smarm-agent 2f011f6ce4 fix(rfc-004): gate spinning on max_spinners>0, killing wake-nobody futex_wake at N=1 ...

`spinning` was `spin_budget_cycles > 0` alone. At N=1 the pool-size cap is 0, so
spinning was enabled while no scheduler could ever enlist: n_spinning pinned at 0,
so the enqueue submit-rule guard (n_spinning==0) fired a wake-nobody futex_wake on
every push. Now gated on `budget > 0 && max_spinners > 0`; a debug_assert guards
the invariant against regression. No behavioural change (a 0-cap runtime never
parks on the futex either way) — single_thread_spinning_is_inert and the AllDone
liveness test (shutdown_releases_pinned_parked_siblings) still pass.

switch_cost N=1 (1-core sandbox): p50 ~466ns -> ~186ns. The wake-nobody syscall
was ~60% of the round-trip in wall-clock terms; the handoff's "~12%" was perf
self-time, not latency share. ROADMAP per-switch section updated to fold in the
N=1 spike correction (shim hypothesis is many-core only). Dropped the throwaway
budget_probe bench (switch_cost reproduces every state via .spin_budget_cycles/
.max_spinners and with percentiles).

2026-06-14 20:01:02 +00:00

..

actor.rs

feat(runtime): phase 2 — fixed slab, per-slot packed-atomic state machine, raw cold locks

2026-06-09 20:08:12 +00:00

channel.rs

feat(select): fd arms in select + timed fd waits (RFC 008 phase 1)

2026-06-11 11:28:58 +00:00

context.rs

docs(context): fix off-by-one-slot error in init_actor_stack layout comment

2026-05-28 05:50:18 +00:00

gen_server.rs

feat(gen_server): handle_down — runtime monitor forwarding via ServerCtx/Watcher

2026-06-10 14:59:44 +00:00

io.rs

fix(runtime): wake idle sibling schedulers at termination

2026-06-11 21:10:30 +00:00

lib.rs

feat(select): fd arms in select + timed fd waits (RFC 008 phase 1)

2026-06-11 11:28:58 +00:00

link.rs

feat(runtime): phase 2 — fixed slab, per-slot packed-atomic state machine, raw cold locks

2026-06-09 20:08:12 +00:00

monitor.rs

feat(channel): migrate Inner to RawMutex; two-class lock order (Leaf -> Channel)

2026-06-09 22:51:37 +00:00

mutex.rs

refactor(wakes): epoch-stamp every registration-based wake; retire per-primitive wait seqs

2026-06-10 07:15:29 +00:00

pid.rs

feat: full runtime redesign (v0.6)

2026-05-23 16:09:35 +00:00

preempt.rs

feat(runtime): phase 1 — peel timers/io/monitor-id out of SharedState; gate stop sentinel behind PREEMPTION_ENABLED

2026-06-09 18:56:17 +00:00

raw_mutex.rs

RFC 004 chunk 1: substrate — queue_len counter, n_spinning/park_seq/spinning state, spin_budget_cycles+max_spinners Config knobs, pub(crate) futex helpers. Behavioural no-op (feature off by default).

2026-06-14 06:47:16 +00:00

registry.rs

feat(registry): named pid registry as a bimap

2026-06-09 22:53:43 +00:00

run_queue.rs

feat(scheduler): RFC 005 wake slot — per-scheduler capacity-one wake cache

2026-06-11 20:20:07 +02:00

runtime.rs

fix(rfc-004): gate spinning on max_spinners>0, killing wake-nobody futex_wake at N=1

2026-06-14 20:01:02 +00:00

scheduler.rs

fix(scheduler): entry-side check_cancelled in park_current — stop against a QUEUED actor was lossy

2026-06-11 21:15:03 +00:00

slot_state.rs

feat(channel): select — ready-index wait over multiple receivers

2026-06-10 07:45:00 +00:00

stack.rs

feat: full runtime redesign (v0.6)

2026-05-23 16:09:35 +00:00

supervisor.rs

docs: mark links/trap_exit (#3) done; refresh supervisor + README

2026-06-07 21:51:56 +00:00

sync_shim.rs

feat(safety): phase 5 — loom model checking + invariant audit/assertion sweep

2026-06-09 21:27:45 +00:00

timer.rs

refactor(wakes): epoch-stamp every registration-based wake; retire per-primitive wait seqs

2026-06-10 07:15:29 +00:00

trace.rs

feat(scheduler): RFC 005 wake slot — per-scheduler capacity-one wake cache

2026-06-11 20:20:07 +02:00