dnrops
/
cutlass
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
cutlass/docs/namespacecutlass_1_1referen...

689 lines
32 KiB
HTML
Raw Permalink Blame History

<!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: cutlass::reference::device::kernel Namespace Reference</title>
<link href="tabs.css" rel="stylesheet" type="text/css"/>
<script type="text/javascript" src="jquery.js"></script>
<script type="text/javascript" src="dynsections.js"></script>
<link href="search/search.css" rel="stylesheet" type="text/css"/>
<script type="text/javascript" src="search/searchdata.js"></script>
<script type="text/javascript" src="search/search.js"></script>
<script type="text/javascript">
$(document).ready(function() { init_search(); });
</script>
<script type="text/x-mathjax-config">
MathJax.Hub.Config({
extensions: ["tex2jax.js"],
jax: ["input/TeX","output/HTML-CSS"],
});
</script><script type="text/javascript" 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&#160;Page</span></a></li>
<li><a href="modules.html"><span>Modules</span></a></li>
<li class="current"><a href="namespaces.html"><span>Namespaces</span></a></li>
<li><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="namespaces.html"><span>Namespace&#160;List</span></a></li>
<li><a href="namespacemembers.html"><span>Namespace&#160;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="namespacecutlass.html">cutlass</a></li><li class="navelem"><a class="el" href="namespacecutlass_1_1reference.html">reference</a></li><li class="navelem"><a class="el" href="namespacecutlass_1_1reference_1_1device.html">device</a></li><li class="navelem"><a class="el" href="namespacecutlass_1_1reference_1_1device_1_1kernel.html">kernel</a></li> </ul>
</div>
</div><!-- top -->
<div class="header">
<div class="summary">
<a href="#namespaces">Namespaces</a> &#124;
<a href="#func-members">Functions</a> </div>
<div class="headertitle">
<div class="title">cutlass::reference::device::kernel Namespace Reference</div> </div>
</div><!--header-->
<div class="contents">
<table class="memberdecls">
<tr class="heading"><td colspan="2"><h2 class="groupheader"><a name="namespaces"></a>
Namespaces</h2></td></tr>
<tr class="memitem:namespacecutlass_1_1reference_1_1device_1_1kernel_1_1detail"><td class="memItemLeft" align="right" valign="top"> &#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="namespacecutlass_1_1reference_1_1device_1_1kernel_1_1detail.html">detail</a></td></tr>
<tr class="memdesc:namespacecutlass_1_1reference_1_1device_1_1kernel_1_1detail"><td class="mdescLeft">&#160;</td><td class="mdescRight">Defines several helpers. <br /></td></tr>
<tr class="separator:"><td class="memSeparator" colspan="2">&#160;</td></tr>
</table><table class="memberdecls">
<tr class="heading"><td colspan="2"><h2 class="groupheader"><a name="func-members"></a>
Functions</h2></td></tr>
<tr class="memitem:a0f44a48b38f56a69beade68adb32df6f"><td class="memTemplParams" colspan="2">template&lt;typename TensorRefA , typename TensorRefB , typename TensorRefC , typename ScalarType , typename AccumulatorType , typename OutputTile , typename InnerProductOp , typename ConvertOp &gt; </td></tr>
<tr class="memitem:a0f44a48b38f56a69beade68adb32df6f"><td class="memTemplItemLeft" align="right" valign="top">__global__ void&#160;</td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacecutlass_1_1reference_1_1device_1_1kernel.html#a0f44a48b38f56a69beade68adb32df6f">Gemm</a> (<a class="el" href="structcutlass_1_1gemm_1_1GemmCoord.html">gemm::GemmCoord</a> problem_size, ScalarType alpha, TensorRefA tensor_a, TensorRefB tensor_b, ScalarType beta, TensorRefC tensor_c, TensorRefC tensor_d, AccumulatorType initial_accum)</td></tr>
<tr class="separator:a0f44a48b38f56a69beade68adb32df6f"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a013cf9aa1c8f98ec2037f242284def7b"><td class="memTemplParams" colspan="2">template&lt;typename TensorRefCollectionA , typename TensorRefCollectionB , typename TensorRefCollectionC , typename ScalarType , typename AccumulatorType , typename OutputTile , typename InnerProductOp , typename ConvertOp &gt; </td></tr>
<tr class="memitem:a013cf9aa1c8f98ec2037f242284def7b"><td class="memTemplItemLeft" align="right" valign="top">__global__ void&#160;</td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacecutlass_1_1reference_1_1device_1_1kernel.html#a013cf9aa1c8f98ec2037f242284def7b">BatchedGemm</a> (<a class="el" href="structcutlass_1_1gemm_1_1GemmCoord.html">gemm::GemmCoord</a> problem_size, ScalarType alpha, TensorRefCollectionA tensor_collection_a, TensorRefCollectionB tensor_collection_b, ScalarType beta, TensorRefCollectionC tensor_collection_c, AccumulatorType initial_accum)</td></tr>
<tr class="separator:a013cf9aa1c8f98ec2037f242284def7b"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a44bffb16758ab0071aac16d203f2d051"><td class="memTemplParams" colspan="2">template&lt;typename T &gt; </td></tr>
<tr class="memitem:a44bffb16758ab0071aac16d203f2d051"><td class="memTemplItemLeft" align="right" valign="top">__global__ void&#160;</td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacecutlass_1_1reference_1_1device_1_1kernel.html#a44bffb16758ab0071aac16d203f2d051">TensorInitializeUniform</a> (<a class="el" href="structcutlass_1_1Distribution.html">Distribution</a> dist, int64_t seed, int dim_contiguous, int dim_strided, T *tensor, int ldm)</td></tr>
<tr class="memdesc:a44bffb16758ab0071aac16d203f2d051"><td class="mdescLeft">&#160;</td><td class="mdescRight">Kernel to initialize tensor to uniform random distribution. <a href="#a44bffb16758ab0071aac16d203f2d051">More...</a><br /></td></tr>
<tr class="separator:a44bffb16758ab0071aac16d203f2d051"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a1e8054d6781358c0faeddfe77f28f23b"><td class="memTemplParams" colspan="2">template&lt;typename T &gt; </td></tr>
<tr class="memitem:a1e8054d6781358c0faeddfe77f28f23b"><td class="memTemplItemLeft" align="right" valign="top">__global__ void&#160;</td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacecutlass_1_1reference_1_1device_1_1kernel.html#a1e8054d6781358c0faeddfe77f28f23b">TensorInitializeGaussian</a> (<a class="el" href="structcutlass_1_1Distribution.html">Distribution</a> dist, int64_t seed, int dim_contiguous, int dim_strided, T *tensor, int ldm)</td></tr>
<tr class="memdesc:a1e8054d6781358c0faeddfe77f28f23b"><td class="mdescLeft">&#160;</td><td class="mdescRight">Kernel to initialize tensor to uniform distribution. <a href="#a1e8054d6781358c0faeddfe77f28f23b">More...</a><br /></td></tr>
<tr class="separator:a1e8054d6781358c0faeddfe77f28f23b"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a1c73ae8819459dba630520208038a2ae"><td class="memTemplParams" colspan="2">template&lt;typename T &gt; </td></tr>
<tr class="memitem:a1c73ae8819459dba630520208038a2ae"><td class="memTemplItemLeft" align="right" valign="top">__global__ void&#160;</td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacecutlass_1_1reference_1_1device_1_1kernel.html#a1c73ae8819459dba630520208038a2ae">TensorInitializeLinear</a> (<a class="el" href="structcutlass_1_1Distribution.html">Distribution</a> dist, int64_t seed, int dim_contiguous, int dim_strided, T *tensor, int ldm)</td></tr>
<tr class="memdesc:a1c73ae8819459dba630520208038a2ae"><td class="mdescLeft">&#160;</td><td class="mdescRight">Kernel to initialize tensor to an identity matrix. <a href="#a1c73ae8819459dba630520208038a2ae">More...</a><br /></td></tr>
<tr class="separator:a1c73ae8819459dba630520208038a2ae"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a7ac1aaf53d1a16e6d9b050471fa08e2c"><td class="memTemplParams" colspan="2">template&lt;typename T &gt; </td></tr>
<tr class="memitem:a7ac1aaf53d1a16e6d9b050471fa08e2c"><td class="memTemplItemLeft" align="right" valign="top">__global__ void&#160;</td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacecutlass_1_1reference_1_1device_1_1kernel.html#a7ac1aaf53d1a16e6d9b050471fa08e2c">TensorInitializeIdentity</a> (<a class="el" href="structcutlass_1_1Distribution.html">Distribution</a> dist, int64_t seed, int dim_contiguous, int dim_strided, T *tensor, int ldm)</td></tr>
<tr class="memdesc:a7ac1aaf53d1a16e6d9b050471fa08e2c"><td class="mdescLeft">&#160;</td><td class="mdescRight">Kernel to initialize tensor to an identity matrix. <a href="#a7ac1aaf53d1a16e6d9b050471fa08e2c">More...</a><br /></td></tr>
<tr class="separator:a7ac1aaf53d1a16e6d9b050471fa08e2c"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:ae22a592321cef9a9f586d3f094933e3f"><td class="memTemplParams" colspan="2">template&lt;typename Func , int Rank, typename Params &gt; </td></tr>
<tr class="memitem:ae22a592321cef9a9f586d3f094933e3f"><td class="memTemplItemLeft" align="right" valign="top">__global__ void&#160;</td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacecutlass_1_1reference_1_1device_1_1kernel.html#ae22a592321cef9a9f586d3f094933e3f">TensorForEach</a> (<a class="el" href="structcutlass_1_1Coord.html">Coord</a>&lt; Rank &gt; size, Params params=Params())</td></tr>
<tr class="memdesc:ae22a592321cef9a9f586d3f094933e3f"><td class="mdescLeft">&#160;</td><td class="mdescRight">Kernel calls a functor for each element in a tensor's index space. <a href="#ae22a592321cef9a9f586d3f094933e3f">More...</a><br /></td></tr>
<tr class="separator:ae22a592321cef9a9f586d3f094933e3f"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:ab3b42b1c0e6f28c3b62b65a373db5fd7"><td class="memTemplParams" colspan="2">template&lt;typename Func , int Rank, typename Params &gt; </td></tr>
<tr class="memitem:ab3b42b1c0e6f28c3b62b65a373db5fd7"><td class="memTemplItemLeft" align="right" valign="top">__global__ void&#160;</td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacecutlass_1_1reference_1_1device_1_1kernel.html#ab3b42b1c0e6f28c3b62b65a373db5fd7">TensorDiagonalForEach</a> (<a class="el" href="structcutlass_1_1Coord.html">Coord</a>&lt; Rank &gt; size, Params params, int start, int end)</td></tr>
<tr class="memdesc:ab3b42b1c0e6f28c3b62b65a373db5fd7"><td class="mdescLeft">&#160;</td><td class="mdescRight">Kernel calls a functor for each element along a tensor's diagonal. <a href="#ab3b42b1c0e6f28c3b62b65a373db5fd7">More...</a><br /></td></tr>
<tr class="separator:ab3b42b1c0e6f28c3b62b65a373db5fd7"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a0100d78891f9e00e75453ef8dc24daa6"><td class="memTemplParams" colspan="2">template&lt;typename Element , typename Func &gt; </td></tr>
<tr class="memitem:a0100d78891f9e00e75453ef8dc24daa6"><td class="memTemplItemLeft" align="right" valign="top">__global__ void&#160;</td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacecutlass_1_1reference_1_1device_1_1kernel.html#a0100d78891f9e00e75453ef8dc24daa6">BlockForEach</a> (Element *ptr, size_t capacity, typename Func::Params params)</td></tr>
<tr class="separator:a0100d78891f9e00e75453ef8dc24daa6"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a4595ede72eddace3c973c7f0f74b001d"><td class="memTemplParams" colspan="2">template&lt;typename Element &gt; </td></tr>
<tr class="memitem:a4595ede72eddace3c973c7f0f74b001d"><td class="memTemplItemLeft" align="right" valign="top">__global__ void&#160;</td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacecutlass_1_1reference_1_1device_1_1kernel.html#a4595ede72eddace3c973c7f0f74b001d">BlockCompareEqual</a> (int *equal, Element const *ptr_A, Element const *ptr_B, size_t capacity)</td></tr>
<tr class="separator:a4595ede72eddace3c973c7f0f74b001d"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a6da13fb683d56d6973af0a97a4023677"><td class="memTemplParams" colspan="2">template&lt;typename Element &gt; </td></tr>
<tr class="memitem:a6da13fb683d56d6973af0a97a4023677"><td class="memTemplItemLeft" align="right" valign="top">__global__ void&#160;</td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacecutlass_1_1reference_1_1device_1_1kernel.html#a6da13fb683d56d6973af0a97a4023677">BlockCompareRelativelyEqual</a> (int *equal, Element const *ptr_A, Element const *ptr_B, size_t capacity, Element epsilon, Element nonzero_floor)</td></tr>
<tr class="separator:a6da13fb683d56d6973af0a97a4023677"><td class="memSeparator" colspan="2">&#160;</td></tr>
</table>
<h2 class="groupheader">Function Documentation</h2>
<a class="anchor" id="a013cf9aa1c8f98ec2037f242284def7b"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
template&lt;typename TensorRefCollectionA , typename TensorRefCollectionB , typename TensorRefCollectionC , typename ScalarType , typename AccumulatorType , typename OutputTile , typename InnerProductOp , typename ConvertOp &gt; </div>
<table class="memname">
<tr>
<td class="memname">__global__ void cutlass::reference::device::kernel::BatchedGemm </td>
<td>(</td>
<td class="paramtype"><a class="el" href="structcutlass_1_1gemm_1_1GemmCoord.html">gemm::GemmCoord</a>&#160;</td>
<td class="paramname"><em>problem_size</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">ScalarType&#160;</td>
<td class="paramname"><em>alpha</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">TensorRefCollectionA&#160;</td>
<td class="paramname"><em>tensor_collection_a</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">TensorRefCollectionB&#160;</td>
<td class="paramname"><em>tensor_collection_b</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">ScalarType&#160;</td>
<td class="paramname"><em>beta</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">TensorRefCollectionC&#160;</td>
<td class="paramname"><em>tensor_collection_c</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">AccumulatorType&#160;</td>
<td class="paramname"><em>initial_accum</em>&#160;</td>
</tr>
<tr>
<td></td>
<td>)</td>
<td></td><td></td>
</tr>
</table>
</div><div class="memdoc">
<p>Computes a general matrix product among matrices (tensors of rank=2) pointed to by <a class="el" href="classcutlass_1_1TensorRef.html">TensorRef</a> objects. </p>
</div>
</div>
<a class="anchor" id="a4595ede72eddace3c973c7f0f74b001d"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
template&lt;typename Element &gt; </div>
<table class="memname">
<tr>
<td class="memname">__global__ void cutlass::reference::device::kernel::BlockCompareEqual </td>
<td>(</td>
<td class="paramtype">int *&#160;</td>
<td class="paramname"><em>equal</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">Element const *&#160;</td>
<td class="paramname"><em>ptr_A</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">Element const *&#160;</td>
<td class="paramname"><em>ptr_B</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">size_t&#160;</td>
<td class="paramname"><em>capacity</em>&#160;</td>
</tr>
<tr>
<td></td>
<td>)</td>
<td></td><td></td>
</tr>
</table>
</div><div class="memdoc">
</div>
</div>
<a class="anchor" id="a6da13fb683d56d6973af0a97a4023677"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
template&lt;typename Element &gt; </div>
<table class="memname">
<tr>
<td class="memname">__global__ void cutlass::reference::device::kernel::BlockCompareRelativelyEqual </td>
<td>(</td>
<td class="paramtype">int *&#160;</td>
<td class="paramname"><em>equal</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">Element const *&#160;</td>
<td class="paramname"><em>ptr_A</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">Element const *&#160;</td>
<td class="paramname"><em>ptr_B</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">size_t&#160;</td>
<td class="paramname"><em>capacity</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">Element&#160;</td>
<td class="paramname"><em>epsilon</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">Element&#160;</td>
<td class="paramname"><em>nonzero_floor</em>&#160;</td>
</tr>
<tr>
<td></td>
<td>)</td>
<td></td><td></td>
</tr>
</table>
</div><div class="memdoc">
</div>
</div>
<a class="anchor" id="a0100d78891f9e00e75453ef8dc24daa6"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
template&lt;typename Element , typename Func &gt; </div>
<table class="memname">
<tr>
<td class="memname">__global__ void cutlass::reference::device::kernel::BlockForEach </td>
<td>(</td>
<td class="paramtype">Element *&#160;</td>
<td class="paramname"><em>ptr</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">size_t&#160;</td>
<td class="paramname"><em>capacity</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">typename Func::Params&#160;</td>
<td class="paramname"><em>params</em>&#160;</td>
</tr>
<tr>
<td></td>
<td>)</td>
<td></td><td></td>
</tr>
</table>
</div><div class="memdoc">
</div>
</div>
<a class="anchor" id="a0f44a48b38f56a69beade68adb32df6f"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
template&lt;typename TensorRefA , typename TensorRefB , typename TensorRefC , typename ScalarType , typename AccumulatorType , typename OutputTile , typename InnerProductOp , typename ConvertOp &gt; </div>
<table class="memname">
<tr>
<td class="memname">__global__ void cutlass::reference::device::kernel::Gemm </td>
<td>(</td>
<td class="paramtype"><a class="el" href="structcutlass_1_1gemm_1_1GemmCoord.html">gemm::GemmCoord</a>&#160;</td>
<td class="paramname"><em>problem_size</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">ScalarType&#160;</td>
<td class="paramname"><em>alpha</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">TensorRefA&#160;</td>
<td class="paramname"><em>tensor_a</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">TensorRefB&#160;</td>
<td class="paramname"><em>tensor_b</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">ScalarType&#160;</td>
<td class="paramname"><em>beta</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">TensorRefC&#160;</td>
<td class="paramname"><em>tensor_c</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">TensorRefC&#160;</td>
<td class="paramname"><em>tensor_d</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">AccumulatorType&#160;</td>
<td class="paramname"><em>initial_accum</em>&#160;</td>
</tr>
<tr>
<td></td>
<td>)</td>
<td></td><td></td>
</tr>
</table>
</div><div class="memdoc">
<p>Computes a general matrix product among matrices (tensors of rank=2) pointed to by <a class="el" href="classcutlass_1_1TensorRef.html">TensorRef</a> objects. </p>
</div>
</div>
<a class="anchor" id="ab3b42b1c0e6f28c3b62b65a373db5fd7"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
template&lt;typename Func , int Rank, typename Params &gt; </div>
<table class="memname">
<tr>
<td class="memname">__global__ void cutlass::reference::device::kernel::TensorDiagonalForEach </td>
<td>(</td>
<td class="paramtype"><a class="el" href="structcutlass_1_1Coord.html">Coord</a>&lt; Rank &gt;&#160;</td>
<td class="paramname"><em>size</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">Params&#160;</td>
<td class="paramname"><em>params</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int&#160;</td>
<td class="paramname"><em>start</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int&#160;</td>
<td class="paramname"><em>end</em>&#160;</td>
</tr>
<tr>
<td></td>
<td>)</td>
<td></td><td></td>
</tr>
</table>
</div><div class="memdoc">
</div>
</div>
<a class="anchor" id="ae22a592321cef9a9f586d3f094933e3f"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
template&lt;typename Func , int Rank, typename Params &gt; </div>
<table class="memname">
<tr>
<td class="memname">__global__ void cutlass::reference::device::kernel::TensorForEach </td>
<td>(</td>
<td class="paramtype"><a class="el" href="structcutlass_1_1Coord.html">Coord</a>&lt; Rank &gt;&#160;</td>
<td class="paramname"><em>size</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">Params&#160;</td>
<td class="paramname"><em>params</em> = <code>Params()</code>&#160;</td>
</tr>
<tr>
<td></td>
<td>)</td>
<td></td><td></td>
</tr>
</table>
</div><div class="memdoc">
</div>
</div>
<a class="anchor" id="a1e8054d6781358c0faeddfe77f28f23b"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
template&lt;typename T &gt; </div>
<table class="memname">
<tr>
<td class="memname">__global__ void cutlass::reference::device::kernel::TensorInitializeGaussian </td>
<td>(</td>
<td class="paramtype"><a class="el" href="structcutlass_1_1Distribution.html">Distribution</a>&#160;</td>
<td class="paramname"><em>dist</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int64_t&#160;</td>
<td class="paramname"><em>seed</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int&#160;</td>
<td class="paramname"><em>dim_contiguous</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int&#160;</td>
<td class="paramname"><em>dim_strided</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">T *&#160;</td>
<td class="paramname"><em>tensor</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int&#160;</td>
<td class="paramname"><em>ldm</em>&#160;</td>
</tr>
<tr>
<td></td>
<td>)</td>
<td></td><td></td>
</tr>
</table>
</div><div class="memdoc">
</div>
</div>
<a class="anchor" id="a7ac1aaf53d1a16e6d9b050471fa08e2c"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
template&lt;typename T &gt; </div>
<table class="memname">
<tr>
<td class="memname">__global__ void cutlass::reference::device::kernel::TensorInitializeIdentity </td>
<td>(</td>
<td class="paramtype"><a class="el" href="structcutlass_1_1Distribution.html">Distribution</a>&#160;</td>
<td class="paramname"><em>dist</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int64_t&#160;</td>
<td class="paramname"><em>seed</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int&#160;</td>
<td class="paramname"><em>dim_contiguous</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int&#160;</td>
<td class="paramname"><em>dim_strided</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">T *&#160;</td>
<td class="paramname"><em>tensor</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int&#160;</td>
<td class="paramname"><em>ldm</em>&#160;</td>
</tr>
<tr>
<td></td>
<td>)</td>
<td></td><td></td>
</tr>
</table>
</div><div class="memdoc">
</div>
</div>
<a class="anchor" id="a1c73ae8819459dba630520208038a2ae"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
template&lt;typename T &gt; </div>
<table class="memname">
<tr>
<td class="memname">__global__ void cutlass::reference::device::kernel::TensorInitializeLinear </td>
<td>(</td>
<td class="paramtype"><a class="el" href="structcutlass_1_1Distribution.html">Distribution</a>&#160;</td>
<td class="paramname"><em>dist</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int64_t&#160;</td>
<td class="paramname"><em>seed</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int&#160;</td>
<td class="paramname"><em>dim_contiguous</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int&#160;</td>
<td class="paramname"><em>dim_strided</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">T *&#160;</td>
<td class="paramname"><em>tensor</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int&#160;</td>
<td class="paramname"><em>ldm</em>&#160;</td>
</tr>
<tr>
<td></td>
<td>)</td>
<td></td><td></td>
</tr>
</table>
</div><div class="memdoc">
</div>
</div>
<a class="anchor" id="a44bffb16758ab0071aac16d203f2d051"></a>
<div class="memitem">
<div class="memproto">
<div class="memtemplate">
template&lt;typename T &gt; </div>
<table class="memname">
<tr>
<td class="memname">__global__ void cutlass::reference::device::kernel::TensorInitializeUniform </td>
<td>(</td>
<td class="paramtype"><a class="el" href="structcutlass_1_1Distribution.html">Distribution</a>&#160;</td>
<td class="paramname"><em>dist</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int64_t&#160;</td>
<td class="paramname"><em>seed</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int&#160;</td>
<td class="paramname"><em>dim_contiguous</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int&#160;</td>
<td class="paramname"><em>dim_strided</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">T *&#160;</td>
<td class="paramname"><em>tensor</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int&#160;</td>
<td class="paramname"><em>ldm</em>&#160;</td>
</tr>
<tr>
<td></td>
<td>)</td>
<td></td><td></td>
</tr>
</table>
</div><div class="memdoc">
</div>
</div>
</div><!-- contents -->
<!-- start footer part -->
<hr class="footer"/><address class="footer"><small>
Generated by &#160;<a href="http://www.doxygen.org/index.html">
<img class="footer" src="doxygen.png" alt="doxygen"/>
</a> 1.8.11
</small></address>
</body>
</html>
Reference in New Issue View Git Blame Copy Permalink