Developer Guide and Reference

Contents

_mm256_cvttps_epi32

Converts packed float32 values to truncated extended 32-bit integer values. The corresponding Intel® AVX instruction is
VCVTTPS2DQ
.

Syntax

extern __m256i _mm256_cvttps_epi32(__m256 m1);
Arguments
m1
float32 source vector
Description
Converts eight packed single-precision floating-point values (float32 values) in the source vector to eight packed signed doubleword integer (extended 32-bit integer) values in the destination.
When a conversion is inexact, a truncated (round towards zero) value is returned. If a converted result is larger than the maximum signed doubleword integer, the floating-point invalid exception is raised; if this exception is masked, the indefinite integer value (
80000000H
) is returned.
Returns
Result of the conversion operation.

Product and Performance Information