Intel® C++ Compiler 16.0 User and Reference Guide
Blends int32 vectors using the instruction mask. Corresponding instruction is VPBLENDMD. This intrinsic only applies to Intel® Many Integrated Core Architecture (Intel® MIC Architecture).
extern __m512i __cdecl _mm512_mask_blend_epi32(__mmask16 k1, __m512i v2, __m512i v3); |
v2 |
int32 vector used for the blend operation |
v3 |
int32 vector used for the blend operation |
k1 |
instruction mask used for blend selection |
This intrinsic performs an element-by-element blending between int32 vectors v2 and v3, using mask k1 as selector. The result is written to int32 vector v2.
The k1 mask is not used as a write-mask for this instruction. Instead, the k1 mask is used as an element selector: every element of the destination is conditionally selected between vectors v2 or v3 using the value of the related mask bit ('0' for source vector v2, '1' for source vector v3).
Returns the result of the blend operation.