compute-runtime VS docker-mods

Well... Fuck! https://github.com/intel/compute-runtime/issues/643

For what it's worth, I was able to get IOMMU enabled and iGPU passthrough working for Plex on an Ubuntu 22.04 LXC container with a fresh install of Proxmox 8.0.3 (kenrel 6.2.16-3-pve) on an Intel i5-13400, ASRock B760M Pro RS/D4, also using Intel drivers released early today. I largely followed this guide.

For this little project unless someone chimes in with experience or can point to regarding the RX 6600 XT / 6650 XT / perhaps 7600 I might go with https://github.com/intel/compute-runtime . (They don't have perfect documentation re supported gpus either, the readme table doesn't list DG2, you need to go to releases to see that) . Phoronix reported it as pretty stable when they tested it with kernel 6.2 in March.

Update: I was able to fix my issue. I'm using Ubuntu 22.04.2 LTS and have the newest available kernel, 6.3.1. Installing the drivers via apt does not work, instead I needed to use https://github.com/intel/compute-runtime/releases/

Drivers: https://github.com/intel/compute-runtime/releases

wget https://github.com/intel/compute-runtime/releases/download/22.43.24595.30/intel-level-zero-gpu-dbgsym_1.3.24595.30_amd64.ddeb

For OpenCL HDR/DV tone-mapping, install the extra compute-runtime if running on host.

Have you tried adding the OpenCL-Intel Dockermod for the LSIO container? I don't believe it'll function properly without it...fair warning it significantly increases startup time. Additionally, it may not matter if you've already verified your hardware acceleration is working as expected, but at one point I did have to specifically indicate both renderD128 and card0 to get everything settled. If it ain't broke, don't fix it, but that's a difference in our compose files I spotted immediately.

Misc - LinuxServer.io images now support mods with multi-digest layers (https://github.com/linuxserver/docker-mods/pull/577)

Are you using a linuxserver container? If so, check this out as well: https://github.com/linuxserver/docker-mods/tree/jellyfin-opencl-intel

If you are running using the lsio container, you could try out this repo which makes the par2 process 2-3 times faster. https://github.com/linuxserver/docker-mods/tree/sabnzbd-par2cmdline-turbo

If you're using the LinuxServer plex image, you're probably facing this issue. https://github.com/linuxserver/docker-mods/issues/575

Following this issue and looking at the code here it appears that the mod identifies running applications on swag by globbing all the conf files (including samples!) and attemting a connection on them. And it does this with ThreadPoolExecutor(max_workers=100) which seems to explain why those DNS queries were being spammed on pihole.

I run a similar set up, but i run jellyfin with linuxserver's docker with this addon: https://github.com/linuxserver/docker-mods/tree/jellyfin-amd