From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932546Ab1DNJ1f (ORCPT ); Thu, 14 Apr 2011 05:27:35 -0400 Received: from mail-pw0-f46.google.com ([209.85.160.46]:38045 "EHLO mail-pw0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757316Ab1DNJ1e convert rfc822-to-8bit (ORCPT ); Thu, 14 Apr 2011 05:27:34 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type :content-transfer-encoding; b=edEgcVsgOA4lQhYU/Q4MvDBdg/xHOAzBVsqmOftTWgsJqQjQWM+3jjgjynWM6IccrL f1zXQiZG4uRMt7aWjisVU5NcA3ILbgGhDnKTc5Xx+g5nhYZT+wrupawLmU//KM5gUarz s6l8Oegs9mS/Lg66RBMOJzq9ujK316K5qhu28= MIME-Version: 1.0 In-Reply-To: <20110414080506.GA23965@elte.hu> References: <20110406223036.GA15721@digium.com> <201104132133.51958.maciej.rutecki@gmail.com> <4DA6011F.7070405@openvz.org> <20110414064737.GB15535@elte.hu> <20110414080506.GA23965@elte.hu> Date: Thu, 14 Apr 2011 13:27:33 +0400 X-Google-Sender-Auth: xLymfwo8TQpKUeDP328QTgufWjY Message-ID: Subject: Re: [regression 2.6.39-rc2][bisected] "perf, x86: P4 PMU - Read proper MSR register to catch" and NMIs From: Cyrill Gorcunov To: Ingo Molnar Cc: Peter Zijlstra , maciej.rutecki@gmail.com, Shaun Ruffell , Don Zickus , linux-kernel@vger.kernel.org, Lin Ming Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Apr 14, 2011 at 12:05 PM, Ingo Molnar wrote: ... >> If there is no counters overflowed I believe we should not poke LVTPC until >> we sure NMI comes from it (and counter overflow is the only sign that NMI >> came from LVTPC as far as I may say, and I see also a possibility for race if >> counter signal reaches LVTPC and it is being processed inside apic chip >> {which might take some time too before real NMI signal appears in cpu} and as >> result hard to tell what we get in output -- double nmi again or something >> else). > > Well, we unmasked unconditionally before. If we unmask conditionally now, we > risk not unmasking. We risk a completely stuck PMU (there wont ever come *any* > NMI from it if we ever forget to unmask) versus spurious NMIs. > > Maybe we can do it - but it will need a lot of testing on a lot of CPU types to > make sure there's no other CPU quirks in this area ... > > So unless the conditional unmasking fixes a real bug (in kgdb or elsewhere) > lets unmask unconditionally now to fix the P4 regression in .39 - and queue up > a *separate* patch that moves it even further down and makes it conditional - > but queue that up for .40. > > Thanks, > >        Ingo > OK. Ingo I'll send a patch from Don with all tested-by (including me) and my ack as only get back home. (I don't mind if Don beat me on this ;)