Re: [PATCH RFC 03/13] mm/gup: Check p4d presence before going on

[Jason Gunthorpe]

On Wed, Mar 06, 2024 at 06:41:37PM +0800, peterx@xxxxxxxxxx wrote:
> From: Peter Xu <peterx@xxxxxxxxxx>
> 
> Currently there should have no p4d swap entries so it may not matter much,
> however this may help us to rule out swap entries in pXd_huge() API, which
> will include p4d_huge().  The p4d_present() checks make it 100% clear that
> we won't rely on p4d_huge() for swap entries.
> 
> Signed-off-by: Peter Xu <peterx@xxxxxxxxxx>
> ---
>  mm/gup.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)

>  	p4dp = p4d_offset(pgdp, address);
>  	p4d = READ_ONCE(*p4dp);
> -	if (p4d_none(p4d))
> +	if (p4d_none(p4d) || !p4d_present(p4d))
>  		return no_page_table(vma, flags);

Like the other place I think this makes more sense as

    if (!p4_present(p4d))
  		return no_page_table(vma, flags);

Since none can never be present.

IOW if the following code doesn't decode swap entries then it should
be guarded with a chceck on present not none..

Jason