From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:48651) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bLaAR-0001aZ-GQ for qemu-devel@nongnu.org; Fri, 08 Jul 2016 14:06:24 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1bLaAM-00085R-FT for qemu-devel@nongnu.org; Fri, 08 Jul 2016 14:06:23 -0400 Received: from mx1.redhat.com ([209.132.183.28]:48917) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bLaAM-00085L-9Y for qemu-devel@nongnu.org; Fri, 08 Jul 2016 14:06:18 -0400 Date: Fri, 8 Jul 2016 20:06:12 +0200 From: Radim =?utf-8?B?S3LEjW3DocWZ?= Message-ID: <20160708180612.GB17087@potion> References: <1467706769-12505-1-git-send-email-peterx@redhat.com> <20160708160129.GA10792@potion> <0fb824f1-b5c8-6dd9-f49b-981e3c9b6a22@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <0fb824f1-b5c8-6dd9-f49b-981e3c9b6a22@redhat.com> Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH v11 00/28] IOMMU: Enable interrupt remapping for Intel IOMMU List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Paolo Bonzini Cc: Peter Xu , ehabkost@redhat.com, mst@redhat.com, jasowang@redhat.com, qemu-devel@nongnu.org, alex.williamson@redhat.com, jan.kiszka@web.de, wexu@redhat.com, imammedo@redhat.com, marcel@redhat.com, davidkiarie4@gmail.com, rth@twiddle.net 2016-07-08 18:36+0200, Paolo Bonzini: > On 08/07/2016 18:01, Radim Kr=C4=8Dm=C3=A1=C5=99 wrote: >> The patchset doesn't work with if you have 16 APICs with IDs 0-15 and >> then some more, because KVM has hacked x2APIC support that translates >> logical interrupt with destination 0xff (first cluster of 16 VCPUs) in= to >> a broadcast (also works with lowest-priority) and then picks VCPU > 16 >> as the destination, leading to messages like >>=20 >> do_IRQ: 17.209 No irq handler for vector >>=20 >> The fun part is that this is how KVM wanted to behave, when it allowed >> x2apic with unremapped ioapic. :( >> KVM could either finally remove x2apic without IR or add a way in whic= h >> userspace could say whether an interrupt is x2APIC or not. >>=20 >> QEMU cannot do anything to work around the bug, so I think it would be >> best to disable EIM for now. >=20 > What's the state of QEMU support for KVM_CAP_X2APIC_API? If it can mak= e > it for hard freeze, it's not a showstopper for Peter's patch. It works and is of postable quality, but the series bases on outdated Igor's ACPI patches. Rebasing KVM_CAP_X2APIC_API on top of IOMMU would not complicate the situation too much and has good chances of getting in before the hard freeze. I'll coordinate with Igor on Monday.