mkldnn_fluid.html



<!DOCTYPE html>
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
<head>
  <meta charset="utf-8">
  
  <meta name="viewport" content="width=device-width, initial-scale=1.0">
  
  <title>Design Doc: Add MKLDNN Kernel in Fluid Operator &mdash; PaddlePaddle  文档</title>
  

    <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
  

        <link rel="index" title="索引"
              href="../../genindex.html"/>
        <link rel="search" title="搜索" href="../../search.html"/>
    <link rel="top" title="PaddlePaddle  文档" href="../../index.html"/> 
<script>
var _hmt = _hmt || [];
(function() {
  var hm = document.createElement("script");
  hm.src = "//hm.baidu.com/hm.js?b9a314ab40d04d805655aab1deee08ba";
  var s = document.getElementsByTagName("script")[0]; 
  s.parentNode.insertBefore(hm, s);
})();
</script>


  <script src="../../_static/js/modernizr.min.js"></script>

</head>

<body class="wy-body-for-nav" role="document">

  <div class="wy-grid-for-nav">

    
    <nav data-toggle="wy-nav-shift" class="wy-nav-side">
      <div class="wy-side-scroll">
        <div class="wy-side-nav-search">
          

            <a href="../../index_cn.html" class="icon icon-home"> PaddlePaddle
          

          </a>

          
<div role="search">
  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
    <input type="text" name="q" placeholder="Search docs" />
    <input type="hidden" name="check_keywords" value="yes" />
    <input type="hidden" name="area" value="default" />
  </form>
</div>

          
        </div>

        <div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
          
<nav class="doc-menu-vertical" role="navigation">

<ul>
<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_cn.html">新手入门</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../getstarted/quickstart_cn.html">快速开始</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../getstarted/concepts/use_concepts_cn.html">基本使用概念</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../build_and_install/index_cn.html">安装与编译</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../build_and_install/pip_install_cn.html">使用pip安装</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../build_and_install/docker_install_cn.html">使用Docker安装运行</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../build_and_install/build_from_source_cn.html">从源码编译</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../howto/index_cn.html">进阶使用</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../howto/cmd_parameter/index_cn.html">命令行参数设置</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../howto/cmd_parameter/use_case_cn.html">使用案例</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/cmd_parameter/arguments_cn.html">参数概述</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/cmd_parameter/detail_introduction_cn.html">细节描述</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/cluster/index_cn.html">分布式训练</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../howto/cluster/preparations_cn.html">环境准备</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/cluster/cmd_argument_cn.html">启动参数说明</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/cluster/multi_cluster/index_cn.html">在不同集群中运行</a><ul>
<li class="toctree-l4"><a class="reference internal" href="../../howto/cluster/multi_cluster/k8s_cn.html">Kubernetes单机训练</a></li>
<li class="toctree-l4"><a class="reference internal" href="../../howto/cluster/multi_cluster/k8s_distributed_cn.html">Kubernetes分布式训练</a></li>
<li class="toctree-l4"><a class="reference internal" href="../../howto/cluster/multi_cluster/openmpi_cn.html">在OpenMPI集群中启动训练</a></li>
<li class="toctree-l4"><a class="reference internal" href="../../howto/cluster/multi_cluster/fabric_cn.html">使用fabric启动集群训练</a></li>
<li class="toctree-l4"><a class="reference internal" href="../../howto/cluster/multi_cluster/k8s_aws_cn.html">Kubernetes on AWS</a></li>
</ul>
</li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/capi/index_cn.html">C-API预测库</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../howto/capi/compile_paddle_lib_cn.html">安装与编译C-API预测库</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/capi/organization_of_the_inputs_cn.html">输入/输出数据组织</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/capi/workflow_of_capi_cn.html">C-API使用流程</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/rnn/index_cn.html">RNN模型</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../howto/rnn/rnn_config_cn.html">RNN配置</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/rnn/recurrent_group_cn.html">Recurrent Group教程</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/rnn/hierarchical_layer_cn.html">支持双层序列作为输入的Layer</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/rnn/hrnn_rnn_api_compare_cn.html">单双层RNN API对比介绍</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/optimization/gpu_profiling_cn.html">GPU性能调优</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../dev/index_cn.html">开发标准</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../dev/contribute_to_paddle_cn.html">如何贡献代码</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../dev/write_docs_cn.html">如何贡献文档</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
</ul>
</li>
</ul>

</nav>

        </div>
      </div>
    </nav>

    <section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">

      
      <nav class="wy-nav-top" role="navigation" aria-label="top navigation">
        <i data-toggle="wy-nav-top" class="fa fa-bars"></i>
        <a href="../../index_cn.html">PaddlePaddle</a>
      </nav>


      <div class="wy-nav-content">
        <div class="rst-content">
          

<div role="navigation" aria-label="breadcrumbs navigation">
  <ul class="wy-breadcrumbs">
    <li><a href="../../index_cn.html">Docs</a> &raquo;</li>
      
    <li>Design Doc: Add MKLDNN Kernel in Fluid Operator</li>
      <li class="wy-breadcrumbs-aside">
        
          
            <a href="../../_sources/design/mkl/mkldnn_fluid.md.txt" rel="nofollow"> View page source</a>
          
        
      </li>
  </ul>
  <hr/>
</div>
          <div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
           <div itemprop="articleBody">
            
  <div class="section" id="design-doc-add-mkldnn-kernel-in-fluid-operator">
<span id="design-doc-add-mkldnn-kernel-in-fluid-operator"></span><h1>Design Doc: Add MKLDNN Kernel in Fluid Operator<a class="headerlink" href="#design-doc-add-mkldnn-kernel-in-fluid-operator" title="永久链接至标题">¶</a></h1>
<div class="section" id="principles">
<span id="principles"></span><h2>Principles<a class="headerlink" href="#principles" title="永久链接至标题">¶</a></h2>
<p>First of all, we should follow some basical principles like:</p>
<ol class="simple">
<li><a class="reference external" href="https://github.com/PaddlePaddle/Paddle/blob/develop/doc/howto/dev/new_op_en.md">How to write a new operator</a>. We are trying to add a new kind of kernel into operators, so basically we should follow this doc.</li>
<li><a class="reference external" href="https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/support_new_device.md">Supporting new Device/Library</a>. Since MKLDNN is a new library to fluid, we should add <code class="docutils literal"><span class="pre">MKLDNNDeviceContext</span></code> and maybe <code class="docutils literal"><span class="pre">mkldnn_helper.h</span></code>, just like <a class="reference external" href="https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/platform/cudnn_helper.h">cudnn_helper.h</a>.</li>
<li><a class="reference external" href="https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/switch_kernel.md">Switch Kernel</a>. Another important point is that we should ensure the data synchronization between different kernel types, which is this <a class="reference external" href="https://github.com/PaddlePaddle/Paddle/issues/6549">topic</a>. So basically we should override <code class="docutils literal"><span class="pre">GetExpectedKernelType</span></code> and <code class="docutils literal"><span class="pre">trans</span></code> functions to support switching kernels.</li>
<li><a class="reference external" href="https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/operator_kernel_type.md">The Keys of Operator Kernel Type</a>. Kernel Type is a pivotal conception which can record the <code class="docutils literal"><span class="pre">Place</span></code>, <code class="docutils literal"><span class="pre">Library</span></code>, <code class="docutils literal"><span class="pre">DataType</span></code> and <code class="docutils literal"><span class="pre">Layout</span></code>.</li>
</ol>
</div>
<div class="section" id="sulution">
<span id="sulution"></span><h2>Sulution<a class="headerlink" href="#sulution" title="永久链接至标题">¶</a></h2>
<p>In general, there are four parts we should follow to run a MKL-DNN primitive.</p>
<ul class="simple">
<li>Create a primitive descriptor that describe this operator</li>
<li>Create a primitive itself by primitive descriptor and the engine</li>
<li>Create all memory buffers that primitive needed</li>
<li>Launch a stream to execute the primitive created
More details can refer to <a class="reference external" href="http://01org.github.io/mkl-dnn">here</a>.</li>
</ul>
<p>It&#8217;s better to avoid reinitialization of primitives and memory handles in the first three stages in every iteration. So we plan to create a map to record all the <code class="docutils literal"><span class="pre">primitive</span></code> and <code class="docutils literal"><span class="pre">memory</span></code>, which should not take too much memories as discussed <a class="reference external" href="https://github.com/PaddlePaddle/Paddle/issues/6822">here</a>.</p>
<p>It&#8217;s assumed that following three conditions should be satisfied.</p>
<ol class="simple">
<li>there is a unique key for each operator instance. May be the actual name of <code class="docutils literal"><span class="pre">Output</span> <span class="pre">Tensor</span></code>.</li>
<li>the <code class="docutils literal"><span class="pre">Input</span> <span class="pre">Tensor</span></code> inside <code class="docutils literal"><span class="pre">Compute</span></code> function is the one after converted.</li>
<li>we can get the phase(eg. <code class="docutils literal"><span class="pre">is_test</span></code>) inside <code class="docutils literal"><span class="pre">Compute</span></code> function, otherwise we need to expose this attribue to user.</li>
</ol>
<div class="section" id="compute">
<span id="compute"></span><h3>Compute<a class="headerlink" href="#compute" title="永久链接至标题">¶</a></h3>
<p>The algorithm of <code class="docutils literal"><span class="pre">Compute</span></code> would be described as follow, let&#8217;s take conv like an example.</p>
<div class="highlight-c++"><div class="highlight"><pre><span></span>  <span class="n">PADDLE_ENFORCE</span><span class="p">(</span><span class="n">platform</span><span class="o">::</span><span class="n">is_cpu_place</span><span class="p">(</span><span class="n">ctx</span><span class="p">.</span><span class="n">GetPlace</span><span class="p">()),</span> <span class="s">&quot;It must use CPUPlace.&quot;</span><span class="p">);</span>
  <span class="n">PADDLE_ENFORCE</span><span class="p">(</span><span class="n">platform</span><span class="o">::</span><span class="n">is_mkldnn_library</span><span class="p">(</span><span class="n">ctx</span><span class="p">.</span><span class="n">GetLibrary</span><span class="p">()),</span> <span class="s">&quot;It must use MKLDNN Library.&quot;</span><span class="p">);</span>

  <span class="k">auto</span><span class="o">&amp;</span> <span class="n">dev_ctx</span> <span class="o">=</span> <span class="n">ctx</span><span class="p">.</span><span class="k">template</span> <span class="n">device_context</span><span class="o">&lt;</span><span class="n">platform</span><span class="o">::</span><span class="n">MKLDNNDeviceContext</span><span class="o">&gt;</span><span class="p">();</span>

  <span class="c1">// find primitive by unique key from mkldnn context</span>
  <span class="c1">// the op_key should be a unique name of this op instance</span>
  <span class="k">auto</span><span class="o">&amp;</span> <span class="n">p</span> <span class="o">=</span> <span class="n">dev_ctx</span><span class="p">.</span><span class="n">findPrimitive</span><span class="p">(</span><span class="n">op_key</span> <span class="o">+</span> <span class="s">&quot;_fwd&quot;</span><span class="p">);</span>

  <span class="c1">// assuming the input tensor inside this compute function is the one after converted</span>
  <span class="c1">// this point should be guarantee by another mechanism</span>
  <span class="k">auto</span><span class="o">&amp;</span> <span class="n">i</span> <span class="o">=</span> <span class="n">dev_ctx</span><span class="p">.</span><span class="n">findMemory</span><span class="p">(</span><span class="n">op_key</span> <span class="o">+</span> <span class="s">&quot;_input&quot;</span><span class="p">);</span>
  
  <span class="k">if</span> <span class="p">(</span><span class="n">p</span> <span class="o">==</span> <span class="k">nullptr</span> <span class="o">||</span> <span class="n">i</span> <span class="o">==</span> <span class="k">nullptr</span> <span class="o">||</span> <span class="n">inputSizeChanged</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">i</span><span class="p">))</span>  <span class="p">{</span>
    <span class="k">auto</span> <span class="n">fwd_primitive_desc</span> <span class="o">=</span> <span class="n">createPrimitiveDesc</span><span class="p">(</span><span class="n">ctx</span><span class="p">);</span>
    <span class="k">auto</span><span class="o">*</span> <span class="n">input</span> <span class="o">=</span> <span class="n">ctx</span><span class="p">.</span><span class="n">Input</span><span class="o">&lt;</span><span class="n">Tensor</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;Input&quot;</span><span class="p">);</span>
    <span class="k">auto</span><span class="o">*</span> <span class="n">filter</span> <span class="o">=</span> <span class="n">ctx</span><span class="p">.</span><span class="n">Input</span><span class="o">&lt;</span><span class="n">Tensor</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;Filter&quot;</span><span class="p">);</span>
    <span class="k">auto</span><span class="o">*</span> <span class="n">output</span> <span class="o">=</span> <span class="n">ctx</span><span class="p">.</span><span class="n">Output</span><span class="o">&lt;</span><span class="n">Tensor</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;Output&quot;</span><span class="p">);</span>
    <span class="n">shared_ptr</span><span class="o">&lt;</span><span class="n">mkldnn</span><span class="o">::</span><span class="n">memory</span><span class="o">&gt;</span> <span class="n">in</span><span class="p">(</span><span class="k">new</span> <span class="n">mkldnn</span><span class="o">::</span><span class="n">memory</span><span class="p">(</span><span class="n">fwd_primitive_desc</span><span class="o">-&gt;</span><span class="n">src_primitive_desc</span><span class="p">(),</span> <span class="n">input</span><span class="o">-&gt;</span><span class="n">data</span><span class="o">&lt;</span><span class="n">T</span><span class="o">&gt;</span><span class="p">()));</span>
    <span class="n">shared_ptr</span><span class="o">&lt;</span><span class="n">mkldnn</span><span class="o">::</span><span class="n">memory</span><span class="o">&gt;</span> <span class="n">wgt</span><span class="p">(</span><span class="k">new</span> <span class="n">mkldnn</span><span class="o">::</span><span class="n">memory</span><span class="p">(</span><span class="n">fwd_primitive_desc</span><span class="o">-&gt;</span><span class="n">weights_primitive_desc</span><span class="p">(),</span> <span class="n">filter</span><span class="o">-&gt;</span><span class="n">data</span><span class="o">&lt;</span><span class="n">T</span><span class="o">&gt;</span><span class="p">()));</span>
    <span class="n">shared_ptr</span><span class="o">&lt;</span><span class="n">mkldnn</span><span class="o">::</span><span class="n">memory</span><span class="o">&gt;</span> <span class="n">out</span><span class="p">(</span><span class="k">new</span> <span class="n">mkldnn</span><span class="o">::</span><span class="n">memory</span><span class="p">(</span><span class="n">fwd_primitive_desc</span><span class="o">-&gt;</span><span class="n">dst_primitive_desc</span><span class="p">(),</span> <span class="n">output</span><span class="o">-&gt;</span><span class="n">mutable_data</span><span class="o">&lt;</span><span class="n">T</span><span class="o">&gt;</span><span class="p">(</span><span class="n">ctx</span><span class="p">.</span><span class="n">GetPlace</span><span class="p">())));</span>
    <span class="n">shared_ptr</span><span class="o">&lt;</span><span class="n">mkldnn</span><span class="o">::</span><span class="n">conv_fwd</span><span class="o">&gt;</span> <span class="n">fwd_primitive</span><span class="p">(</span><span class="k">new</span> <span class="n">mkldnn</span><span class="o">::</span><span class="n">conv_fwd</span><span class="p">(</span><span class="o">*</span><span class="n">fwd_primitive_desc</span><span class="p">,</span> <span class="o">*</span><span class="n">in</span><span class="p">,</span> <span class="o">*</span><span class="n">wgt</span><span class="p">,</span> <span class="o">*</span><span class="n">out</span><span class="p">));</span>

    <span class="n">dev_ctx</span><span class="p">.</span><span class="n">addMemory</span><span class="p">(</span><span class="n">op_key</span><span class="o">+</span><span class="s">&quot;_input&quot;</span><span class="p">,</span> <span class="n">in</span><span class="p">);</span>
    <span class="n">dev_ctx</span><span class="p">.</span><span class="n">addMemory</span><span class="p">(</span><span class="n">op_key</span><span class="o">+</span><span class="s">&quot;_output&quot;</span><span class="p">,</span> <span class="n">out</span><span class="p">);</span>
    <span class="n">dev_ctx</span><span class="p">.</span><span class="n">addMemory</span><span class="p">(</span><span class="n">op_key</span><span class="o">+</span><span class="s">&quot;_filer&quot;</span><span class="p">,</span> <span class="n">wgt</span><span class="p">);</span>
    <span class="n">dev_ctx</span><span class="p">.</span><span class="n">addPrimitive</span><span class="p">(</span><span class="n">op_key</span><span class="o">+</span><span class="s">&quot;_fwd&quot;</span><span class="p">,</span> <span class="n">fwd_primitive</span><span class="p">);</span>
    <span class="n">dev_ctx</span><span class="p">.</span><span class="n">addPrimitiveDesc</span><span class="p">(</span><span class="n">op_key</span><span class="o">+</span><span class="s">&quot;_fwd_PD&quot;</span><span class="p">,</span> <span class="n">fwd_primitive_desc</span><span class="p">);</span>
  <span class="p">}</span>

  <span class="n">p</span> <span class="o">=</span> <span class="n">dev_ctx</span><span class="p">.</span><span class="n">findPrimitive</span><span class="p">(</span><span class="n">op_key</span> <span class="o">+</span> <span class="s">&quot;_fwd&quot;</span><span class="p">);</span>

  <span class="n">PADDLE_ENFORCE</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="s">&quot;Should have forward Primitive&quot;</span><span class="p">);</span>
  <span class="n">PADDLE_ENFORCE</span><span class="p">(</span><span class="n">dev_ctx</span><span class="p">.</span><span class="n">findMemory</span><span class="p">(</span><span class="n">op_unique_key</span><span class="o">+</span><span class="s">&quot;_input&quot;</span><span class="p">),</span> <span class="s">&quot;Should have input memory&quot;</span><span class="p">);</span>
  <span class="n">PADDLE_ENFORCE</span><span class="p">(</span><span class="n">dev_ctx</span><span class="p">.</span><span class="n">findMemory</span><span class="p">(</span><span class="n">op_unique_key</span><span class="o">+</span><span class="s">&quot;_output&quot;</span><span class="p">),</span> <span class="s">&quot;Should have output memory&quot;</span><span class="p">);</span>
  <span class="n">PADDLE_ENFORCE</span><span class="p">(</span><span class="n">dev_ctx</span><span class="p">.</span><span class="n">findMemory</span><span class="p">(</span><span class="n">op_unique_key</span><span class="o">+</span><span class="s">&quot;_filter&quot;</span><span class="p">),</span> <span class="s">&quot;Should have filter memory&quot;</span><span class="p">);</span>
  <span class="n">PADDLE_ENFORCE</span><span class="p">(</span><span class="n">dev_ctx</span><span class="p">.</span><span class="n">findPrimitiveDesc</span><span class="p">(</span><span class="n">op_unique_key</span><span class="o">+</span><span class="s">&quot;_fwd_PD&quot;</span><span class="p">),</span> <span class="s">&quot;Should have forward PrimitiveDesc&quot;</span><span class="p">);</span>
  <span class="n">dev_ctx</span><span class="p">.</span><span class="n">submit</span><span class="p">(</span><span class="n">p</span><span class="p">);</span>
  <span class="n">dev_ctx</span><span class="p">.</span><span class="n">execute</span><span class="p">();</span>  <span class="c1">// the convert primitive should have already contained.</span>
</pre></div>
</div>
<p>The <code class="docutils literal"><span class="pre">createPrimitiveDesc</span></code> returns the primitive descripotor of this operator, would be like this:</p>
<div class="highlight-c++"><div class="highlight"><pre><span></span>  <span class="k">auto</span><span class="o">*</span> <span class="n">input</span> <span class="o">=</span> <span class="n">ctx</span><span class="p">.</span><span class="n">Input</span><span class="o">&lt;</span><span class="n">Tensor</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;Input&quot;</span><span class="p">);</span>
  <span class="k">auto</span><span class="o">*</span> <span class="n">filter</span> <span class="o">=</span> <span class="n">ctx</span><span class="p">.</span><span class="n">Input</span><span class="o">&lt;</span><span class="n">Tensor</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;Filter&quot;</span><span class="p">);</span>
  <span class="k">auto</span><span class="o">*</span> <span class="n">output</span> <span class="o">=</span> <span class="n">ctx</span><span class="p">.</span><span class="n">Output</span><span class="o">&lt;</span><span class="n">Tensor</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;Output&quot;</span><span class="p">);</span>
  <span class="n">std</span><span class="o">::</span><span class="n">vector</span><span class="o">&lt;</span><span class="kt">int</span><span class="o">&gt;</span> <span class="n">strides</span> <span class="o">=</span> <span class="n">ctx</span><span class="p">.</span><span class="n">Attr</span><span class="o">&lt;</span><span class="n">std</span><span class="o">::</span><span class="n">vector</span><span class="o">&lt;</span><span class="kt">int</span><span class="o">&gt;&gt;</span><span class="p">(</span><span class="s">&quot;strides&quot;</span><span class="p">);</span>
  <span class="n">std</span><span class="o">::</span><span class="n">vector</span><span class="o">&lt;</span><span class="kt">int</span><span class="o">&gt;</span> <span class="n">paddings</span> <span class="o">=</span> <span class="n">ctx</span><span class="p">.</span><span class="n">Attr</span><span class="o">&lt;</span><span class="n">std</span><span class="o">::</span><span class="n">vector</span><span class="o">&lt;</span><span class="kt">int</span><span class="o">&gt;&gt;</span><span class="p">(</span><span class="s">&quot;paddings&quot;</span><span class="p">);</span>
  <span class="n">std</span><span class="o">::</span><span class="n">vector</span><span class="o">&lt;</span><span class="kt">int</span><span class="o">&gt;</span> <span class="n">dilations</span> <span class="o">=</span> <span class="n">ctx</span><span class="p">.</span><span class="n">Attr</span><span class="o">&lt;</span><span class="n">std</span><span class="o">::</span><span class="n">vector</span><span class="o">&lt;</span><span class="kt">int</span><span class="o">&gt;&gt;</span><span class="p">(</span><span class="s">&quot;dilations&quot;</span><span class="p">);</span>
  <span class="kt">int</span> <span class="n">groups</span> <span class="o">=</span> <span class="n">ctx</span><span class="p">.</span><span class="n">Attr</span><span class="o">&lt;</span><span class="kt">int</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;groups&quot;</span><span class="p">);</span>
  <span class="n">algorithm</span> <span class="n">algo</span> <span class="o">=</span> <span class="k">static_cast</span><span class="o">&lt;</span><span class="n">algorithm</span><span class="o">&gt;</span><span class="p">(</span><span class="n">ctx</span><span class="p">.</span><span class="n">Attr</span><span class="o">&lt;</span><span class="kt">int</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;convolution_algorithm_option&quot;</span><span class="p">));</span>
  <span class="n">prop_kind</span> <span class="n">pk</span> <span class="o">=</span> <span class="n">ctx</span><span class="p">.</span><span class="n">Attr</span><span class="o">&lt;</span><span class="kt">bool</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;is_test&quot;</span><span class="p">)</span> <span class="o">?</span> <span class="n">prop_kind</span><span class="o">::</span><span class="nl">forward_inference</span> <span class="p">:</span> <span class="n">prop_kind</span><span class="o">::</span><span class="n">forward_training</span><span class="p">;</span>
    
  <span class="k">auto</span> <span class="n">fwd_desc</span> <span class="o">=</span> <span class="n">mkldnn</span><span class="o">::</span><span class="n">conv_fwd</span><span class="o">::</span><span class="n">desc</span><span class="p">(</span><span class="cm">/* all the setting above*/</span><span class="p">);</span>
  <span class="n">shared_ptr</span><span class="o">&lt;</span><span class="n">mkldnn</span><span class="o">::</span><span class="n">conv_fwd</span><span class="o">::</span><span class="n">primitive_desc</span><span class="o">&gt;</span> <span class="n">fwd_primitive_desc</span><span class="p">(</span><span class="k">new</span> <span class="n">mkldnn</span><span class="o">::</span><span class="n">conv_fwd</span><span class="o">::</span><span class="n">primitive_desc</span><span class="p">(</span><span class="n">fwd_desc</span><span class="p">,</span> <span class="n">ctx</span><span class="p">.</span><span class="n">getEngine</span><span class="p">()));</span>

  <span class="k">return</span> <span class="n">fwd_primitive_desc</span><span class="p">;</span>
  <span class="p">}</span>
</pre></div>
</div>
</div>
<div class="section" id="mkldnndevicecontext">
<span id="mkldnndevicecontext"></span><h3>MKLDNNDeviceContext<a class="headerlink" href="#mkldnndevicecontext" title="永久链接至标题">¶</a></h3>
<p><code class="docutils literal"><span class="pre">MKLDNNDeviceContext</span></code>, which is very straightforward, should contain some base information like: <code class="docutils literal"><span class="pre">stream</span></code>, <code class="docutils literal"><span class="pre">engine</span></code> and the map needed.</p>
</div>
<div class="section" id="mkldnn-helper">
<span id="mkldnn-helper"></span><h3>mkldnn_helper<a class="headerlink" href="#mkldnn-helper" title="永久链接至标题">¶</a></h3>
<p>Some functions would be put in <code class="docutils literal"><span class="pre">paddle/platform/mkldnn_helper.h</span></code>.</p>
<ul class="simple">
<li>create MKLDNN memories</li>
<li>create MKLDNN primitives</li>
<li>error check function</li>
<li>etc</li>
</ul>
</div>
<div class="section" id="kernel-switch">
<span id="kernel-switch"></span><h3>Kernel Switch<a class="headerlink" href="#kernel-switch" title="永久链接至标题">¶</a></h3>
<p>We should <code class="docutils literal"><span class="pre">reorder</span></code> the different Layout from other device or to other device. <code class="docutils literal"><span class="pre">GetExpectedKernelType</span></code> and <code class="docutils literal"><span class="pre">trans</span></code> functions can help us to implement it.</p>
<p><code class="docutils literal"><span class="pre">GetExpectedKernelType</span></code> should get the context, and this operator can return the best <code class="docutils literal"><span class="pre">KernelType</span></code>.
<code class="docutils literal"><span class="pre">trans</span></code> would be like this:</p>
<div class="highlight-c++"><div class="highlight"><pre><span></span><span class="kt">void</span> <span class="nf">trans</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">ctx</span><span class="p">)</span> <span class="k">override</span> <span class="p">{</span>
  <span class="k">if</span> <span class="p">(</span><span class="n">NoNeedTrans</span><span class="p">())</span> <span class="p">{</span>
    <span class="k">return</span><span class="p">;</span>
  <span class="p">}</span>
  <span class="c1">// find reorder primitive by op_key from context</span>
  <span class="k">auto</span><span class="o">&amp;</span> <span class="n">dev_ctx</span> <span class="o">=</span> <span class="n">ctx</span><span class="p">.</span><span class="k">template</span> <span class="n">device_context</span><span class="o">&lt;</span><span class="n">platform</span><span class="o">::</span><span class="n">MKLDNNDeviceContext</span><span class="o">&gt;</span><span class="p">();</span>
  <span class="k">auto</span><span class="o">&amp;</span> <span class="n">p</span> <span class="o">=</span> <span class="n">dev_ctx</span><span class="p">.</span><span class="n">findPrimitive</span><span class="p">(</span><span class="n">op_key</span> <span class="o">+</span> <span class="s">&quot;_reorder_input&quot;</span><span class="p">);</span>
  <span class="k">auto</span><span class="o">&amp;</span> <span class="n">i</span> <span class="o">=</span> <span class="n">dev_ctx</span><span class="p">.</span><span class="n">findMemory</span><span class="p">(</span><span class="n">op_key</span> <span class="o">+</span> <span class="s">&quot;_src_input&quot;</span><span class="p">);</span>

  <span class="k">if</span> <span class="p">(</span><span class="n">p</span> <span class="o">==</span> <span class="k">nullptr</span> <span class="o">||</span> <span class="n">i</span> <span class="o">==</span> <span class="k">nullptr</span> <span class="o">||</span> <span class="n">changeSized</span><span class="p">(</span><span class="n">i</span><span class="p">,</span> <span class="n">input</span><span class="p">))</span> <span class="p">{</span>
    <span class="k">auto</span> <span class="n">prim</span> <span class="o">=</span> <span class="n">createPrimitiveDesc</span><span class="p">(</span><span class="n">ctx</span><span class="p">);</span>
    <span class="k">auto</span> <span class="n">src</span> <span class="o">=</span> <span class="n">createMemory</span><span class="p">(</span><span class="n">memoryDesc</span><span class="p">(</span><span class="n">input</span><span class="o">-&gt;</span><span class="n">dims</span><span class="p">(),</span> <span class="n">actual_layout</span><span class="p">),</span> <span class="n">input</span><span class="o">-&gt;</span><span class="n">data</span><span class="p">);</span>
    <span class="k">auto</span> <span class="n">newbuffer</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">::</span><span class="n">memory</span><span class="o">::</span><span class="n">Alloc</span><span class="p">(</span><span class="n">ctx</span><span class="p">.</span><span class="n">GetPlace</span><span class="p">(),</span> <span class="n">input</span><span class="o">-&gt;</span><span class="n">size_in_bytes</span><span class="p">());</span>
    <span class="k">auto</span> <span class="n">dst</span> <span class="o">=</span> <span class="n">createMemory</span><span class="p">(</span><span class="n">p</span><span class="o">-&gt;</span><span class="n">expected_desc</span><span class="p">(),</span> <span class="n">newbuffer</span><span class="o">-&gt;</span><span class="n">data</span><span class="p">);</span>
    <span class="k">auto</span> <span class="n">reorder_primitive</span><span class="p">(</span><span class="k">new</span> <span class="n">mkldnn</span><span class="o">::</span><span class="n">reorder</span><span class="p">(</span><span class="n">src</span><span class="p">,</span> <span class="n">dst</span><span class="p">));</span>

    <span class="n">dev_ctx</span><span class="p">.</span><span class="n">addMemory</span><span class="p">(</span><span class="n">op_key</span><span class="o">+</span><span class="s">&quot;_src_input&quot;</span><span class="p">,</span> <span class="n">src</span><span class="p">);</span>
    <span class="n">dev_ctx</span><span class="p">.</span><span class="n">addMemory</span><span class="p">(</span><span class="n">op_key</span><span class="o">+</span><span class="s">&quot;_input&quot;</span><span class="p">,</span> <span class="n">dst</span><span class="p">);</span>
    <span class="n">dev_ctx</span><span class="p">.</span><span class="n">addPrimitive</span><span class="p">(</span><span class="n">op_key</span><span class="o">+</span><span class="s">&quot;_reorder_input&quot;</span><span class="p">,</span> <span class="n">reorder_primitive</span><span class="p">);</span>
  <span class="p">}</span>

  <span class="n">p</span> <span class="o">=</span> <span class="n">dev_ctx</span><span class="p">.</span><span class="n">findPrimitive</span><span class="p">(</span><span class="n">op_key</span> <span class="o">+</span> <span class="s">&quot;_reorder_input&quot;</span><span class="p">);</span>
  <span class="n">PADDLE_ENFORCE</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="s">&quot;Should have Reorder Primitive&quot;</span><span class="p">);</span>
  <span class="n">dev_ctx</span><span class="p">.</span><span class="n">submit</span><span class="p">(</span><span class="n">p</span><span class="p">);</span>
  <span class="k">if</span> <span class="p">(</span><span class="o">!</span> <span class="k">this</span><span class="o">-&gt;</span><span class="n">isMKLDNNKernel</span><span class="p">())</span> <span class="p">{</span>
    <span class="c1">// execute immediately only if this is not mkldnn kernel function.</span>
    <span class="c1">// otherwise, it can be executed with the operator primitive in Compute</span>
    <span class="n">dev_ctx</span><span class="p">.</span><span class="n">stream</span><span class="p">();</span>
  <span class="p">}</span>
  <span class="c1">// after submit, the input tensor in ExecutionContext should be changed as the converted one</span>
  <span class="c1">// there should be another mechanism to ensure this</span>
<span class="p">}</span>
</pre></div>
</div>
</div>
<div class="section" id="unit-test">
<span id="unit-test"></span><h3>Unit Test<a class="headerlink" href="#unit-test" title="永久链接至标题">¶</a></h3>
<p>All the functions should be tested corresponding.
TBD</p>
</div>
</div>
</div>


           </div>
          </div>
          <footer>
  

  <hr/>

  <div role="contentinfo">
    <p>
        &copy; Copyright 2016, PaddlePaddle developers.

    </p>
  </div>
  Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/snide/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>. 

</footer>

        </div>
      </div>

    </section>

  </div>
  

    <script type="text/javascript">
        var DOCUMENTATION_OPTIONS = {
            URL_ROOT:'../../',
            VERSION:'',
            COLLAPSE_INDEX:false,
            FILE_SUFFIX:'.html',
            HAS_SOURCE:  true
        };
    </script>
      <script type="text/javascript" src="../../_static/jquery.js"></script>
      <script type="text/javascript" src="../../_static/underscore.js"></script>
      <script type="text/javascript" src="../../_static/doctools.js"></script>
      <script type="text/javascript" src="../../_static/translations.js"></script>
      <script type="text/javascript" src="https://cdn.bootcss.com/mathjax/2.7.0/MathJax.js"></script>

  
    <script type="text/javascript" src="../../_static/js/theme.js"></script>
  

  <script type="text/javascript">
      jQuery(function () {
          SphinxRtdTheme.StickyNav.enable();
      });
  </script>
   

</body>
</html>