This was a problem (obstruction) to fix for JDK-8054292 ; InterpreterCodelet is apparently cast into existence in some preallocated memory, and by default its Constructor is not run (the fix for 8054292 explicitly invokes the constructor with placement as a workaround).
It's possible that all that lacks is good documentation.
See also (because 8054292 is confidential, but its fix is not) http://hg.openjdk.java.net/jdk9/hs-comp/hotspot/rev/bb102ba0100f
It's possible that all that lacks is good documentation.
See also (because 8054292 is confidential, but its fix is not) http://hg.openjdk.java.net/jdk9/hs-comp/hotspot/rev/bb102ba0100f