Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932137Ab0DOJse (ORCPT ); Thu, 15 Apr 2010 05:48:34 -0400 Received: from mx1.redhat.com ([209.132.183.28]:37965 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932069Ab0DOJsc (ORCPT ); Thu, 15 Apr 2010 05:48:32 -0400 Message-ID: <4BC6E0D9.1090202@redhat.com> Date: Thu, 15 Apr 2010 12:48:09 +0300 From: Avi Kivity User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.9) Gecko/20100330 Fedora/3.0.4-1.fc12 Thunderbird/3.0.4 MIME-Version: 1.0 To: oerg Roedel CC: "Zhang, Yanmin" , Ingo Molnar , Peter Zijlstra , Sheng Yang , linux-kernel@vger.kernel.org, kvm@vger.kernel.org, Marcelo Tosatti , Jes Sorensen , Gleb Natapov , Zachary Amsden , zhiteng.huang@intel.com, tim.c.chen@intel.com, Arnaldo Carvalho de Melo Subject: Re: [PATCH V3] perf & kvm: Enhance perf to collect KVM guest os statistics from host side References: <1902387910.2078.435.camel@ymzhang.sh.intel.com> <4BC588CF.5010507@redhat.com> <1902445479.2078.458.camel@ymzhang.sh.intel.com> <4BC6C8CD.1020801@redhat.com> <1902473858.2078.481.camel@ymzhang.sh.intel.com> <20100415090403.GA12697@8bytes.org> <4BC6D7C8.9060302@redhat.com> <20100415094407.GB12697@8bytes.org> In-Reply-To: <20100415094407.GB12697@8bytes.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1054 Lines: 33 On 04/15/2010 12:44 PM, oerg Roedel wrote: > >> So, we'd need something like the following: >> >> if (exit == NMI) >> __get_cpu_var(nmi_vcpu) = vcpu; >> >> stgi(); >> >> if (exit == NMI) { >> while (!nmi_handled()) >> cpu_relax(); >> __get_cpu_var(nmi_vcpu) = NULL; >> } >> > Hmm, looks a bit complicated to me. The NMI should happen shortly after > the stgi instruction. Interrupts are still disabled so we stay on this > cpu. Can't we just set and erase the cpu_var at vcpu_load/vcpu_put time? > > That means an NMI that happens outside guest code (for example, in the mmu, or during the exit itself) would be counted as if in guest code. -- I have a truly marvellous patch that fixes the bug which this signature is too narrow to contain. -- 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/