Google Git
Sign in
android / platform / external / ComputeLibrary / 975dfe175 / . / documentation / classarm__compute_1_1_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core.xhtml
blob: fccffc1963a29a83d4d19c1f9355037b878299b0 [file] [log] [blame]
<!-- HTML header for doxygen 1.8.15-->
<!-- Remember to use version doxygen 1.8.15 +-->
<!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 1.8.15"/>
<meta name="robots" content="NOINDEX, NOFOLLOW" /> <!-- Prevent indexing by search engines -->
<title>Compute Library: NEGEMMLowpAssemblyMatrixMultiplyCore Class 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="navtree.css" rel="stylesheet" type="text/css"/>
<script type="text/javascript" src="resize.js"></script>
<script type="text/javascript" src="navtreedata.js"></script>
<script type="text/javascript" src="navtree.js"></script>
<script type="text/javascript">
/* @license magnet:?xt=urn:btih:cf05388f2679ee054f2beb29a391d25f4e673ac3&amp;dn=gpl-2.0.txt GPL-v2 */
$(document).ready(initResizable);
/* @license-end */</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/x-mathjax-config">
MathJax.Hub.Config({
extensions: ["tex2jax.js"],
jax: ["input/TeX","output/HTML-CSS"],
});
</script><script type="text/javascript" async="async" src="http://cdn.mathjax.org/mathjax/latest/MathJax.js"></script>
<link href="doxygen.css" rel="stylesheet" type="text/css" />
<link href="stylesheet.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;">
<img alt="Compute Library" src="https://raw.githubusercontent.com/ARM-software/ComputeLibrary/gh-pages/ACL_logo.png" style="max-width: 100%;margin-top: 15px;margin-left: 10px"/>
<td style="padding-left: 0.5em;">
<div id="projectname">
&#160;<span id="projectnumber">19.08</span>
</div>
</td>
</tr>
</tbody>
</table>
</div>
<!-- end header part -->
<!-- Generated by Doxygen 1.8.15 -->
<script type="text/javascript">
/* @license magnet:?xt=urn:btih:cf05388f2679ee054f2beb29a391d25f4e673ac3&amp;dn=gpl-2.0.txt GPL-v2 */
var searchBox = new SearchBox("searchBox", "search",false,'Search');
/* @license-end */
</script>
<script type="text/javascript" src="menudata.js"></script>
<script type="text/javascript" src="menu.js"></script>
<script type="text/javascript">
/* @license magnet:?xt=urn:btih:cf05388f2679ee054f2beb29a391d25f4e673ac3&amp;dn=gpl-2.0.txt GPL-v2 */
$(function() {
initMenu('',true,false,'search.php','Search');
$(document).ready(function() { init_search(); });
});
/* @license-end */</script>
<div id="main-nav"></div>
</div><!-- top -->
<div id="side-nav" class="ui-resizable side-nav-resizable">
<div id="nav-tree">
<div id="nav-tree-contents">
<div id="nav-sync" class="sync"></div>
</div>
</div>
<div id="splitbar" style="-moz-user-select:none;"
class="ui-resizable-handle">
</div>
</div>
<script type="text/javascript">
/* @license magnet:?xt=urn:btih:cf05388f2679ee054f2beb29a391d25f4e673ac3&amp;dn=gpl-2.0.txt GPL-v2 */
$(document).ready(function(){initNavTree('classarm__compute_1_1_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core.xhtml','');});
/* @license-end */
</script>
<div id="doc-content">
<!-- window showing the filter options -->
<div id="MSearchSelectWindow"
onmouseover="return searchBox.OnSearchSelectShow()"
onmouseout="return searchBox.OnSearchSelectHide()"
onkeydown="return searchBox.OnSearchSelectKey(event)">
</div>
<!-- iframe showing the search results (closed by default) -->
<div id="MSearchResultsWindow">
<iframe src="javascript:void(0)" frameborder="0"
name="MSearchResults" id="MSearchResults">
</iframe>
</div>
<div class="header">
<div class="summary">
<a href="#pub-methods">Public Member Functions</a> </div>
<div class="headertitle">
<div class="title">NEGEMMLowpAssemblyMatrixMultiplyCore Class Reference</div> </div>
</div><!--header-->
<div class="contents">
<p>Basic function to execute matrix multiply assembly kernels.
<a href="classarm__compute_1_1_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core.xhtml#details">More...</a></p>
<p><code>#include &lt;<a class="el" href="_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core_8h_source.xhtml">NEGEMMLowpAssemblyMatrixMultiplyCore.h</a>&gt;</code></p>
<div class="dynheader">
Collaboration diagram for NEGEMMLowpAssemblyMatrixMultiplyCore:</div>
<div class="dyncontent">
<div class="center"><iframe scrolling="no" frameborder="0" src="classarm__compute_1_1_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core__coll__graph.svg" width="224" height="127"><p><b>This browser is not able to show SVG: try Firefox, Chrome, Safari, or Opera instead.</b></p></iframe>
</div>
<center><span class="legend">[<a target="top" href="graph_legend.xhtml">legend</a>]</span></center></div>
<table class="memberdecls">
<tr class="heading"><td colspan="2"><h2 class="groupheader"><a name="pub-methods"></a>
Public Member Functions</h2></td></tr>
<tr class="memitem:af5f476de0574e0225a55db44447c48f4"><td class="memItemLeft" align="right" valign="top">&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core.xhtml#af5f476de0574e0225a55db44447c48f4">NEGEMMLowpAssemblyMatrixMultiplyCore</a> (std::shared_ptr&lt; <a class="el" href="classarm__compute_1_1_i_memory_manager.xhtml">IMemoryManager</a> &gt; memory_manager=nullptr)</td></tr>
<tr class="memdesc:af5f476de0574e0225a55db44447c48f4"><td class="mdescLeft">&#160;</td><td class="mdescRight">Constructor. <a href="#af5f476de0574e0225a55db44447c48f4">More...</a><br /></td></tr>
<tr class="separator:af5f476de0574e0225a55db44447c48f4"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a440dfa49fc99e00b6909e868bf8557fb"><td class="memItemLeft" align="right" valign="top">void&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core.xhtml#a440dfa49fc99e00b6909e868bf8557fb">configure</a> (const <a class="el" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> *a, const <a class="el" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> *b, const <a class="el" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> *c, <a class="el" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> *output)</td></tr>
<tr class="memdesc:a440dfa49fc99e00b6909e868bf8557fb"><td class="mdescLeft">&#160;</td><td class="mdescRight">Initialise the kernel's inputs, output. <a href="#a440dfa49fc99e00b6909e868bf8557fb">More...</a><br /></td></tr>
<tr class="separator:a440dfa49fc99e00b6909e868bf8557fb"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:ad1717410afd0be936c6213a63c8005fb"><td class="memItemLeft" align="right" valign="top">void&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core.xhtml#ad1717410afd0be936c6213a63c8005fb">run</a> () override</td></tr>
<tr class="memdesc:ad1717410afd0be936c6213a63c8005fb"><td class="mdescLeft">&#160;</td><td class="mdescRight">Run the kernels contained in the function. <a href="#ad1717410afd0be936c6213a63c8005fb">More...</a><br /></td></tr>
<tr class="separator:ad1717410afd0be936c6213a63c8005fb"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="inherit_header pub_methods_classarm__compute_1_1_i_function"><td colspan="2" onclick="javascript:toggleInherit('pub_methods_classarm__compute_1_1_i_function')"><img src="closed.png" alt="-"/>&#160;Public Member Functions inherited from <a class="el" href="classarm__compute_1_1_i_function.xhtml">IFunction</a></td></tr>
<tr class="memitem:ab921ecc3f3f6ae2b4bd61f3e1998d8c4 inherit pub_methods_classarm__compute_1_1_i_function"><td class="memItemLeft" align="right" valign="top">virtual&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="classarm__compute_1_1_i_function.xhtml#ab921ecc3f3f6ae2b4bd61f3e1998d8c4">~IFunction</a> ()=default</td></tr>
<tr class="memdesc:ab921ecc3f3f6ae2b4bd61f3e1998d8c4 inherit pub_methods_classarm__compute_1_1_i_function"><td class="mdescLeft">&#160;</td><td class="mdescRight">Destructor. <a href="classarm__compute_1_1_i_function.xhtml#ab921ecc3f3f6ae2b4bd61f3e1998d8c4">More...</a><br /></td></tr>
<tr class="separator:ab921ecc3f3f6ae2b4bd61f3e1998d8c4 inherit pub_methods_classarm__compute_1_1_i_function"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a820f7291c24155a2980512fae45aac26 inherit pub_methods_classarm__compute_1_1_i_function"><td class="memItemLeft" align="right" valign="top">virtual void&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="classarm__compute_1_1_i_function.xhtml#a820f7291c24155a2980512fae45aac26">prepare</a> ()</td></tr>
<tr class="memdesc:a820f7291c24155a2980512fae45aac26 inherit pub_methods_classarm__compute_1_1_i_function"><td class="mdescLeft">&#160;</td><td class="mdescRight">Prepare the function for executing. <a href="classarm__compute_1_1_i_function.xhtml#a820f7291c24155a2980512fae45aac26">More...</a><br /></td></tr>
<tr class="separator:a820f7291c24155a2980512fae45aac26 inherit pub_methods_classarm__compute_1_1_i_function"><td class="memSeparator" colspan="2">&#160;</td></tr>
</table>
<a name="details" id="details"></a><h2 class="groupheader">Detailed Description</h2>
<div class="textblock"><p>Basic function to execute matrix multiply assembly kernels. </p>
<p class="definition">Definition at line <a class="el" href="_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core_8h_source.xhtml#l00042">42</a> of file <a class="el" href="_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core_8h_source.xhtml">NEGEMMLowpAssemblyMatrixMultiplyCore.h</a>.</p>
</div><h2 class="groupheader">Constructor &amp; Destructor Documentation</h2>
<a id="af5f476de0574e0225a55db44447c48f4"></a>
<h2 class="memtitle"><span class="permalink"><a href="#af5f476de0574e0225a55db44447c48f4">&#9670;&nbsp;</a></span>NEGEMMLowpAssemblyMatrixMultiplyCore()</h2>
<div class="memitem">
<div class="memproto">
<table class="memname">
<tr>
<td class="memname"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core.xhtml">NEGEMMLowpAssemblyMatrixMultiplyCore</a> </td>
<td>(</td>
<td class="paramtype">std::shared_ptr&lt; <a class="el" href="classarm__compute_1_1_i_memory_manager.xhtml">IMemoryManager</a> &gt;&#160;</td>
<td class="paramname"><em>memory_manager</em> = <code>nullptr</code></td><td>)</td>
<td></td>
</tr>
</table>
</div><div class="memdoc">
<p>Constructor. </p>
<p class="definition">Definition at line <a class="el" href="_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core_8cpp_source.xhtml#l00041">41</a> of file <a class="el" href="_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core_8cpp_source.xhtml">NEGEMMLowpAssemblyMatrixMultiplyCore.cpp</a>.</p>
<div class="fragment"><div class="line"><a name="l00042"></a><span class="lineno"> 42</span>&#160; : _memory_group(memory_manager), _asm_glue(memory_manager), _mm_kernel(<span class="keyword">nullptr</span>), _mtx_a_reshape_kernel(<span class="keyword">nullptr</span>), _mtx_b_reshape_kernel(<span class="keyword">nullptr</span>), _tmp_a(), _tmp_b()</div><div class="line"><a name="l00043"></a><span class="lineno"> 43</span>&#160;{</div><div class="line"><a name="l00044"></a><span class="lineno"> 44</span>&#160;}</div></div><!-- fragment -->
</div>
</div>
<h2 class="groupheader">Member Function Documentation</h2>
<a id="a440dfa49fc99e00b6909e868bf8557fb"></a>
<h2 class="memtitle"><span class="permalink"><a href="#a440dfa49fc99e00b6909e868bf8557fb">&#9670;&nbsp;</a></span>configure()</h2>
<div class="memitem">
<div class="memproto">
<table class="memname">
<tr>
<td class="memname">void configure </td>
<td>(</td>
<td class="paramtype">const <a class="el" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> *&#160;</td>
<td class="paramname"><em>a</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">const <a class="el" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> *&#160;</td>
<td class="paramname"><em>b</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">const <a class="el" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> *&#160;</td>
<td class="paramname"><em>c</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype"><a class="el" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> *&#160;</td>
<td class="paramname"><em>output</em>&#160;</td>
</tr>
<tr>
<td></td>
<td>)</td>
<td></td><td></td>
</tr>
</table>
</div><div class="memdoc">
<p>Initialise the kernel's inputs, output. </p>
<dl class="params"><dt>Parameters</dt><dd>
<table class="params">
<tr><td class="paramdir">[in]</td><td class="paramname">a</td><td>First input tensor (Matrix A). Data type supported: U8, S8. </td></tr>
<tr><td class="paramdir">[in]</td><td class="paramname">b</td><td>Second input tensor (Matrix B). Data type supported: same as <code>a</code> </td></tr>
<tr><td class="paramdir">[in]</td><td class="paramname">c</td><td>Third input tensor (Matrix C). Data type supported: same as <code>a</code> </td></tr>
<tr><td class="paramdir">[out]</td><td class="paramname">output</td><td>Output tensor. Data type supported: Data type supported: U32, S32 </td></tr>
</table>
</dd>
</dl>
<p class="definition">Definition at line <a class="el" href="_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core_8cpp_source.xhtml#l00046">46</a> of file <a class="el" href="_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core_8cpp_source.xhtml">NEGEMMLowpAssemblyMatrixMultiplyCore.cpp</a>.</p>
<div class="fragment"><div class="line"><a name="l00047"></a><span class="lineno"> 47</span>&#160;{</div><div class="line"><a name="l00048"></a><span class="lineno"> 48</span>&#160; <a class="code" href="_validate_8h.xhtml#aadf5c9cff86327b96d88d04649d9715e">ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN</a>(a, 1, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58a6669348b484e3008dca2bfa8e85e40b5">DataType::U8</a>, <a class="code" href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6aafb0fced528eaac5fe170b763cda5975">DataType::S8</a>);</div><div class="line"><a name="l00049"></a><span class="lineno"> 49</span>&#160; <a class="code" href="_validate_8h.xhtml#aadf5c9cff86327b96d88d04649d9715e">ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN</a>(output, 1, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58ac8bd5bedff8ef192d39a962afc0e19ee">DataType::U32</a>, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">DataType::S32</a>);</div><div class="line"><a name="l00050"></a><span class="lineno"> 50</span>&#160; <a class="code" href="_validate_8h.xhtml#a5befbfaf6bc224eabc58b5e88b1de6d1">ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_TYPES</a>(a, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>);</div><div class="line"><a name="l00051"></a><span class="lineno"> 51</span>&#160; <a class="code" href="_error_8h.xhtml#a5bbdcf574d3f5e412fa6a1117911e67b">ARM_COMPUTE_ERROR_ON_MSG</a>((a)-&gt;<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a4f4125dba5283887b34f889b1c615c0c">info</a>()-&gt;dimension(0) != (<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>)-&gt;<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a4f4125dba5283887b34f889b1c615c0c">info</a>()-&gt;dimension(1), <span class="stringliteral">&quot;The product AB is defined only if the number of columns in A is equal to the number of rows in B&quot;</span>);</div><div class="line"><a name="l00052"></a><span class="lineno"> 52</span>&#160; <a class="code" href="_error_8h.xhtml#a5bbdcf574d3f5e412fa6a1117911e67b">ARM_COMPUTE_ERROR_ON_MSG</a>((a)-&gt;<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a4f4125dba5283887b34f889b1c615c0c">info</a>()-&gt;dimension(1) != (output)-&gt;<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a4f4125dba5283887b34f889b1c615c0c">info</a>()-&gt;dimension(1), <span class="stringliteral">&quot;The output matrix must have the same number of rows as the matrix A&quot;</span>);</div><div class="line"><a name="l00053"></a><span class="lineno"> 53</span>&#160; <a class="code" href="_error_8h.xhtml#a5bbdcf574d3f5e412fa6a1117911e67b">ARM_COMPUTE_ERROR_ON_MSG</a>((<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>)-&gt;<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a4f4125dba5283887b34f889b1c615c0c">info</a>()-&gt;dimension(0) != (output)-&gt;<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a4f4125dba5283887b34f889b1c615c0c">info</a>()-&gt;dimension(0), <span class="stringliteral">&quot;The output matrix must have the same number of columns as the matrix B&quot;</span>);</div><div class="line"><a name="l00054"></a><span class="lineno"> 54</span>&#160;</div><div class="line"><a name="l00055"></a><span class="lineno"> 55</span>&#160; <span class="keywordtype">bool</span> run_optimised = <span class="keyword">false</span>;</div><div class="line"><a name="l00056"></a><span class="lineno"> 56</span>&#160; <span class="keywordflow">switch</span>(a-&gt;<a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()-&gt;<a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a7cfb31af63202568efef5214acfbf3ba">data_type</a>())</div><div class="line"><a name="l00057"></a><span class="lineno"> 57</span>&#160; {</div><div class="line"><a name="l00058"></a><span class="lineno"> 58</span>&#160; <span class="keywordflow">case</span> <a class="code" href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6aafb0fced528eaac5fe170b763cda5975">DataType::S8</a>:</div><div class="line"><a name="l00059"></a><span class="lineno"> 59</span>&#160; <span class="keywordflow">case</span> <a class="code" href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924">DataType::QASYMM8</a>:</div><div class="line"><a name="l00060"></a><span class="lineno"> 60</span>&#160; <span class="keywordflow">case</span> <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58a6669348b484e3008dca2bfa8e85e40b5">DataType::U8</a>:</div><div class="line"><a name="l00061"></a><span class="lineno"> 61</span>&#160; {</div><div class="line"><a name="l00062"></a><span class="lineno"> 62</span>&#160; _asm_glue.<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#a89a324550419139fd5b03254a70bbd36">configure</a>(a, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>, c, output, 1.f, 0.f, <a class="code" href="classarm__compute_1_1_g_e_m_m_info.xhtml">GEMMInfo</a>(<span class="keyword">false</span>, <span class="keyword">false</span>, <span class="keyword">true</span>));</div><div class="line"><a name="l00063"></a><span class="lineno"> 63</span>&#160; run_optimised = _asm_glue.<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ab7c16a89cb470f3fa85818ee85e1e1dd">is_configured</a>();</div><div class="line"><a name="l00064"></a><span class="lineno"> 64</span>&#160; <span class="keywordflow">break</span>;</div><div class="line"><a name="l00065"></a><span class="lineno"> 65</span>&#160; }</div><div class="line"><a name="l00066"></a><span class="lineno"> 66</span>&#160; <span class="keywordflow">default</span>:</div><div class="line"><a name="l00067"></a><span class="lineno"> 67</span>&#160; {</div><div class="line"><a name="l00068"></a><span class="lineno"> 68</span>&#160; <a class="code" href="_error_8h.xhtml#a05b19c75afe9c24200a62b9724734bbd">ARM_COMPUTE_ERROR</a>(<span class="stringliteral">&quot;Datatype not supported&quot;</span>);</div><div class="line"><a name="l00069"></a><span class="lineno"> 69</span>&#160; <span class="keywordflow">break</span>;</div><div class="line"><a name="l00070"></a><span class="lineno"> 70</span>&#160; }</div><div class="line"><a name="l00071"></a><span class="lineno"> 71</span>&#160; }</div><div class="line"><a name="l00072"></a><span class="lineno"> 72</span>&#160; <span class="keywordflow">if</span>(!run_optimised)</div><div class="line"><a name="l00073"></a><span class="lineno"> 73</span>&#160; {</div><div class="line"><a name="l00074"></a><span class="lineno"> 74</span>&#160; <span class="comment">// The interleaved output matrix will have the following shape: [ a_height * 4, ceil(a_width / 4.0f) ]</span></div><div class="line"><a name="l00075"></a><span class="lineno"> 75</span>&#160; <a class="code" href="classarm__compute_1_1_tensor_shape.xhtml">TensorShape</a> shape_tmp_a = a-&gt;<a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()-&gt;<a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a7c66505457d00ece3aa4b34cab80757d">tensor_shape</a>();</div><div class="line"><a name="l00076"></a><span class="lineno"> 76</span>&#160; shape_tmp_a.<a class="code" href="classarm__compute_1_1_tensor_shape.xhtml#a9c54fb6cea3557692fe7c00c40bb40ad">set</a>(0, a-&gt;<a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()-&gt;<a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(0) * 4);</div><div class="line"><a name="l00077"></a><span class="lineno"> 77</span>&#160; shape_tmp_a.<a class="code" href="classarm__compute_1_1_tensor_shape.xhtml#a9c54fb6cea3557692fe7c00c40bb40ad">set</a>(1, std::ceil(a-&gt;<a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()-&gt;<a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(1) / 4.f));</div><div class="line"><a name="l00078"></a><span class="lineno"> 78</span>&#160;</div><div class="line"><a name="l00079"></a><span class="lineno"> 79</span>&#160; <span class="comment">// The transpose1xW output matrix will have the following shape: [ b_height * 16, ceil(b_width / 16.0f) ]</span></div><div class="line"><a name="l00080"></a><span class="lineno"> 80</span>&#160; <a class="code" href="classarm__compute_1_1_tensor_shape.xhtml">TensorShape</a> shape_tmp_b = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>-&gt;info()-&gt;tensor_shape();</div><div class="line"><a name="l00081"></a><span class="lineno"> 81</span>&#160; shape_tmp_b.<a class="code" href="classarm__compute_1_1_tensor_shape.xhtml#a9c54fb6cea3557692fe7c00c40bb40ad">set</a>(0, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>-&gt;info()-&gt;dimension(1) * 16);</div><div class="line"><a name="l00082"></a><span class="lineno"> 82</span>&#160; shape_tmp_b.<a class="code" href="classarm__compute_1_1_tensor_shape.xhtml#a9c54fb6cea3557692fe7c00c40bb40ad">set</a>(1, std::ceil(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>-&gt;info()-&gt;dimension(0) / 16.f));</div><div class="line"><a name="l00083"></a><span class="lineno"> 83</span>&#160;</div><div class="line"><a name="l00084"></a><span class="lineno"> 84</span>&#160; <a class="code" href="classarm__compute_1_1_tensor_info.xhtml">TensorInfo</a> info_a(shape_tmp_a, 1, a-&gt;<a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()-&gt;<a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a7cfb31af63202568efef5214acfbf3ba">data_type</a>());</div><div class="line"><a name="l00085"></a><span class="lineno"> 85</span>&#160; <a class="code" href="classarm__compute_1_1_tensor_info.xhtml">TensorInfo</a> info_b(shape_tmp_b, 1, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>-&gt;info()-&gt;data_type());</div><div class="line"><a name="l00086"></a><span class="lineno"> 86</span>&#160; _tmp_a.<a class="code" href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">allocator</a>()-&gt;<a class="code" href="classarm__compute_1_1_tensor_allocator.xhtml#a3fc6adad84b23f10d54d5a7b6928f872">init</a>(info_a);</div><div class="line"><a name="l00087"></a><span class="lineno"> 87</span>&#160; _tmp_b.<a class="code" href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">allocator</a>()-&gt;<a class="code" href="classarm__compute_1_1_tensor_allocator.xhtml#a3fc6adad84b23f10d54d5a7b6928f872">init</a>(info_b);</div><div class="line"><a name="l00088"></a><span class="lineno"> 88</span>&#160; _memory_group.<a class="code" href="classarm__compute_1_1_memory_group_base.xhtml#ac1f67376afb7822f262a0174ef4a3104">manage</a>(&amp;_tmp_a);</div><div class="line"><a name="l00089"></a><span class="lineno"> 89</span>&#160; _memory_group.<a class="code" href="classarm__compute_1_1_memory_group_base.xhtml#ac1f67376afb7822f262a0174ef4a3104">manage</a>(&amp;_tmp_b);</div><div class="line"><a name="l00090"></a><span class="lineno"> 90</span>&#160;</div><div class="line"><a name="l00091"></a><span class="lineno"> 91</span>&#160; <span class="comment">// Configure interleave kernel</span></div><div class="line"><a name="l00092"></a><span class="lineno"> 92</span>&#160; {</div><div class="line"><a name="l00093"></a><span class="lineno"> 93</span>&#160; <span class="keyword">auto</span> k = arm_compute::support::cpp14::make_unique&lt;NEGEMMInterleave4x4Kernel&gt;();</div><div class="line"><a name="l00094"></a><span class="lineno"> 94</span>&#160; k-&gt;configure(a, &amp;_tmp_a);</div><div class="line"><a name="l00095"></a><span class="lineno"> 95</span>&#160; _mtx_a_reshape_kernel = std::move(k);</div><div class="line"><a name="l00096"></a><span class="lineno"> 96</span>&#160; }</div><div class="line"><a name="l00097"></a><span class="lineno"> 97</span>&#160;</div><div class="line"><a name="l00098"></a><span class="lineno"> 98</span>&#160; <span class="comment">// Configure transpose kernel</span></div><div class="line"><a name="l00099"></a><span class="lineno"> 99</span>&#160; {</div><div class="line"><a name="l00100"></a><span class="lineno"> 100</span>&#160; <span class="keyword">auto</span> k = arm_compute::support::cpp14::make_unique&lt;NEGEMMTranspose1xWKernel&gt;();</div><div class="line"><a name="l00101"></a><span class="lineno"> 101</span>&#160; k-&gt;configure(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>, &amp;_tmp_b);</div><div class="line"><a name="l00102"></a><span class="lineno"> 102</span>&#160; _mtx_b_reshape_kernel = std::move(k);</div><div class="line"><a name="l00103"></a><span class="lineno"> 103</span>&#160; }</div><div class="line"><a name="l00104"></a><span class="lineno"> 104</span>&#160;</div><div class="line"><a name="l00105"></a><span class="lineno"> 105</span>&#160; <span class="comment">// Configure matrix multiply kernel</span></div><div class="line"><a name="l00106"></a><span class="lineno"> 106</span>&#160; {</div><div class="line"><a name="l00107"></a><span class="lineno"> 107</span>&#160; <span class="keyword">auto</span> k = arm_compute::support::cpp14::make_unique&lt;NEGEMMLowpMatrixMultiplyKernel&gt;();</div><div class="line"><a name="l00108"></a><span class="lineno"> 108</span>&#160; k-&gt;configure(&amp;_tmp_a, &amp;_tmp_b, output);</div><div class="line"><a name="l00109"></a><span class="lineno"> 109</span>&#160; _mm_kernel = std::move(k);</div><div class="line"><a name="l00110"></a><span class="lineno"> 110</span>&#160; }</div><div class="line"><a name="l00111"></a><span class="lineno"> 111</span>&#160;</div><div class="line"><a name="l00112"></a><span class="lineno"> 112</span>&#160; <span class="comment">// Allocate tensors</span></div><div class="line"><a name="l00113"></a><span class="lineno"> 113</span>&#160; _tmp_a.<a class="code" href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">allocator</a>()-&gt;<a class="code" href="classarm__compute_1_1_tensor_allocator.xhtml#a6e509c2a177b0b29e9e2369535094dee">allocate</a>();</div><div class="line"><a name="l00114"></a><span class="lineno"> 114</span>&#160; _tmp_b.<a class="code" href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">allocator</a>()-&gt;<a class="code" href="classarm__compute_1_1_tensor_allocator.xhtml#a6e509c2a177b0b29e9e2369535094dee">allocate</a>();</div><div class="line"><a name="l00115"></a><span class="lineno"> 115</span>&#160; }</div><div class="line"><a name="l00116"></a><span class="lineno"> 116</span>&#160;}</div><div class="ttc" id="_error_8h_xhtml_a05b19c75afe9c24200a62b9724734bbd"><div class="ttname"><a href="_error_8h.xhtml#a05b19c75afe9c24200a62b9724734bbd">ARM_COMPUTE_ERROR</a></div><div class="ttdeci">#define ARM_COMPUTE_ERROR(...)</div><div class="ttdoc">Print the given message then throw an std::runtime_error.</div><div class="ttdef"><b>Definition:</b> <a href="_error_8h_source.xhtml#l00261">Error.h:261</a></div></div>
<div class="ttc" id="classarm__compute_1_1_tensor_shape_xhtml"><div class="ttname"><a href="classarm__compute_1_1_tensor_shape.xhtml">arm_compute::TensorShape</a></div><div class="ttdoc">Shape of a tensor.</div><div class="ttdef"><b>Definition:</b> <a href="_tensor_shape_8h_source.xhtml#l00039">TensorShape.h:39</a></div></div>
<div class="ttc" id="classarm__compute_1_1_tensor_allocator_xhtml_a3fc6adad84b23f10d54d5a7b6928f872"><div class="ttname"><a href="classarm__compute_1_1_tensor_allocator.xhtml#a3fc6adad84b23f10d54d5a7b6928f872">arm_compute::TensorAllocator::init</a></div><div class="ttdeci">void init(const TensorAllocator &amp;allocator, const Coordinates &amp;coords, TensorInfo &amp;sub_info)</div><div class="ttdoc">Shares the same backing memory with another tensor allocator, while the tensor info might be differen...</div><div class="ttdef"><b>Definition:</b> <a href="src_2runtime_2_tensor_allocator_8cpp_source.xhtml#l00108">TensorAllocator.cpp:108</a></div></div>
<div class="ttc" id="classarm__compute_1_1_i_tensor_info_xhtml_a178f0d3d87f959e00a743328d95359d2"><div class="ttname"><a href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">arm_compute::ITensorInfo::dimension</a></div><div class="ttdeci">virtual size_t dimension(size_t index) const =0</div><div class="ttdoc">Return the size of the requested dimension.</div></div>
<div class="ttc" id="_validate_8h_xhtml_a5befbfaf6bc224eabc58b5e88b1de6d1"><div class="ttname"><a href="_validate_8h.xhtml#a5befbfaf6bc224eabc58b5e88b1de6d1">ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_TYPES</a></div><div class="ttdeci">#define ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_TYPES(...)</div><div class="ttdef"><b>Definition:</b> <a href="_validate_8h_source.xhtml#l00543">Validate.h:543</a></div></div>
<div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_aa76b4a6e74940dabc5b7fc6b2dab3545"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">arm_compute::test::validation::b</a></div><div class="ttdeci">SimpleTensor&lt; float &gt; b</div><div class="ttdef"><b>Definition:</b> <a href="_c_p_p_2_d_f_t_8cpp_source.xhtml#l00157">DFT.cpp:157</a></div></div>
<div class="ttc" id="namespacearm__compute_xhtml_ab4e88c89b3b7ea1735996cc4def22d58a6669348b484e3008dca2bfa8e85e40b5"><div class="ttname"><a href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58a6669348b484e3008dca2bfa8e85e40b5">arm_compute::Format::U8</a></div><div class="ttdoc">1 channel, 1 U8 per channel</div></div>
<div class="ttc" id="classarm__compute_1_1_i_tensor_info_xhtml_a7cfb31af63202568efef5214acfbf3ba"><div class="ttname"><a href="classarm__compute_1_1_i_tensor_info.xhtml#a7cfb31af63202568efef5214acfbf3ba">arm_compute::ITensorInfo::data_type</a></div><div class="ttdeci">virtual DataType data_type() const =0</div><div class="ttdoc">Data type used for each element of the tensor.</div></div>
<div class="ttc" id="classarm__compute_1_1_tensor_xhtml_adbd0cf83a8e1b335a9bf405a8e5019fa"><div class="ttname"><a href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">arm_compute::Tensor::allocator</a></div><div class="ttdeci">TensorAllocator * allocator()</div><div class="ttdoc">Return a pointer to the tensor's allocator.</div><div class="ttdef"><b>Definition:</b> <a href="runtime_2_tensor_8cpp_source.xhtml#l00048">Tensor.cpp:48</a></div></div>
<div class="ttc" id="namespacearm__compute_xhtml_ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c"><div class="ttname"><a href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">arm_compute::Format::S32</a></div><div class="ttdoc">1 channel, 1 S32 per channel</div></div>
<div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch_xhtml_a89a324550419139fd5b03254a70bbd36"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#a89a324550419139fd5b03254a70bbd36">arm_compute::NEGEMMAssemblyDispatch::configure</a></div><div class="ttdeci">void configure(const ITensor *a, const ITensor *b, const ITensor *c, ITensor *d, float alpha, float beta, const GEMMInfo &amp;gemm_info)</div><div class="ttdoc">If supported create an ACL function else fallback to the arm_gemm function.</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00380">NEGEMMAssemblyDispatch.cpp:380</a></div></div>
<div class="ttc" id="classarm__compute_1_1_memory_group_base_xhtml_ac1f67376afb7822f262a0174ef4a3104"><div class="ttname"><a href="classarm__compute_1_1_memory_group_base.xhtml#ac1f67376afb7822f262a0174ef4a3104">arm_compute::MemoryGroupBase::manage</a></div><div class="ttdeci">void manage(TensorType *obj)</div><div class="ttdoc">Sets a object to be managed by the given memory group.</div><div class="ttdef"><b>Definition:</b> <a href="_memory_group_base_8h_source.xhtml#l00102">MemoryGroupBase.h:102</a></div></div>
<div class="ttc" id="namespacearm__compute_xhtml_ab4e88c89b3b7ea1735996cc4def22d58ac8bd5bedff8ef192d39a962afc0e19ee"><div class="ttname"><a href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58ac8bd5bedff8ef192d39a962afc0e19ee">arm_compute::Format::U32</a></div><div class="ttdoc">1 channel, 1 U32 per channel</div></div>
<div class="ttc" id="classarm__compute_1_1_i_tensor_info_xhtml_a7c66505457d00ece3aa4b34cab80757d"><div class="ttname"><a href="classarm__compute_1_1_i_tensor_info.xhtml#a7c66505457d00ece3aa4b34cab80757d">arm_compute::ITensorInfo::tensor_shape</a></div><div class="ttdeci">virtual const TensorShape &amp; tensor_shape() const =0</div><div class="ttdoc">Size for each dimension of the tensor.</div></div>
<div class="ttc" id="namespacearm__compute_xhtml_ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924"><div class="ttname"><a href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924">arm_compute::DataType::QASYMM8</a></div><div class="ttdoc">quantized, asymmetric fixed-point 8-bit number</div></div>
<div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch_xhtml_ab7c16a89cb470f3fa85818ee85e1e1dd"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ab7c16a89cb470f3fa85818ee85e1e1dd">arm_compute::NEGEMMAssemblyDispatch::is_configured</a></div><div class="ttdeci">bool is_configured() const</div><div class="ttdoc">Was the function successfully configured ?</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00434">NEGEMMAssemblyDispatch.cpp:434</a></div></div>
<div class="ttc" id="classarm__compute_1_1_tensor_allocator_xhtml_a6e509c2a177b0b29e9e2369535094dee"><div class="ttname"><a href="classarm__compute_1_1_tensor_allocator.xhtml#a6e509c2a177b0b29e9e2369535094dee">arm_compute::TensorAllocator::allocate</a></div><div class="ttdeci">void allocate() override</div><div class="ttdoc">Allocate size specified by TensorInfo of CPU memory.</div><div class="ttdef"><b>Definition:</b> <a href="src_2runtime_2_tensor_allocator_8cpp_source.xhtml#l00133">TensorAllocator.cpp:133</a></div></div>
<div class="ttc" id="classarm__compute_1_1_i_tensor_xhtml_a0e95dc1e53c361348314873b168ae237"><div class="ttname"><a href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">arm_compute::ITensor::info</a></div><div class="ttdeci">virtual ITensorInfo * info() const =0</div><div class="ttdoc">Interface to be implemented by the child class to return the tensor's metadata.</div></div>
<div class="ttc" id="_validate_8h_xhtml_aadf5c9cff86327b96d88d04649d9715e"><div class="ttname"><a href="_validate_8h.xhtml#aadf5c9cff86327b96d88d04649d9715e">ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN</a></div><div class="ttdeci">#define ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(t, c,...)</div><div class="ttdef"><b>Definition:</b> <a href="_validate_8h_source.xhtml#l00789">Validate.h:789</a></div></div>
<div class="ttc" id="classarm__compute_1_1_tensor_shape_xhtml_a9c54fb6cea3557692fe7c00c40bb40ad"><div class="ttname"><a href="classarm__compute_1_1_tensor_shape.xhtml#a9c54fb6cea3557692fe7c00c40bb40ad">arm_compute::TensorShape::set</a></div><div class="ttdeci">TensorShape &amp; set(size_t dimension, size_t value, bool apply_dim_correction=true)</div><div class="ttdoc">Accessor to set the value of one of the dimensions.</div><div class="ttdef"><b>Definition:</b> <a href="_tensor_shape_8h_source.xhtml#l00078">TensorShape.h:78</a></div></div>
<div class="ttc" id="classarm__compute_1_1_tensor_info_xhtml"><div class="ttname"><a href="classarm__compute_1_1_tensor_info.xhtml">arm_compute::TensorInfo</a></div><div class="ttdoc">Store the tensor's metadata.</div><div class="ttdef"><b>Definition:</b> <a href="_tensor_info_8h_source.xhtml#l00045">TensorInfo.h:45</a></div></div>
<div class="ttc" id="classarm__compute_1_1_g_e_m_m_info_xhtml"><div class="ttname"><a href="classarm__compute_1_1_g_e_m_m_info.xhtml">arm_compute::GEMMInfo</a></div><div class="ttdoc">GEMM information class.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_types_8h_source.xhtml#l01880">Types.h:1880</a></div></div>
<div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_a4f4125dba5283887b34f889b1c615c0c"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#a4f4125dba5283887b34f889b1c615c0c">arm_compute::test::validation::info</a></div><div class="ttdeci">info</div><div class="ttdef"><b>Definition:</b> <a href="validation_2_c_l_2_convolution_layer_8cpp_source.xhtml#l00174">ConvolutionLayer.cpp:174</a></div></div>
<div class="ttc" id="namespacearm__compute_xhtml_ad8ed01ff3ff33333d8e19db4d2818bb6aafb0fced528eaac5fe170b763cda5975"><div class="ttname"><a href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6aafb0fced528eaac5fe170b763cda5975">arm_compute::DataType::S8</a></div><div class="ttdoc">signed 8-bit number</div></div>
<div class="ttc" id="_error_8h_xhtml_a5bbdcf574d3f5e412fa6a1117911e67b"><div class="ttname"><a href="_error_8h.xhtml#a5bbdcf574d3f5e412fa6a1117911e67b">ARM_COMPUTE_ERROR_ON_MSG</a></div><div class="ttdeci">#define ARM_COMPUTE_ERROR_ON_MSG(cond,...)</div><div class="ttdef"><b>Definition:</b> <a href="_error_8h_source.xhtml#l00328">Error.h:328</a></div></div>
</div><!-- fragment -->
<p class="reference">References <a class="el" href="src_2runtime_2_tensor_allocator_8cpp_source.xhtml#l00133">TensorAllocator::allocate()</a>, <a class="el" href="runtime_2_tensor_8cpp_source.xhtml#l00048">Tensor::allocator()</a>, <a class="el" href="_error_8h_source.xhtml#l00261">ARM_COMPUTE_ERROR</a>, <a class="el" href="_validate_8h_source.xhtml#l00789">ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN</a>, <a class="el" href="_validate_8h_source.xhtml#l00543">ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_TYPES</a>, <a class="el" href="_error_8h_source.xhtml#l00328">ARM_COMPUTE_ERROR_ON_MSG</a>, <a class="el" href="_c_p_p_2_d_f_t_8cpp_source.xhtml#l00157">arm_compute::test::validation::b</a>, <a class="el" href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00380">NEGEMMAssemblyDispatch::configure()</a>, <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml#a7cfb31af63202568efef5214acfbf3ba">ITensorInfo::data_type()</a>, <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">ITensorInfo::dimension()</a>, <a class="el" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">ITensor::info()</a>, <a class="el" href="validation_2_c_l_2_convolution_layer_8cpp_source.xhtml#l00174">arm_compute::test::validation::info</a>, <a class="el" href="src_2runtime_2_tensor_allocator_8cpp_source.xhtml#l00108">TensorAllocator::init()</a>, <a class="el" href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00434">NEGEMMAssemblyDispatch::is_configured()</a>, <a class="el" href="_memory_group_base_8h_source.xhtml#l00102">MemoryGroupBase&lt; TensorType &gt;::manage()</a>, <a class="el" href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924">arm_compute::QASYMM8</a>, <a class="el" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">arm_compute::S32</a>, <a class="el" href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6aafb0fced528eaac5fe170b763cda5975">arm_compute::S8</a>, <a class="el" href="_tensor_shape_8h_source.xhtml#l00078">TensorShape::set()</a>, <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml#a7c66505457d00ece3aa4b34cab80757d">ITensorInfo::tensor_shape()</a>, <a class="el" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58ac8bd5bedff8ef192d39a962afc0e19ee">arm_compute::U32</a>, and <a class="el" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58a6669348b484e3008dca2bfa8e85e40b5">arm_compute::U8</a>.</p>
</div>
</div>
<a id="ad1717410afd0be936c6213a63c8005fb"></a>
<h2 class="memtitle"><span class="permalink"><a href="#ad1717410afd0be936c6213a63c8005fb">&#9670;&nbsp;</a></span>run()</h2>
<div class="memitem">
<div class="memproto">
<table class="mlabels">
<tr>
<td class="mlabels-left">
<table class="memname">
<tr>
<td class="memname">void run </td>
<td>(</td>
<td class="paramname"></td><td>)</td>
<td></td>
</tr>
</table>
</td>
<td class="mlabels-right">
<span class="mlabels"><span class="mlabel">override</span><span class="mlabel">virtual</span></span> </td>
</tr>
</table>
</div><div class="memdoc">
<p>Run the kernels contained in the function. </p>
<p>For NEON kernels:</p><ul>
<li>Multi-threading is used for the kernels which are parallelisable.</li>
<li>By default std::thread::hardware_concurrency() threads are used.</li>
</ul>
<dl class="section note"><dt>Note</dt><dd><a class="el" href="classarm__compute_1_1_c_p_p_scheduler.xhtml#ae64eebaa07f4d2da6cc2ba538c3cb095">CPPScheduler::set_num_threads()</a> can be used to manually set the number of threads</dd></dl>
<p>For OpenCL kernels:</p><ul>
<li>All the kernels are enqueued on the queue associated with <a class="el" href="classarm__compute_1_1_c_l_scheduler.xhtml" title="Provides global access to a CL context and command queue.">CLScheduler</a>.</li>
<li>The queue is then flushed.</li>
</ul>
<dl class="section note"><dt>Note</dt><dd>The function will not block until the kernels are executed. It is the user's responsibility to wait. </dd>
<dd>
Will call <a class="el" href="classarm__compute_1_1_i_function.xhtml#a820f7291c24155a2980512fae45aac26" title="Prepare the function for executing.">prepare()</a> on first run if hasn't been done </dd></dl>
<p>Implements <a class="el" href="classarm__compute_1_1_i_function.xhtml#a18954417d3124a8095783ea13dc6d00b">IFunction</a>.</p>
<p class="definition">Definition at line <a class="el" href="_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core_8cpp_source.xhtml#l00118">118</a> of file <a class="el" href="_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core_8cpp_source.xhtml">NEGEMMLowpAssemblyMatrixMultiplyCore.cpp</a>.</p>
<div class="fragment"><div class="line"><a name="l00119"></a><span class="lineno"> 119</span>&#160;{</div><div class="line"><a name="l00120"></a><span class="lineno"> 120</span>&#160; <a class="code" href="classarm__compute_1_1_memory_group_resource_scope.xhtml">MemoryGroupResourceScope</a> scope_mg(_memory_group);</div><div class="line"><a name="l00121"></a><span class="lineno"> 121</span>&#160; <span class="keywordflow">if</span>(_mtx_a_reshape_kernel)</div><div class="line"><a name="l00122"></a><span class="lineno"> 122</span>&#160; {</div><div class="line"><a name="l00123"></a><span class="lineno"> 123</span>&#160; <a class="code" href="classarm__compute_1_1_scheduler.xhtml#a0d63ca713bab377aabcfb63c192b8429">NEScheduler::get</a>().<a class="code" href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">schedule</a>(_mtx_a_reshape_kernel.get(), <a class="code" href="classarm__compute_1_1_window.xhtml#ad2d402364fa822b0b7775081291eeca9">Window::DimY</a>);</div><div class="line"><a name="l00124"></a><span class="lineno"> 124</span>&#160; }</div><div class="line"><a name="l00125"></a><span class="lineno"> 125</span>&#160;</div><div class="line"><a name="l00126"></a><span class="lineno"> 126</span>&#160; <span class="keywordflow">if</span>(_mtx_b_reshape_kernel)</div><div class="line"><a name="l00127"></a><span class="lineno"> 127</span>&#160; {</div><div class="line"><a name="l00128"></a><span class="lineno"> 128</span>&#160; <a class="code" href="classarm__compute_1_1_scheduler.xhtml#a0d63ca713bab377aabcfb63c192b8429">NEScheduler::get</a>().<a class="code" href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">schedule</a>(_mtx_b_reshape_kernel.get(), <a class="code" href="classarm__compute_1_1_window.xhtml#ad2d402364fa822b0b7775081291eeca9">Window::DimY</a>);</div><div class="line"><a name="l00129"></a><span class="lineno"> 129</span>&#160; }</div><div class="line"><a name="l00130"></a><span class="lineno"> 130</span>&#160;</div><div class="line"><a name="l00131"></a><span class="lineno"> 131</span>&#160; <span class="keywordflow">if</span>(_asm_glue.<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ab7c16a89cb470f3fa85818ee85e1e1dd">is_configured</a>())</div><div class="line"><a name="l00132"></a><span class="lineno"> 132</span>&#160; {</div><div class="line"><a name="l00133"></a><span class="lineno"> 133</span>&#160; _asm_glue.<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ad1717410afd0be936c6213a63c8005fb">run</a>();</div><div class="line"><a name="l00134"></a><span class="lineno"> 134</span>&#160; }</div><div class="line"><a name="l00135"></a><span class="lineno"> 135</span>&#160; <span class="keywordflow">else</span></div><div class="line"><a name="l00136"></a><span class="lineno"> 136</span>&#160; {</div><div class="line"><a name="l00137"></a><span class="lineno"> 137</span>&#160; <a class="code" href="classarm__compute_1_1_scheduler.xhtml#a0d63ca713bab377aabcfb63c192b8429">NEScheduler::get</a>().<a class="code" href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">schedule</a>(_mm_kernel.get(), <a class="code" href="classarm__compute_1_1_window.xhtml#ad2d402364fa822b0b7775081291eeca9">Window::DimY</a>);</div><div class="line"><a name="l00138"></a><span class="lineno"> 138</span>&#160; }</div><div class="line"><a name="l00139"></a><span class="lineno"> 139</span>&#160;}</div><div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch_xhtml_ad1717410afd0be936c6213a63c8005fb"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ad1717410afd0be936c6213a63c8005fb">arm_compute::NEGEMMAssemblyDispatch::run</a></div><div class="ttdeci">void run() override</div><div class="ttdoc">Run the kernels contained in the function.</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00439">NEGEMMAssemblyDispatch.cpp:439</a></div></div>
<div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch_xhtml_ab7c16a89cb470f3fa85818ee85e1e1dd"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ab7c16a89cb470f3fa85818ee85e1e1dd">arm_compute::NEGEMMAssemblyDispatch::is_configured</a></div><div class="ttdeci">bool is_configured() const</div><div class="ttdoc">Was the function successfully configured ?</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00434">NEGEMMAssemblyDispatch.cpp:434</a></div></div>
<div class="ttc" id="classarm__compute_1_1_window_xhtml_ad2d402364fa822b0b7775081291eeca9"><div class="ttname"><a href="classarm__compute_1_1_window.xhtml#ad2d402364fa822b0b7775081291eeca9">arm_compute::Window::DimY</a></div><div class="ttdeci">static constexpr size_t DimY</div><div class="ttdoc">Alias for dimension 1 also known as Y dimension.</div><div class="ttdef"><b>Definition:</b> <a href="_window_8h_source.xhtml#l00045">Window.h:45</a></div></div>
<div class="ttc" id="classarm__compute_1_1_memory_group_resource_scope_xhtml"><div class="ttname"><a href="classarm__compute_1_1_memory_group_resource_scope.xhtml">arm_compute::MemoryGroupResourceScope</a></div><div class="ttdoc">Memory group resources scope handling class.</div><div class="ttdef"><b>Definition:</b> <a href="_i_memory_group_8h_source.xhtml#l00046">IMemoryGroup.h:46</a></div></div>
<div class="ttc" id="classarm__compute_1_1_i_scheduler_xhtml_a4e58f95544bd5ac6559a421671bd9842"><div class="ttname"><a href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">arm_compute::IScheduler::schedule</a></div><div class="ttdeci">virtual void schedule(ICPPKernel *kernel, const Hints &amp;hints)=0</div><div class="ttdoc">Runs the kernel in the same thread as the caller synchronously.</div></div>
<div class="ttc" id="classarm__compute_1_1_scheduler_xhtml_a0d63ca713bab377aabcfb63c192b8429"><div class="ttname"><a href="classarm__compute_1_1_scheduler.xhtml#a0d63ca713bab377aabcfb63c192b8429">arm_compute::Scheduler::get</a></div><div class="ttdeci">static IScheduler &amp; get()</div><div class="ttdoc">Access the scheduler singleton.</div><div class="ttdef"><b>Definition:</b> <a href="_scheduler_8cpp_source.xhtml#l00096">Scheduler.cpp:96</a></div></div>
</div><!-- fragment -->
<p class="reference">References <a class="el" href="_window_8h_source.xhtml#l00045">Window::DimY</a>, <a class="el" href="_scheduler_8cpp_source.xhtml#l00096">Scheduler::get()</a>, <a class="el" href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00434">NEGEMMAssemblyDispatch::is_configured()</a>, <a class="el" href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00439">NEGEMMAssemblyDispatch::run()</a>, and <a class="el" href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">IScheduler::schedule()</a>.</p>
</div>
</div>
<hr/>The documentation for this class was generated from the following files:<ul>
<li>arm_compute/runtime/NEON/functions/<a class="el" href="_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core_8h_source.xhtml">NEGEMMLowpAssemblyMatrixMultiplyCore.h</a></li>
<li>src/runtime/NEON/functions/<a class="el" href="_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core_8cpp_source.xhtml">NEGEMMLowpAssemblyMatrixMultiplyCore.cpp</a></li>
</ul>
</div><!-- contents -->
</div><!-- doc-content -->
<!-- start footer part -->
<div id="nav-path" class="navpath"><!-- id is needed for treeview function! -->
<ul>
<li class="navelem"><a class="el" href="namespacearm__compute.xhtml">arm_compute</a></li><li class="navelem"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_assembly_matrix_multiply_core.xhtml">NEGEMMLowpAssemblyMatrixMultiplyCore</a></li>
<li class="footer">Generated on Mon Sep 2 2019 11:47:37 for Compute Library by
<a href="http://www.doxygen.org/index.html">
<img class="footer" src="doxygen.png" alt="doxygen"/></a> 1.8.15 </li>
</ul>
</div>
</body>
</html>