biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Frontend][torch.compile] CompilationConfig Overhaul (#20283): name change compilation level to compilation mode, deprecation compilation level (#26355)

Browse Source 
Signed-off-by: morrison-turnansky <mturnans@redhat.com>
Signed-off-by: Morrison Turnansky <mturnans@redhat.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
This commit is contained in:
Morrison Turnansky
2025-10-14 22:51:16 -04:00
committed by GitHub
parent e66d787bce
commit 96b9aa5aa0
42 changed files with 270 additions and 248 deletions
Download Patch File Download Diff File Expand all files Collapse all files

30
tests/compile/test_basic_correctness.py
View File

@@ -4,7 +4,7 @@ import dataclasses
import pytest
from vllm.config import CompilationLevel
from vllm.config import CompilationMode
from vllm.utils import cuda_device_count_stateless
from ..utils import compare_all_settings
@@ -21,7 +21,7 @@ class TestSetting:
# we cannot afford testing the full Cartesian product
# of all models and all levels
# of all models and all modes
@pytest.mark.parametrize(
"test_setting",
[
@@ -121,15 +121,13 @@ def test_compile_correctness(
all_args: list[list[str]] = []
all_envs: list[dict[str, str] | None] = []
for comp_level in [
CompilationLevel.DYNAMO_AS_IS,
CompilationLevel.DYNAMO_ONCE,
CompilationLevel.PIECEWISE,
for comp_mode in [
CompilationMode.STOCK_TORCH_COMPILE,
CompilationMode.DYNAMO_TRACE_ONCE,
CompilationMode.VLLM_COMPILE,
]:
for level in [CompilationLevel.NO_COMPILATION, comp_level]:
all_args.append(
final_args + [f"-O.level={level}", "-O.backend=inductor"]
)
for mode in [CompilationMode.NONE, comp_mode]:
all_args.append(final_args + [f"-O.mode={mode}", "-O.backend=inductor"])
# inductor will change the output, so we only compare if the output
# is close, not exactly the same.
@@ -142,13 +140,13 @@ def test_compile_correctness(
all_envs.clear()
all_args.clear()
for level in [
CompilationLevel.NO_COMPILATION,
CompilationLevel.DYNAMO_AS_IS,
CompilationLevel.DYNAMO_ONCE,
CompilationLevel.PIECEWISE,
for mode in [
CompilationMode.NONE,
CompilationMode.STOCK_TORCH_COMPILE,
CompilationMode.DYNAMO_TRACE_ONCE,
CompilationMode.VLLM_COMPILE,
]:
all_args.append(final_args + [f"-O.level={level}", "-O.backend=eager"])
all_args.append(final_args + [f"-O.mode={mode}", "-O.backend=eager"])
all_envs.append({})
all_envs.append({})