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;
Date:   Sat, 04 Sep 2021 09:42:00 +1000
From:   Nicholas Piggin <npiggin@gmail.com>
Subject: Re: Is it possible to implement the per-node page cache for
 programs/libraries?
To:     Matthew Wilcox <willy@infradead.org>
Cc:     akpm@linux-foundation.org, linux-kernel@vger.kernel.org,
        linux-mm@kvack.org,
        Shijie Huang <shijie@amperemail.onmicrosoft.com>,
        song.bao.hua@hisilicon.com, torvalds@linux-foundation.org,
        viro@zeniv.linux.org.uk, Frank Wang <zwang@amperecomputing.com>
References: <a2f423cf-9413-6bc8-e4d8-92374fc0449e@amperemail.onmicrosoft.com>
        <YS7yjcqA6txFHd99@casper.infradead.org>
        <1630552995.2mupnzoqzs.astroid@bobo.none>
        <YTCktV3KF9PzIACU@casper.infradead.org>
        <1630652670.aplcvu6g23.astroid@bobo.none>
        <YTJxFgD0kKPs51dz@casper.infradead.org>
In-Reply-To: <YTJxFgD0kKPs51dz@casper.infradead.org>
MIME-Version: 1.0
Message-Id: <1630712371.zxj5zdhheu.astroid@bobo.none>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Excerpts from Matthew Wilcox's message of September 4, 2021 5:01 am:
> On Fri, Sep 03, 2021 at 05:10:31PM +1000, Nicholas Piggin wrote:
>> Excerpts from Matthew Wilcox's message of September 2, 2021 8:17 pm:
>> > On Thu, Sep 02, 2021 at 01:25:36PM +1000, Nicholas Piggin wrote:
>> >> > I have been thinking about this a bit; one of our internal performa=
nce
>> >> > teams flagged the potential performance win to me a few months ago.
>> >> > I don't have a concrete design for text replication yet; there have=
 been
>> >> > various attempts over the years, but none were particularly compell=
ing.
>> >>=20
>> >> What was not compelling about it?
>> >=20
>> > It wasn't merged, so clearly it wasn't compelling enough?
>>=20
>> Ha ha. It sounded like you had some reasons you didn't find it=20
>> particularly compelling :P
>=20
> I haven't studied it in detail, but it seems to me that your patch (from
> 2007!) chooses whether to store pages or pcache_desc pointers in i_pages.
> Was there a reason you chose to do it that way instead of having per-node
> i_mapping pointers?

What Linus said. The patch was obviously mechanism only and more=20
heuristics would need to be done (in that case you could have per inode=20
hints or whatever).

> (And which way would you choose to do it now, given
> the infrastructure we have now?)

I'm not aware of anything new that would change it fundamentally.

Thanks,
Nick