Skip to content

Boot stuck due to AMDGPU memory leak after NixOS bump #462516

Description

@msanft

After bumping my system from e99366c to 85a6c4a, my system got stuck in my motherboard's boot splash screen after unlocking the disk.

No firmware updates took place, so the cause is likely in userspace. After re-booting into an older generation, everything worked again. I examined the dmesg output and was presented with floods of the following:

Nov 17 09:03:09 bastion kernel: [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Not enough memory for command submission!
Nov 17 09:03:09 bastion kernel: [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Not enough memory for command submission!
Nov 17 09:03:09 bastion kernel: [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Not enough memory for command submission!
Nov 17 09:03:09 bastion kernel: [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Not enough memory for command submission!
Nov 17 09:03:09 bastion kernel: [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Not enough memory for command submission!

Reading about this for a little, it seems that this error is typically caused by userspace programs leaking graphics memory.

This issue occurred on a setup with a AMD Ryzen 9 9950X CPU and a NVIDIA RTX 3070 GPU, running the latest kernel (i.e., linuxPackages_latest).

Metadata

Metadata

Assignees

No one assigned

    Labels

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions