From mboxrd@z Thu Jan 1 00:00:00 1970 From: Radim =?utf-8?B?S3LEjW3DocWZ?= Subject: Re: [PATCH v2 01/14] KVM: x86: change PIT discard tick policy Date: Thu, 18 Feb 2016 17:56:08 +0100 Message-ID: <20160218165608.GC18904@potion.brq.redhat.com> References: <1455736496-374-1-git-send-email-rkrcmar@redhat.com> <1455736496-374-2-git-send-email-rkrcmar@redhat.com> <56C5EDA9.9030204@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org, Yuki Shibuya , Rik van Riel To: Paolo Bonzini Return-path: Received: from mx1.redhat.com ([209.132.183.28]:51628 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1425536AbcBRQ4M (ORCPT ); Thu, 18 Feb 2016 11:56:12 -0500 Content-Disposition: inline In-Reply-To: <56C5EDA9.9030204@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: 2016-02-18 17:13+0100, Paolo Bonzini: > On 17/02/2016 20:14, Radim Kr=C4=8Dm=C3=A1=C5=99 wrote: >> Discard policy uses ack_notifiers to prevent injection of PIT interr= upts >> before EOI from the last one. >>=20 >> This patch changes the policy to always try to deliver the interrupt= , >> which makes a difference when its vector is in ISR. >> Old implementation would drop the interrupt, but proposed one inject= s to >> IRR, like real hardware would. >=20 > This seems like what libvirt calls the "merge" policy: Oops, I never looked beyond QEMU after seeing that the naming in libvir= t doesn't even match ... I think the policy that KVM implements (which I call discard) is "delay= " in libvirt. (https://libvirt.org/formatdomain.html#elementsTime) >=20 > Merge the missed tick(s) into one tick and inject. The guest time > may be delayed, depending on how the OS reacts to the merging of > ticks The "may be delayed" there makes me feel like the timer has to support = a guest visible counter of missed ticks. PIT will definitely be delayed if we get another tick while the previou= s one is still in IRR and there is nothing that the guest can do with it. "catchup" is the other KVM policy and "discard" also needs to allow the guest to handle lost ticks. > where the merged tick is the one placed into IRR. Unlike discard, > "merge" can starve the guest through an interrupt storm. Yeah, starving a VCPU with an interrupt storm is more likely with the changed policy. It's a pretty sad situation if all the time that VCPU gets isn't even enough to run a PIT handler, so I didn't care. The NMI watchdog bug can also be solved without changing the policy. (It's a hack in any case.)