cpu_pm_enter(9)

✨ cpu_pm_enter(9) — like, the CPU low-power entry notifier bestie ✨

okay so real talk, this one is from the Kernel API Manual and it's genuinely iconic in a "the whole system stays alive because of me" way 💅

---

🌸 NAME 🌸

cpu_pm_enter — the CPU low power entry notifier. yes that's her job title, and she's obsessed with it 🎀

---

🦋 SYNOPSIS 🦋

int cpu_pm_enter (void );

---

🍓 ARGUMENTS 🍓

• void — no arguments. zero. none. she takes nothing and still delivers, main character energy fr fr 💀

---

👑 DESCRIPTION 👑

So basically cpu_pm_enter notifies all the listeners that a single CPU is entering a low power state — and heads up, that state may cause some blocks in the same power domain as the cpu to reset. not a drill, just how it is 🫶

⚠️ You must call it on the affected CPU, and you must have interrupts disabled. not optional, not a suggestion, that's the whole vibe check.

The platform is responsible for making sure cpu_pm_enter is not called twice on the same CPU before cpu_pm_exit is called. like, one entry, one exit, take turns, be normal about it 😭

The drivers that get notified can include the VFP co-processor, the interrupt controller and its PM extensions, and local CPU timers context save/restore — all of which shouldn't be interrupted. that's the receipts for why interrupts have to be off. no cap. if you interrupt them it's giving chaos 🔥

And the return conditions are the same as __raw_notifier_call_chain. same rules, same energy, iykyk ✨

---

💖 SEE ALSO 💖

• Kernel file ./kernel/cpu_pm.c
• cpu_pm_register_notifier(9)
• cpu_pm_unregister_notifier(9)
• cpu_pm_exit(9) — her other half, the exit to this entry, go read it 🌈
• cpu_cluster_pm_enter(9)
• cpu_cluster_pm_exit(9)

those are the siblings, the whole friend group honestly 🧚

---

later bestie!! this page lives in ./kernel and it's stamped October 2026 — catch you on the next power state, mwah 💜