On 10 November 2014 18:09, Andrew Haley wrote: > Hmm. That is reasonable enough, but given that GCC nonstandardly > allows atomic_store() et al to work on float, why prefer a GCC > nonstandard builtin? No reason I can see, especially as atomic_store(&f, 1.0f) works with a literal as the second argument while __atomic_store(&f, 1.f) doesn't.
