[v3] atomics

Lawrence Crowl crowl@google.com
Tue Apr 15 18:07:00 GMT 2008


On 4/15/08, Peter Dimov <pdimov@mmltd.net> wrote:
>  As an aside, I'd have expected the implementation to use the __sync
> primitives directly instead of spinlock emulation?

Long term, I think we need to expose the atomic_whatever functions
to the back end so that they can do the appropriate optimizations.

-- 
Lawrence Crowl



More information about the Libstdc++ mailing list