Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20210811031734.GA5193@xsang-OptiPlex-9020> <CAHk-=wiSHHSuSQsCCLOxQA+cbcvjmEeMsTCMWPT1sFVngd9-ig@mail.gmail.com>
 <20210812031910.GA63920@shbuild999.sh.intel.com> <20210816032855.GB72770@shbuild999.sh.intel.com>
 <YRrbpRsvdDoom9iG@cmpxchg.org> <20210817024500.GC72770@shbuild999.sh.intel.com>
 <20210817164737.GA23342@blackbody.suse.cz>
In-Reply-To: <20210817164737.GA23342@blackbody.suse.cz>
From:   Shakeel Butt <shakeelb@google.com>
Date:   Tue, 17 Aug 2021 10:10:43 -0700
Message-ID: <CALvZod4-bsv+Mn19Q9LK3DzL8GC_LuZyJyQ83RiwRiTbCJhCZQ@mail.gmail.com>
Subject: Re: [mm] 2d146aa3aa: vm-scalability.throughput -36.4% regression
To:     =?UTF-8?Q?Michal_Koutn=C3=BD?= <mkoutny@suse.com>
Cc:     Feng Tang <feng.tang@intel.com>,
        Johannes Weiner <hannes@cmpxchg.org>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        kernel test robot <oliver.sang@intel.com>,
        Roman Gushchin <guro@fb.com>, Michal Hocko <mhocko@suse.com>,
        Balbir Singh <bsingharora@gmail.com>,
        Tejun Heo <tj@kernel.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        LKML <linux-kernel@vger.kernel.org>, lkp@lists.01.org,
        kernel test robot <lkp@intel.com>,
        "Huang, Ying" <ying.huang@intel.com>,
        Zhengjun Xing <zhengjun.xing@linux.intel.com>,
        andi.kleen@intel.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Tue, Aug 17, 2021 at 9:47 AM Michal Koutn=C3=BD <mkoutny@suse.com> wrote=
:
>
> On Tue, Aug 17, 2021 at 10:45:00AM +0800, Feng Tang <feng.tang@intel.com>=
 wrote:
> > Initially from the perf-c2c data, the in-cacheline hotspots are only
> > 0x0, and 0x10, and if we extends to 2 cachelines, there is one more
> > offset 0x54 (css.flags), but still I can't figure out which member
> > inside the 128 bytes range is written frequenty.
>
> Is it certain that perf-c2c reported offsets are the cacheline of the
> first bytes of struct cgroup_subsys_state? (Yeah, it looks to me so,
> given what code accesses those and your padding fixing it. I'm just
> raising it in case there was anything non-obvious.)
>
> >
> > /* pah info for cgroup_subsys_state */
> > struct cgroup_subsys_state {
> >       struct cgroup *            cgroup;               /*     0     8 *=
/
> >       struct cgroup_subsys *     ss;                   /*     8     8 *=
/
> >       struct percpu_ref          refcnt;               /*    16    16 *=
/
> >       struct list_head           sibling;              /*    32    16 *=
/
> >       struct list_head           children;             /*    48    16 *=
/
> >       /* --- cacheline 1 boundary (64 bytes) --- */
> >       struct list_head           rstat_css_node;       /*    64    16 *=
/
> >       int                        id;                   /*    80     4 *=
/
> >       unsigned int               flags;                /*    84     4 *=
/
> >       u64                        serial_nr;            /*    88     8 *=
/
> >       atomic_t                   online_cnt;           /*    96     4 *=
/
> >
> >       /* XXX 4 bytes hole, try to pack */
> >
> >       struct work_struct         destroy_work;         /*   104    32 *=
/
> >       /* --- cacheline 2 boundary (128 bytes) was 8 bytes ago --- */
> >
> > Since the test run implies this is cacheline related, and I'm not very
> > familiar with the mem_cgroup code, the original perf-c2c log is attache=
d
> > which may give more hints.
>
> As noted by Johannes, even in atomic mode, the refcnt would have the
> atomic part elsewhere. The other members shouldn't be written frequently
> unless there are some intense modifications of the cgroup tree in
> parallel.
> Does the benchmark create lots of memory cgroups in such a fashion?

From what I know the benchmark is running in the root cgroup and there
is no cgroup manipulation.