[RFC,ftrace] Chose RCU Tasks based on TASKS_RCU rather than PREEMPTION

Message ID f6507b10-5bb5-4407-bd4d-c547193a5a28@paulmck-laptop
State New
Headers
Series [RFC,ftrace] Chose RCU Tasks based on TASKS_RCU rather than PREEMPTION |

Commit Message

Paul E. McKenney Feb. 28, 2024, 7:38 p.m. UTC
  The advent of CONFIG_PREEMPT_AUTO, AKA lazy preemption, will mean that
even kernels built with CONFIG_PREEMPT_NONE or CONFIG_PREEMPT_VOLUNTARY
might see the occasional preemption, and that this preemption just might
happen within a trampoline.

Therefore, update ftrace_shutdown() to invoke synchronize_rcu_tasks()
based on CONFIG_TASKS_RCU instead of CONFIG_PREEMPTION.

Only build tested.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Ankur Arora <ankur.a.arora@oracle.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: <linux-trace-kernel@vger.kernel.org>
  

Comments

Steven Rostedt Feb. 28, 2024, 8:22 p.m. UTC | #1
On Wed, 28 Feb 2024 11:38:29 -0800
"Paul E. McKenney" <paulmck@kernel.org> wrote:

> The advent of CONFIG_PREEMPT_AUTO, AKA lazy preemption, will mean that
> even kernels built with CONFIG_PREEMPT_NONE or CONFIG_PREEMPT_VOLUNTARY
> might see the occasional preemption, and that this preemption just might
> happen within a trampoline.
> 
> Therefore, update ftrace_shutdown() to invoke synchronize_rcu_tasks()
> based on CONFIG_TASKS_RCU instead of CONFIG_PREEMPTION.
> 
> Only build tested.
> 
> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
> Cc: Steven Rostedt <rostedt@goodmis.org>
> Cc: Masami Hiramatsu <mhiramat@kernel.org>
> Cc: Mark Rutland <mark.rutland@arm.com>
> Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
> Cc: Ankur Arora <ankur.a.arora@oracle.com>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: <linux-trace-kernel@vger.kernel.org>
> 
> diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
> index 2da4eaa2777d6..c9e6c69cf3446 100644
> --- a/kernel/trace/ftrace.c
> +++ b/kernel/trace/ftrace.c
> @@ -3156,7 +3156,7 @@ int ftrace_shutdown(struct ftrace_ops *ops, int command)
>  		 * synchronize_rcu_tasks() will wait for those tasks to
>  		 * execute and either schedule voluntarily or enter user space.
>  		 */
> -		if (IS_ENABLED(CONFIG_PREEMPTION))
> +		if (IS_ENABLED(CONFIG_TASKS_RCU))
>  			synchronize_rcu_tasks();

What happens if CONFIG_TASKS_RCU is not enabled? Does
synchronize_rcu_tasks() do anything? Or is it just a synchronize_rcu()?

If that's the case, perhaps just remove the if statement and make it:

	synchronize_rcu_tasks();

Not sure an extra synchronize_rcu() will hurt (especially after doing a
synchronize_rcu_tasks_rude() just before hand!

-- Steve
  
Paul E. McKenney Feb. 28, 2024, 9:16 p.m. UTC | #2
On Wed, Feb 28, 2024 at 03:22:36PM -0500, Steven Rostedt wrote:
> On Wed, 28 Feb 2024 11:38:29 -0800
> "Paul E. McKenney" <paulmck@kernel.org> wrote:
> 
> > The advent of CONFIG_PREEMPT_AUTO, AKA lazy preemption, will mean that
> > even kernels built with CONFIG_PREEMPT_NONE or CONFIG_PREEMPT_VOLUNTARY
> > might see the occasional preemption, and that this preemption just might
> > happen within a trampoline.
> > 
> > Therefore, update ftrace_shutdown() to invoke synchronize_rcu_tasks()
> > based on CONFIG_TASKS_RCU instead of CONFIG_PREEMPTION.
> > 
> > Only build tested.
> > 
> > Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
> > Cc: Steven Rostedt <rostedt@goodmis.org>
> > Cc: Masami Hiramatsu <mhiramat@kernel.org>
> > Cc: Mark Rutland <mark.rutland@arm.com>
> > Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
> > Cc: Ankur Arora <ankur.a.arora@oracle.com>
> > Cc: Thomas Gleixner <tglx@linutronix.de>
> > Cc: <linux-trace-kernel@vger.kernel.org>
> > 
> > diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
> > index 2da4eaa2777d6..c9e6c69cf3446 100644
> > --- a/kernel/trace/ftrace.c
> > +++ b/kernel/trace/ftrace.c
> > @@ -3156,7 +3156,7 @@ int ftrace_shutdown(struct ftrace_ops *ops, int command)
> >  		 * synchronize_rcu_tasks() will wait for those tasks to
> >  		 * execute and either schedule voluntarily or enter user space.
> >  		 */
> > -		if (IS_ENABLED(CONFIG_PREEMPTION))
> > +		if (IS_ENABLED(CONFIG_TASKS_RCU))
> >  			synchronize_rcu_tasks();
> 
> What happens if CONFIG_TASKS_RCU is not enabled? Does
> synchronize_rcu_tasks() do anything? Or is it just a synchronize_rcu()?

It is just a synchronize_rcu().

> If that's the case, perhaps just remove the if statement and make it:
> 
> 	synchronize_rcu_tasks();
> 
> Not sure an extra synchronize_rcu() will hurt (especially after doing a
> synchronize_rcu_tasks_rude() just before hand!

That would work for me.  If there are no objections, I will make this
change.

							Thanx, Paul
  
Paul E. McKenney March 1, 2024, 8:25 p.m. UTC | #3
On Wed, Feb 28, 2024 at 01:16:04PM -0800, Paul E. McKenney wrote:
> On Wed, Feb 28, 2024 at 03:22:36PM -0500, Steven Rostedt wrote:
> > On Wed, 28 Feb 2024 11:38:29 -0800
> > "Paul E. McKenney" <paulmck@kernel.org> wrote:
> > 
> > > The advent of CONFIG_PREEMPT_AUTO, AKA lazy preemption, will mean that
> > > even kernels built with CONFIG_PREEMPT_NONE or CONFIG_PREEMPT_VOLUNTARY
> > > might see the occasional preemption, and that this preemption just might
> > > happen within a trampoline.
> > > 
> > > Therefore, update ftrace_shutdown() to invoke synchronize_rcu_tasks()
> > > based on CONFIG_TASKS_RCU instead of CONFIG_PREEMPTION.
> > > 
> > > Only build tested.
> > > 
> > > Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
> > > Cc: Steven Rostedt <rostedt@goodmis.org>
> > > Cc: Masami Hiramatsu <mhiramat@kernel.org>
> > > Cc: Mark Rutland <mark.rutland@arm.com>
> > > Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
> > > Cc: Ankur Arora <ankur.a.arora@oracle.com>
> > > Cc: Thomas Gleixner <tglx@linutronix.de>
> > > Cc: <linux-trace-kernel@vger.kernel.org>
> > > 
> > > diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
> > > index 2da4eaa2777d6..c9e6c69cf3446 100644
> > > --- a/kernel/trace/ftrace.c
> > > +++ b/kernel/trace/ftrace.c
> > > @@ -3156,7 +3156,7 @@ int ftrace_shutdown(struct ftrace_ops *ops, int command)
> > >  		 * synchronize_rcu_tasks() will wait for those tasks to
> > >  		 * execute and either schedule voluntarily or enter user space.
> > >  		 */
> > > -		if (IS_ENABLED(CONFIG_PREEMPTION))
> > > +		if (IS_ENABLED(CONFIG_TASKS_RCU))
> > >  			synchronize_rcu_tasks();
> > 
> > What happens if CONFIG_TASKS_RCU is not enabled? Does
> > synchronize_rcu_tasks() do anything? Or is it just a synchronize_rcu()?
> 
> It is just a synchronize_rcu().
> 
> > If that's the case, perhaps just remove the if statement and make it:
> > 
> > 	synchronize_rcu_tasks();
> > 
> > Not sure an extra synchronize_rcu() will hurt (especially after doing a
> > synchronize_rcu_tasks_rude() just before hand!
> 
> That would work for me.  If there are no objections, I will make this
> change.

But I did check the latency of synchronize_rcu_tasks_rude() (about 100ms)
and synchronize_rcu() (about 20ms).  This is on a 80-hardware-thread
x86 system that is being flooded with calls to one or the other of
these two functions, but is otherwise idle.  So adding that unnecessary
synchronize_rcu() adds about 20% to that synchronization delay.

Which might still be OK, but...  In the immortal words of MS-DOS,
"Are you sure?".  ;-)

							Thanx, Paul
  
Steven Rostedt March 1, 2024, 8:30 p.m. UTC | #4
On Fri, 1 Mar 2024 12:25:10 -0800
"Paul E. McKenney" <paulmck@kernel.org> wrote:

> > That would work for me.  If there are no objections, I will make this
> > change.  
> 
> But I did check the latency of synchronize_rcu_tasks_rude() (about 100ms)
> and synchronize_rcu() (about 20ms).  This is on a 80-hardware-thread
> x86 system that is being flooded with calls to one or the other of
> these two functions, but is otherwise idle.  So adding that unnecessary
> synchronize_rcu() adds about 20% to that synchronization delay.
> 
> Which might still be OK, but...  In the immortal words of MS-DOS,
> "Are you sure?".  ;-)

It's just safe to keep it. It's definitely not a fast path.

-- Steve
  
Paul E. McKenney March 2, 2024, 1:54 a.m. UTC | #5
On Fri, Mar 01, 2024 at 03:30:01PM -0500, Steven Rostedt wrote:
> On Fri, 1 Mar 2024 12:25:10 -0800
> "Paul E. McKenney" <paulmck@kernel.org> wrote:
> 
> > > That would work for me.  If there are no objections, I will make this
> > > change.  
> > 
> > But I did check the latency of synchronize_rcu_tasks_rude() (about 100ms)
> > and synchronize_rcu() (about 20ms).  This is on a 80-hardware-thread
> > x86 system that is being flooded with calls to one or the other of
> > these two functions, but is otherwise idle.  So adding that unnecessary
> > synchronize_rcu() adds about 20% to that synchronization delay.
> > 
> > Which might still be OK, but...  In the immortal words of MS-DOS,
> > "Are you sure?".  ;-)
> 
> It's just safe to keep it. It's definitely not a fast path.

OK, you got it!  ;-)

							Thanx, Paul
  

Patch

diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 2da4eaa2777d6..c9e6c69cf3446 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -3156,7 +3156,7 @@  int ftrace_shutdown(struct ftrace_ops *ops, int command)
 		 * synchronize_rcu_tasks() will wait for those tasks to
 		 * execute and either schedule voluntarily or enter user space.
 		 */
-		if (IS_ENABLED(CONFIG_PREEMPTION))
+		if (IS_ENABLED(CONFIG_TASKS_RCU))
 			synchronize_rcu_tasks();
 
 		ftrace_trampoline_free(ops);