From 8d54eacd82a0623a963e0c150ad3b02970638b0d Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Tue, 4 Nov 2025 16:12:00 -0800 Subject: xfs: fix delalloc write failures in software-provided atomic writes With the 20 Oct 2025 release of fstests, generic/521 fails for me on regular (aka non-block-atomic-writes) storage: QA output created by 521 dowrite: write: Input/output error LOG DUMP (8553 total operations): 1( 1 mod 256): SKIPPED (no operation) 2( 2 mod 256): WRITE 0x7e000 thru 0x8dfff (0x10000 bytes) HOLE 3( 3 mod 256): READ 0x69000 thru 0x79fff (0x11000 bytes) 4( 4 mod 256): FALLOC 0x53c38 thru 0x5e853 (0xac1b bytes) INTERIOR 5( 5 mod 256): COPY 0x55000 thru 0x59fff (0x5000 bytes) to 0x25000 thru 0x29fff 6( 6 mod 256): WRITE 0x74000 thru 0x88fff (0x15000 bytes) 7( 7 mod 256): ZERO 0xedb1 thru 0x11693 (0x28e3 bytes) with a warning in dmesg from iomap about XFS trying to give it a delalloc mapping for a directio write. Fix the software atomic write iomap_begin code to convert the reservation into a written mapping. This doesn't fix the data corruption problems reported by generic/760, but it's a start. Cc: stable@vger.kernel.org # v6.16 Fixes: bd1d2c21d5d249 ("xfs: add xfs_atomic_write_cow_iomap_begin()") Signed-off-by: Darrick J. Wong Reviewed-by: John Garry Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_iomap.c | 21 +++++++++++++++++++-- 1 file changed, 19 insertions(+), 2 deletions(-) diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index d3f6e3e42a11..788bfdce608a 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -1130,7 +1130,7 @@ xfs_atomic_write_cow_iomap_begin( return -EAGAIN; trace_xfs_iomap_atomic_write_cow(ip, offset, length); - +retry: xfs_ilock(ip, XFS_ILOCK_EXCL); if (!ip->i_cowfp) { @@ -1141,6 +1141,8 @@ xfs_atomic_write_cow_iomap_begin( if (!xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb, &icur, &cmap)) cmap.br_startoff = end_fsb; if (cmap.br_startoff <= offset_fsb) { + if (isnullstartblock(cmap.br_startblock)) + goto convert_delay; xfs_trim_extent(&cmap, offset_fsb, count_fsb); goto found; } @@ -1169,8 +1171,10 @@ xfs_atomic_write_cow_iomap_begin( if (!xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb, &icur, &cmap)) cmap.br_startoff = end_fsb; if (cmap.br_startoff <= offset_fsb) { - xfs_trim_extent(&cmap, offset_fsb, count_fsb); xfs_trans_cancel(tp); + if (isnullstartblock(cmap.br_startblock)) + goto convert_delay; + xfs_trim_extent(&cmap, offset_fsb, count_fsb); goto found; } @@ -1210,6 +1214,19 @@ found: xfs_iunlock(ip, XFS_ILOCK_EXCL); return xfs_bmbt_to_iomap(ip, iomap, &cmap, flags, IOMAP_F_SHARED, seq); +convert_delay: + xfs_iunlock(ip, XFS_ILOCK_EXCL); + error = xfs_bmapi_convert_delalloc(ip, XFS_COW_FORK, offset, iomap, + NULL); + if (error) + return error; + + /* + * Try the lookup again, because the delalloc conversion might have + * turned the COW mapping into unwritten, but we need it to be in + * written state. + */ + goto retry; out_unlock: xfs_iunlock(ip, XFS_ILOCK_EXCL); return error; -- cgit v1.2.3 From 8d7bba1e8314013ecc817a91624104ceb9352ddc Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Tue, 4 Nov 2025 16:15:38 -0800 Subject: xfs: fix various problems in xfs_atomic_write_cow_iomap_begin I think there are several things wrong with this function: A) xfs_bmapi_write can return a much larger unwritten mapping than what the caller asked for. We convert part of that range to written, but return the entire written mapping to iomap even though that's inaccurate. B) The arguments to xfs_reflink_convert_cow_locked are wrong -- an unwritten mapping could be *smaller* than the write range (or even the hole range). In this case, we convert too much file range to written state because we then return a smaller mapping to iomap. C) It doesn't handle delalloc mappings. This I covered in the patch that I already sent to the list. D) Reassigning count_fsb to handle the hole means that if the second cmap lookup attempt succeeds (due to racing with someone else) we trim the mapping more than is strictly necessary. The changing meaning of count_fsb makes this harder to notice. E) The tracepoint is kinda wrong because @length is mutated. That makes it harder to chase the data flows through this function because you can't just grep on the pos/bytecount strings. F) We don't actually check that the br_state = XFS_EXT_NORM assignment is accurate, i.e that the cow fork actually contains a written mapping for the range we're interested in G) Somewhat inadequate documentation of why we need to xfs_trim_extent so aggressively in this function. H) Not sure why xfs_iomap_end_fsb is used here, the vfs already clamped the write range to s_maxbytes. Fix these issues, and then the atomic writes regressions in generic/760, generic/617, generic/091, generic/263, and generic/521 all go away for me. Cc: stable@vger.kernel.org # v6.16 Fixes: bd1d2c21d5d249 ("xfs: add xfs_atomic_write_cow_iomap_begin()") Signed-off-by: Darrick J. Wong Reviewed-by: John Garry Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_iomap.c | 61 ++++++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 50 insertions(+), 11 deletions(-) diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 788bfdce608a..490e12cb99be 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -1091,6 +1091,29 @@ const struct iomap_ops xfs_zoned_direct_write_iomap_ops = { }; #endif /* CONFIG_XFS_RT */ +#ifdef DEBUG +static void +xfs_check_atomic_cow_conversion( + struct xfs_inode *ip, + xfs_fileoff_t offset_fsb, + xfs_filblks_t count_fsb, + const struct xfs_bmbt_irec *cmap) +{ + struct xfs_iext_cursor icur; + struct xfs_bmbt_irec cmap2 = { }; + + if (xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb, &icur, &cmap2)) + xfs_trim_extent(&cmap2, offset_fsb, count_fsb); + + ASSERT(cmap2.br_startoff == cmap->br_startoff); + ASSERT(cmap2.br_blockcount == cmap->br_blockcount); + ASSERT(cmap2.br_startblock == cmap->br_startblock); + ASSERT(cmap2.br_state == cmap->br_state); +} +#else +# define xfs_check_atomic_cow_conversion(...) ((void)0) +#endif + static int xfs_atomic_write_cow_iomap_begin( struct inode *inode, @@ -1102,9 +1125,10 @@ xfs_atomic_write_cow_iomap_begin( { struct xfs_inode *ip = XFS_I(inode); struct xfs_mount *mp = ip->i_mount; - const xfs_fileoff_t offset_fsb = XFS_B_TO_FSBT(mp, offset); - xfs_fileoff_t end_fsb = xfs_iomap_end_fsb(mp, offset, length); - xfs_filblks_t count_fsb = end_fsb - offset_fsb; + const xfs_fileoff_t offset_fsb = XFS_B_TO_FSBT(mp, offset); + const xfs_fileoff_t end_fsb = XFS_B_TO_FSB(mp, offset + length); + const xfs_filblks_t count_fsb = end_fsb - offset_fsb; + xfs_filblks_t hole_count_fsb; int nmaps = 1; xfs_filblks_t resaligned; struct xfs_bmbt_irec cmap; @@ -1143,14 +1167,20 @@ retry: if (cmap.br_startoff <= offset_fsb) { if (isnullstartblock(cmap.br_startblock)) goto convert_delay; + + /* + * cmap could extend outside the write range due to previous + * speculative preallocations. We must trim cmap to the write + * range because the cow fork treats written mappings to mean + * "write in progress". + */ xfs_trim_extent(&cmap, offset_fsb, count_fsb); goto found; } - end_fsb = cmap.br_startoff; - count_fsb = end_fsb - offset_fsb; + hole_count_fsb = cmap.br_startoff - offset_fsb; - resaligned = xfs_aligned_fsb_count(offset_fsb, count_fsb, + resaligned = xfs_aligned_fsb_count(offset_fsb, hole_count_fsb, xfs_get_cowextsz_hint(ip)); xfs_iunlock(ip, XFS_ILOCK_EXCL); @@ -1186,7 +1216,7 @@ retry: * atomic writes to that same range will be aligned (and don't require * this COW-based method). */ - error = xfs_bmapi_write(tp, ip, offset_fsb, count_fsb, + error = xfs_bmapi_write(tp, ip, offset_fsb, hole_count_fsb, XFS_BMAPI_COWFORK | XFS_BMAPI_PREALLOC | XFS_BMAPI_EXTSZALIGN, 0, &cmap, &nmaps); if (error) { @@ -1199,17 +1229,26 @@ retry: if (error) goto out_unlock; + /* + * cmap could map more blocks than the range we passed into bmapi_write + * because of EXTSZALIGN or adjacent pre-existing unwritten mappings + * that were merged. Trim cmap to the original write range so that we + * don't convert more than we were asked to do for this write. + */ + xfs_trim_extent(&cmap, offset_fsb, count_fsb); + found: if (cmap.br_state != XFS_EXT_NORM) { - error = xfs_reflink_convert_cow_locked(ip, offset_fsb, - count_fsb); + error = xfs_reflink_convert_cow_locked(ip, cmap.br_startoff, + cmap.br_blockcount); if (error) goto out_unlock; cmap.br_state = XFS_EXT_NORM; + xfs_check_atomic_cow_conversion(ip, offset_fsb, count_fsb, + &cmap); } - length = XFS_FSB_TO_B(mp, cmap.br_startoff + cmap.br_blockcount); - trace_xfs_iomap_found(ip, offset, length - offset, XFS_COW_FORK, &cmap); + trace_xfs_iomap_found(ip, offset, length, XFS_COW_FORK, &cmap); seq = xfs_iomap_inode_sequence(ip, IOMAP_F_SHARED); xfs_iunlock(ip, XFS_ILOCK_EXCL); return xfs_bmbt_to_iomap(ip, iomap, &cmap, flags, IOMAP_F_SHARED, seq); -- cgit v1.2.3 From f5714a3c1a5658251360603231efe1bee21f9c2c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 4 Nov 2025 08:50:53 -0500 Subject: xfs: fix a rtgroup leak when xfs_init_zone fails Drop the rtgrop reference when xfs_init_zone fails for a conventional device. Fixes: 4e4d52075577 ("xfs: add the zoned space allocator") Signed-off-by: Christoph Hellwig Reviewed-by: Hans Holmberg Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_zone_alloc.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/xfs/xfs_zone_alloc.c b/fs/xfs/xfs_zone_alloc.c index 040402240807..fa76fa4ba0f9 100644 --- a/fs/xfs/xfs_zone_alloc.c +++ b/fs/xfs/xfs_zone_alloc.c @@ -1249,8 +1249,10 @@ xfs_mount_zones( while ((rtg = xfs_rtgroup_next(mp, rtg))) { error = xfs_init_zone(&iz, rtg, NULL); - if (error) + if (error) { + xfs_rtgroup_rele(rtg); goto out_free_zone_info; + } } } -- cgit v1.2.3 From 21ab5179aafa2ded7f3851bfe7e043f8a3b6199d Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 3 Nov 2025 05:15:13 -0500 Subject: xfs: fix zone selection in xfs_select_open_zone_mru xfs_select_open_zone_mru needs to pass XFS_ZONE_ALLOC_OK to xfs_try_use_zone because we only want to tightly pack into zones of the same or a compatible temperature instead of any available zone. This got broken in commit 0301dae732a5 ("xfs: refactor hint based zone allocation"), which failed to update this particular caller when switching to an enum. xfs/638 sometimes, but not reliably fails due to this change. Fixes: 0301dae732a5 ("xfs: refactor hint based zone allocation") Signed-off-by: Christoph Hellwig Reviewed-by: Hans Holmberg Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_zone_alloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/xfs/xfs_zone_alloc.c b/fs/xfs/xfs_zone_alloc.c index fa76fa4ba0f9..ef7a931ebde5 100644 --- a/fs/xfs/xfs_zone_alloc.c +++ b/fs/xfs/xfs_zone_alloc.c @@ -615,7 +615,7 @@ xfs_select_open_zone_mru( lockdep_assert_held(&zi->zi_open_zones_lock); list_for_each_entry_reverse(oz, &zi->zi_open_zones, oz_entry) - if (xfs_try_use_zone(zi, file_hint, oz, false)) + if (xfs_try_use_zone(zi, file_hint, oz, XFS_ZONE_ALLOC_OK)) return oz; cond_resched_lock(&zi->zi_open_zones_lock); -- cgit v1.2.3 From d8a823c6f04ef03e3bd7249d2e796da903e7238d Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 4 Nov 2025 05:43:01 -0500 Subject: xfs: free xfs_busy_extents structure when no RT extents are queued kmemleak occasionally reports leaking xfs_busy_extents structure from xfs_scrub calls after running xfs/528 (but attributed to following tests), which seems to be caused by not freeing the xfs_busy_extents structure when tr.queued is 0 and xfs_trim_rtgroup_extents breaks out of the main loop. Free the structure in this case. Fixes: a3315d11305f ("xfs: use rtgroup busy extent list for FITRIM") Signed-off-by: Christoph Hellwig Reviewed-by: Darrick J. Wong Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_discard.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c index ee49f20875af..6917de832191 100644 --- a/fs/xfs/xfs_discard.c +++ b/fs/xfs/xfs_discard.c @@ -726,8 +726,10 @@ xfs_trim_rtgroup_extents( break; } - if (!tr.queued) + if (!tr.queued) { + kfree(tr.extents); break; + } /* * We hand the extent list to the discard function here so the -- cgit v1.2.3