torch-mlir/projects/CMakeLists.txt

82 lines
3.0 KiB
CMake
Raw Permalink Normal View History

include(AddMLIRPython)
if(TORCH_MLIR_ENABLE_ONNX_C_IMPORTER)
add_subdirectory(onnx_c_importer)
endif()
Upstream the ONNX importer. (#2636) This is part 1 of 2, which will also include upstreaming the FX importer. I started with ONNX because it forces some project layout updates and is more self contained/easier as a first step. Deviating somewhat from the RFCs on project layout, I made the following decisions: * Locating the `onnx_importer.py` into `torch_mlir.extras` as Maks already has opened up that namespace and it seemed to fit. Better to have fewer things at that level. * Setup the build so that the root project only contains MLIR Python and pure Python deps (like the importers), but this can be augmented with the `projects/` adding more depending on which features are enabled. * The default build continues to build everything whereas in `TORCH_MLIR_ENABLE_ONLY_MLIR_PYTHON_BINDINGS=1` mode, it builds a `torch-mlir-core` wheel with the pure contents only. `onnx_importer.py` and `importer_smoke_test.py` are almost verbatim copies from SHARK-Turbine. I made some minor local alterations to adapt to paths and generalize the way they interact with the outer project. I expect I can copy these back to Turbine verbatim from here. I also updated the license boilerplate (they have the same license but slightly different project norms for the headers) but retained the correct copyright. Other updates: * Added the ONNX importer unit test (which also can generate test data) in lit, conditioned on the availability of the Python `onnx` package. In a followup once I know everything is stable, I'll add another env var that the CI can set to always enable this so we know conclusively if tests pass. * Moved the ONNX conversion readme to `docs/`. * Renamed CMake option `TORCH_MLIR_ENABLE_ONLY_MLIR_PYTHON_BINDINGS` -> `TORCH_MLIR_ENABLE_PYTORCH_EXTENSIONS` and inverted the sense. Made the JitIR importer and LTC options `cmake_dependent_options` for robustness.
2023-12-13 11:02:51 +08:00
################################################################################
# PyTorch
# Configure PyTorch if we have any features enabled which require it.
Upstream the ONNX importer. (#2636) This is part 1 of 2, which will also include upstreaming the FX importer. I started with ONNX because it forces some project layout updates and is more self contained/easier as a first step. Deviating somewhat from the RFCs on project layout, I made the following decisions: * Locating the `onnx_importer.py` into `torch_mlir.extras` as Maks already has opened up that namespace and it seemed to fit. Better to have fewer things at that level. * Setup the build so that the root project only contains MLIR Python and pure Python deps (like the importers), but this can be augmented with the `projects/` adding more depending on which features are enabled. * The default build continues to build everything whereas in `TORCH_MLIR_ENABLE_ONLY_MLIR_PYTHON_BINDINGS=1` mode, it builds a `torch-mlir-core` wheel with the pure contents only. `onnx_importer.py` and `importer_smoke_test.py` are almost verbatim copies from SHARK-Turbine. I made some minor local alterations to adapt to paths and generalize the way they interact with the outer project. I expect I can copy these back to Turbine verbatim from here. I also updated the license boilerplate (they have the same license but slightly different project norms for the headers) but retained the correct copyright. Other updates: * Added the ONNX importer unit test (which also can generate test data) in lit, conditioned on the availability of the Python `onnx` package. In a followup once I know everything is stable, I'll add another env var that the CI can set to always enable this so we know conclusively if tests pass. * Moved the ONNX conversion readme to `docs/`. * Renamed CMake option `TORCH_MLIR_ENABLE_ONLY_MLIR_PYTHON_BINDINGS` -> `TORCH_MLIR_ENABLE_PYTORCH_EXTENSIONS` and inverted the sense. Made the JitIR importer and LTC options `cmake_dependent_options` for robustness.
2023-12-13 11:02:51 +08:00
################################################################################
if(TORCH_MLIR_ENABLE_JIT_IR_IMPORTER OR TORCH_MLIR_ENABLE_LTC)
Upstream the ONNX importer. (#2636) This is part 1 of 2, which will also include upstreaming the FX importer. I started with ONNX because it forces some project layout updates and is more self contained/easier as a first step. Deviating somewhat from the RFCs on project layout, I made the following decisions: * Locating the `onnx_importer.py` into `torch_mlir.extras` as Maks already has opened up that namespace and it seemed to fit. Better to have fewer things at that level. * Setup the build so that the root project only contains MLIR Python and pure Python deps (like the importers), but this can be augmented with the `projects/` adding more depending on which features are enabled. * The default build continues to build everything whereas in `TORCH_MLIR_ENABLE_ONLY_MLIR_PYTHON_BINDINGS=1` mode, it builds a `torch-mlir-core` wheel with the pure contents only. `onnx_importer.py` and `importer_smoke_test.py` are almost verbatim copies from SHARK-Turbine. I made some minor local alterations to adapt to paths and generalize the way they interact with the outer project. I expect I can copy these back to Turbine verbatim from here. I also updated the license boilerplate (they have the same license but slightly different project norms for the headers) but retained the correct copyright. Other updates: * Added the ONNX importer unit test (which also can generate test data) in lit, conditioned on the availability of the Python `onnx` package. In a followup once I know everything is stable, I'll add another env var that the CI can set to always enable this so we know conclusively if tests pass. * Moved the ONNX conversion readme to `docs/`. * Renamed CMake option `TORCH_MLIR_ENABLE_ONLY_MLIR_PYTHON_BINDINGS` -> `TORCH_MLIR_ENABLE_PYTORCH_EXTENSIONS` and inverted the sense. Made the JitIR importer and LTC options `cmake_dependent_options` for robustness.
2023-12-13 11:02:51 +08:00
if (NOT TORCH_MLIR_USE_INSTALLED_PYTORCH)
# Source builds
message(STATUS "Building libtorch from source (features depend on it and NOT TORCH_MLIR_USE_INSTALLED_PYTORCH)")
set(ENV{TORCH_MLIR_SRC_PYTORCH_REPO} ${TORCH_MLIR_SRC_PYTORCH_REPO})
set(ENV{TORCH_MLIR_SRC_PYTORCH_BRANCH} ${TORCH_MLIR_SRC_PYTORCH_BRANCH})
set(ENV{TM_PYTORCH_INSTALL_WITHOUT_REBUILD} ${TM_PYTORCH_INSTALL_WITHOUT_REBUILD})
set(ENV{MACOSX_DEPLOYMENT_TARGET} ${MACOSX_DEPLOYMENT_TARGET})
set(ENV{CMAKE_OSX_ARCHITECTURES} ${CMAKE_OSX_ARCHITECTURES})
set(ENV{CMAKE_C_COMPILER_LAUNCHER} ${CMAKE_C_COMPILER_LAUNCHER})
set(ENV{CMAKE_CXX_COMPILER_LAUNCHER} ${CMAKE_CXX_COMPILER_LAUNCHER})
execute_process(
COMMAND ${TORCH_MLIR_SOURCE_DIR}/build_tools/build_libtorch.sh
RESULT_VARIABLE _result
)
if(_result)
message(FATAL_ERROR "Failed to run `build_libtorch.sh`")
endif()
set(TORCH_INSTALL_PREFIX "libtorch")
endif()
message(STATUS "Enabling PyTorch C++ dep (features depend on it)")
include(TorchMLIRPyTorch)
TorchMLIRProbeForPyTorchInstall()
if(TORCH_MLIR_USE_INSTALLED_PYTORCH)
TorchMLIRConfigurePyTorch()
else()
# Assume it is a sibling to the overall project.
set(Torch_DIR "${PROJECT_SOURCE_DIR}/../libtorch/share/cmake/Torch")
message(STATUS "Attempting to locate libtorch as a sibling to the project: ${Torch_DIR}")
if(NOT EXISTS "${Torch_DIR}")
message(FATAL_ERROR "Without TORCH_MLIR_USE_INSTALLED_PYTORCH, expected to find Torch configuration at ${Torch_DIR}, which does not exist")
endif()
endif()
find_package(Torch 1.11 REQUIRED)
set(TORCHGEN_DIR ${Torch_ROOT}/../../../torchgen)
include_directories(BEFORE
${TORCH_INCLUDE_DIRS}
${Python3_INCLUDE_DIRS}
)
link_directories("${TORCH_INSTALL_PREFIX}/lib")
message(STATUS "TORCH_CXXFLAGS is = ${TORCH_CXXFLAGS}")
if(${CMAKE_SYSTEM_NAME} STREQUAL "Linux" AND NOT TORCH_CXXFLAGS)
message(WARNING
"When building on Linux TORCH_CXXFLAGS are almost always required but were not detected. "
"It is very likely this this will produce a non-functional installation. "
"See notes in build_tools/cmake/TorchMLIRPyTorch.cmake.")
endif()
message(STATUS "TORCH_LIBRARIES = ${TORCH_LIBRARIES}")
endif()
# Include jit_ir_common if the jit_ir importer or LTC is enabled,
# since they both require it.
if(TORCH_MLIR_ENABLE_JIT_IR_IMPORTER OR TORCH_MLIR_ENABLE_LTC)
add_subdirectory(jit_ir_common)
endif()
# Include LTC.
if(TORCH_MLIR_ENABLE_LTC)
add_subdirectory(ltc)
endif()
# Include overall PT1 project.
Upstream the ONNX importer. (#2636) This is part 1 of 2, which will also include upstreaming the FX importer. I started with ONNX because it forces some project layout updates and is more self contained/easier as a first step. Deviating somewhat from the RFCs on project layout, I made the following decisions: * Locating the `onnx_importer.py` into `torch_mlir.extras` as Maks already has opened up that namespace and it seemed to fit. Better to have fewer things at that level. * Setup the build so that the root project only contains MLIR Python and pure Python deps (like the importers), but this can be augmented with the `projects/` adding more depending on which features are enabled. * The default build continues to build everything whereas in `TORCH_MLIR_ENABLE_ONLY_MLIR_PYTHON_BINDINGS=1` mode, it builds a `torch-mlir-core` wheel with the pure contents only. `onnx_importer.py` and `importer_smoke_test.py` are almost verbatim copies from SHARK-Turbine. I made some minor local alterations to adapt to paths and generalize the way they interact with the outer project. I expect I can copy these back to Turbine verbatim from here. I also updated the license boilerplate (they have the same license but slightly different project norms for the headers) but retained the correct copyright. Other updates: * Added the ONNX importer unit test (which also can generate test data) in lit, conditioned on the availability of the Python `onnx` package. In a followup once I know everything is stable, I'll add another env var that the CI can set to always enable this so we know conclusively if tests pass. * Moved the ONNX conversion readme to `docs/`. * Renamed CMake option `TORCH_MLIR_ENABLE_ONLY_MLIR_PYTHON_BINDINGS` -> `TORCH_MLIR_ENABLE_PYTORCH_EXTENSIONS` and inverted the sense. Made the JitIR importer and LTC options `cmake_dependent_options` for robustness.
2023-12-13 11:02:51 +08:00
if(TORCH_MLIR_ENABLE_PYTORCH_EXTENSIONS)
add_subdirectory(pt1)
endif()