From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S964938AbbLGXZi (ORCPT ); Mon, 7 Dec 2015 18:25:38 -0500 Received: from mail.skyhub.de ([78.46.96.112]:38269 "EHLO mail.skyhub.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932555AbbLGXZg (ORCPT ); Mon, 7 Dec 2015 18:25:36 -0500 Date: Tue, 8 Dec 2015 00:25:24 +0100 From: Borislav Petkov To: "Raj, Ashok" Cc: "Luck, Tony" , "linux-kernel@vger.kernel.org" , "linux-edac@vger.kernel.org" Subject: Re: [Patch V2] x86, mce: Ensure offline CPU's don't participate in mce rendezvous process. Message-ID: <20151207232524.GK22248@pd.tnic> References: <20151205002930.GA24005@otc-brkl-03.jf.intel.com> <20151207200019.GH22248@pd.tnic> <3908561D78D1C84285E8C5FCA982C28F39F7C24B@ORSMSX114.amr.corp.intel.com> <20151207201951.GI22248@pd.tnic> <3908561D78D1C84285E8C5FCA982C28F39F7C3A4@ORSMSX114.amr.corp.intel.com> <20151207223427.GJ22248@pd.tnic> <20151207234639.GA81526@otc-brkl-03.jf.intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <20151207234639.GA81526@otc-brkl-03.jf.intel.com> User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Dec 07, 2015 at 06:46:40PM -0500, Raj, Ashok wrote: > On Mon, Dec 07, 2015 at 11:34:27PM +0100, Borislav Petkov wrote: > > > > Box logs below. > > Do you have the dmidecode strings to find which platform this is? Is this enough or you want complete dmidecode dump? DMI: Intel Corporation LH Pass/S4600LH...., BIOS SE5C600.86B.99.99.2050.043020121425 04/30/2012 > Not sure how the fix works.. since we excluded only the ones offline. > So unless all online cpu's check in, the code should give you the old > behavior. Did you miss my statement in my previous mail where I said that the MCE is being raised only on the cores of node 0? > What does cat /proc/interrupts | grep MCE Can't. Shell on the box is dead after the injection. > In a system broadcasting, all cpu counts should be the same. Since we didn't > increment the offline stats, if you were to bring the cpu up, it should be one > less than other cpus... See the logs at the end of my previous email. #MC gets raised - or at least output from mce_panic() comes out only - on the cores of node 0. -- Regards/Gruss, Boris. ECO tip #101: Trim your mails when you reply.