Re: [PATCH] mm/mremap_pages: Save a few cycles in 'get_dev_pagemap()'

[Matthew Wilcox]

On Wed, Nov 03, 2021 at 10:35:34PM +0100, Christophe JAILLET wrote:
> Use 'percpu_ref_tryget_live_rcu()' instead of 'percpu_ref_tryget_live()' to
> save a few cycles when it is known that the rcu lock is already
> taken/released.

If this is really important, we can add an __xa_load() which doesn't
take the RCU read lock.

I honestly think that the xarray is the wrong data structure here,
and we'd be better off with a simple array of (start, pointer)
tuples.

> Signed-off-by: Christophe JAILLET <christophe.jaillet@xxxxxxxxxx>
> ---
>  mm/memremap.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/mm/memremap.c b/mm/memremap.c
> index 84de22c14567..012e8d23d365 100644
> --- a/mm/memremap.c
> +++ b/mm/memremap.c
> @@ -506,7 +506,7 @@ struct dev_pagemap *get_dev_pagemap(unsigned long pfn,
>  	/* fall back to slow path lookup */
>  	rcu_read_lock();
>  	pgmap = xa_load(&pgmap_array, PHYS_PFN(phys));
> -	if (pgmap && !percpu_ref_tryget_live(pgmap->ref))
> +	if (pgmap && !percpu_ref_tryget_live_rcu(pgmap->ref))
>  		pgmap = NULL;
>  	rcu_read_unlock();
>  
> -- 
> 2.30.2
> 
>