I strongly suspect that Parrot_PMC_invoke does not need to exist. It is conceivable that an extender could write a custom Sub variant and would want to implement VTABLE_invoke, but I can't think of any real reason for an extender to actually call it directly. All invocations are expected to go through Parrot_pcc_invoke_* functions, where VTABLE_invoke is called internally.
I don't think that VTABLE_invoke is dangerous (unlike Parrot_PMC_destroy, which is very dangerous), but I also don't think it's useful on it's own. --Andrew Whitworth On Fri, May 27, 2011 at 7:19 PM, Jonathan "Duke" Leto <[email protected]> wrote: > Howdy, > > Parrot_PMC_invoke returns a Parrot_Opcode*. It is the only Parrot_PMC_* > function that does this. > > Questions: > 1) Did this function accidentally leak out, or do we actually want > parrot extenders to be fiddling with Parrot_Opcodes ? > > 2) Should I test this or remove it? > > Duke > > -- > Jonathan "Duke" Leto <[email protected]> > 209.691.DUKE // http://leto.net > NOTE: Personal email is only checked twice a day at 10am/2pm PST, > please call/text for time-sensitive matters. > _______________________________________________ > http://lists.parrot.org/mailman/listinfo/parrot-dev > _______________________________________________ http://lists.parrot.org/mailman/listinfo/parrot-dev
