1 files changed, 212 insertions, 59 deletions

diff --git a/tools/perf/builtin-c2c.c b/tools/perf/builtin-c2c.c
index 15e1fce71c72..d390ae4e3ec8 100644
--- a/tools/perf/builtin-c2c.c
+++ b/tools/perf/builtin-c2c.c
@@ -45,6 +45,8 @@
 #include "pmus.h"
 #include "string2.h"
 #include "util/util.h"
+#include "util/symbol.h"
+#include "util/annotate.h"
 
 struct c2c_hists {
 	struct hists		hists;
@@ -62,6 +64,7 @@ struct compute_stats {
 
 struct c2c_hist_entry {
 	struct c2c_hists	*hists;
+	struct evsel		*evsel;
 	struct c2c_stats	 stats;
 	unsigned long		*cpuset;
 	unsigned long		*nodeset;
@@ -195,12 +198,14 @@ static struct hist_entry_ops c2c_entry_ops = {
 
 static int c2c_hists__init(struct c2c_hists *hists,
 			   const char *sort,
-			   int nr_header_lines);
+			   int nr_header_lines,
+			   struct perf_env *env);
 
 static struct c2c_hists*
 he__get_c2c_hists(struct hist_entry *he,
 		  const char *sort,
-		  int nr_header_lines)
+		  int nr_header_lines,
+		  struct perf_env *env)
 {
 	struct c2c_hist_entry *c2c_he;
 	struct c2c_hists *hists;
@@ -214,7 +219,7 @@ he__get_c2c_hists(struct hist_entry *he,
 	if (!hists)
 		return NULL;
 
-	ret = c2c_hists__init(hists, sort, nr_header_lines);
+	ret = c2c_hists__init(hists, sort, nr_header_lines, env);
 	if (ret) {
 		free(hists);
 		return NULL;
@@ -223,6 +228,12 @@ he__get_c2c_hists(struct hist_entry *he,
 	return hists;
 }
 
+static void c2c_he__set_evsel(struct c2c_hist_entry *c2c_he,
+				struct evsel *evsel)
+{
+	c2c_he->evsel = evsel;
+}
+
 static void c2c_he__set_cpu(struct c2c_hist_entry *c2c_he,
 			    struct perf_sample *sample)
 {
@@ -273,6 +284,33 @@ static void compute_stats(struct c2c_hist_entry *c2c_he,
 		update_stats(&cstats->load, weight);
 }
 
+/*
+ * Return true if annotation is possible. When list is NULL,
+ * it means that we are called at the c2c_browser level,
+ * in that case we allow annotation to be initialized. When list
+ * is non-NULL, it means that we are called at the cacheline_browser
+ * level, in that case we allow annotation only if use_browser
+ * is set and symbol information is available.
+ */
+static bool perf_c2c__has_annotation(struct perf_hpp_list *list)
+{
+	if (use_browser != 1)
+		return false;
+	return !list || list->sym;
+}
+
+static void perf_c2c__evsel_hists_inc_stats(struct evsel *evsel,
+					    struct hist_entry *he,
+					    struct perf_sample *sample)
+{
+	struct hists *evsel_hists = evsel__hists(evsel);
+
+	hists__inc_nr_samples(evsel_hists, he->filtered);
+	evsel_hists->stats.total_period += sample->period;
+	if (!he->filtered)
+		evsel_hists->stats.total_non_filtered_period += sample->period;
+}
+
 static int process_sample_event(const struct perf_tool *tool __maybe_unused,
 				union perf_event *event,
 				struct perf_sample *sample,
@@ -284,7 +322,7 @@ static int process_sample_event(const struct perf_tool *tool __maybe_unused,
 	struct c2c_stats stats = { .nr_entries = 0, };
 	struct hist_entry *he;
 	struct addr_location al;
-	struct mem_info *mi, *mi_dup;
+	struct mem_info *mi = NULL;
 	struct callchain_cursor *cursor;
 	int ret;
 
@@ -311,20 +349,15 @@ static int process_sample_event(const struct perf_tool *tool __maybe_unused,
 		goto out;
 	}
 
-	/*
-	 * The mi object is released in hists__add_entry_ops,
-	 * if it gets sorted out into existing data, so we need
-	 * to take the copy now.
-	 */
-	mi_dup = mem_info__get(mi);
-
 	c2c_decode_stats(&stats, mi);
 
 	he = hists__add_entry_ops(&c2c_hists->hists, &c2c_entry_ops,
 				  &al, NULL, NULL, mi, NULL,
 				  sample, true);
-	if (he == NULL)
-		goto free_mi;
+	if (he == NULL) {
+		ret = -ENOMEM;
+		goto out;
+	}
 
 	c2c_he = container_of(he, struct c2c_hist_entry, he);
 	c2c_add_stats(&c2c_he->stats, &stats);
@@ -332,8 +365,15 @@ static int process_sample_event(const struct perf_tool *tool __maybe_unused,
 
 	c2c_he__set_cpu(c2c_he, sample);
 	c2c_he__set_node(c2c_he, sample);
+	c2c_he__set_evsel(c2c_he, evsel);
 
 	hists__inc_nr_samples(&c2c_hists->hists, he->filtered);
+
+	if (perf_c2c__has_annotation(NULL)) {
+		perf_c2c__evsel_hists_inc_stats(evsel, he, sample);
+		addr_map_symbol__inc_samples(mem_info__iaddr(mi), sample, evsel);
+	}
+
 	ret = hist_entry__append_callchain(he, sample);
 
 	if (!ret) {
@@ -348,17 +388,19 @@ static int process_sample_event(const struct perf_tool *tool __maybe_unused,
 		int cpu = sample->cpu == (unsigned int) -1 ? 0 : sample->cpu;
 		int node = c2c.cpu2node[cpu];
 
-		mi = mi_dup;
-
-		c2c_hists = he__get_c2c_hists(he, c2c.cl_sort, 2);
-		if (!c2c_hists)
-			goto free_mi;
+		c2c_hists = he__get_c2c_hists(he, c2c.cl_sort, 2, machine->env);
+		if (!c2c_hists) {
+			ret = -ENOMEM;
+			goto out;
+		}
 
 		he = hists__add_entry_ops(&c2c_hists->hists, &c2c_entry_ops,
 					  &al, NULL, NULL, mi, NULL,
 					  sample, true);
-		if (he == NULL)
-			goto free_mi;
+		if (he == NULL) {
+			ret = -ENOMEM;
+			goto out;
+		}
 
 		c2c_he = container_of(he, struct c2c_hist_entry, he);
 		c2c_add_stats(&c2c_he->stats, &stats);
@@ -369,20 +411,16 @@ static int process_sample_event(const struct perf_tool *tool __maybe_unused,
 
 		c2c_he__set_cpu(c2c_he, sample);
 		c2c_he__set_node(c2c_he, sample);
+		c2c_he__set_evsel(c2c_he, evsel);
 
 		hists__inc_nr_samples(&c2c_hists->hists, he->filtered);
 		ret = hist_entry__append_callchain(he, sample);
 	}
 
 out:
+	mem_info__put(mi);
 	addr_location__exit(&al);
 	return ret;
-
-free_mi:
-	mem_info__put(mi_dup);
-	mem_info__put(mi);
-	ret = -ENOMEM;
-	goto out;
 }
 
 static const char * const c2c_usage[] = {
@@ -1966,33 +2004,38 @@ static struct c2c_fmt *get_format(const char *name)
 	return c2c_fmt;
 }
 
-static int c2c_hists__init_output(struct perf_hpp_list *hpp_list, char *name)
+static int c2c_hists__init_output(struct perf_hpp_list *hpp_list, char *name,
+				  struct perf_env *env __maybe_unused)
 {
 	struct c2c_fmt *c2c_fmt = get_format(name);
+	int level = 0;
 
 	if (!c2c_fmt) {
 		reset_dimensions();
-		return output_field_add(hpp_list, name);
+		return output_field_add(hpp_list, name, &level);
 	}
 
 	perf_hpp_list__column_register(hpp_list, &c2c_fmt->fmt);
 	return 0;
 }
 
-static int c2c_hists__init_sort(struct perf_hpp_list *hpp_list, char *name)
+static int c2c_hists__init_sort(struct perf_hpp_list *hpp_list, char *name, struct perf_env *env)
 {
 	struct c2c_fmt *c2c_fmt = get_format(name);
 	struct c2c_dimension *dim;
 
 	if (!c2c_fmt) {
 		reset_dimensions();
-		return sort_dimension__add(hpp_list, name, NULL, 0);
+		return sort_dimension__add(hpp_list, name, /*evlist=*/NULL, env, /*level=*/0);
 	}
 
 	dim = c2c_fmt->dim;
 	if (dim == &dim_dso)
 		hpp_list->dso = 1;
 
+	if (dim == &dim_symbol || dim == &dim_iaddr)
+		hpp_list->sym = 1;
+
 	perf_hpp_list__register_sort_field(hpp_list, &c2c_fmt->fmt);
 	return 0;
 }
@@ -2007,7 +2050,7 @@ static int c2c_hists__init_sort(struct perf_hpp_list *hpp_list, char *name)
 										\
 		for (tok = strtok_r((char *)_list, ", ", &tmp);			\
 				tok; tok = strtok_r(NULL, ", ", &tmp)) {	\
-			ret = _fn(hpp_list, tok);				\
+			ret = _fn(hpp_list, tok, env);				\
 			if (ret == -EINVAL) {					\
 				pr_err("Invalid --fields key: `%s'", tok);	\
 				break;						\
@@ -2020,7 +2063,8 @@ static int c2c_hists__init_sort(struct perf_hpp_list *hpp_list, char *name)
 
 static int hpp_list__parse(struct perf_hpp_list *hpp_list,
 			   const char *output_,
-			   const char *sort_)
+			   const char *sort_,
+			   struct perf_env *env)
 {
 	char *output = output_ ? strdup(output_) : NULL;
 	char *sort   = sort_   ? strdup(sort_) : NULL;
@@ -2051,7 +2095,8 @@ static int hpp_list__parse(struct perf_hpp_list *hpp_list,
 
 static int c2c_hists__init(struct c2c_hists *hists,
 			   const char *sort,
-			   int nr_header_lines)
+			   int nr_header_lines,
+			   struct perf_env *env)
 {
 	__hists__init(&hists->hists, &hists->list);
 
@@ -2065,15 +2110,16 @@ static int c2c_hists__init(struct c2c_hists *hists,
 	/* Overload number of header lines.*/
 	hists->list.nr_header_lines = nr_header_lines;
 
-	return hpp_list__parse(&hists->list, NULL, sort);
+	return hpp_list__parse(&hists->list, /*output=*/NULL, sort, env);
 }
 
 static int c2c_hists__reinit(struct c2c_hists *c2c_hists,
 			     const char *output,
-			     const char *sort)
+			     const char *sort,
+			     struct perf_env *env)
 {
 	perf_hpp__reset_output_field(&c2c_hists->list);
-	return hpp_list__parse(&c2c_hists->list, output, sort);
+	return hpp_list__parse(&c2c_hists->list, output, sort, env);
 }
 
 #define DISPLAY_LINE_LIMIT  0.001
@@ -2206,8 +2252,9 @@ static int filter_cb(struct hist_entry *he, void *arg __maybe_unused)
 	return 0;
 }
 
-static int resort_cl_cb(struct hist_entry *he, void *arg __maybe_unused)
+static int resort_cl_cb(struct hist_entry *he, void *arg)
 {
+	struct perf_env *env = arg;
 	struct c2c_hist_entry *c2c_he;
 	struct c2c_hists *c2c_hists;
 	bool display = he__display(he, &c2c.shared_clines_stats);
@@ -2221,7 +2268,7 @@ static int resort_cl_cb(struct hist_entry *he, void *arg __maybe_unused)
 		c2c_he->cacheline_idx = idx++;
 		calc_width(c2c_he);
 
-		c2c_hists__reinit(c2c_hists, c2c.cl_output, c2c.cl_resort);
+		c2c_hists__reinit(c2c_hists, c2c.cl_output, c2c.cl_resort, env);
 
 		hists__collapse_resort(&c2c_hists->hists, NULL);
 		hists__output_resort_cb(&c2c_hists->hists, NULL, filter_cb);
@@ -2266,14 +2313,15 @@ static int setup_nodes(struct perf_session *session)
 	int node, idx;
 	struct perf_cpu cpu;
 	int *cpu2node;
+	struct perf_env *env = perf_session__env(session);
 
 	if (c2c.node_info > 2)
 		c2c.node_info = 2;
 
-	c2c.nodes_cnt = session->header.env.nr_numa_nodes;
-	c2c.cpus_cnt  = session->header.env.nr_cpus_avail;
+	c2c.nodes_cnt = env->nr_numa_nodes;
+	c2c.cpus_cnt  = env->nr_cpus_avail;
 
-	n = session->header.env.numa_nodes;
+	n = env->numa_nodes;
 	if (!n)
 		return -EINVAL;
 
@@ -2332,7 +2380,7 @@ static int resort_shared_cl_cb(struct hist_entry *he, void *arg __maybe_unused)
 	return 0;
 }
 
-static int hists__iterate_cb(struct hists *hists, hists__resort_cb_t cb)
+static int hists__iterate_cb(struct hists *hists, hists__resort_cb_t cb, void *arg)
 {
 	struct rb_node *next = rb_first_cached(&hists->entries);
 	int ret = 0;
@@ -2341,7 +2389,7 @@ static int hists__iterate_cb(struct hists *hists, hists__resort_cb_t cb)
 		struct hist_entry *he;
 
 		he = rb_entry(next, struct hist_entry, rb_node);
-		ret = cb(he, NULL);
+		ret = cb(he, arg);
 		if (ret)
 			break;
 		next = rb_next(&he->rb_node);
@@ -2447,7 +2495,7 @@ static void print_cacheline(struct c2c_hists *c2c_hists,
 	hists__fprintf(&c2c_hists->hists, false, 0, 0, 0, out, false);
 }
 
-static void print_pareto(FILE *out)
+static void print_pareto(FILE *out, struct perf_env *env)
 {
 	struct perf_hpp_list hpp_list;
 	struct rb_node *nd;
@@ -2472,7 +2520,7 @@ static void print_pareto(FILE *out)
 			    "dcacheline";
 
 	perf_hpp_list__init(&hpp_list);
-	ret = hpp_list__parse(&hpp_list, cl_output, NULL);
+	ret = hpp_list__parse(&hpp_list, cl_output, /*evlist=*/NULL, env);
 
 	if (WARN_ONCE(ret, "failed to setup sort entries\n"))
 		return;
@@ -2537,10 +2585,48 @@ static void perf_c2c__hists_fprintf(FILE *out, struct perf_session *session)
 	fprintf(out, "=================================================\n");
 	fprintf(out, "#\n");
 
-	print_pareto(out);
+	print_pareto(out, perf_session__env(session));
 }
 
 #ifdef HAVE_SLANG_SUPPORT
+
+static int perf_c2c__toggle_annotation(struct hist_browser *browser)
+{
+	struct hist_entry *he = browser->he_selection;
+	struct symbol *sym = NULL;
+	struct annotated_source *src = NULL;
+	struct c2c_hist_entry *c2c_he = NULL;
+	u64 al_addr = NO_ADDR;
+
+	if (!perf_c2c__has_annotation(he->hists->hpp_list)) {
+		ui_browser__help_window(&browser->b, "No annotation support");
+		return 0;
+	}
+
+	if (he == NULL) {
+		ui_browser__help_window(&browser->b, "No entry selected for annotation");
+		return 0;
+	}
+
+	sym = he->ms.sym;
+	if (sym == NULL) {
+		ui_browser__help_window(&browser->b, "Can not annotate, no symbol found");
+		return 0;
+	}
+
+	src = symbol__hists(sym, 0);
+	if (src == NULL) {
+		ui_browser__help_window(&browser->b, "Failed to initialize annotation source");
+		return 0;
+	}
+
+	if (he->mem_info)
+		al_addr = mem_info__iaddr(he->mem_info)->al_addr;
+
+	c2c_he = container_of(he, struct c2c_hist_entry, he);
+	return hist_entry__tui_annotate(he, c2c_he->evsel, NULL, al_addr);
+}
+
 static void c2c_browser__update_nr_entries(struct hist_browser *hb)
 {
 	u64 nr_entries = 0;
@@ -2608,6 +2694,7 @@ static int perf_c2c__browse_cacheline(struct hist_entry *he)
 	" ENTER         Toggle callchains (if present) \n"
 	" n             Toggle Node details info \n"
 	" s             Toggle full length of symbol and source line columns \n"
+	" a             Toggle annotation view \n"
 	" q             Return back to cacheline list \n";
 
 	if (!he)
@@ -2642,6 +2729,9 @@ static int perf_c2c__browse_cacheline(struct hist_entry *he)
 			c2c.node_info = (c2c.node_info + 1) % 3;
 			setup_nodes_header();
 			break;
+		case 'a':
+			perf_c2c__toggle_annotation(browser);
+			break;
 		case 'q':
 			goto out;
 		case '?':
@@ -2997,6 +3087,7 @@ static int perf_c2c__report(int argc, const char **argv)
 	const char *display = NULL;
 	const char *coalesce = NULL;
 	bool no_source = false;
+	const char *disassembler_style = NULL, *objdump_path = NULL;
 	const struct option options[] = {
 	OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
 		   "file", "vmlinux pathname"),
@@ -3024,11 +3115,22 @@ static int perf_c2c__report(int argc, const char **argv)
 	OPT_BOOLEAN(0, "stitch-lbr", &c2c.stitch_lbr,
 		    "Enable LBR callgraph stitching approach"),
 	OPT_BOOLEAN(0, "double-cl", &chk_double_cl, "Detect adjacent cacheline false sharing"),
+	OPT_STRING('M', "disassembler-style", &disassembler_style, "disassembler style",
+		   "Specify disassembler style (e.g. -M intel for intel syntax)"),
+	OPT_STRING(0, "objdump", &objdump_path, "path",
+		   "objdump binary to use for disassembly and annotations"),
 	OPT_PARENT(c2c_options),
 	OPT_END()
 	};
 	int err = 0;
 	const char *output_str, *sort_str = NULL;
+	struct perf_env *env;
+
+	annotation_options__init();
+
+	err = hists__init();
+	if (err < 0)
+		goto out;
 
 	argc = parse_options(argc, argv, options, report_c2c_usage,
 			     PARSE_OPT_STOP_AT_NON_OPTION);
@@ -3042,6 +3144,27 @@ static int perf_c2c__report(int argc, const char **argv)
 	if (c2c.stats_only)
 		c2c.use_stdio = true;
 
+	/**
+	 * Annotation related options disassembler_style, objdump_path are set
+	 * in the c2c_options, so we can use them here.
+	 */
+	if (disassembler_style) {
+		annotate_opts.disassembler_style = strdup(disassembler_style);
+		if (!annotate_opts.disassembler_style) {
+			err = -ENOMEM;
+			pr_err("Failed to allocate memory for annotation options\n");
+			goto out;
+		}
+	}
+	if (objdump_path) {
+		annotate_opts.objdump_path = strdup(objdump_path);
+		if (!annotate_opts.objdump_path) {
+			err = -ENOMEM;
+			pr_err("Failed to allocate memory for annotation options\n");
+			goto out;
+		}
+	}
+
 	err = symbol__validate_sym_arguments();
 	if (err)
 		goto out;
@@ -3071,14 +3194,14 @@ static int perf_c2c__report(int argc, const char **argv)
 		pr_debug("Error creating perf session\n");
 		goto out;
 	}
-
+	env = perf_session__env(session);
 	/*
 	 * Use the 'tot' as default display type if user doesn't specify it;
 	 * since Arm64 platform doesn't support HITMs flag, use 'peer' as the
 	 * default display type.
 	 */
 	if (!display) {
-		if (!strcmp(perf_env__arch(&session->header.env), "arm64"))
+		if (!strcmp(perf_env__arch(env), "arm64"))
 			display = "peer";
 		else
 			display = "tot";
@@ -3094,7 +3217,7 @@ static int perf_c2c__report(int argc, const char **argv)
 		goto out_session;
 	}
 
-	err = c2c_hists__init(&c2c.hists, "dcacheline", 2);
+	err = c2c_hists__init(&c2c.hists, "dcacheline", 2, perf_session__env(session));
 	if (err) {
 		pr_debug("Failed to initialize hists\n");
 		goto out_session;
@@ -3108,7 +3231,7 @@ static int perf_c2c__report(int argc, const char **argv)
 		goto out_session;
 	}
 
-	err = mem2node__init(&c2c.mem2node, &session->header.env);
+	err = mem2node__init(&c2c.mem2node, env);
 	if (err)
 		goto out_session;
 
@@ -3116,7 +3239,39 @@ static int perf_c2c__report(int argc, const char **argv)
 	if (err)
 		goto out_mem2node;
 
-	if (symbol__init(&session->header.env) < 0)
+	if (c2c.use_stdio)
+		use_browser = 0;
+	else
+		use_browser = 1;
+
+	/*
+	 * Only in the TUI browser we are doing integrated annotation,
+	 * so don't allocate extra space that won't be used in the stdio
+	 * implementation.
+	 */
+	if (perf_c2c__has_annotation(NULL)) {
+		int ret = symbol__annotation_init();
+
+		if (ret < 0)
+			goto out_mem2node;
+		/*
+		 * For searching by name on the "Browse map details".
+		 * providing it only in verbose mode not to bloat too
+		 * much struct symbol.
+		 */
+		if (verbose > 0) {
+			/*
+			 * XXX: Need to provide a less kludgy way to ask for
+			 * more space per symbol, the u32 is for the index on
+			 * the ui browser.
+			 * See symbol__browser_index.
+			 */
+			symbol_conf.priv_size += sizeof(u32);
+		}
+		annotation_config__init();
+	}
+
+	if (symbol__init(env) < 0)
 		goto out_mem2node;
 
 	/* No pipe support at the moment. */
@@ -3125,11 +3280,6 @@ static int perf_c2c__report(int argc, const char **argv)
 		goto out_mem2node;
 	}
 
-	if (c2c.use_stdio)
-		use_browser = 0;
-	else
-		use_browser = 1;
-
 	setup_browser(false);
 
 	err = perf_session__process_events(session);
@@ -3178,13 +3328,13 @@ static int perf_c2c__report(int argc, const char **argv)
 	else if (c2c.display == DISPLAY_SNP_PEER)
 		sort_str = "tot_peer";
 
-	c2c_hists__reinit(&c2c.hists, output_str, sort_str);
+	c2c_hists__reinit(&c2c.hists, output_str, sort_str, perf_session__env(session));
 
 	ui_progress__init(&prog, c2c.hists.hists.nr_entries, "Sorting...");
 
 	hists__collapse_resort(&c2c.hists.hists, NULL);
 	hists__output_resort_cb(&c2c.hists.hists, &prog, resort_shared_cl_cb);
-	hists__iterate_cb(&c2c.hists.hists, resort_cl_cb);
+	hists__iterate_cb(&c2c.hists.hists, resort_cl_cb, perf_session__env(session));
 
 	ui_progress__finish();
 
@@ -3200,6 +3350,7 @@ out_mem2node:
 out_session:
 	perf_session__delete(session);
 out:
+	annotation_options__exit();
 	return err;
 }
 
@@ -3239,6 +3390,7 @@ static int perf_c2c__record(int argc, const char **argv)
 {
 	int rec_argc, i = 0, j;
 	const char **rec_argv;
+	char *event_name_storage = NULL;
 	int ret;
 	bool all_user = false, all_kernel = false;
 	bool event_set = false;
@@ -3300,7 +3452,7 @@ static int perf_c2c__record(int argc, const char **argv)
 	rec_argv[i++] = "--phys-data";
 	rec_argv[i++] = "--sample-cpu";
 
-	ret = perf_mem_events__record_args(rec_argv, &i);
+	ret = perf_mem_events__record_args(rec_argv, &i, &event_name_storage);
 	if (ret)
 		goto out;
 
@@ -3327,6 +3479,7 @@ static int perf_c2c__record(int argc, const char **argv)
 
 	ret = cmd_record(i, rec_argv);
 out:
+	free(event_name_storage);
 	free(rec_argv);
 	return ret;
 }