Pre ce modified (#3335) (#3360)

* Pre ce modified (#3335)

* update

* update

* fix

* fix

* update

* update

* update

* fix

* update

* update

* update

* add ut fix pr(3367)

.github/workflows/_pre_ce_test.yml

@@ -21,6 +21,11 @@ on:
         required: false
         type: string
         default: ""
+      MODEL_CACHE_DIR:
+        description: "Cache Dir Use"
+        required: false
+        type: string
+        default: ""
 
 concurrency:
   group: ${{ github.event.pull_request.number }}
@@ -28,7 +33,7 @@ concurrency:
 
 jobs:
   run_ce_cases:
-    runs-on: [self-hosted, GPU-L20-4Card]
+    runs-on: [self-hosted, PRE_CE_RUN_2Card]
     steps:
       - name: Print current runner name
         run: |
@@ -67,37 +72,51 @@ jobs:
         env:
           docker_image: ${{ inputs.DOCKER_IMAGE }}
           fd_wheel_url: ${{ inputs.FASTDEPLOY_WHEEL_URL }}
+          CACHE_DIR: ${{ inputs.CACHE_DIR }}
+          MODEL_CACHE_DIR: ${{ inputs.MODEL_CACHE_DIR }}
         run: |
           runner_name="${{ runner.name }}"
-          last_char="${runner_name: -1}"
+          CARD_ID=$(echo "${runner_name}" | awk -F'-' '{print $NF}')
+          DEVICES=$(echo "$CARD_ID" | fold -w1 | paste -sd,)
+          DEVICE_PORT=$(echo "$DEVICES" | cut -d',' -f1)
 
-          if [ "${last_char}" = "1" ]; then
+          FLASK_PORT=$((42068 + DEVICE_PORT * 100))
-            gpu_id=2
+          FD_API_PORT=$((42088 + DEVICE_PORT * 100))
-            DEVICES="2,3"
+          FD_ENGINE_QUEUE_PORT=$((42058 + DEVICE_PORT * 100))
-          else
+          FD_METRICS_PORT=$((42078 + DEVICE_PORT * 100))
-            gpu_id=0
+          echo "Test ENV Parameter:"
-            DEVICES="0,1"
+          echo "========================================================="
+          echo "FLASK_PORT=${FLASK_PORT}"
+          echo "FD_API_PORT=${FD_API_PORT}"
+          echo "FD_ENGINE_QUEUE_PORT=${FD_ENGINE_QUEUE_PORT}"
+          echo "FD_METRICS_PORT=${FD_METRICS_PORT}"
+          echo "DEVICES=${DEVICES}"
+          echo "========================================================="
+
+          CACHE_DIR="${CACHE_DIR:-$(dirname "$(dirname "${{ github.workspace }}")")}"
+          echo "CACHE_DIR is set to ${CACHE_DIR}"
+          if [ ! -f "${CACHE_DIR}/gitconfig" ]; then
+            touch "${CACHE_DIR}/gitconfig"
           fi
-          FD_API_PORT=$((9180 + gpu_id * 100))
-          FD_ENGINE_QUEUE_PORT=$((9150 + gpu_id * 100))
-          FD_METRICS_PORT=$((9170 + gpu_id * 100))
 
-          PARENT_DIR=$(dirname "$WORKSPACE")
-          echo "PARENT_DIR:$PARENT_DIR"
           docker run --rm --net=host -v $(pwd):/workspace -w /workspace \
-          -v "/ssd4/GithubActions/gitconfig:/etc/gitconfig:ro" \
+          -v "${CACHE_DIR}/gitconfig:/etc/gitconfig:ro" \
-          -v "/ssd4/GithubActions/ModelData:/ModelData:ro" \
+          -v "${CACHE_DIR}/.cache:/root/.cache" \
-          -v "/ssd4/GithubActions/CacheDir:/root/.cache" \
+          -v "${CACHE_DIR}/ConfigDir:/root/.config" \
-          -v "/ssd4/GithubActions/ConfigDir:/root/.config" \
+          -v "${MODEL_CACHE_DIR}:/ModelData:ro" \
           -e "MODEL_PATH=/ModelData" \
           -e "FD_API_PORT=${FD_API_PORT}" \
           -e "FD_ENGINE_QUEUE_PORT=${FD_ENGINE_QUEUE_PORT}" \
           -e "FD_METRICS_PORT=${FD_METRICS_PORT}" \
+          -e "FLASK_PORT=${FLASK_PORT}" \
           -e "fd_wheel_url=${fd_wheel_url}" \
-          --gpus '"device='"${DEVICES}"'"' ${docker_image} /bin/bash -c '
+          --gpus "\"device=${DEVICES}\"" ${docker_image} /bin/bash -c '
           git config --global --add safe.directory /workspace/FastDeploy
           cd FastDeploy
           python -m pip install --pre paddlepaddle-gpu -i https://www.paddlepaddle.org.cn/packages/nightly/cu126/
           python -m pip install ${fd_wheel_url}
+          for port in $FLASK_PORT $FD_API_PORT $FD_ENGINE_QUEUE_PORT $FD_METRICS_PORT; do
+            lsof -t -i :$port | xargs -r kill -9 || true
+          done
           bash scripts/run_pre_ce.sh
           '

.github/workflows/_unit_test_coverage.yml

@@ -22,6 +22,11 @@ on:
         required: false
         type: string
         default: ""
+      MODEL_CACHE_DIR:
+        description: "Cache Dir Use"
+        required: false
+        type: string
+        default: ""
 
 jobs:
   run_tests_with_coverage:
@@ -67,11 +72,26 @@ jobs:
           fd_wheel_url: ${{ inputs.FASTDEPLOY_WHEEL_URL }}
           CACHE_DIR: ${{ inputs.CACHE_DIR }}
           BASE_REF: ${{ github.event.pull_request.base.ref }}
+          MODEL_CACHE_DIR: ${{ inputs.MODEL_CACHE_DIR }}
         run: |
             set -x
             runner_name="${{ runner.name }}"
             CARD_ID=$(echo "${runner_name}" | awk -F'-' '{print $NF}')
-            gpu_id=$(echo "$CARD_ID" | fold -w1 | paste -sd,)
+            DEVICES=$(echo "$CARD_ID" | fold -w1 | paste -sd,)
+            DEVICE_PORT=$(echo "$DEVICES" | cut -d',' -f1)
+
+            FLASK_PORT=$((42068 + DEVICE_PORT * 100))
+            FD_API_PORT=$((42088 + DEVICE_PORT * 100))
+            FD_ENGINE_QUEUE_PORT=$((42058 + DEVICE_PORT * 100))
+            FD_METRICS_PORT=$((42078 + DEVICE_PORT * 100))
+            echo "Test ENV Parameter:"
+            echo "========================================================="
+            echo "FLASK_PORT=${FLASK_PORT}"
+            echo "FD_API_PORT=${FD_API_PORT}"
+            echo "FD_ENGINE_QUEUE_PORT=${FD_ENGINE_QUEUE_PORT}"
+            echo "FD_METRICS_PORT=${FD_METRICS_PORT}"
+            echo "DEVICES=${DEVICES}"
+            echo "========================================================="
 
             CACHE_DIR="${CACHE_DIR:-$(dirname "$(dirname "${{ github.workspace }}")")}"
             echo "CACHE_DIR is set to ${CACHE_DIR}"
@@ -86,17 +106,22 @@ jobs:
             -v "${CACHE_DIR}/gitconfig:/etc/gitconfig:ro" \
             -v "${CACHE_DIR}/.cache:/root/.cache" \
             -v "${CACHE_DIR}/ConfigDir:/root/.config" \
+            -v "${MODEL_CACHE_DIR}:/ModelData:ro" \
+            -e "MODEL_PATH=/ModelData" \
+            -e "FD_API_PORT=${FD_API_PORT}" \
+            -e "FD_ENGINE_QUEUE_PORT=${FD_ENGINE_QUEUE_PORT}" \
+            -e "FD_METRICS_PORT=${FD_METRICS_PORT}" \
+            -e "FLASK_PORT=${FLASK_PORT}" \
             -e TZ="Asia/Shanghai" \
             -e "fd_wheel_url=${fd_wheel_url}" \
             -e "BASE_REF=${BASE_REF}" \
-            --gpus "\"device=${gpu_id}\"" ${docker_image} /bin/bash -c '
+            --gpus "\"device=${DEVICES}\"" ${docker_image} /bin/bash -c '
 
             git config --global --add safe.directory /workspace/FastDeploy
             cd FastDeploy
             python -m pip install --pre paddlepaddle-gpu -i https://www.paddlepaddle.org.cn/packages/nightly/cu126/
 
-            pip config set global.index-url https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple
+            pip config set global.extra-index-url https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple
-
 
             python -m pip install coverage
             python -m pip install diff-cover
@@ -152,32 +177,40 @@ jobs:
             echo "unittest_failed_url=${UNIT_TEST_RESULT_URL}" >> $GITHUB_OUTPUT
             echo "unittest_failed_url=${UNIT_TEST_RESULT_URL}" >> $GITHUB_ENV
           fi
-      - name: Determine Unit Succ and whether the coverage rate reaches 80%
+      - name: Check Unit Test Success
         shell: bash
         run: |
+          cd FastDeploy
           if [ "$TEST_EXIT_CODE" -eq 8 ]; then
+            filename=$(basename "$unittest_failed_url")
             if [ -z "${unittest_failed_url}" ]; then
               echo "No diff unit failed file URL provided."
             else
-              wget ${unittest_failed_url} || echo "Download unittest file failed, but continuing..."
+              rm -rf "${filename}"
+              wget -O ${filename} ${unittest_failed_url} || echo "Download unittest file failed, but continuing..."
             fi
             echo "Unit tests failed (exit code 8)"
-            filename=$(basename "$unittest_failed_url")
             if [ -f "${filename}" ];then
               echo "Failed test cases:"
               cat "${filename}"
             fi
             exit "$TEST_EXIT_CODE"
           fi
+          echo "All tests passed"
 
+      - name: Verify Code Coverage Threshold (80%)
+        shell: bash
+        run: |
+          cd FastDeploy
           if [ "$COVERAGE_EXIT_CODE" -eq 9 ]; then
             echo "Coverage generation failed (exit code 9)"
+            filename=$(basename "$diff_cov_result_json_url")
             if [ -z "${diff_cov_result_json_url}" ]; then
               echo "No diff cov result file URL provided."
             else
-              wget ${diff_cov_result_json_url} || echo "Download cov json file failed, but continuing..."
+              rm -rf "${filename}"
+              wget -O ${filename} ${diff_cov_result_json_url} || echo "Download cov json file failed, but continuing..."
             fi
-            filename=$(basename "$diff_cov_result_json_url")
             if [ -f "${filename}" ];then
               echo "Failed test cases:"
               if command -v jq >/dev/null 2>&1; then
@@ -188,7 +221,7 @@ jobs:
             fi
             exit "$COVERAGE_EXIT_CODE"
           fi
-          echo "All tests and coverage passed"
+          echo "coverage passed"
           exit 0
 
   diff_coverage_report:

.github/workflows/pr_build_and_test.yml

@@ -39,25 +39,27 @@ jobs:
     needs: [clone,build]
     uses: ./.github/workflows/_unit_test_coverage.yml
     with:
-      DOCKER_IMAGE: ccr-2vdh3abv-pub.cnc.bj.baidubce.com/paddlepaddle/paddleqa:cuda126-py310
+      DOCKER_IMAGE: ccr-2vdh3abv-pub.cnc.bj.baidubce.com/paddlepaddle/paddleqa:fastdeploy-ciuse-cuda126-dailyupdate
       FASTDEPLOY_ARCHIVE_URL: ${{ needs.clone.outputs.repo_archive_url }}
       FASTDEPLOY_WHEEL_URL: ${{ needs.build.outputs.wheel_path }}
+      MODEL_CACHE_DIR: "/ssd2/actions-runner/ModelData"
 
   logprob_test:
     name: Run FastDeploy LogProb Tests
     needs: [build]
     uses: ./.github/workflows/_logprob_test_linux.yml
     with:
-      DOCKER_IMAGE: ccr-2vdh3abv-pub.cnc.bj.baidubce.com/paddlepaddle/paddleqa:cuda126-py310
+      DOCKER_IMAGE: ccr-2vdh3abv-pub.cnc.bj.baidubce.com/paddlepaddle/paddleqa:fastdeploy-ciuse-cuda126-dailyupdate
       PADDLETEST_ARCHIVE_URL: "https://xly-devops.bj.bcebos.com/PaddleTest/PaddleTest.tar.gz"
       FASTDEPLOY_WHEEL_URL: ${{ needs.build.outputs.wheel_path }}
-      MODEL_CACHE_DIR: "/ssd2/actions-runner/ModelCache"
+      MODEL_CACHE_DIR: "/ssd2/actions-runner/ModelData"
 
   pre_ce_test:
     name: Extracted partial CE model tasks to run in CI.
     needs: [clone,build]
     uses: ./.github/workflows/_pre_ce_test.yml
     with:
-      DOCKER_IMAGE: ccr-2vdh3abv-pub.cnc.bj.baidubce.com/paddlepaddle/paddle:fastdeploy-ciuse-cuda126
+      DOCKER_IMAGE: ccr-2vdh3abv-pub.cnc.bj.baidubce.com/paddlepaddle/paddleqa:fastdeploy-ciuse-cuda126-dailyupdate
       FASTDEPLOY_ARCHIVE_URL: ${{ needs.clone.outputs.repo_archive_url }}
       FASTDEPLOY_WHEEL_URL: ${{ needs.build.outputs.wheel_path }}
+      MODEL_CACHE_DIR: "/ssd2/actions-runner/ModelData"

test/entrypoints/openai/test_serving_completion.py

@@ -95,6 +95,7 @@ class TestOpenAIServingCompletion(unittest.TestCase):
             model_name=model_name,
             prompt_batched_token_ids=prompt_batched_token_ids,
             completion_batched_token_ids=completion_batched_token_ids,
+            text_after_process_list=["1", "1"],
         )
 
         assert completion_response.id == request_id