Skip to content

[1.21 cherry-pick] Fix async callback ordering (#1023) #1028

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 2 commits into from
Apr 10, 2025
Merged

[1.21 cherry-pick] Fix async callback ordering (#1023) #1028

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
cbfcf9d
Fix async callback ordering (#1023)
madamczyk-intel Apr 8, 2025
f5ed30f
Merge branch 'v1.21.0_next' into dev/madamczyk/fix_async_clb_ordering…
michalkuligowski Apr 9, 2025
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
5 changes: 4 additions & 1 deletion vllm/worker/hpu_model_runner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2704,6 +2704,8 @@ def try_revert_dummy_output_tokens():

if use_delayed_sampling:
fake_output = self._delayed_sampler_outputs(model_input)
elif model_input.async_callback is not None:
model_input.async_callback()

with self.profiler.record_event(
'internal', ('sample_'
Expand All @@ -2725,7 +2727,8 @@ def try_revert_dummy_output_tokens():
self.cached_step_outputs.append(output)
self.cached_step_inputs.append(model_input)
htorch.core.mark_step()
if model_input.async_callback is not None:
if use_delayed_sampling \
and model_input.async_callback is not None:
model_input.async_callback()
if i < num_steps - 1:
if i == 0:
Expand Down