From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758844Ab2CFKDt (ORCPT ); Tue, 6 Mar 2012 05:03:49 -0500 Received: from e28smtp07.in.ibm.com ([122.248.162.7]:53324 "EHLO e28smtp07.in.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756807Ab2CFKDr (ORCPT ); Tue, 6 Mar 2012 05:03:47 -0500 Date: Tue, 6 Mar 2012 15:33:36 +0530 From: Srivatsa Vaddagiri To: Ingo Molnar Cc: Peter Zijlstra , Mike Galbraith , Suresh Siddha , linux-kernel , Paul Turner Subject: Re: sched: Avoid SMT siblings in select_idle_sibling() if possible Message-ID: <20120306100336.GB14298@linux.vnet.ibm.com> Reply-To: Srivatsa Vaddagiri References: <1329764866.2293.376.camhel@twins> <20120305152443.GE26559@linux.vnet.ibm.com> <20120306091410.GD27238@elte.hu> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline In-Reply-To: <20120306091410.GD27238@elte.hu> User-Agent: Mutt/1.5.21 (2010-09-15) x-cbid: 12030610-8878-0000-0000-0000018E8BD7 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Ingo Molnar [2012-03-06 10:14:11]: > Does volanomark still do yield(), I could count 3.5M yield() calls across 16 cpus over a period of 5.4sec, so yes it does seem to make very good use of yeild()! > thereby invoking a random > shuffle of thread scheduling and pretty much voluntarily > ejecting itself from most scheduler performance considerations? > > If it uses a real locking primitive such as futexes then its > performance matters more. Good point ..I will gather more data for additional benchmarks (tbench, sysbench) and post shortly. - vatsa