Can't suspend or hibernate system

Support
, , ,
22

It roughly means let the firmware handles the ACPI things during boot, not the kernel. It’s a workaround for non-compliant old ACPI.

23

I’d get rid of it - remove it.
Just to have one variable less to deal with while addressing your current problem.

1 Like
24

But overall your problem seems either:

  • Noveau problem:
  • Or non-compliant ACPI as seen from the warnings. Most of the time these can be ignored. But this may actually cause your suspend issue. If this is the real root-cause, then there’s practically nothing you can do.
1 Like
25

Thank u, Indeed I guessed this before by reading this:

> journalctl --boot=-1 --priority=3                                                                                    ✔  16s  
Jan 09 16:22:46 phree-sys kernel: asus_atk0110: Resources not safely usable due to acpi_enforce_resources kernel parameter
Jan 09 16:22:46 phree-sys kernel: asus_atk0110: Resources not safely usable due to acpi_enforce_resources kernel parameter
Jan 09 16:22:46 phree-sys kernel: nouveau 0000:01:00.0: disp: conn 01:0261: func 08 lookup failed, -2
Jan 09 16:22:48 phree-sys avahi-daemon[337]: chroot.c: open() failed: No such file or directory
Jan 09 16:22:58 phree-sys systemd[619]: Failed to start Update XDG user dir configuration.
26

Then try using NVIDIA proprietary driver by sudo mhwd -a pci nonfree 0300. I’m not sure if mhwd supports NVIDIA driver version 340 out-of-the-box.

If not, then you need to install it manually:
https://www.nvidia.com/en-us/drivers/unix/
https://linuxconfig.org/how-to-install-the-nvidia-drivers-on-manjaro-linux

If NVIDIA proprietary driver doesn’t fix the issue, it’s probably the old non-compliant ACPI problem that you have.

27

I tried to install NVIDIA nonfree the manual way, so first I tried to update the system, installed linux515-headers, installed development tools ($ sudo pacman -S base-devel dkms) and added nouveau.modeset=0 to GRUB_CMDLINE_LINUX then updated grub…
But in TTY I got this while installing NVIDIA driver:

...
ERROR: Failed to install the kernel module through DKMS. No kernel module was installed; please try installing again without DKMS, or check the DKMS logs for more information.

Is sth wrong with linux515-headers? (> uname -r >>> 5.15.12-1-MANJARO)

28

That is a parameter which is only useful when you use the open source driver.
The proprietary driver is apparently not supported anymore - at least not the version of it which you’d need.
Not with this kernel version - perhaps with older Kernels.
… that is all I know from reading the forums here - I do not own Nvidia hardware myself …

29

It means that the NVIDIA driver installer failed to install the driver. It’s probably because NVIDIA proprietary driver version 340 isn’t supported for kernel 5.15. It’s supported up until kernel 5.14 and it won’t be supported for newer kernel version anymore.

https://nvidia.custhelp.com/app/answers/detail/a_id/3142/~/support-timeframes-for-unix-legacy-gpu-releases

You need to try install kernel 5.4 and try installing the NVIDIA proprietary driver again (along with adjusting from installing linux515-headers to something like linux514-headers).

Here is how to install multiple kernel and boot a specific kernel version:
https://wiki.manjaro.org/index.php/Manjaro_Kernels

30

Thank you JiaZhang,
I downgraded to 5.4, installed linux54-headers and installed NVIDIA 340 with dkms but not succeeded, then I tried with non dkms and finally installed the driver,
But after reboot It gets stuck in cursor blinking black screen,
What can I do except reinstalling Manjaro?

31

It’s really not safe to install NVIDIA driver in “non dkms” mode. If “dkms” mode fails, the driver is likely not functional for your system.

See below and you need to boot with runlevel 3, everything will be in tty or terminal:

So I guess for now you should revert back to nouveau driver.

For me, it seems that your problem is that your BIOS ACPI is just non-compliant to the ACPI standard. Your system is old, and it’s very likely that the OEM at-that-time disregarded the ACPI standard, only making sure compatibility with Windows BIOS table parser. Maybe Linux kernel 1.x or 2.x may work perfectly without any problem on your system, given your old computer system era. But version 1.x and 2.x are practically ancient.

So there’s really not much choice to accept this suspend problem :frowning:.

1 Like