From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Michael S. Tsirkin" Subject: Re: [PATCHv3 08/14] Add get_fw_dev_path callback for pci bus. Date: Thu, 11 Nov 2010 17:05:11 +0200 Message-ID: <20101111150511.GF25692@redhat.com> References: <1289409261-5418-1-git-send-email-gleb@redhat.com> <1289409261-5418-9-git-send-email-gleb@redhat.com> <20101110173412.GE3891@redhat.com> <4CDBC045.3070902@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Gleb Natapov , qemu-devel@nongnu.org, kvm@vger.kernel.org, blauwirbel@gmail.com, armbru@redhat.com, alex.williamson@redhat.com To: Gerd Hoffmann Return-path: Received: from mx1.redhat.com ([209.132.183.28]:29051 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751233Ab0KKPFW (ORCPT ); Thu, 11 Nov 2010 10:05:22 -0500 Content-Disposition: inline In-Reply-To: <4CDBC045.3070902@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: On Thu, Nov 11, 2010 at 11:07:01AM +0100, Gerd Hoffmann wrote: > On 11/10/10 18:34, Michael S. Tsirkin wrote: > >On Wed, Nov 10, 2010 at 07:14:15PM +0200, Gleb Natapov wrote: > >> > >>Signed-off-by: Gleb Natapov > > > >Good stuff. We should also consider using this for > >CLI and monitor. Some comments below. > > Oh, we already have a table to map pci classes to descriptions for > 'info pci'. You remind me, that one must be fixed to use names from pci_ids.h > I'd strongly suggest to just add the fw names to that > table instead of creating a second one ... > > cheers, > Gerd No, I mean the path. We are currently passing domain:bus:slot.function to point at a device and we know it's broken for nested bridges and kind of broken for multiple domains. It's a bug that we must fix. Class names aren't that interesting to me, and I'd be just as happy with packing these into 2 tables as a single one personally, it all seems to be a matter of style. -- MST