[GIT PULL 00/14] perf/core improvements and fixes

[GIT PULL 00/14] perf/core improvements and fixes

[Arnaldo Carvalho de Melo]

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

Hi Ingo,

	Please consider pulling, will continue processing the backlog tomorrow,

Thanks,

- Arnaldo

The following changes since commit f1ad44884a4c421ceaa9a4a8242aeeee6f686670:

  perf/x86: Remove old MSR perf tracing code (2015-12-06 12:56:14 +0100)

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 cfef25b8daf7e4b49c84e174a904af9d89dc7c46:

  perf annotate: ARM support (2015-12-07 18:13:00 -0300)

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

User visible:

- Fixes and improvements for supporting annotating ARM binaries, support ARM
  call and jump instructions, more work needed to have arch specific stuff
  separated into tools/perf/arch/*/annotate/ (Russell King)

- Fix several 'perf test' entries broken by recent perf/core changes (Jiri Olsa)

Infrastructure:

- Consolidate perf_ev{list,sel}__{enable,disable}() calls (Jiri Olsa)

- Pass correct string to dso__adjust_kmod_long_name (Wang Nan)

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

----------------------------------------------------------------
Jiri Olsa (12):
      perf test: Use machine__new_host in dwarf unwind test
      perf test: Use machine__new_host in mmap thread lookup test
      perf test: Use machine__new_host in mmap thread code reading test
      perf test: Fix cpus and thread maps reference in error path
      perf test: Prevent using bpf-output event in round trip name test
      perf test: Create kernel maps properly for hist entries test
      perf evsel: Use event maps directly in perf_evsel__enable
      perf evsel: Introduce disable() method
      perf evlist: Factor perf_evlist__(enable|disable) functions
      perf stat: Use perf_evlist__enable in handle_initial_delay
      perf stat: Create events as disabled
      perf stat: Move enable_on_exec setup under earlier code

Russell King (1):
      perf annotate: ARM support

Wang Nan (1):
      perf machine: Pass correct string to dso__adjust_kmod_long_name

 tools/perf/builtin-stat.c               | 44 ++++++++++++++++++++-------------
 tools/perf/tests/code-reading.c         | 14 +++++++----
 tools/perf/tests/dwarf-unwind.c         |  8 ++----
 tools/perf/tests/evsel-roundtrip-name.c |  3 ++-
 tools/perf/tests/hists_common.c         |  5 ++++
 tools/perf/tests/mmap-thread-lookup.c   |  6 ++---
 tools/perf/util/annotate.c              | 23 +++++++++++++++++
 tools/perf/util/evlist.c                | 32 ++++++------------------
 tools/perf/util/evsel.c                 | 15 ++++++++++-
 tools/perf/util/evsel.h                 |  3 ++-
 tools/perf/util/machine.c               |  2 +-
 11 files changed, 95 insertions(+), 60 deletions(-)

[PATCH 01/14] perf machine: Pass correct string to dso__adjust_kmod_long_name

[Arnaldo Carvalho de Melo]

From: Wang Nan <wangnan0@huawei.com>

There's a mistake in dso__adjust_kmod_long_name() that it use strdup()
to dup the new long_name of a dso, but passes the original string to
dso__set_long_name(). Which causes random crash during cleanup.

Signed-off-by: Wang Nan <wangnan0@huawei.com>
Reviewed-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Zefan Li <lizefan@huawei.com>
Cc: pi3orama@163.com
Fixes: c03d5184f0e9 ("perf machine: Adjust dso->long_name for offline module")
Link: http://lkml.kernel.org/r/1449455785-42020-1-git-send-email-wangnan0@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/machine.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index 95a7f6087346..bfc289c73c22 100644
--- a/tools/perf/util/machine.c
+++ b/tools/perf/util/machine.c
@@ -576,7 +576,7 @@ static void dso__adjust_kmod_long_name(struct dso *dso, const char *filename)
 	if (!dup_filename)
 		return;
 
-	dso__set_long_name(dso, filename, true);
+	dso__set_long_name(dso, dup_filename, true);
 }
 
 struct map *machine__findnew_module_map(struct machine *machine, u64 start,
-- 
2.1.0

[PATCH 02/14] perf test: Use machine__new_host in dwarf unwind test

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

This is more straightforward than what we have now.

It also fixes a segfault within machine__exit, that's caused by not
creating kernel maps for machine.. We're calling
machine__destroy_kernel_maps in machine__exit since commit:

  ebe9729c8c31 perf machine: Fix to destroy kernel maps when machine exits

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1449131658-1841-2-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/tests/dwarf-unwind.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/tools/perf/tests/dwarf-unwind.c b/tools/perf/tests/dwarf-unwind.c
index 3cce13b19cbb..1c5c0221cea2 100644
--- a/tools/perf/tests/dwarf-unwind.c
+++ b/tools/perf/tests/dwarf-unwind.c
@@ -160,14 +160,11 @@ static int krava_1(struct thread *thread)
 
 int test__dwarf_unwind(int subtest __maybe_unused)
 {
-	struct machines machines;
 	struct machine *machine;
 	struct thread *thread;
 	int err = -1;
 
-	machines__init(&machines);
-
-	machine = machines__find(&machines, HOST_KERNEL_ID);
+	machine = machine__new_host();
 	if (!machine) {
 		pr_err("Could not get machine\n");
 		return -1;
@@ -199,7 +196,6 @@ int test__dwarf_unwind(int subtest __maybe_unused)
 
  out:
 	machine__delete_threads(machine);
-	machine__exit(machine);
-	machines__exit(&machines);
+	machine__delete(machine);
 	return err;
 }
-- 
2.1.0

[PATCH 03/14] perf test: Use machine__new_host in mmap thread lookup test

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

This is more straightforward than what we have now.

It also fixes a segfault within machine__exit, that's caused by not
creating kernel maps for machine.. We're calling
machine__destroy_kernel_maps in machine__exit since commit:

  ebe9729c8c31 perf machine: Fix to destroy kernel maps when machine exits

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1449131658-1841-3-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/tests/mmap-thread-lookup.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/tools/perf/tests/mmap-thread-lookup.c b/tools/perf/tests/mmap-thread-lookup.c
index 6cdb97579c45..0c5ce44f723f 100644
--- a/tools/perf/tests/mmap-thread-lookup.c
+++ b/tools/perf/tests/mmap-thread-lookup.c
@@ -149,7 +149,6 @@ static int synth_process(struct machine *machine)
 
 static int mmap_events(synth_cb synth)
 {
-	struct machines machines;
 	struct machine *machine;
 	int err, i;
 
@@ -162,8 +161,7 @@ static int mmap_events(synth_cb synth)
 	 */
 	TEST_ASSERT_VAL("failed to create threads", !threads_create());
 
-	machines__init(&machines);
-	machine = &machines.host;
+	machine = machine__new_host();
 
 	dump_trace = verbose > 1 ? 1 : 0;
 
@@ -203,7 +201,7 @@ static int mmap_events(synth_cb synth)
 	}
 
 	machine__delete_threads(machine);
-	machines__exit(&machines);
+	machine__delete(machine);
 	return err;
 }
 
-- 
2.1.0

[PATCH 04/14] perf test: Use machine__new_host in mmap thread code reading test

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

This is more straightforward than what we have now.

It also fixes a segfault within machine__exit, that's caused
by not creating kernel maps for machine.. We're calling
machine__destroy_kernel_maps in machine__exit since commit:

  ebe9729c8c31 perf machine: Fix to destroy kernel maps when machine exits

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1449131658-1841-4-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/tests/code-reading.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/tools/perf/tests/code-reading.c b/tools/perf/tests/code-reading.c
index 4417b6a079f0..26182ffcea75 100644
--- a/tools/perf/tests/code-reading.c
+++ b/tools/perf/tests/code-reading.c
@@ -433,7 +433,6 @@ enum {
 
 static int do_test_code_reading(bool try_kcore)
 {
-	struct machines machines;
 	struct machine *machine;
 	struct thread *thread;
 	struct record_opts opts = {
@@ -459,8 +458,7 @@ static int do_test_code_reading(bool try_kcore)
 
 	pid = getpid();
 
-	machines__init(&machines);
-	machine = &machines.host;
+	machine = machine__new_host();
 
 	ret = machine__create_kernel_maps(machine);
 	if (ret < 0) {
@@ -594,9 +592,8 @@ out_err:
 		cpu_map__put(cpus);
 		thread_map__put(threads);
 	}
-	machines__destroy_kernel_maps(&machines);
 	machine__delete_threads(machine);
-	machines__exit(&machines);
+	machine__delete(machine);
 
 	return err;
 }
-- 
2.1.0

[PATCH 05/14] perf test: Fix cpus and thread maps reference in error path

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

In error path to try user space event, both cpus and threads map now
owned by evlist and freed by perf_evlist__set_maps call.  Getting
reference to keep them alive.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1449131658-1841-5-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/tests/code-reading.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/tools/perf/tests/code-reading.c b/tools/perf/tests/code-reading.c
index 26182ffcea75..313a48c6b2bc 100644
--- a/tools/perf/tests/code-reading.c
+++ b/tools/perf/tests/code-reading.c
@@ -547,6 +547,13 @@ static int do_test_code_reading(bool try_kcore)
 		if (ret < 0) {
 			if (!excl_kernel) {
 				excl_kernel = true;
+				/*
+				 * Both cpus and threads are now owned by evlist
+				 * and will be freed by following perf_evlist__set_maps
+				 * call. Getting refference to keep them alive.
+				 */
+				cpu_map__get(cpus);
+				thread_map__get(threads);
 				perf_evlist__set_maps(evlist, NULL, NULL);
 				perf_evlist__delete(evlist);
 				evlist = NULL;
-- 
2.1.0

[PATCH 06/14] perf test: Prevent using bpf-output event in round trip name test

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

The bpf-output is added under software events, but is not parse-able
within parse_events, which is what round trip test is expecting.

Checking software events only until dummy event.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1449131658-1841-6-git-send-email-jolsa@kernel.org
[ Make it a one liner by keeping __perf_evsel__name_array_test() around ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/tests/evsel-roundtrip-name.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tools/perf/tests/evsel-roundtrip-name.c b/tools/perf/tests/evsel-roundtrip-name.c
index 1da92e1159ee..2de4a4f2c3ed 100644
--- a/tools/perf/tests/evsel-roundtrip-name.c
+++ b/tools/perf/tests/evsel-roundtrip-name.c
@@ -103,7 +103,8 @@ int test__perf_evsel__roundtrip_name_test(int subtest __maybe_unused)
 	if (err)
 		ret = err;
 
-	err = perf_evsel__name_array_test(perf_evsel__sw_names);
+	err = __perf_evsel__name_array_test(perf_evsel__sw_names,
+					    PERF_COUNT_SW_DUMMY + 1);
 	if (err)
 		ret = err;
 
-- 
2.1.0

[PATCH 07/14] perf test: Create kernel maps properly for hist entries test

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

It fixes segfault within machine__exit, that's caused
but not creating kernel maps for machine.. We're calling
machine__destroy_kernel_maps in machine__exit since commit:

  ebe9729c8c31 perf machine: Fix to destroy kernel maps when machine exits

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/tip-k4snzv5t4dvdckggzwdzyljo@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/tests/hists_common.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/tools/perf/tests/hists_common.c b/tools/perf/tests/hists_common.c
index ce80b274b097..46f453b1de60 100644
--- a/tools/perf/tests/hists_common.c
+++ b/tools/perf/tests/hists_common.c
@@ -87,6 +87,11 @@ struct machine *setup_fake_machine(struct machines *machines)
 		return NULL;
 	}
 
+	if (machine__create_kernel_maps(machine)) {
+		pr_debug("Not enough memory for machine setup\n");
+		goto out;
+	}
+
 	for (i = 0; i < ARRAY_SIZE(fake_threads); i++) {
 		struct thread *thread;
 
-- 
2.1.0

[PATCH 08/14] perf evsel: Use event maps directly in perf_evsel__enable

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

All events now share proper cpu and thread maps. There's no need to pass
those maps from evlist, it's safe to use evsel maps for enabling event.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
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/1449133606-14429-2-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-stat.c | 5 +----
 tools/perf/util/evsel.c   | 5 ++++-
 tools/perf/util/evsel.h   | 2 +-
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index df2fbf046ee2..813c52ad9303 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -256,12 +256,9 @@ static void handle_initial_delay(void)
 	struct perf_evsel *counter;
 
 	if (initial_delay) {
-		const int ncpus = cpu_map__nr(evsel_list->cpus),
-			nthreads = thread_map__nr(evsel_list->threads);
-
 		usleep(initial_delay * 1000);
 		evlist__for_each(evsel_list, counter)
-			perf_evsel__enable(counter, ncpus, nthreads);
+			perf_evsel__enable(counter);
 	}
 }
 
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 0a1f4d9e52fc..3a9b5068667d 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -981,8 +981,11 @@ int perf_evsel__append_filter(struct perf_evsel *evsel,
 	return -1;
 }
 
-int perf_evsel__enable(struct perf_evsel *evsel, int ncpus, int nthreads)
+int perf_evsel__enable(struct perf_evsel *evsel)
 {
+	int nthreads = thread_map__nr(evsel->threads);
+	int ncpus = cpu_map__nr(evsel->cpus);
+
 	return perf_evsel__run_ioctl(evsel, ncpus, nthreads,
 				     PERF_EVENT_IOC_ENABLE,
 				     0);
diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
index 0e49bd742c63..a721592a3200 100644
--- a/tools/perf/util/evsel.h
+++ b/tools/perf/util/evsel.h
@@ -227,7 +227,7 @@ int perf_evsel__append_filter(struct perf_evsel *evsel,
 			      const char *op, const char *filter);
 int perf_evsel__apply_filter(struct perf_evsel *evsel, int ncpus, int nthreads,
 			     const char *filter);
-int perf_evsel__enable(struct perf_evsel *evsel, int ncpus, int nthreads);
+int perf_evsel__enable(struct perf_evsel *evsel);
 
 int perf_evsel__open_per_cpu(struct perf_evsel *evsel,
 			     struct cpu_map *cpus);
-- 
2.1.0

[PATCH 09/14] perf evsel: Introduce disable() method

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

Adding perf_evsel__disable function to have complement for
perf_evsel__enable function. Both will be used in following patch to
factor perf_evlist__(enable|disable).

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
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/1449133606-14429-3-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evsel.c | 10 ++++++++++
 tools/perf/util/evsel.h |  1 +
 2 files changed, 11 insertions(+)

diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 3a9b5068667d..47f033089349 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -991,6 +991,16 @@ int perf_evsel__enable(struct perf_evsel *evsel)
 				     0);
 }
 
+int perf_evsel__disable(struct perf_evsel *evsel)
+{
+	int nthreads = thread_map__nr(evsel->threads);
+	int ncpus = cpu_map__nr(evsel->cpus);
+
+	return perf_evsel__run_ioctl(evsel, ncpus, nthreads,
+				     PERF_EVENT_IOC_DISABLE,
+				     0);
+}
+
 int perf_evsel__alloc_id(struct perf_evsel *evsel, int ncpus, int nthreads)
 {
 	if (ncpus == 0 || nthreads == 0)
diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
index a721592a3200..5ded1fc0341e 100644
--- a/tools/perf/util/evsel.h
+++ b/tools/perf/util/evsel.h
@@ -228,6 +228,7 @@ int perf_evsel__append_filter(struct perf_evsel *evsel,
 int perf_evsel__apply_filter(struct perf_evsel *evsel, int ncpus, int nthreads,
 			     const char *filter);
 int perf_evsel__enable(struct perf_evsel *evsel);
+int perf_evsel__disable(struct perf_evsel *evsel);
 
 int perf_evsel__open_per_cpu(struct perf_evsel *evsel,
 			     struct cpu_map *cpus);
-- 
2.1.0

[PATCH 10/14] perf evlist: Factor perf_evlist__(enable|disable) functions

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

Use perf_evsel__(enable|disable) functions in perf_evlist__(enable|disable)
functions in order to centralize ioctl enable/disable calls. This way we
eliminate 2 places calling directly ioctl.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
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/1449133606-14429-4-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evlist.c | 32 ++++++++------------------------
 1 file changed, 8 insertions(+), 24 deletions(-)

diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index d1392194a9a9..d1b6c206bb93 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -336,20 +336,12 @@ static int perf_evlist__nr_threads(struct perf_evlist *evlist,
 
 void perf_evlist__disable(struct perf_evlist *evlist)
 {
-	int cpu, thread;
 	struct perf_evsel *pos;
-	int nr_cpus = cpu_map__nr(evlist->cpus);
-	int nr_threads;
 
-	for (cpu = 0; cpu < nr_cpus; cpu++) {
-		evlist__for_each(evlist, pos) {
-			if (!perf_evsel__is_group_leader(pos) || !pos->fd)
-				continue;
-			nr_threads = perf_evlist__nr_threads(evlist, pos);
-			for (thread = 0; thread < nr_threads; thread++)
-				ioctl(FD(pos, cpu, thread),
-				      PERF_EVENT_IOC_DISABLE, 0);
-		}
+	evlist__for_each(evlist, pos) {
+		if (!perf_evsel__is_group_leader(pos) || !pos->fd)
+			continue;
+		perf_evsel__disable(pos);
 	}
 
 	evlist->enabled = false;
@@ -357,20 +349,12 @@ void perf_evlist__disable(struct perf_evlist *evlist)
 
 void perf_evlist__enable(struct perf_evlist *evlist)
 {
-	int cpu, thread;
 	struct perf_evsel *pos;
-	int nr_cpus = cpu_map__nr(evlist->cpus);
-	int nr_threads;
 
-	for (cpu = 0; cpu < nr_cpus; cpu++) {
-		evlist__for_each(evlist, pos) {
-			if (!perf_evsel__is_group_leader(pos) || !pos->fd)
-				continue;
-			nr_threads = perf_evlist__nr_threads(evlist, pos);
-			for (thread = 0; thread < nr_threads; thread++)
-				ioctl(FD(pos, cpu, thread),
-				      PERF_EVENT_IOC_ENABLE, 0);
-		}
+	evlist__for_each(evlist, pos) {
+		if (!perf_evsel__is_group_leader(pos) || !pos->fd)
+			continue;
+		perf_evsel__enable(pos);
 	}
 
 	evlist->enabled = true;
-- 
2.1.0

[PATCH 11/14] perf stat: Use perf_evlist__enable in handle_initial_delay

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

No need to mimic the behaviour of perf_evlist__enable, we can use it
directly.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
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/1449133606-14429-5-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-stat.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 813c52ad9303..8ca40deaa728 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -253,12 +253,9 @@ static void process_interval(void)
 
 static void handle_initial_delay(void)
 {
-	struct perf_evsel *counter;
-
 	if (initial_delay) {
 		usleep(initial_delay * 1000);
-		evlist__for_each(evsel_list, counter)
-			perf_evsel__enable(counter);
+		perf_evlist__enable(evsel_list);
 	}
 }
 
-- 
2.1.0

[PATCH 12/14] perf stat: Create events as disabled

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

Currently we have 2 kinds of stat counters based on when the event is
enabled:

  1) tracee command events, which are enable once the
     tracee executes exec syscall (enable_on_exec bit)
  2) all other events which get alive within the
     perf_event_open syscall

And 2) case could raise a problem in case we want additional filter to
be attached for event. In this case we want the event to be enabled
after it's configured with filter.

Changing the behaviour of 2) events, so they all are created as disabled
(disabled bit). Adding extra enable call to make them alive once they
finish setup.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
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/1449133606-14429-6-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-stat.c | 25 +++++++++++++++++++------
 1 file changed, 19 insertions(+), 6 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 8ca40deaa728..2e70610649a1 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -168,11 +168,18 @@ static int create_perf_stat_counter(struct perf_evsel *evsel)
 	attr->sample_period = 0;
 	attr->sample_type   = 0;
 
+	/*
+	 * Disabling all counters initially, they will be enabled
+	 * either manually by us or by kernel via enable_on_exec
+	 * set later.
+	 */
+	if (perf_evsel__is_group_leader(evsel))
+		attr->disabled = 1;
+
 	if (target__has_cpu(&target))
 		return perf_evsel__open_per_cpu(evsel, perf_evsel__cpus(evsel));
 
 	if (!target__has_task(&target) && perf_evsel__is_group_leader(evsel)) {
-		attr->disabled = 1;
 		if (!initial_delay)
 			attr->enable_on_exec = 1;
 	}
@@ -251,12 +258,18 @@ static void process_interval(void)
 	print_counters(&rs, 0, NULL);
 }
 
-static void handle_initial_delay(void)
+static void enable_counters(void)
 {
-	if (initial_delay) {
+	if (initial_delay)
 		usleep(initial_delay * 1000);
+
+	/*
+	 * We need to enable counters only if:
+	 * - we don't have tracee (attaching to task or cpu)
+	 * - we have initial delay configured
+	 */
+	if (!target__none(&target) || initial_delay)
 		perf_evlist__enable(evsel_list);
-	}
 }
 
 static volatile int workload_exec_errno;
@@ -353,7 +366,7 @@ static int __run_perf_stat(int argc, const char **argv)
 
 	if (forks) {
 		perf_evlist__start_workload(evsel_list);
-		handle_initial_delay();
+		enable_counters();
 
 		if (interval) {
 			while (!waitpid(child_pid, &status, WNOHANG)) {
@@ -372,7 +385,7 @@ static int __run_perf_stat(int argc, const char **argv)
 		if (WIFSIGNALED(status))
 			psignal(WTERMSIG(status), argv[0]);
 	} else {
-		handle_initial_delay();
+		enable_counters();
 		while (!done) {
 			nanosleep(&ts, NULL);
 			if (interval)
-- 
2.1.0

[PATCH 13/14] perf stat: Move enable_on_exec setup under earlier code

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

It's more readable this way and we can save one
perf_evsel__is_group_leader condition in current code.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
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/1449133606-14429-7-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-stat.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 2e70610649a1..e74712dee242 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -173,17 +173,20 @@ static int create_perf_stat_counter(struct perf_evsel *evsel)
 	 * either manually by us or by kernel via enable_on_exec
 	 * set later.
 	 */
-	if (perf_evsel__is_group_leader(evsel))
+	if (perf_evsel__is_group_leader(evsel)) {
 		attr->disabled = 1;
 
-	if (target__has_cpu(&target))
-		return perf_evsel__open_per_cpu(evsel, perf_evsel__cpus(evsel));
-
-	if (!target__has_task(&target) && perf_evsel__is_group_leader(evsel)) {
-		if (!initial_delay)
+		/*
+		 * In case of initial_delay we enable tracee
+		 * events manually.
+		 */
+		if (target__none(&target) && !initial_delay)
 			attr->enable_on_exec = 1;
 	}
 
+	if (target__has_cpu(&target))
+		return perf_evsel__open_per_cpu(evsel, perf_evsel__cpus(evsel));
+
 	return perf_evsel__open_per_thread(evsel, evsel_list->threads);
 }
 
-- 
2.1.0

[PATCH 14/14] perf annotate: ARM support

[Arnaldo Carvalho de Melo]

From: Russell King <rmk+kernel@arm.linux.org.uk>

Add basic support to parse ARM assembly.

This:

* enables perf to correctly show the disassembly, rather than chopping
  some constants off at the '#' (which is not a comment character on
  ARM).

* allows perf to identify ARM instructions that branch to other parts
  within the same function, thereby properly annotating them.

* allows perf to identify function calls, allowing called functions to
  be followed in the annotated view.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Will Deacon <will.deacon@arm.com>
Link: http://lkml.kernel.org/n/tip-owp1uj0nmcgfrlppfyeetuyf@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/annotate.c | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c
index 1dd1949b0e79..b795b6994144 100644
--- a/tools/perf/util/annotate.c
+++ b/tools/perf/util/annotate.c
@@ -65,6 +65,11 @@ static int call__parse(struct ins_operands *ops)
 
 	name++;
 
+#ifdef __arm__
+	if (strchr(name, '+'))
+		return -1;
+#endif
+
 	tok = strchr(name, '>');
 	if (tok == NULL)
 		return -1;
@@ -246,7 +251,11 @@ static int mov__parse(struct ins_operands *ops)
 		return -1;
 
 	target = ++s;
+#ifdef __arm__
+	comment = strchr(s, ';');
+#else
 	comment = strchr(s, '#');
+#endif
 
 	if (comment != NULL)
 		s = comment - 1;
@@ -354,6 +363,20 @@ static struct ins instructions[] = {
 	{ .name = "addq",  .ops  = &mov_ops, },
 	{ .name = "addw",  .ops  = &mov_ops, },
 	{ .name = "and",   .ops  = &mov_ops, },
+#ifdef __arm__
+	{ .name = "b",     .ops  = &jump_ops, }, // might also be a call
+	{ .name = "bcc",   .ops  = &jump_ops, },
+	{ .name = "bcs",   .ops  = &jump_ops, },
+	{ .name = "beq",   .ops  = &jump_ops, },
+	{ .name = "bge",   .ops  = &jump_ops, },
+	{ .name = "bgt",   .ops  = &jump_ops, },
+	{ .name = "bhi",   .ops  = &jump_ops, },
+	{ .name = "bl",    .ops  = &call_ops, },
+	{ .name = "blt",   .ops  = &jump_ops, },
+	{ .name = "bls",   .ops  = &jump_ops, },
+	{ .name = "blx",   .ops  = &call_ops, },
+	{ .name = "bne",   .ops  = &jump_ops, },
+#endif
 	{ .name = "bts",   .ops  = &mov_ops, },
 	{ .name = "call",  .ops  = &call_ops, },
 	{ .name = "callq", .ops  = &call_ops, },
-- 
2.1.0

Re: [GIT PULL 00/14] perf/core improvements and fixes

[Ingo Molnar]

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

> From: Arnaldo Carvalho de Melo <acme@redhat.com>
> 
> Hi Ingo,
> 
> 	Please consider pulling, will continue processing the backlog tomorrow,
> 
> Thanks,
> 
> - Arnaldo
> 
> The following changes since commit f1ad44884a4c421ceaa9a4a8242aeeee6f686670:
> 
>   perf/x86: Remove old MSR perf tracing code (2015-12-06 12:56:14 +0100)
> 
> 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 cfef25b8daf7e4b49c84e174a904af9d89dc7c46:
> 
>   perf annotate: ARM support (2015-12-07 18:13:00 -0300)
> 
> ----------------------------------------------------------------
> perf/core improvements and fixes:
> 
> User visible:
> 
> - Fixes and improvements for supporting annotating ARM binaries, support ARM
>   call and jump instructions, more work needed to have arch specific stuff
>   separated into tools/perf/arch/*/annotate/ (Russell King)
> 
> - Fix several 'perf test' entries broken by recent perf/core changes (Jiri Olsa)
> 
> Infrastructure:
> 
> - Consolidate perf_ev{list,sel}__{enable,disable}() calls (Jiri Olsa)
> 
> - Pass correct string to dso__adjust_kmod_long_name (Wang Nan)
> 
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
> 
> ----------------------------------------------------------------
> Jiri Olsa (12):
>       perf test: Use machine__new_host in dwarf unwind test
>       perf test: Use machine__new_host in mmap thread lookup test
>       perf test: Use machine__new_host in mmap thread code reading test
>       perf test: Fix cpus and thread maps reference in error path
>       perf test: Prevent using bpf-output event in round trip name test
>       perf test: Create kernel maps properly for hist entries test
>       perf evsel: Use event maps directly in perf_evsel__enable
>       perf evsel: Introduce disable() method
>       perf evlist: Factor perf_evlist__(enable|disable) functions
>       perf stat: Use perf_evlist__enable in handle_initial_delay
>       perf stat: Create events as disabled
>       perf stat: Move enable_on_exec setup under earlier code
> 
> Russell King (1):
>       perf annotate: ARM support
> 
> Wang Nan (1):
>       perf machine: Pass correct string to dso__adjust_kmod_long_name
> 
>  tools/perf/builtin-stat.c               | 44 ++++++++++++++++++++-------------
>  tools/perf/tests/code-reading.c         | 14 +++++++----
>  tools/perf/tests/dwarf-unwind.c         |  8 ++----
>  tools/perf/tests/evsel-roundtrip-name.c |  3 ++-
>  tools/perf/tests/hists_common.c         |  5 ++++
>  tools/perf/tests/mmap-thread-lookup.c   |  6 ++---
>  tools/perf/util/annotate.c              | 23 +++++++++++++++++
>  tools/perf/util/evlist.c                | 32 ++++++------------------
>  tools/perf/util/evsel.c                 | 15 ++++++++++-
>  tools/perf/util/evsel.h                 |  3 ++-
>  tools/perf/util/machine.c               |  2 +-
>  11 files changed, 95 insertions(+), 60 deletions(-)

Pulled, thanks a lot Arnaldo!

	Ingo