From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755356Ab3A2JMx (ORCPT ); Tue, 29 Jan 2013 04:12:53 -0500 Received: from mail-bk0-f48.google.com ([209.85.214.48]:36269 "EHLO mail-bk0-f48.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755102Ab3A2JMu (ORCPT ); Tue, 29 Jan 2013 04:12:50 -0500 Date: Tue, 29 Jan 2013 10:12:45 +0100 From: Ingo Molnar To: Yuanhan Liu Cc: Huang Ying , Linus Torvalds , David Howells , linux-kernel@vger.kernel.org, lkp@linux.intel.com Subject: Re: aim7 performance regression by commit 5a50508 report from LKP Message-ID: <20130129091245.GB5775@gmail.com> References: <20130129082537.GS12678@yliu-dev.sh.intel.com> <20130129084400.GA5350@gmail.com> <20130129090620.GT12678@yliu-dev.sh.intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20130129090620.GT12678@yliu-dev.sh.intel.com> 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 * Yuanhan Liu wrote: > On Tue, Jan 29, 2013 at 09:44:00AM +0100, Ingo Molnar wrote: > > > > * Yuanhan Liu wrote: > > > > > [...] > > > > Very nice measurements and analysis, thanks! > > > > > As stated above, anybody can have a chance to own the lock in > > > mutex once somebody release the lock. Well, there is only one > > > to own the lock in rwsem write lock, and the one is known > > > already: the one in the head of wait list. That would result > > > to more contention in rwsem write lock case, especially if the > > > one _will_ own the lock is not running now. > > > > I think we should allow lock-steal between rwsem writers - that > > will not hurt fairness as most rwsem fairness concerns relate to > > reader vs. writer fairness. > > Agreed, and I'm sure this will improve performance and may > make this performance regression go away. > > David, is that Ok to you? If so, I may have a try. I'm not David but please try it :-) Making rwsem behavior and scalability similar to mutexes would have numerous advantages. > > Am I correct to assume that all relevant users in this > > workload are down_write() users? > > Yes, as commit 5a50508 just convert all mutex to down_write. A second track of inquiry would be to see whether any of the key usage sites could be converted to down_read() or whether the lock hold times could be reduced drastically - but I doubt that's really possible on such heavily forking workloads. Thanks, Ingo