Vector Data Types

Built-in vector data types.

Description

The char, unsigned char, short, unsigned short, integer, unsigned integer, long, unsigned long, and float vector data types are supported. The vector data type is defined with the type name i.e. char, uchar, short, ushort, int, uint, float, long, and ulong followed by a literal value n that defines the number of elements in the vector. Supported values of n are 2, 4, 8, and 16.

The built-in vector data types are also declared as appropriate types in the OpenCL API (and header files) that can be used by an application. The following table describes the built-in vector data type in the OpenCL C programming language and the corresponding data type available to the application:

Type in OpenCL Language Description API type for application
charn A 8-bit signed two's complement integer vector. cl_charn
ucharn A 8-bit unsigned integer vector. cl_ucharn
shortn A 16-bit signed two's complement integer vector. cl_shortn
ushortn A 16-bit unsigned integer vector. cl_ushortn
intn A 32-bit signed two's complement integer vector. cl_int
uintn A 32-bit unsigned integer vector. cl_uintn
longn A 64-bit signed two's complement integer vector. cl_longn
ulongn A 64-bit unsigned integer vector. cl_ulongn
floatn A float vector cl_floatn

Notes

Built-in vector data types are supported by the OpenCL implementation even if the underlying compute device does not support any or all of the vector data types. These are to be converted by the device compiler to appropriate instructions that use underlying built-in types supported natively by the compute device.

Optional Double Precision and Half Floating Point

OpenCL 1.0 adds support for double precision and half floating-point as optional extensions. The double data type must confirm to the IEEE-754 double precision storage format.

An application that wants to use double will need to include the #pragma OPENCL EXTENSION cl_khr_fp64 : enable directive before any double precision data type is declared in the kernel code. This will extended the list of built-in vector and scalar data types to include the following:

Type in OpenCL Language Description API type for application
double A double precision float. cl_double
double2 A 2-component double vector. cl_double2
double4 A 4-component double vector. cl_double4
double8 An 8-component double vector. cl_double8
double16 A 16-component double vector. cl_double16

An application that wants to use half and halfn will need to include the #pragma OPENCL EXTENSION cl_khr_fp16 : enable directive. This will extended the list of built-in vector and scalar data types to include the following:

Type in OpenCL Language Description API type for application
half2 A 2-component half-precision floating-point vector. cl_half2
half4 A 4-component half-precision floating-point vector. cl_half4
half8 An 8-component half-precision floating-point vector. cl_half8
half16 A 16-component half-precision floating-point vector. cl_half16

Specification

OpenCL Specification

Also see

Macros and Limits, Scalar Data Types, Reserved Data Types, Other Data Types, Abstract Data Types, cl_khr_fp16, cl_khr_fp64

Copyright © 2007-2009 The Khronos Group Inc. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and/or associated documentation files (the "Materials"), to deal in the Materials without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Materials, and to permit persons to whom the Materials are furnished to do so, subject to the condition that this copyright notice and permission notice shall be included in all copies or substantial portions of the Materials.