fbnic_phylin...plete_notify(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

okay so bestie, we're in the Kernel API Manual, section 9, and this page is literally about the PMD training complete notifier πŸŽ€ (the header up top truncates it all cute like fbnic_phylin...plete_notify(9) but we know her full name, iykyk πŸ’…)


✨ NAME ✨

fbnic_phylink_pmd_training_complete_notify β€” the PMD training complete notifier πŸ¦‹


🧚 SYNOPSIS 🧚

void fbnic_phylink_pmd_training_complete_notify (struct net_device *netdev );

Like, that's the whole vibe. A void, one argument, no returns. Iconic. ⭐


🌸 ARGUMENTS 🌸

  • netdev β€” the netdev struct for the phylink device that's attached 🫢

πŸ’– DESCRIPTION πŸ’–

so when the link first comes up, the PMD is gonna have a period of 2 to 3 seconds where the link just... flutters 😭 like it's nervous or something, all because of link training.

and bestie, we are NOT about to spam the kernel log with a million messages about it. that's giving main character energy of the worst kind πŸ’€ so instead we add a delay of 4 seconds from the time of the last PCS report of link. why? so we can guarantee we're unlikely to see any further link loss events due to link training. periodt. πŸ”₯

no cap, the whole point is the delay: 4 seconds from the last PCS report of link, so the 2 to 3 seconds of link fluttering during training doesn't flood your kernel log 🌈

πŸ¦‹ SEE ALSO πŸ¦‹


and that's the page, bestie! πŸŽ€ this one's living in ./drivers/net/ethernet/meta/fbnic, dated October 2026 β€” go off, you understood the assignment πŸ’œβœ¨πŸ“

πŸ“¦ where this page comes from