From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [kvm-devel] [RFC][PATCH 2/4] kvm: Arch-specifc KVM_EXIT_DEBUG payload Date: Thu, 22 May 2008 16:59:41 +0300 Message-ID: <48357C4D.9070400@qumranet.com> References: <482D9198.7040801@web.de> <482DAFF6.3050403@web.de> <48344820.60103@qumranet.com> <4835782F.60804@web.de> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: kvm-devel , Hollis Blanchard To: Jan Kiszka Return-path: Received: from bzq-179-150-194.static.bezeqint.net ([212.179.150.194]:46845 "EHLO il.qumranet.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756152AbYEVN7p (ORCPT ); Thu, 22 May 2008 09:59:45 -0400 In-Reply-To: <4835782F.60804@web.de> Sender: kvm-owner@vger.kernel.org List-ID: Jan Kiszka wrote: > >> We need to handle branch tracing and last branch recording as well. It >> doesn't have to be in this patchset, though. >> > > IIRC, both features are fairly vendor specific. Do we have the same > level of support on both AMD and Intel CPUs? Moreover, I doubt that the > debug interface would be the right place for them. > Branch tracing is common, and is very much related to debugging (it modifies singlestep to trap only on branches, not on non-branching insns). LBR is either common or very similar, and is also very important for debugging. -- error compiling committee.c: too many arguments to function