Re: [PATCH] iov_iter: optimize page_copy_sane()

[Eric Dumazet]

On 12/08/2018 05:37 AM, Eric Dumazet wrote:
> Avoid cache line miss dereferencing struct page if we can.
> 
> page_copy_sane() mostly deals with order-0 pages.
> 
> Signed-off-by: Eric Dumazet <edumazet@xxxxxxxxxx>
> Cc: Al Viro <viro@xxxxxxxxxxxxxxxxxx>
> ---
>  lib/iov_iter.c | 14 ++++++++++++--
>  1 file changed, 12 insertions(+), 2 deletions(-)
> 
> diff --git a/lib/iov_iter.c b/lib/iov_iter.c
> index 54c248526b55fc498c996f2a5ea651262fcc7f61..2a1e282c3d38e2d03b075f54b7f9313e48b90289 100644
> --- a/lib/iov_iter.c
> +++ b/lib/iov_iter.c
> @@ -854,8 +854,18 @@ EXPORT_SYMBOL(_copy_from_iter_full_nocache);
>  
>  static inline bool page_copy_sane(struct page *page, size_t offset, size_t n)
>  {
> -	struct page *head = compound_head(page);
> -	size_t v = n + offset + page_address(page) - page_address(head);
> +	struct page *head;
> +	size_t v = n + offset;
> +
> +	/*
> +	 * Fast path if this is an order-0 page,
> +	 * or if the copied area fits the first page of a compound one.
> +	 */
> +	if (likely(n <= v && v <= PAGE_SIZE))
> +		return true;
> +
> +	head = compound_head(page);
> +	v += (page - head) << PAGE_SHIFT;
>  
>  	if (likely(n <= v && v <= (PAGE_SIZE << compound_order(head))))
>  		return true;
> 

Hi Alexander !

Any news on this patch ?

Thanks !