xe_sync_in_fence_get(9)
π xe_sync_in_fence_get(9) π
okay bestie, welcome to the kernel API manual for xe_sync_in_fence_get β the xe_sync_in_fence_get page, live from the Kernel API Manual, and yes we are absolutely getting into it π
β¨ NAME β¨
xe_sync_in_fence_get β get a fence from syncs, exec queue, and VM π¦
β¨ SYNOPSIS β¨
struct dma_fence * xe_sync_in_fence_get (struct xe_sync_entry *sync ,
int num_sync , struct xe_exec_queue *q , struct xe_vm *vm );
β¨ ARGUMENTS β¨
syncβ the input syncs, aka the whole vibe you're handing in πnum_syncβ how many syncs there are, bestie. count them. π’qβ the exec queue πvmβ the VM πΈ
β¨ DESCRIPTION β¨
okay so the whole point: this gets a fence from the syncs, the exec queue, and the VM. that's literally the job.
if the syncs contain in-fences, it creates and returns a composite fence made of all the in-fences plus the last fence. iconic behavior, no cap π«Ά
if there are no in-fences, it just returns the last fence on the input exec queue. simple. clean. slay.
β οΈ heads up bestie: the caller must drop the reference to the returned fence. not optional, not vibes-based, actually do it π
β¨ RETURN β¨
- a fence on success π
ERR_PTR(-ENOMEM)on failure π
β¨ SEE ALSO β¨
check the kernel file ./drivers/gpu/drm/xe/xe_sync.c for xe_sync_entry_wait(9), xe_sync_needs_wait(9), __xe_sync_ufence_get(9), xe_sync_ufence_get(9), xe_sync_ufence_put(9), xe_sync_ufence_get_status(9) π
and that's the page, bestie! this one's from ./drivers/gpu/drm/xe, dated October 2026 β go forth and get those fences, you've got main character energy βπ
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