https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91645
Aldy Hernandez <aldyh at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |aldyh at gcc dot gnu.org,
| |amacleod at redhat dot com,
| |jakub at gcc dot gnu.org,
| |law at gcc dot gnu.org
--- Comment #7 from Aldy Hernandez <aldyh at gcc dot gnu.org> ---
> Now, the problem is that GCC doesn't seem to optimize away the call to sqrtf
> based on some surrounding code. As an example, it would be neat to have this
> (or something similar) to get compiled into the same mulss-sqrtss-ret:
>
> float test (float x)
> {
> float y = x*x;
> if (y >= 0.f)
> return std::sqrt(y);
> __builtin_unreachable();
> }
>
> If I understand it correctly, the 'y >= 0.f' excludes 'y' being NaN and 'y'
> being negative (though this is excluded by 'y = x*x'), so there is no need
> to check if the argument to `std::sqrt` is any bad, enabling to just do
> 'sqrtss' and return.
I'm not an FP expert, but I think we have enough information to do this right
now. The evrp dump now has:
=========== BB 2 ============
Imports: y_2
Exports: y_2
<bb 2> :
y_2 = x_1(D) * x_1(D);
if (y_2 >= 0.0)
goto <bb 3>; [INV]
else
goto <bb 4>; [INV]
2->3 (T) y_2 : [frange] float [0.0, Inf] !NAN
2->4 (F) y_2 : [frange] float [ -Inf, 0.0]
=========== BB 3 ============
y_2 [frange] float [0.0, Inf] !NAN
<bb 3> :
_6 = __builtin_sqrtf (y_2);
return _6;
Which means that y_2 is known to be [0.0, Inf] excluding a NAN.
What needs to happen for the call to __builtin_sqrtf to be optimized to sqrtss?
