netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Anton Blanchard <anton@samba.org>
To: Arthur Kepner <akepner@sgi.com>
Cc: Leonid Grossman <leonid.grossman@neterion.com>,
	muhammad.shafiq@neterion.com, ramkrishna.vepa@neterion.com,
	netdev@oss.sgi.com
Subject: Re: [PATCH] s2io: replace readq() with mmiowb() in s2io_xmit()
Date: Tue, 3 May 2005 13:05:15 +1000	[thread overview]
Message-ID: <20050503030515.GD12682@krispykreme> (raw)
In-Reply-To: <Pine.LNX.4.61.0505021844480.16692@linux.site>

 
Hi,

> I thought that an arch was supposed to define mmiowb() if it needs 
> it. In this case, the readq() is only being used for the side-effect 
> of ordering the previous writes (the neterion folks can correct me 
> if I'm wrong). 

The documentation suggests mmiowb does not do write posting:

      <para>
        In addition to write posting, on some large multiprocessing systems
        (e.g. SGI Challenge, Origin and Altix machines) posted writes won't
        be strongly ordered coming from different CPUs.  Thus it's important
        to properly protect parts of your driver that do memory-mapped writes
        with locks and use the <function>mmiowb</function> to make sure they
        arrive in the order intended.  Issuing a regular <function>readX
        </function> will also ensure write ordering, but should only be used
        when the driver has to be sure that the write has actually arrived
        at the device (not that it's simply ordered with respect to other
        writes), since a full <function>readX</function> is a relatively
        expensive operation.
      </para>

So if we require the write to be posted we will need the read. If
however we only need to order wrt subsequent IO writes then I agree
mmiowb can be used.

Anton

      parent reply	other threads:[~2005-05-03  3:05 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <200504051836.j35IanDD005402@guinness.s2io.com>
2005-05-03  1:09 ` [PATCH] s2io: replace readq() with mmiowb() in s2io_xmit() Arthur Kepner
2005-05-03  1:40   ` Anton Blanchard
2005-05-03  2:11     ` Arthur Kepner
2005-05-03  2:58       ` Ramkrishna Vepa
2005-05-03  3:06         ` Anton Blanchard
2005-05-03 14:52           ` Arthur Kepner
2005-05-03 16:23             ` Ramkrishna Vepa
2005-05-03  3:05       ` Anton Blanchard [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=20050503030515.GD12682@krispykreme \
    --to=anton@samba.org \
    --cc=akepner@sgi.com \
    --cc=leonid.grossman@neterion.com \
    --cc=muhammad.shafiq@neterion.com \
    --cc=netdev@oss.sgi.com \
    --cc=ramkrishna.vepa@neterion.com \
    /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;
as well as URLs for NNTP newsgroup(s).