GA104GLM (RTX A5000 Laptop GPU) GSP core crashes after seemingly random amounts of time

NVIDIA Open GPU Kernel Modules Version

610.57.04

Please confirm this issue does not happen with the proprietary driver (of the same version). This issue tracker is only for bugs specific to the open kernel driver.

• I confirm that this does not happen with the proprietary driver package.

Operating System and Version

Fedora Linux 44 (KDE Plasma Desktop Edition)

Kernel Release

7.1.13-200.fc44.x86_64

Please confirm you are running a stable release kernel (e.g. not a -rc). We do not accept bug reports for unreleased kernels.

• I am running on a stable kernel release.

Hardware: GPU

NVIDIA RTX A5000 Laptop GPU (UUID: GPU-57bba7ee-87aa-9a66-e754-e2b6db36d318)

Describe the bug

GPU becomes unresponsive and causes excessive power consumption or prevents proper system resume from S0ix sleep. When this occurs the GPU cannot be reset through nvidia-smi and the machine requires a restart to return GPU functionality.

Sleep issues manifest as the system heating up with the lid closed - upon reopening the lid I do not see anything, the backlight does not light up but the keyboard responds until I attempt to change TTYs, wherein the system will fully crash and the system heats up further rapidly. Only a hard shutdown fixes this.
If the system is open while the GPU crashes, I can still use the system on integrated graphics to grab bug reports/etc. but it will freeze after the screen turns off during a restart. Programs attempting to use the dedicated NVIDIA GPU will hang.

To Reproduce

Indeterminate - bug has occurred both under load and at idle, regardless of CUDA or graphics workload.
Bug however will always occur at least once per system boot.

Bug Incidence

Always

nvidia-bug-report.log.gz

0904.gz from September 4th 2026 UTC -6

More Info

Laptop display configuration has built-in LCD driven off Intel iGPU. Similar behaviours have been exhibited by other cards in other machines - with this machine I have seen these behaviours on both the 580 proprietary driver and the 595 open driver.
System does not support S3 sleep (no sleep options are presented in UEFI - forcing it breaks suspend and resume).

Start by inspecting the linked 0904.gz bug report and comparing the reported GPU failures during idle, load, reset, and S0ix resume. The issue does not name a source file or test; progress would require a reproducible failure and identification of the component or change that prevents the crash.