manager: Only invoke a single sigchld per unit within a cleanup cycle

By default, each iteration of manager_dispatch_sigchld() results in a unit level
sigchld event being invoked. For scope units, this results in a scope_sigchld_event()
which can seemingly stall for workloads that have a large number of PIDs within the
scope. The stall exhibits itself as a SIG_0 being initiated for each u->pids entry
as a result of pid_is_unwaited().

v2:
This patch resolves this condition by only paying to cost of a sigchld in the underlying
scope unit once per sigchld iteration. A new "sigchldgen" member resides within the
Unit struct. The Manager is incremented via the sd event loop, accessed via
sd_event_get_iteration, and the Unit member is set to the same value as the manager each
time that a sigchld event is invoked. If the Manager iteration value and Unit member
match, the sigchld event is not invoked for that iteration.

1 files changed, 1 insertions, 0 deletions

diff --git a/src/core/unit.c b/src/core/unit.c
index 0a1a5321df..8e5395361d 100644
--- a/src/core/unit.c
+++ b/src/core/unit.c
@@ -100,6 +100,7 @@ Unit *unit_new(Manager *m, size_t size) {
         u->on_failure_job_mode = JOB_REPLACE;
         u->cgroup_inotify_wd = -1;
         u->job_timeout = USEC_INFINITY;
+        u->sigchldgen = 0;
 
         RATELIMIT_INIT(u->start_limit, m->default_start_limit_interval, m->default_start_limit_burst);
         RATELIMIT_INIT(u->auto_stop_ratelimit, 10 * USEC_PER_SEC, 16);