[PATCH RFC 05/10] perf,tools: Enable statistic read for perf record

[kan.liang@intel.com]

From: Kan Liang <kan.liang@intel.com>

Using 'N' event/group modifier to specify the event which want to read
counter statistics during sampling. For this event, the sampling will
be disabled unless it's group leader.
The 'N' modifier only be available for in system-wide/CPU mode. If a
group is marked as 'N' modifier, only group members read counter
statistics. Group leader always do sampling.
The other limit is that the first event cannot be stat read event, since
many tools special handle first event.

Signed-off-by: Kan Liang <kan.liang@intel.com>
---
 tools/perf/Documentation/perf-list.txt |  5 +++++
 tools/perf/builtin-record.c            | 31 +++++++++++++++++++++++++++++++
 tools/perf/util/evlist.c               |  3 +++
 3 files changed, 39 insertions(+)

diff --git a/tools/perf/Documentation/perf-list.txt b/tools/perf/Documentation/perf-list.txt
index bada893..133dc2c 100644
--- a/tools/perf/Documentation/perf-list.txt
+++ b/tools/perf/Documentation/perf-list.txt
@@ -31,6 +31,11 @@ counted. The following modifiers exist:
  H - host counting (not in KVM guests)
  p - precise level
  S - read sample value (PERF_SAMPLE_READ)
+ N - read counter statistics during sampling (Can be used to read
+     counter statistics of PMU_A event when PMU_B events are sampling.
+     For example, getting memory bandwidth by uncore events during the
+     CPU PMU events run time)
+     (Only available for group members or non-first event in system-wide/CPU mode)
  D - pin the event to the PMU
 
 The 'p' modifier can be used for specifying how precise the instruction
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index a0178bf..5b09318 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -273,11 +273,42 @@ static int record__open(struct record *rec)
 	struct perf_evlist *evlist = rec->evlist;
 	struct perf_session *session = rec->session;
 	struct record_opts *opts = &rec->opts;
+	struct perf_evsel *first = perf_evlist__first(evlist);
+	struct perf_event_attr *attr;
 	int rc = 0;
 
 	perf_evlist__config(evlist, opts);
 
 	evlist__for_each(evlist, pos) {
+
+		if (pos->stat_read) {
+			if (!target__has_cpu(&opts->target)) {
+				pos->stat_read = 0;
+				ui__warning("Statistics read only available "
+					    "on system-wide/CPU mode. "
+					    "Remove :N modifier for event %s\n",
+					    pos->name);
+				goto out;
+			}
+			/* Don't do stat read for Group leader */
+			if ((pos->leader == pos) && (pos->leader->nr_members > 1))
+				pos->stat_read = 0;
+			else {
+				if (first == pos) {
+					pos->stat_read = 0;
+					ui__warning("The first event cannot "
+						    "be stat read event\n");
+					goto out;
+				}
+				attr = &pos->attr;
+				attr->read_format = PERF_FORMAT_TOTAL_TIME_ENABLED |
+						    PERF_FORMAT_TOTAL_TIME_RUNNING;
+				attr->sample_freq   = 0;
+				attr->sample_period = 0;
+				attr->sample_type = 0;
+				pos->sample_size = 0;
+			}
+		}
 try_again:
 		if (perf_evsel__open(pos, pos->cpus, pos->threads) < 0) {
 			if (perf_evsel__fallback(pos, errno, msg, sizeof(msg))) {
diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index 373f65b..ca7bf8d 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -854,6 +854,9 @@ static int perf_evlist__mmap_per_evsel(struct perf_evlist *evlist, int idx,
 		if (evsel->system_wide && thread)
 			continue;
 
+		if (evsel->stat_read)
+			continue;
+
 		fd = FD(evsel, cpu, thread);
 
 		if (*output == -1) {
-- 
1.8.3.1