Intel® C++ Compiler Classic Developer Guide and Reference

ID 767249
Date 12/16/2022
Public

A newer version of this document is available. Customers should click here to go to the newest version.

Document Table of Contents

Intrinsics for Short Vector Math Library Operations (SVML)

The compiler provides short vector math library (SVML) intrinsics to compute vector math functions. These intrinsics are available for IA-32 and Intel® 64 architectures running on supported operating systems. The prototypes for the SVML intrinsics are available in the immintrin.h file.

To use these intrinsics, include the immintrin.h file as follows:

#include <immintrin.h>

The SVML intrinsics do not have any corresponding instructions.

The SVML intrinsics are vector variants of corresponding scalar math operations using __m128, __m128d, __m256, __m256d, and __m256i data types. They take packed vector arguments, perform the operation on each element of the packed vector argument, and return a packed vector result.

For example, the argument to the _mm_sin_ps intrinsic is a packed 128-bit vector of four 32-bit precision floating point numbers. The intrinsic computes the sine of each of these four numbers and returns the four results in a packed 128-bit vector.

Using SVML intrinsics is faster than repeatedly calling the scalar math functions. However, the intrinsics differ from the scalar functions in accuracy.

NOTE:

Many routines in the Short Vector Math Library (SVML) are more optimized for Intel® microprocessors than for non-Intel microprocessors.

This section also includes information about 512-bit intrinsics for SVML.