diff --git a/develop/doc/_sources/howto/dev/new_op_en.md.txt b/develop/doc/_sources/howto/dev/new_op_en.md.txt
index 57ff7caad19cc6bf2e4a052d306d4fc303c8875d..1e88e1f5b4df710f1b69f0305d8d8a2921c4249a 100644
--- a/develop/doc/_sources/howto/dev/new_op_en.md.txt
+++ b/develop/doc/_sources/howto/dev/new_op_en.md.txt
@@ -205,7 +205,7 @@ The definition of its corresponding backward operator, if applicable, is similar
 
     - `REGISTER_OP` registers the `ops::MulOp` class, type named `mul`, its type `ProtoMaker` is `ops::MulOpMaker`, registering `ops::MulOpGrad` as `mul_grad`.
     - `REGISTER_OP_WITHOUT_GRADIENT` registers an operator without gradient.
-    - `REGISTER_OP_CPU_KERNEL` registers `ops::MulKernel` class and specialized template types `paddle::platform::CPUPlace` and `float`, which also registers `ops::MulKernel`.
+    - `REGISTER_OP_CPU_KERNEL` registers `ops::MulKernel` class and specialized template types `paddle::platform::CPUPlace` and `float`, which also registers `ops::MulGradKernel`.
 
 
 - Registering GPU Kernel in `.cu` files
diff --git a/develop/doc/howto/dev/new_op_en.html b/develop/doc/howto/dev/new_op_en.html
index 51559bad019bcfdc45ce9a9484fe6279ad50041b..5def40c6adfdf54ff84c288d7750ba1b4f6ddbe8 100644
--- a/develop/doc/howto/dev/new_op_en.html
+++ b/develop/doc/howto/dev/new_op_en.html
@@ -359,7 +359,7 @@ Registering the Op           | Ops are registered in <code class="docutils liter
 <ul class="simple">
 <li><code class="docutils literal"><span class="pre">REGISTER_OP</span></code> registers the <code class="docutils literal"><span class="pre">ops::MulOp</span></code> class, type named <code class="docutils literal"><span class="pre">mul</span></code>, its type <code class="docutils literal"><span class="pre">ProtoMaker</span></code> is <code class="docutils literal"><span class="pre">ops::MulOpMaker</span></code>, registering <code class="docutils literal"><span class="pre">ops::MulOpGrad</span></code> as <code class="docutils literal"><span class="pre">mul_grad</span></code>.</li>
 <li><code class="docutils literal"><span class="pre">REGISTER_OP_WITHOUT_GRADIENT</span></code> registers an operator without gradient.</li>
-<li><code class="docutils literal"><span class="pre">REGISTER_OP_CPU_KERNEL</span></code> registers <code class="docutils literal"><span class="pre">ops::MulKernel</span></code> class and specialized template types <code class="docutils literal"><span class="pre">paddle::platform::CPUPlace</span></code> and <code class="docutils literal"><span class="pre">float</span></code>, which also registers <code class="docutils literal"><span class="pre">ops::MulKernel</span></code>.</li>
+<li><code class="docutils literal"><span class="pre">REGISTER_OP_CPU_KERNEL</span></code> registers <code class="docutils literal"><span class="pre">ops::MulKernel</span></code> class and specialized template types <code class="docutils literal"><span class="pre">paddle::platform::CPUPlace</span></code> and <code class="docutils literal"><span class="pre">float</span></code>, which also registers <code class="docutils literal"><span class="pre">ops::MulGradKernel</span></code>.</li>
 </ul>
 </li>
 </ul>
diff --git a/develop/doc_cn/_sources/howto/dev/new_op_cn.md.txt b/develop/doc_cn/_sources/howto/dev/new_op_cn.md.txt
index 9d3d02ffc3116ebec537ab9b890eafccad196ed0..c823d7e9fcd63dd7719ac1403952b03c2d2f03c0 100644
--- a/develop/doc_cn/_sources/howto/dev/new_op_cn.md.txt
+++ b/develop/doc_cn/_sources/howto/dev/new_op_cn.md.txt
@@ -206,7 +206,7 @@ MulOp(const std::string &type, const framework::VariableNameMap &inputs,
 
     - `REGISTER_OP` ： 注册`ops::MulOp`类，类型名为`mul`，该类的`ProtoMaker`为`ops::MulOpMaker`，注册`ops::MulOpGrad`，类型名为`mul_grad`。
     - `REGISTER_OP_WITHOUT_GRADIENT` ： 用于注册没有反向的Op。
-    - `REGISTER_OP_CPU_KERNEL` ：注册`ops::MulKernel`类，并特化模板参数为`paddle::platform::CPUPlace`和`float`类型，同理，注册`ops::MulKernel`类。
+    - `REGISTER_OP_CPU_KERNEL` ：注册`ops::MulKernel`类，并特化模板参数为`paddle::platform::CPUPlace`和`float`类型，同理，注册`ops::MulGradKernel`类。
 
 
 - 在 `.cu`文件中注册GPU Kernel。
diff --git a/develop/doc_cn/howto/dev/new_op_cn.html b/develop/doc_cn/howto/dev/new_op_cn.html
index 162cf1ff7ba2256b2cda5215d86309c6278b38f7..1956d080dadbf8fbfb64fbaf6b61eb8b4d17ef79 100644
--- a/develop/doc_cn/howto/dev/new_op_cn.html
+++ b/develop/doc_cn/howto/dev/new_op_cn.html
@@ -373,7 +373,7 @@ Kernel实现       | CPU、GPU共享Kernel实现在<code class="docutils literal
 <ul class="simple">
 <li><code class="docutils literal"><span class="pre">REGISTER_OP</span></code> ： 注册<code class="docutils literal"><span class="pre">ops::MulOp</span></code>类，类型名为<code class="docutils literal"><span class="pre">mul</span></code>，该类的<code class="docutils literal"><span class="pre">ProtoMaker</span></code>为<code class="docutils literal"><span class="pre">ops::MulOpMaker</span></code>，注册<code class="docutils literal"><span class="pre">ops::MulOpGrad</span></code>，类型名为<code class="docutils literal"><span class="pre">mul_grad</span></code>。</li>
 <li><code class="docutils literal"><span class="pre">REGISTER_OP_WITHOUT_GRADIENT</span></code> ： 用于注册没有反向的Op。</li>
-<li><code class="docutils literal"><span class="pre">REGISTER_OP_CPU_KERNEL</span></code> ：注册<code class="docutils literal"><span class="pre">ops::MulKernel</span></code>类，并特化模板参数为<code class="docutils literal"><span class="pre">paddle::platform::CPUPlace</span></code>和<code class="docutils literal"><span class="pre">float</span></code>类型，同理，注册<code class="docutils literal"><span class="pre">ops::MulKernel</span></code>类。</li>
+<li><code class="docutils literal"><span class="pre">REGISTER_OP_CPU_KERNEL</span></code> ：注册<code class="docutils literal"><span class="pre">ops::MulKernel</span></code>类，并特化模板参数为<code class="docutils literal"><span class="pre">paddle::platform::CPUPlace</span></code>和<code class="docutils literal"><span class="pre">float</span></code>类型，同理，注册<code class="docutils literal"><span class="pre">ops::MulGradKernel</span></code>类。</li>
 </ul>
 </li>
 </ul>