ktime_get_tai_fast_ns(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’– NAME πŸ’–

okay so bestie, welcome to ktime_get_tai_fast_ns β€” this page lives in the Kernel API Manual (section 9), and it's giving β€œNMI safe and fast access to tai clock” ✨ like that's literally the whole name-and-purpose combo.

It's ktime_get_tai_fast_ns β€” NMI safe and fast access to tai clock. periodt πŸ’…


✨ SYNOPSIS ✨

u64 notrace ktime_get_tai_fast_ns (void );

πŸŽ€ ARGUMENTS πŸŽ€

  • void β€” no arguments. literally none, bestie, it said what it said 🫢

πŸ¦‹ DESCRIPTION πŸ¦‹

so here's the tea: the same limitations as described for ktime_get_boot_fast_ns apply here too. you know how it is 😭

the mono time and the TAI offset are not read atomically, which may yield wrong readouts πŸ’€ not me finding out the vibes are slightly out of sync.

⚠️ caveat bestie: the mono time and the TAI offset are not read atomically, so the readouts may be wrong. do not pretend this is atomic β€” it is not. πŸ™ˆ

however β€” and this is the slay part β€” an update of the TAI offset is a rare event, e.g., caused by settime or adjtimex with an offset. so it's not happening every five seconds, thank goodness 🌸

the user of this function has to deal with the possibility of wrong timestamps in post processing. no cap, that's on you to handle in post processing 🫑


🌸 SEE ALSO 🌸

like, go read these too, iykyk πŸ’–


okay that's the page, bestie 🫢 source/version: ./kernel/time β€’ date: October 2026 β€’ page: ktime_get_tai_fast_ns(9) πŸŒˆπŸ’…

πŸ“¦ where this page comes from