2022-11-29 15:45:10

by Mel Gorman

Subject: [PATCH 2/6] mm/page_alloc: Treat RT tasks similar to GFP_HIGH

RT tasks are allowed to dip below the min reserve but ALLOC_HARDER is
typically combined with ALLOC_MIN_RESERVE so RT tasks are a little
unusual. While there is some justification for allowing RT tasks
access to memory reserves, there is a strong chance that a RT task
that is also under memory pressure is at risk of missing deadlines
anyway. Relax how much reserves an RT task can access by treating
it the same as __GFP_HIGH allocations.

Signed-off-by: Mel Gorman <[email protected]>
---
mm/page_alloc.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 3b37909617bc..da746e9eb2cf 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -4852,7 +4852,7 @@ gfp_to_alloc_flags(gfp_t gfp_mask)
*/
alloc_flags &= ~ALLOC_CPUSET;
} else if (unlikely(rt_task(current)) && in_task())
- alloc_flags |= ALLOC_HARDER;
+ alloc_flags |= ALLOC_MIN_RESERVE;

alloc_flags = gfp_to_alloc_flags_cma(gfp_mask, alloc_flags);

--
2.35.3

2022-12-08 16:33:17

by Vlastimil Babka

Subject: Re: [PATCH 2/6] mm/page_alloc: Treat RT tasks similar to GFP_HIGH

Subject should say __GFP_HIGH as there's no GFP_HIGH?

On 11/29/22 16:16, Mel Gorman wrote:
> RT tasks are allowed to dip below the min reserve but ALLOC_HARDER is
> typically combined with ALLOC_MIN_RESERVE so RT tasks are a little
> unusual. While there is some justification for allowing RT tasks
> access to memory reserves, there is a strong chance that a RT task
> that is also under memory pressure is at risk of missing deadlines
> anyway. Relax how much reserves an RT task can access by treating
> it the same as __GFP_HIGH allocations.
>
> Signed-off-by: Mel Gorman <[email protected]>

Acked-by: Vlastimil Babka <[email protected]>

> ---
> mm/page_alloc.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index 3b37909617bc..da746e9eb2cf 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -4852,7 +4852,7 @@ gfp_to_alloc_flags(gfp_t gfp_mask)
> */
> alloc_flags &= ~ALLOC_CPUSET;
> } else if (unlikely(rt_task(current)) && in_task())
> - alloc_flags |= ALLOC_HARDER;
> + alloc_flags |= ALLOC_MIN_RESERVE;
>
> alloc_flags = gfp_to_alloc_flags_cma(gfp_mask, alloc_flags);
>