Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:1 as permitted sender) client-ip=2620:137:e000::3:1;
MIME-Version: 1.0
References: <20231010032117.1577496-4-yosryahmed@google.com>
 <202310202303.c68e7639-oliver.sang@intel.com> <CALvZod5hKvjm3WVSOGc5PpR9eNHFkt=BDmcrBe5CeWgFzP7jgQ@mail.gmail.com>
In-Reply-To: <CALvZod5hKvjm3WVSOGc5PpR9eNHFkt=BDmcrBe5CeWgFzP7jgQ@mail.gmail.com>
From:   Yosry Ahmed <yosryahmed@google.com>
Date:   Fri, 20 Oct 2023 10:42:58 -0700
Message-ID: <CAJD7tkbjZri4ayBOT9rJ0yMAi__c-1SVmRh_5oXezr7U6dvALg@mail.gmail.com>
Subject: Re: [PATCH v2 3/5] mm: memcg: make stats flushing threshold per-memcg
To:     Shakeel Butt <shakeelb@google.com>
Cc:     kernel test robot <oliver.sang@intel.com>, oe-lkp@lists.linux.dev,
        lkp@intel.com, cgroups@vger.kernel.org, linux-mm@kvack.org,
        ying.huang@intel.com, feng.tang@intel.com, fengwei.yin@intel.com,
        Andrew Morton <akpm@linux-foundation.org>,
        Johannes Weiner <hannes@cmpxchg.org>,
        Michal Hocko <mhocko@kernel.org>,
        Roman Gushchin <roman.gushchin@linux.dev>,
        Muchun Song <muchun.song@linux.dev>,
        Ivan Babrou <ivan@cloudflare.com>, Tejun Heo <tj@kernel.org>,
        =?UTF-8?Q?Michal_Koutn=C3=BD?= <mkoutny@suse.com>,
        Waiman Long <longman@redhat.com>, kernel-team@cloudflare.com,
        Wei Xu <weixugc@google.com>, Greg Thelen <gthelen@google.com>,
        linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Fri, Oct 20, 2023 at 10:23=E2=80=AFAM Shakeel Butt <shakeelb@google.com>=
 wrote:
>
> On Fri, Oct 20, 2023 at 9:18=E2=80=AFAM kernel test robot <oliver.sang@in=
tel.com> wrote:
> >
> >
> >
> > Hello,
> >
> > kernel test robot noticed a -25.8% regression of will-it-scale.per_thre=
ad_ops on:
> >
> >
> > commit: 51d74c18a9c61e7ee33bc90b522dd7f6e5b80bb5 ("[PATCH v2 3/5] mm: m=
emcg: make stats flushing threshold per-memcg")
> > url: https://github.com/intel-lab-lkp/linux/commits/Yosry-Ahmed/mm-memc=
g-change-flush_next_time-to-flush_last_time/20231010-112257
> > base: https://git.kernel.org/cgit/linux/kernel/git/akpm/mm.git mm-every=
thing
> > patch link: https://lore.kernel.org/all/20231010032117.1577496-4-yosrya=
hmed@google.com/
> > patch subject: [PATCH v2 3/5] mm: memcg: make stats flushing threshold =
per-memcg
> >
> > testcase: will-it-scale
> > test machine: 104 threads 2 sockets (Skylake) with 192G memory
> > parameters:
> >
> >         nr_task: 100%
> >         mode: thread
> >         test: fallocate1
> >         cpufreq_governor: performance
> >
> >
> > In addition to that, the commit also has significant impact on the foll=
owing tests:
> >
> > +------------------+---------------------------------------------------=
------------+
> > | testcase: change | will-it-scale: will-it-scale.per_thread_ops -30.0%=
 regression |
> > | test machine     | 104 threads 2 sockets (Skylake) with 192G memory  =
            |
> > | test parameters  | cpufreq_governor=3Dperformance                    =
              |
> > |                  | mode=3Dthread                                     =
              |
> > |                  | nr_task=3D50%                                     =
              |
> > |                  | test=3Dfallocate1                                 =
              |
> > +------------------+---------------------------------------------------=
------------+
> >
>
> Yosry, I don't think 25% to 30% regression can be ignored. Unless
> there is a quick fix, IMO this series should be skipped for the
> upcoming kernel open window.

I am currently looking into it. It's reasonable to skip the next merge
window if a quick fix isn't found soon.

I am surprised by the size of the regression given the following:
      1.12 =C4=85  5%      +1.4        2.50 =C4=85  2%
perf-profile.self.cycles-pp.__mod_memcg_lruvec_state

IIUC we are only spending 1% more time in __mod_memcg_lruvec_state().