From d02fc6bcd53721cf8588633409157c232f2418e0 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@linux.intel.com>
Date: Tue, 3 Jan 2017 07:08:23 -0800
Subject: perf pmu: Factor out scale conversion code

Move the scale factor parsing code to an own function to reuse it in an
upcoming patch.

v2: Return error in case strdup returns NULL.

Signed-off-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: http://lkml.kernel.org/r/20170103150833.6694-2-andi@firstfloor.org
[ Keep returning -ENOMEM when strdup() fails in perf_pmu__parse_scale()/convert_scale() ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/pmu.c | 62 ++++++++++++++++++++++++++++-----------------------
 1 file changed, 34 insertions(+), 28 deletions(-)

(limited to 'tools/perf')

diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c
index dc6ccaa4e927..78b16100567d 100644
--- a/tools/perf/util/pmu.c
+++ b/tools/perf/util/pmu.c
@@ -94,32 +94,10 @@ static int pmu_format(const char *name, struct list_head *format)
 	return 0;
 }
 
-static int perf_pmu__parse_scale(struct perf_pmu_alias *alias, char *dir, char *name)
+static int convert_scale(const char *scale, char **end, double *sval)
 {
-	struct stat st;
-	ssize_t sret;
-	char scale[128];
-	int fd, ret = -1;
-	char path[PATH_MAX];
 	char *lc;
-
-	snprintf(path, PATH_MAX, "%s/%s.scale", dir, name);
-
-	fd = open(path, O_RDONLY);
-	if (fd == -1)
-		return -1;
-
-	if (fstat(fd, &st) < 0)
-		goto error;
-
-	sret = read(fd, scale, sizeof(scale)-1);
-	if (sret < 0)
-		goto error;
-
-	if (scale[sret - 1] == '\n')
-		scale[sret - 1] = '\0';
-	else
-		scale[sret] = '\0';
+	int ret = 0;
 
 	/*
 	 * save current locale
@@ -134,7 +112,7 @@ static int perf_pmu__parse_scale(struct perf_pmu_alias *alias, char *dir, char *
 	lc = strdup(lc);
 	if (!lc) {
 		ret = -ENOMEM;
-		goto error;
+		goto out;
 	}
 
 	/*
@@ -144,14 +122,42 @@ static int perf_pmu__parse_scale(struct perf_pmu_alias *alias, char *dir, char *
 	 */
 	setlocale(LC_NUMERIC, "C");
 
-	alias->scale = strtod(scale, NULL);
+	*sval = strtod(scale, end);
 
+out:
 	/* restore locale */
 	setlocale(LC_NUMERIC, lc);
-
 	free(lc);
+	return ret;
+}
+
+static int perf_pmu__parse_scale(struct perf_pmu_alias *alias, char *dir, char *name)
+{
+	struct stat st;
+	ssize_t sret;
+	char scale[128];
+	int fd, ret = -1;
+	char path[PATH_MAX];
+
+	snprintf(path, PATH_MAX, "%s/%s.scale", dir, name);
+
+	fd = open(path, O_RDONLY);
+	if (fd == -1)
+		return -1;
+
+	if (fstat(fd, &st) < 0)
+		goto error;
+
+	sret = read(fd, scale, sizeof(scale)-1);
+	if (sret < 0)
+		goto error;
+
+	if (scale[sret - 1] == '\n')
+		scale[sret - 1] = '\0';
+	else
+		scale[sret] = '\0';
 
-	ret = 0;
+	ret = convert_scale(scale, NULL, &alias->scale);
 error:
 	close(fd);
 	return ret;
-- 
cgit v1.2.3


From 2484c4c58fd1bc0d7711d294c530a1518da6a172 Mon Sep 17 00:00:00 2001
From: David Carrillo-Cisneros <davidcc@google.com>
Date: Thu, 12 Jan 2017 13:01:59 -0800
Subject: perf tools: Remove unneccessary feature-dwarf warning

Don't warn for feature-dwarf==0 if user explicitily disabled DWARF by
using NO_DWARF=1.

Signed-off-by: David Carrillo-Cisneros <davidcc@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/20170112210159.76143-1-davidcc@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Makefile.config | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

(limited to 'tools/perf')

diff --git a/tools/perf/Makefile.config b/tools/perf/Makefile.config
index 76c84f0eec52..03cf947755b9 100644
--- a/tools/perf/Makefile.config
+++ b/tools/perf/Makefile.config
@@ -291,8 +291,10 @@ else
       endif
     endif
     ifneq ($(feature-dwarf), 1)
-      msg := $(warning No libdw.h found or old libdw.h found or elfutils is older than 0.138, disables dwarf support. Please install new elfutils-devel/libdw-dev);
-      NO_DWARF := 1
+      ifndef NO_DWARF
+        msg := $(warning No libdw.h found or old libdw.h found or elfutils is older than 0.138, disables dwarf support. Please install new elfutils-devel/libdw-dev);
+        NO_DWARF := 1
+      endif
     else
       ifneq ($(feature-dwarf_getlocations), 1)
         msg := $(warning Old libdw.h, finding variables at given 'perf probe' point will not work, install elfutils-devel/libdw-dev >= 0.157);
-- 
cgit v1.2.3


From 5c64f99b1d030ce52acce53d83141d88f79e6362 Mon Sep 17 00:00:00 2001
From: Michael Petlan <mpetlan@redhat.com>
Date: Fri, 13 Jan 2017 10:06:52 -0500
Subject: perf script: Fix man page about --dump-raw-trace option

The "--dump-raw-script" is not a valid option, replace it with the valid
one, "--dump-raw-trace"

Signed-off-by: Michael Petlan <mpetlan@redhat.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 133dc4c39c57 ("perf: Rename 'perf trace' to 'perf script'")
LPU-Reference: 728644547.14560155.1484320012612.JavaMail.zimbra@redhat.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-script.txt | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'tools/perf')

diff --git a/tools/perf/Documentation/perf-script.txt b/tools/perf/Documentation/perf-script.txt
index 5dc5c6a09ac4..4ed5f239ba7d 100644
--- a/tools/perf/Documentation/perf-script.txt
+++ b/tools/perf/Documentation/perf-script.txt
@@ -36,7 +36,7 @@ There are several variants of perf script:
 
   'perf script report <script> [args]' to run and display the results
   of <script>.  <script> is the name displayed in the output of 'perf
-  trace --list' i.e. the actual script name minus any language
+  script --list' i.e. the actual script name minus any language
   extension.  The perf.data output from a previous run of 'perf script
   record <script>' is used and should be present for this command to
   succeed.  [args] refers to the (mainly optional) args expected by
@@ -76,7 +76,7 @@ OPTIONS
 	Any command you can specify in a shell.
 
 -D::
---dump-raw-script=::
+--dump-raw-trace=::
         Display verbose dump of the trace data.
 
 -L::
-- 
cgit v1.2.3


From be3d466c7356e574a2aa4e19dd20b19b3a9cc4fc Mon Sep 17 00:00:00 2001
From: Yannick Brosseau <scientist@fb.com>
Date: Fri, 13 Jan 2017 13:25:27 -0500
Subject: perf script: Also allow forcing reading of non-root owned files by
 root

In 2059fc7a5a9e ("perf symbols: Allow forcing reading of non-root owned
files by root") 'perf report' was added the option of forcing reading of
non-root owned symbol file.

This add the same behavior for perf script.

Reported-by: Mark Drayton <mbd@fb.com>
Signed-off-by: Yannick Brosseau <scientist@fb.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: kernel-team@fb.com
Link: http://lkml.kernel.org/r/20170113182527.18625-1-scientist@fb.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-script.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'tools/perf')

diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index 2f3ff69fc4e7..c0783b4f7b6c 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -2180,7 +2180,7 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
 		    "Show the mmap events"),
 	OPT_BOOLEAN('\0', "show-switch-events", &script.show_switch_events,
 		    "Show context switch events (if recorded)"),
-	OPT_BOOLEAN('f', "force", &file.force, "don't complain, do it"),
+	OPT_BOOLEAN('f', "force", &symbol_conf.force, "don't complain, do it"),
 	OPT_BOOLEAN(0, "ns", &nanosecs,
 		    "Use 9 decimal places when displaying time"),
 	OPT_CALLBACK_OPTARG(0, "itrace", &itrace_synth_opts, NULL, "opts",
@@ -2212,6 +2212,7 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
 			     PARSE_OPT_STOP_AT_NON_OPTION);
 
 	file.path = input_name;
+	file.force = symbol_conf.force;
 
 	if (argc > 1 && !strncmp(argv[0], "rec", strlen("rec"))) {
 		rec_script_path = get_script_path(argv[1], RECORD_SUFFIX);
-- 
cgit v1.2.3


From 941bdea79e194dec7e7b42780aa5620020548f8a Mon Sep 17 00:00:00 2001
From: Namhyung Kim <namhyung@kernel.org>
Date: Fri, 13 Jan 2017 19:45:21 +0900
Subject: perf sched timehist: Account thread wait time separately

Separate thread wait time into 3 parts - sleep, iowait and preempt based
on the prev_state of the last event.

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20170113104523.31212-1-namhyung@kernel.org
[ Fix the build on centos:5 where 'wait' shadows a global declaration ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-sched.c | 50 ++++++++++++++++++++++++++++++++++++++++------
 1 file changed, 44 insertions(+), 6 deletions(-)

(limited to 'tools/perf')

diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c
index 5b134b0d1ff3..6d3c3e84881a 100644
--- a/tools/perf/builtin-sched.c
+++ b/tools/perf/builtin-sched.c
@@ -77,6 +77,22 @@ struct sched_atom {
 
 #define TASK_STATE_TO_CHAR_STR "RSDTtZXxKWP"
 
+/* task state bitmask, copied from include/linux/sched.h */
+#define TASK_RUNNING		0
+#define TASK_INTERRUPTIBLE	1
+#define TASK_UNINTERRUPTIBLE	2
+#define __TASK_STOPPED		4
+#define __TASK_TRACED		8
+/* in tsk->exit_state */
+#define EXIT_DEAD		16
+#define EXIT_ZOMBIE		32
+#define EXIT_TRACE		(EXIT_ZOMBIE | EXIT_DEAD)
+/* in tsk->state again */
+#define TASK_DEAD		64
+#define TASK_WAKEKILL		128
+#define TASK_WAKING		256
+#define TASK_PARKED		512
+
 enum thread_state {
 	THREAD_SLEEPING = 0,
 	THREAD_WAIT_CPU,
@@ -216,13 +232,16 @@ struct perf_sched {
 struct thread_runtime {
 	u64 last_time;      /* time of previous sched in/out event */
 	u64 dt_run;         /* run time */
-	u64 dt_wait;        /* time between CPU access (off cpu) */
+	u64 dt_sleep;       /* time between CPU access by sleep (off cpu) */
+	u64 dt_iowait;      /* time between CPU access by iowait (off cpu) */
+	u64 dt_preempt;     /* time between CPU access by preempt (off cpu) */
 	u64 dt_delay;       /* time between wakeup and sched-in */
 	u64 ready_to_run;   /* time of wakeup */
 
 	struct stats run_stats;
 	u64 total_run_time;
 
+	int last_state;
 	u64 migrations;
 };
 
@@ -1858,6 +1877,7 @@ static void timehist_print_sample(struct perf_sched *sched,
 	struct thread_runtime *tr = thread__priv(thread);
 	u32 max_cpus = sched->max_cpu + 1;
 	char tstr[64];
+	u64 wait_time;
 
 	timestamp__scnprintf_usec(t, tstr, sizeof(tstr));
 	printf("%15s [%04d] ", tstr, sample->cpu);
@@ -1880,7 +1900,9 @@ static void timehist_print_sample(struct perf_sched *sched,
 
 	printf(" %-*s ", comm_width, timehist_get_commstr(thread));
 
-	print_sched_time(tr->dt_wait, 6);
+	wait_time = tr->dt_sleep + tr->dt_iowait + tr->dt_preempt;
+	print_sched_time(wait_time, 6);
+
 	print_sched_time(tr->dt_delay, 6);
 	print_sched_time(tr->dt_run, 6);
 
@@ -1930,8 +1952,11 @@ static void timehist_update_runtime_stats(struct thread_runtime *r,
 					 u64 t, u64 tprev)
 {
 	r->dt_delay   = 0;
-	r->dt_wait    = 0;
+	r->dt_sleep   = 0;
+	r->dt_iowait  = 0;
+	r->dt_preempt = 0;
 	r->dt_run     = 0;
+
 	if (tprev) {
 		r->dt_run = t - tprev;
 		if (r->ready_to_run) {
@@ -1943,8 +1968,16 @@ static void timehist_update_runtime_stats(struct thread_runtime *r,
 
 		if (r->last_time > tprev)
 			pr_debug("time travel: last sched out time for task > previous sched_switch event\n");
-		else if (r->last_time)
-			r->dt_wait = tprev - r->last_time;
+		else if (r->last_time) {
+			u64 dt_wait = tprev - r->last_time;
+
+			if (r->last_state == TASK_RUNNING)
+				r->dt_preempt = dt_wait;
+			else if (r->last_state == TASK_UNINTERRUPTIBLE)
+				r->dt_iowait = dt_wait;
+			else
+				r->dt_sleep = dt_wait;
+		}
 	}
 
 	update_stats(&r->run_stats, r->dt_run);
@@ -2447,8 +2480,10 @@ static int timehist_sched_change_event(struct perf_tool *tool,
 			 * time.  we only care total run time and run stat.
 			 */
 			last_tr->dt_run = 0;
-			last_tr->dt_wait = 0;
 			last_tr->dt_delay = 0;
+			last_tr->dt_sleep = 0;
+			last_tr->dt_iowait = 0;
+			last_tr->dt_preempt = 0;
 
 			if (itr->cursor.nr)
 				callchain_append(&itr->callchain, &itr->cursor, t - tprev);
@@ -2470,6 +2505,9 @@ out:
 		/* time of this sched_switch event becomes last time task seen */
 		tr->last_time = sample->time;
 
+		/* last state is used to determine where to account wait time */
+		tr->last_state = perf_evsel__intval(evsel, sample, "prev_state");
+
 		/* sched out event for task so reset ready to run time */
 		tr->ready_to_run = 0;
 	}
-- 
cgit v1.2.3


From 414e050c68ec2a3dd815544ea48ff6016f1a7a11 Mon Sep 17 00:00:00 2001
From: Namhyung Kim <namhyung@kernel.org>
Date: Fri, 13 Jan 2017 19:45:22 +0900
Subject: perf sched timehist: Add --state option

The --state option is to show task state when switched out.  The state
is printed as a single character like in the /proc but I added 'I' for
idle state rather than 'R'.

  $ perf sched timehist --state | head
  Samples do not have callchains.
      time cpu task name              wait time sch delay run time state
               [tid/pid]                 (msec)    (msec)   (msec)
  -------- --- ----------------------- -------- ------------------ -----
  1.753791 [3] <idle>                     0.000     0.000    0.000     I
  1.753834 [1] perf[27469]                0.000     0.000    0.000     S
  1.753904 [3] perf[27470]                0.000     0.006    0.112     S
  1.753914 [1] <idle>                     0.000     0.000    0.079     I
  1.753915 [3] migration/3[23]            0.000     0.002    0.011     S
  1.754287 [2] <idle>                     0.000     0.000    0.000     I
  1.754335 [2] transmission[1773/1739]    0.000     0.004    0.047     S

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20170113104523.31212-2-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-sched.txt |  2 ++
 tools/perf/builtin-sched.c              | 38 +++++++++++++++++++++++++++++----
 2 files changed, 36 insertions(+), 4 deletions(-)

(limited to 'tools/perf')

diff --git a/tools/perf/Documentation/perf-sched.txt b/tools/perf/Documentation/perf-sched.txt
index 76173969ab80..d33deddb0146 100644
--- a/tools/perf/Documentation/perf-sched.txt
+++ b/tools/perf/Documentation/perf-sched.txt
@@ -143,6 +143,8 @@ OPTIONS for 'perf sched timehist'
 	stop time is not given (i.e, time string is 'x.y,') then analysis goes
 	to end of file.
 
+--state::
+	Show task state when it switched out.
 
 SEE ALSO
 --------
diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c
index 6d3c3e84881a..a8ac76602187 100644
--- a/tools/perf/builtin-sched.c
+++ b/tools/perf/builtin-sched.c
@@ -222,6 +222,7 @@ struct perf_sched {
 	bool		show_cpu_visual;
 	bool		show_wakeups;
 	bool		show_migrations;
+	bool		show_state;
 	u64		skipped_samples;
 	const char	*time_str;
 	struct perf_time_interval ptime;
@@ -1840,6 +1841,9 @@ static void timehist_header(struct perf_sched *sched)
 	printf(" %-*s  %9s  %9s  %9s", comm_width,
 		"task name", "wait time", "sch delay", "run time");
 
+	if (sched->show_state)
+		printf("  %s", "state");
+
 	printf("\n");
 
 	/*
@@ -1850,9 +1854,14 @@ static void timehist_header(struct perf_sched *sched)
 	if (sched->show_cpu_visual)
 		printf(" %*s ", ncpus, "");
 
-	printf(" %-*s  %9s  %9s  %9s\n", comm_width,
+	printf(" %-*s  %9s  %9s  %9s", comm_width,
 	       "[tid/pid]", "(msec)", "(msec)", "(msec)");
 
+	if (sched->show_state)
+		printf("  %5s", "");
+
+	printf("\n");
+
 	/*
 	 * separator
 	 */
@@ -1865,14 +1874,29 @@ static void timehist_header(struct perf_sched *sched)
 		graph_dotted_line, graph_dotted_line, graph_dotted_line,
 		graph_dotted_line);
 
+	if (sched->show_state)
+		printf("  %.5s", graph_dotted_line);
+
 	printf("\n");
 }
 
+static char task_state_char(struct thread *thread, int state)
+{
+	static const char state_to_char[] = TASK_STATE_TO_CHAR_STR;
+	unsigned bit = state ? ffs(state) : 0;
+
+	/* 'I' for idle */
+	if (thread->tid == 0)
+		return 'I';
+
+	return bit < sizeof(state_to_char) - 1 ? state_to_char[bit] : '?';
+}
+
 static void timehist_print_sample(struct perf_sched *sched,
 				  struct perf_sample *sample,
 				  struct addr_location *al,
 				  struct thread *thread,
-				  u64 t)
+				  u64 t, int state)
 {
 	struct thread_runtime *tr = thread__priv(thread);
 	u32 max_cpus = sched->max_cpu + 1;
@@ -1906,6 +1930,9 @@ static void timehist_print_sample(struct perf_sched *sched,
 	print_sched_time(tr->dt_delay, 6);
 	print_sched_time(tr->dt_run, 6);
 
+	if (sched->show_state)
+		printf(" %5c ", task_state_char(thread, state));
+
 	if (sched->show_wakeups)
 		printf("  %-*s", comm_width, "");
 
@@ -2406,6 +2433,8 @@ static int timehist_sched_change_event(struct perf_tool *tool,
 	struct thread_runtime *tr = NULL;
 	u64 tprev, t = sample->time;
 	int rc = 0;
+	int state = perf_evsel__intval(evsel, sample, "prev_state");
+
 
 	if (machine__resolve(machine, &al, sample) < 0) {
 		pr_err("problem processing %d event. skipping it\n",
@@ -2493,7 +2522,7 @@ static int timehist_sched_change_event(struct perf_tool *tool,
 	}
 
 	if (!sched->summary_only)
-		timehist_print_sample(sched, sample, &al, thread, t);
+		timehist_print_sample(sched, sample, &al, thread, t, state);
 
 out:
 	if (sched->hist_time.start == 0 && t >= ptime->start)
@@ -2506,7 +2535,7 @@ out:
 		tr->last_time = sample->time;
 
 		/* last state is used to determine where to account wait time */
-		tr->last_state = perf_evsel__intval(evsel, sample, "prev_state");
+		tr->last_state = state;
 
 		/* sched out event for task so reset ready to run time */
 		tr->ready_to_run = 0;
@@ -3278,6 +3307,7 @@ int cmd_sched(int argc, const char **argv, const char *prefix __maybe_unused)
 	OPT_BOOLEAN('I', "idle-hist", &sched.idle_hist, "Show idle events only"),
 	OPT_STRING(0, "time", &sched.time_str, "str",
 		   "Time span for analysis (start,stop)"),
+	OPT_BOOLEAN(0, "state", &sched.show_state, "Show task state when sched-out"),
 	OPT_PARENT(sched_options)
 	};
 
-- 
cgit v1.2.3


From 587782c52a83b35673201fd9a54364fa2b189b33 Mon Sep 17 00:00:00 2001
From: Namhyung Kim <namhyung@kernel.org>
Date: Fri, 13 Jan 2017 19:45:23 +0900
Subject: perf sched timehist: Show total wait times for summary

When --state option is given, the summary will show total run, sleep,
iowait, preempt and delay time instead of statistics of runtime.

  $ perf sched timehist -s --state

  Wait-time summary
            comm  parent sched-in run-time  sleep iowait preempt delay
                          (count)   (msec) (msec) (msec)  (msec) (msec)
  ---------------------------------------------------------------------
       systemd[1]      0        3    0.497   1.685 0.000   0.000 0.061
   ksoftirqd/0[3]      2       21    0.434 989.948 0.000   0.000 0.325
   rcu_preempt[7]      2       28    0.386 993.211 0.000   0.000 0.712
  migration/0[10]      2       12    0.126  50.174 0.000   0.000 0.044
   watchdog/0[11]      2        1    0.009   0.000 0.000   0.000 0.016
  migration/1[13]      2        2    0.029  11.755 0.000   0.000 0.007
  <SNIP>

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20170113104523.31212-3-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-sched.c | 44 +++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 41 insertions(+), 3 deletions(-)

(limited to 'tools/perf')

diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c
index a8ac76602187..daceb3202200 100644
--- a/tools/perf/builtin-sched.c
+++ b/tools/perf/builtin-sched.c
@@ -241,6 +241,10 @@ struct thread_runtime {
 
 	struct stats run_stats;
 	u64 total_run_time;
+	u64 total_sleep_time;
+	u64 total_iowait_time;
+	u64 total_preempt_time;
+	u64 total_delay_time;
 
 	int last_state;
 	u64 migrations;
@@ -2008,7 +2012,12 @@ static void timehist_update_runtime_stats(struct thread_runtime *r,
 	}
 
 	update_stats(&r->run_stats, r->dt_run);
-	r->total_run_time += r->dt_run;
+
+	r->total_run_time     += r->dt_run;
+	r->total_delay_time   += r->dt_delay;
+	r->total_sleep_time   += r->dt_sleep;
+	r->total_iowait_time  += r->dt_iowait;
+	r->total_preempt_time += r->dt_preempt;
 }
 
 static bool is_idle_sample(struct perf_sample *sample,
@@ -2593,7 +2602,26 @@ static void print_thread_runtime(struct thread *t,
 	printf("\n");
 }
 
+static void print_thread_waittime(struct thread *t,
+				  struct thread_runtime *r)
+{
+	printf("%*s   %5d  %9" PRIu64 " ",
+	       comm_width, timehist_get_commstr(t), t->ppid,
+	       (u64) r->run_stats.n);
+
+	print_sched_time(r->total_run_time, 8);
+	print_sched_time(r->total_sleep_time, 6);
+	printf(" ");
+	print_sched_time(r->total_iowait_time, 6);
+	printf(" ");
+	print_sched_time(r->total_preempt_time, 6);
+	printf(" ");
+	print_sched_time(r->total_delay_time, 6);
+	printf("\n");
+}
+
 struct total_run_stats {
+	struct perf_sched *sched;
 	u64  sched_count;
 	u64  task_count;
 	u64  total_run_time;
@@ -2612,7 +2640,11 @@ static int __show_thread_runtime(struct thread *t, void *priv)
 		stats->task_count++;
 		stats->sched_count += r->run_stats.n;
 		stats->total_run_time += r->total_run_time;
-		print_thread_runtime(t, r);
+
+		if (stats->sched->show_state)
+			print_thread_waittime(t, r);
+		else
+			print_thread_runtime(t, r);
 	}
 
 	return 0;
@@ -2700,18 +2732,24 @@ static void timehist_print_summary(struct perf_sched *sched,
 	u64 hist_time = sched->hist_time.end - sched->hist_time.start;
 
 	memset(&totals, 0, sizeof(totals));
+	totals.sched = sched;
 
 	if (sched->idle_hist) {
 		printf("\nIdle-time summary\n");
 		printf("%*s  parent  sched-out  ", comm_width, "comm");
 		printf("  idle-time   min-idle    avg-idle    max-idle  stddev  migrations\n");
+	} else if (sched->show_state) {
+		printf("\nWait-time summary\n");
+		printf("%*s  parent   sched-in  ", comm_width, "comm");
+		printf("   run-time      sleep      iowait     preempt       delay\n");
 	} else {
 		printf("\nRuntime summary\n");
 		printf("%*s  parent   sched-in  ", comm_width, "comm");
 		printf("   run-time    min-run     avg-run     max-run  stddev  migrations\n");
 	}
 	printf("%*s            (count)  ", comm_width, "");
-	printf("     (msec)     (msec)      (msec)      (msec)       %%\n");
+	printf("     (msec)     (msec)      (msec)      (msec)       %s\n",
+	       sched->show_state ? "(msec)" : "%");
 	printf("%.117s\n", graph_dotted_line);
 
 	machine__for_each_thread(m, show_thread_runtime, &totals);
-- 
cgit v1.2.3


From d25ed5d9fad98fd36ffca0bb5c3a0761a11e3f3c Mon Sep 17 00:00:00 2001
From: Soramichi AKIYAMA <akiyama@m.soramichi.jp>
Date: Tue, 17 Jan 2017 00:22:37 +0900
Subject: perf tools: Move two variables usied in libperf from perf.c

The use_browser and perf_version_string variables are both declared in
perf.c but they are also referenced by other functions of libperf.a.

Therefore a user linking an own main() with libperf.a must declare those
two variables in their files even if the files never use the browser or
the version information.

This patch fixes this issue by moving use_browser and
perf_version_string out of perf.c to some other files.

Signed-off-by: Soramichi Akiyama <akiyama@m.soramichi.jp>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20170117002237.c1aec0ce3b4d675dca018deb@m.soramichi.jp
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Build         | 3 +--
 tools/perf/perf.c        | 3 ---
 tools/perf/ui/setup.c    | 1 +
 tools/perf/util/Build    | 1 +
 tools/perf/util/header.c | 2 ++
 5 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'tools/perf')

diff --git a/tools/perf/Build b/tools/perf/Build
index 0b48806f93c2..7039ecb89170 100644
--- a/tools/perf/Build
+++ b/tools/perf/Build
@@ -40,8 +40,7 @@ CFLAGS_builtin-help.o      += $(paths)
 CFLAGS_builtin-timechart.o += $(paths)
 CFLAGS_perf.o              += -DPERF_HTML_PATH="BUILD_STR($(htmldir_SQ))"	\
 			      -DPERF_EXEC_PATH="BUILD_STR($(perfexecdir_SQ))"	\
-			      -DPREFIX="BUILD_STR($(prefix_SQ))"		\
-			      -include $(OUTPUT)PERF-VERSION-FILE
+			      -DPREFIX="BUILD_STR($(prefix_SQ))"
 CFLAGS_builtin-trace.o	   += -DSTRACE_GROUPS_DIR="BUILD_STR($(STRACE_GROUPS_DIR_SQ))"
 CFLAGS_builtin-report.o	   += -DTIPDIR="BUILD_STR($(tipdir_SQ))"
 CFLAGS_builtin-report.o	   += -DDOCDIR="BUILD_STR($(srcdir_SQ)/Documentation)"
diff --git a/tools/perf/perf.c b/tools/perf/perf.c
index 13c8a7db055e..0324cd15e42a 100644
--- a/tools/perf/perf.c
+++ b/tools/perf/perf.c
@@ -29,7 +29,6 @@ const char perf_usage_string[] =
 const char perf_more_info_string[] =
 	"See 'perf help COMMAND' for more information on a specific command.";
 
-int use_browser = -1;
 static int use_pager = -1;
 const char *input_name;
 
@@ -330,8 +329,6 @@ static int handle_alias(int *argcp, const char ***argv)
 	return ret;
 }
 
-const char perf_version_string[] = PERF_VERSION;
-
 #define RUN_SETUP	(1<<0)
 #define USE_PAGER	(1<<1)
 
diff --git a/tools/perf/ui/setup.c b/tools/perf/ui/setup.c
index 1f6b0994f4f4..50d13e58210f 100644
--- a/tools/perf/ui/setup.c
+++ b/tools/perf/ui/setup.c
@@ -7,6 +7,7 @@
 
 pthread_mutex_t ui__lock = PTHREAD_MUTEX_INITIALIZER;
 void *perf_gtk_handle;
+int use_browser = -1;
 
 #ifdef HAVE_GTK2_SUPPORT
 static int setup_gtk_browser(void)
diff --git a/tools/perf/util/Build b/tools/perf/util/Build
index 3840e3a87057..5da376bc1afc 100644
--- a/tools/perf/util/Build
+++ b/tools/perf/util/Build
@@ -162,6 +162,7 @@ CFLAGS_rbtree.o        += -Wno-unused-parameter -DETC_PERFCONFIG="BUILD_STR($(ET
 CFLAGS_libstring.o     += -Wno-unused-parameter -DETC_PERFCONFIG="BUILD_STR($(ETC_PERFCONFIG_SQ))"
 CFLAGS_hweight.o       += -Wno-unused-parameter -DETC_PERFCONFIG="BUILD_STR($(ETC_PERFCONFIG_SQ))"
 CFLAGS_parse-events.o  += -Wno-redundant-decls
+CFLAGS_header.o        += -include $(OUTPUT)PERF-VERSION-FILE
 
 $(OUTPUT)util/kallsyms.o: ../lib/symbol/kallsyms.c FORCE
 	$(call rule_mkdir)
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index d89c9c7ef4e5..00fd8a8850d3 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -41,6 +41,8 @@ static const u64 __perf_magic2_sw = 0x50455246494c4532ULL;
 
 #define PERF_MAGIC	__perf_magic2
 
+const char perf_version_string[] = PERF_VERSION;
+
 struct perf_file_attr {
 	struct perf_event_attr	attr;
 	struct perf_file_section	ids;
-- 
cgit v1.2.3


From d94386f28abad0c5879f0760712e34e71f88a7da Mon Sep 17 00:00:00 2001
From: Soramichi AKIYAMA <akiyama@m.soramichi.jp>
Date: Tue, 17 Jan 2017 22:22:33 +0900
Subject: perf evlist: Fix typo in deliver_sample()

This patch fixes a typo: s/delievery/delivery/

Signed-off-by: Soramichi Akiyama <akiyama@m.soramichi.jp>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20170117222233.dfd92de0ad701e7c53396950@m.soramichi.jp
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/session.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'tools/perf')

diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index f268201048a0..349c68144e55 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -1191,7 +1191,7 @@ static int
 	u64 sample_type = evsel->attr.sample_type;
 	u64 read_format = evsel->attr.read_format;
 
-	/* Standard sample delievery. */
+	/* Standard sample delivery. */
 	if (!(sample_type & PERF_SAMPLE_READ))
 		return tool->sample(tool, event, sample, evsel, machine);
 
-- 
cgit v1.2.3