public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Andi Kleen <ak@suse.de>
To: Manfred Spraul <manfred@colorfullife.com>
Cc: Andi Kleen <ak@suse.de>,
	linux-kernel@vger.kernel.org, Dave Hansen <haveblue@us.ibm.com>
Subject: Re: Horrible L2 cache effects from kernel compile
Date: Tue, 25 Feb 2003 19:42:21 +0100	[thread overview]
Message-ID: <20030225184221.GA9013@wotan.suse.de> (raw)
In-Reply-To: <3E5BB7EE.5090301@colorfullife.com>

> Are you sure that this will help?
> With a smaller table, you might cause fewer cache misses for the table 
> lookup. Instead you get longer hash chains. Walking linked lists 
> probably causes more cache line misses than the single array lookup.

That's true when the hash table has a reasonable size. But with 1MB 
and bigger hash tables you are guaranteed to get a cache miss for most
head bucket access, no matter how many dentries you have. The hash
function is actively working against your cache here.

The dentries are actually more likely to fit into dcache because they
don't get artificially spread out over the cache space.

-Andi

  parent reply	other threads:[~2003-02-25 18:32 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-02-25 18:37 Horrible L2 cache effects from kernel compile Manfred Spraul
2003-02-25 18:41 ` Dave Hansen
2003-02-25 18:42 ` Andi Kleen [this message]
2003-02-25 19:29 ` Martin J. Bligh
2003-02-25 22:18 ` Linus Torvalds
2003-03-03 19:03   ` Benjamin LaHaise
2003-03-03 19:13     ` Linus Torvalds
2003-03-03 23:58       ` Alan Cox
2003-03-03 22:57         ` Andrew Morton
2003-03-03 23:01         ` Benjamin LaHaise
2003-03-03 23:09         ` Linus Torvalds
2003-03-05 20:19         ` dean gaudet
     [not found] <3E5ABBC1.8050203@us.ibm.com.suse.lists.linux.kernel>
2003-02-25 16:16 ` Andi Kleen
2003-02-27  3:24   ` Dave Hansen
2003-02-27 16:36   ` Jan Harkes
     [not found] ` <b3ekil$1cp$1@penguin.transmeta.com.suse.lists.linux.kernel>
     [not found]   ` <20030225170546.GA23772@morningstar.nowhere.lie.suse.lists.linux.kernel>
2003-02-25 17:20     ` Andi Kleen
2003-02-26 18:22       ` Jamie Lokier
  -- strict thread matches above, loose matches on Subject: below --
2003-02-25  0:41 Dave Hansen
2003-02-25  0:59 ` William Lee Irwin III
2003-02-25  1:01   ` Dave Hansen
2003-02-25  3:15   ` John Levon
2003-02-25  3:15     ` William Lee Irwin III
2003-02-25  3:35     ` Andrew Morton
2003-02-25  4:13       ` Martin J. Bligh
2003-02-25 11:57         ` John Levon
2003-02-25  2:31 ` Linus Torvalds
2003-02-25 17:05   ` John W. M. Stevens

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=20030225184221.GA9013@wotan.suse.de \
    --to=ak@suse.de \
    --cc=haveblue@us.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=manfred@colorfullife.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