cik_sdma_enable(9)
β¨ cik_sdma_enable β¨
okay so this is a page from the Kernel API Manual, section 9, and bestieβ¦ it's a little one but she's important π
πΈ NAME πΈ
cik_sdma_enable β stop the async dma engines
like yes, the name says enable and the description says stop, don't come for her, that's literally what the page says π«Ά
π SYNOPSIS π
void cik_sdma_enable (struct radeon_device *rdev , bool enable );
returns void, so no return value to check, no drama, she just does the thing π
π¦ ARGUMENTS π¦
rdevβ aradeon_devicepointer, i.e. your whole radeon device situation πenableβ the enable/disable toggle for the DMA MEs, highkey just a bool doing bool things β¨
π DESCRIPTION π
Halt or unhalt the async dma engines (CIK).
that's the whole vibe: enable flips them, and the async DMA engines on CIK go from halted to unhalted (or the other way around, no cap) π₯
π SEE ALSO π
the kernel file ./drivers/gpu/drm/radeon/cik_sdma.c and honestly the entire friend group:
cik_sdma_get_rptr(9)cik_sdma_get_wptr(9)cik_sdma_set_wptr(9)cik_sdma_ring_ib_execute(9)cik_sdma_hdp_flush_ring_emit(9)cik_sdma_fence_ring_emit(9)cik_sdma_semaphore_ring_emit(9)cik_sdma_gfx_stop(9)cik_sdma_rlc_stop(9)cik_sdma_ctx_switch_enable(9)cik_sdma_gfx_resume(9)cik_sdma_rlc_resume(9)cik_sdma_load_microcode(9)cik_sdma_resume(9)cik_sdma_fini(9)cik_copy_dma(9)cik_sdma_ring_test(9)cik_sdma_ib_test(9)cik_sdma_is_lockup(9)cik_sdma_vm_copy_pages(9)cik_sdma_vm_write_pages(9)cik_sdma_vm_set_pages(9)cik_sdma_vm_pad_ib(9)
every single one, exactly as listed, go read them, iykyk π
ngl the whole page is short but she ate β halt/unhalt, one bool, no cap π
and that's the tea π this one lives in ./drivers/gpu/drm/radeon, dated October 2026, filed under cik_sdma_enable(9) in the Kernel API Manual β mwah, stan, periodt ππ§ββοΈ
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