[GIT PULL 0/9] perf/core fixes and improvements

[GIT PULL 0/9] perf/core fixes and improvements

[Arnaldo Carvalho de Melo]

Hi Ingo,

        Please consider pulling from:

git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux-2.6 perf/core

Regards,

- Arnaldo

Arnaldo Carvalho de Melo (3):
  perf evlist: Remove dependency on debug routines
  perf python: Use exception to propagate errors
  perf evlist: Don't die if sample_{id_all|type} is invalid

David Ahern (4):
  perf stat: clarify unsupported events from uncounted events
  perf script: "sym" field really means show IP data
  perf script: Make printing of dso a separate field option
  perf script: Add printing of sample address

Frederic Weisbecker (2):
  perf python: Fix argument name list of read_on_cpu()
  perf python: Cleanup useless double NULL termination in method arg names

 tools/perf/Documentation/perf-script.txt |   12 ++--
 tools/perf/builtin-script.c              |  108 +++++++++++++++++++++++++++---
 tools/perf/builtin-stat.c                |    9 ++-
 tools/perf/builtin-test.c                |    2 +-
 tools/perf/util/event.c                  |   16 -----
 tools/perf/util/event.h                  |    2 -
 tools/perf/util/evlist.c                 |   68 +++++++++++--------
 tools/perf/util/evlist.h                 |    6 +-
 tools/perf/util/evsel.c                  |   17 +++++
 tools/perf/util/evsel.h                  |    8 ++
 tools/perf/util/python.c                 |   29 ++++----
 tools/perf/util/session.c                |   73 ++++++++++++++-------
 tools/perf/util/session.h                |    5 +-
 13 files changed, 245 insertions(+), 110 deletions(-)

[PATCH 1/9] perf evlist: Remove dependency on debug routines

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

So far we avoided having to link debug.o in the python binding, keep it
that way by not using ui__warning() in evlist.c.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-4wtew8hd3g7ejnlehtspys2t@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evlist.c |   13 ++++---------
 1 files changed, 4 insertions(+), 9 deletions(-)

diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index 50aa348..04c8a60 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -12,7 +12,6 @@
 #include "evlist.h"
 #include "evsel.h"
 #include "util.h"
-#include "debug.h"
 
 #include <sys/mman.h>
 
@@ -257,19 +256,15 @@ int perf_evlist__alloc_mmap(struct perf_evlist *evlist)
 	return evlist->mmap != NULL ? 0 : -ENOMEM;
 }
 
-static int __perf_evlist__mmap(struct perf_evlist *evlist, struct perf_evsel *evsel,
+static int __perf_evlist__mmap(struct perf_evlist *evlist,
 			       int idx, int prot, int mask, int fd)
 {
 	evlist->mmap[idx].prev = 0;
 	evlist->mmap[idx].mask = mask;
 	evlist->mmap[idx].base = mmap(NULL, evlist->mmap_len, prot,
 				      MAP_SHARED, fd, 0);
-	if (evlist->mmap[idx].base == MAP_FAILED) {
-		if (evlist->cpus->map[idx] == -1 && evsel->attr.inherit)
-			ui__warning("Inherit is not allowed on per-task "
-				    "events using mmap.\n");
+	if (evlist->mmap[idx].base == MAP_FAILED)
 		return -1;
-	}
 
 	perf_evlist__add_pollfd(evlist, fd);
 	return 0;
@@ -289,7 +284,7 @@ static int perf_evlist__mmap_per_cpu(struct perf_evlist *evlist, int prot, int m
 
 				if (output == -1) {
 					output = fd;
-					if (__perf_evlist__mmap(evlist, evsel, cpu,
+					if (__perf_evlist__mmap(evlist, cpu,
 								prot, mask, output) < 0)
 						goto out_unmap;
 				} else {
@@ -329,7 +324,7 @@ static int perf_evlist__mmap_per_thread(struct perf_evlist *evlist, int prot, in
 
 			if (output == -1) {
 				output = fd;
-				if (__perf_evlist__mmap(evlist, evsel, thread,
+				if (__perf_evlist__mmap(evlist, thread,
 							prot, mask, output) < 0)
 					goto out_unmap;
 			} else {
-- 
1.6.2.5

[PATCH 2/9] perf python: Use exception to propagate errors

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

We were using pr_debug to tell the user about not being able to parse a sample
where we should really use the python way of reporting errors: exceptions.

Fixes this problem:

[root@emilia ~]# python
>>> import perf
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ImportError: /home/acme/git/build/perf/python/perf.so: undefined symbol: eprintf
>>>
[root@emilia ~]

As we want to keep the objects linked in the python binding (and in the future
in a shared library) minimal.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-m9dba9kaluas0kq8r58z191c@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/python.c |   10 ++++------
 1 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/tools/perf/util/python.c b/tools/perf/util/python.c
index 69436b3..2dd1698 100644
--- a/tools/perf/util/python.c
+++ b/tools/perf/util/python.c
@@ -694,14 +694,12 @@ static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist,
 		err = perf_event__parse_sample(event, first->attr.sample_type,
 					       perf_sample_size(first->attr.sample_type),
 					       sample_id_all, &pevent->sample);
-		if (err) {
-			pr_err("Can't parse sample, err = %d\n", err);
-			goto end;
-		}
-
+		if (err)
+			return PyErr_Format(PyExc_OSError,
+					    "perf: can't parse sample, err=%d", err);
 		return pyevent;
 	}
-end:
+
 	Py_INCREF(Py_None);
 	return Py_None;
 }
-- 
1.6.2.5

[PATCH 3/9] perf evlist: Don't die if sample_{id_all|type} is invalid

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

Fixes two more cases where the python binding would not load:

. Not finding die(), which it shouldn't anyway, not good to just stop the
  world because some particular perf.data file is invalid, just propagate
  the error to the caller.

. Not finding perf_sample_size: fix it by moving it from event.c to evsel,
  where it belongs, as most cases are moving to operate on an evsel object.o

One of the fixed problems:

[root@emilia ~]# python
>>> import perf
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ImportError: /home/acme/git/build/perf/python/perf.so: undefined symbol: perf_sample_size
>>>
[root@emilia ~]#

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-1hkj7b2cvgbfnoizsekjb6c9@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-test.c |    2 +-
 tools/perf/util/event.c   |   16 -------------
 tools/perf/util/event.h   |    2 -
 tools/perf/util/evlist.c  |   55 +++++++++++++++++++++++++++-----------------
 tools/perf/util/evlist.h  |    6 +++-
 tools/perf/util/evsel.c   |   16 +++++++++++++
 tools/perf/util/evsel.h   |    7 +++++
 tools/perf/util/python.c  |    2 +-
 tools/perf/util/session.c |   12 +++++++++-
 9 files changed, 74 insertions(+), 44 deletions(-)

diff --git a/tools/perf/builtin-test.c b/tools/perf/builtin-test.c
index b671862..2da9162 100644
--- a/tools/perf/builtin-test.c
+++ b/tools/perf/builtin-test.c
@@ -474,7 +474,7 @@ static int test__basic_mmap(void)
 	unsigned int nr_events[nsyscalls],
 		     expected_nr_events[nsyscalls], i, j;
 	struct perf_evsel *evsels[nsyscalls], *evsel;
-	int sample_size = perf_sample_size(attr.sample_type);
+	int sample_size = __perf_evsel__sample_size(attr.sample_type);
 
 	for (i = 0; i < nsyscalls; ++i) {
 		char name[64];
diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c
index 0fe9adf..3c1b8a6 100644
--- a/tools/perf/util/event.c
+++ b/tools/perf/util/event.c
@@ -35,22 +35,6 @@ const char *perf_event__name(unsigned int id)
 	return perf_event__names[id];
 }
 
-int perf_sample_size(u64 sample_type)
-{
-	u64 mask = sample_type & PERF_SAMPLE_MASK;
-	int size = 0;
-	int i;
-
-	for (i = 0; i < 64; i++) {
-		if (mask & (1ULL << i))
-			size++;
-	}
-
-	size *= sizeof(u64);
-
-	return size;
-}
-
 static struct perf_sample synth_sample = {
 	.pid	   = -1,
 	.tid	   = -1,
diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h
index c083328..1d7f664 100644
--- a/tools/perf/util/event.h
+++ b/tools/perf/util/event.h
@@ -82,8 +82,6 @@ struct perf_sample {
 	struct ip_callchain *callchain;
 };
 
-int perf_sample_size(u64 sample_type);
-
 #define BUILD_ID_SIZE 20
 
 struct build_id_event {
diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index 04c8a60..b021ea9 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -455,33 +455,46 @@ int perf_evlist__set_filters(struct perf_evlist *evlist)
 	return 0;
 }
 
-u64 perf_evlist__sample_type(struct perf_evlist *evlist)
+bool perf_evlist__valid_sample_type(const struct perf_evlist *evlist)
 {
-	struct perf_evsel *pos;
-	u64 type = 0;
-
-	list_for_each_entry(pos, &evlist->entries, node) {
-		if (!type)
-			type = pos->attr.sample_type;
-		else if (type != pos->attr.sample_type)
-			die("non matching sample_type");
+	struct perf_evsel *pos, *first;
+
+	pos = first = list_entry(evlist->entries.next, struct perf_evsel, node);
+
+	list_for_each_entry_continue(pos, &evlist->entries, node) {
+		if (first->attr.sample_type != pos->attr.sample_type)
+			return false;
 	}
 
-	return type;
+	return true;
 }
 
-bool perf_evlist__sample_id_all(const struct perf_evlist *evlist)
+u64 perf_evlist__sample_type(const struct perf_evlist *evlist)
+{
+	struct perf_evsel *first;
+
+	first = list_entry(evlist->entries.next, struct perf_evsel, node);
+	return first->attr.sample_type;
+}
+
+bool perf_evlist__valid_sample_id_all(const struct perf_evlist *evlist)
 {
-	bool value = false, first = true;
-	struct perf_evsel *pos;
-
-	list_for_each_entry(pos, &evlist->entries, node) {
-		if (first) {
-			value = pos->attr.sample_id_all;
-			first = false;
-		} else if (value != pos->attr.sample_id_all)
-			die("non matching sample_id_all");
+	struct perf_evsel *pos, *first;
+
+	pos = first = list_entry(evlist->entries.next, struct perf_evsel, node);
+
+	list_for_each_entry_continue(pos, &evlist->entries, node) {
+		if (first->attr.sample_id_all != pos->attr.sample_id_all)
+			return false;
 	}
 
-	return value;
+	return true;
+}
+
+bool perf_evlist__sample_id_all(const struct perf_evlist *evlist)
+{
+	struct perf_evsel *first;
+
+	first = list_entry(evlist->entries.next, struct perf_evsel, node);
+	return first->attr.sample_id_all;
 }
diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h
index 0a1ef1f..b2b8623 100644
--- a/tools/perf/util/evlist.h
+++ b/tools/perf/util/evlist.h
@@ -66,7 +66,9 @@ int perf_evlist__create_maps(struct perf_evlist *evlist, pid_t target_pid,
 void perf_evlist__delete_maps(struct perf_evlist *evlist);
 int perf_evlist__set_filters(struct perf_evlist *evlist);
 
-u64 perf_evlist__sample_type(struct perf_evlist *evlist);
-bool perf_evlist__sample_id_all(const struct perf_evlist *evlist);
+u64 perf_evlist__sample_type(const struct perf_evlist *evlist);
+bool perf_evlist__sample_id_all(const const struct perf_evlist *evlist);
 
+bool perf_evlist__valid_sample_type(const struct perf_evlist *evlist);
+bool perf_evlist__valid_sample_id_all(const struct perf_evlist *evlist);
 #endif /* __PERF_EVLIST_H */
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index cca29ed..0239eb8 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -15,6 +15,22 @@
 
 #define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y))
 
+int __perf_evsel__sample_size(u64 sample_type)
+{
+	u64 mask = sample_type & PERF_SAMPLE_MASK;
+	int size = 0;
+	int i;
+
+	for (i = 0; i < 64; i++) {
+		if (mask & (1ULL << i))
+			size++;
+	}
+
+	size *= sizeof(u64);
+
+	return size;
+}
+
 void perf_evsel__init(struct perf_evsel *evsel,
 		      struct perf_event_attr *attr, int idx)
 {
diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
index f79bb2c..7e9366e 100644
--- a/tools/perf/util/evsel.h
+++ b/tools/perf/util/evsel.h
@@ -149,4 +149,11 @@ static inline int perf_evsel__read_scaled(struct perf_evsel *evsel,
 	return __perf_evsel__read(evsel, ncpus, nthreads, true);
 }
 
+int __perf_evsel__sample_size(u64 sample_type);
+
+static inline int perf_evsel__sample_size(struct perf_evsel *evsel)
+{
+	return __perf_evsel__sample_size(evsel->attr.sample_type);
+}
+
 #endif /* __PERF_EVSEL_H */
diff --git a/tools/perf/util/python.c b/tools/perf/util/python.c
index 2dd1698..24063b4 100644
--- a/tools/perf/util/python.c
+++ b/tools/perf/util/python.c
@@ -692,7 +692,7 @@ static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist,
 
 		first = list_entry(evlist->entries.next, struct perf_evsel, node);
 		err = perf_event__parse_sample(event, first->attr.sample_type,
-					       perf_sample_size(first->attr.sample_type),
+					       perf_evsel__sample_size(first),
 					       sample_id_all, &pevent->sample);
 		if (err)
 			return PyErr_Format(PyExc_OSError,
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index 64500fc..f5a8fbd 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -58,6 +58,16 @@ static int perf_session__open(struct perf_session *self, bool force)
 		goto out_close;
 	}
 
+	if (!perf_evlist__valid_sample_type(self->evlist)) {
+		pr_err("non matching sample_type");
+		goto out_close;
+	}
+
+	if (!perf_evlist__valid_sample_id_all(self->evlist)) {
+		pr_err("non matching sample_id_all");
+		goto out_close;
+	}
+
 	self->size = input_stat.st_size;
 	return 0;
 
@@ -97,7 +107,7 @@ out:
 void perf_session__update_sample_type(struct perf_session *self)
 {
 	self->sample_type = perf_evlist__sample_type(self->evlist);
-	self->sample_size = perf_sample_size(self->sample_type);
+	self->sample_size = __perf_evsel__sample_size(self->sample_type);
 	self->sample_id_all = perf_evlist__sample_id_all(self->evlist);
 	perf_session__id_header_size(self);
 }
-- 
1.6.2.5

[PATCH 4/9] perf python: Fix argument name list of read_on_cpu()

[Arnaldo Carvalho de Melo]

From: Frederic Weisbecker <fweisbec@gmail.com>

Mandatory arguments need to be present in the argument name list, as
well as optional arguments, otherwise python barfs:

	# ./python/twatch.py
	Traceback (most recent call last):
	  File "./python/twatch.py", line 41, in <module>
	    main()
	  File "./python/twatch.py", line 32, in main
	    event = evlist.read_on_cpu(cpu)
	RuntimeError: more argument specifiers than keyword list entries

Hence, add cpu to the name list.

Cc: David Ahern <daahern@cisco.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Cc: Tom Zanussi <tzanussi@gmail.com>
Link: http://lkml.kernel.org/r/1301588863-20210-1-git-send-email-fweisbec@gmail.com
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/python.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/tools/perf/util/python.c b/tools/perf/util/python.c
index 24063b4..a9ac050 100644
--- a/tools/perf/util/python.c
+++ b/tools/perf/util/python.c
@@ -674,7 +674,7 @@ static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist,
 	struct perf_evlist *evlist = &pevlist->evlist;
 	union perf_event *event;
 	int sample_id_all = 1, cpu;
-	static char *kwlist[] = {"sample_id_all", NULL, NULL};
+	static char *kwlist[] = {"cpu", "sample_id_all", NULL, NULL};
 	int err;
 
 	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "i|i", kwlist,
-- 
1.6.2.5

[PATCH 5/9] perf python: Cleanup useless double NULL termination in method arg names

[Arnaldo Carvalho de Melo]

From: Frederic Weisbecker <fweisbec@gmail.com>

The list of methods argument names only needs to be NULL terminated
once. Remove the second ones.

Cc: David Ahern <daahern@cisco.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Cc: Tom Zanussi <tzanussi@gmail.com>
Link: http://lkml.kernel.org/r/1301588863-20210-2-git-send-email-fweisbec@gmail.com
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/python.c |   17 +++++++++--------
 1 files changed, 9 insertions(+), 8 deletions(-)

diff --git a/tools/perf/util/python.c b/tools/perf/util/python.c
index a9ac050..8e0b5a3 100644
--- a/tools/perf/util/python.c
+++ b/tools/perf/util/python.c
@@ -247,7 +247,7 @@ struct pyrf_cpu_map {
 static int pyrf_cpu_map__init(struct pyrf_cpu_map *pcpus,
 			      PyObject *args, PyObject *kwargs)
 {
-	static char *kwlist[] = { "cpustr", NULL, NULL, };
+	static char *kwlist[] = { "cpustr", NULL };
 	char *cpustr = NULL;
 
 	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|s",
@@ -316,7 +316,7 @@ struct pyrf_thread_map {
 static int pyrf_thread_map__init(struct pyrf_thread_map *pthreads,
 				 PyObject *args, PyObject *kwargs)
 {
-	static char *kwlist[] = { "pid", "tid", NULL, NULL, };
+	static char *kwlist[] = { "pid", "tid", NULL };
 	int pid = -1, tid = -1;
 
 	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|ii",
@@ -418,7 +418,9 @@ static int pyrf_evsel__init(struct pyrf_evsel *pevsel,
 		"wakeup_events",
 		"bp_type",
 		"bp_addr",
-		"bp_len", NULL, NULL, };
+		"bp_len",
+		 NULL
+	};
 	u64 sample_period = 0;
 	u32 disabled = 0,
 	    inherit = 0,
@@ -499,7 +501,7 @@ static PyObject *pyrf_evsel__open(struct pyrf_evsel *pevsel,
 	struct thread_map *threads = NULL;
 	PyObject *pcpus = NULL, *pthreads = NULL;
 	int group = 0, inherit = 0;
-	static char *kwlist[] = {"cpus", "threads", "group", "inherit", NULL, NULL};
+	static char *kwlist[] = { "cpus", "threads", "group", "inherit", NULL };
 
 	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|OOii", kwlist,
 					 &pcpus, &pthreads, &group, &inherit))
@@ -582,8 +584,7 @@ static PyObject *pyrf_evlist__mmap(struct pyrf_evlist *pevlist,
 				   PyObject *args, PyObject *kwargs)
 {
 	struct perf_evlist *evlist = &pevlist->evlist;
-	static char *kwlist[] = {"pages", "overwrite",
-				  NULL, NULL};
+	static char *kwlist[] = { "pages", "overwrite", NULL };
 	int pages = 128, overwrite = false;
 
 	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|ii", kwlist,
@@ -603,7 +604,7 @@ static PyObject *pyrf_evlist__poll(struct pyrf_evlist *pevlist,
 				   PyObject *args, PyObject *kwargs)
 {
 	struct perf_evlist *evlist = &pevlist->evlist;
-	static char *kwlist[] = {"timeout", NULL, NULL};
+	static char *kwlist[] = { "timeout", NULL };
 	int timeout = -1, n;
 
 	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|i", kwlist, &timeout))
@@ -674,7 +675,7 @@ static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist,
 	struct perf_evlist *evlist = &pevlist->evlist;
 	union perf_event *event;
 	int sample_id_all = 1, cpu;
-	static char *kwlist[] = {"cpu", "sample_id_all", NULL, NULL};
+	static char *kwlist[] = { "cpu", "sample_id_all", NULL };
 	int err;
 
 	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "i|i", kwlist,
-- 
1.6.2.5

[PATCH 6/9] perf stat: clarify unsupported events from uncounted events

[Arnaldo Carvalho de Melo]

From: David Ahern <dsahern@gmail.com>

perf stat continues running even if the event list contains counters
that are not supported. The resulting output then contains <not counted>
for those events which gets confusing as to which events are supported,
but not counted and which are not supported.

Before:

perf stat -ddd -- sleep 1

      Performance counter stats for 'sleep 1':

          0.571283 task-clock                #    0.001 CPUs utilized
                 1 context-switches          #    0.002 M/sec
                 0 CPU-migrations            #    0.000 M/sec
               157 page-faults               #    0.275 M/sec
         1,037,707 cycles                    #    1.816 GHz
     <not counted> stalled-cycles-frontend
     <not counted> stalled-cycles-backend
           654,499 instructions              #    0.63  insns per cycle
           136,129 branches                  #  238.286 M/sec
     <not counted> branch-misses
     <not counted> L1-dcache-loads
     <not counted> L1-dcache-load-misses
     <not counted> LLC-loads
     <not counted> LLC-load-misses
     <not counted> L1-icache-loads
     <not counted> L1-icache-load-misses
     <not counted> dTLB-loads
     <not counted> dTLB-load-misses
     <not counted> iTLB-loads
     <not counted> iTLB-load-misses
     <not counted> L1-dcache-prefetches
     <not counted> L1-dcache-prefetch-misses

       1.001004836 seconds time elapsed

After:

perf stat -ddd -- sleep 1

 Performance counter stats for 'sleep 1':

          1.350326 task-clock                #    0.001 CPUs utilized
                 2 context-switches          #    0.001 M/sec
                 0 CPU-migrations            #    0.000 M/sec
               157 page-faults               #    0.116 M/sec
            11,986 cycles                    #    0.009 GHz
   <not supported> stalled-cycles-frontend
   <not supported> stalled-cycles-backend
           496,986 instructions              #   41.46  insns per cycle
           138,065 branches                  #  102.246 M/sec
             7,245 branch-misses             #    5.25% of all branches
     <not counted> L1-dcache-loads
     <not counted> L1-dcache-load-misses
     <not counted> LLC-loads
     <not counted> LLC-load-misses
     <not counted> L1-icache-loads
     <not counted> L1-icache-load-misses
     <not counted> dTLB-loads
     <not counted> dTLB-load-misses
     <not counted> iTLB-loads
     <not counted> iTLB-load-misses
     <not counted> L1-dcache-prefetches
   <not supported> L1-dcache-prefetch-misses

       1.002397333 seconds time elapsed

v1->v2:
changed supported type from int to bool

v2->v3
fixed vertical alignment of new struct element

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1306767359-13221-1-git-send-email-dsahern@gmail.com
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-stat.c |    9 +++++++--
 tools/perf/util/evsel.h   |    1 +
 2 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index a9f0671..784ed6d 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -61,6 +61,8 @@
 #include <locale.h>
 
 #define DEFAULT_SEPARATOR	" "
+#define CNTR_NOT_SUPPORTED	"<not supported>"
+#define CNTR_NOT_COUNTED	"<not counted>"
 
 static struct perf_event_attr default_attrs[] = {
 
@@ -448,6 +450,7 @@ static int run_perf_stat(int argc __used, const char **argv)
 				if (verbose)
 					ui__warning("%s event is not supported by the kernel.\n",
 						    event_name(counter));
+				counter->supported = false;
 				continue;
 			}
 
@@ -466,6 +469,7 @@ static int run_perf_stat(int argc __used, const char **argv)
 			die("Not all events could be opened.\n");
 			return -1;
 		}
+		counter->supported = true;
 	}
 
 	if (perf_evlist__set_filters(evsel_list)) {
@@ -861,7 +865,7 @@ static void print_counter_aggr(struct perf_evsel *counter)
 	if (scaled == -1) {
 		fprintf(stderr, "%*s%s%*s",
 			csv_output ? 0 : 18,
-			"<not counted>",
+			counter->supported ? CNTR_NOT_COUNTED : CNTR_NOT_SUPPORTED,
 			csv_sep,
 			csv_output ? 0 : -24,
 			event_name(counter));
@@ -914,7 +918,8 @@ static void print_counter(struct perf_evsel *counter)
 				csv_output ? 0 : -4,
 				evsel_list->cpus->map[cpu], csv_sep,
 				csv_output ? 0 : 18,
-				"<not counted>", csv_sep,
+				counter->supported ? CNTR_NOT_COUNTED : CNTR_NOT_SUPPORTED,
+				csv_sep,
 				csv_output ? 0 : -24,
 				event_name(counter));
 
diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
index 7e9366e..e9a3155 100644
--- a/tools/perf/util/evsel.h
+++ b/tools/perf/util/evsel.h
@@ -61,6 +61,7 @@ struct perf_evsel {
 		off_t		id_offset;
 	};
 	struct cgroup_sel	*cgrp;
+	bool 			supported;
 };
 
 struct cpu_map;
-- 
1.6.2.5

[PATCH 7/9] perf script: "sym" field really means show IP data

[Arnaldo Carvalho de Melo]

From: David Ahern <dsahern@gmail.com>

Currently the "sym" output field is used to dump instruction pointers
and callchain stack. Sample addresses can also be converted to symbols,
so the meaning of "sym" needs to be fixed. This patch adds an "ip"
option and if it is selected the user can also opt to dump symbols for
them. If the user opts to dump IP without syms only the address is
shown.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1306528124-25861-2-git-send-email-dsahern@gmail.com
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-script.txt |   10 +++---
 tools/perf/builtin-script.c              |   31 ++++++++++++------
 tools/perf/util/session.c                |   50 +++++++++++++++++------------
 tools/perf/util/session.h                |    5 ++-
 4 files changed, 58 insertions(+), 38 deletions(-)

diff --git a/tools/perf/Documentation/perf-script.txt b/tools/perf/Documentation/perf-script.txt
index 86c87e2..67a4e5c 100644
--- a/tools/perf/Documentation/perf-script.txt
+++ b/tools/perf/Documentation/perf-script.txt
@@ -115,10 +115,10 @@ OPTIONS
 -f::
 --fields::
         Comma separated list of fields to print. Options are:
-        comm, tid, pid, time, cpu, event, trace, sym. Field
+        comm, tid, pid, time, cpu, event, trace, ip, sym. Field
         list can be prepended with the type, trace, sw or hw,
         to indicate to which event type the field list applies.
-        e.g., -f sw:comm,tid,time,sym  and -f trace:time,cpu,trace
+        e.g., -f sw:comm,tid,time,ip,sym  and -f trace:time,cpu,trace
 
 		perf script -f <fields>
 
@@ -132,17 +132,17 @@ OPTIONS
 	The arguments are processed in the order received. A later usage can
 	reset a prior request. e.g.:
     
-		-f trace: -f comm,tid,time,sym
+		-f trace: -f comm,tid,time,ip,sym
     
 	The first -f suppresses trace events (field list is ""), but then the
-	second invocation sets the fields to comm,tid,time,sym. In this case a
+	second invocation sets the fields to comm,tid,time,ip,sym. In this case a
 	warning is given to the user:
     
 		"Overriding previous field request for all events."
     
 	Alternativey, consider the order:
     
-		-f comm,tid,time,sym -f trace:
+		-f comm,tid,time,ip,sym -f trace:
     
 	The first -f sets the fields for all events and the second -f
 	suppresses trace events. The user is given a warning message about
diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index 22747de..0852db2 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -30,7 +30,8 @@ enum perf_output_field {
 	PERF_OUTPUT_CPU             = 1U << 4,
 	PERF_OUTPUT_EVNAME          = 1U << 5,
 	PERF_OUTPUT_TRACE           = 1U << 6,
-	PERF_OUTPUT_SYM             = 1U << 7,
+	PERF_OUTPUT_IP              = 1U << 7,
+	PERF_OUTPUT_SYM             = 1U << 8,
 };
 
 struct output_option {
@@ -44,6 +45,7 @@ struct output_option {
 	{.str = "cpu",   .field = PERF_OUTPUT_CPU},
 	{.str = "event", .field = PERF_OUTPUT_EVNAME},
 	{.str = "trace", .field = PERF_OUTPUT_TRACE},
+	{.str = "ip",    .field = PERF_OUTPUT_IP},
 	{.str = "sym",   .field = PERF_OUTPUT_SYM},
 };
 
@@ -60,7 +62,8 @@ static struct {
 
 		.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
 			      PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
-			      PERF_OUTPUT_EVNAME | PERF_OUTPUT_SYM,
+			      PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
+				  PERF_OUTPUT_SYM,
 
 		.invalid_fields = PERF_OUTPUT_TRACE,
 	},
@@ -70,7 +73,8 @@ static struct {
 
 		.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
 			      PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
-			      PERF_OUTPUT_EVNAME | PERF_OUTPUT_SYM,
+			      PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
+				  PERF_OUTPUT_SYM,
 
 		.invalid_fields = PERF_OUTPUT_TRACE,
 	},
@@ -88,7 +92,8 @@ static struct {
 
 		.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
 			      PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
-			      PERF_OUTPUT_EVNAME | PERF_OUTPUT_SYM,
+			      PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
+				  PERF_OUTPUT_SYM,
 
 		.invalid_fields = PERF_OUTPUT_TRACE,
 	},
@@ -157,15 +162,20 @@ static int perf_evsel__check_attr(struct perf_evsel *evsel,
 		!perf_session__has_traces(session, "record -R"))
 		return -EINVAL;
 
-	if (PRINT_FIELD(SYM)) {
+	if (PRINT_FIELD(IP)) {
 		if (perf_event_attr__check_stype(attr, PERF_SAMPLE_IP, "IP",
-					   PERF_OUTPUT_SYM))
+					   PERF_OUTPUT_IP))
 			return -EINVAL;
 
 		if (!no_callchain &&
 		    !(attr->sample_type & PERF_SAMPLE_CALLCHAIN))
 			symbol_conf.use_callchain = false;
 	}
+	if (PRINT_FIELD(SYM) && !PRINT_FIELD(IP)) {
+		pr_err("Display of symbols requested but IP is not selected.\n"
+		       "No addresses to convert to symbols.\n");
+		return -EINVAL;
+	}
 
 	if ((PRINT_FIELD(PID) || PRINT_FIELD(TID)) &&
 		perf_event_attr__check_stype(attr, PERF_SAMPLE_TID, "TID",
@@ -230,7 +240,7 @@ static void print_sample_start(struct perf_sample *sample,
 	if (PRINT_FIELD(COMM)) {
 		if (latency_format)
 			printf("%8.8s ", thread->comm);
-		else if (PRINT_FIELD(SYM) && symbol_conf.use_callchain)
+		else if (PRINT_FIELD(IP) && symbol_conf.use_callchain)
 			printf("%s ", thread->comm);
 		else
 			printf("%16s ", thread->comm);
@@ -288,12 +298,13 @@ static void process_event(union perf_event *event __unused,
 		print_trace_event(sample->cpu, sample->raw_data,
 				  sample->raw_size);
 
-	if (PRINT_FIELD(SYM)) {
+	if (PRINT_FIELD(IP)) {
 		if (!symbol_conf.use_callchain)
 			printf(" ");
 		else
 			printf("\n");
-		perf_session__print_symbols(event, sample, session);
+		perf_session__print_ip(event, sample, session,
+					      PRINT_FIELD(SYM));
 	}
 
 	printf("\n");
@@ -985,7 +996,7 @@ static const struct option options[] = {
 	OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory",
 		    "Look for files with symbols relative to this directory"),
 	OPT_CALLBACK('f', "fields", NULL, "str",
-		     "comma separated output fields prepend with 'type:'. Valid types: hw,sw,trace,raw. Fields: comm,tid,pid,time,cpu,event,trace,sym",
+		     "comma separated output fields prepend with 'type:'. Valid types: hw,sw,trace,raw. Fields: comm,tid,pid,time,cpu,event,trace,ip,sym",
 		     parse_output_fields),
 
 	OPT_END()
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index f5a8fbd..ad33650 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -1202,9 +1202,10 @@ struct perf_evsel *perf_session__find_first_evtype(struct perf_session *session,
 	return NULL;
 }
 
-void perf_session__print_symbols(union perf_event *event,
-				struct perf_sample *sample,
-				struct perf_session *session)
+void perf_session__print_ip(union perf_event *event,
+			    struct perf_sample *sample,
+			    struct perf_session *session,
+			    int print_sym)
 {
 	struct addr_location al;
 	const char *symname, *dsoname;
@@ -1233,32 +1234,39 @@ void perf_session__print_symbols(union perf_event *event,
 			if (!node)
 				break;
 
-			if (node->sym && node->sym->name)
-				symname = node->sym->name;
-			else
-				symname = "";
+			printf("\t%16" PRIx64, node->ip);
+			if (print_sym) {
+				if (node->sym && node->sym->name)
+					symname = node->sym->name;
+				else
+					symname = "";
 
-			if (node->map && node->map->dso && node->map->dso->name)
-				dsoname = node->map->dso->name;
-			else
-				dsoname = "";
+				if (node->map && node->map->dso && node->map->dso->name)
+					dsoname = node->map->dso->name;
+				else
+					dsoname = "";
 
-			printf("\t%16" PRIx64 " %s (%s)\n", node->ip, symname, dsoname);
+				printf(" %s (%s)", symname, dsoname);
+			}
+			printf("\n");
 
 			callchain_cursor_advance(cursor);
 		}
 
 	} else {
-		if (al.sym && al.sym->name)
-			symname = al.sym->name;
-		else
-			symname = "";
+		printf("%16" PRIx64, al.addr);
+		if (print_sym) {
+			if (al.sym && al.sym->name)
+				symname = al.sym->name;
+			else
+				symname = "";
 
-		if (al.map && al.map->dso && al.map->dso->name)
-			dsoname = al.map->dso->name;
-		else
-			dsoname = "";
+			if (al.map && al.map->dso && al.map->dso->name)
+				dsoname = al.map->dso->name;
+			else
+				dsoname = "";
 
-		printf("%16" PRIx64 " %s (%s)", al.addr, symname, dsoname);
+			printf(" %s (%s)", symname, dsoname);
+		}
 	}
 }
diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h
index 66d4e14..d76af0f 100644
--- a/tools/perf/util/session.h
+++ b/tools/perf/util/session.h
@@ -167,8 +167,9 @@ static inline int perf_session__parse_sample(struct perf_session *session,
 struct perf_evsel *perf_session__find_first_evtype(struct perf_session *session,
 					    unsigned int type);
 
-void perf_session__print_symbols(union perf_event *event,
+void perf_session__print_ip(union perf_event *event,
 				 struct perf_sample *sample,
-				 struct perf_session *session);
+				 struct perf_session *session,
+				 int print_sym);
 
 #endif /* __PERF_SESSION_H */
-- 
1.6.2.5

[PATCH 8/9] perf script: Make printing of dso a separate field option

[Arnaldo Carvalho de Melo]

From: David Ahern <dsahern@gmail.com>

The 'sym' option displays both the function name and the DSO it comes
from. Split the display of the dso into a separate option.  This allows
display of the ip address and symbol without the dso, thus shortening
line lengths - and decluttering the output a bit.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1306528124-25861-3-git-send-email-dsahern@gmail.com
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-script.txt |    2 +-
 tools/perf/builtin-script.c              |   17 ++++++++++++-----
 tools/perf/util/session.c                |   13 ++++++++++---
 tools/perf/util/session.h                |    2 +-
 4 files changed, 24 insertions(+), 10 deletions(-)

diff --git a/tools/perf/Documentation/perf-script.txt b/tools/perf/Documentation/perf-script.txt
index 67a4e5c..1e744c2 100644
--- a/tools/perf/Documentation/perf-script.txt
+++ b/tools/perf/Documentation/perf-script.txt
@@ -115,7 +115,7 @@ OPTIONS
 -f::
 --fields::
         Comma separated list of fields to print. Options are:
-        comm, tid, pid, time, cpu, event, trace, ip, sym. Field
+        comm, tid, pid, time, cpu, event, trace, ip, sym, dso. Field
         list can be prepended with the type, trace, sw or hw,
         to indicate to which event type the field list applies.
         e.g., -f sw:comm,tid,time,ip,sym  and -f trace:time,cpu,trace
diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index 0852db2..a8bd00f 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -32,6 +32,7 @@ enum perf_output_field {
 	PERF_OUTPUT_TRACE           = 1U << 6,
 	PERF_OUTPUT_IP              = 1U << 7,
 	PERF_OUTPUT_SYM             = 1U << 8,
+	PERF_OUTPUT_DSO             = 1U << 9,
 };
 
 struct output_option {
@@ -47,6 +48,7 @@ struct output_option {
 	{.str = "trace", .field = PERF_OUTPUT_TRACE},
 	{.str = "ip",    .field = PERF_OUTPUT_IP},
 	{.str = "sym",   .field = PERF_OUTPUT_SYM},
+	{.str = "dso",   .field = PERF_OUTPUT_DSO},
 };
 
 /* default set to maintain compatibility with current format */
@@ -63,7 +65,7 @@ static struct {
 		.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
 			      PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
 			      PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
-				  PERF_OUTPUT_SYM,
+				  PERF_OUTPUT_SYM | PERF_OUTPUT_DSO,
 
 		.invalid_fields = PERF_OUTPUT_TRACE,
 	},
@@ -74,7 +76,7 @@ static struct {
 		.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
 			      PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
 			      PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
-				  PERF_OUTPUT_SYM,
+				  PERF_OUTPUT_SYM | PERF_OUTPUT_DSO,
 
 		.invalid_fields = PERF_OUTPUT_TRACE,
 	},
@@ -93,7 +95,7 @@ static struct {
 		.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
 			      PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
 			      PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
-				  PERF_OUTPUT_SYM,
+				  PERF_OUTPUT_SYM | PERF_OUTPUT_DSO,
 
 		.invalid_fields = PERF_OUTPUT_TRACE,
 	},
@@ -176,6 +178,11 @@ static int perf_evsel__check_attr(struct perf_evsel *evsel,
 		       "No addresses to convert to symbols.\n");
 		return -EINVAL;
 	}
+	if (PRINT_FIELD(DSO) && !PRINT_FIELD(IP)) {
+		pr_err("Display of DSO requested but IP is not selected.\n"
+		       "No addresses to convert to dso.\n");
+		return -EINVAL;
+	}
 
 	if ((PRINT_FIELD(PID) || PRINT_FIELD(TID)) &&
 		perf_event_attr__check_stype(attr, PERF_SAMPLE_TID, "TID",
@@ -304,7 +311,7 @@ static void process_event(union perf_event *event __unused,
 		else
 			printf("\n");
 		perf_session__print_ip(event, sample, session,
-					      PRINT_FIELD(SYM));
+					      PRINT_FIELD(SYM), PRINT_FIELD(DSO));
 	}
 
 	printf("\n");
@@ -996,7 +1003,7 @@ static const struct option options[] = {
 	OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory",
 		    "Look for files with symbols relative to this directory"),
 	OPT_CALLBACK('f', "fields", NULL, "str",
-		     "comma separated output fields prepend with 'type:'. Valid types: hw,sw,trace,raw. Fields: comm,tid,pid,time,cpu,event,trace,ip,sym",
+		     "comma separated output fields prepend with 'type:'. Valid types: hw,sw,trace,raw. Fields: comm,tid,pid,time,cpu,event,trace,ip,sym,dso",
 		     parse_output_fields),
 
 	OPT_END()
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index ad33650..0dd4182 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -1205,7 +1205,7 @@ struct perf_evsel *perf_session__find_first_evtype(struct perf_session *session,
 void perf_session__print_ip(union perf_event *event,
 			    struct perf_sample *sample,
 			    struct perf_session *session,
-			    int print_sym)
+			    int print_sym, int print_dso)
 {
 	struct addr_location al;
 	const char *symname, *dsoname;
@@ -1241,12 +1241,15 @@ void perf_session__print_ip(union perf_event *event,
 				else
 					symname = "";
 
+				printf(" %s", symname);
+			}
+			if (print_dso) {
 				if (node->map && node->map->dso && node->map->dso->name)
 					dsoname = node->map->dso->name;
 				else
 					dsoname = "";
 
-				printf(" %s (%s)", symname, dsoname);
+				printf(" (%s)", dsoname);
 			}
 			printf("\n");
 
@@ -1261,12 +1264,16 @@ void perf_session__print_ip(union perf_event *event,
 			else
 				symname = "";
 
+			printf(" %s", symname);
+		}
+
+		if (print_dso) {
 			if (al.map && al.map->dso && al.map->dso->name)
 				dsoname = al.map->dso->name;
 			else
 				dsoname = "";
 
-			printf(" %s (%s)", symname, dsoname);
+			printf(" (%s)", dsoname);
 		}
 	}
 }
diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h
index d76af0f..de4178d 100644
--- a/tools/perf/util/session.h
+++ b/tools/perf/util/session.h
@@ -170,6 +170,6 @@ struct perf_evsel *perf_session__find_first_evtype(struct perf_session *session,
 void perf_session__print_ip(union perf_event *event,
 				 struct perf_sample *sample,
 				 struct perf_session *session,
-				 int print_sym);
+				 int print_sym, int print_dso);
 
 #endif /* __PERF_SESSION_H */
-- 
1.6.2.5

[PATCH 9/9] perf script: Add printing of sample address

[Arnaldo Carvalho de Melo]

From: David Ahern <dsahern@gmail.com>

Resolve to a function or variable if possible and if the sym option is
enabled.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1306782503-22002-1-git-send-email-dsahern@gmail.com
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-script.txt |    4 +-
 tools/perf/builtin-script.c              |   84 +++++++++++++++++++++++++++---
 tools/perf/util/evsel.c                  |    1 +
 tools/perf/util/session.c                |    4 +-
 4 files changed, 82 insertions(+), 11 deletions(-)

diff --git a/tools/perf/Documentation/perf-script.txt b/tools/perf/Documentation/perf-script.txt
index 1e744c2..c6068cb 100644
--- a/tools/perf/Documentation/perf-script.txt
+++ b/tools/perf/Documentation/perf-script.txt
@@ -115,8 +115,8 @@ OPTIONS
 -f::
 --fields::
         Comma separated list of fields to print. Options are:
-        comm, tid, pid, time, cpu, event, trace, ip, sym, dso. Field
-        list can be prepended with the type, trace, sw or hw,
+        comm, tid, pid, time, cpu, event, trace, ip, sym, dso, addr.
+        Field list can be prepended with the type, trace, sw or hw,
         to indicate to which event type the field list applies.
         e.g., -f sw:comm,tid,time,ip,sym  and -f trace:time,cpu,trace
 
diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index a8bd00f..3056b45 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -33,6 +33,7 @@ enum perf_output_field {
 	PERF_OUTPUT_IP              = 1U << 7,
 	PERF_OUTPUT_SYM             = 1U << 8,
 	PERF_OUTPUT_DSO             = 1U << 9,
+	PERF_OUTPUT_ADDR            = 1U << 10,
 };
 
 struct output_option {
@@ -49,6 +50,7 @@ struct output_option {
 	{.str = "ip",    .field = PERF_OUTPUT_IP},
 	{.str = "sym",   .field = PERF_OUTPUT_SYM},
 	{.str = "dso",   .field = PERF_OUTPUT_DSO},
+	{.str = "addr",  .field = PERF_OUTPUT_ADDR},
 };
 
 /* default set to maintain compatibility with current format */
@@ -173,14 +175,22 @@ static int perf_evsel__check_attr(struct perf_evsel *evsel,
 		    !(attr->sample_type & PERF_SAMPLE_CALLCHAIN))
 			symbol_conf.use_callchain = false;
 	}
-	if (PRINT_FIELD(SYM) && !PRINT_FIELD(IP)) {
-		pr_err("Display of symbols requested but IP is not selected.\n"
-		       "No addresses to convert to symbols.\n");
+
+	if (PRINT_FIELD(ADDR) &&
+		perf_event_attr__check_stype(attr, PERF_SAMPLE_ADDR, "ADDR",
+				       PERF_OUTPUT_ADDR))
+		return -EINVAL;
+
+	if (PRINT_FIELD(SYM) && !PRINT_FIELD(IP) && !PRINT_FIELD(ADDR)) {
+		pr_err("Display of symbols requested but neither sample IP nor "
+			   "sample address\nis selected. Hence, no addresses to convert "
+		       "to symbols.\n");
 		return -EINVAL;
 	}
-	if (PRINT_FIELD(DSO) && !PRINT_FIELD(IP)) {
-		pr_err("Display of DSO requested but IP is not selected.\n"
-		       "No addresses to convert to dso.\n");
+	if (PRINT_FIELD(DSO) && !PRINT_FIELD(IP) && !PRINT_FIELD(ADDR)) {
+		pr_err("Display of DSO requested but neither sample IP nor "
+			   "sample address\nis selected. Hence, no addresses to convert "
+		       "to DSO.\n");
 		return -EINVAL;
 	}
 
@@ -288,6 +298,63 @@ static void print_sample_start(struct perf_sample *sample,
 	}
 }
 
+static bool sample_addr_correlates_sym(struct perf_event_attr *attr)
+{
+	if ((attr->type == PERF_TYPE_SOFTWARE) &&
+	    ((attr->config == PERF_COUNT_SW_PAGE_FAULTS) ||
+	     (attr->config == PERF_COUNT_SW_PAGE_FAULTS_MIN) ||
+	     (attr->config == PERF_COUNT_SW_PAGE_FAULTS_MAJ)))
+		return true;
+
+	return false;
+}
+
+static void print_sample_addr(union perf_event *event,
+			  struct perf_sample *sample,
+			  struct perf_session *session,
+			  struct thread *thread,
+			  struct perf_event_attr *attr)
+{
+	struct addr_location al;
+	u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+	const char *symname, *dsoname;
+
+	printf("%16" PRIx64, sample->addr);
+
+	if (!sample_addr_correlates_sym(attr))
+		return;
+
+	thread__find_addr_map(thread, session, cpumode, MAP__FUNCTION,
+			      event->ip.pid, sample->addr, &al);
+	if (!al.map)
+		thread__find_addr_map(thread, session, cpumode, MAP__VARIABLE,
+				      event->ip.pid, sample->addr, &al);
+
+	al.cpu = sample->cpu;
+	al.sym = NULL;
+
+	if (al.map)
+		al.sym = map__find_symbol(al.map, al.addr, NULL);
+
+	if (PRINT_FIELD(SYM)) {
+		if (al.sym && al.sym->name)
+			symname = al.sym->name;
+		else
+			symname = "";
+
+		printf(" %16s", symname);
+	}
+
+	if (PRINT_FIELD(DSO)) {
+		if (al.map && al.map->dso && al.map->dso->name)
+			dsoname = al.map->dso->name;
+		else
+			dsoname = "";
+
+		printf(" (%s)", dsoname);
+	}
+}
+
 static void process_event(union perf_event *event __unused,
 			  struct perf_sample *sample,
 			  struct perf_evsel *evsel,
@@ -305,6 +372,9 @@ static void process_event(union perf_event *event __unused,
 		print_trace_event(sample->cpu, sample->raw_data,
 				  sample->raw_size);
 
+	if (PRINT_FIELD(ADDR))
+		print_sample_addr(event, sample, session, thread, attr);
+
 	if (PRINT_FIELD(IP)) {
 		if (!symbol_conf.use_callchain)
 			printf(" ");
@@ -1003,7 +1073,7 @@ static const struct option options[] = {
 	OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory",
 		    "Look for files with symbols relative to this directory"),
 	OPT_CALLBACK('f', "fields", NULL, "str",
-		     "comma separated output fields prepend with 'type:'. Valid types: hw,sw,trace,raw. Fields: comm,tid,pid,time,cpu,event,trace,ip,sym,dso",
+		     "comma separated output fields prepend with 'type:'. Valid types: hw,sw,trace,raw. Fields: comm,tid,pid,time,cpu,event,trace,ip,sym,dso,addr",
 		     parse_output_fields),
 
 	OPT_END()
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 0239eb8..a03a36b 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -377,6 +377,7 @@ int perf_event__parse_sample(const union perf_event *event, u64 type,
 		array++;
 	}
 
+	data->addr = 0;
 	if (type & PERF_SAMPLE_ADDR) {
 		data->addr = *array;
 		array++;
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index 0dd4182..b723f21 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -708,9 +708,9 @@ static void dump_sample(struct perf_session *session, union perf_event *event,
 	if (!dump_trace)
 		return;
 
-	printf("(IP, %d): %d/%d: %#" PRIx64 " period: %" PRIu64 "\n",
+	printf("(IP, %d): %d/%d: %#" PRIx64 " period: %" PRIu64 " addr: %#" PRIx64 "\n",
 	       event->header.misc, sample->pid, sample->tid, sample->ip,
-	       sample->period);
+	       sample->period, sample->addr);
 
 	if (session->sample_type & PERF_SAMPLE_CALLCHAIN)
 		callchain__printf(sample);
-- 
1.6.2.5

Re: [GIT PULL 0/9] perf/core fixes and improvements

[Ingo Molnar]

* Arnaldo Carvalho de Melo <acme@infradead.org> wrote:

> Hi Ingo,
> 
>         Please consider pulling from:
> 
> git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux-2.6 perf/core
> 
> Regards,
> 
> - Arnaldo
> 
> Arnaldo Carvalho de Melo (3):
>   perf evlist: Remove dependency on debug routines
>   perf python: Use exception to propagate errors
>   perf evlist: Don't die if sample_{id_all|type} is invalid
> 
> David Ahern (4):
>   perf stat: clarify unsupported events from uncounted events
>   perf script: "sym" field really means show IP data
>   perf script: Make printing of dso a separate field option
>   perf script: Add printing of sample address
> 
> Frederic Weisbecker (2):
>   perf python: Fix argument name list of read_on_cpu()
>   perf python: Cleanup useless double NULL termination in method arg names
> 
>  tools/perf/Documentation/perf-script.txt |   12 ++--
>  tools/perf/builtin-script.c              |  108 +++++++++++++++++++++++++++---
>  tools/perf/builtin-stat.c                |    9 ++-
>  tools/perf/builtin-test.c                |    2 +-
>  tools/perf/util/event.c                  |   16 -----
>  tools/perf/util/event.h                  |    2 -
>  tools/perf/util/evlist.c                 |   68 +++++++++++--------
>  tools/perf/util/evlist.h                 |    6 +-
>  tools/perf/util/evsel.c                  |   17 +++++
>  tools/perf/util/evsel.h                  |    8 ++
>  tools/perf/util/python.c                 |   29 ++++----
>  tools/perf/util/session.c                |   73 ++++++++++++++-------
>  tools/perf/util/session.h                |    5 +-
>  13 files changed, 245 insertions(+), 110 deletions(-)

Pulled, thanks a lot Arnaldo!

Mind cherry-picking the fixes into perf/urgent?

Thanks,

	Ingo