From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S267684AbUG3PBe (ORCPT ); Fri, 30 Jul 2004 11:01:34 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S267701AbUG3PBd (ORCPT ); Fri, 30 Jul 2004 11:01:33 -0400 Received: from dbl.q-ag.de ([213.172.117.3]:43753 "EHLO dbl.q-ag.de") by vger.kernel.org with ESMTP id S267684AbUG3PB1 (ORCPT ); Fri, 30 Jul 2004 11:01:27 -0400 Message-ID: <410A631B.3020600@colorfullife.com> Date: Fri, 30 Jul 2004 17:02:51 +0200 From: Manfred Spraul User-Agent: Mozilla/5.0 (X11; U; Linux i686; fr-FR; rv:1.6) Gecko/20040510 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Andrew Morton CC: linux-kernel@vger.kernel.org, alan@lxorguk.ukuu.org.uk Subject: Re: [Fastboot] Re: Announce: dumpfs v0.01 - common RAS output API Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Andrew wrote: >btw, if we simply insert a five-second-pause, what problems does that >leave? Network Rx, which is OK. Disk writes will have completed (?). >What remains? > > > I'd disagree: as soon as the IOMMU is reconfigured/reused Network Rx could become a problem: it could point to new io areas of the kexec'ed kernel. Btw, what's the preferred approach to clear the pci master bit: forcedeth writes to freed buffers after ifdown right now. I'll add a reset into the _close function, but disabling the master bit is probably better. -- Manfred