NVIDIA / NVIDIA/open-gpu-kernel-modules
Xid 109 `CTX SWITCH TIMEOUT` on RTX 5090 / Blackwell GB202 during Forza Horizon 6 via Proton — open kernel module 595.71.05
Nobody has claimed this yet.
- Dominant language
- C
- Stars
- 17.4k
- Forks
- 1.9k
- PR merge metrics
- No merged PRs in 30d
Description
NVIDIA Open GPU Kernel Modules Version
595.71.05 (Arch package nvidia-open-dkms 595.71.05-2).
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
CachyOS Linux (rolling).
Kernel Release
Linux 7.0.6-1-cachyos #1 SMP PREEMPT_DYNAMIC Mon, 11 May 2026 15:10:22 +0000 x86_64 GNU/Linux
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
GPU 0: NVIDIA GeForce RTX 5090 (UUID: GPU-[REDACTED-UUID])
Describe the bug
Forza Horizon 6 (Steam AppID 2483190, build 354221, run via Proton) hangs the graphics context after ~3 minutes of open-world gameplay. Every occurrence raises an Xid 109 CTX SWITCH TIMEOUT with a byte-identical signature across runs, then the game receives DXGI DEVICE_REMOVED (HRESULT 0x00000102 → WAIT_TIMEOUT) and writes a crash report.
The Xid is the first and only kernel-side fault. The user-mode crash (ACCESS_VIOLATION_WRITE while writing to a device-removed swapchain resource) is a downstream consequence, not the trigger.
Selected kernel log:
May 16 00:00:33 [REDACTED-HOST] kernel: NVRM: GPU at PCI:0000:01:00: GPU-[REDACTED-UUID]
May 16 00:00:33 [REDACTED-HOST] kernel: NVRM: Xid (PCI:0000:01:00): 109, pid=159175, name=forzahorizon6.e, channel 0x00000029, errorString CTX SWITCH TIMEOUT, Info 0x3c03f
May 16 00:05:20 [REDACTED-HOST] kernel: NVRM: Xid (PCI:0000:01:00): 109, pid=174356, name=forzahorizon6.e, channel 0x00000029, errorString CTX SWITCH TIMEOUT, Info 0x30c03f
May 16 00:42:45 [REDACTED-HOST] kernel: NVRM: Xid (PCI:0000:01:00): 109, pid=98371, name=forzahorizon6.e, channel 0x0000001f, errorString CTX SWITCH TIMEOUT, Info 0x3c02c
(Three confirmed events across two boots — full history in attached log bundle. The channel index varies; Info always has the low byte 0x3f / 0x2c family, suggesting the same fault-class.)
The hang appears to be driver-side rather than a translation-layer artifact:
- Reproduces under both
proton-cachyos-slr(custom Proton with cherry-picked vkd3d-proton commits) and stock Valve Proton (all provided by Valve which can run Forza Horizon 6 is tested; same error produced). - VRAM usage at crash is far below budget — last run reported 4.9 GB used out of 29.0 GB budget (see attached
CrashReport.xml). - Crashes occur both with the in-game High preset and Extreme preset, with DLSS UltraPerformance through Quality, at 4K and 1440p, with/without ray tracing.
- Re-running with the very-low preset and minimum supported settings did not avoid the crash (history: prior session reproduced with Extreme + DLSS UltraPerformance at 4K, current session reproduced with High + DLSS UltraPerformance at 4K and 72 fps cap).
- No NVIDIA assert lines, no GPU-Z page-fault, no DPC watchdog trigger — the channel just stops switching context.
The pattern matches general descriptions of GSP-firmware context scheduler stalls reported elsewhere on Blackwell; this filing is intended to give NVIDIA a clean reproducer plus the requested diagnostic bundle.
To Reproduce
- RTX 5090 / GB202 on the open kernel module 595.71.05, stable linux-cachyos 7.0.6 (Arch-based, KDE Plasma).
- Install Forza Horizon 6 (Steam AppID 2483190).
- Set Steam Compatibility for FH6 to proton-cachyos-11.0-20260429-slr (or stock Proton-Experimental, GE-Proton10-34, Proton 10.0 -4, etc. — both reproduce). Set launch options to
PROTON_LOG="+timestamp,+seh,+loaddll" VKD3D_DEBUG=warn %command%. - Launch the game and reach the open world. Drive any vehicle for 3 to 5 minutes in any region.
- The game freezes; dmesg /
journalctl -k -g XidshowsXid 109 CTX SWITCH TIMEOUT. The game writes~/.../ForzaHorizon6/CrashReport.xmlwithREASON="Video card"andREASON_INFO="0x0000000000000102". - The system remains responsive; only the FH6 process is wedged. No reboot required.
Bug Incidence
Always
nvidia-bug-report.log.gz
More Info
No response
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reproducing the hang with Forza Horizon 6 on the specified RTX 5090 and open module version, then inspect journalctl -k -g Xid and the attached nvidia-bug-report.log.gz. Compare the Xid 109 events and crash report across the listed Proton configurations; the investigation is complete when the context-switch failure is explained or a verified driver fix is identified.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- c, linux
- Domain
- computer-graphics, operating-systems
- Issue type
- Bug
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 32/100