From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [PATCH V2 2/5] ara virt interface of perf to support kvm guest os statistics collection in guest os Date: Tue, 22 Jun 2010 12:41:05 +0300 Message-ID: <4C208531.3040403@redhat.com> References: <1277112686.2096.510.camel@ymzhang.sh.intel.com> <4C1F5452.3000107@redhat.com> <1277172510.2096.585.camel@ymzhang.sh.intel.com> <4C207E63.5010504@redhat.com> <1277198752.2096.721.camel@ymzhang.sh.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: LKML , kvm@vger.kernel.org, Ingo Molnar , Fr??d??ric Weisbecker , Arnaldo Carvalho de Melo , Cyrill Gorcunov , Lin Ming , Sheng Yang , Marcelo Tosatti , oerg Roedel , Jes Sorensen , Gleb Natapov , Zachary Amsden , zhiteng.huang@intel.com, tim.c.chen@intel.com, Peter Zijlstra To: "Zhang, Yanmin" Return-path: In-Reply-To: <1277198752.2096.721.camel@ymzhang.sh.intel.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: kvm.vger.kernel.org On 06/22/2010 12:25 PM, Zhang, Yanmin wrote: > >> I'm talking about the guest/host interface. So you have one vmexit and >> many host perf calls. >> > I understood what you were speaking. I mean, perf generic codes operate perf_event > one by one. At low layer, we just know one perf_event before calling hypercall to > vmexit to host kernel. > Ah. We might fix that by having the perf ops work on guest memory, and add a commit that transfers them to the host. But this is complicated and can be left for later. -- error compiling committee.c: too many arguments to function