Re: [PATCH v4 1/2] compiler_types: Introduce __flex_counter() and family

[Kees Cook]

On Fri, Mar 14, 2025 at 09:53:41PM -0700, Randy Dunlap wrote:
> Hi Kees,
> 
> 
> On 3/14/25 8:15 PM, Kees Cook wrote:
> 
> 
> > diff --git a/include/linux/overflow.h b/include/linux/overflow.h
> > index 0c7e3dcfe867..e2b81cb5576e 100644
> > --- a/include/linux/overflow.h
> > +++ b/include/linux/overflow.h
> > @@ -440,4 +440,40 @@ static inline size_t __must_check size_sub(size_t minuend, size_t subtrahend)
> >  #define DEFINE_FLEX(TYPE, NAME, MEMBER, COUNTER, COUNT)	\
> >  	_DEFINE_FLEX(TYPE, NAME, MEMBER, COUNT, = { .obj.COUNTER = COUNT, })
> >  
> > +/**
> > + * typeof_flex_counter() - Return the type of the counter variable of a given
> > + *                         flexible array member annotated by __counted_by().
> > + * @FAM: Pointer to the flexible array member within a given struct.
> > + *
> > + * Returns "size_t" if no annotation exists.
> 
> Please use
>  * Returns: <text>
> instead so that kernel-doc can make a special doc section for it.

Ah! Thanks -- I hadn't realized that the ":" induced special sections. I
think I have a bunch of other kern-doc clean-ups to do as well.

> 
> Same for patch 2/2.
> 
> > + */
> > +#define typeof_flex_counter(FAM)				\
> > +	typeof(_Generic(__flex_counter(FAM),			\
> > +			void *: (size_t)0,			\
> > +			default: *__flex_counter(FAM)))
> > +
> > +/** can_set_flex_counter() - Check if the counter associated with the given
> 
> Needs a newline between /** and the function name, as in set_flex_counter() below.

Whoops, thanks!

-Kees

-- 
Kees Cook