xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-05-25 05:24:25 +08:00
Code Issues Packages Projects Releases Wiki Activity

[BugFix] Workspace allocation during profile run : DeepEPHighThroughput + DeepGEMM (#30899)

Browse Source
This commit is contained in:
Varun Sundar Rabindranath 2025-12-17 18:00:59 -05:00 committed by GitHub
parent e06d0bf0aa
commit e3fc374a9a
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 4 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
vllm/model_executor/layers/fused_moe/modular_kernel.py
View File

@ -795,7 +795,10 @@ class FusedMoEModularKernel(torch.nn.Module):
top_k, top_k,
global_num_experts, global_num_experts,
local_num_experts, local_num_experts,
expert_tokens_meta, # expert_tokens_meta help in allocating optimal/minimal
# amount of workspace. Mark it None, so we allocate for
# the worst-case scenario.
expert_tokens_meta=None,
) )
) )