-
Notifications
You must be signed in to change notification settings - Fork 1.6k
Open
Description
System Information
- OS: CachyOS (Arch-based)
- Kernel: 6.18.2-2-cachyos
- Driver: nvidia-open 590.48.01
- GPU: NVIDIA GeForce RTX 2080 Ti (TU102)
- Desktop: KDE Plasma 6 (Wayland/KWin)
- Monitor: ASUS XG27ACDMS OLED (2560x1440 @ 144Hz via DisplayPort 1.4)
Description
Random Xid 32 errors causing complete GPU context loss and black screen on KDE Plasma Wayland. The system becomes unresponsive and requires a hard reboot.
Key observation: This issue does NOT occur with an LCD monitor on the same system. The OLED monitor works flawlessly on Windows 11 at full refresh rate (280Hz).
Error Logs
NVRM: GPU at PCI:0000:09:00: GPU-4dfe1525-6c08-3c15-69b4-4141c28dfde6
NVRM: Xid (PCI:0000:09:00): 32, pid=1561, name=kwin_wayland, channel 0x00000059 intr1 00000008 HCE_DBG0 00000c00 HCE_DBG1 00000000
NVRM: Xid (PCI:0000:09:00): 32, pid=1561, name=kwin_wayland, channel 0x00000059 intr1 00000008 HCE_DBG0 00000c04 HCE_DBG1 8210c000
Followed by:
kwin_wayland: Create Context failed "EGL_BAD_CONTEXT"
kwin_wayland: QRhiGles2: Failed to create temporary context
kwin_wayland: Applying output configuration failed!
Steps to Reproduce
- Connect ASUS XG27ACDMS OLED monitor via DisplayPort
- Run KDE Plasma on Wayland with nvidia-open driver
- Use system normally - crash occurs randomly (sometimes after DPMS wake, sometimes during normal use)
Workarounds Attempted (none successful)
NVreg_PreserveVideoMemoryAllocations=1NVreg_EnableGpuFirmware=0- Lower refresh rate (144Hz instead of 280Hz)
- VRR disabled on monitor
- DisplayPort 1.4 mode (not 2.0/2.1)
Additional Context
- Secondary monitor (1080p LCD) connected to AMD RX 6400 on same system
- Issue is specific to OLED + nvidia-open + Wayland combination
- Crashes are random - not always tied to DPMS/suspend
Metadata
Metadata
Assignees
Labels
No labels