pvr_vm_gpuva_remap(9)

📦 version 7.3.1, in nixpkgs unstable

💅 pvr_vm_gpuva_remap(9) — like, the whole glow-up

okay so bestie, welcome to the Kernel API Manual entry for pvr_vm_gpuva_remap. yes that's the whole title, it's in section 9, and it's giving "niche kernel lore that only your driver-dev mutuals know about" ✨

🌸 NAME 🌸

pvr_vm_gpuva_remap — remap a mapping within a memory context. that's it, that's the vibe 🦋

✨ SYNOPSIS ✨

int pvr_vm_gpuva_remap (struct drm_gpuva_op *op , void *op_ctx );

literally the whole signature, no cap. returns an int, takes two args, slays 💅

🎀 ARGUMENTS 🎀

  • op — the gpuva op that's carrying all the remap details. it's the main character of this call, it knows what's being remapped 🔥
  • op_ctx — the operation context. the supporting cast, but like, essential supporting cast 🫶

🧚 CONTEXT 🧚

okay so here's when this gets called: either drm_gpuvm_sm_map or drm_gpuvm_sm_unmap calls it, and it happens when a mapping or unmapping operation causes a region to be split. so you're in the middle of somebody else's business, lowkey 😭

important detail: the op_ctx.vm_ctx mutex is held. so like, don't go trying to grab it again, bestie, that's not the moment 🙈

⭐ RETURN ⭐

  • 0 on success, or
  • any error returned by pvr_vm_gpuva_unmap or pvr_vm_gpuva_unmap.
ngl the original page literally lists pvr_vm_gpuva_unmap twice. iconic of it, honestly. we're keeping it exactly as written — not our bug to fix, we just report it 💀

👑 SEE ALSO 👑

the extended universe, fr fr. all of these are connected and you should probably know them:

and that's the page!! 🎉 sourced straight from ./drivers/gpu/drm/imagination, dated October 2026 — mwah, go read the other nine-section entries, ily, bye 💖✨🎀🦋

📦 where this page comes from