diff --git a/doc/COMPILE.md b/doc/COMPILE.md
index cad747c111230dd594436766bec838f35c49ff3a..737b4e90c6b5140039928c36f002a5c5613d8a44 100644
--- a/doc/COMPILE.md
+++ b/doc/COMPILE.md
@@ -16,7 +16,11 @@ It is recommended to use Docker for compilation. We have prepared the Paddle Ser
 - CPU: `hub.baidubce.com/paddlepaddle/serving:0.2.0-devel`，dockerfile: [Dockerfile.devel](../tools/Dockerfile.devel)
 - GPU: `hub.baidubce.com/paddlepaddle/serving:0.2.0-gpu-devel`，dockerfile: [Dockerfile.gpu.devel](../tools/Dockerfile.gpu.devel)
 
-This document will take Python2 as an example to show how to compile Paddle Serving. If you want to compile with Python 3, just adjust the Python options of cmake.
+This document will take Python2 as an example to show how to compile Paddle Serving. If you want to compile with Python3, just adjust the Python options of cmake:
+
+- Set `DPYTHON_INCLUDE_DIR` to `$PYTHONROOT/include/python3.6m/`
+- Set  `DPYTHON_LIBRARIES` to `$PYTHONROOT/lib64/libpython3.6.so`
+- Set `DPYTHON_EXECUTABLE` to `$PYTHONROOT/bin/python3`
 
 ## Get Code
 
@@ -54,7 +58,7 @@ make -j10
 
 execute `make install` to put targets under directory `./output`
 
-**Attention：**After the compilation is successful, the serving binary file will be generated in the ./core/general-server directory. Before starting the server, export SERVING_BIN = $ {path / to / serving / bin} is required to allow the server to use the compiled serving binary file.
+**Attention：**After the compilation is successful, you need to set the path of `SERVING_BIN`. See [Note](https://github.com/PaddlePaddle/Serving/blob/develop/doc/COMPILE.md#Note) for details.
 
 ## Compile Client
 
diff --git a/doc/COMPILE_CN.md b/doc/COMPILE_CN.md
index afff44b69fe81fa3eb5a4c32692590f6a47d16a1..4488efa14ca83a8717dd709b50edee6ede695aa1 100644
--- a/doc/COMPILE_CN.md
+++ b/doc/COMPILE_CN.md
@@ -16,7 +16,11 @@
 - CPU: `hub.baidubce.com/paddlepaddle/serving:0.2.0-devel`，dockerfile: [Dockerfile.devel](../tools/Dockerfile.devel)
 - GPU: `hub.baidubce.com/paddlepaddle/serving:0.2.0-gpu-devel`，dockerfile: [Dockerfile.gpu.devel](../tools/Dockerfile.gpu.devel)
 
-本文档将以Python2为例介绍如何编译Paddle Serving。如果您想用Python3进行编译，只需要调整cmake的Python相关选项即可。
+本文档将以Python2为例介绍如何编译Paddle Serving。如果您想用Python3进行编译，只需要调整cmake的Python相关选项即可：
+
+- 将`DPYTHON_INCLUDE_DIR`设置为`$PYTHONROOT/include/python3.6m/`
+- 将`DPYTHON_LIBRARIES`设置为`$PYTHONROOT/lib64/libpython3.6.so`
+- 将`DPYTHON_EXECUTABLE`设置为`$PYTHONROOT/bin/python3`
 
 ## 获取代码
 
@@ -54,7 +58,7 @@ make -j10
 
 执行`make install`可以把目标产出放在`./output`目录下。
 
-**注意：** 编译成功后，在./core/general-server目录下会产出serving二进制文件。启动server前需要export SERVING_BIN=${path/to/serving/bin} 来让server端使用编译出的serving二进制文件。
+**注意：** 编译成功后，需要设置`SERVING_BIN`路径，详见后面的[注意事项](https://github.com/PaddlePaddle/Serving/blob/develop/doc/COMPILE_CN.md#注意事项))。
 
 ## 编译Client部分