All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ingo van Lil <inguin@gmx.de>
To: Jeff Dike <jdike@addtoit.com>
Cc: user-mode-linux-devel@lists.sourceforge.net
Subject: Re: [uml-devel] [Patch] Save FPU registers between task switches
Date: Thu, 23 Apr 2009 18:38:58 +0200	[thread overview]
Message-ID: <49F099A2.5070408@gmx.de> (raw)
In-Reply-To: <20090423160659.GD9820@c2.user-mode-linux.org>

Jeff Dike wrote:

>> some time ago Jeff prepared a patch [1] for UML to stop saving the
>> process FP state between task switches. The assumption was that since
>> with SKAS0 every guest process runs inside a host process context the
>> host OS will take care of keeping the proper FP state. Unfortunately
>> this is not true for multi-threaded applications, where all guest
>> threads share a single host process context yet all may use the FPU on
>> their own. Although I haven't verified it I suspect things to be even
>> worse in SKAS3 mode where all guest processes run inside a single host
>> process.
> 
> This seems plausible.  Does reverting the patch fix any process crashes?

Hi Jeff,

the tiny test program I posted a few weeks ago [1] reproducibly crashes 
after a few seconds without the patch. Additionally I've been running 
UML with the patch for about a week now and I haven't seen the 
occasional erratic results that lead me to investigate the issue since then.

I've never been using the SKAS3 patch, so I cannot comment on that. I 
had hoped that Stanislav Meduna could try whether the patch fixes the 
problems he reported last September [2].

Regards,
Ingo

[1] http://marc.info/?l=user-mode-linux-devel&m=123912656515459&w=2
[2] http://marc.info/?l=user-mode-linux-devel&m=121819897821728&w=2

------------------------------------------------------------------------------
Stay on top of everything new and different, both inside and 
around Java (TM) technology - register by April 22, and save
$200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
300 plus technical and hands-on sessions. Register today. 
Use priority code J9JMT32. http://p.sf.net/sfu/p
_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel


      reply	other threads:[~2009-04-23 16:39 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-04-18 15:40 [uml-devel] [Patch] Save FPU registers between task switches Ingo van Lil
2009-04-23 16:06 ` Jeff Dike
2009-04-23 16:38   ` Ingo van Lil [this message]

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=49F099A2.5070408@gmx.de \
    --to=inguin@gmx.de \
    --cc=jdike@addtoit.com \
    --cc=user-mode-linux-devel@lists.sourceforge.net \
    /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.