[Kernel] Add moe_gating_top_k operator support for Ascend NPU (#5579)
### What this PR does / why we need it?
1.replace moe_gating_top_k from torch_npu with custom op
2.enable the renorm function of moe_gating_top_k in softmax scenerio
### Does this PR introduce _any_ user-facing change?
No
### How was this patch tested?
No need test
- vLLM version: v0.13.0
- vLLM main:
7157596103
---------
Signed-off-by: ZCG12345 <2097562023@qq.com>
This commit is contained in:
@@ -24,7 +24,7 @@ elif [[ "$SOC_VERSION" =~ ^ascend910b ]]; then
|
||||
ABSOLUTE_CATLASS_PATH=$(cd "${CATLASS_PATH}" && pwd)
|
||||
export CPATH=${ABSOLUTE_CATLASS_PATH}:${CPATH}
|
||||
|
||||
CUSTOM_OPS="grouped_matmul_swiglu_quant_weight_nz_tensor_list;lightning_indexer;sparse_flash_attention;matmul_allreduce_add_rmsnorm;moe_init_routing_custom"
|
||||
CUSTOM_OPS="grouped_matmul_swiglu_quant_weight_nz_tensor_list;lightning_indexer;sparse_flash_attention;matmul_allreduce_add_rmsnorm;moe_init_routing_custom;moe_gating_top_k;"
|
||||
SOC_ARG="ascend910b"
|
||||
elif [[ "$SOC_VERSION" =~ ^ascend910_93 ]]; then
|
||||
# ASCEND910C (A3) series
|
||||
@@ -70,6 +70,7 @@ elif [[ "$SOC_VERSION" =~ ^ascend910_93 ]]; then
|
||||
"dispatch_layout"
|
||||
"notify_dispatch"
|
||||
"moe_init_routing_custom"
|
||||
"moe_gating_top_k"
|
||||
)
|
||||
CUSTOM_OPS=$(IFS=';'; echo "${CUSTOM_OPS_ARRAY[*]}")
|
||||
SOC_ARG="ascend910_93"
|
||||
|
||||
Reference in New Issue
Block a user