Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Fri, 9 Sep 2022 14:12:24 +0200
From:   Frederic Weisbecker <frederic@kernel.org>
To:     Marcelo Tosatti <mtosatti@redhat.com>
Cc:     atomlin@redhat.com, cl@linux.com, tglx@linutronix.de,
        mingo@kernel.org, peterz@infradead.org, pauld@redhat.com,
        neelx@redhat.com, oleksandr@natalenko.name,
        linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCH v7 2/3] tick/sched: Ensure quiet_vmstat() is called when
 the idle tick was stopped too
Message-ID: <20220909121224.GA220905@lothringen>
References: <20220817191346.287594886@redhat.com>
 <20220817191524.201253713@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20220817191524.201253713@redhat.com>
Precedence: bulk

On Wed, Aug 17, 2022 at 04:13:48PM -0300, Marcelo Tosatti wrote:
> From: Aaron Tomlin <atomlin@redhat.com>
> 
> In the context of the idle task and an adaptive-tick mode/or a nohz_full
> CPU, quiet_vmstat() can be called: before stopping the idle tick,
> entering an idle state and on exit. In particular, for the latter case,
> when the idle task is required to reschedule, the idle tick can remain
> stopped

Since quiet_vmstat() is only called when ts->tick_stopped = false, this
can only happen if the idle loop did not enter into dynticks idle mode
but the exiting idle task eventually stops the tick
(tick_nohz_idle_update_tick()).

This can happen for example if we enter the idle loop with a timer callback
pending in one jiffies, then once that timer fires, which wakes up a task,
we exit the idle loop and then tick_nohz_idle_update_tick() doesn't see any
timer callback pending left and the tick can be stopped.

Or am I missing something?

> and the timer expiration time endless i.e., KTIME_MAX. Now,
> indeed before a nohz_full CPU enters an idle state, CPU-specific vmstat
> counters should be processed to ensure the respective values have been
> reset and folded into the zone specific 'vm_stat[]'. That being said, it
> can only occur when: the idle tick was previously stopped, and
> reprogramming of the timer is not required.
> 
> A customer provided some evidence which indicates that the idle tick was
> stopped; albeit, CPU-specific vmstat counters still remained populated.
> Thus one can only assume quiet_vmstat() was not invoked on return to the
> idle loop.
> 
> If I understand correctly, I suspect this divergence might erroneously
> prevent a reclaim attempt by kswapd. If the number of zone specific free
> pages are below their per-cpu drift value then
> zone_page_state_snapshot() is used to compute a more accurate view of
> the aforementioned statistic.  Thus any task blocked on the NUMA node
> specific pfmemalloc_wait queue will be unable to make significant
> progress via direct reclaim unless it is killed after being woken up by
> kswapd (see throttle_direct_reclaim()).
> 
> Consider the following theoretical scenario:
> 
>         1.      CPU Y migrated running task A to CPU X that was
>                 in an idle state i.e. waiting for an IRQ - not
>                 polling; marked the current task on CPU X to
>                 need/or require a reschedule i.e., set
>                 TIF_NEED_RESCHED and invoked a reschedule IPI to
>                 CPU X (see sched_move_task())

CPU Y is nohz_full right?

> 
>         2.      CPU X acknowledged the reschedule IPI from CPU Y;
>                 generic idle loop code noticed the
>                 TIF_NEED_RESCHED flag against the idle task and
>                 attempts to exit of the loop and calls the main
>                 scheduler function i.e. __schedule().
> 
>                 Since the idle tick was previously stopped no
>                 scheduling-clock tick would occur.
>                 So, no deferred timers would be handled
> 
>         3.      Post transition to kernel execution Task A
>                 running on CPU Y, indirectly released a few pages
>                 (e.g. see __free_one_page()); CPU Y's
>                 'vm_stat_diff[NR_FREE_PAGES]' was updated and zone
>                 specific 'vm_stat[]' update was deferred as per the
>                 CPU-specific stat threshold
> 
>         4.      Task A does invoke exit(2) and the kernel does
>                 remove the task from the run-queue; the idle task
>                 was selected to execute next since there are no
>                 other runnable tasks assigned to the given CPU
>                 (see pick_next_task() and pick_next_task_idle())

This happens on CPU X, right?

> 
>         5.      On return to the idle loop since the idle tick
>                 was already stopped and can remain so (see [1]
>                 below) e.g. no pending soft IRQs, no attempt is
>                 made to zero and fold CPU Y's vmstat counters
>                 since reprogramming of the scheduling-clock tick
>                 is not required/or needed (see [2])

And now back to CPU Y, confused...

[...]
> Index: linux-2.6/kernel/time/tick-sched.c
> ===================================================================
> --- linux-2.6.orig/kernel/time/tick-sched.c
> +++ linux-2.6/kernel/time/tick-sched.c
> @@ -26,6 +26,7 @@
>  #include <linux/posix-timers.h>
>  #include <linux/context_tracking.h>
>  #include <linux/mm.h>
> +#include <linux/rcupdate.h>
>  
>  #include <asm/irq_regs.h>
>  
> @@ -519,6 +520,20 @@ void __tick_nohz_task_switch(void)
>  	}
>  }
>  
> +void __tick_nohz_user_enter_prepare(void)
> +{
> +	struct tick_sched *ts;
> +
> +	if (tick_nohz_full_cpu(smp_processor_id())) {
> +		ts = this_cpu_ptr(&tick_cpu_sched);
> +
> +		if (ts->tick_stopped)
> +			quiet_vmstat();

Wasn't it supposed to be part of the quiescing in task isolation
mode?

Because currently vmstat is a deferrable timer but that deferrability
may not apply to nohz_full anymore (outside idle). And quiet_vmstat()
doesn't cancel the timer so you'll still get the disturbance.

See this patch: https://lore.kernel.org/lkml/20220725104356.GA2950296@lothringen/

> +		rcu_nocb_flush_deferred_wakeup();
> +	}
> +}
>
> +EXPORT_SYMBOL_GPL(__tick_nohz_user_enter_prepare);
> +
>  /* Get the boot-time nohz CPU list from the kernel parameters. */
>  void __init tick_nohz_full_setup(cpumask_var_t cpumask)
>  {
> @@ -890,6 +905,9 @@ static void tick_nohz_stop_tick(struct t
>  		ts->do_timer_last = 0;
>  	}
>  
> +	/* Attempt to fold when the idle tick is stopped or not */
> +	quiet_vmstat();
> +
>  	/* Skip reprogram of event if its not changed */
>  	if (ts->tick_stopped && (expires == ts->next_tick)) {
>  		/* Sanity check: make sure clockevent is actually programmed */

But that chunk looks good.

Thanks.

> @@ -911,7 +929,6 @@ static void tick_nohz_stop_tick(struct t
>  	 */
>  	if (!ts->tick_stopped) {
>  		calc_load_nohz_start();
> -		quiet_vmstat();
>  
>  		ts->last_tick = hrtimer_get_expires(&ts->sched_timer);
>  		ts->tick_stopped = 1;
> 
>