[PATCH 0/3] perf: add a dummy software event to keep tracking

[PATCH 0/3] perf: add a dummy software event to keep tracking

[Adrian Hunter]

Hi

This is an alternative to the 'keep tracking' flag patch
which is here:

	http://marc.info/?l=linux-kernel&m=137242545521246&w=2

perf tools is updated and a test added to demonstrate the
new event.


Adrian Hunter (3):
      perf: add a dummy software event to keep tracking
      perf tools: add support for PERF_COUNT_SW_DUMMY
      perf tools: add 'keep tracking' test

 include/uapi/linux/perf_event.h  |   1 +
 tools/perf/Makefile              |   1 +
 tools/perf/tests/builtin-test.c  |   4 ++
 tools/perf/tests/keep-tracking.c | 150 +++++++++++++++++++++++++++++++++++++++
 tools/perf/tests/tests.h         |   1 +
 tools/perf/util/evlist.c         |  42 ++++++++++-
 tools/perf/util/evlist.h         |   5 ++
 tools/perf/util/parse-events.c   |   4 ++
 tools/perf/util/parse-events.l   |   1 +
 tools/perf/util/python.c         |   1 +
 10 files changed, 208 insertions(+), 2 deletions(-)
 create mode 100644 tools/perf/tests/keep-tracking.c

Regards
Adrian

[PATCH 1/3] perf: add a dummy software event to keep tracking

[Adrian Hunter]

When an event is disabled the "tracking" events
selected by the 'mmap', 'comm' and 'task' bits of
struct perf_event_attr, are also disabled.  However,
the information those events provide is necessary to
resolve symbols for when the main event is re-enabled.

The "tracking" events can be kept enabled by putting
them on another event, but that requires an event
that otherwise does nothing.  A new software event
PERF_COUNT_SW_DUMMY is added for that purpose.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 include/uapi/linux/perf_event.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h
index 62c25a2..9e28422 100644
--- a/include/uapi/linux/perf_event.h
+++ b/include/uapi/linux/perf_event.h
@@ -109,6 +109,7 @@ enum perf_sw_ids {
 	PERF_COUNT_SW_PAGE_FAULTS_MAJ		= 6,
 	PERF_COUNT_SW_ALIGNMENT_FAULTS		= 7,
 	PERF_COUNT_SW_EMULATION_FAULTS		= 8,
+	PERF_COUNT_SW_DUMMY			= 9,
 
 	PERF_COUNT_SW_MAX,			/* non-ABI */
 };
-- 
1.7.11.7

Re: [PATCH 1/3] perf: add a dummy software event to keep tracking

[Adrian Hunter]

On 07/08/13 16:30, Adrian Hunter wrote:
> When an event is disabled the "tracking" events
> selected by the 'mmap', 'comm' and 'task' bits of
> struct perf_event_attr, are also disabled.  However,
> the information those events provide is necessary to
> resolve symbols for when the main event is re-enabled.
> 
> The "tracking" events can be kept enabled by putting
> them on another event, but that requires an event
> that otherwise does nothing.  A new software event
> PERF_COUNT_SW_DUMMY is added for that purpose.
> 
> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>

Any comments?

> ---
>  include/uapi/linux/perf_event.h | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h
> index 62c25a2..9e28422 100644
> --- a/include/uapi/linux/perf_event.h
> +++ b/include/uapi/linux/perf_event.h
> @@ -109,6 +109,7 @@ enum perf_sw_ids {
>  	PERF_COUNT_SW_PAGE_FAULTS_MAJ		= 6,
>  	PERF_COUNT_SW_ALIGNMENT_FAULTS		= 7,
>  	PERF_COUNT_SW_EMULATION_FAULTS		= 8,
> +	PERF_COUNT_SW_DUMMY			= 9,
>  
>  	PERF_COUNT_SW_MAX,			/* non-ABI */
>  };
> 

Re: [PATCH 1/3] perf: add a dummy software event to keep tracking

[Arnaldo Carvalho de Melo]

Em Mon, Aug 26, 2013 at 04:15:34PM +0300, Adrian Hunter escreveu:
> On 07/08/13 16:30, Adrian Hunter wrote:
> > When an event is disabled the "tracking" events
> > selected by the 'mmap', 'comm' and 'task' bits of
> > struct perf_event_attr, are also disabled.  However,
> > the information those events provide is necessary to
> > resolve symbols for when the main event is re-enabled.
> > 
> > The "tracking" events can be kept enabled by putting
> > them on another event, but that requires an event
> > that otherwise does nothing.  A new software event
> > PERF_COUNT_SW_DUMMY is added for that purpose.
> > 
> > Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
> 
> Any comments?

Ok with me, Peter?

- Arnaldo
 
> > ---
> >  include/uapi/linux/perf_event.h | 1 +
> >  1 file changed, 1 insertion(+)
> > 
> > diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h
> > index 62c25a2..9e28422 100644
> > --- a/include/uapi/linux/perf_event.h
> > +++ b/include/uapi/linux/perf_event.h
> > @@ -109,6 +109,7 @@ enum perf_sw_ids {
> >  	PERF_COUNT_SW_PAGE_FAULTS_MAJ		= 6,
> >  	PERF_COUNT_SW_ALIGNMENT_FAULTS		= 7,
> >  	PERF_COUNT_SW_EMULATION_FAULTS		= 8,
> > +	PERF_COUNT_SW_DUMMY			= 9,
> >  
> >  	PERF_COUNT_SW_MAX,			/* non-ABI */
> >  };
> > 

Re: [PATCH 1/3] perf: add a dummy software event to keep tracking

[Peter Zijlstra]

On Mon, Aug 26, 2013 at 10:53:05AM -0300, Arnaldo Carvalho de Melo wrote:
> Em Mon, Aug 26, 2013 at 04:15:34PM +0300, Adrian Hunter escreveu:
> > On 07/08/13 16:30, Adrian Hunter wrote:
> > > When an event is disabled the "tracking" events
> > > selected by the 'mmap', 'comm' and 'task' bits of
> > > struct perf_event_attr, are also disabled.  However,
> > > the information those events provide is necessary to
> > > resolve symbols for when the main event is re-enabled.
> > > 
> > > The "tracking" events can be kept enabled by putting
> > > them on another event, but that requires an event
> > > that otherwise does nothing.  A new software event
> > > PERF_COUNT_SW_DUMMY is added for that purpose.
> > > 
> > > Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
> > 
> > Any comments?
> 
> Ok with me, Peter?

Yeah I suppose so.. I don't really like it and I've made a start at
looking at the problem twice now but never got anywhere, so sure lets go
with this.

Acked-by: Peter Zijlstra <peterz@infradead.org>

[PATCH 2/3] perf tools: add support for PERF_COUNT_SW_DUMMY

[Adrian Hunter]

Add support for the new dummy software event
PERF_COUNT_SW_DUMMY.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 tools/perf/util/parse-events.c | 4 ++++
 tools/perf/util/parse-events.l | 1 +
 tools/perf/util/python.c       | 1 +
 3 files changed, 6 insertions(+)

diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index dba877d..1ef81ea 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -108,6 +108,10 @@ static struct event_symbol event_symbols_sw[PERF_COUNT_SW_MAX] = {
 		.symbol = "emulation-faults",
 		.alias  = "",
 	},
+	[PERF_COUNT_SW_DUMMY] = {
+		.symbol = "dummy",
+		.alias  = "",
+	},
 };
 
 #define __PERF_EVENT_FIELD(config, name) \
diff --git a/tools/perf/util/parse-events.l b/tools/perf/util/parse-events.l
index b36115f..29c5d24 100644
--- a/tools/perf/util/parse-events.l
+++ b/tools/perf/util/parse-events.l
@@ -144,6 +144,7 @@ context-switches|cs				{ return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW
 cpu-migrations|migrations			{ return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_CPU_MIGRATIONS); }
 alignment-faults				{ return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_ALIGNMENT_FAULTS); }
 emulation-faults				{ return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_EMULATION_FAULTS); }
+dummy						{ return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_DUMMY); }
 
 L1-dcache|l1-d|l1d|L1-data		|
 L1-icache|l1-i|l1i|L1-instruction	|
diff --git a/tools/perf/util/python.c b/tools/perf/util/python.c
index 925e0c3..2fa83c0 100644
--- a/tools/perf/util/python.c
+++ b/tools/perf/util/python.c
@@ -967,6 +967,7 @@ static struct {
 	{ "COUNT_SW_PAGE_FAULTS_MAJ",  PERF_COUNT_SW_PAGE_FAULTS_MAJ },
 	{ "COUNT_SW_ALIGNMENT_FAULTS", PERF_COUNT_SW_ALIGNMENT_FAULTS },
 	{ "COUNT_SW_EMULATION_FAULTS", PERF_COUNT_SW_EMULATION_FAULTS },
+	{ "COUNT_SW_DUMMY",            PERF_COUNT_SW_DUMMY },
 
 	{ "SAMPLE_IP",	      PERF_SAMPLE_IP },
 	{ "SAMPLE_TID",	      PERF_SAMPLE_TID },
-- 
1.7.11.7

[PATCH 3/3] perf tools: add 'keep tracking' test

[Adrian Hunter]

Add a test for the newly added PERF_COUNT_SW_DUMMY event.
The test checks that tracking events continue when an
event is disabled but a dummy software event is not
disabled.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 tools/perf/Makefile              |   1 +
 tools/perf/tests/builtin-test.c  |   4 ++
 tools/perf/tests/keep-tracking.c | 150 +++++++++++++++++++++++++++++++++++++++
 tools/perf/tests/tests.h         |   1 +
 tools/perf/util/evlist.c         |  42 ++++++++++-
 tools/perf/util/evlist.h         |   5 ++
 6 files changed, 201 insertions(+), 2 deletions(-)
 create mode 100644 tools/perf/tests/keep-tracking.c

diff --git a/tools/perf/Makefile b/tools/perf/Makefile
index bfd12d0..0193e7c 100644
--- a/tools/perf/Makefile
+++ b/tools/perf/Makefile
@@ -392,6 +392,7 @@ LIB_OBJS += $(OUTPUT)tests/sw-clock.o
 ifeq ($(ARCH),x86)
 LIB_OBJS += $(OUTPUT)tests/perf-time-to-tsc.o
 endif
+LIB_OBJS += $(OUTPUT)tests/keep-tracking.o
 
 BUILTIN_OBJS += $(OUTPUT)builtin-annotate.o
 BUILTIN_OBJS += $(OUTPUT)builtin-bench.o
diff --git a/tools/perf/tests/builtin-test.c b/tools/perf/tests/builtin-test.c
index b7b4049..2a468a1 100644
--- a/tools/perf/tests/builtin-test.c
+++ b/tools/perf/tests/builtin-test.c
@@ -100,6 +100,10 @@ static struct test {
 	},
 #endif
 	{
+		.desc = "Test using a dummy software event to keep tracking",
+		.func = test__keep_tracking,
+	},
+	{
 		.func = NULL,
 	},
 };
diff --git a/tools/perf/tests/keep-tracking.c b/tools/perf/tests/keep-tracking.c
new file mode 100644
index 0000000..74abe00
--- /dev/null
+++ b/tools/perf/tests/keep-tracking.c
@@ -0,0 +1,150 @@
+#include <sys/types.h>
+#include <unistd.h>
+#include <sys/prctl.h>
+
+#include "parse-events.h"
+#include "evlist.h"
+#include "evsel.h"
+#include "thread_map.h"
+#include "cpumap.h"
+#include "tests.h"
+
+#define CHECK__(x) {				\
+	while ((x) < 0) {			\
+		pr_debug(#x " failed!\n");	\
+		goto out_err;			\
+	}					\
+}
+
+#define CHECK_NOT_NULL__(x) {			\
+	while ((x) == NULL) {			\
+		pr_debug(#x " failed!\n");	\
+		goto out_err;			\
+	}					\
+}
+
+static int find_comm(struct perf_evlist *evlist, const char *comm)
+{
+	union perf_event *event;
+	int i, found;
+
+	found = 0;
+	for (i = 0; i < evlist->nr_mmaps; i++) {
+		while ((event = perf_evlist__mmap_read(evlist, i)) != NULL) {
+			if (event->header.type == PERF_RECORD_COMM &&
+			    (pid_t)event->comm.pid == getpid() &&
+			    (pid_t)event->comm.tid == getpid() &&
+			    strcmp(event->comm.comm, comm) == 0)
+				found += 1;
+		}
+	}
+	return found;
+}
+
+/**
+ * test__keep_tracking - test using a dummy software event to keep tracking.
+ *
+ * This function implements a test that checks that tracking events continue
+ * when an event is disabled but a dummy software event is not disabled.  If the
+ * test passes %0 is returned, otherwise %-1 is returned.
+ */
+int test__keep_tracking(void)
+{
+	struct perf_record_opts opts = {
+		.mmap_pages	     = UINT_MAX,
+		.user_freq	     = UINT_MAX,
+		.user_interval	     = ULLONG_MAX,
+		.freq		     = 4000,
+		.target		     = {
+			.uses_mmap   = true,
+		},
+	};
+	struct thread_map *threads = NULL;
+	struct cpu_map *cpus = NULL;
+	struct perf_evlist *evlist = NULL;
+	struct perf_evsel *evsel = NULL;
+	int found, err = -1;
+	const char *comm;
+
+	threads = thread_map__new(-1, getpid(), UINT_MAX);
+	CHECK_NOT_NULL__(threads);
+
+	cpus = cpu_map__new(NULL);
+	CHECK_NOT_NULL__(cpus);
+
+	evlist = perf_evlist__new();
+	CHECK_NOT_NULL__(evlist);
+
+	perf_evlist__set_maps(evlist, cpus, threads);
+
+	CHECK__(parse_events(evlist, "dummy:u"));
+	CHECK__(parse_events(evlist, "cycles:u"));
+
+	perf_evlist__config(evlist, &opts);
+
+	evsel = perf_evlist__first(evlist);
+
+	evsel->attr.comm = 1;
+	evsel->attr.disabled = 1;
+	evsel->attr.enable_on_exec = 0;
+
+	CHECK__(perf_evlist__open(evlist));
+
+	CHECK__(perf_evlist__mmap(evlist, UINT_MAX, false));
+
+	/*
+	 * First, test that a 'comm' event can be found when the event is
+	 * enabled.
+	 */
+
+	perf_evlist__enable(evlist);
+
+	comm = "Test COMM 1";
+	CHECK__(prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0));
+
+	perf_evlist__disable(evlist);
+
+	found = find_comm(evlist, comm);
+	if (found != 1) {
+		pr_debug("First time, failed to find tracking event.\n");
+		goto out_err;
+	}
+
+	/*
+	 * Secondly, test that a 'comm' event can be found when the event is
+	 * disabled with the dummy event still enabled.
+	 */
+
+	perf_evlist__enable(evlist);
+
+	evsel = perf_evlist__last(evlist);
+
+	CHECK__(perf_evlist__disable_event(evlist, evsel));
+
+	comm = "Test COMM 2";
+	CHECK__(prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0));
+
+	perf_evlist__disable(evlist);
+
+	found = find_comm(evlist, comm);
+	if (found != 1) {
+		pr_debug("Seconf time, failed to find tracking event.\n");
+		goto out_err;
+	}
+
+	err = 0;
+
+out_err:
+	if (evlist) {
+		perf_evlist__disable(evlist);
+		perf_evlist__munmap(evlist);
+		perf_evlist__close(evlist);
+		perf_evlist__delete(evlist);
+	}
+	if (cpus)
+		cpu_map__delete(cpus);
+	if (threads)
+		thread_map__delete(threads);
+
+	return err;
+}
diff --git a/tools/perf/tests/tests.h b/tools/perf/tests/tests.h
index d22202a..3cf420c 100644
--- a/tools/perf/tests/tests.h
+++ b/tools/perf/tests/tests.h
@@ -36,5 +36,6 @@ int test__bp_signal_overflow(void);
 int test__task_exit(void);
 int test__sw_clock_freq(void);
 int test__perf_time_to_tsc(void);
+int test__keep_tracking(void);
 
 #endif /* TESTS_H */
diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index da2dd92..c9b4ef3 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -232,7 +232,7 @@ void perf_evlist__disable(struct perf_evlist *evlist)
 
 	for (cpu = 0; cpu < nr_cpus; cpu++) {
 		list_for_each_entry(pos, &evlist->entries, node) {
-			if (!perf_evsel__is_group_leader(pos))
+			if (!perf_evsel__is_group_leader(pos) || !pos->fd)
 				continue;
 			for (thread = 0; thread < nr_threads; thread++)
 				ioctl(FD(pos, cpu, thread),
@@ -250,7 +250,7 @@ void perf_evlist__enable(struct perf_evlist *evlist)
 
 	for (cpu = 0; cpu < nr_cpus; cpu++) {
 		list_for_each_entry(pos, &evlist->entries, node) {
-			if (!perf_evsel__is_group_leader(pos))
+			if (!perf_evsel__is_group_leader(pos) || !pos->fd)
 				continue;
 			for (thread = 0; thread < nr_threads; thread++)
 				ioctl(FD(pos, cpu, thread),
@@ -259,6 +259,44 @@ void perf_evlist__enable(struct perf_evlist *evlist)
 	}
 }
 
+int perf_evlist__disable_event(struct perf_evlist *evlist,
+			       struct perf_evsel *evsel)
+{
+	int cpu, thread, err;
+
+	if (!evsel->fd)
+		return 0;
+
+	for (cpu = 0; cpu < evlist->cpus->nr; cpu++) {
+		for (thread = 0; thread < evlist->threads->nr; thread++) {
+			err = ioctl(FD(evsel, cpu, thread),
+				    PERF_EVENT_IOC_DISABLE, 0);
+			if (err)
+				return err;
+		}
+	}
+	return 0;
+}
+
+int perf_evlist__enable_event(struct perf_evlist *evlist,
+			      struct perf_evsel *evsel)
+{
+	int cpu, thread, err;
+
+	if (!evsel->fd)
+		return -EINVAL;
+
+	for (cpu = 0; cpu < evlist->cpus->nr; cpu++) {
+		for (thread = 0; thread < evlist->threads->nr; thread++) {
+			err = ioctl(FD(evsel, cpu, thread),
+				    PERF_EVENT_IOC_ENABLE, 0);
+			if (err)
+				return err;
+		}
+	}
+	return 0;
+}
+
 static int perf_evlist__alloc_pollfd(struct perf_evlist *evlist)
 {
 	int nr_cpus = cpu_map__nr(evlist->cpus);
diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h
index 327abab..825c90a 100644
--- a/tools/perf/util/evlist.h
+++ b/tools/perf/util/evlist.h
@@ -101,6 +101,11 @@ void perf_evlist__munmap(struct perf_evlist *evlist);
 void perf_evlist__disable(struct perf_evlist *evlist);
 void perf_evlist__enable(struct perf_evlist *evlist);
 
+int perf_evlist__disable_event(struct perf_evlist *evlist,
+			       struct perf_evsel *evsel);
+int perf_evlist__enable_event(struct perf_evlist *evlist,
+			      struct perf_evsel *evsel);
+
 void perf_evlist__set_selected(struct perf_evlist *evlist,
 			       struct perf_evsel *evsel);
 
-- 
1.7.11.7

Re: [PATCH 3/3] perf tools: add 'keep tracking' test

[Namhyung Kim]

On Wed,  7 Aug 2013 16:30:31 +0300, Adrian Hunter wrote:
> Add a test for the newly added PERF_COUNT_SW_DUMMY event.
> The test checks that tracking events continue when an
> event is disabled but a dummy software event is not
> disabled.

[SNIP]

> +int test__keep_tracking(void)
> +{
> +	struct perf_record_opts opts = {
> +		.mmap_pages	     = UINT_MAX,
> +		.user_freq	     = UINT_MAX,
> +		.user_interval	     = ULLONG_MAX,
> +		.freq		     = 4000,
> +		.target		     = {
> +			.uses_mmap   = true,
> +		},
> +	};
> +	struct thread_map *threads = NULL;
> +	struct cpu_map *cpus = NULL;
> +	struct perf_evlist *evlist = NULL;
> +	struct perf_evsel *evsel = NULL;
> +	int found, err = -1;
> +	const char *comm;
> +
> +	threads = thread_map__new(-1, getpid(), UINT_MAX);
> +	CHECK_NOT_NULL__(threads);
> +
> +	cpus = cpu_map__new(NULL);
> +	CHECK_NOT_NULL__(cpus);
> +
> +	evlist = perf_evlist__new();
> +	CHECK_NOT_NULL__(evlist);
> +
> +	perf_evlist__set_maps(evlist, cpus, threads);
> +
> +	CHECK__(parse_events(evlist, "dummy:u"));
> +	CHECK__(parse_events(evlist, "cycles:u"));
> +
> +	perf_evlist__config(evlist, &opts);
> +
> +	evsel = perf_evlist__first(evlist);
> +
> +	evsel->attr.comm = 1;
> +	evsel->attr.disabled = 1;
> +	evsel->attr.enable_on_exec = 0;
> +
> +	CHECK__(perf_evlist__open(evlist));
> +
> +	CHECK__(perf_evlist__mmap(evlist, UINT_MAX, false));
> +
> +	/*
> +	 * First, test that a 'comm' event can be found when the event is
> +	 * enabled.
> +	 */
> +
> +	perf_evlist__enable(evlist);
> +
> +	comm = "Test COMM 1";
> +	CHECK__(prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0));
> +
> +	perf_evlist__disable(evlist);
> +
> +	found = find_comm(evlist, comm);
> +	if (found != 1) {
> +		pr_debug("First time, failed to find tracking event.\n");
> +		goto out_err;
> +	}
> +
> +	/*
> +	 * Secondly, test that a 'comm' event can be found when the event is
> +	 * disabled with the dummy event still enabled.
> +	 */
> +
> +	perf_evlist__enable(evlist);
> +
> +	evsel = perf_evlist__last(evlist);
> +
> +	CHECK__(perf_evlist__disable_event(evlist, evsel));
> +
> +	comm = "Test COMM 2";
> +	CHECK__(prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0));
> +
> +	perf_evlist__disable(evlist);
> +
> +	found = find_comm(evlist, comm);
> +	if (found != 1) {
> +		pr_debug("Seconf time, failed to find tracking event.\n");
> +		goto out_err;
> +	}
> +
> +	err = 0;
> +
> +out_err:
> +	if (evlist) {
> +		perf_evlist__disable(evlist);

It seems the both of goto out_err already called perf_evlist__disable().

Thanks,
Namhyung


> +		perf_evlist__munmap(evlist);
> +		perf_evlist__close(evlist);
> +		perf_evlist__delete(evlist);
> +	}
> +	if (cpus)
> +		cpu_map__delete(cpus);
> +	if (threads)
> +		thread_map__delete(threads);
> +
> +	return err;
> +}

Re: [PATCH 3/3] perf tools: add 'keep tracking' test

[Adrian Hunter]

On 13/08/13 05:33, Namhyung Kim wrote:
> On Wed,  7 Aug 2013 16:30:31 +0300, Adrian Hunter wrote:
>> Add a test for the newly added PERF_COUNT_SW_DUMMY event.
>> The test checks that tracking events continue when an
>> event is disabled but a dummy software event is not
>> disabled.
> 
> [SNIP]
> 
>> +int test__keep_tracking(void)
>> +{
>> +	struct perf_record_opts opts = {
>> +		.mmap_pages	     = UINT_MAX,
>> +		.user_freq	     = UINT_MAX,
>> +		.user_interval	     = ULLONG_MAX,
>> +		.freq		     = 4000,
>> +		.target		     = {
>> +			.uses_mmap   = true,
>> +		},
>> +	};
>> +	struct thread_map *threads = NULL;
>> +	struct cpu_map *cpus = NULL;
>> +	struct perf_evlist *evlist = NULL;
>> +	struct perf_evsel *evsel = NULL;
>> +	int found, err = -1;
>> +	const char *comm;
>> +
>> +	threads = thread_map__new(-1, getpid(), UINT_MAX);
>> +	CHECK_NOT_NULL__(threads);
>> +
>> +	cpus = cpu_map__new(NULL);
>> +	CHECK_NOT_NULL__(cpus);
>> +
>> +	evlist = perf_evlist__new();
>> +	CHECK_NOT_NULL__(evlist);
>> +
>> +	perf_evlist__set_maps(evlist, cpus, threads);
>> +
>> +	CHECK__(parse_events(evlist, "dummy:u"));
>> +	CHECK__(parse_events(evlist, "cycles:u"));
>> +
>> +	perf_evlist__config(evlist, &opts);
>> +
>> +	evsel = perf_evlist__first(evlist);
>> +
>> +	evsel->attr.comm = 1;
>> +	evsel->attr.disabled = 1;
>> +	evsel->attr.enable_on_exec = 0;
>> +
>> +	CHECK__(perf_evlist__open(evlist));
>> +
>> +	CHECK__(perf_evlist__mmap(evlist, UINT_MAX, false));
>> +
>> +	/*
>> +	 * First, test that a 'comm' event can be found when the event is
>> +	 * enabled.
>> +	 */
>> +
>> +	perf_evlist__enable(evlist);
>> +
>> +	comm = "Test COMM 1";
>> +	CHECK__(prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0));
>> +
>> +	perf_evlist__disable(evlist);
>> +
>> +	found = find_comm(evlist, comm);
>> +	if (found != 1) {
>> +		pr_debug("First time, failed to find tracking event.\n");
>> +		goto out_err;
>> +	}
>> +
>> +	/*
>> +	 * Secondly, test that a 'comm' event can be found when the event is
>> +	 * disabled with the dummy event still enabled.
>> +	 */
>> +
>> +	perf_evlist__enable(evlist);
>> +
>> +	evsel = perf_evlist__last(evlist);
>> +
>> +	CHECK__(perf_evlist__disable_event(evlist, evsel));
>> +
>> +	comm = "Test COMM 2";
>> +	CHECK__(prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0));
>> +
>> +	perf_evlist__disable(evlist);
>> +
>> +	found = find_comm(evlist, comm);
>> +	if (found != 1) {
>> +		pr_debug("Seconf time, failed to find tracking event.\n");
>> +		goto out_err;
>> +	}
>> +
>> +	err = 0;
>> +
>> +out_err:
>> +	if (evlist) {
>> +		perf_evlist__disable(evlist);
> 
> It seems the both of goto out_err already called perf_evlist__disable().

There are more "goto out_err"s in the CHECK__ macro

> 
> Thanks,
> Namhyung
> 
> 
>> +		perf_evlist__munmap(evlist);
>> +		perf_evlist__close(evlist);
>> +		perf_evlist__delete(evlist);
>> +	}
>> +	if (cpus)
>> +		cpu_map__delete(cpus);
>> +	if (threads)
>> +		thread_map__delete(threads);
>> +
>> +	return err;
>> +}
> 
> 

Re: [PATCH 0/3] perf: add a dummy software event to keep tracking

[Namhyung Kim]

Hi Adrian,

On Wed,  7 Aug 2013 16:30:28 +0300, Adrian Hunter wrote:
> Hi
>
> This is an alternative to the 'keep tracking' flag patch
> which is here:
>
> 	http://marc.info/?l=linux-kernel&m=137242545521246&w=2
>
> perf tools is updated and a test added to demonstrate the
> new event.

It looks like a perfect fit for tracking fork/mmap events on
perf-ftrace. :)

Thanks,
Namhyung