Re: [PATCH v2] mm/slub: Remove repeated action in calculate_order()

From: Hyeonggon Yoo <42.hyeyoo@gmail.com>
To: Wonhyuk Yang <vvghjk1234@gmail.com>
Cc: Christoph Lameter <cl@linux.com>,
	Pekka Enberg <penberg@kernel.org>,
	David Rientjes <rientjes@google.com>,
	Joonsoo Kim <iamjoonsoo.kim@lge.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Vlastimil Babka <vbabka@suse.cz>,
	Roman Gushchin <roman.gushchin@linux.dev>,
	Ohhoon Kwon <ohkwon1043@gmail.com>,
	JaeSang Yoo <jsyoo5b@gmail.com>,
	Jiyoup Kim <lakroforce@gmail.com>,
	Donghyeok Kim <dthex5d@gmail.com>,
	linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] mm/slub: Remove repeated action in calculate_order()
Date: Sun, 17 Apr 2022 10:43:20 +0900	[thread overview]
Message-ID: <YltwuNz4N5BKRFDT@hyeyoo> (raw)
In-Reply-To: <20220416074059.526970-1-vvghjk1234@gmail.com>

On Sat, Apr 16, 2022 at 04:40:59PM +0900, Wonhyuk Yang wrote:
> To calculate order, calc_slab_order() is called repeatly changing the
> fract_leftover. Thus, the branch which is not dependent on
> fract_leftover is executed repeatly. So make it run only once.
> 
> Plus, when min_object reached to 0, we set fract_leftover to 1. In

Maybe you mean when min_object reached 1.

> this case, we can calculate order by max(slub_min_order,
> get_order(size)) instead of calling calc_slab_order().
> 
> No functional impact expected.
> Signed-off-by: Wonhyuk Yang <vvghjk1234@gmail.com>
> ---
> V1 -> V2: Fix typo miss in a commit message
> 
>  mm/slub.c | 18 +++++++-----------
>  1 file changed, 7 insertions(+), 11 deletions(-)
> 
> diff --git a/mm/slub.c b/mm/slub.c
> index ed5c2c03a47a..e7a394d7b75a 100644
> --- a/mm/slub.c
> +++ b/mm/slub.c
> @@ -3795,9 +3795,6 @@ static inline unsigned int calc_slab_order(unsigned int size,
>  	unsigned int min_order = slub_min_order;
>  	unsigned int order;
>  
> -	if (order_objects(min_order, size) > MAX_OBJS_PER_PAGE)
> -		return get_order(size * MAX_OBJS_PER_PAGE) - 1;
> -
>  	for (order = max(min_order, (unsigned int)get_order(min_objects * size));
>  			order <= max_order; order++) {
>  
> @@ -3820,6 +3817,11 @@ static inline int calculate_order(unsigned int size)
>  	unsigned int max_objects;
>  	unsigned int nr_cpus;
>  
> +	if (unlikely(order_objects(slub_min_order, size) > MAX_OBJS_PER_PAGE)) {
> +		order = get_order(size * MAX_OBJS_PER_PAGE) - 1;
> +		goto out;
> +	}
> +
>  	/*
>  	 * Attempt to find best configuration for a slab. This
>  	 * works by first attempting to generate a layout with
> @@ -3865,14 +3867,8 @@ static inline int calculate_order(unsigned int size)
>  	 * We were unable to place multiple objects in a slab. Now
>  	 * lets see if we can place a single object there.
>  	 */
> -	order = calc_slab_order(size, 1, slub_max_order, 1);
> -	if (order <= slub_max_order)
> -		return order;
> -
> -	/*
> -	 * Doh this slab cannot be placed using slub_max_order.
> -	 */
> -	order = calc_slab_order(size, 1, MAX_ORDER, 1);
> +	order = max_t(unsigned int, slub_min_order, (unsigned int)get_order(size));
> +out:

You don't need to cast value of get_order(size). max_t() does cast both operands.

>  	if (order < MAX_ORDER)
>  		return order;
>  	return -ENOSYS;

For the correctness of the patch, I don't see any problem about the
code.

But to be honest I'm a bit skeptical about saving some cycles in
calculating slab order. It's done only when creating caches (usually in boot
process).

> -- 
> 2.30.2
> 
> 

-- 
Thanks,
Hyeonggon