2005-05-17 19:07:38

by Berck E. Nash

[permalink] [raw]
Subject: Scheduling while atomic lockups, Reiser4, DAC960

Linux luna 2.6.12-rc4-mm1 i686
Debian unstable, AMD Athlon

The error seems to occur with my reiser4 partition. The physical
device is an 8-disk RAID-0 array on a Mylex Extreme RAID 1100 using the
DAC960 driver. The error is sometimes followed by a lockup, but not always.

Not sure if it's a hardware problem or kernel problem.

There's pages and pages of errors that look like this. If there's
anything I can do to provide more information, I will gladly do so.

Just a user, not a hacker. I'm not subscribed to the list, so cc's on
reply would be appreciated.

Thanks,
Berck

scheduling while atomic: pdflush/0xffffffff/92
May 16 03:36:43 luna kernel: [<c02f6ac4>] schedule+0x604/0x610
May 16 03:36:43 luna kernel: [<c02f6265>] __down+0x85/0x120
May 16 03:36:43 luna kernel: [<c0115020>] default_wake_function+0x0/0x10
May 16 03:36:43 luna kernel: [<c02f645f>] __down_failed+0x7/0xc
May 16 03:36:43 luna kernel: [<c024b420>] blk_backing_dev_unplug+0x0/0x10
May 16 03:36:43 luna kernel: [<c01c5ca3>] .text.lock.flush_queue+0x8/0x25
May 16 03:36:43 luna kernel: [<c01c5524>] finish_fq+0x14/0x40
May 16 03:36:43 luna kernel: [<c01c55ae>] finish_all_fq+0x5e/0xa0
May 16 03:36:43 luna kernel: [<c01c560e>]
current_atom_finish_all_fq+0x1e/0x70
May 16 03:36:43 luna kernel: [<c01c167c>] reiser4_write_logs+0x20c/0x2b0
May 16 03:36:43 luna kernel: [<c01ba599>] commit_current_atom+0x109/0x1e0
May 16 03:36:43 luna kernel: [<c01c5957>] release_prepped_list+0xf7/0x130
May 16 03:36:43 luna kernel: [<c01bae23>] try_commit_txnh+0x113/0x190
May 16 03:36:43 luna kernel: [<c01baec8>] commit_txnh+0x28/0xa0
May 16 03:36:43 luna kernel: [<c01b9d7c>] txn_end+0x2c/0x30
May 16 03:36:43 luna kernel: [<c01bab2b>] flush_some_atom+0x1fb/0x290
May 16 03:36:43 luna kernel: [<c01c9368>] writeout+0x68/0xb0
May 16 03:36:43 luna kernel: [<c01c940b>] reiser4_sync_inodes+0x5b/0xa0
May 16 03:36:43 luna kernel: [<c01c93b0>] reiser4_sync_inodes+0x0/0xa0
May 16 03:36:43 luna kernel: [<c0174a59>] sync_sb_inodes+0x19/0x20
May 16 03:36:43 luna kernel: [<c0174b1e>] writeback_inodes+0xbe/0xd0
May 16 03:36:43 luna kernel: [<c013be15>] background_writeout+0x65/0xa0
May 16 03:36:43 luna kernel: [<c013c748>] __pdflush+0xc8/0x1c0
May 16 03:36:43 luna kernel: [<c013c840>] pdflush+0x0/0x20
May 16 03:36:43 luna kernel: [<c013c85a>] pdflush+0x1a/0x20
May 16 03:36:43 luna kernel: [<c013bdb0>] background_writeout+0x0/0xa0
May 16 03:36:43 luna kernel: [<c013c840>] pdflush+0x0/0x20
May 16 03:36:43 luna kernel: [<c012bb44>] kthread+0x94/0xa0
May 16 03:36:43 luna kernel: [<c012bab0>] kthread+0x0/0xa0
May 16 03:36:43 luna kernel: [<c0100f1d>] kernel_thread_helper+0x5/0x18


2005-05-17 22:04:48

by Hans Reiser

[permalink] [raw]
Subject: Re: Scheduling while atomic lockups, Reiser4, DAC960

vs will look at it.

Hans

Berck E. Nash wrote:

>Linux luna 2.6.12-rc4-mm1 i686
>Debian unstable, AMD Athlon
>
>The error seems to occur with my reiser4 partition. The physical
>device is an 8-disk RAID-0 array on a Mylex Extreme RAID 1100 using the
>DAC960 driver. The error is sometimes followed by a lockup, but not always.
>
>Not sure if it's a hardware problem or kernel problem.
>
>There's pages and pages of errors that look like this. If there's
>anything I can do to provide more information, I will gladly do so.
>
>Just a user, not a hacker. I'm not subscribed to the list, so cc's on
>reply would be appreciated.
>
>Thanks,
>Berck
>
> scheduling while atomic: pdflush/0xffffffff/92
>May 16 03:36:43 luna kernel: [<c02f6ac4>] schedule+0x604/0x610
>May 16 03:36:43 luna kernel: [<c02f6265>] __down+0x85/0x120
>May 16 03:36:43 luna kernel: [<c0115020>] default_wake_function+0x0/0x10
>May 16 03:36:43 luna kernel: [<c02f645f>] __down_failed+0x7/0xc
>May 16 03:36:43 luna kernel: [<c024b420>] blk_backing_dev_unplug+0x0/0x10
>May 16 03:36:43 luna kernel: [<c01c5ca3>] .text.lock.flush_queue+0x8/0x25
>May 16 03:36:43 luna kernel: [<c01c5524>] finish_fq+0x14/0x40
>May 16 03:36:43 luna kernel: [<c01c55ae>] finish_all_fq+0x5e/0xa0
>May 16 03:36:43 luna kernel: [<c01c560e>]
>current_atom_finish_all_fq+0x1e/0x70
>May 16 03:36:43 luna kernel: [<c01c167c>] reiser4_write_logs+0x20c/0x2b0
>May 16 03:36:43 luna kernel: [<c01ba599>] commit_current_atom+0x109/0x1e0
>May 16 03:36:43 luna kernel: [<c01c5957>] release_prepped_list+0xf7/0x130
>May 16 03:36:43 luna kernel: [<c01bae23>] try_commit_txnh+0x113/0x190
>May 16 03:36:43 luna kernel: [<c01baec8>] commit_txnh+0x28/0xa0
>May 16 03:36:43 luna kernel: [<c01b9d7c>] txn_end+0x2c/0x30
>May 16 03:36:43 luna kernel: [<c01bab2b>] flush_some_atom+0x1fb/0x290
>May 16 03:36:43 luna kernel: [<c01c9368>] writeout+0x68/0xb0
>May 16 03:36:43 luna kernel: [<c01c940b>] reiser4_sync_inodes+0x5b/0xa0
>May 16 03:36:43 luna kernel: [<c01c93b0>] reiser4_sync_inodes+0x0/0xa0
>May 16 03:36:43 luna kernel: [<c0174a59>] sync_sb_inodes+0x19/0x20
>May 16 03:36:43 luna kernel: [<c0174b1e>] writeback_inodes+0xbe/0xd0
>May 16 03:36:43 luna kernel: [<c013be15>] background_writeout+0x65/0xa0
>May 16 03:36:43 luna kernel: [<c013c748>] __pdflush+0xc8/0x1c0
>May 16 03:36:43 luna kernel: [<c013c840>] pdflush+0x0/0x20
>May 16 03:36:43 luna kernel: [<c013c85a>] pdflush+0x1a/0x20
>May 16 03:36:43 luna kernel: [<c013bdb0>] background_writeout+0x0/0xa0
>May 16 03:36:43 luna kernel: [<c013c840>] pdflush+0x0/0x20
>May 16 03:36:43 luna kernel: [<c012bb44>] kthread+0x94/0xa0
>May 16 03:36:43 luna kernel: [<c012bab0>] kthread+0x0/0xa0
>May 16 03:36:43 luna kernel: [<c0100f1d>] kernel_thread_helper+0x5/0x18
>-
>To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>the body of a message to [email protected]
>More majordomo info at http://vger.kernel.org/majordomo-info.html
>Please read the FAQ at http://www.tux.org/lkml/
>
>
>
>

2005-05-19 06:20:02

by Vladimir V. Saveliev

[permalink] [raw]
Subject: Re: Scheduling while atomic lockups, Reiser4, DAC960

Hello

On Wed, 2005-05-18 at 01:57, Hans Reiser wrote:
> vs will look at it.
>
> Hans
>
> Berck E. Nash wrote:
>
> >Linux luna 2.6.12-rc4-mm1 i686

Please try to patch -R the patch
ftp://ftp.ru.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.12-rc4/2.6.12-rc4-mm1/broken-out/reiser4-sb_sync_inodes-cleanup.patch
or use 2.6.12-rc4-mm2

> >Debian unstable, AMD Athlon
> >
> >The error seems to occur with my reiser4 partition. The physical
> >device is an 8-disk RAID-0 array on a Mylex Extreme RAID 1100 using the
> >DAC960 driver. The error is sometimes followed by a lockup, but not always.
> >
> >Not sure if it's a hardware problem or kernel problem.
> >
> >There's pages and pages of errors that look like this. If there's
> >anything I can do to provide more information, I will gladly do so.
> >
> >Just a user, not a hacker. I'm not subscribed to the list, so cc's on
> >reply would be appreciated.
> >
> >Thanks,
> >Berck
> >
> > scheduling while atomic: pdflush/0xffffffff/92
> >May 16 03:36:43 luna kernel: [<c02f6ac4>] schedule+0x604/0x610
> >May 16 03:36:43 luna kernel: [<c02f6265>] __down+0x85/0x120
> >May 16 03:36:43 luna kernel: [<c0115020>] default_wake_function+0x0/0x10
> >May 16 03:36:43 luna kernel: [<c02f645f>] __down_failed+0x7/0xc
> >May 16 03:36:43 luna kernel: [<c024b420>] blk_backing_dev_unplug+0x0/0x10
> >May 16 03:36:43 luna kernel: [<c01c5ca3>] .text.lock.flush_queue+0x8/0x25
> >May 16 03:36:43 luna kernel: [<c01c5524>] finish_fq+0x14/0x40
> >May 16 03:36:43 luna kernel: [<c01c55ae>] finish_all_fq+0x5e/0xa0
> >May 16 03:36:43 luna kernel: [<c01c560e>]
> >current_atom_finish_all_fq+0x1e/0x70
> >May 16 03:36:43 luna kernel: [<c01c167c>] reiser4_write_logs+0x20c/0x2b0
> >May 16 03:36:43 luna kernel: [<c01ba599>] commit_current_atom+0x109/0x1e0
> >May 16 03:36:43 luna kernel: [<c01c5957>] release_prepped_list+0xf7/0x130
> >May 16 03:36:43 luna kernel: [<c01bae23>] try_commit_txnh+0x113/0x190
> >May 16 03:36:43 luna kernel: [<c01baec8>] commit_txnh+0x28/0xa0
> >May 16 03:36:43 luna kernel: [<c01b9d7c>] txn_end+0x2c/0x30
> >May 16 03:36:43 luna kernel: [<c01bab2b>] flush_some_atom+0x1fb/0x290
> >May 16 03:36:43 luna kernel: [<c01c9368>] writeout+0x68/0xb0
> >May 16 03:36:43 luna kernel: [<c01c940b>] reiser4_sync_inodes+0x5b/0xa0
> >May 16 03:36:43 luna kernel: [<c01c93b0>] reiser4_sync_inodes+0x0/0xa0
> >May 16 03:36:43 luna kernel: [<c0174a59>] sync_sb_inodes+0x19/0x20
> >May 16 03:36:43 luna kernel: [<c0174b1e>] writeback_inodes+0xbe/0xd0
> >May 16 03:36:43 luna kernel: [<c013be15>] background_writeout+0x65/0xa0
> >May 16 03:36:43 luna kernel: [<c013c748>] __pdflush+0xc8/0x1c0
> >May 16 03:36:43 luna kernel: [<c013c840>] pdflush+0x0/0x20
> >May 16 03:36:43 luna kernel: [<c013c85a>] pdflush+0x1a/0x20
> >May 16 03:36:43 luna kernel: [<c013bdb0>] background_writeout+0x0/0xa0
> >May 16 03:36:43 luna kernel: [<c013c840>] pdflush+0x0/0x20
> >May 16 03:36:43 luna kernel: [<c012bb44>] kthread+0x94/0xa0
> >May 16 03:36:43 luna kernel: [<c012bab0>] kthread+0x0/0xa0
> >May 16 03:36:43 luna kernel: [<c0100f1d>] kernel_thread_helper+0x5/0x18
> >-
> >To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> >the body of a message to [email protected]
> >More majordomo info at http://vger.kernel.org/majordomo-info.html
> >Please read the FAQ at http://www.tux.org/lkml/
> >
> >
> >
> >
>
>

2005-05-19 14:11:03

by Berck E. Nash

[permalink] [raw]
Subject: Re: Scheduling while atomic lockups, Reiser4, DAC960

Vladimir Saveliev wrote:

> Please try to patch -R the patch
> ftp://ftp.ru.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.12-rc4/2.6.12-rc4-mm1/broken-out/reiser4-sb_sync_inodes-cleanup.patch
> or use 2.6.12-rc4-mm2

Been using rc-mm2 since yesterday and it all seems good. Thank you!

>
>>>Debian unstable, AMD Athlon
>>>
>>>The error seems to occur with my reiser4 partition. The physical
>>>device is an 8-disk RAID-0 array on a Mylex Extreme RAID 1100 using the
>>>DAC960 driver. The error is sometimes followed by a lockup, but not always.
>>>
>>>Not sure if it's a hardware problem or kernel problem.
>>>
>>>There's pages and pages of errors that look like this. If there's
>>>anything I can do to provide more information, I will gladly do so.
>>>
>>>Just a user, not a hacker. I'm not subscribed to the list, so cc's on
>>>reply would be appreciated.
>>>
>>>Thanks,
>>>Berck
>>>
>>>scheduling while atomic: pdflush/0xffffffff/92
>>>May 16 03:36:43 luna kernel: [<c02f6ac4>] schedule+0x604/0x610
>>>May 16 03:36:43 luna kernel: [<c02f6265>] __down+0x85/0x120
>>>May 16 03:36:43 luna kernel: [<c0115020>] default_wake_function+0x0/0x10
>>>May 16 03:36:43 luna kernel: [<c02f645f>] __down_failed+0x7/0xc
>>>May 16 03:36:43 luna kernel: [<c024b420>] blk_backing_dev_unplug+0x0/0x10
>>>May 16 03:36:43 luna kernel: [<c01c5ca3>] .text.lock.flush_queue+0x8/0x25
>>>May 16 03:36:43 luna kernel: [<c01c5524>] finish_fq+0x14/0x40
>>>May 16 03:36:43 luna kernel: [<c01c55ae>] finish_all_fq+0x5e/0xa0
>>>May 16 03:36:43 luna kernel: [<c01c560e>]
>>>current_atom_finish_all_fq+0x1e/0x70
>>>May 16 03:36:43 luna kernel: [<c01c167c>] reiser4_write_logs+0x20c/0x2b0
>>>May 16 03:36:43 luna kernel: [<c01ba599>] commit_current_atom+0x109/0x1e0
>>>May 16 03:36:43 luna kernel: [<c01c5957>] release_prepped_list+0xf7/0x130
>>>May 16 03:36:43 luna kernel: [<c01bae23>] try_commit_txnh+0x113/0x190
>>>May 16 03:36:43 luna kernel: [<c01baec8>] commit_txnh+0x28/0xa0
>>>May 16 03:36:43 luna kernel: [<c01b9d7c>] txn_end+0x2c/0x30
>>>May 16 03:36:43 luna kernel: [<c01bab2b>] flush_some_atom+0x1fb/0x290
>>>May 16 03:36:43 luna kernel: [<c01c9368>] writeout+0x68/0xb0
>>>May 16 03:36:43 luna kernel: [<c01c940b>] reiser4_sync_inodes+0x5b/0xa0
>>>May 16 03:36:43 luna kernel: [<c01c93b0>] reiser4_sync_inodes+0x0/0xa0
>>>May 16 03:36:43 luna kernel: [<c0174a59>] sync_sb_inodes+0x19/0x20
>>>May 16 03:36:43 luna kernel: [<c0174b1e>] writeback_inodes+0xbe/0xd0
>>>May 16 03:36:43 luna kernel: [<c013be15>] background_writeout+0x65/0xa0
>>>May 16 03:36:43 luna kernel: [<c013c748>] __pdflush+0xc8/0x1c0
>>>May 16 03:36:43 luna kernel: [<c013c840>] pdflush+0x0/0x20
>>>May 16 03:36:43 luna kernel: [<c013c85a>] pdflush+0x1a/0x20
>>>May 16 03:36:43 luna kernel: [<c013bdb0>] background_writeout+0x0/0xa0
>>>May 16 03:36:43 luna kernel: [<c013c840>] pdflush+0x0/0x20
>>>May 16 03:36:43 luna kernel: [<c012bb44>] kthread+0x94/0xa0
>>>May 16 03:36:43 luna kernel: [<c012bab0>] kthread+0x0/0xa0
>>>May 16 03:36:43 luna kernel: [<c0100f1d>] kernel_thread_helper+0x5/0x18
>>>-
>>>To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>>>the body of a message to [email protected]
>>>More majordomo info at http://vger.kernel.org/majordomo-info.html
>>>Please read the FAQ at http://www.tux.org/lkml/
>>>
>>>
>>>
>>>
>>
>>
>