Re: [PATCH] mm, hugetlb: allow hugepage allocations to excessively reclaim

[Vlastimil Babka]

On 10/7/19 9:55 AM, Michal Hocko wrote:
> From: David Rientjes <rientjes@xxxxxxxxxx>
Nit: the subject is still somewhat misleading IMHO, especially in light
of Mike's responses. I would say "reclaim as needed" instead of
"excessively reclaim". The excessive reclaim behavior in hugetlb nr_pages
setting was a bug that was addressed by a different series.
 
...

> [mhocko@xxxxxxxx: reworded changelog]
> Fixes: b39d0ee2632d ("mm, page_alloc: avoid expensive reclaim when compaction may not succeed")
> Cc: Mike Kravetz <mike.kravetz@xxxxxxxxxx>
> Signed-off-by: David Rientjes <rientjes@xxxxxxxxxx>
> Signed-off-by: Michal Hocko <mhocko@xxxxxxxx>

I still believe that using __GFP_NORETRY as needed is a cleaner solution
than a check for pageblock order and __GFP_IO, but that can be always
changed later. This patch does fix the hugetlbfs regression, so

Acked-by: Vlastimil Babka <vbabka@xxxxxxx>