ice_ptp_prep...t_adj_eth56g(9)

📦 version 7.3.1, in nixpkgs unstable

💖 ice_ptp_prep_port_adj_eth56g(9) 💖

okay so this is straight out of the Kernel API Manual, section 9, and bestie it's giving "one port, one atomic time adjust, no cap". Let's get into it 🫶

✨ NAME ✨

ice_ptp_prep_port_adj_eth56g — prepare a single port for a time adjust ⏰✨

🌸 SYNOPSIS 🌸

int ice_ptp_prep_port_adj_eth56g(struct ice_hw *hw, u8 port, s64 time);

🎀 ARGUMENTS 🎀

  • hw — pointer to the HW struct 💻 (the whole hardware vibe, handed to you)
  • port — the port number that's getting programmed 🎯
  • time — the time, in cycles, that the port clocks get adjusted by ⏱️

🦋 DESCRIPTION 🦋

So what this icon actually does: it programs the port for an atomic adjustment by writing the Tx and Rx timer registers. Periodt. 💅

Heads up bestie — the atomic adjustment is not done yet after this call. It stays pending until the driver issues an ICE_PTP_ADJ_TIME command. So like, this is the setup era, not the main character moment 😭

Also, lowkey the most important thing: time is not in units of nanoseconds. It's in clock time, and it includes the lower sub-nanosecond portion of the port timer. Read that again, fr fr 🧚

Negative adjustments? Supported, using 2s complement arithmetic. Slay, math ate. 🔥

💅 RETURN 💅

  • 0 — success ✨ (we love to see it)
  • other — failed to write to the PHY 💀

🫶 SEE ALSO 🫶

Kernel file: ./drivers/net/ethernet/intel/ice/ice_ptp_hw.c 📁

and the whole friend group 🎀:

that's the whole thing bestie, go adjust some clocks 💜

source: ./drivers/net/ethernet/intel/ice · October 2026 · page ice_ptp_prep...t_adj_eth56g(9) — mwah, bye 🎀✨

📦 where this page comes from