sdma_v4_4_2_gfx_resume(9)

📦 version 7.3.1, in nixpkgs unstable

💖 sdma_v4_4_2_gfx_resume(9) 💖

okay so bestie, welcome to the Kernel API Manual — this is the sdma_v4_4_2_gfx_resume(9) page and it's literally about waking up the async dma engines. yes that's the whole vibe. let's get into it 🫶

✨ NAME ✨

sdma_v4_4_2_gfx_resume — setup and start the async dma engines

like, wake the DMA engines up and get them going. that's the assignment. 💅

🎀 SYNOPSIS 🎀

void sdma_v4_4_2_gfx_resume (struct amdgpu_device *adev , unsigned int i , bool restore );

note the void bestie. she's a function. keep that in mind for later, iykyk 👀

🦋 ARGUMENTS 🦋

  • adev — an amdgpu_device pointer. the main character device itself, fr fr.
  • i — the instance to resume. you're picking which one gets to wake up.
  • restore — used to restore wptr when restart. so like, when you're restarting, this is the flag that says "put the write pointer back where it was." 🎯

🌸 DESCRIPTION 🌸

Set up the gfx DMA ring buffers and enable them. that's it, that's the whole thing — ring buffers get set up, then they get enabled, and the engines are alive. ✨

Return-wise: it says 0 for success, error for failure. (yes, the synopsis says void and the description says it returns 0 — that's what the page says, and i'm not gonna gaslight you about it, so both facts are staying right here. 💀)

👑 SEE ALSO 👑

the kernel file for all this lives at:

./drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c

and here's the whole friend group — every single one of them, no cap:

🌈 that's the page!! sourced from ./drivers/gpu/drm/amd/amdgpu, dated October 2026 — go off, kernel docs, you ate. mwah 💋✨

📦 where this page comes from