Re: [PATCH 1/1] mm/khugepaged: bypassing unnecessary scans with MMF_DISABLE_THP check

From: Michal Hocko <mhocko@suse.com>
To: Lance Yang <ioworker0@gmail.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
Subject: Re: [PATCH 1/1] mm/khugepaged: bypassing unnecessary scans with MMF_DISABLE_THP check
Date: Tue, 30 Jan 2024 10:35:40 +0100	[thread overview]
Message-ID: <ZbjC7KyI21ik1xpK@tiehlicka> (raw)
In-Reply-To: <CAK1f24=2YE+BCYiizkqc8rmN8NaFv_Q6ZtE+4DiFK0PpcefOrQ@mail.gmail.com>

On Tue 30-01-24 11:08:10, Lance Yang wrote:
> On Tue, Jan 30, 2024 at 10:12 AM Lance Yang <ioworker0@gmail.com> wrote:
> >
> > 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.
> 
> In the Go standard library, there is no direct encapsulation
> of the fork system call. Instead, fork and execve are
> combined into one through syscall.ForkExec.

OK, this is an important detail. Something that should be a part
of the chnagelog. It is also important to note that this is not
a correctness issue but rather an optimization to save expensive
checks on each VMA when userspace cannot prctl itself before spawning
into the new process.
-- 
Michal Hocko
SUSE Labs