We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent 0453b2c commit 8973189Copy full SHA for 8973189
vllm_ascend/multistream/ms_split.py
@@ -72,8 +72,9 @@ def model_input_split_v1_mla_attn(
72
attn_metadata.query_lens):
73
return [attn_metadata]
74
75
- query_start_loc_cpu: Any = np.zeros(shape=(len(attn_metadata.query_lens) + 1, ),
76
- dtype=int)
+ query_start_loc_cpu: Any = np.zeros(shape=(len(attn_metadata.query_lens) +
+ 1, ),
77
+ dtype=int)
78
np.cumsum(attn_metadata.query_lens, out=query_start_loc_cpu[1:])
79
if attn_metadata.num_prefills > 0:
80
prefill_query_start_loc: Any = np.zeros(
0 commit comments