runtime: wake_slot default ON (RFC 005 accepted, finding 18); supervisor stops survivors sequentially so reverse-order teardown is guaranteed; deflake gen_server/channel tests; baseline.json regenerated slot-on; ROADMAP notes

tests/gen_server.rs

@@ -396,8 +396,11 @@ impl GenServer for Pool {
 }
 
 // A worker spawned and watched from inside a handler delivers its Down to
-// handle_down. Down arms outrank the inbox, so the death is in the log by
-// the time the follow-up call is answered.
+// handle_down. Nothing orders the worker's death before the follow-up
+// call (the worker sits in the shared queue while call/reply wakes ride the
+// wake slot), so poll: the log must become exactly [Panic] within a bounded
+// number of yields. Down-outranks-inbox ordering is covered by
+// `watch_dead_pid_is_noproc_down`.
 #[test]
 fn worker_pool_down_reaches_handle_down() {
     let got = Arc::new(Mutex::new(Vec::new()));
@@ -409,7 +412,15 @@ fn worker_pool_down_reaches_handle_down() {
         });
         server.cast(PoolCast::SpawnDoomedWorker).unwrap();
         let _ = server.call(()).unwrap(); // sync point: cast handled, worker live
-        *got2.lock().unwrap() = server.call(()).unwrap();
+        let mut log = Vec::new();
+        for _ in 0..10_000 {
+            log = server.call(()).unwrap();
+            if !log.is_empty() {
+                break;
+            }
+            smarm::yield_now();
+        }
+        *got2.lock().unwrap() = log;
     });
     assert_eq!(*got.lock().unwrap(), vec![DownReason::Panic]);
 }