- CUDA 11.6 Update 2, TensorRT 8.2 GA Update 3, cuda-nvrtc, remove co…

…py cuda dlls - CUDA gpg-key change cf. NVIDIA/cuda-repo-management#4
mizar · Apr 30, 2022 · 4adee75 · 4adee75
1 parent 9c4ef55
commit 4adee75
Show file tree

Hide file tree

Showing 4 changed files with 35 additions and 78 deletions.
diff --git a/.github/workflows/make-deep-ubuntu.yml b/.github/workflows/make-deep-ubuntu.yml
@@ -102,21 +102,22 @@ jobs:
         if: ${{ matrix.compiler == 'clang++-14' }}
 
       - name: install CUDA, TensorRT
-        # ここではCUDAアプリケーションの開発環境用に、CUDAドライバーを含まない cuda-minimal-build-11-5 をインストールしている。
-        # CUDAアプリケーションの実行環境用にセットアップする場合、 cuda-minimal-build-11-5 の代わりに
-        # CUDAドライバーを含むメタパッケージ、 cuda-11-5 もしくは cuda をインストールする。
+        # ここではCUDAアプリケーションの開発環境用に、CUDAドライバーを含まない cuda-minimal-build-11-6 をインストールしている。
+        # CUDAアプリケーションの実行環境用にセットアップする場合、 cuda-minimal-build-11-6 の代わりに
+        # CUDAドライバーを含むメタパッケージ、 cuda-11-6 もしくは cuda をインストールする。
         # https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html#package-manager-metas
+        # https://github.com/NVIDIA/cuda-repo-management/issues/4
         run: |
           sudo curl "https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/cuda-ubuntu2004.pin" -o /etc/apt/preferences.d/cuda-repository-pin-600
-          sudo curl "https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/7fa2af80.pub" -o /usr/share/keyrings/cuda.gpg.asc
-          echo "deb [signed-by=/usr/share/keyrings/cuda.gpg.asc] https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/ /" | sudo tee /etc/apt/sources.list.d/cuda.list &&\
+          sudo curl "https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/cuda-ubuntu2004-keyring.gpg" -o /usr/share/keyrings/cuda-archive-keyring.gpg
+          echo "deb [signed-by=/usr/share/keyrings/cuda-archive-keyring.gpg] https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/ /" | sudo tee /etc/apt/sources.list.d/cuda-ubuntu2004-x86_64.list &&\
           sudo cat /etc/apt/sources.list
           sudo ls -R /etc/apt/sources.list.d
           sudo apt update
-          sudo apt install cuda-minimal-build-11-5 libnvinfer-dev libnvinfer-plugin-dev libnvonnxparsers-dev libnvparsers-dev
+          sudo apt install cuda-minimal-build-11-6 cuda-nvrtc-dev-11-6 libcublas-11-6 libcublas-dev-11-6 libnvinfer-dev libnvinfer-plugin-dev libnvonnxparsers-dev libnvparsers-dev
 
       - name: make
-        run: ./main/script/build.sh -e ${{ matrix.edition }} -c ${{ matrix.compiler }} -t ${{ matrix.target }} -a ${{ matrix.archcpu }} -x "EXTRA_CPPFLAGS='-I/usr/local/cuda-11.5/include' EXTRA_LDFLAGS='-L/usr/local/cuda-11.5/lib64'"
+        run: ./main/script/build.sh -e ${{ matrix.edition }} -c ${{ matrix.compiler }} -t ${{ matrix.target }} -a ${{ matrix.archcpu }} -x "EXTRA_CPPFLAGS=-I/usr/local/cuda-11.6/include EXTRA_LDFLAGS=-L/usr/local/cuda-11.6/lib64 EXTRA_LDFLAGS+=-L/usr/local/cuda-11.6/lib64/stubs"
 
       - uses: actions/upload-artifact@v2
         with:

diff --git a/.github/workflows/release.yml b/.github/workflows/release.yml
@@ -1111,26 +1111,26 @@ jobs:
                 - ORT-DML : Windows 10 DirectML (DML) 対応のGPU向け。（対応環境であればおそらくORT-TRT や TensorRTのが高速）
                   cf. [DirectML の概要](https://docs.microsoft.com/ja-jp/windows/win32/direct3d12/dml-intro)
                 - ORT-TRT : NVIDIA TensorRT 対応のGPU向け。CUDA, cuDNN, TensorRT の導入も必要です。
-                  - CUDA (開発環境は 11.5.1)
+                  - CUDA (開発環境は 11.6.2)
                     - [Latest CUDA Toolkit Download](https://developer.nvidia.com/cuda-downloads)
                     - [CUDA Toolkit Archive](https://developer.nvidia.com/cuda-toolkit-archive)
                     - [CUDA Installation Guide for Microsoft Windows (CUDAインストール手順, CUDA 最新版)](https://docs.nvidia.com/cuda/cuda-installation-guide-microsoft-windows/index.html)
-                    - CUDA Toolkit 11.5.1 の場合の導入例:
-                      - [Local Installer (2.4GB)](https://developer.download.nvidia.com/compute/cuda/11.5.1/local_installers/cuda_11.5.1_496.13_windows.exe) もしくは [Network Installer (28.7MB)](https://developer.download.nvidia.com/compute/cuda/11.5.1/network_installers/cuda_11.5.1_windows_network.exe) をダウンロード・実行してインストール
-                  - cuDNN (開発環境は v8.2.4 (Archive))
+                    - CUDA Toolkit 11.6.2 の場合の導入例:
+                      - [Local Installer (2.5GB)](https://developer.download.nvidia.com/compute/cuda/11.6.2/local_installers/cuda_11.6.2_511.65_windows.exe) もしくは [Network Installer (33.6MB)](https://developer.download.nvidia.com/compute/cuda/11.6.2/network_installers/cuda_11.6.2_windows_network.exe) をダウンロード・実行してインストール
+                  - cuDNN (開発環境は [v8.2.4 (Archive)](https://developer.nvidia.com/rdp/cudnn-archive))
                     - [cuDNN Download](https://developer.nvidia.com/rdp/cudnn-download) (要ログイン)
-                    - [cuDNN Archive](https://developer.nvidia.com/rdp/cudnn-archive) (要ログイン)
+                    - (注: Windows版では現状、v8.3.x 系は非推奨。過去のバージョンは [cuDNN Archive](https://developer.nvidia.com/rdp/cudnn-archive) (要ログイン) を参照。)
                     - [Installing cuDNN On Windows (cuDNNインストール手順)](https://docs.nvidia.com/deeplearning/cudnn/install-guide/index.html#install-windows)
                     - cuDNN v8.2.4 (`cudnn-11.4-windows-x64-x64-v8.2.4.15.zip`) の場合の導入例:
                       - ダウンロードした `cudnn-11.4-windows-x64-x64-v8.2.4.15.zip` を右クリック → 「すべて展開」 → 展開時に展開されたファイルを表示する
                       - `cudnn-11.4-windows-x64-x64-v8.2.4.15/cuda/bin` フォルダ階層を選択して表示
                       - `cudnn_adv_infer64_8.dll`, `cudnn_cnn_infer64_8.dll`, `cudnn_ops_infer64_8.dll`, `cudnn64_8.dll` (種類: アプリケーション拡張) の4つのファイルを選択し、`YaneuraOu-Deep-ORT-TRT.exe` と同じフォルダにコピー
-                  - TensorRT (開発環境は 8.2.1 GA)
+                  - TensorRT (開発環境は 8.2 GA Update 2)
                     - [NVIDIA TensorRT Download](https://developer.nvidia.com/nvidia-tensorrt-download) (要ログイン)
                     - [Installing TensorRT (TensorRTインストール手順)](https://docs.nvidia.com/deeplearning/tensorrt/install-guide/index.html#installing-zip)
-                    - TensorRT 8.2.1 GA (`TensorRT-8.2.1.8.Windows10.x86_64.cuda-11.4.cudnn8.2.zip`) の場合の導入例:
-                      - ダウンロードした `TensorRT-8.2.1.8.Windows10.x86_64.cuda-11.4.cudnn8.2.zip` を右クリック → 「すべて展開」 → 展開時に展開されたファイルを表示する
-                      - `TensorRT-8.2.1.8/lib` フォルダ階層を選択して表示
+                    - TensorRT 8.2 GA Update 3 (`TensorRT-8.2.4.2.Windows10.x86_64.cuda-11.4.cudnn8.2.zip`) の場合の導入例:
+                      - ダウンロードした `TensorRT-8.2.4.2.Windows10.x86_64.cuda-11.4.cudnn8.2.zip` を右クリック → 「すべて展開」 → 展開時に展開されたファイルを表示する
+                      - `TensorRT-8.2.4.2/lib` フォルダ階層を選択して表示
                       - `nvinfer.dll`, `nvinfer_builder_resource.dll`, `nvinfer_plugin.dll`, `nvonnxparser.dll`, `nvparsers.dll` (種類: アプリケーション拡張) の5つのファイルを選択し、`YaneuraOu-Deep-ORT-TRT.exe` と同じフォルダにコピー
                   - [警告: 以下のような変更があった場合は、古いエンジンやプロファイルのキャッシュファイル(`.engine`, `.profile`)をクリーンアップしてください。](https://onnxruntime.ai/docs/execution-providers/TensorRT-ExecutionProvider.html#environment-variables)
                     - モデルの変更（モデルのトポロジー、opsetのバージョン、演算子などに変更があった場合）

diff --git a/source/props/YaneuraOuEdition-Deep-ORT-TRT.props b/source/props/YaneuraOuEdition-Deep-ORT-TRT.props
@@ -29,59 +29,35 @@
     </BuildMacro>
   </ItemGroup>
   <ItemGroup>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cublas64_11.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cublas64_11.dll">
+    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\cudnn_adv_infer64_8.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\cudnn_adv_infer64_8.dll">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
       <Visible>false</Visible>
     </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cublasLt64_11.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cublasLt64_11.dll">
+    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\cudnn_cnn_infer64_8.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\cudnn_cnn_infer64_8.dll">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
       <Visible>false</Visible>
     </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudart64_110.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudart64_110.dll">
+    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\cudnn_ops_infer64_8.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\cudnn_ops_infer64_8.dll">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
       <Visible>false</Visible>
     </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudnn_adv_infer64_8.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudnn_adv_infer64_8.dll">
+    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\cudnn64_8.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\cudnn64_8.dll">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
       <Visible>false</Visible>
     </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudnn_cnn_infer64_8.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudnn_cnn_infer64_8.dll">
+    <Content Condition="Exists('C:\TensorRT-8.2.4.2\lib\nvinfer_builder_resource.dll')" Include="C:\TensorRT-8.2.4.2\lib\nvinfer_builder_resource.dll">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
       <Visible>false</Visible>
     </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudnn_ops_infer64_8.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudnn_ops_infer64_8.dll">
+    <Content Condition="Exists('C:\TensorRT-8.2.4.2\lib\nvinfer.dll')" Include="C:\TensorRT-8.2.4.2\lib\nvinfer.dll">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
       <Visible>false</Visible>
     </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudnn64_8.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudnn64_8.dll">
+    <Content Condition="Exists('C:\TensorRT-8.2.4.2\lib\nvinfer_plugin.dll')" Include="C:\TensorRT-8.2.4.2\lib\nvinfer_plugin.dll">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
       <Visible>false</Visible>
     </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cufft64_10.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cufft64_10.dll">
-      <CopyToOutputDirectory>Always</CopyToOutputDirectory>
-      <Visible>false</Visible>
-    </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\nvrtc64_112_0.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\nvrtc64_112_0.dll">
-      <CopyToOutputDirectory>Always</CopyToOutputDirectory>
-      <Visible>false</Visible>
-    </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\nvrtc-builtins64_114.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\nvrtc-builtins64_114.dll">
-      <CopyToOutputDirectory>Always</CopyToOutputDirectory>
-      <Visible>false</Visible>
-    </Content>
-    <Content Condition="Exists('C:\TensorRT-8.2.1.8\lib\nvinfer_builder_resource.dll')" Include="C:\TensorRT-8.2.1.8\lib\nvinfer_builder_resource.dll">
-      <CopyToOutputDirectory>Always</CopyToOutputDirectory>
-      <Visible>false</Visible>
-    </Content>
-    <Content Condition="Exists('C:\TensorRT-8.2.1.8\lib\nvinfer.dll')" Include="C:\TensorRT-8.2.1.8\lib\nvinfer.dll">
-      <CopyToOutputDirectory>Always</CopyToOutputDirectory>
-      <Visible>false</Visible>
-    </Content>
-    <Content Condition="Exists('C:\TensorRT-8.2.1.8\lib\nvinfer_plugin.dll')" Include="C:\TensorRT-8.2.1.8\lib\nvinfer_plugin.dll">
-      <CopyToOutputDirectory>Always</CopyToOutputDirectory>
-      <Visible>false</Visible>
-    </Content>
-    <Content Condition="Exists('C:\TensorRT-8.2.1.8\lib\nvonnxparser.dll')" Include="C:\TensorRT-8.2.1.8\lib\nvonnxparser.dll">
+    <Content Condition="Exists('C:\TensorRT-8.2.4.2\lib\nvonnxparser.dll')" Include="C:\TensorRT-8.2.4.2\lib\nvonnxparser.dll">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
       <Visible>false</Visible>
     </Content>

diff --git a/source/props/YaneuraOuEdition-Deep-TensorRT.props b/source/props/YaneuraOuEdition-Deep-TensorRT.props
@@ -9,8 +9,8 @@
     <OutDir>$(OutBaseDir)$(YaneuraOuDir)\</OutDir>
     <TargetName>$(ProjectName)-$(YaneuraOuTarget)</TargetName>
     <_PropertySheetDisplayName>YaneuraOuEdition-Deep-TensorRT</_PropertySheetDisplayName>
-    <IncludePath>$(VC_IncludePath);$(WindowsSDK_IncludePath);C:\TensorRT-8.2.1.8\include;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\include</IncludePath>
-    <LibraryPath>$(VC_LibraryPath_x64);$(WindowsSDK_LibraryPath_x64);C:\TensorRT-8.2.1.8\lib;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\lib\x64</LibraryPath>
+    <IncludePath>$(VC_IncludePath);$(WindowsSDK_IncludePath);C:\TensorRT-8.2.4.2\include;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include</IncludePath>
+    <LibraryPath>$(VC_LibraryPath_x64);$(WindowsSDK_LibraryPath_x64);C:\TensorRT-8.2.4.2\lib;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\lib\x64</LibraryPath>
   </PropertyGroup>
   <ItemDefinitionGroup>
     <ClCompile>
@@ -29,55 +29,35 @@
     </BuildMacro>
   </ItemGroup>
   <ItemGroup>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cublas64_11.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cublas64_11.dll">
+    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\cudnn_adv_infer64_8.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\cudnn_adv_infer64_8.dll">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
       <Visible>false</Visible>
     </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cublasLt64_11.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cublasLt64_11.dll">
+    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\cudnn_cnn_infer64_8.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\cudnn_cnn_infer64_8.dll">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
       <Visible>false</Visible>
     </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudart64_110.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudart64_110.dll">
+    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\cudnn_ops_infer64_8.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\cudnn_ops_infer64_8.dll">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
       <Visible>false</Visible>
     </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudnn_adv_infer64_8.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudnn_adv_infer64_8.dll">
+    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\cudnn64_8.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\cudnn64_8.dll">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
       <Visible>false</Visible>
     </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudnn_cnn_infer64_8.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudnn_cnn_infer64_8.dll">
+    <Content Condition="Exists('C:\TensorRT-8.2.4.2\lib\nvinfer_builder_resource.dll')" Include="C:\TensorRT-8.2.4.2\lib\nvinfer_builder_resource.dll">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
       <Visible>false</Visible>
     </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudnn_ops_infer64_8.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudnn_ops_infer64_8.dll">
+    <Content Include="C:\TensorRT-8.2.4.2\lib\nvinfer.dll">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
       <Visible>false</Visible>
     </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudnn64_8.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\cudnn64_8.dll">
+    <Content Include="C:\TensorRT-8.2.4.2\lib\nvinfer_plugin.dll">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
       <Visible>false</Visible>
     </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\nvrtc64_112_0.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\nvrtc64_112_0.dll">
-      <CopyToOutputDirectory>Always</CopyToOutputDirectory>
-      <Visible>false</Visible>
-    </Content>
-    <Content Condition="Exists('C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\nvrtc-builtins64_115.dll')" Include="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin\nvrtc-builtins64_115.dll">
-      <CopyToOutputDirectory>Always</CopyToOutputDirectory>
-      <Visible>false</Visible>
-    </Content>
-    <Content Condition="Exists('C:\TensorRT-8.2.1.8\lib\nvinfer_builder_resource.dll')" Include="C:\TensorRT-8.2.1.8\lib\nvinfer_builder_resource.dll">
-      <CopyToOutputDirectory>Always</CopyToOutputDirectory>
-      <Visible>false</Visible>
-    </Content>
-    <Content Include="C:\TensorRT-8.2.1.8\lib\nvinfer.dll">
-      <CopyToOutputDirectory>Always</CopyToOutputDirectory>
-      <Visible>false</Visible>
-    </Content>
-    <Content Include="C:\TensorRT-8.2.1.8\lib\nvinfer_plugin.dll">
-      <CopyToOutputDirectory>Always</CopyToOutputDirectory>
-      <Visible>false</Visible>
-    </Content>
-    <Content Include="C:\TensorRT-8.2.1.8\lib\nvonnxparser.dll">
+    <Content Include="C:\TensorRT-8.2.4.2\lib\nvonnxparser.dll">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
       <Visible>false</Visible>
     </Content>