| <!-- HTML header for doxygen 1.8.17--> |
| <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "https://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 1.8.17"/> |
| <meta name="viewport" content="width=device-width, initial-scale=1"/> |
| <title>Arm NN: src/backends/neon/workloads/NeonBatchMatMulWorkload.cpp 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="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> |
| <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="customdoxygen.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="ArmNN" src="Arm_NN_horizontal_blue.png" style="max-width: 15rem; margin-top: .5rem; margin-left 13px"/> |
| <td id="projectalign" style="padding-left: 0.9em;"> |
| <div id="projectname"> |
|  <span id="projectnumber">24.05</span> |
| </div> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </div> |
| <!-- end header part --> |
| <!-- Generated by Doxygen 1.8.17 --> |
| <script type="text/javascript"> |
| /* @license magnet:?xt=urn:btih:cf05388f2679ee054f2beb29a391d25f4e673ac3&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&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&dn=gpl-2.0.txt GPL-v2 */ |
| $(document).ready(function(){initNavTree('_neon_batch_mat_mul_workload_8cpp_source.html',''); initResizable(); }); |
| /* @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="headertitle"> |
| <div class="title">NeonBatchMatMulWorkload.cpp</div> </div> |
| </div><!--header--> |
| <div class="contents"> |
| <a href="_neon_batch_mat_mul_workload_8cpp.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 © 2022-2023 Arm Ltd and Contributors. All rights reserved.</span></div> |
| <div class="line"><a name="l00003"></a><span class="lineno"> 3</span> <span class="comment">// SPDX-License-Identifier: MIT</span></div> |
| <div class="line"><a name="l00004"></a><span class="lineno"> 4</span> <span class="comment">//</span></div> |
| <div class="line"><a name="l00005"></a><span class="lineno"> 5</span>  </div> |
| <div class="line"><a name="l00006"></a><span class="lineno"> 6</span> <span class="preprocessor">#include "<a class="code" href="_neon_batch_mat_mul_workload_8hpp.html">NeonBatchMatMulWorkload.hpp</a>"</span></div> |
| <div class="line"><a name="l00007"></a><span class="lineno"> 7</span>  </div> |
| <div class="line"><a name="l00008"></a><span class="lineno"> 8</span> <span class="preprocessor">#include "<a class="code" href="_neon_workload_utils_8hpp.html">NeonWorkloadUtils.hpp</a>"</span></div> |
| <div class="line"><a name="l00009"></a><span class="lineno"> 9</span>  </div> |
| <div class="line"><a name="l00010"></a><span class="lineno"> 10</span> <span class="preprocessor">#include <<a class="code" href="_polymorphic_downcast_8hpp.html">armnn/utility/PolymorphicDowncast.hpp</a>></span></div> |
| <div class="line"><a name="l00011"></a><span class="lineno"> 11</span> <span class="preprocessor">#include <<a class="code" href="_arm_compute_utils_8hpp.html">aclCommon/ArmComputeUtils.hpp</a>></span></div> |
| <div class="line"><a name="l00012"></a><span class="lineno"> 12</span>  </div> |
| <div class="line"><a name="l00013"></a><span class="lineno"> 13</span> <span class="preprocessor">#include <<a class="code" href="_workload_utils_8hpp.html">backendsCommon/WorkloadUtils.hpp</a>></span></div> |
| <div class="line"><a name="l00014"></a><span class="lineno"> 14</span>  </div> |
| <div class="line"><a name="l00015"></a><span class="lineno"> 15</span> <span class="preprocessor">#include <arm_compute/function_info/MatMulInfo.h></span></div> |
| <div class="line"><a name="l00016"></a><span class="lineno"> 16</span>  </div> |
| <div class="line"><a name="l00017"></a><span class="lineno"> 17</span> <span class="keyword">namespace </span><a class="code" href="namespacearmnn.html">armnn</a></div> |
| <div class="line"><a name="l00018"></a><span class="lineno"> 18</span> {</div> |
| <div class="line"><a name="l00019"></a><span class="lineno"><a class="line" href="namespacearmnn.html#a52cdeda888e3ecf0dbebb533ba0856f1"> 19</a></span> <a class="code" href="namespacearmnn.html#a67a0db04d321a74b7e7fcfd3f1a3f70b">arm_compute::Status</a> <a class="code" href="namespacearmnn.html#a52cdeda888e3ecf0dbebb533ba0856f1">NeonBatchMatMulValidate</a>(<span class="keyword">const</span> <a class="code" href="classarmnn_1_1_tensor_info.html">TensorInfo</a>& inputInfoX,</div> |
| <div class="line"><a name="l00020"></a><span class="lineno"> 20</span>  <span class="keyword">const</span> <a class="code" href="classarmnn_1_1_tensor_info.html">TensorInfo</a>& inputInfoY,</div> |
| <div class="line"><a name="l00021"></a><span class="lineno"> 21</span>  <span class="keyword">const</span> <a class="code" href="classarmnn_1_1_tensor_info.html">TensorInfo</a>& outputInfo,</div> |
| <div class="line"><a name="l00022"></a><span class="lineno"> 22</span>  <span class="keyword">const</span> <a class="code" href="structarmnn_1_1_batch_mat_mul_descriptor.html">BatchMatMulDescriptor</a>& descriptor,</div> |
| <div class="line"><a name="l00023"></a><span class="lineno"> 23</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> isFastMathEnabled,</div> |
| <div class="line"><a name="l00024"></a><span class="lineno"> 24</span>  <span class="keyword">const</span> <a class="code" href="structarmnn_1_1_activation_descriptor.html">ActivationDescriptor</a>* activationDescriptor)</div> |
| <div class="line"><a name="l00025"></a><span class="lineno"> 25</span> {</div> |
| <div class="line"><a name="l00026"></a><span class="lineno"> 26</span>  <span class="keywordflow">if</span> (descriptor.<a class="code" href="structarmnn_1_1_batch_mat_mul_descriptor.html#a0cf8306be7d301de0f095fff9901a525">m_AdjointX</a> || descriptor.<a class="code" href="structarmnn_1_1_batch_mat_mul_descriptor.html#ad945fc98770356dd886a68e98a52e26b">m_AdjointY</a> )</div> |
| <div class="line"><a name="l00027"></a><span class="lineno"> 27</span>  {</div> |
| <div class="line"><a name="l00028"></a><span class="lineno"> 28</span>  <span class="keywordflow">throw</span> <a class="code" href="classarmnn_1_1_exception.html">Exception</a>(<span class="stringliteral">"Support for adjoint not implemented."</span>);</div> |
| <div class="line"><a name="l00029"></a><span class="lineno"> 29</span>  }</div> |
| <div class="line"><a name="l00030"></a><span class="lineno"> 30</span>  <span class="keywordflow">if</span> (descriptor.<a class="code" href="structarmnn_1_1_batch_mat_mul_descriptor.html#aedca000a005e091c23191e82d7e81b1d">m_DataLayoutX</a> != <a class="code" href="namespacearmnn.html#ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f">armnn::DataLayout::NCHW</a> || descriptor.<a class="code" href="structarmnn_1_1_batch_mat_mul_descriptor.html#aaf7828880989b4b9378d3e86aa6dc843">m_DataLayoutY</a> != <a class="code" href="namespacearmnn.html#ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f">armnn::DataLayout::NCHW</a> )</div> |
| <div class="line"><a name="l00031"></a><span class="lineno"> 31</span>  {</div> |
| <div class="line"><a name="l00032"></a><span class="lineno"> 32</span>  <span class="keywordflow">throw</span> <a class="code" href="classarmnn_1_1_exception.html">Exception</a>(<span class="stringliteral">"Only supported the MatMul in the last 2 dimensions"</span>);</div> |
| <div class="line"><a name="l00033"></a><span class="lineno"> 33</span>  }</div> |
| <div class="line"><a name="l00034"></a><span class="lineno"> 34</span>  </div> |
| <div class="line"><a name="l00035"></a><span class="lineno"> 35</span>  arm_compute::TensorInfo aclInputInfoX = armcomputetensorutils::BuildArmComputeTensorInfo(inputInfoX);</div> |
| <div class="line"><a name="l00036"></a><span class="lineno"> 36</span>  arm_compute::TensorInfo aclInputInfoY = armcomputetensorutils::BuildArmComputeTensorInfo(inputInfoY);</div> |
| <div class="line"><a name="l00037"></a><span class="lineno"> 37</span>  arm_compute::TensorInfo aclOutputInfo = armcomputetensorutils::BuildArmComputeTensorInfo(outputInfo);</div> |
| <div class="line"><a name="l00038"></a><span class="lineno"> 38</span>  </div> |
| <div class="line"><a name="l00039"></a><span class="lineno"> 39</span>  <span class="comment">// GeMM dispatches kernel handles dynamic inputs differently to static so this flag needs to be set</span></div> |
| <div class="line"><a name="l00040"></a><span class="lineno"> 40</span>  aclInputInfoX.set_are_values_constant(<span class="keyword">false</span>);</div> |
| <div class="line"><a name="l00041"></a><span class="lineno"> 41</span>  aclInputInfoY.set_are_values_constant(<span class="keyword">false</span>);</div> |
| <div class="line"><a name="l00042"></a><span class="lineno"> 42</span>  </div> |
| <div class="line"><a name="l00043"></a><span class="lineno"> 43</span>  <span class="keyword">const</span> arm_compute::ActivationLayerInfo activationInfo = <a class="code" href="namespacearmnn.html#ad701d0d29baa4266ab4d33b090aa661c">ConvertActivationDescriptorToAclActivationLayerInfo</a>(</div> |
| <div class="line"><a name="l00044"></a><span class="lineno"> 44</span>  activationDescriptor);</div> |
| <div class="line"><a name="l00045"></a><span class="lineno"> 45</span>  </div> |
| <div class="line"><a name="l00046"></a><span class="lineno"> 46</span>  arm_compute::MatMulInfo matMulInfo;</div> |
| <div class="line"><a name="l00047"></a><span class="lineno"> 47</span>  matMulInfo.adj_lhs(descriptor.<a class="code" href="structarmnn_1_1_batch_mat_mul_descriptor.html#acb441bb8db19bcce78d15cdd8ceb5ea0">m_TransposeX</a>);</div> |
| <div class="line"><a name="l00048"></a><span class="lineno"> 48</span>  matMulInfo.adj_rhs(descriptor.<a class="code" href="structarmnn_1_1_batch_mat_mul_descriptor.html#a112b466e5d2ab9d1887178adbe3afa1c">m_TransposeY</a>);</div> |
| <div class="line"><a name="l00049"></a><span class="lineno"> 49</span>  </div> |
| <div class="line"><a name="l00050"></a><span class="lineno"> 50</span>  arm_compute::CpuMatMulSettings settings;</div> |
| <div class="line"><a name="l00051"></a><span class="lineno"> 51</span>  settings.fast_math(isFastMathEnabled);</div> |
| <div class="line"><a name="l00052"></a><span class="lineno"> 52</span>  </div> |
| <div class="line"><a name="l00053"></a><span class="lineno"> 53</span>  <span class="keywordflow">return</span> arm_compute::NEMatMul::validate(&aclInputInfoX, &aclInputInfoY, &aclOutputInfo, matMulInfo, settings,</div> |
| <div class="line"><a name="l00054"></a><span class="lineno"> 54</span>  activationInfo);</div> |
| <div class="line"><a name="l00055"></a><span class="lineno"> 55</span> }</div> |
| <div class="line"><a name="l00056"></a><span class="lineno"> 56</span>  </div> |
| <div class="line"><a name="l00057"></a><span class="lineno"><a class="line" href="classarmnn_1_1_neon_batch_mat_mul_workload.html#adf0e0cd06d407af2ae72ac3613f01a0e"> 57</a></span> <a class="code" href="classarmnn_1_1_neon_batch_mat_mul_workload.html#adf0e0cd06d407af2ae72ac3613f01a0e">NeonBatchMatMulWorkload::NeonBatchMatMulWorkload</a>(<span class="keyword">const</span> <a class="code" href="structarmnn_1_1_batch_mat_mul_queue_descriptor.html">BatchMatMulQueueDescriptor</a>& descriptor,</div> |
| <div class="line"><a name="l00058"></a><span class="lineno"> 58</span>  <span class="keyword">const</span> <a class="code" href="structarmnn_1_1_workload_info.html">WorkloadInfo</a>& info,</div> |
| <div class="line"><a name="l00059"></a><span class="lineno"> 59</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> isFastMathEnabled)</div> |
| <div class="line"><a name="l00060"></a><span class="lineno"> 60</span>  : <a class="code" href="classarmnn_1_1_neon_base_workload.html">NeonBaseWorkload</a><<a class="code" href="structarmnn_1_1_batch_mat_mul_queue_descriptor.html">BatchMatMulQueueDescriptor</a>>(descriptor, <a class="code" href="namespacearmnn.html#a4dc0adc6737b5944e7671bee71788407acaf9b6b99962bf5c2264824231d7a40c">info</a>)</div> |
| <div class="line"><a name="l00061"></a><span class="lineno"> 61</span> {</div> |
| <div class="line"><a name="l00062"></a><span class="lineno"> 62</span>  <span class="keywordflow">if</span> (descriptor.<a class="code" href="structarmnn_1_1_queue_descriptor_with_parameters.html#aad91b9bbf7aa365d304febe79a3d1333">m_Parameters</a>.<a class="code" href="structarmnn_1_1_batch_mat_mul_descriptor.html#a0cf8306be7d301de0f095fff9901a525">m_AdjointX</a> || descriptor.<a class="code" href="structarmnn_1_1_queue_descriptor_with_parameters.html#aad91b9bbf7aa365d304febe79a3d1333">m_Parameters</a>.<a class="code" href="structarmnn_1_1_batch_mat_mul_descriptor.html#ad945fc98770356dd886a68e98a52e26b">m_AdjointY</a> )</div> |
| <div class="line"><a name="l00063"></a><span class="lineno"> 63</span>  {</div> |
| <div class="line"><a name="l00064"></a><span class="lineno"> 64</span>  <span class="keywordflow">throw</span> <a class="code" href="classarmnn_1_1_exception.html">Exception</a>(<span class="stringliteral">"Support for adjoint not implemented."</span>);</div> |
| <div class="line"><a name="l00065"></a><span class="lineno"> 65</span>  }</div> |
| <div class="line"><a name="l00066"></a><span class="lineno"> 66</span>  <span class="keywordflow">if</span> (descriptor.<a class="code" href="structarmnn_1_1_queue_descriptor_with_parameters.html#aad91b9bbf7aa365d304febe79a3d1333">m_Parameters</a>.<a class="code" href="structarmnn_1_1_batch_mat_mul_descriptor.html#aedca000a005e091c23191e82d7e81b1d">m_DataLayoutX</a> != <a class="code" href="namespacearmnn.html#ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f">armnn::DataLayout::NCHW</a></div> |
| <div class="line"><a name="l00067"></a><span class="lineno"> 67</span>  || descriptor.<a class="code" href="structarmnn_1_1_queue_descriptor_with_parameters.html#aad91b9bbf7aa365d304febe79a3d1333">m_Parameters</a>.<a class="code" href="structarmnn_1_1_batch_mat_mul_descriptor.html#aaf7828880989b4b9378d3e86aa6dc843">m_DataLayoutY</a> != <a class="code" href="namespacearmnn.html#ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f">armnn::DataLayout::NCHW</a> )</div> |
| <div class="line"><a name="l00068"></a><span class="lineno"> 68</span>  {</div> |
| <div class="line"><a name="l00069"></a><span class="lineno"> 69</span>  <span class="keywordflow">throw</span> <a class="code" href="classarmnn_1_1_exception.html">Exception</a>(<span class="stringliteral">"Only supported the MatMul in the last 2 dimensions"</span>);</div> |
| <div class="line"><a name="l00070"></a><span class="lineno"> 70</span>  }</div> |
| <div class="line"><a name="l00071"></a><span class="lineno"> 71</span>  </div> |
| <div class="line"><a name="l00072"></a><span class="lineno"> 72</span>  <a class="code" href="classarmnn_1_1_base_workload.html#afb8d2c8817c75de9d01a4c0e0d5c160b">m_Data</a>.<a class="code" href="structarmnn_1_1_queue_descriptor.html#a765d2cee4ccce5b9467e0c2b6d25b84a">ValidateInputsOutputs</a>(<span class="stringliteral">"NeonBatchMatMulWorkload"</span>, 2, 1);</div> |
| <div class="line"><a name="l00073"></a><span class="lineno"> 73</span>  </div> |
| <div class="line"><a name="l00074"></a><span class="lineno"> 74</span>  arm_compute::ITensor& inputX = PolymorphicDowncast<IAclTensorHandle*>(<a class="code" href="classarmnn_1_1_base_workload.html#afb8d2c8817c75de9d01a4c0e0d5c160b">m_Data</a>.<a class="code" href="structarmnn_1_1_queue_descriptor.html#a4b50e46a6810018f3edecfb68b2a76b3">m_Inputs</a>[0])->GetTensor();</div> |
| <div class="line"><a name="l00075"></a><span class="lineno"> 75</span>  arm_compute::ITensor& inputY = PolymorphicDowncast<IAclTensorHandle*>(<a class="code" href="classarmnn_1_1_base_workload.html#afb8d2c8817c75de9d01a4c0e0d5c160b">m_Data</a>.<a class="code" href="structarmnn_1_1_queue_descriptor.html#a4b50e46a6810018f3edecfb68b2a76b3">m_Inputs</a>[1])->GetTensor();</div> |
| <div class="line"><a name="l00076"></a><span class="lineno"> 76</span>  arm_compute::ITensor& output = PolymorphicDowncast<IAclTensorHandle*>(<a class="code" href="classarmnn_1_1_base_workload.html#afb8d2c8817c75de9d01a4c0e0d5c160b">m_Data</a>.<a class="code" href="structarmnn_1_1_queue_descriptor.html#a6abd491bb99ffe88bd472c1ae5a1ed1a">m_Outputs</a>[0])->GetTensor();</div> |
| <div class="line"><a name="l00077"></a><span class="lineno"> 77</span>  </div> |
| <div class="line"><a name="l00078"></a><span class="lineno"> 78</span>  <span class="comment">// GeMM dispatches kernel handles dynamic inputs differently to static so this flag needs to be set</span></div> |
| <div class="line"><a name="l00079"></a><span class="lineno"> 79</span>  inputX.info()->set_are_values_constant(<span class="keyword">false</span>);</div> |
| <div class="line"><a name="l00080"></a><span class="lineno"> 80</span>  inputY.info()->set_are_values_constant(<span class="keyword">false</span>);</div> |
| <div class="line"><a name="l00081"></a><span class="lineno"> 81</span>  </div> |
| <div class="line"><a name="l00082"></a><span class="lineno"> 82</span>  <span class="keyword">const</span> arm_compute::ActivationLayerInfo activationInfo = <a class="code" href="namespacearmnn.html#abfb0841058a8190d30851f07eca3991f">ConvertAdditionalInfoToAclActivationLayerInfo</a>(descriptor);</div> |
| <div class="line"><a name="l00083"></a><span class="lineno"> 83</span>  </div> |
| <div class="line"><a name="l00084"></a><span class="lineno"> 84</span>  arm_compute::MatMulInfo matMulInfo;</div> |
| <div class="line"><a name="l00085"></a><span class="lineno"> 85</span>  matMulInfo.adj_lhs(descriptor.<a class="code" href="structarmnn_1_1_queue_descriptor_with_parameters.html#aad91b9bbf7aa365d304febe79a3d1333">m_Parameters</a>.<a class="code" href="structarmnn_1_1_batch_mat_mul_descriptor.html#acb441bb8db19bcce78d15cdd8ceb5ea0">m_TransposeX</a>);</div> |
| <div class="line"><a name="l00086"></a><span class="lineno"> 86</span>  matMulInfo.adj_rhs(descriptor.<a class="code" href="structarmnn_1_1_queue_descriptor_with_parameters.html#aad91b9bbf7aa365d304febe79a3d1333">m_Parameters</a>.<a class="code" href="structarmnn_1_1_batch_mat_mul_descriptor.html#a112b466e5d2ab9d1887178adbe3afa1c">m_TransposeY</a>);</div> |
| <div class="line"><a name="l00087"></a><span class="lineno"> 87</span>  </div> |
| <div class="line"><a name="l00088"></a><span class="lineno"> 88</span>  arm_compute::CpuMatMulSettings settings;</div> |
| <div class="line"><a name="l00089"></a><span class="lineno"> 89</span>  settings.fast_math(isFastMathEnabled);</div> |
| <div class="line"><a name="l00090"></a><span class="lineno"> 90</span>  </div> |
| <div class="line"><a name="l00091"></a><span class="lineno"> 91</span>  m_MatMulLayer.configure(&inputX, &inputY, &output, matMulInfo, settings, activationInfo);</div> |
| <div class="line"><a name="l00092"></a><span class="lineno"> 92</span>  </div> |
| <div class="line"><a name="l00093"></a><span class="lineno"> 93</span>  <span class="comment">// Report Profiling Details</span></div> |
| <div class="line"><a name="l00094"></a><span class="lineno"> 94</span>  <a class="code" href="structarmnn_1_1_workload_info.html">WorkloadInfo</a> detailsInfo;</div> |
| <div class="line"><a name="l00095"></a><span class="lineno"> 95</span>  detailsInfo.<a class="code" href="structarmnn_1_1_workload_info.html#ac97905bfa0daab357b91df1347600309">m_InputTensorInfos</a> = <a class="code" href="namespacearmnn.html#a4dc0adc6737b5944e7671bee71788407acaf9b6b99962bf5c2264824231d7a40c">info</a>.m_InputTensorInfos;</div> |
| <div class="line"><a name="l00096"></a><span class="lineno"> 96</span>  detailsInfo.<a class="code" href="structarmnn_1_1_workload_info.html#a67b178f8a836bc1e52b8de109760adfd">m_OutputTensorInfos</a> = <a class="code" href="namespacearmnn.html#a4dc0adc6737b5944e7671bee71788407acaf9b6b99962bf5c2264824231d7a40c">info</a>.m_OutputTensorInfos;</div> |
| <div class="line"><a name="l00097"></a><span class="lineno"> 97</span>  <a class="code" href="_profiling_8hpp.html#a786492a3881a4c760ab1eec2149f4aba">ARMNN_REPORT_PROFILING_WORKLOAD_DESC</a>(<span class="stringliteral">"NeonBatchMatMulWorkload_Construct"</span>,</div> |
| <div class="line"><a name="l00098"></a><span class="lineno"> 98</span>  descriptor.<a class="code" href="structarmnn_1_1_queue_descriptor_with_parameters.html#aad91b9bbf7aa365d304febe79a3d1333">m_Parameters</a>,</div> |
| <div class="line"><a name="l00099"></a><span class="lineno"> 99</span>  detailsInfo,</div> |
| <div class="line"><a name="l00100"></a><span class="lineno"> 100</span>  <a class="code" href="classarmnn_1_1_base_workload.html#aaff95a48875d8fb4a616352906660ca9">GetGuid</a>());</div> |
| <div class="line"><a name="l00101"></a><span class="lineno"> 101</span> }</div> |
| <div class="line"><a name="l00102"></a><span class="lineno"> 102</span>  </div> |
| <div class="line"><a name="l00103"></a><span class="lineno"><a class="line" href="classarmnn_1_1_neon_batch_mat_mul_workload.html#ae071e8822437c78baea75c3aef3a263a"> 103</a></span> <span class="keywordtype">void</span> <a class="code" href="classarmnn_1_1_neon_batch_mat_mul_workload.html#ae071e8822437c78baea75c3aef3a263a">NeonBatchMatMulWorkload::Execute</a>()<span class="keyword"> const</span></div> |
| <div class="line"><a name="l00104"></a><span class="lineno"> 104</span> <span class="keyword"></span>{</div> |
| <div class="line"><a name="l00105"></a><span class="lineno"> 105</span>  <a class="code" href="_neon_workload_utils_8hpp.html#a7f97eedf3c9436b110df92c947bbb55d">ARMNN_SCOPED_PROFILING_EVENT_NEON_NAME_GUID</a>(<span class="stringliteral">"NeonBatchMatMulWorkload_Execute"</span>);</div> |
| <div class="line"><a name="l00106"></a><span class="lineno"> 106</span>  m_MatMulLayer.run();</div> |
| <div class="line"><a name="l00107"></a><span class="lineno"> 107</span> }</div> |
| <div class="line"><a name="l00108"></a><span class="lineno"> 108</span> } <span class="comment">//namespace armnn</span></div> |
| </div><!-- fragment --></div><!-- contents --> |
| </div><!-- doc-content --> |
| <div class="ttc" id="astructarmnn_1_1_batch_mat_mul_queue_descriptor_html"><div class="ttname"><a href="structarmnn_1_1_batch_mat_mul_queue_descriptor.html">armnn::BatchMatMulQueueDescriptor</a></div><div class="ttdef"><b>Definition:</b> <a href="_workload_data_8hpp_source.html#l00753">WorkloadData.hpp:753</a></div></div> |
| <div class="ttc" id="astructarmnn_1_1_activation_descriptor_html"><div class="ttname"><a href="structarmnn_1_1_activation_descriptor.html">armnn::ActivationDescriptor</a></div><div class="ttdoc">An ActivationDescriptor for the ActivationLayer.</div><div class="ttdef"><b>Definition:</b> <a href="_descriptors_8hpp_source.html#l00036">Descriptors.hpp:36</a></div></div> |
| <div class="ttc" id="astructarmnn_1_1_batch_mat_mul_descriptor_html_acb441bb8db19bcce78d15cdd8ceb5ea0"><div class="ttname"><a href="structarmnn_1_1_batch_mat_mul_descriptor.html#acb441bb8db19bcce78d15cdd8ceb5ea0">armnn::BatchMatMulDescriptor::m_TransposeX</a></div><div class="ttdeci">bool m_TransposeX</div><div class="ttdoc">Transpose the slices of each input tensor Transpose and Adjoint can not both be set to true for the s...</div><div class="ttdef"><b>Definition:</b> <a href="_descriptors_8hpp_source.html#l01612">Descriptors.hpp:1612</a></div></div> |
| <div class="ttc" id="astructarmnn_1_1_batch_mat_mul_descriptor_html_a0cf8306be7d301de0f095fff9901a525"><div class="ttname"><a href="structarmnn_1_1_batch_mat_mul_descriptor.html#a0cf8306be7d301de0f095fff9901a525">armnn::BatchMatMulDescriptor::m_AdjointX</a></div><div class="ttdeci">bool m_AdjointX</div><div class="ttdoc">Adjoint the slices of each input tensor Transpose and Adjoint can not both be set to true for the sam...</div><div class="ttdef"><b>Definition:</b> <a href="_descriptors_8hpp_source.html#l01617">Descriptors.hpp:1617</a></div></div> |
| <div class="ttc" id="a_workload_utils_8hpp_html"><div class="ttname"><a href="_workload_utils_8hpp.html">WorkloadUtils.hpp</a></div></div> |
| <div class="ttc" id="astructarmnn_1_1_queue_descriptor_html_a765d2cee4ccce5b9467e0c2b6d25b84a"><div class="ttname"><a href="structarmnn_1_1_queue_descriptor.html#a765d2cee4ccce5b9467e0c2b6d25b84a">armnn::QueueDescriptor::ValidateInputsOutputs</a></div><div class="ttdeci">void ValidateInputsOutputs(const std::string &descName, unsigned int numExpectedIn, unsigned int numExpectedOut) const</div><div class="ttdef"><b>Definition:</b> <a href="_workload_data_8cpp_source.html#l00447">WorkloadData.cpp:447</a></div></div> |
| <div class="ttc" id="aclassarmnn_1_1_neon_batch_mat_mul_workload_html_adf0e0cd06d407af2ae72ac3613f01a0e"><div class="ttname"><a href="classarmnn_1_1_neon_batch_mat_mul_workload.html#adf0e0cd06d407af2ae72ac3613f01a0e">armnn::NeonBatchMatMulWorkload::NeonBatchMatMulWorkload</a></div><div class="ttdeci">NeonBatchMatMulWorkload(const BatchMatMulQueueDescriptor &descriptor, const WorkloadInfo &info, const bool isFastMathEnabled)</div><div class="ttdef"><b>Definition:</b> <a href="_neon_batch_mat_mul_workload_8cpp_source.html#l00057">NeonBatchMatMulWorkload.cpp:57</a></div></div> |
| <div class="ttc" id="anamespacearmnn_html_abfb0841058a8190d30851f07eca3991f"><div class="ttname"><a href="namespacearmnn.html#abfb0841058a8190d30851f07eca3991f">armnn::ConvertAdditionalInfoToAclActivationLayerInfo</a></div><div class="ttdeci">arm_compute::ActivationLayerInfo ConvertAdditionalInfoToAclActivationLayerInfo(const QueueDescriptor &queueDescriptor)</div><div class="ttdef"><b>Definition:</b> <a href="_arm_compute_utils_8hpp_source.html#l00105">ArmComputeUtils.hpp:105</a></div></div> |
| <div class="ttc" id="astructarmnn_1_1_batch_mat_mul_descriptor_html_aedca000a005e091c23191e82d7e81b1d"><div class="ttname"><a href="structarmnn_1_1_batch_mat_mul_descriptor.html#aedca000a005e091c23191e82d7e81b1d">armnn::BatchMatMulDescriptor::m_DataLayoutX</a></div><div class="ttdeci">DataLayout m_DataLayoutX</div><div class="ttdoc">Data layout of each input tensor, such as NHWC/NDHWC (leave as default for arbitrary layout)</div><div class="ttdef"><b>Definition:</b> <a href="_descriptors_8hpp_source.html#l01621">Descriptors.hpp:1621</a></div></div> |
| <div class="ttc" id="aclassarmnn_1_1_tensor_info_html"><div class="ttname"><a href="classarmnn_1_1_tensor_info.html">armnn::TensorInfo</a></div><div class="ttdef"><b>Definition:</b> <a href="_tensor_8hpp_source.html#l00152">Tensor.hpp:152</a></div></div> |
| <div class="ttc" id="astructarmnn_1_1_batch_mat_mul_descriptor_html_ad945fc98770356dd886a68e98a52e26b"><div class="ttname"><a href="structarmnn_1_1_batch_mat_mul_descriptor.html#ad945fc98770356dd886a68e98a52e26b">armnn::BatchMatMulDescriptor::m_AdjointY</a></div><div class="ttdeci">bool m_AdjointY</div><div class="ttdef"><b>Definition:</b> <a href="_descriptors_8hpp_source.html#l01618">Descriptors.hpp:1618</a></div></div> |
| <div class="ttc" id="anamespacearmnn_html_a52cdeda888e3ecf0dbebb533ba0856f1"><div class="ttname"><a href="namespacearmnn.html#a52cdeda888e3ecf0dbebb533ba0856f1">armnn::NeonBatchMatMulValidate</a></div><div class="ttdeci">arm_compute::Status NeonBatchMatMulValidate(const TensorInfo &inputInfoX, const TensorInfo &inputInfoY, const TensorInfo &outputInfo, const BatchMatMulDescriptor &descriptor, const bool isFastMathEnabled, const ActivationDescriptor *activationDescriptor)</div><div class="ttdef"><b>Definition:</b> <a href="_neon_batch_mat_mul_workload_8cpp_source.html#l00019">NeonBatchMatMulWorkload.cpp:19</a></div></div> |
| <div class="ttc" id="astructarmnn_1_1_workload_info_html_a67b178f8a836bc1e52b8de109760adfd"><div class="ttname"><a href="structarmnn_1_1_workload_info.html#a67b178f8a836bc1e52b8de109760adfd">armnn::WorkloadInfo::m_OutputTensorInfos</a></div><div class="ttdeci">std::vector< TensorInfo > m_OutputTensorInfos</div><div class="ttdef"><b>Definition:</b> <a href="_workload_info_8hpp_source.html#l00019">WorkloadInfo.hpp:19</a></div></div> |
| <div class="ttc" id="astructarmnn_1_1_queue_descriptor_with_parameters_html_aad91b9bbf7aa365d304febe79a3d1333"><div class="ttname"><a href="structarmnn_1_1_queue_descriptor_with_parameters.html#aad91b9bbf7aa365d304febe79a3d1333">armnn::QueueDescriptorWithParameters::m_Parameters</a></div><div class="ttdeci">LayerDescriptor m_Parameters</div><div class="ttdef"><b>Definition:</b> <a href="_workload_data_8hpp_source.html#l00066">WorkloadData.hpp:66</a></div></div> |
| <div class="ttc" id="astructarmnn_1_1_batch_mat_mul_descriptor_html_a112b466e5d2ab9d1887178adbe3afa1c"><div class="ttname"><a href="structarmnn_1_1_batch_mat_mul_descriptor.html#a112b466e5d2ab9d1887178adbe3afa1c">armnn::BatchMatMulDescriptor::m_TransposeY</a></div><div class="ttdeci">bool m_TransposeY</div><div class="ttdef"><b>Definition:</b> <a href="_descriptors_8hpp_source.html#l01613">Descriptors.hpp:1613</a></div></div> |
| <div class="ttc" id="astructarmnn_1_1_workload_info_html"><div class="ttname"><a href="structarmnn_1_1_workload_info.html">armnn::WorkloadInfo</a></div><div class="ttdoc">Contains information about TensorInfos of a layer.</div><div class="ttdef"><b>Definition:</b> <a href="_workload_info_8hpp_source.html#l00016">WorkloadInfo.hpp:16</a></div></div> |
| <div class="ttc" id="a_polymorphic_downcast_8hpp_html"><div class="ttname"><a href="_polymorphic_downcast_8hpp.html">PolymorphicDowncast.hpp</a></div></div> |
| <div class="ttc" id="astructarmnn_1_1_batch_mat_mul_descriptor_html_aaf7828880989b4b9378d3e86aa6dc843"><div class="ttname"><a href="structarmnn_1_1_batch_mat_mul_descriptor.html#aaf7828880989b4b9378d3e86aa6dc843">armnn::BatchMatMulDescriptor::m_DataLayoutY</a></div><div class="ttdeci">DataLayout m_DataLayoutY</div><div class="ttdef"><b>Definition:</b> <a href="_descriptors_8hpp_source.html#l01622">Descriptors.hpp:1622</a></div></div> |
| <div class="ttc" id="astructarmnn_1_1_batch_mat_mul_descriptor_html"><div class="ttname"><a href="structarmnn_1_1_batch_mat_mul_descriptor.html">armnn::BatchMatMulDescriptor</a></div><div class="ttdoc">A BatchMatMulDescriptor for the BatchMatMul operator.</div><div class="ttdef"><b>Definition:</b> <a href="_descriptors_8hpp_source.html#l01584">Descriptors.hpp:1584</a></div></div> |
| <div class="ttc" id="aclassarmnn_1_1_neon_batch_mat_mul_workload_html_ae071e8822437c78baea75c3aef3a263a"><div class="ttname"><a href="classarmnn_1_1_neon_batch_mat_mul_workload.html#ae071e8822437c78baea75c3aef3a263a">armnn::NeonBatchMatMulWorkload::Execute</a></div><div class="ttdeci">virtual void Execute() const override</div><div class="ttdef"><b>Definition:</b> <a href="_neon_batch_mat_mul_workload_8cpp_source.html#l00103">NeonBatchMatMulWorkload.cpp:103</a></div></div> |
| <div class="ttc" id="a_arm_compute_utils_8hpp_html"><div class="ttname"><a href="_arm_compute_utils_8hpp.html">ArmComputeUtils.hpp</a></div></div> |
| <div class="ttc" id="aclassarmnn_1_1_exception_html"><div class="ttname"><a href="classarmnn_1_1_exception.html">armnn::Exception</a></div><div class="ttdoc">Base class for all ArmNN exceptions so that users can filter to just those.</div><div class="ttdef"><b>Definition:</b> <a href="_exceptions_8hpp_source.html#l00046">Exceptions.hpp:46</a></div></div> |
| <div class="ttc" id="anamespacearmnn_html_ad701d0d29baa4266ab4d33b090aa661c"><div class="ttname"><a href="namespacearmnn.html#ad701d0d29baa4266ab4d33b090aa661c">armnn::ConvertActivationDescriptorToAclActivationLayerInfo</a></div><div class="ttdeci">arm_compute::ActivationLayerInfo ConvertActivationDescriptorToAclActivationLayerInfo(const ActivationDescriptor &actDesc)</div><div class="ttdef"><b>Definition:</b> <a href="_arm_compute_utils_8hpp_source.html#l00087">ArmComputeUtils.hpp:87</a></div></div> |
| <div class="ttc" id="anamespacearmnn_html_a4dc0adc6737b5944e7671bee71788407acaf9b6b99962bf5c2264824231d7a40c"><div class="ttname"><a href="namespacearmnn.html#a4dc0adc6737b5944e7671bee71788407acaf9b6b99962bf5c2264824231d7a40c">armnn::BoostLogSeverityMapping::info</a></div><div class="ttdeci">@ info</div></div> |
| <div class="ttc" id="astructarmnn_1_1_queue_descriptor_html_a6abd491bb99ffe88bd472c1ae5a1ed1a"><div class="ttname"><a href="structarmnn_1_1_queue_descriptor.html#a6abd491bb99ffe88bd472c1ae5a1ed1a">armnn::QueueDescriptor::m_Outputs</a></div><div class="ttdeci">std::vector< ITensorHandle * > m_Outputs</div><div class="ttdef"><b>Definition:</b> <a href="_workload_data_8hpp_source.html#l00027">WorkloadData.hpp:27</a></div></div> |
| <div class="ttc" id="a_profiling_8hpp_html_a786492a3881a4c760ab1eec2149f4aba"><div class="ttname"><a href="_profiling_8hpp.html#a786492a3881a4c760ab1eec2149f4aba">ARMNN_REPORT_PROFILING_WORKLOAD_DESC</a></div><div class="ttdeci">#define ARMNN_REPORT_PROFILING_WORKLOAD_DESC(name, desc, infos, guid)</div><div class="ttdef"><b>Definition:</b> <a href="_profiling_8hpp_source.html#l00227">Profiling.hpp:227</a></div></div> |
| <div class="ttc" id="aclassarmnn_1_1_base_workload_html_aaff95a48875d8fb4a616352906660ca9"><div class="ttname"><a href="classarmnn_1_1_base_workload.html#aaff95a48875d8fb4a616352906660ca9">armnn::BaseWorkload< BatchMatMulQueueDescriptor >::GetGuid</a></div><div class="ttdeci">arm::pipe::ProfilingGuid GetGuid() const final</div><div class="ttdef"><b>Definition:</b> <a href="_workload_8hpp_source.html#l00067">Workload.hpp:67</a></div></div> |
| <div class="ttc" id="anamespacearmnn_html_a67a0db04d321a74b7e7fcfd3f1a3f70b"><div class="ttname"><a href="namespacearmnn.html#a67a0db04d321a74b7e7fcfd3f1a3f70b">armnn::Status</a></div><div class="ttdeci">Status</div><div class="ttdef"><b>Definition:</b> <a href="_types_8hpp_source.html#l00042">Types.hpp:42</a></div></div> |
| <div class="ttc" id="aclassarmnn_1_1_base_workload_html_afb8d2c8817c75de9d01a4c0e0d5c160b"><div class="ttname"><a href="classarmnn_1_1_base_workload.html#afb8d2c8817c75de9d01a4c0e0d5c160b">armnn::BaseWorkload< BatchMatMulQueueDescriptor >::m_Data</a></div><div class="ttdeci">BatchMatMulQueueDescriptor m_Data</div><div class="ttdef"><b>Definition:</b> <a href="_workload_8hpp_source.html#l00089">Workload.hpp:89</a></div></div> |
| <div class="ttc" id="astructarmnn_1_1_workload_info_html_ac97905bfa0daab357b91df1347600309"><div class="ttname"><a href="structarmnn_1_1_workload_info.html#ac97905bfa0daab357b91df1347600309">armnn::WorkloadInfo::m_InputTensorInfos</a></div><div class="ttdeci">std::vector< TensorInfo > m_InputTensorInfos</div><div class="ttdef"><b>Definition:</b> <a href="_workload_info_8hpp_source.html#l00018">WorkloadInfo.hpp:18</a></div></div> |
| <div class="ttc" id="a_neon_workload_utils_8hpp_html"><div class="ttname"><a href="_neon_workload_utils_8hpp.html">NeonWorkloadUtils.hpp</a></div></div> |
| <div class="ttc" id="a_neon_batch_mat_mul_workload_8hpp_html"><div class="ttname"><a href="_neon_batch_mat_mul_workload_8hpp.html">NeonBatchMatMulWorkload.hpp</a></div></div> |
| <div class="ttc" id="anamespacearmnn_html"><div class="ttname"><a href="namespacearmnn.html">armnn</a></div><div class="ttdoc">Copyright (c) 2021 ARM Limited and Contributors.</div><div class="ttdef"><b>Definition:</b> <a href="01__00__quick__start_8dox_source.html#l00006">01_00_quick_start.dox:6</a></div></div> |
| <div class="ttc" id="a_neon_workload_utils_8hpp_html_a7f97eedf3c9436b110df92c947bbb55d"><div class="ttname"><a href="_neon_workload_utils_8hpp.html#a7f97eedf3c9436b110df92c947bbb55d">ARMNN_SCOPED_PROFILING_EVENT_NEON_NAME_GUID</a></div><div class="ttdeci">#define ARMNN_SCOPED_PROFILING_EVENT_NEON_NAME_GUID(label)</div><div class="ttdoc">Creates a profiling event that uses GetGuid() and GetName() from the calling class.</div><div class="ttdef"><b>Definition:</b> <a href="_neon_workload_utils_8hpp_source.html#l00033">NeonWorkloadUtils.hpp:33</a></div></div> |
| <div class="ttc" id="aclassarmnn_1_1_neon_base_workload_html"><div class="ttname"><a href="classarmnn_1_1_neon_base_workload.html">armnn::NeonBaseWorkload</a></div><div class="ttdef"><b>Definition:</b> <a href="_neon_base_workload_8hpp_source.html#l00013">NeonBaseWorkload.hpp:13</a></div></div> |
| <div class="ttc" id="astructarmnn_1_1_queue_descriptor_html_a4b50e46a6810018f3edecfb68b2a76b3"><div class="ttname"><a href="structarmnn_1_1_queue_descriptor.html#a4b50e46a6810018f3edecfb68b2a76b3">armnn::QueueDescriptor::m_Inputs</a></div><div class="ttdeci">std::vector< ITensorHandle * > m_Inputs</div><div class="ttdef"><b>Definition:</b> <a href="_workload_data_8hpp_source.html#l00026">WorkloadData.hpp:26</a></div></div> |
| <div class="ttc" id="anamespacearmnn_html_ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f"><div class="ttname"><a href="namespacearmnn.html#ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f">armnn::DataLayout::NCHW</a></div><div class="ttdeci">@ NCHW</div></div> |
| <!-- start footer part --> |
| <div id="nav-path" class="navpath"><!-- id is needed for treeview function! --> |
| <ul> |
| <li class="navelem"><a class="el" href="dir_68267d1309a1af8e8297ef4c3efbcdba.html">src</a></li><li class="navelem"><a class="el" href="dir_0f3cdec46afbc61a1ded8e1687c9c9a0.html">backends</a></li><li class="navelem"><a class="el" href="dir_d86eb514662c7c08e168285f21d00ea1.html">neon</a></li><li class="navelem"><a class="el" href="dir_369c3c20501d0d10bd0354bf11c2f559.html">workloads</a></li><li class="navelem"><a class="el" href="_neon_batch_mat_mul_workload_8cpp.html">NeonBatchMatMulWorkload.cpp</a></li> |
| <li class="footer">Generated on Thu May 16 2024 09:31:47 for Arm NN by |
| <a href="http://www.doxygen.org/index.html"> |
| <img class="footer" src="doxygen.png" alt="doxygen"/></a> 1.8.17 </li> |
| </ul> |
| </div> |
| </body> |
| </html> |