From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: Non-exiting rdpmc on KVM guests? Date: Wed, 22 Apr 2015 23:05:25 +0200 Message-ID: <55380D15.7010902@redhat.com> References: <20150421150514.GC3182@worktop.meeting.verilan.com> <55367BB6.3050103@redhat.com> <20150421205139.GA2898@worktop.Skamania.guest> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Cc: Andy Lutomirski , kvm list To: Peter Zijlstra Return-path: Received: from mail-wg0-f45.google.com ([74.125.82.45]:33534 "EHLO mail-wg0-f45.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757161AbbDVVFb (ORCPT ); Wed, 22 Apr 2015 17:05:31 -0400 Received: by wgin8 with SMTP id n8so259538316wgi.0 for ; Wed, 22 Apr 2015 14:05:30 -0700 (PDT) In-Reply-To: <20150421205139.GA2898@worktop.Skamania.guest> Sender: kvm-owner@vger.kernel.org List-ID: On 21/04/2015 22:51, Peter Zijlstra wrote: > > However, if you take into account that RDPMC can also be used > > to read an inactive counter, and that multiple guests "fight" for the > > same host counters, it's even harder to ensure that the guest counter > > indices match those on the host. > > That doesn't make sense, only a single vcpu task will ever run at any > one time. Right, but it puts more pressure on the scheduler which could end up going more often through the slow path. Paolo