CardinalDeVille

Been racking my brain on this for the last few days, and looking for some validation that I am going down the right path. Sporadically, when I start my Windows 11 VM with a GTX 1070 passed through (with/without dumped vbios), my server will crash. No super repeatable steps, but tends to happen when I launch/exit a Steam Game (TCG Card Shop Simulator) It doesn't appear to just be this game, as sometimes it will crash just sitting at the desktop (and take the whole server with it) Before this started happening, no changes were made to unRAID, but in no particular order, here are all the things I have tried Adjusting bootflags to include/not include kernel /bzimage append isolcpus=4-15,20-31 initrd=/bzroot pcie_port_pm=off pcie_aspm.policy=performance video=efifb:off (also at one point I abandoned performance for pcie_aspm=off) Updated MB BIOS Redumped VBIOS from GPU-Z in Windows (not running unRAID -- SpaceInvaderOne script would not run for some reason) Factory Reset unRAID New Windows VM New unRAID install (fresh from the creator tool; same flash drive) Memtest for 36~ hours, no errors Reseated GPU, checked cables Typically, nothing has been in syslog, until one particular run today where I got the following: Oct 17 09:10:34 TheVault kernel: [drm] REG_WAIT timeout 1us * 100000 tries - optc1_wait_for_state line:817 Oct 17 09:10:34 TheVault kernel: [drm] REG_WAIT timeout 1us * 100000 tries - optc1_wait_for_state line:817 Oct 17 09:10:35 TheVault kernel: [drm] REG_WAIT timeout 1us * 100000 tries - optc1_wait_for_state line:817 just before the system crashed (again) The GPU is an older GTX 1070. Am I looking at a dead graphics card here? Specs: MB: Asus ROG STRIX X670E-E GAMING WIFI (Latest BIOS) AMD Ryzen 9 7950X3D 128GB Corsair Vengeance RAM Corsair RM1000X PSU EVGA GTX 1070 unRAID v6.12.13

To be fair to the fine folks at Unraid, this is a Linux Kernel issue; the drivers simply do not exist. Yes, it is annoying, but they are smart people and I am sure there are reasons that they don't jump kernels left and right. It was stated that Unraid 7 is being built on 6.5, which contains drivers for these motherboards. Confirmed this to be the case by doing a barebones Linux install on a spare SSD. temp sensors were read with no additional packages on a lightweight distro. The user in me is very annoyed that I have no temp insight on my server, but the tech enthusiast in me is happy that LimeTech is not just running with the latest thing and shipping an unstable OS.

Didn't see this somehow. Unraid 7 is on the horizon. Was announced (ish) when they made changes to the licensing.

In my case, the drivers for this are available in Linux kernel 6.5. Unraid 7 is confirmed to have this, so it’s a waiting game for me at this point. Found no help about this otherwise but if this thread gets bumped and someone sees that’d be sick

I am one of the poor unfortunate souls that experienced hardware failure and bought the “latest and (maybe not) greatest” hardware to replace my aging system. I paired a 7950X3D with an ASUS X670E-E. Unfortunately, this board’s temp drivers not working is causing me a great deal of worry; I’ve had temp issues in the past, and I am not enjoying “flying blind”. This worry is almost enough to shut the system down until it’s supported, or abandon unRAID all together. I’ve read posts on the forum about people compiling their own kernels for unRAID, but there is lots of data on here that I’d rather not have corrupted as part of that process. Does anyone know of any other bloody way to get these drivers into unRAID without doing a custom kernel OR is there any insight into when unRAID will support 6.3, which apparently has support for these mobos and temp sensors