RTX 3050: dmaAllocMapping_GM107 VA space exhaustion → GPU lockup on nvidia-open 610.57.04 (Hyprland/Wayland)

Open
  • Is something wrong?
#1,411 3 comments 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
5/5
Estimated time
Over a week
Newbie friendliness
25/100
Issue type
Bug
Clarity
Needs clarification
Activity status
Active
Tech stack
arch-linux, c, linux
Domain
computer-graphics, operating-systems

Research direction

The report names no source file or test. Start by collecting the full nvidia-bug-report.sh output and journal around the dmaAllocMapping_GM107 failures, then compare the symptom chain with issue #1140; done would require identifying and fixing the DMA VA exhaustion path and verifying that the Wayland video-decode workload no longer locks up the GPU.

Written by the indexing model from the issue text.

Description

System Information

Component Value
GPU NVIDIA GeForce RTX 3050 (GA106)
VBIOS 94.07.A0.00.28
Driver nvidia-open 610.57.04-1 (Arch nvidia-open-dkms, open kernel module)
Kernel 7.2.5-3-omarchy (linux-omarchy)
CPU AMD (x86_64)
OS Arch Linux (Omarchy)
Display server Hyprland (Wayland session)
Module options nvidia_drm modeset=1

Bug Description

After roughly 12 hours of normal desktop use (Wayland compositor, browser, terminals), the nvidia-open driver began failing DMA VA space allocations, eventually locking up the GPU and forcing a hard reboot.

Timeline

  1. Oct 02 01:20:52 — WebKitWebProcess segfault inside libnvidia-eglcore.so (repeating pattern; all of the night)
  2. Oct 02 01:33:45 — Hyprland session restart loop (xkbcomp spam in journal)
  3. Oct 02 01:34:11 — Burst of NVRM: dmaAllocMapping_GM107: can't alloc VA space for mapping. (~13 times), followed by: 
    NVRM: nvAssertOkFailedNoLog: Assertion failed: Out of memory [NV_ERR_NO_MEMORY] (0x00000051) returned from pReuseMappingDb->pMapCb(...) @ mapping_reuse.c:273
[drm] *ERROR* Failed to initialize semaphore for plane fence
[drm] *ERROR* Failed to apply atomic modeset.  Error code: -11
  4. Oct 02 01:34:41 — NVRM: kgmmuInvalidateTlb_GM107: TLB invalidation failed waiting for completion (status=0x00000065) — journal stops, GPU hung, hard reboot required

Notes

  • No Xid errors logged
  • GSP firmware 610.57.04
  • This is the same symptom chain as #1140 (RTX 4090, 595.58.03) and other reports of the "580–595/610 regression" in DMA VA space management
  • Brave SIGILL trap invalid opcode crashes were logged concurrently (browser GPU/media process), same trigger window
  • Reproducible on hardware video-decode workloads under Wayland

Logs

Full coredumpctl + journal snippets available on request. Attached output of nvidia-bug-report.sh if needed.

Dominant language
C
Stars
17.5k
Forks
1.9k
No merged PRs in 30d

Getting set up

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from NVIDIA/open-gpu-kernel-modules

All issues in NVIDIA/open-gpu-kernel-modules

Similar issues

More C issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.