DRM_MODESET_LOCK_ALL_END(9)

💅 DRM_MODESET_LOCK_ALL_END(9) — like, the Kernel API Manual bestie ✨

okay so this is DRM_MODESET_LOCK_ALL_END from the Kernel API Manual, and honestly? she's the cleanup queen we all need. 🎀

---

🌸 NAME 🌸

DRM_MODESET_LOCK_ALL_END — the helper that releases and cleans up your modeset locks. she tidies up after the party. iconic behaviour, no cap. 🧹✨

---

🦋 SYNOPSIS 🦋

DRM_MODESET_LOCK_ALL_END (dev , ctx , ret );

---

💖 ARGUMENTS 💖

• dev — your drm device. the main character of the operation tbh. 🖥️
• ctx — the local modeset acquire context. heads up bestie: it will be dereferenced. 🙈
• ret — your local ret/err/etc variable, there to track the error status and keep you honest. 💀

---

✨ DESCRIPTION ✨

so this is literally the other side of DRM_MODESET_LOCK_ALL_BEGIN. they're a matching set, iykyk. 🫶

it will bounce back to BEGIN if ret is -EDEADLK. yes, it goes back. it's giving boomerang energy. 🪃

⚠️ ok important: you must use the same ret variable for begin and end. like, the same one. this is not a drill. it's so deadlock conditions get handled properly, and skipping it is lowkey a bad time. 😭

---

🔥 RETURN 🔥

ret will be untouched unless it's -EDEADLK on entry. so basically:

• if you successfully acquire the locks, ret stays whatever your code sets it to. slay. 👑
• if there's a deadlock or other failure with acquire or backoff, ret gets set to that failure. 💀

in both of those failure cases, the code between BEGIN/END will not be run — so the failure is just reflecting that you couldn't grab the locks. sad but true. 🌈

---

🌟 SEE ALSO 🌟

• Kernel file ./include/drm/drm_modeset_lock.h struct
• drm_modeset_acquire_ctx(9)
• struct drm_modeset_lock(9)
• drm_modeset_lock_fini(9)
• drm_modeset_is_locked(9)
• drm_modeset_lock_assert_held(9)
• DRM_MODESET_LOCK_ALL_BEGIN(9)

---

🦄 that's the whole page, bestie — brought to you from ./include/drm, October 2026 edition. go forth and release those locks responsibly. periodt. 💜🎀