From mboxrd@z Thu Jan  1 00:00:00 1970
From: Ingo Molnar <mingo@elte.hu>
Subject: Re: 2.6.20->2.6.21 - networking dies after random time
Date: Thu, 26 Jul 2007 10:16:09 +0200
Message-ID: <20070726081609.GA24377@elte.hu>
References: <JKE37W$0126317E099DF36884CC046B6F8A7AA9@xandmail.com> <20070629150759.GC2771@ff.dom.local> <4bacf17f0707222244p664e7a6ap850b3357a57d73c@mail.gmail.com> <20070724080534.GC18740@elte.hu> <20070724094202.GA11610@elte.hu> <alpine.LFD.0.999.0707241228291.3607@woody.linux-foundation.org> <20070724200431.GA22190@elte.hu> <1185322771.4175.102.camel@chaos> <4bacf17f0707260016x14fc1c92s628ae64353663833@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: QUOTED-PRINTABLE
Cc: Thomas Gleixner <tglx@linutronix.de>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Jarek Poplawski <jarkao2@o2.pl>,
	Jean-Baptiste Vignaud <vignaud@xandmail.fr>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	shemminger <shemminger@linux-foundation.org>,
	linux-net <linux-net@vger.kernel.org>,
	netdev <netdev@vger.kernel.org>,
	Andrew Morton <akpm@linux-foundation.org>
To: Marcin =?utf-8?Q?=C5=9Alusarz?= <marcin.slusarz@gmail.com>
Return-path: <netdev-owner@vger.kernel.org>
Received: from mx3.mail.elte.hu ([157.181.1.138]:60337 "EHLO mx3.mail.elte.hu"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1752574AbXGZIQW (ORCPT <rfc822;netdev@vger.kernel.org>);
	Thu, 26 Jul 2007 04:16:22 -0400
Content-Disposition: inline
In-Reply-To: <4bacf17f0707260016x14fc1c92s628ae64353663833@mail.gmail.com>
Sender: netdev-owner@vger.kernel.org
List-Id: netdev.vger.kernel.org


* Marcin =C5=9Alusarz <marcin.slusarz@gmail.com> wrote:

> 2007/7/25, Thomas Gleixner <tglx@linutronix.de>:
> >(...)
>=20
> I've tested Jarek's patch, 2 Ingo's patches (2nd and 3rd) and Thomas'=
=20
> patch (one patch at time of course) - all of them fixed the problem,=20
> but the last one flooded my logs with "Skip resend for irq 17". All=20
> tests were done on 2.6.21.3.

that's great! I think we have two good theories about what might be=20
going on:

 - the driver might be buggy in that it gets confused by the 'resent'=20
   irq.

 - or the chipset/cpu has a bug where it might get confused about the
   resent APIC vector getting mixed up with the same vector coming
   externally too. (Now, it makes little sense to 'resend' a
   level-triggered interrupt on x86 platforms that have flat PIC=20
   hierarchies (other architectures might need more than that to
   retrigger an interrupt) - but there's nothing wrong about it in=20
   theory and it needs fixing for edge irqs anyway.)

in any case, the problem was triggered by our change generating much=20
more resent irqs than before. Nevertheless we'd like to fix that resend=
=20
bug (and if the driver is buggy, the driver bug too). It's really good=20
progress so far - we are working on doing the real fix now.

	Ingo