From: Andrew Morton <akpm@digeo.com>
To: William Lee Irwin III <wli@holomorphy.com>
Cc: Andrea Arcangeli <andrea@suse.de>,
Norman Gaywood <norm@turing.une.edu.au>,
linux-kernel@vger.kernel.org
Subject: Re: Maybe a VM bug in 2.4.18-18 from RH 8.0?
Date: Fri, 06 Dec 2002 16:30:29 -0800 [thread overview]
Message-ID: <3DF14125.966E5899@digeo.com> (raw)
In-Reply-To: 20021207002133.GT9882@holomorphy.com
William Lee Irwin III wrote:
>
> > William Lee Irwin III wrote:
> > >
> > > ...
> > > A 16KB or 64KB kernel allocation unit would then annihilate
> >
> On Fri, Dec 06, 2002 at 04:01:24PM -0800, Andrew Morton wrote:
> > You want to be careful about this:
> > CPU: L1 I cache: 16K, L1 D cache: 16K
> > Because instantiating a 16k page into user pagetables in
> > one hit means that it must all be zeroed. With these large
> > pagesizes that means that the application is likely to get
> > 100% L1 misses against the new page, whereas it currently
> > gets 100% hits.
>
> 16K is reasonable; after that one might as well go all the way.
16k will cause serious slowdowns.
> About the only way to cope is amortizing it by cacheing zeroed pages,
> and that has other downsides.
So will that. You've seen the kernbench profiles...
You will need to find a way to clear the page just before it
is used.
next prev parent reply other threads:[~2002-12-07 0:22 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-12-06 0:13 Maybe a VM bug in 2.4.18-18 from RH 8.0? Norman Gaywood
2002-12-06 1:00 ` Andrew Morton
2002-12-06 1:17 ` Andrea Arcangeli
2002-12-06 1:34 ` Andrew Morton
2002-12-06 1:44 ` Andrea Arcangeli
2002-12-06 2:15 ` William Lee Irwin III
2002-12-06 2:28 ` Andrea Arcangeli
2002-12-06 2:41 ` William Lee Irwin III
2002-12-06 5:25 ` Andrew Morton
2002-12-06 5:48 ` Andrea Arcangeli
2002-12-06 6:14 ` William Lee Irwin III
2002-12-06 6:55 ` Andrew Morton
2002-12-06 7:14 ` GrandMasterLee
2002-12-06 7:25 ` Andrew Morton
2002-12-06 7:34 ` GrandMasterLee
2002-12-06 7:51 ` Andrew Morton
2002-12-06 11:37 ` Christoph Hellwig
2002-12-06 16:19 ` GrandMasterLee
2002-12-06 14:57 ` Andrea Arcangeli
2002-12-06 15:12 ` William Lee Irwin III
2002-12-06 23:32 ` Andrea Arcangeli
2002-12-06 23:45 ` William Lee Irwin III
2002-12-06 23:57 ` Andrea Arcangeli
2002-12-06 6:00 ` William Lee Irwin III
2002-12-06 22:28 ` Andrea Arcangeli
2002-12-06 23:21 ` William Lee Irwin III
2002-12-06 23:50 ` Andrea Arcangeli
2002-12-07 0:30 ` William Lee Irwin III
2002-12-07 0:01 ` Andrew Morton
2002-12-07 0:21 ` William Lee Irwin III
2002-12-07 0:30 ` Andrew Morton [this message]
2002-12-07 2:19 ` Alan Cox
2002-12-07 1:46 ` William Lee Irwin III
2002-12-07 1:56 ` Andrea Arcangeli
2002-12-07 2:31 ` Alan Cox
2002-12-07 2:09 ` William Lee Irwin III
2002-12-07 0:22 ` Andrea Arcangeli
2002-12-07 0:35 ` Andrew Morton
2002-12-07 0:46 ` William Lee Irwin III
2002-12-07 10:55 ` Arjan van de Ven
2002-12-06 10:36 ` Arjan van de Ven
2002-12-06 14:23 ` William Lee Irwin III
2002-12-06 15:12 ` William Lee Irwin III
2002-12-06 22:34 ` Andrea Arcangeli
2002-12-07 18:27 ` Eric W. Biederman
2002-12-06 1:08 ` Andrea Arcangeli
[not found] <mailman.1039133948.27411.linux-kernel2news@redhat.com>
2002-12-06 0:35 ` Pete Zaitcev
2002-12-06 1:27 ` Norman Gaywood
2002-12-06 12:48 ` Rik van Riel
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=3DF14125.966E5899@digeo.com \
--to=akpm@digeo.com \
--cc=andrea@suse.de \
--cc=linux-kernel@vger.kernel.org \
--cc=norm@turing.une.edu.au \
--cc=wli@holomorphy.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