idpf_tx_splitq_clean(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸŽ€ NAME πŸŽ€

okay so bestie, this is idpf_tx_splitq_clean(9) from the Kernel API Manual β€” and the name is literally idpf_tx_splitq_clean : Reclaim resources from buffer queue. Slay. πŸ’…


✨ SYNOPSIS ✨

void idpf_tx_splitq_clean (struct idpf_tx_queue *tx_q , u16 end , int
       napi_budget , struct libeth_sq_napi_stats *cleaned , bool descs_only );

🫢 ARGUMENTS 🫢

  • tx_q β€” the Tx queue to clean, bestie πŸ’–
  • end β€” queue index until which it should be cleaned; like where the cleaning stops ✨
  • napi_budget β€” used to determine if we are in netpoll πŸ‘€
  • cleaned β€” pointer to stats struct to track cleaned packets/bytes πŸ“ˆ
  • descs_only β€” true if queue is using flow-based scheduling and should not clean buffers at this time 🫢

🌸 DESCRIPTION 🌸

Okay so this function cleans the queue descriptor ring. If the queue is using queue-based scheduling, the buffers will be cleaned as well β€” periodt. If the queue is using flow-based scheduling, only the descriptors are cleaned at this time. No cap.

Separate packet completion events will be reported on the completion queue, and the buffers will be cleaned separately. Also lowkey important: the stats are not updated from this function when using flow-based scheduling. That's the vibe check β€” descriptors now, buffers later, stats stay quiet in that mode. πŸ’…


πŸ¦‹ SEE ALSO πŸ¦‹

okay so the cross-reference roll call is:


source/version: ./drivers/net/ethernet/intel/idpf Β· date: October 2026 β€” that's the whole page bestie, you ate πŸ’–πŸ«Άβœ¨

πŸ“¦ where this page comes from