nfsd_stats_io_write_add(9)

📦 version 7.3.1, in nixpkgs unstable

✨ nfsd_stats_io_write_add ✨

Okay bestie, welcome to the nfsd_stats_io_write_add page from the Kernel API Manual — it's a little function with big "I keep the receipts" energy 💅

🌸 NAME 🌸

nfsd_stats_io_write_add — literally counts the number of bytes for an NFS WRITE 🎀

🦋 SYNOPSIS 🦋

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

That's the whole signature, no cap — returns void, takes three things, done ✨

💖 ARGUMENTS 💖

  • nn — the target network namespace, aka where the vibes (and the counters) live 🫶
  • exp — the target export, the specific thing being written to 💜
  • amount — the byte count, just how many bytes we're talking about 🍓

🔥 DESCRIPTION 🔥

So here's the tea: these counters get updated on every single WRITE request 🔥 Like, every one. No skipping, no "oh I'll do it later."

And the readers? They use percpu_counter_sum_positive, which means the local batching that happens under the hood does not mess with read accuracy 👑 Translation: the number you read is the real number, periodt. The batching is a whole internal thing and it stays in its lane 🌈

🌟 SEE ALSO 🌟

Kernel file ./fs/nfsd/stats.h plus the whole friend group:

Go read them too, iykyk 💀

Heads up bestie: this page lives under ./fs/nfsd in the Kernel API Manual, dated October 2026 ⭐ That's the source/version + date situation, so now you know exactly where it came from 🧚

Stay slay, count your bytes, and may your counters always sum positive 💖✨🎀🦋

📦 where this page comes from