On 8/7/22 02:20, Andy Shevchenko wrote:
On Sun, Aug 7, 2022 at 10:38 AM Uwe Kleine-König
<u.kleine-koenig@xxxxxxxxxxxxxx> wrote:
at a quick glance this looks nice. I wonder if it makes sense to split
the patch. For example the change
- ctx->pwm = devm_of_pwm_get(dev, dev->of_node, NULL);
+ ctx->pwm = devm_pwm_get(dev, NULL);
could stand alone. Also I think this change is the relevant part in
patch #1 that makes patches #2 and #3 possible.
True.
When this patch doesn't get split, the series needs some coordination,
as patch #1 is for hwmon and patches #2 and #3 are for pwm.
Splitting the series into:
hwmon: (pwm-fan) Use of devm_pwm_get() instead of devm_of_pwm_get()
pwm: core: Get rid of unused devm_of_pwm_get()
pwm: core: Make of_pwm_get() static
for pwm and the remainder of this patch for hwmon might make application
of the changes here easier to coordinate.
Either way it will need the hwmon maintainer ACKs or alike.
Since we have (plenty of) time I will wait a bit for hwmon maintainers
to react. Guenter, what would you prefer?
I have a substantial number of patches pending for the pwm-fan driver.
Some of those will conflict with this patch. I'll have to spend more time
to be able to understand the implications.
Guenter
