Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752736AbdH2NDf (ORCPT ); Tue, 29 Aug 2017 09:03:35 -0400 Received: from mga02.intel.com ([134.134.136.20]:18872 "EHLO mga02.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751566AbdH2NDd (ORCPT ); Tue, 29 Aug 2017 09:03:33 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.41,444,1498546800"; d="scan'208";a="1189334979" From: Andi Kleen To: Yang Zhang Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org, wanpeng.li@hotmail.com, mst@redhat.com, pbonzini@redhat.com, tglx@linutronix.de, rkrcmar@redhat.com, dmatlack@google.com, agraf@suse.de, peterz@infradead.org, linux-doc@vger.kernel.org Subject: Re: [RFC PATCH v2 0/7] x86/idle: add halt poll support References: <1504007201-12904-1-git-send-email-yang.zhang.wz@gmail.com> Date: Tue, 29 Aug 2017 06:03:26 -0700 In-Reply-To: <1504007201-12904-1-git-send-email-yang.zhang.wz@gmail.com> (Yang Zhang's message of "Tue, 29 Aug 2017 11:46:34 +0000") Message-ID: <87bmmyedb5.fsf@firstfloor.org> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/25.2 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 493 Lines: 14 Yang Zhang writes: > Some latency-intensive workload will see 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. You could test with https://lkml.org/lkml/2017/7/9/204 which optimizes these idle paths with a fast path. If that works it would be a far better solution than explicit tunables and basically giving up by polling. -Andi