[PATCH] hugetlb: fix surplus pages accounting

[Michal Hocko]

Dan Rue has noticed that libhugetlbfs test suite fails counter test:

root@localhost:~# mount_point="/mnt/hugetlb/"
root@localhost:~# echo 200 > /proc/sys/vm/nr_hugepages
root@localhost:~# mkdir -p "${mount_point}"
root@localhost:~# mount -t hugetlbfs hugetlbfs "${mount_point}"
root@localhost:~# export LD_LIBRARY_PATH=/root/libhugetlbfs/libhugetlbfs-2.20/obj64
root@localhost:~# /root/libhugetlbfs/libhugetlbfs-2.20/tests/obj64/counters
Starting testcase "/root/libhugetlbfs/libhugetlbfs-2.20/tests/obj64/counters", pid 3319
Base pool size: 0
Clean...
FAIL    Line 326: Bad HugePages_Total: expected 0, actual 1

The bug was bisected to 0c397daea1d4 ("mm, hugetlb: further simplify
hugetlb allocation API"). The reason is that alloc_surplus_huge_page misaccounts
per node surplus pages. We should increase surplus_huge_pages_node rather than
nr_huge_pages_node which is already handled by alloc_fresh_huge_page.

Fixes: 0c397daea1d4 ("mm, hugetlb: further simplify hugetlb allocation API")
Reported-by: Dan Rue <dan.rue@xxxxxxxxxx>
Tested-by: Dan Rue <dan.rue@xxxxxxxxxx>
Reviewed-by: Mike Kravetz <mike.kravetz@xxxxxxxxxx>
Signed-off-by: Michal Hocko <mhocko@xxxxxxxx>
---
 mm/hugetlb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 7c204e3d132b..a963f2034dfc 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1583,7 +1583,7 @@ static struct page *alloc_surplus_huge_page(struct hstate *h, gfp_t gfp_mask,
 		page = NULL;
 	} else {
 		h->surplus_huge_pages++;
-		h->nr_huge_pages_node[page_to_nid(page)]++;
+		h->surplus_huge_pages_node[page_to_nid(page)]++;
 	}
 
 out_unlock:
-- 
2.16.1

-- 
Michal Hocko
SUSE Labs