amdgpu_dm_is...e_full_power(9)

💅 amdgpu_dm_ism_force_full_power — the Kernel API Manual tea ☕✨

okay so bestie, you're reading the Kernel API Manual page for amdgpu_dm_ism_force_full_power(9) and honestly? it's giving "small function, big responsibility." let's get into it 🫶

---

🌸 NAME 🌸

amdgpu_dm_ism_force_full_power — Force every CRTC's ISM FSM to FULL_POWER ✨

like. every single one. no CRTC left behind, periodt 💅

---

🎀 SYNOPSIS 🎀

void amdgpu_dm_ism_force_full_power (struct amdgpu_display_manager *dm
);

that's it. that's the whole signature. void return, one pointer, iconic behavior 👑

---

🦋 ARGUMENTS 🦋

• dm — the amdgpu display manager 🖥️✨ (yes that's literally the whole description, it's a self-explanatory king)

---

🔥 DESCRIPTION 🔥

so what does it actually do? it sends DM_ISM_EVENT_EXIT_IDLE_REQUESTED to every CRTC's ISM, which leaves each FSM sitting pretty in FULL_POWER_RUNNING ⭐ like, all of them. simultaneously. slay.

it's intended to be paired with amdgpu_dm_ism_disable 🫶 so the flow is:

1. callers should first quiesce workers (and this part is important — that's without dc_lock 🔒🙈)
2. then take dc_lock
3. and then call this helper

⚠️ bestie please read this one twice: this must be called with dc_lock held. not optional. not a suggestion. lock it up before you call it, no cap 💀🔒

---

👑 SEE ALSO 👑

• Kernel file ./drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_ism.c 📁
• dm_ism_next_state(9)
• dm_ism_get_idle_allow_delay(9)
• dm_ism_insert_record(9)
• amdgpu_dm_ism_disable(9)
• amdgpu_dm_ism_enable(9)

go read the whole friend group, they all make more sense together fr fr 🧚💜

---

🌈 and that's the page!! this came from ./drivers/gpu/drm/amd/display/..., dated October 2026, in the Kernel API Manual, filed under amdgpu_dm_is...e_full_power(9) — mwah, go forth and hold that dc_lock 💋🎀