lkml.org 
[lkml]   [2005]   [Jun]   [9]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH] SATA NCQ #4
Jens Axboe wrote:
> On Thu, Jun 09 2005, Jens Axboe wrote:
>
>>On Thu, Jun 09 2005, Grant Coady wrote:
>>
>>>On Wed, 8 Jun 2005 13:45:26 +0200, Jens Axboe <axboe@suse.de> wrote:
>>>
>>>>Any chance you can log the boot process when it fails, using serial
>>>>console or something similar? At least write down the EIP of where it
>>>>fails :-)
>>>
>>>Guess what? I switched box on this morning with monitor off and
>>>the boot completed, 'cos I'd logged in much time later ssh. Didn't
>>>give it enough time yesterday :(
>>>
>>>I have one very large syslog... 139MB
>>>
>>>How much of that would you like :)
>>>
>>>Jun 9 04:27:45 sempro kernel: [<c0100ad3>] default_idle+0x23/0x30
>>>Jun 9 04:27:45 sempro kernel: [<c0100b58>] cpu_idle+0x48/0x60
>>>Jun 9 04:27:45 sempro kernel: [<c04867b8>] start_kernel+0x148/0x170
>>>Jun 9 04:27:45 sempro kernel: [<c04863a0>] unknown_bootoption+0x0/0x1b0
>>>Jun 9 04:27:45 sempro kernel: Badness in __ata_qc_complete at drivers/scsi/libata-core.c:3062
>>
>>Ah duh, I never tested on UP, you have to use assert_spin_locked() there
>>not the direct spin_is_locked().
>>
>>So your system was functioning just fine, you just got a warning for
>>every completed request slowing it down a lot :)
>>
>>This should fix it. Jeff, can you apply that incremental to the ncq
>>branch? Thanks!
>>
>>--- linux-2.6/drivers/scsi/libata-core.c~ 2005-06-09 08:20:34.000000000 +0200
>>+++ linux-2.6/drivers/scsi/libata-core.c 2005-06-09 08:22:24.000000000 +0200
>>@@ -3059,7 +3059,7 @@
>> struct ata_port *ap = qc->ap;
>> unsigned int tag, do_clear = 0;
>>
>>- WARN_ON(!spin_is_locked(&ap->host_set->lock));
>>+ WARN_ON(!assert_spin_locked(&ap->host_set->lock));
>>
>> if (likely(qc->flags & ATA_QCFLAG_ACTIVE)) {
>> assert(ap->queue_depth);
>
>
> Or just kill the check completely, it has served its purpose.

The solution that included the patch won the battle ;-)

I'll be pushing your incremental update out to kernel.org in a sec.

Jeff



-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-06-09 08:51    [W:0.065 / U:2.204 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site