Re: [PATCH] pwm: Rename pwm_get_state() to better reflect its semantic

From: Thierry Reding <thierry.reding@gmail.com>
To: "Uwe Kleine-König" <u.kleine-koenig@pengutronix.de>
Cc: Alexandre Belloni <alexandre.belloni@bootlin.com>,
	linux-doc@vger.kernel.org, David Airlie <airlied@linux.ie>,
	Michael Turquette <mturquette@baylibre.com>,
	linux-fbdev@vger.kernel.org, dri-devel@lists.freedesktop.org,
	Nicolas Ferre <nicolas.ferre@microchip.com>,
	Lee Jones <lee.jones@linaro.org>,
	linux-stm32@st-md-mailman.stormreply.com,
	Daniel Thompson <daniel.thompson@linaro.org>,
	Jonathan Corbet <corbet@lwn.net>,
	Ludovic Desroches <ludovic.desroches@microchip.com>,
	linux-clk@vger.kernel.org, linux-rockchip@lists.infradead.org,
	Chen-Yu Tsai <wens@csie.org>, NXP Linux Team <linux-imx@nxp.com>,
	linux-input@vger.kernel.org,
	Sascha Hauer <s.hauer@pengutronix.de>,
	linux-pwm@vger.kernel.org,
	Alexandre Torgue <alexandre.torgue@st.com>,
	intel-gfx@lists.freedesktop.org, Mark Brown <broonie@kernel.org>,
	Rodrigo Vivi <rodrigo.vivi@intel.com>,
	Fabrice Gasnier <fabrice.gasnier@st.com>,
	Pengutronix Kernel Team <kernel@pengutronix.de>,
	linux-arm-kernel@lists.infradead.org,
	Maxime Coquelin <mcoquelin.stm32@gmail.com>,
	Jernej Skrabec <jernej.skrabec@siol.net>,
	Support Opensource <support.opensource@diasemi.com>,
	Stephen Boyd <sboyd@kernel.org>,
	Jingoo Han <jingoohan1@gmail.com>,
	Dmitry Torokhov <dmitry.torokhov@gmail.com>,
	Liam Girdwood <lgirdwood@gmail.com>,
	linux-kernel@vger.kernel.org, Shawn Guo <shawnguo@kernel.org>,
	Claudiu Beznea <claudiu.beznea@microchip.com>
Subject: Re: [PATCH] pwm: Rename pwm_get_state() to better reflect its semantic
Date: Fri, 9 Apr 2021 14:09:16 +0200	[thread overview]
Message-ID: <YHBD7AhOJGyELpVZ@orome.fritz.box> (raw)
In-Reply-To: <20210406134356.dda74heeshkwdarw@pengutronix.de>

[-- Attachment #1: Type: text/plain, Size: 2984 bytes --]

On Tue, Apr 06, 2021 at 03:43:56PM +0200, Uwe Kleine-König wrote:
> Hello Thierry,
> 
> On Tue, Apr 06, 2021 at 01:16:31PM +0200, Thierry Reding wrote:
> > On Tue, Apr 06, 2021 at 09:30:36AM +0200, Uwe Kleine-König wrote:
> > > Given that lowlevel drivers usually cannot implement exactly what a
> > > consumer requests with pwm_apply_state() there is some rounding involved.
> > > 
> > > pwm_get_state() traditionally returned the setting that was requested most
> > > recently by the consumer (opposed to what was actually implemented in
> > > hardware in reply to the last request). To make this semantic obvious
> > > rename the function.
> > > 
> > > Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> > > ---
> > >  Documentation/driver-api/pwm.rst           |  6 +++-
> > >  drivers/clk/clk-pwm.c                      |  2 +-
> > >  drivers/gpu/drm/i915/display/intel_panel.c |  4 +--
> > >  drivers/input/misc/da7280.c                |  2 +-
> > >  drivers/input/misc/pwm-beeper.c            |  2 +-
> > >  drivers/input/misc/pwm-vibra.c             |  4 +--
> > >  drivers/pwm/core.c                         |  4 +--
> > >  drivers/pwm/pwm-atmel-hlcdc.c              |  2 +-
> > >  drivers/pwm/pwm-atmel.c                    |  2 +-
> > >  drivers/pwm/pwm-imx27.c                    |  2 +-
> > >  drivers/pwm/pwm-rockchip.c                 |  2 +-
> > >  drivers/pwm/pwm-stm32-lp.c                 |  4 +--
> > >  drivers/pwm/pwm-sun4i.c                    |  2 +-
> > >  drivers/pwm/sysfs.c                        | 18 ++++++------
> > >  drivers/regulator/pwm-regulator.c          |  4 +--
> > >  drivers/video/backlight/pwm_bl.c           | 10 +++----
> > >  include/linux/pwm.h                        | 34 ++++++++++++++--------
> > >  17 files changed, 59 insertions(+), 45 deletions(-)
> > 
> > Honestly, I don't think this is worth the churn. If you think people
> > will easily get confused by this then a better solution might be to more
> > explicitly document the pwm_get_state() function to say exactly what it
> > returns.
> 
> I'm not so optimistic that people become aware of the semantic just
> because there is documentation describing it and I strongly believe that
> a good name for functions is more important than accurate documentation.
> 
> If you don't agree, what do you think about the updated wording in
> Documentation/driver-api/pwm.rst?

Yeah, that clarifies this a bit. I can apply that hunk of the patch
separately.

> > But there's no need to make life difficult for everyone by
> > renaming this to something as cumbersome as this.
> 
> I don't expect any merge conflicts (and if still a problem occurs
> resolving should be trivial enough). So I obviously don't agree to your
> weighing.

I wasn't talking about merge conflicts but instead about the extra churn
of changing all consumers and having to type all these extra characters
for no benefit.

Thierry

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]