1 files changed, 303 insertions, 80 deletions

diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 135b7837e6bf..4e5cc266311e 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -10,9 +10,9 @@
 #include "util/util.h"
 
 #include "util/color.h"
-#include "util/list.h"
+#include <linux/list.h>
 #include "util/cache.h"
-#include "util/rbtree.h"
+#include <linux/rbtree.h>
 #include "util/symbol.h"
 #include "util/string.h"
 #include "util/callchain.h"
@@ -46,6 +46,8 @@ static int		dump_trace = 0;
 static int		verbose;
 #define eprintf(x...)	do { if (verbose) fprintf(stderr, x); } while (0)
 
+static int		modules;
+
 static int		full_paths;
 
 static unsigned long	page_size;
@@ -56,8 +58,17 @@ static char		*parent_pattern = default_parent_pattern;
 static regex_t		parent_regex;
 
 static int		exclude_other = 1;
+
+static char		callchain_default_opt[] = "fractal,0.5";
+
 static int		callchain;
 
+static
+struct callchain_param	callchain_param = {
+	.mode	= CHAIN_GRAPH_ABS,
+	.min_percent = 0.5
+};
+
 static u64		sample_type;
 
 struct ip_event {
@@ -121,6 +132,7 @@ typedef union event_union {
 static LIST_HEAD(dsos);
 static struct dso *kernel_dso;
 static struct dso *vdso;
+static struct dso *hypervisor_dso;
 
 static void dsos__add(struct dso *dso)
 {
@@ -176,7 +188,7 @@ static void dsos__fprintf(FILE *fp)
 
 static struct symbol *vdso__find_symbol(struct dso *dso, u64 ip)
 {
-	return dso__find_symbol(kernel_dso, ip);
+	return dso__find_symbol(dso, ip);
 }
 
 static int load_kernel(void)
@@ -187,8 +199,8 @@ static int load_kernel(void)
 	if (!kernel_dso)
 		return -1;
 
-	err = dso__load_kernel(kernel_dso, vmlinux, NULL, verbose);
-	if (err) {
+	err = dso__load_kernel(kernel_dso, vmlinux, NULL, verbose, modules);
+	if (err <= 0) {
 		dso__delete(kernel_dso);
 		kernel_dso = NULL;
 	} else
@@ -202,6 +214,11 @@ static int load_kernel(void)
 
 	dsos__add(vdso);
 
+	hypervisor_dso = dso__new("[hypervisor]", 0);
+	if (!hypervisor_dso)
+		return -1;
+	dsos__add(hypervisor_dso);
+
 	return err;
 }
 
@@ -233,7 +250,7 @@ static u64 map__map_ip(struct map *map, u64 ip)
 	return ip - map->start + map->pgoff;
 }
 
-static u64 vdso__map_ip(struct map *map, u64 ip)
+static u64 vdso__map_ip(struct map *map __used, u64 ip)
 {
 	return ip;
 }
@@ -640,7 +657,11 @@ sort__sym_print(FILE *fp, struct hist_entry *self)
 
 	if (self->sym) {
 		ret += fprintf(fp, "[%c] %s",
-			self->dso == kernel_dso ? 'k' : '.', self->sym->name);
+			self->dso == kernel_dso ? 'k' :
+			self->dso == hypervisor_dso ? 'h' : '.', self->sym->name);
+
+		if (self->sym->module)
+			ret += fprintf(fp, "\t[%s]", self->sym->module->name);
 	} else {
 		ret += fprintf(fp, "%#016llx", (u64)self->ip);
 	}
@@ -705,7 +726,7 @@ static LIST_HEAD(hist_entry__sort_list);
 
 static int sort_dimension__add(char *tok)
 {
-	int i;
+	unsigned int i;
 
 	for (i = 0; i < ARRAY_SIZE(sort_dimensions); i++) {
 		struct sort_dimension *sd = &sort_dimensions[i];
@@ -775,8 +796,109 @@ hist_entry__collapse(struct hist_entry *left, struct hist_entry *right)
 	return cmp;
 }
 
+static size_t ipchain__fprintf_graph_line(FILE *fp, int depth, int depth_mask)
+{
+	int i;
+	size_t ret = 0;
+
+	ret += fprintf(fp, "%s", "                ");
+
+	for (i = 0; i < depth; i++)
+		if (depth_mask & (1 << i))
+			ret += fprintf(fp, "|          ");
+		else
+			ret += fprintf(fp, "           ");
+
+	ret += fprintf(fp, "\n");
+
+	return ret;
+}
+static size_t
+ipchain__fprintf_graph(FILE *fp, struct callchain_list *chain, int depth,
+		       int depth_mask, int count, u64 total_samples,
+		       int hits)
+{
+	int i;
+	size_t ret = 0;
+
+	ret += fprintf(fp, "%s", "                ");
+	for (i = 0; i < depth; i++) {
+		if (depth_mask & (1 << i))
+			ret += fprintf(fp, "|");
+		else
+			ret += fprintf(fp, " ");
+		if (!count && i == depth - 1) {
+			double percent;
+
+			percent = hits * 100.0 / total_samples;
+			ret += percent_color_fprintf(fp, "--%2.2f%%-- ", percent);
+		} else
+			ret += fprintf(fp, "%s", "          ");
+	}
+	if (chain->sym)
+		ret += fprintf(fp, "%s\n", chain->sym->name);
+	else
+		ret += fprintf(fp, "%p\n", (void *)(long)chain->ip);
+
+	return ret;
+}
+
+static size_t
+callchain__fprintf_graph(FILE *fp, struct callchain_node *self,
+			u64 total_samples, int depth, int depth_mask)
+{
+	struct rb_node *node, *next;
+	struct callchain_node *child;
+	struct callchain_list *chain;
+	int new_depth_mask = depth_mask;
+	u64 new_total;
+	size_t ret = 0;
+	int i;
+
+	if (callchain_param.mode == CHAIN_GRAPH_REL)
+		new_total = self->cumul_hit;
+	else
+		new_total = total_samples;
+
+	node = rb_first(&self->rb_root);
+	while (node) {
+		child = rb_entry(node, struct callchain_node, rb_node);
+
+		/*
+		 * The depth mask manages the output of pipes that show
+		 * the depth. We don't want to keep the pipes of the current
+		 * level for the last child of this depth
+		 */
+		next = rb_next(node);
+		if (!next)
+			new_depth_mask &= ~(1 << (depth - 1));
+
+		/*
+		 * But we keep the older depth mask for the line seperator
+		 * to keep the level link until we reach the last child
+		 */
+		ret += ipchain__fprintf_graph_line(fp, depth, depth_mask);
+		i = 0;
+		list_for_each_entry(chain, &child->val, list) {
+			if (chain->ip >= PERF_CONTEXT_MAX)
+				continue;
+			ret += ipchain__fprintf_graph(fp, chain, depth,
+						      new_depth_mask, i++,
+						      new_total,
+						      child->cumul_hit);
+		}
+		ret += callchain__fprintf_graph(fp, child, new_total,
+						depth + 1,
+						new_depth_mask | (1 << depth));
+		node = next;
+	}
+
+	return ret;
+}
+
 static size_t
-callchain__fprintf(FILE *fp, struct callchain_node *self, u64 total_samples)
+callchain__fprintf_flat(FILE *fp, struct callchain_node *self,
+			u64 total_samples)
 {
 	struct callchain_list *chain;
 	size_t ret = 0;
@@ -784,11 +906,18 @@ callchain__fprintf(FILE *fp, struct callchain_node *self, u64 total_samples)
 	if (!self)
 		return 0;
 
-	ret += callchain__fprintf(fp, self->parent, total_samples);
+	ret += callchain__fprintf_flat(fp, self->parent, total_samples);
 
 
-	list_for_each_entry(chain, &self->val, list)
-		ret += fprintf(fp, "                %p\n", (void *)chain->ip);
+	list_for_each_entry(chain, &self->val, list) {
+		if (chain->ip >= PERF_CONTEXT_MAX)
+			continue;
+		if (chain->sym)
+			ret += fprintf(fp, "                %s\n", chain->sym->name);
+		else
+			ret += fprintf(fp, "                %p\n",
+					(void *)(long)chain->ip);
+	}
 
 	return ret;
 }
@@ -807,8 +936,19 @@ hist_entry_callchain__fprintf(FILE *fp, struct hist_entry *self,
 
 		chain = rb_entry(rb_node, struct callchain_node, rb_node);
 		percent = chain->hit * 100.0 / total_samples;
-		ret += fprintf(fp, "           %6.2f%%\n", percent);
-		ret += callchain__fprintf(fp, chain, total_samples);
+		switch (callchain_param.mode) {
+		case CHAIN_FLAT:
+			ret += percent_color_fprintf(fp, "           %6.2f%%\n",
+						     percent);
+			ret += callchain__fprintf_flat(fp, chain, total_samples);
+			break;
+		case CHAIN_GRAPH_ABS: /* Falldown */
+		case CHAIN_GRAPH_REL:
+			ret += callchain__fprintf_graph(fp, chain,
+							total_samples, 1, 1);
+		default:
+			break;
+		}
 		ret += fprintf(fp, "\n");
 		rb_node = rb_next(rb_node);
 	}
@@ -826,25 +966,10 @@ hist_entry__fprintf(FILE *fp, struct hist_entry *self, u64 total_samples)
 	if (exclude_other && !self->parent)
 		return 0;
 
-	if (total_samples) {
-		double percent = self->count * 100.0 / total_samples;
-		char *color = PERF_COLOR_NORMAL;
-
-		/*
-		 * We color high-overhead entries in red, mid-overhead
-		 * entries in green - and keep the low overhead places
-		 * normal:
-		 */
-		if (percent >= 5.0) {
-			color = PERF_COLOR_RED;
-		} else {
-			if (percent >= 0.5)
-				color = PERF_COLOR_GREEN;
-		}
-
-		ret = color_fprintf(fp, color, "   %6.2f%%",
+	if (total_samples)
+		ret = percent_color_fprintf(fp, "   %6.2f%%",
 				(self->count * 100.0) / total_samples);
-	} else
+	else
 		ret = fprintf(fp, "%12Ld ", self->count);
 
 	list_for_each_entry(se, &hist_entry__sort_list, list) {
@@ -923,6 +1048,58 @@ static int call__match(struct symbol *sym)
 	return 0;
 }
 
+static struct symbol **
+resolve_callchain(struct thread *thread, struct map *map __used,
+		    struct ip_callchain *chain, struct hist_entry *entry)
+{
+	u64 context = PERF_CONTEXT_MAX;
+	struct symbol **syms = NULL;
+	unsigned int i;
+
+	if (callchain) {
+		syms = calloc(chain->nr, sizeof(*syms));
+		if (!syms) {
+			fprintf(stderr, "Can't allocate memory for symbols\n");
+			exit(-1);
+		}
+	}
+
+	for (i = 0; i < chain->nr; i++) {
+		u64 ip = chain->ips[i];
+		struct dso *dso = NULL;
+		struct symbol *sym;
+
+		if (ip >= PERF_CONTEXT_MAX) {
+			context = ip;
+			continue;
+		}
+
+		switch (context) {
+		case PERF_CONTEXT_HV:
+			dso = hypervisor_dso;
+			break;
+		case PERF_CONTEXT_KERNEL:
+			dso = kernel_dso;
+			break;
+		default:
+			break;
+		}
+
+		sym = resolve_symbol(thread, NULL, &dso, &ip);
+
+		if (sym) {
+			if (sort__has_parent && call__match(sym) &&
+			    !entry->parent)
+				entry->parent = sym;
+			if (!callchain)
+				break;
+			syms[i] = sym;
+		}
+	}
+
+	return syms;
+}
+
 /*
  * collect histogram counts
  */
@@ -935,6 +1112,7 @@ hist_entry__add(struct thread *thread, struct map *map, struct dso *dso,
 	struct rb_node **p = &hist.rb_node;
 	struct rb_node *parent = NULL;
 	struct hist_entry *he;
+	struct symbol **syms = NULL;
 	struct hist_entry entry = {
 		.thread	= thread,
 		.map	= map,
@@ -948,36 +1126,8 @@ hist_entry__add(struct thread *thread, struct map *map, struct dso *dso,
 	};
 	int cmp;
 
-	if (sort__has_parent && chain) {
-		u64 context = PERF_CONTEXT_MAX;
-		int i;
-
-		for (i = 0; i < chain->nr; i++) {
-			u64 ip = chain->ips[i];
-			struct dso *dso = NULL;
-			struct symbol *sym;
-
-			if (ip >= PERF_CONTEXT_MAX) {
-				context = ip;
-				continue;
-			}
-
-			switch (context) {
-			case PERF_CONTEXT_KERNEL:
-				dso = kernel_dso;
-				break;
-			default:
-				break;
-			}
-
-			sym = resolve_symbol(thread, NULL, &dso, &ip);
-
-			if (sym && call__match(sym)) {
-				entry.parent = sym;
-				break;
-			}
-		}
-	}
+	if ((sort__has_parent || callchain) && chain)
+		syms = resolve_callchain(thread, map, chain, &entry);
 
 	while (*p != NULL) {
 		parent = *p;
@@ -987,8 +1137,10 @@ hist_entry__add(struct thread *thread, struct map *map, struct dso *dso,
 
 		if (!cmp) {
 			he->count += count;
-			if (callchain)
-				append_chain(&he->callchain, chain);
+			if (callchain) {
+				append_chain(&he->callchain, chain, syms);
+				free(syms);
+			}
 			return 0;
 		}
 
@@ -1004,7 +1156,8 @@ hist_entry__add(struct thread *thread, struct map *map, struct dso *dso,
 	*he = entry;
 	if (callchain) {
 		callchain_init(&he->callchain);
-		append_chain(&he->callchain, chain);
+		append_chain(&he->callchain, chain, syms);
+		free(syms);
 	}
 	rb_link_node(&he->rb_node, parent, p);
 	rb_insert_color(&he->rb_node, &hist);
@@ -1076,14 +1229,15 @@ static void collapse__resort(void)
 
 static struct rb_root output_hists;
 
-static void output__insert_entry(struct hist_entry *he)
+static void output__insert_entry(struct hist_entry *he, u64 min_callchain_hits)
 {
 	struct rb_node **p = &output_hists.rb_node;
 	struct rb_node *parent = NULL;
 	struct hist_entry *iter;
 
 	if (callchain)
-		sort_chain_to_rbtree(&he->sorted_chain, &he->callchain);
+		callchain_param.sort(&he->sorted_chain, &he->callchain,
+				      min_callchain_hits, &callchain_param);
 
 	while (*p != NULL) {
 		parent = *p;
@@ -1099,11 +1253,14 @@ static void output__insert_entry(struct hist_entry *he)
 	rb_insert_color(&he->rb_node, &output_hists);
 }
 
-static void output__resort(void)
+static void output__resort(u64 total_samples)
 {
 	struct rb_node *next;
 	struct hist_entry *n;
 	struct rb_root *tree = &hist;
+	u64 min_callchain_hits;
+
+	min_callchain_hits = total_samples * (callchain_param.min_percent / 100);
 
 	if (sort__need_collapse)
 		tree = &collapse_hists;
@@ -1115,7 +1272,7 @@ static void output__resort(void)
 		next = rb_next(&n->rb_node);
 
 		rb_erase(&n->rb_node, tree);
-		output__insert_entry(n);
+		output__insert_entry(n, min_callchain_hits);
 	}
 }
 
@@ -1141,7 +1298,7 @@ static size_t output__fprintf(FILE *fp, u64 total_samples)
 
 	fprintf(fp, "# ........");
 	list_for_each_entry(se, &hist_entry__sort_list, list) {
-		int i;
+		unsigned int i;
 
 		if (exclude_other && (se == &sort_parent))
 			continue;
@@ -1213,6 +1370,7 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 	struct map *map = NULL;
 	void *more_data = event->ip.__more_data;
 	struct ip_callchain *chain = NULL;
+	int cpumode;
 
 	if (sample_type & PERF_SAMPLE_PERIOD) {
 		period = *(u64 *)more_data;
@@ -1228,7 +1386,7 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 		(long long)period);
 
 	if (sample_type & PERF_SAMPLE_CALLCHAIN) {
-		int i;
+		unsigned int i;
 
 		chain = (void *)more_data;
 
@@ -1256,7 +1414,9 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 	if (comm_list && !strlist__has_entry(comm_list, thread->comm))
 		return 0;
 
-	if (event->header.misc & PERF_EVENT_MISC_KERNEL) {
+	cpumode = event->header.misc & PERF_EVENT_MISC_CPUMODE_MASK;
+
+	if (cpumode == PERF_EVENT_MISC_KERNEL) {
 		show = SHOW_KERNEL;
 		level = 'k';
 
@@ -1264,7 +1424,7 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 
 		dprintf(" ...... dso: %s\n", dso->name);
 
-	} else if (event->header.misc & PERF_EVENT_MISC_USER) {
+	} else if (cpumode == PERF_EVENT_MISC_USER) {
 
 		show = SHOW_USER;
 		level = '.';
@@ -1272,6 +1432,9 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 	} else {
 		show = SHOW_HV;
 		level = 'H';
+
+		dso = hypervisor_dso;
+
 		dprintf(" ...... dso: [hypervisor]\n");
 	}
 
@@ -1534,9 +1697,19 @@ static int __cmd_report(void)
 
 	sample_type = perf_header__sample_type();
 
-	if (sort__has_parent && !(sample_type & PERF_SAMPLE_CALLCHAIN)) {
-		fprintf(stderr, "selected --sort parent, but no callchain data\n");
-		exit(-1);
+	if (!(sample_type & PERF_SAMPLE_CALLCHAIN)) {
+		if (sort__has_parent) {
+			fprintf(stderr, "selected --sort parent, but no"
+					" callchain data. Did you call"
+					" perf record without -g?\n");
+			exit(-1);
+		}
+		if (callchain) {
+			fprintf(stderr, "selected -c but no callchain data."
+					" Did you call perf record without"
+					" -g?\n");
+			exit(-1);
+		}
 	}
 
 	if (load_kernel() < 0) {
@@ -1619,7 +1792,7 @@ more:
 	if (offset + head >= header->data_offset + header->data_size)
 		goto done;
 
-	if (offset + head < stat.st_size)
+	if (offset + head < (unsigned long)stat.st_size)
 		goto more;
 
 done:
@@ -1643,12 +1816,58 @@ done:
 		dsos__fprintf(stdout);
 
 	collapse__resort();
-	output__resort();
+	output__resort(total);
 	output__fprintf(stdout, total);
 
 	return rc;
 }
 
+static int
+parse_callchain_opt(const struct option *opt __used, const char *arg,
+		    int unset __used)
+{
+	char *tok;
+	char *endptr;
+
+	callchain = 1;
+
+	if (!arg)
+		return 0;
+
+	tok = strtok((char *)arg, ",");
+	if (!tok)
+		return -1;
+
+	/* get the output mode */
+	if (!strncmp(tok, "graph", strlen(arg)))
+		callchain_param.mode = CHAIN_GRAPH_ABS;
+
+	else if (!strncmp(tok, "flat", strlen(arg)))
+		callchain_param.mode = CHAIN_FLAT;
+
+	else if (!strncmp(tok, "fractal", strlen(arg)))
+		callchain_param.mode = CHAIN_GRAPH_REL;
+
+	else
+		return -1;
+
+	/* get the min percentage */
+	tok = strtok(NULL, ",");
+	if (!tok)
+		goto setup;
+
+	callchain_param.min_percent = strtod(tok, &endptr);
+	if (tok == endptr)
+		return -1;
+
+setup:
+	if (register_callchain_param(&callchain_param) < 0) {
+		fprintf(stderr, "Can't register callchain params\n");
+		return -1;
+	}
+	return 0;
+}
+
 static const char * const report_usage[] = {
 	"perf report [<options>] <command>",
 	NULL
@@ -1662,6 +1881,8 @@ static const struct option options[] = {
 	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
 		    "dump raw trace in ASCII"),
 	OPT_STRING('k', "vmlinux", &vmlinux, "file", "vmlinux pathname"),
+	OPT_BOOLEAN('m', "modules", &modules,
+		    "load module symbols - WARNING: use only with -k and LIVE kernel"),
 	OPT_STRING('s', "sort", &sort_order, "key[,key2...]",
 		   "sort by key(s): pid, comm, dso, symbol, parent"),
 	OPT_BOOLEAN('P', "full-paths", &full_paths,
@@ -1670,7 +1891,9 @@ static const struct option options[] = {
 		   "regex filter to identify parent, see: '--sort parent'"),
 	OPT_BOOLEAN('x', "exclude-other", &exclude_other,
 		    "Only display entries with parent-match"),
-	OPT_BOOLEAN('c', "callchain", &callchain, "Display callchains"),
+	OPT_CALLBACK_DEFAULT('c', "callchain", NULL, "output_type,min_percent",
+		     "Display callchains using output_type and min percent threshold. "
+		     "Default: flat,0", &parse_callchain_opt, callchain_default_opt),
 	OPT_STRING('d', "dsos", &dso_list_str, "dso[,dso...]",
 		   "only consider symbols in these dsos"),
 	OPT_STRING('C', "comms", &comm_list_str, "comm[,comm...]",
@@ -1708,7 +1931,7 @@ static void setup_list(struct strlist **list, const char *list_str,
 	}
 }
 
-int cmd_report(int argc, const char **argv, const char *prefix)
+int cmd_report(int argc, const char **argv, const char *prefix __used)
 {
 	symbol__init();