Received-SPF: pass (google.com: domain of linux-kernel+bounces-120565-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Precedence: bulk
MIME-Version: 1.0
References: <20240326185032.72159-1-ryncsn@gmail.com> <20240326185032.72159-11-ryncsn@gmail.com>
 <CAGsJ_4yLAYUGYzWJwe9_LmqOcCzrz8-UKhdRDjTzgVQF7Z-xZA@mail.gmail.com>
In-Reply-To: <CAGsJ_4yLAYUGYzWJwe9_LmqOcCzrz8-UKhdRDjTzgVQF7Z-xZA@mail.gmail.com>
From: Kairui Song <ryncsn@gmail.com>
Date: Wed, 27 Mar 2024 16:44:14 +0800
Message-ID: <CAMgjq7DOd16kyRngkz2RKQKf71QErAnmZy+PAsWp76V+qwykQA@mail.gmail.com>
Subject: Re: [RFC PATCH 10/10] mm/swap: optimize synchronous swapin
To: Barry Song <21cnbao@gmail.com>
Cc: linux-mm@kvack.org, "Huang, Ying" <ying.huang@intel.com>, Chris Li <chrisl@kernel.org>, 
	Minchan Kim <minchan@kernel.org>, Barry Song <v-songbaohua@oppo.com>, 
	Ryan Roberts <ryan.roberts@arm.com>, Yu Zhao <yuzhao@google.com>, SeongJae Park <sj@kernel.org>, 
	David Hildenbrand <david@redhat.com>, Yosry Ahmed <yosryahmed@google.com>, 
	Johannes Weiner <hannes@cmpxchg.org>, Matthew Wilcox <willy@infradead.org>, Nhat Pham <nphamcs@gmail.com>, 
	Chengming Zhou <zhouchengming@bytedance.com>, Andrew Morton <akpm@linux-foundation.org>, 
	linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Wed, Mar 27, 2024 at 4:09=E2=80=AFPM Barry Song <21cnbao@gmail.com> wrot=
e:
>
> On Wed, Mar 27, 2024 at 8:06=E2=80=AFAM Kairui Song <ryncsn@gmail.com> wr=
ote:
> >
> > From: Kairui Song <kasong@tencent.com>
> >
> > Interestingly the major performance overhead of synchronous is actually
> > from the workingset nodes update, that's because synchronous swap in
> > keeps adding single folios into a xa_node, making the node no longer
> > a shadow node and have to be removed from shadow_nodes, then remove
> > the folio very shortly and making the node a shadow node again,
> > so it has to add back to the shadow_nodes.
>
> Hi Kairui,
>
> Thank you for clarifying this. I'm unsure how it relates to SWP_SYNCHRONO=
US_IO.
> Does this observation apply universally to all instances where
> __swap_count(entry)
> =3D=3D 1, even on devices not using SYNCHRONOUS_IO?

Hi Barry

I was testing using zero pages on ZRAM so the performance issue is
much more obvious.

For non SYNCHRONOUS_IO devices, they don't drop swap cache immediately
unless swap is half full, so a shadow node will be removed from
shadow_nodes on first swapin, but usually won't be added/removed
repeatedly.

I think the logic that "never drop swapcache even if swap count is 1",
then suddenly switch to "always drop swap cache when swap count is 1"
when swap is half full is not a good solution... Maybe some generic
optimization can be applied for that part too.