From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:59867) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1fxLRD-0005DK-Ed for qemu-devel@nongnu.org; Tue, 04 Sep 2018 20:12:56 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1fxLR4-0000xV-Le for qemu-devel@nongnu.org; Tue, 04 Sep 2018 20:12:47 -0400 Received: from mail-oi0-x243.google.com ([2607:f8b0:4003:c06::243]:36539) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1fxLR4-0000rQ-Df for qemu-devel@nongnu.org; Tue, 04 Sep 2018 20:12:42 -0400 Received: by mail-oi0-x243.google.com with SMTP id r69-v6so10318381oie.3 for ; Tue, 04 Sep 2018 17:12:34 -0700 (PDT) Sender: fluxion Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable From: Michael Roth In-Reply-To: <20180831112219.ujfdn4uiya5p7ul4@kamzik.brq.redhat.com> References: <1535630903-146863-1-git-send-email-imammedo@redhat.com> <38b3347b-8923-26e9-82a4-f69002262767@redhat.com> <20180831112219.ujfdn4uiya5p7ul4@kamzik.brq.redhat.com> Message-ID: <153610634684.28231.17938799672560738415@sif> Date: Tue, 04 Sep 2018 19:12:26 -0500 Subject: Re: [Qemu-devel] [PATCH] qga: ignore non present cpus when handling qmp_guest_get_vcpus() List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Andrew Jones , Laszlo Ersek Cc: Igor Mammedov , qemu-devel@nongnu.org Quoting Andrew Jones (2018-08-31 06:22:19) > On Thu, Aug 30, 2018 at 05:51:13PM +0200, Laszlo Ersek wrote: > > Originally these guest agent commands (both getting and setting) were > > meant to be used in the absence of real VCPU hot[un]plug, as a fallback > > / place-holder. > > = > > If the latter (=3D real VCPU hot(un)plug) works, then these guest agent > > commands shouldn't be used at all. > > = > > Drew, do I remember correctly? > = > Yup. This guest agent functionality was merely a stopgap to support "cpu > hotplug", before real cpu hotplug could be supported. > = > > I wonder if, instead of this patch, we should rework > > qmp_guest_get_vcpus(), to silently skip processors for which this > > dirpath ENOENT condition arises (i.e., return a shorter list of > > GuestLogicalProcessor objects). > = > That sounds good to me. +1, otherwise we end up reporting unplugged VCPUs as being present-but-offl= ine by virtue of them being in the returned list, which isn't really the case. > = > Thanks, > drew >=20