linuxppc-dev.lists.ozlabs.org archive mirror
 help / color / mirror / Atom feed
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Guillaume Autran <gautran@mrv.com>
Cc: linux-ppc-embedded <linuxppc-embedded@ozlabs.org>
Subject: Re: [PATCH] 8xx: get_mmu_context() for (very) FEW_CONTEXTS and KERNEL_PREEMPT race/starvation issue
Date: Wed, 29 Jun 2005 14:15:30 +1000	[thread overview]
Message-ID: <1120018530.5133.241.camel@gaston> (raw)
In-Reply-To: <42C153E1.3060004@mrv.com>

On Tue, 2005-06-28 at 09:42 -0400, Guillaume Autran wrote:
> Hi,
> 
> I happen to notice a race condition in the mmu_context code for the 8xx 
> with very few context (16 MMU contexts) and kernel preemption enable. It 
> is hard to reproduce has it shows only when many processes are 
> created/destroy and the system is doing a lot of IRQ processing.
> 
> In short, one process is trying to steal a context that is in the 
> process of being freed (mm->context == NO_CONTEXT) but not completely 
> freed (nr_free_contexts == 0).
> The steal_context() function does not do anything and the process stays 
> in the loop forever.
> 
> Anyway, I got a patch that fixes this part. Does not seem to affect 
> scheduling latency at all.
> 
> Comments are appreciated.

Your patch seems to do a hell lot more than fixing this race ... What
about just calling preempt_disable() in destroy_context() instead ?

Ben.

  reply	other threads:[~2005-06-29  4:20 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-06-25 14:53 [PATCH] 8xx: map_page() skip pinned region and tlbie debugging aid Marcelo Tosatti
2005-06-25 22:24 ` Dan Malek
2005-06-26 14:30   ` Marcelo Tosatti
2005-06-27 13:39     ` Marcelo Tosatti
2005-06-27 20:46       ` Dan Malek
2005-06-28  6:30         ` Benjamin Herrenschmidt
2005-06-28 13:42           ` [PATCH] 8xx: get_mmu_context() for (very) FEW_CONTEXTS and KERNEL_PREEMPT race/starvation issue Guillaume Autran
2005-06-29  4:15             ` Benjamin Herrenschmidt [this message]
2005-06-29 15:32               ` Guillaume Autran
2005-06-29 15:54                 ` Marcelo Tosatti
2005-06-29 21:25                   ` Guillaume Autran
2005-06-29 17:00                     ` Marcelo Tosatti
2005-06-29 23:26                   ` Benjamin Herrenschmidt
2005-06-29 19:38                     ` Marcelo Tosatti
2005-06-30 13:54                       ` Guillaume Autran
2005-07-05 13:12                         ` Guillaume Autran
2005-06-30  0:34                     ` Eugene Surovegin
2005-06-29 23:24                 ` Benjamin Herrenschmidt
2005-06-28 13:53           ` [PATCH] 8xx: map_page() skip pinned region and tlbie debugging aid Dan Malek
2005-06-28 23:47             ` Benjamin Herrenschmidt
2005-06-29 17:19             ` Marcelo Tosatti
2005-06-29 23:31               ` Benjamin Herrenschmidt
2005-06-30 18:05                 ` Dan Malek
2005-06-30 23:29                   ` Benjamin Herrenschmidt
2005-07-01  7:01                     ` Pantelis Antoniou
2005-06-30 17:49               ` Dan Malek
2005-06-27 14:28   ` [PATCH] 8xx: tlbie debugging aid (try #2) Marcelo Tosatti
2005-06-27 20:18     ` Dan Malek
2005-06-27 14:56       ` Marcelo Tosatti
2005-06-27 20:53         ` Dan Malek

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=1120018530.5133.241.camel@gaston \
    --to=benh@kernel.crashing.org \
    --cc=gautran@mrv.com \
    --cc=linuxppc-embedded@ozlabs.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;
as well as URLs for NNTP newsgroup(s).