Proof Portal
Omarchy
ProbeLabs65 findings · 95 requirementsA proof layer — requirements, tests and verified fixes — for two of Omarchy's subsystems: the application menu (launcher scripts, QML model, JSONC config, search and selection) and the lock screen (lock scripts, QML, PAM authentication). Scope is deliberately limited to those components of omacom/omarchy; the rest of the distribution is not covered.
Stable mirror contains broken linux-firmware-amdgpu 20260810-1
Reported by mwheelersmith · filed Aug 19, 2026Closed upstream on 2026-08-21 by its own reporter mwheelersmith. GitHub close reason: completed. Closed by its own reporter — not by itself evidence of a fix.
At a glance
- 1 · What is claimed?
- Only the report text below — no validator has restated it as checkable claims yet.
- 2 · What does it affect?
- Not yet mapped to any requirement or component.
- 3 · Is it reproducible?
- No validation has been run yet.
- 4 · What remains unclear?
- Not validated yet: no claims, no reproducer.
- 5 · What happens next?
- Our staff decide whether the report is valid. Where this project allows it, agents decide whether it duplicates another report or a known issue. No decision has been recorded yet.
Original report
System details
AMD 9850X3D, Radeon RX 9070 XT, Omarchy 3.8.4
What's wrong?
Stable mirror contains broken linux-firmware-amdgpu 20260810-1
After a recent update I noticed colour corruption on some higher-resolution AV1 videos in both Firefox and Chromium on an AMD RX 9070 XT.
The issue appears as rectangular sections of the video displaying incorrect colours. Lower-resolution streams, such as 480p, appear unaffected.
After some investigation, this appears to match an upstream regression in:
linux-firmware-amdgpu 20260810-1My system is currently using the Omarchy stable mirror:
Server = https://stable-mirror.omarchy.org/$repo/os/$archand it currently provides:
Repository : core
Version : 20260810-1Arch has since released 20260810-2, which appears to fix the AV1 corruption reported by other AMD users.
For comparison, the repository metadata currently reports:
Omarchy stable core.db
last-modified: Fri, 14 Aug 2026 15:15:59 GMT
Omarchy edge core.db
last-modified: Tue, 18 Aug 2026 18:28:39 GMTI appreciate that holding packages back is intentional for the stable mirror. I'm raising this because it looks like the stable snapshot happened to capture the broken firmware shortly before the upstream fix became available.
Would it be worth bumping linux-firmware-amdgpu to 20260810-2 on the stable mirror?
Additional testing
- GPU: AMD Radeon RX 9070 XT
- Mesa:
26.1.7-1 - libva:
2.24.1-1 - Reproduced in both Firefox and Chromium
- Affected 4K stream:
av01.0.12M.08 - Unaffected 480p stream:
av01.0.04M.08 - Disabling Firefox hardware video decoding did not resolve the issue
- The corruption is visible in screenshots, not just on the physical display
<img width="1884" height="1159" alt="Image" src="https://github.com/user-attachments/assets/eb3dc8db-ed56-4ee2-aa34-6b93a472a67b" />
References
CachyOS report of the same colour corruption - multiple AMD users, including RX 9070 XT owners, report the same issue with 20260810-1 and confirm it is fixed by 20260810-2.
Arch Linux linux-firmware-amdgpu package - upstream Arch package containing the updated AMD GPU firmware.
View raw text
### System details AMD 9850X3D, Radeon RX 9070 XT, Omarchy 3.8.4 ### What's wrong? # Stable mirror contains broken `linux-firmware-amdgpu 20260810-1` After a recent update I noticed colour corruption on some higher-resolution AV1 videos in both Firefox and Chromium on an AMD RX 9070 XT. The issue appears as rectangular sections of the video displaying incorrect colours. Lower-resolution streams, such as 480p, appear unaffected. After some investigation, this appears to match an upstream regression in: ```text linux-firmware-amdgpu 20260810-1 ``` My system is currently using the Omarchy stable mirror: ```text Server = https://stable-mirror.omarchy.org/$repo/os/$arch ``` and it currently provides: ```text Repository : core Version : 20260810-1 ``` Arch has since released `20260810-2`, which appears to fix the AV1 corruption reported by other AMD users. For comparison, the repository metadata currently reports: ```text Omarchy stable core.db last-modified: Fri, 14 Aug 2026 15:15:59 GMT Omarchy edge core.db last-modified: Tue, 18 Aug 2026 18:28:39 GMT ``` I appreciate that holding packages back is intentional for the stable mirror. I'm raising this because it looks like the stable snapshot happened to capture the broken firmware shortly before the upstream fix became available. Would it be worth bumping `linux-firmware-amdgpu` to `20260810-2` on the stable mirror? ### Additional testing * GPU: AMD Radeon RX 9070 XT * Mesa: `26.1.7-1` * libva: `2.24.1-1` * Reproduced in both Firefox and Chromium * Affected 4K stream: `av01.0.12M.08` * Unaffected 480p stream: `av01.0.04M.08` * Disabling Firefox hardware video decoding did not resolve the issue * The corruption is visible in screenshots, not just on the physical display <img width="1884" height="1159" alt="Image" src="https://github.com/user-attachments/assets/eb3dc8db-ed56-4ee2-aa34-6b93a472a67b" /> References [CachyOS report of the same colour corruption](https://www.reddit.com/r/cachyos/comments/1vne5nm/youtube_maybe_all_videos_look_off_after_updating/) - multiple AMD users, including RX 9070 XT owners, report the same issue with 20260810-1 and confirm it is fixed by 20260810-2. [Arch Linux ](https://archlinux.org/packages/core/any/linux-firmware-amdgpu/)[linux-firmware-amdgpu](https://archlinux.org/packages/core/any/linux-firmware-amdgpu/)[ package](https://archlinux.org/packages/core/any/linux-firmware-amdgpu/) - upstream Arch package containing the updated AMD GPU firmware.
- eb3dc8db-ed56-4ee2-aa34-6b93a472a67b
Validation
Not checked yet. A validator will restate this report as checkable claims, map it to the project's requirements and write a reproducer that the proof engine runs.
Linked
Not linked to any known issue, other report or change yet.
Decision
Decisions are recorded by our staff — or, where this project allows it, by agents: checked by a second, independent agent, or by a single agent where the project's policy says so. Any decision reached so far is shown below.
No decision has been recorded yet.