diff --git a/deploy/cpp/readme.md b/deploy/cpp/readme.md
index 72ce8dacc..68a509c59 100644
--- a/deploy/cpp/readme.md
+++ b/deploy/cpp/readme.md
@@ -122,12 +122,11 @@ build/paddle_inference_install_dir/
 
 * [Paddle预测库官网](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/guides/05_inference_deployment/inference/build_and_install_lib_cn.html#id1)上提供了不同cuda版本的Linux预测库，可以在官网查看并选择合适的预测库版本，注意必须选择`develop`版本。
 
-  以`ubuntu14.04_cuda9.0_cudnn7_avx_mkl`的`develop`版本为例，使用下述命令下载并解压：
+  以`2.1.1-gpu-cuda10.2-cudnn8.1-mkl-gcc8.2`的`develop`版本为例，使用下述命令下载并解压：
 
 
 ```shell
-wget https://paddle-inference-lib.bj.bcebos.com/latest-gpu-cuda9-cudnn7-avx-mkl/paddle_inference.tgz
-
+wget https://paddle-inference-lib.bj.bcebos.com/2.1.1-gpu-cuda10.2-cudnn8.1-mkl-gcc8.2/paddle_inference.tgz
 tar -xvf paddle_inference.tgz
 ```
 
diff --git a/deploy/cpp_shitu/readme.md b/deploy/cpp_shitu/readme.md
index 7ebf45be1..e46b23865 100644
--- a/deploy/cpp_shitu/readme.md
+++ b/deploy/cpp_shitu/readme.md
@@ -229,6 +229,7 @@ cmake .. \
     -DFAISS_WITH_MKL=${FAISS_WITH_MKL}
 
 make -j
+cd ..
 ```
 
 上述命令中，
diff --git a/deploy/cpp_shitu/tools/build.sh b/deploy/cpp_shitu/tools/build.sh
index 5307af061..a9e2f9229 100755
--- a/deploy/cpp_shitu/tools/build.sh
+++ b/deploy/cpp_shitu/tools/build.sh
@@ -21,4 +21,5 @@ cmake .. \
     -DFAISS_DIR=${FAISS_DIR} \
     -DFAISS_WITH_MKL=${FAISS_WITH_MKL}
 
-make -j
\ No newline at end of file
+make -j
+cd ..