Developer Guide and Reference

Contents

_mm_fnmadd_ss

Multiply-adds negated scalar single-precision floating-point values of three float32 vectors. The corresponding FMA instruction is
VFNMADD<XXX>SS
, where XXX could be 132, 213, or 231.

Syntax

extern __m128 _mm_fnmadd_ss(__m128 a, __m128 b, __m128 c);
Arguments
a
float64 vector used for the operation
b
float64 vector also used for the operation
c
float64 vector also used for the operation
Description
Performs a set of scalar SIMD negated multiply-add computation on scalar single-precision floating-point values in the low 32-bits of three source operands,
a
,
b
, and
c
. The float32 values in two operands,
a
and
b
, are multiplied and the negated infinite precision intermediate result obtained is added to the float32 value in the third operand,
c
. The final result is rounded to the nearest float32 value.
The compiler defaults to using the
VFNMADD213SS
instruction and uses the other forms
VFNMADD132SS
or
VFNMADD231SS
only if a low level optimization decides it as useful or necessary. For example, the compiler could change the default if it finds that another instruction form saves a register or eliminates a move.
Returns
Result of the negated multiply-add operation.