From: Dave Olien <dmo@osdl.org>
To: axboe@suse.de
Cc: linux-kernel@vger.kernel.org
Subject: Re: DAC960 crash dequeueing request
Date: Fri, 6 Jun 2003 09:51:53 -0700 [thread overview]
Message-ID: <20030606165153.GA20224@osdl.org> (raw)
In-Reply-To: <20030604220415.GA15621@osdl.org>
Mary,
How's this for a response to Andrew and lkml?
------------------------------------------------------------------------
Andrew,
After running several iterations of the database workload, it seems
the patch below did indeed eliminate the BUG() we were hitting.
In addition, the patched kernel seems to perform somwhat better.
The workload performance is improved by about 2%, and the standard
deviation in performance between iterations of the work load was reduced
from 137 to 48.
A performance results comparison can be found at
http://www.osdl.org/projects/dbt2dev/results/8way/70AKPM/results.html
Thanks to Mary Meredith <maryedie@osdl.org> for the performance
measurements and comparisons, and for hitting the BUG() in the first place.
Andrew Morton wrote:
>
> Dave Olien wrote:
> >
> > In linux 2.5.70, with no patches applied, we've had one BUG of the form:
> >
> > kernel BUG at include/linux/blkdev.h:407!
>
>
>
> The below should fix it.
>
>
> diff -puN drivers/block/deadline-iosched.c~deadline-hash-removal-fix drivers/block/deadline-iosched.c
> --- 25/drivers/block/deadline-iosched.c~deadline-hash-removal-fix 2003-06-04 00:50:36.000000000 -0700
> +++ 25-akpm/drivers/block/deadline-iosched.c 2003-06-04 00:50:36.000000000 -0700
> @@ -121,6 +121,15 @@ static inline void deadline_del_drq_hash
> __deadline_del_drq_hash(drq);
> }
>
> +static void
> +deadline_remove_merge_hints(request_queue_t *q, struct deadline_rq *drq)
> +{
> + deadline_del_drq_hash(drq);
> +
> + if (q->last_merge == &drq->request->queuelist)
> + q->last_merge = NULL;
> +}
> +
> static inline void
> deadline_add_drq_hash(struct deadline_data *dd, struct deadline_rq *drq)
> {
> @@ -310,7 +319,7 @@ static void deadline_remove_request(requ
> struct deadline_data *dd = q->elevator.elevator_data;
>
> list_del_init(&drq->fifo);
> - deadline_del_drq_hash(drq);
> + deadline_remove_merge_hints(q, drq);
> deadline_del_drq_rb(dd, drq);
> }
> }
>
> _
>
> ----- End forwarded message -----
>
> --=-OFfR5BGtLWFFgyRX7TNt--
prev parent reply other threads:[~2003-06-06 16:37 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2003-06-04 22:04 DAC960 crash dequeueing request Dave Olien
2003-06-04 22:08 ` Andrew Morton
2003-06-06 16:51 ` Dave Olien [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20030606165153.GA20224@osdl.org \
--to=dmo@osdl.org \
--cc=axboe@suse.de \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox