llvm
/
torch-mlir
mirror of https://github.com/llvm/torch-mlir
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
torch-mlir/build_tools/update_torch_ods.sh

15 lines
622 B
Bash
Raw Normal View History

[torch-mlir earthmoving (2/N)] Python code movement. This moves the bulk of the Python code (including the Torch interop) from `frontends/pytorch` into `torch-mlir/TorchPlugin`. This also required reconciling a bunch of other Python-related stuff, like the `torch` dialects. As I did this, it was simpler to just remove all the old numpy/basicpy stuff because we were going to delete it anyway and it was faster than debugging an intermediate state that would only last O(days) anyway. torch-mlir has two top-level python packages (built into the `python_packages` directory): - `torch_mlir_dialects`: `torch` dialect Python bindings (does not depend on PyTorch). This also involves building the aggregate CAPI for `torch-mlir`. - `torch_mlir`: bindings to the part of the code that links against PyTorch (or C++ code that transitively does). Additionally, there remain two more Python packages in npcomp (but outside `torch-mlir`): - `npcomp_torch`: Contains the e2e test framework and testing configs that plug into RefBackend and IREE. - `npcomp_core`: Contains the low-level interfaces to RefBackend and IREE that `npcomp_torch` uses, along with its own `MLIR_PYTHON_PACKAGE_PREFIX=npcomp.` aggregation of the core MLIR python bindings. (all other functionality has been stripped out) After all the basicpy/numpy deletions, the `npcomp` C++ code is now very tiny. It basically just contains RefBackend and the `TorchConversion` dialect/passes (e.g. `TorchToLinalg.cpp`). Correspondingly, there are now 4 main testing targets paralleling the Python layering (which is reflective of the deeper underlying dependency structure) - `check-torch-mlir`: checks the `torch-mlir` pure MLIR C++ code. - `check-torch-mlir-plugin`: checks the code in `TorchPlugin` (e.g. TorchScript import) - `check-frontends-pytorch`: Checks the little code we have in `frontends/pytorch` -- mainly things related to the e2e framework itself. - `check-npcomp`: Checks the pure MLIR C++ code inside npcomp. There is a target `check-npcomp-all` that runs all of them. The `torch-mlir/build_standalone.sh` script does a standalone build of `torch-mlir`. The e2e tests (`tools/torchscript_e2e_test.sh`) are working too. The update_torch_ods script now lives in `torch-mlir/build_tools/update_torch_ods.sh` and expects a standalone build. This change also required a fix upstream related to cross-shlib Python dependencies, so we also update llvm-project to 8dca953dd39c0cd8c80decbeb38753f58a4de580 to get https://reviews.llvm.org/D109776 (no other fixes were needed for the integrate, thankfully). This completes most of the large source code changes. Next will be bringing the CI/packaging/examples back to life.
2021-09-11 02:44:38 +08:00
#!/bin/bash
# Updates auto-generated ODS files for the `torch` dialect.
Reduce compilation time for TorchOps.cpp.inc The `assemblyFormat` stuff (which generates unrolled, per-op C++ code) was taking up a lot of compile time, and all the ops are essentially printed with the same logic. So this PR makes them all call the same helper function. This is done by using `let hasCustomAssemblyFormat = 1` and then implementing `FooOp::parse` and `FooOp::print`. Additionally, the `Generated*Ops.td` files are all collapsed into just `GeneratedTorchOps.td` (there is no reason to have the files separate, since the files are very large anyway so one is always having to search within them -- editors don't care that the file to search is now a bit bigger :) ). This reduces TorchOpsODSGenerated.cpp compile time (which is now GeneratedTorchOps.cpp) from 39 to 31 seconds on my machine. This is actually less than I expected, but this PR is an overall cleanup to the code anyway. The next step will be to introduce (better) functionality upstream for sharding the TorchOps.cpp.inc file, so that we can truly parallelize the O(#ops) costs. This is also necessary, because after this PR, TorchDialect.cpp is now the slowest file to compile, due to the `addOperations<... all the ops ...>` call, which needs to be shareded too.
2022-03-19 05:04:47 +08:00
set -euo pipefail
[torch-mlir earthmoving (2/N)] Python code movement. This moves the bulk of the Python code (including the Torch interop) from `frontends/pytorch` into `torch-mlir/TorchPlugin`. This also required reconciling a bunch of other Python-related stuff, like the `torch` dialects. As I did this, it was simpler to just remove all the old numpy/basicpy stuff because we were going to delete it anyway and it was faster than debugging an intermediate state that would only last O(days) anyway. torch-mlir has two top-level python packages (built into the `python_packages` directory): - `torch_mlir_dialects`: `torch` dialect Python bindings (does not depend on PyTorch). This also involves building the aggregate CAPI for `torch-mlir`. - `torch_mlir`: bindings to the part of the code that links against PyTorch (or C++ code that transitively does). Additionally, there remain two more Python packages in npcomp (but outside `torch-mlir`): - `npcomp_torch`: Contains the e2e test framework and testing configs that plug into RefBackend and IREE. - `npcomp_core`: Contains the low-level interfaces to RefBackend and IREE that `npcomp_torch` uses, along with its own `MLIR_PYTHON_PACKAGE_PREFIX=npcomp.` aggregation of the core MLIR python bindings. (all other functionality has been stripped out) After all the basicpy/numpy deletions, the `npcomp` C++ code is now very tiny. It basically just contains RefBackend and the `TorchConversion` dialect/passes (e.g. `TorchToLinalg.cpp`). Correspondingly, there are now 4 main testing targets paralleling the Python layering (which is reflective of the deeper underlying dependency structure) - `check-torch-mlir`: checks the `torch-mlir` pure MLIR C++ code. - `check-torch-mlir-plugin`: checks the code in `TorchPlugin` (e.g. TorchScript import) - `check-frontends-pytorch`: Checks the little code we have in `frontends/pytorch` -- mainly things related to the e2e framework itself. - `check-npcomp`: Checks the pure MLIR C++ code inside npcomp. There is a target `check-npcomp-all` that runs all of them. The `torch-mlir/build_standalone.sh` script does a standalone build of `torch-mlir`. The e2e tests (`tools/torchscript_e2e_test.sh`) are working too. The update_torch_ods script now lives in `torch-mlir/build_tools/update_torch_ods.sh` and expects a standalone build. This change also required a fix upstream related to cross-shlib Python dependencies, so we also update llvm-project to 8dca953dd39c0cd8c80decbeb38753f58a4de580 to get https://reviews.llvm.org/D109776 (no other fixes were needed for the integrate, thankfully). This completes most of the large source code changes. Next will be bringing the CI/packaging/examples back to life.
2021-09-11 02:44:38 +08:00
Fix update_torch_ods.sh
2021-09-30 07:52:20 +08:00
src_dir="$(realpath $(dirname $0)/..)"
[torch-mlir earthmoving (2/N)] Python code movement. This moves the bulk of the Python code (including the Torch interop) from `frontends/pytorch` into `torch-mlir/TorchPlugin`. This also required reconciling a bunch of other Python-related stuff, like the `torch` dialects. As I did this, it was simpler to just remove all the old numpy/basicpy stuff because we were going to delete it anyway and it was faster than debugging an intermediate state that would only last O(days) anyway. torch-mlir has two top-level python packages (built into the `python_packages` directory): - `torch_mlir_dialects`: `torch` dialect Python bindings (does not depend on PyTorch). This also involves building the aggregate CAPI for `torch-mlir`. - `torch_mlir`: bindings to the part of the code that links against PyTorch (or C++ code that transitively does). Additionally, there remain two more Python packages in npcomp (but outside `torch-mlir`): - `npcomp_torch`: Contains the e2e test framework and testing configs that plug into RefBackend and IREE. - `npcomp_core`: Contains the low-level interfaces to RefBackend and IREE that `npcomp_torch` uses, along with its own `MLIR_PYTHON_PACKAGE_PREFIX=npcomp.` aggregation of the core MLIR python bindings. (all other functionality has been stripped out) After all the basicpy/numpy deletions, the `npcomp` C++ code is now very tiny. It basically just contains RefBackend and the `TorchConversion` dialect/passes (e.g. `TorchToLinalg.cpp`). Correspondingly, there are now 4 main testing targets paralleling the Python layering (which is reflective of the deeper underlying dependency structure) - `check-torch-mlir`: checks the `torch-mlir` pure MLIR C++ code. - `check-torch-mlir-plugin`: checks the code in `TorchPlugin` (e.g. TorchScript import) - `check-frontends-pytorch`: Checks the little code we have in `frontends/pytorch` -- mainly things related to the e2e framework itself. - `check-npcomp`: Checks the pure MLIR C++ code inside npcomp. There is a target `check-npcomp-all` that runs all of them. The `torch-mlir/build_standalone.sh` script does a standalone build of `torch-mlir`. The e2e tests (`tools/torchscript_e2e_test.sh`) are working too. The update_torch_ods script now lives in `torch-mlir/build_tools/update_torch_ods.sh` and expects a standalone build. This change also required a fix upstream related to cross-shlib Python dependencies, so we also update llvm-project to 8dca953dd39c0cd8c80decbeb38753f58a4de580 to get https://reviews.llvm.org/D109776 (no other fixes were needed for the integrate, thankfully). This completes most of the large source code changes. Next will be bringing the CI/packaging/examples back to life.
2021-09-11 02:44:38 +08:00
build_dir="$(realpath "${TORCH_MLIR_BUILD_DIR:-$src_dir/build}")"
Reduce compilation time for TorchOps.cpp.inc The `assemblyFormat` stuff (which generates unrolled, per-op C++ code) was taking up a lot of compile time, and all the ops are essentially printed with the same logic. So this PR makes them all call the same helper function. This is done by using `let hasCustomAssemblyFormat = 1` and then implementing `FooOp::parse` and `FooOp::print`. Additionally, the `Generated*Ops.td` files are all collapsed into just `GeneratedTorchOps.td` (there is no reason to have the files separate, since the files are very large anyway so one is always having to search within them -- editors don't care that the file to search is now a bit bigger :) ). This reduces TorchOpsODSGenerated.cpp compile time (which is now GeneratedTorchOps.cpp) from 39 to 31 seconds on my machine. This is actually less than I expected, but this PR is an overall cleanup to the code anyway. The next step will be to introduce (better) functionality upstream for sharding the TorchOps.cpp.inc file, so that we can truly parallelize the O(#ops) costs. This is also necessary, because after this PR, TorchDialect.cpp is now the slowest file to compile, due to the `addOperations<... all the ops ...>` call, which needs to be shareded too.
2022-03-19 05:04:47 +08:00
torch_ir_include_dir="${src_dir}/include/torch-mlir/Dialect/Torch/IR"
Fix update_torch_ods.sh
2021-09-30 07:52:20 +08:00
python_packages_dir="${build_dir}/tools/torch-mlir/python_packages"
[torch-mlir earthmoving (2/N)] Python code movement. This moves the bulk of the Python code (including the Torch interop) from `frontends/pytorch` into `torch-mlir/TorchPlugin`. This also required reconciling a bunch of other Python-related stuff, like the `torch` dialects. As I did this, it was simpler to just remove all the old numpy/basicpy stuff because we were going to delete it anyway and it was faster than debugging an intermediate state that would only last O(days) anyway. torch-mlir has two top-level python packages (built into the `python_packages` directory): - `torch_mlir_dialects`: `torch` dialect Python bindings (does not depend on PyTorch). This also involves building the aggregate CAPI for `torch-mlir`. - `torch_mlir`: bindings to the part of the code that links against PyTorch (or C++ code that transitively does). Additionally, there remain two more Python packages in npcomp (but outside `torch-mlir`): - `npcomp_torch`: Contains the e2e test framework and testing configs that plug into RefBackend and IREE. - `npcomp_core`: Contains the low-level interfaces to RefBackend and IREE that `npcomp_torch` uses, along with its own `MLIR_PYTHON_PACKAGE_PREFIX=npcomp.` aggregation of the core MLIR python bindings. (all other functionality has been stripped out) After all the basicpy/numpy deletions, the `npcomp` C++ code is now very tiny. It basically just contains RefBackend and the `TorchConversion` dialect/passes (e.g. `TorchToLinalg.cpp`). Correspondingly, there are now 4 main testing targets paralleling the Python layering (which is reflective of the deeper underlying dependency structure) - `check-torch-mlir`: checks the `torch-mlir` pure MLIR C++ code. - `check-torch-mlir-plugin`: checks the code in `TorchPlugin` (e.g. TorchScript import) - `check-frontends-pytorch`: Checks the little code we have in `frontends/pytorch` -- mainly things related to the e2e framework itself. - `check-npcomp`: Checks the pure MLIR C++ code inside npcomp. There is a target `check-npcomp-all` that runs all of them. The `torch-mlir/build_standalone.sh` script does a standalone build of `torch-mlir`. The e2e tests (`tools/torchscript_e2e_test.sh`) are working too. The update_torch_ods script now lives in `torch-mlir/build_tools/update_torch_ods.sh` and expects a standalone build. This change also required a fix upstream related to cross-shlib Python dependencies, so we also update llvm-project to 8dca953dd39c0cd8c80decbeb38753f58a4de580 to get https://reviews.llvm.org/D109776 (no other fixes were needed for the integrate, thankfully). This completes most of the large source code changes. Next will be bringing the CI/packaging/examples back to life.
2021-09-11 02:44:38 +08:00
#ninja -C "${build_dir}"
Implement Python relayout from #311 Fixes https://github.com/llvm/mlir-npcomp/issues/311 The key change is that TorchPlugin is folded into `torch_mlir.dialects.torch.importer.jit_ir` (it imports the PyTorch JIT's IR, so that's a good, scoped name for it). The CMake option `-DTORCH_MLIR_ENABLE_JIT_IR_IMPORTER=OFF` disables it, which allows building without a PyTorch native dependency.
2021-09-21 04:55:36 +08:00
PYTHONPATH="${python_packages_dir}/torch_mlir" python \
-m torch_mlir.dialects.torch.importer.jit_ir.build_tools.torch_ods_gen \
Reduce compilation time for TorchOps.cpp.inc The `assemblyFormat` stuff (which generates unrolled, per-op C++ code) was taking up a lot of compile time, and all the ops are essentially printed with the same logic. So this PR makes them all call the same helper function. This is done by using `let hasCustomAssemblyFormat = 1` and then implementing `FooOp::parse` and `FooOp::print`. Additionally, the `Generated*Ops.td` files are all collapsed into just `GeneratedTorchOps.td` (there is no reason to have the files separate, since the files are very large anyway so one is always having to search within them -- editors don't care that the file to search is now a bit bigger :) ). This reduces TorchOpsODSGenerated.cpp compile time (which is now GeneratedTorchOps.cpp) from 39 to 31 seconds on my machine. This is actually less than I expected, but this PR is an overall cleanup to the code anyway. The next step will be to introduce (better) functionality upstream for sharding the TorchOps.cpp.inc file, so that we can truly parallelize the O(#ops) costs. This is also necessary, because after this PR, TorchDialect.cpp is now the slowest file to compile, due to the `addOperations<... all the ops ...>` call, which needs to be shareded too.
2022-03-19 05:04:47 +08:00
--torch_ir_include_dir="${torch_ir_include_dir}" \
--debug_registry_dump="${torch_ir_include_dir}/JITOperatorRegistryDump.txt"