dm_test_ism...e_and_cancel(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… dm_test_ism_dispatch_hysteresis_schedule_and_cancel β€” a Kernel API Manual moment πŸŽ€

okay so bestie, welcome to the Kernel API Manual ✨ β€” today's main character is dm_test_ism_dispatch_hysteresis_schedule_and_cancel(9), and she is literally here to cover the HYSTERESIS_WAITING dispatch arms. no cap, that's the whole vibe. πŸ¦‹


✨ NAME ✨

dm_test_ism_dispatch_hysteresis_schedule_and_cancel β€” cover the HYSTERESIS_WAITING dispatch arms πŸ’–


🌸 SYNOPSIS 🌸

void dm_test_ism_dispatch_hysteresis_schedule_and_cancel (struct kunit
test );

πŸŽ€ ARGUMENTS πŸŽ€

  • test β€” the KUnit test context, aka the little group chat this test is living in 🫢

πŸ”₯ DESCRIPTION πŸ”₯

so here's the lore, bestie. ENTER_IDLE_REQUESTED moves FULL_POWER_RUNNING -> HYSTERESIS_WAITING. slay, transition ate. πŸ’…

then the current-state arm of dm_ism_dispatch_power_state does its thing: it records the idle timestamp, computes a (non-zero!) idle-allow delay, and schedules the delayed worker. iconic behaviour, honestly. ⭐

after that, a subsequent BEGIN_CURSOR_UPDATE (-> HYSTERESIS_BUSY) exercises the previous-state arm, which cancels that pending worker. so we're out here testing the schedule and the cancel β€” hence the name, iykyk. 😭

and the important bit: neither arm reaches the DC-dependent idle-optimization commit path. so like, don't go looking for it, it's not invited to this party πŸ™ˆ


πŸ‘‘ SEE ALSO πŸ‘‘

🧚 Kernel file 🧚

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

🌈 the whole friend group 🌈


πŸ’€ reminder bestie: these are cross-references, not bonus levels. read them if you wanna know the neighbours, but this page is this page. periodt. πŸ“

and that's a wrap on our little manpage date πŸ’œ β€” brought to you by source ./drivers/gpu/drm/amd/display/..., dated October 2026, page dm_test_ism...e_and_cancel(9). go forth and test those arms, queen πŸ‘‘βœ¨

πŸ“¦ where this page comes from