xe_sriov_pac...s_descriptor(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… NAME πŸ’…

ok so bestie, welcome to the Kernel API Manual, section 9, and today's main character is xe_sriov_packet_process_descriptor β€” she's the one who processes the migration data descriptor packet. that's it, that's her whole job, and she does it flawlessly, no cap ✨


πŸ¦‹ SYNOPSIS πŸ¦‹

int xe_sriov_packet_process_descriptor (struct xe_device *xe , unsigned int vfid , struct xe_sriov_packet *data );

like… look at that signature. iconic. the way she takes exactly three arguments and returns an int πŸ’…


🌸 ARGUMENTS 🌸

okay so here's who's who in the group chat:

  • xe β€” the xe_device herself. the queen bee πŸ‘‘ the whole device, that's her.
  • vfid β€” the VF identifier. literally just which VF we're talking to, iykyk.
  • data β€” the xe_sriov_packet that's carrying the descriptor. she's the messenger πŸŽ€

✨ DESCRIPTION ✨

so the descriptor uses the same KLV format as GuC β€” yes, same format, no deviations, we love consistency β€” and it contains metadata used for checking migration data compatibility. that's the whole vibe check: is this migration data compatible or not. the descriptor is literally the compatibility bouncer at the club 🧚


πŸŽ€ RETURN πŸŽ€

  • 0 on success πŸŽ‰ slay, we love a clean return
  • -errno on failure 😭 not me getting an errno back

fr fr that's the whole return contract.


🌈 SEE ALSO 🌈

the kernel file where all this lives:

  • ./drivers/gpu/drm/xe/xe_sriov_packet.c

and her besties, the whole packet friend group 🫢:

like, go read all of them, they're all giving main character energy, okurrr πŸ’œ


real talk though: this whole page lives under ./drivers/gpu/drm/xe, and the timestamp on it is October 2026 β€” so yeah, dated October 2026, filed in that directory, page name xe_sriov_pac...s_descriptor(9) (yes the name's truncated, she's mysterious like that πŸ™ˆ). mwah, bye bestie πŸ’–πŸ¦‹βœ¨
πŸ“¦ where this page comes from