From f0322c7cc05eb23ef034775f9b39254cbd4f3678 Mon Sep 17 00:00:00 2001
From: "Darrick J. Wong" <darrick.wong@oracle.com>
Date: Sat, 23 May 2020 09:43:30 -0700
Subject: [PATCH] xfs: measure all contiguous previous extents for prealloc
 size
Git-commit: f0322c7cc05eb23ef034775f9b39254cbd4f3678
Patch-mainline: v5.8-rc1
References: bsc#1158242

When we're estimating a new speculative preallocation length for an
extending write, we should walk backwards through the extent list to
determine the number of number of blocks that are physically and
logically contiguous with the write offset, and use that as an input to
the preallocation size computation.

This way, preallocation length is truly measured by the effectiveness of
the allocator in giving us contiguous allocations without being
influenced by the state of a given extent.  This fixes both the problem
where ZERO_RANGE within an EOF can reduce preallocation, and prevents
the unnecessary shrinkage of preallocation when delalloc extents are
turned into unwritten extents.

This was found as a regression in xfs/014 after changing delalloc writes
to create unwritten extents during writeback.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Acked-by: Anthony Iliopoulos <ailiop@suse.com>

---
 fs/xfs/xfs_iomap.c |   40 +++++++++++++++++++++++++++-------------
 1 file changed, 27 insertions(+), 13 deletions(-)

--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -397,15 +397,17 @@
 	loff_t			count,
 	struct xfs_iext_cursor	*icur)
 {
+	struct xfs_iext_cursor	ncur = *icur;
+	struct xfs_bmbt_irec	prev, got;
 	struct xfs_mount	*mp = ip->i_mount;
 	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
 	xfs_fileoff_t		offset_fsb = XFS_B_TO_FSBT(mp, offset);
-	struct xfs_bmbt_irec	prev;
-	int			shift = 0;
 	int64_t			freesp;
 	xfs_fsblock_t		qblocks;
-	int			qshift = 0;
 	xfs_fsblock_t		alloc_blocks = 0;
+	xfs_extlen_t		plen;
+	int			shift = 0;
+	int			qshift = 0;
 
 	if (offset + count <= XFS_ISIZE(ip))
 		return 0;
@@ -420,7 +422,7 @@
 	 */
 	if ((mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) ||
 	    XFS_ISIZE(ip) < XFS_FSB_TO_B(mp, mp->m_dalign) ||
-	    !xfs_iext_peek_prev_extent(ifp, icur, &prev) ||
+	    !xfs_iext_prev_extent(ifp, &ncur, &prev) ||
 	    prev.br_startoff + prev.br_blockcount < offset_fsb)
 		return mp->m_writeio_blocks;
 
@@ -433,16 +435,28 @@
 	 * preallocation size.
 	 *
 	 * If the extent is a hole, then preallocation is essentially disabled.
-	 * Otherwise we take the size of the preceding data extent as the basis
-	 * for the preallocation size. If the size of the extent is greater than
-	 * half the maximum extent length, then use the current offset as the
-	 * basis. This ensures that for large files the preallocation size
-	 * always extends to MAXEXTLEN rather than falling short due to things
-	 * like stripe unit/width alignment of real extents.
+	 * Otherwise we take the size of the preceding data extents as the basis
+	 * for the preallocation size. Note that we don't care if the previous
+	 * extents are written or not.
+	 *
+	 * If the size of the extents is greater than half the maximum extent
+	 * length, then use the current offset as the basis. This ensures that
+	 * for large files the preallocation size always extends to MAXEXTLEN
+	 * rather than falling short due to things like stripe unit/width
+	 * alignment of real extents.
 	 */
-	if (prev.br_blockcount <= (MAXEXTLEN >> 1))
-		alloc_blocks = prev.br_blockcount << 1;
-	else
+	plen = prev.br_blockcount;
+	while (xfs_iext_prev_extent(ifp, &ncur, &got)) {
+		if (plen > MAXEXTLEN / 2 ||
+		    isnullstartblock(got.br_startblock) ||
+		    got.br_startoff + got.br_blockcount != prev.br_startoff ||
+		    got.br_startblock + got.br_blockcount != prev.br_startblock)
+			break;
+		plen += got.br_blockcount;
+		prev = got;
+	}
+	alloc_blocks = plen * 2;
+	if (alloc_blocks > MAXEXTLEN)
 		alloc_blocks = XFS_B_TO_FSB(mp, offset);
 	if (!alloc_blocks)
 		goto check_writeio;