* panics with 16port Promise Supertrack EX Controller
@ 2007-07-08 22:52 Flavio Curti
2007-07-08 23:20 ` Michal Piotrowski
0 siblings, 1 reply; 10+ messages in thread
From: Flavio Curti @ 2007-07-08 22:52 UTC (permalink / raw)
To: linux-kernel
[-- Attachment #1: Type: text/plain, Size: 1724 bytes --]
Hello
I have a problem with a server running 2.6.22rc4. The machine panics
after some days of running fine, the machine inst heavy loaded.
The Controller detects as stex device:
scsi0 : stex
scsi 0:0:0:0: Direct-Access Promise 1X2 Mirror 1.10 PQ: 0 ANSI: 3
scsi 0:0:2:0: Direct-Access Promise 12+2 Disk RAID6 1.10 PQ: 0 ANSI: 3
scsi 0:0:16:0: Processor Promise RAID Console 1.00 PQ: 0 ANSI: 3
sd 0:0:0:0: [sda] 976642048 512-byte hardware sectors (500041 MB)
sd 0:0:0:0: [sda] Write Protect is off
sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
sd 0:0:0:0: [sda] 976642048 512-byte hardware sectors (500041 MB)
sd 0:0:0:0: [sda] Write Protect is off
sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
sda: sda1 sda2
sd 0:0:0:0: [sda] Attached SCSI disk
sd 0:0:2:0: [sdb] Very big device. Trying to use READ CAPACITY(16).
sd 0:0:2:0: [sdb] 11719704576 512-byte hardware sectors (6000489 MB)
sd 0:0:2:0: [sdb] Write Protect is off
sd 0:0:2:0: [sdb] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
sd 0:0:2:0: [sdb] Very big device. Trying to use READ CAPACITY(16).
sd 0:0:2:0: [sdb] 11719704576 512-byte hardware sectors (6000489 MB)
sd 0:0:2:0: [sdb] Write Protect is off
sd 0:0:2:0: [sdb] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
sdb: sdb1
sd 0:0:2:0: [sdb] Attached SCSI disk
sd 0:0:0:0: Attached scsi generic sg0 type 0
sd 0:0:2:0: Attached scsi generic sg1 type 0
scsi 0:0:16:0: Attached scsi generic sg2 type 3
Im not sure where the problem is (controller/lvm/ext3), so if anyone has
an idea, Im happy to try it out...
Thank you
Flavio Curti
--
http://no-way.org/~fcu/
[-- Attachment #2: stex-panic.txt --]
[-- Type: text/plain, Size: 5325 bytes --]
Jul 8 00:19:13 dorade.cyberlink.ch ------------[ cut here ]------------
Jul 8 00:19:13 dorade.cyberlink.ch kernel BUG at block/as-iosched.c:1084!
Jul 8 00:19:13 dorade.cyberlink.ch invalid opcode: 0000 [#1]
Jul 8 00:19:13 dorade.cyberlink.ch SMP
Jul 8 00:19:13 dorade.cyberlink.ch
Jul 8 00:19:13 dorade.cyberlink.ch Modules linked in:
Jul 8 00:19:13 dorade.cyberlink.ch i2c_i801
Jul 8 00:19:13 dorade.cyberlink.ch i2c_core
Jul 8 00:19:13 dorade.cyberlink.ch
Jul 8 00:19:13 dorade.cyberlink.ch CPU: 2
Jul 8 00:19:13 dorade.cyberlink.ch EIP: 0060:[<c02a9f17>] Not tainted VLI
Jul 8 00:19:13 dorade.cyberlink.ch EFLAGS: 00210046 (2.6.22-rc7-dorade #1)
Jul 8 00:19:13 dorade.cyberlink.ch EIP is at as_dispatch_request+0x387/0x390
Jul 8 00:19:13 dorade.cyberlink.ch eax: 00000000 ebx: c5ee10c0 ecx: c5ee10d4 edx: 00000000
Jul 8 00:19:13 dorade.cyberlink.ch esi: 00000000 edi: 00000001 ebp: 00000000 esp: cd90bbd4
Jul 8 00:19:13 dorade.cyberlink.ch ds: 007b es: 007b fs: 00d8 gs: 0033 ss: 0068
Jul 8 00:19:13 dorade.cyberlink.ch Process vsftpd (pid: 5332, ti=cd90a000 task=f780d030 task.ti=cd90a000)
Jul 8 00:19:13 dorade.cyberlink.ch
Jul 8 00:19:13 dorade.cyberlink.ch Stack:
Jul 8 00:19:13 dorade.cyberlink.ch c012abe7
Jul 8 00:19:13 dorade.cyberlink.ch f784c5cc
Jul 8 00:19:13 dorade.cyberlink.ch c5f11000
Jul 8 00:19:13 dorade.cyberlink.ch f784c5cc
Jul 8 00:19:13 dorade.cyberlink.ch 00200286
Jul 8 00:19:13 dorade.cyberlink.ch c5ee8be4
Jul 8 00:19:13 dorade.cyberlink.ch c5f11000
Jul 8 00:19:13 dorade.cyberlink.ch f7a16000
Jul 8 00:19:13 dorade.cyberlink.ch
Jul 8 00:19:13 dorade.cyberlink.ch
Jul 8 00:19:13 dorade.cyberlink.ch c5ee8be4
Jul 8 00:19:13 dorade.cyberlink.ch c02a0810
Jul 8 00:19:13 dorade.cyberlink.ch f7a16000
Jul 8 00:19:13 dorade.cyberlink.ch 00200287
Jul 8 00:19:13 dorade.cyberlink.ch c0370076
Jul 8 00:19:13 dorade.cyberlink.ch f784c580
Jul 8 00:19:13 dorade.cyberlink.ch 00000000
Jul 8 00:19:13 dorade.cyberlink.ch c5f11000
Jul 8 00:19:13 dorade.cyberlink.ch c5ee8be4
Jul 8 00:19:13 dorade.cyberlink.ch
Jul 8 00:19:13 dorade.cyberlink.ch Call Trace:
Jul 8 00:19:13 dorade.cyberlink.ch [<c012abe7>]
Jul 8 00:19:13 dorade.cyberlink.ch lock_timer_base+0x27/0x60
Jul 8 00:19:13 dorade.cyberlink.ch [<c02a0810>]
Jul 8 00:19:13 dorade.cyberlink.ch elv_next_request+0x20/0x130
Jul 8 00:19:13 dorade.cyberlink.ch [<c0370076>]
Jul 8 00:19:13 dorade.cyberlink.ch scsi_dispatch_cmd+0x146/0x230
Jul 8 00:19:13 dorade.cyberlink.ch [<c03752b5>]
Jul 8 00:19:13 dorade.cyberlink.ch scsi_request_fn+0x185/0x2c0
Jul 8 00:19:13 dorade.cyberlink.ch [<c02a3e45>]
Jul 8 00:19:13 dorade.cyberlink.ch __generic_unplug_device+0x25/0x30
Jul 8 00:19:13 dorade.cyberlink.ch [<c02a4e55>]
Jul 8 00:19:13 dorade.cyberlink.ch generic_unplug_device+0x15/0x30
Jul 8 00:19:13 dorade.cyberlink.ch [<c03dc582>]
Jul 8 00:19:13 dorade.cyberlink.ch dm_table_unplug_all+0x22/0x30
Jul 8 00:19:13 dorade.cyberlink.ch [<c03da437>]
Jul 8 00:19:13 dorade.cyberlink.ch dm_unplug_all+0x17/0x30
Jul 8 00:19:13 dorade.cyberlink.ch [<c02a163c>]
Jul 8 00:19:13 dorade.cyberlink.ch blk_backing_dev_unplug+0xc/0x10
Jul 8 00:19:13 dorade.cyberlink.ch [<c0192dab>]
Jul 8 00:19:13 dorade.cyberlink.ch sync_buffer+0x2b/0x40
Jul 8 00:19:13 dorade.cyberlink.ch [<c04d99c2>]
Jul 8 00:19:13 dorade.cyberlink.ch __wait_on_bit+0x42/0x70
Jul 8 00:19:13 dorade.cyberlink.ch [<c0192d80>]
Jul 8 00:19:13 dorade.cyberlink.ch sync_buffer+0x0/0x40
Jul 8 00:19:13 dorade.cyberlink.ch __wait_on_buffer+0x24/0x30
Jul 8 00:19:13 dorade.cyberlink.ch sync_dirty_buffer+0x7e/0xd0
Jul 8 00:19:13 dorade.cyberlink.ch [<c01f77e1>]
Jul 8 00:19:13 dorade.cyberlink.ch start_this_handle+0x81/0x3c0
Jul 8 00:19:13 dorade.cyberlink.ch [<c01f86e5>]
Jul 8 00:19:13 dorade.cyberlink.ch journal_dirty_data+0x1c5/0x210
Jul 8 00:19:13 dorade.cyberlink.ch [<c01e70e8>]
Jul 8 00:19:13 dorade.cyberlink.ch ext3_journal_dirty_data+0x18/0x50
Jul 8 00:19:13 dorade.cyberlink.ch [<c01f7bd7>]
Jul 8 00:19:13 dorade.cyberlink.ch journal_start+0xb7/0xe0
Jul 8 00:19:13 dorade.cyberlink.ch [<c01e643e>]
Jul 8 00:19:13 dorade.cyberlink.ch walk_page_buffers+0x3e/0x80
Jul 8 00:19:13 dorade.cyberlink.ch [<c01e8f91>]
Jul 8 00:19:13 dorade.cyberlink.ch ext3_ordered_commit_write+0x51/0xf0
Jul 8 00:19:13 dorade.cyberlink.ch [<c01e70d0>]
Jul 8 00:19:13 dorade.cyberlink.ch ext3_journal_dirty_data+0x0/0x50
Jul 8 00:19:13 dorade.cyberlink.ch [<c0153741>]
Jul 8 00:19:13 dorade.cyberlink.ch generic_file_buffered_write+0x281/0x680
Jul 8 00:19:13 dorade.cyberlink.ch 01
Jul 8 00:19:13 dorade.cyberlink.ch c7
Jul 8 00:19:13 dorade.cyberlink.ch 00
Jul 8 00:19:13 dorade.cyberlink.ch e9
Jul 8 00:19:13 dorade.cyberlink.ch ff
Jul 8 00:19:13 dorade.cyberlink.ch fe
Jul 8 00:19:13 dorade.cyberlink.ch 74
Jul 8 00:19:13 dorade.cyberlink.ch 26
Jul 8 00:19:13 dorade.cyberlink.ch 40
Jul 8 00:19:13 dorade.cyberlink.ch 50
Jul 8 00:19:13 dorade.cyberlink.ch 83
Jul 8 00:19:13 dorade.cyberlink.ch
Jul 8 00:19:13 dorade.cyberlink.ch EIP: [<c02a9f17>]
Jul 8 00:19:13 dorade.cyberlink.ch as_dispatch_request+0x387/0x390
Jul 8 00:19:13 dorade.cyberlink.ch SS:ESP 0068:cd90bbd4
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: panics with 16port Promise Supertrack EX Controller
2007-07-08 22:52 panics with 16port Promise Supertrack EX Controller Flavio Curti
@ 2007-07-08 23:20 ` Michal Piotrowski
2007-07-09 1:59 ` Nick Piggin
2007-07-09 10:45 ` Flavio Curti
0 siblings, 2 replies; 10+ messages in thread
From: Michal Piotrowski @ 2007-07-08 23:20 UTC (permalink / raw)
To: Flavio Curti; +Cc: linux-kernel, Jens Axboe, Nick Piggin
Hi Flavio,
On 09/07/07, Flavio Curti <lists.fcu@no-way.org> wrote:
> Hello
>
> I have a problem with a server running 2.6.22rc4.
Jul 8 00:19:13 dorade.cyberlink.ch EFLAGS: 00210046 (2.6.22-rc7-dorade #1)
Is this a regression?
> The machine panics
> after some days of running fine, the machine inst heavy loaded.
>
> The Controller detects as stex device:
>
> scsi0 : stex
> scsi 0:0:0:0: Direct-Access Promise 1X2 Mirror 1.10 PQ: 0 ANSI: 3
> scsi 0:0:2:0: Direct-Access Promise 12+2 Disk RAID6 1.10 PQ: 0 ANSI: 3
> scsi 0:0:16:0: Processor Promise RAID Console 1.00 PQ: 0 ANSI: 3
> sd 0:0:0:0: [sda] 976642048 512-byte hardware sectors (500041 MB)
> sd 0:0:0:0: [sda] Write Protect is off
> sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
> sd 0:0:0:0: [sda] 976642048 512-byte hardware sectors (500041 MB)
> sd 0:0:0:0: [sda] Write Protect is off
> sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
> sda: sda1 sda2
> sd 0:0:0:0: [sda] Attached SCSI disk
> sd 0:0:2:0: [sdb] Very big device. Trying to use READ CAPACITY(16).
> sd 0:0:2:0: [sdb] 11719704576 512-byte hardware sectors (6000489 MB)
> sd 0:0:2:0: [sdb] Write Protect is off
> sd 0:0:2:0: [sdb] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
> sd 0:0:2:0: [sdb] Very big device. Trying to use READ CAPACITY(16).
> sd 0:0:2:0: [sdb] 11719704576 512-byte hardware sectors (6000489 MB)
> sd 0:0:2:0: [sdb] Write Protect is off
> sd 0:0:2:0: [sdb] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
> sdb: sdb1
> sd 0:0:2:0: [sdb] Attached SCSI disk
> sd 0:0:0:0: Attached scsi generic sg0 type 0
> sd 0:0:2:0: Attached scsi generic sg1 type 0
> scsi 0:0:16:0: Attached scsi generic sg2 type 3
>
> Im not sure where the problem is (controller/lvm/ext3), so if anyone has
> an idea, Im happy to try it out...
kernel BUG at block/as-iosched.c:1084!
BUG_ON(RB_EMPTY_ROOT(&ad->sort_list[REQ_ASYNC]));
>
> Thank you
>
> Flavio Curti
>
> --
> http://no-way.org/~fcu/
>
>
Regards,
Michal
--
LOG
http://www.stardust.webpages.pl/log/
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: panics with 16port Promise Supertrack EX Controller
2007-07-08 23:20 ` Michal Piotrowski
@ 2007-07-09 1:59 ` Nick Piggin
2007-07-15 8:02 ` Flavio Curti
2007-07-15 22:51 ` Flavio Curti
2007-07-09 10:45 ` Flavio Curti
1 sibling, 2 replies; 10+ messages in thread
From: Nick Piggin @ 2007-07-09 1:59 UTC (permalink / raw)
To: Michal Piotrowski; +Cc: Flavio Curti, linux-kernel, Jens Axboe
Michal Piotrowski wrote:
> Hi Flavio,
>
> On 09/07/07, Flavio Curti <lists.fcu@no-way.org> wrote:
>
>> Hello
>>
>> I have a problem with a server running 2.6.22rc4.
>
>
> Jul 8 00:19:13 dorade.cyberlink.ch EFLAGS: 00210046
> (2.6.22-rc7-dorade #1)
>
> Is this a regression?
>
>> The machine panics
>> after some days of running fine, the machine inst heavy loaded.
>>
>> The Controller detects as stex device:
>>
>> scsi0 : stex
>> scsi 0:0:0:0: Direct-Access Promise 1X2 Mirror 1.10 PQ: 0 ANSI: 3
>> scsi 0:0:2:0: Direct-Access Promise 12+2 Disk RAID6 1.10 PQ: 0 ANSI: 3
>> scsi 0:0:16:0: Processor Promise RAID Console 1.00 PQ: 0 ANSI: 3
>> sd 0:0:0:0: [sda] 976642048 512-byte hardware sectors (500041 MB)
>> sd 0:0:0:0: [sda] Write Protect is off
>> sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't
>> support DPO or FUA
>> sd 0:0:0:0: [sda] 976642048 512-byte hardware sectors (500041 MB)
>> sd 0:0:0:0: [sda] Write Protect is off
>> sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't
>> support DPO or FUA
>> sda: sda1 sda2
>> sd 0:0:0:0: [sda] Attached SCSI disk
>> sd 0:0:2:0: [sdb] Very big device. Trying to use READ CAPACITY(16).
>> sd 0:0:2:0: [sdb] 11719704576 512-byte hardware sectors (6000489 MB)
>> sd 0:0:2:0: [sdb] Write Protect is off
>> sd 0:0:2:0: [sdb] Write cache: enabled, read cache: enabled, doesn't
>> support DPO or FUA
>> sd 0:0:2:0: [sdb] Very big device. Trying to use READ CAPACITY(16).
>> sd 0:0:2:0: [sdb] 11719704576 512-byte hardware sectors (6000489 MB)
>> sd 0:0:2:0: [sdb] Write Protect is off
>> sd 0:0:2:0: [sdb] Write cache: enabled, read cache: enabled, doesn't
>> support DPO or FUA
>> sdb: sdb1
>> sd 0:0:2:0: [sdb] Attached SCSI disk
>> sd 0:0:0:0: Attached scsi generic sg0 type 0
>> sd 0:0:2:0: Attached scsi generic sg1 type 0
>> scsi 0:0:16:0: Attached scsi generic sg2 type 3
>>
>> Im not sure where the problem is (controller/lvm/ext3), so if anyone has
>> an idea, Im happy to try it out...
>
>
> kernel BUG at block/as-iosched.c:1084!
>
> BUG_ON(RB_EMPTY_ROOT(&ad->sort_list[REQ_ASYNC]));
Could be a bug in the driver that just happens to be caught by AS checks.
If you could test another scheduler (boot with elevator=deadline or elevator=cfq),
it might help give us an idea.
--
SUSE Labs, Novell Inc.
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: panics with 16port Promise Supertrack EX Controller
2007-07-09 1:59 ` Nick Piggin
@ 2007-07-15 8:02 ` Flavio Curti
2007-07-16 11:17 ` Jens Axboe
2007-07-15 22:51 ` Flavio Curti
1 sibling, 1 reply; 10+ messages in thread
From: Flavio Curti @ 2007-07-15 8:02 UTC (permalink / raw)
To: Nick Piggin; +Cc: Michal Piotrowski, Flavio Curti, linux-kernel, Jens Axboe
[-- Attachment #1: Type: text/plain, Size: 723 bytes --]
Hello
On Mon, Jul 09, 2007 at 11:59:36AM +1000, Nick Piggin wrote:
> >Jul 8 00:19:13 dorade.cyberlink.ch EFLAGS: 00210046
> >(2.6.22-rc7-dorade #1)
> >>The machine panics
> >>after some days of running fine, the machine inst heavy loaded.
> >>The Controller detects as stex device:
> >kernel BUG at block/as-iosched.c:1084!
> >
> >BUG_ON(RB_EMPTY_ROOT(&ad->sort_list[REQ_ASYNC]));
> Could be a bug in the driver that just happens to be caught by AS checks.
> If you could test another scheduler (boot with elevator=deadline or
> elevator=cfq),
> it might help give us an idea.
Ok, I now switched to cfg, and the machine panicd again. Panic attached.
Any help is appreciated.
Flavio Curti
--
http://no-way.org/~fcu/
[-- Attachment #2: dorade-kdump --]
[-- Type: text/plain, Size: 1485 bytes --]
blk_queue_end_tag: attempt to clear non-busy tag (1)
------------[ cut here ]------------
kernel BUG at block/ll_rw_blk.c:2758!
invalid opcode: 0000 [#1]
SMP
Modules linked in:
i2c_i801
i2c_core
CPU: 0
EIP: 0060:[<c02a3164>] Not tainted VLI
EFLAGS: 00010002 (2.6.22.1-dorade #1)
EIP is at __blk_put_request+0x74/0x80
eax: 00000000 ebx: dae3e578 ecx: c02ac600 edx: 00000000
esi: 00030784 edi: f7f28b24 ebp: f7f28b24 esp: c0621e50
ds: 007b es: 007b fs: 00d8 gs: 0000 ss: 0068
Process swapper (pid: 0, ti=c0620000 task=c05d8340 task.ti=c0620000)
Stack:
dae3e578
e2b5aac0
00000001
scsi_end_request+0x94/0xe0
[<c03743c6>]
scsi_io_completion+0x86/0x3e0
[<c012b08a>]
del_timer+0x5a/0x70
[<c038fc7b>]
sd_rw_intr+0x2b/0x200
[<c038bf8e>]
stex_mu_intr+0x12e/0x540
[<c036fe85>]
scsi_finish_command+0x45/0x60
[<c02a4c88>]
blk_done_softirq+0x58/0x70
[<c0126e92>]
__do_softirq+0x82/0x100
[<c0126f47>]
do_softirq+0x37/0x40
[<c0127185>]
irq_exit+0x75/0x80
[<c0106410>]
do_IRQ+0x40/0x80
[<c0139796>]
getnstimeofday+0x36/0xd0
[<c0104503>]
common_interrupt+0x23/0x28
[<c0101d26>]
mwait_idle_with_hints+0x46/0x60
[<c0101e85>]
cpu_idle+0x65/0x90
[<c0628d1f>]
start_kernel+0x2ef/0x370
[<c06284e0>]
unknown_bootoption+0x0/0x250
=======================
Code:
8b
57
24
89
d8
83
e6
01
e8
ca
1b
eb
ff
89
f2
89
f8
89
e9
5b
5e
5f
5d
e9
3b
ff
last message repeated 2 times
89
da
89
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: panics with 16port Promise Supertrack EX Controller
2007-07-15 8:02 ` Flavio Curti
@ 2007-07-16 11:17 ` Jens Axboe
0 siblings, 0 replies; 10+ messages in thread
From: Jens Axboe @ 2007-07-16 11:17 UTC (permalink / raw)
To: Flavio Curti
Cc: Nick Piggin, Michal Piotrowski, Flavio Curti, linux-kernel,
promise_linux
On Sun, Jul 15 2007, Flavio Curti wrote:
> Hello
>
> On Mon, Jul 09, 2007 at 11:59:36AM +1000, Nick Piggin wrote:
> > >Jul 8 00:19:13 dorade.cyberlink.ch EFLAGS: 00210046
> > >(2.6.22-rc7-dorade #1)
> > >>The machine panics
> > >>after some days of running fine, the machine inst heavy loaded.
> > >>The Controller detects as stex device:
> > >kernel BUG at block/as-iosched.c:1084!
> > >
> > >BUG_ON(RB_EMPTY_ROOT(&ad->sort_list[REQ_ASYNC]));
> > Could be a bug in the driver that just happens to be caught by AS checks.
> > If you could test another scheduler (boot with elevator=deadline or
> > elevator=cfq),
> > it might help give us an idea.
>
> Ok, I now switched to cfg, and the machine panicd again. Panic attached.
> Any help is appreciated.
It really looks like a stex bug - perhaps it's doing double completions
of a request?
--
Jens Axboe
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: panics with 16port Promise Supertrack EX Controller
2007-07-09 1:59 ` Nick Piggin
2007-07-15 8:02 ` Flavio Curti
@ 2007-07-15 22:51 ` Flavio Curti
2007-07-16 0:35 ` Randy Dunlap
1 sibling, 1 reply; 10+ messages in thread
From: Flavio Curti @ 2007-07-15 22:51 UTC (permalink / raw)
To: linux-kernel
[-- Attachment #1: Type: text/plain, Size: 723 bytes --]
Hello
On Mon, Jul 09, 2007 at 11:59:36AM +1000, Nick Piggin wrote:
> >Jul 8 00:19:13 dorade.cyberlink.ch EFLAGS: 00210046
> >(2.6.22-rc7-dorade #1)
> >>The machine panics
> >>after some days of running fine, the machine inst heavy loaded.
> >>The Controller detects as stex device:
> >kernel BUG at block/as-iosched.c:1084!
> >
> >BUG_ON(RB_EMPTY_ROOT(&ad->sort_list[REQ_ASYNC]));
> Could be a bug in the driver that just happens to be caught by AS checks.
> If you could test another scheduler (boot with elevator=deadline or
> elevator=cfq),
> it might help give us an idea.
Ok, I now switched to cfg, and the machine panicd again. Panic attached.
Any help is appreciated.
Flavio Curti
--
http://no-way.org/~fcu/
[-- Attachment #2: dorade-kdump --]
[-- Type: text/plain, Size: 1485 bytes --]
blk_queue_end_tag: attempt to clear non-busy tag (1)
------------[ cut here ]------------
kernel BUG at block/ll_rw_blk.c:2758!
invalid opcode: 0000 [#1]
SMP
Modules linked in:
i2c_i801
i2c_core
CPU: 0
EIP: 0060:[<c02a3164>] Not tainted VLI
EFLAGS: 00010002 (2.6.22.1-dorade #1)
EIP is at __blk_put_request+0x74/0x80
eax: 00000000 ebx: dae3e578 ecx: c02ac600 edx: 00000000
esi: 00030784 edi: f7f28b24 ebp: f7f28b24 esp: c0621e50
ds: 007b es: 007b fs: 00d8 gs: 0000 ss: 0068
Process swapper (pid: 0, ti=c0620000 task=c05d8340 task.ti=c0620000)
Stack:
dae3e578
e2b5aac0
00000001
scsi_end_request+0x94/0xe0
[<c03743c6>]
scsi_io_completion+0x86/0x3e0
[<c012b08a>]
del_timer+0x5a/0x70
[<c038fc7b>]
sd_rw_intr+0x2b/0x200
[<c038bf8e>]
stex_mu_intr+0x12e/0x540
[<c036fe85>]
scsi_finish_command+0x45/0x60
[<c02a4c88>]
blk_done_softirq+0x58/0x70
[<c0126e92>]
__do_softirq+0x82/0x100
[<c0126f47>]
do_softirq+0x37/0x40
[<c0127185>]
irq_exit+0x75/0x80
[<c0106410>]
do_IRQ+0x40/0x80
[<c0139796>]
getnstimeofday+0x36/0xd0
[<c0104503>]
common_interrupt+0x23/0x28
[<c0101d26>]
mwait_idle_with_hints+0x46/0x60
[<c0101e85>]
cpu_idle+0x65/0x90
[<c0628d1f>]
start_kernel+0x2ef/0x370
[<c06284e0>]
unknown_bootoption+0x0/0x250
=======================
Code:
8b
57
24
89
d8
83
e6
01
e8
ca
1b
eb
ff
89
f2
89
f8
89
e9
5b
5e
5f
5d
e9
3b
ff
last message repeated 2 times
89
da
89
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: panics with 16port Promise Supertrack EX Controller
2007-07-15 22:51 ` Flavio Curti
@ 2007-07-16 0:35 ` Randy Dunlap
2007-07-16 8:50 ` Flavio Curti
0 siblings, 1 reply; 10+ messages in thread
From: Randy Dunlap @ 2007-07-16 0:35 UTC (permalink / raw)
To: Flavio Curti; +Cc: linux-kernel
On Mon, 16 Jul 2007 00:51:55 +0200 Flavio Curti wrote:
> Hello
>
> On Mon, Jul 09, 2007 at 11:59:36AM +1000, Nick Piggin wrote:
> > >Jul 8 00:19:13 dorade.cyberlink.ch EFLAGS: 00210046
> > >(2.6.22-rc7-dorade #1)
> > >>The machine panics
> > >>after some days of running fine, the machine inst heavy loaded.
> > >>The Controller detects as stex device:
> > >kernel BUG at block/as-iosched.c:1084!
> > >
> > >BUG_ON(RB_EMPTY_ROOT(&ad->sort_list[REQ_ASYNC]));
> > Could be a bug in the driver that just happens to be caught by AS checks.
> > If you could test another scheduler (boot with elevator=deadline or
> > elevator=cfq),
> > it might help give us an idea.
>
> Ok, I now switched to cfg, and the machine panicd again. Panic attached.
> Any help is appreciated.
Make sure that you switched to "cfq" and not "cfg".
---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: panics with 16port Promise Supertrack EX Controller
2007-07-16 0:35 ` Randy Dunlap
@ 2007-07-16 8:50 ` Flavio Curti
0 siblings, 0 replies; 10+ messages in thread
From: Flavio Curti @ 2007-07-16 8:50 UTC (permalink / raw)
To: Randy Dunlap; +Cc: Flavio Curti, linux-kernel
Hello
On Sun, Jul 15, 2007 at 05:35:15PM -0700, Randy Dunlap wrote:
> On Mon, 16 Jul 2007 00:51:55 +0200 Flavio Curti wrote:
> > Ok, I now switched to cfg, and the machine panicd again. Panic attached.
> > Any help is appreciated.
> Make sure that you switched to "cfq" and not "cfg".
Oh, it is. Typo when writing the email, sorry...
kernel /boot/vmlinuz-2.6.22.1-dorade root=/dev/sda1 ro
netconsole=4444@62.12.130.18/eth0,514@213.189.128.9/ elevator=cfq
Kind Regards
Flavio Curti
--
http://no-way.org/~fcu/
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: panics with 16port Promise Supertrack EX Controller
2007-07-08 23:20 ` Michal Piotrowski
2007-07-09 1:59 ` Nick Piggin
@ 2007-07-09 10:45 ` Flavio Curti
1 sibling, 0 replies; 10+ messages in thread
From: Flavio Curti @ 2007-07-09 10:45 UTC (permalink / raw)
To: Michal Piotrowski; +Cc: Flavio Curti, linux-kernel, Jens Axboe, Nick Piggin
On Mon, Jul 09, 2007 at 01:20:51AM +0200, Michal Piotrowski wrote:
> On 09/07/07, Flavio Curti <lists.fcu@no-way.org> wrote:
> Jul 8 00:19:13 dorade.cyberlink.ch EFLAGS: 00210046
> (2.6.22-rc7-dorade #1)
> Is this a regression?
I also tried 2.6.21.5, which showed the same problem. So I doubt it's a
2.6.22rc7 problem.
I'll try a different scheduler as suggested.
Kind regards
Flavio Curti
--
http://no-way.org/~fcu/
^ permalink raw reply [flat|nested] 10+ messages in thread
* RE: panics with 16port Promise Supertrack EX Controller
@ 2007-07-17 1:26 Ed Lin
0 siblings, 0 replies; 10+ messages in thread
From: Ed Lin @ 2007-07-17 1:26 UTC (permalink / raw)
To: Jens Axboe, Flavio Curti
Cc: Nick Piggin, Michal Piotrowski, Flavio Curti, linux-kernel,
Promise_Linux
> -----Original Message-----
> From: Jens Axboe [mailto:jens.axboe@oracle.com]
> Sent: Monday, July 16, 2007 4:18 AM
> To: Flavio Curti
> Cc: Nick Piggin; Michal Piotrowski; Flavio Curti;
> linux-kernel@vger.kernel.org; Promise_Linux
> Subject: Re: panics with 16port Promise Supertrack EX Controller
>
>
> On Sun, Jul 15 2007, Flavio Curti wrote:
> > Hello
> >
> > On Mon, Jul 09, 2007 at 11:59:36AM +1000, Nick Piggin wrote:
> > > >Jul 8 00:19:13 dorade.cyberlink.ch EFLAGS: 00210046
> > > >(2.6.22-rc7-dorade #1)
> > > >>The machine panics
> > > >>after some days of running fine, the machine inst heavy loaded.
> > > >>The Controller detects as stex device:
> > > >kernel BUG at block/as-iosched.c:1084!
> > > >
> > > >BUG_ON(RB_EMPTY_ROOT(&ad->sort_list[REQ_ASYNC]));
> > > Could be a bug in the driver that just happens to be
> caught by AS checks.
> > > If you could test another scheduler (boot with
> elevator=deadline or
> > > elevator=cfq),
> > > it might help give us an idea.
> >
> > Ok, I now switched to cfg, and the machine panicd again.
> Panic attached.
> > Any help is appreciated.
>
> It really looks like a stex bug - perhaps it's doing double
> completions
> of a request?
>
I will come up with something related to tag lock, and see
what happens after that.
Ed Lin
^ permalink raw reply [flat|nested] 10+ messages in thread
end of thread, other threads:[~2007-07-17 1:29 UTC | newest]
Thread overview: 10+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-07-08 22:52 panics with 16port Promise Supertrack EX Controller Flavio Curti
2007-07-08 23:20 ` Michal Piotrowski
2007-07-09 1:59 ` Nick Piggin
2007-07-15 8:02 ` Flavio Curti
2007-07-16 11:17 ` Jens Axboe
2007-07-15 22:51 ` Flavio Curti
2007-07-16 0:35 ` Randy Dunlap
2007-07-16 8:50 ` Flavio Curti
2007-07-09 10:45 ` Flavio Curti
-- strict thread matches above, loose matches on Subject: below --
2007-07-17 1:26 Ed Lin
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox