From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754776AbaEAOv5 (ORCPT ); Thu, 1 May 2014 10:51:57 -0400 Received: from mail-wi0-f170.google.com ([209.85.212.170]:33416 "EHLO mail-wi0-f170.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752974AbaEAOv4 (ORCPT ); Thu, 1 May 2014 10:51:56 -0400 Date: Thu, 1 May 2014 16:51:47 +0200 From: Frederic Weisbecker To: Tejun Heo Cc: LKML , Christoph Lameter , Kevin Hilman , Lai Jiangshan , Mike Galbraith , "Paul E. McKenney" , Viresh Kumar Subject: Re: [PATCH 4/4] workqueue: Handle ordered workqueues on cpumask_unbounds change Message-ID: <20140501145145.GC25369@localhost.localdomain> References: <1398350256-7834-1-git-send-email-fweisbec@gmail.com> <1398350256-7834-5-git-send-email-fweisbec@gmail.com> <20140424153320.GF14460@htj.dyndns.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20140424153320.GF14460@htj.dyndns.org> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Apr 24, 2014 at 11:33:20AM -0400, Tejun Heo wrote: > On Thu, Apr 24, 2014 at 04:37:36PM +0200, Frederic Weisbecker wrote: > > Ordered unbound workqueues need some special care if we want to > > modify their CPU affinity. These can't be simply handled through > > apply_workqueue_attrs() since it works by hot plugging worker pools > > which has parallelism side effects and this would break ordering. > > > > The way we solve this is to change the affinity of the (presumaly > > unique) worker backing the ordered workqueues. > > > > NOTE: Now like Lai said, there may be bad side effects on this because > > ordered wq may share their worker pool with non-ordered workqueues. > > So changing the affinity of the worker itself is not a nice solution. > > This patch is very likely to be replaced by Lai's patch > > "workqueue: allow changing attributions of ordered workqueue" > > https://lkml.org/lkml/2014/4/15/181 > > Yeah, it bothers me that we're taking two completely different > approaches for ordered and !ordered workqueues. The only difference > between them is concurrency and it probably would be a better idea to > address that directly. Yeah I've tried with Lai's patch and it seems to work like a charm so next version will likely be better.