Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
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 v2] locking/osq_lock: annotate a data race in osq_lock
Date:   Mon, 11 May 2020 13:10:27 -0400
Message-Id: <D9C5B7E9-5927-4015-BC7C-202585C5649E@lca.pw>
References: <20200511165441.GB23081@willie-the-truck>
Cc:     "Paul E. McKenney" <paulmck@kernel.org>,
        Elver Marco <elver@google.com>,
        LKML <linux-kernel@vger.kernel.org>,
        Ingo Molnar <mingo@redhat.com>,
        "Peter Zijlstra (Intel)" <peterz@infradead.org>
In-Reply-To: <20200511165441.GB23081@willie-the-truck>
To:     Will Deacon <will@kernel.org>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> On May 11, 2020, at 12:54 PM, Will Deacon <will@kernel.org> wrote:
>=20
> Hmm, I don't see how it can remove the cmpxchg(). Do you have a link to th=
at
> discussion, please?

lore.kernel.org/lkml/20200211124753.GP14914@hirez.programming.kicks-ass.net

Correction =E2=80=94 if compilers could prove =E2=80=9Dprev->next !=3D node=E2=
=80=9D is always true, that cmpxchg() would not run. cpu_relax() should be s=
ufficient to keep that =E2=80=9Cif statement=E2=80=9D been optimized away in=
 any case.=