public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Marcelo Tosatti <marcelo.tosatti@cyclades.com>
To: Andrew Morton <akpm@osdl.org>
Cc: linux-kernel@vger.kernel.org
Subject: Re: [PATCH] cacheline align pagevec structure
Date: Thu, 9 Sep 2004 18:41:13 -0300	[thread overview]
Message-ID: <20040909214113.GB5723@logos.cnet> (raw)
In-Reply-To: <20040909154906.57f9391b.akpm@osdl.org>

On Thu, Sep 09, 2004 at 03:49:06PM -0700, Andrew Morton wrote:
> Marcelo Tosatti <marcelo.tosatti@cyclades.com> wrote:
> >
> > Right now it is 140 bytes on 64-bit and 72 bytes on 32-bit. Thats just a little bit more 
> > than a power of 2 (which will cacheline align), so shrink it to be aligned: 64 bytes on 
> > 32bit and 124bytes on 64-bit. 
> > 
> > It now occupies two cachelines most of the time instead of three. 
> > 
> > I changed nr and cold to "unsigned short" because they'll never reach 2 ^ 16.
> > 
> > I do not see a problem with changing pagevec to "15" page pointers either, 
> > Andrew, is there a special reason for that "16"? Is intentional to align
> > to 64 kbytes (IO device alignment)? I dont think that matters much because
> > of the elevator which sorts and merges requests anyway?
> > 
> > 
> > 
> > Did some reaim benchmarking on 4way PIII (32byte cacheline), with 512MB RAM:
> > 
> > #### stock 2.6.9-rc1-mm4 ####
> > 
> > Peak load Test: Maximum Jobs per Minute 4144.44 (average of 3 runs)
> > Quick Convergence Test: Maximum Jobs per Minute 4007.86 (average of 3 runs)
> > 
> > Peak load Test: Maximum Jobs per Minute 4207.48 (average of 3 runs)
> > Quick Convergence Test: Maximum Jobs per Minute 3999.28 (average of 3 runs)
> > 
> > #### shrink-pagevec #####
> > 
> > Peak load Test: Maximum Jobs per Minute 4717.88 (average of 3 runs)
> > Quick Convergence Test: Maximum Jobs per Minute 4360.59 (average of 3 runs)
> > 
> > Peak load Test: Maximum Jobs per Minute 4493.18 (average of 3 runs)
> > Quick Convergence Test: Maximum Jobs per Minute 4327.77 (average of 3 runs)
> 
> I think the patch make sense, but I'm very sceptical about the benchmarks
> ;)

Why's that? You think changing to the number of pages in the pagevec to "15" instead
"16" is the cause?

Or that the performance increase is not a direct effect of the one cacheline 
saved per pagevec instance?

Or you think such benchmark is too specific to be interpreted as a broad
vision of performance? 

:)


  reply	other threads:[~2004-09-09 23:06 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-09-09 16:39 [PATCH] cacheline align pagevec structure Marcelo Tosatti
2004-09-09 22:49 ` Andrew Morton
2004-09-09 21:41   ` Marcelo Tosatti [this message]
2004-09-09 23:20     ` Andrew Morton
2004-09-09 22:52 ` Andrew Morton
2004-09-09 23:09   ` William Lee Irwin III
2004-09-09 22:12     ` Marcelo Tosatti
2004-09-09 23:59       ` William Lee Irwin III
2004-09-09 23:22     ` Andrew Morton
2004-09-10  0:07       ` [pagevec] resize pagevec to O(lg(NR_CPUS)) William Lee Irwin III
2004-09-10  4:56         ` Nick Piggin
2004-09-10  4:59           ` Nick Piggin
2004-09-10 17:49           ` Marcelo Tosatti
2004-09-12  0:29             ` Nick Piggin
2004-09-12  5:23               ` William Lee Irwin III
2004-09-12  4:36                 ` Nick Piggin
2004-09-12  4:56             ` William Lee Irwin III
2004-09-12  4:28               ` Nick Piggin
2004-09-12  6:27                 ` William Lee Irwin III
2004-09-12  6:03                   ` Nick Piggin
2004-09-12  7:19                     ` William Lee Irwin III
2004-09-12  7:42                       ` Andrew Morton
2004-09-14  2:18                         ` William Lee Irwin III
2004-09-14  2:57                           ` Andrew Morton
2004-09-14  3:12                             ` William Lee Irwin III
2004-09-12  8:57                       ` William Lee Irwin III
2004-09-13 22:21                 ` Marcelo Tosatti
2004-09-14  1:59                   ` Nick Piggin

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=20040909214113.GB5723@logos.cnet \
    --to=marcelo.tosatti@cyclades.com \
    --cc=akpm@osdl.org \
    --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