222 lines
180 KiB
HTML
222 lines
180 KiB
HTML
|
<!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"/>
|
||
|
|
<meta name="generator" content="Doxygen 1.8.11"/>
|
||
|
|
<title>CUTLASS: gemm_batched.h Source File</title>
|
||
|
|
<link href="tabs.css" rel="stylesheet" type="text/css"/>
|
||
|
|
<script type="text/javascript" src="jquery.js"></script>
|
||
|
|
<script type="text/javascript" src="dynsections.js"></script>
|
||
|
|
<link href="search/search.css" rel="stylesheet" type="text/css"/>
|
||
|
|
<script type="text/javascript" src="search/searchdata.js"></script>
|
||
|
|
<script type="text/javascript" src="search/search.js"></script>
|
||
|
|
<script type="text/javascript">
|
||
|
|
$(document).ready(function() { init_search(); });
|
||
|
|
</script>
|
||
|
|
<script type="text/x-mathjax-config">
|
||
|
|
MathJax.Hub.Config({
|
||
|
|
extensions: ["tex2jax.js"],
|
||
|
|
jax: ["input/TeX","output/HTML-CSS"],
|
||
|
|
});
|
||
|
|
</script><script type="text/javascript" src="http://cdn.mathjax.org/mathjax/latest/MathJax.js"></script>
|
||
|
|
<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;">
|
||
|
|
<td id="projectlogo"><img alt="Logo" src="cutlass-logo-small.png"/></td>
|
||
|
|
<td id="projectalign" style="padding-left: 0.5em;">
|
||
|
|
<div id="projectname">CUTLASS
|
||
|
|
</div>
|
||
|
|
<div id="projectbrief">CUDA Templates for Linear Algebra Subroutines and Solvers</div>
|
||
|
|
</td>
|
||
|
|
</tr>
|
||
|
|
</tbody>
|
||
|
|
</table>
|
||
|
|
</div>
|
||
|
|
<!-- end header part -->
|
||
|
|
<!-- Generated by Doxygen 1.8.11 -->
|
||
|
|
<script type="text/javascript">
|
||
|
|
var searchBox = new SearchBox("searchBox", "search",false,'Search');
|
||
|
|
</script>
|
||
|
|
<div id="navrow1" class="tabs">
|
||
|
|
<ul class="tablist">
|
||
|
|
<li><a href="index.html"><span>Main 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><a href="annotated.html"><span>Classes</span></a></li>
|
||
|
|
<li class="current"><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="files.html"><span>File List</span></a></li>
|
||
|
|
<li><a href="globals.html"><span>File Members</span></a></li>
|
||
|
|
</ul>
|
||
|
|
</div>
|
||
|
|
<!-- 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 id="nav-path" class="navpath">
|
||
|
|
<ul>
|
||
|
|
<li class="navelem"><a class="el" href="dir_d44c64559bbebec7f509842c48db8b23.html">include</a></li><li class="navelem"><a class="el" href="dir_6baf2bb612a2f0daa69af3101ede80a1.html">cutlass</a></li><li class="navelem"><a class="el" href="dir_9aa36bd9cfad59a1f88859a38871c977.html">gemm</a></li><li class="navelem"><a class="el" href="dir_36528dc2736efa40b421028b7309c671.html">device</a></li> </ul>
|
||
|
|
</div>
|
||
|
|
</div><!-- top -->
|
||
|
|
<div class="header">
|
||
|
|
<div class="headertitle">
|
||
|
|
<div class="title">device/gemm_batched.h</div> </div>
|
||
|
|
</div><!--header-->
|
||
|
|
<div class="contents">
|
||
|
|
<a href="device_2gemm__batched_8h.html">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno"> 1</span> <span class="comment">/***************************************************************************************************</span></div><div class="line"><a name="l00002"></a><span class="lineno"> 2</span> <span class="comment"> * Copyright (c) 2017-2019, NVIDIA CORPORATION. All rights reserved.</span></div><div class="line"><a name="l00003"></a><span class="lineno"> 3</span> <span class="comment"> *</span></div><div class="line"><a name="l00004"></a><span class="lineno"> 4</span> <span class="comment"> * Redistribution and use in source and binary forms, with or without modification, are permitted</span></div><div class="line"><a name="l00005"></a><span class="lineno"> 5</span> <span class="comment"> * provided that the following conditions are met:</span></div><div class="line"><a name="l00006"></a><span class="lineno"> 6</span> <span class="comment"> * * Redistributions of source code must retain the above copyright notice, this list of</span></div><div class="line"><a name="l00007"></a><span class="lineno"> 7</span> <span class="comment"> * conditions and the following disclaimer.</span></div><div class="line"><a name="l00008"></a><span class="lineno"> 8</span> <span class="comment"> * * Redistributions in binary form must reproduce the above copyright notice, this list of</span></div><div class="line"><a name="l00009"></a><span class="lineno"> 9</span> <span class="comment"> * conditions and the following disclaimer in the documentation and/or other materials</span></div><div class="line"><a name="l00010"></a><span class="lineno"> 10</span> <span class="comment"> * provided with the distribution.</span></div><div class="line"><a name="l00011"></a><span class="lineno"> 11</span> <span class="comment"> * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used</span></div><div class="line"><a name="l00012"></a><span class="lineno"> 12</span> <span class="comment"> * to endorse or promote products derived from this software without specific prior written</span></div><div class="line"><a name="l00013"></a><span class="lineno"> 13</span> <span class="comment"> * permission.</span></div><div class="line"><a name="l00014"></a><span class="lineno"> 14</span> <span class="comment"> *</span></div><div class="line"><a name="l00015"></a><span class="lineno"> 15</span> <span class="comment"> * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR</span></div><div class="line"><a name="l00016"></a><span class="lineno"> 16</span> <span class="comment"> * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND</span></div><div class="line"><a name="l00017"></a><span class="lineno"> 17</span> <span class="comment"> * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE</span></div><div class="line"><a name="l00018"></a><span class="lineno"> 18</span> <span class="comment"> * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,</span></div><div class="line"><a name="l00019"></a><span class="lineno"> 19</span> <span class="comment"> * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;</span></div><div class="line"><a name="l00020"></a><span class="lineno"> 20</span> <span class="comment"> * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,</span></div><div class="line"><a name="l00021"></a><span class="lineno"> 21</span> <span class="comment"> * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE</span></div><div class="line"><a name
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1kernel_1_1DefaultGemm_html"><div class="ttname"><a href="structcutlass_1_1gemm_1_1kernel_1_1DefaultGemm.html">cutlass::gemm::kernel::DefaultGemm</a></div><div class="ttdef"><b>Definition:</b> default_gemm.h:116</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_af8b282788223086b80fbb097b22459ec"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#af8b282788223086b80fbb097b22459ec">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::ThreadblockSwizzle</a></div><div class="ttdeci">ThreadblockSwizzle_ ThreadblockSwizzle</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:540</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a17d34b2884711522fafcfd7c7500955c"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a17d34b2884711522fafcfd7c7500955c">cutlass::gemm::device::GemmBatched::kAlignmentB</a></div><div class="ttdeci">static int const kAlignmentB</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:236</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5_html_a1727630fc0525724df28a75ccf2580b9"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5.html#a1727630fc0525724df28a75ccf2580b9">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::ref_A</a></div><div class="ttdeci">TensorRef< ElementA const, LayoutA > ref_A</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:580</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments_html_ae4450f06a6975191d94026865e445578"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments.html#ae4450f06a6975191d94026865e445578">cutlass::gemm::device::GemmBatched::Arguments::ref_D</a></div><div class="ttdeci">TensorRef< ElementC, LayoutC > ref_D</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:280</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments_html_aab4fc258e38ebcf9b430a5dee6daba5e"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments.html#aab4fc258e38ebcf9b430a5dee6daba5e">cutlass::gemm::device::GemmBatched::Arguments::problem_size</a></div><div class="ttdeci">GemmCoord problem_size</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:273</div></div>
|
||
|
|
<div class="ttc" id="namespacecutlass_html"><div class="ttname"><a href="namespacecutlass.html">cutlass</a></div><div class="ttdef"><b>Definition:</b> aligned_buffer.h:35</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_a3fd5c64783f88a7533801fef7d1375ad"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#a3fd5c64783f88a7533801fef7d1375ad">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::ElementB</a></div><div class="ttdeci">ElementB_ ElementB</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:526</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_a53ca4db66d0d2c96d9036d8eb7c6072b"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#a53ca4db66d0d2c96d9036d8eb7c6072b">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::operator()</a></div><div class="ttdeci">Status operator()(Arguments const &args, void *workspace=nullptr, cudaStream_t stream=nullptr)</div><div class="ttdoc">Runs the kernel using initialized state. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:689</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments_html_a05a1d9720fbb16a20b94049900b0d04f"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments.html#a05a1d9720fbb16a20b94049900b0d04f">cutlass::gemm::device::GemmBatched::Arguments::stride_D</a></div><div class="ttdeci">int64_t stride_D</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:281</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1Params_html_a4f18093b18b0b6dd01a5df0a3813cd40"><div class="ttname"><a href="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1Params.html#a4f18093b18b0b6dd01a5df0a3813cd40">cutlass::gemm::kernel::GemmBatched::Params::ref_D</a></div><div class="ttdeci">Epilogue::OutputTileIterator::TensorRef ref_D</div><div class="ttdef"><b>Definition:</b> kernel/gemm_batched.h:74</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments_html_a0befb9945aadcba460f4d1ad73020e9c"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments.html#a0befb9945aadcba460f4d1ad73020e9c">cutlass::gemm::device::GemmBatched::Arguments::Arguments</a></div><div class="ttdeci">CUTLASS_HOST_DEVICE Arguments(GemmCoord problem_size_, TensorRef< ElementA const, LayoutA > ref_A_, int64_t stride_A_, TensorRef< ElementB const, LayoutB > ref_B_, int64_t stride_B_, TensorRef< ElementC const, LayoutC > ref_C_, int64_t stride_C_, TensorRef< ElementC, LayoutC > ref_D_, int64_t stride_D_, typename EpilogueOutputOp::Params epilogue_, int batch_count_)</div><div class="ttdoc">Constructs an Arguments structure. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:295</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_a4aaaa6ca0e4b9f983fe37b4105fd058f"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#a4aaaa6ca0e4b9f983fe37b4105fd058f">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::LayoutB</a></div><div class="ttdeci">LayoutB_ LayoutB</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:527</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a6acd50cfc477e95dbcf0d4fbba5df65c"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a6acd50cfc477e95dbcf0d4fbba5df65c">cutlass::gemm::device::GemmBatched< ElementB, typename layout::LayoutTranspose< LayoutB >::type, ElementA, typename layout::LayoutTranspose< LayoutA >::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, Stages, kAlignmentB, kAlignmentA >::DefaultGemmKernel</a></div><div class="ttdeci">typename kernel::DefaultGemm< ElementA, LayoutA, kAlignmentA, ElementB, LayoutB, kAlignmentB, ElementC, LayoutC, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, kStages, false, Operator, false >::GemmKernel DefaultGemmKernel</div><div class="ttdoc">Define the kernel. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:262</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_abbd82c0f989a9d07e5e222db96386701"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#abbd82c0f989a9d07e5e222db96386701">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::can_implement</a></div><div class="ttdeci">static Status can_implement(Arguments const &args)</div><div class="ttdoc">Determines whether the GEMM can execute the given problem. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:654</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_a657e50fb03ea4d16f7b904920d9aa000"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#a657e50fb03ea4d16f7b904920d9aa000">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::ThreadblockShape</a></div><div class="ttdeci">ThreadblockShape_ ThreadblockShape</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:536</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_afcdd646be7e79a60bac8dede563c56fa"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#afcdd646be7e79a60bac8dede563c56fa">cutlass::gemm::device::GemmBatched::operator()</a></div><div class="ttdeci">Status operator()(cudaStream_t stream=nullptr)</div><div class="ttdoc">Runs the kernel using initialized state. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:443</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1GemmCoord_html"><div class="ttname"><a href="structcutlass_1_1gemm_1_1GemmCoord.html">cutlass::gemm::GemmCoord</a></div><div class="ttdef"><b>Definition:</b> include/cutlass/gemm/gemm.h:94</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments_html"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments.html">cutlass::gemm::device::GemmBatched::Arguments</a></div><div class="ttdoc">Argument structure. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:267</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a52b7263c5c86e900bcca681d07f19101"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a52b7263c5c86e900bcca681d07f19101">cutlass::gemm::device::GemmBatched< ElementB, typename layout::LayoutTranspose< LayoutB >::type, ElementA, typename layout::LayoutTranspose< LayoutA >::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, Stages, kAlignmentB, kAlignmentA >::Operator</a></div><div class="ttdeci">typename DefaultGemmConfiguration< OperatorClass, ArchTag, ElementB, ElementA, ElementC,ElementAccumulator >::Operator Operator</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:238</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1Params_html_ade55adc311c5561efe76f53ffd56d1f4"><div class="ttname"><a href="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1Params.html#ade55adc311c5561efe76f53ffd56d1f4">cutlass::gemm::kernel::GemmBatched::Params::ref_B</a></div><div class="ttdeci">Mma::IteratorB::TensorRef ref_B</div><div class="ttdef"><b>Definition:</b> kernel/gemm_batched.h:68</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5_html_a9f8a044d7b7439192dfe2bf488558ed3"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5.html#a9f8a044d7b7439192dfe2bf488558ed3">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::stride_C</a></div><div class="ttdeci">int64_t stride_C</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:585</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_a37600c0bf3570bc4b21c26b2b64fc54a"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#a37600c0bf3570bc4b21c26b2b64fc54a">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::OperatorClass</a></div><div class="ttdeci">OperatorClass_ OperatorClass</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:534</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_acb489f1bb4fed9e4314a6b6a3cbd04a9"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#acb489f1bb4fed9e4314a6b6a3cbd04a9">cutlass::gemm::device::GemmBatched< ElementB, typename layout::LayoutTranspose< LayoutB >::type, ElementA, typename layout::LayoutTranspose< LayoutA >::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, Stages, kAlignmentB, kAlignmentA >::LayoutB</a></div><div class="ttdeci">typename layout::LayoutTranspose< LayoutA >::type LayoutB</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:220</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_a681b145a9701109f9d72059bb874895b"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#a681b145a9701109f9d72059bb874895b">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::ArchTag</a></div><div class="ttdeci">ArchTag_ ArchTag</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:535</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_a3947c9b192bec2fad631334f31632353"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#a3947c9b192bec2fad631334f31632353">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::GemmKernel</a></div><div class="ttdeci">typename UnderlyingOperator::GemmKernel GemmKernel</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:570</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_ac2009bb52372115624aa5c4f75b720e5"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#ac2009bb52372115624aa5c4f75b720e5">cutlass::gemm::device::GemmBatched::get_workspace_size</a></div><div class="ttdeci">static size_t get_workspace_size(Arguments const &args)</div><div class="ttdoc">Gets the workspace size. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:361</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5_html_ac181dba327e605b6cde9de5c7f176e7c"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5.html#ac181dba327e605b6cde9de5c7f176e7c">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::stride_D</a></div><div class="ttdeci">int64_t stride_D</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:587</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_a3760f803bd2b31b3fdf47741caa950fa"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#a3760f803bd2b31b3fdf47741caa950fa">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::WarpShape</a></div><div class="ttdeci">WarpShape_ WarpShape</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:537</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5_html_af9c2fa1e0cc0456197c2cc0840c89982"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5.html#af9c2fa1e0cc0456197c2cc0840c89982">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::epilogue</a></div><div class="ttdeci">EpilogueOutputOp::Params epilogue</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:588</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_a9f0c7054068175c1891e4820857603c3"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#a9f0c7054068175c1891e4820857603c3">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::update</a></div><div class="ttdeci">Status update(Arguments const &args, void *workspace=nullptr)</div><div class="ttdoc">Lightweight update given a subset of arguments. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:672</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments_html_a11ef91161a92459d72b56144cd6b4495"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments.html#a11ef91161a92459d72b56144cd6b4495">cutlass::gemm::device::GemmBatched::Arguments::stride_A</a></div><div class="ttdeci">int64_t stride_A</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:275</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_ae073edad6dd4447d7f99c94f4cd0c1c8"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#ae073edad6dd4447d7f99c94f4cd0c1c8">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::InstructionShape</a></div><div class="ttdeci">InstructionShape_ InstructionShape</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:538</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_a428d8b1c4ac36040145a59d8e4cff3d2"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#a428d8b1c4ac36040145a59d8e4cff3d2">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::initialize</a></div><div class="ttdeci">Status initialize(Arguments const &args, void *workspace=nullptr, cudaStream_t stream=nullptr)</div><div class="ttdoc">Initializes GEMM state from arguments. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:666</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_abcae3d15f1ec2ee7ae93690c82fbee8a"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#abcae3d15f1ec2ee7ae93690c82fbee8a">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::run</a></div><div class="ttdeci">Status run(cudaStream_t stream=nullptr)</div><div class="ttdoc">Runs the kernel using initialized state. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:678</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_ae7f006ea8bc324d31de9dfbebc1b9327"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#ae7f006ea8bc324d31de9dfbebc1b9327">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::ElementAccumulator</a></div><div class="ttdeci">ElementAccumulator_ ElementAccumulator</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:533</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1Params_html_a08ecd763b6785dfe872a6e517dc731e6"><div class="ttname"><a href="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1Params.html#a08ecd763b6785dfe872a6e517dc731e6">cutlass::gemm::kernel::GemmBatched::Params::ref_C</a></div><div class="ttdeci">Epilogue::OutputTileIterator::TensorRef ref_C</div><div class="ttdef"><b>Definition:</b> kernel/gemm_batched.h:71</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1layout_1_1ColumnMajor_html"><div class="ttname"><a href="classcutlass_1_1layout_1_1ColumnMajor.html">cutlass::layout::ColumnMajor</a></div><div class="ttdoc">Mapping function for column-major matrices. </div><div class="ttdef"><b>Definition:</b> layout/matrix.h:142</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a24935d746d97b0c994c9a9ade820d2d0"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a24935d746d97b0c994c9a9ade820d2d0">cutlass::gemm::device::GemmBatched< ElementB, typename layout::LayoutTranspose< LayoutB >::type, ElementA, typename layout::LayoutTranspose< LayoutA >::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, Stages, kAlignmentB, kAlignmentA >::OperatorClass</a></div><div class="ttdeci">OperatorClass OperatorClass</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:227</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5_html_a302101a4e5c00c843b3c525ddb94c117"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5.html#a302101a4e5c00c843b3c525ddb94c117">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::stride_B</a></div><div class="ttdeci">int64_t stride_B</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:583</div></div>
|
||
|
|
<div class="ttc" id="default__gemm_8h_html"><div class="ttname"><a href="default__gemm_8h.html">default_gemm.h</a></div><div class="ttdoc">Default kernel-level GEMM definitions combine threadblock-scoped matrix multiply-add with the appropr...</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_ac1f190321a811fa91eec0096829b07ff"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#ac1f190321a811fa91eec0096829b07ff">cutlass::gemm::device::GemmBatched::kAlignmentC</a></div><div class="ttdeci">static int const kAlignmentC</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:237</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5_html_a17c4e381e91229a8ef15b18ee5ec073d"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5.html#a17c4e381e91229a8ef15b18ee5ec073d">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::ref_D</a></div><div class="ttdeci">TensorRef< ElementC, LayoutC > ref_D</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:586</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a74aece33b6fafe58db1b41a6b7b87729"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a74aece33b6fafe58db1b41a6b7b87729">cutlass::gemm::device::GemmBatched< ElementB, typename layout::LayoutTranspose< LayoutB >::type, ElementA, typename layout::LayoutTranspose< LayoutA >::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, Stages, kAlignmentB, kAlignmentA >::ThreadblockShape</a></div><div class="ttdeci">ThreadblockShape ThreadblockShape</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:229</div></div>
|
||
|
|
<div class="ttc" id="unioncutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1SharedStorage_html"><div class="ttname"><a href="unioncutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1SharedStorage.html">cutlass::gemm::kernel::GemmBatched::SharedStorage</a></div><div class="ttdoc">Shared memory storage structure. </div><div class="ttdef"><b>Definition:</b> kernel/gemm_batched.h:124</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments_html_a72bd469f15b44e492cf84658b5f09ad5"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments.html#a72bd469f15b44e492cf84658b5f09ad5">cutlass::gemm::device::GemmBatched::Arguments::Arguments</a></div><div class="ttdeci">CUTLASS_HOST_DEVICE Arguments()</div><div class="ttdoc">Default ctor. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:291</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1Params_html_a186903697a8ad274b6ac5416f7976c97"><div class="ttname"><a href="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1Params.html#a186903697a8ad274b6ac5416f7976c97">cutlass::gemm::kernel::GemmBatched::Params::grid_tiled_shape</a></div><div class="ttdeci">cutlass::gemm::GemmCoord grid_tiled_shape</div><div class="ttdef"><b>Definition:</b> kernel/gemm_batched.h:63</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a0dbb6d5185f223bb8242fc47a3b77757"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a0dbb6d5185f223bb8242fc47a3b77757">cutlass::gemm::device::GemmBatched< ElementB, typename layout::LayoutTranspose< LayoutB >::type, ElementA, typename layout::LayoutTranspose< LayoutA >::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, Stages, kAlignmentB, kAlignmentA >::InstructionShape</a></div><div class="ttdeci">InstructionShape InstructionShape</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:231</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_a75922fd7bcd77fbc714cd87681f692bf"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#a75922fd7bcd77fbc714cd87681f692bf">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::GemmBatched</a></div><div class="ttdeci">GemmBatched()</div><div class="ttdoc">Constructs the GEMM. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:634</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5_html_ad0469cc3e961d21e212d026bccf6fe1a"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5.html#ad0469cc3e961d21e212d026bccf6fe1a">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::problem_size</a></div><div class="ttdeci">GemmCoord problem_size</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:579</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a73837bda9ba209e546f6d996ede1afad"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a73837bda9ba209e546f6d996ede1afad">cutlass::gemm::device::GemmBatched::kStages</a></div><div class="ttdeci">static int const kStages</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:234</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_ab7c3e9a33a1c62513ec6eee3e2598df6"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#ab7c3e9a33a1c62513ec6eee3e2598df6">cutlass::gemm::device::GemmBatched::update</a></div><div class="ttdeci">Status update(Arguments const &args, void *workspace=nullptr)</div><div class="ttdoc">Lightweight update given a subset of arguments. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:396</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_html_aafddaefa35d27c76a89be8e692005615"><div class="ttname"><a href="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched.html#aafddaefa35d27c76a89be8e692005615">cutlass::gemm::kernel::GemmBatched::kThreadCount</a></div><div class="ttdeci">static int const kThreadCount</div><div class="ttdef"><b>Definition:</b> kernel/gemm_batched.h:58</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_acb4d53fbea4366349574091d68594558"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#acb4d53fbea4366349574091d68594558">cutlass::gemm::device::GemmBatched::can_implement</a></div><div class="ttdeci">static Status can_implement(Arguments const &args)</div><div class="ttdoc">Determines whether the GEMM can execute the given problem. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:332</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments_html_a3ce1385631b05430fa5dfc1e9a3671b8"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments.html#a3ce1385631b05430fa5dfc1e9a3671b8">cutlass::gemm::device::GemmBatched::Arguments::stride_C</a></div><div class="ttdeci">int64_t stride_C</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:279</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1Params_html"><div class="ttname"><a href="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1Params.html">cutlass::gemm::kernel::GemmBatched::Params</a></div><div class="ttdoc">Parameters structure. </div><div class="ttdef"><b>Definition:</b> kernel/gemm_batched.h:61</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1layout_1_1LayoutTranspose_html"><div class="ttname"><a href="structcutlass_1_1layout_1_1LayoutTranspose.html">cutlass::layout::LayoutTranspose</a></div><div class="ttdoc">Defines transposes of matrix layouts. </div><div class="ttdef"><b>Definition:</b> layout/matrix.h:921</div></div>
|
||
|
|
<div class="ttc" id="namespacecutlass_html_ac5a88c5840a28a9e0206b9cc7812a18daa4867e1466f5d067dbec566abfe5a67a"><div class="ttname"><a href="namespacecutlass.html#ac5a88c5840a28a9e0206b9cc7812a18daa4867e1466f5d067dbec566abfe5a67a">cutlass::Status::kErrorMisalignedOperand</a></div><div class="ttdoc">operands fail alignment requirements. </div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1TensorRef_html"><div class="ttname"><a href="classcutlass_1_1TensorRef.html">cutlass::TensorRef< ElementA const, LayoutA ></a></div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a008028eec1dd4b8c08128c6dfe44cce5"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a008028eec1dd4b8c08128c6dfe44cce5">cutlass::gemm::device::GemmBatched< ElementB, typename layout::LayoutTranspose< LayoutB >::type, ElementA, typename layout::LayoutTranspose< LayoutA >::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, Stages, kAlignmentB, kAlignmentA >::ElementC</a></div><div class="ttdeci">ElementC ElementC</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:222</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_ab5f57fac13e42a08d351ac48c2cc9992"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#ab5f57fac13e42a08d351ac48c2cc9992">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::UnderlyingArguments</a></div><div class="ttdeci">typename UnderlyingOperator::Arguments UnderlyingArguments</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:569</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_ac4ef1ac1e0876aaee5bff50dc09fe8a9"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#ac4ef1ac1e0876aaee5bff50dc09fe8a9">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::to_underlying_arguments</a></div><div class="ttdeci">static UnderlyingArguments to_underlying_arguments(Arguments const &args)</div><div class="ttdoc">Helper to construct a transposed equivalent for the underying GEMM operator. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:637</div></div>
|
||
|
|
<div class="ttc" id="namespacecutlass_html_ac5a88c5840a28a9e0206b9cc7812a18da4d3b5847a0b17037c6b69faf8b1d4d71"><div class="ttname"><a href="namespacecutlass.html#ac5a88c5840a28a9e0206b9cc7812a18da4d3b5847a0b17037c6b69faf8b1d4d71">cutlass::Status::kErrorInternal</a></div><div class="ttdoc">An error within CUTLASS occurred. </div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments_html_a48844293c34b9c44fe57f577370664ea"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments.html#a48844293c34b9c44fe57f577370664ea">cutlass::gemm::device::GemmBatched::Arguments::ref_B</a></div><div class="ttdeci">TensorRef< ElementB const, LayoutB > ref_B</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:276</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a8976ed5c5e404ee87deaea4455d0d960"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a8976ed5c5e404ee87deaea4455d0d960">cutlass::gemm::device::GemmBatched::kAlignmentA</a></div><div class="ttdeci">static int const kAlignmentA</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:235</div></div>
|
||
|
|
<div class="ttc" id="device__kernel_8h_html"><div class="ttname"><a href="device__kernel_8h.html">device_kernel.h</a></div><div class="ttdoc">Template for generic CUTLASS kernel. </div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_aa2670ac441f48f6a0a2071c67c743ab8"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#aa2670ac441f48f6a0a2071c67c743ab8">cutlass::gemm::device::GemmBatched::initialize</a></div><div class="ttdeci">Status initialize(Arguments const &args, void *workspace=nullptr, cudaStream_t stream=nullptr)</div><div class="ttdoc">Initializes GEMM state from arguments. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:366</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a1b685fd66f6dc2c572be067ef1396a89"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a1b685fd66f6dc2c572be067ef1396a89">cutlass::gemm::device::GemmBatched< ElementB, typename layout::LayoutTranspose< LayoutB >::type, ElementA, typename layout::LayoutTranspose< LayoutA >::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, Stages, kAlignmentB, kAlignmentA >::ThreadblockSwizzle</a></div><div class="ttdeci">ThreadblockSwizzle ThreadblockSwizzle</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:233</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_a494be150d3b809a4ecf66df682481905"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#a494be150d3b809a4ecf66df682481905">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::EpilogueOutputOp</a></div><div class="ttdeci">EpilogueOutputOp_ EpilogueOutputOp</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:539</div></div>
|
||
|
|
<div class="ttc" id="cutlass_8h_html_a28c2443a142676d3d71effdae1a986b1"><div class="ttname"><a href="cutlass_8h.html#a28c2443a142676d3d71effdae1a986b1">CUTLASS_HOST_DEVICE</a></div><div class="ttdeci">#define CUTLASS_HOST_DEVICE</div><div class="ttdef"><b>Definition:</b> cutlass.h:89</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a72a26fb286181aa5ca1fb66d9b385f7f"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a72a26fb286181aa5ca1fb66d9b385f7f">cutlass::gemm::device::GemmBatched::GemmBatched</a></div><div class="ttdeci">GemmBatched()</div><div class="ttdoc">Constructs the GEMM. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:329</div></div>
|
||
|
|
<div class="ttc" id="numeric__types_8h_html"><div class="ttname"><a href="numeric__types_8h.html">numeric_types.h</a></div><div class="ttdoc">Top-level include for all CUTLASS numeric types. </div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_aa2d43ad49fa686ded524cc5f26b36c69"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#aa2d43ad49fa686ded524cc5f26b36c69">cutlass::gemm::device::GemmBatched< ElementB, typename layout::LayoutTranspose< LayoutB >::type, ElementA, typename layout::LayoutTranspose< LayoutA >::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, Stages, kAlignmentB, kAlignmentA >::EpilogueOutputOp</a></div><div class="ttdeci">EpilogueOutputOp EpilogueOutputOp</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:232</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments_html_ac99ca8f9d8a0053e647a6c99b018bda5"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments.html#ac99ca8f9d8a0053e647a6c99b018bda5">cutlass::gemm::device::GemmBatched::Arguments::batch_count</a></div><div class="ttdeci">int batch_count</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:283</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_a00805989734182945f982cab23a5dca8"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#a00805989734182945f982cab23a5dca8">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::operator()</a></div><div class="ttdeci">Status operator()(cudaStream_t stream=nullptr)</div><div class="ttdoc">Runs the kernel using initialized state. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:684</div></div>
|
||
|
|
<div class="ttc" id="default__gemm__configuration_8h_html"><div class="ttname"><a href="default__gemm__configuration_8h.html">default_gemm_configuration.h</a></div><div class="ttdoc">Definitions for GEMM structures. </div></div>
|
||
|
|
<div class="ttc" id="kernel_2gemm__batched_8h_html"><div class="ttname"><a href="kernel_2gemm__batched_8h.html">gemm_batched.h</a></div><div class="ttdoc">Template for a pipelined GEMM kernel. Does not compute batching or support split-K. </div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_a52b9261576b5633e901719f7c21d3369"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#a52b9261576b5633e901719f7c21d3369">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::ElementA</a></div><div class="ttdeci">ElementA_ ElementA</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:523</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a9f9d11529b28ced91c4b05c2530b7a70"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a9f9d11529b28ced91c4b05c2530b7a70">cutlass::gemm::device::GemmBatched< ElementB, typename layout::LayoutTranspose< LayoutB >::type, ElementA, typename layout::LayoutTranspose< LayoutA >::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, Stages, kAlignmentB, kAlignmentA >::LayoutA</a></div><div class="ttdeci">typename layout::LayoutTranspose< LayoutB >::type LayoutA</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:217</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1Params_html_a05909ba49e633c7eeb0707166c72a4ee"><div class="ttname"><a href="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1Params.html#a05909ba49e633c7eeb0707166c72a4ee">cutlass::gemm::kernel::GemmBatched::Params::problem_size</a></div><div class="ttdeci">cutlass::gemm::GemmCoord problem_size</div><div class="ttdef"><b>Definition:</b> kernel/gemm_batched.h:62</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1layout_1_1RowMajor_html"><div class="ttname"><a href="classcutlass_1_1layout_1_1RowMajor.html">cutlass::layout::RowMajor</a></div><div class="ttdoc">Mapping function for row-major matrices. </div><div class="ttdef"><b>Definition:</b> layout/matrix.h:50</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_ac8bb1360bbc57bc63296cd48005c3c42"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#ac8bb1360bbc57bc63296cd48005c3c42">cutlass::gemm::device::GemmBatched< ElementB, typename layout::LayoutTranspose< LayoutB >::type, ElementA, typename layout::LayoutTranspose< LayoutA >::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, Stages, kAlignmentB, kAlignmentA >::ElementB</a></div><div class="ttdeci">ElementA ElementB</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:219</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_a3687659e826ba7f38bb060ad6020a739"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#a3687659e826ba7f38bb060ad6020a739">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::get_workspace_size</a></div><div class="ttdeci">static size_t get_workspace_size(Arguments const &args)</div><div class="ttdoc">Gets the workspace size. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:660</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a3fe2fcad97f15d63fa1a0214ef4861f2"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a3fe2fcad97f15d63fa1a0214ef4861f2">cutlass::gemm::device::GemmBatched< ElementB, typename layout::LayoutTranspose< LayoutB >::type, ElementA, typename layout::LayoutTranspose< LayoutA >::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, Stages, kAlignmentB, kAlignmentA >::ElementA</a></div><div class="ttdeci">ElementB ElementA</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:216</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments_html_ab0955b722ad4ea0217f725e34b3bcfbe"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments.html#ab0955b722ad4ea0217f725e34b3bcfbe">cutlass::gemm::device::GemmBatched::Arguments::ref_C</a></div><div class="ttdeci">TensorRef< ElementC const, LayoutC > ref_C</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:278</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5_html_a2129a4dccbd73f8c0f26b08ce5a5cb28"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5.html#a2129a4dccbd73f8c0f26b08ce5a5cb28">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::Arguments</a></div><div class="ttdeci">CUTLASS_HOST_DEVICE Arguments(GemmCoord problem_size_, TensorRef< ElementA const, LayoutA > ref_A_, int64_t stride_A_, TensorRef< ElementB const, LayoutB > ref_B_, int64_t stride_B_, TensorRef< ElementC const, LayoutC > ref_C_, int64_t stride_C_, TensorRef< ElementC, LayoutC > ref_D_, int64_t stride_D_, typename EpilogueOutputOp::Params epilogue_, int batch_count_)</div><div class="ttdoc">Constructs an Arguments structure. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:601</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5_html_ae86daa985279c77e57e682b64a68d330"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5.html#ae86daa985279c77e57e682b64a68d330">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::Arguments</a></div><div class="ttdeci">CUTLASS_HOST_DEVICE Arguments()</div><div class="ttdoc">Default ctor. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:597</div></div>
|
||
|
|
<div class="ttc" id="namespacecutlass_html_aa43b0a7d59635cb2d9ac96a077c988c3"><div class="ttname"><a href="namespacecutlass.html#aa43b0a7d59635cb2d9ac96a077c988c3">cutlass::TensorRef_aligned</a></div><div class="ttdeci">bool TensorRef_aligned(TensorRef< Element, Layout > const &ref, int alignment)</div><div class="ttdef"><b>Definition:</b> tensor_ref.h:382</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5_html_ad7d2b82b83d7503b9f920ce3bdcdffa5"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5.html#ad7d2b82b83d7503b9f920ce3bdcdffa5">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::ref_B</a></div><div class="ttdeci">TensorRef< ElementB const, LayoutB > ref_B</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:582</div></div>
|
||
|
|
<div class="ttc" id="namespacecutlass_html_ac5a88c5840a28a9e0206b9cc7812a18da8c632159fa131f09d04f94e3cbcd8782"><div class="ttname"><a href="namespacecutlass.html#ac5a88c5840a28a9e0206b9cc7812a18da8c632159fa131f09d04f94e3cbcd8782">cutlass::Status::kSuccess</a></div><div class="ttdoc">Operation was successful. </div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1Params_html_ad1867c0875c10e6327c7fae16acd35a3"><div class="ttname"><a href="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_1_1Params.html#ad1867c0875c10e6327c7fae16acd35a3">cutlass::gemm::kernel::GemmBatched::Params::ref_A</a></div><div class="ttdeci">Mma::IteratorA::TensorRef ref_A</div><div class="ttdef"><b>Definition:</b> kernel/gemm_batched.h:65</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_ad7e9e393be872e401a5a777ceda529d9"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#ad7e9e393be872e401a5a777ceda529d9">cutlass::gemm::device::GemmBatched::run</a></div><div class="ttdeci">Status run(cudaStream_t stream=nullptr)</div><div class="ttdoc">Runs the kernel using initialized state. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:407</div></div>
|
||
|
|
<div class="ttc" id="gemm_2threadblock_2threadblock__swizzle_8h_html"><div class="ttname"><a href="gemm_2threadblock_2threadblock__swizzle_8h.html">threadblock_swizzle.h</a></div><div class="ttdoc">Implements several possible threadblock-swizzling functions mapping blockIdx to GEMM problems...</div></div>
|
||
|
|
<div class="ttc" id="arch_8h_html"><div class="ttname"><a href="arch_8h.html">arch.h</a></div><div class="ttdoc">Defines tags for architecture-specific configurations. </div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched_html"><div class="ttname"><a href="structcutlass_1_1gemm_1_1kernel_1_1GemmBatched.html">cutlass::gemm::kernel::GemmBatched</a></div><div class="ttdef"><b>Definition:</b> kernel/gemm_batched.h:49</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a6e4cff55a6834d43cfbc97df40609eea"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a6e4cff55a6834d43cfbc97df40609eea">cutlass::gemm::device::GemmBatched< ElementB, typename layout::LayoutTranspose< LayoutB >::type, ElementA, typename layout::LayoutTranspose< LayoutA >::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, Stages, kAlignmentB, kAlignmentA >::ElementAccumulator</a></div><div class="ttdeci">ElementAccumulator ElementAccumulator</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:226</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments_html_aa867aa186538d34251d75ccc891453d7"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments.html#aa867aa186538d34251d75ccc891453d7">cutlass::gemm::device::GemmBatched::Arguments::stride_B</a></div><div class="ttdeci">int64_t stride_B</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:277</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a7b8d2dbfa562869deb58c88583951b58"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a7b8d2dbfa562869deb58c88583951b58">cutlass::gemm::device::GemmBatched< ElementB, typename layout::LayoutTranspose< LayoutB >::type, ElementA, typename layout::LayoutTranspose< LayoutA >::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, Stages, kAlignmentB, kAlignmentA >::WarpShape</a></div><div class="ttdeci">WarpShape WarpShape</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:230</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_add5d679e0acf0813a52c209d2448e81b"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#add5d679e0acf0813a52c209d2448e81b">cutlass::gemm::device::GemmBatched< ElementB, typename layout::LayoutTranspose< LayoutB >::type, ElementA, typename layout::LayoutTranspose< LayoutA >::type, ElementC, layout::RowMajor, ElementAccumulator, OperatorClass, ArchTag, ThreadblockShape, WarpShape, InstructionShape, EpilogueOutputOp, ThreadblockSwizzle, Stages, kAlignmentB, kAlignmentA >::ArchTag</a></div><div class="ttdeci">ArchTag ArchTag</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:228</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd_html_af623ca54d9554cdfafc09af7a22cdd62"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_0c9bb6f4463ab6085e6008b5d5ad6abfd.html#af623ca54d9554cdfafc09af7a22cdd62">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::LayoutA</a></div><div class="ttdeci">LayoutA_ LayoutA</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:524</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a45530b940ca86ce39cfc943da5713d80"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a45530b940ca86ce39cfc943da5713d80">cutlass::gemm::device::GemmBatched::operator()</a></div><div class="ttdeci">Status operator()(Arguments const &args, void *workspace=nullptr, cudaStream_t stream=nullptr)</div><div class="ttdoc">Runs the kernel using initialized state. </div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:448</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5_html_aa9e30e41627595590421d8b53941b2b2"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5.html#aa9e30e41627595590421d8b53941b2b2">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::ref_C</a></div><div class="ttdeci">TensorRef< ElementC const, LayoutC > ref_C</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:584</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments_html_aea0c4cd59daee8d3b497be411beb9b3a"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments.html#aea0c4cd59daee8d3b497be411beb9b3a">cutlass::gemm::device::GemmBatched::Arguments::epilogue</a></div><div class="ttdeci">EpilogueOutputOp::Params epilogue</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:282</div></div>
|
||
|
|
<div class="ttc" id="cutlass_8h_html"><div class="ttname"><a href="cutlass_8h.html">cutlass.h</a></div><div class="ttdoc">Basic include for CUTLASS. </div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5_html_adb66f3083f56c15578b139b7935452b5"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5.html#adb66f3083f56c15578b139b7935452b5">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::batch_count</a></div><div class="ttdeci">int batch_count</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:589</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_a35a2fb2e9ad63c316ac6fbb1cc8cf53a"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#a35a2fb2e9ad63c316ac6fbb1cc8cf53a">cutlass::gemm::device::GemmBatched::GemmKernel</a></div><div class="ttdeci">kernel::GemmBatched< typename DefaultGemmKernel::Mma, typename DefaultGemmKernel::Epilogue, ThreadblockSwizzle > GemmKernel</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:264</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html">cutlass::gemm::device::GemmBatched</a></div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:213</div></div>
|
||
|
|
<div class="ttc" id="namespacecutlass_html_ac5a88c5840a28a9e0206b9cc7812a18d"><div class="ttname"><a href="namespacecutlass.html#ac5a88c5840a28a9e0206b9cc7812a18d">cutlass::Status</a></div><div class="ttdeci">Status</div><div class="ttdoc">Status code returned by CUTLASS operations. </div><div class="ttdef"><b>Definition:</b> cutlass.h:39</div></div>
|
||
|
|
<div class="ttc" id="classcutlass_1_1gemm_1_1device_1_1GemmBatched_html_af35efd1f40deeb9d8e295f700fa84dbd"><div class="ttname"><a href="classcutlass_1_1gemm_1_1device_1_1GemmBatched.html#af35efd1f40deeb9d8e295f700fa84dbd">cutlass::gemm::device::GemmBatched::LayoutC</a></div><div class="ttdeci">LayoutC_ LayoutC</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:223</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments_html_a55f32be45559dbf84dcc2db26784f625"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_1_1Arguments.html#a55f32be45559dbf84dcc2db26784f625">cutlass::gemm::device::GemmBatched::Arguments::ref_A</a></div><div class="ttdeci">TensorRef< ElementA const, LayoutA > ref_A</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:274</div></div>
|
||
|
|
<div class="ttc" id="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5_html_ac8830c9ed0e0a8bd7aa2aa4382550a2f"><div class="ttname"><a href="structcutlass_1_1gemm_1_1device_1_1GemmBatched_3_01ElementA___00_01LayoutA___00_01ElementB___00_213d78696663f4231cd52c6a277c60e5.html#ac8830c9ed0e0a8bd7aa2aa4382550a2f">cutlass::gemm::device::GemmBatched< ElementA_, LayoutA_, ElementB_, LayoutB_, ElementC_, layout::ColumnMajor, ElementAccumulator_, OperatorClass_, ArchTag_, ThreadblockShape_, WarpShape_, InstructionShape_, EpilogueOutputOp_, ThreadblockSwizzle_, Stages, AlignmentA, AlignmentB, Operator_ >::Arguments::stride_A</a></div><div class="ttdeci">int64_t stride_A</div><div class="ttdef"><b>Definition:</b> device/gemm_batched.h:581</div></div>
|
||
|
|
</div><!-- fragment --></div><!-- contents -->
|
||
|
|
<!-- start footer part -->
|
||
|
|
<hr class="footer"/><address class="footer"><small>
|
||
|
|
Generated by  <a href="http://www.doxygen.org/index.html">
|
||
|
|
<img class="footer" src="doxygen.png" alt="doxygen"/>
|
||
|
|
</a> 1.8.11
|
||
|
|
</small></address>
|
||
|
|
</body>
|
||
|
|
</html>
|