diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index 9e90d1e7af2c..1de9a6bf8471 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -970,7 +970,6 @@ static DEFINE_PER_CPU(struct bpf_trace_sample_data, bpf_misc_sds);
 u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size,
 		     void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy)
 {
-	int nest_level = this_cpu_inc_return(bpf_event_output_nest_level);
 	struct perf_raw_frag frag = {
 		.copy		= ctx_copy,
 		.size		= ctx_size,
@@ -987,8 +986,12 @@ u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size,
 	};
 	struct perf_sample_data *sd;
 	struct pt_regs *regs;
+	int nest_level;
 	u64 ret;
 
+	preempt_disable();
+	nest_level = this_cpu_inc_return(bpf_event_output_nest_level);
+
 	if (WARN_ON_ONCE(nest_level > ARRAY_SIZE(bpf_misc_sds.sds))) {
 		ret = -EBUSY;
 		goto out;
@@ -1003,6 +1006,7 @@ u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size,
 	ret = __bpf_perf_event_output(regs, map, flags, sd);
 out:
 	this_cpu_dec(bpf_event_output_nest_level);
+	preempt_enable();
 	return ret;
 }