Received-SPF: pass (google.com: domain of linux-kernel+bounces-152435-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.80.249 as permitted sender) client-ip=147.75.80.249;
Precedence: bulk
MIME-Version: 1.0
References: <20240417141111.77855-1-ioworker0@gmail.com> <Zh_mBxJmYe6eCA29@casper.infradead.org>
 <CAK1f24=Mrk7TFnDd=ouCrHaDH9K3VGCUAJbLH9cbn0pGncP+Hw@mail.gmail.com>
 <CAK1f24=0S8qFFNhf5h=wbbAHvYt-n563MOu=bJheLcrfcPAE_Q@mail.gmail.com> <b50af6b1-2fa7-4285-998f-88d28bf57713@redhat.com>
In-Reply-To: <b50af6b1-2fa7-4285-998f-88d28bf57713@redhat.com>
From: Lance Yang <ioworker0@gmail.com>
Date: Sun, 21 Apr 2024 10:00:31 +0800
Message-ID: <CAK1f24nN+DgU7rz0zV4DEwJwWe_CZcpR7GROMxWfWbo+cnr71g@mail.gmail.com>
Subject: Re: [PATCH 1/1] mm/vmscan: avoid split PMD-mapped THP during shrink_folio_list()
To: David Hildenbrand <david@redhat.com>
Cc: Matthew Wilcox <willy@infradead.org>, akpm@linux-foundation.org, maskray@google.com, 
	ziy@nvidia.com, ryan.roberts@arm.com, 21cnbao@gmail.com, mhocko@suse.com, 
	fengwei.yin@intel.com, zokeefe@google.com, shy828301@gmail.com, 
	xiehuan09@gmail.com, wangkefeng.wang@huawei.com, songmuchun@bytedance.com, 
	peterx@redhat.com, minchan@kernel.org, linux-mm@kvack.org, 
	linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Sun, Apr 21, 2024 at 12:32=E2=80=AFAM David Hildenbrand <david@redhat.co=
m> wrote:
>
> On 20.04.24 17:04, Lance Yang wrote:
> > On Sat, Apr 20, 2024 at 12:59=E2=80=AFPM Lance Yang <ioworker0@gmail.co=
m> wrote:
> >>
> >> Hey Matthew,
> >>
> >> Thanks for taking time to review!
> >>
> >> On Wed, Apr 17, 2024 at 11:09=E2=80=AFPM Matthew Wilcox <willy@infrade=
ad.org> wrote:
> >>>
> >>> On Wed, Apr 17, 2024 at 10:11:11PM +0800, Lance Yang wrote:
> >>>> When the user no longer requires the pages, they would use madvise(m=
adv_free)
> >>>> to mark the pages as lazy free. IMO, they would not typically rewrit=
e to the
> >>>> given range.
> >>>>
> >>>> At present, a PMD-mapped THP marked as lazyfree during shrink_folio_=
list()
> >>>> is unconditionally split, which may be unnecessary. If the THP is ex=
clusively
> >>>> mapped and clean, and the PMD associated with it is also clean, then=
 we can
> >>>> attempt to remove the PMD mapping from it. This change will improve =
the
> >>>> efficiency of memory reclamation in this case.
> >>>>
> >>>> On an Intel i5 CPU, reclaiming 1GiB of PMD-mapped THPs using
> >>>> mem_cgroup_force_empty() results in the following runtimes in second=
s
> >>>> (shorter is better):
> >>>>
> >>>> --------------------------------------------
> >>>> |     Old       |      New       |  Change  |
> >>>> --------------------------------------------
> >>>> |   0.683426    |    0.049197    |  -92.80% |
> >>>> --------------------------------------------
> >>>>
> >>>> Signed-off-by: Lance Yang <ioworker0@gmail.com>
> >>>> ---
> >>>>   include/linux/huge_mm.h |  1 +
> >>>>   include/linux/rmap.h    |  1 +
> >>>>   mm/huge_memory.c        |  2 +-
> >>>>   mm/rmap.c               | 81 +++++++++++++++++++++++++++++++++++++=
++++
> >>>>   mm/vmscan.c             |  7 ++++
> >>>>   5 files changed, 91 insertions(+), 1 deletion(-)
> >>>
> >>> I'm confused why we need all this extra code.  If we remove a folio
> >>
> >> Thanks for pointing that out!
> >>
> >> I've added a lot of extra code to rmap.c, and we don't need it
> >> for file pages - sorry. I'll reconsider where to place this code.
> >>
> >>> from the pagecache, we can just call truncate_inode_folio() and
> >>> unmap_mapping_folio() takes care of all the necessary unmappings.
> >>> Why can't you call unmap_mapping_folio() here?
> >>
> >> Thanks for your suggestion.
> >>
> >> But this change only avoids the splitting of *anon* large folios
> >> (PMD-mapped THPs) that are marked as lazyfree during
> >> shrink_folio_list().
> >>
> >> IIUC, in some cases, we cannot unmap the THP marked as lazyfree
> >> here, such as when it's not exclusively mapped, dirty, pinned, etc.
> >
> > I=E2=80=99d like to make a correction.
> >
> > IMO, we can unmap the THP that is not exclusively mapped, but
> > ensuring folio_ref_count() equals folio_mapcount() +1.
>

Hey David,

Thanks a lot for clarifying!

> You must follow the exact same logic as in try_to_unmap_one() I guess.

Agreed. I'll take a closer look at try_to_unmap_one() and follow
the exact same logic - thanks!

>
> That is, unmap the page, syncing against concurrent GUP-fast. Then,
> check mapcount vs. refcount. If there are unexpected references, remap
> the page (set_pte_at).

Yep, I understood. Could you please provide some suggestions on
where to place the exact same logic?

Thanks again for your time!
Lance

>
> --
> Cheers,
>
> David / dhildenb
>