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: <20210726171106.v4.1.I09866d90c6de14f21223a03e9e6a31f8a02ecbaf@changeid>
 <d6668437-5c3b-2dff-bb95-4e3132d13711@redhat.com> <6ff28cfe-1107-347b-0327-ad36e256141b@redhat.com>
 <CAE=gft7567-2Lq7raJKrOpQ8UAvXTFWwPci=_GCRPET3nS=9SA@mail.gmail.com> <20210727141832.86695e7181eb10c6e8fd0191@linux-foundation.org>
In-Reply-To: <20210727141832.86695e7181eb10c6e8fd0191@linux-foundation.org>
From:   Evan Green <evgreen@chromium.org>
Date:   Tue, 3 Aug 2021 11:00:51 -0700
Message-ID: <CAE=gft6LmTKx2JHAj7GBmph_1BrioXhmtn_xQEHtRhK+tPa6qw@mail.gmail.com>
Subject: Re: [PATCH v4] mm: Enable suspend-only swap spaces
To:     Andrew Morton <akpm@linux-foundation.org>
Cc:     David Hildenbrand <david@redhat.com>,
        Michal Hocko <mhocko@suse.com>, Pavel Machek <pavel@ucw.cz>,
        linux-api@vger.kernel.org, Alex Shi <alexs@kernel.org>,
        Alistair Popple <apopple@nvidia.com>,
        Johannes Weiner <hannes@cmpxchg.org>,
        Joonsoo Kim <iamjoonsoo.kim@lge.com>,
        "Matthew Wilcox (Oracle)" <willy@infradead.org>,
        Miaohe Lin <linmiaohe@huawei.com>,
        Minchan Kim <minchan@kernel.org>,
        Suren Baghdasaryan <surenb@google.com>,
        Vlastimil Babka <vbabka@suse.cz>,
        LKML <linux-kernel@vger.kernel.org>, linux-mm@kvack.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Tue, Jul 27, 2021 at 2:18 PM Andrew Morton <akpm@linux-foundation.org> wrote:
>
> On Tue, 27 Jul 2021 09:31:33 -0700 Evan Green <evgreen@chromium.org> wrote:
>
> > > Pavel just mentioned uswsusp, and I wonder if it would be a possible
> > > alternative to this patch.
> >
> > I think you're right that it would be possible to isolate the
> > hibernate image with uswsusp if you avoid using the SNAPSHOT_*SWAP*
> > ioctls. But I'd expect performance to suffer noticeably, since now
> > every page is making a round trip out to usermode and back. I'd still
> > very much use the HIBERNATE_ONLY flag if it were accepted, I think
> > there's value to it.
>
> The uswsusp option makes your patch a performance optimization rather
> than a feature-add.  And we do like to see quantitative testing results
> when considering a performance optimization.  Especially when the
> performance optimization is a bit icky, putting special-case testing
> all over the place, maintenance cost, additional testing effort, etc.
>
> I do think that diligence demands that we quantify the difference.  Is
> this a thing you can help with?

I'm wrong about the performance. Uswsusp is just as fast, and possibly
faster in my use case than kernel-driven hibernate. What's more,
uswsusp also helps me solve several additional problems I hadn't
tackled yet that were looming in front of me. Thanks all for your
patience and thoughtful review on this.

Patch withdrawn.
-Evan