Re: [PATCH 1/1] softirq: Use a dedicated thread for timer wakeups on PREEMPT_RT.

From: Sebastian Andrzej Siewior
Date: Wed Oct 23 2024 - 02:30:27 EST


On 2024-10-23 00:27:34 [+0200], Frederic Weisbecker wrote:
> > Try again without the "ksoftirqd will collect it all" since this won't
> > happen since the revert I mentioned.
>
> I still don't get it, this makes:
>
> """
> Once the ksoftirqd is marked as pending (or is running), a softirq which
> would have been processed at the end of the threaded interrupt, which runs
> at an elevated priority, is now moved to ksoftirqd which runs at SCHED_OTHER
> priority and competes with every regular task for CPU resources.
> """
>
> ksoftirqd raised for timers still doesn't prevent a threaded IRQ from running
> softirqs, unless it preempts ksoftirqd and waits with PI. So is it what you're
> trying to solve?
>
> Or is the problem that timer softirqs are executed with SCHED_NORMAL?

Exactly. It runs at SCHED_NORMAL and competes with everything else. It
can delay tasks wakes depending on system load.

> > Quick question: Do we want this in forced-threaded mode, too? The timer
> > (timer_list timer) and all HRTIMER_MODE_SOFT are handled in ksoftirqd.
>
> It's hard to tell for me as I don't know the !RT usecases for forced-threaded mode.
> "If in doubt say N" ;-)

Oki.

> > > > +void raise_timer_softirq(void)
> > > > +{
> > > > + unsigned long flags;
> > > > +
> > > > + local_irq_save(flags);
> > > > + raise_ktimers_thread(TIMER_SOFTIRQ);
> > > > + wake_timersd();
> > >
> > > This is supposed to be called from hardirq only, right?
> > > Can't irq_exit_rcu() take care of it? Why is it different
> > > from HRTIMER_SOFTIRQ ?
> >
> > Good question. This shouldn't be any different compared to the hrtimer
> > case. This is only raised in hardirq, so yes, the irq_save can go away
> > and the wake call, too.
>
> Cool. You can add lockdep_assert_in_irq() within raise_ktimers_thread() for
> some well deserved relief :-)

If you want to, sure. I would add them to both raise functions.

> Thanks.

Sebastian