[575.57.08] Suspend fails

Hello, so i’m trying to put my laptop into sleep, but NVIDIA driver freaks out and doesn’t let me. This sometimes happen, sometimes don’t. Turned off PreverseVideoMemoryAllocations. I’m attaching the dmesg log, that clearly shows nvidia-modeset shitting itself while trying to suspend. Getting it outside this state requires a hard reboot as systemd also shits itself.

OS: Arch Linux
GPU: RTX 3050 Mobile
Host: ASUS ROG Strix G15 (G513IC)
Kernel version: 6.15.2-arch1-1

dmesg.txt (125.9 KB)

1 Like

I have an asus zephyrus g14 2021 ga401 and I found that, using the drivers installed from the nvidia ubuntu 24.04 repo, hibernation worked when I used kernel 6.11, but when I updated to 6.14, hibernation stopped working.

I have a dell latitude with kernel 6.14 and the nvidia drivers from the repo, and hibernation works there.

Maybe something changed in the kernel between 6.11 and 6.14 that breaks things for asus laptops

Hi @realmotylek and @taco-bell-5-layer-burrito , thanks for reporting this issue.
Is this seen with the latest 575.64 driver?
Could you attach a bug report here, I will take a look and get back to you.

I think i have a similar issue too. its been there ever since i updated to 575.xx its also seems to be there in 575.64 one the laptop isnt used for a while the gpu suspends but after that will not awake again.
typing nvidia-smi in the console will give me this error
nvidia-smi
Unable to determine the device handle for GPU0: 0000:01:00.0: Unknown Error
No devices were found

This was issue was there long back in some other driver version too but has resurfaced now.

Where to get the bug report that i should attach?

From the directory where you launch..

$ sudo nvidia-bug-report.sh
1 Like

I’m not sure if this is related to my issue, because the GPU stops working without putting the laptop to sleep. It happens spontaneously, regardless of what I’m doing.
nvidia-bug-report.log.gz (492.3 KB)

Distribution : “Fedora Linux 42 (KDE Plasma Desktop Edition)”
Kernel : 6.14.11-300.fc42.x86_64
NVIDIA version : 575.64

Hi @yanex013 , I think your issue is different from the one reported above. Could you try after applying @aplattner 's patch : suspend/resume fixes · NVIDIA/open-gpu-kernel-modules@c7e7213 · GitHub . This fix will be present in the next driver release.

1 Like

Not sure if this is the same issue, but my laptop also has trouble suspending. It happens rarely, but when it does, the only way to get the GPU working again is a hard reboot, because SDDM can’t complete the stop job. I’ve tried to figure out what conditions cause the driver to crash, but so far it still seems random. Everything was working fine before a full system update, which included an upgrade of the NVIDIA driver from 570.153.02-2.

Distro: EndeavourOS KDE + Wayland
GPU: RTX 3050 Ti Mobile (v575.64)
Host: Acer Swift X 14 2021 (SFX14-41G-R1S6)
Kernel: 6.15.3-arch1-1

nvidia-bug-report.log.gz (1.2 MB)

I have a Lenovo Legion Pro 5 laptop with an RTX5060. For me, moving from Ubuntu 24.04 with kernel 6.11 to ubuntu 25.04 with kernel 6.14 and typing

sudo systemctl enable nvidia-suspend.service

Solved the issue, with nvidia driver 575.57.08, open flavor for both ubuntu versions.

On both ubuntu versions, sleep works fine before installing nvidia drivers. However, after installing nvidia drivers in ubuntu 24.04, whenever I suspended, it would send me back to the login screen with everything being extremely laggy. When I switched to 25.04, it would also send me back to the login screen, but it would not be laggy. However, when I typed the above command and reset the computer, suspend worked.

UPDATE:

Tried it on ubuntu 22.04 LTS with the same machine (installed my previous laptop’s hard drive) with the same 575.57.08 driver. Had same sleep issue as with 24.04. Running the above systemctl command solved the issue completely. Therefore, it might not be necessary to change distro version or kernel version.

Unfortunately, updating to 575.64.03 did not resolve the issue — after some time, nvidia-smi still returns:

Unable to determine the device handle for GPU0: 0000:01:00.0: Unknown Error
No devices were found.

nvidia-bug-report.log.gz (586.5 KB)