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

[v0.18.0][kernel] Recompilation optimization triggered by triton function parameter optimization (#7647)

Browse Source 
### What this PR does / why we need it?
Some parameters of Triton operators are unnecessarily modified with the
"constexpr" modifier. When these parameters change, recompilation is
triggered, which significantly affects the model performance. Therefore,
these parameters need to be rectified.

- vLLM version: v0.17.0
- vLLM main:
8b6325758c

Signed-off-by: HarpSealCC [844291270@qq.com](mailto:844291270@qq.com)
Signed-off-by: l30072083 <liuchengzhuo1@h-partners.com>
Co-authored-by: l30072083 <liuchengzhuo1@h-partners.com>
This commit is contained in:
HarpsealCC
2026-03-26 19:10:45 +08:00
committed by GitHub
parent d781902ce9
commit d6661c09b6
5 changed files with 20 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm_ascend/ops/triton/spec_decode/utils.py
View File

@@ -18,7 +18,7 @@
from vllm.triton_utils import tl, triton
@triton.jit
@triton.jit(do_not_specialize=["num_reqs"])
def prepare_inputs_padded_kernel(
cu_num_draft_tokens_ptr, # [num_reqs]
valid_sampled_tokens_count_ptr, # [num_reqs]