perf record: collect BPF metadata from new programs

This collects metadata for any BPF programs that were loaded during a
"perf record" run, and emits it at the end of the run.

Signed-off-by: Blake Jones <blakejones@google.com>
Link: https://lore.kernel.org/r/20250612194939.162730-4-blakejones@google.com
Signed-off-by: Namhyung Kim <namhyung@kernel.org>

tools/perf/builtin-record.c

@@ -2162,6 +2162,14 @@ static int record__synthesize(struct record *rec, bool tail)
 	return err;
 }
 
+static void record__synthesize_final_bpf_metadata(struct record *rec __maybe_unused)
+{
+#ifdef HAVE_LIBBPF_SUPPORT
+	perf_event__synthesize_final_bpf_metadata(rec->session,
+						  process_synthesized_event);
+#endif
+}
+
 static int record__process_signal_event(union perf_event *event __maybe_unused, void *data)
 {
 	struct record *rec = data;
@@ -2807,6 +2815,8 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 	trigger_off(&auxtrace_snapshot_trigger);
 	trigger_off(&switch_output_trigger);
 
+	record__synthesize_final_bpf_metadata(rec);
+
 	if (opts->auxtrace_snapshot_on_exit)
 		record__auxtrace_snapshot_exit(rec);
 

tools/perf/util/bpf-event.c

@@ -472,6 +472,49 @@ void bpf_metadata_free(struct bpf_metadata *metadata __maybe_unused)
 
 #endif /* HAVE_LIBBPF_STRINGS_SUPPORT */
 
+struct bpf_metadata_final_ctx {
+	const struct perf_tool *tool;
+	perf_event__handler_t process;
+	struct machine *machine;
+};
+
+static void synthesize_final_bpf_metadata_cb(struct bpf_prog_info_node *node,
+					     void *data)
+{
+	struct bpf_metadata_final_ctx *ctx = (struct bpf_metadata_final_ctx *)data;
+	struct bpf_metadata *metadata = node->metadata;
+	int err;
+
+	if (metadata == NULL)
+		return;
+	err = synthesize_perf_record_bpf_metadata(metadata, ctx->tool,
+						  ctx->process, ctx->machine);
+	if (err != 0) {
+		const char *prog_name = metadata->prog_names[0];
+
+		if (prog_name != NULL)
+			pr_warning("Couldn't synthesize final BPF metadata for %s.\n", prog_name);
+		else
+			pr_warning("Couldn't synthesize final BPF metadata.\n");
+	}
+	bpf_metadata_free(metadata);
+	node->metadata = NULL;
+}
+
+void perf_event__synthesize_final_bpf_metadata(struct perf_session *session,
+					       perf_event__handler_t process)
+{
+	struct perf_env *env = &session->header.env;
+	struct bpf_metadata_final_ctx ctx = {
+		.tool = session->tool,
+		.process = process,
+		.machine = &session->machines.host,
+	};
+
+	perf_env__iterate_bpf_prog_info(env, synthesize_final_bpf_metadata_cb,
+					&ctx);
+}
+
 /*
  * Synthesize PERF_RECORD_KSYMBOL and PERF_RECORD_BPF_EVENT for one bpf
  * program. One PERF_RECORD_BPF_EVENT is generated for the program. And
@@ -612,6 +655,7 @@ static int perf_event__synthesize_one_bpf_prog(struct perf_session *session,
 		}
 
 		info_node->info_linear = info_linear;
+		info_node->metadata = NULL;
 		if (!perf_env__insert_bpf_prog_info(env, info_node)) {
 			free(info_linear);
 			free(info_node);
@@ -803,6 +847,7 @@ static void perf_env__add_bpf_info(struct perf_env *env, u32 id)
 	arrays |= 1UL << PERF_BPIL_JITED_INSNS;
 	arrays |= 1UL << PERF_BPIL_LINE_INFO;
 	arrays |= 1UL << PERF_BPIL_JITED_LINE_INFO;
+	arrays |= 1UL << PERF_BPIL_MAP_IDS;
 
 	info_linear = get_bpf_prog_info_linear(fd, arrays);
 	if (IS_ERR_OR_NULL(info_linear)) {
@@ -815,6 +860,7 @@ static void perf_env__add_bpf_info(struct perf_env *env, u32 id)
 	info_node = malloc(sizeof(struct bpf_prog_info_node));
 	if (info_node) {
 		info_node->info_linear = info_linear;
+		info_node->metadata = bpf_metadata_create(&info_linear->info);
 		if (!perf_env__insert_bpf_prog_info(env, info_node)) {
 			free(info_linear);
 			free(info_node);

tools/perf/util/bpf-event.h

@@ -25,6 +25,7 @@ struct bpf_metadata {
 
 struct bpf_prog_info_node {
 	struct perf_bpil		*info_linear;
+	struct bpf_metadata		*metadata;
 	struct rb_node			rb_node;
 };
 

tools/perf/util/env.c

@@ -3,8 +3,10 @@
 #include "debug.h"
 #include "env.h"
 #include "util/header.h"
-#include "linux/compiler.h"
+#include "util/rwsem.h"
+#include <linux/compiler.h>
 #include <linux/ctype.h>
+#include <linux/rbtree.h>
 #include <linux/string.h>
 #include <linux/zalloc.h>
 #include "cgroup.h"
@@ -89,6 +91,20 @@ struct bpf_prog_info_node *perf_env__find_bpf_prog_info(struct perf_env *env,
 	return node;
 }
 
+void perf_env__iterate_bpf_prog_info(struct perf_env *env,
+				     void (*cb)(struct bpf_prog_info_node *node,
+						void *data),
+				     void *data)
+{
+	struct rb_node *first;
+
+	down_read(&env->bpf_progs.lock);
+	first = rb_first(&env->bpf_progs.infos);
+	for (struct rb_node *node = first; node != NULL; node = rb_next(node))
+		(*cb)(rb_entry(node, struct bpf_prog_info_node, rb_node), data);
+	up_read(&env->bpf_progs.lock);
+}
+
 bool perf_env__insert_btf(struct perf_env *env, struct btf_node *btf_node)
 {
 	bool ret;
@@ -174,6 +190,7 @@ static void perf_env__purge_bpf(struct perf_env *env)
 		next = rb_next(&node->rb_node);
 		rb_erase(&node->rb_node, root);
 		zfree(&node->info_linear);
+		bpf_metadata_free(node->metadata);
 		free(node);
 	}
 

tools/perf/util/env.h

@@ -174,16 +174,22 @@ const char *perf_env__raw_arch(struct perf_env *env);
 int perf_env__nr_cpus_avail(struct perf_env *env);
 
 void perf_env__init(struct perf_env *env);
+#ifdef HAVE_LIBBPF_SUPPORT
 bool __perf_env__insert_bpf_prog_info(struct perf_env *env,
 				      struct bpf_prog_info_node *info_node);
 bool perf_env__insert_bpf_prog_info(struct perf_env *env,
 				    struct bpf_prog_info_node *info_node);
 struct bpf_prog_info_node *perf_env__find_bpf_prog_info(struct perf_env *env,
 							__u32 prog_id);
+void perf_env__iterate_bpf_prog_info(struct perf_env *env,
+				     void (*cb)(struct bpf_prog_info_node *node,
+						void *data),
+				     void *data);
 bool perf_env__insert_btf(struct perf_env *env, struct btf_node *btf_node);
 bool __perf_env__insert_btf(struct perf_env *env, struct btf_node *btf_node);
 struct btf_node *perf_env__find_btf(struct perf_env *env, __u32 btf_id);
 struct btf_node *__perf_env__find_btf(struct perf_env *env, __u32 btf_id);
+#endif // HAVE_LIBBPF_SUPPORT
 
 int perf_env__numa_node(struct perf_env *env, struct perf_cpu cpu);
 char *perf_env__find_pmu_cap(struct perf_env *env, const char *pmu_name,

tools/perf/util/header.c

@@ -3145,6 +3145,7 @@ static int process_bpf_prog_info(struct feat_fd *ff, void *data __maybe_unused)
 		/* after reading from file, translate offset to address */
 		bpil_offs_to_addr(info_linear);
 		info_node->info_linear = info_linear;
+		info_node->metadata = NULL;
 		if (!__perf_env__insert_bpf_prog_info(env, info_node)) {
 			free(info_linear);
 			free(info_node);

tools/perf/util/synthetic-events.h

@@ -92,6 +92,8 @@ int perf_event__synthesize_threads(const struct perf_tool *tool, perf_event__han
 int perf_event__synthesize_tracing_data(const struct perf_tool *tool, int fd, struct evlist *evlist, perf_event__handler_t process);
 int perf_event__synth_time_conv(const struct perf_event_mmap_page *pc, const struct perf_tool *tool, perf_event__handler_t process, struct machine *machine);
 pid_t perf_event__synthesize_comm(const struct perf_tool *tool, union perf_event *event, pid_t pid, perf_event__handler_t process, struct machine *machine);
+void perf_event__synthesize_final_bpf_metadata(struct perf_session *session,
+					       perf_event__handler_t process);
 
 int perf_tool__process_synth_event(const struct perf_tool *tool, union perf_event *event, struct machine *machine, perf_event__handler_t process);