<html lang="en">
<head>
<title>MIPS-3D Built-in Functions - Using the GNU Compiler Collection (GCC)</title>
<meta http-equiv="Content-Type" content="text/html">
<meta name="description" content="Using the GNU Compiler Collection (GCC)">
<meta name="generator" content="makeinfo 4.13">
<link title="Top" rel="start" href="index.html#Top">
<link rel="up" href="MIPS-Loongson-Built_002din-Functions.html#MIPS-Loongson-Built_002din-Functions" title="MIPS Loongson Built-in Functions">
<link rel="prev" href="Paired_002dSingle-Built_002din-Functions.html#Paired_002dSingle-Built_002din-Functions" title="Paired-Single Built-in Functions">
<link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
<!--
Copyright (C) 1988, 1989, 1992, 1993, 1994, 1995, 1996, 1997, 1998,
1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007,
2008 Free Software Foundation, Inc.

Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License, Version 1.2 or
any later version published by the Free Software Foundation; with the
Invariant Sections being ``Funding Free Software'', the Front-Cover
Texts being (a) (see below), and with the Back-Cover Texts being (b)
(see below).  A copy of the license is included in the section entitled
``GNU Free Documentation License''.

(a) The FSF's Front-Cover Text is:

     A GNU Manual

(b) The FSF's Back-Cover Text is:

     You have freedom to copy and modify this GNU Manual, like GNU
     software.  Copies published by the Free Software Foundation raise
     funds for GNU development.-->
<meta http-equiv="Content-Style-Type" content="text/css">
<style type="text/css"><!--
  pre.display { font-family:inherit }
  pre.format  { font-family:inherit }
  pre.smalldisplay { font-family:inherit; font-size:smaller }
  pre.smallformat  { font-family:inherit; font-size:smaller }
  pre.smallexample { font-size:smaller }
  pre.smalllisp    { font-size:smaller }
  span.sc    { font-variant:small-caps }
  span.roman { font-family:serif; font-weight:normal; } 
  span.sansserif { font-family:sans-serif; font-weight:normal; } 
--></style>
<link rel="stylesheet" type="text/css" href="../cs.css">
</head>
<body>
<div class="node">
<a name="MIPS-3D-Built-in-Functions"></a>
<a name="MIPS_002d3D-Built_002din-Functions"></a>
<p>
Previous:&nbsp;<a rel="previous" accesskey="p" href="Paired_002dSingle-Built_002din-Functions.html#Paired_002dSingle-Built_002din-Functions">Paired-Single Built-in Functions</a>,
Up:&nbsp;<a rel="up" accesskey="u" href="MIPS-Loongson-Built_002din-Functions.html#MIPS-Loongson-Built_002din-Functions">MIPS Loongson Built-in Functions</a>
<hr>
</div>

<h5 class="subsubsection">6.52.9.3 MIPS-3D Built-in Functions</h5>

<p>The MIPS-3D Application-Specific Extension (ASE) includes additional
paired-single instructions that are designed to improve the performance
of 3D graphics operations.  Support for these instructions is controlled
by the <samp><span class="option">-mips3d</span></samp> command-line option.

 <p>The functions listed below map directly to a particular MIPS-3D
instruction.  Please refer to the architecture specification for
more details on what each instruction does.

     <dl>
<dt><code>v2sf __builtin_mips_addr_ps (v2sf, v2sf)</code><dd>Reduction add (<code>addr.ps</code>).

     <br><dt><code>v2sf __builtin_mips_mulr_ps (v2sf, v2sf)</code><dd>Reduction multiply (<code>mulr.ps</code>).

     <br><dt><code>v2sf __builtin_mips_cvt_pw_ps (v2sf)</code><dd>Convert paired single to paired word (<code>cvt.pw.ps</code>).

     <br><dt><code>v2sf __builtin_mips_cvt_ps_pw (v2sf)</code><dd>Convert paired word to paired single (<code>cvt.ps.pw</code>).

     <br><dt><code>float __builtin_mips_recip1_s (float)</code><dt><code>double __builtin_mips_recip1_d (double)</code><dt><code>v2sf __builtin_mips_recip1_ps (v2sf)</code><dd>Reduced precision reciprocal (sequence step 1) (<code>recip1.</code><var>fmt</var>).

     <br><dt><code>float __builtin_mips_recip2_s (float, float)</code><dt><code>double __builtin_mips_recip2_d (double, double)</code><dt><code>v2sf __builtin_mips_recip2_ps (v2sf, v2sf)</code><dd>Reduced precision reciprocal (sequence step 2) (<code>recip2.</code><var>fmt</var>).

     <br><dt><code>float __builtin_mips_rsqrt1_s (float)</code><dt><code>double __builtin_mips_rsqrt1_d (double)</code><dt><code>v2sf __builtin_mips_rsqrt1_ps (v2sf)</code><dd>Reduced precision reciprocal square root (sequence step 1)
(<code>rsqrt1.</code><var>fmt</var>).

     <br><dt><code>float __builtin_mips_rsqrt2_s (float, float)</code><dt><code>double __builtin_mips_rsqrt2_d (double, double)</code><dt><code>v2sf __builtin_mips_rsqrt2_ps (v2sf, v2sf)</code><dd>Reduced precision reciprocal square root (sequence step 2)
(<code>rsqrt2.</code><var>fmt</var>). 
</dl>

 <p>The following multi-instruction functions are also available. 
In each case, <var>cond</var> can be any of the 16 floating-point conditions:
<code>f</code>, <code>un</code>, <code>eq</code>, <code>ueq</code>, <code>olt</code>, <code>ult</code>,
<code>ole</code>, <code>ule</code>, <code>sf</code>, <code>ngle</code>, <code>seq</code>,
<code>ngl</code>, <code>lt</code>, <code>nge</code>, <code>le</code> or <code>ngt</code>.

     <dl>
<dt><code>int __builtin_mips_cabs_</code><var>cond</var><code>_s (float </code><var>a</var><code>, float </code><var>b</var><code>)</code><dt><code>int __builtin_mips_cabs_</code><var>cond</var><code>_d (double </code><var>a</var><code>, double </code><var>b</var><code>)</code><dd>Absolute comparison of two scalar values (<code>cabs.</code><var>cond</var><code>.</code><var>fmt</var>,
<code>bc1t</code>/<code>bc1f</code>).

     <p>These functions compare <var>a</var> and <var>b</var> using <code>cabs.</code><var>cond</var><code>.s</code>
or <code>cabs.</code><var>cond</var><code>.d</code> and return the result as a boolean value. 
For example:

     <pre class="smallexample">          float a, b;
          if (__builtin_mips_cabs_eq_s (a, b))
            true ();
          else
            false ();
</pre>
     <br><dt><code>int __builtin_mips_upper_cabs_</code><var>cond</var><code>_ps (v2sf </code><var>a</var><code>, v2sf </code><var>b</var><code>)</code><dt><code>int __builtin_mips_lower_cabs_</code><var>cond</var><code>_ps (v2sf </code><var>a</var><code>, v2sf </code><var>b</var><code>)</code><dd>Absolute comparison of two paired-single values (<code>cabs.</code><var>cond</var><code>.ps</code>,
<code>bc1t</code>/<code>bc1f</code>).

     <p>These functions compare <var>a</var> and <var>b</var> using <code>cabs.</code><var>cond</var><code>.ps</code>
and return either the upper or lower half of the result.  For example:

     <pre class="smallexample">          v2sf a, b;
          if (__builtin_mips_upper_cabs_eq_ps (a, b))
            upper_halves_are_equal ();
          else
            upper_halves_are_unequal ();
          
          if (__builtin_mips_lower_cabs_eq_ps (a, b))
            lower_halves_are_equal ();
          else
            lower_halves_are_unequal ();
</pre>
     <br><dt><code>v2sf __builtin_mips_movt_cabs_</code><var>cond</var><code>_ps (v2sf </code><var>a</var><code>, v2sf </code><var>b</var><code>, v2sf </code><var>c</var><code>, v2sf </code><var>d</var><code>)</code><dt><code>v2sf __builtin_mips_movf_cabs_</code><var>cond</var><code>_ps (v2sf </code><var>a</var><code>, v2sf </code><var>b</var><code>, v2sf </code><var>c</var><code>, v2sf </code><var>d</var><code>)</code><dd>Conditional move based on absolute comparison (<code>cabs.</code><var>cond</var><code>.ps</code>,
<code>movt.ps</code>/<code>movf.ps</code>).

     <p>The <code>movt</code> functions return the value <var>x</var> computed by:

     <pre class="smallexample">          cabs.<var>cond</var>.ps <var>cc</var>,<var>a</var>,<var>b</var>
          mov.ps <var>x</var>,<var>c</var>
          movt.ps <var>x</var>,<var>d</var>,<var>cc</var>
</pre>
     <p>The <code>movf</code> functions are similar but use <code>movf.ps</code> instead
of <code>movt.ps</code>.

     <br><dt><code>int __builtin_mips_any_c_</code><var>cond</var><code>_ps (v2sf </code><var>a</var><code>, v2sf </code><var>b</var><code>)</code><dt><code>int __builtin_mips_all_c_</code><var>cond</var><code>_ps (v2sf </code><var>a</var><code>, v2sf </code><var>b</var><code>)</code><dt><code>int __builtin_mips_any_cabs_</code><var>cond</var><code>_ps (v2sf </code><var>a</var><code>, v2sf </code><var>b</var><code>)</code><dt><code>int __builtin_mips_all_cabs_</code><var>cond</var><code>_ps (v2sf </code><var>a</var><code>, v2sf </code><var>b</var><code>)</code><dd>Comparison of two paired-single values
(<code>c.</code><var>cond</var><code>.ps</code>/<code>cabs.</code><var>cond</var><code>.ps</code>,
<code>bc1any2t</code>/<code>bc1any2f</code>).

     <p>These functions compare <var>a</var> and <var>b</var> using <code>c.</code><var>cond</var><code>.ps</code>
or <code>cabs.</code><var>cond</var><code>.ps</code>.  The <code>any</code> forms return true if either
result is true and the <code>all</code> forms return true if both results are true. 
For example:

     <pre class="smallexample">          v2sf a, b;
          if (__builtin_mips_any_c_eq_ps (a, b))
            one_is_true ();
          else
            both_are_false ();
          
          if (__builtin_mips_all_c_eq_ps (a, b))
            both_are_true ();
          else
            one_is_false ();
</pre>
     <br><dt><code>int __builtin_mips_any_c_</code><var>cond</var><code>_4s (v2sf </code><var>a</var><code>, v2sf </code><var>b</var><code>, v2sf </code><var>c</var><code>, v2sf </code><var>d</var><code>)</code><dt><code>int __builtin_mips_all_c_</code><var>cond</var><code>_4s (v2sf </code><var>a</var><code>, v2sf </code><var>b</var><code>, v2sf </code><var>c</var><code>, v2sf </code><var>d</var><code>)</code><dt><code>int __builtin_mips_any_cabs_</code><var>cond</var><code>_4s (v2sf </code><var>a</var><code>, v2sf </code><var>b</var><code>, v2sf </code><var>c</var><code>, v2sf </code><var>d</var><code>)</code><dt><code>int __builtin_mips_all_cabs_</code><var>cond</var><code>_4s (v2sf </code><var>a</var><code>, v2sf </code><var>b</var><code>, v2sf </code><var>c</var><code>, v2sf </code><var>d</var><code>)</code><dd>Comparison of four paired-single values
(<code>c.</code><var>cond</var><code>.ps</code>/<code>cabs.</code><var>cond</var><code>.ps</code>,
<code>bc1any4t</code>/<code>bc1any4f</code>).

     <p>These functions use <code>c.</code><var>cond</var><code>.ps</code> or <code>cabs.</code><var>cond</var><code>.ps</code>
to compare <var>a</var> with <var>b</var> and to compare <var>c</var> with <var>d</var>. 
The <code>any</code> forms return true if any of the four results are true
and the <code>all</code> forms return true if all four results are true. 
For example:

     <pre class="smallexample">          v2sf a, b, c, d;
          if (__builtin_mips_any_c_eq_4s (a, b, c, d))
            some_are_true ();
          else
            all_are_false ();
          
          if (__builtin_mips_all_c_eq_4s (a, b, c, d))
            all_are_true ();
          else
            some_are_false ();
</pre>
     </dl>

 </body></html>