SymbX/systemd
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

core: drop cgroup v1 synthetic empty event logic

Browse Source 
cgroup v2's empty events are reliable, hence we'd not bother
with validating it again in unit_add_to_cgroup_empty_queue()
either.
This commit is contained in:
Mike Yuan
2025-04-11 23:50:50 +02:00
parent c9eff0bcfc
commit 55e4df21ef
8 changed files with 19 additions and 310 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/core/cgroup.c
+5 -143
View File
@@ -3726,108 +3726,6 @@ int unit_search_main_pid(Unit *u, PidRef *ret) {
return 0;
}
static int unit_watch_pids_in_path(Unit *u, const char *path) {
_cleanup_closedir_ DIR *d = NULL;
_cleanup_fclose_ FILE *f = NULL;
int ret = 0, r;
assert(u);
assert(path);
r = cg_enumerate_processes(SYSTEMD_CGROUP_CONTROLLER, path, &f);
if (r < 0)
RET_GATHER(ret, r);
else {
for (;;) {
_cleanup_(pidref_done) PidRef pid = PIDREF_NULL;
r = cg_read_pidref(f, &pid, /* flags = */ 0);
if (r == 0)
break;
if (r < 0) {
RET_GATHER(ret, r);
break;
}
RET_GATHER(ret, unit_watch_pidref(u, &pid, /* exclusive= */ false));
}
}
r = cg_enumerate_subgroups(SYSTEMD_CGROUP_CONTROLLER, path, &d);
if (r < 0)
RET_GATHER(ret, r);
else {
for (;;) {
_cleanup_free_ char *fn = NULL, *p = NULL;
r = cg_read_subgroup(d, &fn);
if (r == 0)
break;
if (r < 0) {
RET_GATHER(ret, r);
break;
}
p = path_join(empty_to_root(path), fn);
if (!p)
return -ENOMEM;
RET_GATHER(ret, unit_watch_pids_in_path(u, p));
}
}
return ret;
}
int unit_synthesize_cgroup_empty_event(Unit *u) {
int r;
assert(u);
/* Enqueue a synthetic cgroup empty event if this unit doesn't watch any PIDs anymore. This is compatibility
* support for non-unified systems where notifications aren't reliable, and hence need to take whatever we can
* get as notification source as soon as we stopped having any useful PIDs to watch for. */
CGroupRuntime *crt = unit_get_cgroup_runtime(u);
if (!crt || !crt->cgroup_path)
return -ENOENT;
r = cg_unified_controller(SYSTEMD_CGROUP_CONTROLLER);
if (r < 0)
return r;
if (r > 0) /* On unified we have reliable notifications, and don't need this */
return 0;
if (!set_isempty(u->pids))
return 0;
unit_add_to_cgroup_empty_queue(u);
return 0;
}
int unit_watch_all_pids(Unit *u) {
int r;
assert(u);
/* Adds all PIDs from our cgroup to the set of PIDs we
* watch. This is a fallback logic for cases where we do not
* get reliable cgroup empty notifications: we try to use
* SIGCHLD as replacement. */
CGroupRuntime *crt = unit_get_cgroup_runtime(u);
if (!crt || !crt->cgroup_path)
return -ENOENT;
r = cg_unified_controller(SYSTEMD_CGROUP_CONTROLLER);
if (r < 0)
return r;
if (r > 0) /* On unified we can use proper notifications */
return 0;
return unit_watch_pids_in_path(u, crt->cgroup_path);
}
static int on_cgroup_empty_event(sd_event_source *s, void *userdata) {
Manager *m = ASSERT_PTR(userdata);
Unit *u;
@@ -3864,36 +3762,19 @@ static int on_cgroup_empty_event(sd_event_source *s, void *userdata) {
return 0;
}
void unit_add_to_cgroup_empty_queue(Unit *u) {
static void unit_add_to_cgroup_empty_queue(Unit *u) {
int r;
assert(u);
/* Note that there are four different ways how cgroup empty events reach us:
*
* 1. On the unified hierarchy we get an inotify event on the cgroup
*
* 2. On the legacy hierarchy, when running in system mode, we get a datagram on the cgroup agent socket
*
* 3. On the legacy hierarchy, when running in user mode, we get a D-Bus signal on the system bus
*
* 4. On the legacy hierarchy, in service units we start watching all processes of the cgroup for SIGCHLD as
* soon as we get one SIGCHLD, to deal with unreliable cgroup notifications.
*
* Regardless which way we got the notification, we'll verify it here, and then add it to a separate
* queue. This queue will be dispatched at a lower priority than the SIGCHLD handler, so that we always use
* SIGCHLD if we can get it first, and only use the cgroup empty notifications if there's no SIGCHLD pending
* (which might happen if the cgroup doesn't contain processes that are our own child, which is typically the
* case for scope units). */
/* Note that cgroup empty events are dispatched in a separate queue with a lower priority than
* the SIGCHLD handler, so that we always use SIGCHLD if we can get it first, and only use
* the cgroup empty notifications if there's no SIGCHLD pending (which might happen if the cgroup
* doesn't contain processes that are our own child, which is typically the case for scope units). */
if (u->in_cgroup_empty_queue)
return;
/* Let's verify that the cgroup is really empty */
r = unit_cgroup_is_empty(u);
if (r <= 0)
return;
LIST_PREPEND(cgroup_empty_queue, u->manager->cgroup_empty_queue, u);
u->in_cgroup_empty_queue = true;
@@ -4412,25 +4293,6 @@ Unit* manager_get_unit_by_pidref(Manager *m, PidRef *pid) {
return NULL;
}
int manager_notify_cgroup_empty(Manager *m, const char *cgroup) {
Unit *u;
assert(m);
assert(cgroup);
/* Called on the legacy hierarchy whenever we get an explicit cgroup notification from the cgroup agent process
* or from the --system instance */
log_debug("Got cgroup empty notification for: %s", cgroup);
u = manager_get_unit_by_cgroup(m, cgroup);
if (!u)
return 0;
unit_add_to_cgroup_empty_queue(u);
return 1;
}
int unit_get_memory_available(Unit *u, uint64_t *ret) {
uint64_t available = UINT64_MAX, current = 0;
src/core/cgroup.h
-6
View File
@@ -452,7 +452,6 @@ void unit_add_to_cgroup_realize_queue(Unit *u);
int unit_cgroup_is_empty(Unit *u);
void unit_release_cgroup(Unit *u, bool drop_cgroup_runtime);
void unit_add_to_cgroup_empty_queue(Unit *u);
int unit_check_oomd_kill(Unit *u);
int unit_check_oom(Unit *u);
@@ -474,9 +473,6 @@ uint64_t unit_get_ancestor_memory_low(Unit *u);
uint64_t unit_get_ancestor_startup_memory_low(Unit *u);
int unit_search_main_pid(Unit *u, PidRef *ret);
int unit_watch_all_pids(Unit *u);
int unit_synthesize_cgroup_empty_event(Unit *u);
int unit_get_memory_available(Unit *u, uint64_t *ret);
int unit_get_memory_accounting(Unit *u, CGroupMemoryAccountingMetric metric, uint64_t *ret);
@@ -499,8 +495,6 @@ bool unit_has_host_root_cgroup(const Unit *u);
bool unit_has_startup_cgroup_constraints(Unit *u);
int manager_notify_cgroup_empty(Manager *m, const char *group);
void unit_invalidate_cgroup(Unit *u, CGroupMask m);
void unit_invalidate_cgroup_bpf(Unit *u);
src/core/dbus-unit.c
+1 -5
View File
@@ -2644,11 +2644,7 @@ static int bus_unit_track_handler(sd_bus_track *t, void *userdata) {
u->bus_track = sd_bus_track_unref(u->bus_track); /* make sure we aren't called again */
/* If the client that tracks us disappeared, then there's reason to believe that the cgroup is empty now too,
* let's see */
unit_add_to_cgroup_empty_queue(u);
/* Also add the unit to the GC queue, after all if the client left it might be time to GC this unit */
/* Add the unit to the GC queue, after all if the client left it might be time to GC this unit */
unit_add_to_gc_queue(u);
return 0;
src/core/manager.h
+2 -3
View File
@@ -696,8 +696,7 @@ enum {
EVENT_PRIORITY_TIME_CHANGE = SD_EVENT_PRIORITY_NORMAL-1,
EVENT_PRIORITY_TIME_ZONE = SD_EVENT_PRIORITY_NORMAL-1,
EVENT_PRIORITY_IPC = SD_EVENT_PRIORITY_NORMAL,
EVENT_PRIORITY_REWATCH_PIDS = SD_EVENT_PRIORITY_IDLE,
EVENT_PRIORITY_SERVICE_WATCHDOG = SD_EVENT_PRIORITY_IDLE+1,
EVENT_PRIORITY_RUN_QUEUE = SD_EVENT_PRIORITY_IDLE+2,
EVENT_PRIORITY_SERVICE_WATCHDOG = SD_EVENT_PRIORITY_IDLE,
EVENT_PRIORITY_RUN_QUEUE = SD_EVENT_PRIORITY_IDLE+1,
/* … to least important */
};
src/core/scope.c
+9 -37
View File
@@ -93,10 +93,8 @@ static void scope_set_state(Scope *s, ScopeState state) {
if (!IN_SET(state, SCOPE_STOP_SIGTERM, SCOPE_STOP_SIGKILL, SCOPE_START_CHOWN, SCOPE_RUNNING))
s->timer_event_source = sd_event_source_disable_unref(s->timer_event_source);
if (!IN_SET(old_state, SCOPE_DEAD, SCOPE_FAILED) && IN_SET(state, SCOPE_DEAD, SCOPE_FAILED)) {
if (!IN_SET(old_state, SCOPE_DEAD, SCOPE_FAILED) && IN_SET(state, SCOPE_DEAD, SCOPE_FAILED))
unit_unwatch_all_pids(UNIT(s));
unit_dequeue_rewatch_pids(UNIT(s));
}
if (state != old_state)
log_unit_debug(UNIT(s), "Changed %s -> %s",
@@ -237,17 +235,13 @@ static int scope_coldplug(Unit *u) {
if (r < 0)
return r;
if (!IN_SET(s->deserialized_state, SCOPE_DEAD, SCOPE_FAILED)) {
if (u->pids) {
PidRef *pid;
SET_FOREACH(pid, u->pids) {
r = unit_watch_pidref(u, pid, /* exclusive= */ false);
if (r < 0 && r != -EEXIST)
return r;
}
} else
(void) unit_enqueue_rewatch_pids(u);
if (!IN_SET(s->deserialized_state, SCOPE_DEAD, SCOPE_FAILED) && u->pids) {
PidRef *pid;
SET_FOREACH(pid, u->pids) {
r = unit_watch_pidref(u, pid, /* exclusive= */ false);
if (r < 0 && r != -EEXIST)
return r;
}
}
bus_scope_track_controller(s);
@@ -297,13 +291,6 @@ static void scope_enter_signal(Scope *s, ScopeState state, ScopeResult f) {
if (s->result == SCOPE_SUCCESS)
s->result = f;
/* Before sending any signal, make sure we track all members of this cgroup */
(void) unit_watch_all_pids(UNIT(s));
/* Also, enqueue a job that we recheck all our PIDs a bit later, given that it's likely some processes have
* died now */
(void) unit_enqueue_rewatch_pids(UNIT(s));
/* If we have a controller set let's ask the controller nicely to terminate the scope, instead of us going
* directly into SIGTERM berserk mode */
if (state == SCOPE_STOP_SIGTERM)
@@ -435,14 +422,9 @@ static int scope_enter_running(Scope *s) {
/* Set the maximum runtime timeout. */
scope_arm_timer(s, /* relative= */ false, scope_running_timeout(s));
/* On unified we use proper notifications hence we can unwatch the PIDs
* we just attached to the scope. This can also be done on legacy as
* we're going to update the list of the processes we watch with the
* PIDs currently in the scope anyway. */
/* Unwatch all pids we've just added to cgroup. We rely on empty notifications there. */
unit_unwatch_all_pids(u);
/* Start watching the PIDs currently in the scope (legacy hierarchy only) */
(void) unit_enqueue_rewatch_pids(u);
return 1;
fail:
@@ -635,12 +617,6 @@ static void scope_sigchld_event(Unit *u, pid_t pid, int code, int status) {
scope_enter_running(s);
return;
}
/* If we get a SIGCHLD event for one of the processes we were interested in, then we look for others to
* watch, under the assumption that we'll sooner or later get a SIGCHLD for them, as the original
* process we watched was probably the parent of them, and they are hence now our children. */
(void) unit_enqueue_rewatch_pids(u);
}
static int scope_dispatch_timer(sd_event_source *source, usec_t usec, void *userdata) {
@@ -699,10 +675,6 @@ int scope_abandon(Scope *s) {
scope_set_state(s, SCOPE_ABANDONED);
/* The client is no longer watching the remaining processes, so let's step in here, under the assumption that
* the remaining processes will be sooner or later reassigned to us as parent. */
(void) unit_enqueue_rewatch_pids(UNIT(s));
return 0;
}
src/core/service.c
+2 -26
View File
@@ -1310,10 +1310,8 @@ static void service_set_state(Service *s, ServiceState state) {
if (IN_SET(state,
SERVICE_DEAD, SERVICE_FAILED,
SERVICE_DEAD_BEFORE_AUTO_RESTART, SERVICE_FAILED_BEFORE_AUTO_RESTART, SERVICE_AUTO_RESTART, SERVICE_AUTO_RESTART_QUEUED,
SERVICE_DEAD_RESOURCES_PINNED)) {
SERVICE_DEAD_RESOURCES_PINNED))
unit_unwatch_all_pids(u);
unit_dequeue_rewatch_pids(u);
}
if (state != SERVICE_START)
s->exec_fd_event_source = sd_event_source_disable_unref(s->exec_fd_event_source);
@@ -1426,10 +1424,8 @@ static int service_coldplug(Unit *u) {
SERVICE_DEAD, SERVICE_FAILED,
SERVICE_DEAD_BEFORE_AUTO_RESTART, SERVICE_FAILED_BEFORE_AUTO_RESTART, SERVICE_AUTO_RESTART, SERVICE_AUTO_RESTART_QUEUED,
SERVICE_CLEANING,
SERVICE_DEAD_RESOURCES_PINNED)) {
(void) unit_enqueue_rewatch_pids(u);
SERVICE_DEAD_RESOURCES_PINNED))
(void) unit_setup_exec_runtime(u);
}
if (IN_SET(s->deserialized_state, SERVICE_START_POST, SERVICE_RUNNING, SERVICE_RELOAD, SERVICE_RELOAD_SIGNAL, SERVICE_RELOAD_NOTIFY, SERVICE_MOUNTING))
service_start_watchdog(s);
@@ -2189,7 +2185,6 @@ static void service_enter_stop_post(Service *s, ServiceResult f) {
s->result = f;
service_unwatch_control_pid(s);
(void) unit_enqueue_rewatch_pids(UNIT(s));
s->control_command = s->exec_command[SERVICE_EXEC_STOP_POST];
if (s->control_command) {
@@ -2244,13 +2239,6 @@ static void service_enter_signal(Service *s, ServiceState state, ServiceResult f
if (s->result == SERVICE_SUCCESS)
s->result = f;
/* Before sending any signal, make sure we track all members of this cgroup */
(void) unit_watch_all_pids(UNIT(s));
/* Also, enqueue a job that we recheck all our PIDs a bit later, given that it's likely some processes have
* died now */
(void) unit_enqueue_rewatch_pids(UNIT(s));
kill_operation = state_to_kill_operation(s, state);
r = unit_kill_context(UNIT(s), kill_operation);
if (r < 0) {
@@ -2290,8 +2278,6 @@ static void service_enter_stop_by_notify(Service *s) {
assert(s);
(void) unit_enqueue_rewatch_pids(UNIT(s));
r = service_arm_timer(s, /* relative= */ true, s->timeout_stop_usec);
if (r < 0) {
log_unit_warning_errno(UNIT(s), r, "Failed to install timer: %m");
@@ -2312,7 +2298,6 @@ static void service_enter_stop(Service *s, ServiceResult f) {
s->result = f;
service_unwatch_control_pid(s);
(void) unit_enqueue_rewatch_pids(UNIT(s));
s->control_command = s->exec_command[SERVICE_EXEC_STOP];
if (s->control_command) {
@@ -4257,15 +4242,6 @@ static void service_sigchld_event(Unit *u, pid_t pid, int code, int status) {
/* Notify clients about changed exit status */
if (notify_dbus)
unit_add_to_dbus_queue(u);
/* We watch the main/control process otherwise we can't retrieve the unit they
* belong to with cgroupv1. But if they are not our direct child, we won't get a
* SIGCHLD for them. Therefore we need to look for others to watch so we can
* detect when the cgroup becomes empty. Note that the control process is always
* our child so it's pointless to watch all other processes. */
if (!control_pid_good(s))
if (!s->main_pid_known || s->main_pid_alien || unit_cgroup_delegate(u))
(void) unit_enqueue_rewatch_pids(u);
}
static int service_dispatch_timer(sd_event_source *source, usec_t usec, void *userdata) {
src/core/unit.c
-83
View File
@@ -754,8 +754,6 @@ Unit* unit_free(Unit *u) {
unit_done(u);
unit_dequeue_rewatch_pids(u);
u->match_bus_slot = sd_bus_slot_unref(u->match_bus_slot);
u->bus_track = sd_bus_track_unref(u->bus_track);
u->deserialized_refs = strv_free(u->deserialized_refs);
@@ -2890,87 +2888,6 @@ void unit_unwatch_pidref_done(Unit *u, PidRef *pidref) {
pidref_done(pidref);
}
static void unit_tidy_watch_pids(Unit *u) {
PidRef *except1, *except2, *e;
assert(u);
/* Cleans dead PIDs from our list */
except1 = unit_main_pid(u);
except2 = unit_control_pid(u);
SET_FOREACH(e, u->pids) {
if (pidref_equal(except1, e) || pidref_equal(except2, e))
continue;
if (pidref_is_unwaited(e) <= 0)
unit_unwatch_pidref(u, e);
}
}
static int on_rewatch_pids_event(sd_event_source *s, void *userdata) {
Unit *u = ASSERT_PTR(userdata);
assert(s);
unit_tidy_watch_pids(u);
(void) unit_watch_all_pids(u);
/* If the PID set is empty now, then let's finish this off. */
unit_synthesize_cgroup_empty_event(u);
return 0;
}
int unit_enqueue_rewatch_pids(Unit *u) {
int r;
assert(u);
CGroupRuntime *crt = unit_get_cgroup_runtime(u);
if (!crt || !crt->cgroup_path)
return -ENOENT;
r = cg_unified_controller(SYSTEMD_CGROUP_CONTROLLER);
if (r < 0)
return r;
if (r > 0) /* On unified we can use proper notifications */
return 0;
/* Enqueues a low-priority job that will clean up dead PIDs from our list of PIDs to watch and subscribe to new
* PIDs that might have appeared. We do this in a delayed job because the work might be quite slow, as it
* involves issuing kill(pid, 0) on all processes we watch. */
if (!u->rewatch_pids_event_source) {
_cleanup_(sd_event_source_unrefp) sd_event_source *s = NULL;
r = sd_event_add_defer(u->manager->event, &s, on_rewatch_pids_event, u);
if (r < 0)
return log_error_errno(r, "Failed to allocate event source for tidying watched PIDs: %m");
r = sd_event_source_set_priority(s, EVENT_PRIORITY_REWATCH_PIDS);
if (r < 0)
return log_error_errno(r, "Failed to adjust priority of event source for tidying watched PIDs: %m");
(void) sd_event_source_set_description(s, "tidy-watch-pids");
u->rewatch_pids_event_source = TAKE_PTR(s);
}
r = sd_event_source_set_enabled(u->rewatch_pids_event_source, SD_EVENT_ONESHOT);
if (r < 0)
return log_error_errno(r, "Failed to enable event source for tidying watched PIDs: %m");
return 0;
}
void unit_dequeue_rewatch_pids(Unit *u) {
assert(u);
u->rewatch_pids_event_source = sd_event_source_disable_unref(u->rewatch_pids_event_source);
}
bool unit_job_is_applicable(Unit *u, JobType j) {
assert(u);
assert(j >= 0 && j < _JOB_TYPE_MAX);
src/core/unit.h
-7
View File
@@ -373,10 +373,6 @@ typedef struct Unit {
UnitFileState unit_file_state;
PresetAction unit_file_preset;
/* Low-priority event source which is used to remove watched PIDs that have gone away, and subscribe to any new
* ones which might have appeared. */
sd_event_source *rewatch_pids_event_source;
/* How to start OnSuccess=/OnFailure= units */
JobMode on_success_job_mode;
JobMode on_failure_job_mode;
@@ -880,9 +876,6 @@ void unit_unwatch_pidref(Unit *u, const PidRef *pid);
void unit_unwatch_all_pids(Unit *u);
void unit_unwatch_pidref_done(Unit *u, PidRef *pidref);
int unit_enqueue_rewatch_pids(Unit *u);
void unit_dequeue_rewatch_pids(Unit *u);
int unit_install_bus_match(Unit *u, sd_bus *bus, const char *name);
int unit_watch_bus_name(Unit *u, const char *name);
void unit_unwatch_bus_name(Unit *u, const char *name);