Skip to content

Report a missing VAAPI driver when recording cannot initialize it - #7

Open
DanielJBurbridge wants to merge 3 commits into
quattrofrom
cursor/screenrecord-vaapi-detect-f2ea
Open

DanielJBurbridge wants to merge 3 commits into
quattrofrom
cursor/screenrecord-vaapi-detect-f2ea

Conversation

@DanielJBurbridge

@DanielJBurbridge DanielJBurbridge commented Nov 7, 2026 •

Copy link
Copy Markdown
Owner

When recording startup fails because VAAPI cannot load its driver, report the driver failure rather than leaving an unexplained recorder exit. Capture a bounded startup stderr tail, drain later output, and use a deadline that remains bounded with streaming/stalled stderr and non-period clock separators. Keep persistent logs behind debug mode.

Related: omacom#2706.

Validation on 8 November 2026:

  • bash test/shell.d/screenrecording-test.sh passed (isolated fixtures, graphical session disabled).
  • bash test/shell.d/i18n-test.sh passed (isolated fixtures, graphical session disabled).

Recorder/driver behaviour is mocked; no live GPU recording was performed. Translation-completeness check reports its structural-mode skip.

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. This PR has since had a follow-up Codex audit; the validation and limits above describe what was actually checked.

cursoragent and others added 3 commits November 7, 2026 13:29
gpu-screen-recorder logs vaInitialize failed and keeps going on the CPU.
That line was discarded, so a hybrid laptop with no Intel driver never
heard why encoding left the GPU.

Co-authored-by: Daniel Burbridge <DanielJBurbridge@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants