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

[e2e Test][npugraph_ex]add static kernel e2e test case (#6320)

Browse Source 
### What this PR does / why we need it?
Added an E2E test case for the scenario of enabling a static kernel for
npugraph_ex, monitoring its compilation and unloading process.
Also fixed the previously existing spelling errors

- vLLM version: v0.14.1
- vLLM main:
dc917cceb8

---------

Signed-off-by: chencangtao <chencangtao@huawei.com>
Co-authored-by: chencangtao <chencangtao@huawei.com>
This commit is contained in:
ChenCangtao
2026-01-30 16:24:48 +08:00
committed by GitHub
parent 8969b94a14
commit f2990f7741
2 changed files with 34 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
vllm_ascend/compilation/compiler_interface.py
View File

@@ -90,10 +90,10 @@ def npugraph_ex_compile(
# affecting program execution.
num_spec_tokens = vllm_config.speculative_config.num_speculative_token if vllm_config.speculative_config else 0
uniform_decode_query_len = num_spec_tokens + 1
max_num_tokens = vllm_config.scheduler_config.max_num_seq * uniform_decode_query_len
max_num_tokens = vllm_config.scheduler_config.max_num_seqs * uniform_decode_query_len
decode_cudagraph_batch_sizes = [
x
for x in vllm_config.compilation_config.cudagraph_capture_size
for x in vllm_config.compilation_config.cudagraph_capture_sizes
if max_num_tokens >= x >= uniform_decode_query_len
]
config.experimental_config.aclgraph._aclnn_static_shape_kernel_sym_value_range = decode_cudagraph_batch_sizes