From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755259AbaEHRfL (ORCPT ); Thu, 8 May 2014 13:35:11 -0400 Received: from mail-ee0-f41.google.com ([74.125.83.41]:62879 "EHLO mail-ee0-f41.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755022AbaEHRfH (ORCPT ); Thu, 8 May 2014 13:35:07 -0400 Date: Thu, 8 May 2014 19:35:01 +0200 From: Ingo Molnar To: Don Zickus Cc: x86@kernel.org, Peter Zijlstra , ak@linux.intel.com, gong.chen@linux.intel.com, LKML , Thomas Gleixner , =?iso-8859-1?Q?Fr=E9d=E9ric?= Weisbecker , Steven Rostedt , andi@firstfloor.org Subject: Re: [PATCH 1/5] x86, nmi: Add new nmi type 'external' Message-ID: <20140508173501.GA9838@gmail.com> References: <1399476883-98970-1-git-send-email-dzickus@redhat.com> <1399476883-98970-2-git-send-email-dzickus@redhat.com> <20140507153854.GA14926@gmail.com> <20140507160251.GQ39568@redhat.com> <20140507162746.GA15779@gmail.com> <20140508163333.GZ39568@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20140508163333.GZ39568@redhat.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Don Zickus wrote: > > > Again, I don't have a solution to juggle between PMI performance > > > and reliable delivery. We could do away with the spinlocks and > > > go back to single cpu delivery (like it used to be). Then > > > devise a mechanism to switch delivery to another cpu upon > > > hotplug. > > > > > > Thoughts? > > > > I'd say we should do a delayed timer that makes sure that all > > possible handlers are polled after an NMI is triggered, but never > > at a high rate. > > Hmm, I was thinking about it and wanted to avoid a poll as I hear > complaints here and there about the nmi_watchdog constantly wasting > power cycles with its polling. But the polling would only happen if there's NMI traffic, so that's fine. So as long as polling stops some time after the last PMI use, it's a good solution. This would also address a lot of NMI handling related fragility. Thanks, Ingo