Packed QKV and Rotary Embedding Support for sm<80 GQA #27656
Annotations
5 warnings
|
Run reviewdog/action-cpplint@master:
onnxruntime/contrib_ops/cuda/bert/group_query_attention.cc#L175
[cpplint] reported by reviewdog 🐶
Lines should be <= 120 characters long [whitespace/line_length] [2]
Raw Output:
onnxruntime/contrib_ops/cuda/bert/group_query_attention.cc:175: Lines should be <= 120 characters long [whitespace/line_length] [2]
|
|
Run reviewdog/action-cpplint@master:
onnxruntime/contrib_ops/cuda/bert/group_query_attention.cc#L184
[cpplint] reported by reviewdog 🐶
Lines should be <= 120 characters long [whitespace/line_length] [2]
Raw Output:
onnxruntime/contrib_ops/cuda/bert/group_query_attention.cc:184: Lines should be <= 120 characters long [whitespace/line_length] [2]
|
|
Run reviewdog/action-cpplint@master:
onnxruntime/contrib_ops/cuda/bert/group_query_attention_impl.cu#L494
[cpplint] reported by reviewdog 🐶
Lines should be <= 120 characters long [whitespace/line_length] [2]
Raw Output:
onnxruntime/contrib_ops/cuda/bert/group_query_attention_impl.cu:494: Lines should be <= 120 characters long [whitespace/line_length] [2]
|
|
Run reviewdog/action-cpplint@master:
onnxruntime/contrib_ops/cuda/bert/group_query_attention_impl.cu#L497
[cpplint] reported by reviewdog 🐶
Lines should be <= 120 characters long [whitespace/line_length] [2]
Raw Output:
onnxruntime/contrib_ops/cuda/bert/group_query_attention_impl.cu:497: Lines should be <= 120 characters long [whitespace/line_length] [2]
|
|
Run reviewdog/action-cpplint@master:
onnxruntime/contrib_ops/cuda/bert/group_query_attention_impl.cu#L509
[cpplint] reported by reviewdog 🐶
Lines should be <= 120 characters long [whitespace/line_length] [2]
Raw Output:
onnxruntime/contrib_ops/cuda/bert/group_query_attention_impl.cu:509: Lines should be <= 120 characters long [whitespace/line_length] [2]
|
Loading