Re: [PATCH v2 07/28] gup: Remove an assumption of a contiguous memmap

[John Hubbard <jhubbard@nvidia.com>]

On 1/9/22 20:23, Matthew Wilcox (Oracle) wrote:
> This assumption needs the inverse of nth_page(), which I've temporarily
> named page_nth() until someone comes up with a better name.
> 
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
> ---
>   include/linux/mm.h | 2 ++
>   mm/gup.c           | 4 ++--
>   2 files changed, 4 insertions(+), 2 deletions(-)

Reviewed-by: John Hubbard <jhubbard@nvidia.com>

Maybe a better name for page_nth() will come to mind by the time I
reach the end of this series, we'll see. :)


thanks,
-- 
John Hubbard
NVIDIA

> 
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index d8b7d7ed14dd..f2f3400665a4 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -216,8 +216,10 @@ int overcommit_policy_handler(struct ctl_table *, int, void *, size_t *,
>   
>   #if defined(CONFIG_SPARSEMEM) && !defined(CONFIG_SPARSEMEM_VMEMMAP)
>   #define nth_page(page,n) pfn_to_page(page_to_pfn((page)) + (n))
> +#define page_nth(head, tail)	(page_to_pfn(tail) - page_to_pfn(head))
>   #else
>   #define nth_page(page,n) ((page) + (n))
> +#define page_nth(head, tail)	((tail) - (head))
>   #endif
>   
>   /* to align the pointer to the (next) page boundary */
> diff --git a/mm/gup.c b/mm/gup.c
> index 9c0a702a4e03..afb638a30e44 100644
> --- a/mm/gup.c
> +++ b/mm/gup.c
> @@ -238,8 +238,8 @@ static inline struct page *compound_range_next(unsigned long i,
>   	next = nth_page(start, i);
>   	page = compound_head(next);
>   	if (PageHead(page))
> -		nr = min_t(unsigned int,
> -			   page + compound_nr(page) - next, npages - i);
> +		nr = min_t(unsigned int, npages - i,
> +			   compound_nr(page) - page_nth(page, next));
>   
>   	*ntails = nr;
>   	return page;