Skip to content

Revisit Windows VHF health diagnostics after a supported driver test path exists #62

Open

Description

@SYMBaiEX

Status

Deferred from Cycle 8; no milestone assigned. Revisit only after OpenController has a supported Windows virtual-device driver test path with named ownership for the driver and test environment.

Blocked by #57 (provenance only): Cycle 8 gate #57 records this as deferred pending source and repository-fit review. This is not a technical implementation dependency.

Affected users and situation

Windows users streaming controller input through Sunshine/Moonlight may resume from sleep with the virtual gamepad and output-report path unavailable even while the driver broker service remains running. The report below concerns libvirtualhid's VHF/UMDF path, not an OpenController incident.

Evidence

  • libvirtualhid #140, opened 2026-09-11, is one report: after Windows sleep/resume, its diagnostic showed the gamepad and output reports unavailable while mouse support remained available and the broker was still running. The reporter said a full reboot restored the backend; restarting the broker and Sunshine did not.
  • The issue was closed the same day with a maintainer comment that it was fixed by libvirtualhid PR #137. This is counterevidence against treating the upstream defect as unresolved or recurring.

Confidence: Low; one external report, with an upstream fix reported. Potential OpenController fit: Medium, if a supported Windows VHF test path is established.

Repository fit and scope

OpenController currently generates Windows VHF driver and host-bridge source kits, report helpers, and setup guidance. It does not provide a maintained signed and installed virtual-device driver path for a supported suspend/resume test. Without an owned test path, this candidate is not currently buildable or verifiable as an OpenController change.

This is distinct from Cycle 7 #46: #46 concerns a native helper acknowledging its own startup before the first state write. This candidate concerns the OS virtual device becoming unavailable after Windows resume while its broker service remains running.

Acceptance criteria for reconsideration

  • Name the supported Windows virtual-device driver and who owns its build, signing, installation, and test environment.
  • Define a probe that distinguishes a running broker/service from an actively available gamepad and output-report device.
  • Document a post-resume validation procedure and its observed result on the supported path.
  • Keep any recovery behavior out of scope unless separately specified and validated; do not promise automatic recovery based on this report.

Activity

  1. added
    cycle-8Tracked work for OpenController AI and Gaming Improvement Cycle 8
    cycle-8-deferredConsidered but deferred from the active Cycle 8 build wave
    on Oct 5, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    cycle-8Tracked work for OpenController AI and Gaming Improvement Cycle 8cycle-8-deferredConsidered but deferred from the active Cycle 8 build wave

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions