Merge branch 'main' of https://github.com/tile-ai/tilelang into frontend-v2

Author: LeiWang1999

.clang-tidy

@@ -1,6 +1,6 @@
 ---
 InheritParentConfig: true
-ExtraArgs: ['-v']
+ExtraArgs: []
 FormatStyle: file
 UseColor: true
 WarningsAsErrors: '*'

.github/workflows/ci.yml

@@ -22,9 +22,12 @@ env:
   PYTHONDEVMODE: "1"
   PYTHONUNBUFFERED: "1"
   PYTHONPATH: "" # explicit cleanup
+  PIP_USER: "" # explicit cleanup
+  COLUMNS: "100"
   FORCE_COLOR: "1"
   CLICOLOR_FORCE: "1"
   UV_INDEX_STRATEGY: "unsafe-best-match"
+  UV_HTTP_TIMEOUT: "600"
   XDG_CACHE_HOME: "${{ github.workspace }}/.cache" # to be updated
   PIP_CACHE_DIR: "${{ github.workspace }}/.cache/pip" # to be updated
   UV_CACHE_DIR: "${{ github.workspace }}/.cache/uv" # to be updated
@@ -43,15 +46,15 @@ jobs:
           submodules: recursive
 
       - name: Setup Python 3.8
-        id: setup-py38
+        id: setup-pylowest
         uses: actions/setup-python@v6
         with:
           python-version: "3.8" # use lowest supported version for linting
           update-environment: false
 
       - name: Check AST with Python 3.8
         run: |
-          "${{ steps.setup-py38.outputs.python-path }}" -m compileall -q -f tilelang
+          "${{ steps.setup-pylowest.outputs.python-path }}" -m compileall -q -f tilelang
 
       - name: Setup Python 3.12
         uses: actions/setup-python@v6
@@ -297,8 +300,9 @@ jobs:
             echo "Using run-clang-tidy from $(command -v run-clang-tidy)"
             CLANG_TIDY=(run-clang-tidy)
           else
-            echo "Downloading run-clang-tidy script"
-            wget -O run-clang-tidy.py https://raw.githubusercontent.com/llvm/llvm-project/refs/heads/release/21.x/clang-tools-extra/clang-tidy/tool/run-clang-tidy.py
+            RCT_URL=https://raw.githubusercontent.com/llvm/llvm-project/refs/heads/release/21.x/clang-tools-extra/clang-tidy/tool/run-clang-tidy.py
+            echo "Downloading run-clang-tidy script from ${RCT_URL}"
+            echo "import urllib.request; url = '${RCT_URL}'.rstrip('/'); urllib.request.urlretrieve(url, url.split('/')[-1])" | uv run --no-project --script -
             CLANG_TIDY=(uv run --no-project --script -- run-clang-tidy.py)
           fi
           if [[ -x "$(command -v clang-apply-replacements)" ]]; then

.github/workflows/dist.yml

@@ -28,7 +28,74 @@ concurrency:
   group: "${{ github.workflow }}-${{ github.ref }}"
   cancel-in-progress: true
 
+env:
+  PYTHONDEVMODE: "1"
+  PYTHONUNBUFFERED: "1"
+  COLUMNS: "100"
+  FORCE_COLOR: "1"
+  CLICOLOR_FORCE: "1"
+
 jobs:
+  build-sdist:
+    name: Build SDist
+    if: |
+      github.repository_owner == 'tile-ai' &&
+      (github.event_name != 'pull_request' || !github.event.pull_request.draft)
+    runs-on: macos-latest
+    timeout-minutes: 30
+    env:
+      NO_VERSION_LABEL: ${{ github.event_name == 'release' && 'OFF' || 'ON' }}
+      # NO_GIT_VERSION disables embedding the git commit hash in version metadata.
+      # Otherwise, the version of the SDist has a git hash suffix (e.g., 0.1.0+gitabcdef12),
+      # but the package built from the SDist has no way to get the git hash (it is not a git repo),
+      # leading to inconsistent versions between SDist and built packages (+gitabcdef12 vs. +gitunknown).
+      NO_GIT_VERSION: "ON"
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v5
+        with:
+          fetch-depth: 1
+          submodules: recursive
+
+      - name: Setup Python and uv with caching
+        id: setup-uv
+        uses: astral-sh/setup-uv@v7
+        with:
+          python-version: "3.12"
+          activate-environment: true
+
+      - name: Build SDist
+        run: |
+          uv run --no-project --with=build -m -- build --sdist --outdir=dist
+
+      - name: Setup ccache
+        uses: hendrikmuhs/ccache-action@v1
+        with:
+          create-symlink: true
+          key: ccache-${{ runner.os }}-${{ runner.arch }}
+          evict-old-files: "7d"
+
+      - name: Test SDist buildable
+        run: |
+          TEMP_DIR="$(mktemp -d -t tilelang-sdist-test)"
+          cp -r dist "${TEMP_DIR}/dist"
+          uv venv --seed "${TEMP_DIR}/venv"
+          source "${TEMP_DIR}/venv/bin/activate"
+          cd "${TEMP_DIR}"
+          python3 -m pip install --upgrade pip setuptools wheel
+          python3 -m pip install -v dist/*.tar.gz
+          python3 -c "import tilelang; print(tilelang.__version__)"
+
+      - name: Upload SDist
+        # Not PR to save artifact storage, as SDist is only needed for releases.
+        if: github.event_name != 'pull_request'
+        uses: actions/upload-artifact@v4
+        with:
+          name: sdist
+          path: dist/*.tar.gz
+          if-no-files-found: error
+
   build-wheels:
     name: Build wheels for Python ${{ matrix.python-version }} on ${{ matrix.target.runner }} with ${{ matrix.target.toolkit }}
     if: |
@@ -41,14 +108,11 @@ jobs:
           - { runner: ubuntu-24.04-arm, toolkit: "CUDA-12.8" }
           - { runner: macos-latest, toolkit: "Metal" }
         python-version:
-          - "3.8"
-          # TVM is built with Python 3.8 Limited API, it should work with all Python >= 3.8.
-          # - "3.9"
-          # - "3.10"
-          # - "3.11"
-          # - "3.12"
-          # - "3.13"
-          # - "3.14"
+          # Wheels are built with Python 3.8 Limited API, they should work with all Python >= 3.8.
+          # Only build wheels against Python 3.8 Limited API to save CI resources.
+          # FIXME: Here we use Python 3.9 because our dependency `apache-tvm-ffi` claims to support
+          #        Python 3.8 but it depends on a version of `ml-dtypes` that requires Python >= 3.9.
+          - "3.9"
       fail-fast: false
     timeout-minutes: 120
     runs-on: ${{ matrix.target.runner }}
@@ -102,12 +166,20 @@ jobs:
 
   list-artifacts:
     name: List artifacts
-    # Not PR to save artifact storage, as wheels are only needed for releases.
+    # Not PR to save artifact storage, as artifacts are only needed for releases.
     if: github.event_name != 'pull_request'
     runs-on: ubuntu-latest
-    needs: [build-wheels]
+    needs: [build-sdist, build-wheels]
     timeout-minutes: 15
     steps:
+      - name: Download built SDist
+        uses: actions/download-artifact@v5
+        with:
+          # unpacks default artifact into dist/
+          # if `name: artifact` is omitted, the action will create extra parent dir
+          name: sdist
+          path: dist
+
       - name: Download built wheels
         uses: actions/download-artifact@v6
         with:

.github/workflows/pr-perfbench-bot.yml

@@ -12,6 +12,17 @@ concurrency:
   group: "${{ github.workflow }}-${{ github.ref }}"
   cancel-in-progress: true # always cancel in-progress
 
+env:
+  PYTHONDEVMODE: "1"
+  PYTHONUNBUFFERED: "1"
+  PYTHONPATH: "" # explicit cleanup
+  PIP_USER: "" # explicit cleanup
+  COLUMNS: "100"
+  FORCE_COLOR: "1"
+  CLICOLOR_FORCE: "1"
+  XDG_CACHE_HOME: "${{ github.workspace }}/.cache" # to be updated
+  PIP_CACHE_DIR: "${{ github.workspace }}/.cache/pip" # to be updated
+
 jobs:
   perfbench:
     name: Benchmark between PR and main
@@ -31,7 +42,12 @@ jobs:
       - name: Setup Python
         uses: actions/setup-python@v6
         with:
-          python-version: "3.9"
+          python-version: "3.12"
+          update-environment: true
+          cache: pip
+          cache-dependency-path: |
+            pyproject.toml
+            requirements*.txt
 
       - name: Install merged version
         run: |

.gitignore

@@ -20,6 +20,7 @@
 debug/
 build/
 *dist/
+dist*/
 wheelhouse/
 __pycache__
 nnfusion.tar.gz

3rdparty/tvm

@@ -8,6 +8,11 @@ set(CMAKE_CXX_STANDARD 17)
 set(CMAKE_POSITION_INDEPENDENT_CODE ON)
 set(CMAKE_EXPORT_COMPILE_COMMANDS ON)
 
+if(CMAKE_CXX_COMPILER_ID STREQUAL "GNU" AND "$ENV{CIBUILDWHEEL}")
+  # Warning came from tvm submodule
+  string(APPEND CMAKE_CXX_FLAGS " -Wno-dangling-reference")
+endif()
+
 set(CMAKE_MODULE_PATH ${CMAKE_MODULE_PATH} ${CMAKE_CURRENT_SOURCE_DIR}/cmake)
 
 if(EXISTS "${CMAKE_CURRENT_SOURCE_DIR}/.gitmodules" AND EXISTS "${CMAKE_CURRENT_SOURCE_DIR}/.git")
@@ -36,9 +41,18 @@ endif()
 
 find_program(CCACHE_PROGRAM ccache)
 if(CCACHE_PROGRAM)
+  message(STATUS "Using ccache: ${CCACHE_PROGRAM}")
   set(CMAKE_C_COMPILER_LAUNCHER "${CCACHE_PROGRAM}" CACHE STRING "C compiler launcher")
   set(CMAKE_CXX_COMPILER_LAUNCHER "${CCACHE_PROGRAM}" CACHE STRING "CXX compiler launcher")
   set(CMAKE_CUDA_COMPILER_LAUNCHER "${CCACHE_PROGRAM}" CACHE STRING "CUDA compiler launcher")
+else()
+  find_program(SCCACHE_PROGRAM sccache)
+  if(SCCACHE_PROGRAM)
+    message(STATUS "Using sccache: ${SCCACHE_PROGRAM}")
+    set(CMAKE_C_COMPILER_LAUNCHER "${SCCACHE_PROGRAM}" CACHE STRING "C compiler launcher")
+    set(CMAKE_CXX_COMPILER_LAUNCHER "${SCCACHE_PROGRAM}" CACHE STRING "CXX compiler launcher")
+    set(CMAKE_CUDA_COMPILER_LAUNCHER "${SCCACHE_PROGRAM}" CACHE STRING "CUDA compiler launcher")
+  endif()
 endif()
 
 # Configs
@@ -68,8 +82,6 @@ file(GLOB TILE_LANG_SRCS
   src/target/utils.cc
   src/target/codegen_cpp.cc
   src/target/rt_mod_cpp.cc
-  # webgpu doesn't have system dependency
-  src/target/codegen_webgpu.cc
   # intrin_rule doesn't have system dependency
   src/target/intrin_rule*.cc
 )
@@ -181,18 +193,18 @@ install(TARGETS tilelang_cython_wrapper
 
 # let libtilelang to search tvm/tvm_runtime in same dir
 if(APPLE)
-  set_target_properties(tilelang PROPERTIES INSTALL_RPATH "@loader_path")
-  set_target_properties(tilelang_module PROPERTIES INSTALL_RPATH "@loader_path")
-else()
-  set_target_properties(tilelang PROPERTIES INSTALL_RPATH "\$ORIGIN")
-  set_target_properties(tilelang_module PROPERTIES INSTALL_RPATH "\$ORIGIN")
+  set_target_properties(tilelang PROPERTIES INSTALL_RPATH "@loader_path;@loader_path/../../tvm_ffi/lib")
+  set_target_properties(tilelang_module PROPERTIES INSTALL_RPATH "@loader_path;@loader_path/../../tvm_ffi/lib")
+  set_target_properties(tvm PROPERTIES INSTALL_RPATH "@loader_path;@loader_path/../../tvm_ffi/lib")
+  set_target_properties(tvm_runtime PROPERTIES INSTALL_RPATH "@loader_path;@loader_path/../../tvm_ffi/lib")
+elseif(UNIX)
+  set_target_properties(tilelang PROPERTIES INSTALL_RPATH "\$ORIGIN:\$ORIGIN/../../tvm_ffi/lib")
+  set_target_properties(tilelang_module PROPERTIES INSTALL_RPATH "\$ORIGIN:\$ORIGIN/../../tvm_ffi/lib")
+  set_target_properties(tvm PROPERTIES INSTALL_RPATH "\$ORIGIN:\$ORIGIN/../../tvm_ffi/lib")
+  set_target_properties(tvm_runtime PROPERTIES INSTALL_RPATH "\$ORIGIN:\$ORIGIN/../../tvm_ffi/lib")
 endif()
 
-install(TARGETS tvm tvm_runtime tilelang_module tilelang LIBRARY DESTINATION tilelang/lib)
-
-# Copy tvm cython ext for wheels
-# TODO: not necessary for editable builds
-if(TVM_BUILD_FROM_SOURCE)
-  add_dependencies(tilelang tvm_cython)
-  install(FILES "${CMAKE_CURRENT_SOURCE_DIR}/3rdparty/tvm/python/tvm/ffi/core.abi3.so" DESTINATION tilelang/3rdparty/tvm/python/tvm/ffi/)
-endif()
+install(
+  TARGETS tvm tvm_runtime tilelang_module tilelang
+  LIBRARY DESTINATION tilelang/lib
+)

CMakeLists.txt

@@ -1,10 +1,35 @@
+# Reference: https://setuptools.pypa.io/en/latest/userguide/miscellaneous.html
+
+# Include licenses
 include VERSION
-include CMakeLists.txt
-include requirements.txt
-include requirements-test.txt
-include requirements-dev.txt
+include LICENSE
+include THIRDPARTYNOTICES.txt
+
+# Version and dependency files
+include version_provider.py
+include requirements*.txt
 include tilelang/jit/adapter/cython/cython_wrapper.pyx
-recursive-include src *
-recursive-include 3rdparty *
-recursive-exclude 3rdparty/clang* * 
-recursive-exclude 3rdparty/llvm* *
+
+# Include source files in SDist
+include CMakeLists.txt
+graft src
+graft cmake
+graft 3rdparty
+
+# Include test suites in SDist
+graft testing
+graft examples
+global-exclude .coverage .coverage.* coverage.xml coverage-*.xml coverage.*.xml
+global-exclude .junit .junit.* junit.xml junit-*.xml junit.*.xml
+
+# Exclude unneeded files and directories
+prune .git
+prune .github
+prune */.git
+prune */.github
+prune 3rdparty/clang*
+prune 3rdparty/llvm*
+
+# Prune compiled files
+prune */__pycache__
+global-exclude *~ *.py[cod] *.so *.a *.dylib *.pxd *.dll *.lib *.o *.obj

MANIFEST.in

@@ -13,6 +13,7 @@ Tile Language (**tile-lang**) is a concise domain-specific language designed to
 <img src=./images/MatmulExample.png />
 
 ## Latest News
+- 10/30/2025 📦: We have released v0.1.6.post2, which is the last version compatible with Python 3.8.
 - 10/07/2025 🍎: Added Apple Metal Device support, check out [Pull Request #799](https://github.com/tile-ai/tilelang/pull/799) for details.
 - 09/29/2025  🎉: Thrilled to announce that ​​AscendC​​ and ​Ascend​NPU IR​​ backends targeting Huawei Ascend chips are now supported!
 Check out the preview here:

README.md

@@ -1 +1 @@
-0.1.6.post1
+0.1.6.post2