From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751097AbdKTHSV (ORCPT ); Mon, 20 Nov 2017 02:18:21 -0500 Received: from mail-oi0-f68.google.com ([209.85.218.68]:33003 "EHLO mail-oi0-f68.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750795AbdKTHSS (ORCPT ); Mon, 20 Nov 2017 02:18:18 -0500 X-Google-Smtp-Source: AGs4zMY430c2DesnMSchQvX3+1+E/egsqYYMRJCVREFrJOTIEzTw5qDGPtKsa6FmpEfWHyTrtbY+zA== Subject: Re: [Xen-devel] [PATCH RFC v3 0/6] x86/idle: add halt poll support To: Konrad Rzeszutek Wilk , Quan Xu Cc: kvm@vger.kernel.org, linux-doc@vger.kernel.org, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, virtualization@lists.linux-foundation.org, x86@kernel.org, xen-devel@lists.xenproject.org, Yang Zhang References: <1510567565-5118-1-git-send-email-quan.xu0@gmail.com> <20171115213131.GB21113@char.us.oracle.com> From: Quan Xu Message-ID: <3c0c8999-f4e9-d8ae-571f-b62abfe6b4e4@gmail.com> Date: Mon, 20 Nov 2017 15:18:10 +0800 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.4.0 MIME-Version: 1.0 In-Reply-To: <20171115213131.GB21113@char.us.oracle.com> Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 8bit Content-Language: en-US Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 2017-11-16 05:31, Konrad Rzeszutek Wilk wrote: > On Mon, Nov 13, 2017 at 06:05:59PM +0800, Quan Xu wrote: >> From: Yang Zhang >> >> Some latency-intensive workload have seen obviously performance >> drop when running inside VM. The main reason is that the overhead >> is amplified when running inside VM. The most cost I have seen is >> inside idle path. > Meaning an VMEXIT b/c it is an 'halt' operation ? And then going > back in guest (VMRESUME) takes time. And hence your latency gets > all whacked b/c of this?    Konrad, I can't follow 'b/c' here.. sorry. > So if I understand - you want to use your _full_ timeslice (of the guest) > without ever (or as much as possible) to go in the hypervisor?     as much as possible. > Which means in effect you don't care about power-saving or CPUfreq > savings, you just want to eat the full CPU for snack?   actually, we  care about power-saving. The poll duration is self-tuning, otherwise it is almost as the same as   'halt=poll'. Also we always sent out with CPU usage of benchmark netperf/ctxsw. We got much more   performance with limited promotion of CPU usage. >> This patch introduces a new mechanism to poll for a while before >> entering idle state. If schedule is needed during poll, then we >> don't need to goes through the heavy overhead path. > Schedule of what? The guest or the host?   rescheduled of guest scheduler..   it is the guest. Quan Alibaba Cloud > >