public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: "H. Peter Anvin" <hpa@zytor.com>
To: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>,
	Gerhard Brauer <gerhard.brauer@web.de>,
	Ingo Molnar <mingo@elte.hu>,
	linux-kernel@vger.kernel.org
Subject: Re: 2.6.{26.2,27-rc} oops on virtualbox
Date: Tue, 26 Aug 2008 12:52:54 -0700	[thread overview]
Message-ID: <48B45F16.7080203@zytor.com> (raw)
In-Reply-To: <20080826181558.GA16887@Krystal>

One thing that I think really needs to be considered is that the current 
PV stubs are (a) large, and (b) non-atomic.

In the case at hand we have:

c012fc69:       51                      push   %ecx
c012fc6a:       52                      push   %edx
c012fc6b:       ff 15 40 b9 41 c0       call   *0xc041b940
c012fc71:       5a                      pop    %edx
c012fc72:       59                      pop    %ecx

Ten bytes replacing a two-byte native sequence.

If this was done as a call to an out-of-line stub, it would be only five 
bytes, which would reduce native icache overhead from 400% to 150%, but 
perhaps more importantly, it would not be subject to returns inside the 
sequence itself (since the out-of-line stub would still exist.)  As an 
optional bonus, at least on 32 bits the indirect call could be replaced 
with a direct call in the out-of-line stub.

	-hpa

  reply	other threads:[~2008-08-26 19:53 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-08-20 19:29 2.6.{26.2,27-rc} oops on virtualbox Luiz Fernando N. Capitulino
2008-08-21 21:34 ` H. Peter Anvin
2008-08-22  6:42   ` H. Peter Anvin
2008-08-22  6:50     ` Ingo Molnar
2008-08-22 14:39       ` Luiz Fernando N. Capitulino
2008-08-22 15:34         ` Mathieu Desnoyers
2008-08-22 16:29           ` Luiz Fernando N. Capitulino
2008-08-22 16:35             ` Mathieu Desnoyers
2008-08-22 17:20               ` Luiz Fernando N. Capitulino
2008-08-22 18:11                 ` H. Peter Anvin
2008-08-22 19:40                   ` Luiz Fernando N. Capitulino
2008-08-22 20:31                     ` H. Peter Anvin
2008-08-22 20:55                       ` Luiz Fernando N. Capitulino
2008-08-22 20:57                 ` Luiz Fernando N. Capitulino
2008-08-22 21:08                   ` H. Peter Anvin
2008-08-26 14:18                     ` Gerhard Brauer
2008-08-26 14:53                       ` Mathieu Desnoyers
2008-08-26 16:09                         ` Luiz Fernando N. Capitulino
2008-08-26 16:13                         ` Luiz Fernando N. Capitulino
2008-08-26 17:18                           ` Mathieu Desnoyers
2008-08-26 17:32                             ` H. Peter Anvin
2008-08-26 18:02                             ` Luiz Fernando N. Capitulino
2008-08-26 18:15                               ` Mathieu Desnoyers
2008-08-26 19:52                                 ` H. Peter Anvin [this message]
2008-08-26 20:34                                 ` Gerhard Brauer
2008-08-26 20:48                                   ` Mathieu Desnoyers
2008-08-26 21:25                                     ` Gerhard Brauer
2008-08-26 21:35                                       ` Mathieu Desnoyers
2008-08-26 21:51                                         ` H. Peter Anvin
2008-08-27  0:13                                         ` Gerhard Brauer
2008-08-27 19:13                                   ` Luiz Fernando N. Capitulino
2008-08-27 23:33                                     ` Mathieu Desnoyers
2008-08-28 13:30                                       ` Luiz Fernando N. Capitulino
2008-08-31  9:29                                         ` Gerhard Brauer
2008-08-31 13:28                                           ` Stefan Lippers-Hollmann
2008-08-31 14:03                                             ` Gerhard Brauer
2008-08-31 14:09                                           ` Luiz Fernando N. Capitulino
2008-09-21 13:41                                             ` Gerhard Brauer
2008-09-22  9:51                                               ` Ingo Molnar
2008-09-24 13:24                                               ` Luiz Fernando N. Capitulino
2008-08-28 13:50                                       ` Gerhard Brauer
2008-08-26 19:27                         ` Gerhard Brauer
2008-08-26 16:02                       ` Luiz Fernando N. Capitulino
2008-08-26 16:40                         ` Gerhard Brauer
2008-08-22 17:16             ` H. Peter Anvin
2008-08-22 17:45               ` Mathieu Desnoyers
2008-08-22 17:57                 ` H. Peter Anvin
2008-08-22 19:10               ` Luiz Fernando N. Capitulino
2008-08-22 19:14                 ` H. Peter Anvin
2008-08-22 19:18                 ` H. Peter Anvin
2008-08-22 19:42                   ` Luiz Fernando N. Capitulino
2008-08-22 14:28   ` Luiz Fernando N. Capitulino

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=48B45F16.7080203@zytor.com \
    --to=hpa@zytor.com \
    --cc=gerhard.brauer@web.de \
    --cc=lcapitulino@mandriva.com.br \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mathieu.desnoyers@polymtl.ca \
    --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