Date:   Wed, 24 Aug 2022 18:02:25 +0200
From:   simon.guinot@...uanux.org
To:     Hans de Goede <hdegoede@...hat.com>
Cc:     Henning Schild <henning.schild@...mens.com>,
        Andy Shevchenko <andriy.shevchenko@...ux.intel.com>,
        Linus Walleij <linus.walleij@...aro.org>,
        Bartosz Golaszewski <brgl@...ev.pl>,
        Pavel Machek <pavel@....cz>, Mark Gross <markgross@...nel.org>,
        Lee Jones <lee@...nel.org>, linux-gpio@...r.kernel.org,
        linux-kernel@...r.kernel.org, linux-leds@...r.kernel.org,
        platform-driver-x86@...r.kernel.org,
        Sheng-Yuan Huang <syhuang3@...oton.com>,
        Tasanakorn Phaipool <tasanakorn@...il.com>
Subject: Re: [PATCH v4 1/5] gpio-f7188x: Add GPIO support for Nuvoton NCT6116

On Wed, Aug 24, 2022 at 04:24:46PM +0200, Hans de Goede wrote:
> Hi,
> 
> On 8/24/22 16:17, Henning Schild wrote:
> > Am Wed, 24 Aug 2022 15:54:28 +0200
> > schrieb Hans de Goede <hdegoede@...hat.com>:
> > 
> >> Hi Henning,
> >>
> >> On 8/24/22 15:50, Henning Schild wrote:
> >>> Am Wed, 24 Aug 2022 15:10:55 +0200
> >>> schrieb simon.guinot@...uanux.org:
> >>>   
> >>>> On Tue, Aug 23, 2022 at 04:54:59PM +0200, Henning Schild wrote:  
> >>>>> Am Tue, 23 Aug 2022 17:47:38 +0300
> >>>>> schrieb Andy Shevchenko <andriy.shevchenko@...ux.intel.com>:    
> >>>>
> >>>> Hi Andy,
> >>>>
> >>>> Thanks for this new version. It is looking good to me.
> >>>>  
> >>>>>     
> >>>>>> On Tue, Aug 23, 2022 at 12:23:40PM +0200, Henning Schild wrote:
> >>>>>>   
> >>>>>>> Add GPIO support for Nuvoton NCT6116 chip. Nuvoton SuperIO
> >>>>>>> chips are very similar to the ones from Fintek. In other
> >>>>>>> subsystems they also share drivers and are called a family of
> >>>>>>> drivers.
> >>>>>>>
> >>>>>>> For the GPIO subsystem the only difference is that the direction
> >>>>>>> bit is reversed and that there is only one data bit per pin. On
> >>>>>>> the SuperIO level the logical device is another one.
> >>>>>>>
> >>>>>>> On a chip level we do not have a manufacturer ID to check and
> >>>>>>> also no revision.      
> >>>>>>
> >>>>>> ...
> >>>>>>     
> >>>>>>> - * GPIO driver for Fintek Super-I/O F71869, F71869A, F71882,
> >>>>>>> F71889 and F81866
> >>>>>>> + * GPIO driver for Fintek and Nuvoton Super-I/O chips      
> >>>>>>
> >>>>>> I'm not sure it's good idea to drop it from here. It means reader
> >>>>>> has to get this info in a hard way.
> >>>>>>
> >>>>>> ...    
> >>>>>
> >>>>> Let us see what others say. I wanted to keep this in line with
> >>>>> what Kconfig says and the oneliner in the Kconfig was getting
> >>>>> pretty longish. Hence i decided to shorten that. Other drivers
> >>>>> also seem to not list all the possible chips in many places, it
> >>>>> is all maint effort when a new chips is added and the list is in
> >>>>> like 5 places.    
> >>>>
> >>>> I agree with you that we can drop this line. It was already
> >>>> incomplete and the information is quite readable a few lines below
> >>>> in both the define list and the chip enumeration.
> >>>>  
> >>>>>     
> >>>>>>> +#define gpio_dir_invert(type)	((type) == nct6116d)
> >>>>>>> +#define gpio_data_single(type)	((type) == nct6116d)
> >>>>>>>  
> >>>>>>
> >>>>>> What's prevents us to add a proper prefix to these? I don't like
> >>>>>> the idea of them having "gpio" prefix.
> >>>>>>
> >>>>>> ...
> >>>>>>     
> >>>>>>> +		pr_info(DRVNAME ": Unsupported device
> >>>>>>> 0x%04x\n", devid);
> >>>>>>> +			pr_debug(DRVNAME ": Not a Fintek
> >>>>>>> device at 0x%08x\n", addr);
> >>>>>>> +	pr_info(DRVNAME ": Found %s at %#x\n",
> >>>>>>> +		pr_info(DRVNAME ":   revision %d\n",      
> >>>>>>
> >>>>>> Can we, please, utilize pr_fmt()?
> >>>>>>     
> >>>>>>> +			(int)superio_inb(addr,
> >>>>>>> SIO_FINTEK_DEVREV));      
> >>>>>>
> >>>>>> Explicit casting in printf() means wrong specifier in 99% of
> >>>>>> cases.   
> >>>>>
> >>>>> For all the other comments i will wait for a second opinion. I
> >>>>> specifically did not change existing code for more than the
> >>>>> functional changes needed. And a bit of checkpatch.pl fixing.
> >>>>> Beautification could be done on the way but would only cause
> >>>>> inconsistency. That driver is what it is, if someone wants to
> >>>>> overhaul the style ... that should be another patch. One likely
> >>>>> not coming from me.    
> >>>>
> >>>> About the int cast, I think you can drop it while you are updating
> >>>> this line. It is unneeded.  
> >>>
> >>> Ok two voices for doing that one fix along the way. I will send a v5
> >>> and hope nobody insists on me fixing the other findings in code i
> >>> never wrote.  
> >>
> >> You did not write it, but you are using it to do hw-enablement for
> >> your company's products. So being asked to also some touch-ups
> >> left and right while you are at it really is not unexpected IMHO.
> > 
> > Sure thing. Dropping a few characters from a line i touch anyhow is
> > easy enough. But i.e a refactoring to pr_fmt would feel like asking too
> > much in my book. That feels like work of the author or maintainer.
> 
> Right, but that assumes that the original author / maintainer is still
> around and actively contributing which unfortunately is not always
> the case.

Actually the original author is not active but he is still keeping an
eye on the driver :)

I still review and test the patches I catch on the MLs. And I am ready
to do some maintenance work if needed.

Henning, I think you could have done the pr_fmt conversion. It is not
a big deal and it would have been nice. But indeed, you don't have to...

Simon

Download attachment "signature.asc" of type "application/pgp-signature" (834 bytes)

Powered by blists - more mailing lists