CUTLASS 2.0 (#62)

CUTLASS 2.0

Substantially refactored for

- Better performance, particularly for native Turing Tensor Cores
- Robust and durable templates spanning the design space
- Encapsulated functionality embodying modern C++11 programming techniques
- Optimized containers and data types for efficient, generic, portable device code

Updates to:
- Quick start guide
- Documentation
- Utilities
- CUTLASS Profiler

Native Turing Tensor Cores
- Efficient GEMM kernels targeting Turing Tensor Cores
- Mixed-precision floating point, 8-bit integer, 4-bit integer, and binarized operands

Coverage of existing CUTLASS functionality:
- GEMM kernels targeting CUDA and Tensor Cores in NVIDIA GPUs
- Volta Tensor Cores through native mma.sync and through WMMA API
- Optimizations such as parallel reductions, threadblock rasterization, and intra-threadblock reductions
- Batched GEMM operations
- Complex-valued GEMMs

Note: this commit and all that follow require a host compiler supporting C++11 or greater.
This commit is contained in:
Andrew Kerr
2019-11-19 16:55:34 -08:00
committed by GitHub
parent b5cab177a9
commit fb335f6a5f
5434 changed files with 599799 additions and 250176 deletions
@@ -3,21 +3,23 @@
<head>
<meta http-equiv="Content-Type" content="text/xhtml;charset=UTF-8"/>
<meta http-equiv="X-UA-Compatible" content="IE=9"/>
<meta name="generator" content="Doxygen 1.8.14"/>
<meta name="viewport" content="width=device-width, initial-scale=1"/>
<title>Cutlass: cutlass::platform::unique_ptr&lt; T, Deleter &gt; Class Template Reference</title>
<meta name="generator" content="Doxygen 1.8.11"/>
<title>CUTLASS: cutlass::platform::unique_ptr&lt; T, Deleter &gt; Class Template Reference</title>
<link href="tabs.css" rel="stylesheet" type="text/css"/>
<script type="text/javascript" src="jquery.js"></script>
<script type="text/javascript" src="dynsections.js"></script>
<link href="search/search.css" rel="stylesheet" type="text/css"/>
<script type="text/javascript" src="search/searchdata.js"></script>
<script type="text/javascript" src="search/search.js"></script>
<script type="text/javascript">
$(document).ready(function() { init_search(); });
</script>
<script type="text/x-mathjax-config">
MathJax.Hub.Config({
extensions: ["tex2jax.js"],
jax: ["input/TeX","output/HTML-CSS"],
});
</script><script type="text/javascript" async src="http://cdn.mathjax.org/mathjax/latest/MathJax.js"></script>
</script><script type="text/javascript" src="http://cdn.mathjax.org/mathjax/latest/MathJax.js"></script>
<link href="doxygen.css" rel="stylesheet" type="text/css" />
</head>
<body>
@@ -26,8 +28,9 @@
<table cellspacing="0" cellpadding="0">
<tbody>
<tr style="height: 56px;">
<td id="projectlogo"><img alt="Logo" src="cutlass-logo-small.png"/></td>
<td id="projectalign" style="padding-left: 0.5em;">
<div id="projectname">Cutlass
<div id="projectname">CUTLASS
</div>
<div id="projectbrief">CUDA Templates for Linear Algebra Subroutines and Solvers</div>
</td>
@@ -36,22 +39,43 @@
</table>
</div>
<!-- end header part -->
<!-- Generated by Doxygen 1.8.14 -->
<!-- Generated by Doxygen 1.8.11 -->
<script type="text/javascript">
/* @license magnet:?xt=urn:btih:cf05388f2679ee054f2beb29a391d25f4e673ac3&amp;dn=gpl-2.0.txt GPL-v2 */
var searchBox = new SearchBox("searchBox", "search",false,'Search');
/* @license-end */
</script>
<script type="text/javascript" src="menudata.js"></script>
<script type="text/javascript" src="menu.js"></script>
<script type="text/javascript">
/* @license magnet:?xt=urn:btih:cf05388f2679ee054f2beb29a391d25f4e673ac3&amp;dn=gpl-2.0.txt GPL-v2 */
$(function() {
initMenu('',true,false,'search.php','Search');
$(document).ready(function() { init_search(); });
});
/* @license-end */</script>
<div id="main-nav"></div>
<div id="navrow1" class="tabs">
<ul class="tablist">
<li><a href="index.html"><span>Main&#160;Page</span></a></li>
<li><a href="modules.html"><span>Modules</span></a></li>
<li><a href="namespaces.html"><span>Namespaces</span></a></li>
<li class="current"><a href="annotated.html"><span>Classes</span></a></li>
<li><a href="files.html"><span>Files</span></a></li>
<li>
<div id="MSearchBox" class="MSearchBoxInactive">
<span class="left">
<img id="MSearchSelect" src="search/mag_sel.png"
onmouseover="return searchBox.OnSearchSelectShow()"
onmouseout="return searchBox.OnSearchSelectHide()"
alt=""/>
<input type="text" id="MSearchField" value="Search" accesskey="S"
onfocus="searchBox.OnSearchFieldFocus(true)"
onblur="searchBox.OnSearchFieldFocus(false)"
onkeyup="searchBox.OnSearchFieldChange(event)"/>
</span><span class="right">
<a id="MSearchClose" href="javascript:searchBox.CloseResultsWindow()"><img id="MSearchCloseImg" border="0" src="search/close.png" alt=""/></a>
</span>
</div>
</li>
</ul>
</div>
<div id="navrow2" class="tabs2">
<ul class="tablist">
<li><a href="annotated.html"><span>Class&#160;List</span></a></li>
<li><a href="classes.html"><span>Class&#160;Index</span></a></li>
<li><a href="inherits.html"><span>Class&#160;Hierarchy</span></a></li>
<li><a href="functions.html"><span>Class&#160;Members</span></a></li>
</ul>
</div>
<!-- window showing the filter options -->
<div id="MSearchSelectWindow"
onmouseover="return searchBox.OnSearchSelectShow()"
@@ -118,26 +142,24 @@ Public Member Functions</h2></td></tr>
<tr class="memitem:a5b8d8ecafb4da336acd50e40cd42b6e0"><td class="memItemLeft" align="right" valign="top">Deleter &amp;&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="classcutlass_1_1platform_1_1unique__ptr.html#a5b8d8ecafb4da336acd50e40cd42b6e0">get_deleter</a> () <a class="el" href="platform_8h.html#a189faadd7f99f6c354db09acbb2aafcd">noexcept</a></td></tr>
<tr class="memdesc:a5b8d8ecafb4da336acd50e40cd42b6e0"><td class="mdescLeft">&#160;</td><td class="mdescRight">Returns the deleter object. <a href="#a5b8d8ecafb4da336acd50e40cd42b6e0">More...</a><br /></td></tr>
<tr class="separator:a5b8d8ecafb4da336acd50e40cd42b6e0"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:aa427ab4ea4f2336ac6db28d53a4c11ac"><td class="memItemLeft" align="right" valign="top">Deleter const &amp;&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="classcutlass_1_1platform_1_1unique__ptr.html#aa427ab4ea4f2336ac6db28d53a4c11ac">get_deleter</a> () const <a class="el" href="platform_8h.html#a189faadd7f99f6c354db09acbb2aafcd">noexcept</a></td></tr>
<tr class="memitem:aa427ab4ea4f2336ac6db28d53a4c11ac"><td class="memItemLeft" align="right" valign="top">Deleter const &amp;&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="classcutlass_1_1platform_1_1unique__ptr.html#aa427ab4ea4f2336ac6db28d53a4c11ac">get_deleter</a> () const <a class="el" href="platform_8h.html#a189faadd7f99f6c354db09acbb2aafcd">noexcept</a></td></tr>
<tr class="memdesc:aa427ab4ea4f2336ac6db28d53a4c11ac"><td class="mdescLeft">&#160;</td><td class="mdescRight">Returns the deleter object. <a href="#aa427ab4ea4f2336ac6db28d53a4c11ac">More...</a><br /></td></tr>
<tr class="separator:aa427ab4ea4f2336ac6db28d53a4c11ac"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a5791650488ae864f10ad04bec4a31005"><td class="memItemLeft" align="right" valign="top">&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="classcutlass_1_1platform_1_1unique__ptr.html#a5791650488ae864f10ad04bec4a31005">operator bool</a> () const <a class="el" href="platform_8h.html#a189faadd7f99f6c354db09acbb2aafcd">noexcept</a></td></tr>
<tr class="memdesc:a5791650488ae864f10ad04bec4a31005"><td class="mdescLeft">&#160;</td><td class="mdescRight">Checks whether an object is owned. <a href="#a5791650488ae864f10ad04bec4a31005">More...</a><br /></td></tr>
<tr class="separator:a5791650488ae864f10ad04bec4a31005"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a45a3cb6d8641a6130991d56e84cbb38b"><td class="memItemLeft" align="right" valign="top">T &amp;&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="classcutlass_1_1platform_1_1unique__ptr.html#a45a3cb6d8641a6130991d56e84cbb38b">operator*</a> () const</td></tr>
<tr class="memdesc:a45a3cb6d8641a6130991d56e84cbb38b"><td class="mdescLeft">&#160;</td><td class="mdescRight">Dereferences the <a class="el" href="classcutlass_1_1platform_1_1unique__ptr.html" title="std::unique_ptr ">unique_ptr</a>. <a href="#a45a3cb6d8641a6130991d56e84cbb38b">More...</a><br /></td></tr>
<tr class="separator:a45a3cb6d8641a6130991d56e84cbb38b"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a0bd8240db28f3964714da5113cd9e354"><td class="memItemLeft" align="right" valign="top">T &amp;&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="classcutlass_1_1platform_1_1unique__ptr.html#a0bd8240db28f3964714da5113cd9e354">operator*</a> () const </td></tr>
<tr class="memdesc:a0bd8240db28f3964714da5113cd9e354"><td class="mdescLeft">&#160;</td><td class="mdescRight">Dereferences the <a class="el" href="classcutlass_1_1platform_1_1unique__ptr.html" title="std::unique_ptr ">unique_ptr</a>. <a href="#a0bd8240db28f3964714da5113cd9e354">More...</a><br /></td></tr>
<tr class="separator:a0bd8240db28f3964714da5113cd9e354"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:afa52edcaef23461ce1f9c1dac349c24b"><td class="memItemLeft" align="right" valign="top"><a class="el" href="classcutlass_1_1platform_1_1unique__ptr.html#ab6ce60d03d11b269c1e151dfa7c696f9">pointer</a>&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="classcutlass_1_1platform_1_1unique__ptr.html#afa52edcaef23461ce1f9c1dac349c24b">operator-&gt;</a> () const <a class="el" href="platform_8h.html#a189faadd7f99f6c354db09acbb2aafcd">noexcept</a></td></tr>
<tr class="memdesc:afa52edcaef23461ce1f9c1dac349c24b"><td class="mdescLeft">&#160;</td><td class="mdescRight">Returns a pointer to the managed object. <a href="#afa52edcaef23461ce1f9c1dac349c24b">More...</a><br /></td></tr>
<tr class="separator:afa52edcaef23461ce1f9c1dac349c24b"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a5c7a204af07a7d325b0a8303e199a50d"><td class="memItemLeft" align="right" valign="top">T &amp;&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="classcutlass_1_1platform_1_1unique__ptr.html#a5c7a204af07a7d325b0a8303e199a50d">operator[]</a> (size_t i) const</td></tr>
<tr class="memdesc:a5c7a204af07a7d325b0a8303e199a50d"><td class="mdescLeft">&#160;</td><td class="mdescRight">Array access to managed object. <a href="#a5c7a204af07a7d325b0a8303e199a50d">More...</a><br /></td></tr>
<tr class="separator:a5c7a204af07a7d325b0a8303e199a50d"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a353b89f40cb21e0cfd2af2094d417043"><td class="memItemLeft" align="right" valign="top">T &amp;&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="classcutlass_1_1platform_1_1unique__ptr.html#a353b89f40cb21e0cfd2af2094d417043">operator[]</a> (size_t i) const </td></tr>
<tr class="memdesc:a353b89f40cb21e0cfd2af2094d417043"><td class="mdescLeft">&#160;</td><td class="mdescRight">Array access to managed object. <a href="#a353b89f40cb21e0cfd2af2094d417043">More...</a><br /></td></tr>
<tr class="separator:a353b89f40cb21e0cfd2af2094d417043"><td class="memSeparator" colspan="2">&#160;</td></tr>
</table>
<h2 class="groupheader">Member Typedef Documentation</h2>
<a id="a85cab9945c36dc56bd7d6adf30c0d252"></a>
<h2 class="memtitle"><span class="permalink"><a href="#a85cab9945c36dc56bd7d6adf30c0d252">&#9670;&nbsp;</a></span>deleter_type</h2>
<a class="anchor" id="a85cab9945c36dc56bd7d6adf30c0d252"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
@@ -151,9 +173,7 @@ template&lt;class T, class Deleter = default_delete&lt;T&gt;&gt; </div>
</div>
</div>
<a id="a94cea0ebf2ac4bec69dfa1f80ea07d50"></a>
<h2 class="memtitle"><span class="permalink"><a href="#a94cea0ebf2ac4bec69dfa1f80ea07d50">&#9670;&nbsp;</a></span>element_type</h2>
<a class="anchor" id="a94cea0ebf2ac4bec69dfa1f80ea07d50"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
@@ -167,9 +187,7 @@ template&lt;class T, class Deleter = default_delete&lt;T&gt;&gt; </div>
</div>
</div>
<a id="ab6ce60d03d11b269c1e151dfa7c696f9"></a>
<h2 class="memtitle"><span class="permalink"><a href="#ab6ce60d03d11b269c1e151dfa7c696f9">&#9670;&nbsp;</a></span>pointer</h2>
<a class="anchor" id="ab6ce60d03d11b269c1e151dfa7c696f9"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
@@ -184,9 +202,7 @@ template&lt;class T, class Deleter = default_delete&lt;T&gt;&gt; </div>
</div>
</div>
<h2 class="groupheader">Constructor &amp; Destructor Documentation</h2>
<a id="aa8a370bc7e4c2d99eb85e7fea27b3179"></a>
<h2 class="memtitle"><span class="permalink"><a href="#aa8a370bc7e4c2d99eb85e7fea27b3179">&#9670;&nbsp;</a></span>unique_ptr() <span class="overload">[1/2]</span></h2>
<a class="anchor" id="aa8a370bc7e4c2d99eb85e7fea27b3179"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
@@ -211,9 +227,7 @@ template&lt;class T, class Deleter = default_delete&lt;T&gt;&gt; </div>
</div>
</div>
<a id="a14c8bf5a5deefe4a6602ccd5c5af364c"></a>
<h2 class="memtitle"><span class="permalink"><a href="#a14c8bf5a5deefe4a6602ccd5c5af364c">&#9670;&nbsp;</a></span>unique_ptr() <span class="overload">[2/2]</span></h2>
<a class="anchor" id="a14c8bf5a5deefe4a6602ccd5c5af364c"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
@@ -239,9 +253,7 @@ template&lt;class T, class Deleter = default_delete&lt;T&gt;&gt; </div>
</div>
</div>
<a id="a8902399dac4ab64f08f909f2ad9d4bcf"></a>
<h2 class="memtitle"><span class="permalink"><a href="#a8902399dac4ab64f08f909f2ad9d4bcf">&#9670;&nbsp;</a></span>~unique_ptr()</h2>
<a class="anchor" id="a8902399dac4ab64f08f909f2ad9d4bcf"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
@@ -267,9 +279,7 @@ template&lt;class T, class Deleter = default_delete&lt;T&gt;&gt; </div>
</div>
</div>
<h2 class="groupheader">Member Function Documentation</h2>
<a id="a2e7c14b8a118f81c1df46ea5045e297b"></a>
<h2 class="memtitle"><span class="permalink"><a href="#a2e7c14b8a118f81c1df46ea5045e297b">&#9670;&nbsp;</a></span>get()</h2>
<a class="anchor" id="a2e7c14b8a118f81c1df46ea5045e297b"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
@@ -294,9 +304,7 @@ template&lt;class T, class Deleter = default_delete&lt;T&gt;&gt; </div>
</div>
</div>
<a id="a5b8d8ecafb4da336acd50e40cd42b6e0"></a>
<h2 class="memtitle"><span class="permalink"><a href="#a5b8d8ecafb4da336acd50e40cd42b6e0">&#9670;&nbsp;</a></span>get_deleter() <span class="overload">[1/2]</span></h2>
<a class="anchor" id="a5b8d8ecafb4da336acd50e40cd42b6e0"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
@@ -321,9 +329,7 @@ template&lt;class T, class Deleter = default_delete&lt;T&gt;&gt; </div>
</div>
</div>
<a id="aa427ab4ea4f2336ac6db28d53a4c11ac"></a>
<h2 class="memtitle"><span class="permalink"><a href="#aa427ab4ea4f2336ac6db28d53a4c11ac">&#9670;&nbsp;</a></span>get_deleter() <span class="overload">[2/2]</span></h2>
<a class="anchor" id="aa427ab4ea4f2336ac6db28d53a4c11ac"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
@@ -348,9 +354,7 @@ template&lt;class T, class Deleter = default_delete&lt;T&gt;&gt; </div>
</div>
</div>
<a id="a5791650488ae864f10ad04bec4a31005"></a>
<h2 class="memtitle"><span class="permalink"><a href="#a5791650488ae864f10ad04bec4a31005">&#9670;&nbsp;</a></span>operator bool()</h2>
<a class="anchor" id="a5791650488ae864f10ad04bec4a31005"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
@@ -375,9 +379,7 @@ template&lt;class T, class Deleter = default_delete&lt;T&gt;&gt; </div>
</div>
</div>
<a id="a45a3cb6d8641a6130991d56e84cbb38b"></a>
<h2 class="memtitle"><span class="permalink"><a href="#a45a3cb6d8641a6130991d56e84cbb38b">&#9670;&nbsp;</a></span>operator*()</h2>
<a class="anchor" id="a0bd8240db28f3964714da5113cd9e354"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
@@ -402,9 +404,7 @@ template&lt;class T, class Deleter = default_delete&lt;T&gt;&gt; </div>
</div>
</div>
<a id="afa52edcaef23461ce1f9c1dac349c24b"></a>
<h2 class="memtitle"><span class="permalink"><a href="#afa52edcaef23461ce1f9c1dac349c24b">&#9670;&nbsp;</a></span>operator-&gt;()</h2>
<a class="anchor" id="afa52edcaef23461ce1f9c1dac349c24b"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
@@ -429,9 +429,7 @@ template&lt;class T, class Deleter = default_delete&lt;T&gt;&gt; </div>
</div>
</div>
<a id="a5c7a204af07a7d325b0a8303e199a50d"></a>
<h2 class="memtitle"><span class="permalink"><a href="#a5c7a204af07a7d325b0a8303e199a50d">&#9670;&nbsp;</a></span>operator[]()</h2>
<a class="anchor" id="a353b89f40cb21e0cfd2af2094d417043"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
@@ -457,9 +455,7 @@ template&lt;class T, class Deleter = default_delete&lt;T&gt;&gt; </div>
</div>
</div>
<a id="a7ac06ebe7bc66573d3225891e12d2279"></a>
<h2 class="memtitle"><span class="permalink"><a href="#a7ac06ebe7bc66573d3225891e12d2279">&#9670;&nbsp;</a></span>release()</h2>
<a class="anchor" id="a7ac06ebe7bc66573d3225891e12d2279"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
@@ -484,9 +480,7 @@ template&lt;class T, class Deleter = default_delete&lt;T&gt;&gt; </div>
</div>
</div>
<a id="a6740f71511f5495d6038cf8878862331"></a>
<h2 class="memtitle"><span class="permalink"><a href="#a6740f71511f5495d6038cf8878862331">&#9670;&nbsp;</a></span>reset()</h2>
<a class="anchor" id="a6740f71511f5495d6038cf8878862331"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
@@ -512,9 +506,7 @@ template&lt;class T, class Deleter = default_delete&lt;T&gt;&gt; </div>
</div>
</div>
<a id="a748d413c50bdbbe9e2f9986fbc423036"></a>
<h2 class="memtitle"><span class="permalink"><a href="#a748d413c50bdbbe9e2f9986fbc423036">&#9670;&nbsp;</a></span>swap()</h2>
<a class="anchor" id="a748d413c50bdbbe9e2f9986fbc423036"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
@@ -546,9 +538,9 @@ template&lt;class T, class Deleter = default_delete&lt;T&gt;&gt; </div>
</div><!-- contents -->
<!-- start footer part -->
<hr class="footer"/><address class="footer"><small>
Generated on Fri Oct 26 2018 14:53:40 for Cutlass by &#160;<a href="http://www.doxygen.org/index.html">
Generated by &#160;<a href="http://www.doxygen.org/index.html">
<img class="footer" src="doxygen.png" alt="doxygen"/>
</a> 1.8.14
</a> 1.8.11
</small></address>
</body>
</html>