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: <20230421-seilbahn-vorpreschen-bd73ac3c88d7@brauner>
 <CAHk-=wgyL9OujQ72er7oXt_VsMeno4bMKCTydBT1WSaagZ_5CA@mail.gmail.com>
 <6882b74e-874a-c116-62ac-564104c5ad34@kernel.dk> <CAHk-=wiQ8g+B0bCPJ9fxZ+Oa0LPAUAyryw9i+-fBUe72LoA+QQ@mail.gmail.com>
 <CAHk-=wgGzwaz2yGO9_PFv4O1ke_uHg25Ab0UndK+G9vJ9V4=hw@mail.gmail.com>
 <2e7d4f63-7ddd-e4a6-e7eb-fd2a305d442e@kernel.dk> <69ec222c-1b75-cdc1-ac1b-0e9e504db6cb@kernel.dk>
 <CAHk-=wiaFUoHpztu6Zf_4pyzH-gzeJhdCU0MYNw9LzVg1-kx8g@mail.gmail.com>
 <CAHk-=wjSuGTLrmygUSNh==u81iWUtVzJ5GNSz0A-jbr4WGoZyw@mail.gmail.com>
 <20230425194910.GA1350354@hirez.programming.kicks-ass.net> <CAHk-=wjNfkT1oVLGbe2=Vymp66Ht=tk+YKa9gUL4T=_hA_JLjg@mail.gmail.com>
In-Reply-To: <CAHk-=wjNfkT1oVLGbe2=Vymp66Ht=tk+YKa9gUL4T=_hA_JLjg@mail.gmail.com>
From:   Linus Torvalds <torvalds@linux-foundation.org>
Date:   Tue, 25 Apr 2023 13:29:25 -0700
Message-ID: <CAHk-=wjVi5U_DF2Y+fnBuy=RH9OKfK7-MRmpnuZP2wmCdNCqYw@mail.gmail.com>
Subject: Re: [GIT PULL] pipe: nonblocking rw for io_uring
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     Jens Axboe <axboe@kernel.dk>, Ingo Molnar <mingo@redhat.com>,
        Will Deacon <will@kernel.org>,
        Christian Brauner <brauner@kernel.org>,
        linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
        Mark Rutland <mark.rutland@arm.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Tue, Apr 25, 2023 at 12:58=E2=80=AFPM Linus Torvalds
<torvalds@linux-foundation.org> wrote:
>
> Ok. I'll try to remember this, but maybe it might be worth documenting.

We might document it by just making it clear that it's not that we
want to read it "once", it's that we want to have a stable value.

There might be other situations where that is all we want.

IOW, maybe we could have something like

  #define READ_STABLE(x) \
      ({ __auto_type __val =3D (x); OPTIMIZER_HIDE_VAR(__val); __val; })

instead - although from a quick look, the code generation is pretty
much exactly the same.

I dunno. Just throwing that idea out there as a "if reading _once_
isn't the issue, maybe we shouldn't make the code look like it is"...

                Linus