RE: [PATCH v23 3/4] pwm: Add support for RZ/G2L GPT

[Biju Das <biju.das.jz@bp.renesas.com>]

Hi Uwe,

Thanks for the feedback.

> -----Original Message-----
> From: Uwe Kleine-König <ukleinek@kernel.org>
> Sent: 21 February 2025 09:08
> Subject: Re: [PATCH v23 3/4] pwm: Add support for RZ/G2L GPT
> 
> Hello Biju,
> 
> On Thu, Jan 09, 2025 at 09:32:58AM +0000, Biju Das wrote:
> > > On Wed, Jan 08, 2025 at 02:13:09PM +0000, Biju Das wrote:
> > > > Please let me know, if there is any feedback for this patch series
> > > > or any new API to be adapted in next kernel version which
> > > > simplifies the code, So that I can send next
> > > version.
> > >
> > > I didn't look, but if you're ambitious you can convert your driver to the waveform callbacks.
> >
> > Is it ok, after the acceptance of initial version, will switch to
> > waveform callbacks to enable the users of GPT(Reason: lot of customers are using GPT for backlight
> for LCD panels)?
> >
> > Please let me know.
> 
> I'd still accept your driver with the legacy callbacks, mostly because your effort is already that old
> (which I consider my fault mostly).

OK. 

> 
> > Is there any documentation available for waveform callback? What
> > scenarios we can use Waveform callbacks compared to traditional one?
> 
> There is no nice documentation available yet. The improvements include:
> 
>  - arbitrary offsets for the duty cycle, so the active phase doesn't
>    need to be at the start or the end of a period any more
>  - consumers can query the result of a configuration request (before and
>    also after the request was issued).
> 
> Both are of little importance if your focus is mostly backlights.

Mostly RZ/G2L GPT users are backlights. So, from RZ/G2L usecase point, legacy
callback is sufficient.

Later, I will switch to waveform callbacks, for complex GPT usecase support for next
SoCs(eg: RZ/G3E).

Cheers,
Biju