Optimus manager / driver problem

Hello, I have Dell Inspiron laptop with optimus technology. When I was using a 5.4 kernel, optimus-manager worked very well. Only after one update I started to have problems, signaled like this:

[ 324.924574] pci 0000:08:00.0: can’t change power state from D3cold to D0 (config space inaccessible)
[ 324.927735] pci 0000:08:00.0: can’t change power state from D3cold to D0 (config space inaccessible)
[ 324.946387] pci 0000:08:00.0: can’t change power state from D3cold to D0 (config space inaccessible)

Now I’m using a 5.7 kernel and the problem persists. I don’t know if closed nvidia drivers (418.xxx) or optimus-manager are to blame. An interesting fact is that this problem occurs randomly. Specification of my equipment:

System:

  • Host: inspiron3558 Kernel: 5.7.17-2-MANJARO x86_64 bits: 64 *
  • Desktop: Xfce 4.14.2 Distro: Manjaro Linux *
    Machine:
  • Type: Laptop System: Dell product: Inspiron 3558 v: N/A *
  • serial: <superuser/root required> *
  • Mobo: Dell model: 0G1TDD v: A00 serial: <superuser/root required> *
  • UEFI [Legacy]: Dell v: A17 date: 08/27/2018 *
    Battery:
  • ID-1: BAT0 charge: 16.8 Wh condition: 19.1/41.4 Wh (46%) *
    CPU:
  • Dual Core: Intel Core i5-5200U type: MT MCP speed: 798 MHz *
  • min/max: 500/2700 MHz *
    Graphics:
  • Device-1: Intel HD Graphics 5500 driver: i915 v: kernel *
  • Device-2: NVIDIA GK208BM [GeForce 920M] driver: N/A *
  • Device-3: Microdia Integrated Webcam type: USB driver: uvcvideo *
  • Display: x11 server: X Org 1.20.8 driver: intel resolution: 1366x768~60Hz *
  • OpenGL: renderer: Mesa Intel HD Graphics 5500 (BDW GT2) v: 4.6 Mesa 20.1.6 *
    Network:
  • Device-1: Intel Wireless 3160 driver: iwlwifi *
  • Device-2: Realtek RTL810xE PCI Express Fast Ethernet driver: r8169 *
    Drives:
  • Local Storage: total: 465.76 GiB used: 357.75 GiB (76.8%) *
    Info:
  • Processes: 213 Uptime: 8m Memory: 7.69 GiB used: 1.80 GiB (23.4%) *
  • Shell: Bash inxi: 3.1.05*

This is might be problematic, no driver is loaded for the nvidia gpu. Please run sudo mhwd -li and lspci -k.

Because I was using Intel driver. When I switched to Nvidia, output of inxi looks that:

Device-2: NVIDIA GK208BM [GeForce 920M] driver: nvidia v: 418.113

Can you identify which device is at this PCI address?

Yes:

08:00.0 3D controller: NVIDIA Corporation GK208BM [GeForce 920M] (rev ff)
Kernel modules: nouveau, nvidia_drm, nvidia

What are the symptoms of the problem? You cannot suspend or resume? Does it happen in the intel or nvidia mode?

This text is an output of dmesg. If it appears, I can’t switch gpu to nvidia. It ioccurs randomly.

On kernel 5.8 the problem persists, but the fps rate drops by half. Is it possible for the system not to remove kernel 5.7 after each update?

Now I am using the prime method and the problem is gone. Especially on the new 5.10 kernel there was the biggest problem with switching. It turns out that the optimus-manager was to blame. Unfortunately, switching to prime reduced battery life. Subject to be closed.

This topic was automatically closed 15 days after the last reply. New replies are no longer allowed.