Merge tag 'perf-core-for-mingo-20160630' of...

Merge tag 'perf-core-for-mingo-20160630' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core

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

New features:

 - Allow running 'perf test' entries in the same process, not forking to
   test each testcase, useful for debugging (Jiri Olsa)

 - Show number of samples in the stdio annotate header (Peter Zijlstra)

Documentation changes:

 - Add documentation for perf.data on disk format (Andi Kleen)

Build fixes:

 - Fix 'perf trace' build on old systems wrt missing SCHED_RESET_ON_FORK and
   eventfd.h (Arnaldo Carvalho de Melo)

Infrastructure changes:

 - Utility function to fetch arch from evsel/evlist (Ravi Bangoria)

Trivial changes:

 - Fix spelling mistake: "missmatch" -> "mismatch" in libbpf (Colin Ian King)
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>

tools/lib/bpf/libbpf.c

@@ -71,7 +71,7 @@ static const char *libbpf_strerror_table[NR_ERRNO] = {
 	[ERRCODE_OFFSET(LIBELF)]	= "Something wrong in libelf",
 	[ERRCODE_OFFSET(FORMAT)]	= "BPF object format invalid",
 	[ERRCODE_OFFSET(KVERSION)]	= "'version' section incorrect or lost",
-	[ERRCODE_OFFSET(ENDIAN)]	= "Endian missmatch",
+	[ERRCODE_OFFSET(ENDIAN)]	= "Endian mismatch",
 	[ERRCODE_OFFSET(INTERNAL)]	= "Internal error in libbpf",
 	[ERRCODE_OFFSET(RELOC)]		= "Relocation failed",
 	[ERRCODE_OFFSET(VERIFY)]	= "Kernel verifier blocks program loading",

tools/lib/bpf/libbpf.h

@@ -19,7 +19,7 @@ enum libbpf_errno {
 	LIBBPF_ERRNO__LIBELF = __LIBBPF_ERRNO__START,
 	LIBBPF_ERRNO__FORMAT,	/* BPF object format invalid */
 	LIBBPF_ERRNO__KVERSION,	/* Incorrect or no 'version' section */
-	LIBBPF_ERRNO__ENDIAN,	/* Endian missmatch */
+	LIBBPF_ERRNO__ENDIAN,	/* Endian mismatch */
 	LIBBPF_ERRNO__INTERNAL,	/* Internal error in libbpf */
 	LIBBPF_ERRNO__RELOC,	/* Relocation failed */
 	LIBBPF_ERRNO__LOAD,	/* Load program failure for unknown reason */

tools/perf/Documentation/perf-test.txt

@@ -30,3 +30,7 @@ OPTIONS
 -v::
 --verbose::
 	Be more verbose.
+
+-F::
+--dont-fork::
+	Do not fork child for each test, run all tests within single process.

tools/perf/tests/builtin-test.c

@@ -14,6 +14,8 @@
 #include <subcmd/parse-options.h>
 #include "symbol.h"
 
+static bool dont_fork;
+
 struct test __weak arch_tests[] = {
 	{
 		.func = NULL,
@@ -211,6 +213,10 @@ static struct test generic_tests[] = {
 		.desc = "Test backward reading from ring buffer",
 		.func = test__backward_ring_buffer,
 	},
+	{
+		.desc = "Test cpu map print",
+		.func = test__cpu_map_print,
+	},
 	{
 		.func = NULL,
 	},
@@ -247,7 +253,7 @@ static bool perf_test__matches(struct test *test, int curr, int argc, const char
 
 static int run_test(struct test *test, int subtest)
 {
-	int status, err = -1, child = fork();
+	int status, err = -1, child = dont_fork ? 0 : fork();
 	char sbuf[STRERR_BUFSIZE];
 
 	if (child < 0) {
@@ -257,9 +263,12 @@ static int run_test(struct test *test, int subtest)
 	}
 
 	if (!child) {
+		if (!dont_fork) {
 			pr_debug("test child forked, pid %d\n", getpid());
+
 			if (!verbose) {
 				int nullfd = open("/dev/null", O_WRONLY);
+
 				if (nullfd >= 0) {
 					close(STDERR_FILENO);
 					close(STDOUT_FILENO);
@@ -272,11 +281,14 @@ static int run_test(struct test *test, int subtest)
 				signal(SIGSEGV, sighandler_dump_stack);
 				signal(SIGFPE, sighandler_dump_stack);
 			}
+		}
 
 		err = test->func(subtest);
+		if (!dont_fork)
 			exit(err);
 	}
 
+	if (!dont_fork) {
 		wait(&status);
 
 		if (WIFEXITED(status)) {
@@ -286,6 +298,7 @@ static int run_test(struct test *test, int subtest)
 			err = -1;
 			pr_debug("test child interrupted\n");
 		}
+	}
 
 	return err;
 }
@@ -425,6 +438,8 @@ int cmd_test(int argc, const char **argv, const char *prefix __maybe_unused)
 	OPT_STRING('s', "skip", &skip, "tests", "tests to skip"),
 	OPT_INCR('v', "verbose", &verbose,
 		    "be more verbose (show symbol address, etc)"),
+	OPT_BOOLEAN('F', "dont-fork", &dont_fork,
+		    "Do not fork for testcase"),
 	OPT_END()
 	};
 	const char * const test_subcommands[] = { "list", NULL };

tools/perf/tests/cpumap.c

@@ -86,3 +86,27 @@ int test__cpu_map_synthesize(int subtest __maybe_unused)
 	cpu_map__put(cpus);
 	return 0;
 }
+
+static int cpu_map_print(const char *str)
+{
+	struct cpu_map *map = cpu_map__new(str);
+	char buf[100];
+
+	if (!map)
+		return -1;
+
+	cpu_map__snprint(map, buf, sizeof(buf));
+	return !strcmp(buf, str);
+}
+
+int test__cpu_map_print(int subtest __maybe_unused)
+{
+	TEST_ASSERT_VAL("failed to convert map", cpu_map_print("1"));
+	TEST_ASSERT_VAL("failed to convert map", cpu_map_print("1,5"));
+	TEST_ASSERT_VAL("failed to convert map", cpu_map_print("1,3,5,7,9,11,13,15,17,19,21-40"));
+	TEST_ASSERT_VAL("failed to convert map", cpu_map_print("2-5"));
+	TEST_ASSERT_VAL("failed to convert map", cpu_map_print("1,3-6,8-10,24,35-37"));
+	TEST_ASSERT_VAL("failed to convert map", cpu_map_print("1,3-6,8-10,24,35-37"));
+	TEST_ASSERT_VAL("failed to convert map", cpu_map_print("1-10,12-20,22-30,32-40"));
+	return 0;
+}

tools/perf/tests/dso-data.c

@@ -251,6 +251,9 @@ int test__dso_data_cache(int subtest __maybe_unused)
 	long nr_end, nr = open_files_cnt();
 	int dso_cnt, limit, i, fd;
 
+	/* Rest the internal dso open counter limit. */
+	reset_fd_limit();
+
 	memset(&machine, 0, sizeof(machine));
 
 	/* set as system limit */
@@ -312,6 +315,9 @@ int test__dso_data_reopen(int subtest __maybe_unused)
 #define dso_1 (dsos[1])
 #define dso_2 (dsos[2])
 
+	/* Rest the internal dso open counter limit. */
+	reset_fd_limit();
+
 	memset(&machine, 0, sizeof(machine));
 
 	/*

tools/perf/tests/tests.h

@@ -87,6 +87,7 @@ int test__synthesize_stat_round(int subtest);
 int test__event_update(int subtest);
 int test__event_times(int subtest);
 int test__backward_ring_buffer(int subtest);
+int test__cpu_map_print(int subtest);
 
 #if defined(__arm__) || defined(__aarch64__)
 #ifdef HAVE_DWARF_UNWIND_SUPPORT

tools/perf/tests/thread-map.c

 #include <sys/types.h>
 #include <unistd.h>
+#include <sys/prctl.h>
 #include "tests.h"
 #include "thread_map.h"
 #include "debug.h"
 
+#define NAME	(const char *) "perf"
+#define NAMEUL	(unsigned long) NAME
+
 int test__thread_map(int subtest __maybe_unused)
 {
 	struct thread_map *map;
 
+	TEST_ASSERT_VAL("failed to set process name",
+			!prctl(PR_SET_NAME, NAMEUL, 0, 0, 0));
+
 	/* test map on current pid */
 	map = thread_map__new_by_pid(getpid());
 	TEST_ASSERT_VAL("failed to alloc map", map);
@@ -19,7 +26,7 @@ int test__thread_map(int subtest __maybe_unused)
 			thread_map__pid(map, 0) == getpid());
 	TEST_ASSERT_VAL("wrong comm",
 			thread_map__comm(map, 0) &&
-			!strcmp(thread_map__comm(map, 0), "perf"));
+			!strcmp(thread_map__comm(map, 0), NAME));
 	TEST_ASSERT_VAL("wrong refcnt",
 			atomic_read(&map->refcnt) == 1);
 	thread_map__put(map);
@@ -51,7 +58,7 @@ static int process_event(struct perf_tool *tool __maybe_unused,
 
 	TEST_ASSERT_VAL("wrong nr",   map->nr == 1);
 	TEST_ASSERT_VAL("wrong pid",  map->entries[0].pid == (u64) getpid());
-	TEST_ASSERT_VAL("wrong comm", !strcmp(map->entries[0].comm, "perf"));
+	TEST_ASSERT_VAL("wrong comm", !strcmp(map->entries[0].comm, NAME));
 
 	threads = thread_map__new_event(&event->thread_map);
 	TEST_ASSERT_VAL("failed to alloc map", threads);
@@ -61,7 +68,7 @@ static int process_event(struct perf_tool *tool __maybe_unused,
 			thread_map__pid(threads, 0) == getpid());
 	TEST_ASSERT_VAL("wrong comm",
 			thread_map__comm(threads, 0) &&
-			!strcmp(thread_map__comm(threads, 0), "perf"));
+			!strcmp(thread_map__comm(threads, 0), NAME));
 	TEST_ASSERT_VAL("wrong refcnt",
 			atomic_read(&threads->refcnt) == 1);
 	thread_map__put(threads);
@@ -72,6 +79,9 @@ int test__thread_map_synthesize(int subtest __maybe_unused)
 {
 	struct thread_map *threads;
 
+	TEST_ASSERT_VAL("failed to set process name",
+			!prctl(PR_SET_NAME, NAMEUL, 0, 0, 0));
+
 	/* test map on current pid */
 	threads = thread_map__new_by_pid(getpid());
 	TEST_ASSERT_VAL("failed to alloc map", threads);

tools/perf/trace/beauty/eventfd.c

-#include <sys/eventfd.h>
-
 #ifndef EFD_SEMAPHORE
 #define EFD_SEMAPHORE		1
 #endif

tools/perf/trace/beauty/sched_policy.c

@@ -9,6 +9,9 @@
 #ifndef SCHED_DEADLINE
 #define SCHED_DEADLINE 6
 #endif
+#ifndef SCHED_RESET_ON_FORK
+#define SCHED_RESET_ON_FORK 0x40000000
+#endif
 
 static size_t syscall_arg__scnprintf_sched_policy(char *bf, size_t size,
 						  struct syscall_arg *arg)

tools/perf/util/annotate.c

@@ -1522,13 +1522,14 @@ int symbol__annotate_printf(struct symbol *sym, struct map *map,
 	const char *d_filename;
 	const char *evsel_name = perf_evsel__name(evsel);
 	struct annotation *notes = symbol__annotation(sym);
+	struct sym_hist *h = annotation__histogram(notes, evsel->idx);
 	struct disasm_line *pos, *queue = NULL;
 	u64 start = map__rip_2objdump(map, sym->start);
 	int printed = 2, queue_len = 0;
 	int more = 0;
 	u64 len;
 	int width = 8;
-	int namelen, evsel_name_len, graph_dotted_len;
+	int graph_dotted_len;
 
 	filename = strdup(dso->long_name);
 	if (!filename)
@@ -1540,17 +1541,14 @@ int symbol__annotate_printf(struct symbol *sym, struct map *map,
 		d_filename = basename(filename);
 
 	len = symbol__size(sym);
-	namelen = strlen(d_filename);
-	evsel_name_len = strlen(evsel_name);
 
 	if (perf_evsel__is_group_event(evsel))
 		width *= evsel->nr_members;
 
-	printf(" %-*.*s|	Source code & Disassembly of %s for %s\n",
-	       width, width, "Percent", d_filename, evsel_name);
+	graph_dotted_len = printf(" %-*.*s|	Source code & Disassembly of %s for %s (%" PRIu64 " samples)\n",
+	       width, width, "Percent", d_filename, evsel_name, h->sum);
 
-	graph_dotted_len = width + namelen + evsel_name_len;
-	printf("-%-*.*s-----------------------------------------\n",
+	printf("%-*.*s----\n",
 	       graph_dotted_len, graph_dotted_len, graph_dotted_line);
 
 	if (verbose)

tools/perf/util/cpumap.c

@@ -236,13 +236,12 @@ struct cpu_map *cpu_map__new_data(struct cpu_map_data *data)
 
 size_t cpu_map__fprintf(struct cpu_map *map, FILE *fp)
 {
-	int i;
-	size_t printed = fprintf(fp, "%d cpu%s: ",
-				 map->nr, map->nr > 1 ? "s" : "");
-	for (i = 0; i < map->nr; ++i)
-		printed += fprintf(fp, "%s%d", i ? ", " : "", map->map[i]);
+#define BUFSIZE 1024
+	char buf[BUFSIZE];
 
-	return printed + fprintf(fp, "\n");
+	cpu_map__snprint(map, buf, sizeof(buf));
+	return fprintf(fp, "%s\n", buf);
+#undef BUFSIZE
 }
 
 struct cpu_map *cpu_map__dummy_new(void)
@@ -599,3 +598,46 @@ bool cpu_map__has(struct cpu_map *cpus, int cpu)
 
 	return false;
 }
+
+size_t cpu_map__snprint(struct cpu_map *map, char *buf, size_t size)
+{
+	int i, cpu, start = -1;
+	bool first = true;
+	size_t ret = 0;
+
+#define COMMA first ? "" : ","
+
+	for (i = 0; i < map->nr + 1; i++) {
+		bool last = i == map->nr;
+
+		cpu = last ? INT_MAX : map->map[i];
+
+		if (start == -1) {
+			start = i;
+			if (last) {
+				ret += snprintf(buf + ret, size - ret,
+						"%s%d", COMMA,
+						map->map[i]);
+			}
+		} else if (((i - start) != (cpu - map->map[start])) || last) {
+			int end = i - 1;
+
+			if (start == end) {
+				ret += snprintf(buf + ret, size - ret,
+						"%s%d", COMMA,
+						map->map[start]);
+			} else {
+				ret += snprintf(buf + ret, size - ret,
+						"%s%d-%d", COMMA,
+						map->map[start], map->map[end]);
+			}
+			first = false;
+			start = i;
+		}
+	}
+
+#undef COMMA
+
+	pr_debug("cpumask list: %s\n", buf);
+	return ret;
+}

tools/perf/util/cpumap.h

@@ -19,6 +19,7 @@ struct cpu_map *cpu_map__empty_new(int nr);
 struct cpu_map *cpu_map__dummy_new(void);
 struct cpu_map *cpu_map__new_data(struct cpu_map_data *data);
 struct cpu_map *cpu_map__read(FILE *file);
+size_t cpu_map__snprint(struct cpu_map *map, char *buf, size_t size);
 size_t cpu_map__fprintf(struct cpu_map *map, FILE *fp);
 int cpu_map__get_socket_id(int cpu);
 int cpu_map__get_socket(struct cpu_map *map, int idx, void *data);

tools/perf/util/dso.c

@@ -442,17 +442,27 @@ static rlim_t get_fd_limit(void)
 	return limit;
 }
 
-static bool may_cache_fd(void)
+static rlim_t fd_limit;
+
+/*
+ * Used only by tests/dso-data.c to reset the environment
+ * for tests. I dont expect we should change this during
+ * standard runtime.
+ */
+void reset_fd_limit(void)
 {
-	static rlim_t limit;
+	fd_limit = 0;
+}
 
-	if (!limit)
-		limit = get_fd_limit();
+static bool may_cache_fd(void)
+{
+	if (!fd_limit)
+		fd_limit = get_fd_limit();
 
-	if (limit == RLIM_INFINITY)
+	if (fd_limit == RLIM_INFINITY)
 		return true;
 
-	return limit > (rlim_t) dso__data_open_cnt;
+	return fd_limit > (rlim_t) dso__data_open_cnt;
 }
 
 /*

tools/perf/util/dso.h

@@ -360,4 +360,6 @@ enum dso_type dso__type(struct dso *dso, struct machine *machine);
 
 int dso__strerror_load(struct dso *dso, char *buf, size_t buflen);
 
+void reset_fd_limit(void);
+
 #endif /* __PERF_DSO */

tools/perf/util/event.c

@@ -1092,7 +1092,7 @@ size_t perf_event__fprintf_cpu_map(union perf_event *event, FILE *fp)
 	struct cpu_map *cpus = cpu_map__new_data(&event->cpu_map.data);
 	size_t ret;
 
-	ret = fprintf(fp, " nr: ");
+	ret = fprintf(fp, ": ");
 
 	if (cpus)
 		ret += cpu_map__fprintf(cpus, fp);

tools/perf/util/evsel.c

@@ -2422,3 +2422,10 @@ int perf_evsel__open_strerror(struct perf_evsel *evsel, struct target *target,
 			 err, strerror_r(err, sbuf, sizeof(sbuf)),
 			 perf_evsel__name(evsel));
 }
+
+char *perf_evsel__env_arch(struct perf_evsel *evsel)
+{
+	if (evsel && evsel->evlist && evsel->evlist->env)
+		return evsel->evlist->env->arch;
+	return NULL;
+}

tools/perf/util/evsel.h

@@ -435,4 +435,6 @@ typedef int (*attr__fprintf_f)(FILE *, const char *, const char *, void *);
 int perf_event_attr__fprintf(FILE *fp, struct perf_event_attr *attr,
 			     attr__fprintf_f attr__fprintf, void *priv);
 
+char *perf_evsel__env_arch(struct perf_evsel *evsel);
+
 #endif /* __PERF_EVSEL_H */