From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [PATCH 0/2] Introduce panic hypercall Date: Tue, 21 Jun 2011 12:07:26 +0300 Message-ID: <4E005F4E.1060200@redhat.com> References: <1308577094-17551-1-git-send-email-gollub@b1-systems.de> <20110620153825.GH13042@redhat.com> <4DFF6B20.7090107@redhat.com> <201106201826.32975.gollub@b1-systems.de> <4DFF76B1.8020509@redhat.com> <20110621060251.GB491@redhat.com> <4E0051B1.8060801@redhat.com> <20110621084155.GD491@redhat.com> <4E005CBA.3000105@redhat.com> <20110621090348.GE491@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Daniel Gollub , "Daniel P. Berrange" , kvm@vger.kernel.org, qemu-devel To: Gleb Natapov Return-path: Received: from mx1.redhat.com ([209.132.183.28]:46810 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751662Ab1FUJHd (ORCPT ); Tue, 21 Jun 2011 05:07:33 -0400 In-Reply-To: <20110621090348.GE491@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: On 06/21/2011 12:03 PM, Gleb Natapov wrote:> >> > > >Something like steal time, but for watchdog. But this become complicated fast. > > >Watchdog emulation will have to move into kernel for starter. > > > > Why? You can use a performance counter from userspace. > > > Heh, haven't thought about such way of implementing watchdog device. > But the same question again: what impact on performance constantly running > guest under perf is? Same answer. > Doesn't running guest under perf involve a lot of > NMIs (and hence vmexists)? The NMI rate is proportional to how you program the counter. If you program it to exit every 2 billion cycles, that's what you'll get. -- error compiling committee.c: too many arguments to function