llvm
/
torch-mlir
mirror of https://github.com/llvm/torch-mlir
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
torch-mlir/frontends/pytorch/test/node_import/add3.py

27 lines
1.4 KiB
Python
Raw Normal View History

Add TorchScript graph importer. * Does not handle all features yet but should conservatively fail on unsupported things. * Location tracking is still somewhat mismatched between what TorchScript and MLIR do. Likely need a better heuristic for tracking locations from defs for nodes that do not carry location. * Sets the ground-work for a specialized/generic split but only implements the generic side. * Had some evidence that this requires a recent bump of PT nightly (within the last month) to pick up pybind11 2.6, which includes some cross-module symbol fixes (vs the previously sync'd version). No source changes, but older versions fail to cast function types at runtime.
2020-11-21 09:03:23 +08:00
# -*- Python -*-
# This file is licensed under a pytorch-style license
# See frontends/pytorch/LICENSE for license information.
import torch
import torch_mlir
# RUN: %PYTHON %s | npcomp-opt | FileCheck %s
mb = torch_mlir.ModuleBuilder()
# Verify without debug info.
Properly model "derefinement". In terms of IR structure, TorchScript allows types to vary in many circumstances where MLIR requires pointer-identical types. In particular, it is valid to pass any subtype in place of a type. For example, if an `Optional[int]` is required somewhere in the IR, it is legal to pass a value of just `int` (but not the other way around; see `torch.prim.unchecked_cast`). In effect, every *use* can have a different type. We introduce a new op `torch.derefine` that models that impedance mismatch. This op allows casting a value from one type to a type that it is a subtype of to model this behavior. Recommended review order: - TorchOps.td for new torch.derefine (and updated docs for `torch.prim.unchecked_cast`) - new test code in if.py, loop.py, function-derefine.py - new code in node_importer.cpp for handling derefinement insertion - function_importer.cpp and utils changes in torch_to_mlir_utils.cpp Properly handling derefinement on function boundaries required relayering the code so that graph_importer.cpp/.h is now function_importer.cpp/.h because only the `torch::jit::Function` (actually the `c10::FunctionSchema` it holds) knows the derefined types that are actually needed at the boundary (see `function-derefine.py` for a test). Annoyingly, this churns all the functions which are now prefixed with `__torch__.` but that is more correct anyway (that is their linkage name in the `torch::jit::CompilationUnit`; the previous `mb.import_function` was actually buggy in the case of functions calling each other as it would reference their unqualified name). With this change, we can import `resnet18` from `torchvision` :) IR: https://gist.github.com/silvasean/6426a5272d8a6c7caae533fce05ab704
2021-03-02 09:24:15 +08:00
# CHECK-LABEL: func @__torch__.add3
Add TorchScript graph importer. * Does not handle all features yet but should conservatively fail on unsupported things. * Location tracking is still somewhat mismatched between what TorchScript and MLIR do. Likely need a better heuristic for tracking locations from defs for nodes that do not carry location. * Sets the ground-work for a specialized/generic split but only implements the generic side. * Had some evidence that this requires a recent bump of PT nightly (within the last month) to pick up pybind11 2.6, which includes some cross-module symbol fixes (vs the previously sync'd version). No source changes, but older versions fail to cast function types at runtime.
2020-11-21 09:03:23 +08:00
# CHECK-SAME: (%arg0: !numpy.ndarray<*:!numpy.any_dtype>, %arg1: !numpy.ndarray<*:!numpy.any_dtype>, %arg2: !numpy.ndarray<*:!numpy.any_dtype>) -> !numpy.ndarray<*:!numpy.any_dtype> {
# CHECK: %[[C1:.*]] = constant 1 : i64
# CHECK: %[[A0:.*]] = torch.kernel_call "aten::add" %arg0, %arg1, %[[C1]] : (!numpy.ndarray<*:!numpy.any_dtype>, !numpy.ndarray<*:!numpy.any_dtype>, i64) -> !numpy.ndarray<*:!numpy.any_dtype> {sigArgTypes = ["Tensor", "Tensor", "Scalar"], sigIsMutable = false, sigIsVararg = false, sigIsVarret = false, sigRetTypes = ["Tensor"]}
# CHECK: %[[A1:.*]] = torch.kernel_call "aten::add" %[[A0]], %arg2, %[[C1]] : (!numpy.ndarray<*:!numpy.any_dtype>, !numpy.ndarray<*:!numpy.any_dtype>, i64) -> !numpy.ndarray<*:!numpy.any_dtype> {sigArgTypes = ["Tensor", "Tensor", "Scalar"], sigIsMutable = false, sigIsVararg = false, sigIsVarret = false, sigRetTypes = ["Tensor"]}
# CHECK: return %[[A1]] : !numpy.ndarray<*:!numpy.any_dtype>
Add TorchScript import tests missed in previous change.
2020-11-24 06:41:30 +08:00
@mb.import_function
@torch.jit.script
def add3(t0, t1, t2):
return t0 + t1 + t2
assert isinstance(add3, torch.jit.ScriptFunction)
Add TorchScript graph importer. * Does not handle all features yet but should conservatively fail on unsupported things. * Location tracking is still somewhat mismatched between what TorchScript and MLIR do. Likely need a better heuristic for tracking locations from defs for nodes that do not carry location. * Sets the ground-work for a specialized/generic split but only implements the generic side. * Had some evidence that this requires a recent bump of PT nightly (within the last month) to pick up pybind11 2.6, which includes some cross-module symbol fixes (vs the previously sync'd version). No source changes, but older versions fail to cast function types at runtime.
2020-11-21 09:03:23 +08:00
mb.module.operation.print()
print()