From: Jeremy Fitzhardinge <jeremy@goop.org>
To: Ingo Molnar <mingo@elte.hu>
Cc: Andi Kleen <ak@suse.de>, Eric Dumazet <dada1@cosmosbay.com>,
akpm@osdl.org, Arjan van de Ven <arjan@infradead.org>,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH] i386-pda UP optimization
Date: Wed, 15 Nov 2006 09:28:15 -0800 [thread overview]
Message-ID: <455B4E2F.7040408@goop.org> (raw)
In-Reply-To: <20061115172003.GA20403@elte.hu>
Ingo Molnar wrote:
> Eric's test shows a 5% slowdown. That's far from cheap.
>
It seems like an absurdly large difference. PDA references aren't all
that common in the kernel; for the %gs prefix on PDA accesses to be
causing a 5% overall difference in a test like this means that the
prefixes would have to be costing hundreds or thousands of cycles, which
seems absurd. Particularly since Eric's patch doesn't touch head.S, so
the %gs save/restore is still being executed.
Are we sure this isn't a cache layout issue? Eric, did you try evicting
your executable from pagecache between runs to see if you get variation
depending on what physical pages it gets put into? (Making several
copies of the executable should have the same effect.)
J
next prev parent reply other threads:[~2006-11-15 17:28 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-09-12 7:35 i386 PDA patches use of %gs Arjan van de Ven
2006-09-12 7:48 ` Jeremy Fitzhardinge
2006-09-12 7:56 ` Arjan van de Ven
2006-09-12 8:31 ` Jeremy Fitzhardinge
2006-11-15 11:27 ` [PATCH] i386-pda UP optimization Eric Dumazet
2006-11-15 11:32 ` Andi Kleen
2006-11-15 17:20 ` Ingo Molnar
2006-11-15 17:24 ` Andi Kleen
2006-11-15 17:46 ` Eric Dumazet
2006-11-15 17:49 ` Ingo Molnar
2006-11-15 17:58 ` Eric Dumazet
2006-11-15 18:01 ` Ingo Molnar
2006-11-21 11:38 ` Eric Dumazet
2006-11-21 21:42 ` Jeremy Fitzhardinge
2006-11-21 21:52 ` Andi Kleen
2006-11-21 22:10 ` Jeremy Fitzhardinge
2006-11-21 21:58 ` Eric Dumazet
2006-11-21 23:12 ` Jeremy Fitzhardinge
2006-11-15 17:28 ` Jeremy Fitzhardinge [this message]
2006-11-15 17:32 ` Ingo Molnar
2006-11-15 17:59 ` Jeremy Fitzhardinge
2006-11-15 18:05 ` Eric Dumazet
2006-11-15 18:28 ` Jeremy Fitzhardinge
2006-11-15 18:31 ` Ingo Molnar
2006-11-15 18:01 ` Arjan van de Ven
2006-11-15 18:24 ` Jeremy Fitzhardinge
2006-11-15 19:06 ` Ingo Molnar
2006-11-17 0:24 ` Jeremy Fitzhardinge
2006-11-15 17:52 ` Jeremy Fitzhardinge
2006-11-28 23:12 ` Jeremy Fitzhardinge
2006-11-29 9:30 ` Eric Dumazet
2006-11-29 9:56 ` Jeremy Fitzhardinge
2006-09-13 1:00 ` i386 PDA patches use of %gs Jeremy Fitzhardinge
2006-09-13 9:59 ` Ingo Molnar
2006-09-13 16:17 ` Jeremy Fitzhardinge
2006-11-15 18:26 ` Ingo Molnar
2006-11-15 18:29 ` Ingo Molnar
2006-11-15 18:43 ` Jeremy Fitzhardinge
2006-11-15 18:44 ` Ingo Molnar
2006-11-15 18:39 ` Jeremy Fitzhardinge
2006-11-15 18:43 ` Ingo Molnar
2006-11-15 18:49 ` Jeremy Fitzhardinge
2006-11-15 18:49 ` Ingo Molnar
2006-11-15 19:00 ` Jeremy Fitzhardinge
2006-11-15 19:03 ` Ingo Molnar
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=455B4E2F.7040408@goop.org \
--to=jeremy@goop.org \
--cc=ak@suse.de \
--cc=akpm@osdl.org \
--cc=arjan@infradead.org \
--cc=dada1@cosmosbay.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
/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).