From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753367Ab0JLGOn (ORCPT ); Tue, 12 Oct 2010 02:14:43 -0400 Received: from mga01.intel.com ([192.55.52.88]:7237 "EHLO mga01.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752655Ab0JLGOn (ORCPT ); Tue, 12 Oct 2010 02:14:43 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.57,319,1283756400"; d="scan'208";a="615806464" Subject: Re: [PATCH -v3 3/6] x86, NMI, Rewrite NMI handler From: Huang Ying To: Peter Zijlstra Cc: Don Zickus , Ingo Molnar , "H. Peter Anvin" , "linux-kernel@vger.kernel.org" , Andi Kleen , Robert Richter In-Reply-To: <1286863479.2336.1451.camel@twins> References: <1286606987-19879-1-git-send-email-ying.huang@intel.com> <1286606987-19879-3-git-send-email-ying.huang@intel.com> <1286813635.2336.617.camel@twins> <1286844650.7768.137.camel@yhuang-dev> <1286863479.2336.1451.camel@twins> Content-Type: text/plain; charset="UTF-8" Date: Tue, 12 Oct 2010 14:14:40 +0800 Message-ID: <1286864080.7768.179.camel@yhuang-dev> Mime-Version: 1.0 X-Mailer: Evolution 2.30.2 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 2010-10-12 at 14:04 +0800, Peter Zijlstra wrote: > On Tue, 2010-10-12 at 08:50 +0800, Huang Ying wrote: > > On Tue, 2010-10-12 at 00:13 +0800, Peter Zijlstra wrote: > > > On Sat, 2010-10-09 at 14:49 +0800, Huang Ying wrote: > > > > notify_die(DIE_NMI_IPI); > > > > notify_die(DIE_NMI); > > > > /* process io port 0x61 */ > > > > nmi_watchdog_touch(); > > > > unknown_nmi(); > > > > > > Why keep NMI_IPI? What the heck is it for? > > > > DIE_NMI_IPI is used for CPU-specific or CPU-local NMIs, such as perf > > NMI. While DIE_NMI is used for non-CPU-specific or global NMIs, such as > > NMI notification from source bridge. > > > > The order between these two is important. So we use two die value to > > enforce the order. > > But you can't know about that, there is no reason field to distinguish > between these cases, so you might as well fold it into a single notifier > chain and be done with it. NMI users know that. Such as perf uses CPU-specific NMI, while APEI GHES uses non-CPU-specific NMI. Different users expect different die values, such as perf expects DIE_NMI_IPI, while APEI GHES expects DIE_NMI, so that perf can be checked before APEI GHES. > There is no good reason to call two chains when one is enough. > > Also, the IPI name is terrible, its not IPI related at all. Please kill > the thing. I agree with you that IPI is not a good name. Can you suggest a better name? Best Regards, Huang Ying