Re: [PATCH 2/2] rcutorture: Nudge ksoftirqd priority for RCU boost testing

On Thu, Aug 05, 2021 at 12:51:48PM +0100, Valentin Schneider wrote:

On 04/08/21 15:53, Paul E. McKenney wrote:

quoted

On Wed, Aug 04, 2021 at 11:18:11AM +0100, Valentin Schneider wrote:

quoted

Hm so v5.13-rt1 has this commit:

  5e59fba573e6 ("rcutorture: Fix testing of RCU priority boosting")

which gates RCU boost torture testing under CONFIG_PREEMPT_RT. Now, AFAICT
the TIMER_SOFTIRQ priority problem is there regardless of
CONFIG_PREEMPT_RT, so this patch would (should?) make sense even on
!CONFIG_PREEMPT_RT.

What rcutorture scenario TREE03 does is to boot with tree.use_softirq=0
and threadirqs.  I see your point about timers and softirq, but this
does run reliably for me.

Ah, I see why.  Commit ea6d962e80b6 ("rcutorture: Judge RCU priority
boosting on grace periods, not callbacks") includes boosting the priority
of the ksoftirqd kthreads.  But only when running rcutorture builtin,
not as a module.  Here is the code in rcu_torture_init():

              // Testing RCU priority boosting requires rcutorture do
              // some serious abuse.  Counter this by running ksoftirqd
              // at higher priority.
              if (IS_BUILTIN(CONFIG_RCU_TORTURE_TEST)) {
                      for_each_online_cpu(cpu) {
                              struct sched_param sp;
                              struct task_struct *t;

                              t = per_cpu(ksoftirqd, cpu);
                              WARN_ON_ONCE(!t);
                              sp.sched_priority = 2;
                              sched_setscheduler_nocheck(t, SCHED_FIFO, &sp);
                      }
              }

I take it that you were running rcutorture as a module?

This describes how to run it built-in, if that works for you:

https://paulmck.livejournal.com/61432.html

More specifically: https://paulmck.livejournal.com/57769.html

Alternatively, the "IS_BUILTIN(CONFIG_RCU_TORTURE_TEST)" check could be
removed in the above code, and the ksoftirqd kthreads could have their
original priority restored in rcu_torture_cleanup().

Thoughts?

I actually run rcutorture as a builtin, but from what I can tell the above
patch came in v5.14-rc1, and ofc I'm running my tests on v5.13-rt1... I
should have paid closer attention to what was in the latest mainline,
apologies for the noise. 

Not a problem, and thank you for giving rcutorture a try!

FWIW tweaking ksoftirqd priority only when the torture module is builtin
makes sense to me.

Very good, I will stick with the status quo, then.

							Thanx, Paul