From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:44402) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1eK5rW-0003rm-Gy for qemu-devel@nongnu.org; Wed, 29 Nov 2017 12:09:31 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1eK5rR-0006cl-H6 for qemu-devel@nongnu.org; Wed, 29 Nov 2017 12:09:30 -0500 Received: from mx1.redhat.com ([209.132.183.28]:35674) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1eK5rR-0006bX-AT for qemu-devel@nongnu.org; Wed, 29 Nov 2017 12:09:25 -0500 Date: Wed, 29 Nov 2017 15:09:18 -0200 From: Eduardo Habkost Message-ID: <20171129170918.GA3037@localhost.localdomain> References: <1511530010-511740-1-git-send-email-dplotnikov@virtuozzo.com> <20171128195817.GA29077@localhost.localdomain> <2cd31202-27c3-983f-85a9-6814ff504706@virtuozzo.com> <20171128211326.GV3037@localhost.localdomain> <5A1E43A6.1010607@huawei.com> <20171129104122.GY3037@localhost.localdomain> <5A1EA0DB.60303@huawei.com> <20171129133524.GA2279@rkaganb.sw.ru> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20171129133524.GA2279@rkaganb.sw.ru> Subject: Re: [Qemu-devel] [PATCH] i386: turn off l3-cache property by default List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Roman Kagan , "Longpeng (Mike)" , "Denis V. Lunev" , "Michael S. Tsirkin" , Denis Plotnikov , pbonzini@redhat.com, rth@twiddle.net, qemu-devel@nongnu.org, Gonglei , zhaoshenglong On Wed, Nov 29, 2017 at 04:35:25PM +0300, Roman Kagan wrote: > > On 2017/11/29 18:41, Eduardo Habkost wrote: [...] > > > IMO, the long term solution is to make Linux guests not misbehave > > > when we stop lying about the L3 cache. Maybe we could provide a > > > "IPIs are expensive, please avoid them" hint in the KVM CPUID > > > leaf? > > We already have it, it's the hypervisor bit ;) Seriously, I'm unaware > of hypervisors where IPIs aren't expensive. Sounds good enough to me, if we can convince the Linux kernel maintainers that it should avoid IPIs under all hypervisors. -- Eduardo