Review request for JDK-8010731

[Attila Szegedi]

We didn't use those symbol names anywhere, only used .slot() on those CompilerConstant values. No local variable table entries are created for them as it is now. These symbols are only used in constructors of JO classes anyway, and those are quite dull - mostly just initializing fields to undefined.
I agree that someone could add local variable table (LVT) entries for them - in that case, they'd need symbol names. However, as long as there are no LVT entries, I figured we shouldn't have a symbol name if it's not used, lest it misleads someone.

On Apr 23, 2013, at 12:59 PM, "Jim Laskey (Oracle)" <james.laskey at oracle.com> wrote:

> Not sure having this, scope and arguments as anonymous will fly - debuggers need access to these.
> 
> 
> 128     /** "this" used in non-static Java methods; always in slot 0 */
> 
> 129     JAVA_THIS(null, 0),
> 130 
> 131     /** Map parameter in scope object constructors; always in slot 1 */
> 132     INIT_MAP(null, 1),
> 
> 133 
> 
> 134     /** Parent scope parameter in scope object constructors; always in slot 2 */
> 135     INIT_SCOPE(null, 2),
> 
> 136 
> 
> 137     /** Arguments parameter in scope object constructors; in slot 3 when present */
> 138     INIT_ARGUMENTS(null, 3),
> 
> 
> +1 otherwise.
> 
> On 2013-04-23, at 7:45 AM, Attila Szegedi <attila.szegedi at oracle.com> wrote:
> 
>> Please review JDK-8010731 at http://cr.openjdk.java.net/~attila/8010731/webrev.00
>> 
>> Thanks,
>> Attila.
>