Thanks for the patch!
My concern though is that I’m not sure I understand why gcc-4.9 isn’t emitting
the symbols you listed.
There’s a few kinds of symbols that aren’t getting emitted (that I’ve seen in
other gcc-4.9 bug reports).
1) Virtual destructors with an inline body of {}
2) Virtual functions such as OnHatResp() with an inline body of {}
3) stdlibc++ stuff like std::ctype<char>::do_widen(char)
I assume that 3) is just stuff where I need to add some (optional=templinst)
tags.
1) and 2) have me confused though. Shouldn’t GCC have to emit the definition
of a virtual function, even if it can be inlined away? I hardly claim to be an
expert in modern C++, so I am very likely confused … but I’d like to understand
what’s going on a bit better before I go removing symbols.
(Also, if I take this patch, won’t cigi-ccl no longer compile with gcc-4.8? I
don’t have time to verify this today, but hope to in a bit.)
--
To UNSUBSCRIBE, email to [email protected]
with a subject of "unsubscribe". Trouble? Contact [email protected]
