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

[Refactor] Import global var form vllm instead of overwirte it (#5469)

Browse Source 
### What this PR does / why we need it?
Import global var form vllm instead of overwirte it, so that we could
use the correct global variant value

- vLLM version: v0.13.0
- vLLM main:
5326c89803
---------
Signed-off-by: MengqingCao <cmq0113@163.com>
This commit is contained in:
Mengqing Cao
2026-01-07 18:41:45 +08:00
committed by GitHub
parent 380f089fbf
commit 3f4f2b4ae6
10 changed files with 7 additions and 157 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
vllm_ascend/patch/worker/patch_bert.py
View File

@@ -20,7 +20,6 @@ from vllm.model_executor.models import bert
# aclgraph does not support shift operator for now
# TODO: revert me when aclgraph supports shift operator
TOKEN_TYPE_SHIFT = 30
TOKEN_TYPE_MULTIPLIER = 1 << 30
TOKEN_MASK = TOKEN_TYPE_MULTIPLIER - 1