Hi Nick,
> Why is that? The LangRef describes weak in terms of common, and common
> in terms of linkonce. Is there some semantic reason that an alias with
> common or linkonce wouldn't make sense,
Right. We're reflecting linkage specification from gcc, where alias
can be external, weak or internal (undocumented, but created by c++
frontend). As alias is purely linker stuff, there is no 'linkonce' or
'common' there. Symbol can be global, internal or weak.
> part of the verifier was never updated when new linkage types were added?
No, verifier is correct. I'll update langref.
--
With best regards, Anton Korobeynikov
Faculty of Mathematics and Mechanics, Saint Petersburg State University