Rate-limit scheduling of work-events

In general, events are code handling some some condition, which is
scheduled when such condition happened and executed independently from
I/O loop. Work-events are a subgroup of events that are scheduled
repeatedly until some (often significant) work is done (e.g. feeding
routes to protocol). All scheduled events are executed during each
I/O loop iteration.

Separate work-events from regular events to a separate queue and
rate limit their execution to a fixed number per I/O loop iteration.
That should prevent excess latency when many work-events are
scheduled at one time (e.g. simultaneous reload of many BGP sessions).

1 files changed, 3 insertions, 0 deletions

diff --git a/sysdep/unix/io.c b/sysdep/unix/io.c
index 9d54a2c3..3d67d0a7 100644
--- a/sysdep/unix/io.c
+++ b/sysdep/unix/io.c
@@ -2161,6 +2161,7 @@ io_init(void)
 {
   init_list(&sock_list);
   init_list(&global_event_list);
+  init_list(&global_work_list);
   krt_io_init();
   // XXX init_times();
   // XXX update_times();
@@ -2172,6 +2173,7 @@ io_init(void)
 
 static int short_loops = 0;
 #define SHORT_LOOP_MAX 10
+#define WORK_EVENTS_MAX 10
 
 void
 io_loop(void)
@@ -2189,6 +2191,7 @@ io_loop(void)
     {
       times_update(&main_timeloop);
       events = ev_run_list(&global_event_list);
+      events = ev_run_list_limited(&global_work_list, WORK_EVENTS_MAX) || events;
       timers_fire(&main_timeloop);
       io_close_event();