@myswtest Check the output from cat /proc/cmdline is there a resume= entry?
Yep!
# cat /proc/cmdline
BOOT_IMAGE=/boot/vmlinuz-6.6.2-1-default
root=UUID=xxxxxxxx401
resume=/dev/disk/by-id/nvme-Samsxxxx-part2
quiet security=apparmor
amdgpu.ppfeaturemask=0xffffbffb amdgpu.dpm=0
mitigations=auto
# cat /var/log/boot.log | grep bernate
#
.
@myswtest So fire up YaST2 Bootloader and remove the resume=/dev/disk/by-id/nvme-Samsxxxx-part2 entry, save and reboot.
I did have a resume=xxx and tried to remove it and reboot like you’ve suggested, I no longer see any hibernation mentioned in boot.log openSUSE Paste … but the boot problem still persists, this time it also appeared while shutting down the computer …
But checking my journal I can still see a mention of the resume=xxx on a dracut related lines :
nov. 28 18:18:28 noname-opensuse dracut-cmdline[258]: Using kernel command line parameters:
rd.driver.pre=btrfs
resume=UUID=fe888bcc-e7b3-408a-9997-0f92bb766d73
root=UUID=788c0329-b2d2-4765-a3ad-4fc78e9d7340
rootfstype=btrfs rootflags=rw,re>
I have also some lines about hibernation ?
nov. 28 18:18:28 noname-opensuse kernel: PM: hibernation: Registered nosave memory: [mem 0x00000000-0x00000fff]
nov. 28 18:18:28 noname-opensuse kernel: PM: hibernation: Registered nosave memory: [mem 0x0009d000-0x0009dfff]
...
But I digress …
To add to the discussion there is the part in my logs where I have Nvidia error on boot
nov. 28 18:18:31 noname-opensuse kernel: [drm] Initialized nvidia-drm 0.0.0 20160202 for 0000:01:00.0 on minor 0
nov. 28 18:18:31 noname-opensuse kernel: Console: switching to colour dummy device 80x25
nov. 28 18:18:31 noname-opensuse kernel: nvidia 0000:01:00.0: vgaarb: deactivate vga console
nov. 28 18:18:31 noname-opensuse kernel: fbcon: nvidia-drmdrmfb (fb0) is primary device
nov. 28 18:18:31 noname-opensuse systemd[1]: systemd-vconsole-setup.service: Deactivated successfully.
nov. 28 18:18:31 noname-opensuse systemd[1]: Stopped Virtual Console Setup.
nov. 28 18:18:31 noname-opensuse kernel: Console: switching to colour frame buffer device 240x67
nov. 28 18:18:34 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 0
nov. 28 18:18:37 noname-opensuse systemd[1]: Stopping Virtual Console Setup...
nov. 28 18:18:37 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 1
nov. 28 18:18:37 noname-opensuse kernel: nvidia 0000:01:00.0: [drm] fb0: nvidia-drmdrmfb frame buffer device
nov. 28 18:18:37 noname-opensuse systemd[1]: Starting Virtual Console Setup...
nov. 28 18:18:41 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 0
nov. 28 18:18:44 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 1
nov. 28 18:18:47 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 0
nov. 28 18:18:50 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 1
nov. 28 18:18:50 noname-opensuse systemd[1]: Finished dracut initqueue hook.
nov. 28 18:18:53 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 0
nov. 28 18:18:57 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 1
nov. 28 18:19:00 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 0
nov. 28 18:19:03 noname-opensuse systemd[1]: Reached target Preparation for Remote File Systems.
nov. 28 18:19:03 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 1
nov. 28 18:19:06 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 0
nov. 28 18:19:09 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 1
nov. 28 18:19:13 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 0
nov. 28 18:19:16 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 1
nov. 28 18:19:16 noname-opensuse systemd[1]: Reached target Remote File Systems.
nov. 28 18:19:19 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 0
nov. 28 18:19:22 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 1
nov. 28 18:19:25 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 0
nov. 28 18:19:29 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 1
nov. 28 18:19:32 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 0
nov. 28 18:19:35 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 1
nov. 28 18:19:38 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 0
nov. 28 18:19:41 noname-opensuse kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 1
nov. 28 18:19:41 noname-opensuse systemd[1]: dracut pre-mount hook was skipped because no trigger condition checks were met.
nov. 28 18:19:41 noname-opensuse systemd[1]: Starting File System Check on /dev/disk/by-uuid/788c0329-b2d2-4765-a3ad-4fc78e9d7340...
nov. 28 18:19:41 noname-opensuse systemd[1]: Finished Virtual Console Setup.
nov. 28 18:19:41 noname-opensuse systemd[1]: Finished File System Check on /dev/disk/by-uuid/788c0329-b2d2-4765-a3ad-4fc78e9d7340.
nov. 28 18:19:41 noname-opensuse systemd[1]: Mounting /sysroot...
If it can help figuring out where the problem is … the full logs are here :
sudo journalctl --since "18:18" --until "18:19:42"
https://paste.opensuse.org/pastes/17d502a74e42
Resume from hibernation must be done in initrd before root (or any other filesystem) is mounted and dracut stores resume kernel parameter in generated initrd. You need to either recreate initrd or override stored dracut command line with noresume kernel parameter.
Could you try to add nvidia_drm.modeset=1 to the kernel command line?
Thank you for the clarification even if I’m not sure to follow … if I get this right the change I have made in yast → bootloader → kernel Parameters were not fully applied and it will be applied on the next kernel update or if I type a command like dracut --regenerate-all --force ?
To be honest I don’t really know what dracut does and it’s unlikely that this will solve the boot problem, I’ve checked all of no5xmega paste and it seems that it doesn’t have the resume= parameter set in the first place …
I can but just to be sure how do I do that ?
I supposed I should add it on the optional Kernel Command Line Parameters field in yast -> bootloader -> kernel Parameters and then click “OK” ?
Also another newby question, if after adding nvidia_drm.modeset=1 I can’t boot how to I revert the change from Grub ?
I’m really thankful for your help and sorry to have to ask questions about “trivial” procedure…
You can test it emporary. The parameter will only stay for one boot when you do it the following way.
Hit “e” in Grub2 boot menu. Type nvidia_drm.modeset=1 at the end of the line which starts with the word “linux”. After that, press “Ctrl+X” or “F10” to boot.
The same applies for a permanent kernel poarameter (set via yast -> bootloader -> kernel Parameters) which prevents a successful boot. Simply hit “e” at Grub2 boot menu, remove the parameter which prevents booting, hit “Ctrl+X”. After the system booted sucessfully, you can remove the permanent parameter via yast -> bootloader -> kernel Parameters
1 Like
Here an additional informative article how to add temporary/permanent parameters to the kernel comand line:
Thank you very much for the explanation, I tried to add nvidia_drm.modeset=1 from grub following your instruction ( laborious since I have a azerty keyboard and keyboard layout is in qwerty 
) and … no significative change, right after grub I have seen the flip event timeout error, then black screen, then three green dot loader …
Also again when shutting down the computer before making the change it did take times too, the screen turn off … on … off … then I can see the usual Tumbleweed logo with the spinner frozen and only then it’s over.
If that can help or at least confirm that nvidia_drm.modeset=1 was correctly applied
sudo journalctl --since "21:00" | grep "modeset"
nov. 28 21:17:18 noname-opensuse kernel: Command line: BOOT_IMAGE=/boot/vmlinuz-6.6.2-1-default root=UUID=788c0329-b2d2-4765-a3ad-4fc78e9d7340 splash=silent quiet security=apparmor nosimplefb=1 mitigations=auto nvidia_drm.modeset=1
nov. 28 21:17:18 noname-opensuse kernel: Kernel command line: BOOT_IMAGE=/boot/vmlinuz-6.6.2-1-default root=UUID=788c0329-b2d2-4765-a3ad-4fc78e9d7340 splash=silent quiet security=apparmor nosimplefb=1 mitigations=auto nvidia_drm.modeset=1
nov. 28 21:17:18 noname-opensuse dracut-cmdline[261]: Using kernel command line parameters: rd.driver.pre=btrfs resume=UUID=fe888bcc-e7b3-408a-9997-0f92bb766d73 root=UUID=788c0329-b2d2-4765-a3ad-4fc78e9d7340 rootfstype=btrfs rootflags=rw,relatime,ssd,discard=async,space_cache=v2,subvolid=1099,subvol=/@/.snapshots/776/snapshot,subvol=@/.snapshots/776/snapshot BOOT_IMAGE=/boot/vmlinuz-6.6.2-1-default root=UUID=788c0329-b2d2-4765-a3ad-4fc78e9d7340 splash=silent quiet security=apparmor nosimplefb=1 mitigations=auto nvidia_drm.modeset=1
nov. 28 21:17:19 noname-opensuse plymouthd[419]: 00:00:01.546 ply-utils.c:959:ply_get_kernel_command_line : Kernel command line is: 'BOOT_IMAGE=/boot/vmlinuz-6.6.2-1-default root=UUID=788c0329-b2d2-4765-a3ad-4fc78e9d7340 splash=silent quiet security=apparmor nosimplefb=1 mitigations=auto nvidia_drm.modeset=1
nov. 28 21:17:20 noname-opensuse kernel: nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms 545.29.06 Thu Nov 16 01:47:29 UTC 2023
nov. 28 21:18:33 noname-opensuse plymouthd[742]: 00:01:15.619 ply-utils.c:959:ply_get_kernel_command_line : Kernel command line is: 'BOOT_IMAGE=/boot/vmlinuz-6.6.2-1-default root=UUID=788c0329-b2d2-4765-a3ad-4fc78e9d7340 splash=silent quiet security=apparmor nosimplefb=1 mitigations=auto nvidia_drm.modeset=1
Thanks again I will take time to read it, your instructions were pretty clear.
The temporary parameter was applied right.
It was worth a try to test this parameter. But as it didn’t fix the issue…i’m out of ideas…
As arvidjaar explained, resume= is baked into the initrd. resume= as a Grub linu line parameter is simply an override. To defeat resume only for the current boot requires noresume replace it from within the Grub Editing process.
1 Like
Thanks for the additional information I’ve just put back the resume=xxx back … will probably look into it another day just in case I break something else …
I have seen discussion about adding a fbdev=1 in kernel parameters … I supposed it’s nvidia_drm.fbdev=1 or nvidia_drm.modeset=1 nvidia_drm.fbdev=1 … or nvidia_drm.modeset=1 fbdev=1 …
I’m not even sure what is the correct syntax but from the driver changelog from 535 to 545 Linux, Solaris, and FreeBSD driver 545.29.02 [new feature branch release] - Announcements and News - NVIDIA Developer Forums
- Added experimental support for framebuffer consoles provided by nvidia-drm. On kernels that implement drm_fbdev_generic_setup and drm_aperture_remove_conflicting_pci_framebuffers, nvidia-drm will install a framebuffer console when loaded with both
modeset=1andfbdev=1kernel module parameters. This will replace the Linux boot console driven by a system framebuffer driver such as efifb or vesafb.
Note that when an nvidia-drm framebuffer console is enabled, unloading nvidia-drm will cause the screen to turn off.
One user from the reddit thread I’ve posted before apparently resolved it’s issues like that but I have no idea what any of this is or does … I will probably try tomorrow …
In any case thanks again for your help.
You need to put those parameters in /etc/modprobe.d/50-nvidia-power-management.conf
Something like this:
#Power Management
options nvidia NVreg_PreserveVideoMemoryAllocations=1 NVreg_TemporaryFilePath=/path/to/tmp-nvidia
nvidia_drm.fbdev=1
Not sure exactly lf nvidia_drm.fbdev=1 is the correct parameter cause I have not tried it.
After doing those do
dracut -f --regenerate-all
1 Like
Not necessarely. This are parameters which work on kernel command line…the same as if you would put them into /etc/modprobe.d/50-nvidia-power-management.conf
2 Likes
@Aboutduck You can always check out the readme for Nvidia https://download.nvidia.com/XFree86/Linux-x86_64/545.29.06/README/
I use persistence mode here, always powered on 
2 Likes
Ya I know but at least the kernel parameters does not become too long.
Hi using it with the new .run installer if I dont stop it the installer will not succeed with failure to regenerate the initramfs error. I have to disable nvidia persistence first before rebooting to install the .run installer.
@conram yes, this is true… lots of options with the run file… mine are simple --no-systemd --ui=none
1 Like
Hi malcolm, not mean to hijack this thread, just need some clarity with the option --ui=none.
I had used the --systemd option before but with sh NVIDIA--------.run -A I am not able to see --ui=none
or I am just missing it. What is that option stands for. Thanks.