On Tue, 20 Sep 2016 14:54:44 +0800 zhongjiang <zhongjiang@xxxxxxxxxx> wrote:
> From: zhong jiang <zhongjiang@xxxxxxxxxx>
>
> Accoding to HUgh's suggestion, alloc_stable_node() with GFP_KERNEL
> will cause the hungtask, despite less possiblity.
>
> At present, if alloc_stable_node allocate fails, two break_cow may
> want to allocate a couple of pages, and the issue will come up when
> free memory is under pressure.
>
> we fix it by adding the __GFP_HIGH to GFP. because it grant access to
> some of meory reserves. it will make progess to make it allocation
> successful at the utmost.
>
> --- a/mm/ksm.c
> +++ b/mm/ksm.c
> @@ -299,7 +299,7 @@ static inline void free_rmap_item(struct rmap_item *rmap_item)
>
> static inline struct stable_node *alloc_stable_node(void)
> {
> - return kmem_cache_alloc(stable_node_cache, GFP_KERNEL);
> + return kmem_cache_alloc(stable_node_cache, GFP_KERNEL | __GFP_HIGH);
> }
>
> static inline void free_stable_node(struct stable_node *stable_node)
It is very hard for a reader to understand why this __GFP_HIGH is being
used here, so we should have a code comment explaining the reasoning,
please.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>