code review round 0 for JVM/TI version fix (6849968) and phase assertion (6648438)

[David Holmes - Sun Microsystems]

Hi Dan,

In jvmtiEnv.cpp:

  410     return JVMTI_ERROR_NONE;
  411   } if (use_version_1_0_semantics()) {

I assume stylistically that was meant to be an "else if".

Otherwise this seems sound to me.

David

Daniel D. Daugherty said the following on 12/12/09 08:52:
> Greetings,
> 
> I have fixes for the following two JVM/TI bugs ready to go:
> 
> 6648438 4/4 src/share/vm/prims/jvmtiEnv.cpp:457
>            assert(phase == JVMTI_PHASE_LIVE,"sanity check")
> 6849968 3/3 JVMTI tests fails on jdk5.0 with hs14
> 
> 6849968 fixes a problem where JVM/TI version 1.0 semantics
> were not being used when that version was explicitly
> requested. I fixed 6648438 because it happened to be in the
> same place that I already had to tweak for 6849968. It also
> gets rid of an annoying intermittent fastdebug failure.
> 
> Here is the webrev URL:
> 
> http://cr.openjdk.java.net/~dcubed/batch-20091211-webrev/0/
> 
> This webrev is relative to the OpenJDK6 version of the code.
> Once approved, these fixes will be pushed to:
> 
> - OpenJDK6 (HSX-14-??)
> - JDK6-Update train (HSX-16-??)
> - OpenJDK7 (HSX-17-??)
> 
> Comments, questions and suggestions are always welcome!
> 
> Dan