upd

.github/workflows/release_wheel.yml

@@ -25,9 +25,9 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python: ["3.8", "3.9", "3.10", "3.11", "3.12"]
-        cuda: ["11.8", "12.1", "12.4"]
-        torch: ["2.2", "2.3", "2.4"]
+        python: ["3.10"]
+        cuda: ["12.1"]
+        torch: ["2.3"]
         exclude: # for cuda 12.4, we only support torch 2.4+
           - cuda: "12.4"
             torch: "2.2"
@@ -63,75 +63,3 @@ jobs:
         with:
           name: wheel-cuda${{ matrix.cuda }}-torch${{ matrix.torch }}-python${{ matrix.python }}
           path: python/dist/*
-
-  release:
-    needs: build
-    runs-on: [self-hosted]
-    steps:
-      - uses: actions/download-artifact@v4
-        with:
-          path: python/dist/
-          merge-multiple: true
-          pattern: wheel-*
-
-      - run: ls -lah python/dist/
-
-      - uses: softprops/action-gh-release@v1
-        with:
-          tag_name: ${{ inputs.tag_name }}
-          files: |
-            python/dist/flashinfer*cp38*.whl
-
-      - uses: softprops/action-gh-release@v1
-        with:
-          tag_name: ${{ inputs.tag_name }}
-          files: |
-            python/dist/flashinfer*cp39*.whl
-
-      - uses: softprops/action-gh-release@v1
-        with:
-          tag_name: ${{ inputs.tag_name }}
-          files: |
-            python/dist/flashinfer*cp310*.whl
-
-      - uses: softprops/action-gh-release@v1
-        with:
-          tag_name: ${{ inputs.tag_name }}
-          files: |
-            python/dist/flashinfer*cp311*.whl
-
-      - uses: softprops/action-gh-release@v1
-        with:
-          tag_name: ${{ inputs.tag_name }}
-          files: |
-            python/dist/flashinfer*cp312*.whl
-
-      - uses: softprops/action-gh-release@v1
-        with:
-          tag_name: ${{ inputs.tag_name }}
-          files: |
-            python/dist/flashinfer-*.tar.gz
-
-      - name: Clone wheel index
-        run: git clone https://oauth2:${WHL_TOKEN}@github.com/flashinfer-ai/whl.git flashinfer-whl
-        env:
-          WHL_TOKEN: ${{ secrets.WHL_TOKEN }}
-
-      - name: Update wheel index
-        run: python3 scripts/update_whl_index.py
-
-      - name: Push wheel index
-        run: |
-          cd flashinfer-whl
-          git config --local user.name "github-actions[bot]"
-          git config --local user.email "41898282+github-actions[bot]@users.noreply.github.com"
-          git add -A
-          git commit -m "update whl"
-          git push
-
-      # - name: Upload sdist to pypi
-      #   run: |
-      #     pip install twine
-      #     python -m twine upload --repository testpypi --username=__token__ dist/*.tar.gz
-      #   env:
-      #     TWINE_PASSWORD: ${{ secrets.PYPI_TEST_TOKEN }}

python/setup.py

@@ -328,6 +328,7 @@ def __init__(self, *args, **kwargs) -> None:
             "-Xfatbin",
             "-compress-all",
             "-use_fast_math",
+            "--verbose",
         ],
     }
     ext_modules = []