[PATCH] perf/stat: fix bug in handling events in error state

[PATCH] perf/stat: fix bug in handling events in error state

[David Carrillo-Cisneros]

From: Stephane Eranian <eranian@google.com>

(This is a patch has been sitting in the Intel CQM/CMT driver series
 for a while, despite not depend on it. Sending it now independently
 since the series is being discarded.)

When an event is in error state, read() returns 0
instead of sizeof() buffer. In certain modes, such
as interval printing, ignoring the 0 return value
may cause bogus count deltas to be computed and
thus invalid results printed.

This patch fixes this problem by modifying read_counters()
to mark the event as not scaled (scaled = -1) to force
the printout routine to show <NOT COUNTED>.

Signed-off-by: Stephane Eranian <eranian@google.com>
Reviewed-by: David Carrillo-Cisneros <davidcc@google.com>
---
 tools/perf/builtin-stat.c | 12 +++++++++---
 tools/perf/util/evsel.c   |  4 ++--
 2 files changed, 11 insertions(+), 5 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 688dea7cb08f..c3c4b4990f3c 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -310,8 +310,12 @@ static int read_counter(struct perf_evsel *counter)
 			struct perf_counts_values *count;
 
 			count = perf_counts(counter->counts, cpu, thread);
-			if (perf_evsel__read(counter, cpu, thread, count))
+			if (perf_evsel__read(counter, cpu, thread, count)) {
+				counter->counts->scaled = -1;
+				perf_counts(counter->counts, cpu, thread)->ena = 0;
+				perf_counts(counter->counts, cpu, thread)->run = 0;
 				return -1;
+			}
 
 			if (STAT_RECORD) {
 				if (perf_evsel__write_stat_event(counter, cpu, thread, count)) {
@@ -336,12 +340,14 @@ static int read_counter(struct perf_evsel *counter)
 static void read_counters(void)
 {
 	struct perf_evsel *counter;
+	int ret;
 
 	evlist__for_each_entry(evsel_list, counter) {
-		if (read_counter(counter))
+		ret = read_counter(counter);
+		if (ret)
 			pr_debug("failed to read counter %s\n", counter->name);
 
-		if (perf_stat_process_counter(&stat_config, counter))
+		if (ret == 0 && perf_stat_process_counter(&stat_config, counter))
 			pr_warning("failed to process counter %s\n", counter->name);
 	}
 }
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 8bc271141d9d..d54efb5ee8bc 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -1221,7 +1221,7 @@ int perf_evsel__read(struct perf_evsel *evsel, int cpu, int thread,
 	if (FD(evsel, cpu, thread) < 0)
 		return -EINVAL;
 
-	if (readn(FD(evsel, cpu, thread), count, sizeof(*count)) < 0)
+	if (readn(FD(evsel, cpu, thread), count, sizeof(*count)) <= 0)
 		return -errno;
 
 	return 0;
@@ -1239,7 +1239,7 @@ int __perf_evsel__read_on_cpu(struct perf_evsel *evsel,
 	if (evsel->counts == NULL && perf_evsel__alloc_counts(evsel, cpu + 1, thread + 1) < 0)
 		return -ENOMEM;
 
-	if (readn(FD(evsel, cpu, thread), &count, nv * sizeof(u64)) < 0)
+	if (readn(FD(evsel, cpu, thread), &count, nv * sizeof(u64)) <= 0)
 		return -errno;
 
 	perf_evsel__compute_deltas(evsel, cpu, thread, &count);
-- 
2.12.2.715.g7642488e1d-goog

Re: [PATCH] perf/stat: fix bug in handling events in error state

[Jiri Olsa]

On Wed, Apr 12, 2017 at 11:23:01AM -0700, David Carrillo-Cisneros wrote:
> From: Stephane Eranian <eranian@google.com>
> 
> (This is a patch has been sitting in the Intel CQM/CMT driver series
>  for a while, despite not depend on it. Sending it now independently
>  since the series is being discarded.)
> 
> When an event is in error state, read() returns 0
> instead of sizeof() buffer. In certain modes, such
> as interval printing, ignoring the 0 return value
> may cause bogus count deltas to be computed and
> thus invalid results printed.
> 
> This patch fixes this problem by modifying read_counters()
> to mark the event as not scaled (scaled = -1) to force
> the printout routine to show <NOT COUNTED>.
> 
> Signed-off-by: Stephane Eranian <eranian@google.com>
> Reviewed-by: David Carrillo-Cisneros <davidcc@google.com>

Acked-by: Jiri Olsa <jolsa@kernel.org>

thanks,
jirka

[tip:perf/core] perf stat: Fix bug in handling events in error state

[tip-bot for Stephane Eranian]

Commit-ID:  db49a71798a38f3ddf3f3462703328dca39b1ac7
Gitweb:     http://git.kernel.org/tip/db49a71798a38f3ddf3f3462703328dca39b1ac7
Author:     Stephane Eranian <eranian@google.com>
AuthorDate: Wed, 12 Apr 2017 11:23:01 -0700
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Thu, 13 Apr 2017 10:40:36 -0300

perf stat: Fix bug in handling events in error state

(This is a patch has been sitting in the Intel CQM/CMT driver series for
 a while, despite not depend on it. Sending it now independently since
 the series is being discarded.)

When an event is in error state, read() returns 0 instead of sizeof()
buffer. In certain modes, such as interval printing, ignoring the 0
return value may cause bogus count deltas to be computed and thus
invalid results printed.

This patch fixes this problem by modifying read_counters() to mark the
event as not scaled (scaled = -1) to force the printout routine to show
<NOT COUNTED>.

Signed-off-by: Stephane Eranian <eranian@google.com>
Reviewed-by: David Carrillo-Cisneros <davidcc@google.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Paul Turner <pjt@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/20170412182301.44406-1-davidcc@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-stat.c | 12 +++++++++---
 tools/perf/util/evsel.c   |  4 ++--
 2 files changed, 11 insertions(+), 5 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 868e086a..610225b 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -312,8 +312,12 @@ static int read_counter(struct perf_evsel *counter)
 			struct perf_counts_values *count;
 
 			count = perf_counts(counter->counts, cpu, thread);
-			if (perf_evsel__read(counter, cpu, thread, count))
+			if (perf_evsel__read(counter, cpu, thread, count)) {
+				counter->counts->scaled = -1;
+				perf_counts(counter->counts, cpu, thread)->ena = 0;
+				perf_counts(counter->counts, cpu, thread)->run = 0;
 				return -1;
+			}
 
 			if (STAT_RECORD) {
 				if (perf_evsel__write_stat_event(counter, cpu, thread, count)) {
@@ -338,12 +342,14 @@ static int read_counter(struct perf_evsel *counter)
 static void read_counters(void)
 {
 	struct perf_evsel *counter;
+	int ret;
 
 	evlist__for_each_entry(evsel_list, counter) {
-		if (read_counter(counter))
+		ret = read_counter(counter);
+		if (ret)
 			pr_debug("failed to read counter %s\n", counter->name);
 
-		if (perf_stat_process_counter(&stat_config, counter))
+		if (ret == 0 && perf_stat_process_counter(&stat_config, counter))
 			pr_warning("failed to process counter %s\n", counter->name);
 	}
 }
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 8f5d86b..3779b9f 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -1239,7 +1239,7 @@ int perf_evsel__read(struct perf_evsel *evsel, int cpu, int thread,
 	if (FD(evsel, cpu, thread) < 0)
 		return -EINVAL;
 
-	if (readn(FD(evsel, cpu, thread), count, sizeof(*count)) < 0)
+	if (readn(FD(evsel, cpu, thread), count, sizeof(*count)) <= 0)
 		return -errno;
 
 	return 0;
@@ -1257,7 +1257,7 @@ int __perf_evsel__read_on_cpu(struct perf_evsel *evsel,
 	if (evsel->counts == NULL && perf_evsel__alloc_counts(evsel, cpu + 1, thread + 1) < 0)
 		return -ENOMEM;
 
-	if (readn(FD(evsel, cpu, thread), &count, nv * sizeof(u64)) < 0)
+	if (readn(FD(evsel, cpu, thread), &count, nv * sizeof(u64)) <= 0)
 		return -errno;
 
 	perf_evsel__compute_deltas(evsel, cpu, thread, &count);