diff options
| author | Michal Hocko <[email protected]> | 2018-03-09 15:50:55 -0800 |
|---|---|---|
| committer | Linus Torvalds <[email protected]> | 2018-03-09 16:40:01 -0800 |
| commit | 4704dea36dd9e5b4bf37ed20f7f15e70632ccdd0 (patch) | |
| tree | bc2e74779d5e66d124804c62a4f35d806c40f2d6 /tools/perf/scripts/python/export-to-postgresql.py | |
| parent | 719ea86151f31f9d30dc854e028d05dc4afb3224 (diff) | |
hugetlb: fix surplus pages accounting
Dan Rue has noticed that libhugetlbfs test suite fails counter test:
# mount_point="/mnt/hugetlb/"
# echo 200 > /proc/sys/vm/nr_hugepages
# mkdir -p "${mount_point}"
# mount -t hugetlbfs hugetlbfs "${mount_point}"
# export LD_LIBRARY_PATH=/root/libhugetlbfs/libhugetlbfs-2.20/obj64
# /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.
Link: http://lkml.kernel.org/r/[email protected]
Fixes: 0c397daea1d4 ("mm, hugetlb: further simplify hugetlb allocation API")
Signed-off-by: Michal Hocko <[email protected]>
Reported-by: Dan Rue <[email protected]>
Tested-by: Dan Rue <[email protected]>
Reviewed-by: Mike Kravetz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Diffstat (limited to 'tools/perf/scripts/python/export-to-postgresql.py')
0 files changed, 0 insertions, 0 deletions