Clean up multi-thread barrier-sync hold-down timer

Main thread: don't bother with the barrier sync hold-down timer if
none of the worker threads are busy.

Worker threads: avoid epoll_pwait (10ms timeout) when the
control-plane has been active in the last half-second.

Cherry-pick a recent dangling reference fix: pool_elt_at_index after
e.g. rx callback is required, in case the unix file pool expands.

DO NOT cherry-pick the patch into master!

Change-Id: I130dc72fd95c8a22716fa2f0f98ce35cdccc52d7
Signed-off-by: Dave Barach <dave@barachs.net>

1 files changed, 15 insertions, 0 deletions

diff --git a/src/vlib/threads.h b/src/vlib/threads.h
index 0ceb21e1218..5a295e03319 100644
--- a/src/vlib/threads.h
+++ b/src/vlib/threads.h
@@ -423,6 +423,21 @@ vlib_worker_thread_barrier_check (void)
 	}
       while (*vlib_worker_threads->wait_at_barrier)
 	;
+
+      /*
+       * Recompute the offset from thread-0 time.
+       * Note that vlib_time_now adds vm->time_offset, so
+       * clear it first. Save the resulting idea of "now", to
+       * see how well we're doing. See show_clock_command_fn(...)
+       */
+      {
+	f64 now;
+	vm->time_offset = 0.0;
+	now = vlib_time_now (vm);
+	vm->time_offset = vlib_global_main.time_last_barrier_release - now;
+	vm->time_last_barrier_release = vlib_time_now (vm);
+      }
+
       if (CLIB_DEBUG > 0)
 	vm->parked_at_barrier = 0;
       clib_atomic_fetch_add (vlib_worker_threads->workers_at_barrier, -1);