From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755847AbZB1U52 (ORCPT ); Sat, 28 Feb 2009 15:57:28 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753675AbZB1U5T (ORCPT ); Sat, 28 Feb 2009 15:57:19 -0500 Received: from rv-out-0506.google.com ([209.85.198.224]:14652 "EHLO rv-out-0506.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753484AbZB1U5R (ORCPT ); Sat, 28 Feb 2009 15:57:17 -0500 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type :content-transfer-encoding; b=fSbvvjdcEFGOLXV0aZhJW/P3Z6ri00EBA022+SlujileQ9mrkSW8Vm6YFvhP0OrVfd MlqXKrAYUDHKCPZW5x6h/Sxjnhfxe3mQV2qjW3MfBBcsV5umvNn+dr5hkiu0UALd++wW b2ZOTq8TNcVoSSvPkadu0jXbq8s8t56IOs4EA= MIME-Version: 1.0 In-Reply-To: <200902280925.56450.prakash@punnoor.de> References: <200902270752.44078.prakash@punnoor.de> <49A8541C.4030601@kernel.org> <200902280925.56450.prakash@punnoor.de> Date: Sat, 28 Feb 2009 12:57:15 -0800 X-Google-Sender-Auth: ba2fdb6b5c209e68 Message-ID: <86802c440902281257r4b7f6ca3w546310581a3fc6b5@mail.gmail.com> Subject: Re: [PATCH] pci: don't enable too many HT MSI mapping From: Yinghai Lu To: Prakash Punnoor Cc: Jesse Barnes , Andrew Morton , Ingo Molnar , "Eric W. Biederman" , Robert Hancock , david@lang.hm, Matthew Wilcox , linux-kernel , linux-pci@vger.kernel.org Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sat, Feb 28, 2009 at 12:25 AM, Prakash Punnoor wrote: > On Freitag 27 Februar 2009 21:59:08 Yinghai Lu wrote: >> Prakash Punnoor wrote: >> > On Dienstag 24 Februar 2009 18:37:35 Jesse Barnes wrote: >> >> On Monday, February 23, 2009 11:51:59 am Yinghai Lu wrote: >> >>> Impact: fix bug >> >>> >> >>> Prakash reported that his c51-mcp51 system ondie sound card doesn't >> >>> work MSI but if he hack out the HT-MSI on mcp51, the MSI will work well >> >>> with sound card. >> >>> >> >>> this patch rework the nv_msi_ht_cap_quirk() >> >>> and will only enable ht_msi on own root device and try to avoid to >> >>> enable ht_msi on device following that root dev >> >>> >> >>> Reported-by: Prakash Punnoor >> >>> Tested-by: Prakash Punnoor >> >>> Signed-off-by: Yinghai Lu >> >> >> >> Thanks Yinghai & Prakash, applied this fix to my for-linus branch. >> > >> > I am very sorry, but I made a mistake testing this patch. I messed up my >> > kernels and got a false positive and only noticed this now. The patch >> > does NOT work and even makes things worse: >> > >> > dmesg|grep HT >> > pci 0000:00:09.0: Found disabled HT MSI Mapping >> > pci 0000:00:09.0: Enabling HT MSI Mapping >> > pci 0000:00:09.0: Found enabled HT MSI Mapping >> > pci 0000:00:09.0: Found enabled HT MSI Mapping >> > pci 0000:00:09.0: Found enabled HT MSI Mapping >> > pci 0000:00:09.0: Found enabled HT MSI Mapping >> > >> > This is exactly the device which shouldn't be MSI enabled for me. On the >> > other hand, it doesn't enable the needed devices anymore. >> >> please check this one, please power off the system before load the kernel >> with new patch. > > Well, the problem now is that MSI doesn't get enabled on the needed host > bridge 00.0 (device 10de:02f0), so MSI doesn't work at all on my system (but > also doesn't hang my system.) I have no idea how to differentiate the 00.0 > device from the "forbidden" 09.0 (device 10de:0270) on my system. Perhaps a > blacklist is really the best? The original patch by NVidia had a white list, > btw, see also here: > http://kerneltrap.org/mailarchive/linux-kernel/2007/11/25/443612/thread > > The output of your current patch: > pci 0000:00:05.0: Boot video device > pci 0000:00:09.0: Found disabled HT MSI Mapping > pci 0000:00:0e.0: Enabling HT MSI Mapping > pci 0000:00:09.0: Found disabled HT MSI Mapping > pci 0000:00:0f.0: Enabling HT MSI Mapping > pci 0000:00:09.0: Found disabled HT MSI Mapping > pci 0000:00:10.0: Enabling HT MSI Mapping > pci 0000:00:09.0: Found disabled HT MSI Mapping > pci 0000:00:10.1: Enabling HT MSI Mapping > > So, ok, MSI gets enabled on the devices as such (good), not enabled on 09.0 > (good), but also not on the needed host bridge 00.0 (bad). can you post whole bootlog and lspci -vvxxx and lspci -tv and cat /proc/interrupts? YH