GUC2PF_RELAY_FROM_VF(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸŽ€ GUC2PF_RELAY_FROM_VF πŸŽ€

okay bestie, settle in, because this one's from the Kernel API Manual and it's giving very specific firmware-gossip energy πŸ’… no cap.


✨ NAME ✨

GUC2PF_RELAY_FROM_VF β€” the lil' message the GuC firmware uses to pass a VF2PF Relay Message along πŸ’Œ


so here's the tea, and I mean literally the whole tea πŸ«–:

This message is used by the GuC firmware to forward a VF2PF Relay Message received from the Virtual Function (VF) driver to this Physical Function (PF) driver.

and like, it's not a sometimes thing β€” this message is always sent as a CTB HXG Message. always always. periodt ✨


🌈 the bit layout, aka the vibe check

  • Row 0 Β· bit 31 β€” ORIGIN = GUC_HXG_ORIGIN_GUC_
  • Row 0 Β· bits 30:28 β€” TYPE = GUC_HXG_TYPE_EVENT
  • Row 0 Β· bits 27:16 β€” MBZ
  • Row 0 Β· bits 15:0 β€” ACTION = XE_GUC_ACTION_GUC2PF_RELAY_FROM_VF = 0x5100

slay, that ACTION value is 0x5100, don't forget it, it's iconic to me πŸ”₯


πŸ¦‹ and then the payload, obviously

  • 1 Β· bits 31:0 β€” VFID β€” source VF identifier πŸ‘€
  • 2 Β· bits 31:0 β€” RELAY_ID β€” VF/PF message ID
  • 3 Β· bits 31:0 β€” RELAY_DATA1
  • ...
  • n Β· bits 31:0 β€” RELAY_DATAx

so yeah, it just keeps going with the data words until it stops, iykyk 🌸


πŸ‘€ SEE ALSO πŸ‘€

go read these next, bestie, they're the whole friend group 🫢


and that's the whole page, no crumbs left πŸ’€ mwah πŸ’‹

sign-off time: this lil' baddie lives at ./drivers/gpu/drm/xe/abi and it's dated October 2026, and the page you're reading is GUC2PF_RELAY_FROM_VF(9) from the Kernel API Manual, ok bye ily πŸŽ€πŸ’–

πŸ“¦ where this page comes from