2024-04-27 18:27:37 +08:00
|
|
|
# Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
|
|
|
|
# See https://llvm.org/LICENSE.txt for license information.
|
|
|
|
# SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
|
|
|
|
# Also available under a BSD-style license. See LICENSE.
|
|
|
|
from enum import Enum
|
|
|
|
from io import StringIO
|
|
|
|
import os
|
|
|
|
import sys
|
|
|
|
import tempfile
|
|
|
|
from typing import Union
|
|
|
|
|
|
|
|
from torch_mlir.passmanager import PassManager
|
|
|
|
from torch_mlir.ir import StringAttr
|
|
|
|
|
|
|
|
|
|
|
|
def get_module_name_for_debug_dump(module):
|
|
|
|
"""Gets a name suitable for a debug dump.
|
|
|
|
|
|
|
|
The name is not guaranteed to be unique.
|
|
|
|
"""
|
|
|
|
if not "torch.debug_module_name" in module.operation.attributes:
|
|
|
|
return "UnnammedModule"
|
|
|
|
return StringAttr(module.operation.attributes["torch.debug_module_name"]).value
|
|
|
|
|
|
|
|
|
|
|
|
class TorchMlirCompilerError(Exception):
|
|
|
|
pass
|
|
|
|
|
2024-04-28 05:16:31 +08:00
|
|
|
|
|
|
|
def run_pipeline_with_repro_report(
|
|
|
|
module, pipeline: str, description: str, enable_ir_printing: bool = False
|
|
|
|
):
|
2024-04-27 18:27:37 +08:00
|
|
|
"""Runs `pipeline` on `module`, with a nice repro report if it fails."""
|
|
|
|
module_name = get_module_name_for_debug_dump(module)
|
|
|
|
original_stderr = sys.stderr
|
|
|
|
try:
|
|
|
|
sys.stderr = StringIO()
|
|
|
|
asm_for_error_report = module.operation.get_asm(
|
2024-04-28 05:16:31 +08:00
|
|
|
large_elements_limit=10, enable_debug_info=True
|
|
|
|
)
|
2024-04-27 18:27:37 +08:00
|
|
|
# Lower module in place to make it ready for compiler backends.
|
|
|
|
with module.context as ctx:
|
|
|
|
pm = PassManager.parse(pipeline)
|
|
|
|
if enable_ir_printing:
|
|
|
|
ctx.enable_multithreading(False)
|
|
|
|
pm.enable_ir_printing()
|
|
|
|
pm.run(module.operation)
|
|
|
|
except Exception as e:
|
|
|
|
# TODO: More robust.
|
|
|
|
# - don't arbitrarily clutter up /tmp. When a test suite has many
|
|
|
|
# tests, this can be a big disk cost (also, /tmp/ is frequently a
|
|
|
|
# RAM fs, which increases worries about capacity).
|
|
|
|
# - don't have colliding filenames (hard to do without cluttering
|
|
|
|
# up /tmp)
|
|
|
|
# - if we do have have colliding filenames, writes should at least
|
|
|
|
# avoid being racy.
|
|
|
|
filename = os.path.join(tempfile.gettempdir(), module_name + ".mlir")
|
2024-04-28 05:16:31 +08:00
|
|
|
with open(filename, "w") as f:
|
2024-04-27 18:27:37 +08:00
|
|
|
f.write(asm_for_error_report)
|
2024-04-28 05:16:31 +08:00
|
|
|
debug_options = "-mlir-print-ir-after-all -mlir-disable-threading"
|
2024-04-27 18:27:37 +08:00
|
|
|
# Put something descriptive here even if description is empty.
|
|
|
|
description = description or f"{module_name} compile"
|
|
|
|
|
|
|
|
message = f"""\
|
|
|
|
{description} failed with the following diagnostics:
|
|
|
|
{sys.stderr.getvalue()}
|
|
|
|
|
|
|
|
python exception: {e}
|
|
|
|
|
|
|
|
For Torch-MLIR developers, the error can be reproduced with:
|
|
|
|
$ torch-mlir-opt -pass-pipeline='{pipeline}' {filename}
|
|
|
|
Add '{debug_options}' to get the IR dump for debugging purpose.
|
|
|
|
"""
|
2024-04-28 05:16:31 +08:00
|
|
|
trimmed_message = "\n".join([m.lstrip() for m in message.split("\n")])
|
2024-04-27 18:27:37 +08:00
|
|
|
raise TorchMlirCompilerError(trimmed_message) from None
|
|
|
|
finally:
|
|
|
|
sys.stderr = original_stderr
|
|
|
|
|
|
|
|
|
|
|
|
class OutputType(Enum):
|
|
|
|
|
|
|
|
# Output torch dialect. When converting from FX, this will be immediately
|
|
|
|
# after the import from FX to MLIR. When converting from torchscript,
|
|
|
|
# this will come after some cleanup passes which attempt to de-alias,
|
|
|
|
# decompose and infer shapes. These should be roughly the same level of
|
|
|
|
# abstraction since those steps are done within PyTorch itself
|
|
|
|
# when coming directly from Dynamo/FX.
|
|
|
|
TORCH = "torch"
|
|
|
|
|
|
|
|
# The output type contains a mix of `linalg`-on-tensors ops, `scf`, and
|
|
|
|
# `arith` ops (and also `math` and `tm_tensor`). It can be thought of
|
|
|
|
# as taking the `TORCH` output type and lowering it so that tensor
|
|
|
|
# computations are done with `linalg`-on-tensors ops.
|
|
|
|
LINALG_ON_TENSORS = "linalg-on-tensors"
|
|
|
|
|
|
|
|
# This output type consists of `tosa` dialect ops. It can be thought of
|
|
|
|
# as taking the `TORCH` output type and lowering it to TOSA.
|
|
|
|
TOSA = "tosa"
|
|
|
|
|
|
|
|
# This output type consists of `stablehlo` dialect ops. It can be thought of
|
|
|
|
# as taking the `TORCH` output type and lowering it to StableHLO.
|
|
|
|
STABLEHLO = "stablehlo"
|
|
|
|
|
|
|
|
# Raw output of the JIT IR importer. This is not expected to be useful
|
|
|
|
# for end-users, but can be convenient for development or reporting bugs.
|
|
|
|
RAW = "raw"
|
|
|
|
|
|
|
|
@staticmethod
|
|
|
|
def get(spec: Union[str, "OutputType"]) -> "OutputType":
|
|
|
|
"""Gets an OutputType from allowed way to specify one.
|
|
|
|
|
|
|
|
Args:
|
|
|
|
spec: An OutputType instance or the case-insensitive name of one of the
|
|
|
|
enum values.
|
|
|
|
Returns:
|
|
|
|
An OutputType instance.
|
|
|
|
"""
|
|
|
|
if isinstance(spec, OutputType):
|
|
|
|
return spec
|
|
|
|
spec = spec.upper().replace("-", "_")
|
|
|
|
if spec not in OutputType.__members__:
|
2024-04-28 05:16:31 +08:00
|
|
|
raise ValueError(
|
|
|
|
f"For output_type= argument, expected one of: "
|
|
|
|
f"{', '.join(OutputType.__members__.keys())}"
|
|
|
|
)
|
2024-04-27 18:27:37 +08:00
|
|
|
return OutputType[spec]
|
|
|
|
|
|
|
|
|
|
|
|
def lower_mlir_module(verbose, output_type, module):
|
|
|
|
if verbose:
|
|
|
|
print("\n====================")
|
|
|
|
print("Torch Backend IR")
|
|
|
|
print(module)
|
|
|
|
|
|
|
|
if output_type == OutputType.TORCH:
|
|
|
|
return module
|
|
|
|
|
|
|
|
if output_type == OutputType.TOSA:
|
|
|
|
run_pipeline_with_repro_report(
|
2024-04-28 05:16:31 +08:00
|
|
|
module,
|
|
|
|
"builtin.module(torch-backend-to-tosa-backend-pipeline)",
|
|
|
|
"Lowering Torch Backend IR -> TOSA Backend IR",
|
|
|
|
)
|
2024-04-27 18:27:37 +08:00
|
|
|
if verbose:
|
|
|
|
print("\n====================")
|
|
|
|
print("TOSA Backend IR")
|
|
|
|
print(module)
|
|
|
|
return module
|
|
|
|
|
|
|
|
if output_type == OutputType.LINALG_ON_TENSORS:
|
|
|
|
run_pipeline_with_repro_report(
|
|
|
|
module,
|
|
|
|
"builtin.module(torch-backend-to-linalg-on-tensors-backend-pipeline)",
|
2024-04-28 05:16:31 +08:00
|
|
|
"Lowering Torch Backend IR -> Linalg-on-Tensors Backend IR",
|
|
|
|
)
|
2024-04-27 18:27:37 +08:00
|
|
|
if verbose:
|
|
|
|
print("\n====================")
|
|
|
|
print("LINALG Backend IR")
|
|
|
|
print(module)
|
|
|
|
return module
|
|
|
|
|
|
|
|
elif output_type == OutputType.STABLEHLO:
|
|
|
|
run_pipeline_with_repro_report(
|
|
|
|
module,
|
|
|
|
"builtin.module(torch-backend-to-stablehlo-backend-pipeline)",
|
2024-04-28 05:16:31 +08:00
|
|
|
"Lowering Torch Backend IR -> StableHLO Backend IR",
|
|
|
|
)
|
2024-04-27 18:27:37 +08:00
|
|
|
if verbose:
|
|
|
|
print("\n====================")
|
|
|
|
print("StableHLO Backend IR")
|
|
|
|
print(module)
|
|
|
|
return module
|
|
|
|
raise Exception(f"Unknown OutputType: {output_type}")
|