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] tick/sched: fix data races at tick_do_timer_cpu
Date:   Wed, 4 Mar 2020 06:24:51 -0500
Message-Id: <93F1BB73-8891-46EB-99E0-DD6C87209945@lca.pw>
References: <20200304095647.GL2596@hirez.programming.kicks-ass.net>
Cc:     Thomas Gleixner <tglx@linutronix.de>, fweisbec@gmail.com,
        mingo@kernel.org, elver@google.com, linux-kernel@vger.kernel.org
In-Reply-To: <20200304095647.GL2596@hirez.programming.kicks-ass.net>
To:     Peter Zijlstra <peterz@infradead.org>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> On Mar 4, 2020, at 4:56 AM, Peter Zijlstra <peterz@infradead.org> wrote:
>=20
> Worse:
>=20
> +    if (cpu !=3D READ_ONCE(tick_do_timer_cpu) &&
> +        (READ_ONCE(tick_do_timer_cpu) !=3D TICK_DO_TIMER_NONE ||
>=20
> Doing that read twice is just utterly stupid. And the patch is full of
> that :/
>=20
> Please stop this non-thinking 'fix' generation!

That is an oversight that could easily be fixed. However, blaming about =E2=80=
=9Cnon-thinking=E2=80=9D is rather harsh. Do you know I had spent weeks read=
ing about memory-model.txt and alike just to get better at it?=