The Linux Kernel Mailing List
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@zip.com.au>
To: "David S. Miller" <davem@redhat.com>
Cc: mingo@elte.hu, bcrl@redhat.com, velco@fadata.bg,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH] Scalable page cache
Date: Mon, 26 Nov 2001 11:29:19 -0800	[thread overview]
Message-ID: <3C02980E.8EDED15D@zip.com.au> (raw)
In-Reply-To: <Pine.LNX.4.33.0111262115261.17043-100000@localhost.localdomain>, <20011126131641.A13955@redhat.com> <Pine.LNX.4.33.0111262115261.17043-100000@localhost.localdomain> <20011126.103327.18298379.davem@redhat.com>

"David S. Miller" wrote:
> 
>    From: Ingo Molnar <mingo@elte.hu>
>    Date: Mon, 26 Nov 2001 21:29:39 +0100 (CET)
> 
>    so i'm not against removing (or improving) the hash [our patch in fact
>    just left the hash alone], but the patch presented is not a win IMO.
> 
> Maybe you should give it a test to find out for sure :)

umm..  I've never seen any numbers from you, David.

Correct me if I'm wrong, but the pagecache_hash cost is
significant in the following situations:

1: TUX, because its pagecache lookups are not associated with
   a page copy.  This copy makes the benefits of the patch
   unmeasurable with other workloads.

1a: Other sendfile-intensive applications. (Theoretical benefit.
    No benchmark results have been seen).

2: NUMA hardware, where the cost of cacheline transfer is much
   higher.

ergo, there is no point in futzing with the pagecache_lock *at all*
until either TUX is merged, or we decide to support large-scale
NUMA hardware well, which will require changes in other places.

Prove me wrong.  Please.

-

  reply	other threads:[~2001-11-26 19:32 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-11-26 12:31 [PATCH] Scalable page cache Momchil Velikov
2001-11-26 17:22 ` Ingo Molnar
2001-11-26 17:23   ` Momchil Velikov
2001-11-26 18:16     ` Benjamin LaHaise
2001-11-26 18:42       ` Rik van Riel
2001-11-26 20:29       ` Ingo Molnar
2001-11-26 18:33         ` David S. Miller
2001-11-26 19:29           ` Andrew Morton [this message]
2001-11-26 19:35             ` David S. Miller
2001-11-26 18:34         ` Benjamin LaHaise
2001-11-26 20:40       ` Ingo Molnar
2001-11-26 19:19         ` Benjamin LaHaise
2001-11-26 21:00         ` Ingo Molnar
2001-11-26 20:43       ` Daniel Phillips
2001-11-26 19:49     ` Ingo Molnar
2001-11-26 18:08       ` Christoph Hellwig
2001-11-26 20:13         ` Ingo Molnar
2001-11-26 21:09     ` Ingo Molnar
2001-11-26 19:18       ` David S. Miller
2001-11-26 19:45         ` Andrew Morton
2001-11-26 19:57           ` David S. Miller
2001-11-26 20:03             ` Andrew Morton
2001-11-26 20:37             ` Linus Torvalds
2001-11-26 21:02               ` Andrew Morton
2001-11-26 22:23                 ` Linus Torvalds
2001-11-26 22:49                   ` Jeff Garzik
2001-11-26 23:06                     ` Linus Torvalds
2001-11-26 21:10       ` Ingo Molnar
2001-11-26 17:29   ` Linus Torvalds
2001-11-26 20:03     ` Ingo Molnar
2001-11-26 18:02   ` David S. Miller
2001-11-26 18:11     ` Rik van Riel
2001-11-27  8:07     ` benchmark results: " Anton Blanchard
2001-11-26 18:52   ` [PATCH] " Daniel Phillips

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=3C02980E.8EDED15D@zip.com.au \
    --to=akpm@zip.com.au \
    --cc=bcrl@redhat.com \
    --cc=davem@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=velco@fadata.bg \
    /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