Received-SPF: pass (google.com: domain of linux-kernel+bounces-43800-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Precedence: bulk
MIME-Version: 1.0
References: <20240129054551.57728-1-ioworker0@gmail.com> <ZbfSP6xt6JAE2NrH@tiehlicka>
In-Reply-To: <ZbfSP6xt6JAE2NrH@tiehlicka>
From: Lance Yang <ioworker0@gmail.com>
Date: Tue, 30 Jan 2024 10:12:19 +0800
Message-ID: <CAK1f24miiADZCpugOa4QUmerG70kOsMT97Zvmy=5ifOG4mW=+g@mail.gmail.com>
Subject: Re: [PATCH 1/1] mm/khugepaged: bypassing unnecessary scans with
 MMF_DISABLE_THP check
To: Michal Hocko <mhocko@suse.com>
Cc: akpm@linux-foundation.org, zokeefe@google.com, david@redhat.com, 
	songmuchun@bytedance.com, shy828301@gmail.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

Hey Michal,

Thanks for taking time to review!

On some servers within our company, we deploy a
daemon responsible for monitoring and updating
local applications. Some applications prefer not to
use THP, so the daemon calls prctl to disable THP
before fork/exec. Conversely, for other applications,
the daemon calls prctl to enable THP before fork/exec.

Ideally, the daemon should invoke prctl after the fork,
but its current implementation follows the described
approach.

BR,
Lance

On Tue, Jan 30, 2024 at 12:28=E2=80=AFAM Michal Hocko <mhocko@suse.com> wro=
te:
>
> On Mon 29-01-24 13:45:51, Lance Yang wrote:
> > khugepaged scans the entire address space in the
> > background for each given mm, looking for
> > opportunities to merge sequences of basic pages
> > into huge pages. However, when an mm is inserted
> > to the mm_slots list, and the MMF_DISABLE_THP flag
> > is set later, this scanning process becomes
> > unnecessary for that mm and can be skipped to avoid
> > redundant operations, especially in scenarios with
> > a large address space.
>
> Is this a real problem? I thought that the prctl is called
> on the parent before fork/exec. Or are you aware of any
> applications which do call prctl late enough that the race
> would be actually observable?
> --
> Michal Hocko
> SUSE Labs