Request for reviews (L): 6939203: JSR 292 needs method handle constants

[Tom Rodriguez]

In c1_Runtime1.cpp, you've deleted the assignment to k in the ldc code:

-          k = resolved->klass_part()->java_mirror();

I'm still looking through the rest.

tom

On Jun 2, 2010, at 1:29 PM, John Rose wrote:

> On Apr 22, 2010, at 12:22 AM, Christian Thalinger wrote:
> 
>> On Thu, 2010-04-22 at 00:08 -0700, John Rose wrote:
>>> Here's the final webrev (if you agree that I fixed the issues you raised):
>>> http://cr.openjdk.java.net/~jrose/6939203/hs-webrev.01
>> 
>> Yes, that looks good.  -- Christian
> 
> The bug tail from the cpindex changes is tamed, so now it's time to push the "ldc MH" functionality.
> 
> The round of bug fixes on cpindex led me to look more closely at BytecodeTracer and MethodComparator.  This led to some more changes to make to the ldc MH code.
> 
> Here they are, for re-review:
>  http://cr.openjdk.java.net/~jrose/6939203/hs-webrev.03
> 
> Here are the diffs relative to the last reviewed change set (02):
>  http://cr.openjdk.java.net/~jrose/6939203/hs-webrev.03/diff-02-to-03.patch
> 
> Summary:
> - invoke, field ref, and load constant bytecode accessors are regularized; constructors are all the same signature.
> - CP cache indexes are more clearly distinguished from plain pool indexes.
> - BytecodeTracer and MethodComparator now work properly on fast_aldc instructions.
> - BytecodeTracer prints object constants a little more usefully.
> 
> Thanks in advance.
> 
> -- John