Re: [PATCH] dma: get_user_pages -> get_user_pages_fast

[Peter Zijlstra <peterz@infradead.org>]

On Mon, 2010-10-11 at 18:53 +0200, Michael S. Tsirkin wrote:
> There doesn't seem to be any advantage to using
> get_user_pages, so switch iovlock to get_user_pages_fast
> instead.
> 
> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
> ---
> 
> Lightly tested.  This patch is on top of the bugfix patch I posted
> previously.
> 
>  drivers/dma/iovlock.c |   10 ++--------
>  1 files changed, 2 insertions(+), 8 deletions(-)
> 
> diff --git a/drivers/dma/iovlock.c b/drivers/dma/iovlock.c
> index 21ed8f3..c6917e8 100644
> --- a/drivers/dma/iovlock.c
> +++ b/drivers/dma/iovlock.c
> @@ -95,17 +95,11 @@ struct dma_pinned_list *dma_pin_iovec_pages(struct iovec *iov, size_t len)
>  		pages += page_list->nr_pages;
>  
>  		/* pin pages down */
> -		down_read(&current->mm->mmap_sem);
> -		ret = get_user_pages(
> -			current,
> -			current->mm,
> +		ret = get_user_pages_fast(
>  			(unsigned long) iov[i].iov_base,
>  			page_list->nr_pages,

What's the actual nr_pages here? Is it limited to some small number or
can it be arbitrarily large?

>  			1,	/* write */
> -			0,	/* force */
> -			page_list->pages,
> -			NULL);
> -		up_read(&current->mm->mmap_sem);
> +			page_list->pages);
>  
>  		if (unlikely(ret < 0))
>  			goto unpin;