Skip to content

[smoke][bugfix] moe_init_routing_v2 active_expert_range use int type #5521

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
wangxiyuan merged 1 commit into from
Dec 31, 2025
Merged

[smoke][bugfix] moe_init_routing_v2 active_expert_range use int type #5521

Show file tree
Hide file tree
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
1 change: 0 additions & 1 deletion vllm_ascend/eplb/core/eplb_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -104,7 +104,6 @@ def generate_log2phy_map(global_expert_map, ep_rank):
for rankid, map_per_rank in enumerate(global_expert_map):
for idx, val in enumerate(map_per_rank):
val = val.item()
# 计算value：当前值 + i * 有效元素个数
if val != -1:
log2phy_map[idx].append(val + rankid * valid_count)

Expand Down
2 changes: 1 addition & 1 deletion vllm_ascend/ops/fused_moe/fused_moe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -180,7 +180,7 @@ def __init__(self, *args, **kwargs):
or ascend_config.expert_map_record_path) and (
self.log2phy is not None)
self.local_num_experts = (torch.sum(
self._expert_map != -1) if self._expert_map is not None else
self._expert_map != -1).item() if self._expert_map is not None else
self.global_num_experts)
if self._expert_map is not None:
logger.info_once(
Expand Down
4 changes: 3 additions & 1 deletion vllm_ascend/ops/fused_moe/token_dispatcher.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -335,7 +335,9 @@ def __init__(self, **kwargs):
super().__init__(**kwargs)
self.apply_router_weight_on_input = False
self.max_num_tokens = kwargs.get("max_num_tokens")
self.num_experts_local = kwargs.get("num_local_experts", 0)
num_experts_local = kwargs.get("num_local_experts", 0)
self.num_experts_local = num_experts_local.item() if torch.is_tensor(
num_experts_local) else int(num_experts_local)
self.original_shape = None
self.with_quant = False

Expand Down
Loading