Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
From:   Qian Cai <cai@lca.pw>
Mime-Version: 1.0 (1.0)
Subject: Re: [PATCH -next] locking/osq_lock: annotate a data race in osq_lock
Date:   Tue, 11 Feb 2020 06:57:05 -0500
Message-Id: <295818C4-C5B8-43DF-9D5B-445EBA02FC4F@lca.pw>
References: <CANpmjNPWCu+w3O8cg++X4=viVFsWNehTXzTuqbwV8-DcXXpFng@mail.gmail.com>
Cc:     Peter Zijlstra <peterz@infradead.org>,
        Ingo Molnar <mingo@redhat.com>, Will Deacon <will@kernel.org>,
        LKML <linux-kernel@vger.kernel.org>
In-Reply-To: <CANpmjNPWCu+w3O8cg++X4=viVFsWNehTXzTuqbwV8-DcXXpFng@mail.gmail.com>
To:     Marco Elver <elver@google.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> On Feb 11, 2020, at 5:16 AM, Marco Elver <elver@google.com> wrote:
>=20
> I have said this before: we're not just guarding against load/store
> tearing, although on their own, they make it deceptively easy to
> reason about data races.
>=20
> The case here seems to be another instance of a C-CAS, to avoid
> unnecessarily dirtying a cacheline.
>=20
> Here, the loop would make me suspicious, because a compiler could
> optimize out re-loading the value. Due to the smp_load_acquire,
> however, at the least we have 1 implied compiler barrier in this loop
> which means that will likely not happen.
>=20
> Before jumping to 'data_race()', I would ask again: how bad is the
> READ_ONCE? Is the generated code the same? If so, just use the
> READ_ONCE. Do you want to reason about all compiler optimizations? For
> this code here, I certainly don't want to.
>=20
> But in the end it's up to what maintainers prefer, and maybe there is
> a very compelling argument that I missed that makes the fact this is a
> data race always safe.

Yes, I feel like locking maintainers prefer data_race() rather than blindly a=
dding READ_ONCE() unless there is an strong evidence that the later is neede=
d.

Since I can=E2=80=99t prove it is strictly needed to prevent from which spec=
ific optimization, I had chosen the data_race() approach.=