Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

PVC support regression #210

Open
upsj opened this issue Dec 26, 2024 · 3 comments
Open

PVC support regression #210

upsj opened this issue Dec 26, 2024 · 3 comments

Comments

@upsj
Copy link

upsj commented Dec 26, 2024

Rebuilding our system images moving from kernel version 5.14.0-427.42.1.el9_4.x86_64 with version 1.24.4.12.240603.18-1 (intel-i915-dkms) to kernel version 5.14.0-503.19.1.el9_5.x86_64 with version 1.24.7.9.241015.10-1 caused PVC support to break. I also added the necessary firmware files as mentioned in #199

Relevant information before:

$ lspci -knn
...
ca:00.0 Display controller [0380]: Intel Corporation Ponte Vecchio XT (1 Tile) [Data Center GPU Max 1100] [8086:0bda] (rev 2f)
        Subsystem: Intel Corporation Device [8086:0000]
        Kernel driver in use: i915
        Kernel modules: i915
...
$ modprobe -c
...
alias pci:v00008086d00000BDAsv*sd*bc03sc*i* i915
...

After:

$ lspci -knn
...
ca:00.0 Display controller [0380]: Intel Corporation Ponte Vecchio XT (1 Tile) [Data Center GPU Max 1100] [8086:0bda] (rev 2f)
        Subsystem: Intel Corporation Device [8086:0000]
...
$ modprobe -c
...
alias pci:v00008086d00000A2Asv*sd*bc03sc*i* i915
alias pci:v00008086d00000A2Bsv*sd*bc03sc*i* i915
alias pci:v00008086d00000A2Esv*sd*bc03sc*i* i915
alias pci:v00008086d00000A84sv*sd*bc03sc*i* i915
alias pci:v00008086d00000C02sv*sd*bc03sc*i* i915
alias pci:v00008086d00000C06sv*sd*bc03sc*i* i915
alias pci:v00008086d00000C0Asv*sd*bc03sc*i* i915
alias pci:v00008086d00000C0Bsv*sd*bc03sc*i* i915
alias pci:v00008086d00000C0Esv*sd*bc03sc*i* i915
...

Our A770 GPUs are still working, so this is only a partial failure. This is an Intel D50DNP server with 2 single-tile PVC GPUs loaned from Intel. I would like to avoid holding back the kernel version on a single system only.

@upsj
Copy link
Author

upsj commented Dec 27, 2024

I think it is also important to note that the intel-i915-dkms package shipped via https://repositories.intel.com/gpu/rhel/9.4/lts/2350/unified/ also has the same problem (and has had it since Rocky/RHEL 9.3), though IIRC this is version 1.23.

@smuqthya
Copy link

@upsj The kernel you are trying is RHEL9.5 based. which is not yet supported on the version

Please go through Readme doc for the latest OSV support. https://github.com/intel-gpu/intel-gpu-i915-backports/tree/backport/main .

@upsj
Copy link
Author

upsj commented Dec 27, 2024

I am aware, this is why I'm highlighting this regression so it can be handled in the next release.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants