Merge tag 'for-f2fs-3.13' of git://git.kernel.org/pub/scm/linux/kerne…

…l/git/jaegeuk/f2fs

Pull f2fs updates from Jaegeuk Kim:
 "This patch-set includes the following major enhancement patches.
   - add a sysfs to control reclaiming free segments
   - enhance the f2fs global lock procedures
   - enhance the victim selection flow
   - wait for selected node blocks during fsync
   - add some tracepoints
   - add a config to remove abundant BUG_ONs

  The other bug fixes are as follows.
   - fix deadlock on acl operations
   - fix some bugs with respect to orphan inodes

  And, there are a bunch of cleanups"

* tag 'for-f2fs-3.13' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (42 commits)
  f2fs: issue more large discard command
  f2fs: fix memory leak after kobject init failed in fill_super
  f2fs: cleanup waiting routine for writeback pages in cp
  f2fs: avoid to use a NULL point in destroy_segment_manager
  f2fs: remove unnecessary TestClearPageError when wait pages writeback
  f2fs: update f2fs document
  f2fs: avoid to wait all the node blocks during fsync
  f2fs: check all ones or zeros bitmap with bitops for better mount performance
  f2fs: change the method of calculating the number summary blocks
  f2fs: fix calculating incorrect free size when update xattr in __f2fs_setxattr
  f2fs: add an option to avoid unnecessary BUG_ONs
  f2fs: introduce CONFIG_F2FS_CHECK_FS for BUG_ON control
  f2fs: fix a deadlock during init_acl procedure
  f2fs: clean up acl flow for better readability
  f2fs: remove unnecessary segment bitmap updates
  f2fs: add tracepoint for vm_page_mkwrite
  f2fs: add tracepoint for set_page_dirty
  f2fs: remove redundant set_page_dirty from write_compacted_summaries
  f2fs: add reclaiming control by sysfs
  f2fs: introduce f2fs_balance_fs_bg for some background jobs
  ...

Documentation/filesystems/f2fs.txt

@@ -119,6 +119,7 @@ active_logs=%u         Support configuring the number of active logs. In the
                        Default number is 6.
 disable_ext_identify   Disable the extension list configured by mkfs, so f2fs
                        does not aware of cold files such as media files.
+inline_xattr           Enable the inline xattrs feature.
 
 ================================================================================
 DEBUGFS ENTRIES
@@ -164,6 +165,12 @@ Files in /sys/fs/f2fs/<devname>
                               gc_idle = 1 will select the Cost Benefit approach
                               & setting gc_idle = 2 will select the greedy aproach.
 
+ reclaim_segments             This parameter controls the number of prefree
+                              segments to be reclaimed. If the number of prefree
+			      segments is larger than this number, f2fs tries to
+			      conduct checkpoint to reclaim the prefree segments
+			      to free segments. By default, 100 segments, 200MB.
+
 ================================================================================
 USAGE
 ================================================================================

fs/f2fs/Kconfig

@@ -63,3 +63,11 @@ config F2FS_FS_SECURITY
 	  the extended attribute support in advance.
 
 	  If you are not using a security module, say N.
+
+config F2FS_CHECK_FS
+	bool "F2FS consistency checking feature"
+	depends on F2FS_FS
+	help
+	  Enables BUG_ONs which check the file system consistency in runtime.
+
+	  If you want to improve the performance, say N.

fs/f2fs/acl.c

@@ -205,7 +205,8 @@ struct posix_acl *f2fs_get_acl(struct inode *inode, int type)
 	return acl;
 }
 
-static int f2fs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
+static int f2fs_set_acl(struct inode *inode, int type,
+			struct posix_acl *acl, struct page *ipage)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
 	struct f2fs_inode_info *fi = F2FS_I(inode);
@@ -250,7 +251,7 @@ static int f2fs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 		}
 	}
 
-	error = f2fs_setxattr(inode, name_index, "", value, size, NULL);
+	error = f2fs_setxattr(inode, name_index, "", value, size, ipage);
 
 	kfree(value);
 	if (!error)
@@ -260,10 +261,10 @@ static int f2fs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 	return error;
 }
 
-int f2fs_init_acl(struct inode *inode, struct inode *dir)
+int f2fs_init_acl(struct inode *inode, struct inode *dir, struct page *ipage)
 {
-	struct posix_acl *acl = NULL;
 	struct f2fs_sb_info *sbi = F2FS_SB(dir->i_sb);
+	struct posix_acl *acl = NULL;
 	int error = 0;
 
 	if (!S_ISLNK(inode->i_mode)) {
@@ -276,19 +277,19 @@ int f2fs_init_acl(struct inode *inode, struct inode *dir)
 			inode->i_mode &= ~current_umask();
 	}
 
-	if (test_opt(sbi, POSIX_ACL) && acl) {
+	if (!test_opt(sbi, POSIX_ACL) || !acl)
+		goto cleanup;
 
-		if (S_ISDIR(inode->i_mode)) {
-			error = f2fs_set_acl(inode, ACL_TYPE_DEFAULT, acl);
-			if (error)
-				goto cleanup;
-		}
-		error = posix_acl_create(&acl, GFP_KERNEL, &inode->i_mode);
-		if (error < 0)
-			return error;
-		if (error > 0)
-			error = f2fs_set_acl(inode, ACL_TYPE_ACCESS, acl);
+	if (S_ISDIR(inode->i_mode)) {
+		error = f2fs_set_acl(inode, ACL_TYPE_DEFAULT, acl, ipage);
+		if (error)
+			goto cleanup;
 	}
+	error = posix_acl_create(&acl, GFP_KERNEL, &inode->i_mode);
+	if (error < 0)
+		return error;
+	if (error > 0)
+		error = f2fs_set_acl(inode, ACL_TYPE_ACCESS, acl, ipage);
 cleanup:
 	posix_acl_release(acl);
 	return error;
@@ -313,7 +314,8 @@ int f2fs_acl_chmod(struct inode *inode)
 	error = posix_acl_chmod(&acl, GFP_KERNEL, mode);
 	if (error)
 		return error;
-	error = f2fs_set_acl(inode, ACL_TYPE_ACCESS, acl);
+
+	error = f2fs_set_acl(inode, ACL_TYPE_ACCESS, acl, NULL);
 	posix_acl_release(acl);
 	return error;
 }
@@ -388,7 +390,7 @@ static int f2fs_xattr_set_acl(struct dentry *dentry, const char *name,
 		acl = NULL;
 	}
 
-	error = f2fs_set_acl(inode, type, acl);
+	error = f2fs_set_acl(inode, type, acl, NULL);
 
 release_and_out:
 	posix_acl_release(acl);

fs/f2fs/acl.h

@@ -36,9 +36,9 @@ struct f2fs_acl_header {
 
 #ifdef CONFIG_F2FS_FS_POSIX_ACL
 
-extern struct posix_acl *f2fs_get_acl(struct inode *inode, int type);
-extern int f2fs_acl_chmod(struct inode *inode);
-extern int f2fs_init_acl(struct inode *inode, struct inode *dir);
+extern struct posix_acl *f2fs_get_acl(struct inode *, int);
+extern int f2fs_acl_chmod(struct inode *);
+extern int f2fs_init_acl(struct inode *, struct inode *, struct page *);
 #else
 #define f2fs_check_acl	NULL
 #define f2fs_get_acl	NULL
@@ -49,7 +49,8 @@ static inline int f2fs_acl_chmod(struct inode *inode)
 	return 0;
 }
 
-static inline int f2fs_init_acl(struct inode *inode, struct inode *dir)
+static inline int f2fs_init_acl(struct inode *inode, struct inode *dir,
+							struct page *page)
 {
 	return 0;
 }

fs/f2fs/checkpoint.c

@@ -81,7 +81,7 @@ static int f2fs_write_meta_page(struct page *page,
 	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
 
 	/* Should not write any meta pages, if any IO error was occurred */
-	if (wbc->for_reclaim ||
+	if (wbc->for_reclaim || sbi->por_doing ||
 			is_set_ckpt_flags(F2FS_CKPT(sbi), CP_ERROR_FLAG)) {
 		dec_page_count(sbi, F2FS_DIRTY_META);
 		wbc->pages_skipped++;
@@ -142,8 +142,8 @@ long sync_meta_pages(struct f2fs_sb_info *sbi, enum page_type type,
 		for (i = 0; i < nr_pages; i++) {
 			struct page *page = pvec.pages[i];
 			lock_page(page);
-			BUG_ON(page->mapping != mapping);
-			BUG_ON(!PageDirty(page));
+			f2fs_bug_on(page->mapping != mapping);
+			f2fs_bug_on(!PageDirty(page));
 			clear_page_dirty_for_io(page);
 			if (f2fs_write_meta_page(page, &wbc)) {
 				unlock_page(page);
@@ -167,6 +167,8 @@ static int f2fs_set_meta_page_dirty(struct page *page)
 	struct address_space *mapping = page->mapping;
 	struct f2fs_sb_info *sbi = F2FS_SB(mapping->host->i_sb);
 
+	trace_f2fs_set_page_dirty(page, META);
+
 	SetPageUptodate(page);
 	if (!PageDirty(page)) {
 		__set_page_dirty_nobuffers(page);
@@ -206,6 +208,7 @@ int acquire_orphan_inode(struct f2fs_sb_info *sbi)
 void release_orphan_inode(struct f2fs_sb_info *sbi)
 {
 	mutex_lock(&sbi->orphan_inode_mutex);
+	f2fs_bug_on(sbi->n_orphans == 0);
 	sbi->n_orphans--;
 	mutex_unlock(&sbi->orphan_inode_mutex);
 }
@@ -225,12 +228,8 @@ void add_orphan_inode(struct f2fs_sb_info *sbi, nid_t ino)
 			break;
 		orphan = NULL;
 	}
-retry:
-	new = kmem_cache_alloc(orphan_entry_slab, GFP_ATOMIC);
-	if (!new) {
-		cond_resched();
-		goto retry;
-	}
+
+	new = f2fs_kmem_cache_alloc(orphan_entry_slab, GFP_ATOMIC);
 	new->ino = ino;
 
 	/* add new_oentry into list which is sorted by inode number */
@@ -253,6 +252,7 @@ void remove_orphan_inode(struct f2fs_sb_info *sbi, nid_t ino)
 		if (orphan->ino == ino) {
 			list_del(&orphan->list);
 			kmem_cache_free(orphan_entry_slab, orphan);
+			f2fs_bug_on(sbi->n_orphans == 0);
 			sbi->n_orphans--;
 			break;
 		}
@@ -263,7 +263,7 @@ void remove_orphan_inode(struct f2fs_sb_info *sbi, nid_t ino)
 static void recover_orphan_inode(struct f2fs_sb_info *sbi, nid_t ino)
 {
 	struct inode *inode = f2fs_iget(sbi->sb, ino);
-	BUG_ON(IS_ERR(inode));
+	f2fs_bug_on(IS_ERR(inode));
 	clear_nlink(inode);
 
 	/* truncate all the data during iput */
@@ -277,7 +277,7 @@ int recover_orphan_inodes(struct f2fs_sb_info *sbi)
 	if (!is_set_ckpt_flags(F2FS_CKPT(sbi), CP_ORPHAN_PRESENT_FLAG))
 		return 0;
 
-	sbi->por_doing = 1;
+	sbi->por_doing = true;
 	start_blk = __start_cp_addr(sbi) + 1;
 	orphan_blkaddr = __start_sum_addr(sbi) - 1;
 
@@ -294,7 +294,7 @@ int recover_orphan_inodes(struct f2fs_sb_info *sbi)
 	}
 	/* clear Orphan Flag */
 	clear_ckpt_flags(F2FS_CKPT(sbi), CP_ORPHAN_PRESENT_FLAG);
-	sbi->por_doing = 0;
+	sbi->por_doing = false;
 	return 0;
 }
 
@@ -469,9 +469,7 @@ static int __add_dirty_inode(struct inode *inode, struct dir_inode_entry *new)
 			return -EEXIST;
 	}
 	list_add_tail(&new->list, head);
-#ifdef CONFIG_F2FS_STAT_FS
-	sbi->n_dirty_dirs++;
-#endif
+	stat_inc_dirty_dir(sbi);
 	return 0;
 }
 
@@ -482,12 +480,8 @@ void set_dirty_dir_page(struct inode *inode, struct page *page)
 
 	if (!S_ISDIR(inode->i_mode))
 		return;
-retry:
-	new = kmem_cache_alloc(inode_entry_slab, GFP_NOFS);
-	if (!new) {
-		cond_resched();
-		goto retry;
-	}
+
+	new = f2fs_kmem_cache_alloc(inode_entry_slab, GFP_NOFS);
 	new->inode = inode;
 	INIT_LIST_HEAD(&new->list);
 
@@ -504,13 +498,9 @@ void set_dirty_dir_page(struct inode *inode, struct page *page)
 void add_dirty_dir_inode(struct inode *inode)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
-	struct dir_inode_entry *new;
-retry:
-	new = kmem_cache_alloc(inode_entry_slab, GFP_NOFS);
-	if (!new) {
-		cond_resched();
-		goto retry;
-	}
+	struct dir_inode_entry *new =
+			f2fs_kmem_cache_alloc(inode_entry_slab, GFP_NOFS);
+
 	new->inode = inode;
 	INIT_LIST_HEAD(&new->list);
 
@@ -541,9 +531,7 @@ void remove_dirty_dir_inode(struct inode *inode)
 		if (entry->inode == inode) {
 			list_del(&entry->list);
 			kmem_cache_free(inode_entry_slab, entry);
-#ifdef CONFIG_F2FS_STAT_FS
-			sbi->n_dirty_dirs--;
-#endif
+			stat_dec_dirty_dir(sbi);
 			break;
 		}
 	}
@@ -617,11 +605,10 @@ static void block_operations(struct f2fs_sb_info *sbi)
 	blk_start_plug(&plug);
 
 retry_flush_dents:
-	mutex_lock_all(sbi);
-
+	f2fs_lock_all(sbi);
 	/* write all the dirty dentry pages */
 	if (get_pages(sbi, F2FS_DIRTY_DENTS)) {
-		mutex_unlock_all(sbi);
+		f2fs_unlock_all(sbi);
 		sync_dirty_dir_inodes(sbi);
 		goto retry_flush_dents;
 	}
@@ -644,7 +631,22 @@ static void block_operations(struct f2fs_sb_info *sbi)
 static void unblock_operations(struct f2fs_sb_info *sbi)
 {
 	mutex_unlock(&sbi->node_write);
-	mutex_unlock_all(sbi);
+	f2fs_unlock_all(sbi);
+}
+
+static void wait_on_all_pages_writeback(struct f2fs_sb_info *sbi)
+{
+	DEFINE_WAIT(wait);
+
+	for (;;) {
+		prepare_to_wait(&sbi->cp_wait, &wait, TASK_UNINTERRUPTIBLE);
+
+		if (!get_pages(sbi, F2FS_WRITEBACK))
+			break;
+
+		io_schedule();
+	}
+	finish_wait(&sbi->cp_wait, &wait);
 }
 
 static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
@@ -756,8 +758,7 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 	f2fs_put_page(cp_page, 1);
 
 	/* wait for previous submitted node/meta pages writeback */
-	while (get_pages(sbi, F2FS_WRITEBACK))
-		congestion_wait(BLK_RW_ASYNC, HZ / 50);
+	wait_on_all_pages_writeback(sbi);
 
 	filemap_fdatawait_range(sbi->node_inode->i_mapping, 0, LONG_MAX);
 	filemap_fdatawait_range(sbi->meta_inode->i_mapping, 0, LONG_MAX);