[Bug middle-end/51766] [4.7 regression] sync_fetch_and_xxx atomicity

[redi at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51766

--- Comment #10 from Jonathan Wakely <redi at gcc dot gnu.org> 2012-01-10 18:22:01 UTC ---
(In reply to comment #8)
> IBM should have the freedom to choose memory
> models that benefit its architectures.

I meant to add that the new __atomic builtins give that freedom, by design.

Changing the documented behaviour of the __sync builtins to do something they
weren't designed for doesn't seem like a good idea - if they're inefficient
then convince people not to use them.