dm_test_unre..._non_dp_noop(9)

๐Ÿ“ฆ version 7.3.1, in nixpkgs unstable

๐Ÿ’… dm_test_unregister_non_dp_noop โ€” Kernel API Manual ๐Ÿ’…

okay so bestie, welcome to the <pink>dm_test_unregister_non_dp_noop(9)</pink> page โ€” yes the title got squished by the terminal, it's fine, we're fine, everything's fine ๐Ÿ’–. It's a KUnit test and it's lowkey the chillest one in the whole file. Let's get into it.


โœจ NAME โœจ

dm_test_unregister_non_dp_noop โ€” Test unregister tolerates an unregistered connector ๐Ÿซถ

Like, the whole point is "what happens if we try to unregister something that was never registered?" and the answer is <green>nothing bad, babes</green>. Periodt.


๐ŸŒธ SYNOPSIS ๐ŸŒธ

void dm_test_unregister_non_dp_noop (struct kunit *test );

That's it. That's the whole signature. No return value, no drama, main character energy but make it void.


๐ŸŽ€ ARGUMENTS ๐ŸŽ€

  • test โ€” The KUnit test context ๐Ÿ’… (the little test-world the framework hands you so the test can actually exist, yk?)

๐Ÿฆ‹ DESCRIPTION ๐Ÿฆ‹

okay so picture this ๐ŸŽฌ โ€” no sysfs group was ever created, the CEC notifier is NULL, and the DP AUX channel was never registered. Literally none of the setup happened. Zero. Nada.

And despite all of that <purple>nothing was ever set up</purple>, the unregister call still has to be a safe no-op. Not a crash, not a splat, not a "whoops we dereferenced NULL" moment โ€” just a clean little nothing. ๐Ÿงš

The way that unregister just has to politely do nothing and not make it weird? Obsessed. That's the entire vibe check this test is running. โœจ


๐Ÿ’– SEE ALSO ๐Ÿ’–

๐Ÿ“ Kernel file

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

๐ŸŒˆ The rest of the squad (yes, all of them, I would never skip one)

๐Ÿซถ subconnector + content type crew

๐ŸŽจ adjust colour depth era

๐ŸŒธ output color space (the long one, stay with me)

๐Ÿงฎ convert color depth / color_depth_from_info

๐Ÿ”Œ to connector type

๐Ÿชž duplicate mode, encoder crtc masks, aspect ratio

โฑ๏ธ crtc timing decisions

๐Ÿ”„ funcs reset, atomic dup, hdr

โš™๏ธ set_property gang

๐Ÿช„ get_property gang

๐Ÿ”ฅ highest refresh + freesync video mode

๐Ÿ‘‘ cacp caps, part one

๐Ÿ–ฅ๏ธ set panel type, part one

๐Ÿ”— update subconnector (this page's cousin ๐Ÿซถ)

๐Ÿ—œ๏ธ fbc init + detect mst

๐Ÿ” find first crtc

๐Ÿ–ฅ๏ธ set panel type, part two

๐Ÿ‘‘ cacp caps, part two

๐ŸŒŠ fill stream (big mood)

๐Ÿงš create stream

๐Ÿ‘€ detect

๐Ÿ“ poll dac, late register, destroy, encoder disable

โœ… atomic check

๐Ÿ“บ hdmi cec + edid mgmt

๐Ÿงญ to encoder + native mode

๐Ÿงฉ common modes / ddc / freesync modes

๐Ÿงต i2c / vsdb / displayid

๐Ÿšฆ mode valid / cec / s3

๐ŸŒˆ create validate / update after detect / update scaling (final stretch bestie)


ngl if you read that whole SEE ALSO list you're literally unstoppable ๐Ÿ’€. Go drink some water, you earned it. ๐Ÿ“

and that's the page!! this little cutie lives in the source tree at <lavender>./drivers/gpu/drm/amd/display/...</lavender> and it's dated October 2026 ๐ŸŒธ โ€” that's the whole sign-off bestie, ily, go run your tests ๐Ÿซถโœจ๐Ÿ’–

๐Ÿ“ฆ where this page comes from