_mm256_cvtps_ph()

Converts eight single-precision floating point values to half-precision (16-bit) floating point values.

Syntax

extern __m128i _mm_cvtps_ph(__m256 x, int imm);

Arguments

X

a vector containing eight single-precision FP values

Imm

a conversion control constant

Description

This intrinsic converts eight single-precision floating point values to half-precision (16-bit) floating point values. The corresponding Intel® Post 32nm processor extension instruction is VCVTPS2PH.

Parent topic: Intrinsics for Converting Half Floats that Map to Intel(R) Post-32nm Processor Instructions

Submit feedback on this help topic

Copyright © 1996-2011, Intel Corporation. All rights reserved.