Re: [PATCH 1/3] mm, page_alloc: Split buffered_rmqueue

[Vlastimil Babka]

On 01/12/2017 11:42 AM, Mel Gorman wrote:
> buffered_rmqueue removes a page from a given zone and uses the per-cpu
> list for order-0. This is fine but a hypothetical caller that wanted
> multiple order-0 pages has to disable/reenable interrupts multiple
> times. This patch structures buffere_rmqueue such that it's relatively
> easy to build a bulk order-0 page allocator. There is no functional
> change.

Strictly speaking, this will now skip VM_BUG_ON_PAGE(bad_range(...)) for order-0 
allocations. Do we care?

> Signed-off-by: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx>
> Acked-by: Hillf Danton <hillf.zj@xxxxxxxxxxxxxxx>
> ---
>  mm/page_alloc.c | 126 ++++++++++++++++++++++++++++++++++----------------------
>  1 file changed, 77 insertions(+), 49 deletions(-)
>
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index 2c6d5f64feca..d8798583eaf8 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -2610,68 +2610,96 @@ static inline void zone_statistics(struct zone *preferred_zone, struct zone *z,
>  #endif
>  }
>
> +/* Remote page from the per-cpu list, caller must protect the list */

    ^ Remove

> +static struct page *__rmqueue_pcplist(struct zone *zone, unsigned int order,
> +			gfp_t gfp_flags, int migratetype, bool cold,

order and gfp_flags seem unused here

> +			struct per_cpu_pages *pcp, struct list_head *list)
> +{
> +	struct page *page;