From: Alan Cox <alan@lxorguk.ukuu.org.uk>
To: george@mvista.com (george anzinger)
Cc: alan@lxorguk.ukuu.org.uk (Alan Cox), ak@suse.de (Andi Kleen),
linux-kernel@vger.kernel.org
Subject: Re: Why HZ on i386 is 100 ?
Date: Mon, 29 Apr 2002 01:14:26 +0100 (BST) [thread overview]
Message-ID: <E171yoE-0004ym-00@the-village.bc.nu> (raw)
In-Reply-To: <3CCC6EAD.22A439F7@mvista.com> from "george anzinger" at Apr 28, 2002 02:50:37 PM
> The problem is the extra code in the schedule() path, not in the timer
> tick path. It is traversed FAR more often.
Thats still in most cases a single compare. The tick timer will mostly be
going off before our time slice completes. Also importantly the more we
context switch the less timers go off - so it scales correctly.
> The current tick at 1/HZ is really quite relaxed. Given the PIT (ugh!)
> the longest we can put off a tick is about 50 ms. This means that any
> time greater than this will require more than one interrupt, i.e. the
> best case improvement by going tick less (again given the PIT) is about
> 5 times. Other platforms/ hardware, of course, change this.
If you are arguing that the PIT makes it impractical on basic x86 then
we are in violent agreement. I don't propose this kind of stuff for the
PIT but for real computers where a timer reload is a couple of clocks
next prev parent reply other threads:[~2002-04-28 23:55 UTC|newest]
Thread overview: 68+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <3CC4861C.F21859A6@mvista.com.suse.lists.linux.kernel>
[not found] ` <E16zuPf-0007yD-00@the-village.bc.nu.suse.lists.linux.kernel>
2002-04-23 7:17 ` Andi Kleen
2002-04-23 19:09 ` george anzinger
2002-04-24 1:42 ` Alan Cox
2002-04-24 20:20 ` george anzinger
2002-04-27 20:26 ` Alan Cox
2002-04-28 6:02 ` george anzinger
2002-04-28 9:12 ` Alan Cox
2002-04-28 17:34 ` george anzinger
2002-04-28 18:59 ` Alan Cox
2002-04-28 21:50 ` george anzinger
2002-04-29 0:14 ` Alan Cox [this message]
2002-04-23 19:24 ` george anzinger
2002-04-23 19:35 ` Andi Kleen
2002-04-24 17:25 ` Maciej W. Rozycki
2002-04-17 0:33 Chen, Kenneth W
2002-04-17 1:02 ` Davide Libenzi
-- strict thread matches above, loose matches on Subject: below --
2002-04-16 10:41 Cabaniols, Sebastien
2002-04-16 7:47 Olaf Fraczyk
2002-04-16 8:14 ` William Lee Irwin III
2002-04-16 8:18 ` BALBIR SINGH
2002-04-16 10:29 ` Liam Girdwood
2002-04-16 10:01 ` Olaf Fraczyk
2002-04-16 13:35 ` Terje Eggestad
2002-04-16 13:38 ` Mark Mielke
2002-04-16 13:55 ` Terje Eggestad
2002-04-16 15:32 ` Rik van Riel
2002-04-16 16:12 ` Chris Friesen
2002-04-16 17:12 ` Mark Mielke
2002-04-16 13:58 ` Alan Cox
2002-04-17 0:22 ` H. Peter Anvin
2002-04-16 16:27 ` Linus Torvalds
2002-04-16 16:50 ` David Mosberger
2002-04-16 17:18 ` Davide Libenzi
2002-04-16 17:52 ` David Mosberger
2002-04-16 18:10 ` Davide Libenzi
2002-04-17 0:49 ` David Mosberger
2002-04-17 0:57 ` Robert Love
2002-04-17 1:07 ` Davide Libenzi
2002-04-17 5:18 ` Mark Mielke
2002-04-17 5:34 ` Linus Torvalds
2002-04-17 6:01 ` Robert Love
2002-04-17 6:17 ` David Mosberger
2002-04-17 7:59 ` arjan
2002-04-17 8:04 ` Matti Aarnio
2002-04-23 22:42 ` Albert D. Cahalan
2002-04-17 10:12 ` Martin Dalecki
2002-04-18 1:51 ` Dan Mann
2002-04-17 1:22 ` Davide Libenzi
2002-04-17 3:19 ` Ben Greear
2002-04-17 7:55 ` Helge Hafting
2002-04-21 18:00 ` Pavel Machek
2002-04-22 17:20 ` John Alvord
2002-04-22 21:52 ` george anzinger
2002-04-22 23:06 ` J.D. Bakker
2002-04-22 23:26 ` Anton Blanchard
2002-04-23 19:03 ` george anzinger
2002-04-23 7:08 ` Alan Cox
2002-04-22 17:24 ` David Mosberger
2002-04-16 12:42 ` Richard B. Johnson
2002-04-16 12:31 ` Richard B. Johnson
2002-04-16 14:04 ` Bill Davidsen
2002-04-16 21:34 ` bert hubert
2002-04-16 22:21 ` Andreas Dilger
2002-04-16 22:37 ` Herbert Xu
2002-04-16 22:56 ` Andreas Dilger
2002-04-17 0:34 ` J. Dow
2002-04-17 2:40 ` Herbert Xu
2002-04-17 12:44 ` Kent Borg
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=E171yoE-0004ym-00@the-village.bc.nu \
--to=alan@lxorguk.ukuu.org.uk \
--cc=ak@suse.de \
--cc=george@mvista.com \
--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
all inboxes | Powered by JetHome®