Arch Linux

Magniquick

Member

Registered: 2024-10-03

Posts: 1

Cant get nvidia gpu out of D3cold

Sorry in advance if I'm in the wrong place, or am missing something (I'm new here). If so, please tell me what I can do.
Basically, during boot:

❯ doas dmesg | grep nvidia
[    3.138230] nvidia: loading out-of-tree module taints kernel.
[    3.138237] nvidia: module license 'NVIDIA' taints kernel.
[    3.138241] nvidia: module verification failed: signature and/or required key missing - tainting kernel
[    3.138241] nvidia: module license taints kernel.
[    3.434723] nvidia-nvlink: Nvlink Core is being initialized, major device number 234
[    3.438960] nvidia 0000:02:00.0: Unable to change power state from D3cold to D0, device inaccessible
[    3.439053] nvidia 0000:02:00.0: probe with driver nvidia failed with error -1
....

What I've tried:
- using bbswitch to toggle the gpu
    - I can get the /sys/bus/pci/devices/0000:02:00.0/power/runtime_status to show active with this, but /sys/bus/pci/devices/0000:02:00.0/power_state is still reports D3cold
- load with "nvidia NVreg_DynamicPowerManagement=0x0k", k throughout 0 to 2
- disabling linux kernel runtime power management with pcie_aspm=off
    - seems to do nothing
- changing nvidia power managment from on to auto @ ``/sys/bus/pci/devices/0000:02:00.0/power/control``
- I've read the arch wiki nvidia install instructions almost religiously, no dice
lscpi:

❯ lspci -k | grep -A 2 -E "(VGA|3D)"
00:02.0 VGA compatible controller: Intel Corporation WhiskeyLake-U GT2 [UHD Graphics 620] (rev 02)
    Subsystem: Hewlett-Packard Company Device 85c6
    Kernel driver in use: i915
--
02:00.0 3D controller: NVIDIA Corporation GP108M [GeForce MX250] (rev a1)
    Subsystem: Hewlett-Packard Company Device 85c6
    Kernel modules: nouveau, nvidia_drm, nvidia

dmesg: https://0x0.st/XEiP.txt

Last edited by Magniquick (2024-10-03 18:10:42)

seth

Member

From: Won't reply 2 private help req

Registered: 2012-09-03

Posts: 75,158

Re: Cant get nvidia gpu out of D3cold

[    3.342365] NVRM: This is a 64-bit BAR mapped above 4GB by the system
               NVRM: BIOS or the Linux kernel, but the PCI bridge
               NVRM: immediately upstream of this GPU does not define
               NVRM: a matching prefetchable memory window.
[    3.342368] NVRM: This may be due to a known Linux kernel bug.  Please
               NVRM: see the README section on 64-bit BARs for additional
               NVRM: information.
[    3.342390] NVRM: The NVIDIA probe routine failed for 1 device(s).
[    3.342391] NVRM: None of the NVIDIA devices were initialized.
[    4.590903] NVRM: This is a 64-bit BAR mapped above 4GB by the system
               NVRM: BIOS or the Linux kernel, but the PCI bridge
               NVRM: immediately upstream of this GPU does not define
               NVRM: a matching prefetchable memory window.
[    4.594630] NVRM: This may be due to a known Linux kernel bug.  Please
               NVRM: see the README section on 64-bit BARs for additional
               NVRM: information.
[    4.594646] NVRM: The NVIDIA probe routine failed for 1 device(s).
[    4.594646] NVRM: None of the NVIDIA devices were initialized.

Edit:

[    1.562431]  nvme0n1: p1 p2 p3 p4 p5 p6
[  436.487563] rtw_8822be 0000:04:00.0: firmware failed to leave lps state

How many parallel windows installations are there?
Is a trick question, see the 3rd link below. Mandatory.
Disable it (it's NOT the BIOS setting!) and reboot windows and linux twice for voodo reasons.

Last edited by seth (2024-10-04 16:11:34)

---

How to upload text · How to boot w/o GUI · Disable Windows Fast-Start! · Fix your xinitrc