### System details

- Omarchy 4.0.4-1
- CPU: AMD Ryzen 7 3800X
- GPU: AMD Radeon RX 9060 XT (Navi 44), `amdgpu`
- Running kernel: `7.2.5-3-omarchy`
- Hyprland 0.56.2-2; Aquamarine 0.15.0-2
- Quickshell 0.3.1-1
- Mesa 26.2.2-1
- Qt base 6.11.2-3; Qt declarative 6.11.2-1
- Suspend mode: `deep` (`/sys/power/mem_sleep`: `s2idle [deep]`)

### What's wrong?

After waking from system suspend, scattered small colored dots appear on the desktop wallpaper. Rebooting clears the wallpaper dots.

Similar dots have also appeared occasionally in the Steam UI, although much less often. Their timing relative to suspend has not been established. The issue is therefore not confirmed to be confined to the wallpaper. It is also not established whether Steam contributes to the problem or is another affected application.

The dots are visible in a desktop screenshot, not just on the physical display. The original wallpaper image was inspected separately and looks clean. The captured corruption consists of sparse red, green, and blue specks over an otherwise correctly rendered image.

### Steps to reproduce

1. Use a static image as the Omarchy desktop wallpaper.
2. Suspend the system using deep sleep.
3. Resume and inspect the wallpaper: colored specks can appear. Similar specks have occasionally been seen in Steam, but there is no reliable reproduction sequence for that symptom yet.
4. Reboot: the wallpaper renders normally again.

Frequency has not been measured; this is not claimed to occur after every suspend.

### Expected behavior

The wallpaper remains visually unchanged after suspend/resume.

### Diagnostics and investigation

- For the captured occurrence, the screenshot was saved 27 seconds after the logged suspend exit.
- The checked kernel journal showed successful AMD SMU resume and no GPU ring timeout, AMD GPU fault, or explicit VRAM-loss report. This does not rule out a driver problem.
- The installed Omarchy wallpaper plugin uses a cached Qt Quick `Image`. In `shell/plugins/background/Background.qml`, `transitionBackground()` returns early when the requested path matches `currentBackground` and `force` is false. Consequently, requesting the same wallpaper is not necessarily a fresh image load.
- Corrupted textures or render buffers after resume remain a hypothesis, not a confirmed root cause. The additional Steam observation makes a shared graphics-stack problem (such as Mesa/amdgpu or Hyprland) more plausible than a fault confined to Omarchy's wallpaper renderer, but it does not establish the responsible component or prove hardware failure.
- Restarting only the shell, switching wallpapers, and comparing other kernels/rendering backends have **not** been tested. Only full reboot recovery is confirmed.

Related report #12041 describes whole-screen corruption on a hybrid AMD/NVIDIA laptop using s2idle. This report concerns sparse specks on the wallpaper and occasionally in Steam on an AMD desktop using deep sleep; it is not established to be the same bug.

`omarchy debug --no-sudo --print` was attempted, but this installation reports that the command is unknown. Relevant diagnostics are summarized above; full system logs are omitted to avoid disclosing personal information.