Re: [RFCv2 6/6] mm: extend process_madvise syscall to support vector arrary

[Minchan Kim <minchan@kernel.org>]

Hi Yann,

On Fri, May 31, 2019 at 12:06:52PM +0200, Yann Droneaud wrote:
> Hi,
> 
> Le vendredi 31 mai 2019 à 15:43 +0900, Minchan Kim a écrit :
> > 
> > diff --git a/include/uapi/asm-generic/mman-common.h
> > b/include/uapi/asm-generic/mman-common.h
> > index 92e347a89ddc..220c2b5eb961 100644
> > --- a/include/uapi/asm-generic/mman-common.h
> > +++ b/include/uapi/asm-generic/mman-common.h
> > @@ -75,4 +75,15 @@
> >  #define PKEY_ACCESS_MASK	(PKEY_DISABLE_ACCESS |\
> >  				 PKEY_DISABLE_WRITE)
> >  
> > +struct pr_madvise_param {
> > +	int size;		/* the size of this structure */
> > +	int cookie;		/* reserved to support atomicity */
> > +	int nr_elem;		/* count of below arrary fields */
> 
> Those should be unsigned.
> 
> There's an implicit hole here on ABI with 64bits aligned pointers
> 
> > +	int __user *hints;	/* hints for each range */
> > +	/* to store result of each operation */
> > +	const struct iovec __user *results;
> > +	/* input address ranges */
> > +	const struct iovec __user *ranges;
> 
> Using pointer type in uAPI structure require a 'compat' version of the
> syscall need to be provided.
> 
> If using iovec too.

I will fix them when I submit next revision.
Thanks for the review.