Skip to content

Skip redundant moe_sum_reduce for single-expert routing on XPU #22660

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

Open
rahulvijayaraghavan wants to merge 1 commit into
base: main
Choose a base branch
from
+8 −5
Open

Skip redundant moe_sum_reduce for single-expert routing on XPU #22660

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
13 changes: 8 additions & 5 deletions python/sglang/srt/layers/moe/fused_moe_triton/fused_moe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -678,11 +678,14 @@ def fused_experts_impl(
routed_scaling_factor,
)
elif _is_xpu:
moe_sum_reduce(
intermediate_cache3.view(*intermediate_cache3.shape),
out_hidden_states[begin_chunk_idx:end_chunk_idx],
routed_scaling_factor,
)
if topk_ids.shape[1] == 1 and routed_scaling_factor == 1.0:
pass # we write directly into out_hidden_states
else:
moe_sum_reduce(
intermediate_cache3.view(*intermediate_cache3.shape),
out_hidden_states[begin_chunk_idx:end_chunk_idx],
routed_scaling_factor,
)
Comment on lines +681 to +688
Copy link
Copy Markdown
Collaborator

@mingfeima mingfeima Apr 13, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

make sure we have test cases to cover:

  • topk == 1, routed_scaling_factor == 1.0
  • topk == 1, routed_scaling_factor != 1.0

addtionally, is it possible to move topk_ids.shape[1] == 1 and routed_scaling_factor == 1.0 up and skip intermediate_cache3 allocation at the first place.

else:
if _has_vllm_ops:
vllm_ops.moe_sum(
Expand Down
Loading