linux-mips
[Top] [All Lists]

Re: [PATCH] Retry {save,restore}_fp_context if failed in atomic context.

To: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Subject: Re: [PATCH] Retry {save,restore}_fp_context if failed in atomic context.
From: Ralf Baechle <ralf@linux-mips.org>
Date: Wed, 18 Apr 2007 12:30:46 +0100
Cc: linux-mips@linux-mips.org
In-reply-to: <20070416.233235.75185596.anemo@mba.ocn.ne.jp>
Original-recipient: rfc822;linux-mips@linux-mips.org
References: <20070416.231944.41198415.anemo@mba.ocn.ne.jp> <20070416.233235.75185596.anemo@mba.ocn.ne.jp>
Sender: linux-mips-bounce@linux-mips.org
User-agent: Mutt/1.4.2.2i
On Mon, Apr 16, 2007 at 11:32:35PM +0900, Atsushi Nemoto wrote:

> > The save_fp_context()/restore_fp_context() might sleep on accessing
> > user stack and therefore might lose FPU ownership in middle of them.
> > 
> > If these function failed due to "in_atomic" test in do_page_fault,
> > touch the sigcontext area in non-atomic context and retry these
> > save/restore operation.
> > 
> > This is a replacement of a (broken) fix which was titled "Allow CpU
> > exception in kernel partially".
> > 
> > Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
> 
> And this is for 2.6.20-stable.

Both applied, also to older -stable branches except 2.6.16.  In case of
2.6.16 it would have been more time consuming than justifyable and since
the bug this patch fixes is comparable to what we had before starting the
whole surgery I have no problem to leave 2.6.16 as it is.  Anybody still
using 2.6.16 should upgrade anyway ...

  Ralf

<Prev in Thread] Current Thread [Next in Thread>