scx_rescue_try_admit(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… scx_rescue_try_admit(9) ✨

okay so welcome to the Kernel API Manual bestie, today we're doing scx_rescue_try_admit and honestly? it's giving gatekeeper-with-a-heart-of-gold 🫢


🌸 NAME 🌸

scx_rescue_try_admit β€” try to admit a freshly stranded task πŸ₯Ί


✨ SYNOPSIS ✨

bool scx_rescue_try_admit (struct rq *rq , struct task_struct *p );

like, that's the whole signature. no cap. πŸ“


πŸŽ€ ARGUMENTS πŸŽ€

  • rq β€” the rq that p is being inserted on. it's the venue, basically 🏟️
  • p β€” the stranded task being diverted to rescue. poor thing got left on read and now we're rescuing it πŸ’€πŸ«Ά

πŸ¦‹ DESCRIPTION πŸ¦‹

okay so here's the vibe: one rescue at a time, and earlier arrivals go first. line discipline, we love to see it πŸ‘‘

admission is not free, bestie β€” it needs a full quantum of budget, and that budget gets spent as the rescue runs. so like, it's a real commitment, not a situationship πŸ’Έ

return value check, because this matters fr fr:

  • returns true if p was admitted β€” in which case it should be inserted at the tail of rq's local DSQ βœ…
  • returns false if it has to park on the rescue DSQ instead, with the timer armed to admit it later ⏰

so yeah, false isn't a rejection rejection, it's just a "not right now, go wait in the lounge" πŸ™ˆ


🌈 SEE ALSO 🌈

the whole friend group, gathered for the group chat screenshot πŸ“Έ:

kernel file ./kernel/sched/ext/sub.c, plus scx_skip_subtree_pre(9), scx_next_descendant_pre(9), scx_rescue_charge(9), scx_rescue_end(9), scx_rescue_keep(9), scx_rescue_accrue(9), scx_rescue_admit(9), scx_rescue_check_overload(9), scx_rescue_timerfn(9), scx_resolve_local_dsq(9), queue_sync_ecaps(9), scx_process_sync_ecaps(9), scx_unbypass_replay_ecaps(9), scx_discard_stale_ecaps_syncs(9), scx_rehome_task(9), scx_punt_task(9), scx_cgroup_claim_subtree(9), scx_cgroup_return_subtree(9), scx_cgroup_task_migrating(9), scx_cgroup_task_migrated(9), scx_cgroup_task_migrate_canceled(9), scx_bpf_sub_dispatch(9), scx_bpf_sub_grant(9), scx_bpf_sub_revoke(9), scx_bpf_sub_caps(9), scx_bpf_sub_kill_bstr(9).


that's the whole page!! πŸ’– this came from ./kernel/sched/ext and it's dated October 2026 β€” so like, mark your calendars, screenshot the group chat, and slay the scheduler bestie πŸ§šβ­πŸ’œ

πŸ“¦ where this page comes from