Skip to content

[Benchmark] Add fp8_attention to tritonbench integration #319

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Jul 16, 2025
Merged

[Benchmark] Add fp8_attention to tritonbench integration #319

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 10 additions & 2 deletions benchmarks/run.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -70,6 +70,11 @@
"examples.cross_entropy",
"cross_entropy",
),
"fp8_attention": (
"tritonbench.operators.fp8_attention.operator",
"examples.fp8_attention",
"fp8_attention_tritonbench",
),
}


Expand Down Expand Up @@ -282,7 +287,7 @@ def helion_method(
if isinstance(attr, Kernel):
attr.reset()

def _inner() -> Callable[..., Any]:
def _inner() -> Callable[..., Any] | object:
# Force autotuning unless HELION_USE_DEFAULT_CONFIG=1 is set
# This ensures we run autotuning even if the kernel has pre-specified configs
if os.environ.get("HELION_USE_DEFAULT_CONFIG", "0") != "1":
Expand All @@ -292,7 +297,10 @@ def _inner() -> Callable[..., Any]:
if isinstance(attr, Kernel):
attr.settings.force_autotune = True

return kernel_func(*args)
result = kernel_func(*args)
if callable(result):
return result()
return result

return _inner

Expand Down
Loading