From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Gleixner Subject: Re: rt5, sleeping function called from invalid context Date: Thu, 28 Jul 2011 21:45:29 +0200 (CEST) Message-ID: References: <4E31B6F3.80602@am.sony.com> <4E31BA74.3070102@am.sony.com> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Cc: "Rowand, Frank" , "linux-rt-users@vger.kernel.org" To: Frank Rowand Return-path: Received: from www.linutronix.de ([62.245.132.108]:48318 "EHLO Galois.linutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754565Ab1G1Tpc (ORCPT ); Thu, 28 Jul 2011 15:45:32 -0400 In-Reply-To: <4E31BA74.3070102@am.sony.com> Sender: linux-rt-users-owner@vger.kernel.org List-ID: On Thu, 28 Jul 2011, Frank Rowand wrote: > On 07/28/11 12:22, Frank Rowand wrote: > > Thomas, > > > > Here is one of my current BUGs, from rt5 (same panda board, same config): > > And some from rt3 (I don't know if they are still valid, just documenting > them because they reproduce reliably on rt3): > > --------------------------------------------------------------------------- > > Running bash on the console: > > BUG: sleeping function called from invalid context at kernel/rtmutex.c:645 > in_atomic(): 1, irqs_disabled(): 128, pid: 1, name: swapper > 3 locks held by swapper/1: > #0: (&sig->cred_guard_mutex){+.+.+.}, at: [] prepare_bprm_creds+0x20/0x58 > #1: (&p->alloc_lock){+.+...}, at: [] flush_old_exec+0x2dc/0x5d4 > #2: (cpu_asid_lock){......}, at: [] __new_context+0x14/0xbc > irq event stamp: 475085 > hardirqs last enabled at (475084): [] _raw_spin_unlock_irqrestore+0x60/0x74 > hardirqs last disabled at (475085): [] flush_old_exec+0x2e4/0x5d4 > softirqs last enabled at (0): [] copy_process+0x364/0xf84 > softirqs last disabled at (0): [< (null)>] (null) > [] (unwind_backtrace+0x0/0xf0) from [] (rt_spin_lock+0x24/0x5c) > [] (rt_spin_lock+0x24/0x5c) from [] (set_mm_context+0x1c/0x5c) > [] (set_mm_context+0x1c/0x5c) from [] (__new_context+0x54/0xbc) Fixed in rt4/5 already.