From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:42482) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1X0Bi0-00071e-W5 for qemu-devel@nongnu.org; Thu, 26 Jun 2014 11:35:40 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1X0Bht-0002Qb-DC for qemu-devel@nongnu.org; Thu, 26 Jun 2014 11:35:32 -0400 Received: from mx1.redhat.com ([209.132.183.28]:64652) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1X0Bht-0002Pw-5Z for qemu-devel@nongnu.org; Thu, 26 Jun 2014 11:35:25 -0400 Message-ID: <53AC3DB9.7020101@redhat.com> Date: Thu, 26 Jun 2014 17:35:21 +0200 From: Paolo Bonzini MIME-Version: 1.0 References: <1403793280-4353-1-git-send-email-marc.mari.barcelo@gmail.com> <53AC3798.50302@redhat.com> <20140626172735.61327704@crunchbang> In-Reply-To: <20140626172735.61327704@crunchbang> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH] Functions bus_foreach and device_find from libqos virtio API List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: =?UTF-8?B?TWFyYyBNYXLDrQ==?= Cc: qemu-devel@nongnu.org, Stefan Hajnoczi Il 26/06/2014 17:27, Marc Mar=C3=AD ha scritto: > If these functions are just Virtio PCI specific, it makes sense to take > them out. If they are common to other transports, in my opinion is > better to leave them there. Of course, a good solution has to be found > to just having a global bus variable. > > It is also interesting to see that the current > implementation of qpci_device_foreach does only work in one bus, so > probably is not so terrible to have it as global and unique. qpci_device_foreach works for one bus only because libqos doesn't=20 support PCI bridges. But it lets you specify the PCI bus of interest,=20 and it seems a good idea to have this in the virtio-pci implementation. Paolo