From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754010AbYHSLog (ORCPT ); Tue, 19 Aug 2008 07:44:36 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752799AbYHSLo2 (ORCPT ); Tue, 19 Aug 2008 07:44:28 -0400 Received: from mx2.mail.elte.hu ([157.181.151.9]:41495 "EHLO mx2.mail.elte.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752654AbYHSLo2 (ORCPT ); Tue, 19 Aug 2008 07:44:28 -0400 Date: Tue, 19 Aug 2008 13:44:04 +0200 From: Ingo Molnar To: Nick Piggin Cc: Jeremy Fitzhardinge , LKML , x86@kernel.org, Andi Kleen , Jens Axboe Subject: Re: [PATCH 0 of 9] x86/smp function calls: convert x86 tlb flushes to use function calls [POST 2] Message-ID: <20080819114404.GA5354@elte.hu> References: <200808191956.59898.nickpiggin@yahoo.com.au> <20080819102045.GC6722@elte.hu> <200808192108.13338.nickpiggin@yahoo.com.au> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200808192108.13338.nickpiggin@yahoo.com.au> User-Agent: Mutt/1.5.18 (2008-05-17) X-ELTE-VirusStatus: clean X-ELTE-SpamScore: -1.5 X-ELTE-SpamLevel: X-ELTE-SpamCheck: no X-ELTE-SpamVersion: ELTE 2.0 X-ELTE-SpamCheck-Details: score=-1.5 required=5.9 tests=BAYES_00 autolearn=no SpamAssassin version=3.2.3 -1.5 BAYES_00 BODY: Bayesian spam probability is 0 to 1% [score: 0.0000] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Nick Piggin wrote: > On Tuesday 19 August 2008 20:20, Ingo Molnar wrote: > > * Nick Piggin wrote: > > > I'm especially sore about mmap because I have a customer with a > > > database that uses a lot of mmap/munmap and those calls have slowed > > > down something like 50%(!!) from 2.6.early to 2.6.late. > > > > oh, agreed that it all needs to speed up - please send testcases ... > > I'm not sure if I can send the testcase I have. But this issue is > coming to the top of my list to investigate in mainline, so I'll check > out how we're doing here soon. obviously having a repeatable testcase would jump-start any efforts to fix this. It doesnt have to be complex - just a .c thing kernel developers can run and get some quick number out of. A 50% regression shouldnt be too hard to reproduce even with a vastly simplified workload. Ingo