From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:34404) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Vj8Fc-0003oN-77 for qemu-devel@nongnu.org; Wed, 20 Nov 2013 08:55:33 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Vj8FW-0006w5-VV for qemu-devel@nongnu.org; Wed, 20 Nov 2013 08:55:27 -0500 Received: from mx1.redhat.com ([209.132.183.28]:53336) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Vj8FW-0006v5-Nq for qemu-devel@nongnu.org; Wed, 20 Nov 2013 08:55:22 -0500 Message-ID: <528CBF3C.7060004@redhat.com> Date: Wed, 20 Nov 2013 14:55:08 +0100 From: Paolo Bonzini MIME-Version: 1.0 References: <528C83C2.2040903@redhat.com> <20131120110000.GD20821@onelab2.iet.unipi.it> <528C9AB3.905@redhat.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] commit b1bbfe72 causes huge slowdown with no kvm List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Alex Bligh Cc: Luigi Rizzo , qemu-devel Il 20/11/2013 14:47, Alex Bligh ha scritto: > I think you are saying this sounds like another underlying bug > (no pause instruction) uncovered by the timer patches rather than > a bug in the timer patches. > > Reading > http://software.intel.com/file/27087 > perhaps the answer is for pause to yield the lock. > > However, what if an arbitrary guest foolishly /does/ do a tight > spinlock without calling pause? It would be a problem. I don't think it has to be fixed though (and anyway of course not now). The fix could be to make TCG use "real" SMP instead of round-robin. Paolo