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;
MIME-Version: 1.0
References: <20200527161743.15972-1-brgl@bgdev.pl> <58fca7418c8d18392562aaad2c3a6634@kernel.org>
In-Reply-To: <58fca7418c8d18392562aaad2c3a6634@kernel.org>
From:   Bartosz Golaszewski <bgolaszewski@baylibre.com>
Date:   Wed, 27 May 2020 19:06:26 +0200
Message-ID: <CAMpxmJVPB2+hD8mZ+Bo9=UVEist7BSJrZej5O9_9a_SmWyTanA@mail.gmail.com>
Subject: Re: [PATCH] irqchip/irq-mtk-sysirq: drop unnecessary spinlock
To:     Marc Zyngier <maz@kernel.org>
Cc:     Bartosz Golaszewski <brgl@bgdev.pl>,
        Thomas Gleixner <tglx@linutronix.de>,
        Jason Cooper <jason@lakedaemon.net>,
        Matthias Brugger <matthias.bgg@gmail.com>,
        LKML <linux-kernel@vger.kernel.org>,
        arm-soc <linux-arm-kernel@lists.infradead.org>,
        linux-mediatek@lists.infradead.org,
        Fabien Parent <fparent@baylibre.com>,
        Stephane Le Provost <stephane.leprovost@mediatek.com>,
        Pedro Tsai <pedro.tsai@mediatek.com>,
        Andrew Perepech <andrew.perepech@mediatek.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

=C5=9Br., 27 maj 2020 o 18:38 Marc Zyngier <maz@kernel.org> napisa=C5=82(a)=
:
>
> Sight... Do you realize that these two locks do not protect the same
> thing at all? One protects the interrupt data, and the other protects
> the MMIO register which is shared between multiple interrupts, and
> on which the driver performs a RMW.
>
> Thanks to the removal of this spinlock, two irq_set_type() can execute
> in parallel and silently corrupt the register. Not exactly an
> improvement.
>

Eek I missed the fact that the internal lock is tied to the chip, not
the interrupt. In that case I'll convert this spinlock to a raw one.

Bart