Received: from eggs.gnu.org ([2001:4830:134:3::10]:50170) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WFkY1-0003pA-7b for qemu-devel@nongnu.org; Tue, 18 Feb 2014 08:17:22 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1WFkXv-0007ym-Uk for qemu-devel@nongnu.org; Tue, 18 Feb 2014 08:17:17 -0500 Received: from mx1.redhat.com ([209.132.183.28]:36418) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WFkXv-0007ye-Ld for qemu-devel@nongnu.org; Tue, 18 Feb 2014 08:17:11 -0500 From: Markus Armbruster References: <1391699295-8381-1-git-send-email-armbru@redhat.com> <87y518a8da.fsf@blackfin.pond.sub.org> <530358D4.3010505@suse.de> Date: Tue, 18 Feb 2014 14:17:04 +0100 In-Reply-To: <530358D4.3010505@suse.de> ("Andreas =?utf-8?Q?F=C3=A4rber=22?= =?utf-8?Q?'s?= message of "Tue, 18 Feb 2014 13:57:56 +0100") Message-ID: <87wqgs8srz.fsf@blackfin.pond.sub.org> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH v4] qdev: Keep global allocation counter per bus List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Andreas =?utf-8?Q?F=C3=A4rber?= Cc: peter.maydell@linaro.org, "Michael S. Tsirkin" , qemu-devel@nongnu.org, Alexander Graf , Anthony Liguori , Paolo Bonzini Andreas F=C3=A4rber writes: > Am 18.02.2014 13:54, schrieb Markus Armbruster: >> Peter, can you merge this patch? v2 is from December 20, and the >> changes since then have been limited to the commit message. >>=20 >> In case Peter doesn't want to take it directly: Andreas, would you be >> willing to take it through your tree? It's not really QOM, but your >> tree is the closest fit I can find. > > Last thing I read was some open discussions related to migration > problems, so I didn't take it yet. If those are resolved, as a qdev Yes, we discussed migration at some length, but the problem to resolve was documenting the breakage, which we did in v4. We break -device bus=3DFOO.0 in the case where we have multiple FOO.0 before the patch (all but one unusable with -device), and orderly FOO.0, FOO.1, ... after. We (Paolo, Alex and I) agreed that avoiding this breakage isn't worthwhile. The commit message enumerates the affected buses, and explains how to work around the migration breakage. > patch I would take it through qom-next, if mst acks for the PC part. Thanks. Michael, please review :)