OSX GPUOcelot (#8209)

* add patches

* add osx test in ci

* macos specific uvm, gpfifo mask

* only do that for now

* Revert "add patches"

This reverts commit 80d3112a57.

* use fork for now

* workflow only one worker

* merge osxtests with tests

* Revert "merge osxtests with tests"

This reverts commit 3461c8f46c.

* macos pagesize 16384

---------

Co-authored-by: nimlgen <[email protected]>
Co-authored-by: George Hotz <[email protected]>

.github/actions/setup-tinygrad/action.yml

@@ -143,14 +143,20 @@ runs:
 
     # **** CUDA ****
 
-    - name: Install packages (cuda)
-      if: inputs.cuda == 'true'
+    - name: Install cuda packages (Linux)
+      if: inputs.cuda == 'true' && runner.os == 'Linux'
       shell: bash
       run: |
         echo 'Acquire::http::Pipeline-Depth "5";' | sudo tee -a /etc/apt/apt.conf.d/99parallel
         sudo apt update -y || true
         sudo apt install -y --no-install-recommends git g++ cmake ninja-build llvm-15-dev zlib1g-dev libglew-dev \
           flex bison libfl-dev libboost-thread-dev libboost-filesystem-dev nvidia-cuda-toolkit-gcc libzstd-dev
+    - name: Install gpuocelot dependencies (MacOS)
+      if: inputs.cuda == 'true' && runner.os == 'macOS'
+      shell: bash
+      run: |
+        brew update
+        brew install cmake ninja llvm@15 zlib glew flex bison boost zstd ncurses
     - name: Cache gpuocelot
       if: inputs.cuda == 'true'
       id: cache-build
@@ -159,7 +165,7 @@ runs:
         cache-name: cache-gpuocelot-build
       with:
         path: ${{ github.workspace }}/gpuocelot/ocelot
-        key: ubuntu22.04-gpuocelot-b16039dc940dc6bc4ea0a98380495769ff35ed99-rebuild-0
+        key: ${{ runner.os }}-gpuocelot-b16039dc940dc6bc4ea0a98380495769ff35ed99-rebuild-0
     - name: Clone/compile gpuocelot
       if: inputs.cuda == 'true' && steps.cache-build.outputs.cache-hit != 'true'
       shell: bash
@@ -176,7 +182,7 @@ runs:
       shell: bash
       run: |
         cd ${{ github.workspace }}/gpuocelot/ocelot/build
-        sudo cp libgpuocelot.so /usr/lib/libgpuocelot.so
+        sudo cp libgpuocelot.${{ runner.os == 'macOS' && 'dylib' || 'so' }} /usr/${{ runner.os == 'macOS' && 'local/' || ''}}lib/
 
     # **** WebGPU ****
 

.github/workflows/test.yml

@@ -513,6 +513,7 @@ jobs:
         deps: testing
         python-version: '3.11'
         amd: 'true'
+        cuda: 'true'
     - name: Run real world test
       run: JIT=2 METAL=1 python -m pytest -n=auto test/models/test_real_world.py --durations=20
     - name: Test models (Metal)
@@ -540,6 +541,14 @@ jobs:
         FORWARD_ONLY: 1
       run: |
         python3 -m pytest -n=auto test/test_hcq.py test/test_tiny.py --durations=20
+    - name: Run pytest (ptx)
+      env:
+        MOCKGPU: 1
+        PTX: 1
+        NV: 1
+        FORWARD_ONLY: 1
+      run: |
+        python3 -m pytest -n=auto test/test_hcq.py test/test_tiny.py --durations=20
     - name: Run process replay tests
       uses: ./.github/actions/process-replay