From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1161405AbcFBOhQ (ORCPT ); Thu, 2 Jun 2016 10:37:16 -0400 Received: from mail-yw0-f196.google.com ([209.85.161.196]:34229 "EHLO mail-yw0-f196.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932128AbcFBOhL (ORCPT ); Thu, 2 Jun 2016 10:37:11 -0400 Date: Thu, 2 Jun 2016 10:37:03 -0400 From: Tejun Heo To: Jiri Pirko Cc: Bhaktipriya Shridhar , linux-kernel@vger.kernel.org Subject: Re: [patch net-next 16/18] mlxsw: coRE: Add mlxsw specific workqueue and use it for FDB notif. processing Message-ID: <20160602143703.GI14868@mtj.duckdns.org> References: <1460650770-19382-17-git-send-email-jiri@resnulli.us> <20160601194234.GA8615@Karyakshetra> <20160602084106.GE1987@nanopsycho> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20160602084106.GE1987@nanopsycho> User-Agent: Mutt/1.6.1 (2016-04-27) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org (cc'ing linux-kernel) Hello, On Thu, Jun 02, 2016 at 10:41:06AM +0200, Jiri Pirko wrote: > Wed, Jun 01, 2016 at 09:42:34PM CEST, bhaktipriya96@gmail.com wrote: > >Hi Jiri, > > > >I'm working on the workqueue project and I'm updating the legacy > >workqueue interface users to use the new workqueue API. > > > >This patch uses create_workqueue which is deprecated and needs an > >update. > > > >Your commit description says > >"It makes sense to create separate workqueue just for mlxsw driver > >in this case and do not pollute system_wq." > > > >I would like to inquire what "polluting system_wq" means. > > As the number of queued items is much bigger than originally expected, I > wanted to have a separate wq for our driver. I believe it makes sense. Unless it can queue a very large number of concurrent work items and thus requires explicit concurrency limit, using a dedicated wq or not doesn't make any difference. Sharing system_wq doesn't pollute it. All workqueues share the same worker pools anyway. Thanks. -- tejun