In a comparison of 11-b1 .vs. 10-b46 (FCS candidate), the Tools-Javadoc-* measures all regress in the range of -4% to -10%. On all platforms measured: Linux/x64, macOS/x64, windows/x64, Solaris/SPARC.

11 was started essentially from the 10-b38 source base. We do not see a corresponding regression from 10-b38 to 10-b46. Only fixes in 11-b1 that are not in 10 should be candidates.

The regressions reproduce in:
• Triage (ie, re-run on the same machine)
• Re-triage (ie, run on a different machine)

The regressions hold through at least b04.

JVMswap fails, so results are not available. Can't say whether cause is isolated to the JDK or JVM.