Merge tag 'xfs-5.5-fixes-2' of git://git.kernel.org/pub/scm/fs/xfs/xf…

…s-linux

Pull xfs fixes from Darrick Wong:
 "Fix a few bugs that could lead to corrupt files, fsck complaints, and
  filesystem crashes:

   - Minor documentation fixes

   - Fix a file corruption due to read racing with an insert range
     operation.

   - Fix log reservation overflows when allocating large rt extents

   - Fix a buffer log item flags check

   - Don't allow administrators to mount with sunit= options that will
     cause later xfs_repair complaints about the root directory being
     suspicious because the fs geometry appeared inconsistent

   - Fix a non-static helper that should have been static"

* tag 'xfs-5.5-fixes-2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  xfs: Make the symbol 'xfs_rtalloc_log_count' static
  xfs: don't commit sunit/swidth updates to disk if that would cause repair failures
  xfs: split the sunit parameter update into two parts
  xfs: refactor agfl length computation function
  libxfs: resync with the userspace libxfs
  xfs: use bitops interface for buf log item AIL flag check
  xfs: fix log reservation overflows when allocating large rt extents
  xfs: stabilize insert range start boundary to avoid COW writeback race
  xfs: fix Sphinx documentation warning

Documentation/admin-guide/xfs.rst

@@ -253,7 +253,7 @@ The following sysctls are available for the XFS filesystem:
 	pool.
 
   fs.xfs.speculative_prealloc_lifetime
-		(Units: seconds   Min: 1  Default: 300  Max: 86400)
+	(Units: seconds   Min: 1  Default: 300  Max: 86400)
 	The interval at which the background scanning for inodes
 	with unused speculative preallocation runs. The scan
 	removes unused preallocation from clean inodes and releases

fs/xfs/libxfs/xfs_alloc.c

@@ -2248,24 +2248,32 @@ xfs_alloc_longest_free_extent(
 	return pag->pagf_flcount > 0 || pag->pagf_longest > 0;
 }
 
+/*
+ * Compute the minimum length of the AGFL in the given AG.  If @pag is NULL,
+ * return the largest possible minimum length.
+ */
 unsigned int
 xfs_alloc_min_freelist(
 	struct xfs_mount	*mp,
 	struct xfs_perag	*pag)
 {
+	/* AG btrees have at least 1 level. */
+	static const uint8_t	fake_levels[XFS_BTNUM_AGF] = {1, 1, 1};
+	const uint8_t		*levels = pag ? pag->pagf_levels : fake_levels;
 	unsigned int		min_free;
 
+	ASSERT(mp->m_ag_maxlevels > 0);
+
 	/* space needed by-bno freespace btree */
-	min_free = min_t(unsigned int, pag->pagf_levels[XFS_BTNUM_BNOi] + 1,
+	min_free = min_t(unsigned int, levels[XFS_BTNUM_BNOi] + 1,
 				       mp->m_ag_maxlevels);
 	/* space needed by-size freespace btree */
-	min_free += min_t(unsigned int, pag->pagf_levels[XFS_BTNUM_CNTi] + 1,
+	min_free += min_t(unsigned int, levels[XFS_BTNUM_CNTi] + 1,
 				       mp->m_ag_maxlevels);
 	/* space needed reverse mapping used space btree */
 	if (xfs_sb_version_hasrmapbt(&mp->m_sb))
-		min_free += min_t(unsigned int,
-				  pag->pagf_levels[XFS_BTNUM_RMAPi] + 1,
-				  mp->m_rmap_maxlevels);
+		min_free += min_t(unsigned int, levels[XFS_BTNUM_RMAPi] + 1,
+						mp->m_rmap_maxlevels);
 
 	return min_free;
 }

fs/xfs/libxfs/xfs_bmap.c

@@ -4561,7 +4561,7 @@ xfs_bmapi_convert_delalloc(
 	struct xfs_mount	*mp = ip->i_mount;
 	xfs_fileoff_t		offset_fsb = XFS_B_TO_FSBT(mp, offset);
 	struct xfs_bmalloca	bma = { NULL };
-	u16			flags = 0;
+	uint16_t		flags = 0;
 	struct xfs_trans	*tp;
 	int			error;
 
@@ -5972,8 +5972,7 @@ xfs_bmap_insert_extents(
 		goto del_cursor;
 	}
 
-	if (XFS_IS_CORRUPT(mp,
-			   stop_fsb >= got.br_startoff + got.br_blockcount)) {
+	if (XFS_IS_CORRUPT(mp, stop_fsb > got.br_startoff)) {
 		error = -EFSCORRUPTED;
 		goto del_cursor;
 	}

fs/xfs/libxfs/xfs_dir2.c

@@ -724,3 +724,24 @@ xfs_dir2_namecheck(
 	/* There shouldn't be any slashes or nulls here */
 	return !memchr(name, '/', length) && !memchr(name, 0, length);
 }
+
+xfs_dahash_t
+xfs_dir2_hashname(
+	struct xfs_mount	*mp,
+	struct xfs_name		*name)
+{
+	if (unlikely(xfs_sb_version_hasasciici(&mp->m_sb)))
+		return xfs_ascii_ci_hashname(name);
+	return xfs_da_hashname(name->name, name->len);
+}
+
+enum xfs_dacmp
+xfs_dir2_compname(
+	struct xfs_da_args	*args,
+	const unsigned char	*name,
+	int			len)
+{
+	if (unlikely(xfs_sb_version_hasasciici(&args->dp->i_mount->m_sb)))
+		return xfs_ascii_ci_compname(args, name, len);
+	return xfs_da_compname(args, name, len);
+}

fs/xfs/libxfs/xfs_dir2_priv.h

@@ -175,6 +175,12 @@ extern int xfs_dir2_sf_lookup(struct xfs_da_args *args);
 extern int xfs_dir2_sf_removename(struct xfs_da_args *args);
 extern int xfs_dir2_sf_replace(struct xfs_da_args *args);
 extern xfs_failaddr_t xfs_dir2_sf_verify(struct xfs_inode *ip);
+int xfs_dir2_sf_entsize(struct xfs_mount *mp,
+		struct xfs_dir2_sf_hdr *hdr, int len);
+void xfs_dir2_sf_put_ino(struct xfs_mount *mp, struct xfs_dir2_sf_hdr *hdr,
+		struct xfs_dir2_sf_entry *sfep, xfs_ino_t ino);
+void xfs_dir2_sf_put_ftype(struct xfs_mount *mp,
+		struct xfs_dir2_sf_entry *sfep, uint8_t ftype);
 
 /* xfs_dir2_readdir.c */
 extern int xfs_readdir(struct xfs_trans *tp, struct xfs_inode *dp,
@@ -194,25 +200,8 @@ xfs_dir2_data_entsize(
 	return round_up(len, XFS_DIR2_DATA_ALIGN);
 }
 
-static inline xfs_dahash_t
-xfs_dir2_hashname(
-	struct xfs_mount	*mp,
-	struct xfs_name		*name)
-{
-	if (unlikely(xfs_sb_version_hasasciici(&mp->m_sb)))
-		return xfs_ascii_ci_hashname(name);
-	return xfs_da_hashname(name->name, name->len);
-}
-
-static inline enum xfs_dacmp
-xfs_dir2_compname(
-	struct xfs_da_args	*args,
-	const unsigned char	*name,
-	int			len)
-{
-	if (unlikely(xfs_sb_version_hasasciici(&args->dp->i_mount->m_sb)))
-		return xfs_ascii_ci_compname(args, name, len);
-	return xfs_da_compname(args, name, len);
-}
+xfs_dahash_t xfs_dir2_hashname(struct xfs_mount *mp, struct xfs_name *name);
+enum xfs_dacmp xfs_dir2_compname(struct xfs_da_args *args,
+		const unsigned char *name, int len);
 
 #endif /* __XFS_DIR2_PRIV_H__ */

fs/xfs/libxfs/xfs_dir2_sf.c

@@ -37,7 +37,7 @@ static void xfs_dir2_sf_check(xfs_da_args_t *args);
 static void xfs_dir2_sf_toino4(xfs_da_args_t *args);
 static void xfs_dir2_sf_toino8(xfs_da_args_t *args);
 
-static int
+int
 xfs_dir2_sf_entsize(
 	struct xfs_mount	*mp,
 	struct xfs_dir2_sf_hdr	*hdr,
@@ -84,7 +84,7 @@ xfs_dir2_sf_get_ino(
 	return get_unaligned_be64(from) & XFS_MAXINUMBER;
 }
 
-static void
+void
 xfs_dir2_sf_put_ino(
 	struct xfs_mount		*mp,
 	struct xfs_dir2_sf_hdr		*hdr,
@@ -145,7 +145,7 @@ xfs_dir2_sf_get_ftype(
 	return XFS_DIR3_FT_UNKNOWN;
 }
 
-static void
+void
 xfs_dir2_sf_put_ftype(
 	struct xfs_mount	*mp,
 	struct xfs_dir2_sf_entry *sfep,

fs/xfs/libxfs/xfs_ialloc.c

@@ -2909,3 +2909,67 @@ xfs_ialloc_setup_geometry(
 	else
 		igeo->ialloc_align = 0;
 }
+
+/* Compute the location of the root directory inode that is laid out by mkfs. */
+xfs_ino_t
+xfs_ialloc_calc_rootino(
+	struct xfs_mount	*mp,
+	int			sunit)
+{
+	struct xfs_ino_geometry	*igeo = M_IGEO(mp);
+	xfs_agblock_t		first_bno;
+
+	/*
+	 * Pre-calculate the geometry of AG 0.  We know what it looks like
+	 * because libxfs knows how to create allocation groups now.
+	 *
+	 * first_bno is the first block in which mkfs could possibly have
+	 * allocated the root directory inode, once we factor in the metadata
+	 * that mkfs formats before it.  Namely, the four AG headers...
+	 */
+	first_bno = howmany(4 * mp->m_sb.sb_sectsize, mp->m_sb.sb_blocksize);
+
+	/* ...the two free space btree roots... */
+	first_bno += 2;
+
+	/* ...the inode btree root... */
+	first_bno += 1;
+
+	/* ...the initial AGFL... */
+	first_bno += xfs_alloc_min_freelist(mp, NULL);
+
+	/* ...the free inode btree root... */
+	if (xfs_sb_version_hasfinobt(&mp->m_sb))
+		first_bno++;
+
+	/* ...the reverse mapping btree root... */
+	if (xfs_sb_version_hasrmapbt(&mp->m_sb))
+		first_bno++;
+
+	/* ...the reference count btree... */
+	if (xfs_sb_version_hasreflink(&mp->m_sb))
+		first_bno++;
+
+	/*
+	 * ...and the log, if it is allocated in the first allocation group.
+	 *
+	 * This can happen with filesystems that only have a single
+	 * allocation group, or very odd geometries created by old mkfs
+	 * versions on very small filesystems.
+	 */
+	if (mp->m_sb.sb_logstart &&
+	    XFS_FSB_TO_AGNO(mp, mp->m_sb.sb_logstart) == 0)
+		 first_bno += mp->m_sb.sb_logblocks;
+
+	/*
+	 * Now round first_bno up to whatever allocation alignment is given
+	 * by the filesystem or was passed in.
+	 */
+	if (xfs_sb_version_hasdalign(&mp->m_sb) && igeo->ialloc_align > 0)
+		first_bno = roundup(first_bno, sunit);
+	else if (xfs_sb_version_hasalign(&mp->m_sb) &&
+			mp->m_sb.sb_inoalignmt > 1)
+		first_bno = roundup(first_bno, mp->m_sb.sb_inoalignmt);
+
+	return XFS_AGINO_TO_INO(mp, 0, XFS_AGB_TO_AGINO(mp, first_bno));
+}

fs/xfs/libxfs/xfs_ialloc.h

@@ -152,5 +152,6 @@ int xfs_inobt_insert_rec(struct xfs_btree_cur *cur, uint16_t holemask,
 
 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);
 
 #endif	/* __XFS_IALLOC_H__ */

fs/xfs/libxfs/xfs_trans_resv.c

@@ -196,6 +196,24 @@ xfs_calc_inode_chunk_res(
 	return res;
 }
 
+/*
+ * Per-extent log reservation for the btree changes involved in freeing or
+ * allocating a realtime extent.  We have to be able to log as many rtbitmap
+ * blocks as needed to mark inuse MAXEXTLEN blocks' worth of realtime extents,
+ * as well as the realtime summary block.
+ */
+static unsigned int
+xfs_rtalloc_log_count(
+	struct xfs_mount	*mp,
+	unsigned int		num_ops)
+{
+	unsigned int		blksz = XFS_FSB_TO_B(mp, 1);
+	unsigned int		rtbmp_bytes;
+
+	rtbmp_bytes = (MAXEXTLEN / mp->m_sb.sb_rextsize) / NBBY;
+	return (howmany(rtbmp_bytes, blksz) + 1) * num_ops;
+}
+
 /*
  * Various log reservation values.
  *
@@ -218,13 +236,21 @@ xfs_calc_inode_chunk_res(
 
 /*
  * In a write transaction we can allocate a maximum of 2
- * extents.  This gives:
+ * extents.  This gives (t1):
  *    the inode getting the new extents: inode size
  *    the inode's bmap btree: max depth * block size
  *    the agfs of the ags from which the extents are allocated: 2 * sector
  *    the superblock free block counter: sector size
  *    the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size
- * And the bmap_finish transaction can free bmap blocks in a join:
+ * Or, if we're writing to a realtime file (t2):
+ *    the inode getting the new extents: inode size
+ *    the inode's bmap btree: max depth * block size
+ *    the agfs of the ags from which the extents are allocated: 2 * sector
+ *    the superblock free block counter: sector size
+ *    the realtime bitmap: ((MAXEXTLEN / rtextsize) / NBBY) bytes
+ *    the realtime summary: 1 block
+ *    the allocation btrees: 2 trees * (2 * max depth - 1) * block size
+ * And the bmap_finish transaction can free bmap blocks in a join (t3):
  *    the agfs of the ags containing the blocks: 2 * sector size
  *    the agfls of the ags containing the blocks: 2 * sector size
  *    the super block free block counter: sector size
@@ -234,40 +260,72 @@ STATIC uint
 xfs_calc_write_reservation(
 	struct xfs_mount	*mp)
 {
-	return XFS_DQUOT_LOGRES(mp) +
-		max((xfs_calc_inode_res(mp, 1) +
+	unsigned int		t1, t2, t3;
+	unsigned int		blksz = XFS_FSB_TO_B(mp, 1);
+
+	t1 = xfs_calc_inode_res(mp, 1) +
+	     xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK), blksz) +
+	     xfs_calc_buf_res(3, mp->m_sb.sb_sectsize) +
+	     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 2), blksz);
+
+	if (xfs_sb_version_hasrealtime(&mp->m_sb)) {
+		t2 = xfs_calc_inode_res(mp, 1) +
 		     xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK),
-				      XFS_FSB_TO_B(mp, 1)) +
+				     blksz) +
 		     xfs_calc_buf_res(3, mp->m_sb.sb_sectsize) +
-		     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 2),
-				      XFS_FSB_TO_B(mp, 1))),
-		    (xfs_calc_buf_res(5, mp->m_sb.sb_sectsize) +
-		     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 2),
-				      XFS_FSB_TO_B(mp, 1))));
+		     xfs_calc_buf_res(xfs_rtalloc_log_count(mp, 1), blksz) +
+		     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 1), blksz);
+	} else {
+		t2 = 0;
+	}
+
+	t3 = xfs_calc_buf_res(5, mp->m_sb.sb_sectsize) +
+	     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 2), blksz);
+
+	return XFS_DQUOT_LOGRES(mp) + max3(t1, t2, t3);
 }
 
 /*
- * In truncating a file we free up to two extents at once.  We can modify:
+ * In truncating a file we free up to two extents at once.  We can modify (t1):
  *    the inode being truncated: inode size
  *    the inode's bmap btree: (max depth + 1) * block size
- * And the bmap_finish transaction can free the blocks and bmap blocks:
+ * And the bmap_finish transaction can free the blocks and bmap blocks (t2):
  *    the agf for each of the ags: 4 * sector size
  *    the agfl for each of the ags: 4 * sector size
  *    the super block to reflect the freed blocks: sector size
  *    worst case split in allocation btrees per extent assuming 4 extents:
  *		4 exts * 2 trees * (2 * max depth - 1) * block size
+ * Or, if it's a realtime file (t3):
+ *    the agf for each of the ags: 2 * sector size
+ *    the agfl for each of the ags: 2 * sector size
+ *    the super block to reflect the freed blocks: sector size
+ *    the realtime bitmap: 2 exts * ((MAXEXTLEN / rtextsize) / NBBY) bytes
+ *    the realtime summary: 2 exts * 1 block
+ *    worst case split in allocation btrees per extent assuming 2 extents:
+ *		2 exts * 2 trees * (2 * max depth - 1) * block size
  */
 STATIC uint
 xfs_calc_itruncate_reservation(
 	struct xfs_mount	*mp)
 {
-	return XFS_DQUOT_LOGRES(mp) +
-		max((xfs_calc_inode_res(mp, 1) +
-		     xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK) + 1,
-				      XFS_FSB_TO_B(mp, 1))),
-		    (xfs_calc_buf_res(9, mp->m_sb.sb_sectsize) +
-		     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 4),
-				      XFS_FSB_TO_B(mp, 1))));
+	unsigned int		t1, t2, t3;
+	unsigned int		blksz = XFS_FSB_TO_B(mp, 1);
+
+	t1 = xfs_calc_inode_res(mp, 1) +
+	     xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK) + 1, blksz);
+
+	t2 = xfs_calc_buf_res(9, mp->m_sb.sb_sectsize) +
+	     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 4), blksz);
+
+	if (xfs_sb_version_hasrealtime(&mp->m_sb)) {
+		t3 = xfs_calc_buf_res(5, mp->m_sb.sb_sectsize) +
+		     xfs_calc_buf_res(xfs_rtalloc_log_count(mp, 2), blksz) +
+		     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 2), blksz);
+	} else {
+		t3 = 0;
+	}
+
+	return XFS_DQUOT_LOGRES(mp) + max3(t1, t2, t3);
 }
 
 /*

fs/xfs/xfs_bmap_util.c

@@ -992,6 +992,7 @@ xfs_prepare_shift(
 	struct xfs_inode	*ip,
 	loff_t			offset)
 {
+	struct xfs_mount	*mp = ip->i_mount;
 	int			error;
 
 	/*
@@ -1004,6 +1005,17 @@ xfs_prepare_shift(
 			return error;
 	}
 
+	/*
+	 * Shift operations must stabilize the start block offset boundary along
+	 * with the full range of the operation. If we don't, a COW writeback
+	 * completion could race with an insert, front merge with the start
+	 * extent (after split) during the shift and corrupt the file. Start
+	 * with the block just prior to the start to stabilize the boundary.
+	 */
+	offset = round_down(offset, 1 << mp->m_sb.sb_blocklog);
+	if (offset)
+		offset -= (1 << mp->m_sb.sb_blocklog);
+
 	/*
 	 * Writeback and invalidate cache for the remainder of the file as we're
 	 * about to shift down every extent from offset to EOF.