On Thu, Jun 25, 2020 at 11:33:34AM +0100, Lee Jones wrote:
On Thu, 25 Jun 2020, Daniel Thompson wrote:
On Wed, Jun 24, 2020 at 03:57:16PM +0100, Lee Jones wrote:
Kerneldoc syntax is used, but not complete. Descriptions required.
Prevents warnings like:
drivers/video/backlight/ili922x.c:116: warning: Function parameter or member 's' not described in 'CHECK_FREQ_REG' drivers/video/backlight/ili922x.c:116: warning: Function parameter or member 'x' not described in 'CHECK_FREQ_REG'
Cc: stable@vger.kernel.org Cc: Bartlomiej Zolnierkiewicz b.zolnierkie@samsung.com Cc: Software Engineering sbabic@denx.de Signed-off-by: Lee Jones lee.jones@linaro.org
drivers/video/backlight/ili922x.c | 2 ++ 1 file changed, 2 insertions(+)
diff --git a/drivers/video/backlight/ili922x.c b/drivers/video/backlight/ili922x.c index 9c5aa3fbb2842..8cb4b9d3c3bba 100644 --- a/drivers/video/backlight/ili922x.c +++ b/drivers/video/backlight/ili922x.c @@ -107,6 +107,8 @@
- lower frequency when the registers are read/written.
- The macro sets the frequency in the spi_transfer structure if
- the frequency exceeds the maximum value.
- @s: pointer to controller side proxy for an SPI slave device
What's wrong with "a pointer to an SPI device"?
I am aware, having looked it up to find out what the above actually means, that this is how struct spi_device is described in its own kernel doc but quoting at that level of detail of both overkill and confusing.
I figured that using the official description would be better than making something up. However if you think it's better to KISS, then I can change it.
Yes, I'd strongly prefer KISS here.
I know it is an "I am the world" argument[1] but I found using such a dogmatically accurate description out of context to be very confusing and therefore I don't think such a comment improves readability.
Daniel.
[1]: See #3 from http://www.leany.com/logic/Adams.html