Skip to content
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

[MISC] add support custom_op check #8557

Merged
merged 8 commits into from
Sep 21, 2024
Merged

[MISC] add support custom_op check #8557

Changes from 1 commit
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
2ae21c0
add support custom_op check
jikunshang Sep 18, 2024
3749180
fix
jikunshang Sep 19, 2024
9e5fdce
fix all reduce call
jikunshang Sep 19, 2024
b6f524c
add neuron
jikunshang Sep 20, 2024
87c0c95
fix
jikunshang Sep 20, 2024
3a38075
change implementation
youkaichao Sep 20, 2024
383b777
Merge branch 'main' into check_custom_op
youkaichao Sep 20, 2024
d4f215f
polish condition
youkaichao Sep 20, 2024
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
Prev Previous commit
Next Next commit
fix all reduce call
  • Loading branch information
@jikunshang
jikunshang committed Sep 19, 2024
commit 9e5fdcebb08a386520454b48b91093799a0750cf
2 changes: 2 additions & 0 deletions vllm/distributed/parallel_state.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -345,6 +345,8 @@ def all_reduce(self, input_: torch.Tensor) -> torch.Tensor:
if self.ca_comm is not None and self.ca_comm.should_custom_ar(input_):
return torch.ops.vllm.outplace_all_reduce(
input_, group_name=self.unique_name)
elif not supports_custom_op():
return self._all_reduce(input_)
else:
torch.ops.vllm.inplace_all_reduce(input_,
group_name=self.unique_name)
Expand Down
Loading