操作系统版本：centos 7.6
DTK 版本：DTK-24.04.2
驱动版本：rock-5.7.1-6.2.26-V1.5.aio.run
DCU 版类型：z100L
rocminfo时，遇到报错hsa api cal failure at: /data/jenkins_workspace/workspace/galaxy_release/galaxy/rocminfo/rocminfo.cc:1165
Call returned HSA_STATUS_ERROR_OUT_OF_RESOURCES: The runtime failed to allocate the nessary resources. This error may also occur when the core runtime library needs to spawn threads or create internal OS-specific events.
rocm-smi和hy-smi没有报错。dmesg日志报错为"hydcu: rocminfo: failed to get hyuds net namespace."

进入docker image.sourcefind.cn:5000/dcu/admin/base/pytorch:2.1.0-ubuntu20.04-dtk24.04.2-py3.10容器后，也是这个问题，rocm-smi和hysmi能显示显卡，但还多了一个报错MKFD file descriptor not valid in this process.
然后在容器里torch.cuda.is_available()没有成功。

是因为这里说的原因吗，解决方案没看懂，不知道命令行应该是什么，但我的cat /etc/group关于video是
video:x :39:root
ls -l /dev/dri/render*是
crw-rw-rw- 1 root video 226, 128 Nov 8 19:32 /dev/dri/renderD128

https://github.com/ROCm/ROCm/issues/889
https://github.com/ROCm/ROCm/pull/1005
https://github.com/ROCm/ROCm/issues/923