#551: t/pmc/nci.t: 2 test failures
---------------------+------------------------------------------------------
Reporter: jkeenan | Owner:
Type: bug | Status: new
Priority: normal | Milestone:
Component: none | Version:
Severity: medium | Keywords: pmc nci pointer
Lang: | Patch:
Platform: |
---------------------+------------------------------------------------------
Comment(by rg):
I'm not entirely happy with cotto's suggestion. For one, it breaks the
correlation between set_pointer and get_pointer.
I've now attached my other 2 possible solutions to this problem. One is to
extend the Pointer pmc with a new function. This keeps the compatibility,
however it needs some infrastructure changes to support returning a void
**, plus it introduces a new vtable method, that probably no other class
will use.
The other possible solution is to just not handle 'V' arguments with the
jitted call stack, falling back to the predefined signatures list, just
like any non-jitted architecture. While this is removing functionality
that has worked before, it didn't work anymore, because of the conversion
to ATTRs. With nobody really familiar with jit and the would-be examples
(of passing the address of local variables to the called function) also
not handled (with a "don't know how to do it" comment none the less), I
don't think this is a big drawback.
So, who is going to decide which way to go? Can we take a vote? Any other
suggestions?
--
Ticket URL: <https://trac.parrot.org/parrot/ticket/551#comment:9>
Parrot <https://trac.parrot.org/parrot/>
Parrot Development
_______________________________________________
parrot-tickets mailing list
[email protected]
http://lists.parrot.org/mailman/listinfo/parrot-tickets
