Intel® C++ Compiler 16.0 User and Reference Guide

_mm512_fixupnan_ps/ _mm512_mask_fixupnan_ps

Fixes up NaNs from float32 vectors. Corresponding instruction is VFIXUPNANPS. This intrinsic only applies to Intel® Many Integrated Core Architecture (Intel® MIC Architecture).

Syntax

Without Mask

extern __m512 __cdecl _mm512_fixupnan_ps(__m512 v1, __m512 v2, __m512i v3);

With Mask

extern __m512 __cdecl _mm512_mask_fixupnan_ps(__m512 v1_old, __mmask16 k1, __m512 v2, __m512i v3);

Parameters

v1

float32 vector used for the operation

v2

float32 vector used for the operation

v3

int32 vector used for the operation

v1_old

Source vector that retains old values of the destination vector; the resulting vector gets corresponding elements from v1_old for zero mask bits

k1

Writemask; only those elements of the source vectors with corresponding bit set to '1' in the k1 mask are computed and stored in the result; elements in the result vector corresponding to zero bit in k1 are copied from corresponding elements of vector v1_old

Description

Fixes up NaNs from float32 vectors v1 and v2, returning the quietized NaN from v1 as v3.

The masked variant has one additional argument: k1. Only those elements in the source registers with the corresponding bit set in vector mask k1 are used for computing.

Returns

Returns the result of the operation.