From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753754Ab2JDJLY (ORCPT ); Thu, 4 Oct 2012 05:11:24 -0400 Received: from mx1.redhat.com ([209.132.183.28]:60322 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752655Ab2JDJLX (ORCPT ); Thu, 4 Oct 2012 05:11:23 -0400 Date: Thu, 4 Oct 2012 11:10:59 +0200 From: Gleb Natapov To: Andi Kleen Cc: linux-kernel@vger.kernel.org, acme@redhat.com, x86@vger.kernel.org, eranian@google.com, jolsa@redhat.com, a.p.zijlstra@chello.nl, Andi Kleen , avi@redhat.com Subject: Re: [06/31] perf, kvm: Support the intx/intx_cp modifiers in KVM arch perfmon emulation Message-ID: <20121004091059.GY23096@redhat.com> References: <1349221731-15665-7-git-send-email-andi@firstfloor.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1349221731-15665-7-git-send-email-andi@firstfloor.org> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Oct 02, 2012 at 11:48:26PM -0000, Andi Kleen wrote: > From: Andi Kleen > > This is not arch perfmon, but older CPUs will just ignore it. This makes > it possible to do at least some TSX measurements from a KVM guest > > Cc: avi@redhat.com > Signed-off-by: Andi Kleen > > --- > arch/x86/kvm/pmu.c | 13 ++++++++++--- > 1 files changed, 10 insertions(+), 3 deletions(-) > > diff --git a/arch/x86/kvm/pmu.c b/arch/x86/kvm/pmu.c > index 9b7ec11..f72a409 100644 > --- a/arch/x86/kvm/pmu.c > +++ b/arch/x86/kvm/pmu.c > @@ -160,7 +160,7 @@ static void stop_counter(struct kvm_pmc *pmc) > > static void reprogram_counter(struct kvm_pmc *pmc, u32 type, > unsigned config, bool exclude_user, bool exclude_kernel, > - bool intr) > + bool intr, bool intx, bool intx_cp) > { > struct perf_event *event; > struct perf_event_attr attr = { > @@ -173,6 +173,11 @@ static void reprogram_counter(struct kvm_pmc *pmc, u32 type, > .exclude_kernel = exclude_kernel, > .config = config, > }; > + /* Will be ignored on CPUs that don't support this. */ > + if (intx) > + attr.config |= HSW_INTX; > + if (intx_cp) > + attr.config |= HSW_INTX_CHECKPOINTED; > > attr.sample_period = (-pmc->counter) & pmc_bitmask(pmc); > > @@ -239,7 +244,9 @@ static void reprogram_gp_counter(struct kvm_pmc *pmc, u64 eventsel) > reprogram_counter(pmc, type, config, > !(eventsel & ARCH_PERFMON_EVENTSEL_USR), > !(eventsel & ARCH_PERFMON_EVENTSEL_OS), > - eventsel & ARCH_PERFMON_EVENTSEL_INT); > + eventsel & ARCH_PERFMON_EVENTSEL_INT, > + !!(eventsel & HSW_INTX), > + !!(eventsel & HSW_INTX_CHECKPOINTED)); Before reprogram_gp_counter() is called from kvm_pmu_set_msr() eventsel is checked for reserved bits. Both HSW_INTX and HSW_INTX_CHECKPOINTED are marked as reserved currently. Reserved bit mask should depend on cpuid bits provided to a guest. > } > > static void reprogram_fixed_counter(struct kvm_pmc *pmc, u8 en_pmi, int idx) > @@ -256,7 +263,7 @@ static void reprogram_fixed_counter(struct kvm_pmc *pmc, u8 en_pmi, int idx) > arch_events[fixed_pmc_events[idx]].event_type, > !(en & 0x2), /* exclude user */ > !(en & 0x1), /* exclude kernel */ > - pmi); > + pmi, false, false); > } > > static inline u8 fixed_en_pmi(u64 ctrl, int idx) -- Gleb.