Skip to content

[Fix] incorrect assert in EPLB #7575

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
zhyncs merged 1 commit into from
Jun 26, 2025
Merged

[Fix] incorrect assert in EPLB #7575

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
4 changes: 0 additions & 4 deletions python/sglang/srt/managers/expert_distribution.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -479,10 +479,6 @@ def __init__(self, *args, **kwargs):
def on_select_experts(self, layer_idx: int, topk_ids: torch.Tensor):
topk_ids = topk_ids.flatten()
mask = topk_ids != -1
assert self._data[layer_idx, :].shape == topk_ids.shape, (
"Shape mismatch between data and topk_ids."
"Selecting expert is not supported for multiple token prediction at the moment."
)
self._data[layer_idx, :].scatter_add_(
dim=0, index=topk_ids.masked_fill(~mask, 0).long(), src=mask.int()
)
Comment on lines 482 to 484
Copy link
Contributor

@gemini-code-assist gemini-code-assist bot Jun 26, 2025

Choose a reason for hiding this comment

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

medium

Consider adding an assertion to validate that the expert IDs in topk_ids are within the valid range. This can prevent potential out-of-bounds errors in the scatter_add_ operation.

Suggested change
self._data[layer_idx, :].scatter_add_(
dim=0, index=topk_ids.masked_fill(~mask, 0).long(), src=mask.int()
)
mask = topk_ids != -1
if topk_ids.numel() > 0:
assert topk_ids.max() < self._data.shape[1]

Expand Down
Loading