Hi Yasumasa,
Looks good.
Just one minor comment.
http://cr.openjdk.java.net/~ysuenaga/JDK-8248362/webrev.00/src/hotspot/share/prims/jvmtiEnv.cpp.udiff.html
Thanks,
Serguei
On 7/22/20 07:38, Yasumasa Suenaga wrote:
Looks good.
Just one minor comment.
http://cr.openjdk.java.net/~ysuenaga/JDK-8248362/webrev.00/src/hotspot/share/prims/jvmtiEnv.cpp.udiff.html
// JVMTI get java stack frame location at safepoint.Replace: "at safepoint" => "with handshake".
Thanks,
Serguei
On 7/22/20 07:38, Yasumasa Suenaga wrote:
Hi all,
Please review this change:
JBS: https://bugs.openjdk.java.net/browse/JDK-8248362
webrev: http://cr.openjdk.java.net/~ysuenaga/JDK-8248362/webrev.00/
Migrate JVMTI frame operations to Thread-Local Handshakes from VM Operations.
- VM_GetFrameCount
- VM_GetFrameLocation
They affects both GetFrameCount() and GetFrameLocation() JVMTI functions.
This change has passed all tests on submit repo (mach5-one-ysuenaga-JDK-8248362-20200722-1249-12859056), and vmTestbase/nsk/{jdi,jdw,jvmti}, serviceability/{jdwp,jvmti} .
Thanks,
Yasumasa