From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751830AbaCVS4f (ORCPT ); Sat, 22 Mar 2014 14:56:35 -0400 Received: from mail-qa0-f41.google.com ([209.85.216.41]:39281 "EHLO mail-qa0-f41.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751465AbaCVSzz (ORCPT ); Sat, 22 Mar 2014 14:55:55 -0400 Date: Sat, 22 Mar 2014 14:55:51 -0400 From: Tejun Heo To: Frederic Weisbecker Cc: Kevin Hilman , LKML , Christoph Lameter , Mike Galbraith , "Paul E. McKenney" , Viresh Kumar Subject: Re: [PATCH 3/3] workqueue: Add anon workqueue sysfs hierarchy Message-ID: <20140322185551.GA24945@htj.dyndns.org> References: <1394815131-17271-1-git-send-email-fweisbec@gmail.com> <1394815131-17271-4-git-send-email-fweisbec@gmail.com> <7ha9csionc.fsf@paris.lan> <20140322170114.GA20038@localhost.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20140322170114.GA20038@localhost.localdomain> 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 Sat, Mar 22, 2014 at 06:01:18PM +0100, Frederic Weisbecker wrote: > 1) Call a per workqueue mutex when a work execute on an ordered workqueue. Although > contention should be very rare (only while we replace the workqueue attrs and > switch to a new worker), frequent locking may have a visible impact. > > 2) Have a seperate worker for all ordered workqueues. But we may lose a bit of > serialization with other workqueues along the way. Ordered workqueues are always bound to the fallback default worker pool. Why not just adjust cpus_allowed of the worker pool? Thanks. -- tejun