On Fri, 29 Apr 2022 22:12:28 GMT, Chris Plummer <cjplum...@openjdk.org> wrote:
>> The test counts calls of intercepted JNI functions, but doesn't completely >> filter out calls from other threads. >> Function isThreadExpected is used only for ExceptionOccurred function and >> the function checks only some known JFR/Graal threads. >> >> The change: >> - updates the test to count only the calls on the test thread; >> - adds verbose output. > > Also there is > [JDK-8284027](https://bugs.openjdk.java.net/browse/JDK-8284027), which also > is a failure related to unexpected threads and fails because > isThreadExpected() is not filtering all the proper threads (loom makes this > issue worse). It looks like you are trying to make the individual tests > smarter rather than fixing isThreadExpected(). I'm not saying this is the > wrong approach, but we should apply the approach to all tests that use > isThreadExpected() . @plummercj ji01t001 checks that the number of the calls >= of the expected. I thought about make the same condition here, but decided that it would be better to verify for equality, but ensure we count the calls from the test thread only ------------- PR: https://git.openjdk.java.net/jdk/pull/8475
