From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753628Ab3FELwD (ORCPT ); Wed, 5 Jun 2013 07:52:03 -0400 Received: from merlin.infradead.org ([205.233.59.134]:49796 "EHLO merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752543Ab3FELwB (ORCPT ); Wed, 5 Jun 2013 07:52:01 -0400 Date: Wed, 5 Jun 2013 13:51:47 +0200 From: Peter Zijlstra To: Steven Rostedt Cc: "Paul E. McKenney" , LKML , Tejun Heo , Ingo Molnar , Frederic Weisbecker , Jiri Olsa Subject: Re: [RFC][PATCH] ftrace: Use schedule_on_each_cpu() as a heavy synchronize_sched() Message-ID: <20130605115147.GF8923@twins.programming.kicks-ass.net> References: <1369785676.15552.55.camel@gandalf.local.home> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1369785676.15552.55.camel@gandalf.local.home> User-Agent: Mutt/1.5.21 (2012-12-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, May 28, 2013 at 08:01:16PM -0400, Steven Rostedt wrote: > The function tracer uses preempt_disable/enable_notrace() for > synchronization between reading registered ftrace_ops and unregistering > them. > > Most of the ftrace_ops are global permanent structures that do not > require this synchronization. That is, ops may be added and removed from > the hlist but are never freed, and wont hurt if a synchronization is > missed. > > But this is not true for dynamically created ftrace_ops or control_ops, > which are used by the perf function tracing. > > The problem here is that the function tracer can be used to trace > kernel/user context switches as well as going to and from idle. > Basically, it can be used to trace blind spots of the RCU subsystem. > This means that even though preempt_disable() is done, a > synchronize_sched() will ignore CPUs that haven't made it out of user > space or idle. These can include functions that are being traced just > before entering or exiting the kernel sections. > > To implement the RCU synchronization, instead of using > synchronize_sched() the use of schedule_on_each_cpu() is performed. This > means that when a dynamically allocated ftrace_ops, or a control ops is > being unregistered, all CPUs must be touched and execute a ftrace_sync() > stub function via the work queues. This will rip CPUs out from idle or > in dynamic tick mode. This only happens when a user disables perf > function tracing or other dynamically allocated function tracers, but it > allows us to continue to debug RCU and context tracking with function > tracing. > > Signed-off-by: Steven Rostedt > > Index: linux-trace.git/kernel/trace/ftrace.c > =================================================================== > --- linux-trace.git.orig/kernel/trace/ftrace.c > +++ linux-trace.git/kernel/trace/ftrace.c > @@ -413,6 +413,17 @@ static int __register_ftrace_function(st > return 0; > } > > +static void ftrace_sync(struct work_struct *work) > +{ > + /* > + * This function is just a stub to implement a hard force > + * of synchronize_sched(). This requires synchronizing > + * tasks even in userspace and idle. > + * > + * Yes, function tracing is rude. > + */ > +} > + > static int __unregister_ftrace_function(struct ftrace_ops *ops) > { > int ret; > @@ -440,8 +451,12 @@ static int __unregister_ftrace_function( > * so there'll be no new users. We must ensure > * all current users are done before we free > * the control data. > + * Note synchronize_sched() is not enough, as we > + * use preempt_disable() to do RCU, but the function > + * tracer can be called where RCU is not active > + * (before user_exit()). > */ > - synchronize_sched(); > + schedule_on_each_cpu(ftrace_sync); > control_ops_free(ops); > } > } else > @@ -456,9 +471,13 @@ static int __unregister_ftrace_function( > /* > * Dynamic ops may be freed, we must make sure that all > * callers are done before leaving this function. > + * > + * Again, normal synchronize_sched() is not good enough. > + * We need to do a hard force of sched synchronization. > */ > if (ops->flags & FTRACE_OPS_FL_DYNAMIC) > - synchronize_sched(); > + schedule_on_each_cpu(ftrace_sync); > + > > return 0; > } > So what happens if schedule_on_each_cpu() returns -ENOMEM? :-)