For each element, performs MUL and ADD/SUB calculation with 3 operands and store the result in the first operand. (returns the result.)
Which operands to MUL, and which operand to ADD/SUB, depend on the order of the numbers (1, 2, 3) in the instruction name. depend on the order of arguments for intrinsic.
| ????? | instruction name | odd | even |
|---|---|---|---|
| FMADD | VFMADD132PS | (1) * (3) + (2) | |
| VFMADD213PS | (2) * (1) + (3) | ||
| VFMADD231PS | (2) * (3) + (1) | ||
| _mm_fmadd_ps _mm256_fmadd_ps _mm512_fmadd_ps |
a * b + c | ||
| FMSUB | VFMSUB132PS | (1) * (3) - (2) | |
| VFMSUB213PS | (2) * (1) - (3) | ||
| VFMSUB231PS | (2) * (3) - (1) | ||
| _mm_fmsub_ps _mm256_fmsub_ps _mm512_fmsub_ps |
a * b - c | ||
| FMADDSUB | VFMADDSUB132PS | (1) * (3) + (2) | (1) * (3) - (2) |
| VFMADDSUB213PS | (2) * (1) + (3) | (2) * (1) - (3) | |
| VFMADDSUB231PS | (2) * (3) + (1) | (2) * (3) - (1) | |
| _mm_fmaddsub_ps _mm256_fmaddsub_ps _mm512_fmaddsub_ps |
a * b + c | a * b - c | |
| FMSUBADD | VFMSUBADD132PS | (1) * (3) - (2) | (1) * (3) + (2) |
| VFMSUBADD213PS | (2) * (1) - (3) | (2) * (1) + (3) | |
| VFMSUBADD231PS | (2) * (3) - (1) | (2) * (3) + (1) | |
| _mm_fmsubadd_ps _mm256_fmsubadd_ps _mm512_fmsubadd_ps |
a * b - c | a * b + c | |
| FNMADD | VFNMADD132PS | - (1) * (3) + (2) | |
| VFNMADD213PS | - (2) * (1) + (3) | ||
| VFNMADD231PS | - (2) * (3) + (1) | ||
| _mm_fnmadd_ps _mm256_fnmadd_ps _mm512_fnmadd_ps |
- a * b + c | ||
| FNMSUB | VFNMSUB132PS | - (1) * (3) - (2) | |
| VFNMSUB213PS | - (2) * (1) - (3) | ||
| VFNMSUB231PS | - (2) * (3) - (1) | ||
| _mm_fnmsub_ps _mm256_fnmsub_ps _mm512_fnmsub_ps |
- a * b - c | ||
_mask_ if k bit is 0, a is copied.
_mask3_ if k bit is 0, c is copied.
_maskz_ if k bit is 0, zero cleared.