perf/bpf: Call BPF handler directly, not through overflow machinery

To ultimately allow BPF programs attached to perf events to completely
suppress all of the effects of a perf event overflow (rather than just the
sample output, as they do today), call bpf_overflow_handler() from
__perf_event_overflow() directly rather than modifying struct perf_event's
overflow_handler. Return the BPF program's return value from
bpf_overflow_handler() so that __perf_event_overflow() knows how to
proceed. Remove the now unnecessary orig_overflow_handler from struct
perf_event.

This patch is solely a refactoring and results in no behavior change.

Suggested-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Kyle Huey <khuey@kylehuey.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Song Liu <song@kernel.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20240412015019.7060-5-khuey@kylehuey.com

1 files changed, 11 insertions, 16 deletions

diff --git a/kernel/events/core.c b/kernel/events/core.c
index d3f3f552e193..c6a6936183d5 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -9564,9 +9564,9 @@ static inline bool sample_is_allowed(struct perf_event *event, struct pt_regs *r
 }
 
 #ifdef CONFIG_BPF_SYSCALL
-static void bpf_overflow_handler(struct perf_event *event,
-				 struct perf_sample_data *data,
-				 struct pt_regs *regs)
+static int bpf_overflow_handler(struct perf_event *event,
+				struct perf_sample_data *data,
+				struct pt_regs *regs)
 {
 	struct bpf_perf_event_data_kern ctx = {
 		.data = data,
@@ -9587,10 +9587,8 @@ static void bpf_overflow_handler(struct perf_event *event,
 	rcu_read_unlock();
 out:
 	__this_cpu_dec(bpf_prog_active);
-	if (!ret)
-		return;
 
-	event->orig_overflow_handler(event, data, regs);
+	return ret;
 }
 
 static int perf_event_set_bpf_handler(struct perf_event *event,
@@ -9626,8 +9624,6 @@ static int perf_event_set_bpf_handler(struct perf_event *event,
 
 	event->prog = prog;
 	event->bpf_cookie = bpf_cookie;
-	event->orig_overflow_handler = READ_ONCE(event->overflow_handler);
-	WRITE_ONCE(event->overflow_handler, bpf_overflow_handler);
 	return 0;
 }
 
@@ -9638,15 +9634,15 @@ static void perf_event_free_bpf_handler(struct perf_event *event)
 	if (!prog)
 		return;
 
-	WRITE_ONCE(event->overflow_handler, event->orig_overflow_handler);
 	event->prog = NULL;
 	bpf_prog_put(prog);
 }
 #else
-static void bpf_overflow_handler(struct perf_event *event,
-				 struct perf_sample_data *data,
-				 struct pt_regs *regs)
+static int bpf_overflow_handler(struct perf_event *event,
+				struct perf_sample_data *data,
+				struct pt_regs *regs)
 {
+	return 1;
 }
 
 static int perf_event_set_bpf_handler(struct perf_event *event,
@@ -9730,7 +9726,8 @@ static int __perf_event_overflow(struct perf_event *event,
 		irq_work_queue(&event->pending_irq);
 	}
 
-	READ_ONCE(event->overflow_handler)(event, data, regs);
+	if (!(event->prog && !bpf_overflow_handler(event, data, regs)))
+		READ_ONCE(event->overflow_handler)(event, data, regs);
 
 	if (*perf_event_fasync(event) && event->pending_kill) {
 		event->pending_wakeup = 1;
@@ -11997,13 +11994,11 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu,
 		overflow_handler = parent_event->overflow_handler;
 		context = parent_event->overflow_handler_context;
 #if defined(CONFIG_BPF_SYSCALL) && defined(CONFIG_EVENT_TRACING)
-		if (overflow_handler == bpf_overflow_handler) {
+		if (parent_event->prog) {
 			struct bpf_prog *prog = parent_event->prog;
 
 			bpf_prog_inc(prog);
 			event->prog = prog;
-			event->orig_overflow_handler =
-				parent_event->orig_overflow_handler;
 		}
 #endif
 	}