[GIT PULL 00/18] perf/core improvements and fixes

[GIT PULL 00/18] perf/core improvements and fixes

[Arnaldo Carvalho de Melo]

Hi Ingo,

	Please consider pulling,

- Arnaldo

The following changes since commit a11c51acc52822754d66a11c15f6f6edd4d23c55:

  Merge tag 'perf-core-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core (2015-07-21 07:58:06 +0200)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-core-for-mingo

for you to fetch changes up to 7c14898ba9386ee5c939bb418643ac6baff52840:

  perf script: Add option --show-switch-events (2015-07-23 22:51:14 -0300)

----------------------------------------------------------------
perf/core improvements and fixes:

New features:

- Introduce PERF_RECORD_SWITCH(_CPU_WIDE) and use it in 'record' to
  ask for context switches, allowing non priviledged tasks to know
  when they are switched in and out, which wasn't possible with
  the other context switch tracepoint and software events, see the
  patch description for a comprehensive justification (Adrian Hunter)

- Stop collecting /proc/kallsyms in perf.data files, saving about
  4.5MB on a typical x86-64 system, use the symbol resolution
  routines used in all the other tools (report, top, etc) now that
  we can ask libtraceevent to use perf's symbol resolution code.
  (Arnaldo Carvalho de Melo)

User visible fixes:

- Expose perf's symbol resolver to libtraceecent, so that its plugins can
  resolve tracepoint fields to kernel functions, like the 'function' field
  in the "timer:hrtimer_start tracepoint" (Arnaldo Carvalho de Melo)

Infrastructure:

- Map propagation of thread and cpu maps improvements, prep work for
  'perf stat' new features (Jiri Olsa)

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

----------------------------------------------------------------
Adrian Hunter (5):
      perf: Add PERF_RECORD_SWITCH to indicate context switches
      perf tools: Add new PERF_RECORD_SWITCH event
      perf record: Add option --switch-events to select PERF_RECORD_SWITCH events
      perf script: Don't assume evsel position of tracking events
      perf script: Add option --show-switch-events

Arnaldo Carvalho de Melo (8):
      perf symbols: Add front end cache for DSO symbol lookup
      perf symbols: Introduce map__is_(kernel,kmodule)()
      tools lib traceevent: Allow setting an alternative symbol resolver
      perf symbols: Provide libtraceevent callback to resolve kernel symbols
      perf trace: Provide libtracevent with a kernel symbol resolver
      perf script: Switch from perf.data's kallsyms to perf's symbol resolver
      perf tools: Stop reading the kallsyms data from perf.data
      perf tools: Stop copying kallsyms into the perf.data file header

Jiri Olsa (5):
      perf test: Check for refcnt in thread_map test
      perf evlist: Force perf_evlist__set_maps to propagate maps through events
      perf evlist: Use bool instead of target argument in propagate_maps()
      perf evlist: Tolerate NULL maps in propagate_maps
      perf header: Use argv style storage for cmdline feature data

 include/uapi/linux/perf_event.h          |  31 +++++++++-
 kernel/events/core.c                     | 103 +++++++++++++++++++++++++++++++
 tools/lib/traceevent/event-parse.c       |  68 +++++++++++++++++++-
 tools/lib/traceevent/event-parse.h       |   8 +++
 tools/perf/Documentation/perf-record.txt |   4 ++
 tools/perf/Documentation/perf-script.txt |   4 ++
 tools/perf/builtin-inject.c              |   1 +
 tools/perf/builtin-record.c              |   7 +++
 tools/perf/builtin-script.c              |  48 ++++++++++++--
 tools/perf/builtin-trace.c               |   3 +
 tools/perf/perf.h                        |   1 +
 tools/perf/tests/thread-map.c            |   4 ++
 tools/perf/util/dso.h                    |   4 ++
 tools/perf/util/event.c                  |  28 +++++++++
 tools/perf/util/event.h                  |  12 ++++
 tools/perf/util/evlist.c                 |  28 +++++++--
 tools/perf/util/evlist.h                 |  12 ++--
 tools/perf/util/evsel.c                  |   4 ++
 tools/perf/util/header.c                 |  35 ++++++-----
 tools/perf/util/header.h                 |   1 +
 tools/perf/util/machine.c                |  25 ++++++++
 tools/perf/util/machine.h                |   6 ++
 tools/perf/util/map.c                    |  14 +++++
 tools/perf/util/map.h                    |   7 +++
 tools/perf/util/record.c                 |  10 +++
 tools/perf/util/session.c                |  21 +++++++
 tools/perf/util/symbol.c                 |   7 ++-
 tools/perf/util/tool.h                   |   1 +
 tools/perf/util/trace-event-info.c       |  22 +++----
 tools/perf/util/trace-event-parse.c      |  30 ---------
 tools/perf/util/trace-event-read.c       |  28 ++++-----
 tools/perf/util/trace-event.c            |  45 +++++++++-----
 tools/perf/util/trace-event.h            |   1 +
 33 files changed, 513 insertions(+), 110 deletions(-)

[PATCH 01/18] perf test: Check for refcnt in thread_map test

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

Checking also for refcnt in thread_map test.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1437481927-29538-2-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/tests/thread-map.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/tools/perf/tests/thread-map.c b/tools/perf/tests/thread-map.c
index 5acf000939ea..138a0e3431fa 100644
--- a/tools/perf/tests/thread-map.c
+++ b/tools/perf/tests/thread-map.c
@@ -20,6 +20,8 @@ int test__thread_map(void)
 	TEST_ASSERT_VAL("wrong comm",
 			thread_map__comm(map, 0) &&
 			!strcmp(thread_map__comm(map, 0), "perf"));
+	TEST_ASSERT_VAL("wrong refcnt",
+			atomic_read(&map->refcnt) == 1);
 	thread_map__put(map);
 
 	/* test dummy pid */
@@ -33,6 +35,8 @@ int test__thread_map(void)
 	TEST_ASSERT_VAL("wrong comm",
 			thread_map__comm(map, 0) &&
 			!strcmp(thread_map__comm(map, 0), "dummy"));
+	TEST_ASSERT_VAL("wrong refcnt",
+			atomic_read(&map->refcnt) == 1);
 	thread_map__put(map);
 	return 0;
 }
-- 
2.1.0

[PATCH 02/18] perf evlist: Force perf_evlist__set_maps to propagate maps through events

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

Forcing perf_evlist__set_maps to propagate maps through events, so
cpu/thread maps get set within evlist.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1437481927-29538-11-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evlist.c | 17 +++++++++++++++++
 tools/perf/util/evlist.h | 11 +++--------
 2 files changed, 20 insertions(+), 8 deletions(-)

diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index f7d9c77ee31b..6bfcab9b7108 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -1150,6 +1150,23 @@ out_delete_threads:
 	return -1;
 }
 
+int perf_evlist__set_maps(struct perf_evlist *evlist,
+			  struct cpu_map *cpus,
+			  struct thread_map *threads)
+{
+	if (evlist->cpus)
+		cpu_map__put(evlist->cpus);
+
+	evlist->cpus = cpus;
+
+	if (evlist->threads)
+		thread_map__put(evlist->threads);
+
+	evlist->threads = threads;
+
+	return perf_evlist__propagate_maps(evlist, false);
+}
+
 int perf_evlist__apply_filters(struct perf_evlist *evlist, struct perf_evsel **err_evsel)
 {
 	struct perf_evsel *evsel;
diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h
index 037633c1da9d..406a8216a51e 100644
--- a/tools/perf/util/evlist.h
+++ b/tools/perf/util/evlist.h
@@ -152,14 +152,9 @@ int perf_evlist__enable_event_idx(struct perf_evlist *evlist,
 void perf_evlist__set_selected(struct perf_evlist *evlist,
 			       struct perf_evsel *evsel);
 
-static inline void perf_evlist__set_maps(struct perf_evlist *evlist,
-					 struct cpu_map *cpus,
-					 struct thread_map *threads)
-{
-	evlist->cpus	= cpus;
-	evlist->threads	= threads;
-}
-
+int perf_evlist__set_maps(struct perf_evlist *evlist,
+			  struct cpu_map *cpus,
+			  struct thread_map *threads);
 int perf_evlist__create_maps(struct perf_evlist *evlist, struct target *target);
 int perf_evlist__apply_filters(struct perf_evlist *evlist, struct perf_evsel **err_evsel);
 
-- 
2.1.0

[PATCH 03/18] perf evlist: Use bool instead of target argument in propagate_maps()

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

We need only bool info wether user defined her own set of cpus.

Switching target argument to bool so it could be used from places
without target object defined in following patches.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1437481927-29538-9-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evlist.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index 6bfcab9b7108..80ab942afa8a 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -1102,7 +1102,7 @@ int perf_evlist__mmap(struct perf_evlist *evlist, unsigned int pages,
 }
 
 static int perf_evlist__propagate_maps(struct perf_evlist *evlist,
-				       struct target *target)
+				       bool has_user_cpus)
 {
 	struct perf_evsel *evsel;
 
@@ -1111,10 +1111,10 @@ static int perf_evlist__propagate_maps(struct perf_evlist *evlist,
 		 * We already have cpus for evsel (via PMU sysfs) so
 		 * keep it, if there's no target cpu list defined.
 		 */
-		if (evsel->cpus && target->cpu_list)
+		if (evsel->cpus && has_user_cpus)
 			cpu_map__put(evsel->cpus);
 
-		if (!evsel->cpus || target->cpu_list)
+		if (!evsel->cpus || has_user_cpus)
 			evsel->cpus = cpu_map__get(evlist->cpus);
 
 		evsel->threads = thread_map__get(evlist->threads);
@@ -1142,7 +1142,7 @@ int perf_evlist__create_maps(struct perf_evlist *evlist, struct target *target)
 	if (evlist->cpus == NULL)
 		goto out_delete_threads;
 
-	return perf_evlist__propagate_maps(evlist, target);
+	return perf_evlist__propagate_maps(evlist, !!target->cpu_list);
 
 out_delete_threads:
 	thread_map__put(evlist->threads);
-- 
2.1.0

[PATCH 04/18] perf evlist: Tolerate NULL maps in propagate_maps

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

Tolerating NULL maps in perf_evlist__propagate_maps, so we dont need to
pass evlist with both cpus and threads maps defined.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1437481927-29538-10-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evlist.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index 80ab942afa8a..3b9f411a6b46 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -1119,7 +1119,8 @@ static int perf_evlist__propagate_maps(struct perf_evlist *evlist,
 
 		evsel->threads = thread_map__get(evlist->threads);
 
-		if (!evsel->cpus || !evsel->threads)
+		if ((evlist->cpus && !evsel->cpus) ||
+		    (evlist->threads && !evsel->threads))
 			return -ENOMEM;
 	}
 
-- 
2.1.0

[PATCH 05/18] perf header: Use argv style storage for cmdline feature data

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

We will reuse argv style data in following change to display counters
header showing monitored command line.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1437481927-29538-12-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/header.c  | 35 ++++++++++++++++++++---------------
 tools/perf/util/header.h  |  1 +
 tools/perf/util/session.c |  1 +
 3 files changed, 22 insertions(+), 15 deletions(-)

diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index 03ace57a800c..179b2bdd157d 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -923,17 +923,13 @@ static void print_cmdline(struct perf_header *ph, int fd __maybe_unused,
 			  FILE *fp)
 {
 	int nr, i;
-	char *str;
 
 	nr = ph->env.nr_cmdline;
-	str = ph->env.cmdline;
 
 	fprintf(fp, "# cmdline : ");
 
-	for (i = 0; i < nr; i++) {
-		fprintf(fp, "%s ", str);
-		str += strlen(str) + 1;
-	}
+	for (i = 0; i < nr; i++)
+		fprintf(fp, "%s ", ph->env.cmdline_argv[i]);
 	fputc('\n', fp);
 }
 
@@ -1541,14 +1537,13 @@ process_event_desc(struct perf_file_section *section __maybe_unused,
 	return 0;
 }
 
-static int process_cmdline(struct perf_file_section *section __maybe_unused,
+static int process_cmdline(struct perf_file_section *section,
 			   struct perf_header *ph, int fd,
 			   void *data __maybe_unused)
 {
 	ssize_t ret;
-	char *str;
-	u32 nr, i;
-	struct strbuf sb;
+	char *str, *cmdline = NULL, **argv = NULL;
+	u32 nr, i, len = 0;
 
 	ret = readn(fd, &nr, sizeof(nr));
 	if (ret != sizeof(nr))
@@ -1558,22 +1553,32 @@ static int process_cmdline(struct perf_file_section *section __maybe_unused,
 		nr = bswap_32(nr);
 
 	ph->env.nr_cmdline = nr;
-	strbuf_init(&sb, 128);
+
+	cmdline = zalloc(section->size + nr + 1);
+	if (!cmdline)
+		return -1;
+
+	argv = zalloc(sizeof(char *) * (nr + 1));
+	if (!argv)
+		goto error;
 
 	for (i = 0; i < nr; i++) {
 		str = do_read_string(fd, ph);
 		if (!str)
 			goto error;
 
-		/* include a NULL character at the end */
-		strbuf_add(&sb, str, strlen(str) + 1);
+		argv[i] = cmdline + len;
+		memcpy(argv[i], str, strlen(str) + 1);
+		len += strlen(str) + 1;
 		free(str);
 	}
-	ph->env.cmdline = strbuf_detach(&sb, NULL);
+	ph->env.cmdline = cmdline;
+	ph->env.cmdline_argv = (const char **) argv;
 	return 0;
 
 error:
-	strbuf_release(&sb);
+	free(argv);
+	free(cmdline);
 	return -1;
 }
 
diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h
index d4d57962c591..9b53b6525ce8 100644
--- a/tools/perf/util/header.h
+++ b/tools/perf/util/header.h
@@ -84,6 +84,7 @@ struct perf_session_env {
 	int			nr_pmu_mappings;
 	int			nr_groups;
 	char			*cmdline;
+	const char		**cmdline_argv;
 	char			*sibling_cores;
 	char			*sibling_threads;
 	char			*numa_nodes;
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index ed9dc2555ec7..fb1d525ca3d0 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -180,6 +180,7 @@ static void perf_session_env__delete(struct perf_session_env *env)
 	zfree(&env->cpuid);
 
 	zfree(&env->cmdline);
+	zfree(&env->cmdline_argv);
 	zfree(&env->sibling_cores);
 	zfree(&env->sibling_threads);
 	zfree(&env->numa_nodes);
-- 
2.1.0

[PATCH 06/18] perf symbols: Add front end cache for DSO symbol lookup

[Arnaldo Carvalho de Melo]

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

Acked-by: David Ahern <dsahern@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-voo94tow8wpkcc76mlkny6sc@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/dso.h    | 4 ++++
 tools/perf/util/symbol.c | 7 ++++++-
 2 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/tools/perf/util/dso.h b/tools/perf/util/dso.h
index 2fe98bb0e95b..c73276db6d6f 100644
--- a/tools/perf/util/dso.h
+++ b/tools/perf/util/dso.h
@@ -137,6 +137,10 @@ struct dso {
 	struct rb_node	 rb_node;	/* rbtree node sorted by long name */
 	struct rb_root	 symbols[MAP__NR_TYPES];
 	struct rb_root	 symbol_names[MAP__NR_TYPES];
+	struct {
+		u64		addr;
+		struct symbol	*symbol;
+	} last_find_result[MAP__NR_TYPES];
 	void		 *a2l;
 	char		 *symsrc_filename;
 	unsigned int	 a2l_fails;
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index 45c2e3a8316b..725640fd7cd8 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -444,7 +444,12 @@ static struct symbol *symbols__find_by_name(struct rb_root *symbols,
 struct symbol *dso__find_symbol(struct dso *dso,
 				enum map_type type, u64 addr)
 {
-	return symbols__find(&dso->symbols[type], addr);
+	if (dso->last_find_result[type].addr != addr) {
+		dso->last_find_result[type].addr   = addr;
+		dso->last_find_result[type].symbol = symbols__find(&dso->symbols[type], addr);
+	}
+
+	return dso->last_find_result[type].symbol;
 }
 
 struct symbol *dso__first_symbol(struct dso *dso, enum map_type type)
-- 
2.1.0

[PATCH 07/18] perf symbols: Introduce map__is_(kernel,kmodule)()

[Arnaldo Carvalho de Melo]

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

To, with members we already have, check if a kernel level map is for the
kernel proper or for a module.

Acked-by: David Ahern <dsahern@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-m5ic7h0z2crmtj7vi1a1rj3b@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/map.c | 14 ++++++++++++++
 tools/perf/util/map.h |  7 +++++++
 2 files changed, 21 insertions(+)

diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c
index b5a5e9c02437..ce37e95bc513 100644
--- a/tools/perf/util/map.c
+++ b/tools/perf/util/map.c
@@ -224,6 +224,20 @@ struct map *map__new2(u64 start, struct dso *dso, enum map_type type)
 	return map;
 }
 
+/*
+ * Use this and __map__is_kmodule() for map instances that are in
+ * machine->kmaps, and thus have map->groups->machine all properly set, to
+ * disambiguate between the kernel and modules.
+ *
+ * When the need arises, introduce map__is_{kernel,kmodule)() that
+ * checks (map->groups != NULL && map->groups->machine != NULL &&
+ * map->dso->kernel) before calling __map__is_{kernel,kmodule}())
+ */
+bool __map__is_kernel(const struct map *map)
+{
+	return map->groups->machine->vmlinux_maps[map->type] == map;
+}
+
 static void map__exit(struct map *map)
 {
 	BUG_ON(!RB_EMPTY_NODE(&map->rb_node));
diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h
index d73e687b224e..57829e89b78b 100644
--- a/tools/perf/util/map.h
+++ b/tools/perf/util/map.h
@@ -256,4 +256,11 @@ int map_groups__fixup_overlappings(struct map_groups *mg, struct map *map,
 struct map *map_groups__find_by_name(struct map_groups *mg,
 				     enum map_type type, const char *name);
 
+bool __map__is_kernel(const struct map *map);
+
+static inline bool __map__is_kmodule(const struct map *map)
+{
+	return !__map__is_kernel(map);
+}
+
 #endif /* __PERF_MAP_H */
-- 
2.1.0

[PATCH 08/18] tools lib traceevent: Allow setting an alternative symbol resolver

[Arnaldo Carvalho de Melo]

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

The perf tools have a symbol resolver that includes solving kernel
symbols using either kallsyms or ELF symtabs, and it also is using
libtraceevent to format the trace events fields, including via
subsystem specific plugins, like the "timer" one.

To solve fields like "timer:hrtimer_start"'s "function", libtraceevent
needs a way to map from its value to a function name and addr.

This patch provides a way for tools that already have symbol resolving
facilities to ask libtraceevent to use it when needing to resolve
kernel symbols.

Reviewed-by: Steven Rostedt <rostedt@goodmis.org>
Acked-by: David Ahern <dsahern@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-fdx1fazols17w5py26ia3bwh@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/lib/traceevent/event-parse.c | 68 +++++++++++++++++++++++++++++++++++++-
 tools/lib/traceevent/event-parse.h |  8 +++++
 2 files changed, 75 insertions(+), 1 deletion(-)

diff --git a/tools/lib/traceevent/event-parse.c b/tools/lib/traceevent/event-parse.c
index cc25f059ab3d..fcd8a9e3d2e1 100644
--- a/tools/lib/traceevent/event-parse.c
+++ b/tools/lib/traceevent/event-parse.c
@@ -418,7 +418,7 @@ static int func_map_init(struct pevent *pevent)
 }
 
 static struct func_map *
-find_func(struct pevent *pevent, unsigned long long addr)
+__find_func(struct pevent *pevent, unsigned long long addr)
 {
 	struct func_map *func;
 	struct func_map key;
@@ -434,6 +434,71 @@ find_func(struct pevent *pevent, unsigned long long addr)
 	return func;
 }
 
+struct func_resolver {
+	pevent_func_resolver_t *func;
+	void		       *priv;
+	struct func_map	       map;
+};
+
+/**
+ * pevent_set_function_resolver - set an alternative function resolver
+ * @pevent: handle for the pevent
+ * @resolver: function to be used
+ * @priv: resolver function private state.
+ *
+ * Some tools may have already a way to resolve kernel functions, allow them to
+ * keep using it instead of duplicating all the entries inside
+ * pevent->funclist.
+ */
+int pevent_set_function_resolver(struct pevent *pevent,
+				 pevent_func_resolver_t *func, void *priv)
+{
+	struct func_resolver *resolver = malloc(sizeof(*resolver));
+
+	if (resolver == NULL)
+		return -1;
+
+	resolver->func = func;
+	resolver->priv = priv;
+
+	free(pevent->func_resolver);
+	pevent->func_resolver = resolver;
+
+	return 0;
+}
+
+/**
+ * pevent_reset_function_resolver - reset alternative function resolver
+ * @pevent: handle for the pevent
+ *
+ * Stop using whatever alternative resolver was set, use the default
+ * one instead.
+ */
+void pevent_reset_function_resolver(struct pevent *pevent)
+{
+	free(pevent->func_resolver);
+	pevent->func_resolver = NULL;
+}
+
+static struct func_map *
+find_func(struct pevent *pevent, unsigned long long addr)
+{
+	struct func_map *map;
+
+	if (!pevent->func_resolver)
+		return __find_func(pevent, addr);
+
+	map = &pevent->func_resolver->map;
+	map->mod  = NULL;
+	map->addr = addr;
+	map->func = pevent->func_resolver->func(pevent->func_resolver->priv,
+						&map->addr, &map->mod);
+	if (map->func == NULL)
+		return NULL;
+
+	return map;
+}
+
 /**
  * pevent_find_function - find a function by a given address
  * @pevent: handle for the pevent
@@ -6564,6 +6629,7 @@ void pevent_free(struct pevent *pevent)
 	free(pevent->trace_clock);
 	free(pevent->events);
 	free(pevent->sort_events);
+	free(pevent->func_resolver);
 
 	free(pevent);
 }
diff --git a/tools/lib/traceevent/event-parse.h b/tools/lib/traceevent/event-parse.h
index 063b1971eb35..204befb05a17 100644
--- a/tools/lib/traceevent/event-parse.h
+++ b/tools/lib/traceevent/event-parse.h
@@ -453,6 +453,10 @@ struct cmdline_list;
 struct func_map;
 struct func_list;
 struct event_handler;
+struct func_resolver;
+
+typedef char *(pevent_func_resolver_t)(void *priv,
+				       unsigned long long *addrp, char **modp);
 
 struct pevent {
 	int ref_count;
@@ -481,6 +485,7 @@ struct pevent {
 	int cmdline_count;
 
 	struct func_map *func_map;
+	struct func_resolver *func_resolver;
 	struct func_list *funclist;
 	unsigned int func_count;
 
@@ -611,6 +616,9 @@ enum trace_flag_type {
 	TRACE_FLAG_SOFTIRQ		= 0x10,
 };
 
+int pevent_set_function_resolver(struct pevent *pevent,
+				 pevent_func_resolver_t *func, void *priv);
+void pevent_reset_function_resolver(struct pevent *pevent);
 int pevent_register_comm(struct pevent *pevent, const char *comm, int pid);
 int pevent_register_trace_clock(struct pevent *pevent, const char *trace_clock);
 int pevent_register_function(struct pevent *pevent, char *name,
-- 
2.1.0

[PATCH 09/18] perf symbols: Provide libtraceevent callback to resolve kernel symbols

[Arnaldo Carvalho de Melo]

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

That provides the function signature expected by libtraceevent's
pevent_set_function_resolver().

Acked-by: David Ahern <dsahern@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/n/tip-ie6hvlb6u15y4ulg9j1612zg@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/machine.c     | 14 ++++++++++++++
 tools/perf/util/machine.h     |  4 ++++
 tools/perf/util/trace-event.c | 45 ++++++++++++++++++++++++++++---------------
 tools/perf/util/trace-event.h |  1 +
 4 files changed, 49 insertions(+), 15 deletions(-)

diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index d0bf1e590479..22006c15edf4 100644
--- a/tools/perf/util/machine.c
+++ b/tools/perf/util/machine.c
@@ -1993,3 +1993,17 @@ struct dso *machine__findnew_dso(struct machine *machine, const char *filename)
 {
 	return dsos__findnew(&machine->dsos, filename);
 }
+
+char *machine__resolve_kernel_addr(void *vmachine, unsigned long long *addrp, char **modp)
+{
+	struct machine *machine = vmachine;
+	struct map *map;
+	struct symbol *sym = map_groups__find_symbol(&machine->kmaps, MAP__FUNCTION, *addrp, &map,  NULL);
+
+	if (sym == NULL)
+		return NULL;
+
+	*modp = __map__is_kmodule(map) ? (char *)map->dso->short_name : NULL;
+	*addrp = map->unmap_ip(map, sym->start);
+	return sym->name;
+}
diff --git a/tools/perf/util/machine.h b/tools/perf/util/machine.h
index 887798e511e9..ff5f78c886e0 100644
--- a/tools/perf/util/machine.h
+++ b/tools/perf/util/machine.h
@@ -237,5 +237,9 @@ int machine__synthesize_threads(struct machine *machine, struct target *target,
 pid_t machine__get_current_tid(struct machine *machine, int cpu);
 int machine__set_current_tid(struct machine *machine, int cpu, pid_t pid,
 			     pid_t tid);
+/*
+ * For use with libtraceevent's pevent_set_function_resolver()
+ */
+char *machine__resolve_kernel_addr(void *vmachine, unsigned long long *addrp, char **modp);
 
 #endif /* __PERF_MACHINE_H */
diff --git a/tools/perf/util/trace-event.c b/tools/perf/util/trace-event.c
index 6322d37164c5..667bd109d16f 100644
--- a/tools/perf/util/trace-event.c
+++ b/tools/perf/util/trace-event.c
@@ -9,6 +9,7 @@
 #include <linux/kernel.h>
 #include <traceevent/event-parse.h>
 #include "trace-event.h"
+#include "machine.h"
 #include "util.h"
 
 /*
@@ -19,6 +20,7 @@
  * there.
  */
 static struct trace_event tevent;
+static bool tevent_initialized;
 
 int trace_event__init(struct trace_event *t)
 {
@@ -32,6 +34,32 @@ int trace_event__init(struct trace_event *t)
 	return pevent ? 0 : -1;
 }
 
+static int trace_event__init2(void)
+{
+	int be = traceevent_host_bigendian();
+	struct pevent *pevent;
+
+	if (trace_event__init(&tevent))
+		return -1;
+
+	pevent = tevent.pevent;
+	pevent_set_flag(pevent, PEVENT_NSEC_OUTPUT);
+	pevent_set_file_bigendian(pevent, be);
+	pevent_set_host_bigendian(pevent, be);
+	tevent_initialized = true;
+	return 0;
+}
+
+int trace_event__register_resolver(struct machine *machine)
+{
+	if (!tevent_initialized && trace_event__init2())
+		return -1;
+
+	return pevent_set_function_resolver(tevent.pevent,
+					    machine__resolve_kernel_addr,
+					    machine);
+}
+
 void trace_event__cleanup(struct trace_event *t)
 {
 	traceevent_unload_plugins(t->plugin_list, t->pevent);
@@ -62,21 +90,8 @@ tp_format(const char *sys, const char *name)
 struct event_format*
 trace_event__tp_format(const char *sys, const char *name)
 {
-	static bool initialized;
-
-	if (!initialized) {
-		int be = traceevent_host_bigendian();
-		struct pevent *pevent;
-
-		if (trace_event__init(&tevent))
-			return NULL;
-
-		pevent = tevent.pevent;
-		pevent_set_flag(pevent, PEVENT_NSEC_OUTPUT);
-		pevent_set_file_bigendian(pevent, be);
-		pevent_set_host_bigendian(pevent, be);
-		initialized = true;
-	}
+	if (!tevent_initialized && trace_event__init2())
+		return NULL;
 
 	return tp_format(sys, name);
 }
diff --git a/tools/perf/util/trace-event.h b/tools/perf/util/trace-event.h
index d5168f0be4ec..568128c3284a 100644
--- a/tools/perf/util/trace-event.h
+++ b/tools/perf/util/trace-event.h
@@ -18,6 +18,7 @@ struct trace_event {
 
 int trace_event__init(struct trace_event *t);
 void trace_event__cleanup(struct trace_event *t);
+int trace_event__register_resolver(struct machine *machine);
 struct event_format*
 trace_event__tp_format(const char *sys, const char *name);
 
-- 
2.1.0

[PATCH 10/18] perf trace: Provide libtracevent with a kernel symbol resolver

[Arnaldo Carvalho de Melo]

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

So that beautifiers wanting to resolve kernel function addresses to
names can do its work, now, for instance, the 'timer' tracepoints
beautifiers works with 'perf trace', see the "function=tick..." part:

 # perf trace --event timer:hrtimer_start
<SNIP>
  0.000 timer:hrtimer_start:hrtimer=0xffff88026f3101c0 function=tick_sched_timer/0x0 expires=52098339000000 softexpires=52098339000000)
  0.003 timer:hrtimer_start:hrtimer=0xffff88026f3101c0 function=tick_sched_timer/0x0 expires=52098339000000 softexpires=52098339000000)
<SNIP>

Reported-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: David Ahern <dsahern@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/n/tip-n4i0hxpbl1tnleiqkok47fw2@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-trace.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 32b4d280af4f..282841b10f24 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -1489,6 +1489,9 @@ static int trace__symbols_init(struct trace *trace, struct perf_evlist *evlist)
 	if (trace->host == NULL)
 		return -ENOMEM;
 
+	if (trace_event__register_resolver(trace->host) < 0)
+		return -errno;
+
 	err = __machine__synthesize_threads(trace->host, &trace->tool, &trace->opts.target,
 					    evlist->threads, trace__tool_process, false,
 					    trace->opts.proc_map_timeout);
-- 
2.1.0

[PATCH 11/18] perf script: Switch from perf.data's kallsyms to perf's symbol resolver

[Arnaldo Carvalho de Melo]

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

We were storing a copy of kallsyms inside perf.data file so that we
could resolve kernel addresses to function (start, name, mod) tuples,
but that can be achieved using the symbol resolving routines we have
in symbols.c, and that are used elsewhere in tools/perf.

So, do just like 'perf trace' did and ask libtraceevent to use perf's
symbol resolution routines.

The next step is to just skip whatever kallsyms data is embedded in
older perf.data files and finally to stop storing kallsyms in the perf
data file, as the 20-bytes build-id stored in perf.data's header is
enough to find out the right symtab (be it ELF, kcore, kallsyms, etc) to
use.

Acked-by: David Ahern <dsahern@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/n/tip-d0rtb8tk9j72pz0ehw5fnp24@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-script.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index 24809787369f..2777c6444bc3 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -1830,6 +1830,13 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
 	else
 		symbol_conf.use_callchain = false;
 
+	if (pevent_set_function_resolver(session->tevent.pevent,
+					 machine__resolve_kernel_addr,
+					 &session->machines.host) < 0) {
+		pr_err("%s: failed to set libtraceevent function resolver\n", __func__);
+		return -1;
+	}
+
 	if (generate_script_lang) {
 		struct stat perf_stat;
 		int input;
-- 
2.1.0

Re: [PATCH 11/18] perf script: Switch from perf.data's kallsyms to perf's symbol resolver

[Jiri Olsa]

On Thu, Jul 23, 2015 at 10:58:24PM -0300, Arnaldo Carvalho de Melo wrote:
> From: Arnaldo Carvalho de Melo <acme@redhat.com>
> 
> We were storing a copy of kallsyms inside perf.data file so that we
> could resolve kernel addresses to function (start, name, mod) tuples,
> but that can be achieved using the symbol resolving routines we have
> in symbols.c, and that are used elsewhere in tools/perf.
> 
> So, do just like 'perf trace' did and ask libtraceevent to use perf's
> symbol resolution routines.
> 
> The next step is to just skip whatever kallsyms data is embedded in
> older perf.data files and finally to stop storing kallsyms in the perf
> data file, as the 20-bytes build-id stored in perf.data's header is
> enough to find out the right symtab (be it ELF, kcore, kallsyms, etc) to
> use.
> 
> Acked-by: David Ahern <dsahern@gmail.com>
> Cc: Adrian Hunter <adrian.hunter@intel.com>
> Cc: Borislav Petkov <bp@suse.de>
> Cc: Frederic Weisbecker <fweisbec@gmail.com>
> Cc: Jiri Olsa <jolsa@redhat.com>
> Cc: Namhyung Kim <namhyung@kernel.org>
> Cc: Stephane Eranian <eranian@google.com>
> Cc: Steven Rostedt <rostedt@goodmis.org>
> Link: http://lkml.kernel.org/n/tip-d0rtb8tk9j72pz0ehw5fnp24@git.kernel.org
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
> ---
>  tools/perf/builtin-script.c | 7 +++++++
>  1 file changed, 7 insertions(+)
> 
> diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
> index 24809787369f..2777c6444bc3 100644
> --- a/tools/perf/builtin-script.c
> +++ b/tools/perf/builtin-script.c
> @@ -1830,6 +1830,13 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
>  	else
>  		symbol_conf.use_callchain = false;
>  
> +	if (pevent_set_function_resolver(session->tevent.pevent,
> +					 machine__resolve_kernel_addr,
> +					 &session->machines.host) < 0) {
> +		pr_err("%s: failed to set libtraceevent function resolver\n", __func__);
> +		return -1;
> +	}
> +

got segfault because of this:

    [jolsa@krava perf]$ ./perf record ls
    [ perf record: Woken up 1 times to write data ]
     
    SNIP
     
    /proc/kcore requires CAP_SYS_RAWIO capability to access.
    [ perf record: Captured and wrote 0.016 MB perf.data (7 samples) ]
    [jolsa@krava perf]$ ./perf script
    Segmentation fault (core dumped)
     
    [jolsa@krava perf]$ gdb ./perf
    GNU gdb (GDB) Fedora 7.9.1-17.fc22
    Copyright (C) 2015 Free Software Foundation, Inc.
    License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
    This is free software: you are free to change and redistribute it.
    There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
    and "show warranty" for details.
    This GDB was configured as "x86_64-redhat-linux-gnu".
    Type "show configuration" for configuration details.
    For bug reporting instructions, please see:
    <http://www.gnu.org/software/gdb/bugs/>.
    Find the GDB manual and other documentation resources online at:
    <http://www.gnu.org/software/gdb/documentation/>.
    For help, type "help".
    Type "apropos word" to search for commands related to "word"...
    Reading symbols from ./perf...done.
    (gdb) r script
    Starting program: /home/jolsa/kernel/linux-perf/tools/perf/perf script
    Missing separate debuginfos, use: dnf debuginfo-install glibc-2.21-7.fc22.x86_64
    [Thread debugging using libthread_db enabled]
    Using host libthread_db library "/lib64/libthread_db.so.1".
    Detaching after fork from child process 11230.
     
    Program received signal SIGSEGV, Segmentation fault.
    0x00000000004eeace in pevent_set_function_resolver (pevent=0x0, func=0x4a3197 <machine__resolve_kernel_addr>, priv=0x18e1f48) at event-parse.c:464
    464             free(pevent->func_resolver);
    Missing separate debuginfos, use: dnf debuginfo-install audit-libs-2.4.3-1.fc22.x86_64 bzip2-libs-1.0.6-14.fc22.x86_64 elfutils-libelf-0.163-1.fc22.x86_64 elfutils-libs-0.163-1.fc22.x86_64 libunwind-1.1-10.fc22.x86_64 nss-softokn-freebl-3.19.2-1.0.fc22.x86_64 numactl-libs-2.0.10-2.fc22.x86_64 perl-libs-5.20.2-326.fc22.x86_64 python-libs-2.7.10-4.fc22.x86_64 slang-2.3.0-2.fc22.x86_64 xz-libs-5.2.0-2.fc22.x86_64 zlib-1.2.8-7.fc22.x86_64
    (gdb)


I think 'session->tevent.pevent' gets initialized only for tracepoint, right?

jirka

Re: [PATCH 11/18] perf script: Switch from perf.data's kallsyms to perf's symbol resolver

[Arnaldo Carvalho de Melo]

Em Mon, Aug 03, 2015 at 07:41:13PM +0200, Jiri Olsa escreveu:
>     Missing separate debuginfos, use: dnf debuginfo-install glibc-2.21-7.fc22.x86_64
>     [Thread debugging using libthread_db enabled]
>     Using host libthread_db library "/lib64/libthread_db.so.1".
>     Detaching after fork from child process 11230.
>      
>     Program received signal SIGSEGV, Segmentation fault.
>     0x00000000004eeace in pevent_set_function_resolver (pevent=0x0, func=0x4a3197 <machine__resolve_kernel_addr>, priv=0x18e1f48) at event-parse.c:464
>     464             free(pevent->func_resolver);
>     Missing separate debuginfos, use: dnf debuginfo-install audit-libs-2.4.3-1.fc22.x86_64 bzip2-libs-1.0.6-14.fc22.x86_64 elfutils-libelf-0.163-1.fc22.x86_64 elfutils-libs-0.163-1.fc22.x86_64 libunwind-1.1-10.fc22.x86_64 nss-softokn-freebl-3.19.2-1.0.fc22.x86_64 numactl-libs-2.0.10-2.fc22.x86_64 perl-libs-5.20.2-326.fc22.x86_64 python-libs-2.7.10-4.fc22.x86_64 slang-2.3.0-2.fc22.x86_64 xz-libs-5.2.0-2.fc22.x86_64 zlib-1.2.8-7.fc22.x86_64
>     (gdb)

> I think 'session->tevent.pevent' gets initialized only for tracepoint, right?

Reproduced, checking.

- Arnaldo

Re: [PATCH 11/18] perf script: Switch from perf.data's kallsyms to perf's symbol resolver

[Arnaldo Carvalho de Melo]

Em Mon, Aug 03, 2015 at 04:00:09PM -0300, Arnaldo Carvalho de Melo escreveu:
> Em Mon, Aug 03, 2015 at 07:41:13PM +0200, Jiri Olsa escreveu:
> >     Missing separate debuginfos, use: dnf debuginfo-install glibc-2.21-7.fc22.x86_64
> >     [Thread debugging using libthread_db enabled]
> >     Using host libthread_db library "/lib64/libthread_db.so.1".
> >     Detaching after fork from child process 11230.
> >      
> >     Program received signal SIGSEGV, Segmentation fault.
> >     0x00000000004eeace in pevent_set_function_resolver (pevent=0x0, func=0x4a3197 <machine__resolve_kernel_addr>, priv=0x18e1f48) at event-parse.c:464
> >     464             free(pevent->func_resolver);
> >     Missing separate debuginfos, use: dnf debuginfo-install audit-libs-2.4.3-1.fc22.x86_64 bzip2-libs-1.0.6-14.fc22.x86_64 elfutils-libelf-0.163-1.fc22.x86_64 elfutils-libs-0.163-1.fc22.x86_64 libunwind-1.1-10.fc22.x86_64 nss-softokn-freebl-3.19.2-1.0.fc22.x86_64 numactl-libs-2.0.10-2.fc22.x86_64 perl-libs-5.20.2-326.fc22.x86_64 python-libs-2.7.10-4.fc22.x86_64 slang-2.3.0-2.fc22.x86_64 xz-libs-5.2.0-2.fc22.x86_64 zlib-1.2.8-7.fc22.x86_64
> >     (gdb)
> 
> > I think 'session->tevent.pevent' gets initialized only for tracepoint, right?
> 
> Reproduced, checking.

Yeah, I think we only need to set libtraceevent's function resolver if
there are tracepoints, i.e. if libtraceevent is being used. Following
patch should be enough, right?

- Arnaldo

diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index bd31380122f9..7912feb9a024 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -1861,7 +1861,8 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
 	else
 		symbol_conf.use_callchain = false;
 
-	if (pevent_set_function_resolver(session->tevent.pevent,
+	if (session->tevent.pevent &&
+	    pevent_set_function_resolver(session->tevent.pevent,
 					 machine__resolve_kernel_addr,
 					 &session->machines.host) < 0) {
 		pr_err("%s: failed to set libtraceevent function resolver\n", __func__);

Re: [PATCH 11/18] perf script: Switch from perf.data's kallsyms to perf's symbol resolver

[Jiri Olsa]

On Mon, Aug 03, 2015 at 04:27:59PM -0300, Arnaldo Carvalho de Melo wrote:
> Em Mon, Aug 03, 2015 at 04:00:09PM -0300, Arnaldo Carvalho de Melo escreveu:
> > Em Mon, Aug 03, 2015 at 07:41:13PM +0200, Jiri Olsa escreveu:
> > >     Missing separate debuginfos, use: dnf debuginfo-install glibc-2.21-7.fc22.x86_64
> > >     [Thread debugging using libthread_db enabled]
> > >     Using host libthread_db library "/lib64/libthread_db.so.1".
> > >     Detaching after fork from child process 11230.
> > >      
> > >     Program received signal SIGSEGV, Segmentation fault.
> > >     0x00000000004eeace in pevent_set_function_resolver (pevent=0x0, func=0x4a3197 <machine__resolve_kernel_addr>, priv=0x18e1f48) at event-parse.c:464
> > >     464             free(pevent->func_resolver);
> > >     Missing separate debuginfos, use: dnf debuginfo-install audit-libs-2.4.3-1.fc22.x86_64 bzip2-libs-1.0.6-14.fc22.x86_64 elfutils-libelf-0.163-1.fc22.x86_64 elfutils-libs-0.163-1.fc22.x86_64 libunwind-1.1-10.fc22.x86_64 nss-softokn-freebl-3.19.2-1.0.fc22.x86_64 numactl-libs-2.0.10-2.fc22.x86_64 perl-libs-5.20.2-326.fc22.x86_64 python-libs-2.7.10-4.fc22.x86_64 slang-2.3.0-2.fc22.x86_64 xz-libs-5.2.0-2.fc22.x86_64 zlib-1.2.8-7.fc22.x86_64
> > >     (gdb)
> > 
> > > I think 'session->tevent.pevent' gets initialized only for tracepoint, right?
> > 
> > Reproduced, checking.
> 
> Yeah, I think we only need to set libtraceevent's function resolver if
> there are tracepoints, i.e. if libtraceevent is being used. Following
> patch should be enough, right?
> 
> - Arnaldo
> 
> diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
> index bd31380122f9..7912feb9a024 100644
> --- a/tools/perf/builtin-script.c
> +++ b/tools/perf/builtin-script.c
> @@ -1861,7 +1861,8 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
>  	else
>  		symbol_conf.use_callchain = false;
>  
> -	if (pevent_set_function_resolver(session->tevent.pevent,
> +	if (session->tevent.pevent &&
> +	    pevent_set_function_resolver(session->tevent.pevent,
>  					 machine__resolve_kernel_addr,
>  					 &session->machines.host) < 0) {
>  		pr_err("%s: failed to set libtraceevent function resolver\n", __func__);

yep, works nicely

thanks,
jirka

[PATCH 12/18] perf tools: Stop reading the kallsyms data from perf.data

[Arnaldo Carvalho de Melo]

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

As it is not used anymore, since 'perf script' switched to asking
libtraceevent to use tools/perf's symbol resolution routines.

Acked-by: David Ahern <dsahern@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/n/tip-4ilhofz4b7o8yokvutjt9yzz@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/trace-event-parse.c | 30 ------------------------------
 tools/perf/util/trace-event-read.c  | 28 +++++++++++++---------------
 2 files changed, 13 insertions(+), 45 deletions(-)

diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c
index d4957418657e..8ff7d620d942 100644
--- a/tools/perf/util/trace-event-parse.c
+++ b/tools/perf/util/trace-event-parse.c
@@ -135,36 +135,6 @@ void event_format__print(struct event_format *event,
 	return event_format__fprintf(event, cpu, data, size, stdout);
 }
 
-void parse_proc_kallsyms(struct pevent *pevent,
-			 char *file, unsigned int size __maybe_unused)
-{
-	unsigned long long addr;
-	char *func;
-	char *line;
-	char *next = NULL;
-	char *addr_str;
-	char *mod;
-	char *fmt = NULL;
-
-	line = strtok_r(file, "\n", &next);
-	while (line) {
-		mod = NULL;
-		addr_str = strtok_r(line, " ", &fmt);
-		addr = strtoull(addr_str, NULL, 16);
-		/* skip character */
-		strtok_r(NULL, " ", &fmt);
-		func = strtok_r(NULL, "\t", &fmt);
-		mod = strtok_r(NULL, "]", &fmt);
-		/* truncate the extra '[' */
-		if (mod)
-			mod = mod + 1;
-
-		pevent_register_function(pevent, func, addr, mod);
-
-		line = strtok_r(NULL, "\n", &next);
-	}
-}
-
 void parse_ftrace_printk(struct pevent *pevent,
 			 char *file, unsigned int size __maybe_unused)
 {
diff --git a/tools/perf/util/trace-event-read.c b/tools/perf/util/trace-event-read.c
index 54d9e9b548a8..b67a0ccf5ab9 100644
--- a/tools/perf/util/trace-event-read.c
+++ b/tools/perf/util/trace-event-read.c
@@ -162,25 +162,23 @@ out:
 static int read_proc_kallsyms(struct pevent *pevent)
 {
 	unsigned int size;
-	char *buf;
 
 	size = read4(pevent);
 	if (!size)
 		return 0;
-
-	buf = malloc(size + 1);
-	if (buf == NULL)
-		return -1;
-
-	if (do_read(buf, size) < 0) {
-		free(buf);
-		return -1;
-	}
-	buf[size] = '\0';
-
-	parse_proc_kallsyms(pevent, buf, size);
-
-	free(buf);
+	/*
+	 * Just skip it, now that we configure libtraceevent to use the
+	 * tools/perf/ symbol resolver.
+	 *
+	 * We need to skip it so that we can continue parsing old perf.data
+	 * files, that contains this /proc/kallsyms payload.
+	 *
+	 * Newer perf.data files will have just the 4-bytes zeros "kallsyms
+	 * payload", so that older tools can continue reading it and interpret
+	 * it as "no kallsyms payload is present".
+	 */
+	lseek(input_fd, size, SEEK_CUR);
+	trace_data_size += size;
 	return 0;
 }
 
-- 
2.1.0

[PATCH 13/18] perf tools: Stop copying kallsyms into the perf.data file header

[Arnaldo Carvalho de Melo]

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

Since we now ask libtraceevent, the only user of this payload, to use
perf's symbol resolution routines, there is no need to carry about
~4.5MB per perf.data when we can get it from one of the places the perf
symbol resolution looks for that symtab (debuginfo, ~/.debug/,
/proc/kallsyms, --symfs, etc), using the kernel and modules build-ids to
make sure the right table is used.

Acked-by: David Ahern <dsahern@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/n/tip-h89ituf9rso2rv1v7kjrbeda@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/trace-event-info.c | 22 ++++++++--------------
 1 file changed, 8 insertions(+), 14 deletions(-)

diff --git a/tools/perf/util/trace-event-info.c b/tools/perf/util/trace-event-info.c
index eb72716017ac..22245986e59e 100644
--- a/tools/perf/util/trace-event-info.c
+++ b/tools/perf/util/trace-event-info.c
@@ -341,20 +341,14 @@ out:
 
 static int record_proc_kallsyms(void)
 {
-	unsigned int size;
-	const char *path = "/proc/kallsyms";
-	struct stat st;
-	int ret, err = 0;
-
-	ret = stat(path, &st);
-	if (ret < 0) {
-		/* not found */
-		size = 0;
-		if (write(output_fd, &size, 4) != 4)
-			err = -EIO;
-		return err;
-	}
-	return record_file(path, 4);
+	unsigned long long size = 0;
+	/*
+	 * Just to keep older perf.data file parsers happy, record a zero
+	 * sized kallsyms file, i.e. do the same thing that was done when
+	 * /proc/kallsyms (or something specified via --kallsyms, in a
+	 * different path) couldn't be read.
+	 */
+	return write(output_fd, &size, 4) != 4 ? -EIO : 0;
 }
 
 static int record_ftrace_printk(void)
-- 
2.1.0

[PATCH 14/18] perf: Add PERF_RECORD_SWITCH to indicate context switches

[Arnaldo Carvalho de Melo]

From: Adrian Hunter <adrian.hunter@intel.com>

There are already two events for context switches, namely the tracepoint
sched:sched_switch and the software event context_switches.
Unfortunately neither are suitable for use by non-privileged users for
the purpose of synchronizing hardware trace data (e.g. Intel PT) to the
context switch.

Tracepoints are no good at all for non-privileged users because they
need either CAP_SYS_ADMIN or /proc/sys/kernel/perf_event_paranoid <= -1.

On the other hand, kernel software events need either CAP_SYS_ADMIN or
/proc/sys/kernel/perf_event_paranoid <= 1.

Now many distributions do default perf_event_paranoid to 1 making
context_switches a contender, except it has another problem (which is
also shared with sched:sched_switch) which is that it happens before
perf schedules events out instead of after perf schedules events in.
Whereas a privileged user can see all the events anyway, a
non-privileged user only sees events for their own processes, in other
words they see when their process was scheduled out not when it was
scheduled in. That presents two problems to use the event:

1. the information comes too late, so tools have to look ahead in the
   event stream to find out what the current state is

2. if they are unlucky tracing might have stopped before the
   context-switches event is recorded.

This new PERF_RECORD_SWITCH event does not have those problems
and it also has a couple of other small advantages.

It is easier to use because it is an auxiliary event (like mmap, comm
and task events) which can be enabled by setting a single bit. It is
smaller than sched:sched_switch and easier to parse.

To make the event useful for privileged users also, if the
context is cpu-wide then the event record will be
PERF_RECORD_SWITCH_CPU_WIDE which is the same as
PERF_RECORD_SWITCH except it also provides the next or
previous pid/tid.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Jiri Olsa <jolsa@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Pawel Moll <pawel.moll@arm.com>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1437471846-26995-2-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 include/uapi/linux/perf_event.h |  31 +++++++++++-
 kernel/events/core.c            | 103 ++++++++++++++++++++++++++++++++++++++++
 2 files changed, 133 insertions(+), 1 deletion(-)

diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h
index d97f84c080da..022d0acf7df0 100644
--- a/include/uapi/linux/perf_event.h
+++ b/include/uapi/linux/perf_event.h
@@ -330,7 +330,8 @@ struct perf_event_attr {
 				mmap2          :  1, /* include mmap with inode data     */
 				comm_exec      :  1, /* flag comm events that are due to an exec */
 				use_clockid    :  1, /* use @clockid for time fields */
-				__reserved_1   : 38;
+				context_switch :  1, /* context switch data */
+				__reserved_1   : 37;
 
 	union {
 		__u32		wakeup_events;	  /* wakeup every n events */
@@ -572,9 +573,11 @@ struct perf_event_mmap_page {
 /*
  * PERF_RECORD_MISC_MMAP_DATA and PERF_RECORD_MISC_COMM_EXEC are used on
  * different events so can reuse the same bit position.
+ * Ditto PERF_RECORD_MISC_SWITCH_OUT.
  */
 #define PERF_RECORD_MISC_MMAP_DATA		(1 << 13)
 #define PERF_RECORD_MISC_COMM_EXEC		(1 << 13)
+#define PERF_RECORD_MISC_SWITCH_OUT		(1 << 13)
 /*
  * Indicates that the content of PERF_SAMPLE_IP points to
  * the actual instruction that triggered the event. See also
@@ -818,6 +821,32 @@ enum perf_event_type {
 	 */
 	PERF_RECORD_LOST_SAMPLES		= 13,
 
+	/*
+	 * Records a context switch in or out (flagged by
+	 * PERF_RECORD_MISC_SWITCH_OUT). See also
+	 * PERF_RECORD_SWITCH_CPU_WIDE.
+	 *
+	 * struct {
+	 *	struct perf_event_header	header;
+	 *	struct sample_id		sample_id;
+	 * };
+	 */
+	PERF_RECORD_SWITCH			= 14,
+
+	/*
+	 * CPU-wide version of PERF_RECORD_SWITCH with next_prev_pid and
+	 * next_prev_tid that are the next (switching out) or previous
+	 * (switching in) pid/tid.
+	 *
+	 * struct {
+	 *	struct perf_event_header	header;
+	 *	u32				next_prev_pid;
+	 *	u32				next_prev_tid;
+	 *	struct sample_id		sample_id;
+	 * };
+	 */
+	PERF_RECORD_SWITCH_CPU_WIDE		= 15,
+
 	PERF_RECORD_MAX,			/* non-ABI */
 };
 
diff --git a/kernel/events/core.c b/kernel/events/core.c
index d3dae3419b99..ce21143c0d9e 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -163,6 +163,7 @@ static atomic_t nr_mmap_events __read_mostly;
 static atomic_t nr_comm_events __read_mostly;
 static atomic_t nr_task_events __read_mostly;
 static atomic_t nr_freq_events __read_mostly;
+static atomic_t nr_switch_events __read_mostly;
 
 static LIST_HEAD(pmus);
 static DEFINE_MUTEX(pmus_lock);
@@ -2619,6 +2620,9 @@ static void perf_pmu_sched_task(struct task_struct *prev,
 	local_irq_restore(flags);
 }
 
+static void perf_event_switch(struct task_struct *task,
+			      struct task_struct *next_prev, bool sched_in);
+
 #define for_each_task_context_nr(ctxn)					\
 	for ((ctxn) = 0; (ctxn) < perf_nr_task_contexts; (ctxn)++)
 
@@ -2641,6 +2645,9 @@ void __perf_event_task_sched_out(struct task_struct *task,
 	if (__this_cpu_read(perf_sched_cb_usages))
 		perf_pmu_sched_task(task, next, false);
 
+	if (atomic_read(&nr_switch_events))
+		perf_event_switch(task, next, false);
+
 	for_each_task_context_nr(ctxn)
 		perf_event_context_sched_out(task, ctxn, next);
 
@@ -2831,6 +2838,9 @@ void __perf_event_task_sched_in(struct task_struct *prev,
 	if (atomic_read(this_cpu_ptr(&perf_cgroup_events)))
 		perf_cgroup_sched_in(prev, task);
 
+	if (atomic_read(&nr_switch_events))
+		perf_event_switch(task, prev, true);
+
 	if (__this_cpu_read(perf_sched_cb_usages))
 		perf_pmu_sched_task(prev, task, true);
 }
@@ -3454,6 +3464,10 @@ static void unaccount_event(struct perf_event *event)
 		atomic_dec(&nr_task_events);
 	if (event->attr.freq)
 		atomic_dec(&nr_freq_events);
+	if (event->attr.context_switch) {
+		static_key_slow_dec_deferred(&perf_sched_events);
+		atomic_dec(&nr_switch_events);
+	}
 	if (is_cgroup_event(event))
 		static_key_slow_dec_deferred(&perf_sched_events);
 	if (has_branch_stack(event))
@@ -5982,6 +5996,91 @@ void perf_log_lost_samples(struct perf_event *event, u64 lost)
 }
 
 /*
+ * context_switch tracking
+ */
+
+struct perf_switch_event {
+	struct task_struct	*task;
+	struct task_struct	*next_prev;
+
+	struct {
+		struct perf_event_header	header;
+		u32				next_prev_pid;
+		u32				next_prev_tid;
+	} event_id;
+};
+
+static int perf_event_switch_match(struct perf_event *event)
+{
+	return event->attr.context_switch;
+}
+
+static void perf_event_switch_output(struct perf_event *event, void *data)
+{
+	struct perf_switch_event *se = data;
+	struct perf_output_handle handle;
+	struct perf_sample_data sample;
+	int ret;
+
+	if (!perf_event_switch_match(event))
+		return;
+
+	/* Only CPU-wide events are allowed to see next/prev pid/tid */
+	if (event->ctx->task) {
+		se->event_id.header.type = PERF_RECORD_SWITCH;
+		se->event_id.header.size = sizeof(se->event_id.header);
+	} else {
+		se->event_id.header.type = PERF_RECORD_SWITCH_CPU_WIDE;
+		se->event_id.header.size = sizeof(se->event_id);
+		se->event_id.next_prev_pid =
+					perf_event_pid(event, se->next_prev);
+		se->event_id.next_prev_tid =
+					perf_event_tid(event, se->next_prev);
+	}
+
+	perf_event_header__init_id(&se->event_id.header, &sample, event);
+
+	ret = perf_output_begin(&handle, event, se->event_id.header.size);
+	if (ret)
+		return;
+
+	if (event->ctx->task)
+		perf_output_put(&handle, se->event_id.header);
+	else
+		perf_output_put(&handle, se->event_id);
+
+	perf_event__output_id_sample(event, &handle, &sample);
+
+	perf_output_end(&handle);
+}
+
+static void perf_event_switch(struct task_struct *task,
+			      struct task_struct *next_prev, bool sched_in)
+{
+	struct perf_switch_event switch_event;
+
+	/* N.B. caller checks nr_switch_events != 0 */
+
+	switch_event = (struct perf_switch_event){
+		.task		= task,
+		.next_prev	= next_prev,
+		.event_id	= {
+			.header = {
+				/* .type */
+				.misc = sched_in ? 0 : PERF_RECORD_MISC_SWITCH_OUT,
+				/* .size */
+			},
+			/* .next_prev_pid */
+			/* .next_prev_tid */
+		},
+	};
+
+	perf_event_aux(perf_event_switch_output,
+		       &switch_event,
+		       NULL);
+}
+
+/*
  * IRQ throttle logging
  */
 
@@ -7479,6 +7578,10 @@ static void account_event(struct perf_event *event)
 		if (atomic_inc_return(&nr_freq_events) == 1)
 			tick_nohz_full_kick_all();
 	}
+	if (event->attr.context_switch) {
+		atomic_inc(&nr_switch_events);
+		static_key_slow_inc(&perf_sched_events.key);
+	}
 	if (has_branch_stack(event))
 		static_key_slow_inc(&perf_sched_events.key);
 	if (is_cgroup_event(event))
-- 
2.1.0

[PATCH 15/18] perf tools: Add new PERF_RECORD_SWITCH event

[Arnaldo Carvalho de Melo]

From: Adrian Hunter <adrian.hunter@intel.com>

Support processing of PERF_RECORD_SWITCH events and
PERF_RECORD_SWITCH_CPU_WIDE events. There is a single
tools callback for them both so that the tool must
check the event type before using the extra members
in PERF_RECORD_SWITCH_CPU_WIDE.

There is still no way to select the events, though.
That is added in a subsequest patch.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Jiri Olsa <jolsa@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Pawel Moll <pawel.moll@arm.com>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1437471846-26995-3-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-inject.c |  1 +
 tools/perf/util/event.c     | 28 ++++++++++++++++++++++++++++
 tools/perf/util/event.h     | 12 ++++++++++++
 tools/perf/util/evsel.c     |  1 +
 tools/perf/util/machine.c   | 11 +++++++++++
 tools/perf/util/machine.h   |  2 ++
 tools/perf/util/session.c   | 20 ++++++++++++++++++++
 tools/perf/util/tool.h      |  1 +
 8 files changed, 76 insertions(+)

diff --git a/tools/perf/builtin-inject.c b/tools/perf/builtin-inject.c
index 01b06492bd6a..f62c49b35be0 100644
--- a/tools/perf/builtin-inject.c
+++ b/tools/perf/builtin-inject.c
@@ -561,6 +561,7 @@ int cmd_inject(int argc, const char **argv, const char *prefix __maybe_unused)
 			.lost		= perf_event__repipe,
 			.aux		= perf_event__repipe,
 			.itrace_start	= perf_event__repipe,
+			.context_switch	= perf_event__repipe,
 			.read		= perf_event__repipe_sample,
 			.throttle	= perf_event__repipe,
 			.unthrottle	= perf_event__repipe,
diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c
index 67a977e5d0ab..7ff61274ed57 100644
--- a/tools/perf/util/event.c
+++ b/tools/perf/util/event.c
@@ -26,6 +26,8 @@ static const char *perf_event__names[] = {
 	[PERF_RECORD_AUX]			= "AUX",
 	[PERF_RECORD_ITRACE_START]		= "ITRACE_START",
 	[PERF_RECORD_LOST_SAMPLES]		= "LOST_SAMPLES",
+	[PERF_RECORD_SWITCH]			= "SWITCH",
+	[PERF_RECORD_SWITCH_CPU_WIDE]		= "SWITCH_CPU_WIDE",
 	[PERF_RECORD_HEADER_ATTR]		= "ATTR",
 	[PERF_RECORD_HEADER_EVENT_TYPE]		= "EVENT_TYPE",
 	[PERF_RECORD_HEADER_TRACING_DATA]	= "TRACING_DATA",
@@ -749,6 +751,14 @@ int perf_event__process_lost_samples(struct perf_tool *tool __maybe_unused,
 	return machine__process_lost_samples_event(machine, event, sample);
 }
 
+int perf_event__process_switch(struct perf_tool *tool __maybe_unused,
+			       union perf_event *event,
+			       struct perf_sample *sample __maybe_unused,
+			       struct machine *machine)
+{
+	return machine__process_switch_event(machine, event);
+}
+
 size_t perf_event__fprintf_mmap(union perf_event *event, FILE *fp)
 {
 	return fprintf(fp, " %d/%d: [%#" PRIx64 "(%#" PRIx64 ") @ %#" PRIx64 "]: %c %s\n",
@@ -827,6 +837,20 @@ size_t perf_event__fprintf_itrace_start(union perf_event *event, FILE *fp)
 		       event->itrace_start.pid, event->itrace_start.tid);
 }
 
+size_t perf_event__fprintf_switch(union perf_event *event, FILE *fp)
+{
+	bool out = event->header.misc & PERF_RECORD_MISC_SWITCH_OUT;
+	const char *in_out = out ? "OUT" : "IN ";
+
+	if (event->header.type == PERF_RECORD_SWITCH)
+		return fprintf(fp, " %s\n", in_out);
+
+	return fprintf(fp, " %s  %s pid/tid: %5u/%-5u\n",
+		       in_out, out ? "next" : "prev",
+		       event->context_switch.next_prev_pid,
+		       event->context_switch.next_prev_tid);
+}
+
 size_t perf_event__fprintf(union perf_event *event, FILE *fp)
 {
 	size_t ret = fprintf(fp, "PERF_RECORD_%s",
@@ -852,6 +876,10 @@ size_t perf_event__fprintf(union perf_event *event, FILE *fp)
 	case PERF_RECORD_ITRACE_START:
 		ret += perf_event__fprintf_itrace_start(event, fp);
 		break;
+	case PERF_RECORD_SWITCH:
+	case PERF_RECORD_SWITCH_CPU_WIDE:
+		ret += perf_event__fprintf_switch(event, fp);
+		break;
 	default:
 		ret += fprintf(fp, "\n");
 	}
diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h
index c53f36384b64..4bb2ae894c78 100644
--- a/tools/perf/util/event.h
+++ b/tools/perf/util/event.h
@@ -348,6 +348,12 @@ struct itrace_start_event {
 	u32 pid, tid;
 };
 
+struct context_switch_event {
+	struct perf_event_header header;
+	u32 next_prev_pid;
+	u32 next_prev_tid;
+};
+
 union perf_event {
 	struct perf_event_header	header;
 	struct mmap_event		mmap;
@@ -369,6 +375,7 @@ union perf_event {
 	struct auxtrace_error_event	auxtrace_error;
 	struct aux_event		aux;
 	struct itrace_start_event	itrace_start;
+	struct context_switch_event	context_switch;
 };
 
 void perf_event__print_totals(void);
@@ -418,6 +425,10 @@ int perf_event__process_itrace_start(struct perf_tool *tool,
 				     union perf_event *event,
 				     struct perf_sample *sample,
 				     struct machine *machine);
+int perf_event__process_switch(struct perf_tool *tool,
+			       union perf_event *event,
+			       struct perf_sample *sample,
+			       struct machine *machine);
 int perf_event__process_mmap(struct perf_tool *tool,
 			     union perf_event *event,
 			     struct perf_sample *sample,
@@ -480,6 +491,7 @@ size_t perf_event__fprintf_mmap2(union perf_event *event, FILE *fp);
 size_t perf_event__fprintf_task(union perf_event *event, FILE *fp);
 size_t perf_event__fprintf_aux(union perf_event *event, FILE *fp);
 size_t perf_event__fprintf_itrace_start(union perf_event *event, FILE *fp);
+size_t perf_event__fprintf_switch(union perf_event *event, FILE *fp);
 size_t perf_event__fprintf(union perf_event *event, FILE *fp);
 
 u64 kallsyms__get_function_start(const char *kallsyms_filename,
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 49fb7b5feb09..9e6e6f40b787 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -1127,6 +1127,7 @@ int perf_event_attr__fprintf(FILE *fp, struct perf_event_attr *attr,
 	PRINT_ATTRf(mmap2, p_unsigned);
 	PRINT_ATTRf(comm_exec, p_unsigned);
 	PRINT_ATTRf(use_clockid, p_unsigned);
+	PRINT_ATTRf(context_switch, p_unsigned);
 
 	PRINT_ATTRn("{ wakeup_events, wakeup_watermark }", wakeup_events, p_unsigned);
 	PRINT_ATTRf(bp_type, p_unsigned);
diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index 22006c15edf4..be3e00891d22 100644
--- a/tools/perf/util/machine.c
+++ b/tools/perf/util/machine.c
@@ -550,6 +550,14 @@ int machine__process_itrace_start_event(struct machine *machine __maybe_unused,
 	return 0;
 }
 
+int machine__process_switch_event(struct machine *machine __maybe_unused,
+				  union perf_event *event)
+{
+	if (dump_trace)
+		perf_event__fprintf_switch(event, stdout);
+	return 0;
+}
+
 struct map *machine__findnew_module_map(struct machine *machine, u64 start,
 					const char *filename)
 {
@@ -1451,6 +1459,9 @@ int machine__process_event(struct machine *machine, union perf_event *event,
 		ret = machine__process_itrace_start_event(machine, event); break;
 	case PERF_RECORD_LOST_SAMPLES:
 		ret = machine__process_lost_samples_event(machine, event, sample); break;
+	case PERF_RECORD_SWITCH:
+	case PERF_RECORD_SWITCH_CPU_WIDE:
+		ret = machine__process_switch_event(machine, event); break;
 	default:
 		ret = -1;
 		break;
diff --git a/tools/perf/util/machine.h b/tools/perf/util/machine.h
index ff5f78c886e0..ea5cb4a621db 100644
--- a/tools/perf/util/machine.h
+++ b/tools/perf/util/machine.h
@@ -87,6 +87,8 @@ int machine__process_aux_event(struct machine *machine,
 			       union perf_event *event);
 int machine__process_itrace_start_event(struct machine *machine,
 					union perf_event *event);
+int machine__process_switch_event(struct machine *machine __maybe_unused,
+				  union perf_event *event);
 int machine__process_mmap_event(struct machine *machine, union perf_event *event,
 				struct perf_sample *sample);
 int machine__process_mmap2_event(struct machine *machine, union perf_event *event,
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index fb1d525ca3d0..2d9574710543 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -333,6 +333,8 @@ void perf_tool__fill_defaults(struct perf_tool *tool)
 		tool->aux = perf_event__process_aux;
 	if (tool->itrace_start == NULL)
 		tool->itrace_start = perf_event__process_itrace_start;
+	if (tool->context_switch == NULL)
+		tool->context_switch = perf_event__process_switch;
 	if (tool->read == NULL)
 		tool->read = process_event_sample_stub;
 	if (tool->throttle == NULL)
@@ -471,6 +473,19 @@ static void perf_event__itrace_start_swap(union perf_event *event,
 		swap_sample_id_all(event, &event->itrace_start + 1);
 }
 
+static void perf_event__switch_swap(union perf_event *event, bool sample_id_all)
+{
+	if (event->header.type == PERF_RECORD_SWITCH_CPU_WIDE) {
+		event->context_switch.next_prev_pid =
+				bswap_32(event->context_switch.next_prev_pid);
+		event->context_switch.next_prev_tid =
+				bswap_32(event->context_switch.next_prev_tid);
+	}
+
+	if (sample_id_all)
+		swap_sample_id_all(event, &event->context_switch + 1);
+}
+
 static void perf_event__throttle_swap(union perf_event *event,
 				      bool sample_id_all)
 {
@@ -633,6 +648,8 @@ static perf_event__swap_op perf_event__swap_ops[] = {
 	[PERF_RECORD_AUX]		  = perf_event__aux_swap,
 	[PERF_RECORD_ITRACE_START]	  = perf_event__itrace_start_swap,
 	[PERF_RECORD_LOST_SAMPLES]	  = perf_event__all64_swap,
+	[PERF_RECORD_SWITCH]		  = perf_event__switch_swap,
+	[PERF_RECORD_SWITCH_CPU_WIDE]	  = perf_event__switch_swap,
 	[PERF_RECORD_HEADER_ATTR]	  = perf_event__hdr_attr_swap,
 	[PERF_RECORD_HEADER_EVENT_TYPE]	  = perf_event__event_type_swap,
 	[PERF_RECORD_HEADER_TRACING_DATA] = perf_event__tracing_data_swap,
@@ -1094,6 +1111,9 @@ static int machines__deliver_event(struct machines *machines,
 		return tool->aux(tool, event, sample, machine);
 	case PERF_RECORD_ITRACE_START:
 		return tool->itrace_start(tool, event, sample, machine);
+	case PERF_RECORD_SWITCH:
+	case PERF_RECORD_SWITCH_CPU_WIDE:
+		return tool->context_switch(tool, event, sample, machine);
 	default:
 		++evlist->stats.nr_unknown_events;
 		return -1;
diff --git a/tools/perf/util/tool.h b/tools/perf/util/tool.h
index c307dd438286..cab8cc24831b 100644
--- a/tools/perf/util/tool.h
+++ b/tools/perf/util/tool.h
@@ -46,6 +46,7 @@ struct perf_tool {
 			lost_samples,
 			aux,
 			itrace_start,
+			context_switch,
 			throttle,
 			unthrottle;
 	event_attr_op	attr;
-- 
2.1.0

[PATCH 16/18] perf record: Add option --switch-events to select PERF_RECORD_SWITCH events

[Arnaldo Carvalho de Melo]

From: Adrian Hunter <adrian.hunter@intel.com>

Add an option to select PERF_RECORD_SWITCH events.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Tested-by: Jiri Olsa <jolsa@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Pawel Moll <pawel.moll@arm.com>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1437471846-26995-4-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-record.txt |  4 ++++
 tools/perf/builtin-record.c              |  7 +++++++
 tools/perf/perf.h                        |  1 +
 tools/perf/util/evlist.h                 |  1 +
 tools/perf/util/evsel.c                  |  3 +++
 tools/perf/util/record.c                 | 10 ++++++++++
 6 files changed, 26 insertions(+)

diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
index 29e5307945bf..63ee0408761d 100644
--- a/tools/perf/Documentation/perf-record.txt
+++ b/tools/perf/Documentation/perf-record.txt
@@ -293,6 +293,10 @@ When processing pre-existing threads /proc/XXX/mmap, it may take a long time,
 because the file may be huge. A time out is needed in such cases.
 This option sets the time out limit. The default value is 500 ms.
 
+--switch-events::
+Record context switch events i.e. events of type PERF_RECORD_SWITCH or
+PERF_RECORD_SWITCH_CPU_WIDE.
+
 SEE ALSO
 --------
 linkperf:perf-stat[1], linkperf:perf-list[1]
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 1932e27c00d8..445a64d19625 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -1075,6 +1075,8 @@ struct option __record_options[] = {
 			  "opts", "AUX area tracing Snapshot Mode", ""),
 	OPT_UINTEGER(0, "proc-map-timeout", &record.opts.proc_map_timeout,
 			"per thread proc mmap processing timeout in ms"),
+	OPT_BOOLEAN(0, "switch-events", &record.opts.record_switch_events,
+		    "Record context switch events"),
 	OPT_END()
 };
 
@@ -1102,6 +1104,11 @@ int cmd_record(int argc, const char **argv, const char *prefix __maybe_unused)
 			  " system-wide mode\n");
 		usage_with_options(record_usage, record_options);
 	}
+	if (rec->opts.record_switch_events &&
+	    !perf_can_record_switch_events()) {
+		ui__error("kernel does not support recording context switch events (--switch-events option)\n");
+		usage_with_options(record_usage, record_options);
+	}
 
 	if (!rec->itr) {
 		rec->itr = auxtrace_record__init(rec->evlist, &err);
diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index 937b16aa0300..cf459f89fc9b 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -57,6 +57,7 @@ struct record_opts {
 	bool	     running_time;
 	bool	     full_auxtrace;
 	bool	     auxtrace_snapshot_mode;
+	bool	     record_switch_events;
 	unsigned int freq;
 	unsigned int mmap_pages;
 	unsigned int auxtrace_mmap_pages;
diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h
index 406a8216a51e..a8930b68456b 100644
--- a/tools/perf/util/evlist.h
+++ b/tools/perf/util/evlist.h
@@ -114,6 +114,7 @@ void perf_evlist__close(struct perf_evlist *evlist);
 
 void perf_evlist__set_id_pos(struct perf_evlist *evlist);
 bool perf_can_sample_identifier(void);
+bool perf_can_record_switch_events(void);
 void perf_evlist__config(struct perf_evlist *evlist, struct record_opts *opts);
 int record_opts__config(struct record_opts *opts);
 
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 9e6e6f40b787..71f6905c7cb9 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -738,6 +738,9 @@ void perf_evsel__config(struct perf_evsel *evsel, struct record_opts *opts)
 	attr->mmap2 = track && !perf_missing_features.mmap2;
 	attr->comm  = track;
 
+	if (opts->record_switch_events)
+		attr->context_switch = track;
+
 	if (opts->sample_transaction)
 		perf_evsel__set_sample_bit(evsel, TRANSACTION);
 
diff --git a/tools/perf/util/record.c b/tools/perf/util/record.c
index 1f7becbe5e18..0d228a29526d 100644
--- a/tools/perf/util/record.c
+++ b/tools/perf/util/record.c
@@ -85,6 +85,11 @@ static void perf_probe_comm_exec(struct perf_evsel *evsel)
 	evsel->attr.comm_exec = 1;
 }
 
+static void perf_probe_context_switch(struct perf_evsel *evsel)
+{
+	evsel->attr.context_switch = 1;
+}
+
 bool perf_can_sample_identifier(void)
 {
 	return perf_probe_api(perf_probe_sample_identifier);
@@ -95,6 +100,11 @@ static bool perf_can_comm_exec(void)
 	return perf_probe_api(perf_probe_comm_exec);
 }
 
+bool perf_can_record_switch_events(void)
+{
+	return perf_probe_api(perf_probe_context_switch);
+}
+
 void perf_evlist__config(struct perf_evlist *evlist, struct record_opts *opts)
 {
 	struct perf_evsel *evsel;
-- 
2.1.0

[PATCH 17/18] perf script: Don't assume evsel position of tracking events

[Arnaldo Carvalho de Melo]

From: Adrian Hunter <adrian.hunter@intel.com>

The tracking event does not have to be the first event so replace
perf_evlist__first() with perf_evlist__id2evsel() which uses the event
ID to find the correct evsel.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Jiri Olsa <jolsa@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Pawel Moll <pawel.moll@arm.com>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1437471846-26995-5-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-script.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index 2777c6444bc3..8a8e2ead6382 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -661,7 +661,7 @@ static int process_comm_event(struct perf_tool *tool,
 	struct thread *thread;
 	struct perf_script *script = container_of(tool, struct perf_script, tool);
 	struct perf_session *session = script->session;
-	struct perf_evsel *evsel = perf_evlist__first(session->evlist);
+	struct perf_evsel *evsel = perf_evlist__id2evsel(session->evlist, sample->id);
 	int ret = -1;
 
 	thread = machine__findnew_thread(machine, event->comm.pid, event->comm.tid);
@@ -695,7 +695,7 @@ static int process_fork_event(struct perf_tool *tool,
 	struct thread *thread;
 	struct perf_script *script = container_of(tool, struct perf_script, tool);
 	struct perf_session *session = script->session;
-	struct perf_evsel *evsel = perf_evlist__first(session->evlist);
+	struct perf_evsel *evsel = perf_evlist__id2evsel(session->evlist, sample->id);
 
 	if (perf_event__process_fork(tool, event, sample, machine) < 0)
 		return -1;
@@ -727,7 +727,7 @@ static int process_exit_event(struct perf_tool *tool,
 	struct thread *thread;
 	struct perf_script *script = container_of(tool, struct perf_script, tool);
 	struct perf_session *session = script->session;
-	struct perf_evsel *evsel = perf_evlist__first(session->evlist);
+	struct perf_evsel *evsel = perf_evlist__id2evsel(session->evlist, sample->id);
 
 	thread = machine__findnew_thread(machine, event->fork.pid, event->fork.tid);
 	if (thread == NULL) {
@@ -759,7 +759,7 @@ static int process_mmap_event(struct perf_tool *tool,
 	struct thread *thread;
 	struct perf_script *script = container_of(tool, struct perf_script, tool);
 	struct perf_session *session = script->session;
-	struct perf_evsel *evsel = perf_evlist__first(session->evlist);
+	struct perf_evsel *evsel = perf_evlist__id2evsel(session->evlist, sample->id);
 
 	if (perf_event__process_mmap(tool, event, sample, machine) < 0)
 		return -1;
@@ -790,7 +790,7 @@ static int process_mmap2_event(struct perf_tool *tool,
 	struct thread *thread;
 	struct perf_script *script = container_of(tool, struct perf_script, tool);
 	struct perf_session *session = script->session;
-	struct perf_evsel *evsel = perf_evlist__first(session->evlist);
+	struct perf_evsel *evsel = perf_evlist__id2evsel(session->evlist, sample->id);
 
 	if (perf_event__process_mmap2(tool, event, sample, machine) < 0)
 		return -1;
-- 
2.1.0

[PATCH 18/18] perf script: Add option --show-switch-events

[Arnaldo Carvalho de Melo]

From: Adrian Hunter <adrian.hunter@intel.com>

Add option --show-switch-events to show switch events in a similar
fashion to --show-task-events and --show-mmap-events.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Jiri Olsa <jolsa@redhat.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Pawel Moll <pawel.moll@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1437471846-26995-6-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-script.txt |  4 ++++
 tools/perf/builtin-script.c              | 31 +++++++++++++++++++++++++++++++
 2 files changed, 35 insertions(+)

diff --git a/tools/perf/Documentation/perf-script.txt b/tools/perf/Documentation/perf-script.txt
index c82df572fac2..e2fec5fc21e7 100644
--- a/tools/perf/Documentation/perf-script.txt
+++ b/tools/perf/Documentation/perf-script.txt
@@ -222,6 +222,10 @@ OPTIONS
 --show-mmap-events
 	Display mmap related events (e.g. MMAP, MMAP2).
 
+--show-switch-events
+	Display context switch events i.e. events of type PERF_RECORD_SWITCH or
+	PERF_RECORD_SWITCH_CPU_WIDE.
+
 --header
 	Show perf.data header.
 
diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index 8a8e2ead6382..bd31380122f9 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -623,6 +623,7 @@ struct perf_script {
 	struct perf_session	*session;
 	bool			show_task_events;
 	bool			show_mmap_events;
+	bool			show_switch_events;
 };
 
 static int process_attr(struct perf_tool *tool, union perf_event *event,
@@ -813,6 +814,32 @@ static int process_mmap2_event(struct perf_tool *tool,
 	return 0;
 }
 
+static int process_switch_event(struct perf_tool *tool,
+				union perf_event *event,
+				struct perf_sample *sample,
+				struct machine *machine)
+{
+	struct thread *thread;
+	struct perf_script *script = container_of(tool, struct perf_script, tool);
+	struct perf_session *session = script->session;
+	struct perf_evsel *evsel = perf_evlist__id2evsel(session->evlist, sample->id);
+
+	if (perf_event__process_switch(tool, event, sample, machine) < 0)
+		return -1;
+
+	thread = machine__findnew_thread(machine, sample->pid,
+					 sample->tid);
+	if (thread == NULL) {
+		pr_debug("problem processing SWITCH event, skipping it.\n");
+		return -1;
+	}
+
+	print_sample_start(sample, thread, evsel);
+	perf_event__fprintf(event, stdout);
+	thread__put(thread);
+	return 0;
+}
+
 static void sig_handler(int sig __maybe_unused)
 {
 	session_done = 1;
@@ -834,6 +861,8 @@ static int __cmd_script(struct perf_script *script)
 		script->tool.mmap = process_mmap_event;
 		script->tool.mmap2 = process_mmap2_event;
 	}
+	if (script->show_switch_events)
+		script->tool.context_switch = process_switch_event;
 
 	ret = perf_session__process_events(script->session);
 
@@ -1618,6 +1647,8 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
 		    "Show the fork/comm/exit events"),
 	OPT_BOOLEAN('\0', "show-mmap-events", &script.show_mmap_events,
 		    "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_CALLBACK_OPTARG(0, "itrace", &itrace_synth_opts, NULL, "opts",
 			    "Instruction Tracing options",
-- 
2.1.0

Re: [GIT PULL 00/18] perf/core improvements and fixes

[Ingo Molnar]

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

> Hi Ingo,
> 
> 	Please consider pulling,
> 
> - Arnaldo
> 
> The following changes since commit a11c51acc52822754d66a11c15f6f6edd4d23c55:
> 
>   Merge tag 'perf-core-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core (2015-07-21 07:58:06 +0200)
> 
> are available in the git repository at:
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-core-for-mingo
> 
> for you to fetch changes up to 7c14898ba9386ee5c939bb418643ac6baff52840:
> 
>   perf script: Add option --show-switch-events (2015-07-23 22:51:14 -0300)
> 
> ----------------------------------------------------------------
> perf/core improvements and fixes:
> 
> New features:
> 
> - Introduce PERF_RECORD_SWITCH(_CPU_WIDE) and use it in 'record' to
>   ask for context switches, allowing non priviledged tasks to know
>   when they are switched in and out, which wasn't possible with
>   the other context switch tracepoint and software events, see the
>   patch description for a comprehensive justification (Adrian Hunter)
> 
> - Stop collecting /proc/kallsyms in perf.data files, saving about
>   4.5MB on a typical x86-64 system, use the symbol resolution
>   routines used in all the other tools (report, top, etc) now that
>   we can ask libtraceevent to use perf's symbol resolution code.
>   (Arnaldo Carvalho de Melo)
> 
> User visible fixes:
> 
> - Expose perf's symbol resolver to libtraceecent, so that its plugins can
>   resolve tracepoint fields to kernel functions, like the 'function' field
>   in the "timer:hrtimer_start tracepoint" (Arnaldo Carvalho de Melo)
> 
> Infrastructure:
> 
> - Map propagation of thread and cpu maps improvements, prep work for
>   'perf stat' new features (Jiri Olsa)
> 
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
> 
> ----------------------------------------------------------------
> Adrian Hunter (5):
>       perf: Add PERF_RECORD_SWITCH to indicate context switches
>       perf tools: Add new PERF_RECORD_SWITCH event
>       perf record: Add option --switch-events to select PERF_RECORD_SWITCH events
>       perf script: Don't assume evsel position of tracking events
>       perf script: Add option --show-switch-events
> 
> Arnaldo Carvalho de Melo (8):
>       perf symbols: Add front end cache for DSO symbol lookup
>       perf symbols: Introduce map__is_(kernel,kmodule)()
>       tools lib traceevent: Allow setting an alternative symbol resolver
>       perf symbols: Provide libtraceevent callback to resolve kernel symbols
>       perf trace: Provide libtracevent with a kernel symbol resolver
>       perf script: Switch from perf.data's kallsyms to perf's symbol resolver
>       perf tools: Stop reading the kallsyms data from perf.data
>       perf tools: Stop copying kallsyms into the perf.data file header
> 
> Jiri Olsa (5):
>       perf test: Check for refcnt in thread_map test
>       perf evlist: Force perf_evlist__set_maps to propagate maps through events
>       perf evlist: Use bool instead of target argument in propagate_maps()
>       perf evlist: Tolerate NULL maps in propagate_maps
>       perf header: Use argv style storage for cmdline feature data
> 
>  include/uapi/linux/perf_event.h          |  31 +++++++++-
>  kernel/events/core.c                     | 103 +++++++++++++++++++++++++++++++
>  tools/lib/traceevent/event-parse.c       |  68 +++++++++++++++++++-
>  tools/lib/traceevent/event-parse.h       |   8 +++
>  tools/perf/Documentation/perf-record.txt |   4 ++
>  tools/perf/Documentation/perf-script.txt |   4 ++
>  tools/perf/builtin-inject.c              |   1 +
>  tools/perf/builtin-record.c              |   7 +++
>  tools/perf/builtin-script.c              |  48 ++++++++++++--
>  tools/perf/builtin-trace.c               |   3 +
>  tools/perf/perf.h                        |   1 +
>  tools/perf/tests/thread-map.c            |   4 ++
>  tools/perf/util/dso.h                    |   4 ++
>  tools/perf/util/event.c                  |  28 +++++++++
>  tools/perf/util/event.h                  |  12 ++++
>  tools/perf/util/evlist.c                 |  28 +++++++--
>  tools/perf/util/evlist.h                 |  12 ++--
>  tools/perf/util/evsel.c                  |   4 ++
>  tools/perf/util/header.c                 |  35 ++++++-----
>  tools/perf/util/header.h                 |   1 +
>  tools/perf/util/machine.c                |  25 ++++++++
>  tools/perf/util/machine.h                |   6 ++
>  tools/perf/util/map.c                    |  14 +++++
>  tools/perf/util/map.h                    |   7 +++
>  tools/perf/util/record.c                 |  10 +++
>  tools/perf/util/session.c                |  21 +++++++
>  tools/perf/util/symbol.c                 |   7 ++-
>  tools/perf/util/tool.h                   |   1 +
>  tools/perf/util/trace-event-info.c       |  22 +++----
>  tools/perf/util/trace-event-parse.c      |  30 ---------
>  tools/perf/util/trace-event-read.c       |  28 ++++-----
>  tools/perf/util/trace-event.c            |  45 +++++++++-----
>  tools/perf/util/trace-event.h            |   1 +
>  33 files changed, 513 insertions(+), 110 deletions(-)

Pulled, thanks a lot Arnaldo!

	Ingo