Another perf annotate weight patchkit

Another perf annotate weight patchkit

[Andi Kleen]

The v2 version had some bogus white space changes included. Now cleaned
up. This version is ready for review. Sorry for the quick repost.

[PATCH v3 1/4] perf tools record: Modernize -W man page

[Andi Kleen]

Modernize the -W / --weight description in the manpage to cover more
cases that are supported now.

Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 tools/perf/Documentation/perf-record.txt | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
index 178f483140ed..2c6510c5a87b 100644
--- a/tools/perf/Documentation/perf-record.txt
+++ b/tools/perf/Documentation/perf-record.txt
@@ -480,9 +480,11 @@ Note that this feature may not be available on all processors.
 
 -W::
 --weight::
-Enable weightened sampling. An additional weight is recorded per sample and can be
-displayed with the weight and local_weight sort keys.  This currently works for TSX
-abort events and some memory events in precise mode on modern Intel CPUs.
+Enable weighted sampling. One or more additional weights can be recorded per sample
+and can be displayed with the weight, weight2, weight3 and local_weight sort keys.
+Possible uses for weights are memory or instruction retirement latencies.
+Support depends on the PMU and the event, as well as its configuration. Typically
+it requires use of 'p'.
 
 --namespaces::
 Record events of type PERF_RECORD_NAMESPACES.  This enables 'cgroup_id' sort key.
-- 
2.55.0

[PATCH v3 2/4] perf tools top: Add --weight option

[Andi Kleen]

Add a -W/--weight option to perf top to collect weights too. Useful with
follow on patches.

Assisted-by: omp:GPT-5.6-Luna
Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 tools/perf/Documentation/perf-top.txt | 6 ++++++
 tools/perf/builtin-top.c              | 2 ++
 2 files changed, 8 insertions(+)

diff --git a/tools/perf/Documentation/perf-top.txt b/tools/perf/Documentation/perf-top.txt
index af3e4230c72f..2da2a16bbf26 100644
--- a/tools/perf/Documentation/perf-top.txt
+++ b/tools/perf/Documentation/perf-top.txt
@@ -25,6 +25,12 @@ OPTIONS
 --count=<count>::
 	Event period to sample.
 
+-W::
+--weight::
+Enable weighted sampling. One or more additional weights can be recorded per sample
+and can be displayed with the weight, weight2, weight3 and local_weight sort keys.
+Possible uses for weights are memory or instruction retirement latencies.
+
 -C <cpu-list>::
 --cpu=<cpu>::
 Monitor only on the list of CPUs provided. Multiple CPUs can be provided as a
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index 570410599f1b..2888ac3b7b92 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -1493,6 +1493,8 @@ int cmd_top(int argc, const char **argv)
 	OPT_CALLBACK(0, "filter", &top.evlist, "filter",
 		     "event filter", parse_filter),
 	OPT_U64('c', "count", &opts->user_interval, "event period to sample"),
+	OPT_BOOLEAN('W', "weight", &opts->sample_weight,
+		    "sample by weight (on special events only)"),
 	OPT_STRING('p', "pid", &target->pid, "pid",
 		    "profile events on existing process id"),
 	OPT_STRING('t', "tid", &target->tid, "tid",
-- 
2.55.0

[PATCH v3 3/4] perf tools: Add support for displaying weights in annotate

[Andi Kleen]

Add support for showing all the three possible per IP weights in
annotate. The weights are shown by defaults if any are non zero. This
is useful, especially with the new insn lat statistics, but also
for all the existing weights.

Add a hotkey to the interactive browser to turn them off (w), as well
as a perf annotate command line option.

The weights are stored unconditionally in the sym_hist_entry, which
will increase memory consumption somewhat.

Assisted-by: omp:GPT-5.6-Luna
Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 tools/perf/Documentation/perf-annotate.txt |   3 +
 tools/perf/builtin-annotate.c              |   2 +
 tools/perf/ui/browsers/annotate.c          |  17 +++-
 tools/perf/util/annotate.c                 | 110 ++++++++++++++++++---
 tools/perf/util/annotate.h                 |  32 +++++-
 tools/perf/util/symbol_conf.h              |  15 ++-
 6 files changed, 161 insertions(+), 18 deletions(-)

diff --git a/tools/perf/Documentation/perf-annotate.txt b/tools/perf/Documentation/perf-annotate.txt
index a688738809c4..50c6ea680d0d 100644
--- a/tools/perf/Documentation/perf-annotate.txt
+++ b/tools/perf/Documentation/perf-annotate.txt
@@ -79,6 +79,9 @@ include::itrace.txt[]
 
 --stdio:: Use the stdio interface.
 
+--no-weights:: Hide weight columns in annotation output. By default, weight columns
+	are shown when samples contain non-zero weights.
+
 --stdio2:: Use the stdio2 interface, non-interactive, uses the TUI formatting.
 
 --stdio-color=<mode>::
diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c
index 69cb72b2082a..7f489e2af333 100644
--- a/tools/perf/builtin-annotate.c
+++ b/tools/perf/builtin-annotate.c
@@ -719,6 +719,8 @@ int cmd_annotate(int argc, const char **argv)
 	OPT_BOOLEAN(0, "tui", &annotate.use_tui, "Use the TUI interface"),
 #endif
 	OPT_BOOLEAN(0, "stdio", &annotate.use_stdio, "Use the stdio interface"),
+	OPT_BOOLEAN(0, "no-weights", &symbol_conf.hide_weight,
+		    "Hide weight columns in annotation"),
 	OPT_BOOLEAN(0, "stdio2", &annotate.use_stdio2, "Use the stdio interface"),
 	OPT_BOOLEAN(0, "ignore-vmlinux", &symbol_conf.ignore_vmlinux,
                     "don't load vmlinux even if found"),
diff --git a/tools/perf/ui/browsers/annotate.c b/tools/perf/ui/browsers/annotate.c
index e47a46775089..d6126a7d51e4 100644
--- a/tools/perf/ui/browsers/annotate.c
+++ b/tools/perf/ui/browsers/annotate.c
@@ -189,7 +189,7 @@ static void annotate_browser__draw_current_jump(struct ui_browser *browser)
 	struct map_symbol *ms = ab->b.priv;
 	struct symbol *sym = ms->sym;
 	struct annotation *notes = symbol__annotation(sym);
-	u8 pcnt_width = annotation__pcnt_width(notes);
+	u8 pcnt_width = annotation__pcnt_width(notes, ab->evsel);
 	u8 cntr_width = annotation__br_cntr_width();
 	int width;
 	int diff = 0;
@@ -255,7 +255,8 @@ static unsigned int annotate_browser__refresh(struct ui_browser *browser)
 {
 	struct annotation *notes = browser__annotation(browser);
 	int ret = ui_browser__list_head_refresh(browser);
-	int pcnt_width = annotation__pcnt_width(notes);
+	int pcnt_width = annotation__pcnt_width(notes,
+		container_of(browser, struct annotate_browser, b)->evsel);
 
 	if (annotate_opts.jump_arrows)
 		annotate_browser__draw_current_jump(browser);
@@ -972,6 +973,7 @@ static int annotate_browser__run(struct annotate_browser *browser,
 		"O             Bump offset level (jump targets -> +call -> all -> cycle thru)\n"
 		"s             Toggle source code view\n"
 		"t             Circulate percent, total period, samples view\n"
+		"w             Toggle weight columns\n"
 		"c             Show min/max cycle\n"
 		"/             Search string\n"
 		"k             Toggle line numbers\n"
@@ -1091,6 +1093,15 @@ static int annotate_browser__run(struct annotate_browser *browser,
 				symbol_conf.show_total_period = true;
 			annotation__update_column_widths(notes);
 			continue;
+		case 'w':
+			symbol_conf.hide_weight = !symbol_conf.hide_weight;
+			symbol_conf.weight_mask_user = true;
+			symbol_conf.weight_mask = symbol_conf.hide_weight ?
+				0 : sym_hist__weight_mask(annotation__histogram(notes, evsel));
+			browser->b.width = notes->src->widths.max_line_len +
+					   annotation__pcnt_width(notes, evsel);
+			ui_browser__refresh_dimensions(&browser->b);
+			continue;
 		case 'c':
 			if (annotate_opts.show_minmax_cycle)
 				annotate_opts.show_minmax_cycle = false;
@@ -1228,7 +1239,7 @@ int __hist_entry__tui_annotate(struct hist_entry *he, struct map_symbol *ms,
 	browser.b.width = notes->src->widths.max_line_len;
 	browser.b.nr_entries = notes->src->nr_entries;
 	browser.b.entries = &notes->src->source;
-	browser.b.width += 18; /* Percentage */
+	browser.b.width += annotation__pcnt_width(notes, evsel);
 
 	if (annotate_opts.hide_src_code)
 		ui_browser__init_asm_mode(&browser.b);
diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c
index df70e95a8470..d1b691a61253 100644
--- a/tools/perf/util/annotate.c
+++ b/tools/perf/util/annotate.c
@@ -222,6 +222,7 @@ static int __symbol__inc_addr_samples(struct map_symbol *ms,
 	u64 offset;
 	struct sym_hist *h;
 	struct sym_hist_entry *entry;
+	u64 weight = sample->weight ?: sample->ins_lat ?: sample->weight3;
 
 	pr_debug3("%s: addr=%#" PRIx64 "\n", __func__, map__unmap_ip(ms->map, addr));
 
@@ -256,10 +257,28 @@ static int __symbol__inc_addr_samples(struct map_symbol *ms,
 	entry->nr_samples++;
 	entry->period += sample->period;
 
+	if (sample->evsel->core.attr.sample_type & PERF_SAMPLE_WEIGHT_TYPE) {
+		entry->weight_sum[WEIGHT_WEIGHT] += sample->weight;
+		entry->weight_num[WEIGHT_WEIGHT]++;
+		if (sample->weight)
+			sym_hist__set_weight_mask(h, BIT(WEIGHT_WEIGHT));
+	}
+	if (sample->evsel->core.attr.sample_type & PERF_SAMPLE_WEIGHT_STRUCT) {
+		entry->weight_sum[WEIGHT_INSNLAT] += sample->ins_lat;
+		entry->weight_num[WEIGHT_INSNLAT]++;
+		entry->weight_sum[WEIGHT_WEIGHT3] += sample->weight3;
+		entry->weight_num[WEIGHT_WEIGHT3]++;
+		if (sample->ins_lat)
+			sym_hist__set_weight_mask(h, BIT(WEIGHT_INSNLAT));
+		if (sample->weight3)
+			sym_hist__set_weight_mask(h, BIT(WEIGHT_WEIGHT3));
+	}
+
 	pr_debug3("%#" PRIx64 " %s: period++ [addr: %#" PRIx64 ", %#" PRIx64
-		  ", evidx=%d] => nr_samples: %" PRIu64 ", period: %" PRIu64 "\n",
+		  ", evidx=%d] => nr_samples: %" PRIu64 ", period: %" PRIu64
+		  " weight %" PRIu64 "\n",
 		  sym->start, sym->name, addr, addr - sym->start, evsel->core.idx,
-		  entry->nr_samples, entry->period);
+		  entry->nr_samples, entry->period, weight);
 	return 0;
 }
 
@@ -778,6 +797,41 @@ static bool needs_type_info(struct annotated_data_type *data_type)
 	return (data_type != &stackop_type) && (data_type != &canary_type);
 }
 
+static const char *annotation__weight_mode_str(enum symbol__weight_mode mode)
+{
+	switch (mode) {
+	case WEIGHT_NONE:
+		return "";
+	case WEIGHT_WEIGHT:
+		return "Weight";
+	case WEIGHT_INSNLAT:
+		return "InsnLat";
+	case WEIGHT_WEIGHT3:
+		return "Weight3";
+	default:
+		return "";
+	}
+}
+
+static void annotation__column_title(char *buf, size_t size,
+					     struct annotation *notes,
+					     const struct evsel *evsel)
+{
+	const char *base = symbol_conf.show_total_period ? "Period" :
+			symbol_conf.show_nr_samples ? "Samples" : "Percent";
+	int weight;
+	u8 weight_mask = annotation__weight_mask(notes, evsel);
+
+	scnprintf(buf, size, "%s", base);
+	for_each_weight(weight) {
+		if (weight_mask & BIT(weight))
+			scnprintf(buf + strlen(buf),
+				  size - strlen(buf),
+				  " %s",
+				  annotation__weight_mode_str(weight));
+	}
+}
+
 static int
 annotation_line__print(struct annotation_line *al, struct annotation_print_data *apd,
 		       struct annotation_options *opts, int printed,
@@ -833,6 +887,8 @@ annotation_line__print(struct annotation_line *al, struct annotation_print_data
 		for (i = 0; i < nr_percent; i++) {
 			struct annotation_data *data = &al->data[i];
 			double percent;
+			int weight;
+			u8 weight_mask = annotation__weight_mask(notes, apd->evsel);
 
 			percent = annotation_data__percent(data, percent_type);
 			color = get_percent_color(percent);
@@ -845,6 +901,13 @@ annotation_line__print(struct annotation_line *al, struct annotation_print_data
 					      data->he.nr_samples);
 			else
 				color_fprintf(stdout, color, " %7.2f", percent);
+			for_each_weight(weight) {
+				if (weight_mask & BIT(weight))
+					color_fprintf(stdout, color, " %7" PRIu64,
+						      data->he.weight_num[weight] ?
+						      data->he.weight_sum[weight] / data->he.weight_num[weight]
+						      : 0);
+			}
 		}
 
 		printf(" : ");
@@ -891,7 +954,7 @@ annotation_line__print(struct annotation_line *al, struct annotation_print_data
 	} else if (max_lines && printed >= max_lines)
 		return 1;
 	else {
-		int width = annotation__pcnt_width(notes);
+		int width = annotation__pcnt_width(notes, apd->evsel);
 
 		if (queue)
 			return -1;
@@ -915,6 +978,9 @@ static void calc_percent(struct annotation *notes,
 	struct sym_hist *sym_hist = annotation__histogram(notes, evsel);
 	unsigned int hits = 0;
 	u64 period = 0;
+	int i;
+	u64 weight_sum[WEIGHT_WEIGHT3 + 1] = { 0 };
+	u64 weight_num[WEIGHT_WEIGHT3 + 1] = { 0 };
 
 	while (offset < end) {
 		struct sym_hist_entry *entry;
@@ -923,6 +989,10 @@ static void calc_percent(struct annotation *notes,
 		if (entry) {
 			hits   += entry->nr_samples;
 			period += entry->period;
+			for_each_weight(i) {
+				weight_sum[i] += entry->weight_sum[i];
+				weight_num[i] += entry->weight_num[i];
+			}
 		}
 		++offset;
 	}
@@ -930,6 +1000,10 @@ static void calc_percent(struct annotation *notes,
 	if (sym_hist->nr_samples) {
 		data->he.period     = period;
 		data->he.nr_samples = hits;
+		for_each_weight(i) {
+			data->he.weight_sum[i] = weight_sum[i];
+			data->he.weight_num[i] = weight_num[i];
+		}
 		data->percent[PERCENT_HITS_LOCAL] = 100.0 * hits / sym_hist->nr_samples;
 	}
 
@@ -1252,8 +1326,9 @@ int hist_entry__annotate_printf(struct hist_entry *he, struct evsel *evsel)
 	int printed = 2, queue_len = 0;
 	int more = 0;
 	bool context = opts->context;
-	int width = annotation__pcnt_width(notes);
+	int width = annotation__pcnt_width(notes, evsel);
 	int graph_dotted_len;
+	char title[64];
 	char buf[512];
 
 	filename = strdup(dso__long_name(dso));
@@ -1275,10 +1350,10 @@ int hist_entry__annotate_printf(struct hist_entry *he, struct evsel *evsel)
 		return ENOTSUP;
 	}
 
-	graph_dotted_len = printf(" %-*.*s|	Source code & Disassembly of %s for %s (%" PRIu64 " samples, "
+	annotation__column_title(title, sizeof(title), notes, evsel);
+	graph_dotted_len = printf(" %-*.*s|\tSource code & Disassembly of %s for %s (%" PRIu64 " samples, "
 				  "percent: %s)\n",
-				  width, width, symbol_conf.show_total_period ? "Period" :
-				  symbol_conf.show_nr_samples ? "Samples" : "Percent",
+				  width, width, title,
 				  d_filename, evsel_name, h->nr_samples,
 				  percent_type_str(opts->percent_type));
 
@@ -2029,7 +2104,8 @@ static int disasm_line__snprint_type_info(struct disasm_line *dl,
 	return printed;
 }
 
-void annotation_line__write(struct annotation_line *al, struct annotation *notes,
+void annotation_line__write(struct annotation_line *al,
+			    struct annotation *notes,
 			    const struct annotation_write_ops *wops,
 			    struct annotation_print_data *apd)
 {
@@ -2037,7 +2113,8 @@ void annotation_line__write(struct annotation_line *al, struct annotation *notes
 	bool change_color = wops->change_color;
 	double percent_max = annotation_line__max_percent(al, annotate_opts.percent_type);
 	int width = wops->width;
-	int pcnt_width = annotation__pcnt_width(notes);
+	int pcnt_width = annotation__pcnt_width(notes, apd->evsel);
+	u8 weight_mask = annotation__weight_mask(notes, apd->evsel);
 	int cycles_width = annotation__cycles_width(notes);
 	bool show_title = false;
 	char bf[256];
@@ -2062,6 +2139,7 @@ void annotation_line__write(struct annotation_line *al, struct annotation *notes
 
 		for (i = 0; i < al->data_nr; i++) {
 			double percent;
+			int weight;
 
 			percent = annotation_data__percent(&al->data[i],
 							   annotate_opts.percent_type);
@@ -2075,6 +2153,14 @@ void annotation_line__write(struct annotation_line *al, struct annotation *notes
 			} else {
 				obj__printf(obj, "%7.2f ", percent);
 			}
+
+			for_each_weight(weight) {
+				if (weight_mask & BIT(weight))
+					obj__printf(obj, "%7" PRIu64 " ",
+						    al->data[i].he.weight_num[weight] ?
+						    al->data[i].he.weight_sum[weight] /	al->data[i].he.weight_num[weight] :
+						    0);
+			}
 		}
 	} else {
 		obj__set_percent_color(obj, 0, current_entry);
@@ -2082,9 +2168,9 @@ void annotation_line__write(struct annotation_line *al, struct annotation *notes
 		if (!show_title)
 			obj__printf(obj, "%-*s", pcnt_width, " ");
 		else {
-			obj__printf(obj, "%-*s", pcnt_width,
-					   symbol_conf.show_total_period ? "Period" :
-					   symbol_conf.show_nr_samples ? "Samples" : "Percent");
+			char buf[64];
+			annotation__column_title(buf, sizeof(buf), notes, apd->evsel);
+			obj__printf(obj, "%-*s", pcnt_width, buf);
 		}
 	}
 	width -= pcnt_width;
diff --git a/tools/perf/util/annotate.h b/tools/perf/util/annotate.h
index fa08d09b80f7..63bce7a920f3 100644
--- a/tools/perf/util/annotate.h
+++ b/tools/perf/util/annotate.h
@@ -6,6 +6,7 @@
 #include <stdint.h>
 #include <stdio.h>
 #include <linux/types.h>
+#include <linux/bitops.h>
 #include <linux/list.h>
 #include <linux/rbtree.h>
 #include <asm/bug.h>
@@ -86,6 +87,8 @@ struct annotation;
 struct sym_hist_entry {
 	u64		nr_samples;
 	u64		period;
+	u64		weight_sum[WEIGHT_WEIGHT3 + 1];
+	u64		weight_num[WEIGHT_WEIGHT3 + 1];
 };
 
 enum {
@@ -231,8 +234,20 @@ void symbol__calc_percent(struct symbol *sym, struct evsel *evsel);
 struct sym_hist {
 	u64		      nr_samples;
 	u64		      period;
+	u8		      weight_mask;
 };
 
+/* Can be set asynchronously by top. */
+static inline u8 sym_hist__weight_mask(const struct sym_hist *hist)
+{
+	return __atomic_load_n(&hist->weight_mask, __ATOMIC_RELAXED);
+}
+
+static inline void sym_hist__set_weight_mask(struct sym_hist *hist, u8 mask)
+{
+	__atomic_fetch_or(&hist->weight_mask, mask, __ATOMIC_RELAXED);
+}
+
 /**
  * struct cyc_hist - (CPU) cycle histogram for a basic block
  *
@@ -376,9 +391,22 @@ static inline int annotation__cycles_width(struct annotation *notes)
 	return notes->branch ? ANNOTATION__IPC_WIDTH + ANNOTATION__CYCLES_WIDTH : 0;
 }
 
-static inline int annotation__pcnt_width(struct annotation *notes)
+static inline u8 annotation__weight_mask(struct annotation *notes,
+					 const struct evsel *evsel)
+{
+	if (symbol_conf.weight_mask_user)
+		return symbol_conf.weight_mask;
+	if (symbol_conf.hide_weight)
+		return 0;
+	return sym_hist__weight_mask(&notes->src->histograms[evsel->core.idx]);
+}
+
+static inline int annotation__pcnt_width(struct annotation *notes,
+					 const struct evsel *evsel)
 {
-	return (symbol_conf.show_total_period ? 12 : 8) * notes->src->nr_events;
+	int extra = hweight8(annotation__weight_mask(notes, evsel)) * 8;
+	return ((symbol_conf.show_total_period ? 12 : 8) + extra) *
+	       notes->src->nr_events;
 }
 
 static inline bool annotation_line__filter(struct annotation_line *al)
diff --git a/tools/perf/util/symbol_conf.h b/tools/perf/util/symbol_conf.h
index 0dee5aa6a534..6931fa014255 100644
--- a/tools/perf/util/symbol_conf.h
+++ b/tools/perf/util/symbol_conf.h
@@ -27,6 +27,16 @@ enum a2l_style {
 };
 #define MAX_A2L_STYLE (A2L_STYLE_CMD + 1)
 
+enum symbol__weight_mode {
+	WEIGHT_NONE = 0,
+	WEIGHT_WEIGHT,
+	WEIGHT_INSNLAT,
+	WEIGHT_WEIGHT3,
+};
+
+#define for_each_weight(_weight) \
+	for ((_weight) = WEIGHT_WEIGHT; (_weight) <= WEIGHT_WEIGHT3; (_weight)++)
+
 struct symbol_conf {
 	bool		nanosecs;
 	unsigned short	priv_size;
@@ -69,7 +79,10 @@ struct symbol_conf {
 			annotate_data_sample,
 			skip_empty,
 			enable_latency,
-			prefer_latency;
+			prefer_latency,
+			hide_weight,
+			weight_mask_user;
+	u8		weight_mask;
 	const char	*vmlinux_name,
 			*kallsyms_name,
 			*source_prefix,
-- 
2.55.0

[PATCH v3 4/4] perf tools: Add test for weight annotation

[Andi Kleen]

Add a test case to check if --stdio weight annotation works

Assisted-by: omp:GPT-5.6-Luna
Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 tools/perf/tests/shell/annotate_weight.sh | 63 +++++++++++++++++++++++
 1 file changed, 63 insertions(+)
 create mode 100755 tools/perf/tests/shell/annotate_weight.sh

diff --git a/tools/perf/tests/shell/annotate_weight.sh b/tools/perf/tests/shell/annotate_weight.sh
new file mode 100755
index 000000000000..6b8c105c048b
--- /dev/null
+++ b/tools/perf/tests/shell/annotate_weight.sh
@@ -0,0 +1,63 @@
+#!/bin/bash
+# perf annotate weight regression test
+# SPDX-License-Identifier: GPL-2.0
+
+set -e
+
+shelldir=$(dirname "$0")
+# shellcheck source=tools/perf/tests/shell/lib/perf_has_symbol.sh
+. "${shelldir}"/lib/perf_has_symbol.sh
+
+testsym="test_loop"
+skip_test_missing_symbol "${testsym}"
+
+perfdata=$(mktemp /tmp/__perf_test.annotate_weight.XXXXX)
+record_log=$(mktemp /tmp/__perf_test.annotate_weight.XXXXX.log)
+report_out=$(mktemp /tmp/__perf_test.annotate_weight.XXXXX.report)
+annotate_out=$(mktemp /tmp/__perf_test.annotate_weight.XXXXX.annotate)
+
+cleanup() {
+	rm -f "${perfdata}" "${record_log}" "${report_out}" "${annotate_out}"
+	trap - EXIT TERM INT
+}
+
+trap 'cleanup; exit 1' TERM INT
+trap cleanup EXIT
+
+# mem-loads:pu requests a precise user PEBS event whose sample weight should
+# be populated by -W. Unsupported PEBS/weight PMUs are skipped below.
+if ! perf record -W -e mem-loads:pu -o "${perfdata}" -- perf test -w thloop \
+	> /dev/null 2> "${record_log}"; then
+	echo "[SKIP] precise PEBS weight sampling is unavailable"
+	exit 2
+fi
+
+# Confirm the PMU actually produced nonzero weights. A successful record alone
+# is insufficient: some PMUs accept the event but provide no weight payload.
+if ! perf report --stdio -i "${perfdata}" --fields=weight1,weight2,weight3,symbol --percent-limit 0 \
+	> "${report_out}" 2> "${record_log}"; then
+	echo "[SKIP] weighted samples cannot be decoded"
+	exit 2
+fi
+
+if ! awk '$1 ~ /^[0-9]/ && ($1 + 0) > 0 { found = 1 } END { exit !found }' \
+	"${report_out}"; then
+	echo "[SKIP] PEBS weight sampling produced no nonzero weights"
+	exit 2
+fi
+
+perf annotate --stdio -i "${perfdata}" --symbol "${testsym}" \
+	> "${annotate_out}" 2> "${record_log}"
+
+grep -q 'Percent Weight' "${annotate_out}"
+
+# The second numeric column is the rendered average weight. This assertion
+# fails on the original regression because calc_percent() left it at zero.
+if ! awk '$1 ~ /^[0-9]/ && $2 ~ /^[0-9]/ && ($2 + 0) > 0 { found = 1 } END { exit !found }' \
+	"${annotate_out}"; then
+	echo "Annotation output contained no nonzero weight"
+	cat "${annotate_out}"
+	exit 1
+fi
+
+echo "PEBS annotation weights: PASS"
-- 
2.55.0

Re: [PATCH v3 1/4] perf tools record: Modernize -W man page

[sashiko-bot]

> Modernize the -W / --weight description in the manpage to cover more
> cases that are supported now.
> 
> Signed-off-by: Andi Kleen <ak@linux.intel.com>

Sashiko has reviewed this patch and found no issues. It looks great!

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260813214143.253418-1-ak@linux.intel.com?part=1

Re: [PATCH v3 2/4] perf tools top: Add --weight option

[sashiko-bot]

> Add a -W/--weight option to perf top to collect weights too. Useful with
> follow on patches.
> 
> Assisted-by: omp:GPT-5.6-Luna
> Signed-off-by: Andi Kleen <ak@linux.intel.com>

Sashiko has reviewed this patch and found no issues. It looks great!

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260813214143.253418-1-ak@linux.intel.com?part=2

Re: [PATCH v3 3/4] perf tools: Add support for displaying weights in annotate

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] The 'w' hotkey caches the active event's weight mask into a global configuration variable, corrupting the display of weight columns for any subsequently viewed events.
--

--- Patch [3]: [PATCH v3 3/4] perf tools: Add support for displaying weights in annotate ---
commit 4886f581054e1fcacbf3f1649fad566092ced17b
Author: Andi Kleen <ak@linux.intel.com>

perf tools: Add support for displaying weights in annotate

This commit adds support for showing all three possible per-IP weights in
the annotation view, enabled by default if any are non-zero. It also
introduces a hotkey ('w') and a command line option to toggle these weight
columns.

> diff --git a/tools/perf/ui/browsers/annotate.c b/tools/perf/ui/browsers/annotate.c
> index e47a467750890..d6126a7d51e4a 100644
> --- a/tools/perf/ui/browsers/annotate.c
> +++ b/tools/perf/ui/browsers/annotate.c
[ ... ]
> @@ -1091,6 +1093,15 @@ static int annotate_browser__run(struct annotate_browser *browser,
>  				symbol_conf.show_total_period = true;
>  			annotation__update_column_widths(notes);
>  			continue;
> +		case 'w':
> +			symbol_conf.hide_weight = !symbol_conf.hide_weight;
> +			symbol_conf.weight_mask_user = true;
> +			symbol_conf.weight_mask = symbol_conf.hide_weight ?
> +				0 : sym_hist__weight_mask(annotation__histogram(notes, evsel));

[Severity: Medium]
Will caching the active event's weight mask into the global symbol_conf
corrupt the display for subsequently viewed events?

If the user toggles weights with 'w', symbol_conf.weight_mask_user is set
to true and the current event's mask is saved to the global
symbol_conf.weight_mask. If the user then opens an annotation for a
different event, it seems this stale global state will persist and
override the dynamic column visibility for the new event.

> +			browser->b.width = notes->src->widths.max_line_len +
> +					   annotation__pcnt_width(notes, evsel);
> +			ui_browser__refresh_dimensions(&browser->b);
> +			continue;
[ ... ]
> diff --git a/tools/perf/util/annotate.h b/tools/perf/util/annotate.h
> index fa08d09b80f76..63bce7a920f32 100644
> --- a/tools/perf/util/annotate.h
> +++ b/tools/perf/util/annotate.h
[ ... ]
> @@ -376,9 +391,22 @@ static inline int annotation__cycles_width(struct annotation *notes)
>  	return notes->branch ? ANNOTATION__IPC_WIDTH + ANNOTATION__CYCLES_WIDTH : 0;
>  }
>  
> -static inline int annotation__pcnt_width(struct annotation *notes)
> +static inline u8 annotation__weight_mask(struct annotation *notes,
> +					 const struct evsel *evsel)
> +{
> +	if (symbol_conf.weight_mask_user)
> +		return symbol_conf.weight_mask;

[Severity: Medium]
Because symbol_conf is a global variable, does applying it uniformly here
break the dynamic weight mask calculation for other events once the user
has triggered the hotkey?

> +	if (symbol_conf.hide_weight)
> +		return 0;
> +	return sym_hist__weight_mask(&notes->src->histograms[evsel->core.idx]);
> +}

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260813214143.253418-1-ak@linux.intel.com?part=3

Re: [PATCH v3 4/4] perf tools: Add test for weight annotation

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Brittle and silent test assertion due to unisolated `PERF_CONFIG` environment.
--

--- Patch [4]: [PATCH v3 4/4] perf tools: Add test for weight annotation ---
commit ae0853531b9364f64586279c41d3ca0652fe3e71
Author: Andi Kleen <ak@linux.intel.com>

perf tools: Add test for weight annotation

This commit adds a new shell test case to verify that --stdio weight
annotation works correctly. It uses perf record to capture a precise PEBS
event and then tests the perf report and perf annotate tools.

> diff --git a/tools/perf/tests/shell/annotate_weight.sh b/tools/perf/tests/shell/annotate_weight.sh
> new file mode 100755
> --- /dev/null
> +++ b/tools/perf/tests/shell/annotate_weight.sh

[ ... ]

> +perf annotate --stdio -i "${perfdata}" --symbol "${testsym}" \
> +	> "${annotate_out}" 2> "${record_log}"
> +
> +grep -q 'Percent Weight' "${annotate_out}"

[Severity: Medium]
Will this test fail silently for users with a customized ~/.perfconfig?

Because set -e is active at the top of the script, if a user has settings
like annotate.show_nr_samples=true or annotate.show_total_period=true,
perf annotate might change the column header from "Percent Weight" to
something like "Samples Weight" or "Period Weight".

This would cause this standalone grep command to fail and immediately
terminate the test without providing any diagnostics. Does this test need to
run in an isolated environment or parse the output more flexibly?

> +
> +# The second numeric column is the rendered average weight. This assertion
> +# fails on the original regression because calc_percent() left it at zero.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260813214143.253418-1-ak@linux.intel.com?part=4