Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933171AbbD1K2U (ORCPT ); Tue, 28 Apr 2015 06:28:20 -0400 Received: from cn.fujitsu.com ([59.151.112.132]:29144 "EHLO heian.cn.fujitsu.com" rhost-flags-OK-FAIL-OK-FAIL) by vger.kernel.org with ESMTP id S964803AbbD1K2T (ORCPT ); Tue, 28 Apr 2015 06:28:19 -0400 X-IronPort-AV: E=Sophos;i="5.04,848,1406563200"; d="scan'208";a="91352279" Message-ID: <553F6189.2070405@cn.fujitsu.com> Date: Tue, 28 Apr 2015 18:31:37 +0800 From: Lai Jiangshan User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.3.0 MIME-Version: 1.0 To: Mike Galbraith , Tejun Heo CC: , Christoph Lameter , Kevin Hilman , Mike Galbraith , "Paul E. McKenney" , Viresh Kumar , Frederic Weisbecker Subject: Re: [PATCH 3/3 V8] workqueue: Allow modifying low level unbound workqueue cpumask References: <1430128720-3021-1-git-send-email-laijs@cn.fujitsu.com> <1430128720-3021-4-git-send-email-laijs@cn.fujitsu.com> <20150427160756.GE1499@htj.duckdns.org> <553EE60C.1040503@cn.fujitsu.com> <20150428034440.GC2152@htj.duckdns.org> <1430195804.3168.19.camel@gmail.com> In-Reply-To: <1430195804.3168.19.camel@gmail.com> Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit X-Originating-IP: [10.167.226.103] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1096 Lines: 29 On 04/28/2015 12:36 PM, Mike Galbraith wrote: > On Mon, 2015-04-27 at 23:44 -0400, Tejun Heo wrote: > >>> So, we need an API to modify the wq_unbound_cpumask, and I provided >>> this public function. Otherwise, the other code can't modify it. >> >> I see. I don't have too strong an opinion; however, changing the mask >> is a fairly heavy operation. Are there specific reasons why we don't >> want to follow the nohz config right away? > > Isolation is not only applicable to nohz_full. Many loads are > unsuitable for nohz_full, yet require maximum isolation. > > ATM, nohz_full is not dynamic, but hopefully one day will be. In the > here and now, we can isolate cores from the scheduler on the fly via > cpusets, a prime API user candidate. > > -Mike > So, the public function needs to be kept and the EXPORT_SYMBOL_GPL() is killed? -- 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/