dm_test_idl...ed_runs_body(9)

📦 version 7.3.1, in nixpkgs unstable

💅 dm_test_idle_worker_enabled_runs_body — like, the whole enable-body path, tested

okay so bestie, welcome to the Kernel API Manual 🎀 — yes, that's the manual name, and yes, we're doing this. today's main character is dm_test_idle_worker_enabled_runs_body(9), and she's here to test one very specific thing. let's get into it ✨

🌸 NAME 🌸

dm_test_idle_worker_enabled_runs_body — literally tests the enable-body path. that's it, that's the job. 💅

🦋 SYNOPSIS 🦋

void dm_test_idle_worker_enabled_runs_body (struct kunit *test );

✨ ARGUMENTS ✨

  • test — the KUnit test context 🫶 (she's the one holding the vibes, aka the test state)

🔥 DESCRIPTION 🔥

okay so picture this: we've got a headless device, right? and because it's headless, that makes the second branch false — which means the worker goes ahead and runs the enable-body, no cap. and what even is the enable-body? it's dc_post_update_surfaces_to_stream + re-enable, together, as a little duo. 👑

then there's an injected hwss.apply_idle_power_optimizations callback — this is our little stage-manager 🧚 — and what it does is let dc_allow_idle_optimizations clear idle_optimizations_allowed on the disable half. so when the next loop iteration comes around, it breaks at the first branch and the worker just... stops. 💀 iconic behaviour, honestly.

🌈 SEE ALSO 🌈

first up, the kernel file herself — she lives at:

./drivers/gpu/drm/amd/display/amdgpu_dm/tests/amdgpu_dm_crtc_test.c

and now the whole friend group, all of them, in order, fr fr:

💀 reminder bestie: none of these are optional reading, they're all in the original SEE ALSO and they're all staying. iykyk.

and that's the page!! 🌸🦋 sign-off time: this lil cutie ships out of ./drivers/gpu/drm/amd/display/... and she's dated October 2026 ⭐ — love that for her. mwah, go read the kernel source, ily, bye 🫶💖

📦 where this page comes from