Merge tag 'perf-core-for-mingo' of git://git.kernel.org/pub/scm/linux…

…/kernel/git/acme/linux into perf/urgent

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

  * Add summary only option to 'perf trace', suppressing the decoding of
    events, from David Ahern

  * 'perf trace --summary' formatting simplifications, from Pekka Emberg.

  * Beautify fifth argument of mmap() as fd, in 'perf trace', from Namhyung Kim.

  * Fix segfault on perf trace -i perf.data, from Namhyung Kim.

  * Fix segfault with --no-mmap-pages, from David Ahern.

  * Round mmap pages to power 2, from David Ahern.

  * Add direct access to dynamic arrays in libtraceevent, from Steven Rostedt.

  * Handle throttle events in 'object code reading' test, fix from Adrian Hunter.

  * Prevent condition that all sort keys are elided, fix from Namhyung Kim.

  * Synthesize non-exec MMAP records when --data used, allowing the resolution of
    data addresses to symbols (global variables, etc).

  * Don't force a refresh during progress update in the TUI, greatly reducing
    startup costs, fix from Patrick Palka.

  * Fix sw clock event period test wrt not checking if using > max_sample_freq.

  * Code cleanups by David Ahern and Adrian Hunter.

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>

tools/lib/traceevent/event-parse.c

@@ -3435,6 +3435,19 @@ eval_num_arg(void *data, int size, struct event_format *event, struct print_arg
 			goto out_warning_op;
 		}
 		break;
+	case PRINT_DYNAMIC_ARRAY:
+		/* Without [], we pass the address to the dynamic data */
+		offset = pevent_read_number(pevent,
+					    data + arg->dynarray.field->offset,
+					    arg->dynarray.field->size);
+		/*
+		 * The actual length of the dynamic array is stored
+		 * in the top half of the field, and the offset
+		 * is in the bottom half of the 32 bit field.
+		 */
+		offset &= 0xffff;
+		val = (unsigned long long)(data + offset);
+		break;
 	default: /* not sure what to do there */
 		return 0;
 	}

tools/perf/Documentation/perf-trace.txt

@@ -93,9 +93,15 @@ the thread executes on the designated CPUs. Default is to monitor all CPUs.
 --comm::
         Show process COMM right beside its ID, on by default, disable with --no-comm.
 
+-s::
 --summary::
-	Show a summary of syscalls by thread with min, max, and average times (in
-    msec) and relative stddev.
+	Show only a summary of syscalls by thread with min, max, and average times
+    (in msec) and relative stddev.
+
+-S::
+--with-summary::
+	Show all syscalls followed by a summary by thread with min, max, and
+    average times (in msec) and relative stddev.
 
 --tool_stats::
 	Show tool stats such as number of times fd->pathname was discovered thru

tools/perf/builtin-kvm.c

@@ -1510,13 +1510,13 @@ static int kvm_events_live(struct perf_kvm_stat *kvm,
 	/*
 	 * target related setups
 	 */
-	err = perf_target__validate(&kvm->opts.target);
+	err = target__validate(&kvm->opts.target);
 	if (err) {
-		perf_target__strerror(&kvm->opts.target, err, errbuf, BUFSIZ);
+		target__strerror(&kvm->opts.target, err, errbuf, BUFSIZ);
 		ui__warning("%s", errbuf);
 	}
 
-	if (perf_target__none(&kvm->opts.target))
+	if (target__none(&kvm->opts.target))
 		kvm->opts.target.system_wide = true;
 
 
@@ -1544,18 +1544,8 @@ static int kvm_events_live(struct perf_kvm_stat *kvm,
 	}
 	kvm->session->evlist = kvm->evlist;
 	perf_session__set_id_hdr_size(kvm->session);
-
-
-	if (perf_target__has_task(&kvm->opts.target))
-		perf_event__synthesize_thread_map(&kvm->tool,
-						  kvm->evlist->threads,
-						  perf_event__process,
-						  &kvm->session->machines.host);
-	else
-		perf_event__synthesize_threads(&kvm->tool, perf_event__process,
-					       &kvm->session->machines.host);
-
-
+	machine__synthesize_threads(&kvm->session->machines.host, &kvm->opts.target,
+				    kvm->evlist->threads, false);
 	err = kvm_live_open_events(kvm);
 	if (err)
 		goto out;

tools/perf/builtin-record.c

@@ -76,12 +76,12 @@ struct perf_record {
 	long			samples;
 };
 
-static int write_output(struct perf_record *rec, void *buf, size_t size)
+static int do_write_output(struct perf_record *rec, void *buf, size_t size)
 {
 	struct perf_data_file *file = &rec->file;
 
 	while (size) {
-		int ret = write(file->fd, buf, size);
+		ssize_t ret = write(file->fd, buf, size);
 
 		if (ret < 0) {
 			pr_err("failed to write perf data, error: %m\n");
@@ -97,6 +97,11 @@ static int write_output(struct perf_record *rec, void *buf, size_t size)
 	return 0;
 }
 
+static int write_output(struct perf_record *rec, void *buf, size_t size)
+{
+	return do_write_output(rec, buf, size);
+}
+
 static int process_synthesized_event(struct perf_tool *tool,
 				     union perf_event *event,
 				     struct perf_sample *sample __maybe_unused,
@@ -480,16 +485,8 @@ static int __cmd_record(struct perf_record *rec, int argc, const char **argv)
 					 perf_event__synthesize_guest_os, tool);
 	}
 
-	if (perf_target__has_task(&opts->target))
-		err = perf_event__synthesize_thread_map(tool, evsel_list->threads,
-						  process_synthesized_event,
-						  machine);
-	else if (perf_target__has_cpu(&opts->target))
-		err = perf_event__synthesize_threads(tool, process_synthesized_event,
-					       machine);
-	else /* command specified */
-		err = 0;
-
+	err = __machine__synthesize_threads(machine, tool, &opts->target, evsel_list->threads,
+					    process_synthesized_event, opts->sample_address);
 	if (err != 0)
 		goto out_delete_session;
 
@@ -509,7 +506,7 @@ static int __cmd_record(struct perf_record *rec, int argc, const char **argv)
 	 * (apart from group members) have enable_on_exec=1 set,
 	 * so don't spoil it by prematurely enabling them.
 	 */
-	if (!perf_target__none(&opts->target))
+	if (!target__none(&opts->target))
 		perf_evlist__enable(evsel_list);
 
 	/*
@@ -538,7 +535,7 @@ static int __cmd_record(struct perf_record *rec, int argc, const char **argv)
 		 * die with the process and we wait for that. Thus no need to
 		 * disable events in this case.
 		 */
-		if (done && !disabled && !perf_target__none(&opts->target)) {
+		if (done && !disabled && !target__none(&opts->target)) {
 			perf_evlist__disable(evsel_list);
 			disabled = true;
 		}
@@ -909,7 +906,7 @@ int cmd_record(int argc, const char **argv, const char *prefix __maybe_unused)
 
 	argc = parse_options(argc, argv, record_options, record_usage,
 			    PARSE_OPT_STOP_AT_NON_OPTION);
-	if (!argc && perf_target__none(&rec->opts.target))
+	if (!argc && target__none(&rec->opts.target))
 		usage_with_options(record_usage, record_options);
 
 	if (nr_cgroups && !rec->opts.target.system_wide) {
@@ -939,17 +936,17 @@ int cmd_record(int argc, const char **argv, const char *prefix __maybe_unused)
 		goto out_symbol_exit;
 	}
 
-	err = perf_target__validate(&rec->opts.target);
+	err = target__validate(&rec->opts.target);
 	if (err) {
-		perf_target__strerror(&rec->opts.target, err, errbuf, BUFSIZ);
+		target__strerror(&rec->opts.target, err, errbuf, BUFSIZ);
 		ui__warning("%s", errbuf);
 	}
 
-	err = perf_target__parse_uid(&rec->opts.target);
+	err = target__parse_uid(&rec->opts.target);
 	if (err) {
 		int saved_errno = errno;
 
-		perf_target__strerror(&rec->opts.target, err, errbuf, BUFSIZ);
+		target__strerror(&rec->opts.target, err, errbuf, BUFSIZ);
 		ui__error("%s", errbuf);
 
 		err = -saved_errno;

tools/perf/builtin-stat.c

@@ -108,7 +108,7 @@ enum {
 
 static struct perf_evlist	*evsel_list;
 
-static struct perf_target	target = {
+static struct target target = {
 	.uid	= UINT_MAX,
 };
 
@@ -294,11 +294,10 @@ static int create_perf_stat_counter(struct perf_evsel *evsel)
 
 	attr->inherit = !no_inherit;
 
-	if (perf_target__has_cpu(&target))
+	if (target__has_cpu(&target))
 		return perf_evsel__open_per_cpu(evsel, perf_evsel__cpus(evsel));
 
-	if (!perf_target__has_task(&target) &&
-	    perf_evsel__is_group_leader(evsel)) {
+	if (!target__has_task(&target) && perf_evsel__is_group_leader(evsel)) {
 		attr->disabled = 1;
 		if (!initial_delay)
 			attr->enable_on_exec = 1;
@@ -1236,7 +1235,7 @@ static void print_stat(int argc, const char **argv)
 			fprintf(output, "\'system wide");
 		else if (target.cpu_list)
 			fprintf(output, "\'CPU(s) %s", target.cpu_list);
-		else if (!perf_target__has_task(&target)) {
+		else if (!target__has_task(&target)) {
 			fprintf(output, "\'%s", argv[0]);
 			for (i = 1; i < argc; i++)
 				fprintf(output, " %s", argv[i]);
@@ -1667,7 +1666,7 @@ int cmd_stat(int argc, const char **argv, const char *prefix __maybe_unused)
 	} else if (big_num_opt == 0) /* User passed --no-big-num */
 		big_num = false;
 
-	if (!argc && perf_target__none(&target))
+	if (!argc && target__none(&target))
 		usage_with_options(stat_usage, options);
 
 	if (run_count < 0) {
@@ -1680,8 +1679,8 @@ int cmd_stat(int argc, const char **argv, const char *prefix __maybe_unused)
 	}
 
 	/* no_aggr, cgroup are for system-wide only */
-	if ((aggr_mode != AGGR_GLOBAL || nr_cgroups)
-	     && !perf_target__has_cpu(&target)) {
+	if ((aggr_mode != AGGR_GLOBAL || nr_cgroups) &&
+	    !target__has_cpu(&target)) {
 		fprintf(stderr, "both cgroup and no-aggregation "
 			"modes only available in system-wide mode\n");
 
@@ -1694,14 +1693,14 @@ int cmd_stat(int argc, const char **argv, const char *prefix __maybe_unused)
 	if (add_default_attributes())
 		goto out;
 
-	perf_target__validate(&target);
+	target__validate(&target);
 
 	if (perf_evlist__create_maps(evsel_list, &target) < 0) {
-		if (perf_target__has_task(&target)) {
+		if (target__has_task(&target)) {
 			pr_err("Problems finding threads of monitor\n");
 			parse_options_usage(stat_usage, options, "p", 1);
 			parse_options_usage(NULL, options, "t", 1);
-		} else if (perf_target__has_cpu(&target)) {
+		} else if (target__has_cpu(&target)) {
 			perror("failed to parse CPUs map");
 			parse_options_usage(stat_usage, options, "C", 1);
 			parse_options_usage(NULL, options, "a", 1);

tools/perf/builtin-top.c

@@ -950,14 +950,8 @@ static int __cmd_top(struct perf_top *top)
 	if (ret)
 		goto out_delete;
 
-	if (perf_target__has_task(&opts->target))
-		perf_event__synthesize_thread_map(&top->tool, top->evlist->threads,
-						  perf_event__process,
-						  &top->session->machines.host);
-	else
-		perf_event__synthesize_threads(&top->tool, perf_event__process,
-					       &top->session->machines.host);
-
+	machine__synthesize_threads(&top->session->machines.host, &opts->target,
+				    top->evlist->threads, false);
 	ret = perf_top__start_counters(top);
 	if (ret)
 		goto out_delete;
@@ -973,7 +967,7 @@ static int __cmd_top(struct perf_top *top)
 	 * XXX 'top' still doesn't start workloads like record, trace, but should,
 	 * so leave the check here.
 	 */
-        if (!perf_target__none(&opts->target))
+        if (!target__none(&opts->target))
                 perf_evlist__enable(top->evlist);
 
 	/* Wait for a minimal set of events before starting the snapshot */
@@ -1059,7 +1053,7 @@ int cmd_top(int argc, const char **argv, const char *prefix __maybe_unused)
 		.sym_pcnt_filter     = 5,
 	};
 	struct perf_record_opts *opts = &top.record_opts;
-	struct perf_target *target = &opts->target;
+	struct target *target = &opts->target;
 	const struct option options[] = {
 	OPT_CALLBACK('e', "event", &top.evlist, "event",
 		     "event selector. use 'perf list' to list available events",
@@ -1175,24 +1169,24 @@ int cmd_top(int argc, const char **argv, const char *prefix __maybe_unused)
 
 	setup_browser(false);
 
-	status = perf_target__validate(target);
+	status = target__validate(target);
 	if (status) {
-		perf_target__strerror(target, status, errbuf, BUFSIZ);
+		target__strerror(target, status, errbuf, BUFSIZ);
 		ui__warning("%s", errbuf);
 	}
 
-	status = perf_target__parse_uid(target);
+	status = target__parse_uid(target);
 	if (status) {
 		int saved_errno = errno;
 
-		perf_target__strerror(target, status, errbuf, BUFSIZ);
+		target__strerror(target, status, errbuf, BUFSIZ);
 		ui__error("%s", errbuf);
 
 		status = -saved_errno;
 		goto out_delete_evlist;
 	}
 
-	if (perf_target__none(target))
+	if (target__none(target))
 		target->system_wide = true;
 
 	if (perf_evlist__create_maps(top.evlist, target) < 0)