linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Rafael Folco <rfolco@redhat.com>
To: Aaron Tomlin <atomlin@redhat.com>
Cc: frederic@kernel.org, mtosatti@redhat.com, cl@linux.com,
	tglx@linutronix.de,  mingo@kernel.org, peterz@infradead.org,
	pauld@redhat.com, neelx@redhat.com,  oleksandr@natalenko.name,
	atomlin@atomlin.com, akpm@linux-foundation.org,
	 linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCH v8 4/5] tick/nohz_full: Ensure quiet_vmstat() is called on exit to user-mode when the idle tick is stopped
Date: Tue, 27 Sep 2022 13:17:02 -0300	[thread overview]
Message-ID: <CAOr2xH5DjEy7agHYx=5+4cr+OeJz52HUeUttkF9ueS8JHbQMFg@mail.gmail.com> (raw)
In-Reply-To: <20220924152227.819815-5-atomlin@redhat.com>

[-- Attachment #1: Type: text/plain, Size: 3465 bytes --]

Tested this patch w/ nohz_full setup and oslat 8h run on isolated cpus, max
latency is 7us versus 15us without the patch.
Thanks.

On Mon, Sep 26, 2022 at 8:02 PM Aaron Tomlin <atomlin@redhat.com> wrote:

> This patch ensures CPU-specific vmstat differentials do not remain
> when the scheduling-tick is stopped and before exiting to user-mode
> in the context of nohz_full only.
>
> A trivial test program was used to determine the impact of the proposed
> changes and under vanilla. The mlock(2) and munlock(2) system calls was
> used solely to modify vmstat item 'NR_MLOCK'. The following is an average
> count of CPU-cycles across the aforementioned system calls:
>
>                                   Vanilla                 Modified
>
>   Cycles per syscall              8461                    8690    (+2.6%)
>
> Signed-off-by: Aaron Tomlin <atomlin@redhat.com>
> ---
>  include/linux/tick.h     |  5 +++--
>  kernel/time/tick-sched.c | 15 +++++++++++++++
>  2 files changed, 18 insertions(+), 2 deletions(-)
>
> diff --git a/include/linux/tick.h b/include/linux/tick.h
> index bfd571f18cfd..a2bbd6d32e33 100644
> --- a/include/linux/tick.h
> +++ b/include/linux/tick.h
> @@ -11,7 +11,6 @@
>  #include <linux/context_tracking_state.h>
>  #include <linux/cpumask.h>
>  #include <linux/sched.h>
> -#include <linux/rcupdate.h>
>
>  #ifdef CONFIG_GENERIC_CLOCKEVENTS
>  extern void __init tick_init(void);
> @@ -272,6 +271,7 @@ static inline void tick_dep_clear_signal(struct
> signal_struct *signal,
>
>  extern void tick_nohz_full_kick_cpu(int cpu);
>  extern void __tick_nohz_task_switch(void);
> +void __tick_nohz_user_enter_prepare(void);
>  extern void __init tick_nohz_full_setup(cpumask_var_t cpumask);
>  #else
>  static inline bool tick_nohz_full_enabled(void) { return false; }
> @@ -296,6 +296,7 @@ static inline void tick_dep_clear_signal(struct
> signal_struct *signal,
>
>  static inline void tick_nohz_full_kick_cpu(int cpu) { }
>  static inline void __tick_nohz_task_switch(void) { }
> +static inline void __tick_nohz_user_enter_prepare(void) { }
>  static inline void tick_nohz_full_setup(cpumask_var_t cpumask) { }
>  #endif
>
> @@ -308,7 +309,7 @@ static inline void tick_nohz_task_switch(void)
>  static inline void tick_nohz_user_enter_prepare(void)
>  {
>         if (tick_nohz_full_cpu(smp_processor_id()))
> -               rcu_nocb_flush_deferred_wakeup();
> +               __tick_nohz_user_enter_prepare();
>  }
>
>  #endif
> diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
> index b0e3c9205946..634cd0fac267 100644
> --- a/kernel/time/tick-sched.c
> +++ b/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();
> +               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)
>  {
> --
> 2.37.1
>
>
>

-- 
Folco

[-- Attachment #2: Type: text/html, Size: 4439 bytes --]

  reply	other threads:[~2022-09-27 16:17 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-09-24 15:22 [PATCH v8 0/5] Ensure quiet_vmstat() is called when the idle tick was stopped too Aaron Tomlin
2022-09-24 15:22 ` [PATCH v8 1/5] mm/vmstat: Add CPU-specific variable to track a vmstat discrepancy Aaron Tomlin
2022-09-24 15:22 ` [PATCH v8 2/5] mm/vmstat: Use vmstat_dirty to track CPU-specific vmstat discrepancies Aaron Tomlin
2022-10-01 14:55   ` kernel test robot
2022-09-24 15:22 ` [PATCH v8 3/5] mm/vmstat: Do not queue vmstat_update if tick is stopped Aaron Tomlin
2022-10-24 11:03   ` Frederic Weisbecker
2022-11-09 19:40     ` Marcelo Tosatti
2022-11-10 19:15       ` Marcelo Tosatti
2022-11-14 12:12       ` Frederic Weisbecker
2022-10-24 11:54   ` Frederic Weisbecker
2022-09-24 15:22 ` [PATCH v8 4/5] tick/nohz_full: Ensure quiet_vmstat() is called on exit to user-mode when the idle " Aaron Tomlin
2022-09-27 16:17   ` Rafael Folco [this message]
2022-09-29  8:22     ` Aaron Tomlin
2022-09-29 12:49       ` Rafael Folco
2022-10-21 14:50     ` Frederic Weisbecker
2022-11-10 19:14       ` Marcelo Tosatti
2022-09-24 15:24 ` [PATCH v8 5/5] tick/sched: Ensure quiet_vmstat() is called when the idle tick was stopped too Aaron Tomlin
2022-09-25  1:05   ` Hillf Danton
2022-09-26  9:20     ` Aaron Tomlin
2022-10-03 12:44       ` Hillf Danton
2022-10-12 12:41         ` Aaron Tomlin
2022-10-17 16:04         ` Marcelo Tosatti

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='CAOr2xH5DjEy7agHYx=5+4cr+OeJz52HUeUttkF9ueS8JHbQMFg@mail.gmail.com' \
    --to=rfolco@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=atomlin@atomlin.com \
    --cc=atomlin@redhat.com \
    --cc=cl@linux.com \
    --cc=frederic@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mingo@kernel.org \
    --cc=mtosatti@redhat.com \
    --cc=neelx@redhat.com \
    --cc=oleksandr@natalenko.name \
    --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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox