xe_gt_sriov_...tore_produce(9)

📦 version 7.3.1, in nixpkgs unstable

💖 xe_gt_sriov_pf_migration_restore_produce(9) 💖

okay bestie, welcome to the Kernel API Manual (section 9, obviously, she's a kernel function not a command 💅) — today's main character is xe_gt_sriov_pf_migration_restore_produce</p>, and the filename vibes are giving xe_gt_sriov_...tore_produce(9) because it's long enough to need a nickname, iykyk 🎀

✨ NAME ✨

xe_gt_sriov_pf_migration_restore_produce — literally the girl who adds a VF restore data packet to the migration ring. That's it, that's the whole job 🫶

🦋 SYNOPSIS 🦋

int xe_gt_sriov_pf_migration_restore_produce (struct xe_gt *gt ,
       unsigned int vfid , struct xe_sriov_packet *data );

🌸 ARGUMENTS 🌸

  • gt — the xe_gt, no cap. your graphics tile babe ✨
  • vfid — the VF identifier, like her little name tag 🏷️
  • data — the xe_sriov_packet, the actual cargo being shipped 💌

🧚 DESCRIPTION 🧚

So the restore migration data producer (that's userspace, bestie 👀) calls this while it's processing migration data. Here's the tea:

  • if the ring is full, it waits until there's space. like, actually waits. no pushing, no shoving, just queued-up patience 🧘‍♀️
  • then it queues the restore migration data consumer — that's the PF SR-IOV Control worker — which is potentially sitting there waiting for data when the ring was empty 😭

so basically it's the friend who wakes everyone up when the snacks arrive. iconic behaviour, honestly 👑

🔥 RETURN 🔥

0 on success, or a negative error code on failure. simple as. don't @ me 💀

👑 SEE ALSO 👑

the kernel file lives right here, go be nosy:

./drivers/gpu/drm/xe/xe_gt_sriov_pf_migration.c

the whole friend group, in order 💜:

and that's the whole page, bestie 🫶 it ships from ./drivers/gpu/drm/xe, dated October 2026, and answers to xe_gt_sriov_...tore_produce(9). slay, goodbye, go read some kernel source 💖✨🎀

📦 where this page comes from