fix(//docker): Workaround only shared libraries being available in

PyTorch container Signed-off-by: Naren Dasan <[email protected]> Signed-off-by: Naren Dasan <[email protected]>
pytorch · Aug 25, 2020 · 50c7eda · 50c7eda
1 parent 970d775
commit 50c7eda
Show file tree

Hide file tree

Showing 2 changed files with 5 additions and 1 deletion.
diff --git a/docker/Dockerfile.20.07 b/docker/Dockerfile.20.07
@@ -14,6 +14,9 @@ COPY . /workspace/TRTorch
 RUN rm /workspace/TRTorch/WORKSPACE
 COPY ./docker/WORKSPACE.cu11.docker  /workspace/TRTorch/WORKSPACE
 
+# Workaround for bazel expecting both static and shared versions, we only use shared libraries inside container
+RUN cp /usr/lib/x86_64-linux-gnu/libnvinfer.so /usr/lib/x86_64-linux-gnu/libnvinfer_static.a
+
 WORKDIR /workspace/TRTorch
 RUN bazel build //:libtrtorch --compilation_mode opt
 

diff --git a/third_party/tensorrt/local/BUILD b/third_party/tensorrt/local/BUILD
@@ -68,8 +68,9 @@ cc_import(
         "//conditions:default": "lib/x86_64-linux-gnu/libnvinfer.so",
     }),
     static_library = select({
+        ":aarch64_linux": "lib/aarch64-linux-gnu/libnvinfer_static.a",
         ":windows": "lib/nvinfer.lib",
-        "//conditions:default": ""
+        "//conditions:default": "lib/x86_64-linux-gnu/libnvinfer_static.a"
     }),
     visibility = ["//visibility:private"],
 )