sdma_v3_0_ctx_switch_enable(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… sdma_v3_0_ctx_switch_enable(9) πŸ’…

okay bestie, welcome to the Kernel API Manual β€” yes we're literally in the kernel docs rn, and this page is called sdma_v3_0_ctx_switch_enable(9). it's giving niche-but-important, iykyk πŸŽ€


✨ NAME ✨

sdma_v3_0_ctx_switch_enable β€” stop the async dma engines context switch

like, that's the whole vibe. that's what it does. no cap. πŸ¦‹


🌸 SYNOPSIS 🌸

void sdma_v3_0_ctx_switch_enable (struct amdgpu_device *adev , bool
enable );

she takes two things and returns nothing (void, so no receipts, she just does the thing and leaves πŸ’…).


πŸŽ€ ARGUMENTS πŸŽ€

  • adev β€” amdgpu_device pointer. that's your device, your main character, the whole reason we're here 🫢
  • enable β€” enable/disable the DMA MEs context switch. so like, bool energy: turn it on or turn it off, pick a lane ⭐

πŸ”₯ DESCRIPTION πŸ”₯

Halt or unhalt the async dma engines context switch (VI).

that's it, that's the function. you're either pausing that context switch or you're un-pausing it, and it's specifically the (VI) flavor of the async dma engines. no extra lore, no side quests β€” just halt/unhalt, iconic behavior honestly πŸ‘‘


🌈 SEE ALSO 🌈

first up, the source file she lives in β€” literally her address, her home base:

  • Kernel file ./drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c

and now the whole friend group, the entire sdma_v3_0 squad, all present and accounted for (not me listing all of them, but a promise is a promise πŸ’–):


and that's the page!! 🧚 sign-off time: this whole thing lives under ./drivers/gpu/drm/amd/amdgpu, dated October 2026, and yes the title is still sdma_v3_0_ctx_switch_enable(9) because we never forget where we came from πŸ’œ mwah, go halt something (or unhalt it, your call) πŸ“

πŸ“¦ where this page comes from