Update Optional[x] -> x | None and Union[x, y] to x | y (#26633)

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2025-10-12 09:51:31 -07:00
committed by GitHub
parent 9bb38130cb
commit 8fcaaf6a16
944 changed files with 9490 additions and 10121 deletions
+10 -9
View File
@@ -4,9 +4,10 @@
import enum
import hashlib
from collections import Counter
from collections.abc import Callable
from dataclasses import asdict, field
from pathlib import Path
from typing import TYPE_CHECKING, Any, Callable, ClassVar, Optional, Union
from typing import TYPE_CHECKING, Any, ClassVar
from pydantic import TypeAdapter, field_validator
from pydantic.dataclasses import dataclass
@@ -168,7 +169,7 @@ class CompilationConfig:
"""
# Top-level Compilation control
level: Optional[int] = None
level: int | None = None
"""The level of compilation:
- None: If None, we will select the default compilation level.
@@ -177,7 +178,7 @@ class CompilationConfig:
- 1: dynamo as is.
- 2: dynamo once.
- 3: piecewise compilation."""
debug_dump_path: Optional[Path] = None
debug_dump_path: Path | None = None
"""The path to dump the debug information."""
cache_dir: str = ""
"""The directory to store the compiled graph, to accelerate Inductor
@@ -208,7 +209,7 @@ class CompilationConfig:
By default, all custom ops are enabled when running without Inductor and
disabled when running with Inductor: level>=PIECEWISE and use_inductor=True.
Inductor generates (fused) Triton kernels for disabled custom ops."""
splitting_ops: Optional[list[str]] = None
splitting_ops: list[str] | None = None
"""A list of ops to exclude from cudagraphs, used in piecewise compilation.
The behavior depends on use_inductor_graph_partition:
@@ -238,7 +239,7 @@ class CompilationConfig:
are compiled using configurations in inductor_compile_config.
This setting is ignored if level<PIECEWISE."""
compile_sizes: Optional[list[Union[int, str]]] = None
compile_sizes: list[int | str] | None = None
"""Sizes to compile for inductor. In addition
to integers, it also supports "cudagraph_capture_sizes" to
specify the sizes for cudagraph capture."""
@@ -253,7 +254,7 @@ class CompilationConfig:
constructor, e.g. `CompilationConfig(inductor_passes={"a": func})`."""
# CudaGraph compilation
cudagraph_mode: Optional[CUDAGraphMode] = None
cudagraph_mode: CUDAGraphMode | None = None
"""
The mode of the cudagraph:
@@ -308,7 +309,7 @@ class CompilationConfig:
It means the first several runs will be treated as warmup runs.
Only after that, the execution will be recorded, and the recorded
cudagraph will be used for subsequent runs."""
cudagraph_capture_sizes: Optional[list[int]] = None
cudagraph_capture_sizes: list[int] | None = None
"""Sizes to capture cudagraph.
- None (default): capture sizes are inferred from vllm config.
- list[int]: capture sizes are specified as given."""
@@ -320,7 +321,7 @@ class CompilationConfig:
internally managed buffer. Default is False.
Note that this flag is only effective when cudagraph_mode is PIECEWISE.
"""
full_cuda_graph: Optional[bool] = False
full_cuda_graph: bool | None = False
"""whether to use a full cuda graph for the entire forward pass rather than
splitting certain operations such as attention into subgraphs. Thus this
flag cannot be used together with splitting_ops. This may provide
@@ -544,7 +545,7 @@ class CompilationConfig:
"(where 'op' is the registered op name)"
)
def init_backend(self, vllm_config: "VllmConfig") -> Union[str, Callable]:
def init_backend(self, vllm_config: "VllmConfig") -> str | Callable:
if self.level == CompilationLevel.NO_COMPILATION:
raise ValueError("No compilation level is set.")