mem_cgroup_update_lru_size(9)

📦 version 7.3.1, in nixpkgs unstable

💅 mem_cgroup_update_lru_size(9) ✨

okay so bestie, welcome to the Kernel API Manual — yes we're in the kernel API manual today, hold my iced coffee 🧊💖 — and today's main character is mem_cgroup_update_lru_size. she's a little function with a big job and honestly? she ate. let's get into it 🎀

🌸 NAME 🌸

mem_cgroup_update_lru_size — account for adding or removing an lru page 💫

like, literally her whole thing is keeping the books for lru pages going in and out. no cap.

🦋 SYNOPSIS 🦋

void mem_cgroup_update_lru_size (struct lruvec *lruvec , enum lru_list
lru , int zid , long nr_pages );

that's the whole signature, bestie — copy it exactly, don't be out here freehanding it 🙈

✨ ARGUMENTS ✨

  • lruvec — the mem_cgroup per zone lru vector 💜 this is where the accounting lives.
  • lru — the index of the lru list the page is sitting on 🛋️ like which couch is the page chilling on.
  • zid — the zone id of the accounted pages 🗺️
  • nr_pages — positive when adding or negative when removing 🔥 yes she's signed, that's the whole vibe check. plus = coming in, minus = leaving.

👑 DESCRIPTION 👑

This function must be called under lru_lock — not optional, not "if u feel like it", this is a hard requirement bestie 💀

Timing is literally the entire plot: call it just before a page is added to an lru list, or just after a page is removed from an lru list. not during, not whenever, not "approximately" — before and after. periodt 🫶

⚠️ lowkey important: skip the lru_lock and you're not slaying, you're just wrong. the manpage says must, and must means must 😭

🌈 SEE ALSO 🌈

Kernel file ./mm/memcontrol.c — and honestly she has SO many mutuals, look at this friend group 💅 all of these are (9) pages, iykyk:

and that's the page bestie!! 🌸🦋 she's from the kernel source tree at ./mm, and this edition is dated October 2026 — mwah, go forth and lock your lru's 💖🎀👑

📦 where this page comes from