[Cherry-Pick][CI] Sync dev optimizations to 2.4(#7335) (#7346)

* [Cherry-Pick][CI] Sync dev optimizations to 2.4(#7335)

.github/workflows/_unit_test_coverage.yml

@@ -85,12 +85,27 @@ jobs:
             if ls "${REPO_NAME}"* >/dev/null 2>&1; then
               echo "ERROR: Failed to clean ${REPO_NAME}* after multiple attempts"
               ls -ld "${REPO_NAME}"*
-              exit 1
+              echo "Attempting force cleanup with find..."
+              find /workspace -mindepth 1 -maxdepth 1 -name "${REPO_NAME}*" -type d -exec chmod -R u+rwx {} \; -exec rm -rf {} + 2>/dev/null || true
+              if ls "${REPO_NAME}"* >/dev/null 2>&1; then
+                echo "ERROR: Force cleanup still failed"
+                exit 1
+              else
+                echo "Force cleanup succeeded"
+              fi
             fi
           '
 
-          wget -q --no-proxy ${fd_archive_url}
-          tar -xf FastDeploy.tar.gz
+          wget -q --no-proxy ${fd_archive_url} || {
+            echo "ERROR: Failed to download archive from ${fd_archive_url}"
+            exit 1
+          }
+
+          tar --no-same-owner -xf FastDeploy.tar.gz || {
+            echo "ERROR: Failed to extract archive"
+            exit 1
+          }
+
           rm -rf FastDeploy.tar.gz
           cd FastDeploy
           git config --global user.name "FastDeployCI"
@@ -173,12 +188,16 @@ jobs:
           export RDMA_DEVICES=$(find /dev/infiniband/uverbs* -maxdepth 1 -not -type d | xargs -I{} echo '--device {}:{}')
 
           docker run --rm --net=host \
+          --sysctl kernel.msgmax=1048576 \
+          --sysctl kernel.msgmnb=268435456 \
           --name ${runner_name} \
           --cap-add=SYS_PTRACE --cap-add=IPC_LOCK \
-          --shm-size=64G \
+          --shm-size=128G \
           ${RDMA_DEVICES} \
           --device=/dev/infiniband/rdma_cm \
           --ulimit memlock=-1:-1 \
+          --ulimit nofile=65536:65536 \
+          --ulimit nproc=8192:8192 \
           -v $(pwd):/workspace -w /workspace \
           -v "${CACHE_DIR}/gitconfig:/etc/gitconfig:ro" \
           -v "${CACHE_DIR}/.cache:/root/.cache" \
@@ -198,6 +217,7 @@ jobs:
           -e "fd_wheel_url=${fd_wheel_url}" \
           -e "BASE_REF=${BASE_REF}" \
           -e "IS_PR=${IS_PR}" \
+          -e "no_proxy=localhost,127.0.0.1,0.0.0.0,bcebos.com,.bcebos.com,bj.bcebos.com,su.bcebos.com,paddle-ci.gz.bcebos.com,apiin.im.baidu.com,baidu-int.com,.baidu.com,aliyun.com,gitee.com,pypi.tuna.tsinghua.edu.cn,.tuna.tsinghua.edu.cn" \
           --gpus "\"device=${DEVICES}\"" ${docker_image} /bin/bash -c '
 
           git config --global --add safe.directory /workspace/FastDeploy
@@ -205,7 +225,7 @@ jobs:
           git diff origin/${BASE_REF}..HEAD --unified=0 > diff.txt
           # Avoid using pip cache to ensure the wheel is updated to the latest version
           wget -q --no-proxy https://paddle-qa.bj.bcebos.com/paddle-pipeline/Release-TagBuild-Training-Linux-Gpu-Cuda12.6-Cudnn9.5-Trt10.5-Mkl-Avx-Gcc11-SelfBuiltPypiUse/latest/paddlepaddle_gpu-0.0.0-cp310-cp310-linux_x86_64.whl
-          python -m pip install paddlepaddle_gpu-0.0.0-cp310-cp310-linux_x86_64.whl
+          python -m pip install paddlepaddle_gpu-0.0.0-cp310-cp310-linux_x86_64.whl --extra-index-url https://www.paddlepaddle.org.cn/packages/stable/cu126/
           pip config set global.extra-index-url https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple
 
           python -m pip install -r scripts/unittest_requirement.txt
@@ -380,6 +400,13 @@ jobs:
           echo "coverage passed"
           exit 0
 
+      - name: Terminate and delete the container
+        if: always()
+        run: |
+          set +e
+          docker exec -t ${{ runner.name }} /bin/bash -c 'find /workspace -mindepth 1 -delete'
+          docker rm -f ${{ runner.name }}
+
   diff_coverage_report:
     needs: run_tests_with_coverage
     if: always()