Line number support for interpreter

[Bryce McKinlay]

Mark Wielaard wrote:

>I tested it some more and it doesn't give regressions (and as mentioned
>above eclipse 2.1.2 runs with this). But I haven't checked against Mauve
>yet. And I found at least one example which breaks with this code. I
>made the attached Invoke.java test (using reflection to call an
>interface method on an object that implements that interface).
>I cannot figure out what I changed that causes this to fail with my
>patch. Anyone a clue?
>  
>

This was a bug that was fixed recently, by:

2004-04-21  Bryce McKinlay  <mckinlay@redhat.com>

    * java/lang/natClass.cc (_Jv_LayoutInterfaceMethods): New method.
    Set method->index values for interface methods to their itable index.
    (initializeClass): Call _Jv_LayoutInterfaceMethods.

So make sure you've merged correctly, I guess, and didn't change 
something in interpret.cc that would prevent that code from running?

>It might be a good idea to add this small test to testsuite/libjava.lang
>since it seems a strange corner case that isn't normally tested/used.
>  
>

The libjava.lang/InvokeInterface.java test should cover this case 
already. Does that test not fail for you?

Regards

Bryce