[WIP] Add HPU support to vLLM v1 #997
Annotations
10 errors
|
Analysing the code with ruff:
vllm/attention/selector.py#L257
vllm/attention/selector.py:257:81: E501 Line too long (94 > 80)
|
|
Analysing the code with ruff:
vllm/v1/attention/backends/hpu_attn.py#L177
vllm/v1/attention/backends/hpu_attn.py:177:23: E702 Multiple statements on one line (semicolon)
|
|
Analysing the code with ruff:
vllm/v1/worker/hpu_model_runner.py#L11
vllm/v1/worker/hpu_model_runner.py:11:20: F401 `torch.nn` imported but unused
|
|
Analysing the code with ruff:
vllm/v1/worker/hpu_model_runner.py#L14
vllm/v1/worker/hpu_model_runner.py:14:46: F401 `vllm.compilation.compile_context.set_compile_context` imported but unused
|
|
Analysing the code with ruff:
vllm/v1/worker/hpu_model_runner.py#L26
vllm/v1/worker/hpu_model_runner.py:26:81: E501 Line too long (91 > 80)
|
|
Analysing the code with ruff:
vllm/v1/worker/hpu_model_runner.py#L29
vllm/v1/worker/hpu_model_runner.py:29:42: F401 `vllm_hpu_extension.profiler.HabanaHighLevelProfiler` imported but unused
|
|
Analysing the code with ruff:
vllm/v1/worker/hpu_model_runner.py#L30
vllm/v1/worker/hpu_model_runner.py:30:64: F401 `vllm_hpu_extension.profiler.format_bytes` imported but unused
|
|
Analysing the code with ruff:
vllm/v1/worker/hpu_model_runner.py#L167
vllm/v1/worker/hpu_model_runner.py:167:10: SIM223 Use `False` instead of `False and ...`
|
|
Analysing the code with ruff:
vllm/v1/worker/hpu_model_runner.py#L167
vllm/v1/worker/hpu_model_runner.py:167:81: E501 Line too long (88 > 80)
|
|
Analysing the code with ruff:
vllm/v1/worker/hpu_model_runner.py#L375
vllm/v1/worker/hpu_model_runner.py:375:81: E501 Line too long (128 > 80)
|
Loading