Commit dd9b35ef authored by Breno Leitao's avatar Breno Leitao Committed by Andrew Morton

selftests/mm: restore number of hugepages

The test mm `hugetlb_fault_after_madv` selftest needs one and only one
huge page to run, thus it sets `/proc/sys/vm/nr_hugepages` to 1.

The problem is that further tests require the previous number of hugepages
allocated in order to succeed.

Save the number of huge pages before changing it, and restore it once the
test finishes, so, further tests could run successfully.

Link: https://lkml.kernel.org/r/20231103173400.1608403-1-leitao@debian.org
Fixes: 116d5730 ("selftests/mm: add a new test for madv and hugetlb")
Signed-off-by: default avatarBreno Leitao <leitao@debian.org>
Reported-by: default avatarRyan Roberts <ryan.roberts@arm.com>
Closes: https://lore.kernel.org/all/662df57e-47f1-4c15-9b84-f2f2d587fc5c@arm.com/Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
parent 9297e536
...@@ -223,9 +223,12 @@ CATEGORY="hugetlb" run_test ./hugepage-mremap ...@@ -223,9 +223,12 @@ CATEGORY="hugetlb" run_test ./hugepage-mremap
CATEGORY="hugetlb" run_test ./hugepage-vmemmap CATEGORY="hugetlb" run_test ./hugepage-vmemmap
CATEGORY="hugetlb" run_test ./hugetlb-madvise CATEGORY="hugetlb" run_test ./hugetlb-madvise
nr_hugepages_tmp=$(cat /proc/sys/vm/nr_hugepages)
# For this test, we need one and just one huge page # For this test, we need one and just one huge page
echo 1 > /proc/sys/vm/nr_hugepages echo 1 > /proc/sys/vm/nr_hugepages
CATEGORY="hugetlb" run_test ./hugetlb_fault_after_madv CATEGORY="hugetlb" run_test ./hugetlb_fault_after_madv
# Restore the previous number of huge pages, since further tests rely on it
echo "$nr_hugepages_tmp" > /proc/sys/vm/nr_hugepages
if test_selected "hugetlb"; then if test_selected "hugetlb"; then
echo "NOTE: These hugetlb tests provide minimal coverage. Use" echo "NOTE: These hugetlb tests provide minimal coverage. Use"
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment