squall/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity
a8f2c80db0
cutlass/docs/functions_i.html

633 lines
123 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 Members</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><a href="classes.html"><span>Class&#160;Index</span></a></li>
<li><a href="inherits.html"><span>Class&#160;Hierarchy</span></a></li>
<li class="current"><a href="functions.html"><span>Class&#160;Members</span></a></li>
</ul>
</div>
<div id="navrow3" class="tabs2">
<ul class="tablist">
<li class="current"><a href="functions.html"><span>All</span></a></li>
<li><a href="functions_func.html"><span>Functions</span></a></li>
<li><a href="functions_vars.html"><span>Variables</span></a></li>
<li><a href="functions_type.html"><span>Typedefs</span></a></li>
<li><a href="functions_enum.html"><span>Enumerations</span></a></li>
<li><a href="functions_eval.html"><span>Enumerator</span></a></li>
</ul>
</div>
<div id="navrow4" class="tabs3">
<ul class="tablist">
<li><a href="functions.html#index_a"><span>a</span></a></li>
<li><a href="functions_b.html#index_b"><span>b</span></a></li>
<li><a href="functions_c.html#index_c"><span>c</span></a></li>
<li><a href="functions_d.html#index_d"><span>d</span></a></li>
<li><a href="functions_e.html#index_e"><span>e</span></a></li>
<li><a href="functions_f.html#index_f"><span>f</span></a></li>
<li><a href="functions_g.html#index_g"><span>g</span></a></li>
<li><a href="functions_h.html#index_h"><span>h</span></a></li>
<li class="current"><a href="functions_i.html#index_i"><span>i</span></a></li>
<li><a href="functions_k.html#index_k"><span>k</span></a></li>
<li><a href="functions_l.html#index_l"><span>l</span></a></li>
<li><a href="functions_m.html#index_m"><span>m</span></a></li>
<li><a href="functions_n.html#index_n"><span>n</span></a></li>
<li><a href="functions_o.html#index_o"><span>o</span></a></li>
<li><a href="functions_p.html#index_p"><span>p</span></a></li>
<li><a href="functions_q.html#index_q"><span>q</span></a></li>
<li><a href="functions_r.html#index_r"><span>r</span></a></li>
<li><a href="functions_s.html#index_s"><span>s</span></a></li>
<li><a href="functions_t.html#index_t"><span>t</span></a></li>
<li><a href="functions_u.html#index_u"><span>u</span></a></li>
<li><a href="functions_v.html#index_v"><span>v</span></a></li>
<li><a href="functions_w.html#index_w"><span>w</span></a></li>
<li><a href="functions_y.html#index_y"><span>y</span></a></li>
<li><a href="functions_0x7e.html#index_0x7e"><span>~</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="contents">
<div class="textblock">Here is a list of all class members with links to the classes they belong to:</div>
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
<h3><a class="anchor" id="index_i"></a>- i -</h3><ul>
<li>Identity
: <a class="el" href="structcutlass_1_1Distribution.html#a499f4023e0d42356ce71d38cc32bf92aa08adbc111dc827d524a5509c948d8ba6">cutlass::Distribution</a>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</li>
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
<li>IdentityTensorLayout()
: <a class="el" href="classcutlass_1_1IdentityTensorLayout.html#a2aff6124a25853f227cdd7f3b36733c4">cutlass::IdentityTensorLayout&lt; Rank &gt;</a>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</li>
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
<li>imag()
: <a class="el" href="classcutlass_1_1complex.html#aa2c8964e10eeaa99b2d1a18cdcbd0105">cutlass::complex&lt; T &gt;</a>
, <a class="el" href="structcutlass_1_1TypeTraits_3_01complex_3_01double_01_4_01_4_1_1integer__type.html#a48561bb780126ee1ed97a018f07d89e5">cutlass::TypeTraits&lt; complex&lt; double &gt; &gt;::integer_type</a>
, <a class="el" href="structcutlass_1_1TypeTraits_3_01complex_3_01double_01_4_01_4_1_1unsigned__type.html#a8227d537a527c7c577a23c1abd89cee2">cutlass::TypeTraits&lt; complex&lt; double &gt; &gt;::unsigned_type</a>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</li>
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
<li>imag_stride_A
: <a class="el" href="structcutlass_1_1library_1_1GemmPlanarComplexBatchedConfiguration.html#abb6ffb03518c675cfc1a053cbd99fb21">cutlass::library::GemmPlanarComplexBatchedConfiguration</a>
, <a class="el" href="structcutlass_1_1library_1_1GemmPlanarComplexConfiguration.html#a72722870aa853bb9c0418d69aba3fa1f">cutlass::library::GemmPlanarComplexConfiguration</a>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</li>
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
<li>imag_stride_B
: <a class="el" href="structcutlass_1_1library_1_1GemmPlanarComplexBatchedConfiguration.html#a686433eb8204460768f99693b575ed90">cutlass::library::GemmPlanarComplexBatchedConfiguration</a>
, <a class="el" href="structcutlass_1_1library_1_1GemmPlanarComplexConfiguration.html#ad66c23752ee28b8d364487aa6bee4e58">cutlass::library::GemmPlanarComplexConfiguration</a>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</li>
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
<li>imag_stride_C
: <a class="el" href="structcutlass_1_1library_1_1GemmPlanarComplexBatchedConfiguration.html#a9978b56ab1e2c7df6129c7adb66d8cd0">cutlass::library::GemmPlanarComplexBatchedConfiguration</a>
, <a class="el" href="structcutlass_1_1library_1_1GemmPlanarComplexConfiguration.html#a0878decb9bc1f4194e8f029f8a8fdca1">cutlass::library::GemmPlanarComplexConfiguration</a>
Updated doxygen
2018-09-20 05:02:08 +08:00
</li>
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
<li>imag_stride_D
: <a class="el" href="structcutlass_1_1library_1_1GemmPlanarComplexBatchedConfiguration.html#a766b8458393c7c16d74ce3876925f592">cutlass::library::GemmPlanarComplexBatchedConfiguration</a>
, <a class="el" href="structcutlass_1_1library_1_1GemmPlanarComplexConfiguration.html#aa69b38abf8674a34ad61f2e1f60d3324">cutlass::library::GemmPlanarComplexConfiguration</a>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</li>
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
<li>increment_cluster
: <a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1PredicatedTileIterator_1_1Params.html#af3e71c49f5c4830451ba7a70b960b772">cutlass::epilogue::threadblock::PredicatedTileIterator&lt; ThreadMap_, Element_ &gt;::Params</a>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</li>
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
<li>increment_group
: <a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1PredicatedTileIterator_1_1Params.html#a5234c0bbe10bc907f90776dbce50d504">cutlass::epilogue::threadblock::PredicatedTileIterator&lt; ThreadMap_, Element_ &gt;::Params</a>
</li>
<li>increment_row
: <a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1PredicatedTileIterator_1_1Params.html#aca9106ffd4fe4e3d139cf01f3916bcba">cutlass::epilogue::threadblock::PredicatedTileIterator&lt; ThreadMap_, Element_ &gt;::Params</a>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</li>
<li>Index
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 class="el" href="structcutlass_1_1Coord.html#a7a89e5661ef391dd9f4fe81f0c982b75">cutlass::Coord&lt; Rank_, Index_, LongIndex_ &gt;</a>
, <a class="el" href="classcutlass_1_1epilogue_1_1threadblock_1_1InterleavedPredicatedTileIterator.html#acc8e9e6194f2c47232f456b761349bf2">cutlass::epilogue::threadblock::InterleavedPredicatedTileIterator&lt; ThreadMap_, Element_, InterleavedK &gt;</a>
, <a class="el" href="classcutlass_1_1epilogue_1_1threadblock_1_1PredicatedTileIterator.html#ae3464ad454dbb625b0013980998de9ad">cutlass::epilogue::threadblock::PredicatedTileIterator&lt; ThreadMap_, Element_ &gt;</a>
, <a class="el" href="classcutlass_1_1epilogue_1_1threadblock_1_1SharedLoadIterator.html#a82ecb0e54bbcb04364a1924167cbeb15">cutlass::epilogue::threadblock::SharedLoadIterator&lt; ThreadMap_, Element_, MaxAlignment &gt;</a>
, <a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1TileIteratorSimt_3_01WarpShape___00_01Operator___00_01Elemenf2bd262ed3e202b25d5802d83965bf3b.html#a05ff07efb8d79a930472de75ea30938a">cutlass::epilogue::warp::TileIteratorSimt&lt; WarpShape_, Operator_, Element_, layout::RowMajor, MmaSimtPolicy_ &gt;</a>
, <a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1TileIteratorTensorOp_3_01WarpShape___00_01OperatorShape___003cbb32beb84b4984cb7853662096d289.html#a551137bff490b3b6acfd2f0685a81da5">cutlass::epilogue::warp::TileIteratorTensorOp&lt; WarpShape_, OperatorShape_, Element_, layout::RowMajor &gt;</a>
, <a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1TileIteratorVoltaTensorOp_3_01WarpShape___00_01gemm_1_1GemmS2fe0c60b727c738c622c18fc3dd76644.html#a1d327b8913b60b3604924d697ebbaf6a">cutlass::epilogue::warp::TileIteratorVoltaTensorOp&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, float, layout::RowMajor &gt;</a>
, <a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1TileIteratorVoltaTensorOp_3_01WarpShape___00_01gemm_1_1GemmSa0ceeeddc22575876eb977da7f5416a8.html#a3c929d73e2d8f8fba358fab7eaf3b91b">cutlass::epilogue::warp::TileIteratorVoltaTensorOp&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, half_t, layout::RowMajor &gt;</a>
, <a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1TileIteratorWmmaTensorOp_3_01WarpShape___00_01OperatorShape_fd6a91cd8bbd07ecd1344326b830e3a4.html#a6f65512b815630e138f4df3a120d7475">cutlass::epilogue::warp::TileIteratorWmmaTensorOp&lt; WarpShape_, OperatorShape_, OperatorFragment_, layout::RowMajor &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1BatchedGemmCoord.html#a06d52490149675af5ed64f109dc9631d">cutlass::gemm::BatchedGemmCoord</a>
Updated doxygen
2018-09-20 05:02:08 +08:00
, <a class="el" href="structcutlass_1_1gemm_1_1GemmCoord.html#a8e3f0250e2265503862354c729dca892">cutlass::gemm::GemmCoord</a>
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 class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kA_00_01Element_67ca7e11a38e38f2c51b84767654a90f.html#a7a81c5a879e4980ed9d4df58fbf790ee">cutlass::gemm::warp::MmaSimtTileIterator&lt; Shape_, Operand::kA, Element_, layout::ColumnMajor, Policy_, PartitionsK, PartitionGroupSize &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kA_00_01Element_f0ce904a9294556f15e1cc9cf7c99a93.html#a3e026e98c00aea1e4a98067f2b8127b9">cutlass::gemm::warp::MmaSimtTileIterator&lt; Shape_, Operand::kA, Element_, layout::ColumnMajorInterleaved&lt; 4 &gt;, Policy_, PartitionsK, PartitionGroupSize &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kB_00_01Element_ea0a4e7ce3cd5d25cabf79383efdf4d9.html#a835ad161a83d24ae09e907a5f8373edf">cutlass::gemm::warp::MmaSimtTileIterator&lt; Shape_, Operand::kB, Element_, layout::RowMajor, Policy_, PartitionsK, PartitionGroupSize &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kB_00_01Element_ada156b62fcbdce47009c5bf1321c92c.html#a54254a0e5e78d206810b29e0ab2b051a">cutlass::gemm::warp::MmaSimtTileIterator&lt; Shape_, Operand::kB, Element_, layout::RowMajorInterleaved&lt; 4 &gt;, Policy_, PartitionsK, PartitionGroupSize &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kC_00_01Element_4ccafbc821b3a55cd532602442a74031.html#ae86f20baf3ef597fde3b71be9ae9a91a">cutlass::gemm::warp::MmaSimtTileIterator&lt; Shape_, Operand::kC, Element_, layout::ColumnMajor, Policy_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kC_00_01Element_8f92ea79e85febb67169c4b2d94b1b20.html#ab7f320dcdfd5cad606b0701045c31913">cutlass::gemm::warp::MmaSimtTileIterator&lt; Shape_, Operand::kC, Element_, layout::RowMajor, Policy_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpAccumulatorTileIterator_3_01Shape___00_01Element___008f607b871a2b3d854eb4def64712c042.html#aef34e6d38c9039cb9fecb364186e1741">cutlass::gemm::warp::MmaTensorOpAccumulatorTileIterator&lt; Shape_, Element_, cutlass::layout::ColumnMajor, InstructionShape_, OpDelta_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpAccumulatorTileIterator_3_01Shape___00_01Element___00027dabdc144edd6276f664ca74088510.html#a78c667580cfda7d22e63e98dcf5789c4">cutlass::gemm::warp::MmaTensorOpAccumulatorTileIterator&lt; Shape_, Element_, cutlass::layout::ColumnMajorInterleaved&lt; InterleavedN &gt;, InstructionShape_, OpDelta_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpAccumulatorTileIterator_3_01Shape___00_01Element___006c39f57875e0aa9d0ad82c8043ed8b98.html#a9e0746c3f1115c0669dc9d065c50c5f8">cutlass::gemm::warp::MmaTensorOpAccumulatorTileIterator&lt; Shape_, Element_, cutlass::layout::RowMajor, InstructionShape_, OpDelta_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___0b84f53cd44b339eccc12067c9f86e11c.html#ac9e271c881bf768a980f86a04c67932a">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___0e52ad425e1ee3e68544873f66733237b.html#a32d9d7875ad2a5fbb0281115b11ebce9">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___039819fb3ccd43786d556c2c9669508ef.html#a85a6af3ca6d0768de87abfb05be3172f">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___0352e0dcab42bc8360606874e00173556.html#a5afb883b6fd6b897ff852ab93c6ba871">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___0ed7daaeba1c095e77f68533d4d2c475c.html#a3e29a84e9e4f4893492f7e200355087e">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___0c7d419c589d601ce4eb603be566fea21.html#ace4698285e3ccb3d680e721d45c3604a">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpAccumulatorTileIterator.html#a2fc7c23f777567b6b738dcaf0e253298">cutlass::gemm::warp::MmaVoltaTensorOpAccumulatorTileIterator&lt; Shape_, Element_, Layout_, InstructionShape_, OpDelta_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operan0d3248553e52cd61ed8a2b3b12a20343.html#afffcb7e663e505117294da4896e4ad34">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operan34be8e21a40af3ebd2dc3dff460dca72.html#ad07c2e732032e98b818b5c9caac88782">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand734577b7e54a074d143aba59828c2f2.html#a461f3f3f5827b7144952b13e9861e4c0">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operan16c56cdc2dda5eeb996af8ec0242d501.html#aac53debde1f27fd38bb9830ea44309ad">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operan5a221944f4a0e16ccab77ba684856942.html#a0be745dc08b570ca7f102ce17592463c">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operandcc9821c435540895138bc9af495f321.html#afe1ba1cb70dce011a62f8304f8494230">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operana2f40b28f0d2286b84d86f7238d67b52.html#a03dac22a2e5fd81fd0838a4376c4819d">cutlass::gemm::warp::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="classcutlass_1_1HostTensor.html#a33679b84f59829c70457b9ff438101bf">cutlass::HostTensor&lt; Element_, Layout_ &gt;</a>
, <a class="el" href="classcutlass_1_1IdentityTensorLayout.html#ae64fd86033500257785f6923da2558c0">cutlass::IdentityTensorLayout&lt; Rank &gt;</a>
, <a class="el" href="classcutlass_1_1layout_1_1ColumnMajor.html#a6409273d6b40acb27aff0564eb038336">cutlass::layout::ColumnMajor</a>
, <a class="el" href="structcutlass_1_1layout_1_1ColumnMajorBlockLinear.html#a0ed618d4bb2250c01db5bf9ca61e6839">cutlass::layout::ColumnMajorBlockLinear&lt; BlockRows, BlockColumns &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1ColumnMajorInterleaved.html#aa3e25100889b6dfb1a8c1a256e9185ea">cutlass::layout::ColumnMajorInterleaved&lt; Interleave &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1ColumnMajorTensorOpMultiplicandCongruous.html#aa455da7f56592fdb70977b1cad108003">cutlass::layout::ColumnMajorTensorOpMultiplicandCongruous&lt; ElementSize, Crosswise &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1ColumnMajorTensorOpMultiplicandCrosswise.html#af9367fe9d6c5d95367d84dbab02dd0fe">cutlass::layout::ColumnMajorTensorOpMultiplicandCrosswise&lt; ElementSize, Crosswise &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1ColumnMajorVoltaTensorOpMultiplicandBCongruous.html#a19cf61b364213ff4cccaa75ffbfcac52">cutlass::layout::ColumnMajorVoltaTensorOpMultiplicandBCongruous&lt; ElementSize &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1ColumnMajorVoltaTensorOpMultiplicandCongruous.html#a411d7c9444f2d66bf8b8f27d3a6f1226">cutlass::layout::ColumnMajorVoltaTensorOpMultiplicandCongruous&lt; ElementSize &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1ColumnMajorVoltaTensorOpMultiplicandCrosswise.html#a7f6ef66932fda16f2784dd2a1e224ca5">cutlass::layout::ColumnMajorVoltaTensorOpMultiplicandCrosswise&lt; ElementSize, KBlock &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1ContiguousMatrix.html#a5cac7085fc5058fce51bc771b94f49d5">cutlass::layout::ContiguousMatrix</a>
, <a class="el" href="structcutlass_1_1layout_1_1GeneralMatrix.html#ad36a59739ba2eb58f19e2094cfddb24a">cutlass::layout::GeneralMatrix</a>
, <a class="el" href="classcutlass_1_1layout_1_1PackedVectorLayout.html#a359d91ca6d2cae5994db320edeea686e">cutlass::layout::PackedVectorLayout</a>
, <a class="el" href="classcutlass_1_1layout_1_1PitchLinear.html#a9d1dfd7b6d3b2b651009dcba8f5fd5cd">cutlass::layout::PitchLinear</a>
, <a class="el" href="structcutlass_1_1layout_1_1PitchLinearCoord.html#adfb8fc46f2f52a78d0715174ab829dce">cutlass::layout::PitchLinearCoord</a>
, <a class="el" href="classcutlass_1_1layout_1_1RowMajor.html#aa49e242b14b4f482bc6bdd082acfb576">cutlass::layout::RowMajor</a>
, <a class="el" href="structcutlass_1_1layout_1_1RowMajorBlockLinear.html#a4b71b9a7a3e0f9972885798861a18d25">cutlass::layout::RowMajorBlockLinear&lt; BlockRows, BlockColumns &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1RowMajorInterleaved.html#ac827eba11d5fb935273a295d8eb1b972">cutlass::layout::RowMajorInterleaved&lt; Interleave &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1RowMajorTensorOpMultiplicandCongruous.html#a32c8623b9c3c8490263ea0af24a1410f">cutlass::layout::RowMajorTensorOpMultiplicandCongruous&lt; ElementSize, Crosswise &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1RowMajorTensorOpMultiplicandCrosswise.html#afb16d769e699fe88e09420cce6096d86">cutlass::layout::RowMajorTensorOpMultiplicandCrosswise&lt; ElementSize, Crosswise &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1RowMajorVoltaTensorOpMultiplicandBCongruous.html#a80908838bd1135f03e37d342a140bc34">cutlass::layout::RowMajorVoltaTensorOpMultiplicandBCongruous&lt; ElementSize &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1RowMajorVoltaTensorOpMultiplicandCongruous.html#ae9ab9cba1b0b7e8d73677d27aefe4179">cutlass::layout::RowMajorVoltaTensorOpMultiplicandCongruous&lt; ElementSize &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1RowMajorVoltaTensorOpMultiplicandCrosswise.html#a783af2d53a8d6a93c878c9ec4ebdd6b1">cutlass::layout::RowMajorVoltaTensorOpMultiplicandCrosswise&lt; ElementSize, KBlock &gt;</a>
, <a class="el" href="classcutlass_1_1layout_1_1TensorCxRSKx.html#aa7759f0e487cfa2d314951e3dec569e3">cutlass::layout::TensorCxRSKx&lt; Interleave &gt;</a>
, <a class="el" href="classcutlass_1_1layout_1_1TensorNCHW.html#a3d14ee878cfbd3b46a2aef671f9cdfb3">cutlass::layout::TensorNCHW</a>
, <a class="el" href="classcutlass_1_1layout_1_1TensorNCxHWx.html#a57a1e3a60ba44629bbdabe43dc588899">cutlass::layout::TensorNCxHWx&lt; Interleave &gt;</a>
, <a class="el" href="classcutlass_1_1layout_1_1TensorNHWC.html#ac261fc9c833126bb2afa5c70feacf2a3">cutlass::layout::TensorNHWC</a>
, <a class="el" href="structcutlass_1_1layout_1_1TensorOpMultiplicand.html#a68fab741fd24d425c260d274b550bd2f">cutlass::layout::TensorOpMultiplicand&lt; ElementSize, Crosswise &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1TensorOpMultiplicandColumnMajorInterleaved.html#a20c2a57c8a1c6224de7253cfb7239134">cutlass::layout::TensorOpMultiplicandColumnMajorInterleaved&lt; ElementSize, InterleavedK &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1TensorOpMultiplicandCongruous.html#ab498476001a090017735113863e28898">cutlass::layout::TensorOpMultiplicandCongruous&lt; ElementSize, Crosswise &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1TensorOpMultiplicandCongruous_3_0132_00_01Crosswise_01_4.html#adfba7bb7c28bfb734201bac9ca0d2bd0">cutlass::layout::TensorOpMultiplicandCongruous&lt; 32, Crosswise &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1TensorOpMultiplicandCrosswise.html#a4e8aa7fa89ae4ebd8afe284a9786df27">cutlass::layout::TensorOpMultiplicandCrosswise&lt; ElementSize, Crosswise &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1TensorOpMultiplicandRowMajorInterleaved.html#a149aa646854cd7591c7ee528eb5ea3ab">cutlass::layout::TensorOpMultiplicandRowMajorInterleaved&lt; ElementSize, InterleavedK &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1VoltaTensorOpMultiplicandBCongruous.html#aced549eda6008715c78b461396c81b87">cutlass::layout::VoltaTensorOpMultiplicandBCongruous&lt; ElementSize &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1VoltaTensorOpMultiplicandCongruous.html#a53359179aa16d5e938f097a5df36bb71">cutlass::layout::VoltaTensorOpMultiplicandCongruous&lt; ElementSize &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1VoltaTensorOpMultiplicandCrosswise.html#a064fd090cbaa1e359bc44e6edcbf7cde">cutlass::layout::VoltaTensorOpMultiplicandCrosswise&lt; ElementSize, KBlock &gt;</a>
, <a class="el" href="structcutlass_1_1MatrixCoord.html#a25bb16898763baea29b73d8cf13f70d8">cutlass::MatrixCoord</a>
Updating Doxygen docs
2018-10-27 05:54:58 +08:00
, <a class="el" href="structcutlass_1_1reduction_1_1BatchedReductionTraits.html#ab0c9e548f3ee62746e727127e387a8f4">cutlass::reduction::BatchedReductionTraits&lt; ScalarA_, ScalarC_, ScalarD_, ScalarAlphaBeta_, ScalarAccum_, ReductionSize_, OutputTile_, SubTile_, ThreadShape_, Index_, BlockSwizzle_, maxInReg_, maxOutReg_, Functor_ &gt;</a>
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 class="el" href="structcutlass_1_1Tensor4DCoord.html#a4f63c0cc6b642e80624beafb6c3390a1">cutlass::Tensor4DCoord</a>
, <a class="el" href="classcutlass_1_1TensorRef.html#a11ec4b07a2132e647ca2ebe5112ce5ec">cutlass::TensorRef&lt; Element_, Layout_ &gt;</a>
, <a class="el" href="classcutlass_1_1TensorView.html#acf1126519c4d26c85ddd20dc66a13e14">cutlass::TensorView&lt; Element_, Layout_ &gt;</a>
, <a class="el" href="classcutlass_1_1thread_1_1Matrix.html#ae131725fd99cdd6e028ed3cde45d97b8">cutlass::thread::Matrix&lt; Element, Rows, Columns, Layout &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator2dThreadTile_3_01Shape__da632779aba661c0f4cfaaa78126b771.html#aacdd835411c8bbd77201cb3cc9ce18fb">cutlass::transform::threadblock::PredicatedTileAccessIterator2dThreadTile&lt; Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, AccessType_ &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator2dThreadTile_3_01Shape__1790abaa54a01f277d75766d5882fec8.html#a5857c4d9561fa25e2935960197c1a0d5">cutlass::transform::threadblock::PredicatedTileAccessIterator2dThreadTile&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, AccessType_ &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator2dThreadTile_3_01Shape__7327fa15996bcb8502cdfcc192350fe1.html#a89554e5c07581d02000ee5eb8db043c1">cutlass::transform::threadblock::PredicatedTileAccessIterator2dThreadTile&lt; Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, AccessType_ &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator_3_01Shape___00_01Elemen89c687c583745a73cb485041911a4c4e.html#af6b1c61696ecf3a22d8711b66365ec94">cutlass::transform::threadblock::PredicatedTileAccessIterator&lt; Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, AccessType_ &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator_3_01Shape___00_01Elemenab63a1e105bf37f6371516cb9e2c5a7a.html#ad97511a7b656eb221619be8367bac087">cutlass::transform::threadblock::PredicatedTileAccessIterator&lt; Shape_, Element_, layout::ColumnMajorInterleaved&lt; InterleavedK &gt;, AdvanceRank, ThreadMap_, AccessType_ &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator_3_01Shape___00_01Elemen784a0e9da3f55064c47e5613791f51f7.html#a39cef2239175392ec904afcf6d8e5724">cutlass::transform::threadblock::PredicatedTileAccessIterator&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, AccessType_ &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator_3_01Shape___00_01Elemen9838736ad62fae54213fbaf722a989ab.html#aa292ce960f4fba19e5d2b5870b4be082">cutlass::transform::threadblock::PredicatedTileAccessIterator&lt; Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, AccessType_ &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileAccessIterator_3_01Shape___00_01Elemen809793e785fb4211888c6b4e5dcfcb39.html#a0f2a0495da657de176dc6fac3274bf89">cutlass::transform::threadblock::PredicatedTileAccessIterator&lt; Shape_, Element_, layout::RowMajorInterleaved&lt; InterleavedK &gt;, AdvanceRank, ThreadMap_, AccessType_ &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator2dThreadTile_3_01Shape___00_0165b39a630d10785a3558406f9adb99b9.html#a7912240a9802251763983a1fa4a297f1">cutlass::transform::threadblock::PredicatedTileIterator2dThreadTile&lt; Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, Transpose_ &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator2dThreadTile_3_01Shape___00_017a517f3c73efd795ab05059cc9b111e1.html#a715b78e9d8587c4c59e5337d33eb7025">cutlass::transform::threadblock::PredicatedTileIterator2dThreadTile&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, Transpose_ &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator2dThreadTile_3_01Shape___00_013671177d6219bfeb0e1b4dc4c1b5bf11.html#a31ddecaefe6832464ccd170834e398ca">cutlass::transform::threadblock::PredicatedTileIterator2dThreadTile&lt; Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, Transpose_ &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape___00_01Element___0068b3e874b5d93d11f0fa902c7f1d11d9.html#acfd766202c25c0323160a91db74ad82b">cutlass::transform::threadblock::PredicatedTileIterator&lt; Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, AccessSize &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape___00_01Element___00f6b3a9dfab5e7c72d5233f7e5e6e3b9b.html#a910b8ba4e409fb7d26739ee3f0c46a36">cutlass::transform::threadblock::PredicatedTileIterator&lt; Shape_, Element_, layout::ColumnMajorInterleaved&lt; InterleavedK &gt;, AdvanceRank, ThreadMap_, AccessSize &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape___00_01Element___00e7c2c404e7aedfe60ad56bb5571306a1.html#ac68f091bc3ef944e3476c838f2225157">cutlass::transform::threadblock::PredicatedTileIterator&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, AccessSize &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape___00_01Element___0041ea81994f8af0d4d071fdb9e66b5ff0.html#aee0b606c8a3d36df29d4e7a537d3755a">cutlass::transform::threadblock::PredicatedTileIterator&lt; Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, AccessSize &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1PredicatedTileIterator_3_01Shape___00_01Element___00d670f969180a8d182dffb356ebcc957e.html#afa189422b204c68d154c816ae84fa13c">cutlass::transform::threadblock::PredicatedTileIterator&lt; Shape_, Element_, layout::RowMajorInterleaved&lt; InterleavedK &gt;, AdvanceRank, ThreadMap_, AccessSize &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__eb7d20f8b9d69e0ae5e7ef51dc480867.html#a7dc277172871c7f936620dd56a3cedfc">cutlass::transform::threadblock::RegularTileAccessIterator&lt; Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__2c1476eaf582bfe972793e17babfe985.html#a4ba4f46f5286af5b7b647944cf3de8ad">cutlass::transform::threadblock::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="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__a3c11cf1f00ef7a1efb8389ac6e4c6e0.html#a052f3d5d2f3f2f4578a3b59c8e88a676">cutlass::transform::threadblock::RegularTileAccessIterator&lt; Shape_, Element_, layout::ColumnMajorTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__0855e9d9ab619202d2397180c1e4c4a5.html#a17b3fb7aff1db6239249b4bb777655bc">cutlass::transform::threadblock::RegularTileAccessIterator&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__f04332958a49a47d6fb2b25201764630.html#a392334952f932482affbe988e94d0d34">cutlass::transform::threadblock::RegularTileAccessIterator&lt; Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__6baada077236f1a368c61c5e11b45b72.html#a5cc867dbd6b5a122e9951ff492f5238a">cutlass::transform::threadblock::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="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__0184b7188941788a96624510a4b2f876.html#ad619429f4f3710665d47a39bd4fbb78c">cutlass::transform::threadblock::RegularTileAccessIterator&lt; Shape_, Element_, layout::RowMajorTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__ebf4714349612673e8b6609b763eeb6f.html#a636e7123ac8169f41d47e071e460fb81">cutlass::transform::threadblock::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="classcutlass_1_1transform_1_1threadblock_1_1RegularTileAccessIterator_3_01Shape___00_01Element__e9a9e0f4286f652f55eb9b863b21effe.html#abc58f5f5ad3c41b62db504ef05bf7a93">cutlass::transform::threadblock::RegularTileAccessIterator&lt; Shape_, Element_, layout::TensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator2dThreadTile_3_01Shape___00_01Eleb60d066756d1c18f05fceee6a27bdb8a.html#aca10b6f64c20ce65af157f64d121ef1d">cutlass::transform::threadblock::RegularTileIterator2dThreadTile&lt; Shape_, Element_, layout::ColumnMajorInterleaved&lt; 4 &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator2dThreadTile_3_01Shape___00_01Ele76ed82829532ae1c17f4c78158f036c7.html#a583c4c00c3711314310523b17713dbc4">cutlass::transform::threadblock::RegularTileIterator2dThreadTile&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator2dThreadTile_3_01Shape___00_01Ele654c8f6161ae5340f040397a4e2e045c.html#a4ba6c92439dde0bc04b484639e588b73">cutlass::transform::threadblock::RegularTileIterator2dThreadTile&lt; Shape_, Element_, layout::RowMajorInterleaved&lt; 4 &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_011d3637dbd8bc58bcb020b51bf57fbfc0.html#a236cea40e2359a0b5fe35ff1a3519e10">cutlass::transform::threadblock::RegularTileIterator&lt; Shape_, Element_, layout::ColumnMajor, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_017982f81d4ef592e19c8427de2ea933a3.html#a16ffd59718fc0840d097581b97d14249">cutlass::transform::threadblock::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="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_010889a732373c350de9b9a9f6c13cd761.html#a147b0e450c08ac0d819b53a63c7aa99e">cutlass::transform::threadblock::RegularTileIterator&lt; Shape_, Element_, layout::ColumnMajorTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01187f8574e1fe9d7d5e8fbf09bd834bf0.html#a16d6cac906acd65af0137086f8eec3d0">cutlass::transform::threadblock::RegularTileIterator&lt; Shape_, Element_, layout::ColumnMajorVoltaTensorOpMultiplicandBCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01793f74bfd8f116a827948ab01a37349a.html#aa497c5741e2c1a12525fd100478ec34b">cutlass::transform::threadblock::RegularTileIterator&lt; Shape_, Element_, layout::ColumnMajorVoltaTensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01bd31b3810c1fedf2e7e5959ff92b5d3d.html#a1281f6ab20cc0d3823992b2342371d47">cutlass::transform::threadblock::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="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_0184a89653916f5d51ab59d1b386989a17.html#a3a922bd28e511a085e2cd09377c04144">cutlass::transform::threadblock::RegularTileIterator&lt; Shape_, Element_, layout::PitchLinear, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_0149454d361ea5885cf5166a920b5145df.html#af808b8c3fc0bea942874f61c17bbc4f8">cutlass::transform::threadblock::RegularTileIterator&lt; Shape_, Element_, layout::RowMajor, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01c20d35180520077a5a09b1e33543c1a5.html#a2284ad3a34f8db161542da80486e9ff1">cutlass::transform::threadblock::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="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01a31b454d9c930525c1e9ca406a514f40.html#a309465f4c443c5f54e4772226f54bd3c">cutlass::transform::threadblock::RegularTileIterator&lt; Shape_, Element_, layout::RowMajorTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_0104ad31bd559a88cc418ae1cab7492ed5.html#a94e8cc78e38dca1afa16ee123e8c7033">cutlass::transform::threadblock::RegularTileIterator&lt; Shape_, Element_, layout::RowMajorVoltaTensorOpMultiplicandBCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01f6f6511b5033cad31083644ac69c54d8.html#a78f8a422b799001f2c42659b012b722c">cutlass::transform::threadblock::RegularTileIterator&lt; Shape_, Element_, layout::RowMajorVoltaTensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01b3fa5720e807697de61b9f937b269cd0.html#a6f91a8008d60668bc7757654a34efc48">cutlass::transform::threadblock::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="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01efd5013a2503d6567e2bf6b40c97360c.html#af29174087c689128e79bff5b56efb8a2">cutlass::transform::threadblock::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="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_0197fef2242a3454a7d1cebe61aee28b43.html#a8cf9ea1e3e4e31ae93de613116ae0f63">cutlass::transform::threadblock::RegularTileIterator&lt; Shape_, Element_, layout::TensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Crosswise &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01a75d2cd74e722d6ad6a3b41aabfd432d.html#acf73851870e141f73aa1c8c5a251714f">cutlass::transform::threadblock::RegularTileIterator&lt; Shape_, Element_, layout::VoltaTensorOpMultiplicandBCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01f96bbeb63e6d4ce4a2551279de3a9f0e.html#a0adc056cc8ab3d39f8a023085e9ad8c2">cutlass::transform::threadblock::RegularTileIterator&lt; Shape_, Element_, layout::VoltaTensorOpMultiplicandCongruous&lt; sizeof_bits&lt; Element_ &gt;::value &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a>
, <a class="el" href="classcutlass_1_1transform_1_1threadblock_1_1RegularTileIterator_3_01Shape___00_01Element___00_01dbd6b8468d5bd787308d2f615a24d123.html#aed83968f30e942628bbe59a36daaa57e">cutlass::transform::threadblock::RegularTileIterator&lt; Shape_, Element_, layout::VoltaTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, Shape_::kContiguous &gt;, AdvanceRank, ThreadMap_, Alignment &gt;</a>
</li>
<li>infinity()
: <a class="el" href="structstd_1_1numeric__limits_3_01cutlass_1_1half__t_01_4.html#ab7a40820e64282376a050095d5004b74">std::numeric_limits&lt; cutlass::half_t &gt;</a>
</li>
<li>initial_offset()
: <a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1InterleavedOutputTileThreadMap.html#af0ec4c4a46c5a83150b3f1ac80ca8e2d">cutlass::epilogue::threadblock::InterleavedOutputTileThreadMap&lt; WarpCount_, MmaCount_, Threads, ElementsPerAccess, ElementSize &gt;</a>
, <a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1OutputTileOptimalThreadMap_1_1CompactedThreadMap.html#a78883d7cac3a619cce6006dde714cfda">cutlass::epilogue::threadblock::OutputTileOptimalThreadMap&lt; Shape_, Count_, Threads, ElementsPerAccess, ElementSize &gt;::CompactedThreadMap</a>
, <a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1OutputTileOptimalThreadMap.html#a39026f56939737423ace3f6471b4c159">cutlass::epilogue::threadblock::OutputTileOptimalThreadMap&lt; Shape_, Count_, Threads, ElementsPerAccess, ElementSize &gt;</a>
, <a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1OutputTileThreadMap.html#acd308197f2e234388a824273d592d965">cutlass::epilogue::threadblock::OutputTileThreadMap&lt; ThreadMap_, Shape_, Iterations_, Delta_, Count_ &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1PitchLinear2DThreadTileStripminedThreadMap_3_01Shape___00_01Thread0082c3467229b12cc9dd996283ee7160.html#a96ac336cac3a8b4d10f1764a925e0902">cutlass::transform::PitchLinear2DThreadTileStripminedThreadMap&lt; Shape_, Threads, cutlass::layout::PitchLinearShape&lt; 4, 4 &gt; &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1PitchLinearStripminedThreadMap.html#aeb09f5131cac18bfd820d2ab4cb06c49">cutlass::transform::PitchLinearStripminedThreadMap&lt; Shape_, Threads, ElementsPerAccess &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1PitchLinearTilePolicyStripminedThreadContiguous.html#ab397cf8fa1f24bf54352e1352a9af1c9">cutlass::transform::PitchLinearTilePolicyStripminedThreadContiguous&lt; Shape, Threads, ElementsPerAccess &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1PitchLinearTilePolicyStripminedThreadStrided.html#ab970c0505583b4e6f928b9a790c03856">cutlass::transform::PitchLinearTilePolicyStripminedThreadStrided&lt; Shape, Threads, ElementsPerAccess &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1PitchLinearWarpRakedThreadMap.html#a495158bd07a83fa73c78bc4e41b92c86">cutlass::transform::PitchLinearWarpRakedThreadMap&lt; Shape_, Threads, WarpThreadArrangement_, ElementsPerAccess &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1PitchLinearWarpStripedThreadMap.html#acc6d9ccbb71792f3dea2e7c6c88f3a07">cutlass::transform::PitchLinearWarpStripedThreadMap&lt; Shape_, Threads, WarpThreadArrangement_, ElementsPerAccess &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1TransposePitchLinearThreadMap2DThreadTile.html#a93bf84427a6f28f45df317200ee2a404">cutlass::transform::TransposePitchLinearThreadMap2DThreadTile&lt; ThreadMap_ &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1TransposePitchLinearThreadMap.html#a1efbb1ee0b34e0d258fc74c201d9ee02">cutlass::transform::TransposePitchLinearThreadMap&lt; ThreadMap_, WarpThreadArrangement_ &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1TransposePitchLinearThreadMapSimt.html#a7f8daecfdf34c489e01a1fb47fd486a5">cutlass::transform::TransposePitchLinearThreadMapSimt&lt; ThreadMap_ &gt;</a>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</li>
<li>initialize()
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 class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1InterleavedPredicatedTileIterator_1_1Params.html#a6f0ad92d44376e1bf61e7fb6932a3dfd">cutlass::epilogue::threadblock::InterleavedPredicatedTileIterator&lt; ThreadMap_, Element_, InterleavedK &gt;::Params</a>
, <a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1PredicatedTileIterator_1_1Params.html#ad5da25e1dd34da92acbb00b25f4be7f5">cutlass::epilogue::threadblock::PredicatedTileIterator&lt; ThreadMap_, Element_ &gt;::Params</a>
, <a class="el" href="classcutlass_1_1gemm_1_1device_1_1Gemm.html#a53d79d1b434100da1e466e6378ec43ab">cutlass::gemm::device::Gemm&lt; ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, SplitKSerial, Operator_, IsBetaZero &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1device_1_1Gemm_3_01ElementA___00_01LayoutA___00_01ElementB___00_01Layout4d0960ae6b1d1bf19e6239dbd002249c.html#a7a14474e4238d2fac92ad71c6de087d8">cutlass::gemm::device::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="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#aa2670ac441f48f6a0a2071c67c743ab8">cutlass::gemm::device::GemmBatched&lt; ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#a428d8b1c4ac36040145a59d8e4cff3d2">cutlass::gemm::device::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="classcutlass_1_1gemm_1_1device_1_1GemmComplex.html#acec1cbb9d876e0d7ee8e8e9992e592ae">cutlass::gemm::device::GemmComplex&lt; ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, TransformA, TransformB, SplitKSerial &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1device_1_1GemmComplex_3_01ElementA___00_01LayoutA___00_01ElementB___00_07c56401b4df75709ae636675d9980a9a.html#a5c3286631f254746c9eb788b780cdca3">cutlass::gemm::device::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="classcutlass_1_1gemm_1_1device_1_1GemmSplitKParallel.html#a7085b7cf85bc1bcd202ea6928656d966">cutlass::gemm::device::GemmSplitKParallel&lt; ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1device_1_1GemmSplitKParallel_3_01ElementA___00_01LayoutA___00_01ElementBbe7c1f7154ad5b5bf9d4d28301e2b457.html#a4d9f086305f76d7f885bf032f3d2c7c9">cutlass::gemm::device::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="classcutlass_1_1library_1_1Manifest.html#a23feae702cfd606ed14d1407bb9d799d">cutlass::library::Manifest</a>
, <a class="el" href="classcutlass_1_1library_1_1Operation.html#a649274fbec9e5f2e6dbad128f7780166">cutlass::library::Operation</a>
Updating Doxygen docs
2018-10-27 05:54:58 +08:00
, <a class="el" href="structcutlass_1_1reduction_1_1BatchedReductionTraits_1_1Params.html#ac27f42beb3625c5183b76b26677c0cb0">cutlass::reduction::BatchedReductionTraits&lt; ScalarA_, ScalarC_, ScalarD_, ScalarAlphaBeta_, ScalarAccum_, ReductionSize_, OutputTile_, SubTile_, ThreadShape_, Index_, BlockSwizzle_, maxInReg_, maxOutReg_, Functor_ &gt;::Params</a>
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
</li>
<li>instruction_shape
: <a class="el" href="structcutlass_1_1library_1_1MathInstructionDescription.html#ab6c56bb228a4ba93619bf69971dcc82e">cutlass::library::MathInstructionDescription</a>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</li>
<li>InstructionShape
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 class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1DefaultThreadMapWmmaTensorOp.html#a7bbd17d9ed710da442b8f494912028a3">cutlass::epilogue::threadblock::DefaultThreadMapWmmaTensorOp&lt; ThreadblockShape_, WarpShape_, InstructionShape_, PartitionsK, Element_, ElementsPerAccess &gt;</a>
, <a class="el" href="structcutlass_1_1epilogue_1_1warp_1_1VoltaTensorOpPolicy_3_01WarpShape___00_01gemm_1_1GemmShape_136ce744d4c1c6e8707f5a9785196194.html#a2553198647e304e1b2e262169e89adaa">cutlass::epilogue::warp::VoltaTensorOpPolicy&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, float, layout::RowMajor &gt;</a>
, <a class="el" href="structcutlass_1_1epilogue_1_1warp_1_1VoltaTensorOpPolicy_3_01WarpShape___00_01gemm_1_1GemmShape_1d48185f49e4d066f8e9327bf0856b7f.html#aee2b5d5ef9bc8815ad484d95827452f4">cutlass::epilogue::warp::VoltaTensorOpPolicy&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, half_t, layout::RowMajor &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassSimt_00_01ArchTag286687c5e6abe22d241f789fe344a465.html#acc869c97343e89e72df505c6158c6e61">cutlass::gemm::device::DefaultGemmConfiguration&lt; arch::OpClassSimt, ArchTag, ElementA, ElementB, ElementC, ElementAccumulator &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassSimt_00_01ArchTag3026e48abb8c905d1cc6d13d669700e4.html#a7416935776c933afb5391265cb7a24b7">cutlass::gemm::device::DefaultGemmConfiguration&lt; arch::OpClassSimt, ArchTag, int8_t, int8_t, ElementC, int32_t &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arc567cad318a31d04b70ea615d6321decd.html#ae2aad35c09d2ea760d8f2b5539723c95">cutlass::gemm::device::DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm70, ElementA, ElementB, ElementC, ElementAccumulator &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arcde61af9be1337dac1fdb210e7e7a6e01.html#a9cf55b4c5fc4a7c4d19deb4cdba1eb3e">cutlass::gemm::device::DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, ElementA, ElementB, ElementC, ElementAccumulator &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arc485a4f0b5a7d2d4ab2c1a24da6328048.html#aff148c4493980a6539e744a33cda734d">cutlass::gemm::device::DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, int4b_t, int4b_t, ElementC, int32_t &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arc8e2604a56dff3a7595da9ee0604ae55e.html#a656bedbde76370f3ccead9eb39899373">cutlass::gemm::device::DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, int4b_t, uint4b_t, ElementC, int32_t &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arc4fada4957d463c80a2831e47f28157c4.html#a46c9053075836a177db5424d8e640936">cutlass::gemm::device::DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, int8_t, int8_t, ElementC, int32_t &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arc8ab5fd2693c6a6ec43e447acb07f784c.html#af10c5d9ba13aa3596fddb79c0a05fc63">cutlass::gemm::device::DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, int8_t, uint8_t, ElementC, int32_t &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arcffcf31256aed23d4d8d0eab627bc0cad.html#a389b57020ce001af12c96be6783ec81b">cutlass::gemm::device::DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, uint4b_t, int4b_t, ElementC, int32_t &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arcb2e258b7bd321c633dd65d3ebcf6414a.html#afddefd8459e32926c21494d46b363294">cutlass::gemm::device::DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, uint4b_t, uint4b_t, ElementC, int32_t &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arcb27bf218007928652d5b803193eab473.html#a73037a1163049c90943b9a3941e59e88">cutlass::gemm::device::DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, uint8_t, int8_t, ElementC, int32_t &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1device_1_1DefaultGemmConfiguration_3_01arch_1_1OpClassTensorOp_00_01arcfea0f3503156e8e3fba6456f0cedafdd.html#ae2827ac69cfa71a06ba7882b9c9234e6">cutlass::gemm::device::DefaultGemmConfiguration&lt; arch::OpClassTensorOp, arch::Sm75, uint8_t, uint8_t, ElementC, int32_t &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1device_1_1Gemm.html#ac98211b8adf5c18b4b6c54c5d1cdbb1a">cutlass::gemm::device::Gemm&lt; ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, SplitKSerial, Operator_, IsBetaZero &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1device_1_1Gemm_3_01ElementA___00_01LayoutA___00_01ElementB___00_01Layout4d0960ae6b1d1bf19e6239dbd002249c.html#a71b84f983b94b50a48bd0890f1e0ed59">cutlass::gemm::device::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="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a0dbb6d5185f223bb8242fc47a3b77757">cutlass::gemm::device::GemmBatched&lt; ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#ae073edad6dd4447d7f99c94f4cd0c1c8">cutlass::gemm::device::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="classcutlass_1_1gemm_1_1device_1_1GemmComplex.html#a751ed3f77f21dc6ef2508f981cc697b6">cutlass::gemm::device::GemmComplex&lt; ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, TransformA, TransformB, SplitKSerial &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1device_1_1GemmComplex_3_01ElementA___00_01LayoutA___00_01ElementB___00_07c56401b4df75709ae636675d9980a9a.html#af691b4304896f601c34eaedf78493ed5">cutlass::gemm::device::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="classcutlass_1_1gemm_1_1device_1_1GemmSplitKParallel.html#ab2e2468a859f14502fd18013859ec9e6">cutlass::gemm::device::GemmSplitKParallel&lt; ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ConvertScaledOp_, ReductionOp_, ThreadblockSwizzle_, Stages, kAlignmentA, kAlignmentB, Operator_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1device_1_1GemmSplitKParallel_3_01ElementA___00_01LayoutA___00_01ElementBbe7c1f7154ad5b5bf9d4d28301e2b457.html#a8c6c83e045a18b7a3c004e039509576e">cutlass::gemm::device::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="structcutlass_1_1gemm_1_1kernel_1_1DefaultGemm_3_01int8__t_00_01LayoutA_00_01kAlignmentA_00_01inf48440732c1c5f42ddbfaba179861815.html#a4c8d0c1325787bba28ee7bf554839070">cutlass::gemm::kernel::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMma_3_01int8__t_00_01LayoutA_00_01kAlignmentA_00_07e7230d4011ada5e22cfcb29103b696.html#a465987b35751253ff48923ca24fc9457">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShab94a11a77dd0565102710907089acee0.html#a77672784e6e2f17669c018f4caea2700">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShafafd5c61db86cbfe90863578ddd11092.html#a88e589a8ab9399b4116b98bbb872d133">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha46446d1e3871e31d2e728f710d78c8c1.html#a4fb9b534fa9e321c8f8e1565e3d0e39c">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha8da7a0cfbbe859b701fdd9f2b8566aa7.html#af529eb97a1e2d46e301fbaf1dbf4c89c">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha84e9f8afb6a4ca9f5dcd219b182d16e7.html#acd6237299856344061f05c8100b452cc">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha2c0d0b7cdb5c4bcb11e83c058eb65345.html#a2884fa009363296f7404872e354925ad">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha34a52cc7b2942e8c290f0032b6779b52.html#a06edf3a91f98cc7a0e8ab9c01f30b21a">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShaaf312aafe9da92ea9d417bcc12a8e7dc.html#a57c8d3edfc3202db15382b5f19835e3c">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha863d4139ccaa713bc4bde32c425f4067.html#aa3c6e36d439c7ac96d1fb5ae28f7394e">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShaf03a122202ad10acdc96f280106d678b.html#aa55bbfab1326ad19ae3ec1e9cf6fe9fb">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha69bef08ea63dd930f99d9788105873dd.html#a89646f5bc7ad42508f5d0bdc258a758c">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha3adf608332a8c9ee7014fced0da8a9ca.html#a0f38e2bcc7cddf8459419f296d177adf">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShab7edfba3cdf43a07e3c4d719d87565a4.html#a24a4867acd895473f3c6534b205ee6aa">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instruc803d38bc1e4618c07c47f54c87ae2678.html#a4b2e594c3515e4553e648e988ff7b050">cutlass::gemm::threadblock::DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::ColumnMajor, ElementB_, layout::ColumnMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instrucf60fe02fcdd80d28b7fd419133465dcc.html#a06d5949295ea6f22f885df2283265202">cutlass::gemm::threadblock::DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::ColumnMajor, ElementB_, layout::RowMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instruc2bf00737f4ad0a9da9a8be6d3e66c152.html#a363519488ba4a9c4323019b875faa6a0">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instruc24092ddc01fc83dabb7db4c14880fe60.html#ab78bc178f21ba3d6fcf10587133cba08">cutlass::gemm::threadblock::DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::RowMajor, ElementB_, layout::ColumnMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instruc4fee9f2965b8468bfb42b94a74527d22.html#af81cb8a47ce8cab313dffec5d477b335">cutlass::gemm::threadblock::DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::RowMajor, ElementB_, layout::RowMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpAccumulatorTileIterator_3_01Shape___00_01Element___008f607b871a2b3d854eb4def64712c042.html#aabb62dc50c032e4691e6d96ea6740f12">cutlass::gemm::warp::MmaTensorOpAccumulatorTileIterator&lt; Shape_, Element_, cutlass::layout::ColumnMajor, InstructionShape_, OpDelta_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpAccumulatorTileIterator_3_01Shape___00_01Element___00027dabdc144edd6276f664ca74088510.html#a784a554c0b5d90d36e8df22df8e4d80c">cutlass::gemm::warp::MmaTensorOpAccumulatorTileIterator&lt; Shape_, Element_, cutlass::layout::ColumnMajorInterleaved&lt; InterleavedN &gt;, InstructionShape_, OpDelta_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpAccumulatorTileIterator_3_01Shape___00_01Element___006c39f57875e0aa9d0ad82c8043ed8b98.html#a0b99108d7d2ac1b2047c3d6ea51a6423">cutlass::gemm::warp::MmaTensorOpAccumulatorTileIterator&lt; Shape_, Element_, cutlass::layout::RowMajor, InstructionShape_, OpDelta_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___0b84f53cd44b339eccc12067c9f86e11c.html#aa3eecd0ea10efc45a6c11569a3e9d1b5">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___0e52ad425e1ee3e68544873f66733237b.html#a4976d5d69018094eddecb44de711343e">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___039819fb3ccd43786d556c2c9669508ef.html#ad91c3ff6512ec6f63dac11af36f553cb">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___0352e0dcab42bc8360606874e00173556.html#a9f08f1ae1bfbe2b3fdad0105d5f36cfc">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___0ed7daaeba1c095e77f68533d4d2c475c.html#a2229636a51c6233ee4f9de8a0f15107a">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand___0c7d419c589d601ce4eb603be566fea21.html#a276c1d33c1380292ac581eae6afe00f2">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpAccumulatorTileIterator.html#ab907fdc0fef5042e740e24d7dc4e59b7">cutlass::gemm::warp::MmaVoltaTensorOpAccumulatorTileIterator&lt; Shape_, Element_, Layout_, InstructionShape_, OpDelta_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operan0d3248553e52cd61ed8a2b3b12a20343.html#a72f221f2cb137b42d609d9d392cddcd7">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operan34be8e21a40af3ebd2dc3dff460dca72.html#abc3ae2fe83b096488e798d37327ff3c9">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operand734577b7e54a074d143aba59828c2f2.html#a45012fadebb81b49bba98949ea228a0b">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operan16c56cdc2dda5eeb996af8ec0242d501.html#a719ef6becc48d7b2f35bffc781bc946f">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operan5a221944f4a0e16ccab77ba684856942.html#a21877ac7be6b023f10658bd3f74d08dc">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operandcc9821c435540895138bc9af495f321.html#a4fca090ed65e8579f6870a7765e46e3e">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpMultiplicandTileIterator_3_01Shape___00_01Operana2f40b28f0d2286b84d86f7238d67b52.html#a2c060728c43544ae3986db19e789d75e">cutlass::gemm::warp::MmaVoltaTensorOpMultiplicandTileIterator&lt; Shape_, Operand_, Element_, cutlass::layout::VoltaTensorOpMultiplicandCrosswise&lt; sizeof_bits&lt; Element_ &gt;::value, KBlock &gt;, InstructionShape_, OpDelta_, 32 &gt;</a>
</li>
<li>int_scale
: <a class="el" href="structcutlass_1_1Distribution.html#a676b1d8b87691b4218f6ed308e6adfc1">cutlass::Distribution</a>
, <a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1RandomGaussianFunc_1_1Params.html#a0f5b0c8508cd6d143da7bbe20b68a60c">cutlass::reference::device::detail::RandomGaussianFunc&lt; Element &gt;::Params</a>
, <a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1RandomUniformFunc_1_1Params.html#afe8637b103e25ec2e9b731389fa049be">cutlass::reference::device::detail::RandomUniformFunc&lt; Element &gt;::Params</a>
, <a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1RandomGaussianFunc.html#a4c9943f36faab7d4928b1f130d0b784c">cutlass::reference::host::detail::RandomGaussianFunc&lt; Element &gt;</a>
, <a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1RandomGaussianFunc_3_01complex_3_01Element_01_4_01_4.html#a6a906b7ae9c17b6ebe2063d652f3ab50">cutlass::reference::host::detail::RandomGaussianFunc&lt; complex&lt; Element &gt; &gt;</a>
, <a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1RandomUniformFunc.html#a9b410cf9a05a34317a83cce730f9a348">cutlass::reference::host::detail::RandomUniformFunc&lt; Element &gt;</a>
, <a class="el" href="structcutlass_1_1reference_1_1host_1_1detail_1_1RandomUniformFunc_3_01complex_3_01Element_01_4_01_4.html#ad0de7d4946af855288d7f9cccb9a18eb">cutlass::reference::host::detail::RandomUniformFunc&lt; complex&lt; Element &gt; &gt;</a>
</li>
<li>integer_subbyte()
: <a class="el" href="structcutlass_1_1integer__subbyte.html#ab893ce5c4d31dd0485fb327dbcdd4d4c">cutlass::integer_subbyte&lt; Bits, Signed &gt;</a>
</li>
<li>integer_type
: <a class="el" href="structcutlass_1_1TypeTraits_3_01complex_3_01float_01_4_01_4.html#a67120bdef9cad92ac289bf26fa8008d6">cutlass::TypeTraits&lt; complex&lt; float &gt; &gt;</a>
, <a class="el" href="structcutlass_1_1TypeTraits_3_01complex_3_01half_01_4_01_4.html#a7c4b7a2e3265c19a2ff7930fcf23db27">cutlass::TypeTraits&lt; complex&lt; half &gt; &gt;</a>
, <a class="el" href="structcutlass_1_1TypeTraits_3_01complex_3_01half__t_01_4_01_4.html#ac936d3dbbc2cd53b42d6e97fcf0c350f">cutlass::TypeTraits&lt; complex&lt; half_t &gt; &gt;</a>
, <a class="el" href="structcutlass_1_1TypeTraits_3_01double_01_4.html#aeb17f038584dc2584248fa79c2f99835">cutlass::TypeTraits&lt; double &gt;</a>
, <a class="el" href="structcutlass_1_1TypeTraits_3_01float_01_4.html#a571cc731a77f526150254679e8b4e9d3">cutlass::TypeTraits&lt; float &gt;</a>
, <a class="el" href="structcutlass_1_1TypeTraits_3_01half__t_01_4.html#a871eef488ed6c82367cc62fc044b8781">cutlass::TypeTraits&lt; half_t &gt;</a>
, <a class="el" href="structcutlass_1_1TypeTraits_3_01int_01_4.html#a5cdc27bfac5d2e9cc6173af77e492466">cutlass::TypeTraits&lt; int &gt;</a>
, <a class="el" href="structcutlass_1_1TypeTraits_3_01int64__t_01_4.html#afa3be2c14e04b3b5d700f3e7400acc9f">cutlass::TypeTraits&lt; int64_t &gt;</a>
, <a class="el" href="structcutlass_1_1TypeTraits_3_01int8__t_01_4.html#aafce09049d9cd82c79aab4e652e5043b">cutlass::TypeTraits&lt; int8_t &gt;</a>
, <a class="el" href="structcutlass_1_1TypeTraits_3_01uint64__t_01_4.html#aca3cc5f4bfd6cdfc3b032c0f4a190b80">cutlass::TypeTraits&lt; uint64_t &gt;</a>
, <a class="el" href="structcutlass_1_1TypeTraits_3_01uint8__t_01_4.html#ac014e1dc998a6bbd4fcc30f4544ce2be">cutlass::TypeTraits&lt; uint8_t &gt;</a>
, <a class="el" href="structcutlass_1_1TypeTraits_3_01unsigned_01_4.html#ad5047bae9ee9464f636be5269237185b">cutlass::TypeTraits&lt; unsigned &gt;</a>
</li>
<li>InterleavedEpilogue()
: <a class="el" href="classcutlass_1_1epilogue_1_1threadblock_1_1InterleavedEpilogue.html#a2e1f4ab98f9c69170bbcf05037c78eaf">cutlass::epilogue::threadblock::InterleavedEpilogue&lt; Shape_, WarpMmaOperator_, PartitionsK, OutputTileIterator_, AccumulatorFragmentIterator_, OutputOp_, InterleavedK, IsBetaZero &gt;</a>
</li>
<li>InterleavedPredicatedTileIterator()
: <a class="el" href="classcutlass_1_1epilogue_1_1threadblock_1_1InterleavedPredicatedTileIterator.html#aab0960ebd371ed02c4c7c5f8e2c2caf5">cutlass::epilogue::threadblock::InterleavedPredicatedTileIterator&lt; ThreadMap_, Element_, InterleavedK &gt;</a>
</li>
<li>InterleavedTile
: <a class="el" href="structcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOpAccumulatorTileIterator_1_1Policy.html#abc6ad2f60f697528ad075e0f4c7231b6">cutlass::gemm::warp::MmaVoltaTensorOpAccumulatorTileIterator&lt; Shape_, Element_, Layout_, InstructionShape_, OpDelta_ &gt;::Policy</a>
</li>
<li>InterleavedTileShape
: <a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1FragmentIteratorVoltaTensorOp_3_01WarpShape___00_01gemm_1_1Gdb805a2dc5571ac3b66e0fe6ffdcede2.html#a71122ebe81d816eabfc45a0b56572051">cutlass::epilogue::warp::FragmentIteratorVoltaTensorOp&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, float, layout::RowMajor &gt;</a>
, <a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1FragmentIteratorVoltaTensorOp_3_01WarpShape___00_01gemm_1_1G16e08718cffa0989cce3fe8dbc4b075b.html#ae7de02d970a2de2003f5340c0e1c6432">cutlass::epilogue::warp::FragmentIteratorVoltaTensorOp&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, half_t, layout::RowMajor &gt;</a>
, <a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1TileIteratorVoltaTensorOp_3_01WarpShape___00_01gemm_1_1GemmS2fe0c60b727c738c622c18fc3dd76644.html#a1d3db56c73a8f6d7e2be7be014ccc231">cutlass::epilogue::warp::TileIteratorVoltaTensorOp&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, float, layout::RowMajor &gt;</a>
, <a class="el" href="classcutlass_1_1epilogue_1_1warp_1_1TileIteratorVoltaTensorOp_3_01WarpShape___00_01gemm_1_1GemmSa0ceeeddc22575876eb977da7f5416a8.html#aa505394127d3158e94d6b9b7a002056b">cutlass::epilogue::warp::TileIteratorVoltaTensorOp&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, half_t, layout::RowMajor &gt;</a>
, <a class="el" href="structcutlass_1_1epilogue_1_1warp_1_1VoltaTensorOpPolicy_3_01WarpShape___00_01gemm_1_1GemmShape_136ce744d4c1c6e8707f5a9785196194.html#a94d42cddc58ca3112e572c2e7e201d0e">cutlass::epilogue::warp::VoltaTensorOpPolicy&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, float, layout::RowMajor &gt;</a>
, <a class="el" href="structcutlass_1_1epilogue_1_1warp_1_1VoltaTensorOpPolicy_3_01WarpShape___00_01gemm_1_1GemmShape_1d48185f49e4d066f8e9327bf0856b7f.html#a174c54d744433e9fb55d70b9d3298a3f">cutlass::epilogue::warp::VoltaTensorOpPolicy&lt; WarpShape_, gemm::GemmShape&lt; 32, 32, 4 &gt;, half_t, layout::RowMajor &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOp.html#abe80353f8640dd272466e6425925f619">cutlass::gemm::warp::MmaVoltaTensorOp&lt; Shape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, Policy_, Enable &gt;</a>
</li>
<li>IntType
: <a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1RandomGaussianFunc.html#ad4d61c5ff2534d18ce26fed88a17c937">cutlass::reference::device::detail::RandomGaussianFunc&lt; Element &gt;</a>
, <a class="el" href="structcutlass_1_1reference_1_1device_1_1detail_1_1RandomUniformFunc.html#a82cc92fab4650a7a17916b92a940b55c">cutlass::reference::device::detail::RandomUniformFunc&lt; Element &gt;</a>
</li>
<li>Invalid
: <a class="el" href="structcutlass_1_1Distribution.html#a499f4023e0d42356ce71d38cc32bf92aa2ff14122c59a823654b84764f68e597b">cutlass::Distribution</a>
</li>
<li>inverse()
: <a class="el" href="classcutlass_1_1layout_1_1ColumnMajor.html#a4d2ea34ea48b0702da0d20ffcd21be30">cutlass::layout::ColumnMajor</a>
, <a class="el" href="structcutlass_1_1layout_1_1ColumnMajorBlockLinear.html#aaf2b7966340292ba65e3cb6163b4348a">cutlass::layout::ColumnMajorBlockLinear&lt; BlockRows, BlockColumns &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1ColumnMajorInterleaved.html#a7e5a49fed5354bfaf49213bf9b2483dd">cutlass::layout::ColumnMajorInterleaved&lt; Interleave &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1ColumnMajorTensorOpMultiplicandCongruous.html#ae7834e63ec3b61d181f847011be413ab">cutlass::layout::ColumnMajorTensorOpMultiplicandCongruous&lt; ElementSize, Crosswise &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1ColumnMajorTensorOpMultiplicandCrosswise.html#a09eaa824aea4b98af2e71c14f572fd56">cutlass::layout::ColumnMajorTensorOpMultiplicandCrosswise&lt; ElementSize, Crosswise &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1ColumnMajorVoltaTensorOpMultiplicandBCongruous.html#ab83e570cf85596c1639c8eac50ae8597">cutlass::layout::ColumnMajorVoltaTensorOpMultiplicandBCongruous&lt; ElementSize &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1ColumnMajorVoltaTensorOpMultiplicandCongruous.html#adc9104c8495005e5df53accd5aca86d4">cutlass::layout::ColumnMajorVoltaTensorOpMultiplicandCongruous&lt; ElementSize &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1ColumnMajorVoltaTensorOpMultiplicandCrosswise.html#a5b7ac45dcec3ffdee65117ab9d9c439d">cutlass::layout::ColumnMajorVoltaTensorOpMultiplicandCrosswise&lt; ElementSize, KBlock &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1ContiguousMatrix.html#ab85bae46a22d05a7d216aefa68188c7f">cutlass::layout::ContiguousMatrix</a>
, <a class="el" href="classcutlass_1_1layout_1_1PitchLinear.html#a1e71c63f9ba751f1415492ef581fd93a">cutlass::layout::PitchLinear</a>
, <a class="el" href="classcutlass_1_1layout_1_1RowMajor.html#ac2508255cec15a07a2c082f9a58b7e6d">cutlass::layout::RowMajor</a>
, <a class="el" href="structcutlass_1_1layout_1_1RowMajorBlockLinear.html#abcb565659a5e7e3a601eb993d43ea71c">cutlass::layout::RowMajorBlockLinear&lt; BlockRows, BlockColumns &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1RowMajorInterleaved.html#a34770f321b9fefd9c8052abbb1643deb">cutlass::layout::RowMajorInterleaved&lt; Interleave &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1RowMajorTensorOpMultiplicandCongruous.html#a67e0749e995824bc35a8cde478f7c631">cutlass::layout::RowMajorTensorOpMultiplicandCongruous&lt; ElementSize, Crosswise &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1RowMajorTensorOpMultiplicandCrosswise.html#a1fa92e43f3d6b1874f9fe5c1f87b8ee0">cutlass::layout::RowMajorTensorOpMultiplicandCrosswise&lt; ElementSize, Crosswise &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1RowMajorVoltaTensorOpMultiplicandBCongruous.html#ab5bc99d0b654ac8a086e1f8012fe2f16">cutlass::layout::RowMajorVoltaTensorOpMultiplicandBCongruous&lt; ElementSize &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1RowMajorVoltaTensorOpMultiplicandCongruous.html#a49a401c647d68b6f845ccd704e7328bb">cutlass::layout::RowMajorVoltaTensorOpMultiplicandCongruous&lt; ElementSize &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1RowMajorVoltaTensorOpMultiplicandCrosswise.html#a36a240a540307090f18c6ed152deeae2">cutlass::layout::RowMajorVoltaTensorOpMultiplicandCrosswise&lt; ElementSize, KBlock &gt;</a>
, <a class="el" href="classcutlass_1_1layout_1_1TensorNHWC.html#ac172f5707f9eb601046c8fde392a2bf6">cutlass::layout::TensorNHWC</a>
, <a class="el" href="structcutlass_1_1layout_1_1TensorOpMultiplicandCongruous.html#a6ef0968585fcb6bb564e1ea6c4bdc9a3">cutlass::layout::TensorOpMultiplicandCongruous&lt; ElementSize, Crosswise &gt;</a>
, <a class="el" href="structcutlass_1_1layout_1_1TensorOpMultiplicandCrosswise.html#ac4ff3d7bfc5303cb2dee432a7cc430da">cutlass::layout::TensorOpMultiplicandCrosswise&lt; ElementSize, Crosswise &gt;</a>
</li>
<li>is_bounded
: <a class="el" href="structstd_1_1numeric__limits_3_01cutlass_1_1half__t_01_4.html#a59737f49161b87c259683c26737f42c2">std::numeric_limits&lt; cutlass::half_t &gt;</a>
</li>
<li>is_exact
: <a class="el" href="structstd_1_1numeric__limits_3_01cutlass_1_1half__t_01_4.html#a7413e9cd24eb03a86cc5f2d47c49db3e">std::numeric_limits&lt; cutlass::half_t &gt;</a>
</li>
<li>is_iec559
: <a class="el" href="structstd_1_1numeric__limits_3_01cutlass_1_1half__t_01_4.html#afc86b85a1fe209658a50d8c06e54cb77">std::numeric_limits&lt; cutlass::half_t &gt;</a>
</li>
<li>is_integer
: <a class="el" href="structstd_1_1numeric__limits_3_01cutlass_1_1half__t_01_4.html#a2a2ec168a6f0e9f55dc42d3b3e5fff25">std::numeric_limits&lt; cutlass::half_t &gt;</a>
</li>
<li>is_modulo
: <a class="el" href="structstd_1_1numeric__limits_3_01cutlass_1_1half__t_01_4.html#a640a034527a3577039053113bc1c5e46">std::numeric_limits&lt; cutlass::half_t &gt;</a>
</li>
<li>is_signed
: <a class="el" href="structstd_1_1numeric__limits_3_01cutlass_1_1half__t_01_4.html#ab3a169117baca2e7fae33846caa5dbfd">std::numeric_limits&lt; cutlass::half_t &gt;</a>
</li>
<li>is_source_ever_needed()
: <a class="el" href="classcutlass_1_1epilogue_1_1thread_1_1Convert.html#a3dbcb283d1a62392cb0f00d7334952ea">cutlass::epilogue::thread::Convert&lt; ElementOutput_, Count, ElementAccumulator_, Round &gt;</a>
</li>
<li>is_source_needed()
: <a class="el" href="classcutlass_1_1epilogue_1_1thread_1_1Convert.html#a7dc10ee38d2433e1e0d5cb4edac38a42">cutlass::epilogue::thread::Convert&lt; ElementOutput_, Count, ElementAccumulator_, Round &gt;</a>
, <a class="el" href="classcutlass_1_1epilogue_1_1thread_1_1LinearCombination.html#a0c576129fa70b60d7ad81ebbc8b8f22d">cutlass::epilogue::thread::LinearCombination&lt; ElementOutput_, Count, ElementAccumulator_, ElementCompute_, Round &gt;</a>
, <a class="el" href="classcutlass_1_1epilogue_1_1thread_1_1LinearCombinationClamp.html#a3567f90bb09ece3311af0c28e6784c91">cutlass::epilogue::thread::LinearCombinationClamp&lt; ElementOutput_, Count, ElementAccumulator_, ElementCompute_, Round &gt;</a>
, <a class="el" href="classcutlass_1_1epilogue_1_1thread_1_1LinearCombinationRelu.html#a6ba2de177b25375afc33043d665a5114">cutlass::epilogue::thread::LinearCombinationRelu&lt; ElementOutput_, Count, ElementAccumulator_, ElementCompute_, Round &gt;</a>
, <a class="el" href="classcutlass_1_1epilogue_1_1thread_1_1LinearCombinationRelu_3_01ElementOutput___00_01Count_00_01int_00_01float_00_01Round_01_4.html#af360ea56761af5fe2904ad1d7ff799c3">cutlass::epilogue::thread::LinearCombinationRelu&lt; ElementOutput_, Count, int, float, Round &gt;</a>
</li>
<li>is_specialized
: <a class="el" href="structstd_1_1numeric__limits_3_01cutlass_1_1half__t_01_4.html#ade4affb586360c5356a7939c1b343a40">std::numeric_limits&lt; cutlass::half_t &gt;</a>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</li>
<li>is_zero()
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 class="el" href="structcutlass_1_1PredicateVector.html#a29b6a3044b89d0b3ff98fd571e12cdd8">cutlass::PredicateVector&lt; kPredicates_, kPredicatesPerByte_, kPredicateStart_ &gt;</a>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</li>
<li>Iterations
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 class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1InterleavedOutputTileThreadMap.html#a0ea52743b659b1e8a46f8483f2d55837">cutlass::epilogue::threadblock::InterleavedOutputTileThreadMap&lt; WarpCount_, MmaCount_, Threads, ElementsPerAccess, ElementSize &gt;</a>
, <a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1OutputTileOptimalThreadMap_1_1CompactedThreadMap.html#a2185bd2f59bf24db4bced8b5563bf0dd">cutlass::epilogue::threadblock::OutputTileOptimalThreadMap&lt; Shape_, Count_, Threads, ElementsPerAccess, ElementSize &gt;::CompactedThreadMap</a>
, <a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1OutputTileOptimalThreadMap.html#a04eb6ef0d1ec171dd155f4d967b96eec">cutlass::epilogue::threadblock::OutputTileOptimalThreadMap&lt; Shape_, Count_, Threads, ElementsPerAccess, ElementSize &gt;</a>
, <a class="el" href="structcutlass_1_1epilogue_1_1threadblock_1_1OutputTileThreadMap.html#a6abb4ca7c8623a17392f7f90a7a66142">cutlass::epilogue::threadblock::OutputTileThreadMap&lt; ThreadMap_, Shape_, Iterations_, Delta_, Count_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kA_00_01Element_67ca7e11a38e38f2c51b84767654a90f.html#a22a914051aba2f21f3d04e86ce0b937c">cutlass::gemm::warp::MmaSimtTileIterator&lt; Shape_, Operand::kA, Element_, layout::ColumnMajor, Policy_, PartitionsK, PartitionGroupSize &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kA_00_01Element_f0ce904a9294556f15e1cc9cf7c99a93.html#ad9c4b211af0ed6345eddfefe08d870e2">cutlass::gemm::warp::MmaSimtTileIterator&lt; Shape_, Operand::kA, Element_, layout::ColumnMajorInterleaved&lt; 4 &gt;, Policy_, PartitionsK, PartitionGroupSize &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kB_00_01Element_ea0a4e7ce3cd5d25cabf79383efdf4d9.html#ac30fd882cb82be978023c1218a3a4ec0">cutlass::gemm::warp::MmaSimtTileIterator&lt; Shape_, Operand::kB, Element_, layout::RowMajor, Policy_, PartitionsK, PartitionGroupSize &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kB_00_01Element_ada156b62fcbdce47009c5bf1321c92c.html#a6ce04512aab1d8541875a676ad1b2f63">cutlass::gemm::warp::MmaSimtTileIterator&lt; Shape_, Operand::kB, Element_, layout::RowMajorInterleaved&lt; 4 &gt;, Policy_, PartitionsK, PartitionGroupSize &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kC_00_01Element_4ccafbc821b3a55cd532602442a74031.html#a7029869e74e5c70bc70af15981432051">cutlass::gemm::warp::MmaSimtTileIterator&lt; Shape_, Operand::kC, Element_, layout::ColumnMajor, Policy_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaSimtTileIterator_3_01Shape___00_01Operand_1_1kC_00_01Element_8f92ea79e85febb67169c4b2d94b1b20.html#a63e752f34e87c0d2e546ec8abc0727b8">cutlass::gemm::warp::MmaSimtTileIterator&lt; Shape_, Operand::kC, Element_, layout::RowMajor, Policy_ &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1PitchLinear2DThreadTileStripminedThreadMap_3_01Shape___00_01Thread0082c3467229b12cc9dd996283ee7160.html#a1ccecc2d9406a5d052b2ed15a74c1abe">cutlass::transform::PitchLinear2DThreadTileStripminedThreadMap&lt; Shape_, Threads, cutlass::layout::PitchLinearShape&lt; 4, 4 &gt; &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1PitchLinearStripminedThreadMap.html#a784640fa418b1d9b55f912f65925e705">cutlass::transform::PitchLinearStripminedThreadMap&lt; Shape_, Threads, ElementsPerAccess &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1PitchLinearTilePolicyStripminedThreadContiguous.html#ade1dd4814b42ce2dfb113a17170bddae">cutlass::transform::PitchLinearTilePolicyStripminedThreadContiguous&lt; Shape, Threads, ElementsPerAccess &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1PitchLinearTilePolicyStripminedThreadStrided.html#a23ac1cd1182c835a03d220256e4e78d1">cutlass::transform::PitchLinearTilePolicyStripminedThreadStrided&lt; Shape, Threads, ElementsPerAccess &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1PitchLinearWarpRakedThreadMap.html#ac6f4333d33283b867d120c6c43db9036">cutlass::transform::PitchLinearWarpRakedThreadMap&lt; Shape_, Threads, WarpThreadArrangement_, ElementsPerAccess &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1PitchLinearWarpStripedThreadMap.html#a967b87df4c12d47bc73594745efc570d">cutlass::transform::PitchLinearWarpStripedThreadMap&lt; Shape_, Threads, WarpThreadArrangement_, ElementsPerAccess &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1TransposePitchLinearThreadMap2DThreadTile.html#a5efb60f867ee631dbd2b93042deeda7d">cutlass::transform::TransposePitchLinearThreadMap2DThreadTile&lt; ThreadMap_ &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1TransposePitchLinearThreadMap.html#aaff78ad7b55c3043f65db8072489da43">cutlass::transform::TransposePitchLinearThreadMap&lt; ThreadMap_, WarpThreadArrangement_ &gt;</a>
, <a class="el" href="structcutlass_1_1transform_1_1TransposePitchLinearThreadMapSimt.html#a2cb2dcc7a1f80ea28aa7ca248401884f">cutlass::transform::TransposePitchLinearThreadMapSimt&lt; ThreadMap_ &gt;</a>
</li>
<li>iterator()
: <a class="el" href="classcutlass_1_1Array_3_01T_00_01N_00_01false_01_4_1_1iterator.html#adb69680f23a0ba9bbe107900fa537228">cutlass::Array&lt; T, N, false &gt;::iterator</a>
, <a class="el" href="classcutlass_1_1Array_3_01T_00_01N_00_01true_01_4_1_1iterator.html#ac50cf5822b023f4752d80f71963990cb">cutlass::Array&lt; T, N, true &gt;::iterator</a>
</li>
<li>Iterator()
: <a class="el" href="classcutlass_1_1PredicateVector_1_1Iterator.html#a91b7d25cbd64e696ef23c87671f0b077">cutlass::PredicateVector&lt; kPredicates_, kPredicatesPerByte_, kPredicateStart_ &gt;::Iterator</a>
</li>
<li>IteratorA
: <a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1DefaultGemv.html#a359698a3b1768f1ca5ce3032ab133f89">cutlass::gemm::kernel::DefaultGemv&lt; ThreadBlockShape_, ThreadShape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementCD_, LayoutCD_, ElementAccumulator_ &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultGemvCore.html#ac41f77fed60135f81a10adf0eb0b93ea">cutlass::gemm::threadblock::DefaultGemvCore&lt; Shape_, ThreadShape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_ &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMma_3_01ElementA_00_01LayoutA_00_01kAlignmentA_0010764e1fd5a3251a57eddafbd83eab8e.html#aa3a2bf4b00b6e5db0385c49f5a3136c8">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMma_3_01ElementA_00_01LayoutA_00_01kAlignmentA_00c67c16f9881e4f2fda76d8ed83ebabd6.html#aed25b7eeff13af9a122fbf088c655137">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMma_3_01ElementA_00_01LayoutA_00_01kAlignmentA_00ce36642cae579bce6605ff8edde3c6ab.html#a9c1ee7eaf97be7ac194306d11d8171e3">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMma_3_01int8__t_00_01LayoutA_00_01kAlignmentA_00_07e7230d4011ada5e22cfcb29103b696.html#a42f485b11b46f1c9620a29c6095a865f">cutlass::gemm::threadblock::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="classcutlass_1_1gemm_1_1threadblock_1_1Gemv.html#ab636c175e7a3b9e4628ef4766da3938d">cutlass::gemm::threadblock::Gemv&lt; Core_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1threadblock_1_1MmaPipelined.html#a258197bea0d4894569c627c56a64203a">cutlass::gemm::threadblock::MmaPipelined&lt; Shape_, IteratorA_, SmemIteratorA_, IteratorB_, SmemIteratorB_, ElementC_, LayoutC_, Policy_, TransformA_, TransformB_, Enable &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1threadblock_1_1MmaSingleStage.html#a21d381777ba328bf54763a270562a0c0">cutlass::gemm::threadblock::MmaSingleStage&lt; Shape_, IteratorA_, SmemIteratorA_, IteratorB_, SmemIteratorB_, ElementC_, LayoutC_, Policy_, Enable &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaComplexTensorOp_3_01Shape___00_01complex_3_01RealElementA_01_146441010dad1f40eb51b6dae3ded216.html#af9f720bfdb70b44963b60afd41e6a4ed">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaSimt.html#a0b8bd78fb8ff1d7e8a4d4072a43ea3cb">cutlass::gemm::warp::MmaSimt&lt; Shape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, Policy_, PartitionsK, Enable &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOp.html#a794384bbf37bce5504232a862dd1e089">cutlass::gemm::warp::MmaTensorOp&lt; Shape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, Policy_, PartitionsK_, AccumulatorsInRowMajor, PartitionsN_, Enable &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOp.html#a4606c0e308f3e8bdc1ff4c63e951d550">cutlass::gemm::warp::MmaVoltaTensorOp&lt; Shape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, Policy_, Enable &gt;</a>
</li>
<li>IteratorB
: <a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1DefaultGemv.html#a0cfef44420e5daa52b9c15bdbab2a1c5">cutlass::gemm::kernel::DefaultGemv&lt; ThreadBlockShape_, ThreadShape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementCD_, LayoutCD_, ElementAccumulator_ &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultGemvCore.html#a896e3d1e0229839fa873d0b948cd48ef">cutlass::gemm::threadblock::DefaultGemvCore&lt; Shape_, ThreadShape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_ &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMma_3_01ElementA_00_01LayoutA_00_01kAlignmentA_0010764e1fd5a3251a57eddafbd83eab8e.html#a3c848ad770bb5309c2abbe63c19db604">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMma_3_01ElementA_00_01LayoutA_00_01kAlignmentA_00c67c16f9881e4f2fda76d8ed83ebabd6.html#ad3cee865755ea98da79780c60a102f70">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMma_3_01ElementA_00_01LayoutA_00_01kAlignmentA_00ce36642cae579bce6605ff8edde3c6ab.html#a845669c52313e80e4cd6c5f846332d64">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMma_3_01int8__t_00_01LayoutA_00_01kAlignmentA_00_07e7230d4011ada5e22cfcb29103b696.html#af5488e28ff5fec5af06f73b984be04b9">cutlass::gemm::threadblock::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="classcutlass_1_1gemm_1_1threadblock_1_1Gemv.html#aa9c85beae49e786016b644868e47fce9">cutlass::gemm::threadblock::Gemv&lt; Core_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1threadblock_1_1MmaPipelined.html#a3989c584460e66c75a3f37ef91ecf06d">cutlass::gemm::threadblock::MmaPipelined&lt; Shape_, IteratorA_, SmemIteratorA_, IteratorB_, SmemIteratorB_, ElementC_, LayoutC_, Policy_, TransformA_, TransformB_, Enable &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1threadblock_1_1MmaSingleStage.html#af21b952e10f73ed2f5e67c8a8501c562">cutlass::gemm::threadblock::MmaSingleStage&lt; Shape_, IteratorA_, SmemIteratorA_, IteratorB_, SmemIteratorB_, ElementC_, LayoutC_, Policy_, Enable &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaComplexTensorOp_3_01Shape___00_01complex_3_01RealElementA_01_146441010dad1f40eb51b6dae3ded216.html#afca74199eea6e5eaaf791202684b5245">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaSimt.html#a1e8c132a89e5fef8120efe67f968f337">cutlass::gemm::warp::MmaSimt&lt; Shape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, Policy_, PartitionsK, Enable &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOp.html#a83f46087d7f39e4e039e74cbb79e3b89">cutlass::gemm::warp::MmaTensorOp&lt; Shape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, Policy_, PartitionsK_, AccumulatorsInRowMajor, PartitionsN_, Enable &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOp.html#a8b4342792d286cdf9a5e66f8ce486add">cutlass::gemm::warp::MmaVoltaTensorOp&lt; Shape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, Policy_, Enable &gt;</a>
</li>
<li>IteratorC
: <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultGemvCore.html#aa4880e8e43b0f9c704c1e1b15f8a5184">cutlass::gemm::threadblock::DefaultGemvCore&lt; Shape_, ThreadShape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1threadblock_1_1Gemv.html#acddb7f0d5146c1153b93f8ec6c26af06">cutlass::gemm::threadblock::Gemv&lt; Core_ &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaComplexTensorOp_3_01Shape___00_01complex_3_01RealElementA_01_146441010dad1f40eb51b6dae3ded216.html#a91ed53ba48c21c9b8898eeef378e4a33">cutlass::gemm::warp::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="classcutlass_1_1gemm_1_1warp_1_1MmaSimt.html#abf730ef40b6e1ebf604da959025e211f">cutlass::gemm::warp::MmaSimt&lt; Shape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, Policy_, PartitionsK, Enable &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaTensorOp.html#a40ea7a332afaa4e6476020a7357ed541">cutlass::gemm::warp::MmaTensorOp&lt; Shape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, Policy_, PartitionsK_, AccumulatorsInRowMajor, PartitionsN_, Enable &gt;</a>
, <a class="el" href="classcutlass_1_1gemm_1_1warp_1_1MmaVoltaTensorOp.html#a0f73a810a632772db186ce5275311eb9">cutlass::gemm::warp::MmaVoltaTensorOp&lt; Shape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_, Policy_, Enable &gt;</a>
</li>
<li>IteratorCD
: <a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1DefaultGemv.html#aa3c18b202147a4896fc51122a0cb17b3">cutlass::gemm::kernel::DefaultGemv&lt; ThreadBlockShape_, ThreadShape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementCD_, LayoutCD_, ElementAccumulator_ &gt;</a>
</li>
<li>IteratorPolicyA
: <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultGemvCore.html#a1b79b6a56b45f96cde340f7cae11c182">cutlass::gemm::threadblock::DefaultGemvCore&lt; Shape_, ThreadShape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_ &gt;</a>
</li>
<li>IteratorPolicyB
: <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultGemvCore.html#a2c32ed46fa439c4c758fac0618d3eae8">cutlass::gemm::threadblock::DefaultGemvCore&lt; Shape_, ThreadShape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_ &gt;</a>
</li>
<li>IteratorPolicyC
: <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultGemvCore.html#a792e8c1801748e8c3a724f729f540f7e">cutlass::gemm::threadblock::DefaultGemvCore&lt; Shape_, ThreadShape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, LayoutC_ &gt;</a>
</li>
<li>IteratorPolicyCD
: <a class="el" href="structcutlass_1_1gemm_1_1kernel_1_1DefaultGemv.html#aca6a3e28031cffaf13dfa5316c8f0f0b">cutlass::gemm::kernel::DefaultGemv&lt; ThreadBlockShape_, ThreadShape_, ElementA_, LayoutA_, ElementB_, LayoutB_, ElementCD_, LayoutCD_, ElementAccumulator_ &gt;</a>
</li>
<li>IteratorThreadMapA
: <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShab94a11a77dd0565102710907089acee0.html#a0f450843b1adf93a82d8fc2e3c58f769">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShafafd5c61db86cbfe90863578ddd11092.html#a26456fb7a61b7ace3225fc91e9b1f68c">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha46446d1e3871e31d2e728f710d78c8c1.html#ac80eca77d0a29a9178b266a4a5deaae9">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha8da7a0cfbbe859b701fdd9f2b8566aa7.html#aa2ba14171d7b2cef99a3db1419ea41fb">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha84e9f8afb6a4ca9f5dcd219b182d16e7.html#a269432cb8385f4cdc6783d122990410f">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha2c0d0b7cdb5c4bcb11e83c058eb65345.html#a368fe0ace8f4cfcfd886af813344a225">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha34a52cc7b2942e8c290f0032b6779b52.html#a6034c733de72b0f3060f92102298c9d6">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShaaf312aafe9da92ea9d417bcc12a8e7dc.html#a11b7dc567a5f7bb381827b8409530f6a">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha863d4139ccaa713bc4bde32c425f4067.html#ab265cae212878b47d8eb4050e5e63474">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShaf03a122202ad10acdc96f280106d678b.html#ac7be3a6fb81e9c39bc3d4cabd68922d0">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha69bef08ea63dd930f99d9788105873dd.html#a051a850ff5e96734a936a05b8247e69f">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha3adf608332a8c9ee7014fced0da8a9ca.html#ab71d03458d9f47b2d40bfb6bc52016b5">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShab7edfba3cdf43a07e3c4d719d87565a4.html#a5d3406fa412d5c96e6d16f95048c150e">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instruc803d38bc1e4618c07c47f54c87ae2678.html#a7bac08ac303fafc399f88bf74af2b55d">cutlass::gemm::threadblock::DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::ColumnMajor, ElementB_, layout::ColumnMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instrucf60fe02fcdd80d28b7fd419133465dcc.html#a9f433583080cc1b8a5dd35c0d932f02e">cutlass::gemm::threadblock::DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::ColumnMajor, ElementB_, layout::RowMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instruc2bf00737f4ad0a9da9a8be6d3e66c152.html#a507e4ab53b7c31ce1de65d588c74e3ff">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instruc24092ddc01fc83dabb7db4c14880fe60.html#a1ba5e5bc951f24bccc0db30b532ddb62">cutlass::gemm::threadblock::DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::RowMajor, ElementB_, layout::ColumnMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instruc4fee9f2965b8468bfb42b94a74527d22.html#ac6b64933f9b6628ebc1324fd13689143">cutlass::gemm::threadblock::DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::RowMajor, ElementB_, layout::RowMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a>
</li>
<li>IteratorThreadMapB
: <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShab94a11a77dd0565102710907089acee0.html#a89581e14b7ce8bc081a6a6a1ee9857d7">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShafafd5c61db86cbfe90863578ddd11092.html#a09be4f20512d02f18452dab77298b4da">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha46446d1e3871e31d2e728f710d78c8c1.html#a4143b289655ce6b6484be69cf746600b">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha8da7a0cfbbe859b701fdd9f2b8566aa7.html#ae7c43ed31d0702528c90f170bf772293">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha84e9f8afb6a4ca9f5dcd219b182d16e7.html#ad6be000724ac447ecd66dd0720f05162">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha2c0d0b7cdb5c4bcb11e83c058eb65345.html#a50e3846c0e25e25e81bdbfbacd44106c">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha34a52cc7b2942e8c290f0032b6779b52.html#a81b77d546a00570e7941b5e08b37269c">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShaaf312aafe9da92ea9d417bcc12a8e7dc.html#aede8b6f434f8bc14a337820d7c79abf2">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha863d4139ccaa713bc4bde32c425f4067.html#abd3a92eaa779a0148b42921ef1e6b82c">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShaf03a122202ad10acdc96f280106d678b.html#a30b09ef8128d5e6ad601ffba66af08c5">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha69bef08ea63dd930f99d9788105873dd.html#ab1a8a43bc393cdc8b9c690d3f1f4b2cb">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmSha3adf608332a8c9ee7014fced0da8a9ca.html#a84468e2419e1bc35fd1dec28329bf6ca">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01GemmShab7edfba3cdf43a07e3c4d719d87565a4.html#a4fbba145ad9d05c141f26636dfeacbd7">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instruc803d38bc1e4618c07c47f54c87ae2678.html#a1d752a45cea363b3b957ce23da1af694">cutlass::gemm::threadblock::DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::ColumnMajor, ElementB_, layout::ColumnMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instrucf60fe02fcdd80d28b7fd419133465dcc.html#a7d3f221de20d5cbf0a0a31fc8b99a272">cutlass::gemm::threadblock::DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::ColumnMajor, ElementB_, layout::RowMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instruc2bf00737f4ad0a9da9a8be6d3e66c152.html#a2465d25ece980d3652cc02acf0874c11">cutlass::gemm::threadblock::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="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instruc24092ddc01fc83dabb7db4c14880fe60.html#ad4ebd07d52c77fbc01844e197c6d6a09">cutlass::gemm::threadblock::DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::RowMajor, ElementB_, layout::ColumnMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a>
, <a class="el" href="structcutlass_1_1gemm_1_1threadblock_1_1DefaultMmaCore_3_01Shape___00_01WarpShape___00_01Instruc4fee9f2965b8468bfb42b94a74527d22.html#a8e981771cd56b2be97ead8f1ef585b1b">cutlass::gemm::threadblock::DefaultMmaCore&lt; Shape_, WarpShape_, InstructionShape_, ElementA_, layout::RowMajor, ElementB_, layout::RowMajor, ElementC_, LayoutC_, arch::OpClassTensorOp, 2, Operator_ &gt;</a>
CUTLASS v1.0 release
2018-05-17 02:44:56 +08:00
</li>
</ul>
</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