From: Brian Gerst <bgerst@didntduck.org>
To: Andrea Arcangeli <andrea@suse.de>
Cc: "Ingo Molnar" <mingo@elte.hu>,
"Jörn Engel" <joern@wohnheim.fh-wedel.de>,
"Arjan van de Ven" <arjanv@redhat.com>,
"Rik van Riel" <riel@redhat.com>,
"Linus Torvalds" <torvalds@osdl.org>,
linux-kernel@vger.kernel.org
Subject: Re: 4k stacks in 2.6
Date: Thu, 27 May 2004 10:18:40 -0400 [thread overview]
Message-ID: <40B5F8C0.2010005@didntduck.org> (raw)
In-Reply-To: <20040527135930.GC3889@dualathlon.random>
Andrea Arcangeli wrote:
> On Thu, May 27, 2004 at 02:45:51PM +0200, Ingo Molnar wrote:
>
>>are a bit belated. I only reacted to Andrea's mail to clear up apparent
>>misunderstandings about the impact and implementation of this feature.
>
>
> note that there is something relevant to improve in the implementation,
> that is the per-cpu irq stack size should be bigger than 4k, we use 16k
> on x86-64, on x86 it should be 8k. Currently you're decreasing _both_
> the normal kernel context and even the irq stack in some condition.
> There's no good reason to decrease the irq stack too, that's cheap, it's
> per-cpu.
The problem on i386 (unlike x86-64) is that the thread_info struct sits
at the bottom of the stack and is referenced by masking bits off %esp.
So the stack size must be constant whether in process context or IRQ
context.
--
Brian Gerst
next prev parent reply other threads:[~2004-05-27 14:18 UTC|newest]
Thread overview: 106+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-05-23 19:43 4g/4g for 2.6.6 Phy Prabab
2004-05-23 20:32 ` Linus Torvalds
2004-05-23 20:51 ` Jeff Garzik
2004-05-24 1:55 ` Linus Torvalds
2004-05-24 2:19 ` Jeff Garzik
2004-05-24 2:33 ` Wim Coekaerts
2004-05-31 9:51 ` Pavel Machek
2004-05-24 3:30 ` Martin J. Bligh
2004-06-01 5:52 ` Eric W. Biederman
2004-05-23 21:55 ` Phy Prabab
2004-05-24 7:05 ` Arjan van de Ven
2004-05-24 7:11 ` Phy Prabab
2004-05-24 7:24 ` Arjan van de Ven
2004-05-24 7:27 ` Phy Prabab
2004-05-24 12:01 ` Dave Jones
2004-05-24 2:39 ` William Lee Irwin III
2004-05-24 8:25 ` Ingo Molnar
2004-05-24 12:48 ` Andrea Arcangeli
2004-05-25 19:15 ` Rik van Riel
2004-05-25 19:41 ` Andrea Arcangeli
2004-05-25 19:50 ` Rik van Riel
2004-05-25 20:10 ` Rik van Riel
2004-05-25 21:15 ` Andrea Arcangeli
2004-05-26 10:33 ` 4k stacks in 2.6 Ingo Molnar
2004-05-26 12:50 ` Jörn Engel
2004-05-26 12:53 ` Arjan van de Ven
2004-05-26 13:00 ` Jörn Engel
2004-05-26 13:05 ` Arjan van de Ven
2004-05-26 16:41 ` Jörn Engel
2004-05-27 12:45 ` Ingo Molnar
2004-05-27 13:59 ` Andrea Arcangeli
2004-05-27 14:03 ` Arjan van de Ven
2004-05-27 14:42 ` Andrea Arcangeli
2004-06-02 19:40 ` Bill Davidsen
2004-05-27 14:18 ` Brian Gerst [this message]
2004-05-27 14:50 ` Andrea Arcangeli
2004-05-27 14:55 ` Linus Torvalds
2004-05-27 15:39 ` Andrea Arcangeli
2004-05-27 18:31 ` Guy Sotomayor
2004-05-27 19:26 ` Brian Gerst
2004-06-01 5:56 ` 4k stacks in 2.6 [worst offenders] Jörn Engel
2004-06-01 6:02 ` [RFC PATCH] explicitly mark recursion count Jörn Engel
2004-06-01 12:20 ` Pavel Machek
2004-06-01 13:27 ` Jörn Engel
2004-06-01 13:32 ` Pavel Machek
2004-06-01 13:37 ` Jörn Engel
2004-06-01 19:48 ` Horst von Brand
2004-06-01 19:29 ` Horst von Brand
2004-06-01 19:58 ` Linus Torvalds
2004-06-02 13:16 ` Jörn Engel
2004-06-02 14:15 ` Linus Torvalds
2004-06-02 14:27 ` Jörn Engel
2004-06-02 14:45 ` Linus Torvalds
2004-06-02 15:04 ` Jörn Engel
2004-06-02 15:12 ` Linus Torvalds
2004-06-02 15:27 ` Jörn Engel
2004-06-02 15:52 ` Linus Torvalds
2004-06-02 16:17 ` Jörn Engel
2004-06-02 16:25 ` Linus Torvalds
2004-06-02 17:17 ` Jörn Engel
2004-06-02 17:32 ` Greg KH
2004-06-02 17:46 ` Jörn Engel
2004-06-02 14:35 ` Davide Libenzi
2004-06-02 18:20 ` Jörn Engel
2004-06-02 18:37 ` Davide Libenzi
2004-06-02 18:58 ` Jörn Engel
2004-06-02 19:33 ` Valdis.Kletnieks
2004-06-02 19:37 ` viro
2004-06-02 19:45 ` Jörn Engel
2004-06-02 19:59 ` viro
2004-06-03 6:55 ` Jörn Engel
2004-06-02 19:55 ` Valdis.Kletnieks
2004-06-02 23:20 ` Davide Libenzi
2004-06-03 7:29 ` Jörn Engel
2004-06-01 12:39 ` viro
2004-06-01 13:26 ` Jörn Engel
2004-06-07 18:14 ` 4k stacks in 2.6 Timothy Miller
2004-06-08 6:26 ` Arjan van de Ven
2004-06-08 8:45 ` Jörn Engel
2004-05-26 18:12 ` David S. Miller
2004-05-26 19:02 ` Matt Mackall
2004-05-26 19:25 ` Dave Jones
2004-05-25 21:16 ` 4g/4g for 2.6.6 Andrew Morton
2004-05-25 21:48 ` Ingo Molnar
2004-05-25 22:09 ` David S. Miller
2004-05-25 22:20 ` Ingo Molnar
2004-05-25 23:10 ` David S. Miller
2004-05-25 21:04 ` Bill Davidsen
2004-05-24 1:33 ` Martin J. Bligh
2004-05-24 1:38 ` Phy Prabab
[not found] <1ZQpn-1Rx-1@gated-at.bofh.it>
[not found] ` <1ZQz8-1Yh-15@gated-at.bofh.it>
[not found] ` <1ZRFf-2Vt-3@gated-at.bofh.it>
[not found] ` <203Zu-4aT-15@gated-at.bofh.it>
2004-05-26 13:57 ` 4k stacks in 2.6 Andi Kleen
2004-05-26 18:17 ` hch
2004-05-26 18:24 ` Andi Kleen
2004-05-26 20:39 ` Zwane Mwaikambo
[not found] ` <206b3-5WN-33@gated-at.bofh.it>
[not found] ` <20baw-1Lz-15@gated-at.bofh.it>
2004-05-26 19:32 ` Andi Kleen
2004-05-27 11:27 ` Jörn Engel
2004-05-27 13:49 ` Andrea Arcangeli
2004-05-27 14:15 ` Jörn Engel
2004-05-27 14:49 ` Andrea Arcangeli
2004-05-27 14:59 ` Jörn Engel
2004-05-27 15:08 ` Keith Owens
2004-05-27 15:21 ` Jörn Engel
2004-05-27 15:34 ` Arjan van de Ven
2004-05-27 15:46 ` Jörn Engel
2004-06-01 5:25 ` Jörn Engel
-- strict thread matches above, loose matches on Subject: below --
2004-05-26 15:17 Albert Cahalan
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=40B5F8C0.2010005@didntduck.org \
--to=bgerst@didntduck.org \
--cc=andrea@suse.de \
--cc=arjanv@redhat.com \
--cc=joern@wohnheim.fh-wedel.de \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=riel@redhat.com \
--cc=torvalds@osdl.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.