Received-SPF: pass (google.com: domain of linux-kernel+bounces-154383-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Precedence: bulk
MIME-Version: 1.0
References: <cover.1713755580.git.baolin.wang@linux.alibaba.com> <05d0096e4ec3e572d1d52d33a31a661321ac1551.1713755580.git.baolin.wang@linux.alibaba.com>
In-Reply-To: <05d0096e4ec3e572d1d52d33a31a661321ac1551.1713755580.git.baolin.wang@linux.alibaba.com>
From: Barry Song <21cnbao@gmail.com>
Date: Tue, 23 Apr 2024 09:17:27 +0800
Message-ID: <CAGsJ_4xu4iL5pv7T1chyzGC2Vp9q1GwOp3wxb=bYMW-T-pj4dA@mail.gmail.com>
Subject: Re: [RFC PATCH 5/5] mm: shmem: add anonymous share mTHP counters
To: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: akpm@linux-foundation.org, hughd@google.com, willy@infradead.org, 
	david@redhat.com, wangkefeng.wang@huawei.com, ryan.roberts@arm.com, 
	ying.huang@intel.com, shy828301@gmail.com, ziy@nvidia.com, linux-mm@kvack.org, 
	linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Mon, Apr 22, 2024 at 3:03=E2=80=AFPM Baolin Wang
<baolin.wang@linux.alibaba.com> wrote:
>
> Signed-off-by: Baolin Wang <baolin.wang@linux.alibaba.com>
> ---
>  include/linux/huge_mm.h | 2 ++
>  mm/huge_memory.c        | 4 ++++
>  mm/shmem.c              | 5 ++++-
>  3 files changed, 10 insertions(+), 1 deletion(-)
>
> diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h
> index 26b6fa98d8ac..67b9c1acad31 100644
> --- a/include/linux/huge_mm.h
> +++ b/include/linux/huge_mm.h
> @@ -270,6 +270,8 @@ enum mthp_stat_item {
>         MTHP_STAT_ANON_SWPOUT,
>         MTHP_STAT_ANON_SWPOUT_FALLBACK,
>         MTHP_STAT_ANON_SWPIN_REFAULT,
> +       MTHP_STAT_SHMEM_ANON_ALLOC,
> +       MTHP_STAT_SHMEM_ANON_ALLOC_FALLBACK,

not quite sure about this. for 2MB pmd-mapped THP shmem, we count them
as FILE_THP.
here we are counting as SHMEM_ANON. To me, SHMEM_ANON is more correct but
it doesn't align with pmd-mapped THP. David, Ryan, what do you think?


>         __MTHP_STAT_COUNT
>  };
>
> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> index 9e52c0db7580..dc15240c1ab3 100644
> --- a/mm/huge_memory.c
> +++ b/mm/huge_memory.c
> @@ -557,6 +557,8 @@ DEFINE_MTHP_STAT_ATTR(anon_alloc_fallback, MTHP_STAT_=
ANON_ALLOC_FALLBACK);
>  DEFINE_MTHP_STAT_ATTR(anon_swpout, MTHP_STAT_ANON_SWPOUT);
>  DEFINE_MTHP_STAT_ATTR(anon_swpout_fallback, MTHP_STAT_ANON_SWPOUT_FALLBA=
CK);
>  DEFINE_MTHP_STAT_ATTR(anon_swpin_refault, MTHP_STAT_ANON_SWPIN_REFAULT);
> +DEFINE_MTHP_STAT_ATTR(shmem_anon_alloc, MTHP_STAT_SHMEM_ANON_ALLOC);
> +DEFINE_MTHP_STAT_ATTR(shmem_anon_alloc_fallback, MTHP_STAT_SHMEM_ANON_AL=
LOC_FALLBACK);
>
>  static struct attribute *stats_attrs[] =3D {
>         &anon_alloc_attr.attr,
> @@ -564,6 +566,8 @@ static struct attribute *stats_attrs[] =3D {
>         &anon_swpout_attr.attr,
>         &anon_swpout_fallback_attr.attr,
>         &anon_swpin_refault_attr.attr,
> +       &shmem_anon_alloc_attr.attr,
> +       &shmem_anon_alloc_fallback_attr.attr,
>         NULL,
>  };
>
> diff --git a/mm/shmem.c b/mm/shmem.c
> index 8b009e7040b2..4a0aa75ab29c 100644
> --- a/mm/shmem.c
> +++ b/mm/shmem.c
> @@ -1706,11 +1706,14 @@ static struct folio *shmem_alloc_and_add_folio(st=
ruct vm_fault *vmf,
>                         pages =3D 1 << order;
>                         index =3D round_down(index, pages);
>                         folio =3D shmem_alloc_hugefolio(gfp, info, index,=
 order);
> -                       if (folio)
> +                       if (folio) {
> +                               count_mthp_stat(order, MTHP_STAT_SHMEM_AN=
ON_ALLOC);
>                                 goto allocated;
> +                       }
>
>                         if (pages =3D=3D HPAGE_PMD_NR)
>                                 count_vm_event(THP_FILE_FALLBACK);
> +                       count_mthp_stat(order, MTHP_STAT_SHMEM_ANON_ALLOC=
_FALLBACK);
>                         order =3D next_order(&orders, order);
>                 }
>         } else {
> --
> 2.39.3
>

Thanks
Barry