From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1761344AbZCTTf2 (ORCPT ); Fri, 20 Mar 2009 15:35:28 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751101AbZCTTfU (ORCPT ); Fri, 20 Mar 2009 15:35:20 -0400 Received: from bombadil.infradead.org ([18.85.46.34]:46355 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751898AbZCTTfT (ORCPT ); Fri, 20 Mar 2009 15:35:19 -0400 Subject: cpu hotplug and lockdep (was: Re: [tip:core/ipi] generic-ipi: eliminate spurious pointlessWARN_ON()s) From: Peter Zijlstra To: Ingo Molnar Cc: "Eric W. Biederman" , Jan Beulich , tglx@linutronix.de, mingo@redhat.com, hpa@zytor.com, linux-kernel@vger.kernel.org, Gautham R Shenoy In-Reply-To: <1237575134.4667.5.camel@laptop> References: <49B91A7E.76E4.0078.0@novell.com> <1236934491.5188.209.camel@laptop> <49BA33BE.76E4.0078.0@novell.com> <1236937423.22914.3698.camel@twins> <20090313103828.GB31094@elte.hu> <20090320085205.GB16021@elte.hu> <20090320182404.GA31629@elte.hu> <1237575134.4667.5.camel@laptop> Content-Type: text/plain Date: Fri, 20 Mar 2009 20:34:48 +0100 Message-Id: <1237577688.4667.68.camel@laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.26.0 Content-Transfer-Encoding: 7bit X-Bad-Reply: References and In-Reply-To but no 'Re:' in Subject. Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 2009-03-20 at 19:52 +0100, Peter Zijlstra wrote: > On Fri, 2009-03-20 at 19:24 +0100, Ingo Molnar wrote: > > * Eric W. Biederman wrote: > > > > > Now back to my regularly scheduled hotunplug challenges starting > > > with getting lockdep to warn about the dead locks. > > http://programming.kicks-ass.net/kernel-patches/cpu-hotplug/ > > The recursive read code needs more work.. You'll find some network softirq inversion 'fixes' there, but those are false, (soft)irq recursion isn't a problem for recursive read locks, but of course only as long as writes (and thereby all nesting exclusive locks) are irq-safe so as to not create write in read deadlocks. Just never gotten around to actually implementing that, ego also offered to look at that at one point, but I guess he didn't get around to it either ;-)