Re: [issue] low version kernel crashes after adding support for unshare on XFS testing

[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]



On Mon, Nov 11, 2024 at 11:01:58AM +0800, Ma Xinjian wrote:
> Hi all
> 
> Since xfstests-dev commit 34ce77ce867d27ad7ad9a180489e4311c3cd15e4("fsx: support unshare range fallocate mode") and
> ff07e0a329361c752d77d0f446d247f0ebdb5418("fsstress: add support for FALLOC_FL_UNSHARE_RANGE")
> when running the testcases which use fsx or fsstress on XFS filesystem on Rocky 8.10 (e.g. generic/075 and generic/013),
> the kernel crashes with the following call_trace, could someone help take a look?
> 
> [ 4670.432330] INFO: task fsstress:336323 blocked for more than 120 seconds.
> [ 4670.432334]       Not tainted 4.18.0-553.el8_10.x86_64 #1

               Contact your vendor ^^^^^^^^^^^^^^ for support.

They probably need to backport the funshare input validation fixes that
recently landed in 6.12-rc.

--D

> [ 4670.432335] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> [ 4670.432336] task:fsstress        state:D stack:0     pid:336323 ppid:336321 flags:0x00000080
> [ 4670.432339] Call Trace:
> [ 4670.432341]  __schedule+0x2d1/0x870
> [ 4670.432347]  schedule+0x55/0xf0
> [ 4670.432349]  schedule_preempt_disabled+0xa/0x10
> [ 4670.432352]  rwsem_down_write_slowpath+0x370/0x570
> [ 4670.432356]  ? xfs_buffered_write_iomap_end+0x5c/0xe0 [xfs]
> [ 4670.432400]  down_write+0x29/0x50
> [ 4670.432401]  xfs_ilock+0x98/0x100 [xfs]
> [ 4670.432435]  xfs_buffered_write_iomap_end+0x5c/0xe0 [xfs]
> [ 4670.432467]  ? xfs_iunlock+0xcb/0x110 [xfs]
> [ 4670.432499]  iomap_iter+0x77/0x360
> [ 4670.432502]  iomap_file_unshare+0x76/0x1c0
> [ 4670.432504]  ? kmem_cache_free+0x116/0x300
> [ 4670.432509]  xfs_reflink_unshare+0x55/0x150 [xfs]
> [ 4670.432543]  xfs_file_fallocate+0x3b3/0x480 [xfs]
> [ 4670.432575]  ? __inode_security_revalidate+0x63/0x80
> [ 4670.432578]  ? selinux_file_permission+0x10c/0x140
> [ 4670.432581]  vfs_fallocate+0x140/0x280
> [ 4670.432584]  ksys_fallocate+0x3c/0x80
> [ 4670.432586]  __x64_sys_fallocate+0x1a/0x30
> [ 4670.432588]  do_syscall_64+0x5b/0x1b0
> [ 4670.432591]  entry_SYSCALL_64_after_hwframe+0x61/0xc6
> [ 4670.432594] RIP: 0033:0x7f0b8f8d662b
> [ 4670.432598] Code: Unable to access opcode bytes at RIP 0x7f0b8f8d6601.
> [ 4670.432599] RSP: 002b:00007ffee0ad3da8 EFLAGS: 00000246 ORIG_RAX: 000000000000011d
> [ 4670.432601] RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007f0b8f8d662b
> [ 4670.432602] RDX: 00000000001d6c8b RSI: 0000000000000040 RDI: 0000000000000003
> [ 4670.432603] RBP: 0000000000000080 R08: 000000000000002a R09: 0000000000000000
> [ 4670.432604] R10: 00000000000967a4 R11: 0000000000000246 R12: 0000000000000040
> [ 4670.432605] R13: 00000000000967a4 R14: 00000000001d6c8b R15: 00007f0b901906b8
> 
> Best regards
> Ma
> 
> 
> 




[Index of Archives]     [Linux Filesystems Development]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux