[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240525141316.GB24152@redhat.com>
Date: Sat, 25 May 2024 16:13:16 +0200
From: Oleg Nesterov <oleg@...hat.com>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: Frederic Weisbecker <frederic@...nel.org>,
Ingo Molnar <mingo@...hat.com>, Nicholas Piggin <npiggin@...il.com>,
Peter Zijlstra <peterz@...radead.org>, Phil Auld <pauld@...hat.com>,
Chris von Recklinghausen <crecklin@...hat.com>,
linux-kernel@...r.kernel.org
Subject: Re: sched/isolation: tick_take_do_timer_from_boot() calls
smp_call_function_single() with irqs disabled
On 05/25, Oleg Nesterov wrote:
>
> perhaps we can simply kill it along with #ifdef CONFIG_NO_HZ_FULL ?
>
> if (!td->evtdev) {
> tick_cpu = READ_ONCE(tick_do_timer_cpu);
> /*
> * If no cpu took the do_timer update, assign it to
> * this cpu:
> */
> if (tick_cpu == TICK_DO_TIMER_BOOT) {
> WRITE_ONCE(tick_do_timer_cpu, cpu);
> tick_next_period = ktime_get();
> /*
> * The boot CPU may be nohz_full, in which case the
> * first housekeeping secondary will take do_timer
> * from us.
> */
> } else if (tick_nohz_full_cpu(tick_cpu) &&
> !tick_nohz_full_cpu(cpu)) {
> WRITE_ONCE(tick_do_timer_cpu, cpu);
> }
although tick_nohz_full_cpu(tick_cpu) above depends on the fact that
tick_cpu = TICK_DO_TIMER_NONE should not be possible if
tick_nohz_full_enabled(), not good.
Oleg.
Powered by blists - more mailing lists