From: Miaohe Lin <linmiaohe@xxxxxxxxxx> Subject: mm/hugetlb: optimize the surplus state transfer code in move_hugetlb_state() We should not transfer the per-node surplus state when we do not cross the node in order to save some cpu cycles Link: https://lkml.kernel.org/r/20210308112809.26107-3-linmiaohe@xxxxxxxxxx Signed-off-by: Miaohe Lin <linmiaohe@xxxxxxxxxx> Reviewed-by: Mike Kravetz <mike.kravetz@xxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- mm/hugetlb.c | 6 ++++++ 1 file changed, 6 insertions(+) --- a/mm/hugetlb.c~mm-hugetlb-optimize-the-surplus-state-transfer-code-in-move_hugetlb_state +++ a/mm/hugetlb.c @@ -5682,6 +5682,12 @@ void move_hugetlb_state(struct page *old SetHPageTemporary(oldpage); ClearHPageTemporary(newpage); + /* + * There is no need to transfer the per-node surplus state + * when we do not cross the node. + */ + if (new_nid == old_nid) + return; spin_lock(&hugetlb_lock); if (h->surplus_huge_pages_node[old_nid]) { h->surplus_huge_pages_node[old_nid]--; _
