Re: [PATCH 4/8] gpio: tqmx86: introduce _tqmx86_gpio_update_bits() helper

[Andrew Lunn]

On Wed, May 29, 2024 at 09:45:16AM +0200, Matthias Schiffer wrote:
> Simplify a lot of code in the driver by introducing helpers for the
> common RMW pattern. No tqmx86_gpio_update_bits() function with builtin
> locking is added, as it would become redundant with the following fixes,
> which further consolidate interrupt configuration register setup.
> 
> No functional change intended.
> 
> Fixes: b868db94a6a7 ("gpio: tqmx86: Add GPIO from for this IO controller")
> Signed-off-by: Matthias Schiffer <matthias.schiffer@xxxxxxxxxxxxxxx>
> ---
>  drivers/gpio/gpio-tqmx86.c | 40 ++++++++++++++++++++++----------------
>  1 file changed, 23 insertions(+), 17 deletions(-)
> 
> diff --git a/drivers/gpio/gpio-tqmx86.c b/drivers/gpio/gpio-tqmx86.c
> index 613ab9ef2e744..7a851e1730dd1 100644
> --- a/drivers/gpio/gpio-tqmx86.c
> +++ b/drivers/gpio/gpio-tqmx86.c
> @@ -54,6 +54,17 @@ static void tqmx86_gpio_write(struct tqmx86_gpio_data *gd, unsigned int reg,
>  	iowrite8(val, gd->io_base + reg);
>  }
>  
> +static void _tqmx86_gpio_update_bits(struct tqmx86_gpio_data *gd,
> +				     unsigned int reg, u8 mask, u8 val)

Why the _ prefix? This is a local function, it is static, so you don't
have name space issues. Functions starting with _ are those you should
not call without a good reason, there is generally a version without
the _ prefix which is the real function to use. So i would drop the _.

This is also not a fix. Please read:

https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html

and stick to the rules described there.

I don't know how the GPIO tree works, but for netdev, about a week
after fixes are merged, they appear in net-next. So you can then build
on top of them for development work.

	Andrew