Re: [PATCH v8 4/8] dt-bindings: pwm: Support new PWM_USAGE_POWER flag

[Uwe Kleine-König]

On Fri, Apr 16, 2021 at 05:54:29PM +0200, Clemens Gruber wrote:
> On Fri, Apr 16, 2021 at 03:55:11PM +0200, Thierry Reding wrote:
> > On Mon, Apr 12, 2021 at 03:27:41PM +0200, Clemens Gruber wrote:
> > > Add the flag and corresponding documentation for PWM_USAGE_POWER.
> > > 
> > > Cc: Rob Herring <robh+dt@xxxxxxxxxx>
> > > Signed-off-by: Clemens Gruber <clemens.gruber@xxxxxxxxxxxx>
> > > ---
> > >  Documentation/devicetree/bindings/pwm/pwm.txt | 3 +++
> > >  include/dt-bindings/pwm/pwm.h                 | 1 +
> > >  2 files changed, 4 insertions(+)
> > 
> > Rob, what are your thoughts on this? I've been thinking about this some
> > more and I'm having second thoughts about putting this into device tree
> > because it doesn't actually describe a property of the PWM hardware but
> > rather a use-case specific hint. It's a bit of a gray area because this
> > is just part of the PWM specifier which already has use-case specific
> > "configuration", such as the period and the polarity.

This is something I'd prefer over making it part of the device tree API.
I still don't think it's a good idea but when we keep it in-kernel we
can at least easier modify it in the future.

> > Perhaps a better place for this is within the PWM API? We could add the
> > same information into struct pwm_state and then consumers that don't
> > care about specifics of the signal (such as pwm-backlight) can set that
> > flag when they request a state to be applied.
> 
> I just want to note that in my opinion, this is not a flag that is
> changed often, so is it really a good idea to require setting this
> wherever PWM state is applied? Also, this can't be read-out in
> .get_state.

Not being able to read it out isn't a problem in my eyes.

> Thierry: If this discussion carries on and a v10 is required: Could you
> maybe merge the uncontroversial patches 1 to 3 of v9 separately and
> maybe get those in 5.12 ? Patches 4 to 8 can probably wait for 5.13 and
> have some time in linux-next.

I'm ok in getting those into next now and than into the upcoming merge
window. That won't make them part of 5.12 however, but 5.13-rc1. IMHO
patches 7 and 8 can go in, too.

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | https://www.pengutronix.de/ |

Attachment: signature.asc
Description: PGP signature