Repository navigation
Cancel lock fingerprint verification before suspend - #18030
Draft
DanielJBurbridge wants to merge 2 commits into
Draft
DanielJBurbridge wants to merge 2 commits into
DanielJBurbridge wants to merge 2 commits into
Conversation
The lock kept its pam_fprintd verify running across suspend, so fprintd held a stale device claim and every Claim after resume was denied. Abort that verify when logind prepares to sleep, refuse a new one until wake, then arm a single fresh verify. Refs omacom#15619 Co-authored-by: Daniel Burbridge <DanielJBurbridge@users.noreply.github.com>
Contributor
[Medium impact] The PR appears safe to merge; all four previous findings are addressed and no new blocking issue was found. SummaryThe PR releases fingerprint authentication before suspend and makes that release part of the work the delay inhibitor waits for.
All four previous findings are addressed. No new actionable issues were found. This review did not run the tests.
|
DanielJBurbridge
marked this pull request as draft
November 8, 2026 00:39
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Draft pending a Wayland service-fixture run and suspend/resume validation on an affected fingerprint reader.
No visible change: this adjusts fingerprint cancellation and retry scheduling around suspend; lock-screen rendering and layout are unchanged.
Cancel the lock's fingerprint verification inside the existing suspend delay inhibitor.
omarchy-system-sleep-lockcallslock prepareSleep, then polls the lock and fingerprint paused state instead of dropping the inhibitor as soon as the session is secure. Ordinarylock lockcalls keep their normal behaviour. An older shell falls back through the real IPC wrapper.Preserve a pending enrollment retry before stopping its timer, hold off new claims across unlock/relock during the suspend window, and queue another enrollment refresh if the old probe is still exiting at wake.
exec fprintd-listties the tracked process to the actual probe. Start the wake listener for the suspend window rather than keeping it idle all day. A stuck cancellation reaches the existing bounded deadline and logs a fingerprint warning without incorrectly reporting the session unlocked.fingerprintSleepReadymeans the shell is holding new claims, PAM is inactive and the tracked enrollment probe has stopped. PAM inactivity precedes deferred helper destruction: this does not acknowledge fprintd releasing the physical reader after its client disconnects. That remaining hardware behaviour is why this stays draft.Related: #15619.
Validation on 8 November 2026:
bash test/shell.d/sleep-lock-test.shpassed, including held/stuck cancellation and compatibility through the realomarchy-shellwrapper with a stubbed olderqs.bash test/shell.d/lock-fingerprint-retry-test.shpassed. Parser/transition tests now extend this existing suite; the new source-string test file was removed.No real fingerprint authentication, suspend cycle, reader release or GPU/desktop validation was performed. The QML fixture has no active native PAM helper, so even a passing fixture would not prove reader-release timing.
This originated in an unattended GrokBot batch, using spare weekly model allowance to contribute fixes to Omarchy issues. The batch needed more review before submission. Grok 4.7 proposed this follow-up; Codex reviewed and corrected its IPC compatibility and probe-process handling before pushing it. The checks and limits above describe what was actually verified.