Re: [PATCH 36/43] pwm: ep93xx: drop legacy pinctrl

["Uwe Kleine-König" <u.kleine-koenig@pengutronix.de>]

[-- Attachment #1: Type: text/plain, Size: 3123 bytes --]

Hello Nikita,

On Tue, May 16, 2023 at 01:43:27PM +0300, Nikita Shubin wrote:
> On Mon, 2023-05-15 at 16:32 +0200, Uwe Kleine-König wrote:
> > On Mon, Apr 24, 2023 at 03:34:52PM +0300, Nikita Shubin wrote:
> > > diff --git a/drivers/pwm/pwm-ep93xx.c b/drivers/pwm/pwm-ep93xx.c
> > > index 8bfe6cfbb3db..657adb011aeb 100644
> > > --- a/drivers/pwm/pwm-ep93xx.c
> > > +++ b/drivers/pwm/pwm-ep93xx.c
> > > @@ -45,20 +45,6 @@ static inline struct ep93xx_pwm
> > > *to_ep93xx_pwm(struct pwm_chip *chip)
> > >         return container_of(chip, struct ep93xx_pwm, chip);
> > >  }
> > >  
> > > -static int ep93xx_pwm_request(struct pwm_chip *chip, struct
> > > pwm_device *pwm)
> > > -{
> > > -       struct platform_device *pdev = to_platform_device(chip-
> > > >dev);
> > > -
> > > -       return ep93xx_pwm_acquire_gpio(pdev);
> > 
> > I didn't get the whole series and didn't spot a relevant followup
> > change
> 
> https://lore.kernel.org/all/20230424123522.18302-1-nikita.shubin@maquefel.me/

Yeah, I looked there, but didn't find it. Applied the whole series now
and found "ARM: ep93xx: soc: drop defines". A few things I noticed while
doing so:

 - git am warns about new blank lines at EOF in several patches.

 - b4 am 20230424123522.18302-1-nikita.shubin@maquefel.me
   warns about broken DKIM signatures. The copy I got directly via Cc is
   OK though. The relevant problem is that your To: header is empty but
   part of the signed payload + the copy I got via vger.kernel.org had
   the To header mangled to

	To:     unlisted-recipients:; (no To-header on input) 

   This results in:

	$ curl -s https://lore.kernel.org/lkml/20230424123522.18302-37-nikita.shubin@maquefel.me/raw | dkimverify
	signature verification failed
	$ curl -s https://lore.kernel.org/lkml/20230424123522.18302-37-nikita.shubin@maquefel.me/raw | sed 's/^To:.*/To:/' | dkimverify 
	signature ok

   I don't know who is to blame here (i.e. is an empty To allowed?) but
   I'd recommend to put the people you want to merge the patches into
   the To header anyhow.

> > on lore.k.o, so: I assume ep93xx_pwm_acquire_gpio() and
> > ep93xx_pwm_release_gpio() will be unused in the end? Do you drop
> > them?
> > 
> > I assume this series target to be taken via arm-soc (once the review
> > feedback is positive)?

You didn't reply to that one. Still assuming this to be true, I'll mark
this patch as handled-elsewhere in the PWM patchwork.

> > I wonder if this change breaks non-dt machine support?
> 
> The aim for the whole series is fully converting to dt, this means
> platform files will be dropped.
> 
> The v1 series tries not to break anything until platform removal
> commit, before this commit non-dt version should be compilable and
> fully functional.

OK, the pwm patch looks fine to me,

Acked-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>

Best regards and thanks for your efforts,
Uwe

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

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]