simdpp::fmsub(floatint-point)
From libsimdpp-docs
< fp
template<unsigned N, class E1, class E2, class E3>
Ret<N, _DETAIL_> fmsub( Vec1<N,E1> a, Vec2<N,E2> b, Vec3<N,E3> ); |
||
Performs a fused multiply-subtract operation of floating-point in three vectors, as if by a*b-c
for each element. The implementation behaves as if the following set of overloads is provided:
Ret |
Vec1 |
Vec2 |
Vec3
|
---|---|---|---|
float32 | float32 | float32 | float32 |
float64 | float64 | float64 | float64 |
The return type is a vector expression.
Parameters[edit]
a, b, c | - | floating-point vectors |
Return value[edit]
A vector expression evaluating to a * b - c
.
Equivalent operation[edit]
r0 = a0 * b0 - c0 ... rN = aN * bN - cN
See also[edit]
This section is incomplete |