dm_test_com...w_crtc_state(9)
π dm_test_commit_audio_skips_without_new_crtc_state(9) π
okay so this is a page from the Kernel API Manual, section 9, and ngl it's a whole mood β it's a KUnit test for the amdgpu display audio commit path. let's get into it bestie π«Ά
π NAME π
dm_test_commit_audio_skips_without_new_crtc_state β literally just "Test missing CRTC state". that's the whole vibe. it's a test, the thing it tests is a missing CRTC state, periodt. β¨
πΈ SYNOPSIS πΈ
void dm_test_commit_audio_skips_without_new_crtc_state (struct kunit *test );
she takes one argument, she returns void, she does exactly one job π
π¦ ARGUMENTS π¦
testβ the KUnit test context. that's it. that's the whole parameter list. you hand her the test context and she handles the rest π₯
β¨ DESCRIPTION β¨
so here's the situation, bestie. picture a connector that's still attached to the same CRTC β nothing changed, same pairing, we're comfy ποΈ
the rule is: that connector should NOT notify if the atomic state doesn't contain a new CRTC state for that CRTC.
like... not me checking the atomic state and finding no fresh CRTC state for the CRTC in question, and then the connector loudly going off anyway? π no ma'am. if there's no new CRTC state, the notify does not happen. that's the entire point of this test β it proves the skip actually skips.
the way that "no new state for this CRTC" translates to "no notification" β iconic, correct, exactly what we want π
π SEE ALSO π
first up, the actual kernel file this thing lives in β go read it, it's giving main character energy:
./drivers/gpu/drm/amd/display/amdgpu_dm/tests/amdgpu_dm_audio_test.c
and the rest of the squad, all in section 9, all absolute icons in their own right:
dm_test_audio_init_disabled(9)dm_test_audio_init_enabled_success(9)dm_test_audio_fini_without_enabled_audio(9)dm_test_fill_audio_info_ids_name_flags(9)dm_test_fill_audio_info_cea_lt_3_skips_modes(9)dm_test_fill_audio_info_cea_ge_3_copies_modes(9)dm_test_fill_audio_info_latency_present(9)dm_test_fill_audio_info_latency_absent_keeps_zero(9)dm_test_fill_audio_info_cea_ge_3_zero_modes(9)dm_test_audio_component_bind_sets_fields(9)dm_test_audio_component_unbind_clears_fields(9)dm_test_eld_notify_invokes_callback(9)dm_test_eld_notify_no_component(9)dm_test_eld_notify_null_audio_ops(9)dm_test_eld_notify_null_callback(9)dm_test_audio_init_pins_sets_defaults(9)dm_test_audio_init_pins_zero_count(9)dm_test_audio_component_get_eld_copies_matching_connector(9)dm_test_audio_component_get_eld_no_match(9)dm_test_commit_audio_notifies_removed_connector(9)dm_test_commit_audio_notifies_added_connector(9)dm_test_commit_audio_skips_writeback_removal(9)dm_test_commit_audio_skips_without_stream_status(9)dm_test_commit_audio_skips_detached_connector(9)dm_test_commit_audio_skips_without_modeset(9)dm_test_commit_audio_skips_addition_without_stream(9)dm_test_commit_audio_skips_writeback_addition(9)
twenty-seven of them. every single one accounted for. no cap. π§
ok that's the whole page!! head back to your group chat and tell them you now understand CRTC state notification skipping π
that's a wrap, bestie π this little page came from ./drivers/gpu/drm/amd/display/... and it's dated October 2026 β so yeah, she's from the future, she's ahead of her time, and she ate. slay β¨ππ
π¦ where this page comes from
- nixpkgs unstable: linux-manual 7.3.1
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms