FLT_ROUNDS not always 1?

[Richard Henderson]

On Mon, Feb 25, 2002 at 08:39:45PM -0800, Stan Shebs wrote:
> Is it really the case that for all GCC targets, fesetround() is
> a no-op that always returns 1?

No, it is not.

> Or is it just that systems that implement fesetround need to have
> their own config/float-foo.h that defines FLT_ROUNDS in a non-constant
> way, and no one has done this correctly yet?

Correct.

Actually, it would be trivial to add one that uses fegetround,
which will work everywhere that actually has fesetround.

Frankly I think it's silly that FLT_ROUNDS exists at all, 
since it must represent the execution time state of rounding.


r~