Tested 7.0.0-28.28lp2160752 from the PPA on the reporting machine (omen, HP OMEN GT22, Ubuntu 26.04, RealSense D405 FW 5.17.3.10 — the firmware on which the deadlock is deterministic on the first control sequence).
Procedure: verified the in-tree uvcvideo from the test kernel was loaded (no local out-of-tree module present), then ran our XU/async control apply+verify sequence (AE off, exposure, gain, AWB off, WB — the exact trigger from the original report) 15 times against a D405, followed by sustained streaming. Result: 15/15 cycles completed with no hang. No uninterruptible (D-state) tasks, no hung-task warnings in the kernel log, and async control readback completes correctly (previously the first cycle wedged the USB host stack unkillably). Streaming clean afterwards. The cherry-pick fixes the issue on our hardware. Thanks Alice! Tested-by: Jay Chooi <[email protected]> -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/2160752 Title: uvcvideo: unkillable deadlock in status-event/PM path wedges USB stack (RealSense D405) To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2160752/+subscriptions -- ubuntu-bugs mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
