Received-SPF: pass (google.com: domain of linux-kernel+bounces-141645-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Precedence: bulk
MIME-Version: 1.0
References: <20240405102704.77559-1-21cnbao@gmail.com> <20240405102704.77559-3-21cnbao@gmail.com>
 <f56d3f42-c7c9-4c0a-9e1a-b89e92a3cc97@arm.com>
In-Reply-To: <f56d3f42-c7c9-4c0a-9e1a-b89e92a3cc97@arm.com>
From: Barry Song <21cnbao@gmail.com>
Date: Fri, 12 Apr 2024 09:52:51 +1200
Message-ID: <CAGsJ_4x_PKu5QyvPsiX9mz_oH22NyqGjf-XMX=Mm9qN4fSC2WA@mail.gmail.com>
Subject: Re: [PATCH v4 2/2] mm: add per-order mTHP anon_swpout and
 anon_swpout_fallback counters
To: Ryan Roberts <ryan.roberts@arm.com>
Cc: david@redhat.com, akpm@linux-foundation.org, linux-mm@kvack.org, 
	cerasuolodomenico@gmail.com, chrisl@kernel.org, kasong@tencent.com, 
	peterx@redhat.com, surenb@google.com, v-songbaohua@oppo.com, 
	willy@infradead.org, yosryahmed@google.com, yuzhao@google.com, 
	linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Fri, Apr 12, 2024 at 4:43=E2=80=AFAM Ryan Roberts <ryan.roberts@arm.com>=
 wrote:
>
> On 05/04/2024 11:27, Barry Song wrote:
> > From: Barry Song <v-songbaohua@oppo.com>
> >
> > This helps to display the fragmentation situation of the swapfile,
> > knowing the proportion of how much we haven't split large folios.
> > So far, we only support non-split swapout for anon memory, with
> > the possibility of expanding to shmem in the future.  So, we add
> > the "anon" prefix to the counter names.
> >
> > Signed-off-by: Barry Song <v-songbaohua@oppo.com>
> > ---
> >  include/linux/huge_mm.h | 2 ++
> >  mm/huge_memory.c        | 4 ++++
> >  mm/page_io.c            | 6 +++++-
> >  mm/vmscan.c             | 3 +++
> >  4 files changed, 14 insertions(+), 1 deletion(-)
> >
> > diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h
> > index c5d33017a4dd..1d893a358df6 100644
> > --- a/include/linux/huge_mm.h
> > +++ b/include/linux/huge_mm.h
> > @@ -267,6 +267,8 @@ unsigned long thp_vma_allowable_orders(struct vm_ar=
ea_struct *vma,
> >  enum mthp_stat_item {
> >       MTHP_STAT_ANON_ALLOC,
> >       MTHP_STAT_ANON_ALLOC_FALLBACK,
> > +     MTHP_STAT_ANON_SWPOUT,
> > +     MTHP_STAT_ANON_SWPOUT_FALLBACK,
> >       __MTHP_STAT_COUNT
> >  };
> >
> > diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> > index 5b875f0fc923..28113f8fdf18 100644
> > --- a/mm/huge_memory.c
> > +++ b/mm/huge_memory.c
> > @@ -554,10 +554,14 @@ static struct kobj_attribute _name##_attr =3D __A=
TTR_RO(_name)
> >
> >  DEFINE_MTHP_STAT_ATTR(anon_alloc, MTHP_STAT_ANON_ALLOC);
> >  DEFINE_MTHP_STAT_ATTR(anon_alloc_fallback, MTHP_STAT_ANON_ALLOC_FALLBA=
CK);
> > +DEFINE_MTHP_STAT_ATTR(anon_swpout, MTHP_STAT_ANON_SWPOUT);
> > +DEFINE_MTHP_STAT_ATTR(anon_swpout_fallback, MTHP_STAT_ANON_SWPOUT_FALL=
BACK);
> >
> >  static struct attribute *stats_attrs[] =3D {
> >       &anon_alloc_attr.attr,
> >       &anon_alloc_fallback_attr.attr,
> > +     &anon_swpout_attr.attr,
> > +     &anon_swpout_fallback_attr.attr,
> >       NULL,
> >  };
> >
> > diff --git a/mm/page_io.c b/mm/page_io.c
> > index a9a7c236aecc..7669452e8b4d 100644
> > --- a/mm/page_io.c
> > +++ b/mm/page_io.c
> > @@ -212,13 +212,17 @@ int swap_writepage(struct page *page, struct writ=
eback_control *wbc)
> >
> >  static inline void count_swpout_vm_event(struct folio *folio)
> >  {
> > +     long nr_pages =3D folio_nr_pages(folio);
> > +
> >  #ifdef CONFIG_TRANSPARENT_HUGEPAGE
> >       if (unlikely(folio_test_pmd_mappable(folio))) {
> >               count_memcg_folio_events(folio, THP_SWPOUT, 1);
> >               count_vm_event(THP_SWPOUT);
> >       }
> > +     if (nr_pages > 0)
> > +             count_mthp_stat(folio_order(folio), MTHP_STAT_ANON_SWPOUT=
);
>
> I think you mean "nr_pages > 1"? Although as established in my comments a=
gainst
> patch 1, its safe to call for order-0, so could just unconditionally call=
 this
> (as you effectively are at the moment).

yes. nr_pages > 1 is the correct condition. though it doesn't cause
any crash for
nr_pages > 0 as we have memory for order-0. but it is a useless job to coun=
t
for nr_pages=3D=3D1 as there is no place to show it in sysfs.

>
> >  #endif
> > -     count_vm_events(PSWPOUT, folio_nr_pages(folio));
> > +     count_vm_events(PSWPOUT, nr_pages);
> >  }
> >
> >  #if defined(CONFIG_MEMCG) && defined(CONFIG_BLK_CGROUP)
> > diff --git a/mm/vmscan.c b/mm/vmscan.c
> > index ffc4553c8615..b30e6294f82a 100644
> > --- a/mm/vmscan.c
> > +++ b/mm/vmscan.c
> > @@ -1247,6 +1247,9 @@ static unsigned int shrink_folio_list(struct list=
_head *folio_list,
> >                                               count_vm_event(
> >                                                       THP_SWPOUT_FALLBA=
CK);
> >                                       }
> > +                                     if (nr_pages > 0)
> > +                                             count_mthp_stat(get_order=
(nr_pages * PAGE_SIZE),
> > +                                                     MTHP_STAT_ANON_SW=
POUT_FALLBACK);
>
> Same comment.

agreed.  Also, it doesn't cause any crash as we have memory in mthp_stats f=
or
order-0.
i will totally remove  if (nr_pages > 0) and as we are certainly in large f=
olios
cases.

>
> >  #endif
> >                                       if (!add_to_swap(folio))
> >                                               goto activate_locked_spli=
t;
>

Thanks
Barry