scx_bpf_get...smtmask_node(9)

💅 scx_bpf_get_idle_smtmask_node — from the Kernel API Manual, bestie

okay so picture this: it's giving kernel API, it's giving sched_ext, and the manual name is literally Kernel API Manual — so you already know we're in the cool part of the tree 🧚

---

✨ NAME ✨

scx_bpf_get_idle_smtmask_node — grab a referenced kptr to the idle-tracking, per-physical-core cpumask of a target NUMA node. Basically: it tells you whether a whole physical core is free. Obsessed 🫶

---

🌸 SYNOPSIS 🌸

__bpf_kfunc const struct cpumask * scx_bpf_get_idle_smtmask_node (s32 node , const struct bpf_prog_aux *aux );

---

🎀 ARGUMENTS 🎀

• node — the target NUMA node. that's the one, that's the node 👑
• aux — an implicit BPF argument that gives you access to bpf_prog_aux, and it's hidden from BPF progs. so like… it's there, you just don't see it 🙈

---

🦋 DESCRIPTION 🦋

So here's the vibe check: if idle tracking is not enabled, or if node is not valid, or if you're running on a UP kernel, this thing returns an empty cpumask. literally empty. no bits, no nothing 💀

and in that case, the actual error gets reported to the BPF scheduler via scx_error — so don't just sit there wondering, the scheduler gets told, periodt.

---

🌈 SEE ALSO 🌈

Kernel file ./kernel/sched/ext/idle.c, plus this whole friend group:

• scx_bpf_cpu_node(9)
• scx_bpf_select_cpu_dfl(9)
• __scx_bpf_select_cpu_and(9)
• scx_bpf_get_idle_cpumask_node(9)
• scx_bpf_get_idle_cpumask(9)
• scx_bpf_get_idle_smtmask(9)
• scx_bpf_put_idle_cpumask(9)
• scx_bpf_test_and_clear_cpu_idle(9)
• scx_bpf_pick_idle_cpu_node(9)
• scx_bpf_pick_idle_cpu(9)
• scx_bpf_pick_any_cpu_node(9)
• scx_bpf_pick_any_cpu(9)

---

⭐ and that's the whole page, bestie — brought to you from ./kernel/sched/ext, dated October 2026, straight out of the Kernel API Manual. iykyk, now go pick some idle CPUs 💜🎀