From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933723AbbGGTDo (ORCPT ); Tue, 7 Jul 2015 15:03:44 -0400 Received: from smtprelay0108.hostedemail.com ([216.40.44.108]:55286 "EHLO smtprelay.hostedemail.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S932830AbbGGTDf (ORCPT ); Tue, 7 Jul 2015 15:03:35 -0400 X-Session-Marker: 726F737465647440676F6F646D69732E6F7267 X-Spam-Summary: 2,0,0,,d41d8cd98f00b204,rostedt@goodmis.org,:::::,RULES_HIT:41:355:379:541:599:800:960:973:981:988:989:1260:1277:1311:1313:1314:1345:1359:1437:1515:1516:1518:1534:1542:1593:1594:1711:1730:1747:1777:1792:2393:2553:2559:2562:2693:3138:3139:3140:3141:3142:3353:3622:3865:3866:3867:3868:3871:3872:4321:4605:5007:6119:6261:7875:7903:10004:10400:10848:10967:11026:11232:11473:11658:11914:12043:12291:12296:12438:12517:12519:12555:12663:12679:12683:12740:13095:21060:21080,0,RBL:none,CacheIP:none,Bayesian:0.5,0.5,0.5,Netcheck:none,DomainCache:0,MSF:not bulk,SPF:fn,MSBL:0,DNSBL:none,Custom_rules:0:0:0 X-HE-Tag: pain66_764dbea03672c X-Filterd-Recvd-Size: 3452 Date: Tue, 7 Jul 2015 15:03:31 -0400 From: Steven Rostedt To: Fengguang Wu Cc: LKML , LKP Subject: Re: [trace] kernel BUG at kernel/sched/core.c:2881! Message-ID: <20150707150331.6181f431@gandalf.local.home> In-Reply-To: <20150707110627.2a638539@gandalf.local.home> References: <20150630141803.GA28071@wfg-t540p.sh.intel.com> <20150707110627.2a638539@gandalf.local.home> X-Mailer: Claws Mail 3.11.1 (GTK+ 2.24.28; x86_64-pc-linux-gnu) MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 7 Jul 2015 11:06:27 -0400 Steven Rostedt wrote: > On Tue, 30 Jun 2015 22:18:03 +0800 > Fengguang Wu wrote: > > > Hi Rostedt, > > > > FYI, this merge changes kernel crash to some more obvious kernel BUG > > message. If it's still not helpful, I can try bisect the old crash > > or split up the merge and bisect into them. > > > > git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master > > > > This has been a bug for some time. Does this fix it? > Test this patch instead, this is the one I'm testing and will be pushing to Linus. -- Steve --- kernel/trace/trace.h | 1 + kernel/trace/trace_branch.c | 13 ++++++++----- 2 files changed, 9 insertions(+), 5 deletions(-) Index: linux-trace.git/kernel/trace/trace.h =================================================================== --- linux-trace.git.orig/kernel/trace/trace.h 2015-07-07 14:28:27.391715136 -0400 +++ linux-trace.git/kernel/trace/trace.h 2015-07-07 14:29:08.133213493 -0400 @@ -444,6 +444,7 @@ enum { TRACE_CONTROL_BIT, + TRACE_BRANCH_BIT, /* * Abuse of the trace_recursion. * As we need a way to maintain state if we are tracing the function Index: linux-trace.git/kernel/trace/trace_branch.c =================================================================== --- linux-trace.git.orig/kernel/trace/trace_branch.c 2015-07-07 14:28:27.391715136 -0400 +++ linux-trace.git/kernel/trace/trace_branch.c 2015-07-07 14:56:52.036764240 -0400 @@ -36,9 +36,12 @@ probe_likely_condition(struct ftrace_bra struct trace_branch *entry; struct ring_buffer *buffer; unsigned long flags; - int cpu, pc; + int pc; const char *p; + if (current->trace_recursion & TRACE_BRANCH_BIT) + return; + /* * I would love to save just the ftrace_likely_data pointer, but * this code can also be used by modules. Ugly things can happen @@ -50,9 +53,9 @@ probe_likely_condition(struct ftrace_bra return; local_irq_save(flags); - cpu = raw_smp_processor_id(); - data = per_cpu_ptr(tr->trace_buffer.data, cpu); - if (atomic_inc_return(&data->disabled) != 1) + current->trace_recursion |= TRACE_BRANCH_BIT; + data = this_cpu_ptr(tr->trace_buffer.data); + if (atomic_read(&data->disabled)) goto out; pc = preempt_count(); @@ -81,7 +84,7 @@ probe_likely_condition(struct ftrace_bra __buffer_unlock_commit(buffer, event); out: - atomic_dec(&data->disabled); + current->trace_recursion &= ~TRACE_BRANCH_BIT; local_irq_restore(flags); }