Re: [PATCH v4 5/5] spi: s3c64xx: allow controller-data to be optional

[Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxxxxx>]

On 19/01/2022 21:55, Andi Shyti wrote:
> Hi Krzysztof,
> 
> On Wed, Jan 19, 2022 at 09:10:05PM +0100, Krzysztof Kozlowski wrote:
>> The Samsung SoC SPI driver requires to provide controller-data node
>> for each of SPI peripheral device nodes.  Make this controller-data node
>> optional, so DTS could be simpler.
>>
>> Suggested-by: Rob Herring <robh@xxxxxxxxxx>
>> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxxxxx>
>> Reviewed-by: Sam Protsenko <semen.protsenko@xxxxxxxxxx>
>> ---
>>  drivers/spi/spi-s3c64xx.c | 14 ++++++--------
>>  1 file changed, 6 insertions(+), 8 deletions(-)
>>
>> diff --git a/drivers/spi/spi-s3c64xx.c b/drivers/spi/spi-s3c64xx.c
>> index 8755cd85e83c..769d958a2f86 100644
>> --- a/drivers/spi/spi-s3c64xx.c
>> +++ b/drivers/spi/spi-s3c64xx.c
>> @@ -796,16 +796,14 @@ static struct s3c64xx_spi_csinfo *s3c64xx_get_slave_ctrldata(
>>  		return ERR_PTR(-EINVAL);
>>  	}
>>  
>> -	data_np = of_get_child_by_name(slave_np, "controller-data");
>> -	if (!data_np) {
>> -		dev_err(&spi->dev, "child node 'controller-data' not found\n");
>> -		return ERR_PTR(-EINVAL);
>> -	}
>> -
>>  	cs = kzalloc(sizeof(*cs), GFP_KERNEL);
>> -	if (!cs) {
>> -		of_node_put(data_np);
>> +	if (!cs)
>>  		return ERR_PTR(-ENOMEM);
>> +
>> +	data_np = of_get_child_by_name(slave_np, "controller-data");
>> +	if (!data_np) {
>> +		dev_info(&spi->dev, "child node 'controller-data' not found, using defaults\n");
> 
> "not found" sounds like an error; I would just write something
> like "feedback delay set to '0' dfault", you also tell that the
> default value is '0'.

Sure, I will rewrite the message.


Best regards,
Krzysztof