vllm 从源代码安装失败，使用最新代码,

jucafojl 于 3个月前发布在其他

关注(0)|答案(5)|浏览(60)

Looking in indexes: https://pypi.org/simple, https://pypi.ngc.nvidia.com                                                                                                                      Obtaining file:///data/weilong.yu/github/vllm
  Installing build dependencies ... done
  Checking if build backend supports build_editable ... done                                                                                                                                    Getting requirements to build editable ... error
  error: subprocess-exited-with-error

  × Getting requirements to build editable did not run successfully.
  │ exit code: 1
  ╰─> [23 lines of output]
      /tmp/pip-build-env-30uui0su/overlay/local/lib/python3.10/dist-packages/torch/nn/modules/transformer.py:20: UserWarning: Failed to initialize NumPy: No module named 'numpy' (Triggered i
nternally at ../torch/csrc/utils/tensor_numpy.cpp:84.)
        device: torch.device = torch.device(torch._C._get_default_device()),  # torch.device('cpu'),
      <string>:220: UserWarning: Unsupported CUDA architectures ({'5.2', '7.2', '8.7', '6.0', '6.1'}) are excluded from the `TORCH_CUDA_ARCH_LIST` env variable (5.2 6.0 6.1 7.0 7.2 7.5 8.0 8
.6 8.7 9.0+PTX). Supported CUDA architectures are: {'8.6', '9.0', '8.0+PTX', '8.9+PTX', '8.0', '8.9', '8.6+PTX', '7.5+PTX', '7.0+PTX', '7.5', '9.0+PTX', '7.0'}.
      /usr/bin/python: No module named pip
      Traceback (most recent call last):
        File "/usr/local/lib/python3.10/dist-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 353, in <module>
          main()
        File "/usr/local/lib/python3.10/dist-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 335, in main
          json_out['return_val'] = hook(**hook_input['kwargs'])
        File "/usr/local/lib/python3.10/dist-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 132, in get_requires_for_build_editable
          return hook(config_settings)
        File "/tmp/pip-build-env-30uui0su/overlay/local/lib/python3.10/dist-packages/setuptools/build_meta.py", line 448, in get_requires_for_build_editable
          return self.get_requires_for_build_wheel(config_settings)                                                                                                                                   File "/tmp/pip-build-env-30uui0su/overlay/local/lib/python3.10/dist-packages/setuptools/build_meta.py", line 325, in get_requires_for_build_wheel
          return self._get_build_requires(config_settings, requirements=['wheel'])
        File "/tmp/pip-build-env-30uui0su/overlay/local/lib/python3.10/dist-packages/setuptools/build_meta.py", line 295, in _get_build_requires
          self.run_setup()
        File "/tmp/pip-build-env-30uui0su/overlay/local/lib/python3.10/dist-packages/setuptools/build_meta.py", line 311, in run_setup
          exec(code, locals())
        File "<string>", line 335, in <module>
        File "/usr/lib/python3.10/subprocess.py", line 369, in check_call
          raise CalledProcessError(retcode, cmd)
      subprocess.CalledProcessError: Command '['/usr/bin/python', '-m', 'pip', 'install', '-q', '--target=/data/weilong.yu/github/vllm/vllm/thirdparty_files', 'einops', 'flash-attn==2.5.6',
'--no-dependencies']' returned non-zero exit status 1.
      [end of output]

vllm

来源：https://github.com/vllm-project/vllm/issues/3276

5条答案

按热度按时间

yhived7q1#

你使用什么命令从源代码安装？
你安装了哪个版本的cuda?

赞(0）回复(0）举报 3个月前

a11xaf1n2#

这也发生在我身上，在2L4 GCP示例上。
首先在那里运行了正确的docker镜像进行构建。

docker run --gpus all -it --rm --ipc=host nvcr.io/nvidia/pytorch:23.10-py3

运行了命令

export VLLM_INSTALL_PUNICA_KERNELS=1
git clone https://github.com/vllm-project/vllm.git
cd vllm
pip install -e .

Looking in indexes: https://pypi.org/simple, https://pypi.ngc.nvidia.com
Obtaining file:///workspace/vllm
  Installing build dependencies ... done
  Checking if build backend supports build_editable ... done
  Getting requirements to build editable ... error
  error: subprocess-exited-with-error
  
  × Getting requirements to build editable did not run successfully.
  │ exit code: 1
  ╰─> [23 lines of output]
      /tmp/pip-build-env-g6mtcqwi/overlay/local/lib/python3.10/dist-packages/torch/nn/modules/transformer.py:20: UserWarning: Failed to initialize NumPy: No module named 'numpy' (Triggered internally at ../torch/csrc/utils/tensor_numpy.cpp:84.)
        device: torch.device = torch.device(torch._C._get_default_device()),  # torch.device('cpu'),
      <string>:220: UserWarning: Unsupported CUDA architectures ({'6.1', '8.7', '6.0', '5.2', '7.2'}) are excluded from the `TORCH_CUDA_ARCH_LIST` env variable (5.2 6.0 6.1 7.0 7.2 7.5 8.0 8.6 8.7 9.0+PTX). Supported CUDA architectures are: {'8.6', '9.0', '9.0+PTX', '8.9+PTX', '7.5+PTX', '8.6+PTX', '8.0+PTX', '8.9', '7.0+PTX', '7.5', '7.0', '8.0'}.
      /usr/bin/python: No module named pip
      Traceback (most recent call last):
        File "/usr/local/lib/python3.10/dist-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 353, in <module>
          main()
        File "/usr/local/lib/python3.10/dist-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 335, in main
          json_out['return_val'] = hook(**hook_input['kwargs'])
        File "/usr/local/lib/python3.10/dist-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 132, in get_requires_for_build_editable
          return hook(config_settings)
        File "/tmp/pip-build-env-g6mtcqwi/overlay/local/lib/python3.10/dist-packages/setuptools/build_meta.py", line 448, in get_requires_for_build_editable
          return self.get_requires_for_build_wheel(config_settings)
        File "/tmp/pip-build-env-g6mtcqwi/overlay/local/lib/python3.10/dist-packages/setuptools/build_meta.py", line 325, in get_requires_for_build_wheel
          return self._get_build_requires(config_settings, requirements=['wheel'])
        File "/tmp/pip-build-env-g6mtcqwi/overlay/local/lib/python3.10/dist-packages/setuptools/build_meta.py", line 295, in _get_build_requires
          self.run_setup()
        File "/tmp/pip-build-env-g6mtcqwi/overlay/local/lib/python3.10/dist-packages/setuptools/build_meta.py", line 311, in run_setup
          exec(code, locals())
        File "<string>", line 335, in <module>
        File "/usr/lib/python3.10/subprocess.py", line 369, in check_call
          raise CalledProcessError(retcode, cmd)
      subprocess.CalledProcessError: Command '['/usr/bin/python', '-m', 'pip', 'install', '-q', '--target=/workspace/vllm/vllm/thirdparty_files', 'einops', 'flash-attn==2.5.6', '--no-dependencies']' returned non-zero exit status 1.
      [end of output]
  
  note: This error originates from a subprocess, and is likely not a problem with pip.
error: subprocess-exited-with-error

× Getting requirements to build editable did not run successfully.
│ exit code: 1
╰─> See above for output.

note: This error originates from a subprocess, and is likely not a problem with pip.

如果有重复评论，请道歉。

赞(0）回复(0）举报 3个月前

brtdzjyr3#

你使用什么命令从源代码安装？
你安装了哪个版本的cuda?
pip install -e .
| NVIDIA-SMI 535.104.12 Driver Version: 535.104.12 CUDA Version: 12.2 |
commit 1ece1ae (HEAD -> main, origin/main, origin/HEAD)

赞(0）回复(0）举报 3个月前

vulvrdjw4#

这也发生在我身上，在2L4 GCP示例上。
首先在那里运行了正确的docker镜像进行构建。

docker run --gpus all -it --rm --ipc=host nvcr.io/nvidia/pytorch:23.10-py3

运行了命令

export VLLM_INSTALL_PUNICA_KERNELS=1
git clone https://github.com/vllm-project/vllm.git
cd vllm
pip install -e .

Looking in indexes: https://pypi.org/simple, https://pypi.ngc.nvidia.com
Obtaining file:///workspace/vllm
  Installing build dependencies ... done
  Checking if build backend supports build_editable ... done
  Getting requirements to build editable ... error
  error: subprocess-exited-with-error
  
  × Getting requirements to build editable did not run successfully.
  │ exit code: 1
  ╰─> [23 lines of output]
      /tmp/pip-build-env-g6mtcqwi/overlay/local/lib/python3.10/dist-packages/torch/nn/modules/transformer.py:20: UserWarning: Failed to initialize NumPy: No module named 'numpy' (Triggered internally at ../torch/csrc/utils/tensor_numpy.cpp:84.)
        device: torch.device = torch.device(torch._C._get_default_device()),  # torch.device('cpu'),
      <string>:220: UserWarning: Unsupported CUDA architectures ({'6.1', '8.7', '6.0', '5.2', '7.2'}) are excluded from the `TORCH_CUDA_ARCH_LIST` env variable (5.2 6.0 6.1 7.0 7.2 7.5 8.0 8.6 8.7 9.0+PTX). Supported CUDA architectures are: {'8.6', '9.0', '9.0+PTX', '8.9+PTX', '7.5+PTX', '8.6+PTX', '8.0+PTX', '8.9', '7.0+PTX', '7.5', '7.0', '8.0'}.
      /usr/bin/python: No module named pip
      Traceback (most recent call last):
        File "/usr/local/lib/python3.10/dist-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 353, in <module>
          main()
        File "/usr/local/lib/python3.10/dist-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 335, in main
          json_out['return_val'] = hook(**hook_input['kwargs'])
        File "/usr/local/lib/python3.10/dist-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 132, in get_requires_for_build_editable
          return hook(config_settings)
        File "/tmp/pip-build-env-g6mtcqwi/overlay/local/lib/python3.10/dist-packages/setuptools/build_meta.py", line 448, in get_requires_for_build_editable
          return self.get_requires_for_build_wheel(config_settings)
        File "/tmp/pip-build-env-g6mtcqwi/overlay/local/lib/python3.10/dist-packages/setuptools/build_meta.py", line 325, in get_requires_for_build_wheel
          return self._get_build_requires(config_settings, requirements=['wheel'])
        File "/tmp/pip-build-env-g6mtcqwi/overlay/local/lib/python3.10/dist-packages/setuptools/build_meta.py", line 295, in _get_build_requires
          self.run_setup()
        File "/tmp/pip-build-env-g6mtcqwi/overlay/local/lib/python3.10/dist-packages/setuptools/build_meta.py", line 311, in run_setup
          exec(code, locals())
        File "<string>", line 335, in <module>
        File "/usr/lib/python3.10/subprocess.py", line 369, in check_call
          raise CalledProcessError(retcode, cmd)
      subprocess.CalledProcessError: Command '['/usr/bin/python', '-m', 'pip', 'install', '-q', '--target=/workspace/vllm/vllm/thirdparty_files', 'einops', 'flash-attn==2.5.6', '--no-dependencies']' returned non-zero exit status 1.
      [end of output]
  
  note: This error originates from a subprocess, and is likely not a problem with pip.
error: subprocess-exited-with-error

× Getting requirements to build editable did not run successfully.
│ exit code: 1
╰─> See above for output.

note: This error originates from a subprocess, and is likely not a problem with pip.

如果有重复评论，抱歉。
是的，相同。

赞(0）回复(0）举报 3个月前

wydwbb8l5#

相同的问题

赞(0）回复(0）举报 3个月前