On Thu, 6 Oct 2022 02:24:12 GMT, Alex Menkov <amen...@openjdk.org> wrote:
> Test failure is a duplicate of > [JDK-8284777](https://bugs.openjdk.org/browse/JDK-8284777), but the test > needs to be updated: > - the test instruments all loaded classes, so need to updated ASM version to > support records and permits; > Instrumentor.addNativeMethodTrackingInjection is used only by this test; > - return empty array from transforming does not cause test failure, added > code to handle agent errors Marked as reviewed by cjplummer (Reviewer). I wouldn't consider this a duplicate of [JDK-8284777](https://bugs.openjdk.org/browse/JDK-8284777). It was caused changes to make println use ReentrantLock, which resulted in a CCE when a println was done from the transform method. The println often triggered loading of the class being transformed, usually the ExclusiveNode inner class. However, now it seems that ExclusiveNode is always loaded early enough that this circularity doesn't happen. ------------- PR: https://git.openjdk.org/jdk/pull/10589
