Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758053AbZC3Jl4 (ORCPT ); Mon, 30 Mar 2009 05:41:56 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755832AbZC3Jlr (ORCPT ); Mon, 30 Mar 2009 05:41:47 -0400 Received: from mga10.intel.com ([192.55.52.92]:38959 "EHLO fmsmga102.fm.intel.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1755477AbZC3Jlq (ORCPT ); Mon, 30 Mar 2009 05:41:46 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.38,445,1233561600"; d="scan'208";a="677096662" Message-ID: <49D093F4.2080202@linux.intel.com> Date: Mon, 30 Mar 2009 11:42:12 +0200 From: Andi Kleen User-Agent: Thunderbird 2.0.0.21 (Windows/20090302) MIME-Version: 1.0 To: Ingo Molnar CC: Hidetoshi Seto , "H. Peter Anvin" , Thomas Gleixner , linux-kernel@vger.kernel.org Subject: Re: [PATCH -tip 1/3] x86, mce: Add mce_threshold option for intel cmci References: <49CB3F24.8040804@jp.fujitsu.com> <20090328120825.GB14464@elte.hu> In-Reply-To: <20090328120825.GB14464@elte.hu> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1319 Lines: 32 > Makes sense. CMCI is a new CPU feature so having boot controls to > disable it is generally a good idea - while still having old-style > MCE support in place. That's not what the patch does unfortunately. > Applied to tip:x86/mce2, thanks Hidetoshi! If you fear that CMCI is misbehaving you would need a different option to turn it off completely, setting the threshold doesn't help. For example there are banks that only support threshold == 1 and when the enable bit is on then no matter what value you write in there events will cause CMCIs. So this patch cannot turn them off. To turn it off you would need to disable the CMCI enable bit completely. I have no problems in principle with a mce=nocmci that does that but that would be a different patch. However I expect that this will be not a good idea to ever use on Nehalem class systems at least because without CMCI the machine check code cannot handle shared banks correctly and you'll get duplicated events from them. And on non Nehalem systems there is no CMCI anyways, so it'll be always off. -Andi -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/