Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752483AbdGDPWS (ORCPT ); Tue, 4 Jul 2017 11:22:18 -0400 Received: from mx2.suse.de ([195.135.220.15]:50262 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1752219AbdGDPWR (ORCPT ); Tue, 4 Jul 2017 11:22:17 -0400 Date: Tue, 4 Jul 2017 08:22:06 -0700 From: Davidlohr Bueso To: Thomas Gleixner , LKML , linux-mm@kvack.org, Andrey Ryabinin , Michal Hocko , Andrew Morton , Vlastimil Babka , Vladimir Davydov , Peter Zijlstra Subject: Re: [patch V2 2/2] mm/memory-hotplug: Switch locking to a percpu rwsem Message-ID: <20170704152206.GB11168@linux-80c1.suse> Mail-Followup-To: Thomas Gleixner , LKML , linux-mm@kvack.org, Andrey Ryabinin , Michal Hocko , Andrew Morton , Vlastimil Babka , Vladimir Davydov , Peter Zijlstra References: <20170704093232.995040438@linutronix.de> <20170704093421.506836322@linutronix.de> <20170704150106.GA11168@linux-80c1.suse> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Disposition: inline In-Reply-To: <20170704150106.GA11168@linux-80c1.suse> User-Agent: Mutt/1.5.24 (2015-08-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 380 Lines: 11 On Tue, 04 Jul 2017, Davidlohr Bueso wrote: >As a side effect you end up optimizing get/put_online_mems() at the cost >of more overhead for the actual hotplug operation, which is rare and of less >performance importance. So nm this, the reader side actually gets _more_ expensive with pcpu-rwsems due to at least two full barriers for each get/put operation. Thanks, Davidlohr