| 7 Apr 2025 |
 ElvishJerricco | speaking of cmdline, I have no idea what this does: amdgpu.dcdebugmask=0x10 | 20:48:22 |
 nyanbinary 🏳️⚧️ | Disables AMD Psr | 20:49:40 |
| nyanbinary 🏳️⚧️ | its buggy | 20:49:42 |
| nyanbinary 🏳️⚧️ | and causes hangs/crashes | 20:49:46 |
| nyanbinary 🏳️⚧️ | https://github.com/NixOS/nixos-hardware/blob/de6fc5551121c59c01e2a3d45b277a6d05077bc4/framework/16-inch/common/amd.nix#L15 | 20:50:23 |
| ElvishJerricco | oh jfc nixos-hardware means the surface area of possible problems has gone up tremendously | 20:51:30 |
| ElvishJerricco | nyanbinary 🏳️⚧️: can you try it without any of the nixos-hardware stuff? I don't see anything in there that would be critical | 20:53:11 |
| nyanbinary 🏳️⚧️ | sure | 20:54:12 |
| nyanbinary 🏳️⚧️ | * binary login: [ 18.745279] amdgpu 0000:03:00.0: amdgpu: SMU: response:0xFFFFFFFF for index:43 param:0x00000000 message:PouerDownVcn?
[ 18.745323] amdgpu 0000:03:00.0: amdgpu: Failed to power gate VCN instance 0!
18.745347] [drm:vcn_v4_0_stop [andgpu]] ERROR Dpm disable uvd failed, ret = -121
18,745877] amdgpu 0000:03:00.0: amdgpu: SMU: response:0xFFFFFFFF for index:36 param:0x00000001 message:SetWorkloadMask?
[ 18.745917] amdgpu 0000:03:00.0: amdgpu: Failed to set workload mask 0x00000001 | 20:58:54 |
| nyanbinary 🏳️⚧️ | * binary login: [ 18.745279] amdgpu 0000:03:00.0: amdgpu: SMU: response:0xFFFFFFFF for index:43 param:0x00000000 message:PowerDownVcn?
[ 18.745323] amdgpu 0000:03:00.0: amdgpu: Failed to power gate VCN instance 0!
18.745347] [drm:vcn_v4_0_stop [andgpu]] ERROR Dpm disable uvd failed, ret = -121
18,745877] amdgpu 0000:03:00.0: amdgpu: SMU: response:0xFFFFFFFF for index:36 param:0x00000001 message:SetWorkloadMask?
[ 18.745917] amdgpu 0000:03:00.0: amdgpu: Failed to set workload mask 0x00000001 | 20:59:15 |
| ElvishJerricco | nyanbinary 🏳️⚧️: also, your dmesg looks similar to this: https://bbs.archlinux.org/viewtopic.php?pid=2235267 Though your symptoms are probably different. Still, give kernel 6.12 a try | 21:00:21 |
| ElvishJerricco | seems like there may be a regression in 6.13 | 21:01:03 |
| nyanbinary 🏳️⚧️ | Whats that pkg named in nixpkgs again | 21:01:48 |
| nyanbinary 🏳️⚧️ | :P | 21:01:49 |
| nyanbinary 🏳️⚧️ | for 6.12 | 21:01:50 |
| ElvishJerricco | boot.kernelPackages = pkgs.linuxPackages_6_12; | 21:02:17 |
 Makuru | boot.kernelPackages = linuxKernel.kernels.linux_6_13 | 21:02:27 |
| Makuru | Doesnt exist, I think. | 21:02:36 |
| Makuru | boot.kernelPackages = linuxKernel.kernels.linux_6_13; | 21:02:55 |
| ElvishJerricco | uh, no that should exist | 21:02:58 |
| ElvishJerricco | 6.13 is explicitly the one I'm trying to avoid | 21:03:06 |
| ElvishJerricco | $ nix eval github:NixOS/nixpkgs/nixos-unstable#linuxPackages_6_12.kernel.version
"6.12.21"
| 21:03:38 |
| Makuru | Yup, I am wrong, for some reason nixos search doesn't show it. | 21:04:22 |
| ElvishJerricco | linuxKernel.kernels.linux_6_12 is only the kernel. You'd want linuxKernel.packages.linux_6_12 if you wanted the newstyle attr path | 21:04:51 |
| ElvishJerricco | but there's nothing wrong with just linuxPackages_6_12 | 21:05:00 |
| Makuru | Huh | 21:05:44 |
| ElvishJerricco | the search probably just doesn't show the package sets | 21:05:50 |
| ElvishJerricco | the kernel is, of course, just the kernel, not the package set that boot.kernelPackages needs | 21:06:11 |
| nyanbinary 🏳️⚧️ | Erm whats the more 'correct' way of defining it (random question but yeag) | 21:06:38 |
| ElvishJerricco | boot.kernelPackages = pkgs.linuxPackages_6_12; and boot.kernelPackages = pkgs.linuxKernel.packages.linux_6_12; are equally correct | 21:07:06 |
