scx_bpf_dsq_peek(9)

📦 version 7.3.1, in nixpkgs unstable

💅 scx_bpf_dsq_peek(9) — like, the Kernel API Manual page you didn't know you needed 🎀

okay so bestie, gather round, because today we're looking at scx_bpf_dsq_peek from the Kernel API Manual, and honestly? it's giving "sneaky little peek behind the curtain" and I'm obsessed. ✨

🌸 NAME 🌸

scx_bpf_dsq_peek — a lockless little peek at the first element. no locks, no drama, just a vibe check of the front of the queue 🦋

✨ SYNOPSIS ✨

__bpf_kfunc struct task_struct * scx_bpf_dsq_peek (u64 dsq_id , const
struct bpf_prog_aux *aux );

💖 ARGUMENTS 💖

  • dsq_id — the DSQ you wanna examine, bestie. that's the one we're peeking at 👀
  • aux — the implicit BPF argument that gives you access to bpf_prog_aux, and like… it's hidden from BPF progs. so sneaky. so mysterious. 🙈

💫 DESCRIPTION 💫

So what does it do? It reads the first element in the DSQ. Semantically it's the same thing as using the DSQ iterator, except — and this is the iconic part — it's lockfree. no cap. 🔥

But okay, real talk, like any lockless operation it only gives you a point-in-time snapshot. By the time any subsequent locking operation reads the queue, the contents may have changed. it's a photo, not a livestream, yk? 😭

It returns the pointer — or NULL, which means either an empty queue OR an internal error. yes, both. it's not telling you which. main character ambiguity 💀

👑 SEE ALSO 👑

and yeah, the whole thing lives in the kernel file ./kernel/sched/ext/ext.c, iykyk 🧚

ngl bestie, NULL doing double duty as "empty queue" and "internal error" is lowkey chaotic, so don't go assuming the queue's chill just because you got a NULL back 💀

and that's the page!! mwah 💋 this lil rendition came to you from ./kernel/sched/ext, dated October 2026 — slay, periodt, go peek at some queues responsibly 🫶🌈⭐

📦 where this page comes from