xfs: fix livelock in delayed allocation at ENOSPC
PATCH 1/1 xfs: fix livelock in delayed allocation at ENOSPC
Hi Greg,
A regression in 6.3.0 has been identified in XFS that causes filesystem corruption. It has been seen in the wild by a number of users, and bisected down to an issued we'd already fixed in 6.4-rc1 with commit:
9419092fb263 ("xfs: fix livelock in delayed allocation at ENOSPC")
This was reported with much less harmful symptoms (alloctor livelock) and it wasn't realised that it could have other, more impactful symptoms. A reproducer for the corruption was found yesterday and, soon after than, the cause of the corruption reports was identified.
The commit applies cleanly to a 6.3.0 kernel here, so it should also apply cleanly to a current 6.3.x kernel. I've included the entire commit below in case that's easier for you.
Can you please pull this commit into the next 6.3.x release as a matter of priority?
Cheers,
Dave.
Signature · 38 lines
Dave Chinner [email protected] xfs: fix livelock in delayed allocation at ENOSPC From: Dave Chinner <[email protected]> On a filesystem with a non-zero stripe unit and a large sequential write, delayed allocation will set a minimum allocation length of the stripe unit. If allocation fails because there are no extents long enough for an aligned minlen allocation, it is supposed to fall back to unaligned allocation which allows single block extents to be allocated. When the allocator code was rewritting in the 6.3 cycle, this fallback was broken - the old code used args->fsbno as the both the allocation target and the allocation result, the new code passes the target as a separate parameter. The conversion didn't handle the aligned->unaligned fallback path correctly - it reset args->fsbno to the target fsbno on failure which broke allocation failure detection in the high level code and so it never fell back to unaligned allocations. This resulted in a loop in writeback trying to allocate an aligned block, getting a false positive success, trying to insert the result in the BMBT. This did nothing because the extent already was in the BMBT (merge results in an unchanged extent) and so it returned the prior extent to the conversion code as the current iomap. Because the iomap returned didn't cover the offset we tried to map, xfs_convert_blocks() then retries the allocation, which fails in the same way and now we have a livelock. Reported-and-tested-by: Brian Foster <[email protected]> Fixes: 85843327094f ("xfs: factor xfs_bmap_btalloc()") Signed-off-by: Dave Chinner <[email protected]> Reviewed-by: Darrick J. Wong <[email protected]> ---
fs/xfs/libxfs/xfs_bmap.c +0 -1
@@ -3540,7 +3540,6 @@ xfs_bmap_btalloc_at_eof(* original non-aligned state so the caller can proceed on allocation * failure as if this function was never called. */ - args->fsbno = ap->blkno; args->alignment =1 ;return 0 ; }
kernel test robot
build botFYI: kernel test robot notices the stable kernel rule is not satisfied.
Hi,
Thanks for your patch.
FYI: kernel test robot notices the stable kernel rule is not satisfied.
Rule: 'Cc: [email protected]' or 'commit <sha1> upstream.'
Subject: [PATCH, STABLE 6.3.x] xfs: fix livelock in delayed allocation at ENOSPC
The check is based on https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
Signature · 2 lines
0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki
Greg KH
maintainerQuoted 15 lines · Dave Chinner
Hi Greg,
A regression in 6.3.0 has been identified in XFS that causes filesystem corruption. It has been seen in the wild by a number of users, and bisected down to an issued we'd already fixed in 6.4-rc1 with commit:
9419092fb263("xfs: fix livelock in delayed allocation at ENOSPC")This was reported with much less harmful symptoms (alloctor livelock) and it wasn't realised that it could have other, more impactful symptoms. A reproducer for the corruption was found yesterday and, soon after than, the cause of the corruption reports was identified.
Quoted 6 lines
The commit applies cleanly to a 6.3.0 kernel here, so it should also apply cleanly to a current 6.3.x kernel. I've included the entire commit below in case that's easier for you.
Can you please pull this commit into the next 6.3.x release as a matter of priority?
Now queued up, thanks.
greg k-h