1 files changed, 82 insertions, 36 deletions

diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 72a12b69f120..b63b3fb2de70 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -47,7 +47,6 @@
 #include "util/time-utils.h"
 #include "util/auxtrace.h"
 #include "util/units.h"
-#include "util/branch.h"
 #include "util/util.h" // perf_tip()
 #include "ui/ui.h"
 #include "ui/progress.h"
@@ -84,6 +83,7 @@ struct report {
 	bool			header_only;
 	bool			nonany_branch_mode;
 	bool			group_set;
+	bool			stitch_lbr;
 	int			max_stack;
 	struct perf_read_values	show_threads_values;
 	struct annotation_options annotation_opts;
@@ -104,6 +104,7 @@ struct report {
 	bool			symbol_ipc;
 	bool			total_cycles_mode;
 	struct block_report	*block_reports;
+	int			nr_block_reports;
 };
 
 static int report__config(const char *var, const char *value, void *cb)
@@ -185,24 +186,23 @@ static int hist_iter__branch_callback(struct hist_entry_iter *iter,
 {
 	struct hist_entry *he = iter->he;
 	struct report *rep = arg;
-	struct branch_info *bi;
+	struct branch_info *bi = he->branch_info;
 	struct perf_sample *sample = iter->sample;
 	struct evsel *evsel = iter->evsel;
 	int err;
 
+	branch_type_count(&rep->brtype_stat, &bi->flags,
+			  bi->from.addr, bi->to.addr);
+
 	if (!ui__has_annotation() && !rep->symbol_ipc)
 		return 0;
 
-	bi = he->branch_info;
 	err = addr_map_symbol__inc_samples(&bi->from, sample, evsel);
 	if (err)
 		goto out;
 
 	err = addr_map_symbol__inc_samples(&bi->to, sample, evsel);
 
-	branch_type_count(&rep->brtype_stat, &bi->flags,
-			  bi->from.addr, bi->to.addr);
-
 out:
 	return err;
 }
@@ -267,6 +267,9 @@ static int process_sample_event(struct perf_tool *tool,
 		return -1;
 	}
 
+	if (rep->stitch_lbr)
+		al.thread->lbr_stitch_enable = true;
+
 	if (symbol_conf.hide_unresolved && al.sym == NULL)
 		goto out_put;
 
@@ -317,7 +320,7 @@ static int process_read_event(struct perf_tool *tool,
 	struct report *rep = container_of(tool, struct report, tool);
 
 	if (rep->show_threads) {
-		const char *name = perf_evsel__name(evsel);
+		const char *name = evsel__name(evsel);
 		int err = perf_read_values_add_value(&rep->show_threads_values,
 					   event->read.pid, event->read.tid,
 					   evsel->idx,
@@ -339,12 +342,14 @@ static int report__setup_sample_type(struct report *rep)
 	bool is_pipe = perf_data__is_pipe(session->data);
 
 	if (session->itrace_synth_opts->callchain ||
+	    session->itrace_synth_opts->add_callchain ||
 	    (!is_pipe &&
 	     perf_header__has_feat(&session->header, HEADER_AUXTRACE) &&
 	     !session->itrace_synth_opts->set))
 		sample_type |= PERF_SAMPLE_CALLCHAIN;
 
-	if (session->itrace_synth_opts->last_branch)
+	if (session->itrace_synth_opts->last_branch ||
+	    session->itrace_synth_opts->add_last_branch)
 		sample_type |= PERF_SAMPLE_BRANCH_STACK;
 
 	if (!is_pipe && !(sample_type & PERF_SAMPLE_CALLCHAIN)) {
@@ -396,15 +401,12 @@ static int report__setup_sample_type(struct report *rep)
 		}
 	}
 
-	if (symbol_conf.use_callchain || symbol_conf.cumulate_callchain) {
-		if ((sample_type & PERF_SAMPLE_REGS_USER) &&
-		    (sample_type & PERF_SAMPLE_STACK_USER)) {
-			callchain_param.record_mode = CALLCHAIN_DWARF;
-			dwarf_callchain_users = true;
-		} else if (sample_type & PERF_SAMPLE_BRANCH_STACK)
-			callchain_param.record_mode = CALLCHAIN_LBR;
-		else
-			callchain_param.record_mode = CALLCHAIN_FP;
+	callchain_param_setup(sample_type);
+
+	if (rep->stitch_lbr && (callchain_param.record_mode != CALLCHAIN_LBR)) {
+		ui__warning("Can't find LBR callchain. Switch off --stitch-lbr.\n"
+			    "Please apply --call-graph lbr when recording.\n");
+		rep->stitch_lbr = false;
 	}
 
 	/* ??? handle more cases than just ANY? */
@@ -447,10 +449,10 @@ static size_t hists__fprintf_nr_sample_events(struct hists *hists, struct report
 		nr_events = hists->stats.total_non_filtered_period;
 	}
 
-	if (perf_evsel__is_group_event(evsel)) {
+	if (evsel__is_group_event(evsel)) {
 		struct evsel *pos;
 
-		perf_evsel__group_desc(evsel, buf, size);
+		evsel__group_desc(evsel, buf, size);
 		evname = buf;
 
 		for_each_group_member(pos, evsel) {
@@ -525,10 +527,9 @@ static int perf_evlist__tty_browse_hists(struct evlist *evlist,
 
 	evlist__for_each_entry(evlist, pos) {
 		struct hists *hists = evsel__hists(pos);
-		const char *evname = perf_evsel__name(pos);
+		const char *evname = evsel__name(pos);
 
-		if (symbol_conf.event_group &&
-		    !perf_evsel__is_group_leader(pos))
+		if (symbol_conf.event_group && !evsel__is_group_leader(pos))
 			continue;
 
 		hists__fprintf_nr_sample_events(hists, rep, evname, stdout);
@@ -635,7 +636,7 @@ static int report__browse_hists(struct report *rep)
 		 * Usually "ret" is the last pressed key, and we only
 		 * care if the key notifies us to switch data file.
 		 */
-		if (ret != K_SWITCH_INPUT_DATA)
+		if (ret != K_SWITCH_INPUT_DATA && ret != K_RELOAD)
 			ret = 0;
 		break;
 	case 2:
@@ -670,8 +671,7 @@ static int report__collapse_hists(struct report *rep)
 			break;
 
 		/* Non-group events are considered as leader */
-		if (symbol_conf.event_group &&
-		    !perf_evsel__is_group_leader(pos)) {
+		if (symbol_conf.event_group && !evsel__is_group_leader(pos)) {
 			struct hists *leader_hists = evsel__hists(pos->leader);
 
 			hists__match(leader_hists, hists);
@@ -706,8 +706,7 @@ static void report__output_resort(struct report *rep)
 	ui_progress__init(&prog, rep->nr_entries, "Sorting events for output...");
 
 	evlist__for_each_entry(rep->session->evlist, pos) {
-		perf_evsel__output_resort_cb(pos, &prog,
-					     hists__resort_cb, rep);
+		evsel__output_resort_cb(pos, &prog, hists__resort_cb, rep);
 	}
 
 	ui_progress__finish();
@@ -966,8 +965,19 @@ static int __cmd_report(struct report *rep)
 	report__output_resort(rep);
 
 	if (rep->total_cycles_mode) {
+		int block_hpps[6] = {
+			PERF_HPP_REPORT__BLOCK_TOTAL_CYCLES_PCT,
+			PERF_HPP_REPORT__BLOCK_LBR_CYCLES,
+			PERF_HPP_REPORT__BLOCK_CYCLES_PCT,
+			PERF_HPP_REPORT__BLOCK_AVG_CYCLES,
+			PERF_HPP_REPORT__BLOCK_RANGE,
+			PERF_HPP_REPORT__BLOCK_DSO,
+		};
+
 		rep->block_reports = block_info__create_report(session->evlist,
-							       rep->total_cycles);
+							       rep->total_cycles,
+							       block_hpps, 6,
+							       &rep->nr_block_reports);
 		if (!rep->block_reports)
 			return -1;
 	}
@@ -1069,6 +1079,26 @@ parse_percent_limit(const struct option *opt, const char *str,
 	return 0;
 }
 
+static int process_attr(struct perf_tool *tool __maybe_unused,
+			union perf_event *event,
+			struct evlist **pevlist)
+{
+	u64 sample_type;
+	int err;
+
+	err = perf_event__process_attr(tool, event, pevlist);
+	if (err)
+		return err;
+
+	/*
+	 * Check if we need to enable callchains based
+	 * on events sample_type.
+	 */
+	sample_type = perf_evlist__combined_sample_type(*pevlist);
+	callchain_param_setup(sample_type);
+	return 0;
+}
+
 int cmd_report(int argc, const char **argv)
 {
 	struct perf_session *session;
@@ -1094,11 +1124,12 @@ int cmd_report(int argc, const char **argv)
 			.mmap2		 = perf_event__process_mmap2,
 			.comm		 = perf_event__process_comm,
 			.namespaces	 = perf_event__process_namespaces,
+			.cgroup		 = perf_event__process_cgroup,
 			.exit		 = perf_event__process_exit,
 			.fork		 = perf_event__process_fork,
 			.lost		 = perf_event__process_lost,
 			.read		 = process_read_event,
-			.attr		 = perf_event__process_attr,
+			.attr		 = process_attr,
 			.tracing_data	 = perf_event__process_tracing_data,
 			.build_id	 = perf_event__process_build_id,
 			.id_index	 = perf_event__process_id_index,
@@ -1216,6 +1247,10 @@ int cmd_report(int argc, const char **argv)
 		    "Show a column with the sum of periods"),
 	OPT_BOOLEAN_SET(0, "group", &symbol_conf.event_group, &report.group_set,
 		    "Show event group information together"),
+	OPT_INTEGER(0, "group-sort-idx", &symbol_conf.group_sort_idx,
+		    "Sort the output by the event at the index n in group. "
+		    "If n is invalid, sort by the first event. "
+		    "WARNING: should be used on grouped events."),
 	OPT_CALLBACK_NOOPT('b', "branch-stack", &branch_mode, "",
 		    "use branch records for per branch histogram filling",
 		    parse_branch_mode),
@@ -1241,6 +1276,8 @@ int cmd_report(int argc, const char **argv)
 			"Show full source file name path for source lines"),
 	OPT_BOOLEAN(0, "show-ref-call-graph", &symbol_conf.show_ref_callgraph,
 		    "Show callgraph from reference event"),
+	OPT_BOOLEAN(0, "stitch-lbr", &report.stitch_lbr,
+		    "Enable LBR callgraph stitching approach"),
 	OPT_INTEGER(0, "socket-filter", &report.socket_filter,
 		    "only show processor socket that match with this filter"),
 	OPT_BOOLEAN(0, "raw-trace", &symbol_conf.raw_trace,
@@ -1316,7 +1353,7 @@ int cmd_report(int argc, const char **argv)
 	if (symbol_conf.cumulate_callchain && !callchain_param.order_set)
 		callchain_param.order = ORDER_CALLER;
 
-	if (itrace_synth_opts.callchain &&
+	if ((itrace_synth_opts.callchain || itrace_synth_opts.add_callchain) &&
 	    (int)itrace_synth_opts.callchain_sz > report.max_stack)
 		report.max_stack = itrace_synth_opts.callchain_sz;
 
@@ -1358,7 +1395,13 @@ repeat:
 
 	setup_forced_leader(&report, session->evlist);
 
-	if (itrace_synth_opts.last_branch)
+	if (symbol_conf.group_sort_idx && !session->evlist->nr_groups) {
+		parse_options_usage(NULL, options, "group-sort-idx", 0);
+		ret = -EINVAL;
+		goto error;
+	}
+
+	if (itrace_synth_opts.last_branch || itrace_synth_opts.add_last_branch)
 		has_br_stack = true;
 
 	if (has_br_stack && branch_call_mode)
@@ -1378,7 +1421,7 @@ repeat:
 	}
 	if (branch_call_mode) {
 		callchain_param.key = CCKEY_ADDRESS;
-		callchain_param.branch_callstack = 1;
+		callchain_param.branch_callstack = true;
 		symbol_conf.use_callchain = true;
 		callchain_register_param(&callchain_param);
 		if (sort_order == NULL)
@@ -1459,7 +1502,7 @@ repeat:
 		sort_order = sort_tmp;
 	}
 
-	if ((last_key != K_SWITCH_INPUT_DATA) &&
+	if ((last_key != K_SWITCH_INPUT_DATA && last_key != K_RELOAD) &&
 	    (setup_sorting(session->evlist) < 0)) {
 		if (sort_order)
 			parse_options_usage(report_usage, options, "s", 1);
@@ -1538,7 +1581,7 @@ repeat:
 	sort__setup_elide(stdout);
 
 	ret = __cmd_report(&report);
-	if (ret == K_SWITCH_INPUT_DATA) {
+	if (ret == K_SWITCH_INPUT_DATA || ret == K_RELOAD) {
 		perf_session__delete(session);
 		last_key = K_SWITCH_INPUT_DATA;
 		goto repeat;
@@ -1551,8 +1594,11 @@ error:
 		zfree(&report.ptime_range);
 	}
 
-	if (report.block_reports)
-		zfree(&report.block_reports);
+	if (report.block_reports) {
+		block_info__free_report(report.block_reports,
+					report.nr_block_reports);
+		report.block_reports = NULL;
+	}
 
 	zstd_fini(&(session->zstd_data));
 	perf_session__delete(session);