From mboxrd@z Thu Jan  1 00:00:00 1970
From: daniel.lezcano@linaro.org (Daniel Lezcano)
Date: Thu, 15 Jun 2017 22:52:59 +0200
Subject: [PATCH] clockevents/tcb_clksrc: implement suspend/resume
In-Reply-To: <20170615194044.6x6c6icnsrxgpscv@piout.net>
References: <20170411154826.5883-1-alexandre.belloni@free-electrons.com>
 <7100599f-529a-7d56-f43f-01513293d313@linaro.org>
 <20170615194044.6x6c6icnsrxgpscv@piout.net>
Message-ID: <77f1998e-03c5-f1ca-b1c3-ece2f3052683@linaro.org>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

On 15/06/2017 21:40, Alexandre Belloni wrote:
> On 14/04/2017 at 21:13:36 +0200, Daniel Lezcano wrote:
>>> +void tc_clksrc_resume(struct clocksource *cs)
>>> +{
>>> +	int i;
>>> +
>>> +	for (i = 0; i < 3; i++) {
>>
>> s/3/ARRAY_SIZE(tcb_cache)/
>>
>>> +		__raw_writel(tcb_cache[i].cmr, tcaddr + ATMEL_TC_REG(i, CMR));
>>
>> Why __raw_writel?
>>
> 
> Ok, I got to the bottom of that question and I think it is worth
> answering it. __raw_{read,write}l were necessary to make the driver work
> on AVR32, because its core is BE and the IP LE and the regular
> readl/writel are (were) not doing the proper conversion.
> 
> This was supposed to be changed in a patch that was never applied:
> http://lists.infradead.org/pipermail/linux-arm-kernel/2015-March/331775.html
> 
> But everything is fine, as AVR32 is now removed from the kernel. I think
> I'll switch the driver to regular readl/writel, using the _relaxed
> version in the hot path. Is that fine for you?

Yes.

> I'll also do so in the rework if at some point we can agree on some
> bindings, I'll try to address that soon too.

Ok, thanks.

  -- Daniel

-- 
 <http://www.linaro.org/> Linaro.org ? Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog