Multireduce Kernels - prereq refactor (#4173) #1

Workflow file for this run

	name: Unit Tests
	env:
	# increment this when downloads substantially change to avoid the internet
	DOWNLOAD_CACHE_VERSION: '4'

	on:
	push:
	branches:
	- master
	pull_request:
	workflow_dispatch:

	jobs:
	uops:
	name: uops tests
	runs-on: ubuntu-latest
	timeout-minutes: 20
	steps:
	- name: Checkout Code
	uses: actions/checkout@v4
	- name: Set up Python 3.12
	uses: actions/setup-python@v5
	with:
	python-version: 3.12
	- name: Cache python packages
	uses: actions/cache@v4
	with:
	path: ${{ env.Python3_ROOT_DIR }}/lib/python3.12/site-packages
	key: uops-packages-${{ hashFiles('**/setup.py') }}-3.12
	- name: Install dependencies
	run: pip install -e '.[testing]' --extra-index-url https://download.pytorch.org/whl/cpu
	- name: Test IMAGE=2 support
	run: \|
	IMAGE=2 PYTHON=1 python3 test/test_ops.py TestOps.test_gemm
	IMAGE=2 PYTHON=1 python3 test/test_ops.py TestOps.test_simple_conv2d
	- name: Test emulated METAL tensor cores
	run: DEBUG=2 EMULATE_METAL=1 FORWARD_ONLY=1 PYTHON=1 python3 test/test_ops.py TestOps.test_big_gemm
	- name: Test emulated HSA tensor cores
	run: \|
	PYTHONPATH=. DEBUG=2 EMULATE_HSA=1 FORWARD_ONLY=1 PYTHON=1 N=16 HALF=1 ACC_HALF=0 python3 ./extra/gemm/simple_matmul.py
	PYTHONPATH=. DEBUG=2 EMULATE_HSA=1 FORWARD_ONLY=1 PYTHON=1 N=64 HALF=1 ACC_HALF=0 python3 ./extra/gemm/simple_matmul.py
	PYTHONPATH=. DEBUG=2 EMULATE_HSA=1 FORWARD_ONLY=1 PYTHON=1 N=16 HALF=1 ACC_HALF=1 python3 ./extra/gemm/simple_matmul.py
	PYTHONPATH=. DEBUG=2 EMULATE_HSA=1 FORWARD_ONLY=1 PYTHON=1 N=64 HALF=1 ACC_HALF=1 python3 ./extra/gemm/simple_matmul.py
	- name: Test emulated CUDA tensor cores
	run: DEBUG=2 EMULATE_CUDA=1 FORWARD_ONLY=1 PYTHON=1 python3 test/test_ops.py TestOps.test_gemm
	- name: Full test tensor cores
	run: \|
	DEBUG=2 EMULATE_METAL=1 FORWARD_ONLY=1 PYTHON=1 python3 ./test/test_linearizer.py TestLinearizer.test_tensor_cores
	DEBUG=2 EMULATE_HSA=1 FORWARD_ONLY=1 PYTHON=1 python3 ./test/test_linearizer.py TestLinearizer.test_tensor_cores
	DEBUG=2 EMULATE_CUDA=1 FORWARD_ONLY=1 PYTHON=1 python3 ./test/test_linearizer.py TestLinearizer.test_tensor_cores
	- name: Test dtype with Python emulator
	run: DEBUG=1 PYTHONPATH=. PYTHON=1 python3 test/test_dtype.py
	- name: Test ops with Python emulator
	run: DEBUG=2 PYTHON=1 python3 -m pytest test/test_ops.py -k "not (test_split or test_simple_cumsum or test_cumsum or test_einsum or test_dot or test_dot_1d or test_big_gemm or test_broadcastdot or test_multidot or test_var_axis or test_std_axis or test_broadcast_full or test_broadcast_partial or test_simple_conv3d or test_dilated_conv_transpose2d or test_simple_conv_transpose3d or test_large_input_conv2d or test_maxpool2d or test_maxpool2d_simple or test_maxpool2d_bigger_stride or test_avgpool2d or test_cat or test_scaled_product_attention or test_scaled_product_attention_causal or test_slice_fancy_indexing_dim_inject_none or test_slice_fancy_indexing_list_indices or test_slice_fancy_indexing_no_dim_collapse or test_slice_fancy_indexing_tuple_indices or test_slice_fancy_indexing_list_with_tensors or test_slice_fancy_indexing_dim_collapse_int)" --durations=20
	- name: Test uops with Python emulator
	run: PYTHON=1 python3 -m pytest test/test_uops.py --durations=20
	- name: Test symbolic with Python emulator
	run: PYTHONPATH=. PYTHON=1 python3 test/test_symbolic_ops.py
	- name: test_linearizer_failures with Python emulator
	run: PYTHONPATH=. PYTHON=1 python3 -m pytest -rA test/test_linearizer_failures.py::TestLinearizerFailures::test_failure_1

	linter:
	name: Linters
	runs-on: ubuntu-latest
	timeout-minutes: 20

	# TODO: run the pre-commit hook to replace a lot of this
	steps:
	- name: Checkout Code
	uses: actions/checkout@v4
	- name: Set up Python 3.8
	uses: actions/setup-python@v5
	with:
	python-version: 3.8
	- name: Cache python packages
	uses: actions/cache@v4
	with:
	path: ${{ env.Python3_ROOT_DIR }}/lib/python3.8/site-packages
	key: linting-packages-${{ hashFiles('**/setup.py') }}-3.8
	- name: Install dependencies
	run: pip install -e '.[linting,testing]' --extra-index-url https://download.pytorch.org/whl/cpu
	- name: Lint with pylint
	run: python -m pylint --disable=all -e W0311 -e C0303 --jobs=0 --indent-string=' ' */.py
	- name: Lint with ruff
	run: \|
	pip3 install --upgrade --force-reinstall ruff
	python3 -m ruff check . --preview
	- name: Lint tinygrad with pylint
	run: python -m pylint tinygrad/
	- name: Run mypy
	run: python -m mypy --strict-equality
	- name: Test Docs
	run: \|
	python docs/abstractions2.py
	- name: Test Quickstart
	run: awk '/```python/{flag=1;next}/```/{flag=0}flag' docs/quickstart.md > quickstart.py && PYTHONPATH=. python quickstart.py
	- name: Fuzz Test symbolic
	run: python test/external/fuzz_symbolic.py
	- name: Fuzz Test shapetracker
	run: \|
	PYTHONPATH="." python test/external/fuzz_shapetracker.py
	PYTHONPATH="." python test/external/fuzz_shapetracker_math.py
	- name: Test to_movement_ops
	run: PYTHONPATH="." python extra/to_movement_ops.py
	- name: Use as an external package
	run: \|
	mkdir $HOME/test_external_dir
	cd $HOME/test_external_dir
	python -m venv venv
	source venv/bin/activate
	pip install $GITHUB_WORKSPACE
	python -c "from tinygrad.tensor import Tensor; print(Tensor([1,2,3,4,5]))"
	- name: Test DEBUG
	run: DEBUG=100 python3 -c "from tinygrad import Tensor; N = 1024; a, b = Tensor.rand(N, N), Tensor.rand(N, N); c = (a.reshape(N, 1, N) * b.T.reshape(1, N, N)).sum(axis=2); print((c.numpy() - (a.numpy() @ b.numpy())).mean())"
	- name: Repo line count <7000 lines
	run: MAX_LINE_COUNT=7000 python sz.py

	testcpuimagenet:
	name: ImageNet to C Tests
	runs-on: ubuntu-latest
	timeout-minutes: 20

	steps:
	- name: Checkout Code
	uses: actions/checkout@v4
	- name: Set up Python 3.8
	uses: actions/setup-python@v5
	with:
	python-version: 3.8
	- name: Cache python packages
	uses: actions/cache@v4
	with:
	path: ${{ env.Python3_ROOT_DIR }}/lib/python3.8/site-packages
	key: testing-packages-${{ hashFiles('**/setup.py') }}
	- name: Cache downloads
	uses: actions/cache@v4
	with:
	path: ~/.cache/tinygrad/downloads/
	key: downloads-cache-cpu-${{ env.DOWNLOAD_CACHE_VERSION }}
	- name: Install Dependencies
	run: pip install -e '.[testing]' --extra-index-url https://download.pytorch.org/whl/cpu
	#- name: Run Pytest
	# run: python -m pytest -n=auto test/ -k "not (test_efficientnet and models/test_train.py)" --durations=20
	- name: Compile EfficientNet to C
	run: PYTHONPATH="." CLANG=1 python examples/compile_efficientnet.py > recognize.c
	- name: Compile C to native
	run: clang -O2 recognize.c -lm -o recognize
	- name: Test EfficientNet
	run: cat test/models/efficientnet/Chicken.jpg \| ./recognize \| grep cock

	testopencl:
	strategy:
	fail-fast: false
	matrix:
	task: [optimage, openpilot, onnx]
	name: ${{ matrix.task=='optimage'&&'GPU OPT and IMAGE Tests' \|\| matrix.task=='openpilot'&&'openpilot (OpenCL) Tests' \|\| matrix.task=='onnx'&&'ONNX+Optimization Tests' }}
	runs-on: ubuntu-20.04
	timeout-minutes: 20

	steps:
	- name: Checkout Code
	uses: actions/checkout@v4
	- name: Install OpenCL
	run: \|
	echo 'Acquire::http::Pipeline-Depth "5";' \| sudo tee -a /etc/apt/apt.conf.d/99parallel
	echo "deb [ allow-insecure=yes ] https://apt.repos.intel.com/oneapi all main" \| sudo tee /etc/apt/sources.list.d/oneAPI.list
	sudo apt update
	sudo apt install --allow-unauthenticated -y --no-install-recommends \
	intel-oneapi-runtime-openmp=2023.2.1-16 intel-oneapi-runtime-compilers-common=2023.2.1-16 intel-oneapi-runtime-compilers=2023.2.1-16 \
	intel-oneapi-runtime-dpcpp-sycl-opencl-cpu=2023.2.1-16 intel-oneapi-runtime-tbb-common=2021.10.0-49541 \
	intel-oneapi-runtime-tbb=2021.10.0-49541 intel-oneapi-runtime-opencl=2023.2.1-16
	- 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: ${{ env.Python3_ROOT_DIR }}/lib/python3.11/site-packages
	key: testing-packages-${{ hashFiles('**/setup.py') }}
	- name: Cache downloads
	uses: actions/cache@v4
	with:
	path: ~/.cache/tinygrad/downloads/
	key: downloads-cache-${{ matrix.task }}-${{ env.DOWNLOAD_CACHE_VERSION }}
	- name: Install Dependencies
	run: pip install -e '.[testing,testing_tf]' --extra-index-url https://download.pytorch.org/whl/cpu
	- if: ${{ matrix.task == 'optimage' }}
	name: Run Optimizer Test (OPT 2 and 3)
	run: \|
	PYTHONPATH="." OPT=2 GPU=1 python -m pytest -n=auto test/external/external_test_opt.py
	PYTHONPATH="." OPT=3 GPU=1 python -m pytest -n=auto test/external/external_test_opt.py
	- if: ${{ matrix.task == 'optimage'}}
	name: Test WINO=1
	run: GPU=1 DEBUG=2 WINO=1 python3 test/test_ops.py TestOps.test_simple_conv2d
	- if: ${{ matrix.task == 'optimage'}}
	name: Test GPU IMAGE=1 ops
	run: GPU=1 IMAGE=1 python -m pytest -n=auto test/test_ops.py
	- if: ${{ matrix.task == 'optimage'}}
	name: Test GPU IMAGE=2 ops
	run: GPU=1 IMAGE=2 python -m pytest -n=auto test/test_ops.py
	- if: ${{ matrix.task == 'openpilot' }}
	name: Test openpilot model compile and size
	run: \|
	DEBUG=2 ALLOWED_KERNEL_COUNT=208 FLOAT16=1 DEBUGCL=1 GPU=1 IMAGE=2 python openpilot/compile2.py
	#python -c 'import os; assert os.path.getsize("/tmp/output.thneed") < 100_000_000'
	- if: ${{ matrix.task == 'openpilot' }}
	name: Test openpilot model correctness (float32)
	run: FLOAT16=0 DEBUGCL=1 GPU=1 IMAGE=2 python openpilot/compile2.py
	- if: ${{ matrix.task == 'openpilot' }}
	name: Test openpilot alt model correctness (float32)
	run: FLOAT16=0 DEBUGCL=1 GPU=1 IMAGE=2 python openpilot/compile2.py https://github.com/commaai/openpilot/raw/3799fe46b3a629e491d4b8498b8ae83e4c88c304/selfdrive/modeld/models/supercombo.onnx
	- if: ${{ matrix.task == 'openpilot' }}
	name: Test openpilot fastvits model correctness (float32)
	run: FLOAT16=0 DEBUGCL=1 GPU=1 IMAGE=2 python openpilot/compile2.py https://github.com/commaai/openpilot/raw/9118973ed03c1ae1d40cf69a29507ec2cc78efd7/selfdrive/modeld/models/supercombo.onnx
	- if: ${{ matrix.task == 'onnx' }}
	name: Test ONNX (GPU)
	run: GPU=1 python -m pytest -n=auto test/external/external_test_onnx_backend.py --durations=20
	- if: ${{ matrix.task == 'onnx' }}
	name: Test ONNX (CLANG)
	run: CLANG=1 python -m pytest -n=auto test/external/external_test_onnx_backend.py --durations=20
	- if: ${{ matrix.task == 'onnx' }}
	name: Test Action Space
	run: PYTHONPATH="." GPU=1 python3 extra/optimization/get_action_space.py
	- if: ${{ matrix.task == 'onnx' }}
	name: Test Beam Search
	run: PYTHONPATH="." GPU=1 IGNORE_BEAM_CACHE=1 python3 -m pytest extra/optimization/test_beam_search.py
	- if: ${{ matrix.task == 'onnx' }}
	name: Test MLPerf optimizers
	run: GPU=1 python -m pytest -n=auto test/external/external_test_optim.py --durations=20
	- if: ${{ matrix.task == 'onnx' }}
	name: Test THREEFRY
	run: PYTHONPATH=. THREEFRY=1 GPU=1 python3 -m pytest test/test_randomness.py test/test_jit.py

	#testwebgpu:
	# name: WebGPU Tests
	# runs-on: macos-13
	# timeout-minutes: 20
	# steps:
	# - name: Checkout Code
	# uses: actions/checkout@v4
	# - 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: webgpu-testing-user3-packages-${{ hashFiles('**/setup.py') }}
	# - name: Install Dependencies
	# run: pip install --user -e '.[webgpu,testing]' --extra-index-url https://download.pytorch.org/whl/cpu
	# - name: Cache downloads
	# uses: actions/cache@v4
	# with:
	# path: ~/Library/Caches/tinygrad/downloads/
	# key: downloads-cache-webgpu-${{ env.DOWNLOAD_CACHE_VERSION }}
	# - name: Check Device.DEFAULT (WEBGPU) and print some source
	# run: \|
	# WEBGPU=1 python -c "from tinygrad import Device; assert Device.DEFAULT == 'WEBGPU', Device.DEFAULT"
	# WEBGPU=1 DEBUG=4 FORWARD_ONLY=1 python3 test/test_ops.py TestOps.test_add
	#- name: Run webgpu pytest
	# run: WEBGPU=1 WGPU_BACKEND_TYPE=Metal python -m pytest -n=auto
	# - name: Run selected webgpu tests
	# run: \|
	# WEBGPU=1 WGPU_BACKEND_TYPE=Metal python -m pytest -n=auto test/test_ops.py test/test_dtype.py \
	# test/test_jit.py test/test_symbolic_ops.py test/test_symbolic_jit.py test/test_linearizer.py \
	# test/test_linearizer_failures.py test/test_nn.py
	# - name: Build WEBGPU Efficientnet
	# run: WEBGPU=1 WGPU_BACKEND_TYPE=Metal python -m examples.compile_efficientnet
	# - name: Install Puppeteer
	# run: npm install puppeteer
	# - name: Run WEBGPU Efficientnet
	# run: node test/web/test_webgpu.js

	testmetal:
	name: Metal Tests
	runs-on: macos-14
	timeout-minutes: 20

	steps:
	- name: Checkout Code
	uses: actions/checkout@v4
	- 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: metal-m1-testing-user3-packages-${{ hashFiles('**/setup.py') }}
	- name: Install Dependencies
	run: pip install --user -e '.[testing]' --extra-index-url https://download.pytorch.org/whl/cpu
	- name: Cache downloads
	uses: actions/cache@v4
	with:
	path: ~/Library/Caches/tinygrad/downloads/
	key: downloads-cache-metal-only-${{ env.DOWNLOAD_CACHE_VERSION }}
	- name: Check Device.DEFAULT (METAL) and print some source
	run: \|
	METAL=1 python -c "from tinygrad import Device; assert Device.DEFAULT == 'METAL', Device.DEFAULT"
	METAL=1 DEBUG=4 FORWARD_ONLY=1 python3 test/test_ops.py TestOps.test_add
	- name: Run metal test
	run: JIT=2 METAL=1 python -m pytest -n=auto test/ --ignore=test/external --ignore=test/models --durations=20
	- name: Run real world test
	run: JIT=2 METAL=1 python -m pytest -n=auto test/models/test_real_world.py --durations=20
	- name: Run ONNX
	run: JIT=2 METAL=1 python -m pytest -n=auto test/external/external_test_onnx_backend.py --durations=20
	- name: Test tensor core ops (fake)
	run: TC=2 METAL=1 DEBUG=3 python test/test_ops.py TestOps.test_gemm
	- name: Test tensor core ops (real)
	run: METAL=1 DEBUG=3 python test/test_ops.py TestOps.test_big_gemm
	- name: Test LLaMA compile speed
	run: PYTHONPATH="." METAL=1 python test/external/external_test_speed_llama.py
	- name: Test Beam Search
	run: PYTHONPATH="." METAL=1 IGNORE_BEAM_CACHE=1 python3 -m pytest extra/optimization/test_beam_search.py
	- name: Fuzz Test linearizer
	run: PYTHONPATH="." METAL=1 CACHELEVEL=0 FUZZ_ALL_ACTIONS=1 DEPTH=2 FUZZ_N=48 FUZZ_MAX_SIZE=10000000 python test/external/fuzz_linearizer.py


	# testwebgl:
	# name: WebGL Tests
	# runs-on: ubuntu-latest
	# timeout-minutes: 20
	#
	# steps:
	# - name: Checkout Code
	# uses: actions/checkout@v3
	# - name: Set up Python 3.11
	# uses: actions/setup-python@v4
	# with:
	# python-version: 3.11
	# - name: Cache python packages
	# uses: actions/cache@v4
	# with:
	# path: ${{ env.Python3_ROOT_DIR }}/lib/python3.11/site-packages
	# key: webgl-testing-packages-${{ hashFiles('**/setup.py') }}
	# - name: Install Dependencies
	# run: pip install -e '.[webgl,testing]' --extra-index-url https://download.pytorch.org/whl/cpu
	# - name: Cache downloads
	# uses: actions/cache@v4
	# with:
	# path: ~/Library/Caches/tinygrad/downloads/
	# key: downloads-cache-webgl-${{ env.DOWNLOAD_CACHE_VERSION }}
	# - name: Prepare
	# run: \|
	# sudo apt-get -y install xvfb
	# sudo /usr/bin/Xvfb :0 -screen 0 4096x4096x24+32 &
	# - name: Run selected webgl tests
	# run: WEBGL=1 python -m pytest -n=auto test/test_ops.py test/test_dtype.py test/test_jit.py
	# - name: Build WebGL Efficientnet
	# run: WEBGL=1 python -m examples.compile_efficientnet

	tests:
	strategy:
	fail-fast: false
	matrix:
	backend: [llvm, clang, gpu, cuda, hip, ptx] #, triton]

	name: Tests on (${{ matrix.backend }})
	runs-on: ubuntu-latest
	timeout-minutes: 20

	steps:
	- name: Checkout Code
	uses: actions/checkout@v4
	- 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: ${{ env.Python3_ROOT_DIR }}/lib/python3.11/site-packages
	key: ${{ matrix.backend }}-packages-${{ hashFiles('**/setup.py') }}
	- name: Cache downloads
	uses: actions/cache@v4
	with:
	path: ~/.cache/tinygrad/downloads/
	key: downloads-cache-${{ matrix.backend }}-${{ env.DOWNLOAD_CACHE_VERSION }}
	- name: Set env
	run: printf "${{ matrix.backend == 'llvm' && 'LLVM=1' \|\| matrix.backend == 'clang' && 'CLANG=1' \|\| matrix.backend == 'gpu' && 'GPU=1' \|\| matrix.backend == 'cuda' && 'FORWARD_ONLY=1\nJIT=1\nOPT=2\nCUDA=1\nCUDACPU=1\n' \|\| matrix.backend == 'PTX' && 'FORWARD_ONLY=1\nJIT=1\nOPT=2\nCUDA=1\nCUDACPU=1\nPTX=1' \|\| matrix.backend == 'triton' && 'FORWARD_ONLY=1\nJIT=1\nOPT=2\nCUDA=1\nCUDACPU=1\nTRITON=1\nTRITON_PTXAS_PATH=/usr/bin/ptxas' \|\| matrix.backend == 'hip' && 'RHIP=1\nFORWARD_ONLY=1' }}" >> $GITHUB_ENV
	- name: Install OpenCL
	if: matrix.backend == 'gpu'
	run: \|
	echo 'Acquire::http::Pipeline-Depth "5";' \| sudo tee -a /etc/apt/apt.conf.d/99parallel
	echo "deb [ allow-insecure=yes ] https://apt.repos.intel.com/oneapi all main" \| sudo tee /etc/apt/sources.list.d/oneAPI.list
	sudo apt update
	sudo apt install --allow-unauthenticated -y --no-install-recommends opencl-headers \
	intel-oneapi-runtime-openmp=2023.2.1-16 intel-oneapi-runtime-compilers-common=2023.2.1-16 intel-oneapi-runtime-compilers=2023.2.1-16 \
	intel-oneapi-runtime-dpcpp-sycl-opencl-cpu=2023.2.1-16 intel-oneapi-runtime-tbb-common=2021.10.0-49541 \
	intel-oneapi-runtime-tbb=2021.10.0-49541 intel-oneapi-runtime-opencl=2023.2.1-16
	- name: Install packages (cuda)
	if: matrix.backend == 'cuda' \|\| matrix.backend == 'ptx' \|\| matrix.backend == 'triton'
	run: \|
	echo 'Acquire::http::Pipeline-Depth "5";' \| sudo tee -a /etc/apt/apt.conf.d/99parallel
	sudo apt update -y
	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: Cache gpuocelot
	if: matrix.backend == 'cuda' \|\| matrix.backend == 'ptx' \|\| matrix.backend == 'triton'
	id: cache-build
	uses: actions/cache@v4
	env:
	cache-name: cache-gpuocelot-build
	with:
	path: ${{ github.workspace }}/gpuocelot/ocelot
	key: ubuntu22.04-gpuocelot-18401f4245b27ca4b3af433196583cc81ef84480-rebuild-5
	- name: Clone/compile gpuocelot
	if: (matrix.backend == 'cuda' \|\| matrix.backend == 'ptx' \|\| matrix.backend == 'triton') && 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/ocelot
	git checkout 18401f4245b27ca4b3af433196583cc81ef84480
	mkdir build
	cd build
	cmake .. -Wno-dev -G Ninja -DOCELOT_BUILD_TOOLS=OFF -DCMAKE_BUILD_ALWAYS=0 -DBUILD_TESTS_CUDA=OFF
	ninja
	- name: Install gpuocelot
	if: matrix.backend == 'cuda' \|\| matrix.backend == 'ptx' \|\| matrix.backend == 'triton'
	run: \|
	cd ${{ github.workspace }}/gpuocelot/ocelot/build
	sudo ninja install -d explain
	- name: Install packages (hip)
	if: matrix.backend == 'hip'
	run: \|
	echo 'Acquire::http::Pipeline-Depth "5";' \| sudo tee -a /etc/apt/apt.conf.d/99parallel
	wget https://repo.radeon.com/rocm/rocm.gpg.key -O - \| gpg --dearmor \| sudo tee /etc/apt/keyrings/rocm.gpg > /dev/null
	sudo tee /etc/apt/sources.list.d/rocm.list <<'EOF'
	deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/rocm/apt/6.0.2 jammy main
	EOF
	echo -e 'Package: *\nPin: release o=repo.radeon.com\nPin-Priority: 600' \| sudo tee /etc/apt/preferences.d/rocm-pin-600
	sudo apt update
	sudo apt install --no-install-recommends --allow-unauthenticated -y hsa-rocr comgr rocm-llvm hsa-rocr-dev
	curl -s https://api.github.com/repos/Qazalin/remu/releases/latest \| \
	jq -r '.assets[] \| select(.name == "libremu.so").browser_download_url' \| \
	sudo xargs curl -L -o /usr/local/lib/libremu.so
	- name: Install dependencies
	run: pip install -e '.[testing${{matrix.backend=='llvm'&&',llvm'\|\|matrix.backend=='cuda'&&',cuda'\|\|matrix.backend=='ptx'&&',cuda'\|\|matrix.backend=='triton'&&',triton'\|\|''}}]' --extra-index-url https://download.pytorch.org/whl/cpu --extra-index-url https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/Triton-Nightly/pypi/simple/
	- name: Check Device.DEFAULT and print some source
	run: \|
	python -c "from tinygrad import Device; assert Device.DEFAULT in ['LLVM','CLANG','CUDA','GPU','RHIP'], Device.DEFAULT"
	DEBUG=5 PYTHONPATH=${{ github.workspace }} FORWARD_ONLY=1 python3 test/test_ops.py TestOps.test_add
	- name: Verify OpenCL autogen
	if: matrix.backend == 'gpu'
	run: \|
	cp tinygrad/runtime/autogen/opencl.py /tmp/opencl.py.bak
	./autogen_stubs.sh opencl
	diff /tmp/opencl.py.bak tinygrad/runtime/autogen/opencl.py
	- name: Verify CUDA autogen
	if: matrix.backend == 'cuda'
	run: \|
	cp tinygrad/runtime/autogen/cuda.py /tmp/cuda.py.bak
	./autogen_stubs.sh cuda
	diff /tmp/cuda.py.bak tinygrad/runtime/autogen/cuda.py
	- name: Verify HIP autogen
	if: matrix.backend == 'hip'
	run: \|
	cp tinygrad/runtime/autogen/hsa.py /tmp/hsa.py.bak
	cp tinygrad/runtime/autogen/comgr.py /tmp/comgr.py.bak
	./autogen_stubs.sh hsa
	./autogen_stubs.sh comgr
	diff /tmp/hsa.py.bak tinygrad/runtime/autogen/hsa.py
	diff /tmp/comgr.py.bak tinygrad/runtime/autogen/comgr.py
	- name: Run pytest (not cuda or hip)
	if: matrix.backend!='cuda' && matrix.backend!='ptx' && matrix.backend!='triton' && matrix.backend != 'hip'
	run: python -m pytest -n=auto test/ --durations=20
	- name: Run ONNX (only LLVM)
	if: matrix.backend == 'llvm'
	run: python -m pytest -n=auto test/external/external_test_onnx_backend.py --durations=20
	- name: Run pytest (cuda)
	if: matrix.backend=='cuda'\|\|matrix.backend=='ptx'\|\|matrix.backend=='triton'
	run: python -m pytest -n=auto test/ -k 'not (half or test_efficientnet_safetensors)' --ignore=test/external --ignore=test/models --durations=20
	- name: Run pytest (hip)
	if: matrix.backend=='hip'
	run: python -m pytest -n=auto test/test_ops.py test/test_dtype.py test/test_dtype_alu.py test/test_linearizer.py test/test_randomness.py test/imported/test_indexing.py test/external/external_test_hip_compile.py --durations=20

	#testunicorn:
	# name: ARM64 unicorn Test
	# runs-on: ubuntu-latest
	# timeout-minutes: 20
	# steps:
	# - name: Checkout Code
	# uses: actions/checkout@v4
	# - 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: ${{ env.Python3_ROOT_DIR }}/lib/python3.11/site-packages
	# key: testing-arm-packages-${{ hashFiles('**/setup.py') }}
	# - name: Install cross-assembler
	# run: \|
	# sudo apt update -y
	# sudo apt install -y --no-install-recommends gcc-aarch64-linux-gnu
	# - name: Install dependencies
	# 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

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Multireduce Kernels - prereq refactor (#4173) #1

Workflow file

Multireduce Kernels - prereq refactor (#4173) #1

Jobs

Run details

Workflow file for this run