Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
1 change: 1 addition & 0 deletions python/sglang/srt/disaggregation/fake/conn.py

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

disaggregation fake backend modification LGTM

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

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

Thanks for the review! I just squashed the commits into one and force-pushed, so the previous approval got cleared. Could you re-approve when you get a chance? No code changes, just a clean single commit now.

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

Actually, I haven't approved yet since mega_moe part still requires a review from other reviewers. I am not an expert on mega moe, so you should ping other reviewers to check this fix, then we can run the CI.

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

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

Thanks for the clarification! Got it.

Original file line number Diff line number Diff line change
Expand Up @@ -28,6 +28,7 @@ def __init__(
is_mla_backend: Optional[bool] = False,
):
super().__init__(args, disaggregation_mode, server_args, is_mla_backend)
self.kv_args = args
self.req_to_decode_prefix_len = {}

def register_to_bootstrap(self):
Expand Down
4 changes: 2 additions & 2 deletions python/sglang/srt/layers/moe/mega_moe.py
Original file line number Diff line number Diff line change
Expand Up @@ -206,8 +206,8 @@ def _run_mega_routed(
)

if num_tokens > 0:
topk_ids_in = topk_ids
topk_weights_in = topk_weights
topk_ids_in = topk_ids.to(torch.int32)
topk_weights_in = topk_weights.to(torch.float32)
else:
topk_ids_in = hidden_states.new_empty((0, top_k), dtype=torch.int32)
topk_weights_in = hidden_states.new_empty((0, top_k), dtype=torch.float32)
Expand Down
Loading