-
Bug
-
Resolution: Fixed
-
P2
-
10
-
b39
-
b40
-
generic
-
generic
| Issue | Fix Version | Assignee | Priority | Status | Resolution | Resolved In Build |
|---|---|---|---|---|---|---|
| JDK-8195705 | 11 | Hannes Wallnoefer | P2 | Resolved | Fixed | b01 |
| JDK-8197748 | 10u-cpu | Hannes Wallnoefer | P2 | Resolved | Fixed | master |
| JDK-8200975 | 10.0.2 | Unassigned | P2 | Resolved | Fixed | b01 |
| JDK-8195944 | 10.0.1 | Hannes Wallnoefer | P2 | Resolved | Fixed | b01 |
The Octane benchmarks have regressed 5-50% on all platforms when using 10-b39. The Octane benchmarks impacted are:
• Octane-Crypto -41 to -52%
• Octane-DeltaBlue -18 to -22%
• Octane-Gbemu -19 to 32%
• Octane-PdfJS -1 to -6%
• Octane-Raytrace -2 to -5%
• Octane-Richards - 44 to -61%
The regressions reproduce in triage (a repro run on the same machine).
The regressions reproduce in re-triage (a repro run on a different machine).
VMswap results indicate the cause is in the JKD (not the JVM).
There's only one nashorn bugfix in b39:JDK-8193567
Which is the suspected cause (not proven).
Will need the Nashorn engineers to investigate.
• Octane-Crypto -41 to -52%
• Octane-DeltaBlue -18 to -22%
• Octane-Gbemu -19 to 32%
• Octane-PdfJS -1 to -6%
• Octane-Raytrace -2 to -5%
• Octane-Richards - 44 to -61%
The regressions reproduce in triage (a repro run on the same machine).
The regressions reproduce in re-triage (a repro run on a different machine).
VMswap results indicate the cause is in the JKD (not the JVM).
There's only one nashorn bugfix in b39:
Which is the suspected cause (not proven).
Will need the Nashorn engineers to investigate.
- backported by
-
JDK-8195705 Very large regressions in Octane benchmarks using 10-b39
-
- Resolved
-
-
-
- Resolved
-
-
-
- Resolved
-
-
-
- Resolved
-
- relates to
-
-
- Resolved
-