nfsd_stats_io_read_add(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

✨ nfsd_stats_io_read_add(9) ✨

okay so bestie, you're reading a page straight out of the Kernel API Manual πŸŽ€ and it's about nfsd_stats_io_read_add β€” yes the whole name, we're not abbreviating, she's iconic and she knows it πŸ’…


πŸ’– NAME πŸ’–

nfsd_stats_io_read_add β€” literally just counts the number of bytes for an NFS READ πŸ“


πŸ¦‹ SYNOPSIS πŸ¦‹

void nfsd_stats_io_read_add (struct nfsd_net *nn , struct svc_export
*exp , s64 amount );

that's the whole signature, no notes, she ate ✨


🌸 ARGUMENTS 🌸

  • nn β€” the target network namespace 🫢 like, which netns are we doing this for
  • exp β€” the target export, the one getting read from πŸ’œ
  • amount β€” the byte count, the actual number of bytes we're adding up, no cap

πŸ”₯ DESCRIPTION πŸ”₯

so these counters get updated on every single READ request β€” every one, bestie, no skipping πŸ™ˆ

the readers use percpu_counter_sum_positive, and that is the main character detail here: because of it, local batching does not affect read accuracy πŸ‘‘ so you can batch on the local side all you want and the numbers still come out right. slay.


🌈 SEE ALSO 🌈

go look at the kernel file ./fs/nfsd/stats.h πŸ’€ and also these besties right here:


ngl this one's short but she's still doing important work πŸ’…

that's the whole page, mwah πŸ’‹ β€” sourced from ./fs/nfsd, dated October 2026, and remember: nfsd_stats_io_read_add(9) is the moment, iykyk 🌟

πŸ“¦ where this page comes from