Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932912Ab1DMU6M (ORCPT ); Wed, 13 Apr 2011 16:58:12 -0400 Received: from ogre.sisk.pl ([217.79.144.158]:46646 "EHLO ogre.sisk.pl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932627Ab1DMU6L (ORCPT ); Wed, 13 Apr 2011 16:58:11 -0400 From: "Rafael J. Wysocki" To: Mike Frysinger Subject: Re: freezer: should barriers be smp ? Date: Wed, 13 Apr 2011 22:58:17 +0200 User-Agent: KMail/1.13.6 (Linux/2.6.39-rc3+; KDE/4.6.0; x86_64; ; ) Cc: Pavel Machek , linux-kernel@vger.kernel.org, uclinux-dist-devel@blackfin.uclinux.org, linux-pm@lists.linux-foundation.org References: In-Reply-To: MIME-Version: 1.0 Content-Type: Text/Plain; charset="utf-8" Content-Transfer-Encoding: 7bit Message-Id: <201104132258.17705.rjw@sisk.pl> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 951 Lines: 21 On Wednesday, April 13, 2011, Mike Frysinger wrote: > when we suspend/resume Blackfin SMP systems, we notice that the > freezer code runs on multiple cores. this is of course what you want > -- freeze processes in parallel. however, the code only uses non-smp > based barriers which causes us problems ... our cores need software > support to keep caches in sync, so our smp barriers do just that. but > the non-smp barriers do not, and so the frozen/thawed processes > randomly get stuck in the wrong task state. > > thinking about it, shouldnt the freezer code be using smp barriers ? Yes, it should, but rmb() and wmb() are supposed to be SMP barriers. Or do you mean something different? Rafael -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/