/dev/kfd no longer exists

Question

/dev/kfd no longer exists

PhilipDeegan opened this issue 7 years ago · 11 comments

PhilipDeegan commented 7 years ago

Not sure where it went.

Is this expected?

Answer 1 · 2018-01-25T20:36:51.000Z

This is on your host machine, or in your container?

Answer 2 · 2018-01-25T20:45:51.000Z

Host - rocm-dkms became an optional package and was seemingly removed.

Possibly something in the 1.6 to 1.7 upgrade?

Answer 3 · 2018-01-25T20:49:16.000Z

Something went bad with the packages. rocm-dkms should not have been removed. It's a meta package that references a bunch of subpackages. I assume that if you do a lsmod now, you will no longer see the amdgpu or amdkfd kernel modules loaded. The amdkfd module is what provides the /dev/kfd device.

Answer 4 · 2018-01-25T20:50:49.000Z

I tried doing a manual modprobe on amdkfd, but I got an exec error

Answer 5 · 2018-01-25T20:51:07.000Z

doing a full remove/reinstall now will let you know thanks

Answer 6 · 2018-01-25T20:54:07.000Z

That did the trick, can you confirm just for my own curiosity, Is the ROCM 4.11 kernel still supposed to be there? I have it, but I don't see a "rocm-kernel" package anymore so I'm not sure if it's obsolete.

Thanks

Answer 7 · 2018-01-25T20:58:22.000Z

Great. Yes, I think the rocm-kernel package is obsolete. The corresponding equivalent package in our new dkms world is rock-dkms. With dkms, you no longer have a custom monolithic roc kernel, you should have the stock Ubuntu kernel and the rocm stuff loads like a kernel driver with dkms. With a stock ubuntu 16.04 install, if you typed uname -r, it should give you a 4.4 kernel version.

Answer 8 · 2018-01-25T21:00:31.000Z

Yeah I still have 4.11.0-kfd-compute-rocm-rel-1.6-180 somehow

So the normal host kernel should work now? I thought would only work with 4.16

Answer 9 · 2018-01-25T21:24:54.000Z

That kernel is from the 1.6 rocm (dkms does away with custom kernels). You should be able to revert back to the stock ubuntu kernel with:
sudo dpkg --purge linux-headers-4.11.0-kfd-compute-rocm-rel-1.6-180 linux-image-4.11.0-kfd-compute-rocm-rel-1.6-180. Sounds like you may have a mixture of 1.6 and 1.7 packages now, but it would be best (i.e. tested) if you double checked that all the 1.6 packages are removed.

Answer 10 · 2018-01-25T21:27:30.000Z

congrats to all involved, that was a bit of a pain point

Answer 11 · 2018-01-25T21:29:35.000Z

I found the uninstall directions for 1.6 here. Wouldn't hurt to try that. you may unintentionally uninstall a 1.7 package (not sure), but that would be trivial to reinstall with apt install