Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1760435Ab2EPTIC (ORCPT ); Wed, 16 May 2012 15:08:02 -0400 Received: from mx1.redhat.com ([209.132.183.28]:4130 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755502Ab2EPTH7 (ORCPT ); Wed, 16 May 2012 15:07:59 -0400 Date: Wed, 16 May 2012 22:07:58 +0300 From: "Michael S. Tsirkin" To: Gleb Natapov Cc: Marcelo Tosatti , x86@kernel.org, kvm@vger.kernel.org, Ingo Molnar , "H. Peter Anvin" , Avi Kivity , Linus Torvalds , linux-kernel@vger.kernel.org Subject: Re: [PATCHv4 3/5] kvm: host side for eoi optimization Message-ID: <20120516190757.GA12113@redhat.com> References: <42ba7beaa25855997e02fea91a11c9224c51720e.1337168687.git.mst@redhat.com> <20120516154940.GA20549@amt.cnet> <20120516162247.GA10676@redhat.com> <20120516172058.GA22861@amt.cnet> <20120516175856.GD10769@redhat.com> <20120516181500.GA23791@amt.cnet> <20120516182520.GH28798@redhat.com> <20120516182948.GF10769@redhat.com> <20120516183822.GI28798@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20120516183822.GI28798@redhat.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 5253 Lines: 125 On Wed, May 16, 2012 at 09:38:22PM +0300, Gleb Natapov wrote: > On Wed, May 16, 2012 at 09:29:49PM +0300, Michael S. Tsirkin wrote: > > On Wed, May 16, 2012 at 09:25:20PM +0300, Gleb Natapov wrote: > > > On Wed, May 16, 2012 at 03:15:00PM -0300, Marcelo Tosatti wrote: > > > > On Wed, May 16, 2012 at 08:58:57PM +0300, Michael S. Tsirkin wrote: > > > > > On Wed, May 16, 2012 at 02:20:58PM -0300, Marcelo Tosatti wrote: > > > > > > On Wed, May 16, 2012 at 07:22:47PM +0300, Michael S. Tsirkin wrote: > > > > > > > On Wed, May 16, 2012 at 12:49:40PM -0300, Marcelo Tosatti wrote: > > > > > > > > > @@ -1245,9 +1306,20 @@ int kvm_get_apic_interrupt(struct kvm_vcpu *vcpu) > > > > > > > > > int vector = kvm_apic_has_interrupt(vcpu); > > > > > > > > > struct kvm_lapic *apic = vcpu->arch.apic; > > > > > > > > > > > > > > > > > > - if (vector == -1) > > > > > > > > > + /* Detect interrupt nesting and disable EOI optimization */ > > > > > > > > > + if (pv_eoi_enabled(vcpu) && vector == -2) > > > > > > > > > + pv_eoi_clr_pending(vcpu); > > > > > > > > > + > > > > > > > > > + if (vector < 0) > > > > > > > > > > > > > > > > With interrupt window exiting, the guest will exit: > > > > > > > > > > > > > > > > - as soon as it sets RFLAGS.IF=1 and there is any > > > > > > > > interrupt pending in IRR. > > > > > > > > - any new interrupt is set in IRR will kick vcpu > > > > > > > > out of guest mode and recalculate interrupt-window-exiting. > > > > > > > > > > > > > > > > Doesnt this make this bit unnecessary ? > > > > > > > > > > > > > > Looks like we could cut it out. But I'm not sure how architectural it is > > > > > > > that we exit on interrupt window. > > > > > > > I guess there are reasons to exit on interrupt window but > > > > > > > isn't it better to make the feature independent of it? > > > > > > > > > > > > Hum... not sure. Is it helpful for the Hyper-V interface? > > > > > > > > > > > > > This almost never happens in my testing anyway, so > > > > > > > however we handle it is unlikely to affect performance. > > > > > > > > > > > > It decreases the amount of state that must be maintained. > > > > > > > > > > > > BTW there is a bug covered by interrupt window exiting: > > > > > > > > > > > > vcpu0 host > > > > > > - irr 5 set > > > > > > - isr 5 set, irr 5 cleared > > > > > > - eoi_skip bit not set, > > > > > > no other bit set in irr. > > > > > > - enter guest > > > > > > > > > > > > irr 4 set > > > > > > kick vcpu0 out of guest mode > > > > > > > > > > > > - eoi pending bit not set > > > > > > (previous interrupt injection > > > > > > still pending) > > > > > > - skip eoi > > > > > > > > > > > > If it were not for interrupt window exiting, this would > > > > > > inject vector 4 on an unrelated exit who knows how long > > > > > > in the future. > > > > > > > > > > > > Also note optimization depends on the fact that the host > > > > > > kicks vcpu out unconditionally (so it is dependent on > > > > > > certain kvm implementation details). > > > > > > > > > > > > > > > > Look we can summarize as follows: irq windows exit is > > > > > required both before and after this patch. > > > > > But it does not make the check above redundant. > > > > > > > > Right, it is not redundant. > > > > > > > > The above is still a bug: a case where eoi pending bit is not updated > > > > properly. > > > When IRR is set while eoi_skip is enabled, eoi_skip should be cleared. > > > Michael does your patch do that? > > > > I think this does it: > > /* Detect interrupt nesting and disable EOI optimization */ > > if (pv_eoi_enabled(vcpu) && vector == -2) > > pv_eoi_clr_pending(vcpu); > > > > -2 is returned when irr is set. > > > This code is reached from kvm_cpu_get_interrupt(), but this function will > not be called in above scenario. I think I see. So this shall fix it also makes code cleaner (no -2 hack). Right? kvm_apic_has_interrupt is called correct? diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index b4f7013..5a38e34 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -1273,8 +1273,12 @@ int kvm_apic_has_interrupt(struct kvm_vcpu *vcpu) highest_irr = apic_find_highest_irr(apic); if (highest_irr == -1) return -1; - if (((highest_irr & 0xF0) <= apic_get_reg(apic, APIC_PROCPRI))) - return -2; + /* Detect interrupt nesting and disable EOI optimization */ + if ((highest_irr & 0xF0) <= apic_get_reg(apic, APIC_PROCPRI)) { + if (pv_eoi_enabled(vcpu)) + pv_eoi_clr_pending(vcpu); + return -1; + } return highest_irr; } @@ -1306,10 +1310,6 @@ int kvm_get_apic_interrupt(struct kvm_vcpu *vcpu) int vector = kvm_apic_has_interrupt(vcpu); struct kvm_lapic *apic = vcpu->arch.apic; - /* Detect interrupt nesting and disable EOI optimization */ - if (pv_eoi_enabled(vcpu) && vector == -2) - pv_eoi_clr_pending(vcpu); - if (vector < 0) return -1; -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/