Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 5dcefda0 authored by Ingo Molnar's avatar Ingo Molnar
Browse files

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

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

 into perf/core

Fixes and improvements for perf/core:

- perf_target: abstraction for --uid, --pid, --tid, --cpu, --all-cpus handling,
  eliminating code duplicated in the tools, having constraints that apply to
  all of them, from Namhyung Kim

- Fixes for handling fallback to cpu-clock on PPC, from David Ahern

- Fix for processing events with unknown size, from Jiri Olsa

- Compilation fix on 32-bit, from Jiri Olsa

Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: default avatarIngo Molnar <mingo@kernel.org>
parents c4f400e8 5a5626b1

tools/perf/Makefile

+8 −0
Original line number Diff line number Diff line
@@ -300,6 +300,7 @@ LIB_H += util/cpumap.h 
LIB_H += util/top.h

LIB_H += $(ARCH_INCLUDE)

LIB_H += util/cgroup.h

LIB_H += util/target.h



LIB_OBJS += $(OUTPUT)util/abspath.o

LIB_OBJS += $(OUTPUT)util/alias.o
@@ -361,6 +362,7 @@ LIB_OBJS += $(OUTPUT)util/util.o 
LIB_OBJS += $(OUTPUT)util/xyarray.o

LIB_OBJS += $(OUTPUT)util/cpumap.o

LIB_OBJS += $(OUTPUT)util/cgroup.o

LIB_OBJS += $(OUTPUT)util/target.o



BUILTIN_OBJS += $(OUTPUT)builtin-annotate.o
@@ -481,6 +483,7 @@ else 
		LIB_OBJS += $(OUTPUT)ui/helpline.o

		LIB_OBJS += $(OUTPUT)ui/progress.o

		LIB_OBJS += $(OUTPUT)ui/util.o

		LIB_OBJS += $(OUTPUT)ui/tui/setup.o

		LIB_H += ui/browser.h

		LIB_H += ui/browsers/map.h

		LIB_H += ui/helpline.h
@@ -503,6 +506,11 @@ else 
		BASIC_CFLAGS += $(shell pkg-config --cflags gtk+-2.0)

		EXTLIBS += $(shell pkg-config --libs gtk+-2.0)

		LIB_OBJS += $(OUTPUT)ui/gtk/browser.o

		LIB_OBJS += $(OUTPUT)ui/gtk/setup.o

		# Make sure that it'd be included only once.

		ifneq ($(findstring -DNO_NEWT_SUPPORT,$(BASIC_CFLAGS)),)

			LIB_OBJS += $(OUTPUT)ui/setup.o

		endif

	endif

endif

tools/perf/builtin-record.c

+38 −22
Original line number Diff line number Diff line
@@ -44,7 +44,6 @@ struct perf_record { 
	struct perf_evlist	*evlist;

	struct perf_session	*session;

	const char		*progname;

	const char		*uid_str;

	int			output;

	unsigned int		page_size;

	int			realtime_prio;
@@ -218,7 +217,7 @@ static void perf_record__open(struct perf_record *rec) 
			if (err == EPERM || err == EACCES) {

				ui__error_paranoid();

				exit(EXIT_FAILURE);

			} else if (err ==  ENODEV && opts->cpu_list) {

			} else if (err ==  ENODEV && opts->target.cpu_list) {

				die("No such device - did you specify"

					" an out-of-range profile CPU?\n");

			} else if (err == EINVAL) {
@@ -243,9 +242,13 @@ static void perf_record__open(struct perf_record *rec) 
			/*

			 * If it's cycles then fall back to hrtimer

			 * based cpu-clock-tick sw counter, which

			 * is always available even if no PMU support:

			 * is always available even if no PMU support.

			 *

			 * PPC returns ENXIO until 2.6.37 (behavior changed

			 * with commit b0a873e).

			 */

			if (err == ENOENT && attr->type == PERF_TYPE_HARDWARE

			if ((err == ENOENT || err == ENXIO)

					&& attr->type == PERF_TYPE_HARDWARE

					&& attr->config == PERF_COUNT_HW_CPU_CYCLES) {



				if (verbose)
@@ -253,6 +256,10 @@ static void perf_record__open(struct perf_record *rec) 
						    "trying to fall back to cpu-clock-ticks\n");

				attr->type = PERF_TYPE_SOFTWARE;

				attr->config = PERF_COUNT_SW_CPU_CLOCK;

				if (pos->name) {

					free(pos->name);

					pos->name = NULL;

				}

				goto try_again;

			}
@@ -578,7 +585,7 @@ static int __cmd_record(struct perf_record *rec, int argc, const char **argv) 
		perf_session__process_machines(session, tool,

					       perf_event__synthesize_guest_os);



	if (!opts->system_wide)

	if (!opts->target.system_wide)

		perf_event__synthesize_thread_map(tool, evsel_list->threads,

						  process_synthesized_event,

						  machine);
@@ -765,9 +772,9 @@ const struct option record_options[] = { 
		     parse_events_option),

	OPT_CALLBACK(0, "filter", &record.evlist, "filter",

		     "event filter", parse_filter),

	OPT_STRING('p', "pid", &record.opts.target_pid, "pid",

	OPT_STRING('p', "pid", &record.opts.target.pid, "pid",

		    "record events on existing process id"),

	OPT_STRING('t', "tid", &record.opts.target_tid, "tid",

	OPT_STRING('t', "tid", &record.opts.target.tid, "tid",

		    "record events on existing thread id"),

	OPT_INTEGER('r', "realtime", &record.realtime_prio,

		    "collect data with this RT SCHED_FIFO priority"),
@@ -775,11 +782,11 @@ const struct option record_options[] = { 
		    "collect data without buffering"),

	OPT_BOOLEAN('R', "raw-samples", &record.opts.raw_samples,

		    "collect raw sample records from all opened counters"),

	OPT_BOOLEAN('a', "all-cpus", &record.opts.system_wide,

	OPT_BOOLEAN('a', "all-cpus", &record.opts.target.system_wide,

			    "system-wide collection from all CPUs"),

	OPT_BOOLEAN('A', "append", &record.append_file,

			    "append to the output file to do incremental profiling"),

	OPT_STRING('C', "cpu", &record.opts.cpu_list, "cpu",

	OPT_STRING('C', "cpu", &record.opts.target.cpu_list, "cpu",

		    "list of cpus to monitor"),

	OPT_BOOLEAN('f', "force", &record.force,

			"overwrite existing data file (deprecated)"),
@@ -813,7 +820,8 @@ const struct option record_options[] = { 
	OPT_CALLBACK('G', "cgroup", &record.evlist, "name",

		     "monitor event in cgroup name only",

		     parse_cgroups),

	OPT_STRING('u', "uid", &record.uid_str, "user", "user to profile"),

	OPT_STRING('u', "uid", &record.opts.target.uid_str, "user",

		   "user to profile"),



	OPT_CALLBACK_NOOPT('b', "branch-any", &record.opts.branch_stack,

		     "branch any", "sample any taken branches",
@@ -831,6 +839,7 @@ int cmd_record(int argc, const char **argv, const char *prefix __used) 
	struct perf_evsel *pos;

	struct perf_evlist *evsel_list;

	struct perf_record *rec = &record;

	char errbuf[BUFSIZ];



	perf_header__set_cmdline(argc, argv);
@@ -842,8 +851,7 @@ int cmd_record(int argc, const char **argv, const char *prefix __used) 


	argc = parse_options(argc, argv, record_options, record_usage,

			    PARSE_OPT_STOP_AT_NON_OPTION);

	if (!argc && !rec->opts.target_pid && !rec->opts.target_tid &&

		!rec->opts.system_wide && !rec->opts.cpu_list && !rec->uid_str)

	if (!argc && perf_target__none(&rec->opts.target))

		usage_with_options(record_usage, record_options);



	if (rec->force && rec->append_file) {
@@ -856,7 +864,7 @@ int cmd_record(int argc, const char **argv, const char *prefix __used) 
		rec->write_mode = WRITE_FORCE;

	}



	if (nr_cgroups && !rec->opts.system_wide) {

	if (nr_cgroups && !rec->opts.target.system_wide) {

		fprintf(stderr, "cgroup monitoring only available in"

			" system-wide mode\n");

		usage_with_options(record_usage, record_options);
@@ -883,17 +891,25 @@ int cmd_record(int argc, const char **argv, const char *prefix __used) 
		goto out_symbol_exit;

	}



	rec->opts.uid = parse_target_uid(rec->uid_str, rec->opts.target_tid,

					 rec->opts.target_pid);

	if (rec->uid_str != NULL && rec->opts.uid == UINT_MAX - 1)

		goto out_free_fd;

	err = perf_target__validate(&rec->opts.target);

	if (err) {

		perf_target__strerror(&rec->opts.target, err, errbuf, BUFSIZ);

		ui__warning("%s", errbuf);

	}



	if (rec->opts.target_pid)

		rec->opts.target_tid = rec->opts.target_pid;

	err = perf_target__parse_uid(&rec->opts.target);

	if (err) {

		int saved_errno = errno;



		perf_target__strerror(&rec->opts.target, err, errbuf, BUFSIZ);

		ui__warning("%s", errbuf);



		err = -saved_errno;

		goto out_free_fd;

	}



	if (perf_evlist__create_maps(evsel_list, rec->opts.target_pid,

				     rec->opts.target_tid, rec->opts.uid,

				     rec->opts.cpu_list) < 0)

	err = -ENOMEM;

	if (perf_evlist__create_maps(evsel_list, &rec->opts.target) < 0)

		usage_with_options(record_usage, record_options);



	list_for_each_entry(pos, &evsel_list->entries, node) {

tools/perf/builtin-report.c

+4 −8
Original line number Diff line number Diff line
@@ -304,7 +304,7 @@ static size_t hists__fprintf_nr_sample_events(struct hists *self, 
	if (evname != NULL)

		ret += fprintf(fp, " of event '%s'", evname);



	ret += fprintf(fp, "\n# Event count (approx.): %lu", nr_events);

	ret += fprintf(fp, "\n# Event count (approx.): %" PRIu64, nr_events);

	return ret + fprintf(fp, "\n#\n");

}
@@ -676,14 +676,10 @@ int cmd_report(int argc, const char **argv, const char *prefix __used) 


	}



	if (strcmp(report.input_name, "-") != 0) {

		if (report.use_gtk)

			perf_gtk_setup_browser(argc, argv, true);

		else

	if (strcmp(report.input_name, "-") != 0)

		setup_browser(true);

	} else {

	else

		use_browser = 0;

	}



	/*

	 * Only in the newt browser we are doing integrated annotation,

tools/perf/builtin-stat.c

+30 −35
Original line number Diff line number Diff line
@@ -175,22 +175,21 @@ static struct perf_event_attr very_very_detailed_attrs[] = { 


static struct perf_evlist	*evsel_list;



static bool			system_wide			=  false;

static int			run_idx				=  0;

static struct perf_target	target = {

	.uid	= UINT_MAX,

};



static int			run_idx				=  0;

static int			run_count			=  1;

static bool			no_inherit			= false;

static bool			scale				=  true;

static bool			no_aggr				= false;

static const char		*target_pid;

static const char		*target_tid;

static pid_t			child_pid			= -1;

static bool			null_run			=  false;

static int			detailed_run			=  0;

static bool			sync_run			=  false;

static bool			big_num				=  true;

static int			big_num_opt			=  -1;

static const char		*cpu_list;

static const char		*csv_sep			= NULL;

static bool			csv_output			= false;

static bool			group				= false;
@@ -293,10 +292,10 @@ static int create_perf_stat_counter(struct perf_evsel *evsel, 


	attr->inherit = !no_inherit;



	if (system_wide)

	if (!perf_target__no_cpu(&target))

		return perf_evsel__open_per_cpu(evsel, evsel_list->cpus,

						group, group_fd);

	if (!target_pid && !target_tid && (!group || evsel == first)) {

	if (perf_target__no_task(&target) && (!group || evsel == first)) {

		attr->disabled = 1;

		attr->enable_on_exec = 1;

	}
@@ -446,7 +445,7 @@ static int run_perf_stat(int argc __used, const char **argv) 
			exit(-1);

		}



		if (!target_tid && !target_pid && !system_wide)

		if (perf_target__none(&target))

			evsel_list->threads->map[0] = child_pid;



		/*
@@ -463,8 +462,13 @@ static int run_perf_stat(int argc __used, const char **argv) 


	list_for_each_entry(counter, &evsel_list->entries, node) {

		if (create_perf_stat_counter(counter, first) < 0) {

			/*

			 * PPC returns ENXIO for HW counters until 2.6.37

			 * (behavior changed with commit b0a873e).

			 */

			if (errno == EINVAL || errno == ENOSYS ||

			    errno == ENOENT || errno == EOPNOTSUPP) {

			    errno == ENOENT || errno == EOPNOTSUPP ||

			    errno == ENXIO) {

				if (verbose)

					ui__warning("%s event is not supported by the kernel.\n",

						    event_name(counter));
@@ -476,7 +480,7 @@ static int run_perf_stat(int argc __used, const char **argv) 
				error("You may not have permission to collect %sstats.\n"

				      "\t Consider tweaking"

				      " /proc/sys/kernel/perf_event_paranoid or running as root.",

				      system_wide ? "system-wide " : "");

				      target.system_wide ? "system-wide " : "");

			} else {

				error("open_counter returned with %d (%s). "

				      "/bin/dmesg may provide additional information.\n",
@@ -968,14 +972,14 @@ static void print_stat(int argc, const char **argv) 
	if (!csv_output) {

		fprintf(output, "\n");

		fprintf(output, " Performance counter stats for ");

		if (!target_pid && !target_tid) {

		if (perf_target__no_task(&target)) {

			fprintf(output, "\'%s", argv[0]);

			for (i = 1; i < argc; i++)

				fprintf(output, " %s", argv[i]);

		} else if (target_pid)

			fprintf(output, "process id \'%s", target_pid);

		} else if (target.pid)

			fprintf(output, "process id \'%s", target.pid);

		else

			fprintf(output, "thread id \'%s", target_tid);

			fprintf(output, "thread id \'%s", target.tid);



		fprintf(output, "\'");

		if (run_count > 1)
@@ -1049,11 +1053,11 @@ static const struct option options[] = { 
		     "event filter", parse_filter),

	OPT_BOOLEAN('i', "no-inherit", &no_inherit,

		    "child tasks do not inherit counters"),

	OPT_STRING('p', "pid", &target_pid, "pid",

	OPT_STRING('p', "pid", &target.pid, "pid",

		   "stat events on existing process id"),

	OPT_STRING('t', "tid", &target_tid, "tid",

	OPT_STRING('t', "tid", &target.tid, "tid",

		   "stat events on existing thread id"),

	OPT_BOOLEAN('a', "all-cpus", &system_wide,

	OPT_BOOLEAN('a', "all-cpus", &target.system_wide,

		    "system-wide collection from all CPUs"),

	OPT_BOOLEAN('g', "group", &group,

		    "put the counters into a counter group"),
@@ -1072,7 +1076,7 @@ static const struct option options[] = { 
	OPT_CALLBACK_NOOPT('B', "big-num", NULL, NULL, 

			   "print large numbers with thousands\' separators",

			   stat__set_big_num),

	OPT_STRING('C', "cpu", &cpu_list, "cpu",

	OPT_STRING('C', "cpu", &target.cpu_list, "cpu",

		    "list of cpus to monitor in system-wide"),

	OPT_BOOLEAN('A', "no-aggr", &no_aggr,

		    "disable CPU count aggregation"),
@@ -1190,13 +1194,13 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used) 
	} else if (big_num_opt == 0) /* User passed --no-big-num */

		big_num = false;



	if (!argc && !target_pid && !target_tid)

	if (!argc && perf_target__no_task(&target))

		usage_with_options(stat_usage, options);

	if (run_count <= 0)

		usage_with_options(stat_usage, options);



	/* no_aggr, cgroup are for system-wide only */

	if ((no_aggr || nr_cgroups) && !system_wide) {

	if ((no_aggr || nr_cgroups) && perf_target__no_cpu(&target)) {

		fprintf(stderr, "both cgroup and no-aggregation "

			"modes only available in system-wide mode\n");
@@ -1206,23 +1210,14 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used) 
	if (add_default_attributes())

		goto out;



	if (target_pid)

		target_tid = target_pid;

	perf_target__validate(&target);



	evsel_list->threads = thread_map__new_str(target_pid,

						  target_tid, UINT_MAX);

	if (evsel_list->threads == NULL) {

	if (perf_evlist__create_maps(evsel_list, &target) < 0) {

		if (!perf_target__no_task(&target))

			pr_err("Problems finding threads of monitor\n");

		usage_with_options(stat_usage, options);

	}



	if (system_wide)

		evsel_list->cpus = cpu_map__new(cpu_list);

	else

		evsel_list->cpus = cpu_map__dummy_new();



	if (evsel_list->cpus == NULL) {

		if (!perf_target__no_cpu(&target))

			perror("failed to parse CPUs map");



		usage_with_options(stat_usage, options);

		return -1;

	}

tools/perf/builtin-test.c

+6 −6
Original line number Diff line number Diff line
@@ -1165,6 +1165,9 @@ static int sched__get_first_possible_cpu(pid_t pid, cpu_set_t **maskp, 
static int test__PERF_RECORD(void)

{

	struct perf_record_opts opts = {

		.target = {

			.uid = UINT_MAX,

		},

		.no_delay   = true,

		.freq	    = 10,

		.mmap_pages = 256,
@@ -1207,8 +1210,7 @@ static int test__PERF_RECORD(void) 
	 * perf_evlist__prepare_workload we'll fill in the only thread

	 * we're monitoring, the one forked there.

	 */

	err = perf_evlist__create_maps(evlist, opts.target_pid,

				       opts.target_tid, UINT_MAX, opts.cpu_list);

	err = perf_evlist__create_maps(evlist, &opts.target);

	if (err < 0) {

		pr_debug("Not enough memory to create thread/cpu maps\n");

		goto out_delete_evlist;
@@ -1549,8 +1551,6 @@ static int __test__rdpmc(void) 
	sa.sa_sigaction = segfault_handler;

	sigaction(SIGSEGV, &sa, NULL);



	fprintf(stderr, "\n\n");



	fd = sys_perf_event_open(&attr, 0, -1, -1, 0);

	if (fd < 0) {

		die("Error: sys_perf_event_open() syscall returned "
@@ -1575,7 +1575,7 @@ static int __test__rdpmc(void) 
		loops *= 10;



		delta = now - stamp;

		fprintf(stderr, "%14d: %14Lu\n", n, (long long)delta);

		pr_debug("%14d: %14Lu\n", n, (long long)delta);



		delta_sum += delta;

	}
@@ -1583,7 +1583,7 @@ static int __test__rdpmc(void) 
	munmap(addr, page_size);

	close(fd);



	fprintf(stderr, "   ");

	pr_debug("   ");



	if (!delta_sum)

		return -1;