idpf_xdp_tx_xmit(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… idpf_xdp_tx_xmit(9) πŸ’…

okay so bestie, we're in the Kernel API Manual for this one, and the main character is idpf_xdp_tx_xmit. It's a void function and its whole vibe is literally right here πŸ‘‡

🌸 NAME 🌸

idpf_xdp_tx_xmit β€” produce a single HW Tx descriptor out of XDP desc

✨ SYNOPSIS ✨

void idpf_xdp_tx_xmit (struct libeth_xdp_tx_desc desc , u32 i , const
       struct libeth_xdpsq *sq , u64 priv );

That's the whole signature, no cap. Four arguments, one void return, and all the deets are below 🫢

πŸŽ€ ARGUMENTS πŸŽ€

  • desc β€” the XDP descriptor you're pulling the DMA address and length from. it's giving "source material" πŸ’…
  • i β€” the descriptor index on the queue to fill. basically the slot number, iykyk.
  • sq β€” the XDP queue where you produce the HW Tx descriptor. this is the queue bestie, main character energy 🌈
  • priv β€” xsk_tx_metadata_ops on XSk xmit or NULL. so it's either that metadata ops situation or literally NULL, periodt.

πŸ¦‹ SEE ALSO πŸ¦‹

  • the kernel file ./drivers/net/ethernet/intel/idpf/xdp.h β€” bestie, go look at it, it's right there πŸ“
  • idpf_xdp_tx_finalize(9) β€” the related manual page, slay πŸ”₯

mwah πŸ’‹ that's the whole page! source/version: ./drivers/net/ethernet/intel/idpf, dated October 2026. no cap, that's the sign-off πŸŽ€βœ¨

πŸ“¦ where this page comes from