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

[qwen3 next ]add ascend c casual_conv1d_fn (#6661)

Browse Source 
### What this PR does / why we need it?
add ascend c casual_conv1d_fn

- vLLM version: v0.15.0
- vLLM main:
13397841ab
---------
Signed-off-by: ZT-AIA <1028681969@qq.com>
Signed-off-by: ZT-AIA <63220130+ZT-AIA@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
This commit is contained in:
ZT-AIA
2026-03-09 23:29:49 +08:00
committed by GitHub
parent 48b624e4cc
commit ee5347e824
26 changed files with 2504 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
csrc/build_aclnn.sh
View File

@@ -24,7 +24,8 @@ elif [[ "$SOC_VERSION" =~ ^ascend910b ]]; then
ABSOLUTE_CATLASS_PATH=$(cd "${CATLASS_PATH}" && pwd)
export CPATH=${ABSOLUTE_CATLASS_PATH}:${CPATH}
CUSTOM_OPS="moe_grouped_matmul;grouped_matmul_swiglu_quant_weight_nz_tensor_list;lightning_indexer_vllm;sparse_flash_attention;matmul_allreduce_add_rmsnorm;moe_init_routing_custom;moe_gating_top_k;add_rms_norm_bias;apply_top_k_top_p_custom;transpose_kv_cache_by_block;"
CUSTOM_OPS="moe_grouped_matmul;grouped_matmul_swiglu_quant_weight_nz_tensor_list;lightning_indexer_vllm;sparse_flash_attention;matmul_allreduce_add_rmsnorm;moe_init_routing_custom;moe_gating_top_k;add_rms_norm_bias;apply_top_k_top_p_custom;transpose_kv_cache_by_block;causal_conv1d;"
SOC_ARG="ascend910b"
elif [[ "$SOC_VERSION" =~ ^ascend910_93 ]]; then
# ASCEND910C (A3) series
@@ -63,6 +64,7 @@ elif [[ "$SOC_VERSION" =~ ^ascend910_93 ]]; then
"add_rms_norm_bias"
"apply_top_k_top_p_custom"
"transpose_kv_cache_by_block"
"causal_conv1d"
"moe_grouped_matmul"
)
CUSTOM_OPS=$(IFS=';'; echo "${CUSTOM_OPS_ARRAY[*]}")