I implemented such refactoring in JDK-8331253 changes but it caused regression so I have to back out it JDK-8333226.
I need to investigate why it caused regression which I did not expect - refactoring should be calculating about the same value for skipped instructions. This RFE is for do such investigation and redo refactoring.
I need to investigate why it caused regression which I did not expect - refactoring should be calculating about the same value for skipped instructions. This RFE is for do such investigation and redo refactoring.
- relates to
-
JDK-8333792 ZGC: refine estimation of C2's barrier stub size
- Open
-
JDK-8333226 Regressions 2-3% in Compress ZGC after 8331253
- Resolved
-
JDK-8331253 16 bits is not enough for nmethod::_skipped_instructions_size field
- Closed