From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nikunj A Dadhania Subject: Re: [PATCH v4 0/8] KVM paravirt remote flush tlb Date: Tue, 04 Sep 2012 07:00:31 +0530 Message-ID: <87txveefrc.fsf@abhimanyu.in.ibm.com> References: <20120821112346.3512.99814.stgit@abhimanyu.in.ibm.com> <5044BFCA.7060303@redhat.com> Mime-Version: 1.0 Content-Type: text/plain Cc: mtosatti@redhat.com, raghukt@linux.vnet.ibm.com, alex.shi@intel.com, kvm@vger.kernel.org, stefano.stabellini@eu.citrix.com, peterz@infradead.org, hpa@zytor.com, vsrivatsa@gmail.com, mingo@elte.hu To: Avi Kivity Return-path: Received: from e28smtp01.in.ibm.com ([122.248.162.1]:45510 "EHLO e28smtp01.in.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754797Ab2IDBbS (ORCPT ); Mon, 3 Sep 2012 21:31:18 -0400 Received: from /spool/local by e28smtp01.in.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Tue, 4 Sep 2012 07:01:15 +0530 Received: from d28av05.in.ibm.com (d28av05.in.ibm.com [9.184.220.67]) by d28relay01.in.ibm.com (8.13.8/8.13.8/NCO v10.0) with ESMTP id q841V7Ae31326438 for ; Tue, 4 Sep 2012 07:01:07 +0530 Received: from d28av05.in.ibm.com (loopback [127.0.0.1]) by d28av05.in.ibm.com (8.14.4/8.13.1/NCO v10.0 AVout) with ESMTP id q841V5PU008800 for ; Tue, 4 Sep 2012 11:31:07 +1000 In-Reply-To: <5044BFCA.7060303@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: On Mon, 03 Sep 2012 17:33:46 +0300, Avi Kivity wrote: > On 08/21/2012 02:25 PM, Nikunj A. Dadhania wrote: > > > > kernbench(lower is better) > > ========================== > > base pvflushv4 %improvement > > 1VM 48.5800 46.8513 3.55846 > > 2VM 108.1823 104.6410 3.27346 > > 3VM 183.2733 163.3547 10.86825 > > > > ebizzy(higher is better) > > ======================== > > base pvflushv4 %improvement > > 1VM 2414.5000 2089.8750 -13.44481 > > 2VM 2167.6250 2371.7500 9.41699 > > 3VM 1600.1111 2102.5556 31.40060 > > > > The regression is worrying. We're improving the contended case at the > cost of the non-contended case, this is usually the wrong thing to do. > Do we have any clear idea of the cause of the regression? > Previous perf numbers suggest that in 1VM scenario flush_tlb_others_ipi is around 2%, while for contented case its around 10%. That is what is helping contended case. Regards, Nikunj