EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

[BugFix] Check all expert maps when using muilty instance. (#3662)

Browse Source 
### What this PR does / why we need it?
Check all expert maps when using muilty instance.

### Does this PR introduce _any_ user-facing change?
None.

### How was this patch tested?
Qwen 235B in double A3.
case1：master has expert map, slave has not expert map.
case2:   master has expert map, slave has error expert map.
case3:   master has expert map,slave has correct expert map.
- vLLM version: v0.11.0rc3
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.0

---------

Signed-off-by: offline0806 <3337230449@qq.com>
Co-authored-by: offline0806 <3337230449@qq.com>
This commit is contained in:
offline893
2025-10-24 17:10:31 +08:00
committed by GitHub
parent b321e3846a
commit 4e21b1537e
3 changed files with 25 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
vllm_ascend/ops/common_fused_moe.py
View File

@@ -192,6 +192,7 @@ class AscendFusedMoE(FusedMoE):
os.R_OK):
self.expert_load_balancer = ExpertLoadBalancer(
self.expert_map_path, self.global_num_experts)
self.expert_load_balancer.check_expert_map_tensor()
self.global_redundant_expert_num = (
self.expert_load_balancer.get_global_redundant_expert_num())
try: