| <?xml version="1.0" encoding="UTF-8"?> |
| <!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook MathML Module V1.1b1//EN" |
| "http://www.oasis-open.org/docbook/xml/mathml/1.1CR1/dbmathml.dtd"> |
| <refentry> |
| <refentryinfo> |
| <keywordset> |
| <keyword>Scalar Data Types</keyword> |
| </keywordset> |
| </refentryinfo> |
| |
| <refmeta> |
| <refentrytitle>Scalar Data Types</refentrytitle> |
| |
| <refmiscinfo> |
| <copyright> |
| <year>2007-2010</year> |
| <holder>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.</holder> |
| </copyright> |
| </refmiscinfo> |
| <manvolnum>3</manvolnum> |
| </refmeta> |
| |
| <!-- ================================ SYNOPSIS --> |
| |
| <refnamediv id="Built-in Scalar Data Types"> |
| <refname>Scalar Data Types</refname> |
| |
| <refpurpose> |
| Built-in scalar data types. |
| </refpurpose> |
| </refnamediv> |
| |
| |
| |
| <!-- ================================ DESCRIPTION --> |
| |
| <refsect1 id="description"><title>Description</title> |
| |
| <para> |
| These are the data types available in the OpenCL C programming language used to create kernels that are executed on OpenCL device(s). The OpenCL C programming language (also referred to as OpenCL C) is based on the ISO/IEC 9899:1999 C language specification (a.k.a. C99 specification) with specific extensions and restrictions. Please refer to the ISO/IEC 9899:1999 specification for a detailed description of the language grammar. This section describes modifications and restrictions to ISO/IEC 9899:1999 supported in OpenCL C. |
| </para> |
| <para> |
| Most built-in scalar 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 scalar data type in the OpenCL C programming language and the corresponding data type available to the application: |
| </para> |
| |
| <informaltable frame="all"> |
| <tgroup cols="3" align="left" colsep="1" rowsep="1"> |
| <colspec colname="col1" colnum="1" /> |
| <colspec colname="col2" colnum="2" /> |
| <colspec colname="col3" colnum="3" /> |
| |
| <thead> |
| <row> |
| <entry>Type in OpenCL Language</entry> |
| <entry>Description</entry> |
| <entry>API type for application</entry> |
| </row> |
| </thead> |
| |
| <tbody> |
| |
| <row> |
| <entry> |
| <type>bool</type> |
| </entry> |
| |
| <entry>A conditional data type which is either <varname>true</varname> or <varname>false</varname>. |
| The value <varname>true</varname> expands to the integer constant 1 and the value <varname>false</varname> expands to the integer constant 0. When any scalar value is converted to bool, the result is 0 if the value compares equal to 0; otherwise, the result is 1. |
| </entry> |
| |
| <entry> |
| n/a |
| </entry> |
| </row> |
| |
| <row> |
| <entry> |
| <type>char</type> |
| </entry> |
| |
| <entry> |
| A signed two's complement 8-bit integer. |
| </entry> |
| |
| <entry> |
| <type>cl_char</type> |
| </entry> |
| </row> |
| |
| <row> |
| <entry> |
| <literallayout><type>unsigned char</type>, |
| <type>uchar</type></literallayout> |
| </entry> |
| |
| <entry> |
| An unsigned 8-bit integer. |
| </entry> |
| |
| <entry> |
| <type>cl_uchar</type> |
| </entry> |
| </row> |
| |
| <row> |
| <entry> |
| <type>short</type> |
| </entry> |
| |
| <entry> |
| A signed two's complement 16-bit integer. |
| </entry> |
| |
| <entry> |
| <type>cl_short</type> |
| </entry> |
| </row> |
| |
| <row> |
| <entry> |
| <literallayout><type>unsigned short</type>, |
| <type>ushort</type></literallayout> |
| </entry> |
| |
| <entry> |
| An unsigned 16-bit integer. |
| </entry> |
| |
| <entry> |
| <type>cl_ushort</type> |
| </entry> |
| </row> |
| |
| <row> |
| <entry> |
| <type>int</type> |
| </entry> |
| |
| <entry> |
| A signed two's complement 32-bit integer. |
| </entry> |
| |
| <entry> |
| <type>cl_int</type> |
| </entry> |
| </row> |
| |
| <row> |
| <entry> |
| <literallayout><type>unsigned int</type>, |
| <type>uint</type></literallayout> |
| </entry> |
| |
| <entry> |
| An unsigned 32-bit integer. |
| </entry> |
| |
| <entry> |
| <type>cl_uint</type> |
| </entry> |
| </row> |
| |
| <row> |
| <entry> |
| <type>long</type> |
| </entry> |
| |
| <entry> |
| A signed two's complement 64-bit integer. |
| </entry> |
| |
| <entry> |
| <type>cl_long</type> |
| </entry> |
| </row> |
| |
| <row> |
| <entry> |
| <literallayout><type>unsigned long</type>, |
| <type>ulong</type></literallayout> |
| </entry> |
| |
| <entry> |
| An unsigned 64-bit integer. |
| </entry> |
| |
| <entry> |
| <type>cl_ulong</type> |
| </entry> |
| </row> |
| |
| <row> |
| <entry> |
| <type>float</type> |
| </entry> |
| |
| <entry> |
| A single precision float. The float data type must conform to the |
| IEEE 754 single precision storage format. |
| </entry> |
| |
| <entry> |
| <type>cl_float</type> |
| </entry> |
| </row> |
| |
| <row> |
| <entry> |
| <type>half</type> |
| </entry> |
| |
| <entry> |
| A 16-bit float. The half data type must conform to the IEEE 754-2008 half precision storage format. |
| </entry> |
| |
| <entry> |
| <type>cl_half</type> |
| </entry> |
| </row> |
| |
| <row> |
| <entry> |
| <type>size_t</type> |
| </entry> |
| |
| <entry> |
| The unsigned integer type of the result of the <function>sizeof</function> operator. This is a 32-bit unsigned integer if CL_DEVICE_ADDRESS_BITS |
| defined in <citerefentry><refentrytitle>clGetDeviceInfo</refentrytitle></citerefentry> |
| is 32-bits and is a 64-bit unsigned integer if CL_DEVICE_ADDRESS_BITS is 64-bits. |
| </entry> |
| |
| <entry> |
| n/a |
| </entry> |
| </row> |
| |
| <row> |
| <entry> |
| <type>ptrdiff_t</type> |
| </entry> |
| |
| <entry> |
| A signed integer type that is the result of subtracting two pointers. |
| This is a 32-bit signed integer if CL_DEVICE_ADDRESS_BITS defined in <citerefentry><refentrytitle>clGetDeviceInfo</refentrytitle></citerefentry> is 32-bits and is a 64-bit signed integer if CL_DEVICE_ADDRESS_BITS is 64-bits. |
| </entry> |
| |
| <entry> |
| n/a |
| </entry> |
| </row> |
| |
| <row> |
| <entry> |
| <type>intptr_t</type> |
| </entry> |
| |
| <entry> |
| A signed integer type with the property that any valid pointer to |
| <type>void</type> can be converted to this type, then converted back to pointer |
| to <type>void</type>, and the result will compare equal to the original pointer. |
| </entry> |
| |
| <entry> |
| n/a |
| </entry> |
| </row> |
| |
| <row> |
| <entry> |
| <type>uintptr_t</type> |
| </entry> |
| |
| <entry> |
| An unsigned integer type with the property that any valid pointer to |
| <type>void</type> can be converted to this type, then converted back to pointer |
| to <type>void</type>, and the result will compare equal to the original pointer. |
| </entry> |
| |
| <entry> |
| n/a |
| </entry> |
| </row> |
| |
| <row> |
| <entry> |
| <type>void</type> |
| </entry> |
| |
| <entry> |
| The <type>void</type> type comprises an empty set of values; it is an incomplete |
| type that cannot be completed. |
| </entry> |
| |
| <entry> |
| <type>void</type> |
| </entry> |
| </row> |
| |
| |
| </tbody> |
| </tgroup> |
| </informaltable> |
| </refsect1> |
| |
| <!-- OPTIONAL DOUBLE AND HALF DATA TYPES --> |
| |
| <refsect1 id="optional"><title>Optional Double Precision and Half Floating Point</title> |
| <para> |
| OpenCL 1.0 adds support for double precision and half floating-point as optional extensions. The <type>double</type> data type must confirm to the IEEE-754 double precision storage format. |
| </para> |
| <para> |
| An application that wants to use <type>double</type> will need to include the <citerefentry href="cl_khr_fp64"><refentrytitle>#pragma OPENCL |
| EXTENSION cl_khr_fp64 : enable</refentrytitle></citerefentry> 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: |
| </para> |
| <para> |
| <informaltable frame="all"> |
| <tgroup cols="3" align="left" colsep="1" rowsep="1"> |
| <colspec colname="col1" colnum="1" /> |
| <colspec colname="col2" colnum="2" /> |
| <colspec colname="col3" colnum="3" /> |
| |
| <thead> |
| <row> |
| <entry>Type in OpenCL Language</entry> |
| <entry>Description</entry> |
| <entry>API type for application</entry> |
| </row> |
| </thead> |
| <tbody> |
| <row> |
| <entry><type>double</type></entry> |
| <entry>A double precision float.</entry> |
| <entry><type>cl_double</type></entry> |
| </row> |
| <row> |
| <entry><type>double2</type></entry> |
| <entry>A 2-component double vector.</entry> |
| <entry><type>cl_double2</type></entry> |
| </row> |
| <row> |
| <entry><type>double3</type></entry> |
| <entry>A 3-component double vector.</entry> |
| <entry><type>cl_double3</type></entry> |
| </row> |
| <row> |
| <entry><type>double4</type></entry> |
| <entry>A 4-component double vector.</entry> |
| <entry><type>cl_double4</type></entry> |
| </row> |
| <row> |
| <entry><type>double8</type></entry> |
| <entry>An 8-component double vector.</entry> |
| <entry><type>cl_double8</type></entry> |
| </row> |
| <row> |
| <entry><type>double16</type></entry> |
| <entry>A 16-component double vector.</entry> |
| <entry><type>cl_double16</type></entry> |
| </row> |
| </tbody> |
| </tgroup> |
| </informaltable> |
| </para> |
| |
| |
| <para> |
| An application that wants to use <type>half</type> and <type>half<replaceable>n</replaceable></type> will need to include the <citerefentry href="cl_khr_fp16"><refentrytitle>#pragma OPENCL |
| EXTENSION cl_khr_fp16 : enable</refentrytitle></citerefentry> directive. This will extended the list of built-in vector and scalar data types to include the following: |
| </para> |
| <para> |
| <informaltable frame="all"> |
| <tgroup cols="3" align="left" colsep="1" rowsep="1"> |
| <colspec colname="col1" colnum="1" /> |
| <colspec colname="col2" colnum="2" /> |
| <colspec colname="col3" colnum="3" /> |
| |
| <thead> |
| <row> |
| <entry>Type in OpenCL Language</entry> |
| <entry>Description</entry> |
| <entry>API type for application</entry> |
| </row> |
| </thead> |
| <tbody> |
| <row> |
| <entry><type>half2</type></entry> |
| <entry>A 2-component half-precision floating-point vector.</entry> |
| <entry><type>cl_half2</type></entry> |
| </row> |
| <row> |
| <entry><type>half3</type></entry> |
| <entry>A 3-component half-precision floating-point vector.</entry> |
| <entry><type>cl_half3</type></entry> |
| </row> |
| <row> |
| <entry><type>half4</type></entry> |
| <entry>A 4-component half-precision floating-point vector.</entry> |
| <entry><type>cl_half4</type></entry> |
| </row> |
| <row> |
| <entry><type>half8</type></entry> |
| <entry>An 8-component half-precision floating-point vector.</entry> |
| <entry><type>cl_half8</type></entry> |
| </row> |
| <row> |
| <entry><type>half16</type></entry> |
| <entry>A 16-component half-precision floating-point vector.</entry> |
| <entry><type>cl_half16</type></entry> |
| </row> |
| </tbody> |
| </tgroup> |
| </informaltable> |
| </para> |
| |
| |
| </refsect1> |
| |
| |
| |
| <!-- ================================ SPECIFICATION --> |
| <!-- Set the "uri" attribute in the <olink /> element to the "named destination" for the PDF page |
| --> |
| <refsect1 id="specification"><title>Specification</title> |
| <para> |
| <imageobject> |
| <imagedata fileref="pdficon_small1.gif" format="gif" /> |
| </imageobject> |
| |
| <olink uri="scalarDataTypes">OpenCL Specification</olink> |
| </para> |
| </refsect1> |
| |
| |
| <!-- ================================ ALSO SEE --> |
| |
| <refsect1 id="seealso"><title>Also see</title> |
| <para> |
| <citerefentry href="macroLimits"><refentrytitle>Macros and Limits</refentrytitle></citerefentry>, |
| <citerefentry href="vectorDataTypes"><refentrytitle>Vector Data Types</refentrytitle></citerefentry>, |
| <citerefentry href="reservedDataTypes"><refentrytitle>Reserved Data Types</refentrytitle></citerefentry>, |
| <citerefentry href="otherDataTypes"><refentrytitle>Other Data Types</refentrytitle></citerefentry>, |
| <citerefentry href="abstractDataTypes"><refentrytitle>Abstract Data Types</refentrytitle></citerefentry>, |
| <citerefentry><refentrytitle>cl_khr_fp16</refentrytitle></citerefentry>, |
| <citerefentry><refentrytitle>cl_khr_fp64</refentrytitle></citerefentry> |
| </para> |
| </refsect1> |
| |
| |
| |
| <!-- ============================== COPYRIGHT --> |
| <!-- Content included from copyright.inc.xsl --> |
| |
| <refsect3 id="Copyright"><title></title> |
| <imageobject> |
| <imagedata fileref="KhronosLogo.jpg" format="jpg" /> |
| </imageobject> |
| <para /> |
| </refsect3> |
| |
| </refentry> |
| |
