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

[Bugfix] Fix aclgraph not enabled by default (#2590)

Browse Source 
### What this PR does / why we need it?
As vllm will set `cudagraph_mode` to `NONE` before
`check_and_update_config` in post init of `VllmConfig`
(5da4f5d857/vllm/config/__init__.py (L3630)),
we always have `cudagraph_mode` isn't `None`, thus we must remove this
check and add it when the related adaption in vllm is done.

part of https://github.com/vllm-project/vllm-ascend/pull/2577, will add
the e2e test on applying reply after the CI refactor is done

### How was this patch tested?
CI passed with existing test.

- vLLM version: v0.10.1.1
- vLLM main:
f48a9af892

Signed-off-by: MengqingCao <cmq0113@163.com>
This commit is contained in:
Mengqing Cao
2025-08-28 14:08:31 +08:00
committed by GitHub
parent cf96366a39
commit 6c973361fc
3 changed files with 22 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

34
vllm_ascend/platform.py
View File

@@ -146,23 +146,23 @@ class NPUPlatform(Platform):
compilation_config.cudagraph_num_of_warmups = 1
if compilation_config.cudagraph_mode is None:
# if cudagraph_mode is not explicitly set by users, set default value
if compilation_config.level == CompilationLevel.PIECEWISE:
compilation_config.cudagraph_mode = \
CUDAGraphMode.PIECEWISE
elif compilation_config.level not in [
CompilationLevel.NO_COMPILATION, CompilationLevel.PIECEWISE
]:
logger.warning(
"NPU does not support %s compilation level. Setting CUDAGraphMode to NONE",
compilation_config.level)
compilation_config.cudagraph_mode = CUDAGraphMode.NONE
else:
logger.warning(
"compilation_config.level = CompilationLevel.NO_COMPILATION is set, Setting CUDAGraphMode to NONE"
)
compilation_config.cudagraph_mode = CUDAGraphMode.NONE
# TODO: make vllm support oot platform to set `compilation_config.cudagraph_mode`
# if cudagraph_mode is not explicitly set by users, set default value
if compilation_config.level == CompilationLevel.PIECEWISE:
compilation_config.cudagraph_mode = \
CUDAGraphMode.PIECEWISE
elif compilation_config.level not in [
CompilationLevel.NO_COMPILATION, CompilationLevel.PIECEWISE
]:
logger.warning(
"NPU does not support %s compilation level. Setting CUDAGraphMode to NONE",
compilation_config.level)
compilation_config.cudagraph_mode = CUDAGraphMode.NONE
else:
logger.warning(
"compilation_config.level = CompilationLevel.NO_COMPILATION is set, Setting CUDAGraphMode to NONE"
)
compilation_config.cudagraph_mode = CUDAGraphMode.NONE
# set CUDAGraphMode to None when torchair is enabled, no mather what compilation_config.level is.
if ascend_config.torchair_graph_config.enabled: