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

mfix bug when max_seqs=14 in mtp=2 scenario and raise error when cudagraph_capture_sizes can't be an integer multiple of uniform_decode_query_lentp (#3910)

Browse Source 
### What this PR does / why we need it?
1. Revert [bugfix for mtp in
fullgraph](0948483642)
and support it when vllm supports
2. raise error when cudagraph_capture_sizes can't be an integer multiple
of uniform_decode_query_len
3. bugfix when max_num_seqs=14 in mtp=2 scenario

### Does this PR introduce _any_ user-facing change?
no

### How was this patch tested?

- vLLM version: v0.11.0
- vLLM main:
83f478bb19

---------

Signed-off-by: zouyida2052 <zouyida2002@gmail.com>
This commit is contained in:
zouyida2052
2025-10-31 09:24:50 +08:00
committed by GitHub
parent 35a913cf1e
commit 1966885be2
4 changed files with 63 additions and 61 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm_ascend/platform.py
View File

@@ -306,7 +306,6 @@ class NPUPlatform(Platform):
**********************************************************************************\033[0m
"""
logger.warning(warning_message)
update_aclgraph_sizes(vllm_config)
else:
logger.info(
"%s cudagraph_mode is not support on NPU. falling back to NONE",
@@ -344,7 +343,6 @@ class NPUPlatform(Platform):
**********************************************************************************\033[0m
"""
logger.warning(warning_message)
update_aclgraph_sizes(vllm_config)
else:
logger.info(
"%s cudagraph_mode is not support on NPU. falling back to NONE",