From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755376Ab0JNPzu (ORCPT ); Thu, 14 Oct 2010 11:55:50 -0400 Received: from canuck.infradead.org ([134.117.69.58]:60939 "EHLO canuck.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754680Ab0JNPzt convert rfc822-to-8bit (ORCPT ); Thu, 14 Oct 2010 11:55:49 -0400 Subject: Re: [PATCH] mutex: Introduce mutex_cpu_relax() From: Peter Zijlstra To: gerald.schaefer@de.ibm.com Cc: Ingo Molnar , Andrew Morton , Martin Schwidefsky , LKML , linux-s390 , Heiko Carstens In-Reply-To: <1287070392.8344.15.camel@thinkpad> References: <1287070392.8344.15.camel@thinkpad> Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 8BIT Date: Thu, 14 Oct 2010 17:55:29 +0200 Message-ID: <1287071729.29097.278.camel@twins> Mime-Version: 1.0 X-Mailer: Evolution 2.28.3 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 2010-10-14 at 17:33 +0200, Gerald Schaefer wrote: > From: Gerald Schaefer > > The spinning mutex implementation uses cpu_relax() in busy loops as a > compiler barrier. Depending on the architecture, cpu_relax() may do more > than needed in this specific mutex spin loops. On System z we also give > up the time slice of the virtual cpu in cpu_relax(), which prevents > effective spinning on the mutex. > > This patch replaces cpu_relax() in the spinning mutex code with a new > function mutex_cpu_relax(), which can be defined by each architecture > that selects HAVE_MUTEX_CPU_RELAX. The default is still cpu_relax(), so > this patch should not affect other architectures than System z for now. Sure.. I'll take it.