From: "Paul E. McKenney" <paulmck@linux.ibm.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: rcu@vger.kernel.org, linux-kernel@vger.kernel.org,
mingo@kernel.org, jiangshanlai@gmail.com, dipankar@in.ibm.com,
akpm@linux-foundation.org, mathieu.desnoyers@efficios.com,
josh@joshtriplett.org, tglx@linutronix.de, rostedt@goodmis.org,
dhowells@redhat.com, edumazet@google.com, fweisbec@gmail.com,
oleg@redhat.com, joel@joelfernandes.org
Subject: Re: [PATCH RFC tip/core/rcu 14/14] rcu/nohz: Make multi_cpu_stop() enable tick on all online CPUs
Date: Mon, 5 Aug 2019 07:54:48 -0700 [thread overview]
Message-ID: <20190805145448.GI28441@linux.ibm.com> (raw)
In-Reply-To: <20190805080531.GH2349@hirez.programming.kicks-ass.net>
On Mon, Aug 05, 2019 at 10:05:31AM +0200, Peter Zijlstra wrote:
> On Sun, Aug 04, 2019 at 11:41:59AM -0700, Paul E. McKenney wrote:
> > On Sun, Aug 04, 2019 at 04:48:35PM +0200, Peter Zijlstra wrote:
> > > On Sun, Aug 04, 2019 at 04:43:17PM +0200, Peter Zijlstra wrote:
> > > > On Fri, Aug 02, 2019 at 08:15:01AM -0700, Paul E. McKenney wrote:
> > > > > The multi_cpu_stop() function relies on the scheduler to gain control from
> > > > > whatever is running on the various online CPUs, including any nohz_full
> > > > > CPUs running long loops in kernel-mode code. Lack of the scheduler-clock
> > > > > interrupt on such CPUs can delay multi_cpu_stop() for several minutes
> > > > > and can also result in RCU CPU stall warnings. This commit therefore
> > > > > causes multi_cpu_stop() to enable the scheduler-clock interrupt on all
> > > > > online CPUs.
> > > >
> > > > This sounds wrong; should we be fixing sched_can_stop_tick() instead to
> > > > return false when the stop task is runnable?
> >
> > Agreed. However, it is proving surprisingly hard to come up with a
> > code sequence that has the effect of rcu_nocb without nohz_full.
> > And rcu_nocb works just fine. With nohz_full also in place, I am
> > decreasing the failure rate, but it still fails, perhaps a few times
> > per hour of TREE04 rcutorture on an eight-CPU system. (My 12-CPU
> > system stubbornly refuses to fail. Good thing I kept the eight-CPU
> > system around, I guess.)
> >
> > When I arrive at some sequence of actions that actually work reliably,
> > then by all means let's put it somewhere in the NO_HZ_FULL machinery!
>
> I'm confused; what are you arguing? The patch as proposed is just wrong,
> it needs to go.
Eventually, sure. But one dragon at a time. Right now that dragon is
"what is required to get multi_cpu_stop() to work in a timely fashioon".
The "where does that code really go" dragon comes later.
> > > And even without that; I don't understand how we're not instantly
> > > preempted the moment we enqueue the stop task.
> >
> > There is no preemption because CONFIG_PREEMPT=n for the scenarios still
>
> That doesn't make sense; even with CONFIG_PREEMPT=n we set
> TIF_NEED_RESCHED. We'll just not react to it as promptly (only explicit
> rescheduling points and return to userspace). Enabling the tick will not
> make any difference what so ever.
>
> Tick based preemption will not 'fix' the lack of wakeup preemption. If
> the stop task wakeup didn't set TIF_NEED_RESCHED, the OTHER/CFS tick
> will not either.
Seems logical except for the fact that multi_cpu_stop() really is taking
in excess of five minutes on a regular basis.
> > having trouble. Yes, there are cond_resched() calls, but they don't do
> > anything unless the appropriate flags are set, which won't always happen
> > without the tick, apparently. Or without -something- that isn't always
> > happening as it should.
>
> Right; so clearly we're not understanding what's happening. That seems
> like a requirement for actually doing a patch.
Almost but not quite. It is a requirement for a patch *that* *is*
*supposed* *to* *be* *a* *fix*. If you are trying to prohibit me from
writing experimental patches, please feel free to take a long walk on
a short pier.
Understood???
> > > Any enqueue, should go through check_preempt_curr() which will be an
> > > instant resched_curr() when we just woke the stop class.
> >
> > I did try hitting all of the CPUs with resched_cpu(). Ten times on each
> > CPU with a ten-jiffy wait between each. This might have decreased the
> > probability of excessively long CPU-stopper waits by a factor of two or
> > three, but it did not eliminate the excessively long waits.
> >
> > What else should I try?
> >
> > For example, are there any diagnostics I could collect, say from within
> > the CPU stopper when things are taking too long? I see CPU-stopper
> > delays in excess of five -minutes-, so this is anything but subtle.
>
> Catch the whole thing in a function trace?
>
> The chain that should instantly set TIF_NEED_RESCHED:
>
> stop_machine()
> stop_machine_cpuslocked()
> stop_cpus()
> __stop_cpus()
> queue_stop_cpus_work()
> cpu_stop_queue_work()
> wake_up_q()
> wake_up_process()
>
>
> wake_up_process()
> try_to_wake_up()
> ttwu_queue()
> ttwu_queue_remote()
> <- scheduler_ipi()
> sched_ttwu_pending()
> ttwu_do_activate()
>
> ttwu_do_activate()
> activate_task()
> ttwu_do_wakeup()
> check_preempt_curr()
> resched_curr()
>
> You could frob some tracing into __stop_cpus(), before
> wait_for_completion(), at that point all the CPUs in @cpumask should
> either be running the stop task or have TIF_NEED_RESCHED set.
Thank you, this should be quite helpful.
Thanx, Paul
next prev parent reply other threads:[~2019-08-05 14:55 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-08-02 15:14 [PATCH tip/core/rcu 0/14] No-CBs bypass addition for v5.4 Paul E. McKenney
2019-08-02 15:14 ` [PATCH RFC tip/core/rcu 01/14] rcu/nocb: Atomic ->len field in rcu_segcblist structure Paul E. McKenney
2019-08-04 14:50 ` Peter Zijlstra
2019-08-04 14:52 ` Peter Zijlstra
2019-08-04 18:45 ` Paul E. McKenney
2019-08-04 18:42 ` Paul E. McKenney
2019-08-02 15:14 ` [PATCH RFC tip/core/rcu 02/14] rcu/nocb: Add bypass callback queueing Paul E. McKenney
2019-08-07 0:03 ` Joel Fernandes
2019-08-07 0:16 ` Joel Fernandes
2019-08-07 0:35 ` Paul E. McKenney
2019-08-07 0:40 ` Steven Rostedt
2019-08-07 1:17 ` Paul E. McKenney
2019-08-07 1:24 ` Steven Rostedt
2019-08-07 3:47 ` Paul E. McKenney
2019-08-02 15:14 ` [PATCH RFC tip/core/rcu 03/14] rcu/nocb: EXP Check use and usefulness of ->nocb_lock_contended Paul E. McKenney
2019-08-02 15:14 ` [PATCH RFC tip/core/rcu 04/14] rcu/nocb: Print no-CBs diagnostics when rcutorture writer unduly delayed Paul E. McKenney
2019-08-02 15:14 ` [PATCH RFC tip/core/rcu 05/14] rcu/nocb: Avoid synchronous wakeup in __call_rcu_nocb_wake() Paul E. McKenney
2019-08-02 15:14 ` [PATCH RFC tip/core/rcu 06/14] rcu/nocb: Advance CBs after merge in rcutree_migrate_callbacks() Paul E. McKenney
2019-08-02 15:14 ` [PATCH RFC tip/core/rcu 07/14] rcu/nocb: Reduce nocb_cb_wait() leaf rcu_node ->lock contention Paul E. McKenney
2019-08-02 15:14 ` [PATCH RFC tip/core/rcu 08/14] rcu/nocb: Reduce __call_rcu_nocb_wake() " Paul E. McKenney
2019-08-02 15:14 ` [PATCH RFC tip/core/rcu 09/14] rcu/nocb: Don't wake no-CBs GP kthread if timer posted under overload Paul E. McKenney
2019-08-02 15:14 ` [PATCH RFC tip/core/rcu 10/14] rcu: Allow rcu_do_batch() to dynamically adjust batch sizes Paul E. McKenney
2019-08-02 15:14 ` [PATCH RFC tip/core/rcu 11/14] EXP nohz: Add TICK_DEP_BIT_RCU Paul E. McKenney
2019-08-02 15:14 ` [PATCH RFC tip/core/rcu 12/14] rcu/nohz: Force on tick when invoking lots of callbacks Paul E. McKenney
2019-08-02 15:15 ` [PATCH RFC tip/core/rcu 13/14] rcutorture: Force on tick for readers and callback flooders Paul E. McKenney
2019-08-02 15:15 ` [PATCH RFC tip/core/rcu 14/14] rcu/nohz: Make multi_cpu_stop() enable tick on all online CPUs Paul E. McKenney
2019-08-04 14:43 ` Peter Zijlstra
2019-08-04 14:48 ` Peter Zijlstra
2019-08-04 18:41 ` Paul E. McKenney
2019-08-04 20:24 ` Paul E. McKenney
2019-08-05 4:19 ` Paul E. McKenney
2019-08-05 8:07 ` Peter Zijlstra
2019-08-05 14:47 ` Paul E. McKenney
2019-08-05 8:05 ` Peter Zijlstra
2019-08-05 14:54 ` Paul E. McKenney [this message]
2019-08-05 15:50 ` Peter Zijlstra
2019-08-05 17:48 ` Paul E. McKenney
2019-08-06 18:08 ` Paul E. McKenney
2019-08-07 21:41 ` Paul E. McKenney
2019-08-08 20:35 ` Paul E. McKenney
2019-08-08 21:30 ` Paul E. McKenney
2019-08-09 16:51 ` Paul E. McKenney
2019-08-09 18:07 ` Joel Fernandes
2019-08-09 18:39 ` Paul E. McKenney
2019-08-12 21:02 ` Frederic Weisbecker
2019-08-12 23:23 ` Paul E. McKenney
2019-08-13 1:33 ` Joel Fernandes
2019-08-13 12:30 ` Frederic Weisbecker
2019-08-13 14:48 ` Paul E. McKenney
2019-08-14 17:55 ` Joel Fernandes
2019-08-14 22:05 ` Paul E. McKenney
2019-08-15 15:07 ` Joel Fernandes
2019-08-15 17:23 ` Paul E. McKenney
2019-08-15 18:15 ` Joel Fernandes
2019-08-15 18:39 ` Paul E. McKenney
2019-08-15 19:42 ` Joel Fernandes
2019-08-13 21:06 ` 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=20190805145448.GI28441@linux.ibm.com \
--to=paulmck@linux.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=dhowells@redhat.com \
--cc=dipankar@in.ibm.com \
--cc=edumazet@google.com \
--cc=fweisbec@gmail.com \
--cc=jiangshanlai@gmail.com \
--cc=joel@joelfernandes.org \
--cc=josh@joshtriplett.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=mingo@kernel.org \
--cc=oleg@redhat.com \
--cc=peterz@infradead.org \
--cc=rcu@vger.kernel.org \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
/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 NNTP newsgroup(s).