sdma_v4_0_enable(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… sdma_v4_0_enable(9) β€” Kernel API Manual Energy ✨

okay bestie, welcome to the Kernel API Manual page for sdma_v4_0_enable(9) β€” we’re folding that manual name right in here because we’re cute like that. This is the kernel-side vibe check for the async DMA engines, and the signature is giving β€œone bool to rule them all” energy. 🫢


🌸 NAME 🌸

sdma_v4_0_enable β€” stop the async dma engines. That’s the official one-liner, and ngl it’s got main-character energy despite being a one-line summary. πŸ’€

✨ SYNOPSIS ✨

void sdma_v4_0_enable (struct amdgpu_device *adev , bool enable );

This is literally the whole call, bestie: void, takes an amdgpu_device *adev, and a bool enable. No cap. πŸŽ€

πŸ¦‹ ARGUMENTS πŸ¦‹

  • adev β€” amdgpu_device pointer. This is your device pointer, the little handle to the AMDGPU device.
  • enable β€” enable/disable the DMA MEs. It’s a bool, and it’s here to say whether the DMA MEs get enabled or disabled.

πŸ’œ DESCRIPTION πŸ’œ

This function halts or unhalts the async dma engines β€” specifically on VEGA10. That’s it. That’s the whole description. It’s the on/off switch energy for those engines, and the (VEGA10) is important, so don’t come for me if you’re on some other chip. 😭

🌈 SEE ALSO 🌈

okay so the manpage points at the kernel file and a whole friend group of related sdma_v4_0_* functions. Gotta list every single one, because bestie we do not skip the boring parts. πŸ’…


cute sign-off time, bestie: this page comes from source/version ./drivers/gpu/drm/amd/amdgpu, dated October 2026, and the page is sdma_v4_0_enable(9). slay. πŸ’–

πŸ“¦ where this page comes from