freepascal.compiler/docs/mmx.xml

<?xml version="1.0" encoding="ISO8859-1"?>
<fpdoc-descriptions>
<!--  

   $Id$ 
   This file is part of the FPC documentation. 
   Copyright (C) 1997, by Michael Van Canneyt 
 
   The FPC documentation is free text; you can redistribute it and/or 
   modify it under the terms of the GNU Library General Public License as 
   published by the Free Software Foundation; either version 2 of the 
   License, or (at your option) any later version. 
 
   The FPC Documentation is distributed in the hope that it will be useful, 
   but WITHOUT ANY WARRANTY; without even the implied warranty of 
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU 
   Library General Public License for more details. 
 
   You should have received a copy of the GNU Library General Public 
   License along with the FPC documentation; see the file COPYING.LIB.  If not, 
   write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330, 
   Boston, MA 02111-1307, USA. 
 -->
<package name="rtl">
<module name="mmx">
<short>Access to MMX processor functionality</short>
<descr>
This document describes the <file>MMX</file> unit. This unit allows you to use the
<var>MMX</var> capabilities of the Free Pascal compiler. It was written by Florian
Klaempfl for the <var>I386</var> processor. It should work on all platforms that
use the Intel processor.
</descr>

<element name="tmmxshortint">
<short>Array of shortints, 64 bits in size</short>
</element>

<element name="tmmxbyte">
<short>Array of bytes, 64 bits in size</short>
</element>

<element name="tmmxword">
<short>Array of words, 64 bits in size</short>
</element>

<element name="tmmxinteger">
<short>Array of integers, 64 bits in size</short>
</element>

<element name="tmmxfixed">
<short>Array of Fixed16, 64 bits in size</short>
</element>

<element name="tmmxlongint">
<short>Array of longint, 64 bits in size</short>
</element>

<element name="tmmxcardinal">
<short>Array of cardinals, 64 bits in size</short>
</element>

<element name="tmmxsingle">
<short>Array of singles, 64 bits in size</short>
</element>

<element name="pmmxshortint">
<short>Pointer to <link id="tmmxshortint"/> array type</short>
</element>

<element name="pmmxbyte">
<short>Pointer to <link id="tmmxbyte"/> array type</short>
</element>

<element name="pmmxword">
<short>Pointer to <link id="tmmxword"/> array type</short>
</element>

<element name="pmmxinteger">
<short>Pointer to <link id="tmmxinteger"/> array type</short>
</element>

<element name="pmmxfixed">
<short>Pointer to <link id="tmmxfixed"/> array type</short>
</element>

<element name="pmmxlongint">
<short>Pointer to <link id="tmmxlongint"/> array type</short>
</element>

<element name="pmmxcardinal">
<short>Pointer to <link id="tmmxcardinal"/> array type</short>
</element>
<element name="pmmxsingle">
<short>Pointer to <link id="tmmxsingle"/> array type</short>
</element>

<element name="is_mmx_cpu">
<short>Is the current processor equipped with a MMX capable CPU</short>
<descr>
The <var>is_mmx_cpu</var> initialized constant allows you to determine 
if the computer has <var>MMX</var> extensions. It is set correctly in 
the unit's initialization code.
</descr>
</element>

<element name="is_amd_3d_cpu">
<short>Is the current processor equipped with a AMD 3D  CPU</short>
<descr>
The <var>is_amd_3d_cpu</var> initialized constant allows you to determine 
if the computer has the AMD 3D extensions. It is set correctly in 
the unit's initialization code.
</descr>
</element>

<element name="is_amd_3d_mmx_cpu">
<short>Is the current processor equipped with a AMD 3D MMX CPU</short>
<descr>
The <var>is_amd_3d_mmx_cpu</var> initialized constant allows you to determine 
if the computer has the AMD 3D MMX extensions. It is set correctly in 
the unit's initialization code.
</descr>
</element>

<element name="is_amd_3d_dsp_cpu">
<short>Is the current processor equipped with a AMD 3D DSP CPU</short>
<descr>
The <var>is_amd_3d_dsp_cpu</var> initialized constant allows you to determine 
if the computer has the AMD 3D DSP extensions. It is set correctly in 
the unit's initialization code.
</descr>
</element>

<element name="is_sse_cpu">
<short>Is the current processor equipped with a SSE CPU</short>
<descr>
The <var>is_sse_cpu</var> initialized constant allows you to determine 
if the computer has the SSE extensions. It is set correctly in 
the unit's initialization code.
</descr>
</element>



<element name="is_sse2_cpu">
<short>Is the current processor equipped with a SSE2 CPU</short>
<descr>
The <var>is_sse2_cpu</var> initialized constant allows you to determine 
if the computer has the SSE2 extensions. It is set correctly in 
the unit's initialization code.
</descr>
</element>

<element name="Emms">
<short>Reset floating point registers</short>
<descr>
<p>
<var>Emms</var> sets all floating point registers to empty. This procedure must
be called after you have used any <var>MMX</var> instructions, if you want to use
floating point arithmetic. If you just want to move floating point data
around, it isn't necessary to call this function, the compiler doesn't use
the FPU registers when moving data. Only when doing calculations, you should
use this function. The following code demonstrates this:
</p>
<code>
Program MMXDemo;
uses mmx;
var
   d1 : double;
   a : array[0..10000] of double;
   i : longint;
begin
   d1:=1.0;
{$mmx+}
   { floating point data is used, but we do _no_ arithmetic }
   for i:=0 to 10000 do
     a[i]:=d2;  { this is done with 64 bit moves }
{$mmx-}
   emms;   { clear fpu }
   { now we can do floating point arithmetic again }
end. 
</code>
</descr>
<seealso>
<link id="femms"/>
</seealso>
</element>

<element name="femms">
<short>Reset floating point registers - AMD version</short>
<descr>
<var>femms</var> executes the <var>femms</var> assembler instruction for AMD
processors. it is not supported by all assemblers, hence it is coded as byte
codes.
</descr>
<seealso>
<link id="emms"/>
</seealso>
</element>

</module>
</package>
</fpdoc-descriptions>