-Darrick as email seems to bounce. On 19:07-20130306, Rob Landley wrote: > On 03/06/2013 11:57:07 AM, Nishanth Menon wrote: > >Synopsys section might be interesting here: > >Before: http://pastebin.com/bVt3RWKZ > >u32 __pure crc32_le_generic (u32 crc, > > unsigned char const * p, > > size_t len, > > const u32 ( * tab)[256], > > u32 polynomial); > >After: http://pastebin.com/z0NNWHtw > >u32 __pure crc32_le_generic (u32 crc, > > unsigned char const * p, > > size_t len, > > const u32 * tab, > > u32 polynomial); > >Not sure if we like that. > > *blink* *blink* Yep. It is wrong. > > commit 836e2af92503f1642dbc3c3281ec68ec1dd39d2e > Author: Joakim Tjernlund <Joakim.Tjernlund@xxxxxxxxxxxx> > Date: Mon May 24 14:33:31 2010 -0700 > > --- a/lib/crc32.c > +++ b/lib/crc32.c > @@ -48,12 +48,20 @@ MODULE_LICENSE("GPL"); > #if CRC_LE_BITS == 8 || CRC_BE_BITS == 8 > > static inline u32 > -crc32_body(u32 crc, unsigned char const *buf, size_t len, const u32 > *tab) > +crc32_body(u32 crc, unsigned char const *buf, size_t len, const u32 > (*tab)[256] > { > > That's a crazy type to have a in a prototype. (It looks like > function pointers but isn't, I _think_ it's an array of pointers?) > > Reading the code, it's a pointer to a series of arrays. It's basically: > > const u32 tab[][256]; > > Defined in a really weird way. > > So the first form and the second form are not equivalent. The first > form is highly unusual but the compiler accepts it, so... > > Rob yes, I agree - I agree the definition is weird, but kernel-doc script is kind of expected to handle these as well. (*tab)[256] or tab[][256] should be described by * @tab: something. And kernel-doc should be able to convert that into: Synopsis: original prototype u32 __pure crc32_le_generic (u32 crc, unsigned char const * p, size_t len, const u32 ( * tab)[256], u32 polynomial); Arguments: tab something The problem we probably have is that: $args{'parameterlist'} is used to create the synopsis for the function and to print out argument list. If we cleanup the parameter for "Arguments" then Synopsis suffers and viceversa. I got interested in this specific definition as it kind of caught an corner case in kernel-doc script. An alternate solution *might* be to have $args{'parameterlistorig'} which could be used to store the original prototype(for Synopsis) while $args{'parameterlist'} is used elsewhere - but it kinda feels hackish. -- Regards, Nishanth Menon -- To unsubscribe from this list: send the line "unsubscribe linux-doc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
