disable benchmark and move test llvm build

.github/workflows/benchmark.yml

@@ -28,7 +28,7 @@ jobs:
     defaults:
       run:
         shell: bash -o pipefail {0}
-    if: github.repository_owner == 'dyKiU'
+    if: github.repository_owner == 'tinygrad'
     steps:
     - name: Checkout Code
       uses: actions/checkout@v4

.github/workflows/test.yml

@@ -14,6 +14,63 @@ on:
   workflow_dispatch:
 
 jobs:
+  macos_tiny_tests:
+    name: Mac tiny tests on AMD
+    runs-on: [self-hosted, macOS]
+    timeout-minutes: 200
+    steps:
+      - name: Checkout Code
+        uses: actions/checkout@v4
+
+      - name: Install Dependencies
+        run: |
+          brew update
+          brew install cmake ninja llvm lld git
+
+      - name: Set LLVM Environment
+        run: |
+          # Set LLVM_PATH to the Homebrew LLVM installation
+          echo "LLVM_PATH=$(brew --prefix llvm)" >> $GITHUB_ENV
+          echo "LLVM_DIR=$(brew --prefix llvm)/lib/cmake/llvm" >> $GITHUB_ENV
+          echo "LLD_DIR=$(brew --prefix llvm)/lib/cmake/lld" >> $GITHUB_ENV
+          echo "CC=$(brew --prefix llvm)/bin/clang" >> $GITHUB_ENV
+          echo "CXX=$(brew --prefix llvm)/bin/clang++" >> $GITHUB_ENV
+          # Set HIP environment variables
+          echo "HIP_PATH=/usr/local/hip" >> $GITHUB_ENV
+          echo "PATH=/usr/local/hip/bin:$PATH" >> $GITHUB_ENV
+          echo "ROOT_DIR=`pwd`" >> $GITHUB_ENV
+          echo "INSTALL_DIR=$ROOT_DIR/usr/local" >> $GITHUB_ENV
+
+      - name: Cache LLVM Build
+        uses: actions/cache@v4
+        with:
+          path: llvm-project/build
+          key: llvm-build-${{ hashFiles('**/llvm-project/CMakeLists.txt') }}
+
+      - name: Build and Install libamd_comgr with LLVM
+        run: |
+          # Build LLVM project
+          git clone --depth 1 --filter=blob:none --sparse -b amd-staging https://github.com/ROCm/llvm-project.git
+          cd llvm-project
+          git sparse-checkout set llvm clang lld cmake third-party/benchmark third-party/unittest
+          mkdir -p build && cd build
+          cmake ../llvm \
+            -DCMAKE_BUILD_TYPE=Release \
+            -DLLVM_ENABLE_PROJECTS="llvm;clang;lld" \
+            -DLLVM_TARGETS_TO_BUILD="AMDGPU;AArch64" \
+            -DLLVM_ENABLE_RTTI=ON \
+            -DCMAKE_PREFIX_PATH=${LLVM_PATH} \
+            -DCMAKE_MODULE_PATH=${LLVM_PATH}/lib/cmake \
+            -DCMAKE_C_COMPILER=${CC} \
+            -DCMAKE_CXX_COMPILER=${CXX} \
+            -DLLVM_ENABLE_BENCHMARKS=OFF \
+            -DCMAKE_INSTALL_PREFIX=${INSTALL_DIR} \
+            -G Ninja
+          ninja || exit 1  # Exit if build fails
+          ninja install
+          cd ../..
+
+
   autogen:
     name: Autogen+Docs
     runs-on: ubuntu-latest
@@ -625,59 +682,69 @@ jobs:
   #      run: pip install -e '.[testing,arm]' --extra-index-url https://download.pytorch.org/whl/cpu
   #    - name: Test arm
   #      run: CI=1 ARM64=1 CLANG=1 python -m pytest -n=auto test/ -k 'not (test_nn.py and (test_conv_transpose2d or test_conv2d))' --ignore=test/models --ignore=test/test_speed_v_torch.py --ignore=test/test_net_speed.py --ignore=test/test_specific_conv.py  --ignore=test/unit/test_disk_tensor.py
-  macos_tiny_tests:
-    name: Mac tiny tests on AMD
-    runs-on: [self-hosted, macOS]
-    timeout-minutes: 200
-    steps:
-      - name: Checkout Code
-        uses: actions/checkout@v4
 
-      - name: Install Dependencies
-        run: |
-          brew update
-          brew install cmake ninja llvm lld git
 
-      - name: Set LLVM Environment
-        run: |
-          # Set LLVM_PATH to the Homebrew LLVM installation
-          echo "LLVM_PATH=$(brew --prefix llvm)" >> $GITHUB_ENV
-          echo "LLVM_DIR=$(brew --prefix llvm)/lib/cmake/llvm" >> $GITHUB_ENV
-          echo "LLD_DIR=$(brew --prefix llvm)/lib/cmake/lld" >> $GITHUB_ENV
-          echo "CC=$(brew --prefix llvm)/bin/clang" >> $GITHUB_ENV
-          echo "CXX=$(brew --prefix llvm)/bin/clang++" >> $GITHUB_ENV
-          # Set HIP environment variables
-          echo "HIP_PATH=/usr/local/hip" >> $GITHUB_ENV
-          echo "PATH=/usr/local/hip/bin:$PATH" >> $GITHUB_ENV
-          echo "ROOT_DIR=`pwd`" >> $GITHUB_ENV
-          echo "INSTALL_DIR=$ROOT_DIR/usr/local" >> $GITHUB_ENV
+  # osxtests:
+  #   strategy:
+  #     fail-fast: false
+  #     matrix:
+  #       backend: [ptx]
 
-      - name: Cache LLVM Build
-        uses: actions/cache@v4
-        with:
-          path: llvm-project/build
-          key: llvm-build-${{ hashFiles('**/llvm-project/CMakeLists.txt') }}
+  #   name: OSX tiny tests on (${{ matrix.backend }})
+  #   runs-on: macos-15
+  #   timeout-minutes: 20
+
+  #   steps:
+  #     - name: Checkout Code
+  #       uses: actions/checkout@v4
+  #       with:
+  #         fetch-depth: 2 # NOTE: this fetches the HEAD commit of the PR
+  #     - name: Set up Python 3.11
+  #       uses: actions/setup-python@v5
+  #       with:
+  #         python-version: 3.11
+  #     - name: Cache python packages
+  #       uses: actions/cache@v4
+  #       with:
+  #         path: /Users/runner/Library/Python/3.11/lib/python/site-packages
+  #         key: m1-${{ matrix.backend }}-packages-${{ hashFiles('**/setup.py') }}
+  #     - name: Set env
+  #       run: printf "${{ matrix.backend == 'ptx' && 'PTX=1\n\nMOCKGPU=1\nNV=1\nFORWARD_ONLY=1' }}" >> $GITHUB_ENV
+  #     - name: Install packages
+  #       if: matrix.backend == 'ptx' || matrix.backend == 'triton' || matrix.backend == 'nv'
+  #       run: |
+  #         brew update
+  #         brew install cmake ninja llvm@15 zlib glew flex bison boost zstd ncurses
+  #     - name: Cache gpuocelot
+  #       if: matrix.backend == 'ptx'
+  #       id: cache-build
+  #       uses: actions/cache@v4
+  #       env:
+  #         cache-name: cache-gpuocelot-osx-build
+  #       with:
+  #         path: ${{ github.workspace }}/gpuocelot/ocelot
+  #         key: macos-15-gpuocelot-4524e34adb7eaccc6f71262f2e21d7052bb17c2f-rebuild-9
+  #     - name: Clone/compile gpuocelot
+  #       if: matrix.backend == 'ptx' && steps.cache-build.outputs.cache-hit != 'true'
+  #       run: |
+  #         git clone --recurse-submodules https://github.com/gpuocelot/gpuocelot.git ${{ github.workspace }}/gpuocelot
+  #         cd ${{ github.workspace }}/gpuocelot
+  #         git checkout 4524e34adb7eaccc6f71262f2e21d7052bb17c2f
+  #         git apply ${{ github.workspace }}/extra/mockgpu/nv/osx_gpuocelot_patches/*.patch
+  #         cd ocelot
+  #         mkdir build
+  #         cd build
+  #         cmake .. -Wno-dev -G Ninja -DBUILD_TOOLS=OFF -DCMAKE_BUILD_ALWAYS=0 -DBUILD_TESTS_CUDA=OFF
+  #         ninja
+  #     - name: Install gpuocelot
+  #       if: matrix.backend == 'ptx' || matrix.backend == 'triton' || matrix.backend == 'nv'
+  #       run: |
+  #         cd ${{ github.workspace }}/gpuocelot/ocelot/build
+  #         sudo ninja install -d explain
+  #     - name: Install dependencies
+  #       run: pip install --user -e '.[testing]' --extra-index-url https://download.pytorch.org/whl/cpu
+  #     - name: Run tiny test (ptx)
+  #       if: matrix.backend=='ptx'
+  #       run: python3 test/test_tiny.py
 
-      - name: Build and Install libamd_comgr with LLVM
-        run: |
-          # Build LLVM project
-          git clone --depth 1 --filter=blob:none --sparse -b amd-staging https://github.com/ROCm/llvm-project.git
-          cd llvm-project
-          git sparse-checkout set llvm clang lld cmake third-party/benchmark third-party/unittest
-          mkdir -p build && cd build
-          cmake ../llvm \
-            -DCMAKE_BUILD_TYPE=Release \
-            -DLLVM_ENABLE_PROJECTS="llvm;clang;lld" \
-            -DLLVM_TARGETS_TO_BUILD="AMDGPU;AArch64" \
-            -DLLVM_ENABLE_RTTI=ON \
-            -DCMAKE_PREFIX_PATH=${LLVM_PATH} \
-            -DCMAKE_MODULE_PATH=${LLVM_PATH}/lib/cmake \
-            -DCMAKE_C_COMPILER=${CC} \
-            -DCMAKE_CXX_COMPILER=${CXX} \
-            -DLLVM_ENABLE_BENCHMARKS=OFF \
-            -DCMAKE_INSTALL_PREFIX=${INSTALL_DIR} \
-            -G Ninja
-          ninja || exit 1  # Exit if build fails
-          ninja install
-          cd ../..