这边查的问题,但是官网没有对应的驱动包,麻烦看一下,谢谢
root@kq-dp:/home/kq-dp/metax-docker# metax-docker logs -f sglang-deepseek14b
Error response from daemon: No such container: sglang-deepseek14b
root@kq-dp:/home/kq-dp/metax-docker# metax-docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
root@kq-dp:/home/kq-dp/metax-docker# mx-smi
mx-smi version: 2.3.4
=================== MetaX System Management Interface Log ===================
Timestamp : Wed Jul 29 07:00:15 2026
Attached GPUs : 1
+---------------------------------------------------------------------------------+
| MX-SMI 2.3.4 Kernel Mode Driver Version: 3.9.6 |
| MACA Version: 3.8.1.3 BIOS Version: 1.24.3.0 |
|------------------+-----------------+---------------------+----------------------|
| Board Name | GPU Persist-M | Bus-id | GPU-Util sGPU-M |
| Pwr:Usage/Cap | Temp Perf | Memory-Usage | GPU-State |
|==================+=================+=====================+======================|
| 0 MetaX C500 | 0 N/A | 0000:3b:00.0 | 0% Disabled |
| 40W / 350W | 52C N/A | 860/65536 MiB | Available |
+------------------+-----------------+---------------------+----------------------+
+---------------------------------------------------------------------------------+
| Process: |
| GPU PID Process Name GPU Memory |
| Usage(MiB) |
|=================================================================================|
| no process found |
+---------------------------------------------------------------------------------+
End of Log
root@kq-dp:/home/kq-dp/metax-docker# docker images
i Info → U In Use
IMAGE ID DISK USAGE CONTENT SIZE EXTRA
cr.metax-tech.com/public-ai-release/maca/sglang:0.5.12-maca.ai3.8.0.3-torch2.8-py310-ubuntu22.04-amd64
24e06df9d3be 46.7GB 23.2GB
root@kq-dp:/home/kq-dp/metax-docker# ./start_debug.sh
INFO Print the version information of mcoplib during compilation.
Version info:Mcoplib_Version = '0.4.7'
Build_Maca_Version = '3.8.0.23'
GIT_BRANCH = 'HEAD'
GIT_COMMIT = '3c71b39'
Vllm Op Version = 0.22.0
SGlang Op Version = 0.5.12
INFO Staring Check the current MACA version of the operating environment.
INFO: Release major.minor matching, successful:3.8.
/opt/conda/lib/python3.10/site-packages/torch/cuda/init.py:186: UserWarning: CUDA initialization: CUDA driver initialization failed, you might not have a CUDA gpu. (Triggered internally at /workspace/framework/mcPytorch/c10/cuda/CUDAFunctions.cpp:109.)
return torch._C._cuda_getDeviceCount() > 0
W0729 15:11:25.741000 1 site-packages/torch/utils/cpp_extension.py:165] No CUDA runtime is found, using CUDA_HOME='/opt/maca/tools/cu-bridge'
W0729 15:11:25.742000 1 site-packages/torch/utils/cpp_extension.py:2533] TORCH_CUDA_ARCH_LIST is not set, all archs for visible cards are included for compilation.
W0729 15:11:25.742000 1 site-packages/torch/utils/cpp_extension.py:2533] If this is not desired, please set os.environ['TORCH_CUDA_ARCH_LIST'] to specific architectures.
Current Triton version 3.0.0 is below the recommended 3.3.0 version. Errors may occur and these issues will not be fixed. Please consider upgrading Triton.
Current Python version 3.10 is below the recommended 3.11 version. It is recommended to upgrade to Python 3.11 or higher for the best experience.
/opt/conda/lib/python3.10/site-packages/fla/utils.py:425: UserWarning: Triton is not supported on current platform, roll back to CPU.
warnings.warn(('Triton is not supported on current platform, roll back to CPU.'), stacklevel=1)
Unable to execute macainfo command, please ensure MACA is installed and you have an MetaX GPU on your system. using default xcore1000.
/opt/conda/lib/python3.10/site-packages/torchvision/datapoints/init.py:12: UserWarning: The torchvision.datapoints and torchvision.transforms.v2 namespaces are still Beta. While we do not expect major breaking changes, some APIs may still change according to user feedback. Please submit any feedback you may have in this issue: github.com/pytorch/vision/issues/6753, and you can also check out github.com/pytorch/vision/issues/7319 to learn more about the APIs that we suspect might involve future changes. You can silence this warning by calling torchvision.disable_beta_transforms_warning().
warnings.warn(_BETA_TRANSFORMS_WARNING)
/opt/conda/lib/python3.10/site-packages/torchvision/transforms/v2/init.py:54: UserWarning: The torchvision.datapoints and torchvision.transforms.v2 namespaces are still Beta. While we do not expect major breaking changes, some APIs may still change according to user feedback. Please submit any feedback you may have in this issue: github.com/pytorch/vision/issues/6753, and you can also check out github.com/pytorch/vision/issues/7319 to learn more about the APIs that we suspect might involve future changes. You can silence this warning by calling torchvision.disable_beta_transforms_warning().
warnings.warn(_BETA_TRANSFORMS_WARNING)
/opt/conda/lib/python3.10/site-packages/sglang/srt/layers/attention/fla/utils.py:223: UserWarning: Triton is not supported on current platform, roll back to CPU.
warnings.warn(
Traceback (most recent call last):
File "/opt/conda/bin/sglang", line 8, in <module>
sys.exit(main())
File "/opt/conda/lib/python3.10/site-packages/sglang/cli/main.py", line 40, in main
serve(args, extra_argv)
File "/opt/conda/lib/python3.10/site-packages/sglang/cli/serve.py", line 126, in serve
server_args = prepare_server_args(dispatch_argv)
File "/opt/conda/lib/python3.10/site-packages/sglang/srt/server_args.py", line 7810, in prepare_server_args
return ServerArgs.from_cli_args(raw_args)
File "/opt/conda/lib/python3.10/site-packages/sglang/srt/server_args.py", line 7194, in from_cli_args
return cls(**{attr: getattr(args, attr) for attr in attrs})
File "<string>", line 384, in init
File "/opt/conda/lib/python3.10/site-packages/sglang/srt/server_args.py", line 902, in post_init
self._handle_missing_default_values()
File "/opt/conda/lib/python3.10/site-packages/sglang/srt/server_args.py", line 1192, in _handle_missing_default_values
self.device = get_device()
File "/opt/conda/lib/python3.10/site-packages/sglang/srt/utils/common.py", line 2036, in get_device
raise RuntimeError("No accelerator (CUDA, XPU, HPU, NPU, MUSA, MPS) is available.")
RuntimeError: No accelerator (CUDA, XPU, HPU, NPU, MUSA, MPS) is available.
root@kq-dp:/home/kq-dp/metax-docker#
现在提示这个错误,麻烦帮忙看一下。官网没找到对于的版本包~
重启了呀,提示套件版本割裂:用户态 mxsmt 3.8.0.10 + 内核模块 metax-linux 3.9.10,且 3.9.10 内核模块无法在 5.15.0-119-generic 编译加载。
kq-dp@kq-dp:~$ sudo chmod +x metax-driver-3.8.0.10-deb-x86_64.run
[sudo] password for kq-dp:
kq-dp@kq-dp:~$ sudo ./metax-driver-3.8.0.10-deb-x86_64.run -- -f
Verifying archive integrity... 100% MD5 checksums are OK. All good.
Uncompressing MetaX driver installer 100%
uninstalling metax-linux
(Reading database ... 111596 files and directories currently installed.)
Purging configuration files for metax-linux (3.9.10) ...
metax-linux package uninstalled successfully
uninstalling mxsmt
(Reading database ... 111596 files and directories currently installed.)
Purging configuration files for mxsmt (3.8.0.10) ...
mxsmt package uninstalled successfully
uninstalling mxfw
(Reading database ... 111594 files and directories currently installed.)
Purging configuration files for mxfw (3.8.0) ...
mxfw package uninstalled successfully
deb/mxfw_.deb
deb/mxsmt_.deb
deb/metax-linux[-_]*.deb
installing packages under /opt/mxdriver
installing deb/mxfw_3.8.0-1.all.deb
Selecting previously unselected package mxfw.
(Reading database ... 111594 files and directories currently installed.)
Preparing to unpack deb/mxfw_3.8.0-1.all.deb ...
Unpacking mxfw (3.8.0) ...
Setting up mxfw (3.8.0) ...
mxfw package installed successfully
installing deb/mxsmt_3.8.0.10.amd64.deb
Selecting previously unselected package mxsmt.
(Reading database ... 111680 files and directories currently installed.)
Preparing to unpack deb/mxsmt_3.8.0.10.amd64.deb ...
Unpacking mxsmt (3.8.0.10) ...
Setting up mxsmt (3.8.0.10) ...
mxsmt package installed successfully
Processing triggers for libc-bin (2.35-0ubuntu3.14) ...
installing deb/metax-linux_3.9.10-1_amd64.deb
Selecting previously unselected package metax-linux.
(Reading database ... 111693 files and directories currently installed.)
Preparing to unpack .../metax-linux_3.9.10-1_amd64.deb ...
Unpacking metax-linux (3.9.10) ...
Setting up metax-linux (3.9.10) ...
metax-linux package installed successfully
kq-dp@kq-dp:~$ mx-smi
mx-smi: command not found
kq-dp@kq-dp:~$