Babneet Singh

> I can disable it temporarily but that is not a fix. I ran the test locally and in grinders (on internal pLinux machines) to reproduce the issue. But, I...

> Since ub16-ppcle-1 is disabled in jenkins, it's sitting idle. in the process of acquiring access to ssh and run a grinder on the above machine. fyi @AdamBrousseau.

### Summary This issue only happens with the `JIT enabled in JDK17`. The issue is neither seen with `-Xint` nor `JDK11`. I have personally seen high **RES** memory usage on...

> https://openj9-jenkins.osuosl.org/job/Test_openjdk17_j9_sanity.functional_x86-64_linux_criu_Nightly/19/tapResults/ ``` Testing: Create CRIU checkpoint image and restore three times - testSystemNanoTime Test start time: 2022/08/09 03:49:12 Eastern Standard Time Running command: bash /home/jenkins/workspace/Test_openjdk17_j9_sanity.functional_x86-64_linux_criu_Nightly_testList_1/aqa-tests/TKG/../../jvmtest/functional/cmdLineTests/criu/criuScript.sh /home/jenkins/workspace/Test_openjdk17_j9_sanity.functional_x86-64_linux_criu_Nightly_testList_1/aqa-tests/TKG/../../jvmtest/functional/cmdLineTests/criu /home/jenkins/workspace/Test_openjdk17_j9_sanity.functional_x86-64_linux_criu_Nightly_testList_1/openjdkbinary/j2sdk-image/bin/java " -Xjit:count=0...

@gacholio When a virtual thread is being inspected in JVMTI, do we need to protect the virtual thread and corresponding carrier thread from freeing their native resources such as the...

> Is the unmount native called before the vthread ends and when it yields? Both of these need to block if the vthread is being inspected. Yes, `notifyJvmtiUnmountBegin` is called...

> Also wondering if haltThreadForInspection will also need to be modified in a similar fashion. `haltThreadForInspection` is always called after `getVMThread`. So, `haltThreadForInspection` should work with virtual threads as-is because...

> The native should be called before the yield (so that the vmthread stack still represents the continuation, not the carrier). This is true. See [VirtualThread.java#L367](https://github.com/ibmruntimes/openj9-openjdk-jdk19/blob/3b112d4e1c05e85584248cda15d4c3c3c4c1d6be/src/java.base/share/classes/java/lang/VirtualThread.java#L367). The native (`notifyJvmtiUnmountBegin`) is...

@EricYangIBM Can you review these changes? Also, let me know if they work correctly with your TLS changes and related tests.

> can you please take a look at this? Yes. I have a few existing tasks from last year. So, I will take a look at this sometime next week.