public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: "David S. Miller" <davem@redhat.com>
To: ionut@badula.org
Cc: torvalds@transmeta.com, linux-kernel@vger.kernel.org,
	James.Bottomley@steeleye.com
Subject: Re: [PATCH] add new DMA_ADDR_T_SIZE define
Date: Wed, 19 Feb 2003 14:29:52 -0800 (PST)	[thread overview]
Message-ID: <20030219.142952.27404695.davem@redhat.com> (raw)
In-Reply-To: <Pine.LNX.4.44.0302191706330.29393-100000@guppy.limebrokerage.com>

   From: Ion Badulescu <ionut@badula.org>
   Date: Wed, 19 Feb 2003 17:41:39 -0500 (EST)

   > Nearly all cards today are 64-bit DMA address descriptors only.
   > So if anything, this new ifdef will get less and less used over
   > time.
   
   Not true. Even if the only descriptors available are 64-bit, there is no 
   reason why I should have to care about the upper 32 bits when I know that 
   my dma_addr_t will always be 32-bit, at compile time. I can simply 
   memset(0) the entire descriptor and initialize only the bottom 32 bits. 

Yes true, storing the two consequetive 32-bit values is better
for store buffer compression of the cpu.  Using memset is much
more inefficient because you push the full set of data once
then you push non-compressible stores to the same data through
the cpu.

I'm not talking out of my ass, I've measured this.

  reply	other threads:[~2003-02-19 22:35 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-02-19 18:11 [PATCH] add new DMA_ADDR_T_SIZE define James Bottomley
2003-02-19 18:20 ` Ion Badulescu
2003-02-19 22:07   ` David S. Miller
2003-02-19 22:41     ` Ion Badulescu
2003-02-19 22:29       ` David S. Miller [this message]
2003-02-19 22:53         ` Ion Badulescu
2003-02-19 22:38           ` David S. Miller
  -- strict thread matches above, loose matches on Subject: below --
2003-02-23  7:02 Albert Cahalan
2003-02-23  7:00 ` David S. Miller
2003-02-23  7:20   ` Albert Cahalan
2003-02-19 16:26 Ion Badulescu
2003-02-19 17:20 ` Randy.Dunlap
2003-02-19 17:28   ` Ion Badulescu
2003-02-19 22:06     ` David S. Miller
2003-02-19 22:04       ` Ion Badulescu
2003-02-19 22:06   ` David S. Miller
2003-02-19 21:48     ` Jeff Garzik
2003-02-22 10:06       ` Ingo Oeser
2003-02-19 22:01 ` David S. Miller

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=20030219.142952.27404695.davem@redhat.com \
    --to=davem@redhat.com \
    --cc=James.Bottomley@steeleye.com \
    --cc=ionut@badula.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=torvalds@transmeta.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