[cfe-commits] r142797 - in /cfe/trunk: lib/Frontend/CompilerInvocation.cpp test/Lexer/ms-extensions.c

David Blaikie dblaikie at gmail.com
Mon Oct 24 09:53:48 PDT 2011

> Make -fms-compatibility imply -fms-extensions. Fixes PR11204.

Is this the right way to go? I could see extensions implying
compatibility (do we do this already? or is it intentional that one
might use ms-extensions without ms-compatibly?) but the other way
around seems less obvious to me. I may need to be able to compile with
MSVC (compatibility) but I don't want to dig a deeper hole
(extensions). I suppose if I cared I just wouldn't enable either &
check my code on both compilers & use the subset of functionality
across them, is that the line of reasoning? (though that doesn't quite
cover cases where the same code compiles in both compilers but behaves
differently due to lookup, etc).

And if extensions => compatibility and compatibility => extensions,
why have two flags?

- David

More information about the cfe-commits mailing list