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

[torch.compile] Make inductor partition rules respect splitting_ops #25691 (#25845)

Browse Source 
Signed-off-by: baonudesifeizhai <baonudesifeizhai@gmail.com>
Signed-off-by: baonudesifeizhai <85092850+baonudesifeizhai@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
This commit is contained in:
baonudesifeizhai
2025-10-10 12:35:28 -04:00
committed by GitHub
parent e519281920
commit cddce79fda
9 changed files with 267 additions and 112 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
tests/compile/test_decorator.py
View File

@@ -71,7 +71,7 @@ def test_ignore_torch_compile_decorator():
compilation_config=CompilationConfig(
level=CompilationLevel.PIECEWISE,
use_cudagraph=True,
splitting_ops=["silly.attention"],
splitting_ops=["silly::attention"],
cudagraph_capture_sizes=[1, 2],
)
)
@@ -186,7 +186,7 @@ def test_conditional_compile_enable_if():
compilation_config=CompilationConfig(
level=CompilationLevel.PIECEWISE,
use_cudagraph=True,
splitting_ops=["silly.attention"],
splitting_ops=["silly::attention"],
cudagraph_capture_sizes=[1, 2],
),
)
@@ -218,7 +218,7 @@ def test_conditional_compile_enable_if():
compilation_config=CompilationConfig(
level=CompilationLevel.PIECEWISE,
use_cudagraph=True,
splitting_ops=["silly.attention"],
splitting_ops=["silly::attention"],
cudagraph_capture_sizes=[1, 2],
),
)