>
>
> BTW: I have never seen any VAR_DECL here, at least not in my test case, do
> you have test cases for that?
If you call virtual method of undefined variable:
struct A {
virtual void foo ();
};
void
t()
{
struct A *a;
a->foo ();
}
It will make GCC to drop the virtual table lookup and directly call
__builtin_unreachable
>
> The condition here was if TREE_CODE(..) != PARM_DECL, and I extended it to be
> TREE_CODE (..) == PARM_DECL ||| TREE_CODE (..) == RESULT_DECL.
>
> That was purely by heuristic, the intention was to change as little as
> possible, so
> refining that to TREE_CODE (...) == PARM_DECL || (TREE_CODE (...) ==
> RESULT_DECL && !(DECL_BY_REFERENCE (...)))
> would be fully consistent with my heuristic.
>
> I don't have much more than the example where this RESULT_DECL showed up here.
> If this extra condition does not make the test case fail, it would be OK for
> me.
Actually trying to play with a testcase, for scalar return values, we never put
RESULT_DECL
there, only for values returned by reference. So I guess no extra condition is
needed.
So I guess the patch is fine as it is.
Honza
>
> But why is a RESULT_DECL per se undefined?
>
> Do you have a test case, where this extra condition would make a difference?
>
>
> Thanks,
> Bernd.
>
