From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:40269) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1boTzJ-0007M6-KD for qemu-devel@nongnu.org; Mon, 26 Sep 2016 07:22:22 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1boTzF-0003HD-SC for qemu-devel@nongnu.org; Mon, 26 Sep 2016 07:22:20 -0400 Received: from mx1.redhat.com ([209.132.183.28]:41940) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1boTzF-0003H9-My for qemu-devel@nongnu.org; Mon, 26 Sep 2016 07:22:17 -0400 References: <1474548655-157373-1-git-send-email-imammedo@redhat.com> <1474548655-157373-8-git-send-email-imammedo@redhat.com> <20160922180017.1f35fc5a@nial.brq.redhat.com> <9664929b-6b84-ae11-0682-3743e7a78238@redhat.com> <20160926131008.116dfc3f@nial.brq.redhat.com> From: Paolo Bonzini Message-ID: Date: Mon, 26 Sep 2016 13:22:11 +0200 MIME-Version: 1.0 In-Reply-To: <20160926131008.116dfc3f@nial.brq.redhat.com> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH v2 07/14] pc: apic_common: extend APIC ID property to 32bit List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Igor Mammedov Cc: qemu-devel@nongnu.org, rkrcmar@redhat.com, ehabkost@redhat.com, mst@redhat.com, kevin@koconnor.net, lersek@redhat.com, chao.gao@intel.com, peterx@redhat.com, liuxiaojian6@huawei.com On 26/09/2016 13:10, Igor Mammedov wrote: > On Thu, 22 Sep 2016 18:16:47 +0200 > Paolo Bonzini wrote: > >> On 22/09/2016 18:00, Igor Mammedov wrote: >>>> Why not just return initial_apic_id? This is the meaning the property >>>> had before your patch. >>> >>> initial_apic_id is immutable but 'id' could be changed at runtime by guest in xAPIC mode >>> so returned value depends on xAPIC/x2APIC mode >> >> Understood, but this is just a possibly poorly-named property. "id" >> (e.g. from info qtree as opposed to info lapic) used to be the initial >> APIC ID always, even in x2APIC mode. > > 'info qtree' doesn't show CPUs anymore (since ICC bus has been removed), > but if it were it would show effective APIC ID. Same applie[ds] for > reading property value with qom-get. Oh, thanks for correcting me then. >> Not a big deal, but thought I'd mention it since you can keep using >> static properties. > > changing initial APIC ID from guest probably wouldn't work anyway > and beak somewhere else, so we could just continue to ignore > it and use static properties for now if you prefer. No big deal---I'll let other reviewers chime in. Paolo