perf tools: Implement counter output multiplexing

Finish the -M/--multiplex option implementation:

 - separate it out from group_fd

 - correctly set it via the ioctl and dont mmap counters that
   are multiplexed

 - modify the perf record event loop to deal with buffer-less
   counters.

 - remove the -g option from perf sched record

 - account for unordered events in perf sched latency

 - (add -f to perf sched record to ease measurements)

 - skip idle threads (pid==0) in latency output

The result is better latency output by 'perf sched latency':

 -----------------------------------------------------------------------------------
  Task              |  Runtime ms | Switches | Average delay ms | Maximum delay ms |
 -----------------------------------------------------------------------------------
  ksoftirqd/8       |    0.071 ms |        2 | avg:    0.458 ms | max:    0.913 ms |
  at-spi-registry   |    0.609 ms |       19 | avg:    0.013 ms | max:    0.023 ms |
  perf              |    3.316 ms |       16 | avg:    0.013 ms | max:    0.054 ms |
  Xorg              |    0.392 ms |       19 | avg:    0.011 ms | max:    0.018 ms |
  sleep             |    0.537 ms |        2 | avg:    0.009 ms | max:    0.009 ms |
 -----------------------------------------------------------------------------------
  TOTAL:            |    4.925 ms |       58 |
 ---------------------------------------------

Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

tools/perf/builtin-record.c

@@ -49,6 +49,7 @@ static int			inherit_stat			= 0;
 static int			no_samples			= 0;
 static int			sample_address			= 0;
 static int			multiplex			= 0;
+static int			multiplex_fd			= -1;
 
 static long			samples;
 static struct timeval		last_read;
@@ -471,7 +472,15 @@ static void create_counter(int counter, int cpu, pid_t pid)
 	 */
 	if (group && group_fd == -1)
 		group_fd = fd[nr_cpu][counter];
+	if (multiplex && multiplex_fd == -1)
+		multiplex_fd = fd[nr_cpu][counter];
 
+	if (multiplex && fd[nr_cpu][counter] != multiplex_fd) {
+		int ret;
+
+		ret = ioctl(fd[nr_cpu][counter], PERF_COUNTER_IOC_SET_OUTPUT, multiplex_fd);
+		assert(ret != -1);
+	} else {
 		event_array[nr_poll].fd = fd[nr_cpu][counter];
 		event_array[nr_poll].events = POLLIN;
 		nr_poll++;
@@ -485,9 +494,7 @@ static void create_counter(int counter, int cpu, pid_t pid)
 			error("failed to mmap with %d (%s)\n", errno, strerror(errno));
 			exit(-1);
 		}
-
-	if (multiplex && fd[nr_cpu][counter] != group_fd)
-		ioctl(fd[nr_cpu][counter], PERF_COUNTER_IOC_SET_OUTPUT, group_fd);
+	}
 
 	ioctl(fd[nr_cpu][counter], PERF_COUNTER_IOC_ENABLE);
 }
@@ -618,9 +625,11 @@ static int __cmd_record(int argc, const char **argv)
 		int hits = samples;
 
 		for (i = 0; i < nr_cpu; i++) {
-			for (counter = 0; counter < nr_counters; counter++)
+			for (counter = 0; counter < nr_counters; counter++) {
+				if (mmap_array[i][counter].base)
 					mmap_read(&mmap_array[i][counter]);
 			}
+		}
 
 		if (hits == samples) {
 			if (done)

tools/perf/builtin-sched.c

@@ -116,6 +116,8 @@ static u64			sum_fluct;
 static u64			run_avg;
 
 static unsigned long		replay_repeat = 10;
+static unsigned long		nr_timestamps;
+static unsigned long		unordered_timestamps;
 
 #define TASK_STATE_TO_CHAR_STR "RSDTtZX"
 
@@ -1109,8 +1111,11 @@ latency_wakeup_event(struct trace_wakeup_event *wakeup_event,
 	if (atom->state != THREAD_SLEEPING)
 		return;
 
-	if (atom->sched_out_time > timestamp)
+	nr_timestamps++;
+	if (atom->sched_out_time > timestamp) {
+		unordered_timestamps++;
 		return;
+	}
 
 	atom->state = THREAD_WAIT_CPU;
 	atom->wake_up_time = timestamp;
@@ -1130,6 +1135,11 @@ static void output_lat_thread(struct task_atoms *atom_list)
 
 	if (!atom_list->nb_atoms)
 		return;
+	/*
+	 * Ignore idle threads:
+	 */
+	if (!atom_list->thread->pid)
+		return;
 
 	all_runtime += atom_list->total_runtime;
 	all_count += atom_list->nb_atoms;
@@ -1301,8 +1311,16 @@ static void __cmd_lat(void)
 	}
 
 	printf("-----------------------------------------------------------------------------------\n");
-	printf(" TOTAL:            |%9.3f ms |%9Ld |\n",
+	printf(" TOTAL:            |%9.3f ms |%9Ld |",
 		(double)all_runtime/1e6, all_count);
+
+	if (unordered_timestamps && nr_timestamps) {
+		printf(" INFO: %.2f%% unordered events.\n",
+			(double)unordered_timestamps/(double)nr_timestamps*100.0);
+	} else {
+		printf("\n");
+	}
+
 	printf("---------------------------------------------\n");
 }
 
@@ -1667,12 +1685,13 @@ static const char *record_args[] = {
 	"-a",
 	"-R",
 	"-M",
-	"-g",
+	"-f",
 	"-c", "1",
 	"-e", "sched:sched_switch:r",
 	"-e", "sched:sched_stat_wait:r",
 	"-e", "sched:sched_stat_sleep:r",
 	"-e", "sched:sched_stat_iowait:r",
+	"-e", "sched:sched_stat_runtime:r",
 	"-e", "sched:sched_process_exit:r",
 	"-e", "sched:sched_process_fork:r",
 	"-e", "sched:sched_wakeup:r",

tools/perf/util/trace-event-parse.c

@@ -2722,8 +2722,10 @@ void print_event(int cpu, void *data, int size, unsigned long long nsecs,
 	type = trace_parse_common_type(data);
 
 	event = trace_find_event(type);
-	if (!event)
-		die("ug! no event found for type %d", type);
+	if (!event) {
+		printf("ug! no event found for type %d\n", type);
+		return;
+	}
 
 	pid = parse_common_pid(data);