RFC 016 Chunk 2a: per-actor timeslice overrun counter

Tally overruns at the slice-expiry site in preempt.rs (the RFC 006
signal), surfaced as ActorInfo.overruns.

- Counter is a hot-region AtomicU64 on Slot, single-writer Relaxed
  load+store (no atomic RMW), read Relaxed by the snapshot (D5).
- Reached from the rare expiry branch via a stashed *const Slot in a
  preempt thread-local, set/cleared on the same resume/return boundary
  as CURRENT_STOP — one TLS load, no runtime lookup. Shared infra for
  the messages-received counter next.
- Reset in all three slot-lifecycle sites: Slot::vacant, reclaim_slot,
  install_actor (standing invariant, D7) — per-incarnation counts.

src/introspect.rs

@@ -83,6 +83,10 @@ pub struct ActorInfo {
     /// install / spawn_addr / gen_server). 0 for an actor that holds only a
     /// private `channel()` receiver — those are invisible to the registry.
     pub mailbox_depth: u32,
+    /// Timeslice overruns tallied for this incarnation (RFC 016 Chunk 2): how
+    /// many times the actor was preempted for exceeding its slice. Resets on
+    /// restart (per-incarnation, D7).
+    pub overruns: u64,
 }
 
 /// A whole-runtime snapshot. See the module docs for the D2 tearing model.
@@ -159,6 +163,9 @@ fn read_slot(slot: &Slot, idx: u32, mail: Option<&MailboxInfo>) -> Option<ActorI
     let joiners = cold.waiters.len() as u32;
     drop(cold);
 
+    // Counters are hot-region atomics, read lock-free (RFC 016 Chunk 2).
+    let overruns = slot.overruns();
+
     // Names + depth belong to this incarnation only if the registry mailbox's
     // pid matches the slab generation; a stale registry entry (dead prior
     // occupant, not yet pruned) contributes nothing.
@@ -177,6 +184,7 @@ fn read_slot(slot: &Slot, idx: u32, mail: Option<&MailboxInfo>) -> Option<ActorI
         links,
         joiners,
         mailbox_depth,
+        overruns,
     })
 }
 

src/preempt.rs

@@ -58,6 +58,16 @@ thread_local! {
     /// resume path free of atomic ref-count traffic; see `check_cancelled` for
     /// the safety argument.
     static CURRENT_STOP: Cell<*const AtomicBool> = const { Cell::new(std::ptr::null()) };
+
+    /// Raw pointer to the on-CPU actor's slot, set/cleared by the scheduler on
+    /// the same resume/return boundary as `CURRENT_STOP` (RFC 016 Chunk 2).
+    /// Lets the rare slice-expiry site bump that actor's overrun counter with
+    /// one TLS load and no runtime lookup. Null while no actor is on-CPU. The
+    /// slot lives in the fixed slab and is never reclaimed while the actor is
+    /// running, so the pointer is valid for the whole resume (same lifetime
+    /// argument as `CURRENT_STOP`).
+    static CURRENT_SLOT: Cell<*const crate::runtime::Slot> =
+        const { Cell::new(std::ptr::null()) };
 }
 
 // ---------------------------------------------------------------------------
@@ -77,6 +87,33 @@ pub(crate) fn clear_current_stop() {
     CURRENT_STOP.with(|c| c.set(std::ptr::null()));
 }
 
+/// Bind the on-CPU actor's slot. Called by the scheduler immediately before
+/// `switch_to_actor`, beside `set_current_stop`.
+pub(crate) fn set_current_slot(slot: *const crate::runtime::Slot) {
+    CURRENT_SLOT.with(|c| c.set(slot));
+}
+
+/// Unbind the slot pointer on the return path, beside `clear_current_stop`.
+pub(crate) fn clear_current_slot() {
+    CURRENT_SLOT.with(|c| c.set(std::ptr::null()));
+}
+
+/// Tally a timeslice overrun against the on-CPU actor (RFC 016 Chunk 2). A
+/// no-op if no actor is bound (the scheduler's own stack). Reached only from
+/// the slice-expiry branch, which is already the yield path, so its cost is
+/// irrelevant.
+#[inline]
+fn note_overrun() {
+    let p = CURRENT_SLOT.with(|c| c.get());
+    // SAFETY: `p` is null (no actor on-CPU) or a pointer to the on-CPU actor's
+    // slot in the fixed slab. The slot is not reclaimed while the actor runs
+    // (finalize/reclaim happen only after it yields back), so the deref is
+    // valid for the whole resume — the same argument as `check_cancelled`.
+    if !p.is_null() {
+        unsafe { (*p).record_overrun() };
+    }
+}
+
 /// Observation point for cooperative cancellation. If the on-CPU actor has
 /// been flagged for stop, raise the sentinel panic so the trampoline's
 /// `catch_unwind` tears the stack down (running Drop) and reports
@@ -189,6 +226,10 @@ pub fn maybe_preempt() {
                 check_cancelled();
                 let start = TIMESLICE_START.with(|s| s.get());
                 if rdtsc().saturating_sub(start) > CONFIGURED_TIMESLICE_CYCLES.with(|t| t.get()) {
+                    // Tally the overrun (RFC 016 Chunk 2) before handing back —
+                    // this is the slice-expiry site RFC 006 wanted, and it's
+                    // already the yield path, so the counter is near-free.
+                    note_overrun();
                     // SAFETY: reachable only inside an actor (the scheduler
                     // sets PREEMPTION_ENABLED on resume and clears it on
                     // return). The scheduler stack is therefore valid.

src/runtime.rs

@@ -426,6 +426,14 @@ pub(crate) struct Slot {
     /// First-resume closure, double-boxed so it fits an `AtomicPtr`
     /// (`Box<Closure>` is a thin pointer). Swap-to-take; null when absent.
     closure: AtomicPtr<Closure>,
+    /// RFC 016 Chunk 2 — per-actor timeslice overrun tally. Single-writer: only
+    /// the on-CPU actor's scheduler thread increments it (at the slice-expiry
+    /// site in `preempt.rs`, reached via the stashed slot pointer), so the
+    /// writes are plain Relaxed load+store with no atomic-RMW traffic; the
+    /// snapshot reads it Relaxed from any thread. Lives in the hot region rather
+    /// than `SlotCold` so the increment needs no lock; reset across reuse like
+    /// every other slot field (`vacant` / `reclaim_slot` / `install_actor`).
+    overruns: AtomicU64,
     /// Cold lifecycle data. See [`SlotCold`].
     pub(crate) cold: RawMutex<SlotCold>,
 }
@@ -437,6 +445,7 @@ impl Slot {
             sp: AtomicUsize::new(0),
             stop_ptr: AtomicPtr::new(std::ptr::null_mut()),
             closure: AtomicPtr::new(std::ptr::null_mut()),
+            overruns: AtomicU64::new(0),
             cold: RawMutex::new(SlotCold {
                 actor: None,
                 waiters: Vec::new(),
@@ -465,6 +474,31 @@ impl Slot {
         self.word.load()
     }
 
+    /// Tally one timeslice overrun (RFC 016 Chunk 2). Single-writer: only the
+    /// on-CPU actor's own thread calls this, at the slice-expiry site, so a
+    /// Relaxed load+store is sufficient and avoids the cache-line lock of an
+    /// atomic RMW.
+    #[inline]
+    pub(crate) fn record_overrun(&self) {
+        let v = self.overruns.load(Ordering::Relaxed);
+        self.overruns.store(v.wrapping_add(1), Ordering::Relaxed);
+    }
+
+    /// Read the overrun tally (Relaxed; the snapshot reads cross-thread).
+    #[inline]
+    pub(crate) fn overruns(&self) -> u64 {
+        self.overruns.load(Ordering::Relaxed)
+    }
+
+    /// Zero the per-actor introspection counters. Called at every point a slot
+    /// is recycled or freshly occupied (`reclaim_slot`, `install_actor`) so a
+    /// reused slot never carries a previous incarnation's counts — the standing
+    /// slot-lifecycle reset invariant (RFC 016 D7).
+    #[inline]
+    pub(crate) fn reset_counters(&self) {
+        self.overruns.store(0, Ordering::Relaxed);
+    }
+
     /// A pid's-eye snapshot of the slot. Cold paths re-read this under the
     /// cold lock (generation can't change while it is held).
     #[inline]
@@ -1003,6 +1037,7 @@ pub(crate) fn install_actor(
     }
     slot.sp.store(sp, Ordering::Relaxed);
     slot.store_closure(closure);
+    slot.reset_counters();
     inner.live_actors.fetch_add(1, Ordering::Relaxed);
 
     // Publish: only now can pops, unparks, or stops find the actor. The
@@ -1043,6 +1078,7 @@ pub(crate) fn reclaim_slot(inner: &RuntimeInner, pid: Pid) {
         cold.waiters.clear();
         cold.monitors.clear();
         cold.links.clear();
+        slot.reset_counters();
         slot.stop_ptr.store(std::ptr::null_mut(), Ordering::Release);
         // The generation bump IS the reclaim: every stale pid is dead from
         // this store onwards (unpark protocol, pops, cold-path re-verifies).
@@ -1451,6 +1487,7 @@ fn schedule_loop(inner: &Arc<RuntimeInner>, slot_idx: usize) {
         set_actor_sp(sp);
         set_current_pid(pid);
         crate::preempt::set_current_stop(stop_flag);
+        crate::preempt::set_current_slot(slot as *const Slot);
         reset_actor_done();
         YIELD_INTENT.with(|c| c.set(YieldIntent::Yield));
         // RFC 005 timeslice inheritance: a slot-popped actor does NOT get a
@@ -1474,6 +1511,7 @@ fn schedule_loop(inner: &Arc<RuntimeInner>, slot_idx: usize) {
         stats.current_pid_index.store(u32::MAX, Ordering::Relaxed);
         clear_current_pid();
         crate::preempt::clear_current_stop();
+        crate::preempt::clear_current_slot();
 
         let intent = YIELD_INTENT.with(|c| c.get());
         slot.sp.store(get_actor_sp(), Ordering::Relaxed);