linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Yosry Ahmed <yosryahmed@google.com>
To: "Michal Koutný" <mkoutny@suse.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	Johannes Weiner <hannes@cmpxchg.org>,
	 Michal Hocko <mhocko@kernel.org>,
	Roman Gushchin <roman.gushchin@linux.dev>,
	 Shakeel Butt <shakeelb@google.com>,
	Muchun Song <muchun.song@linux.dev>,
	 Ivan Babrou <ivan@cloudflare.com>, Tejun Heo <tj@kernel.org>,
	linux-mm@kvack.org,  cgroups@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH 2/3] mm: memcg: add a helper for non-unified stats flushing
Date: Tue, 22 Aug 2023 09:00:06 -0700	[thread overview]
Message-ID: <CAJD7tkaVuiMU-ifJiyH5d_W1hi9DnAymYJxzBxEKCVX+tU=OCA@mail.gmail.com> (raw)
In-Reply-To: <y7vvyndehadwy5ouounm5oyo52bqhsysiizuphzki2h3bwpdpc@3hvaq6sp6f4k>

On Tue, Aug 22, 2023 at 6:01 AM Michal Koutný <mkoutny@suse.com> wrote:
>
> On Mon, Aug 21, 2023 at 08:54:57PM +0000, Yosry Ahmed <yosryahmed@google.com> wrote:
> > +static void do_stats_flush(struct mem_cgroup *memcg)
> > +{
> > +     cgroup_rstat_flush(memcg->css.cgroup);
>         if(memcg == root_mem_cgroup)
>                 atomic_set(&stats_flush_threshold, 0);
> > +}
> > +
> >  /*
> >   * do_unified_stats_flush - do a unified flush of memory cgroup statistics
> >   *
> > @@ -656,7 +667,7 @@ static void do_unified_stats_flush(void)
> >
> >       WRITE_ONCE(flush_next_time, jiffies_64 + 2*FLUSH_TIME);
> >
> > -     cgroup_rstat_flush(root_mem_cgroup->css.cgroup);
> > +     do_stats_flush(root_mem_cgroup);
> >
>   -     atomic_set(&stats_flush_threshold, 0);
> >       atomic_set(&stats_flush_ongoing, 0);
>
> You may reset stats_flush_threshold to save the unified flushers some
> work.

We can probably also kick FLUSH_TIME forward as well. Perhaps I can
move both into do_stats_flush() then. If I understand correctly this
is what you mean?

static void do_stats_flush(struct mem_cgroup *memcg)
{
       if (mem_cgroup_is_root(memcg))
               WRITE_ONCE(flush_next_time, jiffies_64 + 2*FLUSH_TIME);

       cgroup_rstat_flush(memcg->css.cgroup);

       if (mem_cgroup_is_root(memcg))
               atomic_set(&stats_flush_threshold, 0);
}

  static void do_unified_stats_flush(void)
  {
          if (atomic_read(&stats_flush_ongoing) ||
              atomic_xchg(&stats_flush_ongoing, 1))
                  return;

          do_stats_flush(root_mem_cgroup);
          atomic_set(&stats_flush_ongoing, 0);
  }

, or simplify it further by just resetting stats_flush_threshold
before we flush as well:

static void do_stats_flush(struct mem_cgroup *memcg)
{
       /* for unified flushing, root non-unified flushing can help as well */
       if (mem_cgroup_is_root(memcg)) {
               WRITE_ONCE(flush_next_time, jiffies_64 + 2*FLUSH_TIME);
               atomic_set(&stats_flush_threshold, 0);
       }

       cgroup_rstat_flush(memcg->css.cgroup);
}

  static void do_unified_stats_flush(void)
  {
          if (atomic_read(&stats_flush_ongoing) ||
              atomic_xchg(&stats_flush_ongoing, 1))
                  return;

          do_stats_flush(root_mem_cgroup);
          atomic_set(&stats_flush_ongoing, 0);
  }


What do you think?


  reply	other threads:[~2023-08-22 16:00 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-21 20:54 [PATCH 0/3] memcg: non-unified flushing for userspace stats Yosry Ahmed
2023-08-21 20:54 ` [PATCH 1/3] mm: memcg: properly name and document unified stats flushing Yosry Ahmed
2023-08-21 20:54 ` [PATCH 2/3] mm: memcg: add a helper for non-unified " Yosry Ahmed
2023-08-22 13:01   ` Michal Koutný
2023-08-22 16:00     ` Yosry Ahmed [this message]
2023-08-22 16:35       ` Michal Koutný
2023-08-22 16:48         ` Yosry Ahmed
2023-08-21 20:54 ` [PATCH 3/3] mm: memcg: use non-unified stats flushing for userspace reads Yosry Ahmed
2023-08-22  9:06   ` Michal Hocko
2023-08-22 15:30     ` Yosry Ahmed
2023-08-23  7:33       ` Michal Hocko
2023-08-23 14:55         ` Yosry Ahmed
2023-08-24  7:13           ` Michal Hocko
2023-08-24 18:15             ` Yosry Ahmed
2023-08-24 18:50               ` Yosry Ahmed
2023-08-25  7:05                 ` Michal Hocko
2023-08-25 15:14                   ` Yosry Ahmed
2023-08-25 18:17                     ` Michal Hocko
2023-08-25 18:21                       ` Yosry Ahmed
2023-08-25 18:43                         ` Michal Hocko
2023-08-25 18:44                           ` Michal Hocko
2023-08-28 15:47                     ` Michal Hocko
2023-08-28 16:15                       ` Yosry Ahmed
2023-08-28 17:00                         ` Shakeel Butt
2023-08-28 17:07                           ` Yosry Ahmed
2023-08-28 17:27                             ` Waiman Long
2023-08-28 17:28                               ` Yosry Ahmed
2023-08-28 17:35                                 ` Waiman Long
2023-08-28 17:43                                   ` Waiman Long
2023-08-28 18:35                                     ` Yosry Ahmed
2023-08-29  7:27                               ` Michal Hocko
2023-08-29 15:05                                 ` Waiman Long
2023-08-29 15:17                                   ` Michal Hocko
2023-08-29 16:04                                     ` Yosry Ahmed
2023-08-29 18:44                   ` Tejun Heo
2023-08-29 19:13                     ` Yosry Ahmed
2023-08-29 19:36                       ` Tejun Heo
2023-08-29 19:54                         ` Yosry Ahmed
2023-08-29 20:12                           ` Tejun Heo
2023-08-29 20:20                             ` Yosry Ahmed
2023-08-31  9:05                               ` Michal Hocko
2023-08-22 13:00 ` [PATCH 0/3] memcg: non-unified flushing for userspace stats Michal Koutný
2023-08-22 15:43   ` Yosry Ahmed

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='CAJD7tkaVuiMU-ifJiyH5d_W1hi9DnAymYJxzBxEKCVX+tU=OCA@mail.gmail.com' \
    --to=yosryahmed@google.com \
    --cc=akpm@linux-foundation.org \
    --cc=cgroups@vger.kernel.org \
    --cc=hannes@cmpxchg.org \
    --cc=ivan@cloudflare.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@kernel.org \
    --cc=mkoutny@suse.com \
    --cc=muchun.song@linux.dev \
    --cc=roman.gushchin@linux.dev \
    --cc=shakeelb@google.com \
    --cc=tj@kernel.org \
    /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