Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <1683685251-2059-1-git-send-email-zhaoyang.huang@unisoc.com> <ZFwVypakvK7XJpFl@P9FQF9L96D.corp.robot.car>
In-Reply-To: <ZFwVypakvK7XJpFl@P9FQF9L96D.corp.robot.car>
From:   Zhaoyang Huang <huangzhaoyang@gmail.com>
Date:   Thu, 11 May 2023 10:43:48 +0800
Message-ID: <CAGWkznEa3fgZT6K+bKnd632Mt+3hCvOBoLBJx9MOcRAcq13xKw@mail.gmail.com>
Subject: Re: [PATCHv4] mm: optimization on page allocation when CMA enabled
To:     Roman Gushchin <roman.gushchin@linux.dev>
Cc:     "zhaoyang.huang" <zhaoyang.huang@unisoc.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Roman Gushchin <guro@fb.com>, Minchan Kim <minchan@kernel.org>,
        Joonsoo Kim <iamjoonsoo.kim@lge.com>, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org, ke.wang@unisoc.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Thu, May 11, 2023 at 6:08=E2=80=AFAM Roman Gushchin <roman.gushchin@linu=
x.dev> wrote:
>
> On Wed, May 10, 2023 at 10:20:51AM +0800, zhaoyang.huang wrote:
> > From: Zhaoyang Huang <zhaoyang.huang@unisoc.com>
> >
> > Let us look at the series of scenarios below with WMARK_LOW=3D25MB,WMAR=
K_MIN=3D5MB
> > (managed pages 1.9GB). We can know that current 'fixed 1/2 ratio' start=
 to use
> > CMA since C which actually has caused U&R lower than WMARK_LOW (this sh=
ould be
> > deemed as against current memory policy, that is, UNMOVABLE & RECLAIMAB=
LE should
> > either stay around WATERMARK_LOW when no allocation or do reclaim via e=
ntering
> > slowpath)
> >
> > -- Free_pages
> > |
> > |
> > -- WMARK_LOW
> > |
> > -- Free_CMA
> > |
> > |
> > --
> >
> > Free_CMA/Free_pages(MB)      A(12/30)     B(12/25)     C(12/20)
> > fixed 1/2 ratio                 N             N           Y
> > this commit                     Y             Y           Y
> >
> > Signed-off-by: Zhaoyang Huang <zhaoyang.huang@unisoc.com>
>
> I'm mostly fine with the code. The commit message is still very confusing=
 to me,
> not sure I understand what exactly this table means. And you still use "U=
&R".
I would like to highlight the scenario "A&B" where the previous fixed
1/2 ratio introduces over use of UNMOVABLE & RECLAIMABLE. I will try
to make it more clear by v5
>
> Also I'm a bit concerned about potential performance implications. Would =
be
> great to provide some benchmarks or some data.
> Probably it's ok because of we have pcp caches on top, but I'm not 100% s=
ure.
This patch helps solve my OOM issue in v5.15. Actually, It inherit the
logic of 1/2 ratio and just behave differently when free pages is
around corresponding watermark
>
> Thanks!