From: Mukul Kotwani <mkotwani@yahoo.com>
To: Jens Axboe <axboe@suse.de>
Cc: linux-scsi@vger.kernel.org
Subject: Re: Max IO size
Date: Thu, 1 Aug 2002 16:54:32 -0700 (PDT) [thread overview]
Message-ID: <20020801235432.14372.qmail@web14005.mail.yahoo.com> (raw)
In-Reply-To: <20020801084547.GA12715@suse.de>
Thanks for the reply!
All my values are set high. The sg_tablesize value is
set to 249, max_sectors set to 1024, and I dont think
there is any other value which affects IO size as
such. I didnt see anything called max segment size/max
size in the host template. Any other things I am
missing?
Thanks!
Mukul
--- Jens Axboe <axboe@suse.de> wrote:
> On Wed, Jul 31 2002, Mukul Kotwani wrote:
> > Hello!
> > Couple of other questions..forgot to ask in the
> > previous email!
> >
> > 1) When we do 512 byte sequential reads, the mid
> > level( or someone else) seems to be combining the
> > requests and sending them down as one big 60k+
> > request. Can this combining of requests be
> disabled
> > somehow?
>
> not the mid layer, it's the block layer doing that
> for you. you can
> limit the size of each i/o with the max_sectors part
> of the scsi host
> template.
>
> > 2) If we do 512k IOs, the mid layer seems to be
> > breaking it up into 64 k chunks. The Scatter
> gather
> > size is set to 240, so there is no reason for it
> to
> > break it up. Is there something I am missing? Or
> is
> > there a tunable I can set to increase the IO size
> sent
> > down to the low level driver?
>
> again, the mid layer does not do any combining or
> breaking up of i/o
> requests. the block layer will assemble requests
> according to the rules
> set for it, these rules you control with the host
> template settings.
> these are things like max size, max segments, max
> segment size, etc.
>
> > 3) What exactly does the max_sectors in the host
> > structure do?
>
> see above :-)
>
> --
> Jens Axboe
>
> -
> To unsubscribe from this list: send the line
> "unsubscribe linux-scsi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at
http://vger.kernel.org/majordomo-info.html
__________________________________________________
Do You Yahoo!?
Yahoo! Health - Feel better, live better
http://health.yahoo.com
next prev parent reply other threads:[~2002-08-01 23:54 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-08-01 0:22 Max IO size Mukul Kotwani
2002-08-01 8:45 ` Jens Axboe
2002-08-01 23:54 ` Mukul Kotwani [this message]
2002-08-02 5:17 ` Mukul Kotwani
2002-08-02 6:39 ` Jens Axboe
2002-08-02 6:44 ` Jens Axboe
2002-08-02 6:55 ` Mukul Kotwani
2002-08-02 7:50 ` Performance Mukul Kotwani
2002-08-02 9:16 ` Performance Fabien Salvi
2002-08-02 14:16 ` Performance Craig Tierney
2002-08-02 23:01 ` Performance Mukul Kotwani
2002-08-02 23:06 ` Performance Randy.Dunlap
2002-08-03 8:26 ` Performance Mukul Kotwani
2002-08-03 16:34 ` Performance Randy.Dunlap
2002-08-03 18:16 ` Performance Simon Trimmer
2002-08-03 20:03 ` Performance Mukul Kotwani
2002-08-04 6:27 ` Performance Jeremy Higdon
2002-08-04 7:39 ` Performance Mukul Kotwani
2002-08-04 8:01 ` Performance Jeremy Higdon
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=20020801235432.14372.qmail@web14005.mail.yahoo.com \
--to=mkotwani@yahoo.com \
--cc=axboe@suse.de \
--cc=linux-scsi@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