tracing: Use queue_rcu_work() to free filters

Freeing of filters requires to wait for both an RCU grace period as well as
a RCU task trace wait period after they have been detached from their
lists. The trace task period can be quite large so the freeing of the
filters was moved to use the call_rcu*() routines. The problem with that is
that the callback functions of call_rcu*() is done from a soft irq and can
cause latencies if the callback takes a bit of time.

The filters are freed per event in a system and the syscalls system
contains an event per system call, which can be over 700 events. Freeing 700
filters in a bottom half is undesirable.

Instead, move the freeing to use queue_rcu_work() which is done in task
context.

Link: https://lore.kernel.org/all/9a2f0cd0-1561-4206-8966-f93ccd25927f@paulmck-laptop/
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Link: https://lore.kernel.org/20250609131732.04fd303b@gandalf.local.home
Fixes: a9d0aab5eb33 ("tracing: Fix regression of filter waiting a long time on RCU synchronization")
Suggested-by: "Paul E. McKenney" <paulmck@kernel.org>
Reviewed-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>

diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c
index 3885aadc434d9320a2252f1bbff769fb35dbeb25..196c8bf349704a05a4a8acca4e7838fc067944fb 100644 (file)
--- a/kernel/trace/trace_events_filter.c
+++ b/kernel/trace/trace_events_filter.c
@@ -1344,13 +1344,14 @@ struct filter_list {
 
 struct filter_head {
        struct list_head        list;
-       struct rcu_head         rcu;
+       union {
+               struct rcu_head         rcu;
+               struct rcu_work         rwork;
+       };
 };
 
-
-static void free_filter_list(struct rcu_head *rhp)
+static void free_filter_list(struct filter_head *filter_list)
 {
-       struct filter_head *filter_list = container_of(rhp, struct filter_head, rcu);
        struct filter_list *filter_item, *tmp;
 
        list_for_each_entry_safe(filter_item, tmp, &filter_list->list, list) {
@@ -1361,9 +1362,20 @@ static void free_filter_list(struct rcu_head *rhp)
        kfree(filter_list);
 }
 
+static void free_filter_list_work(struct work_struct *work)
+{
+       struct filter_head *filter_list;
+
+       filter_list = container_of(to_rcu_work(work), struct filter_head, rwork);
+       free_filter_list(filter_list);
+}
+
 static void free_filter_list_tasks(struct rcu_head *rhp)
 {
-       call_rcu(rhp, free_filter_list);
+       struct filter_head *filter_list = container_of(rhp, struct filter_head, rcu);
+
+       INIT_RCU_WORK(&filter_list->rwork, free_filter_list_work);
+       queue_rcu_work(system_wq, &filter_list->rwork);
 }
 
 /*
@@ -1460,7 +1472,7 @@ static void filter_free_subsystem_filters(struct trace_subsystem_dir *dir,
        tracepoint_synchronize_unregister();
 
        if (head)
-               free_filter_list(&head->rcu);
+               free_filter_list(head);
 
        list_for_each_entry(file, &tr->events, list) {
                if (file->system != dir || !file->filter)
@@ -2305,7 +2317,7 @@ static int process_system_preds(struct trace_subsystem_dir *dir,
        return 0;
  fail:
        /* No call succeeded */
-       free_filter_list(&filter_list->rcu);
+       free_filter_list(filter_list);
        parse_error(pe, FILT_ERR_BAD_SUBSYS_FILTER, 0);
        return -EINVAL;
  fail_mem:
@@ -2315,7 +2327,7 @@ static int process_system_preds(struct trace_subsystem_dir *dir,
        if (!fail)
                delay_free_filter(filter_list);
        else
-               free_filter_list(&filter_list->rcu);
+               free_filter_list(filter_list);
 
        return -ENOMEM;
 }