src/sys/vfs ANSIfication

Chris Pressey cpressey at catseye.mine.nu
Tue Mar 30 09:36:44 PST 2004


On Wed, 31 Mar 2004 14:21:33 +0100
Hiten Pandya <hmp at xxxxxxxxxxxxx> wrote:

> David Rhodus wrote:
> > Hmm.
> > I am personally very used to having these structure notes
> > located inside most of the filesystem code like this. Most of
> > other changes that have went on I don't really notice when
> > working with the code, ex. having to read something again.
> > I really don't think the same would be true with this change.
> > 
> > -DR
> 
> 	Seconded, it does not hurt to have these descriptions,
> 	but anyway, I dont have a problem either way.
> 
> 		-- Hiten

If these were normal circumstances, I wouldn't see a need for them, but
these are obviously not normal circumstances:

  http://fxr.watson.org/fxr/ident?v=DFBSD&i=vop_ioctl_args

:)

So it's just a matter of how that information is presented.  If you guys
are OK with Matt's suggestion for a sort of "pseudo-prototype" just
above the function definition, I think I'm leaning towards that.

FWIW, coda does it like this:

	int
	coda_ioctl(void *v)
	{
	/* true args */
	    struct vop_ioctl_args *ap = v;
	    struct vnode *vp = ap->a_vp;
	    int com = ap->a_command;
	    caddr_t data = ap->a_data;
	    int flag = ap->a_fflag;
	    struct ucred *cred = ap->a_cred;
	    struct proc  *p = ap->a_p;
	/* locals */

-Chris





More information about the Submit mailing list