* softlockup in xfstests 083
@ 2012-11-22 19:36 Christoph Hellwig
2012-11-26 13:30 ` Lukáš Czerner
0 siblings, 1 reply; 3+ messages in thread
From: Christoph Hellwig @ 2012-11-22 19:36 UTC (permalink / raw)
To: linux-ext4
Running xfstests on ext4 using current Linus' tree I can reasily
reproduce the following softlockup in xfstests 083:
083 9s ...[ 1124.995740] EXT4-fs (vdb2): mounted filesystem with ordered
data mode. Opts: acl,user_xattr
[ 1156.073356] BUG: soft lockup - CPU#0 stuck for 22s! [fsstress:912]
[ 1156.074733] Modules linked in:
[ 1156.075471] CPU 0
[ 1156.075887] Pid: 912, comm: fsstress Not tainted 3.7.0-rc6+ #212 Bochs Bochs
[ 1156.076558] RIP: 0010:[<ffffffff81152b64>] [<ffffffff81152b64>] kfree+0xc4/0x140
[ 1156.076558] RSP: 0018:ffff880047235c58 EFLAGS: 00000282
[ 1156.076558] RAX: 0000000000000056 RBX: ffff880047235bd0 RCX: 0000000000000065
[ 1156.076558] RDX: 0000000000000055 RSI: 0000000000000046 RDI: 0000000000000000
[ 1156.076558] RBP: ffff880047235c78 R08: ffe2001390ac0000 R09: 002dffe4ef58e42b
[ 1156.076558] R10: ff8707731e20e802 R11: 000000000000086d R12: ffffffff8167f60c
[ 1156.076558] R13: ffff880047235c48 R14: 0000000000000046 R15: ffff880057594400
[ 1156.076558] FS: 0000000000000000(0000) GS:ffff88005d800000(0063) knlGS:00000000f75be6c0
[ 1156.076558] CS: 0010 DS: 002b ES: 002b CR0: 0000000080050033
[ 1156.076558] CR2: 00000000f714b004 CR3: 00000000592cc000 CR4: 00000000000006f0
[ 1156.076558] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 1156.076558] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[ 1156.076558] Process fsstress (pid: 912, threadinfo ffff880047234000, task ffff8800582bc0c0)
[ 1156.076558] Stack:
[ 1156.076558] 0000000000000001 00000000ffffffff 0000000000000008 ffff88005aed6648
[ 1156.076558] ffff880047235d78 ffffffff812547fc ffff880053965390 ffff88005aed6648
[ 1156.076558] ffff880047235d30 ffff88005af8a3a0 ffff88005aed6598 ffff88003a5fe900
[ 1156.076558] Call Trace:
[ 1156.076558] [<ffffffff812547fc>] ext4_ext_remove_space+0x49c/0xeb0
[ 1156.076558] [<ffffffff81257251>] ? ext4_ext_truncate+0x151/0x1b0
[ 1156.076558] [<ffffffff81257273>] ext4_ext_truncate+0x173/0x1b0
[ 1156.076558] [<ffffffff81229445>] ext4_truncate+0xd5/0xe0
[ 1156.076558] [<ffffffff8122c3ba>] ext4_setattr+0x37a/0x570
[ 1156.076558] [<ffffffff810882c1>] ? current_fs_time+0x11/0x60
[ 1156.076558] [<ffffffff8116f024>] notify_change+0x164/0x390
[ 1156.076558] [<ffffffff81164723>] ? user_path_at_empty+0x63/0xa0
[ 1156.076558] [<ffffffff81154a2d>] do_truncate+0x5d/0xa0
[ 1156.076558] [<ffffffff81154bf4>] sys_truncate+0x184/0x1a0
[ 1156.076558] [<ffffffff8107bc93>] sys32_truncate64+0x13/0x20
[ 1156.076558] [<ffffffff81b04e61>] sysenter_dispatch+0x7/0x1e
[ 1156.076558] Code: 1c b0 00 00 48 98 49 8b 9c c6 80 00 00 00 8b 43 04 39 03 73 79 0f 1f 44 00 00 8b 03 89 c2 83 c0 01 4c 89 64 d3 18 89 03 41 55 9d <5b> 41 5c 41 5d 41 5e 5d c3 0f 1f 00 48 8b 1d b9 d6 f4 00 48 85
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: softlockup in xfstests 083
2012-11-22 19:36 softlockup in xfstests 083 Christoph Hellwig
@ 2012-11-26 13:30 ` Lukáš Czerner
2012-12-08 12:45 ` Christoph Hellwig
0 siblings, 1 reply; 3+ messages in thread
From: Lukáš Czerner @ 2012-11-26 13:30 UTC (permalink / raw)
To: Christoph Hellwig; +Cc: linux-ext4
On Thu, 22 Nov 2012, Christoph Hellwig wrote:
> Date: Thu, 22 Nov 2012 14:36:00 -0500
> From: Christoph Hellwig <hch@infradead.org>
> To: linux-ext4@vger.kernel.org
> Subject: softlockup in xfstests 083
>
> Running xfstests on ext4 using current Linus' tree I can reasily
> reproduce the following softlockup in xfstests 083:
Hi Christoph,
I am unable to reproduce the problem even after 70 iteration of 083
on 3.7.0-rc7. It is not immediately obvious what would block you in
ext4_ext_remove_space() other reading the extent tree from the disk,
but it should not take so long, is the trace always the same ?
Thanks!
-Lukas
>
> 083 9s ...[ 1124.995740] EXT4-fs (vdb2): mounted filesystem with ordered
> data mode. Opts: acl,user_xattr
> [ 1156.073356] BUG: soft lockup - CPU#0 stuck for 22s! [fsstress:912]
> [ 1156.074733] Modules linked in:
> [ 1156.075471] CPU 0
> [ 1156.075887] Pid: 912, comm: fsstress Not tainted 3.7.0-rc6+ #212 Bochs Bochs
> [ 1156.076558] RIP: 0010:[<ffffffff81152b64>] [<ffffffff81152b64>] kfree+0xc4/0x140
> [ 1156.076558] RSP: 0018:ffff880047235c58 EFLAGS: 00000282
> [ 1156.076558] RAX: 0000000000000056 RBX: ffff880047235bd0 RCX: 0000000000000065
> [ 1156.076558] RDX: 0000000000000055 RSI: 0000000000000046 RDI: 0000000000000000
> [ 1156.076558] RBP: ffff880047235c78 R08: ffe2001390ac0000 R09: 002dffe4ef58e42b
> [ 1156.076558] R10: ff8707731e20e802 R11: 000000000000086d R12: ffffffff8167f60c
> [ 1156.076558] R13: ffff880047235c48 R14: 0000000000000046 R15: ffff880057594400
> [ 1156.076558] FS: 0000000000000000(0000) GS:ffff88005d800000(0063) knlGS:00000000f75be6c0
> [ 1156.076558] CS: 0010 DS: 002b ES: 002b CR0: 0000000080050033
> [ 1156.076558] CR2: 00000000f714b004 CR3: 00000000592cc000 CR4: 00000000000006f0
> [ 1156.076558] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> [ 1156.076558] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
> [ 1156.076558] Process fsstress (pid: 912, threadinfo ffff880047234000, task ffff8800582bc0c0)
> [ 1156.076558] Stack:
> [ 1156.076558] 0000000000000001 00000000ffffffff 0000000000000008 ffff88005aed6648
> [ 1156.076558] ffff880047235d78 ffffffff812547fc ffff880053965390 ffff88005aed6648
> [ 1156.076558] ffff880047235d30 ffff88005af8a3a0 ffff88005aed6598 ffff88003a5fe900
> [ 1156.076558] Call Trace:
> [ 1156.076558] [<ffffffff812547fc>] ext4_ext_remove_space+0x49c/0xeb0
> [ 1156.076558] [<ffffffff81257251>] ? ext4_ext_truncate+0x151/0x1b0
> [ 1156.076558] [<ffffffff81257273>] ext4_ext_truncate+0x173/0x1b0
> [ 1156.076558] [<ffffffff81229445>] ext4_truncate+0xd5/0xe0
> [ 1156.076558] [<ffffffff8122c3ba>] ext4_setattr+0x37a/0x570
> [ 1156.076558] [<ffffffff810882c1>] ? current_fs_time+0x11/0x60
> [ 1156.076558] [<ffffffff8116f024>] notify_change+0x164/0x390
> [ 1156.076558] [<ffffffff81164723>] ? user_path_at_empty+0x63/0xa0
> [ 1156.076558] [<ffffffff81154a2d>] do_truncate+0x5d/0xa0
> [ 1156.076558] [<ffffffff81154bf4>] sys_truncate+0x184/0x1a0
> [ 1156.076558] [<ffffffff8107bc93>] sys32_truncate64+0x13/0x20
> [ 1156.076558] [<ffffffff81b04e61>] sysenter_dispatch+0x7/0x1e
> [ 1156.076558] Code: 1c b0 00 00 48 98 49 8b 9c c6 80 00 00 00 8b 43 04 39 03 73 79 0f 1f 44 00 00 8b 03 89 c2 83 c0 01 4c 89 64 d3 18 89 03 41 55 9d <5b> 41 5c 41 5d 41 5e 5d c3 0f 1f 00 48 8b 1d b9 d6 f4 00 48 85
> --
> To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: softlockup in xfstests 083
2012-11-26 13:30 ` Lukáš Czerner
@ 2012-12-08 12:45 ` Christoph Hellwig
0 siblings, 0 replies; 3+ messages in thread
From: Christoph Hellwig @ 2012-12-08 12:45 UTC (permalink / raw)
To: Luk?? Czerner; +Cc: linux-ext4
On Mon, Nov 26, 2012 at 02:30:32PM +0100, Luk?? Czerner wrote:
> I am unable to reproduce the problem even after 70 iteration of 083
> on 3.7.0-rc7. It is not immediately obvious what would block you in
> ext4_ext_remove_space() other reading the extent tree from the disk,
> but it should not take so long, is the trace always the same ?
Hi Lukas,
sorry for the delay.
Been retesting this with the current post-pre8 Linus tree and I can
hit it with a single try, same trace as before.
This is an x86-64 kvm guest, single virtual cpu and 1.5 GB memory on
a virtio disk.
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2012-12-08 12:45 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-11-22 19:36 softlockup in xfstests 083 Christoph Hellwig
2012-11-26 13:30 ` Lukáš Czerner
2012-12-08 12:45 ` Christoph Hellwig
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).