On Fri, 12 Jul 2024 03:24:42 GMT, SendaoYan <s...@openjdk.org> wrote:

> Hi all,
> Test TestClhsdbJstackLock.java/TestJhsdbJstackLock.java fails with -Xcomp 
> after [JDK-8335743](https://bugs.openjdk.org/browse/JDK-8335743). I think the 
> new failures was testcase bug.
> 
> https://github.com/openjdk/jdk/blob/627a32d6722c92f814c1ddd1c2fdf9a3b28cd655/src/jdk.hotspot.agent/share/classes/sun/jvm/hotspot/runtime/JavaVFrame.java#L112
> Locals are not always available, e.g., compiled native frames have no scope 
> so there are no locals. So the output `- waiting on <no object reference 
> available>` also acceptable.
> 
> The change has been verified by -Xmixed/-Xcomp(c2)/-Xcomp 
> -XX:TieredStopAtLevel=1(c1), only change the testcase, no risk.

> I just want to clarify that the "" message is coming from the SA 
> JavaVFrame.java file, not the vframe.cpp code you mention in the PR 
> description.

Thanks for the correction. The PR description has been updated.

Fix the testcase bug, I think it's not need the 2rd reviewer.

-------------

PR Comment: https://git.openjdk.org/jdk/pull/20151#issuecomment-2224890815
PR Comment: https://git.openjdk.org/jdk/pull/20151#issuecomment-2224894273

Reply via email to