Re: [PATCH] at86rf230: convert to gpio descriptors

["Arnd Bergmann" <arnd@arndb.de>]

On Thu, Jan 26, 2023, at 15:12, Miquel Raynal wrote:

>> @@ -1682,7 +1650,7 @@ MODULE_DEVICE_TABLE(spi, at86rf230_device_id);
>>  static struct spi_driver at86rf230_driver = {
>>  	.id_table = at86rf230_device_id,
>>  	.driver = {
>> -		.of_match_table = of_match_ptr(at86rf230_of_match),
>> +		.of_match_table = at86rf230_of_match,
>
> Looks like an unrelated change? Or is it a consequence of "not having
> any in-tree users of platform_data" that plays a role here?

I probably did it because I thought I had removed the matching #ifdef
for at86rf230_of_match in the process of making the driver DT-only.

Without this trivial change, building the driver as built-in with
CONFIG_OF=n can result in a warning like

drivers/net/ieee802154/at86rf230.c:1632:28: error: unused variable 'at86rf230_of_match' [-Werror,-Wunused-variable]

It looks like this was already removed in a8b66db804f0 ("at86rf230:
remove #ifdef CONFIG_OF"), which was not technically correct, but
nobody noticed, including me.

I could split this out as a separate patch, but it's probably
not worth it.

> Anyhow, the changes in the driver look good, so:
>
> Reviewed-by: Miquel Raynal <miquel.raynal@bootlin.com>

Thanks,

    Arnd