xe_vma_user...e_invalidate(9)
π xe_vma_userptr_force_invalidate π
okay so bestie, welcome to the Kernel API Manual β yes this whole page lives in xe_vma_user...e_invalidate(9) land and it's lowkey a vibe. sit down, hydrate, and let's talk about forcing a userptr invalidation like the main character you are β¨
πΈ NAME πΈ
xe_vma_userptr_force_invalidate β force invalidate a userptr. that's it, that's the whole gig π
π SYNOPSIS π
void xe_vma_userptr_force_invalidate (struct xe_userptr_vma *uvma );
copy it exactly, don't be cute with the whitespace, iykyk π«Ά
π¦ ARGUMENTS π¦
uvmaβ the userptr vma you're about to invalidate. bestie is the target. handle with love π
β¨ DESCRIPTION β¨
so what does it actually do? it performs a forced userptr invalidation, and the reason is literally for testing purposes. no cap, that's the whole point β it's a testing hook, not your everyday production flow π
ngl, "forced" is doing a lot of work in that sentence. it's intentional, it's for tests, so don't go treating it like a casual little helper in normal code. periodt π
π SEE ALSO π
- the kernel file, right here:
./drivers/gpu/drm/xe/xe_userptr.cβ go read it, it's giving source of truth π xe_vma_userptr_check_repin(9)__xe_vm_userptr_needs_repin(9)xe_vm_userptr_check_repin(9)
like, the whole repin check squad π§ they all hang out together, go say hi π
and that's the page!! mwah π this little slice of the manual came from ./drivers/gpu/drm/xe, dated October 2026 β so when you're out here invalidating userptrs, remember who told you first ππ₯ β
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