Changes in settings

05f1edac - selftests/mm: run all tests from run_vmtests.sh

Mon, 24 Jul 2023 08:25:22 +0000

selftests/mm: run all tests from run_vmtests.shIt is very unclear to me how one is supposed to run all the mm selftestsconsistently and get clear results.Most of the test programs are launched by both run_vmtests.sh andrun_kselftest.sh: hugepage-mmap hugepage-shm map_hugetlb hugepage-mremap hugepage-vmemmap hugetlb-madvise map_fixed_noreplace gup_test gup_longterm uffd-unit-tests uffd-stress compaction_test on-fault-limit map_populate mlock-random-test mlock2-tests mrelease_test mremap_test thuge-gen virtual_address_range va_high_addr_switch mremap_dontunmap hmm-tests madv_populate memfd_secret ksm_tests ksm_functional_tests soft-dirty cowHowever, of this set, when launched by run_vmtests.sh, some of theprograms are invoked multiple times with different arguments. Wheninvoked by run_kselftest.sh, they are invoked without arguments (and asa consequence, some fail immediately).Some test programs are only launched by run_vmtests.sh: test_vmalloc.shAnd some test programs and only launched by run_kselftest.sh: khugepaged migration mkdirty transhuge-stress split_huge_page_test mdwe_test write_to_hugetlbfsFurthermore, run_vmtests.sh is invoked by run_kselftest.sh, so in thiscase all the test programs invoked by both scripts are run twice!Needless to say, this is a bit of a mess. In the absence of fullyunderstanding the history here, it looks to me like the best solution isto launch ALL test programs from run_vmtests.sh, and ONLY invokerun_vmtests.sh from run_kselftest.sh. This way, we get full control overthe parameters, each program is only invoked the intended number oftimes, and regardless of which script is used, the same tests get run inthe same way.The only drawback is that if using run_kselftest.sh, it's top-level tapresult reporting reports only a single test and it fails if any of thecontained tests fail. I don't see this as a big deal though since westill see all the nested reporting from multiple layers. The other issuewith this is that all of run_vmtests.sh must execute within a singlekselftest timeout period, so let's increase that to something moresuitable.In the Makefile, TEST_GEN_PROGS will compile and install the tests andwill add them to the list of tests that run_kselftest.sh will run.TEST_GEN_FILES will compile and install the tests but will not add themto the test list. So let's move all the programs from TEST_GEN_PROGS toTEST_GEN_FILES so that they are built but not executed byrun_kselftest.sh. Note that run_vmtests.sh is added to TEST_PROGS, whichmeans it ends up in the test list. (the lack of "_GEN" means it won't becompiled, but simply copied).Link: https://lkml.kernel.org/r/20230724082522.1202616-9-ryan.roberts@arm.comSigned-off-by: Ryan Roberts Acked-by: David Hildenbrand Acked-by: Peter Xu Cc: Florent Revest Cc: Jérôme Glisse Cc: John Hubbard Cc: Mark Brown Cc: Shuah Khan Signed-off-by: Andrew Morton List of files: /linux-6.15/tools/testing/selftests/mm/settings

00030332 - selftests/mm: make migration test robust to failure

Mon, 24 Jul 2023 08:25:20 +0000

selftests/mm: make migration test robust to failureThe `migration` test currently has a number of robustness problems thatcause it to hang and leak resources.Timeout: There are 3 tests, which each previously ran for 60 seconds. However, the timeout in mm/settings for a single test binary was set to 45seconds. So when run using run_kselftest.sh, the top level timeout wouldtrigger before the test binary was finished. Solve this by meeting in themiddle; each of the 3 tests now runs for 20 seconds (for a total of 60),and the top level timeout is set to 90 seconds.Leaking child processes: the `shared_anon` test fork()s some children butthen an ASSERT() fires before the test kills those children. The assertcauses immediate exit of the parent and leaking of the children. Furthermore, if run using the run_kselftest.sh wrapper, the wrapper wouldget stuck waiting for those children to exit, which never happens. Solvethis by setting the "parent death signal" to SIGHUP in the child, so thatthe child is killed automatically if the parent dies.With these changes, the test binary now runs to completion on arm64, with2 tests passing and the `shared_anon` test failing.Link: https://lkml.kernel.org/r/20230724082522.1202616-7-ryan.roberts@arm.comSigned-off-by: Ryan Roberts Reviewed-by: David Hildenbrand Cc: Florent Revest Cc: Jérôme Glisse Cc: John Hubbard Cc: Mark Brown Cc: Peter Xu Cc: Shuah Khan Signed-off-by: Andrew Morton List of files: /linux-6.15/tools/testing/selftests/mm/settings

baa489fa - selftests/vm: rename selftests/vm to selftests/mm

Tue, 03 Jan 2023 18:07:53 +0000

selftests/vm: rename selftests/vm to selftests/mmRename selftets/vm to selftests/mm for being more consistent with thecode, documentation, and tools directories, and won't be confused withvirtual machines.[sj@kernel.org: convert missing vm->mm changes] Link: https://lkml.kernel.org/r/20230107230643.252273-1-sj@kernel.orgLink: https://lkml.kernel.org/r/20230103180754.129637-5-sj@kernel.orgSigned-off-by: SeongJae Park Cc: Jonathan Corbet Cc: Shuah Khan Signed-off-by: Andrew Morton List of files: /linux-6.15/tools/testing/selftests/mm/settings