From: Joel Fernandes <joel@joelfernandes.org>
To: "Paul E. McKenney" <paulmck@kernel.org>
Cc: Frederic Weisbecker <frederic@kernel.org>, rcu@vger.kernel.org
Subject: Re: [BUG] Random intermittent boost failures (Was Re: [BUG] TREE04..)
Date: Thu, 14 Sep 2023 18:56:27 +0000 [thread overview]
Message-ID: <20230914185627.GA2520229@google.com> (raw)
In-Reply-To: <885bb95b-9068-45f9-ba46-3feb650a3c45@paulmck-laptop>
On Thu, Sep 14, 2023 at 08:23:38AM -0700, Paul E. McKenney wrote:
> On Thu, Sep 14, 2023 at 01:13:51PM +0000, Joel Fernandes wrote:
> > On Thu, Sep 14, 2023 at 04:11:26AM -0700, Paul E. McKenney wrote:
> > > On Wed, Sep 13, 2023 at 04:30:20PM -0400, Joel Fernandes wrote:
> > > > On Mon, Sep 11, 2023 at 4:16 AM Paul E. McKenney <paulmck@kernel.org> wrote:
> > > > [..]
> > > > > > I am digging deeper to see why the rcu_preempt thread cannot be pushed out
> > > > > > and then I'll also look at why is it being pushed out in the first place.
> > > > > >
> > > > > > At least I have a strong repro now running 5 instances of TREE03 in parallel
> > > > > > for several hours.
> > > > >
> > > > > Very good! Then why not boot with rcutorture.onoff_interval=0 and see if
> > > > > the problem still occurs? If yes, then there is definitely some reason
> > > > > other than CPU hotplug that makes this happen.
> > > >
> > > > Hi Paul,
> > > > So looks so far like onoff_interval=0 makes the issue disappear. So
> > > > likely hotplug related. I am ok with doing the cpus_read_lock during
> > > > boost testing and seeing if that fixes it. If it does, I can move on
> > > > to the next thing in my backlog.
> > > >
> > > > What do you think? Or should I spend more time root-causing it? It is
> > > > most like runaway RT threads combined with the CPU hotplug threads,
> > > > making scheduling of the rcu_preempt thread not happen. But I can't
> > > > say for sure without more/better tracing (Speaking of better tracing,
> > > > I am adding core-dump support to rcutorture, but it is not there yet).
> > >
> > > This would not be the first time rcutorture has had trouble with those
> > > threads, so I am for adding the cpus_read_lock().
> > >
> > > Additional root-causing might be helpful, but then again, you might
> > > have higher priority things to worry about. ;-)
> >
> > No worries. Unfortunately putting cpus_read_lock() around the boost test
> > causes hangs. I tried something like the following [1]. If you have a diff, I can
> > quickly try something to see if the issue goes away as well.
>
> The other approaches that occur to me are:
>
> 1. Synchronize with the torture.c CPU-hotplug code. This is a bit
> tricky as well.
>
> 2. Rearrange the testing to convert one of the TREE0* scenarios that
> is not in CFLIST (TREE06 or TREE08) to a real-time configuration,
> with boosting but without CPU hotplug. Then remove boosting
> from TREE04.
>
> Of these, #2 seems most productive. But is there a better way?
We could have the gp thread at higher priority for TREE03. What I see
consistently is that the GP thread gets migrated from CPU M to CPU N only to
be immediately sent back. Dumping the state showed CPU N is running ksoftirqd
which is also a rt priority 2. Making rcu_preempt 3 and ksoftirqd 2 might
give less of a run-around to rcu_preempt maybe enough to prevent the grace
period from stalling. I am not sure if this will fix it, but I am running a
test to see how it goes, will let you know.
thanks,
- Joel
>
> > I am also interested in tracing further the state of CPUs when the back and
> > forth migration happens. That code is really hairy (select_fallback_rq())
> > with some scary code comments about more audit being needed, so I would not
> > be surprised if something is off there. I left some trace prints in there [2]
> > to see if I can gather more information. I'll let you know if I find
> > anything there.
>
> Looking forward to seeing what you come up with!
>
> > [ btw I'm officially OOO from Mon-Thurs next week, that's mostly Ok since the
> > last set stable releases just got done. Though I might cheat and attend to
> > low-hanging fruit type of tasks like documentation during the OOO :-D ]
>
> Have a good time away!
>
> Thanx, Paul
>
> > [1]
> > diff --git a/kernel/rcu/rcutorture.c b/kernel/rcu/rcutorture.c
> > index 9c028b118abc..4354f87f522a 100644
> > --- a/kernel/rcu/rcutorture.c
> > +++ b/kernel/rcu/rcutorture.c
> > @@ -1115,6 +1115,7 @@ static int rcu_torture_boost(void *arg)
> > }
> >
> > // Do one boost-test interval.
> > + cpus_read_lock();
> > endtime = oldstarttime + test_boost_duration * HZ;
> > while (time_before(jiffies, endtime)) {
> > // Has current GP gone too long?
> > @@ -1164,7 +1165,7 @@ static int rcu_torture_boost(void *arg)
> > }
> > schedule_timeout_uninterruptible(1);
> > }
> > -
> > + cpus_read_unlock();
> > /* Go do the stutter. */
> > checkwait: if (stutter_wait("rcu_torture_boost"))
> > sched_set_fifo_low(current);
> > --
> >
> > [2]
> > diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> > index 1569e9591654..1e79e059e578 100644
> > --- a/kernel/sched/core.c
> > +++ b/kernel/sched/core.c
> > @@ -2552,6 +2552,12 @@ static struct rq *move_queued_task(struct rq *rq, struct rq_flags *rf,
> > rq = cpu_rq(new_cpu);
> >
> > rq_lock(rq, rf);
> > + if (rq->curr) {
> > + trace_printk("Curr on new_cpu (%d) is: %s[%d]\n",
> > + new_cpu, rq->curr->comm, task_pid_nr(rq->curr));
> > + } else {
> > + trace_printk("Curr on new_cpu (%d) is NULL\n", new_cpu);
> > + }
> > WARN_ON_ONCE(task_cpu(p) != new_cpu);
> > activate_task(rq, p, 0);
> > check_preempt_curr(rq, p, 0);
> > @@ -9490,6 +9496,11 @@ static int __balance_push_cpu_stop(void *arg)
> > trace_printk("attempt push task %d currently on cpu %d to...\n", task_pid_nr(p), rq->cpu);
> > cpu = select_fallback_rq(rq->cpu, p);
> > trace_printk("to new cpu %d\n", cpu);
> > + trace_printk("is_cpu_allowed(%d), md(p)=%d, pf_kthread=%d, kthread_is_per_cpu=%d, cpu_online=%d, cpu_dying=%d\n",
> > + cpu,
> > + is_migration_disabled(p),
> > + p->flags & PF_KTHREAD, kthread_is_per_cpu(p),
> > + cpu_online(cpu), cpu_dying(cpu));
> > rq = __migrate_task(rq, &rf, p, cpu);
> > trace_printk("After __migrate_task, new cpu = %d\n", rq->cpu);
> > }
> > --
> > 2.42.0.459.ge4e396fd5e-goog
> >
next prev parent reply other threads:[~2023-09-14 18:56 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-10 20:14 [BUG] Random intermittent boost failures (Was Re: [BUG] TREE04..) Joel Fernandes
2023-09-10 21:16 ` Paul E. McKenney
2023-09-10 23:37 ` Joel Fernandes
2023-09-11 2:27 ` Joel Fernandes
2023-09-11 8:16 ` Paul E. McKenney
2023-09-11 13:17 ` Joel Fernandes
2023-09-11 13:49 ` Paul E. McKenney
2023-09-11 16:18 ` Joel Fernandes
2023-09-13 20:30 ` Joel Fernandes
2023-09-14 11:11 ` Paul E. McKenney
2023-09-14 13:13 ` Joel Fernandes
2023-09-14 15:23 ` Paul E. McKenney
2023-09-14 18:56 ` Joel Fernandes [this message]
2023-09-14 21:53 ` Joel Fernandes
2023-09-15 0:13 ` Joel Fernandes
2023-09-15 11:33 ` Joel Fernandes
2023-09-15 14:53 ` Paul E. McKenney
2023-09-15 16:37 ` Joel Fernandes
2023-09-15 16:57 ` Paul E. McKenney
2023-09-15 21:14 ` Joel Fernandes
2023-09-18 6:05 ` Paul E. McKenney
2023-09-15 14:48 ` Paul E. McKenney
2023-09-15 14:45 ` Paul E. McKenney
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230914185627.GA2520229@google.com \
--to=joel@joelfernandes.org \
--cc=frederic@kernel.org \
--cc=paulmck@kernel.org \
--cc=rcu@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).