xfs: fix internal error from AGFL exhaustion
This commit is contained in:
liuh
parent
abbe982872
commit
7d6f4a7c58
107
0002-xfs-fix-internal-error-from-AGFL-exhaustion.patch
Normal file
107
0002-xfs-fix-internal-error-from-AGFL-exhaustion.patch
Normal file
@ -0,0 +1,107 @@
|
||||
From 8fdde9dd48edc32a2f5e5a061a12e8dad54c3c7e Mon Sep 17 00:00:00 2001
|
||||
From: Omar Sandoval <osandov@fb.com>
|
||||
Date: Thu, 15 Feb 2024 09:27:54 +0100
|
||||
Subject: xfs: fix internal error from AGFL exhaustion
|
||||
|
||||
Source kernel commit: f63a5b3769ad7659da4c0420751d78958ab97675
|
||||
|
||||
We've been seeing XFS errors like the following:
|
||||
|
||||
XFS: Internal error i != 1 at line 3526 of file fs/xfs/libxfs/xfs_btree.c. Caller xfs_btree_insert+0x1ec/0x280
|
||||
...
|
||||
Call Trace:
|
||||
xfs_corruption_error+0x94/0xa0
|
||||
xfs_btree_insert+0x221/0x280
|
||||
xfs_alloc_fixup_trees+0x104/0x3e0
|
||||
xfs_alloc_ag_vextent_size+0x667/0x820
|
||||
xfs_alloc_fix_freelist+0x5d9/0x750
|
||||
xfs_free_extent_fix_freelist+0x65/0xa0
|
||||
__xfs_free_extent+0x57/0x180
|
||||
...
|
||||
|
||||
This is the XFS_IS_CORRUPT() check in xfs_btree_insert() when
|
||||
xfs_btree_insrec() fails.
|
||||
|
||||
After converting this into a panic and dissecting the core dump, I found
|
||||
that xfs_btree_insrec() is failing because it's trying to split a leaf
|
||||
node in the cntbt when the AG free list is empty. In particular, it's
|
||||
failing to get a block from the AGFL _while trying to refill the AGFL_.
|
||||
|
||||
If a single operation splits every level of the bnobt and the cntbt (and
|
||||
the rmapbt if it is enabled) at once, the free list will be empty. Then,
|
||||
when the next operation tries to refill the free list, it allocates
|
||||
space. If the allocation does not use a full extent, it will need to
|
||||
insert records for the remaining space in the bnobt and cntbt. And if
|
||||
those new records go in full leaves, the leaves (and potentially more
|
||||
nodes up to the old root) need to be split.
|
||||
|
||||
Fix it by accounting for the additional splits that may be required to
|
||||
refill the free list in the calculation for the minimum free list size.
|
||||
|
||||
P.S. As far as I can tell, this bug has existed for a long time -- maybe
|
||||
back to xfs-history commit afdf80ae7405 ("Add XFS_AG_MAXLEVELS macros
|
||||
...") in April 1994! It requires a very unlucky sequence of events, and
|
||||
in fact we didn't hit it until a particular sparse mmap workload updated
|
||||
from 5.12 to 5.19. But this bug existed in 5.12, so it must've been
|
||||
exposed by some other change in allocation or writeback patterns. It's
|
||||
also much less likely to be hit with the rmapbt enabled, since that
|
||||
increases the minimum free list size and is unlikely to split at the
|
||||
same time as the bnobt and cntbt.
|
||||
|
||||
Reviewed-by: "Darrick J. Wong" <djwong@kernel.org>
|
||||
Reviewed-by: Dave Chinner <dchinner@redhat.com>
|
||||
Signed-off-by: Omar Sandoval <osandov@fb.com>
|
||||
Signed-off-by: Chandan Babu R <chandanbabu@kernel.org>
|
||||
Signed-off-by: Carlos Maiolino <cem@kernel.org>
|
||||
---
|
||||
libxfs/xfs_alloc.c | 27 ++++++++++++++++++++++++---
|
||||
1 file changed, 24 insertions(+), 3 deletions(-)
|
||||
|
||||
diff --git a/libxfs/xfs_alloc.c b/libxfs/xfs_alloc.c
|
||||
index 4519a05551..7ac7c2f6cc 100644
|
||||
--- a/libxfs/xfs_alloc.c
|
||||
+++ b/libxfs/xfs_alloc.c
|
||||
@@ -2271,16 +2271,37 @@ xfs_alloc_min_freelist(
|
||||
|
||||
ASSERT(mp->m_alloc_maxlevels > 0);
|
||||
|
||||
+ /*
|
||||
+ * For a btree shorter than the maximum height, the worst case is that
|
||||
+ * every level gets split and a new level is added, then while inserting
|
||||
+ * another entry to refill the AGFL, every level under the old root gets
|
||||
+ * split again. This is:
|
||||
+ *
|
||||
+ * (full height split reservation) + (AGFL refill split height)
|
||||
+ * = (current height + 1) + (current height - 1)
|
||||
+ * = (new height) + (new height - 2)
|
||||
+ * = 2 * new height - 2
|
||||
+ *
|
||||
+ * For a btree of maximum height, the worst case is that every level
|
||||
+ * under the root gets split, then while inserting another entry to
|
||||
+ * refill the AGFL, every level under the root gets split again. This is
|
||||
+ * also:
|
||||
+ *
|
||||
+ * 2 * (current height - 1)
|
||||
+ * = 2 * (new height - 1)
|
||||
+ * = 2 * new height - 2
|
||||
+ */
|
||||
+
|
||||
/* space needed by-bno freespace btree */
|
||||
min_free = min_t(unsigned int, levels[XFS_BTNUM_BNOi] + 1,
|
||||
- mp->m_alloc_maxlevels);
|
||||
+ mp->m_alloc_maxlevels) * 2 - 2;
|
||||
/* space needed by-size freespace btree */
|
||||
min_free += min_t(unsigned int, levels[XFS_BTNUM_CNTi] + 1,
|
||||
- mp->m_alloc_maxlevels);
|
||||
+ mp->m_alloc_maxlevels) * 2 - 2;
|
||||
/* space needed reverse mapping used space btree */
|
||||
if (xfs_has_rmapbt(mp))
|
||||
min_free += min_t(unsigned int, levels[XFS_BTNUM_RMAPi] + 1,
|
||||
- mp->m_rmap_maxlevels);
|
||||
+ mp->m_rmap_maxlevels) * 2 - 2;
|
||||
|
||||
return min_free;
|
||||
}
|
||||
--
|
||||
cgit 1.2.3-korg
|
||||
|
||||
@ -1,6 +1,6 @@
|
||||
Name: xfsprogs
|
||||
Version: 6.6.0
|
||||
Release: 1
|
||||
Release: 2
|
||||
Summary: Administration and debugging tools for the XFS file system
|
||||
License: GPL+ and LGPLv2+
|
||||
URL: https://xfs.wiki.kernel.org
|
||||
@ -20,6 +20,7 @@ Conflicts: xfsdump < 3.0.1
|
||||
|
||||
Patch0: xfsprogs-5.12.0-default-bigtime-inobtcnt-on.patch
|
||||
Patch1: 0001-xfsprogs-Add-sw64-architecture.patch
|
||||
Patch2: 0002-xfs-fix-internal-error-from-AGFL-exhaustion.patch
|
||||
|
||||
%description
|
||||
xfsprogs are the userspace utilities that manage XFS filesystems.
|
||||
@ -101,6 +102,9 @@ rm -rf %{buildroot}%{_datadir}/doc/xfsprogs/
|
||||
|
||||
|
||||
%changelog
|
||||
* Fri Mar 29 2024 liuh <liuhuan01@kylinos.cn> - 6.6.0-2
|
||||
- sync patch from community
|
||||
|
||||
* Tue Feb 27 2024 wuguanghao <wuguanghao3@huawei.com> - 6.6.0-1
|
||||
- upgrade version to 6.6.0
|
||||
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user