The nsk_jvmti_setFailStatus() sometimes is called after test check results. In these cases the warning logs are generated and hide the real failure reasons. Also, I think it is a error-prone way to set and check error, since check might be just forgotten. Also, the test execution after failure might be incorrect and also make failure analysis harder. So I think it makes sense to always fail when nsk_jvmti_setFailStatus() is called.
If this is going to work I'll rename it later and add add optional message to be more informative. ------------- Commit messages: - more fixed tests - removed empty lines. - set fail status should fails. - 8351375: nsk/jvmti/scenarios/events/EM04/em04t001 test should exit with code 97 if fails in unload phase Changes: https://git.openjdk.org/jdk/pull/24040/files Webrev: https://webrevs.openjdk.org/?repo=jdk&pr=24040&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8351375 Stats: 71 lines in 5 files changed: 47 ins; 19 del; 5 mod Patch: https://git.openjdk.org/jdk/pull/24040.diff Fetch: git fetch https://git.openjdk.org/jdk.git pull/24040/head:pull/24040 PR: https://git.openjdk.org/jdk/pull/24040
