scx_bpf_locked_rq(9)

📦 version 7.3.1, in nixpkgs unstable

💅 scx_bpf_locked_rq(9) — a Kernel API Manual moment

okay bestie, so this is page scx_bpf_locked_rq(9) from the Kernel API Manual, and it's honestly a little diva: one function, one job, zero drama (mostly). let's get into it ✨

🎀 NAME 🎀

scx_bpf_locked_rq — returns the rq that SCX currently has locked. that's it. that's the whole assignment. 👑

🦋 SYNOPSIS 🦋

__bpf_kfunc struct rq * scx_bpf_locked_rq (const struct bpf_prog_aux
*aux );

look at her. beautiful. __bpf_kfunc, returns a struct rq *, takes exactly one arg. no cap, that's the whole signature 💖

🌸 ARGUMENTS 🌸

  • aux — this is the implicit BPF argument that lets you poke at bpf_prog_aux, and it's hidden from BPF progs. so like… it's there, it's required, but it's not for you, babe. it's giving backstage pass you're not supposed to use 🙈

✨ DESCRIPTION ✨

so here's the tea: if a rq lock is currently held by SCX, this bad boy returns the rq. slay. that's your pointer, go be free 🫶

but if no rq lock is being held by SCX? it emits an error and returns NULL. not me getting a NULL handed to me with an error attached 😭 so check that return value, bestie, fr fr. NULL is not a vibe, it's a warning sign.

lowkey important: no lock held by SCX = error + NULL. don't dereference that. periodt. 💀

💜 SEE ALSO 💜

it lives in the kernel file ./kernel/sched/ext/ext.c, and it rolls with a massive friend group. the whole sched_ext extended family, in original order, because I would never do them dirty:

like, that is a stacked friend group. iykyk 🌈

and that's the page!! mwah 💋 this one came from ./kernel/sched/ext, dated October 2026 — screenshot it, cite it, tell your friends, go read the other ninety pages, love u bye 🧚💜🌻

📦 where this page comes from