Proof Portal

Project overview

Omarchy

ProbeLabs65 findings · 95 requirements

A 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.

Inbox
ReportR-855GitHub issuePublicDecided· Out of scope

Tiled dual-DP 5K (LG UltraFine): stock preferred mixed-modeset blacks out laptop and external; modeless recovery misses all-disabled

Reported by bjcatar · filed Sep 7, 2026 · open at source

Decided: Out of scope — decided by agents · verified 3 days ago.

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?
AreaOutside the audited area· Decided by agents · verified
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?
Decided by agents and verified by a second, independent agent — see the decision.

Original report

Tiled dual-DP 5K (LG UltraFine): stock preferred mixed-modeset blacks out laptop and external; modeless recovery misses all-disabledrevision 1 · captured 2026-09-28 21:23 UTC via poll

System details

  • Machine: Dell XPS 14 DA14260
  • CPU/GPU: Intel Core Ultra X7 358H, Panther Lake Arc B390 (xe), DRM connectors eDP-1, DP-1, DP-2, DP-3
  • Display: LG UltraFine 5K over Thunderbolt 3 (boltctl: LG Electronics UltraFine 5K)
  • Omarchy: 4.0.2-1 (Quattro)
  • Hyprland: 0.56.2 / Aquamarine 0.14.0
  • Kernel: 7.1.8-arch1-2-ptl
  • Stock catch-all: hl.monitor({ output = "", mode = "preferred", position = "auto-up", scale = "auto" })

What happened

Plugging the Thunderbolt cable of the LG UltraFine 5K (lid open, wanting laptop + external) makes every output go black. Lock surfaces fail (QScreen name="" / FALLBACK). The only recovery is a sleep/wake cycle (close lid, wait, open — or unplug something).

The UltraFine 5K is a 2×1 tiled panel. Aquamarine logs:

Connector DP-1 tile info: group=1 tiles=2x1 pos=(0,0) size=2560x2880
Connector DP-2 tile info: group=1 tiles=2x1 pos=(1,0) size=2560x2880
Tile group 1 has 2 members, full tiled resolution 5120x2880
Mode 0 on DP-1: [email protected]  (preferred)
Mode 2 on DP-1: [email protected]
Mode 0 on DP-2: [email protected]

Stock mode = "preferred" then does a mixed modeset:

Modesetting DP-1 with [email protected]
Modesetting DP-2 with [email protected]
Disabling output eDP-1

Then both DP tiles drop, eDP stays disabled, and the session has zero usable outputs. Kernel follows with:

thunderbolt ... failed to reach state TB_PORT_UP. Ignoring port...
thunderbolt ... lost during suspend, disconnecting

omarchy-hyprland-monitor-modeless does not recover this: it only treats an enabled monitor with width==0 as sick. All-disabled / FALLBACK exits 1 (“healthy”). omarchy-hyprland-monitor-internal recover also no-ops unless the manual disable toggle is set.

Forcing both tiles to 2560x2880@60 still disabled eDP the moment the second tile came up (three display pipes: laptop + two tiles). The TB link still dropped. Driving DP-1 as 4K SST and disabling the second tile is the only layout that has a chance on this GPU without dropping the laptop panel.

Same class of “stock preferred is the wrong mode on a Thunderbolt 5K” as #7388 (Studio Display 5K@120) and #7939 (LG UltraFine 4K preferred-black). This report is the tiled dual-stream 5K + laptop panel case, plus the recovery hole for all-outputs-disabled.

Also adjacent: #7228 (internal stays disabled after unplug), #9753 (stale ACPI lid → clamshell blanks eDP on hotplug), #7853 (clamshell disable + hyprctl reload kills clients).

What was expected

  • Plug Thunderbolt cable with lid open → laptop panel stays on, external comes up (or 4K SST if 5K tiled + eDP cannot fit on the GPU).
  • If the external modeset fails, eDP must come back without a lid cycle.
  • Charging over the same USB-C/Thunderbolt port must not interact with this (see companion lid/PD report).

Steps to reproduce

  1. Omarchy 4 laptop with Intel Xe, stock monitors.lua catch-all mode = "preferred".
  2. Lid open.
  3. Plug LG UltraFine 5K (or any 2×1 tiled Thunderbolt 5K) into USB-C.
  4. Screen goes black; hyprctl monitors may show FALLBACK / disabled eDP; journalctl shows There are no outputs - creating placeholder screen.

Suggested Omarchy-side fix

  • Do not apply preferred independently to each tile of a 2×1 group. Either pin both tiles to 2560x2880@60 side-by-side, or disable the second tile and use the first tile’s 4K SST mode when an internal panel is also active.
  • Keep eDP enabled on lid-open hotplug (do not steal its CRTC / do not clamshell unless lid is confirmed closed).
  • Teach omarchy-hyprland-monitor-modeless / monitor-watch to recover zero enabled outputs, not only 0×0 enabled heads: re-enable the internal panel + DPMS on, without hyprctl reload (reload races the TB modeset).
  • Document that Thunderbolt 5K tiled displays need an explicit monitors.lua on Intel Xe laptops; stock “preferred/auto” is not safe.

Apple’s 2026 Studio Display / Studio Display XDR are the same tiled-5K architecture, so this will keep hitting people who plug those in on Omarchy laptops.

View raw text
### System details

- **Machine:** Dell XPS 14 DA14260
- **CPU/GPU:** Intel Core Ultra X7 358H, Panther Lake Arc B390 (`xe`), DRM connectors `eDP-1`, `DP-1`, `DP-2`, `DP-3`
- **Display:** LG UltraFine 5K over Thunderbolt 3 (`boltctl`: LG Electronics UltraFine 5K)
- **Omarchy:** 4.0.2-1 (Quattro)
- **Hyprland:** 0.56.2 / Aquamarine 0.14.0
- **Kernel:** 7.1.8-arch1-2-ptl
- **Stock catch-all:** `hl.monitor({ output = "", mode = "preferred", position = "auto-up", scale = "auto" })`

### What happened

Plugging the Thunderbolt cable of the LG UltraFine 5K (lid open, wanting laptop + external) makes **every** output go black. Lock surfaces fail (`QScreen name=""` / `FALLBACK`). The only recovery is a sleep/wake cycle (close lid, wait, open — or unplug something).

The UltraFine 5K is a **2×1 tiled** panel. Aquamarine logs:

```
Connector DP-1 tile info: group=1 tiles=2x1 pos=(0,0) size=2560x2880
Connector DP-2 tile info: group=1 tiles=2x1 pos=(1,0) size=2560x2880
Tile group 1 has 2 members, full tiled resolution 5120x2880
Mode 0 on DP-1: [email protected]  (preferred)
Mode 2 on DP-1: [email protected]
Mode 0 on DP-2: [email protected]
```

Stock `mode = "preferred"` then does a **mixed modeset**:

```
Modesetting DP-1 with [email protected]
Modesetting DP-2 with [email protected]
Disabling output eDP-1
```

Then both DP tiles drop, eDP stays disabled, and the session has **zero usable outputs**. Kernel follows with:

```
thunderbolt ... failed to reach state TB_PORT_UP. Ignoring port...
thunderbolt ... lost during suspend, disconnecting
```

`omarchy-hyprland-monitor-modeless` does not recover this: it only treats an *enabled* monitor with `width==0` as sick. All-disabled / `FALLBACK` exits 1 (“healthy”). `omarchy-hyprland-monitor-internal recover` also no-ops unless the *manual* disable toggle is set.

Forcing both tiles to `2560x2880@60` still disabled eDP the moment the second tile came up (three display pipes: laptop + two tiles). The TB link still dropped. Driving DP-1 as 4K SST and disabling the second tile is the only layout that has a chance on this GPU without dropping the laptop panel.

Same class of “stock preferred is the wrong mode on a Thunderbolt 5K” as #7388 (Studio Display 5K@120) and #7939 (LG UltraFine 4K preferred-black). This report is the **tiled dual-stream 5K + laptop panel** case, plus the recovery hole for all-outputs-disabled.

Also adjacent: #7228 (internal stays disabled after unplug), #9753 (stale ACPI lid → clamshell blanks eDP on hotplug), #7853 (clamshell disable + `hyprctl reload` kills clients).

### What was expected

- Plug Thunderbolt cable with lid open → laptop panel stays on, external comes up (or 4K SST if 5K tiled + eDP cannot fit on the GPU).
- If the external modeset fails, **eDP must come back** without a lid cycle.
- Charging over the same USB-C/Thunderbolt port must not interact with this (see companion lid/PD report).

### Steps to reproduce

1. Omarchy 4 laptop with Intel Xe, stock `monitors.lua` catch-all `mode = "preferred"`.
2. Lid open.
3. Plug LG UltraFine 5K (or any 2×1 tiled Thunderbolt 5K) into USB-C.
4. Screen goes black; `hyprctl monitors` may show FALLBACK / disabled eDP; `journalctl` shows `There are no outputs - creating placeholder screen`.

### Suggested Omarchy-side fix

- Do not apply `preferred` independently to each tile of a 2×1 group. Either pin both tiles to `2560x2880@60` side-by-side, or disable the second tile and use the first tile’s 4K SST mode when an internal panel is also active.
- Keep eDP enabled on lid-open hotplug (do not steal its CRTC / do not clamshell unless lid is *confirmed* closed).
- Teach `omarchy-hyprland-monitor-modeless` / monitor-watch to recover **zero enabled outputs**, not only 0×0 enabled heads: re-enable the internal panel + DPMS on, without `hyprctl reload` (reload races the TB modeset).
- Document that Thunderbolt 5K tiled displays need an explicit `monitors.lua` on Intel Xe laptops; stock “preferred/auto” is not safe.

Apple’s 2026 Studio Display / Studio Display XDR are the same tiled-5K architecture, so this will keep hitting people who plug those in on Omarchy laptops.

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.

  • Out of scopeDecided by agents · verified · 2026-10-09 11:05 UTC