Merge branch 'develop' of https://github.com/PaddlePaddle/Serving into pipeline-auto-batch

doc/LATEST_PACKAGES.md

@@ -14,11 +14,17 @@ https://paddle-serving.bj.bcebos.com/whl/paddle_serving_server-0.3.2-py2-none-an
 ## GPU server
 ### Python 3
 ```
-https://paddle-serving.bj.bcebos.com/whl/paddle_serving_server_gpu-0.3.2-py3-none-any.whl
+#cuda 9.0
+https://paddle-serving.bj.bcebos.com/whl/paddle_serving_server_gpu-0.3.2.post9-py3-none-any.whl
+#cuda 10.0
+https://paddle-serving.bj.bcebos.com/whl/paddle_serving_server_gpu-0.3.2.post10-py3-none-any.whl
 ```
 ### Python 2
 ```
-https://paddle-serving.bj.bcebos.com/whl/paddle_serving_server_gpu-0.3.2-py2-none-any.whl
+#cuda 9.0
+https://paddle-serving.bj.bcebos.com/whl/paddle_serving_server_gpu-0.3.2.post9-py2-none-any.whl
+#cuda 10.0
+https://paddle-serving.bj.bcebos.com/whl/paddle_serving_server_gpu-0.3.2.post10-py2-none-any.whl
 ```
 
 ## Client

python/CMakeLists.txt

@@ -83,6 +83,7 @@ if (SERVER)
             OUTPUT ${PADDLE_SERVING_BINARY_DIR}/.timestamp
             COMMAND cp -r
             ${CMAKE_CURRENT_SOURCE_DIR}/paddle_serving_server_gpu/ ${PADDLE_SERVING_BINARY_DIR}/python/
+            COMMAND env ${py_env} ${PYTHON_EXECUTABLE} paddle_serving_server_gpu/gen_cuda_version.py ${CUDA_VERSION_MAJOR}
             COMMAND env ${py_env} ${PYTHON_EXECUTABLE} setup.py bdist_wheel
             DEPENDS ${SERVING_SERVER_CORE} server_config_py_proto ${PY_FILES})
         add_custom_target(paddle_python ALL DEPENDS ${PADDLE_SERVING_BINARY_DIR}/.timestamp)

python/paddle_serving_server_gpu/__init__.py

@@ -363,7 +363,15 @@ class Server(object):
     def download_bin(self):
         os.chdir(self.module_path)
         need_download = False
-        device_version = "serving-gpu-"
+
+        #acquire lock
+        version_file = open("{}/version.py".format(self.module_path), "r")
+        import re
+        for line in version_file.readlines():
+            if re.match("cuda_version", line):
+                cuda_version = line.split("\"")[1]
+                device_version = "serving-gpu-cuda" + cuda_version + "-"
+
         folder_name = device_version + serving_server_version
         tar_name = folder_name + ".tar.gz"
         bin_url = "https://paddle-serving.bj.bcebos.com/bin/" + tar_name
@@ -372,8 +380,6 @@ class Server(object):
         download_flag = "{}/{}.is_download".format(self.module_path,
                                                    folder_name)
 
-        #acquire lock
-        version_file = open("{}/version.py".format(self.module_path), "r")
         fcntl.flock(version_file, fcntl.LOCK_EX)
 
         if os.path.exists(download_flag):
@@ -385,6 +391,7 @@ class Server(object):
             os.system("touch {}/{}.is_download".format(self.module_path,
                                                        folder_name))
             print('Frist time run, downloading PaddleServing components ...')
+
             r = os.system('wget ' + bin_url + ' --no-check-certificate')
             if r != 0:
                 if os.path.exists(tar_name):

python/paddle_serving_server_gpu/gen_cuda_version.py

+# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import sys
+import re
+import os
+
+new_str = ""
+with open("paddle_serving_server_gpu/version.py", "r") as f:
+    for line in f.readlines():
+        if re.match("cuda_version", line):
+            line = re.sub(r"\d+", sys.argv[1], line)
+        new_str = new_str + line
+
+with open("paddle_serving_server_gpu/version.py", "w") as f:
+    f.write(new_str)

python/paddle_serving_server_gpu/version.py

@@ -15,3 +15,4 @@
 serving_client_version = "0.3.2"
 serving_server_version = "0.3.2"
 module_proto_version = "0.3.2"
+cuda_version = "9"

python/setup.py.server_gpu.in

@@ -41,7 +41,6 @@ REQUIRED_PACKAGES = [
     'paddle_serving_client', 'flask >= 1.1.1', 'paddle_serving_app'
 ]
 
-
 packages=['paddle_serving_server_gpu',
           'paddle_serving_server_gpu.proto',
           'paddle_serving_server_gpu.pipeline',
@@ -58,7 +57,7 @@ package_dir={'paddle_serving_server_gpu':
 
 setup(
     name='paddle-serving-server-gpu',
-    version=serving_server_version.replace('-', ''),
+    version=serving_server_version.replace('-', '') + '.post@CUDA_VERSION_MAJOR@',
     description=
     ('Paddle Serving Package for saved model with PaddlePaddle'),
     url='https://github.com/PaddlePaddle/Serving',