All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jeremy Fitzhardinge <jeremy@goop.org>
To: Ingo Molnar <mingo@elte.hu>
Cc: Andrew Morton <akpm@osdl.org>, Andi Kleen <ak@suse.de>,
	Laurent Riffard <laurent.riffard@free.fr>,
	Arjan van de Ven <arjan@infradead.org>,
	Kernel development list <linux-kernel@vger.kernel.org>,
	Jeremy Fitzhardinge <jeremy@xensource.com>
Subject: Re: 2.6.18-rc6-mm1: GPF loop on early boot
Date: Mon, 11 Sep 2006 01:13:25 -0700	[thread overview]
Message-ID: <45051AA5.6090102@goop.org> (raw)
In-Reply-To: <20060911080115.GB5328@elte.hu>

Ingo Molnar wrote:
> the ratio between the number of syscalls vs. the number of context 
> switches is 1-2 orders of magnitude. So a loss of 9 cycles in the 
> syscall path is roughly equal to a loss of 90-900 cycles in switch_to() 
> costs ...
>   

I agree.

> the TLS ABI is just a gcc stupidity. Why did they pick the _second_ 
> extra selector, instead of the first one ...?

Well, it doesn't matter which they chose if its the same for user and 
kernel space. Why does x86-64 have swapgs rather than swapfs?

>  Anyway, perhaps this could 
> be solved by extending gcc with a switch to also generate __thread code 
> off %fs. Probably not worth the pain though ...

If there's gcc hacking to be done, it would be trying to get it make the 
TLS offsets from the PDA/TCB positive rather than negative...

    J

  reply	other threads:[~2006-09-11  8:13 UTC|newest]

Thread overview: 103+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-09-08  8:13 2.6.18-rc6-mm1 Andrew Morton
2006-09-08 11:49 ` 2.6.18-rc6-mm1 Andy Whitcroft
2006-09-08 12:07 ` 2.6.18-rc6-mm1 Frederik Deweerdt
2006-09-08 12:16 ` [patch -mm] s390: fix save_stack_trace Heiko Carstens
2006-09-09 13:36   ` Andi Kleen
2006-09-08 12:23 ` 2.6.18-rc6-mm1 - x86_64-mm-lockdep-dont-force-framepointer.patch Heiko Carstens
2006-09-09 13:39   ` Andi Kleen
2006-09-11  8:45     ` Martin Schwidefsky
2006-09-08 14:26 ` 2.6.18-rc6-mm1 Rafael J. Wysocki
2006-09-08 20:44   ` [linux-usb-devel] 2.6.18-rc6-mm1 Alan Stern
2006-09-08 22:57     ` Rafael J. Wysocki
2006-09-11 22:08       ` Rafael J. Wysocki
2006-09-12 14:28         ` Alan Stern
2006-09-12 17:22           ` Mattia Dongili
2006-09-12 18:04             ` Mattia Dongili
2006-09-12 20:10               ` Alan Stern
2006-09-13 17:00                 ` Rafael J. Wysocki
2006-09-13 12:07       ` [linux-usb-devel] 2.6.18-rc6-mm1 (-mm2): ohci resume problem Rafael J. Wysocki
2006-09-13 12:42         ` Rafael J. Wysocki
2006-09-13 18:38           ` Alan Stern
2006-09-13 20:00             ` Rafael J. Wysocki
2006-09-13 21:01               ` Alan Stern
2006-09-13 21:32                 ` Rafael J. Wysocki
2006-09-13 21:55                   ` Alan Stern
2006-09-14 13:14                     ` Rafael J. Wysocki
2006-09-14 14:08                       ` Rafael J. Wysocki
2006-09-14 15:04                         ` Alan Stern
2006-09-14 16:17                           ` Alan Stern
2006-09-14 17:08                             ` Rafael J. Wysocki
2006-09-14 17:13                               ` Rafael J. Wysocki
2006-09-14 17:24                                 ` Alan Stern
2006-09-14 17:22                               ` Alan Stern
2006-09-14 17:35                                 ` Rafael J. Wysocki
2006-09-14 18:28                                   ` Alan Stern
     [not found]                                     ` <200609142137.52066.rjw@sisk.pl>
2006-09-14 20:21                                       ` Rafael J. Wysocki
2006-09-14 20:55                                       ` Alan Stern
2006-09-14 21:47                                         ` Rafael J. Wysocki
2006-09-14 22:19                                           ` Alan Stern
2006-09-14 16:48                           ` Rafael J. Wysocki
2006-09-13 20:38             ` Mattia Dongili
2006-09-13 20:54               ` Alan Stern
2006-09-14 20:19                 ` Mattia Dongili
2006-09-14 20:25                   ` Alan Stern
2006-09-14 20:35                     ` Mattia Dongili
2006-09-16 11:58                     ` Mattia Dongili
2006-09-16 14:31                       ` Alan Stern
2006-09-08 17:43 ` 2.6.18-rc6-mm1 Stefan Richter
2006-09-08 18:04   ` 2.6.18-rc6-mm1 Andrew Morton
2006-09-08 18:36     ` 2.6.18-rc6-mm1 Stefan Richter
2006-09-08 19:23 ` 2.6.18-rc6-mm1 Michal Piotrowski
2006-09-08 19:43   ` 2.6.18-rc6-mm1 Andrew Morton
2006-09-08 20:01     ` 2.6.18-rc6-mm1 Michal Piotrowski
2006-09-08 19:30 ` 2.6.18-rc6-mm1 thunder7
2006-09-08 19:44   ` 2.6.18-rc6-mm1 Andrew Morton
2006-09-09  9:04     ` 2.6.18-rc6-mm1 thunder7
2006-09-09 15:31       ` 2.6.18-rc6-mm1 Andrew Morton
2006-09-09 15:31         ` 2.6.18-rc6-mm1 Andrew Morton
2006-09-09 22:02         ` 2.6.18-rc6-mm1 Jean Delvare
2006-09-10  6:30           ` 2.6.18-rc6-mm1 thunder7
2006-09-10  6:30             ` 2.6.18-rc6-mm1 thunder7
2006-09-09  8:35 ` lockdep warning in check_flags() Frederik Deweerdt
2006-09-11  5:43   ` Ingo Molnar
2006-09-12 14:13     ` Frederik Deweerdt
2006-09-12 16:54       ` Ingo Molnar
2006-09-12 20:21         ` Frederik Deweerdt
     [not found] ` <4503DC64.9070007@free.fr>
2006-09-10  8:32   ` 2.6.18-rc6-mm1: GPF loop on early boot Andi Kleen
2006-09-10 10:29     ` Arjan van de Ven
2006-09-10 11:57     ` Ingo Molnar
2006-09-10 11:34       ` Andi Kleen
2006-09-10 13:26         ` Ingo Molnar
2006-09-10 13:55           ` Andi Kleen
2006-09-10 14:02             ` Ingo Molnar
2006-09-10 16:33           ` Andrew Morton
2006-09-10 23:03             ` Jeremy Fitzhardinge
2006-09-11  5:10               ` Ingo Molnar
2006-09-11  7:31                 ` Jeremy Fitzhardinge
2006-09-11  7:29                   ` Ingo Molnar
2006-09-11  7:41                     ` Jeremy Fitzhardinge
2006-09-11  7:36                       ` Ingo Molnar
2006-09-11  7:59                         ` Jeremy Fitzhardinge
2006-09-11  8:01                           ` Ingo Molnar
2006-09-11  8:13                             ` Jeremy Fitzhardinge [this message]
2006-09-11  7:38                       ` Ingo Molnar
2006-09-11  7:56                         ` Jeremy Fitzhardinge
2006-09-11  7:55                           ` Ingo Molnar
2006-09-11  5:21               ` Laurent Riffard
2006-09-11  5:18                 ` Ingo Molnar
2006-09-11  5:25               ` [patch] i386-PDA, lockdep: fix %gs restore Ingo Molnar
2006-09-11  5:41                 ` Andi Kleen
2006-09-11  5:48                   ` Ingo Molnar
2006-09-11  5:46                 ` Ingo Molnar
2006-09-11 16:35                   ` Laurent Riffard
2006-09-11  7:42                 ` Jeremy Fitzhardinge
2006-09-11 19:33                 ` Jeremy Fitzhardinge
2006-09-11 21:25                   ` Jeremy Fitzhardinge
2006-09-11 20:20                 ` Andi Kleen
2006-09-11 21:37                   ` Jeremy Fitzhardinge
2006-09-11 20:48                     ` Andi Kleen
2006-09-10 12:36       ` 2.6.18-rc6-mm1: GPF loop on early boot Laurent Riffard
2006-09-10 13:07         ` Ingo Molnar
2006-09-10 13:11       ` Ingo Molnar
2006-09-11 21:19 ` 2.6.18-rc6-mm1 Mark Haverkamp
2006-09-11 22:16   ` 2.6.18-rc6-mm1 Andrew Morton

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=45051AA5.6090102@goop.org \
    --to=jeremy@goop.org \
    --cc=ak@suse.de \
    --cc=akpm@osdl.org \
    --cc=arjan@infradead.org \
    --cc=jeremy@xensource.com \
    --cc=laurent.riffard@free.fr \
    --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 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.