drm_dev_wedged_event(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

okay so bestie, welcome to drm_dev_wedged_event(9) from the Kernel API Manual πŸ’– this is literally the page where a DRM device gets its β€œwedged” uevent moment, and we’re gonna walk through it together, no cap ✨

πŸŽ€ NAME πŸŽ€

drm_dev_wedged_event - generate a device wedged uevent

✨ SYNOPSIS ✨

int drm_dev_wedged_event (struct drm_device *dev , unsigned long method
, struct drm_wedge_task_info *info );

πŸ¦‹ ARGUMENTS πŸ¦‹

  • dev β€” the DRM device, aka the main character here πŸ‘‘
  • method β€” the method(s) to be used for recovery, so like the recovery vibe check πŸ’…
  • info β€” optional information about the guilty task πŸ•΅οΈβ€β™€οΈ

πŸ”₯ DESCRIPTION πŸ”₯

So this function generates a device wedged uevent for the DRM device specified by dev β€” yes, it’s giving β€œsomething went sideways” energy 😭

The recovery method of choice gets sent in the uevent environment as WEDGED=<method1>[,..,<methodN>], and the order is from less to more side-effects, bestie πŸ’…

If the caller is unsure about recovery or the method is unknown (0), then WEDGED=unknown will be sent instead. No guessing 🫢

For more details, refer to the "Device Wedging" chapter in Documentation/gpu/drm-uapi.rst. It’s literally right there, iykyk πŸ“–βœ¨

🌸 RETURN 🌸

0 on success, negative error code otherwise. Slay when it works, sad little negative number when it doesn’t 😭

πŸ’œ SEE ALSO πŸ’œ


okay that’s the whole vibe πŸŽ€ this page’s source/version corner says ./drivers/gpu/drm, dated October 2026, and the title is drm_dev_wedged_event(9) β€” periodt πŸ’…πŸ’–

πŸ“¦ where this page comes from