* 2.6.2-mm1 problem with umounting reiserfs
@ 2004-02-06 22:02 Steven Cole
2004-02-06 22:39 ` Andrew Morton
0 siblings, 1 reply; 4+ messages in thread
From: Steven Cole @ 2004-02-06 22:02 UTC (permalink / raw)
To: Andrew Morton; +Cc: linux-mm
With kernel 2.6.2-mm1, I got the following when umounting a reiserfs
file system. I did not get this with 2.6.2-rc2-mm1 or stock kernel.
Base distro is Mandrake 10.0 beta2. System is UP PIII.
Steven
[root@spc1 steven]# df -T
Filesystem Type Size Used Avail Use% Mounted on
/dev/hda1 ext3 236M 79M 145M 36% /
/dev/hda9 ext3 20G 16G 4.2G 79% /home
/dev/hda10
reiserfs 4.0G 1.9G 2.1G 47% /share_r
/dev/hda8 ext3 236M 4.1M 220M 2% /tmp
/dev/hda6 ext3 2.9G 2.1G 737M 74% /usr
/dev/hda7 ext3 479M 198M 257M 44% /var
[root@spc1 steven]# umount /dev/hda10
Segmentation fault
[root@spc1 steven]# dmesg | tail --lines 40
kjournald starting. Commit interval 5 seconds
EXT3 FS on hda6, internal journal
EXT3-fs: mounted filesystem with ordered data mode.
kjournald starting. Commit interval 5 seconds
EXT3 FS on hda7, internal journal
EXT3-fs: mounted filesystem with ordered data mode.
atkbd.c: Unknown key released (translated set 2, code 0x7a on isa0060/serio0).
atkbd.c: This is an XFree86 bug. It shouldn't access hardware directly.
atkbd.c: Unknown key released (translated set 2, code 0x7a on isa0060/serio0).
atkbd.c: This is an XFree86 bug. It shouldn't access hardware directly.
Unable to handle kernel NULL pointer dereference at virtual address 00000000
printing eip:
c012a7b2
*pde = 00000000
Oops: 0000 [#1]
PREEMPT
CPU: 0
EIP: 0060:[<c012a7b2>] Not tainted VLI
EFLAGS: 00210202
EIP is at destroy_workqueue+0x72/0xe0
eax: 00000001 ebx: ca55e000 ecx: cfca3364 edx: 00000000
esi: cfca3360 edi: cfca3320 ebp: cf926670 esp: ca55fe90
ds: 007b es: 007b ss: 0068
Process umount (pid: 1743, threadinfo=ca55e000 task=cd6e5940)
Stack: cf926670 00000001 ca55feb8 cfca1200 c04573c0 ca55ff74 c01acc9d cfca3320
cfca1200 cf84d688 c040e963 00000001 00000001 00005c46 cfca1200 cf446d78
ca55fef0 cfca1200 00000000 cfca1200 c019a655 ca55fef0 cfca1200 cf84d688
Call Trace:
[<c01acc9d>] do_journal_release+0x4d/0xe0
[<c019a655>] reiserfs_put_super+0x25/0x180
[<c0154447>] generic_shutdown_super+0x177/0x1e0
[<c01544cd>] kill_block_super+0x1d/0x50
[<c01545df>] deactivate_super+0x5f/0xc0
[<c016b2cb>] sys_umount+0x4b/0x2f0
[<c0141226>] do_munmap+0x296/0x3c0
[<c016b585>] sys_oldumount+0x15/0x19
[<c03f40d2>] sysenter_past_esp+0x43/0x65
Code: ff 4b 14 8b 43 08 a8 08 75 7d 85 ed 74 08 89 2c 24 e8 63 32 00 00 ff 44 24 04 83 c6 40 8b 44 24 04 85 c0 7e c1 8d 4f 44 8b 51 04 <39> 0a 75 4e 8b 47 44 39 48 04 75 3c 89 50 04 89 02 c7 41 04 00
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"aart@kvack.org"> aart@kvack.org </a>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: 2.6.2-mm1 problem with umounting reiserfs
2004-02-06 22:02 2.6.2-mm1 problem with umounting reiserfs Steven Cole
@ 2004-02-06 22:39 ` Andrew Morton
2004-02-06 23:04 ` Steven Cole
2004-02-07 6:23 ` Rusty Russell
0 siblings, 2 replies; 4+ messages in thread
From: Andrew Morton @ 2004-02-06 22:39 UTC (permalink / raw)
To: Steven Cole; +Cc: linux-mm, Rusty Russell
Steven Cole <elenstev@mesatop.com> wrote:
>
> With kernel 2.6.2-mm1, I got the following when umounting a reiserfs
> file system.
> Unable to handle kernel NULL pointer dereference at virtual address 00000000
> printing eip:
> c012a7b2
> *pde = 00000000
> Oops: 0000 [#1]
> PREEMPT
> CPU: 0
> EIP: 0060:[<c012a7b2>] Not tainted VLI
> EFLAGS: 00210202
> EIP is at destroy_workqueue+0x72/0xe0
> eax: 00000001 ebx: ca55e000 ecx: cfca3364 edx: 00000000
> esi: cfca3360 edi: cfca3320 ebp: cf926670 esp: ca55fe90
> ds: 007b es: 007b ss: 0068
> Process umount (pid: 1743, threadinfo=ca55e000 task=cd6e5940)
> Stack: cf926670 00000001 ca55feb8 cfca1200 c04573c0 ca55ff74 c01acc9d cfca3320
> cfca1200 cf84d688 c040e963 00000001 00000001 00005c46 cfca1200 cf446d78
> ca55fef0 cfca1200 00000000 cfca1200 c019a655 ca55fef0 cfca1200 cf84d688
> Call Trace:
> [<c01acc9d>] do_journal_release+0x4d/0xe0
> [<c019a655>] reiserfs_put_super+0x25/0x180
> [<c0154447>] generic_shutdown_super+0x177/0x1e0
> [<c01544cd>] kill_block_super+0x1d/0x50
> [<c01545df>] deactivate_super+0x5f/0xc0
> [<c016b2cb>] sys_umount+0x4b/0x2f0
> [<c0141226>] do_munmap+0x296/0x3c0
> [<c016b585>] sys_oldumount+0x15/0x19
> [<c03f40d2>] sysenter_past_esp+0x43/0x65
Squish. Thanks.
diff -puN kernel/workqueue.c~cpuhotplug-03-core-workqueue-fix kernel/workqueue.c
--- 25/kernel/workqueue.c~cpuhotplug-03-core-workqueue-fix Fri Feb 6 14:36:04 2004
+++ 25-akpm/kernel/workqueue.c Fri Feb 6 14:36:41 2004
@@ -335,7 +335,7 @@ void destroy_workqueue(struct workqueue_
if (cpu_online(cpu))
cleanup_workqueue_thread(wq, cpu);
}
- list_del(&wq->list);
+ del_workqueue(wq);
unlock_cpu_hotplug();
kfree(wq);
}
_
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"aart@kvack.org"> aart@kvack.org </a>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: 2.6.2-mm1 problem with umounting reiserfs
2004-02-06 22:39 ` Andrew Morton
@ 2004-02-06 23:04 ` Steven Cole
2004-02-07 6:23 ` Rusty Russell
1 sibling, 0 replies; 4+ messages in thread
From: Steven Cole @ 2004-02-06 23:04 UTC (permalink / raw)
To: Andrew Morton; +Cc: linux-mm, Rusty Russell
On Fri, 2004-02-06 at 15:39, Andrew Morton wrote:
> Steven Cole <elenstev@mesatop.com> wrote:
> >
> > With kernel 2.6.2-mm1, I got the following when umounting a reiserfs
> > file system.
>
> Squish. Thanks.
>
>
> diff -puN kernel/workqueue.c~cpuhotplug-03-core-workqueue-fix kernel/workqueue.c
> --- 25/kernel/workqueue.c~cpuhotplug-03-core-workqueue-fix Fri Feb 6 14:36:04 2004
> +++ 25-akpm/kernel/workqueue.c Fri Feb 6 14:36:41 2004
> @@ -335,7 +335,7 @@ void destroy_workqueue(struct workqueue_
> if (cpu_online(cpu))
> cleanup_workqueue_thread(wq, cpu);
> }
> - list_del(&wq->list);
> + del_workqueue(wq);
> unlock_cpu_hotplug();
> kfree(wq);
> }
>
Squish confirmed. That was fast!
Now if someone can squish the xfs "i_size_write() called without i_sem"
that we're not supposed to continue to whine about...
Thanks,
Steven
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"aart@kvack.org"> aart@kvack.org </a>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: 2.6.2-mm1 problem with umounting reiserfs
2004-02-06 22:39 ` Andrew Morton
2004-02-06 23:04 ` Steven Cole
@ 2004-02-07 6:23 ` Rusty Russell
1 sibling, 0 replies; 4+ messages in thread
From: Rusty Russell @ 2004-02-07 6:23 UTC (permalink / raw)
To: Andrew Morton; +Cc: Steven Cole, linux-mm
In message <20040206143917.4e39b215.akpm@osdl.org> you write:
> diff -puN kernel/workqueue.c~cpuhotplug-03-core-workqueue-fix kernel/workqueue.c
> --- 25/kernel/workqueue.c~cpuhotplug-03-core-workqueue-fix Fri Feb 6 14:36:04 2004
> +++ 25-akpm/kernel/workqueue.c Fri Feb 6 14:36:41 2004
> @@ -335,7 +335,7 @@ void destroy_workqueue(struct workqueue_
> if (cpu_online(cpu))
> cleanup_workqueue_thread(wq, cpu);
> }
> - list_del(&wq->list);
> + del_workqueue(wq);
Damn. I added that conditional macro at the last minute, trying to
reduce the impact on non-hotplug-CPU.
Thanks for the fix,
Rusty.
--
Anyone who quotes me in their sig is an idiot. -- Rusty Russell.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"aart@kvack.org"> aart@kvack.org </a>
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2004-02-07 6:23 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-02-06 22:02 2.6.2-mm1 problem with umounting reiserfs Steven Cole
2004-02-06 22:39 ` Andrew Morton
2004-02-06 23:04 ` Steven Cole
2004-02-07 6:23 ` Rusty Russell
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox