From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755330Ab0CDKSw (ORCPT ); Thu, 4 Mar 2010 05:18:52 -0500 Received: from smtp.citrix.com ([66.165.176.89]:44001 "EHLO SMTP.CITRIX.COM" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752480Ab0CDKSt (ORCPT ); Thu, 4 Mar 2010 05:18:49 -0500 X-IronPort-AV: E=Sophos;i="4.49,580,1262581200"; d="scan'208";a="7438093" Subject: Re: [Xen-devel] [PATCH 3/7] xen/hvm: Xen PV extension of HVM initialization From: Ian Campbell To: Stefano Stabellini CC: Jeremy Fitzhardinge , Sheng Yang , Keir Fraser , Jeremy Fitzhardinge , Ian Pratt , xen-devel , "Yaozu (Eddie) Dong" , "linux-kernel@vger.kernel.org" In-Reply-To: References: <1267436315-24486-1-git-send-email-sheng@linux.intel.com> <1267436315-24486-4-git-send-email-sheng@linux.intel.com> <4B8C63B0.2090507@goop.org> <201003020938.53630.sheng@linux.intel.com> <1267521735.11737.23107.camel@zakaz.uk.xensource.com> <4B8D7248.5070807@goop.org> <4B8E9DD7.6020408@goop.org> Content-Type: text/plain; charset="UTF-8" Organization: Citrix Systems, Inc. Date: Thu, 4 Mar 2010 10:18:45 +0000 Message-ID: <1267697925.11737.32107.camel@zakaz.uk.xensource.com> MIME-Version: 1.0 X-Mailer: Evolution 2.28.2 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 2010-03-03 at 17:41 +0000, Stefano Stabellini wrote: > On Wed, 3 Mar 2010, Jeremy Fitzhardinge wrote: > > On 03/03/2010 03:35 AM, Stefano Stabellini wrote: > > > Give a look at the fourth patch of the series I just posted: it > > > introduces a simple driver for the Xen PCI platform device to initialize > > > xenbus and gran table later on when running in a HVM domain, at the same > > > time leaving the PV case as it is. > > > > > > > Yes, I noticed that. Do the PV drivers cope with having the PCI > > probe/xenbus setup after their init functions have been called, or do > > you make sure the PCI probe happens early? > > > > Nope, they seem to cope fine so far. xenbus_frontend_register is basically just driver_register so it should be safe to call even before the register_bus(xenbusfoo) call. Things shouldn't kick off until xen_probe() is called which either happens explicitly in xenbus_probe_init() or via probe_work which contains a xenstored_ready check. Ian.