up

.github/workflows/test-gpu.yaml

@@ -21,21 +21,21 @@ jobs:
         env:
           DEBIAN_FRONTEND: noninteractive
         run: |
-          sudo apt-get purge nvidia-* 
-          sudo apt-get autoremove
-          wget -q https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.0-1_all.deb
-          sudo dpkg -i cuda-keyring_1.0-1_all.deb
-          sudo apt-get update
-          sudo apt-get -y install -f nvidia-dkms-555 nvidia-kernel-common-555 nvidia-dkms-555-open nvidia-kernel-common-555 nvidia-driver-555 nvidia-driver-555-open cuda-drivers-555 cuda-drivers cuda-demo-suite-12-1 cuda-runtime-12-1 cuda-12-1
-          cd nccl
-          make -j4 pkg.debian.build
-          sudo dpkg -i build/pkg/deb/libnccl*.deb
+          wget https://developer.download.nvidia.com/compute/cuda/12.1.0/local_installers/cuda_12.1.0_530.30.02_linux.run
+          sudo sh cuda_12.1.0_530.30.02_linux.run --silent --toolkit
+          echo 'export PATH=/usr/local/cuda-12.1/bin:$PATH' >> ~/.bashrc
+          echo 'export LD_LIBRARY_PATH=/usr/local/cuda-12.1/lib64:$LD_LIBRARY_PATH' >> ~/.bashrc
+          source ~/.bashrc
+          /usr/local/cuda-12.1/bin/nvcc --version
+          nvidia-smi
+        shell: bash
       - name: Install Rust nightly
         uses: dtolnay/rust-toolchain@master
         with:
           toolchain: nightly
       - name: E2E Tests
         run: cargo test --release e2e
+        shell: bash
         env:
           NCCL_P2P_DIRECT_DISABLE: 1
           NCCL_NET: Socket