All of lore.kernel.org
 help / color / mirror / Atom feed
From: Oleg Nesterov <oleg@redhat.com>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: Frederic Weisbecker <frederic@kernel.org>,
	Ingo Molnar <mingo@redhat.com>,
	Nicholas Piggin <npiggin@gmail.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Phil Auld <pauld@redhat.com>,
	Chris von Recklinghausen <crecklin@redhat.com>,
	linux-kernel@vger.kernel.org
Subject: Re: sched/isolation: tick_take_do_timer_from_boot() calls smp_call_function_single() with irqs disabled
Date: Sun, 26 May 2024 21:27:58 +0200	[thread overview]
Message-ID: <20240526192758.GA21193@redhat.com> (raw)
In-Reply-To: <87v832dfw1.ffs@tglx>

The more I grep the more I confused.

On 05/25, Thomas Gleixner wrote:
>
> Right. It does not happen because the kernel starts with jiffies as
> clocksource except on S390. The jiffies clocksource is not qualified to
> switch over to NOHZ mode for obvious reasons.

Not obvious for those who never looked at this code ;)

OK, clocksource_jiffies doesn't have CLOCK_SOURCE_VALID_FOR_HRES,

> fs_initcall(clocksource_done_booting)

So clocksource_done_booting() -> clocksource_select() should find another
CLOCK_SOURCE_VALID_FOR_HRES cs, then timekeeping_notify(best)

	- sets tk_core.timekeeper.tkr_mono.clock = best

	- calls tick_clock_notify() which sets .check_clocks on every cpu.

This makes tick_check_oneshot_change(false) return true. (I am ignoring the
highres=n case, tick_nohz_switch_to_nohz() doesn't really differ).

Now, on every CPU the next ->event_handler == tick_handle_periodic path
call hrtimer_switch_to_hres() which

	- sets ->event_handler == hrtimer_interrupt (tick_init_highres)
	  so tick_periodic/do_timer will be never called again

	- calls tick_setup_sched_timer() -> tick_nohz_activate() which
	  sets TS_FLAG_NOHZ

> [    1.918548] clocksource_done_booting: Switched to NOHZ // debug printk
>
> This is the point where tick_nohz_activate() is called first time and
> that does:
>
>   tick_sched_flag_set(ts, TS_FLAG_NOHZ);

See above, but I got lost, most probably I misunderstood these nontrivial
code paths.

> So up to this point the tick is never stopped neither on housekeeping
> nor on NOHZ FULL CPUs:
>
> tick_nohz_full_update_tick()
>   if (!tick_sched_flag_test(ts, TS_FLAG_NOHZ))
>     return;

OK... But tick_nohz_idle_update_tick() doesn't check TS_FLAG_NOHZ and
the tick_nohz_full_cpu() check can't help at boot time.

And I still don't understand why we can rely on can_stop_idle_tick() even
in tick_nohz_idle_stop_tick().

I'll try to read this code again tomorrow, but it will never fit my poor
little brain ;)

Oleg.


  parent reply	other threads:[~2024-05-26 19:29 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-05-22 15:17 sched/isolation: tick_take_do_timer_from_boot() calls smp_call_function_single() with irqs disabled Oleg Nesterov
2024-05-23 13:23 ` Oleg Nesterov
2024-05-24  9:31   ` Thomas Gleixner
2024-05-24 14:10     ` Oleg Nesterov
2024-05-24 15:22       ` Frederic Weisbecker
2024-05-24 15:20     ` Frederic Weisbecker
2024-05-24 17:16       ` Thomas Gleixner
2024-05-24 18:37       ` Oleg Nesterov
2024-05-24 22:06         ` Thomas Gleixner
2024-05-25 13:51           ` Oleg Nesterov
2024-05-25 14:13             ` Oleg Nesterov
2024-05-26 19:27           ` Oleg Nesterov [this message]
2024-05-26 20:52             ` Frederic Weisbecker
2024-05-27 15:57               ` Oleg Nesterov
2024-05-27 11:01             ` Nicholas Piggin
2024-05-27 15:57               ` Oleg Nesterov
2024-05-28  1:02                 ` Nicholas Piggin
2024-05-28 12:19                   ` Oleg Nesterov
2024-05-27 16:13               ` Thomas Gleixner
2024-05-26 20:57           ` Frederic Weisbecker
2024-05-27  9:10           ` Nicholas Piggin
2024-05-27 10:23             ` Thomas Gleixner
2024-05-27 11:16               ` Nicholas Piggin
2024-05-28 12:20 ` [PATCH] tick/nohz_full: don't abuse smp_call_function_single() in tick_setup_device() Oleg Nesterov
2024-05-28 12:22   ` Oleg Nesterov
2024-05-30 12:40   ` [PATCH] tick/nohz_full: turn tick_do_timer_boot_cpu into boot_cpu_is_nohz_full Oleg Nesterov
2024-06-03 15:35     ` [PATCH v2] " Oleg Nesterov
2024-06-03 21:44       ` Frederic Weisbecker
2024-06-04  5:08       ` Nicholas Piggin
2024-05-30 14:52   ` [PATCH] tick/nohz_full: don't abuse smp_call_function_single() in tick_setup_device() Frederic Weisbecker
2024-05-30 16:52     ` Oleg Nesterov
2024-05-30 17:01     ` Oleg Nesterov
2024-06-01 14:03     ` Oleg Nesterov
2024-06-02 21:29       ` Frederic Weisbecker
2024-06-03 15:41         ` Oleg Nesterov
2024-06-03 21:45           ` Frederic Weisbecker
2024-06-10 15:55   ` [PING ;)] " Oleg Nesterov
2024-06-10 18:15     ` Thomas Gleixner
2024-06-10 18:26   ` [tip: timers/urgent] tick/nohz_full: Don't " tip-bot2 for Oleg Nesterov
2024-06-10 19:42     ` Frederic Weisbecker

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=20240526192758.GA21193@redhat.com \
    --to=oleg@redhat.com \
    --cc=crecklin@redhat.com \
    --cc=frederic@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=npiggin@gmail.com \
    --cc=pauld@redhat.com \
    --cc=peterz@infradead.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.