Re: [PATCH 4/4] Documentation: Add device tree bindings for Freescale FTM PWM

[Stephen Warren <swarren@wwwdotorg.org>]

On 08/23/2013 01:36 AM, Thierry Reding wrote:
> On Thu, Aug 22, 2013 at 08:26:10AM +0200, Sascha Hauer wrote:
>> On Thu, Aug 22, 2013 at 02:55:42AM +0000, Xiubo Li-B47053 wrote:
>>> Hi Tomasz,
>>> 
>>> Thanks for your comments.
>>> 
>>> 
>>>> Could you explain meaning of this property more precisely?
>>>> I'm interested especially how is this related to the PWM IP
>>>> block and boards.
>>>> 
>>> 
>>> Yes. There are 8 channels most. While the pinctrls of 4th and
>>> 5th channels could be used by uart's Rx and Tx, then these 2
>>> channels won't be used for pwm output, so there will be 6
>>> channels available by the pwm. Thus, the pwm chip will register
>>> only 6 pwms(6 channels) most("fsl,pwm-channel-orders = {0 1 2 3
>>> 6 7}").And also the "fsl,pwm-channel-number" will be 6.
>> 
>> If the chip has eight PWMs I would register all of them. If some
>> of them are not routed out by the pinmux then just nothing
>> happens if you use them. In a sane devicetree they won't be
>> referenced anyway when they are not routed out of the SoC.
> 
> In that case, shouldn't this be hooked up to the pinctrl subsystem
> as well? As I understand the above, the logical thing would be for
> each PWM channel's .request() operation to configure the pinmuxing
> appropriately. And if it can't be configured as necessary then
> .request() should return an error (or propagate the error from the
> pinctrl subsystem).

I think the pin-muxing should be static, i.e. set up when the PWM
device as a whole probe()s, rather than being twiddled at request/free
time. Certainly the pinmux support in the device core is now set up to
acquire the default state right before probe(). I don't see a need to
do anything custom here.