Randy Dunlap wrote:
> On Mon, 9 Oct 2006 07:28:51 +0400 Alexey Dobriyan wrote:
> > On Sun, Oct 08, 2006 at 07:44:29PM -0700, Randy Dunlap wrote:
> > > /**
> > > - * strlen_user: - Get the size of a string in user space.
> > > + * strnlen_user: - Get the size of a string in user space.
> >
> > It's better to not spend time fixing mismatches, but to teach kernel-doc
> > extract function name from function itself.
> >
> > /**
> > * Get the size of a string in user space.
> > * @foo: bar
> > */
> > size_t strnlen_user()
>
> OK, maybe a good idea. I'll add that to the wish list.
> However, we have seen examples of:
>
> /**
> * doc for foo
> */
> int foo(int arg)
> {
> }
>
> then someone inserts a new function bar() between the kernel-doc
> and foo(). How would we catch that (automated)?
> other than by our review process?
Warn if the function signature of the documented function does not match the
function and throw away the manpage then.
Eike
Attachment:
pgpkjByDQN6JR.pgp
Description: PGP signature
- References:
- [PATCH] kernel-doc: fix function name in usercopy.c
- From: Randy Dunlap <[email protected]>
- Re: [PATCH] kernel-doc: fix function name in usercopy.c
- From: Alexey Dobriyan <[email protected]>
- Re: [PATCH] kernel-doc: fix function name in usercopy.c
- From: Randy Dunlap <[email protected]>
- [PATCH] kernel-doc: fix function name in usercopy.c
- Prev by Date: Re: [PATCH] kernel-doc: drop __must_check and various "inline" qualifiers
- Next by Date: Re: [2.6.19 patch] ATA must depend on BLOCK
- Previous by thread: Re: [PATCH] kernel-doc: fix function name in usercopy.c
- Next by thread: Re: [PATCH] libata: skip reset on bus not a device
- Index(es):
 |