Strategy and Workflow
Due to the amount of data that LKFT produces, test results need to be clear and actionable. This means there is a very low tolerance for any tests which sometimes pass and sometimes fail. Further, any test that is expected to fail is skipped so that results are always in a ‘all green’ state. This makes triaging new regressions straightforward.
This strategy has a few consequences. Skipping inconsistent tests could hide actual problems. For this reason, whenever a test is added to a skiplist, a bug is opened in bugzilla. All bugzilla bugs are regularly reviewed, investigated, and reported internally or upstream.
All regressions are triaged and reported internally and/or upstream. Typically the test will also be added to the skiplist until it is fixed.
LKFT maintains a staging environment that runs without a skiplist, so that it is possible to know when a regression is fixed, at which point it is then removed from the skiplist.
Linux Test Project (LTP)
The following test sets are run in LTP:
The version of LTP used is the most recent release, plus backports that fix regressions, if any.
Linux Kernel Selftests
Linux Kernel Selftests, or kselftests, are tests that are included in the linux kernel tree under tools/testing/selftests/.
The version used depends on the kernel version under test:
- linux-next uses selftest included in linux-next
- mainline uses selftest included in mainline
- Latest stable uses selftest included in latest stable
- All other LTS branches also use the version from latest stable
Running a mismatched kselftest version does create false positives from time to time, but it is the best balance between a supported and up to date set of tests and an older kernel.
libhugetlbfs is run using the latest version in its next branch (02df38e93e as of 2018.03.09).