml.c @@ -168,6 +168,8 @@ static void vhost_user_check_reset(struct virtio_uml_device *vu_dev, if (!vu_dev->registered) return; + vu_dev->registered = 0; + virtio_break_device(&vu_dev->vdev); schedule_work(&pdata->conn_broken_wk); } From abdeb4fa5e1b5b4918034f02236fd886f40c20c1 Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Thu, 9 Feb 2023 10:00:04 +0100 Subject: [PATCH 264/765] um: virtio_uml: move device breaking into workqueue We should not be calling virtio_break_device from an IRQ context. Move breaking the device into the workqueue so that it is done from a reasonable context. Fixes: af9fb41ed315 ("um: virtio_uml: Fix broken device handling in time-travel") Signed-off-by: Benjamin Berg Signed-off-by: Richard Weinberger --- arch/um/drivers/virtio_uml.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) diff --git a/arch/um/drivers/virtio_uml.c b/arch/um/drivers/virtio_uml.c index ecfaae7096e85..8adca2000e519 100644 --- a/arch/um/drivers/virtio_uml.c +++ b/arch/um/drivers/virtio_uml.c @@ -170,7 +170,6 @@ static void vhost_user_check_reset(struct virtio_uml_device *vu_dev, vu_dev->registered = 0; - virtio_break_device(&vu_dev->vdev); schedule_work(&pdata->conn_broken_wk); } @@ -1138,6 +1137,15 @@ void virtio_uml_set_no_vq_suspend(struct virtio_device *vdev, static void vu_of_conn_broken(struct work_struct *wk) { + struct virtio_uml_platform_data *pdata; + struct virtio_uml_device *vu_dev; + + pdata = container_of(wk, struct virtio_uml_platform_data, conn_broken_wk); + + vu_dev = platform_get_drvdata(pdata->pdev); + + virtio_break_device(&vu_dev->vdev); + /* * We can't remove the device from the devicetree so the only thing we * can do is warn. @@ -1268,8 +1276,14 @@ static int vu_unregister_cmdline_device(struct device *dev, void *data) static void vu_conn_broken(struct work_struct *wk) { struct virtio_uml_platform_data *pdata; + struct virtio_uml_device *vu_dev; pdata = container_of(wk, struct virtio_uml_platform_data, conn_broken_wk); + + vu_dev = platform_get_drvdata(pdata->pdev); + + virtio_break_device(&vu_dev->vdev); + vu_unregister_cmdline_device(&pdata->pdev->dev, NULL); } From 339b84dcd7113dd076419ea2a47128cc53450305 Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Thu, 9 Feb 2023 10:00:05 +0100 Subject: [PATCH 265/765] um: virt-pci: properly remove PCI device from bus Triggering a bus rescan will not cause the PCI device to be removed. It is required to explicitly stop and remove the device from the bus. Fixes: 68f5d3f3b654 ("um: add PCI over virtio emulation driver") Signed-off-by: Benjamin Berg Signed-off-by: Richard Weinberger --- arch/um/drivers/virt-pci.c | 21 ++++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-) diff --git a/arch/um/drivers/virt-pci.c b/arch/um/drivers/virt-pci.c index 035c97aadb380..7699ca5f35d48 100644 --- a/arch/um/drivers/virt-pci.c +++ b/arch/um/drivers/virt-pci.c @@ -711,22 +711,33 @@ static void um_pci_virtio_remove(struct virtio_device *vdev) return; } - /* Stop all virtqueues */ - virtio_reset_device(vdev); - vdev->config->del_vqs(vdev); - device_set_wakeup_enable(&vdev->dev, false); mutex_lock(&um_pci_mtx); for (i = 0; i < MAX_DEVICES; i++) { if (um_pci_devices[i].dev != dev) continue; + um_pci_devices[i].dev = NULL; irq_free_desc(dev->irq); + + break; } mutex_unlock(&um_pci_mtx); - um_pci_rescan(); + if (i < MAX_DEVICES) { + struct pci_dev *pci_dev; + + pci_dev = pci_get_slot(bridge->bus, i); + if (pci_dev) + pci_stop_and_remove_bus_device_locked(pci_dev); + } + + /* Stop all virtqueues */ + virtio_reset_device(vdev); + dev->cmd_vq = NULL; + dev->irq_vq = NULL; + vdev->config->del_vqs(vdev); kfree(dev); } From 6c7667c9bc12778d5566137fd4b34f03e1f36909 Mon Sep 17 00:00:00 2001 From: Carlos Bilbao Date: Thu, 9 Feb 2023 12:16:38 -0600 Subject: [PATCH 266/765] uml: vector: Remove unused definitions VECTOR_{WRITE,HEADERS} Remove definitions of VECTOR_WRITE and VECTOR_HEADERS since no longer used. Signed-off-by: Carlos Bilbao Acked-By: Anton Ivanov Signed-off-by: Richard Weinberger --- arch/um/drivers/vector_user.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/arch/um/drivers/vector_user.h b/arch/um/drivers/vector_user.h index 3a73d17a0161d..59ed5f9e6e41d 100644 --- a/arch/um/drivers/vector_user.h +++ b/arch/um/drivers/vector_user.h @@ -68,8 +68,6 @@ struct vector_fds { }; #define VECTOR_READ 1 -#define VECTOR_WRITE (1 < 1) -#define VECTOR_HEADERS (1 < 2) extern struct arglist *uml_parse_vector_ifspec(char *arg); From 04df97e150c83d4640540008e95d0229cb188135 Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Mon, 13 Feb 2023 15:59:20 +0700 Subject: [PATCH 267/765] Documentation: rust: Fix arch support table Stephen Rothwell reported htmldocs warning when merging uml tree: Documentation/rust/arch-support.rst:20: WARNING: Blank line required after table. Fix the arch support table by removing extraneous simple table marker. Link: https://lore.kernel.org/linux-next/20230213152714.78b844f4@canb.auug.org.au/ Fixes: 0438aadfa69a34 ("rust: arch/um: Add support for CONFIG_RUST under x86_64 UML") Reported-by: Stephen Rothwell Signed-off-by: Bagas Sanjaya Signed-off-by: Richard Weinberger --- Documentation/rust/arch-support.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/rust/arch-support.rst b/Documentation/rust/arch-support.rst index a526ca1c688be..ed7f4f5b3cf15 100644 --- a/Documentation/rust/arch-support.rst +++ b/Documentation/rust/arch-support.rst @@ -16,6 +16,6 @@ support corresponds to ``S`` values in the ``MAINTAINERS`` file. Architecture Level of support Constraints ============ ================ ============================================== ``x86`` Maintained ``x86_64`` only. -============ ================ ============================================== ``um`` Maintained ``x86_64`` only. ============ ================ ============================================== + From 273a51e5521213a13e5852e762cc9c03c66b9baa Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Thu, 9 Feb 2023 03:20:10 +0000 Subject: [PATCH 268/765] f2fs: make kobj_type structures constant MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since commit ee6d3dd4ed48 ("driver core: make kobj_type constant.") the driver core allows the usage of const struct kobj_type. Take advantage of this to constify the structure definitions to prevent modification at runtime. Signed-off-by: Thomas Weißschuh Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/sysfs.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c index 6082e132257a3..0b19163c90d41 100644 --- a/fs/f2fs/sysfs.c +++ b/fs/f2fs/sysfs.c @@ -1162,13 +1162,13 @@ static const struct sysfs_ops f2fs_attr_ops = { .store = f2fs_attr_store, }; -static struct kobj_type f2fs_sb_ktype = { +static const struct kobj_type f2fs_sb_ktype = { .default_groups = f2fs_groups, .sysfs_ops = &f2fs_attr_ops, .release = f2fs_sb_release, }; -static struct kobj_type f2fs_ktype = { +static const struct kobj_type f2fs_ktype = { .sysfs_ops = &f2fs_attr_ops, }; @@ -1176,7 +1176,7 @@ static struct kset f2fs_kset = { .kobj = {.ktype = &f2fs_ktype}, }; -static struct kobj_type f2fs_feat_ktype = { +static const struct kobj_type f2fs_feat_ktype = { .default_groups = f2fs_feat_groups, .sysfs_ops = &f2fs_attr_ops, }; @@ -1217,7 +1217,7 @@ static const struct sysfs_ops f2fs_stat_attr_ops = { .store = f2fs_stat_attr_store, }; -static struct kobj_type f2fs_stat_ktype = { +static const struct kobj_type f2fs_stat_ktype = { .default_groups = f2fs_stat_groups, .sysfs_ops = &f2fs_stat_attr_ops, .release = f2fs_stat_kobj_release, @@ -1244,7 +1244,7 @@ static const struct sysfs_ops f2fs_feature_list_attr_ops = { .show = f2fs_sb_feat_attr_show, }; -static struct kobj_type f2fs_feature_list_ktype = { +static const struct kobj_type f2fs_feature_list_ktype = { .default_groups = f2fs_sb_feat_groups, .sysfs_ops = &f2fs_feature_list_attr_ops, .release = f2fs_feature_list_kobj_release, From 1ecf9e390452e73a362ea7fbde8f3f0db83de856 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Thu, 22 Dec 2022 19:33:04 +0000 Subject: [PATCH 269/765] mtd: ubi: wire-up parent MTD device Wire up the device parent pointer of UBI devices to their lower MTD device, typically an MTD partition or whole-chip device. The most noticeable change is that in sysfs, previously ubi devices would be could in /sys/devices/virtual/ubi while after this change they would be correctly attached to their parent MTD device, e.g. /sys/devices/platform/1100d000.spi/spi_master/spi1/spi1.0/mtd/mtd2/ubi0. Locating UBI devices using /sys/class/ubi/ of course still works as well. Signed-off-by: Daniel Golle Signed-off-by: Richard Weinberger --- drivers/mtd/ubi/build.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index 5e90a4423b699..0904eb40c95fa 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -915,6 +915,7 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, ubi->dev.release = dev_release; ubi->dev.class = &ubi_class; ubi->dev.groups = ubi_dev_groups; + ubi->dev.parent = &mtd->dev; ubi->mtd = mtd; ubi->ubi_num = ubi_num; From 05b8773ca33253ea562be145cf3145b05ef19f86 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Thu, 22 Dec 2022 19:33:31 +0000 Subject: [PATCH 270/765] mtd: ubi: block: wire-up device parent ubiblock devices were previously only identifyable by their name, but not connected to their parent UBI volume device e.g. in sysfs. Properly parent ubiblock device as descendant of a UBI volume device to reflect device model hierachy. Signed-off-by: Daniel Golle Signed-off-by: Richard Weinberger --- drivers/mtd/ubi/block.c | 2 +- drivers/mtd/ubi/kapi.c | 1 + include/linux/mtd/ubi.h | 1 + 3 files changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/mtd/ubi/block.c b/drivers/mtd/ubi/block.c index f5d036203fe7f..6a9eb2c860b5a 100644 --- a/drivers/mtd/ubi/block.c +++ b/drivers/mtd/ubi/block.c @@ -419,7 +419,7 @@ int ubiblock_create(struct ubi_volume_info *vi) list_add_tail(&dev->list, &ubiblock_devices); /* Must be the last step: anyone can call file ops from now on */ - ret = add_disk(dev->gd); + ret = device_add_disk(vi->dev, dev->gd, NULL); if (ret) goto out_remove_minor; diff --git a/drivers/mtd/ubi/kapi.c b/drivers/mtd/ubi/kapi.c index 0fce99ff29b58..5db653eacbd45 100644 --- a/drivers/mtd/ubi/kapi.c +++ b/drivers/mtd/ubi/kapi.c @@ -79,6 +79,7 @@ void ubi_do_get_volume_info(struct ubi_device *ubi, struct ubi_volume *vol, vi->name_len = vol->name_len; vi->name = vol->name; vi->cdev = vol->cdev.dev; + vi->dev = &vol->dev; } /** diff --git a/include/linux/mtd/ubi.h b/include/linux/mtd/ubi.h index 7d48ea368c5e5..a529347fd75b2 100644 --- a/include/linux/mtd/ubi.h +++ b/include/linux/mtd/ubi.h @@ -110,6 +110,7 @@ struct ubi_volume_info { int name_len; const char *name; dev_t cdev; + struct device *dev; }; /** From 22d74bc26bbfde150f602b2a6e93fb11df30ce0b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Fri, 10 Feb 2023 02:16:13 +0000 Subject: [PATCH 271/765] ubifs: make kobj_type structures constant MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since commit ee6d3dd4ed48 ("driver core: make kobj_type constant.") the driver core allows the usage of const struct kobj_type. Take advantage of this to constify the structure definitions to prevent modification at runtime. Signed-off-by: Thomas Weißschuh Reviewed-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- fs/ubifs/sysfs.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/ubifs/sysfs.c b/fs/ubifs/sysfs.c index 54270ad36321e..1c958148bb877 100644 --- a/fs/ubifs/sysfs.c +++ b/fs/ubifs/sysfs.c @@ -74,13 +74,13 @@ static const struct sysfs_ops ubifs_attr_ops = { .show = ubifs_attr_show, }; -static struct kobj_type ubifs_sb_ktype = { +static const struct kobj_type ubifs_sb_ktype = { .default_groups = ubifs_groups, .sysfs_ops = &ubifs_attr_ops, .release = ubifs_sb_release, }; -static struct kobj_type ubifs_ktype = { +static const struct kobj_type ubifs_ktype = { .sysfs_ops = &ubifs_attr_ops, }; From 8fcf2d012c8641c18adcd139dba6a1e556338d36 Mon Sep 17 00:00:00 2001 From: Harshit Mogalapalli Date: Tue, 14 Feb 2023 01:38:01 -0800 Subject: [PATCH 272/765] ubi: block: Fix a possible use-after-free bug in ubiblock_create() Smatch warns: drivers/mtd/ubi/block.c:438 ubiblock_create() warn: '&dev->list' not removed from list 'dev' is freed in 'out_free_dev:, but it is still on the list. To fix this, delete the list item before freeing. Fixes: 91cc8fbcc8c7 ("ubi: block: set BLK_MQ_F_BLOCKING") Signed-off-by: Harshit Mogalapalli Reviewed-by: Christoph Hellwig Signed-off-by: Richard Weinberger --- drivers/mtd/ubi/block.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/mtd/ubi/block.c b/drivers/mtd/ubi/block.c index 6a9eb2c860b5a..1de87062c67b9 100644 --- a/drivers/mtd/ubi/block.c +++ b/drivers/mtd/ubi/block.c @@ -429,6 +429,7 @@ int ubiblock_create(struct ubi_volume_info *vi) return 0; out_remove_minor: + list_del(&dev->list); idr_remove(&ubiblock_minor_idr, gd->first_minor); out_cleanup_disk: put_disk(dev->gd); From f2e357893cb7d15994e4ec10838ebb4dccf7eb6e Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Mon, 13 Feb 2023 22:18:24 +0800 Subject: [PATCH 273/765] f2fs: export ipu policy in debugfs Export ipu_policy as a string in debugfs for better readability and it can help us better understand some strategies of the file system. Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/debug.c | 24 ++++++++++++++++++++++++ fs/f2fs/segment.h | 1 + 2 files changed, 25 insertions(+) diff --git a/fs/f2fs/debug.c b/fs/f2fs/debug.c index 32af4f0c57357..ff5995cb95603 100644 --- a/fs/f2fs/debug.c +++ b/fs/f2fs/debug.c @@ -354,6 +354,17 @@ static char *s_flag[] = { [SBI_IS_FREEZING] = " freezefs", }; +static const char *ipu_mode_names[F2FS_IPU_MAX] = { + [F2FS_IPU_FORCE] = "FORCE", + [F2FS_IPU_SSR] = "SSR", + [F2FS_IPU_UTIL] = "UTIL", + [F2FS_IPU_SSR_UTIL] = "SSR_UTIL", + [F2FS_IPU_FSYNC] = "FSYNC", + [F2FS_IPU_ASYNC] = "ASYNC", + [F2FS_IPU_NOCACHE] = "NOCACHE", + [F2FS_IPU_HONOR_OPU_WRITE] = "HONOR_OPU_WRITE", +}; + static int stat_show(struct seq_file *s, void *v) { struct f2fs_stat_info *si; @@ -384,6 +395,19 @@ static int stat_show(struct seq_file *s, void *v) seq_printf(s, "Current Time Sec: %llu / Mounted Time Sec: %llu\n\n", ktime_get_boottime_seconds(), SIT_I(si->sbi)->mounted_time); + + seq_puts(s, "Policy:\n"); + seq_puts(s, " - IPU: ["); + if (IS_F2FS_IPU_DISABLE(si->sbi)) { + seq_puts(s, " DISABLE"); + } else { + unsigned long policy = SM_I(si->sbi)->ipu_policy; + + for_each_set_bit(j, &policy, F2FS_IPU_MAX) + seq_printf(s, " %s", ipu_mode_names[j]); + } + seq_puts(s, " ]\n\n"); + if (test_opt(si->sbi, DISCARD)) seq_printf(s, "Utilization: %u%% (%u valid blocks, %u discard blocks)\n", si->utilization, si->valid_count, si->discard_blks); diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h index 8ee5e5db9287c..92c8be00d3967 100644 --- a/fs/f2fs/segment.h +++ b/fs/f2fs/segment.h @@ -672,6 +672,7 @@ static inline int utilization(struct f2fs_sb_info *sbi) #define F2FS_IPU_DISABLE 0 +/* Modification on enum should be synchronized with ipu_mode_names array */ enum { F2FS_IPU_FORCE, F2FS_IPU_SSR, From dda7d77bcd424ce5da80dd5a7ec92ead00279c02 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Mon, 13 Feb 2023 22:18:25 +0800 Subject: [PATCH 274/765] f2fs: replace si->sbi w/ sbi in stat_show() For each loop add a local f2fs_sb_info pointer insted of looking it up. Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/debug.c | 44 +++++++++++++++++++++++--------------------- 1 file changed, 23 insertions(+), 21 deletions(-) diff --git a/fs/f2fs/debug.c b/fs/f2fs/debug.c index ff5995cb95603..30a77936e3c59 100644 --- a/fs/f2fs/debug.c +++ b/fs/f2fs/debug.c @@ -373,16 +373,18 @@ static int stat_show(struct seq_file *s, void *v) raw_spin_lock_irqsave(&f2fs_stat_lock, flags); list_for_each_entry(si, &f2fs_stat_list, stat_list) { - update_general_status(si->sbi); + struct f2fs_sb_info *sbi = si->sbi; + + update_general_status(sbi); seq_printf(s, "\n=====[ partition info(%pg). #%d, %s, CP: %s]=====\n", - si->sbi->sb->s_bdev, i++, - f2fs_readonly(si->sbi->sb) ? "RO" : "RW", - is_set_ckpt_flags(si->sbi, CP_DISABLED_FLAG) ? - "Disabled" : (f2fs_cp_error(si->sbi) ? "Error" : "Good")); - if (si->sbi->s_flag) { + sbi->sb->s_bdev, i++, + f2fs_readonly(sbi->sb) ? "RO" : "RW", + is_set_ckpt_flags(sbi, CP_DISABLED_FLAG) ? + "Disabled" : (f2fs_cp_error(sbi) ? "Error" : "Good")); + if (sbi->s_flag) { seq_puts(s, "[SBI:"); - for_each_set_bit(j, &si->sbi->s_flag, 32) + for_each_set_bit(j, &sbi->s_flag, 32) seq_puts(s, s_flag[j]); seq_puts(s, "]\n"); } @@ -394,21 +396,21 @@ static int stat_show(struct seq_file *s, void *v) si->overp_segs, si->rsvd_segs); seq_printf(s, "Current Time Sec: %llu / Mounted Time Sec: %llu\n\n", ktime_get_boottime_seconds(), - SIT_I(si->sbi)->mounted_time); + SIT_I(sbi)->mounted_time); seq_puts(s, "Policy:\n"); seq_puts(s, " - IPU: ["); - if (IS_F2FS_IPU_DISABLE(si->sbi)) { + if (IS_F2FS_IPU_DISABLE(sbi)) { seq_puts(s, " DISABLE"); } else { - unsigned long policy = SM_I(si->sbi)->ipu_policy; + unsigned long policy = SM_I(sbi)->ipu_policy; for_each_set_bit(j, &policy, F2FS_IPU_MAX) seq_printf(s, " %s", ipu_mode_names[j]); } seq_puts(s, " ]\n\n"); - if (test_opt(si->sbi, DISCARD)) + if (test_opt(sbi, DISCARD)) seq_printf(s, "Utilization: %u%% (%u valid blocks, %u discard blocks)\n", si->utilization, si->valid_count, si->discard_blks); else @@ -515,15 +517,15 @@ static int stat_show(struct seq_file *s, void *v) seq_printf(s, " - node segments : %d (%d)\n", si->node_segs, si->bg_node_segs); seq_puts(s, " - Reclaimed segs :\n"); - seq_printf(s, " - Normal : %d\n", si->sbi->gc_reclaimed_segs[GC_NORMAL]); - seq_printf(s, " - Idle CB : %d\n", si->sbi->gc_reclaimed_segs[GC_IDLE_CB]); + seq_printf(s, " - Normal : %d\n", sbi->gc_reclaimed_segs[GC_NORMAL]); + seq_printf(s, " - Idle CB : %d\n", sbi->gc_reclaimed_segs[GC_IDLE_CB]); seq_printf(s, " - Idle Greedy : %d\n", - si->sbi->gc_reclaimed_segs[GC_IDLE_GREEDY]); - seq_printf(s, " - Idle AT : %d\n", si->sbi->gc_reclaimed_segs[GC_IDLE_AT]); + sbi->gc_reclaimed_segs[GC_IDLE_GREEDY]); + seq_printf(s, " - Idle AT : %d\n", sbi->gc_reclaimed_segs[GC_IDLE_AT]); seq_printf(s, " - Urgent High : %d\n", - si->sbi->gc_reclaimed_segs[GC_URGENT_HIGH]); - seq_printf(s, " - Urgent Mid : %d\n", si->sbi->gc_reclaimed_segs[GC_URGENT_MID]); - seq_printf(s, " - Urgent Low : %d\n", si->sbi->gc_reclaimed_segs[GC_URGENT_LOW]); + sbi->gc_reclaimed_segs[GC_URGENT_HIGH]); + seq_printf(s, " - Urgent Mid : %d\n", sbi->gc_reclaimed_segs[GC_URGENT_MID]); + seq_printf(s, " - Urgent Low : %d\n", sbi->gc_reclaimed_segs[GC_URGENT_LOW]); seq_printf(s, "Try to move %d blocks (BG: %d)\n", si->tot_blks, si->bg_data_blks + si->bg_node_blks); seq_printf(s, " - data blocks : %d (%d)\n", si->data_blks, @@ -589,7 +591,7 @@ static int stat_show(struct seq_file *s, void *v) si->ndirty_imeta); seq_printf(s, " - fsync mark: %4lld\n", percpu_counter_sum_positive( - &si->sbi->rf_node_block_count)); + &sbi->rf_node_block_count)); seq_printf(s, " - NATs: %9d/%9d\n - SITs: %9d/%9d\n", si->dirty_nats, si->nats, si->dirty_sits, si->sits); seq_printf(s, " - free_nids: %9d/%9d\n - alloc_nids: %9d\n", @@ -616,12 +618,12 @@ static int stat_show(struct seq_file *s, void *v) si->block_count[LFS], si->segment_count[LFS]); /* segment usage info */ - f2fs_update_sit_info(si->sbi); + f2fs_update_sit_info(sbi); seq_printf(s, "\nBDF: %u, avg. vblocks: %u\n", si->bimodal, si->avg_vblocks); /* memory footprint */ - update_mem_info(si->sbi); + update_mem_info(sbi); seq_printf(s, "\nMemory: %llu KB\n", (si->base_mem + si->cache_mem + si->page_mem) >> 10); seq_printf(s, " - static: %llu KB\n", From 7e986855fe13de2c8290c1102292d8e5f29dd769 Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Mon, 13 Feb 2023 09:41:33 -0800 Subject: [PATCH 275/765] f2fs: fix wrong segment count MAIN_SEGS is for data area, while TOTAL_SEGS includes data and metadata. Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.h | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h index 92c8be00d3967..efdb7fc3b7975 100644 --- a/fs/f2fs/segment.h +++ b/fs/f2fs/segment.h @@ -719,9 +719,10 @@ static inline unsigned char curseg_alloc_type(struct f2fs_sb_info *sbi, return curseg->alloc_type; } -static inline void check_seg_range(struct f2fs_sb_info *sbi, unsigned int segno) +static inline bool valid_main_segno(struct f2fs_sb_info *sbi, + unsigned int segno) { - f2fs_bug_on(sbi, segno > TOTAL_SEGS(sbi) - 1); + return segno <= (MAIN_SEGS(sbi) - 1); } static inline void verify_fio_blkaddr(struct f2fs_io_info *fio) @@ -776,7 +777,7 @@ static inline int check_block_count(struct f2fs_sb_info *sbi, /* check segment usage, and check boundary of a given segment number */ if (unlikely(GET_SIT_VBLOCKS(raw_sit) > usable_blks_per_seg - || segno > TOTAL_SEGS(sbi) - 1)) { + || !valid_main_segno(sbi, segno))) { f2fs_err(sbi, "Wrong valid blocks %d or segno %u", GET_SIT_VBLOCKS(raw_sit), segno); set_sbi_flag(sbi, SBI_NEED_FSCK); @@ -793,7 +794,7 @@ static inline pgoff_t current_sit_addr(struct f2fs_sb_info *sbi, unsigned int offset = SIT_BLOCK_OFFSET(start); block_t blk_addr = sit_i->sit_base_addr + offset; - check_seg_range(sbi, start); + f2fs_bug_on(sbi, !valid_main_segno(sbi, start)); #ifdef CONFIG_F2FS_CHECK_FS if (f2fs_test_bit(offset, sit_i->sit_bitmap) != From a46bebd502fe1a3bd1d22f64cedd93e7e7702693 Mon Sep 17 00:00:00 2001 From: Daeho Jeong Date: Thu, 9 Feb 2023 10:18:19 -0800 Subject: [PATCH 276/765] f2fs: synchronize atomic write aborts To fix a race condition between atomic write aborts, I use the inode lock and make COW inode to be re-usable thoroughout the whole atomic file inode lifetime. Reported-by: syzbot+823000d23b3400619f7c@syzkaller.appspotmail.com Fixes: 3db1de0e582c ("f2fs: change the current atomic write way") Signed-off-by: Daeho Jeong Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 44 +++++++++++++++++++++++++++++--------------- fs/f2fs/inode.c | 11 +++++++++-- fs/f2fs/segment.c | 3 --- fs/f2fs/super.c | 2 -- 4 files changed, 38 insertions(+), 22 deletions(-) diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 300eae8b54154..6436c52e7913c 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -1863,7 +1863,10 @@ static int f2fs_release_file(struct inode *inode, struct file *filp) atomic_read(&inode->i_writecount) != 1) return 0; + inode_lock(inode); f2fs_abort_atomic_write(inode, true); + inode_unlock(inode); + return 0; } @@ -1878,8 +1881,12 @@ static int f2fs_file_flush(struct file *file, fl_owner_t id) * before dropping file lock, it needs to do in ->flush. */ if (F2FS_I(inode)->atomic_write_task == current && - (current->flags & PF_EXITING)) + (current->flags & PF_EXITING)) { + inode_lock(inode); f2fs_abort_atomic_write(inode, true); + inode_unlock(inode); + } + return 0; } @@ -2085,19 +2092,28 @@ static int f2fs_ioc_start_atomic_write(struct file *filp, bool truncate) goto out; } - /* Create a COW inode for atomic write */ - pinode = f2fs_iget(inode->i_sb, fi->i_pino); - if (IS_ERR(pinode)) { - f2fs_up_write(&fi->i_gc_rwsem[WRITE]); - ret = PTR_ERR(pinode); - goto out; - } + /* Check if the inode already has a COW inode */ + if (fi->cow_inode == NULL) { + /* Create a COW inode for atomic write */ + pinode = f2fs_iget(inode->i_sb, fi->i_pino); + if (IS_ERR(pinode)) { + f2fs_up_write(&fi->i_gc_rwsem[WRITE]); + ret = PTR_ERR(pinode); + goto out; + } - ret = f2fs_get_tmpfile(mnt_userns, pinode, &fi->cow_inode); - iput(pinode); - if (ret) { - f2fs_up_write(&fi->i_gc_rwsem[WRITE]); - goto out; + ret = f2fs_get_tmpfile(mnt_userns, pinode, &fi->cow_inode); + iput(pinode); + if (ret) { + f2fs_up_write(&fi->i_gc_rwsem[WRITE]); + goto out; + } + + set_inode_flag(fi->cow_inode, FI_COW_FILE); + clear_inode_flag(fi->cow_inode, FI_INLINE_DATA); + } else { + /* Reuse the already created COW inode */ + f2fs_do_truncate_blocks(fi->cow_inode, 0, true); } f2fs_write_inode(inode, NULL); @@ -2105,8 +2121,6 @@ static int f2fs_ioc_start_atomic_write(struct file *filp, bool truncate) stat_inc_atomic_inode(inode); set_inode_flag(inode, FI_ATOMIC_FILE); - set_inode_flag(fi->cow_inode, FI_COW_FILE); - clear_inode_flag(fi->cow_inode, FI_INLINE_DATA); isize = i_size_read(inode); fi->original_i_size = isize; diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c index 2c250120d5da1..7d2e2c0dba65c 100644 --- a/fs/f2fs/inode.c +++ b/fs/f2fs/inode.c @@ -767,11 +767,18 @@ int f2fs_write_inode(struct inode *inode, struct writeback_control *wbc) void f2fs_evict_inode(struct inode *inode) { struct f2fs_sb_info *sbi = F2FS_I_SB(inode); - nid_t xnid = F2FS_I(inode)->i_xattr_nid; + struct f2fs_inode_info *fi = F2FS_I(inode); + nid_t xnid = fi->i_xattr_nid; int err = 0; f2fs_abort_atomic_write(inode, true); + if (fi->cow_inode) { + clear_inode_flag(fi->cow_inode, FI_COW_FILE); + iput(fi->cow_inode); + fi->cow_inode = NULL; + } + trace_f2fs_evict_inode(inode); truncate_inode_pages_final(&inode->i_data); @@ -856,7 +863,7 @@ void f2fs_evict_inode(struct inode *inode) stat_dec_inline_inode(inode); stat_dec_compr_inode(inode); stat_sub_compr_blocks(inode, - atomic_read(&F2FS_I(inode)->i_compr_blocks)); + atomic_read(&fi->i_compr_blocks)); if (likely(!f2fs_cp_error(sbi) && !is_sbi_flag_set(sbi, SBI_CP_DISABLED))) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index ead3f35f501d7..719329c1808c1 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -192,9 +192,6 @@ void f2fs_abort_atomic_write(struct inode *inode, bool clean) if (!f2fs_is_atomic_file(inode)) return; - clear_inode_flag(fi->cow_inode, FI_COW_FILE); - iput(fi->cow_inode); - fi->cow_inode = NULL; release_atomic_write_cnt(inode); clear_inode_flag(inode, FI_ATOMIC_COMMITTED); clear_inode_flag(inode, FI_ATOMIC_REPLACE); diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index c11a161ba5bec..aa55dc12aff2e 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -1427,8 +1427,6 @@ static int f2fs_drop_inode(struct inode *inode) atomic_inc(&inode->i_count); spin_unlock(&inode->i_lock); - f2fs_abort_atomic_write(inode, true); - /* should remain fi->extent_tree for writepage */ f2fs_destroy_extent_node(inode); From 60b730a40c43fbcc034970d3e77eb0f25b8cc1cf Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Tue, 14 Feb 2023 17:51:35 -0800 Subject: [PATCH 277/765] xfs: fix uninitialized variable access If the end position of a GETFSMAP query overlaps an allocated space and we're using the free space info to generate fsmap info, the akeys information gets fed into the fsmap formatter with bad results. Zero-init the space. Reported-by: syzbot+090ae72d552e6bd93cfe@syzkaller.appspotmail.com Signed-off-by: Darrick J. Wong --- fs/xfs/xfs_fsmap.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/xfs/xfs_fsmap.c b/fs/xfs/xfs_fsmap.c index 120d284a03fe4..59e7d1a14b672 100644 --- a/fs/xfs/xfs_fsmap.c +++ b/fs/xfs/xfs_fsmap.c @@ -761,6 +761,7 @@ xfs_getfsmap_datadev_bnobt( { struct xfs_alloc_rec_incore akeys[2]; + memset(akeys, 0, sizeof(akeys)); info->missing_owner = XFS_FMR_OWN_UNKNOWN; return __xfs_getfsmap_datadev(tp, keys, info, xfs_getfsmap_datadev_bnobt_query, &akeys[0]); From 240930fb7e6b52229bdee5b1423bfeab0002fed2 Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Mon, 26 Dec 2022 14:20:15 +0800 Subject: [PATCH 278/765] ext4: dio take shared inode lock when overwriting preallocated blocks In the dio write path, we only take shared inode lock for the case of aligned overwriting initialized blocks inside EOF. But for overwriting preallocated blocks, it may only need to split unwritten extents, this procedure has been protected under i_data_sem lock, it's safe to release the exclusive inode lock and take shared inode lock. This could give a significant speed up for multi-threaded writes. Test on Intel Xeon Gold 6140 and nvme SSD with below fio parameters. direct=1 ioengine=libaio iodepth=10 numjobs=10 runtime=60 rw=randwrite size=100G And the test result are: Before: bs=4k IOPS=11.1k, BW=43.2MiB/s bs=16k IOPS=11.1k, BW=173MiB/s bs=64k IOPS=11.2k, BW=697MiB/s After: bs=4k IOPS=41.4k, BW=162MiB/s bs=16k IOPS=41.3k, BW=646MiB/s bs=64k IOPS=13.5k, BW=843MiB/s Signed-off-by: Zhang Yi Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20221226062015.3479416-1-yi.zhang@huaweicloud.com Signed-off-by: Theodore Ts'o --- fs/ext4/file.c | 34 ++++++++++++++++++++++------------ 1 file changed, 22 insertions(+), 12 deletions(-) diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 7ac0a81bd3713..6e9f198ecacfe 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -202,8 +202,9 @@ ext4_extending_io(struct inode *inode, loff_t offset, size_t len) return false; } -/* Is IO overwriting allocated and initialized blocks? */ -static bool ext4_overwrite_io(struct inode *inode, loff_t pos, loff_t len) +/* Is IO overwriting allocated or initialized blocks? */ +static bool ext4_overwrite_io(struct inode *inode, + loff_t pos, loff_t len, bool *unwritten) { struct ext4_map_blocks map; unsigned int blkbits = inode->i_blkbits; @@ -217,12 +218,15 @@ static bool ext4_overwrite_io(struct inode *inode, loff_t pos, loff_t len) blklen = map.m_len; err = ext4_map_blocks(NULL, inode, &map, 0); + if (err != blklen) + return false; /* * 'err==len' means that all of the blocks have been preallocated, - * regardless of whether they have been initialized or not. To exclude - * unwritten extents, we need to check m_flags. + * regardless of whether they have been initialized or not. We need to + * check m_flags to distinguish the unwritten extents. */ - return err == blklen && (map.m_flags & EXT4_MAP_MAPPED); + *unwritten = !(map.m_flags & EXT4_MAP_MAPPED); + return true; } static ssize_t ext4_generic_write_checks(struct kiocb *iocb, @@ -431,11 +435,16 @@ static const struct iomap_dio_ops ext4_dio_write_ops = { * - For extending writes case we don't take the shared lock, since it requires * updating inode i_disksize and/or orphan handling with exclusive lock. * - * - shared locking will only be true mostly with overwrites. Otherwise we will - * switch to exclusive i_rwsem lock. + * - shared locking will only be true mostly with overwrites, including + * initialized blocks and unwritten blocks. For overwrite unwritten blocks + * we protect splitting extents by i_data_sem in ext4_inode_info, so we can + * also release exclusive i_rwsem lock. + * + * - Otherwise we will switch to exclusive i_rwsem lock. */ static ssize_t ext4_dio_write_checks(struct kiocb *iocb, struct iov_iter *from, - bool *ilock_shared, bool *extend) + bool *ilock_shared, bool *extend, + bool *unwritten) { struct file *file = iocb->ki_filp; struct inode *inode = file_inode(file); @@ -459,7 +468,7 @@ static ssize_t ext4_dio_write_checks(struct kiocb *iocb, struct iov_iter *from, * in file_modified(). */ if (*ilock_shared && (!IS_NOSEC(inode) || *extend || - !ext4_overwrite_io(inode, offset, count))) { + !ext4_overwrite_io(inode, offset, count, unwritten))) { if (iocb->ki_flags & IOCB_NOWAIT) { ret = -EAGAIN; goto out; @@ -491,7 +500,7 @@ static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from) loff_t offset = iocb->ki_pos; size_t count = iov_iter_count(from); const struct iomap_ops *iomap_ops = &ext4_iomap_ops; - bool extend = false, unaligned_io = false; + bool extend = false, unaligned_io = false, unwritten = false; bool ilock_shared = true; /* @@ -534,7 +543,8 @@ static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from) return ext4_buffered_write_iter(iocb, from); } - ret = ext4_dio_write_checks(iocb, from, &ilock_shared, &extend); + ret = ext4_dio_write_checks(iocb, from, + &ilock_shared, &extend, &unwritten); if (ret <= 0) return ret; @@ -582,7 +592,7 @@ static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from) ext4_journal_stop(handle); } - if (ilock_shared) + if (ilock_shared && !unwritten) iomap_ops = &ext4_iomap_overwrite_ops; ret = iomap_dio_rw(iocb, from, iomap_ops, &ext4_dio_write_ops, (unaligned_io || extend) ? IOMAP_DIO_FORCE_WAIT : 0, From c7dbc06688292db34c1bb9c715e29ac4935af994 Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Tue, 14 Feb 2023 15:53:52 -0800 Subject: [PATCH 279/765] f2fs: Revert "f2fs: truncate blocks in batch in __complete_revoke_list()" We should not truncate replaced blocks, and were supposed to truncate the first part as well. This reverts commit 78a99fe6254cad4be310cd84af39f6c46b668c72. Cc: stable@vger.kernel.org Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 719329c1808c1..227e258361734 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -265,19 +265,24 @@ static void __complete_revoke_list(struct inode *inode, struct list_head *head, bool revoke) { struct revoke_entry *cur, *tmp; + pgoff_t start_index = 0; bool truncate = is_inode_flag_set(inode, FI_ATOMIC_REPLACE); list_for_each_entry_safe(cur, tmp, head, list) { - if (revoke) + if (revoke) { __replace_atomic_write_block(inode, cur->index, cur->old_addr, NULL, true); + } else if (truncate) { + f2fs_truncate_hole(inode, start_index, cur->index); + start_index = cur->index + 1; + } list_del(&cur->list); kmem_cache_free(revoke_entry_slab, cur); } if (!revoke && truncate) - f2fs_do_truncate_blocks(inode, 0, false); + f2fs_do_truncate_blocks(inode, start_index * PAGE_SIZE, false); } static int __f2fs_commit_atomic_write(struct inode *inode) From ddf1eca4fc5a4038cb323306f51fbba34ce3f4d2 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Wed, 15 Feb 2023 14:17:01 +0800 Subject: [PATCH 280/765] f2fs: drop unnecessary arg for f2fs_ioc_*() They are not used, let's remove them. Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 6436c52e7913c..ca1720fc1187e 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -2539,7 +2539,7 @@ static int f2fs_ioc_gc_range(struct file *filp, unsigned long arg) return __f2fs_ioc_gc_range(filp, &range); } -static int f2fs_ioc_write_checkpoint(struct file *filp, unsigned long arg) +static int f2fs_ioc_write_checkpoint(struct file *filp) { struct inode *inode = file_inode(filp); struct f2fs_sb_info *sbi = F2FS_I_SB(inode); @@ -3253,7 +3253,7 @@ int f2fs_precache_extents(struct inode *inode) return 0; } -static int f2fs_ioc_precache_extents(struct file *filp, unsigned long arg) +static int f2fs_ioc_precache_extents(struct file *filp) { return f2fs_precache_extents(file_inode(filp)); } @@ -4010,7 +4010,7 @@ static int redirty_blocks(struct inode *inode, pgoff_t page_idx, int len) return ret; } -static int f2fs_ioc_decompress_file(struct file *filp, unsigned long arg) +static int f2fs_ioc_decompress_file(struct file *filp) { struct inode *inode = file_inode(filp); struct f2fs_sb_info *sbi = F2FS_I_SB(inode); @@ -4083,7 +4083,7 @@ static int f2fs_ioc_decompress_file(struct file *filp, unsigned long arg) return ret; } -static int f2fs_ioc_compress_file(struct file *filp, unsigned long arg) +static int f2fs_ioc_compress_file(struct file *filp) { struct inode *inode = file_inode(filp); struct f2fs_sb_info *sbi = F2FS_I_SB(inode); @@ -4199,7 +4199,7 @@ static long __f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) case F2FS_IOC_GARBAGE_COLLECT_RANGE: return f2fs_ioc_gc_range(filp, arg); case F2FS_IOC_WRITE_CHECKPOINT: - return f2fs_ioc_write_checkpoint(filp, arg); + return f2fs_ioc_write_checkpoint(filp); case F2FS_IOC_DEFRAGMENT: return f2fs_ioc_defragment(filp, arg); case F2FS_IOC_MOVE_RANGE: @@ -4213,7 +4213,7 @@ static long __f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) case F2FS_IOC_SET_PIN_FILE: return f2fs_ioc_set_pin_file(filp, arg); case F2FS_IOC_PRECACHE_EXTENTS: - return f2fs_ioc_precache_extents(filp, arg); + return f2fs_ioc_precache_extents(filp); case F2FS_IOC_RESIZE_FS: return f2fs_ioc_resize_fs(filp, arg); case FS_IOC_ENABLE_VERITY: @@ -4239,9 +4239,9 @@ static long __f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) case F2FS_IOC_SET_COMPRESS_OPTION: return f2fs_ioc_set_compress_option(filp, arg); case F2FS_IOC_DECOMPRESS_FILE: - return f2fs_ioc_decompress_file(filp, arg); + return f2fs_ioc_decompress_file(filp); case F2FS_IOC_COMPRESS_FILE: - return f2fs_ioc_compress_file(filp, arg); + return f2fs_ioc_compress_file(filp); default: return -ENOTTY; } From 7fde88eda855952766a74026c181c6270b3392fc Mon Sep 17 00:00:00 2001 From: David Rau Date: Wed, 15 Feb 2023 10:10:45 +0000 Subject: [PATCH 281/765] ASoC: da7219: Improve the IRQ process to increase the stability Remove the sleep control in IRQ thread and create an individual task to handel it for Jack plug in event. This commit improves the control of ground switches in the AAD IRQ. Signed-off-by: David Rau Link: https://lore.kernel.org/r/20230215101045.21456-1-David.Rau.opensource@dm.renesas.com Signed-off-by: Mark Brown --- sound/soc/codecs/da7219-aad.c | 41 +++++++++++++++++++++++++++++------ sound/soc/codecs/da7219-aad.h | 4 ++++ 2 files changed, 38 insertions(+), 7 deletions(-) diff --git a/sound/soc/codecs/da7219-aad.c b/sound/soc/codecs/da7219-aad.c index c55b033d89da2..4a4f09f924bc5 100644 --- a/sound/soc/codecs/da7219-aad.c +++ b/sound/soc/codecs/da7219-aad.c @@ -339,11 +339,39 @@ static void da7219_aad_hptest_work(struct work_struct *work) SND_JACK_HEADSET | SND_JACK_LINEOUT); } +static void da7219_aad_jack_det_work(struct work_struct *work) +{ + struct da7219_aad_priv *da7219_aad = + container_of(work, struct da7219_aad_priv, jack_det_work); + struct snd_soc_component *component = da7219_aad->component; + u8 srm_st; + + mutex_lock(&da7219_aad->jack_det_mutex); + + srm_st = snd_soc_component_read(component, DA7219_PLL_SRM_STS) & DA7219_PLL_SRM_STS_MCLK; + msleep(da7219_aad->gnd_switch_delay * ((srm_st == 0x0) ? 2 : 1) - 4); + /* Enable ground switch */ + snd_soc_component_update_bits(component, 0xFB, 0x01, 0x01); + + mutex_unlock(&da7219_aad->jack_det_mutex); +} + /* * IRQ */ +static irqreturn_t da7219_aad_pre_irq_thread(int irq, void *data) +{ + + struct da7219_aad_priv *da7219_aad = data; + + if (!da7219_aad->jack_inserted) + schedule_work(&da7219_aad->jack_det_work); + + return IRQ_WAKE_THREAD; +} + static irqreturn_t da7219_aad_irq_thread(int irq, void *data) { struct da7219_aad_priv *da7219_aad = data; @@ -351,14 +379,9 @@ static irqreturn_t da7219_aad_irq_thread(int irq, void *data) struct snd_soc_dapm_context *dapm = snd_soc_component_get_dapm(component); struct da7219_priv *da7219 = snd_soc_component_get_drvdata(component); u8 events[DA7219_AAD_IRQ_REG_MAX]; - u8 statusa, srm_st; + u8 statusa; int i, report = 0, mask = 0; - srm_st = snd_soc_component_read(component, DA7219_PLL_SRM_STS) & DA7219_PLL_SRM_STS_MCLK; - msleep(da7219_aad->gnd_switch_delay * ((srm_st == 0x0) ? 2 : 1) - 4); - /* Enable ground switch */ - snd_soc_component_update_bits(component, 0xFB, 0x01, 0x01); - /* Read current IRQ events */ regmap_bulk_read(da7219->regmap, DA7219_ACCDET_IRQ_EVENT_A, events, DA7219_AAD_IRQ_REG_MAX); @@ -377,6 +400,9 @@ static irqreturn_t da7219_aad_irq_thread(int irq, void *data) events[DA7219_AAD_IRQ_REG_A], events[DA7219_AAD_IRQ_REG_B], statusa); + if (!da7219_aad->jack_inserted) + cancel_work_sync(&da7219_aad->jack_det_work); + if (statusa & DA7219_JACK_INSERTION_STS_MASK) { /* Jack Insertion */ if (events[DA7219_AAD_IRQ_REG_A] & @@ -940,8 +966,9 @@ int da7219_aad_init(struct snd_soc_component *component) INIT_WORK(&da7219_aad->btn_det_work, da7219_aad_btn_det_work); INIT_WORK(&da7219_aad->hptest_work, da7219_aad_hptest_work); + INIT_WORK(&da7219_aad->jack_det_work, da7219_aad_jack_det_work); - ret = request_threaded_irq(da7219_aad->irq, NULL, + ret = request_threaded_irq(da7219_aad->irq, da7219_aad_pre_irq_thread, da7219_aad_irq_thread, IRQF_TRIGGER_LOW | IRQF_ONESHOT, "da7219-aad", da7219_aad); diff --git a/sound/soc/codecs/da7219-aad.h b/sound/soc/codecs/da7219-aad.h index 21fdf53095cc2..be87ee47edde1 100644 --- a/sound/soc/codecs/da7219-aad.h +++ b/sound/soc/codecs/da7219-aad.h @@ -11,6 +11,7 @@ #define __DA7219_AAD_H #include +#include #include #include #include @@ -196,6 +197,9 @@ struct da7219_aad_priv { struct work_struct btn_det_work; struct work_struct hptest_work; + struct work_struct jack_det_work; + + struct mutex jack_det_mutex; struct snd_soc_jack *jack; bool micbias_resume_enable; From 4932b1fa61322b737dc3615a298aafdc42f97f79 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Wed, 15 Feb 2023 15:17:41 +0100 Subject: [PATCH 282/765] ASoC: rsnd: adg: Fix BRG typos "BRG" stands for "Baud Rate Generator", but is frequently misspelled as "RBG". Signed-off-by: Geert Uytterhoeven Acked-by: Kuninori Morimoto Link: https://lore.kernel.org/r/ac6365c17861d71fbc89d823089db4aafdb763ed.1676470202.git.geert+renesas@glider.be Signed-off-by: Mark Brown --- sound/soc/sh/rcar/adg.c | 64 ++++++++++++++++++++--------------------- 1 file changed, 32 insertions(+), 32 deletions(-) diff --git a/sound/soc/sh/rcar/adg.c b/sound/soc/sh/rcar/adg.c index 7bc4421835d72..0b8926600d900 100644 --- a/sound/soc/sh/rcar/adg.c +++ b/sound/soc/sh/rcar/adg.c @@ -39,10 +39,10 @@ struct rsnd_adg { int clkin_size; int clkout_size; u32 ckr; - u32 rbga; - u32 rbgb; + u32 brga; + u32 brgb; - int rbg_rate[ADG_HZ_SIZE]; /* RBGA / RBGB */ + int brg_rate[ADG_HZ_SIZE]; /* BRGA / BRGB */ }; #define for_each_rsnd_clkin(pos, adg, i) \ @@ -75,7 +75,7 @@ static const char * const clkout_name_gen2[] = { [CLKOUT3] = "audio_clkout3", }; -static u32 rsnd_adg_calculate_rbgx(unsigned long div) +static u32 rsnd_adg_calculate_brgx(unsigned long div) { int i; @@ -131,8 +131,8 @@ static void __rsnd_adg_get_timesel_ratio(struct rsnd_priv *priv, adg->clkin_rate[CLKA], /* 0000: CLKA */ adg->clkin_rate[CLKB], /* 0001: CLKB */ adg->clkin_rate[CLKC], /* 0010: CLKC */ - adg->rbg_rate[ADG_HZ_441], /* 0011: RBGA */ - adg->rbg_rate[ADG_HZ_48], /* 0100: RBGB */ + adg->brg_rate[ADG_HZ_441], /* 0011: BRGA */ + adg->brg_rate[ADG_HZ_48], /* 0100: BRGB */ }; min = ~0; @@ -323,10 +323,10 @@ int rsnd_adg_clk_query(struct rsnd_priv *priv, unsigned int rate) /* * find divided clock from BRGA/BRGB */ - if (rate == adg->rbg_rate[ADG_HZ_441]) + if (rate == adg->brg_rate[ADG_HZ_441]) return 0x10; - if (rate == adg->rbg_rate[ADG_HZ_48]) + if (rate == adg->brg_rate[ADG_HZ_48]) return 0x20; return -EIO; @@ -358,13 +358,13 @@ int rsnd_adg_ssi_clk_try_start(struct rsnd_mod *ssi_mod, unsigned int rate) ckr = 0x80000000; /* BRGB output = 48kHz */ rsnd_mod_bset(adg_mod, BRGCKR, 0x80770000, adg->ckr | ckr); - rsnd_mod_write(adg_mod, BRRA, adg->rbga); - rsnd_mod_write(adg_mod, BRRB, adg->rbgb); + rsnd_mod_write(adg_mod, BRRA, adg->brga); + rsnd_mod_write(adg_mod, BRRB, adg->brgb); dev_dbg(dev, "CLKOUT is based on BRG%c (= %dHz)\n", (ckr) ? 'B' : 'A', - (ckr) ? adg->rbg_rate[ADG_HZ_48] : - adg->rbg_rate[ADG_HZ_441]); + (ckr) ? adg->brg_rate[ADG_HZ_48] : + adg->brg_rate[ADG_HZ_441]); return 0; } @@ -484,7 +484,7 @@ static int rsnd_adg_get_clkout(struct rsnd_priv *priv) struct device *dev = rsnd_priv_to_dev(priv); struct device_node *np = dev->of_node; struct property *prop; - u32 ckr, rbgx, rbga, rbgb; + u32 ckr, brgx, brga, brgb; u32 rate, div; u32 req_rate[ADG_HZ_SIZE] = {}; uint32_t count = 0; @@ -501,8 +501,8 @@ static int rsnd_adg_get_clkout(struct rsnd_priv *priv) }; ckr = 0; - rbga = 2; /* default 1/6 */ - rbgb = 2; /* default 1/6 */ + brga = 2; /* default 1/6 */ + brgb = 2; /* default 1/6 */ /* * ADG supports BRRA/BRRB output only @@ -543,30 +543,30 @@ static int rsnd_adg_get_clkout(struct rsnd_priv *priv) if (0 == rate) /* not used */ continue; - /* RBGA */ - if (!adg->rbg_rate[ADG_HZ_441] && (0 == rate % 44100)) { + /* BRGA */ + if (!adg->brg_rate[ADG_HZ_441] && (0 == rate % 44100)) { div = 6; if (req_Hz[ADG_HZ_441]) div = rate / req_Hz[ADG_HZ_441]; - rbgx = rsnd_adg_calculate_rbgx(div); - if (BRRx_MASK(rbgx) == rbgx) { - rbga = rbgx; - adg->rbg_rate[ADG_HZ_441] = rate / div; + brgx = rsnd_adg_calculate_brgx(div); + if (BRRx_MASK(brgx) == brgx) { + brga = brgx; + adg->brg_rate[ADG_HZ_441] = rate / div; ckr |= brg_table[i] << 20; if (req_Hz[ADG_HZ_441]) parent_clk_name = __clk_get_name(clk); } } - /* RBGB */ - if (!adg->rbg_rate[ADG_HZ_48] && (0 == rate % 48000)) { + /* BRGB */ + if (!adg->brg_rate[ADG_HZ_48] && (0 == rate % 48000)) { div = 6; if (req_Hz[ADG_HZ_48]) div = rate / req_Hz[ADG_HZ_48]; - rbgx = rsnd_adg_calculate_rbgx(div); - if (BRRx_MASK(rbgx) == rbgx) { - rbgb = rbgx; - adg->rbg_rate[ADG_HZ_48] = rate / div; + brgx = rsnd_adg_calculate_brgx(div); + if (BRRx_MASK(brgx) == brgx) { + brgb = brgx; + adg->brg_rate[ADG_HZ_48] = rate / div; ckr |= brg_table[i] << 16; if (req_Hz[ADG_HZ_48]) parent_clk_name = __clk_get_name(clk); @@ -620,8 +620,8 @@ static int rsnd_adg_get_clkout(struct rsnd_priv *priv) rsnd_adg_get_clkout_end: adg->ckr = ckr; - adg->rbga = rbga; - adg->rbgb = rbgb; + adg->brga = brga; + adg->brgb = brgb; return 0; @@ -663,9 +663,9 @@ void rsnd_adg_clk_dbg_info(struct rsnd_priv *priv, struct seq_file *m) __clk_get_name(clk), clk, clk_get_rate(clk)); dbg_msg(dev, m, "BRGCKR = 0x%08x, BRRA/BRRB = 0x%x/0x%x\n", - adg->ckr, adg->rbga, adg->rbgb); - dbg_msg(dev, m, "BRGA (for 44100 base) = %d\n", adg->rbg_rate[ADG_HZ_441]); - dbg_msg(dev, m, "BRGB (for 48000 base) = %d\n", adg->rbg_rate[ADG_HZ_48]); + adg->ckr, adg->brga, adg->brgb); + dbg_msg(dev, m, "BRGA (for 44100 base) = %d\n", adg->brg_rate[ADG_HZ_441]); + dbg_msg(dev, m, "BRGB (for 48000 base) = %d\n", adg->brg_rate[ADG_HZ_48]); /* * Actual CLKOUT will be exchanged in rsnd_adg_ssi_clk_try_start() From 22ce6843abec19270bf69b176d7ee0a4ef781da5 Mon Sep 17 00:00:00 2001 From: Joseph Hunkeler Date: Thu, 16 Feb 2023 10:50:07 -0500 Subject: [PATCH 283/765] ASoC: amd: yp: Add OMEN by HP Gaming Laptop 16z-n000 to quirks Enables display microphone on the HP OMEN 16z-n000 (8A42) laptop Signed-off-by: Joseph Hunkeler Link: https://lore.kernel.org/r/20230216155007.26143-1-jhunkeler@gmail.com Signed-off-by: Mark Brown --- sound/soc/amd/yc/acp6x-mach.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/sound/soc/amd/yc/acp6x-mach.c b/sound/soc/amd/yc/acp6x-mach.c index 00fb976e0b81e..c061519adfbe4 100644 --- a/sound/soc/amd/yc/acp6x-mach.c +++ b/sound/soc/amd/yc/acp6x-mach.c @@ -248,6 +248,13 @@ static const struct dmi_system_id yc_acp_quirk_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "15NBC1011"), } }, + { + .driver_data = &acp6x_card, + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "HP"), + DMI_MATCH(DMI_PRODUCT_NAME, "OMEN by HP Gaming Laptop 16z-n000"), + } + }, {} }; From 9ec041ea40dbc6425c0ee6ae15786e5ab1d6aad6 Mon Sep 17 00:00:00 2001 From: Luca Boccassi Date: Fri, 10 Feb 2023 01:06:12 +0000 Subject: [PATCH 284/765] sed-opal: add support flag for SUM in status ioctl Not every OPAL drive supports SUM (Single User Mode), so report this information to userspace via the get-status ioctl so that we can adjust the formatting options accordingly. Tested on a kingston drive (which supports it) and a samsung one (which does not). Signed-off-by: Luca Boccassi Link: https://lore.kernel.org/r/20230210010612.28729-1-luca.boccassi@gmail.com Signed-off-by: Jens Axboe --- block/sed-opal.c | 2 ++ include/uapi/linux/sed-opal.h | 1 + 2 files changed, 3 insertions(+) diff --git a/block/sed-opal.c b/block/sed-opal.c index 463873f61e01e..c320093c14f12 100644 --- a/block/sed-opal.c +++ b/block/sed-opal.c @@ -487,6 +487,8 @@ static int opal_discovery0_end(struct opal_dev *dev) break; case FC_SINGLEUSER: single_user = check_sum(body->features); + if (single_user) + dev->flags |= OPAL_FL_SUM_SUPPORTED; break; case FC_GEOMETRY: check_geometry(dev, body); diff --git a/include/uapi/linux/sed-opal.h b/include/uapi/linux/sed-opal.h index 1fed3c9294fc5..d7a1524023dba 100644 --- a/include/uapi/linux/sed-opal.h +++ b/include/uapi/linux/sed-opal.h @@ -144,6 +144,7 @@ struct opal_read_write_table { #define OPAL_FL_LOCKED 0x00000008 #define OPAL_FL_MBR_ENABLED 0x00000010 #define OPAL_FL_MBR_DONE 0x00000020 +#define OPAL_FL_SUM_SUPPORTED 0x00000040 struct opal_status { __u32 flags; From 0f77b29ad14e34a89961f32edc87b92db623bb37 Mon Sep 17 00:00:00 2001 From: Yu Kuai Date: Fri, 17 Feb 2023 10:21:59 +0800 Subject: [PATCH 285/765] block: Revert "block: Do not reread partition table on exclusively open device" This reverts commit 36369f46e91785688a5f39d7a5590e3f07981316. This patch can't fix the problem in a corner case that device can be opened exclusively after the checking and before blkdev_get_by_dev(). We'll use a new solution to fix the problem in the next patch, and the new solution doesn't need to change apis. Signed-off-by: Yu Kuai Acked-by: Jan Kara Link: https://lore.kernel.org/r/20230217022200.3092987-2-yukuai1@huaweicloud.com Signed-off-by: Jens Axboe --- block/blk.h | 2 +- block/genhd.c | 7 ++----- block/ioctl.c | 13 ++++++------- 3 files changed, 9 insertions(+), 13 deletions(-) diff --git a/block/blk.h b/block/blk.h index 4c3b3325219a5..e835f21d48af0 100644 --- a/block/blk.h +++ b/block/blk.h @@ -427,7 +427,7 @@ int bio_add_hw_page(struct request_queue *q, struct bio *bio, struct request_queue *blk_alloc_queue(int node_id); -int disk_scan_partitions(struct gendisk *disk, fmode_t mode, void *owner); +int disk_scan_partitions(struct gendisk *disk, fmode_t mode); int disk_alloc_events(struct gendisk *disk); void disk_add_events(struct gendisk *disk); diff --git a/block/genhd.c b/block/genhd.c index 093ef292e98f7..820e27d88cbff 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -356,7 +356,7 @@ void disk_uevent(struct gendisk *disk, enum kobject_action action) } EXPORT_SYMBOL_GPL(disk_uevent); -int disk_scan_partitions(struct gendisk *disk, fmode_t mode, void *owner) +int disk_scan_partitions(struct gendisk *disk, fmode_t mode) { struct block_device *bdev; @@ -366,9 +366,6 @@ int disk_scan_partitions(struct gendisk *disk, fmode_t mode, void *owner) return -EINVAL; if (disk->open_partitions) return -EBUSY; - /* Someone else has bdev exclusively open? */ - if (disk->part0->bd_holder && disk->part0->bd_holder != owner) - return -EBUSY; set_bit(GD_NEED_PART_SCAN, &disk->state); bdev = blkdev_get_by_dev(disk_devt(disk), mode, NULL); @@ -499,7 +496,7 @@ int __must_check device_add_disk(struct device *parent, struct gendisk *disk, bdev_add(disk->part0, ddev->devt); if (get_capacity(disk)) - disk_scan_partitions(disk, FMODE_READ, NULL); + disk_scan_partitions(disk, FMODE_READ); /* * Announce the disk and partitions after all partitions are diff --git a/block/ioctl.c b/block/ioctl.c index 96617512982e5..6dd49d877584a 100644 --- a/block/ioctl.c +++ b/block/ioctl.c @@ -467,10 +467,10 @@ static int blkdev_bszset(struct block_device *bdev, fmode_t mode, * user space. Note the separate arg/argp parameters that are needed * to deal with the compat_ptr() conversion. */ -static int blkdev_common_ioctl(struct file *file, fmode_t mode, unsigned cmd, - unsigned long arg, void __user *argp) +static int blkdev_common_ioctl(struct block_device *bdev, fmode_t mode, + unsigned int cmd, unsigned long arg, + void __user *argp) { - struct block_device *bdev = I_BDEV(file->f_mapping->host); unsigned int max_sectors; switch (cmd) { @@ -528,8 +528,7 @@ static int blkdev_common_ioctl(struct file *file, fmode_t mode, unsigned cmd, return -EACCES; if (bdev_is_partition(bdev)) return -EINVAL; - return disk_scan_partitions(bdev->bd_disk, mode & ~FMODE_EXCL, - file); + return disk_scan_partitions(bdev->bd_disk, mode & ~FMODE_EXCL); case BLKTRACESTART: case BLKTRACESTOP: case BLKTRACETEARDOWN: @@ -607,7 +606,7 @@ long blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg) break; } - ret = blkdev_common_ioctl(file, mode, cmd, arg, argp); + ret = blkdev_common_ioctl(bdev, mode, cmd, arg, argp); if (ret != -ENOIOCTLCMD) return ret; @@ -676,7 +675,7 @@ long compat_blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg) break; } - ret = blkdev_common_ioctl(file, mode, cmd, arg, argp); + ret = blkdev_common_ioctl(bdev, mode, cmd, arg, argp); if (ret == -ENOIOCTLCMD && disk->fops->compat_ioctl) ret = disk->fops->compat_ioctl(bdev, mode, cmd, arg); From e5cfefa97bccf956ea0bb6464c1f6c84fd7a8d9f Mon Sep 17 00:00:00 2001 From: Yu Kuai Date: Fri, 17 Feb 2023 10:22:00 +0800 Subject: [PATCH 286/765] block: fix scan partition for exclusively open device again As explained in commit 36369f46e917 ("block: Do not reread partition table on exclusively open device"), reread partition on the device that is exclusively opened by someone else is problematic. This patch will make sure partition scan will only be proceed if current thread open the device exclusively, or the device is not opened exclusively, and in the later case, other scanners and exclusive openers will be blocked temporarily until partition scan is done. Fixes: 10c70d95c0f2 ("block: remove the bd_openers checks in blk_drop_partitions") Cc: Suggested-by: Jan Kara Signed-off-by: Yu Kuai Reviewed-by: Christoph Hellwig Link: https://lore.kernel.org/r/20230217022200.3092987-3-yukuai1@huaweicloud.com Signed-off-by: Jens Axboe --- block/genhd.c | 30 ++++++++++++++++++++++++++---- block/ioctl.c | 2 +- 2 files changed, 27 insertions(+), 5 deletions(-) diff --git a/block/genhd.c b/block/genhd.c index 820e27d88cbff..c0a73cd76bb1f 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -359,6 +359,7 @@ EXPORT_SYMBOL_GPL(disk_uevent); int disk_scan_partitions(struct gendisk *disk, fmode_t mode) { struct block_device *bdev; + int ret = 0; if (disk->flags & (GENHD_FL_NO_PART | GENHD_FL_HIDDEN)) return -EINVAL; @@ -368,11 +369,27 @@ int disk_scan_partitions(struct gendisk *disk, fmode_t mode) return -EBUSY; set_bit(GD_NEED_PART_SCAN, &disk->state); - bdev = blkdev_get_by_dev(disk_devt(disk), mode, NULL); + /* + * If the device is opened exclusively by current thread already, it's + * safe to scan partitons, otherwise, use bd_prepare_to_claim() to + * synchronize with other exclusive openers and other partition + * scanners. + */ + if (!(mode & FMODE_EXCL)) { + ret = bd_prepare_to_claim(disk->part0, disk_scan_partitions); + if (ret) + return ret; + } + + bdev = blkdev_get_by_dev(disk_devt(disk), mode & ~FMODE_EXCL, NULL); if (IS_ERR(bdev)) - return PTR_ERR(bdev); - blkdev_put(bdev, mode); - return 0; + ret = PTR_ERR(bdev); + else + blkdev_put(bdev, mode); + + if (!(mode & FMODE_EXCL)) + bd_abort_claiming(disk->part0, disk_scan_partitions); + return ret; } /** @@ -494,6 +511,11 @@ int __must_check device_add_disk(struct device *parent, struct gendisk *disk, if (ret) goto out_unregister_bdi; + /* Make sure the first partition scan will be proceed */ + if (get_capacity(disk) && !(disk->flags & GENHD_FL_NO_PART) && + !test_bit(GD_SUPPRESS_PART_SCAN, &disk->state)) + set_bit(GD_NEED_PART_SCAN, &disk->state); + bdev_add(disk->part0, ddev->devt); if (get_capacity(disk)) disk_scan_partitions(disk, FMODE_READ); diff --git a/block/ioctl.c b/block/ioctl.c index 6dd49d877584a..9c5f637ff153f 100644 --- a/block/ioctl.c +++ b/block/ioctl.c @@ -528,7 +528,7 @@ static int blkdev_common_ioctl(struct block_device *bdev, fmode_t mode, return -EACCES; if (bdev_is_partition(bdev)) return -EINVAL; - return disk_scan_partitions(bdev->bd_disk, mode & ~FMODE_EXCL); + return disk_scan_partitions(bdev->bd_disk, mode); case BLKTRACESTART: case BLKTRACESTOP: case BLKTRACETEARDOWN: From 486dd4e846814016443abfcbfee0b8b3f3b35330 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Wed, 18 Jan 2023 16:48:16 +0100 Subject: [PATCH 287/765] pwm: ab8500: Fix calculation of duty and period MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit After a check of the manual it becomes obvious that the calculations in .apply() are totally bogus: FreqPWMOutx was always written as zero, so the period was fixed at 3413333.33 ns. However state->period wasn't checked at all. The lower 10 bits of duty_cycle were just used as DutyPWMOutx. So if a duty cycle of 512 ns (or 1536 ns) was requested, it actually programmed 1710000 ns. Other values were wrong by the same factor. Signed-off-by: Uwe Kleine-König Signed-off-by: Thierry Reding --- drivers/pwm/pwm-ab8500.c | 69 ++++++++++++++++++++++++++++++++++------ 1 file changed, 60 insertions(+), 9 deletions(-) diff --git a/drivers/pwm/pwm-ab8500.c b/drivers/pwm/pwm-ab8500.c index ad37bc46f2721..c5239eef3b8fb 100644 --- a/drivers/pwm/pwm-ab8500.c +++ b/drivers/pwm/pwm-ab8500.c @@ -3,6 +3,7 @@ * Copyright (C) ST-Ericsson SA 2010 * * Author: Arun R Murthy + * Datasheet: https://web.archive.org/web/20130614115108/http://www.stericsson.com/developers/CD00291561_UM1031_AB8500_user_manual-rev5_CTDS_public.pdf */ #include #include @@ -20,6 +21,8 @@ #define AB8500_PWM_OUT_CTRL2_REG 0x61 #define AB8500_PWM_OUT_CTRL7_REG 0x66 +#define AB8500_PWM_CLKRATE 9600000 + struct ab8500_pwm_chip { struct pwm_chip chip; unsigned int hwid; @@ -35,13 +38,60 @@ static int ab8500_pwm_apply(struct pwm_chip *chip, struct pwm_device *pwm, { int ret; u8 reg; - unsigned int higher_val, lower_val; + u8 higher_val, lower_val; + unsigned int duty_steps, div; struct ab8500_pwm_chip *ab8500 = ab8500_pwm_from_chip(chip); if (state->polarity != PWM_POLARITY_NORMAL) return -EINVAL; - if (!state->enabled) { + if (state->enabled) { + /* + * A time quantum is + * q = (32 - FreqPWMOutx[3:0]) / AB8500_PWM_CLKRATE + * The period is always 1024 q, duty_cycle is between 1q and 1024q. + * + * FreqPWMOutx[3:0] | output frequency | output frequency | 1024q = period + * | (from manual) | (1 / 1024q) | = 1 / freq + * -----------------+------------------+------------------+-------------- + * b0000 | 293 Hz | 292.968750 Hz | 3413333.33 ns + * b0001 | 302 Hz | 302.419355 Hz | 3306666.66 ns + * b0010 | 312 Hz | 312.500000 Hz | 3200000 ns + * b0011 | 323 Hz | 323.275862 Hz | 3093333.33 ns + * b0100 | 334 Hz | 334.821429 Hz | 2986666.66 ns + * b0101 | 347 Hz | 347.222222 Hz | 2880000 ns + * b0110 | 360 Hz | 360.576923 Hz | 2773333.33 ns + * b0111 | 375 Hz | 375.000000 Hz | 2666666.66 ns + * b1000 | 390 Hz | 390.625000 Hz | 2560000 ns + * b1001 | 407 Hz | 407.608696 Hz | 2453333.33 ns + * b1010 | 426 Hz | 426.136364 Hz | 2346666.66 ns + * b1011 | 446 Hz | 446.428571 Hz | 2240000 ns + * b1100 | 468 Hz | 468.750000 Hz | 2133333.33 ns + * b1101 | 493 Hz | 493.421053 Hz | 2026666.66 ns + * b1110 | 520 Hz | 520.833333 Hz | 1920000 ns + * b1111 | 551 Hz | 551.470588 Hz | 1813333.33 ns + * + * + * AB8500_PWM_CLKRATE is a multiple of 1024, so the division by + * 1024 can be done in this factor without loss of precision. + */ + div = min_t(u64, mul_u64_u64_div_u64(state->period, + AB8500_PWM_CLKRATE >> 10, + NSEC_PER_SEC), 32); /* 32 - FreqPWMOutx[3:0] */ + if (div <= 16) + /* requested period < 3413333.33 */ + return -EINVAL; + + duty_steps = max_t(u64, mul_u64_u64_div_u64(state->duty_cycle, + AB8500_PWM_CLKRATE, + (u64)NSEC_PER_SEC * div), 1024); + } + + /* + * The hardware doesn't support duty_steps = 0 explicitly, but emits low + * when disabled. + */ + if (!state->enabled || duty_steps == 0) { ret = abx500_mask_and_set_register_interruptible(chip->dev, AB8500_MISC, AB8500_PWM_OUT_CTRL7_REG, 1 << ab8500->hwid, 0); @@ -53,28 +103,29 @@ static int ab8500_pwm_apply(struct pwm_chip *chip, struct pwm_device *pwm, } /* - * get the first 8 bits that are be written to + * The lower 8 bits of duty_steps is written to ... * AB8500_PWM_OUT_CTRL1_REG[0:7] */ - lower_val = state->duty_cycle & 0x00FF; + lower_val = (duty_steps - 1) & 0x00ff; /* - * get bits [9:10] that are to be written to - * AB8500_PWM_OUT_CTRL2_REG[0:1] + * The two remaining high bits to + * AB8500_PWM_OUT_CTRL2_REG[0:1]; together with FreqPWMOutx. */ - higher_val = ((state->duty_cycle & 0x0300) >> 8); + higher_val = ((duty_steps - 1) & 0x0300) >> 8 | (32 - div) << 4; reg = AB8500_PWM_OUT_CTRL1_REG + (ab8500->hwid * 2); ret = abx500_set_register_interruptible(chip->dev, AB8500_MISC, - reg, (u8)lower_val); + reg, lower_val); if (ret < 0) return ret; ret = abx500_set_register_interruptible(chip->dev, AB8500_MISC, - (reg + 1), (u8)higher_val); + (reg + 1), higher_val); if (ret < 0) return ret; + /* enable */ ret = abx500_mask_and_set_register_interruptible(chip->dev, AB8500_MISC, AB8500_PWM_OUT_CTRL7_REG, 1 << ab8500->hwid, 1 << ab8500->hwid); From 327437884e9a752ccbf759cbab641439ca708f5b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Wed, 18 Jan 2023 16:48:17 +0100 Subject: [PATCH 288/765] pwm: ab8500: Implement .get_state() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The registers are readable, so it's possible to implement the .get_state() callback for this PWM. Signed-off-by: Uwe Kleine-König Signed-off-by: Thierry Reding --- drivers/pwm/pwm-ab8500.c | 43 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 43 insertions(+) diff --git a/drivers/pwm/pwm-ab8500.c b/drivers/pwm/pwm-ab8500.c index c5239eef3b8fb..507ff0d5f7bd8 100644 --- a/drivers/pwm/pwm-ab8500.c +++ b/drivers/pwm/pwm-ab8500.c @@ -136,8 +136,51 @@ static int ab8500_pwm_apply(struct pwm_chip *chip, struct pwm_device *pwm, return ret; } +static int ab8500_pwm_get_state(struct pwm_chip *chip, struct pwm_device *pwm, + struct pwm_state *state) +{ + u8 ctrl7, lower_val, higher_val; + int ret; + struct ab8500_pwm_chip *ab8500 = ab8500_pwm_from_chip(chip); + unsigned int div, duty_steps; + + ret = abx500_get_register_interruptible(chip->dev, AB8500_MISC, + AB8500_PWM_OUT_CTRL7_REG, + &ctrl7); + if (ret) + return ret; + + state->polarity = PWM_POLARITY_NORMAL; + + if (!(ctrl7 & 1 << ab8500->hwid)) { + state->enabled = false; + return 0; + } + + ret = abx500_get_register_interruptible(chip->dev, AB8500_MISC, + AB8500_PWM_OUT_CTRL1_REG + (ab8500->hwid * 2), + &lower_val); + if (ret) + return ret; + + ret = abx500_get_register_interruptible(chip->dev, AB8500_MISC, + AB8500_PWM_OUT_CTRL2_REG + (ab8500->hwid * 2), + &higher_val); + if (ret) + return ret; + + div = 32 - ((higher_val & 0xf0) >> 4); + duty_steps = ((higher_val & 3) << 8 | lower_val) + 1; + + state->period = DIV64_U64_ROUND_UP((u64)div << 10, AB8500_PWM_CLKRATE); + state->duty_cycle = DIV64_U64_ROUND_UP((u64)div * duty_steps, AB8500_PWM_CLKRATE); + + return 0; +} + static const struct pwm_ops ab8500_pwm_ops = { .apply = ab8500_pwm_apply, + .get_state = ab8500_pwm_get_state, .owner = THIS_MODULE, }; From 860793bbdcdfbeae684d552ce0121846cffc4803 Mon Sep 17 00:00:00 2001 From: Jeff LaBundy Date: Sun, 5 Feb 2023 20:56:22 -0600 Subject: [PATCH 289/765] pwm: iqs620a: Replace one remaining instance of regmap_update_bits() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The call to regmap_update_bits() which was responsible for clearing the PWM output enable register bit was recently dropped in favor of a call to regmap_clear_bits(), thereby simplifying the code. Similarly, the call to regmap_update_bits() which sets the same bit can be simplified with a call to regmap_set_bits(). Signed-off-by: Jeff LaBundy Reviewed-by: Uwe Kleine-König Signed-off-by: Thierry Reding --- drivers/pwm/pwm-iqs620a.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/pwm/pwm-iqs620a.c b/drivers/pwm/pwm-iqs620a.c index 4987ca940b648..8362b4870c66c 100644 --- a/drivers/pwm/pwm-iqs620a.c +++ b/drivers/pwm/pwm-iqs620a.c @@ -55,8 +55,8 @@ static int iqs620_pwm_init(struct iqs620_pwm_private *iqs620_pwm, if (ret) return ret; - return regmap_update_bits(iqs62x->regmap, IQS620_PWR_SETTINGS, - IQS620_PWR_SETTINGS_PWM_OUT, 0xff); + return regmap_set_bits(iqs62x->regmap, IQS620_PWR_SETTINGS, + IQS620_PWR_SETTINGS_PWM_OUT); } static int iqs620_pwm_apply(struct pwm_chip *chip, struct pwm_device *pwm, From 0f03bf300833c05d914ab7f5ab3d8bc8564e9912 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Fri, 23 Dec 2022 15:38:11 +0000 Subject: [PATCH 290/765] dt-bindings: pwm: Document Synopsys DesignWare snps,pwm-dw-apb-timers-pwm2 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add documentation for the bindings for Synopsys' DesignWare PWM block as we will be adding DT/platform support to the Linux driver soon. Signed-off-by: Ben Dooks Reviewed-by: Krzysztof Kozlowski Acked-by: Uwe Kleine-König Signed-off-by: Thierry Reding --- .../bindings/pwm/snps,dw-apb-timers-pwm2.yaml | 68 +++++++++++++++++++ 1 file changed, 68 insertions(+) create mode 100644 Documentation/devicetree/bindings/pwm/snps,dw-apb-timers-pwm2.yaml diff --git a/Documentation/devicetree/bindings/pwm/snps,dw-apb-timers-pwm2.yaml b/Documentation/devicetree/bindings/pwm/snps,dw-apb-timers-pwm2.yaml new file mode 100644 index 0000000000000..9aabdb373afa2 --- /dev/null +++ b/Documentation/devicetree/bindings/pwm/snps,dw-apb-timers-pwm2.yaml @@ -0,0 +1,68 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +# Copyright (C) 2022 SiFive, Inc. +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/pwm/snps,dw-apb-timers-pwm2.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Synopsys DW-APB timers PWM controller + +maintainers: + - Ben Dooks + +description: + This describes the DesignWare APB timers module when used in the PWM + mode. The IP core can be generated with various options which can + control the functionality, the number of PWMs available and other + internal controls the designer requires. + + The IP block has a version register so this can be used for detection + instead of having to encode the IP version number in the device tree + comaptible. + +allOf: + - $ref: pwm.yaml# + +properties: + compatible: + const: snps,dw-apb-timers-pwm2 + + reg: + maxItems: 1 + + "#pwm-cells": + const: 3 + + clocks: + items: + - description: Interface bus clock + - description: PWM reference clock + + clock-names: + items: + - const: bus + - const: timer + + snps,pwm-number: + $ref: /schemas/types.yaml#/definitions/uint32 + description: The number of PWM channels configured for this instance + enum: [1, 2, 3, 4, 5, 6, 7, 8] + +required: + - compatible + - reg + - "#pwm-cells" + - clocks + - clock-names + +additionalProperties: false + +examples: + - | + pwm: pwm@180000 { + compatible = "snps,dw-apb-timers-pwm2"; + reg = <0x180000 0x200>; + #pwm-cells = <3>; + clocks = <&bus>, <&timer>; + clock-names = "bus", "timer"; + }; From d88cbbb39b4db057feb1552de31f22c02a21b36f Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Fri, 17 Feb 2023 10:29:10 +0100 Subject: [PATCH 291/765] blk-mq: Reorder fields in 'struct blk_mq_tag_set' Group some variables based on their sizes to reduce hole and avoid padding. On x86_64, this shrinks the size of 'struct blk_mq_tag_set' from 304 to 296 bytes. Signed-off-by: Christophe JAILLET Link: https://lore.kernel.org/r/6f249f9b02a3490283ef0278096556de41aa0cf0.1676626130.git.christophe.jaillet@wanadoo.fr Signed-off-by: Jens Axboe --- include/linux/blk-mq.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index 779fba613bd09..dd5ce1137f04a 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -473,6 +473,7 @@ enum hctx_type { /** * struct blk_mq_tag_set - tag set that can be shared between request queues + * @ops: Pointers to functions that implement block driver behavior. * @map: One or more ctx -> hctx mappings. One map exists for each * hardware queue type (enum hctx_type) that the driver wishes * to support. There are no restrictions on maps being of the @@ -480,7 +481,6 @@ enum hctx_type { * types. * @nr_maps: Number of elements in the @map array. A number in the range * [1, HCTX_MAX_TYPES]. - * @ops: Pointers to functions that implement block driver behavior. * @nr_hw_queues: Number of hardware queues supported by the block driver that * owns this data structure. * @queue_depth: Number of tags per hardware queue, reserved tags included. @@ -505,9 +505,9 @@ enum hctx_type { * (BLK_MQ_F_BLOCKING). */ struct blk_mq_tag_set { + const struct blk_mq_ops *ops; struct blk_mq_queue_map map[HCTX_MAX_TYPES]; unsigned int nr_maps; - const struct blk_mq_ops *ops; unsigned int nr_hw_queues; unsigned int queue_depth; unsigned int reserved_tags; From f8ee39b44b75ac955bb54f46d5370481727c47e7 Mon Sep 17 00:00:00 2001 From: Primoz Fiser Date: Fri, 16 Dec 2022 09:36:45 +0100 Subject: [PATCH 292/765] watchdog: da9062: da9063: use unlocked xfer function in restart Machine resets via da9062/da9063 PMICs are challenging since one needs to use special i2c atomic transfers due to the fact interrupts are disabled in such late system stages. This is the reason both PMICs don't use regmap and have instead opted for i2c_smbus_write_byte_data() in restart handlers. However extensive testing revealed that even using atomic safe function is not enough and occasional resets fail with error message "Failed to shutdown (err = -11)". This is due to the fact that function i2c_smbus_write_byte_data() in turn calls __i2c_lock_bus_helper() which might fail with -EAGAIN when bus lock is already taken and cannot be released anymore. Thus replace i2c_smbus_write_byte_data() with unlocked flavor of i2c_smbus_xfer() function to avoid above dead-lock scenario. At this system stage we don't care about proper locking anymore and only want proper machine reset to be carried out. Signed-off-by: Primoz Fiser Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221216083645.2574077-1-primoz.fiser@norik.com [groeck: Fixed continuation line alignment] Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/da9062_wdt.c | 15 ++++++++++++--- drivers/watchdog/da9063_wdt.c | 15 ++++++++++++--- 2 files changed, 24 insertions(+), 6 deletions(-) diff --git a/drivers/watchdog/da9062_wdt.c b/drivers/watchdog/da9062_wdt.c index f02cbd530538b..426962547df16 100644 --- a/drivers/watchdog/da9062_wdt.c +++ b/drivers/watchdog/da9062_wdt.c @@ -155,11 +155,20 @@ static int da9062_wdt_restart(struct watchdog_device *wdd, unsigned long action, { struct da9062_watchdog *wdt = watchdog_get_drvdata(wdd); struct i2c_client *client = to_i2c_client(wdt->hw->dev); + union i2c_smbus_data msg; int ret; - /* Don't use regmap because it is not atomic safe */ - ret = i2c_smbus_write_byte_data(client, DA9062AA_CONTROL_F, - DA9062AA_SHUTDOWN_MASK); + /* + * Don't use regmap because it is not atomic safe. Additionally, use + * unlocked flavor of i2c_smbus_xfer to avoid scenario where i2c bus + * might be previously locked by some process unable to release the + * lock due to interrupts already being disabled at this late stage. + */ + msg.byte = DA9062AA_SHUTDOWN_MASK; + ret = __i2c_smbus_xfer(client->adapter, client->addr, client->flags, + I2C_SMBUS_WRITE, DA9062AA_CONTROL_F, + I2C_SMBUS_BYTE_DATA, &msg); + if (ret < 0) dev_alert(wdt->hw->dev, "Failed to shutdown (err = %d)\n", ret); diff --git a/drivers/watchdog/da9063_wdt.c b/drivers/watchdog/da9063_wdt.c index 09a4af4c58fc8..684667469b10c 100644 --- a/drivers/watchdog/da9063_wdt.c +++ b/drivers/watchdog/da9063_wdt.c @@ -174,11 +174,20 @@ static int da9063_wdt_restart(struct watchdog_device *wdd, unsigned long action, { struct da9063 *da9063 = watchdog_get_drvdata(wdd); struct i2c_client *client = to_i2c_client(da9063->dev); + union i2c_smbus_data msg; int ret; - /* Don't use regmap because it is not atomic safe */ - ret = i2c_smbus_write_byte_data(client, DA9063_REG_CONTROL_F, - DA9063_SHUTDOWN); + /* + * Don't use regmap because it is not atomic safe. Additionally, use + * unlocked flavor of i2c_smbus_xfer to avoid scenario where i2c bus + * might previously be locked by some process unable to release the + * lock due to interrupts already being disabled at this late stage. + */ + msg.byte = DA9063_SHUTDOWN; + ret = __i2c_smbus_xfer(client->adapter, client->addr, client->flags, + I2C_SMBUS_WRITE, DA9063_REG_CONTROL_F, + I2C_SMBUS_BYTE_DATA, &msg); + if (ret < 0) dev_alert(da9063->dev, "Failed to shutdown (err = %d)\n", ret); From e8b1cb537b8095e1bc528ce2aef56318195d8343 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 13 Jan 2023 11:33:46 +0100 Subject: [PATCH 293/765] dt-bindings: watchdog: allow "timer" as node name On some SoCs the watchdog device is actually mixed with timer, e.g. the qcom,msm-timer on older Qualcomm SoCs where this is actually one hardware block responsible for both system timer and watchdog. Allow calling such device nodes as "timer". Signed-off-by: Krzysztof Kozlowski Reviewed-by: Rob Herring Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221212174933.208900-1-krzysztof.kozlowski@linaro.org Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- Documentation/devicetree/bindings/watchdog/watchdog.yaml | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/watchdog/watchdog.yaml b/Documentation/devicetree/bindings/watchdog/watchdog.yaml index fccae0d001103..519b48889eb14 100644 --- a/Documentation/devicetree/bindings/watchdog/watchdog.yaml +++ b/Documentation/devicetree/bindings/watchdog/watchdog.yaml @@ -14,9 +14,14 @@ description: | This document describes generic bindings which can be used to describe watchdog devices in a device tree. +select: + properties: + $nodename: + pattern: "^watchdog(@.*|-[0-9a-f])?$" + properties: $nodename: - pattern: "^watchdog(@.*|-[0-9a-f])?$" + pattern: "^(timer|watchdog)(@.*|-[0-9a-f])?$" timeout-sec: description: From 4ea6b986dfaff19e014ea634bdca669eebe73def Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Fri, 25 Nov 2022 23:12:40 +0100 Subject: [PATCH 294/765] watchdog: iTCO_wdt: Report firmware_version MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Synchronize the reported information in dmesg and the watchdog APIs. Signed-off-by: Thomas Weißschuh Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221125221240.2818-1-linux@weissschuh.net Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/iTCO_wdt.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/watchdog/iTCO_wdt.c b/drivers/watchdog/iTCO_wdt.c index e937b4dd28be7..264857d314da8 100644 --- a/drivers/watchdog/iTCO_wdt.c +++ b/drivers/watchdog/iTCO_wdt.c @@ -441,11 +441,10 @@ static bool iTCO_wdt_set_running(struct iTCO_wdt_private *p) * Kernel Interfaces */ -static const struct watchdog_info ident = { +static struct watchdog_info ident = { .options = WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING | WDIOF_MAGICCLOSE, - .firmware_version = 0, .identity = DRV_NAME, }; @@ -563,6 +562,7 @@ static int iTCO_wdt_probe(struct platform_device *pdev) break; } + ident.firmware_version = p->iTCO_version; p->wddev.info = &ident, p->wddev.ops = &iTCO_wdt_ops, p->wddev.bootstatus = 0; From 0d9e42e3c8ccde51a9ff7c49684ed7a3805ebd52 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Fri, 18 Nov 2022 23:45:30 +0100 Subject: [PATCH 295/765] watchdog: ziirave_wdt: Convert to i2c's .probe_new() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The probe function doesn't make use of the i2c_device_id * parameter so it can be trivially converted. Signed-off-by: Uwe Kleine-König Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221118224540.619276-597-uwe@kleine-koenig.org Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/ziirave_wdt.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/watchdog/ziirave_wdt.c b/drivers/watchdog/ziirave_wdt.c index d0e88875443ae..21ca08a694ee3 100644 --- a/drivers/watchdog/ziirave_wdt.c +++ b/drivers/watchdog/ziirave_wdt.c @@ -593,8 +593,7 @@ static int ziirave_wdt_init_duration(struct i2c_client *client) reset_duration); } -static int ziirave_wdt_probe(struct i2c_client *client, - const struct i2c_device_id *id) +static int ziirave_wdt_probe(struct i2c_client *client) { int ret; struct ziirave_wdt_data *w_priv; @@ -732,7 +731,7 @@ static struct i2c_driver ziirave_wdt_driver = { .name = "ziirave_wdt", .of_match_table = zrv_wdt_of_match, }, - .probe = ziirave_wdt_probe, + .probe_new = ziirave_wdt_probe, .remove = ziirave_wdt_remove, .id_table = ziirave_wdt_id, }; From 7bfd2747d3ec2c400857084527448c172a720d2a Mon Sep 17 00:00:00 2001 From: Cosmin Tanislav Date: Fri, 18 Nov 2022 17:08:09 +0200 Subject: [PATCH 296/765] watchdog: dw_wdt: stop on reboot HW running watchdogs are just watchdogs that are enabled before the Linux driver is probed, usually by the bootloader (eg. U-Boot). When the system is shutting down, the mechanism for keeping a HW running watchdog pinged is also stopped, but the watchdog itself is not stopped, causing a reset, and preventing the system from being shut down. Opt into stopping watchdogs on reboot. Signed-off-by: Cosmin Tanislav Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221118150809.102505-1-cosmin.tanislav@analog.com Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/dw_wdt.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/watchdog/dw_wdt.c b/drivers/watchdog/dw_wdt.c index 52962e8d11a6f..462f15bd5ffa6 100644 --- a/drivers/watchdog/dw_wdt.c +++ b/drivers/watchdog/dw_wdt.c @@ -663,6 +663,7 @@ static int dw_wdt_drv_probe(struct platform_device *pdev) platform_set_drvdata(pdev, dw_wdt); watchdog_set_restart_priority(wdd, 128); + watchdog_stop_on_reboot(wdd); ret = watchdog_register_device(wdd); if (ret) From 71708daa6f1446248de107e04cf28ffd322c800f Mon Sep 17 00:00:00 2001 From: Robert Marko Date: Fri, 25 Nov 2022 12:29:04 +0100 Subject: [PATCH 297/765] dt-bindings: watchdog: Convert GPIO binding to json-schema Convert the DT binding for GPIO WDT to JSON schema. Cc: luka.perkov@sartura.hr Signed-off-by: Robert Marko Reviewed-by: Rob Herring Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221125112904.48652-1-robert.marko@sartura.hr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- .../bindings/watchdog/gpio-wdt.yaml | 55 +++++++++++++++++++ 1 file changed, 55 insertions(+) create mode 100644 Documentation/devicetree/bindings/watchdog/gpio-wdt.yaml diff --git a/Documentation/devicetree/bindings/watchdog/gpio-wdt.yaml b/Documentation/devicetree/bindings/watchdog/gpio-wdt.yaml new file mode 100644 index 0000000000000..155dc7965e9b5 --- /dev/null +++ b/Documentation/devicetree/bindings/watchdog/gpio-wdt.yaml @@ -0,0 +1,55 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/watchdog/gpio-wdt.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: GPIO controlled watchdog + +maintainers: + - Robert Marko + +properties: + compatible: + const: linux,wdt-gpio + + gpios: + maxItems: 1 + description: GPIO connected to the WDT reset pin + + hw_algo: + $ref: /schemas/types.yaml#/definitions/string + description: Algorithm used by the driver + oneOf: + - description: + Either a high-to-low or a low-to-high transition clears the WDT counter. + The watchdog timer is disabled when GPIO is left floating or connected + to a three-state buffer. + const: toggle + - description: + Low or high level starts counting WDT timeout, the opposite level + disables the WDT. + Active level is determined by the GPIO flags. + const: level + + hw_margin_ms: + $ref: /schemas/types.yaml#/definitions/uint32 + description: Maximum time to reset watchdog circuit (in milliseconds) + minimum: 2 + maximum: 65535 + + always-running: + type: boolean + description: + If the watchdog timer cannot be disabled, add this flag to have the driver + keep toggling the signal without a client. + It will only cease to toggle the signal when the device is open and the + timeout elapsed. + +required: + - compatible + - gpios + - hw_algo + - hw_margin_ms + +unevaluatedProperties: false From 69197b43d9e3b164e847beede62d576ee01ee081 Mon Sep 17 00:00:00 2001 From: Neil Armstrong Date: Mon, 23 Jan 2023 11:10:00 +0100 Subject: [PATCH 298/765] dt-bindings: watchdog: convert meson-wdt.txt to dt-schema Convert the Amlogic Meson6 SoCs Watchdog timer bindings to dt-schema. Take in account the used interrupts property. The "amlogic,meson8-wdt" representation has been simplified since there's no users of this compatible used along the "amlogic,meson6-wdt". Reviewed-by: Krzysztof Kozlowski Reviewed-by: Martin Blumenstingl Signed-off-by: Neil Armstrong Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221117-b4-amlogic-bindings-convert-v3-3-e28dd31e3bed@linaro.org Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- .../bindings/watchdog/amlogic,meson6-wdt.yaml | 50 +++++++++++++++++++ .../bindings/watchdog/meson-wdt.txt | 21 -------- 2 files changed, 50 insertions(+), 21 deletions(-) create mode 100644 Documentation/devicetree/bindings/watchdog/amlogic,meson6-wdt.yaml delete mode 100644 Documentation/devicetree/bindings/watchdog/meson-wdt.txt diff --git a/Documentation/devicetree/bindings/watchdog/amlogic,meson6-wdt.yaml b/Documentation/devicetree/bindings/watchdog/amlogic,meson6-wdt.yaml new file mode 100644 index 0000000000000..84732cb58ec44 --- /dev/null +++ b/Documentation/devicetree/bindings/watchdog/amlogic,meson6-wdt.yaml @@ -0,0 +1,50 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/watchdog/amlogic,meson6-wdt.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Amlogic Meson6 SoCs Watchdog timer + +maintainers: + - Neil Armstrong + - Martin Blumenstingl + +allOf: + - $ref: watchdog.yaml# + +properties: + compatible: + oneOf: + - enum: + - amlogic,meson6-wdt + - amlogic,meson8-wdt + - amlogic,meson8b-wdt + - items: + - const: amlogic,meson8m2-wdt + - const: amlogic,meson8b-wdt + + interrupts: + maxItems: 1 + + reg: + maxItems: 1 + +required: + - compatible + - interrupts + - reg + +unevaluatedProperties: false + +examples: + - | + #include + #include + + wdt: watchdog@c1109900 { + compatible = "amlogic,meson6-wdt"; + reg = <0xc1109900 0x8>; + interrupts = ; + timeout-sec = <10>; + }; diff --git a/Documentation/devicetree/bindings/watchdog/meson-wdt.txt b/Documentation/devicetree/bindings/watchdog/meson-wdt.txt deleted file mode 100644 index 7588cc3971bf6..0000000000000 --- a/Documentation/devicetree/bindings/watchdog/meson-wdt.txt +++ /dev/null @@ -1,21 +0,0 @@ -Meson SoCs Watchdog timer - -Required properties: - -- compatible : depending on the SoC this should be one of: - "amlogic,meson6-wdt" on Meson6 SoCs - "amlogic,meson8-wdt" and "amlogic,meson6-wdt" on Meson8 SoCs - "amlogic,meson8b-wdt" on Meson8b SoCs - "amlogic,meson8m2-wdt" and "amlogic,meson8b-wdt" on Meson8m2 SoCs -- reg : Specifies base physical address and size of the registers. - -Optional properties: -- timeout-sec: contains the watchdog timeout in seconds. - -Example: - -wdt: watchdog@c1109900 { - compatible = "amlogic,meson6-wdt"; - reg = <0xc1109900 0x8>; - timeout-sec = <10>; -}; From d237c8d06d5a08a2dacb07c0ab07af696a58516c Mon Sep 17 00:00:00 2001 From: Lad Prabhakar Date: Fri, 18 Nov 2022 13:38:29 +0000 Subject: [PATCH 299/765] dt-bindings: watchdog: renesas,wdt: Document RZ/Five SoC The WDT block on the RZ/Five SoC is identical to one found on the RZ/G2UL SoC. "renesas,r9a07g043-wdt" compatible string will be used on the RZ/Five SoC so to make this clear, update the comment to include RZ/Five SoC. No driver changes are required as generic compatible string "renesas,rzg2l-wdt" will be used as a fallback on RZ/Five SoC. Signed-off-by: Lad Prabhakar Acked-by: Krzysztof Kozlowski Reviewed-by: Guenter Roeck Reviewed-by: Geert Uytterhoeven Link: https://lore.kernel.org/r/20221118133829.12855-1-prabhakar.mahadev-lad.rj@bp.renesas.com Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- Documentation/devicetree/bindings/watchdog/renesas,wdt.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/watchdog/renesas,wdt.yaml b/Documentation/devicetree/bindings/watchdog/renesas,wdt.yaml index 26b1815a6753a..e2c9bf1aec380 100644 --- a/Documentation/devicetree/bindings/watchdog/renesas,wdt.yaml +++ b/Documentation/devicetree/bindings/watchdog/renesas,wdt.yaml @@ -26,7 +26,7 @@ properties: - items: - enum: - - renesas,r9a07g043-wdt # RZ/G2UL + - renesas,r9a07g043-wdt # RZ/G2UL and RZ/Five - renesas,r9a07g044-wdt # RZ/G2{L,LC} - renesas,r9a07g054-wdt # RZ/V2L - const: renesas,rzg2l-wdt From e9651838fe7727c1c8b193b0217307c4c2260777 Mon Sep 17 00:00:00 2001 From: Konrad Dybcio Date: Thu, 17 Nov 2022 11:58:45 +0100 Subject: [PATCH 300/765] dt-bindings: watchdog: Add MSM8994 watchdog timer Document the MSM8994 watchdog timer which is already used in DT. Signed-off-by: Konrad Dybcio Acked-by: Krzysztof Kozlowski Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221117105845.13644-2-konrad.dybcio@linaro.org Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml b/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml index d8ac0be36e6c8..a1f17c9e02db9 100644 --- a/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml +++ b/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml @@ -17,6 +17,7 @@ properties: oneOf: - items: - enum: + - qcom,apss-wdt-msm8994 - qcom,apss-wdt-qcs404 - qcom,apss-wdt-sc7180 - qcom,apss-wdt-sc7280 From 6ba6f0f5910d5916539268c0ad55657bb8940616 Mon Sep 17 00:00:00 2001 From: Lad Prabhakar Date: Thu, 17 Nov 2022 11:49:06 +0000 Subject: [PATCH 301/765] watchdog: rzg2l_wdt: Issue a reset before we put the PM clocks On RZ/Five SoC it was observed that setting timeout (to say 1 sec) wouldn't reset the system. The procedure described in the HW manual (Procedure for Activating Modules) for activating the target module states we need to start supply of the clock module before applying the reset signal. This patch makes sure we follow the same procedure to clear the registers of the WDT module, fixing the issues seen on RZ/Five SoC. While at it re-used rzg2l_wdt_stop() in rzg2l_wdt_set_timeout() as it has the same function calls. Fixes: 4055ee81009e ("watchdog: rzg2l_wdt: Add set_timeout callback") Signed-off-by: Lad Prabhakar Reviewed-by: Guenter Roeck Reviewed-by: Geert Uytterhoeven Reviewed-by: Biju Das Link: https://lore.kernel.org/r/20221117114907.138583-2-fabrizio.castro.jz@renesas.com Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/rzg2l_wdt.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/drivers/watchdog/rzg2l_wdt.c b/drivers/watchdog/rzg2l_wdt.c index 974a4194a8fd6..ceca42db08374 100644 --- a/drivers/watchdog/rzg2l_wdt.c +++ b/drivers/watchdog/rzg2l_wdt.c @@ -115,25 +115,23 @@ static int rzg2l_wdt_stop(struct watchdog_device *wdev) { struct rzg2l_wdt_priv *priv = watchdog_get_drvdata(wdev); - pm_runtime_put(wdev->parent); reset_control_reset(priv->rstc); + pm_runtime_put(wdev->parent); return 0; } static int rzg2l_wdt_set_timeout(struct watchdog_device *wdev, unsigned int timeout) { - struct rzg2l_wdt_priv *priv = watchdog_get_drvdata(wdev); - wdev->timeout = timeout; /* * If the watchdog is active, reset the module for updating the WDTSET - * register so that it is updated with new timeout values. + * register by calling rzg2l_wdt_stop() (which internally calls reset_control_reset() + * to reset the module) so that it is updated with new timeout values. */ if (watchdog_active(wdev)) { - pm_runtime_put(wdev->parent); - reset_control_reset(priv->rstc); + rzg2l_wdt_stop(wdev); rzg2l_wdt_start(wdev); } From f769f97917c1e756e12ff042a93f6e3167254b5b Mon Sep 17 00:00:00 2001 From: Fabrizio Castro Date: Thu, 17 Nov 2022 11:49:07 +0000 Subject: [PATCH 302/765] watchdog: rzg2l_wdt: Handle TYPE-B reset for RZ/V2M As per section 48.4 of the HW User Manual, IPs in the RZ/V2M SoC need either a TYPE-A reset sequence or a TYPE-B reset sequence. More specifically, the watchdog IP needs a TYPE-B reset sequence. If the proper reset sequence isn't implemented, then resetting IPs may lead to undesired behaviour. In the restart callback of the watchdog driver the reset has basically no effect on the desired funcionality, as the register writes following the reset happen before the IP manages to come out of reset. Implement the TYPE-B reset sequence in the watchdog driver to address the issues with the restart callback on RZ/V2M. Fixes: ec122fd94eeb ("watchdog: rzg2l_wdt: Add rzv2m support") Signed-off-by: Fabrizio Castro Reviewed-by: Guenter Roeck Reviewed-by: Geert Uytterhoeven Link: https://lore.kernel.org/r/20221117114907.138583-3-fabrizio.castro.jz@renesas.com Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/rzg2l_wdt.c | 37 +++++++++++++++++++++++++++++++++++- 1 file changed, 36 insertions(+), 1 deletion(-) diff --git a/drivers/watchdog/rzg2l_wdt.c b/drivers/watchdog/rzg2l_wdt.c index ceca42db08374..d404953d0e0f4 100644 --- a/drivers/watchdog/rzg2l_wdt.c +++ b/drivers/watchdog/rzg2l_wdt.c @@ -8,6 +8,7 @@ #include #include #include +#include #include #include #include @@ -35,6 +36,8 @@ #define F2CYCLE_NSEC(f) (1000000000 / (f)) +#define RZV2M_A_NSEC 730 + static bool nowayout = WATCHDOG_NOWAYOUT; module_param(nowayout, bool, 0); MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default=" @@ -51,11 +54,35 @@ struct rzg2l_wdt_priv { struct reset_control *rstc; unsigned long osc_clk_rate; unsigned long delay; + unsigned long minimum_assertion_period; struct clk *pclk; struct clk *osc_clk; enum rz_wdt_type devtype; }; +static int rzg2l_wdt_reset(struct rzg2l_wdt_priv *priv) +{ + int err, status; + + if (priv->devtype == WDT_RZV2M) { + /* WDT needs TYPE-B reset control */ + err = reset_control_assert(priv->rstc); + if (err) + return err; + ndelay(priv->minimum_assertion_period); + err = reset_control_deassert(priv->rstc); + if (err) + return err; + err = read_poll_timeout(reset_control_status, status, + status != 1, 0, 1000, false, + priv->rstc); + } else { + err = reset_control_reset(priv->rstc); + } + + return err; +} + static void rzg2l_wdt_wait_delay(struct rzg2l_wdt_priv *priv) { /* delay timer when change the setting register */ @@ -115,7 +142,7 @@ static int rzg2l_wdt_stop(struct watchdog_device *wdev) { struct rzg2l_wdt_priv *priv = watchdog_get_drvdata(wdev); - reset_control_reset(priv->rstc); + rzg2l_wdt_reset(priv); pm_runtime_put(wdev->parent); return 0; @@ -154,6 +181,7 @@ static int rzg2l_wdt_restart(struct watchdog_device *wdev, rzg2l_wdt_write(priv, PEEN_FORCE, PEEN); } else { /* RZ/V2M doesn't have parity error registers */ + rzg2l_wdt_reset(priv); wdev->timeout = 0; @@ -251,6 +279,13 @@ static int rzg2l_wdt_probe(struct platform_device *pdev) priv->devtype = (uintptr_t)of_device_get_match_data(dev); + if (priv->devtype == WDT_RZV2M) { + priv->minimum_assertion_period = RZV2M_A_NSEC + + 3 * F2CYCLE_NSEC(pclk_rate) + 5 * + max(F2CYCLE_NSEC(priv->osc_clk_rate), + F2CYCLE_NSEC(pclk_rate)); + } + pm_runtime_enable(&pdev->dev); priv->wdev.info = &rzg2l_wdt_ident; From 07bec0e09c1afbab4c5674fd2341f4f52d594f30 Mon Sep 17 00:00:00 2001 From: ruanjinjie Date: Wed, 16 Nov 2022 17:49:50 +0800 Subject: [PATCH 303/765] watchdog: at91sam9_wdt: use devm_request_irq to avoid missing free_irq() in error path free_irq() is missing in case of error in at91_wdt_init(), use devm_request_irq to fix that. Fixes: 5161b31dc39a ("watchdog: at91sam9_wdt: better watchdog support") Signed-off-by: ruanjinjie Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221116094950.3141943-1-ruanjinjie@huawei.com [groeck: Adjust multi-line alignment] Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/at91sam9_wdt.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/drivers/watchdog/at91sam9_wdt.c b/drivers/watchdog/at91sam9_wdt.c index 292b5a1ca8318..fed7be2464420 100644 --- a/drivers/watchdog/at91sam9_wdt.c +++ b/drivers/watchdog/at91sam9_wdt.c @@ -206,10 +206,9 @@ static int at91_wdt_init(struct platform_device *pdev, struct at91wdt *wdt) "min heartbeat and max heartbeat might be too close for the system to handle it correctly\n"); if ((tmp & AT91_WDT_WDFIEN) && wdt->irq) { - err = request_irq(wdt->irq, wdt_interrupt, - IRQF_SHARED | IRQF_IRQPOLL | - IRQF_NO_SUSPEND, - pdev->name, wdt); + err = devm_request_irq(dev, wdt->irq, wdt_interrupt, + IRQF_SHARED | IRQF_IRQPOLL | IRQF_NO_SUSPEND, + pdev->name, wdt); if (err) return err; } From a243cb9357cc21a441c464e63b151e292a732d56 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 5 Nov 2022 12:09:34 +0100 Subject: [PATCH 304/765] watchdog: Include when appropriate The kstrto() functions have been moved from kernel.h to kstrtox.h. So, in order to eventually remove from , include the latter directly in the appropriate files. Signed-off-by: Christophe JAILLET Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/08fd5512e569558231247515c04c8596a1d11004.1667646547.git.christophe.jaillet@wanadoo.fr Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/aspeed_wdt.c | 1 + drivers/watchdog/watchdog_dev.c | 1 + 2 files changed, 2 insertions(+) diff --git a/drivers/watchdog/aspeed_wdt.c b/drivers/watchdog/aspeed_wdt.c index 86b5331bc4911..c1e79874a2bbc 100644 --- a/drivers/watchdog/aspeed_wdt.c +++ b/drivers/watchdog/aspeed_wdt.c @@ -10,6 +10,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/watchdog/watchdog_dev.c b/drivers/watchdog/watchdog_dev.c index 55574ed425042..f31608f3e3244 100644 --- a/drivers/watchdog/watchdog_dev.c +++ b/drivers/watchdog/watchdog_dev.c @@ -35,6 +35,7 @@ #include /* For __init/__exit/... */ #include /* For hrtimers */ #include /* For printk/panic/... */ +#include /* For kstrto* */ #include /* For kthread_work */ #include /* For handling misc devices */ #include /* For module stuff/... */ From 13721a2ac66b246f5802ba1b75ad8637e53eeecc Mon Sep 17 00:00:00 2001 From: Chen Jun Date: Wed, 16 Nov 2022 01:27:14 +0000 Subject: [PATCH 305/765] watchdog: Fix kmemleak in watchdog_cdev_register kmemleak reports memory leaks in watchdog_dev_register, as follows: unreferenced object 0xffff888116233000 (size 2048): comm ""modprobe"", pid 28147, jiffies 4353426116 (age 61.741s) hex dump (first 32 bytes): 80 fa b9 05 81 88 ff ff 08 30 23 16 81 88 ff ff .........0#..... 08 30 23 16 81 88 ff ff 00 00 00 00 00 00 00 00 .0#............. backtrace: [<000000007f001ffd>] __kmem_cache_alloc_node+0x157/0x220 [<000000006a389304>] kmalloc_trace+0x21/0x110 [<000000008d640eea>] watchdog_dev_register+0x4e/0x780 [watchdog] [<0000000053c9f248>] __watchdog_register_device+0x4f0/0x680 [watchdog] [<00000000b2979824>] watchdog_register_device+0xd2/0x110 [watchdog] [<000000001f730178>] 0xffffffffc10880ae [<000000007a1a8bcc>] do_one_initcall+0xcb/0x4d0 [<00000000b98be325>] do_init_module+0x1ca/0x5f0 [<0000000046d08e7c>] load_module+0x6133/0x70f0 ... unreferenced object 0xffff888105b9fa80 (size 16): comm ""modprobe"", pid 28147, jiffies 4353426116 (age 61.741s) hex dump (first 16 bytes): 77 61 74 63 68 64 6f 67 31 00 b9 05 81 88 ff ff watchdog1....... backtrace: [<000000007f001ffd>] __kmem_cache_alloc_node+0x157/0x220 [<00000000486ab89b>] __kmalloc_node_track_caller+0x44/0x1b0 [<000000005a39aab0>] kvasprintf+0xb5/0x140 [<0000000024806f85>] kvasprintf_const+0x55/0x180 [<000000009276cb7f>] kobject_set_name_vargs+0x56/0x150 [<00000000a92e820b>] dev_set_name+0xab/0xe0 [<00000000cec812c6>] watchdog_dev_register+0x285/0x780 [watchdog] [<0000000053c9f248>] __watchdog_register_device+0x4f0/0x680 [watchdog] [<00000000b2979824>] watchdog_register_device+0xd2/0x110 [watchdog] [<000000001f730178>] 0xffffffffc10880ae [<000000007a1a8bcc>] do_one_initcall+0xcb/0x4d0 [<00000000b98be325>] do_init_module+0x1ca/0x5f0 [<0000000046d08e7c>] load_module+0x6133/0x70f0 ... The reason is that put_device is not be called if cdev_device_add fails and wdd->id != 0. watchdog_cdev_register wd_data = kzalloc [1] err = dev_set_name [2] .. err = cdev_device_add if (err) { if (wdd->id == 0) { // wdd->id != 0 .. } return err; // [1],[2] would be leaked To fix it, call put_device in all wdd->id cases. Fixes: 72139dfa2464 ("watchdog: Fix the race between the release of watchdog_core_data and cdev") Signed-off-by: Chen Jun Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221116012714.102066-1-chenjun102@huawei.com Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/watchdog_dev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/watchdog/watchdog_dev.c b/drivers/watchdog/watchdog_dev.c index f31608f3e3244..455a9a572d7bb 100644 --- a/drivers/watchdog/watchdog_dev.c +++ b/drivers/watchdog/watchdog_dev.c @@ -1062,8 +1062,8 @@ static int watchdog_cdev_register(struct watchdog_device *wdd) if (wdd->id == 0) { misc_deregister(&watchdog_miscdev); old_wd_data = NULL; - put_device(&wd_data->dev); } + put_device(&wd_data->dev); return err; } From 8e103beca676b871e06d282ed140bf5a3631e381 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 13 Jan 2023 11:33:40 +0100 Subject: [PATCH 306/765] dt-bindings: watchdog: qcom-wdt: require fallback for IPQ4019 The device specific compatibles ("qcom,kpss-wdt-ipq4019") should be follwed by fallback "qcom,kpss-wdt", which is actually used by Linux driver for binding. Signed-off-by: Krzysztof Kozlowski Acked-by: Rob Herring Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221212163532.142533-1-krzysztof.kozlowski@linaro.org Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml b/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml index a1f17c9e02db9..e76364c52fc73 100644 --- a/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml +++ b/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml @@ -17,6 +17,7 @@ properties: oneOf: - items: - enum: + - qcom,kpss-wdt-ipq4019 - qcom,apss-wdt-msm8994 - qcom,apss-wdt-qcs404 - qcom,apss-wdt-sc7180 @@ -35,7 +36,6 @@ properties: - qcom,kpss-wdt - qcom,kpss-timer - qcom,kpss-wdt-apq8064 - - qcom,kpss-wdt-ipq4019 - qcom,kpss-wdt-ipq8064 - qcom,kpss-wdt-msm8960 - qcom,scss-timer From f8e6b3d9e3cf60bb111ccc9294eb3fe7c7759193 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 13 Jan 2023 11:33:41 +0100 Subject: [PATCH 307/765] dt-bindings: watchdog: qcom-wdt: do not allow fallback alone The compatible "qcom,kpss-wdt" is too generic and should not be used alone. Mark it as deprecated when not prepended by specific compatible. Signed-off-by: Krzysztof Kozlowski Acked-by: Rob Herring Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221212163532.142533-2-krzysztof.kozlowski@linaro.org Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml b/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml index e76364c52fc73..3e0b30a817d64 100644 --- a/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml +++ b/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml @@ -31,9 +31,10 @@ properties: - qcom,apss-wdt-sm8150 - qcom,apss-wdt-sm8250 - const: qcom,kpss-wdt + - const: qcom,kpss-wdt + deprecated: true - items: - enum: - - qcom,kpss-wdt - qcom,kpss-timer - qcom,kpss-wdt-apq8064 - qcom,kpss-wdt-ipq8064 From b8c8f8649ba1765cbd7e5c92f36c6553a9fadedf Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 13 Jan 2023 11:33:42 +0100 Subject: [PATCH 308/765] dt-bindings: watchdog: qcom-wdt: fix list of MSM timer compatibles The MSM timer ("qcom,msm-timer") is a bit different timer and watchdog device than KPSS watchdog. It has its own generic and specific compatibles, so fix the list to reflect this. Adjust the example to show the newer KPSS watchdog. Signed-off-by: Krzysztof Kozlowski Acked-by: Rob Herring Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221212163532.142533-3-krzysztof.kozlowski@linaro.org Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- .../devicetree/bindings/watchdog/qcom-wdt.yaml | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml b/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml index 3e0b30a817d64..93e4381067ddc 100644 --- a/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml +++ b/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml @@ -33,13 +33,16 @@ properties: - const: qcom,kpss-wdt - const: qcom,kpss-wdt deprecated: true + - items: + - const: qcom,scss-timer + - const: qcom,msm-timer - items: - enum: - - qcom,kpss-timer - qcom,kpss-wdt-apq8064 - qcom,kpss-wdt-ipq8064 - qcom,kpss-wdt-msm8960 - - qcom,scss-timer + - const: qcom,kpss-timer + - const: qcom,msm-timer reg: maxItems: 1 @@ -56,9 +59,11 @@ unevaluatedProperties: false examples: - | - watchdog@208a038 { - compatible = "qcom,kpss-wdt-ipq8064"; - reg = <0x0208a038 0x40>; + #include + + watchdog@17c10000 { + compatible = "qcom,apss-wdt-sm8150", "qcom,kpss-wdt"; + reg = <0x17c10000 0x1000>; clocks = <&sleep_clk>; timeout-sec = <10>; }; From 2bb28d179891cb073ea96fb9238257f871e158c1 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 13 Jan 2023 11:33:43 +0100 Subject: [PATCH 309/765] dt-bindings: watchdog: qcom-wdt: add qcom,kpss-wdt-mdm9615 Document new MDM9615 qcom,kpss-wdt-mdm9615 watchdog/timer compatible. Signed-off-by: Krzysztof Kozlowski Acked-by: Rob Herring Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221212163532.142533-4-krzysztof.kozlowski@linaro.org Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml b/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml index 93e4381067ddc..1828eaf70b3b1 100644 --- a/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml +++ b/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml @@ -40,6 +40,7 @@ properties: - enum: - qcom,kpss-wdt-apq8064 - qcom,kpss-wdt-ipq8064 + - qcom,kpss-wdt-mdm9615 - qcom,kpss-wdt-msm8960 - const: qcom,kpss-timer - const: qcom,msm-timer From 7c631cdff391dd8af4e56fc4cd162d6e6504aeec Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 13 Jan 2023 11:33:44 +0100 Subject: [PATCH 310/765] dt-bindings: watchdog: qcom-wdt: allow interrupts Both of type of watchdogs described in the binding (the KPSS watchdog and APSS WDT timer) have interrupts. Allow interrupts and describe them for KPSS watchdog to fix warnings like: watchdog@17c10000: Unevaluated properties are not allowed ('interrupts' was unexpected) Signed-off-by: Krzysztof Kozlowski Reviewed-by: Rob Herring Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221212163532.142533-5-krzysztof.kozlowski@linaro.org Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- .../bindings/watchdog/qcom-wdt.yaml | 24 ++++++++++++++++--- 1 file changed, 21 insertions(+), 3 deletions(-) diff --git a/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml b/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml index 1828eaf70b3b1..b7fc57f4800e9 100644 --- a/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml +++ b/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml @@ -9,9 +9,6 @@ title: Qualcomm Krait Processor Sub-system (KPSS) Watchdog timer maintainers: - Sai Prakash Ranjan -allOf: - - $ref: watchdog.yaml# - properties: compatible: oneOf: @@ -51,11 +48,31 @@ properties: clocks: maxItems: 1 + interrupts: + minItems: 1 + maxItems: 5 + required: - compatible - reg - clocks +allOf: + - $ref: watchdog.yaml# + + - if: + properties: + compatible: + contains: + const: qcom,kpss-wdt + then: + properties: + interrupts: + minItems: 1 + items: + - description: Bark + - description: Bite + unevaluatedProperties: false examples: @@ -66,5 +83,6 @@ examples: compatible = "qcom,apss-wdt-sm8150", "qcom,kpss-wdt"; reg = <0x17c10000 0x1000>; clocks = <&sleep_clk>; + interrupts = ; timeout-sec = <10>; }; From 733318534849044089de8c818a2fb1c67fa39bd1 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 13 Jan 2023 11:33:45 +0100 Subject: [PATCH 311/765] dt-bindings: watchdog: qcom-wdt: merge MSM timer Merge Qualcomm MSM timer bindings into watchdog, because the timer compatibles are already included here and the hardware is quite similar. While converting the MSM timer bindings, adjust clock-frequency property to take only one frequency, instead of two, because: 1. DT schema does not allow to frequencies, 2. The Linux timer driver reads only first frequency. Signed-off-by: Krzysztof Kozlowski Acked-by: Daniel Lezcano Reviewed-by: Guenter Roeck Reviewed-by: Rob Herring Link: https://lore.kernel.org/r/20221212163532.142533-6-krzysztof.kozlowski@linaro.org Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- .../bindings/timer/qcom,msm-timer.txt | 47 ------------------ .../bindings/watchdog/qcom-wdt.yaml | 49 +++++++++++++++++++ 2 files changed, 49 insertions(+), 47 deletions(-) delete mode 100644 Documentation/devicetree/bindings/timer/qcom,msm-timer.txt diff --git a/Documentation/devicetree/bindings/timer/qcom,msm-timer.txt b/Documentation/devicetree/bindings/timer/qcom,msm-timer.txt deleted file mode 100644 index 5e10c345548f5..0000000000000 --- a/Documentation/devicetree/bindings/timer/qcom,msm-timer.txt +++ /dev/null @@ -1,47 +0,0 @@ -* MSM Timer - -Properties: - -- compatible : Should at least contain "qcom,msm-timer". More specific - properties specify which subsystem the timers are paired with. - - "qcom,kpss-timer" - krait subsystem - "qcom,scss-timer" - scorpion subsystem - -- interrupts : Interrupts for the debug timer, the first general purpose - timer, and optionally a second general purpose timer, and - optionally as well, 2 watchdog interrupts, in that order. - -- reg : Specifies the base address of the timer registers. - -- clocks: Reference to the parent clocks, one per output clock. The parents - must appear in the same order as the clock names. - -- clock-names: The name of the clocks as free-form strings. They should be in - the same order as the clocks. - -- clock-frequency : The frequency of the debug timer and the general purpose - timer(s) in Hz in that order. - -Optional: - -- cpu-offset : per-cpu offset used when the timer is accessed without the - CPU remapping facilities. The offset is - cpu-offset + (0x10000 * cpu-nr). - -Example: - - timer@200a000 { - compatible = "qcom,scss-timer", "qcom,msm-timer"; - interrupts = <1 1 0x301>, - <1 2 0x301>, - <1 3 0x301>, - <1 4 0x301>, - <1 5 0x301>; - reg = <0x0200a000 0x100>; - clock-frequency = <19200000>, - <32768>; - clocks = <&sleep_clk>; - clock-names = "sleep"; - cpu-offset = <0x40000>; - }; diff --git a/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml b/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml index b7fc57f4800e9..837ce9112071e 100644 --- a/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml +++ b/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml @@ -10,6 +10,9 @@ maintainers: - Sai Prakash Ranjan properties: + $nodename: + pattern: "^(watchdog|timer)@[0-9a-f]+$" + compatible: oneOf: - items: @@ -48,6 +51,20 @@ properties: clocks: maxItems: 1 + clock-names: + items: + - const: sleep + + clock-frequency: + description: + The frequency of the general purpose timer in Hz. + + cpu-offset: + $ref: /schemas/types.yaml#/definitions/uint32 + description: + Per-CPU offset used when the timer is accessed without the CPU remapping + facilities. The offset is cpu-offset + (0x10000 * cpu-nr). + interrupts: minItems: 1 maxItems: 5 @@ -67,12 +84,27 @@ allOf: const: qcom,kpss-wdt then: properties: + clock-frequency: false + cpu-offset: false interrupts: minItems: 1 items: - description: Bark - description: Bite + else: + properties: + interrupts: + minItems: 3 + items: + - description: Debug + - description: First general purpose timer + - description: Second general purpose timer + - description: First watchdog + - description: Second watchdog + required: + - clock-frequency + unevaluatedProperties: false examples: @@ -86,3 +118,20 @@ examples: interrupts = ; timeout-sec = <10>; }; + + - | + #include + + watchdog@200a000 { + compatible = "qcom,kpss-wdt-ipq8064", "qcom,kpss-timer", "qcom,msm-timer"; + interrupts = , + , + , + , + ; + reg = <0x0200a000 0x100>; + clock-frequency = <25000000>; + clocks = <&sleep_clk>; + clock-names = "sleep"; + cpu-offset = <0x80000>; + }; From 7d06c07c67100fd0f8e6b3ab7145ce789f788117 Mon Sep 17 00:00:00 2001 From: Li Hua Date: Wed, 16 Nov 2022 10:07:06 +0800 Subject: [PATCH 312/765] watchdog: pcwd_usb: Fix attempting to access uninitialized memory MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The stack variable msb and lsb may be used uninitialized in function usb_pcwd_get_temperature and usb_pcwd_get_timeleft when usb card no response. The build waring is: drivers/watchdog/pcwd_usb.c:336:22: error: ‘lsb’ is used uninitialized in this function [-Werror=uninitialized] *temperature = (lsb * 9 / 5) + 32; ~~~~^~~ drivers/watchdog/pcwd_usb.c:328:21: note: ‘lsb’ was declared here unsigned char msb, lsb; ^~~ cc1: all warnings being treated as errors scripts/Makefile.build:250: recipe for target 'drivers/watchdog/pcwd_usb.o' failed make[3]: *** [drivers/watchdog/pcwd_usb.o] Error 1 Fixes: b7e04f8c61a4 ("mv watchdog tree under drivers") Signed-off-by: Li Hua Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221116020706.70847-1-hucool.lihua@huawei.com Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/pcwd_usb.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/watchdog/pcwd_usb.c b/drivers/watchdog/pcwd_usb.c index 1bdaf17c1d38d..8202f0a6b0935 100644 --- a/drivers/watchdog/pcwd_usb.c +++ b/drivers/watchdog/pcwd_usb.c @@ -325,7 +325,8 @@ static int usb_pcwd_set_heartbeat(struct usb_pcwd_private *usb_pcwd, int t) static int usb_pcwd_get_temperature(struct usb_pcwd_private *usb_pcwd, int *temperature) { - unsigned char msb, lsb; + unsigned char msb = 0x00; + unsigned char lsb = 0x00; usb_pcwd_send_command(usb_pcwd, CMD_READ_TEMP, &msb, &lsb); @@ -341,7 +342,8 @@ static int usb_pcwd_get_temperature(struct usb_pcwd_private *usb_pcwd, static int usb_pcwd_get_timeleft(struct usb_pcwd_private *usb_pcwd, int *time_left) { - unsigned char msb, lsb; + unsigned char msb = 0x00; + unsigned char lsb = 0x00; /* Read the time that's left before rebooting */ /* Note: if the board is not yet armed then we will read 0xFFFF */ From e42c73f1ef0da5970a71f98440193fa9b1e9fb86 Mon Sep 17 00:00:00 2001 From: Andrej Picej Date: Fri, 4 Nov 2022 08:03:56 +0100 Subject: [PATCH 313/765] watchdog: imx2_wdg: suspend watchdog in WAIT mode Putting device into the "Suspend-To-Idle" mode causes watchdog to trigger and resets the board after set watchdog timeout period elapses. Introduce new device-tree property "fsl,suspend-in-wait" which suspends watchdog in WAIT mode. This is done by setting WDW bit in WCR (Watchdog Control Register). Watchdog operation is restored after exiting WAIT mode as expected. WAIT mode corresponds with Linux's "Suspend-To-Idle". Signed-off-by: Andrej Picej Reviewed-by: Fabio Estevam Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221104070358.426657-2-andrej.picej@norik.com Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/imx2_wdt.c | 55 +++++++++++++++++++++++++++++++++++-- 1 file changed, 52 insertions(+), 3 deletions(-) diff --git a/drivers/watchdog/imx2_wdt.c b/drivers/watchdog/imx2_wdt.c index d0c5d47ddede2..19ab7b3d286b9 100644 --- a/drivers/watchdog/imx2_wdt.c +++ b/drivers/watchdog/imx2_wdt.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include #include @@ -35,6 +36,7 @@ #define IMX2_WDT_WCR 0x00 /* Control Register */ #define IMX2_WDT_WCR_WT (0xFF << 8) /* -> Watchdog Timeout Field */ +#define IMX2_WDT_WCR_WDW BIT(7) /* -> Watchdog disable for WAIT */ #define IMX2_WDT_WCR_WDA BIT(5) /* -> External Reset WDOG_B */ #define IMX2_WDT_WCR_SRS BIT(4) /* -> Software Reset Signal */ #define IMX2_WDT_WCR_WRE BIT(3) /* -> WDOG Reset Enable */ @@ -60,13 +62,19 @@ #define WDOG_SEC_TO_COUNT(s) ((s * 2 - 1) << 8) +struct imx2_wdt_data { + bool wdw_supported; +}; + struct imx2_wdt_device { struct clk *clk; struct regmap *regmap; struct watchdog_device wdog; + const struct imx2_wdt_data *data; bool ext_reset; bool clk_is_on; bool no_ping; + bool sleep_wait; }; static bool nowayout = WATCHDOG_NOWAYOUT; @@ -129,6 +137,9 @@ static inline void imx2_wdt_setup(struct watchdog_device *wdog) /* Suspend timer in low power mode, write once-only */ val |= IMX2_WDT_WCR_WDZST; + /* Suspend timer in low power WAIT mode, write once-only */ + if (wdev->sleep_wait) + val |= IMX2_WDT_WCR_WDW; /* Strip the old watchdog Time-Out value */ val &= ~IMX2_WDT_WCR_WT; /* Generate internal chip-level reset if WDOG times out */ @@ -292,6 +303,8 @@ static int __init imx2_wdt_probe(struct platform_device *pdev) wdog->max_hw_heartbeat_ms = IMX2_WDT_MAX_TIME * 1000; wdog->parent = dev; + wdev->data = of_device_get_match_data(dev); + ret = platform_get_irq(pdev, 0); if (ret > 0) if (!devm_request_irq(dev, ret, imx2_wdt_isr, 0, @@ -313,9 +326,18 @@ static int __init imx2_wdt_probe(struct platform_device *pdev) wdev->ext_reset = of_property_read_bool(dev->of_node, "fsl,ext-reset-output"); + + if (of_property_read_bool(dev->of_node, "fsl,suspend-in-wait")) { + if (!wdev->data->wdw_supported) { + dev_err(dev, "suspend-in-wait not supported\n"); + return -EINVAL; + } + wdev->sleep_wait = true; + } + /* * The i.MX7D doesn't support low power mode, so we need to ping the watchdog - * during suspend. + * during suspend. Interaction with "fsl,suspend-in-wait" is unknown! */ wdev->no_ping = !of_device_is_compatible(dev->of_node, "fsl,imx7d-wdt"); platform_set_drvdata(pdev, wdog); @@ -417,9 +439,36 @@ static int __maybe_unused imx2_wdt_resume(struct device *dev) static SIMPLE_DEV_PM_OPS(imx2_wdt_pm_ops, imx2_wdt_suspend, imx2_wdt_resume); +struct imx2_wdt_data imx_wdt = { + .wdw_supported = true, +}; + +struct imx2_wdt_data imx_wdt_legacy = { + .wdw_supported = false, +}; + static const struct of_device_id imx2_wdt_dt_ids[] = { - { .compatible = "fsl,imx21-wdt", }, - { .compatible = "fsl,imx7d-wdt", }, + { .compatible = "fsl,imx21-wdt", .data = &imx_wdt_legacy }, + { .compatible = "fsl,imx25-wdt", .data = &imx_wdt }, + { .compatible = "fsl,imx27-wdt", .data = &imx_wdt_legacy }, + { .compatible = "fsl,imx31-wdt", .data = &imx_wdt_legacy }, + { .compatible = "fsl,imx35-wdt", .data = &imx_wdt }, + { .compatible = "fsl,imx50-wdt", .data = &imx_wdt }, + { .compatible = "fsl,imx51-wdt", .data = &imx_wdt }, + { .compatible = "fsl,imx53-wdt", .data = &imx_wdt }, + { .compatible = "fsl,imx6q-wdt", .data = &imx_wdt }, + { .compatible = "fsl,imx6sl-wdt", .data = &imx_wdt }, + { .compatible = "fsl,imx6sll-wdt", .data = &imx_wdt }, + { .compatible = "fsl,imx6sx-wdt", .data = &imx_wdt }, + { .compatible = "fsl,imx6ul-wdt", .data = &imx_wdt }, + { .compatible = "fsl,imx7d-wdt", .data = &imx_wdt }, + { .compatible = "fsl,imx8mm-wdt", .data = &imx_wdt }, + { .compatible = "fsl,imx8mn-wdt", .data = &imx_wdt }, + { .compatible = "fsl,imx8mp-wdt", .data = &imx_wdt }, + { .compatible = "fsl,imx8mq-wdt", .data = &imx_wdt }, + { .compatible = "fsl,ls1012a-wdt", .data = &imx_wdt_legacy }, + { .compatible = "fsl,ls1043a-wdt", .data = &imx_wdt_legacy }, + { .compatible = "fsl,vf610-wdt", .data = &imx_wdt }, { /* sentinel */ } }; MODULE_DEVICE_TABLE(of, imx2_wdt_dt_ids); From 12878a9fff5a2d45f4f5c71c83fee77f94aa1f8c Mon Sep 17 00:00:00 2001 From: Andrej Picej Date: Fri, 4 Nov 2022 08:03:57 +0100 Subject: [PATCH 314/765] dt-bindings: watchdog: fsl-imx: document suspend in wait mode Property "fsl,suspend-in-wait" suspends watchdog in "WAIT" mode which corresponds to Linux's Suspend-to-Idle S0 mode. If this property is not set and the device is put into Suspend-to-Idle mode, the watchdog triggers a reset after 128 seconds. Signed-off-by: Andrej Picej Reviewed-by: Fabio Estevam Reviewed-by: Krzysztof Kozlowski Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221104070358.426657-3-andrej.picej@norik.com Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- .../bindings/watchdog/fsl-imx-wdt.yaml | 37 +++++++++++++++++-- 1 file changed, 34 insertions(+), 3 deletions(-) diff --git a/Documentation/devicetree/bindings/watchdog/fsl-imx-wdt.yaml b/Documentation/devicetree/bindings/watchdog/fsl-imx-wdt.yaml index fb7695515be1a..181f0cc5b5bde 100644 --- a/Documentation/devicetree/bindings/watchdog/fsl-imx-wdt.yaml +++ b/Documentation/devicetree/bindings/watchdog/fsl-imx-wdt.yaml @@ -9,9 +9,6 @@ title: Freescale i.MX Watchdog Timer (WDT) Controller maintainers: - Anson Huang -allOf: - - $ref: "watchdog.yaml#" - properties: compatible: oneOf: @@ -55,11 +52,45 @@ properties: If present, the watchdog device is configured to assert its external reset (WDOG_B) instead of issuing a software reset. + fsl,suspend-in-wait: + $ref: /schemas/types.yaml#/definitions/flag + description: | + If present, the watchdog device is suspended in WAIT mode + (Suspend-to-Idle). Only supported on certain devices. + required: - compatible - interrupts - reg +allOf: + - $ref: watchdog.yaml# + - if: + not: + properties: + compatible: + contains: + enum: + - fsl,imx25-wdt + - fsl,imx35-wdt + - fsl,imx50-wdt + - fsl,imx51-wdt + - fsl,imx53-wdt + - fsl,imx6q-wdt + - fsl,imx6sl-wdt + - fsl,imx6sll-wdt + - fsl,imx6sx-wdt + - fsl,imx6ul-wdt + - fsl,imx7d-wdt + - fsl,imx8mm-wdt + - fsl,imx8mn-wdt + - fsl,imx8mp-wdt + - fsl,imx8mq-wdt + - fsl,vf610-wdt + then: + properties: + fsl,suspend-in-wait: false + unevaluatedProperties: false examples: From b852e7a4a9ab6a1fb08c5aa36b1043672fd814dc Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Mon, 19 Dec 2022 21:30:39 +0000 Subject: [PATCH 315/765] watchdog: report fw_version in sysfs MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This synchronizes the information reported by ioctl and sysfs. The mismatch is confusing because "wdctl" from util-linux uses the ioctl when used with root privileges and sysfs without. The file is called "fw_version" instead of "firmware_version" as "firmware_version" is already used as custom attribute by single drivers. Signed-off-by: Thomas Weißschuh Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221216-watchdog-sysfs-v2-1-6189311103a9@weissschuh.net Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- Documentation/ABI/testing/sysfs-class-watchdog | 7 +++++++ drivers/watchdog/watchdog_dev.c | 10 ++++++++++ 2 files changed, 17 insertions(+) diff --git a/Documentation/ABI/testing/sysfs-class-watchdog b/Documentation/ABI/testing/sysfs-class-watchdog index 585caecda3a5f..27c000238fe44 100644 --- a/Documentation/ABI/testing/sysfs-class-watchdog +++ b/Documentation/ABI/testing/sysfs-class-watchdog @@ -6,6 +6,13 @@ Description: device at boot. It is equivalent to WDIOC_GETBOOTSTATUS of ioctl interface. +What: /sys/class/watchdog/watchdogn/fw_version +Date: April 2023 +Contact: Thomas Weißschuh +Description: + It is a read only file. It contains firmware version of + watchdog device. + What: /sys/class/watchdog/watchdogn/identity Date: August 2015 Contact: Wim Van Sebroeck diff --git a/drivers/watchdog/watchdog_dev.c b/drivers/watchdog/watchdog_dev.c index 455a9a572d7bb..bd5612fa21902 100644 --- a/drivers/watchdog/watchdog_dev.c +++ b/drivers/watchdog/watchdog_dev.c @@ -547,6 +547,15 @@ static ssize_t pretimeout_show(struct device *dev, } static DEVICE_ATTR_RO(pretimeout); +static ssize_t fw_version_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct watchdog_device *wdd = dev_get_drvdata(dev); + + return sysfs_emit(buf, "%d\n", wdd->info->firmware_version); +} +static DEVICE_ATTR_RO(fw_version); + static ssize_t identity_show(struct device *dev, struct device_attribute *attr, char *buf) { @@ -618,6 +627,7 @@ static umode_t wdt_is_visible(struct kobject *kobj, struct attribute *attr, } static struct attribute *wdt_attrs[] = { &dev_attr_state.attr, + &dev_attr_fw_version.attr, &dev_attr_identity.attr, &dev_attr_timeout.attr, &dev_attr_min_timeout.attr, From ad8bc199b2e3197f53228fba05128069e31893c7 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Mon, 19 Dec 2022 21:30:40 +0000 Subject: [PATCH 316/765] watchdog: report options in sysfs MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This synchronizes the information reported by ioctl and sysfs. The mismatch is confusing because "wdctl" from util-linux uses the ioctl when used with root privileges and sysfs without. Signed-off-by: Thomas Weißschuh Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20221216-watchdog-sysfs-v2-2-6189311103a9@weissschuh.net [groeck: Fixed continuation line alignment] Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- Documentation/ABI/testing/sysfs-class-watchdog | 6 ++++++ drivers/watchdog/watchdog_dev.c | 10 ++++++++++ 2 files changed, 16 insertions(+) diff --git a/Documentation/ABI/testing/sysfs-class-watchdog b/Documentation/ABI/testing/sysfs-class-watchdog index 27c000238fe44..94fb746159512 100644 --- a/Documentation/ABI/testing/sysfs-class-watchdog +++ b/Documentation/ABI/testing/sysfs-class-watchdog @@ -6,6 +6,12 @@ Description: device at boot. It is equivalent to WDIOC_GETBOOTSTATUS of ioctl interface. +What: /sys/class/watchdog/watchdogn/options +Date: April 2023 +Contact: Thomas Weißschuh +Description: + It is a read only file. It contains options of watchdog device. + What: /sys/class/watchdog/watchdogn/fw_version Date: April 2023 Contact: Thomas Weißschuh diff --git a/drivers/watchdog/watchdog_dev.c b/drivers/watchdog/watchdog_dev.c index bd5612fa21902..0122e87968797 100644 --- a/drivers/watchdog/watchdog_dev.c +++ b/drivers/watchdog/watchdog_dev.c @@ -547,6 +547,15 @@ static ssize_t pretimeout_show(struct device *dev, } static DEVICE_ATTR_RO(pretimeout); +static ssize_t options_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct watchdog_device *wdd = dev_get_drvdata(dev); + + return sysfs_emit(buf, "0x%x\n", wdd->info->options); +} +static DEVICE_ATTR_RO(options); + static ssize_t fw_version_show(struct device *dev, struct device_attribute *attr, char *buf) { @@ -627,6 +636,7 @@ static umode_t wdt_is_visible(struct kobject *kobj, struct attribute *attr, } static struct attribute *wdt_attrs[] = { &dev_attr_state.attr, + &dev_attr_options.attr, &dev_attr_fw_version.attr, &dev_attr_identity.attr, &dev_attr_timeout.attr, From c76675f4ef35504eb1df8f0f0f0c1e2675d75f67 Mon Sep 17 00:00:00 2001 From: Bartosz Golaszewski Date: Wed, 1 Feb 2023 16:20:36 +0100 Subject: [PATCH 317/765] dt-bindings: watchdog: qcom-wdt: add qcom,apss-wdt-sa8775p compatible Add a compatible for the sa8775p platform's KPSS watchdog. Signed-off-by: Bartosz Golaszewski Acked-by: Krzysztof Kozlowski Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20230201152038.203387-4-brgl@bgdev.pl Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml b/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml index 837ce9112071e..6448b633c9702 100644 --- a/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml +++ b/Documentation/devicetree/bindings/watchdog/qcom-wdt.yaml @@ -20,6 +20,7 @@ properties: - qcom,kpss-wdt-ipq4019 - qcom,apss-wdt-msm8994 - qcom,apss-wdt-qcs404 + - qcom,apss-wdt-sa8775p - qcom,apss-wdt-sc7180 - qcom,apss-wdt-sc7280 - qcom,apss-wdt-sc8180x From 000987a38b53c172f435142a4026dd71378ca464 Mon Sep 17 00:00:00 2001 From: George Cherian Date: Thu, 9 Feb 2023 02:11:17 +0000 Subject: [PATCH 318/765] watchdog: sbsa_wdog: Make sure the timeout programming is within the limits Make sure to honour the max_hw_heartbeat_ms while programming the timeout value to WOR. Clamp the timeout passed to sbsa_gwdt_set_timeout() to make sure the programmed value is within the permissible range. Fixes: abd3ac7902fb ("watchdog: sbsa: Support architecture version 1") Signed-off-by: George Cherian Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20230209021117.1512097-1-george.cherian@marvell.com Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/sbsa_gwdt.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/watchdog/sbsa_gwdt.c b/drivers/watchdog/sbsa_gwdt.c index 9791c74aebd48..63862803421f1 100644 --- a/drivers/watchdog/sbsa_gwdt.c +++ b/drivers/watchdog/sbsa_gwdt.c @@ -150,6 +150,7 @@ static int sbsa_gwdt_set_timeout(struct watchdog_device *wdd, struct sbsa_gwdt *gwdt = watchdog_get_drvdata(wdd); wdd->timeout = timeout; + timeout = clamp_t(unsigned int, timeout, 1, wdd->max_hw_heartbeat_ms / 1000); if (action) sbsa_gwdt_reg_write(gwdt->clk * timeout, gwdt); From 08abd0466ec9113908e674d042ec2a36dfc2875c Mon Sep 17 00:00:00 2001 From: Tanmay Bhushan <007047221b@gmail.com> Date: Fri, 30 Dec 2022 15:18:58 +0100 Subject: [PATCH 319/765] ext4: remove dead code in updating backup sb ext4_update_backup_sb checks for err having some value after unlocking buffer. But err has not been updated till that point in any code which will lead execution of the code in question. Signed-off-by: Tanmay Bhushan <007047221b@gmail.com> Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20221230141858.3828-1-007047221b@gmail.com Signed-off-by: Theodore Ts'o --- fs/ext4/ioctl.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index 8067ccda34e45..2e8c340363138 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -155,9 +155,6 @@ static int ext4_update_backup_sb(struct super_block *sb, set_buffer_uptodate(bh); unlock_buffer(bh); - if (err) - goto out_bh; - if (handle) { err = ext4_handle_dirty_metadata(handle, NULL, bh); if (err) From 1e9d62d252812575ded7c620d8fc67c32ff06c16 Mon Sep 17 00:00:00 2001 From: Jun Nie Date: Tue, 3 Jan 2023 09:45:16 +0800 Subject: [PATCH 320/765] ext4: optimize ea_inode block expansion Copy ea data from inode entry when expanding ea block if possible. Then remove the ea entry if expansion success. Thus memcpy to a temporary buffer may be avoided. If the expansion fails, we do not need to recovery the removed ea entry neither in this way. Reported-by: syzbot+2dacb8f015bf1420155f@syzkaller.appspotmail.com Link: https://syzkaller.appspot.com/bug?id=3613786cb88c93aa1c6a279b1df6a7b201347d08 Link: https://lore.kernel.org/r/20230103014517.495275-2-jun.nie@linaro.org Cc: stable@kernel.org Signed-off-by: Jun Nie Signed-off-by: Theodore Ts'o --- fs/ext4/xattr.c | 28 +++++++++++++++++----------- 1 file changed, 17 insertions(+), 11 deletions(-) diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index e51052a247cc6..38e08b438ccb9 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -2600,9 +2600,8 @@ static int ext4_xattr_move_to_block(handle_t *handle, struct inode *inode, is = kzalloc(sizeof(struct ext4_xattr_ibody_find), GFP_NOFS); bs = kzalloc(sizeof(struct ext4_xattr_block_find), GFP_NOFS); - buffer = kvmalloc(value_size, GFP_NOFS); b_entry_name = kmalloc(entry->e_name_len + 1, GFP_NOFS); - if (!is || !bs || !buffer || !b_entry_name) { + if (!is || !bs || !b_entry_name) { error = -ENOMEM; goto out; } @@ -2614,12 +2613,18 @@ static int ext4_xattr_move_to_block(handle_t *handle, struct inode *inode, /* Save the entry name and the entry value */ if (entry->e_value_inum) { + buffer = kvmalloc(value_size, GFP_NOFS); + if (!buffer) { + error = -ENOMEM; + goto out; + } + error = ext4_xattr_inode_get(inode, entry, buffer, value_size); if (error) goto out; } else { size_t value_offs = le16_to_cpu(entry->e_value_offs); - memcpy(buffer, (void *)IFIRST(header) + value_offs, value_size); + buffer = (void *)IFIRST(header) + value_offs; } memcpy(b_entry_name, entry->e_name, entry->e_name_len); @@ -2634,25 +2639,26 @@ static int ext4_xattr_move_to_block(handle_t *handle, struct inode *inode, if (error) goto out; - /* Remove the chosen entry from the inode */ - error = ext4_xattr_ibody_set(handle, inode, &i, is); - if (error) - goto out; - i.value = buffer; i.value_len = value_size; error = ext4_xattr_block_find(inode, &i, bs); if (error) goto out; - /* Add entry which was removed from the inode into the block */ + /* Move ea entry from the inode into the block */ error = ext4_xattr_block_set(handle, inode, &i, bs); if (error) goto out; - error = 0; + + /* Remove the chosen entry from the inode */ + i.value = NULL; + i.value_len = 0; + error = ext4_xattr_ibody_set(handle, inode, &i, is); + out: kfree(b_entry_name); - kvfree(buffer); + if (entry->e_value_inum && buffer) + kvfree(buffer); if (is) brelse(is->iloc.bh); if (bs) From f31173c19901a96bb2ebf6bcfec8a08df7095c91 Mon Sep 17 00:00:00 2001 From: Jun Nie Date: Tue, 3 Jan 2023 09:45:17 +0800 Subject: [PATCH 321/765] ext4: refuse to create ea block when umounted The ea block expansion need to access s_root while it is already set as NULL when umount is triggered. Refuse this request to avoid panic. Reported-by: syzbot+2dacb8f015bf1420155f@syzkaller.appspotmail.com Link: https://syzkaller.appspot.com/bug?id=3613786cb88c93aa1c6a279b1df6a7b201347d08 Link: https://lore.kernel.org/r/20230103014517.495275-3-jun.nie@linaro.org Cc: stable@kernel.org Signed-off-by: Jun Nie Signed-off-by: Theodore Ts'o --- fs/ext4/xattr.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 38e08b438ccb9..d8fef540ca9bf 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -1471,6 +1471,13 @@ static struct inode *ext4_xattr_inode_create(handle_t *handle, uid_t owner[2] = { i_uid_read(inode), i_gid_read(inode) }; int err; + if (inode->i_sb->s_root == NULL) { + ext4_warning(inode->i_sb, + "refuse to create EA inode when umounting"); + WARN_ON(1); + return ERR_PTR(-EINVAL); + } + /* * Let the next inode be the goal, so we try and allocate the EA inode * in the same group, or nearby one. From 3f5424790d4377839093b68c12b130077a4e4510 Mon Sep 17 00:00:00 2001 From: zhanchengbin Date: Tue, 3 Jan 2023 10:28:12 +0800 Subject: [PATCH 322/765] ext4: fix inode tree inconsistency caused by ENOMEM If ENOMEM fails when the extent is splitting, we need to restore the length of the split extent. In the ext4_split_extent_at function, only in ext4_ext_create_new_leaf will it alloc memory and change the shape of the extent tree,even if an ENOMEM is returned at this time, the extent tree is still self-consistent, Just restore the split extent lens in the function ext4_split_extent_at. ext4_split_extent_at ext4_ext_insert_extent ext4_ext_create_new_leaf 1)ext4_ext_split ext4_find_extent 2)ext4_ext_grow_indepth ext4_find_extent Signed-off-by: zhanchengbin Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20230103022812.130603-1-zhanchengbin1@huawei.com Signed-off-by: Theodore Ts'o --- fs/ext4/extents.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 9de1c9d1a13d3..3559ea6b07818 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -3251,7 +3251,7 @@ static int ext4_split_extent_at(handle_t *handle, ext4_ext_mark_unwritten(ex2); err = ext4_ext_insert_extent(handle, inode, ppath, &newex, flags); - if (err != -ENOSPC && err != -EDQUOT) + if (err != -ENOSPC && err != -EDQUOT && err != -ENOMEM) goto out; if (EXT4_EXT_MAY_ZEROOUT & split_flag) { From 7fc51f923ea6b1c6f304789965414149eaedb358 Mon Sep 17 00:00:00 2001 From: XU pengfei Date: Wed, 4 Jan 2023 13:52:29 +0800 Subject: [PATCH 323/765] ext4: remove unnecessary variable initialization Variables are assigned first and then used. Initialization is not required. Signed-off-by: XU pengfei Link: https://lore.kernel.org/r/20230104055229.3663-1-xupengfei@nfschina.com --- fs/ext4/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 9d9f414f99fec..9df913bdb4166 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -5788,7 +5788,7 @@ static int ext4_meta_trans_blocks(struct inode *inode, int lblocks, ext4_group_t groups, ngroups = ext4_get_groups_count(inode->i_sb); int gdpblocks; int idxblocks; - int ret = 0; + int ret; /* * How many index blocks need to touch to map @lblocks logical blocks From d99a55a94a0db8eda4a336a6f21730b844b8d2d2 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Wed, 4 Jan 2023 13:09:12 -0800 Subject: [PATCH 324/765] ext4: fix function prototype mismatch for ext4_feat_ktype With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG), indirect call targets are validated against the expected function pointer prototype to make sure the call target is valid to help mitigate ROP attacks. If they are not identical, there is a failure at run time, which manifests as either a kernel panic or thread getting killed. ext4_feat_ktype was setting the "release" handler to "kfree", which doesn't have a matching function prototype. Add a simple wrapper with the correct prototype. This was found as a result of Clang's new -Wcast-function-type-strict flag, which is more sensitive than the simpler -Wcast-function-type, which only checks for type width mismatches. Note that this code is only reached when ext4 is a loadable module and it is being unloaded: CFI failure at kobject_put+0xbb/0x1b0 (target: kfree+0x0/0x180; expected type: 0x7c4aa698) ... RIP: 0010:kobject_put+0xbb/0x1b0 ... Call Trace: ext4_exit_sysfs+0x14/0x60 [ext4] cleanup_module+0x67/0xedb [ext4] Fixes: b99fee58a20a ("ext4: create ext4_feat kobject dynamically") Cc: Theodore Ts'o Cc: Eric Biggers Cc: stable@vger.kernel.org Build-tested-by: Gustavo A. R. Silva Reviewed-by: Gustavo A. R. Silva Reviewed-by: Nathan Chancellor Link: https://lore.kernel.org/r/20230103234616.never.915-kees@kernel.org Signed-off-by: Kees Cook Reviewed-by: Eric Biggers Link: https://lore.kernel.org/r/20230104210908.gonna.388-kees@kernel.org Signed-off-by: Theodore Ts'o --- fs/ext4/sysfs.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c index d233c24ea3425..e2b8b3437c589 100644 --- a/fs/ext4/sysfs.c +++ b/fs/ext4/sysfs.c @@ -491,6 +491,11 @@ static void ext4_sb_release(struct kobject *kobj) complete(&sbi->s_kobj_unregister); } +static void ext4_feat_release(struct kobject *kobj) +{ + kfree(kobj); +} + static const struct sysfs_ops ext4_attr_ops = { .show = ext4_attr_show, .store = ext4_attr_store, @@ -505,7 +510,7 @@ static struct kobj_type ext4_sb_ktype = { static struct kobj_type ext4_feat_ktype = { .default_groups = ext4_feat_groups, .sysfs_ops = &ext4_attr_ops, - .release = (void (*)(struct kobject *))kfree, + .release = ext4_feat_release, }; void ext4_notify_error_sysfs(struct ext4_sb_info *sbi) From 5cd740287ae5e3f9d1c46f5bfe8778972fd6d3fe Mon Sep 17 00:00:00 2001 From: Baokun Li Date: Sat, 7 Jan 2023 11:21:25 +0800 Subject: [PATCH 325/765] ext4: fail ext4_iget if special inode unallocated MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In ext4_fill_super(), EXT4_ORPHAN_FS flag is cleared after ext4_orphan_cleanup() is executed. Therefore, when __ext4_iget() is called to get an inode whose i_nlink is 0 when the flag exists, no error is returned. If the inode is a special inode, a null pointer dereference may occur. If the value of i_nlink is 0 for any inodes (except boot loader inodes) got by using the EXT4_IGET_SPECIAL flag, the current file system is corrupted. Therefore, make the ext4_iget() function return an error if it gets such an abnormal special inode. Link: https://bugzilla.kernel.org/show_bug.cgi?id=199179 Link: https://bugzilla.kernel.org/show_bug.cgi?id=216541 Link: https://bugzilla.kernel.org/show_bug.cgi?id=216539 Reported-by: Luís Henriques Suggested-by: Theodore Ts'o Signed-off-by: Baokun Li Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20230107032126.4165860-2-libaokun1@huawei.com Signed-off-by: Theodore Ts'o --- fs/ext4/inode.c | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-) diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 9df913bdb4166..b65dadfe3b453 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -4872,13 +4872,6 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino, goto bad_inode; raw_inode = ext4_raw_inode(&iloc); - if ((ino == EXT4_ROOT_INO) && (raw_inode->i_links_count == 0)) { - ext4_error_inode(inode, function, line, 0, - "iget: root inode unallocated"); - ret = -EFSCORRUPTED; - goto bad_inode; - } - if ((flags & EXT4_IGET_HANDLE) && (raw_inode->i_links_count == 0) && (raw_inode->i_mode == 0)) { ret = -ESTALE; @@ -4951,11 +4944,16 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino, * NeilBrown 1999oct15 */ if (inode->i_nlink == 0) { - if ((inode->i_mode == 0 || + if ((inode->i_mode == 0 || flags & EXT4_IGET_SPECIAL || !(EXT4_SB(inode->i_sb)->s_mount_state & EXT4_ORPHAN_FS)) && ino != EXT4_BOOT_LOADER_INO) { - /* this inode is deleted */ - ret = -ESTALE; + /* this inode is deleted or unallocated */ + if (flags & EXT4_IGET_SPECIAL) { + ext4_error_inode(inode, function, line, 0, + "iget: special inode unallocated"); + ret = -EFSCORRUPTED; + } else + ret = -ESTALE; goto bad_inode; } /* The only unlinked inodes we let through here have From 3039d8b8692408438a618fac2776b629852663c3 Mon Sep 17 00:00:00 2001 From: Baokun Li Date: Sat, 7 Jan 2023 11:21:26 +0800 Subject: [PATCH 326/765] ext4: update s_journal_inum if it changes after journal replay MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When mounting a crafted ext4 image, s_journal_inum may change after journal replay, which is obviously unreasonable because we have successfully loaded and replayed the journal through the old s_journal_inum. And the new s_journal_inum bypasses some of the checks in ext4_get_journal(), which may trigger a null pointer dereference problem. So if s_journal_inum changes after the journal replay, we ignore the change, and rewrite the current journal_inum to the superblock. Link: https://bugzilla.kernel.org/show_bug.cgi?id=216541 Reported-by: Luís Henriques Signed-off-by: Baokun Li Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20230107032126.4165860-3-libaokun1@huawei.com Signed-off-by: Theodore Ts'o --- fs/ext4/super.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 3b9e30e1afd91..45bcfd35e5591 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -5953,8 +5953,11 @@ static int ext4_load_journal(struct super_block *sb, if (!really_read_only && journal_devnum && journal_devnum != le32_to_cpu(es->s_journal_dev)) { es->s_journal_dev = cpu_to_le32(journal_devnum); - - /* Make sure we flush the recovery flag to disk. */ + ext4_commit_super(sb); + } + if (!really_read_only && journal_inum && + journal_inum != le32_to_cpu(es->s_journal_inum)) { + es->s_journal_inum = cpu_to_le32(journal_inum); ext4_commit_super(sb); } From e6b9bd7290d334451ce054e98e752abc055e0034 Mon Sep 17 00:00:00 2001 From: Zhihao Cheng Date: Tue, 10 Jan 2023 09:53:27 +0800 Subject: [PATCH 327/765] jbd2: fix data missing when reusing bh which is ready to be checkpointed Following process will make data lost and could lead to a filesystem corrupted problem: 1. jh(bh) is inserted into T1->t_checkpoint_list, bh is dirty, and jh->b_transaction = NULL 2. T1 is added into journal->j_checkpoint_transactions. 3. Get bh prepare to write while doing checkpoing: PA PB do_get_write_access jbd2_log_do_checkpoint spin_lock(&jh->b_state_lock) if (buffer_dirty(bh)) clear_buffer_dirty(bh) // clear buffer dirty set_buffer_jbddirty(bh) transaction = journal->j_checkpoint_transactions jh = transaction->t_checkpoint_list if (!buffer_dirty(bh)) __jbd2_journal_remove_checkpoint(jh) // bh won't be flushed jbd2_cleanup_journal_tail __jbd2_journal_file_buffer(jh, transaction, BJ_Reserved) 4. Aborting journal/Power-cut before writing latest bh on journal area. In this way we get a corrupted filesystem with bh's data lost. Fix it by moving the clearing of buffer_dirty bit just before the call to __jbd2_journal_file_buffer(), both bit clearing and jh->b_transaction assignment are under journal->j_list_lock locked, so that jbd2_log_do_checkpoint() will wait until jh's new transaction fininshed even bh is currently not dirty. And journal_shrink_one_cp_list() won't remove jh from checkpoint list if the buffer head is reused in do_get_write_access(). Fetch a reproducer in [Link]. Link: https://bugzilla.kernel.org/show_bug.cgi?id=216898 Cc: Signed-off-by: Zhihao Cheng Signed-off-by: zhanchengbin Suggested-by: Jan Kara Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20230110015327.1181863-1-chengzhihao1@huawei.com Signed-off-by: Theodore Ts'o --- fs/jbd2/transaction.c | 50 +++++++++++++++++++++++++------------------ 1 file changed, 29 insertions(+), 21 deletions(-) diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index 6a404ac1c178f..15de1385012eb 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -1010,36 +1010,28 @@ do_get_write_access(handle_t *handle, struct journal_head *jh, * ie. locked but not dirty) or tune2fs (which may actually have * the buffer dirtied, ugh.) */ - if (buffer_dirty(bh)) { + if (buffer_dirty(bh) && jh->b_transaction) { + warn_dirty_buffer(bh); /* - * First question: is this buffer already part of the current - * transaction or the existing committing transaction? - */ - if (jh->b_transaction) { - J_ASSERT_JH(jh, - jh->b_transaction == transaction || - jh->b_transaction == - journal->j_committing_transaction); - if (jh->b_next_transaction) - J_ASSERT_JH(jh, jh->b_next_transaction == - transaction); - warn_dirty_buffer(bh); - } - /* - * In any case we need to clean the dirty flag and we must - * do it under the buffer lock to be sure we don't race - * with running write-out. + * We need to clean the dirty flag and we must do it under the + * buffer lock to be sure we don't race with running write-out. */ JBUFFER_TRACE(jh, "Journalling dirty buffer"); clear_buffer_dirty(bh); + /* + * The buffer is going to be added to BJ_Reserved list now and + * nothing guarantees jbd2_journal_dirty_metadata() will be + * ever called for it. So we need to set jbddirty bit here to + * make sure the buffer is dirtied and written out when the + * journaling machinery is done with it. + */ set_buffer_jbddirty(bh); } - unlock_buffer(bh); - error = -EROFS; if (is_handle_aborted(handle)) { spin_unlock(&jh->b_state_lock); + unlock_buffer(bh); goto out; } error = 0; @@ -1049,8 +1041,10 @@ do_get_write_access(handle_t *handle, struct journal_head *jh, * b_next_transaction points to it */ if (jh->b_transaction == transaction || - jh->b_next_transaction == transaction) + jh->b_next_transaction == transaction) { + unlock_buffer(bh); goto done; + } /* * this is the first time this transaction is touching this buffer, @@ -1074,10 +1068,24 @@ do_get_write_access(handle_t *handle, struct journal_head *jh, */ smp_wmb(); spin_lock(&journal->j_list_lock); + if (test_clear_buffer_dirty(bh)) { + /* + * Execute buffer dirty clearing and jh->b_transaction + * assignment under journal->j_list_lock locked to + * prevent bh being removed from checkpoint list if + * the buffer is in an intermediate state (not dirty + * and jh->b_transaction is NULL). + */ + JBUFFER_TRACE(jh, "Journalling dirty buffer"); + set_buffer_jbddirty(bh); + } __jbd2_journal_file_buffer(jh, transaction, BJ_Reserved); spin_unlock(&journal->j_list_lock); + unlock_buffer(bh); goto done; } + unlock_buffer(bh); + /* * If there is already a copy-out version of this buffer, then we don't * need to make another one From 0f7bfd6f8164be32dbbdf36aa1e5d00485c53cd7 Mon Sep 17 00:00:00 2001 From: Baokun Li Date: Tue, 10 Jan 2023 21:34:36 +0800 Subject: [PATCH 328/765] ext4: fix task hung in ext4_xattr_delete_inode Syzbot reported a hung task problem: ================================================================== INFO: task syz-executor232:5073 blocked for more than 143 seconds. Not tainted 6.2.0-rc2-syzkaller-00024-g512dee0c00ad #0 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. task:syz-exec232 state:D stack:21024 pid:5073 ppid:5072 flags:0x00004004 Call Trace: context_switch kernel/sched/core.c:5244 [inline] __schedule+0x995/0xe20 kernel/sched/core.c:6555 schedule+0xcb/0x190 kernel/sched/core.c:6631 __wait_on_freeing_inode fs/inode.c:2196 [inline] find_inode_fast+0x35a/0x4c0 fs/inode.c:950 iget_locked+0xb1/0x830 fs/inode.c:1273 __ext4_iget+0x22e/0x3ed0 fs/ext4/inode.c:4861 ext4_xattr_inode_iget+0x68/0x4e0 fs/ext4/xattr.c:389 ext4_xattr_inode_dec_ref_all+0x1a7/0xe50 fs/ext4/xattr.c:1148 ext4_xattr_delete_inode+0xb04/0xcd0 fs/ext4/xattr.c:2880 ext4_evict_inode+0xd7c/0x10b0 fs/ext4/inode.c:296 evict+0x2a4/0x620 fs/inode.c:664 ext4_orphan_cleanup+0xb60/0x1340 fs/ext4/orphan.c:474 __ext4_fill_super fs/ext4/super.c:5516 [inline] ext4_fill_super+0x81cd/0x8700 fs/ext4/super.c:5644 get_tree_bdev+0x400/0x620 fs/super.c:1282 vfs_get_tree+0x88/0x270 fs/super.c:1489 do_new_mount+0x289/0xad0 fs/namespace.c:3145 do_mount fs/namespace.c:3488 [inline] __do_sys_mount fs/namespace.c:3697 [inline] __se_sys_mount+0x2d3/0x3c0 fs/namespace.c:3674 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x3d/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd RIP: 0033:0x7fa5406fd5ea RSP: 002b:00007ffc7232f968 EFLAGS: 00000202 ORIG_RAX: 00000000000000a5 RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007fa5406fd5ea RDX: 0000000020000440 RSI: 0000000020000000 RDI: 00007ffc7232f970 RBP: 00007ffc7232f970 R08: 00007ffc7232f9b0 R09: 0000000000000432 R10: 0000000000804a03 R11: 0000000000000202 R12: 0000000000000004 R13: 0000555556a7a2c0 R14: 00007ffc7232f9b0 R15: 0000000000000000 ================================================================== The problem is that the inode contains an xattr entry with ea_inum of 15 when cleaning up an orphan inode <15>. When evict inode <15>, the reference counting of the corresponding EA inode is decreased. When EA inode <15> is found by find_inode_fast() in __ext4_iget(), it is found that the EA inode holds the I_FREEING flag and waits for the EA inode to complete deletion. As a result, when inode <15> is being deleted, we wait for inode <15> to complete the deletion, resulting in an infinite loop and triggering Hung Task. To solve this problem, we only need to check whether the ino of EA inode and parent is the same before getting EA inode. Link: https://syzkaller.appspot.com/bug?extid=77d6fcc37bbb92f26048 Reported-by: syzbot+77d6fcc37bbb92f26048@syzkaller.appspotmail.com Signed-off-by: Baokun Li Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20230110133436.996350-1-libaokun1@huawei.com Signed-off-by: Theodore Ts'o --- fs/ext4/xattr.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index d8fef540ca9bf..863c15388848a 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -422,6 +422,17 @@ static int ext4_xattr_inode_iget(struct inode *parent, unsigned long ea_ino, struct inode *inode; int err; + /* + * We have to check for this corruption early as otherwise + * iget_locked() could wait indefinitely for the state of our + * parent inode. + */ + if (parent->i_ino == ea_ino) { + ext4_error(parent->i_sb, + "Parent and EA inode have the same ino %lu", ea_ino); + return -EFSCORRUPTED; + } + inode = ext4_iget(parent->i_sb, ea_ino, EXT4_IGET_NORMAL); if (IS_ERR(inode)) { err = PTR_ERR(inode); From f7c843d6d7f82d26ee9efe689f9b3208dbf7ed87 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Fri, 23 Dec 2022 15:38:13 +0000 Subject: [PATCH 329/765] pwm: dwc: Change &pci->dev to dev in probe MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The dwc_pwm_probe() assigns dev to be &pci->dev but then uses &pci->dev throughout the function. Change these all to the 'dev' variable to make lines shorter. Signed-off-by: Ben Dooks Acked-by: Uwe Kleine-König Signed-off-by: Thierry Reding --- drivers/pwm/pwm-dwc.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/drivers/pwm/pwm-dwc.c b/drivers/pwm/pwm-dwc.c index bd2308812096d..de1c497d5b488 100644 --- a/drivers/pwm/pwm-dwc.c +++ b/drivers/pwm/pwm-dwc.c @@ -204,14 +204,13 @@ static int dwc_pwm_probe(struct pci_dev *pci, const struct pci_device_id *id) struct dwc_pwm *dwc; int ret; - dwc = devm_kzalloc(&pci->dev, sizeof(*dwc), GFP_KERNEL); + dwc = devm_kzalloc(dev, sizeof(*dwc), GFP_KERNEL); if (!dwc) return -ENOMEM; ret = pcim_enable_device(pci); if (ret) { - dev_err(&pci->dev, - "Failed to enable device (%pe)\n", ERR_PTR(ret)); + dev_err(dev, "Failed to enable device (%pe)\n", ERR_PTR(ret)); return ret; } @@ -219,14 +218,13 @@ static int dwc_pwm_probe(struct pci_dev *pci, const struct pci_device_id *id) ret = pcim_iomap_regions(pci, BIT(0), pci_name(pci)); if (ret) { - dev_err(&pci->dev, - "Failed to iomap PCI BAR (%pe)\n", ERR_PTR(ret)); + dev_err(dev, "Failed to iomap PCI BAR (%pe)\n", ERR_PTR(ret)); return ret; } dwc->base = pcim_iomap_table(pci)[0]; if (!dwc->base) { - dev_err(&pci->dev, "Base address missing\n"); + dev_err(dev, "Base address missing\n"); return -ENOMEM; } From a357d1493f0c66ce8006dd28c07646d1f891259a Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Fri, 23 Dec 2022 15:38:14 +0000 Subject: [PATCH 330/765] pwm: dwc: Move memory allocation to own function MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In preparation for adding other bus support, move the allocation of the PWM structure out of the main driver code. Signed-off-by: Ben Dooks Acked-by: Uwe Kleine-König Signed-off-by: Thierry Reding --- drivers/pwm/pwm-dwc.c | 24 +++++++++++++++++------- 1 file changed, 17 insertions(+), 7 deletions(-) diff --git a/drivers/pwm/pwm-dwc.c b/drivers/pwm/pwm-dwc.c index de1c497d5b488..e9399df702ed4 100644 --- a/drivers/pwm/pwm-dwc.c +++ b/drivers/pwm/pwm-dwc.c @@ -198,13 +198,29 @@ static const struct pwm_ops dwc_pwm_ops = { .owner = THIS_MODULE, }; +static struct dwc_pwm *dwc_pwm_alloc(struct device *dev) +{ + struct dwc_pwm *dwc; + + dwc = devm_kzalloc(dev, sizeof(*dwc), GFP_KERNEL); + if (!dwc) + return NULL; + + dwc->chip.dev = dev; + dwc->chip.ops = &dwc_pwm_ops; + dwc->chip.npwm = DWC_TIMERS_TOTAL; + + dev_set_drvdata(dev, dwc); + return dwc; +} + static int dwc_pwm_probe(struct pci_dev *pci, const struct pci_device_id *id) { struct device *dev = &pci->dev; struct dwc_pwm *dwc; int ret; - dwc = devm_kzalloc(dev, sizeof(*dwc), GFP_KERNEL); + dwc = dwc_pwm_alloc(dev); if (!dwc) return -ENOMEM; @@ -228,12 +244,6 @@ static int dwc_pwm_probe(struct pci_dev *pci, const struct pci_device_id *id) return -ENOMEM; } - pci_set_drvdata(pci, dwc); - - dwc->chip.dev = dev; - dwc->chip.ops = &dwc_pwm_ops; - dwc->chip.npwm = DWC_TIMERS_TOTAL; - ret = pwmchip_add(&dwc->chip); if (ret) return ret; From cf70d01a62c712ee715df1f7892b58c77474bcfb Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Fri, 23 Dec 2022 15:38:15 +0000 Subject: [PATCH 331/765] pwm: dwc: Use devm_pwmchip_add() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add the PWM chip using devm_pwmchip_add() to avoid having to manually remove it. This is useful for subsequent patches adding platform device support. Signed-off-by: Ben Dooks Reviewed-by: Uwe Kleine-König Signed-off-by: Thierry Reding --- drivers/pwm/pwm-dwc.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/drivers/pwm/pwm-dwc.c b/drivers/pwm/pwm-dwc.c index e9399df702ed4..3bbb26c862c35 100644 --- a/drivers/pwm/pwm-dwc.c +++ b/drivers/pwm/pwm-dwc.c @@ -244,7 +244,7 @@ static int dwc_pwm_probe(struct pci_dev *pci, const struct pci_device_id *id) return -ENOMEM; } - ret = pwmchip_add(&dwc->chip); + ret = devm_pwmchip_add(dev, &dwc->chip); if (ret) return ret; @@ -256,12 +256,8 @@ static int dwc_pwm_probe(struct pci_dev *pci, const struct pci_device_id *id) static void dwc_pwm_remove(struct pci_dev *pci) { - struct dwc_pwm *dwc = pci_get_drvdata(pci); - pm_runtime_forbid(&pci->dev); pm_runtime_get_noresume(&pci->dev); - - pwmchip_remove(&dwc->chip); } #ifdef CONFIG_PM_SLEEP From feabecaff5902f896531dde90646ca5dfa9d4f7d Mon Sep 17 00:00:00 2001 From: Sergey Shtylyov Date: Wed, 17 Aug 2022 23:00:45 +0300 Subject: [PATCH 332/765] genirq/ipi: Fix NULL pointer deref in irq_data_get_affinity_mask() If ipi_send_{mask|single}() is called with an invalid interrupt number, all the local variables there will be NULL. ipi_send_verify() which is invoked from these functions does verify its 'data' parameter, resulting in a kernel oops in irq_data_get_affinity_mask() as the passed NULL pointer gets dereferenced. Add a missing NULL pointer check in ipi_send_verify()... Found by Linux Verification Center (linuxtesting.org) with the SVACE static analysis tool. Fixes: 3b8e29a82dd1 ("genirq: Implement ipi_send_mask/single()") Signed-off-by: Sergey Shtylyov Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/r/b541232d-c2b6-1fe9-79b4-a7129459e4d0@omp.ru --- kernel/irq/ipi.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/kernel/irq/ipi.c b/kernel/irq/ipi.c index bbd945bacef08..961d4af76af37 100644 --- a/kernel/irq/ipi.c +++ b/kernel/irq/ipi.c @@ -188,9 +188,9 @@ EXPORT_SYMBOL_GPL(ipi_get_hwirq); static int ipi_send_verify(struct irq_chip *chip, struct irq_data *data, const struct cpumask *dest, unsigned int cpu) { - const struct cpumask *ipimask = irq_data_get_affinity_mask(data); + const struct cpumask *ipimask; - if (!chip || !ipimask) + if (!chip || !data) return -EINVAL; if (!chip->ipi_send_single && !chip->ipi_send_mask) @@ -199,6 +199,10 @@ static int ipi_send_verify(struct irq_chip *chip, struct irq_data *data, if (cpu >= nr_cpu_ids) return -EINVAL; + ipimask = irq_data_get_affinity_mask(data); + if (!ipimask) + return -EINVAL; + if (dest) { if (!cpumask_subset(dest, ipimask)) return -EINVAL; From 51c58a1ebc047acc4ac2a61a5f215c7a638a685b Mon Sep 17 00:00:00 2001 From: Kiseok Jo Date: Fri, 17 Feb 2023 00:44:03 +0000 Subject: [PATCH 333/765] ASoC: SMA1303: Change the value for right output This device can output mono, left or right. LR data should be swapped to output right data. Signed-off-by: Kiseok Jo Link: https://lore.kernel.org/r/20230217004403.10220-1-kiseok.jo@irondevice.com Signed-off-by: Mark Brown --- sound/soc/codecs/sma1303.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/codecs/sma1303.c b/sound/soc/codecs/sma1303.c index 727c01facf52d..fa4b0a60f8a91 100644 --- a/sound/soc/codecs/sma1303.c +++ b/sound/soc/codecs/sma1303.c @@ -569,7 +569,7 @@ static int sma1303_aif_in_event(struct snd_soc_dapm_widget *w, ret += sma1303_regmap_update_bits(sma1303, SMA1303_11_SYSTEM_CTRL2, SMA1303_LR_DATA_SW_MASK, - SMA1303_LR_DATA_SW_NORMAL, + SMA1303_LR_DATA_SW_SWAP, &temp); if (temp == true) change = true; From 70d1d30355095a22c8aa98dba9ca12486deff020 Mon Sep 17 00:00:00 2001 From: Trevor Wu Date: Wed, 15 Feb 2023 20:50:16 +0800 Subject: [PATCH 334/765] ASoC: mediatek: mt8188: correct etdm control return value In mt8188_etdm_clk_src_sel_put() function, val retrieved by FIELD_PREP is shifted to the corresponding bit filed, so it can compare with the register value directly. Originally, the redundant bit shift of the register value results in the wrong comparison result, so we remove bit shift operation in the patch. Fixes: 2babb4777489 ("ASoC: mediatek: mt8188: support etdm in platform driver") Signed-off-by: Trevor Wu Link: https://lore.kernel.org/r/20230215125017.16044-1-trevor.wu@mediatek.com Signed-off-by: Mark Brown --- sound/soc/mediatek/mt8188/mt8188-dai-etdm.c | 8 -------- 1 file changed, 8 deletions(-) diff --git a/sound/soc/mediatek/mt8188/mt8188-dai-etdm.c b/sound/soc/mediatek/mt8188/mt8188-dai-etdm.c index 071841903c620..1c53d4cb19bbe 100644 --- a/sound/soc/mediatek/mt8188/mt8188-dai-etdm.c +++ b/sound/soc/mediatek/mt8188/mt8188-dai-etdm.c @@ -679,7 +679,6 @@ static int mt8188_etdm_clk_src_sel_put(struct snd_kcontrol *kcontrol, unsigned int old_val; unsigned int mask; unsigned int reg; - unsigned int shift; if (source >= e->items) return -EINVAL; @@ -687,27 +686,22 @@ static int mt8188_etdm_clk_src_sel_put(struct snd_kcontrol *kcontrol, if (!strcmp(kcontrol->id.name, "ETDM_OUT1_Clock_Source")) { reg = ETDM_OUT1_CON4; mask = ETDM_OUT_CON4_CLOCK_MASK; - shift = ETDM_OUT_CON4_CLOCK_SHIFT; val = FIELD_PREP(ETDM_OUT_CON4_CLOCK_MASK, source); } else if (!strcmp(kcontrol->id.name, "ETDM_OUT2_Clock_Source")) { reg = ETDM_OUT2_CON4; mask = ETDM_OUT_CON4_CLOCK_MASK; - shift = ETDM_OUT_CON4_CLOCK_SHIFT; val = FIELD_PREP(ETDM_OUT_CON4_CLOCK_MASK, source); } else if (!strcmp(kcontrol->id.name, "ETDM_OUT3_Clock_Source")) { reg = ETDM_OUT3_CON4; mask = ETDM_OUT_CON4_CLOCK_MASK; - shift = ETDM_OUT_CON4_CLOCK_SHIFT; val = FIELD_PREP(ETDM_OUT_CON4_CLOCK_MASK, source); } else if (!strcmp(kcontrol->id.name, "ETDM_IN1_Clock_Source")) { reg = ETDM_IN1_CON2; mask = ETDM_IN_CON2_CLOCK_MASK; - shift = ETDM_IN_CON2_CLOCK_SHIFT; val = FIELD_PREP(ETDM_IN_CON2_CLOCK_MASK, source); } else if (!strcmp(kcontrol->id.name, "ETDM_IN2_Clock_Source")) { reg = ETDM_IN2_CON2; mask = ETDM_IN_CON2_CLOCK_MASK; - shift = ETDM_IN_CON2_CLOCK_SHIFT; val = FIELD_PREP(ETDM_IN_CON2_CLOCK_MASK, source); } else { return -EINVAL; @@ -715,8 +709,6 @@ static int mt8188_etdm_clk_src_sel_put(struct snd_kcontrol *kcontrol, regmap_read(afe->regmap, reg, &old_val); old_val &= mask; - old_val >>= shift; - if (old_val == val) return 0; From 7e43b75d6a062197b3bf39ddd1b10ce2e2d2a9b0 Mon Sep 17 00:00:00 2001 From: "Dharageswari.R" Date: Mon, 20 Feb 2023 10:06:52 +0200 Subject: [PATCH 335/765] ASoC: Intel: sof_rt5682: Add quirk for Rex board with mx98360a amplifier Add mtl_mx98360a_rt5682 driver data for Chrome Rex board support. Signed-off-by: Dharageswari.R Reviewed-by: Pierre-Louis Bossart Reviewed-by: Ranjani Sridharan Signed-off-by: Peter Ujfalusi Link: https://lore.kernel.org/r/20230220080652.23136-1-peter.ujfalusi@linux.intel.com Signed-off-by: Mark Brown --- sound/soc/intel/boards/sof_rt5682.c | 23 +++++++++++++++++++ .../intel/common/soc-acpi-intel-mtl-match.c | 12 ++++++++++ 2 files changed, 35 insertions(+) diff --git a/sound/soc/intel/boards/sof_rt5682.c b/sound/soc/intel/boards/sof_rt5682.c index 2eabc4b0fafa4..cda30f73eb52d 100644 --- a/sound/soc/intel/boards/sof_rt5682.c +++ b/sound/soc/intel/boards/sof_rt5682.c @@ -223,6 +223,20 @@ static const struct dmi_system_id sof_rt5682_quirk_table[] = { SOF_RT5682_SSP_AMP(2) | SOF_RT5682_NUM_HDMIDEV(4)), }, + { + .callback = sof_rt5682_quirk_cb, + .matches = { + DMI_MATCH(DMI_PRODUCT_FAMILY, "Google_Rex"), + DMI_MATCH(DMI_OEM_STRING, "AUDIO-MAX98360_ALC5682I_I2S"), + }, + .driver_data = (void *)(SOF_RT5682_MCLK_EN | + SOF_RT5682_SSP_CODEC(2) | + SOF_SPEAKER_AMP_PRESENT | + SOF_MAX98360A_SPEAKER_AMP_PRESENT | + SOF_RT5682_SSP_AMP(0) | + SOF_RT5682_NUM_HDMIDEV(4) + ), + }, { .callback = sof_rt5682_quirk_cb, .matches = { @@ -1104,6 +1118,15 @@ static const struct platform_device_id board_ids[] = { SOF_RT5682_SSP_AMP(1) | SOF_RT5682_NUM_HDMIDEV(4)), }, + { + .name = "mtl_mx98360_rt5682", + .driver_data = (kernel_ulong_t)(SOF_RT5682_MCLK_EN | + SOF_RT5682_SSP_CODEC(0) | + SOF_SPEAKER_AMP_PRESENT | + SOF_MAX98360A_SPEAKER_AMP_PRESENT | + SOF_RT5682_SSP_AMP(1) | + SOF_RT5682_NUM_HDMIDEV(4)), + }, { .name = "jsl_rt5682", .driver_data = (kernel_ulong_t)(SOF_RT5682_MCLK_EN | diff --git a/sound/soc/intel/common/soc-acpi-intel-mtl-match.c b/sound/soc/intel/common/soc-acpi-intel-mtl-match.c index b1a66a0f68181..7911c3af8071f 100644 --- a/sound/soc/intel/common/soc-acpi-intel-mtl-match.c +++ b/sound/soc/intel/common/soc-acpi-intel-mtl-match.c @@ -15,6 +15,11 @@ static const struct snd_soc_acpi_codecs mtl_max98357a_amp = { .codecs = {"MX98357A"} }; +static const struct snd_soc_acpi_codecs mtl_max98360a_amp = { + .num_codecs = 1, + .codecs = {"MX98360A"} +}; + static const struct snd_soc_acpi_codecs mtl_rt5682_rt5682s_hp = { .num_codecs = 2, .codecs = {"10EC5682", "RTL5682"}, @@ -28,6 +33,13 @@ struct snd_soc_acpi_mach snd_soc_acpi_intel_mtl_machines[] = { .quirk_data = &mtl_max98357a_amp, .sof_tplg_filename = "sof-mtl-max98357a-rt5682.tplg", }, + { + .comp_ids = &mtl_rt5682_rt5682s_hp, + .drv_name = "mtl_mx98360_rt5682", + .machine_quirk = snd_soc_acpi_codec_list, + .quirk_data = &mtl_max98360a_amp, + .sof_tplg_filename = "sof-mtl-max98360a-rt5682.tplg", + }, {}, }; EXPORT_SYMBOL_GPL(snd_soc_acpi_intel_mtl_machines); From d0566564d483e6576868224286632fd95aafd4ac Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Mon, 20 Feb 2023 18:56:14 +0800 Subject: [PATCH 336/765] regulator: max597x: Fix error return code in max597x_get_status REGULATOR_FAILED_RETRY should not be used in max597x_get_status error path. Othewise, the regulator core will treat it as REGULATOR_STATUS_ON. Signed-off-by: Axel Lin Link: https://lore.kernel.org/r/20230220105614.356187-1-axel.lin@ingics.com Signed-off-by: Mark Brown --- drivers/regulator/max597x-regulator.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/regulator/max597x-regulator.c b/drivers/regulator/max597x-regulator.c index f0fb0f56e4207..648e3641885a8 100644 --- a/drivers/regulator/max597x-regulator.c +++ b/drivers/regulator/max597x-regulator.c @@ -193,7 +193,7 @@ static int max597x_get_status(struct regulator_dev *rdev) ret = regmap_read(rdev->regmap, MAX5970_REG_STATUS3, &val); if (ret) - return REGULATOR_FAILED_RETRY; + return ret; if (val & MAX5970_STATUS3_ALERT) return REGULATOR_STATUS_ERROR; From b361d5d2464a88184f6e17a6462719ba79180b1a Mon Sep 17 00:00:00 2001 From: Janne Grunau Date: Sun, 12 Feb 2023 16:41:24 +0100 Subject: [PATCH 337/765] ASoC: dt-bindings: apple,mca: Add t8112-mca compatible MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The block found on Apple's M2 SoC is compatible with the existing driver so add its per-SoC compatible. Signed-off-by: Janne Grunau Acked-by: Krzysztof Kozlowski Acked-by: Martin Povišer Link: https://lore.kernel.org/r/20230202-asahi-t8112-dt-v1-14-cb5442d1c229@jannau.net Signed-off-by: Mark Brown --- Documentation/devicetree/bindings/sound/apple,mca.yaml | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/sound/apple,mca.yaml b/Documentation/devicetree/bindings/sound/apple,mca.yaml index 40e3a202f4434..5c6ec08c7d247 100644 --- a/Documentation/devicetree/bindings/sound/apple,mca.yaml +++ b/Documentation/devicetree/bindings/sound/apple,mca.yaml @@ -23,6 +23,7 @@ properties: - enum: - apple,t6000-mca - apple,t8103-mca + - apple,t8112-mca - const: apple,mca reg: From 0af2795f936f1ea1f9f1497447145dfcc7ed2823 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Mon, 20 Feb 2023 19:01:01 +0000 Subject: [PATCH 338/765] genirq/msi: Take the per-device MSI lock before validating the control structure Calling msi_ctrl_valid() ultimately results in calling msi_get_device_domain(), which requires holding the device MSI lock. However, in msi_domain_populate_irqs() the lock is taken right after having called msi_ctrl_valid(), which is just a tad too late. Take the lock before invoking msi_ctrl_valid(). Fixes: 40742716f294 ("genirq/msi: Make msi_add_simple_msi_descs() device domain aware") Reported-by: "Russell King (Oracle)" Signed-off-by: Marc Zyngier Signed-off-by: Thomas Gleixner Tested-by: Russell King (Oracle) Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/Y/Opu6ETe3ZzZ/8E@shell.armlinux.org.uk Link: https://lore.kernel.org/r/20230220190101.314446-1-maz@kernel.org --- kernel/irq/msi.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/kernel/irq/msi.c b/kernel/irq/msi.c index 783a3e6a0b107..13d96495e6d09 100644 --- a/kernel/irq/msi.c +++ b/kernel/irq/msi.c @@ -1084,10 +1084,13 @@ int msi_domain_populate_irqs(struct irq_domain *domain, struct device *dev, struct xarray *xa; int ret, virq; - if (!msi_ctrl_valid(dev, &ctrl)) - return -EINVAL; - msi_lock_descs(dev); + + if (!msi_ctrl_valid(dev, &ctrl)) { + ret = -EINVAL; + goto unlock; + } + ret = msi_domain_add_simple_msi_descs(dev, &ctrl); if (ret) goto unlock; From e6cc6f175566dd21a3f6e384c24593b1c751dd74 Mon Sep 17 00:00:00 2001 From: Reinette Chatre Date: Tue, 14 Feb 2023 13:13:20 -0800 Subject: [PATCH 339/765] PCI/MSI: Clarify usage of pci_msix_free_irq() pci_msix_free_irq() is used to free an interrupt on a PCI/MSI-X interrupt domain. The API description specifies that the interrupt to be freed was allocated via pci_msix_alloc_irq_at(). This description limits the usage of pci_msix_free_irq() since pci_msix_free_irq() can also be used to free MSI-X interrupts allocated with, for example, pci_alloc_irq_vectors(). Remove the text stating that the interrupt to be freed had to be allocated with pci_msix_alloc_irq_at(). The needed struct msi_map need not be from pci_msix_alloc_irq_at() but can be created from scratch using pci_irq_vector() to obtain the Linux IRQ number. Highlight that pci_msix_free_irq() cannot be used to disable MSI-X to guide users that, for example, pci_free_irq_vectors() remains to be needed. Signed-off-by: Reinette Chatre Signed-off-by: Thomas Gleixner Reviewed-by: Kevin Tian Link: https://lore.kernel.org/lkml/87r0xsd8j4.ffs@tglx Link: https://lore.kernel.org/r/4c3e7a50d6e70f408812cd7ab199c6b4b326f9de.1676408572.git.reinette.chatre@intel.com --- drivers/pci/msi/api.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/pci/msi/api.c b/drivers/pci/msi/api.c index b8009aa11f3ce..be679aa5db643 100644 --- a/drivers/pci/msi/api.c +++ b/drivers/pci/msi/api.c @@ -163,11 +163,11 @@ EXPORT_SYMBOL_GPL(pci_msix_alloc_irq_at); /** * pci_msix_free_irq - Free an interrupt on a PCI/MSIX interrupt domain - * which was allocated via pci_msix_alloc_irq_at() * * @dev: The PCI device to operate on * @map: A struct msi_map describing the interrupt to free - * as returned from the allocation function. + * + * Undo an interrupt vector allocation. Does not disable MSI-X. */ void pci_msix_free_irq(struct pci_dev *dev, struct msi_map map) { From ce7980ae9080f72f08d50355c4d9084d57aece63 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Fri, 17 Feb 2023 03:16:25 +0000 Subject: [PATCH 340/765] genirq/irqdesc: Make kobj_type structures constant MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since commit ee6d3dd4ed48 ("driver core: make kobj_type constant.") the driver core allows the usage of const struct kobj_type. Take advantage of this to constify the structure definitions which prevents modification at runtime. Signed-off-by: Thomas Weißschuh Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/r/20230217-kobj_type-irq-v1-1-fedfacaf8cdb@weissschuh.net --- kernel/irq/irqdesc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c index fd09962744014..240e145e969fc 100644 --- a/kernel/irq/irqdesc.c +++ b/kernel/irq/irqdesc.c @@ -277,7 +277,7 @@ static struct attribute *irq_attrs[] = { }; ATTRIBUTE_GROUPS(irq); -static struct kobj_type irq_kobj_type = { +static const struct kobj_type irq_kobj_type = { .release = irq_kobj_release, .sysfs_ops = &kobj_sysfs_ops, .default_groups = irq_groups, @@ -335,7 +335,7 @@ postcore_initcall(irq_sysfs_init); #else /* !CONFIG_SYSFS */ -static struct kobj_type irq_kobj_type = { +static const struct kobj_type irq_kobj_type = { .release = irq_kobj_release, }; From 9e0c7efa5ea231d85c0d41693a5115b3b971717c Mon Sep 17 00:00:00 2001 From: Juhyung Park Date: Fri, 3 Feb 2023 11:40:29 +0900 Subject: [PATCH 341/765] block: remove more NULL checks after bdev_get_queue() bdev_get_queue() never returns NULL. Several commits [1][2] have been made before to remove such superfluous checks, but some still remained. For places where bdev_get_queue() is called solely for NULL checks, it is removed entirely. [1] commit ec9fd2a13d74 ("blk-lib: don't check bdev_get_queue() NULL check") [2] commit fea127b36c93 ("block: remove superfluous check for request queue in bdev_is_zoned()") Signed-off-by: Juhyung Park Reviewed-by: Pankaj Raghav Link: https://lore.kernel.org/r/20230203024029.48260-1-qkrwngud825@gmail.com Signed-off-by: Jens Axboe --- block/blk-zoned.c | 10 ---------- include/linux/blkdev.h | 7 +------ kernel/trace/blktrace.c | 6 +----- 3 files changed, 2 insertions(+), 21 deletions(-) diff --git a/block/blk-zoned.c b/block/blk-zoned.c index 614b575be899b..fce9082384d65 100644 --- a/block/blk-zoned.c +++ b/block/blk-zoned.c @@ -334,17 +334,12 @@ int blkdev_report_zones_ioctl(struct block_device *bdev, fmode_t mode, { void __user *argp = (void __user *)arg; struct zone_report_args args; - struct request_queue *q; struct blk_zone_report rep; int ret; if (!argp) return -EINVAL; - q = bdev_get_queue(bdev); - if (!q) - return -ENXIO; - if (!bdev_is_zoned(bdev)) return -ENOTTY; @@ -391,7 +386,6 @@ int blkdev_zone_mgmt_ioctl(struct block_device *bdev, fmode_t mode, unsigned int cmd, unsigned long arg) { void __user *argp = (void __user *)arg; - struct request_queue *q; struct blk_zone_range zrange; enum req_op op; int ret; @@ -399,10 +393,6 @@ int blkdev_zone_mgmt_ioctl(struct block_device *bdev, fmode_t mode, if (!argp) return -EINVAL; - q = bdev_get_queue(bdev); - if (!q) - return -ENXIO; - if (!bdev_is_zoned(bdev)) return -ENOTTY; diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index b9637d63e6f02..89dd9b02b45b2 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -1276,12 +1276,7 @@ static inline bool bdev_nowait(struct block_device *bdev) static inline enum blk_zoned_model bdev_zoned_model(struct block_device *bdev) { - struct request_queue *q = bdev_get_queue(bdev); - - if (q) - return blk_queue_zoned_model(q); - - return BLK_ZONED_NONE; + return blk_queue_zoned_model(bdev_get_queue(bdev)); } static inline bool bdev_is_zoned(struct block_device *bdev) diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 5743be5594153..d5d94510afd3f 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -729,14 +729,10 @@ EXPORT_SYMBOL_GPL(blk_trace_startstop); **/ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg) { - struct request_queue *q; + struct request_queue *q = bdev_get_queue(bdev); int ret, start = 0; char b[BDEVNAME_SIZE]; - q = bdev_get_queue(bdev); - if (!q) - return -ENXIO; - mutex_lock(&q->debugfs_mutex); switch (cmd) { From 9c7c4bc986932218fd0df9d2a100509772028fb1 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Mon, 20 Feb 2023 12:14:13 +0800 Subject: [PATCH 342/765] ublk: remove check IO_URING_F_SQE128 in ublk_ch_uring_cmd sizeof(struct ublksrv_io_cmd) is 16bytes, which can be held in 64byte SQE, so not necessary to check IO_URING_F_SQE128. With this change, we get chance to save half SQ ring memory. Fixed: 71f28f3136af ("ublk_drv: add io_uring based userspace block driver") Signed-off-by: Ming Lei Link: https://lore.kernel.org/r/20230220041413.1524335-1-ming.lei@redhat.com Signed-off-by: Jens Axboe --- drivers/block/ublk_drv.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c index f48d213fb65e8..09d29fa53939c 100644 --- a/drivers/block/ublk_drv.c +++ b/drivers/block/ublk_drv.c @@ -1271,9 +1271,6 @@ static int ublk_ch_uring_cmd(struct io_uring_cmd *cmd, unsigned int issue_flags) __func__, cmd->cmd_op, ub_cmd->q_id, tag, ub_cmd->result); - if (!(issue_flags & IO_URING_F_SQE128)) - goto out; - if (ub_cmd->q_id >= ub->dev_info.nr_hw_queues) goto out; From 9b4f5028e493cb353a5c8f5c45073eeea0303abd Mon Sep 17 00:00:00 2001 From: Tomas Henzl Date: Thu, 2 Feb 2023 17:24:48 +0100 Subject: [PATCH 343/765] scsi: ses: Fix slab-out-of-bounds in ses_enclosure_data_process() A fix for: BUG: KASAN: slab-out-of-bounds in ses_enclosure_data_process+0x949/0xe30 [ses] Read of size 1 at addr ffff88a1b043a451 by task systemd-udevd/3271 Checking after (and before in next loop) addl_desc_ptr[1] is sufficient, we expect the size to be sanitized before first access to addl_desc_ptr[1]. Make sure we don't walk beyond end of page. Link: https://lore.kernel.org/r/20230202162451.15346-2-thenzl@redhat.com Cc: stable@vger.kernel.org Signed-off-by: Tomas Henzl Signed-off-by: Martin K. Petersen --- drivers/scsi/ses.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/ses.c b/drivers/scsi/ses.c index 869ca9c7f23f7..034d605c30765 100644 --- a/drivers/scsi/ses.c +++ b/drivers/scsi/ses.c @@ -609,9 +609,11 @@ static void ses_enclosure_data_process(struct enclosure_device *edev, /* these elements are optional */ type_ptr[0] == ENCLOSURE_COMPONENT_SCSI_TARGET_PORT || type_ptr[0] == ENCLOSURE_COMPONENT_SCSI_INITIATOR_PORT || - type_ptr[0] == ENCLOSURE_COMPONENT_CONTROLLER_ELECTRONICS)) + type_ptr[0] == ENCLOSURE_COMPONENT_CONTROLLER_ELECTRONICS)) { addl_desc_ptr += addl_desc_ptr[1] + 2; - + if (addl_desc_ptr + 1 >= ses_dev->page10 + ses_dev->page10_len) + addl_desc_ptr = NULL; + } } } kfree(buf); From db95d4df71cb55506425b6e4a5f8d68e3a765b63 Mon Sep 17 00:00:00 2001 From: Tomas Henzl Date: Thu, 2 Feb 2023 17:24:49 +0100 Subject: [PATCH 344/765] scsi: ses: Fix possible addl_desc_ptr out-of-bounds accesses Sanitize possible addl_desc_ptr out-of-bounds accesses in ses_enclosure_data_process(). Link: https://lore.kernel.org/r/20230202162451.15346-3-thenzl@redhat.com Cc: stable@vger.kernel.org Signed-off-by: Tomas Henzl Signed-off-by: Martin K. Petersen --- drivers/scsi/ses.c | 35 ++++++++++++++++++++++++++--------- 1 file changed, 26 insertions(+), 9 deletions(-) diff --git a/drivers/scsi/ses.c b/drivers/scsi/ses.c index 034d605c30765..458ca7abbbefa 100644 --- a/drivers/scsi/ses.c +++ b/drivers/scsi/ses.c @@ -439,8 +439,8 @@ int ses_match_host(struct enclosure_device *edev, void *data) } #endif /* 0 */ -static void ses_process_descriptor(struct enclosure_component *ecomp, - unsigned char *desc) +static int ses_process_descriptor(struct enclosure_component *ecomp, + unsigned char *desc, int max_desc_len) { int eip = desc[0] & 0x10; int invalid = desc[0] & 0x80; @@ -451,22 +451,32 @@ static void ses_process_descriptor(struct enclosure_component *ecomp, unsigned char *d; if (invalid) - return; + return 0; switch (proto) { case SCSI_PROTOCOL_FCP: if (eip) { + if (max_desc_len <= 7) + return 1; d = desc + 4; slot = d[3]; } break; case SCSI_PROTOCOL_SAS: + if (eip) { + if (max_desc_len <= 27) + return 1; d = desc + 4; slot = d[3]; d = desc + 8; - } else + } else { + if (max_desc_len <= 23) + return 1; d = desc + 4; + } + + /* only take the phy0 addr */ addr = (u64)d[12] << 56 | (u64)d[13] << 48 | @@ -483,6 +493,8 @@ static void ses_process_descriptor(struct enclosure_component *ecomp, } ecomp->slot = slot; scomp->addr = addr; + + return 0; } struct efd { @@ -555,7 +567,7 @@ static void ses_enclosure_data_process(struct enclosure_device *edev, /* skip past overall descriptor */ desc_ptr += len + 4; } - if (ses_dev->page10) + if (ses_dev->page10 && ses_dev->page10_len > 9) addl_desc_ptr = ses_dev->page10 + 8; type_ptr = ses_dev->page1_types; components = 0; @@ -563,6 +575,7 @@ static void ses_enclosure_data_process(struct enclosure_device *edev, for (j = 0; j < type_ptr[1]; j++) { char *name = NULL; struct enclosure_component *ecomp; + int max_desc_len; if (desc_ptr) { if (desc_ptr >= buf + page7_len) { @@ -589,10 +602,14 @@ static void ses_enclosure_data_process(struct enclosure_device *edev, ecomp = &edev->component[components++]; if (!IS_ERR(ecomp)) { - if (addl_desc_ptr) - ses_process_descriptor( - ecomp, - addl_desc_ptr); + if (addl_desc_ptr) { + max_desc_len = ses_dev->page10_len - + (addl_desc_ptr - ses_dev->page10); + if (ses_process_descriptor(ecomp, + addl_desc_ptr, + max_desc_len)) + addl_desc_ptr = NULL; + } if (create) enclosure_component_register( ecomp); From 801ab13d50cf3d26170ee073ea8bb4eececb76ab Mon Sep 17 00:00:00 2001 From: Tomas Henzl Date: Thu, 2 Feb 2023 17:24:50 +0100 Subject: [PATCH 345/765] scsi: ses: Fix possible desc_ptr out-of-bounds accesses Sanitize possible desc_ptr out-of-bounds accesses in ses_enclosure_data_process(). Link: https://lore.kernel.org/r/20230202162451.15346-4-thenzl@redhat.com Cc: stable@vger.kernel.org Signed-off-by: Tomas Henzl Signed-off-by: Martin K. Petersen --- drivers/scsi/ses.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/drivers/scsi/ses.c b/drivers/scsi/ses.c index 458ca7abbbefa..f8031d0782f7f 100644 --- a/drivers/scsi/ses.c +++ b/drivers/scsi/ses.c @@ -578,15 +578,19 @@ static void ses_enclosure_data_process(struct enclosure_device *edev, int max_desc_len; if (desc_ptr) { - if (desc_ptr >= buf + page7_len) { + if (desc_ptr + 3 >= buf + page7_len) { desc_ptr = NULL; } else { len = (desc_ptr[2] << 8) + desc_ptr[3]; desc_ptr += 4; - /* Add trailing zero - pushes into - * reserved space */ - desc_ptr[len] = '\0'; - name = desc_ptr; + if (desc_ptr + len > buf + page7_len) + desc_ptr = NULL; + else { + /* Add trailing zero - pushes into + * reserved space */ + desc_ptr[len] = '\0'; + name = desc_ptr; + } } } if (type_ptr[0] == ENCLOSURE_COMPONENT_DEVICE || From 578797f0c8cbc2e3ec5fc0dab87087b4c7073686 Mon Sep 17 00:00:00 2001 From: Tomas Henzl Date: Thu, 2 Feb 2023 17:24:51 +0100 Subject: [PATCH 346/765] scsi: ses: Fix slab-out-of-bounds in ses_intf_remove() A fix for: BUG: KASAN: slab-out-of-bounds in ses_intf_remove+0x23f/0x270 [ses] Read of size 8 at addr ffff88a10d32e5d8 by task rmmod/12013 When edev->components is zero, accessing edev->component[0] members is wrong. Link: https://lore.kernel.org/r/20230202162451.15346-5-thenzl@redhat.com Cc: stable@vger.kernel.org Signed-off-by: Tomas Henzl Signed-off-by: Martin K. Petersen --- drivers/scsi/ses.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/scsi/ses.c b/drivers/scsi/ses.c index f8031d0782f7f..9d4fb09acc1ec 100644 --- a/drivers/scsi/ses.c +++ b/drivers/scsi/ses.c @@ -856,7 +856,8 @@ static void ses_intf_remove_enclosure(struct scsi_device *sdev) kfree(ses_dev->page2); kfree(ses_dev); - kfree(edev->component[0].scratch); + if (edev->components) + kfree(edev->component[0].scratch); put_device(&edev->edev); enclosure_unregister(edev); From 3fe97ff3d94934649abb0652028dd7296170c8d0 Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Sat, 28 Nov 2020 15:27:21 -0800 Subject: [PATCH 347/765] scsi: ses: Don't attach if enclosure has no components An enclosure with no components can't usefully be operated by the driver (since effectively it has nothing to manage), so report the problem and don't attach. Not attaching also fixes an oops which could occur if the driver tries to manage a zero component enclosure. [mkp: Switched to KERN_WARNING since this scenario is common] Link: https://lore.kernel.org/r/c5deac044ac409e32d9ad9968ce0dcbc996bfc7a.camel@linux.ibm.com Cc: stable@vger.kernel.org Reported-by: Ding Hui Signed-off-by: James Bottomley Signed-off-by: Martin K. Petersen --- drivers/scsi/ses.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/scsi/ses.c b/drivers/scsi/ses.c index 9d4fb09acc1ec..b11a9162e73aa 100644 --- a/drivers/scsi/ses.c +++ b/drivers/scsi/ses.c @@ -733,6 +733,12 @@ static int ses_intf_add(struct device *cdev, type_ptr[0] == ENCLOSURE_COMPONENT_ARRAY_DEVICE) components += type_ptr[1]; } + + if (components == 0) { + sdev_printk(KERN_WARNING, sdev, "enclosure has no enumerated components\n"); + goto err_free; + } + ses_dev->page1 = buf; ses_dev->page1_len = len; buf = NULL; From 64fd2ba977b1b6b073bcf4c71906f820bc531421 Mon Sep 17 00:00:00 2001 From: Muneendra Date: Wed, 8 Feb 2023 19:43:26 -0800 Subject: [PATCH 348/765] scsi: scsi_transport_fc: Add an additional flag to fc_host_fpin_rcv() The LLDD and the stack currently process FPINs received from the fabric, but the stack is not aware of any action taken by the driver to alleviate congestion. The current interface between the driver and the SCSI stack is limited to passing the notification mainly for statistics and heuristics. The reaction to an FPIN could be handled either by the driver or by the stack (marginal path and failover). Amend the interface to indicate if action on an FPIN has already been reacted to by the LLDDs or not. Add an additional flag to fc_host_fpin_rcv() to indicate if the FPIN has been acknowledged/reacted to by the driver. Also added a new event code FCH_EVT_LINK_FPIN_ACK to notify to the user that the event has been acknowledged/reacted by the LLDD driver Link: https://lore.kernel.org/r/20230209034326.882514-1-muneendra.kumar@broadcom.com Co-developed-by: Anil Gurumurthy Signed-off-by: Anil Gurumurthy Co-developed-by: Nilesh Javali Signed-off-by: Nilesh Javali Signed-off-by: Muneendra Reviewed-by: James Smart Reviewed-by: Himanshu Madhani Reviewed-by: Ewan D. Milne Signed-off-by: Martin K. Petersen --- drivers/scsi/lpfc/lpfc_els.c | 2 +- drivers/scsi/qla2xxx/qla_isr.c | 2 +- drivers/scsi/scsi_transport_fc.c | 10 +++++++--- include/scsi/scsi_transport_fc.h | 4 +++- 4 files changed, 12 insertions(+), 6 deletions(-) diff --git a/drivers/scsi/lpfc/lpfc_els.c b/drivers/scsi/lpfc/lpfc_els.c index 569639dc8b2c8..aee5d0d1187d6 100644 --- a/drivers/scsi/lpfc/lpfc_els.c +++ b/drivers/scsi/lpfc/lpfc_els.c @@ -10287,7 +10287,7 @@ lpfc_els_rcv_fpin(struct lpfc_vport *vport, void *p, u32 fpin_length) /* Send every descriptor individually to the upper layer */ if (deliver) fc_host_fpin_rcv(lpfc_shost_from_vport(vport), - fpin_length, (char *)fpin); + fpin_length, (char *)fpin, 0); desc_cnt++; } } diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c index 46e8b38603f04..030625ebb4e65 100644 --- a/drivers/scsi/qla2xxx/qla_isr.c +++ b/drivers/scsi/qla2xxx/qla_isr.c @@ -45,7 +45,7 @@ qla27xx_process_purex_fpin(struct scsi_qla_host *vha, struct purex_item *item) ql_dump_buffer(ql_dbg_init + ql_dbg_verbose, vha, 0x508f, pkt, pkt_size); - fc_host_fpin_rcv(vha->host, pkt_size, (char *)pkt); + fc_host_fpin_rcv(vha->host, pkt_size, (char *)pkt, 0); } const char *const port_state_str[] = { diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c index 0965f8a7134f0..f12e9467ebb43 100644 --- a/drivers/scsi/scsi_transport_fc.c +++ b/drivers/scsi/scsi_transport_fc.c @@ -137,6 +137,7 @@ static const struct { { FCH_EVT_PORT_FABRIC, "port_fabric" }, { FCH_EVT_LINK_UNKNOWN, "link_unknown" }, { FCH_EVT_LINK_FPIN, "link_FPIN" }, + { FCH_EVT_LINK_FPIN_ACK, "link_FPIN_ACK" }, { FCH_EVT_VENDOR_UNIQUE, "vendor_unique" }, }; fc_enum_name_search(host_event_code, fc_host_event_code, @@ -894,17 +895,20 @@ fc_fpin_congn_stats_update(struct Scsi_Host *shost, * @shost: host the FPIN was received on * @fpin_len: length of FPIN payload, in bytes * @fpin_buf: pointer to FPIN payload - * + * @event_acknowledge: 1, if LLDD handles this event. * Notes: * This routine assumes no locks are held on entry. */ void -fc_host_fpin_rcv(struct Scsi_Host *shost, u32 fpin_len, char *fpin_buf) +fc_host_fpin_rcv(struct Scsi_Host *shost, u32 fpin_len, char *fpin_buf, + u8 event_acknowledge) { struct fc_els_fpin *fpin = (struct fc_els_fpin *)fpin_buf; struct fc_tlv_desc *tlv; u32 desc_cnt = 0, bytes_remain; u32 dtag; + enum fc_host_event_code event_code = + event_acknowledge ? FCH_EVT_LINK_FPIN_ACK : FCH_EVT_LINK_FPIN; /* Update Statistics */ tlv = (struct fc_tlv_desc *)&fpin->fpin_desc[0]; @@ -934,7 +938,7 @@ fc_host_fpin_rcv(struct Scsi_Host *shost, u32 fpin_len, char *fpin_buf) } fc_host_post_fc_event(shost, fc_get_event_number(), - FCH_EVT_LINK_FPIN, fpin_len, fpin_buf, 0); + event_code, fpin_len, fpin_buf, 0); } EXPORT_SYMBOL(fc_host_fpin_rcv); diff --git a/include/scsi/scsi_transport_fc.h b/include/scsi/scsi_transport_fc.h index 3dcda19d35202..483513c575976 100644 --- a/include/scsi/scsi_transport_fc.h +++ b/include/scsi/scsi_transport_fc.h @@ -496,6 +496,7 @@ enum fc_host_event_code { FCH_EVT_PORT_FABRIC = 0x204, FCH_EVT_LINK_UNKNOWN = 0x500, FCH_EVT_LINK_FPIN = 0x501, + FCH_EVT_LINK_FPIN_ACK = 0x502, FCH_EVT_VENDOR_UNIQUE = 0xffff, }; @@ -856,7 +857,8 @@ void fc_host_post_fc_event(struct Scsi_Host *shost, u32 event_number, * Note: when calling fc_host_post_fc_event(), vendor_id may be * specified as 0. */ -void fc_host_fpin_rcv(struct Scsi_Host *shost, u32 fpin_len, char *fpin_buf); +void fc_host_fpin_rcv(struct Scsi_Host *shost, u32 fpin_len, char *fpin_buf, + u8 event_acknowledge); struct fc_vport *fc_vport_create(struct Scsi_Host *shost, int channel, struct fc_vport_identifiers *); int fc_vport_terminate(struct fc_vport *vport); From 473025a6b6f8a4a8de3120c78588d32cf4ba7324 Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Thu, 9 Feb 2023 10:49:03 -0800 Subject: [PATCH 349/765] scsi: ufs: Make the TC G210 driver dependent on CONFIG_OF MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The TC G210 driver only supports devices declared in the device tree. Hence make this driver dependent on CONFIG_OF. This patch fixes the following compiler error: drivers/ufs/host/tc-dwc-g210-pltfrm.c:36:34: error: ‘tc_dwc_g210_pltfm_match’ defined but not used [-Werror=unused-const-variable=] 36 | static const struct of_device_id tc_dwc_g210_pltfm_match[] = { | Link: https://lore.kernel.org/r/20230209184914.2762172-1-bvanassche@acm.org Cc: Joao Pinto Signed-off-by: Bart Van Assche Reviewed-by: Bean Huo Signed-off-by: Martin K. Petersen --- drivers/ufs/host/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ufs/host/Kconfig b/drivers/ufs/host/Kconfig index 6638814379218..8793e34335806 100644 --- a/drivers/ufs/host/Kconfig +++ b/drivers/ufs/host/Kconfig @@ -48,7 +48,7 @@ config SCSI_UFS_CDNS_PLATFORM config SCSI_UFS_DWC_TC_PLATFORM tristate "DesignWare platform support using a G210 Test Chip" - depends on SCSI_UFSHCD_PLATFORM + depends on OF && SCSI_UFSHCD_PLATFORM help Synopsys Test Chip is a PHY for prototyping purposes. From 7dafc3e007918384c8693ff8d70381b5c1e9c247 Mon Sep 17 00:00:00 2001 From: Adrien Thierry Date: Fri, 17 Feb 2023 14:44:22 -0500 Subject: [PATCH 350/765] scsi: ufs: core: Initialize devfreq synchronously During UFS initialization, devfreq initialization is asynchronous: ufshcd_async_scan() calls ufshcd_add_lus(), which in turn initializes devfreq for UFS. The simple ondemand governor is then loaded. If it is built as a module, request_module() is called and throws a warning: WARNING: CPU: 7 PID: 167 at kernel/kmod.c:136 __request_module+0x1e0/0x460 Modules linked in: crct10dif_ce llcc_qcom phy_qcom_qmp_usb ufs_qcom phy_qcom_snps_femto_v2 ufshcd_pltfrm phy_qcom_qmp_combo ufshcd_core phy_qcom_qmp_ufs qcom_wdt socinfo fuse ipv6 CPU: 7 PID: 167 Comm: kworker/u16:3 Not tainted 6.2.0-rc6-00009-g58706f7fb045 #1 Hardware name: Qualcomm SA8540P Ride (DT) Workqueue: events_unbound async_run_entry_fn pstate: 00400005 (nzcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : __request_module+0x1e0/0x460 lr : __request_module+0x1d8/0x460 sp : ffff800009323b90 x29: ffff800009323b90 x28: 0000000000000000 x27: 0000000000000000 x26: ffff800009323d50 x25: ffff7b9045f57810 x24: ffff7b9045f57830 x23: ffffdc5a83e426e8 x22: ffffdc5ae80a9818 x21: 0000000000000001 x20: ffffdc5ae7502f98 x19: ffff7b9045f57800 x18: ffffffffffffffff x17: 312f716572667665 x16: 642f7366752e3030 x15: 0000000000000000 x14: 000000000000021c x13: 0000000000005400 x12: ffff7b9042ed7614 x11: ffff7b9042ed7600 x10: 00000000636c0890 x9 : 0000000000000038 x8 : ffff7b9045f2c880 x7 : ffff7b9045f57c68 x6 : 0000000000000080 x5 : 0000000000000000 x4 : 8000000000000000 x3 : 0000000000000000 x2 : 0000000000000000 x1 : ffffdc5ae5d382f0 x0 : 0000000000000001 Call trace: __request_module+0x1e0/0x460 try_then_request_governor+0x7c/0x100 devfreq_add_device+0x4b0/0x5fc ufshcd_async_scan+0x1d4/0x310 [ufshcd_core] async_run_entry_fn+0x34/0xe0 process_one_work+0x1d0/0x320 worker_thread+0x14c/0x444 kthread+0x10c/0x110 ret_from_fork+0x10/0x20 This occurs because synchronous module loading from async is not allowed. According to __request_module(): /* * We don't allow synchronous module loading from async. Module * init may invoke async_synchronize_full() which will end up * waiting for this task which already is waiting for the module * loading to complete, leading to a deadlock. */ Such a deadlock was experienced on the Qualcomm QDrive3/sa8540p-ride. With DEVFREQ_GOV_SIMPLE_ONDEMAND=m, the boot hangs after the warning. Fix both the warning and the deadlock by moving devfreq initialization out of the async routine. Tested on the sa8540p-ride by using fio to put the UFS under load, and printing the trace generated by /sys/kernel/tracing/events/ufs/ufshcd_clk_scaling events. The trace looks similar with and without the change. Link: https://lore.kernel.org/r/20230217194423.42553-1-athierry@redhat.com Signed-off-by: Adrien Thierry Reviewed-by: Bart Van Assche Signed-off-by: Martin K. Petersen --- drivers/ufs/core/ufshcd.c | 47 ++++++++++++++++++++++++++------------- include/ufs/ufshcd.h | 1 + 2 files changed, 32 insertions(+), 16 deletions(-) diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c index 3b3cf78d3b100..0baeec4ca304f 100644 --- a/drivers/ufs/core/ufshcd.c +++ b/drivers/ufs/core/ufshcd.c @@ -1411,6 +1411,13 @@ static int ufshcd_devfreq_target(struct device *dev, struct ufs_clk_info *clki; unsigned long irq_flags; + /* + * Skip devfreq if UFS initialization is not finished. + * Otherwise ufs could be in a inconsistent state. + */ + if (!smp_load_acquire(&hba->logical_unit_scan_finished)) + return 0; + if (!ufshcd_is_clkscaling_supported(hba)) return -EINVAL; @@ -8384,22 +8391,6 @@ static int ufshcd_add_lus(struct ufs_hba *hba) if (ret) goto out; - /* Initialize devfreq after UFS device is detected */ - if (ufshcd_is_clkscaling_supported(hba)) { - memcpy(&hba->clk_scaling.saved_pwr_info.info, - &hba->pwr_info, - sizeof(struct ufs_pa_layer_attr)); - hba->clk_scaling.saved_pwr_info.is_valid = true; - hba->clk_scaling.is_allowed = true; - - ret = ufshcd_devfreq_init(hba); - if (ret) - goto out; - - hba->clk_scaling.is_enabled = true; - ufshcd_init_clk_scaling_sysfs(hba); - } - ufs_bsg_probe(hba); ufshpb_init(hba); scsi_scan_host(hba->host); @@ -8669,6 +8660,12 @@ static void ufshcd_async_scan(void *data, async_cookie_t cookie) if (ret) { pm_runtime_put_sync(hba->dev); ufshcd_hba_exit(hba); + } else { + /* + * Make sure that when reader code sees UFS initialization has finished, + * all initialization steps have really been executed. + */ + smp_store_release(&hba->logical_unit_scan_finished, true); } } @@ -10309,12 +10306,30 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem *mmio_base, unsigned int irq) */ ufshcd_set_ufs_dev_active(hba); + /* Initialize devfreq */ + if (ufshcd_is_clkscaling_supported(hba)) { + memcpy(&hba->clk_scaling.saved_pwr_info.info, + &hba->pwr_info, + sizeof(struct ufs_pa_layer_attr)); + hba->clk_scaling.saved_pwr_info.is_valid = true; + hba->clk_scaling.is_allowed = true; + + err = ufshcd_devfreq_init(hba); + if (err) + goto rpm_put_sync; + + hba->clk_scaling.is_enabled = true; + ufshcd_init_clk_scaling_sysfs(hba); + } + async_schedule(ufshcd_async_scan, hba); ufs_sysfs_add_nodes(hba->dev); device_enable_async_suspend(dev); return 0; +rpm_put_sync: + pm_runtime_put_sync(dev); free_tmf_queue: blk_mq_destroy_queue(hba->tmf_queue); blk_put_queue(hba->tmf_queue); diff --git a/include/ufs/ufshcd.h b/include/ufs/ufshcd.h index ed9e3d5addb3a..05e416414e41d 100644 --- a/include/ufs/ufshcd.h +++ b/include/ufs/ufshcd.h @@ -978,6 +978,7 @@ struct ufs_hba { struct completion *uic_async_done; enum ufshcd_state ufshcd_state; + bool logical_unit_scan_finished; u32 eh_flags; u32 intr_mask; u16 ee_ctrl_mask; From 19873b03f1daeffd9d5e089b855dd926975ab5b7 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Mon, 20 Feb 2023 22:24:31 +0800 Subject: [PATCH 351/765] scsi: ufs: ufs-mediatek: Guard power management functions with CONFIG_PM Fix the following compilation error when CONFIG_PM is set to 'n': drivers/ufs/host/ufs-mediatek.c: In function `ufs_mtk_runtime_suspend`: drivers/ufs/host/ufs-mediatek.c:1623:8: error: implicit declaration of function `ufshcd_runtime_suspend`; did you mean `ufs_mtk_runtime_suspend`? [-Werror=implicit-function-declaration] 1623 | ret = ufshcd_runtime_suspend(dev); | ^~~~~~~~~~~~~~~~~~~~~~ | ufs_mtk_runtime_suspend drivers/ufs/host/ufs-mediatek.c: In function `ufs_mtk_runtime_resume`: drivers/ufs/host/ufs-mediatek.c:1638:9: error: implicit declaration of function `ufshcd_runtime_resume`; did you mean `ufs_mtk_runtime_resume`? [-Werror=implicit-function-declaration] 1638 | return ufshcd_runtime_resume(dev); | ^~~~~~~~~~~~~~~~~~~~~ | ufs_mtk_runtime_resume At top level: drivers/ufs/host/ufs-mediatek.c:1632:12: error: `ufs_mtk_runtime_resume` defined but not used [-Werror=unused-function] 1632 | static int ufs_mtk_runtime_resume(struct device *dev) | ^~~~~~~~~~~~~~~~~~~~~~ drivers/ufs/host/ufs-mediatek.c:1618:12: error: `ufs_mtk_runtime_suspend` defined but not used [-Werror=unused-function] 1618 | static int ufs_mtk_runtime_suspend(struct device *dev) Link: https://lore.kernel.org/r/20230220142431.54589-1-frank.li@vivo.com Reported-by: k2ci Reported-by: Shida Zhang Signed-off-by: Yangtao Li Reviewed-by: Stanley Chu Signed-off-by: Martin K. Petersen --- drivers/ufs/host/ufs-mediatek.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/ufs/host/ufs-mediatek.c b/drivers/ufs/host/ufs-mediatek.c index 21d9b047539fb..73e217260390e 100644 --- a/drivers/ufs/host/ufs-mediatek.c +++ b/drivers/ufs/host/ufs-mediatek.c @@ -1613,6 +1613,7 @@ static int ufs_mtk_system_resume(struct device *dev) } #endif +#ifdef CONFIG_PM static int ufs_mtk_runtime_suspend(struct device *dev) { struct ufs_hba *hba = dev_get_drvdata(dev); @@ -1635,6 +1636,7 @@ static int ufs_mtk_runtime_resume(struct device *dev) return ufshcd_runtime_resume(dev); } +#endif static const struct dev_pm_ops ufs_mtk_pm_ops = { SET_SYSTEM_SLEEP_PM_OPS(ufs_mtk_system_suspend, From 2076f57f2ca0a2034afac7832257b2bc509c1a87 Mon Sep 17 00:00:00 2001 From: Asutosh Das Date: Thu, 16 Feb 2023 09:13:46 -0800 Subject: [PATCH 352/765] scsi: ufs: mcq: Fix incorrectly set queue depth ufshcd_config_mcq() may change the can_queue value. The current code invokes scsi_add_host() before ufshcd_config_mcq() so the tags are limited to the original can_queue value. Fix this by invoking scsi_add_host() after ufshcd_config_mcq(). Link: https://lore.kernel.org/r/8840cea4a57b46dabce18acc39afc50ab826330f.1676567593.git.quic_asutoshd@quicinc.com Fixes: 2468da61ea09 ("scsi: ufs: core: mcq: Configure operation and runtime interface") Signed-off-by: Asutosh Das Reviewed-by: Bart Van Assche Reviewed-by: Manivannan Sadhasivam Signed-off-by: Martin K. Petersen --- drivers/ufs/core/ufshcd.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c index 0baeec4ca304f..47c7739b9a868 100644 --- a/drivers/ufs/core/ufshcd.c +++ b/drivers/ufs/core/ufshcd.c @@ -8521,7 +8521,9 @@ static int ufshcd_device_init(struct ufs_hba *hba, bool init_dev_params) return ret; if (is_mcq_supported(hba) && !hba->scsi_host_added) { ret = ufshcd_alloc_mcq(hba); - if (ret) { + if (!ret) { + ufshcd_config_mcq(hba); + } else { /* Continue with SDB mode */ use_mcq_mode = false; dev_err(hba->dev, "MCQ mode is disabled, err=%d\n", @@ -8533,10 +8535,10 @@ static int ufshcd_device_init(struct ufs_hba *hba, bool init_dev_params) return ret; } hba->scsi_host_added = true; - } - /* MCQ may be disabled if ufshcd_alloc_mcq() fails */ - if (is_mcq_supported(hba) && use_mcq_mode) + } else if (is_mcq_supported(hba)) { + /* UFSHCD_QUIRK_REINIT_AFTER_MAX_GEAR_SWITCH is set */ ufshcd_config_mcq(hba); + } } ufshcd_tune_unipro_params(hba); From 2b3795167e233dc9974f9f8239afb920f5cd4a32 Mon Sep 17 00:00:00 2001 From: Fengnan Chang Date: Tue, 21 Feb 2023 19:53:40 +0800 Subject: [PATCH 353/765] scsi: sd: Remove unused sd_cdb_cache Since commit ce70fd9a551a ("scsi: core: Remove the cmd field from struct scsi_request") sd_cdb_cache is unused. Remove it. Link: https://lore.kernel.org/r/20230221115340.21201-1-changfengnan@bytedance.com Signed-off-by: Fengnan Chang Reviewed-by: John Garry Reviewed-by: Christoph Hellwig Signed-off-by: Martin K. Petersen --- drivers/scsi/sd.c | 16 +--------------- 1 file changed, 1 insertion(+), 15 deletions(-) diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index 2aa3b0393b96a..a9f7e019e0733 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -121,7 +121,6 @@ static void scsi_disk_release(struct device *cdev); static DEFINE_IDA(sd_index_ida); -static struct kmem_cache *sd_cdb_cache; static mempool_t *sd_page_pool; static struct lock_class_key sd_bio_compl_lkclass; @@ -3853,19 +3852,11 @@ static int __init init_sd(void) if (err) goto err_out; - sd_cdb_cache = kmem_cache_create("sd_ext_cdb", SD_EXT_CDB_SIZE, - 0, 0, NULL); - if (!sd_cdb_cache) { - printk(KERN_ERR "sd: can't init extended cdb cache\n"); - err = -ENOMEM; - goto err_out_class; - } - sd_page_pool = mempool_create_page_pool(SD_MEMPOOL_SIZE, 0); if (!sd_page_pool) { printk(KERN_ERR "sd: can't init discard page pool\n"); err = -ENOMEM; - goto err_out_cache; + goto err_out_class; } err = scsi_register_driver(&sd_template.gendrv); @@ -3876,10 +3867,6 @@ static int __init init_sd(void) err_out_driver: mempool_destroy(sd_page_pool); - -err_out_cache: - kmem_cache_destroy(sd_cdb_cache); - err_out_class: class_unregister(&sd_disk_class); err_out: @@ -3901,7 +3888,6 @@ static void __exit exit_sd(void) scsi_unregister_driver(&sd_template.gendrv); mempool_destroy(sd_page_pool); - kmem_cache_destroy(sd_cdb_cache); class_unregister(&sd_disk_class); From 26a02d972bad946ecbaa12131736c8b6d28893f8 Mon Sep 17 00:00:00 2001 From: Xingui Yang Date: Tue, 21 Feb 2023 08:10:26 +0000 Subject: [PATCH 354/765] scsi: sd: Update DIX config every time sd_revalidate_disk() is called If a controller has DIX is enabled and an attached disk is formatted using a protection type supported by the controller, a block integrity profile is registered to enable protected transfers. If the disk is subsequently reformatted to disable PI, and the controller does not support DIX Type 0, this can lead to failures such as this: [142829.032340] hisi_sas_v3_hw 0000:b4:04.0: erroneous completion iptt=2375 task=00000000bea0970c dev id=5 direct-attached phy4 addr=51c20dbaf642a000 CQ hdr: 0x1023 0x50947 0x0 0x20000 Error info: 0x0 0x0 0x4 0x0 [142829.073883] sas: Enter sas_scsi_recover_host busy: 1 failed: 1 [142829.079783] sas: sas_scsi_find_task: aborting task 0x00000000bea0970c [142829.102342] sas: Internal abort: task to dev 51c20dbaf642a000 response: 0x0 status 0x5 [142829.110319] sas: sas_eh_handle_sas_errors: task 0x00000000bea0970c is done [142829.117275] sd 7:0:5:0: [sdc] tag#2375 UNKNOWN(0x2003) Result: hostbyte=0x05 driverbyte=DRIVER_OK cmd_age=0s [142829.127171] sd 7:0:5:0: [sdc] tag#2375 CDB: opcode=0x2a 2a 00 00 00 00 00 00 00 08 00 [142829.135059] I/O error, dev sdc, sector 0 op 0x1:(WRITE) flags 0x18800 phys_seg 1 prio class 2 This is because the block layer integrity profile is currently only set up the first time a disk is discovered. To address this, remove the first_scan check when configuring protection information during revalidate. Also unregister the block integrity profile if DIX is not supported with a given protection type. [mkp: commit description + printk dedup] Link: https://lore.kernel.org/r/20230221081026.24736-1-yangxingui@huawei.com Signed-off-by: Xingui Yang Signed-off-by: Martin K. Petersen --- drivers/scsi/sd.c | 13 +++++-------- drivers/scsi/sd_dif.c | 10 ++++++---- 2 files changed, 11 insertions(+), 12 deletions(-) diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index a9f7e019e0733..2f9b7c8f6bd0b 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -2253,23 +2253,20 @@ static void sd_config_protection(struct scsi_disk *sdkp) { struct scsi_device *sdp = sdkp->device; - if (!sdkp->first_scan) - return; - sd_dif_config_host(sdkp); if (!sdkp->protection_type) return; if (!scsi_host_dif_capable(sdp->host, sdkp->protection_type)) { - sd_printk(KERN_NOTICE, sdkp, - "Disabling DIF Type %u protection\n", - sdkp->protection_type); + sd_first_printk(KERN_NOTICE, sdkp, + "Disabling DIF Type %u protection\n", + sdkp->protection_type); sdkp->protection_type = 0; } - sd_printk(KERN_NOTICE, sdkp, "Enabling DIF Type %u protection\n", - sdkp->protection_type); + sd_first_printk(KERN_NOTICE, sdkp, "Enabling DIF Type %u protection\n", + sdkp->protection_type); } static void read_capacity_error(struct scsi_disk *sdkp, struct scsi_device *sdp, diff --git a/drivers/scsi/sd_dif.c b/drivers/scsi/sd_dif.c index 968993ee6d5d3..1df847b5f7476 100644 --- a/drivers/scsi/sd_dif.c +++ b/drivers/scsi/sd_dif.c @@ -39,8 +39,10 @@ void sd_dif_config_host(struct scsi_disk *sdkp) dif = 0; dix = 1; } - if (!dix) + if (!dix) { + blk_integrity_unregister(disk); return; + } memset(&bi, 0, sizeof(bi)); @@ -72,9 +74,9 @@ void sd_dif_config_host(struct scsi_disk *sdkp) bi.tag_size = sizeof(u16); } - sd_printk(KERN_NOTICE, sdkp, - "Enabling DIX %s, application tag size %u bytes\n", - bi.profile->name, bi.tag_size); + sd_first_printk(KERN_NOTICE, sdkp, + "Enabling DIX %s, application tag size %u bytes\n", + bi.profile->name, bi.tag_size); out: blk_integrity_register(disk, &bi); } From 3a2d1efaf344c7351945532020774c3fcb134a0a Mon Sep 17 00:00:00 2001 From: Deepak R Varma Date: Sun, 8 Jan 2023 21:58:51 +0530 Subject: [PATCH 355/765] scsi: ipr: Make ipr_probe_ioa_part2() return void Convert function ipr_probe_ioa_part2() to return void instead of int since the current implementation always returns 0 to the caller. The transformation also eliminates the dead code when calling ipr_probe_ioa_part2() function. Issue identified using returnvar Coccinelle semantic patch. Link: https://lore.kernel.org/r/Y7rvQyMOGcPKPTv8@ubun2204.myguest.virtualbox.org Signed-off-by: Deepak R Varma Acked-by: Brian King Signed-off-by: Martin K. Petersen --- drivers/scsi/ipr.c | 13 +++---------- 1 file changed, 3 insertions(+), 10 deletions(-) diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index 7dff517a08586..5feffda0ef68f 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -9500,11 +9500,10 @@ static pci_ers_result_t ipr_pci_error_detected(struct pci_dev *pdev, * This function takes care of initilizing the adapter to the point * where it can accept new commands. * Return value: - * 0 on success / -EIO on failure + * none **/ -static int ipr_probe_ioa_part2(struct ipr_ioa_cfg *ioa_cfg) +static void ipr_probe_ioa_part2(struct ipr_ioa_cfg *ioa_cfg) { - int rc = 0; unsigned long host_lock_flags = 0; ENTER; @@ -9520,7 +9519,6 @@ static int ipr_probe_ioa_part2(struct ipr_ioa_cfg *ioa_cfg) spin_unlock_irqrestore(ioa_cfg->host->host_lock, host_lock_flags); LEAVE; - return rc; } /** @@ -10563,12 +10561,7 @@ static int ipr_probe(struct pci_dev *pdev, const struct pci_device_id *dev_id) return rc; ioa_cfg = pci_get_drvdata(pdev); - rc = ipr_probe_ioa_part2(ioa_cfg); - - if (rc) { - __ipr_remove(pdev); - return rc; - } + ipr_probe_ioa_part2(ioa_cfg); rc = scsi_add_host(ioa_cfg->host, &pdev->dev); From ee4e7dfe4ffc9ca50c6875757bd119abfe22b5c5 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 14 Feb 2023 14:28:08 +0100 Subject: [PATCH 356/765] scsi: ipr: Work around fortify-string warning The ipr_log_vpd_compact() function triggers a fortified memcpy() warning about a potential string overflow with all versions of clang: In file included from drivers/scsi/ipr.c:43: In file included from include/linux/string.h:254: include/linux/fortify-string.h:520:4: error: call to '__write_overflow_field' declared with 'warning' attribute: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Werror,-Wattribute-warning] __write_overflow_field(p_size_field, size); ^ include/linux/fortify-string.h:520:4: error: call to '__write_overflow_field' declared with 'warning' attribute: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Werror,-Wattribute-warning] 2 errors generated. I don't see anything actually wrong with the function, but this is the only instance I can reproduce of the fortification going wrong in the kernel at the moment, so the easiest solution may be to rewrite the function into something that does not trigger the warning. Instead of having a combined buffer for vendor/device/serial strings, use three separate local variables and just truncate the whitespace individually. Link: https://lore.kernel.org/r/20230214132831.2118392-1-arnd@kernel.org Cc: Kees Cook Fixes: 8cf093e275d0 ("[SCSI] ipr: Improved dual adapter errors") Signed-off-by: Arnd Bergmann Reviewed-by: Damien Le Moal Reviewed-by: Kees Cook Acked-by: Brian King Signed-off-by: Martin K. Petersen --- drivers/scsi/ipr.c | 41 +++++++++++++++++++++-------------------- 1 file changed, 21 insertions(+), 20 deletions(-) diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index 5feffda0ef68f..b9f6a16cc1c68 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -1516,23 +1516,22 @@ static void ipr_process_ccn(struct ipr_cmnd *ipr_cmd) } /** - * strip_and_pad_whitespace - Strip and pad trailing whitespace. - * @i: index into buffer - * @buf: string to modify + * strip_whitespace - Strip and pad trailing whitespace. + * @i: size of buffer + * @buf: string to modify * - * This function will strip all trailing whitespace, pad the end - * of the string with a single space, and NULL terminate the string. + * This function will strip all trailing whitespace and + * NUL terminate the string. * - * Return value: - * new length of string **/ -static int strip_and_pad_whitespace(int i, char *buf) +static void strip_whitespace(int i, char *buf) { + if (i < 1) + return; + i--; while (i && buf[i] == ' ') i--; - buf[i+1] = ' '; - buf[i+2] = '\0'; - return i + 2; + buf[i+1] = '\0'; } /** @@ -1547,19 +1546,21 @@ static int strip_and_pad_whitespace(int i, char *buf) static void ipr_log_vpd_compact(char *prefix, struct ipr_hostrcb *hostrcb, struct ipr_vpd *vpd) { - char buffer[IPR_VENDOR_ID_LEN + IPR_PROD_ID_LEN + IPR_SERIAL_NUM_LEN + 3]; - int i = 0; + char vendor_id[IPR_VENDOR_ID_LEN + 1]; + char product_id[IPR_PROD_ID_LEN + 1]; + char sn[IPR_SERIAL_NUM_LEN + 1]; - memcpy(buffer, vpd->vpids.vendor_id, IPR_VENDOR_ID_LEN); - i = strip_and_pad_whitespace(IPR_VENDOR_ID_LEN - 1, buffer); + memcpy(vendor_id, vpd->vpids.vendor_id, IPR_VENDOR_ID_LEN); + strip_whitespace(IPR_VENDOR_ID_LEN, vendor_id); - memcpy(&buffer[i], vpd->vpids.product_id, IPR_PROD_ID_LEN); - i = strip_and_pad_whitespace(i + IPR_PROD_ID_LEN - 1, buffer); + memcpy(product_id, vpd->vpids.product_id, IPR_PROD_ID_LEN); + strip_whitespace(IPR_PROD_ID_LEN, product_id); - memcpy(&buffer[i], vpd->sn, IPR_SERIAL_NUM_LEN); - buffer[IPR_SERIAL_NUM_LEN + i] = '\0'; + memcpy(sn, vpd->sn, IPR_SERIAL_NUM_LEN); + strip_whitespace(IPR_SERIAL_NUM_LEN, sn); - ipr_hcam_err(hostrcb, "%s VPID/SN: %s\n", prefix, buffer); + ipr_hcam_err(hostrcb, "%s VPID/SN: %s %s %s\n", prefix, + vendor_id, product_id, sn); } /** From c6f2e6b6eaaf883df482cb94f302acad9b80a2a4 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Sat, 4 Feb 2023 10:37:16 -0800 Subject: [PATCH 357/765] scsi: mpi3mr: Replace 1-element array with flex-array Nothing else defined MPI3_NVME_ENCAP_CMD_MAX, so the "command" buffer was being defined as a fake flexible array of size 1. Replace this with a proper flex array. Avoids this GCC 13 warning under -fstrict-flex-arrays=3: In function 'fortify_memset_chk', inlined from 'mpi3mr_build_nvme_sgl' at ../drivers/scsi/mpi3mr/mpi3mr_app.c:693:2, inlined from 'mpi3mr_bsg_process_mpt_cmds.constprop' at ../drivers/scsi/mpi3mr/mpi3mr_app.c:1214:8: ../include/linux/fortify-string.h:430:25: warning: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Wattribute-warning] 430 | __write_overflow_field(p_size_field, size); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Link: https://lore.kernel.org/r/20230204183715.never.937-kees@kernel.org Cc: Sumit Saxena Cc: Himanshu Madhani Cc: "Martin K. Petersen" Cc: Stephen Rothwell Cc: kernel test robot Signed-off-by: Kees Cook Signed-off-by: Martin K. Petersen --- include/uapi/scsi/scsi_bsg_mpi3mr.h | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/include/uapi/scsi/scsi_bsg_mpi3mr.h b/include/uapi/scsi/scsi_bsg_mpi3mr.h index fdc3517f9e199..907d345f04f93 100644 --- a/include/uapi/scsi/scsi_bsg_mpi3mr.h +++ b/include/uapi/scsi/scsi_bsg_mpi3mr.h @@ -455,12 +455,6 @@ struct mpi3mr_bsg_packet { } cmd; }; - -/* MPI3: NVMe Encasulation related definitions */ -#ifndef MPI3_NVME_ENCAP_CMD_MAX -#define MPI3_NVME_ENCAP_CMD_MAX (1) -#endif - struct mpi3_nvme_encapsulated_request { __le16 host_tag; __u8 ioc_use_only02; @@ -474,7 +468,7 @@ struct mpi3_nvme_encapsulated_request { __le16 flags; __le32 data_length; __le32 reserved14[3]; - __le32 command[MPI3_NVME_ENCAP_CMD_MAX]; + __le32 command[]; }; struct mpi3_nvme_encapsulated_error_reply { From ae7d45f5283d30274039b95d3e6d53d33c66e991 Mon Sep 17 00:00:00 2001 From: Tomas Henzl Date: Mon, 13 Feb 2023 20:37:52 +0100 Subject: [PATCH 358/765] scsi: mpi3mr: Fix an issue found by KASAN Write only correct size (32 instead of 64 bytes). Link: https://lore.kernel.org/r/20230213193752.6859-1-thenzl@redhat.com Fixes: 42fc9fee116f ("scsi: mpi3mr: Add helper functions to manage device's port") Signed-off-by: Tomas Henzl Acked-by: Sathya Prakash Veerichetty Signed-off-by: Martin K. Petersen --- drivers/scsi/mpi3mr/mpi3mr_transport.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/mpi3mr/mpi3mr_transport.c b/drivers/scsi/mpi3mr/mpi3mr_transport.c index 3fc897336b5e0..3b61815979dab 100644 --- a/drivers/scsi/mpi3mr/mpi3mr_transport.c +++ b/drivers/scsi/mpi3mr/mpi3mr_transport.c @@ -1280,7 +1280,7 @@ void mpi3mr_sas_host_add(struct mpi3mr_ioc *mrioc) if (mrioc->sas_hba.enclosure_handle) { if (!(mpi3mr_cfg_get_enclosure_pg0(mrioc, &ioc_status, - &encl_pg0, sizeof(dev_pg0), + &encl_pg0, sizeof(encl_pg0), MPI3_ENCLOS_PGAD_FORM_HANDLE, mrioc->sas_hba.enclosure_handle)) && (ioc_status == MPI3_IOCSTATUS_SUCCESS)) From fb428a2005fc1260d18b989cc5199f281617f44d Mon Sep 17 00:00:00 2001 From: Shin'ichiro Kawasaki Date: Tue, 14 Feb 2023 09:50:16 +0900 Subject: [PATCH 359/765] scsi: mpi3mr: Fix issues in mpi3mr_get_all_tgt_info() The function mpi3mr_get_all_tgt_info() has four issues: 1) It calculates valid entry length in alltgt_info assuming the header part of the struct mpi3mr_device_map_info would equal to sizeof(u32). The correct size is sizeof(u64). 2) When it calculates the valid entry length kern_entrylen, it excludes one entry by subtracting 1 from num_devices. 3) It copies num_device by calling memcpy(). Substitution is enough. 4) It does not specify the calculated length to sg_copy_from_buffer(). Instead, it specifies the payload length which is larger than the alltgt_info size. It causes "BUG: KASAN: slab-out-of-bounds". Fix the issues by using the correct header size, removing the subtraction from num_devices, replacing the memcpy() with substitution and specifying the correct length to sg_copy_from_buffer(). Link: https://lore.kernel.org/r/20230214005019.1897251-2-shinichiro.kawasaki@wdc.com Cc: stable@vger.kernel.org Fixes: f5e6d5a34376 ("scsi: mpi3mr: Add support for driver commands") Signed-off-by: Shin'ichiro Kawasaki Acked-by: Sathya Prakash Veerichetty Signed-off-by: Martin K. Petersen --- drivers/scsi/mpi3mr/mpi3mr_app.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/drivers/scsi/mpi3mr/mpi3mr_app.c b/drivers/scsi/mpi3mr/mpi3mr_app.c index 9baac224b2135..72054e3a26cb8 100644 --- a/drivers/scsi/mpi3mr/mpi3mr_app.c +++ b/drivers/scsi/mpi3mr/mpi3mr_app.c @@ -312,7 +312,7 @@ static long mpi3mr_get_all_tgt_info(struct mpi3mr_ioc *mrioc, num_devices++; spin_unlock_irqrestore(&mrioc->tgtdev_lock, flags); - if ((job->request_payload.payload_len == sizeof(u32)) || + if ((job->request_payload.payload_len <= sizeof(u64)) || list_empty(&mrioc->tgtdev_list)) { sg_copy_from_buffer(job->request_payload.sg_list, job->request_payload.sg_cnt, @@ -320,14 +320,14 @@ static long mpi3mr_get_all_tgt_info(struct mpi3mr_ioc *mrioc, return 0; } - kern_entrylen = (num_devices - 1) * sizeof(*devmap_info); - size = sizeof(*alltgt_info) + kern_entrylen; + kern_entrylen = num_devices * sizeof(*devmap_info); + size = sizeof(u64) + kern_entrylen; alltgt_info = kzalloc(size, GFP_KERNEL); if (!alltgt_info) return -ENOMEM; devmap_info = alltgt_info->dmi; - memset((u8 *)devmap_info, 0xFF, (kern_entrylen + sizeof(*devmap_info))); + memset((u8 *)devmap_info, 0xFF, kern_entrylen); spin_lock_irqsave(&mrioc->tgtdev_lock, flags); list_for_each_entry(tgtdev, &mrioc->tgtdev_list, list) { if (i < num_devices) { @@ -344,9 +344,10 @@ static long mpi3mr_get_all_tgt_info(struct mpi3mr_ioc *mrioc, num_devices = i; spin_unlock_irqrestore(&mrioc->tgtdev_lock, flags); - memcpy(&alltgt_info->num_devices, &num_devices, sizeof(num_devices)); + alltgt_info->num_devices = num_devices; - usr_entrylen = (job->request_payload.payload_len - sizeof(u32)) / sizeof(*devmap_info); + usr_entrylen = (job->request_payload.payload_len - sizeof(u64)) / + sizeof(*devmap_info); usr_entrylen *= sizeof(*devmap_info); min_entrylen = min(usr_entrylen, kern_entrylen); if (min_entrylen && (!memcpy(&alltgt_info->dmi, devmap_info, min_entrylen))) { @@ -358,7 +359,7 @@ static long mpi3mr_get_all_tgt_info(struct mpi3mr_ioc *mrioc, sg_copy_from_buffer(job->request_payload.sg_list, job->request_payload.sg_cnt, - alltgt_info, job->request_payload.payload_len); + alltgt_info, (min_entrylen + sizeof(u64))); rval = 0; out: kfree(alltgt_info); From eeb270aee3e085411399f129fc14fa04bd6d83cf Mon Sep 17 00:00:00 2001 From: Shin'ichiro Kawasaki Date: Tue, 14 Feb 2023 09:50:17 +0900 Subject: [PATCH 360/765] scsi: mpi3mr: Remove unnecessary memcpy() to alltgt_info->dmi In the function mpi3mr_get_all_tgt_info(), devmap_info points to alltgt_info->dmi then there is no need to memcpy() data from devmap_info to alltgt_info->dmi. Remove the unnecessary memcpy(). This also allows to remove the local variable 'rval' and the goto label 'out'. Link: https://lore.kernel.org/r/20230214005019.1897251-3-shinichiro.kawasaki@wdc.com Cc: stable@vger.kernel.org Fixes: f5e6d5a34376 ("scsi: mpi3mr: Add support for driver commands") Signed-off-by: Shin'ichiro Kawasaki Acked-by: Sathya Prakash Veerichetty Signed-off-by: Martin K. Petersen --- drivers/scsi/mpi3mr/mpi3mr_app.c | 13 ++----------- 1 file changed, 2 insertions(+), 11 deletions(-) diff --git a/drivers/scsi/mpi3mr/mpi3mr_app.c b/drivers/scsi/mpi3mr/mpi3mr_app.c index 72054e3a26cb8..bff6377023979 100644 --- a/drivers/scsi/mpi3mr/mpi3mr_app.c +++ b/drivers/scsi/mpi3mr/mpi3mr_app.c @@ -293,7 +293,6 @@ static long mpi3mr_bsg_pel_enable(struct mpi3mr_ioc *mrioc, static long mpi3mr_get_all_tgt_info(struct mpi3mr_ioc *mrioc, struct bsg_job *job) { - long rval = -EINVAL; u16 num_devices = 0, i = 0, size; unsigned long flags; struct mpi3mr_tgt_dev *tgtdev; @@ -304,7 +303,7 @@ static long mpi3mr_get_all_tgt_info(struct mpi3mr_ioc *mrioc, if (job->request_payload.payload_len < sizeof(u32)) { dprint_bsg_err(mrioc, "%s: invalid size argument\n", __func__); - return rval; + return -EINVAL; } spin_lock_irqsave(&mrioc->tgtdev_lock, flags); @@ -350,20 +349,12 @@ static long mpi3mr_get_all_tgt_info(struct mpi3mr_ioc *mrioc, sizeof(*devmap_info); usr_entrylen *= sizeof(*devmap_info); min_entrylen = min(usr_entrylen, kern_entrylen); - if (min_entrylen && (!memcpy(&alltgt_info->dmi, devmap_info, min_entrylen))) { - dprint_bsg_err(mrioc, "%s:%d: device map info copy failed\n", - __func__, __LINE__); - rval = -EFAULT; - goto out; - } sg_copy_from_buffer(job->request_payload.sg_list, job->request_payload.sg_cnt, alltgt_info, (min_entrylen + sizeof(u64))); - rval = 0; -out: kfree(alltgt_info); - return rval; + return 0; } /** * mpi3mr_get_change_count - Get topology change count From 339e61565f81a6534afdc18fd854b2e2628bf5db Mon Sep 17 00:00:00 2001 From: Shin'ichiro Kawasaki Date: Tue, 14 Feb 2023 09:50:18 +0900 Subject: [PATCH 361/765] scsi: mpi3mr: Use number of bits to manage bitmap sizes To allocate bitmaps, the mpi3mr driver calculates sizes of bitmaps using byte as unit. However, bitmap helper functions assume that bitmaps are allocated using unsigned long as unit. This gap causes memory access beyond the bitmap sizes and results in "BUG: KASAN: slab-out-of-bounds". The BUG was observed at firmware download to eHBA-9600. Call trace indicated that the out-of-bounds access happened in find_first_zero_bit() called from mpi3mr_send_event_ack() for miroc->evtack_cmds_bitmap. To fix the BUG, do not use bytes to manage bitmap sizes. Instead, use number of bits, and call bitmap helper functions which take number of bits as arguments. For memory allocation, call bitmap_zalloc() instead of kzalloc() and krealloc(). For memory free, call bitmap_free() instead of kfree(). For zero clear, call bitmap_clear() instead of memset(). Remove three fields for bitmap byte sizes in struct scmd_priv which are no longer required. Replace the field dev_handle_bitmap_sz with dev_handle_bitmap_bits to keep number of bits of removepend_bitmap across resize. Link: https://lore.kernel.org/r/20230214005019.1897251-4-shinichiro.kawasaki@wdc.com Fixes: c5758fc72b92 ("scsi: mpi3mr: Gracefully handle online FW update operation") Fixes: e844adb1fbdc ("scsi: mpi3mr: Implement SCSI error handler hooks") Fixes: c1af985d27da ("scsi: mpi3mr: Add Event acknowledgment logic") Fixes: 824a156633df ("scsi: mpi3mr: Base driver code") Signed-off-by: Shin'ichiro Kawasaki Acked-by: Sathya Prakash Veerichetty Signed-off-by: Martin K. Petersen --- drivers/scsi/mpi3mr/mpi3mr.h | 10 +---- drivers/scsi/mpi3mr/mpi3mr_fw.c | 75 ++++++++++++++------------------- 2 files changed, 33 insertions(+), 52 deletions(-) diff --git a/drivers/scsi/mpi3mr/mpi3mr.h b/drivers/scsi/mpi3mr/mpi3mr.h index def4c5e15cd89..8a438f248a820 100644 --- a/drivers/scsi/mpi3mr/mpi3mr.h +++ b/drivers/scsi/mpi3mr/mpi3mr.h @@ -955,19 +955,16 @@ struct scmd_priv { * @chain_buf_count: Chain buffer count * @chain_buf_pool: Chain buffer pool * @chain_sgl_list: Chain SGL list - * @chain_bitmap_sz: Chain buffer allocator bitmap size * @chain_bitmap: Chain buffer allocator bitmap * @chain_buf_lock: Chain buffer list lock * @bsg_cmds: Command tracker for BSG command * @host_tm_cmds: Command tracker for task management commands * @dev_rmhs_cmds: Command tracker for device removal commands * @evtack_cmds: Command tracker for event ack commands - * @devrem_bitmap_sz: Device removal bitmap size * @devrem_bitmap: Device removal bitmap - * @dev_handle_bitmap_sz: Device handle bitmap size + * @dev_handle_bitmap_bits: Number of bits in device handle bitmap * @removepend_bitmap: Remove pending bitmap * @delayed_rmhs_list: Delayed device removal list - * @evtack_cmds_bitmap_sz: Event Ack bitmap size * @evtack_cmds_bitmap: Event Ack bitmap * @delayed_evtack_cmds_list: Delayed event acknowledgment list * @ts_update_counter: Timestamp update counter @@ -1128,7 +1125,6 @@ struct mpi3mr_ioc { u32 chain_buf_count; struct dma_pool *chain_buf_pool; struct chain_element *chain_sgl_list; - u16 chain_bitmap_sz; void *chain_bitmap; spinlock_t chain_buf_lock; @@ -1136,12 +1132,10 @@ struct mpi3mr_ioc { struct mpi3mr_drv_cmd host_tm_cmds; struct mpi3mr_drv_cmd dev_rmhs_cmds[MPI3MR_NUM_DEVRMCMD]; struct mpi3mr_drv_cmd evtack_cmds[MPI3MR_NUM_EVTACKCMD]; - u16 devrem_bitmap_sz; void *devrem_bitmap; - u16 dev_handle_bitmap_sz; + u16 dev_handle_bitmap_bits; void *removepend_bitmap; struct list_head delayed_rmhs_list; - u16 evtack_cmds_bitmap_sz; void *evtack_cmds_bitmap; struct list_head delayed_evtack_cmds_list; diff --git a/drivers/scsi/mpi3mr/mpi3mr_fw.c b/drivers/scsi/mpi3mr/mpi3mr_fw.c index 0c4aabaefdcc4..1e4467ea8472a 100644 --- a/drivers/scsi/mpi3mr/mpi3mr_fw.c +++ b/drivers/scsi/mpi3mr/mpi3mr_fw.c @@ -1128,7 +1128,6 @@ static int mpi3mr_issue_and_process_mur(struct mpi3mr_ioc *mrioc, static int mpi3mr_revalidate_factsdata(struct mpi3mr_ioc *mrioc) { - u16 dev_handle_bitmap_sz; void *removepend_bitmap; if (mrioc->facts.reply_sz > mrioc->reply_sz) { @@ -1160,25 +1159,23 @@ mpi3mr_revalidate_factsdata(struct mpi3mr_ioc *mrioc) "\tcontroller while sas transport support is enabled at the\n" "\tdriver, please reboot the system or reload the driver\n"); - dev_handle_bitmap_sz = mrioc->facts.max_devhandle / 8; - if (mrioc->facts.max_devhandle % 8) - dev_handle_bitmap_sz++; - if (dev_handle_bitmap_sz > mrioc->dev_handle_bitmap_sz) { - removepend_bitmap = krealloc(mrioc->removepend_bitmap, - dev_handle_bitmap_sz, GFP_KERNEL); + if (mrioc->facts.max_devhandle > mrioc->dev_handle_bitmap_bits) { + removepend_bitmap = bitmap_zalloc(mrioc->facts.max_devhandle, + GFP_KERNEL); if (!removepend_bitmap) { ioc_err(mrioc, - "failed to increase removepend_bitmap sz from: %d to %d\n", - mrioc->dev_handle_bitmap_sz, dev_handle_bitmap_sz); + "failed to increase removepend_bitmap bits from %d to %d\n", + mrioc->dev_handle_bitmap_bits, + mrioc->facts.max_devhandle); return -EPERM; } - memset(removepend_bitmap + mrioc->dev_handle_bitmap_sz, 0, - dev_handle_bitmap_sz - mrioc->dev_handle_bitmap_sz); + bitmap_free(mrioc->removepend_bitmap); mrioc->removepend_bitmap = removepend_bitmap; ioc_info(mrioc, - "increased dev_handle_bitmap_sz from %d to %d\n", - mrioc->dev_handle_bitmap_sz, dev_handle_bitmap_sz); - mrioc->dev_handle_bitmap_sz = dev_handle_bitmap_sz; + "increased bits of dev_handle_bitmap from %d to %d\n", + mrioc->dev_handle_bitmap_bits, + mrioc->facts.max_devhandle); + mrioc->dev_handle_bitmap_bits = mrioc->facts.max_devhandle; } return 0; @@ -2957,27 +2954,18 @@ static int mpi3mr_alloc_reply_sense_bufs(struct mpi3mr_ioc *mrioc) if (!mrioc->pel_abort_cmd.reply) goto out_failed; - mrioc->dev_handle_bitmap_sz = mrioc->facts.max_devhandle / 8; - if (mrioc->facts.max_devhandle % 8) - mrioc->dev_handle_bitmap_sz++; - mrioc->removepend_bitmap = kzalloc(mrioc->dev_handle_bitmap_sz, - GFP_KERNEL); + mrioc->dev_handle_bitmap_bits = mrioc->facts.max_devhandle; + mrioc->removepend_bitmap = bitmap_zalloc(mrioc->dev_handle_bitmap_bits, + GFP_KERNEL); if (!mrioc->removepend_bitmap) goto out_failed; - mrioc->devrem_bitmap_sz = MPI3MR_NUM_DEVRMCMD / 8; - if (MPI3MR_NUM_DEVRMCMD % 8) - mrioc->devrem_bitmap_sz++; - mrioc->devrem_bitmap = kzalloc(mrioc->devrem_bitmap_sz, - GFP_KERNEL); + mrioc->devrem_bitmap = bitmap_zalloc(MPI3MR_NUM_DEVRMCMD, GFP_KERNEL); if (!mrioc->devrem_bitmap) goto out_failed; - mrioc->evtack_cmds_bitmap_sz = MPI3MR_NUM_EVTACKCMD / 8; - if (MPI3MR_NUM_EVTACKCMD % 8) - mrioc->evtack_cmds_bitmap_sz++; - mrioc->evtack_cmds_bitmap = kzalloc(mrioc->evtack_cmds_bitmap_sz, - GFP_KERNEL); + mrioc->evtack_cmds_bitmap = bitmap_zalloc(MPI3MR_NUM_EVTACKCMD, + GFP_KERNEL); if (!mrioc->evtack_cmds_bitmap) goto out_failed; @@ -3415,10 +3403,7 @@ static int mpi3mr_alloc_chain_bufs(struct mpi3mr_ioc *mrioc) if (!mrioc->chain_sgl_list[i].addr) goto out_failed; } - mrioc->chain_bitmap_sz = num_chains / 8; - if (num_chains % 8) - mrioc->chain_bitmap_sz++; - mrioc->chain_bitmap = kzalloc(mrioc->chain_bitmap_sz, GFP_KERNEL); + mrioc->chain_bitmap = bitmap_zalloc(num_chains, GFP_KERNEL); if (!mrioc->chain_bitmap) goto out_failed; return retval; @@ -4190,10 +4175,11 @@ void mpi3mr_memset_buffers(struct mpi3mr_ioc *mrioc) for (i = 0; i < MPI3MR_NUM_EVTACKCMD; i++) memset(mrioc->evtack_cmds[i].reply, 0, sizeof(*mrioc->evtack_cmds[i].reply)); - memset(mrioc->removepend_bitmap, 0, mrioc->dev_handle_bitmap_sz); - memset(mrioc->devrem_bitmap, 0, mrioc->devrem_bitmap_sz); - memset(mrioc->evtack_cmds_bitmap, 0, - mrioc->evtack_cmds_bitmap_sz); + bitmap_clear(mrioc->removepend_bitmap, 0, + mrioc->dev_handle_bitmap_bits); + bitmap_clear(mrioc->devrem_bitmap, 0, MPI3MR_NUM_DEVRMCMD); + bitmap_clear(mrioc->evtack_cmds_bitmap, 0, + MPI3MR_NUM_EVTACKCMD); } for (i = 0; i < mrioc->num_queues; i++) { @@ -4319,16 +4305,16 @@ void mpi3mr_free_mem(struct mpi3mr_ioc *mrioc) mrioc->evtack_cmds[i].reply = NULL; } - kfree(mrioc->removepend_bitmap); + bitmap_free(mrioc->removepend_bitmap); mrioc->removepend_bitmap = NULL; - kfree(mrioc->devrem_bitmap); + bitmap_free(mrioc->devrem_bitmap); mrioc->devrem_bitmap = NULL; - kfree(mrioc->evtack_cmds_bitmap); + bitmap_free(mrioc->evtack_cmds_bitmap); mrioc->evtack_cmds_bitmap = NULL; - kfree(mrioc->chain_bitmap); + bitmap_free(mrioc->chain_bitmap); mrioc->chain_bitmap = NULL; kfree(mrioc->transport_cmds.reply); @@ -4887,9 +4873,10 @@ int mpi3mr_soft_reset_handler(struct mpi3mr_ioc *mrioc, mpi3mr_flush_delayed_cmd_lists(mrioc); mpi3mr_flush_drv_cmds(mrioc); - memset(mrioc->devrem_bitmap, 0, mrioc->devrem_bitmap_sz); - memset(mrioc->removepend_bitmap, 0, mrioc->dev_handle_bitmap_sz); - memset(mrioc->evtack_cmds_bitmap, 0, mrioc->evtack_cmds_bitmap_sz); + bitmap_clear(mrioc->devrem_bitmap, 0, MPI3MR_NUM_DEVRMCMD); + bitmap_clear(mrioc->removepend_bitmap, 0, + mrioc->dev_handle_bitmap_bits); + bitmap_clear(mrioc->evtack_cmds_bitmap, 0, MPI3MR_NUM_EVTACKCMD); mpi3mr_flush_host_io(mrioc); mpi3mr_cleanup_fwevt_list(mrioc); mpi3mr_invalidate_devhandles(mrioc); From e39ea831ebad4ab15c4748cb62a397a8abcca36e Mon Sep 17 00:00:00 2001 From: Shin'ichiro Kawasaki Date: Tue, 14 Feb 2023 09:50:19 +0900 Subject: [PATCH 362/765] scsi: mpi3mr: Fix missing mrioc->evtack_cmds initialization Commit c1af985d27da ("scsi: mpi3mr: Add Event acknowledgment logic") introduced an array mrioc->evtack_cmds but initialization of the array elements was missed. They are just zero cleared. The function mpi3mr_complete_evt_ack() refers host_tag field of the elements. Due to the zero value of the host_tag field, the function calls clear_bit() for mrico->evtack_cmds_bitmap with wrong bit index. This results in memory access to invalid address and "BUG: KASAN: use-after-free". This BUG was observed at eHBA-9600 firmware update to version 8.3.1.0. To fix it, add the missing initialization of mrioc->evtack_cmds. Link: https://lore.kernel.org/r/20230214005019.1897251-5-shinichiro.kawasaki@wdc.com Cc: stable@vger.kernel.org Fixes: c1af985d27da ("scsi: mpi3mr: Add Event acknowledgment logic") Signed-off-by: Shin'ichiro Kawasaki Reviewed-by: Damien Le Moal Acked-by: Sathya Prakash Veerichetty Signed-off-by: Martin K. Petersen --- drivers/scsi/mpi3mr/mpi3mr_os.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/scsi/mpi3mr/mpi3mr_os.c b/drivers/scsi/mpi3mr/mpi3mr_os.c index 3306de7170f64..6eaeba41072cb 100644 --- a/drivers/scsi/mpi3mr/mpi3mr_os.c +++ b/drivers/scsi/mpi3mr/mpi3mr_os.c @@ -4952,6 +4952,10 @@ mpi3mr_probe(struct pci_dev *pdev, const struct pci_device_id *id) mpi3mr_init_drv_cmd(&mrioc->dev_rmhs_cmds[i], MPI3MR_HOSTTAG_DEVRMCMD_MIN + i); + for (i = 0; i < MPI3MR_NUM_EVTACKCMD; i++) + mpi3mr_init_drv_cmd(&mrioc->evtack_cmds[i], + MPI3MR_HOSTTAG_EVTACKCMD_MIN + i); + if (pdev->revision) mrioc->enable_segqueue = true; From 66b381d874fa440dbcd20e1a9078849f7dd2ff4c Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Thu, 16 Feb 2023 01:51:30 +0100 Subject: [PATCH 363/765] scsi: mpi3mr: Remove unneeded version.h include Remove unneeded version.h include pointed out by 'make versioncheck'. Link: https://lore.kernel.org/r/820137c2-decc-3d78-f170-7f1c0571fbb7@gmail.com Signed-off-by: Jesper Juhl Signed-off-by: Martin K. Petersen --- drivers/scsi/mpi3mr/mpi3mr.h | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/scsi/mpi3mr/mpi3mr.h b/drivers/scsi/mpi3mr/mpi3mr.h index 8a438f248a820..23de2603e71fd 100644 --- a/drivers/scsi/mpi3mr/mpi3mr.h +++ b/drivers/scsi/mpi3mr/mpi3mr.h @@ -29,7 +29,6 @@ #include #include #include -#include #include #include #include From e0aa38c444f2fcf1fceb4fef7902209dd25b48ea Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Thu, 16 Feb 2023 01:51:30 +0100 Subject: [PATCH 364/765] scsi: qedi: Remove unneeded version.h include Remove unneeded version.h include pointed out by 'make versioncheck'. Link: https://lore.kernel.org/r/820137c2-decc-3d78-f170-7f1c0571fbb7@gmail.com Signed-off-by: Jesper Juhl Signed-off-by: Martin K. Petersen --- drivers/scsi/qedi/qedi_dbg.h | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/scsi/qedi/qedi_dbg.h b/drivers/scsi/qedi/qedi_dbg.h index 37d084086fd43..fdda12ef13b0f 100644 --- a/drivers/scsi/qedi/qedi_dbg.h +++ b/drivers/scsi/qedi/qedi_dbg.h @@ -11,7 +11,6 @@ #include #include #include -#include #include #include #include From 5794a3902bde986ade963cdf4db98b6baa433d20 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Thu, 16 Feb 2023 01:51:30 +0100 Subject: [PATCH 365/765] scsi: cxgbi: Remove unneeded version.h include Remove unneeded version.h include pointed out by 'make versioncheck'. Link: https://lore.kernel.org/r/820137c2-decc-3d78-f170-7f1c0571fbb7@gmail.com Signed-off-by: Jesper Juhl Signed-off-by: Martin K. Petersen --- drivers/scsi/cxgbi/libcxgbi.h | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/scsi/cxgbi/libcxgbi.h b/drivers/scsi/cxgbi/libcxgbi.h index 3687b5c0cf905..d8fc7beafa205 100644 --- a/drivers/scsi/cxgbi/libcxgbi.h +++ b/drivers/scsi/cxgbi/libcxgbi.h @@ -24,7 +24,6 @@ #include #include #include -#include #include #include From f4407e6033a69ff633ee9d6acf6d241223c7c9fb Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Fri, 10 Feb 2023 12:51:59 -0800 Subject: [PATCH 366/765] scsi: core: Fix a source code comment Fix an incorrect reference to the scsi_remove_host() function in a source code comment. Link: https://lore.kernel.org/r/20230210205200.36973-2-bvanassche@acm.org Fixes: b49493f99690 ("Fix a memory leak in scsi_host_dev_release()") Signed-off-by: Bart Van Assche Signed-off-by: Martin K. Petersen --- drivers/scsi/hosts.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c index 12346e2297fdb..b28375f9e0194 100644 --- a/drivers/scsi/hosts.c +++ b/drivers/scsi/hosts.c @@ -356,7 +356,7 @@ static void scsi_host_dev_release(struct device *dev) /* * Free the shost_dev device name here if scsi_host_alloc() * and scsi_host_put() have been called but neither - * scsi_host_add() nor scsi_host_remove() has been called. + * scsi_host_add() nor scsi_remove_host() has been called. * This avoids that the memory allocated for the shost_dev * name is leaked. */ From fc663711b94468f4e1427ebe289c9f05669699c9 Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Fri, 10 Feb 2023 12:52:00 -0800 Subject: [PATCH 367/765] scsi: core: Remove the /proc/scsi/${proc_name} directory earlier Remove the /proc/scsi/${proc_name} directory earlier to fix a race condition between unloading and reloading kernel modules. This fixes a bug introduced in 2009 by commit 77c019768f06 ("[SCSI] fix /proc memory leak in the SCSI core"). Fix the following kernel warning: proc_dir_entry 'scsi/scsi_debug' already registered WARNING: CPU: 19 PID: 27986 at fs/proc/generic.c:376 proc_register+0x27d/0x2e0 Call Trace: proc_mkdir+0xb5/0xe0 scsi_proc_hostdir_add+0xb5/0x170 scsi_host_alloc+0x683/0x6c0 sdebug_driver_probe+0x6b/0x2d0 [scsi_debug] really_probe+0x159/0x540 __driver_probe_device+0xdc/0x230 driver_probe_device+0x4f/0x120 __device_attach_driver+0xef/0x180 bus_for_each_drv+0xe5/0x130 __device_attach+0x127/0x290 device_initial_probe+0x17/0x20 bus_probe_device+0x110/0x130 device_add+0x673/0xc80 device_register+0x1e/0x30 sdebug_add_host_helper+0x1a7/0x3b0 [scsi_debug] scsi_debug_init+0x64f/0x1000 [scsi_debug] do_one_initcall+0xd7/0x470 do_init_module+0xe7/0x330 load_module+0x122a/0x12c0 __do_sys_finit_module+0x124/0x1a0 __x64_sys_finit_module+0x46/0x50 do_syscall_64+0x38/0x80 entry_SYSCALL_64_after_hwframe+0x46/0xb0 Link: https://lore.kernel.org/r/20230210205200.36973-3-bvanassche@acm.org Cc: Alan Stern Cc: Yi Zhang Cc: stable@vger.kernel.org Fixes: 77c019768f06 ("[SCSI] fix /proc memory leak in the SCSI core") Reported-by: Yi Zhang Signed-off-by: Bart Van Assche Signed-off-by: Martin K. Petersen --- drivers/scsi/hosts.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c index b28375f9e0194..f7f62e56afcae 100644 --- a/drivers/scsi/hosts.c +++ b/drivers/scsi/hosts.c @@ -181,6 +181,7 @@ void scsi_remove_host(struct Scsi_Host *shost) scsi_forget_host(shost); mutex_unlock(&shost->scan_mutex); scsi_proc_host_rm(shost); + scsi_proc_hostdir_rm(shost->hostt); /* * New SCSI devices cannot be attached anymore because of the SCSI host @@ -340,6 +341,7 @@ static void scsi_host_dev_release(struct device *dev) struct Scsi_Host *shost = dev_to_shost(dev); struct device *parent = dev->parent; + /* In case scsi_remove_host() has not been called. */ scsi_proc_hostdir_rm(shost->hostt); /* Wait for functions invoked through call_rcu(&scmd->rcu, ...) */ From 442336a5a999d4aae6221e3496dad071a152b1fb Mon Sep 17 00:00:00 2001 From: Bo Liu Date: Fri, 17 Feb 2023 03:30:46 -0500 Subject: [PATCH 368/765] scsi: lpfc: Fix double word in comments Remove the repeated word "the" in comments. [mkp: fixed additional typos in the changed lines] Link: https://lore.kernel.org/r/20230217083046.4090-1-liubo03@inspur.com Signed-off-by: Bo Liu Reviewed-by: Randy Dunlap Signed-off-by: Martin K. Petersen --- drivers/scsi/lpfc/lpfc_attr.c | 10 +++++----- drivers/scsi/lpfc/lpfc_els.c | 2 +- drivers/scsi/lpfc/lpfc_hbadisc.c | 2 +- drivers/scsi/lpfc/lpfc_init.c | 4 ++-- drivers/scsi/lpfc/lpfc_mbox.c | 4 ++-- drivers/scsi/lpfc/lpfc_nvmet.c | 2 +- drivers/scsi/lpfc/lpfc_sli.c | 2 +- 7 files changed, 13 insertions(+), 13 deletions(-) diff --git a/drivers/scsi/lpfc/lpfc_attr.c b/drivers/scsi/lpfc/lpfc_attr.c index 76c3434f89766..22f2e046e8eb7 100644 --- a/drivers/scsi/lpfc/lpfc_attr.c +++ b/drivers/scsi/lpfc/lpfc_attr.c @@ -2541,7 +2541,7 @@ lpfc_sriov_hw_max_virtfn_show(struct device *dev, /** * lpfc_enable_bbcr_set: Sets an attribute value. - * @phba: pointer the the adapter structure. + * @phba: pointer to the adapter structure. * @val: integer attribute value. * * Description: @@ -2632,7 +2632,7 @@ lpfc_##attr##_show(struct device *dev, struct device_attribute *attr, \ * takes a default argument, a minimum and maximum argument. * * lpfc_##attr##_init: Initializes an attribute. - * @phba: pointer the the adapter structure. + * @phba: pointer to the adapter structure. * @val: integer attribute value. * * Validates the min and max values then sets the adapter config field @@ -2665,7 +2665,7 @@ lpfc_##attr##_init(struct lpfc_hba *phba, uint val) \ * into a function with the name lpfc_hba_queue_depth_set * * lpfc_##attr##_set: Sets an attribute value. - * @phba: pointer the the adapter structure. + * @phba: pointer to the adapter structure. * @val: integer attribute value. * * Description: @@ -2794,7 +2794,7 @@ lpfc_##attr##_show(struct device *dev, struct device_attribute *attr, \ * lpfc_##attr##_init: validates the min and max values then sets the * adapter config field accordingly, or uses the default if out of range * and prints an error message. - * @phba: pointer the the adapter structure. + * @phba: pointer to the adapter structure. * @val: integer attribute value. * * Returns: @@ -2826,7 +2826,7 @@ lpfc_##attr##_init(struct lpfc_vport *vport, uint val) \ * lpfc_##attr##_set: validates the min and max values then sets the * adapter config field if in the valid range. prints error message * and does not set the parameter if invalid. - * @phba: pointer the the adapter structure. + * @phba: pointer to the adapter structure. * @val: integer attribute value. * * Returns: diff --git a/drivers/scsi/lpfc/lpfc_els.c b/drivers/scsi/lpfc/lpfc_els.c index aee5d0d1187d6..35b252f1ef731 100644 --- a/drivers/scsi/lpfc/lpfc_els.c +++ b/drivers/scsi/lpfc/lpfc_els.c @@ -8886,7 +8886,7 @@ lpfc_els_rcv_rtv(struct lpfc_vport *vport, struct lpfc_iocbq *cmdiocb, * @rrq: Pointer to the rrq struct. * * Build a ELS RRQ command and send it to the target. If the issue_iocb is - * Successful the the completion handler will clear the RRQ. + * successful, the completion handler will clear the RRQ. * * Return codes * 0 - Successfully sent rrq els iocb. diff --git a/drivers/scsi/lpfc/lpfc_hbadisc.c b/drivers/scsi/lpfc/lpfc_hbadisc.c index a6df0a5b40060..66cd0b1dbbd02 100644 --- a/drivers/scsi/lpfc/lpfc_hbadisc.c +++ b/drivers/scsi/lpfc/lpfc_hbadisc.c @@ -2459,7 +2459,7 @@ static void lpfc_sli4_fcf_pri_list_del(struct lpfc_hba *phba, * @phba: pointer to lpfc hba data structure. * @fcf_index: the index of the fcf record to update * This routine acquires the hbalock and then set the LPFC_FCF_FLOGI_FAILED - * flag so the the round robin slection for the particular priority level + * flag so the round robin selection for the particular priority level * will try a different fcf record that does not have this bit set. * If the fcf record is re-read for any reason this flag is cleared brfore * adding it to the priority list. diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c index faaaeae25d447..75737088d0116 100644 --- a/drivers/scsi/lpfc/lpfc_init.c +++ b/drivers/scsi/lpfc/lpfc_init.c @@ -5501,7 +5501,7 @@ lpfc_sli4_async_link_evt(struct lpfc_hba *phba, bf_set(lpfc_mbx_read_top_link_spd, la, (bf_get(lpfc_acqe_link_speed, acqe_link))); - /* Fake the the following irrelvant fields */ + /* Fake the following irrelevant fields */ bf_set(lpfc_mbx_read_top_topology, la, LPFC_TOPOLOGY_PT_PT); bf_set(lpfc_mbx_read_top_alpa_granted, la, 0); bf_set(lpfc_mbx_read_top_il, la, 0); @@ -12548,7 +12548,7 @@ lpfc_cpu_affinity_check(struct lpfc_hba *phba, int vectors) /* Mark CPU as IRQ not assigned by the kernel */ cpup->flag |= LPFC_CPU_MAP_UNASSIGN; - /* If so, find a new_cpup thats on the the SAME + /* If so, find a new_cpup that is on the SAME * phys_id as cpup. start_cpu will start where we * left off so all unassigned entries don't get assgined * the IRQ of the first entry. diff --git a/drivers/scsi/lpfc/lpfc_mbox.c b/drivers/scsi/lpfc/lpfc_mbox.c index 9858b17437697..0dfdc0c4c08c6 100644 --- a/drivers/scsi/lpfc/lpfc_mbox.c +++ b/drivers/scsi/lpfc/lpfc_mbox.c @@ -2509,7 +2509,7 @@ lpfc_sli4_dump_page_a0(struct lpfc_hba *phba, struct lpfcMboxq *mbox) * information via a READ_FCF mailbox command. This mailbox command also is used * to indicate where received unsolicited frames from this FCF will be sent. By * default this routine will set up the FCF to forward all unsolicited frames - * the the RQ ID passed in the @phba. This can be overridden by the caller for + * to the RQ ID passed in the @phba. This can be overridden by the caller for * more complicated setups. **/ void @@ -2577,7 +2577,7 @@ lpfc_reg_fcfi(struct lpfc_hba *phba, struct lpfcMboxq *mbox) * information via a READ_FCF mailbox command. This mailbox command also is used * to indicate where received unsolicited frames from this FCF will be sent. By * default this routine will set up the FCF to forward all unsolicited frames - * the the RQ ID passed in the @phba. This can be overridden by the caller for + * to the RQ ID passed in the @phba. This can be overridden by the caller for * more complicated setups. **/ void diff --git a/drivers/scsi/lpfc/lpfc_nvmet.c b/drivers/scsi/lpfc/lpfc_nvmet.c index f7cfac0da9b6e..7517dd55fe919 100644 --- a/drivers/scsi/lpfc/lpfc_nvmet.c +++ b/drivers/scsi/lpfc/lpfc_nvmet.c @@ -1469,7 +1469,7 @@ lpfc_nvmet_cleanup_io_context(struct lpfc_hba *phba) if (!infop) return; - /* Cycle the the entire CPU context list for every MRQ */ + /* Cycle the entire CPU context list for every MRQ */ for (i = 0; i < phba->cfg_nvmet_mrq; i++) { for_each_present_cpu(j) { infop = lpfc_get_ctx_list(phba, j, i); diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c index edbd81c3b6432..c5b69f313af36 100644 --- a/drivers/scsi/lpfc/lpfc_sli.c +++ b/drivers/scsi/lpfc/lpfc_sli.c @@ -20804,7 +20804,7 @@ lpfc_log_fw_write_cmpl(struct lpfc_hba *phba, u32 shdr_status, * the offset after the write object mailbox has completed. @size is used to * determine the end of the object and whether the eof bit should be set. * - * Return 0 is successful and offset will contain the the new offset to use + * Return 0 is successful and offset will contain the new offset to use * for the next write. * Return negative value for error cases. **/ From 35cd2f5542df569122d48caf606b972642012c50 Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Fri, 10 Feb 2023 11:32:56 -0800 Subject: [PATCH 369/765] scsi: core: Extend struct scsi_exec_args Allow SCSI LLDs to specify SCMD_* flags. Link: https://lore.kernel.org/r/20230210193258.4004923-2-bvanassche@acm.org Cc: Mike Christie Cc: John Garry Reviewed-by: John Garry Signed-off-by: Bart Van Assche Signed-off-by: Martin K. Petersen --- drivers/scsi/scsi_lib.c | 1 + include/scsi/scsi_device.h | 1 + 2 files changed, 2 insertions(+) diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index abe93ec8b7d08..b7c569a42aa47 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -229,6 +229,7 @@ int scsi_execute_cmd(struct scsi_device *sdev, const unsigned char *cmd, scmd->cmd_len = COMMAND_SIZE(cmd[0]); memcpy(scmd->cmnd, cmd, scmd->cmd_len); scmd->allowed = retries; + scmd->flags |= args->scmd_flags; req->timeout = timeout; req->rq_flags |= RQF_QUIET; diff --git a/include/scsi/scsi_device.h b/include/scsi/scsi_device.h index 7e95ec45138fe..de310f21406c5 100644 --- a/include/scsi/scsi_device.h +++ b/include/scsi/scsi_device.h @@ -462,6 +462,7 @@ struct scsi_exec_args { unsigned int sense_len; /* sense buffer len */ struct scsi_sense_hdr *sshdr; /* decoded sense header */ blk_mq_req_flags_t req_flags; /* BLK_MQ_REQ flags */ + int scmd_flags; /* SCMD flags */ int *resid; /* residual length */ }; From 93bc4a5d00e472003ae983bb21febd2519a64f62 Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Fri, 10 Feb 2023 11:32:57 -0800 Subject: [PATCH 370/765] scsi: ufs: core: Rely on the block layer for setting RQF_PM Do not set RQF_PM explicitly since scsi_alloc_request() sets it indirectly if BLK_MQ_REQ_PM is set. The call chain for the code that sets RQF_PM is as follows: scsi_alloc_request() blk_mq_alloc_request() __blk_mq_alloc_requests() blk_mq_rq_ctx_init() if (data->flags & BLK_MQ_REQ_PM) data->rq_flags |= RQF_PM; Link: https://lore.kernel.org/r/20230210193258.4004923-3-bvanassche@acm.org Cc: Mike Christie Cc: John Garry Reviewed-by: John Garry Signed-off-by: Bart Van Assche Signed-off-by: Martin K. Petersen --- drivers/ufs/core/ufshcd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c index 47c7739b9a868..9d8e03b30014e 100644 --- a/drivers/ufs/core/ufshcd.c +++ b/drivers/ufs/core/ufshcd.c @@ -9150,7 +9150,7 @@ static int ufshcd_execute_start_stop(struct scsi_device *sdev, scmd->allowed = 0/*retries*/; scmd->flags |= SCMD_FAIL_IF_RECOVERING; req->timeout = 1 * HZ; - req->rq_flags |= RQF_PM | RQF_QUIET; + req->rq_flags |= RQF_QUIET; blk_execute_rq(req, /*at_head=*/true); From 2702812ae33b38898de6d950cdb6a03888d001af Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Fri, 10 Feb 2023 11:32:58 -0800 Subject: [PATCH 371/765] scsi: ufs: core: Simplify ufshcd_execute_start_stop() Use scsi_execute_cmd() instead of open-coding it. Link: https://lore.kernel.org/r/20230210193258.4004923-4-bvanassche@acm.org Cc: Mike Christie Cc: John Garry Signed-off-by: Bart Van Assche Reviewed-by: John Garry Reviewed-by: Asutosh Das Signed-off-by: Martin K. Petersen --- drivers/ufs/core/ufshcd.c | 35 ++++++++--------------------------- 1 file changed, 8 insertions(+), 27 deletions(-) diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c index 9d8e03b30014e..629442c6bc9cc 100644 --- a/drivers/ufs/core/ufshcd.c +++ b/drivers/ufs/core/ufshcd.c @@ -9134,34 +9134,15 @@ static int ufshcd_execute_start_stop(struct scsi_device *sdev, enum ufs_dev_pwr_mode pwr_mode, struct scsi_sense_hdr *sshdr) { - unsigned char cdb[6] = { START_STOP, 0, 0, 0, pwr_mode << 4, 0 }; - struct request *req; - struct scsi_cmnd *scmd; - int ret; - - req = scsi_alloc_request(sdev->request_queue, REQ_OP_DRV_IN, - BLK_MQ_REQ_PM); - if (IS_ERR(req)) - return PTR_ERR(req); - - scmd = blk_mq_rq_to_pdu(req); - scmd->cmd_len = COMMAND_SIZE(cdb[0]); - memcpy(scmd->cmnd, cdb, scmd->cmd_len); - scmd->allowed = 0/*retries*/; - scmd->flags |= SCMD_FAIL_IF_RECOVERING; - req->timeout = 1 * HZ; - req->rq_flags |= RQF_QUIET; - - blk_execute_rq(req, /*at_head=*/true); - - if (sshdr) - scsi_normalize_sense(scmd->sense_buffer, scmd->sense_len, - sshdr); - ret = scmd->result; - - blk_mq_free_request(req); + const unsigned char cdb[6] = { START_STOP, 0, 0, 0, pwr_mode << 4, 0 }; + const struct scsi_exec_args args = { + .sshdr = sshdr, + .req_flags = BLK_MQ_REQ_PM, + .scmd_flags = SCMD_FAIL_IF_RECOVERING, + }; - return ret; + return scsi_execute_cmd(sdev, cdb, REQ_OP_DRV_IN, /*buffer=*/NULL, + /*bufflen=*/0, /*timeout=*/HZ, /*retries=*/0, &args); } /** From 79f9abd64719cc71ba78a76574e21dc8266c65a3 Mon Sep 17 00:00:00 2001 From: Benjamin Block Date: Tue, 21 Feb 2023 18:55:58 +0100 Subject: [PATCH 372/765] scsi: zfcp: Make the type for accessing request hashtable buckets size_t The appropriate type for array indices is 'size_t' and the current implementation in 'zfcp_reqlist.h' mixes 'int' and 'unsigned int' in different places to access the hashtable buckets of our internal request hash table. To prevent any confusion, change all places to 'size_t'. Link: https://lore.kernel.org/r/64afe93f6263c6b07815937826cd7d5fc4f1a674.1677000450.git.bblock@linux.ibm.com Signed-off-by: Benjamin Block Reviewed-by: Steffen Maier Signed-off-by: Martin K. Petersen --- drivers/s390/scsi/zfcp_reqlist.h | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) diff --git a/drivers/s390/scsi/zfcp_reqlist.h b/drivers/s390/scsi/zfcp_reqlist.h index 9b8ff249e31ca..f4bac61dfbd07 100644 --- a/drivers/s390/scsi/zfcp_reqlist.h +++ b/drivers/s390/scsi/zfcp_reqlist.h @@ -5,14 +5,16 @@ * Data structure and helper functions for tracking pending FSF * requests. * - * Copyright IBM Corp. 2009, 2016 + * Copyright IBM Corp. 2009, 2023 */ #ifndef ZFCP_REQLIST_H #define ZFCP_REQLIST_H +#include + /* number of hash buckets */ -#define ZFCP_REQ_LIST_BUCKETS 128 +#define ZFCP_REQ_LIST_BUCKETS 128u /** * struct zfcp_reqlist - Container for request list (reqlist) @@ -24,7 +26,7 @@ struct zfcp_reqlist { struct list_head buckets[ZFCP_REQ_LIST_BUCKETS]; }; -static inline int zfcp_reqlist_hash(unsigned long req_id) +static inline size_t zfcp_reqlist_hash(unsigned long req_id) { return req_id % ZFCP_REQ_LIST_BUCKETS; } @@ -37,7 +39,7 @@ static inline int zfcp_reqlist_hash(unsigned long req_id) */ static inline struct zfcp_reqlist *zfcp_reqlist_alloc(void) { - unsigned int i; + size_t i; struct zfcp_reqlist *rl; rl = kzalloc(sizeof(struct zfcp_reqlist), GFP_KERNEL); @@ -60,7 +62,7 @@ static inline struct zfcp_reqlist *zfcp_reqlist_alloc(void) */ static inline int zfcp_reqlist_isempty(struct zfcp_reqlist *rl) { - unsigned int i; + size_t i; for (i = 0; i < ZFCP_REQ_LIST_BUCKETS; i++) if (!list_empty(&rl->buckets[i])) @@ -84,7 +86,7 @@ static inline struct zfcp_fsf_req * _zfcp_reqlist_find(struct zfcp_reqlist *rl, unsigned long req_id) { struct zfcp_fsf_req *req; - unsigned int i; + size_t i; i = zfcp_reqlist_hash(req_id); list_for_each_entry(req, &rl->buckets[i], list) @@ -154,7 +156,7 @@ zfcp_reqlist_find_rm(struct zfcp_reqlist *rl, unsigned long req_id) static inline void zfcp_reqlist_add(struct zfcp_reqlist *rl, struct zfcp_fsf_req *req) { - unsigned int i; + size_t i; unsigned long flags; i = zfcp_reqlist_hash(req->req_id); @@ -172,7 +174,7 @@ static inline void zfcp_reqlist_add(struct zfcp_reqlist *rl, static inline void zfcp_reqlist_move(struct zfcp_reqlist *rl, struct list_head *list) { - unsigned int i; + size_t i; unsigned long flags; spin_lock_irqsave(&rl->lock, flags); @@ -200,7 +202,7 @@ zfcp_reqlist_apply_for_all(struct zfcp_reqlist *rl, { struct zfcp_fsf_req *req; unsigned long flags; - unsigned int i; + size_t i; spin_lock_irqsave(&rl->lock, flags); for (i = 0; i < ZFCP_REQ_LIST_BUCKETS; i++) From 3ab01810153b0ca26cfc846ba72996b250adde6d Mon Sep 17 00:00:00 2001 From: Benjamin Block Date: Tue, 21 Feb 2023 18:55:59 +0100 Subject: [PATCH 373/765] scsi: zfcp: Change the type of all fsf request id fields and variables to u64 We use different integer types throughout zfcp to store the FSF request ID and related values; some places use 'unsigned long' and others 'u64'. On s390x these are effectively the same type, but this might cause confusions and is generally inconsistent. The specification for the used hardware specifies this value as a 64-bit number, and ultimately we use this value to communicate with the hardware, so it makes sense to change the type of all these variables to 'u64' where we can. The only exception being when we store it in the 'host_scribble' field of a 'struct scsi_cmnd'; for this case we add a build time check to make sure they are compatible. Link: https://lore.kernel.org/r/9c9cbe5acc2b419a22dce2fed847e3db91b60201.1677000450.git.bblock@linux.ibm.com Signed-off-by: Benjamin Block Reviewed-by: Steffen Maier Signed-off-by: Martin K. Petersen --- drivers/s390/scsi/zfcp_dbf.c | 2 +- drivers/s390/scsi/zfcp_def.h | 6 +++--- drivers/s390/scsi/zfcp_fsf.c | 15 ++++++++------- drivers/s390/scsi/zfcp_qdio.h | 2 +- drivers/s390/scsi/zfcp_reqlist.h | 8 ++++---- drivers/s390/scsi/zfcp_scsi.c | 2 +- 6 files changed, 18 insertions(+), 17 deletions(-) diff --git a/drivers/s390/scsi/zfcp_dbf.c b/drivers/s390/scsi/zfcp_dbf.c index cbc3b62cd9e59..d28c55ae9f015 100644 --- a/drivers/s390/scsi/zfcp_dbf.c +++ b/drivers/s390/scsi/zfcp_dbf.c @@ -649,7 +649,7 @@ void zfcp_dbf_scsi_common(char *tag, int level, struct scsi_device *sdev, rec->scsi_id = sc->device->id; rec->scsi_lun = (u32)sc->device->lun; rec->scsi_lun_64_hi = (u32)(sc->device->lun >> 32); - rec->host_scribble = (unsigned long)sc->host_scribble; + rec->host_scribble = (u64)sc->host_scribble; memcpy(rec->scsi_opcode, sc->cmnd, min_t(int, sc->cmd_len, ZFCP_DBF_SCSI_OPCODE)); diff --git a/drivers/s390/scsi/zfcp_def.h b/drivers/s390/scsi/zfcp_def.h index 94de55304a02b..6c761299a22f4 100644 --- a/drivers/s390/scsi/zfcp_def.h +++ b/drivers/s390/scsi/zfcp_def.h @@ -129,7 +129,7 @@ struct zfcp_erp_action { struct scsi_device *sdev; u32 status; /* recovery status */ enum zfcp_erp_steps step; /* active step of this erp action */ - unsigned long fsf_req_id; + u64 fsf_req_id; struct timer_list timer; }; @@ -163,7 +163,7 @@ struct zfcp_adapter { struct Scsi_Host *scsi_host; /* Pointer to mid-layer */ struct list_head port_list; /* remote port list */ rwlock_t port_list_lock; /* port list lock */ - unsigned long req_no; /* unique FSF req number */ + u64 req_no; /* unique FSF req number */ struct zfcp_reqlist *req_list; u32 fsf_req_seq_no; /* FSF cmnd seq number */ rwlock_t abort_lock; /* Protects against SCSI @@ -325,7 +325,7 @@ static inline u64 zfcp_scsi_dev_lun(struct scsi_device *sdev) */ struct zfcp_fsf_req { struct list_head list; - unsigned long req_id; + u64 req_id; struct zfcp_adapter *adapter; struct zfcp_qdio_req qdio_req; struct completion completion; diff --git a/drivers/s390/scsi/zfcp_fsf.c b/drivers/s390/scsi/zfcp_fsf.c index ab3ea529cca70..71eabcc26cbbf 100644 --- a/drivers/s390/scsi/zfcp_fsf.c +++ b/drivers/s390/scsi/zfcp_fsf.c @@ -884,7 +884,7 @@ static int zfcp_fsf_req_send(struct zfcp_fsf_req *req) const bool is_srb = zfcp_fsf_req_is_status_read_buffer(req); struct zfcp_adapter *adapter = req->adapter; struct zfcp_qdio *qdio = adapter->qdio; - unsigned long req_id = req->req_id; + u64 req_id = req->req_id; zfcp_reqlist_add(adapter->req_list, req); @@ -1042,7 +1042,7 @@ struct zfcp_fsf_req *zfcp_fsf_abort_fcp_cmnd(struct scsi_cmnd *scmnd) struct scsi_device *sdev = scmnd->device; struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(sdev); struct zfcp_qdio *qdio = zfcp_sdev->port->adapter->qdio; - unsigned long old_req_id = (unsigned long) scmnd->host_scribble; + u64 old_req_id = (u64) scmnd->host_scribble; spin_lock_irq(&qdio->req_q_lock); if (zfcp_qdio_sbal_get(qdio)) @@ -1065,7 +1065,7 @@ struct zfcp_fsf_req *zfcp_fsf_abort_fcp_cmnd(struct scsi_cmnd *scmnd) req->handler = zfcp_fsf_abort_fcp_command_handler; req->qtcb->header.lun_handle = zfcp_sdev->lun_handle; req->qtcb->header.port_handle = zfcp_sdev->port->handle; - req->qtcb->bottom.support.req_handle = (u64) old_req_id; + req->qtcb->bottom.support.req_handle = old_req_id; zfcp_fsf_start_timer(req, ZFCP_FSF_SCSI_ER_TIMEOUT); if (!zfcp_fsf_req_send(req)) { @@ -1919,7 +1919,7 @@ int zfcp_fsf_open_wka_port(struct zfcp_fc_wka_port *wka_port) { struct zfcp_qdio *qdio = wka_port->adapter->qdio; struct zfcp_fsf_req *req; - unsigned long req_id = 0; + u64 req_id = 0; int retval = -EIO; spin_lock_irq(&qdio->req_q_lock); @@ -1978,7 +1978,7 @@ int zfcp_fsf_close_wka_port(struct zfcp_fc_wka_port *wka_port) { struct zfcp_qdio *qdio = wka_port->adapter->qdio; struct zfcp_fsf_req *req; - unsigned long req_id = 0; + u64 req_id = 0; int retval = -EIO; spin_lock_irq(&qdio->req_q_lock); @@ -2587,6 +2587,7 @@ int zfcp_fsf_fcp_cmnd(struct scsi_cmnd *scsi_cmnd) goto out; } + BUILD_BUG_ON(sizeof(scsi_cmnd->host_scribble) < sizeof(req->req_id)); scsi_cmnd->host_scribble = (unsigned char *) req->req_id; io = &req->qtcb->bottom.io; @@ -2732,7 +2733,7 @@ void zfcp_fsf_reqid_check(struct zfcp_qdio *qdio, int sbal_idx) struct qdio_buffer *sbal = qdio->res_q[sbal_idx]; struct qdio_buffer_element *sbale; struct zfcp_fsf_req *fsf_req; - unsigned long req_id; + u64 req_id; int idx; for (idx = 0; idx < QDIO_MAX_ELEMENTS_PER_BUFFER; idx++) { @@ -2747,7 +2748,7 @@ void zfcp_fsf_reqid_check(struct zfcp_qdio *qdio, int sbal_idx) * corruption and must stop the machine immediately. */ zfcp_qdio_siosl(adapter); - panic("error: unknown req_id (%lx) on adapter %s.\n", + panic("error: unknown req_id (%llx) on adapter %s.\n", req_id, dev_name(&adapter->ccw_device->dev)); } diff --git a/drivers/s390/scsi/zfcp_qdio.h b/drivers/s390/scsi/zfcp_qdio.h index 390706867df35..90134d9b69a77 100644 --- a/drivers/s390/scsi/zfcp_qdio.h +++ b/drivers/s390/scsi/zfcp_qdio.h @@ -115,7 +115,7 @@ zfcp_qdio_sbale_curr(struct zfcp_qdio *qdio, struct zfcp_qdio_req *q_req) */ static inline void zfcp_qdio_req_init(struct zfcp_qdio *qdio, struct zfcp_qdio_req *q_req, - unsigned long req_id, u8 sbtype, void *data, u32 len) + u64 req_id, u8 sbtype, void *data, u32 len) { struct qdio_buffer_element *sbale; int count = min(atomic_read(&qdio->req_q_free), diff --git a/drivers/s390/scsi/zfcp_reqlist.h b/drivers/s390/scsi/zfcp_reqlist.h index f4bac61dfbd07..59fbb1b128cb9 100644 --- a/drivers/s390/scsi/zfcp_reqlist.h +++ b/drivers/s390/scsi/zfcp_reqlist.h @@ -26,7 +26,7 @@ struct zfcp_reqlist { struct list_head buckets[ZFCP_REQ_LIST_BUCKETS]; }; -static inline size_t zfcp_reqlist_hash(unsigned long req_id) +static inline size_t zfcp_reqlist_hash(u64 req_id) { return req_id % ZFCP_REQ_LIST_BUCKETS; } @@ -83,7 +83,7 @@ static inline void zfcp_reqlist_free(struct zfcp_reqlist *rl) } static inline struct zfcp_fsf_req * -_zfcp_reqlist_find(struct zfcp_reqlist *rl, unsigned long req_id) +_zfcp_reqlist_find(struct zfcp_reqlist *rl, u64 req_id) { struct zfcp_fsf_req *req; size_t i; @@ -104,7 +104,7 @@ _zfcp_reqlist_find(struct zfcp_reqlist *rl, unsigned long req_id) * or NULL if there is no known FSF request with this id. */ static inline struct zfcp_fsf_req * -zfcp_reqlist_find(struct zfcp_reqlist *rl, unsigned long req_id) +zfcp_reqlist_find(struct zfcp_reqlist *rl, u64 req_id) { unsigned long flags; struct zfcp_fsf_req *req; @@ -129,7 +129,7 @@ zfcp_reqlist_find(struct zfcp_reqlist *rl, unsigned long req_id) * NULL if it has not been found. */ static inline struct zfcp_fsf_req * -zfcp_reqlist_find_rm(struct zfcp_reqlist *rl, unsigned long req_id) +zfcp_reqlist_find_rm(struct zfcp_reqlist *rl, u64 req_id) { unsigned long flags; struct zfcp_fsf_req *req; diff --git a/drivers/s390/scsi/zfcp_scsi.c b/drivers/s390/scsi/zfcp_scsi.c index 526ac240d9fe8..3dbf4b21d1276 100644 --- a/drivers/s390/scsi/zfcp_scsi.c +++ b/drivers/s390/scsi/zfcp_scsi.c @@ -170,7 +170,7 @@ static int zfcp_scsi_eh_abort_handler(struct scsi_cmnd *scpnt) (struct zfcp_adapter *) scsi_host->hostdata[0]; struct zfcp_fsf_req *old_req, *abrt_req; unsigned long flags; - unsigned long old_reqid = (unsigned long) scpnt->host_scribble; + u64 old_reqid = (u64) scpnt->host_scribble; int retval = SUCCESS, ret; int retry = 3; char *dbf_tag; From 901b894af5b933cf6576eec05746f34b46e2ac83 Mon Sep 17 00:00:00 2001 From: Benjamin Block Date: Tue, 21 Feb 2023 18:56:00 +0100 Subject: [PATCH 374/765] scsi: zfcp: Trace when request remove fails after qdio send fails When we fail to send a FSF request in 'zfcp_fsf_req_send()' when calling 'zfcp_qdio_send()' we try to remove the request object from our internal hash table again to prevent keeping a stale memory reference. This removal might still - very much theoretically - fail. To store some evidence of when this happens add a new trace record for this case; tag: 'fsrsrmf'. We reuse the 'ZFCP_DBF_HBA_RES' trace ID for this, but mark all fields other then the request ID with ~0, to make fairly obvious that these are invalid values. This faking has to be done because we don't have a valid request object at this point, and can not safely access any of the memory of the old object - we just failed to find it in our hash table, so it might be gone already. Here is an example of a decoded trace record: Timestamp : 2023-02-17-13:09:12:748140 Area : HBA Subarea : 1 Level : - Exception : 000003ff7ff500c2 CPU ID : 0011 Caller : 0x0 Record ID : 1 Tag : fsrsrmf Request ID : 0x0000000080126ab6 Request status : 0xffffffff FSF cmnd : 0xffffffff FSF sequence no: 0xffffffff FSF issued : 2042-09-18-01:53:47:370495 FSF stat : 0xffffffff FSF stat qual : ffffffff ffffffff ffffffff ffffffff Prot stat : 0xffffffff Prot stat qual : ffffffff ffffffff ffffffff ffffffff Port handle : 0xffffffff LUN handle : 0xffffffff This provides at least some basic evidence that this event happened, and what object was affected. Link: https://lore.kernel.org/r/99b8246b2d71b63fa4f9c56333e2037502f7f5af.1677000450.git.bblock@linux.ibm.com Signed-off-by: Benjamin Block Reviewed-by: Steffen Maier Signed-off-by: Martin K. Petersen --- drivers/s390/scsi/zfcp_dbf.c | 44 +++++++++++++++++++++++++++++++++++- drivers/s390/scsi/zfcp_ext.h | 5 +++- drivers/s390/scsi/zfcp_fsf.c | 7 ++++-- 3 files changed, 52 insertions(+), 4 deletions(-) diff --git a/drivers/s390/scsi/zfcp_dbf.c b/drivers/s390/scsi/zfcp_dbf.c index d28c55ae9f015..d904625afd408 100644 --- a/drivers/s390/scsi/zfcp_dbf.c +++ b/drivers/s390/scsi/zfcp_dbf.c @@ -4,7 +4,7 @@ * * Debug traces for zfcp. * - * Copyright IBM Corp. 2002, 2020 + * Copyright IBM Corp. 2002, 2023 */ #define KMSG_COMPONENT "zfcp" @@ -145,6 +145,48 @@ void zfcp_dbf_hba_fsf_fces(char *tag, const struct zfcp_fsf_req *req, u64 wwpn, spin_unlock_irqrestore(&dbf->hba_lock, flags); } +/** + * zfcp_dbf_hba_fsf_reqid - trace only the tag and a request ID + * @tag: tag documenting the source + * @level: trace level + * @adapter: adapter instance the request ID belongs to + * @req_id: the request ID to trace + */ +void zfcp_dbf_hba_fsf_reqid(const char *const tag, const int level, + struct zfcp_adapter *const adapter, + const u64 req_id) +{ + struct zfcp_dbf *const dbf = adapter->dbf; + struct zfcp_dbf_hba *const rec = &dbf->hba_buf; + struct zfcp_dbf_hba_res *const res = &rec->u.res; + unsigned long flags; + + if (unlikely(!debug_level_enabled(dbf->hba, level))) + return; + + spin_lock_irqsave(&dbf->hba_lock, flags); + memset(rec, 0, sizeof(*rec)); + + memcpy(rec->tag, tag, ZFCP_DBF_TAG_LEN); + + rec->id = ZFCP_DBF_HBA_RES; + rec->fsf_req_id = req_id; + rec->fsf_req_status = ~0u; + rec->fsf_cmd = ~0u; + rec->fsf_seq_no = ~0u; + + res->req_issued = ~0ull; + res->prot_status = ~0u; + memset(res->prot_status_qual, 0xff, sizeof(res->prot_status_qual)); + res->fsf_status = ~0u; + memset(res->fsf_status_qual, 0xff, sizeof(res->fsf_status_qual)); + res->port_handle = ~0u; + res->lun_handle = ~0u; + + debug_event(dbf->hba, level, rec, sizeof(*rec)); + spin_unlock_irqrestore(&dbf->hba_lock, flags); +} + /** * zfcp_dbf_hba_fsf_uss - trace event for an unsolicited status buffer * @tag: tag indicating which kind of unsolicited status has been received diff --git a/drivers/s390/scsi/zfcp_ext.h b/drivers/s390/scsi/zfcp_ext.h index c302cbb18a55c..9f5152b42b0ea 100644 --- a/drivers/s390/scsi/zfcp_ext.h +++ b/drivers/s390/scsi/zfcp_ext.h @@ -4,7 +4,7 @@ * * External function declarations. * - * Copyright IBM Corp. 2002, 2020 + * Copyright IBM Corp. 2002, 2023 */ #ifndef ZFCP_EXT_H @@ -46,6 +46,9 @@ extern void zfcp_dbf_hba_fsf_res(char *, int, struct zfcp_fsf_req *); extern void zfcp_dbf_hba_fsf_fces(char *tag, const struct zfcp_fsf_req *req, u64 wwpn, u32 fc_security_old, u32 fc_security_new); +extern void zfcp_dbf_hba_fsf_reqid(const char *const tag, const int level, + struct zfcp_adapter *const adapter, + const u64 req_id); extern void zfcp_dbf_hba_bit_err(char *, struct zfcp_fsf_req *); extern void zfcp_dbf_hba_def_err(struct zfcp_adapter *, u64, u16, void **); extern void zfcp_dbf_san_req(char *, struct zfcp_fsf_req *, u32); diff --git a/drivers/s390/scsi/zfcp_fsf.c b/drivers/s390/scsi/zfcp_fsf.c index 71eabcc26cbbf..ceed1b6f7cb61 100644 --- a/drivers/s390/scsi/zfcp_fsf.c +++ b/drivers/s390/scsi/zfcp_fsf.c @@ -4,7 +4,7 @@ * * Implementation of FSF commands. * - * Copyright IBM Corp. 2002, 2020 + * Copyright IBM Corp. 2002, 2023 */ #define KMSG_COMPONENT "zfcp" @@ -892,8 +892,11 @@ static int zfcp_fsf_req_send(struct zfcp_fsf_req *req) req->issued = get_tod_clock(); if (zfcp_qdio_send(qdio, &req->qdio_req)) { del_timer_sync(&req->timer); + /* lookup request again, list might have changed */ - zfcp_reqlist_find_rm(adapter->req_list, req_id); + if (zfcp_reqlist_find_rm(adapter->req_list, req_id) == NULL) + zfcp_dbf_hba_fsf_reqid("fsrsrmf", 1, adapter, req_id); + zfcp_erp_adapter_reopen(adapter, 0, "fsrs__1"); return -EIO; } From eb7b85853c3866236f9cb378fc68ce5f76efbf9c Mon Sep 17 00:00:00 2001 From: Jacky Bai Date: Wed, 15 Feb 2023 10:41:16 +0800 Subject: [PATCH 375/765] rtc: bbnsm: Add the bbnsm rtc support The BBNSM module includes a real time counter with alarm. Add a RTC driver for this function. Signed-off-by: Jacky Bai Reviewed-by: Peng Fan Link: https://lore.kernel.org/r/20230215024117.3357341-3-ping.bai@nxp.com Signed-off-by: Alexandre Belloni --- drivers/rtc/Kconfig | 12 ++ drivers/rtc/Makefile | 1 + drivers/rtc/rtc-nxp-bbnsm.c | 226 ++++++++++++++++++++++++++++++++++++ 3 files changed, 239 insertions(+) create mode 100644 drivers/rtc/rtc-nxp-bbnsm.c diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig index d2b6d20a67454..d2a7fbc2c9324 100644 --- a/drivers/rtc/Kconfig +++ b/drivers/rtc/Kconfig @@ -1786,6 +1786,18 @@ config RTC_DRV_SNVS This driver can also be built as a module, if so, the module will be called "rtc-snvs". +config RTC_DRV_BBNSM + tristate "NXP BBNSM RTC support" + select REGMAP_MMIO + depends on ARCH_MXC || COMPILE_TEST + depends on HAS_IOMEM + depends on OF + help + If you say yes here you get support for the NXP BBNSM RTC module. + + This driver can also be built as a module, if so, the module + will be called "rtc-bbnsm". + config RTC_DRV_IMX_SC depends on IMX_SCU depends on HAVE_ARM_SMCCC diff --git a/drivers/rtc/Makefile b/drivers/rtc/Makefile index d3c042dcbc733..0f11027a73880 100644 --- a/drivers/rtc/Makefile +++ b/drivers/rtc/Makefile @@ -33,6 +33,7 @@ obj-$(CONFIG_RTC_DRV_ASPEED) += rtc-aspeed.o obj-$(CONFIG_RTC_DRV_AT91RM9200)+= rtc-at91rm9200.o obj-$(CONFIG_RTC_DRV_AT91SAM9) += rtc-at91sam9.o obj-$(CONFIG_RTC_DRV_AU1XXX) += rtc-au1xxx.o +obj-$(CONFIG_RTC_DRV_BBNSM) += rtc-nxp-bbnsm.o obj-$(CONFIG_RTC_DRV_BD70528) += rtc-bd70528.o obj-$(CONFIG_RTC_DRV_BQ32K) += rtc-bq32k.o obj-$(CONFIG_RTC_DRV_BQ4802) += rtc-bq4802.o diff --git a/drivers/rtc/rtc-nxp-bbnsm.c b/drivers/rtc/rtc-nxp-bbnsm.c new file mode 100644 index 0000000000000..acbfbeb8b0700 --- /dev/null +++ b/drivers/rtc/rtc-nxp-bbnsm.c @@ -0,0 +1,226 @@ +// SPDX-License-Identifier: GPL-2.0+ +// +// Copyright 2022 NXP. + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define BBNSM_CTRL 0x8 +#define BBNSM_INT_EN 0x10 +#define BBNSM_EVENTS 0x14 +#define BBNSM_RTC_LS 0x40 +#define BBNSM_RTC_MS 0x44 +#define BBNSM_TA 0x50 + +#define RTC_EN 0x2 +#define RTC_EN_MSK 0x3 +#define TA_EN (0x2 << 2) +#define TA_DIS (0x1 << 2) +#define TA_EN_MSK (0x3 << 2) +#define RTC_INT_EN 0x2 +#define TA_INT_EN (0x2 << 2) + +#define BBNSM_EVENT_TA (0x2 << 2) + +#define CNTR_TO_SECS_SH 15 + +struct bbnsm_rtc { + struct rtc_device *rtc; + struct regmap *regmap; + int irq; + struct clk *clk; +}; + +static u32 bbnsm_read_counter(struct bbnsm_rtc *bbnsm) +{ + u32 rtc_msb, rtc_lsb; + unsigned int timeout = 100; + u32 time; + u32 tmp = 0; + + do { + time = tmp; + /* read the msb */ + regmap_read(bbnsm->regmap, BBNSM_RTC_MS, &rtc_msb); + /* read the lsb */ + regmap_read(bbnsm->regmap, BBNSM_RTC_LS, &rtc_lsb); + /* convert to seconds */ + tmp = (rtc_msb << 17) | (rtc_lsb >> 15); + } while (tmp != time && --timeout); + + return time; +} + +static int bbnsm_rtc_read_time(struct device *dev, struct rtc_time *tm) +{ + struct bbnsm_rtc *bbnsm = dev_get_drvdata(dev); + unsigned long time; + u32 val; + + regmap_read(bbnsm->regmap, BBNSM_CTRL, &val); + if ((val & RTC_EN_MSK) != RTC_EN) + return -EINVAL; + + time = bbnsm_read_counter(bbnsm); + rtc_time64_to_tm(time, tm); + + return 0; +} + +static int bbnsm_rtc_set_time(struct device *dev, struct rtc_time *tm) +{ + struct bbnsm_rtc *bbnsm = dev_get_drvdata(dev); + unsigned long time = rtc_tm_to_time64(tm); + + /* disable the RTC first */ + regmap_update_bits(bbnsm->regmap, BBNSM_CTRL, RTC_EN_MSK, 0); + + /* write the 32bit sec time to 47 bit timer counter, leaving 15 LSBs blank */ + regmap_write(bbnsm->regmap, BBNSM_RTC_LS, time << CNTR_TO_SECS_SH); + regmap_write(bbnsm->regmap, BBNSM_RTC_MS, time >> (32 - CNTR_TO_SECS_SH)); + + /* Enable the RTC again */ + regmap_update_bits(bbnsm->regmap, BBNSM_CTRL, RTC_EN_MSK, RTC_EN); + + return 0; +} + +static int bbnsm_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm) +{ + struct bbnsm_rtc *bbnsm = dev_get_drvdata(dev); + u32 bbnsm_events, bbnsm_ta; + + regmap_read(bbnsm->regmap, BBNSM_TA, &bbnsm_ta); + rtc_time64_to_tm(bbnsm_ta, &alrm->time); + + regmap_read(bbnsm->regmap, BBNSM_EVENTS, &bbnsm_events); + alrm->pending = (bbnsm_events & BBNSM_EVENT_TA) ? 1 : 0; + + return 0; +} + +static int bbnsm_rtc_alarm_irq_enable(struct device *dev, unsigned int enable) +{ + struct bbnsm_rtc *bbnsm = dev_get_drvdata(dev); + + /* enable the alarm event */ + regmap_update_bits(bbnsm->regmap, BBNSM_CTRL, TA_EN_MSK, enable ? TA_EN : TA_DIS); + /* enable the alarm interrupt */ + regmap_update_bits(bbnsm->regmap, BBNSM_INT_EN, TA_EN_MSK, enable ? TA_EN : TA_DIS); + + return 0; +} + +static int bbnsm_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alrm) +{ + struct bbnsm_rtc *bbnsm = dev_get_drvdata(dev); + unsigned long time = rtc_tm_to_time64(&alrm->time); + + /* disable the alarm */ + regmap_update_bits(bbnsm->regmap, BBNSM_CTRL, TA_EN, TA_EN); + + /* write the seconds to TA */ + regmap_write(bbnsm->regmap, BBNSM_TA, time); + + return bbnsm_rtc_alarm_irq_enable(dev, alrm->enabled); +} + +static const struct rtc_class_ops bbnsm_rtc_ops = { + .read_time = bbnsm_rtc_read_time, + .set_time = bbnsm_rtc_set_time, + .read_alarm = bbnsm_rtc_read_alarm, + .set_alarm = bbnsm_rtc_set_alarm, + .alarm_irq_enable = bbnsm_rtc_alarm_irq_enable, +}; + +static irqreturn_t bbnsm_rtc_irq_handler(int irq, void *dev_id) +{ + struct device *dev = dev_id; + struct bbnsm_rtc *bbnsm = dev_get_drvdata(dev); + u32 val; + + regmap_read(bbnsm->regmap, BBNSM_EVENTS, &val); + if (val & BBNSM_EVENT_TA) { + bbnsm_rtc_alarm_irq_enable(dev, false); + /* clear the alarm event */ + regmap_write_bits(bbnsm->regmap, BBNSM_EVENTS, TA_EN_MSK, BBNSM_EVENT_TA); + rtc_update_irq(bbnsm->rtc, 1, RTC_AF | RTC_IRQF); + + return IRQ_HANDLED; + } + + return IRQ_NONE; +} + +static int bbnsm_rtc_probe(struct platform_device *pdev) +{ + struct device_node *np = pdev->dev.of_node; + struct bbnsm_rtc *bbnsm; + int ret; + + bbnsm = devm_kzalloc(&pdev->dev, sizeof(*bbnsm), GFP_KERNEL); + if (!bbnsm) + return -ENOMEM; + + bbnsm->rtc = devm_rtc_allocate_device(&pdev->dev); + if (IS_ERR(bbnsm->rtc)) + return PTR_ERR(bbnsm->rtc); + + bbnsm->regmap = syscon_node_to_regmap(np->parent); + if (IS_ERR(bbnsm->regmap)) { + dev_dbg(&pdev->dev, "bbnsm get regmap failed\n"); + return PTR_ERR(bbnsm->regmap); + } + + bbnsm->irq = platform_get_irq(pdev, 0); + if (bbnsm->irq < 0) + return bbnsm->irq; + + platform_set_drvdata(pdev, bbnsm); + + /* clear all the pending events */ + regmap_write(bbnsm->regmap, BBNSM_EVENTS, 0x7A); + + device_init_wakeup(&pdev->dev, true); + dev_pm_set_wake_irq(&pdev->dev, bbnsm->irq); + + ret = devm_request_irq(&pdev->dev, bbnsm->irq, bbnsm_rtc_irq_handler, + IRQF_SHARED, "rtc alarm", &pdev->dev); + if (ret) { + dev_err(&pdev->dev, "failed to request irq %d: %d\n", + bbnsm->irq, ret); + return ret; + } + + bbnsm->rtc->ops = &bbnsm_rtc_ops; + bbnsm->rtc->range_max = U32_MAX; + + return devm_rtc_register_device(bbnsm->rtc); +} + +static const struct of_device_id bbnsm_dt_ids[] = { + { .compatible = "nxp,imx93-bbnsm-rtc" }, + { /* sentinel */ }, +}; +MODULE_DEVICE_TABLE(of, bbnsm_dt_ids); + +static struct platform_driver bbnsm_rtc_driver = { + .driver = { + .name = "bbnsm_rtc", + .of_match_table = bbnsm_dt_ids, + }, + .probe = bbnsm_rtc_probe, +}; +module_platform_driver(bbnsm_rtc_driver); + +MODULE_AUTHOR("Jacky Bai "); +MODULE_DESCRIPTION("NXP BBNSM RTC Driver"); +MODULE_LICENSE("GPL"); From b6ef5d4a0295bc0567021c0102d1c66f2ed212a2 Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Tue, 14 Feb 2023 21:26:53 +0100 Subject: [PATCH 376/765] rtc: rv3028: add ACPI support The RV-3028 has been assigned the MCRY3028 ACPI ID. Link: https://lore.kernel.org/r/20230214202653.565647-1-alexandre.belloni@bootlin.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-rv3028.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/rtc/rtc-rv3028.c b/drivers/rtc/rtc-rv3028.c index b0099e26e3b05..ec5d7a614e2dd 100644 --- a/drivers/rtc/rtc-rv3028.c +++ b/drivers/rtc/rtc-rv3028.c @@ -982,6 +982,12 @@ static int rv3028_probe(struct i2c_client *client) return 0; } +static const struct acpi_device_id rv3028_i2c_acpi_match[] = { + { "MCRY3028" }, + { } +}; +MODULE_DEVICE_TABLE(acpi, rv3028_i2c_acpi_match); + static const __maybe_unused struct of_device_id rv3028_of_match[] = { { .compatible = "microcrystal,rv3028", }, { } @@ -991,6 +997,7 @@ MODULE_DEVICE_TABLE(of, rv3028_of_match); static struct i2c_driver rv3028_driver = { .driver = { .name = "rtc-rv3028", + .acpi_match_table = rv3028_i2c_acpi_match, .of_match_table = of_match_ptr(rv3028_of_match), }, .probe_new = rv3028_probe, From 2d433e9c897dcb0b70ef17b536f61430b1e151d4 Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Tue, 14 Feb 2023 21:27:15 +0100 Subject: [PATCH 377/765] rtc: rv3032: add ACPI support The RV-3032 has been assigned the MCRY3032 ACPI ID. Link: https://lore.kernel.org/r/20230214202716.565749-1-alexandre.belloni@bootlin.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-rv3032.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/rtc/rtc-rv3032.c b/drivers/rtc/rtc-rv3032.c index bf6954ec5943e..1ff4f2e6fa77e 100644 --- a/drivers/rtc/rtc-rv3032.c +++ b/drivers/rtc/rtc-rv3032.c @@ -980,6 +980,12 @@ static int rv3032_probe(struct i2c_client *client) return 0; } +static const struct acpi_device_id rv3032_i2c_acpi_match[] = { + { "MCRY3032" }, + { } +}; +MODULE_DEVICE_TABLE(acpi, rv3032_i2c_acpi_match); + static const __maybe_unused struct of_device_id rv3032_of_match[] = { { .compatible = "microcrystal,rv3032", }, { } @@ -989,6 +995,7 @@ MODULE_DEVICE_TABLE(of, rv3032_of_match); static struct i2c_driver rv3032_driver = { .driver = { .name = "rtc-rv3032", + .acpi_match_table = rv3032_i2c_acpi_match, .of_match_table = of_match_ptr(rv3032_of_match), }, .probe_new = rv3032_probe, From a783c962619271a8b905efad1d89adfec11ae0c8 Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Tue, 14 Feb 2023 23:27:53 +0100 Subject: [PATCH 378/765] rtc: allow rtc_read_alarm without read_alarm callback .read_alarm is not necessary to read the current alarm because it is recorded in the aie_timer and so rtc_read_alarm() will never call rtc_read_alarm_internal() which is the only function calling the callback. Reported-by: Zhipeng Wang Reported-by: Marcel Ziswiler Fixes: 7ae41220ef58 ("rtc: introduce features bitfield") Tested-by: Philippe Schenker Link: https://lore.kernel.org/r/20230214222754.582582-1-alexandre.belloni@bootlin.com Signed-off-by: Alexandre Belloni --- drivers/rtc/interface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/rtc/interface.c b/drivers/rtc/interface.c index 7c30cb3c764d8..499d89150afc9 100644 --- a/drivers/rtc/interface.c +++ b/drivers/rtc/interface.c @@ -392,7 +392,7 @@ int rtc_read_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm) return err; if (!rtc->ops) { err = -ENODEV; - } else if (!test_bit(RTC_FEATURE_ALARM, rtc->features) || !rtc->ops->read_alarm) { + } else if (!test_bit(RTC_FEATURE_ALARM, rtc->features)) { err = -EINVAL; } else { memset(alarm, 0, sizeof(struct rtc_wkalrm)); From 1b2f85a8bac67b9909f2ee4be1bc11548a7aeaf3 Mon Sep 17 00:00:00 2001 From: Javier Carrasco Date: Wed, 15 Feb 2023 09:18:14 +0100 Subject: [PATCH 379/765] dt-bindings: rtc: nxp,pcf8563: move pcf85263/pcf85363 to a dedicated binding These Real Time Clocks are managed by the rtc-pcf85363 device driver, which now supports the quartz-load-femtofarads property. Signed-off-by: Javier Carrasco Reviewed-by: Krzysztof Kozlowski Link: https://lore.kernel.org/r/20230215081815.3141776-2-javier.carrasco@wolfvision.net Signed-off-by: Alexandre Belloni --- .../devicetree/bindings/rtc/nxp,pcf85363.yaml | 60 +++++++++++++++++++ .../devicetree/bindings/rtc/nxp,pcf8563.yaml | 2 - 2 files changed, 60 insertions(+), 2 deletions(-) create mode 100644 Documentation/devicetree/bindings/rtc/nxp,pcf85363.yaml diff --git a/Documentation/devicetree/bindings/rtc/nxp,pcf85363.yaml b/Documentation/devicetree/bindings/rtc/nxp,pcf85363.yaml new file mode 100644 index 0000000000000..52aa3e2091e92 --- /dev/null +++ b/Documentation/devicetree/bindings/rtc/nxp,pcf85363.yaml @@ -0,0 +1,60 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/rtc/nxp,pcf85363.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Philips PCF85263/PCF85363 Real Time Clock + +maintainers: + - Alexandre Belloni + +allOf: + - $ref: rtc.yaml# + +properties: + compatible: + enum: + - nxp,pcf85263 + - nxp,pcf85363 + + reg: + maxItems: 1 + + "#clock-cells": + const: 0 + + clock-output-names: + maxItems: 1 + + interrupts: + maxItems: 1 + + quartz-load-femtofarads: + description: + The capacitive load of the quartz(x-tal). + enum: [6000, 7000, 12500] + default: 7000 + + start-year: true + wakeup-source: true + +required: + - compatible + - reg + +additionalProperties: false + +examples: + - | + i2c { + #address-cells = <1>; + #size-cells = <0>; + + rtc@51 { + compatible = "nxp,pcf85363"; + reg = <0x51>; + #clock-cells = <0>; + quartz-load-femtofarads = <12500>; + }; + }; diff --git a/Documentation/devicetree/bindings/rtc/nxp,pcf8563.yaml b/Documentation/devicetree/bindings/rtc/nxp,pcf8563.yaml index a98b72752349f..22909a96123e5 100644 --- a/Documentation/devicetree/bindings/rtc/nxp,pcf8563.yaml +++ b/Documentation/devicetree/bindings/rtc/nxp,pcf8563.yaml @@ -19,8 +19,6 @@ properties: - microcrystal,rv8564 - nxp,pca8565 - nxp,pcf8563 - - nxp,pcf85263 - - nxp,pcf85363 reg: maxItems: 1 From fd9a6a13949af81062f4cd04f2c1b28ca5311e71 Mon Sep 17 00:00:00 2001 From: Javier Carrasco Date: Wed, 15 Feb 2023 09:18:15 +0100 Subject: [PATCH 380/765] rtc: pcf85363: add support for the quartz-load-femtofarads property The quartz oscillator load capacitance of the PCF85263 and PCF85363 can be adjusted to 6 pF, 7 pF (default) and 12.5 pF with the CL[1:0] bits in the oscillator control register (address 25h). Signed-off-by: Javier Carrasco Link: https://lore.kernel.org/r/20230215081815.3141776-3-javier.carrasco@wolfvision.net Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pcf85363.c | 37 ++++++++++++++++++++++++++++++++++++- 1 file changed, 36 insertions(+), 1 deletion(-) diff --git a/drivers/rtc/rtc-pcf85363.c b/drivers/rtc/rtc-pcf85363.c index 5de323acd1786..8958eadf1c3ef 100644 --- a/drivers/rtc/rtc-pcf85363.c +++ b/drivers/rtc/rtc-pcf85363.c @@ -101,6 +101,10 @@ #define PIN_IO_INTA_OUT 2 #define PIN_IO_INTA_HIZ 3 +#define OSC_CAP_SEL GENMASK(1, 0) +#define OSC_CAP_6000 0x01 +#define OSC_CAP_12500 0x02 + #define STOP_EN_STOP BIT(0) #define RESET_CPR 0xa4 @@ -117,6 +121,32 @@ struct pcf85x63_config { unsigned int num_nvram; }; +static int pcf85363_load_capacitance(struct pcf85363 *pcf85363, struct device_node *node) +{ + u32 load = 7000; + u8 value = 0; + + of_property_read_u32(node, "quartz-load-femtofarads", &load); + + switch (load) { + default: + dev_warn(&pcf85363->rtc->dev, "Unknown quartz-load-femtofarads value: %d. Assuming 7000", + load); + fallthrough; + case 7000: + break; + case 6000: + value = OSC_CAP_6000; + break; + case 12500: + value = OSC_CAP_12500; + break; + } + + return regmap_update_bits(pcf85363->regmap, CTRL_OSCILLATOR, + OSC_CAP_SEL, value); +} + static int pcf85363_rtc_read_time(struct device *dev, struct rtc_time *tm) { struct pcf85363 *pcf85363 = dev_get_drvdata(dev); @@ -372,7 +402,7 @@ static int pcf85363_probe(struct i2c_client *client) .reg_write = pcf85363_nvram_write, }, }; - int ret, i; + int ret, i, err; if (data) config = data; @@ -394,6 +424,11 @@ static int pcf85363_probe(struct i2c_client *client) if (IS_ERR(pcf85363->rtc)) return PTR_ERR(pcf85363->rtc); + err = pcf85363_load_capacitance(pcf85363, client->dev.of_node); + if (err < 0) + dev_warn(&client->dev, "failed to set xtal load capacitance: %d", + err); + pcf85363->rtc->ops = &rtc_ops; pcf85363->rtc->range_min = RTC_TIMESTAMP_BEGIN_2000; pcf85363->rtc->range_max = RTC_TIMESTAMP_END_2099; From e74a68468062d7ebd8ce17069e12ccc64cc6a58c Mon Sep 17 00:00:00 2001 From: Peter Collingbourne Date: Tue, 14 Feb 2023 21:09:11 -0800 Subject: [PATCH 381/765] arm64: Reset KASAN tag in copy_highpage with HW tags only MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit During page migration, the copy_highpage function is used to copy the page data to the target page. If the source page is a userspace page with MTE tags, the KASAN tag of the target page must have the match-all tag in order to avoid tag check faults during subsequent accesses to the page by the kernel. However, the target page may have been allocated in a number of ways, some of which will use the KASAN allocator and will therefore end up setting the KASAN tag to a non-match-all tag. Therefore, update the target page's KASAN tag to match the source page. We ended up unintentionally fixing this issue as a result of a bad merge conflict resolution between commit e059853d14ca ("arm64: mte: Fix/clarify the PG_mte_tagged semantics") and commit 20794545c146 ("arm64: kasan: Revert "arm64: mte: reset the page tag in page->flags""), which preserved a tag reset for PG_mte_tagged pages which was considered to be unnecessary at the time. Because SW tags KASAN uses separate tag storage, update the code to only reset the tags when HW tags KASAN is enabled. Signed-off-by: Peter Collingbourne Link: https://linux-review.googlesource.com/id/If303d8a709438d3ff5af5fd85706505830f52e0c Reported-by: "Kuan-Ying Lee (李冠穎)" Cc: # 6.1 Fixes: 20794545c146 ("arm64: kasan: Revert "arm64: mte: reset the page tag in page->flags"") Reviewed-by: Andrey Konovalov Link: https://lore.kernel.org/r/20230215050911.1433132-1-pcc@google.com Signed-off-by: Catalin Marinas --- arch/arm64/mm/copypage.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/arm64/mm/copypage.c b/arch/arm64/mm/copypage.c index 8dd5a8fe64b4f..4aadcfb017545 100644 --- a/arch/arm64/mm/copypage.c +++ b/arch/arm64/mm/copypage.c @@ -22,7 +22,8 @@ void copy_highpage(struct page *to, struct page *from) copy_page(kto, kfrom); if (system_supports_mte() && page_mte_tagged(from)) { - page_kasan_tag_reset(to); + if (kasan_hw_tags_enabled()) + page_kasan_tag_reset(to); /* It's a new page, shouldn't have been tagged yet */ WARN_ON_ONCE(!try_page_mte_tagging(to)); mte_copy_page_tags(kto, kfrom); From 0269680e5eb88f6223c53a8b3138cbfa60ba7657 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Thu, 9 Feb 2023 20:04:07 +0000 Subject: [PATCH 382/765] arm64/fpsimd: Remove warning for SME without SVE Support for SME without SVE is architecturally valid and has now been tested well enough so let's remove the warning message that is displayed at boot. Signed-off-by: Mark Brown Link: https://lore.kernel.org/r/20230209-arm64-sme-no-sve-v1-1-74eb3df2f878@kernel.org Signed-off-by: Catalin Marinas --- arch/arm64/kernel/fpsimd.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/arch/arm64/kernel/fpsimd.c b/arch/arm64/kernel/fpsimd.c index c11cb445ffcad..7e823ee7ffa2c 100644 --- a/arch/arm64/kernel/fpsimd.c +++ b/arch/arm64/kernel/fpsimd.c @@ -2122,9 +2122,6 @@ static int __init fpsimd_init(void) pr_notice("Advanced SIMD is not implemented\n"); - if (cpu_have_named_feature(SME) && !cpu_have_named_feature(SVE)) - pr_notice("SME is implemented but not SVE\n"); - sve_sysctl_init(); sme_sysctl_init(); From e034b8a18d4badceecb672c58b488bad1e901d95 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Wed, 22 Feb 2023 13:37:12 +0100 Subject: [PATCH 383/765] drm/msm: Fix possible uninitialized access in fbdev Do not run drm_fb_helper_unprepare() if fbdev allocation fails. Avoids access to an uninitialized pointer. Original bug report is at [1]. Reported-by: kernel test robot Signed-off-by: Thomas Zimmermann Fixes: 3fb1f62f80a1 ("drm/fb-helper: Remove drm_fb_helper_unprepare() from drm_fb_helper_fini()") Link: https://lore.kernel.org/oe-kbuild-all/202302220810.9dymwCQ8-lkp@intel.com/ # 1 Reviewed-by: Dmitry Baryshkov Link: https://patchwork.freedesktop.org/patch/msgid/20230222123712.5049-1-tzimmermann@suse.de --- drivers/gpu/drm/msm/msm_fbdev.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c index c804e5ba682af..d26aa52217ce1 100644 --- a/drivers/gpu/drm/msm/msm_fbdev.c +++ b/drivers/gpu/drm/msm/msm_fbdev.c @@ -136,13 +136,13 @@ static const struct drm_fb_helper_funcs msm_fb_helper_funcs = { struct drm_fb_helper *msm_fbdev_init(struct drm_device *dev) { struct msm_drm_private *priv = dev->dev_private; - struct msm_fbdev *fbdev = NULL; + struct msm_fbdev *fbdev; struct drm_fb_helper *helper; int ret; fbdev = kzalloc(sizeof(*fbdev), GFP_KERNEL); if (!fbdev) - goto fail; + return NULL; helper = &fbdev->base; From b61b82f81e095fe265b0614045d17b08e6ee5c72 Mon Sep 17 00:00:00 2001 From: Sangmoon Kim Date: Mon, 20 Feb 2023 16:34:41 +0900 Subject: [PATCH 384/765] arm64: pass ESR_ELx to die() of cfi_handler Commit 0f2cb928a154 ("arm64: consistently pass ESR_ELx to die()") caused all callers to pass the ESR_ELx value to die(). For consistency, this patch also adds esr to die() call of cfi_handler. Also, when CFI error occurs, die handlers can use ESR_ELx value. Signed-off-by: Sangmoon Kim Acked-by: Mark Rutland Reviewed-by: Mark Brown Link: https://lore.kernel.org/r/20230220073441.2753-1-sangmoon.kim@samsung.com Signed-off-by: Catalin Marinas --- arch/arm64/kernel/traps.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/kernel/traps.c b/arch/arm64/kernel/traps.c index 0ccc063daccb8..4a623e2e982bc 100644 --- a/arch/arm64/kernel/traps.c +++ b/arch/arm64/kernel/traps.c @@ -990,7 +990,7 @@ static int cfi_handler(struct pt_regs *regs, unsigned long esr) switch (report_cfi_failure(regs, regs->pc, &target, type)) { case BUG_TRAP_TYPE_BUG: - die("Oops - CFI", regs, 0); + die("Oops - CFI", regs, esr); break; case BUG_TRAP_TYPE_WARN: From ce8e04f6e5d3b2d14cd00cc4c0b1cc8cbdcf4d12 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 17 Feb 2023 08:22:17 -0700 Subject: [PATCH 385/765] io_uring: consolidate the put_ref-and-return section of adding work We've got a few cases of this, move them to one section and just use gotos to get there. Reduces the text section on both arm64 and x86-64, using gcc-12.2. Signed-off-by: Jens Axboe --- io_uring/io_uring.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 3b915deb4d08f..cbe06deb84ff2 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -1285,17 +1285,15 @@ static void io_req_local_work_add(struct io_kiocb *req) percpu_ref_get(&ctx->refs); - if (!llist_add(&req->io_task_work.node, &ctx->work_llist)) { - percpu_ref_put(&ctx->refs); - return; - } + if (!llist_add(&req->io_task_work.node, &ctx->work_llist)) + goto put_ref; + /* needed for the following wake up */ smp_mb__after_atomic(); if (unlikely(atomic_read(&req->task->io_uring->in_idle))) { io_move_task_work_from_local(ctx); - percpu_ref_put(&ctx->refs); - return; + goto put_ref; } if (ctx->flags & IORING_SETUP_TASKRUN_FLAG) @@ -1305,6 +1303,8 @@ static void io_req_local_work_add(struct io_kiocb *req) if (READ_ONCE(ctx->cq_waiting)) wake_up_state(ctx->submitter_task, TASK_INTERRUPTIBLE); + +put_ref: percpu_ref_put(&ctx->refs); } From 8d664282a03fec09682f10252d3c785c2513691d Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 17 Feb 2023 08:27:23 -0700 Subject: [PATCH 386/765] io_uring: rename 'in_idle' to 'in_cancel' This better describes what it does - it's incremented when the task is currently undergoing a cancelation operation, due to exiting or exec'ing. Signed-off-by: Jens Axboe --- include/linux/io_uring_types.h | 2 +- io_uring/io_uring.c | 18 +++++++++--------- io_uring/tctx.c | 2 +- 3 files changed, 11 insertions(+), 11 deletions(-) diff --git a/include/linux/io_uring_types.h b/include/linux/io_uring_types.h index 0efe4d7843582..00689c12f6abb 100644 --- a/include/linux/io_uring_types.h +++ b/include/linux/io_uring_types.h @@ -58,7 +58,7 @@ struct io_uring_task { struct xarray xa; struct wait_queue_head wait; - atomic_t in_idle; + atomic_t in_cancel; atomic_t inflight_tracked; struct percpu_counter inflight; diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index cbe06deb84ff2..64e07df034d12 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -719,7 +719,7 @@ static void io_put_task_remote(struct task_struct *task, int nr) struct io_uring_task *tctx = task->io_uring; percpu_counter_sub(&tctx->inflight, nr); - if (unlikely(atomic_read(&tctx->in_idle))) + if (unlikely(atomic_read(&tctx->in_cancel))) wake_up(&tctx->wait); put_task_struct_many(task, nr); } @@ -1258,8 +1258,8 @@ void tctx_task_work(struct callback_head *cb) ctx_flush_and_put(ctx, &uring_locked); - /* relaxed read is enough as only the task itself sets ->in_idle */ - if (unlikely(atomic_read(&tctx->in_idle))) + /* relaxed read is enough as only the task itself sets ->in_cancel */ + if (unlikely(atomic_read(&tctx->in_cancel))) io_uring_drop_tctx_refs(current); trace_io_uring_task_work_run(tctx, count, loops); @@ -1291,7 +1291,7 @@ static void io_req_local_work_add(struct io_kiocb *req) /* needed for the following wake up */ smp_mb__after_atomic(); - if (unlikely(atomic_read(&req->task->io_uring->in_idle))) { + if (unlikely(atomic_read(&req->task->io_uring->in_cancel))) { io_move_task_work_from_local(ctx); goto put_ref; } @@ -2937,12 +2937,12 @@ static __cold void io_tctx_exit_cb(struct callback_head *cb) work = container_of(cb, struct io_tctx_exit, task_work); /* - * When @in_idle, we're in cancellation and it's racy to remove the + * When @in_cancel, we're in cancellation and it's racy to remove the * node. It'll be removed by the end of cancellation, just ignore it. * tctx can be NULL if the queueing of this task_work raced with * work cancelation off the exec path. */ - if (tctx && !atomic_read(&tctx->in_idle)) + if (tctx && !atomic_read(&tctx->in_cancel)) io_uring_del_tctx_node((unsigned long)work->ctx); complete(&work->completion); } @@ -3210,7 +3210,7 @@ __cold void io_uring_cancel_generic(bool cancel_all, struct io_sq_data *sqd) if (tctx->io_wq) io_wq_exit_start(tctx->io_wq); - atomic_inc(&tctx->in_idle); + atomic_inc(&tctx->in_cancel); do { bool loop = false; @@ -3261,9 +3261,9 @@ __cold void io_uring_cancel_generic(bool cancel_all, struct io_sq_data *sqd) if (cancel_all) { /* * We shouldn't run task_works after cancel, so just leave - * ->in_idle set for normal exit. + * ->in_cancel set for normal exit. */ - atomic_dec(&tctx->in_idle); + atomic_dec(&tctx->in_cancel); /* for exec all current's requests should be gone, kill tctx */ __io_uring_free(current); } diff --git a/io_uring/tctx.c b/io_uring/tctx.c index 4324b1cf1f6af..3a8d1dd97e1b4 100644 --- a/io_uring/tctx.c +++ b/io_uring/tctx.c @@ -83,7 +83,7 @@ __cold int io_uring_alloc_task_context(struct task_struct *task, xa_init(&tctx->xa); init_waitqueue_head(&tctx->wait); - atomic_set(&tctx->in_idle, 0); + atomic_set(&tctx->in_cancel, 0); atomic_set(&tctx->inflight_tracked, 0); task->io_uring = tctx; init_llist_head(&tctx->task_list); From 6bf65a1b3668b04bb6c8126494d00303104eb9e5 Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Mon, 20 Feb 2023 14:13:52 +0000 Subject: [PATCH 387/765] io_uring/rsrc: fix a comment in io_import_fixed() io_import_fixed() supports offsets, but "may not" means the opposite. Replace it with "might not" so the comments rather speaks about possible cases. Signed-off-by: Pavel Begunkov Reviewed-by: Gabriel Krisman Bertazi Link: https://lore.kernel.org/r/5b5f79958456caa6dc532f6205f75f224b232c81.1676902343.git.asml.silence@gmail.com Signed-off-by: Jens Axboe --- io_uring/rsrc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c index a59fc02de5983..4e6191904c9e5 100644 --- a/io_uring/rsrc.c +++ b/io_uring/rsrc.c @@ -1335,7 +1335,7 @@ int io_import_fixed(int ddir, struct iov_iter *iter, return -EFAULT; /* - * May not be a start of buffer, set size appropriately + * Might not be a start of buffer, set size appropriately * and advance us to the beginning. */ offset = buf_addr - imu->ubuf; From 48ba08374e779421ca34bd14b4834aae19fc3e6a Mon Sep 17 00:00:00 2001 From: Wojciech Lukowicz Date: Sat, 18 Feb 2023 18:41:41 +0000 Subject: [PATCH 388/765] io_uring: fix size calculation when registering buf ring Using struct_size() to calculate the size of io_uring_buf_ring will sum the size of the struct and of the bufs array. However, the struct's fields are overlaid with the array making the calculated size larger than it should be. When registering a ring with N * PAGE_SIZE / sizeof(struct io_uring_buf) entries, i.e. with fully filled pages, the calculated size will span one more page than it should and io_uring will try to pin the following page. Depending on how the application allocated the ring, it might succeed using an unrelated page or fail returning EFAULT. The size of the ring should be the product of ring_entries and the size of io_uring_buf, i.e. the size of the bufs array only. Fixes: c7fb19428d67 ("io_uring: add support for ring mapped supplied buffers") Signed-off-by: Wojciech Lukowicz Reviewed-by: Gabriel Krisman Bertazi Link: https://lore.kernel.org/r/20230218184141.70891-1-wlukowicz01@gmail.com Signed-off-by: Jens Axboe --- io_uring/kbuf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/io_uring/kbuf.c b/io_uring/kbuf.c index 4a6401080c1f8..3002dc8271959 100644 --- a/io_uring/kbuf.c +++ b/io_uring/kbuf.c @@ -505,7 +505,7 @@ int io_register_pbuf_ring(struct io_ring_ctx *ctx, void __user *arg) } pages = io_pin_pages(reg.ring_addr, - struct_size(br, bufs, reg.ring_entries), + flex_array_size(br, bufs, reg.ring_entries), &nr_pages); if (IS_ERR(pages)) { kfree(free_bl); From 9a1563d1720680bdc1d702486b7b73f51c079b32 Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Wed, 22 Feb 2023 14:32:43 +0000 Subject: [PATCH 389/765] io_uring: remove unused wq_list_merge There are no users of wq_list_merge, kill it. Signed-off-by: Pavel Begunkov Link: https://lore.kernel.org/r/5f9ad0301949213230ad9000a8359d591aae615a.1677002255.git.asml.silence@gmail.com Signed-off-by: Jens Axboe --- io_uring/slist.h | 22 ---------------------- 1 file changed, 22 deletions(-) diff --git a/io_uring/slist.h b/io_uring/slist.h index f27601fa46607..7c198a40d5f11 100644 --- a/io_uring/slist.h +++ b/io_uring/slist.h @@ -27,28 +27,6 @@ static inline void wq_list_add_after(struct io_wq_work_node *node, list->last = node; } -/** - * wq_list_merge - merge the second list to the first one. - * @list0: the first list - * @list1: the second list - * Return the first node after mergence. - */ -static inline struct io_wq_work_node *wq_list_merge(struct io_wq_work_list *list0, - struct io_wq_work_list *list1) -{ - struct io_wq_work_node *ret; - - if (!list0->first) { - ret = list1->first; - } else { - ret = list0->first; - list0->last->next = list1->first; - } - INIT_WQ_LIST(list0); - INIT_WQ_LIST(list1); - return ret; -} - static inline void wq_list_add_tail(struct io_wq_work_node *node, struct io_wq_work_list *list) { From edd478269640b360c6f301f2baa04abdda563ef3 Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Wed, 22 Feb 2023 14:36:48 +0000 Subject: [PATCH 390/765] io_uring/rsrc: disallow multi-source reg buffers If two or more mappings go back to back to each other they can be passed into io_uring to be registered as a single registered buffer. That would even work if mappings came from different sources, e.g. it's possible to mix in this way anon pages and pages from shmem or hugetlb. That is not a problem but it'd rather be less prone if we forbid such mixing. Cc: Signed-off-by: Pavel Begunkov Signed-off-by: Jens Axboe --- io_uring/rsrc.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c index 4e6191904c9e5..53845e496881d 100644 --- a/io_uring/rsrc.c +++ b/io_uring/rsrc.c @@ -1162,14 +1162,17 @@ struct page **io_pin_pages(unsigned long ubuf, unsigned long len, int *npages) pret = pin_user_pages(ubuf, nr_pages, FOLL_WRITE | FOLL_LONGTERM, pages, vmas); if (pret == nr_pages) { + struct file *file = vmas[0]->vm_file; + /* don't support file backed memory */ for (i = 0; i < nr_pages; i++) { - struct vm_area_struct *vma = vmas[i]; - - if (vma_is_shmem(vma)) + if (vmas[i]->vm_file != file) { + ret = -EINVAL; + break; + } + if (!file) continue; - if (vma->vm_file && - !is_file_hugepages(vma->vm_file)) { + if (!vma_is_shmem(vmas[i]) && !is_file_hugepages(file)) { ret = -EOPNOTSUPP; break; } From b000ae0ec2d709046ac1a3c5722fea417f8a067e Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Wed, 22 Feb 2023 14:36:50 +0000 Subject: [PATCH 391/765] io_uring/rsrc: optimise single entry advance Iterating within the first bvec entry should be essentially free, but we use iov_iter_advance() for that, which shows up in benchmark profiles taking up to 0.5% of CPU. Replace it with a hand coded version. Signed-off-by: Pavel Begunkov Signed-off-by: Jens Axboe --- io_uring/rsrc.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c index 53845e496881d..ebbd2cea75820 100644 --- a/io_uring/rsrc.c +++ b/io_uring/rsrc.c @@ -1364,7 +1364,10 @@ int io_import_fixed(int ddir, struct iov_iter *iter, const struct bio_vec *bvec = imu->bvec; if (offset <= bvec->bv_len) { - iov_iter_advance(iter, offset); + iter->bvec = bvec; + iter->nr_segs = bvec->bv_len; + iter->count -= offset; + iter->iov_offset = offset; } else { unsigned long seg_skip; From 57bebf807e2abcf87d96b9de1266104ee2d8fc2f Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Wed, 22 Feb 2023 14:36:51 +0000 Subject: [PATCH 392/765] io_uring/rsrc: optimise registered huge pages When registering huge pages, internally io_uring will split them into many PAGE_SIZE bvec entries. That's bad for performance as drivers need to eventually dma-map the data and will do it individually for each bvec entry. Coalesce huge pages into one large bvec. Signed-off-by: Pavel Begunkov Signed-off-by: Jens Axboe --- io_uring/rsrc.c | 38 ++++++++++++++++++++++++++++++++------ 1 file changed, 32 insertions(+), 6 deletions(-) diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c index ebbd2cea75820..aab1bc6883e96 100644 --- a/io_uring/rsrc.c +++ b/io_uring/rsrc.c @@ -1210,6 +1210,7 @@ static int io_sqe_buffer_register(struct io_ring_ctx *ctx, struct iovec *iov, unsigned long off; size_t size; int ret, nr_pages, i; + struct folio *folio; *pimu = ctx->dummy_ubuf; if (!iov->iov_base) @@ -1224,6 +1225,21 @@ static int io_sqe_buffer_register(struct io_ring_ctx *ctx, struct iovec *iov, goto done; } + /* If it's a huge page, try to coalesce them into a single bvec entry */ + if (nr_pages > 1) { + folio = page_folio(pages[0]); + for (i = 1; i < nr_pages; i++) { + if (page_folio(pages[i]) != folio) { + folio = NULL; + break; + } + } + if (folio) { + folio_put_refs(folio, nr_pages - 1); + nr_pages = 1; + } + } + imu = kvmalloc(struct_size(imu, bvec, nr_pages), GFP_KERNEL); if (!imu) goto done; @@ -1236,6 +1252,17 @@ static int io_sqe_buffer_register(struct io_ring_ctx *ctx, struct iovec *iov, off = (unsigned long) iov->iov_base & ~PAGE_MASK; size = iov->iov_len; + /* store original address for later verification */ + imu->ubuf = (unsigned long) iov->iov_base; + imu->ubuf_end = imu->ubuf + iov->iov_len; + imu->nr_bvecs = nr_pages; + *pimu = imu; + ret = 0; + + if (folio) { + bvec_set_page(&imu->bvec[0], pages[0], size, off); + goto done; + } for (i = 0; i < nr_pages; i++) { size_t vec_len; @@ -1244,12 +1271,6 @@ static int io_sqe_buffer_register(struct io_ring_ctx *ctx, struct iovec *iov, off = 0; size -= vec_len; } - /* store original address for later verification */ - imu->ubuf = (unsigned long) iov->iov_base; - imu->ubuf_end = imu->ubuf + iov->iov_len; - imu->nr_bvecs = nr_pages; - *pimu = imu; - ret = 0; done: if (ret) kvfree(imu); @@ -1364,6 +1385,11 @@ int io_import_fixed(int ddir, struct iov_iter *iter, const struct bio_vec *bvec = imu->bvec; if (offset <= bvec->bv_len) { + /* + * Note, huge pages buffers consists of one large + * bvec entry and should always go this way. The other + * branch doesn't expect non PAGE_SIZE'd chunks. + */ iter->bvec = bvec; iter->nr_segs = bvec->bv_len; iter->count -= offset; From 65eb2867f5bf46023ab42ba8347106870d961cf3 Mon Sep 17 00:00:00 2001 From: Vojtech Hejsek Date: Thu, 16 Feb 2023 09:01:15 +0100 Subject: [PATCH 393/765] ACPI: resource: Skip IRQ override on Asus Expertbook B2402FBA The Asus Expertbook B2502FBA has IRQ 1 described as Active_Low in its ACPI table. However, the kernel overrides this and sets it to Edge_High, which prevents the internal keyboard from working properly. Adding this laptop model to the override_table fixes the issue. Signed-off-by: Vojtech Hejsek Signed-off-by: Rafael J. Wysocki --- drivers/acpi/resource.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/acpi/resource.c b/drivers/acpi/resource.c index a222bda7e15b0..7c9125df5a651 100644 --- a/drivers/acpi/resource.c +++ b/drivers/acpi/resource.c @@ -439,6 +439,13 @@ static const struct dmi_system_id asus_laptop[] = { DMI_MATCH(DMI_BOARD_NAME, "B2402CBA"), }, }, + { + .ident = "Asus ExpertBook B2402FBA", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), + DMI_MATCH(DMI_BOARD_NAME, "B2402FBA"), + }, + }, { .ident = "Asus ExpertBook B2502", .matches = { From f525b210e9d4dcefb88594d50e426068e62840f4 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Tue, 10 Jan 2023 16:08:02 +0200 Subject: [PATCH 394/765] rtc: isl12022: Get rid of unneeded private struct isl12022 First of all, the struct rtc_device pointer is kept in the managed resources, no need to keep it outside (no users in the driver). Second, replace private struct isl12022 with a regmap. Signed-off-by: Andy Shevchenko Acked-by: Rasmus Villemoes Link: https://lore.kernel.org/r/20230110140806.87432-2-andriy.shevchenko@linux.intel.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-isl12022.c | 56 ++++++++++++++------------------------ 1 file changed, 21 insertions(+), 35 deletions(-) diff --git a/drivers/rtc/rtc-isl12022.c b/drivers/rtc/rtc-isl12022.c index a3b0de3393f57..44058fa27277c 100644 --- a/drivers/rtc/rtc-isl12022.c +++ b/drivers/rtc/rtc-isl12022.c @@ -46,11 +46,6 @@ static struct i2c_driver isl12022_driver; -struct isl12022 { - struct rtc_device *rtc; - struct regmap *regmap; -}; - static umode_t isl12022_hwmon_is_visible(const void *data, enum hwmon_sensor_types type, u32 attr, int channel) @@ -67,8 +62,7 @@ static umode_t isl12022_hwmon_is_visible(const void *data, */ static int isl12022_hwmon_read_temp(struct device *dev, long *mC) { - struct isl12022 *isl12022 = dev_get_drvdata(dev); - struct regmap *regmap = isl12022->regmap; + struct regmap *regmap = dev_get_drvdata(dev); u8 temp_buf[2]; int temp, ret; @@ -115,23 +109,21 @@ static const struct hwmon_chip_info isl12022_hwmon_chip_info = { static void isl12022_hwmon_register(struct device *dev) { - struct isl12022 *isl12022; + struct regmap *regmap = dev_get_drvdata(dev); struct device *hwmon; int ret; if (!IS_REACHABLE(CONFIG_HWMON)) return; - isl12022 = dev_get_drvdata(dev); - - ret = regmap_update_bits(isl12022->regmap, ISL12022_REG_BETA, + ret = regmap_update_bits(regmap, ISL12022_REG_BETA, ISL12022_BETA_TSE, ISL12022_BETA_TSE); if (ret) { dev_warn(dev, "unable to enable temperature sensor\n"); return; } - hwmon = devm_hwmon_device_register_with_info(dev, "isl12022", isl12022, + hwmon = devm_hwmon_device_register_with_info(dev, "isl12022", regmap, &isl12022_hwmon_chip_info, NULL); if (IS_ERR(hwmon)) @@ -144,8 +136,7 @@ static void isl12022_hwmon_register(struct device *dev) */ static int isl12022_rtc_read_time(struct device *dev, struct rtc_time *tm) { - struct isl12022 *isl12022 = dev_get_drvdata(dev); - struct regmap *regmap = isl12022->regmap; + struct regmap *regmap = dev_get_drvdata(dev); uint8_t buf[ISL12022_REG_INT + 1]; int ret; @@ -190,8 +181,7 @@ static int isl12022_rtc_read_time(struct device *dev, struct rtc_time *tm) static int isl12022_rtc_set_time(struct device *dev, struct rtc_time *tm) { - struct isl12022 *isl12022 = dev_get_drvdata(dev); - struct regmap *regmap = isl12022->regmap; + struct regmap *regmap = dev_get_drvdata(dev); int ret; uint8_t buf[ISL12022_REG_DW + 1]; @@ -218,8 +208,7 @@ static int isl12022_rtc_set_time(struct device *dev, struct rtc_time *tm) buf[ISL12022_REG_DW] = tm->tm_wday & 0x07; - return regmap_bulk_write(isl12022->regmap, ISL12022_REG_SC, - buf, sizeof(buf)); + return regmap_bulk_write(regmap, ISL12022_REG_SC, buf, sizeof(buf)); } static const struct rtc_class_ops isl12022_rtc_ops = { @@ -235,34 +224,31 @@ static const struct regmap_config regmap_config = { static int isl12022_probe(struct i2c_client *client) { - struct isl12022 *isl12022; + struct rtc_device *rtc; + struct regmap *regmap; if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C)) return -ENODEV; - isl12022 = devm_kzalloc(&client->dev, sizeof(struct isl12022), - GFP_KERNEL); - if (!isl12022) - return -ENOMEM; - dev_set_drvdata(&client->dev, isl12022); - - isl12022->regmap = devm_regmap_init_i2c(client, ®map_config); - if (IS_ERR(isl12022->regmap)) { + regmap = devm_regmap_init_i2c(client, ®map_config); + if (IS_ERR(regmap)) { dev_err(&client->dev, "regmap allocation failed\n"); - return PTR_ERR(isl12022->regmap); + return PTR_ERR(regmap); } + dev_set_drvdata(&client->dev, regmap); + isl12022_hwmon_register(&client->dev); - isl12022->rtc = devm_rtc_allocate_device(&client->dev); - if (IS_ERR(isl12022->rtc)) - return PTR_ERR(isl12022->rtc); + rtc = devm_rtc_allocate_device(&client->dev); + if (IS_ERR(rtc)) + return PTR_ERR(rtc); - isl12022->rtc->ops = &isl12022_rtc_ops; - isl12022->rtc->range_min = RTC_TIMESTAMP_BEGIN_2000; - isl12022->rtc->range_max = RTC_TIMESTAMP_END_2099; + rtc->ops = &isl12022_rtc_ops; + rtc->range_min = RTC_TIMESTAMP_BEGIN_2000; + rtc->range_max = RTC_TIMESTAMP_END_2099; - return devm_rtc_register_device(isl12022->rtc); + return devm_rtc_register_device(rtc); } #ifdef CONFIG_OF From 93219a4fb8bdf279f749b5eef40bef1bbe805fc3 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Tue, 10 Jan 2023 16:08:03 +0200 Subject: [PATCH 395/765] rtc: isl12022: Explicitly use __le16 type for ISL12022_REG_TEMP_L We are reading 10-bit value in a 16-bit register in LE format. Make this explicit by using __le16 type for it and corresponding conversion function. Signed-off-by: Andy Shevchenko Acked-by: Rasmus Villemoes Link: https://lore.kernel.org/r/20230110140806.87432-3-andriy.shevchenko@linux.intel.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-isl12022.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/rtc/rtc-isl12022.c b/drivers/rtc/rtc-isl12022.c index 44058fa27277c..bf1aa6f6560de 100644 --- a/drivers/rtc/rtc-isl12022.c +++ b/drivers/rtc/rtc-isl12022.c @@ -19,6 +19,8 @@ #include #include +#include + /* ISL register offsets */ #define ISL12022_REG_SC 0x00 #define ISL12022_REG_MN 0x01 @@ -63,17 +65,16 @@ static umode_t isl12022_hwmon_is_visible(const void *data, static int isl12022_hwmon_read_temp(struct device *dev, long *mC) { struct regmap *regmap = dev_get_drvdata(dev); - u8 temp_buf[2]; int temp, ret; + __le16 buf; - ret = regmap_bulk_read(regmap, ISL12022_REG_TEMP_L, - temp_buf, sizeof(temp_buf)); + ret = regmap_bulk_read(regmap, ISL12022_REG_TEMP_L, &buf, sizeof(buf)); if (ret) return ret; /* * Temperature is represented as a 10-bit number, unit half-Kelvins. */ - temp = (temp_buf[1] << 8) | temp_buf[0]; + temp = le16_to_cpu(buf); temp *= 500; temp -= 273000; From c8ae2735cb3d28892fd734804b60e5abf1f6fa91 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Tue, 10 Jan 2023 16:08:04 +0200 Subject: [PATCH 396/765] rtc: isl12022: Drop unneeded OF guards and of_match_ptr() Drop unneeded OF guards and of_match_ptr(). This allows use of the driver with other types of firmware such as ACPI PRP0001 based probing. Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/r/20230110140806.87432-4-andriy.shevchenko@linux.intel.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-isl12022.c | 10 +--------- 1 file changed, 1 insertion(+), 9 deletions(-) diff --git a/drivers/rtc/rtc-isl12022.c b/drivers/rtc/rtc-isl12022.c index bf1aa6f6560de..77b4763f2a70a 100644 --- a/drivers/rtc/rtc-isl12022.c +++ b/drivers/rtc/rtc-isl12022.c @@ -14,8 +14,6 @@ #include #include #include -#include -#include #include #include @@ -46,8 +44,6 @@ #define ISL12022_BETA_TSE (1 << 7) -static struct i2c_driver isl12022_driver; - static umode_t isl12022_hwmon_is_visible(const void *data, enum hwmon_sensor_types type, u32 attr, int channel) @@ -252,14 +248,12 @@ static int isl12022_probe(struct i2c_client *client) return devm_rtc_register_device(rtc); } -#ifdef CONFIG_OF static const struct of_device_id isl12022_dt_match[] = { { .compatible = "isl,isl12022" }, /* for backward compat., don't use */ { .compatible = "isil,isl12022" }, { }, }; MODULE_DEVICE_TABLE(of, isl12022_dt_match); -#endif static const struct i2c_device_id isl12022_id[] = { { "isl12022", 0 }, @@ -270,9 +264,7 @@ MODULE_DEVICE_TABLE(i2c, isl12022_id); static struct i2c_driver isl12022_driver = { .driver = { .name = "rtc-isl12022", -#ifdef CONFIG_OF - .of_match_table = of_match_ptr(isl12022_dt_match), -#endif + .of_match_table = isl12022_dt_match, }, .probe_new = isl12022_probe, .id_table = isl12022_id, From 9bb1e189d7d2c3838938efcc497333803b946081 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Tue, 10 Jan 2023 16:08:05 +0200 Subject: [PATCH 397/765] rtc: isl12022: Join string literals back For easy grepping on debug purposes join string literals back in the messages. While at it, drop __func__ parameter from unique enough dev_dbg() message as Dynamic Debug can retrieve this at run time. Signed-off-by: Andy Shevchenko Acked-by: Rasmus Villemoes Link: https://lore.kernel.org/r/20230110140806.87432-5-andriy.shevchenko@linux.intel.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-isl12022.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/drivers/rtc/rtc-isl12022.c b/drivers/rtc/rtc-isl12022.c index 77b4763f2a70a..ee38c5067ea87 100644 --- a/drivers/rtc/rtc-isl12022.c +++ b/drivers/rtc/rtc-isl12022.c @@ -143,16 +143,12 @@ static int isl12022_rtc_read_time(struct device *dev, struct rtc_time *tm) if (buf[ISL12022_REG_SR] & (ISL12022_SR_LBAT85 | ISL12022_SR_LBAT75)) { dev_warn(dev, - "voltage dropped below %u%%, " - "date and time is not reliable.\n", + "voltage dropped below %u%%, date and time is not reliable.\n", buf[ISL12022_REG_SR] & ISL12022_SR_LBAT85 ? 85 : 75); } dev_dbg(dev, - "%s: raw data is sec=%02x, min=%02x, hr=%02x, " - "mday=%02x, mon=%02x, year=%02x, wday=%02x, " - "sr=%02x, int=%02x", - __func__, + "raw data is sec=%02x, min=%02x, hr=%02x, mday=%02x, mon=%02x, year=%02x, wday=%02x, sr=%02x, int=%02x", buf[ISL12022_REG_SC], buf[ISL12022_REG_MN], buf[ISL12022_REG_HR], From 303eac653a181be59674920725142cfbdd5ba4cd Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Tue, 10 Jan 2023 16:08:06 +0200 Subject: [PATCH 398/765] rtc: isl12022: sort header inclusion alphabetically Sort header inclusion alphabetically for better maintenance. Signed-off-by: Andy Shevchenko Acked-by: Rasmus Villemoes Link: https://lore.kernel.org/r/20230110140806.87432-6-andriy.shevchenko@linux.intel.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-isl12022.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/rtc/rtc-isl12022.c b/drivers/rtc/rtc-isl12022.c index ee38c5067ea87..e68a79b5e00e5 100644 --- a/drivers/rtc/rtc-isl12022.c +++ b/drivers/rtc/rtc-isl12022.c @@ -8,14 +8,14 @@ * by Alessandro Zummo . */ -#include #include -#include -#include -#include #include -#include #include +#include +#include +#include +#include +#include #include From 668a2abf91143caa226b3ccd0bd4d79ea85935a6 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Fri, 17 Feb 2023 15:23:38 +0100 Subject: [PATCH 399/765] rtc: efi: Avoid spamming the log on RTC read failure There are cases where the EFI runtime services may end up in a funny state, e.g., due to a crash in the variable services, and this affects other EFI runtime services as well. That means that, even though GetTime() should not return an error, there are cases where it might, and there is no point in logging such an occurrence multiple times. This works around an issue where user space -apparently- keeps hitting on /dev/rtc if it fails to read the h/w clock, resulting in a tsunami of log spam and a non-responsive system as a result. Cc: Pierre Gondois Cc: Alexandru Elisei Link: https://lore.kernel.org/all/Y2o1hdZK9GGDVJsS@monolith.localdoman/ Signed-off-by: Ard Biesheuvel Link: https://lore.kernel.org/r/20230217142338.1444509-1-ardb@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-efi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/rtc/rtc-efi.c b/drivers/rtc/rtc-efi.c index 1e8bc6cc1e12d..dc6b0f4a54e2e 100644 --- a/drivers/rtc/rtc-efi.c +++ b/drivers/rtc/rtc-efi.c @@ -164,7 +164,7 @@ static int efi_read_time(struct device *dev, struct rtc_time *tm) if (status != EFI_SUCCESS) { /* should never happen */ - dev_err(dev, "can't read time\n"); + dev_err_once(dev, "can't read time\n"); return -EINVAL; } From 38892c48e54760be41f16519456d868e14933789 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Wed, 15 Feb 2023 18:50:30 +0200 Subject: [PATCH 400/765] rtc: rx6110: Remove unused of_gpio,h of_gpio.h provides a single function, which is not used in this driver. Remove unused header. Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/r/20230215165030.83621-1-andriy.shevchenko@linux.intel.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-rx6110.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/rtc/rtc-rx6110.c b/drivers/rtc/rtc-rx6110.c index 76a49838014ba..37608883a796d 100644 --- a/drivers/rtc/rtc-rx6110.c +++ b/drivers/rtc/rtc-rx6110.c @@ -10,7 +10,6 @@ #include #include #include -#include #include #include #include From 9f6ad5d533d1c71e51bdd06a5712c4fbc8768dfa Mon Sep 17 00:00:00 2001 From: Zhong Jinghua Date: Tue, 21 Feb 2023 17:50:27 +0800 Subject: [PATCH 401/765] loop: loop_set_status_from_info() check before assignment In loop_set_status_from_info(), lo->lo_offset and lo->lo_sizelimit should be checked before reassignment, because if an overflow error occurs, the original correct value will be changed to the wrong value, and it will not be changed back. More, the original patch did not solve the problem, the value was set and ioctl returned an error, but the subsequent io used the value in the loop driver, which still caused an alarm: loop_handle_cmd do_req_filebacked loff_t pos = ((loff_t) blk_rq_pos(rq) << 9) + lo->lo_offset; lo_rw_aio cmd->iocb.ki_pos = pos Fixes: c490a0b5a4f3 ("loop: Check for overflow while configuring loop") Signed-off-by: Zhong Jinghua Reviewed-by: Chaitanya Kulkarni Link: https://lore.kernel.org/r/20230221095027.3656193-1-zhongjinghua@huaweicloud.com Signed-off-by: Jens Axboe --- drivers/block/loop.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 5f04235e4ff75..839373451c2b7 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -977,13 +977,13 @@ loop_set_status_from_info(struct loop_device *lo, return -EINVAL; } + /* Avoid assigning overflow values */ + if (info->lo_offset > LLONG_MAX || info->lo_sizelimit > LLONG_MAX) + return -EOVERFLOW; + lo->lo_offset = info->lo_offset; lo->lo_sizelimit = info->lo_sizelimit; - /* loff_t vars have been assigned __u64 */ - if (lo->lo_offset < 0 || lo->lo_sizelimit < 0) - return -EOVERFLOW; - memcpy(lo->lo_file_name, info->lo_file_name, LO_NAME_SIZE); lo->lo_file_name[LO_NAME_SIZE-1] = 0; lo->lo_flags = info->lo_flags; From 84edc94edb25caf8bcd5f4744bf24b82c6b805df Mon Sep 17 00:00:00 2001 From: Deepak R Varma Date: Thu, 19 Jan 2023 01:53:07 +0530 Subject: [PATCH 402/765] drm/i915/gvt: Avoid full proxy f_ops for debugfs attributes Using DEFINE_SIMPLE_ATTRIBUTE macro with the debugfs_create_file() function adds the overhead of introducing a proxy file operation functions to wrap the original read/write inside file removal protection functions. This adds significant overhead in terms of introducing and managing the proxy factory file operations structure and function wrapping at runtime. As a replacement, a combination of DEFINE_DEBUGFS_ATTRIBUTE macro paired with debugfs_create_file_unsafe() is suggested to be used instead. The DEFINE_DEBUGFS_ATTRIBUTE utilises debugfs_file_get() and debugfs_file_put() wrappers to protect the original read and write function calls for the debug attributes. There is no need for any runtime proxy file operations to be managed by the debugfs core. Following coccicheck make command helped identify this change: make coccicheck M=drivers/gpu/drm/i915/ MODE=patch COCCI=./scripts/coccinelle/api/debugfs/debugfs_simple_attr.cocci Signed-off-by: Deepak R Varma Reviewed-by: Rodrigo Vivi Acked-by: Zhenyu Wang Signed-off-by: Zhenyu Wang Link: http://patchwork.freedesktop.org/patch/msgid/Y8hVK6wuqm50iADP@ubun2204.myguest.virtualbox.org --- drivers/gpu/drm/i915/gvt/debugfs.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/i915/gvt/debugfs.c b/drivers/gpu/drm/i915/gvt/debugfs.c index 0616b73175f3e..baccbf1761b77 100644 --- a/drivers/gpu/drm/i915/gvt/debugfs.c +++ b/drivers/gpu/drm/i915/gvt/debugfs.c @@ -147,9 +147,9 @@ vgpu_scan_nonprivbb_set(void *data, u64 val) return 0; } -DEFINE_SIMPLE_ATTRIBUTE(vgpu_scan_nonprivbb_fops, - vgpu_scan_nonprivbb_get, vgpu_scan_nonprivbb_set, - "0x%llx\n"); +DEFINE_DEBUGFS_ATTRIBUTE(vgpu_scan_nonprivbb_fops, + vgpu_scan_nonprivbb_get, vgpu_scan_nonprivbb_set, + "0x%llx\n"); static int vgpu_status_get(void *data, u64 *val) { @@ -165,7 +165,7 @@ static int vgpu_status_get(void *data, u64 *val) return 0; } -DEFINE_SIMPLE_ATTRIBUTE(vgpu_status_fops, vgpu_status_get, NULL, "0x%llx\n"); +DEFINE_DEBUGFS_ATTRIBUTE(vgpu_status_fops, vgpu_status_get, NULL, "0x%llx\n"); /** * intel_gvt_debugfs_add_vgpu - register debugfs entries for a vGPU @@ -180,10 +180,10 @@ void intel_gvt_debugfs_add_vgpu(struct intel_vgpu *vgpu) debugfs_create_file("mmio_diff", 0444, vgpu->debugfs, vgpu, &vgpu_mmio_diff_fops); - debugfs_create_file("scan_nonprivbb", 0644, vgpu->debugfs, vgpu, - &vgpu_scan_nonprivbb_fops); - debugfs_create_file("status", 0644, vgpu->debugfs, vgpu, - &vgpu_status_fops); + debugfs_create_file_unsafe("scan_nonprivbb", 0644, vgpu->debugfs, vgpu, + &vgpu_scan_nonprivbb_fops); + debugfs_create_file_unsafe("status", 0644, vgpu->debugfs, vgpu, + &vgpu_status_fops); } /** From dd62071ff792cd4c2134b1211ba85efc6cd73ce3 Mon Sep 17 00:00:00 2001 From: Deepak R Varma Date: Sat, 14 Jan 2023 21:12:39 +0530 Subject: [PATCH 403/765] drm/i915/gvt: Remove extra semicolon Remove the extra semicolon at end. Issue identified using semicolon.cocci Coccinelle semantic patch. Signed-off-by: Deepak R Varma Signed-off-by: Zhenyu Wang Link: http://patchwork.freedesktop.org/patch/msgid/Y8LNbzgTf/1kYJX/@ubun2204.myguest.virtualbox.org Reviewed-by: Zhenyu Wang --- drivers/gpu/drm/i915/gvt/vgpu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/gvt/vgpu.c b/drivers/gpu/drm/i915/gvt/vgpu.c index a5497440484f1..08ad1bd651f10 100644 --- a/drivers/gpu/drm/i915/gvt/vgpu.c +++ b/drivers/gpu/drm/i915/gvt/vgpu.c @@ -323,7 +323,7 @@ int intel_gvt_create_vgpu(struct intel_vgpu *vgpu, ret = idr_alloc(&gvt->vgpu_idr, vgpu, IDLE_VGPU_IDR + 1, GVT_MAX_VGPU, GFP_KERNEL); if (ret < 0) - goto out_unlock;; + goto out_unlock; vgpu->id = ret; vgpu->sched_ctl.weight = conf->weight; From d989bf543d8aea77c90a3eb8d2e30f9304570810 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 2 Feb 2023 15:13:09 +0100 Subject: [PATCH 404/765] i915: fix memory leak with using debugfs_lookup() When calling debugfs_lookup() the result must have dput() called on it, otherwise the memory will leak over time. To make things simpler, just call debugfs_lookup_and_remove() instead which handles all of the logic at once. Cc: Zhenyu Wang Cc: Zhi Wang Cc: Jani Nikula Cc: Joonas Lahtinen Cc: Rodrigo Vivi Cc: Tvrtko Ursulin Cc: David Airlie Cc: Daniel Vetter Cc: intel-gvt-dev@lists.freedesktop.org Cc: intel-gfx@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org Signed-off-by: Greg Kroah-Hartman Reviewed-by: Rodrigo Vivi Reviewed-by: Zhenyu Wang Signed-off-by: Zhenyu Wang Link: http://patchwork.freedesktop.org/patch/msgid/20230202141309.2293834-1-gregkh@linuxfoundation.org --- drivers/gpu/drm/i915/gvt/kvmgt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c index 8ae7039b36832..de675d799c7d8 100644 --- a/drivers/gpu/drm/i915/gvt/kvmgt.c +++ b/drivers/gpu/drm/i915/gvt/kvmgt.c @@ -699,7 +699,7 @@ static void intel_vgpu_close_device(struct vfio_device *vfio_dev) clear_bit(INTEL_VGPU_STATUS_ATTACHED, vgpu->status); - debugfs_remove(debugfs_lookup(KVMGT_DEBUGFS_FILENAME, vgpu->debugfs)); + debugfs_lookup_and_remove(KVMGT_DEBUGFS_FILENAME, vgpu->debugfs); kvm_page_track_unregister_notifier(vgpu->vfio_device.kvm, &vgpu->track_node); From 9203a648c951af31b11823056c18b7981135524d Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Thu, 2 Feb 2023 12:50:18 +0000 Subject: [PATCH 405/765] i915/gvt: Fix spelling mistake "vender" -> "vendor" There is a spelling mistake in a literal string. Fix it. Signed-off-by: Colin Ian King Acked-by: Zhenyu Wang Signed-off-by: Zhenyu Wang Link: http://patchwork.freedesktop.org/patch/msgid/20230202125018.285523-1-colin.i.king@gmail.com --- drivers/gpu/drm/i915/gvt/firmware.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/gvt/firmware.c b/drivers/gpu/drm/i915/gvt/firmware.c index a683c22d5b641..9b4b50fa01247 100644 --- a/drivers/gpu/drm/i915/gvt/firmware.c +++ b/drivers/gpu/drm/i915/gvt/firmware.c @@ -171,7 +171,7 @@ static int verify_firmware(struct intel_gvt *gvt, mem = (fw->data + h->cfg_space_offset); id = *(u16 *)(mem + PCI_VENDOR_ID); - VERIFY("vender id", id, pdev->vendor); + VERIFY("vendor id", id, pdev->vendor); id = *(u16 *)(mem + PCI_DEVICE_ID); VERIFY("device id", id, pdev->device); From 0b93efca3659f6d55ed31cff6722dca5f6e4d6e2 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 14 Feb 2023 20:45:33 -0800 Subject: [PATCH 406/765] drm/i915: move a Kconfig symbol to unbreak the menu presentation Inserting a Kconfig symbol that does not have a dependency (DRM_I915_GVT) into a list of other symbols that do have a dependency (on DRM_I915) breaks the driver menu presentation in 'make *config'. Relocate the DRM_I915_GVT symbol so that it does not cause this problem. Fixes: 8b750bf74418 ("drm/i915/gvt: move the gvt code into kvmgt.ko") Signed-off-by: Randy Dunlap Cc: Christoph Hellwig Cc: Zhi Wang Cc: Jani Nikula Cc: Joonas Lahtinen Cc: Rodrigo Vivi Cc: Tvrtko Ursulin Cc: Zhenyu Wang Cc: intel-gfx@lists.freedesktop.org Cc: intel-gvt-dev@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org Reviewed-by: Christoph Hellwig Acked-by: Zhenyu Wang Signed-off-by: Zhenyu Wang Link: http://patchwork.freedesktop.org/patch/msgid/20230215044533.4847-1-rdunlap@infradead.org --- drivers/gpu/drm/i915/Kconfig | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig index 9c0990c0ec876..98e29c92b3425 100644 --- a/drivers/gpu/drm/i915/Kconfig +++ b/drivers/gpu/drm/i915/Kconfig @@ -118,9 +118,6 @@ config DRM_I915_USERPTR If in doubt, say "Y". -config DRM_I915_GVT - bool - config DRM_I915_GVT_KVMGT tristate "Enable KVM host support Intel GVT-g graphics virtualization" depends on DRM_I915 @@ -171,3 +168,6 @@ menu "drm/i915 Unstable Evolution" depends on DRM_I915 source "drivers/gpu/drm/i915/Kconfig.unstable" endmenu + +config DRM_I915_GVT + bool From 20a554638dd2665a88d3d68a68f7981480a27f36 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 8 Feb 2023 18:17:57 +0100 Subject: [PATCH 407/765] objtool: Change arch_decode_instruction() signature MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In preparation to changing struct instruction around a bit, avoid passing it's members by pointer and instead pass the whole thing. A cleanup in it's own right too. Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Acked-by: Josh Poimboeuf Tested-by: Nathan Chancellor # build only Tested-by: Thomas Weißschuh # compile and run Link: https://lore.kernel.org/r/20230208172245.291087549@infradead.org --- tools/objtool/arch/powerpc/decode.c | 22 +++--- tools/objtool/arch/x86/decode.c | 105 +++++++++++++-------------- tools/objtool/check.c | 4 +- tools/objtool/include/objtool/arch.h | 4 +- 4 files changed, 64 insertions(+), 71 deletions(-) diff --git a/tools/objtool/arch/powerpc/decode.c b/tools/objtool/arch/powerpc/decode.c index 9c653805a08a9..53b55690f3204 100644 --- a/tools/objtool/arch/powerpc/decode.c +++ b/tools/objtool/arch/powerpc/decode.c @@ -41,38 +41,36 @@ const char *arch_ret_insn(int len) int arch_decode_instruction(struct objtool_file *file, const struct section *sec, unsigned long offset, unsigned int maxlen, - unsigned int *len, enum insn_type *type, - unsigned long *immediate, - struct list_head *ops_list) + struct instruction *insn) { unsigned int opcode; enum insn_type typ; unsigned long imm; - u32 insn; + u32 ins; - insn = bswap_if_needed(file->elf, *(u32 *)(sec->data->d_buf + offset)); - opcode = insn >> 26; + ins = bswap_if_needed(file->elf, *(u32 *)(sec->data->d_buf + offset)); + opcode = ins >> 26; typ = INSN_OTHER; imm = 0; switch (opcode) { case 18: /* b[l][a] */ - if ((insn & 3) == 1) /* bl */ + if ((ins & 3) == 1) /* bl */ typ = INSN_CALL; - imm = insn & 0x3fffffc; + imm = ins & 0x3fffffc; if (imm & 0x2000000) imm -= 0x4000000; break; } if (opcode == 1) - *len = 8; + insn->len = 8; else - *len = 4; + insn->len = 4; - *type = typ; - *immediate = imm; + insn->type = typ; + insn->immediate = imm; return 0; } diff --git a/tools/objtool/arch/x86/decode.c b/tools/objtool/arch/x86/decode.c index e7b030f7e2a5b..c5c49277cf1a9 100644 --- a/tools/objtool/arch/x86/decode.c +++ b/tools/objtool/arch/x86/decode.c @@ -146,12 +146,11 @@ static bool has_notrack_prefix(struct insn *insn) int arch_decode_instruction(struct objtool_file *file, const struct section *sec, unsigned long offset, unsigned int maxlen, - unsigned int *len, enum insn_type *type, - unsigned long *immediate, - struct list_head *ops_list) + struct instruction *insn) { + struct list_head *ops_list = &insn->stack_ops; const struct elf *elf = file->elf; - struct insn insn; + struct insn ins; int x86_64, ret; unsigned char op1, op2, op3, prefix, rex = 0, rex_b = 0, rex_r = 0, rex_w = 0, rex_x = 0, @@ -165,42 +164,42 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec if (x86_64 == -1) return -1; - ret = insn_decode(&insn, sec->data->d_buf + offset, maxlen, + ret = insn_decode(&ins, sec->data->d_buf + offset, maxlen, x86_64 ? INSN_MODE_64 : INSN_MODE_32); if (ret < 0) { WARN("can't decode instruction at %s:0x%lx", sec->name, offset); return -1; } - *len = insn.length; - *type = INSN_OTHER; + insn->len = ins.length; + insn->type = INSN_OTHER; - if (insn.vex_prefix.nbytes) + if (ins.vex_prefix.nbytes) return 0; - prefix = insn.prefixes.bytes[0]; + prefix = ins.prefixes.bytes[0]; - op1 = insn.opcode.bytes[0]; - op2 = insn.opcode.bytes[1]; - op3 = insn.opcode.bytes[2]; + op1 = ins.opcode.bytes[0]; + op2 = ins.opcode.bytes[1]; + op3 = ins.opcode.bytes[2]; - if (insn.rex_prefix.nbytes) { - rex = insn.rex_prefix.bytes[0]; + if (ins.rex_prefix.nbytes) { + rex = ins.rex_prefix.bytes[0]; rex_w = X86_REX_W(rex) >> 3; rex_r = X86_REX_R(rex) >> 2; rex_x = X86_REX_X(rex) >> 1; rex_b = X86_REX_B(rex); } - if (insn.modrm.nbytes) { - modrm = insn.modrm.bytes[0]; + if (ins.modrm.nbytes) { + modrm = ins.modrm.bytes[0]; modrm_mod = X86_MODRM_MOD(modrm); modrm_reg = X86_MODRM_REG(modrm) + 8*rex_r; modrm_rm = X86_MODRM_RM(modrm) + 8*rex_b; } - if (insn.sib.nbytes) { - sib = insn.sib.bytes[0]; + if (ins.sib.nbytes) { + sib = ins.sib.bytes[0]; /* sib_scale = X86_SIB_SCALE(sib); */ sib_index = X86_SIB_INDEX(sib) + 8*rex_x; sib_base = X86_SIB_BASE(sib) + 8*rex_b; @@ -254,7 +253,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec break; case 0x70 ... 0x7f: - *type = INSN_JUMP_CONDITIONAL; + insn->type = INSN_JUMP_CONDITIONAL; break; case 0x80 ... 0x83: @@ -278,7 +277,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec if (!rm_is_reg(CFI_SP)) break; - imm = insn.immediate.value; + imm = ins.immediate.value; if (op1 & 2) { /* sign extend */ if (op1 & 1) { /* imm32 */ imm <<= 32; @@ -309,7 +308,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec ADD_OP(op) { op->src.type = OP_SRC_AND; op->src.reg = CFI_SP; - op->src.offset = insn.immediate.value; + op->src.offset = ins.immediate.value; op->dest.type = OP_DEST_REG; op->dest.reg = CFI_SP; } @@ -356,7 +355,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec op->src.reg = CFI_SP; op->dest.type = OP_DEST_REG_INDIRECT; op->dest.reg = modrm_rm; - op->dest.offset = insn.displacement.value; + op->dest.offset = ins.displacement.value; } break; } @@ -389,7 +388,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec op->src.reg = modrm_reg; op->dest.type = OP_DEST_REG_INDIRECT; op->dest.reg = CFI_BP; - op->dest.offset = insn.displacement.value; + op->dest.offset = ins.displacement.value; } break; } @@ -402,7 +401,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec op->src.reg = modrm_reg; op->dest.type = OP_DEST_REG_INDIRECT; op->dest.reg = CFI_SP; - op->dest.offset = insn.displacement.value; + op->dest.offset = ins.displacement.value; } break; } @@ -419,7 +418,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec ADD_OP(op) { op->src.type = OP_SRC_REG_INDIRECT; op->src.reg = CFI_BP; - op->src.offset = insn.displacement.value; + op->src.offset = ins.displacement.value; op->dest.type = OP_DEST_REG; op->dest.reg = modrm_reg; } @@ -432,7 +431,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec ADD_OP(op) { op->src.type = OP_SRC_REG_INDIRECT; op->src.reg = CFI_SP; - op->src.offset = insn.displacement.value; + op->src.offset = ins.displacement.value; op->dest.type = OP_DEST_REG; op->dest.reg = modrm_reg; } @@ -464,7 +463,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec /* lea disp(%src), %dst */ ADD_OP(op) { - op->src.offset = insn.displacement.value; + op->src.offset = ins.displacement.value; if (!op->src.offset) { /* lea (%src), %dst */ op->src.type = OP_SRC_REG; @@ -487,7 +486,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec break; case 0x90: - *type = INSN_NOP; + insn->type = INSN_NOP; break; case 0x9c: @@ -511,39 +510,39 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec if (op2 == 0x01) { if (modrm == 0xca) - *type = INSN_CLAC; + insn->type = INSN_CLAC; else if (modrm == 0xcb) - *type = INSN_STAC; + insn->type = INSN_STAC; } else if (op2 >= 0x80 && op2 <= 0x8f) { - *type = INSN_JUMP_CONDITIONAL; + insn->type = INSN_JUMP_CONDITIONAL; } else if (op2 == 0x05 || op2 == 0x07 || op2 == 0x34 || op2 == 0x35) { /* sysenter, sysret */ - *type = INSN_CONTEXT_SWITCH; + insn->type = INSN_CONTEXT_SWITCH; } else if (op2 == 0x0b || op2 == 0xb9) { /* ud2 */ - *type = INSN_BUG; + insn->type = INSN_BUG; } else if (op2 == 0x0d || op2 == 0x1f) { /* nopl/nopw */ - *type = INSN_NOP; + insn->type = INSN_NOP; } else if (op2 == 0x1e) { if (prefix == 0xf3 && (modrm == 0xfa || modrm == 0xfb)) - *type = INSN_ENDBR; + insn->type = INSN_ENDBR; } else if (op2 == 0x38 && op3 == 0xf8) { - if (insn.prefixes.nbytes == 1 && - insn.prefixes.bytes[0] == 0xf2) { + if (ins.prefixes.nbytes == 1 && + ins.prefixes.bytes[0] == 0xf2) { /* ENQCMD cannot be used in the kernel. */ WARN("ENQCMD instruction at %s:%lx", sec->name, offset); @@ -591,29 +590,29 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec case 0xcc: /* int3 */ - *type = INSN_TRAP; + insn->type = INSN_TRAP; break; case 0xe3: /* jecxz/jrcxz */ - *type = INSN_JUMP_CONDITIONAL; + insn->type = INSN_JUMP_CONDITIONAL; break; case 0xe9: case 0xeb: - *type = INSN_JUMP_UNCONDITIONAL; + insn->type = INSN_JUMP_UNCONDITIONAL; break; case 0xc2: case 0xc3: - *type = INSN_RETURN; + insn->type = INSN_RETURN; break; case 0xc7: /* mov imm, r/m */ if (!opts.noinstr) break; - if (insn.length == 3+4+4 && !strncmp(sec->name, ".init.text", 10)) { + if (ins.length == 3+4+4 && !strncmp(sec->name, ".init.text", 10)) { struct reloc *immr, *disp; struct symbol *func; int idx; @@ -661,17 +660,17 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec case 0xca: /* retf */ case 0xcb: /* retf */ - *type = INSN_CONTEXT_SWITCH; + insn->type = INSN_CONTEXT_SWITCH; break; case 0xe0: /* loopne */ case 0xe1: /* loope */ case 0xe2: /* loop */ - *type = INSN_JUMP_CONDITIONAL; + insn->type = INSN_JUMP_CONDITIONAL; break; case 0xe8: - *type = INSN_CALL; + insn->type = INSN_CALL; /* * For the impact on the stack, a CALL behaves like * a PUSH of an immediate value (the return address). @@ -683,30 +682,30 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec break; case 0xfc: - *type = INSN_CLD; + insn->type = INSN_CLD; break; case 0xfd: - *type = INSN_STD; + insn->type = INSN_STD; break; case 0xff: if (modrm_reg == 2 || modrm_reg == 3) { - *type = INSN_CALL_DYNAMIC; - if (has_notrack_prefix(&insn)) + insn->type = INSN_CALL_DYNAMIC; + if (has_notrack_prefix(&ins)) WARN("notrack prefix found at %s:0x%lx", sec->name, offset); } else if (modrm_reg == 4) { - *type = INSN_JUMP_DYNAMIC; - if (has_notrack_prefix(&insn)) + insn->type = INSN_JUMP_DYNAMIC; + if (has_notrack_prefix(&ins)) WARN("notrack prefix found at %s:0x%lx", sec->name, offset); } else if (modrm_reg == 5) { /* jmpf */ - *type = INSN_CONTEXT_SWITCH; + insn->type = INSN_CONTEXT_SWITCH; } else if (modrm_reg == 6) { @@ -723,7 +722,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec break; } - *immediate = insn.immediate.nbytes ? insn.immediate.value : 0; + insn->immediate = ins.immediate.nbytes ? ins.immediate.value : 0; return 0; } diff --git a/tools/objtool/check.c b/tools/objtool/check.c index ba07a8ebaf731..b3b423d33cc24 100644 --- a/tools/objtool/check.c +++ b/tools/objtool/check.c @@ -406,9 +406,7 @@ static int decode_instructions(struct objtool_file *file) ret = arch_decode_instruction(file, sec, offset, sec->sh.sh_size - offset, - &insn->len, &insn->type, - &insn->immediate, - &insn->stack_ops); + insn); if (ret) goto err; diff --git a/tools/objtool/include/objtool/arch.h b/tools/objtool/include/objtool/arch.h index 4ecb480131c7c..73149f8090faf 100644 --- a/tools/objtool/include/objtool/arch.h +++ b/tools/objtool/include/objtool/arch.h @@ -75,9 +75,7 @@ void arch_initial_func_cfi_state(struct cfi_init_state *state); int arch_decode_instruction(struct objtool_file *file, const struct section *sec, unsigned long offset, unsigned int maxlen, - unsigned int *len, enum insn_type *type, - unsigned long *immediate, - struct list_head *ops_list); + struct instruction *insn); bool arch_callee_saved_reg(unsigned char reg); From 3ee88df1b063962e39d7798ccc3b18fd10cea813 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 8 Feb 2023 18:17:58 +0100 Subject: [PATCH 408/765] objtool: Make instruction::stack_ops a single-linked list MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit struct instruction { struct list_head list; /* 0 16 */ struct hlist_node hash; /* 16 16 */ struct list_head call_node; /* 32 16 */ struct section * sec; /* 48 8 */ long unsigned int offset; /* 56 8 */ /* --- cacheline 1 boundary (64 bytes) --- */ unsigned int len; /* 64 4 */ enum insn_type type; /* 68 4 */ long unsigned int immediate; /* 72 8 */ u16 dead_end:1; /* 80: 0 2 */ u16 ignore:1; /* 80: 1 2 */ u16 ignore_alts:1; /* 80: 2 2 */ u16 hint:1; /* 80: 3 2 */ u16 save:1; /* 80: 4 2 */ u16 restore:1; /* 80: 5 2 */ u16 retpoline_safe:1; /* 80: 6 2 */ u16 noendbr:1; /* 80: 7 2 */ u16 entry:1; /* 80: 8 2 */ /* XXX 7 bits hole, try to pack */ s8 instr; /* 82 1 */ u8 visited; /* 83 1 */ /* XXX 4 bytes hole, try to pack */ struct alt_group * alt_group; /* 88 8 */ struct symbol * call_dest; /* 96 8 */ struct instruction * jump_dest; /* 104 8 */ struct instruction * first_jump_src; /* 112 8 */ struct reloc * jump_table; /* 120 8 */ /* --- cacheline 2 boundary (128 bytes) --- */ struct reloc * reloc; /* 128 8 */ struct list_head alts; /* 136 16 */ struct symbol * sym; /* 152 8 */ - struct list_head stack_ops; /* 160 16 */ - struct cfi_state * cfi; /* 176 8 */ + struct stack_op * stack_ops; /* 160 8 */ + struct cfi_state * cfi; /* 168 8 */ - /* size: 184, cachelines: 3, members: 29 */ - /* sum members: 178, holes: 1, sum holes: 4 */ + /* size: 176, cachelines: 3, members: 29 */ + /* sum members: 170, holes: 1, sum holes: 4 */ /* sum bitfield members: 9 bits, bit holes: 1, sum bit holes: 7 bits */ - /* last cacheline: 56 bytes */ + /* last cacheline: 48 bytes */ }; pre: 5:58.22 real, 226.69 user, 131.22 sys, 26221520 mem post: 5:58.50 real, 229.64 user, 128.65 sys, 26221520 mem Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Acked-by: Josh Poimboeuf Tested-by: Nathan Chancellor # build only Tested-by: Thomas Weißschuh # compile and run Link: https://lore.kernel.org/r/20230208172245.362196959@infradead.org --- tools/objtool/arch/x86/decode.c | 4 ++-- tools/objtool/check.c | 11 +++++------ tools/objtool/include/objtool/arch.h | 2 +- tools/objtool/include/objtool/check.h | 2 +- 4 files changed, 9 insertions(+), 10 deletions(-) diff --git a/tools/objtool/arch/x86/decode.c b/tools/objtool/arch/x86/decode.c index c5c49277cf1a9..9ef024fd648c1 100644 --- a/tools/objtool/arch/x86/decode.c +++ b/tools/objtool/arch/x86/decode.c @@ -105,7 +105,7 @@ bool arch_pc_relative_reloc(struct reloc *reloc) #define ADD_OP(op) \ if (!(op = calloc(1, sizeof(*op)))) \ return -1; \ - else for (list_add_tail(&op->list, ops_list); op; op = NULL) + else for (*ops_list = op, ops_list = &op->next; op; op = NULL) /* * Helpers to decode ModRM/SIB: @@ -148,7 +148,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec unsigned long offset, unsigned int maxlen, struct instruction *insn) { - struct list_head *ops_list = &insn->stack_ops; + struct stack_op **ops_list = &insn->stack_ops; const struct elf *elf = file->elf; struct insn ins; int x86_64, ret; diff --git a/tools/objtool/check.c b/tools/objtool/check.c index b3b423d33cc24..8109d7405297c 100644 --- a/tools/objtool/check.c +++ b/tools/objtool/check.c @@ -398,7 +398,6 @@ static int decode_instructions(struct objtool_file *file) } memset(insn, 0, sizeof(*insn)); INIT_LIST_HEAD(&insn->alts); - INIT_LIST_HEAD(&insn->stack_ops); INIT_LIST_HEAD(&insn->call_node); insn->sec = sec; @@ -1331,12 +1330,13 @@ static struct reloc *insn_reloc(struct objtool_file *file, struct instruction *i static void remove_insn_ops(struct instruction *insn) { - struct stack_op *op, *tmp; + struct stack_op *op, *next; - list_for_each_entry_safe(op, tmp, &insn->stack_ops, list) { - list_del(&op->list); + for (op = insn->stack_ops; op; op = next) { + next = op->next; free(op); } + insn->stack_ops = NULL; } static void annotate_call_site(struct objtool_file *file, @@ -1781,7 +1781,6 @@ static int handle_group_alt(struct objtool_file *file, } memset(nop, 0, sizeof(*nop)); INIT_LIST_HEAD(&nop->alts); - INIT_LIST_HEAD(&nop->stack_ops); nop->sec = special_alt->new_sec; nop->offset = special_alt->new_off + special_alt->new_len; @@ -3226,7 +3225,7 @@ static int handle_insn_ops(struct instruction *insn, { struct stack_op *op; - list_for_each_entry(op, &insn->stack_ops, list) { + for (op = insn->stack_ops; op; op = op->next) { if (update_cfi_state(insn, next_insn, &state->cfi, op)) return 1; diff --git a/tools/objtool/include/objtool/arch.h b/tools/objtool/include/objtool/arch.h index 73149f8090faf..2b6d2ce4f9a5b 100644 --- a/tools/objtool/include/objtool/arch.h +++ b/tools/objtool/include/objtool/arch.h @@ -62,9 +62,9 @@ struct op_src { }; struct stack_op { + struct stack_op *next; struct op_dest dest; struct op_src src; - struct list_head list; }; struct instruction; diff --git a/tools/objtool/include/objtool/check.h b/tools/objtool/include/objtool/check.h index acd7fae593484..23e9819993655 100644 --- a/tools/objtool/include/objtool/check.h +++ b/tools/objtool/include/objtool/check.h @@ -68,7 +68,7 @@ struct instruction { struct reloc *reloc; struct list_head alts; struct symbol *sym; - struct list_head stack_ops; + struct stack_op *stack_ops; struct cfi_state *cfi; }; From d54066546121426ecd7ad01a53ae429c4e37a9d5 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 8 Feb 2023 18:17:59 +0100 Subject: [PATCH 409/765] objtool: Make instruction::alts a single-linked list MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit struct instruction { struct list_head list; /* 0 16 */ struct hlist_node hash; /* 16 16 */ struct list_head call_node; /* 32 16 */ struct section * sec; /* 48 8 */ long unsigned int offset; /* 56 8 */ /* --- cacheline 1 boundary (64 bytes) --- */ unsigned int len; /* 64 4 */ enum insn_type type; /* 68 4 */ long unsigned int immediate; /* 72 8 */ u16 dead_end:1; /* 80: 0 2 */ u16 ignore:1; /* 80: 1 2 */ u16 ignore_alts:1; /* 80: 2 2 */ u16 hint:1; /* 80: 3 2 */ u16 save:1; /* 80: 4 2 */ u16 restore:1; /* 80: 5 2 */ u16 retpoline_safe:1; /* 80: 6 2 */ u16 noendbr:1; /* 80: 7 2 */ u16 entry:1; /* 80: 8 2 */ /* XXX 7 bits hole, try to pack */ s8 instr; /* 82 1 */ u8 visited; /* 83 1 */ /* XXX 4 bytes hole, try to pack */ struct alt_group * alt_group; /* 88 8 */ struct symbol * call_dest; /* 96 8 */ struct instruction * jump_dest; /* 104 8 */ struct instruction * first_jump_src; /* 112 8 */ struct reloc * jump_table; /* 120 8 */ /* --- cacheline 2 boundary (128 bytes) --- */ struct reloc * reloc; /* 128 8 */ - struct list_head alts; /* 136 16 */ - struct symbol * sym; /* 152 8 */ - struct stack_op * stack_ops; /* 160 8 */ - struct cfi_state * cfi; /* 168 8 */ + struct alternative * alts; /* 136 8 */ + struct symbol * sym; /* 144 8 */ + struct stack_op * stack_ops; /* 152 8 */ + struct cfi_state * cfi; /* 160 8 */ - /* size: 176, cachelines: 3, members: 29 */ - /* sum members: 170, holes: 1, sum holes: 4 */ + /* size: 168, cachelines: 3, members: 29 */ + /* sum members: 162, holes: 1, sum holes: 4 */ /* sum bitfield members: 9 bits, bit holes: 1, sum bit holes: 7 bits */ - /* last cacheline: 48 bytes */ + /* last cacheline: 40 bytes */ }; pre: 5:58.50 real, 229.64 user, 128.65 sys, 26221520 mem post: 5:48.86 real, 220.30 user, 128.34 sys, 24834672 mem Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Acked-by: Josh Poimboeuf Tested-by: Nathan Chancellor # build only Tested-by: Thomas Weißschuh # compile and run Link: https://lore.kernel.org/r/20230208172245.430556498@infradead.org --- tools/objtool/check.c | 18 +++++++++--------- tools/objtool/include/objtool/check.h | 2 +- 2 files changed, 10 insertions(+), 10 deletions(-) diff --git a/tools/objtool/check.c b/tools/objtool/check.c index 8109d7405297c..9f83e85e2093e 100644 --- a/tools/objtool/check.c +++ b/tools/objtool/check.c @@ -23,7 +23,7 @@ #include struct alternative { - struct list_head list; + struct alternative *next; struct instruction *insn; bool skip_orig; }; @@ -397,7 +397,6 @@ static int decode_instructions(struct objtool_file *file) return -1; } memset(insn, 0, sizeof(*insn)); - INIT_LIST_HEAD(&insn->alts); INIT_LIST_HEAD(&insn->call_node); insn->sec = sec; @@ -1780,7 +1779,6 @@ static int handle_group_alt(struct objtool_file *file, return -1; } memset(nop, 0, sizeof(*nop)); - INIT_LIST_HEAD(&nop->alts); nop->sec = special_alt->new_sec; nop->offset = special_alt->new_off + special_alt->new_len; @@ -1978,7 +1976,8 @@ static int add_special_section_alts(struct objtool_file *file) alt->insn = new_insn; alt->skip_orig = special_alt->skip_orig; orig_insn->ignore_alts |= special_alt->skip_alt; - list_add_tail(&alt->list, &orig_insn->alts); + alt->next = orig_insn->alts; + orig_insn->alts = alt; list_del(&special_alt->list); free(special_alt); @@ -2037,7 +2036,8 @@ static int add_jump_table(struct objtool_file *file, struct instruction *insn, } alt->insn = dest_insn; - list_add_tail(&alt->list, &insn->alts); + alt->next = insn->alts; + insn->alts = alt; prev_offset = reloc->offset; } @@ -3594,10 +3594,10 @@ static int validate_branch(struct objtool_file *file, struct symbol *func, if (propagate_alt_cfi(file, insn)) return 1; - if (!insn->ignore_alts && !list_empty(&insn->alts)) { + if (!insn->ignore_alts && insn->alts) { bool skip_orig = false; - list_for_each_entry(alt, &insn->alts, list) { + for (alt = insn->alts; alt; alt = alt->next) { if (alt->skip_orig) skip_orig = true; @@ -3796,11 +3796,11 @@ static int validate_entry(struct objtool_file *file, struct instruction *insn) insn->visited |= VISITED_ENTRY; - if (!insn->ignore_alts && !list_empty(&insn->alts)) { + if (!insn->ignore_alts && insn->alts) { struct alternative *alt; bool skip_orig = false; - list_for_each_entry(alt, &insn->alts, list) { + for (alt = insn->alts; alt; alt = alt->next) { if (alt->skip_orig) skip_orig = true; diff --git a/tools/objtool/include/objtool/check.h b/tools/objtool/include/objtool/check.h index 23e9819993655..7966f60f858b6 100644 --- a/tools/objtool/include/objtool/check.h +++ b/tools/objtool/include/objtool/check.h @@ -66,7 +66,7 @@ struct instruction { struct instruction *first_jump_src; struct reloc *jump_table; struct reloc *reloc; - struct list_head alts; + struct alternative *alts; struct symbol *sym; struct stack_op *stack_ops; struct cfi_state *cfi; From 8b2de412158ecdb312c707918432e6650df808cc Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 8 Feb 2023 18:18:00 +0100 Subject: [PATCH 410/765] objtool: Shrink instruction::{type,visited} MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since we don't have that many types in enum insn_type, force it into a u8 and re-arrange member to get rid of the holes, saves another 8 bytes. struct instruction { struct list_head list; /* 0 16 */ struct hlist_node hash; /* 16 16 */ struct list_head call_node; /* 32 16 */ struct section * sec; /* 48 8 */ long unsigned int offset; /* 56 8 */ /* --- cacheline 1 boundary (64 bytes) --- */ - unsigned int len; /* 64 4 */ - enum insn_type type; /* 68 4 */ - long unsigned int immediate; /* 72 8 */ - u16 dead_end:1; /* 80: 0 2 */ - u16 ignore:1; /* 80: 1 2 */ - u16 ignore_alts:1; /* 80: 2 2 */ - u16 hint:1; /* 80: 3 2 */ - u16 save:1; /* 80: 4 2 */ - u16 restore:1; /* 80: 5 2 */ - u16 retpoline_safe:1; /* 80: 6 2 */ - u16 noendbr:1; /* 80: 7 2 */ - u16 entry:1; /* 80: 8 2 */ + long unsigned int immediate; /* 64 8 */ + unsigned int len; /* 72 4 */ + u8 type; /* 76 1 */ - /* XXX 7 bits hole, try to pack */ + /* Bitfield combined with previous fields */ - s8 instr; /* 82 1 */ - u8 visited; /* 83 1 */ + u16 dead_end:1; /* 76: 8 2 */ + u16 ignore:1; /* 76: 9 2 */ + u16 ignore_alts:1; /* 76:10 2 */ + u16 hint:1; /* 76:11 2 */ + u16 save:1; /* 76:12 2 */ + u16 restore:1; /* 76:13 2 */ + u16 retpoline_safe:1; /* 76:14 2 */ + u16 noendbr:1; /* 76:15 2 */ + u16 entry:1; /* 78: 0 2 */ + u16 visited:4; /* 78: 1 2 */ - /* XXX 4 bytes hole, try to pack */ + /* XXX 3 bits hole, try to pack */ + /* Bitfield combined with next fields */ - struct alt_group * alt_group; /* 88 8 */ - struct symbol * call_dest; /* 96 8 */ - struct instruction * jump_dest; /* 104 8 */ - struct instruction * first_jump_src; /* 112 8 */ - struct reloc * jump_table; /* 120 8 */ + s8 instr; /* 79 1 */ + struct alt_group * alt_group; /* 80 8 */ + struct symbol * call_dest; /* 88 8 */ + struct instruction * jump_dest; /* 96 8 */ + struct instruction * first_jump_src; /* 104 8 */ + struct reloc * jump_table; /* 112 8 */ + struct reloc * reloc; /* 120 8 */ /* --- cacheline 2 boundary (128 bytes) --- */ - struct reloc * reloc; /* 128 8 */ - struct alternative * alts; /* 136 8 */ - struct symbol * sym; /* 144 8 */ - struct stack_op * stack_ops; /* 152 8 */ - struct cfi_state * cfi; /* 160 8 */ + struct alternative * alts; /* 128 8 */ + struct symbol * sym; /* 136 8 */ + struct stack_op * stack_ops; /* 144 8 */ + struct cfi_state * cfi; /* 152 8 */ - /* size: 168, cachelines: 3, members: 29 */ - /* sum members: 162, holes: 1, sum holes: 4 */ - /* sum bitfield members: 9 bits, bit holes: 1, sum bit holes: 7 bits */ - /* last cacheline: 40 bytes */ + /* size: 160, cachelines: 3, members: 29 */ + /* sum members: 158 */ + /* sum bitfield members: 13 bits, bit holes: 1, sum bit holes: 3 bits */ + /* last cacheline: 32 bytes */ }; pre: 5:48.86 real, 220.30 user, 128.34 sys, 24834672 mem post: 5:48.89 real, 220.96 user, 127.55 sys, 24834672 mem Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Acked-by: Josh Poimboeuf Tested-by: Nathan Chancellor # build only Tested-by: Thomas Weißschuh # compile and run Link: https://lore.kernel.org/r/20230208172245.501847188@infradead.org --- tools/objtool/include/objtool/check.h | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/tools/objtool/include/objtool/check.h b/tools/objtool/include/objtool/check.h index 7966f60f858b6..a497ee7672fbc 100644 --- a/tools/objtool/include/objtool/check.h +++ b/tools/objtool/include/objtool/check.h @@ -42,9 +42,9 @@ struct instruction { struct list_head call_node; struct section *sec; unsigned long offset; - unsigned int len; - enum insn_type type; unsigned long immediate; + unsigned int len; + u8 type; u16 dead_end : 1, ignore : 1, @@ -54,11 +54,11 @@ struct instruction { restore : 1, retpoline_safe : 1, noendbr : 1, - entry : 1; - /* 7 bit hole */ + entry : 1, + visited : 4; + /* 3 bit hole */ s8 instr; - u8 visited; struct alt_group *alt_group; struct symbol *call_dest; From 0932dbe1f5680481e612cafe0c7d0f1796f68612 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 8 Feb 2023 18:18:01 +0100 Subject: [PATCH 411/765] objtool: Remove instruction::reloc MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Instead of caching the reloc for each instruction, only keep a negative cache of not having a reloc (by far the most common case). struct instruction { struct list_head list; /* 0 16 */ struct hlist_node hash; /* 16 16 */ struct list_head call_node; /* 32 16 */ struct section * sec; /* 48 8 */ long unsigned int offset; /* 56 8 */ /* --- cacheline 1 boundary (64 bytes) --- */ long unsigned int immediate; /* 64 8 */ unsigned int len; /* 72 4 */ u8 type; /* 76 1 */ /* Bitfield combined with previous fields */ u16 dead_end:1; /* 76: 8 2 */ u16 ignore:1; /* 76: 9 2 */ u16 ignore_alts:1; /* 76:10 2 */ u16 hint:1; /* 76:11 2 */ u16 save:1; /* 76:12 2 */ u16 restore:1; /* 76:13 2 */ u16 retpoline_safe:1; /* 76:14 2 */ u16 noendbr:1; /* 76:15 2 */ u16 entry:1; /* 78: 0 2 */ u16 visited:4; /* 78: 1 2 */ + u16 no_reloc:1; /* 78: 5 2 */ - /* XXX 3 bits hole, try to pack */ + /* XXX 2 bits hole, try to pack */ /* Bitfield combined with next fields */ s8 instr; /* 79 1 */ struct alt_group * alt_group; /* 80 8 */ struct symbol * call_dest; /* 88 8 */ struct instruction * jump_dest; /* 96 8 */ struct instruction * first_jump_src; /* 104 8 */ struct reloc * jump_table; /* 112 8 */ - struct reloc * reloc; /* 120 8 */ + struct alternative * alts; /* 120 8 */ /* --- cacheline 2 boundary (128 bytes) --- */ - struct alternative * alts; /* 128 8 */ - struct symbol * sym; /* 136 8 */ - struct stack_op * stack_ops; /* 144 8 */ - struct cfi_state * cfi; /* 152 8 */ + struct symbol * sym; /* 128 8 */ + struct stack_op * stack_ops; /* 136 8 */ + struct cfi_state * cfi; /* 144 8 */ - /* size: 160, cachelines: 3, members: 29 */ - /* sum members: 158 */ - /* sum bitfield members: 13 bits, bit holes: 1, sum bit holes: 3 bits */ - /* last cacheline: 32 bytes */ + /* size: 152, cachelines: 3, members: 29 */ + /* sum members: 150 */ + /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */ + /* last cacheline: 24 bytes */ }; pre: 5:48.89 real, 220.96 user, 127.55 sys, 24834672 mem post: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Acked-by: Josh Poimboeuf Tested-by: Nathan Chancellor # build only Tested-by: Thomas Weißschuh # compile and run Link: https://lore.kernel.org/r/20230208172245.572145269@infradead.org --- tools/objtool/check.c | 24 +++++++++++------------- tools/objtool/include/objtool/check.h | 6 +++--- 2 files changed, 14 insertions(+), 16 deletions(-) diff --git a/tools/objtool/check.c b/tools/objtool/check.c index 9f83e85e2093e..6d0ce2395554c 100644 --- a/tools/objtool/check.c +++ b/tools/objtool/check.c @@ -1305,26 +1305,24 @@ __weak bool arch_is_rethunk(struct symbol *sym) return false; } -#define NEGATIVE_RELOC ((void *)-1L) - static struct reloc *insn_reloc(struct objtool_file *file, struct instruction *insn) { - if (insn->reloc == NEGATIVE_RELOC) + struct reloc *reloc; + + if (insn->no_reloc) return NULL; - if (!insn->reloc) { - if (!file) - return NULL; + if (!file) + return NULL; - insn->reloc = find_reloc_by_dest_range(file->elf, insn->sec, - insn->offset, insn->len); - if (!insn->reloc) { - insn->reloc = NEGATIVE_RELOC; - return NULL; - } + reloc = find_reloc_by_dest_range(file->elf, insn->sec, + insn->offset, insn->len); + if (!reloc) { + insn->no_reloc = 1; + return NULL; } - return insn->reloc; + return reloc; } static void remove_insn_ops(struct instruction *insn) diff --git a/tools/objtool/include/objtool/check.h b/tools/objtool/include/objtool/check.h index a497ee7672fbc..fffc8b86f9f37 100644 --- a/tools/objtool/include/objtool/check.h +++ b/tools/objtool/include/objtool/check.h @@ -55,8 +55,9 @@ struct instruction { retpoline_safe : 1, noendbr : 1, entry : 1, - visited : 4; - /* 3 bit hole */ + visited : 4, + no_reloc : 1; + /* 2 bit hole */ s8 instr; @@ -65,7 +66,6 @@ struct instruction { struct instruction *jump_dest; struct instruction *first_jump_src; struct reloc *jump_table; - struct reloc *reloc; struct alternative *alts; struct symbol *sym; struct stack_op *stack_ops; From c6f5dc28fb3d736fa8d7f7d31e0664a9c772c299 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 8 Feb 2023 18:18:02 +0100 Subject: [PATCH 412/765] objtool: Union instruction::{call_dest,jump_table} MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The instruction call_dest and jump_table members can never be used at the same time, their usage depends on type. struct instruction { struct list_head list; /* 0 16 */ struct hlist_node hash; /* 16 16 */ struct list_head call_node; /* 32 16 */ struct section * sec; /* 48 8 */ long unsigned int offset; /* 56 8 */ /* --- cacheline 1 boundary (64 bytes) --- */ long unsigned int immediate; /* 64 8 */ unsigned int len; /* 72 4 */ u8 type; /* 76 1 */ /* Bitfield combined with previous fields */ u16 dead_end:1; /* 76: 8 2 */ u16 ignore:1; /* 76: 9 2 */ u16 ignore_alts:1; /* 76:10 2 */ u16 hint:1; /* 76:11 2 */ u16 save:1; /* 76:12 2 */ u16 restore:1; /* 76:13 2 */ u16 retpoline_safe:1; /* 76:14 2 */ u16 noendbr:1; /* 76:15 2 */ u16 entry:1; /* 78: 0 2 */ u16 visited:4; /* 78: 1 2 */ u16 no_reloc:1; /* 78: 5 2 */ /* XXX 2 bits hole, try to pack */ /* Bitfield combined with next fields */ s8 instr; /* 79 1 */ struct alt_group * alt_group; /* 80 8 */ - struct symbol * call_dest; /* 88 8 */ - struct instruction * jump_dest; /* 96 8 */ - struct instruction * first_jump_src; /* 104 8 */ - struct reloc * jump_table; /* 112 8 */ - struct alternative * alts; /* 120 8 */ + struct instruction * jump_dest; /* 88 8 */ + struct instruction * first_jump_src; /* 96 8 */ + union { + struct symbol * _call_dest; /* 104 8 */ + struct reloc * _jump_table; /* 104 8 */ + }; /* 104 8 */ + struct alternative * alts; /* 112 8 */ + struct symbol * sym; /* 120 8 */ /* --- cacheline 2 boundary (128 bytes) --- */ - struct symbol * sym; /* 128 8 */ - struct stack_op * stack_ops; /* 136 8 */ - struct cfi_state * cfi; /* 144 8 */ + struct stack_op * stack_ops; /* 128 8 */ + struct cfi_state * cfi; /* 136 8 */ - /* size: 152, cachelines: 3, members: 29 */ - /* sum members: 150 */ + /* size: 144, cachelines: 3, members: 28 */ + /* sum members: 142 */ /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */ - /* last cacheline: 24 bytes */ + /* last cacheline: 16 bytes */ }; pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Acked-by: Josh Poimboeuf Tested-by: Nathan Chancellor # build only Tested-by: Thomas Weißschuh # compile and run Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org --- tools/objtool/check.c | 73 +++++++++++++++++---------- tools/objtool/include/objtool/check.h | 6 ++- 2 files changed, 50 insertions(+), 29 deletions(-) diff --git a/tools/objtool/check.c b/tools/objtool/check.c index 6d0ce2395554c..6f0adb2f76c66 100644 --- a/tools/objtool/check.c +++ b/tools/objtool/check.c @@ -114,16 +114,34 @@ static struct instruction *prev_insn_same_sym(struct objtool_file *file, for (insn = next_insn_same_sec(file, insn); insn; \ insn = next_insn_same_sec(file, insn)) +static inline struct symbol *insn_call_dest(struct instruction *insn) +{ + if (insn->type == INSN_JUMP_DYNAMIC || + insn->type == INSN_CALL_DYNAMIC) + return NULL; + + return insn->_call_dest; +} + +static inline struct reloc *insn_jump_table(struct instruction *insn) +{ + if (insn->type == INSN_JUMP_DYNAMIC || + insn->type == INSN_CALL_DYNAMIC) + return insn->_jump_table; + + return NULL; +} + static bool is_jump_table_jump(struct instruction *insn) { struct alt_group *alt_group = insn->alt_group; - if (insn->jump_table) + if (insn_jump_table(insn)) return true; /* Retpoline alternative for a jump table? */ return alt_group && alt_group->orig_group && - alt_group->orig_group->first_insn->jump_table; + insn_jump_table(alt_group->orig_group->first_insn); } static bool is_sibling_call(struct instruction *insn) @@ -137,8 +155,8 @@ static bool is_sibling_call(struct instruction *insn) return !is_jump_table_jump(insn); } - /* add_jump_destinations() sets insn->call_dest for sibling calls. */ - return (is_static_jump(insn) && insn->call_dest); + /* add_jump_destinations() sets insn_call_dest(insn) for sibling calls. */ + return (is_static_jump(insn) && insn_call_dest(insn)); } /* @@ -274,8 +292,8 @@ static void init_insn_state(struct objtool_file *file, struct insn_state *state, /* * We need the full vmlinux for noinstr validation, otherwise we can - * not correctly determine insn->call_dest->sec (external symbols do - * not have a section). + * not correctly determine insn_call_dest(insn)->sec (external symbols + * do not have a section). */ if (opts.link && opts.noinstr && sec) state->noinstr = sec->noinstr; @@ -678,7 +696,7 @@ static int create_static_call_sections(struct objtool_file *file) return -1; /* find key symbol */ - key_name = strdup(insn->call_dest->name); + key_name = strdup(insn_call_dest(insn)->name); if (!key_name) { perror("strdup"); return -1; @@ -709,7 +727,7 @@ static int create_static_call_sections(struct objtool_file *file) * trampoline address. This is fixed up in * static_call_add_module(). */ - key_sym = insn->call_dest; + key_sym = insn_call_dest(insn); } free(key_name); @@ -1340,7 +1358,7 @@ static void annotate_call_site(struct objtool_file *file, struct instruction *insn, bool sibling) { struct reloc *reloc = insn_reloc(file, insn); - struct symbol *sym = insn->call_dest; + struct symbol *sym = insn_call_dest(insn); if (!sym) sym = reloc->sym; @@ -1425,7 +1443,7 @@ static void annotate_call_site(struct objtool_file *file, static void add_call_dest(struct objtool_file *file, struct instruction *insn, struct symbol *dest, bool sibling) { - insn->call_dest = dest; + insn->_call_dest = dest; if (!dest) return; @@ -1683,12 +1701,12 @@ static int add_call_destinations(struct objtool_file *file) if (insn->ignore) continue; - if (!insn->call_dest) { + if (!insn_call_dest(insn)) { WARN_FUNC("unannotated intra-function call", insn->sec, insn->offset); return -1; } - if (insn_func(insn) && insn->call_dest->type != STT_FUNC) { + if (insn_func(insn) && insn_call_dest(insn)->type != STT_FUNC) { WARN_FUNC("unsupported call to non-function", insn->sec, insn->offset); return -1; @@ -2125,7 +2143,7 @@ static void mark_func_jump_tables(struct objtool_file *file, reloc = find_jump_table(file, func, insn); if (reloc) { reloc->jump_table_start = true; - insn->jump_table = reloc; + insn->_jump_table = reloc; } } } @@ -2137,10 +2155,10 @@ static int add_func_jump_tables(struct objtool_file *file, int ret; func_for_each_insn(file, func, insn) { - if (!insn->jump_table) + if (!insn_jump_table(insn)) continue; - ret = add_jump_table(file, insn, insn->jump_table); + ret = add_jump_table(file, insn, insn_jump_table(insn)); if (ret) return ret; } @@ -2612,8 +2630,8 @@ static int decode_sections(struct objtool_file *file) static bool is_fentry_call(struct instruction *insn) { if (insn->type == INSN_CALL && - insn->call_dest && - insn->call_dest->fentry) + insn_call_dest(insn) && + insn_call_dest(insn)->fentry) return true; return false; @@ -3320,8 +3338,8 @@ static inline const char *call_dest_name(struct instruction *insn) struct reloc *rel; int idx; - if (insn->call_dest) - return insn->call_dest->name; + if (insn_call_dest(insn)) + return insn_call_dest(insn)->name; rel = insn_reloc(NULL, insn); if (rel && !strcmp(rel->sym->name, "pv_ops")) { @@ -3403,13 +3421,13 @@ static int validate_call(struct objtool_file *file, struct insn_state *state) { if (state->noinstr && state->instr <= 0 && - !noinstr_call_dest(file, insn, insn->call_dest)) { + !noinstr_call_dest(file, insn, insn_call_dest(insn))) { WARN_FUNC("call to %s() leaves .noinstr.text section", insn->sec, insn->offset, call_dest_name(insn)); return 1; } - if (state->uaccess && !func_uaccess_safe(insn->call_dest)) { + if (state->uaccess && !func_uaccess_safe(insn_call_dest(insn))) { WARN_FUNC("call to %s() with UACCESS enabled", insn->sec, insn->offset, call_dest_name(insn)); return 1; @@ -3847,11 +3865,11 @@ static int validate_entry(struct objtool_file *file, struct instruction *insn) /* fallthrough */ case INSN_CALL: - dest = find_insn(file, insn->call_dest->sec, - insn->call_dest->offset); + dest = find_insn(file, insn_call_dest(insn)->sec, + insn_call_dest(insn)->offset); if (!dest) { WARN("Unresolved function after linking!?: %s", - insn->call_dest->name); + insn_call_dest(insn)->name); return -1; } @@ -3952,13 +3970,13 @@ static int validate_retpoline(struct objtool_file *file) static bool is_kasan_insn(struct instruction *insn) { return (insn->type == INSN_CALL && - !strcmp(insn->call_dest->name, "__asan_handle_no_return")); + !strcmp(insn_call_dest(insn)->name, "__asan_handle_no_return")); } static bool is_ubsan_insn(struct instruction *insn) { return (insn->type == INSN_CALL && - !strcmp(insn->call_dest->name, + !strcmp(insn_call_dest(insn)->name, "__ubsan_handle_builtin_unreachable")); } @@ -4036,7 +4054,8 @@ static bool ignore_unreachable_insn(struct objtool_file *file, struct instructio * It may also insert a UD2 after calling a __noreturn function. */ prev_insn = list_prev_entry(insn, list); - if ((prev_insn->dead_end || dead_end_function(file, prev_insn->call_dest)) && + if ((prev_insn->dead_end || + dead_end_function(file, insn_call_dest(prev_insn))) && (insn->type == INSN_BUG || (insn->type == INSN_JUMP_UNCONDITIONAL && insn->jump_dest && insn->jump_dest->type == INSN_BUG))) diff --git a/tools/objtool/include/objtool/check.h b/tools/objtool/include/objtool/check.h index fffc8b86f9f37..ab6deaed97773 100644 --- a/tools/objtool/include/objtool/check.h +++ b/tools/objtool/include/objtool/check.h @@ -62,10 +62,12 @@ struct instruction { s8 instr; struct alt_group *alt_group; - struct symbol *call_dest; struct instruction *jump_dest; struct instruction *first_jump_src; - struct reloc *jump_table; + union { + struct symbol *_call_dest; + struct reloc *_jump_table; + }; struct alternative *alts; struct symbol *sym; struct stack_op *stack_ops; From a706bb08c81ac878982e41d4b6abcc42258bd39e Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 8 Feb 2023 18:18:03 +0100 Subject: [PATCH 413/765] objtool: Fix overlapping alternatives MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Things like ALTERNATIVE_{2,3}() generate multiple alternatives on the same place, objtool would override the first orig_alt_group with the second (or third), failing to check the CFI among all the different variants. Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Acked-by: Josh Poimboeuf Tested-by: Nathan Chancellor # build only Tested-by: Thomas Weißschuh # compile and run Link: https://lore.kernel.org/r/20230208172245.711471461@infradead.org --- tools/objtool/check.c | 69 +++++++++++++++++++++++++++---------------- 1 file changed, 43 insertions(+), 26 deletions(-) diff --git a/tools/objtool/check.c b/tools/objtool/check.c index 6f0adb2f76c66..7e9d3d3eed654 100644 --- a/tools/objtool/check.c +++ b/tools/objtool/check.c @@ -1744,36 +1744,49 @@ static int handle_group_alt(struct objtool_file *file, struct instruction *orig_insn, struct instruction **new_insn) { - struct instruction *last_orig_insn, *last_new_insn = NULL, *insn, *nop = NULL; + struct instruction *last_new_insn = NULL, *insn, *nop = NULL; struct alt_group *orig_alt_group, *new_alt_group; unsigned long dest_off; - - orig_alt_group = malloc(sizeof(*orig_alt_group)); + orig_alt_group = orig_insn->alt_group; if (!orig_alt_group) { - WARN("malloc failed"); - return -1; - } - orig_alt_group->cfi = calloc(special_alt->orig_len, - sizeof(struct cfi_state *)); - if (!orig_alt_group->cfi) { - WARN("calloc failed"); - return -1; - } + struct instruction *last_orig_insn = NULL; - last_orig_insn = NULL; - insn = orig_insn; - sec_for_each_insn_from(file, insn) { - if (insn->offset >= special_alt->orig_off + special_alt->orig_len) - break; + orig_alt_group = malloc(sizeof(*orig_alt_group)); + if (!orig_alt_group) { + WARN("malloc failed"); + return -1; + } + orig_alt_group->cfi = calloc(special_alt->orig_len, + sizeof(struct cfi_state *)); + if (!orig_alt_group->cfi) { + WARN("calloc failed"); + return -1; + } - insn->alt_group = orig_alt_group; - last_orig_insn = insn; - } - orig_alt_group->orig_group = NULL; - orig_alt_group->first_insn = orig_insn; - orig_alt_group->last_insn = last_orig_insn; + insn = orig_insn; + sec_for_each_insn_from(file, insn) { + if (insn->offset >= special_alt->orig_off + special_alt->orig_len) + break; + insn->alt_group = orig_alt_group; + last_orig_insn = insn; + } + orig_alt_group->orig_group = NULL; + orig_alt_group->first_insn = orig_insn; + orig_alt_group->last_insn = last_orig_insn; + } else { + if (orig_alt_group->last_insn->offset + orig_alt_group->last_insn->len - + orig_alt_group->first_insn->offset != special_alt->orig_len) { + WARN_FUNC("weirdly overlapping alternative! %ld != %d", + orig_insn->sec, orig_insn->offset, + orig_alt_group->last_insn->offset + + orig_alt_group->last_insn->len - + orig_alt_group->first_insn->offset, + special_alt->orig_len); + return -1; + } + } new_alt_group = malloc(sizeof(*new_alt_group)); if (!new_alt_group) { @@ -1848,7 +1861,7 @@ static int handle_group_alt(struct objtool_file *file, dest_off = arch_jump_destination(insn); if (dest_off == special_alt->new_off + special_alt->new_len) { - insn->jump_dest = next_insn_same_sec(file, last_orig_insn); + insn->jump_dest = next_insn_same_sec(file, orig_alt_group->last_insn); if (!insn->jump_dest) { WARN_FUNC("can't find alternative jump destination", insn->sec, insn->offset); @@ -3226,8 +3239,12 @@ static int propagate_alt_cfi(struct objtool_file *file, struct instruction *insn alt_cfi[group_off] = insn->cfi; } else { if (cficmp(alt_cfi[group_off], insn->cfi)) { - WARN_FUNC("stack layout conflict in alternatives", - insn->sec, insn->offset); + struct alt_group *orig_group = insn->alt_group->orig_group ?: insn->alt_group; + struct instruction *orig = orig_group->first_insn; + char *where = offstr(insn->sec, insn->offset); + WARN_FUNC("stack layout conflict in alternatives: %s", + orig->sec, orig->offset, where); + free(where); return -1; } } From 6ea17e848a8ba5138b30e936c4b71877bc972c13 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 8 Feb 2023 18:18:04 +0100 Subject: [PATCH 414/765] x86: Fix FILL_RETURN_BUFFER MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit With overlapping alternative validation fixed, objtool promptly complains: vmlinux.o: warning: objtool: __switch_to_asm+0x2c: stack layout conflict in alternatives: .altinstr_replacement+0x47 .rela.altinstructions: 000000000000009c 0000000200000002 R_X86_64_PC32 0000000000000000 .text + 16dc 00000000000000a0 0000000600000002 R_X86_64_PC32 0000000000000000 .altinstr_replacement + 3a 00000000000000a8 0000000200000002 R_X86_64_PC32 0000000000000000 .text + 16dc 00000000000000ac 0000000600000002 R_X86_64_PC32 0000000000000000 .altinstr_replacement + 66 .text: 00000000000016b0 <__switch_to_asm>: 16b0: f3 0f 1e fa endbr64 16b4: 55 push %rbp 16b5: 53 push %rbx 16b6: 41 54 push %r12 16b8: 41 55 push %r13 16ba: 41 56 push %r14 16bc: 41 57 push %r15 16be: 48 89 a7 18 0b 00 00 mov %rsp,0xb18(%rdi) 16c5: 48 8b a6 18 0b 00 00 mov 0xb18(%rsi),%rsp 16cc: 48 8b 9e 28 05 00 00 mov 0x528(%rsi),%rbx 16d3: 65 48 89 1c 25 00 00 00 00 mov %rbx,%gs:0x0 16d8: R_X86_64_32S fixed_percpu_data+0x28 16dc: eb 2a jmp 1708 <__switch_to_asm+0x58> 16de: 90 nop 16df: 90 nop 16e0: 90 nop 16e1: 90 nop 16e2: 90 nop 16e3: 90 nop 16e4: 90 nop 16e5: 90 nop 16e6: 90 nop 16e7: 90 nop 16e8: 90 nop 16e9: 90 nop 16ea: 90 nop 16eb: 90 nop 16ec: 90 nop 16ed: 90 nop 16ee: 90 nop 16ef: 90 nop 16f0: 90 nop 16f1: 90 nop 16f2: 90 nop 16f3: 90 nop 16f4: 90 nop 16f5: 90 nop 16f6: 90 nop 16f7: 90 nop 16f8: 90 nop 16f9: 90 nop 16fa: 90 nop 16fb: 90 nop 16fc: 90 nop 16fd: 90 nop 16fe: 90 nop 16ff: 90 nop 1700: 90 nop 1701: 90 nop 1702: 90 nop 1703: 90 nop 1704: 90 nop 1705: 90 nop 1706: 90 nop 1707: 90 nop 1708: 41 5f pop %r15 170a: 41 5e pop %r14 170c: 41 5d pop %r13 170e: 41 5c pop %r12 1710: 5b pop %rbx 1711: 5d pop %rbp 1712: e9 00 00 00 00 jmp 1717 <__switch_to_asm+0x67> 1713: R_X86_64_PLT32 __switch_to-0x4 .altinstr_replacement: 3a: 49 c7 c4 10 00 00 00 mov $0x10,%r12 41: e8 01 00 00 00 call 47 <.altinstr_replacement+0x47> 46: cc int3 47: e8 01 00 00 00 call 4d <.altinstr_replacement+0x4d> 4c: cc int3 4d: 48 83 c4 10 add $0x10,%rsp 51: 49 ff cc dec %r12 54: 75 eb jne 41 <.altinstr_replacement+0x41> 56: 0f ae e8 lfence 59: 65 48 c7 04 25 00 00 00 00 ff ff ff ff movq $0xffffffffffffffff,%gs:0x0 5e: R_X86_64_32S pcpu_hot+0x10 66: e8 01 00 00 00 call 6c <.altinstr_replacement+0x6c> 6b: cc int3 6c: 48 83 c4 08 add $0x8,%rsp 70: 0f ae e8 lfence As can be seen from the two alternatives, when overlaid, the NOP after the shorter (starting at 66) coinsides with the call at 47, leading to conflicting CFI state for that instruction. By offsetting the shorter alternative by 2 bytes, this alignment is undone. Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Acked-by: Josh Poimboeuf Tested-by: Nathan Chancellor # build only Tested-by: Thomas Weißschuh # compile and run Link: https://lore.kernel.org/r/20230208172245.783099843@infradead.org --- arch/x86/include/asm/nospec-branch.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index e04313e89f4f5..3ef70e54a858a 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -261,7 +261,7 @@ .macro FILL_RETURN_BUFFER reg:req nr:req ftr:req ftr2=ALT_NOT(X86_FEATURE_ALWAYS) ALTERNATIVE_2 "jmp .Lskip_rsb_\@", \ __stringify(__FILL_RETURN_BUFFER(\reg,\nr)), \ftr, \ - __stringify(__FILL_ONE_RETURN), \ftr2 + __stringify(nop;nop;__FILL_ONE_RETURN), \ftr2 .Lskip_rsb_\@: .endm From 1c34496e5856886d565665fb64029ecdeb080ffb Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 8 Feb 2023 18:18:05 +0100 Subject: [PATCH 415/765] objtool: Remove instruction::list MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Replace the instruction::list by allocating instructions in arrays of 256 entries and stringing them together by (amortized) find_insn(). This shrinks instruction by 16 bytes and brings it down to 128. struct instruction { - struct list_head list; /* 0 16 */ - struct hlist_node hash; /* 16 16 */ - struct list_head call_node; /* 32 16 */ - struct section * sec; /* 48 8 */ - long unsigned int offset; /* 56 8 */ - /* --- cacheline 1 boundary (64 bytes) --- */ - long unsigned int immediate; /* 64 8 */ - unsigned int len; /* 72 4 */ - u8 type; /* 76 1 */ - - /* Bitfield combined with previous fields */ + struct hlist_node hash; /* 0 16 */ + struct list_head call_node; /* 16 16 */ + struct section * sec; /* 32 8 */ + long unsigned int offset; /* 40 8 */ + long unsigned int immediate; /* 48 8 */ + u8 len; /* 56 1 */ + u8 prev_len; /* 57 1 */ + u8 type; /* 58 1 */ + s8 instr; /* 59 1 */ + u32 idx:8; /* 60: 0 4 */ + u32 dead_end:1; /* 60: 8 4 */ + u32 ignore:1; /* 60: 9 4 */ + u32 ignore_alts:1; /* 60:10 4 */ + u32 hint:1; /* 60:11 4 */ + u32 save:1; /* 60:12 4 */ + u32 restore:1; /* 60:13 4 */ + u32 retpoline_safe:1; /* 60:14 4 */ + u32 noendbr:1; /* 60:15 4 */ + u32 entry:1; /* 60:16 4 */ + u32 visited:4; /* 60:17 4 */ + u32 no_reloc:1; /* 60:21 4 */ - u16 dead_end:1; /* 76: 8 2 */ - u16 ignore:1; /* 76: 9 2 */ - u16 ignore_alts:1; /* 76:10 2 */ - u16 hint:1; /* 76:11 2 */ - u16 save:1; /* 76:12 2 */ - u16 restore:1; /* 76:13 2 */ - u16 retpoline_safe:1; /* 76:14 2 */ - u16 noendbr:1; /* 76:15 2 */ - u16 entry:1; /* 78: 0 2 */ - u16 visited:4; /* 78: 1 2 */ - u16 no_reloc:1; /* 78: 5 2 */ + /* XXX 10 bits hole, try to pack */ - /* XXX 2 bits hole, try to pack */ - /* Bitfield combined with next fields */ - - s8 instr; /* 79 1 */ - struct alt_group * alt_group; /* 80 8 */ - struct instruction * jump_dest; /* 88 8 */ - struct instruction * first_jump_src; /* 96 8 */ + /* --- cacheline 1 boundary (64 bytes) --- */ + struct alt_group * alt_group; /* 64 8 */ + struct instruction * jump_dest; /* 72 8 */ + struct instruction * first_jump_src; /* 80 8 */ union { - struct symbol * _call_dest; /* 104 8 */ - struct reloc * _jump_table; /* 104 8 */ - }; /* 104 8 */ - struct alternative * alts; /* 112 8 */ - struct symbol * sym; /* 120 8 */ - /* --- cacheline 2 boundary (128 bytes) --- */ - struct stack_op * stack_ops; /* 128 8 */ - struct cfi_state * cfi; /* 136 8 */ + struct symbol * _call_dest; /* 88 8 */ + struct reloc * _jump_table; /* 88 8 */ + }; /* 88 8 */ + struct alternative * alts; /* 96 8 */ + struct symbol * sym; /* 104 8 */ + struct stack_op * stack_ops; /* 112 8 */ + struct cfi_state * cfi; /* 120 8 */ - /* size: 144, cachelines: 3, members: 28 */ - /* sum members: 142 */ - /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */ - /* last cacheline: 16 bytes */ + /* size: 128, cachelines: 2, members: 29 */ + /* sum members: 124 */ + /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */ }; pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Acked-by: Josh Poimboeuf Tested-by: Nathan Chancellor # build only Tested-by: Thomas Weißschuh # compile and run Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org --- tools/objtool/check.c | 166 +++++++++++++++--------- tools/objtool/include/objtool/check.h | 51 ++++---- tools/objtool/include/objtool/objtool.h | 1 - tools/objtool/objtool.c | 1 - 4 files changed, 133 insertions(+), 86 deletions(-) diff --git a/tools/objtool/check.c b/tools/objtool/check.c index 7e9d3d3eed654..b0b467d9608a1 100644 --- a/tools/objtool/check.c +++ b/tools/objtool/check.c @@ -47,27 +47,29 @@ struct instruction *find_insn(struct objtool_file *file, return NULL; } -static struct instruction *next_insn_same_sec(struct objtool_file *file, - struct instruction *insn) +struct instruction *next_insn_same_sec(struct objtool_file *file, + struct instruction *insn) { - struct instruction *next = list_next_entry(insn, list); + if (insn->idx == INSN_CHUNK_MAX) + return find_insn(file, insn->sec, insn->offset + insn->len); - if (!next || &next->list == &file->insn_list || next->sec != insn->sec) + insn++; + if (!insn->len) return NULL; - return next; + return insn; } static struct instruction *next_insn_same_func(struct objtool_file *file, struct instruction *insn) { - struct instruction *next = list_next_entry(insn, list); + struct instruction *next = next_insn_same_sec(file, insn); struct symbol *func = insn_func(insn); if (!func) return NULL; - if (&next->list != &file->insn_list && insn_func(next) == func) + if (next && insn_func(next) == func) return next; /* Check if we're already in the subfunction: */ @@ -78,17 +80,35 @@ static struct instruction *next_insn_same_func(struct objtool_file *file, return find_insn(file, func->cfunc->sec, func->cfunc->offset); } +static struct instruction *prev_insn_same_sec(struct objtool_file *file, + struct instruction *insn) +{ + if (insn->idx == 0) { + if (insn->prev_len) + return find_insn(file, insn->sec, insn->offset - insn->prev_len); + return NULL; + } + + return insn - 1; +} + static struct instruction *prev_insn_same_sym(struct objtool_file *file, - struct instruction *insn) + struct instruction *insn) { - struct instruction *prev = list_prev_entry(insn, list); + struct instruction *prev = prev_insn_same_sec(file, insn); - if (&prev->list != &file->insn_list && insn_func(prev) == insn_func(insn)) + if (prev && insn_func(prev) == insn_func(insn)) return prev; return NULL; } +#define for_each_insn(file, insn) \ + for (struct section *__sec, *__fake = (struct section *)1; \ + __fake; __fake = NULL) \ + for_each_sec(file, __sec) \ + sec_for_each_insn(file, __sec, insn) + #define func_for_each_insn(file, func, insn) \ for (insn = find_insn(file, func->sec, func->offset); \ insn; \ @@ -96,16 +116,13 @@ static struct instruction *prev_insn_same_sym(struct objtool_file *file, #define sym_for_each_insn(file, sym, insn) \ for (insn = find_insn(file, sym->sec, sym->offset); \ - insn && &insn->list != &file->insn_list && \ - insn->sec == sym->sec && \ - insn->offset < sym->offset + sym->len; \ - insn = list_next_entry(insn, list)) + insn && insn->offset < sym->offset + sym->len; \ + insn = next_insn_same_sec(file, insn)) #define sym_for_each_insn_continue_reverse(file, sym, insn) \ - for (insn = list_prev_entry(insn, list); \ - &insn->list != &file->insn_list && \ - insn->sec == sym->sec && insn->offset >= sym->offset; \ - insn = list_prev_entry(insn, list)) + for (insn = prev_insn_same_sec(file, insn); \ + insn && insn->offset >= sym->offset; \ + insn = prev_insn_same_sec(file, insn)) #define sec_for_each_insn_from(file, insn) \ for (; insn; insn = next_insn_same_sec(file, insn)) @@ -384,6 +401,9 @@ static int decode_instructions(struct objtool_file *file) int ret; for_each_sec(file, sec) { + struct instruction *insns = NULL; + u8 prev_len = 0; + u8 idx = 0; if (!(sec->sh.sh_flags & SHF_EXECINSTR)) continue; @@ -409,22 +429,31 @@ static int decode_instructions(struct objtool_file *file) sec->init = true; for (offset = 0; offset < sec->sh.sh_size; offset += insn->len) { - insn = malloc(sizeof(*insn)); - if (!insn) { - WARN("malloc failed"); - return -1; + if (!insns || idx == INSN_CHUNK_MAX) { + insns = calloc(sizeof(*insn), INSN_CHUNK_SIZE); + if (!insns) { + WARN("malloc failed"); + return -1; + } + idx = 0; + } else { + idx++; } - memset(insn, 0, sizeof(*insn)); - INIT_LIST_HEAD(&insn->call_node); + insn = &insns[idx]; + insn->idx = idx; + INIT_LIST_HEAD(&insn->call_node); insn->sec = sec; insn->offset = offset; + insn->prev_len = prev_len; ret = arch_decode_instruction(file, sec, offset, sec->sh.sh_size - offset, insn); if (ret) - goto err; + return ret; + + prev_len = insn->len; /* * By default, "ud2" is a dead end unless otherwise @@ -435,10 +464,11 @@ static int decode_instructions(struct objtool_file *file) insn->dead_end = true; hash_add(file->insn_hash, &insn->hash, sec_offset_hash(sec, insn->offset)); - list_add_tail(&insn->list, &file->insn_list); nr_insns++; } +// printf("%s: last chunk used: %d\n", sec->name, (int)idx); + list_for_each_entry(func, &sec->symbol_list, list) { if (func->type != STT_NOTYPE && func->type != STT_FUNC) continue; @@ -481,10 +511,6 @@ static int decode_instructions(struct objtool_file *file) printf("nr_insns: %lu\n", nr_insns); return 0; - -err: - free(insn); - return ret; } /* @@ -599,7 +625,7 @@ static int add_dead_ends(struct objtool_file *file) } insn = find_insn(file, reloc->sym->sec, reloc->addend); if (insn) - insn = list_prev_entry(insn, list); + insn = prev_insn_same_sec(file, insn); else if (reloc->addend == reloc->sym->sec->sh.sh_size) { insn = find_last_insn(file, reloc->sym->sec); if (!insn) { @@ -634,7 +660,7 @@ static int add_dead_ends(struct objtool_file *file) } insn = find_insn(file, reloc->sym->sec, reloc->addend); if (insn) - insn = list_prev_entry(insn, list); + insn = prev_insn_same_sec(file, insn); else if (reloc->addend == reloc->sym->sec->sh.sh_size) { insn = find_last_insn(file, reloc->sym->sec); if (!insn) { @@ -1775,6 +1801,7 @@ static int handle_group_alt(struct objtool_file *file, orig_alt_group->orig_group = NULL; orig_alt_group->first_insn = orig_insn; orig_alt_group->last_insn = last_orig_insn; + orig_alt_group->nop = NULL; } else { if (orig_alt_group->last_insn->offset + orig_alt_group->last_insn->len - orig_alt_group->first_insn->offset != special_alt->orig_len) { @@ -1876,12 +1903,11 @@ static int handle_group_alt(struct objtool_file *file, return -1; } - if (nop) - list_add(&nop->list, &last_new_insn->list); end: new_alt_group->orig_group = orig_alt_group; new_alt_group->first_insn = *new_insn; - new_alt_group->last_insn = nop ? : last_new_insn; + new_alt_group->last_insn = last_new_insn; + new_alt_group->nop = nop; new_alt_group->cfi = orig_alt_group->cfi; return 0; } @@ -1931,7 +1957,7 @@ static int handle_jump_alt(struct objtool_file *file, else file->jl_long++; - *new_insn = list_next_entry(orig_insn, list); + *new_insn = next_insn_same_sec(file, orig_insn); return 0; } @@ -3522,11 +3548,28 @@ static struct instruction *next_insn_to_validate(struct objtool_file *file, * Simulate the fact that alternatives are patched in-place. When the * end of a replacement alt_group is reached, redirect objtool flow to * the end of the original alt_group. + * + * insn->alts->insn -> alt_group->first_insn + * ... + * alt_group->last_insn + * [alt_group->nop] -> next(orig_group->last_insn) */ - if (alt_group && insn == alt_group->last_insn && alt_group->orig_group) - return next_insn_same_sec(file, alt_group->orig_group->last_insn); + if (alt_group) { + if (alt_group->nop) { + /* ->nop implies ->orig_group */ + if (insn == alt_group->last_insn) + return alt_group->nop; + if (insn == alt_group->nop) + goto next_orig; + } + if (insn == alt_group->last_insn && alt_group->orig_group) + goto next_orig; + } return next_insn_same_sec(file, insn); + +next_orig: + return next_insn_same_sec(file, alt_group->orig_group->last_insn); } /* @@ -3777,11 +3820,25 @@ static int validate_branch(struct objtool_file *file, struct symbol *func, return 0; } +static int validate_unwind_hint(struct objtool_file *file, + struct instruction *insn, + struct insn_state *state) +{ + if (insn->hint && !insn->visited && !insn->ignore) { + int ret = validate_branch(file, insn_func(insn), insn, *state); + if (ret && opts.backtrace) + BT_FUNC("<=== (hint)", insn); + return ret; + } + + return 0; +} + static int validate_unwind_hints(struct objtool_file *file, struct section *sec) { struct instruction *insn; struct insn_state state; - int ret, warnings = 0; + int warnings = 0; if (!file->hints) return 0; @@ -3789,22 +3846,11 @@ static int validate_unwind_hints(struct objtool_file *file, struct section *sec) init_insn_state(file, &state, sec); if (sec) { - insn = find_insn(file, sec, 0); - if (!insn) - return 0; + sec_for_each_insn(file, sec, insn) + warnings += validate_unwind_hint(file, insn, &state); } else { - insn = list_first_entry(&file->insn_list, typeof(*insn), list); - } - - while (&insn->list != &file->insn_list && (!sec || insn->sec == sec)) { - if (insn->hint && !insn->visited && !insn->ignore) { - ret = validate_branch(file, insn_func(insn), insn, state); - if (ret && opts.backtrace) - BT_FUNC("<=== (hint)", insn); - warnings += ret; - } - - insn = list_next_entry(insn, list); + for_each_insn(file, insn) + warnings += validate_unwind_hint(file, insn, &state); } return warnings; @@ -4070,7 +4116,7 @@ static bool ignore_unreachable_insn(struct objtool_file *file, struct instructio * * It may also insert a UD2 after calling a __noreturn function. */ - prev_insn = list_prev_entry(insn, list); + prev_insn = prev_insn_same_sec(file, insn); if ((prev_insn->dead_end || dead_end_function(file, insn_call_dest(prev_insn))) && (insn->type == INSN_BUG || @@ -4102,7 +4148,7 @@ static bool ignore_unreachable_insn(struct objtool_file *file, struct instructio if (insn->offset + insn->len >= insn_func(insn)->offset + insn_func(insn)->len) break; - insn = list_next_entry(insn, list); + insn = next_insn_same_sec(file, insn); } return false; @@ -4115,10 +4161,10 @@ static int add_prefix_symbol(struct objtool_file *file, struct symbol *func, return 0; for (;;) { - struct instruction *prev = list_prev_entry(insn, list); + struct instruction *prev = prev_insn_same_sec(file, insn); u64 offset; - if (&prev->list == &file->insn_list) + if (!prev) break; if (prev->type != INSN_NOP) @@ -4517,7 +4563,7 @@ int check(struct objtool_file *file) warnings += ret; - if (list_empty(&file->insn_list)) + if (!nr_insns) goto out; if (opts.retpoline) { @@ -4626,7 +4672,7 @@ int check(struct objtool_file *file) warnings += ret; } - if (opts.orc && !list_empty(&file->insn_list)) { + if (opts.orc && nr_insns) { ret = orc_create(file); if (ret < 0) goto out; diff --git a/tools/objtool/include/objtool/check.h b/tools/objtool/include/objtool/check.h index ab6deaed97773..3e7c7004f7df2 100644 --- a/tools/objtool/include/objtool/check.h +++ b/tools/objtool/include/objtool/check.h @@ -27,7 +27,7 @@ struct alt_group { struct alt_group *orig_group; /* First and last instructions in the group */ - struct instruction *first_insn, *last_insn; + struct instruction *first_insn, *last_insn, *nop; /* * Byte-offset-addressed len-sized array of pointers to CFI structs. @@ -36,31 +36,36 @@ struct alt_group { struct cfi_state **cfi; }; +#define INSN_CHUNK_BITS 8 +#define INSN_CHUNK_SIZE (1 << INSN_CHUNK_BITS) +#define INSN_CHUNK_MAX (INSN_CHUNK_SIZE - 1) + struct instruction { - struct list_head list; struct hlist_node hash; struct list_head call_node; struct section *sec; unsigned long offset; unsigned long immediate; - unsigned int len; - u8 type; - - u16 dead_end : 1, - ignore : 1, - ignore_alts : 1, - hint : 1, - save : 1, - restore : 1, - retpoline_safe : 1, - noendbr : 1, - entry : 1, - visited : 4, - no_reloc : 1; - /* 2 bit hole */ + u8 len; + u8 prev_len; + u8 type; s8 instr; + u32 idx : INSN_CHUNK_BITS, + dead_end : 1, + ignore : 1, + ignore_alts : 1, + hint : 1, + save : 1, + restore : 1, + retpoline_safe : 1, + noendbr : 1, + entry : 1, + visited : 4, + no_reloc : 1; + /* 10 bit hole */ + struct alt_group *alt_group; struct instruction *jump_dest; struct instruction *first_jump_src; @@ -109,13 +114,11 @@ static inline bool is_jump(struct instruction *insn) struct instruction *find_insn(struct objtool_file *file, struct section *sec, unsigned long offset); -#define for_each_insn(file, insn) \ - list_for_each_entry(insn, &file->insn_list, list) +struct instruction *next_insn_same_sec(struct objtool_file *file, struct instruction *insn); -#define sec_for_each_insn(file, sec, insn) \ - for (insn = find_insn(file, sec, 0); \ - insn && &insn->list != &file->insn_list && \ - insn->sec == sec; \ - insn = list_next_entry(insn, list)) +#define sec_for_each_insn(file, _sec, insn) \ + for (insn = find_insn(file, _sec, 0); \ + insn && insn->sec == _sec; \ + insn = next_insn_same_sec(file, insn)) #endif /* _CHECK_H */ diff --git a/tools/objtool/include/objtool/objtool.h b/tools/objtool/include/objtool/objtool.h index 6b40977bcdb1a..94a33ee7b3630 100644 --- a/tools/objtool/include/objtool/objtool.h +++ b/tools/objtool/include/objtool/objtool.h @@ -21,7 +21,6 @@ struct pv_state { struct objtool_file { struct elf *elf; - struct list_head insn_list; DECLARE_HASHTABLE(insn_hash, 20); struct list_head retpoline_call_list; struct list_head return_thunk_list; diff --git a/tools/objtool/objtool.c b/tools/objtool/objtool.c index 6affd8067f837..c54f7235c5d94 100644 --- a/tools/objtool/objtool.c +++ b/tools/objtool/objtool.c @@ -99,7 +99,6 @@ struct objtool_file *objtool_open_read(const char *_objname) return NULL; } - INIT_LIST_HEAD(&file.insn_list); hash_init(file.insn_hash); INIT_LIST_HEAD(&file.retpoline_call_list); INIT_LIST_HEAD(&file.return_thunk_list); From 00c8f01c4e84637c3db76f368b8687cb61f4dd9d Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Thu, 16 Feb 2023 12:34:41 -0800 Subject: [PATCH 416/765] objtool: Fix ORC 'signal' propagation There have been some recently reported ORC unwinder warnings like: WARNING: can't access registers at entry_SYSCALL_64_after_hwframe+0x63/0xcd WARNING: stack going in the wrong direction? at __sys_setsockopt+0x2c6/0x5b0 net/socket.c:2271 And a KASAN warning: BUG: KASAN: stack-out-of-bounds in unwind_next_frame (arch/x86/include/asm/ptrace.h:136 arch/x86/kernel/unwind_orc.c:455) It turns out the 'signal' bit isn't getting propagated from the unwind hints to the ORC entries, making the unwinder confused at times. Fixes: ffb1b4a41016 ("x86/unwind/orc: Add 'signal' field to ORC metadata") Reported-by: kernel test robot Reported-by: Dmitry Vyukov Signed-off-by: Josh Poimboeuf Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/97eef9db60cd86d376a9a40d49d77bb67a8f6526.1676579666.git.jpoimboe@kernel.org --- tools/objtool/check.c | 1 + tools/objtool/include/objtool/cfi.h | 1 + tools/objtool/orc_gen.c | 1 + 3 files changed, 3 insertions(+) diff --git a/tools/objtool/check.c b/tools/objtool/check.c index b0b467d9608a1..5822de376d9a8 100644 --- a/tools/objtool/check.c +++ b/tools/objtool/check.c @@ -2330,6 +2330,7 @@ static int read_unwind_hints(struct objtool_file *file) cfi.cfa.offset = bswap_if_needed(file->elf, hint->sp_offset); cfi.type = hint->type; + cfi.signal = hint->signal; cfi.end = hint->end; insn->cfi = cfi_hash_find_or_add(&cfi); diff --git a/tools/objtool/include/objtool/cfi.h b/tools/objtool/include/objtool/cfi.h index f11d1ac1dadf1..b1258e79a1b7e 100644 --- a/tools/objtool/include/objtool/cfi.h +++ b/tools/objtool/include/objtool/cfi.h @@ -34,6 +34,7 @@ struct cfi_state { unsigned char type; bool bp_scratch; bool drap; + bool signal; bool end; }; diff --git a/tools/objtool/orc_gen.c b/tools/objtool/orc_gen.c index 1f22b7ebae588..57a4527d59882 100644 --- a/tools/objtool/orc_gen.c +++ b/tools/objtool/orc_gen.c @@ -27,6 +27,7 @@ static int init_orc_entry(struct orc_entry *orc, struct cfi_state *cfi, } orc->end = cfi->end; + orc->signal = cfi->signal; if (cfi->cfa.base == CFI_UNDEFINED) { orc->sp_reg = ORC_REG_UNDEFINED; From b1a37ed00d7908a991c1d0f18a8cba3c2aa99bdc Mon Sep 17 00:00:00 2001 From: Lee Jones Date: Mon, 23 Jan 2023 12:39:11 +0000 Subject: [PATCH 417/765] HID: core: Provide new max_buffer_size attribute to over-ride the default Presently, when a report is processed, its proposed size, provided by the user of the API (as Report Size * Report Count) is compared against the subsystem default HID_MAX_BUFFER_SIZE (16k). However, some low-level HID drivers allocate a reduced amount of memory to their buffers (e.g. UHID only allocates UHID_DATA_MAX (4k) buffers), rending this check inadequate in some cases. In these circumstances, if the received report ends up being smaller than the proposed report size, the remainder of the buffer is zeroed. That is, the space between sizeof(csize) (size of the current report) and the rsize (size proposed i.e. Report Size * Report Count), which can be handled up to HID_MAX_BUFFER_SIZE (16k). Meaning that memset() shoots straight past the end of the buffer boundary and starts zeroing out in-use values, often resulting in calamity. This patch introduces a new variable into 'struct hid_ll_driver' where individual low-level drivers can over-ride the default maximum value of HID_MAX_BUFFER_SIZE (16k) with something more sympathetic to the interface. Signed-off-by: Lee Jones Signed-off-by: Jiri Kosina --- drivers/hid/hid-core.c | 32 +++++++++++++++++++++++++------- include/linux/hid.h | 3 +++ 2 files changed, 28 insertions(+), 7 deletions(-) diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c index 1ee623c26c49d..7647a946340cb 100644 --- a/drivers/hid/hid-core.c +++ b/drivers/hid/hid-core.c @@ -256,6 +256,7 @@ static int hid_add_field(struct hid_parser *parser, unsigned report_type, unsign { struct hid_report *report; struct hid_field *field; + unsigned int max_buffer_size = HID_MAX_BUFFER_SIZE; unsigned int usages; unsigned int offset; unsigned int i; @@ -286,8 +287,11 @@ static int hid_add_field(struct hid_parser *parser, unsigned report_type, unsign offset = report->size; report->size += parser->global.report_size * parser->global.report_count; + if (parser->device->ll_driver->max_buffer_size) + max_buffer_size = parser->device->ll_driver->max_buffer_size; + /* Total size check: Allow for possible report index byte */ - if (report->size > (HID_MAX_BUFFER_SIZE - 1) << 3) { + if (report->size > (max_buffer_size - 1) << 3) { hid_err(parser->device, "report is too long\n"); return -1; } @@ -1963,6 +1967,7 @@ int hid_report_raw_event(struct hid_device *hid, enum hid_report_type type, u8 * struct hid_report_enum *report_enum = hid->report_enum + type; struct hid_report *report; struct hid_driver *hdrv; + int max_buffer_size = HID_MAX_BUFFER_SIZE; u32 rsize, csize = size; u8 *cdata = data; int ret = 0; @@ -1978,10 +1983,13 @@ int hid_report_raw_event(struct hid_device *hid, enum hid_report_type type, u8 * rsize = hid_compute_report_size(report); - if (report_enum->numbered && rsize >= HID_MAX_BUFFER_SIZE) - rsize = HID_MAX_BUFFER_SIZE - 1; - else if (rsize > HID_MAX_BUFFER_SIZE) - rsize = HID_MAX_BUFFER_SIZE; + if (hid->ll_driver->max_buffer_size) + max_buffer_size = hid->ll_driver->max_buffer_size; + + if (report_enum->numbered && rsize >= max_buffer_size) + rsize = max_buffer_size - 1; + else if (rsize > max_buffer_size) + rsize = max_buffer_size; if (csize < rsize) { dbg_hid("report %d is too short, (%d < %d)\n", report->id, @@ -2396,7 +2404,12 @@ int hid_hw_raw_request(struct hid_device *hdev, unsigned char reportnum, __u8 *buf, size_t len, enum hid_report_type rtype, enum hid_class_request reqtype) { - if (len < 1 || len > HID_MAX_BUFFER_SIZE || !buf) + unsigned int max_buffer_size = HID_MAX_BUFFER_SIZE; + + if (hdev->ll_driver->max_buffer_size) + max_buffer_size = hdev->ll_driver->max_buffer_size; + + if (len < 1 || len > max_buffer_size || !buf) return -EINVAL; return hdev->ll_driver->raw_request(hdev, reportnum, buf, len, @@ -2415,7 +2428,12 @@ EXPORT_SYMBOL_GPL(hid_hw_raw_request); */ int hid_hw_output_report(struct hid_device *hdev, __u8 *buf, size_t len) { - if (len < 1 || len > HID_MAX_BUFFER_SIZE || !buf) + unsigned int max_buffer_size = HID_MAX_BUFFER_SIZE; + + if (hdev->ll_driver->max_buffer_size) + max_buffer_size = hdev->ll_driver->max_buffer_size; + + if (len < 1 || len > max_buffer_size || !buf) return -EINVAL; if (hdev->ll_driver->output_report) diff --git a/include/linux/hid.h b/include/linux/hid.h index eaf8ab1773033..1ea8c7a3570b2 100644 --- a/include/linux/hid.h +++ b/include/linux/hid.h @@ -834,6 +834,7 @@ struct hid_driver { * @output_report: send output report to device * @idle: send idle request to device * @may_wakeup: return if device may act as a wakeup source during system-suspend + * @max_buffer_size: over-ride maximum data buffer size (default: HID_MAX_BUFFER_SIZE) */ struct hid_ll_driver { int (*start)(struct hid_device *hdev); @@ -859,6 +860,8 @@ struct hid_ll_driver { int (*idle)(struct hid_device *hdev, int report, int idle, int reqtype); bool (*may_wakeup)(struct hid_device *hdev); + + unsigned int max_buffer_size; }; extern bool hid_is_usb(const struct hid_device *hdev); From 1c5d4221240a233df2440fe75c881465cdf8da07 Mon Sep 17 00:00:00 2001 From: Lee Jones Date: Mon, 23 Jan 2023 12:39:12 +0000 Subject: [PATCH 418/765] HID: uhid: Over-ride the default maximum data buffer value with our own The default maximum data buffer size for this interface is UHID_DATA_MAX (4k). When data buffers are being processed, ensure this value is used when ensuring the sanity, rather than a value between the user provided value and HID_MAX_BUFFER_SIZE (16k). Signed-off-by: Lee Jones Signed-off-by: Jiri Kosina --- drivers/hid/uhid.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/hid/uhid.c b/drivers/hid/uhid.c index f161c95a1ad2e..4588d2cd4ea44 100644 --- a/drivers/hid/uhid.c +++ b/drivers/hid/uhid.c @@ -395,6 +395,7 @@ static const struct hid_ll_driver uhid_hid_driver = { .parse = uhid_hid_parse, .raw_request = uhid_hid_raw_request, .output_report = uhid_hid_output_report, + .max_buffer_size = UHID_DATA_MAX, }; #ifdef CONFIG_COMPAT From 80d2c29e09e663761c2778167a625b25ffe01b6f Mon Sep 17 00:00:00 2001 From: Matthias Kaehlcke Date: Thu, 23 Feb 2023 00:33:30 +0000 Subject: [PATCH 419/765] regulator: core: Use ktime_get_boottime() to determine how long a regulator was off For regulators with 'off-on-delay-us' the regulator framework currently uses ktime_get() to determine how long the regulator has been off before re-enabling it (after a delay if needed). A problem with using ktime_get() is that it doesn't account for the time the system is suspended. As a result a regulator with a longer 'off-on-delay' (e.g. 500ms) that was switched off during suspend might still incurr in a delay on resume before it is re-enabled, even though the regulator might have been off for hours. ktime_get_boottime() accounts for suspend time, use it instead of ktime_get(). Fixes: a8ce7bd89689 ("regulator: core: Fix off_on_delay handling") Cc: stable@vger.kernel.org # 5.13+ Signed-off-by: Matthias Kaehlcke Reviewed-by: Stephen Boyd Link: https://lore.kernel.org/r/20230223003301.v2.1.I9719661b8eb0a73b8c416f9c26cf5bd8c0563f99@changeid Signed-off-by: Mark Brown --- drivers/regulator/core.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c index ae69e493913da..4fcd36055b025 100644 --- a/drivers/regulator/core.c +++ b/drivers/regulator/core.c @@ -1584,7 +1584,7 @@ static int set_machine_constraints(struct regulator_dev *rdev) } if (rdev->desc->off_on_delay) - rdev->last_off = ktime_get(); + rdev->last_off = ktime_get_boottime(); /* If the constraints say the regulator should be on at this point * and we have control then make sure it is enabled. @@ -2673,7 +2673,7 @@ static int _regulator_do_enable(struct regulator_dev *rdev) * this regulator was disabled. */ ktime_t end = ktime_add_us(rdev->last_off, rdev->desc->off_on_delay); - s64 remaining = ktime_us_delta(end, ktime_get()); + s64 remaining = ktime_us_delta(end, ktime_get_boottime()); if (remaining > 0) _regulator_delay_helper(remaining); @@ -2912,7 +2912,7 @@ static int _regulator_do_disable(struct regulator_dev *rdev) } if (rdev->desc->off_on_delay) - rdev->last_off = ktime_get(); + rdev->last_off = ktime_get_boottime(); trace_regulator_disable_complete(rdev_get_name(rdev)); From 152ac60677aa3760d0850de0db33d495f55e8aba Mon Sep 17 00:00:00 2001 From: Hongbin Ji Date: Wed, 22 Feb 2023 17:21:28 +0800 Subject: [PATCH 420/765] spi: cadence-quadspi: Fix cancel the indirect read mask This is to cancel the indirect read transfer process, so should be use CQSPI_REG_INDIRECTRD_CANCEL_MASK Signed-off-by: Hongbin Ji Link: https://lore.kernel.org/r/20230222092128.4237-1-jhb_ee@163.com Signed-off-by: Mark Brown --- drivers/spi/spi-cadence-quadspi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/spi/spi-cadence-quadspi.c b/drivers/spi/spi-cadence-quadspi.c index 2954c06a7f57f..64b6a460d739b 100644 --- a/drivers/spi/spi-cadence-quadspi.c +++ b/drivers/spi/spi-cadence-quadspi.c @@ -786,7 +786,7 @@ static int cqspi_indirect_read_execute(struct cqspi_flash_pdata *f_pdata, writel(0, reg_base + CQSPI_REG_IRQMASK); /* Cancel the indirect read */ - writel(CQSPI_REG_INDIRECTWR_CANCEL_MASK, + writel(CQSPI_REG_INDIRECTRD_CANCEL_MASK, reg_base + CQSPI_REG_INDIRECTRD); return ret; } From 078a5517d22342eb0474046d3e891427a2552e3c Mon Sep 17 00:00:00 2001 From: Dhruva Gole Date: Thu, 23 Feb 2023 15:22:02 +0530 Subject: [PATCH 421/765] spi: spi-sn-f-ospi: fix duplicate flag while assigning to mode_bits Replace the SPI_TX_OCTAL flag that appeared two time with SPI_RX_OCTAL in the chain of '|' operators while assigning to mode_bits Fixes: 1b74dd64c861 ("spi: Add Socionext F_OSPI SPI flash controller driver") Reported-by: David Binderman Link: https://lore.kernel.org/all/DB6P189MB0568F3BE9384315F5C8C1A3E9CA49@DB6P189MB0568.EURP189.PROD.OUTLOOK.COM/ Cc: stable@vger.kernel.org Signed-off-by: Dhruva Gole Link: https://lore.kernel.org/r/20230223095202.924626-1-d-gole@ti.com Signed-off-by: Mark Brown --- drivers/spi/spi-sn-f-ospi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/spi/spi-sn-f-ospi.c b/drivers/spi/spi-sn-f-ospi.c index 348c6e1edd38a..333b22dfd8dba 100644 --- a/drivers/spi/spi-sn-f-ospi.c +++ b/drivers/spi/spi-sn-f-ospi.c @@ -611,7 +611,7 @@ static int f_ospi_probe(struct platform_device *pdev) return -ENOMEM; ctlr->mode_bits = SPI_TX_DUAL | SPI_TX_QUAD | SPI_TX_OCTAL - | SPI_RX_DUAL | SPI_RX_QUAD | SPI_TX_OCTAL + | SPI_RX_DUAL | SPI_RX_QUAD | SPI_RX_OCTAL | SPI_MODE_0 | SPI_MODE_1 | SPI_LSB_FIRST; ctlr->mem_ops = &f_ospi_mem_ops; ctlr->bus_num = -1; From 33c25354939099b76ecb6c82d1c7c50400fbcca6 Mon Sep 17 00:00:00 2001 From: Matt Roper Date: Mon, 13 Feb 2023 16:19:06 -0800 Subject: [PATCH 422/765] drm/i915/xelpmp: Consider GSI offset when doing MCR lookups MCR range tables use the final MMIO offset of a register (including the 0x380000 GSI offset when applicable). Since the i915_mcr_reg_t passed as a parameter during steering lookup does not include the GSI offset, we need to add it back in for GSI registers before searching the tables. Fixes: a7ec65fc7e83 ("drm/i915/xelpmp: Add multicast steering for media GT") Signed-off-by: Matt Roper Reviewed-by: Radhakrishna Sripada Link: https://patchwork.freedesktop.org/patch/msgid/20230214001906.1477370-1-matthew.d.roper@intel.com (cherry picked from commit d6683bbe70d4cdbf3da6acecf7d569cc6f0b4382) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/gt/intel_gt_mcr.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/gt/intel_gt_mcr.c b/drivers/gpu/drm/i915/gt/intel_gt_mcr.c index 169393a7ad88b..3bb1c701d5ff9 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_mcr.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_mcr.c @@ -559,12 +559,15 @@ static bool reg_needs_read_steering(struct intel_gt *gt, i915_mcr_reg_t reg, enum intel_steering_type type) { - const u32 offset = i915_mmio_reg_offset(reg); + u32 offset = i915_mmio_reg_offset(reg); const struct intel_mmio_range *entry; if (likely(!gt->steering_table[type])) return false; + if (IS_GSI_REG(offset)) + offset += gt->uncore->gsi_offset; + for (entry = gt->steering_table[type]; entry->end; entry++) { if (offset >= entry->start && offset <= entry->end) return true; From 5e438bf7f9a1705ebcae5fa89cdbfbc6932a7871 Mon Sep 17 00:00:00 2001 From: Mavroudis Chatzilaridis Date: Wed, 1 Feb 2023 18:51:25 +0000 Subject: [PATCH 423/765] drm/i915/quirks: Add inverted backlight quirk for HP 14-r206nv This laptop uses inverted backlight PWM. Thus, without this quirk, backlight brightness decreases as the brightness value increases and vice versa. Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/8013 Cc: stable@vger.kernel.org Signed-off-by: Mavroudis Chatzilaridis Reviewed-by: Jani Nikula Signed-off-by: Jani Nikula Link: https://patchwork.freedesktop.org/patch/msgid/20230201184947.8835-1-mavchatz@protonmail.com (cherry picked from commit 83e7d6fd330d413cb2064e680ffea91b0512a520) --- drivers/gpu/drm/i915/display/intel_quirks.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/i915/display/intel_quirks.c b/drivers/gpu/drm/i915/display/intel_quirks.c index 6e48d3bcdfec5..a280448df771a 100644 --- a/drivers/gpu/drm/i915/display/intel_quirks.c +++ b/drivers/gpu/drm/i915/display/intel_quirks.c @@ -199,6 +199,8 @@ static struct intel_quirk intel_quirks[] = { /* ECS Liva Q2 */ { 0x3185, 0x1019, 0xa94d, quirk_increase_ddi_disabled_time }, { 0x3184, 0x1019, 0xa94d, quirk_increase_ddi_disabled_time }, + /* HP Notebook - 14-r206nv */ + { 0x0f31, 0x103c, 0x220f, quirk_invert_brightness }, }; void intel_init_quirks(struct drm_i915_private *i915) From 690e0ec8e63da9a29b39fedc6ed5da09c7c82651 Mon Sep 17 00:00:00 2001 From: John Harrison Date: Wed, 15 Feb 2023 17:11:00 -0800 Subject: [PATCH 424/765] drm/i915: Don't use stolen memory for ring buffers with LLC MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Direction from hardware is that stolen memory should never be used for ring buffer allocations on platforms with LLC. There are too many caching pitfalls due to the way stolen memory accesses are routed. So it is safest to just not use it. Signed-off-by: John Harrison Fixes: c58b735fc762 ("drm/i915: Allocate rings from stolen") Cc: Chris Wilson Cc: Joonas Lahtinen Cc: Jani Nikula Cc: Rodrigo Vivi Cc: Tvrtko Ursulin Cc: intel-gfx@lists.freedesktop.org Cc: # v4.9+ Tested-by: Jouni Högander Reviewed-by: Daniele Ceraolo Spurio Link: https://patchwork.freedesktop.org/patch/msgid/20230216011101.1909009-2-John.C.Harrison@Intel.com (cherry picked from commit f54c1f6c697c4297f7ed94283c184acc338a5cf8) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/gt/intel_ring.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/gt/intel_ring.c b/drivers/gpu/drm/i915/gt/intel_ring.c index 15ec64d881c44..fb1d2595392ed 100644 --- a/drivers/gpu/drm/i915/gt/intel_ring.c +++ b/drivers/gpu/drm/i915/gt/intel_ring.c @@ -116,7 +116,7 @@ static struct i915_vma *create_ring_vma(struct i915_ggtt *ggtt, int size) obj = i915_gem_object_create_lmem(i915, size, I915_BO_ALLOC_VOLATILE | I915_BO_ALLOC_PM_VOLATILE); - if (IS_ERR(obj) && i915_ggtt_has_aperture(ggtt)) + if (IS_ERR(obj) && i915_ggtt_has_aperture(ggtt) && !HAS_LLC(i915)) obj = i915_gem_object_create_stolen(i915, size); if (IS_ERR(obj)) obj = i915_gem_object_create_internal(i915, size); From 85636167e3206c3fbd52254fc432991cc4e90194 Mon Sep 17 00:00:00 2001 From: John Harrison Date: Wed, 15 Feb 2023 17:11:01 -0800 Subject: [PATCH 425/765] drm/i915: Don't use BAR mappings for ring buffers with LLC MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Direction from hardware is that ring buffers should never be mapped via the BAR on systems with LLC. There are too many caching pitfalls due to the way BAR accesses are routed. So it is safest to just not use it. Signed-off-by: John Harrison Fixes: 9d80841ea4c9 ("drm/i915: Allow ringbuffers to be bound anywhere") Cc: Chris Wilson Cc: Joonas Lahtinen Cc: Jani Nikula Cc: Rodrigo Vivi Cc: Tvrtko Ursulin Cc: intel-gfx@lists.freedesktop.org Cc: # v4.9+ Tested-by: Jouni Högander Reviewed-by: Daniele Ceraolo Spurio Link: https://patchwork.freedesktop.org/patch/msgid/20230216011101.1909009-3-John.C.Harrison@Intel.com (cherry picked from commit 65c08339db1ada87afd6cfe7db8e60bb4851d919) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/gt/intel_ring.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/gt/intel_ring.c b/drivers/gpu/drm/i915/gt/intel_ring.c index fb1d2595392ed..fb99143be98e7 100644 --- a/drivers/gpu/drm/i915/gt/intel_ring.c +++ b/drivers/gpu/drm/i915/gt/intel_ring.c @@ -53,7 +53,7 @@ int intel_ring_pin(struct intel_ring *ring, struct i915_gem_ww_ctx *ww) if (unlikely(ret)) goto err_unpin; - if (i915_vma_is_map_and_fenceable(vma)) { + if (i915_vma_is_map_and_fenceable(vma) && !HAS_LLC(vma->vm->i915)) { addr = (void __force *)i915_vma_pin_iomap(vma); } else { int type = i915_coherent_map_type(vma->vm->i915, vma->obj, false); @@ -98,7 +98,7 @@ void intel_ring_unpin(struct intel_ring *ring) return; i915_vma_unset_ggtt_write(vma); - if (i915_vma_is_map_and_fenceable(vma)) + if (i915_vma_is_map_and_fenceable(vma) && !HAS_LLC(vma->vm->i915)) i915_vma_unpin_iomap(vma); else i915_gem_object_unpin_map(vma->obj); From 22ef7d7be4dc071712ddf0ae09df6ee39b4901a0 Mon Sep 17 00:00:00 2001 From: Benjamin Tissoires Date: Thu, 23 Feb 2023 15:17:26 +0100 Subject: [PATCH 426/765] selftest: hid: fix hid_bpf not set in config Now that CONFIG_HID_BPF is not automatically implied by HID, we need to set it properly in the selftests config. Signed-off-by: Benjamin Tissoires Signed-off-by: Jiri Kosina --- tools/testing/selftests/hid/config | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/testing/selftests/hid/config b/tools/testing/selftests/hid/config index 9c5a55abca6b6..5b5cef445b549 100644 --- a/tools/testing/selftests/hid/config +++ b/tools/testing/selftests/hid/config @@ -17,5 +17,6 @@ CONFIG_FTRACE_SYSCALLS=y CONFIG_FUNCTION_TRACER=y CONFIG_HIDRAW=y CONFIG_HID=y +CONFIG_HID_BPF=y CONFIG_INPUT_EVDEV=y CONFIG_UHID=y From ad32ab9604f29827494024828f527228e84fbd2c Mon Sep 17 00:00:00 2001 From: Juergen Gross Date: Thu, 23 Feb 2023 09:38:00 +0100 Subject: [PATCH 427/765] irqdomain: Add missing NULL pointer check in irq_domain_create_hierarchy() The recent switch to per-domain locking caused a NULL dereference in irq_domain_create_hierarchy(), as Xen code is calling msi_create_irq_domain() with a NULL parent pointer. Fix that by testing parent to be set before dereferencing it. For a non-existing parent the irqdomain's root will stay to point to itself. Fixes: 9dbb8e3452ab ("irqdomain: Switch to per-domain locking") Signed-off-by: Juergen Gross Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/r/20230223083800.31347-1-jgross@suse.com --- kernel/irq/irqdomain.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index aa5b7eeeceb8d..6522dfb2e49c4 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -1172,7 +1172,8 @@ struct irq_domain *irq_domain_create_hierarchy(struct irq_domain *parent, domain = __irq_domain_create(fwnode, 0, ~0, 0, ops, host_data); if (domain) { - domain->root = parent->root; + if (parent) + domain->root = parent->root; domain->parent = parent; domain->flags |= flags; From 37f5b858a66543b2b67c0288280af623985abc29 Mon Sep 17 00:00:00 2001 From: Danny Kaehn Date: Fri, 10 Feb 2023 11:00:44 -0600 Subject: [PATCH 428/765] HID: cp2112: Fix driver not registering GPIO IRQ chip as threaded The CP2112 generates interrupts from a polling routine on a thread, and can only support threaded interrupts. This patch configures the gpiochip irq chip with this flag, disallowing consumers to request a hard IRQ from this driver, which resulted in a segfault previously. Signed-off-by: Danny Kaehn Link: https://lore.kernel.org/r/20230210170044.11835-1-kaehndan@gmail.com Signed-off-by: Benjamin Tissoires --- drivers/hid/hid-cp2112.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/hid/hid-cp2112.c b/drivers/hid/hid-cp2112.c index 1e16b0fa310d1..27cadadda7c9d 100644 --- a/drivers/hid/hid-cp2112.c +++ b/drivers/hid/hid-cp2112.c @@ -1354,6 +1354,7 @@ static int cp2112_probe(struct hid_device *hdev, const struct hid_device_id *id) girq->parents = NULL; girq->default_type = IRQ_TYPE_NONE; girq->handler = handle_simple_irq; + girq->threaded = true; ret = gpiochip_add_data(&dev->gc, dev); if (ret < 0) { From f7482d8285b638be87a594a30edaaf1341135c1a Mon Sep 17 00:00:00 2001 From: Krishna Yarlagadda Date: Thu, 23 Feb 2023 21:56:34 +0530 Subject: [PATCH 429/765] spi: tegra210-quad: set half duplex flag Tegra QSPI controller only supports half duplex transfers. Set half duplex constrain flag. Signed-off-by: Krishna Yarlagadda Link: https://lore.kernel.org/r/20230223162635.19747-3-kyarlagadda@nvidia.com Signed-off-by: Mark Brown --- drivers/spi/spi-tegra210-quad.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/spi/spi-tegra210-quad.c b/drivers/spi/spi-tegra210-quad.c index 9f356612ba7e5..258e3b8c9c2cb 100644 --- a/drivers/spi/spi-tegra210-quad.c +++ b/drivers/spi/spi-tegra210-quad.c @@ -1532,6 +1532,7 @@ static int tegra_qspi_probe(struct platform_device *pdev) master->mode_bits = SPI_MODE_0 | SPI_MODE_3 | SPI_CS_HIGH | SPI_TX_DUAL | SPI_RX_DUAL | SPI_TX_QUAD | SPI_RX_QUAD; master->bits_per_word_mask = SPI_BPW_MASK(32) | SPI_BPW_MASK(16) | SPI_BPW_MASK(8); + master->flags = SPI_CONTROLLER_HALF_DUPLEX; master->setup = tegra_qspi_setup; master->transfer_one_message = tegra_qspi_transfer_one_message; master->num_chipselect = 1; From 60675225ebeecea248035fd3a0efc82ae9038a98 Mon Sep 17 00:00:00 2001 From: Srinivas Pandruvada Date: Tue, 21 Feb 2023 22:45:26 -0800 Subject: [PATCH 430/765] cpufreq: intel_pstate: Adjust balance_performance EPP for Sapphire Rapids While the majority of server OS distributions are deployed with the "performance" governor as the default, some distributions like Ubuntu use the "powersave" governor by default. While using the "powersave" governor in its default configuration on Sapphire Rapids systems leads to much lower power, the performance is lower by more than 25% for several workloads relative to the "performance" governor. A 37% difference has been reported by www.Phoronix.com [1]. This is a consequence of using a relatively high EPP value in the default configuration of the "powersave" governor and the performance can be made much closer to the "performance" governor's level by adjusting the default EPP value. Based on experiments, with EPP of 0x00, 0x10, 0x20, the performance delta between the "powersave" governor and the "performance" one is around 12%. However, the EPP of 0x20 reduces average power by 18% with respect to the lower EPP values. [Note that raising min_perf_pct in sysfs as high as 50% in addition to adjusting EPP does not improve the performance any further.] For this reason, change the EPP value corresponding to the the default balance_performance setting for Sapphire Rapids to 0x20, which is straightforward, because analogous default EPP adjustment has been applied to Alder Lake and there is a way to set the balance_performance EPP value in intel_pstate based on the processor model already. The goal here is to limit the mean performance delta between the "powersave" governor in the default configuration and the "performance" governor for a wide variety of server workloadsto to around 10-12%. For some bursty workloads, this delta can be still large, as the frequency ramp-up will still lag when the "powersave" governor is in use irrespective of the EPP setting, because the performance governor always requests the maximum possible frequency. Link: https://www.phoronix.com/review/centos-clear-spr/6 # [1] Signed-off-by: Srinivas Pandruvada [ rjw: Subject and changelog edits ] Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/intel_pstate.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index fd73d6d2b8084..32a4004d155dd 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -3372,6 +3372,7 @@ static const struct x86_cpu_id intel_epp_balance_perf[] = { * AlderLake Mobile CPUs. */ X86_MATCH_INTEL_FAM6_MODEL(ALDERLAKE_L, 102), + X86_MATCH_INTEL_FAM6_MODEL(SAPPHIRERAPIDS_X, 32), {} }; From 7af78020e28a532262d00d94de708a705d636e91 Mon Sep 17 00:00:00 2001 From: Kai-Heng Feng Date: Thu, 23 Feb 2023 15:02:52 +0800 Subject: [PATCH 431/765] cpufreq: amd-pstate: Let user know amd-pstate is disabled Commit 202e683df37c ("cpufreq: amd-pstate: add amd-pstate driver parameter for mode selection") changed the driver to be disabled by default, and this can surprise users. Let users know what happened so they can decide what to do next. Link: https://bugs.launchpad.net/bugs/2006942 Signed-off-by: Kai-Heng Feng Acked-by: Huang Rui Reviewed-by: Yuan Perry Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/amd-pstate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c index 45c88894fd8e6..f965f54f7ae71 100644 --- a/drivers/cpufreq/amd-pstate.c +++ b/drivers/cpufreq/amd-pstate.c @@ -1263,7 +1263,7 @@ static int __init amd_pstate_init(void) * with amd_pstate=passive or other modes in kernel command line */ if (cppc_state == AMD_PSTATE_DISABLE) { - pr_debug("driver load is disabled, boot with specific mode to enable this\n"); + pr_info("driver load is disabled, boot with specific mode to enable this\n"); return -ENODEV; } From 70ba26cbe02635461c91fa7133941da685e2f08d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Mon, 20 Feb 2023 23:28:54 +0000 Subject: [PATCH 432/765] cpufreq: schedutil: make kobj_type structure constant MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since commit ee6d3dd4ed48 ("driver core: make kobj_type constant.") the driver core allows the usage of const struct kobj_type. Take advantage of this to constify the structure definition to prevent modification at runtime. Signed-off-by: Thomas Weißschuh Acked-by: Viresh Kumar Signed-off-by: Rafael J. Wysocki --- kernel/sched/cpufreq_schedutil.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c index 5c840151f3bb2..e3211455b2032 100644 --- a/kernel/sched/cpufreq_schedutil.c +++ b/kernel/sched/cpufreq_schedutil.c @@ -546,7 +546,7 @@ static void sugov_tunables_free(struct kobject *kobj) kfree(to_sugov_tunables(attr_set)); } -static struct kobj_type sugov_tunables_ktype = { +static const struct kobj_type sugov_tunables_ktype = { .default_groups = sugov_groups, .sysfs_ops = &governor_sysfs_ops, .release = &sugov_tunables_free, From fa0746b11ba04d3df04adddb283cab3cc1d90ec3 Mon Sep 17 00:00:00 2001 From: Nick Alcock Date: Fri, 17 Feb 2023 14:10:40 +0000 Subject: [PATCH 433/765] cpufreq: amd-pstate: remove MODULE_LICENSE in non-modules Since commit 8b41fc4454e ("kbuild: create modules.builtin without Makefile.modbuiltin or tristate.conf"), MODULE_LICENSE declarations are used to identify modules. As a consequence, uses of the macro in non-modules will cause modprobe to misidentify their containing object file as a module when it is not (false positives), and modprobe might succeed rather than failing with a suitable error message. So remove it in amd-pstate.c which cannot be built as a module. Signed-off-by: Nick Alcock Suggested-by: Luis Chamberlain [ rjw: Subject and changelog adjustments ] Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/amd-pstate.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c index f965f54f7ae71..73c7643b26972 100644 --- a/drivers/cpufreq/amd-pstate.c +++ b/drivers/cpufreq/amd-pstate.c @@ -1353,4 +1353,3 @@ early_param("amd_pstate", amd_pstate_param); MODULE_AUTHOR("Huang Rui "); MODULE_DESCRIPTION("AMD Processor P-state Frequency Driver"); -MODULE_LICENSE("GPL"); From eb52bc2ae5b8413619a326f47ed635365883343a Mon Sep 17 00:00:00 2001 From: Sumeet Pawnikar Date: Wed, 15 Feb 2023 18:02:49 +0530 Subject: [PATCH 434/765] powercap: RAPL: Add Power Limit4 support for Meteor Lake SoC Add Meteor Lake SoC to the list of processor models for which Power Limit4 is supported by the Intel RAPL driver. Signed-off-by: Sumeet Pawnikar Signed-off-by: Rafael J. Wysocki --- drivers/powercap/intel_rapl_msr.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/powercap/intel_rapl_msr.c b/drivers/powercap/intel_rapl_msr.c index bc6adda588835..a27673706c3d6 100644 --- a/drivers/powercap/intel_rapl_msr.c +++ b/drivers/powercap/intel_rapl_msr.c @@ -143,6 +143,8 @@ static const struct x86_cpu_id pl4_support_ids[] = { { X86_VENDOR_INTEL, 6, INTEL_FAM6_ALDERLAKE_N, X86_FEATURE_ANY }, { X86_VENDOR_INTEL, 6, INTEL_FAM6_RAPTORLAKE, X86_FEATURE_ANY }, { X86_VENDOR_INTEL, 6, INTEL_FAM6_RAPTORLAKE_P, X86_FEATURE_ANY }, + { X86_VENDOR_INTEL, 6, INTEL_FAM6_METEORLAKE, X86_FEATURE_ANY }, + { X86_VENDOR_INTEL, 6, INTEL_FAM6_METEORLAKE_L, X86_FEATURE_ANY }, {} }; From 01178ee713100c7d7e738992134545acff472d1d Mon Sep 17 00:00:00 2001 From: Lukas Bulwahn Date: Mon, 20 Feb 2023 17:57:49 +0100 Subject: [PATCH 435/765] docs: locking: refer to the actual existing config names The config is actually called CONFIG_RT_MUTEXES, not CONFIG_RT_MUTEX. The config CONFIG_LOCK_TORTURE_TEST should be connected by underscore, for the sake of consistent referencing to configs in the kernel documentation. Address those issues. Signed-off-by: Lukas Bulwahn Acked-by: Waiman Long Link: https://lore.kernel.org/r/20230220165749.12850-1-lukas.bulwahn@gmail.com Signed-off-by: Jonathan Corbet --- Documentation/locking/locktorture.rst | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Documentation/locking/locktorture.rst b/Documentation/locking/locktorture.rst index dfaf9fc883f44..7f56fc0d7c316 100644 --- a/Documentation/locking/locktorture.rst +++ b/Documentation/locking/locktorture.rst @@ -5,7 +5,7 @@ Kernel Lock Torture Test Operation CONFIG_LOCK_TORTURE_TEST ======================== -The CONFIG LOCK_TORTURE_TEST config option provides a kernel module +The CONFIG_LOCK_TORTURE_TEST config option provides a kernel module that runs torture tests on core kernel locking primitives. The kernel module, 'locktorture', may be built after the fact on the running kernel to be tested, if desired. The tests periodically output status @@ -67,7 +67,7 @@ torture_type - "rtmutex_lock": rtmutex_lock() and rtmutex_unlock() pairs. - Kernel must have CONFIG_RT_MUTEX=y. + Kernel must have CONFIG_RT_MUTEXES=y. - "rwsem_lock": read/write down() and up() semaphore pairs. From 1481df6cbd0d606fa1a26b752e2e1aba0b650093 Mon Sep 17 00:00:00 2001 From: Carlos Bilbao Date: Tue, 7 Feb 2023 16:08:44 -0600 Subject: [PATCH 436/765] docs/sp_SP: Add process programming-language translation Translate Documentation/process/programming-language.rst into Spanish. Signed-off-by: Carlos Bilbao Link: https://lore.kernel.org/r/20230207220844.2661295-1-carlos.bilbao@amd.com Signed-off-by: Jonathan Corbet --- .../translations/sp_SP/process/index.rst | 1 + .../sp_SP/process/programming-language.rst | 53 +++++++++++++++++++ 2 files changed, 54 insertions(+) create mode 100644 Documentation/translations/sp_SP/process/programming-language.rst diff --git a/Documentation/translations/sp_SP/process/index.rst b/Documentation/translations/sp_SP/process/index.rst index c978a8132ce10..0f1e131b3bb14 100644 --- a/Documentation/translations/sp_SP/process/index.rst +++ b/Documentation/translations/sp_SP/process/index.rst @@ -17,3 +17,4 @@ kernel-enforcement-statement email-clients magic-number + programming-language diff --git a/Documentation/translations/sp_SP/process/programming-language.rst b/Documentation/translations/sp_SP/process/programming-language.rst new file mode 100644 index 0000000000000..301f525372d85 --- /dev/null +++ b/Documentation/translations/sp_SP/process/programming-language.rst @@ -0,0 +1,53 @@ +.. include:: ../disclaimer-sp.rst + +:Original: :ref:`Documentation/process/programming-language.rst ` +:Translator: Carlos Bilbao + +.. _sp_programming_language: + +Lenguaje de programación +======================== + +El kernel está escrito en el lenguaje de programación C [sp-c-language]_. +Más concretamente, el kernel normalmente se compila con ``gcc`` [sp-gcc]_ +bajo ``-std=gnu11`` [sp-gcc-c-dialect-options]_: el dialecto GNU de ISO C11. +``clang`` [sp-clang]_ también es compatible, consulte los documentos en +:ref:`Building Linux with Clang/LLVM `. + +Este dialecto contiene muchas extensiones del lenguaje [sp-gnu-extensions]_, +y muchos de ellos se usan dentro del kernel de forma habitual. + +Hay algo de soporte para compilar el núcleo con ``icc`` [sp-icc]_ para varias +de las arquitecturas, aunque en el momento de escribir este texto, eso no +está terminado y requiere parches de terceros. + +Atributos +--------- + +Una de las comunes extensiones utilizadas en todo el kernel son los atributos +[sp-gcc-attribute-syntax]_. Los atributos permiten introducir semántica +definida por la implementación a las entidades del lenguaje (como variables, +funciones o tipos) sin tener que hacer cambios sintácticos significativos +al idioma (por ejemplo, agregar una nueva palabra clave) [sp-n2049]_. + +En algunos casos, los atributos son opcionales (es decir, hay compiladores +que no los admiten pero de todos modos deben producir el código adecuado, +incluso si es más lento o no realiza tantas comprobaciones/diagnósticos en +tiempo de compilación). + +El kernel define pseudo-palabras clave (por ejemplo, ``__pure``) en lugar +de usar directamente la sintaxis del atributo GNU (por ejemplo, +``__attribute__((__pure__))``) con el fin de detectar cuáles se pueden +utilizar y/o acortar el código. + +Por favor consulte ``include/linux/compiler_attributes.h`` para obtener +más información. + +.. [sp-c-language] http://www.open-std.org/jtc1/sc22/wg14/www/standards +.. [sp-gcc] https://gcc.gnu.org +.. [sp-clang] https://clang.llvm.org +.. [sp-icc] https://software.intel.com/en-us/c-compilers +.. [sp-gcc-c-dialect-options] https://gcc.gnu.org/onlinedocs/gcc/C-Dialect-Options.html +.. [sp-gnu-extensions] https://gcc.gnu.org/onlinedocs/gcc/C-Extensions.html +.. [sp-gcc-attribute-syntax] https://gcc.gnu.org/onlinedocs/gcc/Attribute-Syntax.html +.. [sp-n2049] http://www.open-std.org/jtc1/sc22/wg14/www/docs/n2049.pdf From b8885e2615f4d623334855e7fad20e49b0e0f96d Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 14 Feb 2023 16:57:26 -0800 Subject: [PATCH 437/765] Documentation: front page: use recommended heading adornments Convert the Documentation front page to use the heading adornments that are documented in doc-guide/sphinx.rst for document title and chapters. I.e., convert most section headings to chapters. This leaves "Indices and tables" as a chapter entry at the same level as the other chapters. The only visual difference from before to after is that the "Indices and tables" heading is smaller and has more vertical whitespace preceding it (although that may depend on the web browser being used). Fixes: 0c7b4366f1ab ("docs: Rewrite the front page") Signed-off-by: Randy Dunlap Cc: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/20230215005726.27320-1-rdunlap@infradead.org Signed-off-by: Jonathan Corbet --- Documentation/index.rst | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) diff --git a/Documentation/index.rst b/Documentation/index.rst index bf6aa681c9608..76d1a3ec9be3d 100644 --- a/Documentation/index.rst +++ b/Documentation/index.rst @@ -2,6 +2,7 @@ .. _linux_doc: +============================== The Linux Kernel documentation ============================== @@ -13,7 +14,7 @@ documentation are welcome; join the linux-doc list at vger.kernel.org if you want to help out. Working with the development community --------------------------------------- +====================================== The essential guides for interacting with the kernel's development community and getting your work upstream. @@ -29,7 +30,7 @@ community and getting your work upstream. Internal API manuals --------------------- +==================== Manuals for use by developers working to interface with the rest of the kernel. @@ -43,7 +44,7 @@ kernel. Locking in the kernel Development tools and processes -------------------------------- +=============================== Various other manuals with useful information for all kernel developers. @@ -62,7 +63,7 @@ Various other manuals with useful information for all kernel developers. User-oriented documentation ---------------------------- +=========================== The following manuals are written for *users* of the kernel — those who are trying to get it to work optimally on a given system and application @@ -81,7 +82,7 @@ See also: the `Linux man pages `_, which are kept separately from the kernel's own documentation. Firmware-related documentation ------------------------------- +============================== The following holds information on the kernel's expectations regarding the platform firmwares. @@ -93,7 +94,7 @@ platform firmwares. Architecture-specific documentation ------------------------------------ +=================================== .. toctree:: :maxdepth: 2 @@ -102,7 +103,7 @@ Architecture-specific documentation Other documentation -------------------- +=================== There are several unsorted documents that don't seem to fit on other parts of the documentation body, or may require some adjustments and/or conversion @@ -115,7 +116,7 @@ to ReStructured Text format, or are simply too old. Translations ------------- +============ .. toctree:: :maxdepth: 2 From 901578a45950bcc4d5055a24e9016d61b84dc1a2 Mon Sep 17 00:00:00 2001 From: Thorsten Leemhuis Date: Tue, 14 Feb 2023 11:13:51 +0100 Subject: [PATCH 438/765] docs: recommend using Link: whenever using Reported-by: Encourage developers to place Link: tag pointing to the report when they are using Reported-by: tags. Those links are often extremely useful for any code archaeologist that wants to know more about the backstory of a change than the commit message provides. That includes maintainers higher up in the patch-flow hierarchy, which is why Linus asks developers to add such links [1, 2, 3]. To quote [1]: > Again, the commit has a link to the patch *submission*, which is > almost entirely useless. There's no link to the actual problem the > patch fixes. > > [...] > > Put another way: I can see that > > Reported-by: Zhangfei Gao > > in the commit, but I don't have a clue what the actual report was, and > there really isn't enough information in the commit itself, except for > a fairly handwavy "Device drivers might, for instance, still need to > flush operations.." > > I don't want to know what device drivers _might_ do. I would want to > have an actual pointer to what they do and where. Another reason why these links are wanted: the ongoing regression tracking efforts can only scale with them, as they allow the regression tracking bot 'regzbot' to automatically connect tracked reports with patches that are posted or committed to fix tracked regressions. Link: https://lore.kernel.org/all/CAHk-=wjMmSZzMJ3Xnskdg4+GGz=5p5p+GSYyFBTh0f-DgvdBWg@mail.gmail.com/ [1] Link: https://lore.kernel.org/all/CAHk-=wgs38ZrfPvy=nOwVkVzjpM3VFU1zobP37Fwd_h9iAD5JQ@mail.gmail.com/ [2] Link: https://lore.kernel.org/all/CAHk-=wjxzafG-=J8oT30s7upn4RhBs6TX-uVFZ5rME+L5_DoJA@mail.gmail.com/ [3] Signed-off-by: Thorsten Leemhuis Link: https://lore.kernel.org/r/9a07ec640d809723492f8ade4f54705914e80419.1676369564.git.linux@leemhuis.info Signed-off-by: Jonathan Corbet --- Documentation/process/5.Posting.rst | 3 ++- Documentation/process/submitting-patches.rst | 9 +++++---- 2 files changed, 7 insertions(+), 5 deletions(-) diff --git a/Documentation/process/5.Posting.rst b/Documentation/process/5.Posting.rst index d87f1fee4cbc5..7a670a075ab6f 100644 --- a/Documentation/process/5.Posting.rst +++ b/Documentation/process/5.Posting.rst @@ -251,7 +251,8 @@ The tags in common use are: - Reported-by: names a user who reported a problem which is fixed by this patch; this tag is used to give credit to the (often underappreciated) people who test our code and let us know when things do not work - correctly. + correctly. Note, this tag should be followed by a Link: tag pointing to the + report, unless the report is not available on the web. - Cc: the named person received a copy of the patch and had the opportunity to comment on it. diff --git a/Documentation/process/submitting-patches.rst b/Documentation/process/submitting-patches.rst index 7dc94555417dc..1e009dbe6991b 100644 --- a/Documentation/process/submitting-patches.rst +++ b/Documentation/process/submitting-patches.rst @@ -496,10 +496,11 @@ Using Reported-by:, Tested-by:, Reviewed-by:, Suggested-by: and Fixes: ---------------------------------------------------------------------- The Reported-by tag gives credit to people who find bugs and report them and it -hopefully inspires them to help us again in the future. Please note that if -the bug was reported in private, then ask for permission first before using the -Reported-by tag. The tag is intended for bugs; please do not use it to credit -feature requests. +hopefully inspires them to help us again in the future. The tag is intended for +bugs; please do not use it to credit feature requests. The tag should be +followed by a Link: tag pointing to the report, unless the report is not +available on the web. Please note that if the bug was reported in private, then +ask for permission first before using the Reported-by tag. A Tested-by: tag indicates that the patch has been successfully tested (in some environment) by the person named. This tag informs maintainers that From 2e2b9baf008ec795fe750a48b42e787cf31486df Mon Sep 17 00:00:00 2001 From: Ruili Ji Date: Mon, 6 Feb 2023 18:35:50 +0800 Subject: [PATCH 439/765] drm/amdkfd: To fix sdma page fault issue for GC 11 For the MQD memory, KMD would always allocate 4K memory, and mes scheduler would write to the end of MQD for unmap flag. Signed-off-by: Ruili Ji Signed-off-by: Alex Deucher --- .../gpu/drm/amd/amdkfd/kfd_device_queue_manager.c | 5 +++-- drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c | 15 ++++++++++++++- 2 files changed, 17 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index c06ada0844ba1..7a95698d83f73 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -2373,7 +2373,7 @@ struct device_queue_manager *device_queue_manager_init(struct kfd_dev *dev) if (init_mqd_managers(dqm)) goto out_free; - if (allocate_hiq_sdma_mqd(dqm)) { + if (!dev->shared_resources.enable_mes && allocate_hiq_sdma_mqd(dqm)) { pr_err("Failed to allocate hiq sdma mqd trunk buffer\n"); goto out_free; } @@ -2397,7 +2397,8 @@ static void deallocate_hiq_sdma_mqd(struct kfd_dev *dev, void device_queue_manager_uninit(struct device_queue_manager *dqm) { dqm->ops.uninitialize(dqm); - deallocate_hiq_sdma_mqd(dqm->dev, &dqm->hiq_sdma_mqd); + if (!dqm->dev->shared_resources.enable_mes) + deallocate_hiq_sdma_mqd(dqm->dev, &dqm->hiq_sdma_mqd); kfree(dqm); } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c index 4f6390f3236ef..4a9af800b1f1c 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c @@ -308,11 +308,16 @@ static void init_mqd_sdma(struct mqd_manager *mm, void **mqd, struct queue_properties *q) { struct v11_sdma_mqd *m; + int size; m = (struct v11_sdma_mqd *) mqd_mem_obj->cpu_ptr; - memset(m, 0, sizeof(struct v11_sdma_mqd)); + if (mm->dev->shared_resources.enable_mes) + size = PAGE_SIZE; + else + size = sizeof(struct v11_sdma_mqd); + memset(m, 0, size); *mqd = m; if (gart_addr) *gart_addr = mqd_mem_obj->gpu_addr; @@ -443,6 +448,14 @@ struct mqd_manager *mqd_manager_init_v11(enum KFD_MQD_TYPE type, #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd_sdma; #endif + /* + * To allocate SDMA MQDs by generic functions + * when MES is enabled. + */ + if (dev->shared_resources.enable_mes) { + mqd->allocate_mqd = allocate_mqd; + mqd->free_mqd = kfd_free_mqd_cp; + } pr_debug("%s@%i\n", __func__, __LINE__); break; default: From 08c6ab7fb4d98694df5a9954a42a365cc538f9b0 Mon Sep 17 00:00:00 2001 From: Jesse Zhang Date: Thu, 16 Feb 2023 14:11:33 +0800 Subject: [PATCH 440/765] drm/amdgpu: add tmz support for GC 10.3.6 this patch to add tmz support for GC 10.3.6 Signed-off-by: Jesse Zhang Reviewed-by: Yifan Zhang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 94f10ac0eef74..12a6826caef47 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -552,6 +552,7 @@ void amdgpu_gmc_tmz_set(struct amdgpu_device *adev) case IP_VERSION(10, 3, 2): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): /* VANGOGH */ case IP_VERSION(10, 3, 1): /* YELLOW_CARP*/ From d9e1e14f42337ea11b2dfc0bab99485a8f7fa210 Mon Sep 17 00:00:00 2001 From: Kenneth Feng Date: Wed, 15 Feb 2023 14:42:08 +0800 Subject: [PATCH 441/765] drm/amd/pm: re-enable ac/dc on smu_v13_0_0/10 re-enable ac/dc on smu_v13_0_0/10 Signed-off-by: Kenneth Feng Reviewed-by: Yang Wang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c index 7c906ab3ddd2f..923a9fb3c8873 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c @@ -147,6 +147,7 @@ static struct cmn2asic_msg_mapping smu_v13_0_0_message_map[SMU_MSG_MAX_COUNT] = PPSMC_MSG_SetBadMemoryPagesRetiredFlagsPerChannel, 0), MSG_MAP(AllowGpo, PPSMC_MSG_SetGpoAllow, 0), MSG_MAP(AllowIHHostInterrupt, PPSMC_MSG_AllowIHHostInterrupt, 0), + MSG_MAP(ReenableAcDcInterrupt, PPSMC_MSG_ReenableAcDcInterrupt, 0), }; static struct cmn2asic_mapping smu_v13_0_0_clk_map[SMU_CLK_COUNT] = { From 6d9b6dceaa513c19a968c523f4d68477a33a98c9 Mon Sep 17 00:00:00 2001 From: Hamza Mahfooz Date: Tue, 14 Feb 2023 13:51:08 -0500 Subject: [PATCH 442/765] drm/amd/display: only warn once in dce110_edp_wait_for_hpd_ready() Since, hot plugging eDP displays isn't supported, it is sufficient for us to warn about the lack of a connected display once. So, use ASSERT() in dce110_edp_wait_for_hpd_ready() instead of DC_LOG_WARNING(). Reviewed-by: Harry Wentland Signed-off-by: Hamza Mahfooz Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c index fb3fd5b7c78b9..0d4d3d586166d 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c @@ -779,10 +779,8 @@ void dce110_edp_wait_for_hpd_ready( dal_gpio_destroy_irq(&hpd); - if (false == edp_hpd_high) { - DC_LOG_WARNING( - "%s: wait timed out!\n", __func__); - } + /* ensure that the panel is detected */ + ASSERT(edp_hpd_high); } void dce110_edp_power_control( From 455ad25997ba6e6b4c5fb9b4f3cd54ec415df969 Mon Sep 17 00:00:00 2001 From: Harry Wentland Date: Mon, 13 Feb 2023 13:17:16 -0500 Subject: [PATCH 443/765] drm/amdgpu: Select DRM_DISPLAY_HDCP_HELPER in amdgpu Keeps this selection with the rest of the DRM HELPER selection. Signed-off-by: Harry Wentland Reviewed-by: Rodrigo Siqueira Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/Kconfig | 1 + drivers/gpu/drm/amd/display/Kconfig | 1 - 2 files changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/Kconfig b/drivers/gpu/drm/amd/amdgpu/Kconfig index 5341b6b242c3b..a82d36ea88e25 100644 --- a/drivers/gpu/drm/amd/amdgpu/Kconfig +++ b/drivers/gpu/drm/amd/amdgpu/Kconfig @@ -6,6 +6,7 @@ config DRM_AMDGPU select FW_LOADER select DRM_DISPLAY_DP_HELPER select DRM_DISPLAY_HDMI_HELPER + select DRM_DISPLAY_HDCP_HELPER select DRM_DISPLAY_HELPER select DRM_KMS_HELPER select DRM_SCHED diff --git a/drivers/gpu/drm/amd/display/Kconfig b/drivers/gpu/drm/amd/display/Kconfig index 2efe93f74f840..0c9bd0a53e603 100644 --- a/drivers/gpu/drm/amd/display/Kconfig +++ b/drivers/gpu/drm/amd/display/Kconfig @@ -28,7 +28,6 @@ config DRM_AMD_DC_DCN config DRM_AMD_DC_HDCP bool "Enable HDCP support in DC" depends on DRM_AMD_DC - select DRM_DISPLAY_HDCP_HELPER help Choose this option if you want to support HDCP authentication. From c105518679b6e87232874ffc989ec403bee59664 Mon Sep 17 00:00:00 2001 From: Shane Xiao Date: Wed, 15 Feb 2023 13:23:44 +0800 Subject: [PATCH 444/765] drm/amdgpu: remove TOPDOWN flags when allocating VRAM in large bar system MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since VRAM manager is changed from drm mm to drm buddy, the TOP_DOWN flag should not be set by default in the large bar system. Removing this flag helps improve drm buddy allocator efficiency and reduce the risk of splitting higher order block into lower order. Signed-off-by: Shane Xiao Reviewed-by: Christian K�nig Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index 981010de0a282..e3e1ed4314dd6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -139,7 +139,7 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain) if (flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) places[c].lpfn = visible_pfn; - else + else if (adev->gmc.real_vram_size != adev->gmc.visible_vram_size) places[c].flags |= TTM_PL_FLAG_TOPDOWN; if (flags & AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS) From 2866cc09617991cb4f9f36fbebdbba966fe5a21a Mon Sep 17 00:00:00 2001 From: Shane Xiao Date: Sat, 18 Feb 2023 11:58:45 +0800 Subject: [PATCH 445/765] drm/amdgpu: optimize VRAM allocation when using drm buddy Since the VRAM manager changed from drm mm to drm buddy. It's not necessary to allocate 2MB aligned VRAM for more than 2MB unaligned size, and then do trim. This method improves the allocation efficiency and reduces memory fragmentation. v2: Correct the remainder operation Signed-off-by: Shane Xiao Reviewed-by: Arunpravin Paneer Selvam Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c index 9fa1d814508a6..43d6a9d6a5384 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c @@ -453,7 +453,8 @@ static int amdgpu_vram_mgr_new(struct ttm_resource_manager *man, /* Limit maximum size to 2GiB due to SG table limitations */ size = min(remaining_size, 2ULL << 30); - if (size >= (u64)pages_per_block << PAGE_SHIFT) + if ((size >= (u64)pages_per_block << PAGE_SHIFT) && + !(size & (((u64)pages_per_block << PAGE_SHIFT) - 1))) min_block_size = (u64)pages_per_block << PAGE_SHIFT; cur_size = size; From f9c35f4fffc6cb5bbb23f546f48c045aef012518 Mon Sep 17 00:00:00 2001 From: Hawking Zhang Date: Mon, 20 Feb 2023 09:06:53 +0800 Subject: [PATCH 446/765] drm/amdgpu: fix incorrect active rb bitmap for gfx11 GFX v11 changes RB_BACKEND_DISABLE related registers from per SA to global ones. The approach to query active rb bitmap needs to be changed accordingly. Query per SE setting returns wrong active RB bitmap especially in the case when some of SA are disabled. With the new approach, driver will generate the active rb bitmap based on active SA bitmap and global active RB bitmap. Signed-off-by: Hawking Zhang Reviewed-by: Likun Gao Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 78 +++++++++++++++++--------- 1 file changed, 52 insertions(+), 26 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index 8ad8a0bffcacb..417ab8d1eacea 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -1503,44 +1503,70 @@ static void gfx_v11_0_select_se_sh(struct amdgpu_device *adev, u32 se_num, WREG32_SOC15(GC, 0, regGRBM_GFX_INDEX, data); } -static u32 gfx_v11_0_get_rb_active_bitmap(struct amdgpu_device *adev) +static u32 gfx_v11_0_get_sa_active_bitmap(struct amdgpu_device *adev) { - u32 data, mask; + u32 gc_disabled_sa_mask, gc_user_disabled_sa_mask, sa_mask; + + gc_disabled_sa_mask = RREG32_SOC15(GC, 0, regCC_GC_SA_UNIT_DISABLE); + gc_disabled_sa_mask = REG_GET_FIELD(gc_disabled_sa_mask, + CC_GC_SA_UNIT_DISABLE, + SA_DISABLE); + gc_user_disabled_sa_mask = RREG32_SOC15(GC, 0, regGC_USER_SA_UNIT_DISABLE); + gc_user_disabled_sa_mask = REG_GET_FIELD(gc_user_disabled_sa_mask, + GC_USER_SA_UNIT_DISABLE, + SA_DISABLE); + sa_mask = amdgpu_gfx_create_bitmask(adev->gfx.config.max_sh_per_se * + adev->gfx.config.max_shader_engines); - data = RREG32_SOC15(GC, 0, regCC_RB_BACKEND_DISABLE); - data |= RREG32_SOC15(GC, 0, regGC_USER_RB_BACKEND_DISABLE); + return sa_mask & (~(gc_disabled_sa_mask | gc_user_disabled_sa_mask)); +} - data &= CC_RB_BACKEND_DISABLE__BACKEND_DISABLE_MASK; - data >>= GC_USER_RB_BACKEND_DISABLE__BACKEND_DISABLE__SHIFT; +static u32 gfx_v11_0_get_rb_active_bitmap(struct amdgpu_device *adev) +{ + u32 gc_disabled_rb_mask, gc_user_disabled_rb_mask; + u32 rb_mask; - mask = amdgpu_gfx_create_bitmask(adev->gfx.config.max_backends_per_se / - adev->gfx.config.max_sh_per_se); + gc_disabled_rb_mask = RREG32_SOC15(GC, 0, regCC_RB_BACKEND_DISABLE); + gc_disabled_rb_mask = REG_GET_FIELD(gc_disabled_rb_mask, + CC_RB_BACKEND_DISABLE, + BACKEND_DISABLE); + gc_user_disabled_rb_mask = RREG32_SOC15(GC, 0, regGC_USER_RB_BACKEND_DISABLE); + gc_user_disabled_rb_mask = REG_GET_FIELD(gc_user_disabled_rb_mask, + GC_USER_RB_BACKEND_DISABLE, + BACKEND_DISABLE); + rb_mask = amdgpu_gfx_create_bitmask(adev->gfx.config.max_backends_per_se * + adev->gfx.config.max_shader_engines); - return (~data) & mask; + return rb_mask & (~(gc_disabled_rb_mask | gc_user_disabled_rb_mask)); } static void gfx_v11_0_setup_rb(struct amdgpu_device *adev) { - int i, j; - u32 data; - u32 active_rbs = 0; - u32 rb_bitmap_width_per_sh = adev->gfx.config.max_backends_per_se / - adev->gfx.config.max_sh_per_se; + u32 rb_bitmap_width_per_sa; + u32 max_sa; + u32 active_sa_bitmap; + u32 global_active_rb_bitmap; + u32 active_rb_bitmap = 0; + u32 i; - mutex_lock(&adev->grbm_idx_mutex); - for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { - for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { - gfx_v11_0_select_se_sh(adev, i, j, 0xffffffff); - data = gfx_v11_0_get_rb_active_bitmap(adev); - active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) * - rb_bitmap_width_per_sh); - } + /* query sa bitmap from SA_UNIT_DISABLE registers */ + active_sa_bitmap = gfx_v11_0_get_sa_active_bitmap(adev); + /* query rb bitmap from RB_BACKEND_DISABLE registers */ + global_active_rb_bitmap = gfx_v11_0_get_rb_active_bitmap(adev); + + /* generate active rb bitmap according to active sa bitmap */ + max_sa = adev->gfx.config.max_shader_engines * + adev->gfx.config.max_sh_per_se; + rb_bitmap_width_per_sa = adev->gfx.config.max_backends_per_se / + adev->gfx.config.max_sh_per_se; + for (i = 0; i < max_sa; i++) { + if (active_sa_bitmap & (1 << i)) + active_rb_bitmap |= (0x3 << (i * rb_bitmap_width_per_sa)); } - gfx_v11_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); - mutex_unlock(&adev->grbm_idx_mutex); - adev->gfx.config.backend_enable_mask = active_rbs; - adev->gfx.config.num_rbs = hweight32(active_rbs); + active_rb_bitmap |= global_active_rb_bitmap; + adev->gfx.config.backend_enable_mask = active_rb_bitmap; + adev->gfx.config.num_rbs = hweight32(active_rb_bitmap); } #define DEFAULT_SH_MEM_BASES (0x6000) From ca47518663973083c513cd6b2801dcda0bfaaa99 Mon Sep 17 00:00:00 2001 From: Mario Limonciello Date: Mon, 13 Feb 2023 15:10:30 -0600 Subject: [PATCH 447/765] drm/amd: Don't allow s0ix on APUs older than Raven APUs before Raven didn't support s0ix. As we just relieved some of the safety checks for s0ix to improve power consumption on APUs that support it but that are missing BIOS support a new blind spot was introduced that a user could "try" to run s0ix. Plug this hole so that if users try to run s0ix on anything older than Raven it will just skip suspend of the GPU. Fixes: cf488dcd0ab7 ("drm/amd: Allow s0ix without BIOS support") Suggested-by: Alexander Deucher Signed-off-by: Mario Limonciello Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 3 +++ drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 7 ++++++- 2 files changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c index 458362e4ea011..d4196fcb85a08 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c @@ -1073,6 +1073,9 @@ bool amdgpu_acpi_is_s0ix_active(struct amdgpu_device *adev) (pm_suspend_target_state != PM_SUSPEND_TO_IDLE)) return false; + if (adev->asic_type < CHIP_RAVEN) + return false; + /* * If ACPI_FADT_LOW_POWER_S0 is not set in the FADT, it is generally * risky to do any special firmware-related preparations for entering diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 86fbb41382854..bfb7ed254ee4c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -2414,8 +2414,10 @@ static int amdgpu_pmops_suspend(struct device *dev) if (amdgpu_acpi_is_s0ix_active(adev)) adev->in_s0ix = true; - else + else if (amdgpu_acpi_is_s3_active(adev)) adev->in_s3 = true; + if (!adev->in_s0ix && !adev->in_s3) + return 0; return amdgpu_device_suspend(drm_dev, true); } @@ -2436,6 +2438,9 @@ static int amdgpu_pmops_resume(struct device *dev) struct amdgpu_device *adev = drm_to_adev(drm_dev); int r; + if (!adev->in_s0ix && !adev->in_s3) + return 0; + /* Avoids registers access if device is physically gone */ if (!pci_device_is_present(adev->pdev)) adev->no_hw_access = true; From 6dcb38a19efaa71c95c017652177cecb5be4191d Mon Sep 17 00:00:00 2001 From: Jane Jian Date: Wed, 15 Feb 2023 16:48:47 +0800 Subject: [PATCH 448/765] drm/amdgpu/vcn: set and use harvest config in early init to set harvest config if the vcn0/1 is disabled rather than hard-code the ring attributes as before did Signed-off-by: Jane Jian Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c | 26 +++++++++++++++----------- 1 file changed, 15 insertions(+), 11 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c index 22a41766a8c71..213b43670f23e 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c @@ -78,9 +78,18 @@ static void vcn_v4_0_set_ras_funcs(struct amdgpu_device *adev); static int vcn_v4_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_ring *ring; - if (amdgpu_sriov_vf(adev)) + if (amdgpu_sriov_vf(adev)) { adev->vcn.harvest_config = VCN_HARVEST_MMSCH; + for (int i = 0; i < adev->vcn.num_vcn_inst; ++i) { + ring = &adev->vcn.inst[i].ring_enc[0]; + if (amdgpu_vcn_is_disabled_vcn(adev, VCN_ENCODE_RING, i)) { + adev->vcn.harvest_config |= 1 << i; + dev_info(adev->dev, "VCN%d is disabled by hypervisor\n", i); + } + } + } /* re-use enc ring as unified ring */ adev->vcn.num_enc_rings = 1; @@ -238,16 +247,11 @@ static int vcn_v4_0_hw_init(void *handle) continue; ring = &adev->vcn.inst[i].ring_enc[0]; - if (amdgpu_vcn_is_disabled_vcn(adev, VCN_ENCODE_RING, i)) { - ring->sched.ready = false; - ring->no_scheduler = true; - dev_info(adev->dev, "ring %s is disabled by hypervisor\n", ring->name); - } else { - ring->wptr = 0; - ring->wptr_old = 0; - vcn_v4_0_unified_ring_set_wptr(ring); - ring->sched.ready = true; - } + ring->wptr = 0; + ring->wptr_old = 0; + vcn_v4_0_unified_ring_set_wptr(ring); + ring->sched.ready = true; + } } else { for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { From 4fc8fff378b2f2039f2a666d9f8c570f4e58352c Mon Sep 17 00:00:00 2001 From: Qu Huang Date: Tue, 21 Feb 2023 11:35:16 +0000 Subject: [PATCH 449/765] drm/amdkfd: Fix an illegal memory access In the kfd_wait_on_events() function, the kfd_event_waiter structure is allocated by alloc_event_waiters(), but the event field of the waiter structure is not initialized; When copy_from_user() fails in the kfd_wait_on_events() function, it will enter exception handling to release the previously allocated memory of the waiter structure; Due to the event field of the waiters structure being accessed in the free_waiters() function, this results in illegal memory access and system crash, here is the crash log: localhost kernel: RIP: 0010:native_queued_spin_lock_slowpath+0x185/0x1e0 localhost kernel: RSP: 0018:ffffaa53c362bd60 EFLAGS: 00010082 localhost kernel: RAX: ff3d3d6bff4007cb RBX: 0000000000000282 RCX: 00000000002c0000 localhost kernel: RDX: ffff9e855eeacb80 RSI: 000000000000279c RDI: ffffe7088f6a21d0 localhost kernel: RBP: ffffe7088f6a21d0 R08: 00000000002c0000 R09: ffffaa53c362be64 localhost kernel: R10: ffffaa53c362bbd8 R11: 0000000000000001 R12: 0000000000000002 localhost kernel: R13: ffff9e7ead15d600 R14: 0000000000000000 R15: ffff9e7ead15d698 localhost kernel: FS: 0000152a3d111700(0000) GS:ffff9e855ee80000(0000) knlGS:0000000000000000 localhost kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 localhost kernel: CR2: 0000152938000010 CR3: 000000044d7a4000 CR4: 00000000003506e0 localhost kernel: Call Trace: localhost kernel: _raw_spin_lock_irqsave+0x30/0x40 localhost kernel: remove_wait_queue+0x12/0x50 localhost kernel: kfd_wait_on_events+0x1b6/0x490 [hydcu] localhost kernel: ? ftrace_graph_caller+0xa0/0xa0 localhost kernel: kfd_ioctl+0x38c/0x4a0 [hydcu] localhost kernel: ? kfd_ioctl_set_trap_handler+0x70/0x70 [hydcu] localhost kernel: ? kfd_ioctl_create_queue+0x5a0/0x5a0 [hydcu] localhost kernel: ? ftrace_graph_caller+0xa0/0xa0 localhost kernel: __x64_sys_ioctl+0x8e/0xd0 localhost kernel: ? syscall_trace_enter.isra.18+0x143/0x1b0 localhost kernel: do_syscall_64+0x33/0x80 localhost kernel: entry_SYSCALL_64_after_hwframe+0x44/0xa9 localhost kernel: RIP: 0033:0x152a4dff68d7 Allocate the structure with kcalloc, and remove redundant 0-initialization and a redundant loop condition check. Signed-off-by: Qu Huang Signed-off-by: Felix Kuehling Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdkfd/kfd_events.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c index 729d26d648af3..2880ed96ac2e3 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c @@ -778,16 +778,13 @@ static struct kfd_event_waiter *alloc_event_waiters(uint32_t num_events) struct kfd_event_waiter *event_waiters; uint32_t i; - event_waiters = kmalloc_array(num_events, - sizeof(struct kfd_event_waiter), - GFP_KERNEL); + event_waiters = kcalloc(num_events, sizeof(struct kfd_event_waiter), + GFP_KERNEL); if (!event_waiters) return NULL; - for (i = 0; (event_waiters) && (i < num_events) ; i++) { + for (i = 0; i < num_events; i++) init_wait(&event_waiters[i].wait); - event_waiters[i].activated = false; - } return event_waiters; } From b299221faf9b62166413526be2438d21257f019e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Marek=20Ol=C5=A1=C3=A1k?= Date: Sun, 29 Jan 2023 23:00:59 -0500 Subject: [PATCH 450/765] drm/amdgpu: add more fields into device info, caches sizes, etc. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit AMDGPU_IDS_FLAGS_CONFORMANT_TRUNC_COORD: important for conformance on gfx11 Other fields are exposed from IP discovery. enabled_rb_pipes_mask_hi is added for future chips, currently 0. Mesa MR: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21403 Signed-off-by: Marek Olšák Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 5 ++++- drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h | 2 ++ drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 11 +++++++++++ drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 5 +++++ include/uapi/drm/amdgpu_drm.h | 11 +++++++++++ 5 files changed, 33 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index bfb7ed254ee4c..641bdcdab10ea 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -107,9 +107,12 @@ * - 3.50.0 - Update AMDGPU_INFO_DEV_INFO IOCTL for minimum engine and memory clock * Update AMDGPU_INFO_SENSOR IOCTL for PEAK_PSTATE engine and memory clock * 3.51.0 - Return the PCIe gen and lanes from the INFO ioctl + * 3.52.0 - Add AMDGPU_IDS_FLAGS_CONFORMANT_TRUNC_COORD, add device_info fields: + * tcp_cache_size, num_sqc_per_wgp, sqc_data_cache_size, sqc_inst_cache_size, + * gl1c_cache_size, gl2c_cache_size, mall_size, enabled_rb_pipes_mask_hi */ #define KMS_DRIVER_MAJOR 3 -#define KMS_DRIVER_MINOR 51 +#define KMS_DRIVER_MINOR 52 #define KMS_DRIVER_PATCHLEVEL 0 unsigned int amdgpu_vram_limit = UINT_MAX; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h index 86ec9d0d12c8e..de9e7a00bb150 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h @@ -178,6 +178,8 @@ struct amdgpu_gfx_config { uint32_t num_sc_per_sh; uint32_t num_packer_per_sc; uint32_t pa_sc_tile_steering_override; + /* Whether texture coordinate truncation is conformant. */ + bool ta_cntl2_truncate_coord_mode; uint64_t tcc_disabled_mask; uint32_t gc_num_tcp_per_sa; uint32_t gc_num_sdp_interface; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index ca945055e6836..0efb38539d70c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -808,6 +808,8 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) dev_info->ids_flags |= AMDGPU_IDS_FLAGS_PREEMPTION; if (amdgpu_is_tmz(adev)) dev_info->ids_flags |= AMDGPU_IDS_FLAGS_TMZ; + if (adev->gfx.config.ta_cntl2_truncate_coord_mode) + dev_info->ids_flags |= AMDGPU_IDS_FLAGS_CONFORMANT_TRUNC_COORD; vm_size = adev->vm_manager.max_pfn * AMDGPU_GPU_PAGE_SIZE; vm_size -= AMDGPU_VA_RESERVED_SIZE; @@ -865,6 +867,15 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X4 ? 4 : adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 ? 2 : 1; + dev_info->tcp_cache_size = adev->gfx.config.gc_tcp_l1_size; + dev_info->num_sqc_per_wgp = adev->gfx.config.gc_num_sqc_per_wgp; + dev_info->sqc_data_cache_size = adev->gfx.config.gc_l1_data_cache_size_per_sqc; + dev_info->sqc_inst_cache_size = adev->gfx.config.gc_l1_instruction_cache_size_per_sqc; + dev_info->gl1c_cache_size = adev->gfx.config.gc_gl1c_size_per_instance * + adev->gfx.config.gc_gl1c_per_sa; + dev_info->gl2c_cache_size = adev->gfx.config.gc_gl2c_per_gpu; + dev_info->mall_size = adev->gmc.mall_size; + ret = copy_to_user(out, dev_info, min((size_t)size, sizeof(*dev_info))) ? -EFAULT : 0; kfree(dev_info); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index 417ab8d1eacea..3bf697a80cf2f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -1659,6 +1659,11 @@ static void gfx_v11_0_constants_init(struct amdgpu_device *adev) gfx_v11_0_get_tcc_info(adev); adev->gfx.config.pa_sc_tile_steering_override = 0; + /* Set whether texture coordinate truncation is conformant. */ + tmp = RREG32_SOC15(GC, 0, regTA_CNTL2); + adev->gfx.config.ta_cntl2_truncate_coord_mode = + REG_GET_FIELD(tmp, TA_CNTL2, TRUNCATE_COORD_MODE); + /* XXX SH_MEM regs */ /* where to put LDS, scratch, GPUVM in FSA64 space */ mutex_lock(&adev->srbm_mutex); diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index 973af6d066260..b6eb90df5d052 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -715,6 +715,7 @@ struct drm_amdgpu_cs_chunk_data { #define AMDGPU_IDS_FLAGS_FUSION 0x1 #define AMDGPU_IDS_FLAGS_PREEMPTION 0x2 #define AMDGPU_IDS_FLAGS_TMZ 0x4 +#define AMDGPU_IDS_FLAGS_CONFORMANT_TRUNC_COORD 0x8 /* indicate if acceleration can be working */ #define AMDGPU_INFO_ACCEL_WORKING 0x00 @@ -1115,6 +1116,16 @@ struct drm_amdgpu_info_device { __u64 tcc_disabled_mask; __u64 min_engine_clock; __u64 min_memory_clock; + /* The following fields are only set on gfx11+, older chips set 0. */ + __u32 tcp_cache_size; /* AKA GL0, VMEM cache */ + __u32 num_sqc_per_wgp; + __u32 sqc_data_cache_size; /* AKA SMEM cache */ + __u32 sqc_inst_cache_size; + __u32 gl1c_cache_size; + __u32 gl2c_cache_size; + __u64 mall_size; /* AKA infinity cache */ + /* high 32 bits of the rb pipes mask */ + __u32 enabled_rb_pipes_mask_hi; }; struct drm_amdgpu_info_hw_ip { From edddc6fd542ffbae680c2201bbf6763f1693db4f Mon Sep 17 00:00:00 2001 From: Evan Quan Date: Tue, 21 Feb 2023 15:17:43 +0800 Subject: [PATCH 451/765] drm/amd/pm: correct the baco state setting for ArmD3 scenario The check for baco support relies on the correct baco state. Signed-off-by: Evan Quan Reviewed-by: Feifei Xu Signed-off-by: Alex Deucher --- .../gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c | 21 +++++++++++++++---- 1 file changed, 17 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c index 78945e79dbee1..66e9cb21497b5 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c @@ -2229,10 +2229,23 @@ int smu_v13_0_gfx_ulv_control(struct smu_context *smu, int smu_v13_0_baco_set_armd3_sequence(struct smu_context *smu, enum smu_baco_seq baco_seq) { - return smu_cmn_send_smc_msg_with_param(smu, - SMU_MSG_ArmD3, - baco_seq, - NULL); + struct smu_baco_context *smu_baco = &smu->smu_baco; + int ret; + + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_ArmD3, + baco_seq, + NULL); + if (ret) + return ret; + + if (baco_seq == BACO_SEQ_BAMACO || + baco_seq == BACO_SEQ_BACO) + smu_baco->state = SMU_BACO_STATE_ENTER; + else + smu_baco->state = SMU_BACO_STATE_EXIT; + + return 0; } bool smu_v13_0_baco_is_support(struct smu_context *smu) From 6761c4bfee681c306bbe6599951e74826660be47 Mon Sep 17 00:00:00 2001 From: Evan Quan Date: Tue, 21 Feb 2023 15:21:19 +0800 Subject: [PATCH 452/765] drm/amd/pm: no pptable resetup on runpm exiting It is assumed the pptable used before runpm is same as the one used afterwards. Thus, we can reuse the stored copy and do not need to resetup the pptable again. Signed-off-by: Evan Quan Reviewed-by: Feifei Xu Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c index 834d146c4991f..0652b001ad549 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c @@ -1202,10 +1202,17 @@ static int smu_smc_hw_setup(struct smu_context *smu) return ret; } - ret = smu_setup_pptable(smu); - if (ret) { - dev_err(adev->dev, "Failed to setup pptable!\n"); - return ret; + /* + * It is assumed the pptable used before runpm is same as + * the one used afterwards. Thus, we can reuse the stored + * copy and do not need to resetup the pptable again. + */ + if (!adev->in_runpm) { + ret = smu_setup_pptable(smu); + if (ret) { + dev_err(adev->dev, "Failed to setup pptable!\n"); + return ret; + } } /* smu_dump_pptable(smu); */ From e69c785723ed88a930d332e13bc9140dce48f359 Mon Sep 17 00:00:00 2001 From: Tao Zhou Date: Fri, 17 Feb 2023 11:16:10 +0800 Subject: [PATCH 453/765] drm/amdgpu: add umc retire unit element It records how many bad pages are retired in one uncorrectable error. Signed-off-by: Tao Zhou Reviewed-by: Stanley.Yang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h | 2 ++ drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c | 1 + drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c | 1 + drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 3 +++ 4 files changed, 7 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h index a6951160f13af..f2bf979af5883 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h @@ -74,6 +74,8 @@ struct amdgpu_umc { /* UMC regiser per channel offset */ uint32_t channel_offs; + /* how many pages are retired in one UE */ + uint32_t retire_unit; /* channel index table of interleaved memory */ const uint32_t *channel_idx_tbl; struct ras_common_if *ras_if; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index 7db1f1a7e33c3..ab2556ca984e1 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -692,6 +692,7 @@ static void gmc_v10_0_set_umc_funcs(struct amdgpu_device *adev) adev->umc.channel_inst_num = UMC_V8_7_CHANNEL_INSTANCE_NUM; adev->umc.umc_inst_num = UMC_V8_7_UMC_INSTANCE_NUM; adev->umc.channel_offs = UMC_V8_7_PER_CHANNEL_OFFSET_SIENNA; + adev->umc.retire_unit = 1; adev->umc.channel_idx_tbl = &umc_v8_7_channel_idx_tbl[0][0]; adev->umc.ras = &umc_v8_7_ras; break; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c index 0a31a341aa43b..85e0afc3d4f7f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c @@ -570,6 +570,7 @@ static void gmc_v11_0_set_umc_funcs(struct amdgpu_device *adev) adev->umc.node_inst_num = adev->gmc.num_umc; adev->umc.max_ras_err_cnt_per_query = UMC_V8_10_TOTAL_CHANNEL_NUM(adev); adev->umc.channel_offs = UMC_V8_10_PER_CHANNEL_OFFSET; + adev->umc.retire_unit = UMC_V8_10_NA_COL_2BITS_POWER_OF_2_NUM; if (adev->umc.node_inst_num == 4) adev->umc.channel_idx_tbl = &umc_v8_10_channel_idx_tbl_ext0[0][0][0]; else diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index d65c6cea34451..b06170c00dfca 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -1288,6 +1288,7 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev) adev->umc.channel_inst_num = UMC_V6_1_CHANNEL_INSTANCE_NUM; adev->umc.umc_inst_num = UMC_V6_1_UMC_INSTANCE_NUM; adev->umc.channel_offs = UMC_V6_1_PER_CHANNEL_OFFSET_VG20; + adev->umc.retire_unit = 1; adev->umc.channel_idx_tbl = &umc_v6_1_channel_idx_tbl[0][0]; adev->umc.ras = &umc_v6_1_ras; break; @@ -1296,6 +1297,7 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev) adev->umc.channel_inst_num = UMC_V6_1_CHANNEL_INSTANCE_NUM; adev->umc.umc_inst_num = UMC_V6_1_UMC_INSTANCE_NUM; adev->umc.channel_offs = UMC_V6_1_PER_CHANNEL_OFFSET_ARCT; + adev->umc.retire_unit = 1; adev->umc.channel_idx_tbl = &umc_v6_1_channel_idx_tbl[0][0]; adev->umc.ras = &umc_v6_1_ras; break; @@ -1305,6 +1307,7 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev) adev->umc.channel_inst_num = UMC_V6_7_CHANNEL_INSTANCE_NUM; adev->umc.umc_inst_num = UMC_V6_7_UMC_INSTANCE_NUM; adev->umc.channel_offs = UMC_V6_7_PER_CHANNEL_OFFSET; + adev->umc.retire_unit = (UMC_V6_7_NA_MAP_PA_NUM * 2); if (!adev->gmc.xgmi.connected_to_cpu) adev->umc.ras = &umc_v6_7_ras; if (1 & adev->smuio.funcs->get_die_id(adev)) From 4d33e0f1340b3d08002ff8f9bcbf256cfdc4f3ba Mon Sep 17 00:00:00 2001 From: Tao Zhou Date: Fri, 10 Feb 2023 16:33:58 +0800 Subject: [PATCH 454/765] drm/amdgpu: exclude duplicate pages from UMC RAS UE count If a UMC bad page is reserved but not freed by an application, the application may trigger uncorrectable error repeatly by accessing the page. v2: add specific function to do the check. v3: remove duplicate pages, calculate new added bad page number. v4: reuse save_bad_pages to calculate new added bad page number. Signed-off-by: Tao Zhou Reviewed-by: Stanley.Yang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 16 +++++++++++++--- drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h | 3 ++- drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c | 4 ++-- 3 files changed, 17 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 6e543558386da..5c02c6c9f7732 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -176,7 +176,7 @@ static int amdgpu_reserve_page_direct(struct amdgpu_device *adev, uint64_t addre if (amdgpu_bad_page_threshold != 0) { amdgpu_ras_add_bad_pages(adev, err_data.err_addr, err_data.err_addr_cnt); - amdgpu_ras_save_bad_pages(adev); + amdgpu_ras_save_bad_pages(adev, NULL); } dev_warn(adev->dev, "WARNING: THIS IS ONLY FOR TEST PURPOSES AND WILL CORRUPT RAS EEPROM\n"); @@ -2084,22 +2084,32 @@ int amdgpu_ras_add_bad_pages(struct amdgpu_device *adev, /* * write error record array to eeprom, the function should be * protected by recovery_lock + * new_cnt: new added UE count, excluding reserved bad pages, can be NULL */ -int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev) +int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev, + unsigned long *new_cnt) { struct amdgpu_ras *con = amdgpu_ras_get_context(adev); struct ras_err_handler_data *data; struct amdgpu_ras_eeprom_control *control; int save_count; - if (!con || !con->eh_data) + if (!con || !con->eh_data) { + if (new_cnt) + *new_cnt = 0; + return 0; + } mutex_lock(&con->recovery_lock); control = &con->eeprom_control; data = con->eh_data; save_count = data->count - control->ras_num_recs; mutex_unlock(&con->recovery_lock); + + if (new_cnt) + *new_cnt = save_count / adev->umc.retire_unit; + /* only new entries are saved */ if (save_count > 0) { if (amdgpu_ras_eeprom_append(control, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h index f2ad999993f66..ef38f4c93df0a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h @@ -547,7 +547,8 @@ int amdgpu_ras_query_error_count(struct amdgpu_device *adev, int amdgpu_ras_add_bad_pages(struct amdgpu_device *adev, struct eeprom_table_record *bps, int pages); -int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev); +int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev, + unsigned long *new_cnt); static inline enum ta_ras_block amdgpu_ras_block_to_ta(enum amdgpu_ras_block block) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c index 1c7fcb4f23808..1b8574bc4463d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c @@ -68,7 +68,7 @@ int amdgpu_umc_page_retirement_mca(struct amdgpu_device *adev, if (amdgpu_bad_page_threshold != 0) { amdgpu_ras_add_bad_pages(adev, err_data.err_addr, err_data.err_addr_cnt); - amdgpu_ras_save_bad_pages(adev); + amdgpu_ras_save_bad_pages(adev, NULL); } out: @@ -147,7 +147,7 @@ static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, err_data->err_addr_cnt) { amdgpu_ras_add_bad_pages(adev, err_data->err_addr, err_data->err_addr_cnt); - amdgpu_ras_save_bad_pages(adev); + amdgpu_ras_save_bad_pages(adev, &(err_data->ue_count)); amdgpu_dpm_send_hbm_bad_pages_num(adev, con->eeprom_control.ras_num_recs); From f3cbe70e215a87dcfdf028582a2fa94b24a08efe Mon Sep 17 00:00:00 2001 From: Tao Zhou Date: Tue, 21 Feb 2023 15:25:01 +0800 Subject: [PATCH 455/765] drm/amdgpu: change default behavior of bad_page_threshold parameter Ignore ras umc bad page threshold by default, GPU initialization won't be stopped in this mode. v2: refine the description of bad_page_threshold. Signed-off-by: Tao Zhou Reviewed-by: Stanley.Yang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 7 ++++--- drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c | 4 ++-- 3 files changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 641bdcdab10ea..f5ffca24def40 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -924,7 +924,7 @@ module_param_named(reset_method, amdgpu_reset_method, int, 0444); * result in the GPU entering bad status when the number of total * faulty pages by ECC exceeds the threshold value. */ -MODULE_PARM_DESC(bad_page_threshold, "Bad page threshold(-1 = auto(default value), 0 = disable bad page retirement, -2 = ignore bad page threshold)"); +MODULE_PARM_DESC(bad_page_threshold, "Bad page threshold(-1 = ignore threshold (default value), 0 = disable bad page retirement, -2 = driver sets threshold)"); module_param_named(bad_page_threshold, amdgpu_bad_page_threshold, int, 0444); MODULE_PARM_DESC(num_kcq, "number of kernel compute queue user want to setup (8 if set to greater than 8 or less than 0, only affect gfx 8+)"); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 5c02c6c9f7732..63dfcc98152d5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -2196,11 +2196,12 @@ static void amdgpu_ras_validate_threshold(struct amdgpu_device *adev, /* * Justification of value bad_page_cnt_threshold in ras structure * - * Generally, -1 <= amdgpu_bad_page_threshold <= max record length - * in eeprom, and introduce two scenarios accordingly. + * Generally, 0 <= amdgpu_bad_page_threshold <= max record length + * in eeprom or amdgpu_bad_page_threshold == -2, introduce two + * scenarios accordingly. * * Bad page retirement enablement: - * - If amdgpu_bad_page_threshold = -1, + * - If amdgpu_bad_page_threshold = -2, * bad_page_cnt_threshold = typical value by formula. * * - When the value from user is 0 < amdgpu_bad_page_threshold < diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c index 2d9f3f4cd79e9..9d370465b08d2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c @@ -1191,8 +1191,8 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control, } else { dev_err(adev->dev, "RAS records:%d exceed threshold:%d", control->ras_num_recs, ras->bad_page_cnt_threshold); - if (amdgpu_bad_page_threshold == -2) { - dev_warn(adev->dev, "GPU will be initialized due to bad_page_threshold = -2."); + if (amdgpu_bad_page_threshold == -1) { + dev_warn(adev->dev, "GPU will be initialized due to bad_page_threshold = -1."); res = 0; } else { *exceed_err_limit = true; From 22106ed0be0d6c5b4aa07e18b63c1245bdb719c9 Mon Sep 17 00:00:00 2001 From: Tao Zhou Date: Tue, 21 Feb 2023 16:03:49 +0800 Subject: [PATCH 456/765] drm/amdgpu: add bad_page_threshold check in ras_eeprom_check_err bad_page_threshold controls page retirement behavior and it should be also checked. v2: simplify the condition of bad page handling path. Signed-off-by: Tao Zhou Reviewed-by: Stanley.Yang Signed-off-by: Alex Deucher --- .../gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c | 19 ++++++++++++++----- 1 file changed, 14 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c index 9d370465b08d2..2e08fce875217 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c @@ -417,7 +417,8 @@ bool amdgpu_ras_eeprom_check_err_threshold(struct amdgpu_device *adev) { struct amdgpu_ras *con = amdgpu_ras_get_context(adev); - if (!__is_ras_eeprom_supported(adev)) + if (!__is_ras_eeprom_supported(adev) || + !amdgpu_bad_page_threshold) return false; /* skip check eeprom table for VEGA20 Gaming */ @@ -428,10 +429,18 @@ bool amdgpu_ras_eeprom_check_err_threshold(struct amdgpu_device *adev) return false; if (con->eeprom_control.tbl_hdr.header == RAS_TABLE_HDR_BAD) { - dev_warn(adev->dev, "This GPU is in BAD status."); - dev_warn(adev->dev, "Please retire it or set a larger " - "threshold value when reloading driver.\n"); - return true; + if (amdgpu_bad_page_threshold == -1) { + dev_warn(adev->dev, "RAS records:%d exceed threshold:%d", + con->eeprom_control.ras_num_recs, con->bad_page_cnt_threshold); + dev_warn(adev->dev, + "But GPU can be operated due to bad_page_threshold = -1.\n"); + return false; + } else { + dev_warn(adev->dev, "This GPU is in BAD status."); + dev_warn(adev->dev, "Please retire it or set a larger " + "threshold value when reloading driver.\n"); + return true; + } } return false; From 2d53b579f3f217d5b88fb6708dcaef28f7b9fc0b Mon Sep 17 00:00:00 2001 From: Candice Li Date: Wed, 15 Feb 2023 21:16:56 +0800 Subject: [PATCH 457/765] drm/amdgpu: Add convert_error_address function for umc v8_10 Add convert_error_address for umc v8_10. Signed-off-by: Candice Li Reviewed-by: Tao Zhou Reviewed-by: Stanley.Yang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/umc_v8_10.c | 73 +++++++++++++++----------- 1 file changed, 42 insertions(+), 31 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c b/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c index da394bc06bbaa..293ba39c8a2fd 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c +++ b/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c @@ -209,6 +209,45 @@ static int umc_v8_10_swizzle_mode_na_to_pa(struct amdgpu_device *adev, return 0; } +void umc_v8_10_convert_error_address(struct amdgpu_device *adev, + struct ras_err_data *err_data, uint64_t err_addr, + uint32_t ch_inst, uint32_t umc_inst, + uint32_t node_inst, uint64_t mc_umc_status) +{ + uint64_t na_err_addr_base; + uint64_t na_err_addr, retired_page_addr; + uint32_t channel_index, addr_lsb, col = 0; + int ret = 0; + + channel_index = + adev->umc.channel_idx_tbl[node_inst * adev->umc.umc_inst_num * + adev->umc.channel_inst_num + + umc_inst * adev->umc.channel_inst_num + + ch_inst]; + + /* the lowest lsb bits should be ignored */ + addr_lsb = REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, AddrLsb); + err_addr &= ~((0x1ULL << addr_lsb) - 1); + na_err_addr_base = err_addr & ~(0x3ULL << UMC_V8_10_NA_C5_BIT); + + /* loop for all possibilities of [C6 C5] in normal address. */ + for (col = 0; col < UMC_V8_10_NA_COL_2BITS_POWER_OF_2_NUM; col++) { + na_err_addr = na_err_addr_base | (col << UMC_V8_10_NA_C5_BIT); + + /* Mapping normal error address to retired soc physical address. */ + ret = umc_v8_10_swizzle_mode_na_to_pa(adev, channel_index, + na_err_addr, &retired_page_addr); + if (ret) { + dev_err(adev->dev, "Failed to map pa from umc na.\n"); + break; + } + dev_info(adev->dev, "Error Address(PA): 0x%llx\n", + retired_page_addr); + amdgpu_umc_fill_error_record(err_data, na_err_addr, + retired_page_addr, channel_index, umc_inst); + } +} + static void umc_v8_10_query_error_address(struct amdgpu_device *adev, struct ras_err_data *err_data, uint32_t umc_reg_offset, @@ -218,10 +257,7 @@ static void umc_v8_10_query_error_address(struct amdgpu_device *adev, { uint64_t mc_umc_status_addr; uint64_t mc_umc_status, err_addr; - uint64_t mc_umc_addrt0, na_err_addr_base; - uint64_t na_err_addr, retired_page_addr; - uint32_t channel_index, addr_lsb, col = 0; - int ret = 0; + uint64_t mc_umc_addrt0; mc_umc_status_addr = SOC15_REG_OFFSET(UMC, 0, regMCA_UMC_UMC0_MCUMC_STATUST0); @@ -236,12 +272,6 @@ static void umc_v8_10_query_error_address(struct amdgpu_device *adev, return; } - channel_index = - adev->umc.channel_idx_tbl[node_inst * adev->umc.umc_inst_num * - adev->umc.channel_inst_num + - umc_inst * adev->umc.channel_inst_num + - ch_inst]; - /* calculate error address if ue error is detected */ if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 && REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, AddrV) == 1 && @@ -251,27 +281,8 @@ static void umc_v8_10_query_error_address(struct amdgpu_device *adev, err_addr = RREG64_PCIE((mc_umc_addrt0 + umc_reg_offset) * 4); err_addr = REG_GET_FIELD(err_addr, MCA_UMC_UMC0_MCUMC_ADDRT0, ErrorAddr); - /* the lowest lsb bits should be ignored */ - addr_lsb = REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, AddrLsb); - err_addr &= ~((0x1ULL << addr_lsb) - 1); - na_err_addr_base = err_addr & ~(0x3ULL << UMC_V8_10_NA_C5_BIT); - - /* loop for all possibilities of [C6 C5] in normal address. */ - for (col = 0; col < UMC_V8_10_NA_COL_2BITS_POWER_OF_2_NUM; col++) { - na_err_addr = na_err_addr_base | (col << UMC_V8_10_NA_C5_BIT); - - /* Mapping normal error address to retired soc physical address. */ - ret = umc_v8_10_swizzle_mode_na_to_pa(adev, channel_index, - na_err_addr, &retired_page_addr); - if (ret) { - dev_err(adev->dev, "Failed to map pa from umc na.\n"); - break; - } - dev_info(adev->dev, "Error Address(PA): 0x%llx\n", - retired_page_addr); - amdgpu_umc_fill_error_record(err_data, na_err_addr, - retired_page_addr, channel_index, umc_inst); - } + umc_v8_10_convert_error_address(adev, err_data, err_addr, + ch_inst, umc_inst, node_inst, mc_umc_status); } /* clear umc status */ From b1e9a718af2ec3d21734a8357e8f22aa3bb68bfb Mon Sep 17 00:00:00 2001 From: Candice Li Date: Mon, 16 Jan 2023 16:18:21 +0800 Subject: [PATCH 458/765] drm/amdgpu: Add ecc info query interface for umc v8_10 Support ecc info query for umc v8_10. v2: Simplied by convert_error_address. v3: Remove unused variable and invalid checking. Signed-off-by: Candice Li Reviewed-by: Tao Zhou Reviewed-by: Stanley.Yang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/umc_v8_10.c | 134 +++++++++++++++++++++++++ 1 file changed, 134 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c b/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c index 293ba39c8a2fd..66158219f791c 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c +++ b/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c @@ -360,6 +360,138 @@ static bool umc_v8_10_query_ras_poison_mode(struct amdgpu_device *adev) return true; } +static void umc_v8_10_ecc_info_query_correctable_error_count(struct amdgpu_device *adev, + uint32_t node_inst, uint32_t umc_inst, uint32_t ch_inst, + unsigned long *error_count) +{ + uint64_t mc_umc_status; + uint32_t eccinfo_table_idx; + struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); + + eccinfo_table_idx = node_inst * adev->umc.umc_inst_num * + adev->umc.channel_inst_num + + umc_inst * adev->umc.channel_inst_num + + ch_inst; + + /* check the MCUMC_STATUS */ + mc_umc_status = ras->umc_ecc.ecc[eccinfo_table_idx].mca_umc_status; + if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 && + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, CECC) == 1) { + *error_count += 1; + } +} + +static void umc_v8_10_ecc_info_query_uncorrectable_error_count(struct amdgpu_device *adev, + uint32_t node_inst, uint32_t umc_inst, uint32_t ch_inst, + unsigned long *error_count) +{ + uint64_t mc_umc_status; + uint32_t eccinfo_table_idx; + struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); + + eccinfo_table_idx = node_inst * adev->umc.umc_inst_num * + adev->umc.channel_inst_num + + umc_inst * adev->umc.channel_inst_num + + ch_inst; + + /* check the MCUMC_STATUS */ + mc_umc_status = ras->umc_ecc.ecc[eccinfo_table_idx].mca_umc_status; + if ((REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1) && + (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Deferred) == 1 || + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) == 1 || + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, PCC) == 1 || + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UC) == 1 || + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, TCC) == 1)) { + *error_count += 1; + } +} + +static void umc_v8_10_ecc_info_query_ras_error_count(struct amdgpu_device *adev, + void *ras_error_status) +{ + struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; + + uint32_t node_inst = 0; + uint32_t umc_inst = 0; + uint32_t ch_inst = 0; + + /* TODO: driver needs to toggle DF Cstate to ensure + * safe access of UMC registers. Will add the protection + */ + LOOP_UMC_EACH_NODE_INST_AND_CH(node_inst, umc_inst, ch_inst) { + umc_v8_10_ecc_info_query_correctable_error_count(adev, + node_inst, umc_inst, ch_inst, + &(err_data->ce_count)); + umc_v8_10_ecc_info_query_uncorrectable_error_count(adev, + node_inst, umc_inst, ch_inst, + &(err_data->ue_count)); + } +} + +static void umc_v8_10_ecc_info_query_error_address(struct amdgpu_device *adev, + struct ras_err_data *err_data, + uint32_t ch_inst, + uint32_t umc_inst, + uint32_t node_inst) +{ + uint32_t eccinfo_table_idx, channel_index; + uint64_t mc_umc_status, err_addr; + + struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); + + eccinfo_table_idx = node_inst * adev->umc.umc_inst_num * + adev->umc.channel_inst_num + + umc_inst * adev->umc.channel_inst_num + + ch_inst; + channel_index = + adev->umc.channel_idx_tbl[node_inst * adev->umc.umc_inst_num * + adev->umc.channel_inst_num + + umc_inst * adev->umc.channel_inst_num + + ch_inst]; + + mc_umc_status = ras->umc_ecc.ecc[eccinfo_table_idx].mca_umc_status; + + if (mc_umc_status == 0) + return; + + if (!err_data->err_addr) + return; + + /* calculate error address if ue error is detected */ + if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 && + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, AddrV) == 1 && + (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) == 1)) { + + err_addr = ras->umc_ecc.ecc[eccinfo_table_idx].mca_umc_addr; + err_addr = REG_GET_FIELD(err_addr, MCA_UMC_UMC0_MCUMC_ADDRT0, ErrorAddr); + + umc_v8_10_convert_error_address(adev, err_data, err_addr, + ch_inst, umc_inst, node_inst, mc_umc_status); + } +} + +static void umc_v8_10_ecc_info_query_ras_error_address(struct amdgpu_device *adev, + void *ras_error_status) +{ + struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; + + uint32_t node_inst = 0; + uint32_t umc_inst = 0; + uint32_t ch_inst = 0; + + /* TODO: driver needs to toggle DF Cstate to ensure + * safe access of UMC resgisters. Will add the protection + * when firmware interface is ready + */ + LOOP_UMC_EACH_NODE_INST_AND_CH(node_inst, umc_inst, ch_inst) { + umc_v8_10_ecc_info_query_error_address(adev, + err_data, + ch_inst, + umc_inst, + node_inst); + } +} + const struct amdgpu_ras_block_hw_ops umc_v8_10_ras_hw_ops = { .query_ras_error_count = umc_v8_10_query_ras_error_count, .query_ras_error_address = umc_v8_10_query_ras_error_address, @@ -371,4 +503,6 @@ struct amdgpu_umc_ras umc_v8_10_ras = { }, .err_cnt_init = umc_v8_10_err_cnt_init, .query_ras_poison_mode = umc_v8_10_query_ras_poison_mode, + .ecc_info_query_ras_error_count = umc_v8_10_ecc_info_query_ras_error_count, + .ecc_info_query_ras_error_address = umc_v8_10_ecc_info_query_ras_error_address, }; From 424b3d7582a2a4a7c45d405225ac70cff97f2e4a Mon Sep 17 00:00:00 2001 From: Guchun Chen Date: Tue, 21 Feb 2023 12:35:25 +0800 Subject: [PATCH 459/765] drm/amd/pm: downgrade log level upon SMU IF version mismatch SMU IF version mismatch as a warning message exists widely after asic production, however, due to this log level setting, such mismatch warning will be caught by automation test like IGT and reported as a fake error after checking. As such mismatch does not break anything, to reduce confusion, downgrade it from dev_warn to dev_info. Signed-off-by: Guchun Chen Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c | 4 ++-- drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c | 4 ++-- drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c | 4 ++-- 3 files changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c index 6492d69e2e60f..e1ef88ee1ed39 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c @@ -256,7 +256,7 @@ int smu_v11_0_check_fw_version(struct smu_context *smu) * to be backward compatible. * 2. New fw usually brings some optimizations. But that's visible * only on the paired driver. - * Considering above, we just leave user a warning message instead + * Considering above, we just leave user a verbal message instead * of halt driver loading. */ if (if_version != smu->smc_driver_if_version) { @@ -264,7 +264,7 @@ int smu_v11_0_check_fw_version(struct smu_context *smu) "smu fw program = %d, version = 0x%08x (%d.%d.%d)\n", smu->smc_driver_if_version, if_version, smu_program, smu_version, smu_major, smu_minor, smu_debug); - dev_warn(smu->adev->dev, "SMU driver if version not matched\n"); + dev_info(smu->adev->dev, "SMU driver if version not matched\n"); } return ret; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c index 56a02bc60ceee..c788aa7a99a9e 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c @@ -93,7 +93,7 @@ int smu_v12_0_check_fw_version(struct smu_context *smu) * to be backward compatible. * 2. New fw usually brings some optimizations. But that's visible * only on the paired driver. - * Considering above, we just leave user a warning message instead + * Considering above, we just leave user a verbal message instead * of halt driver loading. */ if (if_version != smu->smc_driver_if_version) { @@ -101,7 +101,7 @@ int smu_v12_0_check_fw_version(struct smu_context *smu) "smu fw program = %d, smu fw version = 0x%08x (%d.%d.%d)\n", smu->smc_driver_if_version, if_version, smu_program, smu_version, smu_major, smu_minor, smu_debug); - dev_warn(smu->adev->dev, "SMU driver if version not matched\n"); + dev_info(smu->adev->dev, "SMU driver if version not matched\n"); } return ret; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c index 66e9cb21497b5..a52ed0580fd7e 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c @@ -311,7 +311,7 @@ int smu_v13_0_check_fw_version(struct smu_context *smu) * to be backward compatible. * 2. New fw usually brings some optimizations. But that's visible * only on the paired driver. - * Considering above, we just leave user a warning message instead + * Considering above, we just leave user a verbal message instead * of halt driver loading. */ if (if_version != smu->smc_driver_if_version) { @@ -319,7 +319,7 @@ int smu_v13_0_check_fw_version(struct smu_context *smu) "smu fw program = %d, smu fw version = 0x%08x (%d.%d.%d)\n", smu->smc_driver_if_version, if_version, smu_program, smu_version, smu_major, smu_minor, smu_debug); - dev_warn(adev->dev, "SMU driver if version not matched\n"); + dev_info(adev->dev, "SMU driver if version not matched\n"); } return ret; From 2a03472262c05f965d7ba394ed35dc9867ba3095 Mon Sep 17 00:00:00 2001 From: Eric Van Hensbergen Date: Mon, 28 Nov 2022 00:38:36 +0000 Subject: [PATCH 460/765] net/9p: Adjust maximum MSIZE to account for p9 header Add maximum p9 header size to MSIZE to make sure we can have page aligned data. Signed-off-by: Eric Van Hensbergen Reviewed-by: Dominique Martinet --- net/9p/client.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/net/9p/client.c b/net/9p/client.c index 622ec6a586eea..6c2a768a6ab15 100644 --- a/net/9p/client.c +++ b/net/9p/client.c @@ -28,7 +28,11 @@ #define CREATE_TRACE_POINTS #include -#define DEFAULT_MSIZE (128 * 1024) +/* DEFAULT MSIZE = 32 pages worth of payload + P9_HDRSZ + + * room for write (16 extra) or read (11 extra) operands. + */ + +#define DEFAULT_MSIZE ((128 * 1024) + P9_IOHDRSZ) /* Client Option Parsing (code inspired by NFS code) * - a little lazy - parse all client options From 344504e912ea49033d012dad9de3f68e20c07634 Mon Sep 17 00:00:00 2001 From: Eric Van Hensbergen Date: Thu, 8 Dec 2022 02:03:32 +0000 Subject: [PATCH 461/765] fs/9p: Expand setup of writeback cache to all levels If cache is enabled, make sure we are putting the right things in place (mainly impacts mmap). This also sets us up for more cache levels. Signed-off-by: Eric Van Hensbergen Reviewed-by: Dominique Martinet --- fs/9p/v9fs.c | 2 +- fs/9p/vfs_addr.c | 2 -- fs/9p/vfs_file.c | 7 ++++--- fs/9p/vfs_inode.c | 3 +-- fs/9p/vfs_inode_dotl.c | 7 ++++--- 5 files changed, 10 insertions(+), 11 deletions(-) diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c index 3a9c4517265fa..61a51b90600dc 100644 --- a/fs/9p/v9fs.c +++ b/fs/9p/v9fs.c @@ -468,7 +468,7 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses, #ifdef CONFIG_9P_FSCACHE /* register the session for caching */ - if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) { + if (v9ses->cache == CACHE_FSCACHE) { rc = v9fs_cache_session_get_cookie(v9ses, dev_name); if (rc < 0) goto err_clnt; diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c index 97599edbc300b..6f46d7e4c7509 100644 --- a/fs/9p/vfs_addr.c +++ b/fs/9p/vfs_addr.c @@ -279,8 +279,6 @@ static int v9fs_write_begin(struct file *filp, struct address_space *mapping, p9_debug(P9_DEBUG_VFS, "filp %p, mapping %p\n", filp, mapping); - BUG_ON(!v9inode->writeback_fid); - /* Prefetch area to be written into the cache if we're caching this * file. We need to do this before we get a lock on the page in case * there's more than one writer competing for the same cache block. diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c index b740017634ef1..3b6458846a0b2 100644 --- a/fs/9p/vfs_file.c +++ b/fs/9p/vfs_file.c @@ -73,8 +73,7 @@ int v9fs_file_open(struct inode *inode, struct file *file) } mutex_lock(&v9inode->v_mutex); - if ((v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) && - !v9inode->writeback_fid && + if ((v9ses->cache) && !v9inode->writeback_fid && ((file->f_flags & O_ACCMODE) != O_RDONLY)) { /* * clone a fid and add it to writeback_fid @@ -92,9 +91,11 @@ int v9fs_file_open(struct inode *inode, struct file *file) v9inode->writeback_fid = (void *) writeback_fid; } mutex_unlock(&v9inode->v_mutex); - if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) +#ifdef CONFIG_9P_FSCACHE + if (v9ses->cache == CACHE_FSCACHE) fscache_use_cookie(v9fs_inode_cookie(v9inode), file->f_mode & FMODE_WRITE); +#endif v9fs_open_fid_add(inode, &fid); return 0; out_error: diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 27a04a226d973..33e521c60e2cc 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -843,8 +843,7 @@ v9fs_vfs_atomic_open(struct inode *dir, struct dentry *dentry, inode = d_inode(dentry); v9inode = V9FS_I(inode); mutex_lock(&v9inode->v_mutex); - if ((v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) && - !v9inode->writeback_fid && + if ((v9ses->cache) && !v9inode->writeback_fid && ((flags & O_ACCMODE) != O_RDONLY)) { /* * clone a fid and add it to writeback_fid diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c index f806b3f116496..bff37a312e64a 100644 --- a/fs/9p/vfs_inode_dotl.c +++ b/fs/9p/vfs_inode_dotl.c @@ -316,8 +316,7 @@ v9fs_vfs_atomic_open_dotl(struct inode *dir, struct dentry *dentry, v9inode = V9FS_I(inode); mutex_lock(&v9inode->v_mutex); - if ((v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) && - !v9inode->writeback_fid && + if ((v9ses->cache) && !v9inode->writeback_fid && ((flags & O_ACCMODE) != O_RDONLY)) { /* * clone a fid and add it to writeback_fid @@ -340,9 +339,11 @@ v9fs_vfs_atomic_open_dotl(struct inode *dir, struct dentry *dentry, if (err) goto out; file->private_data = ofid; - if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) +#ifdef CONFIG_9P_FSCACHE + if (v9ses->cache == CACHE_FSCACHE) fscache_use_cookie(v9fs_inode_cookie(v9inode), file->f_mode & FMODE_WRITE); +#endif v9fs_open_fid_add(inode, &ofid); file->f_mode |= FMODE_CREATED; out: From e90ff8ede777b98b44611b416b1ae6be94258335 Mon Sep 17 00:00:00 2001 From: Sean Anderson Date: Thu, 22 Dec 2022 16:45:31 -0500 Subject: [PATCH 462/765] rtc: abx80x: Add nvmem support This adds support for the 256-byte internal RAM. There are two windows which can be used to access this RAM: 64 bytes at 0x40 (the "standard" address space) and 128 bytes at 0x80 (the "alternate" address space). We use the standard address space because it is also accessible over SPI (if such a port is ever done). We are limited to 32-byte reads for SMBus compatibility, so there's no advantage to using the alternate address space. There are some reserved bits in the EXTRAM register, and the datasheet doesn't say what to do with them. I've opted to skip a read/modify/write and just write the whole thing. If this driver is ever converted to regmap, this would be a good place to use regmap_update_bits. Signed-off-by: Sean Anderson Link: https://lore.kernel.org/r/20221222214532.1873718-1-sean.anderson@seco.com Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-abx80x.c | 77 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 77 insertions(+) diff --git a/drivers/rtc/rtc-abx80x.c b/drivers/rtc/rtc-abx80x.c index 2e0e6432901b8..f34a2e59cac76 100644 --- a/drivers/rtc/rtc-abx80x.c +++ b/drivers/rtc/rtc-abx80x.c @@ -11,6 +11,7 @@ */ #include +#include #include #include #include @@ -88,6 +89,16 @@ #define ABX8XX_TRICKLE_STANDARD_DIODE 0x8 #define ABX8XX_TRICKLE_SCHOTTKY_DIODE 0x4 +#define ABX8XX_REG_EXTRAM 0x3f +#define ABX8XX_EXTRAM_XADS GENMASK(1, 0) + +#define ABX8XX_SRAM_BASE 0x40 +#define ABX8XX_SRAM_WIN_SIZE 0x40 +#define ABX8XX_RAM_SIZE 256 + +#define NVMEM_ADDR_LOWER GENMASK(5, 0) +#define NVMEM_ADDR_UPPER GENMASK(7, 6) + static u8 trickle_resistors[] = {0, 3, 6, 11}; enum abx80x_chip {AB0801, AB0803, AB0804, AB0805, @@ -674,6 +685,68 @@ static int abx80x_setup_watchdog(struct abx80x_priv *priv) } #endif +static int abx80x_nvmem_xfer(struct abx80x_priv *priv, unsigned int offset, + void *val, size_t bytes, bool write) +{ + int ret; + + while (bytes) { + u8 extram, reg, len, lower, upper; + + lower = FIELD_GET(NVMEM_ADDR_LOWER, offset); + upper = FIELD_GET(NVMEM_ADDR_UPPER, offset); + extram = FIELD_PREP(ABX8XX_EXTRAM_XADS, upper); + reg = ABX8XX_SRAM_BASE + lower; + len = min(lower + bytes, (size_t)ABX8XX_SRAM_WIN_SIZE) - lower; + len = min_t(u8, len, I2C_SMBUS_BLOCK_MAX); + + ret = i2c_smbus_write_byte_data(priv->client, ABX8XX_REG_EXTRAM, + extram); + if (ret) + return ret; + + if (write) + ret = i2c_smbus_write_i2c_block_data(priv->client, reg, + len, val); + else + ret = i2c_smbus_read_i2c_block_data(priv->client, reg, + len, val); + if (ret) + return ret; + + offset += len; + val += len; + bytes -= len; + } + + return 0; +} + +static int abx80x_nvmem_read(void *priv, unsigned int offset, void *val, + size_t bytes) +{ + return abx80x_nvmem_xfer(priv, offset, val, bytes, false); +} + +static int abx80x_nvmem_write(void *priv, unsigned int offset, void *val, + size_t bytes) +{ + return abx80x_nvmem_xfer(priv, offset, val, bytes, true); +} + +static int abx80x_setup_nvmem(struct abx80x_priv *priv) +{ + struct nvmem_config config = { + .type = NVMEM_TYPE_BATTERY_BACKED, + .reg_read = abx80x_nvmem_read, + .reg_write = abx80x_nvmem_write, + .size = ABX8XX_RAM_SIZE, + .priv = priv, + }; + + return devm_rtc_nvmem_register(priv->rtc, &config); +} + static const struct i2c_device_id abx80x_id[] = { { "abx80x", ABX80X }, { "ab0801", AB0801 }, @@ -840,6 +913,10 @@ static int abx80x_probe(struct i2c_client *client) return err; } + err = abx80x_setup_nvmem(priv); + if (err) + return err; + if (client->irq > 0) { dev_info(&client->dev, "IRQ %d supplied\n", client->irq); err = devm_request_threaded_irq(&client->dev, client->irq, NULL, From b5bfa7277ee7d944421e0ef193586c6e34d7492c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Nuno=20S=C3=A1?= Date: Fri, 24 Feb 2023 11:45:51 +0100 Subject: [PATCH 463/765] ASoC: adau7118: don't disable regulators on device unbind MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The regulators are supposed to be controlled through the set_bias_level() component callback. Moreover, the regulators are not enabled during probe and so, this would lead to a regulator unbalanced use count. Fixes: ca514c0f12b02 ("ASOC: Add ADAU7118 8 Channel PDM-to-I2S/TDM Converter driver") Signed-off-by: Nuno Sá Link: https://lore.kernel.org/r/20230224104551.1139981-1-nuno.sa@analog.com Signed-off-by: Mark Brown --- sound/soc/codecs/adau7118.c | 19 +------------------ 1 file changed, 1 insertion(+), 18 deletions(-) diff --git a/sound/soc/codecs/adau7118.c b/sound/soc/codecs/adau7118.c index bbb0972498876..a663d37e57760 100644 --- a/sound/soc/codecs/adau7118.c +++ b/sound/soc/codecs/adau7118.c @@ -444,22 +444,6 @@ static const struct snd_soc_component_driver adau7118_component_driver = { .endianness = 1, }; -static void adau7118_regulator_disable(void *data) -{ - struct adau7118_data *st = data; - int ret; - /* - * If we fail to disable DVDD, don't bother in trying IOVDD. We - * actually don't want to be left in the situation where DVDD - * is enabled and IOVDD is disabled. - */ - ret = regulator_disable(st->dvdd); - if (ret) - return; - - regulator_disable(st->iovdd); -} - static int adau7118_regulator_setup(struct adau7118_data *st) { st->iovdd = devm_regulator_get(st->dev, "iovdd"); @@ -481,8 +465,7 @@ static int adau7118_regulator_setup(struct adau7118_data *st) regcache_cache_only(st->map, true); } - return devm_add_action_or_reset(st->dev, adau7118_regulator_disable, - st); + return 0; } static int adau7118_parset_dt(const struct adau7118_data *st) From f1956f4ec15195ec60976d9b5625326285ab102e Mon Sep 17 00:00:00 2001 From: Juergen Gross Date: Mon, 30 Jan 2023 12:30:35 +0100 Subject: [PATCH 464/765] 9p/xen: fix version parsing When connecting the Xen 9pfs frontend to the backend, the "versions" Xenstore entry written by the backend is parsed in a wrong way. The "versions" entry is defined to contain the versions supported by the backend separated by commas (e.g. "1,2"). Today only version "1" is defined. Unfortunately the frontend doesn't look for "1" being listed in the entry, but it is expecting the entry to have the value "1". This will result in failure as soon as the backend will support e.g. versions "1" and "2". Fix that by scanning the entry correctly. Link: https://lkml.kernel.org/r/20230130113036.7087-2-jgross@suse.com Fixes: 71ebd71921e4 ("xen/9pfs: connect to the backend") Signed-off-by: Juergen Gross Reviewed-by: Simon Horman Signed-off-by: Dominique Martinet Signed-off-by: Eric Van Hensbergen --- net/9p/trans_xen.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/net/9p/trans_xen.c b/net/9p/trans_xen.c index 9630b12755579..6298d77311535 100644 --- a/net/9p/trans_xen.c +++ b/net/9p/trans_xen.c @@ -379,13 +379,19 @@ static int xen_9pfs_front_probe(struct xenbus_device *dev, int ret, i; struct xenbus_transaction xbt; struct xen_9pfs_front_priv *priv = NULL; - char *versions; + char *versions, *v; unsigned int max_rings, max_ring_order, len = 0; versions = xenbus_read(XBT_NIL, dev->otherend, "versions", &len); if (IS_ERR(versions)) return PTR_ERR(versions); - if (strcmp(versions, "1")) { + for (v = versions; *v; v++) { + if (simple_strtoul(v, &v, 10) == 1) { + v = NULL; + break; + } + } + if (v) { kfree(versions); return -EINVAL; } From c15fe55d14b3b4ded5af2a3260877460a6ffb8ad Mon Sep 17 00:00:00 2001 From: Juergen Gross Date: Mon, 30 Jan 2023 12:30:36 +0100 Subject: [PATCH 465/765] 9p/xen: fix connection sequence Today the connection sequence of the Xen 9pfs frontend doesn't match the documented sequence. It can work reliably only for a PV 9pfs device having been added at boot time already, as the frontend is not waiting for the backend to have set its state to "XenbusStateInitWait" before reading the backend properties from Xenstore. Fix that by following the documented sequence [1] (the documentation has a bug, so the reference is for the patch fixing that). [1]: https://lore.kernel.org/xen-devel/20230130090937.31623-1-jgross@suse.com/T/#u Link: https://lkml.kernel.org/r/20230130113036.7087-3-jgross@suse.com Fixes: 868eb122739a ("xen/9pfs: introduce Xen 9pfs transport driver") Signed-off-by: Juergen Gross Reviewed-by: Simon Horman Signed-off-by: Dominique Martinet Signed-off-by: Eric Van Hensbergen --- net/9p/trans_xen.c | 38 +++++++++++++++++++++++--------------- 1 file changed, 23 insertions(+), 15 deletions(-) diff --git a/net/9p/trans_xen.c b/net/9p/trans_xen.c index 6298d77311535..70aa613c4cfe4 100644 --- a/net/9p/trans_xen.c +++ b/net/9p/trans_xen.c @@ -373,12 +373,11 @@ static int xen_9pfs_front_alloc_dataring(struct xenbus_device *dev, return ret; } -static int xen_9pfs_front_probe(struct xenbus_device *dev, - const struct xenbus_device_id *id) +static int xen_9pfs_front_init(struct xenbus_device *dev) { int ret, i; struct xenbus_transaction xbt; - struct xen_9pfs_front_priv *priv = NULL; + struct xen_9pfs_front_priv *priv = dev_get_drvdata(&dev->dev); char *versions, *v; unsigned int max_rings, max_ring_order, len = 0; @@ -406,11 +405,6 @@ static int xen_9pfs_front_probe(struct xenbus_device *dev, if (p9_xen_trans.maxsize > XEN_FLEX_RING_SIZE(max_ring_order)) p9_xen_trans.maxsize = XEN_FLEX_RING_SIZE(max_ring_order) / 2; - priv = kzalloc(sizeof(*priv), GFP_KERNEL); - if (!priv) - return -ENOMEM; - - priv->dev = dev; priv->num_rings = XEN_9PFS_NUM_RINGS; priv->rings = kcalloc(priv->num_rings, sizeof(*priv->rings), GFP_KERNEL); @@ -469,23 +463,35 @@ static int xen_9pfs_front_probe(struct xenbus_device *dev, goto error; } - write_lock(&xen_9pfs_lock); - list_add_tail(&priv->list, &xen_9pfs_devs); - write_unlock(&xen_9pfs_lock); - dev_set_drvdata(&dev->dev, priv); - xenbus_switch_state(dev, XenbusStateInitialised); - return 0; error_xenbus: xenbus_transaction_end(xbt, 1); xenbus_dev_fatal(dev, ret, "writing xenstore"); error: - dev_set_drvdata(&dev->dev, NULL); xen_9pfs_front_free(priv); return ret; } +static int xen_9pfs_front_probe(struct xenbus_device *dev, + const struct xenbus_device_id *id) +{ + struct xen_9pfs_front_priv *priv = NULL; + + priv = kzalloc(sizeof(*priv), GFP_KERNEL); + if (!priv) + return -ENOMEM; + + priv->dev = dev; + dev_set_drvdata(&dev->dev, priv); + + write_lock(&xen_9pfs_lock); + list_add_tail(&priv->list, &xen_9pfs_devs); + write_unlock(&xen_9pfs_lock); + + return 0; +} + static int xen_9pfs_front_resume(struct xenbus_device *dev) { dev_warn(&dev->dev, "suspend/resume unsupported\n"); @@ -504,6 +510,8 @@ static void xen_9pfs_front_changed(struct xenbus_device *dev, break; case XenbusStateInitWait: + if (!xen_9pfs_front_init(dev)) + xenbus_switch_state(dev, XenbusStateInitialised); break; case XenbusStateConnected: From 74a25e6e916cb57dab4267a96fbe8864ed21abdb Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Wed, 4 Jan 2023 10:04:24 +0800 Subject: [PATCH 466/765] 9p/rdma: unmap receive dma buffer in rdma_request()/post_recv() When down_interruptible() or ib_post_send() failed in rdma_request(), receive dma buffer is not unmapped. Add unmap action to error path. Also if ib_post_recv() failed in post_recv(), dma buffer is not unmapped. Add unmap action to error path. Link: https://lkml.kernel.org/r/20230104020424.611926-1-shaozhengchao@huawei.com Fixes: fc79d4b104f0 ("9p: rdma: RDMA Transport Support for 9P") Signed-off-by: Zhengchao Shao Reviewed-by: Leon Romanovsky Signed-off-by: Dominique Martinet Signed-off-by: Eric Van Hensbergen --- net/9p/trans_rdma.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/net/9p/trans_rdma.c b/net/9p/trans_rdma.c index 83f9100d46bff..b84748baf9cbe 100644 --- a/net/9p/trans_rdma.c +++ b/net/9p/trans_rdma.c @@ -385,6 +385,7 @@ post_recv(struct p9_client *client, struct p9_rdma_context *c) struct p9_trans_rdma *rdma = client->trans; struct ib_recv_wr wr; struct ib_sge sge; + int ret; c->busa = ib_dma_map_single(rdma->cm_id->device, c->rc.sdata, client->msize, @@ -402,7 +403,12 @@ post_recv(struct p9_client *client, struct p9_rdma_context *c) wr.wr_cqe = &c->cqe; wr.sg_list = &sge; wr.num_sge = 1; - return ib_post_recv(rdma->qp, &wr, NULL); + + ret = ib_post_recv(rdma->qp, &wr, NULL); + if (ret) + ib_dma_unmap_single(rdma->cm_id->device, c->busa, + client->msize, DMA_FROM_DEVICE); + return ret; error: p9_debug(P9_DEBUG_ERROR, "EIO\n"); @@ -499,7 +505,7 @@ static int rdma_request(struct p9_client *client, struct p9_req_t *req) if (down_interruptible(&rdma->sq_sem)) { err = -EINTR; - goto send_error; + goto dma_unmap; } /* Mark request as `sent' *before* we actually send it, @@ -509,11 +515,14 @@ static int rdma_request(struct p9_client *client, struct p9_req_t *req) WRITE_ONCE(req->status, REQ_STATUS_SENT); err = ib_post_send(rdma->qp, &wr, NULL); if (err) - goto send_error; + goto dma_unmap; /* Success */ return 0; +dma_unmap: + ib_dma_unmap_single(rdma->cm_id->device, c->busa, + c->req->tc.size, DMA_TO_DEVICE); /* Handle errors that happened during or while preparing the send: */ send_error: WRITE_ONCE(req->status, REQ_STATUS_ERROR); From 3866584a1c56a2bbc8c0981deb4476d0b801969e Mon Sep 17 00:00:00 2001 From: Eric Van Hensbergen Date: Sun, 18 Dec 2022 17:57:27 +0000 Subject: [PATCH 467/765] net/9p: fix bug in client create for .L We are supposed to set fid->mode to reflect the flags that were used to open the file. We were actually setting it to the creation mode which is the default perms of the file not the flags the file was opened with. Signed-off-by: Eric Van Hensbergen Reviewed-by: Dominique Martinet --- net/9p/client.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/9p/client.c b/net/9p/client.c index 6c2a768a6ab15..2adcb5e7b0e29 100644 --- a/net/9p/client.c +++ b/net/9p/client.c @@ -1293,7 +1293,7 @@ int p9_client_create_dotl(struct p9_fid *ofid, const char *name, u32 flags, qid->type, qid->path, qid->version, iounit); memmove(&ofid->qid, qid, sizeof(struct p9_qid)); - ofid->mode = mode; + ofid->mode = flags; ofid->iounit = iounit; free_and_error: From 89c58cb395ec0fb58df5475dced1093eaf5896ad Mon Sep 17 00:00:00 2001 From: Eric Van Hensbergen Date: Sun, 18 Dec 2022 21:02:24 +0000 Subject: [PATCH 468/765] fs/9p: fix error reporting in v9fs_dir_release Checking the p9_fid_put value allows us to pass back errors involved if we end up clunking the fid as part of dir_release. This can help with more graceful response to errors in writeback among other things. Signed-off-by: Eric Van Hensbergen Reviewed-by: Dominique Martinet --- fs/9p/vfs_dir.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/fs/9p/vfs_dir.c b/fs/9p/vfs_dir.c index 59b0e8948f787..3d74b04fe0de4 100644 --- a/fs/9p/vfs_dir.c +++ b/fs/9p/vfs_dir.c @@ -197,7 +197,7 @@ static int v9fs_dir_readdir_dotl(struct file *file, struct dir_context *ctx) /** - * v9fs_dir_release - close a directory + * v9fs_dir_release - called on a close of a file or directory * @inode: inode of the directory * @filp: file pointer to a directory * @@ -209,6 +209,7 @@ int v9fs_dir_release(struct inode *inode, struct file *filp) struct p9_fid *fid; __le32 version; loff_t i_size; + int retval = 0; fid = filp->private_data; p9_debug(P9_DEBUG_VFS, "inode: %p filp: %p fid: %d\n", @@ -217,7 +218,7 @@ int v9fs_dir_release(struct inode *inode, struct file *filp) spin_lock(&inode->i_lock); hlist_del(&fid->ilist); spin_unlock(&inode->i_lock); - p9_fid_put(fid); + retval = p9_fid_put(fid); } if ((filp->f_mode & FMODE_WRITE)) { @@ -228,7 +229,7 @@ int v9fs_dir_release(struct inode *inode, struct file *filp) } else { fscache_unuse_cookie(v9fs_inode_cookie(v9inode), NULL, NULL); } - return 0; + return retval; } const struct file_operations v9fs_dir_operations = { From 060a2c92d1b627c86c5c42ca69baf00457c00c5a Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Wed, 22 Feb 2023 17:52:32 +0000 Subject: [PATCH 469/765] arm64: mm: hugetlb: Disable HUGETLB_PAGE_OPTIMIZE_VMEMMAP Revert the HUGETLB_PAGE_FREE_VMEMMAP selection from commit 1e63ac088f20 ("arm64: mm: hugetlb: enable HUGETLB_PAGE_FREE_VMEMMAP for arm64") but keep the flush_dcache_page() compound_head() change as it aligns with the corresponding check in the __sync_icache_dcache() function. The original config option was renamed in commit 47010c040dec ("mm: hugetlb_vmemmap: cleanup CONFIG_HUGETLB_PAGE_FREE_VMEMMAP*") to HUGETLB_PAGE_OPTIMIZE_VMEMMAP and the flush_dcache_page() check was further simplified by commit 2da1c30929a2 ("mm: hugetlb_vmemmap: delete hugetlb_optimize_vmemmap_enabled()"). The reason for the revert is that the generic vmemmap_remap_pte() function changes both the permissions (writeable to read-only) and the output address (pfn) of the vmemmap ptes. This is deemed UNPREDICTABLE by the Arm architecture without a break-before-make sequence (make the PTE invalid, TLBI, write the new valid PTE). However, such sequence is not possible since the vmemmap may be concurrently accessed by the kernel. Disable the optimisation until a better solution is found. Fixes: 1e63ac088f20 ("arm64: mm: hugetlb: enable HUGETLB_PAGE_FREE_VMEMMAP for arm64") Cc: # 5.19.x Cc: Muchun Song Cc: Will Deacon Cc: Anshuman Khandual Link: https://lore.kernel.org/r/Y9pZALdn3pKiJUeQ@arm.com Reviewed-by: Anshuman Khandual Link: https://lore.kernel.org/r/20230222175232.540851-1-catalin.marinas@arm.com Signed-off-by: Catalin Marinas --- arch/arm64/Kconfig | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index 619ab046744a4..71c35178e017d 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -100,7 +100,6 @@ config ARM64 select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT select ARCH_WANT_FRAME_POINTERS select ARCH_WANT_HUGE_PMD_SHARE if ARM64_4K_PAGES || (ARM64_16K_PAGES && !ARM64_VA_BITS_36) - select ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP select ARCH_WANT_LD_ORPHAN_WARN select ARCH_WANTS_NO_INSTR select ARCH_WANTS_THP_SWAP if ARM64_4K_PAGES From 1b561d3949f8478c5403c9752b5533211a757226 Mon Sep 17 00:00:00 2001 From: Sudeep Holla Date: Thu, 23 Feb 2023 13:57:42 +0000 Subject: [PATCH 470/765] arm64: acpi: Fix possible memory leak of ffh_ctxt Allocated 'ffh_ctxt' memory leak is possible if the SMCCC version and conduit checks fail and -EOPNOTSUPP is returned without freeing the allocated memory. Fix the same by moving the allocation after the SMCCC version and conduit checks. Fixes: 1d280ce099db ("arm64: Add architecture specific ACPI FFH Opregion callbacks") Cc: # 6.2.x Cc: Will Deacon Reported-by: kernel test robot Reported-by: Dan Carpenter Suggested-by: Dan Carpenter Link: https://lore.kernel.org/r/202302191417.dAl9NuE8-lkp@intel.com/ Signed-off-by: Sudeep Holla Link: https://lore.kernel.org/r/20230223135742.2952091-1-sudeep.holla@arm.com Signed-off-by: Catalin Marinas --- arch/arm64/kernel/acpi.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/arm64/kernel/acpi.c b/arch/arm64/kernel/acpi.c index 378453faa87e1..dba8fcec7f33d 100644 --- a/arch/arm64/kernel/acpi.c +++ b/arch/arm64/kernel/acpi.c @@ -435,10 +435,6 @@ int acpi_ffh_address_space_arch_setup(void *handler_ctxt, void **region_ctxt) enum arm_smccc_conduit conduit; struct acpi_ffh_data *ffh_ctxt; - ffh_ctxt = kzalloc(sizeof(*ffh_ctxt), GFP_KERNEL); - if (!ffh_ctxt) - return -ENOMEM; - if (arm_smccc_get_version() < ARM_SMCCC_VERSION_1_2) return -EOPNOTSUPP; @@ -448,6 +444,10 @@ int acpi_ffh_address_space_arch_setup(void *handler_ctxt, void **region_ctxt) return -EOPNOTSUPP; } + ffh_ctxt = kzalloc(sizeof(*ffh_ctxt), GFP_KERNEL); + if (!ffh_ctxt) + return -ENOMEM; + if (conduit == SMCCC_CONDUIT_SMC) { ffh_ctxt->invoke_ffh_fn = __arm_smccc_smc; ffh_ctxt->invoke_ffh64_fn = arm_smccc_1_2_smc; From aaf5f0d76b6e1870e3674408de2b13a92a4d4059 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Martin=20Povi=C5=A1er?= Date: Fri, 24 Feb 2023 16:33:00 +0100 Subject: [PATCH 471/765] ASoC: apple: mca: Fix final status read on SERDES reset MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit From within the early trigger we are doing a reset of the SERDES unit, but the final status read is on a bad address. Add the missing SERDES unit offset in calculation of the address. Fixes: 3df5d0d97289 ("ASoC: apple: mca: Start new platform driver") Signed-off-by: Martin Povišer Link: https://lore.kernel.org/r/20230224153302.45365-1-povik+lin@cutebit.org Signed-off-by: Mark Brown --- sound/soc/apple/mca.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/apple/mca.c b/sound/soc/apple/mca.c index 24381c42eb54c..9cceeb2599524 100644 --- a/sound/soc/apple/mca.c +++ b/sound/soc/apple/mca.c @@ -210,7 +210,7 @@ static void mca_fe_early_trigger(struct snd_pcm_substream *substream, int cmd, SERDES_CONF_SOME_RST); readl_relaxed(cl->base + serdes_conf); mca_modify(cl, serdes_conf, SERDES_STATUS_RST, 0); - WARN_ON(readl_relaxed(cl->base + REG_SERDES_STATUS) & + WARN_ON(readl_relaxed(cl->base + serdes_unit + REG_SERDES_STATUS) & SERDES_STATUS_RST); break; default: From d8b3e396088d787771f19fd3b7949e080dc31d6f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Martin=20Povi=C5=A1er?= Date: Fri, 24 Feb 2023 16:33:01 +0100 Subject: [PATCH 472/765] ASoC: apple: mca: Fix SERDES reset sequence MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix the reset sequence of reads and writes that we invoke from within the early trigger. It looks like there never was a SERDES_CONF_SOME_RST bit that should be involved in the reset sequence, and its presence in the driver code is a mistake from earlier. Instead, the reset sequence should go as follows: We should switch the the SERDES unit's SYNC_SEL mux to the value of 7 (so outside the range of 1...6 representing cluster's SYNCGEN units), then raise the RST bit in SERDES_STATUS and wait for it to clear. Properly resetting the SERDES unit fixes frame desynchronization hazard in case of long frames (longer than 4 used slots). The desynchronization manifests itself by rotating the PCM channels. Fixes: 3df5d0d97289 ("ASoC: apple: mca: Start new platform driver") Signed-off-by: Martin Povišer Link: https://lore.kernel.org/r/20230224153302.45365-2-povik+lin@cutebit.org Signed-off-by: Mark Brown --- sound/soc/apple/mca.c | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-) diff --git a/sound/soc/apple/mca.c b/sound/soc/apple/mca.c index 9cceeb2599524..aea08c7b2ee85 100644 --- a/sound/soc/apple/mca.c +++ b/sound/soc/apple/mca.c @@ -101,7 +101,6 @@ #define SERDES_CONF_UNK3 BIT(14) #define SERDES_CONF_NO_DATA_FEEDBACK BIT(15) #define SERDES_CONF_SYNC_SEL GENMASK(18, 16) -#define SERDES_CONF_SOME_RST BIT(19) #define REG_TX_SERDES_BITSTART 0x08 #define REG_RX_SERDES_BITSTART 0x0c #define REG_TX_SERDES_SLOTMASK 0x0c @@ -203,15 +202,24 @@ static void mca_fe_early_trigger(struct snd_pcm_substream *substream, int cmd, case SNDRV_PCM_TRIGGER_START: case SNDRV_PCM_TRIGGER_RESUME: case SNDRV_PCM_TRIGGER_PAUSE_RELEASE: + mca_modify(cl, serdes_conf, SERDES_CONF_SYNC_SEL, + FIELD_PREP(SERDES_CONF_SYNC_SEL, 0)); + mca_modify(cl, serdes_conf, SERDES_CONF_SYNC_SEL, + FIELD_PREP(SERDES_CONF_SYNC_SEL, 7)); mca_modify(cl, serdes_unit + REG_SERDES_STATUS, SERDES_STATUS_EN | SERDES_STATUS_RST, SERDES_STATUS_RST); - mca_modify(cl, serdes_conf, SERDES_CONF_SOME_RST, - SERDES_CONF_SOME_RST); - readl_relaxed(cl->base + serdes_conf); - mca_modify(cl, serdes_conf, SERDES_STATUS_RST, 0); + /* + * Experiments suggest that it takes at most ~1 us + * for the bit to clear, so wait 2 us for good measure. + */ + udelay(2); WARN_ON(readl_relaxed(cl->base + serdes_unit + REG_SERDES_STATUS) & SERDES_STATUS_RST); + mca_modify(cl, serdes_conf, SERDES_CONF_SYNC_SEL, + FIELD_PREP(SERDES_CONF_SYNC_SEL, 0)); + mca_modify(cl, serdes_conf, SERDES_CONF_SYNC_SEL, + FIELD_PREP(SERDES_CONF_SYNC_SEL, cl->no + 1)); break; default: break; From fb1847cc460c127b12720119eae5f438ffc62e85 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Martin=20Povi=C5=A1er?= Date: Fri, 24 Feb 2023 16:33:02 +0100 Subject: [PATCH 473/765] ASoC: apple: mca: Improve handling of unavailable DMA channels MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When we fail to obtain a DMA channel, don't return a blanket -EINVAL, instead return the original error code if there's one. This makes deferring work as it should. Also don't print an error message for -EPROBE_DEFER. Fixes: 4ec8179c212f ("ASoC: apple: mca: Postpone requesting of DMA channels") Signed-off-by: Martin Povišer Link: https://lore.kernel.org/r/20230224153302.45365-3-povik+lin@cutebit.org Signed-off-by: Mark Brown --- sound/soc/apple/mca.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/sound/soc/apple/mca.c b/sound/soc/apple/mca.c index aea08c7b2ee85..64750db9b9639 100644 --- a/sound/soc/apple/mca.c +++ b/sound/soc/apple/mca.c @@ -950,10 +950,17 @@ static int mca_pcm_new(struct snd_soc_component *component, chan = mca_request_dma_channel(cl, i); if (IS_ERR_OR_NULL(chan)) { + mca_pcm_free(component, rtd->pcm); + + if (chan && PTR_ERR(chan) == -EPROBE_DEFER) + return PTR_ERR(chan); + dev_err(component->dev, "unable to obtain DMA channel (stream %d cluster %d): %pe\n", i, cl->no, chan); - mca_pcm_free(component, rtd->pcm); - return -EINVAL; + + if (!chan) + return -EINVAL; + return PTR_ERR(chan); } cl->dma_chans[i] = chan; From 047ee71ae4f412d8819e39e4b08c588fa299cfc2 Mon Sep 17 00:00:00 2001 From: Krishna Yarlagadda Date: Fri, 24 Feb 2023 22:10:34 +0530 Subject: [PATCH 474/765] spi: tegra210-quad: Fix validate combined sequence Check for non dma transfers that do not fit in FIFO has issue and skips combined sequence for Tegra234 & Tegra241 which does not have GPCDMA. Fixes: 1b8342cc4a38 ("spi: tegra210-quad: combined sequence mode") Signed-off-by: Krishna Yarlagadda Link: https://lore.kernel.org/r/20230224164034.56933-1-kyarlagadda@nvidia.com Signed-off-by: Mark Brown --- drivers/spi/spi-tegra210-quad.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/spi/spi-tegra210-quad.c b/drivers/spi/spi-tegra210-quad.c index 258e3b8c9c2cb..9aaca2289c596 100644 --- a/drivers/spi/spi-tegra210-quad.c +++ b/drivers/spi/spi-tegra210-quad.c @@ -1297,7 +1297,7 @@ static bool tegra_qspi_validate_cmb_seq(struct tegra_qspi *tqspi, if (xfer->len > 4 || xfer->len < 3) return false; xfer = list_next_entry(xfer, transfer_list); - if (!tqspi->soc_data->has_dma || xfer->len > (QSPI_FIFO_DEPTH << 2)) + if (!tqspi->soc_data->has_dma && xfer->len > (QSPI_FIFO_DEPTH << 2)) return false; return true; From ea9a78c3a7a44e36fa690e1cc90dc2a758c8eb9a Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Fri, 24 Feb 2023 14:05:09 +0100 Subject: [PATCH 475/765] genirq/msi: Drop dead domain name assignment Since commit d59f6617eef0 ("genirq: Allow fwnode to carry name information only") an IRQ domain is always given a name during allocation (e.g. used for the debugfs entry). Drop the unused fallback name assignment when creating MSI domains. Signed-off-by: Johan Hovold Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/r/20230224130509.27814-1-johan+linaro@kernel.org --- kernel/irq/msi.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/kernel/irq/msi.c b/kernel/irq/msi.c index 13d96495e6d09..efd21b79bf322 100644 --- a/kernel/irq/msi.c +++ b/kernel/irq/msi.c @@ -830,11 +830,8 @@ static struct irq_domain *__msi_create_irq_domain(struct fwnode_handle *fwnode, domain = irq_domain_create_hierarchy(parent, flags | IRQ_DOMAIN_FLAG_MSI, 0, fwnode, &msi_domain_ops, info); - if (domain) { - if (!domain->name && info->chip) - domain->name = info->chip->name; + if (domain) irq_domain_update_bus_token(domain, info->bus_token); - } return domain; } From 977bc87356107fb946fb4ff24f1e4c241b5043ec Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 24 Feb 2023 09:54:57 -0700 Subject: [PATCH 476/765] io_uring/rsrc: always initialize 'folio' to NULL Smatch complains that: smatch warnings: io_uring/rsrc.c:1262 io_sqe_buffer_register() error: uninitialized symbol 'folio'. 'folio' may be used uninitialized, which can happen if we end up with a single page mapped. Ensure that we clear folio to NULL at the top so it's always set. Reported-by: kernel test robot Reported-by: Dan Carpenter Link: https://lore.kernel.org/r/202302241432.YML1CD5C-lkp@intel.com/ Signed-off-by: Jens Axboe --- io_uring/rsrc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c index aab1bc6883e96..056f40946ff68 100644 --- a/io_uring/rsrc.c +++ b/io_uring/rsrc.c @@ -1210,7 +1210,7 @@ static int io_sqe_buffer_register(struct io_ring_ctx *ctx, struct iovec *iov, unsigned long off; size_t size; int ret, nr_pages, i; - struct folio *folio; + struct folio *folio = NULL; *pimu = ctx->dummy_ubuf; if (!iov->iov_base) From 7605c43d67face310b4b87dee1a28bc0c8cd8c0f Mon Sep 17 00:00:00 2001 From: David Lamparter Date: Fri, 24 Feb 2023 16:01:24 +0100 Subject: [PATCH 477/765] io_uring: remove MSG_NOSIGNAL from recvmsg MSG_NOSIGNAL is not applicable for the receiving side, SIGPIPE is generated when trying to write to a "broken pipe". AF_PACKET's packet_recvmsg() does enforce this, giving back EINVAL when MSG_NOSIGNAL is set - making it unuseable in io_uring's recvmsg. Remove MSG_NOSIGNAL from io_recvmsg_prep(). Cc: stable@vger.kernel.org # v5.10+ Signed-off-by: David Lamparter Cc: Eric Dumazet Cc: Jens Axboe Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/r/20230224150123.128346-1-equinox@diac24.net Signed-off-by: Jens Axboe --- io_uring/net.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/io_uring/net.c b/io_uring/net.c index cbd4b725f58c9..b7f190ca528e6 100644 --- a/io_uring/net.c +++ b/io_uring/net.c @@ -567,7 +567,7 @@ int io_recvmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) sr->flags = READ_ONCE(sqe->ioprio); if (sr->flags & ~(RECVMSG_FLAGS)) return -EINVAL; - sr->msg_flags = READ_ONCE(sqe->msg_flags) | MSG_NOSIGNAL; + sr->msg_flags = READ_ONCE(sqe->msg_flags); if (sr->msg_flags & MSG_DONTWAIT) req->flags |= REQ_F_NOWAIT; if (sr->msg_flags & MSG_ERRQUEUE) From 11eb695feb636fa5211067189cad25ac073e7fe5 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 24 Feb 2023 09:59:44 -0700 Subject: [PATCH 478/765] block: clear bio->bi_bdev when putting a bio back in the cache This isn't strictly needed in terms of correctness, but it does allow polling to know if the bio has been put already by a different task and hence avoid polling something that we don't need to. Cc: stable@vger.kernel.org Fixes: be4d234d7aeb ("bio: add allocation cache abstraction") Reviewed-by: Keith Busch Signed-off-by: Jens Axboe --- block/bio.c | 1 + 1 file changed, 1 insertion(+) diff --git a/block/bio.c b/block/bio.c index 2693f34afb7e9..605c400250688 100644 --- a/block/bio.c +++ b/block/bio.c @@ -772,6 +772,7 @@ static inline void bio_put_percpu_cache(struct bio *bio) if ((bio->bi_opf & REQ_POLLED) && !WARN_ON_ONCE(in_interrupt())) { bio->bi_next = cache->free_list; + bio->bi_bdev = NULL; cache->free_list = bio; cache->nr++; } else { From 310726c33ad76cebdee312dbfafc12c1b44bf977 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 24 Feb 2023 10:01:19 -0700 Subject: [PATCH 479/765] block: be a bit more careful in checking for NULL bdev while polling Wei reports a crash with an application using polled IO: PGD 14265e067 P4D 14265e067 PUD 47ec50067 PMD 0 Oops: 0000 [#1] SMP CPU: 0 PID: 21915 Comm: iocore_0 Kdump: loaded Tainted: G S 5.12.0-0_fbk12_clang_7346_g1bb6f2e7058f #1 Hardware name: Wiwynn Delta Lake MP T8/Delta Lake-Class2, BIOS Y3DLM08 04/10/2022 RIP: 0010:bio_poll+0x25/0x200 Code: 0f 1f 44 00 00 0f 1f 44 00 00 55 41 57 41 56 41 55 41 54 53 48 83 ec 28 65 48 8b 04 25 28 00 00 00 48 89 44 24 20 48 8b 47 08 <48> 8b 80 70 02 00 00 4c 8b 70 50 8b 6f 34 31 db 83 fd ff 75 25 65 RSP: 0018:ffffc90005fafdf8 EFLAGS: 00010292 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 74b43cd65dd66600 RDX: 0000000000000003 RSI: ffffc90005fafe78 RDI: ffff8884b614e140 RBP: ffff88849964df78 R08: 0000000000000000 R09: 0000000000000008 R10: 0000000000000000 R11: 0000000000000000 R12: ffff88849964df00 R13: ffffc90005fafe78 R14: ffff888137d3c378 R15: 0000000000000001 FS: 00007fd195000640(0000) GS:ffff88903f400000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000270 CR3: 0000000466121001 CR4: 00000000007706f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: iocb_bio_iopoll+0x1d/0x30 io_do_iopoll+0xac/0x250 __se_sys_io_uring_enter+0x3c5/0x5a0 ? __x64_sys_write+0x89/0xd0 do_syscall_64+0x2d/0x40 entry_SYSCALL_64_after_hwframe+0x44/0xae RIP: 0033:0x94f225d Code: 24 cc 00 00 00 41 8b 84 24 d0 00 00 00 c1 e0 04 83 e0 10 41 09 c2 8b 33 8b 53 04 4c 8b 43 18 4c 63 4b 0c b8 aa 01 00 00 0f 05 <85> c0 0f 88 85 00 00 00 29 03 45 84 f6 0f 84 88 00 00 00 41 f6 c7 RSP: 002b:00007fd194ffcd88 EFLAGS: 00000202 ORIG_RAX: 00000000000001aa RAX: ffffffffffffffda RBX: 00007fd194ffcdc0 RCX: 00000000094f225d RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000007 RBP: 00007fd194ffcdb0 R08: 0000000000000000 R09: 0000000000000008 R10: 0000000000000001 R11: 0000000000000202 R12: 00007fd269d68030 R13: 0000000000000000 R14: 0000000000000001 R15: 0000000000000000 which is due to bio->bi_bdev being NULL. This can happen if we have two tasks doing polled IO, and task B ends up completing IO from task A if they are sharing a poll queue. If task B completes the IO and puts the bio into our cache, then it can allocate that bio again before task A is done polling for it. As that would necessitate a preempt between the two tasks, it's enough to just be a bit more careful in checking for whether or not bio->bi_bdev is NULL. Reported-and-tested-by: Wei Zhang Cc: stable@vger.kernel.org Fixes: be4d234d7aeb ("bio: add allocation cache abstraction") Reviewed-by: Keith Busch Signed-off-by: Jens Axboe --- block/blk-core.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/block/blk-core.c b/block/blk-core.c index 5fb6856745b4e..5d7e2ca752340 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -854,10 +854,16 @@ EXPORT_SYMBOL(submit_bio); */ int bio_poll(struct bio *bio, struct io_comp_batch *iob, unsigned int flags) { - struct request_queue *q = bdev_get_queue(bio->bi_bdev); blk_qc_t cookie = READ_ONCE(bio->bi_cookie); + struct block_device *bdev; + struct request_queue *q; int ret = 0; + bdev = READ_ONCE(bio->bi_bdev); + if (!bdev) + return 0; + + q = bdev_get_queue(bdev); if (cookie == BLK_QC_T_NONE || !test_bit(QUEUE_FLAG_POLL, &q->queue_flags)) return 0; @@ -926,7 +932,7 @@ int iocb_bio_iopoll(struct kiocb *kiocb, struct io_comp_batch *iob, */ rcu_read_lock(); bio = READ_ONCE(kiocb->private); - if (bio && bio->bi_bdev) + if (bio) ret = bio_poll(bio, iob, flags); rcu_read_unlock(); From bfaecf465a058b167abc7c0dfad985f167ff3af1 Mon Sep 17 00:00:00 2001 From: qinyu Date: Thu, 23 Feb 2023 16:10:47 +0800 Subject: [PATCH 480/765] power: supply: fix null pointer check order in __power_supply_register There is an null pointer check order issue here: if we have to check !desc and !desc->name anyway, check it before dereferencing it in pr_warn(). Signed-off-by: qinyu Signed-off-by: Sebastian Reichel --- drivers/power/supply/power_supply_core.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/power/supply/power_supply_core.c b/drivers/power/supply/power_supply_core.c index cc5b2e22b42ac..f3d7c1da299fe 100644 --- a/drivers/power/supply/power_supply_core.c +++ b/drivers/power/supply/power_supply_core.c @@ -1207,13 +1207,13 @@ __power_supply_register(struct device *parent, struct power_supply *psy; int rc; + if (!desc || !desc->name || !desc->properties || !desc->num_properties) + return ERR_PTR(-EINVAL); + if (!parent) pr_warn("%s: Expected proper parent device for '%s'\n", __func__, desc->name); - if (!desc || !desc->name || !desc->properties || !desc->num_properties) - return ERR_PTR(-EINVAL); - if (psy_has_property(desc, POWER_SUPPLY_PROP_USB_TYPE) && (!desc->usb_types || !desc->num_usb_types)) return ERR_PTR(-EINVAL); From f0cf05759497dbc90303adcbdf3d13a54bdaf97b Mon Sep 17 00:00:00 2001 From: Bastien Nocera Date: Mon, 20 Feb 2023 14:10:18 +0100 Subject: [PATCH 481/765] ABI: testing: sysfs-class-power: Document absence of "present" property Document how the absence of the "present" property in the sysfs power_supply class should be handled. Link: https://gitlab.freedesktop.org/upower/upower/-/merge_requests/173 Cc: Mario Limonciello Signed-off-by: Bastien Nocera Signed-off-by: Sebastian Reichel --- Documentation/ABI/testing/sysfs-class-power | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/Documentation/ABI/testing/sysfs-class-power b/Documentation/ABI/testing/sysfs-class-power index e434fc523291d..7c81f0a25a487 100644 --- a/Documentation/ABI/testing/sysfs-class-power +++ b/Documentation/ABI/testing/sysfs-class-power @@ -437,7 +437,8 @@ What: /sys/class/power_supply//present Date: May 2007 Contact: linux-pm@vger.kernel.org Description: - Reports whether a battery is present or not in the system. + Reports whether a battery is present or not in the system. If the + property does not exist, the battery is considered to be present. Access: Read From 13af134bdc6a9dacec4687e57b2ea8d3e08ff04f Mon Sep 17 00:00:00 2001 From: ChiaEn Wu Date: Fri, 17 Feb 2023 17:55:55 +0800 Subject: [PATCH 482/765] dt-bindings: power: supply: Revise Richtek RT9467 compatible name Revise RT9467 compatible name from "richtek,rt9467-charger" to "richtek,rt9467" because it has to match the "compatible name" in the source code. Fixes: e1b4620fb503 ("dt-bindings: power: supply: Add Richtek RT9467 battery charger") Reported-by: Rob Herring Signed-off-by: ChiaEn Wu Reviewed-by: Krzysztof Kozlowski Signed-off-by: Sebastian Reichel --- .../{richtek,rt9467-charger.yaml => richtek,rt9467.yaml} | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) rename Documentation/devicetree/bindings/power/supply/{richtek,rt9467-charger.yaml => richtek,rt9467.yaml} (92%) diff --git a/Documentation/devicetree/bindings/power/supply/richtek,rt9467-charger.yaml b/Documentation/devicetree/bindings/power/supply/richtek,rt9467.yaml similarity index 92% rename from Documentation/devicetree/bindings/power/supply/richtek,rt9467-charger.yaml rename to Documentation/devicetree/bindings/power/supply/richtek,rt9467.yaml index 92c570643d2c4..3723717dc1f65 100644 --- a/Documentation/devicetree/bindings/power/supply/richtek,rt9467-charger.yaml +++ b/Documentation/devicetree/bindings/power/supply/richtek,rt9467.yaml @@ -1,7 +1,7 @@ # SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) %YAML 1.2 --- -$id: http://devicetree.org/schemas/power/supply/richtek,rt9467-charger.yaml# +$id: http://devicetree.org/schemas/power/supply/richtek,rt9467.yaml# $schema: http://devicetree.org/meta-schemas/core.yaml# title: Richtek RT9467 Switching Battery Charger with Power Path Management @@ -25,7 +25,7 @@ description: | properties: compatible: - const: richtek,rt9467-charger + const: richtek,rt9467 reg: maxItems: 1 @@ -65,7 +65,7 @@ examples: #size-cells = <0>; charger@5b { - compatible = "richtek,rt9467-charger"; + compatible = "richtek,rt9467"; reg = <0x5b>; wakeup-source; interrupts-extended = <&gpio_intc 32 IRQ_TYPE_LEVEL_LOW>; From e8b812b3e515742a4faaf2e4722bdc5755b98f56 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Tue, 21 Feb 2023 13:53:51 +0100 Subject: [PATCH 483/765] driver core: bus: Handle early calls to bus_to_subsys() When calling soc_device_match() from early_initcall(), bus_kset is still NULL, causing a crash: Unable to handle kernel NULL pointer dereference at virtual address 0000000000000028 ... Call trace: __lock_acquire+0x530/0x20f0 lock_acquire.part.0+0xc8/0x210 lock_acquire+0x64/0x80 _raw_spin_lock+0x4c/0x60 bus_to_subsys+0x24/0xac bus_for_each_dev+0x30/0xcc soc_device_match+0x4c/0xe0 r8a7795_sysc_init+0x18/0x60 rcar_sysc_pd_init+0xb0/0x33c do_one_initcall+0x128/0x2bc Before, bus_for_each_dev() handled this gracefully by checking that the back-pointer to the private structure was valid. Fix this by adding a NULL check for bus_kset to bus_to_subsys(). Fixes: 83b9148df2c95e23 ("driver core: bus: bus iterator cleanups") Signed-off-by: Geert Uytterhoeven Link: https://lore.kernel.org/r/0a92979f6e790737544638e8a4c19b0564e660a2.1676983596.git.geert+renesas@glider.be Signed-off-by: Greg Kroah-Hartman --- drivers/base/bus.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/base/bus.c b/drivers/base/bus.c index cfe8615d5106f..dd4b82d7510f6 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -62,7 +62,7 @@ static struct subsys_private *bus_to_subsys(const struct bus_type *bus) struct subsys_private *sp = NULL; struct kobject *kobj; - if (!bus) + if (!bus || !bus_kset) return NULL; spin_lock(&bus_kset->list_lock); From 6309872413f14f3d58c13ae4dc85b1a7004b4193 Mon Sep 17 00:00:00 2001 From: Saravana Kannan Date: Fri, 24 Feb 2023 22:41:47 -0800 Subject: [PATCH 484/765] driver core: fw_devlink: Avoid spurious error message fw_devlink can sometimes try to create a device link with the consumer and supplier as the same device. These attempts will fail (correctly), but are harmless. So, avoid printing an error for these cases. Also, add more detail to the error message. Fixes: 3fb16866b51d ("driver core: fw_devlink: Make cycle detection more robust") Reported-by: Vladimir Oltean Reported-by: Dmitry Baryshkov Signed-off-by: Saravana Kannan Link: https://lore.kernel.org/r/20230225064148.274376-1-saravanak@google.com Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/base/core.c b/drivers/base/core.c index e54a10b5dbd71..863eb42359dba 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -2046,9 +2046,9 @@ static int fw_devlink_create_devlink(struct device *con, goto out; } - if (!device_link_add(con, sup_dev, flags)) { - dev_err(con, "Failed to create device link with %s\n", - dev_name(sup_dev)); + if (con != sup_dev && !device_link_add(con, sup_dev, flags)) { + dev_err(con, "Failed to create device link (0x%x) with %s\n", + flags, dev_name(sup_dev)); ret = -EINVAL; } From 0c058fb94ae0e2a68639f4569de1c3abf5df7ad7 Mon Sep 17 00:00:00 2001 From: Saravana Kannan Date: Fri, 24 Feb 2023 22:54:42 -0800 Subject: [PATCH 485/765] driver core: fw_devlink: Print full path and name of fwnode Some of the log messages were printing just the fwnode name. While it's short, it's not always uniquely identifiable in system. So print the full path and name to make debugging easier. Signed-off-by: Saravana Kannan Link: https://lore.kernel.org/r/20230225065443.278284-1-saravanak@google.com Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/base/core.c b/drivers/base/core.c index 863eb42359dba..6878dfcbf0d60 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -98,7 +98,7 @@ static int __fwnode_link_add(struct fwnode_handle *con, list_add(&link->s_hook, &sup->consumers); list_add(&link->c_hook, &con->suppliers); - pr_debug("%pfwP Linked as a fwnode consumer to %pfwP\n", + pr_debug("%pfwf Linked as a fwnode consumer to %pfwf\n", con, sup); return 0; @@ -122,7 +122,7 @@ int fwnode_link_add(struct fwnode_handle *con, struct fwnode_handle *sup) */ static void __fwnode_link_del(struct fwnode_link *link) { - pr_debug("%pfwP Dropping the fwnode link to %pfwP\n", + pr_debug("%pfwf Dropping the fwnode link to %pfwf\n", link->consumer, link->supplier); list_del(&link->s_hook); list_del(&link->c_hook); @@ -1062,7 +1062,7 @@ int device_links_check_suppliers(struct device *dev) if (!dev_is_best_effort(dev)) { fwnode_ret = -EPROBE_DEFER; dev_err_probe(dev, -EPROBE_DEFER, - "wait for supplier %pfwP\n", sup_fw); + "wait for supplier %pfwf\n", sup_fw); } else { fwnode_ret = -EAGAIN; } From fd200632d09dcea9fd67714241155b7635505cd6 Mon Sep 17 00:00:00 2001 From: Jinyang He Date: Sat, 25 Feb 2023 15:52:56 +0800 Subject: [PATCH 486/765] LoongArch: Fix Chinese comma in cpu.h Fix Chinese comma introduced by accident in cpu.h. Signed-off-by: Jinyang He Signed-off-by: Huacai Chen --- arch/loongarch/include/asm/cpu.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/loongarch/include/asm/cpu.h b/arch/loongarch/include/asm/cpu.h index 754f285067913..c3da917594728 100644 --- a/arch/loongarch/include/asm/cpu.h +++ b/arch/loongarch/include/asm/cpu.h @@ -36,7 +36,7 @@ #define PRID_SERIES_LA132 0x8000 /* Loongson 32bit */ #define PRID_SERIES_LA264 0xa000 /* Loongson 64bit, 2-issue */ -#define PRID_SERIES_LA364 0xb000 /* Loongson 64bit，3-issue */ +#define PRID_SERIES_LA364 0xb000 /* Loongson 64bit, 3-issue */ #define PRID_SERIES_LA464 0xc000 /* Loongson 64bit, 4-issue */ #define PRID_SERIES_LA664 0xd000 /* Loongson 64bit, 6-issue */ From bb7a78e343468873bf00b2b181fcfd3c02d8cb56 Mon Sep 17 00:00:00 2001 From: Tiezhu Yang Date: Sat, 25 Feb 2023 15:52:56 +0800 Subject: [PATCH 487/765] LoongArch: Only call get_timer_irq() once in constant_clockevent_init() Under CONFIG_DEBUG_ATOMIC_SLEEP=y and CONFIG_DEBUG_PREEMPT=y, we can see the following messages on LoongArch, this is because using might_sleep() in preemption disable context. [ 0.001127] smp: Bringing up secondary CPUs ... [ 0.001222] Booting CPU#1... [ 0.001244] 64-bit Loongson Processor probed (LA464 Core) [ 0.001247] CPU1 revision is: 0014c012 (Loongson-64bit) [ 0.001250] FPU1 revision is: 00000000 [ 0.001252] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:283 [ 0.001255] in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 0, name: swapper/1 [ 0.001257] preempt_count: 1, expected: 0 [ 0.001258] RCU nest depth: 0, expected: 0 [ 0.001259] Preemption disabled at: [ 0.001261] [<9000000000223800>] arch_dup_task_struct+0x20/0x110 [ 0.001272] CPU: 1 PID: 0 Comm: swapper/1 Not tainted 6.2.0-rc7+ #43 [ 0.001275] Hardware name: Loongson Loongson-3A5000-7A1000-1w-A2101/Loongson-LS3A5000-7A1000-1w-A2101, BIOS vUDK2018-LoongArch-V4.0.05132-beta10 12/13/202 [ 0.001277] Stack : 0072617764726148 0000000000000000 9000000000222f1c 90000001001e0000 [ 0.001286] 90000001001e3be0 90000001001e3be8 0000000000000000 0000000000000000 [ 0.001292] 90000001001e3be8 0000000000000040 90000001001e3cb8 90000001001e3a50 [ 0.001297] 9000000001642000 90000001001e3be8 be694d10ce4139dd 9000000100174500 [ 0.001303] 0000000000000001 0000000000000001 00000000ffffe0a2 0000000000000020 [ 0.001309] 000000000000002f 9000000001354116 00000000056b0000 ffffffffffffffff [ 0.001314] 0000000000000000 0000000000000000 90000000014f6e90 9000000001642000 [ 0.001320] 900000000022b69c 0000000000000001 0000000000000000 9000000001736a90 [ 0.001325] 9000000100038000 0000000000000000 9000000000222f34 0000000000000000 [ 0.001331] 00000000000000b0 0000000000000004 0000000000000000 0000000000070000 [ 0.001337] ... [ 0.001339] Call Trace: [ 0.001342] [<9000000000222f34>] show_stack+0x5c/0x180 [ 0.001346] [<90000000010bdd80>] dump_stack_lvl+0x60/0x88 [ 0.001352] [<9000000000266418>] __might_resched+0x180/0x1cc [ 0.001356] [<90000000010c742c>] mutex_lock+0x20/0x64 [ 0.001359] [<90000000002a8ccc>] irq_find_matching_fwspec+0x48/0x124 [ 0.001364] [<90000000002259c4>] constant_clockevent_init+0x68/0x204 [ 0.001368] [<900000000022acf4>] start_secondary+0x40/0xa8 [ 0.001371] [<90000000010c0124>] smpboot_entry+0x60/0x64 Here are the complete call chains: smpboot_entry() start_secondary() constant_clockevent_init() get_timer_irq() irq_find_matching_fwnode() irq_find_matching_fwspec() mutex_lock() might_sleep() __might_sleep() __might_resched() In order to avoid the above issue, we should break the call chains, using timer_irq_installed variable as check condition to only call get_timer_irq() once in constant_clockevent_init() is a simple and proper way. Signed-off-by: Tiezhu Yang Signed-off-by: Huacai Chen --- arch/loongarch/kernel/time.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/arch/loongarch/kernel/time.c b/arch/loongarch/kernel/time.c index a6576dea590c0..4351f69d99501 100644 --- a/arch/loongarch/kernel/time.c +++ b/arch/loongarch/kernel/time.c @@ -140,16 +140,17 @@ static int get_timer_irq(void) int constant_clockevent_init(void) { - int irq; unsigned int cpu = smp_processor_id(); unsigned long min_delta = 0x600; unsigned long max_delta = (1UL << 48) - 1; struct clock_event_device *cd; - static int timer_irq_installed = 0; + static int irq = 0, timer_irq_installed = 0; - irq = get_timer_irq(); - if (irq < 0) - pr_err("Failed to map irq %d (timer)\n", irq); + if (!timer_irq_installed) { + irq = get_timer_irq(); + if (irq < 0) + pr_err("Failed to map irq %d (timer)\n", irq); + } cd = &per_cpu(constant_clockevent_device, cpu); From 41596803302d83a67a80dc1efef4e51ac46acabb Mon Sep 17 00:00:00 2001 From: Huacai Chen Date: Sat, 25 Feb 2023 15:52:56 +0800 Subject: [PATCH 488/765] LoongArch: Make -mstrict-align configurable Introduce Kconfig option ARCH_STRICT_ALIGN to make -mstrict-align be configurable. Not all LoongArch cores support h/w unaligned access, we can use the -mstrict-align build parameter to prevent unaligned accesses. CPUs with h/w unaligned access support: Loongson-2K2000/2K3000/3A5000/3C5000/3D5000. CPUs without h/w unaligned access support: Loongson-2K500/2K1000. This option is enabled by default to make the kernel be able to run on all LoongArch systems. But you can disable it manually if you want to run kernel only on systems with h/w unaligned access support in order to optimise for performance. Reviewed-by: Arnd Bergmann Signed-off-by: Huacai Chen --- arch/loongarch/Kconfig | 19 +++++++++++++++++++ arch/loongarch/Makefile | 5 +++++ arch/loongarch/kernel/Makefile | 4 +++- arch/loongarch/kernel/traps.c | 9 +++++++-- 4 files changed, 34 insertions(+), 3 deletions(-) diff --git a/arch/loongarch/Kconfig b/arch/loongarch/Kconfig index 9cc8b84f7eb03..0c1c6063cc661 100644 --- a/arch/loongarch/Kconfig +++ b/arch/loongarch/Kconfig @@ -94,6 +94,7 @@ config LOONGARCH select HAVE_DYNAMIC_FTRACE_WITH_ARGS select HAVE_DYNAMIC_FTRACE_WITH_REGS select HAVE_EBPF_JIT + select HAVE_EFFICIENT_UNALIGNED_ACCESS if !ARCH_STRICT_ALIGN select HAVE_EXIT_THREAD select HAVE_FAST_GUP select HAVE_FTRACE_MCOUNT_RECORD @@ -441,6 +442,24 @@ config ARCH_IOREMAP protection support. However, you can enable LoongArch DMW-based ioremap() for better performance. +config ARCH_STRICT_ALIGN + bool "Enable -mstrict-align to prevent unaligned accesses" if EXPERT + default y + help + Not all LoongArch cores support h/w unaligned access, we can use + -mstrict-align build parameter to prevent unaligned accesses. + + CPUs with h/w unaligned access support: + Loongson-2K2000/2K3000/3A5000/3C5000/3D5000. + + CPUs without h/w unaligned access support: + Loongson-2K500/2K1000. + + This option is enabled by default to make the kernel be able to run + on all LoongArch systems. But you can disable it manually if you want + to run kernel only on systems with h/w unaligned access support in + order to optimise for performance. + config KEXEC bool "Kexec system call" select KEXEC_CORE diff --git a/arch/loongarch/Makefile b/arch/loongarch/Makefile index 4402387d27551..6e1c931a8507e 100644 --- a/arch/loongarch/Makefile +++ b/arch/loongarch/Makefile @@ -91,10 +91,15 @@ KBUILD_CPPFLAGS += -DVMLINUX_LOAD_ADDRESS=$(load-y) # instead of .eh_frame so we don't discard them. KBUILD_CFLAGS += -fno-asynchronous-unwind-tables +ifdef CONFIG_ARCH_STRICT_ALIGN # Don't emit unaligned accesses. # Not all LoongArch cores support unaligned access, and as kernel we can't # rely on others to provide emulation for these accesses. KBUILD_CFLAGS += $(call cc-option,-mstrict-align) +else +# Optimise for performance on hardware supports unaligned access. +KBUILD_CFLAGS += $(call cc-option,-mno-strict-align) +endif KBUILD_CFLAGS += -isystem $(shell $(CC) -print-file-name=include) diff --git a/arch/loongarch/kernel/Makefile b/arch/loongarch/kernel/Makefile index c8cfbd562921d..df5dbabfe7a6c 100644 --- a/arch/loongarch/kernel/Makefile +++ b/arch/loongarch/kernel/Makefile @@ -8,13 +8,15 @@ extra-y := vmlinux.lds obj-y += head.o cpu-probe.o cacheinfo.o env.o setup.o entry.o genex.o \ traps.o irq.o idle.o process.o dma.o mem.o io.o reset.o switch.o \ elf.o syscall.o signal.o time.o topology.o inst.o ptrace.o vdso.o \ - alternative.o unaligned.o unwind.o + alternative.o unwind.o obj-$(CONFIG_ACPI) += acpi.o obj-$(CONFIG_EFI) += efi.o obj-$(CONFIG_CPU_HAS_FPU) += fpu.o +obj-$(CONFIG_ARCH_STRICT_ALIGN) += unaligned.o + ifdef CONFIG_FUNCTION_TRACER ifndef CONFIG_DYNAMIC_FTRACE obj-y += mcount.o ftrace.o diff --git a/arch/loongarch/kernel/traps.c b/arch/loongarch/kernel/traps.c index c38a146a973b4..05511203732c3 100644 --- a/arch/loongarch/kernel/traps.c +++ b/arch/loongarch/kernel/traps.c @@ -371,9 +371,14 @@ int no_unaligned_warning __read_mostly = 1; /* Only 1 warning by default */ asmlinkage void noinstr do_ale(struct pt_regs *regs) { - unsigned int *pc; irqentry_state_t state = irqentry_enter(regs); +#ifndef CONFIG_ARCH_STRICT_ALIGN + die_if_kernel("Kernel ale access", regs); + force_sig_fault(SIGBUS, BUS_ADRALN, (void __user *)regs->csr_badvaddr); +#else + unsigned int *pc; + perf_sw_event(PERF_COUNT_SW_ALIGNMENT_FAULTS, 1, regs, regs->csr_badvaddr); /* @@ -397,8 +402,8 @@ asmlinkage void noinstr do_ale(struct pt_regs *regs) sigbus: die_if_kernel("Kernel ale access", regs); force_sig_fault(SIGBUS, BUS_ADRALN, (void __user *)regs->csr_badvaddr); - out: +#endif irqentry_exit(regs, state); } From f733f119e9b31088063652a7ad16963d85cb73dd Mon Sep 17 00:00:00 2001 From: Xi Ruoyao Date: Sat, 25 Feb 2023 15:52:56 +0800 Subject: [PATCH 489/765] LoongArch: Use la.pcrel instead of la.abs when it's trivially possible Let's start to kill la.abs in preparation for the subsequent support of the PIE kernel. BTW, Re-tab the indention in arch/loongarch/kernel/entry.S for alignment. Signed-off-by: Xi Ruoyao Signed-off-by: Huacai Chen --- arch/loongarch/include/asm/stackframe.h | 2 +- arch/loongarch/include/asm/uaccess.h | 1 - arch/loongarch/kernel/entry.S | 90 ++++++++++++------------- arch/loongarch/kernel/head.S | 2 +- arch/loongarch/mm/tlbex.S | 3 +- 5 files changed, 48 insertions(+), 50 deletions(-) diff --git a/arch/loongarch/include/asm/stackframe.h b/arch/loongarch/include/asm/stackframe.h index 4ca953062b5be..7deb043ce3873 100644 --- a/arch/loongarch/include/asm/stackframe.h +++ b/arch/loongarch/include/asm/stackframe.h @@ -90,7 +90,7 @@ .endm .macro set_saved_sp stackp temp temp2 - la.abs \temp, kernelsp + la.pcrel \temp, kernelsp #ifdef CONFIG_SMP LONG_ADD \temp, \temp, u0 #endif diff --git a/arch/loongarch/include/asm/uaccess.h b/arch/loongarch/include/asm/uaccess.h index 255899d4a7c36..0d22991ae430d 100644 --- a/arch/loongarch/include/asm/uaccess.h +++ b/arch/loongarch/include/asm/uaccess.h @@ -22,7 +22,6 @@ extern u64 __ua_limit; #define __UA_ADDR ".dword" -#define __UA_LA "la.abs" #define __UA_LIMIT __ua_limit /* diff --git a/arch/loongarch/kernel/entry.S b/arch/loongarch/kernel/entry.S index d53b631c90227..f14ab6a0e0b05 100644 --- a/arch/loongarch/kernel/entry.S +++ b/arch/loongarch/kernel/entry.S @@ -19,70 +19,70 @@ .cfi_sections .debug_frame .align 5 SYM_FUNC_START(handle_syscall) - csrrd t0, PERCPU_BASE_KS - la.abs t1, kernelsp - add.d t1, t1, t0 - move t2, sp - ld.d sp, t1, 0 + csrrd t0, PERCPU_BASE_KS + la.pcrel t1, kernelsp + add.d t1, t1, t0 + move t2, sp + ld.d sp, t1, 0 - addi.d sp, sp, -PT_SIZE - cfi_st t2, PT_R3 + addi.d sp, sp, -PT_SIZE + cfi_st t2, PT_R3 cfi_rel_offset sp, PT_R3 - st.d zero, sp, PT_R0 - csrrd t2, LOONGARCH_CSR_PRMD - st.d t2, sp, PT_PRMD - csrrd t2, LOONGARCH_CSR_CRMD - st.d t2, sp, PT_CRMD - csrrd t2, LOONGARCH_CSR_EUEN - st.d t2, sp, PT_EUEN - csrrd t2, LOONGARCH_CSR_ECFG - st.d t2, sp, PT_ECFG - csrrd t2, LOONGARCH_CSR_ESTAT - st.d t2, sp, PT_ESTAT - cfi_st ra, PT_R1 - cfi_st a0, PT_R4 - cfi_st a1, PT_R5 - cfi_st a2, PT_R6 - cfi_st a3, PT_R7 - cfi_st a4, PT_R8 - cfi_st a5, PT_R9 - cfi_st a6, PT_R10 - cfi_st a7, PT_R11 - csrrd ra, LOONGARCH_CSR_ERA - st.d ra, sp, PT_ERA + st.d zero, sp, PT_R0 + csrrd t2, LOONGARCH_CSR_PRMD + st.d t2, sp, PT_PRMD + csrrd t2, LOONGARCH_CSR_CRMD + st.d t2, sp, PT_CRMD + csrrd t2, LOONGARCH_CSR_EUEN + st.d t2, sp, PT_EUEN + csrrd t2, LOONGARCH_CSR_ECFG + st.d t2, sp, PT_ECFG + csrrd t2, LOONGARCH_CSR_ESTAT + st.d t2, sp, PT_ESTAT + cfi_st ra, PT_R1 + cfi_st a0, PT_R4 + cfi_st a1, PT_R5 + cfi_st a2, PT_R6 + cfi_st a3, PT_R7 + cfi_st a4, PT_R8 + cfi_st a5, PT_R9 + cfi_st a6, PT_R10 + cfi_st a7, PT_R11 + csrrd ra, LOONGARCH_CSR_ERA + st.d ra, sp, PT_ERA cfi_rel_offset ra, PT_ERA - cfi_st tp, PT_R2 - cfi_st u0, PT_R21 - cfi_st fp, PT_R22 + cfi_st tp, PT_R2 + cfi_st u0, PT_R21 + cfi_st fp, PT_R22 SAVE_STATIC - move u0, t0 - li.d tp, ~_THREAD_MASK - and tp, tp, sp + move u0, t0 + li.d tp, ~_THREAD_MASK + and tp, tp, sp - move a0, sp - bl do_syscall + move a0, sp + bl do_syscall RESTORE_ALL_AND_RET SYM_FUNC_END(handle_syscall) SYM_CODE_START(ret_from_fork) - bl schedule_tail # a0 = struct task_struct *prev - move a0, sp - bl syscall_exit_to_user_mode + bl schedule_tail # a0 = struct task_struct *prev + move a0, sp + bl syscall_exit_to_user_mode RESTORE_STATIC RESTORE_SOME RESTORE_SP_AND_RET SYM_CODE_END(ret_from_fork) SYM_CODE_START(ret_from_kernel_thread) - bl schedule_tail # a0 = struct task_struct *prev - move a0, s1 - jirl ra, s0, 0 - move a0, sp - bl syscall_exit_to_user_mode + bl schedule_tail # a0 = struct task_struct *prev + move a0, s1 + jirl ra, s0, 0 + move a0, sp + bl syscall_exit_to_user_mode RESTORE_STATIC RESTORE_SOME RESTORE_SP_AND_RET diff --git a/arch/loongarch/kernel/head.S b/arch/loongarch/kernel/head.S index 57bada6b4e931..aa6181714ec31 100644 --- a/arch/loongarch/kernel/head.S +++ b/arch/loongarch/kernel/head.S @@ -117,7 +117,7 @@ SYM_CODE_START(smpboot_entry) li.w t0, 0x00 # FPE=0, SXE=0, ASXE=0, BTE=0 csrwr t0, LOONGARCH_CSR_EUEN - la.abs t0, cpuboot_data + la.pcrel t0, cpuboot_data ld.d sp, t0, CPU_BOOT_STACK ld.d tp, t0, CPU_BOOT_TINFO diff --git a/arch/loongarch/mm/tlbex.S b/arch/loongarch/mm/tlbex.S index 58781c6e4191a..3dd2a9615cd90 100644 --- a/arch/loongarch/mm/tlbex.S +++ b/arch/loongarch/mm/tlbex.S @@ -24,8 +24,7 @@ move a0, sp REG_S a2, sp, PT_BVADDR li.w a1, \write - la.abs t0, do_page_fault - jirl ra, t0, 0 + bl do_page_fault RESTORE_ALL_AND_RET SYM_FUNC_END(tlb_do_page_fault_\write) .endm From 8cbd5ebfe241699288cb152081854414f6265718 Mon Sep 17 00:00:00 2001 From: Youling Tang Date: Sat, 25 Feb 2023 15:52:56 +0800 Subject: [PATCH 490/765] LoongArch: Add JUMP_VIRT_ADDR macro implementation to avoid using la.abs Add JUMP_VIRT_ADDR macro implementation to avoid using la.abs directly. This is a preparation for subsequent patches. Signed-off-by: Youling Tang Signed-off-by: Huacai Chen --- arch/loongarch/include/asm/stackframe.h | 9 +++++++++ arch/loongarch/kernel/head.S | 12 ++++-------- arch/loongarch/power/suspend_asm.S | 5 ++--- 3 files changed, 15 insertions(+), 11 deletions(-) diff --git a/arch/loongarch/include/asm/stackframe.h b/arch/loongarch/include/asm/stackframe.h index 7deb043ce3873..0274e704c13bf 100644 --- a/arch/loongarch/include/asm/stackframe.h +++ b/arch/loongarch/include/asm/stackframe.h @@ -7,6 +7,7 @@ #include +#include #include #include #include @@ -36,6 +37,14 @@ cfi_restore \reg \offset \docfi .endm +/* Jump to the runtime virtual address. */ + .macro JUMP_VIRT_ADDR temp1 temp2 + li.d \temp1, CACHE_BASE + pcaddi \temp2, 0 + or \temp1, \temp1, \temp2 + jirl zero, \temp1, 0xc + .endm + .macro BACKUP_T0T1 csrwr t0, EXCEPTION_KS0 csrwr t1, EXCEPTION_KS1 diff --git a/arch/loongarch/kernel/head.S b/arch/loongarch/kernel/head.S index aa6181714ec31..d2ac26b5b22bd 100644 --- a/arch/loongarch/kernel/head.S +++ b/arch/loongarch/kernel/head.S @@ -50,11 +50,8 @@ SYM_CODE_START(kernel_entry) # kernel entry point li.d t0, CSR_DMW1_INIT # CA, PLV0, 0x9000 xxxx xxxx xxxx csrwr t0, LOONGARCH_CSR_DMWIN1 - /* We might not get launched at the address the kernel is linked to, - so we jump there. */ - la.abs t0, 0f - jr t0 -0: + JUMP_VIRT_ADDR t0, t1 + /* Enable PG */ li.w t0, 0xb0 # PLV=0, IE=0, PG=1 csrwr t0, LOONGARCH_CSR_CRMD @@ -106,9 +103,8 @@ SYM_CODE_START(smpboot_entry) li.d t0, CSR_DMW1_INIT # CA, PLV0 csrwr t0, LOONGARCH_CSR_DMWIN1 - la.abs t0, 0f - jr t0 -0: + JUMP_VIRT_ADDR t0, t1 + /* Enable PG */ li.w t0, 0xb0 # PLV=0, IE=0, PG=1 csrwr t0, LOONGARCH_CSR_CRMD diff --git a/arch/loongarch/power/suspend_asm.S b/arch/loongarch/power/suspend_asm.S index eb2675642f9f4..90da899c06a19 100644 --- a/arch/loongarch/power/suspend_asm.S +++ b/arch/loongarch/power/suspend_asm.S @@ -78,9 +78,8 @@ SYM_INNER_LABEL(loongarch_wakeup_start, SYM_L_GLOBAL) li.d t0, CSR_DMW1_INIT # CA, PLV0 csrwr t0, LOONGARCH_CSR_DMWIN1 - la.abs t0, 0f - jr t0 -0: + JUMP_VIRT_ADDR t0, t1 + la.pcrel t0, acpi_saved_sp ld.d sp, t0, 0 SETUP_WAKEUP From 396233c650084cb957eb6d87dd4abd3e56a7d499 Mon Sep 17 00:00:00 2001 From: Youling Tang Date: Sat, 25 Feb 2023 15:52:56 +0800 Subject: [PATCH 491/765] LoongArch: Add la_abs macro implementation Use the "la_abs macro" instead of the "la.abs pseudo instruction" to prepare for the subsequent PIE kernel. When PIE is not enabled, la_abs is equivalent to la.abs. Signed-off-by: Youling Tang Signed-off-by: Huacai Chen --- arch/loongarch/include/asm/asmmacro.h | 4 ++++ arch/loongarch/include/asm/stackframe.h | 2 +- arch/loongarch/kernel/genex.S | 8 ++++---- arch/loongarch/mm/tlbex.S | 14 +++++++------- 4 files changed, 16 insertions(+), 12 deletions(-) diff --git a/arch/loongarch/include/asm/asmmacro.h b/arch/loongarch/include/asm/asmmacro.h index be037a40580d4..cdc9935d5554c 100644 --- a/arch/loongarch/include/asm/asmmacro.h +++ b/arch/loongarch/include/asm/asmmacro.h @@ -274,4 +274,8 @@ nor \dst, \src, zero .endm +.macro la_abs reg, sym + la.abs \reg, \sym +.endm + #endif /* _ASM_ASMMACRO_H */ diff --git a/arch/loongarch/include/asm/stackframe.h b/arch/loongarch/include/asm/stackframe.h index 0274e704c13bf..7df80e6ae9d2c 100644 --- a/arch/loongarch/include/asm/stackframe.h +++ b/arch/loongarch/include/asm/stackframe.h @@ -86,7 +86,7 @@ * new value in sp. */ .macro get_saved_sp docfi=0 - la.abs t1, kernelsp + la_abs t1, kernelsp #ifdef CONFIG_SMP csrrd t0, PERCPU_BASE_KS LONG_ADD t1, t1, t0 diff --git a/arch/loongarch/kernel/genex.S b/arch/loongarch/kernel/genex.S index 7e5c293ed89f7..44ff1ff642601 100644 --- a/arch/loongarch/kernel/genex.S +++ b/arch/loongarch/kernel/genex.S @@ -34,7 +34,7 @@ SYM_FUNC_END(__arch_cpu_idle) SYM_FUNC_START(handle_vint) BACKUP_T0T1 SAVE_ALL - la.abs t1, __arch_cpu_idle + la_abs t1, __arch_cpu_idle LONG_L t0, sp, PT_ERA /* 32 byte rollback region */ ori t0, t0, 0x1f @@ -43,7 +43,7 @@ SYM_FUNC_START(handle_vint) LONG_S t0, sp, PT_ERA 1: move a0, sp move a1, sp - la.abs t0, do_vint + la_abs t0, do_vint jirl ra, t0, 0 RESTORE_ALL_AND_RET SYM_FUNC_END(handle_vint) @@ -72,7 +72,7 @@ SYM_FUNC_END(except_vec_cex) SAVE_ALL build_prep_\prep move a0, sp - la.abs t0, do_\handler + la_abs t0, do_\handler jirl ra, t0, 0 668: RESTORE_ALL_AND_RET @@ -93,6 +93,6 @@ SYM_FUNC_END(except_vec_cex) BUILD_HANDLER reserved reserved none /* others */ SYM_FUNC_START(handle_sys) - la.abs t0, handle_syscall + la_abs t0, handle_syscall jr t0 SYM_FUNC_END(handle_sys) diff --git a/arch/loongarch/mm/tlbex.S b/arch/loongarch/mm/tlbex.S index 3dd2a9615cd90..244e2f5aeee56 100644 --- a/arch/loongarch/mm/tlbex.S +++ b/arch/loongarch/mm/tlbex.S @@ -39,7 +39,7 @@ SYM_FUNC_START(handle_tlb_protect) move a1, zero csrrd a2, LOONGARCH_CSR_BADV REG_S a2, sp, PT_BVADDR - la.abs t0, do_page_fault + la_abs t0, do_page_fault jirl ra, t0, 0 RESTORE_ALL_AND_RET SYM_FUNC_END(handle_tlb_protect) @@ -115,7 +115,7 @@ smp_pgtable_change_load: #ifdef CONFIG_64BIT vmalloc_load: - la.abs t1, swapper_pg_dir + la_abs t1, swapper_pg_dir b vmalloc_done_load #endif @@ -186,7 +186,7 @@ tlb_huge_update_load: nopage_tlb_load: dbar 0 csrrd ra, EXCEPTION_KS2 - la.abs t0, tlb_do_page_fault_0 + la_abs t0, tlb_do_page_fault_0 jr t0 SYM_FUNC_END(handle_tlb_load) @@ -262,7 +262,7 @@ smp_pgtable_change_store: #ifdef CONFIG_64BIT vmalloc_store: - la.abs t1, swapper_pg_dir + la_abs t1, swapper_pg_dir b vmalloc_done_store #endif @@ -335,7 +335,7 @@ tlb_huge_update_store: nopage_tlb_store: dbar 0 csrrd ra, EXCEPTION_KS2 - la.abs t0, tlb_do_page_fault_1 + la_abs t0, tlb_do_page_fault_1 jr t0 SYM_FUNC_END(handle_tlb_store) @@ -410,7 +410,7 @@ smp_pgtable_change_modify: #ifdef CONFIG_64BIT vmalloc_modify: - la.abs t1, swapper_pg_dir + la_abs t1, swapper_pg_dir b vmalloc_done_modify #endif @@ -482,7 +482,7 @@ tlb_huge_update_modify: nopage_tlb_modify: dbar 0 csrrd ra, EXCEPTION_KS2 - la.abs t0, tlb_do_page_fault_1 + la_abs t0, tlb_do_page_fault_1 jr t0 SYM_FUNC_END(handle_tlb_modify) From d8da19fbdedd5852592fbba18a7348e3f09500e6 Mon Sep 17 00:00:00 2001 From: Youling Tang Date: Sat, 25 Feb 2023 15:52:56 +0800 Subject: [PATCH 492/765] LoongArch: Add support for kernel relocation This config allows to compile kernel as PIE and to relocate it at any virtual address at runtime: this paves the way to KASLR. Runtime relocation is possible since relocation metadata are embedded into the kernel. Signed-off-by: Youling Tang Signed-off-by: Xi Ruoyao # Use arch_initcall Signed-off-by: Jinyang He # Provide la_abs relocation code Signed-off-by: Huacai Chen --- arch/loongarch/Kconfig | 8 ++ arch/loongarch/Makefile | 5 ++ arch/loongarch/include/asm/asmmacro.h | 13 ++++ arch/loongarch/include/asm/setup.h | 16 ++++ arch/loongarch/kernel/Makefile | 2 + arch/loongarch/kernel/head.S | 4 + arch/loongarch/kernel/relocate.c | 107 ++++++++++++++++++++++++++ arch/loongarch/kernel/vmlinux.lds.S | 20 ++++- 8 files changed, 173 insertions(+), 2 deletions(-) create mode 100644 arch/loongarch/kernel/relocate.c diff --git a/arch/loongarch/Kconfig b/arch/loongarch/Kconfig index 0c1c6063cc661..32ab90dd76e5e 100644 --- a/arch/loongarch/Kconfig +++ b/arch/loongarch/Kconfig @@ -493,6 +493,14 @@ config PHYSICAL_START specified in the "crashkernel=YM@XM" command line boot parameter passed to the panic-ed kernel). +config RELOCATABLE + bool "Relocatable kernel" + help + This builds the kernel as a Position Independent Executable (PIE), + which retains all relocation metadata required, so as to relocate + the kernel binary at runtime to a different virtual address from + its link address. + config SECCOMP bool "Enable seccomp to safely compute untrusted bytecode" depends on PROC_FS diff --git a/arch/loongarch/Makefile b/arch/loongarch/Makefile index 6e1c931a8507e..768592998b391 100644 --- a/arch/loongarch/Makefile +++ b/arch/loongarch/Makefile @@ -71,6 +71,11 @@ KBUILD_AFLAGS_MODULE += -Wa,-mla-global-with-abs KBUILD_CFLAGS_MODULE += -fplt -Wa,-mla-global-with-abs,-mla-local-with-abs endif +ifeq ($(CONFIG_RELOCATABLE),y) +KBUILD_CFLAGS_KERNEL += -fPIE +LDFLAGS_vmlinux += -static -pie --no-dynamic-linker -z notext +endif + cflags-y += -ffreestanding cflags-y += $(call cc-option, -mno-check-zero-division) diff --git a/arch/loongarch/include/asm/asmmacro.h b/arch/loongarch/include/asm/asmmacro.h index cdc9935d5554c..c51a1b43acb44 100644 --- a/arch/loongarch/include/asm/asmmacro.h +++ b/arch/loongarch/include/asm/asmmacro.h @@ -275,7 +275,20 @@ .endm .macro la_abs reg, sym +#ifndef CONFIG_RELOCATABLE la.abs \reg, \sym +#else + 766: + lu12i.w \reg, 0 + ori \reg, \reg, 0 + lu32i.d \reg, 0 + lu52i.d \reg, \reg, 0 + .pushsection ".la_abs", "aw", %progbits + 768: + .dword 768b-766b + .dword \sym + .popsection +#endif .endm #endif /* _ASM_ASMMACRO_H */ diff --git a/arch/loongarch/include/asm/setup.h b/arch/loongarch/include/asm/setup.h index 72ead58039f3e..27d968655b4b8 100644 --- a/arch/loongarch/include/asm/setup.h +++ b/arch/loongarch/include/asm/setup.h @@ -21,4 +21,20 @@ extern void per_cpu_trap_init(int cpu); extern void set_handler(unsigned long offset, void *addr, unsigned long len); extern void set_merr_handler(unsigned long offset, void *addr, unsigned long len); +#ifdef CONFIG_RELOCATABLE + +struct rela_la_abs { + long offset; + long symvalue; +}; + +extern long __la_abs_begin; +extern long __la_abs_end; +extern long __rela_dyn_begin; +extern long __rela_dyn_end; + +extern void __init relocate_kernel(void); + +#endif + #endif /* __SETUP_H */ diff --git a/arch/loongarch/kernel/Makefile b/arch/loongarch/kernel/Makefile index df5dbabfe7a6c..94d33dbd4912e 100644 --- a/arch/loongarch/kernel/Makefile +++ b/arch/loongarch/kernel/Makefile @@ -41,6 +41,8 @@ obj-$(CONFIG_NUMA) += numa.o obj-$(CONFIG_MAGIC_SYSRQ) += sysrq.o +obj-$(CONFIG_RELOCATABLE) += relocate.o + obj-$(CONFIG_KEXEC) += machine_kexec.o relocate_kernel.o obj-$(CONFIG_CRASH_DUMP) += crash_dump.o diff --git a/arch/loongarch/kernel/head.S b/arch/loongarch/kernel/head.S index d2ac26b5b22bd..c5c3ec2b819a9 100644 --- a/arch/loongarch/kernel/head.S +++ b/arch/loongarch/kernel/head.S @@ -86,6 +86,10 @@ SYM_CODE_START(kernel_entry) # kernel entry point PTR_ADD sp, sp, tp set_saved_sp sp, t0, t1 +#ifdef CONFIG_RELOCATABLE + bl relocate_kernel +#endif + bl start_kernel ASM_BUG() diff --git a/arch/loongarch/kernel/relocate.c b/arch/loongarch/kernel/relocate.c new file mode 100644 index 0000000000000..879c40372fbaf --- /dev/null +++ b/arch/loongarch/kernel/relocate.c @@ -0,0 +1,107 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Support for Kernel relocation at boot time + * + * Copyright (C) 2023 Loongson Technology Corporation Limited + */ + +#include +#include +#include +#include +#include +#include +#include + +#define RELOCATED(x) ((void *)((long)x + reloc_offset)) + +static unsigned long reloc_offset; + +static inline void __init relocate_relative(void) +{ + Elf64_Rela *rela, *rela_end; + rela = (Elf64_Rela *)&__rela_dyn_begin; + rela_end = (Elf64_Rela *)&__rela_dyn_end; + + for ( ; rela < rela_end; rela++) { + Elf64_Addr addr = rela->r_offset; + Elf64_Addr relocated_addr = rela->r_addend; + + if (rela->r_info != R_LARCH_RELATIVE) + continue; + + if (relocated_addr >= VMLINUX_LOAD_ADDRESS) + relocated_addr = (Elf64_Addr)RELOCATED(relocated_addr); + + *(Elf64_Addr *)RELOCATED(addr) = relocated_addr; + } +} + +static inline void __init relocate_absolute(void) +{ + void *begin, *end; + struct rela_la_abs *p; + + begin = &__la_abs_begin; + end = &__la_abs_end; + + for (p = begin; (void *)p < end; p++) { + long v = p->symvalue; + uint32_t lu12iw, ori, lu32id, lu52id; + union loongarch_instruction *insn = (void *)p - p->offset; + + lu12iw = (v >> 12) & 0xfffff; + ori = v & 0xfff; + lu32id = (v >> 32) & 0xfffff; + lu52id = v >> 52; + + insn[0].reg1i20_format.immediate = lu12iw; + insn[1].reg2i12_format.immediate = ori; + insn[2].reg1i20_format.immediate = lu32id; + insn[3].reg2i12_format.immediate = lu52id; + } +} + +void __init relocate_kernel(void) +{ + reloc_offset = (unsigned long)_text - VMLINUX_LOAD_ADDRESS; + + if (reloc_offset) + relocate_relative(); + + relocate_absolute(); +} + +/* + * Show relocation information on panic. + */ +static void show_kernel_relocation(const char *level) +{ + if (reloc_offset > 0) { + printk(level); + pr_cont("Kernel relocated by 0x%lx\n", reloc_offset); + pr_cont(" .text @ 0x%px\n", _text); + pr_cont(" .data @ 0x%px\n", _sdata); + pr_cont(" .bss @ 0x%px\n", __bss_start); + } +} + +static int kernel_location_notifier_fn(struct notifier_block *self, + unsigned long v, void *p) +{ + show_kernel_relocation(KERN_EMERG); + return NOTIFY_DONE; +} + +static struct notifier_block kernel_location_notifier = { + .notifier_call = kernel_location_notifier_fn +}; + +static int __init register_kernel_offset_dumper(void) +{ + atomic_notifier_chain_register(&panic_notifier_list, + &kernel_location_notifier); + return 0; +} + +arch_initcall(register_kernel_offset_dumper); diff --git a/arch/loongarch/kernel/vmlinux.lds.S b/arch/loongarch/kernel/vmlinux.lds.S index 733b16e8d55dd..ad2ce1a0908e4 100644 --- a/arch/loongarch/kernel/vmlinux.lds.S +++ b/arch/loongarch/kernel/vmlinux.lds.S @@ -66,10 +66,21 @@ SECTIONS __alt_instructions_end = .; } +#ifdef CONFIG_RELOCATABLE + . = ALIGN(8); + .la_abs : AT(ADDR(.la_abs) - LOAD_OFFSET) { + __la_abs_begin = .; + *(.la_abs) + __la_abs_end = .; + } +#endif + .got : ALIGN(16) { *(.got) } .plt : ALIGN(16) { *(.plt) } .got.plt : ALIGN(16) { *(.got.plt) } + .data.rel : { *(.data.rel*) } + . = ALIGN(PECOFF_SEGMENT_ALIGN); __init_begin = .; __inittext_begin = .; @@ -93,8 +104,6 @@ SECTIONS PERCPU_SECTION(1 << CONFIG_L1_CACHE_SHIFT) #endif - .rela.dyn : ALIGN(8) { *(.rela.dyn) *(.rela*) } - .init.bss : { *(.init.bss) } @@ -107,6 +116,12 @@ SECTIONS RO_DATA(4096) RW_DATA(1 << CONFIG_L1_CACHE_SHIFT, PAGE_SIZE, THREAD_SIZE) + .rela.dyn : ALIGN(8) { + __rela_dyn_begin = .; + *(.rela.dyn) *(.rela*) + __rela_dyn_end = .; + } + .sdata : { *(.sdata) } @@ -133,6 +148,7 @@ SECTIONS DISCARDS /DISCARD/ : { + *(.dynamic .dynsym .dynstr .hash .gnu.hash) *(.gnu.attributes) *(.options) *(.eh_frame) From e5f02b51fa0cb785e352e77271a65e96051b789b Mon Sep 17 00:00:00 2001 From: Youling Tang Date: Sat, 25 Feb 2023 15:52:56 +0800 Subject: [PATCH 493/765] LoongArch: Add support for kernel address space layout randomization (KASLR) This patch adds support for relocating the kernel to a random address. Entropy is derived from the banner, which will change every build and random_get_entropy() which should provide additional runtime entropy. The kernel is relocated by up to RANDOMIZE_BASE_MAX_OFFSET bytes from its link address. Because relocation happens so early during the kernel booting, the amount of physical memory has not yet been determined. This means the only way to limit relocation within the available memory is via Kconfig. So we limit the maximum value of RANDOMIZE_BASE_MAX_OFFSET to 256M (0x10000000) because our memory layout has many holes. Signed-off-by: Youling Tang Signed-off-by: Xi Ruoyao # Fix compiler warnings Signed-off-by: Huacai Chen --- arch/loongarch/Kconfig | 25 +++++ arch/loongarch/include/asm/setup.h | 2 +- arch/loongarch/kernel/head.S | 13 +++ arch/loongarch/kernel/relocate.c | 145 ++++++++++++++++++++++++++++- 4 files changed, 179 insertions(+), 6 deletions(-) diff --git a/arch/loongarch/Kconfig b/arch/loongarch/Kconfig index 32ab90dd76e5e..2fc18a3c565e8 100644 --- a/arch/loongarch/Kconfig +++ b/arch/loongarch/Kconfig @@ -501,6 +501,31 @@ config RELOCATABLE the kernel binary at runtime to a different virtual address from its link address. +config RANDOMIZE_BASE + bool "Randomize the address of the kernel (KASLR)" + depends on RELOCATABLE + help + Randomizes the physical and virtual address at which the + kernel image is loaded, as a security feature that + deters exploit attempts relying on knowledge of the location + of kernel internals. + + The kernel will be offset by up to RANDOMIZE_BASE_MAX_OFFSET. + + If unsure, say N. + +config RANDOMIZE_BASE_MAX_OFFSET + hex "Maximum KASLR offset" if EXPERT + depends on RANDOMIZE_BASE + range 0x0 0x10000000 + default "0x01000000" + help + When KASLR is active, this provides the maximum offset that will + be applied to the kernel image. It should be set according to the + amount of physical RAM available in the target system. + + This is limited by the size of the lower address memory, 256MB. + config SECCOMP bool "Enable seccomp to safely compute untrusted bytecode" depends on PROC_FS diff --git a/arch/loongarch/include/asm/setup.h b/arch/loongarch/include/asm/setup.h index 27d968655b4b8..be05c0e706a2e 100644 --- a/arch/loongarch/include/asm/setup.h +++ b/arch/loongarch/include/asm/setup.h @@ -33,7 +33,7 @@ extern long __la_abs_end; extern long __rela_dyn_begin; extern long __rela_dyn_end; -extern void __init relocate_kernel(void); +extern void * __init relocate_kernel(void); #endif diff --git a/arch/loongarch/kernel/head.S b/arch/loongarch/kernel/head.S index c5c3ec2b819a9..1d35becc01eef 100644 --- a/arch/loongarch/kernel/head.S +++ b/arch/loongarch/kernel/head.S @@ -87,7 +87,20 @@ SYM_CODE_START(kernel_entry) # kernel entry point set_saved_sp sp, t0, t1 #ifdef CONFIG_RELOCATABLE + bl relocate_kernel + +#ifdef CONFIG_RANDOMIZE_BASE + /* Repoint the sp into the new kernel */ + PTR_LI sp, (_THREAD_SIZE - PT_SIZE) + PTR_ADD sp, sp, tp + set_saved_sp sp, t0, t1 +#endif + + /* relocate_kernel() returns the new kernel entry point */ + jr a0 + ASM_BUG() + #endif bl start_kernel diff --git a/arch/loongarch/kernel/relocate.c b/arch/loongarch/kernel/relocate.c index 879c40372fbaf..01f94d1e3edf6 100644 --- a/arch/loongarch/kernel/relocate.c +++ b/arch/loongarch/kernel/relocate.c @@ -9,11 +9,15 @@ #include #include #include +#include +#include +#include #include #include #include #define RELOCATED(x) ((void *)((long)x + reloc_offset)) +#define RELOCATED_KASLR(x) ((void *)((long)x + random_offset)) static unsigned long reloc_offset; @@ -37,13 +41,13 @@ static inline void __init relocate_relative(void) } } -static inline void __init relocate_absolute(void) +static inline void __init relocate_absolute(long random_offset) { void *begin, *end; struct rela_la_abs *p; - begin = &__la_abs_begin; - end = &__la_abs_end; + begin = RELOCATED_KASLR(&__la_abs_begin); + end = RELOCATED_KASLR(&__la_abs_end); for (p = begin; (void *)p < end; p++) { long v = p->symvalue; @@ -62,14 +66,145 @@ static inline void __init relocate_absolute(void) } } -void __init relocate_kernel(void) +#ifdef CONFIG_RANDOMIZE_BASE +static inline __init unsigned long rotate_xor(unsigned long hash, + const void *area, size_t size) { + size_t i, diff; + const typeof(hash) *ptr = PTR_ALIGN(area, sizeof(hash)); + + diff = (void *)ptr - area; + if (size < diff + sizeof(hash)) + return hash; + + size = ALIGN_DOWN(size - diff, sizeof(hash)); + + for (i = 0; i < size / sizeof(hash); i++) { + /* Rotate by odd number of bits and XOR. */ + hash = (hash << ((sizeof(hash) * 8) - 7)) | (hash >> 7); + hash ^= ptr[i]; + } + + return hash; +} + +static inline __init unsigned long get_random_boot(void) +{ + unsigned long hash = 0; + unsigned long entropy = random_get_entropy(); + + /* Attempt to create a simple but unpredictable starting entropy. */ + hash = rotate_xor(hash, linux_banner, strlen(linux_banner)); + + /* Add in any runtime entropy we can get */ + hash = rotate_xor(hash, &entropy, sizeof(entropy)); + + return hash; +} + +static inline __init bool kaslr_disabled(void) +{ + char *str; + const char *builtin_cmdline = CONFIG_CMDLINE; + + str = strstr(builtin_cmdline, "nokaslr"); + if (str == builtin_cmdline || (str > builtin_cmdline && *(str - 1) == ' ')) + return true; + + str = strstr(boot_command_line, "nokaslr"); + if (str == boot_command_line || (str > boot_command_line && *(str - 1) == ' ')) + return true; + + return false; +} + +/* Choose a new address for the kernel */ +static inline void __init *determine_relocation_address(void) +{ + unsigned long kernel_length; + unsigned long random_offset; + void *destination = _text; + + if (kaslr_disabled()) + return destination; + + kernel_length = (long)_end - (long)_text; + + random_offset = get_random_boot() << 16; + random_offset &= (CONFIG_RANDOMIZE_BASE_MAX_OFFSET - 1); + if (random_offset < kernel_length) + random_offset += ALIGN(kernel_length, 0xffff); + + return RELOCATED_KASLR(destination); +} + +static inline int __init relocation_addr_valid(void *location_new) +{ + if ((unsigned long)location_new & 0x00000ffff) + return 0; /* Inappropriately aligned new location */ + + if ((unsigned long)location_new < (unsigned long)_end) + return 0; /* New location overlaps original kernel */ + + return 1; +} +#endif + +static inline void __init update_reloc_offset(unsigned long *addr, long random_offset) +{ + unsigned long *new_addr = (unsigned long *)RELOCATED_KASLR(addr); + + *new_addr = (unsigned long)reloc_offset; +} + +void * __init relocate_kernel(void) +{ + unsigned long kernel_length; + unsigned long random_offset = 0; + void *location_new = _text; /* Default to original kernel start */ + void *kernel_entry = start_kernel; /* Default to original kernel entry point */ + char *cmdline = early_ioremap(fw_arg1, COMMAND_LINE_SIZE); /* Boot command line is passed in fw_arg1 */ + + strscpy(boot_command_line, cmdline, COMMAND_LINE_SIZE); + +#ifdef CONFIG_RANDOMIZE_BASE + location_new = determine_relocation_address(); + + /* Sanity check relocation address */ + if (relocation_addr_valid(location_new)) + random_offset = (unsigned long)location_new - (unsigned long)(_text); +#endif reloc_offset = (unsigned long)_text - VMLINUX_LOAD_ADDRESS; + if (random_offset) { + kernel_length = (long)(_end) - (long)(_text); + + /* Copy the kernel to it's new location */ + memcpy(location_new, _text, kernel_length); + + /* Sync the caches ready for execution of new kernel */ + __asm__ __volatile__ ( + "ibar 0 \t\n" + "dbar 0 \t\n" + ::: "memory"); + + reloc_offset += random_offset; + + /* Return the new kernel's entry point */ + kernel_entry = RELOCATED_KASLR(start_kernel); + + /* The current thread is now within the relocated kernel */ + __current_thread_info = RELOCATED_KASLR(__current_thread_info); + + update_reloc_offset(&reloc_offset, random_offset); + } + if (reloc_offset) relocate_relative(); - relocate_absolute(); + relocate_absolute(random_offset); + + return kernel_entry; } /* From 3f89765d6227fe4645a91e062332b6b4eb24072c Mon Sep 17 00:00:00 2001 From: Youling Tang Date: Sat, 25 Feb 2023 15:52:56 +0800 Subject: [PATCH 494/765] LoongArch: kdump: Add single kernel image implementation This feature depends on the kernel being relocatable. Enable using single kernel image for kdump, and then no longer need to build two kernels (production kernel and capture kernel share a single kernel image). Also enable CONFIG_CRASH_DUMP in loongson3_defconfig. Signed-off-by: Youling Tang Signed-off-by: Huacai Chen --- arch/loongarch/Kconfig | 12 +----------- arch/loongarch/Makefile | 4 ---- arch/loongarch/configs/loongson3_defconfig | 1 + arch/loongarch/include/asm/addrspace.h | 2 ++ arch/loongarch/kernel/head.S | 2 +- 5 files changed, 5 insertions(+), 16 deletions(-) diff --git a/arch/loongarch/Kconfig b/arch/loongarch/Kconfig index 2fc18a3c565e8..c6a02d26b3e5a 100644 --- a/arch/loongarch/Kconfig +++ b/arch/loongarch/Kconfig @@ -473,6 +473,7 @@ config KEXEC config CRASH_DUMP bool "Build kdump crash kernel" + select RELOCATABLE help Generate crash dump after being started by kexec. This should be normally only set in special crash dump kernels which are @@ -482,17 +483,6 @@ config CRASH_DUMP For more details see Documentation/admin-guide/kdump/kdump.rst -config PHYSICAL_START - hex "Physical address where the kernel is loaded" - default "0x90000000a0000000" - depends on CRASH_DUMP - help - This gives the XKPRANGE address where the kernel is loaded. - If you plan to use kernel for capturing the crash dump change - this value to start of the reserved region (the "X" value as - specified in the "crashkernel=YM@XM" command line boot parameter - passed to the panic-ed kernel). - config RELOCATABLE bool "Relocatable kernel" help diff --git a/arch/loongarch/Makefile b/arch/loongarch/Makefile index 768592998b391..f71edf5741011 100644 --- a/arch/loongarch/Makefile +++ b/arch/loongarch/Makefile @@ -79,11 +79,7 @@ endif cflags-y += -ffreestanding cflags-y += $(call cc-option, -mno-check-zero-division) -ifndef CONFIG_PHYSICAL_START load-y = 0x9000000000200000 -else -load-y = $(CONFIG_PHYSICAL_START) -endif bootvars-y = VMLINUX_LOAD_ADDRESS=$(load-y) drivers-$(CONFIG_PCI) += arch/loongarch/pci/ diff --git a/arch/loongarch/configs/loongson3_defconfig b/arch/loongarch/configs/loongson3_defconfig index eb84cae642e58..e18213f01cc47 100644 --- a/arch/loongarch/configs/loongson3_defconfig +++ b/arch/loongarch/configs/loongson3_defconfig @@ -48,6 +48,7 @@ CONFIG_HOTPLUG_CPU=y CONFIG_NR_CPUS=64 CONFIG_NUMA=y CONFIG_KEXEC=y +CONFIG_CRASH_DUMP=y CONFIG_SUSPEND=y CONFIG_HIBERNATION=y CONFIG_ACPI=y diff --git a/arch/loongarch/include/asm/addrspace.h b/arch/loongarch/include/asm/addrspace.h index d342935e5a72d..8fb699b4d40af 100644 --- a/arch/loongarch/include/asm/addrspace.h +++ b/arch/loongarch/include/asm/addrspace.h @@ -125,4 +125,6 @@ extern unsigned long vm_map_base; #define ISA_IOSIZE SZ_16K #define IO_SPACE_LIMIT (PCI_IOSIZE - 1) +#define PHYS_LINK_KADDR PHYSADDR(VMLINUX_LOAD_ADDRESS) + #endif /* _ASM_ADDRSPACE_H */ diff --git a/arch/loongarch/kernel/head.S b/arch/loongarch/kernel/head.S index 1d35becc01eef..aa64b179744f5 100644 --- a/arch/loongarch/kernel/head.S +++ b/arch/loongarch/kernel/head.S @@ -24,7 +24,7 @@ _head: .org 0x8 .dword kernel_entry /* Kernel entry point */ .dword _end - _text /* Kernel image effective size */ - .quad 0 /* Kernel image load offset from start of RAM */ + .quad PHYS_LINK_KADDR /* Kernel image load offset from start of RAM */ .org 0x38 /* 0x20 ~ 0x37 reserved */ .long LINUX_PE_MAGIC .long pe_header - _head /* Offset to the PE header */ From 35c94fab6eee72d60d9cd8d1e6e43e1f77b1dce2 Mon Sep 17 00:00:00 2001 From: Youling Tang Date: Sat, 25 Feb 2023 15:52:56 +0800 Subject: [PATCH 495/765] LoongArch: kdump: Add crashkernel=YM handling When the kernel crashkernel parameter is specified with just a size, we are supposed to allocate a region from RAM to store the crashkernel, "crashkernel=512M" would be recommended for kdump. Fix this by lifting similar code from x86, importing it to LoongArch with LoongArch specific parameters added. We allocate the crashkernel region from the first 4GB of physical memory (because SWIOTLB should be allocated below 4GB). However, LoongArch currently does not implement crashkernel_low and crashkernel_high the same as x86. When X is not specified, crash_base defaults to 0 (crashkernel=YM@XM). Signed-off-by: Youling Tang Signed-off-by: Huacai Chen --- arch/loongarch/kernel/setup.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/arch/loongarch/kernel/setup.c b/arch/loongarch/kernel/setup.c index 4344502c0b317..bae84ccf6d367 100644 --- a/arch/loongarch/kernel/setup.c +++ b/arch/loongarch/kernel/setup.c @@ -234,11 +234,14 @@ static void __init arch_reserve_vmcore(void) #endif } +/* 2MB alignment for crash kernel regions */ +#define CRASH_ALIGN SZ_2M +#define CRASH_ADDR_MAX SZ_4G + static void __init arch_parse_crashkernel(void) { #ifdef CONFIG_KEXEC int ret; - unsigned long long start; unsigned long long total_mem; unsigned long long crash_base, crash_size; @@ -247,8 +250,13 @@ static void __init arch_parse_crashkernel(void) if (ret < 0 || crash_size <= 0) return; - start = memblock_phys_alloc_range(crash_size, 1, crash_base, crash_base + crash_size); - if (start != crash_base) { + if (crash_base <= 0) { + crash_base = memblock_phys_alloc_range(crash_size, CRASH_ALIGN, CRASH_ALIGN, CRASH_ADDR_MAX); + if (!crash_base) { + pr_warn("crashkernel reservation failed - No suitable area found.\n"); + return; + } + } else if (!memblock_phys_alloc_range(crash_size, CRASH_ALIGN, crash_base, crash_base + crash_size)) { pr_warn("Invalid memory region reserved for crash kernel\n"); return; } From edffa33c7bb5a73e90c754c7a497162b77d7c55f Mon Sep 17 00:00:00 2001 From: Qing Zhang Date: Sat, 25 Feb 2023 15:52:57 +0800 Subject: [PATCH 496/765] LoongArch: Add hardware breakpoints/watchpoints support Use perf framework to manage hardware instruction and data breakpoints. LoongArch defines hardware watchpoint functions for instruction fetch and memory load/store operations. After the software configures hardware watchpoints, the processor hardware will monitor the access address of the instruction fetch and load/store operation, and trigger an exception of the watchpoint when it meets the conditions set by the watchpoint. The hardware monitoring points for instruction fetching and load/store operations each have a register for the overall configuration of all monitoring points, a register for recording the status of all monitoring points, and four registers required for configuration of each watchpoint individually. Signed-off-by: Qing Zhang Signed-off-by: Huacai Chen --- arch/loongarch/Kconfig | 1 + arch/loongarch/include/asm/hw_breakpoint.h | 145 ++++++ arch/loongarch/include/asm/loongarch.h | 32 +- arch/loongarch/include/asm/processor.h | 15 +- arch/loongarch/include/asm/switch_to.h | 1 + arch/loongarch/kernel/Makefile | 1 + arch/loongarch/kernel/hw_breakpoint.c | 523 +++++++++++++++++++++ arch/loongarch/kernel/process.c | 7 + arch/loongarch/kernel/traps.c | 10 + 9 files changed, 713 insertions(+), 22 deletions(-) create mode 100644 arch/loongarch/include/asm/hw_breakpoint.h create mode 100644 arch/loongarch/kernel/hw_breakpoint.c diff --git a/arch/loongarch/Kconfig b/arch/loongarch/Kconfig index c6a02d26b3e5a..97423f86384b0 100644 --- a/arch/loongarch/Kconfig +++ b/arch/loongarch/Kconfig @@ -101,6 +101,7 @@ config LOONGARCH select HAVE_FUNCTION_GRAPH_TRACER select HAVE_FUNCTION_TRACER select HAVE_GENERIC_VDSO + select HAVE_HW_BREAKPOINT if PERF_EVENTS select HAVE_IOREMAP_PROT select HAVE_IRQ_EXIT_ON_IRQ_STACK select HAVE_IRQ_TIME_ACCOUNTING diff --git a/arch/loongarch/include/asm/hw_breakpoint.h b/arch/loongarch/include/asm/hw_breakpoint.h new file mode 100644 index 0000000000000..21447fb1efc77 --- /dev/null +++ b/arch/loongarch/include/asm/hw_breakpoint.h @@ -0,0 +1,145 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (C) 2022-2023 Loongson Technology Corporation Limited + */ +#ifndef __ASM_HW_BREAKPOINT_H +#define __ASM_HW_BREAKPOINT_H + +#include + +#ifdef __KERNEL__ + +/* Breakpoint */ +#define LOONGARCH_BREAKPOINT_EXECUTE (0 << 0) + +/* Watchpoints */ +#define LOONGARCH_BREAKPOINT_LOAD (1 << 0) +#define LOONGARCH_BREAKPOINT_STORE (1 << 1) + +struct arch_hw_breakpoint_ctrl { + u32 __reserved : 28, + len : 2, + type : 2; +}; + +struct arch_hw_breakpoint { + u64 address; + u64 mask; + struct arch_hw_breakpoint_ctrl ctrl; +}; + +/* Lengths */ +#define LOONGARCH_BREAKPOINT_LEN_1 0b11 +#define LOONGARCH_BREAKPOINT_LEN_2 0b10 +#define LOONGARCH_BREAKPOINT_LEN_4 0b01 +#define LOONGARCH_BREAKPOINT_LEN_8 0b00 + +/* + * Limits. + * Changing these will require modifications to the register accessors. + */ +#define LOONGARCH_MAX_BRP 8 +#define LOONGARCH_MAX_WRP 8 + +/* Virtual debug register bases. */ +#define CSR_CFG_ADDR 0 +#define CSR_CFG_MASK (CSR_CFG_ADDR + LOONGARCH_MAX_BRP) +#define CSR_CFG_CTRL (CSR_CFG_MASK + LOONGARCH_MAX_BRP) +#define CSR_CFG_ASID (CSR_CFG_CTRL + LOONGARCH_MAX_WRP) + +/* Debug register names. */ +#define LOONGARCH_CSR_NAME_ADDR ADDR +#define LOONGARCH_CSR_NAME_MASK MASK +#define LOONGARCH_CSR_NAME_CTRL CTRL +#define LOONGARCH_CSR_NAME_ASID ASID + +/* Accessor macros for the debug registers. */ +#define LOONGARCH_CSR_WATCH_READ(N, REG, T, VAL) \ +do { \ + if (T == 0) \ + VAL = csr_read64(LOONGARCH_CSR_##IB##N##REG); \ + else \ + VAL = csr_read64(LOONGARCH_CSR_##DB##N##REG); \ +} while (0) + +#define LOONGARCH_CSR_WATCH_WRITE(N, REG, T, VAL) \ +do { \ + if (T == 0) \ + csr_write64(VAL, LOONGARCH_CSR_##IB##N##REG); \ + else \ + csr_write64(VAL, LOONGARCH_CSR_##DB##N##REG); \ +} while (0) + +/* Exact number */ +#define CSR_FWPC_NUM 0x3f +#define CSR_MWPC_NUM 0x3f + +#define CTRL_PLV_ENABLE 0x1e + +#define MWPnCFG3_LoadEn 8 +#define MWPnCFG3_StoreEn 9 + +#define MWPnCFG3_Type_mask 0x3 +#define MWPnCFG3_Size_mask 0x3 + +static inline u32 encode_ctrl_reg(struct arch_hw_breakpoint_ctrl ctrl) +{ + return (ctrl.len << 10) | (ctrl.type << 8); +} + +static inline void decode_ctrl_reg(u32 reg, struct arch_hw_breakpoint_ctrl *ctrl) +{ + reg >>= 8; + ctrl->type = reg & MWPnCFG3_Type_mask; + reg >>= 2; + ctrl->len = reg & MWPnCFG3_Size_mask; +} + +struct task_struct; +struct notifier_block; +struct perf_event; +struct perf_event_attr; + +extern int arch_bp_generic_fields(struct arch_hw_breakpoint_ctrl ctrl, + int *gen_len, int *gen_type, int *offset); +extern int arch_check_bp_in_kernelspace(struct arch_hw_breakpoint *hw); +extern int hw_breakpoint_arch_parse(struct perf_event *bp, + const struct perf_event_attr *attr, + struct arch_hw_breakpoint *hw); +extern int hw_breakpoint_exceptions_notify(struct notifier_block *unused, + unsigned long val, void *data); + +extern int arch_install_hw_breakpoint(struct perf_event *bp); +extern void arch_uninstall_hw_breakpoint(struct perf_event *bp); +extern int hw_breakpoint_slots(int type); +extern void hw_breakpoint_pmu_read(struct perf_event *bp); + +void breakpoint_handler(struct pt_regs *regs); +void watchpoint_handler(struct pt_regs *regs); + +#ifdef CONFIG_HAVE_HW_BREAKPOINT +extern void ptrace_hw_copy_thread(struct task_struct *task); +extern void hw_breakpoint_thread_switch(struct task_struct *next); +#else +static inline void ptrace_hw_copy_thread(struct task_struct *task) +{ +} +static inline void hw_breakpoint_thread_switch(struct task_struct *next) +{ +} +#endif + +/* Determine number of BRP registers available. */ +static inline int get_num_brps(void) +{ + return csr_read64(LOONGARCH_CSR_FWPC) & CSR_FWPC_NUM; +} + +/* Determine number of WRP registers available. */ +static inline int get_num_wrps(void) +{ + return csr_read64(LOONGARCH_CSR_MWPC) & CSR_MWPC_NUM; +} + +#endif /* __KERNEL__ */ +#endif /* __ASM_BREAKPOINT_H */ diff --git a/arch/loongarch/include/asm/loongarch.h b/arch/loongarch/include/asm/loongarch.h index 7f8d57a61c8bd..e9aed583a0643 100644 --- a/arch/loongarch/include/asm/loongarch.h +++ b/arch/loongarch/include/asm/loongarch.h @@ -970,42 +970,42 @@ static __always_inline void iocsr_write64(u64 val, u32 reg) #define LOONGARCH_CSR_DB0ADDR 0x310 /* data breakpoint 0 address */ #define LOONGARCH_CSR_DB0MASK 0x311 /* data breakpoint 0 mask */ -#define LOONGARCH_CSR_DB0CTL 0x312 /* data breakpoint 0 control */ +#define LOONGARCH_CSR_DB0CTRL 0x312 /* data breakpoint 0 control */ #define LOONGARCH_CSR_DB0ASID 0x313 /* data breakpoint 0 asid */ #define LOONGARCH_CSR_DB1ADDR 0x318 /* data breakpoint 1 address */ #define LOONGARCH_CSR_DB1MASK 0x319 /* data breakpoint 1 mask */ -#define LOONGARCH_CSR_DB1CTL 0x31a /* data breakpoint 1 control */ +#define LOONGARCH_CSR_DB1CTRL 0x31a /* data breakpoint 1 control */ #define LOONGARCH_CSR_DB1ASID 0x31b /* data breakpoint 1 asid */ #define LOONGARCH_CSR_DB2ADDR 0x320 /* data breakpoint 2 address */ #define LOONGARCH_CSR_DB2MASK 0x321 /* data breakpoint 2 mask */ -#define LOONGARCH_CSR_DB2CTL 0x322 /* data breakpoint 2 control */ +#define LOONGARCH_CSR_DB2CTRL 0x322 /* data breakpoint 2 control */ #define LOONGARCH_CSR_DB2ASID 0x323 /* data breakpoint 2 asid */ #define LOONGARCH_CSR_DB3ADDR 0x328 /* data breakpoint 3 address */ #define LOONGARCH_CSR_DB3MASK 0x329 /* data breakpoint 3 mask */ -#define LOONGARCH_CSR_DB3CTL 0x32a /* data breakpoint 3 control */ +#define LOONGARCH_CSR_DB3CTRL 0x32a /* data breakpoint 3 control */ #define LOONGARCH_CSR_DB3ASID 0x32b /* data breakpoint 3 asid */ #define LOONGARCH_CSR_DB4ADDR 0x330 /* data breakpoint 4 address */ #define LOONGARCH_CSR_DB4MASK 0x331 /* data breakpoint 4 maks */ -#define LOONGARCH_CSR_DB4CTL 0x332 /* data breakpoint 4 control */ +#define LOONGARCH_CSR_DB4CTRL 0x332 /* data breakpoint 4 control */ #define LOONGARCH_CSR_DB4ASID 0x333 /* data breakpoint 4 asid */ #define LOONGARCH_CSR_DB5ADDR 0x338 /* data breakpoint 5 address */ #define LOONGARCH_CSR_DB5MASK 0x339 /* data breakpoint 5 mask */ -#define LOONGARCH_CSR_DB5CTL 0x33a /* data breakpoint 5 control */ +#define LOONGARCH_CSR_DB5CTRL 0x33a /* data breakpoint 5 control */ #define LOONGARCH_CSR_DB5ASID 0x33b /* data breakpoint 5 asid */ #define LOONGARCH_CSR_DB6ADDR 0x340 /* data breakpoint 6 address */ #define LOONGARCH_CSR_DB6MASK 0x341 /* data breakpoint 6 mask */ -#define LOONGARCH_CSR_DB6CTL 0x342 /* data breakpoint 6 control */ +#define LOONGARCH_CSR_DB6CTRL 0x342 /* data breakpoint 6 control */ #define LOONGARCH_CSR_DB6ASID 0x343 /* data breakpoint 6 asid */ #define LOONGARCH_CSR_DB7ADDR 0x348 /* data breakpoint 7 address */ #define LOONGARCH_CSR_DB7MASK 0x349 /* data breakpoint 7 mask */ -#define LOONGARCH_CSR_DB7CTL 0x34a /* data breakpoint 7 control */ +#define LOONGARCH_CSR_DB7CTRL 0x34a /* data breakpoint 7 control */ #define LOONGARCH_CSR_DB7ASID 0x34b /* data breakpoint 7 asid */ #define LOONGARCH_CSR_FWPC 0x380 /* instruction breakpoint config */ @@ -1013,42 +1013,42 @@ static __always_inline void iocsr_write64(u64 val, u32 reg) #define LOONGARCH_CSR_IB0ADDR 0x390 /* inst breakpoint 0 address */ #define LOONGARCH_CSR_IB0MASK 0x391 /* inst breakpoint 0 mask */ -#define LOONGARCH_CSR_IB0CTL 0x392 /* inst breakpoint 0 control */ +#define LOONGARCH_CSR_IB0CTRL 0x392 /* inst breakpoint 0 control */ #define LOONGARCH_CSR_IB0ASID 0x393 /* inst breakpoint 0 asid */ #define LOONGARCH_CSR_IB1ADDR 0x398 /* inst breakpoint 1 address */ #define LOONGARCH_CSR_IB1MASK 0x399 /* inst breakpoint 1 mask */ -#define LOONGARCH_CSR_IB1CTL 0x39a /* inst breakpoint 1 control */ +#define LOONGARCH_CSR_IB1CTRL 0x39a /* inst breakpoint 1 control */ #define LOONGARCH_CSR_IB1ASID 0x39b /* inst breakpoint 1 asid */ #define LOONGARCH_CSR_IB2ADDR 0x3a0 /* inst breakpoint 2 address */ #define LOONGARCH_CSR_IB2MASK 0x3a1 /* inst breakpoint 2 mask */ -#define LOONGARCH_CSR_IB2CTL 0x3a2 /* inst breakpoint 2 control */ +#define LOONGARCH_CSR_IB2CTRL 0x3a2 /* inst breakpoint 2 control */ #define LOONGARCH_CSR_IB2ASID 0x3a3 /* inst breakpoint 2 asid */ #define LOONGARCH_CSR_IB3ADDR 0x3a8 /* inst breakpoint 3 address */ #define LOONGARCH_CSR_IB3MASK 0x3a9 /* breakpoint 3 mask */ -#define LOONGARCH_CSR_IB3CTL 0x3aa /* inst breakpoint 3 control */ +#define LOONGARCH_CSR_IB3CTRL 0x3aa /* inst breakpoint 3 control */ #define LOONGARCH_CSR_IB3ASID 0x3ab /* inst breakpoint 3 asid */ #define LOONGARCH_CSR_IB4ADDR 0x3b0 /* inst breakpoint 4 address */ #define LOONGARCH_CSR_IB4MASK 0x3b1 /* inst breakpoint 4 mask */ -#define LOONGARCH_CSR_IB4CTL 0x3b2 /* inst breakpoint 4 control */ +#define LOONGARCH_CSR_IB4CTRL 0x3b2 /* inst breakpoint 4 control */ #define LOONGARCH_CSR_IB4ASID 0x3b3 /* inst breakpoint 4 asid */ #define LOONGARCH_CSR_IB5ADDR 0x3b8 /* inst breakpoint 5 address */ #define LOONGARCH_CSR_IB5MASK 0x3b9 /* inst breakpoint 5 mask */ -#define LOONGARCH_CSR_IB5CTL 0x3ba /* inst breakpoint 5 control */ +#define LOONGARCH_CSR_IB5CTRL 0x3ba /* inst breakpoint 5 control */ #define LOONGARCH_CSR_IB5ASID 0x3bb /* inst breakpoint 5 asid */ #define LOONGARCH_CSR_IB6ADDR 0x3c0 /* inst breakpoint 6 address */ #define LOONGARCH_CSR_IB6MASK 0x3c1 /* inst breakpoint 6 mask */ -#define LOONGARCH_CSR_IB6CTL 0x3c2 /* inst breakpoint 6 control */ +#define LOONGARCH_CSR_IB6CTRL 0x3c2 /* inst breakpoint 6 control */ #define LOONGARCH_CSR_IB6ASID 0x3c3 /* inst breakpoint 6 asid */ #define LOONGARCH_CSR_IB7ADDR 0x3c8 /* inst breakpoint 7 address */ #define LOONGARCH_CSR_IB7MASK 0x3c9 /* inst breakpoint 7 mask */ -#define LOONGARCH_CSR_IB7CTL 0x3ca /* inst breakpoint 7 control */ +#define LOONGARCH_CSR_IB7CTRL 0x3ca /* inst breakpoint 7 control */ #define LOONGARCH_CSR_IB7ASID 0x3cb /* inst breakpoint 7 asid */ #define LOONGARCH_CSR_DEBUG 0x500 /* debug config */ diff --git a/arch/loongarch/include/asm/processor.h b/arch/loongarch/include/asm/processor.h index 7184f1dc61f27..5edf78c3a5b4c 100644 --- a/arch/loongarch/include/asm/processor.h +++ b/arch/loongarch/include/asm/processor.h @@ -11,6 +11,7 @@ #include #include +#include #include #include #include @@ -127,10 +128,14 @@ struct thread_struct { struct loongarch_vdso_info *vdso; /* - * FPU & vector registers, must be at last because - * they are conditionally copied at fork(). + * FPU & vector registers, must be at the last of inherited + * context because they are conditionally copied at fork(). */ struct loongarch_fpu fpu FPU_ALIGN; + + /* Hardware breakpoints pinned to this task. */ + struct perf_event *hbp_break[LOONGARCH_MAX_BRP]; + struct perf_event *hbp_watch[LOONGARCH_MAX_WRP]; }; #define thread_saved_ra(tsk) (tsk->thread.sched_ra) @@ -172,6 +177,8 @@ struct thread_struct { .fcc = 0, \ .fpr = {{{0,},},}, \ }, \ + .hbp_break = {0}, \ + .hbp_watch = {0}, \ } struct task_struct; @@ -184,10 +191,6 @@ extern unsigned long boot_option_idle_override; */ extern void start_thread(struct pt_regs *regs, unsigned long pc, unsigned long sp); -static inline void flush_thread(void) -{ -} - unsigned long __get_wchan(struct task_struct *p); #define __KSTK_TOS(tsk) ((unsigned long)task_stack_page(tsk) + \ diff --git a/arch/loongarch/include/asm/switch_to.h b/arch/loongarch/include/asm/switch_to.h index 43a5ab162d38b..24e3094bebab1 100644 --- a/arch/loongarch/include/asm/switch_to.h +++ b/arch/loongarch/include/asm/switch_to.h @@ -34,6 +34,7 @@ extern asmlinkage struct task_struct *__switch_to(struct task_struct *prev, #define switch_to(prev, next, last) \ do { \ lose_fpu_inatomic(1, prev); \ + hw_breakpoint_thread_switch(next); \ (last) = __switch_to(prev, next, task_thread_info(next), \ __builtin_return_address(0), __builtin_frame_address(0)); \ } while (0) diff --git a/arch/loongarch/kernel/Makefile b/arch/loongarch/kernel/Makefile index 94d33dbd4912e..bbb6b047f5db7 100644 --- a/arch/loongarch/kernel/Makefile +++ b/arch/loongarch/kernel/Makefile @@ -50,5 +50,6 @@ obj-$(CONFIG_UNWINDER_GUESS) += unwind_guess.o obj-$(CONFIG_UNWINDER_PROLOGUE) += unwind_prologue.o obj-$(CONFIG_PERF_EVENTS) += perf_event.o perf_regs.o +obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o CPPFLAGS_vmlinux.lds := $(KBUILD_CFLAGS) diff --git a/arch/loongarch/kernel/hw_breakpoint.c b/arch/loongarch/kernel/hw_breakpoint.c new file mode 100644 index 0000000000000..b88efe6c672eb --- /dev/null +++ b/arch/loongarch/kernel/hw_breakpoint.c @@ -0,0 +1,523 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2022-2023 Loongson Technology Corporation Limited + */ +#define pr_fmt(fmt) "hw-breakpoint: " fmt + +#include +#include +#include + +#include + +/* Breakpoint currently in use for each BRP. */ +static DEFINE_PER_CPU(struct perf_event *, bp_on_reg[LOONGARCH_MAX_BRP]); + +/* Watchpoint currently in use for each WRP. */ +static DEFINE_PER_CPU(struct perf_event *, wp_on_reg[LOONGARCH_MAX_WRP]); + +int hw_breakpoint_slots(int type) +{ + /* + * We can be called early, so don't rely on + * our static variables being initialised. + */ + switch (type) { + case TYPE_INST: + return get_num_brps(); + case TYPE_DATA: + return get_num_wrps(); + default: + pr_warn("unknown slot type: %d\n", type); + return 0; + } +} + +#define READ_WB_REG_CASE(OFF, N, REG, T, VAL) \ + case (OFF + N): \ + LOONGARCH_CSR_WATCH_READ(N, REG, T, VAL); \ + break + +#define WRITE_WB_REG_CASE(OFF, N, REG, T, VAL) \ + case (OFF + N): \ + LOONGARCH_CSR_WATCH_WRITE(N, REG, T, VAL); \ + break + +#define GEN_READ_WB_REG_CASES(OFF, REG, T, VAL) \ + READ_WB_REG_CASE(OFF, 0, REG, T, VAL); \ + READ_WB_REG_CASE(OFF, 1, REG, T, VAL); \ + READ_WB_REG_CASE(OFF, 2, REG, T, VAL); \ + READ_WB_REG_CASE(OFF, 3, REG, T, VAL); \ + READ_WB_REG_CASE(OFF, 4, REG, T, VAL); \ + READ_WB_REG_CASE(OFF, 5, REG, T, VAL); \ + READ_WB_REG_CASE(OFF, 6, REG, T, VAL); \ + READ_WB_REG_CASE(OFF, 7, REG, T, VAL); + +#define GEN_WRITE_WB_REG_CASES(OFF, REG, T, VAL) \ + WRITE_WB_REG_CASE(OFF, 0, REG, T, VAL); \ + WRITE_WB_REG_CASE(OFF, 1, REG, T, VAL); \ + WRITE_WB_REG_CASE(OFF, 2, REG, T, VAL); \ + WRITE_WB_REG_CASE(OFF, 3, REG, T, VAL); \ + WRITE_WB_REG_CASE(OFF, 4, REG, T, VAL); \ + WRITE_WB_REG_CASE(OFF, 5, REG, T, VAL); \ + WRITE_WB_REG_CASE(OFF, 6, REG, T, VAL); \ + WRITE_WB_REG_CASE(OFF, 7, REG, T, VAL); + +static u64 read_wb_reg(int reg, int n, int t) +{ + u64 val = 0; + + switch (reg + n) { + GEN_READ_WB_REG_CASES(CSR_CFG_ADDR, ADDR, t, val); + GEN_READ_WB_REG_CASES(CSR_CFG_MASK, MASK, t, val); + GEN_READ_WB_REG_CASES(CSR_CFG_CTRL, CTRL, t, val); + GEN_READ_WB_REG_CASES(CSR_CFG_ASID, ASID, t, val); + default: + pr_warn("Attempt to read from unknown breakpoint register %d\n", n); + } + + return val; +} +NOKPROBE_SYMBOL(read_wb_reg); + +static void write_wb_reg(int reg, int n, int t, u64 val) +{ + switch (reg + n) { + GEN_WRITE_WB_REG_CASES(CSR_CFG_ADDR, ADDR, t, val); + GEN_WRITE_WB_REG_CASES(CSR_CFG_MASK, MASK, t, val); + GEN_WRITE_WB_REG_CASES(CSR_CFG_CTRL, CTRL, t, val); + GEN_WRITE_WB_REG_CASES(CSR_CFG_ASID, ASID, t, val); + default: + pr_warn("Attempt to write to unknown breakpoint register %d\n", n); + } +} +NOKPROBE_SYMBOL(write_wb_reg); + +enum hw_breakpoint_ops { + HW_BREAKPOINT_INSTALL, + HW_BREAKPOINT_UNINSTALL, +}; + +/* + * hw_breakpoint_slot_setup - Find and setup a perf slot according to operations + * + * @slots: pointer to array of slots + * @max_slots: max number of slots + * @bp: perf_event to setup + * @ops: operation to be carried out on the slot + * + * Return: + * slot index on success + * -ENOSPC if no slot is available/matches + * -EINVAL on wrong operations parameter + */ + +static int hw_breakpoint_slot_setup(struct perf_event **slots, int max_slots, + struct perf_event *bp, enum hw_breakpoint_ops ops) +{ + int i; + struct perf_event **slot; + + for (i = 0; i < max_slots; ++i) { + slot = &slots[i]; + switch (ops) { + case HW_BREAKPOINT_INSTALL: + if (!*slot) { + *slot = bp; + return i; + } + break; + case HW_BREAKPOINT_UNINSTALL: + if (*slot == bp) { + *slot = NULL; + return i; + } + break; + default: + pr_warn_once("Unhandled hw breakpoint ops %d\n", ops); + return -EINVAL; + } + } + + return -ENOSPC; +} + +void ptrace_hw_copy_thread(struct task_struct *tsk) +{ + memset(tsk->thread.hbp_break, 0, sizeof(tsk->thread.hbp_break)); + memset(tsk->thread.hbp_watch, 0, sizeof(tsk->thread.hbp_watch)); +} + +/* + * Unregister breakpoints from this task and reset the pointers in the thread_struct. + */ +void flush_ptrace_hw_breakpoint(struct task_struct *tsk) +{ +} + +static int hw_breakpoint_control(struct perf_event *bp, + enum hw_breakpoint_ops ops) +{ + u32 ctrl; + int i, max_slots, enable; + struct perf_event **slots; + struct arch_hw_breakpoint *info = counter_arch_bp(bp); + + if (info->ctrl.type == LOONGARCH_BREAKPOINT_EXECUTE) { + /* Breakpoint */ + slots = this_cpu_ptr(bp_on_reg); + max_slots = boot_cpu_data.watch_ireg_count; + } else { + /* Watchpoint */ + slots = this_cpu_ptr(wp_on_reg); + max_slots = boot_cpu_data.watch_dreg_count; + } + + i = hw_breakpoint_slot_setup(slots, max_slots, bp, ops); + + if (WARN_ONCE(i < 0, "Can't find any breakpoint slot")) + return i; + + switch (ops) { + case HW_BREAKPOINT_INSTALL: + /* Set the FWPnCFG/MWPnCFG 1~4 register. */ + write_wb_reg(CSR_CFG_ADDR, i, 0, info->address); + write_wb_reg(CSR_CFG_ADDR, i, 1, info->address); + write_wb_reg(CSR_CFG_MASK, i, 0, info->mask); + write_wb_reg(CSR_CFG_MASK, i, 1, info->mask); + write_wb_reg(CSR_CFG_ASID, i, 0, 0); + write_wb_reg(CSR_CFG_ASID, i, 1, 0); + if (info->ctrl.type == LOONGARCH_BREAKPOINT_EXECUTE) { + write_wb_reg(CSR_CFG_CTRL, i, 0, CTRL_PLV_ENABLE); + } else { + ctrl = encode_ctrl_reg(info->ctrl); + write_wb_reg(CSR_CFG_CTRL, i, 1, ctrl | CTRL_PLV_ENABLE | + 1 << MWPnCFG3_LoadEn | 1 << MWPnCFG3_StoreEn); + } + enable = csr_read64(LOONGARCH_CSR_CRMD); + csr_write64(CSR_CRMD_WE | enable, LOONGARCH_CSR_CRMD); + break; + case HW_BREAKPOINT_UNINSTALL: + /* Reset the FWPnCFG/MWPnCFG 1~4 register. */ + write_wb_reg(CSR_CFG_ADDR, i, 0, 0); + write_wb_reg(CSR_CFG_ADDR, i, 1, 0); + write_wb_reg(CSR_CFG_MASK, i, 0, 0); + write_wb_reg(CSR_CFG_MASK, i, 1, 0); + write_wb_reg(CSR_CFG_CTRL, i, 0, 0); + write_wb_reg(CSR_CFG_CTRL, i, 1, 0); + write_wb_reg(CSR_CFG_ASID, i, 0, 0); + write_wb_reg(CSR_CFG_ASID, i, 1, 0); + break; + } + + return 0; +} + +/* + * Install a perf counter breakpoint. + */ +int arch_install_hw_breakpoint(struct perf_event *bp) +{ + return hw_breakpoint_control(bp, HW_BREAKPOINT_INSTALL); +} + +void arch_uninstall_hw_breakpoint(struct perf_event *bp) +{ + hw_breakpoint_control(bp, HW_BREAKPOINT_UNINSTALL); +} + +static int get_hbp_len(u8 hbp_len) +{ + unsigned int len_in_bytes = 0; + + switch (hbp_len) { + case LOONGARCH_BREAKPOINT_LEN_1: + len_in_bytes = 1; + break; + case LOONGARCH_BREAKPOINT_LEN_2: + len_in_bytes = 2; + break; + case LOONGARCH_BREAKPOINT_LEN_4: + len_in_bytes = 4; + break; + case LOONGARCH_BREAKPOINT_LEN_8: + len_in_bytes = 8; + break; + } + + return len_in_bytes; +} + +/* + * Check whether bp virtual address is in kernel space. + */ +int arch_check_bp_in_kernelspace(struct arch_hw_breakpoint *hw) +{ + unsigned int len; + unsigned long va; + + va = hw->address; + len = get_hbp_len(hw->ctrl.len); + + return (va >= TASK_SIZE) && ((va + len - 1) >= TASK_SIZE); +} + +/* + * Extract generic type and length encodings from an arch_hw_breakpoint_ctrl. + * Hopefully this will disappear when ptrace can bypass the conversion + * to generic breakpoint descriptions. + */ +int arch_bp_generic_fields(struct arch_hw_breakpoint_ctrl ctrl, + int *gen_len, int *gen_type, int *offset) +{ + /* Type */ + switch (ctrl.type) { + case LOONGARCH_BREAKPOINT_EXECUTE: + *gen_type = HW_BREAKPOINT_X; + break; + case LOONGARCH_BREAKPOINT_LOAD: + *gen_type = HW_BREAKPOINT_R; + break; + case LOONGARCH_BREAKPOINT_STORE: + *gen_type = HW_BREAKPOINT_W; + break; + case LOONGARCH_BREAKPOINT_LOAD | LOONGARCH_BREAKPOINT_STORE: + *gen_type = HW_BREAKPOINT_RW; + break; + default: + return -EINVAL; + } + + if (!ctrl.len) + return -EINVAL; + + *offset = __ffs(ctrl.len); + + /* Len */ + switch (ctrl.len) { + case LOONGARCH_BREAKPOINT_LEN_1: + *gen_len = HW_BREAKPOINT_LEN_1; + break; + case LOONGARCH_BREAKPOINT_LEN_2: + *gen_len = HW_BREAKPOINT_LEN_2; + break; + case LOONGARCH_BREAKPOINT_LEN_4: + *gen_len = HW_BREAKPOINT_LEN_4; + break; + case LOONGARCH_BREAKPOINT_LEN_8: + *gen_len = HW_BREAKPOINT_LEN_8; + break; + default: + return -EINVAL; + } + + return 0; +} + +/* + * Construct an arch_hw_breakpoint from a perf_event. + */ +static int arch_build_bp_info(struct perf_event *bp, + const struct perf_event_attr *attr, + struct arch_hw_breakpoint *hw) +{ + /* Type */ + switch (attr->bp_type) { + case HW_BREAKPOINT_X: + hw->ctrl.type = LOONGARCH_BREAKPOINT_EXECUTE; + break; + case HW_BREAKPOINT_R: + hw->ctrl.type = LOONGARCH_BREAKPOINT_LOAD; + break; + case HW_BREAKPOINT_W: + hw->ctrl.type = LOONGARCH_BREAKPOINT_STORE; + break; + case HW_BREAKPOINT_RW: + hw->ctrl.type = LOONGARCH_BREAKPOINT_LOAD | LOONGARCH_BREAKPOINT_STORE; + break; + default: + return -EINVAL; + } + + /* Len */ + switch (attr->bp_len) { + case HW_BREAKPOINT_LEN_1: + hw->ctrl.len = LOONGARCH_BREAKPOINT_LEN_1; + break; + case HW_BREAKPOINT_LEN_2: + hw->ctrl.len = LOONGARCH_BREAKPOINT_LEN_2; + break; + case HW_BREAKPOINT_LEN_4: + hw->ctrl.len = LOONGARCH_BREAKPOINT_LEN_4; + break; + case HW_BREAKPOINT_LEN_8: + hw->ctrl.len = LOONGARCH_BREAKPOINT_LEN_8; + break; + default: + return -EINVAL; + } + + /* Address */ + hw->address = attr->bp_addr; + + return 0; +} + +/* + * Validate the arch-specific HW Breakpoint register settings. + */ +int hw_breakpoint_arch_parse(struct perf_event *bp, + const struct perf_event_attr *attr, + struct arch_hw_breakpoint *hw) +{ + int ret; + u64 alignment_mask, offset; + + /* Build the arch_hw_breakpoint. */ + ret = arch_build_bp_info(bp, attr, hw); + if (ret) + return ret; + + if (hw->ctrl.type != LOONGARCH_BREAKPOINT_EXECUTE) + alignment_mask = 0x7; + offset = hw->address & alignment_mask; + + hw->address &= ~alignment_mask; + hw->ctrl.len <<= offset; + + return 0; +} + +static void update_bp_registers(struct pt_regs *regs, int enable, int type) +{ + u32 ctrl; + int i, max_slots; + struct perf_event **slots; + struct arch_hw_breakpoint *info; + + switch (type) { + case 0: + slots = this_cpu_ptr(bp_on_reg); + max_slots = boot_cpu_data.watch_ireg_count; + break; + case 1: + slots = this_cpu_ptr(wp_on_reg); + max_slots = boot_cpu_data.watch_dreg_count; + break; + default: + return; + } + + for (i = 0; i < max_slots; ++i) { + if (!slots[i]) + continue; + + info = counter_arch_bp(slots[i]); + if (enable) { + if ((info->ctrl.type == LOONGARCH_BREAKPOINT_EXECUTE) && (type == 0)) { + write_wb_reg(CSR_CFG_CTRL, i, 0, CTRL_PLV_ENABLE); + write_wb_reg(CSR_CFG_CTRL, i, 0, CTRL_PLV_ENABLE); + } else { + ctrl = read_wb_reg(CSR_CFG_CTRL, i, 1); + if (info->ctrl.type == LOONGARCH_BREAKPOINT_LOAD) + ctrl |= 0x1 << MWPnCFG3_LoadEn; + if (info->ctrl.type == LOONGARCH_BREAKPOINT_STORE) + ctrl |= 0x1 << MWPnCFG3_StoreEn; + write_wb_reg(CSR_CFG_CTRL, i, 1, ctrl); + } + regs->csr_prmd |= CSR_PRMD_PWE; + } else { + if ((info->ctrl.type == LOONGARCH_BREAKPOINT_EXECUTE) && (type == 0)) { + write_wb_reg(CSR_CFG_CTRL, i, 0, 0); + } else { + ctrl = read_wb_reg(CSR_CFG_CTRL, i, 1); + if (info->ctrl.type == LOONGARCH_BREAKPOINT_LOAD) + ctrl &= ~0x1 << MWPnCFG3_LoadEn; + if (info->ctrl.type == LOONGARCH_BREAKPOINT_STORE) + ctrl &= ~0x1 << MWPnCFG3_StoreEn; + write_wb_reg(CSR_CFG_CTRL, i, 1, ctrl); + } + regs->csr_prmd &= ~CSR_PRMD_PWE; + } + } +} +NOKPROBE_SYMBOL(update_bp_registers); + +/* + * Debug exception handlers. + */ +void breakpoint_handler(struct pt_regs *regs) +{ + int i; + struct perf_event *bp, **slots; + + slots = this_cpu_ptr(bp_on_reg); + + for (i = 0; i < boot_cpu_data.watch_ireg_count; ++i) { + bp = slots[i]; + if (bp == NULL) + continue; + perf_bp_event(bp, regs); + } + update_bp_registers(regs, 0, 0); +} +NOKPROBE_SYMBOL(breakpoint_handler); + +void watchpoint_handler(struct pt_regs *regs) +{ + int i; + struct perf_event *wp, **slots; + + slots = this_cpu_ptr(wp_on_reg); + + for (i = 0; i < boot_cpu_data.watch_dreg_count; ++i) { + wp = slots[i]; + if (wp == NULL) + continue; + perf_bp_event(wp, regs); + } + update_bp_registers(regs, 0, 1); +} +NOKPROBE_SYMBOL(watchpoint_handler); + +static int __init arch_hw_breakpoint_init(void) +{ + int cpu; + + boot_cpu_data.watch_ireg_count = get_num_brps(); + boot_cpu_data.watch_dreg_count = get_num_wrps(); + + pr_info("Found %d breakpoint and %d watchpoint registers.\n", + boot_cpu_data.watch_ireg_count, boot_cpu_data.watch_dreg_count); + + for (cpu = 1; cpu < NR_CPUS; cpu++) { + cpu_data[cpu].watch_ireg_count = boot_cpu_data.watch_ireg_count; + cpu_data[cpu].watch_dreg_count = boot_cpu_data.watch_dreg_count; + } + + return 0; +} +arch_initcall(arch_hw_breakpoint_init); + +void hw_breakpoint_thread_switch(struct task_struct *next) +{ + struct pt_regs *regs = task_pt_regs(next); + + /* Update breakpoints */ + update_bp_registers(regs, 1, 0); + /* Update watchpoints */ + update_bp_registers(regs, 1, 1); +} + +void hw_breakpoint_pmu_read(struct perf_event *bp) +{ +} + +/* + * Dummy function to register with die_notifier. + */ +int hw_breakpoint_exceptions_notify(struct notifier_block *unused, + unsigned long val, void *data) +{ + return NOTIFY_DONE; +} diff --git a/arch/loongarch/kernel/process.c b/arch/loongarch/kernel/process.c index edfd220a3737a..fa2443c7afb23 100644 --- a/arch/loongarch/kernel/process.c +++ b/arch/loongarch/kernel/process.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -96,6 +97,11 @@ void start_thread(struct pt_regs *regs, unsigned long pc, unsigned long sp) regs->regs[3] = sp; } +void flush_thread(void) +{ + flush_ptrace_hw_breakpoint(current); +} + void exit_thread(struct task_struct *tsk) { } @@ -181,6 +187,7 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args) childregs->regs[2] = tls; out: + ptrace_hw_copy_thread(p); clear_tsk_thread_flag(p, TIF_USEDFPU); clear_tsk_thread_flag(p, TIF_USEDSIMD); clear_tsk_thread_flag(p, TIF_LSX_CTX_LIVE); diff --git a/arch/loongarch/kernel/traps.c b/arch/loongarch/kernel/traps.c index 05511203732c3..ba67d787f0d77 100644 --- a/arch/loongarch/kernel/traps.c +++ b/arch/loongarch/kernel/traps.c @@ -511,7 +511,17 @@ asmlinkage void noinstr do_bp(struct pt_regs *regs) asmlinkage void noinstr do_watch(struct pt_regs *regs) { + irqentry_state_t state = irqentry_enter(regs); + +#ifdef CONFIG_HAVE_HW_BREAKPOINT + breakpoint_handler(regs); + watchpoint_handler(regs); + force_sig(SIGTRAP); +#else pr_warn("Hardware watch point handler not implemented!\n"); +#endif + + irqentry_exit(regs, state); } asmlinkage void noinstr do_ri(struct pt_regs *regs) From 1a69f7a161a78aead07cd4b811d796950e892fa4 Mon Sep 17 00:00:00 2001 From: Qing Zhang Date: Sat, 25 Feb 2023 15:52:57 +0800 Subject: [PATCH 497/765] LoongArch: ptrace: Expose hardware breakpoints to debuggers Implement the regset-based ptrace interface that exposes hardware breakpoints to user-space debuggers to query and set instruction and data breakpoints. Signed-off-by: Qing Zhang Signed-off-by: Huacai Chen --- arch/loongarch/include/uapi/asm/ptrace.h | 9 + arch/loongarch/kernel/ptrace.c | 402 +++++++++++++++++++++++ include/uapi/linux/elf.h | 2 + 3 files changed, 413 insertions(+) diff --git a/arch/loongarch/include/uapi/asm/ptrace.h b/arch/loongarch/include/uapi/asm/ptrace.h index 083193f4a5d5d..cc48ed2620212 100644 --- a/arch/loongarch/include/uapi/asm/ptrace.h +++ b/arch/loongarch/include/uapi/asm/ptrace.h @@ -46,6 +46,15 @@ struct user_fp_state { uint32_t fcsr; }; +struct user_watch_state { + uint16_t dbg_info; + struct { + uint64_t addr; + uint64_t mask; + uint32_t ctrl; + } dbg_regs[8]; +}; + #define PTRACE_SYSEMU 0x1f #define PTRACE_SYSEMU_SINGLESTEP 0x20 diff --git a/arch/loongarch/kernel/ptrace.c b/arch/loongarch/kernel/ptrace.c index dc2b82ea894cd..11b4913956367 100644 --- a/arch/loongarch/kernel/ptrace.c +++ b/arch/loongarch/kernel/ptrace.c @@ -20,7 +20,9 @@ #include #include #include +#include #include +#include #include #include #include @@ -246,6 +248,384 @@ static int cfg_set(struct task_struct *target, return 0; } +#ifdef CONFIG_HAVE_HW_BREAKPOINT + +/* + * Handle hitting a HW-breakpoint. + */ +static void ptrace_hbptriggered(struct perf_event *bp, + struct perf_sample_data *data, + struct pt_regs *regs) +{ + int i; + struct arch_hw_breakpoint *bkpt = counter_arch_bp(bp); + + for (i = 0; i < LOONGARCH_MAX_BRP; ++i) + if (current->thread.hbp_break[i] == bp) + break; + + for (i = 0; i < LOONGARCH_MAX_WRP; ++i) + if (current->thread.hbp_watch[i] == bp) + break; + + force_sig_ptrace_errno_trap(i, (void __user *)bkpt->address); +} + +static struct perf_event *ptrace_hbp_get_event(unsigned int note_type, + struct task_struct *tsk, + unsigned long idx) +{ + struct perf_event *bp; + + switch (note_type) { + case NT_LOONGARCH_HW_BREAK: + if (idx >= LOONGARCH_MAX_BRP) + return ERR_PTR(-EINVAL); + idx = array_index_nospec(idx, LOONGARCH_MAX_BRP); + bp = tsk->thread.hbp_break[idx]; + break; + case NT_LOONGARCH_HW_WATCH: + if (idx >= LOONGARCH_MAX_WRP) + return ERR_PTR(-EINVAL); + idx = array_index_nospec(idx, LOONGARCH_MAX_WRP); + bp = tsk->thread.hbp_watch[idx]; + break; + } + + return bp; +} + +static int ptrace_hbp_set_event(unsigned int note_type, + struct task_struct *tsk, + unsigned long idx, + struct perf_event *bp) +{ + switch (note_type) { + case NT_LOONGARCH_HW_BREAK: + if (idx >= LOONGARCH_MAX_BRP) + return -EINVAL; + idx = array_index_nospec(idx, LOONGARCH_MAX_BRP); + tsk->thread.hbp_break[idx] = bp; + break; + case NT_LOONGARCH_HW_WATCH: + if (idx >= LOONGARCH_MAX_WRP) + return -EINVAL; + idx = array_index_nospec(idx, LOONGARCH_MAX_WRP); + tsk->thread.hbp_watch[idx] = bp; + break; + } + + return 0; +} + +static struct perf_event *ptrace_hbp_create(unsigned int note_type, + struct task_struct *tsk, + unsigned long idx) +{ + int err, type; + struct perf_event *bp; + struct perf_event_attr attr; + + switch (note_type) { + case NT_LOONGARCH_HW_BREAK: + type = HW_BREAKPOINT_X; + break; + case NT_LOONGARCH_HW_WATCH: + type = HW_BREAKPOINT_RW; + break; + default: + return ERR_PTR(-EINVAL); + } + + ptrace_breakpoint_init(&attr); + + /* + * Initialise fields to sane defaults + * (i.e. values that will pass validation). + */ + attr.bp_addr = 0; + attr.bp_len = HW_BREAKPOINT_LEN_4; + attr.bp_type = type; + attr.disabled = 1; + + bp = register_user_hw_breakpoint(&attr, ptrace_hbptriggered, NULL, tsk); + if (IS_ERR(bp)) + return bp; + + err = ptrace_hbp_set_event(note_type, tsk, idx, bp); + if (err) + return ERR_PTR(err); + + return bp; +} + +static int ptrace_hbp_fill_attr_ctrl(unsigned int note_type, + struct arch_hw_breakpoint_ctrl ctrl, + struct perf_event_attr *attr) +{ + int err, len, type, offset; + + err = arch_bp_generic_fields(ctrl, &len, &type, &offset); + if (err) + return err; + + switch (note_type) { + case NT_LOONGARCH_HW_BREAK: + if ((type & HW_BREAKPOINT_X) != type) + return -EINVAL; + break; + case NT_LOONGARCH_HW_WATCH: + if ((type & HW_BREAKPOINT_RW) != type) + return -EINVAL; + break; + default: + return -EINVAL; + } + + attr->bp_len = len; + attr->bp_type = type; + attr->bp_addr += offset; + + return 0; +} + +static int ptrace_hbp_get_resource_info(unsigned int note_type, u16 *info) +{ + u8 num; + u16 reg = 0; + + switch (note_type) { + case NT_LOONGARCH_HW_BREAK: + num = hw_breakpoint_slots(TYPE_INST); + break; + case NT_LOONGARCH_HW_WATCH: + num = hw_breakpoint_slots(TYPE_DATA); + break; + default: + return -EINVAL; + } + + *info = reg | num; + + return 0; +} + +static struct perf_event *ptrace_hbp_get_initialised_bp(unsigned int note_type, + struct task_struct *tsk, + unsigned long idx) +{ + struct perf_event *bp = ptrace_hbp_get_event(note_type, tsk, idx); + + if (!bp) + bp = ptrace_hbp_create(note_type, tsk, idx); + + return bp; +} + +static int ptrace_hbp_get_ctrl(unsigned int note_type, + struct task_struct *tsk, + unsigned long idx, u32 *ctrl) +{ + struct perf_event *bp = ptrace_hbp_get_event(note_type, tsk, idx); + + if (IS_ERR(bp)) + return PTR_ERR(bp); + + *ctrl = bp ? encode_ctrl_reg(counter_arch_bp(bp)->ctrl) : 0; + + return 0; +} + +static int ptrace_hbp_get_mask(unsigned int note_type, + struct task_struct *tsk, + unsigned long idx, u64 *mask) +{ + struct perf_event *bp = ptrace_hbp_get_event(note_type, tsk, idx); + + if (IS_ERR(bp)) + return PTR_ERR(bp); + + *mask = bp ? counter_arch_bp(bp)->mask : 0; + + return 0; +} + +static int ptrace_hbp_get_addr(unsigned int note_type, + struct task_struct *tsk, + unsigned long idx, u64 *addr) +{ + struct perf_event *bp = ptrace_hbp_get_event(note_type, tsk, idx); + + if (IS_ERR(bp)) + return PTR_ERR(bp); + + *addr = bp ? counter_arch_bp(bp)->address : 0; + + return 0; +} + +static int ptrace_hbp_set_ctrl(unsigned int note_type, + struct task_struct *tsk, + unsigned long idx, u32 uctrl) +{ + int err; + struct perf_event *bp; + struct perf_event_attr attr; + struct arch_hw_breakpoint_ctrl ctrl; + + bp = ptrace_hbp_get_initialised_bp(note_type, tsk, idx); + if (IS_ERR(bp)) + return PTR_ERR(bp); + + attr = bp->attr; + decode_ctrl_reg(uctrl, &ctrl); + err = ptrace_hbp_fill_attr_ctrl(note_type, ctrl, &attr); + if (err) + return err; + + return modify_user_hw_breakpoint(bp, &attr); +} + +static int ptrace_hbp_set_mask(unsigned int note_type, + struct task_struct *tsk, + unsigned long idx, u64 mask) +{ + struct perf_event *bp; + struct perf_event_attr attr; + struct arch_hw_breakpoint *info; + + bp = ptrace_hbp_get_initialised_bp(note_type, tsk, idx); + if (IS_ERR(bp)) + return PTR_ERR(bp); + + attr = bp->attr; + info = counter_arch_bp(bp); + info->mask = mask; + + return modify_user_hw_breakpoint(bp, &attr); +} + +static int ptrace_hbp_set_addr(unsigned int note_type, + struct task_struct *tsk, + unsigned long idx, u64 addr) +{ + struct perf_event *bp; + struct perf_event_attr attr; + + bp = ptrace_hbp_get_initialised_bp(note_type, tsk, idx); + if (IS_ERR(bp)) + return PTR_ERR(bp); + + attr = bp->attr; + attr.bp_addr = addr; + + return modify_user_hw_breakpoint(bp, &attr); +} + +#define PTRACE_HBP_CTRL_SZ sizeof(u32) +#define PTRACE_HBP_ADDR_SZ sizeof(u64) +#define PTRACE_HBP_MASK_SZ sizeof(u64) + +static int hw_break_get(struct task_struct *target, + const struct user_regset *regset, + struct membuf to) +{ + u16 info; + u32 ctrl; + u64 addr, mask; + int ret, idx = 0; + unsigned int note_type = regset->core_note_type; + + /* Resource info */ + ret = ptrace_hbp_get_resource_info(note_type, &info); + if (ret) + return ret; + + membuf_write(&to, &info, sizeof(info)); + + /* (address, ctrl) registers */ + while (to.left) { + ret = ptrace_hbp_get_addr(note_type, target, idx, &addr); + if (ret) + return ret; + + ret = ptrace_hbp_get_mask(note_type, target, idx, &mask); + if (ret) + return ret; + + ret = ptrace_hbp_get_ctrl(note_type, target, idx, &ctrl); + if (ret) + return ret; + + membuf_store(&to, addr); + membuf_store(&to, mask); + membuf_store(&to, ctrl); + idx++; + } + + return 0; +} + +static int hw_break_set(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + u32 ctrl; + u64 addr, mask; + int ret, idx = 0, offset, limit; + unsigned int note_type = regset->core_note_type; + + /* Resource info */ + offset = offsetof(struct user_watch_state, dbg_regs); + user_regset_copyin_ignore(&pos, &count, &kbuf, &ubuf, 0, offset); + + /* (address, ctrl) registers */ + limit = regset->n * regset->size; + while (count && offset < limit) { + if (count < PTRACE_HBP_ADDR_SZ) + return -EINVAL; + + ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf, &addr, + offset, offset + PTRACE_HBP_ADDR_SZ); + if (ret) + return ret; + + ret = ptrace_hbp_set_addr(note_type, target, idx, addr); + if (ret) + return ret; + offset += PTRACE_HBP_ADDR_SZ; + + if (!count) + break; + + ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf, &mask, + offset, offset + PTRACE_HBP_ADDR_SZ); + if (ret) + return ret; + + ret = ptrace_hbp_set_mask(note_type, target, idx, mask); + if (ret) + return ret; + offset += PTRACE_HBP_MASK_SZ; + + ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf, &mask, + offset, offset + PTRACE_HBP_MASK_SZ); + if (ret) + return ret; + + ret = ptrace_hbp_set_ctrl(note_type, target, idx, ctrl); + if (ret) + return ret; + offset += PTRACE_HBP_CTRL_SZ; + idx++; + } + + return 0; +} + +#endif + struct pt_regs_offset { const char *name; int offset; @@ -319,6 +699,10 @@ enum loongarch_regset { REGSET_GPR, REGSET_FPR, REGSET_CPUCFG, +#ifdef CONFIG_HAVE_HW_BREAKPOINT + REGSET_HW_BREAK, + REGSET_HW_WATCH, +#endif }; static const struct user_regset loongarch64_regsets[] = { @@ -346,6 +730,24 @@ static const struct user_regset loongarch64_regsets[] = { .regset_get = cfg_get, .set = cfg_set, }, +#ifdef CONFIG_HAVE_HW_BREAKPOINT + [REGSET_HW_BREAK] = { + .core_note_type = NT_LOONGARCH_HW_BREAK, + .n = sizeof(struct user_watch_state) / sizeof(u32), + .size = sizeof(u32), + .align = sizeof(u32), + .regset_get = hw_break_get, + .set = hw_break_set, + }, + [REGSET_HW_WATCH] = { + .core_note_type = NT_LOONGARCH_HW_WATCH, + .n = sizeof(struct user_watch_state) / sizeof(u32), + .size = sizeof(u32), + .align = sizeof(u32), + .regset_get = hw_break_get, + .set = hw_break_set, + }, +#endif }; static const struct user_regset_view user_loongarch64_view = { diff --git a/include/uapi/linux/elf.h b/include/uapi/linux/elf.h index 4c6a8fa5e7ed6..9db473d9405d4 100644 --- a/include/uapi/linux/elf.h +++ b/include/uapi/linux/elf.h @@ -444,6 +444,8 @@ typedef struct elf64_shdr { #define NT_LOONGARCH_LSX 0xa02 /* LoongArch Loongson SIMD Extension registers */ #define NT_LOONGARCH_LASX 0xa03 /* LoongArch Loongson Advanced SIMD Extension registers */ #define NT_LOONGARCH_LBT 0xa04 /* LoongArch Loongson Binary Translation registers */ +#define NT_LOONGARCH_HW_BREAK 0xa05 /* LoongArch hardware breakpoint registers */ +#define NT_LOONGARCH_HW_WATCH 0xa06 /* LoongArch hardware watchpoint registers */ /* Note types with note name "GNU" */ #define NT_GNU_PROPERTY_TYPE_0 5 From 356bd6f23682f11f7afe923d86c7f5f852b97fb2 Mon Sep 17 00:00:00 2001 From: Qing Zhang Date: Sat, 25 Feb 2023 15:52:57 +0800 Subject: [PATCH 498/765] LoongArch: ptrace: Add function argument access API Add regs_get_argument() which returns N th argument of the function call, This enables ftrace kprobe events to access kernel function arguments via $argN syntax for later use. E.g.: echo 'p bio_add_page arg1=$arg1' > kprobe_events bash: echo: write error: Invalid argument Signed-off-by: Qing Zhang Signed-off-by: Huacai Chen --- arch/loongarch/Kconfig | 1 + arch/loongarch/include/asm/ptrace.h | 34 +++++++++++++++++++++++++++++ 2 files changed, 35 insertions(+) diff --git a/arch/loongarch/Kconfig b/arch/loongarch/Kconfig index 97423f86384b0..277ce9b9414e4 100644 --- a/arch/loongarch/Kconfig +++ b/arch/loongarch/Kconfig @@ -98,6 +98,7 @@ config LOONGARCH select HAVE_EXIT_THREAD select HAVE_FAST_GUP select HAVE_FTRACE_MCOUNT_RECORD + select HAVE_FUNCTION_ARG_ACCESS_API select HAVE_FUNCTION_GRAPH_TRACER select HAVE_FUNCTION_TRACER select HAVE_GENERIC_VDSO diff --git a/arch/loongarch/include/asm/ptrace.h b/arch/loongarch/include/asm/ptrace.h index 59c4608de91db..5d38721318663 100644 --- a/arch/loongarch/include/asm/ptrace.h +++ b/arch/loongarch/include/asm/ptrace.h @@ -109,6 +109,40 @@ static inline unsigned long regs_get_kernel_stack_nth(struct pt_regs *regs, unsi struct task_struct; +/** + * regs_get_kernel_argument() - get Nth function argument in kernel + * @regs: pt_regs of that context + * @n: function argument number (start from 0) + * + * regs_get_argument() returns @n th argument of the function call. + * Note that this chooses most probably assignment, in some case + * it can be incorrect. + * This is expected to be called from kprobes or ftrace with regs + * where the top of stack is the return address. + */ +static inline unsigned long regs_get_kernel_argument(struct pt_regs *regs, + unsigned int n) +{ +#define NR_REG_ARGUMENTS 8 + static const unsigned int args[] = { + offsetof(struct pt_regs, regs[4]), + offsetof(struct pt_regs, regs[5]), + offsetof(struct pt_regs, regs[6]), + offsetof(struct pt_regs, regs[7]), + offsetof(struct pt_regs, regs[8]), + offsetof(struct pt_regs, regs[9]), + offsetof(struct pt_regs, regs[10]), + offsetof(struct pt_regs, regs[11]), + }; + + if (n < NR_REG_ARGUMENTS) + return regs_get_register(regs, args[n]); + else { + n -= NR_REG_ARGUMENTS; + return regs_get_kernel_stack_nth(regs, n); + } +} + /* * Does the process account for user or for system time? */ From 424421a7f34c1222d20a6c279f13b9caa71ecc83 Mon Sep 17 00:00:00 2001 From: Qing Zhang Date: Sat, 25 Feb 2023 15:52:57 +0800 Subject: [PATCH 499/765] LoongArch: ptrace: Add hardware single step support Use the generic ptrace_resume code for PTRACE_SYSCALL, PTRACE_CONT, PTRACE_KILL and PTRACE_SINGLESTEP handling. This implies defining arch_has_single_step() and implementing the user_enable_single_step() and user_disable_single_step() functions. LoongArch cannot do hardware single-stepping per se, the hardware single-stepping it is achieved by configuring the instruction fetch watchpoints (FWPS) and specifies that the next instruction must trigger the watch exception by setting the mask bit. In some scenarios CSR.FWPS.Skip is used to ignore the next hit result, avoid endless repeated triggering of the same watchpoint without canceling it. Signed-off-by: Qing Zhang Signed-off-by: Huacai Chen --- arch/loongarch/include/asm/inst.h | 38 ++++++++++++++ arch/loongarch/include/asm/loongarch.h | 3 ++ arch/loongarch/include/asm/processor.h | 1 + arch/loongarch/include/asm/ptrace.h | 4 ++ arch/loongarch/kernel/hw_breakpoint.c | 33 ++++++++++-- arch/loongarch/kernel/ptrace.c | 70 ++++++++++++++++++++++++++ arch/loongarch/kernel/traps.c | 47 ++++++++++++++--- 7 files changed, 186 insertions(+), 10 deletions(-) diff --git a/arch/loongarch/include/asm/inst.h b/arch/loongarch/include/asm/inst.h index 7eedd83fd0d72..36cc5eb4f852b 100644 --- a/arch/loongarch/include/asm/inst.h +++ b/arch/loongarch/include/asm/inst.h @@ -351,6 +351,44 @@ static inline bool is_stack_alloc_ins(union loongarch_instruction *ip) is_imm12_negative(ip->reg2i12_format.immediate); } +static inline bool is_self_loop_ins(union loongarch_instruction *ip, struct pt_regs *regs) +{ + switch (ip->reg0i26_format.opcode) { + case b_op: + case bl_op: + if (ip->reg0i26_format.immediate_l == 0 + && ip->reg0i26_format.immediate_h == 0) + return true; + } + + switch (ip->reg1i21_format.opcode) { + case beqz_op: + case bnez_op: + case bceqz_op: + if (ip->reg1i21_format.immediate_l == 0 + && ip->reg1i21_format.immediate_h == 0) + return true; + } + + switch (ip->reg2i16_format.opcode) { + case beq_op: + case bne_op: + case blt_op: + case bge_op: + case bltu_op: + case bgeu_op: + if (ip->reg2i16_format.immediate == 0) + return true; + break; + case jirl_op: + if (regs->regs[ip->reg2i16_format.rj] + + ((unsigned long)ip->reg2i16_format.immediate << 2) == (unsigned long)ip) + return true; + } + + return false; +} + int larch_insn_read(void *addr, u32 *insnp); int larch_insn_write(void *addr, u32 insn); int larch_insn_patch_text(void *addr, u32 insn); diff --git a/arch/loongarch/include/asm/loongarch.h b/arch/loongarch/include/asm/loongarch.h index e9aed583a0643..65b7dcdea16d0 100644 --- a/arch/loongarch/include/asm/loongarch.h +++ b/arch/loongarch/include/asm/loongarch.h @@ -1055,6 +1055,9 @@ static __always_inline void iocsr_write64(u64 val, u32 reg) #define LOONGARCH_CSR_DERA 0x501 /* debug era */ #define LOONGARCH_CSR_DESAVE 0x502 /* debug save */ +#define CSR_FWPC_SKIP_SHIFT 16 +#define CSR_FWPC_SKIP (_ULCAST_(1) << CSR_FWPC_SKIP_SHIFT) + /* * CSR_ECFG IM */ diff --git a/arch/loongarch/include/asm/processor.h b/arch/loongarch/include/asm/processor.h index 5edf78c3a5b4c..636e1c66398c1 100644 --- a/arch/loongarch/include/asm/processor.h +++ b/arch/loongarch/include/asm/processor.h @@ -125,6 +125,7 @@ struct thread_struct { /* Other stuff associated with the thread. */ unsigned long trap_nr; unsigned long error_code; + unsigned long single_step; /* Used by PTRACE_SINGLESTEP */ struct loongarch_vdso_info *vdso; /* diff --git a/arch/loongarch/include/asm/ptrace.h b/arch/loongarch/include/asm/ptrace.h index 5d38721318663..6eab4393b108f 100644 --- a/arch/loongarch/include/asm/ptrace.h +++ b/arch/loongarch/include/asm/ptrace.h @@ -183,4 +183,8 @@ static inline void user_stack_pointer_set(struct pt_regs *regs, regs->regs[3] = val; } +#ifdef CONFIG_HAVE_HW_BREAKPOINT +#define arch_has_single_step() (1) +#endif + #endif /* _ASM_PTRACE_H */ diff --git a/arch/loongarch/kernel/hw_breakpoint.c b/arch/loongarch/kernel/hw_breakpoint.c index b88efe6c672eb..2406c95b34cc4 100644 --- a/arch/loongarch/kernel/hw_breakpoint.c +++ b/arch/loongarch/kernel/hw_breakpoint.c @@ -153,6 +153,22 @@ void ptrace_hw_copy_thread(struct task_struct *tsk) */ void flush_ptrace_hw_breakpoint(struct task_struct *tsk) { + int i; + struct thread_struct *t = &tsk->thread; + + for (i = 0; i < LOONGARCH_MAX_BRP; i++) { + if (t->hbp_break[i]) { + unregister_hw_breakpoint(t->hbp_break[i]); + t->hbp_break[i] = NULL; + } + } + + for (i = 0; i < LOONGARCH_MAX_WRP; i++) { + if (t->hbp_watch[i]) { + unregister_hw_breakpoint(t->hbp_watch[i]); + t->hbp_watch[i] = NULL; + } + } } static int hw_breakpoint_control(struct perf_event *bp, @@ -501,12 +517,21 @@ arch_initcall(arch_hw_breakpoint_init); void hw_breakpoint_thread_switch(struct task_struct *next) { + u64 addr, mask; struct pt_regs *regs = task_pt_regs(next); - /* Update breakpoints */ - update_bp_registers(regs, 1, 0); - /* Update watchpoints */ - update_bp_registers(regs, 1, 1); + if (test_tsk_thread_flag(next, TIF_SINGLESTEP)) { + addr = read_wb_reg(CSR_CFG_ADDR, 0, 0); + mask = read_wb_reg(CSR_CFG_MASK, 0, 0); + if (!((regs->csr_era ^ addr) & ~mask)) + csr_write32(CSR_FWPC_SKIP, LOONGARCH_CSR_FWPS); + regs->csr_prmd |= CSR_PRMD_PWE; + } else { + /* Update breakpoints */ + update_bp_registers(regs, 1, 0); + /* Update watchpoints */ + update_bp_registers(regs, 1, 1); + } } void hw_breakpoint_pmu_read(struct perf_event *bp) diff --git a/arch/loongarch/kernel/ptrace.c b/arch/loongarch/kernel/ptrace.c index 11b4913956367..06bceae7d1040 100644 --- a/arch/loongarch/kernel/ptrace.c +++ b/arch/loongarch/kernel/ptrace.c @@ -31,6 +31,7 @@ #include #include #include +#include #include #include @@ -41,6 +42,7 @@ #include #include #include +#include #include #include @@ -833,3 +835,71 @@ long arch_ptrace(struct task_struct *child, long request, return ret; } + +#ifdef CONFIG_HAVE_HW_BREAKPOINT +static void ptrace_triggered(struct perf_event *bp, + struct perf_sample_data *data, struct pt_regs *regs) +{ + struct perf_event_attr attr; + + attr = bp->attr; + attr.disabled = true; + modify_user_hw_breakpoint(bp, &attr); +} + +static int set_single_step(struct task_struct *tsk, unsigned long addr) +{ + struct perf_event *bp; + struct perf_event_attr attr; + struct arch_hw_breakpoint *info; + struct thread_struct *thread = &tsk->thread; + + bp = thread->hbp_break[0]; + if (!bp) { + ptrace_breakpoint_init(&attr); + + attr.bp_addr = addr; + attr.bp_len = HW_BREAKPOINT_LEN_8; + attr.bp_type = HW_BREAKPOINT_X; + + bp = register_user_hw_breakpoint(&attr, ptrace_triggered, + NULL, tsk); + if (IS_ERR(bp)) + return PTR_ERR(bp); + + thread->hbp_break[0] = bp; + } else { + int err; + + attr = bp->attr; + attr.bp_addr = addr; + + /* Reenable breakpoint */ + attr.disabled = false; + err = modify_user_hw_breakpoint(bp, &attr); + if (unlikely(err)) + return err; + + csr_write64(attr.bp_addr, LOONGARCH_CSR_IB0ADDR); + } + info = counter_arch_bp(bp); + info->mask = TASK_SIZE - 1; + + return 0; +} + +/* ptrace API */ +void user_enable_single_step(struct task_struct *task) +{ + struct thread_info *ti = task_thread_info(task); + + set_single_step(task, task_pt_regs(task)->csr_era); + task->thread.single_step = task_pt_regs(task)->csr_era; + set_ti_thread_flag(ti, TIF_SINGLESTEP); +} + +void user_disable_single_step(struct task_struct *task) +{ + clear_tsk_thread_flag(task, TIF_SINGLESTEP); +} +#endif diff --git a/arch/loongarch/kernel/traps.c b/arch/loongarch/kernel/traps.c index ba67d787f0d77..1d47747e2154a 100644 --- a/arch/loongarch/kernel/traps.c +++ b/arch/loongarch/kernel/traps.c @@ -513,14 +513,49 @@ asmlinkage void noinstr do_watch(struct pt_regs *regs) { irqentry_state_t state = irqentry_enter(regs); -#ifdef CONFIG_HAVE_HW_BREAKPOINT - breakpoint_handler(regs); - watchpoint_handler(regs); - force_sig(SIGTRAP); -#else +#ifndef CONFIG_HAVE_HW_BREAKPOINT pr_warn("Hardware watch point handler not implemented!\n"); -#endif +#else + if (test_tsk_thread_flag(current, TIF_SINGLESTEP)) { + int llbit = (csr_read32(LOONGARCH_CSR_LLBCTL) & 0x1); + unsigned long pc = instruction_pointer(regs); + union loongarch_instruction *ip = (union loongarch_instruction *)pc; + + if (llbit) { + /* + * When the ll-sc combo is encountered, it is regarded as an single + * instruction. So don't clear llbit and reset CSR.FWPS.Skip until + * the llsc execution is completed. + */ + csr_write32(CSR_FWPC_SKIP, LOONGARCH_CSR_FWPS); + csr_write32(CSR_LLBCTL_KLO, LOONGARCH_CSR_LLBCTL); + goto out; + } + if (pc == current->thread.single_step) { + /* + * Certain insns are occasionally not skipped when CSR.FWPS.Skip is + * set, such as fld.d/fst.d. So singlestep needs to compare whether + * the csr_era is equal to the value of singlestep which last time set. + */ + if (!is_self_loop_ins(ip, regs)) { + /* + * Check if the given instruction the target pc is equal to the + * current pc, If yes, then we should not set the CSR.FWPS.SKIP + * bit to break the original instruction stream. + */ + csr_write32(CSR_FWPC_SKIP, LOONGARCH_CSR_FWPS); + goto out; + } + } + } else { + breakpoint_handler(regs); + watchpoint_handler(regs); + } + + force_sig(SIGTRAP); +out: +#endif irqentry_exit(regs, state); } From 9b3441a6b0e1ed4796e7f5a8586c4d6023c4e3ab Mon Sep 17 00:00:00 2001 From: Tiezhu Yang Date: Sat, 25 Feb 2023 15:52:57 +0800 Subject: [PATCH 500/765] LoongArch: Simulate branch and PC* instructions According to LoongArch Reference Manual, simulate branch and PC* instructions, this is preparation for later patch. Link: https://loongson.github.io/LoongArch-Documentation/LoongArch-Vol1-EN.html#branch-instructions Link: https://loongson.github.io/LoongArch-Documentation/LoongArch-Vol1-EN.html#_pcaddi_pcaddu121_pcaddu18l_pcalau12i Tested-by: Jeff Xie Co-developed-by: Jinyang He Signed-off-by: Jinyang He Signed-off-by: Tiezhu Yang Signed-off-by: Huacai Chen --- arch/loongarch/include/asm/inst.h | 5 ++ arch/loongarch/include/asm/ptrace.h | 1 + arch/loongarch/kernel/inst.c | 123 ++++++++++++++++++++++++++++ 3 files changed, 129 insertions(+) diff --git a/arch/loongarch/include/asm/inst.h b/arch/loongarch/include/asm/inst.h index 36cc5eb4f852b..3ade9ff2bb8ff 100644 --- a/arch/loongarch/include/asm/inst.h +++ b/arch/loongarch/include/asm/inst.h @@ -7,6 +7,7 @@ #include #include +#include #define INSN_NOP 0x03400000 #define INSN_BREAK 0x002a0000 @@ -32,6 +33,7 @@ enum reg1i20_op { lu12iw_op = 0x0a, lu32id_op = 0x0b, pcaddi_op = 0x0c, + pcalau12i_op = 0x0d, pcaddu12i_op = 0x0e, pcaddu18i_op = 0x0f, }; @@ -389,6 +391,9 @@ static inline bool is_self_loop_ins(union loongarch_instruction *ip, struct pt_r return false; } +void simu_pc(struct pt_regs *regs, union loongarch_instruction insn); +void simu_branch(struct pt_regs *regs, union loongarch_instruction insn); + int larch_insn_read(void *addr, u32 *insnp); int larch_insn_write(void *addr, u32 insn); int larch_insn_patch_text(void *addr, u32 insn); diff --git a/arch/loongarch/include/asm/ptrace.h b/arch/loongarch/include/asm/ptrace.h index 6eab4393b108f..d761db943335c 100644 --- a/arch/loongarch/include/asm/ptrace.h +++ b/arch/loongarch/include/asm/ptrace.h @@ -6,6 +6,7 @@ #define _ASM_PTRACE_H #include +#include #include #include diff --git a/arch/loongarch/kernel/inst.c b/arch/loongarch/kernel/inst.c index badc590870423..258ef267cd306 100644 --- a/arch/loongarch/kernel/inst.c +++ b/arch/loongarch/kernel/inst.c @@ -10,6 +10,129 @@ static DEFINE_RAW_SPINLOCK(patch_lock); +void simu_pc(struct pt_regs *regs, union loongarch_instruction insn) +{ + unsigned long pc = regs->csr_era; + unsigned int rd = insn.reg1i20_format.rd; + unsigned int imm = insn.reg1i20_format.immediate; + + if (pc & 3) { + pr_warn("%s: invalid pc 0x%lx\n", __func__, pc); + return; + } + + switch (insn.reg1i20_format.opcode) { + case pcaddi_op: + regs->regs[rd] = pc + sign_extend64(imm << 2, 21); + break; + case pcaddu12i_op: + regs->regs[rd] = pc + sign_extend64(imm << 12, 31); + break; + case pcaddu18i_op: + regs->regs[rd] = pc + sign_extend64(imm << 18, 37); + break; + case pcalau12i_op: + regs->regs[rd] = pc + sign_extend64(imm << 12, 31); + regs->regs[rd] &= ~((1 << 12) - 1); + break; + default: + pr_info("%s: unknown opcode\n", __func__); + return; + } + + regs->csr_era += LOONGARCH_INSN_SIZE; +} + +void simu_branch(struct pt_regs *regs, union loongarch_instruction insn) +{ + unsigned int imm, imm_l, imm_h, rd, rj; + unsigned long pc = regs->csr_era; + + if (pc & 3) { + pr_warn("%s: invalid pc 0x%lx\n", __func__, pc); + return; + } + + imm_l = insn.reg0i26_format.immediate_l; + imm_h = insn.reg0i26_format.immediate_h; + switch (insn.reg0i26_format.opcode) { + case b_op: + regs->csr_era = pc + sign_extend64((imm_h << 16 | imm_l) << 2, 27); + return; + case bl_op: + regs->csr_era = pc + sign_extend64((imm_h << 16 | imm_l) << 2, 27); + regs->regs[1] = pc + LOONGARCH_INSN_SIZE; + return; + } + + imm_l = insn.reg1i21_format.immediate_l; + imm_h = insn.reg1i21_format.immediate_h; + rj = insn.reg1i21_format.rj; + switch (insn.reg1i21_format.opcode) { + case beqz_op: + if (regs->regs[rj] == 0) + regs->csr_era = pc + sign_extend64((imm_h << 16 | imm_l) << 2, 22); + else + regs->csr_era = pc + LOONGARCH_INSN_SIZE; + return; + case bnez_op: + if (regs->regs[rj] != 0) + regs->csr_era = pc + sign_extend64((imm_h << 16 | imm_l) << 2, 22); + else + regs->csr_era = pc + LOONGARCH_INSN_SIZE; + return; + } + + imm = insn.reg2i16_format.immediate; + rj = insn.reg2i16_format.rj; + rd = insn.reg2i16_format.rd; + switch (insn.reg2i16_format.opcode) { + case beq_op: + if (regs->regs[rj] == regs->regs[rd]) + regs->csr_era = pc + sign_extend64(imm << 2, 17); + else + regs->csr_era = pc + LOONGARCH_INSN_SIZE; + break; + case bne_op: + if (regs->regs[rj] != regs->regs[rd]) + regs->csr_era = pc + sign_extend64(imm << 2, 17); + else + regs->csr_era = pc + LOONGARCH_INSN_SIZE; + break; + case blt_op: + if ((long)regs->regs[rj] < (long)regs->regs[rd]) + regs->csr_era = pc + sign_extend64(imm << 2, 17); + else + regs->csr_era = pc + LOONGARCH_INSN_SIZE; + break; + case bge_op: + if ((long)regs->regs[rj] >= (long)regs->regs[rd]) + regs->csr_era = pc + sign_extend64(imm << 2, 17); + else + regs->csr_era = pc + LOONGARCH_INSN_SIZE; + break; + case bltu_op: + if (regs->regs[rj] < regs->regs[rd]) + regs->csr_era = pc + sign_extend64(imm << 2, 17); + else + regs->csr_era = pc + LOONGARCH_INSN_SIZE; + break; + case bgeu_op: + if (regs->regs[rj] >= regs->regs[rd]) + regs->csr_era = pc + sign_extend64(imm << 2, 17); + else + regs->csr_era = pc + LOONGARCH_INSN_SIZE; + break; + case jirl_op: + regs->csr_era = regs->regs[rj] + sign_extend64(imm << 2, 17); + regs->regs[rd] = pc + LOONGARCH_INSN_SIZE; + break; + default: + pr_info("%s: unknown opcode\n", __func__); + return; + } +} + int larch_insn_read(void *addr, u32 *insnp) { int ret; From 6d4cc40fb5f58147defc6c0e9d86e6232fe31616 Mon Sep 17 00:00:00 2001 From: Tiezhu Yang Date: Sat, 25 Feb 2023 15:52:57 +0800 Subject: [PATCH 501/765] LoongArch: Add kprobes support Kprobes allows you to trap at almost any kernel address and execute a callback function, this commit adds kprobes support for LoongArch. Tested-by: Jeff Xie Signed-off-by: Tiezhu Yang Signed-off-by: Huacai Chen --- arch/loongarch/Kconfig | 1 + arch/loongarch/include/asm/inst.h | 15 ++ arch/loongarch/include/asm/kprobes.h | 58 ++++ arch/loongarch/kernel/Makefile | 2 + arch/loongarch/kernel/kprobes.c | 379 +++++++++++++++++++++++++++ arch/loongarch/kernel/traps.c | 14 +- arch/loongarch/mm/fault.c | 3 + 7 files changed, 466 insertions(+), 6 deletions(-) create mode 100644 arch/loongarch/include/asm/kprobes.h create mode 100644 arch/loongarch/kernel/kprobes.c diff --git a/arch/loongarch/Kconfig b/arch/loongarch/Kconfig index 277ce9b9414e4..bb538b68116dc 100644 --- a/arch/loongarch/Kconfig +++ b/arch/loongarch/Kconfig @@ -106,6 +106,7 @@ config LOONGARCH select HAVE_IOREMAP_PROT select HAVE_IRQ_EXIT_ON_IRQ_STACK select HAVE_IRQ_TIME_ACCOUNTING + select HAVE_KPROBES select HAVE_MOD_ARCH_SPECIFIC select HAVE_NMI select HAVE_PCI diff --git a/arch/loongarch/include/asm/inst.h b/arch/loongarch/include/asm/inst.h index 3ade9ff2bb8ff..a04fe755d7193 100644 --- a/arch/loongarch/include/asm/inst.h +++ b/arch/loongarch/include/asm/inst.h @@ -24,6 +24,10 @@ #define ADDR_IMM(addr, INSN) ((addr & ADDR_IMMMASK_##INSN) >> ADDR_IMMSHIFT_##INSN) +enum reg0i15_op { + break_op = 0x54, +}; + enum reg0i26_op { b_op = 0x14, bl_op = 0x15, @@ -180,6 +184,11 @@ enum reg3sa2_op { alsld_op = 0x16, }; +struct reg0i15_format { + unsigned int immediate : 15; + unsigned int opcode : 17; +}; + struct reg0i26_format { unsigned int immediate_h : 10; unsigned int immediate_l : 16; @@ -265,6 +274,7 @@ struct reg3sa2_format { union loongarch_instruction { unsigned int word; + struct reg0i15_format reg0i15_format; struct reg0i26_format reg0i26_format; struct reg1i20_format reg1i20_format; struct reg1i21_format reg1i21_format; @@ -323,6 +333,11 @@ static inline bool is_imm_negative(unsigned long val, unsigned int bit) return val & (1UL << (bit - 1)); } +static inline bool is_break_ins(union loongarch_instruction *ip) +{ + return ip->reg0i15_format.opcode == break_op; +} + static inline bool is_pc_ins(union loongarch_instruction *ip) { return ip->reg1i20_format.opcode >= pcaddi_op && diff --git a/arch/loongarch/include/asm/kprobes.h b/arch/loongarch/include/asm/kprobes.h new file mode 100644 index 0000000000000..7b9fc3ed71c37 --- /dev/null +++ b/arch/loongarch/include/asm/kprobes.h @@ -0,0 +1,58 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +#ifndef __ASM_LOONGARCH_KPROBES_H +#define __ASM_LOONGARCH_KPROBES_H + +#include + +#ifdef CONFIG_KPROBES + +#include +#include + +#define __ARCH_WANT_KPROBES_INSN_SLOT +#define MAX_INSN_SIZE 2 + +#define flush_insn_slot(p) \ +do { \ + if (p->addr) \ + flush_icache_range((unsigned long)p->addr, \ + (unsigned long)p->addr + \ + (MAX_INSN_SIZE * sizeof(kprobe_opcode_t))); \ +} while (0) + +#define kretprobe_blacklist_size 0 + +typedef union loongarch_instruction kprobe_opcode_t; + +/* Architecture specific copy of original instruction */ +struct arch_specific_insn { + /* copy of the original instruction */ + kprobe_opcode_t *insn; + /* restore address after simulation */ + unsigned long restore; +}; + +struct prev_kprobe { + struct kprobe *kp; + unsigned int status; +}; + +/* per-cpu kprobe control block */ +struct kprobe_ctlblk { + unsigned int kprobe_status; + unsigned long saved_status; + struct prev_kprobe prev_kprobe; +}; + +void arch_remove_kprobe(struct kprobe *p); +bool kprobe_fault_handler(struct pt_regs *regs, int trapnr); +bool kprobe_breakpoint_handler(struct pt_regs *regs); +bool kprobe_singlestep_handler(struct pt_regs *regs); + +#else /* !CONFIG_KPROBES */ + +static inline bool kprobe_breakpoint_handler(struct pt_regs *regs) { return false; } +static inline bool kprobe_singlestep_handler(struct pt_regs *regs) { return false; } + +#endif /* CONFIG_KPROBES */ +#endif /* __ASM_LOONGARCH_KPROBES_H */ diff --git a/arch/loongarch/kernel/Makefile b/arch/loongarch/kernel/Makefile index bbb6b047f5db7..0a2f546216c0f 100644 --- a/arch/loongarch/kernel/Makefile +++ b/arch/loongarch/kernel/Makefile @@ -52,4 +52,6 @@ obj-$(CONFIG_UNWINDER_PROLOGUE) += unwind_prologue.o obj-$(CONFIG_PERF_EVENTS) += perf_event.o perf_regs.o obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o +obj-$(CONFIG_KPROBES) += kprobes.o + CPPFLAGS_vmlinux.lds := $(KBUILD_CFLAGS) diff --git a/arch/loongarch/kernel/kprobes.c b/arch/loongarch/kernel/kprobes.c new file mode 100644 index 0000000000000..73eddd76f37cf --- /dev/null +++ b/arch/loongarch/kernel/kprobes.c @@ -0,0 +1,379 @@ +// SPDX-License-Identifier: GPL-2.0-only +#include +#include +#include +#include + +static const union loongarch_instruction breakpoint_insn = { + .reg0i15_format = { + .opcode = break_op, + .immediate = BRK_KPROBE_BP, + } +}; + +static const union loongarch_instruction singlestep_insn = { + .reg0i15_format = { + .opcode = break_op, + .immediate = BRK_KPROBE_SSTEPBP, + } +}; + +DEFINE_PER_CPU(struct kprobe *, current_kprobe); +DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk); + +static bool insns_not_supported(union loongarch_instruction insn) +{ + switch (insn.reg2i14_format.opcode) { + case llw_op: + case lld_op: + case scw_op: + case scd_op: + pr_notice("kprobe: ll and sc instructions are not supported\n"); + return true; + } + + switch (insn.reg1i21_format.opcode) { + case bceqz_op: + pr_notice("kprobe: bceqz and bcnez instructions are not supported\n"); + return true; + } + + return false; +} +NOKPROBE_SYMBOL(insns_not_supported); + +static bool insns_need_simulation(struct kprobe *p) +{ + if (is_pc_ins(&p->opcode)) + return true; + + if (is_branch_ins(&p->opcode)) + return true; + + return false; +} +NOKPROBE_SYMBOL(insns_need_simulation); + +static void arch_simulate_insn(struct kprobe *p, struct pt_regs *regs) +{ + if (is_pc_ins(&p->opcode)) + simu_pc(regs, p->opcode); + else if (is_branch_ins(&p->opcode)) + simu_branch(regs, p->opcode); +} +NOKPROBE_SYMBOL(arch_simulate_insn); + +static void arch_prepare_ss_slot(struct kprobe *p) +{ + p->ainsn.insn[0] = *p->addr; + p->ainsn.insn[1] = singlestep_insn; + p->ainsn.restore = (unsigned long)p->addr + LOONGARCH_INSN_SIZE; +} +NOKPROBE_SYMBOL(arch_prepare_ss_slot); + +static void arch_prepare_simulate(struct kprobe *p) +{ + p->ainsn.restore = 0; +} +NOKPROBE_SYMBOL(arch_prepare_simulate); + +int arch_prepare_kprobe(struct kprobe *p) +{ + if ((unsigned long)p->addr & 0x3) + return -EILSEQ; + + /* copy instruction */ + p->opcode = *p->addr; + + /* decode instruction */ + if (insns_not_supported(p->opcode)) + return -EINVAL; + + if (insns_need_simulation(p)) { + p->ainsn.insn = NULL; + } else { + p->ainsn.insn = get_insn_slot(); + if (!p->ainsn.insn) + return -ENOMEM; + } + + /* prepare the instruction */ + if (p->ainsn.insn) + arch_prepare_ss_slot(p); + else + arch_prepare_simulate(p); + + return 0; +} +NOKPROBE_SYMBOL(arch_prepare_kprobe); + +/* Install breakpoint in text */ +void arch_arm_kprobe(struct kprobe *p) +{ + *p->addr = breakpoint_insn; + flush_insn_slot(p); +} +NOKPROBE_SYMBOL(arch_arm_kprobe); + +/* Remove breakpoint from text */ +void arch_disarm_kprobe(struct kprobe *p) +{ + *p->addr = p->opcode; + flush_insn_slot(p); +} +NOKPROBE_SYMBOL(arch_disarm_kprobe); + +void arch_remove_kprobe(struct kprobe *p) +{ + if (p->ainsn.insn) { + free_insn_slot(p->ainsn.insn, 0); + p->ainsn.insn = NULL; + } +} +NOKPROBE_SYMBOL(arch_remove_kprobe); + +static void save_previous_kprobe(struct kprobe_ctlblk *kcb) +{ + kcb->prev_kprobe.kp = kprobe_running(); + kcb->prev_kprobe.status = kcb->kprobe_status; +} +NOKPROBE_SYMBOL(save_previous_kprobe); + +static void restore_previous_kprobe(struct kprobe_ctlblk *kcb) +{ + __this_cpu_write(current_kprobe, kcb->prev_kprobe.kp); + kcb->kprobe_status = kcb->prev_kprobe.status; +} +NOKPROBE_SYMBOL(restore_previous_kprobe); + +static void set_current_kprobe(struct kprobe *p) +{ + __this_cpu_write(current_kprobe, p); +} +NOKPROBE_SYMBOL(set_current_kprobe); + +/* + * Interrupts need to be disabled before single-step mode is set, + * and not reenabled until after single-step mode ends. + * Without disabling interrupt on local CPU, there is a chance of + * interrupt occurrence in the period of exception return and start + * of out-of-line single-step, that result in wrongly single stepping + * into the interrupt handler. + */ +static void save_local_irqflag(struct kprobe_ctlblk *kcb, + struct pt_regs *regs) +{ + kcb->saved_status = regs->csr_prmd; + regs->csr_prmd &= ~CSR_PRMD_PIE; +} +NOKPROBE_SYMBOL(save_local_irqflag); + +static void restore_local_irqflag(struct kprobe_ctlblk *kcb, + struct pt_regs *regs) +{ + regs->csr_prmd = kcb->saved_status; +} +NOKPROBE_SYMBOL(restore_local_irqflag); + +static void post_kprobe_handler(struct kprobe *cur, struct kprobe_ctlblk *kcb, + struct pt_regs *regs) +{ + /* return addr restore if non-branching insn */ + if (cur->ainsn.restore != 0) + instruction_pointer_set(regs, cur->ainsn.restore); + + /* restore back original saved kprobe variables and continue */ + if (kcb->kprobe_status == KPROBE_REENTER) { + restore_previous_kprobe(kcb); + preempt_enable_no_resched(); + return; + } + + /* + * update the kcb status even if the cur->post_handler is + * not set because reset_curent_kprobe() doesn't update kcb. + */ + kcb->kprobe_status = KPROBE_HIT_SSDONE; + if (cur->post_handler) + cur->post_handler(cur, regs, 0); + + reset_current_kprobe(); + preempt_enable_no_resched(); +} +NOKPROBE_SYMBOL(post_kprobe_handler); + +static void setup_singlestep(struct kprobe *p, struct pt_regs *regs, + struct kprobe_ctlblk *kcb, int reenter) +{ + if (reenter) { + save_previous_kprobe(kcb); + set_current_kprobe(p); + kcb->kprobe_status = KPROBE_REENTER; + } else { + kcb->kprobe_status = KPROBE_HIT_SS; + } + + if (p->ainsn.insn) { + /* IRQs and single stepping do not mix well */ + save_local_irqflag(kcb, regs); + /* set ip register to prepare for single stepping */ + regs->csr_era = (unsigned long)p->ainsn.insn; + } else { + /* simulate single steping */ + arch_simulate_insn(p, regs); + /* now go for post processing */ + post_kprobe_handler(p, kcb, regs); + } +} +NOKPROBE_SYMBOL(setup_singlestep); + +static bool reenter_kprobe(struct kprobe *p, struct pt_regs *regs, + struct kprobe_ctlblk *kcb) +{ + switch (kcb->kprobe_status) { + case KPROBE_HIT_SS: + case KPROBE_HIT_SSDONE: + case KPROBE_HIT_ACTIVE: + kprobes_inc_nmissed_count(p); + setup_singlestep(p, regs, kcb, 1); + break; + case KPROBE_REENTER: + pr_warn("Failed to recover from reentered kprobes.\n"); + dump_kprobe(p); + WARN_ON_ONCE(1); + break; + default: + WARN_ON(1); + return false; + } + + return true; +} +NOKPROBE_SYMBOL(reenter_kprobe); + +bool kprobe_breakpoint_handler(struct pt_regs *regs) +{ + struct kprobe_ctlblk *kcb; + struct kprobe *p, *cur_kprobe; + kprobe_opcode_t *addr = (kprobe_opcode_t *)regs->csr_era; + + /* + * We don't want to be preempted for the entire + * duration of kprobe processing. + */ + preempt_disable(); + kcb = get_kprobe_ctlblk(); + cur_kprobe = kprobe_running(); + + p = get_kprobe(addr); + if (p) { + if (cur_kprobe) { + if (reenter_kprobe(p, regs, kcb)) + return true; + } else { + /* Probe hit */ + set_current_kprobe(p); + kcb->kprobe_status = KPROBE_HIT_ACTIVE; + + /* + * If we have no pre-handler or it returned 0, we + * continue with normal processing. If we have a + * pre-handler and it returned non-zero, it will + * modify the execution path and no need to single + * stepping. Let's just reset current kprobe and exit. + * + * pre_handler can hit a breakpoint and can step thru + * before return. + */ + if (!p->pre_handler || !p->pre_handler(p, regs)) { + setup_singlestep(p, regs, kcb, 0); + } else { + reset_current_kprobe(); + preempt_enable_no_resched(); + } + return true; + } + } + + if (addr->word != breakpoint_insn.word) { + /* + * The breakpoint instruction was removed right + * after we hit it. Another cpu has removed + * either a probepoint or a debugger breakpoint + * at this address. In either case, no further + * handling of this interrupt is appropriate. + * Return back to original instruction, and continue. + */ + regs->csr_era = (unsigned long)addr; + preempt_enable_no_resched(); + return true; + } + + preempt_enable_no_resched(); + return false; +} +NOKPROBE_SYMBOL(kprobe_breakpoint_handler); + +bool kprobe_singlestep_handler(struct pt_regs *regs) +{ + struct kprobe *cur = kprobe_running(); + struct kprobe_ctlblk *kcb = get_kprobe_ctlblk(); + unsigned long addr = instruction_pointer(regs); + + if (cur && (kcb->kprobe_status & (KPROBE_HIT_SS | KPROBE_REENTER)) && + ((unsigned long)&cur->ainsn.insn[1] == addr)) { + restore_local_irqflag(kcb, regs); + post_kprobe_handler(cur, kcb, regs); + return true; + } + + preempt_enable_no_resched(); + return false; +} +NOKPROBE_SYMBOL(kprobe_singlestep_handler); + +bool kprobe_fault_handler(struct pt_regs *regs, int trapnr) +{ + struct kprobe *cur = kprobe_running(); + struct kprobe_ctlblk *kcb = get_kprobe_ctlblk(); + + switch (kcb->kprobe_status) { + case KPROBE_HIT_SS: + case KPROBE_REENTER: + /* + * We are here because the instruction being single + * stepped caused a page fault. We reset the current + * kprobe and the ip points back to the probe address + * and allow the page fault handler to continue as a + * normal page fault. + */ + regs->csr_era = (unsigned long)cur->addr; + WARN_ON_ONCE(!instruction_pointer(regs)); + + if (kcb->kprobe_status == KPROBE_REENTER) { + restore_previous_kprobe(kcb); + } else { + restore_local_irqflag(kcb, regs); + reset_current_kprobe(); + } + preempt_enable_no_resched(); + break; + } + return false; +} +NOKPROBE_SYMBOL(kprobe_fault_handler); + +/* + * Provide a blacklist of symbols identifying ranges which cannot be kprobed. + * This blacklist is exposed to userspace via debugfs (kprobes/blacklist). + */ +int __init arch_populate_kprobe_blacklist(void) +{ + return kprobe_add_area_blacklist((unsigned long)__irqentry_text_start, + (unsigned long)__irqentry_text_end); +} + +int __init arch_init_kprobes(void) +{ + return 0; +} diff --git a/arch/loongarch/kernel/traps.c b/arch/loongarch/kernel/traps.c index 1d47747e2154a..de8ebe20b666c 100644 --- a/arch/loongarch/kernel/traps.c +++ b/arch/loongarch/kernel/traps.c @@ -437,7 +437,9 @@ asmlinkage void noinstr do_bp(struct pt_regs *regs) unsigned long era = exception_era(regs); irqentry_state_t state = irqentry_enter(regs); - local_irq_enable(); + if (regs->csr_prmd & CSR_PRMD_PIE) + local_irq_enable(); + current->thread.trap_nr = read_csr_excode(); if (__get_inst(&opcode, (u32 *)era, user)) goto out_sigsegv; @@ -450,14 +452,12 @@ asmlinkage void noinstr do_bp(struct pt_regs *regs) */ switch (bcode) { case BRK_KPROBE_BP: - if (notify_die(DIE_BREAK, "Kprobe", regs, bcode, - current->thread.trap_nr, SIGTRAP) == NOTIFY_STOP) + if (kprobe_breakpoint_handler(regs)) goto out; else break; case BRK_KPROBE_SSTEPBP: - if (notify_die(DIE_SSTEPBP, "Kprobe_SingleStep", regs, bcode, - current->thread.trap_nr, SIGTRAP) == NOTIFY_STOP) + if (kprobe_singlestep_handler(regs)) goto out; else break; @@ -500,7 +500,9 @@ asmlinkage void noinstr do_bp(struct pt_regs *regs) } out: - local_irq_disable(); + if (regs->csr_prmd & CSR_PRMD_PIE) + local_irq_disable(); + irqentry_exit(regs, state); return; diff --git a/arch/loongarch/mm/fault.c b/arch/loongarch/mm/fault.c index 1ccd53655cab0..449087bd589d3 100644 --- a/arch/loongarch/mm/fault.c +++ b/arch/loongarch/mm/fault.c @@ -135,6 +135,9 @@ static void __kprobes __do_page_fault(struct pt_regs *regs, struct vm_area_struct *vma = NULL; vm_fault_t fault; + if (kprobe_page_fault(regs, current->thread.trap_nr)) + return; + /* * We fault-in kernel-space virtual memory on-demand. The * 'reference' page table is init_mm.pgd. From 3f5536860086d906b01ec5ed68cf50c7edcc40af Mon Sep 17 00:00:00 2001 From: Tiezhu Yang Date: Sat, 25 Feb 2023 15:52:57 +0800 Subject: [PATCH 502/765] LoongArch: Add kretprobes support Use the generic kretprobe trampoline handler to add kretprobes support for LoongArch. Tested-by: Jeff Xie Signed-off-by: Tiezhu Yang Signed-off-by: Huacai Chen --- arch/loongarch/Kconfig | 1 + arch/loongarch/include/asm/kprobes.h | 3 + arch/loongarch/kernel/Makefile | 2 +- arch/loongarch/kernel/kprobes.c | 27 ++++++ arch/loongarch/kernel/kprobes_trampoline.S | 96 ++++++++++++++++++++++ 5 files changed, 128 insertions(+), 1 deletion(-) create mode 100644 arch/loongarch/kernel/kprobes_trampoline.S diff --git a/arch/loongarch/Kconfig b/arch/loongarch/Kconfig index bb538b68116dc..d90bb47f1dc3e 100644 --- a/arch/loongarch/Kconfig +++ b/arch/loongarch/Kconfig @@ -107,6 +107,7 @@ config LOONGARCH select HAVE_IRQ_EXIT_ON_IRQ_STACK select HAVE_IRQ_TIME_ACCOUNTING select HAVE_KPROBES + select HAVE_KRETPROBES select HAVE_MOD_ARCH_SPECIFIC select HAVE_NMI select HAVE_PCI diff --git a/arch/loongarch/include/asm/kprobes.h b/arch/loongarch/include/asm/kprobes.h index 7b9fc3ed71c37..798020ae02c69 100644 --- a/arch/loongarch/include/asm/kprobes.h +++ b/arch/loongarch/include/asm/kprobes.h @@ -49,6 +49,9 @@ bool kprobe_fault_handler(struct pt_regs *regs, int trapnr); bool kprobe_breakpoint_handler(struct pt_regs *regs); bool kprobe_singlestep_handler(struct pt_regs *regs); +void __kretprobe_trampoline(void); +void *trampoline_probe_handler(struct pt_regs *regs); + #else /* !CONFIG_KPROBES */ static inline bool kprobe_breakpoint_handler(struct pt_regs *regs) { return false; } diff --git a/arch/loongarch/kernel/Makefile b/arch/loongarch/kernel/Makefile index 0a2f546216c0f..78d4e33843054 100644 --- a/arch/loongarch/kernel/Makefile +++ b/arch/loongarch/kernel/Makefile @@ -52,6 +52,6 @@ obj-$(CONFIG_UNWINDER_PROLOGUE) += unwind_prologue.o obj-$(CONFIG_PERF_EVENTS) += perf_event.o perf_regs.o obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o -obj-$(CONFIG_KPROBES) += kprobes.o +obj-$(CONFIG_KPROBES) += kprobes.o kprobes_trampoline.o CPPFLAGS_vmlinux.lds := $(KBUILD_CFLAGS) diff --git a/arch/loongarch/kernel/kprobes.c b/arch/loongarch/kernel/kprobes.c index 73eddd76f37cf..56c8c4b09a429 100644 --- a/arch/loongarch/kernel/kprobes.c +++ b/arch/loongarch/kernel/kprobes.c @@ -377,3 +377,30 @@ int __init arch_init_kprobes(void) { return 0; } + +/* ASM function that handles the kretprobes must not be probed */ +NOKPROBE_SYMBOL(__kretprobe_trampoline); + +/* Called from __kretprobe_trampoline */ +void __used *trampoline_probe_handler(struct pt_regs *regs) +{ + return (void *)kretprobe_trampoline_handler(regs, NULL); +} +NOKPROBE_SYMBOL(trampoline_probe_handler); + +void arch_prepare_kretprobe(struct kretprobe_instance *ri, + struct pt_regs *regs) +{ + ri->ret_addr = (kprobe_opcode_t *)regs->regs[1]; + ri->fp = NULL; + + /* Replace the return addr with trampoline addr */ + regs->regs[1] = (unsigned long)&__kretprobe_trampoline; +} +NOKPROBE_SYMBOL(arch_prepare_kretprobe); + +int arch_trampoline_kprobe(struct kprobe *p) +{ + return 0; +} +NOKPROBE_SYMBOL(arch_trampoline_kprobe); diff --git a/arch/loongarch/kernel/kprobes_trampoline.S b/arch/loongarch/kernel/kprobes_trampoline.S new file mode 100644 index 0000000000000..af94b0d213fa9 --- /dev/null +++ b/arch/loongarch/kernel/kprobes_trampoline.S @@ -0,0 +1,96 @@ +/* SPDX-License-Identifier: GPL-2.0+ */ +#include +#include + + .text + + .macro save_all_base_regs + cfi_st ra, PT_R1 + cfi_st tp, PT_R2 + cfi_st a0, PT_R4 + cfi_st a1, PT_R5 + cfi_st a2, PT_R6 + cfi_st a3, PT_R7 + cfi_st a4, PT_R8 + cfi_st a5, PT_R9 + cfi_st a6, PT_R10 + cfi_st a7, PT_R11 + cfi_st t0, PT_R12 + cfi_st t1, PT_R13 + cfi_st t2, PT_R14 + cfi_st t3, PT_R15 + cfi_st t4, PT_R16 + cfi_st t5, PT_R17 + cfi_st t6, PT_R18 + cfi_st t7, PT_R19 + cfi_st t8, PT_R20 + cfi_st u0, PT_R21 + cfi_st fp, PT_R22 + cfi_st s0, PT_R23 + cfi_st s1, PT_R24 + cfi_st s2, PT_R25 + cfi_st s3, PT_R26 + cfi_st s4, PT_R27 + cfi_st s5, PT_R28 + cfi_st s6, PT_R29 + cfi_st s7, PT_R30 + cfi_st s8, PT_R31 + csrrd t0, LOONGARCH_CSR_CRMD + andi t0, t0, 0x7 /* extract bit[1:0] PLV, bit[2] IE */ + LONG_S t0, sp, PT_CRMD + .endm + + .macro restore_all_base_regs + cfi_ld tp, PT_R2 + cfi_ld a0, PT_R4 + cfi_ld a1, PT_R5 + cfi_ld a2, PT_R6 + cfi_ld a3, PT_R7 + cfi_ld a4, PT_R8 + cfi_ld a5, PT_R9 + cfi_ld a6, PT_R10 + cfi_ld a7, PT_R11 + cfi_ld t0, PT_R12 + cfi_ld t1, PT_R13 + cfi_ld t2, PT_R14 + cfi_ld t3, PT_R15 + cfi_ld t4, PT_R16 + cfi_ld t5, PT_R17 + cfi_ld t6, PT_R18 + cfi_ld t7, PT_R19 + cfi_ld t8, PT_R20 + cfi_ld u0, PT_R21 + cfi_ld fp, PT_R22 + cfi_ld s0, PT_R23 + cfi_ld s1, PT_R24 + cfi_ld s2, PT_R25 + cfi_ld s3, PT_R26 + cfi_ld s4, PT_R27 + cfi_ld s5, PT_R28 + cfi_ld s6, PT_R29 + cfi_ld s7, PT_R30 + cfi_ld s8, PT_R31 + LONG_L t0, sp, PT_CRMD + li.d t1, 0x7 /* mask bit[1:0] PLV, bit[2] IE */ + csrxchg t0, t1, LOONGARCH_CSR_CRMD + .endm + +SYM_CODE_START(__kretprobe_trampoline) + addi.d sp, sp, -PT_SIZE + save_all_base_regs + + addi.d t0, sp, PT_SIZE + LONG_S t0, sp, PT_R3 + + move a0, sp /* pt_regs */ + + bl trampoline_probe_handler + + /* use the result as the return-address */ + move ra, a0 + + restore_all_base_regs + addi.d sp, sp, PT_SIZE + + jr ra +SYM_CODE_END(__kretprobe_trampoline) From 09e679c28a4def4ea8d4f618503c1d55c355ae5c Mon Sep 17 00:00:00 2001 From: Tiezhu Yang Date: Sat, 25 Feb 2023 15:52:57 +0800 Subject: [PATCH 503/765] LoongArch: Add kprobes on ftrace support Add kprobe_ftrace_handler() and arch_prepare_kprobe_ftrace() to support kprobes on ftrace, the code is similar with x86 and riscv. Here is a simple example: # echo 'p:myprobe kernel_clone' > /sys/kernel/debug/tracing/kprobe_events # echo 'r:myretprobe kernel_clone $retval' >> /sys/kernel/debug/tracing/kprobe_events # echo 1 > /sys/kernel/debug/tracing/events/kprobes/myprobe/enable # echo 1 > /sys/kernel/debug/tracing/events/kprobes/myretprobe/enable # echo 1 > /sys/kernel/debug/tracing/tracing_on # cat /sys/kernel/debug/tracing/trace # tracer: nop # # entries-in-buffer/entries-written: 2/2 #P:4 # # _-----=> irqs-off/BH-disabled # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / _-=> migrate-disable # |||| / delay # TASK-PID CPU# ||||| TIMESTAMP FUNCTION # | | | ||||| | | bash-488 [002] ..... 2041.190681: myprobe: (kernel_clone+0x0/0x40c) bash-488 [002] ..... 2041.190788: myretprobe: (__do_sys_clone+0x84/0xb8 <- kernel_clone) arg1=0x200 Tested-by: Jeff Xie Signed-off-by: Tiezhu Yang Signed-off-by: Huacai Chen --- arch/loongarch/Kconfig | 1 + arch/loongarch/kernel/ftrace_dyn.c | 64 ++++++++++++++++++++++++++++++ 2 files changed, 65 insertions(+) diff --git a/arch/loongarch/Kconfig b/arch/loongarch/Kconfig index d90bb47f1dc3e..7fd51257e0ed4 100644 --- a/arch/loongarch/Kconfig +++ b/arch/loongarch/Kconfig @@ -107,6 +107,7 @@ config LOONGARCH select HAVE_IRQ_EXIT_ON_IRQ_STACK select HAVE_IRQ_TIME_ACCOUNTING select HAVE_KPROBES + select HAVE_KPROBES_ON_FTRACE select HAVE_KRETPROBES select HAVE_MOD_ARCH_SPECIFIC select HAVE_NMI diff --git a/arch/loongarch/kernel/ftrace_dyn.c b/arch/loongarch/kernel/ftrace_dyn.c index 0f07591cab309..4a3ef8516ccc6 100644 --- a/arch/loongarch/kernel/ftrace_dyn.c +++ b/arch/loongarch/kernel/ftrace_dyn.c @@ -6,6 +6,7 @@ */ #include +#include #include #include @@ -271,3 +272,66 @@ int ftrace_disable_ftrace_graph_caller(void) } #endif /* CONFIG_HAVE_DYNAMIC_FTRACE_WITH_ARGS */ #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ + +#ifdef CONFIG_KPROBES_ON_FTRACE +/* Ftrace callback handler for kprobes -- called under preepmt disabled */ +void kprobe_ftrace_handler(unsigned long ip, unsigned long parent_ip, + struct ftrace_ops *ops, struct ftrace_regs *fregs) +{ + int bit; + struct pt_regs *regs; + struct kprobe *p; + struct kprobe_ctlblk *kcb; + + bit = ftrace_test_recursion_trylock(ip, parent_ip); + if (bit < 0) + return; + + p = get_kprobe((kprobe_opcode_t *)ip); + if (unlikely(!p) || kprobe_disabled(p)) + goto out; + + regs = ftrace_get_regs(fregs); + if (!regs) + goto out; + + kcb = get_kprobe_ctlblk(); + if (kprobe_running()) { + kprobes_inc_nmissed_count(p); + } else { + unsigned long orig_ip = instruction_pointer(regs); + + instruction_pointer_set(regs, ip); + + __this_cpu_write(current_kprobe, p); + kcb->kprobe_status = KPROBE_HIT_ACTIVE; + if (!p->pre_handler || !p->pre_handler(p, regs)) { + /* + * Emulate singlestep (and also recover regs->csr_era) + * as if there is a nop + */ + instruction_pointer_set(regs, (unsigned long)p->addr + MCOUNT_INSN_SIZE); + if (unlikely(p->post_handler)) { + kcb->kprobe_status = KPROBE_HIT_SSDONE; + p->post_handler(p, regs, 0); + } + instruction_pointer_set(regs, orig_ip); + } + + /* + * If pre_handler returns !0, it changes regs->csr_era. We have to + * skip emulating post_handler. + */ + __this_cpu_write(current_kprobe, NULL); + } +out: + ftrace_test_recursion_unlock(bit); +} +NOKPROBE_SYMBOL(kprobe_ftrace_handler); + +int arch_prepare_kprobe_ftrace(struct kprobe *p) +{ + p->ainsn.insn = NULL; + return 0; +} +#endif /* CONFIG_KPROBES_ON_FTRACE */ From fcf77d016216ae75a19078e1079c2511a6319a49 Mon Sep 17 00:00:00 2001 From: Tiezhu Yang Date: Sat, 25 Feb 2023 15:52:57 +0800 Subject: [PATCH 504/765] LoongArch: Mark some assembler symbols as non-kprobe-able Some assembler symbols are not kprobe safe, such as handle_syscall (used as syscall exception handler), *memset*/*memcpy*/*memmove* (may cause recursive exceptions), they can not be instrumented, just blacklist them for kprobing. Here is a related problem and discussion: Link: https://lore.kernel.org/lkml/20230114143859.7ccc45c1c5d9ce302113ab0a@kernel.org/ Tested-by: Jeff Xie Signed-off-by: Tiezhu Yang Signed-off-by: Huacai Chen --- arch/loongarch/include/asm/asm.h | 10 ++++++++++ arch/loongarch/kernel/entry.S | 1 + arch/loongarch/lib/memcpy.S | 3 +++ arch/loongarch/lib/memmove.S | 4 ++++ arch/loongarch/lib/memset.S | 3 +++ 5 files changed, 21 insertions(+) diff --git a/arch/loongarch/include/asm/asm.h b/arch/loongarch/include/asm/asm.h index 40eea6aa469e1..f591b3245def6 100644 --- a/arch/loongarch/include/asm/asm.h +++ b/arch/loongarch/include/asm/asm.h @@ -188,4 +188,14 @@ #define PTRLOG 3 #endif +/* Annotate a function as being unsuitable for kprobes. */ +#ifdef CONFIG_KPROBES +#define _ASM_NOKPROBE(name) \ + .pushsection "_kprobe_blacklist", "aw"; \ + .quad name; \ + .popsection +#else +#define _ASM_NOKPROBE(name) +#endif + #endif /* __ASM_ASM_H */ diff --git a/arch/loongarch/kernel/entry.S b/arch/loongarch/kernel/entry.S index f14ab6a0e0b05..d737e3cf42d3f 100644 --- a/arch/loongarch/kernel/entry.S +++ b/arch/loongarch/kernel/entry.S @@ -67,6 +67,7 @@ SYM_FUNC_START(handle_syscall) RESTORE_ALL_AND_RET SYM_FUNC_END(handle_syscall) +_ASM_NOKPROBE(handle_syscall) SYM_CODE_START(ret_from_fork) bl schedule_tail # a0 = struct task_struct *prev diff --git a/arch/loongarch/lib/memcpy.S b/arch/loongarch/lib/memcpy.S index 7c07d595ee89a..3b7e1dec71094 100644 --- a/arch/loongarch/lib/memcpy.S +++ b/arch/loongarch/lib/memcpy.S @@ -17,6 +17,7 @@ SYM_FUNC_START(memcpy) ALTERNATIVE "b __memcpy_generic", \ "b __memcpy_fast", CPU_FEATURE_UAL SYM_FUNC_END(memcpy) +_ASM_NOKPROBE(memcpy) EXPORT_SYMBOL(memcpy) @@ -41,6 +42,7 @@ SYM_FUNC_START(__memcpy_generic) 2: move a0, a3 jr ra SYM_FUNC_END(__memcpy_generic) +_ASM_NOKPROBE(__memcpy_generic) /* * void *__memcpy_fast(void *dst, const void *src, size_t n) @@ -93,3 +95,4 @@ SYM_FUNC_START(__memcpy_fast) 3: move a0, a3 jr ra SYM_FUNC_END(__memcpy_fast) +_ASM_NOKPROBE(__memcpy_fast) diff --git a/arch/loongarch/lib/memmove.S b/arch/loongarch/lib/memmove.S index 6ffdb46da78fd..b796c3d6da052 100644 --- a/arch/loongarch/lib/memmove.S +++ b/arch/loongarch/lib/memmove.S @@ -29,6 +29,7 @@ SYM_FUNC_START(memmove) b rmemcpy 4: b __rmemcpy_generic SYM_FUNC_END(memmove) +_ASM_NOKPROBE(memmove) EXPORT_SYMBOL(memmove) @@ -39,6 +40,7 @@ SYM_FUNC_START(rmemcpy) ALTERNATIVE "b __rmemcpy_generic", \ "b __rmemcpy_fast", CPU_FEATURE_UAL SYM_FUNC_END(rmemcpy) +_ASM_NOKPROBE(rmemcpy) /* * void *__rmemcpy_generic(void *dst, const void *src, size_t n) @@ -64,6 +66,7 @@ SYM_FUNC_START(__rmemcpy_generic) 2: move a0, a3 jr ra SYM_FUNC_END(__rmemcpy_generic) +_ASM_NOKPROBE(__rmemcpy_generic) /* * void *__rmemcpy_fast(void *dst, const void *src, size_t n) @@ -119,3 +122,4 @@ SYM_FUNC_START(__rmemcpy_fast) 3: move a0, a3 jr ra SYM_FUNC_END(__rmemcpy_fast) +_ASM_NOKPROBE(__rmemcpy_fast) diff --git a/arch/loongarch/lib/memset.S b/arch/loongarch/lib/memset.S index e7cb4ea3747d7..a9eb732ab2adb 100644 --- a/arch/loongarch/lib/memset.S +++ b/arch/loongarch/lib/memset.S @@ -23,6 +23,7 @@ SYM_FUNC_START(memset) ALTERNATIVE "b __memset_generic", \ "b __memset_fast", CPU_FEATURE_UAL SYM_FUNC_END(memset) +_ASM_NOKPROBE(memset) EXPORT_SYMBOL(memset) @@ -45,6 +46,7 @@ SYM_FUNC_START(__memset_generic) 2: move a0, a3 jr ra SYM_FUNC_END(__memset_generic) +_ASM_NOKPROBE(__memset_generic) /* * void *__memset_fast(void *s, int c, size_t n) @@ -89,3 +91,4 @@ SYM_FUNC_START(__memset_fast) 3: move a0, a3 jr ra SYM_FUNC_END(__memset_fast) +_ASM_NOKPROBE(__memset_fast) From a6484baa3b02b9d71bfd6016e7632dd1c02d5cba Mon Sep 17 00:00:00 2001 From: Tiezhu Yang Date: Sat, 25 Feb 2023 15:52:57 +0800 Subject: [PATCH 505/765] samples/kprobes: Add LoongArch support Add LoongArch specific info in handler_pre() and handler_post(). Tested-by: Jeff Xie Signed-off-by: Tiezhu Yang Signed-off-by: Huacai Chen --- samples/kprobes/kprobe_example.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/samples/kprobes/kprobe_example.c b/samples/kprobes/kprobe_example.c index fd346f58ddbaf..ef44c614c6d90 100644 --- a/samples/kprobes/kprobe_example.c +++ b/samples/kprobes/kprobe_example.c @@ -55,6 +55,10 @@ static int __kprobes handler_pre(struct kprobe *p, struct pt_regs *regs) pr_info("<%s> p->addr, 0x%p, ip = 0x%lx, flags = 0x%lx\n", p->symbol_name, p->addr, regs->psw.addr, regs->flags); #endif +#ifdef CONFIG_LOONGARCH + pr_info("<%s> p->addr = 0x%p, era = 0x%lx, estat = 0x%lx\n", + p->symbol_name, p->addr, regs->csr_era, regs->csr_estat); +#endif /* A dump_stack() here will give a stack backtrace */ return 0; @@ -92,6 +96,10 @@ static void __kprobes handler_post(struct kprobe *p, struct pt_regs *regs, pr_info("<%s> p->addr, 0x%p, flags = 0x%lx\n", p->symbol_name, p->addr, regs->flags); #endif +#ifdef CONFIG_LOONGARCH + pr_info("<%s> p->addr = 0x%p, estat = 0x%lx\n", + p->symbol_name, p->addr, regs->csr_estat); +#endif } static int __init kprobe_init(void) From 121ff07bdee06555835e26499f5c125223a6beb3 Mon Sep 17 00:00:00 2001 From: Huacai Chen Date: Sat, 25 Feb 2023 15:52:57 +0800 Subject: [PATCH 506/765] tools: Add LoongArch build infrastructure We will add tools support for LoongArch (bpf, perf, objtool, etc.), add build infrastructure and common headers for preparation. Signed-off-by: Huacai Chen --- tools/arch/loongarch/include/uapi/asm/bitsperlong.h | 9 +++++++++ tools/scripts/Makefile.arch | 11 ++++++++++- 2 files changed, 19 insertions(+), 1 deletion(-) create mode 100644 tools/arch/loongarch/include/uapi/asm/bitsperlong.h diff --git a/tools/arch/loongarch/include/uapi/asm/bitsperlong.h b/tools/arch/loongarch/include/uapi/asm/bitsperlong.h new file mode 100644 index 0000000000000..d4e32b3d48437 --- /dev/null +++ b/tools/arch/loongarch/include/uapi/asm/bitsperlong.h @@ -0,0 +1,9 @@ +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ +#ifndef __ASM_LOONGARCH_BITSPERLONG_H +#define __ASM_LOONGARCH_BITSPERLONG_H + +#define __BITS_PER_LONG (__SIZEOF_POINTER__ * 8) + +#include + +#endif /* __ASM_LOONGARCH_BITSPERLONG_H */ diff --git a/tools/scripts/Makefile.arch b/tools/scripts/Makefile.arch index 0c6c7f4568878..1c72d07cb9fed 100644 --- a/tools/scripts/Makefile.arch +++ b/tools/scripts/Makefile.arch @@ -5,7 +5,7 @@ HOSTARCH := $(shell uname -m | sed -e s/i.86/x86/ -e s/x86_64/x86/ \ -e s/s390x/s390/ -e s/parisc64/parisc/ \ -e s/ppc.*/powerpc/ -e s/mips.*/mips/ \ -e s/sh[234].*/sh/ -e s/aarch64.*/arm64/ \ - -e s/riscv.*/riscv/) + -e s/riscv.*/riscv/ -e s/loongarch.*/loongarch/) ifndef ARCH ARCH := $(HOSTARCH) @@ -34,6 +34,15 @@ ifeq ($(ARCH),sh64) SRCARCH := sh endif +# Additional ARCH settings for loongarch +ifeq ($(ARCH),loongarch32) + SRCARCH := loongarch +endif + +ifeq ($(ARCH),loongarch64) + SRCARCH := loongarch +endif + LP64 := $(shell echo __LP64__ | ${CC} ${CFLAGS} -E -x c - | tail -n 1) ifeq ($(LP64), 1) IS_64_BIT := 1 From eb4071b988397144a35c4aa98061f587cbb3b467 Mon Sep 17 00:00:00 2001 From: Huacai Chen Date: Sat, 25 Feb 2023 15:52:57 +0800 Subject: [PATCH 507/765] selftests/seccomp: Add LoongArch selftesting support BPF for LoongArch is supported now, add the selftesting support in seccomp_bpf.c. Signed-off-by: Tiezhu Yang Signed-off-by: Huacai Chen --- tools/testing/selftests/seccomp/seccomp_bpf.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c index 9c2f448bb3a9d..3933f34b6b050 100644 --- a/tools/testing/selftests/seccomp/seccomp_bpf.c +++ b/tools/testing/selftests/seccomp/seccomp_bpf.c @@ -128,6 +128,8 @@ struct seccomp_data { # define __NR_seccomp 277 # elif defined(__csky__) # define __NR_seccomp 277 +# elif defined(__loongarch__) +# define __NR_seccomp 277 # elif defined(__hppa__) # define __NR_seccomp 338 # elif defined(__powerpc__) @@ -1753,6 +1755,10 @@ TEST_F(TRACE_poke, getpid_runs_normally) NT_ARM_SYSTEM_CALL, &__v)); \ } while (0) # define SYSCALL_RET(_regs) (_regs).regs[0] +#elif defined(__loongarch__) +# define ARCH_REGS struct user_pt_regs +# define SYSCALL_NUM(_regs) (_regs).regs[11] +# define SYSCALL_RET(_regs) (_regs).regs[4] #elif defined(__riscv) && __riscv_xlen == 64 # define ARCH_REGS struct user_regs_struct # define SYSCALL_NUM(_regs) (_regs).a7 From 8883bf83127d533abb415b204eabc064863ae6c9 Mon Sep 17 00:00:00 2001 From: Qing Zhang Date: Sat, 25 Feb 2023 15:52:57 +0800 Subject: [PATCH 508/765] selftests/ftrace: Add LoongArch kprobe args string tests support Before: [5] Kprobe event string type argument [UNTESTED] [7] Kprobe event argument syntax [UNTESTED] After: [5] Kprobe event string type argument [PASS] [7] Kprobe event argument syntax [PASS] Signed-off-by: Qing Zhang Signed-off-by: Huacai Chen --- .../selftests/ftrace/test.d/kprobe/kprobe_args_string.tc | 3 +++ .../selftests/ftrace/test.d/kprobe/kprobe_args_syntax.tc | 4 ++++ 2 files changed, 7 insertions(+) diff --git a/tools/testing/selftests/ftrace/test.d/kprobe/kprobe_args_string.tc b/tools/testing/selftests/ftrace/test.d/kprobe/kprobe_args_string.tc index 4597415652221..a4f8e7c53c1f7 100644 --- a/tools/testing/selftests/ftrace/test.d/kprobe/kprobe_args_string.tc +++ b/tools/testing/selftests/ftrace/test.d/kprobe/kprobe_args_string.tc @@ -28,6 +28,9 @@ s390*) mips*) ARG1=%r4 ;; +loongarch*) + ARG1=%r4 +;; *) echo "Please implement other architecture here" exit_untested diff --git a/tools/testing/selftests/ftrace/test.d/kprobe/kprobe_args_syntax.tc b/tools/testing/selftests/ftrace/test.d/kprobe/kprobe_args_syntax.tc index d4662c8cf407f..1df61e13a812b 100644 --- a/tools/testing/selftests/ftrace/test.d/kprobe/kprobe_args_syntax.tc +++ b/tools/testing/selftests/ftrace/test.d/kprobe/kprobe_args_syntax.tc @@ -40,6 +40,10 @@ mips*) GOODREG=%r4 BADREG=%r12 ;; +loongarch*) + GOODREG=%r4 + BADREG=%r12 +;; *) echo "Please implement other architecture here" exit_untested From 180da5ef84fba0a8ea06870a38d6d657d689944a Mon Sep 17 00:00:00 2001 From: Peter Foley Date: Fri, 13 Jan 2023 12:29:36 -0500 Subject: [PATCH 509/765] scripts: coccicheck: Avoid warning about spurious escape e.g. grep: warning: stray \ before - Signed-off-by: Peter Foley Signed-off-by: Julia Lawall --- scripts/coccicheck | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/coccicheck b/scripts/coccicheck index 2956fce8fa4fc..fb492f032c5fe 100755 --- a/scripts/coccicheck +++ b/scripts/coccicheck @@ -18,7 +18,7 @@ fi SPATCH_VERSION=$($SPATCH --version | head -1 | awk '{print $3}') USE_JOBS="no" -$SPATCH --help | grep "\-\-jobs" > /dev/null && USE_JOBS="yes" +$SPATCH --help | grep -e "--jobs" > /dev/null && USE_JOBS="yes" # The verbosity may be set by the environmental parameter V= # as for example with 'make V=1 coccicheck' From 2b2d50bdd258f8ef37c25f7d5f0d7ba694066dec Mon Sep 17 00:00:00 2001 From: Jarkko Sakkinen Date: Thu, 26 Jan 2023 21:56:12 +0000 Subject: [PATCH 510/765] scripts: coccicheck: Use /usr/bin/env If bash is not located under /bin, coccicheck fails to run. In the real world, this happens for instance when NixOS is used in the host. Instead, use /usr/bin/env to locate the executable binary for bash. Signed-off-by: Jarkko Sakkinen Tested-by: Deepak R Varma Signed-off-by: Julia Lawall --- scripts/coccicheck | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/coccicheck b/scripts/coccicheck index fb492f032c5fe..e52cb43fede60 100755 --- a/scripts/coccicheck +++ b/scripts/coccicheck @@ -1,4 +1,4 @@ -#!/bin/bash +#!/usr/bin/env bash # SPDX-License-Identifier: GPL-2.0 # Linux kernel coccicheck # From 172e344e6f82dc266cb65a69f4bed03428ea8a05 Mon Sep 17 00:00:00 2001 From: Ye Bin Date: Thu, 19 Jan 2023 09:37:11 +0800 Subject: [PATCH 511/765] ext4: init error handle resource before init group descriptors Now, 's_err_report' timer is init after ext4_group_desc_init() when fill super. Theoretically, ext4_group_desc_init() may access to error handle as follows: __ext4_fill_super ext4_group_desc_init ext4_check_descriptors ext4_get_group_desc ext4_error ext4_handle_error ext4_commit_super ext4_update_super if (!es->s_error_count) mod_timer(&sbi->s_err_report, jiffies + 24*60*60*HZ); --> Accessing Uninitialized Variables timer_setup(&sbi->s_err_report, print_daily_error_info, 0); Maybe above issue is just theoretical, as ext4_check_descriptors() didn't judge 'gpd' which get from ext4_get_group_desc(), if access to error handle ext4_get_group_desc() will return NULL, then will trigger null-ptr-deref in ext4_check_descriptors(). However, from the perspective of pure code, it is better to initialize resource that may need to be used first. Signed-off-by: Ye Bin Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20230119013711.86680-1-yebin@huaweicloud.com Signed-off-by: Theodore Ts'o --- fs/ext4/super.c | 19 +++++++------------ 1 file changed, 7 insertions(+), 12 deletions(-) diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 45bcfd35e5591..143e0b573ca58 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -4738,7 +4738,6 @@ static int ext4_group_desc_init(struct super_block *sb, struct ext4_sb_info *sbi = EXT4_SB(sb); unsigned int db_count; ext4_fsblk_t block; - int ret; int i; db_count = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) / @@ -4778,8 +4777,7 @@ static int ext4_group_desc_init(struct super_block *sb, ext4_msg(sb, KERN_ERR, "can't read group descriptor %d", i); sbi->s_gdb_count = i; - ret = PTR_ERR(bh); - goto out; + return PTR_ERR(bh); } rcu_read_lock(); rcu_dereference(sbi->s_group_desc)[i] = bh; @@ -4788,13 +4786,10 @@ static int ext4_group_desc_init(struct super_block *sb, sbi->s_gdb_count = db_count; if (!ext4_check_descriptors(sb, logical_sb_block, first_not_zeroed)) { ext4_msg(sb, KERN_ERR, "group descriptors corrupted!"); - ret = -EFSCORRUPTED; - goto out; + return -EFSCORRUPTED; } + return 0; -out: - ext4_group_desc_free(sbi); - return ret; } static int ext4_load_and_init_journal(struct super_block *sb, @@ -5220,14 +5215,14 @@ static int __ext4_fill_super(struct fs_context *fc, struct super_block *sb) if (ext4_geometry_check(sb, es)) goto failed_mount; - err = ext4_group_desc_init(sb, es, logical_sb_block, &first_not_zeroed); - if (err) - goto failed_mount; - timer_setup(&sbi->s_err_report, print_daily_error_info, 0); spin_lock_init(&sbi->s_error_lock); INIT_WORK(&sbi->s_error_work, flush_stashed_error_work); + err = ext4_group_desc_init(sb, es, logical_sb_block, &first_not_zeroed); + if (err) + goto failed_mount3; + /* Register extent status tree shrinker */ if (ext4_es_register_shrinker(sbi)) goto failed_mount3; From 0813299c586b175d7edb25f56412c54b812d0379 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 26 Jan 2023 12:22:21 +0100 Subject: [PATCH 512/765] ext4: Fix possible corruption when moving a directory When we are renaming a directory to a different directory, we need to update '..' entry in the moved directory. However nothing prevents moved directory from being modified and even converted from the inline format to the normal format. When such race happens the rename code gets confused and we crash. Fix the problem by locking the moved directory. CC: stable@vger.kernel.org Fixes: 32f7f22c0b52 ("ext4: let ext4_rename handle inline dir") Signed-off-by: Jan Kara Link: https://lore.kernel.org/r/20230126112221.11866-1-jack@suse.cz Signed-off-by: Theodore Ts'o --- fs/ext4/namei.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index dd28453d6ea32..270fbcba75b6a 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -3872,9 +3872,16 @@ static int ext4_rename(struct user_namespace *mnt_userns, struct inode *old_dir, if (new.dir != old.dir && EXT4_DIR_LINK_MAX(new.dir)) goto end_rename; } + /* + * We need to protect against old.inode directory getting + * converted from inline directory format into a normal one. + */ + inode_lock_nested(old.inode, I_MUTEX_NONDIR2); retval = ext4_rename_dir_prepare(handle, &old); - if (retval) + if (retval) { + inode_unlock(old.inode); goto end_rename; + } } /* * If we're renaming a file within an inline_data dir and adding or @@ -4006,6 +4013,8 @@ static int ext4_rename(struct user_namespace *mnt_userns, struct inode *old_dir, } else { ext4_journal_stop(handle); } + if (old.dir_bh) + inode_unlock(old.inode); release_bh: brelse(old.dir_bh); brelse(old.bh); From e3645d72f8865ffe36f9dc811540d40aa3c848d3 Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Sun, 29 Jan 2023 11:49:39 +0800 Subject: [PATCH 513/765] ext4: fix incorrect options show of original mount_opt and extend mount_opt2 Current _ext4_show_options() do not distinguish MOPT_2 flag, so it mixed extend sbi->s_mount_opt2 options with sbi->s_mount_opt, it could lead to show incorrect options, e.g. show fc_debug_force if we mount with errors=continue mode and miss it if we set. $ mkfs.ext4 /dev/pmem0 $ mount -o errors=remount-ro /dev/pmem0 /mnt $ cat /proc/fs/ext4/pmem0/options | grep fc_debug_force #empty $ mount -o remount,errors=continue /mnt $ cat /proc/fs/ext4/pmem0/options | grep fc_debug_force fc_debug_force $ mount -o remount,errors=remount-ro,fc_debug_force /mnt $ cat /proc/fs/ext4/pmem0/options | grep fc_debug_force #empty Fixes: 995a3ed67fc8 ("ext4: add fast_commit feature and handling for extended mount options") Signed-off-by: Zhang Yi Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20230129034939.3702550-1-yi.zhang@huaweicloud.com Signed-off-by: Theodore Ts'o --- fs/ext4/ext4.h | 1 + fs/ext4/super.c | 28 +++++++++++++++++++++------- 2 files changed, 22 insertions(+), 7 deletions(-) diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 140e1eb300d17..6479146140d20 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -1529,6 +1529,7 @@ struct ext4_sb_info { unsigned int s_mount_opt2; unsigned long s_mount_flags; unsigned int s_def_mount_opt; + unsigned int s_def_mount_opt2; ext4_fsblk_t s_sb_block; atomic64_t s_resv_clusters; kuid_t s_resuid; diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 143e0b573ca58..2192b41114427 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -2894,7 +2894,7 @@ static int _ext4_show_options(struct seq_file *seq, struct super_block *sb, { struct ext4_sb_info *sbi = EXT4_SB(sb); struct ext4_super_block *es = sbi->s_es; - int def_errors, def_mount_opt = sbi->s_def_mount_opt; + int def_errors; const struct mount_opts *m; char sep = nodefs ? '\n' : ','; @@ -2906,15 +2906,28 @@ static int _ext4_show_options(struct seq_file *seq, struct super_block *sb, for (m = ext4_mount_opts; m->token != Opt_err; m++) { int want_set = m->flags & MOPT_SET; + int opt_2 = m->flags & MOPT_2; + unsigned int mount_opt, def_mount_opt; + if (((m->flags & (MOPT_SET|MOPT_CLEAR)) == 0) || m->flags & MOPT_SKIP) continue; - if (!nodefs && !(m->mount_opt & (sbi->s_mount_opt ^ def_mount_opt))) - continue; /* skip if same as the default */ + + if (opt_2) { + mount_opt = sbi->s_mount_opt2; + def_mount_opt = sbi->s_def_mount_opt2; + } else { + mount_opt = sbi->s_mount_opt; + def_mount_opt = sbi->s_def_mount_opt; + } + /* skip if same as the default */ + if (!nodefs && !(m->mount_opt & (mount_opt ^ def_mount_opt))) + continue; + /* select Opt_noFoo vs Opt_Foo */ if ((want_set && - (sbi->s_mount_opt & m->mount_opt) != m->mount_opt) || - (!want_set && (sbi->s_mount_opt & m->mount_opt))) - continue; /* select Opt_noFoo vs Opt_Foo */ + (mount_opt & m->mount_opt) != m->mount_opt) || + (!want_set && (mount_opt & m->mount_opt))) + continue; SEQ_OPTS_PRINT("%s", token2str(m->token)); } @@ -2942,7 +2955,7 @@ static int _ext4_show_options(struct seq_file *seq, struct super_block *sb, if (nodefs || sbi->s_stripe) SEQ_OPTS_PRINT("stripe=%lu", sbi->s_stripe); if (nodefs || EXT4_MOUNT_DATA_FLAGS & - (sbi->s_mount_opt ^ def_mount_opt)) { + (sbi->s_mount_opt ^ sbi->s_def_mount_opt)) { if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) SEQ_OPTS_PUTS("data=journal"); else if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA) @@ -5081,6 +5094,7 @@ static int __ext4_fill_super(struct fs_context *fc, struct super_block *sb) goto failed_mount; sbi->s_def_mount_opt = sbi->s_mount_opt; + sbi->s_def_mount_opt2 = sbi->s_mount_opt2; err = ext4_check_opt_consistency(fc, sb); if (err < 0) From f195c470f2c2cf3737b2b157a5a2dbe182e374fa Mon Sep 17 00:00:00 2001 From: Jack Chen Date: Thu, 5 Jan 2023 16:29:52 -0500 Subject: [PATCH 514/765] i3c: transfer pid from boardinfo to device info I3C device PID could be defined in device tree and stored in i3c_dev_boardinfo. It should be passed to i3c_device_info when allocating a i3c_dev_desc. Rational behind this change is: when users decide to use SETDASA to assign a dynamic address with exactly the original static address, in step of i3c_master_reattach_i3c_dev, this address is checked to be taken. Then device information retrieving step is skipped. As a result, though the i3c device is registered correctly, its device driver could not be probed. Tested: Tested with a I3C device. If assigned-address is set to be the device's static address, without this change, its device driver could not probed. And with this change, its driver is probed successfully. Signed-off-by: Jack Chen Link: https://lore.kernel.org/r/20230105212952.56321-1-zenghuchen@google.com Signed-off-by: Alexandre Belloni --- drivers/i3c/master.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/i3c/master.c b/drivers/i3c/master.c index d7e6f6c99aea5..4dad80338831d 100644 --- a/drivers/i3c/master.c +++ b/drivers/i3c/master.c @@ -1438,6 +1438,7 @@ static int i3c_master_early_i3c_dev_add(struct i3c_master_controller *master, { struct i3c_device_info info = { .static_addr = boardinfo->static_addr, + .pid = boardinfo->pid, }; struct i3c_dev_desc *i3cdev; int ret; From 430aa33a304654704c16ed7bbbcbec9e9686aa81 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:42 +0100 Subject: [PATCH 515/765] dt-bindings: rtc: qcom-pm8xxx: add nvmem-cell offset On many Qualcomm platforms the PMIC RTC control and time registers are read-only so that the RTC time can not be updated. Instead an offset needs be stored in some machine-specific non-volatile memory, which a driver can take into account. Add an 'offset' nvmem cell which can be used to store a 32-bit offset from the Unix epoch so that the RTC time can be updated on such platforms. Acked-by: Krzysztof Kozlowski Signed-off-by: Johan Hovold Reviewed-by: David Collins Link: https://lore.kernel.org/r/20230202155448.6715-17-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- .../devicetree/bindings/rtc/qcom-pm8xxx-rtc.yaml | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/Documentation/devicetree/bindings/rtc/qcom-pm8xxx-rtc.yaml b/Documentation/devicetree/bindings/rtc/qcom-pm8xxx-rtc.yaml index 21c8ea08ff0a2..b95a69cc9ae0f 100644 --- a/Documentation/devicetree/bindings/rtc/qcom-pm8xxx-rtc.yaml +++ b/Documentation/devicetree/bindings/rtc/qcom-pm8xxx-rtc.yaml @@ -40,6 +40,16 @@ properties: description: Indicates that the setting of RTC time is allowed by the host CPU. + nvmem-cells: + items: + - description: + four-byte nvmem cell holding a little-endian offset from the Unix + epoch representing the time when the RTC timer was last reset + + nvmem-cell-names: + items: + - const: offset + wakeup-source: true required: @@ -69,6 +79,8 @@ examples: compatible = "qcom,pm8921-rtc"; reg = <0x11d>; interrupts = <0x27 0>; + nvmem-cells = <&rtc_offset>; + nvmem-cell-names = "offset"; }; }; }; From 3ca04951b004fa184ff84369448a37bf5df98a79 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 2 Feb 2023 16:54:43 +0100 Subject: [PATCH 516/765] rtc: pm8xxx: add support for nvmem offset On many Qualcomm platforms the PMIC RTC control and time registers are read-only so that the RTC time can not be updated. Instead an offset needs be stored in some machine-specific non-volatile memory, which the driver can take into account. Add support for storing a 32-bit offset from the Epoch in an nvmem cell so that the RTC time can be set on such platforms. Signed-off-by: Johan Hovold Link: https://lore.kernel.org/r/20230202155448.6715-18-johan+linaro@kernel.org Signed-off-by: Alexandre Belloni --- drivers/rtc/rtc-pm8xxx.c | 141 +++++++++++++++++++++++++++++++++++---- 1 file changed, 129 insertions(+), 12 deletions(-) diff --git a/drivers/rtc/rtc-pm8xxx.c b/drivers/rtc/rtc-pm8xxx.c index eff2782beeed6..372494e82f405 100644 --- a/drivers/rtc/rtc-pm8xxx.c +++ b/drivers/rtc/rtc-pm8xxx.c @@ -1,8 +1,13 @@ // SPDX-License-Identifier: GPL-2.0-only -/* Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved. +/* + * pm8xxx RTC driver + * + * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved. + * Copyright (c) 2023, Linaro Limited */ #include #include +#include #include #include #include @@ -49,6 +54,8 @@ struct pm8xxx_rtc_regs { * @alarm_irq: alarm irq number * @regs: register description * @dev: device structure + * @nvmem_cell: nvmem cell for offset + * @offset: offset from epoch in seconds */ struct pm8xxx_rtc { struct rtc_device *rtc; @@ -57,8 +64,60 @@ struct pm8xxx_rtc { int alarm_irq; const struct pm8xxx_rtc_regs *regs; struct device *dev; + struct nvmem_cell *nvmem_cell; + u32 offset; }; +static int pm8xxx_rtc_read_nvmem_offset(struct pm8xxx_rtc *rtc_dd) +{ + size_t len; + void *buf; + int rc; + + buf = nvmem_cell_read(rtc_dd->nvmem_cell, &len); + if (IS_ERR(buf)) { + rc = PTR_ERR(buf); + dev_dbg(rtc_dd->dev, "failed to read nvmem offset: %d\n", rc); + return rc; + } + + if (len != sizeof(u32)) { + dev_dbg(rtc_dd->dev, "unexpected nvmem cell size %zu\n", len); + kfree(buf); + return -EINVAL; + } + + rtc_dd->offset = get_unaligned_le32(buf); + + kfree(buf); + + return 0; +} + +static int pm8xxx_rtc_write_nvmem_offset(struct pm8xxx_rtc *rtc_dd, u32 offset) +{ + u8 buf[sizeof(u32)]; + int rc; + + put_unaligned_le32(offset, buf); + + rc = nvmem_cell_write(rtc_dd->nvmem_cell, buf, sizeof(buf)); + if (rc < 0) { + dev_dbg(rtc_dd->dev, "failed to write nvmem offset: %d\n", rc); + return rc; + } + + return 0; +} + +static int pm8xxx_rtc_read_offset(struct pm8xxx_rtc *rtc_dd) +{ + if (!rtc_dd->nvmem_cell) + return 0; + + return pm8xxx_rtc_read_nvmem_offset(rtc_dd); +} + static int pm8xxx_rtc_read_raw(struct pm8xxx_rtc *rtc_dd, u32 *secs) { const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; @@ -90,6 +149,33 @@ static int pm8xxx_rtc_read_raw(struct pm8xxx_rtc *rtc_dd, u32 *secs) return 0; } +static int pm8xxx_rtc_update_offset(struct pm8xxx_rtc *rtc_dd, u32 secs) +{ + u32 raw_secs; + u32 offset; + int rc; + + if (!rtc_dd->nvmem_cell) + return -ENODEV; + + rc = pm8xxx_rtc_read_raw(rtc_dd, &raw_secs); + if (rc) + return rc; + + offset = secs - raw_secs; + + if (offset == rtc_dd->offset) + return 0; + + rc = pm8xxx_rtc_write_nvmem_offset(rtc_dd, offset); + if (rc) + return rc; + + rtc_dd->offset = offset; + + return 0; +} + /* * Steps to write the RTC registers. * 1. Disable alarm if enabled. @@ -99,23 +185,15 @@ static int pm8xxx_rtc_read_raw(struct pm8xxx_rtc *rtc_dd, u32 *secs) * 5. Enable rtc if disabled in step 2. * 6. Enable alarm if disabled in step 1. */ -static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) +static int __pm8xxx_rtc_set_time(struct pm8xxx_rtc *rtc_dd, u32 secs) { - struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); const struct pm8xxx_rtc_regs *regs = rtc_dd->regs; u8 value[NUM_8_BIT_RTC_REGS]; bool alarm_enabled; - u32 secs; int rc; - if (!rtc_dd->allow_set_time) - return -ENODEV; - - secs = rtc_tm_to_time64(tm); put_unaligned_le32(secs, value); - dev_dbg(dev, "set time: %ptRd %ptRt (%u)\n", tm, tm, secs); - rc = regmap_update_bits_check(rtc_dd->regmap, regs->alarm_ctrl, regs->alarm_en, 0, &alarm_enabled); if (rc) @@ -158,6 +236,27 @@ static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) return 0; } +static int pm8xxx_rtc_set_time(struct device *dev, struct rtc_time *tm) +{ + struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); + u32 secs; + int rc; + + secs = rtc_tm_to_time64(tm); + + if (rtc_dd->allow_set_time) + rc = __pm8xxx_rtc_set_time(rtc_dd, secs); + else + rc = pm8xxx_rtc_update_offset(rtc_dd, secs); + + if (rc) + return rc; + + dev_dbg(dev, "set time: %ptRd %ptRt (%u + %u)\n", tm, tm, + secs - rtc_dd->offset, rtc_dd->offset); + return 0; +} + static int pm8xxx_rtc_read_time(struct device *dev, struct rtc_time *tm) { struct pm8xxx_rtc *rtc_dd = dev_get_drvdata(dev); @@ -168,10 +267,11 @@ static int pm8xxx_rtc_read_time(struct device *dev, struct rtc_time *tm) if (rc) return rc; + secs += rtc_dd->offset; rtc_time64_to_tm(secs, tm); - dev_dbg(dev, "read time: %ptRd %ptRt (%u)\n", tm, tm, secs); - + dev_dbg(dev, "read time: %ptRd %ptRt (%u + %u)\n", tm, tm, + secs - rtc_dd->offset, rtc_dd->offset); return 0; } @@ -184,6 +284,7 @@ static int pm8xxx_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alarm) int rc; secs = rtc_tm_to_time64(&alarm->time); + secs -= rtc_dd->offset; put_unaligned_le32(secs, value); rc = regmap_update_bits(rtc_dd->regmap, regs->alarm_ctrl, @@ -223,6 +324,7 @@ static int pm8xxx_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alarm) return rc; secs = get_unaligned_le32(value); + secs += rtc_dd->offset; rtc_time64_to_tm(secs, &alarm->time); rc = regmap_read(rtc_dd->regmap, regs->alarm_ctrl, &ctrl_reg); @@ -378,9 +480,23 @@ static int pm8xxx_rtc_probe(struct platform_device *pdev) rtc_dd->allow_set_time = of_property_read_bool(pdev->dev.of_node, "allow-set-time"); + rtc_dd->nvmem_cell = devm_nvmem_cell_get(&pdev->dev, "offset"); + if (IS_ERR(rtc_dd->nvmem_cell)) { + rc = PTR_ERR(rtc_dd->nvmem_cell); + if (rc != -ENOENT) + return rc; + rtc_dd->nvmem_cell = NULL; + } + rtc_dd->regs = match->data; rtc_dd->dev = &pdev->dev; + if (!rtc_dd->allow_set_time) { + rc = pm8xxx_rtc_read_offset(rtc_dd); + if (rc) + return rc; + } + rc = pm8xxx_rtc_enable(rtc_dd); if (rc) return rc; @@ -435,3 +551,4 @@ MODULE_ALIAS("platform:rtc-pm8xxx"); MODULE_DESCRIPTION("PMIC8xxx RTC driver"); MODULE_LICENSE("GPL v2"); MODULE_AUTHOR("Anirudh Ghayal "); +MODULE_AUTHOR("Johan Hovold "); From c16bda37594f83147b167d381d54c010024efecf Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Sat, 25 Feb 2023 12:53:53 -0700 Subject: [PATCH 517/765] io_uring/poll: allow some retries for poll triggering spuriously If we get woken spuriously when polling and fail the operation with -EAGAIN again, then we generally only allow polling again if data had been transferred at some point. This is indicated with REQ_F_PARTIAL_IO. However, if the spurious poll triggers when the socket was originally empty, then we haven't transferred data yet and we will fail the poll re-arm. This either punts the socket to io-wq if it's blocking, or it fails the request with -EAGAIN if not. Neither condition is desirable, as the former will slow things down, while the latter will make the application confused. We want to ensure that a repeated poll trigger doesn't lead to infinite work making no progress, that's what the REQ_F_PARTIAL_IO check was for. But it doesn't protect against a loop post the first receive, and it's unnecessarily strict if we started out with an empty socket. Add a somewhat random retry count, just to put an upper limit on the potential number of retries that will be done. This should be high enough that we won't really hit it in practice, unless something needs to be aborted anyway. Cc: stable@vger.kernel.org # v5.10+ Link: https://github.com/axboe/liburing/issues/364 Signed-off-by: Jens Axboe --- io_uring/poll.c | 14 ++++++++++++-- io_uring/poll.h | 1 + 2 files changed, 13 insertions(+), 2 deletions(-) diff --git a/io_uring/poll.c b/io_uring/poll.c index 8339a92b45107..a82d6830bdfdd 100644 --- a/io_uring/poll.c +++ b/io_uring/poll.c @@ -650,6 +650,14 @@ static void io_async_queue_proc(struct file *file, struct wait_queue_head *head, __io_queue_proc(&apoll->poll, pt, head, &apoll->double_poll); } +/* + * We can't reliably detect loops in repeated poll triggers and issue + * subsequently failing. But rather than fail these immediately, allow a + * certain amount of retries before we give up. Given that this condition + * should _rarely_ trigger even once, we should be fine with a larger value. + */ +#define APOLL_MAX_RETRY 128 + static struct async_poll *io_req_alloc_apoll(struct io_kiocb *req, unsigned issue_flags) { @@ -665,14 +673,18 @@ static struct async_poll *io_req_alloc_apoll(struct io_kiocb *req, if (entry == NULL) goto alloc_apoll; apoll = container_of(entry, struct async_poll, cache); + apoll->poll.retries = APOLL_MAX_RETRY; } else { alloc_apoll: apoll = kmalloc(sizeof(*apoll), GFP_ATOMIC); if (unlikely(!apoll)) return NULL; + apoll->poll.retries = APOLL_MAX_RETRY; } apoll->double_poll = NULL; req->apoll = apoll; + if (unlikely(!--apoll->poll.retries)) + return NULL; return apoll; } @@ -694,8 +706,6 @@ int io_arm_poll_handler(struct io_kiocb *req, unsigned issue_flags) return IO_APOLL_ABORTED; if (!file_can_poll(req->file)) return IO_APOLL_ABORTED; - if ((req->flags & (REQ_F_POLLED|REQ_F_PARTIAL_IO)) == REQ_F_POLLED) - return IO_APOLL_ABORTED; if (!(req->flags & REQ_F_APOLL_MULTISHOT)) mask |= EPOLLONESHOT; diff --git a/io_uring/poll.h b/io_uring/poll.h index 5f3bae50fc81a..b2393b403a2c2 100644 --- a/io_uring/poll.h +++ b/io_uring/poll.h @@ -12,6 +12,7 @@ struct io_poll { struct file *file; struct wait_queue_head *head; __poll_t events; + int retries; struct wait_queue_entry wait; }; From e027253c4b77d395798600a90b6a96fe4adf4d5e Mon Sep 17 00:00:00 2001 From: Xiubo Li Date: Mon, 13 Feb 2023 13:56:20 +0800 Subject: [PATCH 518/765] ceph: update the time stamps and try to drop the suid/sgid The fallocate will try to clear the suid/sgid if a unprevileged user changed the file. There is no POSIX item requires that we should clear the suid/sgid in fallocate code path but this is the default behaviour for most of the filesystems and the VFS layer. And also the same for the write code path, which have already support it. And also we need to update the time stamps since the fallocate will change the file contents. Cc: stable@vger.kernel.org Link: https://tracker.ceph.com/issues/58054 Signed-off-by: Xiubo Li Reviewed-by: Jeff Layton Signed-off-by: Ilya Dryomov --- fs/ceph/file.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/fs/ceph/file.c b/fs/ceph/file.c index b5cff85925a10..dc39a4b0ec8ec 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -2102,6 +2102,9 @@ static long ceph_fallocate(struct file *file, int mode, loff_t endoff = 0; loff_t size; + dout("%s %p %llx.%llx mode %x, offset %llu length %llu\n", __func__, + inode, ceph_vinop(inode), mode, offset, length); + if (mode != (FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE)) return -EOPNOTSUPP; @@ -2136,6 +2139,10 @@ static long ceph_fallocate(struct file *file, int mode, if (ret < 0) goto unlock; + ret = file_modified(file); + if (ret) + goto put_caps; + filemap_invalidate_lock(inode->i_mapping); ceph_fscache_invalidate(inode, false); ceph_zero_pagecache_range(inode, offset, length); @@ -2151,6 +2158,7 @@ static long ceph_fallocate(struct file *file, int mode, } filemap_invalidate_unlock(inode->i_mapping); +put_caps: ceph_put_cap_refs(ci, got); unlock: inode_unlock(inode); From f7c4d9b133c7a04ca619355574e96b6abf209fba Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Fri, 24 Feb 2023 18:48:54 +0100 Subject: [PATCH 519/765] rbd: avoid use-after-free in do_rbd_add() when rbd_dev_create() fails If getting an ID or setting up a work queue in rbd_dev_create() fails, use-after-free on rbd_dev->rbd_client, rbd_dev->spec and rbd_dev->opts is triggered in do_rbd_add(). The root cause is that the ownership of these structures is transfered to rbd_dev prematurely and they all end up getting freed when rbd_dev_create() calls rbd_dev_free() prior to returning to do_rbd_add(). Found by Linux Verification Center (linuxtesting.org) with SVACE, an incomplete patch submitted by Natalia Petrova . Cc: stable@vger.kernel.org Fixes: 1643dfa4c2c8 ("rbd: introduce a per-device ordered workqueue") Signed-off-by: Ilya Dryomov --- drivers/block/rbd.c | 20 +++++++++----------- 1 file changed, 9 insertions(+), 11 deletions(-) diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c index 04453f4a319cb..60aed196a2e54 100644 --- a/drivers/block/rbd.c +++ b/drivers/block/rbd.c @@ -5292,8 +5292,7 @@ static void rbd_dev_release(struct device *dev) module_put(THIS_MODULE); } -static struct rbd_device *__rbd_dev_create(struct rbd_client *rbdc, - struct rbd_spec *spec) +static struct rbd_device *__rbd_dev_create(struct rbd_spec *spec) { struct rbd_device *rbd_dev; @@ -5338,9 +5337,6 @@ static struct rbd_device *__rbd_dev_create(struct rbd_client *rbdc, rbd_dev->dev.parent = &rbd_root_dev; device_initialize(&rbd_dev->dev); - rbd_dev->rbd_client = rbdc; - rbd_dev->spec = spec; - return rbd_dev; } @@ -5353,12 +5349,10 @@ static struct rbd_device *rbd_dev_create(struct rbd_client *rbdc, { struct rbd_device *rbd_dev; - rbd_dev = __rbd_dev_create(rbdc, spec); + rbd_dev = __rbd_dev_create(spec); if (!rbd_dev) return NULL; - rbd_dev->opts = opts; - /* get an id and fill in device name */ rbd_dev->dev_id = ida_simple_get(&rbd_dev_id_ida, 0, minor_to_rbd_dev_id(1 << MINORBITS), @@ -5375,6 +5369,10 @@ static struct rbd_device *rbd_dev_create(struct rbd_client *rbdc, /* we have a ref from do_rbd_add() */ __module_get(THIS_MODULE); + rbd_dev->rbd_client = rbdc; + rbd_dev->spec = spec; + rbd_dev->opts = opts; + dout("%s rbd_dev %p dev_id %d\n", __func__, rbd_dev, rbd_dev->dev_id); return rbd_dev; @@ -6736,7 +6734,7 @@ static int rbd_dev_probe_parent(struct rbd_device *rbd_dev, int depth) goto out_err; } - parent = __rbd_dev_create(rbd_dev->rbd_client, rbd_dev->parent_spec); + parent = __rbd_dev_create(rbd_dev->parent_spec); if (!parent) { ret = -ENOMEM; goto out_err; @@ -6746,8 +6744,8 @@ static int rbd_dev_probe_parent(struct rbd_device *rbd_dev, int depth) * Images related by parent/child relationships always share * rbd_client and spec/parent_spec, so bump their refcounts. */ - __rbd_get_client(rbd_dev->rbd_client); - rbd_spec_get(rbd_dev->parent_spec); + parent->rbd_client = __rbd_get_client(rbd_dev->rbd_client); + parent->spec = rbd_spec_get(rbd_dev->parent_spec); __set_bit(RBD_DEV_FLAG_READONLY, &parent->flags); From 250870824c1cf199b032b1ef889c8e8d69d9123a Mon Sep 17 00:00:00 2001 From: Michael Karcher Date: Tue, 24 Jan 2023 22:48:16 +0100 Subject: [PATCH 520/765] sh: intc: Avoid spurious sizeof-pointer-div warning GCC warns about the pattern sizeof(void*)/sizeof(void), as it looks like the abuse of a pattern to calculate the array size. This pattern appears in the unevaluated part of the ternary operator in _INTC_ARRAY if the parameter is NULL. The replacement uses an alternate approach to return 0 in case of NULL which does not generate the pattern sizeof(void*)/sizeof(void), but still emits the warning if _INTC_ARRAY is called with a nonarray parameter. This patch is required for successful compilation with -Werror enabled. The idea to use _Generic for type distinction is taken from Comment #7 in https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108483 by Jakub Jelinek Signed-off-by: Michael Karcher Acked-by: Randy Dunlap # build-tested Link: https://lore.kernel.org/r/619fa552-c988-35e5-b1d7-fe256c46a272@mkarcher.dialup.fu-berlin.de Signed-off-by: John Paul Adrian Glaubitz --- include/linux/sh_intc.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/include/linux/sh_intc.h b/include/linux/sh_intc.h index c255273b02810..37ad81058d6ae 100644 --- a/include/linux/sh_intc.h +++ b/include/linux/sh_intc.h @@ -97,7 +97,10 @@ struct intc_hw_desc { unsigned int nr_subgroups; }; -#define _INTC_ARRAY(a) a, __same_type(a, NULL) ? 0 : sizeof(a)/sizeof(*a) +#define _INTC_SIZEOF_OR_ZERO(a) (_Generic(a, \ + typeof(NULL): 0, \ + default: sizeof(a))) +#define _INTC_ARRAY(a) a, _INTC_SIZEOF_OR_ZERO(a)/sizeof(*a) #define INTC_HW_DESC(vectors, groups, mask_regs, \ prio_regs, sense_regs, ack_regs) \ From ff30bd6a6618e979b16977617371c0f28a95036e Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Thu, 2 Feb 2023 17:20:55 +0100 Subject: [PATCH 521/765] sh: clk: Fix clk_enable() to return 0 on NULL clk On SH, devm_clk_get_optional_enabled() fails with -EINVAL if the clock is not found. This happens because __devm_clk_get() assumes it can pass a NULL clock pointer (as returned by clk_get_optional()) to the init() function (clk_prepare_enable() in this case), while the SH implementation of clk_enable() considers that an error. Fix this by making the SH clk_enable() implementation return zero instead, like the Common Clock Framework does. Reported-by: John Paul Adrian Glaubitz Signed-off-by: Geert Uytterhoeven Tested-by: John Paul Adrian Glaubitz Acked-by: Stephen Boyd Link: https://lore.kernel.org/r/b53e6b557b4240579933b3359dda335ff94ed5af.1675354849.git.geert+renesas@glider.be Signed-off-by: John Paul Adrian Glaubitz --- drivers/sh/clk/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/sh/clk/core.c b/drivers/sh/clk/core.c index d996782a71064..7a73f5e4a1fc7 100644 --- a/drivers/sh/clk/core.c +++ b/drivers/sh/clk/core.c @@ -295,7 +295,7 @@ int clk_enable(struct clk *clk) int ret; if (!clk) - return -EINVAL; + return 0; spin_lock_irqsave(&clock_lock, flags); ret = __clk_enable(clk); From dd8314739a1ff8ed081d3a06f5f87045f7384636 Mon Sep 17 00:00:00 2001 From: Jiaxun Yang Date: Wed, 22 Feb 2023 13:24:22 +0000 Subject: [PATCH 522/765] MIPS: Remove DMA_PERDEV_COHERENT MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit As now we are always managing DMA coherence on per dev bias, there is no need to have such option. And it's not selected by any platform. Signed-off-by: Jiaxun Yang Reviewed-by: Philippe Mathieu-Daudé Signed-off-by: Thomas Bogendoerfer --- arch/mips/Kconfig | 5 ----- 1 file changed, 5 deletions(-) diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 37072e15b2634..a1170f0a0c04d 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -1080,11 +1080,6 @@ config FW_CFE config ARCH_SUPPORTS_UPROBES bool -config DMA_PERDEV_COHERENT - bool - select ARCH_HAS_SETUP_DMA_OPS - select DMA_NONCOHERENT - config DMA_NONCOHERENT bool # From 6cb5d1a16a51d080fbc1649a5144cbc5ca7d6f88 Mon Sep 17 00:00:00 2001 From: Yuezhang Mo Date: Thu, 22 Sep 2022 14:43:47 +0800 Subject: [PATCH 523/765] exfat: fix unexpected EOF while reading dir If the position is not aligned with the dentry size, the return value of readdir() will be NULL and errno is 0, which means the end of the directory stream is reached. If the position is aligned with dentry size, but there is no file or directory at the position, exfat_readdir() will continue to get dentry from the next dentry. So the dentry gotten by readdir() may not be at the position. After this commit, if the position is not aligned with the dentry size, round the position up to the dentry size and continue to get the dentry. Fixes: ca06197382bd ("exfat: add directory operations") Cc: stable@vger.kernel.org # v5.7+ Reported-by: Wang Yugui Signed-off-by: Yuezhang Mo Reviewed-by: Andy Wu Reviewed-by: Aoyama Wataru Reviewed-by: Sungjong Seo Signed-off-by: Namjae Jeon --- fs/exfat/dir.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/fs/exfat/dir.c b/fs/exfat/dir.c index 1dfa67f307f17..1122bee3b6344 100644 --- a/fs/exfat/dir.c +++ b/fs/exfat/dir.c @@ -234,10 +234,7 @@ static int exfat_iterate(struct file *file, struct dir_context *ctx) fake_offset = 1; } - if (cpos & (DENTRY_SIZE - 1)) { - err = -ENOENT; - goto unlock; - } + cpos = round_up(cpos, DENTRY_SIZE); /* name buffer should be allocated before use */ err = exfat_alloc_namebuf(nb); From 706fdcac002316893434d753be8cfb549fe1d40d Mon Sep 17 00:00:00 2001 From: Yuezhang Mo Date: Thu, 20 Oct 2022 14:27:37 +0800 Subject: [PATCH 524/765] exfat: fix reporting fs error when reading dir beyond EOF Since seekdir() does not check whether the position is valid, the position may exceed the size of the directory. We found that for a directory with discontinuous clusters, if the position exceeds the size of the directory and the excess size is greater than or equal to the cluster size, exfat_readdir() will return -EIO, causing a file system error and making the file system unavailable. Reproduce this bug by: seekdir(dir, dir_size + cluster_size); dirent = readdir(dir); The following log will be printed if mount with 'errors=remount-ro'. [11166.712896] exFAT-fs (sdb1): error, invalid access to FAT (entry 0xffffffff) [11166.712905] exFAT-fs (sdb1): Filesystem has been set read-only Fixes: 1e5654de0f51 ("exfat: handle wrong stream entry size in exfat_readdir()") Cc: stable@vger.kernel.org # v5.7+ Signed-off-by: Yuezhang Mo Reviewed-by: Andy Wu Reviewed-by: Aoyama Wataru Reviewed-by: Sungjong Seo Signed-off-by: Namjae Jeon --- fs/exfat/dir.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/exfat/dir.c b/fs/exfat/dir.c index 1122bee3b6344..158427e8124e1 100644 --- a/fs/exfat/dir.c +++ b/fs/exfat/dir.c @@ -100,7 +100,7 @@ static int exfat_readdir(struct inode *inode, loff_t *cpos, struct exfat_dir_ent clu.dir = ei->hint_bmap.clu; } - while (clu_offset > 0) { + while (clu_offset > 0 && clu.dir != EXFAT_EOF_CLUSTER) { if (exfat_get_next_cluster(sb, &(clu.dir))) return -EIO; From bdaadfd343e3cba49ad0b009ff4b148dad0fa404 Mon Sep 17 00:00:00 2001 From: Sungjong Seo Date: Thu, 29 Dec 2022 20:52:38 +0900 Subject: [PATCH 525/765] exfat: redefine DIR_DELETED as the bad cluster number When a file or a directory is deleted, the hint for the cluster of its parent directory in its in-memory inode is set as DIR_DELETED. Therefore, DIR_DELETED must be one of invalid cluster numbers. According to the exFAT specification, a volume can have at most 2^32-11 clusters. However, DIR_DELETED is wrongly defined as 0xFFFF0321, which could be a valid cluster number. To fix it, let's redefine DIR_DELETED as 0xFFFFFFF7, the bad cluster number. Fixes: 1acf1a564b60 ("exfat: add in-memory and on-disk structures and headers") Cc: stable@vger.kernel.org # v5.7+ Reported-by: Yuezhang Mo Signed-off-by: Sungjong Seo Signed-off-by: Namjae Jeon --- fs/exfat/exfat_fs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/exfat/exfat_fs.h b/fs/exfat/exfat_fs.h index bc6d21d7c5adf..25a5df0fdfe01 100644 --- a/fs/exfat/exfat_fs.h +++ b/fs/exfat/exfat_fs.h @@ -50,7 +50,7 @@ enum { #define ES_IDX_LAST_FILENAME(name_len) \ (ES_IDX_FIRST_FILENAME + EXFAT_FILENAME_ENTRY_NUM(name_len) - 1) -#define DIR_DELETED 0xFFFF0321 +#define DIR_DELETED 0xFFFFFFF7 /* type values */ #define TYPE_UNUSED 0x0000 From 39c1ce8eafc0ff64fb9e28536ccc7df6a8e2999d Mon Sep 17 00:00:00 2001 From: Yuezhang Mo Date: Wed, 4 Jan 2023 14:37:47 +0800 Subject: [PATCH 526/765] exfat: fix inode->i_blocks for non-512 byte sector size device inode->i_blocks is not real number of blocks, but 512 byte ones. Fixes: 98d917047e8b ("exfat: add file operations") Cc: stable@vger.kernel.org # v5.7+ Reported-by: Wang Yugui Tested-by: Wang Yugui Signed-off-by: Yuezhang Mo Reviewed-by: Andy Wu Signed-off-by: Namjae Jeon --- fs/exfat/file.c | 3 +-- fs/exfat/inode.c | 6 ++---- fs/exfat/namei.c | 2 +- fs/exfat/super.c | 3 +-- 4 files changed, 5 insertions(+), 9 deletions(-) diff --git a/fs/exfat/file.c b/fs/exfat/file.c index f5b29072775de..b33431c74c8af 100644 --- a/fs/exfat/file.c +++ b/fs/exfat/file.c @@ -209,8 +209,7 @@ void exfat_truncate(struct inode *inode) if (err) goto write_size; - inode->i_blocks = round_up(i_size_read(inode), sbi->cluster_size) >> - inode->i_blkbits; + inode->i_blocks = round_up(i_size_read(inode), sbi->cluster_size) >> 9; write_size: aligned_size = i_size_read(inode); if (aligned_size & (blocksize - 1)) { diff --git a/fs/exfat/inode.c b/fs/exfat/inode.c index 5b644cb057fa8..481dd338f2b8e 100644 --- a/fs/exfat/inode.c +++ b/fs/exfat/inode.c @@ -220,8 +220,7 @@ static int exfat_map_cluster(struct inode *inode, unsigned int clu_offset, num_clusters += num_to_be_allocated; *clu = new_clu.dir; - inode->i_blocks += - num_to_be_allocated << sbi->sect_per_clus_bits; + inode->i_blocks += EXFAT_CLU_TO_B(num_to_be_allocated, sbi) >> 9; /* * Move *clu pointer along FAT chains (hole care) because the @@ -576,8 +575,7 @@ static int exfat_fill_inode(struct inode *inode, struct exfat_dir_entry *info) exfat_save_attr(inode, info->attr); - inode->i_blocks = round_up(i_size_read(inode), sbi->cluster_size) >> - inode->i_blkbits; + inode->i_blocks = round_up(i_size_read(inode), sbi->cluster_size) >> 9; inode->i_mtime = info->mtime; inode->i_ctime = info->mtime; ei->i_crtime = info->crtime; diff --git a/fs/exfat/namei.c b/fs/exfat/namei.c index 5f995eba5dbbe..7442fead02793 100644 --- a/fs/exfat/namei.c +++ b/fs/exfat/namei.c @@ -396,7 +396,7 @@ static int exfat_find_empty_entry(struct inode *inode, ei->i_size_ondisk += sbi->cluster_size; ei->i_size_aligned += sbi->cluster_size; ei->flags = p_dir->flags; - inode->i_blocks += 1 << sbi->sect_per_clus_bits; + inode->i_blocks += sbi->cluster_size >> 9; } return dentry; diff --git a/fs/exfat/super.c b/fs/exfat/super.c index 35f0305cd493c..8c32460e031e8 100644 --- a/fs/exfat/super.c +++ b/fs/exfat/super.c @@ -373,8 +373,7 @@ static int exfat_read_root(struct inode *inode) inode->i_op = &exfat_dir_inode_operations; inode->i_fop = &exfat_dir_operations; - inode->i_blocks = round_up(i_size_read(inode), sbi->cluster_size) >> - inode->i_blkbits; + inode->i_blocks = round_up(i_size_read(inode), sbi->cluster_size) >> 9; ei->i_pos = ((loff_t)sbi->root_dir << 32) | 0xffffffff; ei->i_size_aligned = i_size_read(inode); ei->i_size_ondisk = i_size_read(inode); From 8258ef28001ad30c074e823124e10b9c75a965ff Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Sat, 14 Jan 2023 13:09:48 +0900 Subject: [PATCH 527/765] exfat: handle unreconized benign secondary entries MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Sony PXW-Z280 camera add vendor allocation entries to directory of pictures. Currently, linux exfat does not support it and the file is not visible. This patch handle vendor extension and allocation entries as unreconized benign secondary entries. As described in the specification, it is recognized but ignored, and when deleting directory entry set, the associated clusters allocation are removed as well as benign secondary directory entries. Reported-by: Barócsi Dénes Reviewed-by: Sungjong Seo Reviewed-by: Yuezhang Mo Signed-off-by: Namjae Jeon --- fs/exfat/dir.c | 83 +++++++++++++++++++++++++++++++------------- fs/exfat/exfat_fs.h | 2 ++ fs/exfat/exfat_raw.h | 21 +++++++++++ 3 files changed, 81 insertions(+), 25 deletions(-) diff --git a/fs/exfat/dir.c b/fs/exfat/dir.c index 158427e8124e1..957574180a5e3 100644 --- a/fs/exfat/dir.c +++ b/fs/exfat/dir.c @@ -29,14 +29,15 @@ static int exfat_extract_uni_name(struct exfat_dentry *ep, } -static void exfat_get_uniname_from_ext_entry(struct super_block *sb, +static int exfat_get_uniname_from_ext_entry(struct super_block *sb, struct exfat_chain *p_dir, int entry, unsigned short *uniname) { - int i; + int i, err; struct exfat_entry_set_cache es; - if (exfat_get_dentry_set(&es, sb, p_dir, entry, ES_ALL_ENTRIES)) - return; + err = exfat_get_dentry_set(&es, sb, p_dir, entry, ES_ALL_ENTRIES); + if (err) + return err; /* * First entry : file entry @@ -56,12 +57,13 @@ static void exfat_get_uniname_from_ext_entry(struct super_block *sb, } exfat_put_dentry_set(&es, false); + return 0; } /* read a directory entry from the opened directory */ static int exfat_readdir(struct inode *inode, loff_t *cpos, struct exfat_dir_entry *dir_entry) { - int i, dentries_per_clu, num_ext; + int i, dentries_per_clu, num_ext, err; unsigned int type, clu_offset, max_dentries; struct exfat_chain dir, clu; struct exfat_uni_name uni_name; @@ -146,8 +148,12 @@ static int exfat_readdir(struct inode *inode, loff_t *cpos, struct exfat_dir_ent 0); *uni_name.name = 0x0; - exfat_get_uniname_from_ext_entry(sb, &clu, i, + err = exfat_get_uniname_from_ext_entry(sb, &clu, i, uni_name.name); + if (err) { + brelse(bh); + continue; + } exfat_utf16_to_nls(sb, &uni_name, dir_entry->namebuf.lfn, dir_entry->namebuf.lfnbuf_len); @@ -375,6 +381,12 @@ unsigned int exfat_get_entry_type(struct exfat_dentry *ep) return TYPE_ACL; return TYPE_CRITICAL_SEC; } + + if (ep->type == EXFAT_VENDOR_EXT) + return TYPE_VENDOR_EXT; + if (ep->type == EXFAT_VENDOR_ALLOC) + return TYPE_VENDOR_ALLOC; + return TYPE_BENIGN_SEC; } @@ -518,6 +530,25 @@ int exfat_update_dir_chksum(struct inode *inode, struct exfat_chain *p_dir, return ret; } +static void exfat_free_benign_secondary_clusters(struct inode *inode, + struct exfat_dentry *ep) +{ + struct super_block *sb = inode->i_sb; + struct exfat_chain dir; + unsigned int start_clu = + le32_to_cpu(ep->dentry.generic_secondary.start_clu); + u64 size = le64_to_cpu(ep->dentry.generic_secondary.size); + unsigned char flags = ep->dentry.generic_secondary.flags; + + if (!(flags & ALLOC_POSSIBLE) || !start_clu || !size) + return; + + exfat_chain_set(&dir, start_clu, + EXFAT_B_TO_CLU_ROUND_UP(size, EXFAT_SB(sb)), + flags); + exfat_free_cluster(inode, &dir); +} + int exfat_init_ext_entry(struct inode *inode, struct exfat_chain *p_dir, int entry, int num_entries, struct exfat_uni_name *p_uniname) { @@ -550,6 +581,9 @@ int exfat_init_ext_entry(struct inode *inode, struct exfat_chain *p_dir, if (!ep) return -EIO; + if (exfat_get_entry_type(ep) & TYPE_BENIGN_SEC) + exfat_free_benign_secondary_clusters(inode, ep); + exfat_init_name_entry(ep, uniname); exfat_update_bh(bh, sync); brelse(bh); @@ -573,6 +607,9 @@ int exfat_remove_entries(struct inode *inode, struct exfat_chain *p_dir, if (!ep) return -EIO; + if (exfat_get_entry_type(ep) & TYPE_BENIGN_SEC) + exfat_free_benign_secondary_clusters(inode, ep); + exfat_set_entry_type(ep, TYPE_DELETED); exfat_update_bh(bh, IS_DIRSYNC(inode)); brelse(bh); @@ -741,6 +778,7 @@ enum exfat_validate_dentry_mode { ES_MODE_GET_STRM_ENTRY, ES_MODE_GET_NAME_ENTRY, ES_MODE_GET_CRITICAL_SEC_ENTRY, + ES_MODE_GET_BENIGN_SEC_ENTRY, }; static bool exfat_validate_entry(unsigned int type, @@ -754,36 +792,33 @@ static bool exfat_validate_entry(unsigned int type, if (type != TYPE_FILE && type != TYPE_DIR) return false; *mode = ES_MODE_GET_FILE_ENTRY; - return true; + break; case ES_MODE_GET_FILE_ENTRY: if (type != TYPE_STREAM) return false; *mode = ES_MODE_GET_STRM_ENTRY; - return true; + break; case ES_MODE_GET_STRM_ENTRY: if (type != TYPE_EXTEND) return false; *mode = ES_MODE_GET_NAME_ENTRY; - return true; + break; case ES_MODE_GET_NAME_ENTRY: - if (type == TYPE_STREAM) + if (type & TYPE_BENIGN_SEC) + *mode = ES_MODE_GET_BENIGN_SEC_ENTRY; + else if (type != TYPE_EXTEND) return false; - if (type != TYPE_EXTEND) { - if (!(type & TYPE_CRITICAL_SEC)) - return false; - *mode = ES_MODE_GET_CRITICAL_SEC_ENTRY; - } - return true; - case ES_MODE_GET_CRITICAL_SEC_ENTRY: - if (type == TYPE_EXTEND || type == TYPE_STREAM) - return false; - if ((type & TYPE_CRITICAL_SEC) != TYPE_CRITICAL_SEC) + break; + case ES_MODE_GET_BENIGN_SEC_ENTRY: + /* Assume unreconized benign secondary entry */ + if (!(type & TYPE_BENIGN_SEC)) return false; - return true; + break; default: - WARN_ON_ONCE(1); return false; } + + return true; } struct exfat_dentry *exfat_get_dentry_cached( @@ -1164,10 +1199,8 @@ int exfat_count_ext_entries(struct super_block *sb, struct exfat_chain *p_dir, type = exfat_get_entry_type(ext_ep); brelse(bh); - if (type == TYPE_EXTEND || type == TYPE_STREAM) + if (type & TYPE_CRITICAL_SEC || type & TYPE_BENIGN_SEC) count++; - else - break; } return count; } diff --git a/fs/exfat/exfat_fs.h b/fs/exfat/exfat_fs.h index 25a5df0fdfe01..8a399e234aab3 100644 --- a/fs/exfat/exfat_fs.h +++ b/fs/exfat/exfat_fs.h @@ -71,6 +71,8 @@ enum { #define TYPE_PADDING 0x0402 #define TYPE_ACLTAB 0x0403 #define TYPE_BENIGN_SEC 0x0800 +#define TYPE_VENDOR_EXT 0x0801 +#define TYPE_VENDOR_ALLOC 0x0802 #define MAX_CHARSET_SIZE 6 /* max size of multi-byte character */ #define MAX_NAME_LENGTH 255 /* max len of file name excluding NULL */ diff --git a/fs/exfat/exfat_raw.h b/fs/exfat/exfat_raw.h index 7f39b1c6469c4..0ece2e43cf492 100644 --- a/fs/exfat/exfat_raw.h +++ b/fs/exfat/exfat_raw.h @@ -27,6 +27,7 @@ ((sbi)->num_clusters - EXFAT_RESERVED_CLUSTERS) /* AllocationPossible and NoFatChain field in GeneralSecondaryFlags Field */ +#define ALLOC_POSSIBLE 0x01 #define ALLOC_FAT_CHAIN 0x01 #define ALLOC_NO_FAT_CHAIN 0x03 @@ -50,6 +51,8 @@ #define EXFAT_STREAM 0xC0 /* stream entry */ #define EXFAT_NAME 0xC1 /* file name entry */ #define EXFAT_ACL 0xC2 /* stream entry */ +#define EXFAT_VENDOR_EXT 0xE0 /* vendor extension entry */ +#define EXFAT_VENDOR_ALLOC 0xE1 /* vendor allocation entry */ #define IS_EXFAT_CRITICAL_PRI(x) (x < 0xA0) #define IS_EXFAT_BENIGN_PRI(x) (x < 0xC0) @@ -155,6 +158,24 @@ struct exfat_dentry { __le32 start_clu; __le64 size; } __packed upcase; /* up-case table directory entry */ + struct { + __u8 flags; + __u8 vendor_guid[16]; + __u8 vendor_defined[14]; + } __packed vendor_ext; /* vendor extension directory entry */ + struct { + __u8 flags; + __u8 vendor_guid[16]; + __u8 vendor_defined[2]; + __le32 start_clu; + __le64 size; + } __packed vendor_alloc; /* vendor allocation directory entry */ + struct { + __u8 flags; + __u8 custom_defined[18]; + __le32 start_clu; + __le64 size; + } __packed generic_secondary; /* generic secondary directory entry */ } __packed dentry; } __packed; From c0c33b94cfc23d21f67c7569a8785b33c74d6e3d Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Fri, 24 Feb 2023 07:34:24 -0800 Subject: [PATCH 528/765] nvme: fix sparse warning on effects masking The log entries are stored in le32, so use appropriate byte swapping macros. Reported-by: kernel test robot Link: https://lore.kernel.org/oe-kbuild-all/202302242222.PevBhzvC-lkp@intel.com/ Signed-off-by: Keith Busch Reviewed-by: Chaitanya Kulkarni Signed-off-by: Christoph Hellwig --- drivers/nvme/host/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index 25968b25d0ba5..384138f8f04ce 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -3104,7 +3104,7 @@ static void nvme_init_known_nvm_effects(struct nvme_ctrl *ctrl) * Rather than blindly freezing the IO queues for this effect that * doesn't even apply to IO, mask it off. */ - log->acs[nvme_admin_security_recv] &= ~NVME_CMD_EFFECTS_CSE_MASK; + log->acs[nvme_admin_security_recv] &= cpu_to_le32(~NVME_CMD_EFFECTS_CSE_MASK); log->iocs[nvme_cmd_write] |= cpu_to_le32(NVME_CMD_EFFECTS_LBCC); log->iocs[nvme_cmd_write_zeroes] |= cpu_to_le32(NVME_CMD_EFFECTS_LBCC); From 8ee0d2fb4dfa3465ea2030dec59a6f6fe3005804 Mon Sep 17 00:00:00 2001 From: Vasily Gorbik Date: Fri, 17 Feb 2023 14:00:54 +0100 Subject: [PATCH 529/765] s390/setup: do not complain about parameters handled in decompressor Currently there are several kernel command line parameters which are only parsed and handled in decompressor and not known to the kernel. This leads to the following error message during kernel boot: Unknown kernel command line parameters "mem=3G nokaslr", will be passed to user space. To avoid confusion, register those parameters with an empty stub so that kernel does not complain about them. Reported-by: Gerald Schaefer Reviewed-by: Heiko Carstens Signed-off-by: Vasily Gorbik Signed-off-by: Heiko Carstens --- arch/s390/kernel/early.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c index 59eba19ae0f2a..d26f02495636e 100644 --- a/arch/s390/kernel/early.c +++ b/arch/s390/kernel/early.c @@ -36,6 +36,23 @@ int __bootdata(is_full_image); +#define decompressor_handled_param(param) \ +static int __init ignore_decompressor_param_##param(char *s) \ +{ \ + return 0; \ +} \ +early_param(#param, ignore_decompressor_param_##param) + +decompressor_handled_param(mem); +decompressor_handled_param(vmalloc); +decompressor_handled_param(dfltcc); +decompressor_handled_param(noexec); +decompressor_handled_param(facilities); +decompressor_handled_param(nokaslr); +#if IS_ENABLED(CONFIG_KVM) +decompressor_handled_param(prot_virt); +#endif + static void __init reset_tod_clock(void) { union tod_clock clk; From ae4b60f6b7a8d25c7253cab104468d22efcecf1a Mon Sep 17 00:00:00 2001 From: Nico Boehr Date: Thu, 16 Feb 2023 13:12:08 +0100 Subject: [PATCH 530/765] s390/nmi: fix virtual-physical address confusion When a machine check is received while in SIE, it is reinjected into the guest in some cases. The respective code needs to access the sie_block, which is taken from the backed up R14. Since reinjection only occurs while we are in SIE (i.e. between the labels sie_entry and sie_leave in entry.S and thus if CIF_MCCK_GUEST is set), the backed up R14 will always contain a physical address in s390_backup_mcck_info. This currently works, because virtual and physical addresses are the same. Add phys_to_virt() to resolve the virtual-physical confusion. Signed-off-by: Nico Boehr Reviewed-by: Claudio Imbrenda Reviewed-by: Alexander Gordeev Link: https://lore.kernel.org/r/20230216121208.4390-2-nrb@linux.ibm.com Signed-off-by: Janosch Frank Signed-off-by: Heiko Carstens --- arch/s390/kernel/nmi.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/s390/kernel/nmi.c b/arch/s390/kernel/nmi.c index 5dbf274719a95..56d9c559afa1f 100644 --- a/arch/s390/kernel/nmi.c +++ b/arch/s390/kernel/nmi.c @@ -346,8 +346,7 @@ static void notrace s390_backup_mcck_info(struct pt_regs *regs) struct sie_page *sie_page; /* r14 contains the sie block, which was set in sie64a */ - struct kvm_s390_sie_block *sie_block = - (struct kvm_s390_sie_block *) regs->gprs[14]; + struct kvm_s390_sie_block *sie_block = phys_to_virt(regs->gprs[14]); if (sie_block == NULL) /* Something's seriously wrong, stop system. */ From ebf95e884694b2c796ecb53d80d2b4cff8990d2f Mon Sep 17 00:00:00 2001 From: Harald Freudenberger Date: Fri, 17 Feb 2023 12:05:36 +0100 Subject: [PATCH 531/765] s390/ap,zcrypt,vfio: introduce and use ap_queue_status_reg union Introduce a new ap queue status register wrapper union to access register wide values. So the inline assembler only sees register wide values but the surrounding code may use a more structured view of the same value and a reader of the code (and the compiler) gets a clear understanding about the mapping between fields and register values. All the changes to access the ap queue status are local to the inline functions within ap.h. However, the struct ap_qirq_ctrl has been replaces by a union for same reason and this needed slight adaptions in the calling code. Suggested-by: Halil Pasic Suggested-by: Andreas Arnez Signed-off-by: Harald Freudenberger Acked-by: Heiko Carstens Reviewed-by: Holger Dengler Signed-off-by: Heiko Carstens --- arch/s390/include/asm/ap.h | 100 ++++++++++++++++-------------- drivers/s390/crypto/ap_queue.c | 2 +- drivers/s390/crypto/vfio_ap_ops.c | 4 +- 3 files changed, 55 insertions(+), 51 deletions(-) diff --git a/arch/s390/include/asm/ap.h b/arch/s390/include/asm/ap.h index 57a2d6518d272..c699f251a4648 100644 --- a/arch/s390/include/asm/ap.h +++ b/arch/s390/include/asm/ap.h @@ -49,6 +49,19 @@ struct ap_queue_status { unsigned int _pad2 : 16; }; +/* + * AP queue status reg union to access the reg1 + * register with the lower 32 bits comprising the + * ap queue status. + */ +union ap_queue_status_reg { + unsigned long value; + struct { + u32 _pad; + struct ap_queue_status status; + }; +}; + /** * ap_intructions_available() - Test if AP instructions are available. * @@ -82,7 +95,7 @@ static inline bool ap_instructions_available(void) */ static inline struct ap_queue_status ap_tapq(ap_qid_t qid, unsigned long *info) { - struct ap_queue_status reg1; + union ap_queue_status_reg reg1; unsigned long reg2; asm volatile( @@ -91,12 +104,12 @@ static inline struct ap_queue_status ap_tapq(ap_qid_t qid, unsigned long *info) " .insn rre,0xb2af0000,0,0\n" /* PQAP(TAPQ) */ " lgr %[reg1],1\n" /* gr1 (status) into reg1 */ " lgr %[reg2],2\n" /* gr2 into reg2 */ - : [reg1] "=&d" (reg1), [reg2] "=&d" (reg2) + : [reg1] "=&d" (reg1.value), [reg2] "=&d" (reg2) : [qid] "d" (qid) : "cc", "0", "1", "2"); if (info) *info = reg2; - return reg1; + return reg1.status; } /** @@ -125,16 +138,16 @@ static inline struct ap_queue_status ap_test_queue(ap_qid_t qid, static inline struct ap_queue_status ap_rapq(ap_qid_t qid) { unsigned long reg0 = qid | (1UL << 24); /* fc 1UL is RAPQ */ - struct ap_queue_status reg1; + union ap_queue_status_reg reg1; asm volatile( " lgr 0,%[reg0]\n" /* qid arg into gr0 */ " .insn rre,0xb2af0000,0,0\n" /* PQAP(RAPQ) */ " lgr %[reg1],1\n" /* gr1 (status) into reg1 */ - : [reg1] "=&d" (reg1) + : [reg1] "=&d" (reg1.value) : [reg0] "d" (reg0) : "cc", "0", "1"); - return reg1; + return reg1.status; } /** @@ -146,16 +159,16 @@ static inline struct ap_queue_status ap_rapq(ap_qid_t qid) static inline struct ap_queue_status ap_zapq(ap_qid_t qid) { unsigned long reg0 = qid | (2UL << 24); /* fc 2UL is ZAPQ */ - struct ap_queue_status reg1; + union ap_queue_status_reg reg1; asm volatile( " lgr 0,%[reg0]\n" /* qid arg into gr0 */ " .insn rre,0xb2af0000,0,0\n" /* PQAP(ZAPQ) */ " lgr %[reg1],1\n" /* gr1 (status) into reg1 */ - : [reg1] "=&d" (reg1) + : [reg1] "=&d" (reg1.value) : [reg0] "d" (reg0) : "cc", "0", "1"); - return reg1; + return reg1.status; } /** @@ -209,18 +222,21 @@ static inline int ap_qci(struct ap_config_info *config) * parameter to the PQAP(AQIC) instruction. For details please * see the AR documentation. */ -struct ap_qirq_ctrl { - unsigned int _res1 : 8; - unsigned int zone : 8; /* zone info */ - unsigned int ir : 1; /* ir flag: enable (1) or disable (0) irq */ - unsigned int _res2 : 4; - unsigned int gisc : 3; /* guest isc field */ - unsigned int _res3 : 6; - unsigned int gf : 2; /* gisa format */ - unsigned int _res4 : 1; - unsigned int gisa : 27; /* gisa origin */ - unsigned int _res5 : 1; - unsigned int isc : 3; /* irq sub class */ +union ap_qirq_ctrl { + unsigned long value; + struct { + unsigned int : 8; + unsigned int zone : 8; /* zone info */ + unsigned int ir : 1; /* ir flag: enable (1) or disable (0) irq */ + unsigned int : 4; + unsigned int gisc : 3; /* guest isc field */ + unsigned int : 6; + unsigned int gf : 2; /* gisa format */ + unsigned int : 1; + unsigned int gisa : 27; /* gisa origin */ + unsigned int : 1; + unsigned int isc : 3; /* irq sub class */ + }; }; /** @@ -232,21 +248,14 @@ struct ap_qirq_ctrl { * Returns AP queue status. */ static inline struct ap_queue_status ap_aqic(ap_qid_t qid, - struct ap_qirq_ctrl qirqctrl, + union ap_qirq_ctrl qirqctrl, phys_addr_t pa_ind) { unsigned long reg0 = qid | (3UL << 24); /* fc 3UL is AQIC */ - union { - unsigned long value; - struct ap_qirq_ctrl qirqctrl; - struct { - u32 _pad; - struct ap_queue_status status; - }; - } reg1; + union ap_queue_status_reg reg1; unsigned long reg2 = pa_ind; - reg1.qirqctrl = qirqctrl; + reg1.value = qirqctrl.value; asm volatile( " lgr 0,%[reg0]\n" /* qid param into gr0 */ @@ -254,7 +263,7 @@ static inline struct ap_queue_status ap_aqic(ap_qid_t qid, " lgr 2,%[reg2]\n" /* ni addr into gr2 */ " .insn rre,0xb2af0000,0,0\n" /* PQAP(AQIC) */ " lgr %[reg1],1\n" /* gr1 (status) into reg1 */ - : [reg1] "+&d" (reg1) + : [reg1] "+&d" (reg1.value) : [reg0] "d" (reg0), [reg2] "d" (reg2) : "cc", "memory", "0", "1", "2"); @@ -291,13 +300,7 @@ static inline struct ap_queue_status ap_qact(ap_qid_t qid, int ifbit, union ap_qact_ap_info *apinfo) { unsigned long reg0 = qid | (5UL << 24) | ((ifbit & 0x01) << 22); - union { - unsigned long value; - struct { - u32 _pad; - struct ap_queue_status status; - }; - } reg1; + union ap_queue_status_reg reg1; unsigned long reg2; reg1.value = apinfo->val; @@ -308,7 +311,7 @@ static inline struct ap_queue_status ap_qact(ap_qid_t qid, int ifbit, " .insn rre,0xb2af0000,0,0\n" /* PQAP(QACT) */ " lgr %[reg1],1\n" /* gr1 (status) into reg1 */ " lgr %[reg2],2\n" /* qact out info into reg2 */ - : [reg1] "+&d" (reg1), [reg2] "=&d" (reg2) + : [reg1] "+&d" (reg1.value), [reg2] "=&d" (reg2) : [reg0] "d" (reg0) : "cc", "0", "1", "2"); apinfo->val = reg2; @@ -333,7 +336,7 @@ static inline struct ap_queue_status ap_nqap(ap_qid_t qid, { unsigned long reg0 = qid | 0x40000000UL; /* 0x4... is last msg part */ union register_pair nqap_r1, nqap_r2; - struct ap_queue_status reg1; + union ap_queue_status_reg reg1; nqap_r1.even = (unsigned int)(psmid >> 32); nqap_r1.odd = psmid & 0xffffffff; @@ -345,11 +348,11 @@ static inline struct ap_queue_status ap_nqap(ap_qid_t qid, "0: .insn rre,0xb2ad0000,%[nqap_r1],%[nqap_r2]\n" " brc 2,0b\n" /* handle partial completion */ " lgr %[reg1],1\n" /* gr1 (status) into reg1 */ - : [reg0] "+&d" (reg0), [reg1] "=&d" (reg1), + : [reg0] "+&d" (reg0), [reg1] "=&d" (reg1.value), [nqap_r2] "+&d" (nqap_r2.pair) : [nqap_r1] "d" (nqap_r1.pair) : "cc", "memory", "0", "1"); - return reg1; + return reg1.status; } /** @@ -389,7 +392,7 @@ static inline struct ap_queue_status ap_dqap(ap_qid_t qid, unsigned long *resgr0) { unsigned long reg0 = resgr0 && *resgr0 ? *resgr0 : qid | 0x80000000UL; - struct ap_queue_status reg1; + union ap_queue_status_reg reg1; unsigned long reg2; union register_pair rp1, rp2; @@ -408,8 +411,9 @@ static inline struct ap_queue_status ap_dqap(ap_qid_t qid, "2: lgr %[reg0],0\n" /* gr0 (qid + info) into reg0 */ " lgr %[reg1],1\n" /* gr1 (status) into reg1 */ " lgr %[reg2],2\n" /* gr2 (res length) into reg2 */ - : [reg0] "+&d" (reg0), [reg1] "=&d" (reg1), [reg2] "=&d" (reg2), - [rp1] "+&d" (rp1.pair), [rp2] "+&d" (rp2.pair) + : [reg0] "+&d" (reg0), [reg1] "=&d" (reg1.value), + [reg2] "=&d" (reg2), [rp1] "+&d" (rp1.pair), + [rp2] "+&d" (rp2.pair) : : "cc", "memory", "0", "1", "2"); @@ -421,7 +425,7 @@ static inline struct ap_queue_status ap_dqap(ap_qid_t qid, * Signal the caller that this dqap is only partially received * with a special status response code 0xFF and *resgr0 updated */ - reg1.response_code = 0xFF; + reg1.status.response_code = 0xFF; if (resgr0) *resgr0 = reg0; } else { @@ -430,7 +434,7 @@ static inline struct ap_queue_status ap_dqap(ap_qid_t qid, *resgr0 = 0; } - return reg1; + return reg1.status; } /* diff --git a/drivers/s390/crypto/ap_queue.c b/drivers/s390/crypto/ap_queue.c index a32457b4cbb8a..2637fe1df7277 100644 --- a/drivers/s390/crypto/ap_queue.c +++ b/drivers/s390/crypto/ap_queue.c @@ -29,8 +29,8 @@ static void __ap_flush_queue(struct ap_queue *aq); */ static int ap_queue_enable_irq(struct ap_queue *aq, void *ind) { + union ap_qirq_ctrl qirqctrl = { .value = 0 }; struct ap_queue_status status; - struct ap_qirq_ctrl qirqctrl = { 0 }; qirqctrl.ir = 1; qirqctrl.isc = AP_ISC; diff --git a/drivers/s390/crypto/vfio_ap_ops.c b/drivers/s390/crypto/vfio_ap_ops.c index 28a36e016ea91..72e10abb103a0 100644 --- a/drivers/s390/crypto/vfio_ap_ops.c +++ b/drivers/s390/crypto/vfio_ap_ops.c @@ -301,7 +301,7 @@ static void vfio_ap_free_aqic_resources(struct vfio_ap_queue *q) */ static struct ap_queue_status vfio_ap_irq_disable(struct vfio_ap_queue *q) { - struct ap_qirq_ctrl aqic_gisa = {}; + union ap_qirq_ctrl aqic_gisa = { .value = 0 }; struct ap_queue_status status; int retries = 5; @@ -384,7 +384,7 @@ static struct ap_queue_status vfio_ap_irq_enable(struct vfio_ap_queue *q, int isc, struct kvm_vcpu *vcpu) { - struct ap_qirq_ctrl aqic_gisa = {}; + union ap_qirq_ctrl aqic_gisa = { .value = 0 }; struct ap_queue_status status = {}; struct kvm_s390_gisa *gisa; struct page *h_page; From 6e2985c938e8b765b3de299c561d87f98330c546 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Fri, 17 Feb 2023 15:44:25 -0800 Subject: [PATCH 532/765] xfs: restore old agirotor behavior Prior to the removal of xfs_ialloc_next_ag, we would increment the agi rotor and return the *old* value. atomic_inc_return returns the new value, which causes mkfs to allocate the root directory in AG 1. Put back the old behavior (at least for mkfs) by subtracting 1 here. Fixes: 20a5eab49d35 ("xfs: convert xfs_ialloc_next_ag() to an atomic") Signed-off-by: Darrick J. Wong Reviewed-by: Dave Chinner --- fs/xfs/libxfs/xfs_ialloc.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index 82ea6aa5af107..7ee292aecbeb0 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -1729,7 +1729,8 @@ xfs_dialloc( * an AG has enough space for file creation. */ if (S_ISDIR(mode)) - start_agno = atomic_inc_return(&mp->m_agirotor) % mp->m_maxagi; + start_agno = (atomic_inc_return(&mp->m_agirotor) - 1) % + mp->m_maxagi; else { start_agno = XFS_INO_TO_AGNO(mp, parent); if (start_agno >= mp->m_maxagi) From 6921ed9049bc7457f66c1596c5b78aec0dae4a9d Mon Sep 17 00:00:00 2001 From: KP Singh Date: Mon, 27 Feb 2023 07:05:40 +0100 Subject: [PATCH 533/765] x86/speculation: Allow enabling STIBP with legacy IBRS MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When plain IBRS is enabled (not enhanced IBRS), the logic in spectre_v2_user_select_mitigation() determines that STIBP is not needed. The IBRS bit implicitly protects against cross-thread branch target injection. However, with legacy IBRS, the IBRS bit is cleared on returning to userspace for performance reasons which leaves userspace threads vulnerable to cross-thread branch target injection against which STIBP protects. Exclude IBRS from the spectre_v2_in_ibrs_mode() check to allow for enabling STIBP (through seccomp/prctl() by default or always-on, if selected by spectre_v2_user kernel cmdline parameter). [ bp: Massage. ] Fixes: 7c693f54c873 ("x86/speculation: Add spectre_v2=ibrs option to support Kernel IBRS") Reported-by: José Oliveira Reported-by: Rodrigo Branco Signed-off-by: KP Singh Signed-off-by: Borislav Petkov (AMD) Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20230220120127.1975241-1-kpsingh@kernel.org Link: https://lore.kernel.org/r/20230221184908.2349578-1-kpsingh@kernel.org --- arch/x86/kernel/cpu/bugs.c | 25 ++++++++++++++++++------- 1 file changed, 18 insertions(+), 7 deletions(-) diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index cf81848b72f45..f9d060e71c3ee 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -1133,14 +1133,18 @@ spectre_v2_parse_user_cmdline(void) return SPECTRE_V2_USER_CMD_AUTO; } -static inline bool spectre_v2_in_ibrs_mode(enum spectre_v2_mitigation mode) +static inline bool spectre_v2_in_eibrs_mode(enum spectre_v2_mitigation mode) { - return mode == SPECTRE_V2_IBRS || - mode == SPECTRE_V2_EIBRS || + return mode == SPECTRE_V2_EIBRS || mode == SPECTRE_V2_EIBRS_RETPOLINE || mode == SPECTRE_V2_EIBRS_LFENCE; } +static inline bool spectre_v2_in_ibrs_mode(enum spectre_v2_mitigation mode) +{ + return spectre_v2_in_eibrs_mode(mode) || mode == SPECTRE_V2_IBRS; +} + static void __init spectre_v2_user_select_mitigation(void) { @@ -1203,12 +1207,19 @@ spectre_v2_user_select_mitigation(void) } /* - * If no STIBP, IBRS or enhanced IBRS is enabled, or SMT impossible, - * STIBP is not required. + * If no STIBP, enhanced IBRS is enabled, or SMT impossible, STIBP + * is not required. + * + * Enhanced IBRS also protects against cross-thread branch target + * injection in user-mode as the IBRS bit remains always set which + * implicitly enables cross-thread protections. However, in legacy IBRS + * mode, the IBRS bit is set only on kernel entry and cleared on return + * to userspace. This disables the implicit cross-thread protection, + * so allow for STIBP to be selected in that case. */ if (!boot_cpu_has(X86_FEATURE_STIBP) || !smt_possible || - spectre_v2_in_ibrs_mode(spectre_v2_enabled)) + spectre_v2_in_eibrs_mode(spectre_v2_enabled)) return; /* @@ -2340,7 +2351,7 @@ static ssize_t mmio_stale_data_show_state(char *buf) static char *stibp_state(void) { - if (spectre_v2_in_ibrs_mode(spectre_v2_enabled)) + if (spectre_v2_in_eibrs_mode(spectre_v2_enabled)) return ""; switch (spectre_v2_user_stibp) { From e02b50ca442e88122e1302d4dbc1b71a4808c13f Mon Sep 17 00:00:00 2001 From: KP Singh Date: Mon, 27 Feb 2023 07:05:41 +0100 Subject: [PATCH 534/765] Documentation/hw-vuln: Document the interaction between IBRS and STIBP Explain why STIBP is needed with legacy IBRS as currently implemented (KERNEL_IBRS) and why STIBP is not needed when enhanced IBRS is enabled. Fixes: 7c693f54c873 ("x86/speculation: Add spectre_v2=ibrs option to support Kernel IBRS") Signed-off-by: KP Singh Signed-off-by: Borislav Petkov (AMD) Link: https://lore.kernel.org/r/20230227060541.1939092-2-kpsingh@kernel.org --- Documentation/admin-guide/hw-vuln/spectre.rst | 21 ++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-) diff --git a/Documentation/admin-guide/hw-vuln/spectre.rst b/Documentation/admin-guide/hw-vuln/spectre.rst index 3fe6511c54050..4d186f599d90f 100644 --- a/Documentation/admin-guide/hw-vuln/spectre.rst +++ b/Documentation/admin-guide/hw-vuln/spectre.rst @@ -479,8 +479,16 @@ Spectre variant 2 On Intel Skylake-era systems the mitigation covers most, but not all, cases. See :ref:`[3] ` for more details. - On CPUs with hardware mitigation for Spectre variant 2 (e.g. Enhanced - IBRS on x86), retpoline is automatically disabled at run time. + On CPUs with hardware mitigation for Spectre variant 2 (e.g. IBRS + or enhanced IBRS on x86), retpoline is automatically disabled at run time. + + Systems which support enhanced IBRS (eIBRS) enable IBRS protection once at + boot, by setting the IBRS bit, and they're automatically protected against + Spectre v2 variant attacks, including cross-thread branch target injections + on SMT systems (STIBP). In other words, eIBRS enables STIBP too. + + Legacy IBRS systems clear the IBRS bit on exit to userspace and + therefore explicitly enable STIBP for that The retpoline mitigation is turned on by default on vulnerable CPUs. It can be forced on or off by the administrator @@ -504,9 +512,12 @@ Spectre variant 2 For Spectre variant 2 mitigation, individual user programs can be compiled with return trampolines for indirect branches. This protects them from consuming poisoned entries in the branch - target buffer left by malicious software. Alternatively, the - programs can disable their indirect branch speculation via prctl() - (See :ref:`Documentation/userspace-api/spec_ctrl.rst `). + target buffer left by malicious software. + + On legacy IBRS systems, at return to userspace, implicit STIBP is disabled + because the kernel clears the IBRS bit. In this case, the userspace programs + can disable indirect branch speculation via prctl() (See + :ref:`Documentation/userspace-api/spec_ctrl.rst `). On x86, this will turn on STIBP to guard against attacks from the sibling thread when the user program is running, and use IBPB to flush the branch target buffer when switching to/from the program. From a07484c083eabc809e45a198bd639bfd37ac70b6 Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Wed, 22 Feb 2023 15:35:30 +0700 Subject: [PATCH 535/765] bpf, docs: Fix link to BTF doc Ross reported broken link to BTF documentation (Documentation/bpf/btf.rst) in Documentation/bpf/bpf_devel_QA.rst. The link in question is written using external link syntax, with the target refers to BTF doc in reST source (btf.rst), which doesn't exist in resulting HTML output. Fix the link by replacing external link syntax with simply writing out the target doc, which the link will be generated to the correct HTML doc target. Fixes: 6736aa793c2b5f ("selftests/bpf: Add general instructions for test execution") Reported-by: Ross Zwisler Signed-off-by: Bagas Sanjaya Signed-off-by: Daniel Borkmann Acked-by: Ross Zwisler Link: https://lore.kernel.org/linux-doc/Y++09LKx25dtR4Ow@google.com/ Link: https://lore.kernel.org/bpf/20230222083530.26136-1-bagasdotme@gmail.com --- Documentation/bpf/bpf_devel_QA.rst | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/Documentation/bpf/bpf_devel_QA.rst b/Documentation/bpf/bpf_devel_QA.rst index 03d4993eda6f0..715f7321020f2 100644 --- a/Documentation/bpf/bpf_devel_QA.rst +++ b/Documentation/bpf/bpf_devel_QA.rst @@ -469,7 +469,7 @@ under test should match the config file fragment in tools/testing/selftests/bpf as closely as possible. Finally to ensure support for latest BPF Type Format features - -discussed in `Documentation/bpf/btf.rst`_ - pahole version 1.16 +discussed in Documentation/bpf/btf.rst - pahole version 1.16 is required for kernels built with CONFIG_DEBUG_INFO_BTF=y. pahole is delivered in the dwarves package or can be built from source at @@ -690,6 +690,5 @@ when: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/tools/testing/selftests/bpf/ .. _Documentation/dev-tools/kselftest.rst: https://www.kernel.org/doc/html/latest/dev-tools/kselftest.html -.. _Documentation/bpf/btf.rst: btf.rst Happy BPF hacking! From 2d311f480b52eeb2e1fd432d64b78d82952c3808 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 26 Feb 2023 23:20:16 -0800 Subject: [PATCH 536/765] riscv, bpf: Fix patch_text implicit declaration MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit bpf_jit_comp64.c uses patch_text(), so add to it to prevent the build error: ../arch/riscv/net/bpf_jit_comp64.c: In function 'bpf_arch_text_poke': ../arch/riscv/net/bpf_jit_comp64.c:691:23: error: implicit declaration of function 'patch_text'; did you mean 'path_get'? [-Werror=implicit-function-declaration] 691 | ret = patch_text(ip, new_insns, ninsns); | ^~~~~~~~~~ Fixes: 596f2e6f9cf4 ("riscv, bpf: Add bpf_arch_text_poke support for RV64") Reported-by: kernel test robot Signed-off-by: Randy Dunlap Signed-off-by: Daniel Borkmann Acked-by: Pu Lehui Acked-by: Björn Töpel Link: https://lore.kernel.org/r/202302271000.Aj4nMXbZ-lkp@intel.com Link: https://lore.kernel.org/bpf/20230227072016.14618-1-rdunlap@infradead.org --- arch/riscv/net/bpf_jit_comp64.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/riscv/net/bpf_jit_comp64.c b/arch/riscv/net/bpf_jit_comp64.c index f5a668736c79b..acdc3f040195e 100644 --- a/arch/riscv/net/bpf_jit_comp64.c +++ b/arch/riscv/net/bpf_jit_comp64.c @@ -10,6 +10,7 @@ #include #include #include +#include #include "bpf_jit.h" #define RV_REG_TCC RV_REG_A6 From fe90151c3cc6dbc0836723995f65d371ef4ad514 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 27 Feb 2023 10:58:34 -0500 Subject: [PATCH 537/765] SUNRPC: Let Kunit tests run with some enctypes compiled out Allow the new GSS Kerberos encryption type test suites to run outside of the kunit infrastructure. Replace the assertion that fires when lookup_enctype() so that the case is skipped instead of failing outright. Reported-by: Geert Uytterhoeven Tested-by: Geert Uytterhoeven Signed-off-by: Chuck Lever --- net/sunrpc/auth_gss/gss_krb5_test.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/net/sunrpc/auth_gss/gss_krb5_test.c b/net/sunrpc/auth_gss/gss_krb5_test.c index c287ce15c419f..0a7c5280e4e3d 100644 --- a/net/sunrpc/auth_gss/gss_krb5_test.c +++ b/net/sunrpc/auth_gss/gss_krb5_test.c @@ -49,7 +49,8 @@ static void kdf_case(struct kunit *test) /* Arrange */ gk5e = gss_krb5_lookup_enctype(param->enctype); - KUNIT_ASSERT_NOT_NULL(test, gk5e); + if (!gk5e) + kunit_skip(test, "Encryption type is not available"); derivedkey.data = kunit_kzalloc(test, param->expected_result->len, GFP_KERNEL); @@ -83,7 +84,8 @@ static void checksum_case(struct kunit *test) /* Arrange */ gk5e = gss_krb5_lookup_enctype(param->enctype); - KUNIT_ASSERT_NOT_NULL(test, gk5e); + if (!gk5e) + kunit_skip(test, "Encryption type is not available"); Kc.len = gk5e->Kc_length; Kc.data = kunit_kzalloc(test, Kc.len, GFP_KERNEL); @@ -725,7 +727,8 @@ static void rfc3962_encrypt_case(struct kunit *test) /* Arrange */ gk5e = gss_krb5_lookup_enctype(param->enctype); - KUNIT_ASSERT_NOT_NULL(test, gk5e); + if (!gk5e) + kunit_skip(test, "Encryption type is not available"); cbc_tfm = crypto_alloc_sync_skcipher(gk5e->aux_cipher, 0, 0); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, cbc_tfm); @@ -1319,7 +1322,8 @@ static void rfc6803_encrypt_case(struct kunit *test) /* Arrange */ gk5e = gss_krb5_lookup_enctype(param->enctype); - KUNIT_ASSERT_NOT_NULL(test, gk5e); + if (!gk5e) + kunit_skip(test, "Encryption type is not available"); usage.data[3] = param->constant; @@ -1810,7 +1814,8 @@ static void rfc8009_encrypt_case(struct kunit *test) /* Arrange */ gk5e = gss_krb5_lookup_enctype(param->enctype); - KUNIT_ASSERT_NOT_NULL(test, gk5e); + if (!gk5e) + kunit_skip(test, "Encryption type is not available"); *(__be32 *)usage.data = cpu_to_be32(2); @@ -1975,7 +1980,8 @@ static void encrypt_selftest_case(struct kunit *test) /* Arrange */ gk5e = gss_krb5_lookup_enctype(param->enctype); - KUNIT_ASSERT_NOT_NULL(test, gk5e); + if (!gk5e) + kunit_skip(test, "Encryption type is not available"); cbc_tfm = crypto_alloc_sync_skcipher(gk5e->aux_cipher, 0, 0); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, cbc_tfm); From fb5b855d9f34927579baa0a20b4d0d8ea3740abd Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 27 Feb 2023 11:52:59 -0500 Subject: [PATCH 538/765] SUNRPC: Properly terminate test case arrays Unable to handle kernel paging request at virtual address 73657420 when execute [73657420] *pgd=00000000 Internal error: Oops: 80000005 [#1] ARM CPU: 0 PID: 1 Comm: swapper Tainted: G N 6.2.0-rc7-00133-g373f26a81164-dirty #9 Hardware name: Generic DT based system PC is at 0x73657420 LR is at kunit_run_tests+0x3e0/0x5f4 On x86 with GCC 12, the missing array terminators did not seem to matter. Other platforms appear to be more picky. Reported-by: Geert Uytterhoeven Tested-by: Geert Uytterhoeven Signed-off-by: Chuck Lever --- net/sunrpc/auth_gss/gss_krb5_test.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/net/sunrpc/auth_gss/gss_krb5_test.c b/net/sunrpc/auth_gss/gss_krb5_test.c index 0a7c5280e4e3d..ce0541e32fc98 100644 --- a/net/sunrpc/auth_gss/gss_krb5_test.c +++ b/net/sunrpc/auth_gss/gss_krb5_test.c @@ -519,6 +519,7 @@ static struct kunit_case rfc3961_test_cases[] = { .run_case = kdf_case, .generate_params = rfc3961_kdf_gen_params, }, + {} }; static struct kunit_suite rfc3961_suite = { @@ -780,6 +781,7 @@ static struct kunit_case rfc3962_test_cases[] = { .run_case = rfc3962_encrypt_case, .generate_params = rfc3962_encrypt_gen_params, }, + {} }; static struct kunit_suite rfc3962_suite = { @@ -1415,6 +1417,7 @@ static struct kunit_case rfc6803_test_cases[] = { .run_case = rfc6803_encrypt_case, .generate_params = rfc6803_encrypt_gen_params, }, + {} }; static struct kunit_suite rfc6803_suite = { @@ -1907,6 +1910,7 @@ static struct kunit_case rfc8009_test_cases[] = { .run_case = rfc8009_encrypt_case, .generate_params = rfc8009_encrypt_gen_params, }, + {} }; static struct kunit_suite rfc8009_suite = { @@ -2029,6 +2033,7 @@ static struct kunit_case encryption_test_cases[] = { .run_case = encrypt_selftest_case, .generate_params = encrypt_selftest_gen_params, }, + {} }; static struct kunit_suite encryption_test_suite = { From fea8826d5fdc4ff5c93e883a738597129614039c Mon Sep 17 00:00:00 2001 From: Jiaxun Yang Date: Mon, 27 Feb 2023 18:46:13 +0000 Subject: [PATCH 539/765] MIPS: smp-cps: Don't rely on CP0_CMGCRBASE CP0_CMGCRBASE is not always available on CPS enabled system such as early proAptiv. For early SMP bring up where we can't safely access memeory, we patch the entry of CPS NMI vector to inject CMGCR address directly into register during early core bringup. For VPE bringup as the core is already coherenct at that point we just read the variable to obtain the address. Signed-off-by: Jiaxun Yang Signed-off-by: Thomas Bogendoerfer --- arch/mips/include/asm/smp-cps.h | 4 ++++ arch/mips/kernel/cps-vec.S | 35 ++++++++++++++------------------- arch/mips/kernel/smp-cps.c | 2 ++ 3 files changed, 21 insertions(+), 20 deletions(-) diff --git a/arch/mips/include/asm/smp-cps.h b/arch/mips/include/asm/smp-cps.h index 7e5b9411faee0..22a572b70fe31 100644 --- a/arch/mips/include/asm/smp-cps.h +++ b/arch/mips/include/asm/smp-cps.h @@ -7,6 +7,8 @@ #ifndef __MIPS_ASM_SMP_CPS_H__ #define __MIPS_ASM_SMP_CPS_H__ +#define CPS_ENTRY_PATCH_INSNS 6 + #ifndef __ASSEMBLY__ struct vpe_boot_config { @@ -30,6 +32,8 @@ extern void mips_cps_boot_vpes(struct core_boot_config *cfg, unsigned vpe); extern void mips_cps_pm_save(void); extern void mips_cps_pm_restore(void); +extern void *mips_cps_core_entry_patch_end; + #ifdef CONFIG_MIPS_CPS extern bool mips_cps_smp_in_use(void); diff --git a/arch/mips/kernel/cps-vec.S b/arch/mips/kernel/cps-vec.S index 9753432401487..8ef492da827f8 100644 --- a/arch/mips/kernel/cps-vec.S +++ b/arch/mips/kernel/cps-vec.S @@ -13,6 +13,7 @@ #include #include #include +#include #define GCR_CPC_BASE_OFS 0x0088 #define GCR_CL_COHERENCE_OFS 0x2008 @@ -80,25 +81,20 @@ nop .endm - /* Calculate an uncached address for the CM GCRs */ - .macro cmgcrb dest - .set push - .set noat - MFC0 $1, CP0_CMGCRBASE - PTR_SLL $1, $1, 4 - PTR_LI \dest, UNCAC_BASE - PTR_ADDU \dest, \dest, $1 - .set pop - .endm .balign 0x1000 LEAF(mips_cps_core_entry) /* - * These first 4 bytes will be patched by cps_smp_setup to load the - * CCA to use into register s0. + * These first several instructions will be patched by cps_smp_setup to load the + * CCA to use into register s0 and GCR base address to register s1. */ - .word 0 + .rept CPS_ENTRY_PATCH_INSNS + nop + .endr + + .global mips_cps_core_entry_patch_end +mips_cps_core_entry_patch_end: /* Check whether we're here due to an NMI */ mfc0 k0, CP0_STATUS @@ -121,8 +117,7 @@ not_nmi: mtc0 t0, CP0_STATUS /* Skip cache & coherence setup if we're already coherent */ - cmgcrb v1 - lw s7, GCR_CL_COHERENCE_OFS(v1) + lw s7, GCR_CL_COHERENCE_OFS(s1) bnez s7, 1f nop @@ -132,7 +127,7 @@ not_nmi: /* Enter the coherent domain */ li t0, 0xff - sw t0, GCR_CL_COHERENCE_OFS(v1) + sw t0, GCR_CL_COHERENCE_OFS(s1) ehb /* Set Kseg0 CCA to that in s0 */ @@ -305,8 +300,7 @@ LEAF(mips_cps_core_init) */ LEAF(mips_cps_get_bootcfg) /* Calculate a pointer to this cores struct core_boot_config */ - cmgcrb t0 - lw t0, GCR_CL_ID_OFS(t0) + lw t0, GCR_CL_ID_OFS(s1) li t1, COREBOOTCFG_SIZE mul t0, t0, t1 PTR_LA t1, mips_cps_core_bootcfg @@ -366,8 +360,9 @@ LEAF(mips_cps_boot_vpes) has_vp t0, 5f /* Find base address of CPC */ - cmgcrb t3 - PTR_L t1, GCR_CPC_BASE_OFS(t3) + PTR_LA t1, mips_gcr_base + PTR_L t1, 0(t1) + PTR_L t1, GCR_CPC_BASE_OFS(t1) PTR_LI t2, ~0x7fff and t1, t1, t2 PTR_LI t2, UNCAC_BASE diff --git a/arch/mips/kernel/smp-cps.c b/arch/mips/kernel/smp-cps.c index f2df0cae1b4d9..4fc288bb85b96 100644 --- a/arch/mips/kernel/smp-cps.c +++ b/arch/mips/kernel/smp-cps.c @@ -162,6 +162,8 @@ static void __init cps_prepare_cpus(unsigned int max_cpus) */ entry_code = (u32 *)&mips_cps_core_entry; uasm_i_addiu(&entry_code, 16, 0, cca); + UASM_i_LA(&entry_code, 17, (long)mips_gcr_base); + BUG_ON((void *)entry_code > (void *)&mips_cps_core_entry_patch_end); blast_dcache_range((unsigned long)&mips_cps_core_entry, (unsigned long)entry_code); bc_wback_inv((unsigned long)&mips_cps_core_entry, From 5ae7e037de566c3106c0fa951bbf35fd6370fdf6 Mon Sep 17 00:00:00 2001 From: Jiaxun Yang Date: Mon, 27 Feb 2023 18:46:14 +0000 Subject: [PATCH 540/765] MIPS: cevt-r4k: Offset the value used to clear compare interrupt In c0_compare_int_usable we clear compare interrupt by write value just read out from counter to compare register. However sometimes if those all instructions are graduated together then it's possible that at the time compare register is written, the counter haven't progressed, thus the interrupt is triggered again. It also applies to QEMU that instructions is executed significantly faster then counter. Offset the value used to clear interrupt by one to prevent that happen. Signed-off-by: Jiaxun Yang Signed-off-by: Thomas Bogendoerfer --- arch/mips/kernel/cevt-r4k.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/mips/kernel/cevt-r4k.c b/arch/mips/kernel/cevt-r4k.c index 32ec67c9ab67b..368e8475870f0 100644 --- a/arch/mips/kernel/cevt-r4k.c +++ b/arch/mips/kernel/cevt-r4k.c @@ -200,7 +200,7 @@ int c0_compare_int_usable(void) */ if (c0_compare_int_pending()) { cnt = read_c0_count(); - write_c0_compare(cnt); + write_c0_compare(cnt - 1); back_to_back_c0_hazard(); while (read_c0_count() < (cnt + COMPARE_INT_SEEN_TICKS)) if (!c0_compare_int_pending()) @@ -228,7 +228,7 @@ int c0_compare_int_usable(void) if (!c0_compare_int_pending()) return 0; cnt = read_c0_count(); - write_c0_compare(cnt); + write_c0_compare(cnt - 1); back_to_back_c0_hazard(); while (read_c0_count() < (cnt + COMPARE_INT_SEEN_TICKS)) if (!c0_compare_int_pending()) From f2b95d7a9fa43bd72d442a9df01e29274b1769b2 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ar=C4=B1n=C3=A7=20=C3=9CNAL?= Date: Tue, 21 Feb 2023 12:24:34 +0300 Subject: [PATCH 541/765] mips: remove SYS_HAS_CPU_MIPS32_R1 from RALINK MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit All MIPS processors on the Ralink SoCs implement the MIPS32 Release 2 Architecture. Remove SYS_HAS_CPU_MIPS32_R1. Signed-off-by: Arınç ÜNAL Acked-by: Sergio Paracuellos Signed-off-by: Thomas Bogendoerfer --- arch/mips/Kconfig | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index a1170f0a0c04d..875c29246555b 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -610,7 +610,6 @@ config RALINK select DMA_NONCOHERENT select IRQ_MIPS_CPU select USE_OF - select SYS_HAS_CPU_MIPS32_R1 select SYS_HAS_CPU_MIPS32_R2 select SYS_SUPPORTS_32BIT_KERNEL select SYS_SUPPORTS_LITTLE_ENDIAN From 27fd82726995bd75b68df9ce6a1eda8f6b3ad498 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ar=C4=B1n=C3=A7=20=C3=9CNAL?= Date: Tue, 21 Feb 2023 12:24:35 +0300 Subject: [PATCH 542/765] mips: ralink: make SOC_MT7621 select PINCTRL MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently, out of every Ralink SoC, only the dt-binding of the MT7621 SoC uses pinctrl. Because of this, PINCTRL is not selected at all. Make SOC_MT7621 select PINCTRL. Remove PINCTRL_MT7621, enabling it for the MT7621 SoC will be handled under the PINCTRL_MT7621 option. Signed-off-by: Arınç ÜNAL Acked-by: Sergio Paracuellos Signed-off-by: Thomas Bogendoerfer --- arch/mips/ralink/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/ralink/Kconfig b/arch/mips/ralink/Kconfig index 06031796c87bd..83e61e147b902 100644 --- a/arch/mips/ralink/Kconfig +++ b/arch/mips/ralink/Kconfig @@ -54,7 +54,7 @@ choice select HAVE_PCI select PCI_DRIVERS_GENERIC select SOC_BUS - select PINCTRL_MT7621 + select PINCTRL help The MT7621 system-on-a-chip includes an 880 MHz MIPS1004Kc From 32ff6831cdecd828bd8be9cdfb6c4a3d1feb8f8a Mon Sep 17 00:00:00 2001 From: David Gow Date: Sat, 25 Feb 2023 09:45:30 +0800 Subject: [PATCH 543/765] kunit: Fix 'hooks.o' build by recursing into kunit KUnit's 'hooks.o' file need to be built-in whenever KUnit is enabled (even if CONFIG_KUNIT=m). We'd previously attemtped to do this by adding 'kunit/hooks.o' to obj-y in lib/Makefile, but this caused hooks.c to be rebuilt even when it was unchanged. Instead, always recurse into lib/kunit using obj-y when KUnit is enabled, and add the hooks there. Fixes: 7170b7ed6acb ("kunit: Add "hooks" to call into KUnit when it's built as a module"). Reported-by: Linus Torvalds Link: https://lore.kernel.org/linux-kselftest/CAHk-=wiEf7irTKwPJ0jTMOF3CS-13UXmF6Fns3wuWpOZ_wGyZQ@mail.gmail.com/ Signed-off-by: David Gow Reviewed-by: Brendan Higgins Signed-off-by: Linus Torvalds --- lib/Makefile | 12 ++++-------- lib/kunit/Makefile | 2 +- 2 files changed, 5 insertions(+), 9 deletions(-) diff --git a/lib/Makefile b/lib/Makefile index 469be6240523f..baf2821f7a00f 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -127,14 +127,10 @@ CFLAGS_test_fpu.o += $(FPU_CFLAGS) obj-$(CONFIG_TEST_LIVEPATCH) += livepatch/ -obj-$(CONFIG_KUNIT) += kunit/ -# Include the KUnit hooks unconditionally. They'll compile to nothing if -# CONFIG_KUNIT=n, otherwise will be a small table of static data (static key, -# function pointers) which need to be built-in even when KUnit is a module. -ifeq ($(CONFIG_KUNIT), m) -obj-y += kunit/hooks.o -else -obj-$(CONFIG_KUNIT) += kunit/hooks.o +# Some KUnit files (hooks.o) need to be built-in even when KUnit is a module, +# so we can't just use obj-$(CONFIG_KUNIT). +ifdef CONFIG_KUNIT +obj-y += kunit/ endif ifeq ($(CONFIG_DEBUG_KOBJECT),y) diff --git a/lib/kunit/Makefile b/lib/kunit/Makefile index da665cd4ea12f..cb417f5049962 100644 --- a/lib/kunit/Makefile +++ b/lib/kunit/Makefile @@ -13,7 +13,7 @@ kunit-objs += debugfs.o endif # KUnit 'hooks' are built-in even when KUnit is built as a module. -lib-y += hooks.o +obj-y += hooks.o obj-$(CONFIG_KUNIT_TEST) += kunit-test.o From 46d733d0efc79bc8430d63b57ab88011806d5180 Mon Sep 17 00:00:00 2001 From: George Kennedy Date: Mon, 27 Feb 2023 15:21:41 -0500 Subject: [PATCH 544/765] vc_screen: modify vcs_size() handling in vcs_read() Restore the vcs_size() handling in vcs_read() to what it had been in previous version. Fixes: 226fae124b2d ("vc_screen: move load of struct vc_data pointer in vcs_read() to avoid UAF") Suggested-by: Jiri Slaby Signed-off-by: George Kennedy Signed-off-by: Linus Torvalds --- drivers/tty/vt/vc_screen.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/tty/vt/vc_screen.c b/drivers/tty/vt/vc_screen.c index f566eb1839dc5..c0a2273bb998b 100644 --- a/drivers/tty/vt/vc_screen.c +++ b/drivers/tty/vt/vc_screen.c @@ -414,10 +414,8 @@ vcs_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) */ size = vcs_size(vc, attr, uni_mode); if (size < 0) { - if (read) - break; ret = size; - goto unlock_out; + break; } if (pos >= size) break; From a4eecbae092759537748360299de03e434c9a956 Mon Sep 17 00:00:00 2001 From: Mateusz Guzik Date: Wed, 25 Jan 2023 16:55:56 +0100 Subject: [PATCH 545/765] capability: add cap_isidentical Signed-off-by: Mateusz Guzik Reviewed-by: Serge Hallyn Cc: Al Viro Signed-off-by: Linus Torvalds --- include/linux/capability.h | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/include/linux/capability.h b/include/linux/capability.h index 03c2a613ad404..d3c6c2d1ff452 100644 --- a/include/linux/capability.h +++ b/include/linux/capability.h @@ -157,6 +157,16 @@ static inline bool cap_isclear(const kernel_cap_t a) return true; } +static inline bool cap_isidentical(const kernel_cap_t a, const kernel_cap_t b) +{ + unsigned __capi; + CAP_FOR_EACH_U32(__capi) { + if (a.cap[__capi] != b.cap[__capi]) + return false; + } + return true; +} + /* * Check if "a" is a subset of "set". * return true if ALL of the capabilities in "a" are also in "set" From 981ee95cc1f5905ae4936b0dd501085909cdc14f Mon Sep 17 00:00:00 2001 From: Mateusz Guzik Date: Wed, 25 Jan 2023 16:55:57 +0100 Subject: [PATCH 546/765] vfs: avoid duplicating creds in faccessat if possible access(2) remains commonly used, for example on exec: access("/etc/ld.so.preload", R_OK) or when running gcc: strace -c gcc empty.c % time seconds usecs/call calls errors syscall ------ ----------- ----------- --------- --------- ---------------- 0.00 0.000000 0 42 26 access It falls down to do_faccessat without the AT_EACCESS flag, which in turn results in allocation of new creds in order to modify fsuid/fsgid and caps. This is a very expensive process single-threaded and most notably multi-threaded, with numerous structures getting refed and unrefed on imminent new cred destruction. Turns out for typical consumers the resulting creds would be identical and this can be checked upfront, avoiding the hard work. An access benchmark plugged into will-it-scale running on Cascade Lake shows: test proc before after access1 1 1310582 2908735 (+121%) # distinct files access1 24 4716491 63822173 (+1353%) # distinct files access2 24 2378041 5370335 (+125%) # same file The above benchmarks are not integrated into will-it-scale, but can be found in a pull request: https://github.com/antonblanchard/will-it-scale/pull/36/files Signed-off-by: Mateusz Guzik Cc: Al Viro Signed-off-by: Linus Torvalds --- fs/open.c | 38 +++++++++++++++++++++++++++++++++++++- 1 file changed, 37 insertions(+), 1 deletion(-) diff --git a/fs/open.c b/fs/open.c index 8038cf6525831..4401a73d4032d 100644 --- a/fs/open.c +++ b/fs/open.c @@ -368,7 +368,37 @@ COMPAT_SYSCALL_DEFINE6(fallocate, int, fd, int, mode, compat_arg_u64_dual(offset * access() needs to use the real uid/gid, not the effective uid/gid. * We do this by temporarily clearing all FS-related capabilities and * switching the fsuid/fsgid around to the real ones. + * + * Creating new credentials is expensive, so we try to skip doing it, + * which we can if the result would match what we already got. */ +static bool access_need_override_creds(int flags) +{ + const struct cred *cred; + + if (flags & AT_EACCESS) + return false; + + cred = current_cred(); + if (!uid_eq(cred->fsuid, cred->uid) || + !gid_eq(cred->fsgid, cred->gid)) + return true; + + if (!issecure(SECURE_NO_SETUID_FIXUP)) { + kuid_t root_uid = make_kuid(cred->user_ns, 0); + if (!uid_eq(cred->uid, root_uid)) { + if (!cap_isclear(cred->cap_effective)) + return true; + } else { + if (!cap_isidentical(cred->cap_effective, + cred->cap_permitted)) + return true; + } + } + + return false; +} + static const struct cred *access_override_creds(void) { const struct cred *old_cred; @@ -378,6 +408,12 @@ static const struct cred *access_override_creds(void) if (!override_cred) return NULL; + /* + * XXX access_need_override_creds performs checks in hopes of skipping + * this work. Make sure it stays in sync if making any changes in this + * routine. + */ + override_cred->fsuid = override_cred->uid; override_cred->fsgid = override_cred->gid; @@ -437,7 +473,7 @@ static long do_faccessat(int dfd, const char __user *filename, int mode, int fla if (flags & AT_EMPTY_PATH) lookup_flags |= LOOKUP_EMPTY; - if (!(flags & AT_EACCESS)) { + if (access_need_override_creds(flags)) { old_cred = access_override_creds(); if (!old_cred) return -ENOMEM; From 4c6759967826b87f56c73e0f1deb7b76379ccd23 Mon Sep 17 00:00:00 2001 From: Vlastimil Babka Date: Mon, 27 Feb 2023 17:00:14 -0800 Subject: [PATCH 547/765] mm/mremap: fix dup_anon_vma() in vma_merge() case 4 In case 4, we are shrinking 'prev' (PPPP in the comment) and expanding 'mid' (NNNN). So we need to make sure 'mid' clones the anon_vma from 'prev', if it doesn't have any. After commit 0503ea8f5ba7 ("mm/mmap: remove __vma_adjust()") we can fail to do that due to wrong parameters for dup_anon_vma(). The call is a no-op because res == next, adjust == mid and mid == next. Fix it. Link: https://lkml.kernel.org/r/ad91d62b-37eb-4b73-707a-3c45c9e16256@suse.cz Fixes: 0503ea8f5ba7 ("mm/mmap: remove __vma_adjust()") Signed-off-by: Vlastimil Babka Reviewed-by: Liam R. Howlett Signed-off-by: Andrew Morton --- mm/mmap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/mmap.c b/mm/mmap.c index 20f21f0949ddb..740b54be3ed41 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -973,7 +973,7 @@ struct vm_area_struct *vma_merge(struct vma_iterator *vmi, struct mm_struct *mm, vma_end = addr; adjust = mid; adj_next = -(vma->vm_end - addr); - err = dup_anon_vma(res, adjust); + err = dup_anon_vma(adjust, prev); } else { vma = next; /* case 3 */ vma_start = addr; From 3f98c9a62c338bbe06a215c9491e6166ea39bf82 Mon Sep 17 00:00:00 2001 From: "andrew.yang" Date: Wed, 22 Feb 2023 14:42:20 +0800 Subject: [PATCH 548/765] mm/damon/paddr: fix missing folio_put() damon_get_folio() would always increase folio _refcount and folio_isolate_lru() would increase folio _refcount if the folio's lru flag is set. If an unevictable folio isolated successfully, there will be two more _refcount. The one from folio_isolate_lru() will be decreased in folio_puback_lru(), but the other one from damon_get_folio() will be left behind. This causes a pin page. Whatever the case, the _refcount from damon_get_folio() should be decreased. Link: https://lkml.kernel.org/r/20230222064223.6735-1-andrew.yang@mediatek.com Fixes: 57223ac29584 ("mm/damon/paddr: support the pageout scheme") Signed-off-by: andrew.yang Reviewed-by: SeongJae Park Cc: [5.16.x] Signed-off-by: Andrew Morton --- mm/damon/paddr.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/mm/damon/paddr.c b/mm/damon/paddr.c index 607bb69e526cf..6c655d9b56391 100644 --- a/mm/damon/paddr.c +++ b/mm/damon/paddr.c @@ -250,12 +250,11 @@ static unsigned long damon_pa_pageout(struct damon_region *r, struct damos *s) folio_put(folio); continue; } - if (folio_test_unevictable(folio)) { + if (folio_test_unevictable(folio)) folio_putback_lru(folio); - } else { + else list_add(&folio->lru, &folio_list); - folio_put(folio); - } + folio_put(folio); } applied = reclaim_pages(&folio_list); cond_resched(); From 1c0a0af511effbd75beffa6c5e54fd64d3563860 Mon Sep 17 00:00:00 2001 From: Mikhail Zaslonko Date: Tue, 21 Feb 2023 14:16:17 +0100 Subject: [PATCH 549/765] lib/zlib: DFLTCC deflate does not write all available bits for Z_NO_FLUSH DFLTCC deflate with Z_NO_FLUSH might generate a corrupted stream when the output buffer is not large enough to fit all the deflate output at once. The problem takes place on closing the deflate block since flush_pending() might leave some output bits not written. Similar problem for software deflate with Z_BLOCK flush option (not supported by kernel zlib deflate) has been fixed a while ago in userspace zlib but the fix never got to the kernel. Now flush_pending() flushes the bit buffer before copying out the byte buffer, in order to really flush as much as possible. Currently there are no users of DFLTCC deflate with Z_NO_FLUSH option in the kernel so the problem remained hidden for a while. This commit is based on the old zlib commit: https://github.com/madler/zlib/commit/0b828b4 Link: https://lkml.kernel.org/r/20230221131617.3369978-2-zaslonko@linux.ibm.com Signed-off-by: Mikhail Zaslonko Acked-by: Ilya Leoshkevich Cc: Heiko Carstens Cc: Vasily Gorbik Signed-off-by: Andrew Morton --- lib/zlib_deflate/defutil.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/lib/zlib_deflate/defutil.h b/lib/zlib_deflate/defutil.h index 385333b22ec68..4ea40f5a279fa 100644 --- a/lib/zlib_deflate/defutil.h +++ b/lib/zlib_deflate/defutil.h @@ -420,9 +420,11 @@ static inline void flush_pending( z_streamp strm ) { + unsigned len; deflate_state *s = (deflate_state *) strm->state; - unsigned len = s->pending; + bi_flush(s); + len = s->pending; if (len > strm->avail_out) len = strm->avail_out; if (len == 0) return; From 6da6b1d4a7df8c35770186b53ef65d388398e139 Mon Sep 17 00:00:00 2001 From: Naoya Horiguchi Date: Tue, 21 Feb 2023 17:59:05 +0900 Subject: [PATCH 550/765] mm/hwpoison: convert TTU_IGNORE_HWPOISON to TTU_HWPOISON After a memory error happens on a clean folio, a process unexpectedly receives SIGBUS when it accesses the error page. This SIGBUS killing is pointless and simply degrades the level of RAS of the system, because the clean folio can be dropped without any data lost on memory error handling as we do for a clean pagecache. When memory_failure() is called on a clean folio, try_to_unmap() is called twice (one from split_huge_page() and one from hwpoison_user_mappings()). The root cause of the issue is that pte conversion to hwpoisoned entry is now done in the first call of try_to_unmap() because PageHWPoison is already set at this point, while it's actually expected to be done in the second call. This behavior disturbs the error handling operation like removing pagecache, which results in the malfunction described above. So convert TTU_IGNORE_HWPOISON into TTU_HWPOISON and set TTU_HWPOISON only when we really intend to convert pte to hwpoison entry. This can prevent other callers of try_to_unmap() from accidentally converting to hwpoison entries. Link: https://lkml.kernel.org/r/20230221085905.1465385-1-naoya.horiguchi@linux.dev Fixes: a42634a6c07d ("readahead: Use a folio in read_pages()") Signed-off-by: Naoya Horiguchi Cc: David Hildenbrand Cc: Hugh Dickins Cc: Matthew Wilcox Cc: Miaohe Lin Cc: Minchan Kim Cc: Vlastimil Babka Cc: Signed-off-by: Andrew Morton --- include/linux/rmap.h | 2 +- mm/memory-failure.c | 8 ++++---- mm/rmap.c | 2 +- 3 files changed, 6 insertions(+), 6 deletions(-) diff --git a/include/linux/rmap.h b/include/linux/rmap.h index a4570da03e58a..b87d016604127 100644 --- a/include/linux/rmap.h +++ b/include/linux/rmap.h @@ -94,7 +94,7 @@ enum ttu_flags { TTU_SPLIT_HUGE_PMD = 0x4, /* split huge PMD if any */ TTU_IGNORE_MLOCK = 0x8, /* ignore mlock */ TTU_SYNC = 0x10, /* avoid racy checks with PVMW_SYNC */ - TTU_IGNORE_HWPOISON = 0x20, /* corrupted page is recoverable */ + TTU_HWPOISON = 0x20, /* do convert pte to hwpoison entry */ TTU_BATCH_FLUSH = 0x40, /* Batch TLB flushes where possible * and caller guarantees they will * do a final flush if necessary */ diff --git a/mm/memory-failure.c b/mm/memory-failure.c index a1ede7bdce95e..fae9baf3be162 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -1069,7 +1069,7 @@ static int me_pagecache_dirty(struct page_state *ps, struct page *p) * cache and swap cache(ie. page is freshly swapped in). So it could be * referenced concurrently by 2 types of PTEs: * normal PTEs and swap PTEs. We try to handle them consistently by calling - * try_to_unmap(TTU_IGNORE_HWPOISON) to convert the normal PTEs to swap PTEs, + * try_to_unmap(!TTU_HWPOISON) to convert the normal PTEs to swap PTEs, * and then * - clear dirty bit to prevent IO * - remove from LRU @@ -1486,7 +1486,7 @@ static bool hwpoison_user_mappings(struct page *p, unsigned long pfn, int flags, struct page *hpage) { struct folio *folio = page_folio(hpage); - enum ttu_flags ttu = TTU_IGNORE_MLOCK | TTU_SYNC; + enum ttu_flags ttu = TTU_IGNORE_MLOCK | TTU_SYNC | TTU_HWPOISON; struct address_space *mapping; LIST_HEAD(tokill); bool unmap_success; @@ -1516,7 +1516,7 @@ static bool hwpoison_user_mappings(struct page *p, unsigned long pfn, if (PageSwapCache(p)) { pr_err("%#lx: keeping poisoned page in swap cache\n", pfn); - ttu |= TTU_IGNORE_HWPOISON; + ttu &= ~TTU_HWPOISON; } /* @@ -1531,7 +1531,7 @@ static bool hwpoison_user_mappings(struct page *p, unsigned long pfn, if (page_mkclean(hpage)) { SetPageDirty(hpage); } else { - ttu |= TTU_IGNORE_HWPOISON; + ttu &= ~TTU_HWPOISON; pr_info("%#lx: corrupted page was clean: dropped without side effects\n", pfn); } diff --git a/mm/rmap.c b/mm/rmap.c index 15ae24585fc49..8632e02661ac7 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -1602,7 +1602,7 @@ static bool try_to_unmap_one(struct folio *folio, struct vm_area_struct *vma, /* Update high watermark before we lower rss */ update_hiwater_rss(mm); - if (PageHWPoison(subpage) && !(flags & TTU_IGNORE_HWPOISON)) { + if (PageHWPoison(subpage) && (flags & TTU_HWPOISON)) { pteval = swp_entry_to_pte(make_hwpoison_entry(subpage)); if (folio_test_hugetlb(folio)) { hugetlb_count_sub(folio_nr_pages(folio), mm); From d49b497060387545ca10210590adef6edb296583 Mon Sep 17 00:00:00 2001 From: Konrad Dybcio Date: Fri, 17 Feb 2023 21:35:16 +0100 Subject: [PATCH 551/765] mailmap: map Georgi Djakov's old Linaro address to his current one Georgi's old email is still picked up by the likes of get_maintainer.pl and it keeps bouncing every time one submits an interconnect patch. Map it to his current @kernel.org one. Link: https://lkml.kernel.org/r/20230217203516.826424-1-konrad.dybcio@linaro.org Signed-off-by: Konrad Dybcio Acked-by: Georgi Djakov Cc: Arnd Bergmann Cc: Baolin Wang Cc: Colin Ian King Cc: Kirill Tkhai Cc: Qais Yousef Cc: Vasily Averin Signed-off-by: Andrew Morton --- .mailmap | 1 + 1 file changed, 1 insertion(+) diff --git a/.mailmap b/.mailmap index a872c9683958c..fb65947d671d8 100644 --- a/.mailmap +++ b/.mailmap @@ -150,6 +150,7 @@ Gao Xiang Gao Xiang Gao Xiang Gao Xiang +Georgi Djakov Gerald Schaefer Gerald Schaefer Gerald Schaefer From 60eed1e3d45045623e46944ebc7c42c30a4350f0 Mon Sep 17 00:00:00 2001 From: Heming Zhao via Ocfs2-devel Date: Fri, 17 Feb 2023 08:37:17 +0800 Subject: [PATCH 552/765] ocfs2: fix defrag path triggering jbd2 ASSERT code path: ocfs2_ioctl_move_extents ocfs2_move_extents ocfs2_defrag_extent __ocfs2_move_extent + ocfs2_journal_access_di + ocfs2_split_extent //sub-paths call jbd2_journal_restart + ocfs2_journal_dirty //crash by jbs2 ASSERT crash stacks: PID: 11297 TASK: ffff974a676dcd00 CPU: 67 COMMAND: "defragfs.ocfs2" #0 [ffffb25d8dad3900] machine_kexec at ffffffff8386fe01 #1 [ffffb25d8dad3958] __crash_kexec at ffffffff8395959d #2 [ffffb25d8dad3a20] crash_kexec at ffffffff8395a45d #3 [ffffb25d8dad3a38] oops_end at ffffffff83836d3f #4 [ffffb25d8dad3a58] do_trap at ffffffff83833205 #5 [ffffb25d8dad3aa0] do_invalid_op at ffffffff83833aa6 #6 [ffffb25d8dad3ac0] invalid_op at ffffffff84200d18 [exception RIP: jbd2_journal_dirty_metadata+0x2ba] RIP: ffffffffc09ca54a RSP: ffffb25d8dad3b70 RFLAGS: 00010207 RAX: 0000000000000000 RBX: ffff9706eedc5248 RCX: 0000000000000000 RDX: 0000000000000001 RSI: ffff97337029ea28 RDI: ffff9706eedc5250 RBP: ffff9703c3520200 R8: 000000000f46b0b2 R9: 0000000000000000 R10: 0000000000000001 R11: 00000001000000fe R12: ffff97337029ea28 R13: 0000000000000000 R14: ffff9703de59bf60 R15: ffff9706eedc5250 ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018 #7 [ffffb25d8dad3ba8] ocfs2_journal_dirty at ffffffffc137fb95 [ocfs2] #8 [ffffb25d8dad3be8] __ocfs2_move_extent at ffffffffc139a950 [ocfs2] #9 [ffffb25d8dad3c80] ocfs2_defrag_extent at ffffffffc139b2d2 [ocfs2] Analysis This bug has the same root cause of 'commit 7f27ec978b0e ("ocfs2: call ocfs2_journal_access_di() before ocfs2_journal_dirty() in ocfs2_write_end_nolock()")'. For this bug, jbd2_journal_restart() is called by ocfs2_split_extent() during defragmenting. How to fix For ocfs2_split_extent() can handle journal operations totally by itself. Caller doesn't need to call journal access/dirty pair, and caller only needs to call journal start/stop pair. The fix method is to remove journal access/dirty from __ocfs2_move_extent(). The discussion for this patch: https://oss.oracle.com/pipermail/ocfs2-devel/2023-February/000647.html Link: https://lkml.kernel.org/r/20230217003717.32469-1-heming.zhao@suse.com Signed-off-by: Heming Zhao Reviewed-by: Joseph Qi Cc: Mark Fasheh Cc: Joel Becker Cc: Junxiao Bi Cc: Changwei Ge Cc: Gang He Cc: Jun Piao Cc: Signed-off-by: Andrew Morton --- fs/ocfs2/move_extents.c | 10 ---------- 1 file changed, 10 deletions(-) diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c index 192cad0662d8b..6251748c695b2 100644 --- a/fs/ocfs2/move_extents.c +++ b/fs/ocfs2/move_extents.c @@ -105,14 +105,6 @@ static int __ocfs2_move_extent(handle_t *handle, */ replace_rec.e_flags = ext_flags & ~OCFS2_EXT_REFCOUNTED; - ret = ocfs2_journal_access_di(handle, INODE_CACHE(inode), - context->et.et_root_bh, - OCFS2_JOURNAL_ACCESS_WRITE); - if (ret) { - mlog_errno(ret); - goto out; - } - ret = ocfs2_split_extent(handle, &context->et, path, index, &replace_rec, context->meta_ac, &context->dealloc); @@ -121,8 +113,6 @@ static int __ocfs2_move_extent(handle_t *handle, goto out; } - ocfs2_journal_dirty(handle, context->et.et_root_bh); - context->new_phys_cpos = new_p_cpos; /* From 236b9254f8d1edc273ad88b420aa85fbd84f492d Mon Sep 17 00:00:00 2001 From: Heming Zhao via Ocfs2-devel Date: Mon, 20 Feb 2023 13:05:26 +0800 Subject: [PATCH 553/765] ocfs2: fix non-auto defrag path not working issue This fixes three issues on move extents ioctl without auto defrag: a) In ocfs2_find_victim_alloc_group(), we have to convert bits to block first in case of global bitmap. b) In ocfs2_probe_alloc_group(), when finding enough bits in block group bitmap, we have to back off move_len to start pos as well, otherwise it may corrupt filesystem. c) In ocfs2_ioctl_move_extents(), set me_threshold both for non-auto and auto defrag paths. Otherwise it will set move_max_hop to 0 and finally cause unexpectedly ENOSPC error. Currently there are no tools triggering the above issues since defragfs.ocfs2 enables auto defrag by default. Tested with manually changing defragfs.ocfs2 to run non auto defrag path. Link: https://lkml.kernel.org/r/20230220050526.22020-1-heming.zhao@suse.com Signed-off-by: Heming Zhao Reviewed-by: Joseph Qi Cc: Mark Fasheh Cc: Joel Becker Cc: Junxiao Bi Cc: Changwei Ge Cc: Gang He Cc: Jun Piao Cc: Signed-off-by: Andrew Morton --- fs/ocfs2/move_extents.c | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-) diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c index 6251748c695b2..b1e32ec4a9d41 100644 --- a/fs/ocfs2/move_extents.c +++ b/fs/ocfs2/move_extents.c @@ -434,7 +434,7 @@ static int ocfs2_find_victim_alloc_group(struct inode *inode, bg = (struct ocfs2_group_desc *)gd_bh->b_data; if (vict_blkno < (le64_to_cpu(bg->bg_blkno) + - le16_to_cpu(bg->bg_bits))) { + (le16_to_cpu(bg->bg_bits) << bits_per_unit))) { *ret_bh = gd_bh; *vict_bit = (vict_blkno - blkno) >> @@ -549,6 +549,7 @@ static void ocfs2_probe_alloc_group(struct inode *inode, struct buffer_head *bh, last_free_bits++; if (last_free_bits == move_len) { + i -= move_len; *goal_bit = i; *phys_cpos = base_cpos + i; break; @@ -1020,18 +1021,19 @@ int ocfs2_ioctl_move_extents(struct file *filp, void __user *argp) context->range = ⦥ + /* + * ok, the default theshold for the defragmentation + * is 1M, since our maximum clustersize was 1M also. + * any thought? + */ + if (!range.me_threshold) + range.me_threshold = 1024 * 1024; + + if (range.me_threshold > i_size_read(inode)) + range.me_threshold = i_size_read(inode); + if (range.me_flags & OCFS2_MOVE_EXT_FL_AUTO_DEFRAG) { context->auto_defrag = 1; - /* - * ok, the default theshold for the defragmentation - * is 1M, since our maximum clustersize was 1M also. - * any thought? - */ - if (!range.me_threshold) - range.me_threshold = 1024 * 1024; - - if (range.me_threshold > i_size_read(inode)) - range.me_threshold = i_size_read(inode); if (range.me_flags & OCFS2_MOVE_EXT_FL_PART_DEFRAG) context->partial = 1; From ae3419fbac845b4d3f3a9fae4cc80c68d82cdf6e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Mon, 20 Feb 2023 06:46:12 +0000 Subject: [PATCH 554/765] vc_screen: don't clobber return value in vcs_read MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit 226fae124b2d ("vc_screen: move load of struct vc_data pointer in vcs_read() to avoid UAF") moved the call to vcs_vc() into the loop. While doing this it also moved the unconditional assignment of ret = -ENXIO; This unconditional assignment was valid outside the loop but within it it clobbers the actual value of ret. To avoid this only assign "ret = -ENXIO" when actually needed. [ Also, the 'goto unlock_out" needs to be just a "break", so that it does the right thing when it exits on later iterations when partial success has happened - Linus ] Reported-by: Storm Dragon Link: https://lore.kernel.org/lkml/Y%2FKS6vdql2pIsCiI@hotmail.com/ Fixes: 226fae124b2d ("vc_screen: move load of struct vc_data pointer in vcs_read() to avoid UAF") Signed-off-by: Thomas Weißschuh Link: https://lore.kernel.org/lkml/64981d94-d00c-4b31-9063-43ad0a384bde@t-8ch.de/ Signed-off-by: Linus Torvalds --- drivers/tty/vt/vc_screen.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/tty/vt/vc_screen.c b/drivers/tty/vt/vc_screen.c index c0a2273bb998b..1dc07f9214d57 100644 --- a/drivers/tty/vt/vc_screen.c +++ b/drivers/tty/vt/vc_screen.c @@ -403,10 +403,11 @@ vcs_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) unsigned int this_round, skip = 0; int size; - ret = -ENXIO; vc = vcs_vc(inode, &viewed); - if (!vc) - goto unlock_out; + if (!vc) { + ret = -ENXIO; + break; + } /* Check whether we are above size each round, * as copy_to_user at the end of this loop From d8c32853ebc4f3a49f1658ddfe0a9a42e19fc658 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 21 Feb 2023 21:19:52 +1100 Subject: [PATCH 555/765] powerpc: Drop orphaned VAS MAINTAINERS entry The MAINTAINERS entry for VAS (Virtual Accelerator Switchboard) no longer has any maintainers, it just points to linuxppc-dev, since commit 60496069d0ae ("powerpc: Update MAINTAINERS for ibmvnic and VAS"). So just drop the VAS entry, all the paths are already covered by the main powerpc entry, ie. the output of get_maintainer.pl is unchanged. Signed-off-by: Michael Ellerman Link: https://lore.kernel.org/r/20230221101952.2697101-1-mpe@ellerman.id.au --- MAINTAINERS | 7 ------- 1 file changed, 7 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 324a857e07845..bc24b2e8f0885 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -9789,13 +9789,6 @@ L: netdev@vger.kernel.org S: Supported F: drivers/net/ethernet/ibm/ibmvnic.* -IBM Power Virtual Accelerator Switchboard -L: linuxppc-dev@lists.ozlabs.org -S: Supported -F: arch/powerpc/include/asm/vas.h -F: arch/powerpc/platforms/powernv/copy-paste.h -F: arch/powerpc/platforms/powernv/vas* - IBM Power Virtual Ethernet Device Driver M: Nick Child L: netdev@vger.kernel.org From acd35dbab871d61021284ff06daccdc0ebb51e61 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Wed, 22 Feb 2023 17:00:37 +1100 Subject: [PATCH 556/765] powerpc/vmlinux.lds: Add .text.asan/tsan sections When KASAN/KCSAN are enabled clang generates .text.asan/tsan sections. Because they are not mentioned in the linker script warnings are generated, and when orphan handling is set to error that becomes a build error, eg: ld.lld: error: vmlinux.a(init/main.o):(.text.tsan.module_ctor) is being placed in '.text.tsan.module_ctor' ld.lld: error: vmlinux.a(init/version.o):(.text.tsan.module_ctor) is being placed in '.text.tsan.module_ctor' Fix it by adding the sections to our linker script, similar to the generic change made in 848378812e40 ("vmlinux.lds.h: Handle clang's module.{c,d}tor sections"). Reviewed-by: Nathan Chancellor Signed-off-by: Michael Ellerman Link: https://lore.kernel.org/r/20230222060037.2897169-1-mpe@ellerman.id.au --- arch/powerpc/kernel/vmlinux.lds.S | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/powerpc/kernel/vmlinux.lds.S b/arch/powerpc/kernel/vmlinux.lds.S index f128c7cf9c1da..ee86753e444ea 100644 --- a/arch/powerpc/kernel/vmlinux.lds.S +++ b/arch/powerpc/kernel/vmlinux.lds.S @@ -124,6 +124,7 @@ SECTIONS * included with the main text sections, so put it by itself. */ *(.sfpr); + *(.text.asan.* .text.tsan.*) MEM_KEEP(init.text) MEM_KEEP(exit.text) } :text From f8b2336f15f3bc30e37ce5c052cde5b6319bb6df Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Wed, 22 Feb 2023 00:03:31 +1100 Subject: [PATCH 557/765] powerpc: Avoid dead code/data elimination when using recordmcount Although powerpc now has objtool mcount support, it's not enabled in all configurations due to dependencies. On those configurations, with some linkers (binutils 2.37 at least), it's still possible to hit the dreaded "recordmcount bug", eg. errors such as: CC kernel/kexec_file.o Cannot find symbol for section 10: .text.unlikely. kernel/kexec_file.o: failed make[1]: *** [scripts/Makefile.build:287 : kernel/kexec_file.o] Error 1 Those errors are much more prevalent when building with CONFIG_LD_DEAD_CODE_DATA_ELIMINATION, because it places every function in a separate section. CONFIG_LD_DEAD_CODE_DATA_ELIMINATION is marked experimental and is not enabled in any powerpc defconfigs or by major distros. Although it does have at least some users on 32-bit where kernel size tends to be more important. Avoid the build errors by blocking CONFIG_LD_DEAD_CODE_DATA_ELIMINATION when the build is using recordmcount, rather than objtool. In practice that means for 64-bit big endian builds, or 64-bit clang builds - both because they lack CONFIG_MPROFILE_KERNEL. On 32-bit objtool is always used, so CONFIG_LD_DEAD_CODE_DATA_ELIMINATION is still available there. Signed-off-by: Michael Ellerman Link: https://lore.kernel.org/r/20230221130331.2714199-1-mpe@ellerman.id.au --- arch/powerpc/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 2c9cdf1d87612..a6c4407d3ec83 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -236,7 +236,7 @@ config PPC select HAVE_KPROBES select HAVE_KPROBES_ON_FTRACE select HAVE_KRETPROBES - select HAVE_LD_DEAD_CODE_DATA_ELIMINATION + select HAVE_LD_DEAD_CODE_DATA_ELIMINATION if HAVE_OBJTOOL_MCOUNT select HAVE_LIVEPATCH if HAVE_DYNAMIC_FTRACE_WITH_REGS select HAVE_MOD_ARCH_SPECIFIC select HAVE_NMI if PERF_EVENTS || (PPC64 && PPC_BOOK3S) From 660ca9470f9c613fa2c71a123a9469c80a697ee4 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Mon, 27 Feb 2023 16:25:58 +0800 Subject: [PATCH 558/765] crypto: caam - Fix edesc/iv ordering mixup The attempt to add DMA alignment padding by moving IV to the front of edesc was completely broken as it didn't change the places where edesc was freed. It's also wrong as the IV may still share a cache-line with the edesc. Fix this by restoring the original layout and simply reserving enough memmory so that the IV is on a DMA cache-line by itself. Reported-by: Meenakshi Aggarwal Fixes: 199354d7fb6e ("crypto: caam - Remove GFP_DMA and add DMA alignment padding") Signed-off-by: Herbert Xu --- drivers/crypto/caam/caamalg.c | 26 +++++++++++++++------ drivers/crypto/caam/caamalg_qi.c | 40 +++++++++++++++++++++----------- drivers/crypto/caam/qi.c | 10 ++++++-- 3 files changed, 53 insertions(+), 23 deletions(-) diff --git a/drivers/crypto/caam/caamalg.c b/drivers/crypto/caam/caamalg.c index 4a9b998a8d268..12b1c8346243d 100644 --- a/drivers/crypto/caam/caamalg.c +++ b/drivers/crypto/caam/caamalg.c @@ -60,7 +60,11 @@ #include #include #include +#include +#include #include +#include +#include /* * crypto alg @@ -1000,6 +1004,13 @@ static void aead_crypt_done(struct device *jrdev, u32 *desc, u32 err, crypto_finalize_aead_request(jrp->engine, req, ecode); } +static inline u8 *skcipher_edesc_iv(struct skcipher_edesc *edesc) +{ + + return PTR_ALIGN((u8 *)edesc->sec4_sg + edesc->sec4_sg_bytes, + dma_get_cache_alignment()); +} + static void skcipher_crypt_done(struct device *jrdev, u32 *desc, u32 err, void *context) { @@ -1027,8 +1038,7 @@ static void skcipher_crypt_done(struct device *jrdev, u32 *desc, u32 err, * This is used e.g. by the CTS mode. */ if (ivsize && !ecode) { - memcpy(req->iv, (u8 *)edesc->sec4_sg + edesc->sec4_sg_bytes, - ivsize); + memcpy(req->iv, skcipher_edesc_iv(edesc), ivsize); print_hex_dump_debug("dstiv @" __stringify(__LINE__)": ", DUMP_PREFIX_ADDRESS, 16, 4, req->iv, @@ -1683,18 +1693,19 @@ static struct skcipher_edesc *skcipher_edesc_alloc(struct skcipher_request *req, /* * allocate space for base edesc and hw desc commands, link tables, IV */ - aligned_size = ALIGN(ivsize, __alignof__(*edesc)); - aligned_size += sizeof(*edesc) + desc_bytes + sec4_sg_bytes; + aligned_size = sizeof(*edesc) + desc_bytes + sec4_sg_bytes; aligned_size = ALIGN(aligned_size, dma_get_cache_alignment()); - iv = kzalloc(aligned_size, flags); - if (!iv) { + aligned_size += ~(ARCH_KMALLOC_MINALIGN - 1) & + (dma_get_cache_alignment() - 1); + aligned_size += ALIGN(ivsize, dma_get_cache_alignment()); + edesc = kzalloc(aligned_size, flags); + if (!edesc) { dev_err(jrdev, "could not allocate extended descriptor\n"); caam_unmap(jrdev, req->src, req->dst, src_nents, dst_nents, 0, 0, 0, 0); return ERR_PTR(-ENOMEM); } - edesc = (void *)(iv + ALIGN(ivsize, __alignof__(*edesc))); edesc->src_nents = src_nents; edesc->dst_nents = dst_nents; edesc->mapped_src_nents = mapped_src_nents; @@ -1706,6 +1717,7 @@ static struct skcipher_edesc *skcipher_edesc_alloc(struct skcipher_request *req, /* Make sure IV is located in a DMAable area */ if (ivsize) { + iv = skcipher_edesc_iv(edesc); memcpy(iv, req->iv, ivsize); iv_dma = dma_map_single(jrdev, iv, ivsize, DMA_BIDIRECTIONAL); diff --git a/drivers/crypto/caam/caamalg_qi.c b/drivers/crypto/caam/caamalg_qi.c index 5e218bf20d5bb..743ce50c14f2e 100644 --- a/drivers/crypto/caam/caamalg_qi.c +++ b/drivers/crypto/caam/caamalg_qi.c @@ -20,8 +20,11 @@ #include "caamalg_desc.h" #include #include +#include +#include #include #include +#include /* * crypto alg @@ -1204,6 +1207,12 @@ static int ipsec_gcm_decrypt(struct aead_request *req) false); } +static inline u8 *skcipher_edesc_iv(struct skcipher_edesc *edesc) +{ + return PTR_ALIGN((u8 *)&edesc->sgt[0] + edesc->qm_sg_bytes, + dma_get_cache_alignment()); +} + static void skcipher_done(struct caam_drv_req *drv_req, u32 status) { struct skcipher_edesc *edesc; @@ -1236,8 +1245,7 @@ static void skcipher_done(struct caam_drv_req *drv_req, u32 status) * This is used e.g. by the CTS mode. */ if (!ecode) - memcpy(req->iv, (u8 *)&edesc->sgt[0] + edesc->qm_sg_bytes, - ivsize); + memcpy(req->iv, skcipher_edesc_iv(edesc), ivsize); qi_cache_free(edesc); skcipher_request_complete(req, ecode); @@ -1259,6 +1267,7 @@ static struct skcipher_edesc *skcipher_edesc_alloc(struct skcipher_request *req, int dst_sg_idx, qm_sg_ents, qm_sg_bytes; struct qm_sg_entry *sg_table, *fd_sgt; struct caam_drv_ctx *drv_ctx; + unsigned int len; drv_ctx = get_drv_ctx(ctx, encrypt ? ENCRYPT : DECRYPT); if (IS_ERR(drv_ctx)) @@ -1319,9 +1328,12 @@ static struct skcipher_edesc *skcipher_edesc_alloc(struct skcipher_request *req, qm_sg_ents = 1 + pad_sg_nents(qm_sg_ents); qm_sg_bytes = qm_sg_ents * sizeof(struct qm_sg_entry); - if (unlikely(ALIGN(ivsize, __alignof__(*edesc)) + - offsetof(struct skcipher_edesc, sgt) + qm_sg_bytes > - CAAM_QI_MEMCACHE_SIZE)) { + + len = offsetof(struct skcipher_edesc, sgt) + qm_sg_bytes; + len = ALIGN(len, dma_get_cache_alignment()); + len += ivsize; + + if (unlikely(len > CAAM_QI_MEMCACHE_SIZE)) { dev_err(qidev, "No space for %d S/G entries and/or %dB IV\n", qm_sg_ents, ivsize); caam_unmap(qidev, req->src, req->dst, src_nents, dst_nents, 0, @@ -1330,18 +1342,24 @@ static struct skcipher_edesc *skcipher_edesc_alloc(struct skcipher_request *req, } /* allocate space for base edesc, link tables and IV */ - iv = qi_cache_alloc(flags); - if (unlikely(!iv)) { + edesc = qi_cache_alloc(flags); + if (unlikely(!edesc)) { dev_err(qidev, "could not allocate extended descriptor\n"); caam_unmap(qidev, req->src, req->dst, src_nents, dst_nents, 0, 0, DMA_NONE, 0, 0); return ERR_PTR(-ENOMEM); } - edesc = (void *)(iv + ALIGN(ivsize, __alignof__(*edesc))); + edesc->src_nents = src_nents; + edesc->dst_nents = dst_nents; + edesc->qm_sg_bytes = qm_sg_bytes; + edesc->drv_req.app_ctx = req; + edesc->drv_req.cbk = skcipher_done; + edesc->drv_req.drv_ctx = drv_ctx; /* Make sure IV is located in a DMAable area */ sg_table = &edesc->sgt[0]; + iv = skcipher_edesc_iv(edesc); memcpy(iv, req->iv, ivsize); iv_dma = dma_map_single(qidev, iv, ivsize, DMA_BIDIRECTIONAL); @@ -1353,13 +1371,7 @@ static struct skcipher_edesc *skcipher_edesc_alloc(struct skcipher_request *req, return ERR_PTR(-ENOMEM); } - edesc->src_nents = src_nents; - edesc->dst_nents = dst_nents; edesc->iv_dma = iv_dma; - edesc->qm_sg_bytes = qm_sg_bytes; - edesc->drv_req.app_ctx = req; - edesc->drv_req.cbk = skcipher_done; - edesc->drv_req.drv_ctx = drv_ctx; dma_to_qm_sg_one(sg_table, iv_dma, ivsize, 0); sg_to_qm_sg(req->src, req->cryptlen, sg_table + 1, 0); diff --git a/drivers/crypto/caam/qi.c b/drivers/crypto/caam/qi.c index 4c52c9365558d..2ad2c10358563 100644 --- a/drivers/crypto/caam/qi.c +++ b/drivers/crypto/caam/qi.c @@ -8,7 +8,13 @@ */ #include +#include +#include +#include #include +#include +#include +#include #include #include "debugfs.h" @@ -755,8 +761,8 @@ int caam_qi_init(struct platform_device *caam_pdev) napi_enable(irqtask); } - qi_cache = kmem_cache_create("caamqicache", CAAM_QI_MEMCACHE_SIZE, 0, - 0, NULL); + qi_cache = kmem_cache_create("caamqicache", CAAM_QI_MEMCACHE_SIZE, + dma_get_cache_alignment(), 0, NULL); if (!qi_cache) { dev_err(qidev, "Can't allocate CAAM cache\n"); free_rsp_fqs(); From c176060a4c76ed0043cb9c10435af04ed1ad0560 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Fri, 24 Feb 2023 10:25:12 -0700 Subject: [PATCH 559/765] drm: omapdrm: Do not use helper unininitialized in omap_fbdev_init() Clang warns (or errors with CONFIG_WERROR): ../drivers/gpu/drm/omapdrm/omap_fbdev.c:235:6: error: variable 'helper' is used uninitialized whenever 'if' condition is true [-Werror,-Wsometimes-uninitialized] if (!fbdev) ^~~~~~ ../drivers/gpu/drm/omapdrm/omap_fbdev.c:259:26: note: uninitialized use occurs here drm_fb_helper_unprepare(helper); ^~~~~~ ../drivers/gpu/drm/omapdrm/omap_fbdev.c:235:2: note: remove the 'if' if its condition is always false if (!fbdev) ^~~~~~~~~~~ ../drivers/gpu/drm/omapdrm/omap_fbdev.c:228:30: note: initialize the variable 'helper' to silence this warning struct drm_fb_helper *helper; ^ = NULL 1 error generated. Return early, as there is nothing for the function to do if memory cannot be allocated. There is no point in adding another label to just emit the warning at the end of the function in this case, as memory allocation failures are already logged. Fixes: 3fb1f62f80a1 ("drm/fb-helper: Remove drm_fb_helper_unprepare() from drm_fb_helper_fini()") Link: https://github.com/ClangBuiltLinux/linux/issues/1809 Link: https://lore.kernel.org/oe-kbuild-all/202302250058.fYTe9aTP-lkp@intel.com/ Reported-by: kernel test robot Signed-off-by: Nathan Chancellor Signed-off-by: Thomas Zimmermann Link: https://patchwork.freedesktop.org/patch/msgid/20230224-omapdrm-wsometimes-uninitialized-v1-1-3fec8906ee3a@kernel.org --- drivers/gpu/drm/omapdrm/omap_fbdev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/omapdrm/omap_fbdev.c b/drivers/gpu/drm/omapdrm/omap_fbdev.c index 84429728347f7..a6c8542087ecf 100644 --- a/drivers/gpu/drm/omapdrm/omap_fbdev.c +++ b/drivers/gpu/drm/omapdrm/omap_fbdev.c @@ -233,7 +233,7 @@ void omap_fbdev_init(struct drm_device *dev) fbdev = kzalloc(sizeof(*fbdev), GFP_KERNEL); if (!fbdev) - goto fail; + return; INIT_WORK(&fbdev->work, pan_worker); From 047a754558d640eaa080fce3b22ca9f3d4e04626 Mon Sep 17 00:00:00 2001 From: Asahi Lina Date: Mon, 27 Feb 2023 18:04:21 +0900 Subject: [PATCH 560/765] drm/shmem-helper: Revert accidental non-GPL export The referenced commit added a wrapper for drm_gem_shmem_get_pages_sgt(), but in the process it accidentally changed the export type from GPL to non-GPL. Switch it back to GPL. Reported-by: Dmitry Osipenko Fixes: ddddedaa0db9 ("drm/shmem-helper: Fix locking for drm_gem_shmem_get_pages_sgt()") Signed-off-by: Asahi Lina Signed-off-by: Thomas Zimmermann Link: https://patchwork.freedesktop.org/patch/msgid/20230227-shmem-export-fix-v1-1-8880b2c25e81@asahilina.net --- drivers/gpu/drm/drm_gem_shmem_helper.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/drm_gem_shmem_helper.c b/drivers/gpu/drm/drm_gem_shmem_helper.c index 259176d78f3b9..b05dc62418f7d 100644 --- a/drivers/gpu/drm/drm_gem_shmem_helper.c +++ b/drivers/gpu/drm/drm_gem_shmem_helper.c @@ -744,7 +744,7 @@ struct sg_table *drm_gem_shmem_get_pages_sgt(struct drm_gem_shmem_object *shmem) return sgt; } -EXPORT_SYMBOL(drm_gem_shmem_get_pages_sgt); +EXPORT_SYMBOL_GPL(drm_gem_shmem_get_pages_sgt); /** * drm_gem_shmem_prime_import_sg_table - Produce a shmem GEM object from From b3f11af9b2ce14d4662753d097a21e1d37a06fda Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Mon, 27 Feb 2023 11:58:19 +0000 Subject: [PATCH 561/765] arm64: ftrace: forbid CALL_OPS with CC_OPTIMIZE_FOR_SIZE Florian reports that when building with CONFIG_CC_OPTIMIZE_FOR_SIZE=y, he sees "Misaligned patch-site" warnings at boot, e.g. | Misaligned patch-site bcm2836_arm_irqchip_handle_irq+0x0/0x88 | WARNING: CPU: 0 PID: 0 at arch/arm64/kernel/ftrace.c:120 ftrace_call_adjust+0x4c/0x70 This is because GCC will silently ignore `-falign-functions=N` when passed `-Os`, resulting in functions not being aligned as we expect. This is a known issue, and to account for this we modified the kernel to avoid `-Os` generally. Unfortunately we forgot to account for CONFIG_CC_OPTIMIZE_FOR_SIZE. Forbid the use of CALL_OPS with CONFIG_CC_OPTIMIZE_FOR_SIZE=y to prevent this issue. All exising ftrace features will work as before, though without the performance benefit of CALL_OPS. Reported-by: Florian Fainelli Link: http://lore.kernel.org/linux-arm-kernel/2d9284c3-3805-402b-5423-520ced56d047@gmail.com Signed-off-by: Mark Rutland Cc: Marc Zyngier Cc: Stefan Wahren Cc: Steven Rostedt Cc: Will Deacon Tested-by: Florian Fainelli Link: https://lore.kernel.org/r/20230227115819.365630-1-mark.rutland@arm.com Signed-off-by: Catalin Marinas --- arch/arm64/Kconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index 71c35178e017d..f7521695a4742 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -188,7 +188,8 @@ config ARM64 select HAVE_DYNAMIC_FTRACE_WITH_ARGS \ if $(cc-option,-fpatchable-function-entry=2) select HAVE_DYNAMIC_FTRACE_WITH_CALL_OPS \ - if (DYNAMIC_FTRACE_WITH_ARGS && !CFI_CLANG) + if (DYNAMIC_FTRACE_WITH_ARGS && !CFI_CLANG && \ + !CC_OPTIMIZE_FOR_SIZE) select FTRACE_MCOUNT_USE_PATCHABLE_FUNCTION_ENTRY \ if DYNAMIC_FTRACE_WITH_ARGS select HAVE_EFFICIENT_UNALIGNED_ACCESS From f99e6d7c4ed3be2531bd576425a5bd07fb133bd7 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Mon, 27 Feb 2023 10:11:56 +0100 Subject: [PATCH 562/765] bgmac: fix *initial* chip reset to support BCM5358 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit While bringing hardware up we should perform a full reset including the switch bit (BGMAC_BCMA_IOCTL_SW_RESET aka SICF_SWRST). It's what specification says and what reference driver does. This seems to be critical for the BCM5358. Without this hardware doesn't get initialized properly and doesn't seem to transmit or receive any packets. Originally bgmac was calling bgmac_chip_reset() before setting "has_robosw" property which resulted in expected behaviour. That has changed as a side effect of adding platform device support which regressed BCM5358 support. Fixes: f6a95a24957a ("net: ethernet: bgmac: Add platform device support") Cc: Jon Mason Signed-off-by: Rafał Miłecki Reviewed-by: Leon Romanovsky Reviewed-by: Florian Fainelli Link: https://lore.kernel.org/r/20230227091156.19509-1-zajec5@gmail.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/broadcom/bgmac.c | 8 ++++++-- drivers/net/ethernet/broadcom/bgmac.h | 2 ++ 2 files changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c index 3038386a5afd8..1761df8fb7f96 100644 --- a/drivers/net/ethernet/broadcom/bgmac.c +++ b/drivers/net/ethernet/broadcom/bgmac.c @@ -890,13 +890,13 @@ static void bgmac_chip_reset_idm_config(struct bgmac *bgmac) if (iost & BGMAC_BCMA_IOST_ATTACHED) { flags = BGMAC_BCMA_IOCTL_SW_CLKEN; - if (!bgmac->has_robosw) + if (bgmac->in_init || !bgmac->has_robosw) flags |= BGMAC_BCMA_IOCTL_SW_RESET; } bgmac_clk_enable(bgmac, flags); } - if (iost & BGMAC_BCMA_IOST_ATTACHED && !bgmac->has_robosw) + if (iost & BGMAC_BCMA_IOST_ATTACHED && (bgmac->in_init || !bgmac->has_robosw)) bgmac_idm_write(bgmac, BCMA_IOCTL, bgmac_idm_read(bgmac, BCMA_IOCTL) & ~BGMAC_BCMA_IOCTL_SW_RESET); @@ -1490,6 +1490,8 @@ int bgmac_enet_probe(struct bgmac *bgmac) struct net_device *net_dev = bgmac->net_dev; int err; + bgmac->in_init = true; + bgmac_chip_intrs_off(bgmac); net_dev->irq = bgmac->irq; @@ -1542,6 +1544,8 @@ int bgmac_enet_probe(struct bgmac *bgmac) /* Omit FCS from max MTU size */ net_dev->max_mtu = BGMAC_RX_MAX_FRAME_SIZE - ETH_FCS_LEN; + bgmac->in_init = false; + err = register_netdev(bgmac->net_dev); if (err) { dev_err(bgmac->dev, "Cannot register net device\n"); diff --git a/drivers/net/ethernet/broadcom/bgmac.h b/drivers/net/ethernet/broadcom/bgmac.h index e05ac92c06504..d73ef262991d6 100644 --- a/drivers/net/ethernet/broadcom/bgmac.h +++ b/drivers/net/ethernet/broadcom/bgmac.h @@ -472,6 +472,8 @@ struct bgmac { int irq; u32 int_mask; + bool in_init; + /* Current MAC state */ int mac_speed; int mac_duplex; From 11f180a5d62a51b484e9648f9b310e1bd50b1a57 Mon Sep 17 00:00:00 2001 From: Kang Chen Date: Mon, 27 Feb 2023 17:30:37 +0800 Subject: [PATCH 563/765] nfc: fdp: add null check of devm_kmalloc_array in fdp_nci_i2c_read_device_properties devm_kmalloc_array may fails, *fw_vsc_cfg might be null and cause out-of-bounds write in device_property_read_u8_array later. Fixes: a06347c04c13 ("NFC: Add Intel Fields Peak NFC solution driver") Signed-off-by: Kang Chen Reviewed-by: Krzysztof Kozlowski Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20230227093037.907654-1-void0red@gmail.com Signed-off-by: Paolo Abeni --- drivers/nfc/fdp/i2c.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/nfc/fdp/i2c.c b/drivers/nfc/fdp/i2c.c index 2d53e0f88d2f9..1e0f2297f9c66 100644 --- a/drivers/nfc/fdp/i2c.c +++ b/drivers/nfc/fdp/i2c.c @@ -247,6 +247,9 @@ static void fdp_nci_i2c_read_device_properties(struct device *dev, len, sizeof(**fw_vsc_cfg), GFP_KERNEL); + if (!*fw_vsc_cfg) + goto alloc_err; + r = device_property_read_u8_array(dev, FDP_DP_FW_VSC_CFG_NAME, *fw_vsc_cfg, len); @@ -260,6 +263,7 @@ static void fdp_nci_i2c_read_device_properties(struct device *dev, *fw_vsc_cfg = NULL; } +alloc_err: dev_dbg(dev, "Clock type: %d, clock frequency: %d, VSC: %s", *clock_type, *clock_freq, *fw_vsc_cfg != NULL ? "yes" : "no"); } From 8d2909eeca5ef41f64365f700b1fdde361086609 Mon Sep 17 00:00:00 2001 From: Yuezhang Mo Date: Mon, 6 Feb 2023 18:25:20 +0800 Subject: [PATCH 564/765] exfat: remove unneeded code from exfat_alloc_cluster() In the removed code, num_clusters is 0, nothing is done in exfat_chain_cont_cluster(), so it is unneeded, remove it. Signed-off-by: Yuezhang Mo Reviewed-by: Andy Wu Reviewed-by: Sungjong Seo Signed-off-by: Namjae Jeon --- fs/exfat/fatent.c | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-) diff --git a/fs/exfat/fatent.c b/fs/exfat/fatent.c index 41ae4cce1f420..65a8c9fb072c4 100644 --- a/fs/exfat/fatent.c +++ b/fs/exfat/fatent.c @@ -347,14 +347,7 @@ int exfat_alloc_cluster(struct inode *inode, unsigned int num_alloc, exfat_err(sb, "hint_cluster is invalid (%u)", hint_clu); hint_clu = EXFAT_FIRST_CLUSTER; - if (p_chain->flags == ALLOC_NO_FAT_CHAIN) { - if (exfat_chain_cont_cluster(sb, p_chain->dir, - num_clusters)) { - ret = -EIO; - goto unlock; - } - p_chain->flags = ALLOC_FAT_CHAIN; - } + p_chain->flags = ALLOC_FAT_CHAIN; } p_chain->dir = EXFAT_EOF_CLUSTER; From 3ce937cb8ca9becec406611c6072e93030fdde76 Mon Sep 17 00:00:00 2001 From: Yuezhang Mo Date: Thu, 9 Feb 2023 17:44:50 +0800 Subject: [PATCH 565/765] exfat: don't print error log in normal case When allocating a new cluster, exFAT first allocates from the next cluster of the last cluster of the file. If the last cluster of the file is the last cluster of the volume, allocate from the first cluster. This is a normal case, but the following error log will be printed. It makes users confused, so this commit removes the error log. [1960905.181545] exFAT-fs (sdb1): hint_cluster is invalid (262130) Signed-off-by: Yuezhang Mo Reviewed-by: Andy Wu Reviewed-by: Sungjong Seo Signed-off-by: Namjae Jeon --- fs/exfat/fatent.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/fs/exfat/fatent.c b/fs/exfat/fatent.c index 65a8c9fb072c4..c75c5a2cad425 100644 --- a/fs/exfat/fatent.c +++ b/fs/exfat/fatent.c @@ -344,8 +344,9 @@ int exfat_alloc_cluster(struct inode *inode, unsigned int num_alloc, /* check cluster validation */ if (!is_valid_cluster(sbi, hint_clu)) { - exfat_err(sb, "hint_cluster is invalid (%u)", - hint_clu); + if (hint_clu != sbi->num_clusters) + exfat_err(sb, "hint_cluster is invalid (%u), rewind to the first cluster", + hint_clu); hint_clu = EXFAT_FIRST_CLUSTER; p_chain->flags = ALLOC_FAT_CHAIN; } From d5c514b6a0c0b77ed7e5ef2484e8b20eb09c5f27 Mon Sep 17 00:00:00 2001 From: Yuezhang Mo Date: Fri, 10 Feb 2023 13:33:44 +0800 Subject: [PATCH 566/765] exfat: fix the newly allocated clusters are not freed in error handling In error handling 'free_cluster', before num_alloc clusters allocated, p_chain->size will not updated and always 0, thus the newly allocated clusters are not freed. Signed-off-by: Yuezhang Mo Reviewed-by: Andy Wu Reviewed-by: Sungjong Seo Signed-off-by: Namjae Jeon --- fs/exfat/fatent.c | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-) diff --git a/fs/exfat/fatent.c b/fs/exfat/fatent.c index c75c5a2cad425..56b870d9cc0de 100644 --- a/fs/exfat/fatent.c +++ b/fs/exfat/fatent.c @@ -307,7 +307,7 @@ int exfat_alloc_cluster(struct inode *inode, unsigned int num_alloc, struct exfat_chain *p_chain, bool sync_bmap) { int ret = -ENOSPC; - unsigned int num_clusters = 0, total_cnt; + unsigned int total_cnt; unsigned int hint_clu, new_clu, last_clu = EXFAT_EOF_CLUSTER; struct super_block *sb = inode->i_sb; struct exfat_sb_info *sbi = EXFAT_SB(sb); @@ -358,7 +358,7 @@ int exfat_alloc_cluster(struct inode *inode, unsigned int num_alloc, if (new_clu != hint_clu && p_chain->flags == ALLOC_NO_FAT_CHAIN) { if (exfat_chain_cont_cluster(sb, p_chain->dir, - num_clusters)) { + p_chain->size)) { ret = -EIO; goto free_cluster; } @@ -371,8 +371,6 @@ int exfat_alloc_cluster(struct inode *inode, unsigned int num_alloc, goto free_cluster; } - num_clusters++; - /* update FAT table */ if (p_chain->flags == ALLOC_FAT_CHAIN) { if (exfat_ent_set(sb, new_clu, EXFAT_EOF_CLUSTER)) { @@ -389,13 +387,14 @@ int exfat_alloc_cluster(struct inode *inode, unsigned int num_alloc, goto free_cluster; } } + p_chain->size++; + last_clu = new_clu; - if (--num_alloc == 0) { + if (p_chain->size == num_alloc) { sbi->clu_srch_ptr = hint_clu; - sbi->used_clusters += num_clusters; + sbi->used_clusters += num_alloc; - p_chain->size += num_clusters; mutex_unlock(&sbi->bitmap_lock); return 0; } @@ -406,7 +405,7 @@ int exfat_alloc_cluster(struct inode *inode, unsigned int num_alloc, if (p_chain->flags == ALLOC_NO_FAT_CHAIN) { if (exfat_chain_cont_cluster(sb, p_chain->dir, - num_clusters)) { + p_chain->size)) { ret = -EIO; goto free_cluster; } @@ -415,8 +414,7 @@ int exfat_alloc_cluster(struct inode *inode, unsigned int num_alloc, } } free_cluster: - if (num_clusters) - __exfat_free_cluster(inode, p_chain); + __exfat_free_cluster(inode, p_chain); unlock: mutex_unlock(&sbi->bitmap_lock); return ret; From 010338d729c1090036eb40d2a60b7b7bce2445b8 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Thu, 23 Feb 2023 21:41:01 +0100 Subject: [PATCH 567/765] arm64: kaslr: don't pretend KASLR is enabled if offset < MIN_KIMG_ALIGN Our virtual KASLR displacement is a randomly chosen multiple of 2 MiB plus an offset that is equal to the physical placement modulo 2 MiB. This arrangement ensures that we can always use 2 MiB block mappings (or contiguous PTE mappings for 16k or 64k pages) to map the kernel. This means that a KASLR offset of less than 2 MiB is simply the product of this physical displacement, and no randomization has actually taken place. Currently, we use 'kaslr_offset() > 0' to decide whether or not randomization has occurred, and so we misidentify this case. If the kernel image placement is not randomized, modules are allocated from a dedicated region below the kernel mapping, which is only used for modules and not for other vmalloc() or vmap() calls. When randomization is enabled, the kernel image is vmap()'ed randomly inside the vmalloc region, and modules are allocated in the vicinity of this mapping to ensure that relative references are always in range. However, unlike the dedicated module region below the vmalloc region, this region is not reserved exclusively for modules, and so ordinary vmalloc() calls may end up overlapping with it. This should rarely happen, given that vmalloc allocates bottom up, although it cannot be ruled out entirely. The misidentified case results in a placement of the kernel image within 2 MiB of its default address. However, the logic that randomizes the module region is still invoked, and this could result in the module region overlapping with the start of the vmalloc region, instead of using the dedicated region below it. If this happens, a single large vmalloc() or vmap() call will use up the entire region, and leave no space for loading modules after that. Since commit 82046702e288 ("efi/libstub/arm64: Replace 'preferred' offset with alignment check"), this is much more likely to occur on systems that boot via EFI but lack an implementation of the EFI RNG protocol, as in that case, the EFI stub will decide to leave the image where it found it, and the EFI firmware uses 64k alignment only. Fix this, by correctly identifying the case where the virtual displacement is a result of the physical displacement only. Signed-off-by: Ard Biesheuvel Reviewed-by: Mark Brown Acked-by: Mark Rutland Link: https://lore.kernel.org/r/20230223204101.1500373-1-ardb@kernel.org Signed-off-by: Catalin Marinas --- arch/arm64/include/asm/memory.h | 11 +++++++++++ arch/arm64/kernel/cpufeature.c | 2 +- arch/arm64/kernel/kaslr.c | 2 +- 3 files changed, 13 insertions(+), 2 deletions(-) diff --git a/arch/arm64/include/asm/memory.h b/arch/arm64/include/asm/memory.h index 9dd08cd339c3f..78e5163836a0a 100644 --- a/arch/arm64/include/asm/memory.h +++ b/arch/arm64/include/asm/memory.h @@ -180,6 +180,7 @@ #include #include #include +#include #include #if VA_BITS > 48 @@ -203,6 +204,16 @@ static inline unsigned long kaslr_offset(void) return kimage_vaddr - KIMAGE_VADDR; } +static inline bool kaslr_enabled(void) +{ + /* + * The KASLR offset modulo MIN_KIMG_ALIGN is taken from the physical + * placement of the image rather than from the seed, so a displacement + * of less than MIN_KIMG_ALIGN means that no seed was provided. + */ + return kaslr_offset() >= MIN_KIMG_ALIGN; +} + /* * Allow all memory at the discovery stage. We will clip it later. */ diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c index 45a42cf2191c3..5643a9ca502af 100644 --- a/arch/arm64/kernel/cpufeature.c +++ b/arch/arm64/kernel/cpufeature.c @@ -1633,7 +1633,7 @@ bool kaslr_requires_kpti(void) return false; } - return kaslr_offset() > 0; + return kaslr_enabled(); } static bool __meltdown_safe = true; diff --git a/arch/arm64/kernel/kaslr.c b/arch/arm64/kernel/kaslr.c index 325455d16dbcb..e7477f21a4c9d 100644 --- a/arch/arm64/kernel/kaslr.c +++ b/arch/arm64/kernel/kaslr.c @@ -41,7 +41,7 @@ static int __init kaslr_init(void) return 0; } - if (!kaslr_offset()) { + if (!kaslr_enabled()) { pr_warn("KASLR disabled due to lack of seed\n"); return 0; } From 8f9850dd8d23c1290cb642ce9548a440da5771ec Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 27 Feb 2023 13:03:22 +0300 Subject: [PATCH 568/765] net: phy: unlock on error in phy_probe() If genphy_c45_read_eee_adv() fails then we need to do a reset and unlock the &phydev->lock mutex before returning. Fixes: 3eeca4e199ce ("net: phy: do not force EEE support") Signed-off-by: Dan Carpenter Reviewed-by: Oleksij Rempel Link: https://lore.kernel.org/r/Y/x/6kHCjnQHqOpF@kili Signed-off-by: Paolo Abeni --- drivers/net/phy/phy_device.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 3f8a64fb9d712..9e9fd8ff00f69 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -3146,7 +3146,7 @@ static int phy_probe(struct device *dev) */ err = genphy_c45_read_eee_adv(phydev, phydev->advertising_eee); if (err) - return err; + goto out; /* There is no "enabled" flag. If PHY is advertising, assume it is * kind of enabled. From 434b26605f6cc500c1a995587e5c4bc4bc1693c6 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 24 Feb 2023 11:02:36 +0100 Subject: [PATCH 569/765] s390/rwonce: add READ_ONCE_ALIGNED_128() macro Add an s390 specific READ_ONCE_ALIGNED_128() helper, which can be used for fast block concurrent (atomic) 128-bit accesses. The used lpq instruction requires 128-bit alignment. This is also the reason why the compiler doesn't emit this instruction if __READ_ONCE() is used for 128-bit accesses. Link: https://lore.kernel.org/r/20230224100237.3247871-2-hca@linux.ibm.com Signed-off-by: Heiko Carstens --- arch/s390/include/asm/rwonce.h | 31 +++++++++++++++++++++++++++++++ 1 file changed, 31 insertions(+) create mode 100644 arch/s390/include/asm/rwonce.h diff --git a/arch/s390/include/asm/rwonce.h b/arch/s390/include/asm/rwonce.h new file mode 100644 index 0000000000000..91fc24520e828 --- /dev/null +++ b/arch/s390/include/asm/rwonce.h @@ -0,0 +1,31 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef __ASM_S390_RWONCE_H +#define __ASM_S390_RWONCE_H + +#include + +/* + * Use READ_ONCE_ALIGNED_128() for 128-bit block concurrent (atomic) read + * accesses. Note that x must be 128-bit aligned, otherwise a specification + * exception is generated. + */ +#define READ_ONCE_ALIGNED_128(x) \ +({ \ + union { \ + typeof(x) __x; \ + __uint128_t val; \ + } __u; \ + \ + BUILD_BUG_ON(sizeof(x) != 16); \ + asm volatile( \ + " lpq %[val],%[_x]\n" \ + : [val] "=d" (__u.val) \ + : [_x] "QS" (x) \ + : "memory"); \ + __u.__x; \ +}) + +#include + +#endif /* __ASM_S390_RWONCE_H */ From 5e02c74905cb00184b6c5ae70b1c1bfae5b3bd17 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 24 Feb 2023 11:02:37 +0100 Subject: [PATCH 570/765] s390/cpum_sf: use READ_ONCE_ALIGNED_128() instead of 128-bit cmpxchg Use READ_ONCE_ALIGNED_128() to read the previous value in front of a 128-bit cmpxchg loop, instead of (mis-)using a 128-bit cmpxchg operation to do the same. This makes the code more readable and is faster. Link: https://lore.kernel.org/r/20230224100237.3247871-3-hca@linux.ibm.com Signed-off-by: Heiko Carstens --- arch/s390/kernel/perf_cpum_sf.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c index 79904a839fb9f..e7b867e2f73f8 100644 --- a/arch/s390/kernel/perf_cpum_sf.c +++ b/arch/s390/kernel/perf_cpum_sf.c @@ -1355,8 +1355,7 @@ static void hw_perf_event_update(struct perf_event *event, int flush_all) num_sdb++; /* Reset trailer (using compare-double-and-swap) */ - /* READ_ONCE() 16 byte header */ - prev.val = __cdsg(&te->header.val, 0, 0); + prev.val = READ_ONCE_ALIGNED_128(te->header.val); do { old.val = prev.val; new.val = prev.val; @@ -1558,8 +1557,7 @@ static bool aux_set_alert(struct aux_buffer *aux, unsigned long alert_index, struct hws_trailer_entry *te; te = aux_sdb_trailer(aux, alert_index); - /* READ_ONCE() 16 byte header */ - prev.val = __cdsg(&te->header.val, 0, 0); + prev.val = READ_ONCE_ALIGNED_128(te->header.val); do { old.val = prev.val; new.val = prev.val; @@ -1637,8 +1635,7 @@ static bool aux_reset_buffer(struct aux_buffer *aux, unsigned long range, idx_old = idx = aux->empty_mark + 1; for (i = 0; i < range_scan; i++, idx++) { te = aux_sdb_trailer(aux, idx); - /* READ_ONCE() 16 byte header */ - prev.val = __cdsg(&te->header.val, 0, 0); + prev.val = READ_ONCE_ALIGNED_128(te->header.val); do { old.val = prev.val; new.val = prev.val; From e7ec1d2eac9cad57ff615ef6cc3e324ab7238b82 Mon Sep 17 00:00:00 2001 From: Alexander Gordeev Date: Sat, 17 Dec 2022 11:01:16 +0100 Subject: [PATCH 571/765] s390/mcck: cleanup user process termination path If a machine check interrupt hits while user process is running __s390_handle_mcck() helper function is called directly from the interrupt handler and terminates the current process by calling make_task_dead() routine. The make_task_dead() is not allowed to be called from interrupt context which forces the machine check handler switch to the kernel stack and enable local interrupts first. The __s390_handle_mcck() could also be called to service pending work, but this time from the external interrupts handler. It is the machine check handler that establishes the work and schedules the external interrupt, therefore the machine check interrupt itself should be disabled while reading out the corresponding variable: local_mcck_disable(); mcck = *this_cpu_ptr(&cpu_mcck); memset(this_cpu_ptr(&cpu_mcck), 0, sizeof(mcck)); local_mcck_enable(); However, local_mcck_disable() does not have effect when __s390_handle_mcck() is called directly form the machine check handler, since the machine check interrupt is still disabled. Therefore, it is not the opening bracket to the following local_mcck_enable() function. Simplify the user process termination flow by scheduling the external interrupt and killing the affected process from the interrupt context. Assume a kernel-generated signal is always delivered and ignore a value returned by do_send_sig_info() funciton. Reviewed-by: Heiko Carstens Reviewed-by: Sven Schnelle Signed-off-by: Alexander Gordeev Signed-off-by: Heiko Carstens --- arch/s390/include/asm/nmi.h | 5 ++--- arch/s390/kernel/entry.S | 10 ---------- arch/s390/kernel/nmi.c | 23 +++++------------------ arch/s390/kernel/smp.c | 2 +- 4 files changed, 8 insertions(+), 32 deletions(-) diff --git a/arch/s390/include/asm/nmi.h b/arch/s390/include/asm/nmi.h index af1cd3a6f4060..227466ce9e416 100644 --- a/arch/s390/include/asm/nmi.h +++ b/arch/s390/include/asm/nmi.h @@ -101,9 +101,8 @@ void nmi_alloc_mcesa_early(u64 *mcesad); int nmi_alloc_mcesa(u64 *mcesad); void nmi_free_mcesa(u64 *mcesad); -void s390_handle_mcck(struct pt_regs *regs); -void __s390_handle_mcck(void); -int s390_do_machine_check(struct pt_regs *regs); +void s390_handle_mcck(void); +void s390_do_machine_check(struct pt_regs *regs); #endif /* __ASSEMBLY__ */ #endif /* _ASM_S390_NMI_H */ diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index c8d8c99609367..76a06f3d36711 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -562,16 +562,6 @@ ENTRY(mcck_int_handler) xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) lgr %r2,%r11 # pass pointer to pt_regs brasl %r14,s390_do_machine_check - cghi %r2,0 - je .Lmcck_return - lg %r1,__LC_KERNEL_STACK # switch to kernel stack - mvc STACK_FRAME_OVERHEAD(__PT_SIZE,%r1),0(%r11) - xc __SF_BACKCHAIN(8,%r1),__SF_BACKCHAIN(%r1) - la %r11,STACK_FRAME_OVERHEAD(%r1) - lgr %r2,%r11 - lgr %r15,%r1 - brasl %r14,s390_handle_mcck -.Lmcck_return: lctlg %c1,%c1,__PT_CR1(%r11) lmg %r0,%r10,__PT_R0(%r11) mvc __LC_RETURN_MCCK_PSW(16),__PT_PSW(%r11) # move return PSW diff --git a/arch/s390/kernel/nmi.c b/arch/s390/kernel/nmi.c index 56d9c559afa1f..38ec0487521c4 100644 --- a/arch/s390/kernel/nmi.c +++ b/arch/s390/kernel/nmi.c @@ -156,7 +156,7 @@ NOKPROBE_SYMBOL(s390_handle_damage); * Main machine check handler function. Will be called with interrupts disabled * and machine checks enabled. */ -void __s390_handle_mcck(void) +void s390_handle_mcck(void) { struct mcck_struct mcck; @@ -192,23 +192,16 @@ void __s390_handle_mcck(void) if (mcck.stp_queue) stp_queue_work(); if (mcck.kill_task) { - local_irq_enable(); printk(KERN_EMERG "mcck: Terminating task because of machine " "malfunction (code 0x%016lx).\n", mcck.mcck_code); printk(KERN_EMERG "mcck: task: %s, pid: %d.\n", current->comm, current->pid); - make_task_dead(SIGSEGV); + if (is_global_init(current)) + panic("mcck: Attempting to kill init!\n"); + do_send_sig_info(SIGKILL, SEND_SIG_PRIV, current, PIDTYPE_PID); } } -void noinstr s390_handle_mcck(struct pt_regs *regs) -{ - trace_hardirqs_off(); - pai_kernel_enter(regs); - __s390_handle_mcck(); - pai_kernel_exit(regs); - trace_hardirqs_on(); -} /* * returns 0 if register contents could be validated * returns 1 otherwise @@ -373,7 +366,7 @@ NOKPROBE_SYMBOL(s390_backup_mcck_info); /* * machine check handler. */ -int notrace s390_do_machine_check(struct pt_regs *regs) +void notrace s390_do_machine_check(struct pt_regs *regs) { static int ipd_count; static DEFINE_SPINLOCK(ipd_lock); @@ -503,16 +496,10 @@ int notrace s390_do_machine_check(struct pt_regs *regs) } clear_cpu_flag(CIF_MCCK_GUEST); - if (user_mode(regs) && mcck_pending) { - irqentry_nmi_exit(regs, irq_state); - return 1; - } - if (mcck_pending) schedule_mcck_handler(); irqentry_nmi_exit(regs, irq_state); - return 0; } NOKPROBE_SYMBOL(s390_do_machine_check); diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 23c427284773c..97961522b317a 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -522,7 +522,7 @@ static void smp_handle_ext_call(void) if (test_bit(ec_call_function_single, &bits)) generic_smp_call_function_single_interrupt(); if (test_bit(ec_mcck_pending, &bits)) - __s390_handle_mcck(); + s390_handle_mcck(); if (test_bit(ec_irq_work, &bits)) irq_work_run(); } From e688c6255b742428ea8fa7e4fb8181a6135205e9 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 27 Feb 2023 18:56:10 +0100 Subject: [PATCH 572/765] s390/smp: perform cpu reset before delegating work to target cpu Clear CPU state (e.g. all TLB entries, prefetched instructions, etc.) of the target CPU, however without clearing register contents before starting any work on it. This puts the target CPU in a more defined state compared to the current Stop + Restart sigp orders. Signed-off-by: Heiko Carstens --- arch/s390/kernel/smp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 97961522b317a..d4888453bbf8b 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -333,6 +333,7 @@ static void pcpu_delegate(struct pcpu *pcpu, } /* Stop target cpu (if func returns this stops the current cpu). */ pcpu_sigp_retry(pcpu, SIGP_STOP, 0); + pcpu_sigp_retry(pcpu, SIGP_CPU_RESET, 0); /* Restart func on the target cpu and stop the current cpu. */ if (lc) { lc->restart_stack = stack; From 9b5c37bbf659fe4edb4804bc0aa99840d6798878 Mon Sep 17 00:00:00 2001 From: Vasily Gorbik Date: Wed, 16 Sep 2020 13:00:18 +0200 Subject: [PATCH 573/765] s390/decompressor: add link map saving Produce arch/s390/boot/vmlinux.map link map for the decompressor, when CONFIG_VMLINUX_MAP option is enabled. Link map is quite useful during making kernel changes related to how the decompressor is composed and debugging linker scripts. Acked-by: Heiko Carstens Signed-off-by: Vasily Gorbik Signed-off-by: Heiko Carstens --- arch/s390/boot/Makefile | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/s390/boot/Makefile b/arch/s390/boot/Makefile index 47a397da0498e..cebd4ca169164 100644 --- a/arch/s390/boot/Makefile +++ b/arch/s390/boot/Makefile @@ -52,6 +52,8 @@ targets += vmlinux.bin.zst info.bin syms.bin vmlinux.syms $(obj-all) OBJECTS := $(addprefix $(obj)/,$(obj-y)) OBJECTS_ALL := $(addprefix $(obj)/,$(obj-all)) +clean-files += vmlinux.map + quiet_cmd_section_cmp = SECTCMP $* define cmd_section_cmp s1=`$(OBJDUMP) -t -j "$*" "$<" | sort | \ @@ -71,7 +73,7 @@ $(obj)/bzImage: $(obj)/vmlinux $(obj)/section_cmp.boot.data $(obj)/section_cmp.b $(obj)/section_cmp%: vmlinux $(obj)/vmlinux FORCE $(call if_changed,section_cmp) -LDFLAGS_vmlinux := --oformat $(LD_BFD) -e startup --build-id=sha1 -T +LDFLAGS_vmlinux := --oformat $(LD_BFD) -e startup $(if $(CONFIG_VMLINUX_MAP),-Map=$(obj)/vmlinux.map) --build-id=sha1 -T $(obj)/vmlinux: $(obj)/vmlinux.lds $(OBJECTS_ALL) FORCE $(call if_changed,ld) From e33b93650fc5364f773985a3e961e24349330d97 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Tue, 28 Feb 2023 03:16:54 -0800 Subject: [PATCH 574/765] blk-iocost: Pass gendisk to ioc_refresh_params Current kernel (d2980d8d826554fa6981d621e569a453787472f8) crashes when blk_iocost_init for `nvme1` disk. BUG: kernel NULL pointer dereference, address: 0000000000000050 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page blk_iocost_init (include/asm-generic/qspinlock.h:128 include/linux/spinlock.h:203 include/linux/spinlock_api_smp.h:158 include/linux/spinlock.h:400 block/blk-iocost.c:2884) ioc_qos_write (block/blk-iocost.c:3198) ? kretprobe_perf_func (kernel/trace/trace_kprobe.c:1566) ? kernfs_fop_write_iter (include/linux/slab.h:584 fs/kernfs/file.c:311) ? __kmem_cache_alloc_node (mm/slab.h:? mm/slub.c:3452 mm/slub.c:3491) ? _copy_from_iter (arch/x86/include/asm/uaccess_64.h:46 arch/x86/include/asm/uaccess_64.h:52 lib/iov_iter.c:183 lib/iov_iter.c:628) ? kretprobe_dispatcher (kernel/trace/trace_kprobe.c:1693) cgroup_file_write (kernel/cgroup/cgroup.c:4061) kernfs_fop_write_iter (fs/kernfs/file.c:334) vfs_write (include/linux/fs.h:1849 fs/read_write.c:491 fs/read_write.c:584) ksys_write (fs/read_write.c:637) do_syscall_64 (arch/x86/entry/common.c:50 arch/x86/entry/common.c:80) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:120) This happens because ioc_refresh_params() is being called without a properly initialized ioc->rqos, which is happening later in the callee side. ioc_refresh_params() -> ioc_autop_idx() tries to access ioc->rqos.disk->queue but ioc->rqos.disk is NULL, causing the BUG above. Create function, called ioc_refresh_params_disk(), that is similar to ioc_refresh_params() but where the "struct gendisk" could be passed as an explicit argument. This function will be called when ioc->rqos.disk is not initialized. Fixes: ce57b558604e ("blk-rq-qos: make rq_qos_add and rq_qos_del more useful") Signed-off-by: Breno Leitao Acked-by: Tejun Heo Link: https://lore.kernel.org/r/20230228111654.1778120-1-leitao@debian.org Reviewed-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-iocost.c | 26 ++++++++++++++++++++------ 1 file changed, 20 insertions(+), 6 deletions(-) diff --git a/block/blk-iocost.c b/block/blk-iocost.c index ff534e9d92dca..4442c7a851125 100644 --- a/block/blk-iocost.c +++ b/block/blk-iocost.c @@ -800,7 +800,11 @@ static void ioc_refresh_period_us(struct ioc *ioc) ioc_refresh_margins(ioc); } -static int ioc_autop_idx(struct ioc *ioc) +/* + * ioc->rqos.disk isn't initialized when this function is called from + * the init path. + */ +static int ioc_autop_idx(struct ioc *ioc, struct gendisk *disk) { int idx = ioc->autop_idx; const struct ioc_params *p = &autop[idx]; @@ -808,11 +812,11 @@ static int ioc_autop_idx(struct ioc *ioc) u64 now_ns; /* rotational? */ - if (!blk_queue_nonrot(ioc->rqos.disk->queue)) + if (!blk_queue_nonrot(disk->queue)) return AUTOP_HDD; /* handle SATA SSDs w/ broken NCQ */ - if (blk_queue_depth(ioc->rqos.disk->queue) == 1) + if (blk_queue_depth(disk->queue) == 1) return AUTOP_SSD_QD1; /* use one of the normal ssd sets */ @@ -901,14 +905,19 @@ static void ioc_refresh_lcoefs(struct ioc *ioc) &c[LCOEF_WPAGE], &c[LCOEF_WSEQIO], &c[LCOEF_WRANDIO]); } -static bool ioc_refresh_params(struct ioc *ioc, bool force) +/* + * struct gendisk is required as an argument because ioc->rqos.disk + * is not properly initialized when called from the init path. + */ +static bool ioc_refresh_params_disk(struct ioc *ioc, bool force, + struct gendisk *disk) { const struct ioc_params *p; int idx; lockdep_assert_held(&ioc->lock); - idx = ioc_autop_idx(ioc); + idx = ioc_autop_idx(ioc, disk); p = &autop[idx]; if (idx == ioc->autop_idx && !force) @@ -939,6 +948,11 @@ static bool ioc_refresh_params(struct ioc *ioc, bool force) return true; } +static bool ioc_refresh_params(struct ioc *ioc, bool force) +{ + return ioc_refresh_params_disk(ioc, force, ioc->rqos.disk); +} + /* * When an iocg accumulates too much vtime or gets deactivated, we throw away * some vtime, which lowers the overall device utilization. As the exact amount @@ -2880,7 +2894,7 @@ static int blk_iocost_init(struct gendisk *disk) spin_lock_irq(&ioc->lock); ioc->autop_idx = AUTOP_INVALID; - ioc_refresh_params(ioc, true); + ioc_refresh_params_disk(ioc, true, disk); spin_unlock_irq(&ioc->lock); /* From 54aa7f2330b82884f4a1afce0220add6e8312f8b Mon Sep 17 00:00:00 2001 From: Joseph Qi Date: Tue, 28 Feb 2023 12:54:59 +0800 Subject: [PATCH 575/765] io_uring: fix fget leak when fs don't support nowait buffered read Heming reported a BUG when using io_uring doing link-cp on ocfs2. [1] Do the following steps can reproduce this BUG: mount -t ocfs2 /dev/vdc /mnt/ocfs2 cp testfile /mnt/ocfs2/ ./link-cp /mnt/ocfs2/testfile /mnt/ocfs2/testfile.1 umount /mnt/ocfs2 Then umount will fail, and it outputs: umount: /mnt/ocfs2: target is busy. While tracing umount, it blames mnt_get_count() not return as expected. Do a deep investigation for fget()/fput() on related code flow, I've finally found that fget() leaks since ocfs2 doesn't support nowait buffered read. io_issue_sqe |-io_assign_file // do fget() first |-io_read |-io_iter_do_read |-ocfs2_file_read_iter // return -EOPNOTSUPP |-kiocb_done |-io_rw_done |-__io_complete_rw_common // set REQ_F_REISSUE |-io_resubmit_prep |-io_req_prep_async // override req->file, leak happens This was introduced by commit a196c78b5443 in v5.18. Fix it by don't re-assign req->file if it has already been assigned. [1] https://lore.kernel.org/ocfs2-devel/ab580a75-91c8-d68a-3455-40361be1bfa8@linux.alibaba.com/T/#t Fixes: a196c78b5443 ("io_uring: assign non-fixed early for async work") Cc: Reported-by: Heming Zhao Signed-off-by: Joseph Qi Cc: Xiaoguang Wang Link: https://lore.kernel.org/r/20230228045459.13524-1-joseph.qi@linux.alibaba.com Signed-off-by: Jens Axboe --- io_uring/io_uring.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 64e07df034d12..7625597b5227d 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -1777,7 +1777,7 @@ int io_req_prep_async(struct io_kiocb *req) const struct io_issue_def *def = &io_issue_defs[req->opcode]; /* assign early for deferred execution for non-fixed file */ - if (def->needs_file && !(req->flags & REQ_F_FIXED_FILE)) + if (def->needs_file && !(req->flags & REQ_F_FIXED_FILE) && !req->file) req->file = io_file_get_normal(req, req->cqe.fd); if (!cdef->prep_async) return 0; From 0dd6fff2aad4e35633fef1ea72838bec5b47559a Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 21 Feb 2023 14:02:25 -0800 Subject: [PATCH 576/765] nvme: bring back auto-removal of deleted namespaces during sequential scan Bring back the check of the Identify Namespace return value for the legacy NVMe 1.0-style sequential scanning. While NVMe 1.0 does not support namespace management, there are "modern" cloud solutions like Google Cloud Platform that claim the obsolete 1.0 compliance for no good reason while supporting proprietary sideband namespace management. Fixes: 1a893c2bfef4 ("nvme: refactor namespace probing") Reported-by: Nils Hanke Signed-off-by: Christoph Hellwig Reviewed-by: Keith Busch Reviewed-by: Sagi Grimberg Tested-by: Nils Hanke --- drivers/nvme/host/core.c | 35 ++++++++++++++++++----------------- 1 file changed, 18 insertions(+), 17 deletions(-) diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index 384138f8f04ce..3345f866178e4 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -38,6 +38,7 @@ struct nvme_ns_info { bool is_shared; bool is_readonly; bool is_ready; + bool is_removed; }; unsigned int admin_timeout = 60; @@ -1402,16 +1403,8 @@ static int nvme_identify_ns(struct nvme_ctrl *ctrl, unsigned nsid, error = nvme_submit_sync_cmd(ctrl->admin_q, &c, *id, sizeof(**id)); if (error) { dev_warn(ctrl->device, "Identify namespace failed (%d)\n", error); - goto out_free_id; + kfree(*id); } - - error = NVME_SC_INVALID_NS | NVME_SC_DNR; - if ((*id)->ncap == 0) /* namespace not allocated or attached */ - goto out_free_id; - return 0; - -out_free_id: - kfree(*id); return error; } @@ -1425,6 +1418,13 @@ static int nvme_ns_info_from_identify(struct nvme_ctrl *ctrl, ret = nvme_identify_ns(ctrl, info->nsid, &id); if (ret) return ret; + + if (id->ncap == 0) { + /* namespace not allocated or attached */ + info->is_removed = true; + return -ENODEV; + } + info->anagrpid = id->anagrpid; info->is_shared = id->nmic & NVME_NS_NMIC_SHARED; info->is_readonly = id->nsattr & NVME_NS_ATTR_RO; @@ -4429,6 +4429,7 @@ static void nvme_scan_ns(struct nvme_ctrl *ctrl, unsigned nsid) { struct nvme_ns_info info = { .nsid = nsid }; struct nvme_ns *ns; + int ret; if (nvme_identify_ns_descs(ctrl, &info)) return; @@ -4445,19 +4446,19 @@ static void nvme_scan_ns(struct nvme_ctrl *ctrl, unsigned nsid) * set up a namespace. If not fall back to the legacy version. */ if ((ctrl->cap & NVME_CAP_CRMS_CRIMS) || - (info.ids.csi != NVME_CSI_NVM && info.ids.csi != NVME_CSI_ZNS)) { - if (nvme_ns_info_from_id_cs_indep(ctrl, &info)) - return; - } else { - if (nvme_ns_info_from_identify(ctrl, &info)) - return; - } + (info.ids.csi != NVME_CSI_NVM && info.ids.csi != NVME_CSI_ZNS)) + ret = nvme_ns_info_from_id_cs_indep(ctrl, &info); + else + ret = nvme_ns_info_from_identify(ctrl, &info); + + if (info.is_removed) + nvme_ns_remove_by_nsid(ctrl, nsid); /* * Ignore the namespace if it is not ready. We will get an AEN once it * becomes ready and restart the scan. */ - if (!info.is_ready) + if (ret || !info.is_ready) return; ns = nvme_find_get_ns(ctrl, nsid); From 51d24f701f453c18cb5f4596d8bbe8034e5d3fb4 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 16 Feb 2023 15:14:49 +0300 Subject: [PATCH 577/765] nvme-auth: fix an error code in nvme_auth_process_dhchap_challenge() This function was transitioned from returning NVMe status codes to returning traditional kernel error codes. However, this particular return now accidentally returns positive error codes like ENOMEM instead of negative -ENOMEM. Fixes: b0ef1b11d390 ("nvme-auth: don't use NVMe status codes") Signed-off-by: Dan Carpenter Signed-off-by: Christoph Hellwig --- drivers/nvme/host/auth.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/host/auth.c b/drivers/nvme/host/auth.c index 787537454f7fb..3e96714b8bc07 100644 --- a/drivers/nvme/host/auth.c +++ b/drivers/nvme/host/auth.c @@ -254,7 +254,7 @@ static int nvme_auth_process_dhchap_challenge(struct nvme_ctrl *ctrl, chap->qid, ret, gid_name); chap->status = NVME_AUTH_DHCHAP_FAILURE_DHGROUP_UNUSABLE; chap->dh_tfm = NULL; - return -ret; + return ret; } dev_dbg(ctrl->device, "qid %d: selected DH group %s\n", chap->qid, gid_name); From 76d54bf20cdcc1ed7569a89885e09636e9a8d71d Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Sun, 26 Feb 2023 21:42:54 +0900 Subject: [PATCH 578/765] nvme-tcp: don't access released socket during error recovery While the error recovery work is temporarily failing reconnect attempts, running the 'nvme list' command causes a kernel NULL pointer dereference by calling getsockname() with a released socket. During error recovery work, the nvme tcp socket is released and a new one created, so it is not safe to access the socket without proper check. Signed-off-by: Akinobu Mita Fixes: 02c57a82c008 ("nvme-tcp: print actual source IP address through sysfs "address" attr") Reviewed-by: Martin Belanger Reviewed-by: Hannes Reinecke Signed-off-by: Christoph Hellwig --- drivers/nvme/host/tcp.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c index d6100a787d39e..ef27122acce60 100644 --- a/drivers/nvme/host/tcp.c +++ b/drivers/nvme/host/tcp.c @@ -2489,6 +2489,10 @@ static int nvme_tcp_get_address(struct nvme_ctrl *ctrl, char *buf, int size) len = nvmf_get_address(ctrl, buf, size); + mutex_lock(&queue->queue_lock); + + if (!test_bit(NVME_TCP_Q_LIVE, &queue->flags)) + goto done; ret = kernel_getsockname(queue->sock, (struct sockaddr *)&src_addr); if (ret > 0) { if (len > 0) @@ -2496,6 +2500,8 @@ static int nvme_tcp_get_address(struct nvme_ctrl *ctrl, char *buf, int size) len += scnprintf(buf + len, size - len, "%ssrc_addr=%pISc\n", (len) ? "," : "", &src_addr); } +done: + mutex_unlock(&queue->queue_lock); return len; } From 26a57cb35548ae67c14871cccbf50da3edb01ea4 Mon Sep 17 00:00:00 2001 From: Daniel Wagner Date: Tue, 21 Feb 2023 17:51:06 +0100 Subject: [PATCH 579/765] nvme-fabrics: show well known discovery name The kernel always logs the unique subsystem name for a discovery controller, even in the case user space asked for the well known. This has lead to confusion as the logs of nvme-cli and the kernel logs didn't match. First, nvme-cli connects to the well known discovery controller to figure out if it supports TP8013. If so then nvme-cli disconnects and connects to the unique discovery controller. Currently, the kernel show that user space connected twice to the unique one. To avoid further confusion, show the well known discovery controller if user space asked for it: $ nvme connect-all -v -t tcp -a 192.168.0.1 nvme0: nqn.2014-08.org.nvmexpress.discovery connected nvme0: nqn.2014-08.org.nvmexpress.discovery disconnected nvme0: nqn.discovery connected kernel log: nvme nvme0: new ctrl: NQN "nqn.2014-08.org.nvmexpress.discovery", addr 192.168.0.1:8009 nvme nvme0: Removing ctrl: NQN "nqn.2014-08.org.nvmexpress.discovery" nvme nvme0: new ctrl: NQN "nqn.discovery", addr 192.168.0.1:8009 Fixes: e5ea42faa773 ("nvme: display correct subsystem NQN") Signed-off-by: Daniel Wagner Reviewed-by: Hannes Reinecke Signed-off-by: Christoph Hellwig --- drivers/nvme/host/fabrics.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/nvme/host/fabrics.h b/drivers/nvme/host/fabrics.h index a6e22116e1396..dcac3df8a5f76 100644 --- a/drivers/nvme/host/fabrics.h +++ b/drivers/nvme/host/fabrics.h @@ -189,7 +189,8 @@ nvmf_ctlr_matches_baseopts(struct nvme_ctrl *ctrl, static inline char *nvmf_ctrl_subsysnqn(struct nvme_ctrl *ctrl) { - if (!ctrl->subsys) + if (!ctrl->subsys || + !strcmp(ctrl->opts->subsysnqn, NVME_DISC_SUBSYS_NAME)) return ctrl->opts->subsysnqn; return ctrl->subsys->subnqn; } From 3425ddaea57af77ca96a59a5b8eaa2f9e1b021ba Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Sun, 26 Feb 2023 12:47:56 +0000 Subject: [PATCH 580/765] ASoC: mt6358: Fix event generation for wake on voice stage 2 switch ALSA control put() operations should return 0 if the value changed so that events can be generated appropriately for userspace but the custom control for wake on voice stage 2 doesn't do this, fix it. Reviewed-by: AngeloGioacchino Del Regno Signed-off-by: Mark Brown Link: https://lore.kernel.org/r/20230224-asoc-mt6358-quick-fixes-v1-1-747d9186be4b@kernel.org Signed-off-by: Mark Brown --- sound/soc/codecs/mt6358.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/sound/soc/codecs/mt6358.c b/sound/soc/codecs/mt6358.c index 93f35e8d26fce..9004377461f7a 100644 --- a/sound/soc/codecs/mt6358.c +++ b/sound/soc/codecs/mt6358.c @@ -567,6 +567,8 @@ static int mt6358_put_wov(struct snd_kcontrol *kcontrol, mt6358_disable_wov_phase2(priv); priv->wov_enabled = enabled; + + return 1; } return 0; From 8e847a43c28fca0aaa11fba8f91da7dfd9d6936f Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Sun, 26 Feb 2023 12:47:57 +0000 Subject: [PATCH 581/765] ASoC: mt6358: Validate Wake on Voice 2 writes Currently the Wake on Voice 2 control accepts and stores any value written but it reports that only 0 and 1 are valid values. Reject any out of range values written by userspace. Reviewed-by: AngeloGioacchino Del Regno Signed-off-by: Mark Brown Link: https://lore.kernel.org/r/20230224-asoc-mt6358-quick-fixes-v1-2-747d9186be4b@kernel.org Signed-off-by: Mark Brown --- sound/soc/codecs/mt6358.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/sound/soc/codecs/mt6358.c b/sound/soc/codecs/mt6358.c index 9004377461f7a..89d0dcb2635b5 100644 --- a/sound/soc/codecs/mt6358.c +++ b/sound/soc/codecs/mt6358.c @@ -560,6 +560,9 @@ static int mt6358_put_wov(struct snd_kcontrol *kcontrol, struct mt6358_priv *priv = snd_soc_component_get_drvdata(c); int enabled = ucontrol->value.integer.value[0]; + if (enabled < 0 || enabled > 1) + return -EINVAL; + if (priv->wov_enabled != enabled) { if (enabled) mt6358_enable_wov_phase2(priv); From 8cbd7273a724d4e9615b26d696bb1221a8a48e4c Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Sun, 26 Feb 2023 12:47:58 +0000 Subject: [PATCH 582/765] ASoC: mt6358: Remove undefined HPx Mux enumeration values The HPx Mux enumerations define values 5, 6 and 7 but describe them as "undefined" and map them to the value 0 on writing. Given the descriptions and behaviour it seems that these values are invalid and should not be present in the register, the current behaviour is detected as problematic by mixer-test: # # HPL Mux.0 expected 5 but read 0, is_volatile 0 # # HPL Mux.0 expected 6 but read 0, is_volatile 0 # # HPL Mux.0 expected 7 but read 0, is_volatile 0 Remove the values from the enumeration, this will prevent userspace setting them. Reviewed-by: AngeloGioacchino Del Regno Signed-off-by: Mark Brown Link: https://lore.kernel.org/r/20230224-asoc-mt6358-quick-fixes-v1-3-747d9186be4b@kernel.org Signed-off-by: Mark Brown --- sound/soc/codecs/mt6358.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/sound/soc/codecs/mt6358.c b/sound/soc/codecs/mt6358.c index 89d0dcb2635b5..b54610b279064 100644 --- a/sound/soc/codecs/mt6358.c +++ b/sound/soc/codecs/mt6358.c @@ -637,9 +637,6 @@ static const char * const hp_in_mux_map[] = { "Audio Playback", "Test Mode", "HP Impedance", - "undefined1", - "undefined2", - "undefined3", }; static int hp_in_mux_map_value[] = { @@ -648,9 +645,6 @@ static int hp_in_mux_map_value[] = { HP_MUX_HP, HP_MUX_TEST_MODE, HP_MUX_HP_IMPEDANCE, - HP_MUX_OPEN, - HP_MUX_OPEN, - HP_MUX_OPEN, }; static SOC_VALUE_ENUM_SINGLE_DECL(hpl_in_mux_map_enum, From d71ed1c8f0f458ae6852fdab055790fe1d9d19b6 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Sun, 26 Feb 2023 12:49:56 +0000 Subject: [PATCH 583/765] ASoC: mt8183: Remove spammy logging from I2S DAI driver There is a lot of dev_info() logging in normal operation in the I2S DAI driver, remove it to avoid spamming the console. Reviewed-by: AngeloGioacchino Del Regno Signed-off-by: Mark Brown Link: https://lore.kernel.org/r/20230224-asoc-mt8183-quick-fixes-v1-1-041f29419ed5@kernel.org Signed-off-by: Mark Brown --- sound/soc/mediatek/mt8183/mt8183-dai-i2s.c | 16 ---------------- 1 file changed, 16 deletions(-) diff --git a/sound/soc/mediatek/mt8183/mt8183-dai-i2s.c b/sound/soc/mediatek/mt8183/mt8183-dai-i2s.c index 6a9ace4180d34..38f7fa38ee95d 100644 --- a/sound/soc/mediatek/mt8183/mt8183-dai-i2s.c +++ b/sound/soc/mediatek/mt8183/mt8183-dai-i2s.c @@ -148,9 +148,6 @@ static int mt8183_i2s_hd_set(struct snd_kcontrol *kcontrol, hd_en = ucontrol->value.integer.value[0]; - dev_info(afe->dev, "%s(), kcontrol name %s, hd_en %d\n", - __func__, kcontrol->id.name, hd_en); - i2s_priv = get_i2s_priv_by_name(afe, kcontrol->id.name); if (!i2s_priv) { @@ -276,9 +273,6 @@ static int mtk_apll_event(struct snd_soc_dapm_widget *w, struct snd_soc_component *cmpnt = snd_soc_dapm_to_component(w->dapm); struct mtk_base_afe *afe = snd_soc_component_get_drvdata(cmpnt); - dev_info(cmpnt->dev, "%s(), name %s, event 0x%x\n", - __func__, w->name, event); - switch (event) { case SND_SOC_DAPM_PRE_PMU: if (strcmp(w->name, APLL1_W_NAME) == 0) @@ -307,9 +301,6 @@ static int mtk_mclk_en_event(struct snd_soc_dapm_widget *w, struct mtk_base_afe *afe = snd_soc_component_get_drvdata(cmpnt); struct mtk_afe_i2s_priv *i2s_priv; - dev_info(cmpnt->dev, "%s(), name %s, event 0x%x\n", - __func__, w->name, event); - i2s_priv = get_i2s_priv_by_name(afe, w->name); if (!i2s_priv) { @@ -715,11 +706,6 @@ static int mtk_dai_i2s_config(struct mtk_base_afe *afe, unsigned int i2s_con = 0, fmt_con = I2S_FMT_I2S << I2S_FMT_SFT; int ret = 0; - dev_info(afe->dev, "%s(), id %d, rate %d, format %d\n", - __func__, - i2s_id, - rate, format); - if (i2s_priv) { i2s_priv->rate = rate; @@ -810,8 +796,6 @@ static int mtk_dai_i2s_set_sysclk(struct snd_soc_dai *dai, return -EINVAL; } - dev_info(afe->dev, "%s(), freq %d\n", __func__, freq); - apll = mt8183_get_apll_by_rate(afe, freq); apll_rate = mt8183_get_apll_rate(afe, apll); From 18f51ed09888c8e48bd377d1715d4ff807b4c805 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Sun, 26 Feb 2023 12:49:57 +0000 Subject: [PATCH 584/765] ASoC: mt8183: Fix event generation for I2S DAI operations ALSA control put() operations should return 0 if the value changed so that events can be generated appropriately for userspace but the custom control in the MT8183 I2S DAI driver doesn't do that, fix it. Reviewed-by: AngeloGioacchino Del Regno Signed-off-by: Mark Brown Link: https://lore.kernel.org/r/20230224-asoc-mt8183-quick-fixes-v1-2-041f29419ed5@kernel.org Signed-off-by: Mark Brown --- sound/soc/mediatek/mt8183/mt8183-dai-i2s.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/sound/soc/mediatek/mt8183/mt8183-dai-i2s.c b/sound/soc/mediatek/mt8183/mt8183-dai-i2s.c index 38f7fa38ee95d..8645ab686970f 100644 --- a/sound/soc/mediatek/mt8183/mt8183-dai-i2s.c +++ b/sound/soc/mediatek/mt8183/mt8183-dai-i2s.c @@ -141,7 +141,7 @@ static int mt8183_i2s_hd_set(struct snd_kcontrol *kcontrol, struct mtk_base_afe *afe = snd_soc_component_get_drvdata(cmpnt); struct mtk_afe_i2s_priv *i2s_priv; struct soc_enum *e = (struct soc_enum *)kcontrol->private_value; - int hd_en; + int hd_en, change; if (ucontrol->value.enumerated.item[0] >= e->items) return -EINVAL; @@ -155,9 +155,10 @@ static int mt8183_i2s_hd_set(struct snd_kcontrol *kcontrol, return -EINVAL; } + change = i2s_priv->low_jitter_en != hd_en; i2s_priv->low_jitter_en = hd_en; - return 0; + return change; } static const struct snd_kcontrol_new mtk_dai_i2s_controls[] = { From 54fc4b72b630e1cb92a21140084c6852babbb234 Mon Sep 17 00:00:00 2001 From: Claudiu Beznea Date: Tue, 28 Feb 2023 13:01:43 +0200 Subject: [PATCH 585/765] ASoC: soc-pcm: add option to start DMA after DAI Add option to start DMA component after DAI trigger. This is done by filling the new struct snd_soc_component_driver::start_dma_last. Signed-off-by: Claudiu Beznea Link: https://lore.kernel.org/r/20230228110145.3770525-2-claudiu.beznea@microchip.com Signed-off-by: Mark Brown --- include/sound/soc-component.h | 2 ++ sound/soc/soc-pcm.c | 27 ++++++++++++++++++++++----- 2 files changed, 24 insertions(+), 5 deletions(-) diff --git a/include/sound/soc-component.h b/include/sound/soc-component.h index 3203d35bc8c15..0814ed1438640 100644 --- a/include/sound/soc-component.h +++ b/include/sound/soc-component.h @@ -190,6 +190,8 @@ struct snd_soc_component_driver { bool use_dai_pcm_id; /* use DAI link PCM ID as PCM device number */ int be_pcm_base; /* base device ID for all BE PCMs */ + unsigned int start_dma_last; + #ifdef CONFIG_DEBUG_FS const char *debugfs_prefix; #endif diff --git a/sound/soc/soc-pcm.c b/sound/soc/soc-pcm.c index 005b179a770a0..5eb056b942ce8 100644 --- a/sound/soc/soc-pcm.c +++ b/sound/soc/soc-pcm.c @@ -1088,22 +1088,39 @@ static int soc_pcm_hw_params(struct snd_pcm_substream *substream, static int soc_pcm_trigger(struct snd_pcm_substream *substream, int cmd) { struct snd_soc_pcm_runtime *rtd = asoc_substream_to_rtd(substream); - int ret = -EINVAL, _ret = 0; + struct snd_soc_component *component; + int ret = -EINVAL, _ret = 0, start_dma_last = 0, i; int rollback = 0; switch (cmd) { case SNDRV_PCM_TRIGGER_START: case SNDRV_PCM_TRIGGER_RESUME: case SNDRV_PCM_TRIGGER_PAUSE_RELEASE: + /* Do we need to start dma last? */ + for_each_rtd_components(rtd, i, component) { + if (component->driver->start_dma_last) { + start_dma_last = 1; + break; + } + } + ret = snd_soc_link_trigger(substream, cmd, 0); if (ret < 0) goto start_err; - ret = snd_soc_pcm_component_trigger(substream, cmd, 0); - if (ret < 0) - goto start_err; + if (start_dma_last) { + ret = snd_soc_pcm_dai_trigger(substream, cmd, 0); + if (ret < 0) + goto start_err; + + ret = snd_soc_pcm_component_trigger(substream, cmd, 0); + } else { + ret = snd_soc_pcm_component_trigger(substream, cmd, 0); + if (ret < 0) + goto start_err; - ret = snd_soc_pcm_dai_trigger(substream, cmd, 0); + ret = snd_soc_pcm_dai_trigger(substream, cmd, 0); + } start_err: if (ret < 0) rollback = 1; From 143a2f011c4471511887807822d3fd71f25f5169 Mon Sep 17 00:00:00 2001 From: Claudiu Beznea Date: Tue, 28 Feb 2023 13:01:44 +0200 Subject: [PATCH 586/765] ASoC: dt-bindings: sama7g5-pdmc: add microchip,startup-delay-us binding PDMC can work with different types of microphones, thus different boards could have different microphones. Depending on microphone type the PDMC would need to wait longer or shorter period (at startup) than the default chosen one to filter unwanted noise. Thus add microchip,startup-delay-us binding to let PDMC users to specify startup delay. Signed-off-by: Claudiu Beznea Reviewed-by: Krzysztof Kozlowski Link: https://lore.kernel.org/r/20230228110145.3770525-3-claudiu.beznea@microchip.com Signed-off-by: Mark Brown --- .../devicetree/bindings/sound/microchip,sama7g5-pdmc.yaml | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/Documentation/devicetree/bindings/sound/microchip,sama7g5-pdmc.yaml b/Documentation/devicetree/bindings/sound/microchip,sama7g5-pdmc.yaml index c4cf1e5ab84b0..9b40268537cb2 100644 --- a/Documentation/devicetree/bindings/sound/microchip,sama7g5-pdmc.yaml +++ b/Documentation/devicetree/bindings/sound/microchip,sama7g5-pdmc.yaml @@ -67,6 +67,12 @@ properties: maxItems: 4 uniqueItems: true + microchip,startup-delay-us: + description: | + Specifies the delay in microseconds that needs to be applied after + enabling the PDMC microphones to avoid unwanted noise due to microphones + not being ready. + required: - compatible - reg From c5682e2ba1327d08987a7cabc7b5b40bf3bc131f Mon Sep 17 00:00:00 2001 From: Claudiu Beznea Date: Tue, 28 Feb 2023 13:01:45 +0200 Subject: [PATCH 587/765] ASoC: mchp-pdmc: fix poc noise at capture startup Microchip PDMC IP doesn't filter microphone noises on startup. By default, it captures data received from digital microphones after the MCHP_PDMC_MR.EN bits are set. Thus when enable is set on PDMC side the digital microphones might not be ready yet and PDMC captures data from then in this time. This data captured is poc noise. To avoid this the software workaround is to the following: 1/ enable PDMC channel 2/ wait 150ms (on SAMA7G5-EK setup) 3/ execute 16 dummy reads from RHR 4/ clear interrupts 5/ enable interrupts 6/ enable DMA channel Fixes: 50291652af52 ("ASoC: atmel: mchp-pdmc: add PDMC driver") Signed-off-by: Claudiu Beznea Link: https://lore.kernel.org/r/20230228110145.3770525-4-claudiu.beznea@microchip.com Signed-off-by: Mark Brown --- sound/soc/atmel/mchp-pdmc.c | 53 +++++++++++++++++++++++++++++++++---- 1 file changed, 48 insertions(+), 5 deletions(-) diff --git a/sound/soc/atmel/mchp-pdmc.c b/sound/soc/atmel/mchp-pdmc.c index cf4084dcbd5ee..1aed3baa93697 100644 --- a/sound/soc/atmel/mchp-pdmc.c +++ b/sound/soc/atmel/mchp-pdmc.c @@ -114,6 +114,7 @@ struct mchp_pdmc { struct clk *gclk; u32 pdmcen; u32 suspend_irq; + u32 startup_delay_us; int mic_no; int sinc_order; bool audio_filter_en; @@ -425,6 +426,7 @@ static const struct snd_soc_component_driver mchp_pdmc_dai_component = { .open = &mchp_pdmc_open, .close = &mchp_pdmc_close, .legacy_dai_naming = 1, + .start_dma_last = 1, }; static const unsigned int mchp_pdmc_1mic[] = {1}; @@ -632,6 +634,29 @@ static int mchp_pdmc_hw_params(struct snd_pcm_substream *substream, return 0; } +static void mchp_pdmc_noise_filter_workaround(struct mchp_pdmc *dd) +{ + u32 tmp, steps = 16; + + /* + * PDMC doesn't wait for microphones' startup time thus the acquisition + * may start before the microphones are ready leading to poc noises at + * the beginning of capture. To avoid this, we need to wait 50ms (in + * normal startup procedure) or 150 ms (worst case after resume from sleep + * states) after microphones are enabled and then clear the FIFOs (by + * reading the RHR 16 times) and possible interrupts before continuing. + * Also, for this to work the DMA needs to be started after interrupts + * are enabled. + */ + usleep_range(dd->startup_delay_us, dd->startup_delay_us + 5); + + while (steps--) + regmap_read(dd->regmap, MCHP_PDMC_RHR, &tmp); + + /* Clear interrupts. */ + regmap_read(dd->regmap, MCHP_PDMC_ISR, &tmp); +} + static int mchp_pdmc_trigger(struct snd_pcm_substream *substream, int cmd, struct snd_soc_dai *dai) { @@ -644,15 +669,17 @@ static int mchp_pdmc_trigger(struct snd_pcm_substream *substream, switch (cmd) { case SNDRV_PCM_TRIGGER_RESUME: case SNDRV_PCM_TRIGGER_START: - /* Enable overrun and underrun error interrupts */ - regmap_write(dd->regmap, MCHP_PDMC_IER, dd->suspend_irq | - MCHP_PDMC_IR_RXOVR | MCHP_PDMC_IR_RXUDR); - dd->suspend_irq = 0; - fallthrough; case SNDRV_PCM_TRIGGER_PAUSE_RELEASE: snd_soc_component_update_bits(cpu, MCHP_PDMC_MR, MCHP_PDMC_MR_PDMCEN_MASK, dd->pdmcen); + + mchp_pdmc_noise_filter_workaround(dd); + + /* Enable interrupts. */ + regmap_write(dd->regmap, MCHP_PDMC_IER, dd->suspend_irq | + MCHP_PDMC_IR_RXOVR | MCHP_PDMC_IR_RXUDR); + dd->suspend_irq = 0; break; case SNDRV_PCM_TRIGGER_SUSPEND: regmap_read(dd->regmap, MCHP_PDMC_IMR, &dd->suspend_irq); @@ -796,6 +823,7 @@ static bool mchp_pdmc_readable_reg(struct device *dev, unsigned int reg) case MCHP_PDMC_CFGR: case MCHP_PDMC_IMR: case MCHP_PDMC_ISR: + case MCHP_PDMC_RHR: case MCHP_PDMC_VER: return true; default: @@ -817,6 +845,17 @@ static bool mchp_pdmc_writeable_reg(struct device *dev, unsigned int reg) } } +static bool mchp_pdmc_volatile_reg(struct device *dev, unsigned int reg) +{ + switch (reg) { + case MCHP_PDMC_ISR: + case MCHP_PDMC_RHR: + return true; + default: + return false; + } +} + static bool mchp_pdmc_precious_reg(struct device *dev, unsigned int reg) { switch (reg) { @@ -836,6 +875,7 @@ static const struct regmap_config mchp_pdmc_regmap_config = { .readable_reg = mchp_pdmc_readable_reg, .writeable_reg = mchp_pdmc_writeable_reg, .precious_reg = mchp_pdmc_precious_reg, + .volatile_reg = mchp_pdmc_volatile_reg, .cache_type = REGCACHE_FLAT, }; @@ -918,6 +958,9 @@ static int mchp_pdmc_dt_init(struct mchp_pdmc *dd) dd->channel_mic_map[i].clk_edge = edge; } + dd->startup_delay_us = 150000; + of_property_read_u32(np, "microchip,startup-delay-us", &dd->startup_delay_us); + return 0; } From 5df1a5d28449f2b98ff84f69dea74b06f9b8e362 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Fri, 24 Feb 2023 14:03:55 +0000 Subject: [PATCH 588/765] ASoC: mt8192: Remove spammy log messages MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There are a lot of info level log messages in the mt8192 ADDA driver which are trivially triggerable from userspace, many in normal operation. Remove these to avoid spamming the console. Reviewed-by: Nícolas F. R. A. Prado Tested-by: Nícolas F. R. A. Prado Signed-off-by: Mark Brown Link: https://lore.kernel.org/r/20230223-asoc-mt8192-quick-fixes-v1-1-9a85f90368e1@kernel.org Signed-off-by: Mark Brown --- sound/soc/mediatek/mt8192/mt8192-dai-adda.c | 33 --------------------- 1 file changed, 33 deletions(-) diff --git a/sound/soc/mediatek/mt8192/mt8192-dai-adda.c b/sound/soc/mediatek/mt8192/mt8192-dai-adda.c index f8c73e8624df2..bc8753f1001cd 100644 --- a/sound/soc/mediatek/mt8192/mt8192-dai-adda.c +++ b/sound/soc/mediatek/mt8192/mt8192-dai-adda.c @@ -303,9 +303,6 @@ static int mtk_adda_ul_event(struct snd_soc_dapm_widget *w, struct mt8192_afe_private *afe_priv = afe->platform_priv; int mtkaif_dmic = afe_priv->mtkaif_dmic; - dev_info(afe->dev, "%s(), name %s, event 0x%x, mtkaif_dmic %d\n", - __func__, w->name, event, mtkaif_dmic); - switch (event) { case SND_SOC_DAPM_PRE_PMU: mt8192_afe_gpio_request(afe->dev, true, MT8192_DAI_ADDA, 1); @@ -345,10 +342,6 @@ static int mtk_adda_ch34_ul_event(struct snd_soc_dapm_widget *w, int mtkaif_dmic = afe_priv->mtkaif_dmic_ch34; int mtkaif_adda6_only = afe_priv->mtkaif_adda6_only; - dev_info(afe->dev, - "%s(), name %s, event 0x%x, mtkaif_dmic %d, mtkaif_adda6_only %d\n", - __func__, w->name, event, mtkaif_dmic, mtkaif_adda6_only); - switch (event) { case SND_SOC_DAPM_PRE_PMU: mt8192_afe_gpio_request(afe->dev, true, MT8192_DAI_ADDA_CH34, @@ -538,9 +531,6 @@ static int mtk_adda_dl_event(struct snd_soc_dapm_widget *w, struct snd_soc_component *cmpnt = snd_soc_dapm_to_component(w->dapm); struct mtk_base_afe *afe = snd_soc_component_get_drvdata(cmpnt); - dev_info(afe->dev, "%s(), name %s, event 0x%x\n", - __func__, w->name, event); - switch (event) { case SND_SOC_DAPM_PRE_PMU: mt8192_afe_gpio_request(afe->dev, true, MT8192_DAI_ADDA, 0); @@ -564,9 +554,6 @@ static int mtk_adda_ch34_dl_event(struct snd_soc_dapm_widget *w, struct snd_soc_component *cmpnt = snd_soc_dapm_to_component(w->dapm); struct mtk_base_afe *afe = snd_soc_component_get_drvdata(cmpnt); - dev_info(afe->dev, "%s(), name %s, event 0x%x\n", - __func__, w->name, event); - switch (event) { case SND_SOC_DAPM_PRE_PMU: mt8192_afe_gpio_request(afe->dev, true, MT8192_DAI_ADDA_CH34, @@ -612,9 +599,6 @@ static int stf_positive_gain_set(struct snd_kcontrol *kcontrol, AFE_SIDETONE_GAIN, POSITIVE_GAIN_MASK_SFT, (gain_db / 6) << POSITIVE_GAIN_SFT); - } else { - dev_warn(afe->dev, "%s(), gain_db %d invalid\n", - __func__, gain_db); } return 0; } @@ -640,9 +624,6 @@ static int mt8192_adda_dmic_set(struct snd_kcontrol *kcontrol, dmic_on = ucontrol->value.integer.value[0]; - dev_info(afe->dev, "%s(), kcontrol name %s, dmic_on %d\n", - __func__, kcontrol->id.name, dmic_on); - afe_priv->mtkaif_dmic = dmic_on; afe_priv->mtkaif_dmic_ch34 = dmic_on; return 0; @@ -669,9 +650,6 @@ static int mt8192_adda6_only_set(struct snd_kcontrol *kcontrol, mtkaif_adda6_only = ucontrol->value.integer.value[0]; - dev_info(afe->dev, "%s(), kcontrol name %s, mtkaif_adda6_only %d\n", - __func__, kcontrol->id.name, mtkaif_adda6_only); - afe_priv->mtkaif_adda6_only = mtkaif_adda6_only; return 0; } @@ -750,9 +728,6 @@ static int mtk_stf_event(struct snd_soc_dapm_widget *w, regmap_read(afe->regmap, AFE_SIDETONE_CON1, ®_value); - dev_info(afe->dev, "%s(), name %s, event 0x%x, ul_rate 0x%x, AFE_SIDETONE_CON1 0x%x\n", - __func__, w->name, event, ul_rate, reg_value); - switch (event) { case SND_SOC_DAPM_PRE_PMU: /* set side tone gain = 0 */ @@ -1163,12 +1138,6 @@ static int mtk_dai_adda_hw_params(struct snd_pcm_substream *substream, unsigned int rate = params_rate(params); int id = dai->id; - dev_info(afe->dev, "%s(), id %d, stream %d, rate %d\n", - __func__, - id, - substream->stream, - rate); - if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) { unsigned int dl_src2_con0 = 0; unsigned int dl_src2_con1 = 0; @@ -1441,8 +1410,6 @@ int mt8192_dai_adda_register(struct mtk_base_afe *afe) struct mtk_base_afe_dai *dai; struct mt8192_afe_private *afe_priv = afe->platform_priv; - dev_info(afe->dev, "%s()\n", __func__); - dai = devm_kzalloc(afe->dev, sizeof(*dai), GFP_KERNEL); if (!dai) return -ENOMEM; From b373076f609993d333dbbc3283b65320c7a41834 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Fri, 24 Feb 2023 14:03:56 +0000 Subject: [PATCH 589/765] ASoC: mt8192: Fix event generation for controls MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit ALSA controls put() operations should return 1 if the value changed and 0 if it remains the same, fix the mt8192 driver to do so. Reviewed-by: Nícolas F. R. A. Prado Tested-by: Nícolas F. R. A. Prado Signed-off-by: Mark Brown Link: https://lore.kernel.org/r/20230223-asoc-mt8192-quick-fixes-v1-2-9a85f90368e1@kernel.org Signed-off-by: Mark Brown --- sound/soc/mediatek/mt8192/mt8192-dai-adda.c | 25 +++++++++++++++------ 1 file changed, 18 insertions(+), 7 deletions(-) diff --git a/sound/soc/mediatek/mt8192/mt8192-dai-adda.c b/sound/soc/mediatek/mt8192/mt8192-dai-adda.c index bc8753f1001cd..a33d1ce333498 100644 --- a/sound/soc/mediatek/mt8192/mt8192-dai-adda.c +++ b/sound/soc/mediatek/mt8192/mt8192-dai-adda.c @@ -591,16 +591,19 @@ static int stf_positive_gain_set(struct snd_kcontrol *kcontrol, struct mtk_base_afe *afe = snd_soc_component_get_drvdata(cmpnt); struct mt8192_afe_private *afe_priv = afe->platform_priv; int gain_db = ucontrol->value.integer.value[0]; + bool change = false; afe_priv->stf_positive_gain_db = gain_db; if (gain_db >= 0 && gain_db <= 24) { - regmap_update_bits(afe->regmap, - AFE_SIDETONE_GAIN, - POSITIVE_GAIN_MASK_SFT, - (gain_db / 6) << POSITIVE_GAIN_SFT); + regmap_update_bits_check(afe->regmap, + AFE_SIDETONE_GAIN, + POSITIVE_GAIN_MASK_SFT, + (gain_db / 6) << POSITIVE_GAIN_SFT, + &change); } - return 0; + + return change; } static int mt8192_adda_dmic_get(struct snd_kcontrol *kcontrol, @@ -621,12 +624,17 @@ static int mt8192_adda_dmic_set(struct snd_kcontrol *kcontrol, struct mtk_base_afe *afe = snd_soc_component_get_drvdata(cmpnt); struct mt8192_afe_private *afe_priv = afe->platform_priv; int dmic_on; + bool change; dmic_on = ucontrol->value.integer.value[0]; + change = (afe_priv->mtkaif_dmic != dmic_on) || + (afe_priv->mtkaif_dmic_ch34 != dmic_on); + afe_priv->mtkaif_dmic = dmic_on; afe_priv->mtkaif_dmic_ch34 = dmic_on; - return 0; + + return change; } static int mt8192_adda6_only_get(struct snd_kcontrol *kcontrol, @@ -647,11 +655,14 @@ static int mt8192_adda6_only_set(struct snd_kcontrol *kcontrol, struct mtk_base_afe *afe = snd_soc_component_get_drvdata(cmpnt); struct mt8192_afe_private *afe_priv = afe->platform_priv; int mtkaif_adda6_only; + bool change; mtkaif_adda6_only = ucontrol->value.integer.value[0]; + change = afe_priv->mtkaif_adda6_only != mtkaif_adda6_only; afe_priv->mtkaif_adda6_only = mtkaif_adda6_only; - return 0; + + return change; } static const struct snd_kcontrol_new mtk_adda_controls[] = { From 05437a91173b8780692ac35313f98cac68be7c42 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Fri, 24 Feb 2023 14:03:57 +0000 Subject: [PATCH 590/765] ASoC: mt8192: Report an error if when an invalid sidetone gain is written MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Reporting an error on invalid values is optional but helpful to userspace so do so. Reviewed-by: Nícolas F. R. A. Prado Tested-by: Nícolas F. R. A. Prado Signed-off-by: Mark Brown Link: https://lore.kernel.org/r/20230223-asoc-mt8192-quick-fixes-v1-3-9a85f90368e1@kernel.org Signed-off-by: Mark Brown --- sound/soc/mediatek/mt8192/mt8192-dai-adda.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/sound/soc/mediatek/mt8192/mt8192-dai-adda.c b/sound/soc/mediatek/mt8192/mt8192-dai-adda.c index a33d1ce333498..a02a297c04507 100644 --- a/sound/soc/mediatek/mt8192/mt8192-dai-adda.c +++ b/sound/soc/mediatek/mt8192/mt8192-dai-adda.c @@ -601,6 +601,8 @@ static int stf_positive_gain_set(struct snd_kcontrol *kcontrol, POSITIVE_GAIN_MASK_SFT, (gain_db / 6) << POSITIVE_GAIN_SFT, &change); + } else { + return -EINVAL; } return change; From ce40d93b062c0bdcd29218c12ab1dba544382dd8 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Fri, 24 Feb 2023 14:03:58 +0000 Subject: [PATCH 591/765] ASoC: mt8192: Fix range for sidetone positive gain MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The Sidetone_Positive_Gain_dB control reports a range of 0..100 as valid but the put() function rejects anything larger than 24. Fix this. There are numerous other problems with this control, the name is very non idiomatic and it should be a TLV, but it's ABI so probably we should leave those alone. Reviewed-by: Nícolas F. R. A. Prado Tested-by: Nícolas F. R. A. Prado Signed-off-by: Mark Brown Link: https://lore.kernel.org/r/20230223-asoc-mt8192-quick-fixes-v1-4-9a85f90368e1@kernel.org Signed-off-by: Mark Brown --- sound/soc/mediatek/mt8192/mt8192-dai-adda.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/mediatek/mt8192/mt8192-dai-adda.c b/sound/soc/mediatek/mt8192/mt8192-dai-adda.c index a02a297c04507..4919535e2759d 100644 --- a/sound/soc/mediatek/mt8192/mt8192-dai-adda.c +++ b/sound/soc/mediatek/mt8192/mt8192-dai-adda.c @@ -670,7 +670,7 @@ static int mt8192_adda6_only_set(struct snd_kcontrol *kcontrol, static const struct snd_kcontrol_new mtk_adda_controls[] = { SOC_SINGLE("Sidetone_Gain", AFE_SIDETONE_GAIN, SIDE_TONE_GAIN_SFT, SIDE_TONE_GAIN_MASK, 0), - SOC_SINGLE_EXT("Sidetone_Positive_Gain_dB", SND_SOC_NOPM, 0, 100, 0, + SOC_SINGLE_EXT("Sidetone_Positive_Gain_dB", SND_SOC_NOPM, 0, 24, 0, stf_positive_gain_get, stf_positive_gain_set), SOC_SINGLE("ADDA_DL_GAIN", AFE_ADDA_DL_SRC2_CON1, DL_2_GAIN_CTL_PRE_SFT, DL_2_GAIN_CTL_PRE_MASK, 0), From c769fb6bcc485d752d492064a9005525a8d5fa24 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Sat, 25 Feb 2023 21:48:13 +0000 Subject: [PATCH 592/765] ASoC: sam9g20ek: Disable capture unless building with microphone input Without modification the AT91SAM9G20-EK has no capture support, none of the inputs of the CODEC are wired to anything to useful and there are no paths supporting loopback. Since the audio is clocked from the CODEC and the DAPM inputs are marked as unusable this means that capture will fail to transfer any data as the ADC path can't be powered up. Flag this in the device description so apps don't see unusable capture support, guarded with the existing optional define for mic input. Reviewed-by: Claudiu Beznea Signed-off-by: Mark Brown Link: https://lore.kernel.org/r/20230225-asoc-sam9g20ek-v1-1-9baeb4893142@kernel.org Signed-off-by: Mark Brown --- sound/soc/atmel/sam9g20_wm8731.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/sound/soc/atmel/sam9g20_wm8731.c b/sound/soc/atmel/sam9g20_wm8731.c index 1430642c8433a..785b9d01d8afe 100644 --- a/sound/soc/atmel/sam9g20_wm8731.c +++ b/sound/soc/atmel/sam9g20_wm8731.c @@ -98,6 +98,9 @@ static struct snd_soc_dai_link at91sam9g20ek_dai = { .init = at91sam9g20ek_wm8731_init, .dai_fmt = SND_SOC_DAIFMT_I2S | SND_SOC_DAIFMT_NB_NF | SND_SOC_DAIFMT_CBP_CFP, +#ifndef ENABLE_MIC_INPUT + .playback_only = true, +#endif SND_SOC_DAILINK_REG(pcm), }; From 0de2cc3707b6b6e2ad40bd24ce09a5c1f65d01e1 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 27 Feb 2023 09:58:26 +0100 Subject: [PATCH 593/765] ASoC: zl38060 add gpiolib dependency Without gpiolib, this driver fails to link: arm-linux-gnueabi-ld: sound/soc/codecs/zl38060.o: in function `chip_gpio_get': zl38060.c:(.text+0x30): undefined reference to `gpiochip_get_data' arm-linux-gnueabi-ld: sound/soc/codecs/zl38060.o: in function `zl38_spi_probe': zl38060.c:(.text+0xa18): undefined reference to `devm_gpiochip_add_data_with_key' This appears to have been in the driver since the start, but is hard to hit in randconfig testing since gpiolib is almost always selected by something else. Fixes: 52e8a94baf90 ("ASoC: Add initial ZL38060 driver") Signed-off-by: Arnd Bergmann Link: https://lore.kernel.org/r/20230227085850.2503725-1-arnd@kernel.org Signed-off-by: Mark Brown --- sound/soc/codecs/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/soc/codecs/Kconfig b/sound/soc/codecs/Kconfig index bd72c426a93d1..07747565c3b51 100644 --- a/sound/soc/codecs/Kconfig +++ b/sound/soc/codecs/Kconfig @@ -2103,6 +2103,7 @@ config SND_SOC_WSA883X config SND_SOC_ZL38060 tristate "Microsemi ZL38060 Connected Home Audio Processor" depends on SPI_MASTER + depends on GPIOLIB select REGMAP help Support for ZL38060 Connected Home Audio Processor from Microsemi, From 2449d436681d40bc63ec2c766fd51b632270d8a7 Mon Sep 17 00:00:00 2001 From: Krishna Yarlagadda Date: Tue, 28 Feb 2023 01:34:28 +0530 Subject: [PATCH 594/765] spi: tegra210-quad: Fix iterator outside loop Fix warn: iterator used outside loop: 'xfer'. 'xfer' variable contain invalid value in few conditions. Complete transfer within DATA phase in successful case and at the end for failed transfer. Reported-by: Dan Carpenter Link:https://lore.kernel.org/all/202210191211.46FkzKmv-lkp@intel.com/ Fixes: 8777dd9dff40 ("spi: tegra210-quad: Fix combined sequence") Signed-off-by: Krishna Yarlagadda Link: https://lore.kernel.org/r/20230227200428.45832-1-kyarlagadda@nvidia.com Signed-off-by: Mark Brown --- drivers/spi/spi-tegra210-quad.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/spi/spi-tegra210-quad.c b/drivers/spi/spi-tegra210-quad.c index 9aaca2289c596..0b9bc3b7f53a7 100644 --- a/drivers/spi/spi-tegra210-quad.c +++ b/drivers/spi/spi-tegra210-quad.c @@ -1156,6 +1156,10 @@ static int tegra_qspi_combined_seq_xfer(struct tegra_qspi *tqspi, ret = -EIO; goto exit; } + if (!xfer->cs_change) { + tegra_qspi_transfer_end(spi); + spi_transfer_delay_exec(xfer); + } break; default: ret = -EINVAL; @@ -1164,14 +1168,14 @@ static int tegra_qspi_combined_seq_xfer(struct tegra_qspi *tqspi, msg->actual_length += xfer->len; transfer_phase++; } - if (!xfer->cs_change) { - tegra_qspi_transfer_end(spi); - spi_transfer_delay_exec(xfer); - } ret = 0; exit: msg->status = ret; + if (ret < 0) { + tegra_qspi_transfer_end(spi); + spi_transfer_delay_exec(xfer); + } return ret; } From d52279d5c9204a041e9ba02a66a353573b2f96e4 Mon Sep 17 00:00:00 2001 From: Duc Anh Le Date: Tue, 28 Feb 2023 00:49:21 +0100 Subject: [PATCH 595/765] ASoC: amd: yc: Add DMI entries to support HP OMEN 16-n0xxx (8A43) This model requires an additional detection quirk to enable the internal microphone. Signed-off-by: Duc Anh Le Link: https://lore.kernel.org/r/20230227234921.7784-1-lub.the.studio@gmail.com Signed-off-by: Mark Brown --- sound/soc/amd/yc/acp6x-mach.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/sound/soc/amd/yc/acp6x-mach.c b/sound/soc/amd/yc/acp6x-mach.c index c061519adfbe4..9c183d4c3ce64 100644 --- a/sound/soc/amd/yc/acp6x-mach.c +++ b/sound/soc/amd/yc/acp6x-mach.c @@ -255,6 +255,13 @@ static const struct dmi_system_id yc_acp_quirk_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "OMEN by HP Gaming Laptop 16z-n000"), } }, + { + .driver_data = &acp6x_card, + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "HP"), + DMI_MATCH(DMI_BOARD_NAME, "8A43"), + } + }, {} }; From 6934cf8a3e0b4a8318ce8f1342348e967e29192f Mon Sep 17 00:00:00 2001 From: Heiko Stuebner Date: Wed, 8 Feb 2023 23:53:28 +0100 Subject: [PATCH 596/765] RISC-V: improve string-function assembly Adapt the suggestions for the assembly string functions that Andrew suggested but that I didn't manage to include into the series that got applied. This includes improvements to two comments, removal of unneeded labels and moving one instruction slightly higher to contradict an explanatory comment. Suggested-by: Andrew Jones Signed-off-by: Heiko Stuebner Reviewed-by: Andrew Jones Tested-by: Conor Dooley Link: https://lore.kernel.org/r/20230208225328.1636017-3-heiko@sntech.de Signed-off-by: Palmer Dabbelt --- arch/riscv/lib/strcmp.S | 6 ++++-- arch/riscv/lib/strlen.S | 10 +++++----- arch/riscv/lib/strncmp.S | 16 +++++++--------- 3 files changed, 16 insertions(+), 16 deletions(-) diff --git a/arch/riscv/lib/strcmp.S b/arch/riscv/lib/strcmp.S index 986ab23fe7877..c42a8412547f6 100644 --- a/arch/riscv/lib/strcmp.S +++ b/arch/riscv/lib/strcmp.S @@ -40,7 +40,9 @@ SYM_FUNC_START(strcmp) ret /* - * Variant of strcmp using the ZBB extension if available + * Variant of strcmp using the ZBB extension if available. + * The code was published as part of the bitmanip manual + * in Appendix A. */ #ifdef CONFIG_RISCV_ISA_ZBB strcmp_zbb: @@ -57,7 +59,7 @@ strcmp_zbb: * a1 - string2 * * Clobbers - * t0, t1, t2, t3, t4, t5 + * t0, t1, t2, t3, t4 */ or t2, a0, a1 diff --git a/arch/riscv/lib/strlen.S b/arch/riscv/lib/strlen.S index 8345ceeee3f65..15bb8f3aa959e 100644 --- a/arch/riscv/lib/strlen.S +++ b/arch/riscv/lib/strlen.S @@ -96,7 +96,7 @@ strlen_zbb: * of valid bytes in this chunk. */ srli a0, t1, 3 - bgtu t3, a0, 3f + bgtu t3, a0, 2f /* Prepare for the word comparison loop. */ addi t2, t0, SZREG @@ -112,20 +112,20 @@ strlen_zbb: addi t0, t0, SZREG orc.b t1, t1 beq t1, t3, 1b -2: + not t1, t1 CZ t1, t1 + srli t1, t1, 3 - /* Get number of processed words. */ + /* Get number of processed bytes. */ sub t2, t0, t2 /* Add number of characters in the first word. */ add a0, a0, t2 - srli t1, t1, 3 /* Add number of characters in the last word. */ add a0, a0, t1 -3: +2: ret .option pop diff --git a/arch/riscv/lib/strncmp.S b/arch/riscv/lib/strncmp.S index ee49595075bee..e09de79e62b96 100644 --- a/arch/riscv/lib/strncmp.S +++ b/arch/riscv/lib/strncmp.S @@ -70,7 +70,7 @@ strncmp_zbb: li t5, -1 and t2, t2, SZREG-1 add t4, a0, a2 - bnez t2, 4f + bnez t2, 3f /* Adjust limit for fast-path. */ andi t6, t4, -SZREG @@ -114,23 +114,21 @@ strncmp_zbb: ret /* Simple loop for misaligned strings. */ -3: - /* Restore limit for slow-path. */ .p2align 3 -4: - bge a0, t4, 6f +3: + bge a0, t4, 5f lbu t0, 0(a0) lbu t1, 0(a1) addi a0, a0, 1 addi a1, a1, 1 - bne t0, t1, 5f - bnez t0, 4b + bne t0, t1, 4f + bnez t0, 3b -5: +4: sub a0, t0, t1 ret -6: +5: li a0, 0 ret From e32d546c483a2a0f607687f5b521c2a2f942ffbe Mon Sep 17 00:00:00 2001 From: Mario Limonciello Date: Thu, 23 Feb 2023 14:26:22 -0600 Subject: [PATCH 597/765] ACPI: x86: Drop quirk for HP Elitebook There was a quirk in `acpi/x86/s2idle.c` for an HP Elitebook G9 platforms to force AMD GUID codepath instead of Microsoft codepath. This was due to a bug with WCN6855 WLAN firmware interaction with the system. This bug is fixed by WCN6855 firmware: WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.23 Remove the quirk as it's no longer necessary with this firmware. Link: https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git/commit/?id=c7a57ef688f7d99d8338a5d8edddc8836ff0e6de Tested-by: Anson Tsao Signed-off-by: Mario Limonciello Signed-off-by: Rafael J. Wysocki --- drivers/acpi/x86/s2idle.c | 24 ------------------------ 1 file changed, 24 deletions(-) diff --git a/drivers/acpi/x86/s2idle.c b/drivers/acpi/x86/s2idle.c index c7afce465a071..e499c60c45791 100644 --- a/drivers/acpi/x86/s2idle.c +++ b/drivers/acpi/x86/s2idle.c @@ -384,29 +384,6 @@ static const struct acpi_device_id amd_hid_ids[] = { {} }; -static int lps0_prefer_amd(const struct dmi_system_id *id) -{ - pr_debug("Using AMD GUID w/ _REV 2.\n"); - rev_id = 2; - return 0; -} -static const struct dmi_system_id s2idle_dmi_table[] __initconst = { - { - /* - * AMD Rembrandt based HP EliteBook 835/845/865 G9 - * Contains specialized AML in AMD/_REV 2 path to avoid - * triggering a bug in Qualcomm WLAN firmware. This may be - * removed in the future if that firmware is fixed. - */ - .callback = lps0_prefer_amd, - .matches = { - DMI_MATCH(DMI_BOARD_VENDOR, "HP"), - DMI_MATCH(DMI_BOARD_NAME, "8990"), - }, - }, - {} -}; - static int lps0_device_attach(struct acpi_device *adev, const struct acpi_device_id *not_used) { @@ -586,7 +563,6 @@ static const struct platform_s2idle_ops acpi_s2idle_ops_lps0 = { void __init acpi_s2idle_setup(void) { - dmi_check_system(s2idle_dmi_table); acpi_scan_add_handler(&lps0_handler); s2idle_set_ops(&acpi_s2idle_ops_lps0); } From 5bd289f69bc145fa10927a6883502a0e7cbcb811 Mon Sep 17 00:00:00 2001 From: Nick Alcock Date: Fri, 24 Feb 2023 15:07:49 +0000 Subject: [PATCH 598/765] cpufreq: intel_pstate: remove MODULE_LICENSE in non-modules Since commit 8b41fc4454e ("kbuild: create modules.builtin without Makefile.modbuiltin or tristate.conf"), MODULE_LICENSE declarations are used to identify modules. As a consequence, uses of the macro in non-modules will cause modprobe to misidentify their containing object file as a module when it is not (false positives), and modprobe might succeed rather than failing with a suitable error message. So remove it in the files in this commit, none of which can be built as modules. Signed-off-by: Nick Alcock Suggested-by: Luis Chamberlain Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/intel_pstate.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index 32a4004d155dd..61d297199389e 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -3531,4 +3531,3 @@ early_param("intel_pstate", intel_pstate_setup); MODULE_AUTHOR("Dirk Brandewie "); MODULE_DESCRIPTION("'intel_pstate' - P state driver Intel Core processors"); -MODULE_LICENSE("GPL"); From d25d01b4e59bde815df52a333f9b0f7558cc3281 Mon Sep 17 00:00:00 2001 From: Nick Alcock Date: Tue, 28 Feb 2023 13:02:01 +0000 Subject: [PATCH 599/765] powercap: remove MODULE_LICENSE in non-modules Since commit 8b41fc4454e ("kbuild: create modules.builtin without Makefile.modbuiltin or tristate.conf"), MODULE_LICENSE declarations are used to identify modules. As a consequence, uses of the macro in non-modules will cause modprobe to misidentify their containing object file as a module when it is not (false positives), and modprobe might succeed rather than failing with a suitable error message. So remove it in the files in this commit, none of which can be built as modules. Signed-off-by: Nick Alcock Suggested-by: Luis Chamberlain Signed-off-by: Rafael J. Wysocki --- drivers/powercap/powercap_sys.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/powercap/powercap_sys.c b/drivers/powercap/powercap_sys.c index e180dee0f83d0..52c32dcbf7d84 100644 --- a/drivers/powercap/powercap_sys.c +++ b/drivers/powercap/powercap_sys.c @@ -679,4 +679,3 @@ fs_initcall(powercap_init); MODULE_DESCRIPTION("PowerCap sysfs Driver"); MODULE_AUTHOR("Srinivas Pandruvada "); -MODULE_LICENSE("GPL v2"); From 79f76e574c3690261d6165f008b7054e54c6dca9 Mon Sep 17 00:00:00 2001 From: Sergio Paracuellos Date: Tue, 14 Feb 2023 11:39:33 +0100 Subject: [PATCH 600/765] mips: dts: ralink: mt7621: rename watchdog node from 'wdt' into 'watchdog' MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Watchdog nodes must use 'watchdog' for node name. When a 'make dtbs_check' is performed the following warning appears: wdt@100: $nodename:0: 'wdt@100' does not match '^watchdog(@.*|-[0-9a-f])?$' Fix this warning up properly renaming the node into 'watchdog'. Reviewed-by: Arınç ÜNAL Reviewed-by: Philippe Mathieu-Daudé Signed-off-by: Sergio Paracuellos Acked-by: Guenter Roeck Signed-off-by: Thomas Bogendoerfer --- arch/mips/boot/dts/ralink/mt7621.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/boot/dts/ralink/mt7621.dtsi b/arch/mips/boot/dts/ralink/mt7621.dtsi index 290d47fbcfbb7..c16295d27fa67 100644 --- a/arch/mips/boot/dts/ralink/mt7621.dtsi +++ b/arch/mips/boot/dts/ralink/mt7621.dtsi @@ -70,7 +70,7 @@ "250m", "270m"; }; - wdt: wdt@100 { + wdt: watchdog@100 { compatible = "mediatek,mt7621-wdt"; reg = <0x100 0x100>; }; From 70f864d1084734f8816a247c24e6876d2dfb5f89 Mon Sep 17 00:00:00 2001 From: Sergio Paracuellos Date: Tue, 14 Feb 2023 11:39:34 +0100 Subject: [PATCH 601/765] mips: dts: ralink: mt7621: add phandle to system controller node for watchdog MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit To allow to access system controller registers from watchdog driver code add a phandle in the watchdog 'wdt' node. This avoid using arch dependent operations in driver code. Reviewed-by: Arınç ÜNAL Signed-off-by: Sergio Paracuellos Acked-by: Guenter Roeck Signed-off-by: Thomas Bogendoerfer --- arch/mips/boot/dts/ralink/mt7621.dtsi | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/mips/boot/dts/ralink/mt7621.dtsi b/arch/mips/boot/dts/ralink/mt7621.dtsi index c16295d27fa67..7caed0d14f11a 100644 --- a/arch/mips/boot/dts/ralink/mt7621.dtsi +++ b/arch/mips/boot/dts/ralink/mt7621.dtsi @@ -73,6 +73,7 @@ wdt: watchdog@100 { compatible = "mediatek,mt7621-wdt"; reg = <0x100 0x100>; + mediatek,sysctl = <&sysc>; }; gpio: gpio@600 { From 1a2c73f4834dd79e4f2c590ac75358fb44137650 Mon Sep 17 00:00:00 2001 From: Jiaxun Yang Date: Tue, 28 Feb 2023 19:34:59 +0000 Subject: [PATCH 602/765] MIPS: Workaround clang inline compat branch issue Clang is unable to handle the situation that a chunk of inline assembly ends with a compat branch instruction and then compiler generates another control transfer instruction immediately after this compat branch. The later instruction will end up in forbidden slot and cause exception. Workaround by add a option to control the use of compact branch. Currently it's selected by CC_IS_CLANG and hopefully we can change it to a version check in future if clang manages to fix it. Fix boot on boston board. Link: https://github.com/llvm/llvm-project/issues/61045 Signed-off-by: Jiaxun Yang Acked-by: Nathan Chancellor Acked-by: Nick Desaulniers Signed-off-by: Thomas Bogendoerfer --- arch/mips/Kconfig | 4 ++++ arch/mips/include/asm/asm.h | 2 +- 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 875c29246555b..e2f3ca73f40d6 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -3200,6 +3200,10 @@ config CC_HAS_MNO_BRANCH_LIKELY def_bool y depends on $(cc-option,-mno-branch-likely) +# https://github.com/llvm/llvm-project/issues/61045 +config CC_HAS_BROKEN_INLINE_COMPAT_BRANCH + def_bool y if CC_IS_CLANG + menu "Power management options" config ARCH_HIBERNATION_POSSIBLE diff --git a/arch/mips/include/asm/asm.h b/arch/mips/include/asm/asm.h index 336ac9b652350..2e99450f42284 100644 --- a/arch/mips/include/asm/asm.h +++ b/arch/mips/include/asm/asm.h @@ -336,7 +336,7 @@ symbol = value */ #ifdef CONFIG_WAR_R10000_LLSC # define SC_BEQZ beqzl -#elif MIPS_ISA_REV >= 6 +#elif !defined(CONFIG_CC_HAS_BROKEN_INLINE_COMPAT_BRANCH) && MIPS_ISA_REV >= 6 # define SC_BEQZ beqzc #else # define SC_BEQZ beqz From ae44f1c9d1fc54aeceb335fedb1e73b2c3ee4561 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 24 Feb 2023 17:59:39 +0200 Subject: [PATCH 603/765] powerpc: dts: t1040rdb: fix compatible string for Rev A boards It looks like U-Boot fails to start the kernel properly when the compatible string of the board isn't fsl,T1040RDB, so stop overriding it from the rev-a.dts. Fixes: 5ebb74749202 ("powerpc: dts: t1040rdb: fix ports names for Seville Ethernet switch") Signed-off-by: Vladimir Oltean Signed-off-by: Jakub Kicinski --- arch/powerpc/boot/dts/fsl/t1040rdb-rev-a.dts | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/powerpc/boot/dts/fsl/t1040rdb-rev-a.dts b/arch/powerpc/boot/dts/fsl/t1040rdb-rev-a.dts index 73f8c998c64df..d4f5f159d6f23 100644 --- a/arch/powerpc/boot/dts/fsl/t1040rdb-rev-a.dts +++ b/arch/powerpc/boot/dts/fsl/t1040rdb-rev-a.dts @@ -10,7 +10,6 @@ / { model = "fsl,T1040RDB-REV-A"; - compatible = "fsl,T1040RDB-REV-A"; }; &seville_port0 { From 8b322f9fdb35ecee0d4a40de8af923444bd624ea Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 24 Feb 2023 17:59:40 +0200 Subject: [PATCH 604/765] powerpc: dts: t1040rdb: enable both CPU ports Since commit eca70102cfb1 ("net: dsa: felix: add support for changing DSA master") included in kernel v6.1, the driver supports 2 CPU ports, and they can be put in a LAG, for example (see Documentation/networking/dsa/configuration.rst for more details). Defining the second CPU port in the device tree should not cause any compatibility issue, because the default CPU port was &seville_port8 before this change, and still is &seville_port8 now (the numerically first CPU port is used by default). Signed-off-by: Vladimir Oltean Signed-off-by: Jakub Kicinski --- arch/powerpc/boot/dts/fsl/t1040rdb.dts | 5 ++++- arch/powerpc/boot/dts/fsl/t1040si-post.dtsi | 2 ++ 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/boot/dts/fsl/t1040rdb.dts b/arch/powerpc/boot/dts/fsl/t1040rdb.dts index b6733e7e65805..dd3aab81e9dea 100644 --- a/arch/powerpc/boot/dts/fsl/t1040rdb.dts +++ b/arch/powerpc/boot/dts/fsl/t1040rdb.dts @@ -180,6 +180,9 @@ }; &seville_port8 { - ethernet = <&enet0>; + status = "okay"; +}; + +&seville_port9 { status = "okay"; }; diff --git a/arch/powerpc/boot/dts/fsl/t1040si-post.dtsi b/arch/powerpc/boot/dts/fsl/t1040si-post.dtsi index f58eb820eb5ec..ad0ab33336b88 100644 --- a/arch/powerpc/boot/dts/fsl/t1040si-post.dtsi +++ b/arch/powerpc/boot/dts/fsl/t1040si-post.dtsi @@ -686,6 +686,7 @@ seville_port8: port@8 { reg = <8>; phy-mode = "internal"; + ethernet = <&enet0>; status = "disabled"; fixed-link { @@ -697,6 +698,7 @@ seville_port9: port@9 { reg = <9>; phy-mode = "internal"; + ethernet = <&enet1>; status = "disabled"; fixed-link { From 4d42cd6bc2ac1b9be50ade13771daec90c9d18b1 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 27 Feb 2023 10:12:01 -0800 Subject: [PATCH 605/765] tls: rx: fix return value for async crypto Gaurav reports that TLS Rx is broken with async crypto accelerators. The commit under fixes missed updating the retval byte counting logic when updating how records are stored. Even tho both before and after the change 'decrypted' was updated inside the main loop, it was completely overwritten when processing the async completions. Now that the rx_list only holds non-zero-copy records we need to add, not overwrite. Reported-and-bisected-by: Gaurav Jain Fixes: cbbdee9918a2 ("tls: rx: async: don't put async zc on the list") Link: https://bugzilla.kernel.org/show_bug.cgi?id=217064 Tested-by: Gaurav Jain Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20230227181201.1793772-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/tls/tls_sw.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index 782d3701b86f4..021d760f91335 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -2127,7 +2127,7 @@ int tls_sw_recvmsg(struct sock *sk, else err = process_rx_list(ctx, msg, &control, 0, async_copy_bytes, is_peek); - decrypted = max(err, 0); + decrypted += max(err, 0); } copied += decrypted; From 510d2358c466bf6588034f0d3b2266eed2bc0a51 Mon Sep 17 00:00:00 2001 From: Jack Chen Date: Thu, 16 Feb 2023 10:10:57 -0500 Subject: [PATCH 606/765] i3c: master: dw: stop hardcoding initial speed Bus-speed could be default(12.5MHz) or defined by users in dts. Dw-i3c-master should not hard-code the initial speed to be I3C_BUS_TYP_I3C_SCL_RATE (12.5MHz) And because of Synopsys's I3C controller limit (hcnt/lcnt register length) and core-clk provided, there is a limit to bus speed, too. For example, when core-clk is 250 MHz, the bus speed cannot be lowered below 1MHz. Tested: tested with an i3c sensor and captured with a logic analyzer. Signed-off-by: Jack Chen Link: https://lore.kernel.org/r/20230216151057.293764-1-zenghuchen@google.com Signed-off-by: Alexandre Belloni --- drivers/i3c/master/dw-i3c-master.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/i3c/master/dw-i3c-master.c b/drivers/i3c/master/dw-i3c-master.c index 51a8608203de7..48954d3e65714 100644 --- a/drivers/i3c/master/dw-i3c-master.c +++ b/drivers/i3c/master/dw-i3c-master.c @@ -531,7 +531,7 @@ static int dw_i3c_clk_cfg(struct dw_i3c_master *master) if (hcnt < SCL_I3C_TIMING_CNT_MIN) hcnt = SCL_I3C_TIMING_CNT_MIN; - lcnt = DIV_ROUND_UP(core_rate, I3C_BUS_TYP_I3C_SCL_RATE) - hcnt; + lcnt = DIV_ROUND_UP(core_rate, master->base.bus.scl_rate.i3c) - hcnt; if (lcnt < SCL_I3C_TIMING_CNT_MIN) lcnt = SCL_I3C_TIMING_CNT_MIN; @@ -541,7 +541,8 @@ static int dw_i3c_clk_cfg(struct dw_i3c_master *master) if (!(readl(master->regs + DEVICE_CTRL) & DEV_CTRL_I2C_SLAVE_PRESENT)) writel(BUS_I3C_MST_FREE(lcnt), master->regs + BUS_FREE_TIMING); - lcnt = DIV_ROUND_UP(I3C_BUS_TLOW_OD_MIN_NS, core_period); + lcnt = max_t(u8, + DIV_ROUND_UP(I3C_BUS_TLOW_OD_MIN_NS, core_period), lcnt); scl_timing = SCL_I3C_TIMING_HCNT(hcnt) | SCL_I3C_TIMING_LCNT(lcnt); writel(scl_timing, master->regs + SCL_I3C_OD_TIMING); From 81a1dd10b072fd432f37cd5dc9eb3ed6ec5d386e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Bj=C3=B6rn=20T=C3=B6pel?= Date: Tue, 28 Feb 2023 19:42:10 +0100 Subject: [PATCH 607/765] riscv, lib: Fix Zbb strncmp MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The Zbb optimized strncmp has two parts; a fast path that does XLEN/8B per iteration, and a slow that does one byte per iteration. The idea is to compare aligned XLEN chunks for most of strings, and do the remainder tail in the slow path. The Zbb strncmp has two issues in the fast path: Incorrect remainder handling (wrong compare): Assume that the string length is 9. On 64b systems, the fast path should do one iteration, and one iteration in the slow path. Instead, both were done in the fast path, which lead to incorrect results. An example: strncmp("/dev/vda", "/dev/", 5); Correct by changing "bgt" to "bge". Missing NULL checks in the second string: This could lead to incorrect results for: strncmp("/dev/vda", "/dev/vda\0", 8); Correct by adding an additional check. Fixes: b6fcdb191e36 ("RISC-V: add zbb support to string functions") Suggested-by: Heiko Stuebner Signed-off-by: Björn Töpel Tested-by: Conor Dooley Tested-by: Guenter Roeck Link: https://lore.kernel.org/r/20230228184211.1585641-1-bjorn@kernel.org Signed-off-by: Palmer Dabbelt --- arch/riscv/lib/strncmp.S | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/riscv/lib/strncmp.S b/arch/riscv/lib/strncmp.S index e09de79e62b96..7ac2f667285ab 100644 --- a/arch/riscv/lib/strncmp.S +++ b/arch/riscv/lib/strncmp.S @@ -78,11 +78,13 @@ strncmp_zbb: /* Main loop for aligned string. */ .p2align 3 1: - bgt a0, t6, 3f + bge a0, t6, 3f REG_L t0, 0(a0) REG_L t1, 0(a1) orc.b t3, t0 bne t3, t5, 2f + orc.b t3, t1 + bne t3, t5, 2f addi a0, a0, SZREG addi a1, a1, SZREG beq t0, t1, 1b From 880ce5f20033cd6ecb2c0edfe0376c9e45220012 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 27 Feb 2023 14:17:06 +0000 Subject: [PATCH 608/765] net: avoid skb end_offset change in __skb_unclone_keeptruesize() Once initial skb->head has been allocated from skb_small_head_cache, we need to make sure to use the same strategy whenever skb->head has to be re-allocated, as found by syzbot [1] This means kmalloc_reserve() can not fallback from using skb_small_head_cache to generic (power-of-two) kmem caches. It seems that we probably want to rework things in the future, to partially revert following patch, because we no longer use ksize() for skb allocated in TX path. 2b88cba55883 ("net: preserve skb_end_offset() in skb_unclone_keeptruesize()") Ideally, TCP stack should never put payload in skb->head, this effort has to be completed. In the mean time, add a sanity check. [1] BUG: KASAN: invalid-free in slab_free mm/slub.c:3787 [inline] BUG: KASAN: invalid-free in kmem_cache_free+0xee/0x5c0 mm/slub.c:3809 Free of addr ffff88806cdee800 by task syz-executor239/5189 CPU: 0 PID: 5189 Comm: syz-executor239 Not tainted 6.2.0-rc8-syzkaller-02400-gd1fabc68f8e0 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/21/2023 Call Trace: __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0xd1/0x138 lib/dump_stack.c:106 print_address_description mm/kasan/report.c:306 [inline] print_report+0x15e/0x45d mm/kasan/report.c:417 kasan_report_invalid_free+0x9b/0x1b0 mm/kasan/report.c:482 ____kasan_slab_free+0x1a5/0x1c0 mm/kasan/common.c:216 kasan_slab_free include/linux/kasan.h:177 [inline] slab_free_hook mm/slub.c:1781 [inline] slab_free_freelist_hook+0x8b/0x1c0 mm/slub.c:1807 slab_free mm/slub.c:3787 [inline] kmem_cache_free+0xee/0x5c0 mm/slub.c:3809 skb_kfree_head net/core/skbuff.c:857 [inline] skb_kfree_head net/core/skbuff.c:853 [inline] skb_free_head+0x16f/0x1a0 net/core/skbuff.c:872 skb_release_data+0x57a/0x820 net/core/skbuff.c:901 skb_release_all net/core/skbuff.c:966 [inline] __kfree_skb+0x4f/0x70 net/core/skbuff.c:980 tcp_wmem_free_skb include/net/tcp.h:302 [inline] tcp_rtx_queue_purge net/ipv4/tcp.c:3061 [inline] tcp_write_queue_purge+0x617/0xcf0 net/ipv4/tcp.c:3074 tcp_v4_destroy_sock+0x125/0x810 net/ipv4/tcp_ipv4.c:2302 inet_csk_destroy_sock+0x19a/0x440 net/ipv4/inet_connection_sock.c:1195 __tcp_close+0xb96/0xf50 net/ipv4/tcp.c:3021 tcp_close+0x2d/0xc0 net/ipv4/tcp.c:3033 inet_release+0x132/0x270 net/ipv4/af_inet.c:426 __sock_release+0xcd/0x280 net/socket.c:651 sock_close+0x1c/0x20 net/socket.c:1393 __fput+0x27c/0xa90 fs/file_table.c:320 task_work_run+0x16f/0x270 kernel/task_work.c:179 resume_user_mode_work include/linux/resume_user_mode.h:49 [inline] exit_to_user_mode_loop kernel/entry/common.c:171 [inline] exit_to_user_mode_prepare+0x23c/0x250 kernel/entry/common.c:203 __syscall_exit_to_user_mode_work kernel/entry/common.c:285 [inline] syscall_exit_to_user_mode+0x1d/0x50 kernel/entry/common.c:296 do_syscall_64+0x46/0xb0 arch/x86/entry/common.c:86 entry_SYSCALL_64_after_hwframe+0x63/0xcd RIP: 0033:0x7f2511f546c3 Code: c7 c2 c0 ff ff ff f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b7 0f 1f 00 64 8b 04 25 18 00 00 00 85 c0 75 14 b8 03 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 45 c3 0f 1f 40 00 48 83 ec 18 89 7c 24 0c e8 RSP: 002b:00007ffef0103d48 EFLAGS: 00000246 ORIG_RAX: 0000000000000003 RAX: 0000000000000000 RBX: 0000000000000004 RCX: 00007f2511f546c3 RDX: 0000000000000978 RSI: 00000000200000c0 RDI: 0000000000000003 RBP: 0000000000000000 R08: 0000000000000002 R09: 0000000000003434 R10: 0000000000000000 R11: 0000000000000246 R12: 00007ffef0103d6c R13: 00007ffef0103d80 R14: 00007ffef0103dc0 R15: 0000000000000003 Allocated by task 5189: kasan_save_stack+0x22/0x40 mm/kasan/common.c:45 kasan_set_track+0x25/0x30 mm/kasan/common.c:52 ____kasan_kmalloc mm/kasan/common.c:374 [inline] ____kasan_kmalloc mm/kasan/common.c:333 [inline] __kasan_kmalloc+0xa5/0xb0 mm/kasan/common.c:383 kasan_kmalloc include/linux/kasan.h:211 [inline] __do_kmalloc_node mm/slab_common.c:968 [inline] __kmalloc_node_track_caller+0x5b/0xc0 mm/slab_common.c:988 kmalloc_reserve+0xf1/0x230 net/core/skbuff.c:539 pskb_expand_head+0x237/0x1160 net/core/skbuff.c:1995 __skb_unclone_keeptruesize+0x93/0x220 net/core/skbuff.c:2094 skb_unclone_keeptruesize include/linux/skbuff.h:1910 [inline] skb_prepare_for_shift net/core/skbuff.c:3804 [inline] skb_shift+0xef8/0x1e20 net/core/skbuff.c:3877 tcp_skb_shift net/ipv4/tcp_input.c:1538 [inline] tcp_shift_skb_data net/ipv4/tcp_input.c:1646 [inline] tcp_sacktag_walk+0x93b/0x18a0 net/ipv4/tcp_input.c:1713 tcp_sacktag_write_queue+0x1599/0x31d0 net/ipv4/tcp_input.c:1974 tcp_ack+0x2e9f/0x5a10 net/ipv4/tcp_input.c:3847 tcp_rcv_established+0x667/0x2230 net/ipv4/tcp_input.c:6006 tcp_v4_do_rcv+0x670/0x9b0 net/ipv4/tcp_ipv4.c:1721 sk_backlog_rcv include/net/sock.h:1113 [inline] __release_sock+0x133/0x3b0 net/core/sock.c:2921 release_sock+0x58/0x1b0 net/core/sock.c:3488 tcp_sendmsg+0x3a/0x50 net/ipv4/tcp.c:1485 inet_sendmsg+0x9d/0xe0 net/ipv4/af_inet.c:825 sock_sendmsg_nosec net/socket.c:722 [inline] sock_sendmsg+0xde/0x190 net/socket.c:745 sock_write_iter+0x295/0x3d0 net/socket.c:1136 call_write_iter include/linux/fs.h:2189 [inline] new_sync_write fs/read_write.c:491 [inline] vfs_write+0x9ed/0xdd0 fs/read_write.c:584 ksys_write+0x1ec/0x250 fs/read_write.c:637 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x39/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd The buggy address belongs to the object at ffff88806cdee800 which belongs to the cache kmalloc-1k of size 1024 The buggy address is located 0 bytes inside of 1024-byte region [ffff88806cdee800, ffff88806cdeec00) The buggy address belongs to the physical page: page:ffffea0001b37a00 refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x6cde8 head:ffffea0001b37a00 order:3 compound_mapcount:0 subpages_mapcount:0 compound_pincount:0 flags: 0xfff00000010200(slab|head|node=0|zone=1|lastcpupid=0x7ff) raw: 00fff00000010200 ffff888012441dc0 dead000000000122 0000000000000000 raw: 0000000000000000 0000000000100010 00000001ffffffff 0000000000000000 page dumped because: kasan: bad access detected page_owner tracks the page as allocated page last allocated via order 3, migratetype Unmovable, gfp_mask 0x1f2a20(GFP_ATOMIC|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC|__GFP_MEMALLOC|__GFP_HARDWALL), pid 75, tgid 75 (kworker/u4:4), ts 96369578780, free_ts 26734162530 prep_new_page mm/page_alloc.c:2531 [inline] get_page_from_freelist+0x119c/0x2ce0 mm/page_alloc.c:4283 __alloc_pages+0x1cb/0x5b0 mm/page_alloc.c:5549 alloc_pages+0x1aa/0x270 mm/mempolicy.c:2287 alloc_slab_page mm/slub.c:1851 [inline] allocate_slab+0x25f/0x350 mm/slub.c:1998 new_slab mm/slub.c:2051 [inline] ___slab_alloc+0xa91/0x1400 mm/slub.c:3193 __slab_alloc.constprop.0+0x56/0xa0 mm/slub.c:3292 __slab_alloc_node mm/slub.c:3345 [inline] slab_alloc_node mm/slub.c:3442 [inline] __kmem_cache_alloc_node+0x1a4/0x430 mm/slub.c:3491 __do_kmalloc_node mm/slab_common.c:967 [inline] __kmalloc_node_track_caller+0x4b/0xc0 mm/slab_common.c:988 kmalloc_reserve+0xf1/0x230 net/core/skbuff.c:539 __alloc_skb+0x129/0x330 net/core/skbuff.c:608 __netdev_alloc_skb+0x74/0x410 net/core/skbuff.c:672 __netdev_alloc_skb_ip_align include/linux/skbuff.h:3203 [inline] netdev_alloc_skb_ip_align include/linux/skbuff.h:3213 [inline] batadv_iv_ogm_aggregate_new+0x106/0x4e0 net/batman-adv/bat_iv_ogm.c:558 batadv_iv_ogm_queue_add net/batman-adv/bat_iv_ogm.c:670 [inline] batadv_iv_ogm_schedule_buff+0xe6b/0x1450 net/batman-adv/bat_iv_ogm.c:849 batadv_iv_ogm_schedule net/batman-adv/bat_iv_ogm.c:868 [inline] batadv_iv_ogm_schedule net/batman-adv/bat_iv_ogm.c:861 [inline] batadv_iv_send_outstanding_bat_ogm_packet+0x744/0x910 net/batman-adv/bat_iv_ogm.c:1712 process_one_work+0x9bf/0x1710 kernel/workqueue.c:2289 worker_thread+0x669/0x1090 kernel/workqueue.c:2436 page last free stack trace: reset_page_owner include/linux/page_owner.h:24 [inline] free_pages_prepare mm/page_alloc.c:1446 [inline] free_pcp_prepare+0x66a/0xc20 mm/page_alloc.c:1496 free_unref_page_prepare mm/page_alloc.c:3369 [inline] free_unref_page+0x1d/0x490 mm/page_alloc.c:3464 free_contig_range+0xb5/0x180 mm/page_alloc.c:9488 destroy_args+0xa8/0x64c mm/debug_vm_pgtable.c:998 debug_vm_pgtable+0x28de/0x296f mm/debug_vm_pgtable.c:1318 do_one_initcall+0x141/0x790 init/main.c:1306 do_initcall_level init/main.c:1379 [inline] do_initcalls init/main.c:1395 [inline] do_basic_setup init/main.c:1414 [inline] kernel_init_freeable+0x6f9/0x782 init/main.c:1634 kernel_init+0x1e/0x1d0 init/main.c:1522 ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:308 Memory state around the buggy address: ffff88806cdee700: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc ffff88806cdee780: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc >ffff88806cdee800: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ^ ffff88806cdee880: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 Fixes: bf9f1baa279f ("net: add dedicated kmem_cache for typical/small skb->head") Reported-by: syzbot Signed-off-by: Eric Dumazet Tested-by: Christoph Paasch Signed-off-by: David S. Miller --- net/core/skbuff.c | 31 +++++++++++++++++++++++-------- 1 file changed, 23 insertions(+), 8 deletions(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index eb7d33b41e710..1a31815104d61 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -517,18 +517,16 @@ static void *kmalloc_reserve(unsigned int *size, gfp_t flags, int node, #ifdef HAVE_SKB_SMALL_HEAD_CACHE if (obj_size <= SKB_SMALL_HEAD_CACHE_SIZE && !(flags & KMALLOC_NOT_NORMAL_BITS)) { - - /* skb_small_head_cache has non power of two size, - * likely forcing SLUB to use order-3 pages. - * We deliberately attempt a NOMEMALLOC allocation only. - */ obj = kmem_cache_alloc_node(skb_small_head_cache, flags | __GFP_NOMEMALLOC | __GFP_NOWARN, node); - if (obj) { - *size = SKB_SMALL_HEAD_CACHE_SIZE; + *size = SKB_SMALL_HEAD_CACHE_SIZE; + if (obj || !(gfp_pfmemalloc_allowed(flags))) goto out; - } + /* Try again but now we are using pfmemalloc reserves */ + ret_pfmemalloc = true; + obj = kmem_cache_alloc_node(skb_small_head_cache, flags, node); + goto out; } #endif *size = obj_size = kmalloc_size_roundup(obj_size); @@ -2082,6 +2080,7 @@ struct sk_buff *skb_realloc_headroom(struct sk_buff *skb, unsigned int headroom) } EXPORT_SYMBOL(skb_realloc_headroom); +/* Note: We plan to rework this in linux-6.4 */ int __skb_unclone_keeptruesize(struct sk_buff *skb, gfp_t pri) { unsigned int saved_end_offset, saved_truesize; @@ -2100,6 +2099,22 @@ int __skb_unclone_keeptruesize(struct sk_buff *skb, gfp_t pri) if (likely(skb_end_offset(skb) == saved_end_offset)) return 0; +#ifdef HAVE_SKB_SMALL_HEAD_CACHE + /* We can not change skb->end if the original or new value + * is SKB_SMALL_HEAD_HEADROOM, as it might break skb_kfree_head(). + */ + if (saved_end_offset == SKB_SMALL_HEAD_HEADROOM || + skb_end_offset(skb) == SKB_SMALL_HEAD_HEADROOM) { + /* We think this path should not be taken. + * Add a temporary trace to warn us just in case. + */ + pr_err_once("__skb_unclone_keeptruesize() skb_end_offset() %u -> %u\n", + saved_end_offset, skb_end_offset(skb)); + WARN_ON_ONCE(1); + return 0; + } +#endif + shinfo = skb_shinfo(skb); /* We are about to change back skb->end, From fb07390463c95e6eef254044d6dde050bfb9807a Mon Sep 17 00:00:00 2001 From: Pedro Tammela Date: Mon, 27 Feb 2023 12:23:52 -0300 Subject: [PATCH 609/765] net/sched: act_connmark: handle errno on tcf_idr_check_alloc Smatch reports that 'ci' can be used uninitialized. The current code ignores errno coming from tcf_idr_check_alloc, which will lead to the incorrect usage of 'ci'. Handle the errno as it should. Fixes: 288864effe33 ("net/sched: act_connmark: transition to percpu stats and rcu") Reviewed-by: Jamal Hadi Salim Signed-off-by: Pedro Tammela Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/act_connmark.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/sched/act_connmark.c b/net/sched/act_connmark.c index 8dabfb52ea3db..0d7aee8933c5f 100644 --- a/net/sched/act_connmark.c +++ b/net/sched/act_connmark.c @@ -158,6 +158,9 @@ static int tcf_connmark_init(struct net *net, struct nlattr *nla, nparms->zone = parm->zone; ret = 0; + } else { + err = ret; + goto out_free; } err = tcf_action_check_ctrlact(parm->action, tp, &goto_ch, extack); From 693aa2c0d9b6d5b1f2745d31b6e70d09dbbaf06e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 27 Feb 2023 15:30:24 +0000 Subject: [PATCH 610/765] ila: do not generate empty messages in ila_xlat_nl_cmd_get_mapping() ila_xlat_nl_cmd_get_mapping() generates an empty skb, triggerring a recent sanity check [1]. Instead, return an error code, so that user space can get it. [1] skb_assert_len WARNING: CPU: 0 PID: 5923 at include/linux/skbuff.h:2527 skb_assert_len include/linux/skbuff.h:2527 [inline] WARNING: CPU: 0 PID: 5923 at include/linux/skbuff.h:2527 __dev_queue_xmit+0x1bc0/0x3488 net/core/dev.c:4156 Modules linked in: CPU: 0 PID: 5923 Comm: syz-executor269 Not tainted 6.2.0-syzkaller-18300-g2ebd1fbb946d #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/21/2023 pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : skb_assert_len include/linux/skbuff.h:2527 [inline] pc : __dev_queue_xmit+0x1bc0/0x3488 net/core/dev.c:4156 lr : skb_assert_len include/linux/skbuff.h:2527 [inline] lr : __dev_queue_xmit+0x1bc0/0x3488 net/core/dev.c:4156 sp : ffff80001e0d6c40 x29: ffff80001e0d6e60 x28: dfff800000000000 x27: ffff0000c86328c0 x26: dfff800000000000 x25: ffff0000c8632990 x24: ffff0000c8632a00 x23: 0000000000000000 x22: 1fffe000190c6542 x21: ffff0000c8632a10 x20: ffff0000c8632a00 x19: ffff80001856e000 x18: ffff80001e0d5fc0 x17: 0000000000000000 x16: ffff80001235d16c x15: 0000000000000000 x14: 0000000000000000 x13: 0000000000000001 x12: 0000000000000001 x11: ff80800008353a30 x10: 0000000000000000 x9 : 21567eaf25bfb600 x8 : 21567eaf25bfb600 x7 : 0000000000000001 x6 : 0000000000000001 x5 : ffff80001e0d6558 x4 : ffff800015c74760 x3 : ffff800008596744 x2 : 0000000000000001 x1 : 0000000100000000 x0 : 000000000000000e Call trace: skb_assert_len include/linux/skbuff.h:2527 [inline] __dev_queue_xmit+0x1bc0/0x3488 net/core/dev.c:4156 dev_queue_xmit include/linux/netdevice.h:3033 [inline] __netlink_deliver_tap_skb net/netlink/af_netlink.c:307 [inline] __netlink_deliver_tap+0x45c/0x6f8 net/netlink/af_netlink.c:325 netlink_deliver_tap+0xf4/0x174 net/netlink/af_netlink.c:338 __netlink_sendskb net/netlink/af_netlink.c:1283 [inline] netlink_sendskb+0x6c/0x154 net/netlink/af_netlink.c:1292 netlink_unicast+0x334/0x8d4 net/netlink/af_netlink.c:1380 nlmsg_unicast include/net/netlink.h:1099 [inline] genlmsg_unicast include/net/genetlink.h:433 [inline] genlmsg_reply include/net/genetlink.h:443 [inline] ila_xlat_nl_cmd_get_mapping+0x620/0x7d0 net/ipv6/ila/ila_xlat.c:493 genl_family_rcv_msg_doit net/netlink/genetlink.c:968 [inline] genl_family_rcv_msg net/netlink/genetlink.c:1048 [inline] genl_rcv_msg+0x938/0xc1c net/netlink/genetlink.c:1065 netlink_rcv_skb+0x214/0x3c4 net/netlink/af_netlink.c:2574 genl_rcv+0x38/0x50 net/netlink/genetlink.c:1076 netlink_unicast_kernel net/netlink/af_netlink.c:1339 [inline] netlink_unicast+0x660/0x8d4 net/netlink/af_netlink.c:1365 netlink_sendmsg+0x800/0xae0 net/netlink/af_netlink.c:1942 sock_sendmsg_nosec net/socket.c:714 [inline] sock_sendmsg net/socket.c:734 [inline] ____sys_sendmsg+0x558/0x844 net/socket.c:2479 ___sys_sendmsg net/socket.c:2533 [inline] __sys_sendmsg+0x26c/0x33c net/socket.c:2562 __do_sys_sendmsg net/socket.c:2571 [inline] __se_sys_sendmsg net/socket.c:2569 [inline] __arm64_sys_sendmsg+0x80/0x94 net/socket.c:2569 __invoke_syscall arch/arm64/kernel/syscall.c:38 [inline] invoke_syscall+0x98/0x2c0 arch/arm64/kernel/syscall.c:52 el0_svc_common+0x138/0x258 arch/arm64/kernel/syscall.c:142 do_el0_svc+0x64/0x198 arch/arm64/kernel/syscall.c:193 el0_svc+0x58/0x168 arch/arm64/kernel/entry-common.c:637 el0t_64_sync_handler+0x84/0xf0 arch/arm64/kernel/entry-common.c:655 el0t_64_sync+0x190/0x194 arch/arm64/kernel/entry.S:591 irq event stamp: 136484 hardirqs last enabled at (136483): [] __up_console_sem+0x60/0xb4 kernel/printk/printk.c:345 hardirqs last disabled at (136484): [] el1_dbg+0x24/0x80 arch/arm64/kernel/entry-common.c:405 softirqs last enabled at (136418): [] softirq_handle_end kernel/softirq.c:414 [inline] softirqs last enabled at (136418): [] __do_softirq+0xd4c/0xfa4 kernel/softirq.c:600 softirqs last disabled at (136371): [] ____do_softirq+0x14/0x20 arch/arm64/kernel/irq.c:80 ---[ end trace 0000000000000000 ]--- skb len=0 headroom=0 headlen=0 tailroom=192 mac=(0,0) net=(0,-1) trans=-1 shinfo(txflags=0 nr_frags=0 gso(size=0 type=0 segs=0)) csum(0x0 ip_summed=0 complete_sw=0 valid=0 level=0) hash(0x0 sw=0 l4=0) proto=0x0010 pkttype=6 iif=0 dev name=nlmon0 feat=0x0000000000005861 Fixes: 7f00feaf1076 ("ila: Add generic ILA translation facility") Reported-by: syzbot Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv6/ila/ila_xlat.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv6/ila/ila_xlat.c b/net/ipv6/ila/ila_xlat.c index 47447f0241df6..bee45dfeb1874 100644 --- a/net/ipv6/ila/ila_xlat.c +++ b/net/ipv6/ila/ila_xlat.c @@ -477,6 +477,7 @@ int ila_xlat_nl_cmd_get_mapping(struct sk_buff *skb, struct genl_info *info) rcu_read_lock(); + ret = -ESRCH; ila = ila_lookup_by_params(&xp, ilan); if (ila) { ret = ila_dump_info(ila, From dfd2f0eb2347dbdf391fd5b8255fefc58a745472 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 27 Feb 2023 18:44:36 +0000 Subject: [PATCH 611/765] net/sched: flower: fix fl_change() error recovery path The two "goto errout;" paths in fl_change() became wrong after cited commit. Indeed we only must not call __fl_put() until the net pointer has been set in tcf_exts_init_ex() This is a minimal fix. We might in the future validate TCA_FLOWER_FLAGS before we allocate @fnew. BUG: KASAN: null-ptr-deref in instrument_atomic_read include/linux/instrumented.h:72 [inline] BUG: KASAN: null-ptr-deref in atomic_read include/linux/atomic/atomic-instrumented.h:27 [inline] BUG: KASAN: null-ptr-deref in refcount_read include/linux/refcount.h:147 [inline] BUG: KASAN: null-ptr-deref in __refcount_add_not_zero include/linux/refcount.h:152 [inline] BUG: KASAN: null-ptr-deref in __refcount_inc_not_zero include/linux/refcount.h:227 [inline] BUG: KASAN: null-ptr-deref in refcount_inc_not_zero include/linux/refcount.h:245 [inline] BUG: KASAN: null-ptr-deref in maybe_get_net include/net/net_namespace.h:269 [inline] BUG: KASAN: null-ptr-deref in tcf_exts_get_net include/net/pkt_cls.h:260 [inline] BUG: KASAN: null-ptr-deref in __fl_put net/sched/cls_flower.c:513 [inline] BUG: KASAN: null-ptr-deref in __fl_put+0x13e/0x3b0 net/sched/cls_flower.c:508 Read of size 4 at addr 000000000000014c by task syz-executor548/5082 CPU: 0 PID: 5082 Comm: syz-executor548 Not tainted 6.2.0-syzkaller-05251-g5b7c4cabbb65 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/21/2023 Call Trace: __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0xd9/0x150 lib/dump_stack.c:106 print_report mm/kasan/report.c:420 [inline] kasan_report+0xec/0x130 mm/kasan/report.c:517 check_region_inline mm/kasan/generic.c:183 [inline] kasan_check_range+0x141/0x190 mm/kasan/generic.c:189 instrument_atomic_read include/linux/instrumented.h:72 [inline] atomic_read include/linux/atomic/atomic-instrumented.h:27 [inline] refcount_read include/linux/refcount.h:147 [inline] __refcount_add_not_zero include/linux/refcount.h:152 [inline] __refcount_inc_not_zero include/linux/refcount.h:227 [inline] refcount_inc_not_zero include/linux/refcount.h:245 [inline] maybe_get_net include/net/net_namespace.h:269 [inline] tcf_exts_get_net include/net/pkt_cls.h:260 [inline] __fl_put net/sched/cls_flower.c:513 [inline] __fl_put+0x13e/0x3b0 net/sched/cls_flower.c:508 fl_change+0x101b/0x4ab0 net/sched/cls_flower.c:2341 tc_new_tfilter+0x97c/0x2290 net/sched/cls_api.c:2310 rtnetlink_rcv_msg+0x996/0xd50 net/core/rtnetlink.c:6165 netlink_rcv_skb+0x165/0x440 net/netlink/af_netlink.c:2574 netlink_unicast_kernel net/netlink/af_netlink.c:1339 [inline] netlink_unicast+0x547/0x7f0 net/netlink/af_netlink.c:1365 netlink_sendmsg+0x925/0xe30 net/netlink/af_netlink.c:1942 sock_sendmsg_nosec net/socket.c:722 [inline] sock_sendmsg+0xde/0x190 net/socket.c:745 ____sys_sendmsg+0x334/0x900 net/socket.c:2504 ___sys_sendmsg+0x110/0x1b0 net/socket.c:2558 __sys_sendmmsg+0x18f/0x460 net/socket.c:2644 __do_sys_sendmmsg net/socket.c:2673 [inline] __se_sys_sendmmsg net/socket.c:2670 [inline] __x64_sys_sendmmsg+0x9d/0x100 net/socket.c:2670 Fixes: 08a0063df3ae ("net/sched: flower: Move filter handle initialization earlier") Reported-by: syzbot+baabf3efa7c1e57d28b2@syzkaller.appspotmail.com Reported-by: syzbot Signed-off-by: Eric Dumazet Cc: Paul Blakey Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/cls_flower.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index e960a46b05205..475fe222a8556 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -2200,8 +2200,9 @@ static int fl_change(struct net *net, struct sk_buff *in_skb, fnew->flags = nla_get_u32(tb[TCA_FLOWER_FLAGS]); if (!tc_flags_valid(fnew->flags)) { + kfree(fnew); err = -EINVAL; - goto errout; + goto errout_tb; } } @@ -2226,8 +2227,10 @@ static int fl_change(struct net *net, struct sk_buff *in_skb, } spin_unlock(&tp->lock); - if (err) - goto errout; + if (err) { + kfree(fnew); + goto errout_tb; + } } fnew->handle = handle; @@ -2337,7 +2340,6 @@ static int fl_change(struct net *net, struct sk_buff *in_skb, fl_mask_put(head, fnew->mask); errout_idr: idr_remove(&head->handle_idr, fnew->handle); -errout: __fl_put(fnew); errout_tb: kfree(tb); From dd093fb08e8f8a958fec4eef36f9f09eac047f60 Mon Sep 17 00:00:00 2001 From: Tom Lendacky Date: Wed, 22 Feb 2023 10:39:39 -0600 Subject: [PATCH 612/765] virt/sev-guest: Return -EIO if certificate buffer is not large enough Commit 47894e0fa6a5 ("virt/sev-guest: Prevent IV reuse in the SNP guest driver") changed the behavior associated with the return value when the caller does not supply a large enough certificate buffer. Prior to the commit a value of -EIO was returned. Now, 0 is returned. This breaks the established ABI with the user. Change the code to detect the buffer size error and return -EIO. Fixes: 47894e0fa6a5 ("virt/sev-guest: Prevent IV reuse in the SNP guest driver") Reported-by: Larry Dewey Signed-off-by: Tom Lendacky Signed-off-by: Borislav Petkov (AMD) Tested-by: Larry Dewey Cc: Link: https://lore.kernel.org/r/2afbcae6daf13f7ad5a4296692e0a0fe1bc1e4ee.1677083979.git.thomas.lendacky@amd.com --- drivers/virt/coco/sev-guest/sev-guest.c | 20 +++++++++++++++++--- 1 file changed, 17 insertions(+), 3 deletions(-) diff --git a/drivers/virt/coco/sev-guest/sev-guest.c b/drivers/virt/coco/sev-guest/sev-guest.c index 4ec4174e05a3c..7b4e9009f3355 100644 --- a/drivers/virt/coco/sev-guest/sev-guest.c +++ b/drivers/virt/coco/sev-guest/sev-guest.c @@ -377,9 +377,26 @@ static int handle_guest_request(struct snp_guest_dev *snp_dev, u64 exit_code, in snp_dev->input.data_npages = certs_npages; } + /* + * Increment the message sequence number. There is no harm in doing + * this now because decryption uses the value stored in the response + * structure and any failure will wipe the VMPCK, preventing further + * use anyway. + */ + snp_inc_msg_seqno(snp_dev); + if (fw_err) *fw_err = err; + /* + * If an extended guest request was issued and the supplied certificate + * buffer was not large enough, a standard guest request was issued to + * prevent IV reuse. If the standard request was successful, return -EIO + * back to the caller as would have originally been returned. + */ + if (!rc && err == SNP_GUEST_REQ_INVALID_LEN) + return -EIO; + if (rc) { dev_alert(snp_dev->dev, "Detected error from ASP request. rc: %d, fw_err: %llu\n", @@ -395,9 +412,6 @@ static int handle_guest_request(struct snp_guest_dev *snp_dev, u64 exit_code, in goto disable_vmpck; } - /* Increment to new message sequence after payload decryption was successful. */ - snp_inc_msg_seqno(snp_dev); - return 0; disable_vmpck: From 81563d8548b0478075c720666be348d4199b8591 Mon Sep 17 00:00:00 2001 From: Horatiu Vultur Date: Tue, 28 Feb 2023 21:47:42 +0100 Subject: [PATCH 613/765] net: lan966x: Fix port police support using tc-matchall When the police was removed from the port, then it was trying to remove the police from the police id and not from the actual police index. The police id represents the id of the police and police index represents the position in HW where the police is situated. The port police id can be any number while the port police index is a number based on the port chip port. Fix this by deleting the police from HW that is situated at the police index and not police id. Fixes: 5390334b59a3 ("net: lan966x: Add port police support using tc-matchall") Signed-off-by: Horatiu Vultur Reviewed-by: Simon Horman Reviewed-by: Vladimir Oltean Signed-off-by: David S. Miller --- drivers/net/ethernet/microchip/lan966x/lan966x_police.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_police.c b/drivers/net/ethernet/microchip/lan966x/lan966x_police.c index a9aec900d608d..7d66fe75cd3bf 100644 --- a/drivers/net/ethernet/microchip/lan966x/lan966x_police.c +++ b/drivers/net/ethernet/microchip/lan966x/lan966x_police.c @@ -194,7 +194,7 @@ int lan966x_police_port_del(struct lan966x_port *port, return -EINVAL; } - err = lan966x_police_del(port, port->tc.police_id); + err = lan966x_police_del(port, POL_IDX_PORT + port->chip_port); if (err) { NL_SET_ERR_MSG_MOD(extack, "Failed to add policer to port"); From 23badca4248a9a467f630adbd1557f664585e1db Mon Sep 17 00:00:00 2001 From: Trevor Wu Date: Wed, 1 Mar 2023 19:01:59 +0800 Subject: [PATCH 614/765] ASoC: mediatek: mt8188: add missing initialization In etdm dai driver, dai_etdm_parse_of() function is used to parse dts properties to get parameters. There are two for-loops which are sepearately for all etdm and etdm input only cases. In etdm in only loop, dai_id is not initialized, so it keeps the value intiliazed in another loop. In the patch, add the missing initialization to fix the unexpected parsing problem. Fixes: 2babb4777489 ("ASoC: mediatek: mt8188: support etdm in platform driver") Signed-off-by: Trevor Wu Reviewed-by: AngeloGioacchino Del Regno Link: https://lore.kernel.org/r/20230301110200.26177-2-trevor.wu@mediatek.com Signed-off-by: Mark Brown --- sound/soc/mediatek/mt8188/mt8188-dai-etdm.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/sound/soc/mediatek/mt8188/mt8188-dai-etdm.c b/sound/soc/mediatek/mt8188/mt8188-dai-etdm.c index 1c53d4cb19bbe..7a37752d42444 100644 --- a/sound/soc/mediatek/mt8188/mt8188-dai-etdm.c +++ b/sound/soc/mediatek/mt8188/mt8188-dai-etdm.c @@ -2498,6 +2498,9 @@ static void mt8188_dai_etdm_parse_of(struct mtk_base_afe *afe) /* etdm in only */ for (i = 0; i < 2; i++) { + dai_id = ETDM_TO_DAI_ID(i); + etdm_data = afe_priv->dai_priv[dai_id]; + snprintf(prop, sizeof(prop), "mediatek,%s-chn-disabled", of_afe_etdms[i].name); From b56ec2992a2e43bc3e60d6db86849d31640e791f Mon Sep 17 00:00:00 2001 From: Trevor Wu Date: Wed, 1 Mar 2023 19:02:00 +0800 Subject: [PATCH 615/765] ASoC: mediatek: mt8195: add missing initialization In etdm dai driver, dai_etdm_parse_of() function is used to parse dts properties to get parameters. There are two for-loops which are sepearately for all etdm and etdm input only cases. In etdm in only loop, dai_id is not initialized, so it keeps the value intiliazed in another loop. In the patch, add the missing initialization to fix the unexpected parsing problem. Fixes: 1de9a54acafb ("ASoC: mediatek: mt8195: support etdm in platform driver") Signed-off-by: Trevor Wu Reviewed-by: AngeloGioacchino Del Regno Link: https://lore.kernel.org/r/20230301110200.26177-3-trevor.wu@mediatek.com Signed-off-by: Mark Brown --- sound/soc/mediatek/mt8195/mt8195-dai-etdm.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/sound/soc/mediatek/mt8195/mt8195-dai-etdm.c b/sound/soc/mediatek/mt8195/mt8195-dai-etdm.c index c2e268054773d..f2c9a1fdbe0d0 100644 --- a/sound/soc/mediatek/mt8195/mt8195-dai-etdm.c +++ b/sound/soc/mediatek/mt8195/mt8195-dai-etdm.c @@ -2567,6 +2567,9 @@ static void mt8195_dai_etdm_parse_of(struct mtk_base_afe *afe) /* etdm in only */ for (i = 0; i < 2; i++) { + dai_id = ETDM_TO_DAI_ID(i); + etdm_data = afe_priv->dai_priv[dai_id]; + ret = snprintf(prop, sizeof(prop), "mediatek,%s-chn-disabled", of_afe_etdms[i].name); From 2067e7a00aa604b94de31d64f29b8893b1696f26 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Mon, 27 Feb 2023 17:36:46 +0800 Subject: [PATCH 616/765] selftests: nft_nat: ensuring the listening side is up before starting the client The test_local_dnat_portonly() function initiates the client-side as soon as it sets the listening side to the background. This could lead to a race condition where the server may not be ready to listen. To ensure that the server-side is up and running before initiating the client-side, a delay is introduced to the test_local_dnat_portonly() function. Before the fix: # ./nft_nat.sh PASS: netns routing/connectivity: ns0-rthlYrBU can reach ns1-rthlYrBU and ns2-rthlYrBU PASS: ping to ns1-rthlYrBU was ip NATted to ns2-rthlYrBU PASS: ping to ns1-rthlYrBU OK after ip nat output chain flush PASS: ipv6 ping to ns1-rthlYrBU was ip6 NATted to ns2-rthlYrBU 2023/02/27 04:11:03 socat[6055] E connect(5, AF=2 10.0.1.99:2000, 16): Connection refused ERROR: inet port rewrite After the fix: # ./nft_nat.sh PASS: netns routing/connectivity: ns0-9sPJV6JJ can reach ns1-9sPJV6JJ and ns2-9sPJV6JJ PASS: ping to ns1-9sPJV6JJ was ip NATted to ns2-9sPJV6JJ PASS: ping to ns1-9sPJV6JJ OK after ip nat output chain flush PASS: ipv6 ping to ns1-9sPJV6JJ was ip6 NATted to ns2-9sPJV6JJ PASS: inet port rewrite without l3 address Fixes: 282e5f8fe907 ("netfilter: nat: really support inet nat without l3 address") Signed-off-by: Hangbin Liu Signed-off-by: Pablo Neira Ayuso --- tools/testing/selftests/netfilter/nft_nat.sh | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tools/testing/selftests/netfilter/nft_nat.sh b/tools/testing/selftests/netfilter/nft_nat.sh index 924ecb3f1f737..dd40d9f6f2599 100755 --- a/tools/testing/selftests/netfilter/nft_nat.sh +++ b/tools/testing/selftests/netfilter/nft_nat.sh @@ -404,6 +404,8 @@ EOF echo SERVER-$family | ip netns exec "$ns1" timeout 5 socat -u STDIN TCP-LISTEN:2000 & sc_s=$! + sleep 1 + result=$(ip netns exec "$ns0" timeout 1 socat TCP:$daddr:2000 STDOUT) if [ "$result" = "SERVER-inet" ];then From 860e874290fb3be08e966c9c8ffc510c5b0f2bd8 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 28 Feb 2023 17:09:03 +0100 Subject: [PATCH 617/765] netfilter: nft_last: copy content when cloning expression If the ruleset contains last timestamps, restore them accordingly. Otherwise, listing after restoration shows never used items. Fixes: 33a24de37e81 ("netfilter: nft_last: move stateful fields out of expression data") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_last.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/netfilter/nft_last.c b/net/netfilter/nft_last.c index 7f2bda6641bd8..8e6d7eaf9dc8b 100644 --- a/net/netfilter/nft_last.c +++ b/net/netfilter/nft_last.c @@ -105,11 +105,15 @@ static void nft_last_destroy(const struct nft_ctx *ctx, static int nft_last_clone(struct nft_expr *dst, const struct nft_expr *src) { struct nft_last_priv *priv_dst = nft_expr_priv(dst); + struct nft_last_priv *priv_src = nft_expr_priv(src); priv_dst->last = kzalloc(sizeof(*priv_dst->last), GFP_ATOMIC); if (!priv_dst->last) return -ENOMEM; + priv_dst->last->set = priv_src->last->set; + priv_dst->last->jiffies = priv_src->last->jiffies; + return 0; } From aabef97a35160461e9c576848ded737558d89055 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 28 Feb 2023 20:43:02 +0100 Subject: [PATCH 618/765] netfilter: nft_quota: copy content when cloning expression If the ruleset contains consumed quota, restore them accordingly. Otherwise, listing after restoration shows never used items. Restore the user-defined quota and flags too. Fixes: ed0a0c60f0e5 ("netfilter: nft_quota: move stateful fields out of expression data") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_quota.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/net/netfilter/nft_quota.c b/net/netfilter/nft_quota.c index 123578e289179..3ba12a7471b0f 100644 --- a/net/netfilter/nft_quota.c +++ b/net/netfilter/nft_quota.c @@ -236,12 +236,16 @@ static void nft_quota_destroy(const struct nft_ctx *ctx, static int nft_quota_clone(struct nft_expr *dst, const struct nft_expr *src) { struct nft_quota *priv_dst = nft_expr_priv(dst); + struct nft_quota *priv_src = nft_expr_priv(src); + + priv_dst->quota = priv_src->quota; + priv_dst->flags = priv_src->flags; priv_dst->consumed = kmalloc(sizeof(*priv_dst->consumed), GFP_ATOMIC); if (!priv_dst->consumed) return -ENOMEM; - atomic64_set(priv_dst->consumed, 0); + *priv_dst->consumed = *priv_src->consumed; return 0; } From 1947ddf9b3d5b886ba227bbfd3d6f501af08b5b0 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 27 Feb 2023 09:41:20 -0700 Subject: [PATCH 619/765] io_uring/poll: don't pass in wake func to io_init_poll_iocb() We only use one, and it's io_poll_wake(). Hardwire that in the initial init, as well as in __io_queue_proc() if we're setting up for double poll. Signed-off-by: Jens Axboe --- io_uring/poll.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/io_uring/poll.c b/io_uring/poll.c index a82d6830bdfdd..795facbd0e9f1 100644 --- a/io_uring/poll.c +++ b/io_uring/poll.c @@ -51,6 +51,9 @@ struct io_poll_table { #define IO_WQE_F_DOUBLE 1 +static int io_poll_wake(struct wait_queue_entry *wait, unsigned mode, int sync, + void *key); + static inline struct io_kiocb *wqe_to_req(struct wait_queue_entry *wqe) { unsigned long priv = (unsigned long)wqe->private; @@ -164,15 +167,14 @@ static void io_poll_tw_hash_eject(struct io_kiocb *req, bool *locked) } } -static void io_init_poll_iocb(struct io_poll *poll, __poll_t events, - wait_queue_func_t wake_func) +static void io_init_poll_iocb(struct io_poll *poll, __poll_t events) { poll->head = NULL; #define IO_POLL_UNMASK (EPOLLERR|EPOLLHUP|EPOLLNVAL|EPOLLRDHUP) /* mask in events that we always want/need */ poll->events = events | IO_POLL_UNMASK; INIT_LIST_HEAD(&poll->wait.entry); - init_waitqueue_func_entry(&poll->wait, wake_func); + init_waitqueue_func_entry(&poll->wait, io_poll_wake); } static inline void io_poll_remove_entry(struct io_poll *poll) @@ -508,7 +510,7 @@ static void __io_queue_proc(struct io_poll *poll, struct io_poll_table *pt, /* mark as double wq entry */ wqe_private |= IO_WQE_F_DOUBLE; - io_init_poll_iocb(poll, first->events, first->wait.func); + io_init_poll_iocb(poll, first->events); if (!io_poll_double_prepare(req)) { /* the request is completing, just back off */ kfree(poll); @@ -569,7 +571,7 @@ static int __io_arm_poll_handler(struct io_kiocb *req, INIT_HLIST_NODE(&req->hash_node); req->work.cancel_seq = atomic_read(&ctx->cancel_seq); - io_init_poll_iocb(poll, mask, io_poll_wake); + io_init_poll_iocb(poll, mask); poll->file = req->file; req->apoll_events = poll->events; From f122a08b197d076ccf136c73fae0146875812a88 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 28 Feb 2023 11:39:09 -0800 Subject: [PATCH 620/765] capability: just use a 'u64' instead of a 'u32[2]' array Back in 2008 we extended the capability bits from 32 to 64, and we did it by extending the single 32-bit capability word from one word to an array of two words. It was then obfuscated by hiding the "2" behind two macro expansions, with the reasoning being that maybe it gets extended further some day. That reasoning may have been valid at the time, but the last thing we want to do is to extend the capability set any more. And the array of values not only causes source code oddities (with loops to deal with it), but also results in worse code generation. It's a lose-lose situation. So just change the 'u32[2]' into a 'u64' and be done with it. We still have to deal with the fact that the user space interface is designed around an array of these 32-bit values, but that was the case before too, since the array layouts were different (ie user space doesn't use an array of 32-bit values for individual capability masks, but an array of 32-bit slices of multiple masks). So that marshalling of data is actually simplified too, even if it does remain somewhat obscure and odd. This was all triggered by my reaction to the new "cap_isidentical()" introduced recently. By just using a saner data structure, it went from unsigned __capi; CAP_FOR_EACH_U32(__capi) { if (a.cap[__capi] != b.cap[__capi]) return false; } return true; to just being return a.val == b.val; instead. Which is rather more obvious both to humans and to compilers. Cc: Mateusz Guzik Cc: Casey Schaufler Cc: Serge Hallyn Cc: Al Viro Cc: Paul Moore Signed-off-by: Linus Torvalds --- fs/proc/array.c | 7 +- include/linux/capability.h | 131 ++++-------------- io_uring/fdinfo.c | 4 +- kernel/auditsc.c | 6 +- kernel/capability.c | 104 ++++++-------- kernel/umh.c | 41 +++--- security/apparmor/policy_unpack.c | 40 ++++-- security/commoncap.c | 49 +++---- .../selftests/bpf/progs/test_deny_namespace.c | 7 +- 9 files changed, 150 insertions(+), 239 deletions(-) diff --git a/fs/proc/array.c b/fs/proc/array.c index 49283b8103c7e..9b0315d34c58f 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -300,13 +300,8 @@ static inline void task_sig(struct seq_file *m, struct task_struct *p) static void render_cap_t(struct seq_file *m, const char *header, kernel_cap_t *a) { - unsigned __capi; - seq_puts(m, header); - CAP_FOR_EACH_U32(__capi) { - seq_put_hex_ll(m, NULL, - a->cap[CAP_LAST_U32 - __capi], 8); - } + seq_put_hex_ll(m, NULL, a->val, 16); seq_putc(m, '\n'); } diff --git a/include/linux/capability.h b/include/linux/capability.h index d3c6c2d1ff452..0c356a5179917 100644 --- a/include/linux/capability.h +++ b/include/linux/capability.h @@ -15,28 +15,25 @@ #include #include +#include #define _KERNEL_CAPABILITY_VERSION _LINUX_CAPABILITY_VERSION_3 -#define _KERNEL_CAPABILITY_U32S _LINUX_CAPABILITY_U32S_3 extern int file_caps_enabled; -typedef struct kernel_cap_struct { - __u32 cap[_KERNEL_CAPABILITY_U32S]; -} kernel_cap_t; +typedef struct { u64 val; } kernel_cap_t; /* same as vfs_ns_cap_data but in cpu endian and always filled completely */ struct cpu_vfs_cap_data { __u32 magic_etc; + kuid_t rootid; kernel_cap_t permitted; kernel_cap_t inheritable; - kuid_t rootid; }; #define _USER_CAP_HEADER_SIZE (sizeof(struct __user_cap_header_struct)) #define _KERNEL_CAP_T_SIZE (sizeof(kernel_cap_t)) - struct file; struct inode; struct dentry; @@ -44,16 +41,6 @@ struct task_struct; struct user_namespace; struct mnt_idmap; -extern const kernel_cap_t __cap_empty_set; -extern const kernel_cap_t __cap_init_eff_set; - -/* - * Internal kernel functions only - */ - -#define CAP_FOR_EACH_U32(__capi) \ - for (__capi = 0; __capi < _KERNEL_CAPABILITY_U32S; ++__capi) - /* * CAP_FS_MASK and CAP_NFSD_MASKS: * @@ -67,104 +54,52 @@ extern const kernel_cap_t __cap_init_eff_set; * 2. The security.* and trusted.* xattrs are fs-related MAC permissions */ -# define CAP_FS_MASK_B0 (CAP_TO_MASK(CAP_CHOWN) \ - | CAP_TO_MASK(CAP_MKNOD) \ - | CAP_TO_MASK(CAP_DAC_OVERRIDE) \ - | CAP_TO_MASK(CAP_DAC_READ_SEARCH) \ - | CAP_TO_MASK(CAP_FOWNER) \ - | CAP_TO_MASK(CAP_FSETID)) - -# define CAP_FS_MASK_B1 (CAP_TO_MASK(CAP_MAC_OVERRIDE)) - -#if _KERNEL_CAPABILITY_U32S != 2 -# error Fix up hand-coded capability macro initializers -#else /* HAND-CODED capability initializers */ +# define CAP_FS_MASK (BIT_ULL(CAP_CHOWN) \ + | BIT_ULL(CAP_MKNOD) \ + | BIT_ULL(CAP_DAC_OVERRIDE) \ + | BIT_ULL(CAP_DAC_READ_SEARCH) \ + | BIT_ULL(CAP_FOWNER) \ + | BIT_ULL(CAP_FSETID) \ + | BIT_ULL(CAP_MAC_OVERRIDE)) +#define CAP_VALID_MASK (BIT_ULL(CAP_LAST_CAP+1)-1) -#define CAP_LAST_U32 ((_KERNEL_CAPABILITY_U32S) - 1) -#define CAP_LAST_U32_VALID_MASK (CAP_TO_MASK(CAP_LAST_CAP + 1) -1) +# define CAP_EMPTY_SET ((kernel_cap_t) { 0 }) +# define CAP_FULL_SET ((kernel_cap_t) { CAP_VALID_MASK }) +# define CAP_FS_SET ((kernel_cap_t) { CAP_FS_MASK | BIT_ULL(CAP_LINUX_IMMUTABLE) }) +# define CAP_NFSD_SET ((kernel_cap_t) { CAP_FS_MASK | BIT_ULL(CAP_SYS_RESOURCE) }) -# define CAP_EMPTY_SET ((kernel_cap_t){{ 0, 0 }}) -# define CAP_FULL_SET ((kernel_cap_t){{ ~0, CAP_LAST_U32_VALID_MASK }}) -# define CAP_FS_SET ((kernel_cap_t){{ CAP_FS_MASK_B0 \ - | CAP_TO_MASK(CAP_LINUX_IMMUTABLE), \ - CAP_FS_MASK_B1 } }) -# define CAP_NFSD_SET ((kernel_cap_t){{ CAP_FS_MASK_B0 \ - | CAP_TO_MASK(CAP_SYS_RESOURCE), \ - CAP_FS_MASK_B1 } }) +# define cap_clear(c) do { (c).val = 0; } while (0) -#endif /* _KERNEL_CAPABILITY_U32S != 2 */ - -# define cap_clear(c) do { (c) = __cap_empty_set; } while (0) - -#define cap_raise(c, flag) ((c).cap[CAP_TO_INDEX(flag)] |= CAP_TO_MASK(flag)) -#define cap_lower(c, flag) ((c).cap[CAP_TO_INDEX(flag)] &= ~CAP_TO_MASK(flag)) -#define cap_raised(c, flag) ((c).cap[CAP_TO_INDEX(flag)] & CAP_TO_MASK(flag)) - -#define CAP_BOP_ALL(c, a, b, OP) \ -do { \ - unsigned __capi; \ - CAP_FOR_EACH_U32(__capi) { \ - c.cap[__capi] = a.cap[__capi] OP b.cap[__capi]; \ - } \ -} while (0) - -#define CAP_UOP_ALL(c, a, OP) \ -do { \ - unsigned __capi; \ - CAP_FOR_EACH_U32(__capi) { \ - c.cap[__capi] = OP a.cap[__capi]; \ - } \ -} while (0) +#define cap_raise(c, flag) ((c).val |= BIT_ULL(flag)) +#define cap_lower(c, flag) ((c).val &= ~BIT_ULL(flag)) +#define cap_raised(c, flag) (((c).val & BIT_ULL(flag)) != 0) static inline kernel_cap_t cap_combine(const kernel_cap_t a, const kernel_cap_t b) { - kernel_cap_t dest; - CAP_BOP_ALL(dest, a, b, |); - return dest; + return (kernel_cap_t) { a.val | b.val }; } static inline kernel_cap_t cap_intersect(const kernel_cap_t a, const kernel_cap_t b) { - kernel_cap_t dest; - CAP_BOP_ALL(dest, a, b, &); - return dest; + return (kernel_cap_t) { a.val & b.val }; } static inline kernel_cap_t cap_drop(const kernel_cap_t a, const kernel_cap_t drop) { - kernel_cap_t dest; - CAP_BOP_ALL(dest, a, drop, &~); - return dest; -} - -static inline kernel_cap_t cap_invert(const kernel_cap_t c) -{ - kernel_cap_t dest; - CAP_UOP_ALL(dest, c, ~); - return dest; + return (kernel_cap_t) { a.val &~ drop.val }; } static inline bool cap_isclear(const kernel_cap_t a) { - unsigned __capi; - CAP_FOR_EACH_U32(__capi) { - if (a.cap[__capi] != 0) - return false; - } - return true; + return !a.val; } static inline bool cap_isidentical(const kernel_cap_t a, const kernel_cap_t b) { - unsigned __capi; - CAP_FOR_EACH_U32(__capi) { - if (a.cap[__capi] != b.cap[__capi]) - return false; - } - return true; + return a.val == b.val; } /* @@ -176,39 +111,31 @@ static inline bool cap_isidentical(const kernel_cap_t a, const kernel_cap_t b) */ static inline bool cap_issubset(const kernel_cap_t a, const kernel_cap_t set) { - kernel_cap_t dest; - dest = cap_drop(a, set); - return cap_isclear(dest); + return !(a.val & ~set.val); } /* Used to decide between falling back on the old suser() or fsuser(). */ static inline kernel_cap_t cap_drop_fs_set(const kernel_cap_t a) { - const kernel_cap_t __cap_fs_set = CAP_FS_SET; - return cap_drop(a, __cap_fs_set); + return cap_drop(a, CAP_FS_SET); } static inline kernel_cap_t cap_raise_fs_set(const kernel_cap_t a, const kernel_cap_t permitted) { - const kernel_cap_t __cap_fs_set = CAP_FS_SET; - return cap_combine(a, - cap_intersect(permitted, __cap_fs_set)); + return cap_combine(a, cap_intersect(permitted, CAP_FS_SET)); } static inline kernel_cap_t cap_drop_nfsd_set(const kernel_cap_t a) { - const kernel_cap_t __cap_fs_set = CAP_NFSD_SET; - return cap_drop(a, __cap_fs_set); + return cap_drop(a, CAP_NFSD_SET); } static inline kernel_cap_t cap_raise_nfsd_set(const kernel_cap_t a, const kernel_cap_t permitted) { - const kernel_cap_t __cap_nfsd_set = CAP_NFSD_SET; - return cap_combine(a, - cap_intersect(permitted, __cap_nfsd_set)); + return cap_combine(a, cap_intersect(permitted, CAP_NFSD_SET)); } #ifdef CONFIG_MULTIUSER diff --git a/io_uring/fdinfo.c b/io_uring/fdinfo.c index 882bd56b01ed0..76c279b13aee4 100644 --- a/io_uring/fdinfo.c +++ b/io_uring/fdinfo.c @@ -22,7 +22,6 @@ static __cold int io_uring_show_cred(struct seq_file *m, unsigned int id, struct user_namespace *uns = seq_user_ns(m); struct group_info *gi; kernel_cap_t cap; - unsigned __capi; int g; seq_printf(m, "%5d\n", id); @@ -42,8 +41,7 @@ static __cold int io_uring_show_cred(struct seq_file *m, unsigned int id, } seq_puts(m, "\n\tCapEff:\t"); cap = cred->cap_effective; - CAP_FOR_EACH_U32(__capi) - seq_put_hex_ll(m, NULL, cap.cap[CAP_LAST_U32 - __capi], 8); + seq_put_hex_ll(m, NULL, cap.val, 16); seq_putc(m, '\n'); return 0; } diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 93d0b87f32838..addeed3df15d3 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -1295,15 +1295,11 @@ static void audit_log_execve_info(struct audit_context *context, static void audit_log_cap(struct audit_buffer *ab, char *prefix, kernel_cap_t *cap) { - int i; - if (cap_isclear(*cap)) { audit_log_format(ab, " %s=0", prefix); return; } - audit_log_format(ab, " %s=", prefix); - CAP_FOR_EACH_U32(i) - audit_log_format(ab, "%08x", cap->cap[CAP_LAST_U32 - i]); + audit_log_format(ab, " %s=%016llx", prefix, cap->val); } static void audit_log_fcaps(struct audit_buffer *ab, struct audit_names *name) diff --git a/kernel/capability.c b/kernel/capability.c index 339a44dfe2f46..3e058f41df32d 100644 --- a/kernel/capability.c +++ b/kernel/capability.c @@ -20,13 +20,6 @@ #include #include -/* - * Leveraged for setting/resetting capabilities - */ - -const kernel_cap_t __cap_empty_set = CAP_EMPTY_SET; -EXPORT_SYMBOL(__cap_empty_set); - int file_caps_enabled = 1; static int __init file_caps_disable(char *str) @@ -151,6 +144,7 @@ SYSCALL_DEFINE2(capget, cap_user_header_t, header, cap_user_data_t, dataptr) pid_t pid; unsigned tocopy; kernel_cap_t pE, pI, pP; + struct __user_cap_data_struct kdata[2]; ret = cap_validate_magic(header, &tocopy); if ((dataptr == NULL) || (ret != 0)) @@ -163,42 +157,46 @@ SYSCALL_DEFINE2(capget, cap_user_header_t, header, cap_user_data_t, dataptr) return -EINVAL; ret = cap_get_target_pid(pid, &pE, &pI, &pP); - if (!ret) { - struct __user_cap_data_struct kdata[_KERNEL_CAPABILITY_U32S]; - unsigned i; - - for (i = 0; i < tocopy; i++) { - kdata[i].effective = pE.cap[i]; - kdata[i].permitted = pP.cap[i]; - kdata[i].inheritable = pI.cap[i]; - } - - /* - * Note, in the case, tocopy < _KERNEL_CAPABILITY_U32S, - * we silently drop the upper capabilities here. This - * has the effect of making older libcap - * implementations implicitly drop upper capability - * bits when they perform a: capget/modify/capset - * sequence. - * - * This behavior is considered fail-safe - * behavior. Upgrading the application to a newer - * version of libcap will enable access to the newer - * capabilities. - * - * An alternative would be to return an error here - * (-ERANGE), but that causes legacy applications to - * unexpectedly fail; the capget/modify/capset aborts - * before modification is attempted and the application - * fails. - */ - if (copy_to_user(dataptr, kdata, tocopy - * sizeof(struct __user_cap_data_struct))) { - return -EFAULT; - } - } + if (ret) + return ret; - return ret; + /* + * Annoying legacy format with 64-bit capabilities exposed + * as two sets of 32-bit fields, so we need to split the + * capability values up. + */ + kdata[0].effective = pE.val; kdata[1].effective = pE.val >> 32; + kdata[0].permitted = pP.val; kdata[1].permitted = pP.val >> 32; + kdata[0].inheritable = pI.val; kdata[1].inheritable = pI.val >> 32; + + /* + * Note, in the case, tocopy < _KERNEL_CAPABILITY_U32S, + * we silently drop the upper capabilities here. This + * has the effect of making older libcap + * implementations implicitly drop upper capability + * bits when they perform a: capget/modify/capset + * sequence. + * + * This behavior is considered fail-safe + * behavior. Upgrading the application to a newer + * version of libcap will enable access to the newer + * capabilities. + * + * An alternative would be to return an error here + * (-ERANGE), but that causes legacy applications to + * unexpectedly fail; the capget/modify/capset aborts + * before modification is attempted and the application + * fails. + */ + if (copy_to_user(dataptr, kdata, tocopy * sizeof(kdata[0]))) + return -EFAULT; + + return 0; +} + +static kernel_cap_t mk_kernel_cap(u32 low, u32 high) +{ + return (kernel_cap_t) { (low | ((u64)high << 32)) & CAP_VALID_MASK }; } /** @@ -221,8 +219,8 @@ SYSCALL_DEFINE2(capget, cap_user_header_t, header, cap_user_data_t, dataptr) */ SYSCALL_DEFINE2(capset, cap_user_header_t, header, const cap_user_data_t, data) { - struct __user_cap_data_struct kdata[_KERNEL_CAPABILITY_U32S]; - unsigned i, tocopy, copybytes; + struct __user_cap_data_struct kdata[2] = { { 0, }, }; + unsigned tocopy, copybytes; kernel_cap_t inheritable, permitted, effective; struct cred *new; int ret; @@ -246,21 +244,9 @@ SYSCALL_DEFINE2(capset, cap_user_header_t, header, const cap_user_data_t, data) if (copy_from_user(&kdata, data, copybytes)) return -EFAULT; - for (i = 0; i < tocopy; i++) { - effective.cap[i] = kdata[i].effective; - permitted.cap[i] = kdata[i].permitted; - inheritable.cap[i] = kdata[i].inheritable; - } - while (i < _KERNEL_CAPABILITY_U32S) { - effective.cap[i] = 0; - permitted.cap[i] = 0; - inheritable.cap[i] = 0; - i++; - } - - effective.cap[CAP_LAST_U32] &= CAP_LAST_U32_VALID_MASK; - permitted.cap[CAP_LAST_U32] &= CAP_LAST_U32_VALID_MASK; - inheritable.cap[CAP_LAST_U32] &= CAP_LAST_U32_VALID_MASK; + effective = mk_kernel_cap(kdata[0].effective, kdata[1].effective); + permitted = mk_kernel_cap(kdata[0].permitted, kdata[1].permitted); + inheritable = mk_kernel_cap(kdata[0].inheritable, kdata[1].inheritable); new = prepare_creds(); if (!new) diff --git a/kernel/umh.c b/kernel/umh.c index fbf872c624cbc..2a47082773352 100644 --- a/kernel/umh.c +++ b/kernel/umh.c @@ -501,9 +501,9 @@ static int proc_cap_handler(struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos) { struct ctl_table t; - unsigned long cap_array[_KERNEL_CAPABILITY_U32S]; - kernel_cap_t new_cap; - int err, i; + unsigned long cap_array[2]; + kernel_cap_t new_cap, *cap; + int err; if (write && (!capable(CAP_SETPCAP) || !capable(CAP_SYS_MODULE))) @@ -514,14 +514,16 @@ static int proc_cap_handler(struct ctl_table *table, int write, * userspace if this is a read. */ spin_lock(&umh_sysctl_lock); - for (i = 0; i < _KERNEL_CAPABILITY_U32S; i++) { - if (table->data == CAP_BSET) - cap_array[i] = usermodehelper_bset.cap[i]; - else if (table->data == CAP_PI) - cap_array[i] = usermodehelper_inheritable.cap[i]; - else - BUG(); - } + if (table->data == CAP_BSET) + cap = &usermodehelper_bset; + else if (table->data == CAP_PI) + cap = &usermodehelper_inheritable; + else + BUG(); + + /* Legacy format: capabilities are exposed as two 32-bit values */ + cap_array[0] = (u32) cap->val; + cap_array[1] = cap->val >> 32; spin_unlock(&umh_sysctl_lock); t = *table; @@ -535,22 +537,15 @@ static int proc_cap_handler(struct ctl_table *table, int write, if (err < 0) return err; - /* - * convert from the sysctl array of ulongs to the kernel_cap_t - * internal representation - */ - for (i = 0; i < _KERNEL_CAPABILITY_U32S; i++) - new_cap.cap[i] = cap_array[i]; + new_cap.val = (u32)cap_array[0]; + new_cap.val += (u64)cap_array[1] << 32; /* * Drop everything not in the new_cap (but don't add things) */ if (write) { spin_lock(&umh_sysctl_lock); - if (table->data == CAP_BSET) - usermodehelper_bset = cap_intersect(usermodehelper_bset, new_cap); - if (table->data == CAP_PI) - usermodehelper_inheritable = cap_intersect(usermodehelper_inheritable, new_cap); + *cap = cap_intersect(*cap, new_cap); spin_unlock(&umh_sysctl_lock); } @@ -561,14 +556,14 @@ struct ctl_table usermodehelper_table[] = { { .procname = "bset", .data = CAP_BSET, - .maxlen = _KERNEL_CAPABILITY_U32S * sizeof(unsigned long), + .maxlen = 2 * sizeof(unsigned long), .mode = 0600, .proc_handler = proc_cap_handler, }, { .procname = "inheritable", .data = CAP_PI, - .maxlen = _KERNEL_CAPABILITY_U32S * sizeof(unsigned long), + .maxlen = 2 * sizeof(unsigned long), .mode = 0600, .proc_handler = proc_cap_handler, }, diff --git a/security/apparmor/policy_unpack.c b/security/apparmor/policy_unpack.c index 5e9949832af62..cf2ceec40b28a 100644 --- a/security/apparmor/policy_unpack.c +++ b/security/apparmor/policy_unpack.c @@ -304,6 +304,26 @@ VISIBLE_IF_KUNIT bool aa_unpack_u64(struct aa_ext *e, u64 *data, const char *nam } EXPORT_SYMBOL_IF_KUNIT(aa_unpack_u64); +static bool aa_unpack_cap_low(struct aa_ext *e, kernel_cap_t *data, const char *name) +{ + u32 val; + + if (!aa_unpack_u32(e, &val, name)) + return false; + data->val = val; + return true; +} + +static bool aa_unpack_cap_high(struct aa_ext *e, kernel_cap_t *data, const char *name) +{ + u32 val; + + if (!aa_unpack_u32(e, &val, name)) + return false; + data->val = (u32)data->val | ((u64)val << 32); + return true; +} + VISIBLE_IF_KUNIT bool aa_unpack_array(struct aa_ext *e, const char *name, u16 *size) { void *pos = e->pos; @@ -897,25 +917,25 @@ static struct aa_profile *unpack_profile(struct aa_ext *e, char **ns_name) profile->path_flags = PATH_MEDIATE_DELETED; info = "failed to unpack profile capabilities"; - if (!aa_unpack_u32(e, &(rules->caps.allow.cap[0]), NULL)) + if (!aa_unpack_cap_low(e, &rules->caps.allow, NULL)) goto fail; - if (!aa_unpack_u32(e, &(rules->caps.audit.cap[0]), NULL)) + if (!aa_unpack_cap_low(e, &rules->caps.audit, NULL)) goto fail; - if (!aa_unpack_u32(e, &(rules->caps.quiet.cap[0]), NULL)) + if (!aa_unpack_cap_low(e, &rules->caps.quiet, NULL)) goto fail; - if (!aa_unpack_u32(e, &tmpcap.cap[0], NULL)) + if (!aa_unpack_cap_low(e, &tmpcap, NULL)) goto fail; info = "failed to unpack upper profile capabilities"; if (aa_unpack_nameX(e, AA_STRUCT, "caps64")) { /* optional upper half of 64 bit caps */ - if (!aa_unpack_u32(e, &(rules->caps.allow.cap[1]), NULL)) + if (!aa_unpack_cap_high(e, &rules->caps.allow, NULL)) goto fail; - if (!aa_unpack_u32(e, &(rules->caps.audit.cap[1]), NULL)) + if (!aa_unpack_cap_high(e, &rules->caps.audit, NULL)) goto fail; - if (!aa_unpack_u32(e, &(rules->caps.quiet.cap[1]), NULL)) + if (!aa_unpack_cap_high(e, &rules->caps.quiet, NULL)) goto fail; - if (!aa_unpack_u32(e, &(tmpcap.cap[1]), NULL)) + if (!aa_unpack_cap_high(e, &tmpcap, NULL)) goto fail; if (!aa_unpack_nameX(e, AA_STRUCTEND, NULL)) goto fail; @@ -924,9 +944,9 @@ static struct aa_profile *unpack_profile(struct aa_ext *e, char **ns_name) info = "failed to unpack extended profile capabilities"; if (aa_unpack_nameX(e, AA_STRUCT, "capsx")) { /* optional extended caps mediation mask */ - if (!aa_unpack_u32(e, &(rules->caps.extended.cap[0]), NULL)) + if (!aa_unpack_cap_low(e, &rules->caps.extended, NULL)) goto fail; - if (!aa_unpack_u32(e, &(rules->caps.extended.cap[1]), NULL)) + if (!aa_unpack_cap_high(e, &rules->caps.extended, NULL)) goto fail; if (!aa_unpack_nameX(e, AA_STRUCTEND, NULL)) goto fail; diff --git a/security/commoncap.c b/security/commoncap.c index aec62db552710..5bb7d1e962772 100644 --- a/security/commoncap.c +++ b/security/commoncap.c @@ -589,7 +589,6 @@ static inline int bprm_caps_from_vfs_caps(struct cpu_vfs_cap_data *caps, bool *has_fcap) { struct cred *new = bprm->cred; - unsigned i; int ret = 0; if (caps->magic_etc & VFS_CAP_FLAGS_EFFECTIVE) @@ -598,22 +597,17 @@ static inline int bprm_caps_from_vfs_caps(struct cpu_vfs_cap_data *caps, if (caps->magic_etc & VFS_CAP_REVISION_MASK) *has_fcap = true; - CAP_FOR_EACH_U32(i) { - __u32 permitted = caps->permitted.cap[i]; - __u32 inheritable = caps->inheritable.cap[i]; - - /* - * pP' = (X & fP) | (pI & fI) - * The addition of pA' is handled later. - */ - new->cap_permitted.cap[i] = - (new->cap_bset.cap[i] & permitted) | - (new->cap_inheritable.cap[i] & inheritable); + /* + * pP' = (X & fP) | (pI & fI) + * The addition of pA' is handled later. + */ + new->cap_permitted.val = + (new->cap_bset.val & caps->permitted.val) | + (new->cap_inheritable.val & caps->inheritable.val); - if (permitted & ~new->cap_permitted.cap[i]) - /* insufficient to execute correctly */ - ret = -EPERM; - } + if (caps->permitted.val & ~new->cap_permitted.val) + /* insufficient to execute correctly */ + ret = -EPERM; /* * For legacy apps, with no internal support for recognizing they @@ -644,7 +638,6 @@ int get_vfs_caps_from_disk(struct mnt_idmap *idmap, { struct inode *inode = d_backing_inode(dentry); __u32 magic_etc; - unsigned tocopy, i; int size; struct vfs_ns_cap_data data, *nscaps = &data; struct vfs_cap_data *caps = (struct vfs_cap_data *) &data; @@ -677,17 +670,14 @@ int get_vfs_caps_from_disk(struct mnt_idmap *idmap, case VFS_CAP_REVISION_1: if (size != XATTR_CAPS_SZ_1) return -EINVAL; - tocopy = VFS_CAP_U32_1; break; case VFS_CAP_REVISION_2: if (size != XATTR_CAPS_SZ_2) return -EINVAL; - tocopy = VFS_CAP_U32_2; break; case VFS_CAP_REVISION_3: if (size != XATTR_CAPS_SZ_3) return -EINVAL; - tocopy = VFS_CAP_U32_3; rootkuid = make_kuid(fs_ns, le32_to_cpu(nscaps->rootid)); break; @@ -705,15 +695,20 @@ int get_vfs_caps_from_disk(struct mnt_idmap *idmap, if (!rootid_owns_currentns(rootvfsuid)) return -ENODATA; - CAP_FOR_EACH_U32(i) { - if (i >= tocopy) - break; - cpu_caps->permitted.cap[i] = le32_to_cpu(caps->data[i].permitted); - cpu_caps->inheritable.cap[i] = le32_to_cpu(caps->data[i].inheritable); + cpu_caps->permitted.val = le32_to_cpu(caps->data[0].permitted); + cpu_caps->inheritable.val = le32_to_cpu(caps->data[0].inheritable); + + /* + * Rev1 had just a single 32-bit word, later expanded + * to a second one for the high bits + */ + if ((magic_etc & VFS_CAP_REVISION_MASK) != VFS_CAP_REVISION_1) { + cpu_caps->permitted.val += (u64)le32_to_cpu(caps->data[1].permitted) << 32; + cpu_caps->inheritable.val += (u64)le32_to_cpu(caps->data[1].inheritable) << 32; } - cpu_caps->permitted.cap[CAP_LAST_U32] &= CAP_LAST_U32_VALID_MASK; - cpu_caps->inheritable.cap[CAP_LAST_U32] &= CAP_LAST_U32_VALID_MASK; + cpu_caps->permitted.val &= CAP_VALID_MASK; + cpu_caps->inheritable.val &= CAP_VALID_MASK; cpu_caps->rootid = vfsuid_into_kuid(rootvfsuid); diff --git a/tools/testing/selftests/bpf/progs/test_deny_namespace.c b/tools/testing/selftests/bpf/progs/test_deny_namespace.c index 09ad5a4ebd1f1..591104e79812e 100644 --- a/tools/testing/selftests/bpf/progs/test_deny_namespace.c +++ b/tools/testing/selftests/bpf/progs/test_deny_namespace.c @@ -6,7 +6,7 @@ #include struct kernel_cap_struct { - __u32 cap[_LINUX_CAPABILITY_U32S_3]; + __u64 val; } __attribute__((preserve_access_index)); struct cred { @@ -19,14 +19,13 @@ SEC("lsm.s/userns_create") int BPF_PROG(test_userns_create, const struct cred *cred, int ret) { struct kernel_cap_struct caps = cred->cap_effective; - int cap_index = CAP_TO_INDEX(CAP_SYS_ADMIN); - __u32 cap_mask = CAP_TO_MASK(CAP_SYS_ADMIN); + __u64 cap_mask = BIT_LL(CAP_SYS_ADMIN); if (ret) return 0; ret = -EPERM; - if (caps.cap[cap_index] & cap_mask) + if (caps.val & cap_mask) return 0; return -EPERM; From e2a56364485e7789e7b8f342637c7f3a219f7ede Mon Sep 17 00:00:00 2001 From: Mario Limonciello Date: Tue, 28 Feb 2023 16:11:28 -0600 Subject: [PATCH 621/765] ACPI: x86: utils: Add Cezanne to the list for forcing StorageD3Enable commit 018d6711c26e4 ("ACPI: x86: Add a quirk for Dell Inspiron 14 2-in-1 for StorageD3Enable") introduced a quirk to allow a system with ambiguous use of _ADR 0 to force StorageD3Enable. It was reported that several more Dell systems suffered the same symptoms. As the list is continuing to grow but these are all Cezanne systems, instead add Cezanne to the CPU list to apply the StorageD3Enable property and remove the whole list. It was also reported that an HP system only has StorageD3Enable on the ACPI device for the first NVME disk, not the second. Link: https://bugzilla.kernel.org/show_bug.cgi?id=217003 Link: https://bugzilla.kernel.org/show_bug.cgi?id=216773 Reported-by: David Alvarez Lombardi Reported-by: dbilios@stdio.gr Reported-and-tested-by: Elvis Angelaccio Tested-by: victor.bonnelle@proton.me Tested-by: hurricanepootis@protonmail.com Signed-off-by: Mario Limonciello Signed-off-by: Rafael J. Wysocki --- drivers/acpi/x86/utils.c | 37 +++++++++++++------------------------ 1 file changed, 13 insertions(+), 24 deletions(-) diff --git a/drivers/acpi/x86/utils.c b/drivers/acpi/x86/utils.c index 4e816bb402f68..e45285d4e62a4 100644 --- a/drivers/acpi/x86/utils.c +++ b/drivers/acpi/x86/utils.c @@ -200,39 +200,28 @@ bool acpi_device_override_status(struct acpi_device *adev, unsigned long long *s * a hardcoded allowlist for D3 support, which was used for these platforms. * * This allows quirking on Linux in a similar fashion. + * + * Cezanne systems shouldn't *normally* need this as the BIOS includes + * StorageD3Enable. But for two reasons we have added it. + * 1) The BIOS on a number of Dell systems have ambiguity + * between the same value used for _ADR on ACPI nodes GPP1.DEV0 and GPP1.NVME. + * GPP1.NVME is needed to get StorageD3Enable node set properly. + * https://bugzilla.kernel.org/show_bug.cgi?id=216440 + * https://bugzilla.kernel.org/show_bug.cgi?id=216773 + * https://bugzilla.kernel.org/show_bug.cgi?id=217003 + * 2) On at least one HP system StorageD3Enable is missing on the second NVME + disk in the system. */ static const struct x86_cpu_id storage_d3_cpu_ids[] = { X86_MATCH_VENDOR_FAM_MODEL(AMD, 23, 96, NULL), /* Renoir */ X86_MATCH_VENDOR_FAM_MODEL(AMD, 23, 104, NULL), /* Lucienne */ - {} -}; - -static const struct dmi_system_id force_storage_d3_dmi[] = { - { - /* - * _ADR is ambiguous between GPP1.DEV0 and GPP1.NVME - * but .NVME is needed to get StorageD3Enable node - * https://bugzilla.kernel.org/show_bug.cgi?id=216440 - */ - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), - DMI_MATCH(DMI_PRODUCT_NAME, "Inspiron 14 7425 2-in-1"), - } - }, - { - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), - DMI_MATCH(DMI_PRODUCT_NAME, "Inspiron 16 5625"), - } - }, + X86_MATCH_VENDOR_FAM_MODEL(AMD, 25, 80, NULL), /* Cezanne */ {} }; bool force_storage_d3(void) { - const struct dmi_system_id *dmi_id = dmi_first_match(force_storage_d3_dmi); - - return dmi_id || x86_match_cpu(storage_d3_cpu_ids); + return x86_match_cpu(storage_d3_cpu_ids); } /* From f1b930e740811d416de4d2074da48b6633a672c8 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 27 Feb 2023 13:06:50 +0300 Subject: [PATCH 622/765] thermal: intel: quark_dts: fix error pointer dereference If alloc_soc_dts() fails, then we can just return. Trying to free "soc_dts" will lead to an Oops. Fixes: 8c1876939663 ("thermal: intel Quark SoC X1000 DTS thermal driver") Signed-off-by: Dan Carpenter Signed-off-by: Rafael J. Wysocki --- drivers/thermal/intel/intel_quark_dts_thermal.c | 12 ++---------- 1 file changed, 2 insertions(+), 10 deletions(-) diff --git a/drivers/thermal/intel/intel_quark_dts_thermal.c b/drivers/thermal/intel/intel_quark_dts_thermal.c index 97b843fa75680..ffdc95047838f 100644 --- a/drivers/thermal/intel/intel_quark_dts_thermal.c +++ b/drivers/thermal/intel/intel_quark_dts_thermal.c @@ -400,22 +400,14 @@ MODULE_DEVICE_TABLE(x86cpu, qrk_thermal_ids); static int __init intel_quark_thermal_init(void) { - int err = 0; - if (!x86_match_cpu(qrk_thermal_ids) || !iosf_mbi_available()) return -ENODEV; soc_dts = alloc_soc_dts(); - if (IS_ERR(soc_dts)) { - err = PTR_ERR(soc_dts); - goto err_free; - } + if (IS_ERR(soc_dts)) + return PTR_ERR(soc_dts); return 0; - -err_free: - free_soc_dts(soc_dts); - return err; } static void __exit intel_quark_thermal_exit(void) From 1467fb960349dfa5e300658f1a409dde2cfb0c51 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 25 Feb 2023 21:39:52 -0800 Subject: [PATCH 623/765] thermal: intel: BXT_PMIC: select REGMAP instead of depending on it REGMAP is a hidden (not user visible) symbol. Users cannot set it directly thru "make *config", so drivers should select it instead of depending on it if they need it. Consistently using "select" or "depends on" can also help reduce Kconfig circular dependency issues. Therefore, change the use of "depends on REGMAP" to "select REGMAP". Fixes: b474303ffd57 ("thermal: add Intel BXT WhiskeyCove PMIC thermal driver") Signed-off-by: Randy Dunlap Signed-off-by: Rafael J. Wysocki --- drivers/thermal/intel/Kconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/thermal/intel/Kconfig b/drivers/thermal/intel/Kconfig index b5808f92702df..cb7e7697cf1e3 100644 --- a/drivers/thermal/intel/Kconfig +++ b/drivers/thermal/intel/Kconfig @@ -73,7 +73,8 @@ endmenu config INTEL_BXT_PMIC_THERMAL tristate "Intel Broxton PMIC thermal driver" - depends on X86 && INTEL_SOC_PMIC_BXTWC && REGMAP + depends on X86 && INTEL_SOC_PMIC_BXTWC + select REGMAP help Select this driver for Intel Broxton PMIC with ADC channels monitoring system temperature measurements and alerts. From f43523620f646c89ffd8ada840a0068290e51266 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 27 Feb 2023 13:07:09 +0300 Subject: [PATCH 624/765] cpufreq: apple-soc: Fix an IS_ERR() vs NULL check The of_iomap() function returns NULL if it fails. It never returns error pointers. Fix the check accordingly. Fixes: 6286bbb40576 ("cpufreq: apple-soc: Add new driver to control Apple SoC CPU P-states") Signed-off-by: Dan Carpenter Reviewed-by: Eric Curtin Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/apple-soc-cpufreq.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/cpufreq/apple-soc-cpufreq.c b/drivers/cpufreq/apple-soc-cpufreq.c index c11d22fd84c37..021f423705e1b 100644 --- a/drivers/cpufreq/apple-soc-cpufreq.c +++ b/drivers/cpufreq/apple-soc-cpufreq.c @@ -189,8 +189,8 @@ static int apple_soc_cpufreq_find_cluster(struct cpufreq_policy *policy, *info = match->data; *reg_base = of_iomap(args.np, 0); - if (IS_ERR(*reg_base)) - return PTR_ERR(*reg_base); + if (!*reg_base) + return -ENOMEM; return 0; } From 92304df83b943776492309f42452effea0cc1089 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 14 Feb 2023 14:20:42 +0100 Subject: [PATCH 625/765] power: supply: qcom_battmgr: remove bogus do_div() The argument to do_div() is a 32-bit integer, and it was read from a 32-bit register so there is no point in doing a 64-bit division on it. On 32-bit arm, do_div() causes a compile-time warning here: include/asm-generic/div64.h:238:22: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Werror=incompatible-pointer-types] 238 | __rem = __div64_32(&(n), __base); \ | ^~~~ | | | unsigned int * drivers/power/supply/qcom_battmgr.c:1130:4: note: in expansion of macro 'do_div' 1130 | do_div(battmgr->status.percent, 100); Signed-off-by: Arnd Bergmann Reviewed-by: Geert Uytterhoeven Acked-by: Sebastian Reichel Reviewed-by: Konrad Dybcio Signed-off-by: Linus Torvalds --- drivers/power/supply/qcom_battmgr.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/power/supply/qcom_battmgr.c b/drivers/power/supply/qcom_battmgr.c index ec31f887184fd..de77df97b3a44 100644 --- a/drivers/power/supply/qcom_battmgr.c +++ b/drivers/power/supply/qcom_battmgr.c @@ -1126,8 +1126,7 @@ static void qcom_battmgr_sm8350_callback(struct qcom_battmgr *battmgr, battmgr->info.charge_type = le32_to_cpu(resp->intval.value); break; case BATT_CAPACITY: - battmgr->status.percent = le32_to_cpu(resp->intval.value); - do_div(battmgr->status.percent, 100); + battmgr->status.percent = le32_to_cpu(resp->intval.value) / 100; break; case BATT_VOLT_OCV: battmgr->status.voltage_ocv = le32_to_cpu(resp->intval.value); From 8c42dd78df148c90e48efff204cce38743906a79 Mon Sep 17 00:00:00 2001 From: Gerald Schaefer Date: Mon, 27 Feb 2023 20:03:00 +0100 Subject: [PATCH 626/765] s390/extmem: return correct segment type in __segment_load() Commit f05f62d04271f ("s390/vmem: get rid of memory segment list") reshuffled the call to vmem_add_mapping() in __segment_load(), which now overwrites rc after it was set to contain the segment type code. As result, __segment_load() will now always return 0 on success, which corresponds to the segment type code SEG_TYPE_SW, i.e. a writeable segment. This results in a kernel crash when loading a read-only segment as dcssblk block device, and trying to write to it. Instead of reshuffling code again, make sure to return the segment type on success, and also describe this rather delicate and unexpected logic in the function comment. Also initialize new segtype variable with invalid value, to prevent possible future confusion. Fixes: f05f62d04271 ("s390/vmem: get rid of memory segment list") Cc: # 5.9+ Signed-off-by: Gerald Schaefer Reviewed-by: Heiko Carstens Signed-off-by: Heiko Carstens --- arch/s390/mm/extmem.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/arch/s390/mm/extmem.c b/arch/s390/mm/extmem.c index 5060956b8e7d6..1bc42ce265990 100644 --- a/arch/s390/mm/extmem.c +++ b/arch/s390/mm/extmem.c @@ -289,15 +289,17 @@ segment_overlaps_others (struct dcss_segment *seg) /* * real segment loading function, called from segment_load + * Must return either an error code < 0, or the segment type code >= 0 */ static int __segment_load (char *name, int do_nonshared, unsigned long *addr, unsigned long *end) { unsigned long start_addr, end_addr, dummy; struct dcss_segment *seg; - int rc, diag_cc; + int rc, diag_cc, segtype; start_addr = end_addr = 0; + segtype = -1; seg = kmalloc(sizeof(*seg), GFP_KERNEL | GFP_DMA); if (seg == NULL) { rc = -ENOMEM; @@ -326,9 +328,9 @@ __segment_load (char *name, int do_nonshared, unsigned long *addr, unsigned long seg->res_name[8] = '\0'; strlcat(seg->res_name, " (DCSS)", sizeof(seg->res_name)); seg->res->name = seg->res_name; - rc = seg->vm_segtype; - if (rc == SEG_TYPE_SC || - ((rc == SEG_TYPE_SR || rc == SEG_TYPE_ER) && !do_nonshared)) + segtype = seg->vm_segtype; + if (segtype == SEG_TYPE_SC || + ((segtype == SEG_TYPE_SR || segtype == SEG_TYPE_ER) && !do_nonshared)) seg->res->flags |= IORESOURCE_READONLY; /* Check for overlapping resources before adding the mapping. */ @@ -386,7 +388,7 @@ __segment_load (char *name, int do_nonshared, unsigned long *addr, unsigned long out_free: kfree(seg); out: - return rc; + return rc < 0 ? rc : segtype; } /* From 6ca6b58107a8891e4b08087843188fdc5737ec08 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 1 Mar 2023 09:41:06 +0100 Subject: [PATCH 627/765] s390/Kconfig: sort config S390 select list again Keep the config S390 select list sorted. Signed-off-by: Heiko Carstens --- arch/s390/Kconfig | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 933771b0b07a0..e2c2f1516c269 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -125,8 +125,8 @@ config S390 select ARCH_WANTS_DYNAMIC_TASK_STRUCT select ARCH_WANTS_NO_INSTR select ARCH_WANT_DEFAULT_BPF_JIT - select ARCH_WANT_IPC_PARSE_VERSION select ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP + select ARCH_WANT_IPC_PARSE_VERSION select BUILDTIME_TABLE_SORT select CLONE_BACKWARDS2 select DMA_OPS if PCI @@ -187,7 +187,6 @@ config S390 select HAVE_KPROBES select HAVE_KPROBES_ON_FTRACE select HAVE_KRETPROBES - select HAVE_RETHOOK select HAVE_KVM select HAVE_LIVEPATCH select HAVE_MEMBLOCK_PHYS_MAP @@ -200,6 +199,7 @@ config S390 select HAVE_PERF_USER_STACK_DUMP select HAVE_REGS_AND_STACK_ACCESS_API select HAVE_RELIABLE_STACKTRACE + select HAVE_RETHOOK select HAVE_RSEQ select HAVE_SAMPLE_FTRACE_DIRECT select HAVE_SAMPLE_FTRACE_DIRECT_MULTI @@ -210,9 +210,9 @@ config S390 select HAVE_VIRT_CPU_ACCOUNTING_IDLE select IOMMU_HELPER if PCI select IOMMU_SUPPORT if PCI + select MMU_GATHER_MERGE_VMAS select MMU_GATHER_NO_GATHER select MMU_GATHER_RCU_TABLE_FREE - select MMU_GATHER_MERGE_VMAS select MODULES_USE_ELF_RELA select NEED_DMA_MAP_STATE if PCI select NEED_PER_CPU_EMBED_FIRST_CHUNK From 9f828bc3fb900c5d39ec13ea150341d28fb1f158 Mon Sep 17 00:00:00 2001 From: Mayuresh Chitale Date: Wed, 8 Feb 2023 13:13:14 +0530 Subject: [PATCH 628/765] drivers/perf: RISC-V: Allow programming custom firmware events Applications need to be able to program the SBI implementation specific or custom firmware events in addition to the standard firmware events. Remove a check in the driver that prohibits the programming of the custom firmware events. Signed-off-by: Mayuresh Chitale Reviewed-by: Andrew Jones Link: https://lore.kernel.org/r/20230208074314.3661406-1-mchitale@ventanamicro.com Signed-off-by: Palmer Dabbelt --- drivers/perf/riscv_pmu_sbi.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/drivers/perf/riscv_pmu_sbi.c b/drivers/perf/riscv_pmu_sbi.c index 7b2288d4b1eca..8bd27d795253c 100644 --- a/drivers/perf/riscv_pmu_sbi.c +++ b/drivers/perf/riscv_pmu_sbi.c @@ -385,11 +385,8 @@ static int pmu_sbi_event_map(struct perf_event *event, u64 *econfig) bSoftware = config >> 63; raw_config_val = config & RISCV_PMU_RAW_EVENT_MASK; if (bSoftware) { - if (raw_config_val < SBI_PMU_FW_MAX) - ret = (raw_config_val & 0xFFFF) | - (SBI_PMU_EVENT_TYPE_FW << 16); - else - return -EINVAL; + ret = (raw_config_val & 0xFFFF) | + (SBI_PMU_EVENT_TYPE_FW << 16); } else { ret = RISCV_PMU_RAW_EVENT_IDX; *econfig = raw_config_val; From 1907e0fec10c2498729d27550c460528e49abe1e Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 23 Feb 2023 08:15:38 +0000 Subject: [PATCH 629/765] cifs: Add some missing xas_retry() calls The xas_for_each loops added into fs/cifs/file.c need to go round again if indicated by xas_retry(). Fixes: b8713c4dbfa3 ("cifs: Add some helper functions") Signed-off-by: David Howells Reviewed-by: Paulo Alcantara (SUSE) cc: Shyam Prasad N cc: Rohith Surabattula cc: Tom Talpey cc: Jeff Layton cc: linux-cifs@vger.kernel.org Signed-off-by: Steve French --- fs/cifs/file.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/fs/cifs/file.c b/fs/cifs/file.c index ebfcaae8c4373..879524bf7528b 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -52,6 +52,8 @@ static void cifs_undirty_folios(struct inode *inode, loff_t start, unsigned int end = (start + len - 1) / PAGE_SIZE; xas_for_each_marked(&xas, folio, end, PAGECACHE_TAG_DIRTY) { + if (xas_retry(&xas, folio)) + continue; xas_pause(&xas); rcu_read_unlock(); folio_lock(folio); @@ -81,6 +83,8 @@ void cifs_pages_written_back(struct inode *inode, loff_t start, unsigned int len end = (start + len - 1) / PAGE_SIZE; xas_for_each(&xas, folio, end) { + if (xas_retry(&xas, folio)) + continue; if (!folio_test_writeback(folio)) { WARN_ONCE(1, "bad %x @%llx page %lx %lx\n", len, start, folio_index(folio), end); @@ -112,6 +116,8 @@ void cifs_pages_write_failed(struct inode *inode, loff_t start, unsigned int len end = (start + len - 1) / PAGE_SIZE; xas_for_each(&xas, folio, end) { + if (xas_retry(&xas, folio)) + continue; if (!folio_test_writeback(folio)) { WARN_ONCE(1, "bad %x @%llx page %lx %lx\n", len, start, folio_index(folio), end); From 4bd3e4c7e4e552aa148edd3f2a2326ccc3011122 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 23 Feb 2023 08:15:39 +0000 Subject: [PATCH 630/765] cifs: Fix an uninitialised variable Fix an uninitialised variable introduced in cifs. Fixes: 3d78fe73fa12 ("cifs: Build the RDMA SGE list directly from an iterator") Signed-off-by: David Howells Reviewed-by: Paulo Alcantara (SUSE) cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Tom Talpey cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: linux-rdma@vger.kernel.org Signed-off-by: Steve French --- fs/cifs/smbdirect.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/cifs/smbdirect.c b/fs/cifs/smbdirect.c index 55b6e319a61dc..0362ebd4fa0fc 100644 --- a/fs/cifs/smbdirect.c +++ b/fs/cifs/smbdirect.c @@ -837,7 +837,7 @@ static int smbd_post_send_iter(struct smbd_connection *info, int data_length; struct smbd_request *request; struct smbd_data_transfer *packet; - int new_credits; + int new_credits = 0; wait_credit: /* Wait for send credits. A SMBD packet needs one credit */ From a21be1f5df85831147d5a1cbb5f2a13ec8c6852b Mon Sep 17 00:00:00 2001 From: Shyam Prasad N Date: Tue, 27 Dec 2022 14:04:29 +0000 Subject: [PATCH 631/765] cifs: match even the scope id for ipv6 addresses match_address function matches the scope id for ipv6 addresses, but cifs_match_ipaddr (which is another function used for comparison) does not use scope id. Doing so with this change. Signed-off-by: Shyam Prasad N Signed-off-by: Steve French --- fs/cifs/connect.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index ec020d860be32..5dabd8dc3e8e2 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -1294,7 +1294,8 @@ cifs_match_ipaddr(struct sockaddr *srcaddr, struct sockaddr *rhs) case AF_INET6: { struct sockaddr_in6 *saddr6 = (struct sockaddr_in6 *)srcaddr; struct sockaddr_in6 *vaddr6 = (struct sockaddr_in6 *)rhs; - return ipv6_addr_equal(&saddr6->sin6_addr, &vaddr6->sin6_addr); + return (ipv6_addr_equal(&saddr6->sin6_addr, &vaddr6->sin6_addr) + && saddr6->sin6_scope_id == vaddr6->sin6_scope_id); } default: WARN_ON(1); From 410612b0726b2ee68808da7bd27103d96b4cf898 Mon Sep 17 00:00:00 2001 From: Shyam Prasad N Date: Tue, 27 Dec 2022 14:09:32 +0000 Subject: [PATCH 632/765] cifs: reuse cifs_match_ipaddr for comparison of dstaddr too We have two pieces of code that does pretty much the same comparison. This change reuses cifs_match_ipaddr within match_address. Signed-off-by: Shyam Prasad N Signed-off-by: Steve French --- fs/cifs/connect.c | 28 ++-------------------------- 1 file changed, 2 insertions(+), 26 deletions(-) diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 5dabd8dc3e8e2..5233f14f0636a 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -1344,32 +1344,8 @@ match_port(struct TCP_Server_Info *server, struct sockaddr *addr) static bool match_server_address(struct TCP_Server_Info *server, struct sockaddr *addr) { - switch (addr->sa_family) { - case AF_INET: { - struct sockaddr_in *addr4 = (struct sockaddr_in *)addr; - struct sockaddr_in *srv_addr4 = - (struct sockaddr_in *)&server->dstaddr; - - if (addr4->sin_addr.s_addr != srv_addr4->sin_addr.s_addr) - return false; - break; - } - case AF_INET6: { - struct sockaddr_in6 *addr6 = (struct sockaddr_in6 *)addr; - struct sockaddr_in6 *srv_addr6 = - (struct sockaddr_in6 *)&server->dstaddr; - - if (!ipv6_addr_equal(&addr6->sin6_addr, - &srv_addr6->sin6_addr)) - return false; - if (addr6->sin6_scope_id != srv_addr6->sin6_scope_id) - return false; - break; - } - default: - WARN_ON(1); - return false; /* don't expect to be here */ - } + if (!cifs_match_ipaddr(addr, (struct sockaddr *)&server->dstaddr)) + return false; return true; } From 0268792f77d2e4ba8e056173bdf2f9af9963be76 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 27 Feb 2023 13:04:53 +0000 Subject: [PATCH 633/765] cifs: Fix cifs_write_back_from_locked_folio() cifs_write_back_from_locked_folio() should return the number of bytes read, but returns the result of ->async_writev(), which will be 0 on success. As it happens, this doesn't prevent cifs_writepages_region() from working as it will then examine and ignore the pages that are no longer dirty rather than just skipping over them. Fixes: d08089f649a0 ("cifs: Change the I/O paths to use an iterator rather than a page list") Signed-off-by: David Howells cc: Shyam Prasad N cc: Rohith Surabattula cc: Tom Talpey cc: Jeff Layton cc: linux-cifs@vger.kernel.org Reviewed-by: Paulo Alcantara (SUSE) Signed-off-by: Steve French --- fs/cifs/file.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 879524bf7528b..ec0694a65c7b6 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -2845,6 +2845,7 @@ static ssize_t cifs_write_back_from_locked_folio(struct address_space *mapping, free_xid(xid); if (rc == 0) { wbc->nr_to_write = count; + rc = len; } else if (is_retryable_error(rc)) { cifs_pages_write_redirty(inode, start, len); } else { From 4c0421fa6df136ff869a078594b4b7b7637e566a Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 27 Feb 2023 13:04:54 +0000 Subject: [PATCH 634/765] iov: Fix netfs_extract_user_to_sg() Fix the loop check in netfs_extract_user_to_sg() for extraction from user-backed iterators to do the body if npages > 0, not if npages < 0 (which it can never be). This isn't currently used by cifs, which only ever extracts data from BVEC, KVEC and XARRAY iterators at this level, user-backed iterators having being decanted into BVEC iterators at a higher level to accommodate the work being done in a kernel thread. Found by smatch: fs/netfs/iterator.c:139 netfs_extract_user_to_sg() warn: unsigned 'npages' is never less than zero. Fixes: 018584697533 ("netfs: Add a function to extract an iterator into a scatterlist") Reported-by: kernel test robot Link: https://lore.kernel.org/oe-kbuild-all/202302261115.P3TQi1ZO-lkp@intel.com/ Reported-by: Dan Carpenter Link: https://lore.kernel.org/r/Y/yYnAhoAYDBKixX@kili Reviewed-by: Paulo Alcantara (SUSE) Signed-off-by: David Howells cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: linux-cachefs@redhat.com Signed-off-by: Steve French --- fs/netfs/iterator.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/netfs/iterator.c b/fs/netfs/iterator.c index f00d43b8ac0a8..e9a45dea748a2 100644 --- a/fs/netfs/iterator.c +++ b/fs/netfs/iterator.c @@ -134,7 +134,7 @@ static ssize_t netfs_extract_user_to_sg(struct iov_iter *iter, npages = DIV_ROUND_UP(off + len, PAGE_SIZE); sg_max -= npages; - for (; npages < 0; npages--) { + for (; npages > 0; npages--) { struct page *page = *pages; size_t seg = min_t(size_t, PAGE_SIZE - off, len); From b9ee2e307c6b06384b6f9e393a9b8e048e8fc277 Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Tue, 28 Feb 2023 19:01:54 -0300 Subject: [PATCH 635/765] cifs: improve checking of DFS links over STATUS_OBJECT_NAME_INVALID Do not map STATUS_OBJECT_NAME_INVALID to -EREMOTE under non-DFS shares, or 'nodfs' mounts or CONFIG_CIFS_DFS_UPCALL=n builds. Otherwise, in the slow path, get a referral to figure out whether it is an actual DFS link. This could be simply reproduced under a non-DFS share by running the following $ mount.cifs //srv/share /mnt -o ... $ cat /mnt/$(printf '\U110000') cat: '/mnt/'$'\364\220\200\200': Object is remote Fixes: c877ce47e137 ("cifs: reduce roundtrips on create/qinfo requests") CC: stable@vger.kernel.org # 6.2 Signed-off-by: Paulo Alcantara (SUSE) Reviewed-by: Ronnie Sahlberg Signed-off-by: Steve French --- fs/cifs/cifsproto.h | 20 ++++++++++---- fs/cifs/misc.c | 67 +++++++++++++++++++++++++++++++++++++++++++++ fs/cifs/smb2inode.c | 21 +++++++------- fs/cifs/smb2ops.c | 23 +++++++++------- 4 files changed, 106 insertions(+), 25 deletions(-) diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index b7a36ebd0f2f5..20a2f0f3f6825 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -667,11 +667,21 @@ static inline int get_dfs_path(const unsigned int xid, struct cifs_ses *ses, int match_target_ip(struct TCP_Server_Info *server, const char *share, size_t share_len, bool *result); - -int cifs_dfs_query_info_nonascii_quirk(const unsigned int xid, - struct cifs_tcon *tcon, - struct cifs_sb_info *cifs_sb, - const char *dfs_link_path); +int cifs_inval_name_dfs_link_error(const unsigned int xid, + struct cifs_tcon *tcon, + struct cifs_sb_info *cifs_sb, + const char *full_path, + bool *islink); +#else +static inline int cifs_inval_name_dfs_link_error(const unsigned int xid, + struct cifs_tcon *tcon, + struct cifs_sb_info *cifs_sb, + const char *full_path, + bool *islink) +{ + *islink = false; + return 0; +} #endif static inline int cifs_create_options(struct cifs_sb_info *cifs_sb, int options) diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 2905734eb289b..0c6c1fc8dae98 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -21,6 +21,7 @@ #include "cifsfs.h" #ifdef CONFIG_CIFS_DFS_UPCALL #include "dns_resolve.h" +#include "dfs_cache.h" #endif #include "fs_context.h" #include "cached_dir.h" @@ -1198,4 +1199,70 @@ int cifs_update_super_prepath(struct cifs_sb_info *cifs_sb, char *prefix) cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_USE_PREFIX_PATH; return 0; } + +/* + * Handle weird Windows SMB server behaviour. It responds with + * STATUS_OBJECT_NAME_INVALID code to SMB2 QUERY_INFO request for + * "\\\" DFS reference, where contains + * non-ASCII unicode symbols. + */ +int cifs_inval_name_dfs_link_error(const unsigned int xid, + struct cifs_tcon *tcon, + struct cifs_sb_info *cifs_sb, + const char *full_path, + bool *islink) +{ + struct cifs_ses *ses = tcon->ses; + size_t len; + char *path; + char *ref_path; + + *islink = false; + + /* + * Fast path - skip check when @full_path doesn't have a prefix path to + * look up or tcon is not DFS. + */ + if (strlen(full_path) < 2 || !cifs_sb || + (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_DFS) || + !is_tcon_dfs(tcon) || !ses->server->origin_fullpath) + return 0; + + /* + * Slow path - tcon is DFS and @full_path has prefix path, so attempt + * to get a referral to figure out whether it is an DFS link. + */ + len = strnlen(tcon->tree_name, MAX_TREE_SIZE + 1) + strlen(full_path) + 1; + path = kmalloc(len, GFP_KERNEL); + if (!path) + return -ENOMEM; + + scnprintf(path, len, "%s%s", tcon->tree_name, full_path); + ref_path = dfs_cache_canonical_path(path + 1, cifs_sb->local_nls, + cifs_remap(cifs_sb)); + kfree(path); + + if (IS_ERR(ref_path)) { + if (PTR_ERR(ref_path) != -EINVAL) + return PTR_ERR(ref_path); + } else { + struct dfs_info3_param *refs = NULL; + int num_refs = 0; + + /* + * XXX: we are not using dfs_cache_find() here because we might + * end filling all the DFS cache and thus potentially + * removing cached DFS targets that the client would eventually + * need during failover. + */ + if (ses->server->ops->get_dfs_refer && + !ses->server->ops->get_dfs_refer(xid, ses, ref_path, &refs, + &num_refs, cifs_sb->local_nls, + cifs_remap(cifs_sb))) + *islink = refs[0].server_type == DFS_TYPE_LINK; + free_dfs_info_array(refs, num_refs); + kfree(ref_path); + } + return 0; +} #endif diff --git a/fs/cifs/smb2inode.c b/fs/cifs/smb2inode.c index 37b4cd59245d6..9b956294e8643 100644 --- a/fs/cifs/smb2inode.c +++ b/fs/cifs/smb2inode.c @@ -527,12 +527,13 @@ int smb2_query_path_info(const unsigned int xid, struct cifs_tcon *tcon, struct cifs_sb_info *cifs_sb, const char *full_path, struct cifs_open_info_data *data, bool *adjust_tz, bool *reparse) { - int rc; __u32 create_options = 0; struct cifsFileInfo *cfile; struct cached_fid *cfid = NULL; struct kvec err_iov[3] = {}; int err_buftype[3] = {}; + bool islink; + int rc, rc2; *adjust_tz = false; *reparse = false; @@ -580,15 +581,15 @@ int smb2_query_path_info(const unsigned int xid, struct cifs_tcon *tcon, SMB2_OP_QUERY_INFO, cfile, NULL, NULL, NULL, NULL); goto out; - } else if (rc != -EREMOTE && IS_ENABLED(CONFIG_CIFS_DFS_UPCALL) && - hdr->Status == STATUS_OBJECT_NAME_INVALID) { - /* - * Handle weird Windows SMB server behaviour. It responds with - * STATUS_OBJECT_NAME_INVALID code to SMB2 QUERY_INFO request - * for "\\\" DFS reference, - * where contains non-ASCII unicode symbols. - */ - rc = -EREMOTE; + } else if (rc != -EREMOTE && hdr->Status == STATUS_OBJECT_NAME_INVALID) { + rc2 = cifs_inval_name_dfs_link_error(xid, tcon, cifs_sb, + full_path, &islink); + if (rc2) { + rc = rc2; + goto out; + } + if (islink) + rc = -EREMOTE; } if (rc == -EREMOTE && IS_ENABLED(CONFIG_CIFS_DFS_UPCALL) && cifs_sb && (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_DFS)) diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c index f79b075f2992f..6dfb865ee9d75 100644 --- a/fs/cifs/smb2ops.c +++ b/fs/cifs/smb2ops.c @@ -796,7 +796,6 @@ static int smb2_is_path_accessible(const unsigned int xid, struct cifs_tcon *tcon, struct cifs_sb_info *cifs_sb, const char *full_path) { - int rc; __le16 *utf16_path; __u8 oplock = SMB2_OPLOCK_LEVEL_NONE; int err_buftype = CIFS_NO_BUFFER; @@ -804,6 +803,8 @@ smb2_is_path_accessible(const unsigned int xid, struct cifs_tcon *tcon, struct kvec err_iov = {}; struct cifs_fid fid; struct cached_fid *cfid; + bool islink; + int rc, rc2; rc = open_cached_dir(xid, tcon, full_path, cifs_sb, true, &cfid); if (!rc) { @@ -833,15 +834,17 @@ smb2_is_path_accessible(const unsigned int xid, struct cifs_tcon *tcon, if (unlikely(!hdr || err_buftype == CIFS_NO_BUFFER)) goto out; - /* - * Handle weird Windows SMB server behaviour. It responds with - * STATUS_OBJECT_NAME_INVALID code to SMB2 QUERY_INFO request - * for "\\\" DFS reference, - * where contains non-ASCII unicode symbols. - */ - if (rc != -EREMOTE && IS_ENABLED(CONFIG_CIFS_DFS_UPCALL) && - hdr->Status == STATUS_OBJECT_NAME_INVALID) - rc = -EREMOTE; + + if (rc != -EREMOTE && hdr->Status == STATUS_OBJECT_NAME_INVALID) { + rc2 = cifs_inval_name_dfs_link_error(xid, tcon, cifs_sb, + full_path, &islink); + if (rc2) { + rc = rc2; + goto out; + } + if (islink) + rc = -EREMOTE; + } if (rc == -EREMOTE && IS_ENABLED(CONFIG_CIFS_DFS_UPCALL) && cifs_sb && (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_DFS)) rc = -EOPNOTSUPP; From 1bcd548d935a33c6fc58331405eb1b82fd6150de Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Tue, 28 Feb 2023 19:01:55 -0300 Subject: [PATCH 636/765] cifs: prevent data race in cifs_reconnect_tcon() Make sure to get an up-to-date TCP_Server_Info::nr_targets value prior to waiting the server to be reconnected in cifs_reconnect_tcon(). It is set in cifs_tcp_ses_needs_reconnect() and protected by TCP_Server_Info::srv_lock. Create a new cifs_wait_for_server_reconnect() helper that can be used by both SMB2+ and CIFS reconnect code. Signed-off-by: Paulo Alcantara (SUSE) Signed-off-by: Steve French --- fs/cifs/cifsproto.h | 1 + fs/cifs/cifssmb.c | 43 ++---------------------- fs/cifs/misc.c | 44 ++++++++++++++++++++++++ fs/cifs/smb2pdu.c | 82 ++++++++++++--------------------------------- 4 files changed, 69 insertions(+), 101 deletions(-) diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 20a2f0f3f6825..e2eff66eefabf 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -694,5 +694,6 @@ static inline int cifs_create_options(struct cifs_sb_info *cifs_sb, int options) struct super_block *cifs_get_tcon_super(struct cifs_tcon *tcon); void cifs_put_tcon_super(struct super_block *sb); +int cifs_wait_for_server_reconnect(struct TCP_Server_Info *server, bool retry); #endif /* _CIFSPROTO_H */ diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index a24e4ddf80432..a43c78396dd88 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -72,7 +72,6 @@ cifs_reconnect_tcon(struct cifs_tcon *tcon, int smb_command) struct cifs_ses *ses; struct TCP_Server_Info *server; struct nls_table *nls_codepage; - int retries; /* * SMBs NegProt, SessSetup, uLogoff do not have tcon yet so check for @@ -102,45 +101,9 @@ cifs_reconnect_tcon(struct cifs_tcon *tcon, int smb_command) } spin_unlock(&tcon->tc_lock); - retries = server->nr_targets; - - /* - * Give demultiplex thread up to 10 seconds to each target available for - * reconnect -- should be greater than cifs socket timeout which is 7 - * seconds. - */ - while (server->tcpStatus == CifsNeedReconnect) { - rc = wait_event_interruptible_timeout(server->response_q, - (server->tcpStatus != CifsNeedReconnect), - 10 * HZ); - if (rc < 0) { - cifs_dbg(FYI, "%s: aborting reconnect due to a received signal by the process\n", - __func__); - return -ERESTARTSYS; - } - - /* are we still trying to reconnect? */ - spin_lock(&server->srv_lock); - if (server->tcpStatus != CifsNeedReconnect) { - spin_unlock(&server->srv_lock); - break; - } - spin_unlock(&server->srv_lock); - - if (retries && --retries) - continue; - - /* - * on "soft" mounts we wait once. Hard mounts keep - * retrying until process is killed or server comes - * back on-line - */ - if (!tcon->retry) { - cifs_dbg(FYI, "gave up waiting on reconnect in smb_init\n"); - return -EHOSTDOWN; - } - retries = server->nr_targets; - } + rc = cifs_wait_for_server_reconnect(server, tcon->retry); + if (rc) + return rc; spin_lock(&ses->chan_lock); if (!cifs_chan_needs_reconnect(ses, server) && !tcon->need_reconnect) { diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 0c6c1fc8dae98..a0d286ee723dd 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -1266,3 +1266,47 @@ int cifs_inval_name_dfs_link_error(const unsigned int xid, return 0; } #endif + +int cifs_wait_for_server_reconnect(struct TCP_Server_Info *server, bool retry) +{ + int timeout = 10; + int rc; + + spin_lock(&server->srv_lock); + if (server->tcpStatus != CifsNeedReconnect) { + spin_unlock(&server->srv_lock); + return 0; + } + timeout *= server->nr_targets; + spin_unlock(&server->srv_lock); + + /* + * Give demultiplex thread up to 10 seconds to each target available for + * reconnect -- should be greater than cifs socket timeout which is 7 + * seconds. + * + * On "soft" mounts we wait once. Hard mounts keep retrying until + * process is killed or server comes back on-line. + */ + do { + rc = wait_event_interruptible_timeout(server->response_q, + (server->tcpStatus != CifsNeedReconnect), + timeout * HZ); + if (rc < 0) { + cifs_dbg(FYI, "%s: aborting reconnect due to received signal\n", + __func__); + return -ERESTARTSYS; + } + + /* are we still trying to reconnect? */ + spin_lock(&server->srv_lock); + if (server->tcpStatus != CifsNeedReconnect) { + spin_unlock(&server->srv_lock); + return 0; + } + spin_unlock(&server->srv_lock); + } while (retry); + + cifs_dbg(FYI, "%s: gave up waiting on reconnect\n", __func__); + return -EHOSTDOWN; +} diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c index ca9d7110ddcbe..0e53265e1462a 100644 --- a/fs/cifs/smb2pdu.c +++ b/fs/cifs/smb2pdu.c @@ -139,66 +139,6 @@ smb2_hdr_assemble(struct smb2_hdr *shdr, __le16 smb2_cmd, return; } -static int wait_for_server_reconnect(struct TCP_Server_Info *server, - __le16 smb2_command, bool retry) -{ - int timeout = 10; - int rc; - - spin_lock(&server->srv_lock); - if (server->tcpStatus != CifsNeedReconnect) { - spin_unlock(&server->srv_lock); - return 0; - } - timeout *= server->nr_targets; - spin_unlock(&server->srv_lock); - - /* - * Return to caller for TREE_DISCONNECT and LOGOFF and CLOSE - * here since they are implicitly done when session drops. - */ - switch (smb2_command) { - /* - * BB Should we keep oplock break and add flush to exceptions? - */ - case SMB2_TREE_DISCONNECT: - case SMB2_CANCEL: - case SMB2_CLOSE: - case SMB2_OPLOCK_BREAK: - return -EAGAIN; - } - - /* - * Give demultiplex thread up to 10 seconds to each target available for - * reconnect -- should be greater than cifs socket timeout which is 7 - * seconds. - * - * On "soft" mounts we wait once. Hard mounts keep retrying until - * process is killed or server comes back on-line. - */ - do { - rc = wait_event_interruptible_timeout(server->response_q, - (server->tcpStatus != CifsNeedReconnect), - timeout * HZ); - if (rc < 0) { - cifs_dbg(FYI, "%s: aborting reconnect due to received signal\n", - __func__); - return -ERESTARTSYS; - } - - /* are we still trying to reconnect? */ - spin_lock(&server->srv_lock); - if (server->tcpStatus != CifsNeedReconnect) { - spin_unlock(&server->srv_lock); - return 0; - } - spin_unlock(&server->srv_lock); - } while (retry); - - cifs_dbg(FYI, "%s: gave up waiting on reconnect\n", __func__); - return -EHOSTDOWN; -} - static int smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon, struct TCP_Server_Info *server) @@ -243,7 +183,27 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon, (!tcon->ses->server) || !server) return -EIO; - rc = wait_for_server_reconnect(server, smb2_command, tcon->retry); + spin_lock(&server->srv_lock); + if (server->tcpStatus == CifsNeedReconnect) { + /* + * Return to caller for TREE_DISCONNECT and LOGOFF and CLOSE + * here since they are implicitly done when session drops. + */ + switch (smb2_command) { + /* + * BB Should we keep oplock break and add flush to exceptions? + */ + case SMB2_TREE_DISCONNECT: + case SMB2_CANCEL: + case SMB2_CLOSE: + case SMB2_OPLOCK_BREAK: + spin_unlock(&server->srv_lock); + return -EAGAIN; + } + } + spin_unlock(&server->srv_lock); + + rc = cifs_wait_for_server_reconnect(server, tcon->retry); if (rc) return rc; From 71562809e401b2f5ad371d99ce0323e988406fd6 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 28 Feb 2023 22:38:38 +0000 Subject: [PATCH 637/765] cifs: Fix memory leak in direct I/O When __cifs_readv() and __cifs_writev() extract pages from a user-backed iterator into a BVEC-type iterator, they set ->bv_need_unpin to note whether they need to unpin the pages later. However, in both cases they examine the BVEC-type iterator and not the source iterator - and so bv_need_unpin doesn't get set and the pages are leaked. I think this may be responsible for the generic/208 xfstest failing occasionally with: WARNING: CPU: 0 PID: 3064 at mm/gup.c:218 try_grab_page+0x65/0x100 RIP: 0010:try_grab_page+0x65/0x100 follow_page_pte+0x1a7/0x570 __get_user_pages+0x1a2/0x650 __gup_longterm_locked+0xdc/0xb50 internal_get_user_pages_fast+0x17f/0x310 pin_user_pages_fast+0x46/0x60 iov_iter_extract_pages+0xc9/0x510 ? __kmalloc_large_node+0xb1/0x120 ? __kmalloc_node+0xbe/0x130 netfs_extract_user_iter+0xbf/0x200 [netfs] __cifs_writev+0x150/0x330 [cifs] vfs_write+0x2a8/0x3c0 ksys_pwrite64+0x65/0xa0 with the page refcount going negative. This is less unlikely than it seems because the page is being pinned, not simply got, and so the refcount increased by 1024 each time, and so only needs to be called around ~2097152 for the refcount to go negative. Further, the test program (aio-dio-invalidate-failure) uses a 32MiB static buffer and all the PTEs covering it refer to the same page because it's never written to. The warning in try_grab_page(): if (WARN_ON_ONCE(folio_ref_count(folio) <= 0)) return -ENOMEM; then trips and prevents us ever using the page again for DIO at least. Fixes: d08089f649a0 ("cifs: Change the I/O paths to use an iterator rather than a page list") Reported-by: Murphy Zhou Link: https://lore.kernel.org/r/CAH2r5mvaTsJ---n=265a4zqRA7pP+o4MJ36WCQUS6oPrOij8cw@mail.gmail.com Signed-off-by: David Howells Reviewed-by: Paulo Alcantara (SUSE) cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org Signed-off-by: Steve French --- fs/cifs/file.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/cifs/file.c b/fs/cifs/file.c index ec0694a65c7b6..4d4a2d82636d2 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -3612,7 +3612,7 @@ static ssize_t __cifs_writev( ctx->nr_pinned_pages = rc; ctx->bv = (void *)ctx->iter.bvec; - ctx->bv_need_unpin = iov_iter_extract_will_pin(&ctx->iter); + ctx->bv_need_unpin = iov_iter_extract_will_pin(from); } else if ((iov_iter_is_bvec(from) || iov_iter_is_kvec(from)) && !is_sync_kiocb(iocb)) { /* @@ -4148,7 +4148,7 @@ static ssize_t __cifs_readv( ctx->nr_pinned_pages = rc; ctx->bv = (void *)ctx->iter.bvec; - ctx->bv_need_unpin = iov_iter_extract_will_pin(&ctx->iter); + ctx->bv_need_unpin = iov_iter_extract_will_pin(to); ctx->should_dirty = true; } else if ((iov_iter_is_bvec(to) || iov_iter_is_kvec(to)) && !is_sync_kiocb(iocb)) { From 61fc1ee8be26bc192d691932b0a67eabee45d12f Mon Sep 17 00:00:00 2001 From: Alexandre Ghiti Date: Tue, 16 Mar 2021 15:34:20 -0400 Subject: [PATCH 638/765] riscv: Bump COMMAND_LINE_SIZE value to 1024 Increase COMMAND_LINE_SIZE as the current default value is too low for syzbot kernel command line. There has been considerable discussion on this patch that has led to a larger patch set removing COMMAND_LINE_SIZE from the uapi headers on all ports. That's not quite done yet, but it's gotten far enough we're confident this is not a uABI change so this is safe. Reported-by: Dmitry Vyukov Signed-off-by: Alexandre Ghiti Link: https://lore.kernel.org/r/20210316193420.904-1-alex@ghiti.fr [Palmer: it's not uabi] Link: https://lore.kernel.org/linux-riscv/874b8076-b0d1-4aaa-bcd8-05d523060152@app.fastmail.com/#t Signed-off-by: Palmer Dabbelt --- arch/riscv/include/uapi/asm/setup.h | 8 ++++++++ 1 file changed, 8 insertions(+) create mode 100644 arch/riscv/include/uapi/asm/setup.h diff --git a/arch/riscv/include/uapi/asm/setup.h b/arch/riscv/include/uapi/asm/setup.h new file mode 100644 index 0000000000000..66b13a5228808 --- /dev/null +++ b/arch/riscv/include/uapi/asm/setup.h @@ -0,0 +1,8 @@ +/* SPDX-License-Identifier: GPL-2.0-only WITH Linux-syscall-note */ + +#ifndef _UAPI_ASM_RISCV_SETUP_H +#define _UAPI_ASM_RISCV_SETUP_H + +#define COMMAND_LINE_SIZE 1024 + +#endif /* _UAPI_ASM_RISCV_SETUP_H */ From 05eacc198c68cbb35a7281ce4011f8899ee1cfb8 Mon Sep 17 00:00:00 2001 From: Mark Hawrylak Date: Sun, 19 Feb 2023 16:02:00 +1100 Subject: [PATCH 639/765] drm/radeon: Fix eDP for single-display iMac11,2 Apple iMac11,2 (mid 2010) also with Radeon HD-4670 that has the same issue as iMac10,1 (late 2009) where the internal eDP panel stays dark on driver load. This patch treats iMac11,2 the same as iMac10,1, so the eDP panel stays active. Additional steps: Kernel boot parameter radeon.nomodeset=0 required to keep the eDP panel active. This patch is an extension of commit 564d8a2cf3ab ("drm/radeon: Fix eDP for single-display iMac10,1 (v2)") Link: https://lore.kernel.org/all/lsq.1507553064.833262317@decadent.org.uk/ Signed-off-by: Mark Hawrylak Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org --- drivers/gpu/drm/radeon/atombios_encoders.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c index 1471c3a966020..4aca09cab4b8c 100644 --- a/drivers/gpu/drm/radeon/atombios_encoders.c +++ b/drivers/gpu/drm/radeon/atombios_encoders.c @@ -2123,11 +2123,12 @@ int radeon_atom_pick_dig_encoder(struct drm_encoder *encoder, int fe_idx) /* * On DCE32 any encoder can drive any block so usually just use crtc id, - * but Apple thinks different at least on iMac10,1, so there use linkb, + * but Apple thinks different at least on iMac10,1 and iMac11,2, so there use linkb, * otherwise the internal eDP panel will stay dark. */ if (ASIC_IS_DCE32(rdev)) { - if (dmi_match(DMI_PRODUCT_NAME, "iMac10,1")) + if (dmi_match(DMI_PRODUCT_NAME, "iMac10,1") || + dmi_match(DMI_PRODUCT_NAME, "iMac11,2")) enc_idx = (dig->linkb) ? 1 : 0; else enc_idx = radeon_crtc->crtc_id; From 1bf56f25258871db5bfad7aebe19e46148eda159 Mon Sep 17 00:00:00 2001 From: Candice Li Date: Fri, 24 Feb 2023 12:15:57 +0800 Subject: [PATCH 640/765] drm/amdgpu: Make umc_v8_10_convert_error_address static and remove unused variable Fixes following warnings: warning: no previous prototype for 'umc_v8_10_convert_error_address' warning: variable 'channel_index' set but not used Reported-by: kernel test robot Signed-off-by: Candice Li Reviewed-by: Tao Zhou Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/umc_v8_10.c | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c b/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c index 66158219f791c..fb55e8cb9967a 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c +++ b/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c @@ -209,10 +209,10 @@ static int umc_v8_10_swizzle_mode_na_to_pa(struct amdgpu_device *adev, return 0; } -void umc_v8_10_convert_error_address(struct amdgpu_device *adev, - struct ras_err_data *err_data, uint64_t err_addr, - uint32_t ch_inst, uint32_t umc_inst, - uint32_t node_inst, uint64_t mc_umc_status) +static void umc_v8_10_convert_error_address(struct amdgpu_device *adev, + struct ras_err_data *err_data, uint64_t err_addr, + uint32_t ch_inst, uint32_t umc_inst, + uint32_t node_inst, uint64_t mc_umc_status) { uint64_t na_err_addr_base; uint64_t na_err_addr, retired_page_addr; @@ -434,7 +434,7 @@ static void umc_v8_10_ecc_info_query_error_address(struct amdgpu_device *adev, uint32_t umc_inst, uint32_t node_inst) { - uint32_t eccinfo_table_idx, channel_index; + uint32_t eccinfo_table_idx; uint64_t mc_umc_status, err_addr; struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); @@ -443,11 +443,6 @@ static void umc_v8_10_ecc_info_query_error_address(struct amdgpu_device *adev, adev->umc.channel_inst_num + umc_inst * adev->umc.channel_inst_num + ch_inst; - channel_index = - adev->umc.channel_idx_tbl[node_inst * adev->umc.umc_inst_num * - adev->umc.channel_inst_num + - umc_inst * adev->umc.channel_inst_num + - ch_inst]; mc_umc_status = ras->umc_ecc.ecc[eccinfo_table_idx].mca_umc_status; From c76e483cd9163138e8fc44d829c986819f072d4f Mon Sep 17 00:00:00 2001 From: Harry Wentland Date: Mon, 12 Dec 2022 13:02:25 -0500 Subject: [PATCH 641/765] drm/amd/display: Don't restrict bpc to 8 bpc MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This will let us pass the kms_hdr.bpc_switch IGT test. The reason the bpc restriction was required is historical. At one point in time we were not falling back to a lower bpc when we didn't have enough bandwidth for the maximum bpc reported by a display. This meant that we couldn't enable some high refresh modes unless we limitted the bpc. Starting with this patch the issue is fixed: commit cbd14ae7ea93 ("drm/amd/display: Fix incorrectly pruned modes with deep color") This patch implemented a fallback mechanism if mode validation failed at the max bpc. This means users now automatically get all modes that can be supported by at least 6 bpc. The driver will enable the mode with the highest possible bpc that is supported by the display. v2: - explain why this is no longer needed (Michel) - refer to commit that fixed bpc fallback (Michel) Signed-off-by: Harry Wentland Cc: Pekka Paalanen Cc: Sebastian Wick Cc: Vitaly.Prosyak@amd.com Cc: Joshua Ashton Cc: dri-devel@lists.freedesktop.org Cc: amd-gfx@lists.freedesktop.org Cc: Michel Dänzer Reviewed-by: Joshua Ashton Reviewed-by: Michel Dänzer Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index c420bce47acb5..700d170d52c04 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -7235,7 +7235,7 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm, drm_connector_attach_max_bpc_property(&aconnector->base, 8, 16); /* This defaults to the max in the range, but we want 8bpc for non-edp. */ - aconnector->base.state->max_bpc = (connector_type == DRM_MODE_CONNECTOR_eDP) ? 16 : 8; + aconnector->base.state->max_bpc = 16; aconnector->base.state->max_requested_bpc = aconnector->base.state->max_bpc; if (connector_type == DRM_MODE_CONNECTOR_eDP && From 65a24000808f70ac69bd2a96381fa0c7341f20c0 Mon Sep 17 00:00:00 2001 From: Mario Limonciello Date: Sun, 19 Feb 2023 23:04:04 -0600 Subject: [PATCH 642/765] drm/amd: Fix initialization for nbio 7.5.1 A mistake has been made in the BIOS for some ASICs with NBIO 7.5.1 where some NBIO registers aren't properly setup. Ensure that they're set during initialization. Tested-by: Richard Gong Signed-off-by: Mario Limonciello Acked-by: Alex Deucher Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org # 6.1.x --- drivers/gpu/drm/amd/amdgpu/nbio_v7_2.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_2.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_2.c index 31776b12e4c45..4b0d563c6522c 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_2.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_2.c @@ -382,6 +382,11 @@ static void nbio_v7_2_init_registers(struct amdgpu_device *adev) if (def != data) WREG32_PCIE_PORT(SOC15_REG_OFFSET(NBIO, 0, regBIF1_PCIE_MST_CTRL_3), data); break; + case IP_VERSION(7, 5, 1): + data = RREG32_SOC15(NBIO, 0, regRCC_DEV2_EPF0_STRAP2); + data &= ~RCC_DEV2_EPF0_STRAP2__STRAP_NO_SOFT_RESET_DEV2_F0_MASK; + WREG32_SOC15(NBIO, 0, regRCC_DEV2_EPF0_STRAP2, data); + fallthrough; default: def = data = RREG32_PCIE_PORT(SOC15_REG_OFFSET(NBIO, 0, regPCIE_CONFIG_CNTL)); data = REG_SET_FIELD(data, PCIE_CONFIG_CNTL, From ca87c9ae70566c651dcf09c1b080db259e20f9ee Mon Sep 17 00:00:00 2001 From: tiancyin Date: Wed, 8 Feb 2023 14:10:04 +0800 Subject: [PATCH 643/765] drm/amd/display: fix dm irq error message in gpu recover [Why] Variable adev->crtc_irq.num_types was initialized as the value of adev->mode_info.num_crtc at early_init stage, later at hw_init stage, the num_crtc changed due to the display pipe harvest on some SKUs, but the num_types was not updated accordingly, that cause below error in gpu recover. *ERROR* amdgpu_dm_set_crtc_irq_state: crtc is NULL at id :3 *ERROR* amdgpu_dm_set_crtc_irq_state: crtc is NULL at id :3 *ERROR* amdgpu_dm_set_crtc_irq_state: crtc is NULL at id :3 *ERROR* amdgpu_dm_set_pflip_irq_state: crtc is NULL at id :3 *ERROR* amdgpu_dm_set_pflip_irq_state: crtc is NULL at id :3 *ERROR* amdgpu_dm_set_pflip_irq_state: crtc is NULL at id :3 *ERROR* amdgpu_dm_set_pflip_irq_state: crtc is NULL at id :3 *ERROR* amdgpu_dm_set_vupdate_irq_state: crtc is NULL at id :3 *ERROR* amdgpu_dm_set_vupdate_irq_state: crtc is NULL at id :3 *ERROR* amdgpu_dm_set_vupdate_irq_state: crtc is NULL at id :3 [How] Defer the initialization of num_types to eliminate the error logs. Signed-off-by: tiancyin Reviewed-by: Hamza Mahfooz Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 700d170d52c04..79ae01ac4b19e 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -4265,6 +4265,8 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) /* Update the actual used number of crtc */ adev->mode_info.num_crtc = adev->dm.display_indexes_num; + amdgpu_dm_set_irq_funcs(adev); + link_cnt = dm->dc->caps.max_links; if (amdgpu_dm_mode_config_init(dm->adev)) { DRM_ERROR("DM: Failed to initialize mode config\n"); @@ -4757,8 +4759,6 @@ static int dm_early_init(void *handle) break; } - amdgpu_dm_set_irq_funcs(adev); - if (adev->mode_info.funcs == NULL) adev->mode_info.funcs = &dm_display_funcs; From cca3306488f71465f8c5e920e5a4e24fa461c72b Mon Sep 17 00:00:00 2001 From: Tom Rix Date: Fri, 24 Feb 2023 11:45:19 -0500 Subject: [PATCH 644/765] drm/amdgpu: remove unused variable ring MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit building with gcc and W=1 reports drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c:81:29: error: variable ‘ring’ set but not used [-Werror=unused-but-set-variable] 81 | struct amdgpu_ring *ring; | ^~~~ ring is not used so remove it. Signed-off-by: Tom Rix Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c index 213b43670f23e..023a1fffa6a9b 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c @@ -78,12 +78,10 @@ static void vcn_v4_0_set_ras_funcs(struct amdgpu_device *adev); static int vcn_v4_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring; if (amdgpu_sriov_vf(adev)) { adev->vcn.harvest_config = VCN_HARVEST_MMSCH; for (int i = 0; i < adev->vcn.num_vcn_inst; ++i) { - ring = &adev->vcn.inst[i].ring_enc[0]; if (amdgpu_vcn_is_disabled_vcn(adev, VCN_ENCODE_RING, i)) { adev->vcn.harvest_config |= 1 << i; dev_info(adev->dev, "VCN%d is disabled by hypervisor\n", i); From 23f4a2d29ba57bf88095f817de5809d427fcbe7e Mon Sep 17 00:00:00 2001 From: Horatio Zhang Date: Fri, 24 Feb 2023 13:55:44 +0800 Subject: [PATCH 645/765] drm/amdgpu: fix ttm_bo calltrace warning in psp_hw_fini The call trace occurs when the amdgpu is removed after the mode1 reset. During mode1 reset, from suspend to resume, there is no need to reinitialize the ta firmware buffer which caused the bo pin_count increase redundantly. [ 489.885525] Call Trace: [ 489.885525] [ 489.885526] amdttm_bo_put+0x34/0x50 [amdttm] [ 489.885529] amdgpu_bo_free_kernel+0xe8/0x130 [amdgpu] [ 489.885620] psp_free_shared_bufs+0xb7/0x150 [amdgpu] [ 489.885720] psp_hw_fini+0xce/0x170 [amdgpu] [ 489.885815] amdgpu_device_fini_hw+0x2ff/0x413 [amdgpu] [ 489.885960] ? blocking_notifier_chain_unregister+0x56/0xb0 [ 489.885962] amdgpu_driver_unload_kms+0x51/0x60 [amdgpu] [ 489.886049] amdgpu_pci_remove+0x5a/0x140 [amdgpu] [ 489.886132] ? __pm_runtime_resume+0x60/0x90 [ 489.886134] pci_device_remove+0x3e/0xb0 [ 489.886135] __device_release_driver+0x1ab/0x2a0 [ 489.886137] driver_detach+0xf3/0x140 [ 489.886138] bus_remove_driver+0x6c/0xf0 [ 489.886140] driver_unregister+0x31/0x60 [ 489.886141] pci_unregister_driver+0x40/0x90 [ 489.886142] amdgpu_exit+0x15/0x451 [amdgpu] Signed-off-by: Horatio Zhang Signed-off-by: longlyao Reviewed-by: Guchun Chen Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index 15e601f096486..28fe6d9410540 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -1683,7 +1683,7 @@ static int psp_hdcp_initialize(struct psp_context *psp) psp->hdcp_context.context.mem_context.shared_mem_size = PSP_HDCP_SHARED_MEM_SIZE; psp->hdcp_context.context.ta_load_type = GFX_CMD_ID_LOAD_TA; - if (!psp->hdcp_context.context.initialized) { + if (!psp->hdcp_context.context.mem_context.shared_buf) { ret = psp_ta_init_shared_buf(psp, &psp->hdcp_context.context.mem_context); if (ret) return ret; @@ -1750,7 +1750,7 @@ static int psp_dtm_initialize(struct psp_context *psp) psp->dtm_context.context.mem_context.shared_mem_size = PSP_DTM_SHARED_MEM_SIZE; psp->dtm_context.context.ta_load_type = GFX_CMD_ID_LOAD_TA; - if (!psp->dtm_context.context.initialized) { + if (!psp->dtm_context.context.mem_context.shared_buf) { ret = psp_ta_init_shared_buf(psp, &psp->dtm_context.context.mem_context); if (ret) return ret; @@ -1818,7 +1818,7 @@ static int psp_rap_initialize(struct psp_context *psp) psp->rap_context.context.mem_context.shared_mem_size = PSP_RAP_SHARED_MEM_SIZE; psp->rap_context.context.ta_load_type = GFX_CMD_ID_LOAD_TA; - if (!psp->rap_context.context.initialized) { + if (!psp->rap_context.context.mem_context.shared_buf) { ret = psp_ta_init_shared_buf(psp, &psp->rap_context.context.mem_context); if (ret) return ret; From 01a18aa309aec12461fb5e6aecb76f8b33810658 Mon Sep 17 00:00:00 2001 From: Ryan Lin Date: Tue, 7 Feb 2023 23:03:48 +0800 Subject: [PATCH 646/765] drm/amd/display: Ext displays with dock can't recognized after resume [Why] Needs to set the default value of the LTTPR timeout after resume. [How] Set the default (3.2ms) timeout at resuming if the sink supports LTTPR Reviewed-by: Jerry Zuo Acked-by: Qingqing Zhuo Signed-off-by: Ryan Lin Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 10 ++++++++++ .../gpu/drm/amd/display/dc/link/protocols/link_ddc.h | 1 + .../amd/display/dc/link/protocols/link_dp_capability.c | 2 -- 3 files changed, 11 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 79ae01ac4b19e..009ef917dad47 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -41,6 +41,8 @@ #include "dpcd_defs.h" #include "link/protocols/link_dpcd.h" #include "link_service_types.h" +#include "link/protocols/link_dp_capability.h" +#include "link/protocols/link_ddc.h" #include "vid.h" #include "amdgpu.h" @@ -2302,6 +2304,14 @@ static void s3_handle_mst(struct drm_device *dev, bool suspend) if (suspend) { drm_dp_mst_topology_mgr_suspend(mgr); } else { + /* if extended timeout is supported in hardware, + * default to LTTPR timeout (3.2ms) first as a W/A for DP link layer + * CTS 4.2.1.1 regression introduced by CTS specs requirement update. + */ + try_to_configure_aux_timeout(aconnector->dc_link->ddc, LINK_AUX_DEFAULT_LTTPR_TIMEOUT_PERIOD); + if (!dp_is_lttpr_present(aconnector->dc_link)) + try_to_configure_aux_timeout(aconnector->dc_link->ddc, LINK_AUX_DEFAULT_TIMEOUT_PERIOD); + ret = drm_dp_mst_topology_mgr_resume(mgr, true); if (ret < 0) { dm_helpers_dp_mst_stop_top_mgr(aconnector->dc_link->ctx, diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.h index 86e9d2e886d6f..aaa5064408ba4 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.h +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.h @@ -33,6 +33,7 @@ #define DPVGA_DONGLE_AUX_DEFER_WA_DELAY 40 #define I2C_OVER_AUX_DEFER_WA_DELAY_1MS 1 #define LINK_AUX_DEFAULT_LTTPR_TIMEOUT_PERIOD 3200 /*us*/ +#define LINK_AUX_DEFAULT_TIMEOUT_PERIOD 552 /*us*/ #define EDID_SEGMENT_SIZE 256 diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c index 4874d1bf1dcb0..d4370856f164a 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c @@ -60,8 +60,6 @@ #define MIN(X, Y) ((X) < (Y) ? (X) : (Y)) #endif -#define LINK_AUX_DEFAULT_TIMEOUT_PERIOD 552 /*us*/ - struct dp_lt_fallback_entry { enum dc_lane_count lane_count; enum dc_link_rate link_rate; From 031f196d1b1b6d5dfcb0533b431e3ab1750e6189 Mon Sep 17 00:00:00 2001 From: Alex Hung Date: Wed, 11 Jan 2023 09:54:11 -0700 Subject: [PATCH 647/765] drm/amd/display: fix shift-out-of-bounds in CalculateVMAndRowBytes [WHY] When PTEBufferSizeInRequests is zero, UBSAN reports the following warning because dml_log2 returns an unexpected negative value: shift exponent 4294966273 is too large for 32-bit type 'int' [HOW] In the case PTEBufferSizeInRequests is zero, skip the dml_log2() and assign the result directly. Reviewed-by: Jun Lei Acked-by: Qingqing Zhuo Signed-off-by: Alex Hung Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher --- .../gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c index 379729b028474..c3d75e56410cc 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c @@ -1802,7 +1802,10 @@ static unsigned int CalculateVMAndRowBytes( } if (SurfaceTiling == dm_sw_linear) { - *dpte_row_height = dml_min(128, 1 << (unsigned int) dml_floor(dml_log2(PTEBufferSizeInRequests * *PixelPTEReqWidth / Pitch), 1)); + if (PTEBufferSizeInRequests == 0) + *dpte_row_height = 1; + else + *dpte_row_height = dml_min(128, 1 << (unsigned int) dml_floor(dml_log2(PTEBufferSizeInRequests * *PixelPTEReqWidth / Pitch), 1)); *dpte_row_width_ub = (dml_ceil(((double) SwathWidth - 1) / *PixelPTEReqWidth, 1) + 1) * *PixelPTEReqWidth; *PixelPTEBytesPerRow = *dpte_row_width_ub / *PixelPTEReqWidth * *PTERequestSize; } else if (ScanDirection != dm_vert) { From 1fa0d424a1d50aebbd87d40a0cb41995ba336f27 Mon Sep 17 00:00:00 2001 From: Aric Cyr Date: Thu, 9 Feb 2023 20:03:33 -0500 Subject: [PATCH 648/765] Revert "drm/amd/display: Do not set DRR on pipe commit" This reverts commit 4f1b5e739dfd1edde33329e3f376733a131fb1ff. [Why & How] Original change causes a regression. Revert until fix is available. Reviewed-by: Aric Cyr Acked-by: Qingqing Zhuo Signed-off-by: Aric Cyr Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c index df787fcf8e86e..3b4d4d68359bb 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c @@ -998,5 +998,8 @@ void dcn30_prepare_bandwidth(struct dc *dc, dc->clk_mgr->funcs->set_max_memclk(dc->clk_mgr, dc->clk_mgr->bw_params->clk_table.entries[dc->clk_mgr->bw_params->clk_table.num_entries - 1].memclk_mhz); dcn20_prepare_bandwidth(dc, context); + + dc_dmub_srv_p_state_delegate(dc, + context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching, context); } From a8af68f79d149796609a679b00a34762249c6a5b Mon Sep 17 00:00:00 2001 From: Sung Joon Kim Date: Fri, 10 Feb 2023 14:39:49 -0500 Subject: [PATCH 649/765] drm/amd/display: Extend Freesync over PCon support for more devices [why] More branch devices are able to support Freesync over PCon so include them in the list of supporting devices. [how] Add more compatible PCon devices in the whitelist for Freesync over Pcon. Reviewed-by: Harry Wentland Acked-by: Qingqing Zhuo Signed-off-by: Sung Joon Kim Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index 6fdc2027c2b47..1583157da355b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -1149,6 +1149,8 @@ static bool dm_is_freesync_pcon_whitelist(const uint32_t branch_dev_id) switch (branch_dev_id) { case DP_BRANCH_DEVICE_ID_0060AD: + case DP_BRANCH_DEVICE_ID_00E04C: + case DP_BRANCH_DEVICE_ID_90CC24: ret_val = true; break; default: From 6bb811d0ee3e1fe9f22a028c89b3472c999b70bc Mon Sep 17 00:00:00 2001 From: bobzhou Date: Mon, 27 Feb 2023 15:30:54 +0800 Subject: [PATCH 650/765] drm/amdgpu/vcn: fix compilation issue with legacy gcc MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch is used to fix following compilation issue with legacy gcc error: ‘for’ loop initial declarations are only allowed in C99 mode for (int i = 0; i < adev->vcn.num_vcn_inst; ++i) { Signed-off-by: bobzhou Reviewed-by: Guchun Chen Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c index 023a1fffa6a9b..43d587404c3e1 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c @@ -78,10 +78,11 @@ static void vcn_v4_0_set_ras_funcs(struct amdgpu_device *adev); static int vcn_v4_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int i; if (amdgpu_sriov_vf(adev)) { adev->vcn.harvest_config = VCN_HARVEST_MMSCH; - for (int i = 0; i < adev->vcn.num_vcn_inst; ++i) { + for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { if (amdgpu_vcn_is_disabled_vcn(adev, VCN_ENCODE_RING, i)) { adev->vcn.harvest_config |= 1 << i; dev_info(adev->dev, "VCN%d is disabled by hypervisor\n", i); From 49c47cc21b5b7a3d8deb18fc57b0aa2ab1286962 Mon Sep 17 00:00:00 2001 From: Hangyu Hua Date: Tue, 28 Feb 2023 10:33:44 +0800 Subject: [PATCH 651/765] net: tls: fix possible race condition between do_tls_getsockopt_conf() and do_tls_setsockopt_conf() ctx->crypto_send.info is not protected by lock_sock in do_tls_getsockopt_conf(). A race condition between do_tls_getsockopt_conf() and error paths of do_tls_setsockopt_conf() may lead to a use-after-free or null-deref. More discussion: https://lore.kernel.org/all/Y/ht6gQL+u6fj3dG@hog/ Fixes: 3c4d7559159b ("tls: kernel TLS support") Signed-off-by: Hangyu Hua Link: https://lore.kernel.org/r/20230228023344.9623-1-hbh25y@gmail.com Signed-off-by: Jakub Kicinski --- net/tls/tls_main.c | 23 +++++------------------ 1 file changed, 5 insertions(+), 18 deletions(-) diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c index 3735cb00905df..b32c112984dd9 100644 --- a/net/tls/tls_main.c +++ b/net/tls/tls_main.c @@ -405,13 +405,11 @@ static int do_tls_getsockopt_conf(struct sock *sk, char __user *optval, rc = -EINVAL; goto out; } - lock_sock(sk); memcpy(crypto_info_aes_gcm_128->iv, cctx->iv + TLS_CIPHER_AES_GCM_128_SALT_SIZE, TLS_CIPHER_AES_GCM_128_IV_SIZE); memcpy(crypto_info_aes_gcm_128->rec_seq, cctx->rec_seq, TLS_CIPHER_AES_GCM_128_REC_SEQ_SIZE); - release_sock(sk); if (copy_to_user(optval, crypto_info_aes_gcm_128, sizeof(*crypto_info_aes_gcm_128))) @@ -429,13 +427,11 @@ static int do_tls_getsockopt_conf(struct sock *sk, char __user *optval, rc = -EINVAL; goto out; } - lock_sock(sk); memcpy(crypto_info_aes_gcm_256->iv, cctx->iv + TLS_CIPHER_AES_GCM_256_SALT_SIZE, TLS_CIPHER_AES_GCM_256_IV_SIZE); memcpy(crypto_info_aes_gcm_256->rec_seq, cctx->rec_seq, TLS_CIPHER_AES_GCM_256_REC_SEQ_SIZE); - release_sock(sk); if (copy_to_user(optval, crypto_info_aes_gcm_256, sizeof(*crypto_info_aes_gcm_256))) @@ -451,13 +447,11 @@ static int do_tls_getsockopt_conf(struct sock *sk, char __user *optval, rc = -EINVAL; goto out; } - lock_sock(sk); memcpy(aes_ccm_128->iv, cctx->iv + TLS_CIPHER_AES_CCM_128_SALT_SIZE, TLS_CIPHER_AES_CCM_128_IV_SIZE); memcpy(aes_ccm_128->rec_seq, cctx->rec_seq, TLS_CIPHER_AES_CCM_128_REC_SEQ_SIZE); - release_sock(sk); if (copy_to_user(optval, aes_ccm_128, sizeof(*aes_ccm_128))) rc = -EFAULT; break; @@ -472,13 +466,11 @@ static int do_tls_getsockopt_conf(struct sock *sk, char __user *optval, rc = -EINVAL; goto out; } - lock_sock(sk); memcpy(chacha20_poly1305->iv, cctx->iv + TLS_CIPHER_CHACHA20_POLY1305_SALT_SIZE, TLS_CIPHER_CHACHA20_POLY1305_IV_SIZE); memcpy(chacha20_poly1305->rec_seq, cctx->rec_seq, TLS_CIPHER_CHACHA20_POLY1305_REC_SEQ_SIZE); - release_sock(sk); if (copy_to_user(optval, chacha20_poly1305, sizeof(*chacha20_poly1305))) rc = -EFAULT; @@ -493,13 +485,11 @@ static int do_tls_getsockopt_conf(struct sock *sk, char __user *optval, rc = -EINVAL; goto out; } - lock_sock(sk); memcpy(sm4_gcm_info->iv, cctx->iv + TLS_CIPHER_SM4_GCM_SALT_SIZE, TLS_CIPHER_SM4_GCM_IV_SIZE); memcpy(sm4_gcm_info->rec_seq, cctx->rec_seq, TLS_CIPHER_SM4_GCM_REC_SEQ_SIZE); - release_sock(sk); if (copy_to_user(optval, sm4_gcm_info, sizeof(*sm4_gcm_info))) rc = -EFAULT; break; @@ -513,13 +503,11 @@ static int do_tls_getsockopt_conf(struct sock *sk, char __user *optval, rc = -EINVAL; goto out; } - lock_sock(sk); memcpy(sm4_ccm_info->iv, cctx->iv + TLS_CIPHER_SM4_CCM_SALT_SIZE, TLS_CIPHER_SM4_CCM_IV_SIZE); memcpy(sm4_ccm_info->rec_seq, cctx->rec_seq, TLS_CIPHER_SM4_CCM_REC_SEQ_SIZE); - release_sock(sk); if (copy_to_user(optval, sm4_ccm_info, sizeof(*sm4_ccm_info))) rc = -EFAULT; break; @@ -535,13 +523,11 @@ static int do_tls_getsockopt_conf(struct sock *sk, char __user *optval, rc = -EINVAL; goto out; } - lock_sock(sk); memcpy(crypto_info_aria_gcm_128->iv, cctx->iv + TLS_CIPHER_ARIA_GCM_128_SALT_SIZE, TLS_CIPHER_ARIA_GCM_128_IV_SIZE); memcpy(crypto_info_aria_gcm_128->rec_seq, cctx->rec_seq, TLS_CIPHER_ARIA_GCM_128_REC_SEQ_SIZE); - release_sock(sk); if (copy_to_user(optval, crypto_info_aria_gcm_128, sizeof(*crypto_info_aria_gcm_128))) @@ -559,13 +545,11 @@ static int do_tls_getsockopt_conf(struct sock *sk, char __user *optval, rc = -EINVAL; goto out; } - lock_sock(sk); memcpy(crypto_info_aria_gcm_256->iv, cctx->iv + TLS_CIPHER_ARIA_GCM_256_SALT_SIZE, TLS_CIPHER_ARIA_GCM_256_IV_SIZE); memcpy(crypto_info_aria_gcm_256->rec_seq, cctx->rec_seq, TLS_CIPHER_ARIA_GCM_256_REC_SEQ_SIZE); - release_sock(sk); if (copy_to_user(optval, crypto_info_aria_gcm_256, sizeof(*crypto_info_aria_gcm_256))) @@ -614,11 +598,9 @@ static int do_tls_getsockopt_no_pad(struct sock *sk, char __user *optval, if (len < sizeof(value)) return -EINVAL; - lock_sock(sk); value = -EINVAL; if (ctx->rx_conf == TLS_SW || ctx->rx_conf == TLS_HW) value = ctx->rx_no_pad; - release_sock(sk); if (value < 0) return value; @@ -635,6 +617,8 @@ static int do_tls_getsockopt(struct sock *sk, int optname, { int rc = 0; + lock_sock(sk); + switch (optname) { case TLS_TX: case TLS_RX: @@ -651,6 +635,9 @@ static int do_tls_getsockopt(struct sock *sk, int optname, rc = -ENOPROTOOPT; break; } + + release_sock(sk); + return rc; } From f3221361dc85d4de22586ce8441ec2c67b454f5d Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 28 Feb 2023 16:28:57 -0800 Subject: [PATCH 652/765] net: tls: avoid hanging tasks on the tx_lock syzbot sent a hung task report and Eric explains that adversarial receiver may keep RWIN at 0 for a long time, so we are not guaranteed to make forward progress. Thread which took tx_lock and went to sleep may not release tx_lock for hours. Use interruptible sleep where possible and reschedule the work if it can't take the lock. Testing: existing selftest passes Reported-by: syzbot+9c0268252b8ef967c62e@syzkaller.appspotmail.com Fixes: 79ffe6087e91 ("net/tls: add a TX lock") Link: https://lore.kernel.org/all/000000000000e412e905f5b46201@google.com/ Cc: stable@vger.kernel.org # wait 4 weeks Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/r/20230301002857.2101894-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/tls/tls_sw.c | 26 +++++++++++++++++++------- 1 file changed, 19 insertions(+), 7 deletions(-) diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index 021d760f91335..635b8bf6b937c 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -956,7 +956,9 @@ int tls_sw_sendmsg(struct sock *sk, struct msghdr *msg, size_t size) MSG_CMSG_COMPAT)) return -EOPNOTSUPP; - mutex_lock(&tls_ctx->tx_lock); + ret = mutex_lock_interruptible(&tls_ctx->tx_lock); + if (ret) + return ret; lock_sock(sk); if (unlikely(msg->msg_controllen)) { @@ -1290,7 +1292,9 @@ int tls_sw_sendpage(struct sock *sk, struct page *page, MSG_SENDPAGE_NOTLAST | MSG_SENDPAGE_NOPOLICY)) return -EOPNOTSUPP; - mutex_lock(&tls_ctx->tx_lock); + ret = mutex_lock_interruptible(&tls_ctx->tx_lock); + if (ret) + return ret; lock_sock(sk); ret = tls_sw_do_sendpage(sk, page, offset, size, flags); release_sock(sk); @@ -2435,11 +2439,19 @@ static void tx_work_handler(struct work_struct *work) if (!test_and_clear_bit(BIT_TX_SCHEDULED, &ctx->tx_bitmask)) return; - mutex_lock(&tls_ctx->tx_lock); - lock_sock(sk); - tls_tx_records(sk, -1); - release_sock(sk); - mutex_unlock(&tls_ctx->tx_lock); + + if (mutex_trylock(&tls_ctx->tx_lock)) { + lock_sock(sk); + tls_tx_records(sk, -1); + release_sock(sk); + mutex_unlock(&tls_ctx->tx_lock); + } else if (!test_and_set_bit(BIT_TX_SCHEDULED, &ctx->tx_bitmask)) { + /* Someone is holding the tx_lock, they will likely run Tx + * and cancel the work on their way out of the lock section. + * Schedule a long delay just in case. + */ + schedule_delayed_work(&ctx->tx_work.work, msecs_to_jiffies(10)); + } } static bool tls_is_tx_ready(struct tls_sw_context_tx *ctx) From ff75e4eb71cb1b64da07a5a7e8140e202b365887 Mon Sep 17 00:00:00 2001 From: Sergio Paracuellos Date: Tue, 14 Feb 2023 11:39:32 +0100 Subject: [PATCH 653/765] dt-bindings: watchdog: mt7621-wdt: add phandle to access system controller registers MT7621 SoC provides a system controller node for accessing to some registers. Add a phandle in this node to avoid using MIPS related arch operations and includes in watchdog driver code. Acked-by: Krzysztof Kozlowski Signed-off-by: Sergio Paracuellos Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20230214103936.1061078-2-sergio.paracuellos@gmail.com Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- .../devicetree/bindings/watchdog/mediatek,mt7621-wdt.yaml | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/Documentation/devicetree/bindings/watchdog/mediatek,mt7621-wdt.yaml b/Documentation/devicetree/bindings/watchdog/mediatek,mt7621-wdt.yaml index b2b17fdf4e398..a668d0c2f14b8 100644 --- a/Documentation/devicetree/bindings/watchdog/mediatek,mt7621-wdt.yaml +++ b/Documentation/devicetree/bindings/watchdog/mediatek,mt7621-wdt.yaml @@ -19,6 +19,12 @@ properties: reg: maxItems: 1 + mediatek,sysctl: + $ref: /schemas/types.yaml#/definitions/phandle + description: + phandle to system controller 'sysc' syscon node which + controls system registers + required: - compatible - reg @@ -30,4 +36,5 @@ examples: watchdog@100 { compatible = "mediatek,mt7621-wdt"; reg = <0x100 0x100>; + mediatek,sysctl = <&sysc>; }; From 783c7cb4659b53b5e1b809dac5e8cdf250145919 Mon Sep 17 00:00:00 2001 From: Sergio Paracuellos Date: Tue, 14 Feb 2023 11:39:35 +0100 Subject: [PATCH 654/765] watchdog: mt7621-wdt: avoid static global declarations Instead of using static global definitions in driver code, refactor code introducing a new watchdog driver data structure and use it along the code. Reviewed-by: Guenter Roeck Signed-off-by: Sergio Paracuellos Link: https://lore.kernel.org/r/20230214103936.1061078-5-sergio.paracuellos@gmail.com [groeck: unsigned -> unsigned int] Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/mt7621_wdt.c | 102 ++++++++++++++++++++++------------ 1 file changed, 65 insertions(+), 37 deletions(-) diff --git a/drivers/watchdog/mt7621_wdt.c b/drivers/watchdog/mt7621_wdt.c index a8aa3522cfda8..b02c1415fc716 100644 --- a/drivers/watchdog/mt7621_wdt.c +++ b/drivers/watchdog/mt7621_wdt.c @@ -31,8 +31,11 @@ #define TMR1CTL_RESTART BIT(9) #define TMR1CTL_PRESCALE_SHIFT 16 -static void __iomem *mt7621_wdt_base; -static struct reset_control *mt7621_wdt_reset; +struct mt7621_wdt_data { + void __iomem *base; + struct reset_control *rst; + struct watchdog_device wdt; +}; static bool nowayout = WATCHDOG_NOWAYOUT; module_param(nowayout, bool, 0); @@ -40,27 +43,31 @@ MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default=" __MODULE_STRING(WATCHDOG_NOWAYOUT) ")"); -static inline void rt_wdt_w32(unsigned reg, u32 val) +static inline void rt_wdt_w32(void __iomem *base, unsigned int reg, u32 val) { - iowrite32(val, mt7621_wdt_base + reg); + iowrite32(val, base + reg); } -static inline u32 rt_wdt_r32(unsigned reg) +static inline u32 rt_wdt_r32(void __iomem *base, unsigned int reg) { - return ioread32(mt7621_wdt_base + reg); + return ioread32(base + reg); } static int mt7621_wdt_ping(struct watchdog_device *w) { - rt_wdt_w32(TIMER_REG_TMRSTAT, TMR1CTL_RESTART); + struct mt7621_wdt_data *drvdata = watchdog_get_drvdata(w); + + rt_wdt_w32(drvdata->base, TIMER_REG_TMRSTAT, TMR1CTL_RESTART); return 0; } static int mt7621_wdt_set_timeout(struct watchdog_device *w, unsigned int t) { + struct mt7621_wdt_data *drvdata = watchdog_get_drvdata(w); + w->timeout = t; - rt_wdt_w32(TIMER_REG_TMR1LOAD, t * 1000); + rt_wdt_w32(drvdata->base, TIMER_REG_TMR1LOAD, t * 1000); mt7621_wdt_ping(w); return 0; @@ -68,29 +75,31 @@ static int mt7621_wdt_set_timeout(struct watchdog_device *w, unsigned int t) static int mt7621_wdt_start(struct watchdog_device *w) { + struct mt7621_wdt_data *drvdata = watchdog_get_drvdata(w); u32 t; /* set the prescaler to 1ms == 1000us */ - rt_wdt_w32(TIMER_REG_TMR1CTL, 1000 << TMR1CTL_PRESCALE_SHIFT); + rt_wdt_w32(drvdata->base, TIMER_REG_TMR1CTL, 1000 << TMR1CTL_PRESCALE_SHIFT); mt7621_wdt_set_timeout(w, w->timeout); - t = rt_wdt_r32(TIMER_REG_TMR1CTL); + t = rt_wdt_r32(drvdata->base, TIMER_REG_TMR1CTL); t |= TMR1CTL_ENABLE; - rt_wdt_w32(TIMER_REG_TMR1CTL, t); + rt_wdt_w32(drvdata->base, TIMER_REG_TMR1CTL, t); return 0; } static int mt7621_wdt_stop(struct watchdog_device *w) { + struct mt7621_wdt_data *drvdata = watchdog_get_drvdata(w); u32 t; mt7621_wdt_ping(w); - t = rt_wdt_r32(TIMER_REG_TMR1CTL); + t = rt_wdt_r32(drvdata->base, TIMER_REG_TMR1CTL); t &= ~TMR1CTL_ENABLE; - rt_wdt_w32(TIMER_REG_TMR1CTL, t); + rt_wdt_w32(drvdata->base, TIMER_REG_TMR1CTL, t); return 0; } @@ -105,7 +114,9 @@ static int mt7621_wdt_bootcause(void) static int mt7621_wdt_is_running(struct watchdog_device *w) { - return !!(rt_wdt_r32(TIMER_REG_TMR1CTL) & TMR1CTL_ENABLE); + struct mt7621_wdt_data *drvdata = watchdog_get_drvdata(w); + + return !!(rt_wdt_r32(drvdata->base, TIMER_REG_TMR1CTL) & TMR1CTL_ENABLE); } static const struct watchdog_info mt7621_wdt_info = { @@ -121,30 +132,39 @@ static const struct watchdog_ops mt7621_wdt_ops = { .set_timeout = mt7621_wdt_set_timeout, }; -static struct watchdog_device mt7621_wdt_dev = { - .info = &mt7621_wdt_info, - .ops = &mt7621_wdt_ops, - .min_timeout = 1, - .max_timeout = 0xfffful / 1000, -}; - static int mt7621_wdt_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; - mt7621_wdt_base = devm_platform_ioremap_resource(pdev, 0); - if (IS_ERR(mt7621_wdt_base)) - return PTR_ERR(mt7621_wdt_base); + struct watchdog_device *mt7621_wdt; + struct mt7621_wdt_data *drvdata; + int err; + + drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL); + if (!drvdata) + return -ENOMEM; - mt7621_wdt_reset = devm_reset_control_get_exclusive(dev, NULL); - if (!IS_ERR(mt7621_wdt_reset)) - reset_control_deassert(mt7621_wdt_reset); + drvdata->base = devm_platform_ioremap_resource(pdev, 0); + if (IS_ERR(drvdata->base)) + return PTR_ERR(drvdata->base); - mt7621_wdt_dev.bootstatus = mt7621_wdt_bootcause(); + drvdata->rst = devm_reset_control_get_exclusive(dev, NULL); + if (!IS_ERR(drvdata->rst)) + reset_control_deassert(drvdata->rst); - watchdog_init_timeout(&mt7621_wdt_dev, mt7621_wdt_dev.max_timeout, - dev); - watchdog_set_nowayout(&mt7621_wdt_dev, nowayout); - if (mt7621_wdt_is_running(&mt7621_wdt_dev)) { + mt7621_wdt = &drvdata->wdt; + mt7621_wdt->info = &mt7621_wdt_info; + mt7621_wdt->ops = &mt7621_wdt_ops; + mt7621_wdt->min_timeout = 1; + mt7621_wdt->max_timeout = 0xfffful / 1000; + mt7621_wdt->parent = dev; + + mt7621_wdt->bootstatus = mt7621_wdt_bootcause(); + + watchdog_init_timeout(mt7621_wdt, mt7621_wdt->max_timeout, dev); + watchdog_set_nowayout(mt7621_wdt, nowayout); + watchdog_set_drvdata(mt7621_wdt, drvdata); + + if (mt7621_wdt_is_running(mt7621_wdt)) { /* * Make sure to apply timeout from watchdog core, taking * the prescaler of this driver here into account (the @@ -154,17 +174,25 @@ static int mt7621_wdt_probe(struct platform_device *pdev) * we first disable the watchdog, set the new prescaler * and timeout, and then re-enable the watchdog. */ - mt7621_wdt_stop(&mt7621_wdt_dev); - mt7621_wdt_start(&mt7621_wdt_dev); - set_bit(WDOG_HW_RUNNING, &mt7621_wdt_dev.status); + mt7621_wdt_stop(mt7621_wdt); + mt7621_wdt_start(mt7621_wdt); + set_bit(WDOG_HW_RUNNING, &mt7621_wdt->status); } - return devm_watchdog_register_device(dev, &mt7621_wdt_dev); + err = devm_watchdog_register_device(dev, &drvdata->wdt); + if (err) + return err; + + platform_set_drvdata(pdev, drvdata); + + return 0; } static void mt7621_wdt_shutdown(struct platform_device *pdev) { - mt7621_wdt_stop(&mt7621_wdt_dev); + struct mt7621_wdt_data *drvdata = platform_get_drvdata(pdev); + + mt7621_wdt_stop(&drvdata->wdt); } static const struct of_device_id mt7621_wdt_match[] = { From ff8ec4ac39ad413b580d611dbf68e1d8a82eba56 Mon Sep 17 00:00:00 2001 From: Sergio Paracuellos Date: Tue, 14 Feb 2023 11:39:36 +0100 Subject: [PATCH 655/765] watchdog: mt7621-wdt: avoid ralink architecture dependent code MT7621 SoC has a system controller node. Watchdog need to access to reset status register. Ralink architecture and related driver are old and from the beggining they are using some architecture dependent operations for accessing this shared registers through 'asm/mach-ralink/ralink_regs.h' header file. However this is not ideal from a driver perspective which can just access to the system controller registers in an arch independent way using regmap syscon APIs. Update Kconfig accordingly to select new added dependencies and allow driver to be compile tested. Signed-off-by: Sergio Paracuellos Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20230214103936.1061078-6-sergio.paracuellos@gmail.com Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/Kconfig | 4 +++- drivers/watchdog/mt7621_wdt.c | 22 +++++++++++++++++----- 2 files changed, 20 insertions(+), 6 deletions(-) diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig index 0bc40b763b065..3df3638ac8b0c 100644 --- a/drivers/watchdog/Kconfig +++ b/drivers/watchdog/Kconfig @@ -1872,7 +1872,9 @@ config GXP_WATCHDOG config MT7621_WDT tristate "Mediatek SoC watchdog" select WATCHDOG_CORE - depends on SOC_MT7620 || SOC_MT7621 + select REGMAP_MMIO + select MFD_SYSCON + depends on SOC_MT7620 || SOC_MT7621 || COMPILE_TEST help Hardware driver for the Mediatek/Ralink MT7621/8 SoC Watchdog Timer. diff --git a/drivers/watchdog/mt7621_wdt.c b/drivers/watchdog/mt7621_wdt.c index b02c1415fc716..442731bba1946 100644 --- a/drivers/watchdog/mt7621_wdt.c +++ b/drivers/watchdog/mt7621_wdt.c @@ -15,8 +15,8 @@ #include #include #include - -#include +#include +#include #define SYSC_RSTSTAT 0x38 #define WDT_RST_CAUSE BIT(1) @@ -34,6 +34,7 @@ struct mt7621_wdt_data { void __iomem *base; struct reset_control *rst; + struct regmap *sysc; struct watchdog_device wdt; }; @@ -104,9 +105,12 @@ static int mt7621_wdt_stop(struct watchdog_device *w) return 0; } -static int mt7621_wdt_bootcause(void) +static int mt7621_wdt_bootcause(struct mt7621_wdt_data *d) { - if (rt_sysc_r32(SYSC_RSTSTAT) & WDT_RST_CAUSE) + u32 val; + + regmap_read(d->sysc, SYSC_RSTSTAT, &val); + if (val & WDT_RST_CAUSE) return WDIOF_CARDRESET; return 0; @@ -134,6 +138,7 @@ static const struct watchdog_ops mt7621_wdt_ops = { static int mt7621_wdt_probe(struct platform_device *pdev) { + struct device_node *np = pdev->dev.of_node; struct device *dev = &pdev->dev; struct watchdog_device *mt7621_wdt; struct mt7621_wdt_data *drvdata; @@ -143,6 +148,13 @@ static int mt7621_wdt_probe(struct platform_device *pdev) if (!drvdata) return -ENOMEM; + drvdata->sysc = syscon_regmap_lookup_by_phandle(np, "mediatek,sysctl"); + if (IS_ERR(drvdata->sysc)) { + drvdata->sysc = syscon_regmap_lookup_by_compatible("mediatek,mt7621-sysc"); + if (IS_ERR(drvdata->sysc)) + return PTR_ERR(drvdata->sysc); + } + drvdata->base = devm_platform_ioremap_resource(pdev, 0); if (IS_ERR(drvdata->base)) return PTR_ERR(drvdata->base); @@ -158,7 +170,7 @@ static int mt7621_wdt_probe(struct platform_device *pdev) mt7621_wdt->max_timeout = 0xfffful / 1000; mt7621_wdt->parent = dev; - mt7621_wdt->bootstatus = mt7621_wdt_bootcause(); + mt7621_wdt->bootstatus = mt7621_wdt_bootcause(drvdata); watchdog_init_timeout(mt7621_wdt, mt7621_wdt->max_timeout, dev); watchdog_set_nowayout(mt7621_wdt, nowayout); From cf3be7e82b129ed34f811f116f2b113f6299d449 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Fri, 17 Feb 2023 10:53:17 +0100 Subject: [PATCH 656/765] watchdog: at91rm9200: Only warn once about problems in .remove() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The single difference between returning 0 and returning an error code in a platform remove callback is that in the latter case the platform core emits a warning about the error being ignored. at91wdt_remove() already emits a warning in the error case, so suppress the more generic (and less helpful) one by returning 0. Signed-off-by: Uwe Kleine-König Reviewed-by: Claudiu Beznea Reviewed-by: Guenter Roeck Link: https://lore.kernel.org/r/20230217095317.1213387-1-u.kleine-koenig@pengutronix.de Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/at91rm9200_wdt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/watchdog/at91rm9200_wdt.c b/drivers/watchdog/at91rm9200_wdt.c index 5126454bb8613..d57409c1a4d12 100644 --- a/drivers/watchdog/at91rm9200_wdt.c +++ b/drivers/watchdog/at91rm9200_wdt.c @@ -270,7 +270,7 @@ static int at91wdt_remove(struct platform_device *pdev) misc_deregister(&at91wdt_miscdev); at91wdt_miscdev.parent = NULL; - return res; + return 0; } static void at91wdt_shutdown(struct platform_device *pdev) From 5c1ebbfabcd61142a4551bfc0e51840f9bdae7af Mon Sep 17 00:00:00 2001 From: Brian Vazquez Date: Wed, 1 Mar 2023 13:32:47 +0000 Subject: [PATCH 657/765] net: use indirect calls helpers for sk_exit_memory_pressure() Florian reported a regression and sent a patch with the following changelog: There is a noticeable tcp performance regression (loopback or cross-netns), seen with iperf3 -Z (sendfile mode) when generic retpolines are needed. With SK_RECLAIM_THRESHOLD checks gone number of calls to enter/leave memory pressure happen much more often. For TCP indirect calls are used. We can't remove the if-set-return short-circuit check in tcp_enter_memory_pressure because there are callers other than sk_enter_memory_pressure. Doing a check in the sk wrapper too reduces the indirect calls enough to recover some performance. Before, 0.00-60.00 sec 322 GBytes 46.1 Gbits/sec receiver After: 0.00-60.04 sec 359 GBytes 51.4 Gbits/sec receiver "iperf3 -c $peer -t 60 -Z -f g", connected via veth in another netns. It seems we forgot to upstream this indirect call mitigation we had for years, lets do this instead. [edumazet] - It seems we forgot to upstream this indirect call mitigation we had for years, let's do this instead. - Changed to INDIRECT_CALL_INET_1() to avoid bots reports. Fixes: 4890b686f408 ("net: keep sk->sk_forward_alloc as small as possible") Reported-by: Florian Westphal Link: https://lore.kernel.org/netdev/20230227152741.4a53634b@kernel.org/T/ Signed-off-by: Brian Vazquez Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20230301133247.2346111-1-edumazet@google.com Signed-off-by: Paolo Abeni --- net/core/sock.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/core/sock.c b/net/core/sock.c index 341c565dbc262..c258887953905 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -2818,7 +2818,8 @@ static void sk_enter_memory_pressure(struct sock *sk) static void sk_leave_memory_pressure(struct sock *sk) { if (sk->sk_prot->leave_memory_pressure) { - sk->sk_prot->leave_memory_pressure(sk); + INDIRECT_CALL_INET_1(sk->sk_prot->leave_memory_pressure, + tcp_leave_memory_pressure, sk); } else { unsigned long *memory_pressure = sk->sk_prot->memory_pressure; From 6c993779ea1d0cccdb3a5d7d45446dd229e610a3 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Thu, 16 Feb 2023 23:25:04 -0500 Subject: [PATCH 658/765] ca8210: fix mac_len negative array access This patch fixes a buffer overflow access of skb->data if ieee802154_hdr_peek_addrs() fails. Reported-by: lianhui tang Signed-off-by: Alexander Aring Link: https://lore.kernel.org/r/20230217042504.3303396-1-aahringo@redhat.com Signed-off-by: Stefan Schmidt --- drivers/net/ieee802154/ca8210.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ieee802154/ca8210.c b/drivers/net/ieee802154/ca8210.c index e1a569b99e4a6..0b0c6c0764fe9 100644 --- a/drivers/net/ieee802154/ca8210.c +++ b/drivers/net/ieee802154/ca8210.c @@ -1913,6 +1913,8 @@ static int ca8210_skb_tx( * packet */ mac_len = ieee802154_hdr_peek_addrs(skb, &header); + if (mac_len < 0) + return mac_len; secspec.security_level = header.sec.level; secspec.key_id_mode = header.sec.key_id_mode; From 02f18662f6c671382345fcb696e808d78f4c194a Mon Sep 17 00:00:00 2001 From: Miquel Raynal Date: Wed, 1 Mar 2023 16:44:50 +0100 Subject: [PATCH 659/765] ieee802154: Prevent user from crashing the host Avoid crashing the machine by checking info->attrs[NL802154_ATTR_SCAN_TYPE] presence before de-referencing it, which was the primary intend of the blamed patch. Reported-by: Sanan Hasanov Suggested-by: Eric Dumazet Fixes: a0b6106672b5 ("ieee802154: Convert scan error messages to extack") Signed-off-by: Miquel Raynal Link: https://lore.kernel.org/r/20230301154450.547716-1-miquel.raynal@bootlin.com Signed-off-by: Stefan Schmidt --- net/ieee802154/nl802154.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ieee802154/nl802154.c b/net/ieee802154/nl802154.c index 2215f576ee378..d8f4379d4fa68 100644 --- a/net/ieee802154/nl802154.c +++ b/net/ieee802154/nl802154.c @@ -1412,7 +1412,7 @@ static int nl802154_trigger_scan(struct sk_buff *skb, struct genl_info *info) return -EOPNOTSUPP; } - if (!nla_get_u8(info->attrs[NL802154_ATTR_SCAN_TYPE])) { + if (!info->attrs[NL802154_ATTR_SCAN_TYPE]) { NL_SET_ERR_MSG(info->extack, "Malformed request, missing scan type"); return -EINVAL; } From 42e19e6f04984088b6f9f0507c4c89a8152d9730 Mon Sep 17 00:00:00 2001 From: Vasily Gorbik Date: Wed, 1 Mar 2023 02:23:08 +0100 Subject: [PATCH 660/765] s390/kprobes: fix irq mask clobbering on kprobe reenter from post_handler Recent test_kprobe_missed kprobes kunit test uncovers the following error (reported when CONFIG_DEBUG_ATOMIC_SLEEP is enabled): BUG: sleeping function called from invalid context at kernel/locking/mutex.c:580 in_atomic(): 0, irqs_disabled(): 1, non_block: 0, pid: 662, name: kunit_try_catch preempt_count: 0, expected: 0 RCU nest depth: 0, expected: 0 no locks held by kunit_try_catch/662. irq event stamp: 280 hardirqs last enabled at (279): [<00000003e60a3d42>] __do_pgm_check+0x17a/0x1c0 hardirqs last disabled at (280): [<00000003e3bd774a>] kprobe_exceptions_notify+0x27a/0x318 softirqs last enabled at (0): [<00000003e3c5c890>] copy_process+0x14a8/0x4c80 softirqs last disabled at (0): [<0000000000000000>] 0x0 CPU: 46 PID: 662 Comm: kunit_try_catch Tainted: G N 6.2.0-173644-g44c18d77f0c0 #2 Hardware name: IBM 3931 A01 704 (LPAR) Call Trace: [<00000003e60a3a00>] dump_stack_lvl+0x120/0x198 [<00000003e3d02e82>] __might_resched+0x60a/0x668 [<00000003e60b9908>] __mutex_lock+0xc0/0x14e0 [<00000003e60bad5a>] mutex_lock_nested+0x32/0x40 [<00000003e3f7b460>] unregister_kprobe+0x30/0xd8 [<00000003e51b2602>] test_kprobe_missed+0xf2/0x268 [<00000003e51b5406>] kunit_try_run_case+0x10e/0x290 [<00000003e51b7dfa>] kunit_generic_run_threadfn_adapter+0x62/0xb8 [<00000003e3ce30f8>] kthread+0x2d0/0x398 [<00000003e3b96afa>] __ret_from_fork+0x8a/0xe8 [<00000003e60ccada>] ret_from_fork+0xa/0x40 The reason for this error report is that kprobes handling code failed to restore irqs. The problem is that when kprobe is triggered from another kprobe post_handler current sequence of enable_singlestep / disable_singlestep is the following: enable_singlestep <- original kprobe (saves kprobe_saved_imask) enable_singlestep <- kprobe triggered from post_handler (clobbers kprobe_saved_imask) disable_singlestep <- kprobe triggered from post_handler (restores kprobe_saved_imask) disable_singlestep <- original kprobe (restores wrong clobbered kprobe_saved_imask) There is just one kprobe_ctlblk per cpu and both calls saves and loads irq mask to kprobe_saved_imask. To fix the problem simply move resume_execution (which calls disable_singlestep) before calling post_handler. This also fixes the problem that post_handler is called with pt_regs which were not yet adjusted after single-stepping. Cc: stable@vger.kernel.org Fixes: 4ba069b802c2 ("[S390] add kprobes support.") Reviewed-by: Heiko Carstens Signed-off-by: Vasily Gorbik Signed-off-by: Heiko Carstens --- arch/s390/kernel/kprobes.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c index 5e713f318de3e..698fce57a2c8e 100644 --- a/arch/s390/kernel/kprobes.c +++ b/arch/s390/kernel/kprobes.c @@ -402,12 +402,11 @@ static int post_kprobe_handler(struct pt_regs *regs) if (!p) return 0; + resume_execution(p, regs); if (kcb->kprobe_status != KPROBE_REENTER && p->post_handler) { kcb->kprobe_status = KPROBE_HIT_SSDONE; p->post_handler(p, regs, 0); } - - resume_execution(p, regs); pop_kprobe(kcb); preempt_enable_no_resched(); From cd57953936f2213dfaccce10d20f396956222c7d Mon Sep 17 00:00:00 2001 From: Vasily Gorbik Date: Wed, 1 Mar 2023 17:58:06 +0100 Subject: [PATCH 661/765] s390/kprobes: fix current_kprobe never cleared after kprobes reenter Recent test_kprobe_missed kprobes kunit test uncovers the following problem. Once kprobe is triggered from another kprobe (kprobe reenter), all future kprobes on this cpu are considered as kprobe reenter, thus pre_handler and post_handler are not being called and kprobes are counted as "missed". Commit b9599798f953 ("[S390] kprobes: activation and deactivation") introduced a simpler scheme for kprobes (de)activation and status tracking by using push_kprobe/pop_kprobe, which supposed to work for both initial kprobe entry as well as kprobe reentry and helps to avoid handling those two cases differently. The problem is that a sequence of calls in case of kprobes reenter: push_kprobe() <- NULL (current_kprobe) push_kprobe() <- kprobe1 (current_kprobe) pop_kprobe() -> kprobe1 (current_kprobe) pop_kprobe() -> kprobe1 (current_kprobe) leaves "kprobe1" as "current_kprobe" on this cpu, instead of setting it to NULL. In fact push_kprobe/pop_kprobe can only store a single state (there is just one prev_kprobe in kprobe_ctlblk). Which is a hack but sufficient, there is no need to have another prev_kprobe just to store NULL. To make a simple and backportable fix simply reset "prev_kprobe" when kprobe is poped from this "stack". No need to worry about "kprobe_status" in this case, because its value is only checked when current_kprobe != NULL. Cc: stable@vger.kernel.org Fixes: b9599798f953 ("[S390] kprobes: activation and deactivation") Reviewed-by: Heiko Carstens Signed-off-by: Vasily Gorbik Signed-off-by: Heiko Carstens --- arch/s390/kernel/kprobes.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c index 698fce57a2c8e..7b41ceecbb253 100644 --- a/arch/s390/kernel/kprobes.c +++ b/arch/s390/kernel/kprobes.c @@ -278,6 +278,7 @@ static void pop_kprobe(struct kprobe_ctlblk *kcb) { __this_cpu_write(current_kprobe, kcb->prev_kprobe.kp); kcb->kprobe_status = kcb->prev_kprobe.status; + kcb->prev_kprobe.kp = NULL; } NOKPROBE_SYMBOL(pop_kprobe); From 0fb7fb713461e44b12e72c292bf90ee300f40710 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 1 Mar 2023 22:07:48 +0100 Subject: [PATCH 662/765] genirq/msi, platform-msi: Ensure that MSI descriptors are unreferenced Miquel reported a warning in the MSI core which is triggered when interrupts are freed via platform_msi_device_domain_free(). This code got reworked to use core functions for freeing the MSI descriptors, but nothing took care to clear the msi_desc->irq entry, which then triggers the warning in msi_free_msi_desc() which uses desc->irq to validate that the descriptor has been torn down. The same issue exists in msi_domain_populate_irqs(). Up to the point that msi_free_msi_descs() grew a warning for this case, this went un-noticed. Provide the counterpart of msi_domain_populate_irqs() and invoke it in platform_msi_device_domain_free() before freeing the interrupts and MSI descriptors and also in the error path of msi_domain_populate_irqs(). Fixes: 2f2940d16823 ("genirq/msi: Remove filter from msi_free_descs_free_range()") Reported-by: Miquel Raynal Signed-off-by: Thomas Gleixner Tested-by: Miquel Raynal Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/87mt4wkwnv.ffs@tglx --- drivers/base/platform-msi.c | 1 + include/linux/msi.h | 2 ++ kernel/irq/msi.c | 23 ++++++++++++++++++++++- 3 files changed, 25 insertions(+), 1 deletion(-) diff --git a/drivers/base/platform-msi.c b/drivers/base/platform-msi.c index 5883e7634a2b7..f37ad34c80ec4 100644 --- a/drivers/base/platform-msi.c +++ b/drivers/base/platform-msi.c @@ -324,6 +324,7 @@ void platform_msi_device_domain_free(struct irq_domain *domain, unsigned int vir struct platform_msi_priv_data *data = domain->host_data; msi_lock_descs(data->dev); + msi_domain_depopulate_descs(data->dev, virq, nr_irqs); irq_domain_free_irqs_common(domain, virq, nr_irqs); msi_free_msi_descs_range(data->dev, virq, virq + nr_irqs - 1); msi_unlock_descs(data->dev); diff --git a/include/linux/msi.h b/include/linux/msi.h index a112b913fff94..15dd71817996f 100644 --- a/include/linux/msi.h +++ b/include/linux/msi.h @@ -631,6 +631,8 @@ int msi_domain_prepare_irqs(struct irq_domain *domain, struct device *dev, int nvec, msi_alloc_info_t *args); int msi_domain_populate_irqs(struct irq_domain *domain, struct device *dev, int virq, int nvec, msi_alloc_info_t *args); +void msi_domain_depopulate_descs(struct device *dev, int virq, int nvec); + struct irq_domain * __platform_msi_create_device_domain(struct device *dev, unsigned int nvec, diff --git a/kernel/irq/msi.c b/kernel/irq/msi.c index efd21b79bf322..d169ee0c1799b 100644 --- a/kernel/irq/msi.c +++ b/kernel/irq/msi.c @@ -1109,14 +1109,35 @@ int msi_domain_populate_irqs(struct irq_domain *domain, struct device *dev, return 0; fail: - for (--virq; virq >= virq_base; virq--) + for (--virq; virq >= virq_base; virq--) { + msi_domain_depopulate_descs(dev, virq, 1); irq_domain_free_irqs_common(domain, virq, 1); + } msi_domain_free_descs(dev, &ctrl); unlock: msi_unlock_descs(dev); return ret; } +void msi_domain_depopulate_descs(struct device *dev, int virq_base, int nvec) +{ + struct msi_ctrl ctrl = { + .domid = MSI_DEFAULT_DOMAIN, + .first = virq_base, + .last = virq_base + nvec - 1, + }; + struct msi_desc *desc; + struct xarray *xa; + unsigned long idx; + + if (!msi_ctrl_valid(dev, &ctrl)) + return; + + xa = &dev->msi.data->__domains[ctrl.domid].store; + xa_for_each_range(xa, idx, desc, ctrl.first, ctrl.last) + desc->irq = 0; +} + /* * Carefully check whether the device can use reservation mode. If * reservation mode is enabled then the early activation will assign a From bd75497a77ccae9a5e1b0f1a3fa8283d67b21f4c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 30 Jan 2023 23:48:54 -0500 Subject: [PATCH 663/765] m68k: fix livelock in uaccess m68k equivalent of 26178ec11ef3 "x86: mm: consolidate VM_FAULT_RETRY handling" If e.g. get_user() triggers a page fault and a fatal signal is caught, we might end up with handle_mm_fault() returning VM_FAULT_RETRY and not doing anything to page tables. In such case we must *not* return to the faulting insn - that would repeat the entire thing without making any progress; what we need instead is to treat that as failed (user) memory access. Tested-by: Finn Thain Tested-by: Geert Uytterhoeven Acked-by: Geert Uytterhoeven Signed-off-by: Al Viro --- arch/m68k/mm/fault.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/m68k/mm/fault.c b/arch/m68k/mm/fault.c index 4d2837eb3e2a3..228128e45c673 100644 --- a/arch/m68k/mm/fault.c +++ b/arch/m68k/mm/fault.c @@ -138,8 +138,11 @@ int do_page_fault(struct pt_regs *regs, unsigned long address, fault = handle_mm_fault(vma, address, flags, regs); pr_debug("handle_mm_fault returns %x\n", fault); - if (fault_signal_pending(fault, regs)) + if (fault_signal_pending(fault, regs)) { + if (!user_mode(regs)) + goto no_context; return 0; + } /* The fault is fully completed (including releasing mmap lock) */ if (fault & VM_FAULT_COMPLETED) From d835eb3a57de50d16dd4a6b53f95dbc2bfa8ef48 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 30 Jan 2023 23:55:52 -0500 Subject: [PATCH 664/765] riscv: fix livelock in uaccess MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit riscv equivalent of 26178ec11ef3 "x86: mm: consolidate VM_FAULT_RETRY handling" If e.g. get_user() triggers a page fault and a fatal signal is caught, we might end up with handle_mm_fault() returning VM_FAULT_RETRY and not doing anything to page tables. In such case we must *not* return to the faulting insn - that would repeat the entire thing without making any progress; what we need instead is to treat that as failed (user) memory access. Tested-by: Björn Töpel Tested-by: Geert Uytterhoeven Signed-off-by: Al Viro --- arch/riscv/mm/fault.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/riscv/mm/fault.c b/arch/riscv/mm/fault.c index d86f7cebd4a7e..c91d85349d39a 100644 --- a/arch/riscv/mm/fault.c +++ b/arch/riscv/mm/fault.c @@ -324,8 +324,11 @@ asmlinkage void do_page_fault(struct pt_regs *regs) * signal first. We do not need to release the mmap_lock because it * would already be released in __lock_page_or_retry in mm/filemap.c. */ - if (fault_signal_pending(fault, regs)) + if (fault_signal_pending(fault, regs)) { + if (!user_mode(regs)) + no_context(regs, addr); return; + } /* The fault is fully completed (including releasing mmap lock) */ if (fault & VM_FAULT_COMPLETED) From 0b92ed09cb9ffada29ca05510ef28252b6059d6d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 30 Jan 2023 23:46:10 -0500 Subject: [PATCH 665/765] hexagon: fix livelock in uaccess hexagon equivalent of 26178ec11ef3 "x86: mm: consolidate VM_FAULT_RETRY handling" If e.g. get_user() triggers a page fault and a fatal signal is caught, we might end up with handle_mm_fault() returning VM_FAULT_RETRY and not doing anything to page tables. In such case we must *not* return to the faulting insn - that would repeat the entire thing without making any progress; what we need instead is to treat that as failed (user) memory access. Acked-by: Brian Cain Signed-off-by: Al Viro --- arch/hexagon/mm/vm_fault.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/hexagon/mm/vm_fault.c b/arch/hexagon/mm/vm_fault.c index f73c7cbfe3260..4b578d02fd01a 100644 --- a/arch/hexagon/mm/vm_fault.c +++ b/arch/hexagon/mm/vm_fault.c @@ -93,8 +93,11 @@ void do_page_fault(unsigned long address, long cause, struct pt_regs *regs) fault = handle_mm_fault(vma, address, flags, regs); - if (fault_signal_pending(fault, regs)) + if (fault_signal_pending(fault, regs)) { + if (!user_mode(regs)) + goto no_context; return; + } /* The fault is fully completed (including releasing mmap lock) */ if (fault & VM_FAULT_COMPLETED) From 15261678a8c2ff093ffedad730b1646d34a1716a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 30 Jan 2023 23:53:52 -0500 Subject: [PATCH 666/765] parisc: fix livelock in uaccess parisc equivalent of 26178ec11ef3 "x86: mm: consolidate VM_FAULT_RETRY handling" If e.g. get_user() triggers a page fault and a fatal signal is caught, we might end up with handle_mm_fault() returning VM_FAULT_RETRY and not doing anything to page tables. In such case we must *not* return to the faulting insn - that would repeat the entire thing without making any progress; what we need instead is to treat that as failed (user) memory access. Tested-by: Helge Deller Signed-off-by: Al Viro --- arch/parisc/mm/fault.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/arch/parisc/mm/fault.c b/arch/parisc/mm/fault.c index 869204e97ec9d..6941fdbf25173 100644 --- a/arch/parisc/mm/fault.c +++ b/arch/parisc/mm/fault.c @@ -308,8 +308,13 @@ void do_page_fault(struct pt_regs *regs, unsigned long code, fault = handle_mm_fault(vma, address, flags, regs); - if (fault_signal_pending(fault, regs)) + if (fault_signal_pending(fault, regs)) { + if (!user_mode(regs)) { + msg = "Page fault: fault signal on kernel memory"; + goto no_context; + } return; + } /* The fault is fully completed (including releasing mmap lock) */ if (fault & VM_FAULT_COMPLETED) From dce45493aff3fdd57fed2a0da264e585dba88433 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 30 Jan 2023 23:34:16 -0500 Subject: [PATCH 667/765] alpha: fix livelock in uaccess alpha equivalent of 26178ec11ef3 "x86: mm: consolidate VM_FAULT_RETRY handling" If e.g. get_user() triggers a page fault and a fatal signal is caught, we might end up with handle_mm_fault() returning VM_FAULT_RETRY and not doing anything to page tables. In such case we must *not* return to the faulting insn - that would repeat the entire thing without making any progress; what we need instead is to treat that as failed (user) memory access. Signed-off-by: Al Viro --- arch/alpha/mm/fault.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/alpha/mm/fault.c b/arch/alpha/mm/fault.c index ef427a6bdd1ab..7b01ae4f3bc6c 100644 --- a/arch/alpha/mm/fault.c +++ b/arch/alpha/mm/fault.c @@ -152,8 +152,11 @@ do_page_fault(unsigned long address, unsigned long mmcsr, the fault. */ fault = handle_mm_fault(vma, address, flags, regs); - if (fault_signal_pending(fault, regs)) + if (fault_signal_pending(fault, regs)) { + if (!user_mode(regs)) + goto no_context; return; + } /* The fault is fully completed (including releasing mmap lock) */ if (fault & VM_FAULT_COMPLETED) From 79c54c97c7735e09f9e2193a713f30e56eec00dc Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 31 Jan 2023 00:03:40 -0500 Subject: [PATCH 668/765] sparc: fix livelock in uaccess sparc equivalent of 26178ec11ef3 "x86: mm: consolidate VM_FAULT_RETRY handling" If e.g. get_user() triggers a page fault and a fatal signal is caught, we might end up with handle_mm_fault() returning VM_FAULT_RETRY and not doing anything to page tables. In such case we must *not* return to the faulting insn - that would repeat the entire thing without making any progress; what we need instead is to treat that as failed (user) memory access. Signed-off-by: Al Viro --- arch/sparc/mm/fault_32.c | 5 ++++- arch/sparc/mm/fault_64.c | 7 ++++++- 2 files changed, 10 insertions(+), 2 deletions(-) diff --git a/arch/sparc/mm/fault_32.c b/arch/sparc/mm/fault_32.c index 91259f291c540..179295b14664a 100644 --- a/arch/sparc/mm/fault_32.c +++ b/arch/sparc/mm/fault_32.c @@ -187,8 +187,11 @@ asmlinkage void do_sparc_fault(struct pt_regs *regs, int text_fault, int write, */ fault = handle_mm_fault(vma, address, flags, regs); - if (fault_signal_pending(fault, regs)) + if (fault_signal_pending(fault, regs)) { + if (!from_user) + goto no_context; return; + } /* The fault is fully completed (including releasing mmap lock) */ if (fault & VM_FAULT_COMPLETED) diff --git a/arch/sparc/mm/fault_64.c b/arch/sparc/mm/fault_64.c index 4acc12eafbf54..d91305de694c5 100644 --- a/arch/sparc/mm/fault_64.c +++ b/arch/sparc/mm/fault_64.c @@ -424,8 +424,13 @@ asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs) fault = handle_mm_fault(vma, address, flags, regs); - if (fault_signal_pending(fault, regs)) + if (fault_signal_pending(fault, regs)) { + if (regs->tstate & TSTATE_PRIV) { + insn = get_fault_insn(regs, insn); + goto handle_kernel_fault; + } goto exit_exception; + } /* The fault is fully completed (including releasing mmap lock) */ if (fault & VM_FAULT_COMPLETED) From d088af1e221c32bd67825c5dccf664f699e827b4 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 30 Jan 2023 23:47:28 -0500 Subject: [PATCH 669/765] ia64: fix livelock in uaccess ia64 equivalent of 26178ec11ef3 "x86: mm: consolidate VM_FAULT_RETRY handling" If e.g. get_user() triggers a page fault and a fatal signal is caught, we might end up with handle_mm_fault() returning VM_FAULT_RETRY and not doing anything to page tables. In such case we must *not* return to the faulting insn - that would repeat the entire thing without making any progress; what we need instead is to treat that as failed (user) memory access. Signed-off-by: Al Viro --- arch/ia64/mm/fault.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/ia64/mm/fault.c b/arch/ia64/mm/fault.c index ef78c2d66cdde..85c4d9ac8686d 100644 --- a/arch/ia64/mm/fault.c +++ b/arch/ia64/mm/fault.c @@ -136,8 +136,11 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re */ fault = handle_mm_fault(vma, address, flags, regs); - if (fault_signal_pending(fault, regs)) + if (fault_signal_pending(fault, regs)) { + if (!user_mode(regs)) + goto no_context; return; + } /* The fault is fully completed (including releasing mmap lock) */ if (fault & VM_FAULT_COMPLETED) From a1179ac743e8f2044b67848b5109614619b65366 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 30 Jan 2023 23:50:22 -0500 Subject: [PATCH 670/765] microblaze: fix livelock in uaccess microblaze equivalent of 26178ec11ef3 "x86: mm: consolidate VM_FAULT_RETRY handling" If e.g. get_user() triggers a page fault and a fatal signal is caught, we might end up with handle_mm_fault() returning VM_FAULT_RETRY and not doing anything to page tables. In such case we must *not* return to the faulting insn - that would repeat the entire thing without making any progress; what we need instead is to treat that as failed (user) memory access. Signed-off-by: Al Viro --- arch/microblaze/mm/fault.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/microblaze/mm/fault.c b/arch/microblaze/mm/fault.c index 5c40c3ebe52f7..687714db6f4d0 100644 --- a/arch/microblaze/mm/fault.c +++ b/arch/microblaze/mm/fault.c @@ -219,8 +219,11 @@ void do_page_fault(struct pt_regs *regs, unsigned long address, */ fault = handle_mm_fault(vma, address, flags, regs); - if (fault_signal_pending(fault, regs)) + if (fault_signal_pending(fault, regs)) { + if (!user_mode(regs)) + bad_page_fault(regs, address, SIGBUS); return; + } /* The fault is fully completed (including releasing mmap lock) */ if (fault & VM_FAULT_COMPLETED) From e902e508c5b280488c62f119a05ad88f9f7eb1cb Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 30 Jan 2023 23:51:40 -0500 Subject: [PATCH 671/765] nios2: fix livelock in uaccess nios2 equivalent of 26178ec11ef3 "x86: mm: consolidate VM_FAULT_RETRY handling" If e.g. get_user() triggers a page fault and a fatal signal is caught, we might end up with handle_mm_fault() returning VM_FAULT_RETRY and not doing anything to page tables. In such case we must *not* return to the faulting insn - that would repeat the entire thing without making any progress; what we need instead is to treat that as failed (user) memory access. Signed-off-by: Al Viro --- arch/nios2/mm/fault.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/nios2/mm/fault.c b/arch/nios2/mm/fault.c index edaca0a6c1c1c..ca64eccea5511 100644 --- a/arch/nios2/mm/fault.c +++ b/arch/nios2/mm/fault.c @@ -136,8 +136,11 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long cause, */ fault = handle_mm_fault(vma, address, flags, regs); - if (fault_signal_pending(fault, regs)) + if (fault_signal_pending(fault, regs)) { + if (!user_mode(regs)) + goto no_context; return; + } /* The fault is fully completed (including releasing mmap lock) */ if (fault & VM_FAULT_COMPLETED) From caa82ae7ef52b7cf5f80a2b2fbcbdbcfd16426cc Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 30 Jan 2023 23:52:40 -0500 Subject: [PATCH 672/765] openrisc: fix livelock in uaccess openrisc equivalent of 26178ec11ef3 "x86: mm: consolidate VM_FAULT_RETRY handling" If e.g. get_user() triggers a page fault and a fatal signal is caught, we might end up with handle_mm_fault() returning VM_FAULT_RETRY and not doing anything to page tables. In such case we must *not* return to the faulting insn - that would repeat the entire thing without making any progress; what we need instead is to treat that as failed (user) memory access. Signed-off-by: Al Viro --- arch/openrisc/mm/fault.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/openrisc/mm/fault.c b/arch/openrisc/mm/fault.c index b4762d66e9efe..6734fee3134f4 100644 --- a/arch/openrisc/mm/fault.c +++ b/arch/openrisc/mm/fault.c @@ -162,8 +162,11 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long address, fault = handle_mm_fault(vma, address, flags, regs); - if (fault_signal_pending(fault, regs)) + if (fault_signal_pending(fault, regs)) { + if (!user_mode(regs)) + goto no_context; return; + } /* The fault is fully completed (including releasing mmap lock) */ if (fault & VM_FAULT_COMPLETED) From 3098cb655e7c9ea1c2919e61234f63ebaab6bb21 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Thu, 2 Mar 2023 11:16:36 -0300 Subject: [PATCH 673/765] rust: bindgen: Add `alt_instr` as opaque type To address this build error: BINDGEN rust/bindings/bindings_generated.rs BINDGEN rust/bindings/bindings_helpers_generated.rs EXPORTS rust/exports_core_generated.h RUSTC P rust/libmacros.so RUSTC L rust/compiler_builtins.o RUSTC L rust/alloc.o RUSTC L rust/bindings.o RUSTC L rust/build_error.o EXPORTS rust/exports_alloc_generated.h error[E0588]: packed type cannot transitively contain a `#[repr(align)]` type --> /var/home/acme/git/linux/rust/bindings/bindings_generated.rs:10094:1 | 10094 | / pub struct alt_instr { 10095 | | pub instr_offset: s32, 10096 | | pub repl_offset: s32, 10097 | | pub __bindgen_anon_1: alt_instr__bindgen_ty_1, 10098 | | pub instrlen: u8_, 10099 | | pub replacementlen: u8_, 10100 | | } | |_^ | note: `alt_instr__bindgen_ty_1__bindgen_ty_1` has a `#[repr(align)]` attribute --> /var/home/acme/git/linux/rust/bindings/bindings_generated.rs:10111:1 | 10111 | / pub struct alt_instr__bindgen_ty_1__bindgen_ty_1 { 10112 | | pub _bitfield_1: __BindgenBitfieldUnit<[u8; 4usize], u16>, 10113 | | } | |_^ note: `alt_instr` contains a field of type `alt_instr__bindgen_ty_1` --> /var/home/acme/git/linux/rust/bindings/bindings_generated.rs:10097:9 | 10097 | pub __bindgen_anon_1: alt_instr__bindgen_ty_1, | ^^^^^^^^^^^^^^^^ note: ...which contains a field of type `alt_instr__bindgen_ty_1__bindgen_ty_1` --> /var/home/acme/git/linux/rust/bindings/bindings_generated.rs:10104:9 | 10104 | pub __bindgen_anon_1: alt_instr__bindgen_ty_1__bindgen_ty_1, | ^^^^^^^^^^^^^^^^ error: aborting due to previous error For more information about this error, try `rustc --explain E0588`. make[1]: *** [rust/Makefile:389: rust/bindings.o] Error 1 make: *** [Makefile:1293: prepare] Error 2 Cc: Derek Barbosa Cc: Miguel Ojeda Signed-off-by: Arnaldo Carvalho de Melo Fixes: 5d1dd961e743 ("x86/alternatives: Add alt_instr.flags") Reported-by: kernel test robot Reported-by: Vincenzo Palazzo Reviewed-by: Martin Rodriguez Reboredo Reviewed-by: Vincenzo Palazzo Signed-off-by: Miguel Ojeda --- rust/bindgen_parameters | 1 + 1 file changed, 1 insertion(+) diff --git a/rust/bindgen_parameters b/rust/bindgen_parameters index be4963bf72030..552d9a85925b9 100644 --- a/rust/bindgen_parameters +++ b/rust/bindgen_parameters @@ -6,6 +6,7 @@ --opaque-type local_apic # Packed type cannot transitively contain a `#[repr(align)]` type. +--opaque-type alt_instr --opaque-type x86_msi_data --opaque-type x86_msi_addr_lo From 49d24398327e32265eccdeec4baeb5a6a609c0bd Mon Sep 17 00:00:00 2001 From: Uday Shankar Date: Tue, 28 Feb 2023 17:06:55 -0700 Subject: [PATCH 674/765] blk-mq: enforce op-specific segment limits in blk_insert_cloned_request The block layer might merge together discard requests up until the max_discard_segments limit is hit, but blk_insert_cloned_request checks the segment count against max_segments regardless of the req op. This can result in errors like the following when discards are issued through a DM device and max_discard_segments exceeds max_segments for the queue of the chosen underlying device. blk_insert_cloned_request: over max segments limit. (256 > 129) Fix this by looking at the req_op and enforcing the appropriate segment limit - max_discard_segments for REQ_OP_DISCARDs and max_segments for everything else. Signed-off-by: Uday Shankar Reviewed-by: Keith Busch Reviewed-by: Ming Lei Link: https://lore.kernel.org/r/20230301000655.48112-1-ushankar@purestorage.com Signed-off-by: Jens Axboe --- block/blk-merge.c | 7 ------- block/blk-mq.c | 7 ++++--- block/blk.h | 7 +++++++ 3 files changed, 11 insertions(+), 10 deletions(-) diff --git a/block/blk-merge.c b/block/blk-merge.c index 808b58129d3e4..ff72edd7ee033 100644 --- a/block/blk-merge.c +++ b/block/blk-merge.c @@ -586,13 +586,6 @@ int __blk_rq_map_sg(struct request_queue *q, struct request *rq, } EXPORT_SYMBOL(__blk_rq_map_sg); -static inline unsigned int blk_rq_get_max_segments(struct request *rq) -{ - if (req_op(rq) == REQ_OP_DISCARD) - return queue_max_discard_segments(rq->q); - return queue_max_segments(rq->q); -} - static inline unsigned int blk_rq_get_max_sectors(struct request *rq, sector_t offset) { diff --git a/block/blk-mq.c b/block/blk-mq.c index 08093d4348dd6..7c6f812323afd 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -3000,6 +3000,7 @@ blk_status_t blk_insert_cloned_request(struct request *rq) { struct request_queue *q = rq->q; unsigned int max_sectors = blk_queue_get_max_sectors(q, req_op(rq)); + unsigned int max_segments = blk_rq_get_max_segments(rq); blk_status_t ret; if (blk_rq_sectors(rq) > max_sectors) { @@ -3026,9 +3027,9 @@ blk_status_t blk_insert_cloned_request(struct request *rq) * original queue. */ rq->nr_phys_segments = blk_recalc_rq_segments(rq); - if (rq->nr_phys_segments > queue_max_segments(q)) { - printk(KERN_ERR "%s: over max segments limit. (%hu > %hu)\n", - __func__, rq->nr_phys_segments, queue_max_segments(q)); + if (rq->nr_phys_segments > max_segments) { + printk(KERN_ERR "%s: over max segments limit. (%u > %u)\n", + __func__, rq->nr_phys_segments, max_segments); return BLK_STS_IOERR; } diff --git a/block/blk.h b/block/blk.h index e835f21d48af0..cc4e8873dfdea 100644 --- a/block/blk.h +++ b/block/blk.h @@ -156,6 +156,13 @@ static inline bool blk_discard_mergable(struct request *req) return false; } +static inline unsigned int blk_rq_get_max_segments(struct request *rq) +{ + if (req_op(rq) == REQ_OP_DISCARD) + return queue_max_discard_segments(rq->q); + return queue_max_segments(rq->q); +} + static inline unsigned int blk_queue_get_max_sectors(struct request_queue *q, enum req_op op) { From 51287dcb00cc715c27bf6a6b4dbd431621c5b65a Mon Sep 17 00:00:00 2001 From: Marco Elver Date: Fri, 24 Feb 2023 09:59:39 +0100 Subject: [PATCH 675/765] kasan: emit different calls for instrumentable memintrinsics Clang 15 provides an option to prefix memcpy/memset/memmove calls with __asan_/__hwasan_ in instrumented functions: https://reviews.llvm.org/D122724 GCC will add support in future: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108777 Use it to regain KASAN instrumentation of memcpy/memset/memmove on architectures that require noinstr to be really free from instrumented mem*() functions (all GENERIC_ENTRY architectures). Link: https://lkml.kernel.org/r/20230224085942.1791837-1-elver@google.com Fixes: 69d4c0d32186 ("entry, kasan, x86: Disallow overriding mem*() functions") Signed-off-by: Marco Elver Acked-by: Peter Zijlstra (Intel) Reviewed-by: Andrey Konovalov Tested-by: Linux Kernel Functional Testing Tested-by: Naresh Kamboju Cc: Alexander Potapenko Cc: Andrey Ryabinin Cc: Borislav Petkov (AMD) Cc: Dave Hansen Cc: Dmitry Vyukov Cc: Ingo Molnar Cc: Jakub Jelinek Cc: kasan-dev@googlegroups.com Cc: Kees Cook Cc: Linux Kernel Functional Testing Cc: Nathan Chancellor # build only Cc: Nick Desaulniers Cc: Nicolas Schier Cc: Thomas Gleixner Cc: Vincenzo Frascino Signed-off-by: Andrew Morton --- mm/kasan/kasan.h | 4 ++++ mm/kasan/shadow.c | 11 +++++++++++ scripts/Makefile.kasan | 8 ++++++++ 3 files changed, 23 insertions(+) diff --git a/mm/kasan/kasan.h b/mm/kasan/kasan.h index 9377b0789edc2..a61eeee3095a9 100644 --- a/mm/kasan/kasan.h +++ b/mm/kasan/kasan.h @@ -666,4 +666,8 @@ void __hwasan_storeN_noabort(unsigned long addr, size_t size); void __hwasan_tag_memory(unsigned long addr, u8 tag, unsigned long size); +void *__hwasan_memset(void *addr, int c, size_t len); +void *__hwasan_memmove(void *dest, const void *src, size_t len); +void *__hwasan_memcpy(void *dest, const void *src, size_t len); + #endif /* __MM_KASAN_KASAN_H */ diff --git a/mm/kasan/shadow.c b/mm/kasan/shadow.c index 3703983a8e556..90186122b21b0 100644 --- a/mm/kasan/shadow.c +++ b/mm/kasan/shadow.c @@ -107,6 +107,17 @@ void *__asan_memcpy(void *dest, const void *src, size_t len) } EXPORT_SYMBOL(__asan_memcpy); +#ifdef CONFIG_KASAN_SW_TAGS +void *__hwasan_memset(void *addr, int c, size_t len) __alias(__asan_memset); +EXPORT_SYMBOL(__hwasan_memset); +#ifdef __HAVE_ARCH_MEMMOVE +void *__hwasan_memmove(void *dest, const void *src, size_t len) __alias(__asan_memmove); +EXPORT_SYMBOL(__hwasan_memmove); +#endif +void *__hwasan_memcpy(void *dest, const void *src, size_t len) __alias(__asan_memcpy); +EXPORT_SYMBOL(__hwasan_memcpy); +#endif + void kasan_poison(const void *addr, size_t size, u8 value, bool init) { void *shadow_start, *shadow_end; diff --git a/scripts/Makefile.kasan b/scripts/Makefile.kasan index b9e94c5e70970..fa9f836f8039d 100644 --- a/scripts/Makefile.kasan +++ b/scripts/Makefile.kasan @@ -38,6 +38,11 @@ endif CFLAGS_KASAN += $(call cc-param,asan-stack=$(stack_enable)) +# Instrument memcpy/memset/memmove calls by using instrumented __asan_mem*() +# instead. With compilers that don't support this option, compiler-inserted +# memintrinsics won't be checked by KASAN on GENERIC_ENTRY architectures. +CFLAGS_KASAN += $(call cc-param,asan-kernel-mem-intrinsic-prefix=1) + endif # CONFIG_KASAN_GENERIC ifdef CONFIG_KASAN_SW_TAGS @@ -54,6 +59,9 @@ CFLAGS_KASAN := -fsanitize=kernel-hwaddress \ $(call cc-param,hwasan-inline-all-checks=0) \ $(instrumentation_flags) +# Instrument memcpy/memset/memmove calls by using instrumented __hwasan_mem*(). +CFLAGS_KASAN += $(call cc-param,hwasan-kernel-mem-intrinsic-prefix=1) + endif # CONFIG_KASAN_SW_TAGS export CFLAGS_KASAN CFLAGS_KASAN_NOSANITIZE From 36be5cba99f6f9984a9a9f0454f95a38f4184d3e Mon Sep 17 00:00:00 2001 From: Marco Elver Date: Fri, 24 Feb 2023 09:59:40 +0100 Subject: [PATCH 676/765] kasan: treat meminstrinsic as builtins in uninstrumented files Where the compiler instruments meminstrinsics by generating calls to __asan/__hwasan_ prefixed functions, let the compiler consider memintrinsics as builtin again. To do so, never override memset/memmove/memcpy if the compiler does the correct instrumentation - even on !GENERIC_ENTRY architectures. [elver@google.com: powerpc: don't rename memintrinsics if compiler adds prefixes] Link: https://lore.kernel.org/all/20230224085942.1791837-1-elver@google.com/ [1] Link: https://lkml.kernel.org/r/20230227094726.3833247-1-elver@google.com Link: https://lkml.kernel.org/r/20230224085942.1791837-2-elver@google.com Fixes: 69d4c0d32186 ("entry, kasan, x86: Disallow overriding mem*() functions") Signed-off-by: Marco Elver Reviewed-by: Andrey Konovalov Tested-by: Linux Kernel Functional Testing Tested-by: Naresh Kamboju Acked-by: Michael Ellerman (powerpc) Cc: Alexander Potapenko Cc: Andrey Ryabinin Cc: Borislav Petkov (AMD) Cc: Dave Hansen Cc: Dmitry Vyukov Cc: Ingo Molnar Cc: Jakub Jelinek Cc: Kees Cook Cc: Nathan Chancellor Cc: Nick Desaulniers Cc: Nicolas Schier Cc: Peter Zijlstra (Intel) Cc: Thomas Gleixner Cc: Vincenzo Frascino Signed-off-by: Andrew Morton --- lib/Kconfig.kasan | 9 +++++++++ mm/kasan/shadow.c | 5 ++++- scripts/Makefile.kasan | 9 +++++++++ 3 files changed, 22 insertions(+), 1 deletion(-) diff --git a/lib/Kconfig.kasan b/lib/Kconfig.kasan index be6ee60202908..fdca89c057452 100644 --- a/lib/Kconfig.kasan +++ b/lib/Kconfig.kasan @@ -49,6 +49,15 @@ menuconfig KASAN if KASAN +config CC_HAS_KASAN_MEMINTRINSIC_PREFIX + def_bool (CC_IS_CLANG && $(cc-option,-fsanitize=kernel-address -mllvm -asan-kernel-mem-intrinsic-prefix=1)) || \ + (CC_IS_GCC && $(cc-option,-fsanitize=kernel-address --param asan-kernel-mem-intrinsic-prefix=1)) + # Don't define it if we don't need it: compilation of the test uses + # this variable to decide how the compiler should treat builtins. + depends on !KASAN_HW_TAGS + help + The compiler is able to prefix memintrinsics with __asan or __hwasan. + choice prompt "KASAN mode" default KASAN_GENERIC diff --git a/mm/kasan/shadow.c b/mm/kasan/shadow.c index 90186122b21b0..c8b86f3273b50 100644 --- a/mm/kasan/shadow.c +++ b/mm/kasan/shadow.c @@ -38,11 +38,14 @@ bool __kasan_check_write(const volatile void *p, unsigned int size) } EXPORT_SYMBOL(__kasan_check_write); -#ifndef CONFIG_GENERIC_ENTRY +#if !defined(CONFIG_CC_HAS_KASAN_MEMINTRINSIC_PREFIX) && !defined(CONFIG_GENERIC_ENTRY) /* * CONFIG_GENERIC_ENTRY relies on compiler emitted mem*() calls to not be * instrumented. KASAN enabled toolchains should emit __asan_mem*() functions * for the sites they want to instrument. + * + * If we have a compiler that can instrument meminstrinsics, never override + * these, so that non-instrumented files can safely consider them as builtins. */ #undef memset void *memset(void *addr, int c, size_t len) diff --git a/scripts/Makefile.kasan b/scripts/Makefile.kasan index fa9f836f8039d..c186110ffa209 100644 --- a/scripts/Makefile.kasan +++ b/scripts/Makefile.kasan @@ -1,5 +1,14 @@ # SPDX-License-Identifier: GPL-2.0 + +ifdef CONFIG_CC_HAS_KASAN_MEMINTRINSIC_PREFIX +# Safe for compiler to generate meminstrinsic calls in uninstrumented files. +CFLAGS_KASAN_NOSANITIZE := +else +# Don't let compiler generate memintrinsic calls in uninstrumented files +# because they are instrumented. CFLAGS_KASAN_NOSANITIZE := -fno-builtin +endif + KASAN_SHADOW_OFFSET ?= $(CONFIG_KASAN_SHADOW_OFFSET) cc-param = $(call cc-option, -mllvm -$(1), $(call cc-option, --param $(1))) From 85f195b12d8b769fa14ef37aa8a5e6bd07458122 Mon Sep 17 00:00:00 2001 From: Marco Elver Date: Fri, 24 Feb 2023 09:59:41 +0100 Subject: [PATCH 677/765] kasan: test: fix test for new meminstrinsic instrumentation The tests for memset/memmove have been failing since they haven't been instrumented in 69d4c0d32186. Fix the test to recognize when memintrinsics aren't instrumented, and skip test cases accordingly. We also need to conditionally pass -fno-builtin to the test, otherwise the instrumentation pass won't recognize memintrinsics and end up not instrumenting them either. Link: https://lkml.kernel.org/r/20230224085942.1791837-3-elver@google.com Fixes: 69d4c0d32186 ("entry, kasan, x86: Disallow overriding mem*() functions") Reported-by: Linux Kernel Functional Testing Signed-off-by: Marco Elver Reviewed-by: Andrey Konovalov Tested-by: Linux Kernel Functional Testing Tested-by: Naresh Kamboju Cc: Alexander Potapenko Cc: Andrey Ryabinin Cc: Borislav Petkov (AMD) Cc: Dave Hansen Cc: Dmitry Vyukov Cc: Ingo Molnar Cc: Jakub Jelinek Cc: Kees Cook Cc: Nathan Chancellor Cc: Nick Desaulniers Cc: Nicolas Schier Cc: Peter Zijlstra (Intel) Cc: Thomas Gleixner Cc: Vincenzo Frascino Signed-off-by: Andrew Morton --- mm/kasan/Makefile | 9 ++++++++- mm/kasan/kasan_test.c | 29 +++++++++++++++++++++++++++++ 2 files changed, 37 insertions(+), 1 deletion(-) diff --git a/mm/kasan/Makefile b/mm/kasan/Makefile index d4837bff3b60f..7634dd2a61285 100644 --- a/mm/kasan/Makefile +++ b/mm/kasan/Makefile @@ -35,7 +35,14 @@ CFLAGS_shadow.o := $(CC_FLAGS_KASAN_RUNTIME) CFLAGS_hw_tags.o := $(CC_FLAGS_KASAN_RUNTIME) CFLAGS_sw_tags.o := $(CC_FLAGS_KASAN_RUNTIME) -CFLAGS_KASAN_TEST := $(CFLAGS_KASAN) -fno-builtin $(call cc-disable-warning, vla) +CFLAGS_KASAN_TEST := $(CFLAGS_KASAN) $(call cc-disable-warning, vla) +ifndef CONFIG_CC_HAS_KASAN_MEMINTRINSIC_PREFIX +# If compiler instruments memintrinsics by prefixing them with __asan/__hwasan, +# we need to treat them normally (as builtins), otherwise the compiler won't +# recognize them as instrumentable. If it doesn't instrument them, we need to +# pass -fno-builtin, so the compiler doesn't inline them. +CFLAGS_KASAN_TEST += -fno-builtin +endif CFLAGS_kasan_test.o := $(CFLAGS_KASAN_TEST) CFLAGS_kasan_test_module.o := $(CFLAGS_KASAN_TEST) diff --git a/mm/kasan/kasan_test.c b/mm/kasan/kasan_test.c index 74cd80c12b251..627eaf1ee1db1 100644 --- a/mm/kasan/kasan_test.c +++ b/mm/kasan/kasan_test.c @@ -165,6 +165,15 @@ static void kasan_test_exit(struct kunit *test) kunit_skip((test), "Test requires " #config "=n"); \ } while (0) +#define KASAN_TEST_NEEDS_CHECKED_MEMINTRINSICS(test) do { \ + if (IS_ENABLED(CONFIG_KASAN_HW_TAGS)) \ + break; /* No compiler instrumentation. */ \ + if (IS_ENABLED(CONFIG_CC_HAS_KASAN_MEMINTRINSIC_PREFIX)) \ + break; /* Should always be instrumented! */ \ + if (IS_ENABLED(CONFIG_GENERIC_ENTRY)) \ + kunit_skip((test), "Test requires checked mem*()"); \ +} while (0) + static void kmalloc_oob_right(struct kunit *test) { char *ptr; @@ -454,6 +463,8 @@ static void kmalloc_oob_16(struct kunit *test) u64 words[2]; } *ptr1, *ptr2; + KASAN_TEST_NEEDS_CHECKED_MEMINTRINSICS(test); + /* This test is specifically crafted for the generic mode. */ KASAN_TEST_NEEDS_CONFIG_ON(test, CONFIG_KASAN_GENERIC); @@ -476,6 +487,8 @@ static void kmalloc_uaf_16(struct kunit *test) u64 words[2]; } *ptr1, *ptr2; + KASAN_TEST_NEEDS_CHECKED_MEMINTRINSICS(test); + ptr1 = kmalloc(sizeof(*ptr1), GFP_KERNEL); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ptr1); @@ -498,6 +511,8 @@ static void kmalloc_oob_memset_2(struct kunit *test) char *ptr; size_t size = 128 - KASAN_GRANULE_SIZE; + KASAN_TEST_NEEDS_CHECKED_MEMINTRINSICS(test); + ptr = kmalloc(size, GFP_KERNEL); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ptr); @@ -511,6 +526,8 @@ static void kmalloc_oob_memset_4(struct kunit *test) char *ptr; size_t size = 128 - KASAN_GRANULE_SIZE; + KASAN_TEST_NEEDS_CHECKED_MEMINTRINSICS(test); + ptr = kmalloc(size, GFP_KERNEL); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ptr); @@ -524,6 +541,8 @@ static void kmalloc_oob_memset_8(struct kunit *test) char *ptr; size_t size = 128 - KASAN_GRANULE_SIZE; + KASAN_TEST_NEEDS_CHECKED_MEMINTRINSICS(test); + ptr = kmalloc(size, GFP_KERNEL); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ptr); @@ -537,6 +556,8 @@ static void kmalloc_oob_memset_16(struct kunit *test) char *ptr; size_t size = 128 - KASAN_GRANULE_SIZE; + KASAN_TEST_NEEDS_CHECKED_MEMINTRINSICS(test); + ptr = kmalloc(size, GFP_KERNEL); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ptr); @@ -550,6 +571,8 @@ static void kmalloc_oob_in_memset(struct kunit *test) char *ptr; size_t size = 128 - KASAN_GRANULE_SIZE; + KASAN_TEST_NEEDS_CHECKED_MEMINTRINSICS(test); + ptr = kmalloc(size, GFP_KERNEL); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ptr); @@ -566,6 +589,8 @@ static void kmalloc_memmove_negative_size(struct kunit *test) size_t size = 64; size_t invalid_size = -2; + KASAN_TEST_NEEDS_CHECKED_MEMINTRINSICS(test); + /* * Hardware tag-based mode doesn't check memmove for negative size. * As a result, this test introduces a side-effect memory corruption, @@ -590,6 +615,8 @@ static void kmalloc_memmove_invalid_size(struct kunit *test) size_t size = 64; size_t invalid_size = size; + KASAN_TEST_NEEDS_CHECKED_MEMINTRINSICS(test); + ptr = kmalloc(size, GFP_KERNEL); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ptr); @@ -618,6 +645,8 @@ static void kmalloc_uaf_memset(struct kunit *test) char *ptr; size_t size = 33; + KASAN_TEST_NEEDS_CHECKED_MEMINTRINSICS(test); + /* * Only generic KASAN uses quarantine, which is required to avoid a * kernel memory corruption this test causes. From 4ec4190be4cf9cc3e0ccaf5f155a5f9066d18950 Mon Sep 17 00:00:00 2001 From: Marco Elver Date: Fri, 24 Feb 2023 09:59:42 +0100 Subject: [PATCH 678/765] kasan, x86: don't rename memintrinsics in uninstrumented files Now that memcpy/memset/memmove are no longer overridden by KASAN, we can just use the normal symbol names in uninstrumented files. Drop the preprocessor redefinitions. Link: https://lkml.kernel.org/r/20230224085942.1791837-4-elver@google.com Fixes: 69d4c0d32186 ("entry, kasan, x86: Disallow overriding mem*() functions") Signed-off-by: Marco Elver Reviewed-by: Andrey Konovalov Cc: Alexander Potapenko Cc: Andrey Ryabinin Cc: Borislav Petkov (AMD) Cc: Dave Hansen Cc: Dmitry Vyukov Cc: Ingo Molnar Cc: Jakub Jelinek Cc: Kees Cook Cc: Linux Kernel Functional Testing Cc: Naresh Kamboju Cc: Nathan Chancellor Cc: Nick Desaulniers Cc: Nicolas Schier Cc: Peter Zijlstra (Intel) Cc: Thomas Gleixner Cc: Vincenzo Frascino Signed-off-by: Andrew Morton --- arch/x86/include/asm/string_64.h | 19 ------------------- 1 file changed, 19 deletions(-) diff --git a/arch/x86/include/asm/string_64.h b/arch/x86/include/asm/string_64.h index 888731ccf1f67..c1e14cee0722d 100644 --- a/arch/x86/include/asm/string_64.h +++ b/arch/x86/include/asm/string_64.h @@ -85,25 +85,6 @@ char *strcpy(char *dest, const char *src); char *strcat(char *dest, const char *src); int strcmp(const char *cs, const char *ct); -#if (defined(CONFIG_KASAN) && !defined(__SANITIZE_ADDRESS__)) -/* - * For files that not instrumented (e.g. mm/slub.c) we - * should use not instrumented version of mem* functions. - */ - -#undef memcpy -#define memcpy(dst, src, len) __memcpy(dst, src, len) -#undef memmove -#define memmove(dst, src, len) __memmove(dst, src, len) -#undef memset -#define memset(s, c, n) __memset(s, c, n) - -#ifndef __NO_FORTIFY -#define __NO_FORTIFY /* FORTIFY_SOURCE uses __builtin_memcpy, etc. */ -#endif - -#endif - #ifdef CONFIG_ARCH_HAS_UACCESS_FLUSHCACHE #define __HAVE_ARCH_MEMCPY_FLUSHCACHE 1 void __memcpy_flushcache(void *dst, const void *src, size_t cnt); From 359d62559f578dc1ac86fd2d198f1455e9d8ce04 Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Thu, 23 Feb 2023 20:26:18 -0800 Subject: [PATCH 679/765] lib: parser: update documentation for match_NUMBER functions commit 67222c4ba8af ("lib: parser: optimize match_NUMBER apis to use local array") removed -ENOMEM as a possible return value, so update the comments accordingly. Link: https://lkml.kernel.org/r/20230224042618.9092-1-ebiggers@kernel.org Fixes: 67222c4ba8af ("lib: parser: optimize match_NUMBER apis to use local array") Signed-off-by: Eric Biggers Cc: Li Lingfeng Cc: Tejun Heo Cc: Yu Kuai Signed-off-by: Andrew Morton --- lib/parser.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/lib/parser.c b/lib/parser.c index 2b5e2b480253f..f4eafb9d74e6f 100644 --- a/lib/parser.c +++ b/lib/parser.c @@ -133,7 +133,7 @@ EXPORT_SYMBOL(match_token); * as a number in that base. * * Return: On success, sets @result to the integer represented by the - * string and returns 0. Returns -ENOMEM, -EINVAL, or -ERANGE on failure. + * string and returns 0. Returns -EINVAL or -ERANGE on failure. */ static int match_number(substring_t *s, int *result, int base) { @@ -165,7 +165,7 @@ static int match_number(substring_t *s, int *result, int base) * as a number in that base. * * Return: On success, sets @result to the integer represented by the - * string and returns 0. Returns -ENOMEM, -EINVAL, or -ERANGE on failure. + * string and returns 0. Returns -EINVAL or -ERANGE on failure. */ static int match_u64int(substring_t *s, u64 *result, int base) { @@ -189,7 +189,7 @@ static int match_u64int(substring_t *s, u64 *result, int base) * Description: Attempts to parse the &substring_t @s as a decimal integer. * * Return: On success, sets @result to the integer represented by the string - * and returns 0. Returns -ENOMEM, -EINVAL, or -ERANGE on failure. + * and returns 0. Returns -EINVAL or -ERANGE on failure. */ int match_int(substring_t *s, int *result) { @@ -205,7 +205,7 @@ EXPORT_SYMBOL(match_int); * Description: Attempts to parse the &substring_t @s as a decimal integer. * * Return: On success, sets @result to the integer represented by the string - * and returns 0. Returns -ENOMEM, -EINVAL, or -ERANGE on failure. + * and returns 0. Returns -EINVAL or -ERANGE on failure. */ int match_uint(substring_t *s, unsigned int *result) { @@ -228,7 +228,7 @@ EXPORT_SYMBOL(match_uint); * integer. * * Return: On success, sets @result to the integer represented by the string - * and returns 0. Returns -ENOMEM, -EINVAL, or -ERANGE on failure. + * and returns 0. Returns -EINVAL or -ERANGE on failure. */ int match_u64(substring_t *s, u64 *result) { @@ -244,7 +244,7 @@ EXPORT_SYMBOL(match_u64); * Description: Attempts to parse the &substring_t @s as an octal integer. * * Return: On success, sets @result to the integer represented by the string - * and returns 0. Returns -ENOMEM, -EINVAL, or -ERANGE on failure. + * and returns 0. Returns -EINVAL or -ERANGE on failure. */ int match_octal(substring_t *s, int *result) { @@ -260,7 +260,7 @@ EXPORT_SYMBOL(match_octal); * Description: Attempts to parse the &substring_t @s as a hexadecimal integer. * * Return: On success, sets @result to the integer represented by the string - * and returns 0. Returns -ENOMEM, -EINVAL, or -ERANGE on failure. + * and returns 0. Returns -EINVAL or -ERANGE on failure. */ int match_hex(substring_t *s, int *result) { From b905039e428d639adeebb719b76f98865ea38d4d Mon Sep 17 00:00:00 2001 From: "Guilherme G. Piccoli" Date: Sun, 26 Feb 2023 13:08:38 -0300 Subject: [PATCH 680/765] panic: fix the panic_print NMI backtrace setting Commit 8d470a45d1a6 ("panic: add option to dump all CPUs backtraces in panic_print") introduced a setting for the "panic_print" kernel parameter to allow users to request a NMI backtrace on panic. Problem is that the panic_print handling happens after the secondary CPUs are already disabled, hence this option ended-up being kind of a no-op - kernel skips the NMI trace in idling CPUs, which is the case of offline CPUs. Fix it by checking the NMI backtrace bit in the panic_print prior to the CPU disabling function. Link: https://lkml.kernel.org/r/20230226160838.414257-1-gpiccoli@igalia.com Fixes: 8d470a45d1a6 ("panic: add option to dump all CPUs backtraces in panic_print") Signed-off-by: Guilherme G. Piccoli Cc: Cc: Baoquan He Cc: Dave Young Cc: Feng Tang Cc: HATAYAMA Daisuke Cc: Hidehiro Kawai Cc: Kees Cook Cc: Michael Kelley Cc: Petr Mladek Cc: Vivek Goyal Signed-off-by: Andrew Morton --- kernel/panic.c | 44 ++++++++++++++++++++++++++------------------ 1 file changed, 26 insertions(+), 18 deletions(-) diff --git a/kernel/panic.c b/kernel/panic.c index 487f5b03bf835..5cfea8302d23a 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -212,9 +212,6 @@ static void panic_print_sys_info(bool console_flush) return; } - if (panic_print & PANIC_PRINT_ALL_CPU_BT) - trigger_all_cpu_backtrace(); - if (panic_print & PANIC_PRINT_TASK_INFO) show_state(); @@ -244,6 +241,30 @@ void check_panic_on_warn(const char *origin) origin, limit); } +/* + * Helper that triggers the NMI backtrace (if set in panic_print) + * and then performs the secondary CPUs shutdown - we cannot have + * the NMI backtrace after the CPUs are off! + */ +static void panic_other_cpus_shutdown(bool crash_kexec) +{ + if (panic_print & PANIC_PRINT_ALL_CPU_BT) + trigger_all_cpu_backtrace(); + + /* + * Note that smp_send_stop() is the usual SMP shutdown function, + * which unfortunately may not be hardened to work in a panic + * situation. If we want to do crash dump after notifier calls + * and kmsg_dump, we will need architecture dependent extra + * bits in addition to stopping other CPUs, hence we rely on + * crash_smp_send_stop() for that. + */ + if (!crash_kexec) + smp_send_stop(); + else + crash_smp_send_stop(); +} + /** * panic - halt the system * @fmt: The text string to print @@ -334,23 +355,10 @@ void panic(const char *fmt, ...) * * Bypass the panic_cpu check and call __crash_kexec directly. */ - if (!_crash_kexec_post_notifiers) { + if (!_crash_kexec_post_notifiers) __crash_kexec(NULL); - /* - * Note smp_send_stop is the usual smp shutdown function, which - * unfortunately means it may not be hardened to work in a - * panic situation. - */ - smp_send_stop(); - } else { - /* - * If we want to do crash dump after notifier calls and - * kmsg_dump, we will need architecture dependent extra - * works in addition to stopping other CPUs. - */ - crash_smp_send_stop(); - } + panic_other_cpus_shutdown(_crash_kexec_post_notifiers); /* * Run any panic handlers, including those that might need to From 07db5e247ab5858439b14dd7cc1fe538b9efcf32 Mon Sep 17 00:00:00 2001 From: Dongliang Mu Date: Sun, 26 Feb 2023 20:49:47 +0800 Subject: [PATCH 681/765] fs: hfsplus: fix UAF issue in hfsplus_put_super The current hfsplus_put_super first calls hfs_btree_close on sbi->ext_tree, then invokes iput on sbi->hidden_dir, resulting in an use-after-free issue in hfsplus_release_folio. As shown in hfsplus_fill_super, the error handling code also calls iput before hfs_btree_close. To fix this error, we move all iput calls before hfsplus_btree_close. Note that this patch is tested on Syzbot. Link: https://lkml.kernel.org/r/20230226124948.3175736-1-mudongliangabcd@gmail.com Reported-by: syzbot+57e3e98f7e3b80f64d56@syzkaller.appspotmail.com Tested-by: Dongliang Mu Signed-off-by: Dongliang Mu Cc: Bart Van Assche Cc: Jens Axboe Cc: Muchun Song Cc: Roman Gushchin Cc: "Theodore Ts'o" Cc: Signed-off-by: Andrew Morton --- fs/hfsplus/super.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 122ed89ebf9f2..1986b4f18a901 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -295,11 +295,11 @@ static void hfsplus_put_super(struct super_block *sb) hfsplus_sync_fs(sb, 1); } + iput(sbi->alloc_file); + iput(sbi->hidden_dir); hfs_btree_close(sbi->attr_tree); hfs_btree_close(sbi->cat_tree); hfs_btree_close(sbi->ext_tree); - iput(sbi->alloc_file); - iput(sbi->hidden_dir); kfree(sbi->s_vhdr_buf); kfree(sbi->s_backup_vhdr_buf); unload_nls(sbi->nls); From 3e35102666f873a135d31a726ac1ec8af4905206 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sun, 26 Feb 2023 12:31:11 -0800 Subject: [PATCH 682/765] fs/cramfs/inode.c: initialize file_ra_state file_ra_state_init() assumes that the file_ra_state has been zeroed out. Fixes a KMSAN used-unintialized issue (at least). Fixes: cf948cbc35e80 ("cramfs: read_mapping_page() is synchronous") Reported-by: syzbot Link: https://lkml.kernel.org/r/0000000000008f74e905f56df987@google.com Cc: Matthew Wilcox Cc: Nicolas Pitre Cc: Signed-off-by: Andrew Morton --- fs/cramfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index e3d168911dbe1..006ef68d7ff6a 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -183,7 +183,7 @@ static void *cramfs_blkdev_read(struct super_block *sb, unsigned int offset, unsigned int len) { struct address_space *mapping = sb->s_bdev->bd_inode->i_mapping; - struct file_ra_state ra; + struct file_ra_state ra = {}; struct page *pages[BLKS_PER_BUF]; unsigned i, blocknr, buffer; unsigned long devsize; From 6a41de1eb2168c78079cc5e6bbbec6d1490c54ed Mon Sep 17 00:00:00 2001 From: Konrad Dybcio Date: Tue, 28 Feb 2023 16:33:35 +0100 Subject: [PATCH 683/765] mailmap: map Vikash Garodia's old address to his current one Vikash's old email is still picked up by the likes of get_maintainer.pl and keeps bouncing. Map it to his current one. Link: https://lkml.kernel.org/r/20230228153335.907164-3-konrad.dybcio@linaro.org Signed-off-by: Konrad Dybcio Cc: Vikash Garodia Signed-off-by: Andrew Morton --- .mailmap | 1 + 1 file changed, 1 insertion(+) diff --git a/.mailmap b/.mailmap index fb65947d671d8..3366d9b630807 100644 --- a/.mailmap +++ b/.mailmap @@ -442,6 +442,7 @@ Vasily Averin Vasily Averin Vasily Averin Valentin Schneider +Vikash Garodia Vinod Koul Vinod Koul Vinod Koul From ecf1d926661bec4080a79c0ac9dbfe02b31702cf Mon Sep 17 00:00:00 2001 From: Konrad Dybcio Date: Tue, 28 Feb 2023 16:33:34 +0100 Subject: [PATCH 684/765] mailmap: map Dikshita Agarwal's old address to his current one Dikshita's old email is still picked up by the likes of get_maintainer.pl and keeps bouncing. Map it to his current one. Link: https://lkml.kernel.org/r/20230228153335.907164-2-konrad.dybcio@linaro.org Signed-off-by: Konrad Dybcio Cc: Dikshita Agarwal Signed-off-by: Andrew Morton --- .mailmap | 1 + 1 file changed, 1 insertion(+) diff --git a/.mailmap b/.mailmap index 3366d9b630807..5367faaf78312 100644 --- a/.mailmap +++ b/.mailmap @@ -121,6 +121,7 @@ Dengcheng Zhu Dengcheng Zhu Dengcheng Zhu +Dikshita Agarwal Dmitry Baryshkov Dmitry Baryshkov <[dbaryshkov@gmail.com]> Dmitry Baryshkov From e57cf3639c323eeed05d3725fd82f91b349adca8 Mon Sep 17 00:00:00 2001 From: Yuiko Oshino Date: Wed, 1 Mar 2023 08:43:07 -0700 Subject: [PATCH 685/765] net: lan78xx: fix accessing the LAN7800's internal phy specific registers from the MAC driver Move the LAN7800 internal phy (phy ID 0x0007c132) specific register accesses to the phy driver (microchip.c). Fix the error reported by Enguerrand de Ribaucourt in December 2022, "Some operations during the cable switch workaround modify the register LAN88XX_INT_MASK of the PHY. However, this register is specific to the LAN8835 PHY. For instance, if a DP8322I PHY is connected to the LAN7801, that register (0x19), corresponds to the LED and MAC address configuration, resulting in unapropriate behavior." I did not test with the DP8322I PHY, but I tested with an EVB-LAN7800 with the internal PHY. Fixes: 14437e3fa284 ("lan78xx: workaround of forced 100 Full/Half duplex mode error") Signed-off-by: Yuiko Oshino Reviewed-by: Andrew Lunn Link: https://lore.kernel.org/r/20230301154307.30438-1-yuiko.oshino@microchip.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/microchip.c | 32 ++++++++++++++++++++++++++++++++ drivers/net/usb/lan78xx.c | 27 +-------------------------- 2 files changed, 33 insertions(+), 26 deletions(-) diff --git a/drivers/net/phy/microchip.c b/drivers/net/phy/microchip.c index ccecee2524ce6..0b88635f4fbca 100644 --- a/drivers/net/phy/microchip.c +++ b/drivers/net/phy/microchip.c @@ -342,6 +342,37 @@ static int lan88xx_config_aneg(struct phy_device *phydev) return genphy_config_aneg(phydev); } +static void lan88xx_link_change_notify(struct phy_device *phydev) +{ + int temp; + + /* At forced 100 F/H mode, chip may fail to set mode correctly + * when cable is switched between long(~50+m) and short one. + * As workaround, set to 10 before setting to 100 + * at forced 100 F/H mode. + */ + if (!phydev->autoneg && phydev->speed == 100) { + /* disable phy interrupt */ + temp = phy_read(phydev, LAN88XX_INT_MASK); + temp &= ~LAN88XX_INT_MASK_MDINTPIN_EN_; + phy_write(phydev, LAN88XX_INT_MASK, temp); + + temp = phy_read(phydev, MII_BMCR); + temp &= ~(BMCR_SPEED100 | BMCR_SPEED1000); + phy_write(phydev, MII_BMCR, temp); /* set to 10 first */ + temp |= BMCR_SPEED100; + phy_write(phydev, MII_BMCR, temp); /* set to 100 later */ + + /* clear pending interrupt generated while workaround */ + temp = phy_read(phydev, LAN88XX_INT_STS); + + /* enable phy interrupt back */ + temp = phy_read(phydev, LAN88XX_INT_MASK); + temp |= LAN88XX_INT_MASK_MDINTPIN_EN_; + phy_write(phydev, LAN88XX_INT_MASK, temp); + } +} + static struct phy_driver microchip_phy_driver[] = { { .phy_id = 0x0007c132, @@ -359,6 +390,7 @@ static struct phy_driver microchip_phy_driver[] = { .config_init = lan88xx_config_init, .config_aneg = lan88xx_config_aneg, + .link_change_notify = lan88xx_link_change_notify, .config_intr = lan88xx_phy_config_intr, .handle_interrupt = lan88xx_handle_interrupt, diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c index f18ab8e220db7..068488890d57b 100644 --- a/drivers/net/usb/lan78xx.c +++ b/drivers/net/usb/lan78xx.c @@ -2115,33 +2115,8 @@ static void lan78xx_remove_mdio(struct lan78xx_net *dev) static void lan78xx_link_status_change(struct net_device *net) { struct phy_device *phydev = net->phydev; - int temp; - - /* At forced 100 F/H mode, chip may fail to set mode correctly - * when cable is switched between long(~50+m) and short one. - * As workaround, set to 10 before setting to 100 - * at forced 100 F/H mode. - */ - if (!phydev->autoneg && (phydev->speed == 100)) { - /* disable phy interrupt */ - temp = phy_read(phydev, LAN88XX_INT_MASK); - temp &= ~LAN88XX_INT_MASK_MDINTPIN_EN_; - phy_write(phydev, LAN88XX_INT_MASK, temp); - temp = phy_read(phydev, MII_BMCR); - temp &= ~(BMCR_SPEED100 | BMCR_SPEED1000); - phy_write(phydev, MII_BMCR, temp); /* set to 10 first */ - temp |= BMCR_SPEED100; - phy_write(phydev, MII_BMCR, temp); /* set to 100 later */ - - /* clear pending interrupt generated while workaround */ - temp = phy_read(phydev, LAN88XX_INT_STS); - - /* enable phy interrupt back */ - temp = phy_read(phydev, LAN88XX_INT_MASK); - temp |= LAN88XX_INT_MASK_MDINTPIN_EN_; - phy_write(phydev, LAN88XX_INT_MASK, temp); - } + phy_print_status(phydev); } static int irq_map(struct irq_domain *d, unsigned int irq, From 9781e98a97110f5e76999058368b4be76a788484 Mon Sep 17 00:00:00 2001 From: Shigeru Yoshida Date: Thu, 2 Mar 2023 01:39:13 +0900 Subject: [PATCH 686/765] net: caif: Fix use-after-free in cfusbl_device_notify() syzbot reported use-after-free in cfusbl_device_notify() [1]. This causes a stack trace like below: BUG: KASAN: use-after-free in cfusbl_device_notify+0x7c9/0x870 net/caif/caif_usb.c:138 Read of size 8 at addr ffff88807ac4e6f0 by task kworker/u4:6/1214 CPU: 0 PID: 1214 Comm: kworker/u4:6 Not tainted 5.19.0-rc3-syzkaller-00146-g92f20ff72066 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Workqueue: netns cleanup_net Call Trace: __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0xcd/0x134 lib/dump_stack.c:106 print_address_description.constprop.0.cold+0xeb/0x467 mm/kasan/report.c:313 print_report mm/kasan/report.c:429 [inline] kasan_report.cold+0xf4/0x1c6 mm/kasan/report.c:491 cfusbl_device_notify+0x7c9/0x870 net/caif/caif_usb.c:138 notifier_call_chain+0xb5/0x200 kernel/notifier.c:87 call_netdevice_notifiers_info+0xb5/0x130 net/core/dev.c:1945 call_netdevice_notifiers_extack net/core/dev.c:1983 [inline] call_netdevice_notifiers net/core/dev.c:1997 [inline] netdev_wait_allrefs_any net/core/dev.c:10227 [inline] netdev_run_todo+0xbc0/0x10f0 net/core/dev.c:10341 default_device_exit_batch+0x44e/0x590 net/core/dev.c:11334 ops_exit_list+0x125/0x170 net/core/net_namespace.c:167 cleanup_net+0x4ea/0xb00 net/core/net_namespace.c:594 process_one_work+0x996/0x1610 kernel/workqueue.c:2289 worker_thread+0x665/0x1080 kernel/workqueue.c:2436 kthread+0x2e9/0x3a0 kernel/kthread.c:376 ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:302 When unregistering a net device, unregister_netdevice_many_notify() sets the device's reg_state to NETREG_UNREGISTERING, calls notifiers with NETDEV_UNREGISTER, and adds the device to the todo list. Later on, devices in the todo list are processed by netdev_run_todo(). netdev_run_todo() waits devices' reference count become 1 while rebdoadcasting NETDEV_UNREGISTER notification. When cfusbl_device_notify() is called with NETDEV_UNREGISTER multiple times, the parent device might be freed. This could cause UAF. Processing NETDEV_UNREGISTER multiple times also causes inbalance of reference count for the module. This patch fixes the issue by accepting only first NETDEV_UNREGISTER notification. Fixes: 7ad65bf68d70 ("caif: Add support for CAIF over CDC NCM USB interface") CC: sjur.brandeland@stericsson.com Reported-by: syzbot+b563d33852b893653a9e@syzkaller.appspotmail.com Link: https://syzkaller.appspot.com/bug?id=c3bfd8e2450adab3bffe4d80821fbbced600407f [1] Signed-off-by: Shigeru Yoshida Link: https://lore.kernel.org/r/20230301163913.391304-1-syoshida@redhat.com Signed-off-by: Jakub Kicinski --- net/caif/caif_usb.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/caif/caif_usb.c b/net/caif/caif_usb.c index ebc202ffdd8d8..bf61ea4b8132d 100644 --- a/net/caif/caif_usb.c +++ b/net/caif/caif_usb.c @@ -134,6 +134,9 @@ static int cfusbl_device_notify(struct notifier_block *me, unsigned long what, struct usb_device *usbdev; int res; + if (what == NETDEV_UNREGISTER && dev->reg_state >= NETREG_UNREGISTERED) + return 0; + /* Check whether we have a NCM device, and find its VID/PID. */ if (!(dev->dev.parent && dev->dev.parent->driver && strcmp(dev->dev.parent->driver->name, "cdc_ncm") == 0)) From 7cf93538e087a792cb476008a757ab7c1c23b68c Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 1 Mar 2023 10:36:40 -0800 Subject: [PATCH 687/765] tools: ynl: fully inherit attrs in subsets To avoid having to repeat the entire definition of an attribute (including the value) use the Attr object from the original set. In fact this is already the documented expectation. Fixes: be5bea1cc0bf ("net: add basic C code generators for Netlink") Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- Documentation/userspace-api/netlink/specs.rst | 3 ++- tools/net/ynl/lib/nlspec.py | 23 ++++++++++++------- 2 files changed, 17 insertions(+), 9 deletions(-) diff --git a/Documentation/userspace-api/netlink/specs.rst b/Documentation/userspace-api/netlink/specs.rst index 6ffe8137cd902..1424ab1b9b337 100644 --- a/Documentation/userspace-api/netlink/specs.rst +++ b/Documentation/userspace-api/netlink/specs.rst @@ -199,7 +199,8 @@ The ``value`` property can be skipped, in which case the attribute ID will be the value of the previous attribute plus one (recursively) and ``0`` for the first attribute in the attribute set. -Note that the ``value`` of an attribute is defined only in its main set. +Note that the ``value`` of an attribute is defined only in its main set +(not in subsets). enum ~~~~ diff --git a/tools/net/ynl/lib/nlspec.py b/tools/net/ynl/lib/nlspec.py index 71da568e2c28f..dff31dad36c50 100644 --- a/tools/net/ynl/lib/nlspec.py +++ b/tools/net/ynl/lib/nlspec.py @@ -95,15 +95,22 @@ def __init__(self, family, yaml): self.attrs = collections.OrderedDict() self.attrs_by_val = collections.OrderedDict() - val = 0 - for elem in self.yaml['attributes']: - if 'value' in elem: - val = elem['value'] + if self.subset_of is None: + val = 0 + for elem in self.yaml['attributes']: + if 'value' in elem: + val = elem['value'] - attr = self.new_attr(elem, val) - self.attrs[attr.name] = attr - self.attrs_by_val[attr.value] = attr - val += 1 + attr = self.new_attr(elem, val) + self.attrs[attr.name] = attr + self.attrs_by_val[attr.value] = attr + val += 1 + else: + real_set = family.attr_sets[self.subset_of] + for elem in self.yaml['attributes']: + attr = real_set[elem['name']] + self.attrs[attr.name] = attr + self.attrs_by_val[attr.value] = attr def new_attr(self, elem, value): return SpecAttr(self.family, self, elem, value) From ad4fafcde5bc1badb8fc2c7f260a5d6b83a038e4 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 1 Mar 2023 10:36:41 -0800 Subject: [PATCH 688/765] tools: ynl: use 1 as the default for first entry in attrs/ops Pretty much all families use value: 1 or reserve as unspec the first entry in attribute set and the first operation. Make this the default. Update documentation (the doc for values of operations just refers back to doc for attrs so updating only attrs). Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- Documentation/userspace-api/netlink/specs.rst | 7 ++++++- tools/net/ynl/lib/nlspec.py | 6 +++--- tools/net/ynl/ynl-gen-c.py | 7 +++++-- 3 files changed, 14 insertions(+), 6 deletions(-) diff --git a/Documentation/userspace-api/netlink/specs.rst b/Documentation/userspace-api/netlink/specs.rst index 1424ab1b9b337..32e53328d113d 100644 --- a/Documentation/userspace-api/netlink/specs.rst +++ b/Documentation/userspace-api/netlink/specs.rst @@ -197,7 +197,12 @@ value Numerical attribute ID, used in serialized Netlink messages. The ``value`` property can be skipped, in which case the attribute ID will be the value of the previous attribute plus one (recursively) -and ``0`` for the first attribute in the attribute set. +and ``1`` for the first attribute in the attribute set. + +Attributes (and operations) use ``1`` as the default value for the first +entry (unlike enums in definitions which start from ``0``) because +entry ``0`` is almost always reserved as undefined. Spec can explicitly +set value to ``0`` if needed. Note that the ``value`` of an attribute is defined only in its main set (not in subsets). diff --git a/tools/net/ynl/lib/nlspec.py b/tools/net/ynl/lib/nlspec.py index dff31dad36c50..9d394e50de23d 100644 --- a/tools/net/ynl/lib/nlspec.py +++ b/tools/net/ynl/lib/nlspec.py @@ -96,7 +96,7 @@ def __init__(self, family, yaml): self.attrs_by_val = collections.OrderedDict() if self.subset_of is None: - val = 0 + val = 1 for elem in self.yaml['attributes']: if 'value' in elem: val = elem['value'] @@ -252,7 +252,7 @@ def add_unresolved(self, elem): self._resolution_list.append(elem) def _dictify_ops_unified(self): - val = 0 + val = 1 for elem in self.yaml['operations']['list']: if 'value' in elem: val = elem['value'] @@ -263,7 +263,7 @@ def _dictify_ops_unified(self): self.msgs[op.name] = op def _dictify_ops_directional(self): - req_val = rsp_val = 0 + req_val = rsp_val = 1 for elem in self.yaml['operations']['list']: if 'notify' in elem: if 'value' in elem: diff --git a/tools/net/ynl/ynl-gen-c.py b/tools/net/ynl/ynl-gen-c.py index 274e9c566f61b..62f8f2c3c56c8 100755 --- a/tools/net/ynl/ynl-gen-c.py +++ b/tools/net/ynl/ynl-gen-c.py @@ -2044,14 +2044,17 @@ def render_uapi(family, cw): max_value = f"({cnt_name} - 1)" uapi_enum_start(family, cw, family['operations'], 'enum-name') + val = 0 for op in family.msgs.values(): if separate_ntf and ('notify' in op or 'event' in op): continue suffix = ',' - if 'value' in op: - suffix = f" = {op['value']}," + if op.value != val: + suffix = f" = {op.value}," + val = op.value cw.p(op.enum_name + suffix) + val += 1 cw.nl() cw.p(cnt_name + ('' if max_by_define else ',')) if not max_by_define: From bcec7171eba901cc5f427ebbad9fe17ed4749b8f Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 1 Mar 2023 10:36:42 -0800 Subject: [PATCH 689/765] netlink: specs: update for codegen enumerating from 1 Now that the codegen rules had been changed we can update the specs to reflect the new default. Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- Documentation/netlink/specs/ethtool.yaml | 15 --------------- Documentation/netlink/specs/fou.yaml | 2 ++ Documentation/netlink/specs/netdev.yaml | 2 -- 3 files changed, 2 insertions(+), 17 deletions(-) diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml index 08b776908d152..35c462bce56f6 100644 --- a/Documentation/netlink/specs/ethtool.yaml +++ b/Documentation/netlink/specs/ethtool.yaml @@ -11,7 +11,6 @@ attribute-sets: - name: dev-index type: u32 - value: 1 - name: dev-name type: string @@ -25,7 +24,6 @@ attribute-sets: - name: index type: u32 - value: 1 - name: name type: string @@ -39,14 +37,12 @@ attribute-sets: name: bit type: nest nested-attributes: bitset-bit - value: 1 - name: bitset attributes: - name: nomask type: flag - value: 1 - name: size type: u32 @@ -61,7 +57,6 @@ attribute-sets: - name: index type: u32 - value: 1 - name: value type: string @@ -71,7 +66,6 @@ attribute-sets: - name: string type: nest - value: 1 multi-attr: true nested-attributes: string - @@ -80,7 +74,6 @@ attribute-sets: - name: id type: u32 - value: 1 - name: count type: u32 @@ -96,14 +89,12 @@ attribute-sets: name: stringset type: nest multi-attr: true - value: 1 nested-attributes: stringset - name: strset attributes: - name: header - value: 1 type: nest nested-attributes: header - @@ -119,7 +110,6 @@ attribute-sets: attributes: - name: header - value: 1 type: nest nested-attributes: header - @@ -132,7 +122,6 @@ attribute-sets: attributes: - name: header - value: 1 type: nest nested-attributes: header - @@ -180,7 +169,6 @@ attribute-sets: attributes: - name: pad - value: 1 type: pad - name: reassembly-errors @@ -205,7 +193,6 @@ attribute-sets: attributes: - name: header - value: 1 type: nest nested-attributes: header - @@ -251,13 +238,11 @@ operations: do: &strset-get-op request: - value: 1 attributes: - header - stringsets - counts-only reply: - value: 1 attributes: - header - stringsets diff --git a/Documentation/netlink/specs/fou.yaml b/Documentation/netlink/specs/fou.yaml index 266c386eedf3a..cca4cf98f03ab 100644 --- a/Documentation/netlink/specs/fou.yaml +++ b/Documentation/netlink/specs/fou.yaml @@ -26,6 +26,7 @@ attribute-sets: - name: unspec type: unused + value: 0 - name: port type: u16 @@ -71,6 +72,7 @@ operations: - name: unspec doc: unused + value: 0 - name: add diff --git a/Documentation/netlink/specs/netdev.yaml b/Documentation/netlink/specs/netdev.yaml index cffef09729f15..ba9ee13cf7296 100644 --- a/Documentation/netlink/specs/netdev.yaml +++ b/Documentation/netlink/specs/netdev.yaml @@ -48,7 +48,6 @@ attribute-sets: name: ifindex doc: netdev ifindex type: u32 - value: 1 checks: min: 1 - @@ -66,7 +65,6 @@ operations: - name: dev-get doc: Get / dump information about a netdev. - value: 1 attribute-set: dev do: request: From 84cba1840e68430325ac133a11be06bfb2f7acd8 Mon Sep 17 00:00:00 2001 From: Petr Oros Date: Wed, 1 Mar 2023 21:47:07 +0100 Subject: [PATCH 690/765] ice: copy last block omitted in ice_get_module_eeprom() ice_get_module_eeprom() is broken since commit e9c9692c8a81 ("ice: Reimplement module reads used by ethtool") In this refactor, ice_get_module_eeprom() reads the eeprom in blocks of size 8. But the condition that should protect the buffer overflow ignores the last block. The last block always contains zeros. Bug uncovered by ethtool upstream commit 9538f384b535 ("netlink: eeprom: Defer page requests to individual parsers") After this commit, ethtool reads a block with length = 1; to read the SFF-8024 identifier value. unpatched driver: $ ethtool -m enp65s0f0np0 offset 0x90 length 8 Offset Values ------ ------ 0x0090: 00 00 00 00 00 00 00 00 $ ethtool -m enp65s0f0np0 offset 0x90 length 12 Offset Values ------ ------ 0x0090: 00 00 01 a0 4d 65 6c 6c 00 00 00 00 $ $ ethtool -m enp65s0f0np0 Offset Values ------ ------ 0x0000: 11 06 06 00 00 00 00 00 00 00 00 00 00 00 00 00 0x0010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x0020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x0030: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x0040: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x0050: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x0060: 00 00 00 00 00 00 00 00 00 00 00 00 00 01 08 00 0x0070: 00 10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 patched driver: $ ethtool -m enp65s0f0np0 offset 0x90 length 8 Offset Values ------ ------ 0x0090: 00 00 01 a0 4d 65 6c 6c $ ethtool -m enp65s0f0np0 offset 0x90 length 12 Offset Values ------ ------ 0x0090: 00 00 01 a0 4d 65 6c 6c 61 6e 6f 78 $ ethtool -m enp65s0f0np0 Identifier : 0x11 (QSFP28) Extended identifier : 0x00 Extended identifier description : 1.5W max. Power consumption Extended identifier description : No CDR in TX, No CDR in RX Extended identifier description : High Power Class (> 3.5 W) not enabled Connector : 0x23 (No separable connector) Transceiver codes : 0x88 0x00 0x00 0x00 0x00 0x00 0x00 0x00 Transceiver type : 40G Ethernet: 40G Base-CR4 Transceiver type : 25G Ethernet: 25G Base-CR CA-N Encoding : 0x05 (64B/66B) BR, Nominal : 25500Mbps Rate identifier : 0x00 Length (SMF,km) : 0km Length (OM3 50um) : 0m Length (OM2 50um) : 0m Length (OM1 62.5um) : 0m Length (Copper or Active cable) : 1m Transmitter technology : 0xa0 (Copper cable unequalized) Attenuation at 2.5GHz : 4db Attenuation at 5.0GHz : 5db Attenuation at 7.0GHz : 7db Attenuation at 12.9GHz : 10db ........ .... Fixes: e9c9692c8a81 ("ice: Reimplement module reads used by ethtool") Signed-off-by: Petr Oros Reviewed-by: Jesse Brandeburg Tested-by: Jesse Brandeburg Signed-off-by: David S. Miller --- drivers/net/ethernet/intel/ice/ice_ethtool.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c index b360bd8f15998..f86e814354a31 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -4331,6 +4331,8 @@ ice_get_module_eeprom(struct net_device *netdev, * SFP modules only ever use page 0. */ if (page == 0 || !(data[0x2] & 0x4)) { + u32 copy_len; + /* If i2c bus is busy due to slow page change or * link management access, call can fail. This is normal. * So we retry this a few times. @@ -4354,8 +4356,8 @@ ice_get_module_eeprom(struct net_device *netdev, } /* Make sure we have enough room for the new block */ - if ((i + SFF_READ_BLOCK_SIZE) < ee->len) - memcpy(data + i, value, SFF_READ_BLOCK_SIZE); + copy_len = min_t(u32, SFF_READ_BLOCK_SIZE, ee->len - i); + memcpy(data + i, value, copy_len); } } return 0; From 3e04419cbe2d0bc10ffc20c006c6513ffa4b1ca3 Mon Sep 17 00:00:00 2001 From: Huanhuan Wang Date: Thu, 2 Mar 2023 10:58:28 +0100 Subject: [PATCH 691/765] nfp: fix incorrectly set csum flag for nfd3 path The csum flag of IPsec packet are set repeatedly. Therefore, the csum flag set of IPsec and non-IPsec packet need to be distinguished. As the ipv6 header does not have a csum field, so l3-csum flag is not required to be set for ipv6 case. L4-csum flag include the tcp csum flag and udp csum flag, we shouldn't set the udp and tcp csum flag at the same time for one packet, should set l4-csum flag according to the transport layer is tcp or udp. Fixes: 57f273adbcd4 ("nfp: add framework to support ipsec offloading") Signed-off-by: Huanhuan Wang Reviewed-by: Louis Peens Signed-off-by: Simon Horman Signed-off-by: David S. Miller --- drivers/net/ethernet/netronome/nfp/nfd3/dp.c | 7 +++--- .../net/ethernet/netronome/nfp/nfd3/ipsec.c | 25 +++++++++++++++++-- 2 files changed, 27 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/netronome/nfp/nfd3/dp.c b/drivers/net/ethernet/netronome/nfp/nfd3/dp.c index 59fb0583cc080..0cc026b0aefd4 100644 --- a/drivers/net/ethernet/netronome/nfp/nfd3/dp.c +++ b/drivers/net/ethernet/netronome/nfp/nfd3/dp.c @@ -324,14 +324,15 @@ netdev_tx_t nfp_nfd3_tx(struct sk_buff *skb, struct net_device *netdev) /* Do not reorder - tso may adjust pkt cnt, vlan may override fields */ nfp_nfd3_tx_tso(r_vec, txbuf, txd, skb, md_bytes); - nfp_nfd3_tx_csum(dp, r_vec, txbuf, txd, skb); + if (ipsec) + nfp_nfd3_ipsec_tx(txd, skb); + else + nfp_nfd3_tx_csum(dp, r_vec, txbuf, txd, skb); if (skb_vlan_tag_present(skb) && dp->ctrl & NFP_NET_CFG_CTRL_TXVLAN) { txd->flags |= NFD3_DESC_TX_VLAN; txd->vlan = cpu_to_le16(skb_vlan_tag_get(skb)); } - if (ipsec) - nfp_nfd3_ipsec_tx(txd, skb); /* Gather DMA */ if (nr_frags > 0) { __le64 second_half; diff --git a/drivers/net/ethernet/netronome/nfp/nfd3/ipsec.c b/drivers/net/ethernet/netronome/nfp/nfd3/ipsec.c index e90f8c975903b..51087693072c2 100644 --- a/drivers/net/ethernet/netronome/nfp/nfd3/ipsec.c +++ b/drivers/net/ethernet/netronome/nfp/nfd3/ipsec.c @@ -10,9 +10,30 @@ void nfp_nfd3_ipsec_tx(struct nfp_nfd3_tx_desc *txd, struct sk_buff *skb) { struct xfrm_state *x = xfrm_input_state(skb); + struct xfrm_offload *xo = xfrm_offload(skb); + struct iphdr *iph = ip_hdr(skb); + int l4_proto; if (x->xso.dev && (x->xso.dev->features & NETIF_F_HW_ESP_TX_CSUM)) { - txd->flags |= NFD3_DESC_TX_CSUM | NFD3_DESC_TX_IP4_CSUM | - NFD3_DESC_TX_TCP_CSUM | NFD3_DESC_TX_UDP_CSUM; + txd->flags |= NFD3_DESC_TX_CSUM; + + if (iph->version == 4) + txd->flags |= NFD3_DESC_TX_IP4_CSUM; + + if (x->props.mode == XFRM_MODE_TRANSPORT) + l4_proto = xo->proto; + else if (x->props.mode == XFRM_MODE_TUNNEL) + l4_proto = xo->inner_ipproto; + else + return; + + switch (l4_proto) { + case IPPROTO_UDP: + txd->flags |= NFD3_DESC_TX_UDP_CSUM; + return; + case IPPROTO_TCP: + txd->flags |= NFD3_DESC_TX_TCP_CSUM; + return; + } } } From 8b46168ca79c3aad04bc20605e523127266624d4 Mon Sep 17 00:00:00 2001 From: Huanhuan Wang Date: Thu, 2 Mar 2023 10:58:29 +0100 Subject: [PATCH 692/765] nfp: fix incorrectly set csum flag for nfdk path The csum flag of IPsec packet are set repeatedly. Therefore, the csum flag set of IPsec and non-IPsec packet need to be distinguished. As the ipv6 header does not have a csum field, so l3-csum flag is not required to be set for ipv6 case. Fixes: 436396f26d50 ("nfp: support IPsec offloading for NFP3800") Signed-off-by: Huanhuan Wang Reviewed-by: Louis Peens Signed-off-by: Simon Horman Signed-off-by: David S. Miller --- drivers/net/ethernet/netronome/nfp/nfdk/dp.c | 6 ++++-- drivers/net/ethernet/netronome/nfp/nfdk/ipsec.c | 8 ++++++-- 2 files changed, 10 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/netronome/nfp/nfdk/dp.c b/drivers/net/ethernet/netronome/nfp/nfdk/dp.c index d60c0e991a91c..33b6d74adb4b5 100644 --- a/drivers/net/ethernet/netronome/nfp/nfdk/dp.c +++ b/drivers/net/ethernet/netronome/nfp/nfdk/dp.c @@ -387,7 +387,8 @@ netdev_tx_t nfp_nfdk_tx(struct sk_buff *skb, struct net_device *netdev) if (!skb_is_gso(skb)) { real_len = skb->len; /* Metadata desc */ - metadata = nfp_nfdk_tx_csum(dp, r_vec, 1, skb, metadata); + if (!ipsec) + metadata = nfp_nfdk_tx_csum(dp, r_vec, 1, skb, metadata); txd->raw = cpu_to_le64(metadata); txd++; } else { @@ -395,7 +396,8 @@ netdev_tx_t nfp_nfdk_tx(struct sk_buff *skb, struct net_device *netdev) (txd + 1)->raw = nfp_nfdk_tx_tso(r_vec, txbuf, skb); real_len = txbuf->real_len; /* Metadata desc */ - metadata = nfp_nfdk_tx_csum(dp, r_vec, txbuf->pkt_cnt, skb, metadata); + if (!ipsec) + metadata = nfp_nfdk_tx_csum(dp, r_vec, txbuf->pkt_cnt, skb, metadata); txd->raw = cpu_to_le64(metadata); txd += 2; txbuf++; diff --git a/drivers/net/ethernet/netronome/nfp/nfdk/ipsec.c b/drivers/net/ethernet/netronome/nfp/nfdk/ipsec.c index 58d8f59eb8854..cec199f4c852f 100644 --- a/drivers/net/ethernet/netronome/nfp/nfdk/ipsec.c +++ b/drivers/net/ethernet/netronome/nfp/nfdk/ipsec.c @@ -9,9 +9,13 @@ u64 nfp_nfdk_ipsec_tx(u64 flags, struct sk_buff *skb) { struct xfrm_state *x = xfrm_input_state(skb); + struct iphdr *iph = ip_hdr(skb); - if (x->xso.dev && (x->xso.dev->features & NETIF_F_HW_ESP_TX_CSUM)) - flags |= NFDK_DESC_TX_L3_CSUM | NFDK_DESC_TX_L4_CSUM; + if (x->xso.dev && (x->xso.dev->features & NETIF_F_HW_ESP_TX_CSUM)) { + if (iph->version == 4) + flags |= NFDK_DESC_TX_L3_CSUM; + flags |= NFDK_DESC_TX_L4_CSUM; + } return flags; } From 1cf78d4c4144ffdbc2c9d505db482cb204bb480b Mon Sep 17 00:00:00 2001 From: Huanhuan Wang Date: Thu, 2 Mar 2023 10:58:30 +0100 Subject: [PATCH 693/765] nfp: fix esp-tx-csum-offload doesn't take effect When esp-tx-csum-offload is set to on, the protocol stack shouldn't calculate the IPsec offload packet's csum, but it does. Because the callback `.ndo_features_check` incorrectly masked NETIF_F_CSUM_MASK bit. Fixes: 57f273adbcd4 ("nfp: add framework to support ipsec offloading") Signed-off-by: Huanhuan Wang Signed-off-by: Simon Horman Signed-off-by: David S. Miller --- drivers/net/ethernet/netronome/nfp/nfp_net_common.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c index 81b7ca0ad222e..62f0bf91d1e1e 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c @@ -38,6 +38,7 @@ #include #include #include +#include #include "nfpcore/nfp_dev.h" #include "nfpcore/nfp_nsp.h" @@ -1897,6 +1898,9 @@ nfp_net_features_check(struct sk_buff *skb, struct net_device *dev, features &= ~NETIF_F_GSO_MASK; } + if (xfrm_offload(skb)) + return features; + /* VXLAN/GRE check */ switch (vlan_get_protocol(skb)) { case htons(ETH_P_IP): From 6210038aeaf49c395c2da57572246d93ec67f6d4 Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Fri, 3 Mar 2023 18:29:07 +0900 Subject: [PATCH 694/765] ata: ahci: Revert "ata: ahci: Add Tiger Lake UP{3,4} AHCI controller" Commit 104ff59af73a ("ata: ahci: Add Tiger Lake UP{3,4} AHCI controller") enabled low power mode for the Tiger Lake AHIC adapter in the author system but created regressions for others. Revert this patch for now until a better solution is found to make this adapter eco-friendly. Link: https://bugzilla.kernel.org/show_bug.cgi?id=217114 CC: stable@vger.kernel.org Signed-off-by: Damien Le Moal --- drivers/ata/ahci.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index 3bb9bb483fe37..14a1c0d14916f 100644 --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -421,7 +421,6 @@ static const struct pci_device_id ahci_pci_tbl[] = { { PCI_VDEVICE(INTEL, 0x34d3), board_ahci_low_power }, /* Ice Lake LP AHCI */ { PCI_VDEVICE(INTEL, 0x02d3), board_ahci_low_power }, /* Comet Lake PCH-U AHCI */ { PCI_VDEVICE(INTEL, 0x02d7), board_ahci_low_power }, /* Comet Lake PCH RAID */ - { PCI_VDEVICE(INTEL, 0xa0d3), board_ahci_low_power }, /* Tiger Lake UP{3,4} AHCI */ /* JMicron 360/1/3/5/6, match class to avoid IDE function */ { PCI_VENDOR_ID_JMICRON, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, From 5911d78fabbbbc02fb2b3f6907e0b3f6da120781 Mon Sep 17 00:00:00 2001 From: Jaroslav Kysela Date: Tue, 21 Feb 2023 11:21:57 +0100 Subject: [PATCH 695/765] ALSA: hda/realtek: Improve support for Dell Precision 3260 The headset jack works better with model=alc283-dac-wcaps. Without this option, the headset insertion (separate physical jack) may not be handled correctly (re-insertion is required). It seems that it follows the "Intel Reference Board" defaults. Reported-by: steven_wu2@dell.com Signed-off-by: Jaroslav Kysela Link: https://lore.kernel.org/r/20230221102157.515852-1-perex@perex.cz Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index e103bb3693c06..7afbb708b7055 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -9260,6 +9260,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x1028, 0x0a62, "Dell Precision 5560", ALC289_FIXUP_DUAL_SPK), SND_PCI_QUIRK(0x1028, 0x0a9d, "Dell Latitude 5430", ALC269_FIXUP_DELL4_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x1028, 0x0a9e, "Dell Latitude 5430", ALC269_FIXUP_DELL4_MIC_NO_PRESENCE), + SND_PCI_QUIRK(0x1028, 0x0ac9, "Dell Precision 3260", ALC295_FIXUP_CHROME_BOOK), SND_PCI_QUIRK(0x1028, 0x0b19, "Dell XPS 15 9520", ALC289_FIXUP_DUAL_SPK), SND_PCI_QUIRK(0x1028, 0x0b1a, "Dell Precision 5570", ALC289_FIXUP_DUAL_SPK), SND_PCI_QUIRK(0x1028, 0x0b37, "Dell Inspiron 16 Plus 7620 2-in-1", ALC295_FIXUP_DELL_INSPIRON_TOP_SPEAKERS), From ea24b9953bcd3889f77a66e7f1d7e86e995dd9c3 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=C5=81ukasz=20Stelmach?= Date: Thu, 23 Feb 2023 08:47:48 +0100 Subject: [PATCH 696/765] ALSA: hda/realtek: Add quirk for HP EliteDesk 800 G6 Tower PC MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit HP EliteDesk 800 G6 Tower PC (103c:870c) requires a quirk for enabling headset-mic. Signed-off-by: Łukasz Stelmach Cc: Link: https://bugzilla.kernel.org/show_bug.cgi?id=217008 Link: https://lore.kernel.org/r/20230223074749.1026060-1-l.stelmach@samsung.com Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 7afbb708b7055..3c629f4ae0807 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -11618,6 +11618,7 @@ static const struct snd_pci_quirk alc662_fixup_tbl[] = { SND_PCI_QUIRK(0x1028, 0x0698, "Dell", ALC668_FIXUP_DELL_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x1028, 0x069f, "Dell", ALC668_FIXUP_DELL_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x103c, 0x1632, "HP RP5800", ALC662_FIXUP_HP_RP5800), + SND_PCI_QUIRK(0x103c, 0x870c, "HP", ALC897_FIXUP_HP_HSMIC_VERB), SND_PCI_QUIRK(0x103c, 0x8719, "HP", ALC897_FIXUP_HP_HSMIC_VERB), SND_PCI_QUIRK(0x103c, 0x873e, "HP", ALC671_FIXUP_HP_HEADSET_MIC2), SND_PCI_QUIRK(0x103c, 0x877e, "HP 288 Pro G6", ALC671_FIXUP_HP_HEADSET_MIC2), From 951606a14a8901e3551fe4d8d3cedd73fe954ce1 Mon Sep 17 00:00:00 2001 From: Dmitry Fomin Date: Sat, 25 Feb 2023 21:43:21 +0300 Subject: [PATCH 697/765] ALSA: ice1712: Do not left ice->gpio_mutex locked in aureon_add_controls() If snd_ctl_add() fails in aureon_add_controls(), it immediately returns and leaves ice->gpio_mutex locked. ice->gpio_mutex locks in snd_ice1712_save_gpio_status and unlocks in snd_ice1712_restore_gpio_status(ice). It seems that the mutex is required only for aureon_cs8415_get(), so snd_ice1712_restore_gpio_status(ice) can be placed just after that. Compile tested only. Found by Linux Verification Center (linuxtesting.org) with SVACE. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Dmitry Fomin Cc: Link: https://lore.kernel.org/r/20230225184322.6286-1-fomindmitriyfoma@mail.ru Signed-off-by: Takashi Iwai --- sound/pci/ice1712/aureon.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/pci/ice1712/aureon.c b/sound/pci/ice1712/aureon.c index 9a30f6d35d135..40a0e00950301 100644 --- a/sound/pci/ice1712/aureon.c +++ b/sound/pci/ice1712/aureon.c @@ -1892,6 +1892,7 @@ static int aureon_add_controls(struct snd_ice1712 *ice) unsigned char id; snd_ice1712_save_gpio_status(ice); id = aureon_cs8415_get(ice, CS8415_ID); + snd_ice1712_restore_gpio_status(ice); if (id != 0x41) dev_info(ice->card->dev, "No CS8415 chip. Skipping CS8415 controls.\n"); @@ -1909,7 +1910,6 @@ static int aureon_add_controls(struct snd_ice1712 *ice) kctl->id.device = ice->pcm->device; } } - snd_ice1712_restore_gpio_status(ice); } return 0; From a8e98f3448e1a4b6848f213cf51720e29dcc774b Mon Sep 17 00:00:00 2001 From: Dmitry Fomin Date: Sat, 25 Feb 2023 21:43:22 +0300 Subject: [PATCH 698/765] ALSA: ice1712: Delete unreachable code in aureon_add_controls() If the check (id != 0x41) fails, then id == 0x41 and the other check in 'else' branch also fails: id & 0x0F = 0b01000001 & 0b00001111 = 0b00000001. Found by Linux Verification Center (linuxtesting.org) with SVACE. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Dmitry Fomin Link: https://lore.kernel.org/r/20230225184322.6286-2-fomindmitriyfoma@mail.ru Signed-off-by: Takashi Iwai --- sound/pci/ice1712/aureon.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/sound/pci/ice1712/aureon.c b/sound/pci/ice1712/aureon.c index 40a0e00950301..24b9782340001 100644 --- a/sound/pci/ice1712/aureon.c +++ b/sound/pci/ice1712/aureon.c @@ -1896,10 +1896,6 @@ static int aureon_add_controls(struct snd_ice1712 *ice) if (id != 0x41) dev_info(ice->card->dev, "No CS8415 chip. Skipping CS8415 controls.\n"); - else if ((id & 0x0F) != 0x01) - dev_info(ice->card->dev, - "Detected unsupported CS8415 rev. (%c)\n", - (char)((id & 0x0F) + 'A' - 1)); else { for (i = 0; i < ARRAY_SIZE(cs8415_controls); i++) { struct snd_kcontrol *kctl; From db50f7a3983f0154e730f1147ef729e0c5c2f90c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Szalecki?= Date: Wed, 1 Mar 2023 02:23:56 +0100 Subject: [PATCH 699/765] HID: logitech-hidpp: Add support for Logitech MX Master 3S mouse MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add signature for the Logitech MX Master 3S mouse over Bluetooth. Signed-off-by: Rafał Szalecki Reviewed-by: Bastien Nocera Signed-off-by: Jiri Kosina --- drivers/hid/hid-logitech-hidpp.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/hid/hid-logitech-hidpp.c b/drivers/hid/hid-logitech-hidpp.c index 25dcda76d6c7b..5fc88a0632978 100644 --- a/drivers/hid/hid-logitech-hidpp.c +++ b/drivers/hid/hid-logitech-hidpp.c @@ -4399,6 +4399,8 @@ static const struct hid_device_id hidpp_devices[] = { HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_LOGITECH, 0xb02a) }, { /* MX Master 3 mouse over Bluetooth */ HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_LOGITECH, 0xb023) }, + { /* MX Master 3S mouse over Bluetooth */ + HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_LOGITECH, 0xb034) }, {} }; From 8ae2f2b0a28416ed2f6d8478ac8b9f7862f36785 Mon Sep 17 00:00:00 2001 From: Reka Norman Date: Mon, 27 Feb 2023 13:49:38 +1100 Subject: [PATCH 700/765] HID: intel-ish-hid: ipc: Fix potential use-after-free in work function When a reset notify IPC message is received, the ISR schedules a work function and passes the ISHTP device to it via a global pointer ishtp_dev. If ish_probe() fails, the devm-managed device resources including ishtp_dev are freed, but the work is not cancelled, causing a use-after-free when the work function tries to access ishtp_dev. Use devm_work_autocancel() instead, so that the work is automatically cancelled if probe fails. Signed-off-by: Reka Norman Acked-by: Srinivas Pandruvada Signed-off-by: Jiri Kosina --- drivers/hid/intel-ish-hid/ipc/ipc.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/hid/intel-ish-hid/ipc/ipc.c b/drivers/hid/intel-ish-hid/ipc/ipc.c index 15e14239af829..a49c6affd7c4c 100644 --- a/drivers/hid/intel-ish-hid/ipc/ipc.c +++ b/drivers/hid/intel-ish-hid/ipc/ipc.c @@ -5,6 +5,7 @@ * Copyright (c) 2014-2016, Intel Corporation. */ +#include #include #include #include @@ -621,7 +622,6 @@ static void recv_ipc(struct ishtp_device *dev, uint32_t doorbell_val) case MNG_RESET_NOTIFY: if (!ishtp_dev) { ishtp_dev = dev; - INIT_WORK(&fw_reset_work, fw_reset_work_fn); } schedule_work(&fw_reset_work); break; @@ -940,6 +940,7 @@ struct ishtp_device *ish_dev_init(struct pci_dev *pdev) { struct ishtp_device *dev; int i; + int ret; dev = devm_kzalloc(&pdev->dev, sizeof(struct ishtp_device) + sizeof(struct ish_hw), @@ -975,6 +976,12 @@ struct ishtp_device *ish_dev_init(struct pci_dev *pdev) list_add_tail(&tx_buf->link, &dev->wr_free_list); } + ret = devm_work_autocancel(&pdev->dev, &fw_reset_work, fw_reset_work_fn); + if (ret) { + dev_err(dev->devc, "Failed to initialise FW reset work\n"); + return NULL; + } + dev->ops = &ish_hw_ops; dev->devc = &pdev->dev; dev->mtu = IPC_PAYLOAD_SIZE - sizeof(struct ishtp_msg_hdr); From d900f3d20cc3169ce42ec72acc850e662a4d4db2 Mon Sep 17 00:00:00 2001 From: Liu Jian Date: Fri, 3 Mar 2023 16:09:46 +0800 Subject: [PATCH 701/765] bpf, sockmap: Fix an infinite loop error when len is 0 in tcp_bpf_recvmsg_parser() When the buffer length of the recvmsg system call is 0, we got the flollowing soft lockup problem: watchdog: BUG: soft lockup - CPU#3 stuck for 27s! [a.out:6149] CPU: 3 PID: 6149 Comm: a.out Kdump: loaded Not tainted 6.2.0+ #30 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.15.0-1 04/01/2014 RIP: 0010:remove_wait_queue+0xb/0xc0 Code: 5e 41 5f c3 cc cc cc cc 0f 1f 80 00 00 00 00 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 f3 0f 1e fa 0f 1f 44 00 00 41 57 <41> 56 41 55 41 54 55 48 89 fd 53 48 89 f3 4c 8d 6b 18 4c 8d 73 20 RSP: 0018:ffff88811b5978b8 EFLAGS: 00000246 RAX: 0000000000000000 RBX: ffff88811a7d3780 RCX: ffffffffb7a4d768 RDX: dffffc0000000000 RSI: ffff88811b597908 RDI: ffff888115408040 RBP: 1ffff110236b2f1b R08: 0000000000000000 R09: ffff88811a7d37e7 R10: ffffed10234fa6fc R11: 0000000000000001 R12: ffff88811179b800 R13: 0000000000000001 R14: ffff88811a7d38a8 R15: ffff88811a7d37e0 FS: 00007f6fb5398740(0000) GS:ffff888237180000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000020000000 CR3: 000000010b6ba002 CR4: 0000000000370ee0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: tcp_msg_wait_data+0x279/0x2f0 tcp_bpf_recvmsg_parser+0x3c6/0x490 inet_recvmsg+0x280/0x290 sock_recvmsg+0xfc/0x120 ____sys_recvmsg+0x160/0x3d0 ___sys_recvmsg+0xf0/0x180 __sys_recvmsg+0xea/0x1a0 do_syscall_64+0x3f/0x90 entry_SYSCALL_64_after_hwframe+0x72/0xdc The logic in tcp_bpf_recvmsg_parser is as follows: msg_bytes_ready: copied = sk_msg_recvmsg(sk, psock, msg, len, flags); if (!copied) { wait data; goto msg_bytes_ready; } In this case, "copied" always is 0, the infinite loop occurs. According to the Linux system call man page, 0 should be returned in this case. Therefore, in tcp_bpf_recvmsg_parser(), if the length is 0, directly return. Also modify several other functions with the same problem. Fixes: 1f5be6b3b063 ("udp: Implement udp_bpf_recvmsg() for sockmap") Fixes: 9825d866ce0d ("af_unix: Implement unix_dgram_bpf_recvmsg()") Fixes: c5d2177a72a1 ("bpf, sockmap: Fix race in ingress receive verdict with redirect to self") Fixes: 604326b41a6f ("bpf, sockmap: convert to generic sk_msg interface") Signed-off-by: Liu Jian Signed-off-by: Daniel Borkmann Acked-by: John Fastabend Cc: Jakub Sitnicki Link: https://lore.kernel.org/bpf/20230303080946.1146638-1-liujian56@huawei.com --- net/ipv4/tcp_bpf.c | 6 ++++++ net/ipv4/udp_bpf.c | 3 +++ net/unix/unix_bpf.c | 3 +++ 3 files changed, 12 insertions(+) diff --git a/net/ipv4/tcp_bpf.c b/net/ipv4/tcp_bpf.c index cf26d65ca3893..ebf9175119370 100644 --- a/net/ipv4/tcp_bpf.c +++ b/net/ipv4/tcp_bpf.c @@ -186,6 +186,9 @@ static int tcp_bpf_recvmsg_parser(struct sock *sk, if (unlikely(flags & MSG_ERRQUEUE)) return inet_recv_error(sk, msg, len, addr_len); + if (!len) + return 0; + psock = sk_psock_get(sk); if (unlikely(!psock)) return tcp_recvmsg(sk, msg, len, flags, addr_len); @@ -244,6 +247,9 @@ static int tcp_bpf_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, if (unlikely(flags & MSG_ERRQUEUE)) return inet_recv_error(sk, msg, len, addr_len); + if (!len) + return 0; + psock = sk_psock_get(sk); if (unlikely(!psock)) return tcp_recvmsg(sk, msg, len, flags, addr_len); diff --git a/net/ipv4/udp_bpf.c b/net/ipv4/udp_bpf.c index e5dc91d0e0793..0735d820e413f 100644 --- a/net/ipv4/udp_bpf.c +++ b/net/ipv4/udp_bpf.c @@ -68,6 +68,9 @@ static int udp_bpf_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, if (unlikely(flags & MSG_ERRQUEUE)) return inet_recv_error(sk, msg, len, addr_len); + if (!len) + return 0; + psock = sk_psock_get(sk); if (unlikely(!psock)) return sk_udp_recvmsg(sk, msg, len, flags, addr_len); diff --git a/net/unix/unix_bpf.c b/net/unix/unix_bpf.c index e9bf155139612..2f9d8271c6ec7 100644 --- a/net/unix/unix_bpf.c +++ b/net/unix/unix_bpf.c @@ -54,6 +54,9 @@ static int unix_bpf_recvmsg(struct sock *sk, struct msghdr *msg, struct sk_psock *psock; int copied; + if (!len) + return 0; + psock = sk_psock_get(sk); if (unlikely(!psock)) return __unix_recvmsg(sk, msg, len, flags); From a76d19e6acb1ef16561b0682cd1a566463fa3d98 Mon Sep 17 00:00:00 2001 From: Benjamin Gray Date: Tue, 28 Feb 2023 10:33:17 +1100 Subject: [PATCH 702/765] i2c: Disable I2C_APPLE when I2C_PASEMI is a builtin The ppc64le_allmodconfig sets I2C_PASEMI=y and leaves COMPILE_TEST to default to y and I2C_APPLE to default to m, running into a known incompatible configuration that breaks the build [1]. Specifically, a common dependency (i2c-pasemi-core.o in this case) cannot be used by both builtin and module consumers. Disable I2C_APPLE when I2C_PASEMI is a builtin to prevent this. [1]: https://lore.kernel.org/all/202112061809.XT99aPrf-lkp@intel.com Suggested-by: Arnd Bergmann Signed-off-by: Benjamin Gray Acked-by: Arnd Bergmann Acked-by: Sven Peter Signed-off-by: Wolfram Sang --- drivers/i2c/busses/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig index 9b8e84f20604f..25eb4e8fd22fd 100644 --- a/drivers/i2c/busses/Kconfig +++ b/drivers/i2c/busses/Kconfig @@ -914,6 +914,7 @@ config I2C_PASEMI config I2C_APPLE tristate "Apple SMBus platform driver" + depends on !I2C_PASEMI depends on ARCH_APPLE || COMPILE_TEST default ARCH_APPLE help From 1d092308ce223bb1403475737b8fb847e9e8704c Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Fri, 17 Feb 2023 23:13:30 +0100 Subject: [PATCH 703/765] i2c: gxp: remove "empty" switch statement There used to be error messages which had to go. Now, it only consists of 'break's, so it can go. Signed-off-by: Wolfram Sang --- drivers/i2c/busses/i2c-gxp.c | 13 +------------ 1 file changed, 1 insertion(+), 12 deletions(-) diff --git a/drivers/i2c/busses/i2c-gxp.c b/drivers/i2c/busses/i2c-gxp.c index da4c8e5a80390..352dedf6292f1 100644 --- a/drivers/i2c/busses/i2c-gxp.c +++ b/drivers/i2c/busses/i2c-gxp.c @@ -126,19 +126,8 @@ static int gxp_i2c_master_xfer(struct i2c_adapter *adapter, time_left = wait_for_completion_timeout(&drvdata->completion, adapter->timeout); ret = num - drvdata->msgs_remaining; - if (time_left == 0) { - switch (drvdata->state) { - case GXP_I2C_WDATA_PHASE: - break; - case GXP_I2C_RDATA_PHASE: - break; - case GXP_I2C_ADDR_PHASE: - break; - default: - break; - } + if (time_left == 0) return -ETIMEDOUT; - } if (drvdata->state == GXP_I2C_ADDR_NACK || drvdata->state == GXP_I2C_DATA_NACK) From 4b3dfb0ed6336dea4a763ce9fa30a42763eb3800 Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Mon, 20 Feb 2023 15:40:59 +0100 Subject: [PATCH 704/765] i2c: gxp: return proper error on address NACK According to Documentation/i2c/fault-codes.rst, NACK after sending an address should be -ENXIO. Signed-off-by: Wolfram Sang --- drivers/i2c/busses/i2c-gxp.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/i2c/busses/i2c-gxp.c b/drivers/i2c/busses/i2c-gxp.c index 352dedf6292f1..fda7809e72baa 100644 --- a/drivers/i2c/busses/i2c-gxp.c +++ b/drivers/i2c/busses/i2c-gxp.c @@ -129,8 +129,10 @@ static int gxp_i2c_master_xfer(struct i2c_adapter *adapter, if (time_left == 0) return -ETIMEDOUT; - if (drvdata->state == GXP_I2C_ADDR_NACK || - drvdata->state == GXP_I2C_DATA_NACK) + if (drvdata->state == GXP_I2C_ADDR_NACK) + return -ENXIO; + + if (drvdata->state == GXP_I2C_DATA_NACK) return -EIO; return ret; From 65609d3206f784489eb1ebd6fce64b84a42cc63c Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 27 Feb 2023 13:06:33 +0300 Subject: [PATCH 705/765] i2c: gxp: fix an error code in probe This is passing IS_ERR() instead of PTR_ERR() so instead of an error code it prints and returns the number 1. Fixes: 4a55ed6f89f5 ("i2c: Add GXP SoC I2C Controller") Signed-off-by: Dan Carpenter Reviewed-by: Nick Hawkins Signed-off-by: Wolfram Sang --- drivers/i2c/busses/i2c-gxp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/i2c/busses/i2c-gxp.c b/drivers/i2c/busses/i2c-gxp.c index fda7809e72baa..d4b55d989a268 100644 --- a/drivers/i2c/busses/i2c-gxp.c +++ b/drivers/i2c/busses/i2c-gxp.c @@ -516,7 +516,7 @@ static int gxp_i2c_probe(struct platform_device *pdev) i2cg_map = syscon_regmap_lookup_by_phandle(pdev->dev.of_node, "hpe,sysreg"); if (IS_ERR(i2cg_map)) { - return dev_err_probe(&pdev->dev, IS_ERR(i2cg_map), + return dev_err_probe(&pdev->dev, PTR_ERR(i2cg_map), "failed to map i2cg_handle\n"); } From e778361555713826481be6234fd1aa030bdb035e Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 2 Mar 2023 15:49:44 -0800 Subject: [PATCH 706/765] umh: simplify the capability pointer logic The usermodehelper code uses two fake pointers for the two capability cases: CAP_BSET for reading and writing 'usermodehelper_bset', and CAP_PI to read and write 'usermodehelper_inheritable'. This seems to be a completely unnecessary indirection, since we could instead just use the pointers themselves, and never have to do any "if this then that" kind of logic. So just get rid of the fake pointer values, and use the real pointer values instead. Reviewed-by: Luis Chamberlain Cc: Eric Biederman Cc: Peter Zijlstra Cc: Christoph Hellwig Cc: Kees Cook Cc: Iurii Zaikin Signed-off-by: Linus Torvalds --- kernel/umh.c | 18 +++++------------- 1 file changed, 5 insertions(+), 13 deletions(-) diff --git a/kernel/umh.c b/kernel/umh.c index 2a47082773352..60aa9e764a38f 100644 --- a/kernel/umh.c +++ b/kernel/umh.c @@ -32,9 +32,6 @@ #include -#define CAP_BSET (void *)1 -#define CAP_PI (void *)2 - static kernel_cap_t usermodehelper_bset = CAP_FULL_SET; static kernel_cap_t usermodehelper_inheritable = CAP_FULL_SET; static DEFINE_SPINLOCK(umh_sysctl_lock); @@ -512,16 +509,11 @@ static int proc_cap_handler(struct ctl_table *table, int write, /* * convert from the global kernel_cap_t to the ulong array to print to * userspace if this is a read. + * + * Legacy format: capabilities are exposed as two 32-bit values */ + cap = table->data; spin_lock(&umh_sysctl_lock); - if (table->data == CAP_BSET) - cap = &usermodehelper_bset; - else if (table->data == CAP_PI) - cap = &usermodehelper_inheritable; - else - BUG(); - - /* Legacy format: capabilities are exposed as two 32-bit values */ cap_array[0] = (u32) cap->val; cap_array[1] = cap->val >> 32; spin_unlock(&umh_sysctl_lock); @@ -555,14 +547,14 @@ static int proc_cap_handler(struct ctl_table *table, int write, struct ctl_table usermodehelper_table[] = { { .procname = "bset", - .data = CAP_BSET, + .data = &usermodehelper_bset, .maxlen = 2 * sizeof(unsigned long), .mode = 0600, .proc_handler = proc_cap_handler, }, { .procname = "inheritable", - .data = CAP_PI, + .data = &usermodehelper_inheritable, .maxlen = 2 * sizeof(unsigned long), .mode = 0600, .proc_handler = proc_cap_handler, From e77d587a2c04e82c6a0dffa4a32c874a4029385d Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sat, 4 Mar 2023 14:03:27 -0800 Subject: [PATCH 707/765] mm: avoid gcc complaint about pointer casting MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The migration code ends up temporarily stashing information of the wrong type in unused fields of the newly allocated destination folio. That all works fine, but gcc does complain about the pointer type mis-use: mm/migrate.c: In function ‘__migrate_folio_extract’: mm/migrate.c:1050:20: note: randstruct: casting between randomized structure pointer types (ssa): ‘struct anon_vma’ and ‘struct address_space’ 1050 | *anon_vmap = (void *)dst->mapping; | ~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~ and gcc is actually right to complain since it really doesn't understand that this is a very temporary special case where this is ok. This could be fixed in different ways by just obfuscating the assignment sufficiently that gcc doesn't see what is going on, but the truly "proper C" way to do this is by explicitly using a union. Using unions for type conversions like this is normally hugely ugly and syntactically nasty, but this really is one of the few cases where we want to make it clear that we're not doing type conversion, we're really re-using the value bit-for-bit just using another type. IOW, this should not become a common pattern, but in this one case using that odd union is probably the best way to document to the compiler what is conceptually going on here. [ Side note: there are valid cases where we convert pointers to other pointer types, notably the whole "folio vs page" situation, where the types actually have fundamental commonalities. The fact that the gcc note is limited to just randomized structures means that we don't see equivalent warnings for those cases, but it migth also mean that we miss other cases where we do play these kinds of dodgy games, and this kind of explicit conversion might be a good idea. ] I verified that at least for an allmodconfig build on x86-64, this generates the exact same code, apart from line numbers and assembler comment changes. Fixes: 64c8902ed441 ("migrate_pages: split unmap_and_move() to _unmap() and _move()") Cc: Huang, Ying Cc: Andrew Morton Signed-off-by: Linus Torvalds --- mm/migrate.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/mm/migrate.c b/mm/migrate.c index 37865f85df6d4..98f1c11197a8c 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1035,11 +1035,16 @@ static int move_to_new_folio(struct folio *dst, struct folio *src, * destination folio. This is safe because nobody is using them * except us. */ +union migration_ptr { + struct anon_vma *anon_vma; + struct address_space *mapping; +}; static void __migrate_folio_record(struct folio *dst, unsigned long page_was_mapped, struct anon_vma *anon_vma) { - dst->mapping = (void *)anon_vma; + union migration_ptr ptr = { .anon_vma = anon_vma }; + dst->mapping = ptr.mapping; dst->private = (void *)page_was_mapped; } @@ -1047,7 +1052,8 @@ static void __migrate_folio_extract(struct folio *dst, int *page_was_mappedp, struct anon_vma **anon_vmap) { - *anon_vmap = (void *)dst->mapping; + union migration_ptr ptr = { .mapping = dst->mapping }; + *anon_vmap = ptr.anon_vma; *page_was_mappedp = (unsigned long)dst->private; dst->mapping = NULL; dst->private = NULL; From 3304f18bfcf588df0fe3b703b6a6c1129d80bdc7 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 4 Mar 2023 20:27:29 -0500 Subject: [PATCH 708/765] Adding VFS co-maintainer Acked-by: Christian Brauner Signed-off-by: Al Viro --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index 135d93368d36e..bcc2b508c8e72 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -8043,6 +8043,7 @@ F: include/trace/events/fs_dax.h FILESYSTEMS (VFS and infrastructure) M: Alexander Viro +M: Christian Brauner L: linux-fsdevel@vger.kernel.org S: Maintained F: fs/* From 95207db8166ab95c42a03fdc5e3abd212c9987dc Mon Sep 17 00:00:00 2001 From: Masahiro Yamada Date: Mon, 17 Oct 2022 03:23:49 +0900 Subject: [PATCH 709/765] Remove Intel compiler support include/linux/compiler-intel.h had no update in the past 3 years. We often forget about the third C compiler to build the kernel. For example, commit a0a12c3ed057 ("asm goto: eradicate CC_HAS_ASM_GOTO") only mentioned GCC and Clang. init/Kconfig defines CC_IS_GCC and CC_IS_CLANG but not CC_IS_ICC, and nobody has reported any issue. I guess the Intel Compiler support is broken, and nobody is caring about it. Harald Arnesen pointed out ICC (classic Intel C/C++ compiler) is deprecated: $ icc -v icc: remark #10441: The Intel(R) C++ Compiler Classic (ICC) is deprecated and will be removed from product release in the second half of 2023. The Intel(R) oneAPI DPC++/C++ Compiler (ICX) is the recommended compiler moving forward. Please transition to use this compiler. Use '-diag-disable=10441' to disable this message. icc version 2021.7.0 (gcc version 12.1.0 compatibility) Arnd Bergmann provided a link to the article, "Intel C/C++ compilers complete adoption of LLVM". lib/zstd/common/compiler.h and lib/zstd/compress/zstd_fast.c were kept untouched for better sync with https://github.com/facebook/zstd Link: https://www.intel.com/content/www/us/en/developer/articles/technical/adoption-of-llvm-complete-icx.html Signed-off-by: Masahiro Yamada Acked-by: Arnd Bergmann Reviewed-by: Nick Desaulniers Reviewed-by: Nathan Chancellor Reviewed-by: Miguel Ojeda Signed-off-by: Linus Torvalds --- arch/ia64/include/uapi/asm/cmpxchg.h | 6 +- arch/ia64/include/uapi/asm/intel_intrin.h | 162 ---------------------- arch/ia64/include/uapi/asm/intrinsics.h | 6 +- include/acpi/platform/acenv.h | 5 +- include/acpi/platform/acenvex.h | 2 +- include/acpi/platform/acintel.h | 55 -------- include/linux/compiler-intel.h | 34 ----- include/linux/compiler_attributes.h | 14 +- include/linux/compiler_types.h | 2 - scripts/cc-version.sh | 2 - scripts/min-tool-version.sh | 4 - 11 files changed, 5 insertions(+), 287 deletions(-) delete mode 100644 arch/ia64/include/uapi/asm/intel_intrin.h delete mode 100644 include/acpi/platform/acintel.h delete mode 100644 include/linux/compiler-intel.h diff --git a/arch/ia64/include/uapi/asm/cmpxchg.h b/arch/ia64/include/uapi/asm/cmpxchg.h index ca2e026853438..259ae57570bfe 100644 --- a/arch/ia64/include/uapi/asm/cmpxchg.h +++ b/arch/ia64/include/uapi/asm/cmpxchg.h @@ -15,11 +15,7 @@ #include /* include compiler specific intrinsics */ #include -#ifdef __INTEL_COMPILER -# include -#else -# include -#endif +#include /* * This function doesn't exist, so you'll get a linker error if diff --git a/arch/ia64/include/uapi/asm/intel_intrin.h b/arch/ia64/include/uapi/asm/intel_intrin.h deleted file mode 100644 index dc1884dc54b5e..0000000000000 --- a/arch/ia64/include/uapi/asm/intel_intrin.h +++ /dev/null @@ -1,162 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ -#ifndef _ASM_IA64_INTEL_INTRIN_H -#define _ASM_IA64_INTEL_INTRIN_H -/* - * Intel Compiler Intrinsics - * - * Copyright (C) 2002,2003 Jun Nakajima - * Copyright (C) 2002,2003 Suresh Siddha - * Copyright (C) 2005,2006 Hongjiu Lu - * - */ -#include - -#define ia64_barrier() __memory_barrier() - -#define ia64_stop() /* Nothing: As of now stop bit is generated for each - * intrinsic - */ - -#define ia64_getreg __getReg -#define ia64_setreg __setReg - -#define ia64_hint __hint -#define ia64_hint_pause __hint_pause - -#define ia64_mux1_brcst _m64_mux1_brcst -#define ia64_mux1_mix _m64_mux1_mix -#define ia64_mux1_shuf _m64_mux1_shuf -#define ia64_mux1_alt _m64_mux1_alt -#define ia64_mux1_rev _m64_mux1_rev - -#define ia64_mux1(x,v) _m_to_int64(_m64_mux1(_m_from_int64(x), (v))) -#define ia64_popcnt _m64_popcnt -#define ia64_getf_exp __getf_exp -#define ia64_shrp _m64_shrp - -#define ia64_tpa __tpa -#define ia64_invala __invala -#define ia64_invala_gr __invala_gr -#define ia64_invala_fr __invala_fr -#define ia64_nop __nop -#define ia64_sum __sum -#define ia64_ssm __ssm -#define ia64_rum __rum -#define ia64_rsm __rsm -#define ia64_fc __fc - -#define ia64_ldfs __ldfs -#define ia64_ldfd __ldfd -#define ia64_ldfe __ldfe -#define ia64_ldf8 __ldf8 -#define ia64_ldf_fill __ldf_fill - -#define ia64_stfs __stfs -#define ia64_stfd __stfd -#define ia64_stfe __stfe -#define ia64_stf8 __stf8 -#define ia64_stf_spill __stf_spill - -#define ia64_mf __mf -#define ia64_mfa __mfa - -#define ia64_fetchadd4_acq __fetchadd4_acq -#define ia64_fetchadd4_rel __fetchadd4_rel -#define ia64_fetchadd8_acq __fetchadd8_acq -#define ia64_fetchadd8_rel __fetchadd8_rel - -#define ia64_xchg1 _InterlockedExchange8 -#define ia64_xchg2 _InterlockedExchange16 -#define ia64_xchg4 _InterlockedExchange -#define ia64_xchg8 _InterlockedExchange64 - -#define ia64_cmpxchg1_rel _InterlockedCompareExchange8_rel -#define ia64_cmpxchg1_acq _InterlockedCompareExchange8_acq -#define ia64_cmpxchg2_rel _InterlockedCompareExchange16_rel -#define ia64_cmpxchg2_acq _InterlockedCompareExchange16_acq -#define ia64_cmpxchg4_rel _InterlockedCompareExchange_rel -#define ia64_cmpxchg4_acq _InterlockedCompareExchange_acq -#define ia64_cmpxchg8_rel _InterlockedCompareExchange64_rel -#define ia64_cmpxchg8_acq _InterlockedCompareExchange64_acq - -#define __ia64_set_dbr(index, val) \ - __setIndReg(_IA64_REG_INDR_DBR, index, val) -#define ia64_set_ibr(index, val) \ - __setIndReg(_IA64_REG_INDR_IBR, index, val) -#define ia64_set_pkr(index, val) \ - __setIndReg(_IA64_REG_INDR_PKR, index, val) -#define ia64_set_pmc(index, val) \ - __setIndReg(_IA64_REG_INDR_PMC, index, val) -#define ia64_set_pmd(index, val) \ - __setIndReg(_IA64_REG_INDR_PMD, index, val) -#define ia64_set_rr(index, val) \ - __setIndReg(_IA64_REG_INDR_RR, index, val) - -#define ia64_get_cpuid(index) \ - __getIndReg(_IA64_REG_INDR_CPUID, index) -#define __ia64_get_dbr(index) __getIndReg(_IA64_REG_INDR_DBR, index) -#define ia64_get_ibr(index) __getIndReg(_IA64_REG_INDR_IBR, index) -#define ia64_get_pkr(index) __getIndReg(_IA64_REG_INDR_PKR, index) -#define ia64_get_pmc(index) __getIndReg(_IA64_REG_INDR_PMC, index) -#define ia64_get_pmd(index) __getIndReg(_IA64_REG_INDR_PMD, index) -#define ia64_get_rr(index) __getIndReg(_IA64_REG_INDR_RR, index) - -#define ia64_srlz_d __dsrlz -#define ia64_srlz_i __isrlz - -#define ia64_dv_serialize_data() -#define ia64_dv_serialize_instruction() - -#define ia64_st1_rel __st1_rel -#define ia64_st2_rel __st2_rel -#define ia64_st4_rel __st4_rel -#define ia64_st8_rel __st8_rel - -/* FIXME: need st4.rel.nta intrinsic */ -#define ia64_st4_rel_nta __st4_rel - -#define ia64_ld1_acq __ld1_acq -#define ia64_ld2_acq __ld2_acq -#define ia64_ld4_acq __ld4_acq -#define ia64_ld8_acq __ld8_acq - -#define ia64_sync_i __synci -#define ia64_thash __thash -#define ia64_ttag __ttag -#define ia64_itcd __itcd -#define ia64_itci __itci -#define ia64_itrd __itrd -#define ia64_itri __itri -#define ia64_ptce __ptce -#define ia64_ptcl __ptcl -#define ia64_ptcg __ptcg -#define ia64_ptcga __ptcga -#define ia64_ptri __ptri -#define ia64_ptrd __ptrd -#define ia64_dep_mi _m64_dep_mi - -/* Values for lfhint in __lfetch and __lfetch_fault */ - -#define ia64_lfhint_none __lfhint_none -#define ia64_lfhint_nt1 __lfhint_nt1 -#define ia64_lfhint_nt2 __lfhint_nt2 -#define ia64_lfhint_nta __lfhint_nta - -#define ia64_lfetch __lfetch -#define ia64_lfetch_excl __lfetch_excl -#define ia64_lfetch_fault __lfetch_fault -#define ia64_lfetch_fault_excl __lfetch_fault_excl - -#define ia64_intrin_local_irq_restore(x) \ -do { \ - if ((x) != 0) { \ - ia64_ssm(IA64_PSR_I); \ - ia64_srlz_d(); \ - } else { \ - ia64_rsm(IA64_PSR_I); \ - } \ -} while (0) - -#define __builtin_trap() __break(0); - -#endif /* _ASM_IA64_INTEL_INTRIN_H */ diff --git a/arch/ia64/include/uapi/asm/intrinsics.h b/arch/ia64/include/uapi/asm/intrinsics.h index a0e0a064f5b12..63f27c4ec739f 100644 --- a/arch/ia64/include/uapi/asm/intrinsics.h +++ b/arch/ia64/include/uapi/asm/intrinsics.h @@ -14,11 +14,7 @@ #include /* include compiler specific intrinsics */ #include -#ifdef __INTEL_COMPILER -# include -#else -# include -#endif +#include #include #define ia64_set_rr0_to_rr4(val0, val1, val2, val3, val4) \ diff --git a/include/acpi/platform/acenv.h b/include/acpi/platform/acenv.h index 03eb3d9770759..9e4f7564201a0 100644 --- a/include/acpi/platform/acenv.h +++ b/include/acpi/platform/acenv.h @@ -148,15 +148,12 @@ * *****************************************************************************/ -#if defined(__GNUC__) && !defined(__INTEL_COMPILER) +#if defined(__GNUC__) #include #elif defined(_MSC_VER) #include "acmsvc.h" -#elif defined(__INTEL_COMPILER) -#include - #endif #if defined(_LINUX) || defined(__linux__) diff --git a/include/acpi/platform/acenvex.h b/include/acpi/platform/acenvex.h index 3a6b1db9a984d..72cc7bab469e9 100644 --- a/include/acpi/platform/acenvex.h +++ b/include/acpi/platform/acenvex.h @@ -35,7 +35,7 @@ #endif -#if defined(__GNUC__) && !defined(__INTEL_COMPILER) +#if defined(__GNUC__) #include "acgccex.h" #elif defined(_MSC_VER) diff --git a/include/acpi/platform/acintel.h b/include/acpi/platform/acintel.h deleted file mode 100644 index 85b1ae86ee636..0000000000000 --- a/include/acpi/platform/acintel.h +++ /dev/null @@ -1,55 +0,0 @@ -/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 */ -/****************************************************************************** - * - * Name: acintel.h - VC specific defines, etc. - * - * Copyright (C) 2000 - 2022, Intel Corp. - * - *****************************************************************************/ - -#ifndef __ACINTEL_H__ -#define __ACINTEL_H__ - -/* - * Use compiler specific is a good practice for even when - * -nostdinc is specified (i.e., ACPI_USE_STANDARD_HEADERS undefined. - */ -#ifndef va_arg -#include -#endif - -/* Configuration specific to Intel 64-bit C compiler */ - -#define COMPILER_DEPENDENT_INT64 __int64 -#define COMPILER_DEPENDENT_UINT64 unsigned __int64 -#define ACPI_INLINE __inline - -/* - * Calling conventions: - * - * ACPI_SYSTEM_XFACE - Interfaces to host OS (handlers, threads) - * ACPI_EXTERNAL_XFACE - External ACPI interfaces - * ACPI_INTERNAL_XFACE - Internal ACPI interfaces - * ACPI_INTERNAL_VAR_XFACE - Internal variable-parameter list interfaces - */ -#define ACPI_SYSTEM_XFACE -#define ACPI_EXTERNAL_XFACE -#define ACPI_INTERNAL_XFACE -#define ACPI_INTERNAL_VAR_XFACE - -/* remark 981 - operands evaluated in no particular order */ -#pragma warning(disable:981) - -/* warn C4100: unreferenced formal parameter */ -#pragma warning(disable:4100) - -/* warn C4127: conditional expression is constant */ -#pragma warning(disable:4127) - -/* warn C4706: assignment within conditional expression */ -#pragma warning(disable:4706) - -/* warn C4214: bit field types other than int */ -#pragma warning(disable:4214) - -#endif /* __ACINTEL_H__ */ diff --git a/include/linux/compiler-intel.h b/include/linux/compiler-intel.h deleted file mode 100644 index b17f3cd18334d..0000000000000 --- a/include/linux/compiler-intel.h +++ /dev/null @@ -1,34 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef __LINUX_COMPILER_TYPES_H -#error "Please don't include directly, include instead." -#endif - -#ifdef __ECC - -/* Compiler specific definitions for Intel ECC compiler */ - -#include - -/* Intel ECC compiler doesn't support gcc specific asm stmts. - * It uses intrinsics to do the equivalent things. - */ - -#define barrier() __memory_barrier() -#define barrier_data(ptr) barrier() - -#define RELOC_HIDE(ptr, off) \ - ({ unsigned long __ptr; \ - __ptr = (unsigned long) (ptr); \ - (typeof(ptr)) (__ptr + (off)); }) - -/* This should act as an optimization barrier on var. - * Given that this compiler does not have inline assembly, a compiler barrier - * is the best we can do. - */ -#define OPTIMIZER_HIDE_VAR(var) barrier() - -#endif - -/* icc has this, but it's called _bswap16 */ -#define __HAVE_BUILTIN_BSWAP16__ -#define __builtin_bswap16 _bswap16 diff --git a/include/linux/compiler_attributes.h b/include/linux/compiler_attributes.h index 4a3bd114a24fa..e659cb6fded39 100644 --- a/include/linux/compiler_attributes.h +++ b/include/linux/compiler_attributes.h @@ -64,16 +64,10 @@ * compiler should see some alignment anyway, when the return value is * massaged by 'flags = ptr & 3; ptr &= ~3;'). * - * Optional: not supported by icc - * * gcc: https://gcc.gnu.org/onlinedocs/gcc/Common-Function-Attributes.html#index-assume_005faligned-function-attribute * clang: https://clang.llvm.org/docs/AttributeReference.html#assume-aligned */ -#if __has_attribute(__assume_aligned__) -# define __assume_aligned(a, ...) __attribute__((__assume_aligned__(a, ## __VA_ARGS__))) -#else -# define __assume_aligned(a, ...) -#endif +#define __assume_aligned(a, ...) __attribute__((__assume_aligned__(a, ## __VA_ARGS__))) /* * Note the long name. @@ -85,7 +79,6 @@ /* * Optional: only supported since gcc >= 9 * Optional: not supported by clang - * Optional: not supported by icc * * gcc: https://gcc.gnu.org/onlinedocs/gcc/Common-Function-Attributes.html#index-copy-function-attribute */ @@ -98,7 +91,6 @@ /* * Optional: not supported by gcc * Optional: only supported since clang >= 14.0 - * Optional: not supported by icc * * clang: https://clang.llvm.org/docs/AttributeReference.html#diagnose_as_builtin */ @@ -122,7 +114,6 @@ /* * Optional: not supported by clang - * Optional: not supported by icc * * gcc: https://gcc.gnu.org/onlinedocs/gcc/Common-Type-Attributes.html#index-designated_005finit-type-attribute */ @@ -236,7 +227,6 @@ /* * Optional: only supported since gcc >= 8 * Optional: not supported by clang - * Optional: not supported by icc * * gcc: https://gcc.gnu.org/onlinedocs/gcc/Common-Variable-Attributes.html#index-nonstring-variable-attribute */ @@ -267,7 +257,6 @@ /* * Optional: not supported by gcc. - * Optional: not supported by icc. * * clang: https://clang.llvm.org/docs/AttributeReference.html#overloadable */ @@ -287,7 +276,6 @@ * Note: the "type" argument should match any __builtin_object_size(p, type) usage. * * Optional: not supported by gcc. - * Optional: not supported by icc. * * clang: https://clang.llvm.org/docs/AttributeReference.html#pass-object-size-pass-dynamic-object-size */ diff --git a/include/linux/compiler_types.h b/include/linux/compiler_types.h index 690c7c826fbfd..547ea1ff806eb 100644 --- a/include/linux/compiler_types.h +++ b/include/linux/compiler_types.h @@ -120,8 +120,6 @@ static inline void __chk_io_ptr(const volatile void __iomem *ptr) { } /* Compiler specific macros. */ #ifdef __clang__ #include -#elif defined(__INTEL_COMPILER) -#include #elif defined(__GNUC__) /* The above compilers also define __GNUC__, so order is important here. */ #include diff --git a/scripts/cc-version.sh b/scripts/cc-version.sh index 2401c86fcf533..0573c92e841d3 100755 --- a/scripts/cc-version.sh +++ b/scripts/cc-version.sh @@ -12,8 +12,6 @@ get_c_compiler_info() cat <<- EOF | "$@" -E -P -x c - 2>/dev/null #if defined(__clang__) Clang __clang_major__ __clang_minor__ __clang_patchlevel__ - #elif defined(__INTEL_COMPILER) - ICC __INTEL_COMPILER __INTEL_COMPILER_UPDATE #elif defined(__GNUC__) GCC __GNUC__ __GNUC_MINOR__ __GNUC_PATCHLEVEL__ #else diff --git a/scripts/min-tool-version.sh b/scripts/min-tool-version.sh index a814f1efb39d5..20d483ec6f5fc 100755 --- a/scripts/min-tool-version.sh +++ b/scripts/min-tool-version.sh @@ -19,10 +19,6 @@ binutils) gcc) echo 5.1.0 ;; -icc) - # temporary - echo 16.0.3 - ;; llvm) if [ "$SRCARCH" = s390 ]; then echo 15.0.0 From 596ff4a09b8981790e15572e8e7bc904df5835e7 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sat, 4 Mar 2023 13:35:43 -0800 Subject: [PATCH 710/765] cpumask: re-introduce constant-sized cpumask optimizations Commit aa47a7c215e7 ("lib/cpumask: deprecate nr_cpumask_bits") resulted in the cpumask operations potentially becoming hugely less efficient, because suddenly the cpumask was always considered to be variable-sized. The optimization was then later added back in a limited form by commit 6f9c07be9d02 ("lib/cpumask: add FORCE_NR_CPUS config option"), but that FORCE_NR_CPUS option is not useful in a generic kernel and more of a special case for embedded situations with fixed hardware. Instead, just re-introduce the optimization, with some changes. Instead of depending on CPUMASK_OFFSTACK being false, and then always using the full constant cpumask width, this introduces three different cpumask "sizes": - the exact size (nr_cpumask_bits) remains identical to nr_cpu_ids. This is used for situations where we should use the exact size. - the "small" size (small_cpumask_bits) is the NR_CPUS constant if it fits in a single word and the bitmap operations thus end up able to trigger the "small_const_nbits()" optimizations. This is used for the operations that have optimized single-word cases that get inlined, notably the bit find and scanning functions. - the "large" size (large_cpumask_bits) is the NR_CPUS constant if it is an sufficiently small constant that makes simple "copy" and "clear" operations more efficient. This is arbitrarily set at four words or less. As a an example of this situation, without this fixed size optimization, cpumask_clear() will generate code like movl nr_cpu_ids(%rip), %edx addq $63, %rdx shrq $3, %rdx andl $-8, %edx callq memset@PLT on x86-64, because it would calculate the "exact" number of longwords that need to be cleared. In contrast, with this patch, using a MAX_CPU of 64 (which is quite a reasonable value to use), the above becomes a single movq $0,cpumask instruction instead, because instead of caring to figure out exactly how many CPU's the system has, it just knows that the cpumask will be a single word and can just clear it all. Note that this does end up tightening the rules a bit from the original version in another way: operations that set bits in the cpumask are now limited to the actual nr_cpu_ids limit, whereas we used to do the nr_cpumask_bits thing almost everywhere in the cpumask code. But if you just clear bits, or scan for bits, we can use the simpler compile-time constants. In the process, remove 'cpumask_complement()' and 'for_each_cpu_not()' which were not useful, and which fundamentally have to be limited to 'nr_cpu_ids'. Better remove them now than have somebody introduce use of them later. Of course, on x86-64 with MAXSMP there is no sane small compile-time constant for the cpumask sizes, and we end up using the actual CPU bits, and will generate the above kind of horrors regardless. Please don't use MAXSMP unless you really expect to have machines with thousands of cores. Signed-off-by: Linus Torvalds --- .clang-format | 1 - arch/ia64/kernel/acpi.c | 4 +- include/linux/cpumask.h | 125 ++++++++++++++++++++++------------------ lib/cpumask_kunit.c | 14 +---- 4 files changed, 72 insertions(+), 72 deletions(-) diff --git a/.clang-format b/.clang-format index 2c61b4553374b..d988e9fa9b265 100644 --- a/.clang-format +++ b/.clang-format @@ -226,7 +226,6 @@ ForEachMacros: - 'for_each_console_srcu' - 'for_each_cpu' - 'for_each_cpu_and' - - 'for_each_cpu_not' - 'for_each_cpu_wrap' - 'for_each_dapm_widgets' - 'for_each_dedup_cand' diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c index 96d13cb7c19f0..15f6cfddcc080 100644 --- a/arch/ia64/kernel/acpi.c +++ b/arch/ia64/kernel/acpi.c @@ -783,11 +783,9 @@ __init void prefill_possible_map(void) static int _acpi_map_lsapic(acpi_handle handle, int physid, int *pcpu) { - cpumask_t tmp_map; int cpu; - cpumask_complement(&tmp_map, cpu_present_mask); - cpu = cpumask_first(&tmp_map); + cpu = cpumask_first_zero(cpu_present_mask); if (cpu >= nr_cpu_ids) return -EINVAL; diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h index 10c92bd9b8070..8fbe76607965e 100644 --- a/include/linux/cpumask.h +++ b/include/linux/cpumask.h @@ -50,8 +50,41 @@ static inline void set_nr_cpu_ids(unsigned int nr) #endif } -/* Deprecated. Always use nr_cpu_ids. */ -#define nr_cpumask_bits nr_cpu_ids +/* + * We have several different "preferred sizes" for the cpumask + * operations, depending on operation. + * + * For example, the bitmap scanning and operating operations have + * optimized routines that work for the single-word case, but only when + * the size is constant. So if NR_CPUS fits in one single word, we are + * better off using that small constant, in order to trigger the + * optimized bit finding. That is 'small_cpumask_size'. + * + * The clearing and copying operations will similarly perform better + * with a constant size, but we limit that size arbitrarily to four + * words. We call this 'large_cpumask_size'. + * + * Finally, some operations just want the exact limit, either because + * they set bits or just don't have any faster fixed-sized versions. We + * call this just 'nr_cpumask_size'. + * + * Note that these optional constants are always guaranteed to be at + * least as big as 'nr_cpu_ids' itself is, and all our cpumask + * allocations are at least that size (see cpumask_size()). The + * optimization comes from being able to potentially use a compile-time + * constant instead of a run-time generated exact number of CPUs. + */ +#if NR_CPUS <= BITS_PER_LONG + #define small_cpumask_bits ((unsigned int)NR_CPUS) + #define large_cpumask_bits ((unsigned int)NR_CPUS) +#elif NR_CPUS <= 4*BITS_PER_LONG + #define small_cpumask_bits nr_cpu_ids + #define large_cpumask_bits ((unsigned int)NR_CPUS) +#else + #define small_cpumask_bits nr_cpu_ids + #define large_cpumask_bits nr_cpu_ids +#endif +#define nr_cpumask_bits nr_cpu_ids /* * The following particular system cpumasks and operations manage @@ -126,7 +159,7 @@ static __always_inline unsigned int cpumask_check(unsigned int cpu) */ static inline unsigned int cpumask_first(const struct cpumask *srcp) { - return find_first_bit(cpumask_bits(srcp), nr_cpumask_bits); + return find_first_bit(cpumask_bits(srcp), small_cpumask_bits); } /** @@ -137,7 +170,7 @@ static inline unsigned int cpumask_first(const struct cpumask *srcp) */ static inline unsigned int cpumask_first_zero(const struct cpumask *srcp) { - return find_first_zero_bit(cpumask_bits(srcp), nr_cpumask_bits); + return find_first_zero_bit(cpumask_bits(srcp), small_cpumask_bits); } /** @@ -150,7 +183,7 @@ static inline unsigned int cpumask_first_zero(const struct cpumask *srcp) static inline unsigned int cpumask_first_and(const struct cpumask *srcp1, const struct cpumask *srcp2) { - return find_first_and_bit(cpumask_bits(srcp1), cpumask_bits(srcp2), nr_cpumask_bits); + return find_first_and_bit(cpumask_bits(srcp1), cpumask_bits(srcp2), small_cpumask_bits); } /** @@ -161,7 +194,7 @@ unsigned int cpumask_first_and(const struct cpumask *srcp1, const struct cpumask */ static inline unsigned int cpumask_last(const struct cpumask *srcp) { - return find_last_bit(cpumask_bits(srcp), nr_cpumask_bits); + return find_last_bit(cpumask_bits(srcp), small_cpumask_bits); } /** @@ -177,7 +210,7 @@ unsigned int cpumask_next(int n, const struct cpumask *srcp) /* -1 is a legal arg here. */ if (n != -1) cpumask_check(n); - return find_next_bit(cpumask_bits(srcp), nr_cpumask_bits, n + 1); + return find_next_bit(cpumask_bits(srcp), small_cpumask_bits, n + 1); } /** @@ -192,7 +225,7 @@ static inline unsigned int cpumask_next_zero(int n, const struct cpumask *srcp) /* -1 is a legal arg here. */ if (n != -1) cpumask_check(n); - return find_next_zero_bit(cpumask_bits(srcp), nr_cpumask_bits, n+1); + return find_next_zero_bit(cpumask_bits(srcp), small_cpumask_bits, n+1); } #if NR_CPUS == 1 @@ -235,7 +268,7 @@ unsigned int cpumask_next_and(int n, const struct cpumask *src1p, if (n != -1) cpumask_check(n); return find_next_and_bit(cpumask_bits(src1p), cpumask_bits(src2p), - nr_cpumask_bits, n + 1); + small_cpumask_bits, n + 1); } /** @@ -246,17 +279,7 @@ unsigned int cpumask_next_and(int n, const struct cpumask *src1p, * After the loop, cpu is >= nr_cpu_ids. */ #define for_each_cpu(cpu, mask) \ - for_each_set_bit(cpu, cpumask_bits(mask), nr_cpumask_bits) - -/** - * for_each_cpu_not - iterate over every cpu in a complemented mask - * @cpu: the (optionally unsigned) integer iterator - * @mask: the cpumask pointer - * - * After the loop, cpu is >= nr_cpu_ids. - */ -#define for_each_cpu_not(cpu, mask) \ - for_each_clear_bit(cpu, cpumask_bits(mask), nr_cpumask_bits) + for_each_set_bit(cpu, cpumask_bits(mask), small_cpumask_bits) #if NR_CPUS == 1 static inline @@ -290,7 +313,7 @@ unsigned int __pure cpumask_next_wrap(int n, const struct cpumask *mask, int sta * After the loop, cpu is >= nr_cpu_ids. */ #define for_each_cpu_wrap(cpu, mask, start) \ - for_each_set_bit_wrap(cpu, cpumask_bits(mask), nr_cpumask_bits, start) + for_each_set_bit_wrap(cpu, cpumask_bits(mask), small_cpumask_bits, start) /** * for_each_cpu_and - iterate over every cpu in both masks @@ -307,7 +330,7 @@ unsigned int __pure cpumask_next_wrap(int n, const struct cpumask *mask, int sta * After the loop, cpu is >= nr_cpu_ids. */ #define for_each_cpu_and(cpu, mask1, mask2) \ - for_each_and_bit(cpu, cpumask_bits(mask1), cpumask_bits(mask2), nr_cpumask_bits) + for_each_and_bit(cpu, cpumask_bits(mask1), cpumask_bits(mask2), small_cpumask_bits) /** * for_each_cpu_andnot - iterate over every cpu present in one mask, excluding @@ -325,7 +348,7 @@ unsigned int __pure cpumask_next_wrap(int n, const struct cpumask *mask, int sta * After the loop, cpu is >= nr_cpu_ids. */ #define for_each_cpu_andnot(cpu, mask1, mask2) \ - for_each_andnot_bit(cpu, cpumask_bits(mask1), cpumask_bits(mask2), nr_cpumask_bits) + for_each_andnot_bit(cpu, cpumask_bits(mask1), cpumask_bits(mask2), small_cpumask_bits) /** * cpumask_any_but - return a "random" in a cpumask, but not this one. @@ -356,7 +379,7 @@ unsigned int cpumask_any_but(const struct cpumask *mask, unsigned int cpu) */ static inline unsigned int cpumask_nth(unsigned int cpu, const struct cpumask *srcp) { - return find_nth_bit(cpumask_bits(srcp), nr_cpumask_bits, cpumask_check(cpu)); + return find_nth_bit(cpumask_bits(srcp), small_cpumask_bits, cpumask_check(cpu)); } /** @@ -372,7 +395,7 @@ unsigned int cpumask_nth_and(unsigned int cpu, const struct cpumask *srcp1, const struct cpumask *srcp2) { return find_nth_and_bit(cpumask_bits(srcp1), cpumask_bits(srcp2), - nr_cpumask_bits, cpumask_check(cpu)); + small_cpumask_bits, cpumask_check(cpu)); } /** @@ -388,7 +411,7 @@ unsigned int cpumask_nth_andnot(unsigned int cpu, const struct cpumask *srcp1, const struct cpumask *srcp2) { return find_nth_andnot_bit(cpumask_bits(srcp1), cpumask_bits(srcp2), - nr_cpumask_bits, cpumask_check(cpu)); + small_cpumask_bits, cpumask_check(cpu)); } /** @@ -408,7 +431,7 @@ unsigned int cpumask_nth_and_andnot(unsigned int cpu, const struct cpumask *srcp return find_nth_and_andnot_bit(cpumask_bits(srcp1), cpumask_bits(srcp2), cpumask_bits(srcp3), - nr_cpumask_bits, cpumask_check(cpu)); + small_cpumask_bits, cpumask_check(cpu)); } #define CPU_BITS_NONE \ @@ -495,10 +518,14 @@ static __always_inline bool cpumask_test_and_clear_cpu(int cpu, struct cpumask * /** * cpumask_setall - set all cpus (< nr_cpu_ids) in a cpumask * @dstp: the cpumask pointer + * + * Note: since we set bits, we should use the tighter 'bitmap_set()' with + * the eact number of bits, not 'bitmap_fill()' that will fill past the + * end. */ static inline void cpumask_setall(struct cpumask *dstp) { - bitmap_fill(cpumask_bits(dstp), nr_cpumask_bits); + bitmap_set(cpumask_bits(dstp), 0, nr_cpumask_bits); } /** @@ -507,7 +534,7 @@ static inline void cpumask_setall(struct cpumask *dstp) */ static inline void cpumask_clear(struct cpumask *dstp) { - bitmap_zero(cpumask_bits(dstp), nr_cpumask_bits); + bitmap_zero(cpumask_bits(dstp), large_cpumask_bits); } /** @@ -523,7 +550,7 @@ static inline bool cpumask_and(struct cpumask *dstp, const struct cpumask *src2p) { return bitmap_and(cpumask_bits(dstp), cpumask_bits(src1p), - cpumask_bits(src2p), nr_cpumask_bits); + cpumask_bits(src2p), small_cpumask_bits); } /** @@ -536,7 +563,7 @@ static inline void cpumask_or(struct cpumask *dstp, const struct cpumask *src1p, const struct cpumask *src2p) { bitmap_or(cpumask_bits(dstp), cpumask_bits(src1p), - cpumask_bits(src2p), nr_cpumask_bits); + cpumask_bits(src2p), small_cpumask_bits); } /** @@ -550,7 +577,7 @@ static inline void cpumask_xor(struct cpumask *dstp, const struct cpumask *src2p) { bitmap_xor(cpumask_bits(dstp), cpumask_bits(src1p), - cpumask_bits(src2p), nr_cpumask_bits); + cpumask_bits(src2p), small_cpumask_bits); } /** @@ -566,19 +593,7 @@ static inline bool cpumask_andnot(struct cpumask *dstp, const struct cpumask *src2p) { return bitmap_andnot(cpumask_bits(dstp), cpumask_bits(src1p), - cpumask_bits(src2p), nr_cpumask_bits); -} - -/** - * cpumask_complement - *dstp = ~*srcp - * @dstp: the cpumask result - * @srcp: the input to invert - */ -static inline void cpumask_complement(struct cpumask *dstp, - const struct cpumask *srcp) -{ - bitmap_complement(cpumask_bits(dstp), cpumask_bits(srcp), - nr_cpumask_bits); + cpumask_bits(src2p), small_cpumask_bits); } /** @@ -590,7 +605,7 @@ static inline bool cpumask_equal(const struct cpumask *src1p, const struct cpumask *src2p) { return bitmap_equal(cpumask_bits(src1p), cpumask_bits(src2p), - nr_cpumask_bits); + small_cpumask_bits); } /** @@ -604,7 +619,7 @@ static inline bool cpumask_or_equal(const struct cpumask *src1p, const struct cpumask *src3p) { return bitmap_or_equal(cpumask_bits(src1p), cpumask_bits(src2p), - cpumask_bits(src3p), nr_cpumask_bits); + cpumask_bits(src3p), small_cpumask_bits); } /** @@ -616,7 +631,7 @@ static inline bool cpumask_intersects(const struct cpumask *src1p, const struct cpumask *src2p) { return bitmap_intersects(cpumask_bits(src1p), cpumask_bits(src2p), - nr_cpumask_bits); + small_cpumask_bits); } /** @@ -630,7 +645,7 @@ static inline bool cpumask_subset(const struct cpumask *src1p, const struct cpumask *src2p) { return bitmap_subset(cpumask_bits(src1p), cpumask_bits(src2p), - nr_cpumask_bits); + small_cpumask_bits); } /** @@ -639,7 +654,7 @@ static inline bool cpumask_subset(const struct cpumask *src1p, */ static inline bool cpumask_empty(const struct cpumask *srcp) { - return bitmap_empty(cpumask_bits(srcp), nr_cpumask_bits); + return bitmap_empty(cpumask_bits(srcp), small_cpumask_bits); } /** @@ -657,7 +672,7 @@ static inline bool cpumask_full(const struct cpumask *srcp) */ static inline unsigned int cpumask_weight(const struct cpumask *srcp) { - return bitmap_weight(cpumask_bits(srcp), nr_cpumask_bits); + return bitmap_weight(cpumask_bits(srcp), small_cpumask_bits); } /** @@ -668,7 +683,7 @@ static inline unsigned int cpumask_weight(const struct cpumask *srcp) static inline unsigned int cpumask_weight_and(const struct cpumask *srcp1, const struct cpumask *srcp2) { - return bitmap_weight_and(cpumask_bits(srcp1), cpumask_bits(srcp2), nr_cpumask_bits); + return bitmap_weight_and(cpumask_bits(srcp1), cpumask_bits(srcp2), small_cpumask_bits); } /** @@ -681,7 +696,7 @@ static inline void cpumask_shift_right(struct cpumask *dstp, const struct cpumask *srcp, int n) { bitmap_shift_right(cpumask_bits(dstp), cpumask_bits(srcp), n, - nr_cpumask_bits); + small_cpumask_bits); } /** @@ -705,7 +720,7 @@ static inline void cpumask_shift_left(struct cpumask *dstp, static inline void cpumask_copy(struct cpumask *dstp, const struct cpumask *srcp) { - bitmap_copy(cpumask_bits(dstp), cpumask_bits(srcp), nr_cpumask_bits); + bitmap_copy(cpumask_bits(dstp), cpumask_bits(srcp), large_cpumask_bits); } /** @@ -789,7 +804,7 @@ static inline int cpulist_parse(const char *buf, struct cpumask *dstp) */ static inline unsigned int cpumask_size(void) { - return BITS_TO_LONGS(nr_cpumask_bits) * sizeof(long); + return BITS_TO_LONGS(large_cpumask_bits) * sizeof(long); } /* diff --git a/lib/cpumask_kunit.c b/lib/cpumask_kunit.c index d1fc6ece21f37..a105e6369efc9 100644 --- a/lib/cpumask_kunit.c +++ b/lib/cpumask_kunit.c @@ -23,16 +23,6 @@ KUNIT_EXPECT_EQ_MSG((test), mask_weight, iter, MASK_MSG(mask)); \ } while (0) -#define EXPECT_FOR_EACH_CPU_NOT_EQ(test, mask) \ - do { \ - const cpumask_t *m = (mask); \ - int mask_weight = cpumask_weight(m); \ - int cpu, iter = 0; \ - for_each_cpu_not(cpu, m) \ - iter++; \ - KUNIT_EXPECT_EQ_MSG((test), nr_cpu_ids - mask_weight, iter, MASK_MSG(mask)); \ - } while (0) - #define EXPECT_FOR_EACH_CPU_OP_EQ(test, op, mask1, mask2) \ do { \ const cpumask_t *m1 = (mask1); \ @@ -77,7 +67,7 @@ static void test_cpumask_weight(struct kunit *test) KUNIT_EXPECT_EQ_MSG(test, 0, cpumask_weight(&mask_empty), MASK_MSG(&mask_empty)); KUNIT_EXPECT_EQ_MSG(test, nr_cpu_ids, cpumask_weight(cpu_possible_mask), MASK_MSG(cpu_possible_mask)); - KUNIT_EXPECT_EQ_MSG(test, nr_cpumask_bits, cpumask_weight(&mask_all), MASK_MSG(&mask_all)); + KUNIT_EXPECT_EQ_MSG(test, nr_cpu_ids, cpumask_weight(&mask_all), MASK_MSG(&mask_all)); } static void test_cpumask_first(struct kunit *test) @@ -113,14 +103,12 @@ static void test_cpumask_next(struct kunit *test) static void test_cpumask_iterators(struct kunit *test) { EXPECT_FOR_EACH_CPU_EQ(test, &mask_empty); - EXPECT_FOR_EACH_CPU_NOT_EQ(test, &mask_empty); EXPECT_FOR_EACH_CPU_WRAP_EQ(test, &mask_empty); EXPECT_FOR_EACH_CPU_OP_EQ(test, and, &mask_empty, &mask_empty); EXPECT_FOR_EACH_CPU_OP_EQ(test, and, cpu_possible_mask, &mask_empty); EXPECT_FOR_EACH_CPU_OP_EQ(test, andnot, &mask_empty, &mask_empty); EXPECT_FOR_EACH_CPU_EQ(test, cpu_possible_mask); - EXPECT_FOR_EACH_CPU_NOT_EQ(test, cpu_possible_mask); EXPECT_FOR_EACH_CPU_WRAP_EQ(test, cpu_possible_mask); EXPECT_FOR_EACH_CPU_OP_EQ(test, and, cpu_possible_mask, cpu_possible_mask); EXPECT_FOR_EACH_CPU_OP_EQ(test, andnot, cpu_possible_mask, &mask_empty); From fe15c26ee26efa11741a7b632e9f23b01aca4cc6 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 5 Mar 2023 14:52:03 -0800 Subject: [PATCH 711/765] Linux 6.3-rc1 --- Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Makefile b/Makefile index aa26f34a9b522..d7bd0eb9b3463 100644 --- a/Makefile +++ b/Makefile @@ -1,8 +1,8 @@ # SPDX-License-Identifier: GPL-2.0 VERSION = 6 -PATCHLEVEL = 2 +PATCHLEVEL = 3 SUBLEVEL = 0 -EXTRAVERSION = +EXTRAVERSION = -rc1 NAME = Hurr durr I'ma ninja sloth # *DOCUMENTATION* From a9334b702a03b693f54ebd3b98f67bf722b74870 Mon Sep 17 00:00:00 2001 From: Rongguang Wei Date: Thu, 2 Mar 2023 14:21:43 +0800 Subject: [PATCH 712/765] net: stmmac: add to set device wake up flag when stmmac init phy When MAC is not support PMT, driver will check PHY's WoL capability and set device wakeup capability in stmmac_init_phy(). We can enable the WoL through ethtool, the driver would enable the device wake up flag. Now the device_may_wakeup() return true. But if there is a way which enable the PHY's WoL capability derectly, like in BIOS. The driver would not know the enable thing and would not set the device wake up flag. The phy_suspend may failed like this: [ 32.409063] PM: dpm_run_callback(): mdio_bus_phy_suspend+0x0/0x50 returns -16 [ 32.409065] PM: Device stmmac-1:00 failed to suspend: error -16 [ 32.409067] PM: Some devices failed to suspend, or early wake event detected Add to set the device wakeup enable flag according to the get_wol function result in PHY can fix the error in this scene. v2: add a Fixes tag. Fixes: 1d8e5b0f3f2c ("net: stmmac: Support WOL with phy") Signed-off-by: Rongguang Wei Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index e4902a7bb61ed..8f543c3ab5c56 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -1170,6 +1170,7 @@ static int stmmac_init_phy(struct net_device *dev) phylink_ethtool_get_wol(priv->phylink, &wol); device_set_wakeup_capable(priv->device, !!wol.supported); + device_set_wakeup_enable(priv->device, !!wol.wolopts); } return ret; From f4b47a2e9463950df3e7c8b70e017877c1d4eb11 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 3 Mar 2023 16:37:54 +0000 Subject: [PATCH 713/765] net: phylib: get rid of unnecessary locking The locking in phy_probe() and phy_remove() does very little to prevent any races with e.g. phy_attach_direct(), but instead causes lockdep ABBA warnings. Remove it. ====================================================== WARNING: possible circular locking dependency detected 6.2.0-dirty #1108 Tainted: G W E ------------------------------------------------------ ip/415 is trying to acquire lock: ffff5c268f81ef50 (&dev->lock){+.+.}-{3:3}, at: phy_attach_direct+0x17c/0x3a0 [libphy] but task is already holding lock: ffffaef6496cb518 (rtnl_mutex){+.+.}-{3:3}, at: rtnetlink_rcv_msg+0x154/0x560 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (rtnl_mutex){+.+.}-{3:3}: __lock_acquire+0x35c/0x6c0 lock_acquire.part.0+0xcc/0x220 lock_acquire+0x68/0x84 __mutex_lock+0x8c/0x414 mutex_lock_nested+0x34/0x40 rtnl_lock+0x24/0x30 sfp_bus_add_upstream+0x34/0x150 phy_sfp_probe+0x4c/0x94 [libphy] mv3310_probe+0x148/0x184 [marvell10g] phy_probe+0x8c/0x200 [libphy] call_driver_probe+0xbc/0x15c really_probe+0xc0/0x320 __driver_probe_device+0x84/0x120 driver_probe_device+0x44/0x120 __device_attach_driver+0xc4/0x160 bus_for_each_drv+0x80/0xe0 __device_attach+0xb0/0x1f0 device_initial_probe+0x1c/0x2c bus_probe_device+0xa4/0xb0 device_add+0x360/0x53c phy_device_register+0x60/0xa4 [libphy] fwnode_mdiobus_phy_device_register+0xc0/0x190 [fwnode_mdio] fwnode_mdiobus_register_phy+0x160/0xd80 [fwnode_mdio] of_mdiobus_register+0x140/0x340 [of_mdio] orion_mdio_probe+0x298/0x3c0 [mvmdio] platform_probe+0x70/0xe0 call_driver_probe+0x34/0x15c really_probe+0xc0/0x320 __driver_probe_device+0x84/0x120 driver_probe_device+0x44/0x120 __driver_attach+0x104/0x210 bus_for_each_dev+0x78/0xdc driver_attach+0x2c/0x3c bus_add_driver+0x184/0x240 driver_register+0x80/0x13c __platform_driver_register+0x30/0x3c xt_compat_calc_jump+0x28/0xa4 [x_tables] do_one_initcall+0x50/0x1b0 do_init_module+0x50/0x1fc load_module+0x684/0x744 __do_sys_finit_module+0xc4/0x140 __arm64_sys_finit_module+0x28/0x34 invoke_syscall+0x50/0x120 el0_svc_common.constprop.0+0x6c/0x1b0 do_el0_svc+0x34/0x44 el0_svc+0x48/0xf0 el0t_64_sync_handler+0xb8/0xc0 el0t_64_sync+0x1a0/0x1a4 -> #0 (&dev->lock){+.+.}-{3:3}: check_prev_add+0xb4/0xc80 validate_chain+0x414/0x47c __lock_acquire+0x35c/0x6c0 lock_acquire.part.0+0xcc/0x220 lock_acquire+0x68/0x84 __mutex_lock+0x8c/0x414 mutex_lock_nested+0x34/0x40 phy_attach_direct+0x17c/0x3a0 [libphy] phylink_fwnode_phy_connect.part.0+0x70/0xe4 [phylink] phylink_fwnode_phy_connect+0x48/0x60 [phylink] mvpp2_open+0xec/0x2e0 [mvpp2] __dev_open+0x104/0x214 __dev_change_flags+0x1d4/0x254 dev_change_flags+0x2c/0x7c do_setlink+0x254/0xa50 __rtnl_newlink+0x430/0x514 rtnl_newlink+0x58/0x8c rtnetlink_rcv_msg+0x17c/0x560 netlink_rcv_skb+0x64/0x150 rtnetlink_rcv+0x20/0x30 netlink_unicast+0x1d4/0x2b4 netlink_sendmsg+0x1a4/0x400 ____sys_sendmsg+0x228/0x290 ___sys_sendmsg+0x88/0xec __sys_sendmsg+0x70/0xd0 __arm64_sys_sendmsg+0x2c/0x40 invoke_syscall+0x50/0x120 el0_svc_common.constprop.0+0x6c/0x1b0 do_el0_svc+0x34/0x44 el0_svc+0x48/0xf0 el0t_64_sync_handler+0xb8/0xc0 el0t_64_sync+0x1a0/0x1a4 other info that might help us debug this: Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(rtnl_mutex); lock(&dev->lock); lock(rtnl_mutex); lock(&dev->lock); *** DEADLOCK *** Fixes: 298e54fa810e ("net: phy: add core phylib sfp support") Reported-by: Marc Zyngier Signed-off-by: Russell King (Oracle) Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/phy/phy_device.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 9e9fd8ff00f69..1785f1cead971 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -3098,8 +3098,6 @@ static int phy_probe(struct device *dev) if (phydrv->flags & PHY_IS_INTERNAL) phydev->is_internal = true; - mutex_lock(&phydev->lock); - /* Deassert the reset signal */ phy_device_reset(phydev, 0); @@ -3188,12 +3186,10 @@ static int phy_probe(struct device *dev) phydev->state = PHY_READY; out: - /* Assert the reset signal */ + /* Re-assert the reset signal on error */ if (err) phy_device_reset(phydev, 1); - mutex_unlock(&phydev->lock); - return err; } @@ -3203,9 +3199,7 @@ static int phy_remove(struct device *dev) cancel_delayed_work_sync(&phydev->state_queue); - mutex_lock(&phydev->lock); phydev->state = PHY_DOWN; - mutex_unlock(&phydev->lock); sfp_bus_del_upstream(phydev->sfp_bus); phydev->sfp_bus = NULL; From accd7e23693aaaa9aa0d3e9eca0ae77d1be80ab3 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Fri, 3 Mar 2023 18:43:57 -0800 Subject: [PATCH 714/765] bnxt_en: Avoid order-5 memory allocation for TPA data The driver needs to keep track of all the possible concurrent TPA (GRO/LRO) completions on the aggregation ring. On P5 chips, the maximum number of concurrent TPA is 256 and the amount of memory we allocate is order-5 on systems using 4K pages. Memory allocation failure has been reported: NetworkManager: page allocation failure: order:5, mode:0x40dc0(GFP_KERNEL|__GFP_COMP|__GFP_ZERO), nodemask=(null),cpuset=/,mems_allowed=0-1 CPU: 15 PID: 2995 Comm: NetworkManager Kdump: loaded Not tainted 5.10.156 #1 Hardware name: Dell Inc. PowerEdge R660/0M1CC5, BIOS 0.2.25 08/12/2022 Call Trace: dump_stack+0x57/0x6e warn_alloc.cold.120+0x7b/0xdd ? _cond_resched+0x15/0x30 ? __alloc_pages_direct_compact+0x15f/0x170 __alloc_pages_slowpath.constprop.108+0xc58/0xc70 __alloc_pages_nodemask+0x2d0/0x300 kmalloc_order+0x24/0xe0 kmalloc_order_trace+0x19/0x80 bnxt_alloc_mem+0x1150/0x15c0 [bnxt_en] ? bnxt_get_func_stat_ctxs+0x13/0x60 [bnxt_en] __bnxt_open_nic+0x12e/0x780 [bnxt_en] bnxt_open+0x10b/0x240 [bnxt_en] __dev_open+0xe9/0x180 __dev_change_flags+0x1af/0x220 dev_change_flags+0x21/0x60 do_setlink+0x35c/0x1100 Instead of allocating this big chunk of memory and dividing it up for the concurrent TPA instances, allocate each small chunk separately for each TPA instance. This will reduce it to order-0 allocations. Fixes: 79632e9ba386 ("bnxt_en: Expand bnxt_tpa_info struct to support 57500 chips.") Reviewed-by: Somnath Kotur Reviewed-by: Damodharam Ammepalli Reviewed-by: Pavan Chebbi Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 23 ++++++++++++----------- 1 file changed, 12 insertions(+), 11 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 5d4b1f2ebeac7..f96d539b469c9 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -3145,7 +3145,7 @@ static int bnxt_alloc_ring(struct bnxt *bp, struct bnxt_ring_mem_info *rmem) static void bnxt_free_tpa_info(struct bnxt *bp) { - int i; + int i, j; for (i = 0; i < bp->rx_nr_rings; i++) { struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i]; @@ -3153,8 +3153,10 @@ static void bnxt_free_tpa_info(struct bnxt *bp) kfree(rxr->rx_tpa_idx_map); rxr->rx_tpa_idx_map = NULL; if (rxr->rx_tpa) { - kfree(rxr->rx_tpa[0].agg_arr); - rxr->rx_tpa[0].agg_arr = NULL; + for (j = 0; j < bp->max_tpa; j++) { + kfree(rxr->rx_tpa[j].agg_arr); + rxr->rx_tpa[j].agg_arr = NULL; + } } kfree(rxr->rx_tpa); rxr->rx_tpa = NULL; @@ -3163,14 +3165,13 @@ static void bnxt_free_tpa_info(struct bnxt *bp) static int bnxt_alloc_tpa_info(struct bnxt *bp) { - int i, j, total_aggs = 0; + int i, j; bp->max_tpa = MAX_TPA; if (bp->flags & BNXT_FLAG_CHIP_P5) { if (!bp->max_tpa_v2) return 0; bp->max_tpa = max_t(u16, bp->max_tpa_v2, MAX_TPA_P5); - total_aggs = bp->max_tpa * MAX_SKB_FRAGS; } for (i = 0; i < bp->rx_nr_rings; i++) { @@ -3184,12 +3185,12 @@ static int bnxt_alloc_tpa_info(struct bnxt *bp) if (!(bp->flags & BNXT_FLAG_CHIP_P5)) continue; - agg = kcalloc(total_aggs, sizeof(*agg), GFP_KERNEL); - rxr->rx_tpa[0].agg_arr = agg; - if (!agg) - return -ENOMEM; - for (j = 1; j < bp->max_tpa; j++) - rxr->rx_tpa[j].agg_arr = agg + j * MAX_SKB_FRAGS; + for (j = 0; j < bp->max_tpa; j++) { + agg = kcalloc(MAX_SKB_FRAGS, sizeof(*agg), GFP_KERNEL); + if (!agg) + return -ENOMEM; + rxr->rx_tpa[j].agg_arr = agg; + } rxr->rx_tpa_idx_map = kzalloc(sizeof(*rxr->rx_tpa_idx_map), GFP_KERNEL); if (!rxr->rx_tpa_idx_map) From 89b59a84cb166f1ab5b6de9830e61324937c661e Mon Sep 17 00:00:00 2001 From: Selvin Xavier Date: Fri, 3 Mar 2023 18:43:58 -0800 Subject: [PATCH 715/765] bnxt_en: Fix the double free during device removal Following warning reported by KASAN during driver unload ================================================================== BUG: KASAN: double-free in bnxt_remove_one+0x103/0x200 [bnxt_en] Free of addr ffff88814e8dd4c0 by task rmmod/17469 CPU: 47 PID: 17469 Comm: rmmod Kdump: loaded Tainted: G S 6.2.0-rc7+ #2 Hardware name: Dell Inc. PowerEdge R740/01YM03, BIOS 2.3.10 08/15/2019 Call Trace: dump_stack_lvl+0x33/0x46 print_report+0x17b/0x4b3 ? __call_rcu_common.constprop.79+0x27e/0x8c0 ? __pfx_free_object_rcu+0x10/0x10 ? __virt_addr_valid+0xe3/0x160 ? bnxt_remove_one+0x103/0x200 [bnxt_en] kasan_report_invalid_free+0x64/0xd0 ? bnxt_remove_one+0x103/0x200 [bnxt_en] ? bnxt_remove_one+0x103/0x200 [bnxt_en] __kasan_slab_free+0x179/0x1c0 ? bnxt_remove_one+0x103/0x200 [bnxt_en] __kmem_cache_free+0x194/0x350 bnxt_remove_one+0x103/0x200 [bnxt_en] pci_device_remove+0x62/0x110 device_release_driver_internal+0xf6/0x1c0 driver_detach+0x76/0xe0 bus_remove_driver+0x89/0x160 pci_unregister_driver+0x26/0x110 ? strncpy_from_user+0x188/0x1c0 bnxt_exit+0xc/0x24 [bnxt_en] __x64_sys_delete_module+0x21f/0x390 ? __pfx___x64_sys_delete_module+0x10/0x10 ? __pfx_mem_cgroup_handle_over_high+0x10/0x10 ? _raw_spin_lock+0x87/0xe0 ? __pfx__raw_spin_lock+0x10/0x10 ? __audit_syscall_entry+0x185/0x210 ? ktime_get_coarse_real_ts64+0x51/0x80 ? syscall_trace_enter.isra.18+0x126/0x1a0 do_syscall_64+0x37/0x90 entry_SYSCALL_64_after_hwframe+0x72/0xdc RIP: 0033:0x7effcb6fd71b Code: 73 01 c3 48 8b 0d 6d 17 2c 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa b8 b0 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 3d 17 2c 00 f7 d8 64 89 01 48 RSP: 002b:00007ffeada270b8 EFLAGS: 00000206 ORIG_RAX: 00000000000000b0 RAX: ffffffffffffffda RBX: 00005623660e0750 RCX: 00007effcb6fd71b RDX: 000000000000000a RSI: 0000000000000800 RDI: 00005623660e07b8 RBP: 0000000000000000 R08: 00007ffeada26031 R09: 0000000000000000 R10: 00007effcb771280 R11: 0000000000000206 R12: 00007ffeada272e0 R13: 00007ffeada28bc4 R14: 00005623660e02a0 R15: 00005623660e0750 Auxiliary device structures are freed in bnxt_aux_dev_release. So avoid calling kfree from bnxt_remove_one. Also, set bp->edev to NULL before freeing the auxilary private structure. Fixes: d80d88b0dfff ("bnxt_en: Add auxiliary driver support") Reviewed-by: Ajit Khaparde Reviewed-by: Andy Gospodarek Signed-off-by: Selvin Xavier Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 -- drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c | 2 ++ 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index f96d539b469c9..808236dc898b8 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -13205,8 +13205,6 @@ static void bnxt_remove_one(struct pci_dev *pdev) bnxt_free_hwrm_resources(bp); bnxt_ethtool_free(bp); bnxt_dcb_free(bp); - kfree(bp->edev); - bp->edev = NULL; kfree(bp->ptp_cfg); bp->ptp_cfg = NULL; kfree(bp->fw_health); diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c index d4cc9c371e7bc..e7b5e28ee29f1 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c @@ -317,9 +317,11 @@ static void bnxt_aux_dev_release(struct device *dev) { struct bnxt_aux_priv *aux_priv = container_of(dev, struct bnxt_aux_priv, aux_dev.dev); + struct bnxt *bp = netdev_priv(aux_priv->edev->net); ida_free(&bnxt_aux_dev_ids, aux_priv->id); kfree(aux_priv->edev->ulp_tbl); + bp->edev = NULL; kfree(aux_priv->edev); kfree(aux_priv); } From 9f7dd42f0db1dc6915a52d4a8a96ca18dd8cc34e Mon Sep 17 00:00:00 2001 From: Ivan Delalande Date: Thu, 2 Mar 2023 17:48:31 -0800 Subject: [PATCH 716/765] netfilter: ctnetlink: revert to dumping mark regardless of event type It seems that change was unintentional, we have userspace code that needs the mark while listening for events like REPLY, DESTROY, etc. Also include 0-marks in requested dumps, as they were before that fix. Fixes: 1feeae071507 ("netfilter: ctnetlink: fix compilation warning after data race fixes in ct mark") Signed-off-by: Ivan Delalande Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_netlink.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index c11dff91d52d2..bfc3aaa2c872b 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -328,11 +328,12 @@ ctnetlink_dump_timestamp(struct sk_buff *skb, const struct nf_conn *ct) } #ifdef CONFIG_NF_CONNTRACK_MARK -static int ctnetlink_dump_mark(struct sk_buff *skb, const struct nf_conn *ct) +static int ctnetlink_dump_mark(struct sk_buff *skb, const struct nf_conn *ct, + bool dump) { u32 mark = READ_ONCE(ct->mark); - if (!mark) + if (!mark && !dump) return 0; if (nla_put_be32(skb, CTA_MARK, htonl(mark))) @@ -343,7 +344,7 @@ static int ctnetlink_dump_mark(struct sk_buff *skb, const struct nf_conn *ct) return -1; } #else -#define ctnetlink_dump_mark(a, b) (0) +#define ctnetlink_dump_mark(a, b, c) (0) #endif #ifdef CONFIG_NF_CONNTRACK_SECMARK @@ -548,7 +549,7 @@ static int ctnetlink_dump_extinfo(struct sk_buff *skb, static int ctnetlink_dump_info(struct sk_buff *skb, struct nf_conn *ct) { if (ctnetlink_dump_status(skb, ct) < 0 || - ctnetlink_dump_mark(skb, ct) < 0 || + ctnetlink_dump_mark(skb, ct, true) < 0 || ctnetlink_dump_secctx(skb, ct) < 0 || ctnetlink_dump_id(skb, ct) < 0 || ctnetlink_dump_use(skb, ct) < 0 || @@ -831,8 +832,7 @@ ctnetlink_conntrack_event(unsigned int events, const struct nf_ct_event *item) } #ifdef CONFIG_NF_CONNTRACK_MARK - if (events & (1 << IPCT_MARK) && - ctnetlink_dump_mark(skb, ct) < 0) + if (ctnetlink_dump_mark(skb, ct, events & (1 << IPCT_MARK))) goto nla_put_failure; #endif nlmsg_end(skb, nlh); @@ -2735,7 +2735,7 @@ static int __ctnetlink_glue_build(struct sk_buff *skb, struct nf_conn *ct) goto nla_put_failure; #ifdef CONFIG_NF_CONNTRACK_MARK - if (ctnetlink_dump_mark(skb, ct) < 0) + if (ctnetlink_dump_mark(skb, ct, true) < 0) goto nla_put_failure; #endif if (ctnetlink_dump_labels(skb, ct) < 0) From 4a02426787bf024dafdb79b362285ee325de3f5e Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 3 Mar 2023 10:58:56 +0100 Subject: [PATCH 717/765] netfilter: tproxy: fix deadlock due to missing BH disable MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The xtables packet traverser performs an unconditional local_bh_disable(), but the nf_tables evaluation loop does not. Functions that are called from either xtables or nftables must assume that they can be called in process context. inet_twsk_deschedule_put() assumes that no softirq interrupt can occur. If tproxy is used from nf_tables its possible that we'll deadlock trying to aquire a lock already held in process context. Add a small helper that takes care of this and use it. Link: https://lore.kernel.org/netfilter-devel/401bd6ed-314a-a196-1cdc-e13c720cc8f2@balasys.hu/ Fixes: 4ed8eb6570a4 ("netfilter: nf_tables: Add native tproxy support") Reported-and-tested-by: Major Dávid Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tproxy.h | 7 +++++++ net/ipv4/netfilter/nf_tproxy_ipv4.c | 2 +- net/ipv6/netfilter/nf_tproxy_ipv6.c | 2 +- 3 files changed, 9 insertions(+), 2 deletions(-) diff --git a/include/net/netfilter/nf_tproxy.h b/include/net/netfilter/nf_tproxy.h index 82d0e41b76f22..faa108b1ba675 100644 --- a/include/net/netfilter/nf_tproxy.h +++ b/include/net/netfilter/nf_tproxy.h @@ -17,6 +17,13 @@ static inline bool nf_tproxy_sk_is_transparent(struct sock *sk) return false; } +static inline void nf_tproxy_twsk_deschedule_put(struct inet_timewait_sock *tw) +{ + local_bh_disable(); + inet_twsk_deschedule_put(tw); + local_bh_enable(); +} + /* assign a socket to the skb -- consumes sk */ static inline void nf_tproxy_assign_sock(struct sk_buff *skb, struct sock *sk) { diff --git a/net/ipv4/netfilter/nf_tproxy_ipv4.c b/net/ipv4/netfilter/nf_tproxy_ipv4.c index b22b2c745c76c..69e3317996043 100644 --- a/net/ipv4/netfilter/nf_tproxy_ipv4.c +++ b/net/ipv4/netfilter/nf_tproxy_ipv4.c @@ -38,7 +38,7 @@ nf_tproxy_handle_time_wait4(struct net *net, struct sk_buff *skb, hp->source, lport ? lport : hp->dest, skb->dev, NF_TPROXY_LOOKUP_LISTENER); if (sk2) { - inet_twsk_deschedule_put(inet_twsk(sk)); + nf_tproxy_twsk_deschedule_put(inet_twsk(sk)); sk = sk2; } } diff --git a/net/ipv6/netfilter/nf_tproxy_ipv6.c b/net/ipv6/netfilter/nf_tproxy_ipv6.c index 929502e51203b..52f828bb5a83d 100644 --- a/net/ipv6/netfilter/nf_tproxy_ipv6.c +++ b/net/ipv6/netfilter/nf_tproxy_ipv6.c @@ -63,7 +63,7 @@ nf_tproxy_handle_time_wait6(struct sk_buff *skb, int tproto, int thoff, lport ? lport : hp->dest, skb->dev, NF_TPROXY_LOOKUP_LISTENER); if (sk2) { - inet_twsk_deschedule_put(inet_twsk(sk)); + nf_tproxy_twsk_deschedule_put(inet_twsk(sk)); sk = sk2; } } From 0d9fad91abfd723ea5070a46d98a9f4496c93ba9 Mon Sep 17 00:00:00 2001 From: Kars de Jong Date: Thu, 23 Feb 2023 12:23:49 +0100 Subject: [PATCH 718/765] m68k: mm: Fix systems with memory at end of 32-bit address space The calculation of end addresses of memory chunks overflowed to 0 when a memory chunk is located at the end of 32-bit address space. This is the case for the HP300 architecture. Link: https://lore.kernel.org/linux-m68k/CACz-3rhUo5pgNwdWHaPWmz+30Qo9xCg70wNxdf7o5x-6tXq8QQ@mail.gmail.com/ Signed-off-by: Kars de Jong Reviewed-by: Geert Uytterhoeven Link: https://lore.kernel.org/r/20230223112349.26675-1-jongk@linux-m68k.org Signed-off-by: Geert Uytterhoeven --- arch/m68k/mm/motorola.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/arch/m68k/mm/motorola.c b/arch/m68k/mm/motorola.c index 2a375637e0077..9113012240789 100644 --- a/arch/m68k/mm/motorola.c +++ b/arch/m68k/mm/motorola.c @@ -437,7 +437,7 @@ void __init paging_init(void) } min_addr = m68k_memory[0].addr; - max_addr = min_addr + m68k_memory[0].size; + max_addr = min_addr + m68k_memory[0].size - 1; memblock_add_node(m68k_memory[0].addr, m68k_memory[0].size, 0, MEMBLOCK_NONE); for (i = 1; i < m68k_num_memory;) { @@ -452,21 +452,21 @@ void __init paging_init(void) } memblock_add_node(m68k_memory[i].addr, m68k_memory[i].size, i, MEMBLOCK_NONE); - addr = m68k_memory[i].addr + m68k_memory[i].size; + addr = m68k_memory[i].addr + m68k_memory[i].size - 1; if (addr > max_addr) max_addr = addr; i++; } m68k_memoffset = min_addr - PAGE_OFFSET; - m68k_virt_to_node_shift = fls(max_addr - min_addr - 1) - 6; + m68k_virt_to_node_shift = fls(max_addr - min_addr) - 6; module_fixup(NULL, __start_fixup, __stop_fixup); flush_icache(); - high_memory = phys_to_virt(max_addr); + high_memory = phys_to_virt(max_addr) + 1; min_low_pfn = availmem >> PAGE_SHIFT; - max_pfn = max_low_pfn = max_addr >> PAGE_SHIFT; + max_pfn = max_low_pfn = (max_addr >> PAGE_SHIFT) + 1; /* Reserve kernel text/data/bss and the memory allocated in head.S */ memblock_reserve(m68k_memory[0].addr, availmem - m68k_memory[0].addr); From d4b97925e87eb133e400fe4a482d750c74ce392f Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 27 Feb 2023 21:14:13 +0100 Subject: [PATCH 719/765] m68k: mm: Move initrd phys_to_virt handling after paging_init() When booting with an initial ramdisk on platforms where physical memory does not start at address zero (e.g. on Amiga): initrd: 0ef0602c - 0f800000 Zone ranges: DMA [mem 0x0000000008000000-0x000000f7ffffffff] Normal empty Movable zone start for each node Early memory node ranges node 0: [mem 0x0000000008000000-0x000000000f7fffff] Initmem setup node 0 [mem 0x0000000008000000-0x000000000f7fffff] Unable to handle kernel access at virtual address (ptrval) Oops: 00000000 Modules linked in: PC: [<00201d3c>] memcmp+0x28/0x56 As phys_to_virt() relies on m68k_memoffset and module_fixup(), it must not be called before paging_init(). Hence postpone the phys_to_virt handling for the initial ramdisk until after calling paging_init(). While at it, reduce #ifdef clutter by using IS_ENABLED() instead. Fixes: 376e3fdecb0dcae2 ("m68k: Enable memtest functionality") Reported-by: Stephen Walsh Link: https://lists.debian.org/debian-68k/2022/09/msg00007.html Reported-by: John Paul Adrian Glaubitz Link: https://lore.kernel.org/r/4f45f05f377bf3f5baf88dbd5c3c8aeac59d94f0.camel@physik.fu-berlin.de Signed-off-by: Geert Uytterhoeven Acked-by: Finn Thain Link: https://lore.kernel.org/r/dff216da09ab7a60217c3fc2147e671ae07d636f.1677528627.git.geert@linux-m68k.org --- arch/m68k/kernel/setup_mm.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/arch/m68k/kernel/setup_mm.c b/arch/m68k/kernel/setup_mm.c index 3a2bb2e8fdad4..fbff1cea62caa 100644 --- a/arch/m68k/kernel/setup_mm.c +++ b/arch/m68k/kernel/setup_mm.c @@ -326,16 +326,16 @@ void __init setup_arch(char **cmdline_p) panic("No configuration setup"); } -#ifdef CONFIG_BLK_DEV_INITRD - if (m68k_ramdisk.size) { + if (IS_ENABLED(CONFIG_BLK_DEV_INITRD) && m68k_ramdisk.size) memblock_reserve(m68k_ramdisk.addr, m68k_ramdisk.size); + + paging_init(); + + if (IS_ENABLED(CONFIG_BLK_DEV_INITRD) && m68k_ramdisk.size) { initrd_start = (unsigned long)phys_to_virt(m68k_ramdisk.addr); initrd_end = initrd_start + m68k_ramdisk.size; pr_info("initrd: %08lx - %08lx\n", initrd_start, initrd_end); } -#endif - - paging_init(); #ifdef CONFIG_NATFEAT nf_init(); From e36a82bebbf7da814530d5a179bef9df5934b717 Mon Sep 17 00:00:00 2001 From: Michael Schmitz Date: Wed, 1 Mar 2023 15:11:07 +1300 Subject: [PATCH 720/765] m68k: Only force 030 bus error if PC not in exception table __get_kernel_nofault() does copy data in supervisor mode when forcing a task backtrace log through /proc/sysrq_trigger. This is expected cause a bus error exception on e.g. NULL pointer dereferencing when logging a kernel task has no workqueue associated. This bus error ought to be ignored. Our 030 bus error handler is ill equipped to deal with this: Whenever ssw indicates a kernel mode access on a data fault, we don't even attempt to handle the fault and instead always send a SEGV signal (or panic). As a result, the check for exception handling at the fault PC (buried in send_sig_fault() which gets called from do_page_fault() eventually) is never used. In contrast, both 040 and 060 access error handlers do not care whether a fault happened on supervisor mode access, and will call do_page_fault() on those, ultimately honoring the exception table. Add a check in bus_error030 to call do_page_fault() in case we do have an entry for the fault PC in our exception table. I had attempted a fix for this earlier in 2019 that did rely on testing pagefault_disabled() (see link below) to achieve the same thing, but this patch should be more generic. Tested on 030 Atari Falcon. Reported-by: Eero Tamminen Link: https://lore.kernel.org/r/alpine.LNX.2.21.1904091023540.25@nippy.intranet Link: https://lore.kernel.org/r/63130691-1984-c423-c1f2-73bfd8d3dcd3@gmail.com Signed-off-by: Michael Schmitz Reviewed-by: Geert Uytterhoeven Link: https://lore.kernel.org/r/20230301021107.26307-1-schmitzmic@gmail.com Signed-off-by: Geert Uytterhoeven --- arch/m68k/kernel/traps.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/m68k/kernel/traps.c b/arch/m68k/kernel/traps.c index 5c8cba0efc63e..a700807c9b6d9 100644 --- a/arch/m68k/kernel/traps.c +++ b/arch/m68k/kernel/traps.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include @@ -545,7 +546,8 @@ static inline void bus_error030 (struct frame *fp) errorcode |= 2; if (mmusr & (MMU_I | MMU_WP)) { - if (ssw & 4) { + /* We might have an exception table for this PC */ + if (ssw & 4 && !search_exception_tables(fp->ptregs.pc)) { pr_err("Data %s fault at %#010lx in %s (pc=%#lx)\n", ssw & RW ? "read" : "write", fp->un.fmtb.daddr, From 49854d3ccc55efd7e6873e0c39f360bdbe251c51 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 27 Feb 2023 19:32:19 +0100 Subject: [PATCH 721/765] udf: Fix lost writes in udf_adinicb_writepage() The patch converting udf_adinicb_writepage() to avoid manually kmapping the page used memcpy_to_page() however that copies in the wrong direction (effectively overwriting file data with the old contents). What we should be using is memcpy_from_page() to copy data from the page into the inode and then mark inode dirty to store the data. Fixes: 5cfc45321a6d ("udf: Convert udf_adinicb_writepage() to memcpy_to_page()") Signed-off-by: Jan Kara --- fs/udf/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/udf/inode.c b/fs/udf/inode.c index f7a9607c2b957..facaf3a206251 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -193,7 +193,7 @@ static int udf_adinicb_writepage(struct folio *folio, struct udf_inode_info *iinfo = UDF_I(inode); BUG_ON(!PageLocked(page)); - memcpy_to_page(page, 0, iinfo->i_data + iinfo->i_lenEAttr, + memcpy_from_page(iinfo->i_data + iinfo->i_lenEAttr, page, 0, i_size_read(inode)); unlock_page(page); mark_inode_dirty(inode); From cecb1f06541e12ec68805dbddb2013ee720dfe3d Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 28 Feb 2023 12:00:25 +0100 Subject: [PATCH 722/765] udf: Fix reading of in-ICB files After merging address space operations of normal and in-ICB files, readahead could get called for in-ICB files which resulted in udf_get_block() being called for these files. udf_get_block() is not prepared to be called for in-ICB files and ends up returning garbage results as it interprets file data as extent list. Fix the problem by skipping readahead for in-ICB files. Fixes: 37a8a39f7ad3 ("udf: Switch to single address_space_operations") Signed-off-by: Jan Kara --- fs/udf/inode.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/fs/udf/inode.c b/fs/udf/inode.c index facaf3a206251..0cb7d8fba2c85 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -241,6 +241,15 @@ static int udf_read_folio(struct file *file, struct folio *folio) static void udf_readahead(struct readahead_control *rac) { + struct udf_inode_info *iinfo = UDF_I(rac->mapping->host); + + /* + * No readahead needed for in-ICB files and udf_get_block() would get + * confused for such file anyway. + */ + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) + return; + mpage_readahead(rac, udf_get_block); } From 63bceed808c5cafbac4e20b5a40012a0ec6c6529 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 28 Feb 2023 12:11:38 +0100 Subject: [PATCH 723/765] udf: Warn if block mapping is done for in-ICB files Now that address space operations are merge dfor in-ICB and normal files, it is more likely some code mistakenly tries to map blocks for in-ICB files. WARN and return error instead of silently returning garbage. Signed-off-by: Jan Kara --- fs/udf/inode.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/fs/udf/inode.c b/fs/udf/inode.c index 0cb7d8fba2c85..2210e5eb1ea06 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -416,6 +416,9 @@ static int udf_map_block(struct inode *inode, struct udf_map_rq *map) int err; struct udf_inode_info *iinfo = UDF_I(inode); + if (WARN_ON_ONCE(iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB)) + return -EFSCORRUPTED; + map->oflags = 0; if (!(map->iflags & UDF_MAP_CREATE)) { struct kernel_lb_addr eloc; From 32db18d606c9a6eac7ac8bb12b4bf0e2eb1d5d14 Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Tue, 28 Feb 2023 14:45:22 +0700 Subject: [PATCH 724/765] bpf, doc: Do not link to docs.kernel.org for kselftest link The question on how to run BPF selftests have a reference link to kernel selftest documentation (Documentation/dev-tools/kselftest.rst). However, it uses external link to the documentation at kernel.org/docs (aka docs.kernel.org) instead, which requires Internet access. Fix this and replace the link with internal linking, by using :doc: directive while keeping the anchor text. Fixes: b7a27c3aafa252 ("bpf, doc: howto use/run the BPF selftests") Signed-off-by: Bagas Sanjaya Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20230228074523.11493-2-bagasdotme@gmail.com --- Documentation/bpf/bpf_devel_QA.rst | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/Documentation/bpf/bpf_devel_QA.rst b/Documentation/bpf/bpf_devel_QA.rst index 715f7321020f2..eb087c599ffa4 100644 --- a/Documentation/bpf/bpf_devel_QA.rst +++ b/Documentation/bpf/bpf_devel_QA.rst @@ -461,8 +461,8 @@ needed:: $ sudo make run_tests -See the kernels selftest `Documentation/dev-tools/kselftest.rst`_ -document for further documentation. +See :doc:`kernel selftest documentation ` +for details. To maximize the number of tests passing, the .config of the kernel under test should match the config file fragment in @@ -688,7 +688,5 @@ when: .. _netdev-FAQ: Documentation/process/maintainer-netdev.rst .. _selftests: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/tools/testing/selftests/bpf/ -.. _Documentation/dev-tools/kselftest.rst: - https://www.kernel.org/doc/html/latest/dev-tools/kselftest.html Happy BPF hacking! From b7abcd9c656b982a99e18f795bb1cf81f84b656d Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Tue, 28 Feb 2023 14:45:23 +0700 Subject: [PATCH 725/765] bpf, doc: Link to submitting-patches.rst for general patch submission info The link for patch submission information in general refers to index page for "Working with the kernel development community" section of kernel docs, whereas the link should have been Documentation/process/submitting-patches.rst instead. Fix it by replacing the index target with the appropriate doc. Fixes: 542228384888f5 ("bpf, doc: convert bpf_devel_QA.rst to use RST formatting") Signed-off-by: Bagas Sanjaya Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20230228074523.11493-3-bagasdotme@gmail.com --- Documentation/bpf/bpf_devel_QA.rst | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/Documentation/bpf/bpf_devel_QA.rst b/Documentation/bpf/bpf_devel_QA.rst index eb087c599ffa4..b421d94dc9f21 100644 --- a/Documentation/bpf/bpf_devel_QA.rst +++ b/Documentation/bpf/bpf_devel_QA.rst @@ -7,8 +7,8 @@ workflows related to reporting bugs, submitting patches, and queueing patches for stable kernels. For general information about submitting patches, please refer to -`Documentation/process/`_. This document only describes additional specifics -related to BPF. +Documentation/process/submitting-patches.rst. This document only describes +additional specifics related to BPF. .. contents:: :local: @@ -684,7 +684,6 @@ when: .. Links -.. _Documentation/process/: https://www.kernel.org/doc/html/latest/process/ .. _netdev-FAQ: Documentation/process/maintainer-netdev.rst .. _selftests: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/tools/testing/selftests/bpf/ From 80c16b2b121fbc3380dbffa9bab7559acbaaa2ed Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 6 Mar 2023 17:22:04 +0200 Subject: [PATCH 726/765] cpumask: Fix typo nr_cpumask_size --> nr_cpumask_bits The never used nr_cpumask_size is just a typo, hence use existing redefinition that's called nr_cpumask_bits. Signed-off-by: Andy Shevchenko Signed-off-by: Linus Torvalds --- include/linux/cpumask.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h index 8fbe76607965e..ce8eb7ef21072 100644 --- a/include/linux/cpumask.h +++ b/include/linux/cpumask.h @@ -66,7 +66,7 @@ static inline void set_nr_cpu_ids(unsigned int nr) * * Finally, some operations just want the exact limit, either because * they set bits or just don't have any faster fixed-sized versions. We - * call this just 'nr_cpumask_size'. + * call this just 'nr_cpumask_bits'. * * Note that these optional constants are always guaranteed to be at * least as big as 'nr_cpu_ids' itself is, and all our cpumask From 294635a8165a31408a8b3a24f9c74849ca3d8701 Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Fri, 24 Feb 2023 17:36:07 +0100 Subject: [PATCH 727/765] bpf, test_run: fix &xdp_frame misplacement for LIVE_FRAMES MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit &xdp_buff and &xdp_frame are bound in a way that xdp_buff->data_hard_start == xdp_frame It's always the case and e.g. xdp_convert_buff_to_frame() relies on this. IOW, the following: for (u32 i = 0; i < 0xdead; i++) { xdpf = xdp_convert_buff_to_frame(&xdp); xdp_convert_frame_to_buff(xdpf, &xdp); } shouldn't ever modify @xdpf's contents or the pointer itself. However, "live packet" code wrongly treats &xdp_frame as part of its context placed *before* the data_hard_start. With such flow, data_hard_start is sizeof(*xdpf) off to the right and no longer points to the XDP frame. Instead of replacing `sizeof(ctx)` with `offsetof(ctx, xdpf)` in several places and praying that there are no more miscalcs left somewhere in the code, unionize ::frm with ::data in a flex array, so that both starts pointing to the actual data_hard_start and the XDP frame actually starts being a part of it, i.e. a part of the headroom, not the context. A nice side effect is that the maximum frame size for this mode gets increased by 40 bytes, as xdp_buff::frame_sz includes everything from data_hard_start (-> includes xdpf already) to the end of XDP/skb shared info. Also update %MAX_PKT_SIZE accordingly in the selftests code. Leave it hardcoded for 64 bit && 4k pages, it can be made more flexible later on. Minor: align `&head->data` with how `head->frm` is assigned for consistency. Minor #2: rename 'frm' to 'frame' in &xdp_page_head while at it for clarity. (was found while testing XDP traffic generator on ice, which calls xdp_convert_frame_to_buff() for each XDP frame) Fixes: b530e9e1063e ("bpf: Add "live packet" mode for XDP in BPF_PROG_RUN") Acked-by: Toke Høiland-Jørgensen Signed-off-by: Alexander Lobakin Link: https://lore.kernel.org/r/20230224163607.2994755-1-aleksander.lobakin@intel.com Signed-off-by: Martin KaFai Lau --- net/bpf/test_run.c | 19 +++++++++++++------ .../bpf/prog_tests/xdp_do_redirect.c | 7 ++++--- 2 files changed, 17 insertions(+), 9 deletions(-) diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c index 6f3d654b3339c..f81b24320a36e 100644 --- a/net/bpf/test_run.c +++ b/net/bpf/test_run.c @@ -97,8 +97,11 @@ static bool bpf_test_timer_continue(struct bpf_test_timer *t, int iterations, struct xdp_page_head { struct xdp_buff orig_ctx; struct xdp_buff ctx; - struct xdp_frame frm; - u8 data[]; + union { + /* ::data_hard_start starts here */ + DECLARE_FLEX_ARRAY(struct xdp_frame, frame); + DECLARE_FLEX_ARRAY(u8, data); + }; }; struct xdp_test_data { @@ -113,6 +116,10 @@ struct xdp_test_data { u32 frame_cnt; }; +/* tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c:%MAX_PKT_SIZE + * must be updated accordingly this gets changed, otherwise BPF selftests + * will fail. + */ #define TEST_XDP_FRAME_SIZE (PAGE_SIZE - sizeof(struct xdp_page_head)) #define TEST_XDP_MAX_BATCH 256 @@ -132,8 +139,8 @@ static void xdp_test_run_init_page(struct page *page, void *arg) headroom -= meta_len; new_ctx = &head->ctx; - frm = &head->frm; - data = &head->data; + frm = head->frame; + data = head->data; memcpy(data + headroom, orig_ctx->data_meta, frm_len); xdp_init_buff(new_ctx, TEST_XDP_FRAME_SIZE, &xdp->rxq); @@ -223,7 +230,7 @@ static void reset_ctx(struct xdp_page_head *head) head->ctx.data = head->orig_ctx.data; head->ctx.data_meta = head->orig_ctx.data_meta; head->ctx.data_end = head->orig_ctx.data_end; - xdp_update_frame_from_buff(&head->ctx, &head->frm); + xdp_update_frame_from_buff(&head->ctx, head->frame); } static int xdp_recv_frames(struct xdp_frame **frames, int nframes, @@ -285,7 +292,7 @@ static int xdp_test_run_batch(struct xdp_test_data *xdp, struct bpf_prog *prog, head = phys_to_virt(page_to_phys(page)); reset_ctx(head); ctx = &head->ctx; - frm = &head->frm; + frm = head->frame; xdp->frame_cnt++; act = bpf_prog_run_xdp(prog, ctx); diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c b/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c index 2666c84dbd014..7271a18ab3e22 100644 --- a/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c +++ b/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c @@ -65,12 +65,13 @@ static int attach_tc_prog(struct bpf_tc_hook *hook, int fd) } /* The maximum permissible size is: PAGE_SIZE - sizeof(struct xdp_page_head) - - * sizeof(struct skb_shared_info) - XDP_PACKET_HEADROOM = 3368 bytes + * SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) - XDP_PACKET_HEADROOM = + * 3408 bytes for 64-byte cacheline and 3216 for 256-byte one. */ #if defined(__s390x__) -#define MAX_PKT_SIZE 3176 +#define MAX_PKT_SIZE 3216 #else -#define MAX_PKT_SIZE 3368 +#define MAX_PKT_SIZE 3408 #endif static void test_max_pkt_size(int fd) { From 9b459804ff9973e173fabafba2a1319f771e85fa Mon Sep 17 00:00:00 2001 From: Lorenz Bauer Date: Mon, 6 Mar 2023 11:21:37 +0000 Subject: [PATCH 728/765] btf: fix resolving BTF_KIND_VAR after ARRAY, STRUCT, UNION, PTR btf_datasec_resolve contains a bug that causes the following BTF to fail loading: [1] DATASEC a size=2 vlen=2 type_id=4 offset=0 size=1 type_id=7 offset=1 size=1 [2] INT (anon) size=1 bits_offset=0 nr_bits=8 encoding=(none) [3] PTR (anon) type_id=2 [4] VAR a type_id=3 linkage=0 [5] INT (anon) size=1 bits_offset=0 nr_bits=8 encoding=(none) [6] TYPEDEF td type_id=5 [7] VAR b type_id=6 linkage=0 This error message is printed during btf_check_all_types: [1] DATASEC a size=2 vlen=2 type_id=7 offset=1 size=1 Invalid type By tracing btf_*_resolve we can pinpoint the problem: btf_datasec_resolve(depth: 1, type_id: 1, mode: RESOLVE_TBD) = 0 btf_var_resolve(depth: 2, type_id: 4, mode: RESOLVE_TBD) = 0 btf_ptr_resolve(depth: 3, type_id: 3, mode: RESOLVE_PTR) = 0 btf_var_resolve(depth: 2, type_id: 4, mode: RESOLVE_PTR) = 0 btf_datasec_resolve(depth: 1, type_id: 1, mode: RESOLVE_PTR) = -22 The last invocation of btf_datasec_resolve should invoke btf_var_resolve by means of env_stack_push, instead it returns EINVAL. The reason is that env_stack_push is never executed for the second VAR. if (!env_type_is_resolve_sink(env, var_type) && !env_type_is_resolved(env, var_type_id)) { env_stack_set_next_member(env, i + 1); return env_stack_push(env, var_type, var_type_id); } env_type_is_resolve_sink() changes its behaviour based on resolve_mode. For RESOLVE_PTR, we can simplify the if condition to the following: (btf_type_is_modifier() || btf_type_is_ptr) && !env_type_is_resolved() Since we're dealing with a VAR the clause evaluates to false. This is not sufficient to trigger the bug however. The log output and EINVAL are only generated if btf_type_id_size() fails. if (!btf_type_id_size(btf, &type_id, &type_size)) { btf_verifier_log_vsi(env, v->t, vsi, "Invalid type"); return -EINVAL; } Most types are sized, so for example a VAR referring to an INT is not a problem. The bug is only triggered if a VAR points at a modifier. Since we skipped btf_var_resolve that modifier was also never resolved, which means that btf_resolved_type_id returns 0 aka VOID for the modifier. This in turn causes btf_type_id_size to return NULL, triggering EINVAL. To summarise, the following conditions are necessary: - VAR pointing at PTR, STRUCT, UNION or ARRAY - Followed by a VAR pointing at TYPEDEF, VOLATILE, CONST, RESTRICT or TYPE_TAG The fix is to reset resolve_mode to RESOLVE_TBD before attempting to resolve a VAR from a DATASEC. Fixes: 1dc92851849c ("bpf: kernel side support for BTF Var and DataSec") Signed-off-by: Lorenz Bauer Link: https://lore.kernel.org/r/20230306112138.155352-2-lmb@isovalent.com Signed-off-by: Martin KaFai Lau --- kernel/bpf/btf.c | 1 + 1 file changed, 1 insertion(+) diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c index fa22ec79ac0e1..73780748404c2 100644 --- a/kernel/bpf/btf.c +++ b/kernel/bpf/btf.c @@ -4569,6 +4569,7 @@ static int btf_datasec_resolve(struct btf_verifier_env *env, struct btf *btf = env->btf; u16 i; + env->resolve_mode = RESOLVE_TBD; for_each_vsi_from(i, v->next_member, v->t, vsi) { u32 var_type_id = vsi->type, type_id, type_size = 0; const struct btf_type *var_type = btf_type_by_id(env->btf, From dfdd608c3b365f0fd49d7e13911ebcde06b9865b Mon Sep 17 00:00:00 2001 From: Lorenz Bauer Date: Mon, 6 Mar 2023 11:21:38 +0000 Subject: [PATCH 729/765] selftests/bpf: check that modifier resolves after pointer Add a regression test that ensures that a VAR pointing at a modifier which follows a PTR (or STRUCT or ARRAY) is resolved correctly by the datasec validator. Signed-off-by: Lorenz Bauer Link: https://lore.kernel.org/r/20230306112138.155352-3-lmb@isovalent.com Signed-off-by: Martin KaFai Lau --- tools/testing/selftests/bpf/prog_tests/btf.c | 28 ++++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/tools/testing/selftests/bpf/prog_tests/btf.c b/tools/testing/selftests/bpf/prog_tests/btf.c index cbb600be943d3..210d643fda6c7 100644 --- a/tools/testing/selftests/bpf/prog_tests/btf.c +++ b/tools/testing/selftests/bpf/prog_tests/btf.c @@ -879,6 +879,34 @@ static struct btf_raw_test raw_tests[] = { .btf_load_err = true, .err_str = "Invalid elem", }, +{ + .descr = "var after datasec, ptr followed by modifier", + .raw_types = { + /* .bss section */ /* [1] */ + BTF_TYPE_ENC(NAME_TBD, BTF_INFO_ENC(BTF_KIND_DATASEC, 0, 2), + sizeof(void*)+4), + BTF_VAR_SECINFO_ENC(4, 0, sizeof(void*)), + BTF_VAR_SECINFO_ENC(6, sizeof(void*), 4), + /* int */ /* [2] */ + BTF_TYPE_INT_ENC(0, BTF_INT_SIGNED, 0, 32, 4), + /* int* */ /* [3] */ + BTF_TYPE_ENC(0, BTF_INFO_ENC(BTF_KIND_PTR, 0, 0), 2), + BTF_VAR_ENC(NAME_TBD, 3, 0), /* [4] */ + /* const int */ /* [5] */ + BTF_TYPE_ENC(0, BTF_INFO_ENC(BTF_KIND_CONST, 0, 0), 2), + BTF_VAR_ENC(NAME_TBD, 5, 0), /* [6] */ + BTF_END_RAW, + }, + .str_sec = "\0a\0b\0c\0", + .str_sec_size = sizeof("\0a\0b\0c\0"), + .map_type = BPF_MAP_TYPE_ARRAY, + .map_name = ".bss", + .key_size = sizeof(int), + .value_size = sizeof(void*)+4, + .key_type_id = 0, + .value_type_id = 1, + .max_entries = 1, +}, /* Test member exceeds the size of struct. * * struct A { From 8ca09d5fa3549d142c2080a72a4c70ce389163cd Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 6 Mar 2023 12:15:13 -0800 Subject: [PATCH 730/765] cpumask: fix incorrect cpumask scanning result checks It turns out that commit 596ff4a09b89 ("cpumask: re-introduce constant-sized cpumask optimizations") exposed a number of cases of drivers not checking the result of "cpumask_next()" and friends correctly. The documented correct check for "no more cpus in the cpumask" is to check for the result being equal or larger than the number of possible CPU ids, exactly _because_ we've always done those constant-sized cpumask scans using a widened type before. So the return value of a cpumask scan should be checked with if (cpu >= nr_cpu_ids) ... because the cpumask scan did not necessarily stop exactly *at* that maximum CPU id. But a few cases ended up instead using checks like if (cpu == nr_cpumask_bits) ... which used that internal "widened" number of bits. And that used to work pretty much by accident (ok, in this case "by accident" is simply because it matched the historical internal implementation of the cpumask scanning, so it was more of a "intentionally using implementation details rather than an accident"). But the extended constant-sized optimizations then did that internal implementation differently, and now that code that did things wrong but matched the old implementation no longer worked at all. Which then causes subsequent odd problems due to using what ends up being an invalid CPU ID. Most of these cases require either unusual hardware or special uses to hit, but the random.c one triggers quite easily. All you really need is to have a sufficiently small CONFIG_NR_CPUS value for the bit scanning optimization to be triggered, but not enough CPUs to then actually fill that widened cpumask. At that point, the cpumask scanning will return the NR_CPUS constant, which is _not_ the same as nr_cpumask_bits. This just does the mindless fix with sed -i 's/== nr_cpumask_bits/>= nr_cpu_ids/' to fix the incorrect uses. The ones in the SCSI lpfc driver in particular could probably be fixed more cleanly by just removing that repeated pattern entirely, but I am not emptionally invested enough in that driver to care. Reported-and-tested-by: Guenter Roeck Link: https://lore.kernel.org/lkml/481b19b5-83a0-4793-b4fd-194ad7b978c3@roeck-us.net/ Reported-and-tested-by: Geert Uytterhoeven Link: https://lore.kernel.org/lkml/CAMuHMdUKo_Sf7TjKzcNDa8Ve+6QrK+P8nSQrSQ=6LTRmcBKNww@mail.gmail.com/ Reported-by: Vernon Yang Link: https://lore.kernel.org/lkml/20230306160651.2016767-1-vernon2gm@gmail.com/ Cc: Yury Norov Cc: Jason A. Donenfeld Signed-off-by: Linus Torvalds --- arch/powerpc/xmon/xmon.c | 2 +- drivers/char/random.c | 2 +- drivers/net/wireguard/queueing.h | 2 +- drivers/scsi/lpfc/lpfc_init.c | 14 +++++++------- 4 files changed, 10 insertions(+), 10 deletions(-) diff --git a/arch/powerpc/xmon/xmon.c b/arch/powerpc/xmon/xmon.c index 73c620c2a3a16..e753a6bd48881 100644 --- a/arch/powerpc/xmon/xmon.c +++ b/arch/powerpc/xmon/xmon.c @@ -1275,7 +1275,7 @@ static int xmon_batch_next_cpu(void) while (!cpumask_empty(&xmon_batch_cpus)) { cpu = cpumask_next_wrap(smp_processor_id(), &xmon_batch_cpus, xmon_batch_start_cpu, true); - if (cpu == nr_cpumask_bits) + if (cpu >= nr_cpu_ids) break; if (xmon_batch_start_cpu == -1) xmon_batch_start_cpu = cpu; diff --git a/drivers/char/random.c b/drivers/char/random.c index ce3ccd172cc86..253f2ddb89130 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -1311,7 +1311,7 @@ static void __cold try_to_generate_entropy(void) /* Basic CPU round-robin, which avoids the current CPU. */ do { cpu = cpumask_next(cpu, &timer_cpus); - if (cpu == nr_cpumask_bits) + if (cpu >= nr_cpu_ids) cpu = cpumask_first(&timer_cpus); } while (cpu == smp_processor_id() && num_cpus > 1); diff --git a/drivers/net/wireguard/queueing.h b/drivers/net/wireguard/queueing.h index 583adb37ee1e3..125284b346a77 100644 --- a/drivers/net/wireguard/queueing.h +++ b/drivers/net/wireguard/queueing.h @@ -106,7 +106,7 @@ static inline int wg_cpumask_choose_online(int *stored_cpu, unsigned int id) { unsigned int cpu = *stored_cpu, cpu_index, i; - if (unlikely(cpu == nr_cpumask_bits || + if (unlikely(cpu >= nr_cpu_ids || !cpumask_test_cpu(cpu, cpu_online_mask))) { cpu_index = id % cpumask_weight(cpu_online_mask); cpu = cpumask_first(cpu_online_mask); diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c index 61958a24a43d9..73b544bfbb2e6 100644 --- a/drivers/scsi/lpfc/lpfc_init.c +++ b/drivers/scsi/lpfc/lpfc_init.c @@ -12563,7 +12563,7 @@ lpfc_cpu_affinity_check(struct lpfc_hba *phba, int vectors) goto found_same; new_cpu = cpumask_next( new_cpu, cpu_present_mask); - if (new_cpu == nr_cpumask_bits) + if (new_cpu >= nr_cpu_ids) new_cpu = first_cpu; } /* At this point, we leave the CPU as unassigned */ @@ -12577,7 +12577,7 @@ lpfc_cpu_affinity_check(struct lpfc_hba *phba, int vectors) * selecting the same IRQ. */ start_cpu = cpumask_next(new_cpu, cpu_present_mask); - if (start_cpu == nr_cpumask_bits) + if (start_cpu >= nr_cpu_ids) start_cpu = first_cpu; lpfc_printf_log(phba, KERN_INFO, LOG_INIT, @@ -12613,7 +12613,7 @@ lpfc_cpu_affinity_check(struct lpfc_hba *phba, int vectors) goto found_any; new_cpu = cpumask_next( new_cpu, cpu_present_mask); - if (new_cpu == nr_cpumask_bits) + if (new_cpu >= nr_cpu_ids) new_cpu = first_cpu; } /* We should never leave an entry unassigned */ @@ -12631,7 +12631,7 @@ lpfc_cpu_affinity_check(struct lpfc_hba *phba, int vectors) * selecting the same IRQ. */ start_cpu = cpumask_next(new_cpu, cpu_present_mask); - if (start_cpu == nr_cpumask_bits) + if (start_cpu >= nr_cpu_ids) start_cpu = first_cpu; lpfc_printf_log(phba, KERN_INFO, LOG_INIT, @@ -12704,7 +12704,7 @@ lpfc_cpu_affinity_check(struct lpfc_hba *phba, int vectors) goto found_hdwq; } new_cpu = cpumask_next(new_cpu, cpu_present_mask); - if (new_cpu == nr_cpumask_bits) + if (new_cpu >= nr_cpu_ids) new_cpu = first_cpu; } @@ -12719,7 +12719,7 @@ lpfc_cpu_affinity_check(struct lpfc_hba *phba, int vectors) goto found_hdwq; new_cpu = cpumask_next(new_cpu, cpu_present_mask); - if (new_cpu == nr_cpumask_bits) + if (new_cpu >= nr_cpu_ids) new_cpu = first_cpu; } @@ -12730,7 +12730,7 @@ lpfc_cpu_affinity_check(struct lpfc_hba *phba, int vectors) found_hdwq: /* We found an available entry, copy the IRQ info */ start_cpu = cpumask_next(new_cpu, cpu_present_mask); - if (start_cpu == nr_cpumask_bits) + if (start_cpu >= nr_cpu_ids) start_cpu = first_cpu; cpup->hdwq = new_cpup->hdwq; logit: From 58aac3a2ef414fea6d7fdf823ea177744a087d13 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Sat, 4 Mar 2023 11:52:44 +0100 Subject: [PATCH 731/765] net: phy: smsc: fix link up detection in forced irq mode Currently link up can't be detected in forced mode if polling isn't used. Only link up interrupt source we have is aneg complete which isn't applicable in forced mode. Therefore we have to use energy-on as link up indicator. Fixes: 7365494550f6 ("net: phy: smsc: skip ENERGYON interrupt if disabled") Signed-off-by: Heiner Kallweit Signed-off-by: David S. Miller --- drivers/net/phy/smsc.c | 14 +++----------- 1 file changed, 3 insertions(+), 11 deletions(-) diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c index ac7481ce2fc16..00d9eff91dcfa 100644 --- a/drivers/net/phy/smsc.c +++ b/drivers/net/phy/smsc.c @@ -44,7 +44,6 @@ static struct smsc_hw_stat smsc_hw_stats[] = { }; struct smsc_phy_priv { - u16 intmask; bool energy_enable; }; @@ -57,7 +56,6 @@ static int smsc_phy_ack_interrupt(struct phy_device *phydev) static int smsc_phy_config_intr(struct phy_device *phydev) { - struct smsc_phy_priv *priv = phydev->priv; int rc; if (phydev->interrupts == PHY_INTERRUPT_ENABLED) { @@ -65,14 +63,9 @@ static int smsc_phy_config_intr(struct phy_device *phydev) if (rc) return rc; - priv->intmask = MII_LAN83C185_ISF_INT4 | MII_LAN83C185_ISF_INT6; - if (priv->energy_enable) - priv->intmask |= MII_LAN83C185_ISF_INT7; - - rc = phy_write(phydev, MII_LAN83C185_IM, priv->intmask); + rc = phy_write(phydev, MII_LAN83C185_IM, + MII_LAN83C185_ISF_INT_PHYLIB_EVENTS); } else { - priv->intmask = 0; - rc = phy_write(phydev, MII_LAN83C185_IM, 0); if (rc) return rc; @@ -85,7 +78,6 @@ static int smsc_phy_config_intr(struct phy_device *phydev) static irqreturn_t smsc_phy_handle_interrupt(struct phy_device *phydev) { - struct smsc_phy_priv *priv = phydev->priv; int irq_status; irq_status = phy_read(phydev, MII_LAN83C185_ISF); @@ -96,7 +88,7 @@ static irqreturn_t smsc_phy_handle_interrupt(struct phy_device *phydev) return IRQ_NONE; } - if (!(irq_status & priv->intmask)) + if (!(irq_status & MII_LAN83C185_ISF_INT_PHYLIB_EVENTS)) return IRQ_NONE; phy_trigger_machine(phydev); From 193250ace270fecd586dd2d0dfbd9cbd2ade977f Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Sat, 4 Mar 2023 13:43:20 +0000 Subject: [PATCH 732/765] net: ethernet: mtk_eth_soc: fix RX data corruption issue MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix data corruption issue with SerDes connected PHYs operating at 1.25 Gbps speed where we could previously observe about 30% packet loss while the bad packet counter was increasing. As almost all boards with MediaTek MT7622 or MT7986 use either the MT7531 switch IC operating at 3.125Gbps SerDes rate or single-port PHYs using rate-adaptation to 2500Base-X mode, this issue only got exposed now when we started trying to use SFP modules operating with 1.25 Gbps with the BananaPi R3 board. The fix is to set bit 12 which disables the RX FIFO clear function when setting up MAC MCR, MediaTek SDK did the same change stating: "If without this patch, kernel might receive invalid packets that are corrupted by GMAC."[1] [1]: https://git01.mediatek.com/plugins/gitiles/openwrt/feeds/mtk-openwrt-feeds/+/d8a2975939a12686c4a95c40db21efdc3f821f63 Fixes: 42c03844e93d ("net-next: mediatek: add support for MediaTek MT7622 SoC") Tested-by: Bjørn Mork Signed-off-by: Daniel Golle Reviewed-by: Vladimir Oltean Reviewed-by: Florian Fainelli Link: https://lore.kernel.org/r/138da2735f92c8b6f8578ec2e5a794ee515b665f.1677937317.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 3 ++- drivers/net/ethernet/mediatek/mtk_eth_soc.h | 1 + 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 14be6ea51b889..3cb43623d3dbe 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -616,7 +616,8 @@ static int mtk_mac_finish(struct phylink_config *config, unsigned int mode, mcr_cur = mtk_r32(mac->hw, MTK_MAC_MCR(mac->id)); mcr_new = mcr_cur; mcr_new |= MAC_MCR_IPG_CFG | MAC_MCR_FORCE_MODE | - MAC_MCR_BACKOFF_EN | MAC_MCR_BACKPR_EN | MAC_MCR_FORCE_LINK; + MAC_MCR_BACKOFF_EN | MAC_MCR_BACKPR_EN | MAC_MCR_FORCE_LINK | + MAC_MCR_RX_FIFO_CLR_DIS; /* Only update control register when needed! */ if (mcr_new != mcr_cur) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h index afc9d52e79bfc..b65de174c3d9b 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h @@ -397,6 +397,7 @@ #define MAC_MCR_FORCE_MODE BIT(15) #define MAC_MCR_TX_EN BIT(14) #define MAC_MCR_RX_EN BIT(13) +#define MAC_MCR_RX_FIFO_CLR_DIS BIT(12) #define MAC_MCR_BACKOFF_EN BIT(9) #define MAC_MCR_BACKPR_EN BIT(8) #define MAC_MCR_FORCE_RX_FC BIT(5) From e539a105f947b9db470fec39fe91d85fe737a432 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Sat, 4 Mar 2023 11:26:10 -0800 Subject: [PATCH 733/765] net: tls: fix device-offloaded sendpage straddling records Adrien reports that incorrect data is transmitted when a single page straddles multiple records. We would transmit the same data in all iterations of the loop. Reported-by: Adrien Moulin Link: https://lore.kernel.org/all/61481278.42813558.1677845235112.JavaMail.zimbra@corp.free.fr Fixes: c1318b39c7d3 ("tls: Add opt-in zerocopy mode of sendfile()") Tested-by: Adrien Moulin Reviewed-by: Tariq Toukan Acked-by: Maxim Mikityanskiy Link: https://lore.kernel.org/r/20230304192610.3818098-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/tls/tls_device.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/tls/tls_device.c b/net/tls/tls_device.c index 6c593788dc250..a7cc4f9faac28 100644 --- a/net/tls/tls_device.c +++ b/net/tls/tls_device.c @@ -508,6 +508,8 @@ static int tls_push_data(struct sock *sk, zc_pfrag.offset = iter_offset.offset; zc_pfrag.size = copy; tls_append_frag(record, &zc_pfrag, copy); + + iter_offset.offset += copy; } else if (copy) { copy = min_t(size_t, copy, pfrag->size - pfrag->offset); From c77737b736ceb50fdf150434347dbd81ec76dbb1 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 7 Mar 2023 05:22:54 +0000 Subject: [PATCH 734/765] netfilter: conntrack: adopt safer max chain length Customers using GKE 1.25 and 1.26 are facing conntrack issues root caused to commit c9c3b6811f74 ("netfilter: conntrack: make max chain length random"). Even if we assume Uniform Hashing, a bucket often reachs 8 chained items while the load factor of the hash table is smaller than 0.5 With a limit of 16, we reach load factors of 3. With a limit of 32, we reach load factors of 11. With a limit of 40, we reach load factors of 15. With a limit of 50, we reach load factors of 24. This patch changes MIN_CHAINLEN to 50, to minimize risks. Ideally, we could in the future add a cushion based on expected load factor (2 * nf_conntrack_max / nf_conntrack_buckets), because some setups might expect unusual values. Fixes: c9c3b6811f74 ("netfilter: conntrack: make max chain length random") Signed-off-by: Eric Dumazet Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 7250082e7de56..c6a6a6099b4e2 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -96,8 +96,8 @@ static DEFINE_MUTEX(nf_conntrack_mutex); #define GC_SCAN_MAX_DURATION msecs_to_jiffies(10) #define GC_SCAN_EXPIRED_MAX (64000u / HZ) -#define MIN_CHAINLEN 8u -#define MAX_CHAINLEN (32u - MIN_CHAINLEN) +#define MIN_CHAINLEN 50u +#define MAX_CHAINLEN (80u - MIN_CHAINLEN) static struct conntrack_gc_work conntrack_gc_work; From 24efcdf03d85bb73df0ba99f69c8d238e7ada0e5 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 14 Feb 2023 16:25:07 +0100 Subject: [PATCH 735/765] platform/x86/amd: pmc: remove CONFIG_SUSPEND checks The amd_pmc_write_stb() function was previously hidden in an ifdef to avoid a warning when CONFIG_SUSPEND is disabled, but now there is an additional caller: drivers/platform/x86/amd/pmc.c: In function 'amd_pmc_stb_debugfs_open_v2': drivers/platform/x86/amd/pmc.c:256:8: error: implicit declaration of function 'amd_pmc_write_stb'; did you mean 'amd_pmc_read_stb'? [-Werror=implicit-function-declaration] 256 | ret = amd_pmc_write_stb(dev, AMD_PMC_STB_DUMMY_PC); | ^~~~~~~~~~~~~~~~~ | amd_pmc_read_stb There is now an easier way to handle this using DEFINE_SIMPLE_DEV_PM_OPS() to replace all the #ifdefs, letting gcc drop any of the unused functions silently. Fixes: b0d4bb973539 ("platform/x86/amd: pmc: Write dummy postcode into the STB DRAM") Signed-off-by: Arnd Bergmann Link: https://lore.kernel.org/r/20230214152512.806188-1-arnd@kernel.org Signed-off-by: Hans de Goede Reviewed-by: Hans de Goede --- drivers/platform/x86/amd/pmc.c | 30 +++++++++--------------------- 1 file changed, 9 insertions(+), 21 deletions(-) diff --git a/drivers/platform/x86/amd/pmc.c b/drivers/platform/x86/amd/pmc.c index ab05b9ee6655a..2edaae04a6912 100644 --- a/drivers/platform/x86/amd/pmc.c +++ b/drivers/platform/x86/amd/pmc.c @@ -171,9 +171,7 @@ MODULE_PARM_DESC(disable_workarounds, "Disable workarounds for platform bugs"); static struct amd_pmc_dev pmc; static int amd_pmc_send_cmd(struct amd_pmc_dev *dev, u32 arg, u32 *data, u8 msg, bool ret); static int amd_pmc_read_stb(struct amd_pmc_dev *dev, u32 *buf); -#ifdef CONFIG_SUSPEND static int amd_pmc_write_stb(struct amd_pmc_dev *dev, u32 data); -#endif static inline u32 amd_pmc_reg_read(struct amd_pmc_dev *dev, int reg_offset) { @@ -386,7 +384,6 @@ static int get_metrics_table(struct amd_pmc_dev *pdev, struct smu_metrics *table return 0; } -#ifdef CONFIG_SUSPEND static void amd_pmc_validate_deepest(struct amd_pmc_dev *pdev) { struct smu_metrics table; @@ -400,7 +397,6 @@ static void amd_pmc_validate_deepest(struct amd_pmc_dev *pdev) dev_dbg(pdev->dev, "Last suspend in deepest state for %lluus\n", table.timein_s0i3_lastcapture); } -#endif static int amd_pmc_get_smu_version(struct amd_pmc_dev *dev) { @@ -673,7 +669,6 @@ static int amd_pmc_send_cmd(struct amd_pmc_dev *dev, u32 arg, u32 *data, u8 msg, return rc; } -#ifdef CONFIG_SUSPEND static int amd_pmc_get_os_hint(struct amd_pmc_dev *dev) { switch (dev->cpu_id) { @@ -861,9 +856,7 @@ static int __maybe_unused amd_pmc_suspend_handler(struct device *dev) return 0; } -static SIMPLE_DEV_PM_OPS(amd_pmc_pm, amd_pmc_suspend_handler, NULL); - -#endif +static DEFINE_SIMPLE_DEV_PM_OPS(amd_pmc_pm, amd_pmc_suspend_handler, NULL); static const struct pci_device_id pmc_pci_ids[] = { { PCI_DEVICE(PCI_VENDOR_ID_AMD, AMD_CPU_ID_PS) }, @@ -905,7 +898,6 @@ static int amd_pmc_s2d_init(struct amd_pmc_dev *dev) return 0; } -#ifdef CONFIG_SUSPEND static int amd_pmc_write_stb(struct amd_pmc_dev *dev, u32 data) { int err; @@ -926,7 +918,6 @@ static int amd_pmc_write_stb(struct amd_pmc_dev *dev, u32 data) return 0; } -#endif static int amd_pmc_read_stb(struct amd_pmc_dev *dev, u32 *buf) { @@ -1017,11 +1008,11 @@ static int amd_pmc_probe(struct platform_device *pdev) } platform_set_drvdata(pdev, dev); -#ifdef CONFIG_SUSPEND - err = acpi_register_lps0_dev(&amd_pmc_s2idle_dev_ops); - if (err) - dev_warn(dev->dev, "failed to register LPS0 sleep handler, expect increased power consumption\n"); -#endif + if (IS_ENABLED(CONFIG_SUSPEND)) { + err = acpi_register_lps0_dev(&amd_pmc_s2idle_dev_ops); + if (err) + dev_warn(dev->dev, "failed to register LPS0 sleep handler, expect increased power consumption\n"); + } amd_pmc_dbgfs_register(dev); return 0; @@ -1035,9 +1026,8 @@ static int amd_pmc_remove(struct platform_device *pdev) { struct amd_pmc_dev *dev = platform_get_drvdata(pdev); -#ifdef CONFIG_SUSPEND - acpi_unregister_lps0_dev(&amd_pmc_s2idle_dev_ops); -#endif + if (IS_ENABLED(CONFIG_SUSPEND)) + acpi_unregister_lps0_dev(&amd_pmc_s2idle_dev_ops); amd_pmc_dbgfs_unregister(dev); pci_dev_put(dev->rdev); mutex_destroy(&dev->lock); @@ -1061,9 +1051,7 @@ static struct platform_driver amd_pmc_driver = { .name = "amd_pmc", .acpi_match_table = amd_pmc_acpi_ids, .dev_groups = pmc_groups, -#ifdef CONFIG_SUSPEND - .pm = &amd_pmc_pm, -#endif + .pm = pm_sleep_ptr(&amd_pmc_pm), }, .probe = amd_pmc_probe, .remove = amd_pmc_remove, From 001f61c4688f1dad18cfaa35ae364b35fb3cd66c Mon Sep 17 00:00:00 2001 From: Armin Wolf Date: Sat, 18 Feb 2023 12:53:17 +0100 Subject: [PATCH 736/765] platform/x86: dell-ddv: Fix cache invalidation on resume If one or both sensor buffers could not be initialized, either due to missing hardware support or due to some error during probing, the resume handler will encounter undefined behaviour when attempting to lock buffers then protected by an uninitialized or destroyed mutex. Fix this by introducing a "active" flag which is set during probe, and only invalidate buffers which where flaged as "active". Tested on a Dell Inspiron 3505. Fixes: 3b7eeff93d29 ("platform/x86: dell-ddv: Add hwmon support") Signed-off-by: Armin Wolf Link: https://lore.kernel.org/r/20230218115318.20662-1-W_Armin@gmx.de Signed-off-by: Hans de Goede Reviewed-by: Hans de Goede --- drivers/platform/x86/dell/dell-wmi-ddv.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/platform/x86/dell/dell-wmi-ddv.c b/drivers/platform/x86/dell/dell-wmi-ddv.c index d547c9d097256..eff4e9649faf2 100644 --- a/drivers/platform/x86/dell/dell-wmi-ddv.c +++ b/drivers/platform/x86/dell/dell-wmi-ddv.c @@ -96,6 +96,7 @@ struct combined_chip_info { }; struct dell_wmi_ddv_sensors { + bool active; struct mutex lock; /* protect caching */ unsigned long timestamp; union acpi_object *obj; @@ -520,6 +521,9 @@ static struct hwmon_channel_info *dell_wmi_ddv_channel_create(struct device *dev static void dell_wmi_ddv_hwmon_cache_invalidate(struct dell_wmi_ddv_sensors *sensors) { + if (!sensors->active) + return; + mutex_lock(&sensors->lock); kfree(sensors->obj); sensors->obj = NULL; @@ -530,6 +534,7 @@ static void dell_wmi_ddv_hwmon_cache_destroy(void *data) { struct dell_wmi_ddv_sensors *sensors = data; + sensors->active = false; mutex_destroy(&sensors->lock); kfree(sensors->obj); } @@ -549,6 +554,7 @@ static struct hwmon_channel_info *dell_wmi_ddv_channel_init(struct wmi_device *w return ERR_PTR(ret); mutex_init(&sensors->lock); + sensors->active = true; ret = devm_add_action_or_reset(&wdev->dev, dell_wmi_ddv_hwmon_cache_destroy, sensors); if (ret < 0) @@ -852,7 +858,7 @@ static int dell_wmi_ddv_resume(struct device *dev) { struct dell_wmi_ddv_data *data = dev_get_drvdata(dev); - /* Force re-reading of all sensors */ + /* Force re-reading of all active sensors */ dell_wmi_ddv_hwmon_cache_invalidate(&data->fans); dell_wmi_ddv_hwmon_cache_invalidate(&data->temps); From 0331b1b0ba65376ecf1c69414aa7696fef0930cb Mon Sep 17 00:00:00 2001 From: Armin Wolf Date: Sat, 18 Feb 2023 12:53:18 +0100 Subject: [PATCH 737/765] platform/x86: dell-ddv: Fix temperature scaling After using the built-in UEFI hardware diagnostics to compare the measured battery temperature, i noticed that the temperature is actually expressed in tenth degree kelvin, similar to the SBS-Data standard. For example, a value of 2992 is displayed as 26 degrees celsius. Fix the scaling so that the correct values are being displayed. Tested on a Dell Inspiron 3505. Fixes: a77272c16041 ("platform/x86: dell: Add new dell-wmi-ddv driver") Signed-off-by: Armin Wolf Link: https://lore.kernel.org/r/20230218115318.20662-2-W_Armin@gmx.de Signed-off-by: Hans de Goede Reviewed-by: Hans de Goede --- drivers/platform/x86/dell/dell-wmi-ddv.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/platform/x86/dell/dell-wmi-ddv.c b/drivers/platform/x86/dell/dell-wmi-ddv.c index eff4e9649faf2..2750dee99c3e2 100644 --- a/drivers/platform/x86/dell/dell-wmi-ddv.c +++ b/drivers/platform/x86/dell/dell-wmi-ddv.c @@ -17,7 +17,6 @@ #include #include #include -#include #include #include #include @@ -665,7 +664,8 @@ static ssize_t temp_show(struct device *dev, struct device_attribute *attr, char if (ret < 0) return ret; - return sysfs_emit(buf, "%d\n", DIV_ROUND_CLOSEST(value, 10)); + /* Use 2731 instead of 2731.5 to avoid unnecessary rounding */ + return sysfs_emit(buf, "%d\n", value - 2731); } static ssize_t eppid_show(struct device *dev, struct device_attribute *attr, char *buf) From 95ecf90158522269749f1b7ce98b1eed66ca087b Mon Sep 17 00:00:00 2001 From: Srinivas Pandruvada Date: Sun, 26 Feb 2023 21:35:04 -0800 Subject: [PATCH 738/765] platform/x86: ISST: Increase range of valid mail box commands A new command CONFIG_TDP_GET_RATIO_INFO is added, with sub command type of 0x0C. The previous range of valid sub commands was from 0x00 to 0x0B. Change the valid range from 0x00 to 0x0C. Signed-off-by: Srinivas Pandruvada Link: https://lore.kernel.org/r/20230227053504.2734214-1-srinivas.pandruvada@linux.intel.com Signed-off-by: Hans de Goede Reviewed-by: Hans de Goede --- drivers/platform/x86/intel/speed_select_if/isst_if_common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/platform/x86/intel/speed_select_if/isst_if_common.c b/drivers/platform/x86/intel/speed_select_if/isst_if_common.c index a7e02b24a87ad..0829e793a8fce 100644 --- a/drivers/platform/x86/intel/speed_select_if/isst_if_common.c +++ b/drivers/platform/x86/intel/speed_select_if/isst_if_common.c @@ -47,7 +47,7 @@ struct isst_cmd_set_req_type { static const struct isst_valid_cmd_ranges isst_valid_cmds[] = { {0xD0, 0x00, 0x03}, - {0x7F, 0x00, 0x0B}, + {0x7F, 0x00, 0x0C}, {0x7F, 0x10, 0x12}, {0x7F, 0x20, 0x23}, {0x94, 0x03, 0x03}, From 6a192c0cbf38a6ba10847590d680975086844dbb Mon Sep 17 00:00:00 2001 From: Srinivas Pandruvada Date: Mon, 27 Feb 2023 06:06:14 -0800 Subject: [PATCH 739/765] platform/x86/intel/tpmi: Fix double free reported by Smatch Fix warning: drivers/platform/x86/intel/tpmi.c:253 tpmi_create_device() warn: 'feature_vsec_dev' was already freed. If there is some error, feature_vsec_dev memory is freed as part of resource managed call intel_vsec_add_aux(). So, additional kfree() call is not required. Reordered res allocation and feature_vsec_dev, so that on error only res is freed. Reported-by: Dan Carpenter Link: https://lore.kernel.org/platform-driver-x86/Y%2FxYR7WGiPayZu%2FR@kili/T/#u Fixes: 47731fd2865f ("platform/x86/intel: Intel TPMI enumeration driver") Signed-off-by: Srinivas Pandruvada Link: https://lore.kernel.org/r/20230227140614.2913474-1-srinivas.pandruvada@linux.intel.com Signed-off-by: Hans de Goede Reviewed-by: Hans de Goede --- drivers/platform/x86/intel/tpmi.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/platform/x86/intel/tpmi.c b/drivers/platform/x86/intel/tpmi.c index c60733261c89c..c999732b0f1e5 100644 --- a/drivers/platform/x86/intel/tpmi.c +++ b/drivers/platform/x86/intel/tpmi.c @@ -209,14 +209,14 @@ static int tpmi_create_device(struct intel_tpmi_info *tpmi_info, if (!name) return -EOPNOTSUPP; - feature_vsec_dev = kzalloc(sizeof(*feature_vsec_dev), GFP_KERNEL); - if (!feature_vsec_dev) + res = kcalloc(pfs->pfs_header.num_entries, sizeof(*res), GFP_KERNEL); + if (!res) return -ENOMEM; - res = kcalloc(pfs->pfs_header.num_entries, sizeof(*res), GFP_KERNEL); - if (!res) { + feature_vsec_dev = kzalloc(sizeof(*feature_vsec_dev), GFP_KERNEL); + if (!feature_vsec_dev) { ret = -ENOMEM; - goto free_vsec; + goto free_res; } snprintf(feature_id_name, sizeof(feature_id_name), "tpmi-%s", name); @@ -239,6 +239,8 @@ static int tpmi_create_device(struct intel_tpmi_info *tpmi_info, /* * intel_vsec_add_aux() is resource managed, no explicit * delete is required on error or on module unload. + * feature_vsec_dev memory is also freed as part of device + * delete. */ ret = intel_vsec_add_aux(vsec_dev->pcidev, &vsec_dev->auxdev.dev, feature_vsec_dev, feature_id_name); @@ -249,8 +251,6 @@ static int tpmi_create_device(struct intel_tpmi_info *tpmi_info, free_res: kfree(res); -free_vsec: - kfree(feature_vsec_dev); return ret; } From 03f5eb300ad1241f854269a3e521b119189a4493 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 25 Feb 2023 21:39:50 -0800 Subject: [PATCH 740/765] platform: mellanox: select REGMAP instead of depending on it REGMAP is a hidden (not user visible) symbol. Users cannot set it directly thru "make *config", so drivers should select it instead of depending on it if they need it. Consistently using "select" or "depends on" can also help reduce Kconfig circular dependency issues. Therefore, change the use of "depends on REGMAP" to "select REGMAP". For NVSW_SN2201, select REGMAP_I2C instead of depending on it. Fixes: c6acad68eb2d ("platform/mellanox: mlxreg-hotplug: Modify to use a regmap interface") Fixes: 5ec4a8ace06c ("platform/mellanox: Introduce support for Mellanox register access driver") Fixes: 62f9529b8d5c ("platform/mellanox: mlxreg-lc: Add initial support for Nvidia line card devices") Fixes: 662f24826f95 ("platform/mellanox: Add support for new SN2201 system") Signed-off-by: Randy Dunlap Cc: Darren Hart Cc: Hans de Goede Cc: Michael Shych Cc: Mark Gross Cc: Vadim Pasternak Cc: platform-driver-x86@vger.kernel.org Link: https://lore.kernel.org/r/20230226053953.4681-6-rdunlap@infradead.org Signed-off-by: Hans de Goede Reviewed-by: Hans de Goede --- drivers/platform/mellanox/Kconfig | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/platform/mellanox/Kconfig b/drivers/platform/mellanox/Kconfig index 09c7829e95c4b..382793e73a60a 100644 --- a/drivers/platform/mellanox/Kconfig +++ b/drivers/platform/mellanox/Kconfig @@ -16,17 +16,17 @@ if MELLANOX_PLATFORM config MLXREG_HOTPLUG tristate "Mellanox platform hotplug driver support" - depends on REGMAP depends on HWMON depends on I2C + select REGMAP help This driver handles hot-plug events for the power suppliers, power cables and fans on the wide range Mellanox IB and Ethernet systems. config MLXREG_IO tristate "Mellanox platform register access driver support" - depends on REGMAP depends on HWMON + select REGMAP help This driver allows access to Mellanox programmable device register space through sysfs interface. The sets of registers for sysfs access @@ -36,9 +36,9 @@ config MLXREG_IO config MLXREG_LC tristate "Mellanox line card platform driver support" - depends on REGMAP depends on HWMON depends on I2C + select REGMAP help This driver provides support for the Mellanox MSN4800-XX line cards, which are the part of MSN4800 Ethernet modular switch systems @@ -80,10 +80,9 @@ config MLXBF_PMC config NVSW_SN2201 tristate "Nvidia SN2201 platform driver support" - depends on REGMAP depends on HWMON depends on I2C - depends on REGMAP_I2C + select REGMAP_I2C help This driver provides support for the Nvidia SN2201 platform. The SN2201 is a highly integrated for one rack unit system with From 7e7e1541c91615e9950d0b96bcd1806d297e970e Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 25 Feb 2023 21:39:51 -0800 Subject: [PATCH 741/765] platform: x86: MLX_PLATFORM: select REGMAP instead of depending on it REGMAP is a hidden (not user visible) symbol. Users cannot set it directly thru "make *config", so drivers should select it instead of depending on it if they need it. Consistently using "select" or "depends on" can also help reduce Kconfig circular dependency issues. Therefore, change the use of "depends on REGMAP" to "select REGMAP". Fixes: ef0f62264b2a ("platform/x86: mlx-platform: Add physical bus number auto detection") Signed-off-by: Randy Dunlap Cc: Vadim Pasternak Cc: Darren Hart Cc: Hans de Goede Cc: Mark Gross Cc: platform-driver-x86@vger.kernel.org Link: https://lore.kernel.org/r/20230226053953.4681-7-rdunlap@infradead.org Signed-off-by: Hans de Goede Reviewed-by: Hans de Goede --- drivers/platform/x86/Kconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/platform/x86/Kconfig b/drivers/platform/x86/Kconfig index ec7c2b4e1721c..4a01b315e0a91 100644 --- a/drivers/platform/x86/Kconfig +++ b/drivers/platform/x86/Kconfig @@ -955,7 +955,8 @@ config SERIAL_MULTI_INSTANTIATE config MLX_PLATFORM tristate "Mellanox Technologies platform support" - depends on I2C && REGMAP + depends on I2C + select REGMAP help This option enables system support for the Mellanox Technologies platform. The Mellanox systems provide data center networking From 94e9cbda06a317f886f63fab511dedf441c81c54 Mon Sep 17 00:00:00 2001 From: Srinivas Pandruvada Date: Fri, 10 Feb 2023 22:32:46 -0800 Subject: [PATCH 742/765] platform/x86: ISST: Fix kernel documentation warnings Fix warning displayed for "make W=1" for kernel documentation. Signed-off-by: Srinivas Pandruvada Link: https://lore.kernel.org/r/20230211063257.311746-2-srinivas.pandruvada@linux.intel.com Signed-off-by: Hans de Goede Reviewed-by: Hans de Goede --- drivers/platform/x86/intel/speed_select_if/isst_if_common.c | 3 ++- drivers/platform/x86/intel/speed_select_if/isst_if_common.h | 1 + 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/platform/x86/intel/speed_select_if/isst_if_common.c b/drivers/platform/x86/intel/speed_select_if/isst_if_common.c index 0829e793a8fce..0954a04623edf 100644 --- a/drivers/platform/x86/intel/speed_select_if/isst_if_common.c +++ b/drivers/platform/x86/intel/speed_select_if/isst_if_common.c @@ -112,6 +112,7 @@ static void isst_delete_hash(void) * isst_store_cmd() - Store command to a hash table * @cmd: Mailbox command. * @sub_cmd: Mailbox sub-command or MSR id. + * @cpu: Target CPU for the command * @mbox_cmd_type: Mailbox or MSR command. * @param: Mailbox parameter. * @data: Mailbox request data or MSR data. @@ -363,7 +364,7 @@ static struct pci_dev *_isst_if_get_pci_dev(int cpu, int bus_no, int dev, int fn /** * isst_if_get_pci_dev() - Get the PCI device instance for a CPU * @cpu: Logical CPU number. - * @bus_number: The bus number assigned by the hardware. + * @bus_no: The bus number assigned by the hardware. * @dev: The device number assigned by the hardware. * @fn: The function number assigned by the hardware. * diff --git a/drivers/platform/x86/intel/speed_select_if/isst_if_common.h b/drivers/platform/x86/intel/speed_select_if/isst_if_common.h index fdecdae248d77..35ff506b402e1 100644 --- a/drivers/platform/x86/intel/speed_select_if/isst_if_common.h +++ b/drivers/platform/x86/intel/speed_select_if/isst_if_common.h @@ -40,6 +40,7 @@ * @offset: Offset to the first valid member in command structure. * This will be the offset of the start of the command * after command count field + * @owner: Registered module owner * @cmd_callback: Callback function to handle IOCTL. The callback has the * command pointer with data for command. There is a pointer * called write_only, which when set, will not copy the From e8059d393158e927e36898aca89986a52025b9f3 Mon Sep 17 00:00:00 2001 From: Daniel Scally Date: Thu, 2 Mar 2023 10:26:11 +0000 Subject: [PATCH 743/765] platform/x86: int3472: Add GPIOs to Surface Go 3 Board data Add the INT347E GPIO lookup table to the board data for the Surface Go 3. This is necessary to allow the ov7251 IR camera to probe properly on that platform. Signed-off-by: Daniel Scally Link: https://lore.kernel.org/r/20230302102611.314341-1-dan.scally@ideasonboard.com Reviewed-by: Hans de Goede Signed-off-by: Hans de Goede --- drivers/platform/x86/intel/int3472/tps68470_board_data.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/platform/x86/intel/int3472/tps68470_board_data.c b/drivers/platform/x86/intel/int3472/tps68470_board_data.c index 309eab9c05588..322237e056f32 100644 --- a/drivers/platform/x86/intel/int3472/tps68470_board_data.c +++ b/drivers/platform/x86/intel/int3472/tps68470_board_data.c @@ -159,9 +159,10 @@ static const struct int3472_tps68470_board_data surface_go_tps68470_board_data = static const struct int3472_tps68470_board_data surface_go3_tps68470_board_data = { .dev_name = "i2c-INT3472:01", .tps68470_regulator_pdata = &surface_go_tps68470_pdata, - .n_gpiod_lookups = 1, + .n_gpiod_lookups = 2, .tps68470_gpio_lookup_tables = { - &surface_go_int347a_gpios + &surface_go_int347a_gpios, + &surface_go_int347e_gpios, }, }; From 1a0009abfa7893b9cfcd3884658af1cbee6b26ce Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Tue, 7 Mar 2023 11:58:42 +0100 Subject: [PATCH 744/765] platform: mellanox: mlx-platform: Initialize shift variable to 0 Initialize shift variable in mlxplat_mlxcpld_verify_bus_topology() to 0 to avoid the following compile error: drivers/platform/x86/mlx-platform.c:6013 mlxplat_mlxcpld_verify_bus_topology() error: uninitialized symbol 'shift'. Fixes: 50b823fdd357 ("platform: mellanox: mlx-platform: Move bus shift assignment out of the loop") Cc: Vadim Pasternak Cc: Michael Shych Signed-off-by: Hans de Goede Link: https://lore.kernel.org/r/20230307105842.286118-1-hdegoede@redhat.com --- drivers/platform/x86/mlx-platform.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/platform/x86/mlx-platform.c b/drivers/platform/x86/mlx-platform.c index 7b6779cdb1349..67367f010139e 100644 --- a/drivers/platform/x86/mlx-platform.c +++ b/drivers/platform/x86/mlx-platform.c @@ -5980,7 +5980,7 @@ MODULE_DEVICE_TABLE(dmi, mlxplat_dmi_table); static int mlxplat_mlxcpld_verify_bus_topology(int *nr) { struct i2c_adapter *search_adap; - int shift, i; + int i, shift = 0; /* Scan adapters from expected id to verify it is free. */ *nr = MLXPLAT_CPLD_PHYS_ADAPTER_DEF_NR; From 418383e6ed6b4624a54ec05c535f13d184fbf33b Mon Sep 17 00:00:00 2001 From: Enrico Sau Date: Mon, 6 Mar 2023 12:59:33 +0100 Subject: [PATCH 745/765] net: usb: cdc_mbim: avoid altsetting toggling for Telit FE990 Add quirk CDC_MBIM_FLAG_AVOID_ALTSETTING_TOGGLE for Telit FE990 0x1081 composition in order to avoid bind error. Signed-off-by: Enrico Sau Link: https://lore.kernel.org/r/20230306115933.198259-1-enrico.sau@gmail.com Signed-off-by: Paolo Abeni --- drivers/net/usb/cdc_mbim.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/usb/cdc_mbim.c b/drivers/net/usb/cdc_mbim.c index c89639381eca3..cd4083e0b3b9e 100644 --- a/drivers/net/usb/cdc_mbim.c +++ b/drivers/net/usb/cdc_mbim.c @@ -665,6 +665,11 @@ static const struct usb_device_id mbim_devs[] = { .driver_info = (unsigned long)&cdc_mbim_info_avoid_altsetting_toggle, }, + /* Telit FE990 */ + { USB_DEVICE_AND_INTERFACE_INFO(0x1bc7, 0x1081, USB_CLASS_COMM, USB_CDC_SUBCLASS_MBIM, USB_CDC_PROTO_NONE), + .driver_info = (unsigned long)&cdc_mbim_info_avoid_altsetting_toggle, + }, + /* default entry */ { USB_INTERFACE_INFO(USB_CLASS_COMM, USB_CDC_SUBCLASS_MBIM, USB_CDC_PROTO_NONE), .driver_info = (unsigned long)&cdc_mbim_info_zlp, From 382e363d5bed0cec5807b35761d14e55955eee63 Mon Sep 17 00:00:00 2001 From: Enrico Sau Date: Mon, 6 Mar 2023 13:05:28 +0100 Subject: [PATCH 746/765] net: usb: qmi_wwan: add Telit 0x1080 composition Add the following Telit FE990 composition: 0x1080: tty, adb, rmnet, tty, tty, tty, tty Signed-off-by: Enrico Sau Link: https://lore.kernel.org/r/20230306120528.198842-1-enrico.sau@gmail.com Signed-off-by: Paolo Abeni --- drivers/net/usb/qmi_wwan.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c index a808d718c0123..571e37e67f9ce 100644 --- a/drivers/net/usb/qmi_wwan.c +++ b/drivers/net/usb/qmi_wwan.c @@ -1364,6 +1364,7 @@ static const struct usb_device_id products[] = { {QMI_QUIRK_SET_DTR(0x1bc7, 0x1057, 2)}, /* Telit FN980 */ {QMI_QUIRK_SET_DTR(0x1bc7, 0x1060, 2)}, /* Telit LN920 */ {QMI_QUIRK_SET_DTR(0x1bc7, 0x1070, 2)}, /* Telit FN990 */ + {QMI_QUIRK_SET_DTR(0x1bc7, 0x1080, 2)}, /* Telit FE990 */ {QMI_FIXED_INTF(0x1bc7, 0x1100, 3)}, /* Telit ME910 */ {QMI_FIXED_INTF(0x1bc7, 0x1101, 3)}, /* Telit ME910 dual modem */ {QMI_FIXED_INTF(0x1bc7, 0x1200, 5)}, /* Telit LE920 */ From 63355b9884b3d1677de6bd1517cd2b8a9bf53978 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 7 Mar 2023 12:16:18 -0800 Subject: [PATCH 747/765] cpumask: be more careful with 'cpumask_setall()' Commit 596ff4a09b89 ("cpumask: re-introduce constant-sized cpumask optimizations") changed cpumask_setall() to use "bitmap_set()" instead of "bitmap_fill()", because bitmap_fill() would explicitly set all the bits of a constant sized small bitmap, and that's exactly what we don't want: we want to only set bits up to 'nr_cpu_ids', which is what "bitmap_set()" does. However, Yury correctly points out that while "bitmap_set()" does indeed only set bits up to the required bitmap size, it doesn't _clear_ bits above that size, so the upper bits would still not have well-defined values. Now, none of this should really matter, since any bits set past 'nr_cpu_ids' should always be ignored in the first place. Yes, the bit scanning functions might return them as a result, but since users should always consider the ">= nr_cpu_ids" condition to mean "no more bits", that shouldn't have any actual effect (see previous commit 8ca09d5fa354 "cpumask: fix incorrect cpumask scanning result checks"). But let's just do it right, the way the code was _intended_ to work. We have had enough lazy code that works but bites us in the *rse later (again, see previous commit) that there's no reason to not just do this properly. It turns out that "bitmap_fill()" gets this all right for the complex case, and really only fails for the inlined optimized case that just fills the whole word. And while we could just fix bitmap_fill() to use the proper last word mask, there's two issues with that: - the cpumask case wants to do the _optimization_ based on "NR_CPUS is a small constant", but then wants to do the actual bit _fill_ based on "nr_cpu_ids" that isn't necessarily that same constant - we have lots of non-cpumask users of bitmap_fill(), and while they hopefully don't care, and probably would want the proper semantics anyway ("only set bits up to the limit"), I do not want the cpumask changes to impact other parts So this ends up just doing the single-word optimization by hand in the cpumask code. If our cpumask is fundamentally limited to a single word, just do the proper "fill in that word" exactly. And if it's the more complex multi-word case, then the generic bitmap_fill() will DTRT. This is all an example of how our bitmap function optimizations really are somewhat broken. They conflate the "this is size of the bitmap" optimizations with the actual bit(s) we want to set. In many cases we really want to have the two be separate things: sometimes we base our optimizations on the size of the whole bitmap ("I know this whole bitmap fits in a single word, so I'll just use single-word accesses"), and sometimes we base them on the bit we are looking at ("this is just acting on bits that are in the first word, so I'll use single-word accesses"). Notice how the end result of the two optimizations are the same, but the way we get to them are quite different. And all our cpumask optimization games are really about that fundamental distinction, and we'd often really want to pass in both the "this is the bit I'm working on" (which _can_ be a small constant but might be variable), and "I know it's in this range even if it's variable" (based on CONFIG_NR_CPUS). So this cpumask_setall() implementation just makes that explicit. It checks the "I statically know the size is small" using the known static size of the cpumask (which is what that 'small_cpumask_bits' is all about), but then sets the actual bits using the exact number of cpus we have (ie 'nr_cpumask_bits') Of course, in a perfect world, the compiler would have done all the range analysis (possibly with help from us just telling it that "this value is always in this range"), and would do all of this for us. But that is not the world we live in. While we dream of that perfect world, this does that manual logic to make it all work out. And this was a very long explanation for a small code change that shouldn't even matter. Reported-by: Yury Norov Link: https://lore.kernel.org/lkml/ZAV9nGG9e1%2FrV+L%2F@yury-laptop/ Signed-off-by: Linus Torvalds --- include/linux/cpumask.h | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h index ce8eb7ef21072..63d637d18e799 100644 --- a/include/linux/cpumask.h +++ b/include/linux/cpumask.h @@ -518,14 +518,14 @@ static __always_inline bool cpumask_test_and_clear_cpu(int cpu, struct cpumask * /** * cpumask_setall - set all cpus (< nr_cpu_ids) in a cpumask * @dstp: the cpumask pointer - * - * Note: since we set bits, we should use the tighter 'bitmap_set()' with - * the eact number of bits, not 'bitmap_fill()' that will fill past the - * end. */ static inline void cpumask_setall(struct cpumask *dstp) { - bitmap_set(cpumask_bits(dstp), 0, nr_cpumask_bits); + if (small_const_nbits(small_cpumask_bits)) { + cpumask_bits(dstp)[0] = BITMAP_LAST_WORD_MASK(nr_cpumask_bits); + return; + } + bitmap_fill(cpumask_bits(dstp), nr_cpumask_bits); } /** From fef3f92e8a4214652d8f33f50330dc5a92efbf11 Mon Sep 17 00:00:00 2001 From: Dave Ertman Date: Fri, 27 Jan 2023 14:24:10 +0100 Subject: [PATCH 748/765] ice: Fix DSCP PFC TLV creation When creating the TLV to send to the FW for configuring DSCP mode PFC,the PFCENABLE field was being masked with a 4 bit mask (0xF), but this is an 8 bit bitmask for enabled classes for PFC. This means that traffic classes 4-7 could not be enabled for PFC. Remove the mask completely, as it is not necessary, as we are assigning 8 bits to an 8 bit field. Fixes: 2a87bd73e50d ("ice: Add DSCP support") Signed-off-by: Dave Ertman Signed-off-by: Karen Ostrowska Tested-by: Gurucharan G (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_dcb.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/ice/ice_dcb.c b/drivers/net/ethernet/intel/ice/ice_dcb.c index c557dfc50aadd..396e555023aae 100644 --- a/drivers/net/ethernet/intel/ice/ice_dcb.c +++ b/drivers/net/ethernet/intel/ice/ice_dcb.c @@ -1411,7 +1411,7 @@ ice_add_dscp_pfc_tlv(struct ice_lldp_org_tlv *tlv, struct ice_dcbx_cfg *dcbcfg) tlv->ouisubtype = htonl(ouisubtype); buf[0] = dcbcfg->pfc.pfccap & 0xF; - buf[1] = dcbcfg->pfc.pfcena & 0xF; + buf[1] = dcbcfg->pfc.pfcena; } /** From c4a9c8e78aad0fd9b79c9b1e83bab3288a15ce3c Mon Sep 17 00:00:00 2001 From: Michal Swiatkowski Date: Mon, 13 Feb 2023 12:27:33 +0100 Subject: [PATCH 749/765] ice: don't ignore return codes in VSI related code There were few smatch warnings reported by Dan: - ice_vsi_cfg_xdp_txqs can return 0 instead of ret, which is cleaner - return values in ice_vsi_cfg_def were ignored - in ice_vsi_rebuild return value was ignored in case rebuild failed, it was a never reached code, however, rewrite it for clarity. - ice_vsi_cfg_tc can return 0 instead of ret Fixes: 6624e780a577 ("ice: split ice_vsi_setup into smaller functions") Reported-by: Dan Carpenter Signed-off-by: Michal Swiatkowski Tested-by: Gurucharan G (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_lib.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c index 781475480ff27..0f52ea38b6f3a 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_lib.c @@ -2126,7 +2126,7 @@ int ice_vsi_cfg_xdp_txqs(struct ice_vsi *vsi) ice_for_each_rxq(vsi, i) ice_tx_xsk_pool(vsi, i); - return ret; + return 0; } /** @@ -2693,12 +2693,14 @@ ice_vsi_cfg_def(struct ice_vsi *vsi, struct ice_vsi_cfg_params *params) return ret; /* allocate memory for Tx/Rx ring stat pointers */ - if (ice_vsi_alloc_stat_arrays(vsi)) + ret = ice_vsi_alloc_stat_arrays(vsi); + if (ret) goto unroll_vsi_alloc; ice_alloc_fd_res(vsi); - if (ice_vsi_get_qs(vsi)) { + ret = ice_vsi_get_qs(vsi); + if (ret) { dev_err(dev, "Failed to allocate queues. vsi->idx = %d\n", vsi->idx); goto unroll_vsi_alloc_stat; @@ -2811,6 +2813,7 @@ ice_vsi_cfg_def(struct ice_vsi *vsi, struct ice_vsi_cfg_params *params) break; default: /* clean up the resources and exit */ + ret = -EINVAL; goto unroll_vsi_init; } @@ -3508,10 +3511,10 @@ int ice_vsi_rebuild(struct ice_vsi *vsi, u32 vsi_flags) if (vsi_flags & ICE_VSI_FLAG_INIT) { ret = -EIO; goto err_vsi_cfg_tc_lan; - } else { - kfree(coalesce); - return ice_schedule_reset(pf, ICE_RESET_PFR); } + + kfree(coalesce); + return ice_schedule_reset(pf, ICE_RESET_PFR); } ice_vsi_realloc_stat_arrays(vsi, prev_txq, prev_rxq); @@ -3759,7 +3762,7 @@ int ice_vsi_cfg_tc(struct ice_vsi *vsi, u8 ena_tc) dev = ice_pf_to_dev(pf); if (vsi->tc_cfg.ena_tc == ena_tc && vsi->mqprio_qopt.mode != TC_MQPRIO_MODE_CHANNEL) - return ret; + return 0; ice_for_each_traffic_class(i) { /* build bitmap of enabled TCs */ From 8f5c5a790e3025d6eca96bf7ee5e3873dc92373f Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 14 Feb 2023 16:25:36 +0100 Subject: [PATCH 750/765] ethernet: ice: avoid gcc-9 integer overflow warning With older compilers like gcc-9, the calculation of the vlan priority field causes a false-positive warning from the byteswap: In file included from drivers/net/ethernet/intel/ice/ice_tc_lib.c:4: drivers/net/ethernet/intel/ice/ice_tc_lib.c: In function 'ice_parse_cls_flower': include/uapi/linux/swab.h:15:15: error: integer overflow in expression '(int)(short unsigned int)((int)match.key->..vlan_priority << 13) & 57344 & 255' of type 'int' results in '0' [-Werror=overflow] 15 | (((__u16)(x) & (__u16)0x00ffU) << 8) | \ | ~~~~~~~~~~~~^~~~~~~~~~~~~~~~~ include/uapi/linux/swab.h:106:2: note: in expansion of macro '___constant_swab16' 106 | ___constant_swab16(x) : \ | ^~~~~~~~~~~~~~~~~~ include/uapi/linux/byteorder/little_endian.h:42:43: note: in expansion of macro '__swab16' 42 | #define __cpu_to_be16(x) ((__force __be16)__swab16((x))) | ^~~~~~~~ include/linux/byteorder/generic.h:96:21: note: in expansion of macro '__cpu_to_be16' 96 | #define cpu_to_be16 __cpu_to_be16 | ^~~~~~~~~~~~~ drivers/net/ethernet/intel/ice/ice_tc_lib.c:1458:5: note: in expansion of macro 'cpu_to_be16' 1458 | cpu_to_be16((match.key->vlan_priority << | ^~~~~~~~~~~ After a change to be16_encode_bits(), the code becomes more readable to both people and compilers, which avoids the warning. Fixes: 34800178b302 ("ice: Add support for VLAN priority filters in switchdev") Suggested-by: Alexander Lobakin Signed-off-by: Arnd Bergmann Reviewed-by: Alexander Lobakin Tested-by: Sujai Buvaneswaran Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_tc_lib.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_tc_lib.c b/drivers/net/ethernet/intel/ice/ice_tc_lib.c index 6b48cbc049c67..76f29a5bf8d73 100644 --- a/drivers/net/ethernet/intel/ice/ice_tc_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_tc_lib.c @@ -1455,8 +1455,8 @@ ice_parse_cls_flower(struct net_device *filter_dev, struct ice_vsi *vsi, if (match.mask->vlan_priority) { fltr->flags |= ICE_TC_FLWR_FIELD_VLAN_PRIO; headers->vlan_hdr.vlan_prio = - cpu_to_be16((match.key->vlan_priority << - VLAN_PRIO_SHIFT) & VLAN_PRIO_MASK); + be16_encode_bits(match.key->vlan_priority, + VLAN_PRIO_MASK); } if (match.mask->vlan_tpid) @@ -1489,8 +1489,8 @@ ice_parse_cls_flower(struct net_device *filter_dev, struct ice_vsi *vsi, if (match.mask->vlan_priority) { fltr->flags |= ICE_TC_FLWR_FIELD_CVLAN_PRIO; headers->cvlan_hdr.vlan_prio = - cpu_to_be16((match.key->vlan_priority << - VLAN_PRIO_SHIFT) & VLAN_PRIO_MASK); + be16_encode_bits(match.key->vlan_priority, + VLAN_PRIO_MASK); } } From 7d834b4d1ab66c48e8c0810fdeadaabb80fa2c81 Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Tue, 7 Mar 2023 00:26:50 +0300 Subject: [PATCH 751/765] nfc: change order inside nfc_se_io error path cb_context should be freed on the error path in nfc_se_io as stated by commit 25ff6f8a5a3b ("nfc: fix memory leak of se_io context in nfc_genl_se_io"). Make the error path in nfc_se_io unwind everything in reverse order, i.e. free the cb_context after unlocking the device. Suggested-by: Krzysztof Kozlowski Signed-off-by: Fedor Pchelkin Reviewed-by: Krzysztof Kozlowski Link: https://lore.kernel.org/r/20230306212650.230322-1-pchelkin@ispras.ru Signed-off-by: Jakub Kicinski --- net/nfc/netlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/nfc/netlink.c b/net/nfc/netlink.c index 348bf561bc9fb..b9264e730fd93 100644 --- a/net/nfc/netlink.c +++ b/net/nfc/netlink.c @@ -1446,8 +1446,8 @@ static int nfc_se_io(struct nfc_dev *dev, u32 se_idx, return rc; error: - kfree(cb_context); device_unlock(&dev->dev); + kfree(cb_context); return rc; } From e7b15acdc10f74872fdae1482b3f837818362085 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 6 Mar 2023 11:20:18 -0800 Subject: [PATCH 752/765] mailmap: add entry for Maxim Mikityanskiy Map Maxim's old corporate addresses to his personal one. Link: https://lore.kernel.org/r/20230306192018.3894988-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- .mailmap | 2 ++ 1 file changed, 2 insertions(+) diff --git a/.mailmap b/.mailmap index a872c9683958c..04bd44774ad08 100644 --- a/.mailmap +++ b/.mailmap @@ -304,6 +304,8 @@ Mauro Carvalho Chehab Mauro Carvalho Chehab Mauro Carvalho Chehab Mauro Carvalho Chehab +Maxim Mikityanskiy +Maxim Mikityanskiy Maxime Ripard Maxime Ripard Mayuresh Janorkar From b1649b0fe98c2c7eed156c957f6f524d5660d859 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Mon, 6 Mar 2023 11:44:05 -0800 Subject: [PATCH 753/765] mailmap: update entries for Stephen Hemminger Map all my old email addresses to current address. Signed-off-by: Stephen Hemminger Link: https://lore.kernel.org/r/20230306194405.108236-1-stephen@networkplumber.org Signed-off-by: Jakub Kicinski --- .mailmap | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/.mailmap b/.mailmap index 04bd44774ad08..4f37ff13a3464 100644 --- a/.mailmap +++ b/.mailmap @@ -411,7 +411,10 @@ Shuah Khan Simon Arlott Simon Kelley Stéphane Witzmann -Stephen Hemminger +Stephen Hemminger +Stephen Hemminger +Stephen Hemminger +Stephen Hemminger Steve Wise Steve Wise Subash Abhinov Kasiviswanathan From 37d9df224d1eec1b434fe9ffa40104c756478c29 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 6 Mar 2023 12:04:57 -0800 Subject: [PATCH 754/765] ynl: re-license uniformly under GPL-2.0 OR BSD-3-Clause I was intending to make all the Netlink Spec code BSD-3-Clause to ease the adoption but it appears that: - I fumbled the uAPI and used "GPL WITH uAPI note" there - it gives people pause as they expect GPL in the kernel As suggested by Chuck re-license under dual. This gives us benefit of full BSD freedom while fulfilling the broad "kernel is under GPL" expectations. Link: https://lore.kernel.org/all/20230304120108.05dd44c5@kernel.org/ Link: https://lore.kernel.org/r/20230306200457.3903854-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- Documentation/netlink/genetlink-c.yaml | 2 +- Documentation/netlink/genetlink-legacy.yaml | 2 +- Documentation/netlink/genetlink.yaml | 2 +- Documentation/netlink/specs/ethtool.yaml | 2 ++ Documentation/netlink/specs/fou.yaml | 2 ++ Documentation/netlink/specs/netdev.yaml | 2 ++ Documentation/userspace-api/netlink/specs.rst | 3 +++ include/uapi/linux/fou.h | 2 +- include/uapi/linux/netdev.h | 2 +- net/core/netdev-genl-gen.c | 2 +- net/core/netdev-genl-gen.h | 2 +- net/ipv4/fou_nl.c | 2 +- net/ipv4/fou_nl.h | 2 +- tools/include/uapi/linux/netdev.h | 2 +- tools/net/ynl/cli.py | 2 +- tools/net/ynl/lib/__init__.py | 2 +- tools/net/ynl/lib/nlspec.py | 2 +- tools/net/ynl/lib/ynl.py | 2 +- tools/net/ynl/ynl-gen-c.py | 7 ++++--- tools/net/ynl/ynl-regen.sh | 2 +- 20 files changed, 28 insertions(+), 18 deletions(-) diff --git a/Documentation/netlink/genetlink-c.yaml b/Documentation/netlink/genetlink-c.yaml index bbcfa2472b047..f082a5ad7cf1d 100644 --- a/Documentation/netlink/genetlink-c.yaml +++ b/Documentation/netlink/genetlink-c.yaml @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: GPL-2.0 +# SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause %YAML 1.2 --- $id: http://kernel.org/schemas/netlink/genetlink-c.yaml# diff --git a/Documentation/netlink/genetlink-legacy.yaml b/Documentation/netlink/genetlink-legacy.yaml index 5642925c4ceb1..c6b8c77f7d12e 100644 --- a/Documentation/netlink/genetlink-legacy.yaml +++ b/Documentation/netlink/genetlink-legacy.yaml @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: GPL-2.0 +# SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause %YAML 1.2 --- $id: http://kernel.org/schemas/netlink/genetlink-legacy.yaml# diff --git a/Documentation/netlink/genetlink.yaml b/Documentation/netlink/genetlink.yaml index 62a922755ce2d..b2d56ab9e615d 100644 --- a/Documentation/netlink/genetlink.yaml +++ b/Documentation/netlink/genetlink.yaml @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: GPL-2.0 +# SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause %YAML 1.2 --- $id: http://kernel.org/schemas/netlink/genetlink-legacy.yaml# diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml index 35c462bce56f6..18ecb7d90cbe5 100644 --- a/Documentation/netlink/specs/ethtool.yaml +++ b/Documentation/netlink/specs/ethtool.yaml @@ -1,3 +1,5 @@ +# SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause + name: ethtool protocol: genetlink-legacy diff --git a/Documentation/netlink/specs/fou.yaml b/Documentation/netlink/specs/fou.yaml index cca4cf98f03ab..cff104288723b 100644 --- a/Documentation/netlink/specs/fou.yaml +++ b/Documentation/netlink/specs/fou.yaml @@ -1,3 +1,5 @@ +# SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause + name: fou protocol: genetlink-legacy diff --git a/Documentation/netlink/specs/netdev.yaml b/Documentation/netlink/specs/netdev.yaml index ba9ee13cf7296..24de747b53443 100644 --- a/Documentation/netlink/specs/netdev.yaml +++ b/Documentation/netlink/specs/netdev.yaml @@ -1,3 +1,5 @@ +# SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause + name: netdev doc: diff --git a/Documentation/userspace-api/netlink/specs.rst b/Documentation/userspace-api/netlink/specs.rst index 32e53328d113d..2122e0c4a3998 100644 --- a/Documentation/userspace-api/netlink/specs.rst +++ b/Documentation/userspace-api/netlink/specs.rst @@ -24,6 +24,9 @@ YAML specifications can be found under ``Documentation/netlink/specs/`` This document describes details of the schema. See :doc:`intro-specs` for a practical starting guide. +All specs must be licensed under ``GPL-2.0-only OR BSD-3-Clause`` +to allow for easy adoption in user space code. + Compatibility levels ==================== diff --git a/include/uapi/linux/fou.h b/include/uapi/linux/fou.h index 19ebbef41a63c..5041c35984938 100644 --- a/include/uapi/linux/fou.h +++ b/include/uapi/linux/fou.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ +/* SPDX-License-Identifier: (GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause */ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/fou.yaml */ /* YNL-GEN uapi header */ diff --git a/include/uapi/linux/netdev.h b/include/uapi/linux/netdev.h index 588391447bfb0..8c4e3e536c042 100644 --- a/include/uapi/linux/netdev.h +++ b/include/uapi/linux/netdev.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ +/* SPDX-License-Identifier: (GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause */ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/netdev.yaml */ /* YNL-GEN uapi header */ diff --git a/net/core/netdev-genl-gen.c b/net/core/netdev-genl-gen.c index 48812ec843f55..9e10802587fc3 100644 --- a/net/core/netdev-genl-gen.c +++ b/net/core/netdev-genl-gen.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: BSD-3-Clause +// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/netdev.yaml */ /* YNL-GEN kernel source */ diff --git a/net/core/netdev-genl-gen.h b/net/core/netdev-genl-gen.h index b16dc7e026bb1..2c5fc7d1e8a74 100644 --- a/net/core/netdev-genl-gen.h +++ b/net/core/netdev-genl-gen.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: BSD-3-Clause */ +/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/netdev.yaml */ /* YNL-GEN kernel header */ diff --git a/net/ipv4/fou_nl.c b/net/ipv4/fou_nl.c index 6c3820f41dd5d..5c14fe030eda5 100644 --- a/net/ipv4/fou_nl.c +++ b/net/ipv4/fou_nl.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: BSD-3-Clause +// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/fou.yaml */ /* YNL-GEN kernel source */ diff --git a/net/ipv4/fou_nl.h b/net/ipv4/fou_nl.h index b7a68121ce6f7..58b1e1ed4b3bb 100644 --- a/net/ipv4/fou_nl.h +++ b/net/ipv4/fou_nl.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: BSD-3-Clause */ +/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/fou.yaml */ /* YNL-GEN kernel header */ diff --git a/tools/include/uapi/linux/netdev.h b/tools/include/uapi/linux/netdev.h index 588391447bfb0..8c4e3e536c042 100644 --- a/tools/include/uapi/linux/netdev.h +++ b/tools/include/uapi/linux/netdev.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ +/* SPDX-License-Identifier: (GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause */ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/netdev.yaml */ /* YNL-GEN uapi header */ diff --git a/tools/net/ynl/cli.py b/tools/net/ynl/cli.py index db410b74d5395..ffaa8038aa8c6 100755 --- a/tools/net/ynl/cli.py +++ b/tools/net/ynl/cli.py @@ -1,5 +1,5 @@ #!/usr/bin/env python3 -# SPDX-License-Identifier: BSD-3-Clause +# SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause import argparse import json diff --git a/tools/net/ynl/lib/__init__.py b/tools/net/ynl/lib/__init__.py index 3c73f59eabab9..a2cb8b16d6f10 100644 --- a/tools/net/ynl/lib/__init__.py +++ b/tools/net/ynl/lib/__init__.py @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: BSD-3-Clause +# SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause from .nlspec import SpecAttr, SpecAttrSet, SpecFamily, SpecOperation from .ynl import YnlFamily diff --git a/tools/net/ynl/lib/nlspec.py b/tools/net/ynl/lib/nlspec.py index 9d394e50de23d..0a2cfb5862aa7 100644 --- a/tools/net/ynl/lib/nlspec.py +++ b/tools/net/ynl/lib/nlspec.py @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: BSD-3-Clause +# SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause import collections import importlib diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py index 1c7411ee04dc8..a842adc8e87e3 100644 --- a/tools/net/ynl/lib/ynl.py +++ b/tools/net/ynl/lib/ynl.py @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: BSD-3-Clause +# SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause import functools import os diff --git a/tools/net/ynl/ynl-gen-c.py b/tools/net/ynl/ynl-gen-c.py index 62f8f2c3c56c8..c940ca834d3f3 100755 --- a/tools/net/ynl/ynl-gen-c.py +++ b/tools/net/ynl/ynl-gen-c.py @@ -1,4 +1,5 @@ #!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause import argparse import collections @@ -2127,12 +2128,12 @@ def main(): _, spec_kernel = find_kernel_root(args.spec) if args.mode == 'uapi': - cw.p('/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */') + cw.p('/* SPDX-License-Identifier: (GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause */') else: if args.header: - cw.p('/* SPDX-License-Identifier: BSD-3-Clause */') + cw.p('/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */') else: - cw.p('// SPDX-License-Identifier: BSD-3-Clause') + cw.p('// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause') cw.p("/* Do not edit directly, auto-generated from: */") cw.p(f"/*\t{spec_kernel} */") cw.p(f"/* YNL-GEN {args.mode} {'header' if args.header else 'source'} */") diff --git a/tools/net/ynl/ynl-regen.sh b/tools/net/ynl/ynl-regen.sh index 43989ae48ed05..74f5de1c23994 100755 --- a/tools/net/ynl/ynl-regen.sh +++ b/tools/net/ynl/ynl-regen.sh @@ -1,5 +1,5 @@ #!/bin/bash -# SPDX-License-Identifier: BSD-3-Clause +# SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause TOOL=$(dirname $(realpath $0))/ynl-gen-c.py From ce7ca794712f186da99719e8b4e97bd5ddbb04c3 Mon Sep 17 00:00:00 2001 From: "D. Wythe" Date: Tue, 7 Mar 2023 11:23:46 +0800 Subject: [PATCH 755/765] net/smc: fix fallback failed while sendmsg with fastopen Before determining whether the msg has unsupported options, it has been prematurely terminated by the wrong status check. For the application, the general usages of MSG_FASTOPEN likes fd = socket(...) /* rather than connect */ sendto(fd, data, len, MSG_FASTOPEN) Hence, We need to check the flag before state check, because the sock state here is always SMC_INIT when applications tries MSG_FASTOPEN. Once we found unsupported options, fallback it to TCP. Fixes: ee9dfbef02d1 ("net/smc: handle sockopts forcing fallback") Signed-off-by: D. Wythe Signed-off-by: Simon Horman v2 -> v1: Optimize code style Reviewed-by: Tony Lu Signed-off-by: David S. Miller --- net/smc/af_smc.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index a4cccdfdc00a2..ff6dd86bdc9f3 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -2657,16 +2657,14 @@ static int smc_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) { struct sock *sk = sock->sk; struct smc_sock *smc; - int rc = -EPIPE; + int rc; smc = smc_sk(sk); lock_sock(sk); - if ((sk->sk_state != SMC_ACTIVE) && - (sk->sk_state != SMC_APPCLOSEWAIT1) && - (sk->sk_state != SMC_INIT)) - goto out; + /* SMC does not support connect with fastopen */ if (msg->msg_flags & MSG_FASTOPEN) { + /* not connected yet, fallback */ if (sk->sk_state == SMC_INIT && !smc->connect_nonblock) { rc = smc_switch_to_fallback(smc, SMC_CLC_DECL_OPTUNSUPP); if (rc) @@ -2675,6 +2673,11 @@ static int smc_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) rc = -EINVAL; goto out; } + } else if ((sk->sk_state != SMC_ACTIVE) && + (sk->sk_state != SMC_APPCLOSEWAIT1) && + (sk->sk_state != SMC_INIT)) { + rc = -EPIPE; + goto out; } if (smc->use_fallback) { From ea9dd2e5c6d12c8b65ce7514c8359a70eeaa0e70 Mon Sep 17 00:00:00 2001 From: Suman Ghosh Date: Tue, 7 Mar 2023 16:19:08 +0530 Subject: [PATCH 756/765] octeontx2-af: Unlock contexts in the queue context cache in case of fault detection NDC caches contexts of frequently used queue's (Rx and Tx queues) contexts. Due to a HW errata when NDC detects fault/poision while accessing contexts it could go into an illegal state where a cache line could get locked forever. To makesure all cache lines in NDC are available for optimum performance upon fault/lockerror/posion errors scan through all cache lines in NDC and clear the lock bit. Fixes: 4a3581cd5995 ("octeontx2-af: NPA AQ instruction enqueue support") Signed-off-by: Suman Ghosh Signed-off-by: Sunil Kovvuri Goutham Signed-off-by: Sai Krishna Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- .../net/ethernet/marvell/octeontx2/af/rvu.h | 5 ++ .../marvell/octeontx2/af/rvu_debugfs.c | 7 +-- .../ethernet/marvell/octeontx2/af/rvu_nix.c | 16 ++++- .../ethernet/marvell/octeontx2/af/rvu_npa.c | 58 ++++++++++++++++++- .../ethernet/marvell/octeontx2/af/rvu_reg.h | 3 + 5 files changed, 82 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h index 389663a13d1d1..ef721caeac49b 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h @@ -884,6 +884,9 @@ int rvu_cpt_lf_teardown(struct rvu *rvu, u16 pcifunc, int blkaddr, int lf, int rvu_cpt_ctx_flush(struct rvu *rvu, u16 pcifunc); int rvu_cpt_init(struct rvu *rvu); +#define NDC_AF_BANK_MASK GENMASK_ULL(7, 0) +#define NDC_AF_BANK_LINE_MASK GENMASK_ULL(31, 16) + /* CN10K RVU */ int rvu_set_channels_base(struct rvu *rvu); void rvu_program_channels(struct rvu *rvu); @@ -902,6 +905,8 @@ static inline void rvu_dbg_init(struct rvu *rvu) {} static inline void rvu_dbg_exit(struct rvu *rvu) {} #endif +int rvu_ndc_fix_locked_cacheline(struct rvu *rvu, int blkaddr); + /* RVU Switch */ void rvu_switch_enable(struct rvu *rvu); void rvu_switch_disable(struct rvu *rvu); diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c index fa280ebd3052b..26cfa501f1a11 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c @@ -198,9 +198,6 @@ enum cpt_eng_type { CPT_IE_TYPE = 3, }; -#define NDC_MAX_BANK(rvu, blk_addr) (rvu_read64(rvu, \ - blk_addr, NDC_AF_CONST) & 0xFF) - #define rvu_dbg_NULL NULL #define rvu_dbg_open_NULL NULL @@ -1448,6 +1445,7 @@ static int ndc_blk_hits_miss_stats(struct seq_file *s, int idx, int blk_addr) struct nix_hw *nix_hw; struct rvu *rvu; int bank, max_bank; + u64 ndc_af_const; if (blk_addr == BLKADDR_NDC_NPA0) { rvu = s->private; @@ -1456,7 +1454,8 @@ static int ndc_blk_hits_miss_stats(struct seq_file *s, int idx, int blk_addr) rvu = nix_hw->rvu; } - max_bank = NDC_MAX_BANK(rvu, blk_addr); + ndc_af_const = rvu_read64(rvu, blk_addr, NDC_AF_CONST); + max_bank = FIELD_GET(NDC_AF_BANK_MASK, ndc_af_const); for (bank = 0; bank < max_bank; bank++) { seq_printf(s, "BANK:%d\n", bank); seq_printf(s, "\tHits:\t%lld\n", diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c index 26e639e57dae3..4ad707e758b9f 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c @@ -790,6 +790,7 @@ static int nix_aq_enqueue_wait(struct rvu *rvu, struct rvu_block *block, struct nix_aq_res_s *result; int timeout = 1000; u64 reg, head; + int ret; result = (struct nix_aq_res_s *)aq->res->base; @@ -813,9 +814,22 @@ static int nix_aq_enqueue_wait(struct rvu *rvu, struct rvu_block *block, return -EBUSY; } - if (result->compcode != NIX_AQ_COMP_GOOD) + if (result->compcode != NIX_AQ_COMP_GOOD) { /* TODO: Replace this with some error code */ + if (result->compcode == NIX_AQ_COMP_CTX_FAULT || + result->compcode == NIX_AQ_COMP_LOCKERR || + result->compcode == NIX_AQ_COMP_CTX_POISON) { + ret = rvu_ndc_fix_locked_cacheline(rvu, BLKADDR_NDC_NIX0_RX); + ret |= rvu_ndc_fix_locked_cacheline(rvu, BLKADDR_NDC_NIX0_TX); + ret |= rvu_ndc_fix_locked_cacheline(rvu, BLKADDR_NDC_NIX1_RX); + ret |= rvu_ndc_fix_locked_cacheline(rvu, BLKADDR_NDC_NIX1_TX); + if (ret) + dev_err(rvu->dev, + "%s: Not able to unlock cachelines\n", __func__); + } + return -EBUSY; + } return 0; } diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npa.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npa.c index 70bd036ed76e4..4f5ca5ab13a40 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npa.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npa.c @@ -4,7 +4,7 @@ * Copyright (C) 2018 Marvell. * */ - +#include #include #include @@ -42,9 +42,18 @@ static int npa_aq_enqueue_wait(struct rvu *rvu, struct rvu_block *block, return -EBUSY; } - if (result->compcode != NPA_AQ_COMP_GOOD) + if (result->compcode != NPA_AQ_COMP_GOOD) { /* TODO: Replace this with some error code */ + if (result->compcode == NPA_AQ_COMP_CTX_FAULT || + result->compcode == NPA_AQ_COMP_LOCKERR || + result->compcode == NPA_AQ_COMP_CTX_POISON) { + if (rvu_ndc_fix_locked_cacheline(rvu, BLKADDR_NDC_NPA0)) + dev_err(rvu->dev, + "%s: Not able to unlock cachelines\n", __func__); + } + return -EBUSY; + } return 0; } @@ -545,3 +554,48 @@ void rvu_npa_lf_teardown(struct rvu *rvu, u16 pcifunc, int npalf) npa_ctx_free(rvu, pfvf); } + +/* Due to an Hardware errata, in some corner cases, AQ context lock + * operations can result in a NDC way getting into an illegal state + * of not valid but locked. + * + * This API solves the problem by clearing the lock bit of the NDC block. + * The operation needs to be done for each line of all the NDC banks. + */ +int rvu_ndc_fix_locked_cacheline(struct rvu *rvu, int blkaddr) +{ + int bank, max_bank, line, max_line, err; + u64 reg, ndc_af_const; + + /* Set the ENABLE bit(63) to '0' */ + reg = rvu_read64(rvu, blkaddr, NDC_AF_CAMS_RD_INTERVAL); + rvu_write64(rvu, blkaddr, NDC_AF_CAMS_RD_INTERVAL, reg & GENMASK_ULL(62, 0)); + + /* Poll until the BUSY bits(47:32) are set to '0' */ + err = rvu_poll_reg(rvu, blkaddr, NDC_AF_CAMS_RD_INTERVAL, GENMASK_ULL(47, 32), true); + if (err) { + dev_err(rvu->dev, "Timed out while polling for NDC CAM busy bits.\n"); + return err; + } + + ndc_af_const = rvu_read64(rvu, blkaddr, NDC_AF_CONST); + max_bank = FIELD_GET(NDC_AF_BANK_MASK, ndc_af_const); + max_line = FIELD_GET(NDC_AF_BANK_LINE_MASK, ndc_af_const); + for (bank = 0; bank < max_bank; bank++) { + for (line = 0; line < max_line; line++) { + /* Check if 'cache line valid bit(63)' is not set + * but 'cache line lock bit(60)' is set and on + * success, reset the lock bit(60). + */ + reg = rvu_read64(rvu, blkaddr, + NDC_AF_BANKX_LINEX_METADATA(bank, line)); + if (!(reg & BIT_ULL(63)) && (reg & BIT_ULL(60))) { + rvu_write64(rvu, blkaddr, + NDC_AF_BANKX_LINEX_METADATA(bank, line), + reg & ~BIT_ULL(60)); + } + } + } + + return 0; +} diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_reg.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu_reg.h index 1729b22580ce1..7007f0b8e6591 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_reg.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_reg.h @@ -694,6 +694,7 @@ #define NDC_AF_INTR_ENA_W1S (0x00068) #define NDC_AF_INTR_ENA_W1C (0x00070) #define NDC_AF_ACTIVE_PC (0x00078) +#define NDC_AF_CAMS_RD_INTERVAL (0x00080) #define NDC_AF_BP_TEST_ENABLE (0x001F8) #define NDC_AF_BP_TEST(a) (0x00200 | (a) << 3) #define NDC_AF_BLK_RST (0x002F0) @@ -709,6 +710,8 @@ (0x00F00 | (a) << 5 | (b) << 4) #define NDC_AF_BANKX_HIT_PC(a) (0x01000 | (a) << 3) #define NDC_AF_BANKX_MISS_PC(a) (0x01100 | (a) << 3) +#define NDC_AF_BANKX_LINEX_METADATA(a, b) \ + (0x10000 | (a) << 12 | (b) << 3) /* LBK */ #define LBK_CONST (0x10ull) From cdd28833100c18a469c85a1cc3de9f6bbbe6caa0 Mon Sep 17 00:00:00 2001 From: Daniel Machon Date: Tue, 7 Mar 2023 12:21:03 +0100 Subject: [PATCH 757/765] net: microchip: sparx5: fix deletion of existing DSCP mappings Fix deletion of existing DSCP mappings in the APP table. Adding and deleting DSCP entries are replicated per-port, since the mapping table is global for all ports in the chip. Whenever a mapping for a DSCP value already exists, the old mapping is deleted first. However, it is only deleted for the specified port. Fix this by calling sparx5_dcb_ieee_delapp() instead of dcb_ieee_delapp() as it ought to be. Reproduce: // Map and remap DSCP value 63 $ dcb app add dev eth0 dscp-prio 63:1 $ dcb app add dev eth0 dscp-prio 63:2 $ dcb app show dev eth0 dscp-prio dscp-prio 63:2 $ dcb app show dev eth1 dscp-prio dscp-prio 63:1 63:2 <-- 63:1 should not be there Fixes: 8dcf69a64118 ("net: microchip: sparx5: add support for offloading dscp table") Signed-off-by: Daniel Machon Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- .../ethernet/microchip/sparx5/sparx5_dcb.c | 32 +++++++++---------- 1 file changed, 16 insertions(+), 16 deletions(-) diff --git a/drivers/net/ethernet/microchip/sparx5/sparx5_dcb.c b/drivers/net/ethernet/microchip/sparx5/sparx5_dcb.c index 871a3e62f8527..2d763664dcda1 100644 --- a/drivers/net/ethernet/microchip/sparx5/sparx5_dcb.c +++ b/drivers/net/ethernet/microchip/sparx5/sparx5_dcb.c @@ -249,6 +249,21 @@ static int sparx5_dcb_ieee_dscp_setdel(struct net_device *dev, return 0; } +static int sparx5_dcb_ieee_delapp(struct net_device *dev, struct dcb_app *app) +{ + int err; + + if (app->selector == IEEE_8021QAZ_APP_SEL_DSCP) + err = sparx5_dcb_ieee_dscp_setdel(dev, app, dcb_ieee_delapp); + else + err = dcb_ieee_delapp(dev, app); + + if (err < 0) + return err; + + return sparx5_dcb_app_update(dev); +} + static int sparx5_dcb_ieee_setapp(struct net_device *dev, struct dcb_app *app) { struct dcb_app app_itr; @@ -264,7 +279,7 @@ static int sparx5_dcb_ieee_setapp(struct net_device *dev, struct dcb_app *app) if (prio) { app_itr = *app; app_itr.priority = prio; - dcb_ieee_delapp(dev, &app_itr); + sparx5_dcb_ieee_delapp(dev, &app_itr); } if (app->selector == IEEE_8021QAZ_APP_SEL_DSCP) @@ -281,21 +296,6 @@ static int sparx5_dcb_ieee_setapp(struct net_device *dev, struct dcb_app *app) return err; } -static int sparx5_dcb_ieee_delapp(struct net_device *dev, struct dcb_app *app) -{ - int err; - - if (app->selector == IEEE_8021QAZ_APP_SEL_DSCP) - err = sparx5_dcb_ieee_dscp_setdel(dev, app, dcb_ieee_delapp); - else - err = dcb_ieee_delapp(dev, app); - - if (err < 0) - return err; - - return sparx5_dcb_app_update(dev); -} - static int sparx5_dcb_setapptrust(struct net_device *dev, u8 *selectors, int nselectors) { From 7fef099702527c3b2c5234a2ea6a24411485a13a Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 7 Mar 2023 13:06:29 -0800 Subject: [PATCH 758/765] x86/resctl: fix scheduler confusion with 'current' The implementation of 'current' on x86 is very intentionally special: it is a very common thing to look up, and it uses 'this_cpu_read_stable()' to get the current thread pointer efficiently from per-cpu storage. And the keyword in there is 'stable': the current thread pointer never changes as far as a single thread is concerned. Even if when a thread is preempted, or moved to another CPU, or even across an explicit call 'schedule()' that thread will still have the same value for 'current'. It is, after all, the kernel base pointer to thread-local storage. That's why it's stable to begin with, but it's also why it's important enough that we have that special 'this_cpu_read_stable()' access for it. So this is all done very intentionally to allow the compiler to treat 'current' as a value that never visibly changes, so that the compiler can do CSE and combine multiple different 'current' accesses into one. However, there is obviously one very special situation when the currently running thread does actually change: inside the scheduler itself. So the scheduler code paths are special, and do not have a 'current' thread at all. Instead there are _two_ threads: the previous and the next thread - typically called 'prev' and 'next' (or prev_p/next_p) internally. So this is all actually quite straightforward and simple, and not all that complicated. Except for when you then have special code that is run in scheduler context, that code then has to be aware that 'current' isn't really a valid thing. Did you mean 'prev'? Did you mean 'next'? In fact, even if then look at the code, and you use 'current' after the new value has been assigned to the percpu variable, we have explicitly told the compiler that 'current' is magical and always stable. So the compiler is quite free to use an older (or newer) value of 'current', and the actual assignment to the percpu storage is not relevant even if it might look that way. Which is exactly what happened in the resctl code, that blithely used 'current' in '__resctrl_sched_in()' when it really wanted the new process state (as implied by the name: we're scheduling 'into' that new resctl state). And clang would end up just using the old thread pointer value at least in some configurations. This could have happened with gcc too, and purely depends on random compiler details. Clang just seems to have been more aggressive about moving the read of the per-cpu current_task pointer around. The fix is trivial: just make the resctl code adhere to the scheduler rules of using the prev/next thread pointer explicitly, instead of using 'current' in a situation where it just wasn't valid. That same code is then also used outside of the scheduler context (when a thread resctl state is explicitly changed), and then we will just pass in 'current' as that pointer, of course. There is no ambiguity in that case. The fix may be trivial, but noticing and figuring out what went wrong was not. The credit for that goes to Stephane Eranian. Reported-by: Stephane Eranian Link: https://lore.kernel.org/lkml/20230303231133.1486085-1-eranian@google.com/ Link: https://lore.kernel.org/lkml/alpine.LFD.2.01.0908011214330.3304@localhost.localdomain/ Reviewed-by: Nick Desaulniers Tested-by: Tony Luck Tested-by: Stephane Eranian Tested-by: Babu Moger Cc: stable@kernel.org Signed-off-by: Linus Torvalds --- arch/x86/include/asm/resctrl.h | 12 ++++++------ arch/x86/kernel/cpu/resctrl/rdtgroup.c | 4 ++-- arch/x86/kernel/process_32.c | 2 +- arch/x86/kernel/process_64.c | 2 +- 4 files changed, 10 insertions(+), 10 deletions(-) diff --git a/arch/x86/include/asm/resctrl.h b/arch/x86/include/asm/resctrl.h index 52788f79786fa..255a78d9d9067 100644 --- a/arch/x86/include/asm/resctrl.h +++ b/arch/x86/include/asm/resctrl.h @@ -49,7 +49,7 @@ DECLARE_STATIC_KEY_FALSE(rdt_mon_enable_key); * simple as possible. * Must be called with preemption disabled. */ -static void __resctrl_sched_in(void) +static inline void __resctrl_sched_in(struct task_struct *tsk) { struct resctrl_pqr_state *state = this_cpu_ptr(&pqr_state); u32 closid = state->default_closid; @@ -61,13 +61,13 @@ static void __resctrl_sched_in(void) * Else use the closid/rmid assigned to this cpu. */ if (static_branch_likely(&rdt_alloc_enable_key)) { - tmp = READ_ONCE(current->closid); + tmp = READ_ONCE(tsk->closid); if (tmp) closid = tmp; } if (static_branch_likely(&rdt_mon_enable_key)) { - tmp = READ_ONCE(current->rmid); + tmp = READ_ONCE(tsk->rmid); if (tmp) rmid = tmp; } @@ -88,17 +88,17 @@ static inline unsigned int resctrl_arch_round_mon_val(unsigned int val) return val * scale; } -static inline void resctrl_sched_in(void) +static inline void resctrl_sched_in(struct task_struct *tsk) { if (static_branch_likely(&rdt_enable_key)) - __resctrl_sched_in(); + __resctrl_sched_in(tsk); } void resctrl_cpu_detect(struct cpuinfo_x86 *c); #else -static inline void resctrl_sched_in(void) {} +static inline void resctrl_sched_in(struct task_struct *tsk) {} static inline void resctrl_cpu_detect(struct cpuinfo_x86 *c) {} #endif /* CONFIG_X86_CPU_RESCTRL */ diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c index e2c1599d1b373..884b6e9a7e31c 100644 --- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c +++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c @@ -314,7 +314,7 @@ static void update_cpu_closid_rmid(void *info) * executing task might have its own closid selected. Just reuse * the context switch code. */ - resctrl_sched_in(); + resctrl_sched_in(current); } /* @@ -530,7 +530,7 @@ static void _update_task_closid_rmid(void *task) * Otherwise, the MSR is updated when the task is scheduled in. */ if (task == current) - resctrl_sched_in(); + resctrl_sched_in(task); } static void update_task_closid_rmid(struct task_struct *t) diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 470c128759eab..708c87b88cc15 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -212,7 +212,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) switch_fpu_finish(); /* Load the Intel cache allocation PQR MSR. */ - resctrl_sched_in(); + resctrl_sched_in(next_p); return prev_p; } diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 4e34b3b68ebdc..bb65a68b4b499 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -656,7 +656,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) } /* Load the Intel cache allocation PQR MSR. */ - resctrl_sched_in(); + resctrl_sched_in(next_p); return prev_p; } From c8b8a3c601f2cfad25ab5ce5b04df700048aef6e Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 7 Mar 2023 17:54:11 +0200 Subject: [PATCH 759/765] net: dsa: mt7530: permit port 5 to work without port 6 on MT7621 SoC MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The MT7530 switch from the MT7621 SoC has 2 ports which can be set up as internal: port 5 and 6. Arınç reports that the GMAC1 attached to port 5 receives corrupted frames, unless port 6 (attached to GMAC0) has been brought up by the driver. This is true regardless of whether port 5 is used as a user port or as a CPU port (carrying DSA tags). Offline debugging (blind for me) which began in the linked thread showed experimentally that the configuration done by the driver for port 6 contains a step which is needed by port 5 as well - the write to CORE_GSWPLL_GRP2 (note that I've no idea as to what it does, apart from the comment "Set core clock into 500Mhz"). Prints put by Arınç show that the reset value of CORE_GSWPLL_GRP2 is RG_GSWPLL_POSDIV_500M(1) | RG_GSWPLL_FBKDIV_500M(40) (0x128), both on the MCM MT7530 from the MT7621 SoC, as well as on the standalone MT7530 from MT7623NI Bananapi BPI-R2. Apparently, port 5 on the standalone MT7530 can work under both values of the register, while on the MT7621 SoC it cannot. The call path that triggers the register write is: mt753x_phylink_mac_config() for port 6 -> mt753x_pad_setup() -> mt7530_pad_clk_setup() so this fully explains the behavior noticed by Arınç, that bringing port 6 up is necessary. The simplest fix for the problem is to extract the register writes which are needed for both port 5 and 6 into a common mt7530_pll_setup() function, which is called at mt7530_setup() time, immediately after switch reset. We can argue that this mirrors the code layout introduced in mt7531_setup() by commit 42bc4fafe359 ("net: mt7531: only do PLL once after the reset"), in that the PLL setup has the exact same positioning, and further work to consolidate the separate setup() functions is not hindered. Testing confirms that: - the slight reordering of writes to MT7530_P6ECR and to CORE_GSWPLL_GRP1 / CORE_GSWPLL_GRP2 introduced by this change does not appear to cause problems for the operation of port 6 on MT7621 and on MT7623 (where port 5 also always worked) - packets sent through port 5 are not corrupted anymore, regardless of whether port 6 is enabled by phylink or not (or even present in the device tree) My algorithm for determining the Fixes: tag is as follows. Testing shows that some logic from mt7530_pad_clk_setup() is needed even for port 5. Prior to commit ca366d6c889b ("net: dsa: mt7530: Convert to PHYLINK API"), a call did exist for all phy_is_pseudo_fixed_link() ports - so port 5 included. That commit replaced it with a temporary "Port 5 is not supported!" comment, and the following commit 38f790a80560 ("net: dsa: mt7530: Add support for port 5") replaced that comment with a configuration procedure in mt7530_setup_port5() which was insufficient for port 5 to work. I'm laying the blame on the patch that claimed support for port 5, although one would have also needed the change from commit c3b8e07909db ("net: dsa: mt7530: setup core clock even in TRGMII mode") for the write to be performed completely independently from port 6's configuration. Thanks go to Arınç for describing the problem, for debugging and for testing. Reported-by: Arınç ÜNAL Link: https://lore.kernel.org/netdev/f297c2c4-6e7c-57ac-2394-f6025d309b9d@arinc9.com/ Fixes: 38f790a80560 ("net: dsa: mt7530: Add support for port 5") Signed-off-by: Vladimir Oltean Tested-by: Arınç ÜNAL Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20230307155411.868573-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/mt7530.c | 35 ++++++++++++++++++++--------------- 1 file changed, 20 insertions(+), 15 deletions(-) diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c index 3a15015bc409e..a508402c4ecbf 100644 --- a/drivers/net/dsa/mt7530.c +++ b/drivers/net/dsa/mt7530.c @@ -393,6 +393,24 @@ mt7530_fdb_write(struct mt7530_priv *priv, u16 vid, mt7530_write(priv, MT7530_ATA1 + (i * 4), reg[i]); } +/* Set up switch core clock for MT7530 */ +static void mt7530_pll_setup(struct mt7530_priv *priv) +{ + /* Disable PLL */ + core_write(priv, CORE_GSWPLL_GRP1, 0); + + /* Set core clock into 500Mhz */ + core_write(priv, CORE_GSWPLL_GRP2, + RG_GSWPLL_POSDIV_500M(1) | + RG_GSWPLL_FBKDIV_500M(25)); + + /* Enable PLL */ + core_write(priv, CORE_GSWPLL_GRP1, + RG_GSWPLL_EN_PRE | + RG_GSWPLL_POSDIV_200M(2) | + RG_GSWPLL_FBKDIV_200M(32)); +} + /* Setup TX circuit including relevant PAD and driving */ static int mt7530_pad_clk_setup(struct dsa_switch *ds, phy_interface_t interface) @@ -453,21 +471,6 @@ mt7530_pad_clk_setup(struct dsa_switch *ds, phy_interface_t interface) core_clear(priv, CORE_TRGMII_GSW_CLK_CG, REG_GSWCK_EN | REG_TRGMIICK_EN); - /* Setup core clock for MT7530 */ - /* Disable PLL */ - core_write(priv, CORE_GSWPLL_GRP1, 0); - - /* Set core clock into 500Mhz */ - core_write(priv, CORE_GSWPLL_GRP2, - RG_GSWPLL_POSDIV_500M(1) | - RG_GSWPLL_FBKDIV_500M(25)); - - /* Enable PLL */ - core_write(priv, CORE_GSWPLL_GRP1, - RG_GSWPLL_EN_PRE | - RG_GSWPLL_POSDIV_200M(2) | - RG_GSWPLL_FBKDIV_200M(32)); - /* Setup the MT7530 TRGMII Tx Clock */ core_write(priv, CORE_PLL_GROUP5, RG_LCDDS_PCW_NCPO1(ncpo1)); core_write(priv, CORE_PLL_GROUP6, RG_LCDDS_PCW_NCPO0(0)); @@ -2196,6 +2199,8 @@ mt7530_setup(struct dsa_switch *ds) SYS_CTRL_PHY_RST | SYS_CTRL_SW_RST | SYS_CTRL_REG_RST); + mt7530_pll_setup(priv); + /* Enable Port 6 only; P5 as GMAC5 which currently is not supported */ val = mt7530_read(priv, MT7530_MHWTRAP); val &= ~MHWTRAP_P6_DIS & ~MHWTRAP_PHY_ACCESS; From 8f14820801042c221bb9fe51643a2585cac5dec2 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 7 Mar 2023 09:19:30 -0800 Subject: [PATCH 760/765] eth: fealnx: bring back this old driver This reverts commit d5e2d038dbece821f1af57acbeded3aa9a1832c1. We have a report of this chip being used on a SURECOM EP-320X-S 100/10M Ethernet PCI Adapter which could still have been purchased in some parts of the world 3 years ago. Cc: stable@vger.kernel.org Link: https://bugzilla.kernel.org/show_bug.cgi?id=217151 Fixes: d5e2d038dbec ("eth: fealnx: delete the driver for Myson MTD-800") Link: https://lore.kernel.org/r/20230307171930.4008454-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- arch/mips/configs/mtx1_defconfig | 1 + arch/powerpc/configs/ppc6xx_defconfig | 1 + drivers/net/ethernet/Kconfig | 10 + drivers/net/ethernet/Makefile | 1 + drivers/net/ethernet/fealnx.c | 1953 +++++++++++++++++++++++++ 5 files changed, 1966 insertions(+) create mode 100644 drivers/net/ethernet/fealnx.c diff --git a/arch/mips/configs/mtx1_defconfig b/arch/mips/configs/mtx1_defconfig index 89a1511d2ee47..edf9634aa8ee1 100644 --- a/arch/mips/configs/mtx1_defconfig +++ b/arch/mips/configs/mtx1_defconfig @@ -284,6 +284,7 @@ CONFIG_IXGB=m CONFIG_SKGE=m CONFIG_SKY2=m CONFIG_MYRI10GE=m +CONFIG_FEALNX=m CONFIG_NATSEMI=m CONFIG_NS83820=m CONFIG_S2IO=m diff --git a/arch/powerpc/configs/ppc6xx_defconfig b/arch/powerpc/configs/ppc6xx_defconfig index 1102582779599..f73c98be56c8f 100644 --- a/arch/powerpc/configs/ppc6xx_defconfig +++ b/arch/powerpc/configs/ppc6xx_defconfig @@ -461,6 +461,7 @@ CONFIG_MV643XX_ETH=m CONFIG_SKGE=m CONFIG_SKY2=m CONFIG_MYRI10GE=m +CONFIG_FEALNX=m CONFIG_NATSEMI=m CONFIG_NS83820=m CONFIG_PCMCIA_AXNET=m diff --git a/drivers/net/ethernet/Kconfig b/drivers/net/ethernet/Kconfig index 323ec56e8a74c..1917da7841919 100644 --- a/drivers/net/ethernet/Kconfig +++ b/drivers/net/ethernet/Kconfig @@ -132,6 +132,16 @@ source "drivers/net/ethernet/mscc/Kconfig" source "drivers/net/ethernet/microsoft/Kconfig" source "drivers/net/ethernet/moxa/Kconfig" source "drivers/net/ethernet/myricom/Kconfig" + +config FEALNX + tristate "Myson MTD-8xx PCI Ethernet support" + depends on PCI + select CRC32 + select MII + help + Say Y here to support the Myson MTD-800 family of PCI-based Ethernet + cards. + source "drivers/net/ethernet/ni/Kconfig" source "drivers/net/ethernet/natsemi/Kconfig" source "drivers/net/ethernet/neterion/Kconfig" diff --git a/drivers/net/ethernet/Makefile b/drivers/net/ethernet/Makefile index 2fedbaa545eb1..0d872d4efcd10 100644 --- a/drivers/net/ethernet/Makefile +++ b/drivers/net/ethernet/Makefile @@ -64,6 +64,7 @@ obj-$(CONFIG_NET_VENDOR_MICROCHIP) += microchip/ obj-$(CONFIG_NET_VENDOR_MICROSEMI) += mscc/ obj-$(CONFIG_NET_VENDOR_MOXART) += moxa/ obj-$(CONFIG_NET_VENDOR_MYRI) += myricom/ +obj-$(CONFIG_FEALNX) += fealnx.o obj-$(CONFIG_NET_VENDOR_NATSEMI) += natsemi/ obj-$(CONFIG_NET_VENDOR_NETERION) += neterion/ obj-$(CONFIG_NET_VENDOR_NETRONOME) += netronome/ diff --git a/drivers/net/ethernet/fealnx.c b/drivers/net/ethernet/fealnx.c new file mode 100644 index 0000000000000..ed18450fd2cc5 --- /dev/null +++ b/drivers/net/ethernet/fealnx.c @@ -0,0 +1,1953 @@ +/* + Written 1998-2000 by Donald Becker. + + This software may be used and distributed according to the terms of + the GNU General Public License (GPL), incorporated herein by reference. + Drivers based on or derived from this code fall under the GPL and must + retain the authorship, copyright and license notice. This file is not + a complete program and may only be used when the entire operating + system is licensed under the GPL. + + The author may be reached as becker@scyld.com, or C/O + Scyld Computing Corporation + 410 Severn Ave., Suite 210 + Annapolis MD 21403 + + Support information and updates available at + http://www.scyld.com/network/pci-skeleton.html + + Linux kernel updates: + + Version 2.51, Nov 17, 2001 (jgarzik): + - Add ethtool support + - Replace some MII-related magic numbers with constants + +*/ + +#define DRV_NAME "fealnx" + +static int debug; /* 1-> print debug message */ +static int max_interrupt_work = 20; + +/* Maximum number of multicast addresses to filter (vs. Rx-all-multicast). */ +static int multicast_filter_limit = 32; + +/* Set the copy breakpoint for the copy-only-tiny-frames scheme. */ +/* Setting to > 1518 effectively disables this feature. */ +static int rx_copybreak; + +/* Used to pass the media type, etc. */ +/* Both 'options[]' and 'full_duplex[]' should exist for driver */ +/* interoperability. */ +/* The media type is usually passed in 'options[]'. */ +#define MAX_UNITS 8 /* More are supported, limit only on options */ +static int options[MAX_UNITS] = { -1, -1, -1, -1, -1, -1, -1, -1 }; +static int full_duplex[MAX_UNITS] = { -1, -1, -1, -1, -1, -1, -1, -1 }; + +/* Operational parameters that are set at compile time. */ +/* Keep the ring sizes a power of two for compile efficiency. */ +/* The compiler will convert '%'<2^N> into a bit mask. */ +/* Making the Tx ring too large decreases the effectiveness of channel */ +/* bonding and packet priority. */ +/* There are no ill effects from too-large receive rings. */ +// 88-12-9 modify, +// #define TX_RING_SIZE 16 +// #define RX_RING_SIZE 32 +#define TX_RING_SIZE 6 +#define RX_RING_SIZE 12 +#define TX_TOTAL_SIZE TX_RING_SIZE*sizeof(struct fealnx_desc) +#define RX_TOTAL_SIZE RX_RING_SIZE*sizeof(struct fealnx_desc) + +/* Operational parameters that usually are not changed. */ +/* Time in jiffies before concluding the transmitter is hung. */ +#define TX_TIMEOUT (2*HZ) + +#define PKT_BUF_SZ 1536 /* Size of each temporary Rx buffer. */ + + +/* Include files, designed to support most kernel versions 2.0.0 and later. */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include /* Processor type for cache alignment. */ +#include +#include +#include + +/* This driver was written to use PCI memory space, however some x86 systems + work only with I/O space accesses. */ +#ifndef __alpha__ +#define USE_IO_OPS +#endif + +/* Kernel compatibility defines, some common to David Hinds' PCMCIA package. */ +/* This is only in the support-all-kernels source code. */ + +#define RUN_AT(x) (jiffies + (x)) + +MODULE_AUTHOR("Myson or whoever"); +MODULE_DESCRIPTION("Myson MTD-8xx 100/10M Ethernet PCI Adapter Driver"); +MODULE_LICENSE("GPL"); +module_param(max_interrupt_work, int, 0); +module_param(debug, int, 0); +module_param(rx_copybreak, int, 0); +module_param(multicast_filter_limit, int, 0); +module_param_array(options, int, NULL, 0); +module_param_array(full_duplex, int, NULL, 0); +MODULE_PARM_DESC(max_interrupt_work, "fealnx maximum events handled per interrupt"); +MODULE_PARM_DESC(debug, "fealnx enable debugging (0-1)"); +MODULE_PARM_DESC(rx_copybreak, "fealnx copy breakpoint for copy-only-tiny-frames"); +MODULE_PARM_DESC(multicast_filter_limit, "fealnx maximum number of filtered multicast addresses"); +MODULE_PARM_DESC(options, "fealnx: Bits 0-3: media type, bit 17: full duplex"); +MODULE_PARM_DESC(full_duplex, "fealnx full duplex setting(s) (1)"); + +enum { + MIN_REGION_SIZE = 136, +}; + +/* A chip capabilities table, matching the entries in pci_tbl[] above. */ +enum chip_capability_flags { + HAS_MII_XCVR, + HAS_CHIP_XCVR, +}; + +/* 89/6/13 add, */ +/* for different PHY */ +enum phy_type_flags { + MysonPHY = 1, + AhdocPHY = 2, + SeeqPHY = 3, + MarvellPHY = 4, + Myson981 = 5, + LevelOnePHY = 6, + OtherPHY = 10, +}; + +struct chip_info { + char *chip_name; + int flags; +}; + +static const struct chip_info skel_netdrv_tbl[] = { + { "100/10M Ethernet PCI Adapter", HAS_MII_XCVR }, + { "100/10M Ethernet PCI Adapter", HAS_CHIP_XCVR }, + { "1000/100/10M Ethernet PCI Adapter", HAS_MII_XCVR }, +}; + +/* Offsets to the Command and Status Registers. */ +enum fealnx_offsets { + PAR0 = 0x0, /* physical address 0-3 */ + PAR1 = 0x04, /* physical address 4-5 */ + MAR0 = 0x08, /* multicast address 0-3 */ + MAR1 = 0x0C, /* multicast address 4-7 */ + FAR0 = 0x10, /* flow-control address 0-3 */ + FAR1 = 0x14, /* flow-control address 4-5 */ + TCRRCR = 0x18, /* receive & transmit configuration */ + BCR = 0x1C, /* bus command */ + TXPDR = 0x20, /* transmit polling demand */ + RXPDR = 0x24, /* receive polling demand */ + RXCWP = 0x28, /* receive current word pointer */ + TXLBA = 0x2C, /* transmit list base address */ + RXLBA = 0x30, /* receive list base address */ + ISR = 0x34, /* interrupt status */ + IMR = 0x38, /* interrupt mask */ + FTH = 0x3C, /* flow control high/low threshold */ + MANAGEMENT = 0x40, /* bootrom/eeprom and mii management */ + TALLY = 0x44, /* tally counters for crc and mpa */ + TSR = 0x48, /* tally counter for transmit status */ + BMCRSR = 0x4c, /* basic mode control and status */ + PHYIDENTIFIER = 0x50, /* phy identifier */ + ANARANLPAR = 0x54, /* auto-negotiation advertisement and link + partner ability */ + ANEROCR = 0x58, /* auto-negotiation expansion and pci conf. */ + BPREMRPSR = 0x5c, /* bypass & receive error mask and phy status */ +}; + +/* Bits in the interrupt status/enable registers. */ +/* The bits in the Intr Status/Enable registers, mostly interrupt sources. */ +enum intr_status_bits { + RFCON = 0x00020000, /* receive flow control xon packet */ + RFCOFF = 0x00010000, /* receive flow control xoff packet */ + LSCStatus = 0x00008000, /* link status change */ + ANCStatus = 0x00004000, /* autonegotiation completed */ + FBE = 0x00002000, /* fatal bus error */ + FBEMask = 0x00001800, /* mask bit12-11 */ + ParityErr = 0x00000000, /* parity error */ + TargetErr = 0x00001000, /* target abort */ + MasterErr = 0x00000800, /* master error */ + TUNF = 0x00000400, /* transmit underflow */ + ROVF = 0x00000200, /* receive overflow */ + ETI = 0x00000100, /* transmit early int */ + ERI = 0x00000080, /* receive early int */ + CNTOVF = 0x00000040, /* counter overflow */ + RBU = 0x00000020, /* receive buffer unavailable */ + TBU = 0x00000010, /* transmit buffer unavilable */ + TI = 0x00000008, /* transmit interrupt */ + RI = 0x00000004, /* receive interrupt */ + RxErr = 0x00000002, /* receive error */ +}; + +/* Bits in the NetworkConfig register, W for writing, R for reading */ +/* FIXME: some names are invented by me. Marked with (name?) */ +/* If you have docs and know bit names, please fix 'em */ +enum rx_mode_bits { + CR_W_ENH = 0x02000000, /* enhanced mode (name?) */ + CR_W_FD = 0x00100000, /* full duplex */ + CR_W_PS10 = 0x00080000, /* 10 mbit */ + CR_W_TXEN = 0x00040000, /* tx enable (name?) */ + CR_W_PS1000 = 0x00010000, /* 1000 mbit */ + /* CR_W_RXBURSTMASK= 0x00000e00, Im unsure about this */ + CR_W_RXMODEMASK = 0x000000e0, + CR_W_PROM = 0x00000080, /* promiscuous mode */ + CR_W_AB = 0x00000040, /* accept broadcast */ + CR_W_AM = 0x00000020, /* accept mutlicast */ + CR_W_ARP = 0x00000008, /* receive runt pkt */ + CR_W_ALP = 0x00000004, /* receive long pkt */ + CR_W_SEP = 0x00000002, /* receive error pkt */ + CR_W_RXEN = 0x00000001, /* rx enable (unicast?) (name?) */ + + CR_R_TXSTOP = 0x04000000, /* tx stopped (name?) */ + CR_R_FD = 0x00100000, /* full duplex detected */ + CR_R_PS10 = 0x00080000, /* 10 mbit detected */ + CR_R_RXSTOP = 0x00008000, /* rx stopped (name?) */ +}; + +/* The Tulip Rx and Tx buffer descriptors. */ +struct fealnx_desc { + s32 status; + s32 control; + u32 buffer; + u32 next_desc; + struct fealnx_desc *next_desc_logical; + struct sk_buff *skbuff; + u32 reserved1; + u32 reserved2; +}; + +/* Bits in network_desc.status */ +enum rx_desc_status_bits { + RXOWN = 0x80000000, /* own bit */ + FLNGMASK = 0x0fff0000, /* frame length */ + FLNGShift = 16, + MARSTATUS = 0x00004000, /* multicast address received */ + BARSTATUS = 0x00002000, /* broadcast address received */ + PHYSTATUS = 0x00001000, /* physical address received */ + RXFSD = 0x00000800, /* first descriptor */ + RXLSD = 0x00000400, /* last descriptor */ + ErrorSummary = 0x80, /* error summary */ + RUNTPKT = 0x40, /* runt packet received */ + LONGPKT = 0x20, /* long packet received */ + FAE = 0x10, /* frame align error */ + CRC = 0x08, /* crc error */ + RXER = 0x04, /* receive error */ +}; + +enum rx_desc_control_bits { + RXIC = 0x00800000, /* interrupt control */ + RBSShift = 0, +}; + +enum tx_desc_status_bits { + TXOWN = 0x80000000, /* own bit */ + JABTO = 0x00004000, /* jabber timeout */ + CSL = 0x00002000, /* carrier sense lost */ + LC = 0x00001000, /* late collision */ + EC = 0x00000800, /* excessive collision */ + UDF = 0x00000400, /* fifo underflow */ + DFR = 0x00000200, /* deferred */ + HF = 0x00000100, /* heartbeat fail */ + NCRMask = 0x000000ff, /* collision retry count */ + NCRShift = 0, +}; + +enum tx_desc_control_bits { + TXIC = 0x80000000, /* interrupt control */ + ETIControl = 0x40000000, /* early transmit interrupt */ + TXLD = 0x20000000, /* last descriptor */ + TXFD = 0x10000000, /* first descriptor */ + CRCEnable = 0x08000000, /* crc control */ + PADEnable = 0x04000000, /* padding control */ + RetryTxLC = 0x02000000, /* retry late collision */ + PKTSMask = 0x3ff800, /* packet size bit21-11 */ + PKTSShift = 11, + TBSMask = 0x000007ff, /* transmit buffer bit 10-0 */ + TBSShift = 0, +}; + +/* BootROM/EEPROM/MII Management Register */ +#define MASK_MIIR_MII_READ 0x00000000 +#define MASK_MIIR_MII_WRITE 0x00000008 +#define MASK_MIIR_MII_MDO 0x00000004 +#define MASK_MIIR_MII_MDI 0x00000002 +#define MASK_MIIR_MII_MDC 0x00000001 + +/* ST+OP+PHYAD+REGAD+TA */ +#define OP_READ 0x6000 /* ST:01+OP:10+PHYAD+REGAD+TA:Z0 */ +#define OP_WRITE 0x5002 /* ST:01+OP:01+PHYAD+REGAD+TA:10 */ + +/* ------------------------------------------------------------------------- */ +/* Constants for Myson PHY */ +/* ------------------------------------------------------------------------- */ +#define MysonPHYID 0xd0000302 +/* 89-7-27 add, (begin) */ +#define MysonPHYID0 0x0302 +#define StatusRegister 18 +#define SPEED100 0x0400 // bit10 +#define FULLMODE 0x0800 // bit11 +/* 89-7-27 add, (end) */ + +/* ------------------------------------------------------------------------- */ +/* Constants for Seeq 80225 PHY */ +/* ------------------------------------------------------------------------- */ +#define SeeqPHYID0 0x0016 + +#define MIIRegister18 18 +#define SPD_DET_100 0x80 +#define DPLX_DET_FULL 0x40 + +/* ------------------------------------------------------------------------- */ +/* Constants for Ahdoc 101 PHY */ +/* ------------------------------------------------------------------------- */ +#define AhdocPHYID0 0x0022 + +#define DiagnosticReg 18 +#define DPLX_FULL 0x0800 +#define Speed_100 0x0400 + +/* 89/6/13 add, */ +/* -------------------------------------------------------------------------- */ +/* Constants */ +/* -------------------------------------------------------------------------- */ +#define MarvellPHYID0 0x0141 +#define LevelOnePHYID0 0x0013 + +#define MII1000BaseTControlReg 9 +#define MII1000BaseTStatusReg 10 +#define SpecificReg 17 + +/* for 1000BaseT Control Register */ +#define PHYAbletoPerform1000FullDuplex 0x0200 +#define PHYAbletoPerform1000HalfDuplex 0x0100 +#define PHY1000AbilityMask 0x300 + +// for phy specific status register, marvell phy. +#define SpeedMask 0x0c000 +#define Speed_1000M 0x08000 +#define Speed_100M 0x4000 +#define Speed_10M 0 +#define Full_Duplex 0x2000 + +// 89/12/29 add, for phy specific status register, levelone phy, (begin) +#define LXT1000_100M 0x08000 +#define LXT1000_1000M 0x0c000 +#define LXT1000_Full 0x200 +// 89/12/29 add, for phy specific status register, levelone phy, (end) + +/* for 3-in-1 case, BMCRSR register */ +#define LinkIsUp2 0x00040000 + +/* for PHY */ +#define LinkIsUp 0x0004 + + +struct netdev_private { + /* Descriptor rings first for alignment. */ + struct fealnx_desc *rx_ring; + struct fealnx_desc *tx_ring; + + dma_addr_t rx_ring_dma; + dma_addr_t tx_ring_dma; + + spinlock_t lock; + + /* Media monitoring timer. */ + struct timer_list timer; + + /* Reset timer */ + struct timer_list reset_timer; + int reset_timer_armed; + unsigned long crvalue_sv; + unsigned long imrvalue_sv; + + /* Frequently used values: keep some adjacent for cache effect. */ + int flags; + struct pci_dev *pci_dev; + unsigned long crvalue; + unsigned long bcrvalue; + unsigned long imrvalue; + struct fealnx_desc *cur_rx; + struct fealnx_desc *lack_rxbuf; + int really_rx_count; + struct fealnx_desc *cur_tx; + struct fealnx_desc *cur_tx_copy; + int really_tx_count; + int free_tx_count; + unsigned int rx_buf_sz; /* Based on MTU+slack. */ + + /* These values are keep track of the transceiver/media in use. */ + unsigned int linkok; + unsigned int line_speed; + unsigned int duplexmode; + unsigned int default_port:4; /* Last dev->if_port value. */ + unsigned int PHYType; + + /* MII transceiver section. */ + int mii_cnt; /* MII device addresses. */ + unsigned char phys[2]; /* MII device addresses. */ + struct mii_if_info mii; + void __iomem *mem; +}; + + +static int mdio_read(struct net_device *dev, int phy_id, int location); +static void mdio_write(struct net_device *dev, int phy_id, int location, int value); +static int netdev_open(struct net_device *dev); +static void getlinktype(struct net_device *dev); +static void getlinkstatus(struct net_device *dev); +static void netdev_timer(struct timer_list *t); +static void reset_timer(struct timer_list *t); +static void fealnx_tx_timeout(struct net_device *dev, unsigned int txqueue); +static void init_ring(struct net_device *dev); +static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev); +static irqreturn_t intr_handler(int irq, void *dev_instance); +static int netdev_rx(struct net_device *dev); +static void set_rx_mode(struct net_device *dev); +static void __set_rx_mode(struct net_device *dev); +static struct net_device_stats *get_stats(struct net_device *dev); +static int mii_ioctl(struct net_device *dev, struct ifreq *rq, int cmd); +static const struct ethtool_ops netdev_ethtool_ops; +static int netdev_close(struct net_device *dev); +static void reset_rx_descriptors(struct net_device *dev); +static void reset_tx_descriptors(struct net_device *dev); + +static void stop_nic_rx(void __iomem *ioaddr, long crvalue) +{ + int delay = 0x1000; + iowrite32(crvalue & ~(CR_W_RXEN), ioaddr + TCRRCR); + while (--delay) { + if ( (ioread32(ioaddr + TCRRCR) & CR_R_RXSTOP) == CR_R_RXSTOP) + break; + } +} + + +static void stop_nic_rxtx(void __iomem *ioaddr, long crvalue) +{ + int delay = 0x1000; + iowrite32(crvalue & ~(CR_W_RXEN+CR_W_TXEN), ioaddr + TCRRCR); + while (--delay) { + if ( (ioread32(ioaddr + TCRRCR) & (CR_R_RXSTOP+CR_R_TXSTOP)) + == (CR_R_RXSTOP+CR_R_TXSTOP) ) + break; + } +} + +static const struct net_device_ops netdev_ops = { + .ndo_open = netdev_open, + .ndo_stop = netdev_close, + .ndo_start_xmit = start_tx, + .ndo_get_stats = get_stats, + .ndo_set_rx_mode = set_rx_mode, + .ndo_eth_ioctl = mii_ioctl, + .ndo_tx_timeout = fealnx_tx_timeout, + .ndo_set_mac_address = eth_mac_addr, + .ndo_validate_addr = eth_validate_addr, +}; + +static int fealnx_init_one(struct pci_dev *pdev, + const struct pci_device_id *ent) +{ + struct netdev_private *np; + int i, option, err, irq; + static int card_idx = -1; + char boardname[12]; + void __iomem *ioaddr; + unsigned long len; + unsigned int chip_id = ent->driver_data; + struct net_device *dev; + void *ring_space; + dma_addr_t ring_dma; + u8 addr[ETH_ALEN]; +#ifdef USE_IO_OPS + int bar = 0; +#else + int bar = 1; +#endif + + card_idx++; + sprintf(boardname, "fealnx%d", card_idx); + + option = card_idx < MAX_UNITS ? options[card_idx] : 0; + + i = pci_enable_device(pdev); + if (i) return i; + pci_set_master(pdev); + + len = pci_resource_len(pdev, bar); + if (len < MIN_REGION_SIZE) { + dev_err(&pdev->dev, + "region size %ld too small, aborting\n", len); + return -ENODEV; + } + + i = pci_request_regions(pdev, boardname); + if (i) + return i; + + irq = pdev->irq; + + ioaddr = pci_iomap(pdev, bar, len); + if (!ioaddr) { + err = -ENOMEM; + goto err_out_res; + } + + dev = alloc_etherdev(sizeof(struct netdev_private)); + if (!dev) { + err = -ENOMEM; + goto err_out_unmap; + } + SET_NETDEV_DEV(dev, &pdev->dev); + + /* read ethernet id */ + for (i = 0; i < 6; ++i) + addr[i] = ioread8(ioaddr + PAR0 + i); + eth_hw_addr_set(dev, addr); + + /* Reset the chip to erase previous misconfiguration. */ + iowrite32(0x00000001, ioaddr + BCR); + + /* Make certain the descriptor lists are aligned. */ + np = netdev_priv(dev); + np->mem = ioaddr; + spin_lock_init(&np->lock); + np->pci_dev = pdev; + np->flags = skel_netdrv_tbl[chip_id].flags; + pci_set_drvdata(pdev, dev); + np->mii.dev = dev; + np->mii.mdio_read = mdio_read; + np->mii.mdio_write = mdio_write; + np->mii.phy_id_mask = 0x1f; + np->mii.reg_num_mask = 0x1f; + + ring_space = dma_alloc_coherent(&pdev->dev, RX_TOTAL_SIZE, &ring_dma, + GFP_KERNEL); + if (!ring_space) { + err = -ENOMEM; + goto err_out_free_dev; + } + np->rx_ring = ring_space; + np->rx_ring_dma = ring_dma; + + ring_space = dma_alloc_coherent(&pdev->dev, TX_TOTAL_SIZE, &ring_dma, + GFP_KERNEL); + if (!ring_space) { + err = -ENOMEM; + goto err_out_free_rx; + } + np->tx_ring = ring_space; + np->tx_ring_dma = ring_dma; + + /* find the connected MII xcvrs */ + if (np->flags == HAS_MII_XCVR) { + int phy, phy_idx = 0; + + for (phy = 1; phy < 32 && phy_idx < ARRAY_SIZE(np->phys); + phy++) { + int mii_status = mdio_read(dev, phy, 1); + + if (mii_status != 0xffff && mii_status != 0x0000) { + np->phys[phy_idx++] = phy; + dev_info(&pdev->dev, + "MII PHY found at address %d, status " + "0x%4.4x.\n", phy, mii_status); + /* get phy type */ + { + unsigned int data; + + data = mdio_read(dev, np->phys[0], 2); + if (data == SeeqPHYID0) + np->PHYType = SeeqPHY; + else if (data == AhdocPHYID0) + np->PHYType = AhdocPHY; + else if (data == MarvellPHYID0) + np->PHYType = MarvellPHY; + else if (data == MysonPHYID0) + np->PHYType = Myson981; + else if (data == LevelOnePHYID0) + np->PHYType = LevelOnePHY; + else + np->PHYType = OtherPHY; + } + } + } + + np->mii_cnt = phy_idx; + if (phy_idx == 0) + dev_warn(&pdev->dev, + "MII PHY not found -- this device may " + "not operate correctly.\n"); + } else { + np->phys[0] = 32; +/* 89/6/23 add, (begin) */ + /* get phy type */ + if (ioread32(ioaddr + PHYIDENTIFIER) == MysonPHYID) + np->PHYType = MysonPHY; + else + np->PHYType = OtherPHY; + } + np->mii.phy_id = np->phys[0]; + + if (dev->mem_start) + option = dev->mem_start; + + /* The lower four bits are the media type. */ + if (option > 0) { + if (option & 0x200) + np->mii.full_duplex = 1; + np->default_port = option & 15; + } + + if (card_idx < MAX_UNITS && full_duplex[card_idx] > 0) + np->mii.full_duplex = full_duplex[card_idx]; + + if (np->mii.full_duplex) { + dev_info(&pdev->dev, "Media type forced to Full Duplex.\n"); +/* 89/6/13 add, (begin) */ +// if (np->PHYType==MarvellPHY) + if ((np->PHYType == MarvellPHY) || (np->PHYType == LevelOnePHY)) { + unsigned int data; + + data = mdio_read(dev, np->phys[0], 9); + data = (data & 0xfcff) | 0x0200; + mdio_write(dev, np->phys[0], 9, data); + } +/* 89/6/13 add, (end) */ + if (np->flags == HAS_MII_XCVR) + mdio_write(dev, np->phys[0], MII_ADVERTISE, ADVERTISE_FULL); + else + iowrite32(ADVERTISE_FULL, ioaddr + ANARANLPAR); + np->mii.force_media = 1; + } + + dev->netdev_ops = &netdev_ops; + dev->ethtool_ops = &netdev_ethtool_ops; + dev->watchdog_timeo = TX_TIMEOUT; + + err = register_netdev(dev); + if (err) + goto err_out_free_tx; + + printk(KERN_INFO "%s: %s at %p, %pM, IRQ %d.\n", + dev->name, skel_netdrv_tbl[chip_id].chip_name, ioaddr, + dev->dev_addr, irq); + + return 0; + +err_out_free_tx: + dma_free_coherent(&pdev->dev, TX_TOTAL_SIZE, np->tx_ring, + np->tx_ring_dma); +err_out_free_rx: + dma_free_coherent(&pdev->dev, RX_TOTAL_SIZE, np->rx_ring, + np->rx_ring_dma); +err_out_free_dev: + free_netdev(dev); +err_out_unmap: + pci_iounmap(pdev, ioaddr); +err_out_res: + pci_release_regions(pdev); + return err; +} + + +static void fealnx_remove_one(struct pci_dev *pdev) +{ + struct net_device *dev = pci_get_drvdata(pdev); + + if (dev) { + struct netdev_private *np = netdev_priv(dev); + + dma_free_coherent(&pdev->dev, TX_TOTAL_SIZE, np->tx_ring, + np->tx_ring_dma); + dma_free_coherent(&pdev->dev, RX_TOTAL_SIZE, np->rx_ring, + np->rx_ring_dma); + unregister_netdev(dev); + pci_iounmap(pdev, np->mem); + free_netdev(dev); + pci_release_regions(pdev); + } else + printk(KERN_ERR "fealnx: remove for unknown device\n"); +} + + +static ulong m80x_send_cmd_to_phy(void __iomem *miiport, int opcode, int phyad, int regad) +{ + ulong miir; + int i; + unsigned int mask, data; + + /* enable MII output */ + miir = (ulong) ioread32(miiport); + miir &= 0xfffffff0; + + miir |= MASK_MIIR_MII_WRITE + MASK_MIIR_MII_MDO; + + /* send 32 1's preamble */ + for (i = 0; i < 32; i++) { + /* low MDC; MDO is already high (miir) */ + miir &= ~MASK_MIIR_MII_MDC; + iowrite32(miir, miiport); + + /* high MDC */ + miir |= MASK_MIIR_MII_MDC; + iowrite32(miir, miiport); + } + + /* calculate ST+OP+PHYAD+REGAD+TA */ + data = opcode | (phyad << 7) | (regad << 2); + + /* sent out */ + mask = 0x8000; + while (mask) { + /* low MDC, prepare MDO */ + miir &= ~(MASK_MIIR_MII_MDC + MASK_MIIR_MII_MDO); + if (mask & data) + miir |= MASK_MIIR_MII_MDO; + + iowrite32(miir, miiport); + /* high MDC */ + miir |= MASK_MIIR_MII_MDC; + iowrite32(miir, miiport); + udelay(30); + + /* next */ + mask >>= 1; + if (mask == 0x2 && opcode == OP_READ) + miir &= ~MASK_MIIR_MII_WRITE; + } + return miir; +} + + +static int mdio_read(struct net_device *dev, int phyad, int regad) +{ + struct netdev_private *np = netdev_priv(dev); + void __iomem *miiport = np->mem + MANAGEMENT; + ulong miir; + unsigned int mask, data; + + miir = m80x_send_cmd_to_phy(miiport, OP_READ, phyad, regad); + + /* read data */ + mask = 0x8000; + data = 0; + while (mask) { + /* low MDC */ + miir &= ~MASK_MIIR_MII_MDC; + iowrite32(miir, miiport); + + /* read MDI */ + miir = ioread32(miiport); + if (miir & MASK_MIIR_MII_MDI) + data |= mask; + + /* high MDC, and wait */ + miir |= MASK_MIIR_MII_MDC; + iowrite32(miir, miiport); + udelay(30); + + /* next */ + mask >>= 1; + } + + /* low MDC */ + miir &= ~MASK_MIIR_MII_MDC; + iowrite32(miir, miiport); + + return data & 0xffff; +} + + +static void mdio_write(struct net_device *dev, int phyad, int regad, int data) +{ + struct netdev_private *np = netdev_priv(dev); + void __iomem *miiport = np->mem + MANAGEMENT; + ulong miir; + unsigned int mask; + + miir = m80x_send_cmd_to_phy(miiport, OP_WRITE, phyad, regad); + + /* write data */ + mask = 0x8000; + while (mask) { + /* low MDC, prepare MDO */ + miir &= ~(MASK_MIIR_MII_MDC + MASK_MIIR_MII_MDO); + if (mask & data) + miir |= MASK_MIIR_MII_MDO; + iowrite32(miir, miiport); + + /* high MDC */ + miir |= MASK_MIIR_MII_MDC; + iowrite32(miir, miiport); + + /* next */ + mask >>= 1; + } + + /* low MDC */ + miir &= ~MASK_MIIR_MII_MDC; + iowrite32(miir, miiport); +} + + +static int netdev_open(struct net_device *dev) +{ + struct netdev_private *np = netdev_priv(dev); + void __iomem *ioaddr = np->mem; + const int irq = np->pci_dev->irq; + int rc, i; + + iowrite32(0x00000001, ioaddr + BCR); /* Reset */ + + rc = request_irq(irq, intr_handler, IRQF_SHARED, dev->name, dev); + if (rc) + return -EAGAIN; + + for (i = 0; i < 3; i++) + iowrite16(((const unsigned short *)dev->dev_addr)[i], + ioaddr + PAR0 + i*2); + + init_ring(dev); + + iowrite32(np->rx_ring_dma, ioaddr + RXLBA); + iowrite32(np->tx_ring_dma, ioaddr + TXLBA); + + /* Initialize other registers. */ + /* Configure the PCI bus bursts and FIFO thresholds. + 486: Set 8 longword burst. + 586: no burst limit. + Burst length 5:3 + 0 0 0 1 + 0 0 1 4 + 0 1 0 8 + 0 1 1 16 + 1 0 0 32 + 1 0 1 64 + 1 1 0 128 + 1 1 1 256 + Wait the specified 50 PCI cycles after a reset by initializing + Tx and Rx queues and the address filter list. + FIXME (Ueimor): optimistic for alpha + posted writes ? */ + + np->bcrvalue = 0x10; /* little-endian, 8 burst length */ +#ifdef __BIG_ENDIAN + np->bcrvalue |= 0x04; /* big-endian */ +#endif + +#if defined(__i386__) && !defined(MODULE) && !defined(CONFIG_UML) + if (boot_cpu_data.x86 <= 4) + np->crvalue = 0xa00; + else +#endif + np->crvalue = 0xe00; /* rx 128 burst length */ + + +// 89/12/29 add, +// 90/1/16 modify, +// np->imrvalue=FBE|TUNF|CNTOVF|RBU|TI|RI; + np->imrvalue = TUNF | CNTOVF | RBU | TI | RI; + if (np->pci_dev->device == 0x891) { + np->bcrvalue |= 0x200; /* set PROG bit */ + np->crvalue |= CR_W_ENH; /* set enhanced bit */ + np->imrvalue |= ETI; + } + iowrite32(np->bcrvalue, ioaddr + BCR); + + if (dev->if_port == 0) + dev->if_port = np->default_port; + + iowrite32(0, ioaddr + RXPDR); +// 89/9/1 modify, +// np->crvalue = 0x00e40001; /* tx store and forward, tx/rx enable */ + np->crvalue |= 0x00e40001; /* tx store and forward, tx/rx enable */ + np->mii.full_duplex = np->mii.force_media; + getlinkstatus(dev); + if (np->linkok) + getlinktype(dev); + __set_rx_mode(dev); + + netif_start_queue(dev); + + /* Clear and Enable interrupts by setting the interrupt mask. */ + iowrite32(FBE | TUNF | CNTOVF | RBU | TI | RI, ioaddr + ISR); + iowrite32(np->imrvalue, ioaddr + IMR); + + if (debug) + printk(KERN_DEBUG "%s: Done netdev_open().\n", dev->name); + + /* Set the timer to check for link beat. */ + timer_setup(&np->timer, netdev_timer, 0); + np->timer.expires = RUN_AT(3 * HZ); + + /* timer handler */ + add_timer(&np->timer); + + timer_setup(&np->reset_timer, reset_timer, 0); + np->reset_timer_armed = 0; + return rc; +} + + +static void getlinkstatus(struct net_device *dev) +/* function: Routine will read MII Status Register to get link status. */ +/* input : dev... pointer to the adapter block. */ +/* output : none. */ +{ + struct netdev_private *np = netdev_priv(dev); + unsigned int i, DelayTime = 0x1000; + + np->linkok = 0; + + if (np->PHYType == MysonPHY) { + for (i = 0; i < DelayTime; ++i) { + if (ioread32(np->mem + BMCRSR) & LinkIsUp2) { + np->linkok = 1; + return; + } + udelay(100); + } + } else { + for (i = 0; i < DelayTime; ++i) { + if (mdio_read(dev, np->phys[0], MII_BMSR) & BMSR_LSTATUS) { + np->linkok = 1; + return; + } + udelay(100); + } + } +} + + +static void getlinktype(struct net_device *dev) +{ + struct netdev_private *np = netdev_priv(dev); + + if (np->PHYType == MysonPHY) { /* 3-in-1 case */ + if (ioread32(np->mem + TCRRCR) & CR_R_FD) + np->duplexmode = 2; /* full duplex */ + else + np->duplexmode = 1; /* half duplex */ + if (ioread32(np->mem + TCRRCR) & CR_R_PS10) + np->line_speed = 1; /* 10M */ + else + np->line_speed = 2; /* 100M */ + } else { + if (np->PHYType == SeeqPHY) { /* this PHY is SEEQ 80225 */ + unsigned int data; + + data = mdio_read(dev, np->phys[0], MIIRegister18); + if (data & SPD_DET_100) + np->line_speed = 2; /* 100M */ + else + np->line_speed = 1; /* 10M */ + if (data & DPLX_DET_FULL) + np->duplexmode = 2; /* full duplex mode */ + else + np->duplexmode = 1; /* half duplex mode */ + } else if (np->PHYType == AhdocPHY) { + unsigned int data; + + data = mdio_read(dev, np->phys[0], DiagnosticReg); + if (data & Speed_100) + np->line_speed = 2; /* 100M */ + else + np->line_speed = 1; /* 10M */ + if (data & DPLX_FULL) + np->duplexmode = 2; /* full duplex mode */ + else + np->duplexmode = 1; /* half duplex mode */ + } +/* 89/6/13 add, (begin) */ + else if (np->PHYType == MarvellPHY) { + unsigned int data; + + data = mdio_read(dev, np->phys[0], SpecificReg); + if (data & Full_Duplex) + np->duplexmode = 2; /* full duplex mode */ + else + np->duplexmode = 1; /* half duplex mode */ + data &= SpeedMask; + if (data == Speed_1000M) + np->line_speed = 3; /* 1000M */ + else if (data == Speed_100M) + np->line_speed = 2; /* 100M */ + else + np->line_speed = 1; /* 10M */ + } +/* 89/6/13 add, (end) */ +/* 89/7/27 add, (begin) */ + else if (np->PHYType == Myson981) { + unsigned int data; + + data = mdio_read(dev, np->phys[0], StatusRegister); + + if (data & SPEED100) + np->line_speed = 2; + else + np->line_speed = 1; + + if (data & FULLMODE) + np->duplexmode = 2; + else + np->duplexmode = 1; + } +/* 89/7/27 add, (end) */ +/* 89/12/29 add */ + else if (np->PHYType == LevelOnePHY) { + unsigned int data; + + data = mdio_read(dev, np->phys[0], SpecificReg); + if (data & LXT1000_Full) + np->duplexmode = 2; /* full duplex mode */ + else + np->duplexmode = 1; /* half duplex mode */ + data &= SpeedMask; + if (data == LXT1000_1000M) + np->line_speed = 3; /* 1000M */ + else if (data == LXT1000_100M) + np->line_speed = 2; /* 100M */ + else + np->line_speed = 1; /* 10M */ + } + np->crvalue &= (~CR_W_PS10) & (~CR_W_FD) & (~CR_W_PS1000); + if (np->line_speed == 1) + np->crvalue |= CR_W_PS10; + else if (np->line_speed == 3) + np->crvalue |= CR_W_PS1000; + if (np->duplexmode == 2) + np->crvalue |= CR_W_FD; + } +} + + +/* Take lock before calling this */ +static void allocate_rx_buffers(struct net_device *dev) +{ + struct netdev_private *np = netdev_priv(dev); + + /* allocate skb for rx buffers */ + while (np->really_rx_count != RX_RING_SIZE) { + struct sk_buff *skb; + + skb = netdev_alloc_skb(dev, np->rx_buf_sz); + if (skb == NULL) + break; /* Better luck next round. */ + + while (np->lack_rxbuf->skbuff) + np->lack_rxbuf = np->lack_rxbuf->next_desc_logical; + + np->lack_rxbuf->skbuff = skb; + np->lack_rxbuf->buffer = dma_map_single(&np->pci_dev->dev, + skb->data, + np->rx_buf_sz, + DMA_FROM_DEVICE); + np->lack_rxbuf->status = RXOWN; + ++np->really_rx_count; + } +} + + +static void netdev_timer(struct timer_list *t) +{ + struct netdev_private *np = from_timer(np, t, timer); + struct net_device *dev = np->mii.dev; + void __iomem *ioaddr = np->mem; + int old_crvalue = np->crvalue; + unsigned int old_linkok = np->linkok; + unsigned long flags; + + if (debug) + printk(KERN_DEBUG "%s: Media selection timer tick, status %8.8x " + "config %8.8x.\n", dev->name, ioread32(ioaddr + ISR), + ioread32(ioaddr + TCRRCR)); + + spin_lock_irqsave(&np->lock, flags); + + if (np->flags == HAS_MII_XCVR) { + getlinkstatus(dev); + if ((old_linkok == 0) && (np->linkok == 1)) { /* we need to detect the media type again */ + getlinktype(dev); + if (np->crvalue != old_crvalue) { + stop_nic_rxtx(ioaddr, np->crvalue); + iowrite32(np->crvalue, ioaddr + TCRRCR); + } + } + } + + allocate_rx_buffers(dev); + + spin_unlock_irqrestore(&np->lock, flags); + + np->timer.expires = RUN_AT(10 * HZ); + add_timer(&np->timer); +} + + +/* Take lock before calling */ +/* Reset chip and disable rx, tx and interrupts */ +static void reset_and_disable_rxtx(struct net_device *dev) +{ + struct netdev_private *np = netdev_priv(dev); + void __iomem *ioaddr = np->mem; + int delay=51; + + /* Reset the chip's Tx and Rx processes. */ + stop_nic_rxtx(ioaddr, 0); + + /* Disable interrupts by clearing the interrupt mask. */ + iowrite32(0, ioaddr + IMR); + + /* Reset the chip to erase previous misconfiguration. */ + iowrite32(0x00000001, ioaddr + BCR); + + /* Ueimor: wait for 50 PCI cycles (and flush posted writes btw). + We surely wait too long (address+data phase). Who cares? */ + while (--delay) { + ioread32(ioaddr + BCR); + rmb(); + } +} + + +/* Take lock before calling */ +/* Restore chip after reset */ +static void enable_rxtx(struct net_device *dev) +{ + struct netdev_private *np = netdev_priv(dev); + void __iomem *ioaddr = np->mem; + + reset_rx_descriptors(dev); + + iowrite32(np->tx_ring_dma + ((char*)np->cur_tx - (char*)np->tx_ring), + ioaddr + TXLBA); + iowrite32(np->rx_ring_dma + ((char*)np->cur_rx - (char*)np->rx_ring), + ioaddr + RXLBA); + + iowrite32(np->bcrvalue, ioaddr + BCR); + + iowrite32(0, ioaddr + RXPDR); + __set_rx_mode(dev); /* changes np->crvalue, writes it into TCRRCR */ + + /* Clear and Enable interrupts by setting the interrupt mask. */ + iowrite32(FBE | TUNF | CNTOVF | RBU | TI | RI, ioaddr + ISR); + iowrite32(np->imrvalue, ioaddr + IMR); + + iowrite32(0, ioaddr + TXPDR); +} + + +static void reset_timer(struct timer_list *t) +{ + struct netdev_private *np = from_timer(np, t, reset_timer); + struct net_device *dev = np->mii.dev; + unsigned long flags; + + printk(KERN_WARNING "%s: resetting tx and rx machinery\n", dev->name); + + spin_lock_irqsave(&np->lock, flags); + np->crvalue = np->crvalue_sv; + np->imrvalue = np->imrvalue_sv; + + reset_and_disable_rxtx(dev); + /* works for me without this: + reset_tx_descriptors(dev); */ + enable_rxtx(dev); + netif_start_queue(dev); /* FIXME: or netif_wake_queue(dev); ? */ + + np->reset_timer_armed = 0; + + spin_unlock_irqrestore(&np->lock, flags); +} + + +static void fealnx_tx_timeout(struct net_device *dev, unsigned int txqueue) +{ + struct netdev_private *np = netdev_priv(dev); + void __iomem *ioaddr = np->mem; + unsigned long flags; + int i; + + printk(KERN_WARNING + "%s: Transmit timed out, status %8.8x, resetting...\n", + dev->name, ioread32(ioaddr + ISR)); + + { + printk(KERN_DEBUG " Rx ring %p: ", np->rx_ring); + for (i = 0; i < RX_RING_SIZE; i++) + printk(KERN_CONT " %8.8x", + (unsigned int) np->rx_ring[i].status); + printk(KERN_CONT "\n"); + printk(KERN_DEBUG " Tx ring %p: ", np->tx_ring); + for (i = 0; i < TX_RING_SIZE; i++) + printk(KERN_CONT " %4.4x", np->tx_ring[i].status); + printk(KERN_CONT "\n"); + } + + spin_lock_irqsave(&np->lock, flags); + + reset_and_disable_rxtx(dev); + reset_tx_descriptors(dev); + enable_rxtx(dev); + + spin_unlock_irqrestore(&np->lock, flags); + + netif_trans_update(dev); /* prevent tx timeout */ + dev->stats.tx_errors++; + netif_wake_queue(dev); /* or .._start_.. ?? */ +} + + +/* Initialize the Rx and Tx rings, along with various 'dev' bits. */ +static void init_ring(struct net_device *dev) +{ + struct netdev_private *np = netdev_priv(dev); + int i; + + /* initialize rx variables */ + np->rx_buf_sz = (dev->mtu <= 1500 ? PKT_BUF_SZ : dev->mtu + 32); + np->cur_rx = &np->rx_ring[0]; + np->lack_rxbuf = np->rx_ring; + np->really_rx_count = 0; + + /* initial rx descriptors. */ + for (i = 0; i < RX_RING_SIZE; i++) { + np->rx_ring[i].status = 0; + np->rx_ring[i].control = np->rx_buf_sz << RBSShift; + np->rx_ring[i].next_desc = np->rx_ring_dma + + (i + 1)*sizeof(struct fealnx_desc); + np->rx_ring[i].next_desc_logical = &np->rx_ring[i + 1]; + np->rx_ring[i].skbuff = NULL; + } + + /* for the last rx descriptor */ + np->rx_ring[i - 1].next_desc = np->rx_ring_dma; + np->rx_ring[i - 1].next_desc_logical = np->rx_ring; + + /* allocate skb for rx buffers */ + for (i = 0; i < RX_RING_SIZE; i++) { + struct sk_buff *skb = netdev_alloc_skb(dev, np->rx_buf_sz); + + if (skb == NULL) { + np->lack_rxbuf = &np->rx_ring[i]; + break; + } + + ++np->really_rx_count; + np->rx_ring[i].skbuff = skb; + np->rx_ring[i].buffer = dma_map_single(&np->pci_dev->dev, + skb->data, + np->rx_buf_sz, + DMA_FROM_DEVICE); + np->rx_ring[i].status = RXOWN; + np->rx_ring[i].control |= RXIC; + } + + /* initialize tx variables */ + np->cur_tx = &np->tx_ring[0]; + np->cur_tx_copy = &np->tx_ring[0]; + np->really_tx_count = 0; + np->free_tx_count = TX_RING_SIZE; + + for (i = 0; i < TX_RING_SIZE; i++) { + np->tx_ring[i].status = 0; + /* do we need np->tx_ring[i].control = XXX; ?? */ + np->tx_ring[i].next_desc = np->tx_ring_dma + + (i + 1)*sizeof(struct fealnx_desc); + np->tx_ring[i].next_desc_logical = &np->tx_ring[i + 1]; + np->tx_ring[i].skbuff = NULL; + } + + /* for the last tx descriptor */ + np->tx_ring[i - 1].next_desc = np->tx_ring_dma; + np->tx_ring[i - 1].next_desc_logical = &np->tx_ring[0]; +} + + +static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev) +{ + struct netdev_private *np = netdev_priv(dev); + unsigned long flags; + + spin_lock_irqsave(&np->lock, flags); + + np->cur_tx_copy->skbuff = skb; + +#define one_buffer +#define BPT 1022 +#if defined(one_buffer) + np->cur_tx_copy->buffer = dma_map_single(&np->pci_dev->dev, skb->data, + skb->len, DMA_TO_DEVICE); + np->cur_tx_copy->control = TXIC | TXLD | TXFD | CRCEnable | PADEnable; + np->cur_tx_copy->control |= (skb->len << PKTSShift); /* pkt size */ + np->cur_tx_copy->control |= (skb->len << TBSShift); /* buffer size */ +// 89/12/29 add, + if (np->pci_dev->device == 0x891) + np->cur_tx_copy->control |= ETIControl | RetryTxLC; + np->cur_tx_copy->status = TXOWN; + np->cur_tx_copy = np->cur_tx_copy->next_desc_logical; + --np->free_tx_count; +#elif defined(two_buffer) + if (skb->len > BPT) { + struct fealnx_desc *next; + + /* for the first descriptor */ + np->cur_tx_copy->buffer = dma_map_single(&np->pci_dev->dev, + skb->data, BPT, + DMA_TO_DEVICE); + np->cur_tx_copy->control = TXIC | TXFD | CRCEnable | PADEnable; + np->cur_tx_copy->control |= (skb->len << PKTSShift); /* pkt size */ + np->cur_tx_copy->control |= (BPT << TBSShift); /* buffer size */ + + /* for the last descriptor */ + next = np->cur_tx_copy->next_desc_logical; + next->skbuff = skb; + next->control = TXIC | TXLD | CRCEnable | PADEnable; + next->control |= (skb->len << PKTSShift); /* pkt size */ + next->control |= ((skb->len - BPT) << TBSShift); /* buf size */ +// 89/12/29 add, + if (np->pci_dev->device == 0x891) + np->cur_tx_copy->control |= ETIControl | RetryTxLC; + next->buffer = dma_map_single(&ep->pci_dev->dev, + skb->data + BPT, skb->len - BPT, + DMA_TO_DEVICE); + + next->status = TXOWN; + np->cur_tx_copy->status = TXOWN; + + np->cur_tx_copy = next->next_desc_logical; + np->free_tx_count -= 2; + } else { + np->cur_tx_copy->buffer = dma_map_single(&np->pci_dev->dev, + skb->data, skb->len, + DMA_TO_DEVICE); + np->cur_tx_copy->control = TXIC | TXLD | TXFD | CRCEnable | PADEnable; + np->cur_tx_copy->control |= (skb->len << PKTSShift); /* pkt size */ + np->cur_tx_copy->control |= (skb->len << TBSShift); /* buffer size */ +// 89/12/29 add, + if (np->pci_dev->device == 0x891) + np->cur_tx_copy->control |= ETIControl | RetryTxLC; + np->cur_tx_copy->status = TXOWN; + np->cur_tx_copy = np->cur_tx_copy->next_desc_logical; + --np->free_tx_count; + } +#endif + + if (np->free_tx_count < 2) + netif_stop_queue(dev); + ++np->really_tx_count; + iowrite32(0, np->mem + TXPDR); + + spin_unlock_irqrestore(&np->lock, flags); + return NETDEV_TX_OK; +} + + +/* Take lock before calling */ +/* Chip probably hosed tx ring. Clean up. */ +static void reset_tx_descriptors(struct net_device *dev) +{ + struct netdev_private *np = netdev_priv(dev); + struct fealnx_desc *cur; + int i; + + /* initialize tx variables */ + np->cur_tx = &np->tx_ring[0]; + np->cur_tx_copy = &np->tx_ring[0]; + np->really_tx_count = 0; + np->free_tx_count = TX_RING_SIZE; + + for (i = 0; i < TX_RING_SIZE; i++) { + cur = &np->tx_ring[i]; + if (cur->skbuff) { + dma_unmap_single(&np->pci_dev->dev, cur->buffer, + cur->skbuff->len, DMA_TO_DEVICE); + dev_kfree_skb_any(cur->skbuff); + cur->skbuff = NULL; + } + cur->status = 0; + cur->control = 0; /* needed? */ + /* probably not needed. We do it for purely paranoid reasons */ + cur->next_desc = np->tx_ring_dma + + (i + 1)*sizeof(struct fealnx_desc); + cur->next_desc_logical = &np->tx_ring[i + 1]; + } + /* for the last tx descriptor */ + np->tx_ring[TX_RING_SIZE - 1].next_desc = np->tx_ring_dma; + np->tx_ring[TX_RING_SIZE - 1].next_desc_logical = &np->tx_ring[0]; +} + + +/* Take lock and stop rx before calling this */ +static void reset_rx_descriptors(struct net_device *dev) +{ + struct netdev_private *np = netdev_priv(dev); + struct fealnx_desc *cur = np->cur_rx; + int i; + + allocate_rx_buffers(dev); + + for (i = 0; i < RX_RING_SIZE; i++) { + if (cur->skbuff) + cur->status = RXOWN; + cur = cur->next_desc_logical; + } + + iowrite32(np->rx_ring_dma + ((char*)np->cur_rx - (char*)np->rx_ring), + np->mem + RXLBA); +} + + +/* The interrupt handler does all of the Rx thread work and cleans up + after the Tx thread. */ +static irqreturn_t intr_handler(int irq, void *dev_instance) +{ + struct net_device *dev = (struct net_device *) dev_instance; + struct netdev_private *np = netdev_priv(dev); + void __iomem *ioaddr = np->mem; + long boguscnt = max_interrupt_work; + unsigned int num_tx = 0; + int handled = 0; + + spin_lock(&np->lock); + + iowrite32(0, ioaddr + IMR); + + do { + u32 intr_status = ioread32(ioaddr + ISR); + + /* Acknowledge all of the current interrupt sources ASAP. */ + iowrite32(intr_status, ioaddr + ISR); + + if (debug) + printk(KERN_DEBUG "%s: Interrupt, status %4.4x.\n", dev->name, + intr_status); + + if (!(intr_status & np->imrvalue)) + break; + + handled = 1; + +// 90/1/16 delete, +// +// if (intr_status & FBE) +// { /* fatal error */ +// stop_nic_tx(ioaddr, 0); +// stop_nic_rx(ioaddr, 0); +// break; +// }; + + if (intr_status & TUNF) + iowrite32(0, ioaddr + TXPDR); + + if (intr_status & CNTOVF) { + /* missed pkts */ + dev->stats.rx_missed_errors += + ioread32(ioaddr + TALLY) & 0x7fff; + + /* crc error */ + dev->stats.rx_crc_errors += + (ioread32(ioaddr + TALLY) & 0x7fff0000) >> 16; + } + + if (intr_status & (RI | RBU)) { + if (intr_status & RI) + netdev_rx(dev); + else { + stop_nic_rx(ioaddr, np->crvalue); + reset_rx_descriptors(dev); + iowrite32(np->crvalue, ioaddr + TCRRCR); + } + } + + while (np->really_tx_count) { + long tx_status = np->cur_tx->status; + long tx_control = np->cur_tx->control; + + if (!(tx_control & TXLD)) { /* this pkt is combined by two tx descriptors */ + struct fealnx_desc *next; + + next = np->cur_tx->next_desc_logical; + tx_status = next->status; + tx_control = next->control; + } + + if (tx_status & TXOWN) + break; + + if (!(np->crvalue & CR_W_ENH)) { + if (tx_status & (CSL | LC | EC | UDF | HF)) { + dev->stats.tx_errors++; + if (tx_status & EC) + dev->stats.tx_aborted_errors++; + if (tx_status & CSL) + dev->stats.tx_carrier_errors++; + if (tx_status & LC) + dev->stats.tx_window_errors++; + if (tx_status & UDF) + dev->stats.tx_fifo_errors++; + if ((tx_status & HF) && np->mii.full_duplex == 0) + dev->stats.tx_heartbeat_errors++; + + } else { + dev->stats.tx_bytes += + ((tx_control & PKTSMask) >> PKTSShift); + + dev->stats.collisions += + ((tx_status & NCRMask) >> NCRShift); + dev->stats.tx_packets++; + } + } else { + dev->stats.tx_bytes += + ((tx_control & PKTSMask) >> PKTSShift); + dev->stats.tx_packets++; + } + + /* Free the original skb. */ + dma_unmap_single(&np->pci_dev->dev, + np->cur_tx->buffer, + np->cur_tx->skbuff->len, + DMA_TO_DEVICE); + dev_consume_skb_irq(np->cur_tx->skbuff); + np->cur_tx->skbuff = NULL; + --np->really_tx_count; + if (np->cur_tx->control & TXLD) { + np->cur_tx = np->cur_tx->next_desc_logical; + ++np->free_tx_count; + } else { + np->cur_tx = np->cur_tx->next_desc_logical; + np->cur_tx = np->cur_tx->next_desc_logical; + np->free_tx_count += 2; + } + num_tx++; + } /* end of for loop */ + + if (num_tx && np->free_tx_count >= 2) + netif_wake_queue(dev); + + /* read transmit status for enhanced mode only */ + if (np->crvalue & CR_W_ENH) { + long data; + + data = ioread32(ioaddr + TSR); + dev->stats.tx_errors += (data & 0xff000000) >> 24; + dev->stats.tx_aborted_errors += + (data & 0xff000000) >> 24; + dev->stats.tx_window_errors += + (data & 0x00ff0000) >> 16; + dev->stats.collisions += (data & 0x0000ffff); + } + + if (--boguscnt < 0) { + printk(KERN_WARNING "%s: Too much work at interrupt, " + "status=0x%4.4x.\n", dev->name, intr_status); + if (!np->reset_timer_armed) { + np->reset_timer_armed = 1; + np->reset_timer.expires = RUN_AT(HZ/2); + add_timer(&np->reset_timer); + stop_nic_rxtx(ioaddr, 0); + netif_stop_queue(dev); + /* or netif_tx_disable(dev); ?? */ + /* Prevent other paths from enabling tx,rx,intrs */ + np->crvalue_sv = np->crvalue; + np->imrvalue_sv = np->imrvalue; + np->crvalue &= ~(CR_W_TXEN | CR_W_RXEN); /* or simply = 0? */ + np->imrvalue = 0; + } + + break; + } + } while (1); + + /* read the tally counters */ + /* missed pkts */ + dev->stats.rx_missed_errors += ioread32(ioaddr + TALLY) & 0x7fff; + + /* crc error */ + dev->stats.rx_crc_errors += + (ioread32(ioaddr + TALLY) & 0x7fff0000) >> 16; + + if (debug) + printk(KERN_DEBUG "%s: exiting interrupt, status=%#4.4x.\n", + dev->name, ioread32(ioaddr + ISR)); + + iowrite32(np->imrvalue, ioaddr + IMR); + + spin_unlock(&np->lock); + + return IRQ_RETVAL(handled); +} + + +/* This routine is logically part of the interrupt handler, but separated + for clarity and better register allocation. */ +static int netdev_rx(struct net_device *dev) +{ + struct netdev_private *np = netdev_priv(dev); + void __iomem *ioaddr = np->mem; + + /* If EOP is set on the next entry, it's a new packet. Send it up. */ + while (!(np->cur_rx->status & RXOWN) && np->cur_rx->skbuff) { + s32 rx_status = np->cur_rx->status; + + if (np->really_rx_count == 0) + break; + + if (debug) + printk(KERN_DEBUG " netdev_rx() status was %8.8x.\n", rx_status); + + if ((!((rx_status & RXFSD) && (rx_status & RXLSD))) || + (rx_status & ErrorSummary)) { + if (rx_status & ErrorSummary) { /* there was a fatal error */ + if (debug) + printk(KERN_DEBUG + "%s: Receive error, Rx status %8.8x.\n", + dev->name, rx_status); + + dev->stats.rx_errors++; /* end of a packet. */ + if (rx_status & (LONGPKT | RUNTPKT)) + dev->stats.rx_length_errors++; + if (rx_status & RXER) + dev->stats.rx_frame_errors++; + if (rx_status & CRC) + dev->stats.rx_crc_errors++; + } else { + int need_to_reset = 0; + int desno = 0; + + if (rx_status & RXFSD) { /* this pkt is too long, over one rx buffer */ + struct fealnx_desc *cur; + + /* check this packet is received completely? */ + cur = np->cur_rx; + while (desno <= np->really_rx_count) { + ++desno; + if ((!(cur->status & RXOWN)) && + (cur->status & RXLSD)) + break; + /* goto next rx descriptor */ + cur = cur->next_desc_logical; + } + if (desno > np->really_rx_count) + need_to_reset = 1; + } else /* RXLSD did not find, something error */ + need_to_reset = 1; + + if (need_to_reset == 0) { + int i; + + dev->stats.rx_length_errors++; + + /* free all rx descriptors related this long pkt */ + for (i = 0; i < desno; ++i) { + if (!np->cur_rx->skbuff) { + printk(KERN_DEBUG + "%s: I'm scared\n", dev->name); + break; + } + np->cur_rx->status = RXOWN; + np->cur_rx = np->cur_rx->next_desc_logical; + } + continue; + } else { /* rx error, need to reset this chip */ + stop_nic_rx(ioaddr, np->crvalue); + reset_rx_descriptors(dev); + iowrite32(np->crvalue, ioaddr + TCRRCR); + } + break; /* exit the while loop */ + } + } else { /* this received pkt is ok */ + + struct sk_buff *skb; + /* Omit the four octet CRC from the length. */ + short pkt_len = ((rx_status & FLNGMASK) >> FLNGShift) - 4; + +#ifndef final_version + if (debug) + printk(KERN_DEBUG " netdev_rx() normal Rx pkt length %d" + " status %x.\n", pkt_len, rx_status); +#endif + + /* Check if the packet is long enough to accept without copying + to a minimally-sized skbuff. */ + if (pkt_len < rx_copybreak && + (skb = netdev_alloc_skb(dev, pkt_len + 2)) != NULL) { + skb_reserve(skb, 2); /* 16 byte align the IP header */ + dma_sync_single_for_cpu(&np->pci_dev->dev, + np->cur_rx->buffer, + np->rx_buf_sz, + DMA_FROM_DEVICE); + /* Call copy + cksum if available. */ + +#if ! defined(__alpha__) + skb_copy_to_linear_data(skb, + np->cur_rx->skbuff->data, pkt_len); + skb_put(skb, pkt_len); +#else + skb_put_data(skb, np->cur_rx->skbuff->data, + pkt_len); +#endif + dma_sync_single_for_device(&np->pci_dev->dev, + np->cur_rx->buffer, + np->rx_buf_sz, + DMA_FROM_DEVICE); + } else { + dma_unmap_single(&np->pci_dev->dev, + np->cur_rx->buffer, + np->rx_buf_sz, + DMA_FROM_DEVICE); + skb_put(skb = np->cur_rx->skbuff, pkt_len); + np->cur_rx->skbuff = NULL; + --np->really_rx_count; + } + skb->protocol = eth_type_trans(skb, dev); + netif_rx(skb); + dev->stats.rx_packets++; + dev->stats.rx_bytes += pkt_len; + } + + np->cur_rx = np->cur_rx->next_desc_logical; + } /* end of while loop */ + + /* allocate skb for rx buffers */ + allocate_rx_buffers(dev); + + return 0; +} + + +static struct net_device_stats *get_stats(struct net_device *dev) +{ + struct netdev_private *np = netdev_priv(dev); + void __iomem *ioaddr = np->mem; + + /* The chip only need report frame silently dropped. */ + if (netif_running(dev)) { + dev->stats.rx_missed_errors += + ioread32(ioaddr + TALLY) & 0x7fff; + dev->stats.rx_crc_errors += + (ioread32(ioaddr + TALLY) & 0x7fff0000) >> 16; + } + + return &dev->stats; +} + + +/* for dev->set_multicast_list */ +static void set_rx_mode(struct net_device *dev) +{ + spinlock_t *lp = &((struct netdev_private *)netdev_priv(dev))->lock; + unsigned long flags; + spin_lock_irqsave(lp, flags); + __set_rx_mode(dev); + spin_unlock_irqrestore(lp, flags); +} + + +/* Take lock before calling */ +static void __set_rx_mode(struct net_device *dev) +{ + struct netdev_private *np = netdev_priv(dev); + void __iomem *ioaddr = np->mem; + u32 mc_filter[2]; /* Multicast hash filter */ + u32 rx_mode; + + if (dev->flags & IFF_PROMISC) { /* Set promiscuous. */ + memset(mc_filter, 0xff, sizeof(mc_filter)); + rx_mode = CR_W_PROM | CR_W_AB | CR_W_AM; + } else if ((netdev_mc_count(dev) > multicast_filter_limit) || + (dev->flags & IFF_ALLMULTI)) { + /* Too many to match, or accept all multicasts. */ + memset(mc_filter, 0xff, sizeof(mc_filter)); + rx_mode = CR_W_AB | CR_W_AM; + } else { + struct netdev_hw_addr *ha; + + memset(mc_filter, 0, sizeof(mc_filter)); + netdev_for_each_mc_addr(ha, dev) { + unsigned int bit; + bit = (ether_crc(ETH_ALEN, ha->addr) >> 26) ^ 0x3F; + mc_filter[bit >> 5] |= (1 << bit); + } + rx_mode = CR_W_AB | CR_W_AM; + } + + stop_nic_rxtx(ioaddr, np->crvalue); + + iowrite32(mc_filter[0], ioaddr + MAR0); + iowrite32(mc_filter[1], ioaddr + MAR1); + np->crvalue &= ~CR_W_RXMODEMASK; + np->crvalue |= rx_mode; + iowrite32(np->crvalue, ioaddr + TCRRCR); +} + +static void netdev_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) +{ + struct netdev_private *np = netdev_priv(dev); + + strscpy(info->driver, DRV_NAME, sizeof(info->driver)); + strscpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info)); +} + +static int netdev_get_link_ksettings(struct net_device *dev, + struct ethtool_link_ksettings *cmd) +{ + struct netdev_private *np = netdev_priv(dev); + + spin_lock_irq(&np->lock); + mii_ethtool_get_link_ksettings(&np->mii, cmd); + spin_unlock_irq(&np->lock); + + return 0; +} + +static int netdev_set_link_ksettings(struct net_device *dev, + const struct ethtool_link_ksettings *cmd) +{ + struct netdev_private *np = netdev_priv(dev); + int rc; + + spin_lock_irq(&np->lock); + rc = mii_ethtool_set_link_ksettings(&np->mii, cmd); + spin_unlock_irq(&np->lock); + + return rc; +} + +static int netdev_nway_reset(struct net_device *dev) +{ + struct netdev_private *np = netdev_priv(dev); + return mii_nway_restart(&np->mii); +} + +static u32 netdev_get_link(struct net_device *dev) +{ + struct netdev_private *np = netdev_priv(dev); + return mii_link_ok(&np->mii); +} + +static u32 netdev_get_msglevel(struct net_device *dev) +{ + return debug; +} + +static void netdev_set_msglevel(struct net_device *dev, u32 value) +{ + debug = value; +} + +static const struct ethtool_ops netdev_ethtool_ops = { + .get_drvinfo = netdev_get_drvinfo, + .nway_reset = netdev_nway_reset, + .get_link = netdev_get_link, + .get_msglevel = netdev_get_msglevel, + .set_msglevel = netdev_set_msglevel, + .get_link_ksettings = netdev_get_link_ksettings, + .set_link_ksettings = netdev_set_link_ksettings, +}; + +static int mii_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) +{ + struct netdev_private *np = netdev_priv(dev); + int rc; + + if (!netif_running(dev)) + return -EINVAL; + + spin_lock_irq(&np->lock); + rc = generic_mii_ioctl(&np->mii, if_mii(rq), cmd, NULL); + spin_unlock_irq(&np->lock); + + return rc; +} + + +static int netdev_close(struct net_device *dev) +{ + struct netdev_private *np = netdev_priv(dev); + void __iomem *ioaddr = np->mem; + int i; + + netif_stop_queue(dev); + + /* Disable interrupts by clearing the interrupt mask. */ + iowrite32(0x0000, ioaddr + IMR); + + /* Stop the chip's Tx and Rx processes. */ + stop_nic_rxtx(ioaddr, 0); + + del_timer_sync(&np->timer); + del_timer_sync(&np->reset_timer); + + free_irq(np->pci_dev->irq, dev); + + /* Free all the skbuffs in the Rx queue. */ + for (i = 0; i < RX_RING_SIZE; i++) { + struct sk_buff *skb = np->rx_ring[i].skbuff; + + np->rx_ring[i].status = 0; + if (skb) { + dma_unmap_single(&np->pci_dev->dev, + np->rx_ring[i].buffer, np->rx_buf_sz, + DMA_FROM_DEVICE); + dev_kfree_skb(skb); + np->rx_ring[i].skbuff = NULL; + } + } + + for (i = 0; i < TX_RING_SIZE; i++) { + struct sk_buff *skb = np->tx_ring[i].skbuff; + + if (skb) { + dma_unmap_single(&np->pci_dev->dev, + np->tx_ring[i].buffer, skb->len, + DMA_TO_DEVICE); + dev_kfree_skb(skb); + np->tx_ring[i].skbuff = NULL; + } + } + + return 0; +} + +static const struct pci_device_id fealnx_pci_tbl[] = { + {0x1516, 0x0800, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + {0x1516, 0x0803, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1}, + {0x1516, 0x0891, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 2}, + {} /* terminate list */ +}; +MODULE_DEVICE_TABLE(pci, fealnx_pci_tbl); + + +static struct pci_driver fealnx_driver = { + .name = "fealnx", + .id_table = fealnx_pci_tbl, + .probe = fealnx_init_one, + .remove = fealnx_remove_one, +}; + +module_pci_driver(fealnx_driver); From 2aab4b96900272885bc157f8b236abf1cdc02e08 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 7 Mar 2023 16:45:30 +0000 Subject: [PATCH 761/765] af_unix: fix struct pid leaks in OOB support syzbot reported struct pid leak [1]. Issue is that queue_oob() calls maybe_add_creds() which potentially holds a reference on a pid. But skb->destructor is not set (either directly or by calling unix_scm_to_skb()) This means that subsequent kfree_skb() or consume_skb() would leak this reference. In this fix, I chose to fully support scm even for the OOB message. [1] BUG: memory leak unreferenced object 0xffff8881053e7f80 (size 128): comm "syz-executor242", pid 5066, jiffies 4294946079 (age 13.220s) hex dump (first 32 bytes): 01 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [] alloc_pid+0x6a/0x560 kernel/pid.c:180 [] copy_process+0x169f/0x26c0 kernel/fork.c:2285 [] kernel_clone+0xf7/0x610 kernel/fork.c:2684 [] __do_sys_clone+0x7c/0xb0 kernel/fork.c:2825 [] do_syscall_x64 arch/x86/entry/common.c:50 [inline] [] do_syscall_64+0x39/0xb0 arch/x86/entry/common.c:80 [] entry_SYSCALL_64_after_hwframe+0x63/0xcd Fixes: 314001f0bf92 ("af_unix: Add OOB support") Reported-by: syzbot+7699d9e5635c10253a27@syzkaller.appspotmail.com Signed-off-by: Eric Dumazet Cc: Rao Shoaib Reviewed-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20230307164530.771896-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/unix/af_unix.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 347122c3575ea..0b0f18ecce447 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -2105,7 +2105,8 @@ static int unix_dgram_sendmsg(struct socket *sock, struct msghdr *msg, #define UNIX_SKB_FRAGS_SZ (PAGE_SIZE << get_order(32768)) #if IS_ENABLED(CONFIG_AF_UNIX_OOB) -static int queue_oob(struct socket *sock, struct msghdr *msg, struct sock *other) +static int queue_oob(struct socket *sock, struct msghdr *msg, struct sock *other, + struct scm_cookie *scm, bool fds_sent) { struct unix_sock *ousk = unix_sk(other); struct sk_buff *skb; @@ -2116,6 +2117,11 @@ static int queue_oob(struct socket *sock, struct msghdr *msg, struct sock *other if (!skb) return err; + err = unix_scm_to_skb(scm, skb, !fds_sent); + if (err < 0) { + kfree_skb(skb); + return err; + } skb_put(skb, 1); err = skb_copy_datagram_from_iter(skb, 0, &msg->msg_iter, 1); @@ -2243,7 +2249,7 @@ static int unix_stream_sendmsg(struct socket *sock, struct msghdr *msg, #if IS_ENABLED(CONFIG_AF_UNIX_OOB) if (msg->msg_flags & MSG_OOB) { - err = queue_oob(sock, msg, other); + err = queue_oob(sock, msg, other, &scm, fds_sent); if (err) goto out_err; sent++; From 649c15c7691e9b13cbe9bf6c65c365350e056067 Mon Sep 17 00:00:00 2001 From: Thadeu Lima de Souza Cascardo Date: Tue, 7 Mar 2023 14:37:07 -0300 Subject: [PATCH 762/765] net: avoid double iput when sock_alloc_file fails When sock_alloc_file fails to allocate a file, it will call sock_release. __sys_socket_file should then not call sock_release again, otherwise there will be a double free. [ 89.319884] ------------[ cut here ]------------ [ 89.320286] kernel BUG at fs/inode.c:1764! [ 89.320656] invalid opcode: 0000 [#1] PREEMPT SMP NOPTI [ 89.321051] CPU: 7 PID: 125 Comm: iou-sqp-124 Not tainted 6.2.0+ #361 [ 89.321535] RIP: 0010:iput+0x1ff/0x240 [ 89.321808] Code: d1 83 e1 03 48 83 f9 02 75 09 48 81 fa 00 10 00 00 77 05 83 e2 01 75 1f 4c 89 ef e8 fb d2 ba 00 e9 80 fe ff ff c3 cc cc cc cc <0f> 0b 0f 0b e9 d0 fe ff ff 0f 0b eb 8d 49 8d b4 24 08 01 00 00 48 [ 89.322760] RSP: 0018:ffffbdd60068bd50 EFLAGS: 00010202 [ 89.323036] RAX: 0000000000000000 RBX: ffff9d7ad3cacac0 RCX: 0000000000001107 [ 89.323412] RDX: 000000000003af00 RSI: 0000000000000000 RDI: ffff9d7ad3cacb40 [ 89.323785] RBP: ffffbdd60068bd68 R08: ffffffffffffffff R09: ffffffffab606438 [ 89.324157] R10: ffffffffacb3dfa0 R11: 6465686361657256 R12: ffff9d7ad3cacb40 [ 89.324529] R13: 0000000080000001 R14: 0000000080000001 R15: 0000000000000002 [ 89.324904] FS: 00007f7b28516740(0000) GS:ffff9d7aeb1c0000(0000) knlGS:0000000000000000 [ 89.325328] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 89.325629] CR2: 00007f0af52e96c0 CR3: 0000000002a02006 CR4: 0000000000770ee0 [ 89.326004] PKRU: 55555554 [ 89.326161] Call Trace: [ 89.326298] [ 89.326419] __sock_release+0xb5/0xc0 [ 89.326632] __sys_socket_file+0xb2/0xd0 [ 89.326844] io_socket+0x88/0x100 [ 89.327039] ? io_issue_sqe+0x6a/0x430 [ 89.327258] io_issue_sqe+0x67/0x430 [ 89.327450] io_submit_sqes+0x1fe/0x670 [ 89.327661] io_sq_thread+0x2e6/0x530 [ 89.327859] ? __pfx_autoremove_wake_function+0x10/0x10 [ 89.328145] ? __pfx_io_sq_thread+0x10/0x10 [ 89.328367] ret_from_fork+0x29/0x50 [ 89.328576] RIP: 0033:0x0 [ 89.328732] Code: Unable to access opcode bytes at 0xffffffffffffffd6. [ 89.329073] RSP: 002b:0000000000000000 EFLAGS: 00000202 ORIG_RAX: 00000000000001a9 [ 89.329477] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 00007f7b28637a3d [ 89.329845] RDX: 00007fff4e4318a8 RSI: 00007fff4e4318b0 RDI: 0000000000000400 [ 89.330216] RBP: 00007fff4e431830 R08: 00007fff4e431711 R09: 00007fff4e4318b0 [ 89.330584] R10: 0000000000000000 R11: 0000000000000202 R12: 00007fff4e441b38 [ 89.330950] R13: 0000563835e3e725 R14: 0000563835e40d10 R15: 00007f7b28784040 [ 89.331318] [ 89.331441] Modules linked in: [ 89.331617] ---[ end trace 0000000000000000 ]--- Fixes: da214a475f8b ("net: add __sys_socket_file()") Signed-off-by: Thadeu Lima de Souza Cascardo Reviewed-by: Jens Axboe Reviewed-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20230307173707.468744-1-cascardo@canonical.com Signed-off-by: Jakub Kicinski --- net/socket.c | 11 ++++------- 1 file changed, 4 insertions(+), 7 deletions(-) diff --git a/net/socket.c b/net/socket.c index 6bae8ce7059ee..9c92c0e6c4da8 100644 --- a/net/socket.c +++ b/net/socket.c @@ -450,7 +450,9 @@ static struct file_system_type sock_fs_type = { * * Returns the &file bound with @sock, implicitly storing it * in sock->file. If dname is %NULL, sets to "". - * On failure the return is a ERR pointer (see linux/err.h). + * + * On failure @sock is released, and an ERR pointer is returned. + * * This function uses GFP_KERNEL internally. */ @@ -1638,7 +1640,6 @@ static struct socket *__sys_socket_create(int family, int type, int protocol) struct file *__sys_socket_file(int family, int type, int protocol) { struct socket *sock; - struct file *file; int flags; sock = __sys_socket_create(family, type, protocol); @@ -1649,11 +1650,7 @@ struct file *__sys_socket_file(int family, int type, int protocol) if (SOCK_NONBLOCK != O_NONBLOCK && (flags & SOCK_NONBLOCK)) flags = (flags & ~SOCK_NONBLOCK) | O_NONBLOCK; - file = sock_alloc_file(sock, flags, NULL); - if (IS_ERR(file)) - sock_release(sock); - - return file; + return sock_alloc_file(sock, flags, NULL); } int __sys_socket(int family, int type, int protocol) From 6517a60b0307abdcca80133c237551cc7cc8e6ae Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 7 Mar 2023 16:39:22 -0800 Subject: [PATCH 763/765] tools: ynl: move the enum classes to shared code Move bulk of the EnumSet and EnumEntry code to shared code for reuse by cli. Signed-off-by: Jakub Kicinski --- tools/net/ynl/lib/__init__.py | 7 ++- tools/net/ynl/lib/nlspec.py | 96 ++++++++++++++++++++++++++++++ tools/net/ynl/ynl-gen-c.py | 107 +++++++--------------------------- 3 files changed, 121 insertions(+), 89 deletions(-) diff --git a/tools/net/ynl/lib/__init__.py b/tools/net/ynl/lib/__init__.py index a2cb8b16d6f10..4b3797fe784ba 100644 --- a/tools/net/ynl/lib/__init__.py +++ b/tools/net/ynl/lib/__init__.py @@ -1,7 +1,8 @@ # SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause -from .nlspec import SpecAttr, SpecAttrSet, SpecFamily, SpecOperation +from .nlspec import SpecAttr, SpecAttrSet, SpecEnumEntry, SpecEnumSet, \ + SpecFamily, SpecOperation from .ynl import YnlFamily -__all__ = ["SpecAttr", "SpecAttrSet", "SpecFamily", "SpecOperation", - "YnlFamily"] +__all__ = ["SpecAttr", "SpecAttrSet", "SpecEnumEntry", "SpecEnumSet", + "SpecFamily", "SpecOperation", "YnlFamily"] diff --git a/tools/net/ynl/lib/nlspec.py b/tools/net/ynl/lib/nlspec.py index 0a2cfb5862aa7..7c1cf6c1499e1 100644 --- a/tools/net/ynl/lib/nlspec.py +++ b/tools/net/ynl/lib/nlspec.py @@ -57,6 +57,91 @@ def resolve(self): pass +class SpecEnumEntry(SpecElement): + """ Entry within an enum declared in the Netlink spec. + + Attributes: + doc documentation string + enum_set back reference to the enum + value numerical value of this enum (use accessors in most situations!) + + Methods: + raw_value raw value, i.e. the id in the enum, unlike user value which is a mask for flags + user_value user value, same as raw value for enums, for flags it's the mask + """ + def __init__(self, enum_set, yaml, prev, value_start): + if isinstance(yaml, str): + yaml = {'name': yaml} + super().__init__(enum_set.family, yaml) + + self.doc = yaml.get('doc', '') + self.enum_set = enum_set + + if 'value' in yaml: + self.value = yaml['value'] + elif prev: + self.value = prev.value + 1 + else: + self.value = value_start + + def has_doc(self): + return bool(self.doc) + + def raw_value(self): + return self.value + + def user_value(self): + if self.enum_set['type'] == 'flags': + return 1 << self.value + else: + return self.value + + +class SpecEnumSet(SpecElement): + """ Enum type + + Represents an enumeration (list of numerical constants) + as declared in the "definitions" section of the spec. + + Attributes: + type enum or flags + entries entries by name + Methods: + get_mask for flags compute the mask of all defined values + """ + def __init__(self, family, yaml): + super().__init__(family, yaml) + + self.type = yaml['type'] + + prev_entry = None + value_start = self.yaml.get('value-start', 0) + self.entries = dict() + for entry in self.yaml['entries']: + e = self.new_entry(entry, prev_entry, value_start) + self.entries[e.name] = e + prev_entry = e + + def new_entry(self, entry, prev_entry, value_start): + return SpecEnumEntry(self, entry, prev_entry, value_start) + + def has_doc(self): + if 'doc' in self.yaml: + return True + for entry in self.entries.values(): + if entry.has_doc(): + return True + return False + + def get_mask(self): + mask = 0 + idx = self.yaml.get('value-start', 0) + for _ in self.entries.values(): + mask |= 1 << idx + idx += 1 + return mask + + class SpecAttr(SpecElement): """ Single Netlink atttribute type @@ -193,6 +278,7 @@ class SpecFamily(SpecElement): msgs dict of all messages (index by name) msgs_by_value dict of all messages (indexed by name) ops dict of all valid requests / responses + consts dict of all constants/enums """ def __init__(self, spec_path, schema_path=None): with open(spec_path, "r") as stream: @@ -222,6 +308,7 @@ def __init__(self, spec_path, schema_path=None): self.req_by_value = collections.OrderedDict() self.rsp_by_value = collections.OrderedDict() self.ops = collections.OrderedDict() + self.consts = collections.OrderedDict() last_exception = None while len(self._resolution_list) > 0: @@ -242,6 +329,9 @@ def __init__(self, spec_path, schema_path=None): if len(resolved) == 0: raise last_exception + def new_enum(self, elem): + return SpecEnumSet(self, elem) + def new_attr_set(self, elem): return SpecAttrSet(self, elem) @@ -296,6 +386,12 @@ def _dictify_ops_directional(self): def resolve(self): self.resolve_up(super()) + for elem in self.yaml['definitions']: + if elem['type'] == 'enum' or elem['type'] == 'flags': + self.consts[elem['name']] = self.new_enum(elem) + else: + self.consts[elem['name']] = elem + for elem in self.yaml['attribute-sets']: attr_set = self.new_attr_set(elem) self.attr_sets[elem['name']] = attr_set diff --git a/tools/net/ynl/ynl-gen-c.py b/tools/net/ynl/ynl-gen-c.py index c940ca834d3f3..1bcc5354d8000 100755 --- a/tools/net/ynl/ynl-gen-c.py +++ b/tools/net/ynl/ynl-gen-c.py @@ -6,7 +6,7 @@ import os import yaml -from lib import SpecFamily, SpecAttrSet, SpecAttr, SpecOperation +from lib import SpecFamily, SpecAttrSet, SpecAttr, SpecOperation, SpecEnumSet, SpecEnumEntry def c_upper(name): @@ -567,97 +567,37 @@ def set_inherited(self, new_inherited): self.inherited = [c_lower(x) for x in sorted(self._inherited)] -class EnumEntry: +class EnumEntry(SpecEnumEntry): def __init__(self, enum_set, yaml, prev, value_start): - if isinstance(yaml, str): - self.name = yaml - yaml = {} - self.doc = '' - else: - self.name = yaml['name'] - self.doc = yaml.get('doc', '') - - self.yaml = yaml - self.enum_set = enum_set - self.c_name = c_upper(enum_set.value_pfx + self.name) - - if 'value' in yaml: - self.value = yaml['value'] - if prev: - self.value_change = (self.value != prev.value + 1) - elif prev: - self.value_change = False - self.value = prev.value + 1 + super().__init__(enum_set, yaml, prev, value_start) + + if prev: + self.value_change = (self.value != prev.value + 1) else: - self.value = value_start self.value_change = (self.value != 0) - self.value_change = self.value_change or self.enum_set['type'] == 'flags' - def __getitem__(self, key): - return self.yaml[key] - - def __contains__(self, key): - return key in self.yaml - - def has_doc(self): - return bool(self.doc) + # Added by resolve: + self.c_name = None + delattr(self, "c_name") - # raw value, i.e. the id in the enum, unlike user value which is a mask for flags - def raw_value(self): - return self.value + def resolve(self): + self.resolve_up(super()) - # user value, same as raw value for enums, for flags it's the mask - def user_value(self): - if self.enum_set['type'] == 'flags': - return 1 << self.value - else: - return self.value + self.c_name = c_upper(self.enum_set.value_pfx + self.name) -class EnumSet: +class EnumSet(SpecEnumSet): def __init__(self, family, yaml): - self.yaml = yaml - self.family = family - self.render_name = c_lower(family.name + '-' + yaml['name']) self.enum_name = 'enum ' + self.render_name self.value_pfx = yaml.get('name-prefix', f"{family.name}-{yaml['name']}-") - self.type = yaml['type'] - - prev_entry = None - value_start = self.yaml.get('value-start', 0) - self.entries = {} - self.entry_list = [] - for entry in self.yaml['entries']: - e = EnumEntry(self, entry, prev_entry, value_start) - self.entries[e.name] = e - self.entry_list.append(e) - prev_entry = e - - def __getitem__(self, key): - return self.yaml[key] - - def __contains__(self, key): - return key in self.yaml - - def has_doc(self): - if 'doc' in self.yaml: - return True - for entry in self.entry_list: - if entry.has_doc(): - return True - return False + super().__init__(family, yaml) - def get_mask(self): - mask = 0 - idx = self.yaml.get('value-start', 0) - for _ in self.entry_list: - mask |= 1 << idx - idx += 1 - return mask + def new_entry(self, entry, prev_entry, value_start): + return EnumEntry(self, entry, prev_entry, value_start) class AttrSet(SpecAttrSet): @@ -792,8 +732,6 @@ def resolve(self): self.mcgrps = self.yaml.get('mcast-groups', {'list': []}) - self.consts = dict() - self.hooks = dict() for when in ['pre', 'post']: self.hooks[when] = dict() @@ -820,6 +758,9 @@ def resolve(self): if self.kernel_policy == 'global': self._load_global_policy() + def new_enum(self, elem): + return EnumSet(self, elem) + def new_attr_set(self, elem): return AttrSet(self, elem) @@ -837,12 +778,6 @@ def _mock_up_events(self): } def _dictify(self): - for elem in self.yaml['definitions']: - if elem['type'] == 'enum' or elem['type'] == 'flags': - self.consts[elem['name']] = EnumSet(self, elem) - else: - self.consts[elem['name']] = elem - ntf = [] for msg in self.msgs.values(): if 'notify' in msg: @@ -1980,7 +1915,7 @@ def render_uapi(family, cw): if 'doc' in enum: doc = ' - ' + enum['doc'] cw.write_doc_line(enum.enum_name + doc) - for entry in enum.entry_list: + for entry in enum.entries.values(): if entry.has_doc(): doc = '@' + entry.c_name + ': ' + entry['doc'] cw.write_doc_line(doc) @@ -1988,7 +1923,7 @@ def render_uapi(family, cw): uapi_enum_start(family, cw, const, 'name') name_pfx = const.get('name-prefix', f"{family.name}-{const['name']}-") - for entry in enum.entry_list: + for entry in enum.entries.values(): suffix = ',' if entry.value_change: suffix = f" = {entry.user_value()}" + suffix From c311aaa74ca18bd0781d1d3bebd08484799f840c Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 7 Mar 2023 16:39:23 -0800 Subject: [PATCH 764/765] tools: ynl: fix enum-as-flags in the generic CLI Lorenzo points out that the generic CLI is broken for the netdev family. When I added the support for documentation of enums (and sparse enums) the client script was not updated. It expects the values in enum to be a list of names, now it can also be a dict (YAML object). Reported-by: Lorenzo Bianconi Fixes: e4b48ed460d3 ("tools: ynl: add a completely generic client") Signed-off-by: Jakub Kicinski --- tools/net/ynl/lib/nlspec.py | 7 +++++-- tools/net/ynl/lib/ynl.py | 9 ++------- 2 files changed, 7 insertions(+), 9 deletions(-) diff --git a/tools/net/ynl/lib/nlspec.py b/tools/net/ynl/lib/nlspec.py index 7c1cf6c1499e1..a34d088f67432 100644 --- a/tools/net/ynl/lib/nlspec.py +++ b/tools/net/ynl/lib/nlspec.py @@ -104,8 +104,9 @@ class SpecEnumSet(SpecElement): as declared in the "definitions" section of the spec. Attributes: - type enum or flags - entries entries by name + type enum or flags + entries entries by name + entries_by_val entries by value Methods: get_mask for flags compute the mask of all defined values """ @@ -117,9 +118,11 @@ def __init__(self, family, yaml): prev_entry = None value_start = self.yaml.get('value-start', 0) self.entries = dict() + self.entries_by_val = dict() for entry in self.yaml['entries']: e = self.new_entry(entry, prev_entry, value_start) self.entries[e.name] = e + self.entries_by_val[e.raw_value()] = e prev_entry = e def new_entry(self, entry, prev_entry, value_start): diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py index a842adc8e87e3..90764a83c6461 100644 --- a/tools/net/ynl/lib/ynl.py +++ b/tools/net/ynl/lib/ynl.py @@ -303,11 +303,6 @@ def __init__(self, def_path, schema=None): self.sock.setsockopt(Netlink.SOL_NETLINK, Netlink.NETLINK_CAP_ACK, 1) self.sock.setsockopt(Netlink.SOL_NETLINK, Netlink.NETLINK_EXT_ACK, 1) - self._types = dict() - - for elem in self.yaml.get('definitions', []): - self._types[elem['name']] = elem - self.async_msg_ids = set() self.async_msg_queue = [] @@ -353,13 +348,13 @@ def _add_attr(self, space, name, value): def _decode_enum(self, rsp, attr_spec): raw = rsp[attr_spec['name']] - enum = self._types[attr_spec['enum']] + enum = self.consts[attr_spec['enum']] i = attr_spec.get('value-start', 0) if 'enum-as-flags' in attr_spec and attr_spec['enum-as-flags']: value = set() while raw: if raw & 1: - value.add(enum['entries'][i]) + value.add(enum.entries_by_val[i].name) raw >>= 1 i += 1 else: From 573b22ccb7ce9ab7f0539a2e11a9d3609a8783f5 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 6 Mar 2023 01:20:30 +0000 Subject: [PATCH 765/765] sh: sanitize the flags on sigreturn We fetch %SR value from sigframe; it might have been modified by signal handler, so we can't trust it with any bits that are not modifiable in user mode. Signed-off-by: Al Viro Cc: Rich Felker Signed-off-by: Linus Torvalds --- arch/sh/include/asm/processor_32.h | 1 + arch/sh/kernel/signal_32.c | 3 +++ 2 files changed, 4 insertions(+) diff --git a/arch/sh/include/asm/processor_32.h b/arch/sh/include/asm/processor_32.h index 27aebf1e75a20..3ef7adf739c83 100644 --- a/arch/sh/include/asm/processor_32.h +++ b/arch/sh/include/asm/processor_32.h @@ -50,6 +50,7 @@ #define SR_FD 0x00008000 #define SR_MD 0x40000000 +#define SR_USER_MASK 0x00000303 // M, Q, S, T bits /* * DSP structure and data */ diff --git a/arch/sh/kernel/signal_32.c b/arch/sh/kernel/signal_32.c index 90f495d35db29..a6bfc6f374911 100644 --- a/arch/sh/kernel/signal_32.c +++ b/arch/sh/kernel/signal_32.c @@ -115,6 +115,7 @@ static int restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *r0_p) { unsigned int err = 0; + unsigned int sr = regs->sr & ~SR_USER_MASK; #define COPY(x) err |= __get_user(regs->x, &sc->sc_##x) COPY(regs[1]); @@ -130,6 +131,8 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *r0_p COPY(sr); COPY(pc); #undef COPY + regs->sr = (regs->sr & SR_USER_MASK) | sr; + #ifdef CONFIG_SH_FPU if (boot_cpu_data.flags & CPU_HAS_FPU) { int owned_fp;