Re: [PATCH 6/6] mm, migration: do not trigger OOM killer when migrating memory

[Andrew Morton]

On Fri, 23 Jun 2017 10:53:45 +0200 Michal Hocko <mhocko@xxxxxxxxxx> wrote:

> From: Michal Hocko <mhocko@xxxxxxxx>
> 
> Page migration (for memory hotplug, soft_offline_page or mbind) needs
> to allocate a new memory. This can trigger an oom killer if the target
> memory is depleated. Although quite unlikely, still possible, especially
> for the memory hotplug (offlining of memoery). Up to now we didn't
> really have reasonable means to back off. __GFP_NORETRY can fail just
> too easily and __GFP_THISNODE sticks to a single node and that is not
> suitable for all callers.
> 
> But now that we have __GFP_RETRY_MAYFAIL we should use it.  It is
> preferable to fail the migration than disrupt the system by killing some
> processes.

I'm not sure which tree this is against...

> --- a/mm/memory-failure.c
> +++ b/mm/memory-failure.c
> @@ -1492,7 +1492,8 @@ static struct page *new_page(struct page *p, unsigned long private, int **x)
>  
>  		return alloc_huge_page_node(hstate, nid);
>  	} else {
> -		return __alloc_pages_node(nid, GFP_HIGHUSER_MOVABLE, 0);
> +		return __alloc_pages_node(nid,
> +				GFP_HIGHUSER_MOVABLE | __GFP_RETRY_MAYFAIL, 0);
>  	}
>  }

new_page() is now

static struct page *new_page(struct page *p, unsigned long private, int **x)
{
	int nid = page_to_nid(p);

	return new_page_nodemask(p, nid, &node_states[N_MEMORY]);
}

and new_page_nodemask() uses __GFP_RETRY_MAYFAIL so I simply dropped
the above hunk.