[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <49c5a522-585f-4c05-6acf-0930a3b0670b@arm.com>
Date: Tue, 9 Apr 2019 08:43:01 +0100
From: Marc Zyngier <marc.zyngier@....com>
To: Guenter Roeck <linux@...ck-us.net>
Cc: linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
Russell King <linux@....linux.org.uk>,
Will Deacon <will.deacon@....com>,
Catalin Marinas <catalin.marinas@....com>,
Mark Rutland <mark.rutland@....com>,
Daniel Lezcano <daniel.lezcano@...aro.org>,
Wim Van Sebroeck <wim@...ux-watchdog.org>,
Valentin Schneider <valentin.schneider@....com>
Subject: Re: [PATCH 2/7] watchdog/sbsa: Use arch_timer_read_counter instead of
arch_counter_get_cntvct
On 08/04/2019 19:07, Guenter Roeck wrote:
> On Mon, Apr 08, 2019 at 04:49:02PM +0100, Marc Zyngier wrote:
>> Only arch_timer_read_counter will guarantee that workarounds are
>> applied. So let's use this one instead of arch_counter_get_cntvct.
>>
>> Signed-off-by: Marc Zyngier <marc.zyngier@....com>
>
> Reviewed-by: Guenter Roeck <linux@...ck-us.net>
>
> ... assuming/hoping that those counters are actually the same.
There is only a single counter. It is just that in a number of cases,
the HW will return nonsensical values, which is a bit annoying if you
end-up feeding this garbage to a watchdog.
arch_timer_read_counter() guarantees that *if* there is a workaround for
the timer, it gets applied. arch_counter_get_cntvct() only returns the
raw value, with potential side effects.
Thanks,
M.
--
Jazz is not dead. It just smells funny...
Powered by blists - more mailing lists