From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Peverley Subject: Use of CONFIG_PROVE_LOCKING With RT patch Date: Fri, 7 Jan 2011 14:15:10 +0000 Message-ID: Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 To: linux-rt-users@vger.kernel.org Return-path: Received: from mail-gy0-f174.google.com ([209.85.160.174]:34370 "EHLO mail-gy0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753440Ab1AGOPL (ORCPT ); Fri, 7 Jan 2011 09:15:11 -0500 Received: by gyb11 with SMTP id 11so6576965gyb.19 for ; Fri, 07 Jan 2011 06:15:11 -0800 (PST) Sender: linux-rt-users-owner@vger.kernel.org List-ID: Hi all, I'm working with (a modified) linux-2.6.31.12-rt21 and have been trying to enable CONFIG_PROVE_LOCKING but for my build this comepletely stops the kernel coming up. Doing a bit of reading through the config, it seems that when I select PROVE_LOCKING my .config has the following added : CONFIG_DEBUG_MUTEXES=y CONFIG_DEBUG_LOCK_ALLOC=y CONFIG_PROVE_LOCKING=y CONFIG_LOCKDEP=y CONFIG_TRACE_IRQFLAGS=y So one of them is likely to be the individual cause. However, reading through the config I'm confused... If you read lib/Kconfig.debug you see that DEBUG_LOCK_ALLOC selects DEBUG_MUTEXES. However DEBUG_MUTEXES has : depends on DEBUG_KERNEL && !PREEMPT_RT BUT this kernel has PREEMPT_RT set.... Should the config system have actually set it? The real question of course is that if CONFIG_DEBUG_MUTEXES shouldn't be included for a PREEMPT_RT kernel, could this be part of the issue I'm seeing, i.e. its causing the lockup? Cheers, ~Pev