> (In reply to Justin Wood (:Callek) from comment #4)
> > Do we have a reason for diverging from m-c here, would this break cross
> > compiles?
> > ...
> > Other than that r+, but I want an answer as to "why" on the XPIDL_COMPILE
> > bit.
> Actually looked again, and I see that we are safe to remove that in c-c, I
> do see it in use in m-c, in ONLY one spot, and as such probably defeats the
> purpose of this test.