Merge tag 'perf-core-for-mingo-4.14-20170728' of...

Merge tag 'perf-core-for-mingo-4.14-20170728' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core

Pull perf/core improvements and fixes for 4.14 from Arnaldo Carvalho de Melo:

New features:

 - Add PERF_SAMPLE_CALLCHAIN and PERF_RECORD_MMAP[2] to 'perf data' CTF
   conversion, allowing CTF trace visualization tools to show callchains
   and to resolve symbols (Geneviève Bastien)

Improvements:

 - Use group read for event groups in 'perf stat', reducing overhead when
   groups are defined in the event specification, i.e. when using {} to
   enclose a list of events, asking them to be read at the same time,
   e.g.: "perf stat -e '{cycles,instructions}'" (Jiri Olsa)

Fixes:

 - Do not overwrite perf_sample->weight in 'perf annotate' when
   processing samples, use whatever came from the kernel when
   perf_event_attr.sample_type has PERF_SAMPLE_WEIGHT set or just handle
   its default value, 0, when that is not set and "weight" is one of the
   sort orders chosen (Arnaldo Carvalho de Melo)

 - 'perf annotate --show-total-period' fixes:
    - TUI should show period, not nr_samples
    - Set appropriate column width for period/percent
    - Fix the column header to show "Period" when when that is what
      is being asked for
   (Taeung Song, Arnaldo Carvalho de Melo)

 - Use default sort if evlist is empty, fixing pipe mode (David Carrillo-Cisneros)
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>

tools/perf/builtin-annotate.c

@@ -177,8 +177,6 @@ static int perf_evsel__add_sample(struct perf_evsel *evsel,
 	 */
 	process_branch_stack(sample->branch_stack, al, sample);
 
-	sample->weight = 1;
-
 	he = hists__add_entry(hists, al, NULL, NULL, NULL, sample, true);
 	if (he == NULL)
 		return -ENOMEM;

tools/perf/builtin-data.c

@@ -69,7 +69,7 @@ static int cmd_data_convert(int argc, const char **argv)
 	};
 
 #ifndef HAVE_LIBBABELTRACE_SUPPORT
-	pr_err("No conversion support compiled in.\n");
+	pr_err("No conversion support compiled in. perf should be compiled with environment variables LIBBABELTRACE=1 and LIBBABELTRACE_DIR=/path/to/libbabeltrace/\n");
 	return -1;
 #endif
 

tools/perf/builtin-stat.c

@@ -213,10 +213,20 @@ static void perf_stat__reset_stats(void)
 static int create_perf_stat_counter(struct perf_evsel *evsel)
 {
 	struct perf_event_attr *attr = &evsel->attr;
+	struct perf_evsel *leader = evsel->leader;
 
-	if (stat_config.scale)
+	if (stat_config.scale) {
 		attr->read_format = PERF_FORMAT_TOTAL_TIME_ENABLED |
 				    PERF_FORMAT_TOTAL_TIME_RUNNING;
+	}
+
+	/*
+	 * The event is part of non trivial group, let's enable
+	 * the group read (for leader) and ID retrieval for all
+	 * members.
+	 */
+	if (leader->nr_members > 1)
+		attr->read_format |= PERF_FORMAT_ID|PERF_FORMAT_GROUP;
 
 	attr->inherit = !no_inherit;
 
@@ -333,13 +343,21 @@ static int read_counter(struct perf_evsel *counter)
 			struct perf_counts_values *count;
 
 			count = perf_counts(counter->counts, cpu, thread);
-			if (perf_evsel__read(counter, cpu, thread, count)) {
+
+			/*
+			 * The leader's group read loads data into its group members
+			 * (via perf_evsel__read_counter) and sets threir count->loaded.
+			 */
+			if (!count->loaded &&
+			    perf_evsel__read_counter(counter, cpu, thread)) {
 				counter->counts->scaled = -1;
 				perf_counts(counter->counts, cpu, thread)->ena = 0;
 				perf_counts(counter->counts, cpu, thread)->run = 0;
 				return -1;
 			}
 
+			count->loaded = false;
+
 			if (STAT_RECORD) {
 				if (perf_evsel__write_stat_event(counter, cpu, thread, count)) {
 					pr_err("failed to write stat event\n");
@@ -559,6 +577,11 @@ static int store_counter_ids(struct perf_evsel *counter)
 	return __store_counter_ids(counter, cpus, threads);
 }
 
+static bool perf_evsel__should_store_id(struct perf_evsel *counter)
+{
+	return STAT_RECORD || counter->attr.read_format & PERF_FORMAT_ID;
+}
+
 static int __run_perf_stat(int argc, const char **argv)
 {
 	int interval = stat_config.interval;
@@ -631,7 +654,8 @@ static int __run_perf_stat(int argc, const char **argv)
 		if (l > unit_width)
 			unit_width = l;
 
-		if (STAT_RECORD && store_counter_ids(counter))
+		if (perf_evsel__should_store_id(counter) &&
+		    store_counter_ids(counter))
 			return -1;
 	}
 

tools/perf/ui/browsers/annotate.c

@@ -17,8 +17,8 @@
 #include <sys/ttydefaults.h>
 
 struct disasm_line_samples {
-	double		percent;
-	u64		nr;
+	double		      percent;
+	struct sym_hist_entry he;
 };
 
 #define IPC_WIDTH 6
@@ -110,11 +110,12 @@ static int annotate_browser__set_jumps_percent_color(struct annotate_browser *br
 
 static int annotate_browser__pcnt_width(struct annotate_browser *ab)
 {
-	int w = 7 * ab->nr_events;
+	return (annotate_browser__opts.show_total_period ? 12 : 7) * ab->nr_events;
+}
 
-	if (ab->have_cycles)
-		w += IPC_WIDTH + CYCLES_WIDTH;
-	return w;
+static int annotate_browser__cycles_width(struct annotate_browser *ab)
+{
+	return ab->have_cycles ? IPC_WIDTH + CYCLES_WIDTH : 0;
 }
 
 static void annotate_browser__write(struct ui_browser *browser, void *entry, int row)
@@ -127,7 +128,8 @@ static void annotate_browser__write(struct ui_browser *browser, void *entry, int
 			     (!current_entry || (browser->use_navkeypressed &&
 					         !browser->navkeypressed)));
 	int width = browser->width, printed;
-	int i, pcnt_width = annotate_browser__pcnt_width(ab);
+	int i, pcnt_width = annotate_browser__pcnt_width(ab),
+	       cycles_width = annotate_browser__cycles_width(ab);
 	double percent_max = 0.0;
 	char bf[256];
 	bool show_title = false;
@@ -151,8 +153,8 @@ static void annotate_browser__write(struct ui_browser *browser, void *entry, int
 						bdl->samples[i].percent,
 						current_entry);
 			if (annotate_browser__opts.show_total_period) {
-				ui_browser__printf(browser, "%6" PRIu64 " ",
-						   bdl->samples[i].nr);
+				ui_browser__printf(browser, "%11" PRIu64 " ",
+						   bdl->samples[i].he.period);
 			} else {
 				ui_browser__printf(browser, "%6.2f ",
 						   bdl->samples[i].percent);
@@ -162,9 +164,11 @@ static void annotate_browser__write(struct ui_browser *browser, void *entry, int
 		ui_browser__set_percent_color(browser, 0, current_entry);
 
 		if (!show_title)
-			ui_browser__write_nstring(browser, " ", 7 * ab->nr_events);
-		else
-			ui_browser__printf(browser, "%*s", 7, "Percent");
+			ui_browser__write_nstring(browser, " ", pcnt_width);
+		else {
+			ui_browser__printf(browser, "%*s", pcnt_width,
+					   annotate_browser__opts.show_total_period ? "Period" : "Percent");
+		}
 	}
 	if (ab->have_cycles) {
 		if (dl->ipc)
@@ -190,7 +194,7 @@ static void annotate_browser__write(struct ui_browser *browser, void *entry, int
 		width += 1;
 
 	if (!*dl->line)
-		ui_browser__write_nstring(browser, " ", width - pcnt_width);
+		ui_browser__write_nstring(browser, " ", width - pcnt_width - cycles_width);
 	else if (dl->offset == -1) {
 		if (dl->line_nr && annotate_browser__opts.show_linenr)
 			printed = scnprintf(bf, sizeof(bf), "%-*d ",
@@ -199,7 +203,7 @@ static void annotate_browser__write(struct ui_browser *browser, void *entry, int
 			printed = scnprintf(bf, sizeof(bf), "%*s  ",
 				    ab->addr_width, " ");
 		ui_browser__write_nstring(browser, bf, printed);
-		ui_browser__write_nstring(browser, dl->line, width - printed - pcnt_width + 1);
+		ui_browser__write_nstring(browser, dl->line, width - printed - pcnt_width - cycles_width + 1);
 	} else {
 		u64 addr = dl->offset;
 		int color = -1;
@@ -256,7 +260,7 @@ static void annotate_browser__write(struct ui_browser *browser, void *entry, int
 		}
 
 		disasm_line__scnprintf(dl, bf, sizeof(bf), !annotate_browser__opts.use_offset);
-		ui_browser__write_nstring(browser, bf, width - pcnt_width - 3 - printed);
+		ui_browser__write_nstring(browser, bf, width - pcnt_width - cycles_width - 3 - printed);
 	}
 
 	if (current_entry)
@@ -457,7 +461,7 @@ static void annotate_browser__calc_percent(struct annotate_browser *browser,
 						pos->offset,
 						next ? next->offset : len,
 						&path, &sample);
-			bpos->samples[i].nr = sample.nr_samples;
+			bpos->samples[i].he = sample;
 
 			if (max_percent < bpos->samples[i].percent)
 				max_percent = bpos->samples[i].percent;

tools/perf/util/annotate.c

@@ -963,8 +963,9 @@ double disasm__calc_percent(struct annotation *notes, int evidx, s64 offset,
 		u64 period = 0;
 
 		while (offset < end) {
-			hits += h->addr[offset++].nr_samples;
-			period += h->addr[offset++].period;
+			hits   += h->addr[offset].nr_samples;
+			period += h->addr[offset].period;
+			++offset;
 		}
 
 		if (h->nr_samples) {
@@ -1142,7 +1143,7 @@ static int disasm_line__print(struct disasm_line *dl, struct symbol *sym, u64 st
 			color = get_percent_color(percent);
 
 			if (symbol_conf.show_total_period)
-				color_fprintf(stdout, color, " %7" PRIu64,
+				color_fprintf(stdout, color, " %11" PRIu64,
 					      sample.period);
 			else
 				color_fprintf(stdout, color, " %7.2f", percent);
@@ -1165,7 +1166,7 @@ static int disasm_line__print(struct disasm_line *dl, struct symbol *sym, u64 st
 	} else if (max_lines && printed >= max_lines)
 		return 1;
 	else {
-		int width = 8;
+		int width = symbol_conf.show_total_period ? 12 : 8;
 
 		if (queue)
 			return -1;
@@ -1806,7 +1807,7 @@ int symbol__annotate_printf(struct symbol *sym, struct map *map,
 	int printed = 2, queue_len = 0;
 	int more = 0;
 	u64 len;
-	int width = 8;
+	int width = symbol_conf.show_total_period ? 12 : 8;
 	int graph_dotted_len;
 
 	filename = strdup(dso->long_name);

tools/perf/util/counts.h

@@ -12,6 +12,7 @@ struct perf_counts_values {
 		};
 		u64 values[3];
 	};
+	bool	loaded;
 };
 
 struct perf_counts {

tools/perf/util/data-convert-bt.c

@@ -76,6 +76,8 @@ struct ctf_writer {
 	struct bt_ctf_event_class	*comm_class;
 	struct bt_ctf_event_class	*exit_class;
 	struct bt_ctf_event_class	*fork_class;
+	struct bt_ctf_event_class	*mmap_class;
+	struct bt_ctf_event_class	*mmap2_class;
 };
 
 struct convert {
@@ -506,6 +508,81 @@ add_bpf_output_values(struct bt_ctf_event_class *event_class,
 	return ret;
 }
 
+static int
+add_callchain_output_values(struct bt_ctf_event_class *event_class,
+		      struct bt_ctf_event *event,
+		      struct ip_callchain *callchain)
+{
+	struct bt_ctf_field_type *len_type, *seq_type;
+	struct bt_ctf_field *len_field, *seq_field;
+	unsigned int nr_elements = callchain->nr;
+	unsigned int i;
+	int ret;
+
+	len_type = bt_ctf_event_class_get_field_by_name(
+			event_class, "perf_callchain_size");
+	len_field = bt_ctf_field_create(len_type);
+	if (!len_field) {
+		pr_err("failed to create 'perf_callchain_size' for callchain output event\n");
+		ret = -1;
+		goto put_len_type;
+	}
+
+	ret = bt_ctf_field_unsigned_integer_set_value(len_field, nr_elements);
+	if (ret) {
+		pr_err("failed to set field value for perf_callchain_size\n");
+		goto put_len_field;
+	}
+	ret = bt_ctf_event_set_payload(event, "perf_callchain_size", len_field);
+	if (ret) {
+		pr_err("failed to set payload to perf_callchain_size\n");
+		goto put_len_field;
+	}
+
+	seq_type = bt_ctf_event_class_get_field_by_name(
+			event_class, "perf_callchain");
+	seq_field = bt_ctf_field_create(seq_type);
+	if (!seq_field) {
+		pr_err("failed to create 'perf_callchain' for callchain output event\n");
+		ret = -1;
+		goto put_seq_type;
+	}
+
+	ret = bt_ctf_field_sequence_set_length(seq_field, len_field);
+	if (ret) {
+		pr_err("failed to set length of 'perf_callchain'\n");
+		goto put_seq_field;
+	}
+
+	for (i = 0; i < nr_elements; i++) {
+		struct bt_ctf_field *elem_field =
+			bt_ctf_field_sequence_get_field(seq_field, i);
+
+		ret = bt_ctf_field_unsigned_integer_set_value(elem_field,
+				((u64 *)(callchain->ips))[i]);
+
+		bt_ctf_field_put(elem_field);
+		if (ret) {
+			pr_err("failed to set callchain[%d]\n", i);
+			goto put_seq_field;
+		}
+	}
+
+	ret = bt_ctf_event_set_payload(event, "perf_callchain", seq_field);
+	if (ret)
+		pr_err("failed to set payload for raw_data\n");
+
+put_seq_field:
+	bt_ctf_field_put(seq_field);
+put_seq_type:
+	bt_ctf_field_type_put(seq_type);
+put_len_field:
+	bt_ctf_field_put(len_field);
+put_len_type:
+	bt_ctf_field_type_put(len_type);
+	return ret;
+}
+
 static int add_generic_values(struct ctf_writer *cw,
 			      struct bt_ctf_event *event,
 			      struct perf_evsel *evsel,
@@ -519,7 +596,6 @@ static int add_generic_values(struct ctf_writer *cw,
 	 *   PERF_SAMPLE_TIME         - not needed as we have it in
 	 *                              ctf event header
 	 *   PERF_SAMPLE_READ         - TODO
-	 *   PERF_SAMPLE_CALLCHAIN    - TODO
 	 *   PERF_SAMPLE_RAW          - tracepoint fields are handled separately
 	 *   PERF_SAMPLE_BRANCH_STACK - TODO
 	 *   PERF_SAMPLE_REGS_USER    - TODO
@@ -720,6 +796,7 @@ static int process_sample_event(struct perf_tool *tool,
 	struct bt_ctf_event_class *event_class;
 	struct bt_ctf_event *event;
 	int ret;
+	unsigned long type = evsel->attr.sample_type;
 
 	if (WARN_ONCE(!priv, "Failed to setup all events.\n"))
 		return 0;
@@ -751,6 +828,13 @@ static int process_sample_event(struct perf_tool *tool,
 			return -1;
 	}
 
+	if (type & PERF_SAMPLE_CALLCHAIN) {
+		ret = add_callchain_output_values(event_class,
+				event, sample->callchain);
+		if (ret)
+			return -1;
+	}
+
 	if (perf_evsel__is_bpf_output(evsel)) {
 		ret = add_bpf_output_values(event_class, event, sample);
 		if (ret)
@@ -833,6 +917,18 @@ __FUNC_PROCESS_NON_SAMPLE(exit,
 	__NON_SAMPLE_SET_FIELD(fork, u32, ptid);
 	__NON_SAMPLE_SET_FIELD(fork, u64, time);
 )
+__FUNC_PROCESS_NON_SAMPLE(mmap,
+	__NON_SAMPLE_SET_FIELD(mmap, u32, pid);
+	__NON_SAMPLE_SET_FIELD(mmap, u32, tid);
+	__NON_SAMPLE_SET_FIELD(mmap, u64_hex, start);
+	__NON_SAMPLE_SET_FIELD(mmap, string, filename);
+)
+__FUNC_PROCESS_NON_SAMPLE(mmap2,
+	__NON_SAMPLE_SET_FIELD(mmap2, u32, pid);
+	__NON_SAMPLE_SET_FIELD(mmap2, u32, tid);
+	__NON_SAMPLE_SET_FIELD(mmap2, u64_hex, start);
+	__NON_SAMPLE_SET_FIELD(mmap2, string, filename);
+)
 #undef __NON_SAMPLE_SET_FIELD
 #undef __FUNC_PROCESS_NON_SAMPLE
 
@@ -1043,6 +1139,14 @@ static int add_generic_types(struct ctf_writer *cw, struct perf_evsel *evsel,
 	if (type & PERF_SAMPLE_TRANSACTION)
 		ADD_FIELD(event_class, cw->data.u64, "perf_transaction");
 
+	if (type & PERF_SAMPLE_CALLCHAIN) {
+		ADD_FIELD(event_class, cw->data.u32, "perf_callchain_size");
+		ADD_FIELD(event_class,
+			bt_ctf_field_type_sequence_create(
+				cw->data.u64_hex, "perf_callchain_size"),
+			"perf_callchain");
+	}
+
 #undef ADD_FIELD
 	return 0;
 }
@@ -1164,6 +1268,19 @@ __FUNC_ADD_NON_SAMPLE_EVENT_CLASS(exit,
 	__NON_SAMPLE_ADD_FIELD(u64, time);
 )
 
+__FUNC_ADD_NON_SAMPLE_EVENT_CLASS(mmap,
+	__NON_SAMPLE_ADD_FIELD(u32, pid);
+	__NON_SAMPLE_ADD_FIELD(u32, tid);
+	__NON_SAMPLE_ADD_FIELD(u64_hex, start);
+	__NON_SAMPLE_ADD_FIELD(string, filename);
+)
+
+__FUNC_ADD_NON_SAMPLE_EVENT_CLASS(mmap2,
+	__NON_SAMPLE_ADD_FIELD(u32, pid);
+	__NON_SAMPLE_ADD_FIELD(u32, tid);
+	__NON_SAMPLE_ADD_FIELD(u64_hex, start);
+	__NON_SAMPLE_ADD_FIELD(string, filename);
+)
 #undef __NON_SAMPLE_ADD_FIELD
 #undef __FUNC_ADD_NON_SAMPLE_EVENT_CLASS
 
@@ -1179,6 +1296,12 @@ static int setup_non_sample_events(struct ctf_writer *cw,
 	if (ret)
 		return ret;
 	ret = add_fork_event(cw);
+	if (ret)
+		return ret;
+	ret = add_mmap_event(cw);
+	if (ret)
+		return ret;
+	ret = add_mmap2_event(cw);
 	if (ret)
 		return ret;
 	return 0;
@@ -1482,6 +1605,8 @@ int bt_convert__perf2ctf(const char *input, const char *path,
 		c.tool.comm = process_comm_event;
 		c.tool.exit = process_exit_event;
 		c.tool.fork = process_fork_event;
+		c.tool.mmap = process_mmap_event;
+		c.tool.mmap2 = process_mmap2_event;
 	}
 
 	err = perf_config(convert__config, &c);

tools/perf/util/evlist.h

@@ -265,6 +265,11 @@ bool perf_evlist__valid_read_format(struct perf_evlist *evlist);
 void perf_evlist__splice_list_tail(struct perf_evlist *evlist,
 				   struct list_head *list);
 
+static inline bool perf_evlist__empty(struct perf_evlist *evlist)
+{
+	return list_empty(&evlist->entries);
+}
+
 static inline struct perf_evsel *perf_evlist__first(struct perf_evlist *evlist)
 {
 	return list_entry(evlist->entries.next, struct perf_evsel, node);

tools/perf/util/evsel.c

@@ -49,6 +49,7 @@ static struct {
 	bool clockid_wrong;
 	bool lbr_flags;
 	bool write_backward;
+	bool group_read;
 } perf_missing_features;
 
 static clockid_t clockid;
@@ -1261,20 +1262,148 @@ void perf_counts_values__scale(struct perf_counts_values *count,
 		*pscaled = scaled;
 }
 
+static int perf_evsel__read_size(struct perf_evsel *evsel)
+{
+	u64 read_format = evsel->attr.read_format;
+	int entry = sizeof(u64); /* value */
+	int size = 0;
+	int nr = 1;
+
+	if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
+		size += sizeof(u64);
+
+	if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING)
+		size += sizeof(u64);
+
+	if (read_format & PERF_FORMAT_ID)
+		entry += sizeof(u64);
+
+	if (read_format & PERF_FORMAT_GROUP) {
+		nr = evsel->nr_members;
+		size += sizeof(u64);
+	}
+
+	size += entry * nr;
+	return size;
+}
+
 int perf_evsel__read(struct perf_evsel *evsel, int cpu, int thread,
 		     struct perf_counts_values *count)
 {
+	size_t size = perf_evsel__read_size(evsel);
+
 	memset(count, 0, sizeof(*count));
 
 	if (FD(evsel, cpu, thread) < 0)
 		return -EINVAL;
 
-	if (readn(FD(evsel, cpu, thread), count, sizeof(*count)) <= 0)
+	if (readn(FD(evsel, cpu, thread), count->values, size) <= 0)
 		return -errno;
 
 	return 0;
 }
 
+static int
+perf_evsel__read_one(struct perf_evsel *evsel, int cpu, int thread)
+{
+	struct perf_counts_values *count = perf_counts(evsel->counts, cpu, thread);
+
+	return perf_evsel__read(evsel, cpu, thread, count);
+}
+
+static void
+perf_evsel__set_count(struct perf_evsel *counter, int cpu, int thread,
+		      u64 val, u64 ena, u64 run)
+{
+	struct perf_counts_values *count;
+
+	count = perf_counts(counter->counts, cpu, thread);
+
+	count->val    = val;
+	count->ena    = ena;
+	count->run    = run;
+	count->loaded = true;
+}
+
+static int
+perf_evsel__process_group_data(struct perf_evsel *leader,
+			       int cpu, int thread, u64 *data)
+{
+	u64 read_format = leader->attr.read_format;
+	struct sample_read_value *v;
+	u64 nr, ena = 0, run = 0, i;
+
+	nr = *data++;
+
+	if (nr != (u64) leader->nr_members)
+		return -EINVAL;
+
+	if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
+		ena = *data++;
+
+	if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING)
+		run = *data++;
+
+	v = (struct sample_read_value *) data;
+
+	perf_evsel__set_count(leader, cpu, thread,
+			      v[0].value, ena, run);
+
+	for (i = 1; i < nr; i++) {
+		struct perf_evsel *counter;
+
+		counter = perf_evlist__id2evsel(leader->evlist, v[i].id);
+		if (!counter)
+			return -EINVAL;
+
+		perf_evsel__set_count(counter, cpu, thread,
+				      v[i].value, ena, run);
+	}
+
+	return 0;
+}
+
+static int
+perf_evsel__read_group(struct perf_evsel *leader, int cpu, int thread)
+{
+	struct perf_stat_evsel *ps = leader->priv;
+	u64 read_format = leader->attr.read_format;
+	int size = perf_evsel__read_size(leader);
+	u64 *data = ps->group_data;
+
+	if (!(read_format & PERF_FORMAT_ID))
+		return -EINVAL;
+
+	if (!perf_evsel__is_group_leader(leader))
+		return -EINVAL;
+
+	if (!data) {
+		data = zalloc(size);
+		if (!data)
+			return -ENOMEM;
+
+		ps->group_data = data;
+	}
+
+	if (FD(leader, cpu, thread) < 0)
+		return -EINVAL;
+
+	if (readn(FD(leader, cpu, thread), data, size) <= 0)
+		return -errno;
+
+	return perf_evsel__process_group_data(leader, cpu, thread, data);
+}
+
+int perf_evsel__read_counter(struct perf_evsel *evsel, int cpu, int thread)
+{
+	u64 read_format = evsel->attr.read_format;
+
+	if (read_format & PERF_FORMAT_GROUP)
+		return perf_evsel__read_group(evsel, cpu, thread);
+	else
+		return perf_evsel__read_one(evsel, cpu, thread);
+}
+
 int __perf_evsel__read_on_cpu(struct perf_evsel *evsel,
 			      int cpu, int thread, bool scale)
 {
@@ -1550,6 +1679,8 @@ int perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
 	if (perf_missing_features.lbr_flags)
 		evsel->attr.branch_sample_type &= ~(PERF_SAMPLE_BRANCH_NO_FLAGS |
 				     PERF_SAMPLE_BRANCH_NO_CYCLES);
+	if (perf_missing_features.group_read && evsel->attr.inherit)
+		evsel->attr.read_format &= ~(PERF_FORMAT_GROUP|PERF_FORMAT_ID);
 retry_sample_id:
 	if (perf_missing_features.sample_id_all)
 		evsel->attr.sample_id_all = 0;
@@ -1705,6 +1836,12 @@ int perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
 		perf_missing_features.lbr_flags = true;
 		pr_debug2("switching off branch sample type no (cycles/flags)\n");
 		goto fallback_missing_features;
+	} else if (!perf_missing_features.group_read &&
+		    evsel->attr.inherit &&
+		   (evsel->attr.read_format & PERF_FORMAT_GROUP)) {
+		perf_missing_features.group_read = true;
+		pr_debug2("switching off group read\n");
+		goto fallback_missing_features;
 	}
 out_close:
 	do {

tools/perf/util/evsel.h

@@ -299,6 +299,8 @@ static inline bool perf_evsel__match2(struct perf_evsel *e1,
 int perf_evsel__read(struct perf_evsel *evsel, int cpu, int thread,
 		     struct perf_counts_values *count);
 
+int perf_evsel__read_counter(struct perf_evsel *evsel, int cpu, int thread);
+
 int __perf_evsel__read_on_cpu(struct perf_evsel *evsel,
 			      int cpu, int thread, bool scale);
 

tools/perf/util/sort.c

@@ -2563,7 +2563,7 @@ static const char *get_default_sort_order(struct perf_evlist *evlist)
 
 	BUG_ON(sort__mode >= ARRAY_SIZE(default_sort_orders));
 
-	if (evlist == NULL)
+	if (evlist == NULL || perf_evlist__empty(evlist))
 		goto out_no_evlist;
 
 	evlist__for_each_entry(evlist, evsel) {

tools/perf/util/stat.c

@@ -128,6 +128,10 @@ static int perf_evsel__alloc_stat_priv(struct perf_evsel *evsel)
 
 static void perf_evsel__free_stat_priv(struct perf_evsel *evsel)
 {
+	struct perf_stat_evsel *ps = evsel->priv;
+
+	if (ps)
+		free(ps->group_data);
 	zfree(&evsel->priv);
 }
 

tools/perf/util/stat.h

@@ -28,8 +28,9 @@ enum perf_stat_evsel_id {
 };
 
 struct perf_stat_evsel {
-	struct stats		res_stats[3];
-	enum perf_stat_evsel_id	id;
+	struct stats		 res_stats[3];
+	enum perf_stat_evsel_id	 id;
+	u64			*group_data;
 };
 
 enum aggr_mode {