squall/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity
06eb90cc0d
cutlass/docs/classes.html

300 lines
226 KiB
HTML
Raw Normal View History

CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta http-equiv="Content-Type" content="text/xhtml;charset=UTF-8"/>
<meta http-equiv="X-UA-Compatible" content="IE=9"/>
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.
2019-11-20 08:55:34 +08:00
<meta name="generator" content="Doxygen 1.8.11"/>
<title>CUTLASS: Class Index</title>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
<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>
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.
2019-11-20 08:55:34 +08:00
<script type="text/javascript">
$(document).ready(function() { init_search(); });
</script>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
<script type="text/x-mathjax-config">
MathJax.Hub.Config({
extensions: ["tex2jax.js"],
jax: ["input/TeX","output/HTML-CSS"],
});
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.
2019-11-20 08:55:34 +08:00
</script><script type="text/javascript" src="http://cdn.mathjax.org/mathjax/latest/MathJax.js"></script>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
<link href="doxygen.css" rel="stylesheet" type="text/css" />
</head>
<body>
<div id="top"><!-- do not remove this div, it is closed by doxygen! -->
<div id="titlearea">
<table cellspacing="0" cellpadding="0">
<tbody>
<tr style="height: 56px;">
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.
2019-11-20 08:55:34 +08:00
<td id="projectlogo"><img alt="Logo" src="cutlass-logo-small.png"/></td>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
<td id="projectalign" style="padding-left: 0.5em;">
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.
2019-11-20 08:55:34 +08:00
<div id="projectname">CUTLASS
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</div>
<div id="projectbrief">CUDA Templates for Linear Algebra Subroutines and Solvers</div>
</td>
</tr>
</tbody>
</table>
</div>
<!-- end header part -->
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.
2019-11-20 08:55:34 +08:00
<!-- Generated by Doxygen 1.8.11 -->
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
<script type="text/javascript">
var searchBox = new SearchBox("searchBox", "search",false,'Search');
</script>
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.
2019-11-20 08:55:34 +08:00
<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 class="current"><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>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</div><!-- top -->
<!-- window showing the filter options -->
<div id="MSearchSelectWindow"
onmouseover="return searchBox.OnSearchSelectShow()"
onmouseout="return searchBox.OnSearchSelectHide()"
onkeydown="return searchBox.OnSearchSelectKey(event)">
</div>
<!-- iframe showing the search results (closed by default) -->
<div id="MSearchResultsWindow">
<iframe src="javascript:void(0)" frameborder="0"
name="MSearchResults" id="MSearchResults">
</iframe>
</div>
<div class="header">
<div class="headertitle">
<div class="title">Class Index</div> </div>
</div><!--header-->
<div class="contents">
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.
2019-11-20 08:55:34 +08:00
<div class="qindex"><a class="qindex" href="#letter_A">A</a>&#160;|&#160;<a class="qindex" href="#letter_B">B</a>&#160;|&#160;<a class="qindex" href="#letter_C">C</a>&#160;|&#160;<a class="qindex" href="#letter_D">D</a>&#160;|&#160;<a class="qindex" href="#letter_E">E</a>&#160;|&#160;<a class="qindex" href="#letter_F">F</a>&#160;|&#160;<a class="qindex" href="#letter_G">G</a>&#160;|&#160;<a class="qindex" href="#letter_H">H</a>&#160;|&#160;<a class="qindex" href="#letter_I">I</a>&#160;|&#160;<a class="qindex" href="#letter_K">K</a>&#160;|&#160;<a class="qindex" href="#letter_L">L</a>&#160;|&#160;<a class="qindex" href="#letter_M">M</a>&#160;|&#160;<a class="qindex" href="#letter_N">N</a>&#160;|&#160;<a class="qindex" href="#letter_O">O</a>&#160;|&#160;<a class="qindex" href="#letter_P">P</a>&#160;|&#160;<a class="qindex" href="#letter_R">R</a>&#160;|&#160;<a class="qindex" href="#letter_S">S</a>&#160;|&#160;<a class="qindex" href="#letter_T">T</a>&#160;|&#160;<a class="qindex" href="#letter_U">U</a>&#160;|&#160;<a class="qindex" href="#letter_V">V</a>&#160;|&#160;<a class="qindex" href="#letter_W">W</a>&#160;|&#160;<a class="qindex" href="#letter_X">X</a></div>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
<table class="classindex">
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.
2019-11-20 08:55:34 +08:00
<tr><td rowspan="2" valign="bottom"><a name="letter_A"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;A&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1FragmentIteratorVoltaTensorOp.html">FragmentIteratorVoltaTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1detail_1_1Mma__HFMA2_3_01Shape_00_01LayoutA_00_01LayoutB_00_0e1104c65871c539155bd3a0c7631928b.html">Mma_HFMA2&lt; Shape, LayoutA, LayoutB, layout::ColumnMajor, false &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread_1_1detail.html">cutlass::gemm::thread::detail</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_R"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;R&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="structcutlass_1_1TypeTraits_3_01double_01_4.html">TypeTraits&lt; double &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1FragmentIteratorVoltaTensorOp_3_01WarpShape___00_01gemm_1_1Gdb805a2dc5571ac3b66e0fe6ffdcede2.html">FragmentIteratorVoltaTensorOp&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, float, layout::RowMajor &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1detail_1_1Mma__HFMA2_3_01Shape_00_01LayoutA_00_01LayoutB_00_07ac147cb320ee0d28ff8e78eb4cd330e.html">Mma_HFMA2&lt; Shape, LayoutA, LayoutB, layout::RowMajor, false &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread_1_1detail.html">cutlass::gemm::thread::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1TypeTraits_3_01float_01_4.html">TypeTraits&lt; float &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator2dThreadTile_3_01Shape___00_0b878062cc0cd214bf7e17d74ff17e246.html">PredicatedTileIterator2dThreadTile&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, Transpose_ &gt;::AccessType</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1FragmentIteratorVoltaTensorOp_3_01WarpShape___00_01gemm_1_1G16e08718cffa0989cce3fe8dbc4b075b.html">FragmentIteratorVoltaTensorOp&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, half_t, layout::RowMajor &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1threadblock_1_1MmaBase.html">MmaBase</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1RandomGaussianFunc.html">RandomGaussianFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1TypeTraits_3_01half__t_01_4.html">TypeTraits&lt; half_t &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1AlignedArray.html">AlignedArray</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1FragmentIteratorWmmaTensorOp.html">FragmentIteratorWmmaTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaComplexTensorOp.html">MmaComplexTensorOp</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1RandomGaussianFunc.html">RandomGaussianFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1TypeTraits_3_01int_01_4.html">TypeTraits&lt; int &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1AlignedBuffer.html">AlignedBuffer</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1FragmentIteratorWmmaTensorOp_3_01WarpShape___00_01OperatorShfdb1f120c6797383663f9fd11d0fc599.html">FragmentIteratorWmmaTensorOp&lt; WarpShape_, OperatorShape_, OperatorElementC_, OperatorFragmentC_, layout::RowMajor &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaComplexTensorOp_3_01Shape___00_01complex_3_01RealElementA_01_146441010dad1f40eb51b6dae3ded216.html">MmaComplexTensorOp&lt; Shape_, complex&lt; RealElementA &gt;, LayoutA_, complex&lt; RealElementB &gt;, LayoutB_, complex&lt; RealElementC &gt;, LayoutC_, Policy_, TransformA, TransformB, Enable &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1RandomGaussianFunc_3_01complex_3_01Element_01_4_01_4.html">RandomGaussianFunc&lt; complex&lt; Element &gt; &gt;</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1TypeTraits_3_01int64__t_01_4.html">TypeTraits&lt; int64_t &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1Gemm_1_1Arguments.html">Gemm::Arguments</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_G"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;G&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1MmaGeneric.html">MmaGeneric</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread.html">cutlass::gemm::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1RandomUniformFunc.html">RandomUniformFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1TypeTraits_3_01int8__t_01_4.html">TypeTraits&lt; int8_t &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1Gemm_3_01ElementA___00_01LayoutA___00_01ElementB___00_01Layou1b211cc9c97c022d8fe10f2dd32c8709.html">Gemm&lt; ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, SplitKSerial, Operator_, IsBetaZero &gt;::Arguments</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1threadblock_1_1MmaPipelined.html">MmaPipelined</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1RandomUniformFunc.html">RandomUniformFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1TypeTraits_3_01uint64__t_01_4.html">TypeTraits&lt; uint64_t &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments.html">GemmBatched::Arguments</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1device_1_1Gemm.html">Gemm</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1MmaPolicy.html">MmaPolicy</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1RandomUniformFunc_3_01complex_3_01Element_01_4_01_4.html">RandomUniformFunc&lt; complex&lt; Element &gt; &gt;</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1TypeTraits_3_01uint8__t_01_4.html">TypeTraits&lt; uint8_t &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5.html">GemmBatched&lt; ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ &gt;::Arguments</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1Gemm.html">Gemm</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimt.html">MmaSimt</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1RealType.html">RealType</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1TypeTraits_3_01unsigned_01_4.html">TypeTraits&lt; unsigned &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1GemmComplex_1_1Arguments.html">GemmComplex::Arguments</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1Gemm.html">Gemm</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device.html">cutlass::reference::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1warp_1_1MmaSimtPolicy.html">MmaSimtPolicy</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1RealType_3_01complex_3_01T_01_4_01_4.html">RealType&lt; complex&lt; T &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_V"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;V&#160;&#160;</div></td></tr></table>
</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1GemmComplex_3_01ElementA___00_01LayoutA___00_01ElementB___00_a3923967cafb5cb9774c320dc24baa77.html">GemmComplex&lt; ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, TransformA, TransformB, SplitKSerial &gt;::Arguments</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1thread_1_1Gemm.html">Gemm</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1thread.html">cutlass::reference::device::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator.html">MmaSimtTileIterator</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reduction_1_1thread_1_1Reduce.html">Reduce</a> (<a class="el" href="namespacecutlass_1_1reduction_1_1thread.html">cutlass::reduction::thread</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1GemmSplitKParallel_1_1Arguments.html">GemmSplitKParallel::Arguments</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1Gemm.html">Gemm</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host.html">cutlass::reference::host</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kA_00_01Element_67ca7e11a38e38f2c51b84767654a90f.html">MmaSimtTileIterator&lt; Shape_, Operand::kA, Element_, layout::ColumnMajor, Policy_, PartitionsK, PartitionGroupSize &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reduction_1_1thread_1_1Reduce_3_01plus_3_01half__t_01_4_00_01AlignedArray_3_01half__t_00_01N_01_4_01_4.html">Reduce&lt; plus&lt; half_t &gt;, AlignedArray&lt; half_t, N &gt; &gt;</a> (<a class="el" href="namespacecutlass_1_1reduction_1_1thread.html">cutlass::reduction::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1VoltaTensorOpMultiplicandBCongruous.html">VoltaTensorOpMultiplicandBCongruous</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1GemmSplitKParallel_3_01ElementA___00_01LayoutA___00_01Elementafcb1aeaf2035a7ac769d7acc233423b.html">GemmSplitKParallel&lt; ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ &gt;::Arguments</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1Gemm_3_01ElementA_00_01LayoutA_00_01ElementB_00_01Layout4e016ab7cfc644acd7cb4ae770339773.html">Gemm&lt; ElementA, LayoutA, ElementB, LayoutB, ElementC, LayoutC, ScalarType, AccumulatorType, arch::OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device.html">cutlass::reference::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kA_00_01Element_f0ce904a9294556f15e1cc9cf7c99a93.html">MmaSimtTileIterator&lt; Shape_, Operand::kA, Element_, layout::ColumnMajorInterleaved&lt; 4 &gt;, Policy_, PartitionsK, PartitionGroupSize &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reduction_1_1thread_1_1Reduce_3_01plus_3_01half__t_01_4_00_01Array_3_01half__t_00_01N_01_4_01_4.html">Reduce&lt; plus&lt; half_t &gt;, Array&lt; half_t, N &gt; &gt;</a> (<a class="el" href="namespacecutlass_1_1reduction_1_1thread.html">cutlass::reduction::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1VoltaTensorOpMultiplicandCongruous.html">VoltaTensorOpMultiplicandCongruous</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1Array_3_01T_00_01N_00_01false_01_4.html">Array&lt; T, N, false &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1Gemm_3_01ElementA_00_01LayoutA_00_01ElementB_00_01Layout30b72addd464a2ca4a26785cbfd77a8e.html">Gemm&lt; ElementA, LayoutA, ElementB, LayoutB, ElementC, LayoutC, ScalarType, AccumulatorType, arch::OpMultiplyAddSaturate &gt;</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device.html">cutlass::reference::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kB_00_01Element_ea0a4e7ce3cd5d25cabf79383efdf4d9.html">MmaSimtTileIterator&lt; Shape_, Operand::kB, Element_, layout::RowMajor, Policy_, PartitionsK, PartitionGroupSize &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reduction_1_1thread_1_1Reduce_3_01plus_3_01T_01_4_00_01Array_3_01T_00_01N_01_4_01_4.html">Reduce&lt; plus&lt; T &gt;, Array&lt; T, N &gt; &gt;</a> (<a class="el" href="namespacecutlass_1_1reduction_1_1thread.html">cutlass::reduction::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1VoltaTensorOpMultiplicandCrosswise.html">VoltaTensorOpMultiplicandCrosswise</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1Array_3_01T_00_01N_00_01true_01_4.html">Array&lt; T, N, true &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1Gemm_3_01ElementA_00_01LayoutA_00_01ElementB_00_01Layout660562b232f408218828ca5915b7e73a.html">Gemm&lt; ElementA, LayoutA, ElementB, LayoutB, ElementC, LayoutC, ScalarType, AccumulatorType, arch::OpXorPopc &gt;</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device.html">cutlass::reference::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kB_00_01Element_ada156b62fcbdce47009c5bf1321c92c.html">MmaSimtTileIterator&lt; Shape_, Operand::kB, Element_, layout::RowMajorInterleaved&lt; 4 &gt;, Policy_, PartitionsK, PartitionGroupSize &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reduction_1_1thread_1_1Reduce_3_01plus_3_01T_01_4_00_01T_01_4.html">Reduce&lt; plus&lt; T &gt;, T &gt;</a> (<a class="el" href="namespacecutlass_1_1reduction_1_1thread.html">cutlass::reduction::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1warp_1_1VoltaTensorOpPolicy.html">VoltaTensorOpPolicy</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td></tr>
<tr><td rowspan="2" valign="bottom"><a name="letter_B"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;B&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1Gemm_3_01ElementA_00_01LayoutA_00_01ElementB_00_01LayoutB_193dd3a37f00deff1e5dcd7c310afb1f.html">Gemm&lt; ElementA, LayoutA, ElementB, LayoutB, ElementC, LayoutC, ScalarType, ComputeType, arch::OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host.html">cutlass::reference::host</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kC_00_01Element_4ccafbc821b3a55cd532602442a74031.html">MmaSimtTileIterator&lt; Shape_, Operand::kC, Element_, layout::ColumnMajor, Policy_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reduction_1_1thread_1_1ReduceAdd.html">ReduceAdd</a> (<a class="el" href="namespacecutlass_1_1reduction_1_1thread.html">cutlass::reduction::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1warp_1_1VoltaTensorOpPolicy_3_01WarpShape___00_01gemm_1_1GemmShape_136ce744d4c1c6e8707f5a9785196194.html">VoltaTensorOpPolicy&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, float, layout::RowMajor &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1Gemm_3_01ElementA_00_01LayoutA_00_01ElementB_00_01LayoutB_55729eac7dbd6bf311ea36f680e83e93.html">Gemm&lt; ElementA, LayoutA, ElementB, LayoutB, ElementC, LayoutC, ScalarType, ComputeType, arch::OpMultiplyAddSaturate &gt;</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host.html">cutlass::reference::host</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kC_00_01Element_8f92ea79e85febb67169c4b2d94b1b20.html">MmaSimtTileIterator&lt; Shape_, Operand::kC, Element_, layout::RowMajor, Policy_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1reduction_1_1kernel_1_1ReduceSplitK.html">ReduceSplitK</a> (<a class="el" href="namespacecutlass_1_1reduction_1_1kernel.html">cutlass::reduction::kernel</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1warp_1_1VoltaTensorOpPolicy_3_01WarpShape___00_01gemm_1_1GemmShape_1d48185f49e4d066f8e9327bf0856b7f.html">VoltaTensorOpPolicy&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, half_t, layout::RowMajor &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1BatchedGemmCoord.html">BatchedGemmCoord</a> (<a class="el" href="namespacecutlass_1_1gemm.html">cutlass::gemm</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1Gemm_3_01ElementA_00_01LayoutA_00_01ElementB_00_01LayoutB_4f3f32c4b336238abfd741e87bfced46.html">Gemm&lt; ElementA, LayoutA, ElementB, LayoutB, ElementC, LayoutC, ScalarType, ComputeType, arch::OpXorPopc &gt;</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host.html">cutlass::reference::host</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1threadblock_1_1MmaSingleStage.html">MmaSingleStage</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1thread_1_1ReductionOpPlus.html">ReductionOpPlus</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1thread.html">cutlass::epilogue::thread</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_W"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;W&#160;&#160;</div></td></tr></table>
</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1reduction_1_1BatchedReduction.html">BatchedReduction</a> (<a class="el" href="namespacecutlass_1_1reduction.html">cutlass::reduction</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1device_1_1Gemm_3_01ElementA___00_01LayoutA___00_01ElementB___00_01Layout4d0960ae6b1d1bf19e6239dbd002249c.html">Gemm&lt; ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, SplitKSerial, Operator_, IsBetaZero &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOp.html">MmaTensorOp</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1ReferenceFactory.html">ReferenceFactory</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1reduction_1_1BatchedReductionTraits.html">BatchedReductionTraits</a> (<a class="el" href="namespacecutlass_1_1reduction.html">cutlass::reduction</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1library_1_1GemmArguments.html">GemmArguments</a> (<a class="el" href="namespacecutlass_1_1library.html">cutlass::library</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpAccumulatorTileIterator.html">MmaTensorOpAccumulatorTileIterator</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1ReferenceFactory_3_01Element_00_01false_01_4.html">ReferenceFactory&lt; Element, false &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1warp_1_1WarpSize.html">WarpSize</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1BlockForEach.html">BlockForEach</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device.html">cutlass::reference::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1library_1_1GemmArrayArguments.html">GemmArrayArguments</a> (<a class="el" href="namespacecutlass_1_1library.html">cutlass::library</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpAccumulatorTileIterator_3_01Shape___00_01Element___008f607b871a2b3d854eb4def64712c042.html">MmaTensorOpAccumulatorTileIterator&lt; Shape_, Element_, cutlass::layout::ColumnMajor, InstructionShape_, OpDelta_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1ReferenceFactory_3_01Element_00_01true_01_4.html">ReferenceFactory&lt; Element, true &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Wmma_3_01Shape___00_01cutlass_1_1half__t_00_01LayoutA___00_01cutlass_1_84e30c8cc93eeb7ca02f651bd16d4c38.html">Wmma&lt; Shape_, cutlass::half_t, LayoutA_, cutlass::half_t, LayoutB_, ElementC_, LayoutC_, cutlass::arch::OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1BlockForEach.html">BlockForEach</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host.html">cutlass::reference::host</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1library_1_1GemmArrayConfiguration.html">GemmArrayConfiguration</a> (<a class="el" href="namespacecutlass_1_1library.html">cutlass::library</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpAccumulatorTileIterator_3_01Shape___00_01Element___00027dabdc144edd6276f664ca74088510.html">MmaTensorOpAccumulatorTileIterator&lt; Shape_, Element_, cutlass::layout::ColumnMajorInterleaved&lt; InterleavedN &gt;, InstructionShape_, OpDelta_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator.html">RegularTileAccessIterator</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Wmma_3_01Shape___00_01cutlass_1_1int4b__t_00_01LayoutA___00_01cutlass_16fd808a90b3cf9d7cfc99f30888ca3fe.html">Wmma&lt; Shape_, cutlass::int4b_t, LayoutA_, cutlass::int4b_t, LayoutB_, int32_t, LayoutC_, cutlass::arch::OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td></tr>
<tr><td rowspan="2" valign="bottom"><a name="letter_C"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;C&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html">GemmBatched</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpAccumulatorTileIterator_3_01Shape___00_01Element___006c39f57875e0aa9d0ad82c8043ed8b98.html">MmaTensorOpAccumulatorTileIterator&lt; Shape_, Element_, cutlass::layout::RowMajor, InstructionShape_, OpDelta_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__eb7d20f8b9d69e0ae5e7ef51dc480867.html">RegularTileAccessIterator&lt; Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Wmma_3_01Shape___00_01cutlass_1_1uint1b__t_00_01LayoutA___00_01cutlass_c80a7ea4d219cd9b13b560b493338028.html">Wmma&lt; Shape_, cutlass::uint1b_t, LayoutA_, cutlass::uint1b_t, LayoutB_, int32_t, LayoutC_, cutlass::arch::OpXorPopc &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched.html">GemmBatched</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator.html">MmaTensorOpMultiplicandTileIterator</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__2c1476eaf582bfe972793e17babfe985.html">RegularTileAccessIterator&lt; Shape_, Element_, layout::ColumnMajorTensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value, int(128/sizeof(Element_))&gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Wmma_3_01Shape___00_01int8__t_00_01LayoutA___00_01int8__t_00_01LayoutB_505c57bb6818a941dc16f00cf35a9ec0.html">Wmma&lt; Shape_, int8_t, LayoutA_, int8_t, LayoutB_, int32_t, LayoutC_, cutlass::arch::OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1detail_1_1Cast.html">Cast</a> (<a class="el" href="namespacecutlass_1_1reference_1_1detail.html">cutlass::reference::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html">GemmBatched&lt; ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___0b84f53cd44b339eccc12067c9f86e11c.html">MmaTensorOpMultiplicandTileIterator&lt; Shape_, Operand_, Element_, cutlass::layout::ColumnMajorTensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value, int(128/sizeof(Element_))&gt;, InstructionShape_, OpDelta_, 32, PartitionsK_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__a3c11cf1f00ef7a1efb8389ac6e4c6e0.html">RegularTileAccessIterator&lt; Shape_, Element_, layout::ColumnMajorTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Wmma_3_01Shape___00_01uint8__t_00_01LayoutA___00_01uint8__t_00_01Layout219a464a1248ebfc37aa29bcb10cb1b0.html">Wmma&lt; Shape_, uint8_t, LayoutA_, uint8_t, LayoutB_, int32_t, LayoutC_, cutlass::arch::OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1detail_1_1Cast_3_01float_00_01int8__t_01_4.html">Cast&lt; float, int8_t &gt;</a> (<a class="el" href="namespacecutlass_1_1reference_1_1detail.html">cutlass::reference::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1library_1_1GemmBatchedConfiguration.html">GemmBatchedConfiguration</a> (<a class="el" href="namespacecutlass_1_1library.html">cutlass::library</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___0e52ad425e1ee3e68544873f66733237b.html">MmaTensorOpMultiplicandTileIterator&lt; Shape_, Operand_, Element_, cutlass::layout::ColumnMajorTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, InstructionShape_, OpDelta_, 32, PartitionsK_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__0855e9d9ab619202d2397180c1e4c4a5.html">RegularTileAccessIterator&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_a"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;a&#160;&#160;</div></td></tr></table>
</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1detail_1_1Cast_3_01float_00_01uint8__t_01_4.html">Cast&lt; float, uint8_t &gt;</a> (<a class="el" href="namespacecutlass_1_1reference_1_1detail.html">cutlass::reference::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1GemmBatchedIdentityThreadblockSwizzle.html">GemmBatchedIdentityThreadblockSwizzle</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___039819fb3ccd43786d556c2c9669508ef.html">MmaTensorOpMultiplicandTileIterator&lt; Shape_, Operand_, Element_, cutlass::layout::RowMajorTensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value, int(128/sizeof(Element_))&gt;, InstructionShape_, OpDelta_, 32, PartitionsK_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__f04332958a49a47d6fb2b25201764630.html">RegularTileAccessIterator&lt; Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1layout_1_1ColumnMajor.html">ColumnMajor</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1device_1_1GemmComplex.html">GemmComplex</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___0352e0dcab42bc8360606874e00173556.html">MmaTensorOpMultiplicandTileIterator&lt; Shape_, Operand_, Element_, cutlass::layout::RowMajorTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, InstructionShape_, OpDelta_, 32, PartitionsK_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__6baada077236f1a368c61c5e11b45b72.html">RegularTileAccessIterator&lt; Shape_, Element_, layout::RowMajorTensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value, int(128/sizeof(Element_))&gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1aligned__chunk.html">aligned_chunk</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1ColumnMajorBlockLinear.html">ColumnMajorBlockLinear</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1device_1_1GemmComplex_3_01ElementA___00_01LayoutA___00_01ElementB___00_07c56401b4df75709ae636675d9980a9a.html">GemmComplex&lt; ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, TransformA, TransformB, SplitKSerial &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___0ed7daaeba1c095e77f68533d4d2c475c.html">MmaTensorOpMultiplicandTileIterator&lt; Shape_, Operand_, Element_, cutlass::layout::TensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value, 64 &gt;, InstructionShape_, OpDelta_, 32, PartitionsK_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__0184b7188941788a96624510a4b2f876.html">RegularTileAccessIterator&lt; Shape_, Element_, layout::RowMajorTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1aligned__storage.html">aligned_storage</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1ColumnMajorInterleaved.html">ColumnMajorInterleaved</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1library_1_1GemmConfiguration.html">GemmConfiguration</a> (<a class="el" href="namespacecutlass_1_1library.html">cutlass::library</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___0c7d419c589d601ce4eb603be566fea21.html">MmaTensorOpMultiplicandTileIterator&lt; Shape_, Operand_, Element_, cutlass::layout::TensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, InstructionShape_, OpDelta_, 32, PartitionsK_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__ebf4714349612673e8b6609b763eeb6f.html">RegularTileAccessIterator&lt; Shape_, Element_, layout::TensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value, int(128/sizeof(Element_))&gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1alignment__of.html">alignment_of</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1ColumnMajorTensorOpMultiplicandCongruous.html">ColumnMajorTensorOpMultiplicandCongruous</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1GemmCoord.html">GemmCoord</a> (<a class="el" href="namespacecutlass_1_1gemm.html">cutlass::gemm</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1warp_1_1MmaTensorOpPolicy.html">MmaTensorOpPolicy</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__e9a9e0f4286f652f55eb9b863b21effe.html">RegularTileAccessIterator&lt; Shape_, Element_, layout::TensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1alignment__of_3_01const_01value__t_01_4.html">alignment_of&lt; const value_t &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1ColumnMajorTensorOpMultiplicandCrosswise.html">ColumnMajorTensorOpMultiplicandCrosswise</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1library_1_1GemmDescription.html">GemmDescription</a> (<a class="el" href="namespacecutlass_1_1library.html">cutlass::library</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOp.html">MmaVoltaTensorOp</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator.html">RegularTileIterator</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1alignment__of_3_01const_01volatile_01value__t_01_4.html">alignment_of&lt; const volatile value_t &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1ColumnMajorVoltaTensorOpMultiplicandBCongruous.html">ColumnMajorVoltaTensorOpMultiplicandBCongruous</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1GemmHorizontalThreadblockSwizzle.html">GemmHorizontalThreadblockSwizzle</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpAccumulatorTileIterator.html">MmaVoltaTensorOpAccumulatorTileIterator</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator2dThreadTile.html">RegularTileIterator2dThreadTile</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1alignment__of_3_01double2_01_4.html">alignment_of&lt; double2 &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1ColumnMajorVoltaTensorOpMultiplicandCongruous.html">ColumnMajorVoltaTensorOpMultiplicandCongruous</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1GemmIdentityThreadblockSwizzle.html">GemmIdentityThreadblockSwizzle</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator.html">MmaVoltaTensorOpMultiplicandTileIterator</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator2dThreadTile_3_01Shape___00_01Eleb60d066756d1c18f05fceee6a27bdb8a.html">RegularTileIterator2dThreadTile&lt; Shape_, Element_, layout::ColumnMajorInterleaved&lt; 4 &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1alignment__of_3_01double4_01_4.html">alignment_of&lt; double4 &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1ColumnMajorVoltaTensorOpMultiplicandCrosswise.html">ColumnMajorVoltaTensorOpMultiplicandCrosswise</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1library_1_1GemmPlanarComplexBatchedConfiguration.html">GemmPlanarComplexBatchedConfiguration</a> (<a class="el" href="namespacecutlass_1_1library.html">cutlass::library</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operan0d3248553e52cd61ed8a2b3b12a20343.html">MmaVoltaTensorOpMultiplicandTileIterator&lt; Shape_, Operand::kA, Element_, cutlass::layout::ColumnMajorVoltaTensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, InstructionShape_, OpDelta_, 32 &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator2dThreadTile_3_01Shape___00_01Ele76ed82829532ae1c17f4c78158f036c7.html">RegularTileIterator2dThreadTile&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1alignment__of_3_01float4_01_4.html">alignment_of&lt; float4 &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1CommandLine.html">CommandLine</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1library_1_1GemmPlanarComplexConfiguration.html">GemmPlanarComplexConfiguration</a> (<a class="el" href="namespacecutlass_1_1library.html">cutlass::library</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operan34be8e21a40af3ebd2dc3dff460dca72.html">MmaVoltaTensorOpMultiplicandTileIterator&lt; Shape_, Operand::kA, Element_, cutlass::layout::VoltaTensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, InstructionShape_, OpDelta_, 32 &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator2dThreadTile_3_01Shape___00_01Ele654c8f6161ae5340f040397a4e2e045c.html">RegularTileIterator2dThreadTile&lt; Shape_, Element_, layout::RowMajorInterleaved&lt; 4 &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1alignment__of_3_01int4_01_4.html">alignment_of&lt; int4 &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1OutputTileOptimalThreadMap_1_1CompactedThreadMap.html">OutputTileOptimalThreadMap::CompactedThreadMap</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1GemmShape.html">GemmShape</a> (<a class="el" href="namespacecutlass_1_1gemm.html">cutlass::gemm</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand734577b7e54a074d143aba59828c2f2.html">MmaVoltaTensorOpMultiplicandTileIterator&lt; Shape_, Operand::kB, Element_, cutlass::layout::RowMajorVoltaTensorOpMultiplicandBCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, InstructionShape_, OpDelta_, 32 &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_011d3637dbd8bc58bcb020b51bf57fbfc0.html">RegularTileIterator&lt; Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1alignment__of_3_01long4_01_4.html">alignment_of&lt; long4 &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1PredicateVector_1_1ConstIterator.html">PredicateVector::ConstIterator</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1GemmSplitKHorizontalThreadblockSwizzle.html">GemmSplitKHorizontalThreadblockSwizzle</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operan16c56cdc2dda5eeb996af8ec0242d501.html">MmaVoltaTensorOpMultiplicandTileIterator&lt; Shape_, Operand::kB, Element_, cutlass::layout::VoltaTensorOpMultiplicandBCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, InstructionShape_, OpDelta_, 32 &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_017982f81d4ef592e19c8427de2ea933a3.html">RegularTileIterator&lt; Shape_, Element_, layout::ColumnMajorTensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value, int(128/sizeof(Element_))&gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1alignment__of_3_01longlong2_01_4.html">alignment_of&lt; longlong2 &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1ConstSubbyteReference.html">ConstSubbyteReference</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1GemmSplitKIdentityThreadblockSwizzle.html">GemmSplitKIdentityThreadblockSwizzle</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operan5a221944f4a0e16ccab77ba684856942.html">MmaVoltaTensorOpMultiplicandTileIterator&lt; Shape_, Operand_, Element_, cutlass::layout::ColumnMajorVoltaTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, KBlock &gt;, InstructionShape_, OpDelta_, 32 &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_010889a732373c350de9b9a9f6c13cd761.html">RegularTileIterator&lt; Shape_, Element_, layout::ColumnMajorTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1alignment__of_3_01longlong4_01_4.html">alignment_of&lt; longlong4 &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1ContiguousMatrix.html">ContiguousMatrix</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1device_1_1GemmSplitKParallel.html">GemmSplitKParallel</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operandcc9821c435540895138bc9af495f321.html">MmaVoltaTensorOpMultiplicandTileIterator&lt; Shape_, Operand_, Element_, cutlass::layout::RowMajorVoltaTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, KBlock &gt;, InstructionShape_, OpDelta_, 32 &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01187f8574e1fe9d7d5e8fbf09bd834bf0.html">RegularTileIterator&lt; Shape_, Element_, layout::ColumnMajorVoltaTensorOpMultiplicandBCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1alignment__of_3_01uint4_01_4.html">alignment_of&lt; uint4 &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1thread_1_1Convert.html">Convert</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1thread.html">cutlass::epilogue::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1GemmSplitKParallel.html">GemmSplitKParallel</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operana2f40b28f0d2286b84d86f7238d67b52.html">MmaVoltaTensorOpMultiplicandTileIterator&lt; Shape_, Operand_, Element_, cutlass::layout::VoltaTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, KBlock &gt;, InstructionShape_, OpDelta_, 32 &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01793f74bfd8f116a827948ab01a37349a.html">RegularTileIterator&lt; Shape_, Element_, layout::ColumnMajorVoltaTensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1alignment__of_3_01ulong4_01_4.html">alignment_of&lt; ulong4 &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1Coord.html">Coord</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1device_1_1GemmSplitKParallel_3_01ElementA___00_01LayoutA___00_01ElementBbe7c1f7154ad5b5bf9d4d28301e2b457.html">GemmSplitKParallel&lt; ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_N"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;N&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01bd31b3810c1fedf2e7e5959ff92b5d3d.html">RegularTileIterator&lt; Shape_, Element_, layout::ColumnMajorVoltaTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Shape_::kRow &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1alignment__of_3_01ulonglong2_01_4.html">alignment_of&lt; ulonglong2 &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td rowspan="2" valign="bottom"><a name="letter_D"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;D&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1threadblock_1_1Gemv.html">Gemv</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_0184a89653916f5d51ab59d1b386989a17.html">RegularTileIterator&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1alignment__of_3_01ulonglong4_01_4.html">alignment_of&lt; ulonglong4 &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1detail_1_1GemvBatchedStridedEpilogueScaling.html">GemvBatchedStridedEpilogueScaling</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel_1_1detail.html">cutlass::gemm::kernel::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1NumericArrayConverter.html">NumericArrayConverter</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_0149454d361ea5885cf5166a920b5145df.html">RegularTileIterator&lt; Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1alignment__of_3_01volatile_01value__t_01_4.html">alignment_of&lt; volatile value_t &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structDebugType.html">DebugType</a>&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1GemvBatchedStridedThreadblockDefaultSwizzle.html">GemvBatchedStridedThreadblockDefaultSwizzle</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1NumericArrayConverter_3_01float_00_01half__t_00_012_00_01Round_01_4.html">NumericArrayConverter&lt; float, half_t, 2, Round &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01c20d35180520077a5a09b1e33543c1a5.html">RegularTileIterator&lt; Shape_, Element_, layout::RowMajorTensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value, int(128/sizeof(Element_))&gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1device__memory_1_1allocation.html">allocation</a> (<a class="el" href="namespacecutlass_1_1device__memory.html">cutlass::device_memory</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structDebugValue.html">DebugValue</a>&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1GeneralMatrix.html">GeneralMatrix</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1NumericArrayConverter_3_01float_00_01half__t_00_01N_00_01Round_01_4.html">NumericArrayConverter&lt; float, half_t, N, Round &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01a31b454d9c930525c1e9ca406a514f40.html">RegularTileIterator&lt; Shape_, Element_, layout::RowMajorTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_b"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;b&#160;&#160;</div></td></tr></table>
</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1reduction_1_1DefaultBlockSwizzle.html">DefaultBlockSwizzle</a> (<a class="el" href="namespacecutlass_1_1reduction.html">cutlass::reduction</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_H"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;H&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="structcutlass_1_1NumericArrayConverter_3_01half__t_00_01float_00_012_00_01FloatRoundStyle_1_1round__to__nearest_01_4.html">NumericArrayConverter&lt; half_t, float, 2, FloatRoundStyle::round_to_nearest &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_0104ad31bd559a88cc418ae1cab7492ed5.html">RegularTileIterator&lt; Shape_, Element_, layout::RowMajorVoltaTensorOpMultiplicandBCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultEpilogueComplexTensorOp.html">DefaultEpilogueComplexTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1NumericArrayConverter_3_01half__t_00_01float_00_01N_00_01Round_01_4.html">NumericArrayConverter&lt; half_t, float, N, Round &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01f6f6511b5033cad31083644ac69c54d8.html">RegularTileIterator&lt; Shape_, Element_, layout::RowMajorVoltaTensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1bool__constant.html">bool_constant</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultEpilogueSimt.html">DefaultEpilogueSimt</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1HostTensor.html">HostTensor</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1NumericConverter.html">NumericConverter</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01b3fa5720e807697de61b9f937b269cd0.html">RegularTileIterator&lt; Shape_, Element_, layout::RowMajorVoltaTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Shape_::kColumn &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_c"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;c&#160;&#160;</div></td></tr></table>
</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultEpilogueTensorOp.html">DefaultEpilogueTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_I"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;I&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="structcutlass_1_1NumericConverter_3_01float_00_01half__t_00_01Round_01_4.html">NumericConverter&lt; float, half_t, Round &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01efd5013a2503d6567e2bf6b40c97360c.html">RegularTileIterator&lt; Shape_, Element_, layout::TensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value, int(128/sizeof(Element_))&gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultEpilogueVoltaTensorOp.html">DefaultEpilogueVoltaTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1NumericConverter_3_01half__t_00_01float_00_01FloatRoundStyle_1_1round__to__nearest_01_4.html">NumericConverter&lt; half_t, float, FloatRoundStyle::round_to_nearest &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_0197fef2242a3454a7d1cebe61aee28b43.html">RegularTileIterator&lt; Shape_, Element_, layout::TensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1complex.html">complex</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultEpilogueWmmaTensorOp.html">DefaultEpilogueWmmaTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1IdentityTensorLayout.html">IdentityTensorLayout</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1NumericConverter_3_01half__t_00_01float_00_01FloatRoundStyle_1_1round__toward__zero_01_4.html">NumericConverter&lt; half_t, float, FloatRoundStyle::round_toward_zero &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01a75d2cd74e722d6ad6a3b41aabfd432d.html">RegularTileIterator&lt; Shape_, Element_, layout::VoltaTensorOpMultiplicandBCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1conditional.html">conditional</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1DefaultGemm.html">DefaultGemm</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1IntegerType.html">IntegerType</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1NumericConverter_3_01int8__t_00_01float_00_01Round_01_4.html">NumericConverter&lt; int8_t, float, Round &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01f96bbeb63e6d4ce4a2551279de3a9f0e.html">RegularTileIterator&lt; Shape_, Element_, layout::VoltaTensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1conditional_3_01false_00_01T_00_01F_01_4.html">conditional&lt; false, T, F &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1DefaultGemm_3_01ElementA_00_01layout_1_1ColumnMajorInterleave661fe54d13cc2c9153dcdf31e4beaa30.html">DefaultGemm&lt; ElementA, layout::ColumnMajorInterleaved&lt; InterleavedK &gt;, kAlignmentA, ElementB, layout::RowMajorInterleaved&lt; InterleavedK &gt;, kAlignmentB, ElementC, layout::ColumnMajorInterleaved&lt; InterleavedK &gt;, int32_t, arch::OpClassTensorOp, arch::Sm75, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, 2, SplitKSerial, Operator, IsBetaZero &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1IntegerType_3_011_00_01false_01_4.html">IntegerType&lt; 1, false &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1NumericConverter_3_01T_00_01T_00_01Round_01_4.html">NumericConverter&lt; T, T, Round &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01dbd6b8468d5bd787308d2f615a24d123.html">RegularTileIterator&lt; Shape_, Element_, layout::VoltaTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Shape_::kContiguous &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1Array_3_01T_00_01N_00_01true_01_4_1_1const__iterator.html">Array&lt; T, N, true &gt;::const_iterator</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1DefaultGemm_3_01ElementA_00_01LayoutA_00_01kAlignmentA_00_01Edd80343e6570718ed237122e4ebf7fb5.html">DefaultGemm&lt; ElementA, LayoutA, kAlignmentA, ElementB, LayoutB, kAlignmentB, ElementC, layout::RowMajor, ElementAccumulator, arch::OpClassSimt, ArchTag, ThreadblockShape, WarpShape, GemmShape&lt; 1, 1, 1 &gt;, EpilogueOutputOp, ThreadblockSwizzle, 2, SplitKSerial, Operator &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1IntegerType_3_011_00_01true_01_4.html">IntegerType&lt; 1, true &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1NumericConverterClamp.html">NumericConverterClamp</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1detail_1_1RowArrangement.html">RowArrangement</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock_1_1detail.html">cutlass::epilogue::threadblock::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1Array_3_01T_00_01N_00_01false_01_4_1_1const__iterator.html">Array&lt; T, N, false &gt;::const_iterator</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1DefaultGemm_3_01ElementA_00_01LayoutA_00_01kAlignmentA_00_01E044b039b2fe402f29b04a9f5feee5342.html">DefaultGemm&lt; ElementA, LayoutA, kAlignmentA, ElementB, LayoutB, kAlignmentB, ElementC, layout::RowMajor, ElementAccumulator, arch::OpClassTensorOp, arch::Sm70, ThreadblockShape, WarpShape, GemmShape&lt; 8, 8, 4 &gt;, EpilogueOutputOp, ThreadblockSwizzle, 2, SplitKSerial, Operator &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1IntegerType_3_0116_00_01false_01_4.html">IntegerType&lt; 16, false &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_O"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;O&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1detail_1_1RowArrangement_3_01Shape_00_01WarpsRemaini91159e6f7e123d881e3ec45101fa4f81.html">RowArrangement&lt; Shape, WarpsRemaining, ElementsPerAccess, ElementSize, false &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock_1_1detail.html">cutlass::epilogue::threadblock::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1Array_3_01T_00_01N_00_01false_01_4_1_1const__reference.html">Array&lt; T, N, false &gt;::const_reference</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1DefaultGemm_3_01ElementA_00_01LayoutA_00_01kAlignmentA_00_01E5d78d37a9ae2ec08d7d477d571df036e.html">DefaultGemm&lt; ElementA, LayoutA, kAlignmentA, ElementB, LayoutB, kAlignmentB, ElementC, layout::RowMajor, ElementAccumulator, arch::OpClassTensorOp, arch::Sm75, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, 2, SplitKSerial, Operator &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1IntegerType_3_0116_00_01true_01_4.html">IntegerType&lt; 16, true &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1detail_1_1RowArrangement_3_01Shape_00_01WarpsRemaini6d8790249bf12cac580da73bb37eb791.html">RowArrangement&lt; Shape, WarpsRemaining, ElementsPerAccess, ElementSize, true &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock_1_1detail.html">cutlass::epilogue::threadblock::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1Array_3_01T_00_01N_00_01true_01_4_1_1const__reverse__iterator.html">Array&lt; T, N, true &gt;::const_reverse_iterator</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1DefaultGemm_3_01int8__t_00_01LayoutA_00_01kAlignmentA_00_01inf48440732c1c5f42ddbfaba179861815.html">DefaultGemm&lt; int8_t, LayoutA, kAlignmentA, int8_t, LayoutB, kAlignmentB, ElementC, LayoutC, ElementAccumulator, arch::OpClassSimt, ArchTag, ThreadblockShape, WarpShape, GemmShape&lt; 1, 1, 4 &gt;, EpilogueOutputOp, ThreadblockSwizzle, 2, SplitKSerial, Operator, false &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1IntegerType_3_0132_00_01false_01_4.html">IntegerType&lt; 32, false &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1library_1_1Operation.html">Operation</a> (<a class="el" href="namespacecutlass_1_1library.html">cutlass::library</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1layout_1_1RowMajor.html">RowMajor</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1Array_3_01T_00_01N_00_01false_01_4_1_1const__reverse__iterator.html">Array&lt; T, N, false &gt;::const_reverse_iterator</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration.html">DefaultGemmConfiguration</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1IntegerType_3_0132_00_01true_01_4.html">IntegerType&lt; 32, true &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1library_1_1OperationDescription.html">OperationDescription</a> (<a class="el" href="namespacecutlass_1_1library.html">cutlass::library</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1RowMajorBlockLinear.html">RowMajorBlockLinear</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1cuda__exception.html">cuda_exception</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassSimt_00_01ArchTag286687c5e6abe22d241f789fe344a465.html">DefaultGemmConfiguration&lt; arch::OpClassSimt, ArchTag, ElementA, ElementB, ElementC, ElementAccumulator &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1IntegerType_3_014_00_01false_01_4.html">IntegerType&lt; 4, false &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1OutputTileOptimalThreadMap.html">OutputTileOptimalThreadMap</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1RowMajorInterleaved.html">RowMajorInterleaved</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_d"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;d&#160;&#160;</div></td></tr></table>
</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassSimt_00_01ArchTag3026e48abb8c905d1cc6d13d669700e4.html">DefaultGemmConfiguration&lt; arch::OpClassSimt, ArchTag, int8_t, int8_t, ElementC, int32_t &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1IntegerType_3_014_00_01true_01_4.html">IntegerType&lt; 4, true &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1OutputTileShape.html">OutputTileShape</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1RowMajorTensorOpMultiplicandCongruous.html">RowMajorTensorOpMultiplicandCongruous</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arc567cad318a31d04b70ea615d6321decd.html">DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm70, ElementA, ElementB, ElementC, ElementAccumulator &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1IntegerType_3_0164_00_01false_01_4.html">IntegerType&lt; 64, false &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1OutputTileThreadMap.html">OutputTileThreadMap</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1RowMajorTensorOpMultiplicandCrosswise.html">RowMajorTensorOpMultiplicandCrosswise</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1default__delete.html">default_delete</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arcde61af9be1337dac1fdb210e7e7a6e01.html">DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, ElementA, ElementB, ElementC, ElementAccumulator &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1IntegerType_3_0164_00_01true_01_4.html">IntegerType&lt; 64, true &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_P"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;P&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1RowMajorVoltaTensorOpMultiplicandBCongruous.html">RowMajorVoltaTensorOpMultiplicandBCongruous</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1default__delete_3_01T[]_4.html">default_delete&lt; T[]&gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arc485a4f0b5a7d2d4ab2c1a24da6328048.html">DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, int4b_t, int4b_t, ElementC, int32_t &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1IntegerType_3_018_00_01false_01_4.html">IntegerType&lt; 8, false &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1RowMajorVoltaTensorOpMultiplicandCongruous.html">RowMajorVoltaTensorOpMultiplicandCongruous</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1device__memory_1_1allocation_1_1deleter.html">allocation::deleter</a> (<a class="el" href="namespacecutlass_1_1device__memory.html">cutlass::device_memory</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arc8e2604a56dff3a7595da9ee0604ae55e.html">DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, int4b_t, uint4b_t, ElementC, int32_t &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1IntegerType_3_018_00_01true_01_4.html">IntegerType&lt; 8, true &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1layout_1_1PackedVectorLayout.html">PackedVectorLayout</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1RowMajorVoltaTensorOpMultiplicandCrosswise.html">RowMajorVoltaTensorOpMultiplicandCrosswise</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1divide__assert.html">divide_assert</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arc4fada4957d463c80a2831e47f28157c4.html">DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, int8_t, int8_t, ElementC, int32_t &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1threadblock_1_1InterleavedEpilogue.html">InterleavedEpilogue</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1EpilogueWorkspace_1_1Params.html">EpilogueWorkspace::Params</a> (<a class="el" href="namespacecutlass_1_1epilogue.html">cutlass::epilogue</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_S"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;S&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="structcutlass_1_1divides.html">divides</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arc8ab5fd2693c6a6ec43e447acb07f784c.html">DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, int8_t, uint8_t, ElementC, int32_t &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1InterleavedOutputTileThreadMap.html">InterleavedOutputTileThreadMap</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1PredicatedTileIterator_1_1Params.html">PredicatedTileIterator::Params</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1divides_3_01Array_3_01half__t_00_01N_01_4_01_4.html">divides&lt; Array&lt; half_t, N &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arcffcf31256aed23d4d8d0eab627bc0cad.html">DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, uint4b_t, int4b_t, ElementC, int32_t &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1threadblock_1_1InterleavedPredicatedTileIterator.html">InterleavedPredicatedTileIterator</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1InterleavedPredicatedTileIterator_1_1Params.html">InterleavedPredicatedTileIterator::Params</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1ScalarIO.html">ScalarIO</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1divides_3_01Array_3_01T_00_01N_01_4_01_4.html">divides&lt; Array&lt; T, N &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arcb2e258b7bd321c633dd65d3ebcf6414a.html">DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, uint4b_t, uint4b_t, ElementC, int32_t &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1PredicateVector_1_1Iterator.html">PredicateVector::Iterator</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reduction_1_1thread_1_1ReduceAdd_1_1Params.html">ReduceAdd::Params</a> (<a class="el" href="namespacecutlass_1_1reduction_1_1thread.html">cutlass::reduction::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1Semaphore.html">Semaphore</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__base__of__helper_1_1dummy.html">is_base_of_helper::dummy</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arcb27bf218007928652d5b803193eab473.html">DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, uint8_t, int8_t, ElementC, int32_t &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_K"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;K&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="structcutlass_1_1reduction_1_1kernel_1_1ReduceSplitK_1_1Params.html">ReduceSplitK::Params</a> (<a class="el" href="namespacecutlass_1_1reduction_1_1kernel.html">cutlass::reduction::kernel</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1threadblock_1_1SharedLoadIterator.html">SharedLoadIterator</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_e"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;e&#160;&#160;</div></td></tr></table>
</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arcfea0f3503156e8e3fba6456f0cedafdd.html">DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, uint8_t, uint8_t, ElementC, int32_t &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reduction_1_1BatchedReductionTraits_1_1Params.html">BatchedReductionTraits::Params</a> (<a class="el" href="namespacecutlass_1_1reduction.html">cutlass::reduction</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1EpilogueWorkspace_1_1SharedStorage.html">EpilogueWorkspace::SharedStorage</a> (<a class="el" href="namespacecutlass_1_1epilogue.html">cutlass::epilogue</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassWmmaTensorOp_00_0884059ecad03bea3e86c4cf722226097.html">DefaultGemmConfiguration&lt; arch::OpClassWmmaTensorOp, ArchTag, ElementA, ElementB, ElementC, ElementAccumulator &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1device.html">cutlass::gemm::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1KernelLaunchConfiguration.html">KernelLaunchConfiguration</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator_3_01Shape___00_01Elemen41e459f664d17473570cf22fb616845f.html">PredicatedTileAccessIterator&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, AccessType_ &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DirectEpilogueTensorOp_1_1SharedStorage.html">DirectEpilogueTensorOp::SharedStorage</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1enable__if.html">enable_if</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1DefaultGemmSplitKParallel.html">DefaultGemmSplitKParallel</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_L"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;L&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator_3_01Shape___00_01Elemen44ce348364e78f5a56fa0c2cef6af930.html">PredicatedTileAccessIterator&lt; Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, AccessType_ &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1InterleavedEpilogue_1_1SharedStorage.html">InterleavedEpilogue::SharedStorage</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1enable__if_3_01false_00_01T_01_4.html">enable_if&lt; false, T &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1DefaultGemv.html">DefaultGemv</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator_3_01Shape___00_01Elemena9b06926a275b569ee9f7f142604b997.html">PredicatedTileAccessIterator&lt; Shape_, Element_, layout::ColumnMajorInterleaved&lt; InterleavedK &gt;, AdvanceRank, ThreadMap_, AccessType_ &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1EpilogueBase_1_1SharedStorage.html">EpilogueBase::SharedStorage</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_h"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;h&#160;&#160;</div></td></tr></table>
</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultGemvCore.html">DefaultGemvCore</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1LayoutTranspose.html">LayoutTranspose</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator_3_01Shape___00_01Elemen058417e2cdd86f3cd6ad5458581571c8.html">PredicatedTileAccessIterator&lt; Shape_, Element_, layout::RowMajorInterleaved&lt; InterleavedK &gt;, AdvanceRank, ThreadMap_, AccessType_ &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reduction_1_1kernel_1_1ReduceSplitK_1_1SharedStorage.html">ReduceSplitK::SharedStorage</a> (<a class="el" href="namespacecutlass_1_1reduction_1_1kernel.html">cutlass::reduction::kernel</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultInterleavedEpilogueTensorOp.html">DefaultInterleavedEpilogueTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1LayoutTranspose_3_01layout_1_1ColumnMajor_01_4.html">LayoutTranspose&lt; layout::ColumnMajor &gt;</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator2dThreadTile_3_01Shape__8ccc62d47a092afc8bee32ffe9d1e4ba.html">PredicatedTileAccessIterator2dThreadTile&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, AccessType_ &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="unioncutlass_1_1gemm_1_1kernel_1_1GemmSplitKParallel_1_1SharedStorage.html">GemmSplitKParallel::SharedStorage</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1half__t.html">half_t</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultInterleavedThreadMapTensorOp.html">DefaultInterleavedThreadMapTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1LayoutTranspose_3_01layout_1_1RowMajor_01_4.html">LayoutTranspose&lt; layout::RowMajor &gt;</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1thread_1_1Convert_1_1Params.html">Convert::Params</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1thread.html">cutlass::epilogue::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="unioncutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1SharedStorage.html">GemmBatched::SharedStorage</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_i"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;i&#160;&#160;</div></td></tr></table>
</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMma.html">DefaultMma</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1thread_1_1LinearCombination.html">LinearCombination</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1thread.html">cutlass::epilogue::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator2dThreadTile_3_01Shape__a56cbccec33ee916292ad9d068474609.html">PredicatedTileAccessIterator2dThreadTile&lt; Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, AccessType_ &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="unioncutlass_1_1gemm_1_1kernel_1_1Gemm_1_1SharedStorage.html">Gemm::SharedStorage</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMma_3_01ElementA_00_01LayoutA_00_01kAlignmentA_0010764e1fd5a3251a57eddafbd83eab8e.html">DefaultMma&lt; ElementA, LayoutA, kAlignmentA, ElementB, LayoutB, kAlignmentB, ElementAccumulator, layout::ColumnMajorInterleaved&lt; InterleavedK &gt;, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, 2, Operator, true &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1thread_1_1LinearCombinationClamp.html">LinearCombinationClamp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1thread.html">cutlass::epilogue::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape___00_01Element___006a5f2f7a8271031e6cdc5daa5441f2af.html">PredicatedTileIterator&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, AccessSize &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1gemm_1_1threadblock_1_1MmaBase_1_1SharedStorage.html">MmaBase::SharedStorage</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1integer__subbyte.html">integer_subbyte</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMma_3_01ElementA_00_01LayoutA_00_01kAlignmentA_00c67c16f9881e4f2fda76d8ed83ebabd6.html">DefaultMma&lt; ElementA, LayoutA, kAlignmentA, ElementB, LayoutB, kAlignmentB, ElementAccumulator, layout::RowMajor, arch::OpClassSimt, ArchTag, ThreadblockShape, WarpShape, InstructionShape, 2, Operator, false &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1thread_1_1LinearCombinationRelu.html">LinearCombinationRelu</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1thread.html">cutlass::epilogue::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape___00_01Element___00a6b756b1bcfbb35fe4a3e68ff074e380.html">PredicatedTileIterator&lt; Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, AccessSize &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1warp_1_1SimtPolicy.html">SimtPolicy</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1TypeTraits_3_01complex_3_01double_01_4_01_4_1_1integer__type.html">TypeTraits&lt; complex&lt; double &gt; &gt;::integer_type</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMma_3_01ElementA_00_01LayoutA_00_01kAlignmentA_00ce36642cae579bce6605ff8edde3c6ab.html">DefaultMma&lt; ElementA, LayoutA, kAlignmentA, ElementB, LayoutB, kAlignmentB, ElementAccumulator, layout::RowMajor, arch::OpClassTensorOp, ArchTag, ThreadblockShape, WarpShape, InstructionShape, 2, Operator, false &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1thread_1_1LinearCombinationRelu_3_01ElementOutput___00_01Count_00_01int_00_01float_00_01Round_01_4.html">LinearCombinationRelu&lt; ElementOutput_, Count, int, float, Round &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1thread.html">cutlass::epilogue::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape___00_01Element___004d0f9b5e19c29acc17bcdc360dafebbd.html">PredicatedTileIterator&lt; Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, AccessSize &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1warp_1_1SimtPolicy_3_01WarpShape___00_01Operator___00_01layout_1_1Rcef1c60e23e997017ae176c92931151d.html">SimtPolicy&lt; WarpShape_, Operator_, layout::RowMajor, MmaSimtPolicy_ &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1integral__constant.html">integral_constant</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMma_3_01int8__t_00_01LayoutA_00_01kAlignmentA_00_07e7230d4011ada5e22cfcb29103b696.html">DefaultMma&lt; int8_t, LayoutA, kAlignmentA, int8_t, LayoutB, kAlignmentB, ElementAccumulator, layout::RowMajor, arch::OpClassSimt, ArchTag, ThreadblockShape, WarpShape, GemmShape&lt; 1, 1, 4 &gt;, 2, Operator, false &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_M"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;M&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape___00_01Element___00ebd1a63351e1085d0b718582ec7b06c8.html">PredicatedTileIterator&lt; Shape_, Element_, layout::ColumnMajorInterleaved&lt; InterleavedK &gt;, AdvanceRank, ThreadMap_, AccessSize &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Sm50.html">Sm50</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__arithmetic.html">is_arithmetic</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore.html">DefaultMmaCore</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape___00_01Element___009fd89f6dad84238fd7d63df0a0c0364f.html">PredicatedTileIterator&lt; Shape_, Element_, layout::RowMajorInterleaved&lt; InterleavedK &gt;, AdvanceRank, ThreadMap_, AccessSize &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Sm60.html">Sm60</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__base__of.html">is_base_of</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShab94a11a77dd0565102710907089acee0.html">DefaultMmaCore&lt; Shape_, WarpShape_, GemmShape&lt; 1, 1, 1 &gt;, ElementA_, layout::ColumnMajor, ElementB_, layout::ColumnMajor, ElementC_, LayoutC_, arch::OpClassSimt, 2, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1library_1_1Manifest.html">Manifest</a> (<a class="el" href="namespacecutlass_1_1library.html">cutlass::library</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator2dThreadTile_3_01Shape___00_0145ef045e8f7d57dc718098adcb00cf3d.html">PredicatedTileIterator2dThreadTile&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, Transpose_ &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Sm61.html">Sm61</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__base__of__helper.html">is_base_of_helper</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShafafd5c61db86cbfe90863578ddd11092.html">DefaultMmaCore&lt; Shape_, WarpShape_, GemmShape&lt; 1, 1, 1 &gt;, ElementA_, layout::ColumnMajor, ElementB_, layout::RowMajor, ElementC_, LayoutC_, arch::OpClassSimt, 2, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1PredicatedTileIterator_1_1Mask.html">PredicatedTileIterator::Mask</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator2dThreadTile_3_01Shape___00_01e11ed7192af5d7ad1bce5641fa13112e.html">PredicatedTileIterator2dThreadTile&lt; Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, Transpose_ &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Sm70.html">Sm70</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__floating__point.html">is_floating_point</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha46446d1e3871e31d2e728f710d78c8c1.html">DefaultMmaCore&lt; Shape_, WarpShape_, GemmShape&lt; 1, 1, 1 &gt;, ElementA_, layout::ColumnMajor, ElementB_, layout::RowMajor, ElementC_, LayoutC_, arch::OpClassSimt, 2, Operator_, &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1InterleavedPredicatedTileIterator_1_1Mask.html">InterleavedPredicatedTileIterator::Mask</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator2dThreadTile_3_01Shape___00_0102e766863c6ac9ec2063a02c4803eecb.html">PredicatedTileIterator2dThreadTile&lt; Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, Transpose_ &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Sm72.html">Sm72</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__fundamental.html">is_fundamental</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha8da7a0cfbbe859b701fdd9f2b8566aa7.html">DefaultMmaCore&lt; Shape_, WarpShape_, GemmShape&lt; 1, 1, 1 &gt;, ElementA_, layout::RowMajor, ElementB_, layout::ColumnMajor, ElementC_, LayoutC_, arch::OpClassSimt, 2, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1library_1_1MathInstructionDescription.html">MathInstructionDescription</a> (<a class="el" href="namespacecutlass_1_1library.html">cutlass::library</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1thread_1_1LinearCombination_1_1Params.html">LinearCombination::Params</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1thread.html">cutlass::epilogue::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Sm75.html">Sm75</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__integral.html">is_integral</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha84e9f8afb6a4ca9f5dcd219b182d16e7.html">DefaultMmaCore&lt; Shape_, WarpShape_, GemmShape&lt; 1, 1, 1 &gt;, ElementA_, layout::RowMajor, ElementB_, layout::RowMajor, ElementC_, LayoutC_, arch::OpClassSimt, 2, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1thread_1_1Matrix.html">Matrix</a> (<a class="el" href="namespacecutlass_1_1thread.html">cutlass::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator_3_01Shape___00_01Elemenc07b5ec72f83e782121ac629288d61fe.html">PredicatedTileAccessIterator&lt; Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, AccessType_ &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1SubbyteReference.html">SubbyteReference</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__integral_3_01char_01_4.html">is_integral&lt; char &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha2c0d0b7cdb5c4bcb11e83c058eb65345.html">DefaultMmaCore&lt; Shape_, WarpShape_, GemmShape&lt; 1, 1, 4 &gt;, int8_t, layout::ColumnMajor, int8_t, layout::ColumnMajor, ElementC_, LayoutC_, arch::OpClassSimt, 2, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1MatrixCoord.html">MatrixCoord</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator2dThreadTile_3_01Shape__18e9cf25bb3b8edfaad595241a6dc2d7.html">PredicatedTileAccessIterator2dThreadTile&lt; Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, AccessType_ &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_T"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;T&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__integral_3_01const_01T_01_4.html">is_integral&lt; const T &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha34a52cc7b2942e8c290f0032b6779b52.html">DefaultMmaCore&lt; Shape_, WarpShape_, GemmShape&lt; 1, 1, 4 &gt;, int8_t, layout::ColumnMajor, int8_t, layout::RowMajor, ElementC_, LayoutC_, arch::OpClassSimt, 2, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1MatrixShape.html">MatrixShape</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1GemmSplitKParallel_1_1Params.html">GemmSplitKParallel::Params</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__integral_3_01const_01volatile_01T_01_4.html">is_integral&lt; const volatile T &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShaaf312aafe9da92ea9d417bcc12a8e7dc.html">DefaultMmaCore&lt; Shape_, WarpShape_, GemmShape&lt; 1, 1, 4 &gt;, int8_t, layout::RowMajor, int8_t, layout::ColumnMajor, ElementC_, LayoutC_, arch::OpClassSimt, 2, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1Max.html">Max</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1Gemm_1_1Params.html">Gemm::Params</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1Tensor4DCoord.html">Tensor4DCoord</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__integral_3_01int_01_4.html">is_integral&lt; int &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha863d4139ccaa713bc4bde32c425f4067.html">DefaultMmaCore&lt; Shape_, WarpShape_, GemmShape&lt; 1, 1, 4 &gt;, int8_t, layout::RowMajor, int8_t, layout::RowMajor, ElementC_, LayoutC_, arch::OpClassSimt, 2, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1Min.html">Min</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1Params.html">GemmBatched::Params</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1kernel.html">cutlass::gemm::kernel</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1TensorContainsFunc.html">TensorContainsFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__integral_3_01long_01_4.html">is_integral&lt; long &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShaf03a122202ad10acdc96f280106d678b.html">DefaultMmaCore&lt; Shape_, WarpShape_, GemmShape&lt; 8, 8, 4 &gt;, ElementA_, layout::ColumnMajor, ElementB_, layout::ColumnMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma.html">Mma</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1RandomGaussianFunc_1_1Params.html">RandomGaussianFunc::Params</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1TensorCopyDiagonalInFunc.html">TensorCopyDiagonalInFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__integral_3_01long_01long_01_4.html">is_integral&lt; long long &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha69bef08ea63dd930f99d9788105873dd.html">DefaultMmaCore&lt; Shape_, WarpShape_, GemmShape&lt; 8, 8, 4 &gt;, ElementA_, layout::ColumnMajor, ElementB_, layout::RowMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1Mma.html">Mma</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread.html">cutlass::gemm::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1TensorFillRandomGaussianFunc_1_1Params.html">TensorFillRandomGaussianFunc::Params</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1TensorCopyDiagonalOutFunc.html">TensorCopyDiagonalOutFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__integral_3_01short_01_4.html">is_integral&lt; short &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha3adf608332a8c9ee7014fced0da8a9ca.html">DefaultMmaCore&lt; Shape_, WarpShape_, GemmShape&lt; 8, 8, 4 &gt;, ElementA_, layout::RowMajor, ElementB_, layout::ColumnMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_011_00_011_00_011_01_4_00_011_00_01complex_30fa42e1ad201df010637cd22fc070a1.html">Mma&lt; gemm::GemmShape&lt; 1, 1, 1 &gt;, 1, complex&lt; double &gt;, LayoutA, complex&lt; double &gt;, LayoutB, complex&lt; double &gt;, LayoutC, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1RandomUniformFunc_1_1Params.html">RandomUniformFunc::Params</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1TensorCopyIf.html">TensorCopyIf</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__integral_3_01signed_01char_01_4.html">is_integral&lt; signed char &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShab7edfba3cdf43a07e3c4d719d87565a4.html">DefaultMmaCore&lt; Shape_, WarpShape_, GemmShape&lt; 8, 8, 4 &gt;, ElementA_, layout::RowMajor, ElementB_, layout::RowMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_011_00_011_00_011_01_4_00_011_00_01complex_48b3a43bc03fff93a111ac01abe7e40d.html">Mma&lt; gemm::GemmShape&lt; 1, 1, 1 &gt;, 1, complex&lt; double &gt;, LayoutA, double, LayoutB, complex&lt; double &gt;, LayoutC, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1TensorFillRandomUniformFunc_1_1Params.html">TensorFillRandomUniformFunc::Params</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1layout_1_1TensorCxRSKx.html">TensorCxRSKx</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__integral_3_01unsigned_01char_01_4.html">is_integral&lt; unsigned char &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instruc803d38bc1e4618c07c47f54c87ae2678.html">DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::ColumnMajor, ElementB_, layout::ColumnMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_011_00_011_00_011_01_4_00_011_00_01complex_76f9d24016e1b4167b16f4d7628c9546.html">Mma&lt; gemm::GemmShape&lt; 1, 1, 1 &gt;, 1, complex&lt; float &gt;, LayoutA, complex&lt; float &gt;, LayoutB, complex&lt; float &gt;, LayoutC, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1TensorFillDiagonalFunc_1_1Params.html">TensorFillDiagonalFunc::Params</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1library_1_1TensorDescription.html">TensorDescription</a> (<a class="el" href="namespacecutlass_1_1library.html">cutlass::library</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__integral_3_01unsigned_01int_01_4.html">is_integral&lt; unsigned int &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instrucf60fe02fcdd80d28b7fd419133465dcc.html">DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::ColumnMajor, ElementB_, layout::RowMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_011_00_011_00_011_01_4_00_011_00_01complex_f1c9d2ee842455cd0c5b71d56108d468.html">Mma&lt; gemm::GemmShape&lt; 1, 1, 1 &gt;, 1, complex&lt; float &gt;, LayoutA, float, LayoutB, complex&lt; float &gt;, LayoutC, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1TensorUpdateDiagonalFunc_1_1Params.html">TensorUpdateDiagonalFunc::Params</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1TensorDiagonalForEach.html">TensorDiagonalForEach</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device.html">cutlass::reference::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__integral_3_01unsigned_01long_01_4.html">is_integral&lt; unsigned long &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instruc2bf00737f4ad0a9da9a8be6d3e66c152.html">DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::ColumnMajorInterleaved&lt; InterleavedK &gt;, ElementB_, layout::RowMajorInterleaved&lt; InterleavedK &gt;, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_, AccumulatorsInRowMajor &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_011_00_011_00_011_01_4_00_011_00_01double_070b94670e040ed5855e5b42d5ca8a443.html">Mma&lt; gemm::GemmShape&lt; 1, 1, 1 &gt;, 1, double, LayoutA, complex&lt; double &gt;, LayoutB, complex&lt; double &gt;, LayoutC, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1TensorFillLinearFunc_1_1Params.html">TensorFillLinearFunc::Params</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1TensorEqualsFunc.html">TensorEqualsFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__integral_3_01unsigned_01long_01long_01_4.html">is_integral&lt; unsigned long long &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instruc24092ddc01fc83dabb7db4c14880fe60.html">DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::RowMajor, ElementB_, layout::ColumnMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_011_00_011_00_011_01_4_00_011_00_01double_0aa57e6a2e6b5da37d10688bf99419a23.html">Mma&lt; gemm::GemmShape&lt; 1, 1, 1 &gt;, 1, double, LayoutA, double, LayoutB, double, LayoutC, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1TensorCopyDiagonalInFunc_1_1Params.html">TensorCopyDiagonalInFunc::Params</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1TensorFillDiagonalFunc.html">TensorFillDiagonalFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__integral_3_01unsigned_01short_01_4.html">is_integral&lt; unsigned short &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instruc4fee9f2965b8468bfb42b94a74527d22.html">DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::RowMajor, ElementB_, layout::RowMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1threadblock.html">cutlass::gemm::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_011_00_011_00_011_01_4_00_011_00_01ElementAb6e65b2cf5ede7f41cb070a767158dee.html">Mma&lt; gemm::GemmShape&lt; 1, 1, 1 &gt;, 1, ElementA, LayoutA, ElementB, LayoutB, ElementC, LayoutC, Operator &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1TensorCopyDiagonalOutFunc_1_1Params.html">TensorCopyDiagonalOutFunc::Params</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1TensorFillDiagonalFunc.html">TensorFillDiagonalFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__integral_3_01volatile_01T_01_4.html">is_integral&lt; volatile T &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1warp_1_1DefaultMmaTensorOp.html">DefaultMmaTensorOp</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_011_00_011_00_011_01_4_00_011_00_01float_00e3e12e263df6506b8cf06c3f4d478b8e.html">Mma&lt; gemm::GemmShape&lt; 1, 1, 1 &gt;, 1, float, LayoutA, complex&lt; float &gt;, LayoutB, complex&lt; float &gt;, LayoutC, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1thread_1_1LinearCombinationClamp_1_1Params.html">LinearCombinationClamp::Params</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1thread.html">cutlass::epilogue::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1TensorFillFunc.html">TensorFillFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__pointer.html">is_pointer</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultThreadMapSimt.html">DefaultThreadMapSimt</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_011_00_011_00_011_01_4_00_011_00_01float_004bb3fd76ca2af7b3210676fa9644d95b.html">Mma&lt; gemm::GemmShape&lt; 1, 1, 1 &gt;, 1, float, LayoutA, float, LayoutB, float, LayoutC, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1thread_1_1LinearCombinationRelu_1_1Params.html">LinearCombinationRelu::Params</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1thread.html">cutlass::epilogue::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1TensorFillGaussianFunc.html">TensorFillGaussianFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__pointer__helper.html">is_pointer_helper</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultThreadMapTensorOp.html">DefaultThreadMapTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_011_00_011_00_011_01_4_00_011_00_01half__t_4f30ee91f7bb3844ff7579c68d078818.html">Mma&lt; gemm::GemmShape&lt; 1, 1, 1 &gt;, 1, half_t, LayoutA, half_t, LayoutB, float, LayoutC, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1thread_1_1LinearCombinationRelu_3_01ElementOutput___00_01Count_00_00274a94522c46cd041d0b10d484e2ef3.html">LinearCombinationRelu&lt; ElementOutput_, Count, int, float, Round &gt;::Params</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1thread.html">cutlass::epilogue::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1TensorFillLinearFunc.html">TensorFillLinearFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__pointer__helper_3_01T_01_5_01_4.html">is_pointer_helper&lt; T * &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultThreadMapVoltaTensorOp.html">DefaultThreadMapVoltaTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_011_00_011_00_011_01_4_00_011_00_01int_00_00b2dff9ce8caad9aff5bc6a355539161.html">Mma&lt; gemm::GemmShape&lt; 1, 1, 1 &gt;, 1, int, LayoutA, int, LayoutB, int, LayoutC, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1thread_1_1ReductionOpPlus_1_1Params.html">ReductionOpPlus::Params</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1thread.html">cutlass::epilogue::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1TensorFillLinearFunc.html">TensorFillLinearFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1is__pow2.html">is_pow2</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultThreadMapVoltaTensorOp_3_01ThreadblockShape__95db04b7b72e34283958bd7fbf851d16.html">DefaultThreadMapVoltaTensorOp&lt; ThreadblockShape_, WarpShape_, PartitionsK, ElementOutput_, ElementsPerAccess, float &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_011_00_011_00_012_01_4_00_011_00_01int16__t8c4bac365710598317a69c489f7239db.html">Mma&lt; gemm::GemmShape&lt; 1, 1, 2 &gt;, 1, int16_t, layout::RowMajor, int16_t, layout::ColumnMajor, int, LayoutC, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1TensorUpdateOffDiagonalFunc_1_1Params.html">TensorUpdateOffDiagonalFunc::Params</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1TensorFillRandomGaussianFunc.html">TensorFillRandomGaussianFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__same.html">is_same</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultThreadMapVoltaTensorOp_3_01ThreadblockShape__d58c94abc36b7c5c109b55202c6992e7.html">DefaultThreadMapVoltaTensorOp&lt; ThreadblockShape_, WarpShape_, PartitionsK, ElementOutput_, ElementsPerAccess, half_t &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_011_00_011_00_014_01_4_00_011_00_01int8__t_a1ef6624fc8c10126f17f4ee88283d72.html">Mma&lt; gemm::GemmShape&lt; 1, 1, 4 &gt;, 1, int8_t, LayoutA, int8_t, LayoutB, int, LayoutC, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DirectEpilogueTensorOp_1_1Params.html">DirectEpilogueTensorOp::Params</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1TensorFillRandomUniformFunc.html">TensorFillRandomUniformFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__same_3_01A_00_01A_01_4.html">is_same&lt; A, A &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultThreadMapWmmaTensorOp.html">DefaultThreadMapWmmaTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_011_00_012_00_011_01_4_00_011_00_01half__t_f3dc2e59f857ada163d1e0781ea8f391.html">Mma&lt; gemm::GemmShape&lt; 1, 2, 1 &gt;, 1, half_t, LayoutA, half_t, LayoutB, half_t, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1layout_1_1PitchLinear.html">PitchLinear</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1TensorFillRandomUniformFunc.html">TensorFillRandomUniformFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__trivially__copyable.html">is_trivially_copyable</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1detail_1_1RowArrangement_3_01Shape_00_01WarpsRemainief28e98b3f284469f271d28aba73de2e.html">RowArrangement&lt; Shape, WarpsRemaining, ElementsPerAccess, ElementSize, true &gt;::Detail</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock_1_1detail.html">cutlass::epilogue::threadblock::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_0116_00_0116_00_014_01_4_00_0132_00_01half_0bcc4d05f9811035f08cc1b7f0154a4d.html">Mma&lt; gemm::GemmShape&lt; 16, 16, 4 &gt;, 32, half_t, LayoutA, half_t, LayoutB, ElementC, LayoutC, Operator &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1PitchLinear2DThreadTileStripminedThreadMap.html">PitchLinear2DThreadTileStripminedThreadMap</a> (<a class="el" href="namespacecutlass_1_1transform.html">cutlass::transform</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1TensorForEach.html">TensorForEach</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device.html">cutlass::reference::device</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__void.html">is_void</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1OutputTileOptimalThreadMap_1_1Detail.html">OutputTileOptimalThreadMap::Detail</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_0116_00_018_00_018_01_4_00_0132_00_01half__02a3f19a78995f97d793a668e0e4d4f0.html">Mma&lt; gemm::GemmShape&lt; 16, 8, 8 &gt;, 32, half_t, layout::RowMajor, half_t, layout::ColumnMajor, float, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1PitchLinear2DThreadTileStripminedThreadMap_3_01Shape___00_01Thread0082c3467229b12cc9dd996283ee7160.html">PitchLinear2DThreadTileStripminedThreadMap&lt; Shape_, Threads, cutlass::layout::PitchLinearShape&lt; 4, 4 &gt; &gt;</a> (<a class="el" href="namespacecutlass_1_1transform.html">cutlass::transform</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1kernel_1_1detail_1_1TensorForEachHelper.html">TensorForEachHelper</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1kernel_1_1detail.html">cutlass::reference::device::kernel::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__volatile.html">is_volatile</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1InterleavedOutputTileThreadMap_1_1Detail.html">InterleavedOutputTileThreadMap::Detail</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_0116_00_018_00_018_01_4_00_0132_00_01half__96363097c47b056f0ca1911afd7f8b7a.html">Mma&lt; gemm::GemmShape&lt; 16, 8, 8 &gt;, 32, half_t, layout::RowMajor, half_t, layout::ColumnMajor, half_t, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1PitchLinearCoord.html">PitchLinearCoord</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1TensorForEachHelper.html">TensorForEachHelper</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1is__volatile_3_01volatile_01T_01_4.html">is_volatile&lt; volatile T &gt;</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1warp_1_1TileIteratorTensorOp_3_01WarpShape___00_01OperatorShape___05f11e023c9e6ee5f7a888fa4c5bbf6d1.html">TileIteratorTensorOp&lt; WarpShape_, OperatorShape_, Element_, layout::RowMajor &gt;::Detail</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_012_00_011_00_011_01_4_00_011_00_01half__t_8cf78649807b93684f3d431bfa34ee28.html">Mma&lt; gemm::GemmShape&lt; 2, 1, 1 &gt;, 1, half_t, LayoutA, half_t, LayoutB, half_t, LayoutC, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1PitchLinearShape.html">PitchLinearShape</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1kernel_1_1detail_1_1TensorForEachHelper_3_01Func_00_01Rank_00_010_01_4.html">TensorForEachHelper&lt; Func, Rank, 0 &gt;</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1kernel_1_1detail.html">cutlass::reference::device::kernel::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1Array_3_01T_00_01N_00_01true_01_4_1_1iterator.html">Array&lt; T, N, true &gt;::iterator</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1warp_1_1TileIteratorVoltaTensorOp_3_01WarpShape___00_01gemm_1_1Gemmffcab2297c8de8d0013602a39c525b78.html">TileIteratorVoltaTensorOp&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, half_t, layout::RowMajor &gt;::Detail</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_012_00_012_00_011_01_4_00_011_00_01half__t_ccde11d1bbbdab3702772ce44eb9729a.html">Mma&lt; gemm::GemmShape&lt; 2, 2, 1 &gt;, 1, half_t, layout::ColumnMajor, half_t, layout::RowMajor, half_t, layout::ColumnMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1PitchLinearStripminedThreadMap.html">PitchLinearStripminedThreadMap</a> (<a class="el" href="namespacecutlass_1_1transform.html">cutlass::transform</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1TensorForEachHelper_3_01Func_00_01Rank_00_010_01_4.html">TensorForEachHelper&lt; Func, Rank, 0 &gt;</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1Array_3_01T_00_01N_00_01false_01_4_1_1iterator.html">Array&lt; T, N, false &gt;::iterator</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1warp_1_1TileIteratorVoltaTensorOp_3_01WarpShape___00_01gemm_1_1Gemm770cbca45441d295d5d7433e8222a700.html">TileIteratorVoltaTensorOp&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, float, layout::RowMajor &gt;::Detail</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_012_00_012_00_011_01_4_00_011_00_01half__t_c07cc6439298fa5486a719e577be2538.html">Mma&lt; gemm::GemmShape&lt; 2, 2, 1 &gt;, 1, half_t, layout::ColumnMajor, half_t, layout::RowMajor, half_t, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1PitchLinearTilePolicyStripminedThreadContiguous.html">PitchLinearTilePolicyStripminedThreadContiguous</a> (<a class="el" href="namespacecutlass_1_1transform.html">cutlass::transform</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1TensorFuncBinaryOp.html">TensorFuncBinaryOp</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_l"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;l&#160;&#160;</div></td></tr></table>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</td></tr>
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.
2019-11-20 08:55:34 +08:00
<tr><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1PitchLinearStripminedThreadMap_1_1Detail.html">PitchLinearStripminedThreadMap::Detail</a> (<a class="el" href="namespacecutlass_1_1transform.html">cutlass::transform</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_01128_01_4_00_0132_00_01uint15918972b95027764b3a849b03075ed2b.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 128 &gt;, 32, uint1b_t, layout::RowMajor, uint1b_t, layout::ColumnMajor, int, layout::RowMajor, OpXorPopc &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1PitchLinearTilePolicyStripminedThreadStrided.html">PitchLinearTilePolicyStripminedThreadStrided</a> (<a class="el" href="namespacecutlass_1_1transform.html">cutlass::transform</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1layout_1_1TensorNCHW.html">TensorNCHW</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1PitchLinearWarpRakedThreadMap_1_1Detail.html">PitchLinearWarpRakedThreadMap::Detail</a> (<a class="el" href="namespacecutlass_1_1transform.html">cutlass::transform</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_0116_01_4_00_0132_00_01int8__927179f46017ea5f58f859f1196c4829.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 16 &gt;, 32, int8_t, layout::RowMajor, int8_t, layout::ColumnMajor, int, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1PitchLinearWarpRakedThreadMap.html">PitchLinearWarpRakedThreadMap</a> (<a class="el" href="namespacecutlass_1_1transform.html">cutlass::transform</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1layout_1_1TensorNCxHWx.html">TensorNCxHWx</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1log2__down.html">log2_down</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1TransposePitchLinearThreadMap_1_1Detail.html">TransposePitchLinearThreadMap::Detail</a> (<a class="el" href="namespacecutlass_1_1transform.html">cutlass::transform</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_0116_01_4_00_0132_00_01int8__8ebae0cbdf333fddfe5c24d35ebe8e02.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 16 &gt;, 32, int8_t, layout::RowMajor, int8_t, layout::ColumnMajor, int, layout::RowMajor, OpMultiplyAddSaturate &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1PitchLinearWarpStripedThreadMap.html">PitchLinearWarpStripedThreadMap</a> (<a class="el" href="namespacecutlass_1_1transform.html">cutlass::transform</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1layout_1_1TensorNHWC.html">TensorNHWC</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1log2__down_3_01N_00_011_00_01Count_01_4.html">log2_down&lt; N, 1, Count &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1PitchLinearWarpStripedThreadMap_1_1Detail.html">PitchLinearWarpStripedThreadMap::Detail</a> (<a class="el" href="namespacecutlass_1_1transform.html">cutlass::transform</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_0116_01_4_00_0132_00_01int8__5299c9c90c8f2f521be0c8cec1c3eb08.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 16 &gt;, 32, int8_t, layout::RowMajor, uint8_t, layout::ColumnMajor, int, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Opera33cdf53848564e894d4407637dc86caf.html">MmaVoltaTensorOpMultiplicandTileIterator&lt; Shape_, Operand::kA, Element_, cutlass::layout::VoltaTensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, InstructionShape_, OpDelta_, 32 &gt;::Policy</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1TensorOpMultiplicand.html">TensorOpMultiplicand</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1log2__up.html">log2_up</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1PitchLinear2DThreadTileStripminedThreadMap_3_01Shape___00_01Thread896c01a3c466da1bf392e0cdfced4d53.html">PitchLinear2DThreadTileStripminedThreadMap&lt; Shape_, Threads, cutlass::layout::PitchLinearShape&lt; 4, 4 &gt; &gt;::Detail</a> (<a class="el" href="namespacecutlass_1_1transform.html">cutlass::transform</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_0116_01_4_00_0132_00_01int8__f083347e265b1e9eea5572d86ddb6bf9.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 16 &gt;, 32, int8_t, layout::RowMajor, uint8_t, layout::ColumnMajor, int, layout::RowMajor, OpMultiplyAddSaturate &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1warp_1_1MmaTensorOpAccumulatorTileIterator_3_01Shape___00_01Element___093b5d2838ac5a742704ef62b5c8688f0.html">MmaTensorOpAccumulatorTileIterator&lt; Shape_, Element_, cutlass::layout::RowMajor, InstructionShape_, OpDelta_ &gt;::Policy</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1TensorOpMultiplicandColumnMajorInterleaved.html">TensorOpMultiplicandColumnMajorInterleaved</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1log2__up_3_01N_00_011_00_01Count_01_4.html">log2_up&lt; N, 1, Count &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element_0a9491607d11be8e1780e79ad711aa42.html">RegularTileAccessIterator&lt; Shape_, Element_, layout::TensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value, int(128/sizeof(Element_))&gt;, AdvanceRank, ThreadMap_, Alignment &gt;::Detail</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_0116_01_4_00_0132_00_01uint8_a62aa63a212985df306fb27e8a50aeae.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 16 &gt;, 32, uint8_t, layout::RowMajor, int8_t, layout::ColumnMajor, int, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operafa294175b280756dd8388f9ffe7b72c4.html">MmaVoltaTensorOpMultiplicandTileIterator&lt; Shape_, Operand_, Element_, cutlass::layout::VoltaTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, KBlock &gt;, InstructionShape_, OpDelta_, 32 &gt;::Policy</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1TensorOpMultiplicandCongruous.html">TensorOpMultiplicandCongruous</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_m"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;m&#160;&#160;</div></td></tr></table>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</td></tr>
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.
2019-11-20 08:55:34 +08:00
<tr><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element_3be8b96d170d886f39b6b30acab65e7a.html">RegularTileAccessIterator&lt; Shape_, Element_, layout::TensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, AdvanceRank, ThreadMap_, Alignment &gt;::Detail</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_0116_01_4_00_0132_00_01uint8_ab741d81fdc991345cb9e43c29fca573.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 16 &gt;, 32, uint8_t, layout::RowMajor, int8_t, layout::ColumnMajor, int, layout::RowMajor, OpMultiplyAddSaturate &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___0784c74bd670999ec23ad8ef9dc55777.html">MmaTensorOpMultiplicandTileIterator&lt; Shape_, Operand_, Element_, cutlass::layout::TensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, InstructionShape_, OpDelta_, 32, PartitionsK_ &gt;::Policy</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1TensorOpMultiplicandCongruous_3_0132_00_01Crosswise_01_4.html">TensorOpMultiplicandCongruous&lt; 32, Crosswise &gt;</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_052caec9d5bceeb59b9a13cb3338ce64d.html">RegularTileIterator&lt; Shape_, Element_, layout::TensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value, int(128/sizeof(Element_))&gt;, AdvanceRank, ThreadMap_, Alignment &gt;::Detail</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_0116_01_4_00_0132_00_01uint8_5221708cec5828d35db1d1c47cb4964e.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 16 &gt;, 32, uint8_t, layout::RowMajor, uint8_t, layout::ColumnMajor, int, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___07638f8b7761f6e2e2e6918e2c05e739.html">MmaTensorOpMultiplicandTileIterator&lt; Shape_, Operand_, Element_, cutlass::layout::TensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value, 64 &gt;, InstructionShape_, OpDelta_, 32, PartitionsK_ &gt;::Policy</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1TensorOpMultiplicandCrosswise.html">TensorOpMultiplicandCrosswise</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1maximum.html">maximum</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_039093927f4b1ee61538c569bf1ae4efd.html">RegularTileIterator&lt; Shape_, Element_, layout::TensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, AdvanceRank, ThreadMap_, Alignment &gt;::Detail</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_0116_01_4_00_0132_00_01uint8_bef0c048bc0f8ba2d875cb7ab26d363b.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 16 &gt;, 32, uint8_t, layout::RowMajor, uint8_t, layout::ColumnMajor, int, layout::RowMajor, OpMultiplyAddSaturate &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1warp_1_1MmaTensorOpAccumulatorTileIterator_3_01Shape___00_01Element___0d35fa5dc4e4b4f72784c943fd857fc1d.html">MmaTensorOpAccumulatorTileIterator&lt; Shape_, Element_, cutlass::layout::ColumnMajor, InstructionShape_, OpDelta_ &gt;::Policy</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1layout_1_1TensorOpMultiplicandRowMajorInterleaved.html">TensorOpMultiplicandRowMajorInterleaved</a> (<a class="el" href="namespacecutlass_1_1layout.html">cutlass::layout</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1maximum_3_01Array_3_01T_00_01N_01_4_01_4.html">maximum&lt; Array&lt; T, N &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_032f88d1be8b209e44a4815c707ba35bb.html">RegularTileIterator&lt; Shape_, Element_, layout::VoltaTensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, AdvanceRank, ThreadMap_, Alignment &gt;::Detail</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_0132_01_4_00_0132_00_01int4b_6e513ccbc44ae7909a60d93b9b5435b3.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 32 &gt;, 32, int4b_t, layout::RowMajor, int4b_t, layout::ColumnMajor, int, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1warp_1_1MmaTensorOpAccumulatorTileIterator_3_01Shape___00_01Element___03822d9be37f3725022005a5434441f22.html">MmaTensorOpAccumulatorTileIterator&lt; Shape_, Element_, cutlass::layout::ColumnMajorInterleaved&lt; InterleavedN &gt;, InstructionShape_, OpDelta_ &gt;::Policy</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1warp_1_1TensorOpPolicy.html">TensorOpPolicy</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1maximum_3_01float_01_4.html">maximum&lt; float &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_02d305cfb0b55c6fb236a52cf2240651e.html">RegularTileIterator&lt; Shape_, Element_, layout::VoltaTensorOpMultiplicandBCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, AdvanceRank, ThreadMap_, Alignment &gt;::Detail</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_0132_01_4_00_0132_00_01int4b_0ee08a4520882d24ba9026879265e892.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 32 &gt;, 32, int4b_t, layout::RowMajor, int4b_t, layout::ColumnMajor, int, layout::RowMajor, OpMultiplyAddSaturate &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Opera6fa6d2d3725bb3ec613d5c527ea3ffe7.html">MmaVoltaTensorOpMultiplicandTileIterator&lt; Shape_, Operand::kB, Element_, cutlass::layout::VoltaTensorOpMultiplicandBCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, InstructionShape_, OpDelta_, 32 &gt;::Policy</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1warp_1_1TensorOpPolicy_3_01WarpShape_00_01OperatorShape_00_01layout69549d10c3610d943987eb90e827bc05.html">TensorOpPolicy&lt; WarpShape, OperatorShape, layout::ColumnMajorInterleaved&lt; InterleavedK &gt; &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1minimum.html">minimum</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_0390833403016f5d817416e20828845df.html">RegularTileIterator&lt; Shape_, Element_, layout::VoltaTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Shape_::kContiguous &gt;, AdvanceRank, ThreadMap_, Alignment &gt;::Detail</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_0132_01_4_00_0132_00_01int4b_4746fc55e614df0016c518d3fda2677e.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 32 &gt;, 32, int4b_t, layout::RowMajor, uint4b_t, layout::ColumnMajor, int, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpAccumulatorTileIterator_1_1Policy.html">MmaVoltaTensorOpAccumulatorTileIterator::Policy</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1warp.html">cutlass::gemm::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1warp_1_1TensorOpPolicy_3_01WarpShape_00_01OperatorShape_00_01layout_1_1RowMajor_01_4.html">TensorOpPolicy&lt; WarpShape, OperatorShape, layout::RowMajor &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1minimum_3_01Array_3_01T_00_01N_01_4_01_4.html">minimum&lt; Array&lt; T, N &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultThreadMapSimt_1_1Detail.html">DefaultThreadMapSimt::Detail</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_0132_01_4_00_0132_00_01int4b_546e9ec6de6a5970b326da6f6280f1d4.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 32 &gt;, 32, int4b_t, layout::RowMajor, uint4b_t, layout::ColumnMajor, int, layout::RowMajor, OpMultiplyAddSaturate &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator.html">PredicatedTileAccessIterator</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1TensorRef.html">TensorRef</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1minimum_3_01float_01_4.html">minimum&lt; float &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultThreadMapTensorOp_1_1Detail.html">DefaultThreadMapTensorOp::Detail</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_0132_01_4_00_0132_00_01uint4b03e3b50dbcb30d0d1ac062f3a9d5abef.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 32 &gt;, 32, uint4b_t, layout::RowMajor, int4b_t, layout::ColumnMajor, int, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator2dThreadTile.html">PredicatedTileAccessIterator2dThreadTile</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1TensorUpdateDiagonalFunc.html">TensorUpdateDiagonalFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1minus.html">minus</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultInterleavedThreadMapTensorOp_1_1Detail.html">DefaultInterleavedThreadMapTensorOp::Detail</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_0132_01_4_00_0132_00_01uint4b6d968039dde5c9f062ab15f90a8049fe.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 32 &gt;, 32, uint4b_t, layout::RowMajor, int4b_t, layout::ColumnMajor, int, layout::RowMajor, OpMultiplyAddSaturate &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator2dThreadTile_3_01Shape__da632779aba661c0f4cfaaa78126b771.html">PredicatedTileAccessIterator2dThreadTile&lt; Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, AccessType_ &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1TensorUpdateOffDiagonalFunc.html">TensorUpdateOffDiagonalFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1device_1_1detail.html">cutlass::reference::device::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1minus_3_01Array_3_01half__t_00_01N_01_4_01_4.html">minus&lt; Array&lt; half_t, N &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultThreadMapVoltaTensorOp_3_01ThreadblockShape__4433cc988100e98097a748d2670fb0fc.html">DefaultThreadMapVoltaTensorOp&lt; ThreadblockShape_, WarpShape_, PartitionsK, ElementOutput_, ElementsPerAccess, half_t &gt;::Detail</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_0132_01_4_00_0132_00_01uint4bc4b6ba004e25c44bfd9266c61f937dfb.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 32 &gt;, 32, uint4b_t, layout::RowMajor, uint4b_t, layout::ColumnMajor, int, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator2dThreadTile_3_01Shape__1790abaa54a01f277d75766d5882fec8.html">PredicatedTileAccessIterator2dThreadTile&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, AccessType_ &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1TensorUpdateOffDiagonalFunc.html">TensorUpdateOffDiagonalFunc</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1minus_3_01Array_3_01T_00_01N_01_4_01_4.html">minus&lt; Array&lt; T, N &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultThreadMapVoltaTensorOp_3_01ThreadblockShape__52116c60c62f0fd520071558e42b814f.html">DefaultThreadMapVoltaTensorOp&lt; ThreadblockShape_, WarpShape_, PartitionsK, ElementOutput_, ElementsPerAccess, float &gt;::Detail</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_0132_01_4_00_0132_00_01uint4b451d5cf5d7e8cbbe476afe3dab5c09b2.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 32 &gt;, 32, uint4b_t, layout::RowMajor, uint4b_t, layout::ColumnMajor, int, layout::RowMajor, OpMultiplyAddSaturate &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator2dThreadTile_3_01Shape__7327fa15996bcb8502cdfcc192350fe1.html">PredicatedTileAccessIterator2dThreadTile&lt; Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, AccessType_ &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1TensorView.html">TensorView</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1multiplies.html">multiplies</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultThreadMapWmmaTensorOp_1_1Detail.html">DefaultThreadMapWmmaTensorOp::Detail</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_014_01_4_00_018_00_01half__t_b0242d7a01097510effbc4718040d3e5.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 4 &gt;, 8, half_t, layout::ColumnMajor, half_t, layout::ColumnMajor, float, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator_3_01Shape___00_01Elemen89c687c583745a73cb485041911a4c4e.html">PredicatedTileAccessIterator&lt; Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, AccessType_ &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1library_1_1TileDescription.html">TileDescription</a> (<a class="el" href="namespacecutlass_1_1library.html">cutlass::library</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1multiplies_3_01Array_3_01half__t_00_01N_01_4_01_4.html">multiplies&lt; Array&lt; half_t, N &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1threadblock_1_1DirectEpilogueTensorOp.html">DirectEpilogueTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_014_01_4_00_018_00_01half__t_c7f88bfd32a544fba8111d2dcadeab11.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 4 &gt;, 8, half_t, layout::ColumnMajor, half_t, layout::ColumnMajor, half_t, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator_3_01Shape___00_01Elemenab63a1e105bf37f6371516cb9e2c5a7a.html">PredicatedTileAccessIterator&lt; Shape_, Element_, layout::ColumnMajorInterleaved&lt; InterleavedK &gt;, AdvanceRank, ThreadMap_, AccessType_ &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1TileIteratorSimt.html">TileIteratorSimt</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1multiplies_3_01Array_3_01T_00_01N_01_4_01_4.html">multiplies&lt; Array&lt; T, N &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1Distribution.html">Distribution</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_014_01_4_00_018_00_01half__t_44a3b2a8df88a2b067f1284515cb5371.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 4 &gt;, 8, half_t, layout::ColumnMajor, half_t, layout::RowMajor, float, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator_3_01Shape___00_01Elemen784a0e9da3f55064c47e5613791f51f7.html">PredicatedTileAccessIterator&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, AccessType_ &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1TileIteratorSimt_3_01WarpShape___00_01Operator___00_01Elemenf2bd262ed3e202b25d5802d83965bf3b.html">TileIteratorSimt&lt; WarpShape_, Operator_, Element_, layout::RowMajor, MmaSimtPolicy_ &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1multiply__add.html">multiply_add</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td rowspan="2" valign="bottom"><a name="letter_E"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;E&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_014_01_4_00_018_00_01half__t_4b7308177b308a272c1889fbe9670275.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 4 &gt;, 8, half_t, layout::ColumnMajor, half_t, layout::RowMajor, half_t, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator_3_01Shape___00_01Elemen9838736ad62fae54213fbaf722a989ab.html">PredicatedTileAccessIterator&lt; Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, AccessType_ &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1TileIteratorTensorOp.html">TileIteratorTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1multiply__add_3_01Array_3_01half__t_00_01N_01_4_00_01Array_3_01half__t_00_01N_01adaeadb27c0e4439444709c0eb30963.html">multiply_add&lt; Array&lt; half_t, N &gt;, Array&lt; half_t, N &gt;, Array&lt; half_t, N &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_014_01_4_00_018_00_01half__t_5a9888862cebd333ecaf11f7262f77d4.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 4 &gt;, 8, half_t, layout::RowMajor, half_t, layout::ColumnMajor, float, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator_3_01Shape___00_01Elemen809793e785fb4211888c6b4e5dcfcb39.html">PredicatedTileAccessIterator&lt; Shape_, Element_, layout::RowMajorInterleaved&lt; InterleavedK &gt;, AdvanceRank, ThreadMap_, AccessType_ &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1TileIteratorTensorOp_3_01WarpShape___00_01OperatorShape___003cbb32beb84b4984cb7853662096d289.html">TileIteratorTensorOp&lt; WarpShape_, OperatorShape_, Element_, layout::RowMajor &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1multiply__add_3_01Array_3_01T_00_01N_01_4_00_01Array_3_01T_00_01N_01_4_00_01Array_3_01T_00_01N_01_4_01_4.html">multiply_add&lt; Array&lt; T, N &gt;, Array&lt; T, N &gt;, Array&lt; T, N &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1detail_1_1EnableMma__Crow__SM60.html">EnableMma_Crow_SM60</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread_1_1detail.html">cutlass::gemm::thread::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_014_01_4_00_018_00_01half__t_31defda8ea2b7d855642ffd77da1a411.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 4 &gt;, 8, half_t, layout::RowMajor, half_t, layout::ColumnMajor, half_t, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1threadblock_1_1PredicatedTileIterator.html">PredicatedTileIterator</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1epilogue_1_1warp_1_1TileIteratorVoltaTensorOp.html">TileIteratorVoltaTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1multiply__add_3_01complex_3_01T_01_4_00_01complex_3_01T_01_4_00_01complex_3_01T_01_4_01_4.html">multiply_add&lt; complex&lt; T &gt;, complex&lt; T &gt;, complex&lt; T &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1threadblock_1_1Epilogue.html">Epilogue</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_014_01_4_00_018_00_01half__t_839a7c8bb938d1661f4611e68f85d8cb.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 4 &gt;, 8, half_t, layout::RowMajor, half_t, layout::RowMajor, float, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator.html">PredicatedTileIterator</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1TileIteratorVoltaTensorOp_3_01WarpShape___00_01gemm_1_1GemmS2fe0c60b727c738c622c18fc3dd76644.html">TileIteratorVoltaTensorOp&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, float, layout::RowMajor &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1multiply__add_3_01complex_3_01T_01_4_00_01T_00_01complex_3_01T_01_4_01_4.html">multiply_add&lt; complex&lt; T &gt;, T, complex&lt; T &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1threadblock_1_1EpilogueBase.html">EpilogueBase</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1threadblock.html">cutlass::epilogue::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1Mma_3_01gemm_1_1GemmShape_3_018_00_018_00_014_01_4_00_018_00_01half__t_73d9802d6b944a5299bc255887db6bbc.html">Mma&lt; gemm::GemmShape&lt; 8, 8, 4 &gt;, 8, half_t, layout::RowMajor, half_t, layout::RowMajor, half_t, layout::RowMajor, OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator2dThreadTile.html">PredicatedTileIterator2dThreadTile</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1TileIteratorVoltaTensorOp_3_01WarpShape___00_01gemm_1_1GemmSa0ceeeddc22575876eb977da7f5416a8.html">TileIteratorVoltaTensorOp&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, half_t, layout::RowMajor &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1multiply__add_3_01T_00_01complex_3_01T_01_4_00_01complex_3_01T_01_4_01_4.html">multiply_add&lt; T, complex&lt; T &gt;, complex&lt; T &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1EpilogueWorkspace.html">EpilogueWorkspace</a> (<a class="el" href="namespacecutlass_1_1epilogue.html">cutlass::epilogue</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1Mma_3_01Shape___00_01ElementA___00_01LayoutA___00_01ElementB_e41c1cd6078b6d1347fac239b0639d56.html">Mma&lt; Shape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, arch::OpMultiplyAdd, bool &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread.html">cutlass::gemm::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator2dThreadTile_3_01Shape___00_0165b39a630d10785a3558406f9adb99b9.html">PredicatedTileIterator2dThreadTile&lt; Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, Transpose_ &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1TileIteratorWmmaTensorOp.html">TileIteratorWmmaTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_n"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;n&#160;&#160;</div></td></tr></table>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</td></tr>
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.
2019-11-20 08:55:34 +08:00
<tr><td rowspan="2" valign="bottom"><a name="letter_F"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;F&#160;&#160;</div></td></tr></table>
</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1Mma_3_01Shape___00_01half__t_00_01LayoutA_00_01half__t_00_01L066c9d2371712cdf0cac099ca9bcc578.html">Mma&lt; Shape_, half_t, LayoutA, half_t, LayoutB, half_t, LayoutC, arch::OpMultiplyAdd &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread.html">cutlass::gemm::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator2dThreadTile_3_01Shape___00_017a517f3c73efd795ab05059cc9b111e1.html">PredicatedTileIterator2dThreadTile&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, Transpose_ &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1TileIteratorWmmaTensorOp_3_01WarpShape___00_01OperatorShape_fd6a91cd8bbd07ecd1344326b830e3a4.html">TileIteratorWmmaTensorOp&lt; WarpShape_, OperatorShape_, OperatorFragment_, layout::RowMajor &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1Mma_3_01Shape___00_01half__t_00_01LayoutA___00_01half__t_00_088f0e99e501b6012297eb30b4e89bcea.html">Mma&lt; Shape_, half_t, LayoutA_, half_t, LayoutB_, half_t, layout::RowMajor, arch::OpMultiplyAdd, typename platform::enable_if&lt; detail::EnableMma_Crow_SM60&lt; LayoutA_, LayoutB_ &gt;::value &gt;::type &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread.html">cutlass::gemm::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator2dThreadTile_3_01Shape___00_013671177d6219bfeb0e1b4dc4c1b5bf11.html">PredicatedTileIterator2dThreadTile&lt; Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, Transpose_ &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1thread_1_1Transpose.html">Transpose</a> (<a class="el" href="namespacecutlass_1_1transform_1_1thread.html">cutlass::transform::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1negate.html">negate</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1FloatType.html">FloatType</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1Mma_3_01Shape___00_01int8__t_00_01layout_1_1ColumnMajor_00_013f3785e722edc6e9aab6f866309b8623.html">Mma&lt; Shape_, int8_t, layout::ColumnMajor, int8_t, layout::RowMajor, int32_t, LayoutC_, arch::OpMultiplyAdd, int8_t &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread.html">cutlass::gemm::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape___00_01Element___0068b3e874b5d93d11f0fa902c7f1d11d9.html">PredicatedTileIterator&lt; Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, AccessSize &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1thread_1_1Transpose_3_01ElementCount___00_01layout_1_1PitchLinearS99f8e05faf0bb5ed48a0154afe740d81.html">Transpose&lt; ElementCount_, layout::PitchLinearShape&lt; 4, 4 &gt;, int8_t &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1thread.html">cutlass::transform::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1negate_3_01Array_3_01half__t_00_01N_01_4_01_4.html">negate&lt; Array&lt; half_t, N &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1FloatType_3_0111_00_0152_01_4.html">FloatType&lt; 11, 52 &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1Mma_3_01Shape___00_01int8__t_00_01layout_1_1RowMajor_00_01int89c659e7faf47264972bdba6cd80f42b.html">Mma&lt; Shape_, int8_t, layout::RowMajor, int8_t, layout::ColumnMajor, int32_t, LayoutC_, arch::OpMultiplyAdd, bool &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread.html">cutlass::gemm::thread</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape___00_01Element___00f6b3a9dfab5e7c72d5233f7e5e6e3b9b.html">PredicatedTileIterator&lt; Shape_, Element_, layout::ColumnMajorInterleaved&lt; InterleavedK &gt;, AdvanceRank, ThreadMap_, AccessSize &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1TransposePitchLinearThreadMap.html">TransposePitchLinearThreadMap</a> (<a class="el" href="namespacecutlass_1_1transform.html">cutlass::transform</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1negate_3_01Array_3_01T_00_01N_01_4_01_4.html">negate&lt; Array&lt; T, N &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1FloatType_3_015_00_0110_01_4.html">FloatType&lt; 5, 10 &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1detail_1_1Mma__HFMA2.html">Mma_HFMA2</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread_1_1detail.html">cutlass::gemm::thread::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape___00_01Element___00e7c2c404e7aedfe60ad56bb5571306a1.html">PredicatedTileIterator&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, AccessSize &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1TransposePitchLinearThreadMap2DThreadTile.html">TransposePitchLinearThreadMap2DThreadTile</a> (<a class="el" href="namespacecutlass_1_1transform.html">cutlass::transform</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1nullptr__t.html">nullptr_t</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="structcutlass_1_1FloatType_3_018_00_0123_01_4.html">FloatType&lt; 8, 23 &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1detail_1_1Mma__HFMA2_3_01Shape_00_01layout_1_1ColumnMajor_00_72621f7ab9ae4a4ba4fe9725cf8e89c1.html">Mma_HFMA2&lt; Shape, layout::ColumnMajor, layout::ColumnMajor, layout::ColumnMajor, true &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread_1_1detail.html">cutlass::gemm::thread::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape___00_01Element___0041ea81994f8af0d4d071fdb9e66b5ff0.html">PredicatedTileIterator&lt; Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, AccessSize &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1transform_1_1TransposePitchLinearThreadMapSimt.html">TransposePitchLinearThreadMapSimt</a> (<a class="el" href="namespacecutlass_1_1transform.html">cutlass::transform</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structstd_1_1numeric__limits_3_01cutlass_1_1half__t_01_4.html">numeric_limits&lt; cutlass::half_t &gt;</a> (std)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1FragmentIteratorComplexTensorOp.html">FragmentIteratorComplexTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1detail_1_1Mma__HFMA2_3_01Shape_00_01layout_1_1ColumnMajor_00_94c813e3bbfb6f9857c155166f772687.html">Mma_HFMA2&lt; Shape, layout::ColumnMajor, layout::ColumnMajor, layout::RowMajor, true &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread_1_1detail.html">cutlass::gemm::thread::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape___00_01Element___00d670f969180a8d182dffb356ebcc957e.html">PredicatedTileIterator&lt; Shape_, Element_, layout::RowMajorInterleaved&lt; InterleavedK &gt;, AdvanceRank, ThreadMap_, AccessSize &gt;</a> (<a class="el" href="namespacecutlass_1_1transform_1_1threadblock.html">cutlass::transform::threadblock</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1TrivialConvert.html">TrivialConvert</a> (<a class="el" href="namespacecutlass_1_1reference_1_1host_1_1detail.html">cutlass::reference::host::detail</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_p"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;p&#160;&#160;</div></td></tr></table>
Updated doxygen
2018-09-20 05:02:08 +08:00
</td></tr>
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.
2019-11-20 08:55:34 +08:00
<tr><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1FragmentIteratorComplexTensorOp_3_01WarpShape___00_01Operato8cf03c624cf3210c71b7cbd580b080f8.html">FragmentIteratorComplexTensorOp&lt; WarpShape_, OperatorShape_, OperatorElementC_, OperatorFragmentC_, layout::RowMajor &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1detail_1_1Mma__HFMA2_3_01Shape_00_01layout_1_1ColumnMajor_00_17070298bc4cced0a1b98aee2bb6b455.html">Mma_HFMA2&lt; Shape, layout::ColumnMajor, layout::RowMajor, layout::ColumnMajor, true &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread_1_1detail.html">cutlass::gemm::thread::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1PredicateVector.html">PredicateVector</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1PredicateVector_1_1TrivialIterator.html">PredicateVector::TrivialIterator</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1FragmentIteratorSimt.html">FragmentIteratorSimt</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1detail_1_1Mma__HFMA2_3_01Shape_00_01layout_1_1ColumnMajor_00_bf6d29bb09a025e7b96942809743e28a.html">Mma_HFMA2&lt; Shape, layout::ColumnMajor, layout::RowMajor, layout::RowMajor, true &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread_1_1detail.html">cutlass::gemm::thread::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1PtxWmma.html">PtxWmma</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1TypeTraits.html">TypeTraits</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1platform_1_1alignment__of_1_1pad.html">alignment_of::pad</a> (<a class="el" href="namespacecutlass_1_1platform.html">cutlass::platform</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1FragmentIteratorSimt_3_01WarpShape___00_01Operator___00_01la3f2abc523201c1b0228df99119ab88e1.html">FragmentIteratorSimt&lt; WarpShape_, Operator_, layout::RowMajor, MmaSimtPolicy_ &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1detail_1_1Mma__HFMA2_3_01Shape_00_01layout_1_1RowMajor_00_01l26a133b13650c1d058273e3649f60f04.html">Mma_HFMA2&lt; Shape, layout::RowMajor, layout::ColumnMajor, layout::ColumnMajor, true &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread_1_1detail.html">cutlass::gemm::thread::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1PtxWmmaLoadA.html">PtxWmmaLoadA</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1TypeTraits_3_01complex_3_01double_01_4_01_4.html">TypeTraits&lt; complex&lt; double &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1plus.html">plus</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1FragmentIteratorTensorOp.html">FragmentIteratorTensorOp</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1detail_1_1Mma__HFMA2_3_01Shape_00_01layout_1_1RowMajor_00_01lbba3a796be96a0276693ef6b259ecc4a.html">Mma_HFMA2&lt; Shape, layout::RowMajor, layout::ColumnMajor, layout::RowMajor, true &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread_1_1detail.html">cutlass::gemm::thread::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1PtxWmmaLoadB.html">PtxWmmaLoadB</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1TypeTraits_3_01complex_3_01float_01_4_01_4.html">TypeTraits&lt; complex&lt; float &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1plus_3_01Array_3_01half__t_00_01N_01_4_01_4.html">plus&lt; Array&lt; half_t, N &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1FragmentIteratorTensorOp_3_01WarpShape___00_01OperatorShape_e459aab140a2ce78336e584f95886726.html">FragmentIteratorTensorOp&lt; WarpShape_, OperatorShape_, OperatorElementC_, OperatorFragmentC_, layout::ColumnMajorInterleaved&lt; InterleavedK &gt; &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1detail_1_1Mma__HFMA2_3_01Shape_00_01layout_1_1RowMajor_00_01l2aa4d2fd2e940e0d0cf7c47bc8f6017c.html">Mma_HFMA2&lt; Shape, layout::RowMajor, layout::RowMajor, layout::ColumnMajor, true &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread_1_1detail.html">cutlass::gemm::thread::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1PtxWmmaLoadC.html">PtxWmmaLoadC</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1TypeTraits_3_01complex_3_01half_01_4_01_4.html">TypeTraits&lt; complex&lt; half &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1plus_3_01Array_3_01T_00_01N_01_4_01_4.html">plus&lt; Array&lt; T, N &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
<tr><td valign="top"><a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1FragmentIteratorTensorOp_3_01WarpShape___00_01OperatorShape_5e78dabe303f20d76b00c600aab61eda.html">FragmentIteratorTensorOp&lt; WarpShape_, OperatorShape_, OperatorElementC_, OperatorFragmentC_, layout::RowMajor &gt;</a> (<a class="el" href="namespacecutlass_1_1epilogue_1_1warp.html">cutlass::epilogue::warp</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1gemm_1_1thread_1_1detail_1_1Mma__HFMA2_3_01Shape_00_01layout_1_1RowMajor_00_01l086c058a15d6c79558e4f3d9ff1dc148.html">Mma_HFMA2&lt; Shape, layout::RowMajor, layout::RowMajor, layout::RowMajor, true &gt;</a> (<a class="el" href="namespacecutlass_1_1gemm_1_1thread_1_1detail.html">cutlass::gemm::thread::detail</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1arch_1_1PtxWmmaStoreD.html">PtxWmmaStoreD</a> (<a class="el" href="namespacecutlass_1_1arch.html">cutlass::arch</a>)&#160;&#160;&#160;</td><td valign="top"><a class="el" href="structcutlass_1_1TypeTraits_3_01complex_3_01half__t_01_4_01_4.html">TypeTraits&lt; complex&lt; half_t &gt; &gt;</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td><td rowspan="2" valign="bottom"><a name="letter_r"></a><table border="0" cellspacing="0" cellpadding="0"><tr><td><div class="ah">&#160;&#160;r&#160;&#160;</div></td></tr></table>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</td></tr>
Updating Doxygen docs
2018-10-27 05:54:58 +08:00
<tr><td></td><td></td><td></td><td></td></tr>
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.
2019-11-20 08:55:34 +08:00
<tr><td></td><td></td><td></td><td></td><td valign="top"><a class="el" href="classcutlass_1_1Array_3_01T_00_01N_00_01false_01_4_1_1reference.html">Array&lt; T, N, false &gt;::reference</a> (<a class="el" href="namespacecutlass.html">cutlass</a>)&#160;&#160;&#160;</td></tr>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
<tr><td></td><td></td><td></td><td></td><td></td></tr>
</table>
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.
2019-11-20 08:55:34 +08:00
<div class="qindex"><a class="qindex" href="#letter_A">A</a>&#160;|&#160;<a class="qindex" href="#letter_B">B</a>&#160;|&#160;<a class="qindex" href="#letter_C">C</a>&#160;|&#160;<a class="qindex" href="#letter_D">D</a>&#160;|&#160;<a class="qindex" href="#letter_E">E</a>&#160;|&#160;<a class="qindex" href="#letter_F">F</a>&#160;|&#160;<a class="qindex" href="#letter_G">G</a>&#160;|&#160;<a class="qindex" href="#letter_H">H</a>&#160;|&#160;<a class="qindex" href="#letter_I">I</a>&#160;|&#160;<a class="qindex" href="#letter_K">K</a>&#160;|&#160;<a class="qindex" href="#letter_L">L</a>&#160;|&#160;<a class="qindex" href="#letter_M">M</a>&#160;|&#160;<a class="qindex" href="#letter_N">N</a>&#160;|&#160;<a class="qindex" href="#letter_O">O</a>&#160;|&#160;<a class="qindex" href="#letter_P">P</a>&#160;|&#160;<a class="qindex" href="#letter_R">R</a>&#160;|&#160;<a class="qindex" href="#letter_S">S</a>&#160;|&#160;<a class="qindex" href="#letter_T">T</a>&#160;|&#160;<a class="qindex" href="#letter_U">U</a>&#160;|&#160;<a class="qindex" href="#letter_V">V</a>&#160;|&#160;<a class="qindex" href="#letter_W">W</a>&#160;|&#160;<a class="qindex" href="#letter_X">X</a></div>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</div><!-- contents -->
<!-- start footer part -->
<hr class="footer"/><address class="footer"><small>
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.
2019-11-20 08:55:34 +08:00
Generated by &#160;<a href="http://www.doxygen.org/index.html">
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
<img class="footer" src="doxygen.png" alt="doxygen"/>
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.
2019-11-20 08:55:34 +08:00
</a> 1.8.11
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</small></address>
</body>
</html>
Reference in New Issue Copy Permalink