On Wed, 06 Dec 2023, Daniel Thompson wrote: > On Tue, Dec 05, 2023 at 02:56:38PM -0800, Randy Dunlap wrote: > > Fix kernel-doc warnings found when using "W=1". > > > > ili922x.c:85: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst > > ili922x.c:85: warning: missing initial short description on line: > > * START_BYTE(id, rs, rw) > > ili922x.c:91: warning: contents before sections > > ili922x.c:118: warning: expecting prototype for CHECK_FREQ_REG(spi_device s, spi_transfer x)(). Prototype was for CHECK_FREQ_REG() instead > > > > Signed-off-by: Randy Dunlap <rdunlap@xxxxxxxxxxxxx> > > Cc: Lee Jones <lee@xxxxxxxxxx> > > Cc: Daniel Thompson <daniel.thompson@xxxxxxxxxx> > > Cc: Jingoo Han <jingoohan1@xxxxxxxxx> > > Cc: Helge Deller <deller@xxxxxx> > > Cc: linux-fbdev@xxxxxxxxxxxxxxx > > --- > > drivers/video/backlight/ili922x.c | 9 ++++----- > > 1 file changed, 4 insertions(+), 5 deletions(-) > > > > diff -- a/drivers/video/backlight/ili922x.c b/drivers/video/backlight/ili922x.c > > --- a/drivers/video/backlight/ili922x.c > > +++ b/drivers/video/backlight/ili922x.c > > @@ -82,13 +82,12 @@ > > #define START_RW_READ 1 > > > > /** > > - * START_BYTE(id, rs, rw) > > - * > > - * Set the start byte according to the required operation. > > + * START_BYTE() - Set the start byte according to the required operation. > > * The start byte is defined as: > > * ---------------------------------- > > * | 0 | 1 | 1 | 1 | 0 | ID | RS | RW | > > * ---------------------------------- > > I'm not sure we want "The start byte is defined as" in the brief > description. Needs a blank line between the brief and full description > (or hoist the argument descriptions up to match the idiomatic > form for a kernel-doc comment in the docs if you prefer). I'd consider dropping the kernel-docness of this header entirely. Kerneldoc is designed for documenting exported (or at least externally available) functions and data structures, with allowances for static functions in the name of consistency or in cases of excessive complication. I've fixed A LOT of kernel-doc headers in my time and I can't say I remember coming across MACROs being documented this way before now. -- Lee Jones [李琼斯]
