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;
From:   Thomas Gleixner <tglx@linutronix.de>
To:     Paul Cercueil <paul@crapouillou.net>
Cc:     Maarten ter Huurne <maarten@treewalker.org>,
        Daniel Lezcano <daniel.lezcano@linaro.org>, od@zcrc.me,
        linux-kernel@vger.kernel.org, Mathieu Malaterre <malat@debian.org>,
        Artur Rojek <contact@artur-rojek.eu>
Subject: Re: [PATCH v3] clocksource: Add driver for the Ingenic JZ47xx OST
In-Reply-To: <1579121936.3.1@crapouillou.net>
References: <1579110897.3.0@crapouillou.net> <87y2u8xzq0.fsf@nanos.tec.linutronix.de> <1579121936.3.1@crapouillou.net>
Date:   Wed, 15 Jan 2020 22:50:19 +0100
Message-ID: <87muaoxuck.fsf@nanos.tec.linutronix.de>
MIME-Version: 1.0
Content-Type: text/plain
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Paul Cercueil <paul@crapouillou.net> writes:
> Le mer., janv. 15, 2020 at 20:54, Thomas Gleixner <tglx@linutronix.de> 
>> That lock still a massive contention point as clock readouts can be 
>> pretty
>> frequent depending on workloads. Just think about tracing ...
>> 
>> So I really would avoid both the lock and that ugly 64bit readout 
>> thing.
>
> The 64bit readout thing is gone in V3.
>
> The lock cannot go away unless we have a way to retrieve the underlying 
> mmio pointer from the regmap, which the regmap maintainers will never 
> accept. So I can't really change that now. Besides, 
> drivers/clocksource/ingenic-timer.c also registers a clocksource that's 
> read with the regmap, and nobody complained.

I don't complain. I just told you that a spinlock in that code path is
really suboptimal.

I missed the one in the other driver, but the same problem exists there.

Thanks,

        tglx