> Great, I think diagnosing at CodeGen is the better choice here.
>
> Rafael, in the new patch could you swap around one of the tests to prove
> that checking isn't dependent on parse order?
>
> -void alias1() {}
> void alias2 [[gnu::alias("_Z6alias1v")]] ();
> +void alias1() {}
Since that was already passing, I just added to the test in r193099.
Cheers,
Rafael
_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
