Repository navigation
Conversation
Five seconds runs out while a slow monitor is still waking from DPMS, so the password prompt disappears as soon as it appears. Ten seconds leaves time to see it and start typing.
[Medium impact] The change affects the lock screen’s idle display behavior. The code change appears sound, but the repository's required running-UI timing check must be completed before merging. Findings
SummaryExtends the lock screen's idle blank delay from five to ten seconds.
Reviews (1) · Last reviewed commit: "Give the lock screen ten seconds before ..." · Reviewed by Greptile |
| // Five seconds runs out while a slow panel is still resyncing after DPMS | ||
| // off, so the lock blanks again as the picture appears. Ten seconds leaves | ||
| // a visible window to start the password. | ||
| readonly property int idleMs: 10000 |
There was a problem hiding this comment.
The new ten-second window has not been checked in the running lock screen. AGENTS.md requires running-UI verification for visual changes, and its linked guide requires a short recording for timing changes. The fingerprint fixture checks service behavior, not how long a waking display stays visible.
Exercise the changed countdown, check that input restarts it and an untouched screen blanks, and attach the required timing evidence before merging.
File Used: AGENTS.md (source)
Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!
|
Tested on a Dell XPS 9320 (built-in screen), on top of quattro:
I couldn't test the slow external monitor case, but 10s doesn't feel too long on a laptop either. |
|
Reviewed at Competing fix. #8561 (opened 2026-08-27) fixes the same re-blank the same way: it raises the idle blank interval for every arm, to 30 seconds instead of 10. When two fixes for one bug are similar, the older one is preferred unless the newer one shows that it misses a case, breaks something, or does not fix the cause. I found none of those in #8561. Its merge conflict comes from predating the Codex Medium and muse-wonder (via pi, thinking high) were asked to compare all five without being given my answer. Both preferred #8561, and both found no defect in this diff and nothing here that #8561 lacks. They can read this session's notes, so their agreement may not be independent. muse-wonder added a point about margin. The countdown still starts at the wake keypress ( Not run. Because this one loses to the older fix, I did not run it on a worker or label it Next: this waits on the maintainer to choose among #8561, this pull request and the other approaches above. Nothing is needed from you in the meantime. |
What changes
The lock screen powers the display off five seconds after the last input (
idleBlankTimer.idleMs). On a monitor that takes about that long to finish waking from DPMS, the picture appears and is switched off again immediately. Keeping the mouse moving or typing is what restarts the countdown and holds the panel up.This raises that delay to ten seconds. A wake still restarts the countdown, and an untouched lock screen still blanks. The longer grace after a suspend (
resumeMs, 15 seconds) is unchanged.Seen on an HP OMEN 27q over HDMI with an NVIDIA RTX 3070, Omarchy 4.5.0-3. One key or mouse move turns the panel on, the panel spends about five seconds resyncing, and DPMS off lands as the lock screen becomes visible.
Related reports: #7399, #7749, #14227. Other open pull requests approach the same blank differently (#17544, #17271, #8561, #8773).
Verification
bash test/shell.d/lock-fingerprint-service-test.shpassed. That fixture finds the blank timer by its interval, which is now 10000.The running session was not locked to time a wake, so the ten-second window is not measured on hardware here.
Screenshots
No layout change. The lock screen stays lit for 10 seconds after the last input instead of 5.