From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934666AbZFOVMY (ORCPT ); Mon, 15 Jun 2009 17:12:24 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S934568AbZFOVMK (ORCPT ); Mon, 15 Jun 2009 17:12:10 -0400 Received: from tomts43.bellnexxia.net ([209.226.175.110]:61542 "EHLO tomts43-srv.bellnexxia.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932232AbZFOVMI (ORCPT ); Mon, 15 Jun 2009 17:12:08 -0400 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AokFAC9TNkpMQWQl/2dsb2JhbACBT9NshA0F Date: Mon, 15 Jun 2009 17:12:07 -0400 From: Mathieu Desnoyers To: Ingo Molnar Cc: "H. Peter Anvin" , Peter Zijlstra , Linus Torvalds , mingo@redhat.com, paulus@samba.org, acme@redhat.com, linux-kernel@vger.kernel.org, penberg@cs.helsinki.fi, vegard.nossum@gmail.com, efault@gmx.de, jeremy@goop.org, npiggin@suse.de, tglx@linutronix.de, linux-tip-commits@vger.kernel.org Subject: Re: [tip:perfcounters/core] perf_counter: x86: Fix call-chain support to use NMI-safe methods Message-ID: <20090615211207.GB12919@Krystal> References: <4A369508.2090707@zytor.com> <20090615184858.GD6520@Krystal> <1245091917.6741.185.camel@laptop> <20090615185907.GF6520@Krystal> <1245092561.6741.205.camel@laptop> <4A369CD8.3090505@zytor.com> <20090615192720.GA9056@Krystal> <4A36A1C7.6080005@zytor.com> <20090615210119.GD24554@elte.hu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline In-Reply-To: <20090615210119.GD24554@elte.hu> X-Editor: vi X-Info: http://krystal.dyndns.org:8080 X-Operating-System: Linux/2.6.21.3-grsec (i686) X-Uptime: 17:05:36 up 107 days, 17:31, 3 users, load average: 0.06, 0.14, 0.27 User-Agent: Mutt/1.5.18 (2008-05-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Ingo Molnar (mingo@elte.hu) wrote: > > * H. Peter Anvin wrote: > > > Mathieu Desnoyers wrote: > > >>> > > >> Writing control registers is serializing, so it's a lot more expensive > > >> than writing a normal register; my *guess* is that it will be on the > > >> order of 100-200 cycles. > > >> > > >> That is not based on any actual information. > > >> > > > > > > Then how about just writing to the cr2 register *if* it has changed > > > while the NMI handler was running ? > > > > > > if (unlikely(read_cr2() != saved_cr2))) > > > write_cr2(saved_cr2) > > > > > > Mathieu > > > > > > > That works fine, obviously, and although it's probably overkill > > it's also a trivially cheap optimization. > > Writing cr2 costs 84 cycles so it's not overkill at all - it's a > nice optimization! > > Btw., we dont have to re-read it - we actually _know_ when we got a > fault (the fault handler gives us back an error code). > Just for the sake of making NMI handlers less tricky, supporting page faults caused by faulting kernel instructions (rather than only supporting explicit faulting from get_user_pages_inatomic) would be rather nice design-wise if it only costs 2-3 cycles. And I would not want to touch the page fault handler itself to write the saved cr2 value before the handler exits, because this would add a branch on a very hot path. Mathieu > So we can do this common optimization and avoid the cr2 write > usually. We only need the cr2 read. > > Hm, tempting ... > > Ingo -- Mathieu Desnoyers OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68