linux-mips
[Top] [All Lists]

Re: preempt_schedule_irq missing from mfinfo[]?

To: djohnson+linuxmips@sw.starentnetworks.com
Subject: Re: preempt_schedule_irq missing from mfinfo[]?
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Sun, 03 Jul 2005 00:59:21 +0900 (JST)
Cc: linux-mips@linux-mips.org, ralf@linux-mips.org
In-reply-to: <17093.19241.353160.946039@cortez.sw.starentnetworks.com>
Original-recipient: rfc822;linux-mips@linux-mips.org
References: <17092.5345.75666.403044@cortez.sw.starentnetworks.com> <20050701.114358.21591461.nemoto@toshiba-tops.co.jp> <17093.19241.353160.946039@cortez.sw.starentnetworks.com>
Sender: linux-mips-bounce@linux-mips.org
>>>>> On Fri, 1 Jul 2005 09:54:49 -0400, Dave Johnson 
>>>>> <djohnson+linuxmips@sw.starentnetworks.com> said:

dave> That'll do it.  My patch wasn't enough.  I added some sanity
dave> checks to get_wchan and it hit one while running overnight.

dave> The task being examined transitioned from !TASK_RUNNING to
dave> TASK_RUNNING while it was being examined. Doh!

dave> Definately not SMP/preempt safe as written today.

Perhaps you can make it SMP/preempt safe by doing stack_page test in
the unwinding loop as done on i386, etc.

But anyway I think just calling thread_saved_pc is enough.

Ralf, how do you think about this?

---
Atsushi Nemoto

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