All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jes Sorensen <jes@sgi.com>
To: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Cc: Alan Cox <alan@redhat.com>, Linus Torvalds <torvalds@osdl.org>,
	linux-kernel@vger.kernel.org, Jeremy Higdon <jeremy@sgi.com>
Subject: Re: [patch] SGIIOC4 limit request size
Date: Wed, 01 Feb 2006 11:41:16 +0100	[thread overview]
Message-ID: <43E0904C.1020201@sgi.com> (raw)
In-Reply-To: <58cb370e0602010234p62521a00h6d8920c84cac44d5@mail.gmail.com>

Bartlomiej Zolnierkiewicz wrote:
> On 01 Feb 2006 03:59:16 -0500, Jes Sorensen <jes@sgi.com> wrote:
>>This one takes care of a problem with the SGI IOC4 driver where it
>>hits DMA problems if the request grows too large.
> 
> 
> Does this happen only for CONFIG_IA64_PAGE_SIZE_4KB=y
> or CONFIG_IA64_PAGE_SIZE_8KB=y?
> 
> from sgiioc4.c:
> 
> /* Each Physical Region Descriptor Entry size is 16 bytes (2 * 64 bits) */
> /* IOC4 has only 1 IDE channel */
> #define IOC4_PRD_BYTES       16
> #define IOC4_PRD_ENTRIES     (PAGE_SIZE /(4*IOC4_PRD_BYTES))
> 
> As limiting request size to 127 sectors punishes performance
> wouldn't it be better to define IOC4_PRD_ENTRIES to 256
> if this is possible (would need 4 pages for PAGE_SIZE=4096
> and 2 for PAGE_SIZE=8192)?

This happens with the default page size which is 16KB, ie.
IOC4_PRD_ENTRIES=256, the problem is not due to the request
going beyond the number of PRD_ENTRIES. I haven't tried with smaller
page sizes but I would assume the problem would be the same.

Even with this patch performance seems very reasonable.

Jes

  reply	other threads:[~2006-02-01 10:41 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-02-01  8:59 [patch] SGIIOC4 limit request size Jes Sorensen
2006-02-01 10:34 ` Bartlomiej Zolnierkiewicz
2006-02-01 10:41   ` Jes Sorensen [this message]
2006-02-01 10:49   ` Jeremy Higdon
2006-02-01 11:08     ` Bartlomiej Zolnierkiewicz
2006-02-01 11:17       ` Jeremy Higdon
2006-02-01 11:26         ` Bartlomiej Zolnierkiewicz
2006-02-01 11:36           ` Jeremy Higdon
2006-02-01 12:44             ` Bartlomiej Zolnierkiewicz
2006-02-02  8:00               ` [patch] Fix DMA timeouts with sgiioc4 Jeremy Higdon
2006-02-02  8:45                 ` Jes Sorensen
2006-02-01 13:39             ` [patch] SGIIOC4 limit request size Alan Cox
2006-02-01 14:53               ` Bartlomiej Zolnierkiewicz

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=43E0904C.1020201@sgi.com \
    --to=jes@sgi.com \
    --cc=alan@redhat.com \
    --cc=bzolnier@gmail.com \
    --cc=jeremy@sgi.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=torvalds@osdl.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.