perf report: Add weight[123] output fields

Add weight1, weight2 and weight3 fields to -F/--fields and their aliases
like 'ins_lat', 'p_stage_cyc' and 'retire_lat'.  Note that they are in
the sort keys too but the difference is that output fields will sum up
the weight values and display the average.

In the sort key, users can see the distribution of weight value and I
think it's confusing we have local vs. global weight for the same weight.

For example, I experiment with mem-loads events to get the weights.  On
my laptop, it seems only weight1 field is supported.

  $ perf mem record -- perf test -w noploop

Let's look at the noploop function only.  It has 7 samples.

  $ perf script -F event,ip,sym,weight | grep noploop
  # event                         weight     ip           sym
  cpu/mem-loads,ldlat=30/P:           43     55b3c122bffc noploop
  cpu/mem-loads,ldlat=30/P:           48     55b3c122bffc noploop
  cpu/mem-loads,ldlat=30/P:           38     55b3c122bffc noploop    <--- same weight
  cpu/mem-loads,ldlat=30/P:           38     55b3c122bffc noploop    <--- same weight
  cpu/mem-loads,ldlat=30/P:           59     55b3c122bffc noploop
  cpu/mem-loads,ldlat=30/P:           33     55b3c122bffc noploop
  cpu/mem-loads,ldlat=30/P:           38     55b3c122bffc noploop    <--- same weight

When you use the 'weight' sort key, it'd show entries with a separate
weight value separately.  Also note that the first entry has 3 samples
with weight value 38, so they are displayed together and the weight
value is the sum of 3 samples (114 = 38 * 3).

  $ perf report -n -s +weight | grep -e Weight -e noploop
  # Overhead  Samples  Command   Shared Object   Symbol         Weight
       0.53%        3     perf   perf            [.] noploop    114
       0.18%        1     perf   perf            [.] noploop    59
       0.18%        1     perf   perf            [.] noploop    48
       0.18%        1     perf   perf            [.] noploop    43
       0.18%        1     perf   perf            [.] noploop    33

If you use 'local_weight' sort key, you can see the actual weight.

  $ perf report -n -s +local_weight | grep -e Weight -e noploop
  # Overhead  Samples  Command   Shared Object   Symbol         Local Weight
       0.53%        3     perf   perf            [.] noploop    38
       0.18%        1     perf   perf            [.] noploop    59
       0.18%        1     perf   perf            [.] noploop    48
       0.18%        1     perf   perf            [.] noploop    43
       0.18%        1     perf   perf            [.] noploop    33

But when you use the -F/--field option instead, you can see the average
weight for the while noploop function (as it won't group samples by
weight value and use the default 'comm,dso,sym' sort keys).

  $ perf report -n -F +weight | grep -e Weight -e noploop
  Warning:
  --fields weight shows the average value unlike in the --sort key.
  # Overhead  Samples  Weight1  Command  Shared Object  Symbol
       1.23%        7     42.4  perf     perf           [.] noploop

The weight1 field shows the average value:
  (38 * 3 + 59 + 48 + 43 + 33) / 7 = 42.4

Also it'd show the warning that 'weight' field has the average value.
Using 'weight1' can remove the warning.

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: https://lore.kernel.org/r/20240411181718.2367948-3-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

1 files changed, 74 insertions, 18 deletions

diff --git a/tools/perf/ui/hist.c b/tools/perf/ui/hist.c
index 2bf959d08354..685ba2a54fd8 100644
--- a/tools/perf/ui/hist.c
+++ b/tools/perf/ui/hist.c
@@ -25,7 +25,7 @@
 
 static int __hpp__fmt(struct perf_hpp *hpp, struct hist_entry *he,
 		      hpp_field_fn get_field, const char *fmt, int len,
-		      hpp_snprint_fn print_fn, bool fmt_percent)
+		      hpp_snprint_fn print_fn, enum perf_hpp_fmt_type fmtype)
 {
 	int ret;
 	struct hists *hists = he->hists;
@@ -33,7 +33,7 @@ static int __hpp__fmt(struct perf_hpp *hpp, struct hist_entry *he,
 	char *buf = hpp->buf;
 	size_t size = hpp->size;
 
-	if (fmt_percent) {
+	if (fmtype == PERF_HPP_FMT_TYPE__PERCENT) {
 		double percent = 0.0;
 		u64 total = hists__total_period(hists);
 
@@ -41,8 +41,16 @@ static int __hpp__fmt(struct perf_hpp *hpp, struct hist_entry *he,
 			percent = 100.0 * get_field(he) / total;
 
 		ret = hpp__call_print_fn(hpp, print_fn, fmt, len, percent);
-	} else
+	} else if (fmtype == PERF_HPP_FMT_TYPE__AVERAGE) {
+		double average = 0;
+
+		if (he->stat.nr_events)
+			average = 1.0 * get_field(he) / he->stat.nr_events;
+
+		ret = hpp__call_print_fn(hpp, print_fn, fmt, len, average);
+	} else {
 		ret = hpp__call_print_fn(hpp, print_fn, fmt, len, get_field(he));
+	}
 
 	if (evsel__is_group_event(evsel)) {
 		int prev_idx, idx_delta;
@@ -54,6 +62,7 @@ static int __hpp__fmt(struct perf_hpp *hpp, struct hist_entry *he,
 		list_for_each_entry(pair, &he->pairs.head, pairs.node) {
 			u64 period = get_field(pair);
 			u64 total = hists__total_period(pair->hists);
+			int nr_samples = pair->stat.nr_events;
 
 			if (!total)
 				continue;
@@ -66,7 +75,7 @@ static int __hpp__fmt(struct perf_hpp *hpp, struct hist_entry *he,
 				 * zero-fill group members in the middle which
 				 * have no sample
 				 */
-				if (fmt_percent) {
+				if (fmtype != PERF_HPP_FMT_TYPE__RAW) {
 					ret += hpp__call_print_fn(hpp, print_fn,
 								  fmt, len, 0.0);
 				} else {
@@ -75,9 +84,14 @@ static int __hpp__fmt(struct perf_hpp *hpp, struct hist_entry *he,
 				}
 			}
 
-			if (fmt_percent) {
+			if (fmtype == PERF_HPP_FMT_TYPE__PERCENT) {
 				ret += hpp__call_print_fn(hpp, print_fn, fmt, len,
 							  100.0 * period / total);
+			} else if (fmtype == PERF_HPP_FMT_TYPE__AVERAGE) {
+				double avg = nr_samples ? (period / nr_samples) : 0;
+
+				ret += hpp__call_print_fn(hpp, print_fn, fmt,
+							  len, avg);
 			} else {
 				ret += hpp__call_print_fn(hpp, print_fn, fmt,
 							  len, period);
@@ -92,7 +106,7 @@ static int __hpp__fmt(struct perf_hpp *hpp, struct hist_entry *he,
 			/*
 			 * zero-fill group members at last which have no sample
 			 */
-			if (fmt_percent) {
+			if (fmtype != PERF_HPP_FMT_TYPE__RAW) {
 				ret += hpp__call_print_fn(hpp, print_fn,
 							  fmt, len, 0.0);
 			} else {
@@ -114,33 +128,35 @@ static int __hpp__fmt(struct perf_hpp *hpp, struct hist_entry *he,
 
 int hpp__fmt(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp,
 	     struct hist_entry *he, hpp_field_fn get_field,
-	     const char *fmtstr, hpp_snprint_fn print_fn, bool fmt_percent)
+	     const char *fmtstr, hpp_snprint_fn print_fn,
+	     enum perf_hpp_fmt_type fmtype)
 {
 	int len = fmt->user_len ?: fmt->len;
 
 	if (symbol_conf.field_sep) {
 		return __hpp__fmt(hpp, he, get_field, fmtstr, 1,
-				  print_fn, fmt_percent);
+				  print_fn, fmtype);
 	}
 
-	if (fmt_percent)
+	if (fmtype == PERF_HPP_FMT_TYPE__PERCENT)
 		len -= 2; /* 2 for a space and a % sign */
 	else
 		len -= 1;
 
-	return  __hpp__fmt(hpp, he, get_field, fmtstr, len, print_fn, fmt_percent);
+	return  __hpp__fmt(hpp, he, get_field, fmtstr, len, print_fn, fmtype);
 }
 
 int hpp__fmt_acc(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp,
 		 struct hist_entry *he, hpp_field_fn get_field,
-		 const char *fmtstr, hpp_snprint_fn print_fn, bool fmt_percent)
+		 const char *fmtstr, hpp_snprint_fn print_fn,
+		 enum perf_hpp_fmt_type fmtype)
 {
 	if (!symbol_conf.cumulate_callchain) {
 		int len = fmt->user_len ?: fmt->len;
 		return snprintf(hpp->buf, hpp->size, " %*s", len - 1, "N/A");
 	}
 
-	return hpp__fmt(fmt, hpp, he, get_field, fmtstr, print_fn, fmt_percent);
+	return hpp__fmt(fmt, hpp, he, get_field, fmtstr, print_fn, fmtype);
 }
 
 static int field_cmp(u64 field_a, u64 field_b)
@@ -350,7 +366,7 @@ static int hpp__color_##_type(struct perf_hpp_fmt *fmt,				\
 			      struct perf_hpp *hpp, struct hist_entry *he) 	\
 {										\
 	return hpp__fmt(fmt, hpp, he, he_get_##_field, " %*.2f%%",		\
-			hpp_color_scnprintf, true);				\
+			hpp_color_scnprintf, PERF_HPP_FMT_TYPE__PERCENT);	\
 }
 
 #define __HPP_ENTRY_PERCENT_FN(_type, _field)					\
@@ -358,7 +374,7 @@ static int hpp__entry_##_type(struct perf_hpp_fmt *fmt,				\
 			      struct perf_hpp *hpp, struct hist_entry *he) 	\
 {										\
 	return hpp__fmt(fmt, hpp, he, he_get_##_field, " %*.2f%%",		\
-			hpp_entry_scnprintf, true);				\
+			hpp_entry_scnprintf, PERF_HPP_FMT_TYPE__PERCENT);	\
 }
 
 #define __HPP_SORT_FN(_type, _field)						\
@@ -378,7 +394,7 @@ static int hpp__color_##_type(struct perf_hpp_fmt *fmt,				\
 			      struct perf_hpp *hpp, struct hist_entry *he) 	\
 {										\
 	return hpp__fmt_acc(fmt, hpp, he, he_get_acc_##_field, " %*.2f%%", 	\
-			    hpp_color_scnprintf, true);				\
+			    hpp_color_scnprintf, PERF_HPP_FMT_TYPE__PERCENT);	\
 }
 
 #define __HPP_ENTRY_ACC_PERCENT_FN(_type, _field)				\
@@ -386,7 +402,7 @@ static int hpp__entry_##_type(struct perf_hpp_fmt *fmt,				\
 			      struct perf_hpp *hpp, struct hist_entry *he) 	\
 {										\
 	return hpp__fmt_acc(fmt, hpp, he, he_get_acc_##_field, " %*.2f%%",	\
-			    hpp_entry_scnprintf, true);				\
+			    hpp_entry_scnprintf, PERF_HPP_FMT_TYPE__PERCENT);	\
 }
 
 #define __HPP_SORT_ACC_FN(_type, _field)					\
@@ -406,7 +422,7 @@ static int hpp__entry_##_type(struct perf_hpp_fmt *fmt,				\
 			      struct perf_hpp *hpp, struct hist_entry *he) 	\
 {										\
 	return hpp__fmt(fmt, hpp, he, he_get_raw_##_field, " %*"PRIu64, 	\
-			hpp_entry_scnprintf, false);				\
+			hpp_entry_scnprintf, PERF_HPP_FMT_TYPE__RAW);		\
 }
 
 #define __HPP_SORT_RAW_FN(_type, _field)					\
@@ -416,6 +432,26 @@ static int64_t hpp__sort_##_type(struct perf_hpp_fmt *fmt __maybe_unused, 	\
 	return __hpp__sort(a, b, he_get_raw_##_field);				\
 }
 
+#define __HPP_ENTRY_AVERAGE_FN(_type, _field)					\
+static u64 he_get_##_field(struct hist_entry *he)				\
+{										\
+	return he->stat._field;							\
+}										\
+										\
+static int hpp__entry_##_type(struct perf_hpp_fmt *fmt,				\
+			      struct perf_hpp *hpp, struct hist_entry *he) 	\
+{										\
+	return hpp__fmt(fmt, hpp, he, he_get_##_field, " %*.1f",		\
+			hpp_entry_scnprintf, PERF_HPP_FMT_TYPE__AVERAGE);	\
+}
+
+#define __HPP_SORT_AVERAGE_FN(_type, _field)					\
+static int64_t hpp__sort_##_type(struct perf_hpp_fmt *fmt __maybe_unused, 	\
+				 struct hist_entry *a, struct hist_entry *b) 	\
+{										\
+	return __hpp__sort(a, b, he_get_##_field);				\
+}
+
 
 #define HPP_PERCENT_FNS(_type, _field)					\
 __HPP_COLOR_PERCENT_FN(_type, _field)					\
@@ -431,6 +467,10 @@ __HPP_SORT_ACC_FN(_type, _field)
 __HPP_ENTRY_RAW_FN(_type, _field)					\
 __HPP_SORT_RAW_FN(_type, _field)
 
+#define HPP_AVERAGE_FNS(_type, _field)					\
+__HPP_ENTRY_AVERAGE_FN(_type, _field)					\
+__HPP_SORT_AVERAGE_FN(_type, _field)
+
 HPP_PERCENT_FNS(overhead, period)
 HPP_PERCENT_FNS(overhead_sys, period_sys)
 HPP_PERCENT_FNS(overhead_us, period_us)
@@ -441,6 +481,10 @@ HPP_PERCENT_ACC_FNS(overhead_acc, period)
 HPP_RAW_FNS(samples, nr_events)
 HPP_RAW_FNS(period, period)
 
+HPP_AVERAGE_FNS(weight1, weight1)
+HPP_AVERAGE_FNS(weight2, weight2)
+HPP_AVERAGE_FNS(weight3, weight3)
+
 static int64_t hpp__nop_cmp(struct perf_hpp_fmt *fmt __maybe_unused,
 			    struct hist_entry *a __maybe_unused,
 			    struct hist_entry *b __maybe_unused)
@@ -510,7 +554,10 @@ struct perf_hpp_fmt perf_hpp__format[] = {
 	HPP__COLOR_PRINT_FNS("guest usr", overhead_guest_us, OVERHEAD_GUEST_US),
 	HPP__COLOR_ACC_PRINT_FNS("Children", overhead_acc, OVERHEAD_ACC),
 	HPP__PRINT_FNS("Samples", samples, SAMPLES),
-	HPP__PRINT_FNS("Period", period, PERIOD)
+	HPP__PRINT_FNS("Period", period, PERIOD),
+	HPP__PRINT_FNS("Weight1", weight1, WEIGHT1),
+	HPP__PRINT_FNS("Weight2", weight2, WEIGHT2),
+	HPP__PRINT_FNS("Weight3", weight3, WEIGHT3),
 };
 
 struct perf_hpp_list perf_hpp_list = {
@@ -526,6 +573,7 @@ struct perf_hpp_list perf_hpp_list = {
 #undef HPP_PERCENT_FNS
 #undef HPP_PERCENT_ACC_FNS
 #undef HPP_RAW_FNS
+#undef HPP_AVERAGE_FNS
 
 #undef __HPP_HEADER_FN
 #undef __HPP_WIDTH_FN
@@ -534,9 +582,11 @@ struct perf_hpp_list perf_hpp_list = {
 #undef __HPP_COLOR_ACC_PERCENT_FN
 #undef __HPP_ENTRY_ACC_PERCENT_FN
 #undef __HPP_ENTRY_RAW_FN
+#undef __HPP_ENTRY_AVERAGE_FN
 #undef __HPP_SORT_FN
 #undef __HPP_SORT_ACC_FN
 #undef __HPP_SORT_RAW_FN
+#undef __HPP_SORT_AVERAGE_FN
 
 static void fmt_free(struct perf_hpp_fmt *fmt)
 {
@@ -785,6 +835,12 @@ void perf_hpp__reset_width(struct perf_hpp_fmt *fmt, struct hists *hists)
 		fmt->len = 12;
 		break;
 
+	case PERF_HPP__WEIGHT1:
+	case PERF_HPP__WEIGHT2:
+	case PERF_HPP__WEIGHT3:
+		fmt->len = 8;
+		break;
+
 	default:
 		break;
 	}