initial_commit

143 files changed, 34086 insertions, 0 deletions

diff --git a/tools/perf/util/PERF-VERSION-GEN b/tools/perf/util/PERF-VERSION-GEN
new file mode 100755
index 00000000..ad73300f
--- /dev/null
+++ b/tools/perf/util/PERF-VERSION-GEN
@@ -0,0 +1,42 @@
+#!/bin/sh
+
+if [ $# -eq 1 ]  ; then
+	OUTPUT=$1
+fi
+
+GVF=${OUTPUT}PERF-VERSION-FILE
+
+LF='
+'
+
+# First check if there is a .git to get the version from git describe
+# otherwise try to get the version from the kernel makefile
+if test -d ../../.git -o -f ../../.git &&
+	VN=$(git describe --abbrev=4 HEAD 2>/dev/null) &&
+	case "$VN" in
+	*$LF*) (exit 1) ;;
+	v[0-9]*)
+		git update-index -q --refresh
+		test -z "$(git diff-index --name-only HEAD --)" ||
+		VN="$VN-dirty" ;;
+	esac
+then
+	VN=$(echo "$VN" | sed -e 's/-/./g');
+else
+	VN=$(MAKEFLAGS= make -sC ../.. kernelversion)
+fi
+
+VN=$(expr "$VN" : v*'\(.*\)')
+
+if test -r $GVF
+then
+	VC=$(sed -e 's/^PERF_VERSION = //' <$GVF)
+else
+	VC=unset
+fi
+test "$VN" = "$VC" || {
+	echo >&2 "PERF_VERSION = $VN"
+	echo "PERF_VERSION = $VN" >$GVF
+}
+
+
diff --git a/tools/perf/util/abspath.c b/tools/perf/util/abspath.c
new file mode 100644
index 00000000..0e76affe
--- /dev/null
+++ b/tools/perf/util/abspath.c
@@ -0,0 +1,37 @@
+#include "cache.h"
+
+static const char *get_pwd_cwd(void)
+{
+	static char cwd[PATH_MAX + 1];
+	char *pwd;
+	struct stat cwd_stat, pwd_stat;
+	if (getcwd(cwd, PATH_MAX) == NULL)
+		return NULL;
+	pwd = getenv("PWD");
+	if (pwd && strcmp(pwd, cwd)) {
+		stat(cwd, &cwd_stat);
+		if (!stat(pwd, &pwd_stat) &&
+		    pwd_stat.st_dev == cwd_stat.st_dev &&
+		    pwd_stat.st_ino == cwd_stat.st_ino) {
+			strlcpy(cwd, pwd, PATH_MAX);
+		}
+	}
+	return cwd;
+}
+
+const char *make_nonrelative_path(const char *path)
+{
+	static char buf[PATH_MAX + 1];
+
+	if (is_absolute_path(path)) {
+		if (strlcpy(buf, path, PATH_MAX) >= PATH_MAX)
+			die("Too long path: %.*s", 60, path);
+	} else {
+		const char *cwd = get_pwd_cwd();
+		if (!cwd)
+			die("Cannot determine the current working directory");
+		if (snprintf(buf, PATH_MAX, "%s/%s", cwd, path) >= PATH_MAX)
+			die("Too long path: %.*s", 60, path);
+	}
+	return buf;
+}
diff --git a/tools/perf/util/alias.c b/tools/perf/util/alias.c
new file mode 100644
index 00000000..b8144e80
--- /dev/null
+++ b/tools/perf/util/alias.c
@@ -0,0 +1,77 @@
+#include "cache.h"
+
+static const char *alias_key;
+static char *alias_val;
+
+static int alias_lookup_cb(const char *k, const char *v, void *cb __used)
+{
+	if (!prefixcmp(k, "alias.") && !strcmp(k+6, alias_key)) {
+		if (!v)
+			return config_error_nonbool(k);
+		alias_val = strdup(v);
+		return 0;
+	}
+	return 0;
+}
+
+char *alias_lookup(const char *alias)
+{
+	alias_key = alias;
+	alias_val = NULL;
+	perf_config(alias_lookup_cb, NULL);
+	return alias_val;
+}
+
+int split_cmdline(char *cmdline, const char ***argv)
+{
+	int src, dst, count = 0, size = 16;
+	char quoted = 0;
+
+	*argv = malloc(sizeof(char*) * size);
+
+	/* split alias_string */
+	(*argv)[count++] = cmdline;
+	for (src = dst = 0; cmdline[src];) {
+		char c = cmdline[src];
+		if (!quoted && isspace(c)) {
+			cmdline[dst++] = 0;
+			while (cmdline[++src]
+					&& isspace(cmdline[src]))
+				; /* skip */
+			if (count >= size) {
+				size += 16;
+				*argv = realloc(*argv, sizeof(char*) * size);
+			}
+			(*argv)[count++] = cmdline + dst;
+		} else if (!quoted && (c == '\'' || c == '"')) {
+			quoted = c;
+			src++;
+		} else if (c == quoted) {
+			quoted = 0;
+			src++;
+		} else {
+			if (c == '\\' && quoted != '\'') {
+				src++;
+				c = cmdline[src];
+				if (!c) {
+					free(*argv);
+					*argv = NULL;
+					return error("cmdline ends with \\");
+				}
+			}
+			cmdline[dst++] = c;
+			src++;
+		}
+	}
+
+	cmdline[dst] = 0;
+
+	if (quoted) {
+		free(*argv);
+		*argv = NULL;
+		return error("unclosed quote");
+	}
+
+	return count;
+}
+
diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c
new file mode 100644
index 00000000..e01af2b1
--- /dev/null
+++ b/tools/perf/util/annotate.c
@@ -0,0 +1,605 @@
+/*
+ * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Parts came from builtin-annotate.c, see those files for further
+ * copyright notes.
+ *
+ * Released under the GPL v2. (and only v2, not any later version)
+ */
+
+#include "util.h"
+#include "build-id.h"
+#include "color.h"
+#include "cache.h"
+#include "symbol.h"
+#include "debug.h"
+#include "annotate.h"
+#include <pthread.h>
+
+int symbol__annotate_init(struct map *map __used, struct symbol *sym)
+{
+	struct annotation *notes = symbol__annotation(sym);
+	pthread_mutex_init(&notes->lock, NULL);
+	return 0;
+}
+
+int symbol__alloc_hist(struct symbol *sym, int nevents)
+{
+	struct annotation *notes = symbol__annotation(sym);
+	size_t sizeof_sym_hist = (sizeof(struct sym_hist) +
+				  (sym->end - sym->start) * sizeof(u64));
+
+	notes->src = zalloc(sizeof(*notes->src) + nevents * sizeof_sym_hist);
+	if (notes->src == NULL)
+		return -1;
+	notes->src->sizeof_sym_hist = sizeof_sym_hist;
+	notes->src->nr_histograms   = nevents;
+	INIT_LIST_HEAD(&notes->src->source);
+	return 0;
+}
+
+void symbol__annotate_zero_histograms(struct symbol *sym)
+{
+	struct annotation *notes = symbol__annotation(sym);
+
+	pthread_mutex_lock(&notes->lock);
+	if (notes->src != NULL)
+		memset(notes->src->histograms, 0,
+		       notes->src->nr_histograms * notes->src->sizeof_sym_hist);
+	pthread_mutex_unlock(&notes->lock);
+}
+
+int symbol__inc_addr_samples(struct symbol *sym, struct map *map,
+			     int evidx, u64 addr)
+{
+	unsigned offset;
+	struct annotation *notes;
+	struct sym_hist *h;
+
+	notes = symbol__annotation(sym);
+	if (notes->src == NULL)
+		return -ENOMEM;
+
+	pr_debug3("%s: addr=%#" PRIx64 "\n", __func__, map->unmap_ip(map, addr));
+
+	if (addr >= sym->end)
+		return 0;
+
+	offset = addr - sym->start;
+	h = annotation__histogram(notes, evidx);
+	h->sum++;
+	h->addr[offset]++;
+
+	pr_debug3("%#" PRIx64 " %s: period++ [addr: %#" PRIx64 ", %#" PRIx64
+		  ", evidx=%d] => %" PRIu64 "\n", sym->start, sym->name,
+		  addr, addr - sym->start, evidx, h->addr[offset]);
+	return 0;
+}
+
+static struct objdump_line *objdump_line__new(s64 offset, char *line, size_t privsize)
+{
+	struct objdump_line *self = malloc(sizeof(*self) + privsize);
+
+	if (self != NULL) {
+		self->offset = offset;
+		self->line = line;
+	}
+
+	return self;
+}
+
+void objdump_line__free(struct objdump_line *self)
+{
+	free(self->line);
+	free(self);
+}
+
+static void objdump__add_line(struct list_head *head, struct objdump_line *line)
+{
+	list_add_tail(&line->node, head);
+}
+
+struct objdump_line *objdump__get_next_ip_line(struct list_head *head,
+					       struct objdump_line *pos)
+{
+	list_for_each_entry_continue(pos, head, node)
+		if (pos->offset >= 0)
+			return pos;
+
+	return NULL;
+}
+
+static int objdump_line__print(struct objdump_line *oline, struct symbol *sym,
+			       int evidx, u64 len, int min_pcnt,
+			       int printed, int max_lines,
+			       struct objdump_line *queue)
+{
+	static const char *prev_line;
+	static const char *prev_color;
+
+	if (oline->offset != -1) {
+		const char *path = NULL;
+		unsigned int hits = 0;
+		double percent = 0.0;
+		const char *color;
+		struct annotation *notes = symbol__annotation(sym);
+		struct source_line *src_line = notes->src->lines;
+		struct sym_hist *h = annotation__histogram(notes, evidx);
+		s64 offset = oline->offset;
+		struct objdump_line *next;
+
+		next = objdump__get_next_ip_line(&notes->src->source, oline);
+
+		while (offset < (s64)len &&
+		       (next == NULL || offset < next->offset)) {
+			if (src_line) {
+				if (path == NULL)
+					path = src_line[offset].path;
+				percent += src_line[offset].percent;
+			} else
+				hits += h->addr[offset];
+
+			++offset;
+		}
+
+		if (src_line == NULL && h->sum)
+			percent = 100.0 * hits / h->sum;
+
+		if (percent < min_pcnt)
+			return -1;
+
+		if (max_lines && printed >= max_lines)
+			return 1;
+
+		if (queue != NULL) {
+			list_for_each_entry_from(queue, &notes->src->source, node) {
+				if (queue == oline)
+					break;
+				objdump_line__print(queue, sym, evidx, len,
+						    0, 0, 1, NULL);
+			}
+		}
+
+		color = get_percent_color(percent);
+
+		/*
+		 * Also color the filename and line if needed, with
+		 * the same color than the percentage. Don't print it
+		 * twice for close colored addr with the same filename:line
+		 */
+		if (path) {
+			if (!prev_line || strcmp(prev_line, path)
+				       || color != prev_color) {
+				color_fprintf(stdout, color, " %s", path);
+				prev_line = path;
+				prev_color = color;
+			}
+		}
+
+		color_fprintf(stdout, color, " %7.2f", percent);
+		printf(" :	");
+		color_fprintf(stdout, PERF_COLOR_BLUE, "%s\n", oline->line);
+	} else if (max_lines && printed >= max_lines)
+		return 1;
+	else {
+		if (queue)
+			return -1;
+
+		if (!*oline->line)
+			printf("         :\n");
+		else
+			printf("         :	%s\n", oline->line);
+	}
+
+	return 0;
+}
+
+static int symbol__parse_objdump_line(struct symbol *sym, struct map *map,
+				      FILE *file, size_t privsize)
+{
+	struct annotation *notes = symbol__annotation(sym);
+	struct objdump_line *objdump_line;
+	char *line = NULL, *tmp, *tmp2, *c;
+	size_t line_len;
+	s64 line_ip, offset = -1;
+
+	if (getline(&line, &line_len, file) < 0)
+		return -1;
+
+	if (!line)
+		return -1;
+
+	while (line_len != 0 && isspace(line[line_len - 1]))
+		line[--line_len] = '\0';
+
+	c = strchr(line, '\n');
+	if (c)
+		*c = 0;
+
+	line_ip = -1;
+
+	/*
+	 * Strip leading spaces:
+	 */
+	tmp = line;
+	while (*tmp) {
+		if (*tmp != ' ')
+			break;
+		tmp++;
+	}
+
+	if (*tmp) {
+		/*
+		 * Parse hexa addresses followed by ':'
+		 */
+		line_ip = strtoull(tmp, &tmp2, 16);
+		if (*tmp2 != ':' || tmp == tmp2 || tmp2[1] == '\0')
+			line_ip = -1;
+	}
+
+	if (line_ip != -1) {
+		u64 start = map__rip_2objdump(map, sym->start),
+		    end = map__rip_2objdump(map, sym->end);
+
+		offset = line_ip - start;
+		if (offset < 0 || (u64)line_ip > end)
+			offset = -1;
+	}
+
+	objdump_line = objdump_line__new(offset, line, privsize);
+	if (objdump_line == NULL) {
+		free(line);
+		return -1;
+	}
+	objdump__add_line(&notes->src->source, objdump_line);
+
+	return 0;
+}
+
+int symbol__annotate(struct symbol *sym, struct map *map, size_t privsize)
+{
+	struct dso *dso = map->dso;
+	char *filename = dso__build_id_filename(dso, NULL, 0);
+	bool free_filename = true;
+	char command[PATH_MAX * 2];
+	FILE *file;
+	int err = 0;
+	char symfs_filename[PATH_MAX];
+
+	if (filename) {
+		snprintf(symfs_filename, sizeof(symfs_filename), "%s%s",
+			 symbol_conf.symfs, filename);
+	}
+
+	if (filename == NULL) {
+		if (dso->has_build_id) {
+			pr_err("Can't annotate %s: not enough memory\n",
+			       sym->name);
+			return -ENOMEM;
+		}
+		goto fallback;
+	} else if (readlink(symfs_filename, command, sizeof(command)) < 0 ||
+		   strstr(command, "[kernel.kallsyms]") ||
+		   access(symfs_filename, R_OK)) {
+		free(filename);
+fallback:
+		/*
+		 * If we don't have build-ids or the build-id file isn't in the
+		 * cache, or is just a kallsyms file, well, lets hope that this
+		 * DSO is the same as when 'perf record' ran.
+		 */
+		filename = dso->long_name;
+		snprintf(symfs_filename, sizeof(symfs_filename), "%s%s",
+			 symbol_conf.symfs, filename);
+		free_filename = false;
+	}
+
+	if (dso->symtab_type == SYMTAB__KALLSYMS) {
+		char bf[BUILD_ID_SIZE * 2 + 16] = " with build id ";
+		char *build_id_msg = NULL;
+
+		if (dso->annotate_warned)
+			goto out_free_filename;
+
+		if (dso->has_build_id) {
+			build_id__sprintf(dso->build_id,
+					  sizeof(dso->build_id), bf + 15);
+			build_id_msg = bf;
+		}
+		err = -ENOENT;
+		dso->annotate_warned = 1;
+		pr_err("Can't annotate %s: No vmlinux file%s was found in the "
+		       "path.\nPlease use 'perf buildid-cache -av vmlinux' or "
+		       "--vmlinux vmlinux.\n",
+		       sym->name, build_id_msg ?: "");
+		goto out_free_filename;
+	}
+
+	pr_debug("%s: filename=%s, sym=%s, start=%#" PRIx64 ", end=%#" PRIx64 "\n", __func__,
+		 filename, sym->name, map->unmap_ip(map, sym->start),
+		 map->unmap_ip(map, sym->end));
+
+	pr_debug("annotating [%p] %30s : [%p] %30s\n",
+		 dso, dso->long_name, sym, sym->name);
+
+	snprintf(command, sizeof(command),
+		 "objdump --start-address=0x%016" PRIx64
+		 " --stop-address=0x%016" PRIx64 " -dS -C %s|grep -v %s|expand",
+		 map__rip_2objdump(map, sym->start),
+		 map__rip_2objdump(map, sym->end),
+		 symfs_filename, filename);
+
+	pr_debug("Executing: %s\n", command);
+
+	file = popen(command, "r");
+	if (!file)
+		goto out_free_filename;
+
+	while (!feof(file))
+		if (symbol__parse_objdump_line(sym, map, file, privsize) < 0)
+			break;
+
+	pclose(file);
+out_free_filename:
+	if (free_filename)
+		free(filename);
+	return err;
+}
+
+static void insert_source_line(struct rb_root *root, struct source_line *src_line)
+{
+	struct source_line *iter;
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+
+	while (*p != NULL) {
+		parent = *p;
+		iter = rb_entry(parent, struct source_line, node);
+
+		if (src_line->percent > iter->percent)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	rb_link_node(&src_line->node, parent, p);
+	rb_insert_color(&src_line->node, root);
+}
+
+static void symbol__free_source_line(struct symbol *sym, int len)
+{
+	struct annotation *notes = symbol__annotation(sym);
+	struct source_line *src_line = notes->src->lines;
+	int i;
+
+	for (i = 0; i < len; i++)
+		free(src_line[i].path);
+
+	free(src_line);
+	notes->src->lines = NULL;
+}
+
+/* Get the filename:line for the colored entries */
+static int symbol__get_source_line(struct symbol *sym, struct map *map,
+				   int evidx, struct rb_root *root, int len,
+				   const char *filename)
+{
+	u64 start;
+	int i;
+	char cmd[PATH_MAX * 2];
+	struct source_line *src_line;
+	struct annotation *notes = symbol__annotation(sym);
+	struct sym_hist *h = annotation__histogram(notes, evidx);
+
+	if (!h->sum)
+		return 0;
+
+	src_line = notes->src->lines = calloc(len, sizeof(struct source_line));
+	if (!notes->src->lines)
+		return -1;
+
+	start = map->unmap_ip(map, sym->start);
+
+	for (i = 0; i < len; i++) {
+		char *path = NULL;
+		size_t line_len;
+		u64 offset;
+		FILE *fp;
+
+		src_line[i].percent = 100.0 * h->addr[i] / h->sum;
+		if (src_line[i].percent <= 0.5)
+			continue;
+
+		offset = start + i;
+		sprintf(cmd, "addr2line -e %s %016" PRIx64, filename, offset);
+		fp = popen(cmd, "r");
+		if (!fp)
+			continue;
+
+		if (getline(&path, &line_len, fp) < 0 || !line_len)
+			goto next;
+
+		src_line[i].path = malloc(sizeof(char) * line_len + 1);
+		if (!src_line[i].path)
+			goto next;
+
+		strcpy(src_line[i].path, path);
+		insert_source_line(root, &src_line[i]);
+
+	next:
+		pclose(fp);
+	}
+
+	return 0;
+}
+
+static void print_summary(struct rb_root *root, const char *filename)
+{
+	struct source_line *src_line;
+	struct rb_node *node;
+
+	printf("\nSorted summary for file %s\n", filename);
+	printf("----------------------------------------------\n\n");
+
+	if (RB_EMPTY_ROOT(root)) {
+		printf(" Nothing higher than %1.1f%%\n", MIN_GREEN);
+		return;
+	}
+
+	node = rb_first(root);
+	while (node) {
+		double percent;
+		const char *color;
+		char *path;
+
+		src_line = rb_entry(node, struct source_line, node);
+		percent = src_line->percent;
+		color = get_percent_color(percent);
+		path = src_line->path;
+
+		color_fprintf(stdout, color, " %7.2f %s", percent, path);
+		node = rb_next(node);
+	}
+}
+
+static void symbol__annotate_hits(struct symbol *sym, int evidx)
+{
+	struct annotation *notes = symbol__annotation(sym);
+	struct sym_hist *h = annotation__histogram(notes, evidx);
+	u64 len = sym->end - sym->start, offset;
+
+	for (offset = 0; offset < len; ++offset)
+		if (h->addr[offset] != 0)
+			printf("%*" PRIx64 ": %" PRIu64 "\n", BITS_PER_LONG / 2,
+			       sym->start + offset, h->addr[offset]);
+	printf("%*s: %" PRIu64 "\n", BITS_PER_LONG / 2, "h->sum", h->sum);
+}
+
+int symbol__annotate_printf(struct symbol *sym, struct map *map, int evidx,
+			    bool full_paths, int min_pcnt, int max_lines,
+			    int context)
+{
+	struct dso *dso = map->dso;
+	const char *filename = dso->long_name, *d_filename;
+	struct annotation *notes = symbol__annotation(sym);
+	struct objdump_line *pos, *queue = NULL;
+	int printed = 2, queue_len = 0;
+	int more = 0;
+	u64 len;
+
+	if (full_paths)
+		d_filename = filename;
+	else
+		d_filename = basename(filename);
+
+	len = sym->end - sym->start;
+
+	printf(" Percent |	Source code & Disassembly of %s\n", d_filename);
+	printf("------------------------------------------------\n");
+
+	if (verbose)
+		symbol__annotate_hits(sym, evidx);
+
+	list_for_each_entry(pos, &notes->src->source, node) {
+		if (context && queue == NULL) {
+			queue = pos;
+			queue_len = 0;
+		}
+
+		switch (objdump_line__print(pos, sym, evidx, len, min_pcnt,
+					    printed, max_lines, queue)) {
+		case 0:
+			++printed;
+			if (context) {
+				printed += queue_len;
+				queue = NULL;
+				queue_len = 0;
+			}
+			break;
+		case 1:
+			/* filtered by max_lines */
+			++more;
+			break;
+		case -1:
+		default:
+			/*
+			 * Filtered by min_pcnt or non IP lines when
+			 * context != 0
+			 */
+			if (!context)
+				break;
+			if (queue_len == context)
+				queue = list_entry(queue->node.next, typeof(*queue), node);
+			else
+				++queue_len;
+			break;
+		}
+	}
+
+	return more;
+}
+
+void symbol__annotate_zero_histogram(struct symbol *sym, int evidx)
+{
+	struct annotation *notes = symbol__annotation(sym);
+	struct sym_hist *h = annotation__histogram(notes, evidx);
+
+	memset(h, 0, notes->src->sizeof_sym_hist);
+}
+
+void symbol__annotate_decay_histogram(struct symbol *sym, int evidx)
+{
+	struct annotation *notes = symbol__annotation(sym);
+	struct sym_hist *h = annotation__histogram(notes, evidx);
+	struct objdump_line *pos;
+	int len = sym->end - sym->start;
+
+	h->sum = 0;
+
+	list_for_each_entry(pos, &notes->src->source, node) {
+		if (pos->offset != -1 && pos->offset < len) {
+			h->addr[pos->offset] = h->addr[pos->offset] * 7 / 8;
+			h->sum += h->addr[pos->offset];
+		}
+	}
+}
+
+void objdump_line_list__purge(struct list_head *head)
+{
+	struct objdump_line *pos, *n;
+
+	list_for_each_entry_safe(pos, n, head, node) {
+		list_del(&pos->node);
+		objdump_line__free(pos);
+	}
+}
+
+int symbol__tty_annotate(struct symbol *sym, struct map *map, int evidx,
+			 bool print_lines, bool full_paths, int min_pcnt,
+			 int max_lines)
+{
+	struct dso *dso = map->dso;
+	const char *filename = dso->long_name;
+	struct rb_root source_line = RB_ROOT;
+	u64 len;
+
+	if (symbol__annotate(sym, map, 0) < 0)
+		return -1;
+
+	len = sym->end - sym->start;
+
+	if (print_lines) {
+		symbol__get_source_line(sym, map, evidx, &source_line,
+					len, filename);
+		print_summary(&source_line, filename);
+	}
+
+	symbol__annotate_printf(sym, map, evidx, full_paths,
+				min_pcnt, max_lines, 0);
+	if (print_lines)
+		symbol__free_source_line(sym, len);
+
+	objdump_line_list__purge(&symbol__annotation(sym)->src->source);
+
+	return 0;
+}
diff --git a/tools/perf/util/annotate.h b/tools/perf/util/annotate.h
new file mode 100644
index 00000000..c2c28689
--- /dev/null
+++ b/tools/perf/util/annotate.h
@@ -0,0 +1,103 @@
+#ifndef __PERF_ANNOTATE_H
+#define __PERF_ANNOTATE_H
+
+#include <stdbool.h>
+#include "types.h"
+#include "symbol.h"
+#include <linux/list.h>
+#include <linux/rbtree.h>
+
+struct objdump_line {
+	struct list_head node;
+	s64		 offset;
+	char		 *line;
+};
+
+void objdump_line__free(struct objdump_line *self);
+struct objdump_line *objdump__get_next_ip_line(struct list_head *head,
+					       struct objdump_line *pos);
+
+struct sym_hist {
+	u64		sum;
+	u64		addr[0];
+};
+
+struct source_line {
+	struct rb_node	node;
+	double		percent;
+	char		*path;
+};
+
+/** struct annotated_source - symbols with hits have this attached as in sannotation
+ *
+ * @histogram: Array of addr hit histograms per event being monitored
+ * @lines: If 'print_lines' is specified, per source code line percentages
+ * @source: source parsed from objdump -dS
+ *
+ * lines is allocated, percentages calculated and all sorted by percentage
+ * when the annotation is about to be presented, so the percentages are for
+ * one of the entries in the histogram array, i.e. for the event/counter being
+ * presented. It is deallocated right after symbol__{tui,tty,etc}_annotate
+ * returns.
+ */
+struct annotated_source {
+	struct list_head   source;
+	struct source_line *lines;
+	int    		   nr_histograms;
+	int    		   sizeof_sym_hist;
+	struct sym_hist	   histograms[0];
+};
+
+struct annotation {
+	pthread_mutex_t		lock;
+	struct annotated_source *src;
+};
+
+struct sannotation {
+	struct annotation annotation;
+	struct symbol	  symbol;
+};
+
+static inline struct sym_hist *annotation__histogram(struct annotation *notes, int idx)
+{
+	return (((void *)&notes->src->histograms) +
+	 	(notes->src->sizeof_sym_hist * idx));
+}
+
+static inline struct annotation *symbol__annotation(struct symbol *sym)
+{
+	struct sannotation *a = container_of(sym, struct sannotation, symbol);
+	return &a->annotation;
+}
+
+int symbol__inc_addr_samples(struct symbol *sym, struct map *map,
+			     int evidx, u64 addr);
+int symbol__alloc_hist(struct symbol *sym, int nevents);
+void symbol__annotate_zero_histograms(struct symbol *sym);
+
+int symbol__annotate(struct symbol *sym, struct map *map, size_t privsize);
+int symbol__annotate_init(struct map *map __used, struct symbol *sym);
+int symbol__annotate_printf(struct symbol *sym, struct map *map, int evidx,
+			    bool full_paths, int min_pcnt, int max_lines,
+			    int context);
+void symbol__annotate_zero_histogram(struct symbol *sym, int evidx);
+void symbol__annotate_decay_histogram(struct symbol *sym, int evidx);
+void objdump_line_list__purge(struct list_head *head);
+
+int symbol__tty_annotate(struct symbol *sym, struct map *map, int evidx,
+			 bool print_lines, bool full_paths, int min_pcnt,
+			 int max_lines);
+
+#ifdef NO_NEWT_SUPPORT
+static inline int symbol__tui_annotate(struct symbol *sym __used,
+				       struct map *map __used,
+				       int evidx __used, int refresh __used)
+{
+	return 0;
+}
+#else
+int symbol__tui_annotate(struct symbol *sym, struct map *map, int evidx,
+			 int refresh);
+#endif
+
+#endif	/* __PERF_ANNOTATE_H */
diff --git a/tools/perf/util/bitmap.c b/tools/perf/util/bitmap.c
new file mode 100644
index 00000000..5e230aca
--- /dev/null
+++ b/tools/perf/util/bitmap.c
@@ -0,0 +1,21 @@
+/*
+ * From lib/bitmap.c
+ * Helper functions for bitmap.h.
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2.  See the file COPYING for more details.
+ */
+#include <linux/bitmap.h>
+
+int __bitmap_weight(const unsigned long *bitmap, int bits)
+{
+	int k, w = 0, lim = bits/BITS_PER_LONG;
+
+	for (k = 0; k < lim; k++)
+		w += hweight_long(bitmap[k]);
+
+	if (bits % BITS_PER_LONG)
+		w += hweight_long(bitmap[k] & BITMAP_LAST_WORD_MASK(bits));
+
+	return w;
+}
diff --git a/tools/perf/util/build-id.c b/tools/perf/util/build-id.c
new file mode 100644
index 00000000..a91cd99f
--- /dev/null
+++ b/tools/perf/util/build-id.c
@@ -0,0 +1,82 @@
+/*
+ * build-id.c
+ *
+ * build-id support
+ *
+ * Copyright (C) 2009, 2010 Red Hat Inc.
+ * Copyright (C) 2009, 2010 Arnaldo Carvalho de Melo <acme@redhat.com>
+ */
+#include "util.h"
+#include <stdio.h>
+#include "build-id.h"
+#include "event.h"
+#include "symbol.h"
+#include <linux/kernel.h>
+#include "debug.h"
+
+static int build_id__mark_dso_hit(union perf_event *event,
+				  struct perf_sample *sample __used,
+				  struct perf_evsel *evsel __used,
+				  struct perf_session *session)
+{
+	struct addr_location al;
+	u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+	struct thread *thread = perf_session__findnew(session, event->ip.pid);
+
+	if (thread == NULL) {
+		pr_err("problem processing %d event, skipping it.\n",
+			event->header.type);
+		return -1;
+	}
+
+	thread__find_addr_map(thread, session, cpumode, MAP__FUNCTION,
+			      event->ip.pid, event->ip.ip, &al);
+
+	if (al.map != NULL)
+		al.map->dso->hit = 1;
+
+	return 0;
+}
+
+static int perf_event__exit_del_thread(union perf_event *event,
+				       struct perf_sample *sample __used,
+				       struct perf_session *session)
+{
+	struct thread *thread = perf_session__findnew(session, event->fork.tid);
+
+	dump_printf("(%d:%d):(%d:%d)\n", event->fork.pid, event->fork.tid,
+		    event->fork.ppid, event->fork.ptid);
+
+	if (thread) {
+		rb_erase(&thread->rb_node, &session->threads);
+		session->last_match = NULL;
+		thread__delete(thread);
+	}
+
+	return 0;
+}
+
+struct perf_event_ops build_id__mark_dso_hit_ops = {
+	.sample	= build_id__mark_dso_hit,
+	.mmap	= perf_event__process_mmap,
+	.fork	= perf_event__process_task,
+	.exit	= perf_event__exit_del_thread,
+};
+
+char *dso__build_id_filename(struct dso *self, char *bf, size_t size)
+{
+	char build_id_hex[BUILD_ID_SIZE * 2 + 1];
+
+	if (!self->has_build_id)
+		return NULL;
+
+	build_id__sprintf(self->build_id, sizeof(self->build_id), build_id_hex);
+	if (bf == NULL) {
+		if (asprintf(&bf, "%s/.build-id/%.2s/%s", buildid_dir,
+			     build_id_hex, build_id_hex + 2) < 0)
+			return NULL;
+	} else
+		snprintf(bf, size, "%s/.build-id/%.2s/%s", buildid_dir,
+			 build_id_hex, build_id_hex + 2);
+	return bf;
+}
diff --git a/tools/perf/util/build-id.h b/tools/perf/util/build-id.h
new file mode 100644
index 00000000..5dafb00e
--- /dev/null
+++ b/tools/perf/util/build-id.h
@@ -0,0 +1,10 @@
+#ifndef PERF_BUILD_ID_H_
+#define PERF_BUILD_ID_H_ 1
+
+#include "session.h"
+
+extern struct perf_event_ops build_id__mark_dso_hit_ops;
+
+char *dso__build_id_filename(struct dso *self, char *bf, size_t size);
+
+#endif
diff --git a/tools/perf/util/cache.h b/tools/perf/util/cache.h
new file mode 100644
index 00000000..fc5e5a09
--- /dev/null
+++ b/tools/perf/util/cache.h
@@ -0,0 +1,90 @@
+#ifndef __PERF_CACHE_H
+#define __PERF_CACHE_H
+
+#include <stdbool.h>
+#include "util.h"
+#include "strbuf.h"
+#include "../perf.h"
+
+#define CMD_EXEC_PATH "--exec-path"
+#define CMD_PERF_DIR "--perf-dir="
+#define CMD_WORK_TREE "--work-tree="
+#define CMD_DEBUGFS_DIR "--debugfs-dir="
+
+#define PERF_DIR_ENVIRONMENT "PERF_DIR"
+#define PERF_WORK_TREE_ENVIRONMENT "PERF_WORK_TREE"
+#define EXEC_PATH_ENVIRONMENT "PERF_EXEC_PATH"
+#define DEFAULT_PERF_DIR_ENVIRONMENT ".perf"
+#define PERF_DEBUGFS_ENVIRONMENT "PERF_DEBUGFS_DIR"
+
+typedef int (*config_fn_t)(const char *, const char *, void *);
+extern int perf_default_config(const char *, const char *, void *);
+extern int perf_config(config_fn_t fn, void *);
+extern int perf_config_int(const char *, const char *);
+extern int perf_config_bool(const char *, const char *);
+extern int config_error_nonbool(const char *);
+extern const char *perf_config_dirname(const char *, const char *);
+
+/* pager.c */
+extern void setup_pager(void);
+extern const char *pager_program;
+extern int pager_in_use(void);
+extern int pager_use_color;
+
+extern int use_browser;
+
+#ifdef NO_NEWT_SUPPORT
+static inline void setup_browser(bool fallback_to_pager)
+{
+	if (fallback_to_pager)
+		setup_pager();
+}
+static inline void exit_browser(bool wait_for_ok __used) {}
+#else
+void setup_browser(bool fallback_to_pager);
+void exit_browser(bool wait_for_ok);
+#endif
+
+char *alias_lookup(const char *alias);
+int split_cmdline(char *cmdline, const char ***argv);
+
+#define alloc_nr(x) (((x)+16)*3/2)
+
+/*
+ * Realloc the buffer pointed at by variable 'x' so that it can hold
+ * at least 'nr' entries; the number of entries currently allocated
+ * is 'alloc', using the standard growing factor alloc_nr() macro.
+ *
+ * DO NOT USE any expression with side-effect for 'x' or 'alloc'.
+ */
+#define ALLOC_GROW(x, nr, alloc) \
+	do { \
+		if ((nr) > alloc) { \
+			if (alloc_nr(alloc) < (nr)) \
+				alloc = (nr); \
+			else \
+				alloc = alloc_nr(alloc); \
+			x = xrealloc((x), alloc * sizeof(*(x))); \
+		} \
+	} while(0)
+
+
+static inline int is_absolute_path(const char *path)
+{
+	return path[0] == '/';
+}
+
+const char *make_nonrelative_path(const char *path);
+char *strip_path_suffix(const char *path, const char *suffix);
+
+extern char *mkpath(const char *fmt, ...) __attribute__((format (printf, 1, 2)));
+extern char *perf_path(const char *fmt, ...) __attribute__((format (printf, 1, 2)));
+
+extern char *perf_pathdup(const char *fmt, ...)
+	__attribute__((format (printf, 1, 2)));
+
+#ifdef NO_STRLCPY
+extern size_t strlcpy(char *dest, const char *src, size_t size);
+#endif
+
+#endif /* __PERF_CACHE_H */
diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c
new file mode 100644
index 00000000..9f7106a8
--- /dev/null
+++ b/tools/perf/util/callchain.c
@@ -0,0 +1,461 @@
+/*
+ * Copyright (C) 2009-2011, Frederic Weisbecker <fweisbec@gmail.com>
+ *
+ * Handle the callchains from the stream in an ad-hoc radix tree and then
+ * sort them in an rbtree.
+ *
+ * Using a radix for code path provides a fast retrieval and factorizes
+ * memory use. Also that lets us use the paths in a hierarchical graph view.
+ *
+ */
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <stdbool.h>
+#include <errno.h>
+#include <math.h>
+
+#include "util.h"
+#include "callchain.h"
+
+bool ip_callchain__valid(struct ip_callchain *chain,
+			 const union perf_event *event)
+{
+	unsigned int chain_size = event->header.size;
+	chain_size -= (unsigned long)&event->ip.__more_data - (unsigned long)event;
+	return chain->nr * sizeof(u64) <= chain_size;
+}
+
+#define chain_for_each_child(child, parent)	\
+	list_for_each_entry(child, &parent->children, siblings)
+
+#define chain_for_each_child_safe(child, next, parent)	\
+	list_for_each_entry_safe(child, next, &parent->children, siblings)
+
+static void
+rb_insert_callchain(struct rb_root *root, struct callchain_node *chain,
+		    enum chain_mode mode)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	struct callchain_node *rnode;
+	u64 chain_cumul = callchain_cumul_hits(chain);
+
+	while (*p) {
+		u64 rnode_cumul;
+
+		parent = *p;
+		rnode = rb_entry(parent, struct callchain_node, rb_node);
+		rnode_cumul = callchain_cumul_hits(rnode);
+
+		switch (mode) {
+		case CHAIN_FLAT:
+			if (rnode->hit < chain->hit)
+				p = &(*p)->rb_left;
+			else
+				p = &(*p)->rb_right;
+			break;
+		case CHAIN_GRAPH_ABS: /* Falldown */
+		case CHAIN_GRAPH_REL:
+			if (rnode_cumul < chain_cumul)
+				p = &(*p)->rb_left;
+			else
+				p = &(*p)->rb_right;
+			break;
+		case CHAIN_NONE:
+		default:
+			break;
+		}
+	}
+
+	rb_link_node(&chain->rb_node, parent, p);
+	rb_insert_color(&chain->rb_node, root);
+}
+
+static void
+__sort_chain_flat(struct rb_root *rb_root, struct callchain_node *node,
+		  u64 min_hit)
+{
+	struct callchain_node *child;
+
+	chain_for_each_child(child, node)
+		__sort_chain_flat(rb_root, child, min_hit);
+
+	if (node->hit && node->hit >= min_hit)
+		rb_insert_callchain(rb_root, node, CHAIN_FLAT);
+}
+
+/*
+ * Once we get every callchains from the stream, we can now
+ * sort them by hit
+ */
+static void
+sort_chain_flat(struct rb_root *rb_root, struct callchain_root *root,
+		u64 min_hit, struct callchain_param *param __used)
+{
+	__sort_chain_flat(rb_root, &root->node, min_hit);
+}
+
+static void __sort_chain_graph_abs(struct callchain_node *node,
+				   u64 min_hit)
+{
+	struct callchain_node *child;
+
+	node->rb_root = RB_ROOT;
+
+	chain_for_each_child(child, node) {
+		__sort_chain_graph_abs(child, min_hit);
+		if (callchain_cumul_hits(child) >= min_hit)
+			rb_insert_callchain(&node->rb_root, child,
+					    CHAIN_GRAPH_ABS);
+	}
+}
+
+static void
+sort_chain_graph_abs(struct rb_root *rb_root, struct callchain_root *chain_root,
+		     u64 min_hit, struct callchain_param *param __used)
+{
+	__sort_chain_graph_abs(&chain_root->node, min_hit);
+	rb_root->rb_node = chain_root->node.rb_root.rb_node;
+}
+
+static void __sort_chain_graph_rel(struct callchain_node *node,
+				   double min_percent)
+{
+	struct callchain_node *child;
+	u64 min_hit;
+
+	node->rb_root = RB_ROOT;
+	min_hit = ceil(node->children_hit * min_percent);
+
+	chain_for_each_child(child, node) {
+		__sort_chain_graph_rel(child, min_percent);
+		if (callchain_cumul_hits(child) >= min_hit)
+			rb_insert_callchain(&node->rb_root, child,
+					    CHAIN_GRAPH_REL);
+	}
+}
+
+static void
+sort_chain_graph_rel(struct rb_root *rb_root, struct callchain_root *chain_root,
+		     u64 min_hit __used, struct callchain_param *param)
+{
+	__sort_chain_graph_rel(&chain_root->node, param->min_percent / 100.0);
+	rb_root->rb_node = chain_root->node.rb_root.rb_node;
+}
+
+int callchain_register_param(struct callchain_param *param)
+{
+	switch (param->mode) {
+	case CHAIN_GRAPH_ABS:
+		param->sort = sort_chain_graph_abs;
+		break;
+	case CHAIN_GRAPH_REL:
+		param->sort = sort_chain_graph_rel;
+		break;
+	case CHAIN_FLAT:
+		param->sort = sort_chain_flat;
+		break;
+	case CHAIN_NONE:
+	default:
+		return -1;
+	}
+	return 0;
+}
+
+/*
+ * Create a child for a parent. If inherit_children, then the new child
+ * will become the new parent of it's parent children
+ */
+static struct callchain_node *
+create_child(struct callchain_node *parent, bool inherit_children)
+{
+	struct callchain_node *new;
+
+	new = zalloc(sizeof(*new));
+	if (!new) {
+		perror("not enough memory to create child for code path tree");
+		return NULL;
+	}
+	new->parent = parent;
+	INIT_LIST_HEAD(&new->children);
+	INIT_LIST_HEAD(&new->val);
+
+	if (inherit_children) {
+		struct callchain_node *next;
+
+		list_splice(&parent->children, &new->children);
+		INIT_LIST_HEAD(&parent->children);
+
+		chain_for_each_child(next, new)
+			next->parent = new;
+	}
+	list_add_tail(&new->siblings, &parent->children);
+
+	return new;
+}
+
+
+/*
+ * Fill the node with callchain values
+ */
+static void
+fill_node(struct callchain_node *node, struct callchain_cursor *cursor)
+{
+	struct callchain_cursor_node *cursor_node;
+
+	node->val_nr = cursor->nr - cursor->pos;
+	if (!node->val_nr)
+		pr_warning("Warning: empty node in callchain tree\n");
+
+	cursor_node = callchain_cursor_current(cursor);
+
+	while (cursor_node) {
+		struct callchain_list *call;
+
+		call = zalloc(sizeof(*call));
+		if (!call) {
+			perror("not enough memory for the code path tree");
+			return;
+		}
+		call->ip = cursor_node->ip;
+		call->ms.sym = cursor_node->sym;
+		call->ms.map = cursor_node->map;
+		list_add_tail(&call->list, &node->val);
+
+		callchain_cursor_advance(cursor);
+		cursor_node = callchain_cursor_current(cursor);
+	}
+}
+
+static void
+add_child(struct callchain_node *parent,
+	  struct callchain_cursor *cursor,
+	  u64 period)
+{
+	struct callchain_node *new;
+
+	new = create_child(parent, false);
+	fill_node(new, cursor);
+
+	new->children_hit = 0;
+	new->hit = period;
+}
+
+/*
+ * Split the parent in two parts (a new child is created) and
+ * give a part of its callchain to the created child.
+ * Then create another child to host the given callchain of new branch
+ */
+static void
+split_add_child(struct callchain_node *parent,
+		struct callchain_cursor *cursor,
+		struct callchain_list *to_split,
+		u64 idx_parents, u64 idx_local, u64 period)
+{
+	struct callchain_node *new;
+	struct list_head *old_tail;
+	unsigned int idx_total = idx_parents + idx_local;
+
+	/* split */
+	new = create_child(parent, true);
+
+	/* split the callchain and move a part to the new child */
+	old_tail = parent->val.prev;
+	list_del_range(&to_split->list, old_tail);
+	new->val.next = &to_split->list;
+	new->val.prev = old_tail;
+	to_split->list.prev = &new->val;
+	old_tail->next = &new->val;
+
+	/* split the hits */
+	new->hit = parent->hit;
+	new->children_hit = parent->children_hit;
+	parent->children_hit = callchain_cumul_hits(new);
+	new->val_nr = parent->val_nr - idx_local;
+	parent->val_nr = idx_local;
+
+	/* create a new child for the new branch if any */
+	if (idx_total < cursor->nr) {
+		parent->hit = 0;
+		add_child(parent, cursor, period);
+		parent->children_hit += period;
+	} else {
+		parent->hit = period;
+	}
+}
+
+static int
+append_chain(struct callchain_node *root,
+	     struct callchain_cursor *cursor,
+	     u64 period);
+
+static void
+append_chain_children(struct callchain_node *root,
+		      struct callchain_cursor *cursor,
+		      u64 period)
+{
+	struct callchain_node *rnode;
+
+	/* lookup in childrens */
+	chain_for_each_child(rnode, root) {
+		unsigned int ret = append_chain(rnode, cursor, period);
+
+		if (!ret)
+			goto inc_children_hit;
+	}
+	/* nothing in children, add to the current node */
+	add_child(root, cursor, period);
+
+inc_children_hit:
+	root->children_hit += period;
+}
+
+static int
+append_chain(struct callchain_node *root,
+	     struct callchain_cursor *cursor,
+	     u64 period)
+{
+	struct callchain_cursor_node *curr_snap = cursor->curr;
+	struct callchain_list *cnode;
+	u64 start = cursor->pos;
+	bool found = false;
+	u64 matches;
+
+	/*
+	 * Lookup in the current node
+	 * If we have a symbol, then compare the start to match
+	 * anywhere inside a function.
+	 */
+	list_for_each_entry(cnode, &root->val, list) {
+		struct callchain_cursor_node *node;
+		struct symbol *sym;
+
+		node = callchain_cursor_current(cursor);
+		if (!node)
+			break;
+
+		sym = node->sym;
+
+		if (cnode->ms.sym && sym) {
+			if (cnode->ms.sym->start != sym->start)
+				break;
+		} else if (cnode->ip != node->ip)
+			break;
+
+		if (!found)
+			found = true;
+
+		callchain_cursor_advance(cursor);
+	}
+
+	/* matches not, relay on the parent */
+	if (!found) {
+		cursor->curr = curr_snap;
+		cursor->pos = start;
+		return -1;
+	}
+
+	matches = cursor->pos - start;
+
+	/* we match only a part of the node. Split it and add the new chain */
+	if (matches < root->val_nr) {
+		split_add_child(root, cursor, cnode, start, matches, period);
+		return 0;
+	}
+
+	/* we match 100% of the path, increment the hit */
+	if (matches == root->val_nr && cursor->pos == cursor->nr) {
+		root->hit += period;
+		return 0;
+	}
+
+	/* We match the node and still have a part remaining */
+	append_chain_children(root, cursor, period);
+
+	return 0;
+}
+
+int callchain_append(struct callchain_root *root,
+		     struct callchain_cursor *cursor,
+		     u64 period)
+{
+	if (!cursor->nr)
+		return 0;
+
+	callchain_cursor_commit(cursor);
+
+	append_chain_children(&root->node, cursor, period);
+
+	if (cursor->nr > root->max_depth)
+		root->max_depth = cursor->nr;
+
+	return 0;
+}
+
+static int
+merge_chain_branch(struct callchain_cursor *cursor,
+		   struct callchain_node *dst, struct callchain_node *src)
+{
+	struct callchain_cursor_node **old_last = cursor->last;
+	struct callchain_node *child, *next_child;
+	struct callchain_list *list, *next_list;
+	int old_pos = cursor->nr;
+	int err = 0;
+
+	list_for_each_entry_safe(list, next_list, &src->val, list) {
+		callchain_cursor_append(cursor, list->ip,
+					list->ms.map, list->ms.sym);
+		list_del(&list->list);
+		free(list);
+	}
+
+	if (src->hit) {
+		callchain_cursor_commit(cursor);
+		append_chain_children(dst, cursor, src->hit);
+	}
+
+	chain_for_each_child_safe(child, next_child, src) {
+		err = merge_chain_branch(cursor, dst, child);
+		if (err)
+			break;
+
+		list_del(&child->siblings);
+		free(child);
+	}
+
+	cursor->nr = old_pos;
+	cursor->last = old_last;
+
+	return err;
+}
+
+int callchain_merge(struct callchain_cursor *cursor,
+		    struct callchain_root *dst, struct callchain_root *src)
+{
+	return merge_chain_branch(cursor, &dst->node, &src->node);
+}
+
+int callchain_cursor_append(struct callchain_cursor *cursor,
+			    u64 ip, struct map *map, struct symbol *sym)
+{
+	struct callchain_cursor_node *node = *cursor->last;
+
+	if (!node) {
+		node = calloc(sizeof(*node), 1);
+		if (!node)
+			return -ENOMEM;
+
+		*cursor->last = node;
+	}
+
+	node->ip = ip;
+	node->map = map;
+	node->sym = sym;
+
+	cursor->nr++;
+
+	cursor->last = &node->next;
+
+	return 0;
+}
diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h
new file mode 100644
index 00000000..1a79df9f
--- /dev/null
+++ b/tools/perf/util/callchain.h
@@ -0,0 +1,135 @@
+#ifndef __PERF_CALLCHAIN_H
+#define __PERF_CALLCHAIN_H
+
+#include "../perf.h"
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include "event.h"
+#include "symbol.h"
+
+enum chain_mode {
+	CHAIN_NONE,
+	CHAIN_FLAT,
+	CHAIN_GRAPH_ABS,
+	CHAIN_GRAPH_REL
+};
+
+struct callchain_node {
+	struct callchain_node	*parent;
+	struct list_head	siblings;
+	struct list_head	children;
+	struct list_head	val;
+	struct rb_node		rb_node; /* to sort nodes in an rbtree */
+	struct rb_root		rb_root; /* sorted tree of children */
+	unsigned int		val_nr;
+	u64			hit;
+	u64			children_hit;
+};
+
+struct callchain_root {
+	u64			max_depth;
+	struct callchain_node	node;
+};
+
+struct callchain_param;
+
+typedef void (*sort_chain_func_t)(struct rb_root *, struct callchain_root *,
+				 u64, struct callchain_param *);
+
+struct callchain_param {
+	enum chain_mode 	mode;
+	u32			print_limit;
+	double			min_percent;
+	sort_chain_func_t	sort;
+};
+
+struct callchain_list {
+	u64			ip;
+	struct map_symbol	ms;
+	struct list_head	list;
+};
+
+/*
+ * A callchain cursor is a single linked list that
+ * let one feed a callchain progressively.
+ * It keeps persitent allocated entries to minimize
+ * allocations.
+ */
+struct callchain_cursor_node {
+	u64				ip;
+	struct map			*map;
+	struct symbol			*sym;
+	struct callchain_cursor_node	*next;
+};
+
+struct callchain_cursor {
+	u64				nr;
+	struct callchain_cursor_node	*first;
+	struct callchain_cursor_node	**last;
+	u64				pos;
+	struct callchain_cursor_node	*curr;
+};
+
+static inline void callchain_init(struct callchain_root *root)
+{
+	INIT_LIST_HEAD(&root->node.siblings);
+	INIT_LIST_HEAD(&root->node.children);
+	INIT_LIST_HEAD(&root->node.val);
+
+	root->node.parent = NULL;
+	root->node.hit = 0;
+	root->node.children_hit = 0;
+	root->max_depth = 0;
+}
+
+static inline u64 callchain_cumul_hits(struct callchain_node *node)
+{
+	return node->hit + node->children_hit;
+}
+
+int callchain_register_param(struct callchain_param *param);
+int callchain_append(struct callchain_root *root,
+		     struct callchain_cursor *cursor,
+		     u64 period);
+
+int callchain_merge(struct callchain_cursor *cursor,
+		    struct callchain_root *dst, struct callchain_root *src);
+
+bool ip_callchain__valid(struct ip_callchain *chain,
+			 const union perf_event *event);
+/*
+ * Initialize a cursor before adding entries inside, but keep
+ * the previously allocated entries as a cache.
+ */
+static inline void callchain_cursor_reset(struct callchain_cursor *cursor)
+{
+	cursor->nr = 0;
+	cursor->last = &cursor->first;
+}
+
+int callchain_cursor_append(struct callchain_cursor *cursor, u64 ip,
+			    struct map *map, struct symbol *sym);
+
+/* Close a cursor writing session. Initialize for the reader */
+static inline void callchain_cursor_commit(struct callchain_cursor *cursor)
+{
+	cursor->curr = cursor->first;
+	cursor->pos = 0;
+}
+
+/* Cursor reading iteration helpers */
+static inline struct callchain_cursor_node *
+callchain_cursor_current(struct callchain_cursor *cursor)
+{
+	if (cursor->pos == cursor->nr)
+		return NULL;
+
+	return cursor->curr;
+}
+
+static inline void callchain_cursor_advance(struct callchain_cursor *cursor)
+{
+	cursor->curr = cursor->curr->next;
+	cursor->pos++;
+}
+#endif	/* __PERF_CALLCHAIN_H */
diff --git a/tools/perf/util/cgroup.c b/tools/perf/util/cgroup.c
new file mode 100644
index 00000000..96bee5c4
--- /dev/null
+++ b/tools/perf/util/cgroup.c
@@ -0,0 +1,178 @@
+#include "util.h"
+#include "../perf.h"
+#include "parse-options.h"
+#include "evsel.h"
+#include "cgroup.h"
+#include "debugfs.h" /* MAX_PATH, STR() */
+#include "evlist.h"
+
+int nr_cgroups;
+
+static int
+cgroupfs_find_mountpoint(char *buf, size_t maxlen)
+{
+	FILE *fp;
+	char mountpoint[MAX_PATH+1], tokens[MAX_PATH+1], type[MAX_PATH+1];
+	char *token, *saved_ptr = NULL;
+	int found = 0;
+
+	fp = fopen("/proc/mounts", "r");
+	if (!fp)
+		return -1;
+
+	/*
+	 * in order to handle split hierarchy, we need to scan /proc/mounts
+	 * and inspect every cgroupfs mount point to find one that has
+	 * perf_event subsystem
+	 */
+	while (fscanf(fp, "%*s %"STR(MAX_PATH)"s %"STR(MAX_PATH)"s %"
+				STR(MAX_PATH)"s %*d %*d\n",
+				mountpoint, type, tokens) == 3) {
+
+		if (!strcmp(type, "cgroup")) {
+
+			token = strtok_r(tokens, ",", &saved_ptr);
+
+			while (token != NULL) {
+				if (!strcmp(token, "perf_event")) {
+					found = 1;
+					break;
+				}
+				token = strtok_r(NULL, ",", &saved_ptr);
+			}
+		}
+		if (found)
+			break;
+	}
+	fclose(fp);
+	if (!found)
+		return -1;
+
+	if (strlen(mountpoint) < maxlen) {
+		strcpy(buf, mountpoint);
+		return 0;
+	}
+	return -1;
+}
+
+static int open_cgroup(char *name)
+{
+	char path[MAX_PATH+1];
+	char mnt[MAX_PATH+1];
+	int fd;
+
+
+	if (cgroupfs_find_mountpoint(mnt, MAX_PATH+1))
+		return -1;
+
+	snprintf(path, MAX_PATH, "%s/%s", mnt, name);
+
+	fd = open(path, O_RDONLY);
+	if (fd == -1)
+		fprintf(stderr, "no access to cgroup %s\n", path);
+
+	return fd;
+}
+
+static int add_cgroup(struct perf_evlist *evlist, char *str)
+{
+	struct perf_evsel *counter;
+	struct cgroup_sel *cgrp = NULL;
+	int n;
+	/*
+	 * check if cgrp is already defined, if so we reuse it
+	 */
+	list_for_each_entry(counter, &evlist->entries, node) {
+		cgrp = counter->cgrp;
+		if (!cgrp)
+			continue;
+		if (!strcmp(cgrp->name, str))
+			break;
+
+		cgrp = NULL;
+	}
+
+	if (!cgrp) {
+		cgrp = zalloc(sizeof(*cgrp));
+		if (!cgrp)
+			return -1;
+
+		cgrp->name = str;
+
+		cgrp->fd = open_cgroup(str);
+		if (cgrp->fd == -1) {
+			free(cgrp);
+			return -1;
+		}
+	}
+
+	/*
+	 * find corresponding event
+	 * if add cgroup N, then need to find event N
+	 */
+	n = 0;
+	list_for_each_entry(counter, &evlist->entries, node) {
+		if (n == nr_cgroups)
+			goto found;
+		n++;
+	}
+	if (cgrp->refcnt == 0)
+		free(cgrp);
+
+	return -1;
+found:
+	cgrp->refcnt++;
+	counter->cgrp = cgrp;
+	return 0;
+}
+
+void close_cgroup(struct cgroup_sel *cgrp)
+{
+	if (!cgrp)
+		return;
+
+	/* XXX: not reentrant */
+	if (--cgrp->refcnt == 0) {
+		close(cgrp->fd);
+		free(cgrp->name);
+		free(cgrp);
+	}
+}
+
+int parse_cgroups(const struct option *opt __used, const char *str,
+		  int unset __used)
+{
+	struct perf_evlist *evlist = *(struct perf_evlist **)opt->value;
+	const char *p, *e, *eos = str + strlen(str);
+	char *s;
+	int ret;
+
+	if (list_empty(&evlist->entries)) {
+		fprintf(stderr, "must define events before cgroups\n");
+		return -1;
+	}
+
+	for (;;) {
+		p = strchr(str, ',');
+		e = p ? p : eos;
+
+		/* allow empty cgroups, i.e., skip */
+		if (e - str) {
+			/* termination added */
+			s = strndup(str, e - str);
+			if (!s)
+				return -1;
+			ret = add_cgroup(evlist, s);
+			if (ret) {
+				free(s);
+				return -1;
+			}
+		}
+		/* nr_cgroups is increased een for empty cgroups */
+		nr_cgroups++;
+		if (!p)
+			break;
+		str = p+1;
+	}
+	return 0;
+}
diff --git a/tools/perf/util/cgroup.h b/tools/perf/util/cgroup.h
new file mode 100644
index 00000000..89acd6de
--- /dev/null
+++ b/tools/perf/util/cgroup.h
@@ -0,0 +1,17 @@
+#ifndef __CGROUP_H__
+#define __CGROUP_H__
+
+struct option;
+
+struct cgroup_sel {
+	char *name;
+	int fd;
+	int refcnt;
+};
+
+
+extern int nr_cgroups; /* number of explicit cgroups defined */
+extern void close_cgroup(struct cgroup_sel *cgrp);
+extern int parse_cgroups(const struct option *opt, const char *str, int unset);
+
+#endif /* __CGROUP_H__ */
diff --git a/tools/perf/util/color.c b/tools/perf/util/color.c
new file mode 100644
index 00000000..e191eb9a
--- /dev/null
+++ b/tools/perf/util/color.c
@@ -0,0 +1,324 @@
+#include "cache.h"
+#include "color.h"
+
+int perf_use_color_default = -1;
+
+static int parse_color(const char *name, int len)
+{
+	static const char * const color_names[] = {
+		"normal", "black", "red", "green", "yellow",
+		"blue", "magenta", "cyan", "white"
+	};
+	char *end;
+	int i;
+
+	for (i = 0; i < (int)ARRAY_SIZE(color_names); i++) {
+		const char *str = color_names[i];
+		if (!strncasecmp(name, str, len) && !str[len])
+			return i - 1;
+	}
+	i = strtol(name, &end, 10);
+	if (end - name == len && i >= -1 && i <= 255)
+		return i;
+	return -2;
+}
+
+static int parse_attr(const char *name, int len)
+{
+	static const int attr_values[] = { 1, 2, 4, 5, 7 };
+	static const char * const attr_names[] = {
+		"bold", "dim", "ul", "blink", "reverse"
+	};
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(attr_names); i++) {
+		const char *str = attr_names[i];
+		if (!strncasecmp(name, str, len) && !str[len])
+			return attr_values[i];
+	}
+	return -1;
+}
+
+void color_parse(const char *value, const char *var, char *dst)
+{
+	color_parse_mem(value, strlen(value), var, dst);
+}
+
+void color_parse_mem(const char *value, int value_len, const char *var,
+		char *dst)
+{
+	const char *ptr = value;
+	int len = value_len;
+	int attr = -1;
+	int fg = -2;
+	int bg = -2;
+
+	if (!strncasecmp(value, "reset", len)) {
+		strcpy(dst, PERF_COLOR_RESET);
+		return;
+	}
+
+	/* [fg [bg]] [attr] */
+	while (len > 0) {
+		const char *word = ptr;
+		int val, wordlen = 0;
+
+		while (len > 0 && !isspace(word[wordlen])) {
+			wordlen++;
+			len--;
+		}
+
+		ptr = word + wordlen;
+		while (len > 0 && isspace(*ptr)) {
+			ptr++;
+			len--;
+		}
+
+		val = parse_color(word, wordlen);
+		if (val >= -1) {
+			if (fg == -2) {
+				fg = val;
+				continue;
+			}
+			if (bg == -2) {
+				bg = val;
+				continue;
+			}
+			goto bad;
+		}
+		val = parse_attr(word, wordlen);
+		if (val < 0 || attr != -1)
+			goto bad;
+		attr = val;
+	}
+
+	if (attr >= 0 || fg >= 0 || bg >= 0) {
+		int sep = 0;
+
+		*dst++ = '\033';
+		*dst++ = '[';
+		if (attr >= 0) {
+			*dst++ = '0' + attr;
+			sep++;
+		}
+		if (fg >= 0) {
+			if (sep++)
+				*dst++ = ';';
+			if (fg < 8) {
+				*dst++ = '3';
+				*dst++ = '0' + fg;
+			} else {
+				dst += sprintf(dst, "38;5;%d", fg);
+			}
+		}
+		if (bg >= 0) {
+			if (sep++)
+				*dst++ = ';';
+			if (bg < 8) {
+				*dst++ = '4';
+				*dst++ = '0' + bg;
+			} else {
+				dst += sprintf(dst, "48;5;%d", bg);
+			}
+		}
+		*dst++ = 'm';
+	}
+	*dst = 0;
+	return;
+bad:
+	die("bad color value '%.*s' for variable '%s'", value_len, value, var);
+}
+
+int perf_config_colorbool(const char *var, const char *value, int stdout_is_tty)
+{
+	if (value) {
+		if (!strcasecmp(value, "never"))
+			return 0;
+		if (!strcasecmp(value, "always"))
+			return 1;
+		if (!strcasecmp(value, "auto"))
+			goto auto_color;
+	}
+
+	/* Missing or explicit false to turn off colorization */
+	if (!perf_config_bool(var, value))
+		return 0;
+
+	/* any normal truth value defaults to 'auto' */
+ auto_color:
+	if (stdout_is_tty < 0)
+		stdout_is_tty = isatty(1);
+	if (stdout_is_tty || (pager_in_use() && pager_use_color)) {
+		char *term = getenv("TERM");
+		if (term && strcmp(term, "dumb"))
+			return 1;
+	}
+	return 0;
+}
+
+int perf_color_default_config(const char *var, const char *value, void *cb)
+{
+	if (!strcmp(var, "color.ui")) {
+		perf_use_color_default = perf_config_colorbool(var, value, -1);
+		return 0;
+	}
+
+	return perf_default_config(var, value, cb);
+}
+
+static int __color_vsnprintf(char *bf, size_t size, const char *color,
+			     const char *fmt, va_list args, const char *trail)
+{
+	int r = 0;
+
+	/*
+	 * Auto-detect:
+	 */
+	if (perf_use_color_default < 0) {
+		if (isatty(1) || pager_in_use())
+			perf_use_color_default = 1;
+		else
+			perf_use_color_default = 0;
+	}
+
+	if (perf_use_color_default && *color)
+		r += snprintf(bf, size, "%s", color);
+	r += vsnprintf(bf + r, size - r, fmt, args);
+	if (perf_use_color_default && *color)
+		r += snprintf(bf + r, size - r, "%s", PERF_COLOR_RESET);
+	if (trail)
+		r += snprintf(bf + r, size - r, "%s", trail);
+	return r;
+}
+
+static int __color_vfprintf(FILE *fp, const char *color, const char *fmt,
+		va_list args, const char *trail)
+{
+	int r = 0;
+
+	/*
+	 * Auto-detect:
+	 */
+	if (perf_use_color_default < 0) {
+		if (isatty(1) || pager_in_use())
+			perf_use_color_default = 1;
+		else
+			perf_use_color_default = 0;
+	}
+
+	if (perf_use_color_default && *color)
+		r += fprintf(fp, "%s", color);
+	r += vfprintf(fp, fmt, args);
+	if (perf_use_color_default && *color)
+		r += fprintf(fp, "%s", PERF_COLOR_RESET);
+	if (trail)
+		r += fprintf(fp, "%s", trail);
+	return r;
+}
+
+int color_vsnprintf(char *bf, size_t size, const char *color,
+		    const char *fmt, va_list args)
+{
+	return __color_vsnprintf(bf, size, color, fmt, args, NULL);
+}
+
+int color_vfprintf(FILE *fp, const char *color, const char *fmt, va_list args)
+{
+	return __color_vfprintf(fp, color, fmt, args, NULL);
+}
+
+int color_snprintf(char *bf, size_t size, const char *color,
+		   const char *fmt, ...)
+{
+	va_list args;
+	int r;
+
+	va_start(args, fmt);
+	r = color_vsnprintf(bf, size, color, fmt, args);
+	va_end(args);
+	return r;
+}
+
+int color_fprintf(FILE *fp, const char *color, const char *fmt, ...)
+{
+	va_list args;
+	int r;
+
+	va_start(args, fmt);
+	r = color_vfprintf(fp, color, fmt, args);
+	va_end(args);
+	return r;
+}
+
+int color_fprintf_ln(FILE *fp, const char *color, const char *fmt, ...)
+{
+	va_list args;
+	int r;
+	va_start(args, fmt);
+	r = __color_vfprintf(fp, color, fmt, args, "\n");
+	va_end(args);
+	return r;
+}
+
+/*
+ * This function splits the buffer by newlines and colors the lines individually.
+ *
+ * Returns 0 on success.
+ */
+int color_fwrite_lines(FILE *fp, const char *color,
+		size_t count, const char *buf)
+{
+	if (!*color)
+		return fwrite(buf, count, 1, fp) != 1;
+
+	while (count) {
+		char *p = memchr(buf, '\n', count);
+
+		if (p != buf && (fputs(color, fp) < 0 ||
+				fwrite(buf, p ? (size_t)(p - buf) : count, 1, fp) != 1 ||
+				fputs(PERF_COLOR_RESET, fp) < 0))
+			return -1;
+		if (!p)
+			return 0;
+		if (fputc('\n', fp) < 0)
+			return -1;
+		count -= p + 1 - buf;
+		buf = p + 1;
+	}
+	return 0;
+}
+
+const char *get_percent_color(double percent)
+{
+	const 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 >= MIN_RED)
+		color = PERF_COLOR_RED;
+	else {
+		if (percent > MIN_GREEN)
+			color = PERF_COLOR_GREEN;
+	}
+	return color;
+}
+
+int percent_color_fprintf(FILE *fp, const char *fmt, double percent)
+{
+	int r;
+	const char *color;
+
+	color = get_percent_color(percent);
+	r = color_fprintf(fp, color, fmt, percent);
+
+	return r;
+}
+
+int percent_color_snprintf(char *bf, size_t size, const char *fmt, double percent)
+{
+	const char *color = get_percent_color(percent);
+	return color_snprintf(bf, size, color, fmt, percent);
+}
diff --git a/tools/perf/util/color.h b/tools/perf/util/color.h
new file mode 100644
index 00000000..dea082b7
--- /dev/null
+++ b/tools/perf/util/color.h
@@ -0,0 +1,46 @@
+#ifndef __PERF_COLOR_H
+#define __PERF_COLOR_H
+
+/* "\033[1;38;5;2xx;48;5;2xxm\0" is 23 bytes */
+#define COLOR_MAXLEN 24
+
+#define PERF_COLOR_NORMAL	""
+#define PERF_COLOR_RESET	"\033[m"
+#define PERF_COLOR_BOLD		"\033[1m"
+#define PERF_COLOR_RED		"\033[31m"
+#define PERF_COLOR_GREEN	"\033[32m"
+#define PERF_COLOR_YELLOW	"\033[33m"
+#define PERF_COLOR_BLUE		"\033[34m"
+#define PERF_COLOR_MAGENTA	"\033[35m"
+#define PERF_COLOR_CYAN		"\033[36m"
+#define PERF_COLOR_BG_RED	"\033[41m"
+
+#define MIN_GREEN	0.5
+#define MIN_RED		5.0
+
+/*
+ * This variable stores the value of color.ui
+ */
+extern int perf_use_color_default;
+
+
+/*
+ * Use this instead of perf_default_config if you need the value of color.ui.
+ */
+int perf_color_default_config(const char *var, const char *value, void *cb);
+
+int perf_config_colorbool(const char *var, const char *value, int stdout_is_tty);
+void color_parse(const char *value, const char *var, char *dst);
+void color_parse_mem(const char *value, int len, const char *var, char *dst);
+int color_vsnprintf(char *bf, size_t size, const char *color,
+		    const char *fmt, va_list args);
+int color_vfprintf(FILE *fp, const char *color, const char *fmt, va_list args);
+int color_fprintf(FILE *fp, const char *color, const char *fmt, ...);
+int color_snprintf(char *bf, size_t size, const char *color, const char *fmt, ...);
+int color_fprintf_ln(FILE *fp, const char *color, const char *fmt, ...);
+int color_fwrite_lines(FILE *fp, const char *color, size_t count, const char *buf);
+int percent_color_snprintf(char *bf, size_t size, const char *fmt, double percent);
+int percent_color_fprintf(FILE *fp, const char *fmt, double percent);
+const char *get_percent_color(double percent);
+
+#endif /* __PERF_COLOR_H */
diff --git a/tools/perf/util/config.c b/tools/perf/util/config.c
new file mode 100644
index 00000000..6c86eca8
--- /dev/null
+++ b/tools/perf/util/config.c
@@ -0,0 +1,485 @@
+/*
+ * GIT - The information manager from hell
+ *
+ * Copyright (C) Linus Torvalds, 2005
+ * Copyright (C) Johannes Schindelin, 2005
+ *
+ */
+#include "util.h"
+#include "cache.h"
+#include "exec_cmd.h"
+
+#define MAXNAME (256)
+
+#define DEBUG_CACHE_DIR ".debug"
+
+
+char buildid_dir[MAXPATHLEN]; /* root dir for buildid, binary cache */
+
+static FILE *config_file;
+static const char *config_file_name;
+static int config_linenr;
+static int config_file_eof;
+
+static const char *config_exclusive_filename;
+
+static int get_next_char(void)
+{
+	int c;
+	FILE *f;
+
+	c = '\n';
+	if ((f = config_file) != NULL) {
+		c = fgetc(f);
+		if (c == '\r') {
+			/* DOS like systems */
+			c = fgetc(f);
+			if (c != '\n') {
+				ungetc(c, f);
+				c = '\r';
+			}
+		}
+		if (c == '\n')
+			config_linenr++;
+		if (c == EOF) {
+			config_file_eof = 1;
+			c = '\n';
+		}
+	}
+	return c;
+}
+
+static char *parse_value(void)
+{
+	static char value[1024];
+	int quote = 0, comment = 0, space = 0;
+	size_t len = 0;
+
+	for (;;) {
+		int c = get_next_char();
+
+		if (len >= sizeof(value) - 1)
+			return NULL;
+		if (c == '\n') {
+			if (quote)
+				return NULL;
+			value[len] = 0;
+			return value;
+		}
+		if (comment)
+			continue;
+		if (isspace(c) && !quote) {
+			space = 1;
+			continue;
+		}
+		if (!quote) {
+			if (c == ';' || c == '#') {
+				comment = 1;
+				continue;
+			}
+		}
+		if (space) {
+			if (len)
+				value[len++] = ' ';
+			space = 0;
+		}
+		if (c == '\\') {
+			c = get_next_char();
+			switch (c) {
+			case '\n':
+				continue;
+			case 't':
+				c = '\t';
+				break;
+			case 'b':
+				c = '\b';
+				break;
+			case 'n':
+				c = '\n';
+				break;
+			/* Some characters escape as themselves */
+			case '\\': case '"':
+				break;
+			/* Reject unknown escape sequences */
+			default:
+				return NULL;
+			}
+			value[len++] = c;
+			continue;
+		}
+		if (c == '"') {
+			quote = 1-quote;
+			continue;
+		}
+		value[len++] = c;
+	}
+}
+
+static inline int iskeychar(int c)
+{
+	return isalnum(c) || c == '-';
+}
+
+static int get_value(config_fn_t fn, void *data, char *name, unsigned int len)
+{
+	int c;
+	char *value;
+
+	/* Get the full name */
+	for (;;) {
+		c = get_next_char();
+		if (config_file_eof)
+			break;
+		if (!iskeychar(c))
+			break;
+		name[len++] = c;
+		if (len >= MAXNAME)
+			return -1;
+	}
+	name[len] = 0;
+	while (c == ' ' || c == '\t')
+		c = get_next_char();
+
+	value = NULL;
+	if (c != '\n') {
+		if (c != '=')
+			return -1;
+		value = parse_value();
+		if (!value)
+			return -1;
+	}
+	return fn(name, value, data);
+}
+
+static int get_extended_base_var(char *name, int baselen, int c)
+{
+	do {
+		if (c == '\n')
+			return -1;
+		c = get_next_char();
+	} while (isspace(c));
+
+	/* We require the format to be '[base "extension"]' */
+	if (c != '"')
+		return -1;
+	name[baselen++] = '.';
+
+	for (;;) {
+		int ch = get_next_char();
+
+		if (ch == '\n')
+			return -1;
+		if (ch == '"')
+			break;
+		if (ch == '\\') {
+			ch = get_next_char();
+			if (ch == '\n')
+				return -1;
+		}
+		name[baselen++] = ch;
+		if (baselen > MAXNAME / 2)
+			return -1;
+	}
+
+	/* Final ']' */
+	if (get_next_char() != ']')
+		return -1;
+	return baselen;
+}
+
+static int get_base_var(char *name)
+{
+	int baselen = 0;
+
+	for (;;) {
+		int c = get_next_char();
+		if (config_file_eof)
+			return -1;
+		if (c == ']')
+			return baselen;
+		if (isspace(c))
+			return get_extended_base_var(name, baselen, c);
+		if (!iskeychar(c) && c != '.')
+			return -1;
+		if (baselen > MAXNAME / 2)
+			return -1;
+		name[baselen++] = tolower(c);
+	}
+}
+
+static int perf_parse_file(config_fn_t fn, void *data)
+{
+	int comment = 0;
+	int baselen = 0;
+	static char var[MAXNAME];
+
+	/* U+FEFF Byte Order Mark in UTF8 */
+	static const unsigned char *utf8_bom = (unsigned char *) "\xef\xbb\xbf";
+	const unsigned char *bomptr = utf8_bom;
+
+	for (;;) {
+		int c = get_next_char();
+		if (bomptr && *bomptr) {
+			/* We are at the file beginning; skip UTF8-encoded BOM
+			 * if present. Sane editors won't put this in on their
+			 * own, but e.g. Windows Notepad will do it happily. */
+			if ((unsigned char) c == *bomptr) {
+				bomptr++;
+				continue;
+			} else {
+				/* Do not tolerate partial BOM. */
+				if (bomptr != utf8_bom)
+					break;
+				/* No BOM at file beginning. Cool. */
+				bomptr = NULL;
+			}
+		}
+		if (c == '\n') {
+			if (config_file_eof)
+				return 0;
+			comment = 0;
+			continue;
+		}
+		if (comment || isspace(c))
+			continue;
+		if (c == '#' || c == ';') {
+			comment = 1;
+			continue;
+		}
+		if (c == '[') {
+			baselen = get_base_var(var);
+			if (baselen <= 0)
+				break;
+			var[baselen++] = '.';
+			var[baselen] = 0;
+			continue;
+		}
+		if (!isalpha(c))
+			break;
+		var[baselen] = tolower(c);
+		if (get_value(fn, data, var, baselen+1) < 0)
+			break;
+	}
+	die("bad config file line %d in %s", config_linenr, config_file_name);
+}
+
+static int parse_unit_factor(const char *end, unsigned long *val)
+{
+	if (!*end)
+		return 1;
+	else if (!strcasecmp(end, "k")) {
+		*val *= 1024;
+		return 1;
+	}
+	else if (!strcasecmp(end, "m")) {
+		*val *= 1024 * 1024;
+		return 1;
+	}
+	else if (!strcasecmp(end, "g")) {
+		*val *= 1024 * 1024 * 1024;
+		return 1;
+	}
+	return 0;
+}
+
+static int perf_parse_long(const char *value, long *ret)
+{
+	if (value && *value) {
+		char *end;
+		long val = strtol(value, &end, 0);
+		unsigned long factor = 1;
+		if (!parse_unit_factor(end, &factor))
+			return 0;
+		*ret = val * factor;
+		return 1;
+	}
+	return 0;
+}
+
+static void die_bad_config(const char *name)
+{
+	if (config_file_name)
+		die("bad config value for '%s' in %s", name, config_file_name);
+	die("bad config value for '%s'", name);
+}
+
+int perf_config_int(const char *name, const char *value)
+{
+	long ret = 0;
+	if (!perf_parse_long(value, &ret))
+		die_bad_config(name);
+	return ret;
+}
+
+static int perf_config_bool_or_int(const char *name, const char *value, int *is_bool)
+{
+	*is_bool = 1;
+	if (!value)
+		return 1;
+	if (!*value)
+		return 0;
+	if (!strcasecmp(value, "true") || !strcasecmp(value, "yes") || !strcasecmp(value, "on"))
+		return 1;
+	if (!strcasecmp(value, "false") || !strcasecmp(value, "no") || !strcasecmp(value, "off"))
+		return 0;
+	*is_bool = 0;
+	return perf_config_int(name, value);
+}
+
+int perf_config_bool(const char *name, const char *value)
+{
+	int discard;
+	return !!perf_config_bool_or_int(name, value, &discard);
+}
+
+const char *perf_config_dirname(const char *name, const char *value)
+{
+	if (!name)
+		return NULL;
+	return value;
+}
+
+static int perf_default_core_config(const char *var __used, const char *value __used)
+{
+	/* Add other config variables here and to Documentation/config.txt. */
+	return 0;
+}
+
+int perf_default_config(const char *var, const char *value, void *dummy __used)
+{
+	if (!prefixcmp(var, "core."))
+		return perf_default_core_config(var, value);
+
+	/* Add other config variables here and to Documentation/config.txt. */
+	return 0;
+}
+
+static int perf_config_from_file(config_fn_t fn, const char *filename, void *data)
+{
+	int ret;
+	FILE *f = fopen(filename, "r");
+
+	ret = -1;
+	if (f) {
+		config_file = f;
+		config_file_name = filename;
+		config_linenr = 1;
+		config_file_eof = 0;
+		ret = perf_parse_file(fn, data);
+		fclose(f);
+		config_file_name = NULL;
+	}
+	return ret;
+}
+
+static const char *perf_etc_perfconfig(void)
+{
+	static const char *system_wide;
+	if (!system_wide)
+		system_wide = system_path(ETC_PERFCONFIG);
+	return system_wide;
+}
+
+static int perf_env_bool(const char *k, int def)
+{
+	const char *v = getenv(k);
+	return v ? perf_config_bool(k, v) : def;
+}
+
+static int perf_config_system(void)
+{
+	return !perf_env_bool("PERF_CONFIG_NOSYSTEM", 0);
+}
+
+static int perf_config_global(void)
+{
+	return !perf_env_bool("PERF_CONFIG_NOGLOBAL", 0);
+}
+
+int perf_config(config_fn_t fn, void *data)
+{
+	int ret = 0, found = 0;
+	const char *home = NULL;
+
+	/* Setting $PERF_CONFIG makes perf read _only_ the given config file. */
+	if (config_exclusive_filename)
+		return perf_config_from_file(fn, config_exclusive_filename, data);
+	if (perf_config_system() && !access(perf_etc_perfconfig(), R_OK)) {
+		ret += perf_config_from_file(fn, perf_etc_perfconfig(),
+					    data);
+		found += 1;
+	}
+
+	home = getenv("HOME");
+	if (perf_config_global() && home) {
+		char *user_config = strdup(mkpath("%s/.perfconfig", home));
+		if (!access(user_config, R_OK)) {
+			ret += perf_config_from_file(fn, user_config, data);
+			found += 1;
+		}
+		free(user_config);
+	}
+
+	if (found == 0)
+		return -1;
+	return ret;
+}
+
+/*
+ * Call this to report error for your variable that should not
+ * get a boolean value (i.e. "[my] var" means "true").
+ */
+int config_error_nonbool(const char *var)
+{
+	return error("Missing value for '%s'", var);
+}
+
+struct buildid_dir_config {
+	char *dir;
+};
+
+static int buildid_dir_command_config(const char *var, const char *value,
+				      void *data)
+{
+	struct buildid_dir_config *c = data;
+	const char *v;
+
+	/* same dir for all commands */
+	if (!prefixcmp(var, "buildid.") && !strcmp(var + 8, "dir")) {
+		v = perf_config_dirname(var, value);
+		if (!v)
+			return -1;
+		strncpy(c->dir, v, MAXPATHLEN-1);
+		c->dir[MAXPATHLEN-1] = '\0';
+	}
+	return 0;
+}
+
+static void check_buildid_dir_config(void)
+{
+	struct buildid_dir_config c;
+	c.dir = buildid_dir;
+	perf_config(buildid_dir_command_config, &c);
+}
+
+void set_buildid_dir(void)
+{
+	buildid_dir[0] = '\0';
+
+	/* try config file */
+	check_buildid_dir_config();
+
+	/* default to $HOME/.debug */
+	if (buildid_dir[0] == '\0') {
+		char *v = getenv("HOME");
+		if (v) {
+			snprintf(buildid_dir, MAXPATHLEN-1, "%s/%s",
+				 v, DEBUG_CACHE_DIR);
+		} else {
+			strncpy(buildid_dir, DEBUG_CACHE_DIR, MAXPATHLEN-1);
+		}
+		buildid_dir[MAXPATHLEN-1] = '\0';
+	}
+	/* for communicating with external commands */
+	setenv("PERF_BUILDID_DIR", buildid_dir, 1);
+}
diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c
new file mode 100644
index 00000000..6893eec6
--- /dev/null
+++ b/tools/perf/util/cpumap.c
@@ -0,0 +1,184 @@
+#include "util.h"
+#include "../perf.h"
+#include "cpumap.h"
+#include <assert.h>
+#include <stdio.h>
+
+static struct cpu_map *cpu_map__default_new(void)
+{
+	struct cpu_map *cpus;
+	int nr_cpus;
+
+	nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
+	if (nr_cpus < 0)
+		return NULL;
+
+	cpus = malloc(sizeof(*cpus) + nr_cpus * sizeof(int));
+	if (cpus != NULL) {
+		int i;
+		for (i = 0; i < nr_cpus; ++i)
+			cpus->map[i] = i;
+
+		cpus->nr = nr_cpus;
+	}
+
+	return cpus;
+}
+
+static struct cpu_map *cpu_map__trim_new(int nr_cpus, int *tmp_cpus)
+{
+	size_t payload_size = nr_cpus * sizeof(int);
+	struct cpu_map *cpus = malloc(sizeof(*cpus) + payload_size);
+
+	if (cpus != NULL) {
+		cpus->nr = nr_cpus;
+		memcpy(cpus->map, tmp_cpus, payload_size);
+	}
+
+	return cpus;
+}
+
+static struct cpu_map *cpu_map__read_all_cpu_map(void)
+{
+	struct cpu_map *cpus = NULL;
+	FILE *onlnf;
+	int nr_cpus = 0;
+	int *tmp_cpus = NULL, *tmp;
+	int max_entries = 0;
+	int n, cpu, prev;
+	char sep;
+
+	onlnf = fopen("/sys/devices/system/cpu/online", "r");
+	if (!onlnf)
+		return cpu_map__default_new();
+
+	sep = 0;
+	prev = -1;
+	for (;;) {
+		n = fscanf(onlnf, "%u%c", &cpu, &sep);
+		if (n <= 0)
+			break;
+		if (prev >= 0) {
+			int new_max = nr_cpus + cpu - prev - 1;
+
+			if (new_max >= max_entries) {
+				max_entries = new_max + MAX_NR_CPUS / 2;
+				tmp = realloc(tmp_cpus, max_entries * sizeof(int));
+				if (tmp == NULL)
+					goto out_free_tmp;
+				tmp_cpus = tmp;
+			}
+
+			while (++prev < cpu)
+				tmp_cpus[nr_cpus++] = prev;
+		}
+		if (nr_cpus == max_entries) {
+			max_entries += MAX_NR_CPUS;
+			tmp = realloc(tmp_cpus, max_entries * sizeof(int));
+			if (tmp == NULL)
+				goto out_free_tmp;
+			tmp_cpus = tmp;
+		}
+
+		tmp_cpus[nr_cpus++] = cpu;
+		if (n == 2 && sep == '-')
+			prev = cpu;
+		else
+			prev = -1;
+		if (n == 1 || sep == '\n')
+			break;
+	}
+
+	if (nr_cpus > 0)
+		cpus = cpu_map__trim_new(nr_cpus, tmp_cpus);
+	else
+		cpus = cpu_map__default_new();
+out_free_tmp:
+	free(tmp_cpus);
+	fclose(onlnf);
+	return cpus;
+}
+
+struct cpu_map *cpu_map__new(const char *cpu_list)
+{
+	struct cpu_map *cpus = NULL;
+	unsigned long start_cpu, end_cpu = 0;
+	char *p = NULL;
+	int i, nr_cpus = 0;
+	int *tmp_cpus = NULL, *tmp;
+	int max_entries = 0;
+
+	if (!cpu_list)
+		return cpu_map__read_all_cpu_map();
+
+	if (!isdigit(*cpu_list))
+		goto out;
+
+	while (isdigit(*cpu_list)) {
+		p = NULL;
+		start_cpu = strtoul(cpu_list, &p, 0);
+		if (start_cpu >= INT_MAX
+		    || (*p != '\0' && *p != ',' && *p != '-'))
+			goto invalid;
+
+		if (*p == '-') {
+			cpu_list = ++p;
+			p = NULL;
+			end_cpu = strtoul(cpu_list, &p, 0);
+
+			if (end_cpu >= INT_MAX || (*p != '\0' && *p != ','))
+				goto invalid;
+
+			if (end_cpu < start_cpu)
+				goto invalid;
+		} else {
+			end_cpu = start_cpu;
+		}
+
+		for (; start_cpu <= end_cpu; start_cpu++) {
+			/* check for duplicates */
+			for (i = 0; i < nr_cpus; i++)
+				if (tmp_cpus[i] == (int)start_cpu)
+					goto invalid;
+
+			if (nr_cpus == max_entries) {
+				max_entries += MAX_NR_CPUS;
+				tmp = realloc(tmp_cpus, max_entries * sizeof(int));
+				if (tmp == NULL)
+					goto invalid;
+				tmp_cpus = tmp;
+			}
+			tmp_cpus[nr_cpus++] = (int)start_cpu;
+		}
+		if (*p)
+			++p;
+
+		cpu_list = p;
+	}
+
+	if (nr_cpus > 0)
+		cpus = cpu_map__trim_new(nr_cpus, tmp_cpus);
+	else
+		cpus = cpu_map__default_new();
+invalid:
+	free(tmp_cpus);
+out:
+	return cpus;
+}
+
+struct cpu_map *cpu_map__dummy_new(void)
+{
+	struct cpu_map *cpus = malloc(sizeof(*cpus) + sizeof(int));
+
+	if (cpus != NULL) {
+		cpus->nr = 1;
+		cpus->map[0] = -1;
+	}
+
+	return cpus;
+}
+
+void cpu_map__delete(struct cpu_map *map)
+{
+	free(map);
+}
diff --git a/tools/perf/util/cpumap.h b/tools/perf/util/cpumap.h
new file mode 100644
index 00000000..072c0a37
--- /dev/null
+++ b/tools/perf/util/cpumap.h
@@ -0,0 +1,13 @@
+#ifndef __PERF_CPUMAP_H
+#define __PERF_CPUMAP_H
+
+struct cpu_map {
+	int nr;
+	int map[];
+};
+
+struct cpu_map *cpu_map__new(const char *cpu_list);
+struct cpu_map *cpu_map__dummy_new(void);
+void cpu_map__delete(struct cpu_map *map);
+
+#endif /* __PERF_CPUMAP_H */
diff --git a/tools/perf/util/ctype.c b/tools/perf/util/ctype.c
new file mode 100644
index 00000000..35073621
--- /dev/null
+++ b/tools/perf/util/ctype.c
@@ -0,0 +1,39 @@
+/*
+ * Sane locale-independent, ASCII ctype.
+ *
+ * No surprises, and works with signed and unsigned chars.
+ */
+#include "cache.h"
+
+enum {
+	S = GIT_SPACE,
+	A = GIT_ALPHA,
+	D = GIT_DIGIT,
+	G = GIT_GLOB_SPECIAL,	/* *, ?, [, \\ */
+	R = GIT_REGEX_SPECIAL,	/* $, (, ), +, ., ^, {, | * */
+	P = GIT_PRINT_EXTRA,	/* printable - alpha - digit - glob - regex */
+
+	PS = GIT_SPACE | GIT_PRINT_EXTRA,
+};
+
+unsigned char sane_ctype[256] = {
+/*	0  1  2  3  4  5  6  7  8  9  A  B  C  D  E  F			    */
+
+	0, 0, 0, 0, 0, 0, 0, 0, 0, S, S, 0, 0, S, 0, 0,		/*   0.. 15 */
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,		/*  16.. 31 */
+	PS,P, P, P, R, P, P, P, R, R, G, R, P, P, R, P,		/*  32.. 47 */
+	D, D, D, D, D, D, D, D, D, D, P, P, P, P, P, G,		/*  48.. 63 */
+	P, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,		/*  64.. 79 */
+	A, A, A, A, A, A, A, A, A, A, A, G, G, P, R, P,		/*  80.. 95 */
+	P, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,		/*  96..111 */
+	A, A, A, A, A, A, A, A, A, A, A, R, R, P, P, 0,		/* 112..127 */
+	/* Nothing in the 128.. range */
+};
+
+const char *graph_line =
+	"_____________________________________________________________________"
+	"_____________________________________________________________________";
+const char *graph_dotted_line =
+	"---------------------------------------------------------------------"
+	"---------------------------------------------------------------------"
+	"---------------------------------------------------------------------";
diff --git a/tools/perf/util/debug.c b/tools/perf/util/debug.c
new file mode 100644
index 00000000..155749d7
--- /dev/null
+++ b/tools/perf/util/debug.c
@@ -0,0 +1,104 @@
+/* For general debugging purposes */
+
+#include "../perf.h"
+
+#include <string.h>
+#include <stdarg.h>
+#include <stdio.h>
+
+#include "cache.h"
+#include "color.h"
+#include "event.h"
+#include "debug.h"
+#include "util.h"
+
+int verbose;
+bool dump_trace = false, quiet = false;
+
+int eprintf(int level, const char *fmt, ...)
+{
+	va_list args;
+	int ret = 0;
+
+	if (verbose >= level) {
+		va_start(args, fmt);
+		if (use_browser > 0)
+			ret = ui_helpline__show_help(fmt, args);
+		else
+			ret = vfprintf(stderr, fmt, args);
+		va_end(args);
+	}
+
+	return ret;
+}
+
+int dump_printf(const char *fmt, ...)
+{
+	va_list args;
+	int ret = 0;
+
+	if (dump_trace) {
+		va_start(args, fmt);
+		ret = vprintf(fmt, args);
+		va_end(args);
+	}
+
+	return ret;
+}
+
+#ifdef NO_NEWT_SUPPORT
+void ui__warning(const char *format, ...)
+{
+	va_list args;
+
+	va_start(args, format);
+	vfprintf(stderr, format, args);
+	va_end(args);
+}
+#endif
+
+void ui__warning_paranoid(void)
+{
+	ui__warning("Permission error - are you root?\n"
+		    "Consider tweaking /proc/sys/kernel/perf_event_paranoid:\n"
+		    " -1 - Not paranoid at all\n"
+		    "  0 - Disallow raw tracepoint access for unpriv\n"
+		    "  1 - Disallow cpu events for unpriv\n"
+		    "  2 - Disallow kernel profiling for unpriv\n");
+}
+
+void trace_event(union perf_event *event)
+{
+	unsigned char *raw_event = (void *)event;
+	const char *color = PERF_COLOR_BLUE;
+	int i, j;
+
+	if (!dump_trace)
+		return;
+
+	printf(".");
+	color_fprintf(stdout, color, "\n. ... raw event: size %d bytes\n",
+		      event->header.size);
+
+	for (i = 0; i < event->header.size; i++) {
+		if ((i & 15) == 0) {
+			printf(".");
+			color_fprintf(stdout, color, "  %04x: ", i);
+		}
+
+		color_fprintf(stdout, color, " %02x", raw_event[i]);
+
+		if (((i & 15) == 15) || i == event->header.size-1) {
+			color_fprintf(stdout, color, "  ");
+			for (j = 0; j < 15-(i & 15); j++)
+				color_fprintf(stdout, color, "   ");
+			for (j = i & ~15; j <= i; j++) {
+				color_fprintf(stdout, color, "%c",
+					      isprint(raw_event[j]) ?
+					      raw_event[j] : '.');
+			}
+			color_fprintf(stdout, color, "\n");
+		}
+	}
+	printf(".\n");
+}
diff --git a/tools/perf/util/debug.h b/tools/perf/util/debug.h
new file mode 100644
index 00000000..fd53db47
--- /dev/null
+++ b/tools/perf/util/debug.h
@@ -0,0 +1,41 @@
+/* For debugging general purposes */
+#ifndef __PERF_DEBUG_H
+#define __PERF_DEBUG_H
+
+#include <stdbool.h>
+#include "event.h"
+
+extern int verbose;
+extern bool quiet, dump_trace;
+
+int dump_printf(const char *fmt, ...) __attribute__((format(printf, 1, 2)));
+void trace_event(union perf_event *event);
+
+struct ui_progress;
+
+#ifdef NO_NEWT_SUPPORT
+static inline int ui_helpline__show_help(const char *format __used, va_list ap __used)
+{
+	return 0;
+}
+
+static inline struct ui_progress *ui_progress__new(const char *title __used,
+						   u64 total __used)
+{
+	return (struct ui_progress *)1;
+}
+
+static inline void ui_progress__update(struct ui_progress *self __used,
+				       u64 curr __used) {}
+
+static inline void ui_progress__delete(struct ui_progress *self __used) {}
+#else
+extern char ui_helpline__last_msg[];
+int ui_helpline__show_help(const char *format, va_list ap);
+#include "ui/progress.h"
+#endif
+
+void ui__warning(const char *format, ...) __attribute__((format(printf, 1, 2)));
+void ui__warning_paranoid(void);
+
+#endif	/* __PERF_DEBUG_H */
diff --git a/tools/perf/util/debugfs.c b/tools/perf/util/debugfs.c
new file mode 100644
index 00000000..a88fefc0
--- /dev/null
+++ b/tools/perf/util/debugfs.c
@@ -0,0 +1,240 @@
+#include "util.h"
+#include "debugfs.h"
+#include "cache.h"
+
+static int debugfs_premounted;
+static char debugfs_mountpoint[MAX_PATH+1];
+
+static const char *debugfs_known_mountpoints[] = {
+	"/sys/kernel/debug/",
+	"/debug/",
+	0,
+};
+
+/* use this to force a umount */
+void debugfs_force_cleanup(void)
+{
+	debugfs_find_mountpoint();
+	debugfs_premounted = 0;
+	debugfs_umount();
+}
+
+/* construct a full path to a debugfs element */
+int debugfs_make_path(const char *element, char *buffer, int size)
+{
+	int len;
+
+	if (strlen(debugfs_mountpoint) == 0) {
+		buffer[0] = '\0';
+		return -1;
+	}
+
+	len = strlen(debugfs_mountpoint) + strlen(element) + 1;
+	if (len >= size)
+		return len+1;
+
+	snprintf(buffer, size-1, "%s/%s", debugfs_mountpoint, element);
+	return 0;
+}
+
+static int debugfs_found;
+
+/* find the path to the mounted debugfs */
+const char *debugfs_find_mountpoint(void)
+{
+	const char **ptr;
+	char type[100];
+	FILE *fp;
+
+	if (debugfs_found)
+		return (const char *) debugfs_mountpoint;
+
+	ptr = debugfs_known_mountpoints;
+	while (*ptr) {
+		if (debugfs_valid_mountpoint(*ptr) == 0) {
+			debugfs_found = 1;
+			strcpy(debugfs_mountpoint, *ptr);
+			return debugfs_mountpoint;
+		}
+		ptr++;
+	}
+
+	/* give up and parse /proc/mounts */
+	fp = fopen("/proc/mounts", "r");
+	if (fp == NULL)
+		die("Can't open /proc/mounts for read");
+
+	while (fscanf(fp, "%*s %"
+		      STR(MAX_PATH)
+		      "s %99s %*s %*d %*d\n",
+		      debugfs_mountpoint, type) == 2) {
+		if (strcmp(type, "debugfs") == 0)
+			break;
+	}
+	fclose(fp);
+
+	if (strcmp(type, "debugfs") != 0)
+		return NULL;
+
+	debugfs_found = 1;
+
+	return debugfs_mountpoint;
+}
+
+/* verify that a mountpoint is actually a debugfs instance */
+
+int debugfs_valid_mountpoint(const char *debugfs)
+{
+	struct statfs st_fs;
+
+	if (statfs(debugfs, &st_fs) < 0)
+		return -ENOENT;
+	else if (st_fs.f_type != (long) DEBUGFS_MAGIC)
+		return -ENOENT;
+
+	return 0;
+}
+
+
+int debugfs_valid_entry(const char *path)
+{
+	struct stat st;
+
+	if (stat(path, &st))
+		return -errno;
+
+	return 0;
+}
+
+/* mount the debugfs somewhere if it's not mounted */
+
+char *debugfs_mount(const char *mountpoint)
+{
+	/* see if it's already mounted */
+	if (debugfs_find_mountpoint()) {
+		debugfs_premounted = 1;
+		return debugfs_mountpoint;
+	}
+
+	/* if not mounted and no argument */
+	if (mountpoint == NULL) {
+		/* see if environment variable set */
+		mountpoint = getenv(PERF_DEBUGFS_ENVIRONMENT);
+		/* if no environment variable, use default */
+		if (mountpoint == NULL)
+			mountpoint = "/sys/kernel/debug";
+	}
+
+	if (mount(NULL, mountpoint, "debugfs", 0, NULL) < 0)
+		return NULL;
+
+	/* save the mountpoint */
+	strncpy(debugfs_mountpoint, mountpoint, sizeof(debugfs_mountpoint));
+	debugfs_found = 1;
+
+	return debugfs_mountpoint;
+}
+
+/* umount the debugfs */
+
+int debugfs_umount(void)
+{
+	char umountcmd[128];
+	int ret;
+
+	/* if it was already mounted, leave it */
+	if (debugfs_premounted)
+		return 0;
+
+	/* make sure it's a valid mount point */
+	ret = debugfs_valid_mountpoint(debugfs_mountpoint);
+	if (ret)
+		return ret;
+
+	snprintf(umountcmd, sizeof(umountcmd),
+		 "/bin/umount %s", debugfs_mountpoint);
+	return system(umountcmd);
+}
+
+int debugfs_write(const char *entry, const char *value)
+{
+	char path[MAX_PATH+1];
+	int ret, count;
+	int fd;
+
+	/* construct the path */
+	snprintf(path, sizeof(path), "%s/%s", debugfs_mountpoint, entry);
+
+	/* verify that it exists */
+	ret = debugfs_valid_entry(path);
+	if (ret)
+		return ret;
+
+	/* get how many chars we're going to write */
+	count = strlen(value);
+
+	/* open the debugfs entry */
+	fd = open(path, O_RDWR);
+	if (fd < 0)
+		return -errno;
+
+	while (count > 0) {
+		/* write it */
+		ret = write(fd, value, count);
+		if (ret <= 0) {
+			if (ret == EAGAIN)
+				continue;
+			close(fd);
+			return -errno;
+		}
+		count -= ret;
+	}
+
+	/* close it */
+	close(fd);
+
+	/* return success */
+	return 0;
+}
+
+/*
+ * read a debugfs entry
+ * returns the number of chars read or a negative errno
+ */
+int debugfs_read(const char *entry, char *buffer, size_t size)
+{
+	char path[MAX_PATH+1];
+	int ret;
+	int fd;
+
+	/* construct the path */
+	snprintf(path, sizeof(path), "%s/%s", debugfs_mountpoint, entry);
+
+	/* verify that it exists */
+	ret = debugfs_valid_entry(path);
+	if (ret)
+		return ret;
+
+	/* open the debugfs entry */
+	fd = open(path, O_RDONLY);
+	if (fd < 0)
+		return -errno;
+
+	do {
+		/* read it */
+		ret = read(fd, buffer, size);
+		if (ret == 0) {
+			close(fd);
+			return EOF;
+		}
+	} while (ret < 0 && errno == EAGAIN);
+
+	/* close it */
+	close(fd);
+
+	/* make *sure* there's a null character at the end */
+	buffer[ret] = '\0';
+
+	/* return the number of chars read */
+	return ret;
+}
diff --git a/tools/perf/util/debugfs.h b/tools/perf/util/debugfs.h
new file mode 100644
index 00000000..83a02879
--- /dev/null
+++ b/tools/perf/util/debugfs.h
@@ -0,0 +1,25 @@
+#ifndef __DEBUGFS_H__
+#define __DEBUGFS_H__
+
+#include <sys/mount.h>
+
+#ifndef MAX_PATH
+# define MAX_PATH 256
+#endif
+
+#ifndef STR
+# define _STR(x) #x
+# define STR(x) _STR(x)
+#endif
+
+extern const char *debugfs_find_mountpoint(void);
+extern int debugfs_valid_mountpoint(const char *debugfs);
+extern int debugfs_valid_entry(const char *path);
+extern char *debugfs_mount(const char *mountpoint);
+extern int debugfs_umount(void);
+extern int debugfs_write(const char *entry, const char *value);
+extern int debugfs_read(const char *entry, char *buffer, size_t size);
+extern void debugfs_force_cleanup(void);
+extern int debugfs_make_path(const char *element, char *buffer, int size);
+
+#endif /* __DEBUGFS_H__ */
diff --git a/tools/perf/util/environment.c b/tools/perf/util/environment.c
new file mode 100644
index 00000000..275b0ee3
--- /dev/null
+++ b/tools/perf/util/environment.c
@@ -0,0 +1,9 @@
+/*
+ * We put all the perf config variables in this same object
+ * file, so that programs can link against the config parser
+ * without having to link against all the rest of perf.
+ */
+#include "cache.h"
+
+const char *pager_program;
+int pager_use_color = 1;
diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c
new file mode 100644
index 00000000..3c1b8a63
--- /dev/null
+++ b/tools/perf/util/event.c
@@ -0,0 +1,802 @@
+#include <linux/types.h>
+#include "event.h"
+#include "debug.h"
+#include "session.h"
+#include "sort.h"
+#include "string.h"
+#include "strlist.h"
+#include "thread.h"
+#include "thread_map.h"
+
+static const char *perf_event__names[] = {
+	[0]					= "TOTAL",
+	[PERF_RECORD_MMAP]			= "MMAP",
+	[PERF_RECORD_LOST]			= "LOST",
+	[PERF_RECORD_COMM]			= "COMM",
+	[PERF_RECORD_EXIT]			= "EXIT",
+	[PERF_RECORD_THROTTLE]			= "THROTTLE",
+	[PERF_RECORD_UNTHROTTLE]		= "UNTHROTTLE",
+	[PERF_RECORD_FORK]			= "FORK",
+	[PERF_RECORD_READ]			= "READ",
+	[PERF_RECORD_SAMPLE]			= "SAMPLE",
+	[PERF_RECORD_HEADER_ATTR]		= "ATTR",
+	[PERF_RECORD_HEADER_EVENT_TYPE]		= "EVENT_TYPE",
+	[PERF_RECORD_HEADER_TRACING_DATA]	= "TRACING_DATA",
+	[PERF_RECORD_HEADER_BUILD_ID]		= "BUILD_ID",
+	[PERF_RECORD_FINISHED_ROUND]		= "FINISHED_ROUND",
+};
+
+const char *perf_event__name(unsigned int id)
+{
+	if (id >= ARRAY_SIZE(perf_event__names))
+		return "INVALID";
+	if (!perf_event__names[id])
+		return "UNKNOWN";
+	return perf_event__names[id];
+}
+
+static struct perf_sample synth_sample = {
+	.pid	   = -1,
+	.tid	   = -1,
+	.time	   = -1,
+	.stream_id = -1,
+	.cpu	   = -1,
+	.period	   = 1,
+};
+
+static pid_t perf_event__synthesize_comm(union perf_event *event, pid_t pid,
+					 int full, perf_event__handler_t process,
+					 struct perf_session *session)
+{
+	char filename[PATH_MAX];
+	char bf[BUFSIZ];
+	FILE *fp;
+	size_t size = 0;
+	DIR *tasks;
+	struct dirent dirent, *next;
+	pid_t tgid = 0;
+
+	snprintf(filename, sizeof(filename), "/proc/%d/status", pid);
+
+	fp = fopen(filename, "r");
+	if (fp == NULL) {
+out_race:
+		/*
+		 * We raced with a task exiting - just return:
+		 */
+		pr_debug("couldn't open %s\n", filename);
+		return 0;
+	}
+
+	memset(&event->comm, 0, sizeof(event->comm));
+
+	while (!event->comm.comm[0] || !event->comm.pid) {
+		if (fgets(bf, sizeof(bf), fp) == NULL) {
+			pr_warning("couldn't get COMM and pgid, malformed %s\n", filename);
+			goto out;
+		}
+
+		if (memcmp(bf, "Name:", 5) == 0) {
+			char *name = bf + 5;
+			while (*name && isspace(*name))
+				++name;
+			size = strlen(name) - 1;
+			memcpy(event->comm.comm, name, size++);
+		} else if (memcmp(bf, "Tgid:", 5) == 0) {
+			char *tgids = bf + 5;
+			while (*tgids && isspace(*tgids))
+				++tgids;
+			tgid = event->comm.pid = atoi(tgids);
+		}
+	}
+
+	event->comm.header.type = PERF_RECORD_COMM;
+	size = ALIGN(size, sizeof(u64));
+	memset(event->comm.comm + size, 0, session->id_hdr_size);
+	event->comm.header.size = (sizeof(event->comm) -
+				(sizeof(event->comm.comm) - size) +
+				session->id_hdr_size);
+	if (!full) {
+		event->comm.tid = pid;
+
+		process(event, &synth_sample, session);
+		goto out;
+	}
+
+	snprintf(filename, sizeof(filename), "/proc/%d/task", pid);
+
+	tasks = opendir(filename);
+	if (tasks == NULL)
+		goto out_race;
+
+	while (!readdir_r(tasks, &dirent, &next) && next) {
+		char *end;
+		pid = strtol(dirent.d_name, &end, 10);
+		if (*end)
+			continue;
+
+		event->comm.tid = pid;
+
+		process(event, &synth_sample, session);
+	}
+
+	closedir(tasks);
+out:
+	fclose(fp);
+
+	return tgid;
+}
+
+static int perf_event__synthesize_mmap_events(union perf_event *event,
+					      pid_t pid, pid_t tgid,
+					      perf_event__handler_t process,
+					      struct perf_session *session)
+{
+	char filename[PATH_MAX];
+	FILE *fp;
+
+	snprintf(filename, sizeof(filename), "/proc/%d/maps", pid);
+
+	fp = fopen(filename, "r");
+	if (fp == NULL) {
+		/*
+		 * We raced with a task exiting - just return:
+		 */
+		pr_debug("couldn't open %s\n", filename);
+		return -1;
+	}
+
+	event->header.type = PERF_RECORD_MMAP;
+	/*
+	 * Just like the kernel, see __perf_event_mmap in kernel/perf_event.c
+	 */
+	event->header.misc = PERF_RECORD_MISC_USER;
+
+	while (1) {
+		char bf[BUFSIZ], *pbf = bf;
+		int n;
+		size_t size;
+		if (fgets(bf, sizeof(bf), fp) == NULL)
+			break;
+
+		/* 00400000-0040c000 r-xp 00000000 fd:01 41038  /bin/cat */
+		n = hex2u64(pbf, &event->mmap.start);
+		if (n < 0)
+			continue;
+		pbf += n + 1;
+		n = hex2u64(pbf, &event->mmap.len);
+		if (n < 0)
+			continue;
+		pbf += n + 3;
+		if (*pbf == 'x') { /* vm_exec */
+			char *execname = strchr(bf, '/');
+
+			/* Catch VDSO */
+			if (execname == NULL)
+				execname = strstr(bf, "[vdso]");
+
+			if (execname == NULL)
+				continue;
+
+			pbf += 3;
+			n = hex2u64(pbf, &event->mmap.pgoff);
+
+			size = strlen(execname);
+			execname[size - 1] = '\0'; /* Remove \n */
+			memcpy(event->mmap.filename, execname, size);
+			size = ALIGN(size, sizeof(u64));
+			event->mmap.len -= event->mmap.start;
+			event->mmap.header.size = (sizeof(event->mmap) -
+					        (sizeof(event->mmap.filename) - size));
+			memset(event->mmap.filename + size, 0, session->id_hdr_size);
+			event->mmap.header.size += session->id_hdr_size;
+			event->mmap.pid = tgid;
+			event->mmap.tid = pid;
+
+			process(event, &synth_sample, session);
+		}
+	}
+
+	fclose(fp);
+	return 0;
+}
+
+int perf_event__synthesize_modules(perf_event__handler_t process,
+				   struct perf_session *session,
+				   struct machine *machine)
+{
+	struct rb_node *nd;
+	struct map_groups *kmaps = &machine->kmaps;
+	union perf_event *event = zalloc((sizeof(event->mmap) +
+					  session->id_hdr_size));
+	if (event == NULL) {
+		pr_debug("Not enough memory synthesizing mmap event "
+			 "for kernel modules\n");
+		return -1;
+	}
+
+	event->header.type = PERF_RECORD_MMAP;
+
+	/*
+	 * kernel uses 0 for user space maps, see kernel/perf_event.c
+	 * __perf_event_mmap
+	 */
+	if (machine__is_host(machine))
+		event->header.misc = PERF_RECORD_MISC_KERNEL;
+	else
+		event->header.misc = PERF_RECORD_MISC_GUEST_KERNEL;
+
+	for (nd = rb_first(&kmaps->maps[MAP__FUNCTION]);
+	     nd; nd = rb_next(nd)) {
+		size_t size;
+		struct map *pos = rb_entry(nd, struct map, rb_node);
+
+		if (pos->dso->kernel)
+			continue;
+
+		size = ALIGN(pos->dso->long_name_len + 1, sizeof(u64));
+		event->mmap.header.type = PERF_RECORD_MMAP;
+		event->mmap.header.size = (sizeof(event->mmap) -
+				        (sizeof(event->mmap.filename) - size));
+		memset(event->mmap.filename + size, 0, session->id_hdr_size);
+		event->mmap.header.size += session->id_hdr_size;
+		event->mmap.start = pos->start;
+		event->mmap.len   = pos->end - pos->start;
+		event->mmap.pid   = machine->pid;
+
+		memcpy(event->mmap.filename, pos->dso->long_name,
+		       pos->dso->long_name_len + 1);
+		process(event, &synth_sample, session);
+	}
+
+	free(event);
+	return 0;
+}
+
+static int __event__synthesize_thread(union perf_event *comm_event,
+				      union perf_event *mmap_event,
+				      pid_t pid, perf_event__handler_t process,
+				      struct perf_session *session)
+{
+	pid_t tgid = perf_event__synthesize_comm(comm_event, pid, 1, process,
+					    session);
+	if (tgid == -1)
+		return -1;
+	return perf_event__synthesize_mmap_events(mmap_event, pid, tgid,
+					     process, session);
+}
+
+int perf_event__synthesize_thread_map(struct thread_map *threads,
+				      perf_event__handler_t process,
+				      struct perf_session *session)
+{
+	union perf_event *comm_event, *mmap_event;
+	int err = -1, thread;
+
+	comm_event = malloc(sizeof(comm_event->comm) + session->id_hdr_size);
+	if (comm_event == NULL)
+		goto out;
+
+	mmap_event = malloc(sizeof(mmap_event->mmap) + session->id_hdr_size);
+	if (mmap_event == NULL)
+		goto out_free_comm;
+
+	err = 0;
+	for (thread = 0; thread < threads->nr; ++thread) {
+		if (__event__synthesize_thread(comm_event, mmap_event,
+					       threads->map[thread],
+					       process, session)) {
+			err = -1;
+			break;
+		}
+	}
+	free(mmap_event);
+out_free_comm:
+	free(comm_event);
+out:
+	return err;
+}
+
+int perf_event__synthesize_threads(perf_event__handler_t process,
+				   struct perf_session *session)
+{
+	DIR *proc;
+	struct dirent dirent, *next;
+	union perf_event *comm_event, *mmap_event;
+	int err = -1;
+
+	comm_event = malloc(sizeof(comm_event->comm) + session->id_hdr_size);
+	if (comm_event == NULL)
+		goto out;
+
+	mmap_event = malloc(sizeof(mmap_event->mmap) + session->id_hdr_size);
+	if (mmap_event == NULL)
+		goto out_free_comm;
+
+	proc = opendir("/proc");
+	if (proc == NULL)
+		goto out_free_mmap;
+
+	while (!readdir_r(proc, &dirent, &next) && next) {
+		char *end;
+		pid_t pid = strtol(dirent.d_name, &end, 10);
+
+		if (*end) /* only interested in proper numerical dirents */
+			continue;
+
+		__event__synthesize_thread(comm_event, mmap_event, pid,
+					   process, session);
+	}
+
+	closedir(proc);
+	err = 0;
+out_free_mmap:
+	free(mmap_event);
+out_free_comm:
+	free(comm_event);
+out:
+	return err;
+}
+
+struct process_symbol_args {
+	const char *name;
+	u64	   start;
+};
+
+static int find_symbol_cb(void *arg, const char *name, char type,
+			  u64 start, u64 end __used)
+{
+	struct process_symbol_args *args = arg;
+
+	/*
+	 * Must be a function or at least an alias, as in PARISC64, where "_text" is
+	 * an 'A' to the same address as "_stext".
+	 */
+	if (!(symbol_type__is_a(type, MAP__FUNCTION) ||
+	      type == 'A') || strcmp(name, args->name))
+		return 0;
+
+	args->start = start;
+	return 1;
+}
+
+int perf_event__synthesize_kernel_mmap(perf_event__handler_t process,
+				       struct perf_session *session,
+				       struct machine *machine,
+				       const char *symbol_name)
+{
+	size_t size;
+	const char *filename, *mmap_name;
+	char path[PATH_MAX];
+	char name_buff[PATH_MAX];
+	struct map *map;
+	int err;
+	/*
+	 * We should get this from /sys/kernel/sections/.text, but till that is
+	 * available use this, and after it is use this as a fallback for older
+	 * kernels.
+	 */
+	struct process_symbol_args args = { .name = symbol_name, };
+	union perf_event *event = zalloc((sizeof(event->mmap) +
+					  session->id_hdr_size));
+	if (event == NULL) {
+		pr_debug("Not enough memory synthesizing mmap event "
+			 "for kernel modules\n");
+		return -1;
+	}
+
+	mmap_name = machine__mmap_name(machine, name_buff, sizeof(name_buff));
+	if (machine__is_host(machine)) {
+		/*
+		 * kernel uses PERF_RECORD_MISC_USER for user space maps,
+		 * see kernel/perf_event.c __perf_event_mmap
+		 */
+		event->header.misc = PERF_RECORD_MISC_KERNEL;
+		filename = "/proc/kallsyms";
+	} else {
+		event->header.misc = PERF_RECORD_MISC_GUEST_KERNEL;
+		if (machine__is_default_guest(machine))
+			filename = (char *) symbol_conf.default_guest_kallsyms;
+		else {
+			sprintf(path, "%s/proc/kallsyms", machine->root_dir);
+			filename = path;
+		}
+	}
+
+	if (kallsyms__parse(filename, &args, find_symbol_cb) <= 0)
+		return -ENOENT;
+
+	map = machine->vmlinux_maps[MAP__FUNCTION];
+	size = snprintf(event->mmap.filename, sizeof(event->mmap.filename),
+			"%s%s", mmap_name, symbol_name) + 1;
+	size = ALIGN(size, sizeof(u64));
+	event->mmap.header.type = PERF_RECORD_MMAP;
+	event->mmap.header.size = (sizeof(event->mmap) -
+			(sizeof(event->mmap.filename) - size) + session->id_hdr_size);
+	event->mmap.pgoff = args.start;
+	event->mmap.start = map->start;
+	event->mmap.len   = map->end - event->mmap.start;
+	event->mmap.pid   = machine->pid;
+
+	err = process(event, &synth_sample, session);
+	free(event);
+
+	return err;
+}
+
+int perf_event__process_comm(union perf_event *event,
+			     struct perf_sample *sample __used,
+			     struct perf_session *session)
+{
+	struct thread *thread = perf_session__findnew(session, event->comm.tid);
+
+	dump_printf(": %s:%d\n", event->comm.comm, event->comm.tid);
+
+	if (thread == NULL || thread__set_comm(thread, event->comm.comm)) {
+		dump_printf("problem processing PERF_RECORD_COMM, skipping event.\n");
+		return -1;
+	}
+
+	return 0;
+}
+
+int perf_event__process_lost(union perf_event *event,
+			     struct perf_sample *sample __used,
+			     struct perf_session *session)
+{
+	dump_printf(": id:%" PRIu64 ": lost:%" PRIu64 "\n",
+		    event->lost.id, event->lost.lost);
+	session->hists.stats.total_lost += event->lost.lost;
+	return 0;
+}
+
+static void perf_event__set_kernel_mmap_len(union perf_event *event,
+					    struct map **maps)
+{
+	maps[MAP__FUNCTION]->start = event->mmap.start;
+	maps[MAP__FUNCTION]->end   = event->mmap.start + event->mmap.len;
+	/*
+	 * Be a bit paranoid here, some perf.data file came with
+	 * a zero sized synthesized MMAP event for the kernel.
+	 */
+	if (maps[MAP__FUNCTION]->end == 0)
+		maps[MAP__FUNCTION]->end = ~0ULL;
+}
+
+static int perf_event__process_kernel_mmap(union perf_event *event,
+					   struct perf_session *session)
+{
+	struct map *map;
+	char kmmap_prefix[PATH_MAX];
+	struct machine *machine;
+	enum dso_kernel_type kernel_type;
+	bool is_kernel_mmap;
+
+	machine = perf_session__findnew_machine(session, event->mmap.pid);
+	if (!machine) {
+		pr_err("Can't find id %d's machine\n", event->mmap.pid);
+		goto out_problem;
+	}
+
+	machine__mmap_name(machine, kmmap_prefix, sizeof(kmmap_prefix));
+	if (machine__is_host(machine))
+		kernel_type = DSO_TYPE_KERNEL;
+	else
+		kernel_type = DSO_TYPE_GUEST_KERNEL;
+
+	is_kernel_mmap = memcmp(event->mmap.filename,
+				kmmap_prefix,
+				strlen(kmmap_prefix)) == 0;
+	if (event->mmap.filename[0] == '/' ||
+	    (!is_kernel_mmap && event->mmap.filename[0] == '[')) {
+
+		char short_module_name[1024];
+		char *name, *dot;
+
+		if (event->mmap.filename[0] == '/') {
+			name = strrchr(event->mmap.filename, '/');
+			if (name == NULL)
+				goto out_problem;
+
+			++name; /* skip / */
+			dot = strrchr(name, '.');
+			if (dot == NULL)
+				goto out_problem;
+			snprintf(short_module_name, sizeof(short_module_name),
+					"[%.*s]", (int)(dot - name), name);
+			strxfrchar(short_module_name, '-', '_');
+		} else
+			strcpy(short_module_name, event->mmap.filename);
+
+		map = machine__new_module(machine, event->mmap.start,
+					  event->mmap.filename);
+		if (map == NULL)
+			goto out_problem;
+
+		name = strdup(short_module_name);
+		if (name == NULL)
+			goto out_problem;
+
+		map->dso->short_name = name;
+		map->dso->sname_alloc = 1;
+		map->end = map->start + event->mmap.len;
+	} else if (is_kernel_mmap) {
+		const char *symbol_name = (event->mmap.filename +
+				strlen(kmmap_prefix));
+		/*
+		 * Should be there already, from the build-id table in
+		 * the header.
+		 */
+		struct dso *kernel = __dsos__findnew(&machine->kernel_dsos,
+						     kmmap_prefix);
+		if (kernel == NULL)
+			goto out_problem;
+
+		kernel->kernel = kernel_type;
+		if (__machine__create_kernel_maps(machine, kernel) < 0)
+			goto out_problem;
+
+		perf_event__set_kernel_mmap_len(event, machine->vmlinux_maps);
+
+		/*
+		 * Avoid using a zero address (kptr_restrict) for the ref reloc
+		 * symbol. Effectively having zero here means that at record
+		 * time /proc/sys/kernel/kptr_restrict was non zero.
+		 */
+		if (event->mmap.pgoff != 0) {
+			perf_session__set_kallsyms_ref_reloc_sym(machine->vmlinux_maps,
+								 symbol_name,
+								 event->mmap.pgoff);
+		}
+
+		if (machine__is_default_guest(machine)) {
+			/*
+			 * preload dso of guest kernel and modules
+			 */
+			dso__load(kernel, machine->vmlinux_maps[MAP__FUNCTION],
+				  NULL);
+		}
+	}
+	return 0;
+out_problem:
+	return -1;
+}
+
+int perf_event__process_mmap(union perf_event *event,
+			     struct perf_sample *sample __used,
+			     struct perf_session *session)
+{
+	struct machine *machine;
+	struct thread *thread;
+	struct map *map;
+	u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+	int ret = 0;
+
+	dump_printf(" %d/%d: [%#" PRIx64 "(%#" PRIx64 ") @ %#" PRIx64 "]: %s\n",
+			event->mmap.pid, event->mmap.tid, event->mmap.start,
+			event->mmap.len, event->mmap.pgoff, event->mmap.filename);
+
+	if (cpumode == PERF_RECORD_MISC_GUEST_KERNEL ||
+	    cpumode == PERF_RECORD_MISC_KERNEL) {
+		ret = perf_event__process_kernel_mmap(event, session);
+		if (ret < 0)
+			goto out_problem;
+		return 0;
+	}
+
+	machine = perf_session__find_host_machine(session);
+	if (machine == NULL)
+		goto out_problem;
+	thread = perf_session__findnew(session, event->mmap.pid);
+	if (thread == NULL)
+		goto out_problem;
+	map = map__new(&machine->user_dsos, event->mmap.start,
+			event->mmap.len, event->mmap.pgoff,
+			event->mmap.pid, event->mmap.filename,
+			MAP__FUNCTION);
+	if (map == NULL)
+		goto out_problem;
+
+	thread__insert_map(thread, map);
+	return 0;
+
+out_problem:
+	dump_printf("problem processing PERF_RECORD_MMAP, skipping event.\n");
+	return 0;
+}
+
+int perf_event__process_task(union perf_event *event,
+			     struct perf_sample *sample __used,
+			     struct perf_session *session)
+{
+	struct thread *thread = perf_session__findnew(session, event->fork.tid);
+	struct thread *parent = perf_session__findnew(session, event->fork.ptid);
+
+	dump_printf("(%d:%d):(%d:%d)\n", event->fork.pid, event->fork.tid,
+		    event->fork.ppid, event->fork.ptid);
+
+	if (event->header.type == PERF_RECORD_EXIT) {
+		perf_session__remove_thread(session, thread);
+		return 0;
+	}
+
+	if (thread == NULL || parent == NULL ||
+	    thread__fork(thread, parent) < 0) {
+		dump_printf("problem processing PERF_RECORD_FORK, skipping event.\n");
+		return -1;
+	}
+
+	return 0;
+}
+
+int perf_event__process(union perf_event *event, struct perf_sample *sample,
+			struct perf_session *session)
+{
+	switch (event->header.type) {
+	case PERF_RECORD_COMM:
+		perf_event__process_comm(event, sample, session);
+		break;
+	case PERF_RECORD_MMAP:
+		perf_event__process_mmap(event, sample, session);
+		break;
+	case PERF_RECORD_FORK:
+	case PERF_RECORD_EXIT:
+		perf_event__process_task(event, sample, session);
+		break;
+	case PERF_RECORD_LOST:
+		perf_event__process_lost(event, sample, session);
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+void thread__find_addr_map(struct thread *self,
+			   struct perf_session *session, u8 cpumode,
+			   enum map_type type, pid_t pid, u64 addr,
+			   struct addr_location *al)
+{
+	struct map_groups *mg = &self->mg;
+	struct machine *machine = NULL;
+
+	al->thread = self;
+	al->addr = addr;
+	al->cpumode = cpumode;
+	al->filtered = false;
+
+	if (cpumode == PERF_RECORD_MISC_KERNEL && perf_host) {
+		al->level = 'k';
+		machine = perf_session__find_host_machine(session);
+		if (machine == NULL) {
+			al->map = NULL;
+			return;
+		}
+		mg = &machine->kmaps;
+	} else if (cpumode == PERF_RECORD_MISC_USER && perf_host) {
+		al->level = '.';
+		machine = perf_session__find_host_machine(session);
+	} else if (cpumode == PERF_RECORD_MISC_GUEST_KERNEL && perf_guest) {
+		al->level = 'g';
+		machine = perf_session__find_machine(session, pid);
+		if (machine == NULL) {
+			al->map = NULL;
+			return;
+		}
+		mg = &machine->kmaps;
+	} else {
+		/*
+		 * 'u' means guest os user space.
+		 * TODO: We don't support guest user space. Might support late.
+		 */
+		if (cpumode == PERF_RECORD_MISC_GUEST_USER && perf_guest)
+			al->level = 'u';
+		else
+			al->level = 'H';
+		al->map = NULL;
+
+		if ((cpumode == PERF_RECORD_MISC_GUEST_USER ||
+			cpumode == PERF_RECORD_MISC_GUEST_KERNEL) &&
+			!perf_guest)
+			al->filtered = true;
+		if ((cpumode == PERF_RECORD_MISC_USER ||
+			cpumode == PERF_RECORD_MISC_KERNEL) &&
+			!perf_host)
+			al->filtered = true;
+
+		return;
+	}
+try_again:
+	al->map = map_groups__find(mg, type, al->addr);
+	if (al->map == NULL) {
+		/*
+		 * If this is outside of all known maps, and is a negative
+		 * address, try to look it up in the kernel dso, as it might be
+		 * a vsyscall or vdso (which executes in user-mode).
+		 *
+		 * XXX This is nasty, we should have a symbol list in the
+		 * "[vdso]" dso, but for now lets use the old trick of looking
+		 * in the whole kernel symbol list.
+		 */
+		if ((long long)al->addr < 0 &&
+		    cpumode == PERF_RECORD_MISC_USER &&
+		    machine && mg != &machine->kmaps) {
+			mg = &machine->kmaps;
+			goto try_again;
+		}
+	} else
+		al->addr = al->map->map_ip(al->map, al->addr);
+}
+
+void thread__find_addr_location(struct thread *self,
+				struct perf_session *session, u8 cpumode,
+				enum map_type type, pid_t pid, u64 addr,
+				struct addr_location *al,
+				symbol_filter_t filter)
+{
+	thread__find_addr_map(self, session, cpumode, type, pid, addr, al);
+	if (al->map != NULL)
+		al->sym = map__find_symbol(al->map, al->addr, filter);
+	else
+		al->sym = NULL;
+}
+
+int perf_event__preprocess_sample(const union perf_event *event,
+				  struct perf_session *session,
+				  struct addr_location *al,
+				  struct perf_sample *sample,
+				  symbol_filter_t filter)
+{
+	u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+	struct thread *thread = perf_session__findnew(session, event->ip.pid);
+
+	if (thread == NULL)
+		return -1;
+
+	if (symbol_conf.comm_list &&
+	    !strlist__has_entry(symbol_conf.comm_list, thread->comm))
+		goto out_filtered;
+
+	dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid);
+	/*
+	 * Have we already created the kernel maps for the host machine?
+	 *
+	 * This should have happened earlier, when we processed the kernel MMAP
+	 * events, but for older perf.data files there was no such thing, so do
+	 * it now.
+	 */
+	if (cpumode == PERF_RECORD_MISC_KERNEL &&
+	    session->host_machine.vmlinux_maps[MAP__FUNCTION] == NULL)
+		machine__create_kernel_maps(&session->host_machine);
+
+	thread__find_addr_map(thread, session, cpumode, MAP__FUNCTION,
+			      event->ip.pid, event->ip.ip, al);
+	dump_printf(" ...... dso: %s\n",
+		    al->map ? al->map->dso->long_name :
+			al->level == 'H' ? "[hypervisor]" : "<not found>");
+	al->sym = NULL;
+	al->cpu = sample->cpu;
+
+	if (al->map) {
+		if (symbol_conf.dso_list &&
+		    (!al->map || !al->map->dso ||
+		     !(strlist__has_entry(symbol_conf.dso_list,
+					  al->map->dso->short_name) ||
+		       (al->map->dso->short_name != al->map->dso->long_name &&
+			strlist__has_entry(symbol_conf.dso_list,
+					   al->map->dso->long_name)))))
+			goto out_filtered;
+
+		al->sym = map__find_symbol(al->map, al->addr, filter);
+	}
+
+	if (symbol_conf.sym_list && al->sym &&
+	    !strlist__has_entry(symbol_conf.sym_list, al->sym->name))
+		goto out_filtered;
+
+	return 0;
+
+out_filtered:
+	al->filtered = true;
+	return 0;
+}
diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h
new file mode 100644
index 00000000..1d7f6648
--- /dev/null
+++ b/tools/perf/util/event.h
@@ -0,0 +1,191 @@
+#ifndef __PERF_RECORD_H
+#define __PERF_RECORD_H
+
+#include <limits.h>
+
+#include "../perf.h"
+#include "map.h"
+
+/*
+ * PERF_SAMPLE_IP | PERF_SAMPLE_TID | *
+ */
+struct ip_event {
+	struct perf_event_header header;
+	u64 ip;
+	u32 pid, tid;
+	unsigned char __more_data[];
+};
+
+struct mmap_event {
+	struct perf_event_header header;
+	u32 pid, tid;
+	u64 start;
+	u64 len;
+	u64 pgoff;
+	char filename[PATH_MAX];
+};
+
+struct comm_event {
+	struct perf_event_header header;
+	u32 pid, tid;
+	char comm[16];
+};
+
+struct fork_event {
+	struct perf_event_header header;
+	u32 pid, ppid;
+	u32 tid, ptid;
+	u64 time;
+};
+
+struct lost_event {
+	struct perf_event_header header;
+	u64 id;
+	u64 lost;
+};
+
+/*
+ * PERF_FORMAT_ENABLED | PERF_FORMAT_RUNNING | PERF_FORMAT_ID
+ */
+struct read_event {
+	struct perf_event_header header;
+	u32 pid, tid;
+	u64 value;
+	u64 time_enabled;
+	u64 time_running;
+	u64 id;
+};
+
+
+#define PERF_SAMPLE_MASK				\
+	(PERF_SAMPLE_IP | PERF_SAMPLE_TID |		\
+	 PERF_SAMPLE_TIME | PERF_SAMPLE_ADDR |		\
+	PERF_SAMPLE_ID | PERF_SAMPLE_STREAM_ID |	\
+	 PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD)
+
+struct sample_event {
+	struct perf_event_header        header;
+	u64 array[];
+};
+
+struct perf_sample {
+	u64 ip;
+	u32 pid, tid;
+	u64 time;
+	u64 addr;
+	u64 id;
+	u64 stream_id;
+	u64 period;
+	u32 cpu;
+	u32 raw_size;
+	void *raw_data;
+	struct ip_callchain *callchain;
+};
+
+#define BUILD_ID_SIZE 20
+
+struct build_id_event {
+	struct perf_event_header header;
+	pid_t			 pid;
+	u8			 build_id[ALIGN(BUILD_ID_SIZE, sizeof(u64))];
+	char			 filename[];
+};
+
+enum perf_user_event_type { /* above any possible kernel type */
+	PERF_RECORD_USER_TYPE_START		= 64,
+	PERF_RECORD_HEADER_ATTR			= 64,
+	PERF_RECORD_HEADER_EVENT_TYPE		= 65,
+	PERF_RECORD_HEADER_TRACING_DATA		= 66,
+	PERF_RECORD_HEADER_BUILD_ID		= 67,
+	PERF_RECORD_FINISHED_ROUND		= 68,
+	PERF_RECORD_HEADER_MAX
+};
+
+struct attr_event {
+	struct perf_event_header header;
+	struct perf_event_attr attr;
+	u64 id[];
+};
+
+#define MAX_EVENT_NAME 64
+
+struct perf_trace_event_type {
+	u64	event_id;
+	char	name[MAX_EVENT_NAME];
+};
+
+struct event_type_event {
+	struct perf_event_header header;
+	struct perf_trace_event_type event_type;
+};
+
+struct tracing_data_event {
+	struct perf_event_header header;
+	u32 size;
+};
+
+union perf_event {
+	struct perf_event_header	header;
+	struct ip_event			ip;
+	struct mmap_event		mmap;
+	struct comm_event		comm;
+	struct fork_event		fork;
+	struct lost_event		lost;
+	struct read_event		read;
+	struct sample_event		sample;
+	struct attr_event		attr;
+	struct event_type_event		event_type;
+	struct tracing_data_event	tracing_data;
+	struct build_id_event		build_id;
+};
+
+void perf_event__print_totals(void);
+
+struct perf_session;
+struct thread_map;
+
+typedef int (*perf_event__handler_synth_t)(union perf_event *event, 
+					   struct perf_session *session);
+typedef int (*perf_event__handler_t)(union perf_event *event,
+				     struct perf_sample *sample,
+				      struct perf_session *session);
+
+int perf_event__synthesize_thread_map(struct thread_map *threads,
+				      perf_event__handler_t process,
+				      struct perf_session *session);
+int perf_event__synthesize_threads(perf_event__handler_t process,
+				   struct perf_session *session);
+int perf_event__synthesize_kernel_mmap(perf_event__handler_t process,
+				       struct perf_session *session,
+				       struct machine *machine,
+				       const char *symbol_name);
+
+int perf_event__synthesize_modules(perf_event__handler_t process,
+				   struct perf_session *session,
+				   struct machine *machine);
+
+int perf_event__process_comm(union perf_event *event, struct perf_sample *sample,
+			     struct perf_session *session);
+int perf_event__process_lost(union perf_event *event, struct perf_sample *sample,
+			     struct perf_session *session);
+int perf_event__process_mmap(union perf_event *event, struct perf_sample *sample,
+			     struct perf_session *session);
+int perf_event__process_task(union perf_event *event, struct perf_sample *sample,
+			     struct perf_session *session);
+int perf_event__process(union perf_event *event, struct perf_sample *sample,
+			struct perf_session *session);
+
+struct addr_location;
+int perf_event__preprocess_sample(const union perf_event *self,
+				  struct perf_session *session,
+				  struct addr_location *al,
+				  struct perf_sample *sample,
+				  symbol_filter_t filter);
+
+const char *perf_event__name(unsigned int id);
+
+int perf_event__parse_sample(const union perf_event *event, u64 type,
+			     int sample_size, bool sample_id_all,
+			     struct perf_sample *sample);
+
+#endif /* __PERF_RECORD_H */
diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
new file mode 100644
index 00000000..b021ea92
--- /dev/null
+++ b/tools/perf/util/evlist.c
@@ -0,0 +1,500 @@
+/*
+ * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Parts came from builtin-{top,stat,record}.c, see those files for further
+ * copyright notes.
+ *
+ * Released under the GPL v2. (and only v2, not any later version)
+ */
+#include <poll.h>
+#include "cpumap.h"
+#include "thread_map.h"
+#include "evlist.h"
+#include "evsel.h"
+#include "util.h"
+
+#include <sys/mman.h>
+
+#include <linux/bitops.h>
+#include <linux/hash.h>
+
+#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y))
+#define SID(e, x, y) xyarray__entry(e->sample_id, x, y)
+
+void perf_evlist__init(struct perf_evlist *evlist, struct cpu_map *cpus,
+		       struct thread_map *threads)
+{
+	int i;
+
+	for (i = 0; i < PERF_EVLIST__HLIST_SIZE; ++i)
+		INIT_HLIST_HEAD(&evlist->heads[i]);
+	INIT_LIST_HEAD(&evlist->entries);
+	perf_evlist__set_maps(evlist, cpus, threads);
+}
+
+struct perf_evlist *perf_evlist__new(struct cpu_map *cpus,
+				     struct thread_map *threads)
+{
+	struct perf_evlist *evlist = zalloc(sizeof(*evlist));
+
+	if (evlist != NULL)
+		perf_evlist__init(evlist, cpus, threads);
+
+	return evlist;
+}
+
+static void perf_evlist__purge(struct perf_evlist *evlist)
+{
+	struct perf_evsel *pos, *n;
+
+	list_for_each_entry_safe(pos, n, &evlist->entries, node) {
+		list_del_init(&pos->node);
+		perf_evsel__delete(pos);
+	}
+
+	evlist->nr_entries = 0;
+}
+
+void perf_evlist__exit(struct perf_evlist *evlist)
+{
+	free(evlist->mmap);
+	free(evlist->pollfd);
+	evlist->mmap = NULL;
+	evlist->pollfd = NULL;
+}
+
+void perf_evlist__delete(struct perf_evlist *evlist)
+{
+	perf_evlist__purge(evlist);
+	perf_evlist__exit(evlist);
+	free(evlist);
+}
+
+void perf_evlist__add(struct perf_evlist *evlist, struct perf_evsel *entry)
+{
+	list_add_tail(&entry->node, &evlist->entries);
+	++evlist->nr_entries;
+}
+
+int perf_evlist__add_default(struct perf_evlist *evlist)
+{
+	struct perf_event_attr attr = {
+		.type = PERF_TYPE_HARDWARE,
+		.config = PERF_COUNT_HW_CPU_CYCLES,
+	};
+	struct perf_evsel *evsel = perf_evsel__new(&attr, 0);
+
+	if (evsel == NULL)
+		return -ENOMEM;
+
+	perf_evlist__add(evlist, evsel);
+	return 0;
+}
+
+int perf_evlist__alloc_pollfd(struct perf_evlist *evlist)
+{
+	int nfds = evlist->cpus->nr * evlist->threads->nr * evlist->nr_entries;
+	evlist->pollfd = malloc(sizeof(struct pollfd) * nfds);
+	return evlist->pollfd != NULL ? 0 : -ENOMEM;
+}
+
+void perf_evlist__add_pollfd(struct perf_evlist *evlist, int fd)
+{
+	fcntl(fd, F_SETFL, O_NONBLOCK);
+	evlist->pollfd[evlist->nr_fds].fd = fd;
+	evlist->pollfd[evlist->nr_fds].events = POLLIN;
+	evlist->nr_fds++;
+}
+
+static void perf_evlist__id_hash(struct perf_evlist *evlist,
+				 struct perf_evsel *evsel,
+				 int cpu, int thread, u64 id)
+{
+	int hash;
+	struct perf_sample_id *sid = SID(evsel, cpu, thread);
+
+	sid->id = id;
+	sid->evsel = evsel;
+	hash = hash_64(sid->id, PERF_EVLIST__HLIST_BITS);
+	hlist_add_head(&sid->node, &evlist->heads[hash]);
+}
+
+void perf_evlist__id_add(struct perf_evlist *evlist, struct perf_evsel *evsel,
+			 int cpu, int thread, u64 id)
+{
+	perf_evlist__id_hash(evlist, evsel, cpu, thread, id);
+	evsel->id[evsel->ids++] = id;
+}
+
+static int perf_evlist__id_add_fd(struct perf_evlist *evlist,
+				  struct perf_evsel *evsel,
+				  int cpu, int thread, int fd)
+{
+	u64 read_data[4] = { 0, };
+	int id_idx = 1; /* The first entry is the counter value */
+
+	if (!(evsel->attr.read_format & PERF_FORMAT_ID) ||
+	    read(fd, &read_data, sizeof(read_data)) == -1)
+		return -1;
+
+	if (evsel->attr.read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
+		++id_idx;
+	if (evsel->attr.read_format & PERF_FORMAT_TOTAL_TIME_RUNNING)
+		++id_idx;
+
+	perf_evlist__id_add(evlist, evsel, cpu, thread, read_data[id_idx]);
+	return 0;
+}
+
+struct perf_evsel *perf_evlist__id2evsel(struct perf_evlist *evlist, u64 id)
+{
+	struct hlist_head *head;
+	struct hlist_node *pos;
+	struct perf_sample_id *sid;
+	int hash;
+
+	if (evlist->nr_entries == 1)
+		return list_entry(evlist->entries.next, struct perf_evsel, node);
+
+	hash = hash_64(id, PERF_EVLIST__HLIST_BITS);
+	head = &evlist->heads[hash];
+
+	hlist_for_each_entry(sid, pos, head, node)
+		if (sid->id == id)
+			return sid->evsel;
+	return NULL;
+}
+
+union perf_event *perf_evlist__mmap_read(struct perf_evlist *evlist, int idx)
+{
+	/* XXX Move this to perf.c, making it generally available */
+	unsigned int page_size = sysconf(_SC_PAGE_SIZE);
+	struct perf_mmap *md = &evlist->mmap[idx];
+	unsigned int head = perf_mmap__read_head(md);
+	unsigned int old = md->prev;
+	unsigned char *data = md->base + page_size;
+	union perf_event *event = NULL;
+
+	if (evlist->overwrite) {
+		/*
+		 * If we're further behind than half the buffer, there's a chance
+		 * the writer will bite our tail and mess up the samples under us.
+		 *
+		 * If we somehow ended up ahead of the head, we got messed up.
+		 *
+		 * In either case, truncate and restart at head.
+		 */
+		int diff = head - old;
+		if (diff > md->mask / 2 || diff < 0) {
+			fprintf(stderr, "WARNING: failed to keep up with mmap data.\n");
+
+			/*
+			 * head points to a known good entry, start there.
+			 */
+			old = head;
+		}
+	}
+
+	if (old != head) {
+		size_t size;
+
+		event = (union perf_event *)&data[old & md->mask];
+		size = event->header.size;
+
+		/*
+		 * Event straddles the mmap boundary -- header should always
+		 * be inside due to u64 alignment of output.
+		 */
+		if ((old & md->mask) + size != ((old + size) & md->mask)) {
+			unsigned int offset = old;
+			unsigned int len = min(sizeof(*event), size), cpy;
+			void *dst = &evlist->event_copy;
+
+			do {
+				cpy = min(md->mask + 1 - (offset & md->mask), len);
+				memcpy(dst, &data[offset & md->mask], cpy);
+				offset += cpy;
+				dst += cpy;
+				len -= cpy;
+			} while (len);
+
+			event = &evlist->event_copy;
+		}
+
+		old += size;
+	}
+
+	md->prev = old;
+
+	if (!evlist->overwrite)
+		perf_mmap__write_tail(md, old);
+
+	return event;
+}
+
+void perf_evlist__munmap(struct perf_evlist *evlist)
+{
+	int i;
+
+	for (i = 0; i < evlist->nr_mmaps; i++) {
+		if (evlist->mmap[i].base != NULL) {
+			munmap(evlist->mmap[i].base, evlist->mmap_len);
+			evlist->mmap[i].base = NULL;
+		}
+	}
+
+	free(evlist->mmap);
+	evlist->mmap = NULL;
+}
+
+int perf_evlist__alloc_mmap(struct perf_evlist *evlist)
+{
+	evlist->nr_mmaps = evlist->cpus->nr;
+	if (evlist->cpus->map[0] == -1)
+		evlist->nr_mmaps = evlist->threads->nr;
+	evlist->mmap = zalloc(evlist->nr_mmaps * sizeof(struct perf_mmap));
+	return evlist->mmap != NULL ? 0 : -ENOMEM;
+}
+
+static int __perf_evlist__mmap(struct perf_evlist *evlist,
+			       int idx, int prot, int mask, int fd)
+{
+	evlist->mmap[idx].prev = 0;
+	evlist->mmap[idx].mask = mask;
+	evlist->mmap[idx].base = mmap(NULL, evlist->mmap_len, prot,
+				      MAP_SHARED, fd, 0);
+	if (evlist->mmap[idx].base == MAP_FAILED)
+		return -1;
+
+	perf_evlist__add_pollfd(evlist, fd);
+	return 0;
+}
+
+static int perf_evlist__mmap_per_cpu(struct perf_evlist *evlist, int prot, int mask)
+{
+	struct perf_evsel *evsel;
+	int cpu, thread;
+
+	for (cpu = 0; cpu < evlist->cpus->nr; cpu++) {
+		int output = -1;
+
+		for (thread = 0; thread < evlist->threads->nr; thread++) {
+			list_for_each_entry(evsel, &evlist->entries, node) {
+				int fd = FD(evsel, cpu, thread);
+
+				if (output == -1) {
+					output = fd;
+					if (__perf_evlist__mmap(evlist, cpu,
+								prot, mask, output) < 0)
+						goto out_unmap;
+				} else {
+					if (ioctl(fd, PERF_EVENT_IOC_SET_OUTPUT, output) != 0)
+						goto out_unmap;
+				}
+
+				if ((evsel->attr.read_format & PERF_FORMAT_ID) &&
+				    perf_evlist__id_add_fd(evlist, evsel, cpu, thread, fd) < 0)
+					goto out_unmap;
+			}
+		}
+	}
+
+	return 0;
+
+out_unmap:
+	for (cpu = 0; cpu < evlist->cpus->nr; cpu++) {
+		if (evlist->mmap[cpu].base != NULL) {
+			munmap(evlist->mmap[cpu].base, evlist->mmap_len);
+			evlist->mmap[cpu].base = NULL;
+		}
+	}
+	return -1;
+}
+
+static int perf_evlist__mmap_per_thread(struct perf_evlist *evlist, int prot, int mask)
+{
+	struct perf_evsel *evsel;
+	int thread;
+
+	for (thread = 0; thread < evlist->threads->nr; thread++) {
+		int output = -1;
+
+		list_for_each_entry(evsel, &evlist->entries, node) {
+			int fd = FD(evsel, 0, thread);
+
+			if (output == -1) {
+				output = fd;
+				if (__perf_evlist__mmap(evlist, thread,
+							prot, mask, output) < 0)
+					goto out_unmap;
+			} else {
+				if (ioctl(fd, PERF_EVENT_IOC_SET_OUTPUT, output) != 0)
+					goto out_unmap;
+			}
+
+			if ((evsel->attr.read_format & PERF_FORMAT_ID) &&
+			    perf_evlist__id_add_fd(evlist, evsel, 0, thread, fd) < 0)
+				goto out_unmap;
+		}
+	}
+
+	return 0;
+
+out_unmap:
+	for (thread = 0; thread < evlist->threads->nr; thread++) {
+		if (evlist->mmap[thread].base != NULL) {
+			munmap(evlist->mmap[thread].base, evlist->mmap_len);
+			evlist->mmap[thread].base = NULL;
+		}
+	}
+	return -1;
+}
+
+/** perf_evlist__mmap - Create per cpu maps to receive events
+ *
+ * @evlist - list of events
+ * @pages - map length in pages
+ * @overwrite - overwrite older events?
+ *
+ * If overwrite is false the user needs to signal event consuption using:
+ *
+ *	struct perf_mmap *m = &evlist->mmap[cpu];
+ *	unsigned int head = perf_mmap__read_head(m);
+ *
+ *	perf_mmap__write_tail(m, head)
+ *
+ * Using perf_evlist__read_on_cpu does this automatically.
+ */
+int perf_evlist__mmap(struct perf_evlist *evlist, int pages, bool overwrite)
+{
+	unsigned int page_size = sysconf(_SC_PAGE_SIZE);
+	int mask = pages * page_size - 1;
+	struct perf_evsel *evsel;
+	const struct cpu_map *cpus = evlist->cpus;
+	const struct thread_map *threads = evlist->threads;
+	int prot = PROT_READ | (overwrite ? 0 : PROT_WRITE);
+
+	if (evlist->mmap == NULL && perf_evlist__alloc_mmap(evlist) < 0)
+		return -ENOMEM;
+
+	if (evlist->pollfd == NULL && perf_evlist__alloc_pollfd(evlist) < 0)
+		return -ENOMEM;
+
+	evlist->overwrite = overwrite;
+	evlist->mmap_len = (pages + 1) * page_size;
+
+	list_for_each_entry(evsel, &evlist->entries, node) {
+		if ((evsel->attr.read_format & PERF_FORMAT_ID) &&
+		    evsel->sample_id == NULL &&
+		    perf_evsel__alloc_id(evsel, cpus->nr, threads->nr) < 0)
+			return -ENOMEM;
+	}
+
+	if (evlist->cpus->map[0] == -1)
+		return perf_evlist__mmap_per_thread(evlist, prot, mask);
+
+	return perf_evlist__mmap_per_cpu(evlist, prot, mask);
+}
+
+int perf_evlist__create_maps(struct perf_evlist *evlist, pid_t target_pid,
+			     pid_t target_tid, const char *cpu_list)
+{
+	evlist->threads = thread_map__new(target_pid, target_tid);
+
+	if (evlist->threads == NULL)
+		return -1;
+
+	if (cpu_list == NULL && target_tid != -1)
+		evlist->cpus = cpu_map__dummy_new();
+	else
+		evlist->cpus = cpu_map__new(cpu_list);
+
+	if (evlist->cpus == NULL)
+		goto out_delete_threads;
+
+	return 0;
+
+out_delete_threads:
+	thread_map__delete(evlist->threads);
+	return -1;
+}
+
+void perf_evlist__delete_maps(struct perf_evlist *evlist)
+{
+	cpu_map__delete(evlist->cpus);
+	thread_map__delete(evlist->threads);
+	evlist->cpus	= NULL;
+	evlist->threads = NULL;
+}
+
+int perf_evlist__set_filters(struct perf_evlist *evlist)
+{
+	const struct thread_map *threads = evlist->threads;
+	const struct cpu_map *cpus = evlist->cpus;
+	struct perf_evsel *evsel;
+	char *filter;
+	int thread;
+	int cpu;
+	int err;
+	int fd;
+
+	list_for_each_entry(evsel, &evlist->entries, node) {
+		filter = evsel->filter;
+		if (!filter)
+			continue;
+		for (cpu = 0; cpu < cpus->nr; cpu++) {
+			for (thread = 0; thread < threads->nr; thread++) {
+				fd = FD(evsel, cpu, thread);
+				err = ioctl(fd, PERF_EVENT_IOC_SET_FILTER, filter);
+				if (err)
+					return err;
+			}
+		}
+	}
+
+	return 0;
+}
+
+bool perf_evlist__valid_sample_type(const struct perf_evlist *evlist)
+{
+	struct perf_evsel *pos, *first;
+
+	pos = first = list_entry(evlist->entries.next, struct perf_evsel, node);
+
+	list_for_each_entry_continue(pos, &evlist->entries, node) {
+		if (first->attr.sample_type != pos->attr.sample_type)
+			return false;
+	}
+
+	return true;
+}
+
+u64 perf_evlist__sample_type(const struct perf_evlist *evlist)
+{
+	struct perf_evsel *first;
+
+	first = list_entry(evlist->entries.next, struct perf_evsel, node);
+	return first->attr.sample_type;
+}
+
+bool perf_evlist__valid_sample_id_all(const struct perf_evlist *evlist)
+{
+	struct perf_evsel *pos, *first;
+
+	pos = first = list_entry(evlist->entries.next, struct perf_evsel, node);
+
+	list_for_each_entry_continue(pos, &evlist->entries, node) {
+		if (first->attr.sample_id_all != pos->attr.sample_id_all)
+			return false;
+	}
+
+	return true;
+}
+
+bool perf_evlist__sample_id_all(const struct perf_evlist *evlist)
+{
+	struct perf_evsel *first;
+
+	first = list_entry(evlist->entries.next, struct perf_evsel, node);
+	return first->attr.sample_id_all;
+}
diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h
new file mode 100644
index 00000000..b2b86237
--- /dev/null
+++ b/tools/perf/util/evlist.h
@@ -0,0 +1,74 @@
+#ifndef __PERF_EVLIST_H
+#define __PERF_EVLIST_H 1
+
+#include <linux/list.h>
+#include "../perf.h"
+#include "event.h"
+
+struct pollfd;
+struct thread_map;
+struct cpu_map;
+
+#define PERF_EVLIST__HLIST_BITS 8
+#define PERF_EVLIST__HLIST_SIZE (1 << PERF_EVLIST__HLIST_BITS)
+
+struct perf_evlist {
+	struct list_head entries;
+	struct hlist_head heads[PERF_EVLIST__HLIST_SIZE];
+	int		 nr_entries;
+	int		 nr_fds;
+	int		 nr_mmaps;
+	int		 mmap_len;
+	bool		 overwrite;
+	union perf_event event_copy;
+	struct perf_mmap *mmap;
+	struct pollfd	 *pollfd;
+	struct thread_map *threads;
+	struct cpu_map	  *cpus;
+};
+
+struct perf_evsel;
+
+struct perf_evlist *perf_evlist__new(struct cpu_map *cpus,
+				     struct thread_map *threads);
+void perf_evlist__init(struct perf_evlist *evlist, struct cpu_map *cpus,
+		       struct thread_map *threads);
+void perf_evlist__exit(struct perf_evlist *evlist);
+void perf_evlist__delete(struct perf_evlist *evlist);
+
+void perf_evlist__add(struct perf_evlist *evlist, struct perf_evsel *entry);
+int perf_evlist__add_default(struct perf_evlist *evlist);
+
+void perf_evlist__id_add(struct perf_evlist *evlist, struct perf_evsel *evsel,
+			 int cpu, int thread, u64 id);
+
+int perf_evlist__alloc_pollfd(struct perf_evlist *evlist);
+void perf_evlist__add_pollfd(struct perf_evlist *evlist, int fd);
+
+struct perf_evsel *perf_evlist__id2evsel(struct perf_evlist *evlist, u64 id);
+
+union perf_event *perf_evlist__mmap_read(struct perf_evlist *self, int idx);
+
+int perf_evlist__alloc_mmap(struct perf_evlist *evlist);
+int perf_evlist__mmap(struct perf_evlist *evlist, int pages, bool overwrite);
+void perf_evlist__munmap(struct perf_evlist *evlist);
+
+static inline void perf_evlist__set_maps(struct perf_evlist *evlist,
+					 struct cpu_map *cpus,
+					 struct thread_map *threads)
+{
+	evlist->cpus	= cpus;
+	evlist->threads	= threads;
+}
+
+int perf_evlist__create_maps(struct perf_evlist *evlist, pid_t target_pid,
+			     pid_t target_tid, const char *cpu_list);
+void perf_evlist__delete_maps(struct perf_evlist *evlist);
+int perf_evlist__set_filters(struct perf_evlist *evlist);
+
+u64 perf_evlist__sample_type(const struct perf_evlist *evlist);
+bool perf_evlist__sample_id_all(const const struct perf_evlist *evlist);
+
+bool perf_evlist__valid_sample_type(const struct perf_evlist *evlist);
+bool perf_evlist__valid_sample_id_all(const struct perf_evlist *evlist);
+#endif /* __PERF_EVLIST_H */
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
new file mode 100644
index 00000000..ad2183c9
--- /dev/null
+++ b/tools/perf/util/evsel.c
@@ -0,0 +1,441 @@
+/*
+ * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Parts came from builtin-{top,stat,record}.c, see those files for further
+ * copyright notes.
+ *
+ * Released under the GPL v2. (and only v2, not any later version)
+ */
+
+#include "evsel.h"
+#include "evlist.h"
+#include "util.h"
+#include "cpumap.h"
+#include "thread_map.h"
+
+#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y))
+
+int __perf_evsel__sample_size(u64 sample_type)
+{
+	u64 mask = sample_type & PERF_SAMPLE_MASK;
+	int size = 0;
+	int i;
+
+	for (i = 0; i < 64; i++) {
+		if (mask & (1ULL << i))
+			size++;
+	}
+
+	size *= sizeof(u64);
+
+	return size;
+}
+
+void perf_evsel__init(struct perf_evsel *evsel,
+		      struct perf_event_attr *attr, int idx)
+{
+	evsel->idx	   = idx;
+	evsel->attr	   = *attr;
+	INIT_LIST_HEAD(&evsel->node);
+}
+
+struct perf_evsel *perf_evsel__new(struct perf_event_attr *attr, int idx)
+{
+	struct perf_evsel *evsel = zalloc(sizeof(*evsel));
+
+	if (evsel != NULL)
+		perf_evsel__init(evsel, attr, idx);
+
+	return evsel;
+}
+
+int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
+{
+	int cpu, thread;
+	evsel->fd = xyarray__new(ncpus, nthreads, sizeof(int));
+
+	if (evsel->fd) {
+		for (cpu = 0; cpu < ncpus; cpu++) {
+			for (thread = 0; thread < nthreads; thread++) {
+				FD(evsel, cpu, thread) = -1;
+			}
+		}
+	}
+
+	return evsel->fd != NULL ? 0 : -ENOMEM;
+}
+
+int perf_evsel__alloc_id(struct perf_evsel *evsel, int ncpus, int nthreads)
+{
+	evsel->sample_id = xyarray__new(ncpus, nthreads, sizeof(struct perf_sample_id));
+	if (evsel->sample_id == NULL)
+		return -ENOMEM;
+
+	evsel->id = zalloc(ncpus * nthreads * sizeof(u64));
+	if (evsel->id == NULL) {
+		xyarray__delete(evsel->sample_id);
+		evsel->sample_id = NULL;
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+int perf_evsel__alloc_counts(struct perf_evsel *evsel, int ncpus)
+{
+	evsel->counts = zalloc((sizeof(*evsel->counts) +
+				(ncpus * sizeof(struct perf_counts_values))));
+	return evsel->counts != NULL ? 0 : -ENOMEM;
+}
+
+void perf_evsel__free_fd(struct perf_evsel *evsel)
+{
+	xyarray__delete(evsel->fd);
+	evsel->fd = NULL;
+}
+
+void perf_evsel__free_id(struct perf_evsel *evsel)
+{
+	xyarray__delete(evsel->sample_id);
+	evsel->sample_id = NULL;
+	free(evsel->id);
+	evsel->id = NULL;
+}
+
+void perf_evsel__close_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
+{
+	int cpu, thread;
+
+	for (cpu = 0; cpu < ncpus; cpu++)
+		for (thread = 0; thread < nthreads; ++thread) {
+			close(FD(evsel, cpu, thread));
+			FD(evsel, cpu, thread) = -1;
+		}
+}
+
+void perf_evsel__exit(struct perf_evsel *evsel)
+{
+	assert(list_empty(&evsel->node));
+	xyarray__delete(evsel->fd);
+	xyarray__delete(evsel->sample_id);
+	free(evsel->id);
+}
+
+void perf_evsel__delete(struct perf_evsel *evsel)
+{
+	perf_evsel__exit(evsel);
+	close_cgroup(evsel->cgrp);
+	free(evsel->name);
+	free(evsel);
+}
+
+int __perf_evsel__read_on_cpu(struct perf_evsel *evsel,
+			      int cpu, int thread, bool scale)
+{
+	struct perf_counts_values count;
+	size_t nv = scale ? 3 : 1;
+
+	if (FD(evsel, cpu, thread) < 0)
+		return -EINVAL;
+
+	if (evsel->counts == NULL && perf_evsel__alloc_counts(evsel, cpu + 1) < 0)
+		return -ENOMEM;
+
+	if (readn(FD(evsel, cpu, thread), &count, nv * sizeof(u64)) < 0)
+		return -errno;
+
+	if (scale) {
+		if (count.run == 0)
+			count.val = 0;
+		else if (count.run < count.ena)
+			count.val = (u64)((double)count.val * count.ena / count.run + 0.5);
+	} else
+		count.ena = count.run = 0;
+
+	evsel->counts->cpu[cpu] = count;
+	return 0;
+}
+
+int __perf_evsel__read(struct perf_evsel *evsel,
+		       int ncpus, int nthreads, bool scale)
+{
+	size_t nv = scale ? 3 : 1;
+	int cpu, thread;
+	struct perf_counts_values *aggr = &evsel->counts->aggr, count;
+
+	aggr->val = aggr->ena = aggr->run = 0;
+
+	for (cpu = 0; cpu < ncpus; cpu++) {
+		for (thread = 0; thread < nthreads; thread++) {
+			if (FD(evsel, cpu, thread) < 0)
+				continue;
+
+			if (readn(FD(evsel, cpu, thread),
+				  &count, nv * sizeof(u64)) < 0)
+				return -errno;
+
+			aggr->val += count.val;
+			if (scale) {
+				aggr->ena += count.ena;
+				aggr->run += count.run;
+			}
+		}
+	}
+
+	evsel->counts->scaled = 0;
+	if (scale) {
+		if (aggr->run == 0) {
+			evsel->counts->scaled = -1;
+			aggr->val = 0;
+			return 0;
+		}
+
+		if (aggr->run < aggr->ena) {
+			evsel->counts->scaled = 1;
+			aggr->val = (u64)((double)aggr->val * aggr->ena / aggr->run + 0.5);
+		}
+	} else
+		aggr->ena = aggr->run = 0;
+
+	return 0;
+}
+
+static int __perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
+			      struct thread_map *threads, bool group)
+{
+	int cpu, thread;
+	unsigned long flags = 0;
+	int pid = -1;
+
+	if (evsel->fd == NULL &&
+	    perf_evsel__alloc_fd(evsel, cpus->nr, threads->nr) < 0)
+		return -1;
+
+	if (evsel->cgrp) {
+		flags = PERF_FLAG_PID_CGROUP;
+		pid = evsel->cgrp->fd;
+	}
+
+	for (cpu = 0; cpu < cpus->nr; cpu++) {
+		int group_fd = -1;
+
+		for (thread = 0; thread < threads->nr; thread++) {
+
+			if (!evsel->cgrp)
+				pid = threads->map[thread];
+
+			FD(evsel, cpu, thread) = sys_perf_event_open(&evsel->attr,
+								     pid,
+								     cpus->map[cpu],
+								     group_fd, flags);
+			if (FD(evsel, cpu, thread) < 0)
+				goto out_close;
+
+			if (group && group_fd == -1)
+				group_fd = FD(evsel, cpu, thread);
+		}
+	}
+
+	return 0;
+
+out_close:
+	do {
+		while (--thread >= 0) {
+			close(FD(evsel, cpu, thread));
+			FD(evsel, cpu, thread) = -1;
+		}
+		thread = threads->nr;
+	} while (--cpu >= 0);
+	return -1;
+}
+
+static struct {
+	struct cpu_map map;
+	int cpus[1];
+} empty_cpu_map = {
+	.map.nr	= 1,
+	.cpus	= { -1, },
+};
+
+static struct {
+	struct thread_map map;
+	int threads[1];
+} empty_thread_map = {
+	.map.nr	 = 1,
+	.threads = { -1, },
+};
+
+int perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
+		     struct thread_map *threads, bool group)
+{
+	if (cpus == NULL) {
+		/* Work around old compiler warnings about strict aliasing */
+		cpus = &empty_cpu_map.map;
+	}
+
+	if (threads == NULL)
+		threads = &empty_thread_map.map;
+
+	return __perf_evsel__open(evsel, cpus, threads, group);
+}
+
+int perf_evsel__open_per_cpu(struct perf_evsel *evsel,
+			     struct cpu_map *cpus, bool group)
+{
+	return __perf_evsel__open(evsel, cpus, &empty_thread_map.map, group);
+}
+
+int perf_evsel__open_per_thread(struct perf_evsel *evsel,
+				struct thread_map *threads, bool group)
+{
+	return __perf_evsel__open(evsel, &empty_cpu_map.map, threads, group);
+}
+
+static int perf_event__parse_id_sample(const union perf_event *event, u64 type,
+				       struct perf_sample *sample)
+{
+	const u64 *array = event->sample.array;
+
+	array += ((event->header.size -
+		   sizeof(event->header)) / sizeof(u64)) - 1;
+
+	if (type & PERF_SAMPLE_CPU) {
+		u32 *p = (u32 *)array;
+		sample->cpu = *p;
+		array--;
+	}
+
+	if (type & PERF_SAMPLE_STREAM_ID) {
+		sample->stream_id = *array;
+		array--;
+	}
+
+	if (type & PERF_SAMPLE_ID) {
+		sample->id = *array;
+		array--;
+	}
+
+	if (type & PERF_SAMPLE_TIME) {
+		sample->time = *array;
+		array--;
+	}
+
+	if (type & PERF_SAMPLE_TID) {
+		u32 *p = (u32 *)array;
+		sample->pid = p[0];
+		sample->tid = p[1];
+	}
+
+	return 0;
+}
+
+static bool sample_overlap(const union perf_event *event,
+			   const void *offset, u64 size)
+{
+	const void *base = event;
+
+	if (offset + size > base + event->header.size)
+		return true;
+
+	return false;
+}
+
+int perf_event__parse_sample(const union perf_event *event, u64 type,
+			     int sample_size, bool sample_id_all,
+			     struct perf_sample *data)
+{
+	const u64 *array;
+
+	data->cpu = data->pid = data->tid = -1;
+	data->stream_id = data->id = data->time = -1ULL;
+	data->period = 1;
+
+	if (event->header.type != PERF_RECORD_SAMPLE) {
+		if (!sample_id_all)
+			return 0;
+		return perf_event__parse_id_sample(event, type, data);
+	}
+
+	array = event->sample.array;
+
+	if (sample_size + sizeof(event->header) > event->header.size)
+		return -EFAULT;
+
+	if (type & PERF_SAMPLE_IP) {
+		data->ip = event->ip.ip;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_TID) {
+		u32 *p = (u32 *)array;
+		data->pid = p[0];
+		data->tid = p[1];
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_TIME) {
+		data->time = *array;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_ADDR) {
+		data->addr = *array;
+		array++;
+	}
+
+	data->id = -1ULL;
+	if (type & PERF_SAMPLE_ID) {
+		data->id = *array;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_STREAM_ID) {
+		data->stream_id = *array;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_CPU) {
+		u32 *p = (u32 *)array;
+		data->cpu = *p;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_PERIOD) {
+		data->period = *array;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_READ) {
+		fprintf(stderr, "PERF_SAMPLE_READ is unsuported for now\n");
+		return -1;
+	}
+
+	if (type & PERF_SAMPLE_CALLCHAIN) {
+		if (sample_overlap(event, array, sizeof(data->callchain->nr)))
+			return -EFAULT;
+
+		data->callchain = (struct ip_callchain *)array;
+
+		if (sample_overlap(event, array, data->callchain->nr))
+			return -EFAULT;
+
+		array += 1 + data->callchain->nr;
+	}
+
+	if (type & PERF_SAMPLE_RAW) {
+		u32 *p = (u32 *)array;
+
+		if (sample_overlap(event, array, sizeof(u32)))
+			return -EFAULT;
+
+		data->raw_size = *p;
+		p++;
+
+		if (sample_overlap(event, p, data->raw_size))
+			return -EFAULT;
+
+		data->raw_data = p;
+	}
+
+	return 0;
+}
diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
new file mode 100644
index 00000000..7e9366e4
--- /dev/null
+++ b/tools/perf/util/evsel.h
@@ -0,0 +1,159 @@
+#ifndef __PERF_EVSEL_H
+#define __PERF_EVSEL_H 1
+
+#include <linux/list.h>
+#include <stdbool.h>
+#include "../../../include/linux/perf_event.h"
+#include "types.h"
+#include "xyarray.h"
+#include "cgroup.h"
+#include "hist.h"
+ 
+struct perf_counts_values {
+	union {
+		struct {
+			u64 val;
+			u64 ena;
+			u64 run;
+		};
+		u64 values[3];
+	};
+};
+
+struct perf_counts {
+	s8		   	  scaled;
+	struct perf_counts_values aggr;
+	struct perf_counts_values cpu[];
+};
+
+struct perf_evsel;
+
+/*
+ * Per fd, to map back from PERF_SAMPLE_ID to evsel, only used when there are
+ * more than one entry in the evlist.
+ */
+struct perf_sample_id {
+	struct hlist_node 	node;
+	u64		 	id;
+	struct perf_evsel	*evsel;
+};
+
+/** struct perf_evsel - event selector
+ *
+ * @name - Can be set to retain the original event name passed by the user,
+ *         so that when showing results in tools such as 'perf stat', we
+ *         show the name used, not some alias.
+ */
+struct perf_evsel {
+	struct list_head	node;
+	struct perf_event_attr	attr;
+	char			*filter;
+	struct xyarray		*fd;
+	struct xyarray		*sample_id;
+	u64			*id;
+	struct perf_counts	*counts;
+	int			idx;
+	int			ids;
+	struct hists		hists;
+	char			*name;
+	union {
+		void		*priv;
+		off_t		id_offset;
+	};
+	struct cgroup_sel	*cgrp;
+};
+
+struct cpu_map;
+struct thread_map;
+struct perf_evlist;
+
+struct perf_evsel *perf_evsel__new(struct perf_event_attr *attr, int idx);
+void perf_evsel__init(struct perf_evsel *evsel,
+		      struct perf_event_attr *attr, int idx);
+void perf_evsel__exit(struct perf_evsel *evsel);
+void perf_evsel__delete(struct perf_evsel *evsel);
+
+int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads);
+int perf_evsel__alloc_id(struct perf_evsel *evsel, int ncpus, int nthreads);
+int perf_evsel__alloc_counts(struct perf_evsel *evsel, int ncpus);
+void perf_evsel__free_fd(struct perf_evsel *evsel);
+void perf_evsel__free_id(struct perf_evsel *evsel);
+void perf_evsel__close_fd(struct perf_evsel *evsel, int ncpus, int nthreads);
+
+int perf_evsel__open_per_cpu(struct perf_evsel *evsel,
+			     struct cpu_map *cpus, bool group);
+int perf_evsel__open_per_thread(struct perf_evsel *evsel,
+				struct thread_map *threads, bool group);
+int perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
+		     struct thread_map *threads, bool group);
+
+#define perf_evsel__match(evsel, t, c)		\
+	(evsel->attr.type == PERF_TYPE_##t &&	\
+	 evsel->attr.config == PERF_COUNT_##c)
+
+int __perf_evsel__read_on_cpu(struct perf_evsel *evsel,
+			      int cpu, int thread, bool scale);
+
+/**
+ * perf_evsel__read_on_cpu - Read out the results on a CPU and thread
+ *
+ * @evsel - event selector to read value
+ * @cpu - CPU of interest
+ * @thread - thread of interest
+ */
+static inline int perf_evsel__read_on_cpu(struct perf_evsel *evsel,
+					  int cpu, int thread)
+{
+	return __perf_evsel__read_on_cpu(evsel, cpu, thread, false);
+}
+
+/**
+ * perf_evsel__read_on_cpu_scaled - Read out the results on a CPU and thread, scaled
+ *
+ * @evsel - event selector to read value
+ * @cpu - CPU of interest
+ * @thread - thread of interest
+ */
+static inline int perf_evsel__read_on_cpu_scaled(struct perf_evsel *evsel,
+						 int cpu, int thread)
+{
+	return __perf_evsel__read_on_cpu(evsel, cpu, thread, true);
+}
+
+int __perf_evsel__read(struct perf_evsel *evsel, int ncpus, int nthreads,
+		       bool scale);
+
+/**
+ * perf_evsel__read - Read the aggregate results on all CPUs
+ *
+ * @evsel - event selector to read value
+ * @ncpus - Number of cpus affected, from zero
+ * @nthreads - Number of threads affected, from zero
+ */
+static inline int perf_evsel__read(struct perf_evsel *evsel,
+				    int ncpus, int nthreads)
+{
+	return __perf_evsel__read(evsel, ncpus, nthreads, false);
+}
+
+/**
+ * perf_evsel__read_scaled - Read the aggregate results on all CPUs, scaled
+ *
+ * @evsel - event selector to read value
+ * @ncpus - Number of cpus affected, from zero
+ * @nthreads - Number of threads affected, from zero
+ */
+static inline int perf_evsel__read_scaled(struct perf_evsel *evsel,
+					  int ncpus, int nthreads)
+{
+	return __perf_evsel__read(evsel, ncpus, nthreads, true);
+}
+
+int __perf_evsel__sample_size(u64 sample_type);
+
+static inline int perf_evsel__sample_size(struct perf_evsel *evsel)
+{
+	return __perf_evsel__sample_size(evsel->attr.sample_type);
+}
+
+#endif /* __PERF_EVSEL_H */
diff --git a/tools/perf/util/exec_cmd.c b/tools/perf/util/exec_cmd.c
new file mode 100644
index 00000000..7adf4ad1
--- /dev/null
+++ b/tools/perf/util/exec_cmd.c
@@ -0,0 +1,148 @@
+#include "cache.h"
+#include "exec_cmd.h"
+#include "quote.h"
+
+#include <string.h>
+
+#define MAX_ARGS	32
+
+static const char *argv_exec_path;
+static const char *argv0_path;
+
+const char *system_path(const char *path)
+{
+	static const char *prefix = PREFIX;
+	struct strbuf d = STRBUF_INIT;
+
+	if (is_absolute_path(path))
+		return path;
+
+	strbuf_addf(&d, "%s/%s", prefix, path);
+	path = strbuf_detach(&d, NULL);
+	return path;
+}
+
+const char *perf_extract_argv0_path(const char *argv0)
+{
+	const char *slash;
+
+	if (!argv0 || !*argv0)
+		return NULL;
+	slash = argv0 + strlen(argv0);
+
+	while (argv0 <= slash && !is_dir_sep(*slash))
+		slash--;
+
+	if (slash >= argv0) {
+		argv0_path = strndup(argv0, slash - argv0);
+		return argv0_path ? slash + 1 : NULL;
+	}
+
+	return argv0;
+}
+
+void perf_set_argv_exec_path(const char *exec_path)
+{
+	argv_exec_path = exec_path;
+	/*
+	 * Propagate this setting to external programs.
+	 */
+	setenv(EXEC_PATH_ENVIRONMENT, exec_path, 1);
+}
+
+
+/* Returns the highest-priority, location to look for perf programs. */
+const char *perf_exec_path(void)
+{
+	const char *env;
+
+	if (argv_exec_path)
+		return argv_exec_path;
+
+	env = getenv(EXEC_PATH_ENVIRONMENT);
+	if (env && *env) {
+		return env;
+	}
+
+	return system_path(PERF_EXEC_PATH);
+}
+
+static void add_path(struct strbuf *out, const char *path)
+{
+	if (path && *path) {
+		if (is_absolute_path(path))
+			strbuf_addstr(out, path);
+		else
+			strbuf_addstr(out, make_nonrelative_path(path));
+
+		strbuf_addch(out, PATH_SEP);
+	}
+}
+
+void setup_path(void)
+{
+	const char *old_path = getenv("PATH");
+	struct strbuf new_path = STRBUF_INIT;
+
+	add_path(&new_path, perf_exec_path());
+	add_path(&new_path, argv0_path);
+
+	if (old_path)
+		strbuf_addstr(&new_path, old_path);
+	else
+		strbuf_addstr(&new_path, "/usr/local/bin:/usr/bin:/bin");
+
+	setenv("PATH", new_path.buf, 1);
+
+	strbuf_release(&new_path);
+}
+
+static const char **prepare_perf_cmd(const char **argv)
+{
+	int argc;
+	const char **nargv;
+
+	for (argc = 0; argv[argc]; argc++)
+		; /* just counting */
+	nargv = malloc(sizeof(*nargv) * (argc + 2));
+
+	nargv[0] = "perf";
+	for (argc = 0; argv[argc]; argc++)
+		nargv[argc + 1] = argv[argc];
+	nargv[argc + 1] = NULL;
+	return nargv;
+}
+
+int execv_perf_cmd(const char **argv) {
+	const char **nargv = prepare_perf_cmd(argv);
+
+	/* execvp() can only ever return if it fails */
+	execvp("perf", (char **)nargv);
+
+	free(nargv);
+	return -1;
+}
+
+
+int execl_perf_cmd(const char *cmd,...)
+{
+	int argc;
+	const char *argv[MAX_ARGS + 1];
+	const char *arg;
+	va_list param;
+
+	va_start(param, cmd);
+	argv[0] = cmd;
+	argc = 1;
+	while (argc < MAX_ARGS) {
+		arg = argv[argc++] = va_arg(param, char *);
+		if (!arg)
+			break;
+	}
+	va_end(param);
+	if (MAX_ARGS <= argc)
+		return error("too many args to run %s", cmd);
+
+	argv[argc] = NULL;
+	return execv_perf_cmd(argv);
+}
diff --git a/tools/perf/util/exec_cmd.h b/tools/perf/util/exec_cmd.h
new file mode 100644
index 00000000..bc4b9159
--- /dev/null
+++ b/tools/perf/util/exec_cmd.h
@@ -0,0 +1,12 @@
+#ifndef __PERF_EXEC_CMD_H
+#define __PERF_EXEC_CMD_H
+
+extern void perf_set_argv_exec_path(const char *exec_path);
+extern const char *perf_extract_argv0_path(const char *path);
+extern const char *perf_exec_path(void);
+extern void setup_path(void);
+extern int execv_perf_cmd(const char **argv); /* NULL terminated */
+extern int execl_perf_cmd(const char *cmd, ...);
+extern const char *system_path(const char *path);
+
+#endif /* __PERF_EXEC_CMD_H */
diff --git a/tools/perf/util/generate-cmdlist.sh b/tools/perf/util/generate-cmdlist.sh
new file mode 100755
index 00000000..f06f6fd1
--- /dev/null
+++ b/tools/perf/util/generate-cmdlist.sh
@@ -0,0 +1,24 @@
+#!/bin/sh
+
+echo "/* Automatically generated by $0 */
+struct cmdname_help
+{
+    char name[16];
+    char help[80];
+};
+
+static struct cmdname_help common_cmds[] = {"
+
+sed -n -e 's/^perf-\([^ 	]*\)[ 	].* common.*/\1/p' command-list.txt |
+sort |
+while read cmd
+do
+     sed -n '
+     /^NAME/,/perf-'"$cmd"'/H
+     ${
+            x
+            s/.*perf-'"$cmd"' - \(.*\)/  {"'"$cmd"'", "\1"},/
+	    p
+     }' "Documentation/perf-$cmd.txt"
+done
+echo "};"
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
new file mode 100644
index 00000000..cb2959a3
--- /dev/null
+++ b/tools/perf/util/header.c
@@ -0,0 +1,1185 @@
+#define _FILE_OFFSET_BITS 64
+
+#include <sys/types.h>
+#include <byteswap.h>
+#include <unistd.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <linux/list.h>
+#include <linux/kernel.h>
+
+#include "evlist.h"
+#include "evsel.h"
+#include "util.h"
+#include "header.h"
+#include "../perf.h"
+#include "trace-event.h"
+#include "session.h"
+#include "symbol.h"
+#include "debug.h"
+
+static bool no_buildid_cache = false;
+
+static int event_count;
+static struct perf_trace_event_type *events;
+
+int perf_header__push_event(u64 id, const char *name)
+{
+	if (strlen(name) > MAX_EVENT_NAME)
+		pr_warning("Event %s will be truncated\n", name);
+
+	if (!events) {
+		events = malloc(sizeof(struct perf_trace_event_type));
+		if (events == NULL)
+			return -ENOMEM;
+	} else {
+		struct perf_trace_event_type *nevents;
+
+		nevents = realloc(events, (event_count + 1) * sizeof(*events));
+		if (nevents == NULL)
+			return -ENOMEM;
+		events = nevents;
+	}
+	memset(&events[event_count], 0, sizeof(struct perf_trace_event_type));
+	events[event_count].event_id = id;
+	strncpy(events[event_count].name, name, MAX_EVENT_NAME - 1);
+	event_count++;
+	return 0;
+}
+
+char *perf_header__find_event(u64 id)
+{
+	int i;
+	for (i = 0 ; i < event_count; i++) {
+		if (events[i].event_id == id)
+			return events[i].name;
+	}
+	return NULL;
+}
+
+static const char *__perf_magic = "PERFFILE";
+
+#define PERF_MAGIC	(*(u64 *)__perf_magic)
+
+struct perf_file_attr {
+	struct perf_event_attr	attr;
+	struct perf_file_section	ids;
+};
+
+void perf_header__set_feat(struct perf_header *header, int feat)
+{
+	set_bit(feat, header->adds_features);
+}
+
+void perf_header__clear_feat(struct perf_header *header, int feat)
+{
+	clear_bit(feat, header->adds_features);
+}
+
+bool perf_header__has_feat(const struct perf_header *header, int feat)
+{
+	return test_bit(feat, header->adds_features);
+}
+
+static int do_write(int fd, const void *buf, size_t size)
+{
+	while (size) {
+		int ret = write(fd, buf, size);
+
+		if (ret < 0)
+			return -errno;
+
+		size -= ret;
+		buf += ret;
+	}
+
+	return 0;
+}
+
+#define NAME_ALIGN 64
+
+static int write_padded(int fd, const void *bf, size_t count,
+			size_t count_aligned)
+{
+	static const char zero_buf[NAME_ALIGN];
+	int err = do_write(fd, bf, count);
+
+	if (!err)
+		err = do_write(fd, zero_buf, count_aligned - count);
+
+	return err;
+}
+
+#define dsos__for_each_with_build_id(pos, head)	\
+	list_for_each_entry(pos, head, node)	\
+		if (!pos->has_build_id)		\
+			continue;		\
+		else
+
+static int __dsos__write_buildid_table(struct list_head *head, pid_t pid,
+				u16 misc, int fd)
+{
+	struct dso *pos;
+
+	dsos__for_each_with_build_id(pos, head) {
+		int err;
+		struct build_id_event b;
+		size_t len;
+
+		if (!pos->hit)
+			continue;
+		len = pos->long_name_len + 1;
+		len = ALIGN(len, NAME_ALIGN);
+		memset(&b, 0, sizeof(b));
+		memcpy(&b.build_id, pos->build_id, sizeof(pos->build_id));
+		b.pid = pid;
+		b.header.misc = misc;
+		b.header.size = sizeof(b) + len;
+		err = do_write(fd, &b, sizeof(b));
+		if (err < 0)
+			return err;
+		err = write_padded(fd, pos->long_name,
+				   pos->long_name_len + 1, len);
+		if (err < 0)
+			return err;
+	}
+
+	return 0;
+}
+
+static int machine__write_buildid_table(struct machine *machine, int fd)
+{
+	int err;
+	u16 kmisc = PERF_RECORD_MISC_KERNEL,
+	    umisc = PERF_RECORD_MISC_USER;
+
+	if (!machine__is_host(machine)) {
+		kmisc = PERF_RECORD_MISC_GUEST_KERNEL;
+		umisc = PERF_RECORD_MISC_GUEST_USER;
+	}
+
+	err = __dsos__write_buildid_table(&machine->kernel_dsos, machine->pid,
+					  kmisc, fd);
+	if (err == 0)
+		err = __dsos__write_buildid_table(&machine->user_dsos,
+						  machine->pid, umisc, fd);
+	return err;
+}
+
+static int dsos__write_buildid_table(struct perf_header *header, int fd)
+{
+	struct perf_session *session = container_of(header,
+			struct perf_session, header);
+	struct rb_node *nd;
+	int err = machine__write_buildid_table(&session->host_machine, fd);
+
+	if (err)
+		return err;
+
+	for (nd = rb_first(&session->machines); nd; nd = rb_next(nd)) {
+		struct machine *pos = rb_entry(nd, struct machine, rb_node);
+		err = machine__write_buildid_table(pos, fd);
+		if (err)
+			break;
+	}
+	return err;
+}
+
+int build_id_cache__add_s(const char *sbuild_id, const char *debugdir,
+			  const char *name, bool is_kallsyms)
+{
+	const size_t size = PATH_MAX;
+	char *realname, *filename = malloc(size),
+	     *linkname = malloc(size), *targetname;
+	int len, err = -1;
+
+	if (is_kallsyms) {
+		if (symbol_conf.kptr_restrict) {
+			pr_debug("Not caching a kptr_restrict'ed /proc/kallsyms\n");
+			return 0;
+		}
+		realname = (char *)name;
+	} else
+		realname = realpath(name, NULL);
+
+	if (realname == NULL || filename == NULL || linkname == NULL)
+		goto out_free;
+
+	len = snprintf(filename, size, "%s%s%s",
+		       debugdir, is_kallsyms ? "/" : "", realname);
+	if (mkdir_p(filename, 0755))
+		goto out_free;
+
+	snprintf(filename + len, sizeof(filename) - len, "/%s", sbuild_id);
+
+	if (access(filename, F_OK)) {
+		if (is_kallsyms) {
+			 if (copyfile("/proc/kallsyms", filename))
+				goto out_free;
+		} else if (link(realname, filename) && copyfile(name, filename))
+			goto out_free;
+	}
+
+	len = snprintf(linkname, size, "%s/.build-id/%.2s",
+		       debugdir, sbuild_id);
+
+	if (access(linkname, X_OK) && mkdir_p(linkname, 0755))
+		goto out_free;
+
+	snprintf(linkname + len, size - len, "/%s", sbuild_id + 2);
+	targetname = filename + strlen(debugdir) - 5;
+	memcpy(targetname, "../..", 5);
+
+	if (symlink(targetname, linkname) == 0)
+		err = 0;
+out_free:
+	if (!is_kallsyms)
+		free(realname);
+	free(filename);
+	free(linkname);
+	return err;
+}
+
+static int build_id_cache__add_b(const u8 *build_id, size_t build_id_size,
+				 const char *name, const char *debugdir,
+				 bool is_kallsyms)
+{
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+	build_id__sprintf(build_id, build_id_size, sbuild_id);
+
+	return build_id_cache__add_s(sbuild_id, debugdir, name, is_kallsyms);
+}
+
+int build_id_cache__remove_s(const char *sbuild_id, const char *debugdir)
+{
+	const size_t size = PATH_MAX;
+	char *filename = malloc(size),
+	     *linkname = malloc(size);
+	int err = -1;
+
+	if (filename == NULL || linkname == NULL)
+		goto out_free;
+
+	snprintf(linkname, size, "%s/.build-id/%.2s/%s",
+		 debugdir, sbuild_id, sbuild_id + 2);
+
+	if (access(linkname, F_OK))
+		goto out_free;
+
+	if (readlink(linkname, filename, size) < 0)
+		goto out_free;
+
+	if (unlink(linkname))
+		goto out_free;
+
+	/*
+	 * Since the link is relative, we must make it absolute:
+	 */
+	snprintf(linkname, size, "%s/.build-id/%.2s/%s",
+		 debugdir, sbuild_id, filename);
+
+	if (unlink(linkname))
+		goto out_free;
+
+	err = 0;
+out_free:
+	free(filename);
+	free(linkname);
+	return err;
+}
+
+static int dso__cache_build_id(struct dso *dso, const char *debugdir)
+{
+	bool is_kallsyms = dso->kernel && dso->long_name[0] != '/';
+
+	return build_id_cache__add_b(dso->build_id, sizeof(dso->build_id),
+				     dso->long_name, debugdir, is_kallsyms);
+}
+
+static int __dsos__cache_build_ids(struct list_head *head, const char *debugdir)
+{
+	struct dso *pos;
+	int err = 0;
+
+	dsos__for_each_with_build_id(pos, head)
+		if (dso__cache_build_id(pos, debugdir))
+			err = -1;
+
+	return err;
+}
+
+static int machine__cache_build_ids(struct machine *machine, const char *debugdir)
+{
+	int ret = __dsos__cache_build_ids(&machine->kernel_dsos, debugdir);
+	ret |= __dsos__cache_build_ids(&machine->user_dsos, debugdir);
+	return ret;
+}
+
+static int perf_session__cache_build_ids(struct perf_session *session)
+{
+	struct rb_node *nd;
+	int ret;
+	char debugdir[PATH_MAX];
+
+	snprintf(debugdir, sizeof(debugdir), "%s", buildid_dir);
+
+	if (mkdir(debugdir, 0755) != 0 && errno != EEXIST)
+		return -1;
+
+	ret = machine__cache_build_ids(&session->host_machine, debugdir);
+
+	for (nd = rb_first(&session->machines); nd; nd = rb_next(nd)) {
+		struct machine *pos = rb_entry(nd, struct machine, rb_node);
+		ret |= machine__cache_build_ids(pos, debugdir);
+	}
+	return ret ? -1 : 0;
+}
+
+static bool machine__read_build_ids(struct machine *machine, bool with_hits)
+{
+	bool ret = __dsos__read_build_ids(&machine->kernel_dsos, with_hits);
+	ret |= __dsos__read_build_ids(&machine->user_dsos, with_hits);
+	return ret;
+}
+
+static bool perf_session__read_build_ids(struct perf_session *session, bool with_hits)
+{
+	struct rb_node *nd;
+	bool ret = machine__read_build_ids(&session->host_machine, with_hits);
+
+	for (nd = rb_first(&session->machines); nd; nd = rb_next(nd)) {
+		struct machine *pos = rb_entry(nd, struct machine, rb_node);
+		ret |= machine__read_build_ids(pos, with_hits);
+	}
+
+	return ret;
+}
+
+static int perf_header__adds_write(struct perf_header *header,
+				   struct perf_evlist *evlist, int fd)
+{
+	int nr_sections;
+	struct perf_session *session;
+	struct perf_file_section *feat_sec;
+	int sec_size;
+	u64 sec_start;
+	int idx = 0, err;
+
+	session = container_of(header, struct perf_session, header);
+
+	if (perf_header__has_feat(header, HEADER_BUILD_ID &&
+	    !perf_session__read_build_ids(session, true)))
+		perf_header__clear_feat(header, HEADER_BUILD_ID);
+
+	nr_sections = bitmap_weight(header->adds_features, HEADER_FEAT_BITS);
+	if (!nr_sections)
+		return 0;
+
+	feat_sec = calloc(sizeof(*feat_sec), nr_sections);
+	if (feat_sec == NULL)
+		return -ENOMEM;
+
+	sec_size = sizeof(*feat_sec) * nr_sections;
+
+	sec_start = header->data_offset + header->data_size;
+	lseek(fd, sec_start + sec_size, SEEK_SET);
+
+	if (perf_header__has_feat(header, HEADER_TRACE_INFO)) {
+		struct perf_file_section *trace_sec;
+
+		trace_sec = &feat_sec[idx++];
+
+		/* Write trace info */
+		trace_sec->offset = lseek(fd, 0, SEEK_CUR);
+		read_tracing_data(fd, &evlist->entries);
+		trace_sec->size = lseek(fd, 0, SEEK_CUR) - trace_sec->offset;
+	}
+
+	if (perf_header__has_feat(header, HEADER_BUILD_ID)) {
+		struct perf_file_section *buildid_sec;
+
+		buildid_sec = &feat_sec[idx++];
+
+		/* Write build-ids */
+		buildid_sec->offset = lseek(fd, 0, SEEK_CUR);
+		err = dsos__write_buildid_table(header, fd);
+		if (err < 0) {
+			pr_debug("failed to write buildid table\n");
+			goto out_free;
+		}
+		buildid_sec->size = lseek(fd, 0, SEEK_CUR) -
+					  buildid_sec->offset;
+		if (!no_buildid_cache)
+			perf_session__cache_build_ids(session);
+	}
+
+	lseek(fd, sec_start, SEEK_SET);
+	err = do_write(fd, feat_sec, sec_size);
+	if (err < 0)
+		pr_debug("failed to write feature section\n");
+out_free:
+	free(feat_sec);
+	return err;
+}
+
+int perf_header__write_pipe(int fd)
+{
+	struct perf_pipe_file_header f_header;
+	int err;
+
+	f_header = (struct perf_pipe_file_header){
+		.magic	   = PERF_MAGIC,
+		.size	   = sizeof(f_header),
+	};
+
+	err = do_write(fd, &f_header, sizeof(f_header));
+	if (err < 0) {
+		pr_debug("failed to write perf pipe header\n");
+		return err;
+	}
+
+	return 0;
+}
+
+int perf_session__write_header(struct perf_session *session,
+			       struct perf_evlist *evlist,
+			       int fd, bool at_exit)
+{
+	struct perf_file_header f_header;
+	struct perf_file_attr   f_attr;
+	struct perf_header *header = &session->header;
+	struct perf_evsel *attr, *pair = NULL;
+	int err;
+
+	lseek(fd, sizeof(f_header), SEEK_SET);
+
+	if (session->evlist != evlist)
+		pair = list_entry(session->evlist->entries.next, struct perf_evsel, node);
+
+	list_for_each_entry(attr, &evlist->entries, node) {
+		attr->id_offset = lseek(fd, 0, SEEK_CUR);
+		err = do_write(fd, attr->id, attr->ids * sizeof(u64));
+		if (err < 0) {
+out_err_write:
+			pr_debug("failed to write perf header\n");
+			return err;
+		}
+		if (session->evlist != evlist) {
+			err = do_write(fd, pair->id, pair->ids * sizeof(u64));
+			if (err < 0)
+				goto out_err_write;
+			attr->ids += pair->ids;
+			pair = list_entry(pair->node.next, struct perf_evsel, node);
+		}
+	}
+
+	header->attr_offset = lseek(fd, 0, SEEK_CUR);
+
+	list_for_each_entry(attr, &evlist->entries, node) {
+		f_attr = (struct perf_file_attr){
+			.attr = attr->attr,
+			.ids  = {
+				.offset = attr->id_offset,
+				.size   = attr->ids * sizeof(u64),
+			}
+		};
+		err = do_write(fd, &f_attr, sizeof(f_attr));
+		if (err < 0) {
+			pr_debug("failed to write perf header attribute\n");
+			return err;
+		}
+	}
+
+	header->event_offset = lseek(fd, 0, SEEK_CUR);
+	header->event_size = event_count * sizeof(struct perf_trace_event_type);
+	if (events) {
+		err = do_write(fd, events, header->event_size);
+		if (err < 0) {
+			pr_debug("failed to write perf header events\n");
+			return err;
+		}
+	}
+
+	header->data_offset = lseek(fd, 0, SEEK_CUR);
+
+	if (at_exit) {
+		err = perf_header__adds_write(header, evlist, fd);
+		if (err < 0)
+			return err;
+	}
+
+	f_header = (struct perf_file_header){
+		.magic	   = PERF_MAGIC,
+		.size	   = sizeof(f_header),
+		.attr_size = sizeof(f_attr),
+		.attrs = {
+			.offset = header->attr_offset,
+			.size   = evlist->nr_entries * sizeof(f_attr),
+		},
+		.data = {
+			.offset = header->data_offset,
+			.size	= header->data_size,
+		},
+		.event_types = {
+			.offset = header->event_offset,
+			.size	= header->event_size,
+		},
+	};
+
+	memcpy(&f_header.adds_features, &header->adds_features, sizeof(header->adds_features));
+
+	lseek(fd, 0, SEEK_SET);
+	err = do_write(fd, &f_header, sizeof(f_header));
+	if (err < 0) {
+		pr_debug("failed to write perf header\n");
+		return err;
+	}
+	lseek(fd, header->data_offset + header->data_size, SEEK_SET);
+
+	header->frozen = 1;
+	return 0;
+}
+
+static int perf_header__getbuffer64(struct perf_header *header,
+				    int fd, void *buf, size_t size)
+{
+	if (readn(fd, buf, size) <= 0)
+		return -1;
+
+	if (header->needs_swap)
+		mem_bswap_64(buf, size);
+
+	return 0;
+}
+
+int perf_header__process_sections(struct perf_header *header, int fd,
+				  int (*process)(struct perf_file_section *section,
+						 struct perf_header *ph,
+						 int feat, int fd))
+{
+	struct perf_file_section *feat_sec;
+	int nr_sections;
+	int sec_size;
+	int idx = 0;
+	int err = -1, feat = 1;
+
+	nr_sections = bitmap_weight(header->adds_features, HEADER_FEAT_BITS);
+	if (!nr_sections)
+		return 0;
+
+	feat_sec = calloc(sizeof(*feat_sec), nr_sections);
+	if (!feat_sec)
+		return -1;
+
+	sec_size = sizeof(*feat_sec) * nr_sections;
+
+	lseek(fd, header->data_offset + header->data_size, SEEK_SET);
+
+	if (perf_header__getbuffer64(header, fd, feat_sec, sec_size))
+		goto out_free;
+
+	err = 0;
+	while (idx < nr_sections && feat < HEADER_LAST_FEATURE) {
+		if (perf_header__has_feat(header, feat)) {
+			struct perf_file_section *sec = &feat_sec[idx++];
+
+			err = process(sec, header, feat, fd);
+			if (err < 0)
+				break;
+		}
+		++feat;
+	}
+out_free:
+	free(feat_sec);
+	return err;
+}
+
+int perf_file_header__read(struct perf_file_header *header,
+			   struct perf_header *ph, int fd)
+{
+	lseek(fd, 0, SEEK_SET);
+
+	if (readn(fd, header, sizeof(*header)) <= 0 ||
+	    memcmp(&header->magic, __perf_magic, sizeof(header->magic)))
+		return -1;
+
+	if (header->attr_size != sizeof(struct perf_file_attr)) {
+		u64 attr_size = bswap_64(header->attr_size);
+
+		if (attr_size != sizeof(struct perf_file_attr))
+			return -1;
+
+		mem_bswap_64(header, offsetof(struct perf_file_header,
+					    adds_features));
+		ph->needs_swap = true;
+	}
+
+	if (header->size != sizeof(*header)) {
+		/* Support the previous format */
+		if (header->size == offsetof(typeof(*header), adds_features))
+			bitmap_zero(header->adds_features, HEADER_FEAT_BITS);
+		else
+			return -1;
+	}
+
+	memcpy(&ph->adds_features, &header->adds_features,
+	       sizeof(ph->adds_features));
+	/*
+	 * FIXME: hack that assumes that if we need swap the perf.data file
+	 * may be coming from an arch with a different word-size, ergo different
+	 * DEFINE_BITMAP format, investigate more later, but for now its mostly
+	 * safe to assume that we have a build-id section. Trace files probably
+	 * have several other issues in this realm anyway...
+	 */
+	if (ph->needs_swap) {
+		memset(&ph->adds_features, 0, sizeof(ph->adds_features));
+		perf_header__set_feat(ph, HEADER_BUILD_ID);
+	}
+
+	ph->event_offset = header->event_types.offset;
+	ph->event_size   = header->event_types.size;
+	ph->data_offset  = header->data.offset;
+	ph->data_size	 = header->data.size;
+	return 0;
+}
+
+static int __event_process_build_id(struct build_id_event *bev,
+				    char *filename,
+				    struct perf_session *session)
+{
+	int err = -1;
+	struct list_head *head;
+	struct machine *machine;
+	u16 misc;
+	struct dso *dso;
+	enum dso_kernel_type dso_type;
+
+	machine = perf_session__findnew_machine(session, bev->pid);
+	if (!machine)
+		goto out;
+
+	misc = bev->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+
+	switch (misc) {
+	case PERF_RECORD_MISC_KERNEL:
+		dso_type = DSO_TYPE_KERNEL;
+		head = &machine->kernel_dsos;
+		break;
+	case PERF_RECORD_MISC_GUEST_KERNEL:
+		dso_type = DSO_TYPE_GUEST_KERNEL;
+		head = &machine->kernel_dsos;
+		break;
+	case PERF_RECORD_MISC_USER:
+	case PERF_RECORD_MISC_GUEST_USER:
+		dso_type = DSO_TYPE_USER;
+		head = &machine->user_dsos;
+		break;
+	default:
+		goto out;
+	}
+
+	dso = __dsos__findnew(head, filename);
+	if (dso != NULL) {
+		char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+		dso__set_build_id(dso, &bev->build_id);
+
+		if (filename[0] == '[')
+			dso->kernel = dso_type;
+
+		build_id__sprintf(dso->build_id, sizeof(dso->build_id),
+				  sbuild_id);
+		pr_debug("build id event received for %s: %s\n",
+			 dso->long_name, sbuild_id);
+	}
+
+	err = 0;
+out:
+	return err;
+}
+
+static int perf_header__read_build_ids_abi_quirk(struct perf_header *header,
+						 int input, u64 offset, u64 size)
+{
+	struct perf_session *session = container_of(header, struct perf_session, header);
+	struct {
+		struct perf_event_header   header;
+		u8			   build_id[ALIGN(BUILD_ID_SIZE, sizeof(u64))];
+		char			   filename[0];
+	} old_bev;
+	struct build_id_event bev;
+	char filename[PATH_MAX];
+	u64 limit = offset + size;
+
+	while (offset < limit) {
+		ssize_t len;
+
+		if (read(input, &old_bev, sizeof(old_bev)) != sizeof(old_bev))
+			return -1;
+
+		if (header->needs_swap)
+			perf_event_header__bswap(&old_bev.header);
+
+		len = old_bev.header.size - sizeof(old_bev);
+		if (read(input, filename, len) != len)
+			return -1;
+
+		bev.header = old_bev.header;
+		bev.pid	   = 0;
+		memcpy(bev.build_id, old_bev.build_id, sizeof(bev.build_id));
+		__event_process_build_id(&bev, filename, session);
+
+		offset += bev.header.size;
+	}
+
+	return 0;
+}
+
+static int perf_header__read_build_ids(struct perf_header *header,
+				       int input, u64 offset, u64 size)
+{
+	struct perf_session *session = container_of(header, struct perf_session, header);
+	struct build_id_event bev;
+	char filename[PATH_MAX];
+	u64 limit = offset + size, orig_offset = offset;
+	int err = -1;
+
+	while (offset < limit) {
+		ssize_t len;
+
+		if (read(input, &bev, sizeof(bev)) != sizeof(bev))
+			goto out;
+
+		if (header->needs_swap)
+			perf_event_header__bswap(&bev.header);
+
+		len = bev.header.size - sizeof(bev);
+		if (read(input, filename, len) != len)
+			goto out;
+		/*
+		 * The a1645ce1 changeset:
+		 *
+		 * "perf: 'perf kvm' tool for monitoring guest performance from host"
+		 *
+		 * Added a field to struct build_id_event that broke the file
+		 * format.
+		 *
+		 * Since the kernel build-id is the first entry, process the
+		 * table using the old format if the well known
+		 * '[kernel.kallsyms]' string for the kernel build-id has the
+		 * first 4 characters chopped off (where the pid_t sits).
+		 */
+		if (memcmp(filename, "nel.kallsyms]", 13) == 0) {
+			if (lseek(input, orig_offset, SEEK_SET) == (off_t)-1)
+				return -1;
+			return perf_header__read_build_ids_abi_quirk(header, input, offset, size);
+		}
+
+		__event_process_build_id(&bev, filename, session);
+
+		offset += bev.header.size;
+	}
+	err = 0;
+out:
+	return err;
+}
+
+static int perf_file_section__process(struct perf_file_section *section,
+				      struct perf_header *ph,
+				      int feat, int fd)
+{
+	if (lseek(fd, section->offset, SEEK_SET) == (off_t)-1) {
+		pr_debug("Failed to lseek to %" PRIu64 " offset for feature "
+			  "%d, continuing...\n", section->offset, feat);
+		return 0;
+	}
+
+	switch (feat) {
+	case HEADER_TRACE_INFO:
+		trace_report(fd, false);
+		break;
+
+	case HEADER_BUILD_ID:
+		if (perf_header__read_build_ids(ph, fd, section->offset, section->size))
+			pr_debug("Failed to read buildids, continuing...\n");
+		break;
+	default:
+		pr_debug("unknown feature %d, continuing...\n", feat);
+	}
+
+	return 0;
+}
+
+static int perf_file_header__read_pipe(struct perf_pipe_file_header *header,
+				       struct perf_header *ph, int fd,
+				       bool repipe)
+{
+	if (readn(fd, header, sizeof(*header)) <= 0 ||
+	    memcmp(&header->magic, __perf_magic, sizeof(header->magic)))
+		return -1;
+
+	if (repipe && do_write(STDOUT_FILENO, header, sizeof(*header)) < 0)
+		return -1;
+
+	if (header->size != sizeof(*header)) {
+		u64 size = bswap_64(header->size);
+
+		if (size != sizeof(*header))
+			return -1;
+
+		ph->needs_swap = true;
+	}
+
+	return 0;
+}
+
+static int perf_header__read_pipe(struct perf_session *session, int fd)
+{
+	struct perf_header *header = &session->header;
+	struct perf_pipe_file_header f_header;
+
+	if (perf_file_header__read_pipe(&f_header, header, fd,
+					session->repipe) < 0) {
+		pr_debug("incompatible file format\n");
+		return -EINVAL;
+	}
+
+	session->fd = fd;
+
+	return 0;
+}
+
+int perf_session__read_header(struct perf_session *session, int fd)
+{
+	struct perf_header *header = &session->header;
+	struct perf_file_header	f_header;
+	struct perf_file_attr	f_attr;
+	u64			f_id;
+	int nr_attrs, nr_ids, i, j;
+
+	session->evlist = perf_evlist__new(NULL, NULL);
+	if (session->evlist == NULL)
+		return -ENOMEM;
+
+	if (session->fd_pipe)
+		return perf_header__read_pipe(session, fd);
+
+	if (perf_file_header__read(&f_header, header, fd) < 0) {
+		pr_debug("incompatible file format\n");
+		return -EINVAL;
+	}
+
+	nr_attrs = f_header.attrs.size / sizeof(f_attr);
+	lseek(fd, f_header.attrs.offset, SEEK_SET);
+
+	for (i = 0; i < nr_attrs; i++) {
+		struct perf_evsel *evsel;
+		off_t tmp;
+
+		if (readn(fd, &f_attr, sizeof(f_attr)) <= 0)
+			goto out_errno;
+
+		if (header->needs_swap)
+			perf_event__attr_swap(&f_attr.attr);
+
+		tmp = lseek(fd, 0, SEEK_CUR);
+		evsel = perf_evsel__new(&f_attr.attr, i);
+
+		if (evsel == NULL)
+			goto out_delete_evlist;
+		/*
+		 * Do it before so that if perf_evsel__alloc_id fails, this
+		 * entry gets purged too at perf_evlist__delete().
+		 */
+		perf_evlist__add(session->evlist, evsel);
+
+		nr_ids = f_attr.ids.size / sizeof(u64);
+		/*
+		 * We don't have the cpu and thread maps on the header, so
+		 * for allocating the perf_sample_id table we fake 1 cpu and
+		 * hattr->ids threads.
+		 */
+		if (perf_evsel__alloc_id(evsel, 1, nr_ids))
+			goto out_delete_evlist;
+
+		lseek(fd, f_attr.ids.offset, SEEK_SET);
+
+		for (j = 0; j < nr_ids; j++) {
+			if (perf_header__getbuffer64(header, fd, &f_id, sizeof(f_id)))
+				goto out_errno;
+
+			perf_evlist__id_add(session->evlist, evsel, 0, j, f_id);
+		}
+
+		lseek(fd, tmp, SEEK_SET);
+	}
+
+	if (f_header.event_types.size) {
+		lseek(fd, f_header.event_types.offset, SEEK_SET);
+		events = malloc(f_header.event_types.size);
+		if (events == NULL)
+			return -ENOMEM;
+		if (perf_header__getbuffer64(header, fd, events,
+					     f_header.event_types.size))
+			goto out_errno;
+		event_count =  f_header.event_types.size / sizeof(struct perf_trace_event_type);
+	}
+
+	perf_header__process_sections(header, fd, perf_file_section__process);
+
+	lseek(fd, header->data_offset, SEEK_SET);
+
+	header->frozen = 1;
+	return 0;
+out_errno:
+	return -errno;
+
+out_delete_evlist:
+	perf_evlist__delete(session->evlist);
+	session->evlist = NULL;
+	return -ENOMEM;
+}
+
+int perf_event__synthesize_attr(struct perf_event_attr *attr, u16 ids, u64 *id,
+				perf_event__handler_t process,
+				struct perf_session *session)
+{
+	union perf_event *ev;
+	size_t size;
+	int err;
+
+	size = sizeof(struct perf_event_attr);
+	size = ALIGN(size, sizeof(u64));
+	size += sizeof(struct perf_event_header);
+	size += ids * sizeof(u64);
+
+	ev = malloc(size);
+
+	if (ev == NULL)
+		return -ENOMEM;
+
+	ev->attr.attr = *attr;
+	memcpy(ev->attr.id, id, ids * sizeof(u64));
+
+	ev->attr.header.type = PERF_RECORD_HEADER_ATTR;
+	ev->attr.header.size = size;
+
+	err = process(ev, NULL, session);
+
+	free(ev);
+
+	return err;
+}
+
+int perf_session__synthesize_attrs(struct perf_session *session,
+				   perf_event__handler_t process)
+{
+	struct perf_evsel *attr;
+	int err = 0;
+
+	list_for_each_entry(attr, &session->evlist->entries, node) {
+		err = perf_event__synthesize_attr(&attr->attr, attr->ids,
+						  attr->id, process, session);
+		if (err) {
+			pr_debug("failed to create perf header attribute\n");
+			return err;
+		}
+	}
+
+	return err;
+}
+
+int perf_event__process_attr(union perf_event *event,
+			     struct perf_session *session)
+{
+	unsigned int i, ids, n_ids;
+	struct perf_evsel *evsel;
+
+	if (session->evlist == NULL) {
+		session->evlist = perf_evlist__new(NULL, NULL);
+		if (session->evlist == NULL)
+			return -ENOMEM;
+	}
+
+	evsel = perf_evsel__new(&event->attr.attr,
+				session->evlist->nr_entries);
+	if (evsel == NULL)
+		return -ENOMEM;
+
+	perf_evlist__add(session->evlist, evsel);
+
+	ids = event->header.size;
+	ids -= (void *)&event->attr.id - (void *)event;
+	n_ids = ids / sizeof(u64);
+	/*
+	 * We don't have the cpu and thread maps on the header, so
+	 * for allocating the perf_sample_id table we fake 1 cpu and
+	 * hattr->ids threads.
+	 */
+	if (perf_evsel__alloc_id(evsel, 1, n_ids))
+		return -ENOMEM;
+
+	for (i = 0; i < n_ids; i++) {
+		perf_evlist__id_add(session->evlist, evsel, 0, i,
+				    event->attr.id[i]);
+	}
+
+	perf_session__update_sample_type(session);
+
+	return 0;
+}
+
+int perf_event__synthesize_event_type(u64 event_id, char *name,
+				      perf_event__handler_t process,
+				      struct perf_session *session)
+{
+	union perf_event ev;
+	size_t size = 0;
+	int err = 0;
+
+	memset(&ev, 0, sizeof(ev));
+
+	ev.event_type.event_type.event_id = event_id;
+	memset(ev.event_type.event_type.name, 0, MAX_EVENT_NAME);
+	strncpy(ev.event_type.event_type.name, name, MAX_EVENT_NAME - 1);
+
+	ev.event_type.header.type = PERF_RECORD_HEADER_EVENT_TYPE;
+	size = strlen(name);
+	size = ALIGN(size, sizeof(u64));
+	ev.event_type.header.size = sizeof(ev.event_type) -
+		(sizeof(ev.event_type.event_type.name) - size);
+
+	err = process(&ev, NULL, session);
+
+	return err;
+}
+
+int perf_event__synthesize_event_types(perf_event__handler_t process,
+				       struct perf_session *session)
+{
+	struct perf_trace_event_type *type;
+	int i, err = 0;
+
+	for (i = 0; i < event_count; i++) {
+		type = &events[i];
+
+		err = perf_event__synthesize_event_type(type->event_id,
+							type->name, process,
+							session);
+		if (err) {
+			pr_debug("failed to create perf header event type\n");
+			return err;
+		}
+	}
+
+	return err;
+}
+
+int perf_event__process_event_type(union perf_event *event,
+				   struct perf_session *session __unused)
+{
+	if (perf_header__push_event(event->event_type.event_type.event_id,
+				    event->event_type.event_type.name) < 0)
+		return -ENOMEM;
+
+	return 0;
+}
+
+int perf_event__synthesize_tracing_data(int fd, struct perf_evlist *evlist,
+					 perf_event__handler_t process,
+				   struct perf_session *session __unused)
+{
+	union perf_event ev;
+	ssize_t size = 0, aligned_size = 0, padding;
+	int err __used = 0;
+
+	memset(&ev, 0, sizeof(ev));
+
+	ev.tracing_data.header.type = PERF_RECORD_HEADER_TRACING_DATA;
+	size = read_tracing_data_size(fd, &evlist->entries);
+	if (size <= 0)
+		return size;
+	aligned_size = ALIGN(size, sizeof(u64));
+	padding = aligned_size - size;
+	ev.tracing_data.header.size = sizeof(ev.tracing_data);
+	ev.tracing_data.size = aligned_size;
+
+	process(&ev, NULL, session);
+
+	err = read_tracing_data(fd, &evlist->entries);
+	write_padded(fd, NULL, 0, padding);
+
+	return aligned_size;
+}
+
+int perf_event__process_tracing_data(union perf_event *event,
+				     struct perf_session *session)
+{
+	ssize_t size_read, padding, size = event->tracing_data.size;
+	off_t offset = lseek(session->fd, 0, SEEK_CUR);
+	char buf[BUFSIZ];
+
+	/* setup for reading amidst mmap */
+	lseek(session->fd, offset + sizeof(struct tracing_data_event),
+	      SEEK_SET);
+
+	size_read = trace_report(session->fd, session->repipe);
+
+	padding = ALIGN(size_read, sizeof(u64)) - size_read;
+
+	if (read(session->fd, buf, padding) < 0)
+		die("reading input file");
+	if (session->repipe) {
+		int retw = write(STDOUT_FILENO, buf, padding);
+		if (retw <= 0 || retw != padding)
+			die("repiping tracing data padding");
+	}
+
+	if (size_read + padding != size)
+		die("tracing data size mismatch");
+
+	return size_read + padding;
+}
+
+int perf_event__synthesize_build_id(struct dso *pos, u16 misc,
+				    perf_event__handler_t process,
+				    struct machine *machine,
+				    struct perf_session *session)
+{
+	union perf_event ev;
+	size_t len;
+	int err = 0;
+
+	if (!pos->hit)
+		return err;
+
+	memset(&ev, 0, sizeof(ev));
+
+	len = pos->long_name_len + 1;
+	len = ALIGN(len, NAME_ALIGN);
+	memcpy(&ev.build_id.build_id, pos->build_id, sizeof(pos->build_id));
+	ev.build_id.header.type = PERF_RECORD_HEADER_BUILD_ID;
+	ev.build_id.header.misc = misc;
+	ev.build_id.pid = machine->pid;
+	ev.build_id.header.size = sizeof(ev.build_id) + len;
+	memcpy(&ev.build_id.filename, pos->long_name, pos->long_name_len);
+
+	err = process(&ev, NULL, session);
+
+	return err;
+}
+
+int perf_event__process_build_id(union perf_event *event,
+				 struct perf_session *session)
+{
+	__event_process_build_id(&event->build_id,
+				 event->build_id.filename,
+				 session);
+	return 0;
+}
+
+void disable_buildid_cache(void)
+{
+	no_buildid_cache = true;
+}
diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h
new file mode 100644
index 00000000..18862567
--- /dev/null
+++ b/tools/perf/util/header.h
@@ -0,0 +1,107 @@
+#ifndef __PERF_HEADER_H
+#define __PERF_HEADER_H
+
+#include "../../../include/linux/perf_event.h"
+#include <sys/types.h>
+#include <stdbool.h>
+#include "types.h"
+#include "event.h"
+
+#include <linux/bitmap.h>
+
+enum {
+	HEADER_TRACE_INFO = 1,
+	HEADER_BUILD_ID,
+	HEADER_LAST_FEATURE,
+};
+
+#define HEADER_FEAT_BITS			256
+
+struct perf_file_section {
+	u64 offset;
+	u64 size;
+};
+
+struct perf_file_header {
+	u64				magic;
+	u64				size;
+	u64				attr_size;
+	struct perf_file_section	attrs;
+	struct perf_file_section	data;
+	struct perf_file_section	event_types;
+	DECLARE_BITMAP(adds_features, HEADER_FEAT_BITS);
+};
+
+struct perf_pipe_file_header {
+	u64				magic;
+	u64				size;
+};
+
+struct perf_header;
+
+int perf_file_header__read(struct perf_file_header *header,
+			   struct perf_header *ph, int fd);
+
+struct perf_header {
+	int			frozen;
+	bool			needs_swap;
+	s64			attr_offset;
+	u64			data_offset;
+	u64			data_size;
+	u64			event_offset;
+	u64			event_size;
+	DECLARE_BITMAP(adds_features, HEADER_FEAT_BITS);
+};
+
+struct perf_evlist;
+
+int perf_session__read_header(struct perf_session *session, int fd);
+int perf_session__write_header(struct perf_session *session,
+			       struct perf_evlist *evlist,
+			       int fd, bool at_exit);
+int perf_header__write_pipe(int fd);
+
+int perf_header__push_event(u64 id, const char *name);
+char *perf_header__find_event(u64 id);
+
+void perf_header__set_feat(struct perf_header *header, int feat);
+void perf_header__clear_feat(struct perf_header *header, int feat);
+bool perf_header__has_feat(const struct perf_header *header, int feat);
+
+int perf_header__process_sections(struct perf_header *header, int fd,
+				  int (*process)(struct perf_file_section *section,
+						 struct perf_header *ph,
+						 int feat, int fd));
+
+int build_id_cache__add_s(const char *sbuild_id, const char *debugdir,
+			  const char *name, bool is_kallsyms);
+int build_id_cache__remove_s(const char *sbuild_id, const char *debugdir);
+
+int perf_event__synthesize_attr(struct perf_event_attr *attr, u16 ids, u64 *id,
+				perf_event__handler_t process,
+				struct perf_session *session);
+int perf_session__synthesize_attrs(struct perf_session *session,
+				   perf_event__handler_t process);
+int perf_event__process_attr(union perf_event *event, struct perf_session *session);
+
+int perf_event__synthesize_event_type(u64 event_id, char *name,
+				      perf_event__handler_t process,
+				      struct perf_session *session);
+int perf_event__synthesize_event_types(perf_event__handler_t process,
+				       struct perf_session *session);
+int perf_event__process_event_type(union perf_event *event,
+				   struct perf_session *session);
+
+int perf_event__synthesize_tracing_data(int fd, struct perf_evlist *evlist,
+					perf_event__handler_t process,
+					struct perf_session *session);
+int perf_event__process_tracing_data(union perf_event *event,
+				     struct perf_session *session);
+
+int perf_event__synthesize_build_id(struct dso *pos, u16 misc,
+				    perf_event__handler_t process,
+				    struct machine *machine,
+				    struct perf_session *session);
+int perf_event__process_build_id(union perf_event *event,
+				 struct perf_session *session);
+#endif /* __PERF_HEADER_H */
diff --git a/tools/perf/util/help.c b/tools/perf/util/help.c
new file mode 100644
index 00000000..6f2975a0
--- /dev/null
+++ b/tools/perf/util/help.c
@@ -0,0 +1,338 @@
+#include "cache.h"
+#include "../builtin.h"
+#include "exec_cmd.h"
+#include "levenshtein.h"
+#include "help.h"
+
+void add_cmdname(struct cmdnames *cmds, const char *name, size_t len)
+{
+	struct cmdname *ent = malloc(sizeof(*ent) + len + 1);
+
+	ent->len = len;
+	memcpy(ent->name, name, len);
+	ent->name[len] = 0;
+
+	ALLOC_GROW(cmds->names, cmds->cnt + 1, cmds->alloc);
+	cmds->names[cmds->cnt++] = ent;
+}
+
+static void clean_cmdnames(struct cmdnames *cmds)
+{
+	unsigned int i;
+
+	for (i = 0; i < cmds->cnt; ++i)
+		free(cmds->names[i]);
+	free(cmds->names);
+	cmds->cnt = 0;
+	cmds->alloc = 0;
+}
+
+static int cmdname_compare(const void *a_, const void *b_)
+{
+	struct cmdname *a = *(struct cmdname **)a_;
+	struct cmdname *b = *(struct cmdname **)b_;
+	return strcmp(a->name, b->name);
+}
+
+static void uniq(struct cmdnames *cmds)
+{
+	unsigned int i, j;
+
+	if (!cmds->cnt)
+		return;
+
+	for (i = j = 1; i < cmds->cnt; i++)
+		if (strcmp(cmds->names[i]->name, cmds->names[i-1]->name))
+			cmds->names[j++] = cmds->names[i];
+
+	cmds->cnt = j;
+}
+
+void exclude_cmds(struct cmdnames *cmds, struct cmdnames *excludes)
+{
+	size_t ci, cj, ei;
+	int cmp;
+
+	ci = cj = ei = 0;
+	while (ci < cmds->cnt && ei < excludes->cnt) {
+		cmp = strcmp(cmds->names[ci]->name, excludes->names[ei]->name);
+		if (cmp < 0)
+			cmds->names[cj++] = cmds->names[ci++];
+		else if (cmp == 0)
+			ci++, ei++;
+		else if (cmp > 0)
+			ei++;
+	}
+
+	while (ci < cmds->cnt)
+		cmds->names[cj++] = cmds->names[ci++];
+
+	cmds->cnt = cj;
+}
+
+static void pretty_print_string_list(struct cmdnames *cmds, int longest)
+{
+	int cols = 1, rows;
+	int space = longest + 1; /* min 1 SP between words */
+	struct winsize win;
+	int max_cols;
+	int i, j;
+
+	get_term_dimensions(&win);
+	max_cols = win.ws_col - 1; /* don't print *on* the edge */
+
+	if (space < max_cols)
+		cols = max_cols / space;
+	rows = (cmds->cnt + cols - 1) / cols;
+
+	for (i = 0; i < rows; i++) {
+		printf("  ");
+
+		for (j = 0; j < cols; j++) {
+			unsigned int n = j * rows + i;
+			unsigned int size = space;
+
+			if (n >= cmds->cnt)
+				break;
+			if (j == cols-1 || n + rows >= cmds->cnt)
+				size = 1;
+			printf("%-*s", size, cmds->names[n]->name);
+		}
+		putchar('\n');
+	}
+}
+
+static int is_executable(const char *name)
+{
+	struct stat st;
+
+	if (stat(name, &st) || /* stat, not lstat */
+	    !S_ISREG(st.st_mode))
+		return 0;
+
+	return st.st_mode & S_IXUSR;
+}
+
+static void list_commands_in_dir(struct cmdnames *cmds,
+					 const char *path,
+					 const char *prefix)
+{
+	int prefix_len;
+	DIR *dir = opendir(path);
+	struct dirent *de;
+	struct strbuf buf = STRBUF_INIT;
+	int len;
+
+	if (!dir)
+		return;
+	if (!prefix)
+		prefix = "perf-";
+	prefix_len = strlen(prefix);
+
+	strbuf_addf(&buf, "%s/", path);
+	len = buf.len;
+
+	while ((de = readdir(dir)) != NULL) {
+		int entlen;
+
+		if (prefixcmp(de->d_name, prefix))
+			continue;
+
+		strbuf_setlen(&buf, len);
+		strbuf_addstr(&buf, de->d_name);
+		if (!is_executable(buf.buf))
+			continue;
+
+		entlen = strlen(de->d_name) - prefix_len;
+		if (has_extension(de->d_name, ".exe"))
+			entlen -= 4;
+
+		add_cmdname(cmds, de->d_name + prefix_len, entlen);
+	}
+	closedir(dir);
+	strbuf_release(&buf);
+}
+
+void load_command_list(const char *prefix,
+		struct cmdnames *main_cmds,
+		struct cmdnames *other_cmds)
+{
+	const char *env_path = getenv("PATH");
+	const char *exec_path = perf_exec_path();
+
+	if (exec_path) {
+		list_commands_in_dir(main_cmds, exec_path, prefix);
+		qsort(main_cmds->names, main_cmds->cnt,
+		      sizeof(*main_cmds->names), cmdname_compare);
+		uniq(main_cmds);
+	}
+
+	if (env_path) {
+		char *paths, *path, *colon;
+		path = paths = strdup(env_path);
+		while (1) {
+			if ((colon = strchr(path, PATH_SEP)))
+				*colon = 0;
+			if (!exec_path || strcmp(path, exec_path))
+				list_commands_in_dir(other_cmds, path, prefix);
+
+			if (!colon)
+				break;
+			path = colon + 1;
+		}
+		free(paths);
+
+		qsort(other_cmds->names, other_cmds->cnt,
+		      sizeof(*other_cmds->names), cmdname_compare);
+		uniq(other_cmds);
+	}
+	exclude_cmds(other_cmds, main_cmds);
+}
+
+void list_commands(const char *title, struct cmdnames *main_cmds,
+		   struct cmdnames *other_cmds)
+{
+	unsigned int i, longest = 0;
+
+	for (i = 0; i < main_cmds->cnt; i++)
+		if (longest < main_cmds->names[i]->len)
+			longest = main_cmds->names[i]->len;
+	for (i = 0; i < other_cmds->cnt; i++)
+		if (longest < other_cmds->names[i]->len)
+			longest = other_cmds->names[i]->len;
+
+	if (main_cmds->cnt) {
+		const char *exec_path = perf_exec_path();
+		printf("available %s in '%s'\n", title, exec_path);
+		printf("----------------");
+		mput_char('-', strlen(title) + strlen(exec_path));
+		putchar('\n');
+		pretty_print_string_list(main_cmds, longest);
+		putchar('\n');
+	}
+
+	if (other_cmds->cnt) {
+		printf("%s available from elsewhere on your $PATH\n", title);
+		printf("---------------------------------------");
+		mput_char('-', strlen(title));
+		putchar('\n');
+		pretty_print_string_list(other_cmds, longest);
+		putchar('\n');
+	}
+}
+
+int is_in_cmdlist(struct cmdnames *c, const char *s)
+{
+	unsigned int i;
+
+	for (i = 0; i < c->cnt; i++)
+		if (!strcmp(s, c->names[i]->name))
+			return 1;
+	return 0;
+}
+
+static int autocorrect;
+static struct cmdnames aliases;
+
+static int perf_unknown_cmd_config(const char *var, const char *value, void *cb)
+{
+	if (!strcmp(var, "help.autocorrect"))
+		autocorrect = perf_config_int(var,value);
+	/* Also use aliases for command lookup */
+	if (!prefixcmp(var, "alias."))
+		add_cmdname(&aliases, var + 6, strlen(var + 6));
+
+	return perf_default_config(var, value, cb);
+}
+
+static int levenshtein_compare(const void *p1, const void *p2)
+{
+	const struct cmdname *const *c1 = p1, *const *c2 = p2;
+	const char *s1 = (*c1)->name, *s2 = (*c2)->name;
+	int l1 = (*c1)->len;
+	int l2 = (*c2)->len;
+	return l1 != l2 ? l1 - l2 : strcmp(s1, s2);
+}
+
+static void add_cmd_list(struct cmdnames *cmds, struct cmdnames *old)
+{
+	unsigned int i;
+
+	ALLOC_GROW(cmds->names, cmds->cnt + old->cnt, cmds->alloc);
+
+	for (i = 0; i < old->cnt; i++)
+		cmds->names[cmds->cnt++] = old->names[i];
+	free(old->names);
+	old->cnt = 0;
+	old->names = NULL;
+}
+
+const char *help_unknown_cmd(const char *cmd)
+{
+	unsigned int i, n = 0, best_similarity = 0;
+	struct cmdnames main_cmds, other_cmds;
+
+	memset(&main_cmds, 0, sizeof(main_cmds));
+	memset(&other_cmds, 0, sizeof(main_cmds));
+	memset(&aliases, 0, sizeof(aliases));
+
+	perf_config(perf_unknown_cmd_config, NULL);
+
+	load_command_list("perf-", &main_cmds, &other_cmds);
+
+	add_cmd_list(&main_cmds, &aliases);
+	add_cmd_list(&main_cmds, &other_cmds);
+	qsort(main_cmds.names, main_cmds.cnt,
+	      sizeof(main_cmds.names), cmdname_compare);
+	uniq(&main_cmds);
+
+	if (main_cmds.cnt) {
+		/* This reuses cmdname->len for similarity index */
+		for (i = 0; i < main_cmds.cnt; ++i)
+			main_cmds.names[i]->len =
+				levenshtein(cmd, main_cmds.names[i]->name, 0, 2, 1, 4);
+
+		qsort(main_cmds.names, main_cmds.cnt,
+		      sizeof(*main_cmds.names), levenshtein_compare);
+
+		best_similarity = main_cmds.names[0]->len;
+		n = 1;
+		while (n < main_cmds.cnt && best_similarity == main_cmds.names[n]->len)
+			++n;
+	}
+
+	if (autocorrect && n == 1) {
+		const char *assumed = main_cmds.names[0]->name;
+
+		main_cmds.names[0] = NULL;
+		clean_cmdnames(&main_cmds);
+		fprintf(stderr, "WARNING: You called a perf program named '%s', "
+			"which does not exist.\n"
+			"Continuing under the assumption that you meant '%s'\n",
+			cmd, assumed);
+		if (autocorrect > 0) {
+			fprintf(stderr, "in %0.1f seconds automatically...\n",
+				(float)autocorrect/10.0);
+			poll(NULL, 0, autocorrect * 100);
+		}
+		return assumed;
+	}
+
+	fprintf(stderr, "perf: '%s' is not a perf-command. See 'perf --help'.\n", cmd);
+
+	if (main_cmds.cnt && best_similarity < 6) {
+		fprintf(stderr, "\nDid you mean %s?\n",
+			n < 2 ? "this": "one of these");
+
+		for (i = 0; i < n; i++)
+			fprintf(stderr, "\t%s\n", main_cmds.names[i]->name);
+	}
+
+	exit(1);
+}
+
+int cmd_version(int argc __used, const char **argv __used, const char *prefix __used)
+{
+	printf("perf version %s\n", perf_version_string);
+	return 0;
+}
diff --git a/tools/perf/util/help.h b/tools/perf/util/help.h
new file mode 100644
index 00000000..7f5c6ded
--- /dev/null
+++ b/tools/perf/util/help.h
@@ -0,0 +1,29 @@
+#ifndef __PERF_HELP_H
+#define __PERF_HELP_H
+
+struct cmdnames {
+	size_t alloc;
+	size_t cnt;
+	struct cmdname {
+		size_t len; /* also used for similarity index in help.c */
+		char name[FLEX_ARRAY];
+	} **names;
+};
+
+static inline void mput_char(char c, unsigned int num)
+{
+	while(num--)
+		putchar(c);
+}
+
+void load_command_list(const char *prefix,
+		struct cmdnames *main_cmds,
+		struct cmdnames *other_cmds);
+void add_cmdname(struct cmdnames *cmds, const char *name, size_t len);
+/* Here we require that excludes is a sorted list. */
+void exclude_cmds(struct cmdnames *cmds, struct cmdnames *excludes);
+int is_in_cmdlist(struct cmdnames *c, const char *s);
+void list_commands(const char *title, struct cmdnames *main_cmds,
+		   struct cmdnames *other_cmds);
+
+#endif /* __PERF_HELP_H */
diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
new file mode 100644
index 00000000..fb695991
--- /dev/null
+++ b/tools/perf/util/hist.c
@@ -0,0 +1,1013 @@
+#include "annotate.h"
+#include "util.h"
+#include "build-id.h"
+#include "hist.h"
+#include "session.h"
+#include "sort.h"
+#include <math.h>
+
+enum hist_filter {
+	HIST_FILTER__DSO,
+	HIST_FILTER__THREAD,
+	HIST_FILTER__PARENT,
+};
+
+struct callchain_param	callchain_param = {
+	.mode	= CHAIN_GRAPH_REL,
+	.min_percent = 0.5
+};
+
+u16 hists__col_len(struct hists *self, enum hist_column col)
+{
+	return self->col_len[col];
+}
+
+void hists__set_col_len(struct hists *self, enum hist_column col, u16 len)
+{
+	self->col_len[col] = len;
+}
+
+bool hists__new_col_len(struct hists *self, enum hist_column col, u16 len)
+{
+	if (len > hists__col_len(self, col)) {
+		hists__set_col_len(self, col, len);
+		return true;
+	}
+	return false;
+}
+
+static void hists__reset_col_len(struct hists *self)
+{
+	enum hist_column col;
+
+	for (col = 0; col < HISTC_NR_COLS; ++col)
+		hists__set_col_len(self, col, 0);
+}
+
+static void hists__calc_col_len(struct hists *self, struct hist_entry *h)
+{
+	u16 len;
+
+	if (h->ms.sym)
+		hists__new_col_len(self, HISTC_SYMBOL, h->ms.sym->namelen);
+	else {
+		const unsigned int unresolved_col_width = BITS_PER_LONG / 4;
+
+		if (hists__col_len(self, HISTC_DSO) < unresolved_col_width &&
+		    !symbol_conf.col_width_list_str && !symbol_conf.field_sep &&
+		    !symbol_conf.dso_list)
+			hists__set_col_len(self, HISTC_DSO,
+					   unresolved_col_width);
+	}
+
+	len = thread__comm_len(h->thread);
+	if (hists__new_col_len(self, HISTC_COMM, len))
+		hists__set_col_len(self, HISTC_THREAD, len + 6);
+
+	if (h->ms.map) {
+		len = dso__name_len(h->ms.map->dso);
+		hists__new_col_len(self, HISTC_DSO, len);
+	}
+}
+
+static void hist_entry__add_cpumode_period(struct hist_entry *self,
+					   unsigned int cpumode, u64 period)
+{
+	switch (cpumode) {
+	case PERF_RECORD_MISC_KERNEL:
+		self->period_sys += period;
+		break;
+	case PERF_RECORD_MISC_USER:
+		self->period_us += period;
+		break;
+	case PERF_RECORD_MISC_GUEST_KERNEL:
+		self->period_guest_sys += period;
+		break;
+	case PERF_RECORD_MISC_GUEST_USER:
+		self->period_guest_us += period;
+		break;
+	default:
+		break;
+	}
+}
+
+/*
+ * histogram, sorted on item, collects periods
+ */
+
+static struct hist_entry *hist_entry__new(struct hist_entry *template)
+{
+	size_t callchain_size = symbol_conf.use_callchain ? sizeof(struct callchain_root) : 0;
+	struct hist_entry *self = malloc(sizeof(*self) + callchain_size);
+
+	if (self != NULL) {
+		*self = *template;
+		self->nr_events = 1;
+		if (self->ms.map)
+			self->ms.map->referenced = true;
+		if (symbol_conf.use_callchain)
+			callchain_init(self->callchain);
+	}
+
+	return self;
+}
+
+static void hists__inc_nr_entries(struct hists *self, struct hist_entry *h)
+{
+	if (!h->filtered) {
+		hists__calc_col_len(self, h);
+		++self->nr_entries;
+	}
+}
+
+static u8 symbol__parent_filter(const struct symbol *parent)
+{
+	if (symbol_conf.exclude_other && parent == NULL)
+		return 1 << HIST_FILTER__PARENT;
+	return 0;
+}
+
+struct hist_entry *__hists__add_entry(struct hists *self,
+				      struct addr_location *al,
+				      struct symbol *sym_parent, u64 period)
+{
+	struct rb_node **p = &self->entries.rb_node;
+	struct rb_node *parent = NULL;
+	struct hist_entry *he;
+	struct hist_entry entry = {
+		.thread	= al->thread,
+		.ms = {
+			.map	= al->map,
+			.sym	= al->sym,
+		},
+		.cpu	= al->cpu,
+		.ip	= al->addr,
+		.level	= al->level,
+		.period	= period,
+		.parent = sym_parent,
+		.filtered = symbol__parent_filter(sym_parent),
+	};
+	int cmp;
+
+	while (*p != NULL) {
+		parent = *p;
+		he = rb_entry(parent, struct hist_entry, rb_node);
+
+		cmp = hist_entry__cmp(&entry, he);
+
+		if (!cmp) {
+			he->period += period;
+			++he->nr_events;
+
+			/* If the map of an existing hist_entry has
+			 * become out-of-date due to an exec() or
+			 * similar, update it.  Otherwise we will
+			 * mis-adjust symbol addresses when computing
+			 * the history counter to increment.
+			 */
+			if (he->ms.map != entry.ms.map) {
+				he->ms.map = entry.ms.map;
+				if (he->ms.map)
+					he->ms.map->referenced = true;
+			}
+			goto out;
+		}
+
+		if (cmp < 0)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	he = hist_entry__new(&entry);
+	if (!he)
+		return NULL;
+	rb_link_node(&he->rb_node, parent, p);
+	rb_insert_color(&he->rb_node, &self->entries);
+	hists__inc_nr_entries(self, he);
+out:
+	hist_entry__add_cpumode_period(he, al->cpumode, period);
+	return he;
+}
+
+int64_t
+hist_entry__cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	struct sort_entry *se;
+	int64_t cmp = 0;
+
+	list_for_each_entry(se, &hist_entry__sort_list, list) {
+		cmp = se->se_cmp(left, right);
+		if (cmp)
+			break;
+	}
+
+	return cmp;
+}
+
+int64_t
+hist_entry__collapse(struct hist_entry *left, struct hist_entry *right)
+{
+	struct sort_entry *se;
+	int64_t cmp = 0;
+
+	list_for_each_entry(se, &hist_entry__sort_list, list) {
+		int64_t (*f)(struct hist_entry *, struct hist_entry *);
+
+		f = se->se_collapse ?: se->se_cmp;
+
+		cmp = f(left, right);
+		if (cmp)
+			break;
+	}
+
+	return cmp;
+}
+
+void hist_entry__free(struct hist_entry *he)
+{
+	free(he);
+}
+
+/*
+ * collapse the histogram
+ */
+
+static bool hists__collapse_insert_entry(struct hists *self,
+					 struct rb_root *root,
+					 struct hist_entry *he)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	struct hist_entry *iter;
+	int64_t cmp;
+
+	while (*p != NULL) {
+		parent = *p;
+		iter = rb_entry(parent, struct hist_entry, rb_node);
+
+		cmp = hist_entry__collapse(iter, he);
+
+		if (!cmp) {
+			iter->period += he->period;
+			if (symbol_conf.use_callchain) {
+				callchain_cursor_reset(&self->callchain_cursor);
+				callchain_merge(&self->callchain_cursor, iter->callchain,
+						he->callchain);
+			}
+			hist_entry__free(he);
+			return false;
+		}
+
+		if (cmp < 0)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	rb_link_node(&he->rb_node, parent, p);
+	rb_insert_color(&he->rb_node, root);
+	return true;
+}
+
+void hists__collapse_resort(struct hists *self)
+{
+	struct rb_root tmp;
+	struct rb_node *next;
+	struct hist_entry *n;
+
+	if (!sort__need_collapse)
+		return;
+
+	tmp = RB_ROOT;
+	next = rb_first(&self->entries);
+	self->nr_entries = 0;
+	hists__reset_col_len(self);
+
+	while (next) {
+		n = rb_entry(next, struct hist_entry, rb_node);
+		next = rb_next(&n->rb_node);
+
+		rb_erase(&n->rb_node, &self->entries);
+		if (hists__collapse_insert_entry(self, &tmp, n))
+			hists__inc_nr_entries(self, n);
+	}
+
+	self->entries = tmp;
+}
+
+/*
+ * reverse the map, sort on period.
+ */
+
+static void __hists__insert_output_entry(struct rb_root *entries,
+					 struct hist_entry *he,
+					 u64 min_callchain_hits)
+{
+	struct rb_node **p = &entries->rb_node;
+	struct rb_node *parent = NULL;
+	struct hist_entry *iter;
+
+	if (symbol_conf.use_callchain)
+		callchain_param.sort(&he->sorted_chain, he->callchain,
+				      min_callchain_hits, &callchain_param);
+
+	while (*p != NULL) {
+		parent = *p;
+		iter = rb_entry(parent, struct hist_entry, rb_node);
+
+		if (he->period > iter->period)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	rb_link_node(&he->rb_node, parent, p);
+	rb_insert_color(&he->rb_node, entries);
+}
+
+void hists__output_resort(struct hists *self)
+{
+	struct rb_root tmp;
+	struct rb_node *next;
+	struct hist_entry *n;
+	u64 min_callchain_hits;
+
+	min_callchain_hits = self->stats.total_period * (callchain_param.min_percent / 100);
+
+	tmp = RB_ROOT;
+	next = rb_first(&self->entries);
+
+	self->nr_entries = 0;
+	hists__reset_col_len(self);
+
+	while (next) {
+		n = rb_entry(next, struct hist_entry, rb_node);
+		next = rb_next(&n->rb_node);
+
+		rb_erase(&n->rb_node, &self->entries);
+		__hists__insert_output_entry(&tmp, n, min_callchain_hits);
+		hists__inc_nr_entries(self, n);
+	}
+
+	self->entries = tmp;
+}
+
+static size_t callchain__fprintf_left_margin(FILE *fp, int left_margin)
+{
+	int i;
+	int ret = fprintf(fp, "            ");
+
+	for (i = 0; i < left_margin; i++)
+		ret += fprintf(fp, " ");
+
+	return ret;
+}
+
+static size_t ipchain__fprintf_graph_line(FILE *fp, int depth, int depth_mask,
+					  int left_margin)
+{
+	int i;
+	size_t ret = callchain__fprintf_left_margin(fp, left_margin);
+
+	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 period,
+				     u64 total_samples, u64 hits,
+				     int left_margin)
+{
+	int i;
+	size_t ret = 0;
+
+	ret += callchain__fprintf_left_margin(fp, left_margin);
+	for (i = 0; i < depth; i++) {
+		if (depth_mask & (1 << i))
+			ret += fprintf(fp, "|");
+		else
+			ret += fprintf(fp, " ");
+		if (!period && 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->ms.sym)
+		ret += fprintf(fp, "%s\n", chain->ms.sym->name);
+	else
+		ret += fprintf(fp, "%p\n", (void *)(long)chain->ip);
+
+	return ret;
+}
+
+static struct symbol *rem_sq_bracket;
+static struct callchain_list rem_hits;
+
+static void init_rem_hits(void)
+{
+	rem_sq_bracket = malloc(sizeof(*rem_sq_bracket) + 6);
+	if (!rem_sq_bracket) {
+		fprintf(stderr, "Not enough memory to display remaining hits\n");
+		return;
+	}
+
+	strcpy(rem_sq_bracket->name, "[...]");
+	rem_hits.ms.sym = rem_sq_bracket;
+}
+
+static size_t __callchain__fprintf_graph(FILE *fp, struct callchain_node *self,
+					 u64 total_samples, int depth,
+					 int depth_mask, int left_margin)
+{
+	struct rb_node *node, *next;
+	struct callchain_node *child;
+	struct callchain_list *chain;
+	int new_depth_mask = depth_mask;
+	u64 new_total;
+	u64 remaining;
+	size_t ret = 0;
+	int i;
+	uint entries_printed = 0;
+
+	if (callchain_param.mode == CHAIN_GRAPH_REL)
+		new_total = self->children_hit;
+	else
+		new_total = total_samples;
+
+	remaining = new_total;
+
+	node = rb_first(&self->rb_root);
+	while (node) {
+		u64 cumul;
+
+		child = rb_entry(node, struct callchain_node, rb_node);
+		cumul = callchain_cumul_hits(child);
+		remaining -= cumul;
+
+		/*
+		 * 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.
+		 * Except if we have remaining filtered hits. They will
+		 * supersede the last child
+		 */
+		next = rb_next(node);
+		if (!next && (callchain_param.mode != CHAIN_GRAPH_REL || !remaining))
+			new_depth_mask &= ~(1 << (depth - 1));
+
+		/*
+		 * But we keep the older depth mask for the line separator
+		 * to keep the level link until we reach the last child
+		 */
+		ret += ipchain__fprintf_graph_line(fp, depth, depth_mask,
+						   left_margin);
+		i = 0;
+		list_for_each_entry(chain, &child->val, list) {
+			ret += ipchain__fprintf_graph(fp, chain, depth,
+						      new_depth_mask, i++,
+						      new_total,
+						      cumul,
+						      left_margin);
+		}
+		ret += __callchain__fprintf_graph(fp, child, new_total,
+						  depth + 1,
+						  new_depth_mask | (1 << depth),
+						  left_margin);
+		node = next;
+		if (++entries_printed == callchain_param.print_limit)
+			break;
+	}
+
+	if (callchain_param.mode == CHAIN_GRAPH_REL &&
+		remaining && remaining != new_total) {
+
+		if (!rem_sq_bracket)
+			return ret;
+
+		new_depth_mask &= ~(1 << (depth - 1));
+
+		ret += ipchain__fprintf_graph(fp, &rem_hits, depth,
+					      new_depth_mask, 0, new_total,
+					      remaining, left_margin);
+	}
+
+	return ret;
+}
+
+static size_t callchain__fprintf_graph(FILE *fp, struct callchain_node *self,
+				       u64 total_samples, int left_margin)
+{
+	struct callchain_list *chain;
+	bool printed = false;
+	int i = 0;
+	int ret = 0;
+	u32 entries_printed = 0;
+
+	list_for_each_entry(chain, &self->val, list) {
+		if (!i++ && sort__first_dimension == SORT_SYM)
+			continue;
+
+		if (!printed) {
+			ret += callchain__fprintf_left_margin(fp, left_margin);
+			ret += fprintf(fp, "|\n");
+			ret += callchain__fprintf_left_margin(fp, left_margin);
+			ret += fprintf(fp, "---");
+
+			left_margin += 3;
+			printed = true;
+		} else
+			ret += callchain__fprintf_left_margin(fp, left_margin);
+
+		if (chain->ms.sym)
+			ret += fprintf(fp, " %s\n", chain->ms.sym->name);
+		else
+			ret += fprintf(fp, " %p\n", (void *)(long)chain->ip);
+
+		if (++entries_printed == callchain_param.print_limit)
+			break;
+	}
+
+	ret += __callchain__fprintf_graph(fp, self, total_samples, 1, 1, left_margin);
+
+	return ret;
+}
+
+static size_t callchain__fprintf_flat(FILE *fp, struct callchain_node *self,
+				      u64 total_samples)
+{
+	struct callchain_list *chain;
+	size_t ret = 0;
+
+	if (!self)
+		return 0;
+
+	ret += callchain__fprintf_flat(fp, self->parent, total_samples);
+
+
+	list_for_each_entry(chain, &self->val, list) {
+		if (chain->ip >= PERF_CONTEXT_MAX)
+			continue;
+		if (chain->ms.sym)
+			ret += fprintf(fp, "                %s\n", chain->ms.sym->name);
+		else
+			ret += fprintf(fp, "                %p\n",
+					(void *)(long)chain->ip);
+	}
+
+	return ret;
+}
+
+static size_t hist_entry_callchain__fprintf(FILE *fp, struct hist_entry *self,
+					    u64 total_samples, int left_margin)
+{
+	struct rb_node *rb_node;
+	struct callchain_node *chain;
+	size_t ret = 0;
+	u32 entries_printed = 0;
+
+	rb_node = rb_first(&self->sorted_chain);
+	while (rb_node) {
+		double percent;
+
+		chain = rb_entry(rb_node, struct callchain_node, rb_node);
+		percent = chain->hit * 100.0 / 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,
+							left_margin);
+		case CHAIN_NONE:
+		default:
+			break;
+		}
+		ret += fprintf(fp, "\n");
+		if (++entries_printed == callchain_param.print_limit)
+			break;
+		rb_node = rb_next(rb_node);
+	}
+
+	return ret;
+}
+
+int hist_entry__snprintf(struct hist_entry *self, char *s, size_t size,
+			 struct hists *hists, struct hists *pair_hists,
+			 bool show_displacement, long displacement,
+			 bool color, u64 session_total)
+{
+	struct sort_entry *se;
+	u64 period, total, period_sys, period_us, period_guest_sys, period_guest_us;
+	u64 nr_events;
+	const char *sep = symbol_conf.field_sep;
+	int ret;
+
+	if (symbol_conf.exclude_other && !self->parent)
+		return 0;
+
+	if (pair_hists) {
+		period = self->pair ? self->pair->period : 0;
+		nr_events = self->pair ? self->pair->nr_events : 0;
+		total = pair_hists->stats.total_period;
+		period_sys = self->pair ? self->pair->period_sys : 0;
+		period_us = self->pair ? self->pair->period_us : 0;
+		period_guest_sys = self->pair ? self->pair->period_guest_sys : 0;
+		period_guest_us = self->pair ? self->pair->period_guest_us : 0;
+	} else {
+		period = self->period;
+		nr_events = self->nr_events;
+		total = session_total;
+		period_sys = self->period_sys;
+		period_us = self->period_us;
+		period_guest_sys = self->period_guest_sys;
+		period_guest_us = self->period_guest_us;
+	}
+
+	if (total) {
+		if (color)
+			ret = percent_color_snprintf(s, size,
+						     sep ? "%.2f" : "   %6.2f%%",
+						     (period * 100.0) / total);
+		else
+			ret = snprintf(s, size, sep ? "%.2f" : "   %6.2f%%",
+				       (period * 100.0) / total);
+		if (symbol_conf.show_cpu_utilization) {
+			ret += percent_color_snprintf(s + ret, size - ret,
+					sep ? "%.2f" : "   %6.2f%%",
+					(period_sys * 100.0) / total);
+			ret += percent_color_snprintf(s + ret, size - ret,
+					sep ? "%.2f" : "   %6.2f%%",
+					(period_us * 100.0) / total);
+			if (perf_guest) {
+				ret += percent_color_snprintf(s + ret,
+						size - ret,
+						sep ? "%.2f" : "   %6.2f%%",
+						(period_guest_sys * 100.0) /
+								total);
+				ret += percent_color_snprintf(s + ret,
+						size - ret,
+						sep ? "%.2f" : "   %6.2f%%",
+						(period_guest_us * 100.0) /
+								total);
+			}
+		}
+	} else
+		ret = snprintf(s, size, sep ? "%" PRIu64 : "%12" PRIu64 " ", period);
+
+	if (symbol_conf.show_nr_samples) {
+		if (sep)
+			ret += snprintf(s + ret, size - ret, "%c%" PRIu64, *sep, nr_events);
+		else
+			ret += snprintf(s + ret, size - ret, "%11" PRIu64, nr_events);
+	}
+
+	if (pair_hists) {
+		char bf[32];
+		double old_percent = 0, new_percent = 0, diff;
+
+		if (total > 0)
+			old_percent = (period * 100.0) / total;
+		if (session_total > 0)
+			new_percent = (self->period * 100.0) / session_total;
+
+		diff = new_percent - old_percent;
+
+		if (fabs(diff) >= 0.01)
+			snprintf(bf, sizeof(bf), "%+4.2F%%", diff);
+		else
+			snprintf(bf, sizeof(bf), " ");
+
+		if (sep)
+			ret += snprintf(s + ret, size - ret, "%c%s", *sep, bf);
+		else
+			ret += snprintf(s + ret, size - ret, "%11.11s", bf);
+
+		if (show_displacement) {
+			if (displacement)
+				snprintf(bf, sizeof(bf), "%+4ld", displacement);
+			else
+				snprintf(bf, sizeof(bf), " ");
+
+			if (sep)
+				ret += snprintf(s + ret, size - ret, "%c%s", *sep, bf);
+			else
+				ret += snprintf(s + ret, size - ret, "%6.6s", bf);
+		}
+	}
+
+	list_for_each_entry(se, &hist_entry__sort_list, list) {
+		if (se->elide)
+			continue;
+
+		ret += snprintf(s + ret, size - ret, "%s", sep ?: "  ");
+		ret += se->se_snprintf(self, s + ret, size - ret,
+				       hists__col_len(hists, se->se_width_idx));
+	}
+
+	return ret;
+}
+
+int hist_entry__fprintf(struct hist_entry *self, struct hists *hists,
+			struct hists *pair_hists, bool show_displacement,
+			long displacement, FILE *fp, u64 session_total)
+{
+	char bf[512];
+	hist_entry__snprintf(self, bf, sizeof(bf), hists, pair_hists,
+			     show_displacement, displacement,
+			     true, session_total);
+	return fprintf(fp, "%s\n", bf);
+}
+
+static size_t hist_entry__fprintf_callchain(struct hist_entry *self,
+					    struct hists *hists, FILE *fp,
+					    u64 session_total)
+{
+	int left_margin = 0;
+
+	if (sort__first_dimension == SORT_COMM) {
+		struct sort_entry *se = list_first_entry(&hist_entry__sort_list,
+							 typeof(*se), list);
+		left_margin = hists__col_len(hists, se->se_width_idx);
+		left_margin -= thread__comm_len(self->thread);
+	}
+
+	return hist_entry_callchain__fprintf(fp, self, session_total,
+					     left_margin);
+}
+
+size_t hists__fprintf(struct hists *self, struct hists *pair,
+		      bool show_displacement, FILE *fp)
+{
+	struct sort_entry *se;
+	struct rb_node *nd;
+	size_t ret = 0;
+	unsigned long position = 1;
+	long displacement = 0;
+	unsigned int width;
+	const char *sep = symbol_conf.field_sep;
+	const char *col_width = symbol_conf.col_width_list_str;
+
+	init_rem_hits();
+
+	fprintf(fp, "# %s", pair ? "Baseline" : "Overhead");
+
+	if (symbol_conf.show_nr_samples) {
+		if (sep)
+			fprintf(fp, "%cSamples", *sep);
+		else
+			fputs("  Samples  ", fp);
+	}
+
+	if (symbol_conf.show_cpu_utilization) {
+		if (sep) {
+			ret += fprintf(fp, "%csys", *sep);
+			ret += fprintf(fp, "%cus", *sep);
+			if (perf_guest) {
+				ret += fprintf(fp, "%cguest sys", *sep);
+				ret += fprintf(fp, "%cguest us", *sep);
+			}
+		} else {
+			ret += fprintf(fp, "  sys  ");
+			ret += fprintf(fp, "  us  ");
+			if (perf_guest) {
+				ret += fprintf(fp, "  guest sys  ");
+				ret += fprintf(fp, "  guest us  ");
+			}
+		}
+	}
+
+	if (pair) {
+		if (sep)
+			ret += fprintf(fp, "%cDelta", *sep);
+		else
+			ret += fprintf(fp, "  Delta    ");
+
+		if (show_displacement) {
+			if (sep)
+				ret += fprintf(fp, "%cDisplacement", *sep);
+			else
+				ret += fprintf(fp, " Displ");
+		}
+	}
+
+	list_for_each_entry(se, &hist_entry__sort_list, list) {
+		if (se->elide)
+			continue;
+		if (sep) {
+			fprintf(fp, "%c%s", *sep, se->se_header);
+			continue;
+		}
+		width = strlen(se->se_header);
+		if (symbol_conf.col_width_list_str) {
+			if (col_width) {
+				hists__set_col_len(self, se->se_width_idx,
+						   atoi(col_width));
+				col_width = strchr(col_width, ',');
+				if (col_width)
+					++col_width;
+			}
+		}
+		if (!hists__new_col_len(self, se->se_width_idx, width))
+			width = hists__col_len(self, se->se_width_idx);
+		fprintf(fp, "  %*s", width, se->se_header);
+	}
+	fprintf(fp, "\n");
+
+	if (sep)
+		goto print_entries;
+
+	fprintf(fp, "# ........");
+	if (symbol_conf.show_nr_samples)
+		fprintf(fp, " ..........");
+	if (pair) {
+		fprintf(fp, " ..........");
+		if (show_displacement)
+			fprintf(fp, " .....");
+	}
+	list_for_each_entry(se, &hist_entry__sort_list, list) {
+		unsigned int i;
+
+		if (se->elide)
+			continue;
+
+		fprintf(fp, "  ");
+		width = hists__col_len(self, se->se_width_idx);
+		if (width == 0)
+			width = strlen(se->se_header);
+		for (i = 0; i < width; i++)
+			fprintf(fp, ".");
+	}
+
+	fprintf(fp, "\n#\n");
+
+print_entries:
+	for (nd = rb_first(&self->entries); nd; nd = rb_next(nd)) {
+		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+
+		if (show_displacement) {
+			if (h->pair != NULL)
+				displacement = ((long)h->pair->position -
+					        (long)position);
+			else
+				displacement = 0;
+			++position;
+		}
+		ret += hist_entry__fprintf(h, self, pair, show_displacement,
+					   displacement, fp, self->stats.total_period);
+
+		if (symbol_conf.use_callchain)
+			ret += hist_entry__fprintf_callchain(h, self, fp,
+							     self->stats.total_period);
+		if (h->ms.map == NULL && verbose > 1) {
+			__map_groups__fprintf_maps(&h->thread->mg,
+						   MAP__FUNCTION, verbose, fp);
+			fprintf(fp, "%.10s end\n", graph_dotted_line);
+		}
+	}
+
+	free(rem_sq_bracket);
+
+	return ret;
+}
+
+/*
+ * See hists__fprintf to match the column widths
+ */
+unsigned int hists__sort_list_width(struct hists *self)
+{
+	struct sort_entry *se;
+	int ret = 9; /* total % */
+
+	if (symbol_conf.show_cpu_utilization) {
+		ret += 7; /* count_sys % */
+		ret += 6; /* count_us % */
+		if (perf_guest) {
+			ret += 13; /* count_guest_sys % */
+			ret += 12; /* count_guest_us % */
+		}
+	}
+
+	if (symbol_conf.show_nr_samples)
+		ret += 11;
+
+	list_for_each_entry(se, &hist_entry__sort_list, list)
+		if (!se->elide)
+			ret += 2 + hists__col_len(self, se->se_width_idx);
+
+	if (verbose) /* Addr + origin */
+		ret += 3 + BITS_PER_LONG / 4;
+
+	return ret;
+}
+
+static void hists__remove_entry_filter(struct hists *self, struct hist_entry *h,
+				       enum hist_filter filter)
+{
+	h->filtered &= ~(1 << filter);
+	if (h->filtered)
+		return;
+
+	++self->nr_entries;
+	if (h->ms.unfolded)
+		self->nr_entries += h->nr_rows;
+	h->row_offset = 0;
+	self->stats.total_period += h->period;
+	self->stats.nr_events[PERF_RECORD_SAMPLE] += h->nr_events;
+
+	hists__calc_col_len(self, h);
+}
+
+void hists__filter_by_dso(struct hists *self, const struct dso *dso)
+{
+	struct rb_node *nd;
+
+	self->nr_entries = self->stats.total_period = 0;
+	self->stats.nr_events[PERF_RECORD_SAMPLE] = 0;
+	hists__reset_col_len(self);
+
+	for (nd = rb_first(&self->entries); nd; nd = rb_next(nd)) {
+		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+
+		if (symbol_conf.exclude_other && !h->parent)
+			continue;
+
+		if (dso != NULL && (h->ms.map == NULL || h->ms.map->dso != dso)) {
+			h->filtered |= (1 << HIST_FILTER__DSO);
+			continue;
+		}
+
+		hists__remove_entry_filter(self, h, HIST_FILTER__DSO);
+	}
+}
+
+void hists__filter_by_thread(struct hists *self, const struct thread *thread)
+{
+	struct rb_node *nd;
+
+	self->nr_entries = self->stats.total_period = 0;
+	self->stats.nr_events[PERF_RECORD_SAMPLE] = 0;
+	hists__reset_col_len(self);
+
+	for (nd = rb_first(&self->entries); nd; nd = rb_next(nd)) {
+		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+
+		if (thread != NULL && h->thread != thread) {
+			h->filtered |= (1 << HIST_FILTER__THREAD);
+			continue;
+		}
+
+		hists__remove_entry_filter(self, h, HIST_FILTER__THREAD);
+	}
+}
+
+int hist_entry__inc_addr_samples(struct hist_entry *he, int evidx, u64 ip)
+{
+	return symbol__inc_addr_samples(he->ms.sym, he->ms.map, evidx, ip);
+}
+
+int hist_entry__annotate(struct hist_entry *he, size_t privsize)
+{
+	return symbol__annotate(he->ms.sym, he->ms.map, privsize);
+}
+
+void hists__inc_nr_events(struct hists *self, u32 type)
+{
+	++self->stats.nr_events[0];
+	++self->stats.nr_events[type];
+}
+
+size_t hists__fprintf_nr_events(struct hists *self, FILE *fp)
+{
+	int i;
+	size_t ret = 0;
+
+	for (i = 0; i < PERF_RECORD_HEADER_MAX; ++i) {
+		const char *name;
+
+		if (self->stats.nr_events[i] == 0)
+			continue;
+
+		name = perf_event__name(i);
+		if (!strcmp(name, "UNKNOWN"))
+			continue;
+
+		ret += fprintf(fp, "%16s events: %10d\n", name,
+			       self->stats.nr_events[i]);
+	}
+
+	return ret;
+}
diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h
new file mode 100644
index 00000000..3beb97c4
--- /dev/null
+++ b/tools/perf/util/hist.h
@@ -0,0 +1,117 @@
+#ifndef __PERF_HIST_H
+#define __PERF_HIST_H
+
+#include <linux/types.h>
+#include "callchain.h"
+
+extern struct callchain_param callchain_param;
+
+struct hist_entry;
+struct addr_location;
+struct symbol;
+
+/*
+ * The kernel collects the number of events it couldn't send in a stretch and
+ * when possible sends this number in a PERF_RECORD_LOST event. The number of
+ * such "chunks" of lost events is stored in .nr_events[PERF_EVENT_LOST] while
+ * total_lost tells exactly how many events the kernel in fact lost, i.e. it is
+ * the sum of all struct lost_event.lost fields reported.
+ *
+ * The total_period is needed because by default auto-freq is used, so
+ * multipling nr_events[PERF_EVENT_SAMPLE] by a frequency isn't possible to get
+ * the total number of low level events, it is necessary to to sum all struct
+ * sample_event.period and stash the result in total_period.
+ */
+struct events_stats {
+	u64 total_period;
+	u64 total_lost;
+	u64 total_invalid_chains;
+	u32 nr_events[PERF_RECORD_HEADER_MAX];
+	u32 nr_unknown_events;
+	u32 nr_invalid_chains;
+	u32 nr_unknown_id;
+};
+
+enum hist_column {
+	HISTC_SYMBOL,
+	HISTC_DSO,
+	HISTC_THREAD,
+	HISTC_COMM,
+	HISTC_PARENT,
+	HISTC_CPU,
+	HISTC_NR_COLS, /* Last entry */
+};
+
+struct hists {
+	struct rb_root		entries;
+	u64			nr_entries;
+	struct events_stats	stats;
+	u64			event_stream;
+	u16			col_len[HISTC_NR_COLS];
+	/* Best would be to reuse the session callchain cursor */
+	struct callchain_cursor	callchain_cursor;
+};
+
+struct hist_entry *__hists__add_entry(struct hists *self,
+				      struct addr_location *al,
+				      struct symbol *parent, u64 period);
+extern int64_t hist_entry__cmp(struct hist_entry *, struct hist_entry *);
+extern int64_t hist_entry__collapse(struct hist_entry *, struct hist_entry *);
+int hist_entry__fprintf(struct hist_entry *self, struct hists *hists,
+			struct hists *pair_hists, bool show_displacement,
+			long displacement, FILE *fp, u64 total);
+int hist_entry__snprintf(struct hist_entry *self, char *bf, size_t size,
+			 struct hists *hists, struct hists *pair_hists,
+			 bool show_displacement, long displacement,
+			 bool color, u64 total);
+void hist_entry__free(struct hist_entry *);
+
+void hists__output_resort(struct hists *self);
+void hists__collapse_resort(struct hists *self);
+
+void hists__inc_nr_events(struct hists *self, u32 type);
+size_t hists__fprintf_nr_events(struct hists *self, FILE *fp);
+
+size_t hists__fprintf(struct hists *self, struct hists *pair,
+		      bool show_displacement, FILE *fp);
+
+int hist_entry__inc_addr_samples(struct hist_entry *self, int evidx, u64 addr);
+int hist_entry__annotate(struct hist_entry *self, size_t privsize);
+
+void hists__filter_by_dso(struct hists *self, const struct dso *dso);
+void hists__filter_by_thread(struct hists *self, const struct thread *thread);
+
+u16 hists__col_len(struct hists *self, enum hist_column col);
+void hists__set_col_len(struct hists *self, enum hist_column col, u16 len);
+bool hists__new_col_len(struct hists *self, enum hist_column col, u16 len);
+
+struct perf_evlist;
+
+#ifdef NO_NEWT_SUPPORT
+static inline
+int perf_evlist__tui_browse_hists(struct perf_evlist *evlist __used,
+				  const char *help __used)
+{
+	return 0;
+}
+
+static inline int hist_entry__tui_annotate(struct hist_entry *self __used,
+					   int evidx __used)
+{
+	return 0;
+}
+#define KEY_LEFT -1
+#define KEY_RIGHT -2
+#else
+#include <newt.h>
+int hist_entry__tui_annotate(struct hist_entry *self, int evidx);
+
+#define KEY_LEFT NEWT_KEY_LEFT
+#define KEY_RIGHT NEWT_KEY_RIGHT
+
+int perf_evlist__tui_browse_hists(struct perf_evlist *evlist, const char *help);
+#endif
+
+unsigned int hists__sort_list_width(struct hists *self);
+
+#endif	/* __PERF_HIST_H */
diff --git a/tools/perf/util/hweight.c b/tools/perf/util/hweight.c
new file mode 100644
index 00000000..5c1d0d09
--- /dev/null
+++ b/tools/perf/util/hweight.c
@@ -0,0 +1,31 @@
+#include <linux/bitops.h>
+
+/**
+ * hweightN - returns the hamming weight of a N-bit word
+ * @x: the word to weigh
+ *
+ * The Hamming Weight of a number is the total number of bits set in it.
+ */
+
+unsigned int hweight32(unsigned int w)
+{
+	unsigned int res = w - ((w >> 1) & 0x55555555);
+	res = (res & 0x33333333) + ((res >> 2) & 0x33333333);
+	res = (res + (res >> 4)) & 0x0F0F0F0F;
+	res = res + (res >> 8);
+	return (res + (res >> 16)) & 0x000000FF;
+}
+
+unsigned long hweight64(__u64 w)
+{
+#if BITS_PER_LONG == 32
+	return hweight32((unsigned int)(w >> 32)) + hweight32((unsigned int)w);
+#elif BITS_PER_LONG == 64
+	__u64 res = w - ((w >> 1) & 0x5555555555555555ul);
+	res = (res & 0x3333333333333333ul) + ((res >> 2) & 0x3333333333333333ul);
+	res = (res + (res >> 4)) & 0x0F0F0F0F0F0F0F0Ful;
+	res = res + (res >> 8);
+	res = res + (res >> 16);
+	return (res + (res >> 32)) & 0x00000000000000FFul;
+#endif
+}
diff --git a/tools/perf/util/include/asm/alternative-asm.h b/tools/perf/util/include/asm/alternative-asm.h
new file mode 100644
index 00000000..6789d788
--- /dev/null
+++ b/tools/perf/util/include/asm/alternative-asm.h
@@ -0,0 +1,8 @@
+#ifndef _PERF_ASM_ALTERNATIVE_ASM_H
+#define _PERF_ASM_ALTERNATIVE_ASM_H
+
+/* Just disable it so we can build arch/x86/lib/memcpy_64.S for perf bench: */
+
+#define altinstruction_entry #
+
+#endif
diff --git a/tools/perf/util/include/asm/asm-offsets.h b/tools/perf/util/include/asm/asm-offsets.h
new file mode 100644
index 00000000..ed538942
--- /dev/null
+++ b/tools/perf/util/include/asm/asm-offsets.h
@@ -0,0 +1 @@
+/* stub */
diff --git a/tools/perf/util/include/asm/bug.h b/tools/perf/util/include/asm/bug.h
new file mode 100644
index 00000000..7fcc6810
--- /dev/null
+++ b/tools/perf/util/include/asm/bug.h
@@ -0,0 +1,22 @@
+#ifndef _PERF_ASM_GENERIC_BUG_H
+#define _PERF_ASM_GENERIC_BUG_H
+
+#define __WARN_printf(arg...)	do { fprintf(stderr, arg); } while (0)
+
+#define WARN(condition, format...) ({		\
+	int __ret_warn_on = !!(condition);	\
+	if (unlikely(__ret_warn_on))		\
+		__WARN_printf(format);		\
+	unlikely(__ret_warn_on);		\
+})
+
+#define WARN_ONCE(condition, format...)	({	\
+	static int __warned;			\
+	int __ret_warn_once = !!(condition);	\
+						\
+	if (unlikely(__ret_warn_once))		\
+		if (WARN(!__warned, format)) 	\
+			__warned = 1;		\
+	unlikely(__ret_warn_once);		\
+})
+#endif
diff --git a/tools/perf/util/include/asm/byteorder.h b/tools/perf/util/include/asm/byteorder.h
new file mode 100644
index 00000000..b722abe3
--- /dev/null
+++ b/tools/perf/util/include/asm/byteorder.h
@@ -0,0 +1,2 @@
+#include <asm/types.h>
+#include "../../../../include/linux/swab.h"
diff --git a/tools/perf/util/include/asm/cpufeature.h b/tools/perf/util/include/asm/cpufeature.h
new file mode 100644
index 00000000..acffd5e4
--- /dev/null
+++ b/tools/perf/util/include/asm/cpufeature.h
@@ -0,0 +1,9 @@
+
+#ifndef PERF_CPUFEATURE_H
+#define PERF_CPUFEATURE_H
+
+/* cpufeature.h ... dummy header file for including arch/x86/lib/memcpy_64.S */
+
+#define X86_FEATURE_REP_GOOD 0
+
+#endif	/* PERF_CPUFEATURE_H */
diff --git a/tools/perf/util/include/asm/dwarf2.h b/tools/perf/util/include/asm/dwarf2.h
new file mode 100644
index 00000000..bb4198e7
--- /dev/null
+++ b/tools/perf/util/include/asm/dwarf2.h
@@ -0,0 +1,11 @@
+
+#ifndef PERF_DWARF2_H
+#define PERF_DWARF2_H
+
+/* dwarf2.h ... dummy header file for including arch/x86/lib/memcpy_64.S */
+
+#define CFI_STARTPROC
+#define CFI_ENDPROC
+
+#endif	/* PERF_DWARF2_H */
+
diff --git a/tools/perf/util/include/asm/hweight.h b/tools/perf/util/include/asm/hweight.h
new file mode 100644
index 00000000..36cf26d4
--- /dev/null
+++ b/tools/perf/util/include/asm/hweight.h
@@ -0,0 +1,8 @@
+#ifndef PERF_HWEIGHT_H
+#define PERF_HWEIGHT_H
+
+#include <linux/types.h>
+unsigned int hweight32(unsigned int w);
+unsigned long hweight64(__u64 w);
+
+#endif /* PERF_HWEIGHT_H */
diff --git a/tools/perf/util/include/asm/swab.h b/tools/perf/util/include/asm/swab.h
new file mode 100644
index 00000000..ed538942
--- /dev/null
+++ b/tools/perf/util/include/asm/swab.h
@@ -0,0 +1 @@
+/* stub */
diff --git a/tools/perf/util/include/asm/system.h b/tools/perf/util/include/asm/system.h
new file mode 100644
index 00000000..710cecca
--- /dev/null
+++ b/tools/perf/util/include/asm/system.h
@@ -0,0 +1 @@
+/* Empty */
diff --git a/tools/perf/util/include/asm/uaccess.h b/tools/perf/util/include/asm/uaccess.h
new file mode 100644
index 00000000..d0f72b8f
--- /dev/null
+++ b/tools/perf/util/include/asm/uaccess.h
@@ -0,0 +1,14 @@
+#ifndef _PERF_ASM_UACCESS_H_
+#define _PERF_ASM_UACCESS_H_
+
+#define __get_user(src, dest)						\
+({									\
+	(src) = *dest;							\
+	0;								\
+})
+
+#define get_user	__get_user
+
+#define access_ok(type, addr, size)	1
+
+#endif
diff --git a/tools/perf/util/include/dwarf-regs.h b/tools/perf/util/include/dwarf-regs.h
new file mode 100644
index 00000000..cf6727e9
--- /dev/null
+++ b/tools/perf/util/include/dwarf-regs.h
@@ -0,0 +1,8 @@
+#ifndef _PERF_DWARF_REGS_H_
+#define _PERF_DWARF_REGS_H_
+
+#ifdef DWARF_SUPPORT
+const char *get_arch_regstr(unsigned int n);
+#endif
+
+#endif
diff --git a/tools/perf/util/include/linux/bitmap.h b/tools/perf/util/include/linux/bitmap.h
new file mode 100644
index 00000000..eda4416e
--- /dev/null
+++ b/tools/perf/util/include/linux/bitmap.h
@@ -0,0 +1,35 @@
+#ifndef _PERF_BITOPS_H
+#define _PERF_BITOPS_H
+
+#include <string.h>
+#include <linux/bitops.h>
+
+int __bitmap_weight(const unsigned long *bitmap, int bits);
+
+#define BITMAP_LAST_WORD_MASK(nbits)					\
+(									\
+	((nbits) % BITS_PER_LONG) ?					\
+		(1UL<<((nbits) % BITS_PER_LONG))-1 : ~0UL		\
+)
+
+#define small_const_nbits(nbits) \
+	(__builtin_constant_p(nbits) && (nbits) <= BITS_PER_LONG)
+
+static inline void bitmap_zero(unsigned long *dst, int nbits)
+{
+	if (small_const_nbits(nbits))
+		*dst = 0UL;
+	else {
+		int len = BITS_TO_LONGS(nbits) * sizeof(unsigned long);
+		memset(dst, 0, len);
+	}
+}
+
+static inline int bitmap_weight(const unsigned long *src, int nbits)
+{
+	if (small_const_nbits(nbits))
+		return hweight_long(*src & BITMAP_LAST_WORD_MASK(nbits));
+	return __bitmap_weight(src, nbits);
+}
+
+#endif /* _PERF_BITOPS_H */
diff --git a/tools/perf/util/include/linux/bitops.h b/tools/perf/util/include/linux/bitops.h
new file mode 100644
index 00000000..305c8484
--- /dev/null
+++ b/tools/perf/util/include/linux/bitops.h
@@ -0,0 +1,33 @@
+#ifndef _PERF_LINUX_BITOPS_H_
+#define _PERF_LINUX_BITOPS_H_
+
+#include <linux/kernel.h>
+#include <linux/compiler.h>
+#include <asm/hweight.h>
+
+#define BITS_PER_LONG __WORDSIZE
+#define BITS_PER_BYTE           8
+#define BITS_TO_LONGS(nr)       DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(long))
+
+static inline void set_bit(int nr, unsigned long *addr)
+{
+	addr[nr / BITS_PER_LONG] |= 1UL << (nr % BITS_PER_LONG);
+}
+
+static inline void clear_bit(int nr, unsigned long *addr)
+{
+	addr[nr / BITS_PER_LONG] &= ~(1UL << (nr % BITS_PER_LONG));
+}
+
+static __always_inline int test_bit(unsigned int nr, const unsigned long *addr)
+{
+	return ((1UL << (nr % BITS_PER_LONG)) &
+		(((unsigned long *)addr)[nr / BITS_PER_LONG])) != 0;
+}
+
+static inline unsigned long hweight_long(unsigned long w)
+{
+	return sizeof(w) == 4 ? hweight32(w) : hweight64(w);
+}
+
+#endif
diff --git a/tools/perf/util/include/linux/compiler.h b/tools/perf/util/include/linux/compiler.h
new file mode 100644
index 00000000..547628e9
--- /dev/null
+++ b/tools/perf/util/include/linux/compiler.h
@@ -0,0 +1,14 @@
+#ifndef _PERF_LINUX_COMPILER_H_
+#define _PERF_LINUX_COMPILER_H_
+
+#ifndef __always_inline
+#define __always_inline	inline
+#endif
+#define __user
+#ifndef __attribute_const__
+#define __attribute_const__
+#endif
+
+#define __used		__attribute__((__unused__))
+
+#endif
diff --git a/tools/perf/util/include/linux/const.h b/tools/perf/util/include/linux/const.h
new file mode 100644
index 00000000..1b476c9a
--- /dev/null
+++ b/tools/perf/util/include/linux/const.h
@@ -0,0 +1 @@
+#include "../../../../include/linux/const.h"
diff --git a/tools/perf/util/include/linux/ctype.h b/tools/perf/util/include/linux/ctype.h
new file mode 100644
index 00000000..a53d4ee1
--- /dev/null
+++ b/tools/perf/util/include/linux/ctype.h
@@ -0,0 +1 @@
+#include "../util.h"
diff --git a/tools/perf/util/include/linux/hash.h b/tools/perf/util/include/linux/hash.h
new file mode 100644
index 00000000..201f5739
--- /dev/null
+++ b/tools/perf/util/include/linux/hash.h
@@ -0,0 +1,5 @@
+#include "../../../../include/linux/hash.h"
+
+#ifndef PERF_HASH_H
+#define PERF_HASH_H
+#endif
diff --git a/tools/perf/util/include/linux/kernel.h b/tools/perf/util/include/linux/kernel.h
new file mode 100644
index 00000000..1eb804fd
--- /dev/null
+++ b/tools/perf/util/include/linux/kernel.h
@@ -0,0 +1,111 @@
+#ifndef PERF_LINUX_KERNEL_H_
+#define PERF_LINUX_KERNEL_H_
+
+#include <stdarg.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <assert.h>
+
+#define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
+
+#define ALIGN(x,a)		__ALIGN_MASK(x,(typeof(x))(a)-1)
+#define __ALIGN_MASK(x,mask)	(((x)+(mask))&~(mask))
+
+#ifndef offsetof
+#define offsetof(TYPE, MEMBER) ((size_t) &((TYPE *)0)->MEMBER)
+#endif
+
+#ifndef container_of
+/**
+ * container_of - cast a member of a structure out to the containing structure
+ * @ptr:	the pointer to the member.
+ * @type:	the type of the container struct this is embedded in.
+ * @member:	the name of the member within the struct.
+ *
+ */
+#define container_of(ptr, type, member) ({			\
+	const typeof(((type *)0)->member) * __mptr = (ptr);	\
+	(type *)((char *)__mptr - offsetof(type, member)); })
+#endif
+
+#define BUILD_BUG_ON_ZERO(e) (sizeof(struct { int:-!!(e); }))
+
+#ifndef max
+#define max(x, y) ({				\
+	typeof(x) _max1 = (x);			\
+	typeof(y) _max2 = (y);			\
+	(void) (&_max1 == &_max2);		\
+	_max1 > _max2 ? _max1 : _max2; })
+#endif
+
+#ifndef min
+#define min(x, y) ({				\
+	typeof(x) _min1 = (x);			\
+	typeof(y) _min2 = (y);			\
+	(void) (&_min1 == &_min2);		\
+	_min1 < _min2 ? _min1 : _min2; })
+#endif
+
+#ifndef BUG_ON
+#define BUG_ON(cond) assert(!(cond))
+#endif
+
+/*
+ * Both need more care to handle endianness
+ * (Don't use bitmap_copy_le() for now)
+ */
+#define cpu_to_le64(x)	(x)
+#define cpu_to_le32(x)	(x)
+
+static inline int
+vscnprintf(char *buf, size_t size, const char *fmt, va_list args)
+{
+	int i;
+	ssize_t ssize = size;
+
+	i = vsnprintf(buf, size, fmt, args);
+
+	return (i >= ssize) ? (ssize - 1) : i;
+}
+
+static inline int scnprintf(char * buf, size_t size, const char * fmt, ...)
+{
+	va_list args;
+	ssize_t ssize = size;
+	int i;
+
+	va_start(args, fmt);
+	i = vsnprintf(buf, size, fmt, args);
+	va_end(args);
+
+	return (i >= ssize) ? (ssize - 1) : i;
+}
+
+static inline unsigned long
+simple_strtoul(const char *nptr, char **endptr, int base)
+{
+	return strtoul(nptr, endptr, base);
+}
+
+int eprintf(int level,
+	    const char *fmt, ...) __attribute__((format(printf, 2, 3)));
+
+#ifndef pr_fmt
+#define pr_fmt(fmt) fmt
+#endif
+
+#define pr_err(fmt, ...) \
+	eprintf(0, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_warning(fmt, ...) \
+	eprintf(0, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_info(fmt, ...) \
+	eprintf(0, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_debug(fmt, ...) \
+	eprintf(1, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_debugN(n, fmt, ...) \
+	eprintf(n, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_debug2(fmt, ...) pr_debugN(2, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_debug3(fmt, ...) pr_debugN(3, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_debug4(fmt, ...) pr_debugN(4, pr_fmt(fmt), ##__VA_ARGS__)
+
+#endif
diff --git a/tools/perf/util/include/linux/linkage.h b/tools/perf/util/include/linux/linkage.h
new file mode 100644
index 00000000..06387cff
--- /dev/null
+++ b/tools/perf/util/include/linux/linkage.h
@@ -0,0 +1,13 @@
+
+#ifndef PERF_LINUX_LINKAGE_H_
+#define PERF_LINUX_LINKAGE_H_
+
+/* linkage.h ... for including arch/x86/lib/memcpy_64.S */
+
+#define ENTRY(name)				\
+	.globl name;				\
+	name:
+
+#define ENDPROC(name)
+
+#endif	/* PERF_LINUX_LINKAGE_H_ */
diff --git a/tools/perf/util/include/linux/list.h b/tools/perf/util/include/linux/list.h
new file mode 100644
index 00000000..1d928a0c
--- /dev/null
+++ b/tools/perf/util/include/linux/list.h
@@ -0,0 +1,29 @@
+#include <linux/kernel.h>
+#include <linux/prefetch.h>
+
+#include "../../../../include/linux/list.h"
+
+#ifndef PERF_LIST_H
+#define PERF_LIST_H
+/**
+ * list_del_range - deletes range of entries from list.
+ * @begin: first element in the range to delete from the list.
+ * @end: last element in the range to delete from the list.
+ * Note: list_empty on the range of entries does not return true after this,
+ * the entries is in an undefined state.
+ */
+static inline void list_del_range(struct list_head *begin,
+				  struct list_head *end)
+{
+	begin->prev->next = end->next;
+	end->next->prev = begin->prev;
+}
+
+/**
+ * list_for_each_from	-	iterate over a list from one of its nodes
+ * @pos:  the &struct list_head to use as a loop cursor, from where to start
+ * @head: the head for your list.
+ */
+#define list_for_each_from(pos, head) \
+	for (; pos != (head); pos = pos->next)
+#endif
diff --git a/tools/perf/util/include/linux/module.h b/tools/perf/util/include/linux/module.h
new file mode 100644
index 00000000..b43e2dc2
--- /dev/null
+++ b/tools/perf/util/include/linux/module.h
@@ -0,0 +1,6 @@
+#ifndef PERF_LINUX_MODULE_H
+#define PERF_LINUX_MODULE_H
+
+#define EXPORT_SYMBOL(name)
+
+#endif
diff --git a/tools/perf/util/include/linux/poison.h b/tools/perf/util/include/linux/poison.h
new file mode 100644
index 00000000..fef6dbc9
--- /dev/null
+++ b/tools/perf/util/include/linux/poison.h
@@ -0,0 +1 @@
+#include "../../../../include/linux/poison.h"
diff --git a/tools/perf/util/include/linux/prefetch.h b/tools/perf/util/include/linux/prefetch.h
new file mode 100644
index 00000000..7841e485
--- /dev/null
+++ b/tools/perf/util/include/linux/prefetch.h
@@ -0,0 +1,6 @@
+#ifndef PERF_LINUX_PREFETCH_H
+#define PERF_LINUX_PREFETCH_H
+
+static inline void prefetch(void *a __attribute__((unused))) { }
+
+#endif
diff --git a/tools/perf/util/include/linux/rbtree.h b/tools/perf/util/include/linux/rbtree.h
new file mode 100644
index 00000000..7a243a14
--- /dev/null
+++ b/tools/perf/util/include/linux/rbtree.h
@@ -0,0 +1 @@
+#include "../../../../include/linux/rbtree.h"
diff --git a/tools/perf/util/include/linux/string.h b/tools/perf/util/include/linux/string.h
new file mode 100644
index 00000000..3b2f5900
--- /dev/null
+++ b/tools/perf/util/include/linux/string.h
@@ -0,0 +1 @@
+#include <string.h>
diff --git a/tools/perf/util/include/linux/types.h b/tools/perf/util/include/linux/types.h
new file mode 100644
index 00000000..12de3b81
--- /dev/null
+++ b/tools/perf/util/include/linux/types.h
@@ -0,0 +1,21 @@
+#ifndef _PERF_LINUX_TYPES_H_
+#define _PERF_LINUX_TYPES_H_
+
+#include <asm/types.h>
+
+#define DECLARE_BITMAP(name,bits) \
+	unsigned long name[BITS_TO_LONGS(bits)]
+
+struct list_head {
+	struct list_head *next, *prev;
+};
+
+struct hlist_head {
+	struct hlist_node *first;
+};
+
+struct hlist_node {
+	struct hlist_node *next, **pprev;
+};
+
+#endif
diff --git a/tools/perf/util/levenshtein.c b/tools/perf/util/levenshtein.c
new file mode 100644
index 00000000..e521d151
--- /dev/null
+++ b/tools/perf/util/levenshtein.c
@@ -0,0 +1,84 @@
+#include "cache.h"
+#include "levenshtein.h"
+
+/*
+ * This function implements the Damerau-Levenshtein algorithm to
+ * calculate a distance between strings.
+ *
+ * Basically, it says how many letters need to be swapped, substituted,
+ * deleted from, or added to string1, at least, to get string2.
+ *
+ * The idea is to build a distance matrix for the substrings of both
+ * strings.  To avoid a large space complexity, only the last three rows
+ * are kept in memory (if swaps had the same or higher cost as one deletion
+ * plus one insertion, only two rows would be needed).
+ *
+ * At any stage, "i + 1" denotes the length of the current substring of
+ * string1 that the distance is calculated for.
+ *
+ * row2 holds the current row, row1 the previous row (i.e. for the substring
+ * of string1 of length "i"), and row0 the row before that.
+ *
+ * In other words, at the start of the big loop, row2[j + 1] contains the
+ * Damerau-Levenshtein distance between the substring of string1 of length
+ * "i" and the substring of string2 of length "j + 1".
+ *
+ * All the big loop does is determine the partial minimum-cost paths.
+ *
+ * It does so by calculating the costs of the path ending in characters
+ * i (in string1) and j (in string2), respectively, given that the last
+ * operation is a substition, a swap, a deletion, or an insertion.
+ *
+ * This implementation allows the costs to be weighted:
+ *
+ * - w (as in "sWap")
+ * - s (as in "Substitution")
+ * - a (for insertion, AKA "Add")
+ * - d (as in "Deletion")
+ *
+ * Note that this algorithm calculates a distance _iff_ d == a.
+ */
+int levenshtein(const char *string1, const char *string2,
+		int w, int s, int a, int d)
+{
+	int len1 = strlen(string1), len2 = strlen(string2);
+	int *row0 = malloc(sizeof(int) * (len2 + 1));
+	int *row1 = malloc(sizeof(int) * (len2 + 1));
+	int *row2 = malloc(sizeof(int) * (len2 + 1));
+	int i, j;
+
+	for (j = 0; j <= len2; j++)
+		row1[j] = j * a;
+	for (i = 0; i < len1; i++) {
+		int *dummy;
+
+		row2[0] = (i + 1) * d;
+		for (j = 0; j < len2; j++) {
+			/* substitution */
+			row2[j + 1] = row1[j] + s * (string1[i] != string2[j]);
+			/* swap */
+			if (i > 0 && j > 0 && string1[i - 1] == string2[j] &&
+					string1[i] == string2[j - 1] &&
+					row2[j + 1] > row0[j - 1] + w)
+				row2[j + 1] = row0[j - 1] + w;
+			/* deletion */
+			if (row2[j + 1] > row1[j + 1] + d)
+				row2[j + 1] = row1[j + 1] + d;
+			/* insertion */
+			if (row2[j + 1] > row2[j] + a)
+				row2[j + 1] = row2[j] + a;
+		}
+
+		dummy = row0;
+		row0 = row1;
+		row1 = row2;
+		row2 = dummy;
+	}
+
+	i = row1[len2];
+	free(row0);
+	free(row1);
+	free(row2);
+
+	return i;
+}
diff --git a/tools/perf/util/levenshtein.h b/tools/perf/util/levenshtein.h
new file mode 100644
index 00000000..b0fcb6d8
--- /dev/null
+++ b/tools/perf/util/levenshtein.h
@@ -0,0 +1,8 @@
+#ifndef __PERF_LEVENSHTEIN_H
+#define __PERF_LEVENSHTEIN_H
+
+int levenshtein(const char *string1, const char *string2,
+	int swap_penalty, int substition_penalty,
+	int insertion_penalty, int deletion_penalty);
+
+#endif /* __PERF_LEVENSHTEIN_H */
diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c
new file mode 100644
index 00000000..a16ecab5
--- /dev/null
+++ b/tools/perf/util/map.c
@@ -0,0 +1,683 @@
+#include "symbol.h"
+#include <errno.h>
+#include <inttypes.h>
+#include <limits.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdio.h>
+#include <unistd.h>
+#include "map.h"
+
+const char *map_type__name[MAP__NR_TYPES] = {
+	[MAP__FUNCTION] = "Functions",
+	[MAP__VARIABLE] = "Variables",
+};
+
+static inline int is_anon_memory(const char *filename)
+{
+	return strcmp(filename, "//anon") == 0;
+}
+
+void map__init(struct map *self, enum map_type type,
+	       u64 start, u64 end, u64 pgoff, struct dso *dso)
+{
+	self->type     = type;
+	self->start    = start;
+	self->end      = end;
+	self->pgoff    = pgoff;
+	self->dso      = dso;
+	self->map_ip   = map__map_ip;
+	self->unmap_ip = map__unmap_ip;
+	RB_CLEAR_NODE(&self->rb_node);
+	self->groups   = NULL;
+	self->referenced = false;
+}
+
+struct map *map__new(struct list_head *dsos__list, u64 start, u64 len,
+		     u64 pgoff, u32 pid, char *filename,
+		     enum map_type type)
+{
+	struct map *self = malloc(sizeof(*self));
+
+	if (self != NULL) {
+		char newfilename[PATH_MAX];
+		struct dso *dso;
+		int anon;
+
+		anon = is_anon_memory(filename);
+
+		if (anon) {
+			snprintf(newfilename, sizeof(newfilename), "/tmp/perf-%d.map", pid);
+			filename = newfilename;
+		}
+
+		dso = __dsos__findnew(dsos__list, filename);
+		if (dso == NULL)
+			goto out_delete;
+
+		map__init(self, type, start, start + len, pgoff, dso);
+
+		if (anon) {
+set_identity:
+			self->map_ip = self->unmap_ip = identity__map_ip;
+		} else if (strcmp(filename, "[vdso]") == 0) {
+			dso__set_loaded(dso, self->type);
+			goto set_identity;
+		}
+	}
+	return self;
+out_delete:
+	free(self);
+	return NULL;
+}
+
+void map__delete(struct map *self)
+{
+	free(self);
+}
+
+void map__fixup_start(struct map *self)
+{
+	struct rb_root *symbols = &self->dso->symbols[self->type];
+	struct rb_node *nd = rb_first(symbols);
+	if (nd != NULL) {
+		struct symbol *sym = rb_entry(nd, struct symbol, rb_node);
+		self->start = sym->start;
+	}
+}
+
+void map__fixup_end(struct map *self)
+{
+	struct rb_root *symbols = &self->dso->symbols[self->type];
+	struct rb_node *nd = rb_last(symbols);
+	if (nd != NULL) {
+		struct symbol *sym = rb_entry(nd, struct symbol, rb_node);
+		self->end = sym->end;
+	}
+}
+
+#define DSO__DELETED "(deleted)"
+
+int map__load(struct map *self, symbol_filter_t filter)
+{
+	const char *name = self->dso->long_name;
+	int nr;
+
+	if (dso__loaded(self->dso, self->type))
+		return 0;
+
+	nr = dso__load(self->dso, self, filter);
+	if (nr < 0) {
+		if (self->dso->has_build_id) {
+			char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+			build_id__sprintf(self->dso->build_id,
+					  sizeof(self->dso->build_id),
+					  sbuild_id);
+			pr_warning("%s with build id %s not found",
+				   name, sbuild_id);
+		} else
+			pr_warning("Failed to open %s", name);
+
+		pr_warning(", continuing without symbols\n");
+		return -1;
+	} else if (nr == 0) {
+		const size_t len = strlen(name);
+		const size_t real_len = len - sizeof(DSO__DELETED);
+
+		if (len > sizeof(DSO__DELETED) &&
+		    strcmp(name + real_len + 1, DSO__DELETED) == 0) {
+			pr_warning("%.*s was updated, restart the long "
+				   "running apps that use it!\n",
+				   (int)real_len, name);
+		} else {
+			pr_warning("no symbols found in %s, maybe install "
+				   "a debug package?\n", name);
+		}
+
+		return -1;
+	}
+	/*
+	 * Only applies to the kernel, as its symtabs aren't relative like the
+	 * module ones.
+	 */
+	if (self->dso->kernel)
+		map__reloc_vmlinux(self);
+
+	return 0;
+}
+
+struct symbol *map__find_symbol(struct map *self, u64 addr,
+				symbol_filter_t filter)
+{
+	if (map__load(self, filter) < 0)
+		return NULL;
+
+	return dso__find_symbol(self->dso, self->type, addr);
+}
+
+struct symbol *map__find_symbol_by_name(struct map *self, const char *name,
+					symbol_filter_t filter)
+{
+	if (map__load(self, filter) < 0)
+		return NULL;
+
+	if (!dso__sorted_by_name(self->dso, self->type))
+		dso__sort_by_name(self->dso, self->type);
+
+	return dso__find_symbol_by_name(self->dso, self->type, name);
+}
+
+struct map *map__clone(struct map *self)
+{
+	struct map *map = malloc(sizeof(*self));
+
+	if (!map)
+		return NULL;
+
+	memcpy(map, self, sizeof(*self));
+
+	return map;
+}
+
+int map__overlap(struct map *l, struct map *r)
+{
+	if (l->start > r->start) {
+		struct map *t = l;
+		l = r;
+		r = t;
+	}
+
+	if (l->end > r->start)
+		return 1;
+
+	return 0;
+}
+
+size_t map__fprintf(struct map *self, FILE *fp)
+{
+	return fprintf(fp, " %" PRIx64 "-%" PRIx64 " %" PRIx64 " %s\n",
+		       self->start, self->end, self->pgoff, self->dso->name);
+}
+
+/*
+ * objdump wants/reports absolute IPs for ET_EXEC, and RIPs for ET_DYN.
+ * map->dso->adjust_symbols==1 for ET_EXEC-like cases.
+ */
+u64 map__rip_2objdump(struct map *map, u64 rip)
+{
+	u64 addr = map->dso->adjust_symbols ?
+			map->unmap_ip(map, rip) :	/* RIP -> IP */
+			rip;
+	return addr;
+}
+
+u64 map__objdump_2ip(struct map *map, u64 addr)
+{
+	u64 ip = map->dso->adjust_symbols ?
+			addr :
+			map->unmap_ip(map, addr);	/* RIP -> IP */
+	return ip;
+}
+
+void map_groups__init(struct map_groups *self)
+{
+	int i;
+	for (i = 0; i < MAP__NR_TYPES; ++i) {
+		self->maps[i] = RB_ROOT;
+		INIT_LIST_HEAD(&self->removed_maps[i]);
+	}
+	self->machine = NULL;
+}
+
+static void maps__delete(struct rb_root *self)
+{
+	struct rb_node *next = rb_first(self);
+
+	while (next) {
+		struct map *pos = rb_entry(next, struct map, rb_node);
+
+		next = rb_next(&pos->rb_node);
+		rb_erase(&pos->rb_node, self);
+		map__delete(pos);
+	}
+}
+
+static void maps__delete_removed(struct list_head *self)
+{
+	struct map *pos, *n;
+
+	list_for_each_entry_safe(pos, n, self, node) {
+		list_del(&pos->node);
+		map__delete(pos);
+	}
+}
+
+void map_groups__exit(struct map_groups *self)
+{
+	int i;
+
+	for (i = 0; i < MAP__NR_TYPES; ++i) {
+		maps__delete(&self->maps[i]);
+		maps__delete_removed(&self->removed_maps[i]);
+	}
+}
+
+void map_groups__flush(struct map_groups *self)
+{
+	int type;
+
+	for (type = 0; type < MAP__NR_TYPES; type++) {
+		struct rb_root *root = &self->maps[type];
+		struct rb_node *next = rb_first(root);
+
+		while (next) {
+			struct map *pos = rb_entry(next, struct map, rb_node);
+			next = rb_next(&pos->rb_node);
+			rb_erase(&pos->rb_node, root);
+			/*
+			 * We may have references to this map, for
+			 * instance in some hist_entry instances, so
+			 * just move them to a separate list.
+			 */
+			list_add_tail(&pos->node, &self->removed_maps[pos->type]);
+		}
+	}
+}
+
+struct symbol *map_groups__find_symbol(struct map_groups *self,
+				       enum map_type type, u64 addr,
+				       struct map **mapp,
+				       symbol_filter_t filter)
+{
+	struct map *map = map_groups__find(self, type, addr);
+
+	if (map != NULL) {
+		if (mapp != NULL)
+			*mapp = map;
+		return map__find_symbol(map, map->map_ip(map, addr), filter);
+	}
+
+	return NULL;
+}
+
+struct symbol *map_groups__find_symbol_by_name(struct map_groups *self,
+					       enum map_type type,
+					       const char *name,
+					       struct map **mapp,
+					       symbol_filter_t filter)
+{
+	struct rb_node *nd;
+
+	for (nd = rb_first(&self->maps[type]); nd; nd = rb_next(nd)) {
+		struct map *pos = rb_entry(nd, struct map, rb_node);
+		struct symbol *sym = map__find_symbol_by_name(pos, name, filter);
+
+		if (sym == NULL)
+			continue;
+		if (mapp != NULL)
+			*mapp = pos;
+		return sym;
+	}
+
+	return NULL;
+}
+
+size_t __map_groups__fprintf_maps(struct map_groups *self,
+				  enum map_type type, int verbose, FILE *fp)
+{
+	size_t printed = fprintf(fp, "%s:\n", map_type__name[type]);
+	struct rb_node *nd;
+
+	for (nd = rb_first(&self->maps[type]); nd; nd = rb_next(nd)) {
+		struct map *pos = rb_entry(nd, struct map, rb_node);
+		printed += fprintf(fp, "Map:");
+		printed += map__fprintf(pos, fp);
+		if (verbose > 2) {
+			printed += dso__fprintf(pos->dso, type, fp);
+			printed += fprintf(fp, "--\n");
+		}
+	}
+
+	return printed;
+}
+
+size_t map_groups__fprintf_maps(struct map_groups *self, int verbose, FILE *fp)
+{
+	size_t printed = 0, i;
+	for (i = 0; i < MAP__NR_TYPES; ++i)
+		printed += __map_groups__fprintf_maps(self, i, verbose, fp);
+	return printed;
+}
+
+static size_t __map_groups__fprintf_removed_maps(struct map_groups *self,
+						 enum map_type type,
+						 int verbose, FILE *fp)
+{
+	struct map *pos;
+	size_t printed = 0;
+
+	list_for_each_entry(pos, &self->removed_maps[type], node) {
+		printed += fprintf(fp, "Map:");
+		printed += map__fprintf(pos, fp);
+		if (verbose > 1) {
+			printed += dso__fprintf(pos->dso, type, fp);
+			printed += fprintf(fp, "--\n");
+		}
+	}
+	return printed;
+}
+
+static size_t map_groups__fprintf_removed_maps(struct map_groups *self,
+					       int verbose, FILE *fp)
+{
+	size_t printed = 0, i;
+	for (i = 0; i < MAP__NR_TYPES; ++i)
+		printed += __map_groups__fprintf_removed_maps(self, i, verbose, fp);
+	return printed;
+}
+
+size_t map_groups__fprintf(struct map_groups *self, int verbose, FILE *fp)
+{
+	size_t printed = map_groups__fprintf_maps(self, verbose, fp);
+	printed += fprintf(fp, "Removed maps:\n");
+	return printed + map_groups__fprintf_removed_maps(self, verbose, fp);
+}
+
+int map_groups__fixup_overlappings(struct map_groups *self, struct map *map,
+				   int verbose, FILE *fp)
+{
+	struct rb_root *root = &self->maps[map->type];
+	struct rb_node *next = rb_first(root);
+	int err = 0;
+
+	while (next) {
+		struct map *pos = rb_entry(next, struct map, rb_node);
+		next = rb_next(&pos->rb_node);
+
+		if (!map__overlap(pos, map))
+			continue;
+
+		if (verbose >= 2) {
+			fputs("overlapping maps:\n", fp);
+			map__fprintf(map, fp);
+			map__fprintf(pos, fp);
+		}
+
+		rb_erase(&pos->rb_node, root);
+		/*
+		 * Now check if we need to create new maps for areas not
+		 * overlapped by the new map:
+		 */
+		if (map->start > pos->start) {
+			struct map *before = map__clone(pos);
+
+			if (before == NULL) {
+				err = -ENOMEM;
+				goto move_map;
+			}
+
+			before->end = map->start - 1;
+			map_groups__insert(self, before);
+			if (verbose >= 2)
+				map__fprintf(before, fp);
+		}
+
+		if (map->end < pos->end) {
+			struct map *after = map__clone(pos);
+
+			if (after == NULL) {
+				err = -ENOMEM;
+				goto move_map;
+			}
+
+			after->start = map->end + 1;
+			map_groups__insert(self, after);
+			if (verbose >= 2)
+				map__fprintf(after, fp);
+		}
+move_map:
+		/*
+		 * If we have references, just move them to a separate list.
+		 */
+		if (pos->referenced)
+			list_add_tail(&pos->node, &self->removed_maps[map->type]);
+		else
+			map__delete(pos);
+
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+/*
+ * XXX This should not really _copy_ te maps, but refcount them.
+ */
+int map_groups__clone(struct map_groups *self,
+		      struct map_groups *parent, enum map_type type)
+{
+	struct rb_node *nd;
+	for (nd = rb_first(&parent->maps[type]); nd; nd = rb_next(nd)) {
+		struct map *map = rb_entry(nd, struct map, rb_node);
+		struct map *new = map__clone(map);
+		if (new == NULL)
+			return -ENOMEM;
+		map_groups__insert(self, new);
+	}
+	return 0;
+}
+
+static u64 map__reloc_map_ip(struct map *map, u64 ip)
+{
+	return ip + (s64)map->pgoff;
+}
+
+static u64 map__reloc_unmap_ip(struct map *map, u64 ip)
+{
+	return ip - (s64)map->pgoff;
+}
+
+void map__reloc_vmlinux(struct map *self)
+{
+	struct kmap *kmap = map__kmap(self);
+	s64 reloc;
+
+	if (!kmap->ref_reloc_sym || !kmap->ref_reloc_sym->unrelocated_addr)
+		return;
+
+	reloc = (kmap->ref_reloc_sym->unrelocated_addr -
+		 kmap->ref_reloc_sym->addr);
+
+	if (!reloc)
+		return;
+
+	self->map_ip   = map__reloc_map_ip;
+	self->unmap_ip = map__reloc_unmap_ip;
+	self->pgoff    = reloc;
+}
+
+void maps__insert(struct rb_root *maps, struct map *map)
+{
+	struct rb_node **p = &maps->rb_node;
+	struct rb_node *parent = NULL;
+	const u64 ip = map->start;
+	struct map *m;
+
+	while (*p != NULL) {
+		parent = *p;
+		m = rb_entry(parent, struct map, rb_node);
+		if (ip < m->start)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	rb_link_node(&map->rb_node, parent, p);
+	rb_insert_color(&map->rb_node, maps);
+}
+
+void maps__remove(struct rb_root *self, struct map *map)
+{
+	rb_erase(&map->rb_node, self);
+}
+
+struct map *maps__find(struct rb_root *maps, u64 ip)
+{
+	struct rb_node **p = &maps->rb_node;
+	struct rb_node *parent = NULL;
+	struct map *m;
+
+	while (*p != NULL) {
+		parent = *p;
+		m = rb_entry(parent, struct map, rb_node);
+		if (ip < m->start)
+			p = &(*p)->rb_left;
+		else if (ip > m->end)
+			p = &(*p)->rb_right;
+		else
+			return m;
+	}
+
+	return NULL;
+}
+
+int machine__init(struct machine *self, const char *root_dir, pid_t pid)
+{
+	map_groups__init(&self->kmaps);
+	RB_CLEAR_NODE(&self->rb_node);
+	INIT_LIST_HEAD(&self->user_dsos);
+	INIT_LIST_HEAD(&self->kernel_dsos);
+
+	self->kmaps.machine = self;
+	self->pid	    = pid;
+	self->root_dir      = strdup(root_dir);
+	return self->root_dir == NULL ? -ENOMEM : 0;
+}
+
+static void dsos__delete(struct list_head *self)
+{
+	struct dso *pos, *n;
+
+	list_for_each_entry_safe(pos, n, self, node) {
+		list_del(&pos->node);
+		dso__delete(pos);
+	}
+}
+
+void machine__exit(struct machine *self)
+{
+	map_groups__exit(&self->kmaps);
+	dsos__delete(&self->user_dsos);
+	dsos__delete(&self->kernel_dsos);
+	free(self->root_dir);
+	self->root_dir = NULL;
+}
+
+void machine__delete(struct machine *self)
+{
+	machine__exit(self);
+	free(self);
+}
+
+struct machine *machines__add(struct rb_root *self, pid_t pid,
+			      const char *root_dir)
+{
+	struct rb_node **p = &self->rb_node;
+	struct rb_node *parent = NULL;
+	struct machine *pos, *machine = malloc(sizeof(*machine));
+
+	if (!machine)
+		return NULL;
+
+	if (machine__init(machine, root_dir, pid) != 0) {
+		free(machine);
+		return NULL;
+	}
+
+	while (*p != NULL) {
+		parent = *p;
+		pos = rb_entry(parent, struct machine, rb_node);
+		if (pid < pos->pid)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	rb_link_node(&machine->rb_node, parent, p);
+	rb_insert_color(&machine->rb_node, self);
+
+	return machine;
+}
+
+struct machine *machines__find(struct rb_root *self, pid_t pid)
+{
+	struct rb_node **p = &self->rb_node;
+	struct rb_node *parent = NULL;
+	struct machine *machine;
+	struct machine *default_machine = NULL;
+
+	while (*p != NULL) {
+		parent = *p;
+		machine = rb_entry(parent, struct machine, rb_node);
+		if (pid < machine->pid)
+			p = &(*p)->rb_left;
+		else if (pid > machine->pid)
+			p = &(*p)->rb_right;
+		else
+			return machine;
+		if (!machine->pid)
+			default_machine = machine;
+	}
+
+	return default_machine;
+}
+
+struct machine *machines__findnew(struct rb_root *self, pid_t pid)
+{
+	char path[PATH_MAX];
+	const char *root_dir;
+	struct machine *machine = machines__find(self, pid);
+
+	if (!machine || machine->pid != pid) {
+		if (pid == HOST_KERNEL_ID || pid == DEFAULT_GUEST_KERNEL_ID)
+			root_dir = "";
+		else {
+			if (!symbol_conf.guestmount)
+				goto out;
+			sprintf(path, "%s/%d", symbol_conf.guestmount, pid);
+			if (access(path, R_OK)) {
+				pr_err("Can't access file %s\n", path);
+				goto out;
+			}
+			root_dir = path;
+		}
+		machine = machines__add(self, pid, root_dir);
+	}
+
+out:
+	return machine;
+}
+
+void machines__process(struct rb_root *self, machine__process_t process, void *data)
+{
+	struct rb_node *nd;
+
+	for (nd = rb_first(self); nd; nd = rb_next(nd)) {
+		struct machine *pos = rb_entry(nd, struct machine, rb_node);
+		process(pos, data);
+	}
+}
+
+char *machine__mmap_name(struct machine *self, char *bf, size_t size)
+{
+	if (machine__is_host(self))
+		snprintf(bf, size, "[%s]", "kernel.kallsyms");
+	else if (machine__is_default_guest(self))
+		snprintf(bf, size, "[%s]", "guest.kernel.kallsyms");
+	else
+		snprintf(bf, size, "[%s.%d]", "guest.kernel.kallsyms", self->pid);
+
+	return bf;
+}
diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h
new file mode 100644
index 00000000..b397c038
--- /dev/null
+++ b/tools/perf/util/map.h
@@ -0,0 +1,237 @@
+#ifndef __PERF_MAP_H
+#define __PERF_MAP_H
+
+#include <linux/compiler.h>
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include <stdio.h>
+#include <stdbool.h>
+#include "types.h"
+
+enum map_type {
+	MAP__FUNCTION = 0,
+	MAP__VARIABLE,
+};
+
+#define MAP__NR_TYPES (MAP__VARIABLE + 1)
+
+extern const char *map_type__name[MAP__NR_TYPES];
+
+struct dso;
+struct ref_reloc_sym;
+struct map_groups;
+struct machine;
+
+struct map {
+	union {
+		struct rb_node	rb_node;
+		struct list_head node;
+	};
+	u64			start;
+	u64			end;
+	u8 /* enum map_type */	type;
+	bool			referenced;
+	u32			priv;
+	u64			pgoff;
+
+	/* ip -> dso rip */
+	u64			(*map_ip)(struct map *, u64);
+	/* dso rip -> ip */
+	u64			(*unmap_ip)(struct map *, u64);
+
+	struct dso		*dso;
+	struct map_groups	*groups;
+};
+
+struct kmap {
+	struct ref_reloc_sym	*ref_reloc_sym;
+	struct map_groups	*kmaps;
+};
+
+struct map_groups {
+	struct rb_root	 maps[MAP__NR_TYPES];
+	struct list_head removed_maps[MAP__NR_TYPES];
+	struct machine	 *machine;
+};
+
+/* Native host kernel uses -1 as pid index in machine */
+#define	HOST_KERNEL_ID			(-1)
+#define	DEFAULT_GUEST_KERNEL_ID		(0)
+
+struct machine {
+	struct rb_node	  rb_node;
+	pid_t		  pid;
+	char		  *root_dir;
+	struct list_head  user_dsos;
+	struct list_head  kernel_dsos;
+	struct map_groups kmaps;
+	struct map	  *vmlinux_maps[MAP__NR_TYPES];
+};
+
+static inline
+struct map *machine__kernel_map(struct machine *self, enum map_type type)
+{
+	return self->vmlinux_maps[type];
+}
+
+static inline struct kmap *map__kmap(struct map *self)
+{
+	return (struct kmap *)(self + 1);
+}
+
+static inline u64 map__map_ip(struct map *map, u64 ip)
+{
+	return ip - map->start + map->pgoff;
+}
+
+static inline u64 map__unmap_ip(struct map *map, u64 ip)
+{
+	return ip + map->start - map->pgoff;
+}
+
+static inline u64 identity__map_ip(struct map *map __used, u64 ip)
+{
+	return ip;
+}
+
+
+/* rip/ip <-> addr suitable for passing to `objdump --start-address=` */
+u64 map__rip_2objdump(struct map *map, u64 rip);
+u64 map__objdump_2ip(struct map *map, u64 addr);
+
+struct symbol;
+
+typedef int (*symbol_filter_t)(struct map *map, struct symbol *sym);
+
+void map__init(struct map *self, enum map_type type,
+	       u64 start, u64 end, u64 pgoff, struct dso *dso);
+struct map *map__new(struct list_head *dsos__list, u64 start, u64 len,
+		     u64 pgoff, u32 pid, char *filename,
+		     enum map_type type);
+void map__delete(struct map *self);
+struct map *map__clone(struct map *self);
+int map__overlap(struct map *l, struct map *r);
+size_t map__fprintf(struct map *self, FILE *fp);
+
+int map__load(struct map *self, symbol_filter_t filter);
+struct symbol *map__find_symbol(struct map *self,
+				u64 addr, symbol_filter_t filter);
+struct symbol *map__find_symbol_by_name(struct map *self, const char *name,
+					symbol_filter_t filter);
+void map__fixup_start(struct map *self);
+void map__fixup_end(struct map *self);
+
+void map__reloc_vmlinux(struct map *self);
+
+size_t __map_groups__fprintf_maps(struct map_groups *self,
+				  enum map_type type, int verbose, FILE *fp);
+void maps__insert(struct rb_root *maps, struct map *map);
+void maps__remove(struct rb_root *self, struct map *map);
+struct map *maps__find(struct rb_root *maps, u64 addr);
+void map_groups__init(struct map_groups *self);
+void map_groups__exit(struct map_groups *self);
+int map_groups__clone(struct map_groups *self,
+		      struct map_groups *parent, enum map_type type);
+size_t map_groups__fprintf(struct map_groups *self, int verbose, FILE *fp);
+size_t map_groups__fprintf_maps(struct map_groups *self, int verbose, FILE *fp);
+
+typedef void (*machine__process_t)(struct machine *self, void *data);
+
+void machines__process(struct rb_root *self, machine__process_t process, void *data);
+struct machine *machines__add(struct rb_root *self, pid_t pid,
+			      const char *root_dir);
+struct machine *machines__find_host(struct rb_root *self);
+struct machine *machines__find(struct rb_root *self, pid_t pid);
+struct machine *machines__findnew(struct rb_root *self, pid_t pid);
+char *machine__mmap_name(struct machine *self, char *bf, size_t size);
+int machine__init(struct machine *self, const char *root_dir, pid_t pid);
+void machine__exit(struct machine *self);
+void machine__delete(struct machine *self);
+
+/*
+ * Default guest kernel is defined by parameter --guestkallsyms
+ * and --guestmodules
+ */
+static inline bool machine__is_default_guest(struct machine *self)
+{
+	return self ? self->pid == DEFAULT_GUEST_KERNEL_ID : false;
+}
+
+static inline bool machine__is_host(struct machine *self)
+{
+	return self ? self->pid == HOST_KERNEL_ID : false;
+}
+
+static inline void map_groups__insert(struct map_groups *self, struct map *map)
+{
+	maps__insert(&self->maps[map->type], map);
+	map->groups = self;
+}
+
+static inline void map_groups__remove(struct map_groups *self, struct map *map)
+{
+	maps__remove(&self->maps[map->type], map);
+}
+
+static inline struct map *map_groups__find(struct map_groups *self,
+					   enum map_type type, u64 addr)
+{
+	return maps__find(&self->maps[type], addr);
+}
+
+struct symbol *map_groups__find_symbol(struct map_groups *self,
+				       enum map_type type, u64 addr,
+				       struct map **mapp,
+				       symbol_filter_t filter);
+
+struct symbol *map_groups__find_symbol_by_name(struct map_groups *self,
+					       enum map_type type,
+					       const char *name,
+					       struct map **mapp,
+					       symbol_filter_t filter);
+
+static inline
+struct symbol *machine__find_kernel_symbol(struct machine *self,
+					   enum map_type type, u64 addr,
+					   struct map **mapp,
+					   symbol_filter_t filter)
+{
+	return map_groups__find_symbol(&self->kmaps, type, addr, mapp, filter);
+}
+
+static inline
+struct symbol *machine__find_kernel_function(struct machine *self, u64 addr,
+					     struct map **mapp,
+					     symbol_filter_t filter)
+{
+	return machine__find_kernel_symbol(self, MAP__FUNCTION, addr, mapp, filter);
+}
+
+static inline
+struct symbol *map_groups__find_function_by_name(struct map_groups *self,
+						 const char *name, struct map **mapp,
+						 symbol_filter_t filter)
+{
+	return map_groups__find_symbol_by_name(self, MAP__FUNCTION, name, mapp, filter);
+}
+
+static inline
+struct symbol *machine__find_kernel_function_by_name(struct machine *self,
+						     const char *name,
+						     struct map **mapp,
+						     symbol_filter_t filter)
+{
+	return map_groups__find_function_by_name(&self->kmaps, name, mapp,
+						 filter);
+}
+
+int map_groups__fixup_overlappings(struct map_groups *self, struct map *map,
+				   int verbose, FILE *fp);
+
+struct map *map_groups__find_by_name(struct map_groups *self,
+				     enum map_type type, const char *name);
+struct map *machine__new_module(struct machine *self, u64 start, const char *filename);
+
+void map_groups__flush(struct map_groups *self);
+
+#endif /* __PERF_MAP_H */
diff --git a/tools/perf/util/pager.c b/tools/perf/util/pager.c
new file mode 100644
index 00000000..1915de20
--- /dev/null
+++ b/tools/perf/util/pager.c
@@ -0,0 +1,96 @@
+#include "cache.h"
+#include "run-command.h"
+#include "sigchain.h"
+
+/*
+ * This is split up from the rest of git so that we can do
+ * something different on Windows.
+ */
+
+static int spawned_pager;
+
+static void pager_preexec(void)
+{
+	/*
+	 * Work around bug in "less" by not starting it until we
+	 * have real input
+	 */
+	fd_set in;
+
+	FD_ZERO(&in);
+	FD_SET(0, &in);
+	select(1, &in, NULL, &in, NULL);
+
+	setenv("LESS", "FRSX", 0);
+}
+
+static const char *pager_argv[] = { "sh", "-c", NULL, NULL };
+static struct child_process pager_process;
+
+static void wait_for_pager(void)
+{
+	fflush(stdout);
+	fflush(stderr);
+	/* signal EOF to pager */
+	close(1);
+	close(2);
+	finish_command(&pager_process);
+}
+
+static void wait_for_pager_signal(int signo)
+{
+	wait_for_pager();
+	sigchain_pop(signo);
+	raise(signo);
+}
+
+void setup_pager(void)
+{
+	const char *pager = getenv("PERF_PAGER");
+
+	if (!isatty(1))
+		return;
+	if (!pager) {
+		if (!pager_program)
+			perf_config(perf_default_config, NULL);
+		pager = pager_program;
+	}
+	if (!pager)
+		pager = getenv("PAGER");
+	if (!pager)
+		pager = "less";
+	else if (!*pager || !strcmp(pager, "cat"))
+		return;
+
+	spawned_pager = 1; /* means we are emitting to terminal */
+
+	/* spawn the pager */
+	pager_argv[2] = pager;
+	pager_process.argv = pager_argv;
+	pager_process.in = -1;
+	pager_process.preexec_cb = pager_preexec;
+
+	if (start_command(&pager_process))
+		return;
+
+	/* original process continues, but writes to the pipe */
+	dup2(pager_process.in, 1);
+	if (isatty(2))
+		dup2(pager_process.in, 2);
+	close(pager_process.in);
+
+	/* this makes sure that the parent terminates after the pager */
+	sigchain_push_common(wait_for_pager_signal);
+	atexit(wait_for_pager);
+}
+
+int pager_in_use(void)
+{
+	const char *env;
+
+	if (spawned_pager)
+		return 1;
+
+	env = getenv("PERF_PAGER_IN_USE");
+	return env ? perf_config_bool("PERF_PAGER_IN_USE", env) : 0;
+}
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
new file mode 100644
index 00000000..41982c37
--- /dev/null
+++ b/tools/perf/util/parse-events.c
@@ -0,0 +1,1093 @@
+#include "../../../include/linux/hw_breakpoint.h"
+#include "util.h"
+#include "../perf.h"
+#include "evlist.h"
+#include "evsel.h"
+#include "parse-options.h"
+#include "parse-events.h"
+#include "exec_cmd.h"
+#include "string.h"
+#include "symbol.h"
+#include "cache.h"
+#include "header.h"
+#include "debugfs.h"
+
+struct event_symbol {
+	u8		type;
+	u64		config;
+	const char	*symbol;
+	const char	*alias;
+};
+
+enum event_result {
+	EVT_FAILED,
+	EVT_HANDLED,
+	EVT_HANDLED_ALL
+};
+
+char debugfs_path[MAXPATHLEN];
+
+#define CHW(x) .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_##x
+#define CSW(x) .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_##x
+
+static struct event_symbol event_symbols[] = {
+  { CHW(CPU_CYCLES),			"cpu-cycles",			"cycles"		},
+  { CHW(STALLED_CYCLES_FRONTEND),	"stalled-cycles-frontend",	"idle-cycles-frontend"	},
+  { CHW(STALLED_CYCLES_BACKEND),	"stalled-cycles-backend",	"idle-cycles-backend"	},
+  { CHW(INSTRUCTIONS),			"instructions",			""			},
+  { CHW(CACHE_REFERENCES),		"cache-references",		""			},
+  { CHW(CACHE_MISSES),			"cache-misses",			""			},
+  { CHW(BRANCH_INSTRUCTIONS),		"branch-instructions",		"branches"		},
+  { CHW(BRANCH_MISSES),			"branch-misses",		""			},
+  { CHW(BUS_CYCLES),			"bus-cycles",			""			},
+
+  { CSW(CPU_CLOCK),			"cpu-clock",			""			},
+  { CSW(TASK_CLOCK),			"task-clock",			""			},
+  { CSW(PAGE_FAULTS),			"page-faults",			"faults"		},
+  { CSW(PAGE_FAULTS_MIN),		"minor-faults",			""			},
+  { CSW(PAGE_FAULTS_MAJ),		"major-faults",			""			},
+  { CSW(CONTEXT_SWITCHES),		"context-switches",		"cs"			},
+  { CSW(CPU_MIGRATIONS),		"cpu-migrations",		"migrations"		},
+  { CSW(ALIGNMENT_FAULTS),		"alignment-faults",		""			},
+  { CSW(EMULATION_FAULTS),		"emulation-faults",		""			},
+};
+
+#define __PERF_EVENT_FIELD(config, name) \
+	((config & PERF_EVENT_##name##_MASK) >> PERF_EVENT_##name##_SHIFT)
+
+#define PERF_EVENT_RAW(config)		__PERF_EVENT_FIELD(config, RAW)
+#define PERF_EVENT_CONFIG(config)	__PERF_EVENT_FIELD(config, CONFIG)
+#define PERF_EVENT_TYPE(config)		__PERF_EVENT_FIELD(config, TYPE)
+#define PERF_EVENT_ID(config)		__PERF_EVENT_FIELD(config, EVENT)
+
+static const char *hw_event_names[PERF_COUNT_HW_MAX] = {
+	"cycles",
+	"instructions",
+	"cache-references",
+	"cache-misses",
+	"branches",
+	"branch-misses",
+	"bus-cycles",
+	"stalled-cycles-frontend",
+	"stalled-cycles-backend",
+};
+
+static const char *sw_event_names[PERF_COUNT_SW_MAX] = {
+	"cpu-clock",
+	"task-clock",
+	"page-faults",
+	"context-switches",
+	"CPU-migrations",
+	"minor-faults",
+	"major-faults",
+	"alignment-faults",
+	"emulation-faults",
+};
+
+#define MAX_ALIASES 8
+
+static const char *hw_cache[][MAX_ALIASES] = {
+ { "L1-dcache",	"l1-d",		"l1d",		"L1-data",		},
+ { "L1-icache",	"l1-i",		"l1i",		"L1-instruction",	},
+ { "LLC",	"L2"							},
+ { "dTLB",	"d-tlb",	"Data-TLB",				},
+ { "iTLB",	"i-tlb",	"Instruction-TLB",			},
+ { "branch",	"branches",	"bpu",		"btb",		"bpc",	},
+};
+
+static const char *hw_cache_op[][MAX_ALIASES] = {
+ { "load",	"loads",	"read",					},
+ { "store",	"stores",	"write",				},
+ { "prefetch",	"prefetches",	"speculative-read", "speculative-load",	},
+};
+
+static const char *hw_cache_result[][MAX_ALIASES] = {
+ { "refs",	"Reference",	"ops",		"access",		},
+ { "misses",	"miss",							},
+};
+
+#define C(x)		PERF_COUNT_HW_CACHE_##x
+#define CACHE_READ	(1 << C(OP_READ))
+#define CACHE_WRITE	(1 << C(OP_WRITE))
+#define CACHE_PREFETCH	(1 << C(OP_PREFETCH))
+#define COP(x)		(1 << x)
+
+/*
+ * cache operartion stat
+ * L1I : Read and prefetch only
+ * ITLB and BPU : Read-only
+ */
+static unsigned long hw_cache_stat[C(MAX)] = {
+ [C(L1D)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
+ [C(L1I)]	= (CACHE_READ | CACHE_PREFETCH),
+ [C(LL)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
+ [C(DTLB)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
+ [C(ITLB)]	= (CACHE_READ),
+ [C(BPU)]	= (CACHE_READ),
+};
+
+#define for_each_subsystem(sys_dir, sys_dirent, sys_next)	       \
+	while (!readdir_r(sys_dir, &sys_dirent, &sys_next) && sys_next)	       \
+	if (sys_dirent.d_type == DT_DIR &&				       \
+	   (strcmp(sys_dirent.d_name, ".")) &&				       \
+	   (strcmp(sys_dirent.d_name, "..")))
+
+static int tp_event_has_id(struct dirent *sys_dir, struct dirent *evt_dir)
+{
+	char evt_path[MAXPATHLEN];
+	int fd;
+
+	snprintf(evt_path, MAXPATHLEN, "%s/%s/%s/id", debugfs_path,
+			sys_dir->d_name, evt_dir->d_name);
+	fd = open(evt_path, O_RDONLY);
+	if (fd < 0)
+		return -EINVAL;
+	close(fd);
+
+	return 0;
+}
+
+#define for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next)	       \
+	while (!readdir_r(evt_dir, &evt_dirent, &evt_next) && evt_next)        \
+	if (evt_dirent.d_type == DT_DIR &&				       \
+	   (strcmp(evt_dirent.d_name, ".")) &&				       \
+	   (strcmp(evt_dirent.d_name, "..")) &&				       \
+	   (!tp_event_has_id(&sys_dirent, &evt_dirent)))
+
+#define MAX_EVENT_LENGTH 512
+
+
+struct tracepoint_path *tracepoint_id_to_path(u64 config)
+{
+	struct tracepoint_path *path = NULL;
+	DIR *sys_dir, *evt_dir;
+	struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent;
+	char id_buf[4];
+	int fd;
+	u64 id;
+	char evt_path[MAXPATHLEN];
+	char dir_path[MAXPATHLEN];
+
+	if (debugfs_valid_mountpoint(debugfs_path))
+		return NULL;
+
+	sys_dir = opendir(debugfs_path);
+	if (!sys_dir)
+		return NULL;
+
+	for_each_subsystem(sys_dir, sys_dirent, sys_next) {
+
+		snprintf(dir_path, MAXPATHLEN, "%s/%s", debugfs_path,
+			 sys_dirent.d_name);
+		evt_dir = opendir(dir_path);
+		if (!evt_dir)
+			continue;
+
+		for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) {
+
+			snprintf(evt_path, MAXPATHLEN, "%s/%s/id", dir_path,
+				 evt_dirent.d_name);
+			fd = open(evt_path, O_RDONLY);
+			if (fd < 0)
+				continue;
+			if (read(fd, id_buf, sizeof(id_buf)) < 0) {
+				close(fd);
+				continue;
+			}
+			close(fd);
+			id = atoll(id_buf);
+			if (id == config) {
+				closedir(evt_dir);
+				closedir(sys_dir);
+				path = zalloc(sizeof(*path));
+				path->system = malloc(MAX_EVENT_LENGTH);
+				if (!path->system) {
+					free(path);
+					return NULL;
+				}
+				path->name = malloc(MAX_EVENT_LENGTH);
+				if (!path->name) {
+					free(path->system);
+					free(path);
+					return NULL;
+				}
+				strncpy(path->system, sys_dirent.d_name,
+					MAX_EVENT_LENGTH);
+				strncpy(path->name, evt_dirent.d_name,
+					MAX_EVENT_LENGTH);
+				return path;
+			}
+		}
+		closedir(evt_dir);
+	}
+
+	closedir(sys_dir);
+	return NULL;
+}
+
+#define TP_PATH_LEN (MAX_EVENT_LENGTH * 2 + 1)
+static const char *tracepoint_id_to_name(u64 config)
+{
+	static char buf[TP_PATH_LEN];
+	struct tracepoint_path *path;
+
+	path = tracepoint_id_to_path(config);
+	if (path) {
+		snprintf(buf, TP_PATH_LEN, "%s:%s", path->system, path->name);
+		free(path->name);
+		free(path->system);
+		free(path);
+	} else
+		snprintf(buf, TP_PATH_LEN, "%s:%s", "unknown", "unknown");
+
+	return buf;
+}
+
+static int is_cache_op_valid(u8 cache_type, u8 cache_op)
+{
+	if (hw_cache_stat[cache_type] & COP(cache_op))
+		return 1;	/* valid */
+	else
+		return 0;	/* invalid */
+}
+
+static char *event_cache_name(u8 cache_type, u8 cache_op, u8 cache_result)
+{
+	static char name[50];
+
+	if (cache_result) {
+		sprintf(name, "%s-%s-%s", hw_cache[cache_type][0],
+			hw_cache_op[cache_op][0],
+			hw_cache_result[cache_result][0]);
+	} else {
+		sprintf(name, "%s-%s", hw_cache[cache_type][0],
+			hw_cache_op[cache_op][1]);
+	}
+
+	return name;
+}
+
+const char *event_type(int type)
+{
+	switch (type) {
+	case PERF_TYPE_HARDWARE:
+		return "hardware";
+
+	case PERF_TYPE_SOFTWARE:
+		return "software";
+
+	case PERF_TYPE_TRACEPOINT:
+		return "tracepoint";
+
+	case PERF_TYPE_HW_CACHE:
+		return "hardware-cache";
+
+	default:
+		break;
+	}
+
+	return "unknown";
+}
+
+const char *event_name(struct perf_evsel *evsel)
+{
+	u64 config = evsel->attr.config;
+	int type = evsel->attr.type;
+
+	if (evsel->name)
+		return evsel->name;
+
+	return __event_name(type, config);
+}
+
+const char *__event_name(int type, u64 config)
+{
+	static char buf[32];
+
+	if (type == PERF_TYPE_RAW) {
+		sprintf(buf, "raw 0x%" PRIx64, config);
+		return buf;
+	}
+
+	switch (type) {
+	case PERF_TYPE_HARDWARE:
+		if (config < PERF_COUNT_HW_MAX && hw_event_names[config])
+			return hw_event_names[config];
+		return "unknown-hardware";
+
+	case PERF_TYPE_HW_CACHE: {
+		u8 cache_type, cache_op, cache_result;
+
+		cache_type   = (config >>  0) & 0xff;
+		if (cache_type > PERF_COUNT_HW_CACHE_MAX)
+			return "unknown-ext-hardware-cache-type";
+
+		cache_op     = (config >>  8) & 0xff;
+		if (cache_op > PERF_COUNT_HW_CACHE_OP_MAX)
+			return "unknown-ext-hardware-cache-op";
+
+		cache_result = (config >> 16) & 0xff;
+		if (cache_result > PERF_COUNT_HW_CACHE_RESULT_MAX)
+			return "unknown-ext-hardware-cache-result";
+
+		if (!is_cache_op_valid(cache_type, cache_op))
+			return "invalid-cache";
+
+		return event_cache_name(cache_type, cache_op, cache_result);
+	}
+
+	case PERF_TYPE_SOFTWARE:
+		if (config < PERF_COUNT_SW_MAX && sw_event_names[config])
+			return sw_event_names[config];
+		return "unknown-software";
+
+	case PERF_TYPE_TRACEPOINT:
+		return tracepoint_id_to_name(config);
+
+	default:
+		break;
+	}
+
+	return "unknown";
+}
+
+static int parse_aliases(const char **str, const char *names[][MAX_ALIASES], int size)
+{
+	int i, j;
+	int n, longest = -1;
+
+	for (i = 0; i < size; i++) {
+		for (j = 0; j < MAX_ALIASES && names[i][j]; j++) {
+			n = strlen(names[i][j]);
+			if (n > longest && !strncasecmp(*str, names[i][j], n))
+				longest = n;
+		}
+		if (longest > 0) {
+			*str += longest;
+			return i;
+		}
+	}
+
+	return -1;
+}
+
+static enum event_result
+parse_generic_hw_event(const char **str, struct perf_event_attr *attr)
+{
+	const char *s = *str;
+	int cache_type = -1, cache_op = -1, cache_result = -1;
+
+	cache_type = parse_aliases(&s, hw_cache, PERF_COUNT_HW_CACHE_MAX);
+	/*
+	 * No fallback - if we cannot get a clear cache type
+	 * then bail out:
+	 */
+	if (cache_type == -1)
+		return EVT_FAILED;
+
+	while ((cache_op == -1 || cache_result == -1) && *s == '-') {
+		++s;
+
+		if (cache_op == -1) {
+			cache_op = parse_aliases(&s, hw_cache_op,
+						PERF_COUNT_HW_CACHE_OP_MAX);
+			if (cache_op >= 0) {
+				if (!is_cache_op_valid(cache_type, cache_op))
+					return 0;
+				continue;
+			}
+		}
+
+		if (cache_result == -1) {
+			cache_result = parse_aliases(&s, hw_cache_result,
+						PERF_COUNT_HW_CACHE_RESULT_MAX);
+			if (cache_result >= 0)
+				continue;
+		}
+
+		/*
+		 * Can't parse this as a cache op or result, so back up
+		 * to the '-'.
+		 */
+		--s;
+		break;
+	}
+
+	/*
+	 * Fall back to reads:
+	 */
+	if (cache_op == -1)
+		cache_op = PERF_COUNT_HW_CACHE_OP_READ;
+
+	/*
+	 * Fall back to accesses:
+	 */
+	if (cache_result == -1)
+		cache_result = PERF_COUNT_HW_CACHE_RESULT_ACCESS;
+
+	attr->config = cache_type | (cache_op << 8) | (cache_result << 16);
+	attr->type = PERF_TYPE_HW_CACHE;
+
+	*str = s;
+	return EVT_HANDLED;
+}
+
+static enum event_result
+parse_single_tracepoint_event(char *sys_name,
+			      const char *evt_name,
+			      unsigned int evt_length,
+			      struct perf_event_attr *attr,
+			      const char **strp)
+{
+	char evt_path[MAXPATHLEN];
+	char id_buf[4];
+	u64 id;
+	int fd;
+
+	snprintf(evt_path, MAXPATHLEN, "%s/%s/%s/id", debugfs_path,
+		 sys_name, evt_name);
+
+	fd = open(evt_path, O_RDONLY);
+	if (fd < 0)
+		return EVT_FAILED;
+
+	if (read(fd, id_buf, sizeof(id_buf)) < 0) {
+		close(fd);
+		return EVT_FAILED;
+	}
+
+	close(fd);
+	id = atoll(id_buf);
+	attr->config = id;
+	attr->type = PERF_TYPE_TRACEPOINT;
+	*strp += strlen(sys_name) + evt_length + 1; /* + 1 for the ':' */
+
+	attr->sample_type |= PERF_SAMPLE_RAW;
+	attr->sample_type |= PERF_SAMPLE_TIME;
+	attr->sample_type |= PERF_SAMPLE_CPU;
+
+	attr->sample_period = 1;
+
+
+	return EVT_HANDLED;
+}
+
+/* sys + ':' + event + ':' + flags*/
+#define MAX_EVOPT_LEN	(MAX_EVENT_LENGTH * 2 + 2 + 128)
+static enum event_result
+parse_multiple_tracepoint_event(const struct option *opt, char *sys_name,
+				const char *evt_exp, char *flags)
+{
+	char evt_path[MAXPATHLEN];
+	struct dirent *evt_ent;
+	DIR *evt_dir;
+
+	snprintf(evt_path, MAXPATHLEN, "%s/%s", debugfs_path, sys_name);
+	evt_dir = opendir(evt_path);
+
+	if (!evt_dir) {
+		perror("Can't open event dir");
+		return EVT_FAILED;
+	}
+
+	while ((evt_ent = readdir(evt_dir))) {
+		char event_opt[MAX_EVOPT_LEN + 1];
+		int len;
+
+		if (!strcmp(evt_ent->d_name, ".")
+		    || !strcmp(evt_ent->d_name, "..")
+		    || !strcmp(evt_ent->d_name, "enable")
+		    || !strcmp(evt_ent->d_name, "filter"))
+			continue;
+
+		if (!strglobmatch(evt_ent->d_name, evt_exp))
+			continue;
+
+		len = snprintf(event_opt, MAX_EVOPT_LEN, "%s:%s%s%s", sys_name,
+			       evt_ent->d_name, flags ? ":" : "",
+			       flags ?: "");
+		if (len < 0)
+			return EVT_FAILED;
+
+		if (parse_events(opt, event_opt, 0))
+			return EVT_FAILED;
+	}
+
+	return EVT_HANDLED_ALL;
+}
+
+static enum event_result
+parse_tracepoint_event(const struct option *opt, const char **strp,
+		       struct perf_event_attr *attr)
+{
+	const char *evt_name;
+	char *flags = NULL, *comma_loc;
+	char sys_name[MAX_EVENT_LENGTH];
+	unsigned int sys_length, evt_length;
+
+	if (debugfs_valid_mountpoint(debugfs_path))
+		return 0;
+
+	evt_name = strchr(*strp, ':');
+	if (!evt_name)
+		return EVT_FAILED;
+
+	sys_length = evt_name - *strp;
+	if (sys_length >= MAX_EVENT_LENGTH)
+		return 0;
+
+	strncpy(sys_name, *strp, sys_length);
+	sys_name[sys_length] = '\0';
+	evt_name = evt_name + 1;
+
+	comma_loc = strchr(evt_name, ',');
+	if (comma_loc) {
+		/* take the event name up to the comma */
+		evt_name = strndup(evt_name, comma_loc - evt_name);
+	}
+	flags = strchr(evt_name, ':');
+	if (flags) {
+		/* split it out: */
+		evt_name = strndup(evt_name, flags - evt_name);
+		flags++;
+	}
+
+	evt_length = strlen(evt_name);
+	if (evt_length >= MAX_EVENT_LENGTH)
+		return EVT_FAILED;
+	if (strpbrk(evt_name, "*?")) {
+		*strp += strlen(sys_name) + evt_length + 1; /* 1 == the ':' */
+		return parse_multiple_tracepoint_event(opt, sys_name, evt_name,
+						       flags);
+	} else {
+		return parse_single_tracepoint_event(sys_name, evt_name,
+						     evt_length, attr, strp);
+	}
+}
+
+static enum event_result
+parse_breakpoint_type(const char *type, const char **strp,
+		      struct perf_event_attr *attr)
+{
+	int i;
+
+	for (i = 0; i < 3; i++) {
+		if (!type[i])
+			break;
+
+		switch (type[i]) {
+		case 'r':
+			attr->bp_type |= HW_BREAKPOINT_R;
+			break;
+		case 'w':
+			attr->bp_type |= HW_BREAKPOINT_W;
+			break;
+		case 'x':
+			attr->bp_type |= HW_BREAKPOINT_X;
+			break;
+		default:
+			return EVT_FAILED;
+		}
+	}
+	if (!attr->bp_type) /* Default */
+		attr->bp_type = HW_BREAKPOINT_R | HW_BREAKPOINT_W;
+
+	*strp = type + i;
+
+	return EVT_HANDLED;
+}
+
+static enum event_result
+parse_breakpoint_event(const char **strp, struct perf_event_attr *attr)
+{
+	const char *target;
+	const char *type;
+	char *endaddr;
+	u64 addr;
+	enum event_result err;
+
+	target = strchr(*strp, ':');
+	if (!target)
+		return EVT_FAILED;
+
+	if (strncmp(*strp, "mem", target - *strp) != 0)
+		return EVT_FAILED;
+
+	target++;
+
+	addr = strtoull(target, &endaddr, 0);
+	if (target == endaddr)
+		return EVT_FAILED;
+
+	attr->bp_addr = addr;
+	*strp = endaddr;
+
+	type = strchr(target, ':');
+
+	/* If no type is defined, just rw as default */
+	if (!type) {
+		attr->bp_type = HW_BREAKPOINT_R | HW_BREAKPOINT_W;
+	} else {
+		err = parse_breakpoint_type(++type, strp, attr);
+		if (err == EVT_FAILED)
+			return EVT_FAILED;
+	}
+
+	/*
+	 * We should find a nice way to override the access length
+	 * Provide some defaults for now
+	 */
+	if (attr->bp_type == HW_BREAKPOINT_X)
+		attr->bp_len = sizeof(long);
+	else
+		attr->bp_len = HW_BREAKPOINT_LEN_4;
+
+	attr->type = PERF_TYPE_BREAKPOINT;
+
+	return EVT_HANDLED;
+}
+
+static int check_events(const char *str, unsigned int i)
+{
+	int n;
+
+	n = strlen(event_symbols[i].symbol);
+	if (!strncasecmp(str, event_symbols[i].symbol, n))
+		return n;
+
+	n = strlen(event_symbols[i].alias);
+	if (n) {
+		if (!strncasecmp(str, event_symbols[i].alias, n))
+			return n;
+	}
+
+	return 0;
+}
+
+static enum event_result
+parse_symbolic_event(const char **strp, struct perf_event_attr *attr)
+{
+	const char *str = *strp;
+	unsigned int i;
+	int n;
+
+	for (i = 0; i < ARRAY_SIZE(event_symbols); i++) {
+		n = check_events(str, i);
+		if (n > 0) {
+			attr->type = event_symbols[i].type;
+			attr->config = event_symbols[i].config;
+			*strp = str + n;
+			return EVT_HANDLED;
+		}
+	}
+	return EVT_FAILED;
+}
+
+static enum event_result
+parse_raw_event(const char **strp, struct perf_event_attr *attr)
+{
+	const char *str = *strp;
+	u64 config;
+	int n;
+
+	if (*str != 'r')
+		return EVT_FAILED;
+	n = hex2u64(str + 1, &config);
+	if (n > 0) {
+		*strp = str + n + 1;
+		attr->type = PERF_TYPE_RAW;
+		attr->config = config;
+		return EVT_HANDLED;
+	}
+	return EVT_FAILED;
+}
+
+static enum event_result
+parse_numeric_event(const char **strp, struct perf_event_attr *attr)
+{
+	const char *str = *strp;
+	char *endp;
+	unsigned long type;
+	u64 config;
+
+	type = strtoul(str, &endp, 0);
+	if (endp > str && type < PERF_TYPE_MAX && *endp == ':') {
+		str = endp + 1;
+		config = strtoul(str, &endp, 0);
+		if (endp > str) {
+			attr->type = type;
+			attr->config = config;
+			*strp = endp;
+			return EVT_HANDLED;
+		}
+	}
+	return EVT_FAILED;
+}
+
+static int
+parse_event_modifier(const char **strp, struct perf_event_attr *attr)
+{
+	const char *str = *strp;
+	int exclude = 0;
+	int eu = 0, ek = 0, eh = 0, precise = 0;
+
+	if (!*str)
+		return 0;
+
+	if (*str == ',')
+		return 0;
+
+	if (*str++ != ':')
+		return -1;
+
+	while (*str) {
+		if (*str == 'u') {
+			if (!exclude)
+				exclude = eu = ek = eh = 1;
+			eu = 0;
+		} else if (*str == 'k') {
+			if (!exclude)
+				exclude = eu = ek = eh = 1;
+			ek = 0;
+		} else if (*str == 'h') {
+			if (!exclude)
+				exclude = eu = ek = eh = 1;
+			eh = 0;
+		} else if (*str == 'p') {
+			precise++;
+		} else
+			break;
+
+		++str;
+	}
+	if (str < *strp + 2)
+		return -1;
+
+	*strp = str;
+
+	attr->exclude_user   = eu;
+	attr->exclude_kernel = ek;
+	attr->exclude_hv     = eh;
+	attr->precise_ip     = precise;
+
+	return 0;
+}
+
+/*
+ * Each event can have multiple symbolic names.
+ * Symbolic names are (almost) exactly matched.
+ */
+static enum event_result
+parse_event_symbols(const struct option *opt, const char **str,
+		    struct perf_event_attr *attr)
+{
+	enum event_result ret;
+
+	ret = parse_tracepoint_event(opt, str, attr);
+	if (ret != EVT_FAILED)
+		goto modifier;
+
+	ret = parse_raw_event(str, attr);
+	if (ret != EVT_FAILED)
+		goto modifier;
+
+	ret = parse_numeric_event(str, attr);
+	if (ret != EVT_FAILED)
+		goto modifier;
+
+	ret = parse_symbolic_event(str, attr);
+	if (ret != EVT_FAILED)
+		goto modifier;
+
+	ret = parse_generic_hw_event(str, attr);
+	if (ret != EVT_FAILED)
+		goto modifier;
+
+	ret = parse_breakpoint_event(str, attr);
+	if (ret != EVT_FAILED)
+		goto modifier;
+
+	fprintf(stderr, "invalid or unsupported event: '%s'\n", *str);
+	fprintf(stderr, "Run 'perf list' for a list of valid events\n");
+	return EVT_FAILED;
+
+modifier:
+	if (parse_event_modifier(str, attr) < 0) {
+		fprintf(stderr, "invalid event modifier: '%s'\n", *str);
+		fprintf(stderr, "Run 'perf list' for a list of valid events and modifiers\n");
+
+		return EVT_FAILED;
+	}
+
+	return ret;
+}
+
+int parse_events(const struct option *opt, const char *str, int unset __used)
+{
+	struct perf_evlist *evlist = *(struct perf_evlist **)opt->value;
+	struct perf_event_attr attr;
+	enum event_result ret;
+	const char *ostr;
+
+	for (;;) {
+		ostr = str;
+		memset(&attr, 0, sizeof(attr));
+		ret = parse_event_symbols(opt, &str, &attr);
+		if (ret == EVT_FAILED)
+			return -1;
+
+		if (!(*str == 0 || *str == ',' || isspace(*str)))
+			return -1;
+
+		if (ret != EVT_HANDLED_ALL) {
+			struct perf_evsel *evsel;
+			evsel = perf_evsel__new(&attr, evlist->nr_entries);
+			if (evsel == NULL)
+				return -1;
+			perf_evlist__add(evlist, evsel);
+
+			evsel->name = calloc(str - ostr + 1, 1);
+			if (!evsel->name)
+				return -1;
+			strncpy(evsel->name, ostr, str - ostr);
+		}
+
+		if (*str == 0)
+			break;
+		if (*str == ',')
+			++str;
+		while (isspace(*str))
+			++str;
+	}
+
+	return 0;
+}
+
+int parse_filter(const struct option *opt, const char *str,
+		 int unset __used)
+{
+	struct perf_evlist *evlist = *(struct perf_evlist **)opt->value;
+	struct perf_evsel *last = NULL;
+
+	if (evlist->nr_entries > 0)
+		last = list_entry(evlist->entries.prev, struct perf_evsel, node);
+
+	if (last == NULL || last->attr.type != PERF_TYPE_TRACEPOINT) {
+		fprintf(stderr,
+			"-F option should follow a -e tracepoint option\n");
+		return -1;
+	}
+
+	last->filter = strdup(str);
+	if (last->filter == NULL) {
+		fprintf(stderr, "not enough memory to hold filter string\n");
+		return -1;
+	}
+
+	return 0;
+}
+
+static const char * const event_type_descriptors[] = {
+	"Hardware event",
+	"Software event",
+	"Tracepoint event",
+	"Hardware cache event",
+	"Raw hardware event descriptor",
+	"Hardware breakpoint",
+};
+
+/*
+ * Print the events from <debugfs_mount_point>/tracing/events
+ */
+
+void print_tracepoint_events(const char *subsys_glob, const char *event_glob)
+{
+	DIR *sys_dir, *evt_dir;
+	struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent;
+	char evt_path[MAXPATHLEN];
+	char dir_path[MAXPATHLEN];
+
+	if (debugfs_valid_mountpoint(debugfs_path))
+		return;
+
+	sys_dir = opendir(debugfs_path);
+	if (!sys_dir)
+		return;
+
+	for_each_subsystem(sys_dir, sys_dirent, sys_next) {
+		if (subsys_glob != NULL && 
+		    !strglobmatch(sys_dirent.d_name, subsys_glob))
+			continue;
+
+		snprintf(dir_path, MAXPATHLEN, "%s/%s", debugfs_path,
+			 sys_dirent.d_name);
+		evt_dir = opendir(dir_path);
+		if (!evt_dir)
+			continue;
+
+		for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) {
+			if (event_glob != NULL && 
+			    !strglobmatch(evt_dirent.d_name, event_glob))
+				continue;
+
+			snprintf(evt_path, MAXPATHLEN, "%s:%s",
+				 sys_dirent.d_name, evt_dirent.d_name);
+			printf("  %-50s [%s]\n", evt_path,
+				event_type_descriptors[PERF_TYPE_TRACEPOINT]);
+		}
+		closedir(evt_dir);
+	}
+	closedir(sys_dir);
+}
+
+/*
+ * Check whether event is in <debugfs_mount_point>/tracing/events
+ */
+
+int is_valid_tracepoint(const char *event_string)
+{
+	DIR *sys_dir, *evt_dir;
+	struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent;
+	char evt_path[MAXPATHLEN];
+	char dir_path[MAXPATHLEN];
+
+	if (debugfs_valid_mountpoint(debugfs_path))
+		return 0;
+
+	sys_dir = opendir(debugfs_path);
+	if (!sys_dir)
+		return 0;
+
+	for_each_subsystem(sys_dir, sys_dirent, sys_next) {
+
+		snprintf(dir_path, MAXPATHLEN, "%s/%s", debugfs_path,
+			 sys_dirent.d_name);
+		evt_dir = opendir(dir_path);
+		if (!evt_dir)
+			continue;
+
+		for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) {
+			snprintf(evt_path, MAXPATHLEN, "%s:%s",
+				 sys_dirent.d_name, evt_dirent.d_name);
+			if (!strcmp(evt_path, event_string)) {
+				closedir(evt_dir);
+				closedir(sys_dir);
+				return 1;
+			}
+		}
+		closedir(evt_dir);
+	}
+	closedir(sys_dir);
+	return 0;
+}
+
+void print_events_type(u8 type)
+{
+	struct event_symbol *syms = event_symbols;
+	unsigned int i;
+	char name[64];
+
+	for (i = 0; i < ARRAY_SIZE(event_symbols); i++, syms++) {
+		if (type != syms->type)
+			continue;
+
+		if (strlen(syms->alias))
+			snprintf(name, sizeof(name),  "%s OR %s",
+				 syms->symbol, syms->alias);
+		else
+			snprintf(name, sizeof(name), "%s", syms->symbol);
+
+		printf("  %-50s [%s]\n", name,
+			event_type_descriptors[type]);
+	}
+}
+
+int print_hwcache_events(const char *event_glob)
+{
+	unsigned int type, op, i, printed = 0;
+
+	for (type = 0; type < PERF_COUNT_HW_CACHE_MAX; type++) {
+		for (op = 0; op < PERF_COUNT_HW_CACHE_OP_MAX; op++) {
+			/* skip invalid cache type */
+			if (!is_cache_op_valid(type, op))
+				continue;
+
+			for (i = 0; i < PERF_COUNT_HW_CACHE_RESULT_MAX; i++) {
+				char *name = event_cache_name(type, op, i);
+
+				if (event_glob != NULL && !strglobmatch(name, event_glob))
+					continue;
+
+				printf("  %-50s [%s]\n", name,
+					event_type_descriptors[PERF_TYPE_HW_CACHE]);
+				++printed;
+			}
+		}
+	}
+
+	return printed;
+}
+
+#define MAX_NAME_LEN 100
+
+/*
+ * Print the help text for the event symbols:
+ */
+void print_events(const char *event_glob)
+{
+	unsigned int i, type, prev_type = -1, printed = 0, ntypes_printed = 0;
+	struct event_symbol *syms = event_symbols;
+	char name[MAX_NAME_LEN];
+
+	printf("\n");
+	printf("List of pre-defined events (to be used in -e):\n");
+
+	for (i = 0; i < ARRAY_SIZE(event_symbols); i++, syms++) {
+		type = syms->type;
+
+		if (type != prev_type && printed) {
+			printf("\n");
+			printed = 0;
+			ntypes_printed++;
+		}
+
+		if (event_glob != NULL && 
+		    !(strglobmatch(syms->symbol, event_glob) ||
+		      (syms->alias && strglobmatch(syms->alias, event_glob))))
+			continue;
+
+		if (strlen(syms->alias))
+			snprintf(name, MAX_NAME_LEN, "%s OR %s", syms->symbol, syms->alias);
+		else
+			strncpy(name, syms->symbol, MAX_NAME_LEN);
+		printf("  %-50s [%s]\n", name,
+			event_type_descriptors[type]);
+
+		prev_type = type;
+		++printed;
+	}
+
+	if (ntypes_printed) {
+		printed = 0;
+		printf("\n");
+	}
+	print_hwcache_events(event_glob);
+
+	if (event_glob != NULL)
+		return;
+
+	printf("\n");
+	printf("  %-50s [%s]\n",
+		"rNNN (see 'perf list --help' on how to encode it)",
+	       event_type_descriptors[PERF_TYPE_RAW]);
+	printf("\n");
+
+	printf("  %-50s [%s]\n",
+			"mem:<addr>[:access]",
+			event_type_descriptors[PERF_TYPE_BREAKPOINT]);
+	printf("\n");
+
+	print_tracepoint_events(NULL, NULL);
+
+	exit(129);
+}
diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h
new file mode 100644
index 00000000..746d3fcb
--- /dev/null
+++ b/tools/perf/util/parse-events.h
@@ -0,0 +1,41 @@
+#ifndef __PERF_PARSE_EVENTS_H
+#define __PERF_PARSE_EVENTS_H
+/*
+ * Parse symbolic events/counts passed in as options:
+ */
+
+#include "../../../include/linux/perf_event.h"
+
+struct list_head;
+struct perf_evsel;
+
+struct option;
+
+struct tracepoint_path {
+	char *system;
+	char *name;
+	struct tracepoint_path *next;
+};
+
+extern struct tracepoint_path *tracepoint_id_to_path(u64 config);
+extern bool have_tracepoints(struct list_head *evlist);
+
+const char *event_type(int type);
+const char *event_name(struct perf_evsel *event);
+extern const char *__event_name(int type, u64 config);
+
+extern int parse_events(const struct option *opt, const char *str, int unset);
+extern int parse_filter(const struct option *opt, const char *str, int unset);
+
+#define EVENTS_HELP_MAX (128*1024)
+
+void print_events(const char *event_glob);
+void print_events_type(u8 type);
+void print_tracepoint_events(const char *subsys_glob, const char *event_glob);
+int print_hwcache_events(const char *event_glob);
+extern int is_valid_tracepoint(const char *event_string);
+
+extern char debugfs_path[];
+extern int valid_debugfs_mount(const char *debugfs);
+
+#endif /* __PERF_PARSE_EVENTS_H */
diff --git a/tools/perf/util/parse-options.c b/tools/perf/util/parse-options.c
new file mode 100644
index 00000000..99d02aa5
--- /dev/null
+++ b/tools/perf/util/parse-options.c
@@ -0,0 +1,577 @@
+#include "util.h"
+#include "parse-options.h"
+#include "cache.h"
+
+#define OPT_SHORT 1
+#define OPT_UNSET 2
+
+static int opterror(const struct option *opt, const char *reason, int flags)
+{
+	if (flags & OPT_SHORT)
+		return error("switch `%c' %s", opt->short_name, reason);
+	if (flags & OPT_UNSET)
+		return error("option `no-%s' %s", opt->long_name, reason);
+	return error("option `%s' %s", opt->long_name, reason);
+}
+
+static int get_arg(struct parse_opt_ctx_t *p, const struct option *opt,
+		   int flags, const char **arg)
+{
+	if (p->opt) {
+		*arg = p->opt;
+		p->opt = NULL;
+	} else if ((opt->flags & PARSE_OPT_LASTARG_DEFAULT) && (p->argc == 1 ||
+		    **(p->argv + 1) == '-')) {
+		*arg = (const char *)opt->defval;
+	} else if (p->argc > 1) {
+		p->argc--;
+		*arg = *++p->argv;
+	} else
+		return opterror(opt, "requires a value", flags);
+	return 0;
+}
+
+static int get_value(struct parse_opt_ctx_t *p,
+		     const struct option *opt, int flags)
+{
+	const char *s, *arg = NULL;
+	const int unset = flags & OPT_UNSET;
+
+	if (unset && p->opt)
+		return opterror(opt, "takes no value", flags);
+	if (unset && (opt->flags & PARSE_OPT_NONEG))
+		return opterror(opt, "isn't available", flags);
+
+	if (!(flags & OPT_SHORT) && p->opt) {
+		switch (opt->type) {
+		case OPTION_CALLBACK:
+			if (!(opt->flags & PARSE_OPT_NOARG))
+				break;
+			/* FALLTHROUGH */
+		case OPTION_BOOLEAN:
+		case OPTION_INCR:
+		case OPTION_BIT:
+		case OPTION_SET_UINT:
+		case OPTION_SET_PTR:
+			return opterror(opt, "takes no value", flags);
+		case OPTION_END:
+		case OPTION_ARGUMENT:
+		case OPTION_GROUP:
+		case OPTION_STRING:
+		case OPTION_INTEGER:
+		case OPTION_UINTEGER:
+		case OPTION_LONG:
+		case OPTION_U64:
+		default:
+			break;
+		}
+	}
+
+	switch (opt->type) {
+	case OPTION_BIT:
+		if (unset)
+			*(int *)opt->value &= ~opt->defval;
+		else
+			*(int *)opt->value |= opt->defval;
+		return 0;
+
+	case OPTION_BOOLEAN:
+		*(bool *)opt->value = unset ? false : true;
+		return 0;
+
+	case OPTION_INCR:
+		*(int *)opt->value = unset ? 0 : *(int *)opt->value + 1;
+		return 0;
+
+	case OPTION_SET_UINT:
+		*(unsigned int *)opt->value = unset ? 0 : opt->defval;
+		return 0;
+
+	case OPTION_SET_PTR:
+		*(void **)opt->value = unset ? NULL : (void *)opt->defval;
+		return 0;
+
+	case OPTION_STRING:
+		if (unset)
+			*(const char **)opt->value = NULL;
+		else if (opt->flags & PARSE_OPT_OPTARG && !p->opt)
+			*(const char **)opt->value = (const char *)opt->defval;
+		else
+			return get_arg(p, opt, flags, (const char **)opt->value);
+		return 0;
+
+	case OPTION_CALLBACK:
+		if (unset)
+			return (*opt->callback)(opt, NULL, 1) ? (-1) : 0;
+		if (opt->flags & PARSE_OPT_NOARG)
+			return (*opt->callback)(opt, NULL, 0) ? (-1) : 0;
+		if (opt->flags & PARSE_OPT_OPTARG && !p->opt)
+			return (*opt->callback)(opt, NULL, 0) ? (-1) : 0;
+		if (get_arg(p, opt, flags, &arg))
+			return -1;
+		return (*opt->callback)(opt, arg, 0) ? (-1) : 0;
+
+	case OPTION_INTEGER:
+		if (unset) {
+			*(int *)opt->value = 0;
+			return 0;
+		}
+		if (opt->flags & PARSE_OPT_OPTARG && !p->opt) {
+			*(int *)opt->value = opt->defval;
+			return 0;
+		}
+		if (get_arg(p, opt, flags, &arg))
+			return -1;
+		*(int *)opt->value = strtol(arg, (char **)&s, 10);
+		if (*s)
+			return opterror(opt, "expects a numerical value", flags);
+		return 0;
+
+	case OPTION_UINTEGER:
+		if (unset) {
+			*(unsigned int *)opt->value = 0;
+			return 0;
+		}
+		if (opt->flags & PARSE_OPT_OPTARG && !p->opt) {
+			*(unsigned int *)opt->value = opt->defval;
+			return 0;
+		}
+		if (get_arg(p, opt, flags, &arg))
+			return -1;
+		*(unsigned int *)opt->value = strtol(arg, (char **)&s, 10);
+		if (*s)
+			return opterror(opt, "expects a numerical value", flags);
+		return 0;
+
+	case OPTION_LONG:
+		if (unset) {
+			*(long *)opt->value = 0;
+			return 0;
+		}
+		if (opt->flags & PARSE_OPT_OPTARG && !p->opt) {
+			*(long *)opt->value = opt->defval;
+			return 0;
+		}
+		if (get_arg(p, opt, flags, &arg))
+			return -1;
+		*(long *)opt->value = strtol(arg, (char **)&s, 10);
+		if (*s)
+			return opterror(opt, "expects a numerical value", flags);
+		return 0;
+
+	case OPTION_U64:
+		if (unset) {
+			*(u64 *)opt->value = 0;
+			return 0;
+		}
+		if (opt->flags & PARSE_OPT_OPTARG && !p->opt) {
+			*(u64 *)opt->value = opt->defval;
+			return 0;
+		}
+		if (get_arg(p, opt, flags, &arg))
+			return -1;
+		*(u64 *)opt->value = strtoull(arg, (char **)&s, 10);
+		if (*s)
+			return opterror(opt, "expects a numerical value", flags);
+		return 0;
+
+	case OPTION_END:
+	case OPTION_ARGUMENT:
+	case OPTION_GROUP:
+	default:
+		die("should not happen, someone must be hit on the forehead");
+	}
+}
+
+static int parse_short_opt(struct parse_opt_ctx_t *p, const struct option *options)
+{
+	for (; options->type != OPTION_END; options++) {
+		if (options->short_name == *p->opt) {
+			p->opt = p->opt[1] ? p->opt + 1 : NULL;
+			return get_value(p, options, OPT_SHORT);
+		}
+	}
+	return -2;
+}
+
+static int parse_long_opt(struct parse_opt_ctx_t *p, const char *arg,
+                          const struct option *options)
+{
+	const char *arg_end = strchr(arg, '=');
+	const struct option *abbrev_option = NULL, *ambiguous_option = NULL;
+	int abbrev_flags = 0, ambiguous_flags = 0;
+
+	if (!arg_end)
+		arg_end = arg + strlen(arg);
+
+	for (; options->type != OPTION_END; options++) {
+		const char *rest;
+		int flags = 0;
+
+		if (!options->long_name)
+			continue;
+
+		rest = skip_prefix(arg, options->long_name);
+		if (options->type == OPTION_ARGUMENT) {
+			if (!rest)
+				continue;
+			if (*rest == '=')
+				return opterror(options, "takes no value", flags);
+			if (*rest)
+				continue;
+			p->out[p->cpidx++] = arg - 2;
+			return 0;
+		}
+		if (!rest) {
+			/* abbreviated? */
+			if (!strncmp(options->long_name, arg, arg_end - arg)) {
+is_abbreviated:
+				if (abbrev_option) {
+					/*
+					 * If this is abbreviated, it is
+					 * ambiguous. So when there is no
+					 * exact match later, we need to
+					 * error out.
+					 */
+					ambiguous_option = abbrev_option;
+					ambiguous_flags = abbrev_flags;
+				}
+				if (!(flags & OPT_UNSET) && *arg_end)
+					p->opt = arg_end + 1;
+				abbrev_option = options;
+				abbrev_flags = flags;
+				continue;
+			}
+			/* negated and abbreviated very much? */
+			if (!prefixcmp("no-", arg)) {
+				flags |= OPT_UNSET;
+				goto is_abbreviated;
+			}
+			/* negated? */
+			if (strncmp(arg, "no-", 3))
+				continue;
+			flags |= OPT_UNSET;
+			rest = skip_prefix(arg + 3, options->long_name);
+			/* abbreviated and negated? */
+			if (!rest && !prefixcmp(options->long_name, arg + 3))
+				goto is_abbreviated;
+			if (!rest)
+				continue;
+		}
+		if (*rest) {
+			if (*rest != '=')
+				continue;
+			p->opt = rest + 1;
+		}
+		return get_value(p, options, flags);
+	}
+
+	if (ambiguous_option)
+		return error("Ambiguous option: %s "
+			"(could be --%s%s or --%s%s)",
+			arg,
+			(ambiguous_flags & OPT_UNSET) ?  "no-" : "",
+			ambiguous_option->long_name,
+			(abbrev_flags & OPT_UNSET) ?  "no-" : "",
+			abbrev_option->long_name);
+	if (abbrev_option)
+		return get_value(p, abbrev_option, abbrev_flags);
+	return -2;
+}
+
+static void check_typos(const char *arg, const struct option *options)
+{
+	if (strlen(arg) < 3)
+		return;
+
+	if (!prefixcmp(arg, "no-")) {
+		error ("did you mean `--%s` (with two dashes ?)", arg);
+		exit(129);
+	}
+
+	for (; options->type != OPTION_END; options++) {
+		if (!options->long_name)
+			continue;
+		if (!prefixcmp(options->long_name, arg)) {
+			error ("did you mean `--%s` (with two dashes ?)", arg);
+			exit(129);
+		}
+	}
+}
+
+void parse_options_start(struct parse_opt_ctx_t *ctx,
+			 int argc, const char **argv, int flags)
+{
+	memset(ctx, 0, sizeof(*ctx));
+	ctx->argc = argc - 1;
+	ctx->argv = argv + 1;
+	ctx->out  = argv;
+	ctx->cpidx = ((flags & PARSE_OPT_KEEP_ARGV0) != 0);
+	ctx->flags = flags;
+	if ((flags & PARSE_OPT_KEEP_UNKNOWN) &&
+	    (flags & PARSE_OPT_STOP_AT_NON_OPTION))
+		die("STOP_AT_NON_OPTION and KEEP_UNKNOWN don't go together");
+}
+
+static int usage_with_options_internal(const char * const *,
+				       const struct option *, int);
+
+int parse_options_step(struct parse_opt_ctx_t *ctx,
+		       const struct option *options,
+		       const char * const usagestr[])
+{
+	int internal_help = !(ctx->flags & PARSE_OPT_NO_INTERNAL_HELP);
+
+	/* we must reset ->opt, unknown short option leave it dangling */
+	ctx->opt = NULL;
+
+	for (; ctx->argc; ctx->argc--, ctx->argv++) {
+		const char *arg = ctx->argv[0];
+
+		if (*arg != '-' || !arg[1]) {
+			if (ctx->flags & PARSE_OPT_STOP_AT_NON_OPTION)
+				break;
+			ctx->out[ctx->cpidx++] = ctx->argv[0];
+			continue;
+		}
+
+		if (arg[1] != '-') {
+			ctx->opt = arg + 1;
+			if (internal_help && *ctx->opt == 'h')
+				return parse_options_usage(usagestr, options);
+			switch (parse_short_opt(ctx, options)) {
+			case -1:
+				return parse_options_usage(usagestr, options);
+			case -2:
+				goto unknown;
+			default:
+				break;
+			}
+			if (ctx->opt)
+				check_typos(arg + 1, options);
+			while (ctx->opt) {
+				if (internal_help && *ctx->opt == 'h')
+					return parse_options_usage(usagestr, options);
+				switch (parse_short_opt(ctx, options)) {
+				case -1:
+					return parse_options_usage(usagestr, options);
+				case -2:
+					/* fake a short option thing to hide the fact that we may have
+					 * started to parse aggregated stuff
+					 *
+					 * This is leaky, too bad.
+					 */
+					ctx->argv[0] = strdup(ctx->opt - 1);
+					*(char *)ctx->argv[0] = '-';
+					goto unknown;
+				default:
+					break;
+				}
+			}
+			continue;
+		}
+
+		if (!arg[2]) { /* "--" */
+			if (!(ctx->flags & PARSE_OPT_KEEP_DASHDASH)) {
+				ctx->argc--;
+				ctx->argv++;
+			}
+			break;
+		}
+
+		if (internal_help && !strcmp(arg + 2, "help-all"))
+			return usage_with_options_internal(usagestr, options, 1);
+		if (internal_help && !strcmp(arg + 2, "help"))
+			return parse_options_usage(usagestr, options);
+		switch (parse_long_opt(ctx, arg + 2, options)) {
+		case -1:
+			return parse_options_usage(usagestr, options);
+		case -2:
+			goto unknown;
+		default:
+			break;
+		}
+		continue;
+unknown:
+		if (!(ctx->flags & PARSE_OPT_KEEP_UNKNOWN))
+			return PARSE_OPT_UNKNOWN;
+		ctx->out[ctx->cpidx++] = ctx->argv[0];
+		ctx->opt = NULL;
+	}
+	return PARSE_OPT_DONE;
+}
+
+int parse_options_end(struct parse_opt_ctx_t *ctx)
+{
+	memmove(ctx->out + ctx->cpidx, ctx->argv, ctx->argc * sizeof(*ctx->out));
+	ctx->out[ctx->cpidx + ctx->argc] = NULL;
+	return ctx->cpidx + ctx->argc;
+}
+
+int parse_options(int argc, const char **argv, const struct option *options,
+		  const char * const usagestr[], int flags)
+{
+	struct parse_opt_ctx_t ctx;
+
+	parse_options_start(&ctx, argc, argv, flags);
+	switch (parse_options_step(&ctx, options, usagestr)) {
+	case PARSE_OPT_HELP:
+		exit(129);
+	case PARSE_OPT_DONE:
+		break;
+	default: /* PARSE_OPT_UNKNOWN */
+		if (ctx.argv[0][1] == '-') {
+			error("unknown option `%s'", ctx.argv[0] + 2);
+		} else {
+			error("unknown switch `%c'", *ctx.opt);
+		}
+		usage_with_options(usagestr, options);
+	}
+
+	return parse_options_end(&ctx);
+}
+
+#define USAGE_OPTS_WIDTH 24
+#define USAGE_GAP         2
+
+int usage_with_options_internal(const char * const *usagestr,
+				const struct option *opts, int full)
+{
+	if (!usagestr)
+		return PARSE_OPT_HELP;
+
+	fprintf(stderr, "\n usage: %s\n", *usagestr++);
+	while (*usagestr && **usagestr)
+		fprintf(stderr, "    or: %s\n", *usagestr++);
+	while (*usagestr) {
+		fprintf(stderr, "%s%s\n",
+				**usagestr ? "    " : "",
+				*usagestr);
+		usagestr++;
+	}
+
+	if (opts->type != OPTION_GROUP)
+		fputc('\n', stderr);
+
+	for (; opts->type != OPTION_END; opts++) {
+		size_t pos;
+		int pad;
+
+		if (opts->type == OPTION_GROUP) {
+			fputc('\n', stderr);
+			if (*opts->help)
+				fprintf(stderr, "%s\n", opts->help);
+			continue;
+		}
+		if (!full && (opts->flags & PARSE_OPT_HIDDEN))
+			continue;
+
+		pos = fprintf(stderr, "    ");
+		if (opts->short_name)
+			pos += fprintf(stderr, "-%c", opts->short_name);
+		else
+			pos += fprintf(stderr, "    ");
+
+		if (opts->long_name && opts->short_name)
+			pos += fprintf(stderr, ", ");
+		if (opts->long_name)
+			pos += fprintf(stderr, "--%s", opts->long_name);
+
+		switch (opts->type) {
+		case OPTION_ARGUMENT:
+			break;
+		case OPTION_LONG:
+		case OPTION_U64:
+		case OPTION_INTEGER:
+		case OPTION_UINTEGER:
+			if (opts->flags & PARSE_OPT_OPTARG)
+				if (opts->long_name)
+					pos += fprintf(stderr, "[=<n>]");
+				else
+					pos += fprintf(stderr, "[<n>]");
+			else
+				pos += fprintf(stderr, " <n>");
+			break;
+		case OPTION_CALLBACK:
+			if (opts->flags & PARSE_OPT_NOARG)
+				break;
+			/* FALLTHROUGH */
+		case OPTION_STRING:
+			if (opts->argh) {
+				if (opts->flags & PARSE_OPT_OPTARG)
+					if (opts->long_name)
+						pos += fprintf(stderr, "[=<%s>]", opts->argh);
+					else
+						pos += fprintf(stderr, "[<%s>]", opts->argh);
+				else
+					pos += fprintf(stderr, " <%s>", opts->argh);
+			} else {
+				if (opts->flags & PARSE_OPT_OPTARG)
+					if (opts->long_name)
+						pos += fprintf(stderr, "[=...]");
+					else
+						pos += fprintf(stderr, "[...]");
+				else
+					pos += fprintf(stderr, " ...");
+			}
+			break;
+		default: /* OPTION_{BIT,BOOLEAN,SET_UINT,SET_PTR} */
+		case OPTION_END:
+		case OPTION_GROUP:
+		case OPTION_BIT:
+		case OPTION_BOOLEAN:
+		case OPTION_INCR:
+		case OPTION_SET_UINT:
+		case OPTION_SET_PTR:
+			break;
+		}
+
+		if (pos <= USAGE_OPTS_WIDTH)
+			pad = USAGE_OPTS_WIDTH - pos;
+		else {
+			fputc('\n', stderr);
+			pad = USAGE_OPTS_WIDTH;
+		}
+		fprintf(stderr, "%*s%s\n", pad + USAGE_GAP, "", opts->help);
+	}
+	fputc('\n', stderr);
+
+	return PARSE_OPT_HELP;
+}
+
+void usage_with_options(const char * const *usagestr,
+			const struct option *opts)
+{
+	exit_browser(false);
+	usage_with_options_internal(usagestr, opts, 0);
+	exit(129);
+}
+
+int parse_options_usage(const char * const *usagestr,
+			const struct option *opts)
+{
+	return usage_with_options_internal(usagestr, opts, 0);
+}
+
+
+int parse_opt_verbosity_cb(const struct option *opt, const char *arg __used,
+			   int unset)
+{
+	int *target = opt->value;
+
+	if (unset)
+		/* --no-quiet, --no-verbose */
+		*target = 0;
+	else if (opt->short_name == 'v') {
+		if (*target >= 0)
+			(*target)++;
+		else
+			*target = 1;
+	} else {
+		if (*target <= 0)
+			(*target)--;
+		else
+			*target = -1;
+	}
+	return 0;
+}
diff --git a/tools/perf/util/parse-options.h b/tools/perf/util/parse-options.h
new file mode 100644
index 00000000..abc31a1d
--- /dev/null
+++ b/tools/perf/util/parse-options.h
@@ -0,0 +1,192 @@
+#ifndef __PERF_PARSE_OPTIONS_H
+#define __PERF_PARSE_OPTIONS_H
+
+#include <linux/kernel.h>
+#include <stdbool.h>
+
+enum parse_opt_type {
+	/* special types */
+	OPTION_END,
+	OPTION_ARGUMENT,
+	OPTION_GROUP,
+	/* options with no arguments */
+	OPTION_BIT,
+	OPTION_BOOLEAN,
+	OPTION_INCR,
+	OPTION_SET_UINT,
+	OPTION_SET_PTR,
+	/* options with arguments (usually) */
+	OPTION_STRING,
+	OPTION_INTEGER,
+	OPTION_LONG,
+	OPTION_CALLBACK,
+	OPTION_U64,
+	OPTION_UINTEGER,
+};
+
+enum parse_opt_flags {
+	PARSE_OPT_KEEP_DASHDASH = 1,
+	PARSE_OPT_STOP_AT_NON_OPTION = 2,
+	PARSE_OPT_KEEP_ARGV0 = 4,
+	PARSE_OPT_KEEP_UNKNOWN = 8,
+	PARSE_OPT_NO_INTERNAL_HELP = 16,
+};
+
+enum parse_opt_option_flags {
+	PARSE_OPT_OPTARG  = 1,
+	PARSE_OPT_NOARG   = 2,
+	PARSE_OPT_NONEG   = 4,
+	PARSE_OPT_HIDDEN  = 8,
+	PARSE_OPT_LASTARG_DEFAULT = 16,
+};
+
+struct option;
+typedef int parse_opt_cb(const struct option *, const char *arg, int unset);
+
+/*
+ * `type`::
+ *   holds the type of the option, you must have an OPTION_END last in your
+ *   array.
+ *
+ * `short_name`::
+ *   the character to use as a short option name, '\0' if none.
+ *
+ * `long_name`::
+ *   the long option name, without the leading dashes, NULL if none.
+ *
+ * `value`::
+ *   stores pointers to the values to be filled.
+ *
+ * `argh`::
+ *   token to explain the kind of argument this option wants. Keep it
+ *   homogenous across the repository.
+ *
+ * `help`::
+ *   the short help associated to what the option does.
+ *   Must never be NULL (except for OPTION_END).
+ *   OPTION_GROUP uses this pointer to store the group header.
+ *
+ * `flags`::
+ *   mask of parse_opt_option_flags.
+ *   PARSE_OPT_OPTARG: says that the argument is optionnal (not for BOOLEANs)
+ *   PARSE_OPT_NOARG: says that this option takes no argument, for CALLBACKs
+ *   PARSE_OPT_NONEG: says that this option cannot be negated
+ *   PARSE_OPT_HIDDEN this option is skipped in the default usage, showed in
+ *                    the long one.
+ *
+ * `callback`::
+ *   pointer to the callback to use for OPTION_CALLBACK.
+ *
+ * `defval`::
+ *   default value to fill (*->value) with for PARSE_OPT_OPTARG.
+ *   OPTION_{BIT,SET_UINT,SET_PTR} store the {mask,integer,pointer} to put in
+ *   the value when met.
+ *   CALLBACKS can use it like they want.
+ */
+struct option {
+	enum parse_opt_type type;
+	int short_name;
+	const char *long_name;
+	void *value;
+	const char *argh;
+	const char *help;
+
+	int flags;
+	parse_opt_cb *callback;
+	intptr_t defval;
+};
+
+#define check_vtype(v, type) ( BUILD_BUG_ON_ZERO(!__builtin_types_compatible_p(typeof(v), type)) + v )
+
+#define OPT_END()                   { .type = OPTION_END }
+#define OPT_ARGUMENT(l, h)          { .type = OPTION_ARGUMENT, .long_name = (l), .help = (h) }
+#define OPT_GROUP(h)                { .type = OPTION_GROUP, .help = (h) }
+#define OPT_BIT(s, l, v, h, b)      { .type = OPTION_BIT, .short_name = (s), .long_name = (l), .value = check_vtype(v, int *), .help = (h), .defval = (b) }
+#define OPT_BOOLEAN(s, l, v, h)     { .type = OPTION_BOOLEAN, .short_name = (s), .long_name = (l), .value = check_vtype(v, bool *), .help = (h) }
+#define OPT_INCR(s, l, v, h)        { .type = OPTION_INCR, .short_name = (s), .long_name = (l), .value = check_vtype(v, int *), .help = (h) }
+#define OPT_SET_UINT(s, l, v, h, i)  { .type = OPTION_SET_UINT, .short_name = (s), .long_name = (l), .value = check_vtype(v, unsigned int *), .help = (h), .defval = (i) }
+#define OPT_SET_PTR(s, l, v, h, p)  { .type = OPTION_SET_PTR, .short_name = (s), .long_name = (l), .value = (v), .help = (h), .defval = (p) }
+#define OPT_INTEGER(s, l, v, h)     { .type = OPTION_INTEGER, .short_name = (s), .long_name = (l), .value = check_vtype(v, int *), .help = (h) }
+#define OPT_UINTEGER(s, l, v, h)    { .type = OPTION_UINTEGER, .short_name = (s), .long_name = (l), .value = check_vtype(v, unsigned int *), .help = (h) }
+#define OPT_LONG(s, l, v, h)        { .type = OPTION_LONG, .short_name = (s), .long_name = (l), .value = check_vtype(v, long *), .help = (h) }
+#define OPT_U64(s, l, v, h)         { .type = OPTION_U64, .short_name = (s), .long_name = (l), .value = check_vtype(v, u64 *), .help = (h) }
+#define OPT_STRING(s, l, v, a, h)   { .type = OPTION_STRING,  .short_name = (s), .long_name = (l), .value = check_vtype(v, const char **), (a), .help = (h) }
+#define OPT_DATE(s, l, v, h) \
+	{ .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), .argh = "time", .help = (h), .callback = parse_opt_approxidate_cb }
+#define OPT_CALLBACK(s, l, v, a, h, f) \
+	{ .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), (a), .help = (h), .callback = (f) }
+#define OPT_CALLBACK_NOOPT(s, l, v, a, h, f) \
+	{ .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), (a), .help = (h), .callback = (f), .flags = PARSE_OPT_NOARG }
+#define OPT_CALLBACK_DEFAULT(s, l, v, a, h, f, d) \
+	{ .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), (a), .help = (h), .callback = (f), .defval = (intptr_t)d, .flags = PARSE_OPT_LASTARG_DEFAULT }
+#define OPT_CALLBACK_DEFAULT_NOOPT(s, l, v, a, h, f, d) \
+	{ .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l),\
+	.value = (v), (a), .help = (h), .callback = (f), .defval = (intptr_t)d,\
+	.flags = PARSE_OPT_LASTARG_DEFAULT | PARSE_OPT_NOARG}
+
+/* parse_options() will filter out the processed options and leave the
+ * non-option argments in argv[].
+ * Returns the number of arguments left in argv[].
+ */
+extern int parse_options(int argc, const char **argv,
+                         const struct option *options,
+                         const char * const usagestr[], int flags);
+
+extern NORETURN void usage_with_options(const char * const *usagestr,
+                                        const struct option *options);
+
+/*----- incremantal advanced APIs -----*/
+
+enum {
+	PARSE_OPT_HELP = -1,
+	PARSE_OPT_DONE,
+	PARSE_OPT_UNKNOWN,
+};
+
+/*
+ * It's okay for the caller to consume argv/argc in the usual way.
+ * Other fields of that structure are private to parse-options and should not
+ * be modified in any way.
+ */
+struct parse_opt_ctx_t {
+	const char **argv;
+	const char **out;
+	int argc, cpidx;
+	const char *opt;
+	int flags;
+};
+
+extern int parse_options_usage(const char * const *usagestr,
+			       const struct option *opts);
+
+extern void parse_options_start(struct parse_opt_ctx_t *ctx,
+				int argc, const char **argv, int flags);
+
+extern int parse_options_step(struct parse_opt_ctx_t *ctx,
+			      const struct option *options,
+			      const char * const usagestr[]);
+
+extern int parse_options_end(struct parse_opt_ctx_t *ctx);
+
+
+/*----- some often used options -----*/
+extern int parse_opt_abbrev_cb(const struct option *, const char *, int);
+extern int parse_opt_approxidate_cb(const struct option *, const char *, int);
+extern int parse_opt_verbosity_cb(const struct option *, const char *, int);
+
+#define OPT__VERBOSE(var)  OPT_BOOLEAN('v', "verbose", (var), "be verbose")
+#define OPT__QUIET(var)    OPT_BOOLEAN('q', "quiet",   (var), "be quiet")
+#define OPT__VERBOSITY(var) \
+	{ OPTION_CALLBACK, 'v', "verbose", (var), NULL, "be more verbose", \
+	  PARSE_OPT_NOARG, &parse_opt_verbosity_cb, 0 }, \
+	{ OPTION_CALLBACK, 'q', "quiet", (var), NULL, "be more quiet", \
+	  PARSE_OPT_NOARG, &parse_opt_verbosity_cb, 0 }
+#define OPT__DRY_RUN(var)  OPT_BOOLEAN('n', "dry-run", (var), "dry run")
+#define OPT__ABBREV(var)  \
+	{ OPTION_CALLBACK, 0, "abbrev", (var), "n", \
+	  "use <n> digits to display SHA-1s", \
+	  PARSE_OPT_OPTARG, &parse_opt_abbrev_cb, 0 }
+
+extern const char *parse_options_fix_filename(const char *prefix, const char *file);
+
+#endif /* __PERF_PARSE_OPTIONS_H */
diff --git a/tools/perf/util/path.c b/tools/perf/util/path.c
new file mode 100644
index 00000000..bd749771
--- /dev/null
+++ b/tools/perf/util/path.c
@@ -0,0 +1,157 @@
+/*
+ * I'm tired of doing "vsnprintf()" etc just to open a
+ * file, so here's a "return static buffer with printf"
+ * interface for paths.
+ *
+ * It's obviously not thread-safe. Sue me. But it's quite
+ * useful for doing things like
+ *
+ *   f = open(mkpath("%s/%s.perf", base, name), O_RDONLY);
+ *
+ * which is what it's designed for.
+ */
+#include "cache.h"
+
+static char bad_path[] = "/bad-path/";
+/*
+ * Two hacks:
+ */
+
+static const char *get_perf_dir(void)
+{
+	return ".";
+}
+
+#ifdef NO_STRLCPY
+size_t strlcpy(char *dest, const char *src, size_t size)
+{
+	size_t ret = strlen(src);
+
+	if (size) {
+		size_t len = (ret >= size) ? size - 1 : ret;
+		memcpy(dest, src, len);
+		dest[len] = '\0';
+	}
+	return ret;
+}
+#endif
+
+static char *get_pathname(void)
+{
+	static char pathname_array[4][PATH_MAX];
+	static int idx;
+
+	return pathname_array[3 & ++idx];
+}
+
+static char *cleanup_path(char *path)
+{
+	/* Clean it up */
+	if (!memcmp(path, "./", 2)) {
+		path += 2;
+		while (*path == '/')
+			path++;
+	}
+	return path;
+}
+
+static char *perf_vsnpath(char *buf, size_t n, const char *fmt, va_list args)
+{
+	const char *perf_dir = get_perf_dir();
+	size_t len;
+
+	len = strlen(perf_dir);
+	if (n < len + 1)
+		goto bad;
+	memcpy(buf, perf_dir, len);
+	if (len && !is_dir_sep(perf_dir[len-1]))
+		buf[len++] = '/';
+	len += vsnprintf(buf + len, n - len, fmt, args);
+	if (len >= n)
+		goto bad;
+	return cleanup_path(buf);
+bad:
+	strlcpy(buf, bad_path, n);
+	return buf;
+}
+
+char *perf_pathdup(const char *fmt, ...)
+{
+	char path[PATH_MAX];
+	va_list args;
+	va_start(args, fmt);
+	(void)perf_vsnpath(path, sizeof(path), fmt, args);
+	va_end(args);
+	return xstrdup(path);
+}
+
+char *mkpath(const char *fmt, ...)
+{
+	va_list args;
+	unsigned len;
+	char *pathname = get_pathname();
+
+	va_start(args, fmt);
+	len = vsnprintf(pathname, PATH_MAX, fmt, args);
+	va_end(args);
+	if (len >= PATH_MAX)
+		return bad_path;
+	return cleanup_path(pathname);
+}
+
+char *perf_path(const char *fmt, ...)
+{
+	const char *perf_dir = get_perf_dir();
+	char *pathname = get_pathname();
+	va_list args;
+	unsigned len;
+
+	len = strlen(perf_dir);
+	if (len > PATH_MAX-100)
+		return bad_path;
+	memcpy(pathname, perf_dir, len);
+	if (len && perf_dir[len-1] != '/')
+		pathname[len++] = '/';
+	va_start(args, fmt);
+	len += vsnprintf(pathname + len, PATH_MAX - len, fmt, args);
+	va_end(args);
+	if (len >= PATH_MAX)
+		return bad_path;
+	return cleanup_path(pathname);
+}
+
+/* strip arbitrary amount of directory separators at end of path */
+static inline int chomp_trailing_dir_sep(const char *path, int len)
+{
+	while (len && is_dir_sep(path[len - 1]))
+		len--;
+	return len;
+}
+
+/*
+ * If path ends with suffix (complete path components), returns the
+ * part before suffix (sans trailing directory separators).
+ * Otherwise returns NULL.
+ */
+char *strip_path_suffix(const char *path, const char *suffix)
+{
+	int path_len = strlen(path), suffix_len = strlen(suffix);
+
+	while (suffix_len) {
+		if (!path_len)
+			return NULL;
+
+		if (is_dir_sep(path[path_len - 1])) {
+			if (!is_dir_sep(suffix[suffix_len - 1]))
+				return NULL;
+			path_len = chomp_trailing_dir_sep(path, path_len);
+			suffix_len = chomp_trailing_dir_sep(suffix, suffix_len);
+		}
+		else if (path[--path_len] != suffix[--suffix_len])
+			return NULL;
+	}
+
+	if (path_len && !is_dir_sep(path[path_len - 1]))
+		return NULL;
+	return strndup(path, chomp_trailing_dir_sep(path, path_len));
+}
diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
new file mode 100644
index 00000000..8c50da86
--- /dev/null
+++ b/tools/perf/util/probe-event.c
@@ -0,0 +1,2007 @@
+/*
+ * probe-event.c : perf-probe definition to probe_events format converter
+ *
+ * Written by Masami Hiramatsu <mhiramat@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+
+#define _GNU_SOURCE
+#include <sys/utsname.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <errno.h>
+#include <stdio.h>
+#include <unistd.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdarg.h>
+#include <limits.h>
+#include <elf.h>
+
+#undef _GNU_SOURCE
+#include "util.h"
+#include "event.h"
+#include "string.h"
+#include "strlist.h"
+#include "debug.h"
+#include "cache.h"
+#include "color.h"
+#include "symbol.h"
+#include "thread.h"
+#include "debugfs.h"
+#include "trace-event.h"	/* For __unused */
+#include "probe-event.h"
+#include "probe-finder.h"
+
+#define MAX_CMDLEN 256
+#define MAX_PROBE_ARGS 128
+#define PERFPROBE_GROUP "probe"
+
+bool probe_event_dry_run;	/* Dry run flag */
+
+#define semantic_error(msg ...) pr_err("Semantic error :" msg)
+
+/* If there is no space to write, returns -E2BIG. */
+static int e_snprintf(char *str, size_t size, const char *format, ...)
+	__attribute__((format(printf, 3, 4)));
+
+static int e_snprintf(char *str, size_t size, const char *format, ...)
+{
+	int ret;
+	va_list ap;
+	va_start(ap, format);
+	ret = vsnprintf(str, size, format, ap);
+	va_end(ap);
+	if (ret >= (int)size)
+		ret = -E2BIG;
+	return ret;
+}
+
+static char *synthesize_perf_probe_point(struct perf_probe_point *pp);
+static struct machine machine;
+
+/* Initialize symbol maps and path of vmlinux/modules */
+static int init_vmlinux(void)
+{
+	int ret;
+
+	symbol_conf.sort_by_name = true;
+	if (symbol_conf.vmlinux_name == NULL)
+		symbol_conf.try_vmlinux_path = true;
+	else
+		pr_debug("Use vmlinux: %s\n", symbol_conf.vmlinux_name);
+	ret = symbol__init();
+	if (ret < 0) {
+		pr_debug("Failed to init symbol map.\n");
+		goto out;
+	}
+
+	ret = machine__init(&machine, "", HOST_KERNEL_ID);
+	if (ret < 0)
+		goto out;
+
+	if (machine__create_kernel_maps(&machine) < 0) {
+		pr_debug("machine__create_kernel_maps() failed.\n");
+		goto out;
+	}
+out:
+	if (ret < 0)
+		pr_warning("Failed to init vmlinux path.\n");
+	return ret;
+}
+
+static struct symbol *__find_kernel_function_by_name(const char *name,
+						     struct map **mapp)
+{
+	return machine__find_kernel_function_by_name(&machine, name, mapp,
+						     NULL);
+}
+
+static struct map *kernel_get_module_map(const char *module)
+{
+	struct rb_node *nd;
+	struct map_groups *grp = &machine.kmaps;
+
+	if (!module)
+		module = "kernel";
+
+	for (nd = rb_first(&grp->maps[MAP__FUNCTION]); nd; nd = rb_next(nd)) {
+		struct map *pos = rb_entry(nd, struct map, rb_node);
+		if (strncmp(pos->dso->short_name + 1, module,
+			    pos->dso->short_name_len - 2) == 0) {
+			return pos;
+		}
+	}
+	return NULL;
+}
+
+static struct dso *kernel_get_module_dso(const char *module)
+{
+	struct dso *dso;
+	struct map *map;
+	const char *vmlinux_name;
+
+	if (module) {
+		list_for_each_entry(dso, &machine.kernel_dsos, node) {
+			if (strncmp(dso->short_name + 1, module,
+				    dso->short_name_len - 2) == 0)
+				goto found;
+		}
+		pr_debug("Failed to find module %s.\n", module);
+		return NULL;
+	}
+
+	map = machine.vmlinux_maps[MAP__FUNCTION];
+	dso = map->dso;
+
+	vmlinux_name = symbol_conf.vmlinux_name;
+	if (vmlinux_name) {
+		if (dso__load_vmlinux(dso, map, vmlinux_name, NULL) <= 0)
+			return NULL;
+	} else {
+		if (dso__load_vmlinux_path(dso, map, NULL) <= 0) {
+			pr_debug("Failed to load kernel map.\n");
+			return NULL;
+		}
+	}
+found:
+	return dso;
+}
+
+const char *kernel_get_module_path(const char *module)
+{
+	struct dso *dso = kernel_get_module_dso(module);
+	return (dso) ? dso->long_name : NULL;
+}
+
+#ifdef DWARF_SUPPORT
+static int open_vmlinux(const char *module)
+{
+	const char *path = kernel_get_module_path(module);
+	if (!path) {
+		pr_err("Failed to find path of %s module.\n",
+		       module ?: "kernel");
+		return -ENOENT;
+	}
+	pr_debug("Try to open %s\n", path);
+	return open(path, O_RDONLY);
+}
+
+/*
+ * Convert trace point to probe point with debuginfo
+ * Currently only handles kprobes.
+ */
+static int kprobe_convert_to_perf_probe(struct probe_trace_point *tp,
+					struct perf_probe_point *pp)
+{
+	struct symbol *sym;
+	struct map *map;
+	u64 addr;
+	int ret = -ENOENT;
+
+	sym = __find_kernel_function_by_name(tp->symbol, &map);
+	if (sym) {
+		addr = map->unmap_ip(map, sym->start + tp->offset);
+		pr_debug("try to find %s+%ld@%" PRIx64 "\n", tp->symbol,
+			 tp->offset, addr);
+		ret = find_perf_probe_point((unsigned long)addr, pp);
+	}
+	if (ret <= 0) {
+		pr_debug("Failed to find corresponding probes from "
+			 "debuginfo. Use kprobe event information.\n");
+		pp->function = strdup(tp->symbol);
+		if (pp->function == NULL)
+			return -ENOMEM;
+		pp->offset = tp->offset;
+	}
+	pp->retprobe = tp->retprobe;
+
+	return 0;
+}
+
+/* Try to find perf_probe_event with debuginfo */
+static int try_to_find_probe_trace_events(struct perf_probe_event *pev,
+					   struct probe_trace_event **tevs,
+					   int max_tevs, const char *module)
+{
+	bool need_dwarf = perf_probe_event_need_dwarf(pev);
+	int fd, ntevs;
+
+	fd = open_vmlinux(module);
+	if (fd < 0) {
+		if (need_dwarf) {
+			pr_warning("Failed to open debuginfo file.\n");
+			return fd;
+		}
+		pr_debug("Could not open vmlinux. Try to use symbols.\n");
+		return 0;
+	}
+
+	/* Searching trace events corresponding to probe event */
+	ntevs = find_probe_trace_events(fd, pev, tevs, max_tevs);
+
+	if (ntevs > 0) {	/* Succeeded to find trace events */
+		pr_debug("find %d probe_trace_events.\n", ntevs);
+		return ntevs;
+	}
+
+	if (ntevs == 0)	{	/* No error but failed to find probe point. */
+		pr_warning("Probe point '%s' not found.\n",
+			   synthesize_perf_probe_point(&pev->point));
+		return -ENOENT;
+	}
+	/* Error path : ntevs < 0 */
+	pr_debug("An error occurred in debuginfo analysis (%d).\n", ntevs);
+	if (ntevs == -EBADF) {
+		pr_warning("Warning: No dwarf info found in the vmlinux - "
+			"please rebuild kernel with CONFIG_DEBUG_INFO=y.\n");
+		if (!need_dwarf) {
+			pr_debug("Trying to use symbols.\n");
+			return 0;
+		}
+	}
+	return ntevs;
+}
+
+/*
+ * Find a src file from a DWARF tag path. Prepend optional source path prefix
+ * and chop off leading directories that do not exist. Result is passed back as
+ * a newly allocated path on success.
+ * Return 0 if file was found and readable, -errno otherwise.
+ */
+static int get_real_path(const char *raw_path, const char *comp_dir,
+			 char **new_path)
+{
+	const char *prefix = symbol_conf.source_prefix;
+
+	if (!prefix) {
+		if (raw_path[0] != '/' && comp_dir)
+			/* If not an absolute path, try to use comp_dir */
+			prefix = comp_dir;
+		else {
+			if (access(raw_path, R_OK) == 0) {
+				*new_path = strdup(raw_path);
+				return 0;
+			} else
+				return -errno;
+		}
+	}
+
+	*new_path = malloc((strlen(prefix) + strlen(raw_path) + 2));
+	if (!*new_path)
+		return -ENOMEM;
+
+	for (;;) {
+		sprintf(*new_path, "%s/%s", prefix, raw_path);
+
+		if (access(*new_path, R_OK) == 0)
+			return 0;
+
+		if (!symbol_conf.source_prefix)
+			/* In case of searching comp_dir, don't retry */
+			return -errno;
+
+		switch (errno) {
+		case ENAMETOOLONG:
+		case ENOENT:
+		case EROFS:
+		case EFAULT:
+			raw_path = strchr(++raw_path, '/');
+			if (!raw_path) {
+				free(*new_path);
+				*new_path = NULL;
+				return -ENOENT;
+			}
+			continue;
+
+		default:
+			free(*new_path);
+			*new_path = NULL;
+			return -errno;
+		}
+	}
+}
+
+#define LINEBUF_SIZE 256
+#define NR_ADDITIONAL_LINES 2
+
+static int __show_one_line(FILE *fp, int l, bool skip, bool show_num)
+{
+	char buf[LINEBUF_SIZE];
+	const char *color = show_num ? "" : PERF_COLOR_BLUE;
+	const char *prefix = NULL;
+
+	do {
+		if (fgets(buf, LINEBUF_SIZE, fp) == NULL)
+			goto error;
+		if (skip)
+			continue;
+		if (!prefix) {
+			prefix = show_num ? "%7d  " : "         ";
+			color_fprintf(stdout, color, prefix, l);
+		}
+		color_fprintf(stdout, color, "%s", buf);
+
+	} while (strchr(buf, '\n') == NULL);
+
+	return 1;
+error:
+	if (ferror(fp)) {
+		pr_warning("File read error: %s\n", strerror(errno));
+		return -1;
+	}
+	return 0;
+}
+
+static int _show_one_line(FILE *fp, int l, bool skip, bool show_num)
+{
+	int rv = __show_one_line(fp, l, skip, show_num);
+	if (rv == 0) {
+		pr_warning("Source file is shorter than expected.\n");
+		rv = -1;
+	}
+	return rv;
+}
+
+#define show_one_line_with_num(f,l)	_show_one_line(f,l,false,true)
+#define show_one_line(f,l)		_show_one_line(f,l,false,false)
+#define skip_one_line(f,l)		_show_one_line(f,l,true,false)
+#define show_one_line_or_eof(f,l)	__show_one_line(f,l,false,false)
+
+/*
+ * Show line-range always requires debuginfo to find source file and
+ * line number.
+ */
+int show_line_range(struct line_range *lr, const char *module)
+{
+	int l = 1;
+	struct line_node *ln;
+	FILE *fp;
+	int fd, ret;
+	char *tmp;
+
+	/* Search a line range */
+	ret = init_vmlinux();
+	if (ret < 0)
+		return ret;
+
+	fd = open_vmlinux(module);
+	if (fd < 0) {
+		pr_warning("Failed to open debuginfo file.\n");
+		return fd;
+	}
+
+	ret = find_line_range(fd, lr);
+	if (ret == 0) {
+		pr_warning("Specified source line is not found.\n");
+		return -ENOENT;
+	} else if (ret < 0) {
+		pr_warning("Debuginfo analysis failed. (%d)\n", ret);
+		return ret;
+	}
+
+	/* Convert source file path */
+	tmp = lr->path;
+	ret = get_real_path(tmp, lr->comp_dir, &lr->path);
+	free(tmp);	/* Free old path */
+	if (ret < 0) {
+		pr_warning("Failed to find source file. (%d)\n", ret);
+		return ret;
+	}
+
+	setup_pager();
+
+	if (lr->function)
+		fprintf(stdout, "<%s@%s:%d>\n", lr->function, lr->path,
+			lr->start - lr->offset);
+	else
+		fprintf(stdout, "<%s:%d>\n", lr->path, lr->start);
+
+	fp = fopen(lr->path, "r");
+	if (fp == NULL) {
+		pr_warning("Failed to open %s: %s\n", lr->path,
+			   strerror(errno));
+		return -errno;
+	}
+	/* Skip to starting line number */
+	while (l < lr->start) {
+		ret = skip_one_line(fp, l++);
+		if (ret < 0)
+			goto end;
+	}
+
+	list_for_each_entry(ln, &lr->line_list, list) {
+		for (; ln->line > l; l++) {
+			ret = show_one_line(fp, l - lr->offset);
+			if (ret < 0)
+				goto end;
+		}
+		ret = show_one_line_with_num(fp, l++ - lr->offset);
+		if (ret < 0)
+			goto end;
+	}
+
+	if (lr->end == INT_MAX)
+		lr->end = l + NR_ADDITIONAL_LINES;
+	while (l <= lr->end) {
+		ret = show_one_line_or_eof(fp, l++ - lr->offset);
+		if (ret <= 0)
+			break;
+	}
+end:
+	fclose(fp);
+	return ret;
+}
+
+static int show_available_vars_at(int fd, struct perf_probe_event *pev,
+				  int max_vls, struct strfilter *_filter,
+				  bool externs)
+{
+	char *buf;
+	int ret, i, nvars;
+	struct str_node *node;
+	struct variable_list *vls = NULL, *vl;
+	const char *var;
+
+	buf = synthesize_perf_probe_point(&pev->point);
+	if (!buf)
+		return -EINVAL;
+	pr_debug("Searching variables at %s\n", buf);
+
+	ret = find_available_vars_at(fd, pev, &vls, max_vls, externs);
+	if (ret <= 0) {
+		pr_err("Failed to find variables at %s (%d)\n", buf, ret);
+		goto end;
+	}
+	/* Some variables are found */
+	fprintf(stdout, "Available variables at %s\n", buf);
+	for (i = 0; i < ret; i++) {
+		vl = &vls[i];
+		/*
+		 * A probe point might be converted to
+		 * several trace points.
+		 */
+		fprintf(stdout, "\t@<%s+%lu>\n", vl->point.symbol,
+			vl->point.offset);
+		free(vl->point.symbol);
+		nvars = 0;
+		if (vl->vars) {
+			strlist__for_each(node, vl->vars) {
+				var = strchr(node->s, '\t') + 1;
+				if (strfilter__compare(_filter, var)) {
+					fprintf(stdout, "\t\t%s\n", node->s);
+					nvars++;
+				}
+			}
+			strlist__delete(vl->vars);
+		}
+		if (nvars == 0)
+			fprintf(stdout, "\t\t(No matched variables)\n");
+	}
+	free(vls);
+end:
+	free(buf);
+	return ret;
+}
+
+/* Show available variables on given probe point */
+int show_available_vars(struct perf_probe_event *pevs, int npevs,
+			int max_vls, const char *module,
+			struct strfilter *_filter, bool externs)
+{
+	int i, fd, ret = 0;
+
+	ret = init_vmlinux();
+	if (ret < 0)
+		return ret;
+
+	setup_pager();
+
+	for (i = 0; i < npevs && ret >= 0; i++) {
+		fd = open_vmlinux(module);
+		if (fd < 0) {
+			pr_warning("Failed to open debug information file.\n");
+			ret = fd;
+			break;
+		}
+		ret = show_available_vars_at(fd, &pevs[i], max_vls, _filter,
+					     externs);
+	}
+	return ret;
+}
+
+#else	/* !DWARF_SUPPORT */
+
+static int kprobe_convert_to_perf_probe(struct probe_trace_point *tp,
+					struct perf_probe_point *pp)
+{
+	struct symbol *sym;
+
+	sym = __find_kernel_function_by_name(tp->symbol, NULL);
+	if (!sym) {
+		pr_err("Failed to find symbol %s in kernel.\n", tp->symbol);
+		return -ENOENT;
+	}
+	pp->function = strdup(tp->symbol);
+	if (pp->function == NULL)
+		return -ENOMEM;
+	pp->offset = tp->offset;
+	pp->retprobe = tp->retprobe;
+
+	return 0;
+}
+
+static int try_to_find_probe_trace_events(struct perf_probe_event *pev,
+				struct probe_trace_event **tevs __unused,
+				int max_tevs __unused, const char *mod __unused)
+{
+	if (perf_probe_event_need_dwarf(pev)) {
+		pr_warning("Debuginfo-analysis is not supported.\n");
+		return -ENOSYS;
+	}
+	return 0;
+}
+
+int show_line_range(struct line_range *lr __unused, const char *module __unused)
+{
+	pr_warning("Debuginfo-analysis is not supported.\n");
+	return -ENOSYS;
+}
+
+int show_available_vars(struct perf_probe_event *pevs __unused,
+			int npevs __unused, int max_vls __unused,
+			const char *module __unused,
+			struct strfilter *filter __unused,
+			bool externs __unused)
+{
+	pr_warning("Debuginfo-analysis is not supported.\n");
+	return -ENOSYS;
+}
+#endif
+
+static int parse_line_num(char **ptr, int *val, const char *what)
+{
+	const char *start = *ptr;
+
+	errno = 0;
+	*val = strtol(*ptr, ptr, 0);
+	if (errno || *ptr == start) {
+		semantic_error("'%s' is not a valid number.\n", what);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+/*
+ * Stuff 'lr' according to the line range described by 'arg'.
+ * The line range syntax is described by:
+ *
+ *         SRC[:SLN[+NUM|-ELN]]
+ *         FNC[@SRC][:SLN[+NUM|-ELN]]
+ */
+int parse_line_range_desc(const char *arg, struct line_range *lr)
+{
+	char *range, *file, *name = strdup(arg);
+	int err;
+
+	if (!name)
+		return -ENOMEM;
+
+	lr->start = 0;
+	lr->end = INT_MAX;
+
+	range = strchr(name, ':');
+	if (range) {
+		*range++ = '\0';
+
+		err = parse_line_num(&range, &lr->start, "start line");
+		if (err)
+			goto err;
+
+		if (*range == '+' || *range == '-') {
+			const char c = *range++;
+
+			err = parse_line_num(&range, &lr->end, "end line");
+			if (err)
+				goto err;
+
+			if (c == '+') {
+				lr->end += lr->start;
+				/*
+				 * Adjust the number of lines here.
+				 * If the number of lines == 1, the
+				 * the end of line should be equal to
+				 * the start of line.
+				 */
+				lr->end--;
+			}
+		}
+
+		pr_debug("Line range is %d to %d\n", lr->start, lr->end);
+
+		err = -EINVAL;
+		if (lr->start > lr->end) {
+			semantic_error("Start line must be smaller"
+				       " than end line.\n");
+			goto err;
+		}
+		if (*range != '\0') {
+			semantic_error("Tailing with invalid str '%s'.\n", range);
+			goto err;
+		}
+	}
+
+	file = strchr(name, '@');
+	if (file) {
+		*file = '\0';
+		lr->file = strdup(++file);
+		if (lr->file == NULL) {
+			err = -ENOMEM;
+			goto err;
+		}
+		lr->function = name;
+	} else if (strchr(name, '.'))
+		lr->file = name;
+	else
+		lr->function = name;
+
+	return 0;
+err:
+	free(name);
+	return err;
+}
+
+/* Check the name is good for event/group */
+static bool check_event_name(const char *name)
+{
+	if (!isalpha(*name) && *name != '_')
+		return false;
+	while (*++name != '\0') {
+		if (!isalpha(*name) && !isdigit(*name) && *name != '_')
+			return false;
+	}
+	return true;
+}
+
+/* Parse probepoint definition. */
+static int parse_perf_probe_point(char *arg, struct perf_probe_event *pev)
+{
+	struct perf_probe_point *pp = &pev->point;
+	char *ptr, *tmp;
+	char c, nc = 0;
+	/*
+	 * <Syntax>
+	 * perf probe [EVENT=]SRC[:LN|;PTN]
+	 * perf probe [EVENT=]FUNC[@SRC][+OFFS|%return|:LN|;PAT]
+	 *
+	 * TODO:Group name support
+	 */
+
+	ptr = strpbrk(arg, ";=@+%");
+	if (ptr && *ptr == '=') {	/* Event name */
+		*ptr = '\0';
+		tmp = ptr + 1;
+		if (strchr(arg, ':')) {
+			semantic_error("Group name is not supported yet.\n");
+			return -ENOTSUP;
+		}
+		if (!check_event_name(arg)) {
+			semantic_error("%s is bad for event name -it must "
+				       "follow C symbol-naming rule.\n", arg);
+			return -EINVAL;
+		}
+		pev->event = strdup(arg);
+		if (pev->event == NULL)
+			return -ENOMEM;
+		pev->group = NULL;
+		arg = tmp;
+	}
+
+	ptr = strpbrk(arg, ";:+@%");
+	if (ptr) {
+		nc = *ptr;
+		*ptr++ = '\0';
+	}
+
+	tmp = strdup(arg);
+	if (tmp == NULL)
+		return -ENOMEM;
+
+	/* Check arg is function or file and copy it */
+	if (strchr(tmp, '.'))	/* File */
+		pp->file = tmp;
+	else			/* Function */
+		pp->function = tmp;
+
+	/* Parse other options */
+	while (ptr) {
+		arg = ptr;
+		c = nc;
+		if (c == ';') {	/* Lazy pattern must be the last part */
+			pp->lazy_line = strdup(arg);
+			if (pp->lazy_line == NULL)
+				return -ENOMEM;
+			break;
+		}
+		ptr = strpbrk(arg, ";:+@%");
+		if (ptr) {
+			nc = *ptr;
+			*ptr++ = '\0';
+		}
+		switch (c) {
+		case ':':	/* Line number */
+			pp->line = strtoul(arg, &tmp, 0);
+			if (*tmp != '\0') {
+				semantic_error("There is non-digit char"
+					       " in line number.\n");
+				return -EINVAL;
+			}
+			break;
+		case '+':	/* Byte offset from a symbol */
+			pp->offset = strtoul(arg, &tmp, 0);
+			if (*tmp != '\0') {
+				semantic_error("There is non-digit character"
+						" in offset.\n");
+				return -EINVAL;
+			}
+			break;
+		case '@':	/* File name */
+			if (pp->file) {
+				semantic_error("SRC@SRC is not allowed.\n");
+				return -EINVAL;
+			}
+			pp->file = strdup(arg);
+			if (pp->file == NULL)
+				return -ENOMEM;
+			break;
+		case '%':	/* Probe places */
+			if (strcmp(arg, "return") == 0) {
+				pp->retprobe = 1;
+			} else {	/* Others not supported yet */
+				semantic_error("%%%s is not supported.\n", arg);
+				return -ENOTSUP;
+			}
+			break;
+		default:	/* Buggy case */
+			pr_err("This program has a bug at %s:%d.\n",
+				__FILE__, __LINE__);
+			return -ENOTSUP;
+			break;
+		}
+	}
+
+	/* Exclusion check */
+	if (pp->lazy_line && pp->line) {
+		semantic_error("Lazy pattern can't be used with"
+			       " line number.\n");
+		return -EINVAL;
+	}
+
+	if (pp->lazy_line && pp->offset) {
+		semantic_error("Lazy pattern can't be used with offset.\n");
+		return -EINVAL;
+	}
+
+	if (pp->line && pp->offset) {
+		semantic_error("Offset can't be used with line number.\n");
+		return -EINVAL;
+	}
+
+	if (!pp->line && !pp->lazy_line && pp->file && !pp->function) {
+		semantic_error("File always requires line number or "
+			       "lazy pattern.\n");
+		return -EINVAL;
+	}
+
+	if (pp->offset && !pp->function) {
+		semantic_error("Offset requires an entry function.\n");
+		return -EINVAL;
+	}
+
+	if (pp->retprobe && !pp->function) {
+		semantic_error("Return probe requires an entry function.\n");
+		return -EINVAL;
+	}
+
+	if ((pp->offset || pp->line || pp->lazy_line) && pp->retprobe) {
+		semantic_error("Offset/Line/Lazy pattern can't be used with "
+			       "return probe.\n");
+		return -EINVAL;
+	}
+
+	pr_debug("symbol:%s file:%s line:%d offset:%lu return:%d lazy:%s\n",
+		 pp->function, pp->file, pp->line, pp->offset, pp->retprobe,
+		 pp->lazy_line);
+	return 0;
+}
+
+/* Parse perf-probe event argument */
+static int parse_perf_probe_arg(char *str, struct perf_probe_arg *arg)
+{
+	char *tmp, *goodname;
+	struct perf_probe_arg_field **fieldp;
+
+	pr_debug("parsing arg: %s into ", str);
+
+	tmp = strchr(str, '=');
+	if (tmp) {
+		arg->name = strndup(str, tmp - str);
+		if (arg->name == NULL)
+			return -ENOMEM;
+		pr_debug("name:%s ", arg->name);
+		str = tmp + 1;
+	}
+
+	tmp = strchr(str, ':');
+	if (tmp) {	/* Type setting */
+		*tmp = '\0';
+		arg->type = strdup(tmp + 1);
+		if (arg->type == NULL)
+			return -ENOMEM;
+		pr_debug("type:%s ", arg->type);
+	}
+
+	tmp = strpbrk(str, "-.[");
+	if (!is_c_varname(str) || !tmp) {
+		/* A variable, register, symbol or special value */
+		arg->var = strdup(str);
+		if (arg->var == NULL)
+			return -ENOMEM;
+		pr_debug("%s\n", arg->var);
+		return 0;
+	}
+
+	/* Structure fields or array element */
+	arg->var = strndup(str, tmp - str);
+	if (arg->var == NULL)
+		return -ENOMEM;
+	goodname = arg->var;
+	pr_debug("%s, ", arg->var);
+	fieldp = &arg->field;
+
+	do {
+		*fieldp = zalloc(sizeof(struct perf_probe_arg_field));
+		if (*fieldp == NULL)
+			return -ENOMEM;
+		if (*tmp == '[') {	/* Array */
+			str = tmp;
+			(*fieldp)->index = strtol(str + 1, &tmp, 0);
+			(*fieldp)->ref = true;
+			if (*tmp != ']' || tmp == str + 1) {
+				semantic_error("Array index must be a"
+						" number.\n");
+				return -EINVAL;
+			}
+			tmp++;
+			if (*tmp == '\0')
+				tmp = NULL;
+		} else {		/* Structure */
+			if (*tmp == '.') {
+				str = tmp + 1;
+				(*fieldp)->ref = false;
+			} else if (tmp[1] == '>') {
+				str = tmp + 2;
+				(*fieldp)->ref = true;
+			} else {
+				semantic_error("Argument parse error: %s\n",
+					       str);
+				return -EINVAL;
+			}
+			tmp = strpbrk(str, "-.[");
+		}
+		if (tmp) {
+			(*fieldp)->name = strndup(str, tmp - str);
+			if ((*fieldp)->name == NULL)
+				return -ENOMEM;
+			if (*str != '[')
+				goodname = (*fieldp)->name;
+			pr_debug("%s(%d), ", (*fieldp)->name, (*fieldp)->ref);
+			fieldp = &(*fieldp)->next;
+		}
+	} while (tmp);
+	(*fieldp)->name = strdup(str);
+	if ((*fieldp)->name == NULL)
+		return -ENOMEM;
+	if (*str != '[')
+		goodname = (*fieldp)->name;
+	pr_debug("%s(%d)\n", (*fieldp)->name, (*fieldp)->ref);
+
+	/* If no name is specified, set the last field name (not array index)*/
+	if (!arg->name) {
+		arg->name = strdup(goodname);
+		if (arg->name == NULL)
+			return -ENOMEM;
+	}
+	return 0;
+}
+
+/* Parse perf-probe event command */
+int parse_perf_probe_command(const char *cmd, struct perf_probe_event *pev)
+{
+	char **argv;
+	int argc, i, ret = 0;
+
+	argv = argv_split(cmd, &argc);
+	if (!argv) {
+		pr_debug("Failed to split arguments.\n");
+		return -ENOMEM;
+	}
+	if (argc - 1 > MAX_PROBE_ARGS) {
+		semantic_error("Too many probe arguments (%d).\n", argc - 1);
+		ret = -ERANGE;
+		goto out;
+	}
+	/* Parse probe point */
+	ret = parse_perf_probe_point(argv[0], pev);
+	if (ret < 0)
+		goto out;
+
+	/* Copy arguments and ensure return probe has no C argument */
+	pev->nargs = argc - 1;
+	pev->args = zalloc(sizeof(struct perf_probe_arg) * pev->nargs);
+	if (pev->args == NULL) {
+		ret = -ENOMEM;
+		goto out;
+	}
+	for (i = 0; i < pev->nargs && ret >= 0; i++) {
+		ret = parse_perf_probe_arg(argv[i + 1], &pev->args[i]);
+		if (ret >= 0 &&
+		    is_c_varname(pev->args[i].var) && pev->point.retprobe) {
+			semantic_error("You can't specify local variable for"
+				       " kretprobe.\n");
+			ret = -EINVAL;
+		}
+	}
+out:
+	argv_free(argv);
+
+	return ret;
+}
+
+/* Return true if this perf_probe_event requires debuginfo */
+bool perf_probe_event_need_dwarf(struct perf_probe_event *pev)
+{
+	int i;
+
+	if (pev->point.file || pev->point.line || pev->point.lazy_line)
+		return true;
+
+	for (i = 0; i < pev->nargs; i++)
+		if (is_c_varname(pev->args[i].var))
+			return true;
+
+	return false;
+}
+
+/* Parse probe_events event into struct probe_point */
+static int parse_probe_trace_command(const char *cmd,
+					struct probe_trace_event *tev)
+{
+	struct probe_trace_point *tp = &tev->point;
+	char pr;
+	char *p;
+	int ret, i, argc;
+	char **argv;
+
+	pr_debug("Parsing probe_events: %s\n", cmd);
+	argv = argv_split(cmd, &argc);
+	if (!argv) {
+		pr_debug("Failed to split arguments.\n");
+		return -ENOMEM;
+	}
+	if (argc < 2) {
+		semantic_error("Too few probe arguments.\n");
+		ret = -ERANGE;
+		goto out;
+	}
+
+	/* Scan event and group name. */
+	ret = sscanf(argv[0], "%c:%a[^/ \t]/%a[^ \t]",
+		     &pr, (float *)(void *)&tev->group,
+		     (float *)(void *)&tev->event);
+	if (ret != 3) {
+		semantic_error("Failed to parse event name: %s\n", argv[0]);
+		ret = -EINVAL;
+		goto out;
+	}
+	pr_debug("Group:%s Event:%s probe:%c\n", tev->group, tev->event, pr);
+
+	tp->retprobe = (pr == 'r');
+
+	/* Scan function name and offset */
+	ret = sscanf(argv[1], "%a[^+]+%lu", (float *)(void *)&tp->symbol,
+		     &tp->offset);
+	if (ret == 1)
+		tp->offset = 0;
+
+	tev->nargs = argc - 2;
+	tev->args = zalloc(sizeof(struct probe_trace_arg) * tev->nargs);
+	if (tev->args == NULL) {
+		ret = -ENOMEM;
+		goto out;
+	}
+	for (i = 0; i < tev->nargs; i++) {
+		p = strchr(argv[i + 2], '=');
+		if (p)	/* We don't need which register is assigned. */
+			*p++ = '\0';
+		else
+			p = argv[i + 2];
+		tev->args[i].name = strdup(argv[i + 2]);
+		/* TODO: parse regs and offset */
+		tev->args[i].value = strdup(p);
+		if (tev->args[i].name == NULL || tev->args[i].value == NULL) {
+			ret = -ENOMEM;
+			goto out;
+		}
+	}
+	ret = 0;
+out:
+	argv_free(argv);
+	return ret;
+}
+
+/* Compose only probe arg */
+int synthesize_perf_probe_arg(struct perf_probe_arg *pa, char *buf, size_t len)
+{
+	struct perf_probe_arg_field *field = pa->field;
+	int ret;
+	char *tmp = buf;
+
+	if (pa->name && pa->var)
+		ret = e_snprintf(tmp, len, "%s=%s", pa->name, pa->var);
+	else
+		ret = e_snprintf(tmp, len, "%s", pa->name ? pa->name : pa->var);
+	if (ret <= 0)
+		goto error;
+	tmp += ret;
+	len -= ret;
+
+	while (field) {
+		if (field->name[0] == '[')
+			ret = e_snprintf(tmp, len, "%s", field->name);
+		else
+			ret = e_snprintf(tmp, len, "%s%s",
+					 field->ref ? "->" : ".", field->name);
+		if (ret <= 0)
+			goto error;
+		tmp += ret;
+		len -= ret;
+		field = field->next;
+	}
+
+	if (pa->type) {
+		ret = e_snprintf(tmp, len, ":%s", pa->type);
+		if (ret <= 0)
+			goto error;
+		tmp += ret;
+		len -= ret;
+	}
+
+	return tmp - buf;
+error:
+	pr_debug("Failed to synthesize perf probe argument: %s\n",
+		 strerror(-ret));
+	return ret;
+}
+
+/* Compose only probe point (not argument) */
+static char *synthesize_perf_probe_point(struct perf_probe_point *pp)
+{
+	char *buf, *tmp;
+	char offs[32] = "", line[32] = "", file[32] = "";
+	int ret, len;
+
+	buf = zalloc(MAX_CMDLEN);
+	if (buf == NULL) {
+		ret = -ENOMEM;
+		goto error;
+	}
+	if (pp->offset) {
+		ret = e_snprintf(offs, 32, "+%lu", pp->offset);
+		if (ret <= 0)
+			goto error;
+	}
+	if (pp->line) {
+		ret = e_snprintf(line, 32, ":%d", pp->line);
+		if (ret <= 0)
+			goto error;
+	}
+	if (pp->file) {
+		tmp = pp->file;
+		len = strlen(tmp);
+		if (len > 30) {
+			tmp = strchr(pp->file + len - 30, '/');
+			tmp = tmp ? tmp + 1 : pp->file + len - 30;
+		}
+		ret = e_snprintf(file, 32, "@%s", tmp);
+		if (ret <= 0)
+			goto error;
+	}
+
+	if (pp->function)
+		ret = e_snprintf(buf, MAX_CMDLEN, "%s%s%s%s%s", pp->function,
+				 offs, pp->retprobe ? "%return" : "", line,
+				 file);
+	else
+		ret = e_snprintf(buf, MAX_CMDLEN, "%s%s", file, line);
+	if (ret <= 0)
+		goto error;
+
+	return buf;
+error:
+	pr_debug("Failed to synthesize perf probe point: %s\n",
+		 strerror(-ret));
+	if (buf)
+		free(buf);
+	return NULL;
+}
+
+#if 0
+char *synthesize_perf_probe_command(struct perf_probe_event *pev)
+{
+	char *buf;
+	int i, len, ret;
+
+	buf = synthesize_perf_probe_point(&pev->point);
+	if (!buf)
+		return NULL;
+
+	len = strlen(buf);
+	for (i = 0; i < pev->nargs; i++) {
+		ret = e_snprintf(&buf[len], MAX_CMDLEN - len, " %s",
+				 pev->args[i].name);
+		if (ret <= 0) {
+			free(buf);
+			return NULL;
+		}
+		len += ret;
+	}
+
+	return buf;
+}
+#endif
+
+static int __synthesize_probe_trace_arg_ref(struct probe_trace_arg_ref *ref,
+					     char **buf, size_t *buflen,
+					     int depth)
+{
+	int ret;
+	if (ref->next) {
+		depth = __synthesize_probe_trace_arg_ref(ref->next, buf,
+							 buflen, depth + 1);
+		if (depth < 0)
+			goto out;
+	}
+
+	ret = e_snprintf(*buf, *buflen, "%+ld(", ref->offset);
+	if (ret < 0)
+		depth = ret;
+	else {
+		*buf += ret;
+		*buflen -= ret;
+	}
+out:
+	return depth;
+
+}
+
+static int synthesize_probe_trace_arg(struct probe_trace_arg *arg,
+				       char *buf, size_t buflen)
+{
+	struct probe_trace_arg_ref *ref = arg->ref;
+	int ret, depth = 0;
+	char *tmp = buf;
+
+	/* Argument name or separator */
+	if (arg->name)
+		ret = e_snprintf(buf, buflen, " %s=", arg->name);
+	else
+		ret = e_snprintf(buf, buflen, " ");
+	if (ret < 0)
+		return ret;
+	buf += ret;
+	buflen -= ret;
+
+	/* Special case: @XXX */
+	if (arg->value[0] == '@' && arg->ref)
+			ref = ref->next;
+
+	/* Dereferencing arguments */
+	if (ref) {
+		depth = __synthesize_probe_trace_arg_ref(ref, &buf,
+							  &buflen, 1);
+		if (depth < 0)
+			return depth;
+	}
+
+	/* Print argument value */
+	if (arg->value[0] == '@' && arg->ref)
+		ret = e_snprintf(buf, buflen, "%s%+ld", arg->value,
+				 arg->ref->offset);
+	else
+		ret = e_snprintf(buf, buflen, "%s", arg->value);
+	if (ret < 0)
+		return ret;
+	buf += ret;
+	buflen -= ret;
+
+	/* Closing */
+	while (depth--) {
+		ret = e_snprintf(buf, buflen, ")");
+		if (ret < 0)
+			return ret;
+		buf += ret;
+		buflen -= ret;
+	}
+	/* Print argument type */
+	if (arg->type) {
+		ret = e_snprintf(buf, buflen, ":%s", arg->type);
+		if (ret <= 0)
+			return ret;
+		buf += ret;
+	}
+
+	return buf - tmp;
+}
+
+char *synthesize_probe_trace_command(struct probe_trace_event *tev)
+{
+	struct probe_trace_point *tp = &tev->point;
+	char *buf;
+	int i, len, ret;
+
+	buf = zalloc(MAX_CMDLEN);
+	if (buf == NULL)
+		return NULL;
+
+	len = e_snprintf(buf, MAX_CMDLEN, "%c:%s/%s %s+%lu",
+			 tp->retprobe ? 'r' : 'p',
+			 tev->group, tev->event,
+			 tp->symbol, tp->offset);
+	if (len <= 0)
+		goto error;
+
+	for (i = 0; i < tev->nargs; i++) {
+		ret = synthesize_probe_trace_arg(&tev->args[i], buf + len,
+						  MAX_CMDLEN - len);
+		if (ret <= 0)
+			goto error;
+		len += ret;
+	}
+
+	return buf;
+error:
+	free(buf);
+	return NULL;
+}
+
+static int convert_to_perf_probe_event(struct probe_trace_event *tev,
+				       struct perf_probe_event *pev)
+{
+	char buf[64] = "";
+	int i, ret;
+
+	/* Convert event/group name */
+	pev->event = strdup(tev->event);
+	pev->group = strdup(tev->group);
+	if (pev->event == NULL || pev->group == NULL)
+		return -ENOMEM;
+
+	/* Convert trace_point to probe_point */
+	ret = kprobe_convert_to_perf_probe(&tev->point, &pev->point);
+	if (ret < 0)
+		return ret;
+
+	/* Convert trace_arg to probe_arg */
+	pev->nargs = tev->nargs;
+	pev->args = zalloc(sizeof(struct perf_probe_arg) * pev->nargs);
+	if (pev->args == NULL)
+		return -ENOMEM;
+	for (i = 0; i < tev->nargs && ret >= 0; i++) {
+		if (tev->args[i].name)
+			pev->args[i].name = strdup(tev->args[i].name);
+		else {
+			ret = synthesize_probe_trace_arg(&tev->args[i],
+							  buf, 64);
+			pev->args[i].name = strdup(buf);
+		}
+		if (pev->args[i].name == NULL && ret >= 0)
+			ret = -ENOMEM;
+	}
+
+	if (ret < 0)
+		clear_perf_probe_event(pev);
+
+	return ret;
+}
+
+void clear_perf_probe_event(struct perf_probe_event *pev)
+{
+	struct perf_probe_point *pp = &pev->point;
+	struct perf_probe_arg_field *field, *next;
+	int i;
+
+	if (pev->event)
+		free(pev->event);
+	if (pev->group)
+		free(pev->group);
+	if (pp->file)
+		free(pp->file);
+	if (pp->function)
+		free(pp->function);
+	if (pp->lazy_line)
+		free(pp->lazy_line);
+	for (i = 0; i < pev->nargs; i++) {
+		if (pev->args[i].name)
+			free(pev->args[i].name);
+		if (pev->args[i].var)
+			free(pev->args[i].var);
+		if (pev->args[i].type)
+			free(pev->args[i].type);
+		field = pev->args[i].field;
+		while (field) {
+			next = field->next;
+			if (field->name)
+				free(field->name);
+			free(field);
+			field = next;
+		}
+	}
+	if (pev->args)
+		free(pev->args);
+	memset(pev, 0, sizeof(*pev));
+}
+
+static void clear_probe_trace_event(struct probe_trace_event *tev)
+{
+	struct probe_trace_arg_ref *ref, *next;
+	int i;
+
+	if (tev->event)
+		free(tev->event);
+	if (tev->group)
+		free(tev->group);
+	if (tev->point.symbol)
+		free(tev->point.symbol);
+	for (i = 0; i < tev->nargs; i++) {
+		if (tev->args[i].name)
+			free(tev->args[i].name);
+		if (tev->args[i].value)
+			free(tev->args[i].value);
+		if (tev->args[i].type)
+			free(tev->args[i].type);
+		ref = tev->args[i].ref;
+		while (ref) {
+			next = ref->next;
+			free(ref);
+			ref = next;
+		}
+	}
+	if (tev->args)
+		free(tev->args);
+	memset(tev, 0, sizeof(*tev));
+}
+
+static int open_kprobe_events(bool readwrite)
+{
+	char buf[PATH_MAX];
+	const char *__debugfs;
+	int ret;
+
+	__debugfs = debugfs_find_mountpoint();
+	if (__debugfs == NULL) {
+		pr_warning("Debugfs is not mounted.\n");
+		return -ENOENT;
+	}
+
+	ret = e_snprintf(buf, PATH_MAX, "%stracing/kprobe_events", __debugfs);
+	if (ret >= 0) {
+		pr_debug("Opening %s write=%d\n", buf, readwrite);
+		if (readwrite && !probe_event_dry_run)
+			ret = open(buf, O_RDWR, O_APPEND);
+		else
+			ret = open(buf, O_RDONLY, 0);
+	}
+
+	if (ret < 0) {
+		if (errno == ENOENT)
+			pr_warning("kprobe_events file does not exist - please"
+				 " rebuild kernel with CONFIG_KPROBE_EVENT.\n");
+		else
+			pr_warning("Failed to open kprobe_events file: %s\n",
+				   strerror(errno));
+	}
+	return ret;
+}
+
+/* Get raw string list of current kprobe_events */
+static struct strlist *get_probe_trace_command_rawlist(int fd)
+{
+	int ret, idx;
+	FILE *fp;
+	char buf[MAX_CMDLEN];
+	char *p;
+	struct strlist *sl;
+
+	sl = strlist__new(true, NULL);
+
+	fp = fdopen(dup(fd), "r");
+	while (!feof(fp)) {
+		p = fgets(buf, MAX_CMDLEN, fp);
+		if (!p)
+			break;
+
+		idx = strlen(p) - 1;
+		if (p[idx] == '\n')
+			p[idx] = '\0';
+		ret = strlist__add(sl, buf);
+		if (ret < 0) {
+			pr_debug("strlist__add failed: %s\n", strerror(-ret));
+			strlist__delete(sl);
+			return NULL;
+		}
+	}
+	fclose(fp);
+
+	return sl;
+}
+
+/* Show an event */
+static int show_perf_probe_event(struct perf_probe_event *pev)
+{
+	int i, ret;
+	char buf[128];
+	char *place;
+
+	/* Synthesize only event probe point */
+	place = synthesize_perf_probe_point(&pev->point);
+	if (!place)
+		return -EINVAL;
+
+	ret = e_snprintf(buf, 128, "%s:%s", pev->group, pev->event);
+	if (ret < 0)
+		return ret;
+
+	printf("  %-20s (on %s", buf, place);
+
+	if (pev->nargs > 0) {
+		printf(" with");
+		for (i = 0; i < pev->nargs; i++) {
+			ret = synthesize_perf_probe_arg(&pev->args[i],
+							buf, 128);
+			if (ret < 0)
+				break;
+			printf(" %s", buf);
+		}
+	}
+	printf(")\n");
+	free(place);
+	return ret;
+}
+
+/* List up current perf-probe events */
+int show_perf_probe_events(void)
+{
+	int fd, ret;
+	struct probe_trace_event tev;
+	struct perf_probe_event pev;
+	struct strlist *rawlist;
+	struct str_node *ent;
+
+	setup_pager();
+	ret = init_vmlinux();
+	if (ret < 0)
+		return ret;
+
+	memset(&tev, 0, sizeof(tev));
+	memset(&pev, 0, sizeof(pev));
+
+	fd = open_kprobe_events(false);
+	if (fd < 0)
+		return fd;
+
+	rawlist = get_probe_trace_command_rawlist(fd);
+	close(fd);
+	if (!rawlist)
+		return -ENOENT;
+
+	strlist__for_each(ent, rawlist) {
+		ret = parse_probe_trace_command(ent->s, &tev);
+		if (ret >= 0) {
+			ret = convert_to_perf_probe_event(&tev, &pev);
+			if (ret >= 0)
+				ret = show_perf_probe_event(&pev);
+		}
+		clear_perf_probe_event(&pev);
+		clear_probe_trace_event(&tev);
+		if (ret < 0)
+			break;
+	}
+	strlist__delete(rawlist);
+
+	return ret;
+}
+
+/* Get current perf-probe event names */
+static struct strlist *get_probe_trace_event_names(int fd, bool include_group)
+{
+	char buf[128];
+	struct strlist *sl, *rawlist;
+	struct str_node *ent;
+	struct probe_trace_event tev;
+	int ret = 0;
+
+	memset(&tev, 0, sizeof(tev));
+	rawlist = get_probe_trace_command_rawlist(fd);
+	sl = strlist__new(true, NULL);
+	strlist__for_each(ent, rawlist) {
+		ret = parse_probe_trace_command(ent->s, &tev);
+		if (ret < 0)
+			break;
+		if (include_group) {
+			ret = e_snprintf(buf, 128, "%s:%s", tev.group,
+					tev.event);
+			if (ret >= 0)
+				ret = strlist__add(sl, buf);
+		} else
+			ret = strlist__add(sl, tev.event);
+		clear_probe_trace_event(&tev);
+		if (ret < 0)
+			break;
+	}
+	strlist__delete(rawlist);
+
+	if (ret < 0) {
+		strlist__delete(sl);
+		return NULL;
+	}
+	return sl;
+}
+
+static int write_probe_trace_event(int fd, struct probe_trace_event *tev)
+{
+	int ret = 0;
+	char *buf = synthesize_probe_trace_command(tev);
+
+	if (!buf) {
+		pr_debug("Failed to synthesize probe trace event.\n");
+		return -EINVAL;
+	}
+
+	pr_debug("Writing event: %s\n", buf);
+	if (!probe_event_dry_run) {
+		ret = write(fd, buf, strlen(buf));
+		if (ret <= 0)
+			pr_warning("Failed to write event: %s\n",
+				   strerror(errno));
+	}
+	free(buf);
+	return ret;
+}
+
+static int get_new_event_name(char *buf, size_t len, const char *base,
+			      struct strlist *namelist, bool allow_suffix)
+{
+	int i, ret;
+
+	/* Try no suffix */
+	ret = e_snprintf(buf, len, "%s", base);
+	if (ret < 0) {
+		pr_debug("snprintf() failed: %s\n", strerror(-ret));
+		return ret;
+	}
+	if (!strlist__has_entry(namelist, buf))
+		return 0;
+
+	if (!allow_suffix) {
+		pr_warning("Error: event \"%s\" already exists. "
+			   "(Use -f to force duplicates.)\n", base);
+		return -EEXIST;
+	}
+
+	/* Try to add suffix */
+	for (i = 1; i < MAX_EVENT_INDEX; i++) {
+		ret = e_snprintf(buf, len, "%s_%d", base, i);
+		if (ret < 0) {
+			pr_debug("snprintf() failed: %s\n", strerror(-ret));
+			return ret;
+		}
+		if (!strlist__has_entry(namelist, buf))
+			break;
+	}
+	if (i == MAX_EVENT_INDEX) {
+		pr_warning("Too many events are on the same function.\n");
+		ret = -ERANGE;
+	}
+
+	return ret;
+}
+
+static int __add_probe_trace_events(struct perf_probe_event *pev,
+				     struct probe_trace_event *tevs,
+				     int ntevs, bool allow_suffix)
+{
+	int i, fd, ret;
+	struct probe_trace_event *tev = NULL;
+	char buf[64];
+	const char *event, *group;
+	struct strlist *namelist;
+
+	fd = open_kprobe_events(true);
+	if (fd < 0)
+		return fd;
+	/* Get current event names */
+	namelist = get_probe_trace_event_names(fd, false);
+	if (!namelist) {
+		pr_debug("Failed to get current event list.\n");
+		return -EIO;
+	}
+
+	ret = 0;
+	printf("Add new event%s\n", (ntevs > 1) ? "s:" : ":");
+	for (i = 0; i < ntevs; i++) {
+		tev = &tevs[i];
+		if (pev->event)
+			event = pev->event;
+		else
+			if (pev->point.function)
+				event = pev->point.function;
+			else
+				event = tev->point.symbol;
+		if (pev->group)
+			group = pev->group;
+		else
+			group = PERFPROBE_GROUP;
+
+		/* Get an unused new event name */
+		ret = get_new_event_name(buf, 64, event,
+					 namelist, allow_suffix);
+		if (ret < 0)
+			break;
+		event = buf;
+
+		tev->event = strdup(event);
+		tev->group = strdup(group);
+		if (tev->event == NULL || tev->group == NULL) {
+			ret = -ENOMEM;
+			break;
+		}
+		ret = write_probe_trace_event(fd, tev);
+		if (ret < 0)
+			break;
+		/* Add added event name to namelist */
+		strlist__add(namelist, event);
+
+		/* Trick here - save current event/group */
+		event = pev->event;
+		group = pev->group;
+		pev->event = tev->event;
+		pev->group = tev->group;
+		show_perf_probe_event(pev);
+		/* Trick here - restore current event/group */
+		pev->event = (char *)event;
+		pev->group = (char *)group;
+
+		/*
+		 * Probes after the first probe which comes from same
+		 * user input are always allowed to add suffix, because
+		 * there might be several addresses corresponding to
+		 * one code line.
+		 */
+		allow_suffix = true;
+	}
+
+	if (ret >= 0) {
+		/* Show how to use the event. */
+		printf("\nYou can now use it on all perf tools, such as:\n\n");
+		printf("\tperf record -e %s:%s -aR sleep 1\n\n", tev->group,
+			 tev->event);
+	}
+
+	strlist__delete(namelist);
+	close(fd);
+	return ret;
+}
+
+static int convert_to_probe_trace_events(struct perf_probe_event *pev,
+					  struct probe_trace_event **tevs,
+					  int max_tevs, const char *module)
+{
+	struct symbol *sym;
+	int ret = 0, i;
+	struct probe_trace_event *tev;
+
+	/* Convert perf_probe_event with debuginfo */
+	ret = try_to_find_probe_trace_events(pev, tevs, max_tevs, module);
+	if (ret != 0)
+		return ret;
+
+	/* Allocate trace event buffer */
+	tev = *tevs = zalloc(sizeof(struct probe_trace_event));
+	if (tev == NULL)
+		return -ENOMEM;
+
+	/* Copy parameters */
+	tev->point.symbol = strdup(pev->point.function);
+	if (tev->point.symbol == NULL) {
+		ret = -ENOMEM;
+		goto error;
+	}
+	tev->point.offset = pev->point.offset;
+	tev->point.retprobe = pev->point.retprobe;
+	tev->nargs = pev->nargs;
+	if (tev->nargs) {
+		tev->args = zalloc(sizeof(struct probe_trace_arg)
+				   * tev->nargs);
+		if (tev->args == NULL) {
+			ret = -ENOMEM;
+			goto error;
+		}
+		for (i = 0; i < tev->nargs; i++) {
+			if (pev->args[i].name) {
+				tev->args[i].name = strdup(pev->args[i].name);
+				if (tev->args[i].name == NULL) {
+					ret = -ENOMEM;
+					goto error;
+				}
+			}
+			tev->args[i].value = strdup(pev->args[i].var);
+			if (tev->args[i].value == NULL) {
+				ret = -ENOMEM;
+				goto error;
+			}
+			if (pev->args[i].type) {
+				tev->args[i].type = strdup(pev->args[i].type);
+				if (tev->args[i].type == NULL) {
+					ret = -ENOMEM;
+					goto error;
+				}
+			}
+		}
+	}
+
+	/* Currently just checking function name from symbol map */
+	sym = __find_kernel_function_by_name(tev->point.symbol, NULL);
+	if (!sym) {
+		pr_warning("Kernel symbol \'%s\' not found.\n",
+			   tev->point.symbol);
+		ret = -ENOENT;
+		goto error;
+	}
+
+	return 1;
+error:
+	clear_probe_trace_event(tev);
+	free(tev);
+	*tevs = NULL;
+	return ret;
+}
+
+struct __event_package {
+	struct perf_probe_event		*pev;
+	struct probe_trace_event	*tevs;
+	int				ntevs;
+};
+
+int add_perf_probe_events(struct perf_probe_event *pevs, int npevs,
+			  int max_tevs, const char *module, bool force_add)
+{
+	int i, j, ret;
+	struct __event_package *pkgs;
+
+	pkgs = zalloc(sizeof(struct __event_package) * npevs);
+	if (pkgs == NULL)
+		return -ENOMEM;
+
+	/* Init vmlinux path */
+	ret = init_vmlinux();
+	if (ret < 0) {
+		free(pkgs);
+		return ret;
+	}
+
+	/* Loop 1: convert all events */
+	for (i = 0; i < npevs; i++) {
+		pkgs[i].pev = &pevs[i];
+		/* Convert with or without debuginfo */
+		ret  = convert_to_probe_trace_events(pkgs[i].pev,
+						     &pkgs[i].tevs,
+						     max_tevs,
+						     module);
+		if (ret < 0)
+			goto end;
+		pkgs[i].ntevs = ret;
+	}
+
+	/* Loop 2: add all events */
+	for (i = 0; i < npevs; i++) {
+		ret = __add_probe_trace_events(pkgs[i].pev, pkgs[i].tevs,
+						pkgs[i].ntevs, force_add);
+		if (ret < 0)
+			break;
+	}
+end:
+	/* Loop 3: cleanup and free trace events  */
+	for (i = 0; i < npevs; i++) {
+		for (j = 0; j < pkgs[i].ntevs; j++)
+			clear_probe_trace_event(&pkgs[i].tevs[j]);
+		free(pkgs[i].tevs);
+	}
+	free(pkgs);
+
+	return ret;
+}
+
+static int __del_trace_probe_event(int fd, struct str_node *ent)
+{
+	char *p;
+	char buf[128];
+	int ret;
+
+	/* Convert from perf-probe event to trace-probe event */
+	ret = e_snprintf(buf, 128, "-:%s", ent->s);
+	if (ret < 0)
+		goto error;
+
+	p = strchr(buf + 2, ':');
+	if (!p) {
+		pr_debug("Internal error: %s should have ':' but not.\n",
+			 ent->s);
+		ret = -ENOTSUP;
+		goto error;
+	}
+	*p = '/';
+
+	pr_debug("Writing event: %s\n", buf);
+	ret = write(fd, buf, strlen(buf));
+	if (ret < 0) {
+		ret = -errno;
+		goto error;
+	}
+
+	printf("Remove event: %s\n", ent->s);
+	return 0;
+error:
+	pr_warning("Failed to delete event: %s\n", strerror(-ret));
+	return ret;
+}
+
+static int del_trace_probe_event(int fd, const char *group,
+				  const char *event, struct strlist *namelist)
+{
+	char buf[128];
+	struct str_node *ent, *n;
+	int found = 0, ret = 0;
+
+	ret = e_snprintf(buf, 128, "%s:%s", group, event);
+	if (ret < 0) {
+		pr_err("Failed to copy event.\n");
+		return ret;
+	}
+
+	if (strpbrk(buf, "*?")) { /* Glob-exp */
+		strlist__for_each_safe(ent, n, namelist)
+			if (strglobmatch(ent->s, buf)) {
+				found++;
+				ret = __del_trace_probe_event(fd, ent);
+				if (ret < 0)
+					break;
+				strlist__remove(namelist, ent);
+			}
+	} else {
+		ent = strlist__find(namelist, buf);
+		if (ent) {
+			found++;
+			ret = __del_trace_probe_event(fd, ent);
+			if (ret >= 0)
+				strlist__remove(namelist, ent);
+		}
+	}
+	if (found == 0 && ret >= 0)
+		pr_info("Info: Event \"%s\" does not exist.\n", buf);
+
+	return ret;
+}
+
+int del_perf_probe_events(struct strlist *dellist)
+{
+	int fd, ret = 0;
+	const char *group, *event;
+	char *p, *str;
+	struct str_node *ent;
+	struct strlist *namelist;
+
+	fd = open_kprobe_events(true);
+	if (fd < 0)
+		return fd;
+
+	/* Get current event names */
+	namelist = get_probe_trace_event_names(fd, true);
+	if (namelist == NULL)
+		return -EINVAL;
+
+	strlist__for_each(ent, dellist) {
+		str = strdup(ent->s);
+		if (str == NULL) {
+			ret = -ENOMEM;
+			break;
+		}
+		pr_debug("Parsing: %s\n", str);
+		p = strchr(str, ':');
+		if (p) {
+			group = str;
+			*p = '\0';
+			event = p + 1;
+		} else {
+			group = "*";
+			event = str;
+		}
+		pr_debug("Group: %s, Event: %s\n", group, event);
+		ret = del_trace_probe_event(fd, group, event, namelist);
+		free(str);
+		if (ret < 0)
+			break;
+	}
+	strlist__delete(namelist);
+	close(fd);
+
+	return ret;
+}
+/* TODO: don't use a global variable for filter ... */
+static struct strfilter *available_func_filter;
+
+/*
+ * If a symbol corresponds to a function with global binding and
+ * matches filter return 0. For all others return 1.
+ */
+static int filter_available_functions(struct map *map __unused,
+				      struct symbol *sym)
+{
+	if (sym->binding == STB_GLOBAL &&
+	    strfilter__compare(available_func_filter, sym->name))
+		return 0;
+	return 1;
+}
+
+int show_available_funcs(const char *module, struct strfilter *_filter)
+{
+	struct map *map;
+	int ret;
+
+	setup_pager();
+
+	ret = init_vmlinux();
+	if (ret < 0)
+		return ret;
+
+	map = kernel_get_module_map(module);
+	if (!map) {
+		pr_err("Failed to find %s map.\n", (module) ? : "kernel");
+		return -EINVAL;
+	}
+	available_func_filter = _filter;
+	if (map__load(map, filter_available_functions)) {
+		pr_err("Failed to load map.\n");
+		return -EINVAL;
+	}
+	if (!dso__sorted_by_name(map->dso, map->type))
+		dso__sort_by_name(map->dso, map->type);
+
+	dso__fprintf_symbols_by_name(map->dso, map->type, stdout);
+	return 0;
+}
diff --git a/tools/perf/util/probe-event.h b/tools/perf/util/probe-event.h
new file mode 100644
index 00000000..3434fc9d
--- /dev/null
+++ b/tools/perf/util/probe-event.h
@@ -0,0 +1,137 @@
+#ifndef _PROBE_EVENT_H
+#define _PROBE_EVENT_H
+
+#include <stdbool.h>
+#include "strlist.h"
+#include "strfilter.h"
+
+extern bool probe_event_dry_run;
+
+/* kprobe-tracer tracing point */
+struct probe_trace_point {
+	char		*symbol;	/* Base symbol */
+	unsigned long	offset;		/* Offset from symbol */
+	bool		retprobe;	/* Return probe flag */
+};
+
+/* probe-tracer tracing argument referencing offset */
+struct probe_trace_arg_ref {
+	struct probe_trace_arg_ref	*next;	/* Next reference */
+	long				offset;	/* Offset value */
+};
+
+/* kprobe-tracer tracing argument */
+struct probe_trace_arg {
+	char				*name;	/* Argument name */
+	char				*value;	/* Base value */
+	char				*type;	/* Type name */
+	struct probe_trace_arg_ref	*ref;	/* Referencing offset */
+};
+
+/* kprobe-tracer tracing event (point + arg) */
+struct probe_trace_event {
+	char				*event;	/* Event name */
+	char				*group;	/* Group name */
+	struct probe_trace_point	point;	/* Trace point */
+	int				nargs;	/* Number of args */
+	struct probe_trace_arg		*args;	/* Arguments */
+};
+
+/* Perf probe probing point */
+struct perf_probe_point {
+	char		*file;		/* File path */
+	char		*function;	/* Function name */
+	int		line;		/* Line number */
+	bool		retprobe;	/* Return probe flag */
+	char		*lazy_line;	/* Lazy matching pattern */
+	unsigned long	offset;		/* Offset from function entry */
+};
+
+/* Perf probe probing argument field chain */
+struct perf_probe_arg_field {
+	struct perf_probe_arg_field	*next;	/* Next field */
+	char				*name;	/* Name of the field */
+	long				index;	/* Array index number */
+	bool				ref;	/* Referencing flag */
+};
+
+/* Perf probe probing argument */
+struct perf_probe_arg {
+	char				*name;	/* Argument name */
+	char				*var;	/* Variable name */
+	char				*type;	/* Type name */
+	struct perf_probe_arg_field	*field;	/* Structure fields */
+};
+
+/* Perf probe probing event (point + arg) */
+struct perf_probe_event {
+	char			*event;	/* Event name */
+	char			*group;	/* Group name */
+	struct perf_probe_point	point;	/* Probe point */
+	int			nargs;	/* Number of arguments */
+	struct perf_probe_arg	*args;	/* Arguments */
+};
+
+
+/* Line number container */
+struct line_node {
+	struct list_head	list;
+	int			line;
+};
+
+/* Line range */
+struct line_range {
+	char			*file;		/* File name */
+	char			*function;	/* Function name */
+	int			start;		/* Start line number */
+	int			end;		/* End line number */
+	int			offset;		/* Start line offset */
+	char			*path;		/* Real path name */
+	char			*comp_dir;	/* Compile directory */
+	struct list_head	line_list;	/* Visible lines */
+};
+
+/* List of variables */
+struct variable_list {
+	struct probe_trace_point	point;	/* Actual probepoint */
+	struct strlist			*vars;	/* Available variables */
+};
+
+/* Command string to events */
+extern int parse_perf_probe_command(const char *cmd,
+				    struct perf_probe_event *pev);
+
+/* Events to command string */
+extern char *synthesize_perf_probe_command(struct perf_probe_event *pev);
+extern char *synthesize_probe_trace_command(struct probe_trace_event *tev);
+extern int synthesize_perf_probe_arg(struct perf_probe_arg *pa, char *buf,
+				     size_t len);
+
+/* Check the perf_probe_event needs debuginfo */
+extern bool perf_probe_event_need_dwarf(struct perf_probe_event *pev);
+
+/* Release event contents */
+extern void clear_perf_probe_event(struct perf_probe_event *pev);
+
+/* Command string to line-range */
+extern int parse_line_range_desc(const char *cmd, struct line_range *lr);
+
+/* Internal use: Return kernel/module path */
+extern const char *kernel_get_module_path(const char *module);
+
+extern int add_perf_probe_events(struct perf_probe_event *pevs, int npevs,
+				 int max_probe_points, const char *module,
+				 bool force_add);
+extern int del_perf_probe_events(struct strlist *dellist);
+extern int show_perf_probe_events(void);
+extern int show_line_range(struct line_range *lr, const char *module);
+extern int show_available_vars(struct perf_probe_event *pevs, int npevs,
+			       int max_probe_points, const char *module,
+			       struct strfilter *filter, bool externs);
+extern int show_available_funcs(const char *module, struct strfilter *filter);
+
+
+/* Maximum index number of event-name postfix */
+#define MAX_EVENT_INDEX	1024
+
+#endif /*_PROBE_EVENT_H */
diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
new file mode 100644
index 00000000..3b9d0b80
--- /dev/null
+++ b/tools/perf/util/probe-finder.c
@@ -0,0 +1,2064 @@
+/*
+ * probe-finder.c : C expression to kprobe event converter
+ *
+ * Written by Masami Hiramatsu <mhiramat@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+
+#include <sys/utsname.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <errno.h>
+#include <stdio.h>
+#include <unistd.h>
+#include <getopt.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdarg.h>
+#include <ctype.h>
+#include <dwarf-regs.h>
+
+#include <linux/bitops.h>
+#include "event.h"
+#include "debug.h"
+#include "util.h"
+#include "symbol.h"
+#include "probe-finder.h"
+
+/* Kprobe tracer basic type is up to u64 */
+#define MAX_BASIC_TYPE_BITS	64
+
+/*
+ * Compare the tail of two strings.
+ * Return 0 if whole of either string is same as another's tail part.
+ */
+static int strtailcmp(const char *s1, const char *s2)
+{
+	int i1 = strlen(s1);
+	int i2 = strlen(s2);
+	while (--i1 >= 0 && --i2 >= 0) {
+		if (s1[i1] != s2[i2])
+			return s1[i1] - s2[i2];
+	}
+	return 0;
+}
+
+/* Line number list operations */
+
+/* Add a line to line number list */
+static int line_list__add_line(struct list_head *head, int line)
+{
+	struct line_node *ln;
+	struct list_head *p;
+
+	/* Reverse search, because new line will be the last one */
+	list_for_each_entry_reverse(ln, head, list) {
+		if (ln->line < line) {
+			p = &ln->list;
+			goto found;
+		} else if (ln->line == line)	/* Already exist */
+			return 1;
+	}
+	/* List is empty, or the smallest entry */
+	p = head;
+found:
+	pr_debug("line list: add a line %u\n", line);
+	ln = zalloc(sizeof(struct line_node));
+	if (ln == NULL)
+		return -ENOMEM;
+	ln->line = line;
+	INIT_LIST_HEAD(&ln->list);
+	list_add(&ln->list, p);
+	return 0;
+}
+
+/* Check if the line in line number list */
+static int line_list__has_line(struct list_head *head, int line)
+{
+	struct line_node *ln;
+
+	/* Reverse search, because new line will be the last one */
+	list_for_each_entry(ln, head, list)
+		if (ln->line == line)
+			return 1;
+
+	return 0;
+}
+
+/* Init line number list */
+static void line_list__init(struct list_head *head)
+{
+	INIT_LIST_HEAD(head);
+}
+
+/* Free line number list */
+static void line_list__free(struct list_head *head)
+{
+	struct line_node *ln;
+	while (!list_empty(head)) {
+		ln = list_first_entry(head, struct line_node, list);
+		list_del(&ln->list);
+		free(ln);
+	}
+}
+
+/* Dwarf FL wrappers */
+static char *debuginfo_path;	/* Currently dummy */
+
+static const Dwfl_Callbacks offline_callbacks = {
+	.find_debuginfo = dwfl_standard_find_debuginfo,
+	.debuginfo_path = &debuginfo_path,
+
+	.section_address = dwfl_offline_section_address,
+
+	/* We use this table for core files too.  */
+	.find_elf = dwfl_build_id_find_elf,
+};
+
+/* Get a Dwarf from offline image */
+static Dwarf *dwfl_init_offline_dwarf(int fd, Dwfl **dwflp, Dwarf_Addr *bias)
+{
+	Dwfl_Module *mod;
+	Dwarf *dbg = NULL;
+
+	if (!dwflp)
+		return NULL;
+
+	*dwflp = dwfl_begin(&offline_callbacks);
+	if (!*dwflp)
+		return NULL;
+
+	mod = dwfl_report_offline(*dwflp, "", "", fd);
+	if (!mod)
+		goto error;
+
+	dbg = dwfl_module_getdwarf(mod, bias);
+	if (!dbg) {
+error:
+		dwfl_end(*dwflp);
+		*dwflp = NULL;
+	}
+	return dbg;
+}
+
+#if _ELFUTILS_PREREQ(0, 148)
+/* This method is buggy if elfutils is older than 0.148 */
+static int __linux_kernel_find_elf(Dwfl_Module *mod,
+				   void **userdata,
+				   const char *module_name,
+				   Dwarf_Addr base,
+				   char **file_name, Elf **elfp)
+{
+	int fd;
+	const char *path = kernel_get_module_path(module_name);
+
+	pr_debug2("Use file %s for %s\n", path, module_name);
+	if (path) {
+		fd = open(path, O_RDONLY);
+		if (fd >= 0) {
+			*file_name = strdup(path);
+			return fd;
+		}
+	}
+	/* If failed, try to call standard method */
+	return dwfl_linux_kernel_find_elf(mod, userdata, module_name, base,
+					  file_name, elfp);
+}
+
+static const Dwfl_Callbacks kernel_callbacks = {
+	.find_debuginfo = dwfl_standard_find_debuginfo,
+	.debuginfo_path = &debuginfo_path,
+
+	.find_elf = __linux_kernel_find_elf,
+	.section_address = dwfl_linux_kernel_module_section_address,
+};
+
+/* Get a Dwarf from live kernel image */
+static Dwarf *dwfl_init_live_kernel_dwarf(Dwarf_Addr addr, Dwfl **dwflp,
+					  Dwarf_Addr *bias)
+{
+	Dwarf *dbg;
+
+	if (!dwflp)
+		return NULL;
+
+	*dwflp = dwfl_begin(&kernel_callbacks);
+	if (!*dwflp)
+		return NULL;
+
+	/* Load the kernel dwarves: Don't care the result here */
+	dwfl_linux_kernel_report_kernel(*dwflp);
+	dwfl_linux_kernel_report_modules(*dwflp);
+
+	dbg = dwfl_addrdwarf(*dwflp, addr, bias);
+	/* Here, check whether we could get a real dwarf */
+	if (!dbg) {
+		pr_debug("Failed to find kernel dwarf at %lx\n",
+			 (unsigned long)addr);
+		dwfl_end(*dwflp);
+		*dwflp = NULL;
+	}
+	return dbg;
+}
+#else
+/* With older elfutils, this just support kernel module... */
+static Dwarf *dwfl_init_live_kernel_dwarf(Dwarf_Addr addr __used, Dwfl **dwflp,
+					  Dwarf_Addr *bias)
+{
+	int fd;
+	const char *path = kernel_get_module_path("kernel");
+
+	if (!path) {
+		pr_err("Failed to find vmlinux path\n");
+		return NULL;
+	}
+
+	pr_debug2("Use file %s for debuginfo\n", path);
+	fd = open(path, O_RDONLY);
+	if (fd < 0)
+		return NULL;
+
+	return dwfl_init_offline_dwarf(fd, dwflp, bias);
+}
+#endif
+
+/* Dwarf wrappers */
+
+/* Find the realpath of the target file. */
+static const char *cu_find_realpath(Dwarf_Die *cu_die, const char *fname)
+{
+	Dwarf_Files *files;
+	size_t nfiles, i;
+	const char *src = NULL;
+	int ret;
+
+	if (!fname)
+		return NULL;
+
+	ret = dwarf_getsrcfiles(cu_die, &files, &nfiles);
+	if (ret != 0)
+		return NULL;
+
+	for (i = 0; i < nfiles; i++) {
+		src = dwarf_filesrc(files, i, NULL, NULL);
+		if (strtailcmp(src, fname) == 0)
+			break;
+	}
+	if (i == nfiles)
+		return NULL;
+	return src;
+}
+
+/* Get DW_AT_comp_dir (should be NULL with older gcc) */
+static const char *cu_get_comp_dir(Dwarf_Die *cu_die)
+{
+	Dwarf_Attribute attr;
+	if (dwarf_attr(cu_die, DW_AT_comp_dir, &attr) == NULL)
+		return NULL;
+	return dwarf_formstring(&attr);
+}
+
+/* Get a line number and file name for given address */
+static int cu_find_lineinfo(Dwarf_Die *cudie, unsigned long addr,
+			    const char **fname, int *lineno)
+{
+	Dwarf_Line *line;
+	Dwarf_Addr laddr;
+
+	line = dwarf_getsrc_die(cudie, (Dwarf_Addr)addr);
+	if (line && dwarf_lineaddr(line, &laddr) == 0 &&
+	    addr == (unsigned long)laddr && dwarf_lineno(line, lineno) == 0) {
+		*fname = dwarf_linesrc(line, NULL, NULL);
+		if (!*fname)
+			/* line number is useless without filename */
+			*lineno = 0;
+	}
+
+	return *lineno ?: -ENOENT;
+}
+
+/* Compare diename and tname */
+static bool die_compare_name(Dwarf_Die *dw_die, const char *tname)
+{
+	const char *name;
+	name = dwarf_diename(dw_die);
+	return name ? (strcmp(tname, name) == 0) : false;
+}
+
+/* Get callsite line number of inline-function instance */
+static int die_get_call_lineno(Dwarf_Die *in_die)
+{
+	Dwarf_Attribute attr;
+	Dwarf_Word ret;
+
+	if (!dwarf_attr(in_die, DW_AT_call_line, &attr))
+		return -ENOENT;
+
+	dwarf_formudata(&attr, &ret);
+	return (int)ret;
+}
+
+/* Get type die */
+static Dwarf_Die *die_get_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem)
+{
+	Dwarf_Attribute attr;
+
+	if (dwarf_attr_integrate(vr_die, DW_AT_type, &attr) &&
+	    dwarf_formref_die(&attr, die_mem))
+		return die_mem;
+	else
+		return NULL;
+}
+
+/* Get a type die, but skip qualifiers */
+static Dwarf_Die *__die_get_real_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem)
+{
+	int tag;
+
+	do {
+		vr_die = die_get_type(vr_die, die_mem);
+		if (!vr_die)
+			break;
+		tag = dwarf_tag(vr_die);
+	} while (tag == DW_TAG_const_type ||
+		 tag == DW_TAG_restrict_type ||
+		 tag == DW_TAG_volatile_type ||
+		 tag == DW_TAG_shared_type);
+
+	return vr_die;
+}
+
+/* Get a type die, but skip qualifiers and typedef */
+static Dwarf_Die *die_get_real_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem)
+{
+	do {
+		vr_die = __die_get_real_type(vr_die, die_mem);
+	} while (vr_die && dwarf_tag(vr_die) == DW_TAG_typedef);
+
+	return vr_die;
+}
+
+static int die_get_attr_udata(Dwarf_Die *tp_die, unsigned int attr_name,
+			      Dwarf_Word *result)
+{
+	Dwarf_Attribute attr;
+
+	if (dwarf_attr(tp_die, attr_name, &attr) == NULL ||
+	    dwarf_formudata(&attr, result) != 0)
+		return -ENOENT;
+
+	return 0;
+}
+
+static bool die_is_signed_type(Dwarf_Die *tp_die)
+{
+	Dwarf_Word ret;
+
+	if (die_get_attr_udata(tp_die, DW_AT_encoding, &ret))
+		return false;
+
+	return (ret == DW_ATE_signed_char || ret == DW_ATE_signed ||
+		ret == DW_ATE_signed_fixed);
+}
+
+static int die_get_byte_size(Dwarf_Die *tp_die)
+{
+	Dwarf_Word ret;
+
+	if (die_get_attr_udata(tp_die, DW_AT_byte_size, &ret))
+		return 0;
+
+	return (int)ret;
+}
+
+static int die_get_bit_size(Dwarf_Die *tp_die)
+{
+	Dwarf_Word ret;
+
+	if (die_get_attr_udata(tp_die, DW_AT_bit_size, &ret))
+		return 0;
+
+	return (int)ret;
+}
+
+static int die_get_bit_offset(Dwarf_Die *tp_die)
+{
+	Dwarf_Word ret;
+
+	if (die_get_attr_udata(tp_die, DW_AT_bit_offset, &ret))
+		return 0;
+
+	return (int)ret;
+}
+
+/* Get data_member_location offset */
+static int die_get_data_member_location(Dwarf_Die *mb_die, Dwarf_Word *offs)
+{
+	Dwarf_Attribute attr;
+	Dwarf_Op *expr;
+	size_t nexpr;
+	int ret;
+
+	if (dwarf_attr(mb_die, DW_AT_data_member_location, &attr) == NULL)
+		return -ENOENT;
+
+	if (dwarf_formudata(&attr, offs) != 0) {
+		/* DW_AT_data_member_location should be DW_OP_plus_uconst */
+		ret = dwarf_getlocation(&attr, &expr, &nexpr);
+		if (ret < 0 || nexpr == 0)
+			return -ENOENT;
+
+		if (expr[0].atom != DW_OP_plus_uconst || nexpr != 1) {
+			pr_debug("Unable to get offset:Unexpected OP %x (%zd)\n",
+				 expr[0].atom, nexpr);
+			return -ENOTSUP;
+		}
+		*offs = (Dwarf_Word)expr[0].number;
+	}
+	return 0;
+}
+
+/* Return values for die_find callbacks */
+enum {
+	DIE_FIND_CB_FOUND = 0,		/* End of Search */
+	DIE_FIND_CB_CHILD = 1,		/* Search only children */
+	DIE_FIND_CB_SIBLING = 2,	/* Search only siblings */
+	DIE_FIND_CB_CONTINUE = 3,	/* Search children and siblings */
+};
+
+/* Search a child die */
+static Dwarf_Die *die_find_child(Dwarf_Die *rt_die,
+				 int (*callback)(Dwarf_Die *, void *),
+				 void *data, Dwarf_Die *die_mem)
+{
+	Dwarf_Die child_die;
+	int ret;
+
+	ret = dwarf_child(rt_die, die_mem);
+	if (ret != 0)
+		return NULL;
+
+	do {
+		ret = callback(die_mem, data);
+		if (ret == DIE_FIND_CB_FOUND)
+			return die_mem;
+
+		if ((ret & DIE_FIND_CB_CHILD) &&
+		    die_find_child(die_mem, callback, data, &child_die)) {
+			memcpy(die_mem, &child_die, sizeof(Dwarf_Die));
+			return die_mem;
+		}
+	} while ((ret & DIE_FIND_CB_SIBLING) &&
+		 dwarf_siblingof(die_mem, die_mem) == 0);
+
+	return NULL;
+}
+
+struct __addr_die_search_param {
+	Dwarf_Addr	addr;
+	Dwarf_Die	*die_mem;
+};
+
+static int __die_search_func_cb(Dwarf_Die *fn_die, void *data)
+{
+	struct __addr_die_search_param *ad = data;
+
+	if (dwarf_tag(fn_die) == DW_TAG_subprogram &&
+	    dwarf_haspc(fn_die, ad->addr)) {
+		memcpy(ad->die_mem, fn_die, sizeof(Dwarf_Die));
+		return DWARF_CB_ABORT;
+	}
+	return DWARF_CB_OK;
+}
+
+/* Search a real subprogram including this line, */
+static Dwarf_Die *die_find_real_subprogram(Dwarf_Die *cu_die, Dwarf_Addr addr,
+					   Dwarf_Die *die_mem)
+{
+	struct __addr_die_search_param ad;
+	ad.addr = addr;
+	ad.die_mem = die_mem;
+	/* dwarf_getscopes can't find subprogram. */
+	if (!dwarf_getfuncs(cu_die, __die_search_func_cb, &ad, 0))
+		return NULL;
+	else
+		return die_mem;
+}
+
+/* die_find callback for inline function search */
+static int __die_find_inline_cb(Dwarf_Die *die_mem, void *data)
+{
+	Dwarf_Addr *addr = data;
+
+	if (dwarf_tag(die_mem) == DW_TAG_inlined_subroutine &&
+	    dwarf_haspc(die_mem, *addr))
+		return DIE_FIND_CB_FOUND;
+
+	return DIE_FIND_CB_CONTINUE;
+}
+
+/* Similar to dwarf_getfuncs, but returns inlined_subroutine if exists. */
+static Dwarf_Die *die_find_inlinefunc(Dwarf_Die *sp_die, Dwarf_Addr addr,
+				      Dwarf_Die *die_mem)
+{
+	Dwarf_Die tmp_die;
+
+	sp_die = die_find_child(sp_die, __die_find_inline_cb, &addr, &tmp_die);
+	if (!sp_die)
+		return NULL;
+
+	/* Inlined function could be recursive. Trace it until fail */
+	while (sp_die) {
+		memcpy(die_mem, sp_die, sizeof(Dwarf_Die));
+		sp_die = die_find_child(sp_die, __die_find_inline_cb, &addr,
+					&tmp_die);
+	}
+
+	return die_mem;
+}
+
+/* Walker on lines (Note: line number will not be sorted) */
+typedef int (* line_walk_handler_t) (const char *fname, int lineno,
+				     Dwarf_Addr addr, void *data);
+
+struct __line_walk_param {
+	const char *fname;
+	line_walk_handler_t handler;
+	void *data;
+	int retval;
+};
+
+static int __die_walk_funclines_cb(Dwarf_Die *in_die, void *data)
+{
+	struct __line_walk_param *lw = data;
+	Dwarf_Addr addr;
+	int lineno;
+
+	if (dwarf_tag(in_die) == DW_TAG_inlined_subroutine) {
+		lineno = die_get_call_lineno(in_die);
+		if (lineno > 0 && dwarf_entrypc(in_die, &addr) == 0) {
+			lw->retval = lw->handler(lw->fname, lineno, addr,
+						 lw->data);
+			if (lw->retval != 0)
+				return DIE_FIND_CB_FOUND;
+		}
+	}
+	return DIE_FIND_CB_SIBLING;
+}
+
+/* Walk on lines of blocks included in given DIE */
+static int __die_walk_funclines(Dwarf_Die *sp_die,
+				line_walk_handler_t handler, void *data)
+{
+	struct __line_walk_param lw = {
+		.handler = handler,
+		.data = data,
+		.retval = 0,
+	};
+	Dwarf_Die die_mem;
+	Dwarf_Addr addr;
+	int lineno;
+
+	/* Handle function declaration line */
+	lw.fname = dwarf_decl_file(sp_die);
+	if (lw.fname && dwarf_decl_line(sp_die, &lineno) == 0 &&
+	    dwarf_entrypc(sp_die, &addr) == 0) {
+		lw.retval = handler(lw.fname, lineno, addr, data);
+		if (lw.retval != 0)
+			goto done;
+	}
+	die_find_child(sp_die, __die_walk_funclines_cb, &lw, &die_mem);
+done:
+	return lw.retval;
+}
+
+static int __die_walk_culines_cb(Dwarf_Die *sp_die, void *data)
+{
+	struct __line_walk_param *lw = data;
+
+	lw->retval = __die_walk_funclines(sp_die, lw->handler, lw->data);
+	if (lw->retval != 0)
+		return DWARF_CB_ABORT;
+
+	return DWARF_CB_OK;
+}
+
+/*
+ * Walk on lines inside given PDIE. If the PDIE is subprogram, walk only on
+ * the lines inside the subprogram, otherwise PDIE must be a CU DIE.
+ */
+static int die_walk_lines(Dwarf_Die *pdie, line_walk_handler_t handler,
+			  void *data)
+{
+	Dwarf_Lines *lines;
+	Dwarf_Line *line;
+	Dwarf_Addr addr;
+	const char *fname;
+	int lineno, ret = 0;
+	Dwarf_Die die_mem, *cu_die;
+	size_t nlines, i;
+
+	/* Get the CU die */
+	if (dwarf_tag(pdie) == DW_TAG_subprogram)
+		cu_die = dwarf_diecu(pdie, &die_mem, NULL, NULL);
+	else
+		cu_die = pdie;
+	if (!cu_die) {
+		pr_debug2("Failed to get CU from subprogram\n");
+		return -EINVAL;
+	}
+
+	/* Get lines list in the CU */
+	if (dwarf_getsrclines(cu_die, &lines, &nlines) != 0) {
+		pr_debug2("Failed to get source lines on this CU.\n");
+		return -ENOENT;
+	}
+	pr_debug2("Get %zd lines from this CU\n", nlines);
+
+	/* Walk on the lines on lines list */
+	for (i = 0; i < nlines; i++) {
+		line = dwarf_onesrcline(lines, i);
+		if (line == NULL ||
+		    dwarf_lineno(line, &lineno) != 0 ||
+		    dwarf_lineaddr(line, &addr) != 0) {
+			pr_debug2("Failed to get line info. "
+				  "Possible error in debuginfo.\n");
+			continue;
+		}
+		/* Filter lines based on address */
+		if (pdie != cu_die)
+			/*
+			 * Address filtering
+			 * The line is included in given function, and
+			 * no inline block includes it.
+			 */
+			if (!dwarf_haspc(pdie, addr) ||
+			    die_find_inlinefunc(pdie, addr, &die_mem))
+				continue;
+		/* Get source line */
+		fname = dwarf_linesrc(line, NULL, NULL);
+
+		ret = handler(fname, lineno, addr, data);
+		if (ret != 0)
+			return ret;
+	}
+
+	/*
+	 * Dwarf lines doesn't include function declarations and inlined
+	 * subroutines. We have to check functions list or given function.
+	 */
+	if (pdie != cu_die)
+		ret = __die_walk_funclines(pdie, handler, data);
+	else {
+		struct __line_walk_param param = {
+			.handler = handler,
+			.data = data,
+			.retval = 0,
+		};
+		dwarf_getfuncs(cu_die, __die_walk_culines_cb, &param, 0);
+		ret = param.retval;
+	}
+
+	return ret;
+}
+
+struct __find_variable_param {
+	const char *name;
+	Dwarf_Addr addr;
+};
+
+static int __die_find_variable_cb(Dwarf_Die *die_mem, void *data)
+{
+	struct __find_variable_param *fvp = data;
+	int tag;
+
+	tag = dwarf_tag(die_mem);
+	if ((tag == DW_TAG_formal_parameter ||
+	     tag == DW_TAG_variable) &&
+	    die_compare_name(die_mem, fvp->name))
+		return DIE_FIND_CB_FOUND;
+
+	if (dwarf_haspc(die_mem, fvp->addr))
+		return DIE_FIND_CB_CONTINUE;
+	else
+		return DIE_FIND_CB_SIBLING;
+}
+
+/* Find a variable called 'name' at given address */
+static Dwarf_Die *die_find_variable_at(Dwarf_Die *sp_die, const char *name,
+				       Dwarf_Addr addr, Dwarf_Die *die_mem)
+{
+	struct __find_variable_param fvp = { .name = name, .addr = addr};
+
+	return die_find_child(sp_die, __die_find_variable_cb, (void *)&fvp,
+			      die_mem);
+}
+
+static int __die_find_member_cb(Dwarf_Die *die_mem, void *data)
+{
+	const char *name = data;
+
+	if ((dwarf_tag(die_mem) == DW_TAG_member) &&
+	    die_compare_name(die_mem, name))
+		return DIE_FIND_CB_FOUND;
+
+	return DIE_FIND_CB_SIBLING;
+}
+
+/* Find a member called 'name' */
+static Dwarf_Die *die_find_member(Dwarf_Die *st_die, const char *name,
+				  Dwarf_Die *die_mem)
+{
+	return die_find_child(st_die, __die_find_member_cb, (void *)name,
+			      die_mem);
+}
+
+/* Get the name of given variable DIE */
+static int die_get_typename(Dwarf_Die *vr_die, char *buf, int len)
+{
+	Dwarf_Die type;
+	int tag, ret, ret2;
+	const char *tmp = "";
+
+	if (__die_get_real_type(vr_die, &type) == NULL)
+		return -ENOENT;
+
+	tag = dwarf_tag(&type);
+	if (tag == DW_TAG_array_type || tag == DW_TAG_pointer_type)
+		tmp = "*";
+	else if (tag == DW_TAG_subroutine_type) {
+		/* Function pointer */
+		ret = snprintf(buf, len, "(function_type)");
+		return (ret >= len) ? -E2BIG : ret;
+	} else {
+		if (!dwarf_diename(&type))
+			return -ENOENT;
+		if (tag == DW_TAG_union_type)
+			tmp = "union ";
+		else if (tag == DW_TAG_structure_type)
+			tmp = "struct ";
+		/* Write a base name */
+		ret = snprintf(buf, len, "%s%s", tmp, dwarf_diename(&type));
+		return (ret >= len) ? -E2BIG : ret;
+	}
+	ret = die_get_typename(&type, buf, len);
+	if (ret > 0) {
+		ret2 = snprintf(buf + ret, len - ret, "%s", tmp);
+		ret = (ret2 >= len - ret) ? -E2BIG : ret2 + ret;
+	}
+	return ret;
+}
+
+/* Get the name and type of given variable DIE, stored as "type\tname" */
+static int die_get_varname(Dwarf_Die *vr_die, char *buf, int len)
+{
+	int ret, ret2;
+
+	ret = die_get_typename(vr_die, buf, len);
+	if (ret < 0) {
+		pr_debug("Failed to get type, make it unknown.\n");
+		ret = snprintf(buf, len, "(unknown_type)");
+	}
+	if (ret > 0) {
+		ret2 = snprintf(buf + ret, len - ret, "\t%s",
+				dwarf_diename(vr_die));
+		ret = (ret2 >= len - ret) ? -E2BIG : ret2 + ret;
+	}
+	return ret;
+}
+
+/*
+ * Probe finder related functions
+ */
+
+static struct probe_trace_arg_ref *alloc_trace_arg_ref(long offs)
+{
+	struct probe_trace_arg_ref *ref;
+	ref = zalloc(sizeof(struct probe_trace_arg_ref));
+	if (ref != NULL)
+		ref->offset = offs;
+	return ref;
+}
+
+/*
+ * Convert a location into trace_arg.
+ * If tvar == NULL, this just checks variable can be converted.
+ */
+static int convert_variable_location(Dwarf_Die *vr_die, Dwarf_Addr addr,
+				     Dwarf_Op *fb_ops,
+				     struct probe_trace_arg *tvar)
+{
+	Dwarf_Attribute attr;
+	Dwarf_Op *op;
+	size_t nops;
+	unsigned int regn;
+	Dwarf_Word offs = 0;
+	bool ref = false;
+	const char *regs;
+	int ret;
+
+	if (dwarf_attr(vr_die, DW_AT_external, &attr) != NULL)
+		goto static_var;
+
+	/* TODO: handle more than 1 exprs */
+	if (dwarf_attr(vr_die, DW_AT_location, &attr) == NULL ||
+	    dwarf_getlocation_addr(&attr, addr, &op, &nops, 1) <= 0 ||
+	    nops == 0) {
+		/* TODO: Support const_value */
+		return -ENOENT;
+	}
+
+	if (op->atom == DW_OP_addr) {
+static_var:
+		if (!tvar)
+			return 0;
+		/* Static variables on memory (not stack), make @varname */
+		ret = strlen(dwarf_diename(vr_die));
+		tvar->value = zalloc(ret + 2);
+		if (tvar->value == NULL)
+			return -ENOMEM;
+		snprintf(tvar->value, ret + 2, "@%s", dwarf_diename(vr_die));
+		tvar->ref = alloc_trace_arg_ref((long)offs);
+		if (tvar->ref == NULL)
+			return -ENOMEM;
+		return 0;
+	}
+
+	/* If this is based on frame buffer, set the offset */
+	if (op->atom == DW_OP_fbreg) {
+		if (fb_ops == NULL)
+			return -ENOTSUP;
+		ref = true;
+		offs = op->number;
+		op = &fb_ops[0];
+	}
+
+	if (op->atom >= DW_OP_breg0 && op->atom <= DW_OP_breg31) {
+		regn = op->atom - DW_OP_breg0;
+		offs += op->number;
+		ref = true;
+	} else if (op->atom >= DW_OP_reg0 && op->atom <= DW_OP_reg31) {
+		regn = op->atom - DW_OP_reg0;
+	} else if (op->atom == DW_OP_bregx) {
+		regn = op->number;
+		offs += op->number2;
+		ref = true;
+	} else if (op->atom == DW_OP_regx) {
+		regn = op->number;
+	} else {
+		pr_debug("DW_OP %x is not supported.\n", op->atom);
+		return -ENOTSUP;
+	}
+
+	if (!tvar)
+		return 0;
+
+	regs = get_arch_regstr(regn);
+	if (!regs) {
+		/* This should be a bug in DWARF or this tool */
+		pr_warning("Mapping for the register number %u "
+			   "missing on this architecture.\n", regn);
+		return -ERANGE;
+	}
+
+	tvar->value = strdup(regs);
+	if (tvar->value == NULL)
+		return -ENOMEM;
+
+	if (ref) {
+		tvar->ref = alloc_trace_arg_ref((long)offs);
+		if (tvar->ref == NULL)
+			return -ENOMEM;
+	}
+	return 0;
+}
+
+#define BYTES_TO_BITS(nb)	((nb) * BITS_PER_LONG / sizeof(long))
+
+static int convert_variable_type(Dwarf_Die *vr_die,
+				 struct probe_trace_arg *tvar,
+				 const char *cast)
+{
+	struct probe_trace_arg_ref **ref_ptr = &tvar->ref;
+	Dwarf_Die type;
+	char buf[16];
+	int ret;
+
+	/* TODO: check all types */
+	if (cast && strcmp(cast, "string") != 0) {
+		/* Non string type is OK */
+		tvar->type = strdup(cast);
+		return (tvar->type == NULL) ? -ENOMEM : 0;
+	}
+
+	if (die_get_bit_size(vr_die) != 0) {
+		/* This is a bitfield */
+		ret = snprintf(buf, 16, "b%d@%d/%zd", die_get_bit_size(vr_die),
+				die_get_bit_offset(vr_die),
+				BYTES_TO_BITS(die_get_byte_size(vr_die)));
+		goto formatted;
+	}
+
+	if (die_get_real_type(vr_die, &type) == NULL) {
+		pr_warning("Failed to get a type information of %s.\n",
+			   dwarf_diename(vr_die));
+		return -ENOENT;
+	}
+
+	pr_debug("%s type is %s.\n",
+		 dwarf_diename(vr_die), dwarf_diename(&type));
+
+	if (cast && strcmp(cast, "string") == 0) {	/* String type */
+		ret = dwarf_tag(&type);
+		if (ret != DW_TAG_pointer_type &&
+		    ret != DW_TAG_array_type) {
+			pr_warning("Failed to cast into string: "
+				   "%s(%s) is not a pointer nor array.\n",
+				   dwarf_diename(vr_die), dwarf_diename(&type));
+			return -EINVAL;
+		}
+		if (ret == DW_TAG_pointer_type) {
+			if (die_get_real_type(&type, &type) == NULL) {
+				pr_warning("Failed to get a type"
+					   " information.\n");
+				return -ENOENT;
+			}
+			while (*ref_ptr)
+				ref_ptr = &(*ref_ptr)->next;
+			/* Add new reference with offset +0 */
+			*ref_ptr = zalloc(sizeof(struct probe_trace_arg_ref));
+			if (*ref_ptr == NULL) {
+				pr_warning("Out of memory error\n");
+				return -ENOMEM;
+			}
+		}
+		if (!die_compare_name(&type, "char") &&
+		    !die_compare_name(&type, "unsigned char")) {
+			pr_warning("Failed to cast into string: "
+				   "%s is not (unsigned) char *.\n",
+				   dwarf_diename(vr_die));
+			return -EINVAL;
+		}
+		tvar->type = strdup(cast);
+		return (tvar->type == NULL) ? -ENOMEM : 0;
+	}
+
+	ret = BYTES_TO_BITS(die_get_byte_size(&type));
+	if (!ret)
+		/* No size ... try to use default type */
+		return 0;
+
+	/* Check the bitwidth */
+	if (ret > MAX_BASIC_TYPE_BITS) {
+		pr_info("%s exceeds max-bitwidth. Cut down to %d bits.\n",
+			dwarf_diename(&type), MAX_BASIC_TYPE_BITS);
+		ret = MAX_BASIC_TYPE_BITS;
+	}
+	ret = snprintf(buf, 16, "%c%d",
+		       die_is_signed_type(&type) ? 's' : 'u', ret);
+
+formatted:
+	if (ret < 0 || ret >= 16) {
+		if (ret >= 16)
+			ret = -E2BIG;
+		pr_warning("Failed to convert variable type: %s\n",
+			   strerror(-ret));
+		return ret;
+	}
+	tvar->type = strdup(buf);
+	if (tvar->type == NULL)
+		return -ENOMEM;
+	return 0;
+}
+
+static int convert_variable_fields(Dwarf_Die *vr_die, const char *varname,
+				    struct perf_probe_arg_field *field,
+				    struct probe_trace_arg_ref **ref_ptr,
+				    Dwarf_Die *die_mem)
+{
+	struct probe_trace_arg_ref *ref = *ref_ptr;
+	Dwarf_Die type;
+	Dwarf_Word offs;
+	int ret, tag;
+
+	pr_debug("converting %s in %s\n", field->name, varname);
+	if (die_get_real_type(vr_die, &type) == NULL) {
+		pr_warning("Failed to get the type of %s.\n", varname);
+		return -ENOENT;
+	}
+	pr_debug2("Var real type: (%x)\n", (unsigned)dwarf_dieoffset(&type));
+	tag = dwarf_tag(&type);
+
+	if (field->name[0] == '[' &&
+	    (tag == DW_TAG_array_type || tag == DW_TAG_pointer_type)) {
+		if (field->next)
+			/* Save original type for next field */
+			memcpy(die_mem, &type, sizeof(*die_mem));
+		/* Get the type of this array */
+		if (die_get_real_type(&type, &type) == NULL) {
+			pr_warning("Failed to get the type of %s.\n", varname);
+			return -ENOENT;
+		}
+		pr_debug2("Array real type: (%x)\n",
+			 (unsigned)dwarf_dieoffset(&type));
+		if (tag == DW_TAG_pointer_type) {
+			ref = zalloc(sizeof(struct probe_trace_arg_ref));
+			if (ref == NULL)
+				return -ENOMEM;
+			if (*ref_ptr)
+				(*ref_ptr)->next = ref;
+			else
+				*ref_ptr = ref;
+		}
+		ref->offset += die_get_byte_size(&type) * field->index;
+		if (!field->next)
+			/* Save vr_die for converting types */
+			memcpy(die_mem, vr_die, sizeof(*die_mem));
+		goto next;
+	} else if (tag == DW_TAG_pointer_type) {
+		/* Check the pointer and dereference */
+		if (!field->ref) {
+			pr_err("Semantic error: %s must be referred by '->'\n",
+			       field->name);
+			return -EINVAL;
+		}
+		/* Get the type pointed by this pointer */
+		if (die_get_real_type(&type, &type) == NULL) {
+			pr_warning("Failed to get the type of %s.\n", varname);
+			return -ENOENT;
+		}
+		/* Verify it is a data structure  */
+		if (dwarf_tag(&type) != DW_TAG_structure_type) {
+			pr_warning("%s is not a data structure.\n", varname);
+			return -EINVAL;
+		}
+
+		ref = zalloc(sizeof(struct probe_trace_arg_ref));
+		if (ref == NULL)
+			return -ENOMEM;
+		if (*ref_ptr)
+			(*ref_ptr)->next = ref;
+		else
+			*ref_ptr = ref;
+	} else {
+		/* Verify it is a data structure  */
+		if (tag != DW_TAG_structure_type) {
+			pr_warning("%s is not a data structure.\n", varname);
+			return -EINVAL;
+		}
+		if (field->name[0] == '[') {
+			pr_err("Semantic error: %s is not a pointor"
+			       " nor array.\n", varname);
+			return -EINVAL;
+		}
+		if (field->ref) {
+			pr_err("Semantic error: %s must be referred by '.'\n",
+			       field->name);
+			return -EINVAL;
+		}
+		if (!ref) {
+			pr_warning("Structure on a register is not "
+				   "supported yet.\n");
+			return -ENOTSUP;
+		}
+	}
+
+	if (die_find_member(&type, field->name, die_mem) == NULL) {
+		pr_warning("%s(tyep:%s) has no member %s.\n", varname,
+			   dwarf_diename(&type), field->name);
+		return -EINVAL;
+	}
+
+	/* Get the offset of the field */
+	ret = die_get_data_member_location(die_mem, &offs);
+	if (ret < 0) {
+		pr_warning("Failed to get the offset of %s.\n", field->name);
+		return ret;
+	}
+	ref->offset += (long)offs;
+
+next:
+	/* Converting next field */
+	if (field->next)
+		return convert_variable_fields(die_mem, field->name,
+					field->next, &ref, die_mem);
+	else
+		return 0;
+}
+
+/* Show a variables in kprobe event format */
+static int convert_variable(Dwarf_Die *vr_die, struct probe_finder *pf)
+{
+	Dwarf_Die die_mem;
+	int ret;
+
+	pr_debug("Converting variable %s into trace event.\n",
+		 dwarf_diename(vr_die));
+
+	ret = convert_variable_location(vr_die, pf->addr, pf->fb_ops,
+					pf->tvar);
+	if (ret == -ENOENT)
+		pr_err("Failed to find the location of %s at this address.\n"
+		       " Perhaps, it has been optimized out.\n", pf->pvar->var);
+	else if (ret == -ENOTSUP)
+		pr_err("Sorry, we don't support this variable location yet.\n");
+	else if (pf->pvar->field) {
+		ret = convert_variable_fields(vr_die, pf->pvar->var,
+					      pf->pvar->field, &pf->tvar->ref,
+					      &die_mem);
+		vr_die = &die_mem;
+	}
+	if (ret == 0)
+		ret = convert_variable_type(vr_die, pf->tvar, pf->pvar->type);
+	/* *expr will be cached in libdw. Don't free it. */
+	return ret;
+}
+
+/* Find a variable in a subprogram die */
+static int find_variable(Dwarf_Die *sp_die, struct probe_finder *pf)
+{
+	Dwarf_Die vr_die, *scopes;
+	char buf[32], *ptr;
+	int ret, nscopes;
+
+	if (!is_c_varname(pf->pvar->var)) {
+		/* Copy raw parameters */
+		pf->tvar->value = strdup(pf->pvar->var);
+		if (pf->tvar->value == NULL)
+			return -ENOMEM;
+		if (pf->pvar->type) {
+			pf->tvar->type = strdup(pf->pvar->type);
+			if (pf->tvar->type == NULL)
+				return -ENOMEM;
+		}
+		if (pf->pvar->name) {
+			pf->tvar->name = strdup(pf->pvar->name);
+			if (pf->tvar->name == NULL)
+				return -ENOMEM;
+		} else
+			pf->tvar->name = NULL;
+		return 0;
+	}
+
+	if (pf->pvar->name)
+		pf->tvar->name = strdup(pf->pvar->name);
+	else {
+		ret = synthesize_perf_probe_arg(pf->pvar, buf, 32);
+		if (ret < 0)
+			return ret;
+		ptr = strchr(buf, ':');	/* Change type separator to _ */
+		if (ptr)
+			*ptr = '_';
+		pf->tvar->name = strdup(buf);
+	}
+	if (pf->tvar->name == NULL)
+		return -ENOMEM;
+
+	pr_debug("Searching '%s' variable in context.\n",
+		 pf->pvar->var);
+	/* Search child die for local variables and parameters. */
+	if (die_find_variable_at(sp_die, pf->pvar->var, pf->addr, &vr_die))
+		ret = convert_variable(&vr_die, pf);
+	else {
+		/* Search upper class */
+		nscopes = dwarf_getscopes_die(sp_die, &scopes);
+		while (nscopes-- > 1) {
+			pr_debug("Searching variables in %s\n",
+				 dwarf_diename(&scopes[nscopes]));
+			/* We should check this scope, so give dummy address */
+			if (die_find_variable_at(&scopes[nscopes],
+						 pf->pvar->var, 0,
+						 &vr_die)) {
+				ret = convert_variable(&vr_die, pf);
+				goto found;
+			}
+		}
+		if (scopes)
+			free(scopes);
+		ret = -ENOENT;
+	}
+found:
+	if (ret < 0)
+		pr_warning("Failed to find '%s' in this function.\n",
+			   pf->pvar->var);
+	return ret;
+}
+
+/* Convert subprogram DIE to trace point */
+static int convert_to_trace_point(Dwarf_Die *sp_die, Dwarf_Addr paddr,
+				  bool retprobe, struct probe_trace_point *tp)
+{
+	Dwarf_Addr eaddr;
+	const char *name;
+
+	/* Copy the name of probe point */
+	name = dwarf_diename(sp_die);
+	if (name) {
+		if (dwarf_entrypc(sp_die, &eaddr) != 0) {
+			pr_warning("Failed to get entry address of %s\n",
+				   dwarf_diename(sp_die));
+			return -ENOENT;
+		}
+		tp->symbol = strdup(name);
+		if (tp->symbol == NULL)
+			return -ENOMEM;
+		tp->offset = (unsigned long)(paddr - eaddr);
+	} else
+		/* This function has no name. */
+		tp->offset = (unsigned long)paddr;
+
+	/* Return probe must be on the head of a subprogram */
+	if (retprobe) {
+		if (eaddr != paddr) {
+			pr_warning("Return probe must be on the head of"
+				   " a real function.\n");
+			return -EINVAL;
+		}
+		tp->retprobe = true;
+	}
+
+	return 0;
+}
+
+/* Call probe_finder callback with real subprogram DIE */
+static int call_probe_finder(Dwarf_Die *sp_die, struct probe_finder *pf)
+{
+	Dwarf_Die die_mem;
+	Dwarf_Attribute fb_attr;
+	size_t nops;
+	int ret;
+
+	/* If no real subprogram, find a real one */
+	if (!sp_die || dwarf_tag(sp_die) != DW_TAG_subprogram) {
+		sp_die = die_find_real_subprogram(&pf->cu_die,
+						  pf->addr, &die_mem);
+		if (!sp_die) {
+			pr_warning("Failed to find probe point in any "
+				   "functions.\n");
+			return -ENOENT;
+		}
+	}
+
+	/* Get the frame base attribute/ops */
+	dwarf_attr(sp_die, DW_AT_frame_base, &fb_attr);
+	ret = dwarf_getlocation_addr(&fb_attr, pf->addr, &pf->fb_ops, &nops, 1);
+	if (ret <= 0 || nops == 0) {
+		pf->fb_ops = NULL;
+#if _ELFUTILS_PREREQ(0, 142)
+	} else if (nops == 1 && pf->fb_ops[0].atom == DW_OP_call_frame_cfa &&
+		   pf->cfi != NULL) {
+		Dwarf_Frame *frame;
+		if (dwarf_cfi_addrframe(pf->cfi, pf->addr, &frame) != 0 ||
+		    dwarf_frame_cfa(frame, &pf->fb_ops, &nops) != 0) {
+			pr_warning("Failed to get call frame on 0x%jx\n",
+				   (uintmax_t)pf->addr);
+			return -ENOENT;
+		}
+#endif
+	}
+
+	/* Call finder's callback handler */
+	ret = pf->callback(sp_die, pf);
+
+	/* *pf->fb_ops will be cached in libdw. Don't free it. */
+	pf->fb_ops = NULL;
+
+	return ret;
+}
+
+static int probe_point_line_walker(const char *fname, int lineno,
+				   Dwarf_Addr addr, void *data)
+{
+	struct probe_finder *pf = data;
+	int ret;
+
+	if (lineno != pf->lno || strtailcmp(fname, pf->fname) != 0)
+		return 0;
+
+	pf->addr = addr;
+	ret = call_probe_finder(NULL, pf);
+
+	/* Continue if no error, because the line will be in inline function */
+	return ret < 0 ? ret : 0;
+}
+
+/* Find probe point from its line number */
+static int find_probe_point_by_line(struct probe_finder *pf)
+{
+	return die_walk_lines(&pf->cu_die, probe_point_line_walker, pf);
+}
+
+/* Find lines which match lazy pattern */
+static int find_lazy_match_lines(struct list_head *head,
+				 const char *fname, const char *pat)
+{
+	FILE *fp;
+	char *line = NULL;
+	size_t line_len;
+	ssize_t len;
+	int count = 0, linenum = 1;
+
+	fp = fopen(fname, "r");
+	if (!fp) {
+		pr_warning("Failed to open %s: %s\n", fname, strerror(errno));
+		return -errno;
+	}
+
+	while ((len = getline(&line, &line_len, fp)) > 0) {
+
+		if (line[len - 1] == '\n')
+			line[len - 1] = '\0';
+
+		if (strlazymatch(line, pat)) {
+			line_list__add_line(head, linenum);
+			count++;
+		}
+		linenum++;
+	}
+
+	if (ferror(fp))
+		count = -errno;
+	free(line);
+	fclose(fp);
+
+	if (count == 0)
+		pr_debug("No matched lines found in %s.\n", fname);
+	return count;
+}
+
+static int probe_point_lazy_walker(const char *fname, int lineno,
+				   Dwarf_Addr addr, void *data)
+{
+	struct probe_finder *pf = data;
+	int ret;
+
+	if (!line_list__has_line(&pf->lcache, lineno) ||
+	    strtailcmp(fname, pf->fname) != 0)
+		return 0;
+
+	pr_debug("Probe line found: line:%d addr:0x%llx\n",
+		 lineno, (unsigned long long)addr);
+	pf->addr = addr;
+	ret = call_probe_finder(NULL, pf);
+
+	/*
+	 * Continue if no error, because the lazy pattern will match
+	 * to other lines
+	 */
+	return ret < 0 ? ret : 0;
+}
+
+/* Find probe points from lazy pattern  */
+static int find_probe_point_lazy(Dwarf_Die *sp_die, struct probe_finder *pf)
+{
+	int ret = 0;
+
+	if (list_empty(&pf->lcache)) {
+		/* Matching lazy line pattern */
+		ret = find_lazy_match_lines(&pf->lcache, pf->fname,
+					    pf->pev->point.lazy_line);
+		if (ret <= 0)
+			return ret;
+	}
+
+	return die_walk_lines(sp_die, probe_point_lazy_walker, pf);
+}
+
+/* Callback parameter with return value */
+struct dwarf_callback_param {
+	void *data;
+	int retval;
+};
+
+static int probe_point_inline_cb(Dwarf_Die *in_die, void *data)
+{
+	struct dwarf_callback_param *param = data;
+	struct probe_finder *pf = param->data;
+	struct perf_probe_point *pp = &pf->pev->point;
+	Dwarf_Addr addr;
+
+	if (pp->lazy_line)
+		param->retval = find_probe_point_lazy(in_die, pf);
+	else {
+		/* Get probe address */
+		if (dwarf_entrypc(in_die, &addr) != 0) {
+			pr_warning("Failed to get entry address of %s.\n",
+				   dwarf_diename(in_die));
+			param->retval = -ENOENT;
+			return DWARF_CB_ABORT;
+		}
+		pf->addr = addr;
+		pf->addr += pp->offset;
+		pr_debug("found inline addr: 0x%jx\n",
+			 (uintmax_t)pf->addr);
+
+		param->retval = call_probe_finder(in_die, pf);
+		if (param->retval < 0)
+			return DWARF_CB_ABORT;
+	}
+
+	return DWARF_CB_OK;
+}
+
+/* Search function from function name */
+static int probe_point_search_cb(Dwarf_Die *sp_die, void *data)
+{
+	struct dwarf_callback_param *param = data;
+	struct probe_finder *pf = param->data;
+	struct perf_probe_point *pp = &pf->pev->point;
+
+	/* Check tag and diename */
+	if (dwarf_tag(sp_die) != DW_TAG_subprogram ||
+	    !die_compare_name(sp_die, pp->function))
+		return DWARF_CB_OK;
+
+	/* Check declared file */
+	if (pp->file && strtailcmp(pp->file, dwarf_decl_file(sp_die)))
+		return DWARF_CB_OK;
+
+	pf->fname = dwarf_decl_file(sp_die);
+	if (pp->line) { /* Function relative line */
+		dwarf_decl_line(sp_die, &pf->lno);
+		pf->lno += pp->line;
+		param->retval = find_probe_point_by_line(pf);
+	} else if (!dwarf_func_inline(sp_die)) {
+		/* Real function */
+		if (pp->lazy_line)
+			param->retval = find_probe_point_lazy(sp_die, pf);
+		else {
+			if (dwarf_entrypc(sp_die, &pf->addr) != 0) {
+				pr_warning("Failed to get entry address of "
+					   "%s.\n", dwarf_diename(sp_die));
+				param->retval = -ENOENT;
+				return DWARF_CB_ABORT;
+			}
+			pf->addr += pp->offset;
+			/* TODO: Check the address in this function */
+			param->retval = call_probe_finder(sp_die, pf);
+		}
+	} else {
+		struct dwarf_callback_param _param = {.data = (void *)pf,
+						      .retval = 0};
+		/* Inlined function: search instances */
+		dwarf_func_inline_instances(sp_die, probe_point_inline_cb,
+					    &_param);
+		param->retval = _param.retval;
+	}
+
+	return DWARF_CB_ABORT; /* Exit; no same symbol in this CU. */
+}
+
+static int find_probe_point_by_func(struct probe_finder *pf)
+{
+	struct dwarf_callback_param _param = {.data = (void *)pf,
+					      .retval = 0};
+	dwarf_getfuncs(&pf->cu_die, probe_point_search_cb, &_param, 0);
+	return _param.retval;
+}
+
+struct pubname_callback_param {
+	char *function;
+	char *file;
+	Dwarf_Die *cu_die;
+	Dwarf_Die *sp_die;
+	int found;
+};
+
+static int pubname_search_cb(Dwarf *dbg, Dwarf_Global *gl, void *data)
+{
+	struct pubname_callback_param *param = data;
+
+	if (dwarf_offdie(dbg, gl->die_offset, param->sp_die)) {
+		if (dwarf_tag(param->sp_die) != DW_TAG_subprogram)
+			return DWARF_CB_OK;
+
+		if (die_compare_name(param->sp_die, param->function)) {
+			if (!dwarf_offdie(dbg, gl->cu_offset, param->cu_die))
+				return DWARF_CB_OK;
+
+			if (param->file &&
+			    strtailcmp(param->file, dwarf_decl_file(param->sp_die)))
+				return DWARF_CB_OK;
+
+			param->found = 1;
+			return DWARF_CB_ABORT;
+		}
+	}
+
+	return DWARF_CB_OK;
+}
+
+/* Find probe points from debuginfo */
+static int find_probes(int fd, struct probe_finder *pf)
+{
+	struct perf_probe_point *pp = &pf->pev->point;
+	Dwarf_Off off, noff;
+	size_t cuhl;
+	Dwarf_Die *diep;
+	Dwarf *dbg = NULL;
+	Dwfl *dwfl;
+	Dwarf_Addr bias;	/* Currently ignored */
+	int ret = 0;
+
+	dbg = dwfl_init_offline_dwarf(fd, &dwfl, &bias);
+	if (!dbg) {
+		pr_warning("No debug information found in the vmlinux - "
+			"please rebuild with CONFIG_DEBUG_INFO=y.\n");
+		close(fd);	/* Without dwfl_end(), fd isn't closed. */
+		return -EBADF;
+	}
+
+#if _ELFUTILS_PREREQ(0, 142)
+	/* Get the call frame information from this dwarf */
+	pf->cfi = dwarf_getcfi(dbg);
+#endif
+
+	off = 0;
+	line_list__init(&pf->lcache);
+
+	/* Fastpath: lookup by function name from .debug_pubnames section */
+	if (pp->function) {
+		struct pubname_callback_param pubname_param = {
+			.function = pp->function,
+			.file	  = pp->file,
+			.cu_die	  = &pf->cu_die,
+			.sp_die	  = &pf->sp_die,
+			.found	  = 0,
+		};
+		struct dwarf_callback_param probe_param = {
+			.data = pf,
+		};
+
+		dwarf_getpubnames(dbg, pubname_search_cb, &pubname_param, 0);
+		if (pubname_param.found) {
+			ret = probe_point_search_cb(&pf->sp_die, &probe_param);
+			if (ret)
+				goto found;
+		}
+	}
+
+	/* Loop on CUs (Compilation Unit) */
+	while (!dwarf_nextcu(dbg, off, &noff, &cuhl, NULL, NULL, NULL)) {
+		/* Get the DIE(Debugging Information Entry) of this CU */
+		diep = dwarf_offdie(dbg, off + cuhl, &pf->cu_die);
+		if (!diep)
+			continue;
+
+		/* Check if target file is included. */
+		if (pp->file)
+			pf->fname = cu_find_realpath(&pf->cu_die, pp->file);
+		else
+			pf->fname = NULL;
+
+		if (!pp->file || pf->fname) {
+			if (pp->function)
+				ret = find_probe_point_by_func(pf);
+			else if (pp->lazy_line)
+				ret = find_probe_point_lazy(NULL, pf);
+			else {
+				pf->lno = pp->line;
+				ret = find_probe_point_by_line(pf);
+			}
+			if (ret < 0)
+				break;
+		}
+		off = noff;
+	}
+
+found:
+	line_list__free(&pf->lcache);
+	if (dwfl)
+		dwfl_end(dwfl);
+
+	return ret;
+}
+
+/* Add a found probe point into trace event list */
+static int add_probe_trace_event(Dwarf_Die *sp_die, struct probe_finder *pf)
+{
+	struct trace_event_finder *tf =
+			container_of(pf, struct trace_event_finder, pf);
+	struct probe_trace_event *tev;
+	int ret, i;
+
+	/* Check number of tevs */
+	if (tf->ntevs == tf->max_tevs) {
+		pr_warning("Too many( > %d) probe point found.\n",
+			   tf->max_tevs);
+		return -ERANGE;
+	}
+	tev = &tf->tevs[tf->ntevs++];
+
+	ret = convert_to_trace_point(sp_die, pf->addr, pf->pev->point.retprobe,
+				     &tev->point);
+	if (ret < 0)
+		return ret;
+
+	pr_debug("Probe point found: %s+%lu\n", tev->point.symbol,
+		 tev->point.offset);
+
+	/* Find each argument */
+	tev->nargs = pf->pev->nargs;
+	tev->args = zalloc(sizeof(struct probe_trace_arg) * tev->nargs);
+	if (tev->args == NULL)
+		return -ENOMEM;
+	for (i = 0; i < pf->pev->nargs; i++) {
+		pf->pvar = &pf->pev->args[i];
+		pf->tvar = &tev->args[i];
+		ret = find_variable(sp_die, pf);
+		if (ret != 0)
+			return ret;
+	}
+
+	return 0;
+}
+
+/* Find probe_trace_events specified by perf_probe_event from debuginfo */
+int find_probe_trace_events(int fd, struct perf_probe_event *pev,
+			    struct probe_trace_event **tevs, int max_tevs)
+{
+	struct trace_event_finder tf = {
+			.pf = {.pev = pev, .callback = add_probe_trace_event},
+			.max_tevs = max_tevs};
+	int ret;
+
+	/* Allocate result tevs array */
+	*tevs = zalloc(sizeof(struct probe_trace_event) * max_tevs);
+	if (*tevs == NULL)
+		return -ENOMEM;
+
+	tf.tevs = *tevs;
+	tf.ntevs = 0;
+
+	ret = find_probes(fd, &tf.pf);
+	if (ret < 0) {
+		free(*tevs);
+		*tevs = NULL;
+		return ret;
+	}
+
+	return (ret < 0) ? ret : tf.ntevs;
+}
+
+#define MAX_VAR_LEN 64
+
+/* Collect available variables in this scope */
+static int collect_variables_cb(Dwarf_Die *die_mem, void *data)
+{
+	struct available_var_finder *af = data;
+	struct variable_list *vl;
+	char buf[MAX_VAR_LEN];
+	int tag, ret;
+
+	vl = &af->vls[af->nvls - 1];
+
+	tag = dwarf_tag(die_mem);
+	if (tag == DW_TAG_formal_parameter ||
+	    tag == DW_TAG_variable) {
+		ret = convert_variable_location(die_mem, af->pf.addr,
+						af->pf.fb_ops, NULL);
+		if (ret == 0) {
+			ret = die_get_varname(die_mem, buf, MAX_VAR_LEN);
+			pr_debug2("Add new var: %s\n", buf);
+			if (ret > 0)
+				strlist__add(vl->vars, buf);
+		}
+	}
+
+	if (af->child && dwarf_haspc(die_mem, af->pf.addr))
+		return DIE_FIND_CB_CONTINUE;
+	else
+		return DIE_FIND_CB_SIBLING;
+}
+
+/* Add a found vars into available variables list */
+static int add_available_vars(Dwarf_Die *sp_die, struct probe_finder *pf)
+{
+	struct available_var_finder *af =
+			container_of(pf, struct available_var_finder, pf);
+	struct variable_list *vl;
+	Dwarf_Die die_mem, *scopes = NULL;
+	int ret, nscopes;
+
+	/* Check number of tevs */
+	if (af->nvls == af->max_vls) {
+		pr_warning("Too many( > %d) probe point found.\n", af->max_vls);
+		return -ERANGE;
+	}
+	vl = &af->vls[af->nvls++];
+
+	ret = convert_to_trace_point(sp_die, pf->addr, pf->pev->point.retprobe,
+				     &vl->point);
+	if (ret < 0)
+		return ret;
+
+	pr_debug("Probe point found: %s+%lu\n", vl->point.symbol,
+		 vl->point.offset);
+
+	/* Find local variables */
+	vl->vars = strlist__new(true, NULL);
+	if (vl->vars == NULL)
+		return -ENOMEM;
+	af->child = true;
+	die_find_child(sp_die, collect_variables_cb, (void *)af, &die_mem);
+
+	/* Find external variables */
+	if (!af->externs)
+		goto out;
+	/* Don't need to search child DIE for externs. */
+	af->child = false;
+	nscopes = dwarf_getscopes_die(sp_die, &scopes);
+	while (nscopes-- > 1)
+		die_find_child(&scopes[nscopes], collect_variables_cb,
+			       (void *)af, &die_mem);
+	if (scopes)
+		free(scopes);
+
+out:
+	if (strlist__empty(vl->vars)) {
+		strlist__delete(vl->vars);
+		vl->vars = NULL;
+	}
+
+	return ret;
+}
+
+/* Find available variables at given probe point */
+int find_available_vars_at(int fd, struct perf_probe_event *pev,
+			   struct variable_list **vls, int max_vls,
+			   bool externs)
+{
+	struct available_var_finder af = {
+			.pf = {.pev = pev, .callback = add_available_vars},
+			.max_vls = max_vls, .externs = externs};
+	int ret;
+
+	/* Allocate result vls array */
+	*vls = zalloc(sizeof(struct variable_list) * max_vls);
+	if (*vls == NULL)
+		return -ENOMEM;
+
+	af.vls = *vls;
+	af.nvls = 0;
+
+	ret = find_probes(fd, &af.pf);
+	if (ret < 0) {
+		/* Free vlist for error */
+		while (af.nvls--) {
+			if (af.vls[af.nvls].point.symbol)
+				free(af.vls[af.nvls].point.symbol);
+			if (af.vls[af.nvls].vars)
+				strlist__delete(af.vls[af.nvls].vars);
+		}
+		free(af.vls);
+		*vls = NULL;
+		return ret;
+	}
+
+	return (ret < 0) ? ret : af.nvls;
+}
+
+/* Reverse search */
+int find_perf_probe_point(unsigned long addr, struct perf_probe_point *ppt)
+{
+	Dwarf_Die cudie, spdie, indie;
+	Dwarf *dbg = NULL;
+	Dwfl *dwfl = NULL;
+	Dwarf_Addr _addr, baseaddr, bias = 0;
+	const char *fname = NULL, *func = NULL, *tmp;
+	int baseline = 0, lineno = 0, ret = 0;
+
+	/* Open the live linux kernel */
+	dbg = dwfl_init_live_kernel_dwarf(addr, &dwfl, &bias);
+	if (!dbg) {
+		pr_warning("No debug information found in the vmlinux - "
+			"please rebuild with CONFIG_DEBUG_INFO=y.\n");
+		ret = -EINVAL;
+		goto end;
+	}
+
+	/* Adjust address with bias */
+	addr += bias;
+	/* Find cu die */
+	if (!dwarf_addrdie(dbg, (Dwarf_Addr)addr - bias, &cudie)) {
+		pr_warning("Failed to find debug information for address %lx\n",
+			   addr);
+		ret = -EINVAL;
+		goto end;
+	}
+
+	/* Find a corresponding line (filename and lineno) */
+	cu_find_lineinfo(&cudie, addr, &fname, &lineno);
+	/* Don't care whether it failed or not */
+
+	/* Find a corresponding function (name, baseline and baseaddr) */
+	if (die_find_real_subprogram(&cudie, (Dwarf_Addr)addr, &spdie)) {
+		/* Get function entry information */
+		tmp = dwarf_diename(&spdie);
+		if (!tmp ||
+		    dwarf_entrypc(&spdie, &baseaddr) != 0 ||
+		    dwarf_decl_line(&spdie, &baseline) != 0)
+			goto post;
+		func = tmp;
+
+		if (addr == (unsigned long)baseaddr)
+			/* Function entry - Relative line number is 0 */
+			lineno = baseline;
+		else if (die_find_inlinefunc(&spdie, (Dwarf_Addr)addr,
+					     &indie)) {
+			if (dwarf_entrypc(&indie, &_addr) == 0 &&
+			    _addr == addr)
+				/*
+				 * addr is at an inline function entry.
+				 * In this case, lineno should be the call-site
+				 * line number.
+				 */
+				lineno = die_get_call_lineno(&indie);
+			else {
+				/*
+				 * addr is in an inline function body.
+				 * Since lineno points one of the lines
+				 * of the inline function, baseline should
+				 * be the entry line of the inline function.
+				 */
+				tmp = dwarf_diename(&indie);
+				if (tmp &&
+				    dwarf_decl_line(&spdie, &baseline) == 0)
+					func = tmp;
+			}
+		}
+	}
+
+post:
+	/* Make a relative line number or an offset */
+	if (lineno)
+		ppt->line = lineno - baseline;
+	else if (func)
+		ppt->offset = addr - (unsigned long)baseaddr;
+
+	/* Duplicate strings */
+	if (func) {
+		ppt->function = strdup(func);
+		if (ppt->function == NULL) {
+			ret = -ENOMEM;
+			goto end;
+		}
+	}
+	if (fname) {
+		ppt->file = strdup(fname);
+		if (ppt->file == NULL) {
+			if (ppt->function) {
+				free(ppt->function);
+				ppt->function = NULL;
+			}
+			ret = -ENOMEM;
+			goto end;
+		}
+	}
+end:
+	if (dwfl)
+		dwfl_end(dwfl);
+	if (ret == 0 && (fname || func))
+		ret = 1;	/* Found a point */
+	return ret;
+}
+
+/* Add a line and store the src path */
+static int line_range_add_line(const char *src, unsigned int lineno,
+			       struct line_range *lr)
+{
+	/* Copy source path */
+	if (!lr->path) {
+		lr->path = strdup(src);
+		if (lr->path == NULL)
+			return -ENOMEM;
+	}
+	return line_list__add_line(&lr->line_list, lineno);
+}
+
+static int line_range_walk_cb(const char *fname, int lineno,
+			      Dwarf_Addr addr __used,
+			      void *data)
+{
+	struct line_finder *lf = data;
+
+	if ((strtailcmp(fname, lf->fname) != 0) ||
+	    (lf->lno_s > lineno || lf->lno_e < lineno))
+		return 0;
+
+	if (line_range_add_line(fname, lineno, lf->lr) < 0)
+		return -EINVAL;
+
+	return 0;
+}
+
+/* Find line range from its line number */
+static int find_line_range_by_line(Dwarf_Die *sp_die, struct line_finder *lf)
+{
+	int ret;
+
+	ret = die_walk_lines(sp_die ?: &lf->cu_die, line_range_walk_cb, lf);
+
+	/* Update status */
+	if (ret >= 0)
+		if (!list_empty(&lf->lr->line_list))
+			ret = lf->found = 1;
+		else
+			ret = 0;	/* Lines are not found */
+	else {
+		free(lf->lr->path);
+		lf->lr->path = NULL;
+	}
+	return ret;
+}
+
+static int line_range_inline_cb(Dwarf_Die *in_die, void *data)
+{
+	struct dwarf_callback_param *param = data;
+
+	param->retval = find_line_range_by_line(in_die, param->data);
+	return DWARF_CB_ABORT;	/* No need to find other instances */
+}
+
+/* Search function from function name */
+static int line_range_search_cb(Dwarf_Die *sp_die, void *data)
+{
+	struct dwarf_callback_param *param = data;
+	struct line_finder *lf = param->data;
+	struct line_range *lr = lf->lr;
+
+	/* Check declared file */
+	if (lr->file && strtailcmp(lr->file, dwarf_decl_file(sp_die)))
+		return DWARF_CB_OK;
+
+	if (dwarf_tag(sp_die) == DW_TAG_subprogram &&
+	    die_compare_name(sp_die, lr->function)) {
+		lf->fname = dwarf_decl_file(sp_die);
+		dwarf_decl_line(sp_die, &lr->offset);
+		pr_debug("fname: %s, lineno:%d\n", lf->fname, lr->offset);
+		lf->lno_s = lr->offset + lr->start;
+		if (lf->lno_s < 0)	/* Overflow */
+			lf->lno_s = INT_MAX;
+		lf->lno_e = lr->offset + lr->end;
+		if (lf->lno_e < 0)	/* Overflow */
+			lf->lno_e = INT_MAX;
+		pr_debug("New line range: %d to %d\n", lf->lno_s, lf->lno_e);
+		lr->start = lf->lno_s;
+		lr->end = lf->lno_e;
+		if (dwarf_func_inline(sp_die)) {
+			struct dwarf_callback_param _param;
+			_param.data = (void *)lf;
+			_param.retval = 0;
+			dwarf_func_inline_instances(sp_die,
+						    line_range_inline_cb,
+						    &_param);
+			param->retval = _param.retval;
+		} else
+			param->retval = find_line_range_by_line(sp_die, lf);
+		return DWARF_CB_ABORT;
+	}
+	return DWARF_CB_OK;
+}
+
+static int find_line_range_by_func(struct line_finder *lf)
+{
+	struct dwarf_callback_param param = {.data = (void *)lf, .retval = 0};
+	dwarf_getfuncs(&lf->cu_die, line_range_search_cb, &param, 0);
+	return param.retval;
+}
+
+int find_line_range(int fd, struct line_range *lr)
+{
+	struct line_finder lf = {.lr = lr, .found = 0};
+	int ret = 0;
+	Dwarf_Off off = 0, noff;
+	size_t cuhl;
+	Dwarf_Die *diep;
+	Dwarf *dbg = NULL;
+	Dwfl *dwfl;
+	Dwarf_Addr bias;	/* Currently ignored */
+	const char *comp_dir;
+
+	dbg = dwfl_init_offline_dwarf(fd, &dwfl, &bias);
+	if (!dbg) {
+		pr_warning("No debug information found in the vmlinux - "
+			"please rebuild with CONFIG_DEBUG_INFO=y.\n");
+		close(fd);	/* Without dwfl_end(), fd isn't closed. */
+		return -EBADF;
+	}
+
+	/* Fastpath: lookup by function name from .debug_pubnames section */
+	if (lr->function) {
+		struct pubname_callback_param pubname_param = {
+			.function = lr->function, .file = lr->file,
+			.cu_die = &lf.cu_die, .sp_die = &lf.sp_die, .found = 0};
+		struct dwarf_callback_param line_range_param = {
+			.data = (void *)&lf, .retval = 0};
+
+		dwarf_getpubnames(dbg, pubname_search_cb, &pubname_param, 0);
+		if (pubname_param.found) {
+			line_range_search_cb(&lf.sp_die, &line_range_param);
+			if (lf.found)
+				goto found;
+		}
+	}
+
+	/* Loop on CUs (Compilation Unit) */
+	while (!lf.found && ret >= 0) {
+		if (dwarf_nextcu(dbg, off, &noff, &cuhl, NULL, NULL, NULL) != 0)
+			break;
+
+		/* Get the DIE(Debugging Information Entry) of this CU */
+		diep = dwarf_offdie(dbg, off + cuhl, &lf.cu_die);
+		if (!diep)
+			continue;
+
+		/* Check if target file is included. */
+		if (lr->file)
+			lf.fname = cu_find_realpath(&lf.cu_die, lr->file);
+		else
+			lf.fname = 0;
+
+		if (!lr->file || lf.fname) {
+			if (lr->function)
+				ret = find_line_range_by_func(&lf);
+			else {
+				lf.lno_s = lr->start;
+				lf.lno_e = lr->end;
+				ret = find_line_range_by_line(NULL, &lf);
+			}
+		}
+		off = noff;
+	}
+
+found:
+	/* Store comp_dir */
+	if (lf.found) {
+		comp_dir = cu_get_comp_dir(&lf.cu_die);
+		if (comp_dir) {
+			lr->comp_dir = strdup(comp_dir);
+			if (!lr->comp_dir)
+				ret = -ENOMEM;
+		}
+	}
+
+	pr_debug("path: %s\n", lr->path);
+	dwfl_end(dwfl);
+	return (ret < 0) ? ret : lf.found;
+}
+
diff --git a/tools/perf/util/probe-finder.h b/tools/perf/util/probe-finder.h
new file mode 100644
index 00000000..605730a3
--- /dev/null
+++ b/tools/perf/util/probe-finder.h
@@ -0,0 +1,93 @@
+#ifndef _PROBE_FINDER_H
+#define _PROBE_FINDER_H
+
+#include <stdbool.h>
+#include "util.h"
+#include "probe-event.h"
+
+#define MAX_PATH_LEN		 256
+#define MAX_PROBE_BUFFER	1024
+#define MAX_PROBES		 128
+
+static inline int is_c_varname(const char *name)
+{
+	/* TODO */
+	return isalpha(name[0]) || name[0] == '_';
+}
+
+#ifdef DWARF_SUPPORT
+/* Find probe_trace_events specified by perf_probe_event from debuginfo */
+extern int find_probe_trace_events(int fd, struct perf_probe_event *pev,
+				    struct probe_trace_event **tevs,
+				    int max_tevs);
+
+/* Find a perf_probe_point from debuginfo */
+extern int find_perf_probe_point(unsigned long addr,
+				 struct perf_probe_point *ppt);
+
+/* Find a line range */
+extern int find_line_range(int fd, struct line_range *lr);
+
+/* Find available variables */
+extern int find_available_vars_at(int fd, struct perf_probe_event *pev,
+				  struct variable_list **vls, int max_points,
+				  bool externs);
+
+#include <dwarf.h>
+#include <elfutils/libdw.h>
+#include <elfutils/libdwfl.h>
+#include <elfutils/version.h>
+
+struct probe_finder {
+	struct perf_probe_event	*pev;		/* Target probe event */
+
+	/* Callback when a probe point is found */
+	int (*callback)(Dwarf_Die *sp_die, struct probe_finder *pf);
+
+	/* For function searching */
+	int			lno;		/* Line number */
+	Dwarf_Addr		addr;		/* Address */
+	const char		*fname;		/* Real file name */
+	Dwarf_Die		cu_die;		/* Current CU */
+	Dwarf_Die		sp_die;
+	struct list_head	lcache;		/* Line cache for lazy match */
+
+	/* For variable searching */
+#if _ELFUTILS_PREREQ(0, 142)
+	Dwarf_CFI		*cfi;		/* Call Frame Information */
+#endif
+	Dwarf_Op		*fb_ops;	/* Frame base attribute */
+	struct perf_probe_arg	*pvar;		/* Current target variable */
+	struct probe_trace_arg	*tvar;		/* Current result variable */
+};
+
+struct trace_event_finder {
+	struct probe_finder	pf;
+	struct probe_trace_event *tevs;		/* Found trace events */
+	int			ntevs;		/* Number of trace events */
+	int			max_tevs;	/* Max number of trace events */
+};
+
+struct available_var_finder {
+	struct probe_finder	pf;
+	struct variable_list	*vls;		/* Found variable lists */
+	int			nvls;		/* Number of variable lists */
+	int			max_vls;	/* Max no. of variable lists */
+	bool			externs;	/* Find external vars too */
+	bool			child;		/* Search child scopes */
+};
+
+struct line_finder {
+	struct line_range	*lr;		/* Target line range */
+
+	const char		*fname;		/* File name */
+	int			lno_s;		/* Start line number */
+	int			lno_e;		/* End line number */
+	Dwarf_Die		cu_die;		/* Current CU */
+	Dwarf_Die		sp_die;
+	int			found;
+};
+
+#endif /* DWARF_SUPPORT */
+
+#endif /*_PROBE_FINDER_H */
diff --git a/tools/perf/util/pstack.c b/tools/perf/util/pstack.c
new file mode 100644
index 00000000..13d36faf
--- /dev/null
+++ b/tools/perf/util/pstack.c
@@ -0,0 +1,75 @@
+/*
+ * Simple pointer stack
+ *
+ * (c) 2010 Arnaldo Carvalho de Melo <acme@redhat.com>
+ */
+
+#include "util.h"
+#include "pstack.h"
+#include <linux/kernel.h>
+#include <stdlib.h>
+
+struct pstack {
+	unsigned short	top;
+	unsigned short	max_nr_entries;
+	void		*entries[0];
+};
+
+struct pstack *pstack__new(unsigned short max_nr_entries)
+{
+	struct pstack *self = zalloc((sizeof(*self) +
+				     max_nr_entries * sizeof(void *)));
+	if (self != NULL)
+		self->max_nr_entries = max_nr_entries;
+	return self;
+}
+
+void pstack__delete(struct pstack *self)
+{
+	free(self);
+}
+
+bool pstack__empty(const struct pstack *self)
+{
+	return self->top == 0;
+}
+
+void pstack__remove(struct pstack *self, void *key)
+{
+	unsigned short i = self->top, last_index = self->top - 1;
+
+	while (i-- != 0) {
+		if (self->entries[i] == key) {
+			if (i < last_index)
+				memmove(self->entries + i,
+					self->entries + i + 1,
+					(last_index - i) * sizeof(void *));
+			--self->top;
+			return;
+		}
+	}
+	pr_err("%s: %p not on the pstack!\n", __func__, key);
+}
+
+void pstack__push(struct pstack *self, void *key)
+{
+	if (self->top == self->max_nr_entries) {
+		pr_err("%s: top=%d, overflow!\n", __func__, self->top);
+		return;
+	}
+	self->entries[self->top++] = key;
+}
+
+void *pstack__pop(struct pstack *self)
+{
+	void *ret;
+
+	if (self->top == 0) {
+		pr_err("%s: underflow!\n", __func__);
+		return NULL;
+	}
+
+	ret = self->entries[--self->top];
+	self->entries[self->top] = NULL;
+	return ret;
+}
diff --git a/tools/perf/util/pstack.h b/tools/perf/util/pstack.h
new file mode 100644
index 00000000..4cedea59
--- /dev/null
+++ b/tools/perf/util/pstack.h
@@ -0,0 +1,14 @@
+#ifndef _PERF_PSTACK_
+#define _PERF_PSTACK_
+
+#include <stdbool.h>
+
+struct pstack;
+struct pstack *pstack__new(unsigned short max_nr_entries);
+void pstack__delete(struct pstack *self);
+bool pstack__empty(const struct pstack *self);
+void pstack__remove(struct pstack *self, void *key);
+void pstack__push(struct pstack *self, void *key);
+void *pstack__pop(struct pstack *self);
+
+#endif /* _PERF_PSTACK_ */
diff --git a/tools/perf/util/python.c b/tools/perf/util/python.c
new file mode 100644
index 00000000..a9ac0504
--- /dev/null
+++ b/tools/perf/util/python.c
@@ -0,0 +1,905 @@
+#include <Python.h>
+#include <structmember.h>
+#include <inttypes.h>
+#include <poll.h>
+#include "evlist.h"
+#include "evsel.h"
+#include "event.h"
+#include "cpumap.h"
+#include "thread_map.h"
+
+/* Define PyVarObject_HEAD_INIT for python 2.5 */
+#ifndef PyVarObject_HEAD_INIT
+# define PyVarObject_HEAD_INIT(type, size) PyObject_HEAD_INIT(type) size,
+#endif
+
+struct throttle_event {
+	struct perf_event_header header;
+	u64			 time;
+	u64			 id;
+	u64			 stream_id;
+};
+
+PyMODINIT_FUNC initperf(void);
+
+#define member_def(type, member, ptype, help) \
+	{ #member, ptype, \
+	  offsetof(struct pyrf_event, event) + offsetof(struct type, member), \
+	  0, help }
+
+#define sample_member_def(name, member, ptype, help) \
+	{ #name, ptype, \
+	  offsetof(struct pyrf_event, sample) + offsetof(struct perf_sample, member), \
+	  0, help }
+
+struct pyrf_event {
+	PyObject_HEAD
+	struct perf_sample sample;
+	union perf_event   event;
+};
+
+#define sample_members \
+	sample_member_def(sample_ip, ip, T_ULONGLONG, "event type"),			 \
+	sample_member_def(sample_pid, pid, T_INT, "event pid"),			 \
+	sample_member_def(sample_tid, tid, T_INT, "event tid"),			 \
+	sample_member_def(sample_time, time, T_ULONGLONG, "event timestamp"),		 \
+	sample_member_def(sample_addr, addr, T_ULONGLONG, "event addr"),		 \
+	sample_member_def(sample_id, id, T_ULONGLONG, "event id"),			 \
+	sample_member_def(sample_stream_id, stream_id, T_ULONGLONG, "event stream id"), \
+	sample_member_def(sample_period, period, T_ULONGLONG, "event period"),		 \
+	sample_member_def(sample_cpu, cpu, T_UINT, "event cpu"),
+
+static char pyrf_mmap_event__doc[] = PyDoc_STR("perf mmap event object.");
+
+static PyMemberDef pyrf_mmap_event__members[] = {
+	sample_members
+	member_def(perf_event_header, type, T_UINT, "event type"),
+	member_def(mmap_event, pid, T_UINT, "event pid"),
+	member_def(mmap_event, tid, T_UINT, "event tid"),
+	member_def(mmap_event, start, T_ULONGLONG, "start of the map"),
+	member_def(mmap_event, len, T_ULONGLONG, "map length"),
+	member_def(mmap_event, pgoff, T_ULONGLONG, "page offset"),
+	member_def(mmap_event, filename, T_STRING_INPLACE, "backing store"),
+	{ .name = NULL, },
+};
+
+static PyObject *pyrf_mmap_event__repr(struct pyrf_event *pevent)
+{
+	PyObject *ret;
+	char *s;
+
+	if (asprintf(&s, "{ type: mmap, pid: %u, tid: %u, start: %#" PRIx64 ", "
+			 "length: %#" PRIx64 ", offset: %#" PRIx64 ", "
+			 "filename: %s }",
+		     pevent->event.mmap.pid, pevent->event.mmap.tid,
+		     pevent->event.mmap.start, pevent->event.mmap.len,
+		     pevent->event.mmap.pgoff, pevent->event.mmap.filename) < 0) {
+		ret = PyErr_NoMemory();
+	} else {
+		ret = PyString_FromString(s);
+		free(s);
+	}
+	return ret;
+}
+
+static PyTypeObject pyrf_mmap_event__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.mmap_event",
+	.tp_basicsize	= sizeof(struct pyrf_event),
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_mmap_event__doc,
+	.tp_members	= pyrf_mmap_event__members,
+	.tp_repr	= (reprfunc)pyrf_mmap_event__repr,
+};
+
+static char pyrf_task_event__doc[] = PyDoc_STR("perf task (fork/exit) event object.");
+
+static PyMemberDef pyrf_task_event__members[] = {
+	sample_members
+	member_def(perf_event_header, type, T_UINT, "event type"),
+	member_def(fork_event, pid, T_UINT, "event pid"),
+	member_def(fork_event, ppid, T_UINT, "event ppid"),
+	member_def(fork_event, tid, T_UINT, "event tid"),
+	member_def(fork_event, ptid, T_UINT, "event ptid"),
+	member_def(fork_event, time, T_ULONGLONG, "timestamp"),
+	{ .name = NULL, },
+};
+
+static PyObject *pyrf_task_event__repr(struct pyrf_event *pevent)
+{
+	return PyString_FromFormat("{ type: %s, pid: %u, ppid: %u, tid: %u, "
+				   "ptid: %u, time: %" PRIu64 "}",
+				   pevent->event.header.type == PERF_RECORD_FORK ? "fork" : "exit",
+				   pevent->event.fork.pid,
+				   pevent->event.fork.ppid,
+				   pevent->event.fork.tid,
+				   pevent->event.fork.ptid,
+				   pevent->event.fork.time);
+}
+
+static PyTypeObject pyrf_task_event__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.task_event",
+	.tp_basicsize	= sizeof(struct pyrf_event),
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_task_event__doc,
+	.tp_members	= pyrf_task_event__members,
+	.tp_repr	= (reprfunc)pyrf_task_event__repr,
+};
+
+static char pyrf_comm_event__doc[] = PyDoc_STR("perf comm event object.");
+
+static PyMemberDef pyrf_comm_event__members[] = {
+	sample_members
+	member_def(perf_event_header, type, T_UINT, "event type"),
+	member_def(comm_event, pid, T_UINT, "event pid"),
+	member_def(comm_event, tid, T_UINT, "event tid"),
+	member_def(comm_event, comm, T_STRING_INPLACE, "process name"),
+	{ .name = NULL, },
+};
+
+static PyObject *pyrf_comm_event__repr(struct pyrf_event *pevent)
+{
+	return PyString_FromFormat("{ type: comm, pid: %u, tid: %u, comm: %s }",
+				   pevent->event.comm.pid,
+				   pevent->event.comm.tid,
+				   pevent->event.comm.comm);
+}
+
+static PyTypeObject pyrf_comm_event__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.comm_event",
+	.tp_basicsize	= sizeof(struct pyrf_event),
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_comm_event__doc,
+	.tp_members	= pyrf_comm_event__members,
+	.tp_repr	= (reprfunc)pyrf_comm_event__repr,
+};
+
+static char pyrf_throttle_event__doc[] = PyDoc_STR("perf throttle event object.");
+
+static PyMemberDef pyrf_throttle_event__members[] = {
+	sample_members
+	member_def(perf_event_header, type, T_UINT, "event type"),
+	member_def(throttle_event, time, T_ULONGLONG, "timestamp"),
+	member_def(throttle_event, id, T_ULONGLONG, "event id"),
+	member_def(throttle_event, stream_id, T_ULONGLONG, "event stream id"),
+	{ .name = NULL, },
+};
+
+static PyObject *pyrf_throttle_event__repr(struct pyrf_event *pevent)
+{
+	struct throttle_event *te = (struct throttle_event *)(&pevent->event.header + 1);
+
+	return PyString_FromFormat("{ type: %sthrottle, time: %" PRIu64 ", id: %" PRIu64
+				   ", stream_id: %" PRIu64 " }",
+				   pevent->event.header.type == PERF_RECORD_THROTTLE ? "" : "un",
+				   te->time, te->id, te->stream_id);
+}
+
+static PyTypeObject pyrf_throttle_event__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.throttle_event",
+	.tp_basicsize	= sizeof(struct pyrf_event),
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_throttle_event__doc,
+	.tp_members	= pyrf_throttle_event__members,
+	.tp_repr	= (reprfunc)pyrf_throttle_event__repr,
+};
+
+static int pyrf_event__setup_types(void)
+{
+	int err;
+	pyrf_mmap_event__type.tp_new =
+	pyrf_task_event__type.tp_new =
+	pyrf_comm_event__type.tp_new =
+	pyrf_throttle_event__type.tp_new = PyType_GenericNew;
+	err = PyType_Ready(&pyrf_mmap_event__type);
+	if (err < 0)
+		goto out;
+	err = PyType_Ready(&pyrf_task_event__type);
+	if (err < 0)
+		goto out;
+	err = PyType_Ready(&pyrf_comm_event__type);
+	if (err < 0)
+		goto out;
+	err = PyType_Ready(&pyrf_throttle_event__type);
+	if (err < 0)
+		goto out;
+out:
+	return err;
+}
+
+static PyTypeObject *pyrf_event__type[] = {
+	[PERF_RECORD_MMAP]	 = &pyrf_mmap_event__type,
+	[PERF_RECORD_LOST]	 = &pyrf_mmap_event__type,
+	[PERF_RECORD_COMM]	 = &pyrf_comm_event__type,
+	[PERF_RECORD_EXIT]	 = &pyrf_task_event__type,
+	[PERF_RECORD_THROTTLE]	 = &pyrf_throttle_event__type,
+	[PERF_RECORD_UNTHROTTLE] = &pyrf_throttle_event__type,
+	[PERF_RECORD_FORK]	 = &pyrf_task_event__type,
+	[PERF_RECORD_READ]	 = &pyrf_mmap_event__type,
+	[PERF_RECORD_SAMPLE]	 = &pyrf_mmap_event__type,
+};
+
+static PyObject *pyrf_event__new(union perf_event *event)
+{
+	struct pyrf_event *pevent;
+	PyTypeObject *ptype;
+
+	if (event->header.type < PERF_RECORD_MMAP ||
+	    event->header.type > PERF_RECORD_SAMPLE)
+		return NULL;
+
+	ptype = pyrf_event__type[event->header.type];
+	pevent = PyObject_New(struct pyrf_event, ptype);
+	if (pevent != NULL)
+		memcpy(&pevent->event, event, event->header.size);
+	return (PyObject *)pevent;
+}
+
+struct pyrf_cpu_map {
+	PyObject_HEAD
+
+	struct cpu_map *cpus;
+};
+
+static int pyrf_cpu_map__init(struct pyrf_cpu_map *pcpus,
+			      PyObject *args, PyObject *kwargs)
+{
+	static char *kwlist[] = { "cpustr", NULL, NULL, };
+	char *cpustr = NULL;
+
+	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|s",
+					 kwlist, &cpustr))
+		return -1;
+
+	pcpus->cpus = cpu_map__new(cpustr);
+	if (pcpus->cpus == NULL)
+		return -1;
+	return 0;
+}
+
+static void pyrf_cpu_map__delete(struct pyrf_cpu_map *pcpus)
+{
+	cpu_map__delete(pcpus->cpus);
+	pcpus->ob_type->tp_free((PyObject*)pcpus);
+}
+
+static Py_ssize_t pyrf_cpu_map__length(PyObject *obj)
+{
+	struct pyrf_cpu_map *pcpus = (void *)obj;
+
+	return pcpus->cpus->nr;
+}
+
+static PyObject *pyrf_cpu_map__item(PyObject *obj, Py_ssize_t i)
+{
+	struct pyrf_cpu_map *pcpus = (void *)obj;
+
+	if (i >= pcpus->cpus->nr)
+		return NULL;
+
+	return Py_BuildValue("i", pcpus->cpus->map[i]);
+}
+
+static PySequenceMethods pyrf_cpu_map__sequence_methods = {
+	.sq_length = pyrf_cpu_map__length,
+	.sq_item   = pyrf_cpu_map__item,
+};
+
+static char pyrf_cpu_map__doc[] = PyDoc_STR("cpu map object.");
+
+static PyTypeObject pyrf_cpu_map__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.cpu_map",
+	.tp_basicsize	= sizeof(struct pyrf_cpu_map),
+	.tp_dealloc	= (destructor)pyrf_cpu_map__delete,
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_cpu_map__doc,
+	.tp_as_sequence	= &pyrf_cpu_map__sequence_methods,
+	.tp_init	= (initproc)pyrf_cpu_map__init,
+};
+
+static int pyrf_cpu_map__setup_types(void)
+{
+	pyrf_cpu_map__type.tp_new = PyType_GenericNew;
+	return PyType_Ready(&pyrf_cpu_map__type);
+}
+
+struct pyrf_thread_map {
+	PyObject_HEAD
+
+	struct thread_map *threads;
+};
+
+static int pyrf_thread_map__init(struct pyrf_thread_map *pthreads,
+				 PyObject *args, PyObject *kwargs)
+{
+	static char *kwlist[] = { "pid", "tid", NULL, NULL, };
+	int pid = -1, tid = -1;
+
+	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|ii",
+					 kwlist, &pid, &tid))
+		return -1;
+
+	pthreads->threads = thread_map__new(pid, tid);
+	if (pthreads->threads == NULL)
+		return -1;
+	return 0;
+}
+
+static void pyrf_thread_map__delete(struct pyrf_thread_map *pthreads)
+{
+	thread_map__delete(pthreads->threads);
+	pthreads->ob_type->tp_free((PyObject*)pthreads);
+}
+
+static Py_ssize_t pyrf_thread_map__length(PyObject *obj)
+{
+	struct pyrf_thread_map *pthreads = (void *)obj;
+
+	return pthreads->threads->nr;
+}
+
+static PyObject *pyrf_thread_map__item(PyObject *obj, Py_ssize_t i)
+{
+	struct pyrf_thread_map *pthreads = (void *)obj;
+
+	if (i >= pthreads->threads->nr)
+		return NULL;
+
+	return Py_BuildValue("i", pthreads->threads->map[i]);
+}
+
+static PySequenceMethods pyrf_thread_map__sequence_methods = {
+	.sq_length = pyrf_thread_map__length,
+	.sq_item   = pyrf_thread_map__item,
+};
+
+static char pyrf_thread_map__doc[] = PyDoc_STR("thread map object.");
+
+static PyTypeObject pyrf_thread_map__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.thread_map",
+	.tp_basicsize	= sizeof(struct pyrf_thread_map),
+	.tp_dealloc	= (destructor)pyrf_thread_map__delete,
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_thread_map__doc,
+	.tp_as_sequence	= &pyrf_thread_map__sequence_methods,
+	.tp_init	= (initproc)pyrf_thread_map__init,
+};
+
+static int pyrf_thread_map__setup_types(void)
+{
+	pyrf_thread_map__type.tp_new = PyType_GenericNew;
+	return PyType_Ready(&pyrf_thread_map__type);
+}
+
+struct pyrf_evsel {
+	PyObject_HEAD
+
+	struct perf_evsel evsel;
+};
+
+static int pyrf_evsel__init(struct pyrf_evsel *pevsel,
+			    PyObject *args, PyObject *kwargs)
+{
+	struct perf_event_attr attr = {
+		.type = PERF_TYPE_HARDWARE,
+		.config = PERF_COUNT_HW_CPU_CYCLES,
+		.sample_type = PERF_SAMPLE_PERIOD | PERF_SAMPLE_TID,
+	};
+	static char *kwlist[] = {
+		"type",
+		"config",
+		"sample_freq",
+		"sample_period",
+		"sample_type",
+		"read_format",
+		"disabled",
+		"inherit",
+		"pinned",
+		"exclusive",
+		"exclude_user",
+		"exclude_kernel",
+		"exclude_hv",
+		"exclude_idle",
+		"mmap",
+		"comm",
+		"freq",
+		"inherit_stat",
+		"enable_on_exec",
+		"task",
+		"watermark",
+		"precise_ip",
+		"mmap_data",
+		"sample_id_all",
+		"wakeup_events",
+		"bp_type",
+		"bp_addr",
+		"bp_len", NULL, NULL, };
+	u64 sample_period = 0;
+	u32 disabled = 0,
+	    inherit = 0,
+	    pinned = 0,
+	    exclusive = 0,
+	    exclude_user = 0,
+	    exclude_kernel = 0,
+	    exclude_hv = 0,
+	    exclude_idle = 0,
+	    mmap = 0,
+	    comm = 0,
+	    freq = 1,
+	    inherit_stat = 0,
+	    enable_on_exec = 0,
+	    task = 0,
+	    watermark = 0,
+	    precise_ip = 0,
+	    mmap_data = 0,
+	    sample_id_all = 1;
+	int idx = 0;
+
+	if (!PyArg_ParseTupleAndKeywords(args, kwargs,
+					 "|iKiKKiiiiiiiiiiiiiiiiiiiiiKK", kwlist,
+					 &attr.type, &attr.config, &attr.sample_freq,
+					 &sample_period, &attr.sample_type,
+					 &attr.read_format, &disabled, &inherit,
+					 &pinned, &exclusive, &exclude_user,
+					 &exclude_kernel, &exclude_hv, &exclude_idle,
+					 &mmap, &comm, &freq, &inherit_stat,
+					 &enable_on_exec, &task, &watermark,
+					 &precise_ip, &mmap_data, &sample_id_all,
+					 &attr.wakeup_events, &attr.bp_type,
+					 &attr.bp_addr, &attr.bp_len, &idx))
+		return -1;
+
+	/* union... */
+	if (sample_period != 0) {
+		if (attr.sample_freq != 0)
+			return -1; /* FIXME: throw right exception */
+		attr.sample_period = sample_period;
+	}
+
+	/* Bitfields */
+	attr.disabled	    = disabled;
+	attr.inherit	    = inherit;
+	attr.pinned	    = pinned;
+	attr.exclusive	    = exclusive;
+	attr.exclude_user   = exclude_user;
+	attr.exclude_kernel = exclude_kernel;
+	attr.exclude_hv	    = exclude_hv;
+	attr.exclude_idle   = exclude_idle;
+	attr.mmap	    = mmap;
+	attr.comm	    = comm;
+	attr.freq	    = freq;
+	attr.inherit_stat   = inherit_stat;
+	attr.enable_on_exec = enable_on_exec;
+	attr.task	    = task;
+	attr.watermark	    = watermark;
+	attr.precise_ip	    = precise_ip;
+	attr.mmap_data	    = mmap_data;
+	attr.sample_id_all  = sample_id_all;
+
+	perf_evsel__init(&pevsel->evsel, &attr, idx);
+	return 0;
+}
+
+static void pyrf_evsel__delete(struct pyrf_evsel *pevsel)
+{
+	perf_evsel__exit(&pevsel->evsel);
+	pevsel->ob_type->tp_free((PyObject*)pevsel);
+}
+
+static PyObject *pyrf_evsel__open(struct pyrf_evsel *pevsel,
+				  PyObject *args, PyObject *kwargs)
+{
+	struct perf_evsel *evsel = &pevsel->evsel;
+	struct cpu_map *cpus = NULL;
+	struct thread_map *threads = NULL;
+	PyObject *pcpus = NULL, *pthreads = NULL;
+	int group = 0, inherit = 0;
+	static char *kwlist[] = {"cpus", "threads", "group", "inherit", NULL, NULL};
+
+	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|OOii", kwlist,
+					 &pcpus, &pthreads, &group, &inherit))
+		return NULL;
+
+	if (pthreads != NULL)
+		threads = ((struct pyrf_thread_map *)pthreads)->threads;
+
+	if (pcpus != NULL)
+		cpus = ((struct pyrf_cpu_map *)pcpus)->cpus;
+
+	evsel->attr.inherit = inherit;
+	if (perf_evsel__open(evsel, cpus, threads, group) < 0) {
+		PyErr_SetFromErrno(PyExc_OSError);
+		return NULL;
+	}
+
+	Py_INCREF(Py_None);
+	return Py_None;
+}
+
+static PyMethodDef pyrf_evsel__methods[] = {
+	{
+		.ml_name  = "open",
+		.ml_meth  = (PyCFunction)pyrf_evsel__open,
+		.ml_flags = METH_VARARGS | METH_KEYWORDS,
+		.ml_doc	  = PyDoc_STR("open the event selector file descriptor table.")
+	},
+	{ .ml_name = NULL, }
+};
+
+static char pyrf_evsel__doc[] = PyDoc_STR("perf event selector list object.");
+
+static PyTypeObject pyrf_evsel__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.evsel",
+	.tp_basicsize	= sizeof(struct pyrf_evsel),
+	.tp_dealloc	= (destructor)pyrf_evsel__delete,
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_evsel__doc,
+	.tp_methods	= pyrf_evsel__methods,
+	.tp_init	= (initproc)pyrf_evsel__init,
+};
+
+static int pyrf_evsel__setup_types(void)
+{
+	pyrf_evsel__type.tp_new = PyType_GenericNew;
+	return PyType_Ready(&pyrf_evsel__type);
+}
+
+struct pyrf_evlist {
+	PyObject_HEAD
+
+	struct perf_evlist evlist;
+};
+
+static int pyrf_evlist__init(struct pyrf_evlist *pevlist,
+			     PyObject *args, PyObject *kwargs __used)
+{
+	PyObject *pcpus = NULL, *pthreads = NULL;
+	struct cpu_map *cpus;
+	struct thread_map *threads;
+
+	if (!PyArg_ParseTuple(args, "OO", &pcpus, &pthreads))
+		return -1;
+
+	threads = ((struct pyrf_thread_map *)pthreads)->threads;
+	cpus = ((struct pyrf_cpu_map *)pcpus)->cpus;
+	perf_evlist__init(&pevlist->evlist, cpus, threads);
+	return 0;
+}
+
+static void pyrf_evlist__delete(struct pyrf_evlist *pevlist)
+{
+	perf_evlist__exit(&pevlist->evlist);
+	pevlist->ob_type->tp_free((PyObject*)pevlist);
+}
+
+static PyObject *pyrf_evlist__mmap(struct pyrf_evlist *pevlist,
+				   PyObject *args, PyObject *kwargs)
+{
+	struct perf_evlist *evlist = &pevlist->evlist;
+	static char *kwlist[] = {"pages", "overwrite",
+				  NULL, NULL};
+	int pages = 128, overwrite = false;
+
+	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|ii", kwlist,
+					 &pages, &overwrite))
+		return NULL;
+
+	if (perf_evlist__mmap(evlist, pages, overwrite) < 0) {
+		PyErr_SetFromErrno(PyExc_OSError);
+		return NULL;
+	}
+
+	Py_INCREF(Py_None);
+	return Py_None;
+}
+
+static PyObject *pyrf_evlist__poll(struct pyrf_evlist *pevlist,
+				   PyObject *args, PyObject *kwargs)
+{
+	struct perf_evlist *evlist = &pevlist->evlist;
+	static char *kwlist[] = {"timeout", NULL, NULL};
+	int timeout = -1, n;
+
+	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|i", kwlist, &timeout))
+		return NULL;
+
+	n = poll(evlist->pollfd, evlist->nr_fds, timeout);
+	if (n < 0) {
+		PyErr_SetFromErrno(PyExc_OSError);
+		return NULL;
+	}
+
+	return Py_BuildValue("i", n);
+}
+
+static PyObject *pyrf_evlist__get_pollfd(struct pyrf_evlist *pevlist,
+					 PyObject *args __used, PyObject *kwargs __used)
+{
+	struct perf_evlist *evlist = &pevlist->evlist;
+        PyObject *list = PyList_New(0);
+	int i;
+
+	for (i = 0; i < evlist->nr_fds; ++i) {
+		PyObject *file;
+		FILE *fp = fdopen(evlist->pollfd[i].fd, "r");
+
+		if (fp == NULL)
+			goto free_list;
+
+		file = PyFile_FromFile(fp, "perf", "r", NULL);
+		if (file == NULL)
+			goto free_list;
+
+		if (PyList_Append(list, file) != 0) {
+			Py_DECREF(file);
+			goto free_list;
+		}
+			
+		Py_DECREF(file);
+	}
+
+	return list;
+free_list:
+	return PyErr_NoMemory();
+}
+
+
+static PyObject *pyrf_evlist__add(struct pyrf_evlist *pevlist,
+				  PyObject *args, PyObject *kwargs __used)
+{
+	struct perf_evlist *evlist = &pevlist->evlist;
+	PyObject *pevsel;
+	struct perf_evsel *evsel;
+
+	if (!PyArg_ParseTuple(args, "O", &pevsel))
+		return NULL;
+
+	Py_INCREF(pevsel);
+	evsel = &((struct pyrf_evsel *)pevsel)->evsel;
+	evsel->idx = evlist->nr_entries;
+	perf_evlist__add(evlist, evsel);
+
+	return Py_BuildValue("i", evlist->nr_entries);
+}
+
+static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist,
+					  PyObject *args, PyObject *kwargs)
+{
+	struct perf_evlist *evlist = &pevlist->evlist;
+	union perf_event *event;
+	int sample_id_all = 1, cpu;
+	static char *kwlist[] = {"cpu", "sample_id_all", NULL, NULL};
+	int err;
+
+	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "i|i", kwlist,
+					 &cpu, &sample_id_all))
+		return NULL;
+
+	event = perf_evlist__mmap_read(evlist, cpu);
+	if (event != NULL) {
+		struct perf_evsel *first;
+		PyObject *pyevent = pyrf_event__new(event);
+		struct pyrf_event *pevent = (struct pyrf_event *)pyevent;
+
+		if (pyevent == NULL)
+			return PyErr_NoMemory();
+
+		first = list_entry(evlist->entries.next, struct perf_evsel, node);
+		err = perf_event__parse_sample(event, first->attr.sample_type,
+					       perf_evsel__sample_size(first),
+					       sample_id_all, &pevent->sample);
+		if (err)
+			return PyErr_Format(PyExc_OSError,
+					    "perf: can't parse sample, err=%d", err);
+		return pyevent;
+	}
+
+	Py_INCREF(Py_None);
+	return Py_None;
+}
+
+static PyMethodDef pyrf_evlist__methods[] = {
+	{
+		.ml_name  = "mmap",
+		.ml_meth  = (PyCFunction)pyrf_evlist__mmap,
+		.ml_flags = METH_VARARGS | METH_KEYWORDS,
+		.ml_doc	  = PyDoc_STR("mmap the file descriptor table.")
+	},
+	{
+		.ml_name  = "poll",
+		.ml_meth  = (PyCFunction)pyrf_evlist__poll,
+		.ml_flags = METH_VARARGS | METH_KEYWORDS,
+		.ml_doc	  = PyDoc_STR("poll the file descriptor table.")
+	},
+	{
+		.ml_name  = "get_pollfd",
+		.ml_meth  = (PyCFunction)pyrf_evlist__get_pollfd,
+		.ml_flags = METH_VARARGS | METH_KEYWORDS,
+		.ml_doc	  = PyDoc_STR("get the poll file descriptor table.")
+	},
+	{
+		.ml_name  = "add",
+		.ml_meth  = (PyCFunction)pyrf_evlist__add,
+		.ml_flags = METH_VARARGS | METH_KEYWORDS,
+		.ml_doc	  = PyDoc_STR("adds an event selector to the list.")
+	},
+	{
+		.ml_name  = "read_on_cpu",
+		.ml_meth  = (PyCFunction)pyrf_evlist__read_on_cpu,
+		.ml_flags = METH_VARARGS | METH_KEYWORDS,
+		.ml_doc	  = PyDoc_STR("reads an event.")
+	},
+	{ .ml_name = NULL, }
+};
+
+static Py_ssize_t pyrf_evlist__length(PyObject *obj)
+{
+	struct pyrf_evlist *pevlist = (void *)obj;
+
+	return pevlist->evlist.nr_entries;
+}
+
+static PyObject *pyrf_evlist__item(PyObject *obj, Py_ssize_t i)
+{
+	struct pyrf_evlist *pevlist = (void *)obj;
+	struct perf_evsel *pos;
+
+	if (i >= pevlist->evlist.nr_entries)
+		return NULL;
+
+	list_for_each_entry(pos, &pevlist->evlist.entries, node)
+		if (i-- == 0)
+			break;
+
+	return Py_BuildValue("O", container_of(pos, struct pyrf_evsel, evsel));
+}
+
+static PySequenceMethods pyrf_evlist__sequence_methods = {
+	.sq_length = pyrf_evlist__length,
+	.sq_item   = pyrf_evlist__item,
+};
+
+static char pyrf_evlist__doc[] = PyDoc_STR("perf event selector list object.");
+
+static PyTypeObject pyrf_evlist__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.evlist",
+	.tp_basicsize	= sizeof(struct pyrf_evlist),
+	.tp_dealloc	= (destructor)pyrf_evlist__delete,
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_as_sequence	= &pyrf_evlist__sequence_methods,
+	.tp_doc		= pyrf_evlist__doc,
+	.tp_methods	= pyrf_evlist__methods,
+	.tp_init	= (initproc)pyrf_evlist__init,
+};
+
+static int pyrf_evlist__setup_types(void)
+{
+	pyrf_evlist__type.tp_new = PyType_GenericNew;
+	return PyType_Ready(&pyrf_evlist__type);
+}
+
+static struct {
+	const char *name;
+	int	    value;
+} perf__constants[] = {
+	{ "TYPE_HARDWARE",   PERF_TYPE_HARDWARE },
+	{ "TYPE_SOFTWARE",   PERF_TYPE_SOFTWARE },
+	{ "TYPE_TRACEPOINT", PERF_TYPE_TRACEPOINT },
+	{ "TYPE_HW_CACHE",   PERF_TYPE_HW_CACHE },
+	{ "TYPE_RAW",	     PERF_TYPE_RAW },
+	{ "TYPE_BREAKPOINT", PERF_TYPE_BREAKPOINT },
+
+	{ "COUNT_HW_CPU_CYCLES",	  PERF_COUNT_HW_CPU_CYCLES },
+	{ "COUNT_HW_INSTRUCTIONS",	  PERF_COUNT_HW_INSTRUCTIONS },
+	{ "COUNT_HW_CACHE_REFERENCES",	  PERF_COUNT_HW_CACHE_REFERENCES },
+	{ "COUNT_HW_CACHE_MISSES",	  PERF_COUNT_HW_CACHE_MISSES },
+	{ "COUNT_HW_BRANCH_INSTRUCTIONS", PERF_COUNT_HW_BRANCH_INSTRUCTIONS },
+	{ "COUNT_HW_BRANCH_MISSES",	  PERF_COUNT_HW_BRANCH_MISSES },
+	{ "COUNT_HW_BUS_CYCLES",	  PERF_COUNT_HW_BUS_CYCLES },
+	{ "COUNT_HW_CACHE_L1D",		  PERF_COUNT_HW_CACHE_L1D },
+	{ "COUNT_HW_CACHE_L1I",		  PERF_COUNT_HW_CACHE_L1I },
+	{ "COUNT_HW_CACHE_LL",	  	  PERF_COUNT_HW_CACHE_LL },
+	{ "COUNT_HW_CACHE_DTLB",	  PERF_COUNT_HW_CACHE_DTLB },
+	{ "COUNT_HW_CACHE_ITLB",	  PERF_COUNT_HW_CACHE_ITLB },
+	{ "COUNT_HW_CACHE_BPU",		  PERF_COUNT_HW_CACHE_BPU },
+	{ "COUNT_HW_CACHE_OP_READ",	  PERF_COUNT_HW_CACHE_OP_READ },
+	{ "COUNT_HW_CACHE_OP_WRITE",	  PERF_COUNT_HW_CACHE_OP_WRITE },
+	{ "COUNT_HW_CACHE_OP_PREFETCH",	  PERF_COUNT_HW_CACHE_OP_PREFETCH },
+	{ "COUNT_HW_CACHE_RESULT_ACCESS", PERF_COUNT_HW_CACHE_RESULT_ACCESS },
+	{ "COUNT_HW_CACHE_RESULT_MISS",   PERF_COUNT_HW_CACHE_RESULT_MISS },
+
+	{ "COUNT_HW_STALLED_CYCLES_FRONTEND",	  PERF_COUNT_HW_STALLED_CYCLES_FRONTEND },
+	{ "COUNT_HW_STALLED_CYCLES_BACKEND",	  PERF_COUNT_HW_STALLED_CYCLES_BACKEND },
+
+	{ "COUNT_SW_CPU_CLOCK",	       PERF_COUNT_SW_CPU_CLOCK },
+	{ "COUNT_SW_TASK_CLOCK",       PERF_COUNT_SW_TASK_CLOCK },
+	{ "COUNT_SW_PAGE_FAULTS",      PERF_COUNT_SW_PAGE_FAULTS },
+	{ "COUNT_SW_CONTEXT_SWITCHES", PERF_COUNT_SW_CONTEXT_SWITCHES },
+	{ "COUNT_SW_CPU_MIGRATIONS",   PERF_COUNT_SW_CPU_MIGRATIONS },
+	{ "COUNT_SW_PAGE_FAULTS_MIN",  PERF_COUNT_SW_PAGE_FAULTS_MIN },
+	{ "COUNT_SW_PAGE_FAULTS_MAJ",  PERF_COUNT_SW_PAGE_FAULTS_MAJ },
+	{ "COUNT_SW_ALIGNMENT_FAULTS", PERF_COUNT_SW_ALIGNMENT_FAULTS },
+	{ "COUNT_SW_EMULATION_FAULTS", PERF_COUNT_SW_EMULATION_FAULTS },
+
+	{ "SAMPLE_IP",	      PERF_SAMPLE_IP },
+	{ "SAMPLE_TID",	      PERF_SAMPLE_TID },
+	{ "SAMPLE_TIME",      PERF_SAMPLE_TIME },
+	{ "SAMPLE_ADDR",      PERF_SAMPLE_ADDR },
+	{ "SAMPLE_READ",      PERF_SAMPLE_READ },
+	{ "SAMPLE_CALLCHAIN", PERF_SAMPLE_CALLCHAIN },
+	{ "SAMPLE_ID",	      PERF_SAMPLE_ID },
+	{ "SAMPLE_CPU",	      PERF_SAMPLE_CPU },
+	{ "SAMPLE_PERIOD",    PERF_SAMPLE_PERIOD },
+	{ "SAMPLE_STREAM_ID", PERF_SAMPLE_STREAM_ID },
+	{ "SAMPLE_RAW",	      PERF_SAMPLE_RAW },
+
+	{ "FORMAT_TOTAL_TIME_ENABLED", PERF_FORMAT_TOTAL_TIME_ENABLED },
+	{ "FORMAT_TOTAL_TIME_RUNNING", PERF_FORMAT_TOTAL_TIME_RUNNING },
+	{ "FORMAT_ID",		       PERF_FORMAT_ID },
+	{ "FORMAT_GROUP",	       PERF_FORMAT_GROUP },
+
+	{ "RECORD_MMAP",       PERF_RECORD_MMAP },
+	{ "RECORD_LOST",       PERF_RECORD_LOST },
+	{ "RECORD_COMM",       PERF_RECORD_COMM },
+	{ "RECORD_EXIT",       PERF_RECORD_EXIT },
+	{ "RECORD_THROTTLE",   PERF_RECORD_THROTTLE },
+	{ "RECORD_UNTHROTTLE", PERF_RECORD_UNTHROTTLE },
+	{ "RECORD_FORK",       PERF_RECORD_FORK },
+	{ "RECORD_READ",       PERF_RECORD_READ },
+	{ "RECORD_SAMPLE",     PERF_RECORD_SAMPLE },
+	{ .name = NULL, },
+};
+
+static PyMethodDef perf__methods[] = {
+	{ .ml_name = NULL, }
+};
+
+PyMODINIT_FUNC initperf(void)
+{
+	PyObject *obj;
+	int i;
+	PyObject *dict, *module = Py_InitModule("perf", perf__methods);
+
+	if (module == NULL ||
+	    pyrf_event__setup_types() < 0 ||
+	    pyrf_evlist__setup_types() < 0 ||
+	    pyrf_evsel__setup_types() < 0 ||
+	    pyrf_thread_map__setup_types() < 0 ||
+	    pyrf_cpu_map__setup_types() < 0)
+		return;
+
+	Py_INCREF(&pyrf_evlist__type);
+	PyModule_AddObject(module, "evlist", (PyObject*)&pyrf_evlist__type);
+
+	Py_INCREF(&pyrf_evsel__type);
+	PyModule_AddObject(module, "evsel", (PyObject*)&pyrf_evsel__type);
+
+	Py_INCREF(&pyrf_thread_map__type);
+	PyModule_AddObject(module, "thread_map", (PyObject*)&pyrf_thread_map__type);
+
+	Py_INCREF(&pyrf_cpu_map__type);
+	PyModule_AddObject(module, "cpu_map", (PyObject*)&pyrf_cpu_map__type);
+
+	dict = PyModule_GetDict(module);
+	if (dict == NULL)
+		goto error;
+
+	for (i = 0; perf__constants[i].name != NULL; i++) {
+		obj = PyInt_FromLong(perf__constants[i].value);
+		if (obj == NULL)
+			goto error;
+		PyDict_SetItemString(dict, perf__constants[i].name, obj);
+		Py_DECREF(obj);
+	}
+
+error:
+	if (PyErr_Occurred())
+		PyErr_SetString(PyExc_ImportError, "perf: Init failed!");
+}
diff --git a/tools/perf/util/quote.c b/tools/perf/util/quote.c
new file mode 100644
index 00000000..01f03242
--- /dev/null
+++ b/tools/perf/util/quote.c
@@ -0,0 +1,54 @@
+#include "cache.h"
+#include "quote.h"
+
+/* Help to copy the thing properly quoted for the shell safety.
+ * any single quote is replaced with '\'', any exclamation point
+ * is replaced with '\!', and the whole thing is enclosed in a
+ *
+ * E.g.
+ *  original     sq_quote     result
+ *  name     ==> name      ==> 'name'
+ *  a b      ==> a b       ==> 'a b'
+ *  a'b      ==> a'\''b    ==> 'a'\''b'
+ *  a!b      ==> a'\!'b    ==> 'a'\!'b'
+ */
+static inline int need_bs_quote(char c)
+{
+	return (c == '\'' || c == '!');
+}
+
+static void sq_quote_buf(struct strbuf *dst, const char *src)
+{
+	char *to_free = NULL;
+
+	if (dst->buf == src)
+		to_free = strbuf_detach(dst, NULL);
+
+	strbuf_addch(dst, '\'');
+	while (*src) {
+		size_t len = strcspn(src, "'!");
+		strbuf_add(dst, src, len);
+		src += len;
+		while (need_bs_quote(*src)) {
+			strbuf_addstr(dst, "'\\");
+			strbuf_addch(dst, *src++);
+			strbuf_addch(dst, '\'');
+		}
+	}
+	strbuf_addch(dst, '\'');
+	free(to_free);
+}
+
+void sq_quote_argv(struct strbuf *dst, const char** argv, size_t maxlen)
+{
+	int i;
+
+	/* Copy into destination buffer. */
+	strbuf_grow(dst, 255);
+	for (i = 0; argv[i]; ++i) {
+		strbuf_addch(dst, ' ');
+		sq_quote_buf(dst, argv[i]);
+		if (maxlen && dst->len > maxlen)
+			die("Too many or long arguments");
+	}
+}
diff --git a/tools/perf/util/quote.h b/tools/perf/util/quote.h
new file mode 100644
index 00000000..172889ea
--- /dev/null
+++ b/tools/perf/util/quote.h
@@ -0,0 +1,29 @@
+#ifndef __PERF_QUOTE_H
+#define __PERF_QUOTE_H
+
+#include <stddef.h>
+#include <stdio.h>
+
+/* Help to copy the thing properly quoted for the shell safety.
+ * any single quote is replaced with '\'', any exclamation point
+ * is replaced with '\!', and the whole thing is enclosed in a
+ * single quote pair.
+ *
+ * For example, if you are passing the result to system() as an
+ * argument:
+ *
+ * sprintf(cmd, "foobar %s %s", sq_quote(arg0), sq_quote(arg1))
+ *
+ * would be appropriate.  If the system() is going to call ssh to
+ * run the command on the other side:
+ *
+ * sprintf(cmd, "git-diff-tree %s %s", sq_quote(arg0), sq_quote(arg1));
+ * sprintf(rcmd, "ssh %s %s", sq_util/quote.host), sq_quote(cmd));
+ *
+ * Note that the above examples leak memory!  Remember to free result from
+ * sq_quote() in a real application.
+ */
+
+extern void sq_quote_argv(struct strbuf *, const char **argv, size_t maxlen);
+
+#endif /* __PERF_QUOTE_H */
diff --git a/tools/perf/util/run-command.c b/tools/perf/util/run-command.c
new file mode 100644
index 00000000..da8e9b28
--- /dev/null
+++ b/tools/perf/util/run-command.c
@@ -0,0 +1,214 @@
+#include "cache.h"
+#include "run-command.h"
+#include "exec_cmd.h"
+
+static inline void close_pair(int fd[2])
+{
+	close(fd[0]);
+	close(fd[1]);
+}
+
+static inline void dup_devnull(int to)
+{
+	int fd = open("/dev/null", O_RDWR);
+	dup2(fd, to);
+	close(fd);
+}
+
+int start_command(struct child_process *cmd)
+{
+	int need_in, need_out, need_err;
+	int fdin[2], fdout[2], fderr[2];
+
+	/*
+	 * In case of errors we must keep the promise to close FDs
+	 * that have been passed in via ->in and ->out.
+	 */
+
+	need_in = !cmd->no_stdin && cmd->in < 0;
+	if (need_in) {
+		if (pipe(fdin) < 0) {
+			if (cmd->out > 0)
+				close(cmd->out);
+			return -ERR_RUN_COMMAND_PIPE;
+		}
+		cmd->in = fdin[1];
+	}
+
+	need_out = !cmd->no_stdout
+		&& !cmd->stdout_to_stderr
+		&& cmd->out < 0;
+	if (need_out) {
+		if (pipe(fdout) < 0) {
+			if (need_in)
+				close_pair(fdin);
+			else if (cmd->in)
+				close(cmd->in);
+			return -ERR_RUN_COMMAND_PIPE;
+		}
+		cmd->out = fdout[0];
+	}
+
+	need_err = !cmd->no_stderr && cmd->err < 0;
+	if (need_err) {
+		if (pipe(fderr) < 0) {
+			if (need_in)
+				close_pair(fdin);
+			else if (cmd->in)
+				close(cmd->in);
+			if (need_out)
+				close_pair(fdout);
+			else if (cmd->out)
+				close(cmd->out);
+			return -ERR_RUN_COMMAND_PIPE;
+		}
+		cmd->err = fderr[0];
+	}
+
+	fflush(NULL);
+	cmd->pid = fork();
+	if (!cmd->pid) {
+		if (cmd->no_stdin)
+			dup_devnull(0);
+		else if (need_in) {
+			dup2(fdin[0], 0);
+			close_pair(fdin);
+		} else if (cmd->in) {
+			dup2(cmd->in, 0);
+			close(cmd->in);
+		}
+
+		if (cmd->no_stderr)
+			dup_devnull(2);
+		else if (need_err) {
+			dup2(fderr[1], 2);
+			close_pair(fderr);
+		}
+
+		if (cmd->no_stdout)
+			dup_devnull(1);
+		else if (cmd->stdout_to_stderr)
+			dup2(2, 1);
+		else if (need_out) {
+			dup2(fdout[1], 1);
+			close_pair(fdout);
+		} else if (cmd->out > 1) {
+			dup2(cmd->out, 1);
+			close(cmd->out);
+		}
+
+		if (cmd->dir && chdir(cmd->dir))
+			die("exec %s: cd to %s failed (%s)", cmd->argv[0],
+			    cmd->dir, strerror(errno));
+		if (cmd->env) {
+			for (; *cmd->env; cmd->env++) {
+				if (strchr(*cmd->env, '='))
+					putenv((char*)*cmd->env);
+				else
+					unsetenv(*cmd->env);
+			}
+		}
+		if (cmd->preexec_cb)
+			cmd->preexec_cb();
+		if (cmd->perf_cmd) {
+			execv_perf_cmd(cmd->argv);
+		} else {
+			execvp(cmd->argv[0], (char *const*) cmd->argv);
+		}
+		exit(127);
+	}
+
+	if (cmd->pid < 0) {
+		int err = errno;
+		if (need_in)
+			close_pair(fdin);
+		else if (cmd->in)
+			close(cmd->in);
+		if (need_out)
+			close_pair(fdout);
+		else if (cmd->out)
+			close(cmd->out);
+		if (need_err)
+			close_pair(fderr);
+		return err == ENOENT ?
+			-ERR_RUN_COMMAND_EXEC :
+			-ERR_RUN_COMMAND_FORK;
+	}
+
+	if (need_in)
+		close(fdin[0]);
+	else if (cmd->in)
+		close(cmd->in);
+
+	if (need_out)
+		close(fdout[1]);
+	else if (cmd->out)
+		close(cmd->out);
+
+	if (need_err)
+		close(fderr[1]);
+
+	return 0;
+}
+
+static int wait_or_whine(pid_t pid)
+{
+	for (;;) {
+		int status, code;
+		pid_t waiting = waitpid(pid, &status, 0);
+
+		if (waiting < 0) {
+			if (errno == EINTR)
+				continue;
+			error("waitpid failed (%s)", strerror(errno));
+			return -ERR_RUN_COMMAND_WAITPID;
+		}
+		if (waiting != pid)
+			return -ERR_RUN_COMMAND_WAITPID_WRONG_PID;
+		if (WIFSIGNALED(status))
+			return -ERR_RUN_COMMAND_WAITPID_SIGNAL;
+
+		if (!WIFEXITED(status))
+			return -ERR_RUN_COMMAND_WAITPID_NOEXIT;
+		code = WEXITSTATUS(status);
+		switch (code) {
+		case 127:
+			return -ERR_RUN_COMMAND_EXEC;
+		case 0:
+			return 0;
+		default:
+			return -code;
+		}
+	}
+}
+
+int finish_command(struct child_process *cmd)
+{
+	return wait_or_whine(cmd->pid);
+}
+
+int run_command(struct child_process *cmd)
+{
+	int code = start_command(cmd);
+	if (code)
+		return code;
+	return finish_command(cmd);
+}
+
+static void prepare_run_command_v_opt(struct child_process *cmd,
+				      const char **argv,
+				      int opt)
+{
+	memset(cmd, 0, sizeof(*cmd));
+	cmd->argv = argv;
+	cmd->no_stdin = opt & RUN_COMMAND_NO_STDIN ? 1 : 0;
+	cmd->perf_cmd = opt & RUN_PERF_CMD ? 1 : 0;
+	cmd->stdout_to_stderr = opt & RUN_COMMAND_STDOUT_TO_STDERR ? 1 : 0;
+}
+
+int run_command_v_opt(const char **argv, int opt)
+{
+	struct child_process cmd;
+	prepare_run_command_v_opt(&cmd, argv, opt);
+	return run_command(&cmd);
+}
diff --git a/tools/perf/util/run-command.h b/tools/perf/util/run-command.h
new file mode 100644
index 00000000..1ef264d5
--- /dev/null
+++ b/tools/perf/util/run-command.h
@@ -0,0 +1,58 @@
+#ifndef __PERF_RUN_COMMAND_H
+#define __PERF_RUN_COMMAND_H
+
+enum {
+	ERR_RUN_COMMAND_FORK = 10000,
+	ERR_RUN_COMMAND_EXEC,
+	ERR_RUN_COMMAND_PIPE,
+	ERR_RUN_COMMAND_WAITPID,
+	ERR_RUN_COMMAND_WAITPID_WRONG_PID,
+	ERR_RUN_COMMAND_WAITPID_SIGNAL,
+	ERR_RUN_COMMAND_WAITPID_NOEXIT,
+};
+#define IS_RUN_COMMAND_ERR(x) (-(x) >= ERR_RUN_COMMAND_FORK)
+
+struct child_process {
+	const char **argv;
+	pid_t pid;
+	/*
+	 * Using .in, .out, .err:
+	 * - Specify 0 for no redirections (child inherits stdin, stdout,
+	 *   stderr from parent).
+	 * - Specify -1 to have a pipe allocated as follows:
+	 *     .in: returns the writable pipe end; parent writes to it,
+	 *          the readable pipe end becomes child's stdin
+	 *     .out, .err: returns the readable pipe end; parent reads from
+	 *          it, the writable pipe end becomes child's stdout/stderr
+	 *   The caller of start_command() must close the returned FDs
+	 *   after it has completed reading from/writing to it!
+	 * - Specify > 0 to set a channel to a particular FD as follows:
+	 *     .in: a readable FD, becomes child's stdin
+	 *     .out: a writable FD, becomes child's stdout/stderr
+	 *     .err > 0 not supported
+	 *   The specified FD is closed by start_command(), even in case
+	 *   of errors!
+	 */
+	int in;
+	int out;
+	int err;
+	const char *dir;
+	const char *const *env;
+	unsigned no_stdin:1;
+	unsigned no_stdout:1;
+	unsigned no_stderr:1;
+	unsigned perf_cmd:1; /* if this is to be perf sub-command */
+	unsigned stdout_to_stderr:1;
+	void (*preexec_cb)(void);
+};
+
+int start_command(struct child_process *);
+int finish_command(struct child_process *);
+int run_command(struct child_process *);
+
+#define RUN_COMMAND_NO_STDIN 1
+#define RUN_PERF_CMD	     2	/*If this is to be perf sub-command */
+#define RUN_COMMAND_STDOUT_TO_STDERR 4
+int run_command_v_opt(const char **argv, int opt);
+
+#endif /* __PERF_RUN_COMMAND_H */
diff --git a/tools/perf/util/scripting-engines/trace-event-perl.c b/tools/perf/util/scripting-engines/trace-event-perl.c
new file mode 100644
index 00000000..74350ffb
--- /dev/null
+++ b/tools/perf/util/scripting-engines/trace-event-perl.c
@@ -0,0 +1,571 @@
+/*
+ * trace-event-perl.  Feed perf script events to an embedded Perl interpreter.
+ *
+ * Copyright (C) 2009 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+#include <errno.h>
+
+#include "../../perf.h"
+#include "../util.h"
+#include "../trace-event.h"
+
+#include <EXTERN.h>
+#include <perl.h>
+
+void boot_Perf__Trace__Context(pTHX_ CV *cv);
+void boot_DynaLoader(pTHX_ CV *cv);
+typedef PerlInterpreter * INTERP;
+
+void xs_init(pTHX);
+
+void xs_init(pTHX)
+{
+	const char *file = __FILE__;
+	dXSUB_SYS;
+
+	newXS("Perf::Trace::Context::bootstrap", boot_Perf__Trace__Context,
+	      file);
+	newXS("DynaLoader::boot_DynaLoader", boot_DynaLoader, file);
+}
+
+INTERP my_perl;
+
+#define FTRACE_MAX_EVENT				\
+	((1 << (sizeof(unsigned short) * 8)) - 1)
+
+struct event *events[FTRACE_MAX_EVENT];
+
+extern struct scripting_context *scripting_context;
+
+static char *cur_field_name;
+static int zero_flag_atom;
+
+static void define_symbolic_value(const char *ev_name,
+				  const char *field_name,
+				  const char *field_value,
+				  const char *field_str)
+{
+	unsigned long long value;
+	dSP;
+
+	value = eval_flag(field_value);
+
+	ENTER;
+	SAVETMPS;
+	PUSHMARK(SP);
+
+	XPUSHs(sv_2mortal(newSVpv(ev_name, 0)));
+	XPUSHs(sv_2mortal(newSVpv(field_name, 0)));
+	XPUSHs(sv_2mortal(newSVuv(value)));
+	XPUSHs(sv_2mortal(newSVpv(field_str, 0)));
+
+	PUTBACK;
+	if (get_cv("main::define_symbolic_value", 0))
+		call_pv("main::define_symbolic_value", G_SCALAR);
+	SPAGAIN;
+	PUTBACK;
+	FREETMPS;
+	LEAVE;
+}
+
+static void define_symbolic_values(struct print_flag_sym *field,
+				   const char *ev_name,
+				   const char *field_name)
+{
+	define_symbolic_value(ev_name, field_name, field->value, field->str);
+	if (field->next)
+		define_symbolic_values(field->next, ev_name, field_name);
+}
+
+static void define_symbolic_field(const char *ev_name,
+				  const char *field_name)
+{
+	dSP;
+
+	ENTER;
+	SAVETMPS;
+	PUSHMARK(SP);
+
+	XPUSHs(sv_2mortal(newSVpv(ev_name, 0)));
+	XPUSHs(sv_2mortal(newSVpv(field_name, 0)));
+
+	PUTBACK;
+	if (get_cv("main::define_symbolic_field", 0))
+		call_pv("main::define_symbolic_field", G_SCALAR);
+	SPAGAIN;
+	PUTBACK;
+	FREETMPS;
+	LEAVE;
+}
+
+static void define_flag_value(const char *ev_name,
+			      const char *field_name,
+			      const char *field_value,
+			      const char *field_str)
+{
+	unsigned long long value;
+	dSP;
+
+	value = eval_flag(field_value);
+
+	ENTER;
+	SAVETMPS;
+	PUSHMARK(SP);
+
+	XPUSHs(sv_2mortal(newSVpv(ev_name, 0)));
+	XPUSHs(sv_2mortal(newSVpv(field_name, 0)));
+	XPUSHs(sv_2mortal(newSVuv(value)));
+	XPUSHs(sv_2mortal(newSVpv(field_str, 0)));
+
+	PUTBACK;
+	if (get_cv("main::define_flag_value", 0))
+		call_pv("main::define_flag_value", G_SCALAR);
+	SPAGAIN;
+	PUTBACK;
+	FREETMPS;
+	LEAVE;
+}
+
+static void define_flag_values(struct print_flag_sym *field,
+			       const char *ev_name,
+			       const char *field_name)
+{
+	define_flag_value(ev_name, field_name, field->value, field->str);
+	if (field->next)
+		define_flag_values(field->next, ev_name, field_name);
+}
+
+static void define_flag_field(const char *ev_name,
+			      const char *field_name,
+			      const char *delim)
+{
+	dSP;
+
+	ENTER;
+	SAVETMPS;
+	PUSHMARK(SP);
+
+	XPUSHs(sv_2mortal(newSVpv(ev_name, 0)));
+	XPUSHs(sv_2mortal(newSVpv(field_name, 0)));
+	XPUSHs(sv_2mortal(newSVpv(delim, 0)));
+
+	PUTBACK;
+	if (get_cv("main::define_flag_field", 0))
+		call_pv("main::define_flag_field", G_SCALAR);
+	SPAGAIN;
+	PUTBACK;
+	FREETMPS;
+	LEAVE;
+}
+
+static void define_event_symbols(struct event *event,
+				 const char *ev_name,
+				 struct print_arg *args)
+{
+	switch (args->type) {
+	case PRINT_NULL:
+		break;
+	case PRINT_ATOM:
+		define_flag_value(ev_name, cur_field_name, "0",
+				  args->atom.atom);
+		zero_flag_atom = 0;
+		break;
+	case PRINT_FIELD:
+		if (cur_field_name)
+			free(cur_field_name);
+		cur_field_name = strdup(args->field.name);
+		break;
+	case PRINT_FLAGS:
+		define_event_symbols(event, ev_name, args->flags.field);
+		define_flag_field(ev_name, cur_field_name, args->flags.delim);
+		define_flag_values(args->flags.flags, ev_name, cur_field_name);
+		break;
+	case PRINT_SYMBOL:
+		define_event_symbols(event, ev_name, args->symbol.field);
+		define_symbolic_field(ev_name, cur_field_name);
+		define_symbolic_values(args->symbol.symbols, ev_name,
+				       cur_field_name);
+		break;
+	case PRINT_STRING:
+		break;
+	case PRINT_TYPE:
+		define_event_symbols(event, ev_name, args->typecast.item);
+		break;
+	case PRINT_OP:
+		if (strcmp(args->op.op, ":") == 0)
+			zero_flag_atom = 1;
+		define_event_symbols(event, ev_name, args->op.left);
+		define_event_symbols(event, ev_name, args->op.right);
+		break;
+	default:
+		/* we should warn... */
+		return;
+	}
+
+	if (args->next)
+		define_event_symbols(event, ev_name, args->next);
+}
+
+static inline struct event *find_cache_event(int type)
+{
+	static char ev_name[256];
+	struct event *event;
+
+	if (events[type])
+		return events[type];
+
+	events[type] = event = trace_find_event(type);
+	if (!event)
+		return NULL;
+
+	sprintf(ev_name, "%s::%s", event->system, event->name);
+
+	define_event_symbols(event, ev_name, event->print_fmt.args);
+
+	return event;
+}
+
+static void perl_process_event(union perf_event *pevent __unused,
+			       struct perf_sample *sample,
+			       struct perf_evsel *evsel,
+			       struct perf_session *session __unused,
+			       struct thread *thread)
+{
+	struct format_field *field;
+	static char handler[256];
+	unsigned long long val;
+	unsigned long s, ns;
+	struct event *event;
+	int type;
+	int pid;
+	int cpu = sample->cpu;
+	void *data = sample->raw_data;
+	unsigned long long nsecs = sample->time;
+	char *comm = thread->comm;
+
+	dSP;
+
+	type = trace_parse_common_type(data);
+
+	event = find_cache_event(type);
+	if (!event)
+		die("ug! no event found for type %d", type);
+
+	pid = trace_parse_common_pid(data);
+
+	sprintf(handler, "%s::%s", event->system, event->name);
+
+	s = nsecs / NSECS_PER_SEC;
+	ns = nsecs - s * NSECS_PER_SEC;
+
+	scripting_context->event_data = data;
+
+	ENTER;
+	SAVETMPS;
+	PUSHMARK(SP);
+
+	XPUSHs(sv_2mortal(newSVpv(handler, 0)));
+	XPUSHs(sv_2mortal(newSViv(PTR2IV(scripting_context))));
+	XPUSHs(sv_2mortal(newSVuv(cpu)));
+	XPUSHs(sv_2mortal(newSVuv(s)));
+	XPUSHs(sv_2mortal(newSVuv(ns)));
+	XPUSHs(sv_2mortal(newSViv(pid)));
+	XPUSHs(sv_2mortal(newSVpv(comm, 0)));
+
+	/* common fields other than pid can be accessed via xsub fns */
+
+	for (field = event->format.fields; field; field = field->next) {
+		if (field->flags & FIELD_IS_STRING) {
+			int offset;
+			if (field->flags & FIELD_IS_DYNAMIC) {
+				offset = *(int *)(data + field->offset);
+				offset &= 0xffff;
+			} else
+				offset = field->offset;
+			XPUSHs(sv_2mortal(newSVpv((char *)data + offset, 0)));
+		} else { /* FIELD_IS_NUMERIC */
+			val = read_size(data + field->offset, field->size);
+			if (field->flags & FIELD_IS_SIGNED) {
+				XPUSHs(sv_2mortal(newSViv(val)));
+			} else {
+				XPUSHs(sv_2mortal(newSVuv(val)));
+			}
+		}
+	}
+
+	PUTBACK;
+
+	if (get_cv(handler, 0))
+		call_pv(handler, G_SCALAR);
+	else if (get_cv("main::trace_unhandled", 0)) {
+		XPUSHs(sv_2mortal(newSVpv(handler, 0)));
+		XPUSHs(sv_2mortal(newSViv(PTR2IV(scripting_context))));
+		XPUSHs(sv_2mortal(newSVuv(cpu)));
+		XPUSHs(sv_2mortal(newSVuv(nsecs)));
+		XPUSHs(sv_2mortal(newSViv(pid)));
+		XPUSHs(sv_2mortal(newSVpv(comm, 0)));
+		call_pv("main::trace_unhandled", G_SCALAR);
+	}
+	SPAGAIN;
+	PUTBACK;
+	FREETMPS;
+	LEAVE;
+}
+
+static void run_start_sub(void)
+{
+	dSP; /* access to Perl stack */
+	PUSHMARK(SP);
+
+	if (get_cv("main::trace_begin", 0))
+		call_pv("main::trace_begin", G_DISCARD | G_NOARGS);
+}
+
+/*
+ * Start trace script
+ */
+static int perl_start_script(const char *script, int argc, const char **argv)
+{
+	const char **command_line;
+	int i, err = 0;
+
+	command_line = malloc((argc + 2) * sizeof(const char *));
+	command_line[0] = "";
+	command_line[1] = script;
+	for (i = 2; i < argc + 2; i++)
+		command_line[i] = argv[i - 2];
+
+	my_perl = perl_alloc();
+	perl_construct(my_perl);
+
+	if (perl_parse(my_perl, xs_init, argc + 2, (char **)command_line,
+		       (char **)NULL)) {
+		err = -1;
+		goto error;
+	}
+
+	if (perl_run(my_perl)) {
+		err = -1;
+		goto error;
+	}
+
+	if (SvTRUE(ERRSV)) {
+		err = -1;
+		goto error;
+	}
+
+	run_start_sub();
+
+	free(command_line);
+	return 0;
+error:
+	perl_free(my_perl);
+	free(command_line);
+
+	return err;
+}
+
+/*
+ * Stop trace script
+ */
+static int perl_stop_script(void)
+{
+	dSP; /* access to Perl stack */
+	PUSHMARK(SP);
+
+	if (get_cv("main::trace_end", 0))
+		call_pv("main::trace_end", G_DISCARD | G_NOARGS);
+
+	perl_destruct(my_perl);
+	perl_free(my_perl);
+
+	return 0;
+}
+
+static int perl_generate_script(const char *outfile)
+{
+	struct event *event = NULL;
+	struct format_field *f;
+	char fname[PATH_MAX];
+	int not_first, count;
+	FILE *ofp;
+
+	sprintf(fname, "%s.pl", outfile);
+	ofp = fopen(fname, "w");
+	if (ofp == NULL) {
+		fprintf(stderr, "couldn't open %s\n", fname);
+		return -1;
+	}
+
+	fprintf(ofp, "# perf script event handlers, "
+		"generated by perf script -g perl\n");
+
+	fprintf(ofp, "# Licensed under the terms of the GNU GPL"
+		" License version 2\n\n");
+
+	fprintf(ofp, "# The common_* event handler fields are the most useful "
+		"fields common to\n");
+
+	fprintf(ofp, "# all events.  They don't necessarily correspond to "
+		"the 'common_*' fields\n");
+
+	fprintf(ofp, "# in the format files.  Those fields not available as "
+		"handler params can\n");
+
+	fprintf(ofp, "# be retrieved using Perl functions of the form "
+		"common_*($context).\n");
+
+	fprintf(ofp, "# See Context.pm for the list of available "
+		"functions.\n\n");
+
+	fprintf(ofp, "use lib \"$ENV{'PERF_EXEC_PATH'}/scripts/perl/"
+		"Perf-Trace-Util/lib\";\n");
+
+	fprintf(ofp, "use lib \"./Perf-Trace-Util/lib\";\n");
+	fprintf(ofp, "use Perf::Trace::Core;\n");
+	fprintf(ofp, "use Perf::Trace::Context;\n");
+	fprintf(ofp, "use Perf::Trace::Util;\n\n");
+
+	fprintf(ofp, "sub trace_begin\n{\n\t# optional\n}\n\n");
+	fprintf(ofp, "sub trace_end\n{\n\t# optional\n}\n\n");
+
+	while ((event = trace_find_next_event(event))) {
+		fprintf(ofp, "sub %s::%s\n{\n", event->system, event->name);
+		fprintf(ofp, "\tmy (");
+
+		fprintf(ofp, "$event_name, ");
+		fprintf(ofp, "$context, ");
+		fprintf(ofp, "$common_cpu, ");
+		fprintf(ofp, "$common_secs, ");
+		fprintf(ofp, "$common_nsecs,\n");
+		fprintf(ofp, "\t    $common_pid, ");
+		fprintf(ofp, "$common_comm,\n\t    ");
+
+		not_first = 0;
+		count = 0;
+
+		for (f = event->format.fields; f; f = f->next) {
+			if (not_first++)
+				fprintf(ofp, ", ");
+			if (++count % 5 == 0)
+				fprintf(ofp, "\n\t    ");
+
+			fprintf(ofp, "$%s", f->name);
+		}
+		fprintf(ofp, ") = @_;\n\n");
+
+		fprintf(ofp, "\tprint_header($event_name, $common_cpu, "
+			"$common_secs, $common_nsecs,\n\t             "
+			"$common_pid, $common_comm);\n\n");
+
+		fprintf(ofp, "\tprintf(\"");
+
+		not_first = 0;
+		count = 0;
+
+		for (f = event->format.fields; f; f = f->next) {
+			if (not_first++)
+				fprintf(ofp, ", ");
+			if (count && count % 4 == 0) {
+				fprintf(ofp, "\".\n\t       \"");
+			}
+			count++;
+
+			fprintf(ofp, "%s=", f->name);
+			if (f->flags & FIELD_IS_STRING ||
+			    f->flags & FIELD_IS_FLAG ||
+			    f->flags & FIELD_IS_SYMBOLIC)
+				fprintf(ofp, "%%s");
+			else if (f->flags & FIELD_IS_SIGNED)
+				fprintf(ofp, "%%d");
+			else
+				fprintf(ofp, "%%u");
+		}
+
+		fprintf(ofp, "\\n\",\n\t       ");
+
+		not_first = 0;
+		count = 0;
+
+		for (f = event->format.fields; f; f = f->next) {
+			if (not_first++)
+				fprintf(ofp, ", ");
+
+			if (++count % 5 == 0)
+				fprintf(ofp, "\n\t       ");
+
+			if (f->flags & FIELD_IS_FLAG) {
+				if ((count - 1) % 5 != 0) {
+					fprintf(ofp, "\n\t       ");
+					count = 4;
+				}
+				fprintf(ofp, "flag_str(\"");
+				fprintf(ofp, "%s::%s\", ", event->system,
+					event->name);
+				fprintf(ofp, "\"%s\", $%s)", f->name,
+					f->name);
+			} else if (f->flags & FIELD_IS_SYMBOLIC) {
+				if ((count - 1) % 5 != 0) {
+					fprintf(ofp, "\n\t       ");
+					count = 4;
+				}
+				fprintf(ofp, "symbol_str(\"");
+				fprintf(ofp, "%s::%s\", ", event->system,
+					event->name);
+				fprintf(ofp, "\"%s\", $%s)", f->name,
+					f->name);
+			} else
+				fprintf(ofp, "$%s", f->name);
+		}
+
+		fprintf(ofp, ");\n");
+		fprintf(ofp, "}\n\n");
+	}
+
+	fprintf(ofp, "sub trace_unhandled\n{\n\tmy ($event_name, $context, "
+		"$common_cpu, $common_secs, $common_nsecs,\n\t    "
+		"$common_pid, $common_comm) = @_;\n\n");
+
+	fprintf(ofp, "\tprint_header($event_name, $common_cpu, "
+		"$common_secs, $common_nsecs,\n\t             $common_pid, "
+		"$common_comm);\n}\n\n");
+
+	fprintf(ofp, "sub print_header\n{\n"
+		"\tmy ($event_name, $cpu, $secs, $nsecs, $pid, $comm) = @_;\n\n"
+		"\tprintf(\"%%-20s %%5u %%05u.%%09u %%8u %%-20s \",\n\t       "
+		"$event_name, $cpu, $secs, $nsecs, $pid, $comm);\n}");
+
+	fclose(ofp);
+
+	fprintf(stderr, "generated Perl script: %s\n", fname);
+
+	return 0;
+}
+
+struct scripting_ops perl_scripting_ops = {
+	.name = "Perl",
+	.start_script = perl_start_script,
+	.stop_script = perl_stop_script,
+	.process_event = perl_process_event,
+	.generate_script = perl_generate_script,
+};
diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c
new file mode 100644
index 00000000..6ccf70e8
--- /dev/null
+++ b/tools/perf/util/scripting-engines/trace-event-python.c
@@ -0,0 +1,599 @@
+/*
+ * trace-event-python.  Feed trace events to an embedded Python interpreter.
+ *
+ * Copyright (C) 2010 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <Python.h>
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+#include <errno.h>
+
+#include "../../perf.h"
+#include "../util.h"
+#include "../trace-event.h"
+
+PyMODINIT_FUNC initperf_trace_context(void);
+
+#define FTRACE_MAX_EVENT				\
+	((1 << (sizeof(unsigned short) * 8)) - 1)
+
+struct event *events[FTRACE_MAX_EVENT];
+
+#define MAX_FIELDS	64
+#define N_COMMON_FIELDS	7
+
+extern struct scripting_context *scripting_context;
+
+static char *cur_field_name;
+static int zero_flag_atom;
+
+static PyObject *main_module, *main_dict;
+
+static void handler_call_die(const char *handler_name)
+{
+	PyErr_Print();
+	Py_FatalError("problem in Python trace event handler");
+}
+
+static void define_value(enum print_arg_type field_type,
+			 const char *ev_name,
+			 const char *field_name,
+			 const char *field_value,
+			 const char *field_str)
+{
+	const char *handler_name = "define_flag_value";
+	PyObject *handler, *t, *retval;
+	unsigned long long value;
+	unsigned n = 0;
+
+	if (field_type == PRINT_SYMBOL)
+		handler_name = "define_symbolic_value";
+
+	t = PyTuple_New(4);
+	if (!t)
+		Py_FatalError("couldn't create Python tuple");
+
+	value = eval_flag(field_value);
+
+	PyTuple_SetItem(t, n++, PyString_FromString(ev_name));
+	PyTuple_SetItem(t, n++, PyString_FromString(field_name));
+	PyTuple_SetItem(t, n++, PyInt_FromLong(value));
+	PyTuple_SetItem(t, n++, PyString_FromString(field_str));
+
+	handler = PyDict_GetItemString(main_dict, handler_name);
+	if (handler && PyCallable_Check(handler)) {
+		retval = PyObject_CallObject(handler, t);
+		if (retval == NULL)
+			handler_call_die(handler_name);
+	}
+
+	Py_DECREF(t);
+}
+
+static void define_values(enum print_arg_type field_type,
+			  struct print_flag_sym *field,
+			  const char *ev_name,
+			  const char *field_name)
+{
+	define_value(field_type, ev_name, field_name, field->value,
+		     field->str);
+
+	if (field->next)
+		define_values(field_type, field->next, ev_name, field_name);
+}
+
+static void define_field(enum print_arg_type field_type,
+			 const char *ev_name,
+			 const char *field_name,
+			 const char *delim)
+{
+	const char *handler_name = "define_flag_field";
+	PyObject *handler, *t, *retval;
+	unsigned n = 0;
+
+	if (field_type == PRINT_SYMBOL)
+		handler_name = "define_symbolic_field";
+
+	if (field_type == PRINT_FLAGS)
+		t = PyTuple_New(3);
+	else
+		t = PyTuple_New(2);
+	if (!t)
+		Py_FatalError("couldn't create Python tuple");
+
+	PyTuple_SetItem(t, n++, PyString_FromString(ev_name));
+	PyTuple_SetItem(t, n++, PyString_FromString(field_name));
+	if (field_type == PRINT_FLAGS)
+		PyTuple_SetItem(t, n++, PyString_FromString(delim));
+
+	handler = PyDict_GetItemString(main_dict, handler_name);
+	if (handler && PyCallable_Check(handler)) {
+		retval = PyObject_CallObject(handler, t);
+		if (retval == NULL)
+			handler_call_die(handler_name);
+	}
+
+	Py_DECREF(t);
+}
+
+static void define_event_symbols(struct event *event,
+				 const char *ev_name,
+				 struct print_arg *args)
+{
+	switch (args->type) {
+	case PRINT_NULL:
+		break;
+	case PRINT_ATOM:
+		define_value(PRINT_FLAGS, ev_name, cur_field_name, "0",
+			     args->atom.atom);
+		zero_flag_atom = 0;
+		break;
+	case PRINT_FIELD:
+		if (cur_field_name)
+			free(cur_field_name);
+		cur_field_name = strdup(args->field.name);
+		break;
+	case PRINT_FLAGS:
+		define_event_symbols(event, ev_name, args->flags.field);
+		define_field(PRINT_FLAGS, ev_name, cur_field_name,
+			     args->flags.delim);
+		define_values(PRINT_FLAGS, args->flags.flags, ev_name,
+			      cur_field_name);
+		break;
+	case PRINT_SYMBOL:
+		define_event_symbols(event, ev_name, args->symbol.field);
+		define_field(PRINT_SYMBOL, ev_name, cur_field_name, NULL);
+		define_values(PRINT_SYMBOL, args->symbol.symbols, ev_name,
+			      cur_field_name);
+		break;
+	case PRINT_STRING:
+		break;
+	case PRINT_TYPE:
+		define_event_symbols(event, ev_name, args->typecast.item);
+		break;
+	case PRINT_OP:
+		if (strcmp(args->op.op, ":") == 0)
+			zero_flag_atom = 1;
+		define_event_symbols(event, ev_name, args->op.left);
+		define_event_symbols(event, ev_name, args->op.right);
+		break;
+	default:
+		/* we should warn... */
+		return;
+	}
+
+	if (args->next)
+		define_event_symbols(event, ev_name, args->next);
+}
+
+static inline struct event *find_cache_event(int type)
+{
+	static char ev_name[256];
+	struct event *event;
+
+	if (events[type])
+		return events[type];
+
+	events[type] = event = trace_find_event(type);
+	if (!event)
+		return NULL;
+
+	sprintf(ev_name, "%s__%s", event->system, event->name);
+
+	define_event_symbols(event, ev_name, event->print_fmt.args);
+
+	return event;
+}
+
+static void python_process_event(union perf_event *pevent __unused,
+				 struct perf_sample *sample,
+				 struct perf_evsel *evsel __unused,
+				 struct perf_session *session __unused,
+				 struct thread *thread)
+{
+	PyObject *handler, *retval, *context, *t, *obj, *dict = NULL;
+	static char handler_name[256];
+	struct format_field *field;
+	unsigned long long val;
+	unsigned long s, ns;
+	struct event *event;
+	unsigned n = 0;
+	int type;
+	int pid;
+	int cpu = sample->cpu;
+	void *data = sample->raw_data;
+	unsigned long long nsecs = sample->time;
+	char *comm = thread->comm;
+
+	t = PyTuple_New(MAX_FIELDS);
+	if (!t)
+		Py_FatalError("couldn't create Python tuple");
+
+	type = trace_parse_common_type(data);
+
+	event = find_cache_event(type);
+	if (!event)
+		die("ug! no event found for type %d", type);
+
+	pid = trace_parse_common_pid(data);
+
+	sprintf(handler_name, "%s__%s", event->system, event->name);
+
+	handler = PyDict_GetItemString(main_dict, handler_name);
+	if (handler && !PyCallable_Check(handler))
+		handler = NULL;
+	if (!handler) {
+		dict = PyDict_New();
+		if (!dict)
+			Py_FatalError("couldn't create Python dict");
+	}
+	s = nsecs / NSECS_PER_SEC;
+	ns = nsecs - s * NSECS_PER_SEC;
+
+	scripting_context->event_data = data;
+
+	context = PyCObject_FromVoidPtr(scripting_context, NULL);
+
+	PyTuple_SetItem(t, n++, PyString_FromString(handler_name));
+	PyTuple_SetItem(t, n++, context);
+
+	if (handler) {
+		PyTuple_SetItem(t, n++, PyInt_FromLong(cpu));
+		PyTuple_SetItem(t, n++, PyInt_FromLong(s));
+		PyTuple_SetItem(t, n++, PyInt_FromLong(ns));
+		PyTuple_SetItem(t, n++, PyInt_FromLong(pid));
+		PyTuple_SetItem(t, n++, PyString_FromString(comm));
+	} else {
+		PyDict_SetItemString(dict, "common_cpu", PyInt_FromLong(cpu));
+		PyDict_SetItemString(dict, "common_s", PyInt_FromLong(s));
+		PyDict_SetItemString(dict, "common_ns", PyInt_FromLong(ns));
+		PyDict_SetItemString(dict, "common_pid", PyInt_FromLong(pid));
+		PyDict_SetItemString(dict, "common_comm", PyString_FromString(comm));
+	}
+	for (field = event->format.fields; field; field = field->next) {
+		if (field->flags & FIELD_IS_STRING) {
+			int offset;
+			if (field->flags & FIELD_IS_DYNAMIC) {
+				offset = *(int *)(data + field->offset);
+				offset &= 0xffff;
+			} else
+				offset = field->offset;
+			obj = PyString_FromString((char *)data + offset);
+		} else { /* FIELD_IS_NUMERIC */
+			val = read_size(data + field->offset, field->size);
+			if (field->flags & FIELD_IS_SIGNED) {
+				if ((long long)val >= LONG_MIN &&
+				    (long long)val <= LONG_MAX)
+					obj = PyInt_FromLong(val);
+				else
+					obj = PyLong_FromLongLong(val);
+			} else {
+				if (val <= LONG_MAX)
+					obj = PyInt_FromLong(val);
+				else
+					obj = PyLong_FromUnsignedLongLong(val);
+			}
+		}
+		if (handler)
+			PyTuple_SetItem(t, n++, obj);
+		else
+			PyDict_SetItemString(dict, field->name, obj);
+
+	}
+	if (!handler)
+		PyTuple_SetItem(t, n++, dict);
+
+	if (_PyTuple_Resize(&t, n) == -1)
+		Py_FatalError("error resizing Python tuple");
+
+	if (handler) {
+		retval = PyObject_CallObject(handler, t);
+		if (retval == NULL)
+			handler_call_die(handler_name);
+	} else {
+		handler = PyDict_GetItemString(main_dict, "trace_unhandled");
+		if (handler && PyCallable_Check(handler)) {
+
+			retval = PyObject_CallObject(handler, t);
+			if (retval == NULL)
+				handler_call_die("trace_unhandled");
+		}
+		Py_DECREF(dict);
+	}
+
+	Py_DECREF(t);
+}
+
+static int run_start_sub(void)
+{
+	PyObject *handler, *retval;
+	int err = 0;
+
+	main_module = PyImport_AddModule("__main__");
+	if (main_module == NULL)
+		return -1;
+	Py_INCREF(main_module);
+
+	main_dict = PyModule_GetDict(main_module);
+	if (main_dict == NULL) {
+		err = -1;
+		goto error;
+	}
+	Py_INCREF(main_dict);
+
+	handler = PyDict_GetItemString(main_dict, "trace_begin");
+	if (handler == NULL || !PyCallable_Check(handler))
+		goto out;
+
+	retval = PyObject_CallObject(handler, NULL);
+	if (retval == NULL)
+		handler_call_die("trace_begin");
+
+	Py_DECREF(retval);
+	return err;
+error:
+	Py_XDECREF(main_dict);
+	Py_XDECREF(main_module);
+out:
+	return err;
+}
+
+/*
+ * Start trace script
+ */
+static int python_start_script(const char *script, int argc, const char **argv)
+{
+	const char **command_line;
+	char buf[PATH_MAX];
+	int i, err = 0;
+	FILE *fp;
+
+	command_line = malloc((argc + 1) * sizeof(const char *));
+	command_line[0] = script;
+	for (i = 1; i < argc + 1; i++)
+		command_line[i] = argv[i - 1];
+
+	Py_Initialize();
+
+	initperf_trace_context();
+
+	PySys_SetArgv(argc + 1, (char **)command_line);
+
+	fp = fopen(script, "r");
+	if (!fp) {
+		sprintf(buf, "Can't open python script \"%s\"", script);
+		perror(buf);
+		err = -1;
+		goto error;
+	}
+
+	err = PyRun_SimpleFile(fp, script);
+	if (err) {
+		fprintf(stderr, "Error running python script %s\n", script);
+		goto error;
+	}
+
+	err = run_start_sub();
+	if (err) {
+		fprintf(stderr, "Error starting python script %s\n", script);
+		goto error;
+	}
+
+	free(command_line);
+
+	return err;
+error:
+	Py_Finalize();
+	free(command_line);
+
+	return err;
+}
+
+/*
+ * Stop trace script
+ */
+static int python_stop_script(void)
+{
+	PyObject *handler, *retval;
+	int err = 0;
+
+	handler = PyDict_GetItemString(main_dict, "trace_end");
+	if (handler == NULL || !PyCallable_Check(handler))
+		goto out;
+
+	retval = PyObject_CallObject(handler, NULL);
+	if (retval == NULL)
+		handler_call_die("trace_end");
+	else
+		Py_DECREF(retval);
+out:
+	Py_XDECREF(main_dict);
+	Py_XDECREF(main_module);
+	Py_Finalize();
+
+	return err;
+}
+
+static int python_generate_script(const char *outfile)
+{
+	struct event *event = NULL;
+	struct format_field *f;
+	char fname[PATH_MAX];
+	int not_first, count;
+	FILE *ofp;
+
+	sprintf(fname, "%s.py", outfile);
+	ofp = fopen(fname, "w");
+	if (ofp == NULL) {
+		fprintf(stderr, "couldn't open %s\n", fname);
+		return -1;
+	}
+	fprintf(ofp, "# perf script event handlers, "
+		"generated by perf script -g python\n");
+
+	fprintf(ofp, "# Licensed under the terms of the GNU GPL"
+		" License version 2\n\n");
+
+	fprintf(ofp, "# The common_* event handler fields are the most useful "
+		"fields common to\n");
+
+	fprintf(ofp, "# all events.  They don't necessarily correspond to "
+		"the 'common_*' fields\n");
+
+	fprintf(ofp, "# in the format files.  Those fields not available as "
+		"handler params can\n");
+
+	fprintf(ofp, "# be retrieved using Python functions of the form "
+		"common_*(context).\n");
+
+	fprintf(ofp, "# See the perf-trace-python Documentation for the list "
+		"of available functions.\n\n");
+
+	fprintf(ofp, "import os\n");
+	fprintf(ofp, "import sys\n\n");
+
+	fprintf(ofp, "sys.path.append(os.environ['PERF_EXEC_PATH'] + \\\n");
+	fprintf(ofp, "\t'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')\n");
+	fprintf(ofp, "\nfrom perf_trace_context import *\n");
+	fprintf(ofp, "from Core import *\n\n\n");
+
+	fprintf(ofp, "def trace_begin():\n");
+	fprintf(ofp, "\tprint \"in trace_begin\"\n\n");
+
+	fprintf(ofp, "def trace_end():\n");
+	fprintf(ofp, "\tprint \"in trace_end\"\n\n");
+
+	while ((event = trace_find_next_event(event))) {
+		fprintf(ofp, "def %s__%s(", event->system, event->name);
+		fprintf(ofp, "event_name, ");
+		fprintf(ofp, "context, ");
+		fprintf(ofp, "common_cpu,\n");
+		fprintf(ofp, "\tcommon_secs, ");
+		fprintf(ofp, "common_nsecs, ");
+		fprintf(ofp, "common_pid, ");
+		fprintf(ofp, "common_comm,\n\t");
+
+		not_first = 0;
+		count = 0;
+
+		for (f = event->format.fields; f; f = f->next) {
+			if (not_first++)
+				fprintf(ofp, ", ");
+			if (++count % 5 == 0)
+				fprintf(ofp, "\n\t");
+
+			fprintf(ofp, "%s", f->name);
+		}
+		fprintf(ofp, "):\n");
+
+		fprintf(ofp, "\t\tprint_header(event_name, common_cpu, "
+			"common_secs, common_nsecs,\n\t\t\t"
+			"common_pid, common_comm)\n\n");
+
+		fprintf(ofp, "\t\tprint \"");
+
+		not_first = 0;
+		count = 0;
+
+		for (f = event->format.fields; f; f = f->next) {
+			if (not_first++)
+				fprintf(ofp, ", ");
+			if (count && count % 3 == 0) {
+				fprintf(ofp, "\" \\\n\t\t\"");
+			}
+			count++;
+
+			fprintf(ofp, "%s=", f->name);
+			if (f->flags & FIELD_IS_STRING ||
+			    f->flags & FIELD_IS_FLAG ||
+			    f->flags & FIELD_IS_SYMBOLIC)
+				fprintf(ofp, "%%s");
+			else if (f->flags & FIELD_IS_SIGNED)
+				fprintf(ofp, "%%d");
+			else
+				fprintf(ofp, "%%u");
+		}
+
+		fprintf(ofp, "\\n\" %% \\\n\t\t(");
+
+		not_first = 0;
+		count = 0;
+
+		for (f = event->format.fields; f; f = f->next) {
+			if (not_first++)
+				fprintf(ofp, ", ");
+
+			if (++count % 5 == 0)
+				fprintf(ofp, "\n\t\t");
+
+			if (f->flags & FIELD_IS_FLAG) {
+				if ((count - 1) % 5 != 0) {
+					fprintf(ofp, "\n\t\t");
+					count = 4;
+				}
+				fprintf(ofp, "flag_str(\"");
+				fprintf(ofp, "%s__%s\", ", event->system,
+					event->name);
+				fprintf(ofp, "\"%s\", %s)", f->name,
+					f->name);
+			} else if (f->flags & FIELD_IS_SYMBOLIC) {
+				if ((count - 1) % 5 != 0) {
+					fprintf(ofp, "\n\t\t");
+					count = 4;
+				}
+				fprintf(ofp, "symbol_str(\"");
+				fprintf(ofp, "%s__%s\", ", event->system,
+					event->name);
+				fprintf(ofp, "\"%s\", %s)", f->name,
+					f->name);
+			} else
+				fprintf(ofp, "%s", f->name);
+		}
+
+		fprintf(ofp, "),\n\n");
+	}
+
+	fprintf(ofp, "def trace_unhandled(event_name, context, "
+		"event_fields_dict):\n");
+
+	fprintf(ofp, "\t\tprint ' '.join(['%%s=%%s'%%(k,str(v))"
+		"for k,v in sorted(event_fields_dict.items())])\n\n");
+
+	fprintf(ofp, "def print_header("
+		"event_name, cpu, secs, nsecs, pid, comm):\n"
+		"\tprint \"%%-20s %%5u %%05u.%%09u %%8u %%-20s \" %% \\\n\t"
+		"(event_name, cpu, secs, nsecs, pid, comm),\n");
+
+	fclose(ofp);
+
+	fprintf(stderr, "generated Python script: %s\n", fname);
+
+	return 0;
+}
+
+struct scripting_ops python_scripting_ops = {
+	.name = "Python",
+	.start_script = python_start_script,
+	.stop_script = python_stop_script,
+	.process_event = python_process_event,
+	.generate_script = python_generate_script,
+};
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
new file mode 100644
index 00000000..2dbf0abd
--- /dev/null
+++ b/tools/perf/util/session.c
@@ -0,0 +1,1270 @@
+#define _FILE_OFFSET_BITS 64
+
+#include <linux/kernel.h>
+
+#include <byteswap.h>
+#include <unistd.h>
+#include <sys/types.h>
+#include <sys/mman.h>
+
+#include "evlist.h"
+#include "evsel.h"
+#include "session.h"
+#include "sort.h"
+#include "util.h"
+
+static int perf_session__open(struct perf_session *self, bool force)
+{
+	struct stat input_stat;
+
+	if (!strcmp(self->filename, "-")) {
+		self->fd_pipe = true;
+		self->fd = STDIN_FILENO;
+
+		if (perf_session__read_header(self, self->fd) < 0)
+			pr_err("incompatible file format");
+
+		return 0;
+	}
+
+	self->fd = open(self->filename, O_RDONLY);
+	if (self->fd < 0) {
+		int err = errno;
+
+		pr_err("failed to open %s: %s", self->filename, strerror(err));
+		if (err == ENOENT && !strcmp(self->filename, "perf.data"))
+			pr_err("  (try 'perf record' first)");
+		pr_err("\n");
+		return -errno;
+	}
+
+	if (fstat(self->fd, &input_stat) < 0)
+		goto out_close;
+
+	if (!force && input_stat.st_uid && (input_stat.st_uid != geteuid())) {
+		pr_err("file %s not owned by current user or root\n",
+		       self->filename);
+		goto out_close;
+	}
+
+	if (!input_stat.st_size) {
+		pr_info("zero-sized file (%s), nothing to do!\n",
+			self->filename);
+		goto out_close;
+	}
+
+	if (perf_session__read_header(self, self->fd) < 0) {
+		pr_err("incompatible file format");
+		goto out_close;
+	}
+
+	if (!perf_evlist__valid_sample_type(self->evlist)) {
+		pr_err("non matching sample_type");
+		goto out_close;
+	}
+
+	if (!perf_evlist__valid_sample_id_all(self->evlist)) {
+		pr_err("non matching sample_id_all");
+		goto out_close;
+	}
+
+	self->size = input_stat.st_size;
+	return 0;
+
+out_close:
+	close(self->fd);
+	self->fd = -1;
+	return -1;
+}
+
+static void perf_session__id_header_size(struct perf_session *session)
+{
+       struct perf_sample *data;
+       u64 sample_type = session->sample_type;
+       u16 size = 0;
+
+	if (!session->sample_id_all)
+		goto out;
+
+       if (sample_type & PERF_SAMPLE_TID)
+               size += sizeof(data->tid) * 2;
+
+       if (sample_type & PERF_SAMPLE_TIME)
+               size += sizeof(data->time);
+
+       if (sample_type & PERF_SAMPLE_ID)
+               size += sizeof(data->id);
+
+       if (sample_type & PERF_SAMPLE_STREAM_ID)
+               size += sizeof(data->stream_id);
+
+       if (sample_type & PERF_SAMPLE_CPU)
+               size += sizeof(data->cpu) * 2;
+out:
+       session->id_hdr_size = size;
+}
+
+void perf_session__update_sample_type(struct perf_session *self)
+{
+	self->sample_type = perf_evlist__sample_type(self->evlist);
+	self->sample_size = __perf_evsel__sample_size(self->sample_type);
+	self->sample_id_all = perf_evlist__sample_id_all(self->evlist);
+	perf_session__id_header_size(self);
+}
+
+int perf_session__create_kernel_maps(struct perf_session *self)
+{
+	int ret = machine__create_kernel_maps(&self->host_machine);
+
+	if (ret >= 0)
+		ret = machines__create_guest_kernel_maps(&self->machines);
+	return ret;
+}
+
+static void perf_session__destroy_kernel_maps(struct perf_session *self)
+{
+	machine__destroy_kernel_maps(&self->host_machine);
+	machines__destroy_guest_kernel_maps(&self->machines);
+}
+
+struct perf_session *perf_session__new(const char *filename, int mode,
+				       bool force, bool repipe,
+				       struct perf_event_ops *ops)
+{
+	size_t len = filename ? strlen(filename) + 1 : 0;
+	struct perf_session *self = zalloc(sizeof(*self) + len);
+
+	if (self == NULL)
+		goto out;
+
+	memcpy(self->filename, filename, len);
+	self->threads = RB_ROOT;
+	INIT_LIST_HEAD(&self->dead_threads);
+	self->last_match = NULL;
+	/*
+	 * On 64bit we can mmap the data file in one go. No need for tiny mmap
+	 * slices. On 32bit we use 32MB.
+	 */
+#if BITS_PER_LONG == 64
+	self->mmap_window = ULLONG_MAX;
+#else
+	self->mmap_window = 32 * 1024 * 1024ULL;
+#endif
+	self->machines = RB_ROOT;
+	self->repipe = repipe;
+	INIT_LIST_HEAD(&self->ordered_samples.samples);
+	INIT_LIST_HEAD(&self->ordered_samples.sample_cache);
+	INIT_LIST_HEAD(&self->ordered_samples.to_free);
+	machine__init(&self->host_machine, "", HOST_KERNEL_ID);
+
+	if (mode == O_RDONLY) {
+		if (perf_session__open(self, force) < 0)
+			goto out_delete;
+		perf_session__update_sample_type(self);
+	} else if (mode == O_WRONLY) {
+		/*
+		 * In O_RDONLY mode this will be performed when reading the
+		 * kernel MMAP event, in perf_event__process_mmap().
+		 */
+		if (perf_session__create_kernel_maps(self) < 0)
+			goto out_delete;
+	}
+
+	if (ops && ops->ordering_requires_timestamps &&
+	    ops->ordered_samples && !self->sample_id_all) {
+		dump_printf("WARNING: No sample_id_all support, falling back to unordered processing\n");
+		ops->ordered_samples = false;
+	}
+
+out:
+	return self;
+out_delete:
+	perf_session__delete(self);
+	return NULL;
+}
+
+static void perf_session__delete_dead_threads(struct perf_session *self)
+{
+	struct thread *n, *t;
+
+	list_for_each_entry_safe(t, n, &self->dead_threads, node) {
+		list_del(&t->node);
+		thread__delete(t);
+	}
+}
+
+static void perf_session__delete_threads(struct perf_session *self)
+{
+	struct rb_node *nd = rb_first(&self->threads);
+
+	while (nd) {
+		struct thread *t = rb_entry(nd, struct thread, rb_node);
+
+		rb_erase(&t->rb_node, &self->threads);
+		nd = rb_next(nd);
+		thread__delete(t);
+	}
+}
+
+void perf_session__delete(struct perf_session *self)
+{
+	perf_session__destroy_kernel_maps(self);
+	perf_session__delete_dead_threads(self);
+	perf_session__delete_threads(self);
+	machine__exit(&self->host_machine);
+	close(self->fd);
+	free(self);
+}
+
+void perf_session__remove_thread(struct perf_session *self, struct thread *th)
+{
+	self->last_match = NULL;
+	rb_erase(&th->rb_node, &self->threads);
+	/*
+	 * We may have references to this thread, for instance in some hist_entry
+	 * instances, so just move them to a separate list.
+	 */
+	list_add_tail(&th->node, &self->dead_threads);
+}
+
+static bool symbol__match_parent_regex(struct symbol *sym)
+{
+	if (sym->name && !regexec(&parent_regex, sym->name, 0, NULL, 0))
+		return 1;
+
+	return 0;
+}
+
+int perf_session__resolve_callchain(struct perf_session *self,
+				    struct thread *thread,
+				    struct ip_callchain *chain,
+				    struct symbol **parent)
+{
+	u8 cpumode = PERF_RECORD_MISC_USER;
+	unsigned int i;
+	int err;
+
+	callchain_cursor_reset(&self->callchain_cursor);
+
+	for (i = 0; i < chain->nr; i++) {
+		u64 ip = chain->ips[i];
+		struct addr_location al;
+
+		if (ip >= PERF_CONTEXT_MAX) {
+			switch (ip) {
+			case PERF_CONTEXT_HV:
+				cpumode = PERF_RECORD_MISC_HYPERVISOR;	break;
+			case PERF_CONTEXT_KERNEL:
+				cpumode = PERF_RECORD_MISC_KERNEL;	break;
+			case PERF_CONTEXT_USER:
+				cpumode = PERF_RECORD_MISC_USER;	break;
+			default:
+				break;
+			}
+			continue;
+		}
+
+		al.filtered = false;
+		thread__find_addr_location(thread, self, cpumode,
+				MAP__FUNCTION, thread->pid, ip, &al, NULL);
+		if (al.sym != NULL) {
+			if (sort__has_parent && !*parent &&
+			    symbol__match_parent_regex(al.sym))
+				*parent = al.sym;
+			if (!symbol_conf.use_callchain)
+				break;
+		}
+
+		err = callchain_cursor_append(&self->callchain_cursor,
+					      ip, al.map, al.sym);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int process_event_synth_stub(union perf_event *event __used,
+				    struct perf_session *session __used)
+{
+	dump_printf(": unhandled!\n");
+	return 0;
+}
+
+static int process_event_sample_stub(union perf_event *event __used,
+				     struct perf_sample *sample __used,
+				     struct perf_evsel *evsel __used,
+				     struct perf_session *session __used)
+{
+	dump_printf(": unhandled!\n");
+	return 0;
+}
+
+static int process_event_stub(union perf_event *event __used,
+			      struct perf_sample *sample __used,
+			      struct perf_session *session __used)
+{
+	dump_printf(": unhandled!\n");
+	return 0;
+}
+
+static int process_finished_round_stub(union perf_event *event __used,
+				       struct perf_session *session __used,
+				       struct perf_event_ops *ops __used)
+{
+	dump_printf(": unhandled!\n");
+	return 0;
+}
+
+static int process_finished_round(union perf_event *event,
+				  struct perf_session *session,
+				  struct perf_event_ops *ops);
+
+static void perf_event_ops__fill_defaults(struct perf_event_ops *handler)
+{
+	if (handler->sample == NULL)
+		handler->sample = process_event_sample_stub;
+	if (handler->mmap == NULL)
+		handler->mmap = process_event_stub;
+	if (handler->comm == NULL)
+		handler->comm = process_event_stub;
+	if (handler->fork == NULL)
+		handler->fork = process_event_stub;
+	if (handler->exit == NULL)
+		handler->exit = process_event_stub;
+	if (handler->lost == NULL)
+		handler->lost = perf_event__process_lost;
+	if (handler->read == NULL)
+		handler->read = process_event_stub;
+	if (handler->throttle == NULL)
+		handler->throttle = process_event_stub;
+	if (handler->unthrottle == NULL)
+		handler->unthrottle = process_event_stub;
+	if (handler->attr == NULL)
+		handler->attr = process_event_synth_stub;
+	if (handler->event_type == NULL)
+		handler->event_type = process_event_synth_stub;
+	if (handler->tracing_data == NULL)
+		handler->tracing_data = process_event_synth_stub;
+	if (handler->build_id == NULL)
+		handler->build_id = process_event_synth_stub;
+	if (handler->finished_round == NULL) {
+		if (handler->ordered_samples)
+			handler->finished_round = process_finished_round;
+		else
+			handler->finished_round = process_finished_round_stub;
+	}
+}
+
+void mem_bswap_64(void *src, int byte_size)
+{
+	u64 *m = src;
+
+	while (byte_size > 0) {
+		*m = bswap_64(*m);
+		byte_size -= sizeof(u64);
+		++m;
+	}
+}
+
+static void perf_event__all64_swap(union perf_event *event)
+{
+	struct perf_event_header *hdr = &event->header;
+	mem_bswap_64(hdr + 1, event->header.size - sizeof(*hdr));
+}
+
+static void perf_event__comm_swap(union perf_event *event)
+{
+	event->comm.pid = bswap_32(event->comm.pid);
+	event->comm.tid = bswap_32(event->comm.tid);
+}
+
+static void perf_event__mmap_swap(union perf_event *event)
+{
+	event->mmap.pid	  = bswap_32(event->mmap.pid);
+	event->mmap.tid	  = bswap_32(event->mmap.tid);
+	event->mmap.start = bswap_64(event->mmap.start);
+	event->mmap.len	  = bswap_64(event->mmap.len);
+	event->mmap.pgoff = bswap_64(event->mmap.pgoff);
+}
+
+static void perf_event__task_swap(union perf_event *event)
+{
+	event->fork.pid	 = bswap_32(event->fork.pid);
+	event->fork.tid	 = bswap_32(event->fork.tid);
+	event->fork.ppid = bswap_32(event->fork.ppid);
+	event->fork.ptid = bswap_32(event->fork.ptid);
+	event->fork.time = bswap_64(event->fork.time);
+}
+
+static void perf_event__read_swap(union perf_event *event)
+{
+	event->read.pid		 = bswap_32(event->read.pid);
+	event->read.tid		 = bswap_32(event->read.tid);
+	event->read.value	 = bswap_64(event->read.value);
+	event->read.time_enabled = bswap_64(event->read.time_enabled);
+	event->read.time_running = bswap_64(event->read.time_running);
+	event->read.id		 = bswap_64(event->read.id);
+}
+
+/* exported for swapping attributes in file header */
+void perf_event__attr_swap(struct perf_event_attr *attr)
+{
+	attr->type		= bswap_32(attr->type);
+	attr->size		= bswap_32(attr->size);
+	attr->config		= bswap_64(attr->config);
+	attr->sample_period	= bswap_64(attr->sample_period);
+	attr->sample_type	= bswap_64(attr->sample_type);
+	attr->read_format	= bswap_64(attr->read_format);
+	attr->wakeup_events	= bswap_32(attr->wakeup_events);
+	attr->bp_type		= bswap_32(attr->bp_type);
+	attr->bp_addr		= bswap_64(attr->bp_addr);
+	attr->bp_len		= bswap_64(attr->bp_len);
+}
+
+static void perf_event__hdr_attr_swap(union perf_event *event)
+{
+	size_t size;
+
+	perf_event__attr_swap(&event->attr.attr);
+
+	size = event->header.size;
+	size -= (void *)&event->attr.id - (void *)event;
+	mem_bswap_64(event->attr.id, size);
+}
+
+static void perf_event__event_type_swap(union perf_event *event)
+{
+	event->event_type.event_type.event_id =
+		bswap_64(event->event_type.event_type.event_id);
+}
+
+static void perf_event__tracing_data_swap(union perf_event *event)
+{
+	event->tracing_data.size = bswap_32(event->tracing_data.size);
+}
+
+typedef void (*perf_event__swap_op)(union perf_event *event);
+
+static perf_event__swap_op perf_event__swap_ops[] = {
+	[PERF_RECORD_MMAP]		  = perf_event__mmap_swap,
+	[PERF_RECORD_COMM]		  = perf_event__comm_swap,
+	[PERF_RECORD_FORK]		  = perf_event__task_swap,
+	[PERF_RECORD_EXIT]		  = perf_event__task_swap,
+	[PERF_RECORD_LOST]		  = perf_event__all64_swap,
+	[PERF_RECORD_READ]		  = perf_event__read_swap,
+	[PERF_RECORD_SAMPLE]		  = perf_event__all64_swap,
+	[PERF_RECORD_HEADER_ATTR]	  = perf_event__hdr_attr_swap,
+	[PERF_RECORD_HEADER_EVENT_TYPE]	  = perf_event__event_type_swap,
+	[PERF_RECORD_HEADER_TRACING_DATA] = perf_event__tracing_data_swap,
+	[PERF_RECORD_HEADER_BUILD_ID]	  = NULL,
+	[PERF_RECORD_HEADER_MAX]	  = NULL,
+};
+
+struct sample_queue {
+	u64			timestamp;
+	u64			file_offset;
+	union perf_event	*event;
+	struct list_head	list;
+};
+
+static void perf_session_free_sample_buffers(struct perf_session *session)
+{
+	struct ordered_samples *os = &session->ordered_samples;
+
+	while (!list_empty(&os->to_free)) {
+		struct sample_queue *sq;
+
+		sq = list_entry(os->to_free.next, struct sample_queue, list);
+		list_del(&sq->list);
+		free(sq);
+	}
+}
+
+static int perf_session_deliver_event(struct perf_session *session,
+				      union perf_event *event,
+				      struct perf_sample *sample,
+				      struct perf_event_ops *ops,
+				      u64 file_offset);
+
+static void flush_sample_queue(struct perf_session *s,
+			       struct perf_event_ops *ops)
+{
+	struct ordered_samples *os = &s->ordered_samples;
+	struct list_head *head = &os->samples;
+	struct sample_queue *tmp, *iter;
+	struct perf_sample sample;
+	u64 limit = os->next_flush;
+	u64 last_ts = os->last_sample ? os->last_sample->timestamp : 0ULL;
+	int ret;
+
+	if (!ops->ordered_samples || !limit)
+		return;
+
+	list_for_each_entry_safe(iter, tmp, head, list) {
+		if (iter->timestamp > limit)
+			break;
+
+		ret = perf_session__parse_sample(s, iter->event, &sample);
+		if (ret)
+			pr_err("Can't parse sample, err = %d\n", ret);
+		else
+			perf_session_deliver_event(s, iter->event, &sample, ops,
+						   iter->file_offset);
+
+		os->last_flush = iter->timestamp;
+		list_del(&iter->list);
+		list_add(&iter->list, &os->sample_cache);
+	}
+
+	if (list_empty(head)) {
+		os->last_sample = NULL;
+	} else if (last_ts <= limit) {
+		os->last_sample =
+			list_entry(head->prev, struct sample_queue, list);
+	}
+}
+
+/*
+ * When perf record finishes a pass on every buffers, it records this pseudo
+ * event.
+ * We record the max timestamp t found in the pass n.
+ * Assuming these timestamps are monotonic across cpus, we know that if
+ * a buffer still has events with timestamps below t, they will be all
+ * available and then read in the pass n + 1.
+ * Hence when we start to read the pass n + 2, we can safely flush every
+ * events with timestamps below t.
+ *
+ *    ============ PASS n =================
+ *       CPU 0         |   CPU 1
+ *                     |
+ *    cnt1 timestamps  |   cnt2 timestamps
+ *          1          |         2
+ *          2          |         3
+ *          -          |         4  <--- max recorded
+ *
+ *    ============ PASS n + 1 ==============
+ *       CPU 0         |   CPU 1
+ *                     |
+ *    cnt1 timestamps  |   cnt2 timestamps
+ *          3          |         5
+ *          4          |         6
+ *          5          |         7 <---- max recorded
+ *
+ *      Flush every events below timestamp 4
+ *
+ *    ============ PASS n + 2 ==============
+ *       CPU 0         |   CPU 1
+ *                     |
+ *    cnt1 timestamps  |   cnt2 timestamps
+ *          6          |         8
+ *          7          |         9
+ *          -          |         10
+ *
+ *      Flush every events below timestamp 7
+ *      etc...
+ */
+static int process_finished_round(union perf_event *event __used,
+				  struct perf_session *session,
+				  struct perf_event_ops *ops)
+{
+	flush_sample_queue(session, ops);
+	session->ordered_samples.next_flush = session->ordered_samples.max_timestamp;
+
+	return 0;
+}
+
+/* The queue is ordered by time */
+static void __queue_event(struct sample_queue *new, struct perf_session *s)
+{
+	struct ordered_samples *os = &s->ordered_samples;
+	struct sample_queue *sample = os->last_sample;
+	u64 timestamp = new->timestamp;
+	struct list_head *p;
+
+	os->last_sample = new;
+
+	if (!sample) {
+		list_add(&new->list, &os->samples);
+		os->max_timestamp = timestamp;
+		return;
+	}
+
+	/*
+	 * last_sample might point to some random place in the list as it's
+	 * the last queued event. We expect that the new event is close to
+	 * this.
+	 */
+	if (sample->timestamp <= timestamp) {
+		while (sample->timestamp <= timestamp) {
+			p = sample->list.next;
+			if (p == &os->samples) {
+				list_add_tail(&new->list, &os->samples);
+				os->max_timestamp = timestamp;
+				return;
+			}
+			sample = list_entry(p, struct sample_queue, list);
+		}
+		list_add_tail(&new->list, &sample->list);
+	} else {
+		while (sample->timestamp > timestamp) {
+			p = sample->list.prev;
+			if (p == &os->samples) {
+				list_add(&new->list, &os->samples);
+				return;
+			}
+			sample = list_entry(p, struct sample_queue, list);
+		}
+		list_add(&new->list, &sample->list);
+	}
+}
+
+#define MAX_SAMPLE_BUFFER	(64 * 1024 / sizeof(struct sample_queue))
+
+static int perf_session_queue_event(struct perf_session *s, union perf_event *event,
+				    struct perf_sample *sample, u64 file_offset)
+{
+	struct ordered_samples *os = &s->ordered_samples;
+	struct list_head *sc = &os->sample_cache;
+	u64 timestamp = sample->time;
+	struct sample_queue *new;
+
+	if (!timestamp || timestamp == ~0ULL)
+		return -ETIME;
+
+	if (timestamp < s->ordered_samples.last_flush) {
+		printf("Warning: Timestamp below last timeslice flush\n");
+		return -EINVAL;
+	}
+
+	if (!list_empty(sc)) {
+		new = list_entry(sc->next, struct sample_queue, list);
+		list_del(&new->list);
+	} else if (os->sample_buffer) {
+		new = os->sample_buffer + os->sample_buffer_idx;
+		if (++os->sample_buffer_idx == MAX_SAMPLE_BUFFER)
+			os->sample_buffer = NULL;
+	} else {
+		os->sample_buffer = malloc(MAX_SAMPLE_BUFFER * sizeof(*new));
+		if (!os->sample_buffer)
+			return -ENOMEM;
+		list_add(&os->sample_buffer->list, &os->to_free);
+		os->sample_buffer_idx = 2;
+		new = os->sample_buffer + 1;
+	}
+
+	new->timestamp = timestamp;
+	new->file_offset = file_offset;
+	new->event = event;
+
+	__queue_event(new, s);
+
+	return 0;
+}
+
+static void callchain__printf(struct perf_sample *sample)
+{
+	unsigned int i;
+
+	printf("... chain: nr:%" PRIu64 "\n", sample->callchain->nr);
+
+	for (i = 0; i < sample->callchain->nr; i++)
+		printf("..... %2d: %016" PRIx64 "\n",
+		       i, sample->callchain->ips[i]);
+}
+
+static void perf_session__print_tstamp(struct perf_session *session,
+				       union perf_event *event,
+				       struct perf_sample *sample)
+{
+	if (event->header.type != PERF_RECORD_SAMPLE &&
+	    !session->sample_id_all) {
+		fputs("-1 -1 ", stdout);
+		return;
+	}
+
+	if ((session->sample_type & PERF_SAMPLE_CPU))
+		printf("%u ", sample->cpu);
+
+	if (session->sample_type & PERF_SAMPLE_TIME)
+		printf("%" PRIu64 " ", sample->time);
+}
+
+static void dump_event(struct perf_session *session, union perf_event *event,
+		       u64 file_offset, struct perf_sample *sample)
+{
+	if (!dump_trace)
+		return;
+
+	printf("\n%#" PRIx64 " [%#x]: event: %d\n",
+	       file_offset, event->header.size, event->header.type);
+
+	trace_event(event);
+
+	if (sample)
+		perf_session__print_tstamp(session, event, sample);
+
+	printf("%#" PRIx64 " [%#x]: PERF_RECORD_%s", file_offset,
+	       event->header.size, perf_event__name(event->header.type));
+}
+
+static void dump_sample(struct perf_session *session, union perf_event *event,
+			struct perf_sample *sample)
+{
+	if (!dump_trace)
+		return;
+
+	printf("(IP, %d): %d/%d: %#" PRIx64 " period: %" PRIu64 "\n",
+	       event->header.misc, sample->pid, sample->tid, sample->ip,
+	       sample->period);
+
+	if (session->sample_type & PERF_SAMPLE_CALLCHAIN)
+		callchain__printf(sample);
+}
+
+static int perf_session_deliver_event(struct perf_session *session,
+				      union perf_event *event,
+				      struct perf_sample *sample,
+				      struct perf_event_ops *ops,
+				      u64 file_offset)
+{
+	struct perf_evsel *evsel;
+
+	dump_event(session, event, file_offset, sample);
+
+	switch (event->header.type) {
+	case PERF_RECORD_SAMPLE:
+		dump_sample(session, event, sample);
+		evsel = perf_evlist__id2evsel(session->evlist, sample->id);
+		if (evsel == NULL) {
+			++session->hists.stats.nr_unknown_id;
+			return -1;
+		}
+		return ops->sample(event, sample, evsel, session);
+	case PERF_RECORD_MMAP:
+		return ops->mmap(event, sample, session);
+	case PERF_RECORD_COMM:
+		return ops->comm(event, sample, session);
+	case PERF_RECORD_FORK:
+		return ops->fork(event, sample, session);
+	case PERF_RECORD_EXIT:
+		return ops->exit(event, sample, session);
+	case PERF_RECORD_LOST:
+		return ops->lost(event, sample, session);
+	case PERF_RECORD_READ:
+		return ops->read(event, sample, session);
+	case PERF_RECORD_THROTTLE:
+		return ops->throttle(event, sample, session);
+	case PERF_RECORD_UNTHROTTLE:
+		return ops->unthrottle(event, sample, session);
+	default:
+		++session->hists.stats.nr_unknown_events;
+		return -1;
+	}
+}
+
+static int perf_session__preprocess_sample(struct perf_session *session,
+					   union perf_event *event, struct perf_sample *sample)
+{
+	if (event->header.type != PERF_RECORD_SAMPLE ||
+	    !(session->sample_type & PERF_SAMPLE_CALLCHAIN))
+		return 0;
+
+	if (!ip_callchain__valid(sample->callchain, event)) {
+		pr_debug("call-chain problem with event, skipping it.\n");
+		++session->hists.stats.nr_invalid_chains;
+		session->hists.stats.total_invalid_chains += sample->period;
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int perf_session__process_user_event(struct perf_session *session, union perf_event *event,
+					    struct perf_event_ops *ops, u64 file_offset)
+{
+	dump_event(session, event, file_offset, NULL);
+
+	/* These events are processed right away */
+	switch (event->header.type) {
+	case PERF_RECORD_HEADER_ATTR:
+		return ops->attr(event, session);
+	case PERF_RECORD_HEADER_EVENT_TYPE:
+		return ops->event_type(event, session);
+	case PERF_RECORD_HEADER_TRACING_DATA:
+		/* setup for reading amidst mmap */
+		lseek(session->fd, file_offset, SEEK_SET);
+		return ops->tracing_data(event, session);
+	case PERF_RECORD_HEADER_BUILD_ID:
+		return ops->build_id(event, session);
+	case PERF_RECORD_FINISHED_ROUND:
+		return ops->finished_round(event, session, ops);
+	default:
+		return -EINVAL;
+	}
+}
+
+static int perf_session__process_event(struct perf_session *session,
+				       union perf_event *event,
+				       struct perf_event_ops *ops,
+				       u64 file_offset)
+{
+	struct perf_sample sample;
+	int ret;
+
+	if (session->header.needs_swap &&
+	    perf_event__swap_ops[event->header.type])
+		perf_event__swap_ops[event->header.type](event);
+
+	if (event->header.type >= PERF_RECORD_HEADER_MAX)
+		return -EINVAL;
+
+	hists__inc_nr_events(&session->hists, event->header.type);
+
+	if (event->header.type >= PERF_RECORD_USER_TYPE_START)
+		return perf_session__process_user_event(session, event, ops, file_offset);
+
+	/*
+	 * For all kernel events we get the sample data
+	 */
+	ret = perf_session__parse_sample(session, event, &sample);
+	if (ret)
+		return ret;
+
+	/* Preprocess sample records - precheck callchains */
+	if (perf_session__preprocess_sample(session, event, &sample))
+		return 0;
+
+	if (ops->ordered_samples) {
+		ret = perf_session_queue_event(session, event, &sample,
+					       file_offset);
+		if (ret != -ETIME)
+			return ret;
+	}
+
+	return perf_session_deliver_event(session, event, &sample, ops,
+					  file_offset);
+}
+
+void perf_event_header__bswap(struct perf_event_header *self)
+{
+	self->type = bswap_32(self->type);
+	self->misc = bswap_16(self->misc);
+	self->size = bswap_16(self->size);
+}
+
+static struct thread *perf_session__register_idle_thread(struct perf_session *self)
+{
+	struct thread *thread = perf_session__findnew(self, 0);
+
+	if (thread == NULL || thread__set_comm(thread, "swapper")) {
+		pr_err("problem inserting idle task.\n");
+		thread = NULL;
+	}
+
+	return thread;
+}
+
+static void perf_session__warn_about_errors(const struct perf_session *session,
+					    const struct perf_event_ops *ops)
+{
+	if (ops->lost == perf_event__process_lost &&
+	    session->hists.stats.total_lost != 0) {
+		ui__warning("Processed %" PRIu64 " events and LOST %" PRIu64
+			    "!\n\nCheck IO/CPU overload!\n\n",
+			    session->hists.stats.total_period,
+			    session->hists.stats.total_lost);
+	}
+
+	if (session->hists.stats.nr_unknown_events != 0) {
+		ui__warning("Found %u unknown events!\n\n"
+			    "Is this an older tool processing a perf.data "
+			    "file generated by a more recent tool?\n\n"
+			    "If that is not the case, consider "
+			    "reporting to linux-kernel@vger.kernel.org.\n\n",
+			    session->hists.stats.nr_unknown_events);
+	}
+
+	if (session->hists.stats.nr_unknown_id != 0) {
+		ui__warning("%u samples with id not present in the header\n",
+			    session->hists.stats.nr_unknown_id);
+	}
+
+ 	if (session->hists.stats.nr_invalid_chains != 0) {
+ 		ui__warning("Found invalid callchains!\n\n"
+ 			    "%u out of %u events were discarded for this reason.\n\n"
+ 			    "Consider reporting to linux-kernel@vger.kernel.org.\n\n",
+ 			    session->hists.stats.nr_invalid_chains,
+ 			    session->hists.stats.nr_events[PERF_RECORD_SAMPLE]);
+ 	}
+}
+
+#define session_done()	(*(volatile int *)(&session_done))
+volatile int session_done;
+
+static int __perf_session__process_pipe_events(struct perf_session *self,
+					       struct perf_event_ops *ops)
+{
+	union perf_event event;
+	uint32_t size;
+	int skip = 0;
+	u64 head;
+	int err;
+	void *p;
+
+	perf_event_ops__fill_defaults(ops);
+
+	head = 0;
+more:
+	err = readn(self->fd, &event, sizeof(struct perf_event_header));
+	if (err <= 0) {
+		if (err == 0)
+			goto done;
+
+		pr_err("failed to read event header\n");
+		goto out_err;
+	}
+
+	if (self->header.needs_swap)
+		perf_event_header__bswap(&event.header);
+
+	size = event.header.size;
+	if (size == 0)
+		size = 8;
+
+	p = &event;
+	p += sizeof(struct perf_event_header);
+
+	if (size - sizeof(struct perf_event_header)) {
+		err = readn(self->fd, p, size - sizeof(struct perf_event_header));
+		if (err <= 0) {
+			if (err == 0) {
+				pr_err("unexpected end of event stream\n");
+				goto done;
+			}
+
+			pr_err("failed to read event data\n");
+			goto out_err;
+		}
+	}
+
+	if (size == 0 ||
+	    (skip = perf_session__process_event(self, &event, ops, head)) < 0) {
+		dump_printf("%#" PRIx64 " [%#x]: skipping unknown header type: %d\n",
+			    head, event.header.size, event.header.type);
+		/*
+		 * assume we lost track of the stream, check alignment, and
+		 * increment a single u64 in the hope to catch on again 'soon'.
+		 */
+		if (unlikely(head & 7))
+			head &= ~7ULL;
+
+		size = 8;
+	}
+
+	head += size;
+
+	if (skip > 0)
+		head += skip;
+
+	if (!session_done())
+		goto more;
+done:
+	err = 0;
+out_err:
+	perf_session__warn_about_errors(self, ops);
+	perf_session_free_sample_buffers(self);
+	return err;
+}
+
+static union perf_event *
+fetch_mmaped_event(struct perf_session *session,
+		   u64 head, size_t mmap_size, char *buf)
+{
+	union perf_event *event;
+
+	/*
+	 * Ensure we have enough space remaining to read
+	 * the size of the event in the headers.
+	 */
+	if (head + sizeof(event->header) > mmap_size)
+		return NULL;
+
+	event = (union perf_event *)(buf + head);
+
+	if (session->header.needs_swap)
+		perf_event_header__bswap(&event->header);
+
+	if (head + event->header.size > mmap_size)
+		return NULL;
+
+	return event;
+}
+
+int __perf_session__process_events(struct perf_session *session,
+				   u64 data_offset, u64 data_size,
+				   u64 file_size, struct perf_event_ops *ops)
+{
+	u64 head, page_offset, file_offset, file_pos, progress_next;
+	int err, mmap_prot, mmap_flags, map_idx = 0;
+	struct ui_progress *progress;
+	size_t	page_size, mmap_size;
+	char *buf, *mmaps[8];
+	union perf_event *event;
+	uint32_t size;
+
+	perf_event_ops__fill_defaults(ops);
+
+	page_size = sysconf(_SC_PAGESIZE);
+
+	page_offset = page_size * (data_offset / page_size);
+	file_offset = page_offset;
+	head = data_offset - page_offset;
+
+	if (data_offset + data_size < file_size)
+		file_size = data_offset + data_size;
+
+	progress_next = file_size / 16;
+	progress = ui_progress__new("Processing events...", file_size);
+	if (progress == NULL)
+		return -1;
+
+	mmap_size = session->mmap_window;
+	if (mmap_size > file_size)
+		mmap_size = file_size;
+
+	memset(mmaps, 0, sizeof(mmaps));
+
+	mmap_prot  = PROT_READ;
+	mmap_flags = MAP_SHARED;
+
+	if (session->header.needs_swap) {
+		mmap_prot  |= PROT_WRITE;
+		mmap_flags = MAP_PRIVATE;
+	}
+remap:
+	buf = mmap(NULL, mmap_size, mmap_prot, mmap_flags, session->fd,
+		   file_offset);
+	if (buf == MAP_FAILED) {
+		pr_err("failed to mmap file\n");
+		err = -errno;
+		goto out_err;
+	}
+	mmaps[map_idx] = buf;
+	map_idx = (map_idx + 1) & (ARRAY_SIZE(mmaps) - 1);
+	file_pos = file_offset + head;
+
+more:
+	event = fetch_mmaped_event(session, head, mmap_size, buf);
+	if (!event) {
+		if (mmaps[map_idx]) {
+			munmap(mmaps[map_idx], mmap_size);
+			mmaps[map_idx] = NULL;
+		}
+
+		page_offset = page_size * (head / page_size);
+		file_offset += page_offset;
+		head -= page_offset;
+		goto remap;
+	}
+
+	size = event->header.size;
+
+	if (size == 0 ||
+	    perf_session__process_event(session, event, ops, file_pos) < 0) {
+		dump_printf("%#" PRIx64 " [%#x]: skipping unknown header type: %d\n",
+			    file_offset + head, event->header.size,
+			    event->header.type);
+		/*
+		 * assume we lost track of the stream, check alignment, and
+		 * increment a single u64 in the hope to catch on again 'soon'.
+		 */
+		if (unlikely(head & 7))
+			head &= ~7ULL;
+
+		size = 8;
+	}
+
+	head += size;
+	file_pos += size;
+
+	if (file_pos >= progress_next) {
+		progress_next += file_size / 16;
+		ui_progress__update(progress, file_pos);
+	}
+
+	if (file_pos < file_size)
+		goto more;
+
+	err = 0;
+	/* do the final flush for ordered samples */
+	session->ordered_samples.next_flush = ULLONG_MAX;
+	flush_sample_queue(session, ops);
+out_err:
+	ui_progress__delete(progress);
+	perf_session__warn_about_errors(session, ops);
+	perf_session_free_sample_buffers(session);
+	return err;
+}
+
+int perf_session__process_events(struct perf_session *self,
+				 struct perf_event_ops *ops)
+{
+	int err;
+
+	if (perf_session__register_idle_thread(self) == NULL)
+		return -ENOMEM;
+
+	if (!self->fd_pipe)
+		err = __perf_session__process_events(self,
+						     self->header.data_offset,
+						     self->header.data_size,
+						     self->size, ops);
+	else
+		err = __perf_session__process_pipe_events(self, ops);
+
+	return err;
+}
+
+bool perf_session__has_traces(struct perf_session *self, const char *msg)
+{
+	if (!(self->sample_type & PERF_SAMPLE_RAW)) {
+		pr_err("No trace sample to read. Did you call 'perf %s'?\n", msg);
+		return false;
+	}
+
+	return true;
+}
+
+int perf_session__set_kallsyms_ref_reloc_sym(struct map **maps,
+					     const char *symbol_name,
+					     u64 addr)
+{
+	char *bracket;
+	enum map_type i;
+	struct ref_reloc_sym *ref;
+
+	ref = zalloc(sizeof(struct ref_reloc_sym));
+	if (ref == NULL)
+		return -ENOMEM;
+
+	ref->name = strdup(symbol_name);
+	if (ref->name == NULL) {
+		free(ref);
+		return -ENOMEM;
+	}
+
+	bracket = strchr(ref->name, ']');
+	if (bracket)
+		*bracket = '\0';
+
+	ref->addr = addr;
+
+	for (i = 0; i < MAP__NR_TYPES; ++i) {
+		struct kmap *kmap = map__kmap(maps[i]);
+		kmap->ref_reloc_sym = ref;
+	}
+
+	return 0;
+}
+
+size_t perf_session__fprintf_dsos(struct perf_session *self, FILE *fp)
+{
+	return __dsos__fprintf(&self->host_machine.kernel_dsos, fp) +
+	       __dsos__fprintf(&self->host_machine.user_dsos, fp) +
+	       machines__fprintf_dsos(&self->machines, fp);
+}
+
+size_t perf_session__fprintf_dsos_buildid(struct perf_session *self, FILE *fp,
+					  bool with_hits)
+{
+	size_t ret = machine__fprintf_dsos_buildid(&self->host_machine, fp, with_hits);
+	return ret + machines__fprintf_dsos_buildid(&self->machines, fp, with_hits);
+}
+
+size_t perf_session__fprintf_nr_events(struct perf_session *session, FILE *fp)
+{
+	struct perf_evsel *pos;
+	size_t ret = fprintf(fp, "Aggregated stats:\n");
+
+	ret += hists__fprintf_nr_events(&session->hists, fp);
+
+	list_for_each_entry(pos, &session->evlist->entries, node) {
+		ret += fprintf(fp, "%s stats:\n", event_name(pos));
+		ret += hists__fprintf_nr_events(&pos->hists, fp);
+	}
+
+	return ret;
+}
+
+struct perf_evsel *perf_session__find_first_evtype(struct perf_session *session,
+					      unsigned int type)
+{
+	struct perf_evsel *pos;
+
+	list_for_each_entry(pos, &session->evlist->entries, node) {
+		if (pos->attr.type == type)
+			return pos;
+	}
+	return NULL;
+}
+
+void perf_session__print_symbols(union perf_event *event,
+				struct perf_sample *sample,
+				struct perf_session *session)
+{
+	struct addr_location al;
+	const char *symname, *dsoname;
+	struct callchain_cursor *cursor = &session->callchain_cursor;
+	struct callchain_cursor_node *node;
+
+	if (perf_event__preprocess_sample(event, session, &al, sample,
+					  NULL) < 0) {
+		error("problem processing %d event, skipping it.\n",
+			event->header.type);
+		return;
+	}
+
+	if (symbol_conf.use_callchain && sample->callchain) {
+
+		if (perf_session__resolve_callchain(session, al.thread,
+						sample->callchain, NULL) != 0) {
+			if (verbose)
+				error("Failed to resolve callchain. Skipping\n");
+			return;
+		}
+		callchain_cursor_commit(cursor);
+
+		while (1) {
+			node = callchain_cursor_current(cursor);
+			if (!node)
+				break;
+
+			if (node->sym && node->sym->name)
+				symname = node->sym->name;
+			else
+				symname = "";
+
+			if (node->map && node->map->dso && node->map->dso->name)
+				dsoname = node->map->dso->name;
+			else
+				dsoname = "";
+
+			printf("\t%16" PRIx64 " %s (%s)\n", node->ip, symname, dsoname);
+
+			callchain_cursor_advance(cursor);
+		}
+
+	} else {
+		if (al.sym && al.sym->name)
+			symname = al.sym->name;
+		else
+			symname = "";
+
+		if (al.map && al.map->dso && al.map->dso->name)
+			dsoname = al.map->dso->name;
+		else
+			dsoname = "";
+
+		printf("%16" PRIx64 " %s (%s)", al.addr, symname, dsoname);
+	}
+}
diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h
new file mode 100644
index 00000000..b84c0031
--- /dev/null
+++ b/tools/perf/util/session.h
@@ -0,0 +1,175 @@
+#ifndef __PERF_SESSION_H
+#define __PERF_SESSION_H
+
+#include "hist.h"
+#include "event.h"
+#include "header.h"
+#include "symbol.h"
+#include "thread.h"
+#include <linux/rbtree.h>
+#include "../../../include/linux/perf_event.h"
+
+struct sample_queue;
+struct ip_callchain;
+struct thread;
+
+struct ordered_samples {
+	u64			last_flush;
+	u64			next_flush;
+	u64			max_timestamp;
+	struct list_head	samples;
+	struct list_head	sample_cache;
+	struct list_head	to_free;
+	struct sample_queue	*sample_buffer;
+	struct sample_queue	*last_sample;
+	int			sample_buffer_idx;
+};
+
+struct perf_session {
+	struct perf_header	header;
+	unsigned long		size;
+	unsigned long		mmap_window;
+	struct rb_root		threads;
+	struct list_head	dead_threads;
+	struct thread		*last_match;
+	struct machine		host_machine;
+	struct rb_root		machines;
+	struct perf_evlist	*evlist;
+	/*
+	 * FIXME: Need to split this up further, we need global
+	 *	  stats + per event stats. 'perf diff' also needs
+	 *	  to properly support multiple events in a single
+	 *	  perf.data file.
+	 */
+	struct hists		hists;
+	u64			sample_type;
+	int			sample_size;
+	int			fd;
+	bool			fd_pipe;
+	bool			repipe;
+	bool			sample_id_all;
+	u16			id_hdr_size;
+	int			cwdlen;
+	char			*cwd;
+	struct ordered_samples	ordered_samples;
+	struct callchain_cursor	callchain_cursor;
+	char			filename[0];
+};
+
+struct perf_evsel;
+struct perf_event_ops;
+
+typedef int (*event_sample)(union perf_event *event, struct perf_sample *sample,
+			    struct perf_evsel *evsel, struct perf_session *session);
+typedef int (*event_op)(union perf_event *self, struct perf_sample *sample,
+			struct perf_session *session);
+typedef int (*event_synth_op)(union perf_event *self,
+			      struct perf_session *session);
+typedef int (*event_op2)(union perf_event *self, struct perf_session *session,
+			 struct perf_event_ops *ops);
+
+struct perf_event_ops {
+	event_sample	sample;
+	event_op	mmap,
+			comm,
+			fork,
+			exit,
+			lost,
+			read,
+			throttle,
+			unthrottle;
+	event_synth_op	attr,
+			event_type,
+			tracing_data,
+			build_id;
+	event_op2	finished_round;
+	bool		ordered_samples;
+	bool		ordering_requires_timestamps;
+};
+
+struct perf_session *perf_session__new(const char *filename, int mode,
+				       bool force, bool repipe,
+				       struct perf_event_ops *ops);
+void perf_session__delete(struct perf_session *self);
+
+void perf_event_header__bswap(struct perf_event_header *self);
+
+int __perf_session__process_events(struct perf_session *self,
+				   u64 data_offset, u64 data_size, u64 size,
+				   struct perf_event_ops *ops);
+int perf_session__process_events(struct perf_session *self,
+				 struct perf_event_ops *event_ops);
+
+int perf_session__resolve_callchain(struct perf_session *self,
+				    struct thread *thread,
+				    struct ip_callchain *chain,
+				    struct symbol **parent);
+
+bool perf_session__has_traces(struct perf_session *self, const char *msg);
+
+int perf_session__set_kallsyms_ref_reloc_sym(struct map **maps,
+					     const char *symbol_name,
+					     u64 addr);
+
+void mem_bswap_64(void *src, int byte_size);
+void perf_event__attr_swap(struct perf_event_attr *attr);
+
+int perf_session__create_kernel_maps(struct perf_session *self);
+
+void perf_session__update_sample_type(struct perf_session *self);
+void perf_session__remove_thread(struct perf_session *self, struct thread *th);
+
+static inline
+struct machine *perf_session__find_host_machine(struct perf_session *self)
+{
+	return &self->host_machine;
+}
+
+static inline
+struct machine *perf_session__find_machine(struct perf_session *self, pid_t pid)
+{
+	if (pid == HOST_KERNEL_ID)
+		return &self->host_machine;
+	return machines__find(&self->machines, pid);
+}
+
+static inline
+struct machine *perf_session__findnew_machine(struct perf_session *self, pid_t pid)
+{
+	if (pid == HOST_KERNEL_ID)
+		return &self->host_machine;
+	return machines__findnew(&self->machines, pid);
+}
+
+static inline
+void perf_session__process_machines(struct perf_session *self,
+				    machine__process_t process)
+{
+	process(&self->host_machine, self);
+	return machines__process(&self->machines, process, self);
+}
+
+size_t perf_session__fprintf_dsos(struct perf_session *self, FILE *fp);
+
+size_t perf_session__fprintf_dsos_buildid(struct perf_session *self,
+					  FILE *fp, bool with_hits);
+
+size_t perf_session__fprintf_nr_events(struct perf_session *session, FILE *fp);
+
+static inline int perf_session__parse_sample(struct perf_session *session,
+					     const union perf_event *event,
+					     struct perf_sample *sample)
+{
+	return perf_event__parse_sample(event, session->sample_type,
+					session->sample_size,
+					session->sample_id_all, sample);
+}
+
+struct perf_evsel *perf_session__find_first_evtype(struct perf_session *session,
+					    unsigned int type);
+
+void perf_session__print_symbols(union perf_event *event,
+				 struct perf_sample *sample,
+				 struct perf_session *session);
+
+#endif /* __PERF_SESSION_H */
diff --git a/tools/perf/util/setup.py b/tools/perf/util/setup.py
new file mode 100644
index 00000000..bbc982f5
--- /dev/null
+++ b/tools/perf/util/setup.py
@@ -0,0 +1,24 @@
+#!/usr/bin/python2
+
+from distutils.core import setup, Extension
+from os import getenv
+
+cflags = ['-fno-strict-aliasing', '-Wno-write-strings']
+cflags += getenv('CFLAGS', '').split()
+
+perf = Extension('perf',
+		  sources = ['util/python.c', 'util/ctype.c', 'util/evlist.c',
+			     'util/evsel.c', 'util/cpumap.c', 'util/thread_map.c',
+			     'util/util.c', 'util/xyarray.c', 'util/cgroup.c'],
+		  include_dirs = ['util/include'],
+		  extra_compile_args = cflags,
+                 )
+
+setup(name='perf',
+      version='0.1',
+      description='Interface with the Linux profiling infrastructure',
+      author='Arnaldo Carvalho de Melo',
+      author_email='acme@redhat.com',
+      license='GPLv2',
+      url='http://perf.wiki.kernel.org',
+      ext_modules=[perf])
diff --git a/tools/perf/util/sigchain.c b/tools/perf/util/sigchain.c
new file mode 100644
index 00000000..ba785e9b
--- /dev/null
+++ b/tools/perf/util/sigchain.c
@@ -0,0 +1,52 @@
+#include "sigchain.h"
+#include "cache.h"
+
+#define SIGCHAIN_MAX_SIGNALS 32
+
+struct sigchain_signal {
+	sigchain_fun *old;
+	int n;
+	int alloc;
+};
+static struct sigchain_signal signals[SIGCHAIN_MAX_SIGNALS];
+
+static void check_signum(int sig)
+{
+	if (sig < 1 || sig >= SIGCHAIN_MAX_SIGNALS)
+		die("BUG: signal out of range: %d", sig);
+}
+
+static int sigchain_push(int sig, sigchain_fun f)
+{
+	struct sigchain_signal *s = signals + sig;
+	check_signum(sig);
+
+	ALLOC_GROW(s->old, s->n + 1, s->alloc);
+	s->old[s->n] = signal(sig, f);
+	if (s->old[s->n] == SIG_ERR)
+		return -1;
+	s->n++;
+	return 0;
+}
+
+int sigchain_pop(int sig)
+{
+	struct sigchain_signal *s = signals + sig;
+	check_signum(sig);
+	if (s->n < 1)
+		return 0;
+
+	if (signal(sig, s->old[s->n - 1]) == SIG_ERR)
+		return -1;
+	s->n--;
+	return 0;
+}
+
+void sigchain_push_common(sigchain_fun f)
+{
+	sigchain_push(SIGINT, f);
+	sigchain_push(SIGHUP, f);
+	sigchain_push(SIGTERM, f);
+	sigchain_push(SIGQUIT, f);
+	sigchain_push(SIGPIPE, f);
+}
diff --git a/tools/perf/util/sigchain.h b/tools/perf/util/sigchain.h
new file mode 100644
index 00000000..959d64eb
--- /dev/null
+++ b/tools/perf/util/sigchain.h
@@ -0,0 +1,10 @@
+#ifndef __PERF_SIGCHAIN_H
+#define __PERF_SIGCHAIN_H
+
+typedef void (*sigchain_fun)(int);
+
+int sigchain_pop(int sig);
+
+void sigchain_push_common(sigchain_fun f);
+
+#endif /* __PERF_SIGCHAIN_H */
diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c
new file mode 100644
index 00000000..0ffbc8e5
--- /dev/null
+++ b/tools/perf/util/sort.c
@@ -0,0 +1,348 @@
+#include "sort.h"
+#include "hist.h"
+
+regex_t		parent_regex;
+const char	default_parent_pattern[] = "^sys_|^do_page_fault";
+const char	*parent_pattern = default_parent_pattern;
+const char	default_sort_order[] = "comm,dso,symbol";
+const char	*sort_order = default_sort_order;
+int		sort__need_collapse = 0;
+int		sort__has_parent = 0;
+
+enum sort_type	sort__first_dimension;
+
+char * field_sep;
+
+LIST_HEAD(hist_entry__sort_list);
+
+static int hist_entry__thread_snprintf(struct hist_entry *self, char *bf,
+				       size_t size, unsigned int width);
+static int hist_entry__comm_snprintf(struct hist_entry *self, char *bf,
+				     size_t size, unsigned int width);
+static int hist_entry__dso_snprintf(struct hist_entry *self, char *bf,
+				    size_t size, unsigned int width);
+static int hist_entry__sym_snprintf(struct hist_entry *self, char *bf,
+				    size_t size, unsigned int width);
+static int hist_entry__parent_snprintf(struct hist_entry *self, char *bf,
+				       size_t size, unsigned int width);
+static int hist_entry__cpu_snprintf(struct hist_entry *self, char *bf,
+				    size_t size, unsigned int width);
+
+struct sort_entry sort_thread = {
+	.se_header	= "Command:  Pid",
+	.se_cmp		= sort__thread_cmp,
+	.se_snprintf	= hist_entry__thread_snprintf,
+	.se_width_idx	= HISTC_THREAD,
+};
+
+struct sort_entry sort_comm = {
+	.se_header	= "Command",
+	.se_cmp		= sort__comm_cmp,
+	.se_collapse	= sort__comm_collapse,
+	.se_snprintf	= hist_entry__comm_snprintf,
+	.se_width_idx	= HISTC_COMM,
+};
+
+struct sort_entry sort_dso = {
+	.se_header	= "Shared Object",
+	.se_cmp		= sort__dso_cmp,
+	.se_snprintf	= hist_entry__dso_snprintf,
+	.se_width_idx	= HISTC_DSO,
+};
+
+struct sort_entry sort_sym = {
+	.se_header	= "Symbol",
+	.se_cmp		= sort__sym_cmp,
+	.se_snprintf	= hist_entry__sym_snprintf,
+	.se_width_idx	= HISTC_SYMBOL,
+};
+
+struct sort_entry sort_parent = {
+	.se_header	= "Parent symbol",
+	.se_cmp		= sort__parent_cmp,
+	.se_snprintf	= hist_entry__parent_snprintf,
+	.se_width_idx	= HISTC_PARENT,
+};
+ 
+struct sort_entry sort_cpu = {
+	.se_header      = "CPU",
+	.se_cmp	        = sort__cpu_cmp,
+	.se_snprintf    = hist_entry__cpu_snprintf,
+	.se_width_idx	= HISTC_CPU,
+};
+
+struct sort_dimension {
+	const char		*name;
+	struct sort_entry	*entry;
+	int			taken;
+};
+
+static struct sort_dimension sort_dimensions[] = {
+	{ .name = "pid",	.entry = &sort_thread,	},
+	{ .name = "comm",	.entry = &sort_comm,	},
+	{ .name = "dso",	.entry = &sort_dso,	},
+	{ .name = "symbol",	.entry = &sort_sym,	},
+	{ .name = "parent",	.entry = &sort_parent,	},
+	{ .name = "cpu",	.entry = &sort_cpu,	},
+};
+
+int64_t cmp_null(void *l, void *r)
+{
+	if (!l && !r)
+		return 0;
+	else if (!l)
+		return -1;
+	else
+		return 1;
+}
+
+/* --sort pid */
+
+int64_t
+sort__thread_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	return right->thread->pid - left->thread->pid;
+}
+
+static int repsep_snprintf(char *bf, size_t size, const char *fmt, ...)
+{
+	int n;
+	va_list ap;
+
+	va_start(ap, fmt);
+	n = vsnprintf(bf, size, fmt, ap);
+	if (field_sep && n > 0) {
+		char *sep = bf;
+
+		while (1) {
+			sep = strchr(sep, *field_sep);
+			if (sep == NULL)
+				break;
+			*sep = '.';
+		}
+	}
+	va_end(ap);
+
+	if (n >= (int)size)
+		return size - 1;
+	return n;
+}
+
+static int hist_entry__thread_snprintf(struct hist_entry *self, char *bf,
+				       size_t size, unsigned int width)
+{
+	return repsep_snprintf(bf, size, "%*s:%5d", width,
+			      self->thread->comm ?: "", self->thread->pid);
+}
+
+static int hist_entry__comm_snprintf(struct hist_entry *self, char *bf,
+				     size_t size, unsigned int width)
+{
+	return repsep_snprintf(bf, size, "%*s", width, self->thread->comm);
+}
+
+/* --sort dso */
+
+int64_t
+sort__dso_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	struct dso *dso_l = left->ms.map ? left->ms.map->dso : NULL;
+	struct dso *dso_r = right->ms.map ? right->ms.map->dso : NULL;
+	const char *dso_name_l, *dso_name_r;
+
+	if (!dso_l || !dso_r)
+		return cmp_null(dso_l, dso_r);
+
+	if (verbose) {
+		dso_name_l = dso_l->long_name;
+		dso_name_r = dso_r->long_name;
+	} else {
+		dso_name_l = dso_l->short_name;
+		dso_name_r = dso_r->short_name;
+	}
+
+	return strcmp(dso_name_l, dso_name_r);
+}
+
+static int hist_entry__dso_snprintf(struct hist_entry *self, char *bf,
+				    size_t size, unsigned int width)
+{
+	if (self->ms.map && self->ms.map->dso) {
+		const char *dso_name = !verbose ? self->ms.map->dso->short_name :
+						  self->ms.map->dso->long_name;
+		return repsep_snprintf(bf, size, "%-*s", width, dso_name);
+	}
+
+	return repsep_snprintf(bf, size, "%-*s", width, "[unknown]");
+}
+
+/* --sort symbol */
+
+int64_t
+sort__sym_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	u64 ip_l, ip_r;
+
+	if (left->ms.sym == right->ms.sym)
+		return 0;
+
+	ip_l = left->ms.sym ? left->ms.sym->start : left->ip;
+	ip_r = right->ms.sym ? right->ms.sym->start : right->ip;
+
+	return (int64_t)(ip_r - ip_l);
+}
+
+static int hist_entry__sym_snprintf(struct hist_entry *self, char *bf,
+				    size_t size, unsigned int width __used)
+{
+	size_t ret = 0;
+
+	if (verbose) {
+		char o = self->ms.map ? dso__symtab_origin(self->ms.map->dso) : '!';
+		ret += repsep_snprintf(bf, size, "%-#*llx %c ",
+				       BITS_PER_LONG / 4, self->ip, o);
+	}
+
+	ret += repsep_snprintf(bf + ret, size - ret, "[%c] ", self->level);
+	if (self->ms.sym)
+		ret += repsep_snprintf(bf + ret, size - ret, "%s",
+				       self->ms.sym->name);
+	else
+		ret += repsep_snprintf(bf + ret, size - ret, "%-#*llx",
+				       BITS_PER_LONG / 4, self->ip);
+
+	return ret;
+}
+
+/* --sort comm */
+
+int64_t
+sort__comm_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	return right->thread->pid - left->thread->pid;
+}
+
+int64_t
+sort__comm_collapse(struct hist_entry *left, struct hist_entry *right)
+{
+	char *comm_l = left->thread->comm;
+	char *comm_r = right->thread->comm;
+
+	if (!comm_l || !comm_r)
+		return cmp_null(comm_l, comm_r);
+
+	return strcmp(comm_l, comm_r);
+}
+
+/* --sort parent */
+
+int64_t
+sort__parent_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	struct symbol *sym_l = left->parent;
+	struct symbol *sym_r = right->parent;
+
+	if (!sym_l || !sym_r)
+		return cmp_null(sym_l, sym_r);
+
+	return strcmp(sym_l->name, sym_r->name);
+}
+
+static int hist_entry__parent_snprintf(struct hist_entry *self, char *bf,
+				       size_t size, unsigned int width)
+{
+	return repsep_snprintf(bf, size, "%-*s", width,
+			      self->parent ? self->parent->name : "[other]");
+}
+
+/* --sort cpu */
+
+int64_t
+sort__cpu_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	return right->cpu - left->cpu;
+}
+
+static int hist_entry__cpu_snprintf(struct hist_entry *self, char *bf,
+				       size_t size, unsigned int width)
+{
+	return repsep_snprintf(bf, size, "%-*d", width, self->cpu);
+}
+
+int sort_dimension__add(const char *tok)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(sort_dimensions); i++) {
+		struct sort_dimension *sd = &sort_dimensions[i];
+
+		if (sd->taken)
+			continue;
+
+		if (strncasecmp(tok, sd->name, strlen(tok)))
+			continue;
+
+		if (sd->entry->se_collapse)
+			sort__need_collapse = 1;
+
+		if (sd->entry == &sort_parent) {
+			int ret = regcomp(&parent_regex, parent_pattern, REG_EXTENDED);
+			if (ret) {
+				char err[BUFSIZ];
+
+				regerror(ret, &parent_regex, err, sizeof(err));
+				pr_err("Invalid regex: %s\n%s", parent_pattern, err);
+				return -EINVAL;
+			}
+			sort__has_parent = 1;
+		}
+
+		if (list_empty(&hist_entry__sort_list)) {
+			if (!strcmp(sd->name, "pid"))
+				sort__first_dimension = SORT_PID;
+			else if (!strcmp(sd->name, "comm"))
+				sort__first_dimension = SORT_COMM;
+			else if (!strcmp(sd->name, "dso"))
+				sort__first_dimension = SORT_DSO;
+			else if (!strcmp(sd->name, "symbol"))
+				sort__first_dimension = SORT_SYM;
+			else if (!strcmp(sd->name, "parent"))
+				sort__first_dimension = SORT_PARENT;
+			else if (!strcmp(sd->name, "cpu"))
+				sort__first_dimension = SORT_CPU;
+		}
+
+		list_add_tail(&sd->entry->list, &hist_entry__sort_list);
+		sd->taken = 1;
+
+		return 0;
+	}
+
+	return -ESRCH;
+}
+
+void setup_sorting(const char * const usagestr[], const struct option *opts)
+{
+	char *tmp, *tok, *str = strdup(sort_order);
+
+	for (tok = strtok_r(str, ", ", &tmp);
+			tok; tok = strtok_r(NULL, ", ", &tmp)) {
+		if (sort_dimension__add(tok) < 0) {
+			error("Unknown --sort key: `%s'", tok);
+			usage_with_options(usagestr, opts);
+		}
+	}
+
+	free(str);
+}
+
+void sort_entry__setup_elide(struct sort_entry *self, struct strlist *list,
+			     const char *list_name, FILE *fp)
+{
+	if (list && strlist__nr_entries(list) == 1) {
+		if (fp != NULL)
+			fprintf(fp, "# %s: %s\n", list_name,
+				strlist__entry(list, 0)->s);
+		self->elide = true;
+	}
+}
diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h
new file mode 100644
index 00000000..0b91053a
--- /dev/null
+++ b/tools/perf/util/sort.h
@@ -0,0 +1,124 @@
+#ifndef __PERF_SORT_H
+#define __PERF_SORT_H
+#include "../builtin.h"
+
+#include "util.h"
+
+#include "color.h"
+#include <linux/list.h>
+#include "cache.h"
+#include <linux/rbtree.h>
+#include "symbol.h"
+#include "string.h"
+#include "callchain.h"
+#include "strlist.h"
+#include "values.h"
+
+#include "../perf.h"
+#include "debug.h"
+#include "header.h"
+
+#include "parse-options.h"
+#include "parse-events.h"
+
+#include "thread.h"
+#include "sort.h"
+
+extern regex_t parent_regex;
+extern const char *sort_order;
+extern const char default_parent_pattern[];
+extern const char *parent_pattern;
+extern const char default_sort_order[];
+extern int sort__need_collapse;
+extern int sort__has_parent;
+extern char *field_sep;
+extern struct sort_entry sort_comm;
+extern struct sort_entry sort_dso;
+extern struct sort_entry sort_sym;
+extern struct sort_entry sort_parent;
+extern enum sort_type sort__first_dimension;
+
+/**
+ * struct hist_entry - histogram entry
+ *
+ * @row_offset - offset from the first callchain expanded to appear on screen
+ * @nr_rows - rows expanded in callchain, recalculated on folding/unfolding
+ */
+struct hist_entry {
+	struct rb_node		rb_node;
+	u64			period;
+	u64			period_sys;
+	u64			period_us;
+	u64			period_guest_sys;
+	u64			period_guest_us;
+	struct map_symbol	ms;
+	struct thread		*thread;
+	u64			ip;
+	s32			cpu;
+	u32			nr_events;
+
+	/* XXX These two should move to some tree widget lib */
+	u16			row_offset;
+	u16			nr_rows;
+
+	bool			init_have_children;
+	char			level;
+	u8			filtered;
+	struct symbol		*parent;
+	union {
+		unsigned long	  position;
+		struct hist_entry *pair;
+		struct rb_root	  sorted_chain;
+	};
+	struct callchain_root	callchain[0];
+};
+
+enum sort_type {
+	SORT_PID,
+	SORT_COMM,
+	SORT_DSO,
+	SORT_SYM,
+	SORT_PARENT,
+	SORT_CPU,
+};
+
+/*
+ * configurable sorting bits
+ */
+
+struct sort_entry {
+	struct list_head list;
+
+	const char *se_header;
+
+	int64_t (*se_cmp)(struct hist_entry *, struct hist_entry *);
+	int64_t (*se_collapse)(struct hist_entry *, struct hist_entry *);
+	int	(*se_snprintf)(struct hist_entry *self, char *bf, size_t size,
+			       unsigned int width);
+	u8	se_width_idx;
+	bool	elide;
+};
+
+extern struct sort_entry sort_thread;
+extern struct list_head hist_entry__sort_list;
+
+void setup_sorting(const char * const usagestr[], const struct option *opts);
+
+extern size_t sort__thread_print(FILE *, struct hist_entry *, unsigned int);
+extern size_t sort__comm_print(FILE *, struct hist_entry *, unsigned int);
+extern size_t sort__dso_print(FILE *, struct hist_entry *, unsigned int);
+extern size_t sort__sym_print(FILE *, struct hist_entry *, unsigned int __used);
+extern int64_t cmp_null(void *, void *);
+extern int64_t sort__thread_cmp(struct hist_entry *, struct hist_entry *);
+extern int64_t sort__comm_cmp(struct hist_entry *, struct hist_entry *);
+extern int64_t sort__comm_collapse(struct hist_entry *, struct hist_entry *);
+extern int64_t sort__dso_cmp(struct hist_entry *, struct hist_entry *);
+extern int64_t sort__sym_cmp(struct hist_entry *, struct hist_entry *);
+extern int64_t sort__parent_cmp(struct hist_entry *, struct hist_entry *);
+int64_t sort__cpu_cmp(struct hist_entry *left, struct hist_entry *right);
+extern size_t sort__parent_print(FILE *, struct hist_entry *, unsigned int);
+extern int sort_dimension__add(const char *);
+void sort_entry__setup_elide(struct sort_entry *self, struct strlist *list,
+			     const char *list_name, FILE *fp);
+
+#endif	/* __PERF_SORT_H */
diff --git a/tools/perf/util/strbuf.c b/tools/perf/util/strbuf.c
new file mode 100644
index 00000000..92e06851
--- /dev/null
+++ b/tools/perf/util/strbuf.c
@@ -0,0 +1,133 @@
+#include "cache.h"
+
+int prefixcmp(const char *str, const char *prefix)
+{
+	for (; ; str++, prefix++)
+		if (!*prefix)
+			return 0;
+		else if (*str != *prefix)
+			return (unsigned char)*prefix - (unsigned char)*str;
+}
+
+/*
+ * Used as the default ->buf value, so that people can always assume
+ * buf is non NULL and ->buf is NUL terminated even for a freshly
+ * initialized strbuf.
+ */
+char strbuf_slopbuf[1];
+
+void strbuf_init(struct strbuf *sb, ssize_t hint)
+{
+	sb->alloc = sb->len = 0;
+	sb->buf = strbuf_slopbuf;
+	if (hint)
+		strbuf_grow(sb, hint);
+}
+
+void strbuf_release(struct strbuf *sb)
+{
+	if (sb->alloc) {
+		free(sb->buf);
+		strbuf_init(sb, 0);
+	}
+}
+
+char *strbuf_detach(struct strbuf *sb, size_t *sz)
+{
+	char *res = sb->alloc ? sb->buf : NULL;
+	if (sz)
+		*sz = sb->len;
+	strbuf_init(sb, 0);
+	return res;
+}
+
+void strbuf_grow(struct strbuf *sb, size_t extra)
+{
+	if (sb->len + extra + 1 <= sb->len)
+		die("you want to use way too much memory");
+	if (!sb->alloc)
+		sb->buf = NULL;
+	ALLOC_GROW(sb->buf, sb->len + extra + 1, sb->alloc);
+}
+
+static void strbuf_splice(struct strbuf *sb, size_t pos, size_t len,
+				   const void *data, size_t dlen)
+{
+	if (pos + len < pos)
+		die("you want to use way too much memory");
+	if (pos > sb->len)
+		die("`pos' is too far after the end of the buffer");
+	if (pos + len > sb->len)
+		die("`pos + len' is too far after the end of the buffer");
+
+	if (dlen >= len)
+		strbuf_grow(sb, dlen - len);
+	memmove(sb->buf + pos + dlen,
+			sb->buf + pos + len,
+			sb->len - pos - len);
+	memcpy(sb->buf + pos, data, dlen);
+	strbuf_setlen(sb, sb->len + dlen - len);
+}
+
+void strbuf_remove(struct strbuf *sb, size_t pos, size_t len)
+{
+	strbuf_splice(sb, pos, len, NULL, 0);
+}
+
+void strbuf_add(struct strbuf *sb, const void *data, size_t len)
+{
+	strbuf_grow(sb, len);
+	memcpy(sb->buf + sb->len, data, len);
+	strbuf_setlen(sb, sb->len + len);
+}
+
+void strbuf_addf(struct strbuf *sb, const char *fmt, ...)
+{
+	int len;
+	va_list ap;
+
+	if (!strbuf_avail(sb))
+		strbuf_grow(sb, 64);
+	va_start(ap, fmt);
+	len = vsnprintf(sb->buf + sb->len, sb->alloc - sb->len, fmt, ap);
+	va_end(ap);
+	if (len < 0)
+		die("your vsnprintf is broken");
+	if (len > strbuf_avail(sb)) {
+		strbuf_grow(sb, len);
+		va_start(ap, fmt);
+		len = vsnprintf(sb->buf + sb->len, sb->alloc - sb->len, fmt, ap);
+		va_end(ap);
+		if (len > strbuf_avail(sb)) {
+			die("this should not happen, your snprintf is broken");
+		}
+	}
+	strbuf_setlen(sb, sb->len + len);
+}
+
+ssize_t strbuf_read(struct strbuf *sb, int fd, ssize_t hint)
+{
+	size_t oldlen = sb->len;
+	size_t oldalloc = sb->alloc;
+
+	strbuf_grow(sb, hint ? hint : 8192);
+	for (;;) {
+		ssize_t cnt;
+
+		cnt = read(fd, sb->buf + sb->len, sb->alloc - sb->len - 1);
+		if (cnt < 0) {
+			if (oldalloc == 0)
+				strbuf_release(sb);
+			else
+				strbuf_setlen(sb, oldlen);
+			return -1;
+		}
+		if (!cnt)
+			break;
+		sb->len += cnt;
+		strbuf_grow(sb, 8192);
+	}
+
+	sb->buf[sb->len] = '\0';
+	return sb->len - oldlen;
+}
diff --git a/tools/perf/util/strbuf.h b/tools/perf/util/strbuf.h
new file mode 100644
index 00000000..436ac319
--- /dev/null
+++ b/tools/perf/util/strbuf.h
@@ -0,0 +1,92 @@
+#ifndef __PERF_STRBUF_H
+#define __PERF_STRBUF_H
+
+/*
+ * Strbuf's can be use in many ways: as a byte array, or to store arbitrary
+ * long, overflow safe strings.
+ *
+ * Strbufs has some invariants that are very important to keep in mind:
+ *
+ * 1. the ->buf member is always malloc-ed, hence strbuf's can be used to
+ *    build complex strings/buffers whose final size isn't easily known.
+ *
+ *    It is NOT legal to copy the ->buf pointer away.
+ *    `strbuf_detach' is the operation that detachs a buffer from its shell
+ *    while keeping the shell valid wrt its invariants.
+ *
+ * 2. the ->buf member is a byte array that has at least ->len + 1 bytes
+ *    allocated. The extra byte is used to store a '\0', allowing the ->buf
+ *    member to be a valid C-string. Every strbuf function ensure this
+ *    invariant is preserved.
+ *
+ *    Note that it is OK to "play" with the buffer directly if you work it
+ *    that way:
+ *
+ *    strbuf_grow(sb, SOME_SIZE);
+ *       ... Here, the memory array starting at sb->buf, and of length
+ *       ... strbuf_avail(sb) is all yours, and you are sure that
+ *       ... strbuf_avail(sb) is at least SOME_SIZE.
+ *    strbuf_setlen(sb, sb->len + SOME_OTHER_SIZE);
+ *
+ *    Of course, SOME_OTHER_SIZE must be smaller or equal to strbuf_avail(sb).
+ *
+ *    Doing so is safe, though if it has to be done in many places, adding the
+ *    missing API to the strbuf module is the way to go.
+ *
+ *    XXX: do _not_ assume that the area that is yours is of size ->alloc - 1
+ *         even if it's true in the current implementation. Alloc is somehow a
+ *         "private" member that should not be messed with.
+ */
+
+#include <assert.h>
+
+extern char strbuf_slopbuf[];
+struct strbuf {
+	size_t alloc;
+	size_t len;
+	char *buf;
+};
+
+#define STRBUF_INIT  { 0, 0, strbuf_slopbuf }
+
+/*----- strbuf life cycle -----*/
+extern void strbuf_init(struct strbuf *buf, ssize_t hint);
+extern void strbuf_release(struct strbuf *);
+extern char *strbuf_detach(struct strbuf *, size_t *);
+
+/*----- strbuf size related -----*/
+static inline ssize_t strbuf_avail(const struct strbuf *sb) {
+	return sb->alloc ? sb->alloc - sb->len - 1 : 0;
+}
+
+extern void strbuf_grow(struct strbuf *, size_t);
+
+static inline void strbuf_setlen(struct strbuf *sb, size_t len) {
+	if (!sb->alloc)
+		strbuf_grow(sb, 0);
+	assert(len < sb->alloc);
+	sb->len = len;
+	sb->buf[len] = '\0';
+}
+
+/*----- add data in your buffer -----*/
+static inline void strbuf_addch(struct strbuf *sb, int c) {
+	strbuf_grow(sb, 1);
+	sb->buf[sb->len++] = c;
+	sb->buf[sb->len] = '\0';
+}
+
+extern void strbuf_remove(struct strbuf *, size_t pos, size_t len);
+
+extern void strbuf_add(struct strbuf *, const void *, size_t);
+static inline void strbuf_addstr(struct strbuf *sb, const char *s) {
+	strbuf_add(sb, s, strlen(s));
+}
+
+__attribute__((format(printf,2,3)))
+extern void strbuf_addf(struct strbuf *sb, const char *fmt, ...);
+
+/* XXX: if read fails, any partial read is undone */
+extern ssize_t strbuf_read(struct strbuf *, int fd, ssize_t hint);
+
+#endif /* __PERF_STRBUF_H */
diff --git a/tools/perf/util/strfilter.c b/tools/perf/util/strfilter.c
new file mode 100644
index 00000000..834c8ebf
--- /dev/null
+++ b/tools/perf/util/strfilter.c
@@ -0,0 +1,199 @@
+#include "util.h"
+#include "string.h"
+#include "strfilter.h"
+
+/* Operators */
+static const char *OP_and	= "&";	/* Logical AND */
+static const char *OP_or	= "|";	/* Logical OR */
+static const char *OP_not	= "!";	/* Logical NOT */
+
+#define is_operator(c)	((c) == '|' || (c) == '&' || (c) == '!')
+#define is_separator(c)	(is_operator(c) || (c) == '(' || (c) == ')')
+
+static void strfilter_node__delete(struct strfilter_node *self)
+{
+	if (self) {
+		if (self->p && !is_operator(*self->p))
+			free((char *)self->p);
+		strfilter_node__delete(self->l);
+		strfilter_node__delete(self->r);
+		free(self);
+	}
+}
+
+void strfilter__delete(struct strfilter *self)
+{
+	if (self) {
+		strfilter_node__delete(self->root);
+		free(self);
+	}
+}
+
+static const char *get_token(const char *s, const char **e)
+{
+	const char *p;
+
+	while (isspace(*s))	/* Skip spaces */
+		s++;
+
+	if (*s == '\0') {
+		p = s;
+		goto end;
+	}
+
+	p = s + 1;
+	if (!is_separator(*s)) {
+		/* End search */
+retry:
+		while (*p && !is_separator(*p) && !isspace(*p))
+			p++;
+		/* Escape and special case: '!' is also used in glob pattern */
+		if (*(p - 1) == '\\' || (*p == '!' && *(p - 1) == '[')) {
+			p++;
+			goto retry;
+		}
+	}
+end:
+	*e = p;
+	return s;
+}
+
+static struct strfilter_node *strfilter_node__alloc(const char *op,
+						    struct strfilter_node *l,
+						    struct strfilter_node *r)
+{
+	struct strfilter_node *ret = zalloc(sizeof(struct strfilter_node));
+
+	if (ret) {
+		ret->p = op;
+		ret->l = l;
+		ret->r = r;
+	}
+
+	return ret;
+}
+
+static struct strfilter_node *strfilter_node__new(const char *s,
+						  const char **ep)
+{
+	struct strfilter_node root, *cur, *last_op;
+	const char *e;
+
+	if (!s)
+		return NULL;
+
+	memset(&root, 0, sizeof(root));
+	last_op = cur = &root;
+
+	s = get_token(s, &e);
+	while (*s != '\0' && *s != ')') {
+		switch (*s) {
+		case '&':	/* Exchg last OP->r with AND */
+			if (!cur->r || !last_op->r)
+				goto error;
+			cur = strfilter_node__alloc(OP_and, last_op->r, NULL);
+			if (!cur)
+				goto nomem;
+			last_op->r = cur;
+			last_op = cur;
+			break;
+		case '|':	/* Exchg the root with OR */
+			if (!cur->r || !root.r)
+				goto error;
+			cur = strfilter_node__alloc(OP_or, root.r, NULL);
+			if (!cur)
+				goto nomem;
+			root.r = cur;
+			last_op = cur;
+			break;
+		case '!':	/* Add NOT as a leaf node */
+			if (cur->r)
+				goto error;
+			cur->r = strfilter_node__alloc(OP_not, NULL, NULL);
+			if (!cur->r)
+				goto nomem;
+			cur = cur->r;
+			break;
+		case '(':	/* Recursively parses inside the parenthesis */
+			if (cur->r)
+				goto error;
+			cur->r = strfilter_node__new(s + 1, &s);
+			if (!s)
+				goto nomem;
+			if (!cur->r || *s != ')')
+				goto error;
+			e = s + 1;
+			break;
+		default:
+			if (cur->r)
+				goto error;
+			cur->r = strfilter_node__alloc(NULL, NULL, NULL);
+			if (!cur->r)
+				goto nomem;
+			cur->r->p = strndup(s, e - s);
+			if (!cur->r->p)
+				goto nomem;
+		}
+		s = get_token(e, &e);
+	}
+	if (!cur->r)
+		goto error;
+	*ep = s;
+	return root.r;
+nomem:
+	s = NULL;
+error:
+	*ep = s;
+	strfilter_node__delete(root.r);
+	return NULL;
+}
+
+/*
+ * Parse filter rule and return new strfilter.
+ * Return NULL if fail, and *ep == NULL if memory allocation failed.
+ */
+struct strfilter *strfilter__new(const char *rules, const char **err)
+{
+	struct strfilter *ret = zalloc(sizeof(struct strfilter));
+	const char *ep = NULL;
+
+	if (ret)
+		ret->root = strfilter_node__new(rules, &ep);
+
+	if (!ret || !ret->root || *ep != '\0') {
+		if (err)
+			*err = ep;
+		strfilter__delete(ret);
+		ret = NULL;
+	}
+
+	return ret;
+}
+
+static bool strfilter_node__compare(struct strfilter_node *self,
+				    const char *str)
+{
+	if (!self || !self->p)
+		return false;
+
+	switch (*self->p) {
+	case '|':	/* OR */
+		return strfilter_node__compare(self->l, str) ||
+			strfilter_node__compare(self->r, str);
+	case '&':	/* AND */
+		return strfilter_node__compare(self->l, str) &&
+			strfilter_node__compare(self->r, str);
+	case '!':	/* NOT */
+		return !strfilter_node__compare(self->r, str);
+	default:
+		return strglobmatch(str, self->p);
+	}
+}
+
+/* Return true if STR matches the filter rules */
+bool strfilter__compare(struct strfilter *self, const char *str)
+{
+	if (!self)
+		return false;
+	return strfilter_node__compare(self->root, str);
+}
diff --git a/tools/perf/util/strfilter.h b/tools/perf/util/strfilter.h
new file mode 100644
index 00000000..00f58a75
--- /dev/null
+++ b/tools/perf/util/strfilter.h
@@ -0,0 +1,48 @@
+#ifndef __PERF_STRFILTER_H
+#define __PERF_STRFILTER_H
+/* General purpose glob matching filter */
+
+#include <linux/list.h>
+#include <stdbool.h>
+
+/* A node of string filter */
+struct strfilter_node {
+	struct strfilter_node *l;	/* Tree left branche (for &,|) */
+	struct strfilter_node *r;	/* Tree right branche (for !,&,|) */
+	const char *p;		/* Operator or rule */
+};
+
+/* String filter */
+struct strfilter {
+	struct strfilter_node *root;
+};
+
+/**
+ * strfilter__new - Create a new string filter
+ * @rules: Filter rule, which is a combination of glob expressions.
+ * @err: Pointer which points an error detected on @rules
+ *
+ * Parse @rules and return new strfilter. Return NULL if an error detected.
+ * In that case, *@err will indicate where it is detected, and *@err is NULL
+ * if a memory allocation is failed.
+ */
+struct strfilter *strfilter__new(const char *rules, const char **err);
+
+/**
+ * strfilter__compare - compare given string and a string filter
+ * @self: String filter
+ * @str: target string
+ *
+ * Compare @str and @self. Return true if the str match the rule
+ */
+bool strfilter__compare(struct strfilter *self, const char *str);
+
+/**
+ * strfilter__delete - delete a string filter
+ * @self: String filter to delete
+ *
+ * Delete @self.
+ */
+void strfilter__delete(struct strfilter *self);
+
+#endif
diff --git a/tools/perf/util/string.c b/tools/perf/util/string.c
new file mode 100644
index 00000000..b9a985da
--- /dev/null
+++ b/tools/perf/util/string.c
@@ -0,0 +1,296 @@
+#include "util.h"
+#include "string.h"
+
+#define K 1024LL
+/*
+ * perf_atoll()
+ * Parse (\d+)(b|B|kb|KB|mb|MB|gb|GB|tb|TB) (e.g. "256MB")
+ * and return its numeric value
+ */
+s64 perf_atoll(const char *str)
+{
+	unsigned int i;
+	s64 length = -1, unit = 1;
+
+	if (!isdigit(str[0]))
+		goto out_err;
+
+	for (i = 1; i < strlen(str); i++) {
+		switch (str[i]) {
+		case 'B':
+		case 'b':
+			break;
+		case 'K':
+			if (str[i + 1] != 'B')
+				goto out_err;
+			else
+				goto kilo;
+		case 'k':
+			if (str[i + 1] != 'b')
+				goto out_err;
+kilo:
+			unit = K;
+			break;
+		case 'M':
+			if (str[i + 1] != 'B')
+				goto out_err;
+			else
+				goto mega;
+		case 'm':
+			if (str[i + 1] != 'b')
+				goto out_err;
+mega:
+			unit = K * K;
+			break;
+		case 'G':
+			if (str[i + 1] != 'B')
+				goto out_err;
+			else
+				goto giga;
+		case 'g':
+			if (str[i + 1] != 'b')
+				goto out_err;
+giga:
+			unit = K * K * K;
+			break;
+		case 'T':
+			if (str[i + 1] != 'B')
+				goto out_err;
+			else
+				goto tera;
+		case 't':
+			if (str[i + 1] != 'b')
+				goto out_err;
+tera:
+			unit = K * K * K * K;
+			break;
+		case '\0':	/* only specified figures */
+			unit = 1;
+			break;
+		default:
+			if (!isdigit(str[i]))
+				goto out_err;
+			break;
+		}
+	}
+
+	length = atoll(str) * unit;
+	goto out;
+
+out_err:
+	length = -1;
+out:
+	return length;
+}
+
+/*
+ * Helper function for splitting a string into an argv-like array.
+ * originally copied from lib/argv_split.c
+ */
+static const char *skip_sep(const char *cp)
+{
+	while (*cp && isspace(*cp))
+		cp++;
+
+	return cp;
+}
+
+static const char *skip_arg(const char *cp)
+{
+	while (*cp && !isspace(*cp))
+		cp++;
+
+	return cp;
+}
+
+static int count_argc(const char *str)
+{
+	int count = 0;
+
+	while (*str) {
+		str = skip_sep(str);
+		if (*str) {
+			count++;
+			str = skip_arg(str);
+		}
+	}
+
+	return count;
+}
+
+/**
+ * argv_free - free an argv
+ * @argv - the argument vector to be freed
+ *
+ * Frees an argv and the strings it points to.
+ */
+void argv_free(char **argv)
+{
+	char **p;
+	for (p = argv; *p; p++)
+		free(*p);
+
+	free(argv);
+}
+
+/**
+ * argv_split - split a string at whitespace, returning an argv
+ * @str: the string to be split
+ * @argcp: returned argument count
+ *
+ * Returns an array of pointers to strings which are split out from
+ * @str.  This is performed by strictly splitting on white-space; no
+ * quote processing is performed.  Multiple whitespace characters are
+ * considered to be a single argument separator.  The returned array
+ * is always NULL-terminated.  Returns NULL on memory allocation
+ * failure.
+ */
+char **argv_split(const char *str, int *argcp)
+{
+	int argc = count_argc(str);
+	char **argv = zalloc(sizeof(*argv) * (argc+1));
+	char **argvp;
+
+	if (argv == NULL)
+		goto out;
+
+	if (argcp)
+		*argcp = argc;
+
+	argvp = argv;
+
+	while (*str) {
+		str = skip_sep(str);
+
+		if (*str) {
+			const char *p = str;
+			char *t;
+
+			str = skip_arg(str);
+
+			t = strndup(p, str-p);
+			if (t == NULL)
+				goto fail;
+			*argvp++ = t;
+		}
+	}
+	*argvp = NULL;
+
+out:
+	return argv;
+
+fail:
+	argv_free(argv);
+	return NULL;
+}
+
+/* Character class matching */
+static bool __match_charclass(const char *pat, char c, const char **npat)
+{
+	bool complement = false, ret = true;
+
+	if (*pat == '!') {
+		complement = true;
+		pat++;
+	}
+	if (*pat++ == c)	/* First character is special */
+		goto end;
+
+	while (*pat && *pat != ']') {	/* Matching */
+		if (*pat == '-' && *(pat + 1) != ']') {	/* Range */
+			if (*(pat - 1) <= c && c <= *(pat + 1))
+				goto end;
+			if (*(pat - 1) > *(pat + 1))
+				goto error;
+			pat += 2;
+		} else if (*pat++ == c)
+			goto end;
+	}
+	if (!*pat)
+		goto error;
+	ret = false;
+
+end:
+	while (*pat && *pat != ']')	/* Searching closing */
+		pat++;
+	if (!*pat)
+		goto error;
+	*npat = pat + 1;
+	return complement ? !ret : ret;
+
+error:
+	return false;
+}
+
+/* Glob/lazy pattern matching */
+static bool __match_glob(const char *str, const char *pat, bool ignore_space)
+{
+	while (*str && *pat && *pat != '*') {
+		if (ignore_space) {
+			/* Ignore spaces for lazy matching */
+			if (isspace(*str)) {
+				str++;
+				continue;
+			}
+			if (isspace(*pat)) {
+				pat++;
+				continue;
+			}
+		}
+		if (*pat == '?') {	/* Matches any single character */
+			str++;
+			pat++;
+			continue;
+		} else if (*pat == '[')	/* Character classes/Ranges */
+			if (__match_charclass(pat + 1, *str, &pat)) {
+				str++;
+				continue;
+			} else
+				return false;
+		else if (*pat == '\\') /* Escaped char match as normal char */
+			pat++;
+		if (*str++ != *pat++)
+			return false;
+	}
+	/* Check wild card */
+	if (*pat == '*') {
+		while (*pat == '*')
+			pat++;
+		if (!*pat)	/* Tail wild card matches all */
+			return true;
+		while (*str)
+			if (__match_glob(str++, pat, ignore_space))
+				return true;
+	}
+	return !*str && !*pat;
+}
+
+/**
+ * strglobmatch - glob expression pattern matching
+ * @str: the target string to match
+ * @pat: the pattern string to match
+ *
+ * This returns true if the @str matches @pat. @pat can includes wildcards
+ * ('*','?') and character classes ([CHARS], complementation and ranges are
+ * also supported). Also, this supports escape character ('\') to use special
+ * characters as normal character.
+ *
+ * Note: if @pat syntax is broken, this always returns false.
+ */
+bool strglobmatch(const char *str, const char *pat)
+{
+	return __match_glob(str, pat, false);
+}
+
+/**
+ * strlazymatch - matching pattern strings lazily with glob pattern
+ * @str: the target string to match
+ * @pat: the pattern string to match
+ *
+ * This is similar to strglobmatch, except this ignores spaces in
+ * the target string.
+ */
+bool strlazymatch(const char *str, const char *pat)
+{
+	return __match_glob(str, pat, true);
+}
diff --git a/tools/perf/util/strlist.c b/tools/perf/util/strlist.c
new file mode 100644
index 00000000..6783a204
--- /dev/null
+++ b/tools/perf/util/strlist.c
@@ -0,0 +1,200 @@
+/*
+ * (c) 2009 Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Licensed under the GPLv2.
+ */
+
+#include "strlist.h"
+#include <errno.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+static struct str_node *str_node__new(const char *s, bool dupstr)
+{
+	struct str_node *self = malloc(sizeof(*self));
+
+	if (self != NULL) {
+		if (dupstr) {
+			s = strdup(s);
+			if (s == NULL)
+				goto out_delete;
+		}
+		self->s = s;
+	}
+
+	return self;
+
+out_delete:
+	free(self);
+	return NULL;
+}
+
+static void str_node__delete(struct str_node *self, bool dupstr)
+{
+	if (dupstr)
+		free((void *)self->s);
+	free(self);
+}
+
+int strlist__add(struct strlist *self, const char *new_entry)
+{
+	struct rb_node **p = &self->entries.rb_node;
+	struct rb_node *parent = NULL;
+	struct str_node *sn;
+
+	while (*p != NULL) {
+		int rc;
+
+		parent = *p;
+		sn = rb_entry(parent, struct str_node, rb_node);
+		rc = strcmp(sn->s, new_entry);
+
+		if (rc > 0)
+			p = &(*p)->rb_left;
+		else if (rc < 0)
+			p = &(*p)->rb_right;
+		else
+			return -EEXIST;
+	}
+
+	sn = str_node__new(new_entry, self->dupstr);
+	if (sn == NULL)
+		return -ENOMEM;
+
+	rb_link_node(&sn->rb_node, parent, p);
+	rb_insert_color(&sn->rb_node, &self->entries);
+	++self->nr_entries;
+
+	return 0;
+}
+
+int strlist__load(struct strlist *self, const char *filename)
+{
+	char entry[1024];
+	int err;
+	FILE *fp = fopen(filename, "r");
+
+	if (fp == NULL)
+		return errno;
+
+	while (fgets(entry, sizeof(entry), fp) != NULL) {
+		const size_t len = strlen(entry);
+
+		if (len == 0)
+			continue;
+		entry[len - 1] = '\0';
+
+		err = strlist__add(self, entry);
+		if (err != 0)
+			goto out;
+	}
+
+	err = 0;
+out:
+	fclose(fp);
+	return err;
+}
+
+void strlist__remove(struct strlist *self, struct str_node *sn)
+{
+	rb_erase(&sn->rb_node, &self->entries);
+	str_node__delete(sn, self->dupstr);
+}
+
+struct str_node *strlist__find(struct strlist *self, const char *entry)
+{
+	struct rb_node **p = &self->entries.rb_node;
+	struct rb_node *parent = NULL;
+
+	while (*p != NULL) {
+		struct str_node *sn;
+		int rc;
+
+		parent = *p;
+		sn = rb_entry(parent, struct str_node, rb_node);
+		rc = strcmp(sn->s, entry);
+
+		if (rc > 0)
+			p = &(*p)->rb_left;
+		else if (rc < 0)
+			p = &(*p)->rb_right;
+		else
+			return sn;
+	}
+
+	return NULL;
+}
+
+static int strlist__parse_list_entry(struct strlist *self, const char *s)
+{
+	if (strncmp(s, "file://", 7) == 0)
+		return strlist__load(self, s + 7);
+
+	return strlist__add(self, s);
+}
+
+int strlist__parse_list(struct strlist *self, const char *s)
+{
+	char *sep;
+	int err;
+
+	while ((sep = strchr(s, ',')) != NULL) {
+		*sep = '\0';
+		err = strlist__parse_list_entry(self, s);
+		*sep = ',';
+		if (err != 0)
+			return err;
+		s = sep + 1;
+	}
+
+	return *s ? strlist__parse_list_entry(self, s) : 0;
+}
+
+struct strlist *strlist__new(bool dupstr, const char *slist)
+{
+	struct strlist *self = malloc(sizeof(*self));
+
+	if (self != NULL) {
+		self->entries	 = RB_ROOT;
+		self->dupstr	 = dupstr;
+		self->nr_entries = 0;
+		if (slist && strlist__parse_list(self, slist) != 0)
+			goto out_error;
+	}
+
+	return self;
+out_error:
+	free(self);
+	return NULL;
+}
+
+void strlist__delete(struct strlist *self)
+{
+	if (self != NULL) {
+		struct str_node *pos;
+		struct rb_node *next = rb_first(&self->entries);
+
+		while (next) {
+			pos = rb_entry(next, struct str_node, rb_node);
+			next = rb_next(&pos->rb_node);
+			strlist__remove(self, pos);
+		}
+		self->entries = RB_ROOT;
+		free(self);
+	}
+}
+
+struct str_node *strlist__entry(const struct strlist *self, unsigned int idx)
+{
+	struct rb_node *nd;
+
+	for (nd = rb_first(&self->entries); nd; nd = rb_next(nd)) {
+		struct str_node *pos = rb_entry(nd, struct str_node, rb_node);
+
+		if (!idx--)
+			return pos;
+	}
+
+	return NULL;
+}
diff --git a/tools/perf/util/strlist.h b/tools/perf/util/strlist.h
new file mode 100644
index 00000000..3ba83900
--- /dev/null
+++ b/tools/perf/util/strlist.h
@@ -0,0 +1,78 @@
+#ifndef __PERF_STRLIST_H
+#define __PERF_STRLIST_H
+
+#include <linux/rbtree.h>
+#include <stdbool.h>
+
+struct str_node {
+	struct rb_node rb_node;
+	const char     *s;
+};
+
+struct strlist {
+	struct rb_root entries;
+	unsigned int   nr_entries;
+	bool	       dupstr;
+};
+
+struct strlist *strlist__new(bool dupstr, const char *slist);
+void strlist__delete(struct strlist *self);
+
+void strlist__remove(struct strlist *self, struct str_node *sn);
+int strlist__load(struct strlist *self, const char *filename);
+int strlist__add(struct strlist *self, const char *str);
+
+struct str_node *strlist__entry(const struct strlist *self, unsigned int idx);
+struct str_node *strlist__find(struct strlist *self, const char *entry);
+
+static inline bool strlist__has_entry(struct strlist *self, const char *entry)
+{
+	return strlist__find(self, entry) != NULL;
+}
+
+static inline bool strlist__empty(const struct strlist *self)
+{
+	return self->nr_entries == 0;
+}
+
+static inline unsigned int strlist__nr_entries(const struct strlist *self)
+{
+	return self->nr_entries;
+}
+
+/* For strlist iteration */
+static inline struct str_node *strlist__first(struct strlist *self)
+{
+	struct rb_node *rn = rb_first(&self->entries);
+	return rn ? rb_entry(rn, struct str_node, rb_node) : NULL;
+}
+static inline struct str_node *strlist__next(struct str_node *sn)
+{
+	struct rb_node *rn;
+	if (!sn)
+		return NULL;
+	rn = rb_next(&sn->rb_node);
+	return rn ? rb_entry(rn, struct str_node, rb_node) : NULL;
+}
+
+/**
+ * strlist_for_each      - iterate over a strlist
+ * @pos:	the &struct str_node to use as a loop cursor.
+ * @self:	the &struct strlist for loop.
+ */
+#define strlist__for_each(pos, self)	\
+	for (pos = strlist__first(self); pos; pos = strlist__next(pos))
+
+/**
+ * strlist_for_each_safe - iterate over a strlist safe against removal of
+ *                         str_node
+ * @pos:	the &struct str_node to use as a loop cursor.
+ * @n:		another &struct str_node to use as temporary storage.
+ * @self:	the &struct strlist for loop.
+ */
+#define strlist__for_each_safe(pos, n, self)	\
+	for (pos = strlist__first(self), n = strlist__next(pos); pos;\
+	     pos = n, n = strlist__next(n))
+
+int strlist__parse_list(struct strlist *self, const char *s);
+#endif /* __PERF_STRLIST_H */
diff --git a/tools/perf/util/svghelper.c b/tools/perf/util/svghelper.c
new file mode 100644
index 00000000..96c86604
--- /dev/null
+++ b/tools/perf/util/svghelper.c
@@ -0,0 +1,501 @@
+/*
+ * svghelper.c - helper functions for outputting svg
+ *
+ * (C) Copyright 2009 Intel Corporation
+ *
+ * Authors:
+ *     Arjan van de Ven <arjan@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <inttypes.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <string.h>
+
+#include "svghelper.h"
+
+static u64 first_time, last_time;
+static u64 turbo_frequency, max_freq;
+
+
+#define SLOT_MULT 30.0
+#define SLOT_HEIGHT 25.0
+
+int svg_page_width = 1000;
+
+#define MIN_TEXT_SIZE 0.01
+
+static u64 total_height;
+static FILE *svgfile;
+
+static double cpu2slot(int cpu)
+{
+	return 2 * cpu + 1;
+}
+
+static double cpu2y(int cpu)
+{
+	return cpu2slot(cpu) * SLOT_MULT;
+}
+
+static double time2pixels(u64 __time)
+{
+	double X;
+
+	X = 1.0 * svg_page_width * (__time - first_time) / (last_time - first_time);
+	return X;
+}
+
+/*
+ * Round text sizes so that the svg viewer only needs a discrete
+ * number of renderings of the font
+ */
+static double round_text_size(double size)
+{
+	int loop = 100;
+	double target = 10.0;
+
+	if (size >= 10.0)
+		return size;
+	while (loop--) {
+		if (size >= target)
+			return target;
+		target = target / 2.0;
+	}
+	return size;
+}
+
+void open_svg(const char *filename, int cpus, int rows, u64 start, u64 end)
+{
+	int new_width;
+
+	svgfile = fopen(filename, "w");
+	if (!svgfile) {
+		fprintf(stderr, "Cannot open %s for output\n", filename);
+		return;
+	}
+	first_time = start;
+	first_time = first_time / 100000000 * 100000000;
+	last_time = end;
+
+	/*
+	 * if the recording is short, we default to a width of 1000, but
+	 * for longer recordings we want at least 200 units of width per second
+	 */
+	new_width = (last_time - first_time) / 5000000;
+
+	if (new_width > svg_page_width)
+		svg_page_width = new_width;
+
+	total_height = (1 + rows + cpu2slot(cpus)) * SLOT_MULT;
+	fprintf(svgfile, "<?xml version=\"1.0\" standalone=\"no\"?> \n");
+	fprintf(svgfile, "<svg width=\"%i\" height=\"%" PRIu64 "\" version=\"1.1\" xmlns=\"http://www.w3.org/2000/svg\">\n", svg_page_width, total_height);
+
+	fprintf(svgfile, "<defs>\n  <style type=\"text/css\">\n    <![CDATA[\n");
+
+	fprintf(svgfile, "      rect          { stroke-width: 1; }\n");
+	fprintf(svgfile, "      rect.process  { fill:rgb(180,180,180); fill-opacity:0.9; stroke-width:1;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.process2 { fill:rgb(180,180,180); fill-opacity:0.9; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.sample   { fill:rgb(  0,  0,255); fill-opacity:0.8; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.blocked  { fill:rgb(255,  0,  0); fill-opacity:0.5; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.waiting  { fill:rgb(224,214,  0); fill-opacity:0.8; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.WAITING  { fill:rgb(255,214, 48); fill-opacity:0.6; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.cpu      { fill:rgb(192,192,192); fill-opacity:0.2; stroke-width:0.5; stroke:rgb(128,128,128); } \n");
+	fprintf(svgfile, "      rect.pstate   { fill:rgb(128,128,128); fill-opacity:0.8; stroke-width:0; } \n");
+	fprintf(svgfile, "      rect.c1       { fill:rgb(255,214,214); fill-opacity:0.5; stroke-width:0; } \n");
+	fprintf(svgfile, "      rect.c2       { fill:rgb(255,172,172); fill-opacity:0.5; stroke-width:0; } \n");
+	fprintf(svgfile, "      rect.c3       { fill:rgb(255,130,130); fill-opacity:0.5; stroke-width:0; } \n");
+	fprintf(svgfile, "      rect.c4       { fill:rgb(255, 88, 88); fill-opacity:0.5; stroke-width:0; } \n");
+	fprintf(svgfile, "      rect.c5       { fill:rgb(255, 44, 44); fill-opacity:0.5; stroke-width:0; } \n");
+	fprintf(svgfile, "      rect.c6       { fill:rgb(255,  0,  0); fill-opacity:0.5; stroke-width:0; } \n");
+	fprintf(svgfile, "      line.pstate   { stroke:rgb(255,255,  0); stroke-opacity:0.8; stroke-width:2; } \n");
+
+	fprintf(svgfile, "    ]]>\n   </style>\n</defs>\n");
+}
+
+void svg_box(int Yslot, u64 start, u64 end, const char *type)
+{
+	if (!svgfile)
+		return;
+
+	fprintf(svgfile, "<rect x=\"%4.8f\" width=\"%4.8f\" y=\"%4.1f\" height=\"%4.1f\" class=\"%s\"/>\n",
+		time2pixels(start), time2pixels(end)-time2pixels(start), Yslot * SLOT_MULT, SLOT_HEIGHT, type);
+}
+
+void svg_sample(int Yslot, int cpu, u64 start, u64 end)
+{
+	double text_size;
+	if (!svgfile)
+		return;
+
+	fprintf(svgfile, "<rect x=\"%4.8f\" width=\"%4.8f\" y=\"%4.1f\" height=\"%4.1f\" class=\"sample\"/>\n",
+		time2pixels(start), time2pixels(end)-time2pixels(start), Yslot * SLOT_MULT, SLOT_HEIGHT);
+
+	text_size = (time2pixels(end)-time2pixels(start));
+	if (cpu > 9)
+		text_size = text_size/2;
+	if (text_size > 1.25)
+		text_size = 1.25;
+	text_size = round_text_size(text_size);
+
+	if (text_size > MIN_TEXT_SIZE)
+		fprintf(svgfile, "<text x=\"%1.8f\" y=\"%1.8f\" font-size=\"%1.8fpt\">%i</text>\n",
+			time2pixels(start), Yslot *  SLOT_MULT + SLOT_HEIGHT - 1, text_size,  cpu + 1);
+
+}
+
+static char *time_to_string(u64 duration)
+{
+	static char text[80];
+
+	text[0] = 0;
+
+	if (duration < 1000) /* less than 1 usec */
+		return text;
+
+	if (duration < 1000 * 1000) { /* less than 1 msec */
+		sprintf(text, "%4.1f us", duration / 1000.0);
+		return text;
+	}
+	sprintf(text, "%4.1f ms", duration / 1000.0 / 1000);
+
+	return text;
+}
+
+void svg_waiting(int Yslot, u64 start, u64 end)
+{
+	char *text;
+	const char *style;
+	double font_size;
+
+	if (!svgfile)
+		return;
+
+	style = "waiting";
+
+	if (end-start > 10 * 1000000) /* 10 msec */
+		style = "WAITING";
+
+	text = time_to_string(end-start);
+
+	font_size = 1.0 * (time2pixels(end)-time2pixels(start));
+
+	if (font_size > 3)
+		font_size = 3;
+
+	font_size = round_text_size(font_size);
+
+	fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\">\n", time2pixels(start), Yslot * SLOT_MULT);
+	fprintf(svgfile, "<rect x=\"0\" width=\"%4.8f\" y=\"0\" height=\"%4.1f\" class=\"%s\"/>\n",
+		time2pixels(end)-time2pixels(start), SLOT_HEIGHT, style);
+	if (font_size > MIN_TEXT_SIZE)
+		fprintf(svgfile, "<text transform=\"rotate(90)\" font-size=\"%1.8fpt\"> %s</text>\n",
+			font_size, text);
+	fprintf(svgfile, "</g>\n");
+}
+
+static char *cpu_model(void)
+{
+	static char cpu_m[255];
+	char buf[256];
+	FILE *file;
+
+	cpu_m[0] = 0;
+	/* CPU type */
+	file = fopen("/proc/cpuinfo", "r");
+	if (file) {
+		while (fgets(buf, 255, file)) {
+			if (strstr(buf, "model name")) {
+				strncpy(cpu_m, &buf[13], 255);
+				break;
+			}
+		}
+		fclose(file);
+	}
+
+	/* CPU type */
+	file = fopen("/sys/devices/system/cpu/cpu0/cpufreq/scaling_available_frequencies", "r");
+	if (file) {
+		while (fgets(buf, 255, file)) {
+			unsigned int freq;
+			freq = strtoull(buf, NULL, 10);
+			if (freq > max_freq)
+				max_freq = freq;
+		}
+		fclose(file);
+	}
+	return cpu_m;
+}
+
+void svg_cpu_box(int cpu, u64 __max_freq, u64 __turbo_freq)
+{
+	char cpu_string[80];
+	if (!svgfile)
+		return;
+
+	max_freq = __max_freq;
+	turbo_frequency = __turbo_freq;
+
+	fprintf(svgfile, "<rect x=\"%4.8f\" width=\"%4.8f\" y=\"%4.1f\" height=\"%4.1f\" class=\"cpu\"/>\n",
+		time2pixels(first_time),
+		time2pixels(last_time)-time2pixels(first_time),
+		cpu2y(cpu), SLOT_MULT+SLOT_HEIGHT);
+
+	sprintf(cpu_string, "CPU %i", (int)cpu+1);
+	fprintf(svgfile, "<text x=\"%4.8f\" y=\"%4.8f\">%s</text>\n",
+		10+time2pixels(first_time), cpu2y(cpu) + SLOT_HEIGHT/2, cpu_string);
+
+	fprintf(svgfile, "<text transform=\"translate(%4.8f,%4.8f)\" font-size=\"1.25pt\">%s</text>\n",
+		10+time2pixels(first_time), cpu2y(cpu) + SLOT_MULT + SLOT_HEIGHT - 4, cpu_model());
+}
+
+void svg_process(int cpu, u64 start, u64 end, const char *type, const char *name)
+{
+	double width;
+
+	if (!svgfile)
+		return;
+
+
+	fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\">\n", time2pixels(start), cpu2y(cpu));
+	fprintf(svgfile, "<rect x=\"0\" width=\"%4.8f\" y=\"0\" height=\"%4.1f\" class=\"%s\"/>\n",
+		time2pixels(end)-time2pixels(start), SLOT_MULT+SLOT_HEIGHT, type);
+	width = time2pixels(end)-time2pixels(start);
+	if (width > 6)
+		width = 6;
+
+	width = round_text_size(width);
+
+	if (width > MIN_TEXT_SIZE)
+		fprintf(svgfile, "<text transform=\"rotate(90)\" font-size=\"%3.8fpt\">%s</text>\n",
+			width, name);
+
+	fprintf(svgfile, "</g>\n");
+}
+
+void svg_cstate(int cpu, u64 start, u64 end, int type)
+{
+	double width;
+	char style[128];
+
+	if (!svgfile)
+		return;
+
+
+	if (type > 6)
+		type = 6;
+	sprintf(style, "c%i", type);
+
+	fprintf(svgfile, "<rect class=\"%s\" x=\"%4.8f\" width=\"%4.8f\" y=\"%4.1f\" height=\"%4.1f\"/>\n",
+		style,
+		time2pixels(start), time2pixels(end)-time2pixels(start),
+		cpu2y(cpu), SLOT_MULT+SLOT_HEIGHT);
+
+	width = (time2pixels(end)-time2pixels(start))/2.0;
+	if (width > 6)
+		width = 6;
+
+	width = round_text_size(width);
+
+	if (width > MIN_TEXT_SIZE)
+		fprintf(svgfile, "<text x=\"%4.8f\" y=\"%4.8f\" font-size=\"%3.8fpt\">C%i</text>\n",
+			time2pixels(start), cpu2y(cpu)+width, width, type);
+}
+
+static char *HzToHuman(unsigned long hz)
+{
+	static char buffer[1024];
+	unsigned long long Hz;
+
+	memset(buffer, 0, 1024);
+
+	Hz = hz;
+
+	/* default: just put the Number in */
+	sprintf(buffer, "%9lli", Hz);
+
+	if (Hz > 1000)
+		sprintf(buffer, " %6lli Mhz", (Hz+500)/1000);
+
+	if (Hz > 1500000)
+		sprintf(buffer, " %6.2f Ghz", (Hz+5000.0)/1000000);
+
+	if (Hz == turbo_frequency)
+		sprintf(buffer, "Turbo");
+
+	return buffer;
+}
+
+void svg_pstate(int cpu, u64 start, u64 end, u64 freq)
+{
+	double height = 0;
+
+	if (!svgfile)
+		return;
+
+	if (max_freq)
+		height = freq * 1.0 / max_freq * (SLOT_HEIGHT + SLOT_MULT);
+	height = 1 + cpu2y(cpu) + SLOT_MULT + SLOT_HEIGHT - height;
+	fprintf(svgfile, "<line x1=\"%4.8f\" x2=\"%4.8f\" y1=\"%4.1f\" y2=\"%4.1f\" class=\"pstate\"/>\n",
+		time2pixels(start), time2pixels(end), height, height);
+	fprintf(svgfile, "<text x=\"%4.8f\" y=\"%4.8f\" font-size=\"0.25pt\">%s</text>\n",
+		time2pixels(start), height+0.9, HzToHuman(freq));
+
+}
+
+
+void svg_partial_wakeline(u64 start, int row1, char *desc1, int row2, char *desc2)
+{
+	double height;
+
+	if (!svgfile)
+		return;
+
+
+	if (row1 < row2) {
+		if (row1) {
+			fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+				time2pixels(start), row1 * SLOT_MULT + SLOT_HEIGHT,  time2pixels(start), row1 * SLOT_MULT + SLOT_HEIGHT + SLOT_MULT/32);
+			if (desc2)
+				fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\"><text transform=\"rotate(90)\" font-size=\"0.02pt\">%s &gt;</text></g>\n",
+					time2pixels(start), row1 * SLOT_MULT + SLOT_HEIGHT + SLOT_HEIGHT/48, desc2);
+		}
+		if (row2) {
+			fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+				time2pixels(start), row2 * SLOT_MULT - SLOT_MULT/32,  time2pixels(start), row2 * SLOT_MULT);
+			if (desc1)
+				fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\"><text transform=\"rotate(90)\" font-size=\"0.02pt\">%s &gt;</text></g>\n",
+					time2pixels(start), row2 * SLOT_MULT - SLOT_MULT/32, desc1);
+		}
+	} else {
+		if (row2) {
+			fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+				time2pixels(start), row2 * SLOT_MULT + SLOT_HEIGHT,  time2pixels(start), row2 * SLOT_MULT + SLOT_HEIGHT + SLOT_MULT/32);
+			if (desc1)
+				fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\"><text transform=\"rotate(90)\" font-size=\"0.02pt\">%s &lt;</text></g>\n",
+					time2pixels(start), row2 * SLOT_MULT + SLOT_HEIGHT + SLOT_MULT/48, desc1);
+		}
+		if (row1) {
+			fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+				time2pixels(start), row1 * SLOT_MULT - SLOT_MULT/32,  time2pixels(start), row1 * SLOT_MULT);
+			if (desc2)
+				fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\"><text transform=\"rotate(90)\" font-size=\"0.02pt\">%s &lt;</text></g>\n",
+					time2pixels(start), row1 * SLOT_MULT - SLOT_HEIGHT/32, desc2);
+		}
+	}
+	height = row1 * SLOT_MULT;
+	if (row2 > row1)
+		height += SLOT_HEIGHT;
+	if (row1)
+		fprintf(svgfile, "<circle  cx=\"%4.8f\" cy=\"%4.2f\" r = \"0.01\"  style=\"fill:rgb(32,255,32)\"/>\n",
+			time2pixels(start), height);
+}
+
+void svg_wakeline(u64 start, int row1, int row2)
+{
+	double height;
+
+	if (!svgfile)
+		return;
+
+
+	if (row1 < row2)
+		fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+			time2pixels(start), row1 * SLOT_MULT + SLOT_HEIGHT,  time2pixels(start), row2 * SLOT_MULT);
+	else
+		fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+			time2pixels(start), row2 * SLOT_MULT + SLOT_HEIGHT,  time2pixels(start), row1 * SLOT_MULT);
+
+	height = row1 * SLOT_MULT;
+	if (row2 > row1)
+		height += SLOT_HEIGHT;
+	fprintf(svgfile, "<circle  cx=\"%4.8f\" cy=\"%4.2f\" r = \"0.01\"  style=\"fill:rgb(32,255,32)\"/>\n",
+			time2pixels(start), height);
+}
+
+void svg_interrupt(u64 start, int row)
+{
+	if (!svgfile)
+		return;
+
+	fprintf(svgfile, "<circle  cx=\"%4.8f\" cy=\"%4.2f\" r = \"0.01\"  style=\"fill:rgb(255,128,128)\"/>\n",
+			time2pixels(start), row * SLOT_MULT);
+	fprintf(svgfile, "<circle  cx=\"%4.8f\" cy=\"%4.2f\" r = \"0.01\"  style=\"fill:rgb(255,128,128)\"/>\n",
+			time2pixels(start), row * SLOT_MULT + SLOT_HEIGHT);
+}
+
+void svg_text(int Yslot, u64 start, const char *text)
+{
+	if (!svgfile)
+		return;
+
+	fprintf(svgfile, "<text x=\"%4.8f\" y=\"%4.8f\">%s</text>\n",
+		time2pixels(start), Yslot * SLOT_MULT+SLOT_HEIGHT/2, text);
+}
+
+static void svg_legenda_box(int X, const char *text, const char *style)
+{
+	double boxsize;
+	boxsize = SLOT_HEIGHT / 2;
+
+	fprintf(svgfile, "<rect x=\"%i\" width=\"%4.8f\" y=\"0\" height=\"%4.1f\" class=\"%s\"/>\n",
+		X, boxsize, boxsize, style);
+	fprintf(svgfile, "<text transform=\"translate(%4.8f, %4.8f)\" font-size=\"%4.8fpt\">%s</text>\n",
+		X + boxsize + 5, boxsize, 0.8 * boxsize, text);
+}
+
+void svg_legenda(void)
+{
+	if (!svgfile)
+		return;
+
+	svg_legenda_box(0,	"Running", "sample");
+	svg_legenda_box(100,	"Idle","c1");
+	svg_legenda_box(200,	"Deeper Idle", "c3");
+	svg_legenda_box(350,	"Deepest Idle", "c6");
+	svg_legenda_box(550,	"Sleeping", "process2");
+	svg_legenda_box(650,	"Waiting for cpu", "waiting");
+	svg_legenda_box(800,	"Blocked on IO", "blocked");
+}
+
+void svg_time_grid(void)
+{
+	u64 i;
+
+	if (!svgfile)
+		return;
+
+	i = first_time;
+	while (i < last_time) {
+		int color = 220;
+		double thickness = 0.075;
+		if ((i % 100000000) == 0) {
+			thickness = 0.5;
+			color = 192;
+		}
+		if ((i % 1000000000) == 0) {
+			thickness = 2.0;
+			color = 128;
+		}
+
+		fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%" PRIu64 "\" style=\"stroke:rgb(%i,%i,%i);stroke-width:%1.3f\"/>\n",
+			time2pixels(i), SLOT_MULT/2, time2pixels(i), total_height, color, color, color, thickness);
+
+		i += 10000000;
+	}
+}
+
+void svg_close(void)
+{
+	if (svgfile) {
+		fprintf(svgfile, "</svg>\n");
+		fclose(svgfile);
+		svgfile = NULL;
+	}
+}
diff --git a/tools/perf/util/svghelper.h b/tools/perf/util/svghelper.h
new file mode 100644
index 00000000..e0781989
--- /dev/null
+++ b/tools/perf/util/svghelper.h
@@ -0,0 +1,28 @@
+#ifndef __PERF_SVGHELPER_H
+#define __PERF_SVGHELPER_H
+
+#include "types.h"
+
+extern void open_svg(const char *filename, int cpus, int rows, u64 start, u64 end);
+extern void svg_box(int Yslot, u64 start, u64 end, const char *type);
+extern void svg_sample(int Yslot, int cpu, u64 start, u64 end);
+extern void svg_waiting(int Yslot, u64 start, u64 end);
+extern void svg_cpu_box(int cpu, u64 max_frequency, u64 turbo_frequency);
+
+
+extern void svg_process(int cpu, u64 start, u64 end, const char *type, const char *name);
+extern void svg_cstate(int cpu, u64 start, u64 end, int type);
+extern void svg_pstate(int cpu, u64 start, u64 end, u64 freq);
+
+
+extern void svg_time_grid(void);
+extern void svg_legenda(void);
+extern void svg_wakeline(u64 start, int row1, int row2);
+extern void svg_partial_wakeline(u64 start, int row1, char *desc1, int row2, char *desc2);
+extern void svg_interrupt(u64 start, int row);
+extern void svg_text(int Yslot, u64 start, const char *text);
+extern void svg_close(void);
+
+extern int svg_page_width;
+
+#endif /* __PERF_SVGHELPER_H */
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
new file mode 100644
index 00000000..40fd1c7c
--- /dev/null
+++ b/tools/perf/util/symbol.c
@@ -0,0 +1,2669 @@
+#define _GNU_SOURCE
+#include <ctype.h>
+#include <dirent.h>
+#include <errno.h>
+#include <libgen.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include <string.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/param.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include <inttypes.h>
+#include "build-id.h"
+#include "debug.h"
+#include "symbol.h"
+#include "strlist.h"
+
+#include <libelf.h>
+#include <gelf.h>
+#include <elf.h>
+#include <limits.h>
+#include <sys/utsname.h>
+
+#ifndef KSYM_NAME_LEN
+#define KSYM_NAME_LEN 128
+#endif
+
+#ifndef NT_GNU_BUILD_ID
+#define NT_GNU_BUILD_ID 3
+#endif
+
+static bool dso__build_id_equal(const struct dso *dso, u8 *build_id);
+static int elf_read_build_id(Elf *elf, void *bf, size_t size);
+static void dsos__add(struct list_head *head, struct dso *dso);
+static struct map *map__new2(u64 start, struct dso *dso, enum map_type type);
+static int dso__load_kernel_sym(struct dso *dso, struct map *map,
+				symbol_filter_t filter);
+static int dso__load_guest_kernel_sym(struct dso *dso, struct map *map,
+			symbol_filter_t filter);
+static int vmlinux_path__nr_entries;
+static char **vmlinux_path;
+
+struct symbol_conf symbol_conf = {
+	.exclude_other	  = true,
+	.use_modules	  = true,
+	.try_vmlinux_path = true,
+	.symfs            = "",
+};
+
+int dso__name_len(const struct dso *dso)
+{
+	if (verbose)
+		return dso->long_name_len;
+
+	return dso->short_name_len;
+}
+
+bool dso__loaded(const struct dso *dso, enum map_type type)
+{
+	return dso->loaded & (1 << type);
+}
+
+bool dso__sorted_by_name(const struct dso *dso, enum map_type type)
+{
+	return dso->sorted_by_name & (1 << type);
+}
+
+static void dso__set_sorted_by_name(struct dso *dso, enum map_type type)
+{
+	dso->sorted_by_name |= (1 << type);
+}
+
+bool symbol_type__is_a(char symbol_type, enum map_type map_type)
+{
+	switch (map_type) {
+	case MAP__FUNCTION:
+		return symbol_type == 'T' || symbol_type == 'W';
+	case MAP__VARIABLE:
+		return symbol_type == 'D' || symbol_type == 'd';
+	default:
+		return false;
+	}
+}
+
+static void symbols__fixup_end(struct rb_root *symbols)
+{
+	struct rb_node *nd, *prevnd = rb_first(symbols);
+	struct symbol *curr, *prev;
+
+	if (prevnd == NULL)
+		return;
+
+	curr = rb_entry(prevnd, struct symbol, rb_node);
+
+	for (nd = rb_next(prevnd); nd; nd = rb_next(nd)) {
+		prev = curr;
+		curr = rb_entry(nd, struct symbol, rb_node);
+
+		if (prev->end == prev->start && prev->end != curr->start)
+			prev->end = curr->start - 1;
+	}
+
+	/* Last entry */
+	if (curr->end == curr->start)
+		curr->end = roundup(curr->start, 4096);
+}
+
+static void __map_groups__fixup_end(struct map_groups *mg, enum map_type type)
+{
+	struct map *prev, *curr;
+	struct rb_node *nd, *prevnd = rb_first(&mg->maps[type]);
+
+	if (prevnd == NULL)
+		return;
+
+	curr = rb_entry(prevnd, struct map, rb_node);
+
+	for (nd = rb_next(prevnd); nd; nd = rb_next(nd)) {
+		prev = curr;
+		curr = rb_entry(nd, struct map, rb_node);
+		prev->end = curr->start - 1;
+	}
+
+	/*
+	 * We still haven't the actual symbols, so guess the
+	 * last map final address.
+	 */
+	curr->end = ~0ULL;
+}
+
+static void map_groups__fixup_end(struct map_groups *mg)
+{
+	int i;
+	for (i = 0; i < MAP__NR_TYPES; ++i)
+		__map_groups__fixup_end(mg, i);
+}
+
+static struct symbol *symbol__new(u64 start, u64 len, u8 binding,
+				  const char *name)
+{
+	size_t namelen = strlen(name) + 1;
+	struct symbol *sym = calloc(1, (symbol_conf.priv_size +
+					sizeof(*sym) + namelen));
+	if (sym == NULL)
+		return NULL;
+
+	if (symbol_conf.priv_size)
+		sym = ((void *)sym) + symbol_conf.priv_size;
+
+	sym->start   = start;
+	sym->end     = len ? start + len - 1 : start;
+	sym->binding = binding;
+	sym->namelen = namelen - 1;
+
+	pr_debug4("%s: %s %#" PRIx64 "-%#" PRIx64 "\n",
+		  __func__, name, start, sym->end);
+	memcpy(sym->name, name, namelen);
+
+	return sym;
+}
+
+void symbol__delete(struct symbol *sym)
+{
+	free(((void *)sym) - symbol_conf.priv_size);
+}
+
+static size_t symbol__fprintf(struct symbol *sym, FILE *fp)
+{
+	return fprintf(fp, " %" PRIx64 "-%" PRIx64 " %c %s\n",
+		       sym->start, sym->end,
+		       sym->binding == STB_GLOBAL ? 'g' :
+		       sym->binding == STB_LOCAL  ? 'l' : 'w',
+		       sym->name);
+}
+
+void dso__set_long_name(struct dso *dso, char *name)
+{
+	if (name == NULL)
+		return;
+	dso->long_name = name;
+	dso->long_name_len = strlen(name);
+}
+
+static void dso__set_short_name(struct dso *dso, const char *name)
+{
+	if (name == NULL)
+		return;
+	dso->short_name = name;
+	dso->short_name_len = strlen(name);
+}
+
+static void dso__set_basename(struct dso *dso)
+{
+	dso__set_short_name(dso, basename(dso->long_name));
+}
+
+struct dso *dso__new(const char *name)
+{
+	struct dso *dso = calloc(1, sizeof(*dso) + strlen(name) + 1);
+
+	if (dso != NULL) {
+		int i;
+		strcpy(dso->name, name);
+		dso__set_long_name(dso, dso->name);
+		dso__set_short_name(dso, dso->name);
+		for (i = 0; i < MAP__NR_TYPES; ++i)
+			dso->symbols[i] = dso->symbol_names[i] = RB_ROOT;
+		dso->symtab_type = SYMTAB__NOT_FOUND;
+		dso->loaded = 0;
+		dso->sorted_by_name = 0;
+		dso->has_build_id = 0;
+		dso->kernel = DSO_TYPE_USER;
+		INIT_LIST_HEAD(&dso->node);
+	}
+
+	return dso;
+}
+
+static void symbols__delete(struct rb_root *symbols)
+{
+	struct symbol *pos;
+	struct rb_node *next = rb_first(symbols);
+
+	while (next) {
+		pos = rb_entry(next, struct symbol, rb_node);
+		next = rb_next(&pos->rb_node);
+		rb_erase(&pos->rb_node, symbols);
+		symbol__delete(pos);
+	}
+}
+
+void dso__delete(struct dso *dso)
+{
+	int i;
+	for (i = 0; i < MAP__NR_TYPES; ++i)
+		symbols__delete(&dso->symbols[i]);
+	if (dso->sname_alloc)
+		free((char *)dso->short_name);
+	if (dso->lname_alloc)
+		free(dso->long_name);
+	free(dso);
+}
+
+void dso__set_build_id(struct dso *dso, void *build_id)
+{
+	memcpy(dso->build_id, build_id, sizeof(dso->build_id));
+	dso->has_build_id = 1;
+}
+
+static void symbols__insert(struct rb_root *symbols, struct symbol *sym)
+{
+	struct rb_node **p = &symbols->rb_node;
+	struct rb_node *parent = NULL;
+	const u64 ip = sym->start;
+	struct symbol *s;
+
+	while (*p != NULL) {
+		parent = *p;
+		s = rb_entry(parent, struct symbol, rb_node);
+		if (ip < s->start)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+	rb_link_node(&sym->rb_node, parent, p);
+	rb_insert_color(&sym->rb_node, symbols);
+}
+
+static struct symbol *symbols__find(struct rb_root *symbols, u64 ip)
+{
+	struct rb_node *n;
+
+	if (symbols == NULL)
+		return NULL;
+
+	n = symbols->rb_node;
+
+	while (n) {
+		struct symbol *s = rb_entry(n, struct symbol, rb_node);
+
+		if (ip < s->start)
+			n = n->rb_left;
+		else if (ip > s->end)
+			n = n->rb_right;
+		else
+			return s;
+	}
+
+	return NULL;
+}
+
+struct symbol_name_rb_node {
+	struct rb_node	rb_node;
+	struct symbol	sym;
+};
+
+static void symbols__insert_by_name(struct rb_root *symbols, struct symbol *sym)
+{
+	struct rb_node **p = &symbols->rb_node;
+	struct rb_node *parent = NULL;
+	struct symbol_name_rb_node *symn, *s;
+
+	symn = container_of(sym, struct symbol_name_rb_node, sym);
+
+	while (*p != NULL) {
+		parent = *p;
+		s = rb_entry(parent, struct symbol_name_rb_node, rb_node);
+		if (strcmp(sym->name, s->sym.name) < 0)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+	rb_link_node(&symn->rb_node, parent, p);
+	rb_insert_color(&symn->rb_node, symbols);
+}
+
+static void symbols__sort_by_name(struct rb_root *symbols,
+				  struct rb_root *source)
+{
+	struct rb_node *nd;
+
+	for (nd = rb_first(source); nd; nd = rb_next(nd)) {
+		struct symbol *pos = rb_entry(nd, struct symbol, rb_node);
+		symbols__insert_by_name(symbols, pos);
+	}
+}
+
+static struct symbol *symbols__find_by_name(struct rb_root *symbols,
+					    const char *name)
+{
+	struct rb_node *n;
+
+	if (symbols == NULL)
+		return NULL;
+
+	n = symbols->rb_node;
+
+	while (n) {
+		struct symbol_name_rb_node *s;
+		int cmp;
+
+		s = rb_entry(n, struct symbol_name_rb_node, rb_node);
+		cmp = strcmp(name, s->sym.name);
+
+		if (cmp < 0)
+			n = n->rb_left;
+		else if (cmp > 0)
+			n = n->rb_right;
+		else
+			return &s->sym;
+	}
+
+	return NULL;
+}
+
+struct symbol *dso__find_symbol(struct dso *dso,
+				enum map_type type, u64 addr)
+{
+	return symbols__find(&dso->symbols[type], addr);
+}
+
+struct symbol *dso__find_symbol_by_name(struct dso *dso, enum map_type type,
+					const char *name)
+{
+	return symbols__find_by_name(&dso->symbol_names[type], name);
+}
+
+void dso__sort_by_name(struct dso *dso, enum map_type type)
+{
+	dso__set_sorted_by_name(dso, type);
+	return symbols__sort_by_name(&dso->symbol_names[type],
+				     &dso->symbols[type]);
+}
+
+int build_id__sprintf(const u8 *build_id, int len, char *bf)
+{
+	char *bid = bf;
+	const u8 *raw = build_id;
+	int i;
+
+	for (i = 0; i < len; ++i) {
+		sprintf(bid, "%02x", *raw);
+		++raw;
+		bid += 2;
+	}
+
+	return raw - build_id;
+}
+
+size_t dso__fprintf_buildid(struct dso *dso, FILE *fp)
+{
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+	build_id__sprintf(dso->build_id, sizeof(dso->build_id), sbuild_id);
+	return fprintf(fp, "%s", sbuild_id);
+}
+
+size_t dso__fprintf_symbols_by_name(struct dso *dso,
+				    enum map_type type, FILE *fp)
+{
+	size_t ret = 0;
+	struct rb_node *nd;
+	struct symbol_name_rb_node *pos;
+
+	for (nd = rb_first(&dso->symbol_names[type]); nd; nd = rb_next(nd)) {
+		pos = rb_entry(nd, struct symbol_name_rb_node, rb_node);
+		fprintf(fp, "%s\n", pos->sym.name);
+	}
+
+	return ret;
+}
+
+size_t dso__fprintf(struct dso *dso, enum map_type type, FILE *fp)
+{
+	struct rb_node *nd;
+	size_t ret = fprintf(fp, "dso: %s (", dso->short_name);
+
+	if (dso->short_name != dso->long_name)
+		ret += fprintf(fp, "%s, ", dso->long_name);
+	ret += fprintf(fp, "%s, %sloaded, ", map_type__name[type],
+		       dso->loaded ? "" : "NOT ");
+	ret += dso__fprintf_buildid(dso, fp);
+	ret += fprintf(fp, ")\n");
+	for (nd = rb_first(&dso->symbols[type]); nd; nd = rb_next(nd)) {
+		struct symbol *pos = rb_entry(nd, struct symbol, rb_node);
+		ret += symbol__fprintf(pos, fp);
+	}
+
+	return ret;
+}
+
+int kallsyms__parse(const char *filename, void *arg,
+		    int (*process_symbol)(void *arg, const char *name,
+					  char type, u64 start, u64 end))
+{
+	char *line = NULL;
+	size_t n;
+	int err = -1;
+	u64 prev_start = 0;
+	char prev_symbol_type = 0;
+	char *prev_symbol_name;
+	FILE *file = fopen(filename, "r");
+
+	if (file == NULL)
+		goto out_failure;
+
+	prev_symbol_name = malloc(KSYM_NAME_LEN);
+	if (prev_symbol_name == NULL)
+		goto out_close;
+
+	err = 0;
+
+	while (!feof(file)) {
+		u64 start;
+		int line_len, len;
+		char symbol_type;
+		char *symbol_name;
+
+		line_len = getline(&line, &n, file);
+		if (line_len < 0 || !line)
+			break;
+
+		line[--line_len] = '\0'; /* \n */
+
+		len = hex2u64(line, &start);
+
+		len++;
+		if (len + 2 >= line_len)
+			continue;
+
+		symbol_type = toupper(line[len]);
+		len += 2;
+		symbol_name = line + len;
+		len = line_len - len;
+
+		if (len >= KSYM_NAME_LEN) {
+			err = -1;
+			break;
+		}
+
+		if (prev_symbol_type) {
+			u64 end = start;
+			if (end != prev_start)
+				--end;
+			err = process_symbol(arg, prev_symbol_name,
+					     prev_symbol_type, prev_start, end);
+			if (err)
+				break;
+		}
+
+		memcpy(prev_symbol_name, symbol_name, len + 1);
+		prev_symbol_type = symbol_type;
+		prev_start = start;
+	}
+
+	free(prev_symbol_name);
+	free(line);
+out_close:
+	fclose(file);
+	return err;
+
+out_failure:
+	return -1;
+}
+
+struct process_kallsyms_args {
+	struct map *map;
+	struct dso *dso;
+};
+
+static u8 kallsyms2elf_type(char type)
+{
+	if (type == 'W')
+		return STB_WEAK;
+
+	return isupper(type) ? STB_GLOBAL : STB_LOCAL;
+}
+
+static int map__process_kallsym_symbol(void *arg, const char *name,
+				       char type, u64 start, u64 end)
+{
+	struct symbol *sym;
+	struct process_kallsyms_args *a = arg;
+	struct rb_root *root = &a->dso->symbols[a->map->type];
+
+	if (!symbol_type__is_a(type, a->map->type))
+		return 0;
+
+	sym = symbol__new(start, end - start + 1,
+			  kallsyms2elf_type(type), name);
+	if (sym == NULL)
+		return -ENOMEM;
+	/*
+	 * We will pass the symbols to the filter later, in
+	 * map__split_kallsyms, when we have split the maps per module
+	 */
+	symbols__insert(root, sym);
+
+	return 0;
+}
+
+/*
+ * Loads the function entries in /proc/kallsyms into kernel_map->dso,
+ * so that we can in the next step set the symbol ->end address and then
+ * call kernel_maps__split_kallsyms.
+ */
+static int dso__load_all_kallsyms(struct dso *dso, const char *filename,
+				  struct map *map)
+{
+	struct process_kallsyms_args args = { .map = map, .dso = dso, };
+	return kallsyms__parse(filename, &args, map__process_kallsym_symbol);
+}
+
+/*
+ * Split the symbols into maps, making sure there are no overlaps, i.e. the
+ * kernel range is broken in several maps, named [kernel].N, as we don't have
+ * the original ELF section names vmlinux have.
+ */
+static int dso__split_kallsyms(struct dso *dso, struct map *map,
+			       symbol_filter_t filter)
+{
+	struct map_groups *kmaps = map__kmap(map)->kmaps;
+	struct machine *machine = kmaps->machine;
+	struct map *curr_map = map;
+	struct symbol *pos;
+	int count = 0, moved = 0;	
+	struct rb_root *root = &dso->symbols[map->type];
+	struct rb_node *next = rb_first(root);
+	int kernel_range = 0;
+
+	while (next) {
+		char *module;
+
+		pos = rb_entry(next, struct symbol, rb_node);
+		next = rb_next(&pos->rb_node);
+
+		module = strchr(pos->name, '\t');
+		if (module) {
+			if (!symbol_conf.use_modules)
+				goto discard_symbol;
+
+			*module++ = '\0';
+
+			if (strcmp(curr_map->dso->short_name, module)) {
+				if (curr_map != map &&
+				    dso->kernel == DSO_TYPE_GUEST_KERNEL &&
+				    machine__is_default_guest(machine)) {
+					/*
+					 * We assume all symbols of a module are
+					 * continuous in * kallsyms, so curr_map
+					 * points to a module and all its
+					 * symbols are in its kmap. Mark it as
+					 * loaded.
+					 */
+					dso__set_loaded(curr_map->dso,
+							curr_map->type);
+				}
+
+				curr_map = map_groups__find_by_name(kmaps,
+							map->type, module);
+				if (curr_map == NULL) {
+					pr_debug("%s/proc/{kallsyms,modules} "
+					         "inconsistency while looking "
+						 "for \"%s\" module!\n",
+						 machine->root_dir, module);
+					curr_map = map;
+					goto discard_symbol;
+				}
+
+				if (curr_map->dso->loaded &&
+				    !machine__is_default_guest(machine))
+					goto discard_symbol;
+			}
+			/*
+			 * So that we look just like we get from .ko files,
+			 * i.e. not prelinked, relative to map->start.
+			 */
+			pos->start = curr_map->map_ip(curr_map, pos->start);
+			pos->end   = curr_map->map_ip(curr_map, pos->end);
+		} else if (curr_map != map) {
+			char dso_name[PATH_MAX];
+			struct dso *ndso;
+
+			if (count == 0) {
+				curr_map = map;
+				goto filter_symbol;
+			}
+
+			if (dso->kernel == DSO_TYPE_GUEST_KERNEL)
+				snprintf(dso_name, sizeof(dso_name),
+					"[guest.kernel].%d",
+					kernel_range++);
+			else
+				snprintf(dso_name, sizeof(dso_name),
+					"[kernel].%d",
+					kernel_range++);
+
+			ndso = dso__new(dso_name);
+			if (ndso == NULL)
+				return -1;
+
+			ndso->kernel = dso->kernel;
+
+			curr_map = map__new2(pos->start, ndso, map->type);
+			if (curr_map == NULL) {
+				dso__delete(ndso);
+				return -1;
+			}
+
+			curr_map->map_ip = curr_map->unmap_ip = identity__map_ip;
+			map_groups__insert(kmaps, curr_map);
+			++kernel_range;
+		}
+filter_symbol:
+		if (filter && filter(curr_map, pos)) {
+discard_symbol:		rb_erase(&pos->rb_node, root);
+			symbol__delete(pos);
+		} else {
+			if (curr_map != map) {
+				rb_erase(&pos->rb_node, root);
+				symbols__insert(&curr_map->dso->symbols[curr_map->type], pos);
+				++moved;
+			} else
+				++count;
+		}
+	}
+
+	if (curr_map != map &&
+	    dso->kernel == DSO_TYPE_GUEST_KERNEL &&
+	    machine__is_default_guest(kmaps->machine)) {
+		dso__set_loaded(curr_map->dso, curr_map->type);
+	}
+
+	return count + moved;
+}
+
+static bool symbol__restricted_filename(const char *filename,
+					const char *restricted_filename)
+{
+	bool restricted = false;
+
+	if (symbol_conf.kptr_restrict) {
+		char *r = realpath(filename, NULL);
+
+		if (r != NULL) {
+			restricted = strcmp(r, restricted_filename) == 0;
+			free(r);
+			return restricted;
+		}
+	}
+
+	return restricted;
+}
+
+int dso__load_kallsyms(struct dso *dso, const char *filename,
+		       struct map *map, symbol_filter_t filter)
+{
+	if (symbol__restricted_filename(filename, "/proc/kallsyms"))
+		return -1;
+
+	if (dso__load_all_kallsyms(dso, filename, map) < 0)
+		return -1;
+
+	if (dso->kernel == DSO_TYPE_GUEST_KERNEL)
+		dso->symtab_type = SYMTAB__GUEST_KALLSYMS;
+	else
+		dso->symtab_type = SYMTAB__KALLSYMS;
+
+	return dso__split_kallsyms(dso, map, filter);
+}
+
+static int dso__load_perf_map(struct dso *dso, struct map *map,
+			      symbol_filter_t filter)
+{
+	char *line = NULL;
+	size_t n;
+	FILE *file;
+	int nr_syms = 0;
+
+	file = fopen(dso->long_name, "r");
+	if (file == NULL)
+		goto out_failure;
+
+	while (!feof(file)) {
+		u64 start, size;
+		struct symbol *sym;
+		int line_len, len;
+
+		line_len = getline(&line, &n, file);
+		if (line_len < 0)
+			break;
+
+		if (!line)
+			goto out_failure;
+
+		line[--line_len] = '\0'; /* \n */
+
+		len = hex2u64(line, &start);
+
+		len++;
+		if (len + 2 >= line_len)
+			continue;
+
+		len += hex2u64(line + len, &size);
+
+		len++;
+		if (len + 2 >= line_len)
+			continue;
+
+		sym = symbol__new(start, size, STB_GLOBAL, line + len);
+
+		if (sym == NULL)
+			goto out_delete_line;
+
+		if (filter && filter(map, sym))
+			symbol__delete(sym);
+		else {
+			symbols__insert(&dso->symbols[map->type], sym);
+			nr_syms++;
+		}
+	}
+
+	free(line);
+	fclose(file);
+
+	return nr_syms;
+
+out_delete_line:
+	free(line);
+out_failure:
+	return -1;
+}
+
+/**
+ * elf_symtab__for_each_symbol - iterate thru all the symbols
+ *
+ * @syms: struct elf_symtab instance to iterate
+ * @idx: uint32_t idx
+ * @sym: GElf_Sym iterator
+ */
+#define elf_symtab__for_each_symbol(syms, nr_syms, idx, sym) \
+	for (idx = 0, gelf_getsym(syms, idx, &sym);\
+	     idx < nr_syms; \
+	     idx++, gelf_getsym(syms, idx, &sym))
+
+static inline uint8_t elf_sym__type(const GElf_Sym *sym)
+{
+	return GELF_ST_TYPE(sym->st_info);
+}
+
+static inline int elf_sym__is_function(const GElf_Sym *sym)
+{
+	return elf_sym__type(sym) == STT_FUNC &&
+	       sym->st_name != 0 &&
+	       sym->st_shndx != SHN_UNDEF;
+}
+
+static inline bool elf_sym__is_object(const GElf_Sym *sym)
+{
+	return elf_sym__type(sym) == STT_OBJECT &&
+		sym->st_name != 0 &&
+		sym->st_shndx != SHN_UNDEF;
+}
+
+static inline int elf_sym__is_label(const GElf_Sym *sym)
+{
+	return elf_sym__type(sym) == STT_NOTYPE &&
+		sym->st_name != 0 &&
+		sym->st_shndx != SHN_UNDEF &&
+		sym->st_shndx != SHN_ABS;
+}
+
+static inline const char *elf_sec__name(const GElf_Shdr *shdr,
+					const Elf_Data *secstrs)
+{
+	return secstrs->d_buf + shdr->sh_name;
+}
+
+static inline int elf_sec__is_text(const GElf_Shdr *shdr,
+					const Elf_Data *secstrs)
+{
+	return strstr(elf_sec__name(shdr, secstrs), "text") != NULL;
+}
+
+static inline bool elf_sec__is_data(const GElf_Shdr *shdr,
+				    const Elf_Data *secstrs)
+{
+	return strstr(elf_sec__name(shdr, secstrs), "data") != NULL;
+}
+
+static inline const char *elf_sym__name(const GElf_Sym *sym,
+					const Elf_Data *symstrs)
+{
+	return symstrs->d_buf + sym->st_name;
+}
+
+static Elf_Scn *elf_section_by_name(Elf *elf, GElf_Ehdr *ep,
+				    GElf_Shdr *shp, const char *name,
+				    size_t *idx)
+{
+	Elf_Scn *sec = NULL;
+	size_t cnt = 1;
+
+	while ((sec = elf_nextscn(elf, sec)) != NULL) {
+		char *str;
+
+		gelf_getshdr(sec, shp);
+		str = elf_strptr(elf, ep->e_shstrndx, shp->sh_name);
+		if (!strcmp(name, str)) {
+			if (idx)
+				*idx = cnt;
+			break;
+		}
+		++cnt;
+	}
+
+	return sec;
+}
+
+#define elf_section__for_each_rel(reldata, pos, pos_mem, idx, nr_entries) \
+	for (idx = 0, pos = gelf_getrel(reldata, 0, &pos_mem); \
+	     idx < nr_entries; \
+	     ++idx, pos = gelf_getrel(reldata, idx, &pos_mem))
+
+#define elf_section__for_each_rela(reldata, pos, pos_mem, idx, nr_entries) \
+	for (idx = 0, pos = gelf_getrela(reldata, 0, &pos_mem); \
+	     idx < nr_entries; \
+	     ++idx, pos = gelf_getrela(reldata, idx, &pos_mem))
+
+/*
+ * We need to check if we have a .dynsym, so that we can handle the
+ * .plt, synthesizing its symbols, that aren't on the symtabs (be it
+ * .dynsym or .symtab).
+ * And always look at the original dso, not at debuginfo packages, that
+ * have the PLT data stripped out (shdr_rel_plt.sh_type == SHT_NOBITS).
+ */
+static int dso__synthesize_plt_symbols(struct  dso *dso, struct map *map,
+				       symbol_filter_t filter)
+{
+	uint32_t nr_rel_entries, idx;
+	GElf_Sym sym;
+	u64 plt_offset;
+	GElf_Shdr shdr_plt;
+	struct symbol *f;
+	GElf_Shdr shdr_rel_plt, shdr_dynsym;
+	Elf_Data *reldata, *syms, *symstrs;
+	Elf_Scn *scn_plt_rel, *scn_symstrs, *scn_dynsym;
+	size_t dynsym_idx;
+	GElf_Ehdr ehdr;
+	char sympltname[1024];
+	Elf *elf;
+	int nr = 0, symidx, fd, err = 0;
+	char name[PATH_MAX];
+
+	snprintf(name, sizeof(name), "%s%s",
+		 symbol_conf.symfs, dso->long_name);
+	fd = open(name, O_RDONLY);
+	if (fd < 0)
+		goto out;
+
+	elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL);
+	if (elf == NULL)
+		goto out_close;
+
+	if (gelf_getehdr(elf, &ehdr) == NULL)
+		goto out_elf_end;
+
+	scn_dynsym = elf_section_by_name(elf, &ehdr, &shdr_dynsym,
+					 ".dynsym", &dynsym_idx);
+	if (scn_dynsym == NULL)
+		goto out_elf_end;
+
+	scn_plt_rel = elf_section_by_name(elf, &ehdr, &shdr_rel_plt,
+					  ".rela.plt", NULL);
+	if (scn_plt_rel == NULL) {
+		scn_plt_rel = elf_section_by_name(elf, &ehdr, &shdr_rel_plt,
+						  ".rel.plt", NULL);
+		if (scn_plt_rel == NULL)
+			goto out_elf_end;
+	}
+
+	err = -1;
+
+	if (shdr_rel_plt.sh_link != dynsym_idx)
+		goto out_elf_end;
+
+	if (elf_section_by_name(elf, &ehdr, &shdr_plt, ".plt", NULL) == NULL)
+		goto out_elf_end;
+
+	/*
+	 * Fetch the relocation section to find the idxes to the GOT
+	 * and the symbols in the .dynsym they refer to.
+	 */
+	reldata = elf_getdata(scn_plt_rel, NULL);
+	if (reldata == NULL)
+		goto out_elf_end;
+
+	syms = elf_getdata(scn_dynsym, NULL);
+	if (syms == NULL)
+		goto out_elf_end;
+
+	scn_symstrs = elf_getscn(elf, shdr_dynsym.sh_link);
+	if (scn_symstrs == NULL)
+		goto out_elf_end;
+
+	symstrs = elf_getdata(scn_symstrs, NULL);
+	if (symstrs == NULL)
+		goto out_elf_end;
+
+	nr_rel_entries = shdr_rel_plt.sh_size / shdr_rel_plt.sh_entsize;
+	plt_offset = shdr_plt.sh_offset;
+
+	if (shdr_rel_plt.sh_type == SHT_RELA) {
+		GElf_Rela pos_mem, *pos;
+
+		elf_section__for_each_rela(reldata, pos, pos_mem, idx,
+					   nr_rel_entries) {
+			symidx = GELF_R_SYM(pos->r_info);
+			plt_offset += shdr_plt.sh_entsize;
+			gelf_getsym(syms, symidx, &sym);
+			snprintf(sympltname, sizeof(sympltname),
+				 "%s@plt", elf_sym__name(&sym, symstrs));
+
+			f = symbol__new(plt_offset, shdr_plt.sh_entsize,
+					STB_GLOBAL, sympltname);
+			if (!f)
+				goto out_elf_end;
+
+			if (filter && filter(map, f))
+				symbol__delete(f);
+			else {
+				symbols__insert(&dso->symbols[map->type], f);
+				++nr;
+			}
+		}
+	} else if (shdr_rel_plt.sh_type == SHT_REL) {
+		GElf_Rel pos_mem, *pos;
+		elf_section__for_each_rel(reldata, pos, pos_mem, idx,
+					  nr_rel_entries) {
+			symidx = GELF_R_SYM(pos->r_info);
+			plt_offset += shdr_plt.sh_entsize;
+			gelf_getsym(syms, symidx, &sym);
+			snprintf(sympltname, sizeof(sympltname),
+				 "%s@plt", elf_sym__name(&sym, symstrs));
+
+			f = symbol__new(plt_offset, shdr_plt.sh_entsize,
+					STB_GLOBAL, sympltname);
+			if (!f)
+				goto out_elf_end;
+
+			if (filter && filter(map, f))
+				symbol__delete(f);
+			else {
+				symbols__insert(&dso->symbols[map->type], f);
+				++nr;
+			}
+		}
+	}
+
+	err = 0;
+out_elf_end:
+	elf_end(elf);
+out_close:
+	close(fd);
+
+	if (err == 0)
+		return nr;
+out:
+	pr_debug("%s: problems reading %s PLT info.\n",
+		 __func__, dso->long_name);
+	return 0;
+}
+
+static bool elf_sym__is_a(GElf_Sym *sym, enum map_type type)
+{
+	switch (type) {
+	case MAP__FUNCTION:
+		return elf_sym__is_function(sym);
+	case MAP__VARIABLE:
+		return elf_sym__is_object(sym);
+	default:
+		return false;
+	}
+}
+
+static bool elf_sec__is_a(GElf_Shdr *shdr, Elf_Data *secstrs,
+			  enum map_type type)
+{
+	switch (type) {
+	case MAP__FUNCTION:
+		return elf_sec__is_text(shdr, secstrs);
+	case MAP__VARIABLE:
+		return elf_sec__is_data(shdr, secstrs);
+	default:
+		return false;
+	}
+}
+
+static size_t elf_addr_to_index(Elf *elf, GElf_Addr addr)
+{
+	Elf_Scn *sec = NULL;
+	GElf_Shdr shdr;
+	size_t cnt = 1;
+
+	while ((sec = elf_nextscn(elf, sec)) != NULL) {
+		gelf_getshdr(sec, &shdr);
+
+		if ((addr >= shdr.sh_addr) &&
+		    (addr < (shdr.sh_addr + shdr.sh_size)))
+			return cnt;
+
+		++cnt;
+	}
+
+	return -1;
+}
+
+static int dso__load_sym(struct dso *dso, struct map *map, const char *name,
+			 int fd, symbol_filter_t filter, int kmodule,
+			 int want_symtab)
+{
+	struct kmap *kmap = dso->kernel ? map__kmap(map) : NULL;
+	struct map *curr_map = map;
+	struct dso *curr_dso = dso;
+	Elf_Data *symstrs, *secstrs;
+	uint32_t nr_syms;
+	int err = -1;
+	uint32_t idx;
+	GElf_Ehdr ehdr;
+	GElf_Shdr shdr, opdshdr;
+	Elf_Data *syms, *opddata = NULL;
+	GElf_Sym sym;
+	Elf_Scn *sec, *sec_strndx, *opdsec;
+	Elf *elf;
+	int nr = 0;
+	size_t opdidx = 0;
+
+	elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL);
+	if (elf == NULL) {
+		pr_debug("%s: cannot read %s ELF file.\n", __func__, name);
+		goto out_close;
+	}
+
+	if (gelf_getehdr(elf, &ehdr) == NULL) {
+		pr_debug("%s: cannot get elf header.\n", __func__);
+		goto out_elf_end;
+	}
+
+	/* Always reject images with a mismatched build-id: */
+	if (dso->has_build_id) {
+		u8 build_id[BUILD_ID_SIZE];
+
+		if (elf_read_build_id(elf, build_id,
+				      BUILD_ID_SIZE) != BUILD_ID_SIZE)
+			goto out_elf_end;
+
+		if (!dso__build_id_equal(dso, build_id))
+			goto out_elf_end;
+	}
+
+	sec = elf_section_by_name(elf, &ehdr, &shdr, ".symtab", NULL);
+	if (sec == NULL) {
+		if (want_symtab)
+			goto out_elf_end;
+
+		sec = elf_section_by_name(elf, &ehdr, &shdr, ".dynsym", NULL);
+		if (sec == NULL)
+			goto out_elf_end;
+	}
+
+	opdsec = elf_section_by_name(elf, &ehdr, &opdshdr, ".opd", &opdidx);
+	if (opdshdr.sh_type != SHT_PROGBITS)
+		opdsec = NULL;
+	if (opdsec)
+		opddata = elf_rawdata(opdsec, NULL);
+
+	syms = elf_getdata(sec, NULL);
+	if (syms == NULL)
+		goto out_elf_end;
+
+	sec = elf_getscn(elf, shdr.sh_link);
+	if (sec == NULL)
+		goto out_elf_end;
+
+	symstrs = elf_getdata(sec, NULL);
+	if (symstrs == NULL)
+		goto out_elf_end;
+
+	sec_strndx = elf_getscn(elf, ehdr.e_shstrndx);
+	if (sec_strndx == NULL)
+		goto out_elf_end;
+
+	secstrs = elf_getdata(sec_strndx, NULL);
+	if (secstrs == NULL)
+		goto out_elf_end;
+
+	nr_syms = shdr.sh_size / shdr.sh_entsize;
+
+	memset(&sym, 0, sizeof(sym));
+	if (dso->kernel == DSO_TYPE_USER) {
+		dso->adjust_symbols = (ehdr.e_type == ET_EXEC ||
+				elf_section_by_name(elf, &ehdr, &shdr,
+						     ".gnu.prelink_undo",
+						     NULL) != NULL);
+	} else {
+		dso->adjust_symbols = 0;
+	}
+	elf_symtab__for_each_symbol(syms, nr_syms, idx, sym) {
+		struct symbol *f;
+		const char *elf_name = elf_sym__name(&sym, symstrs);
+		char *demangled = NULL;
+		int is_label = elf_sym__is_label(&sym);
+		const char *section_name;
+
+		if (kmap && kmap->ref_reloc_sym && kmap->ref_reloc_sym->name &&
+		    strcmp(elf_name, kmap->ref_reloc_sym->name) == 0)
+			kmap->ref_reloc_sym->unrelocated_addr = sym.st_value;
+
+		if (!is_label && !elf_sym__is_a(&sym, map->type))
+			continue;
+
+		/* Reject ARM ELF "mapping symbols": these aren't unique and
+		 * don't identify functions, so will confuse the profile
+		 * output: */
+		if (ehdr.e_machine == EM_ARM) {
+			if (!strcmp(elf_name, "$a") ||
+			    !strcmp(elf_name, "$d") ||
+			    !strcmp(elf_name, "$t"))
+				continue;
+		}
+
+		if (opdsec && sym.st_shndx == opdidx) {
+			u32 offset = sym.st_value - opdshdr.sh_addr;
+			u64 *opd = opddata->d_buf + offset;
+			sym.st_value = *opd;
+			sym.st_shndx = elf_addr_to_index(elf, sym.st_value);
+		}
+
+		sec = elf_getscn(elf, sym.st_shndx);
+		if (!sec)
+			goto out_elf_end;
+
+		gelf_getshdr(sec, &shdr);
+
+		if (is_label && !elf_sec__is_a(&shdr, secstrs, map->type))
+			continue;
+
+		section_name = elf_sec__name(&shdr, secstrs);
+
+		/* On ARM, symbols for thumb functions have 1 added to
+		 * the symbol address as a flag - remove it */
+		if ((ehdr.e_machine == EM_ARM) &&
+		    (map->type == MAP__FUNCTION) &&
+		    (sym.st_value & 1))
+			--sym.st_value;
+
+		if (dso->kernel != DSO_TYPE_USER || kmodule) {
+			char dso_name[PATH_MAX];
+
+			if (strcmp(section_name,
+				   (curr_dso->short_name +
+				    dso->short_name_len)) == 0)
+				goto new_symbol;
+
+			if (strcmp(section_name, ".text") == 0) {
+				curr_map = map;
+				curr_dso = dso;
+				goto new_symbol;
+			}
+
+			snprintf(dso_name, sizeof(dso_name),
+				 "%s%s", dso->short_name, section_name);
+
+			curr_map = map_groups__find_by_name(kmap->kmaps, map->type, dso_name);
+			if (curr_map == NULL) {
+				u64 start = sym.st_value;
+
+				if (kmodule)
+					start += map->start + shdr.sh_offset;
+
+				curr_dso = dso__new(dso_name);
+				if (curr_dso == NULL)
+					goto out_elf_end;
+				curr_dso->kernel = dso->kernel;
+				curr_dso->long_name = dso->long_name;
+				curr_dso->long_name_len = dso->long_name_len;
+				curr_map = map__new2(start, curr_dso,
+						     map->type);
+				if (curr_map == NULL) {
+					dso__delete(curr_dso);
+					goto out_elf_end;
+				}
+				curr_map->map_ip = identity__map_ip;
+				curr_map->unmap_ip = identity__map_ip;
+				curr_dso->symtab_type = dso->symtab_type;
+				map_groups__insert(kmap->kmaps, curr_map);
+				dsos__add(&dso->node, curr_dso);
+				dso__set_loaded(curr_dso, map->type);
+			} else
+				curr_dso = curr_map->dso;
+
+			goto new_symbol;
+		}
+
+		if (curr_dso->adjust_symbols) {
+			pr_debug4("%s: adjusting symbol: st_value: %#" PRIx64 " "
+				  "sh_addr: %#" PRIx64 " sh_offset: %#" PRIx64 "\n", __func__,
+				  (u64)sym.st_value, (u64)shdr.sh_addr,
+				  (u64)shdr.sh_offset);
+			sym.st_value -= shdr.sh_addr - shdr.sh_offset;
+		}
+		/*
+		 * We need to figure out if the object was created from C++ sources
+		 * DWARF DW_compile_unit has this, but we don't always have access
+		 * to it...
+		 */
+		demangled = bfd_demangle(NULL, elf_name, DMGL_PARAMS | DMGL_ANSI);
+		if (demangled != NULL)
+			elf_name = demangled;
+new_symbol:
+		f = symbol__new(sym.st_value, sym.st_size,
+				GELF_ST_BIND(sym.st_info), elf_name);
+		free(demangled);
+		if (!f)
+			goto out_elf_end;
+
+		if (filter && filter(curr_map, f))
+			symbol__delete(f);
+		else {
+			symbols__insert(&curr_dso->symbols[curr_map->type], f);
+			nr++;
+		}
+	}
+
+	/*
+	 * For misannotated, zeroed, ASM function sizes.
+	 */
+	if (nr > 0) {
+		symbols__fixup_end(&dso->symbols[map->type]);
+		if (kmap) {
+			/*
+			 * We need to fixup this here too because we create new
+			 * maps here, for things like vsyscall sections.
+			 */
+			__map_groups__fixup_end(kmap->kmaps, map->type);
+		}
+	}
+	err = nr;
+out_elf_end:
+	elf_end(elf);
+out_close:
+	return err;
+}
+
+static bool dso__build_id_equal(const struct dso *dso, u8 *build_id)
+{
+	return memcmp(dso->build_id, build_id, sizeof(dso->build_id)) == 0;
+}
+
+bool __dsos__read_build_ids(struct list_head *head, bool with_hits)
+{
+	bool have_build_id = false;
+	struct dso *pos;
+
+	list_for_each_entry(pos, head, node) {
+		if (with_hits && !pos->hit)
+			continue;
+		if (pos->has_build_id) {
+			have_build_id = true;
+			continue;
+		}
+		if (filename__read_build_id(pos->long_name, pos->build_id,
+					    sizeof(pos->build_id)) > 0) {
+			have_build_id	  = true;
+			pos->has_build_id = true;
+		}
+	}
+
+	return have_build_id;
+}
+
+/*
+ * Align offset to 4 bytes as needed for note name and descriptor data.
+ */
+#define NOTE_ALIGN(n) (((n) + 3) & -4U)
+
+static int elf_read_build_id(Elf *elf, void *bf, size_t size)
+{
+	int err = -1;
+	GElf_Ehdr ehdr;
+	GElf_Shdr shdr;
+	Elf_Data *data;
+	Elf_Scn *sec;
+	Elf_Kind ek;
+	void *ptr;
+
+	if (size < BUILD_ID_SIZE)
+		goto out;
+
+	ek = elf_kind(elf);
+	if (ek != ELF_K_ELF)
+		goto out;
+
+	if (gelf_getehdr(elf, &ehdr) == NULL) {
+		pr_err("%s: cannot get elf header.\n", __func__);
+		goto out;
+	}
+
+	sec = elf_section_by_name(elf, &ehdr, &shdr,
+				  ".note.gnu.build-id", NULL);
+	if (sec == NULL) {
+		sec = elf_section_by_name(elf, &ehdr, &shdr,
+					  ".notes", NULL);
+		if (sec == NULL)
+			goto out;
+	}
+
+	data = elf_getdata(sec, NULL);
+	if (data == NULL)
+		goto out;
+
+	ptr = data->d_buf;
+	while (ptr < (data->d_buf + data->d_size)) {
+		GElf_Nhdr *nhdr = ptr;
+		int namesz = NOTE_ALIGN(nhdr->n_namesz),
+		    descsz = NOTE_ALIGN(nhdr->n_descsz);
+		const char *name;
+
+		ptr += sizeof(*nhdr);
+		name = ptr;
+		ptr += namesz;
+		if (nhdr->n_type == NT_GNU_BUILD_ID &&
+		    nhdr->n_namesz == sizeof("GNU")) {
+			if (memcmp(name, "GNU", sizeof("GNU")) == 0) {
+				memcpy(bf, ptr, BUILD_ID_SIZE);
+				err = BUILD_ID_SIZE;
+				break;
+			}
+		}
+		ptr += descsz;
+	}
+
+out:
+	return err;
+}
+
+int filename__read_build_id(const char *filename, void *bf, size_t size)
+{
+	int fd, err = -1;
+	Elf *elf;
+
+	if (size < BUILD_ID_SIZE)
+		goto out;
+
+	fd = open(filename, O_RDONLY);
+	if (fd < 0)
+		goto out;
+
+	elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL);
+	if (elf == NULL) {
+		pr_debug2("%s: cannot read %s ELF file.\n", __func__, filename);
+		goto out_close;
+	}
+
+	err = elf_read_build_id(elf, bf, size);
+
+	elf_end(elf);
+out_close:
+	close(fd);
+out:
+	return err;
+}
+
+int sysfs__read_build_id(const char *filename, void *build_id, size_t size)
+{
+	int fd, err = -1;
+
+	if (size < BUILD_ID_SIZE)
+		goto out;
+
+	fd = open(filename, O_RDONLY);
+	if (fd < 0)
+		goto out;
+
+	while (1) {
+		char bf[BUFSIZ];
+		GElf_Nhdr nhdr;
+		int namesz, descsz;
+
+		if (read(fd, &nhdr, sizeof(nhdr)) != sizeof(nhdr))
+			break;
+
+		namesz = NOTE_ALIGN(nhdr.n_namesz);
+		descsz = NOTE_ALIGN(nhdr.n_descsz);
+		if (nhdr.n_type == NT_GNU_BUILD_ID &&
+		    nhdr.n_namesz == sizeof("GNU")) {
+			if (read(fd, bf, namesz) != namesz)
+				break;
+			if (memcmp(bf, "GNU", sizeof("GNU")) == 0) {
+				if (read(fd, build_id,
+				    BUILD_ID_SIZE) == BUILD_ID_SIZE) {
+					err = 0;
+					break;
+				}
+			} else if (read(fd, bf, descsz) != descsz)
+				break;
+		} else {
+			int n = namesz + descsz;
+			if (read(fd, bf, n) != n)
+				break;
+		}
+	}
+	close(fd);
+out:
+	return err;
+}
+
+char dso__symtab_origin(const struct dso *dso)
+{
+	static const char origin[] = {
+		[SYMTAB__KALLSYMS]	      = 'k',
+		[SYMTAB__JAVA_JIT]	      = 'j',
+		[SYMTAB__BUILD_ID_CACHE]      = 'B',
+		[SYMTAB__FEDORA_DEBUGINFO]    = 'f',
+		[SYMTAB__UBUNTU_DEBUGINFO]    = 'u',
+		[SYMTAB__BUILDID_DEBUGINFO]   = 'b',
+		[SYMTAB__SYSTEM_PATH_DSO]     = 'd',
+		[SYMTAB__SYSTEM_PATH_KMODULE] = 'K',
+		[SYMTAB__GUEST_KALLSYMS]      =  'g',
+		[SYMTAB__GUEST_KMODULE]	      =  'G',
+	};
+
+	if (dso == NULL || dso->symtab_type == SYMTAB__NOT_FOUND)
+		return '!';
+	return origin[dso->symtab_type];
+}
+
+int dso__load(struct dso *dso, struct map *map, symbol_filter_t filter)
+{
+	int size = PATH_MAX;
+	char *name;
+	int ret = -1;
+	int fd;
+	struct machine *machine;
+	const char *root_dir;
+	int want_symtab;
+
+	dso__set_loaded(dso, map->type);
+
+	if (dso->kernel == DSO_TYPE_KERNEL)
+		return dso__load_kernel_sym(dso, map, filter);
+	else if (dso->kernel == DSO_TYPE_GUEST_KERNEL)
+		return dso__load_guest_kernel_sym(dso, map, filter);
+
+	if (map->groups && map->groups->machine)
+		machine = map->groups->machine;
+	else
+		machine = NULL;
+
+	name = malloc(size);
+	if (!name)
+		return -1;
+
+	dso->adjust_symbols = 0;
+
+	if (strncmp(dso->name, "/tmp/perf-", 10) == 0) {
+		ret = dso__load_perf_map(dso, map, filter);
+		dso->symtab_type = ret > 0 ? SYMTAB__JAVA_JIT :
+					      SYMTAB__NOT_FOUND;
+		return ret;
+	}
+
+	/* Iterate over candidate debug images.
+	 * On the first pass, only load images if they have a full symtab.
+	 * Failing that, do a second pass where we accept .dynsym also
+	 */
+	want_symtab = 1;
+restart:
+	for (dso->symtab_type = SYMTAB__BUILD_ID_CACHE;
+	     dso->symtab_type != SYMTAB__NOT_FOUND;
+	     dso->symtab_type++) {
+		switch (dso->symtab_type) {
+		case SYMTAB__BUILD_ID_CACHE:
+			/* skip the locally configured cache if a symfs is given */
+			if (symbol_conf.symfs[0] ||
+			    (dso__build_id_filename(dso, name, size) == NULL)) {
+				continue;
+			}
+			break;
+		case SYMTAB__FEDORA_DEBUGINFO:
+			snprintf(name, size, "%s/usr/lib/debug%s.debug",
+				 symbol_conf.symfs, dso->long_name);
+			break;
+		case SYMTAB__UBUNTU_DEBUGINFO:
+			snprintf(name, size, "%s/usr/lib/debug%s",
+				 symbol_conf.symfs, dso->long_name);
+			break;
+		case SYMTAB__BUILDID_DEBUGINFO: {
+			char build_id_hex[BUILD_ID_SIZE * 2 + 1];
+
+			if (!dso->has_build_id)
+				continue;
+
+			build_id__sprintf(dso->build_id,
+					  sizeof(dso->build_id),
+					  build_id_hex);
+			snprintf(name, size,
+				 "%s/usr/lib/debug/.build-id/%.2s/%s.debug",
+				 symbol_conf.symfs, build_id_hex, build_id_hex + 2);
+			}
+			break;
+		case SYMTAB__SYSTEM_PATH_DSO:
+			snprintf(name, size, "%s%s",
+			     symbol_conf.symfs, dso->long_name);
+			break;
+		case SYMTAB__GUEST_KMODULE:
+			if (map->groups && machine)
+				root_dir = machine->root_dir;
+			else
+				root_dir = "";
+			snprintf(name, size, "%s%s%s", symbol_conf.symfs,
+				 root_dir, dso->long_name);
+			break;
+
+		case SYMTAB__SYSTEM_PATH_KMODULE:
+			snprintf(name, size, "%s%s", symbol_conf.symfs,
+				 dso->long_name);
+			break;
+		default:;
+		}
+
+		/* Name is now the name of the next image to try */
+		fd = open(name, O_RDONLY);
+		if (fd < 0)
+			continue;
+
+		ret = dso__load_sym(dso, map, name, fd, filter, 0,
+				    want_symtab);
+		close(fd);
+
+		/*
+		 * Some people seem to have debuginfo files _WITHOUT_ debug
+		 * info!?!?
+		 */
+		if (!ret)
+			continue;
+
+		if (ret > 0) {
+			int nr_plt = dso__synthesize_plt_symbols(dso, map,
+								 filter);
+			if (nr_plt > 0)
+				ret += nr_plt;
+			break;
+		}
+	}
+
+	/*
+	 * If we wanted a full symtab but no image had one,
+	 * relax our requirements and repeat the search.
+	 */
+	if (ret <= 0 && want_symtab) {
+		want_symtab = 0;
+		goto restart;
+	}
+
+	free(name);
+	if (ret < 0 && strstr(dso->name, " (deleted)") != NULL)
+		return 0;
+	return ret;
+}
+
+struct map *map_groups__find_by_name(struct map_groups *mg,
+				     enum map_type type, const char *name)
+{
+	struct rb_node *nd;
+
+	for (nd = rb_first(&mg->maps[type]); nd; nd = rb_next(nd)) {
+		struct map *map = rb_entry(nd, struct map, rb_node);
+
+		if (map->dso && strcmp(map->dso->short_name, name) == 0)
+			return map;
+	}
+
+	return NULL;
+}
+
+static int dso__kernel_module_get_build_id(struct dso *dso,
+					   const char *root_dir)
+{
+	char filename[PATH_MAX];
+	/*
+	 * kernel module short names are of the form "[module]" and
+	 * we need just "module" here.
+	 */
+	const char *name = dso->short_name + 1;
+
+	snprintf(filename, sizeof(filename),
+		 "%s/sys/module/%.*s/notes/.note.gnu.build-id",
+		 root_dir, (int)strlen(name) - 1, name);
+
+	if (sysfs__read_build_id(filename, dso->build_id,
+				 sizeof(dso->build_id)) == 0)
+		dso->has_build_id = true;
+
+	return 0;
+}
+
+static int map_groups__set_modules_path_dir(struct map_groups *mg,
+				const char *dir_name)
+{
+	struct dirent *dent;
+	DIR *dir = opendir(dir_name);
+	int ret = 0;
+
+	if (!dir) {
+		pr_debug("%s: cannot open %s dir\n", __func__, dir_name);
+		return -1;
+	}
+
+	while ((dent = readdir(dir)) != NULL) {
+		char path[PATH_MAX];
+		struct stat st;
+
+		/*sshfs might return bad dent->d_type, so we have to stat*/
+		sprintf(path, "%s/%s", dir_name, dent->d_name);
+		if (stat(path, &st))
+			continue;
+
+		if (S_ISDIR(st.st_mode)) {
+			if (!strcmp(dent->d_name, ".") ||
+			    !strcmp(dent->d_name, ".."))
+				continue;
+
+			snprintf(path, sizeof(path), "%s/%s",
+				 dir_name, dent->d_name);
+			ret = map_groups__set_modules_path_dir(mg, path);
+			if (ret < 0)
+				goto out;
+		} else {
+			char *dot = strrchr(dent->d_name, '.'),
+			     dso_name[PATH_MAX];
+			struct map *map;
+			char *long_name;
+
+			if (dot == NULL || strcmp(dot, ".ko"))
+				continue;
+			snprintf(dso_name, sizeof(dso_name), "[%.*s]",
+				 (int)(dot - dent->d_name), dent->d_name);
+
+			strxfrchar(dso_name, '-', '_');
+			map = map_groups__find_by_name(mg, MAP__FUNCTION,
+						       dso_name);
+			if (map == NULL)
+				continue;
+
+			snprintf(path, sizeof(path), "%s/%s",
+				 dir_name, dent->d_name);
+
+			long_name = strdup(path);
+			if (long_name == NULL) {
+				ret = -1;
+				goto out;
+			}
+			dso__set_long_name(map->dso, long_name);
+			map->dso->lname_alloc = 1;
+			dso__kernel_module_get_build_id(map->dso, "");
+		}
+	}
+
+out:
+	closedir(dir);
+	return ret;
+}
+
+static char *get_kernel_version(const char *root_dir)
+{
+	char version[PATH_MAX];
+	FILE *file;
+	char *name, *tmp;
+	const char *prefix = "Linux version ";
+
+	sprintf(version, "%s/proc/version", root_dir);
+	file = fopen(version, "r");
+	if (!file)
+		return NULL;
+
+	version[0] = '\0';
+	tmp = fgets(version, sizeof(version), file);
+	fclose(file);
+
+	name = strstr(version, prefix);
+	if (!name)
+		return NULL;
+	name += strlen(prefix);
+	tmp = strchr(name, ' ');
+	if (tmp)
+		*tmp = '\0';
+
+	return strdup(name);
+}
+
+static int machine__set_modules_path(struct machine *machine)
+{
+	char *version;
+	char modules_path[PATH_MAX];
+
+	version = get_kernel_version(machine->root_dir);
+	if (!version)
+		return -1;
+
+	snprintf(modules_path, sizeof(modules_path), "%s/lib/modules/%s/kernel",
+		 machine->root_dir, version);
+	free(version);
+
+	return map_groups__set_modules_path_dir(&machine->kmaps, modules_path);
+}
+
+/*
+ * Constructor variant for modules (where we know from /proc/modules where
+ * they are loaded) and for vmlinux, where only after we load all the
+ * symbols we'll know where it starts and ends.
+ */
+static struct map *map__new2(u64 start, struct dso *dso, enum map_type type)
+{
+	struct map *map = calloc(1, (sizeof(*map) +
+				     (dso->kernel ? sizeof(struct kmap) : 0)));
+	if (map != NULL) {
+		/*
+		 * ->end will be filled after we load all the symbols
+		 */
+		map__init(map, type, start, 0, 0, dso);
+	}
+
+	return map;
+}
+
+struct map *machine__new_module(struct machine *machine, u64 start,
+				const char *filename)
+{
+	struct map *map;
+	struct dso *dso = __dsos__findnew(&machine->kernel_dsos, filename);
+
+	if (dso == NULL)
+		return NULL;
+
+	map = map__new2(start, dso, MAP__FUNCTION);
+	if (map == NULL)
+		return NULL;
+
+	if (machine__is_host(machine))
+		dso->symtab_type = SYMTAB__SYSTEM_PATH_KMODULE;
+	else
+		dso->symtab_type = SYMTAB__GUEST_KMODULE;
+	map_groups__insert(&machine->kmaps, map);
+	return map;
+}
+
+static int machine__create_modules(struct machine *machine)
+{
+	char *line = NULL;
+	size_t n;
+	FILE *file;
+	struct map *map;
+	const char *modules;
+	char path[PATH_MAX];
+
+	if (machine__is_default_guest(machine))
+		modules = symbol_conf.default_guest_modules;
+	else {
+		sprintf(path, "%s/proc/modules", machine->root_dir);
+		modules = path;
+	}
+
+	if (symbol__restricted_filename(path, "/proc/modules"))
+		return -1;
+
+	file = fopen(modules, "r");
+	if (file == NULL)
+		return -1;
+
+	while (!feof(file)) {
+		char name[PATH_MAX];
+		u64 start;
+		char *sep;
+		int line_len;
+
+		line_len = getline(&line, &n, file);
+		if (line_len < 0)
+			break;
+
+		if (!line)
+			goto out_failure;
+
+		line[--line_len] = '\0'; /* \n */
+
+		sep = strrchr(line, 'x');
+		if (sep == NULL)
+			continue;
+
+		hex2u64(sep + 1, &start);
+
+		sep = strchr(line, ' ');
+		if (sep == NULL)
+			continue;
+
+		*sep = '\0';
+
+		snprintf(name, sizeof(name), "[%s]", line);
+		map = machine__new_module(machine, start, name);
+		if (map == NULL)
+			goto out_delete_line;
+		dso__kernel_module_get_build_id(map->dso, machine->root_dir);
+	}
+
+	free(line);
+	fclose(file);
+
+	return machine__set_modules_path(machine);
+
+out_delete_line:
+	free(line);
+out_failure:
+	return -1;
+}
+
+int dso__load_vmlinux(struct dso *dso, struct map *map,
+		      const char *vmlinux, symbol_filter_t filter)
+{
+	int err = -1, fd;
+	char symfs_vmlinux[PATH_MAX];
+
+	snprintf(symfs_vmlinux, sizeof(symfs_vmlinux), "%s%s",
+		 symbol_conf.symfs, vmlinux);
+	fd = open(symfs_vmlinux, O_RDONLY);
+	if (fd < 0)
+		return -1;
+
+	dso__set_long_name(dso, (char *)vmlinux);
+	dso__set_loaded(dso, map->type);
+	err = dso__load_sym(dso, map, symfs_vmlinux, fd, filter, 0, 0);
+	close(fd);
+
+	if (err > 0)
+		pr_debug("Using %s for symbols\n", symfs_vmlinux);
+
+	return err;
+}
+
+int dso__load_vmlinux_path(struct dso *dso, struct map *map,
+			   symbol_filter_t filter)
+{
+	int i, err = 0;
+	char *filename;
+
+	pr_debug("Looking at the vmlinux_path (%d entries long)\n",
+		 vmlinux_path__nr_entries + 1);
+
+	filename = dso__build_id_filename(dso, NULL, 0);
+	if (filename != NULL) {
+		err = dso__load_vmlinux(dso, map, filename, filter);
+		if (err > 0) {
+			dso__set_long_name(dso, filename);
+			goto out;
+		}
+		free(filename);
+	}
+
+	for (i = 0; i < vmlinux_path__nr_entries; ++i) {
+		err = dso__load_vmlinux(dso, map, vmlinux_path[i], filter);
+		if (err > 0) {
+			dso__set_long_name(dso, strdup(vmlinux_path[i]));
+			break;
+		}
+	}
+out:
+	return err;
+}
+
+static int dso__load_kernel_sym(struct dso *dso, struct map *map,
+				symbol_filter_t filter)
+{
+	int err;
+	const char *kallsyms_filename = NULL;
+	char *kallsyms_allocated_filename = NULL;
+	/*
+	 * Step 1: if the user specified a kallsyms or vmlinux filename, use
+	 * it and only it, reporting errors to the user if it cannot be used.
+	 *
+	 * For instance, try to analyse an ARM perf.data file _without_ a
+	 * build-id, or if the user specifies the wrong path to the right
+	 * vmlinux file, obviously we can't fallback to another vmlinux (a
+	 * x86_86 one, on the machine where analysis is being performed, say),
+	 * or worse, /proc/kallsyms.
+	 *
+	 * If the specified file _has_ a build-id and there is a build-id
+	 * section in the perf.data file, we will still do the expected
+	 * validation in dso__load_vmlinux and will bail out if they don't
+	 * match.
+	 */
+	if (symbol_conf.kallsyms_name != NULL) {
+		kallsyms_filename = symbol_conf.kallsyms_name;
+		goto do_kallsyms;
+	}
+
+	if (symbol_conf.vmlinux_name != NULL) {
+		err = dso__load_vmlinux(dso, map,
+					symbol_conf.vmlinux_name, filter);
+		if (err > 0) {
+			dso__set_long_name(dso,
+					   strdup(symbol_conf.vmlinux_name));
+			goto out_fixup;
+		}
+		return err;
+	}
+
+	if (vmlinux_path != NULL) {
+		err = dso__load_vmlinux_path(dso, map, filter);
+		if (err > 0)
+			goto out_fixup;
+	}
+
+	/* do not try local files if a symfs was given */
+	if (symbol_conf.symfs[0] != 0)
+		return -1;
+
+	/*
+	 * Say the kernel DSO was created when processing the build-id header table,
+	 * we have a build-id, so check if it is the same as the running kernel,
+	 * using it if it is.
+	 */
+	if (dso->has_build_id) {
+		u8 kallsyms_build_id[BUILD_ID_SIZE];
+		char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+		if (sysfs__read_build_id("/sys/kernel/notes", kallsyms_build_id,
+					 sizeof(kallsyms_build_id)) == 0) {
+			if (dso__build_id_equal(dso, kallsyms_build_id)) {
+				kallsyms_filename = "/proc/kallsyms";
+				goto do_kallsyms;
+			}
+		}
+		/*
+		 * Now look if we have it on the build-id cache in
+		 * $HOME/.debug/[kernel.kallsyms].
+		 */
+		build_id__sprintf(dso->build_id, sizeof(dso->build_id),
+				  sbuild_id);
+
+		if (asprintf(&kallsyms_allocated_filename,
+			     "%s/.debug/[kernel.kallsyms]/%s",
+			     getenv("HOME"), sbuild_id) == -1) {
+			pr_err("Not enough memory for kallsyms file lookup\n");
+			return -1;
+		}
+
+		kallsyms_filename = kallsyms_allocated_filename;
+
+		if (access(kallsyms_filename, F_OK)) {
+			pr_err("No kallsyms or vmlinux with build-id %s "
+			       "was found\n", sbuild_id);
+			free(kallsyms_allocated_filename);
+			return -1;
+		}
+	} else {
+		/*
+		 * Last resort, if we don't have a build-id and couldn't find
+		 * any vmlinux file, try the running kernel kallsyms table.
+		 */
+		kallsyms_filename = "/proc/kallsyms";
+	}
+
+do_kallsyms:
+	err = dso__load_kallsyms(dso, kallsyms_filename, map, filter);
+	if (err > 0)
+		pr_debug("Using %s for symbols\n", kallsyms_filename);
+	free(kallsyms_allocated_filename);
+
+	if (err > 0) {
+out_fixup:
+		if (kallsyms_filename != NULL)
+			dso__set_long_name(dso, strdup("[kernel.kallsyms]"));
+		map__fixup_start(map);
+		map__fixup_end(map);
+	}
+
+	return err;
+}
+
+static int dso__load_guest_kernel_sym(struct dso *dso, struct map *map,
+				      symbol_filter_t filter)
+{
+	int err;
+	const char *kallsyms_filename = NULL;
+	struct machine *machine;
+	char path[PATH_MAX];
+
+	if (!map->groups) {
+		pr_debug("Guest kernel map hasn't the point to groups\n");
+		return -1;
+	}
+	machine = map->groups->machine;
+
+	if (machine__is_default_guest(machine)) {
+		/*
+		 * if the user specified a vmlinux filename, use it and only
+		 * it, reporting errors to the user if it cannot be used.
+		 * Or use file guest_kallsyms inputted by user on commandline
+		 */
+		if (symbol_conf.default_guest_vmlinux_name != NULL) {
+			err = dso__load_vmlinux(dso, map,
+				symbol_conf.default_guest_vmlinux_name, filter);
+			goto out_try_fixup;
+		}
+
+		kallsyms_filename = symbol_conf.default_guest_kallsyms;
+		if (!kallsyms_filename)
+			return -1;
+	} else {
+		sprintf(path, "%s/proc/kallsyms", machine->root_dir);
+		kallsyms_filename = path;
+	}
+
+	err = dso__load_kallsyms(dso, kallsyms_filename, map, filter);
+	if (err > 0)
+		pr_debug("Using %s for symbols\n", kallsyms_filename);
+
+out_try_fixup:
+	if (err > 0) {
+		if (kallsyms_filename != NULL) {
+			machine__mmap_name(machine, path, sizeof(path));
+			dso__set_long_name(dso, strdup(path));
+		}
+		map__fixup_start(map);
+		map__fixup_end(map);
+	}
+
+	return err;
+}
+
+static void dsos__add(struct list_head *head, struct dso *dso)
+{
+	list_add_tail(&dso->node, head);
+}
+
+static struct dso *dsos__find(struct list_head *head, const char *name)
+{
+	struct dso *pos;
+
+	list_for_each_entry(pos, head, node)
+		if (strcmp(pos->long_name, name) == 0)
+			return pos;
+	return NULL;
+}
+
+struct dso *__dsos__findnew(struct list_head *head, const char *name)
+{
+	struct dso *dso = dsos__find(head, name);
+
+	if (!dso) {
+		dso = dso__new(name);
+		if (dso != NULL) {
+			dsos__add(head, dso);
+			dso__set_basename(dso);
+		}
+	}
+
+	return dso;
+}
+
+size_t __dsos__fprintf(struct list_head *head, FILE *fp)
+{
+	struct dso *pos;
+	size_t ret = 0;
+
+	list_for_each_entry(pos, head, node) {
+		int i;
+		for (i = 0; i < MAP__NR_TYPES; ++i)
+			ret += dso__fprintf(pos, i, fp);
+	}
+
+	return ret;
+}
+
+size_t machines__fprintf_dsos(struct rb_root *machines, FILE *fp)
+{
+	struct rb_node *nd;
+	size_t ret = 0;
+
+	for (nd = rb_first(machines); nd; nd = rb_next(nd)) {
+		struct machine *pos = rb_entry(nd, struct machine, rb_node);
+		ret += __dsos__fprintf(&pos->kernel_dsos, fp);
+		ret += __dsos__fprintf(&pos->user_dsos, fp);
+	}
+
+	return ret;
+}
+
+static size_t __dsos__fprintf_buildid(struct list_head *head, FILE *fp,
+				      bool with_hits)
+{
+	struct dso *pos;
+	size_t ret = 0;
+
+	list_for_each_entry(pos, head, node) {
+		if (with_hits && !pos->hit)
+			continue;
+		ret += dso__fprintf_buildid(pos, fp);
+		ret += fprintf(fp, " %s\n", pos->long_name);
+	}
+	return ret;
+}
+
+size_t machine__fprintf_dsos_buildid(struct machine *machine, FILE *fp,
+				     bool with_hits)
+{
+	return __dsos__fprintf_buildid(&machine->kernel_dsos, fp, with_hits) +
+	       __dsos__fprintf_buildid(&machine->user_dsos, fp, with_hits);
+}
+
+size_t machines__fprintf_dsos_buildid(struct rb_root *machines,
+				      FILE *fp, bool with_hits)
+{
+	struct rb_node *nd;
+	size_t ret = 0;
+
+	for (nd = rb_first(machines); nd; nd = rb_next(nd)) {
+		struct machine *pos = rb_entry(nd, struct machine, rb_node);
+		ret += machine__fprintf_dsos_buildid(pos, fp, with_hits);
+	}
+	return ret;
+}
+
+struct dso *dso__new_kernel(const char *name)
+{
+	struct dso *dso = dso__new(name ?: "[kernel.kallsyms]");
+
+	if (dso != NULL) {
+		dso__set_short_name(dso, "[kernel]");
+		dso->kernel = DSO_TYPE_KERNEL;
+	}
+
+	return dso;
+}
+
+static struct dso *dso__new_guest_kernel(struct machine *machine,
+					const char *name)
+{
+	char bf[PATH_MAX];
+	struct dso *dso = dso__new(name ?: machine__mmap_name(machine, bf,
+							      sizeof(bf)));
+	if (dso != NULL) {
+		dso__set_short_name(dso, "[guest.kernel]");
+		dso->kernel = DSO_TYPE_GUEST_KERNEL;
+	}
+
+	return dso;
+}
+
+void dso__read_running_kernel_build_id(struct dso *dso, struct machine *machine)
+{
+	char path[PATH_MAX];
+
+	if (machine__is_default_guest(machine))
+		return;
+	sprintf(path, "%s/sys/kernel/notes", machine->root_dir);
+	if (sysfs__read_build_id(path, dso->build_id,
+				 sizeof(dso->build_id)) == 0)
+		dso->has_build_id = true;
+}
+
+static struct dso *machine__create_kernel(struct machine *machine)
+{
+	const char *vmlinux_name = NULL;
+	struct dso *kernel;
+
+	if (machine__is_host(machine)) {
+		vmlinux_name = symbol_conf.vmlinux_name;
+		kernel = dso__new_kernel(vmlinux_name);
+	} else {
+		if (machine__is_default_guest(machine))
+			vmlinux_name = symbol_conf.default_guest_vmlinux_name;
+		kernel = dso__new_guest_kernel(machine, vmlinux_name);
+	}
+
+	if (kernel != NULL) {
+		dso__read_running_kernel_build_id(kernel, machine);
+		dsos__add(&machine->kernel_dsos, kernel);
+	}
+	return kernel;
+}
+
+struct process_args {
+	u64 start;
+};
+
+static int symbol__in_kernel(void *arg, const char *name,
+			     char type __used, u64 start, u64 end __used)
+{
+	struct process_args *args = arg;
+
+	if (strchr(name, '['))
+		return 0;
+
+	args->start = start;
+	return 1;
+}
+
+/* Figure out the start address of kernel map from /proc/kallsyms */
+static u64 machine__get_kernel_start_addr(struct machine *machine)
+{
+	const char *filename;
+	char path[PATH_MAX];
+	struct process_args args;
+
+	if (machine__is_host(machine)) {
+		filename = "/proc/kallsyms";
+	} else {
+		if (machine__is_default_guest(machine))
+			filename = (char *)symbol_conf.default_guest_kallsyms;
+		else {
+			sprintf(path, "%s/proc/kallsyms", machine->root_dir);
+			filename = path;
+		}
+	}
+
+	if (symbol__restricted_filename(filename, "/proc/kallsyms"))
+		return 0;
+
+	if (kallsyms__parse(filename, &args, symbol__in_kernel) <= 0)
+		return 0;
+
+	return args.start;
+}
+
+int __machine__create_kernel_maps(struct machine *machine, struct dso *kernel)
+{
+	enum map_type type;
+	u64 start = machine__get_kernel_start_addr(machine);
+
+	for (type = 0; type < MAP__NR_TYPES; ++type) {
+		struct kmap *kmap;
+
+		machine->vmlinux_maps[type] = map__new2(start, kernel, type);
+		if (machine->vmlinux_maps[type] == NULL)
+			return -1;
+
+		machine->vmlinux_maps[type]->map_ip =
+			machine->vmlinux_maps[type]->unmap_ip =
+				identity__map_ip;
+		kmap = map__kmap(machine->vmlinux_maps[type]);
+		kmap->kmaps = &machine->kmaps;
+		map_groups__insert(&machine->kmaps,
+				   machine->vmlinux_maps[type]);
+	}
+
+	return 0;
+}
+
+void machine__destroy_kernel_maps(struct machine *machine)
+{
+	enum map_type type;
+
+	for (type = 0; type < MAP__NR_TYPES; ++type) {
+		struct kmap *kmap;
+
+		if (machine->vmlinux_maps[type] == NULL)
+			continue;
+
+		kmap = map__kmap(machine->vmlinux_maps[type]);
+		map_groups__remove(&machine->kmaps,
+				   machine->vmlinux_maps[type]);
+		if (kmap->ref_reloc_sym) {
+			/*
+			 * ref_reloc_sym is shared among all maps, so free just
+			 * on one of them.
+			 */
+			if (type == MAP__FUNCTION) {
+				free((char *)kmap->ref_reloc_sym->name);
+				kmap->ref_reloc_sym->name = NULL;
+				free(kmap->ref_reloc_sym);
+			}
+			kmap->ref_reloc_sym = NULL;
+		}
+
+		map__delete(machine->vmlinux_maps[type]);
+		machine->vmlinux_maps[type] = NULL;
+	}
+}
+
+int machine__create_kernel_maps(struct machine *machine)
+{
+	struct dso *kernel = machine__create_kernel(machine);
+
+	if (kernel == NULL ||
+	    __machine__create_kernel_maps(machine, kernel) < 0)
+		return -1;
+
+	if (symbol_conf.use_modules && machine__create_modules(machine) < 0)
+		pr_debug("Problems creating module maps, continuing anyway...\n");
+	/*
+	 * Now that we have all the maps created, just set the ->end of them:
+	 */
+	map_groups__fixup_end(&machine->kmaps);
+	return 0;
+}
+
+static void vmlinux_path__exit(void)
+{
+	while (--vmlinux_path__nr_entries >= 0) {
+		free(vmlinux_path[vmlinux_path__nr_entries]);
+		vmlinux_path[vmlinux_path__nr_entries] = NULL;
+	}
+
+	free(vmlinux_path);
+	vmlinux_path = NULL;
+}
+
+static int vmlinux_path__init(void)
+{
+	struct utsname uts;
+	char bf[PATH_MAX];
+
+	vmlinux_path = malloc(sizeof(char *) * 5);
+	if (vmlinux_path == NULL)
+		return -1;
+
+	vmlinux_path[vmlinux_path__nr_entries] = strdup("vmlinux");
+	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+		goto out_fail;
+	++vmlinux_path__nr_entries;
+	vmlinux_path[vmlinux_path__nr_entries] = strdup("/boot/vmlinux");
+	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+		goto out_fail;
+	++vmlinux_path__nr_entries;
+
+	/* only try running kernel version if no symfs was given */
+	if (symbol_conf.symfs[0] != 0)
+		return 0;
+
+	if (uname(&uts) < 0)
+		return -1;
+
+	snprintf(bf, sizeof(bf), "/boot/vmlinux-%s", uts.release);
+	vmlinux_path[vmlinux_path__nr_entries] = strdup(bf);
+	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+		goto out_fail;
+	++vmlinux_path__nr_entries;
+	snprintf(bf, sizeof(bf), "/lib/modules/%s/build/vmlinux", uts.release);
+	vmlinux_path[vmlinux_path__nr_entries] = strdup(bf);
+	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+		goto out_fail;
+	++vmlinux_path__nr_entries;
+	snprintf(bf, sizeof(bf), "/usr/lib/debug/lib/modules/%s/vmlinux",
+		 uts.release);
+	vmlinux_path[vmlinux_path__nr_entries] = strdup(bf);
+	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+		goto out_fail;
+	++vmlinux_path__nr_entries;
+
+	return 0;
+
+out_fail:
+	vmlinux_path__exit();
+	return -1;
+}
+
+size_t machine__fprintf_vmlinux_path(struct machine *machine, FILE *fp)
+{
+	int i;
+	size_t printed = 0;
+	struct dso *kdso = machine->vmlinux_maps[MAP__FUNCTION]->dso;
+
+	if (kdso->has_build_id) {
+		char filename[PATH_MAX];
+		if (dso__build_id_filename(kdso, filename, sizeof(filename)))
+			printed += fprintf(fp, "[0] %s\n", filename);
+	}
+
+	for (i = 0; i < vmlinux_path__nr_entries; ++i)
+		printed += fprintf(fp, "[%d] %s\n",
+				   i + kdso->has_build_id, vmlinux_path[i]);
+
+	return printed;
+}
+
+static int setup_list(struct strlist **list, const char *list_str,
+		      const char *list_name)
+{
+	if (list_str == NULL)
+		return 0;
+
+	*list = strlist__new(true, list_str);
+	if (!*list) {
+		pr_err("problems parsing %s list\n", list_name);
+		return -1;
+	}
+	return 0;
+}
+
+static bool symbol__read_kptr_restrict(void)
+{
+	bool value = false;
+
+	if (geteuid() != 0) {
+		FILE *fp = fopen("/proc/sys/kernel/kptr_restrict", "r");
+		if (fp != NULL) {
+			char line[8];
+
+			if (fgets(line, sizeof(line), fp) != NULL)
+				value = atoi(line) != 0;
+
+			fclose(fp);
+		}
+	}
+
+	return value;
+}
+
+int symbol__init(void)
+{
+	const char *symfs;
+
+	if (symbol_conf.initialized)
+		return 0;
+
+	symbol_conf.priv_size = ALIGN(symbol_conf.priv_size, sizeof(u64));
+
+	elf_version(EV_CURRENT);
+	if (symbol_conf.sort_by_name)
+		symbol_conf.priv_size += (sizeof(struct symbol_name_rb_node) -
+					  sizeof(struct symbol));
+
+	if (symbol_conf.try_vmlinux_path && vmlinux_path__init() < 0)
+		return -1;
+
+	if (symbol_conf.field_sep && *symbol_conf.field_sep == '.') {
+		pr_err("'.' is the only non valid --field-separator argument\n");
+		return -1;
+	}
+
+	if (setup_list(&symbol_conf.dso_list,
+		       symbol_conf.dso_list_str, "dso") < 0)
+		return -1;
+
+	if (setup_list(&symbol_conf.comm_list,
+		       symbol_conf.comm_list_str, "comm") < 0)
+		goto out_free_dso_list;
+
+	if (setup_list(&symbol_conf.sym_list,
+		       symbol_conf.sym_list_str, "symbol") < 0)
+		goto out_free_comm_list;
+
+	/*
+	 * A path to symbols of "/" is identical to ""
+	 * reset here for simplicity.
+	 */
+	symfs = realpath(symbol_conf.symfs, NULL);
+	if (symfs == NULL)
+		symfs = symbol_conf.symfs;
+	if (strcmp(symfs, "/") == 0)
+		symbol_conf.symfs = "";
+	if (symfs != symbol_conf.symfs)
+		free((void *)symfs);
+
+	symbol_conf.kptr_restrict = symbol__read_kptr_restrict();
+
+	symbol_conf.initialized = true;
+	return 0;
+
+out_free_dso_list:
+	strlist__delete(symbol_conf.dso_list);
+out_free_comm_list:
+	strlist__delete(symbol_conf.comm_list);
+	return -1;
+}
+
+void symbol__exit(void)
+{
+	if (!symbol_conf.initialized)
+		return;
+	strlist__delete(symbol_conf.sym_list);
+	strlist__delete(symbol_conf.dso_list);
+	strlist__delete(symbol_conf.comm_list);
+	vmlinux_path__exit();
+	symbol_conf.sym_list = symbol_conf.dso_list = symbol_conf.comm_list = NULL;
+	symbol_conf.initialized = false;
+}
+
+int machines__create_kernel_maps(struct rb_root *machines, pid_t pid)
+{
+	struct machine *machine = machines__findnew(machines, pid);
+
+	if (machine == NULL)
+		return -1;
+
+	return machine__create_kernel_maps(machine);
+}
+
+static int hex(char ch)
+{
+	if ((ch >= '0') && (ch <= '9'))
+		return ch - '0';
+	if ((ch >= 'a') && (ch <= 'f'))
+		return ch - 'a' + 10;
+	if ((ch >= 'A') && (ch <= 'F'))
+		return ch - 'A' + 10;
+	return -1;
+}
+
+/*
+ * While we find nice hex chars, build a long_val.
+ * Return number of chars processed.
+ */
+int hex2u64(const char *ptr, u64 *long_val)
+{
+	const char *p = ptr;
+	*long_val = 0;
+
+	while (*p) {
+		const int hex_val = hex(*p);
+
+		if (hex_val < 0)
+			break;
+
+		*long_val = (*long_val << 4) | hex_val;
+		p++;
+	}
+
+	return p - ptr;
+}
+
+char *strxfrchar(char *s, char from, char to)
+{
+	char *p = s;
+
+	while ((p = strchr(p, from)) != NULL)
+		*p++ = to;
+
+	return s;
+}
+
+int machines__create_guest_kernel_maps(struct rb_root *machines)
+{
+	int ret = 0;
+	struct dirent **namelist = NULL;
+	int i, items = 0;
+	char path[PATH_MAX];
+	pid_t pid;
+
+	if (symbol_conf.default_guest_vmlinux_name ||
+	    symbol_conf.default_guest_modules ||
+	    symbol_conf.default_guest_kallsyms) {
+		machines__create_kernel_maps(machines, DEFAULT_GUEST_KERNEL_ID);
+	}
+
+	if (symbol_conf.guestmount) {
+		items = scandir(symbol_conf.guestmount, &namelist, NULL, NULL);
+		if (items <= 0)
+			return -ENOENT;
+		for (i = 0; i < items; i++) {
+			if (!isdigit(namelist[i]->d_name[0])) {
+				/* Filter out . and .. */
+				continue;
+			}
+			pid = atoi(namelist[i]->d_name);
+			sprintf(path, "%s/%s/proc/kallsyms",
+				symbol_conf.guestmount,
+				namelist[i]->d_name);
+			ret = access(path, R_OK);
+			if (ret) {
+				pr_debug("Can't access file %s\n", path);
+				goto failure;
+			}
+			machines__create_kernel_maps(machines, pid);
+		}
+failure:
+		free(namelist);
+	}
+
+	return ret;
+}
+
+void machines__destroy_guest_kernel_maps(struct rb_root *machines)
+{
+	struct rb_node *next = rb_first(machines);
+
+	while (next) {
+		struct machine *pos = rb_entry(next, struct machine, rb_node);
+
+		next = rb_next(&pos->rb_node);
+		rb_erase(&pos->rb_node, machines);
+		machine__delete(pos);
+	}
+}
+
+int machine__load_kallsyms(struct machine *machine, const char *filename,
+			   enum map_type type, symbol_filter_t filter)
+{
+	struct map *map = machine->vmlinux_maps[type];
+	int ret = dso__load_kallsyms(map->dso, filename, map, filter);
+
+	if (ret > 0) {
+		dso__set_loaded(map->dso, type);
+		/*
+		 * Since /proc/kallsyms will have multiple sessions for the
+		 * kernel, with modules between them, fixup the end of all
+		 * sections.
+		 */
+		__map_groups__fixup_end(&machine->kmaps, type);
+	}
+
+	return ret;
+}
+
+int machine__load_vmlinux_path(struct machine *machine, enum map_type type,
+			       symbol_filter_t filter)
+{
+	struct map *map = machine->vmlinux_maps[type];
+	int ret = dso__load_vmlinux_path(map->dso, map, filter);
+
+	if (ret > 0) {
+		dso__set_loaded(map->dso, type);
+		map__reloc_vmlinux(map);
+	}
+
+	return ret;
+}
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
new file mode 100644
index 00000000..325ee36a
--- /dev/null
+++ b/tools/perf/util/symbol.h
@@ -0,0 +1,245 @@
+#ifndef __PERF_SYMBOL
+#define __PERF_SYMBOL 1
+
+#include <linux/types.h>
+#include <stdbool.h>
+#include <stdint.h>
+#include "map.h"
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include <stdio.h>
+
+#ifdef HAVE_CPLUS_DEMANGLE
+extern char *cplus_demangle(const char *, int);
+
+static inline char *bfd_demangle(void __used *v, const char *c, int i)
+{
+	return cplus_demangle(c, i);
+}
+#else
+#ifdef NO_DEMANGLE
+static inline char *bfd_demangle(void __used *v, const char __used *c,
+				 int __used i)
+{
+	return NULL;
+}
+#else
+#include <bfd.h>
+#endif
+#endif
+
+int hex2u64(const char *ptr, u64 *val);
+char *strxfrchar(char *s, char from, char to);
+
+/*
+ * libelf 0.8.x and earlier do not support ELF_C_READ_MMAP;
+ * for newer versions we can use mmap to reduce memory usage:
+ */
+#ifdef LIBELF_NO_MMAP
+# define PERF_ELF_C_READ_MMAP ELF_C_READ
+#else
+# define PERF_ELF_C_READ_MMAP ELF_C_READ_MMAP
+#endif
+
+#ifndef DMGL_PARAMS
+#define DMGL_PARAMS      (1 << 0)       /* Include function args */
+#define DMGL_ANSI        (1 << 1)       /* Include const, volatile, etc */
+#endif
+
+#define BUILD_ID_SIZE 20
+
+/** struct symbol - symtab entry
+ *
+ * @ignore - resolvable but tools ignore it (e.g. idle routines)
+ */
+struct symbol {
+	struct rb_node	rb_node;
+	u64		start;
+	u64		end;
+	u16		namelen;
+	u8		binding;
+	bool		ignore;
+	char		name[0];
+};
+
+void symbol__delete(struct symbol *sym);
+
+struct strlist;
+
+struct symbol_conf {
+	unsigned short	priv_size;
+	bool		try_vmlinux_path,
+			use_modules,
+			sort_by_name,
+			show_nr_samples,
+			use_callchain,
+			exclude_other,
+			show_cpu_utilization,
+			initialized,
+			kptr_restrict;
+	const char	*vmlinux_name,
+			*kallsyms_name,
+			*source_prefix,
+			*field_sep;
+	const char	*default_guest_vmlinux_name,
+			*default_guest_kallsyms,
+			*default_guest_modules;
+	const char	*guestmount;
+	const char	*dso_list_str,
+			*comm_list_str,
+			*sym_list_str,
+			*col_width_list_str;
+       struct strlist	*dso_list,
+			*comm_list,
+			*sym_list;
+	const char	*symfs;
+};
+
+extern struct symbol_conf symbol_conf;
+
+static inline void *symbol__priv(struct symbol *sym)
+{
+	return ((void *)sym) - symbol_conf.priv_size;
+}
+
+struct ref_reloc_sym {
+	const char	*name;
+	u64		addr;
+	u64		unrelocated_addr;
+};
+
+struct map_symbol {
+	struct map    *map;
+	struct symbol *sym;
+	bool	      unfolded;
+	bool	      has_children;
+};
+
+struct addr_location {
+	struct thread *thread;
+	struct map    *map;
+	struct symbol *sym;
+	u64	      addr;
+	char	      level;
+	bool	      filtered;
+	u8	      cpumode;
+	s32	      cpu;
+};
+
+enum dso_kernel_type {
+	DSO_TYPE_USER = 0,
+	DSO_TYPE_KERNEL,
+	DSO_TYPE_GUEST_KERNEL
+};
+
+struct dso {
+	struct list_head node;
+	struct rb_root	 symbols[MAP__NR_TYPES];
+	struct rb_root	 symbol_names[MAP__NR_TYPES];
+	enum dso_kernel_type	kernel;
+	u8		 adjust_symbols:1;
+	u8		 has_build_id:1;
+	u8		 hit:1;
+	u8		 annotate_warned:1;
+	u8		 sname_alloc:1;
+	u8		 lname_alloc:1;
+	unsigned char	 symtab_type;
+	u8		 sorted_by_name;
+	u8		 loaded;
+	u8		 build_id[BUILD_ID_SIZE];
+	const char	 *short_name;
+	char	 	 *long_name;
+	u16		 long_name_len;
+	u16		 short_name_len;
+	char		 name[0];
+};
+
+struct dso *dso__new(const char *name);
+struct dso *dso__new_kernel(const char *name);
+void dso__delete(struct dso *dso);
+
+int dso__name_len(const struct dso *dso);
+
+bool dso__loaded(const struct dso *dso, enum map_type type);
+bool dso__sorted_by_name(const struct dso *dso, enum map_type type);
+
+static inline void dso__set_loaded(struct dso *dso, enum map_type type)
+{
+	dso->loaded |= (1 << type);
+}
+
+void dso__sort_by_name(struct dso *dso, enum map_type type);
+
+struct dso *__dsos__findnew(struct list_head *head, const char *name);
+
+int dso__load(struct dso *dso, struct map *map, symbol_filter_t filter);
+int dso__load_vmlinux(struct dso *dso, struct map *map,
+		      const char *vmlinux, symbol_filter_t filter);
+int dso__load_vmlinux_path(struct dso *dso, struct map *map,
+			   symbol_filter_t filter);
+int dso__load_kallsyms(struct dso *dso, const char *filename, struct map *map,
+		       symbol_filter_t filter);
+int machine__load_kallsyms(struct machine *machine, const char *filename,
+			   enum map_type type, symbol_filter_t filter);
+int machine__load_vmlinux_path(struct machine *machine, enum map_type type,
+			       symbol_filter_t filter);
+
+size_t __dsos__fprintf(struct list_head *head, FILE *fp);
+
+size_t machine__fprintf_dsos_buildid(struct machine *machine,
+				     FILE *fp, bool with_hits);
+size_t machines__fprintf_dsos(struct rb_root *machines, FILE *fp);
+size_t machines__fprintf_dsos_buildid(struct rb_root *machines,
+				      FILE *fp, bool with_hits);
+size_t dso__fprintf_buildid(struct dso *dso, FILE *fp);
+size_t dso__fprintf_symbols_by_name(struct dso *dso,
+				    enum map_type type, FILE *fp);
+size_t dso__fprintf(struct dso *dso, enum map_type type, FILE *fp);
+
+enum symtab_type {
+	SYMTAB__KALLSYMS = 0,
+	SYMTAB__GUEST_KALLSYMS,
+	SYMTAB__JAVA_JIT,
+	SYMTAB__BUILD_ID_CACHE,
+	SYMTAB__FEDORA_DEBUGINFO,
+	SYMTAB__UBUNTU_DEBUGINFO,
+	SYMTAB__BUILDID_DEBUGINFO,
+	SYMTAB__SYSTEM_PATH_DSO,
+	SYMTAB__GUEST_KMODULE,
+	SYMTAB__SYSTEM_PATH_KMODULE,
+	SYMTAB__NOT_FOUND,
+};
+
+char dso__symtab_origin(const struct dso *dso);
+void dso__set_long_name(struct dso *dso, char *name);
+void dso__set_build_id(struct dso *dso, void *build_id);
+void dso__read_running_kernel_build_id(struct dso *dso,
+				       struct machine *machine);
+struct symbol *dso__find_symbol(struct dso *dso, enum map_type type,
+				u64 addr);
+struct symbol *dso__find_symbol_by_name(struct dso *dso, enum map_type type,
+					const char *name);
+
+int filename__read_build_id(const char *filename, void *bf, size_t size);
+int sysfs__read_build_id(const char *filename, void *bf, size_t size);
+bool __dsos__read_build_ids(struct list_head *head, bool with_hits);
+int build_id__sprintf(const u8 *build_id, int len, char *bf);
+int kallsyms__parse(const char *filename, void *arg,
+		    int (*process_symbol)(void *arg, const char *name,
+					  char type, u64 start, u64 end));
+
+void machine__destroy_kernel_maps(struct machine *machine);
+int __machine__create_kernel_maps(struct machine *machine, struct dso *kernel);
+int machine__create_kernel_maps(struct machine *machine);
+
+int machines__create_kernel_maps(struct rb_root *machines, pid_t pid);
+int machines__create_guest_kernel_maps(struct rb_root *machines);
+void machines__destroy_guest_kernel_maps(struct rb_root *machines);
+
+int symbol__init(void);
+void symbol__exit(void);
+bool symbol_type__is_a(char symbol_type, enum map_type map_type);
+
+size_t machine__fprintf_vmlinux_path(struct machine *machine, FILE *fp);
+
+#endif /* __PERF_SYMBOL */
diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c
new file mode 100644
index 00000000..d5d3b222
--- /dev/null
+++ b/tools/perf/util/thread.c
@@ -0,0 +1,140 @@
+#include "../perf.h"
+#include <stdlib.h>
+#include <stdio.h>
+#include <string.h>
+#include "session.h"
+#include "thread.h"
+#include "util.h"
+#include "debug.h"
+
+static struct thread *thread__new(pid_t pid)
+{
+	struct thread *self = zalloc(sizeof(*self));
+
+	if (self != NULL) {
+		map_groups__init(&self->mg);
+		self->pid = pid;
+		self->comm = malloc(32);
+		if (self->comm)
+			snprintf(self->comm, 32, ":%d", self->pid);
+	}
+
+	return self;
+}
+
+void thread__delete(struct thread *self)
+{
+	map_groups__exit(&self->mg);
+	free(self->comm);
+	free(self);
+}
+
+int thread__set_comm(struct thread *self, const char *comm)
+{
+	int err;
+
+	if (self->comm)
+		free(self->comm);
+	self->comm = strdup(comm);
+	err = self->comm == NULL ? -ENOMEM : 0;
+	if (!err) {
+		self->comm_set = true;
+		map_groups__flush(&self->mg);
+	}
+	return err;
+}
+
+int thread__comm_len(struct thread *self)
+{
+	if (!self->comm_len) {
+		if (!self->comm)
+			return 0;
+		self->comm_len = strlen(self->comm);
+	}
+
+	return self->comm_len;
+}
+
+static size_t thread__fprintf(struct thread *self, FILE *fp)
+{
+	return fprintf(fp, "Thread %d %s\n", self->pid, self->comm) +
+	       map_groups__fprintf(&self->mg, verbose, fp);
+}
+
+struct thread *perf_session__findnew(struct perf_session *self, pid_t pid)
+{
+	struct rb_node **p = &self->threads.rb_node;
+	struct rb_node *parent = NULL;
+	struct thread *th;
+
+	/*
+	 * Font-end cache - PID lookups come in blocks,
+	 * so most of the time we dont have to look up
+	 * the full rbtree:
+	 */
+	if (self->last_match && self->last_match->pid == pid)
+		return self->last_match;
+
+	while (*p != NULL) {
+		parent = *p;
+		th = rb_entry(parent, struct thread, rb_node);
+
+		if (th->pid == pid) {
+			self->last_match = th;
+			return th;
+		}
+
+		if (pid < th->pid)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	th = thread__new(pid);
+	if (th != NULL) {
+		rb_link_node(&th->rb_node, parent, p);
+		rb_insert_color(&th->rb_node, &self->threads);
+		self->last_match = th;
+	}
+
+	return th;
+}
+
+void thread__insert_map(struct thread *self, struct map *map)
+{
+	map_groups__fixup_overlappings(&self->mg, map, verbose, stderr);
+	map_groups__insert(&self->mg, map);
+}
+
+int thread__fork(struct thread *self, struct thread *parent)
+{
+	int i;
+
+	if (parent->comm_set) {
+		if (self->comm)
+			free(self->comm);
+		self->comm = strdup(parent->comm);
+		if (!self->comm)
+			return -ENOMEM;
+		self->comm_set = true;
+	}
+
+	for (i = 0; i < MAP__NR_TYPES; ++i)
+		if (map_groups__clone(&self->mg, &parent->mg, i) < 0)
+			return -ENOMEM;
+	return 0;
+}
+
+size_t perf_session__fprintf(struct perf_session *self, FILE *fp)
+{
+	size_t ret = 0;
+	struct rb_node *nd;
+
+	for (nd = rb_first(&self->threads); nd; nd = rb_next(nd)) {
+		struct thread *pos = rb_entry(nd, struct thread, rb_node);
+
+		ret += thread__fprintf(pos, fp);
+	}
+
+	return ret;
+}
diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h
new file mode 100644
index 00000000..e5f2401c
--- /dev/null
+++ b/tools/perf/util/thread.h
@@ -0,0 +1,48 @@
+#ifndef __PERF_THREAD_H
+#define __PERF_THREAD_H
+
+#include <linux/rbtree.h>
+#include <unistd.h>
+#include "symbol.h"
+
+struct thread {
+	union {
+		struct rb_node	 rb_node;
+		struct list_head node;
+	};
+	struct map_groups	mg;
+	pid_t			pid;
+	char			shortname[3];
+	bool			comm_set;
+	char			*comm;
+	int			comm_len;
+};
+
+struct perf_session;
+
+void thread__delete(struct thread *self);
+
+int thread__set_comm(struct thread *self, const char *comm);
+int thread__comm_len(struct thread *self);
+struct thread *perf_session__findnew(struct perf_session *self, pid_t pid);
+void thread__insert_map(struct thread *self, struct map *map);
+int thread__fork(struct thread *self, struct thread *parent);
+size_t perf_session__fprintf(struct perf_session *self, FILE *fp);
+
+static inline struct map *thread__find_map(struct thread *self,
+					   enum map_type type, u64 addr)
+{
+	return self ? map_groups__find(&self->mg, type, addr) : NULL;
+}
+
+void thread__find_addr_map(struct thread *self,
+			   struct perf_session *session, u8 cpumode,
+			   enum map_type type, pid_t pid, u64 addr,
+			   struct addr_location *al);
+
+void thread__find_addr_location(struct thread *self,
+				struct perf_session *session, u8 cpumode,
+				enum map_type type, pid_t pid, u64 addr,
+				struct addr_location *al,
+				symbol_filter_t filter);
+#endif	/* __PERF_THREAD_H */
diff --git a/tools/perf/util/thread_map.c b/tools/perf/util/thread_map.c
new file mode 100644
index 00000000..a5df131b
--- /dev/null
+++ b/tools/perf/util/thread_map.c
@@ -0,0 +1,64 @@
+#include <dirent.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include "thread_map.h"
+
+/* Skip "." and ".." directories */
+static int filter(const struct dirent *dir)
+{
+	if (dir->d_name[0] == '.')
+		return 0;
+	else
+		return 1;
+}
+
+struct thread_map *thread_map__new_by_pid(pid_t pid)
+{
+	struct thread_map *threads;
+	char name[256];
+	int items;
+	struct dirent **namelist = NULL;
+	int i;
+
+	sprintf(name, "/proc/%d/task", pid);
+	items = scandir(name, &namelist, filter, NULL);
+	if (items <= 0)
+                return NULL;
+
+	threads = malloc(sizeof(*threads) + sizeof(pid_t) * items);
+	if (threads != NULL) {
+		for (i = 0; i < items; i++)
+			threads->map[i] = atoi(namelist[i]->d_name);
+		threads->nr = items;
+	}
+
+	for (i=0; i<items; i++)
+		free(namelist[i]);
+	free(namelist);
+
+	return threads;
+}
+
+struct thread_map *thread_map__new_by_tid(pid_t tid)
+{
+	struct thread_map *threads = malloc(sizeof(*threads) + sizeof(pid_t));
+
+	if (threads != NULL) {
+		threads->map[0] = tid;
+		threads->nr	= 1;
+	}
+
+	return threads;
+}
+
+struct thread_map *thread_map__new(pid_t pid, pid_t tid)
+{
+	if (pid != -1)
+		return thread_map__new_by_pid(pid);
+	return thread_map__new_by_tid(tid);
+}
+
+void thread_map__delete(struct thread_map *threads)
+{
+	free(threads);
+}
diff --git a/tools/perf/util/thread_map.h b/tools/perf/util/thread_map.h
new file mode 100644
index 00000000..3cb90731
--- /dev/null
+++ b/tools/perf/util/thread_map.h
@@ -0,0 +1,15 @@
+#ifndef __PERF_THREAD_MAP_H
+#define __PERF_THREAD_MAP_H
+
+#include <sys/types.h>
+
+struct thread_map {
+	int nr;
+	int map[];
+};
+
+struct thread_map *thread_map__new_by_pid(pid_t pid);
+struct thread_map *thread_map__new_by_tid(pid_t tid);
+struct thread_map *thread_map__new(pid_t pid, pid_t tid);
+void thread_map__delete(struct thread_map *threads);
+#endif	/* __PERF_THREAD_MAP_H */
diff --git a/tools/perf/util/top.c b/tools/perf/util/top.c
new file mode 100644
index 00000000..a11f6073
--- /dev/null
+++ b/tools/perf/util/top.c
@@ -0,0 +1,238 @@
+/*
+ * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Refactored from builtin-top.c, see that files for further copyright notes.
+ *
+ * Released under the GPL v2. (and only v2, not any later version)
+ */
+
+#include "cpumap.h"
+#include "event.h"
+#include "evlist.h"
+#include "evsel.h"
+#include "parse-events.h"
+#include "symbol.h"
+#include "top.h"
+#include <inttypes.h>
+
+/*
+ * Ordering weight: count-1 * count-2 * ... / count-n
+ */
+static double sym_weight(const struct sym_entry *sym, struct perf_top *top)
+{
+	double weight = sym->snap_count;
+	int counter;
+
+	if (!top->display_weighted)
+		return weight;
+
+	for (counter = 1; counter < top->evlist->nr_entries - 1; counter++)
+		weight *= sym->count[counter];
+
+	weight /= (sym->count[counter] + 1);
+
+	return weight;
+}
+
+static void perf_top__remove_active_sym(struct perf_top *top, struct sym_entry *syme)
+{
+	pthread_mutex_lock(&top->active_symbols_lock);
+	list_del_init(&syme->node);
+	pthread_mutex_unlock(&top->active_symbols_lock);
+}
+
+static void rb_insert_active_sym(struct rb_root *tree, struct sym_entry *se)
+{
+	struct rb_node **p = &tree->rb_node;
+	struct rb_node *parent = NULL;
+	struct sym_entry *iter;
+
+	while (*p != NULL) {
+		parent = *p;
+		iter = rb_entry(parent, struct sym_entry, rb_node);
+
+		if (se->weight > iter->weight)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	rb_link_node(&se->rb_node, parent, p);
+	rb_insert_color(&se->rb_node, tree);
+}
+
+#define SNPRINTF(buf, size, fmt, args...) \
+({ \
+	size_t r = snprintf(buf, size, fmt, ## args); \
+	r > size ?  size : r; \
+})
+
+size_t perf_top__header_snprintf(struct perf_top *top, char *bf, size_t size)
+{
+	struct perf_evsel *counter;
+	float samples_per_sec = top->samples / top->delay_secs;
+	float ksamples_per_sec = top->kernel_samples / top->delay_secs;
+	float esamples_percent = (100.0 * top->exact_samples) / top->samples;
+	size_t ret = 0;
+
+	if (!perf_guest) {
+		ret = SNPRINTF(bf, size,
+			       "   PerfTop:%8.0f irqs/sec  kernel:%4.1f%%"
+			       "  exact: %4.1f%% [", samples_per_sec,
+			       100.0 - (100.0 * ((samples_per_sec - ksamples_per_sec) /
+					samples_per_sec)),
+				esamples_percent);
+	} else {
+		float us_samples_per_sec = top->us_samples / top->delay_secs;
+		float guest_kernel_samples_per_sec = top->guest_kernel_samples / top->delay_secs;
+		float guest_us_samples_per_sec = top->guest_us_samples / top->delay_secs;
+
+		ret = SNPRINTF(bf, size,
+			       "   PerfTop:%8.0f irqs/sec  kernel:%4.1f%% us:%4.1f%%"
+			       " guest kernel:%4.1f%% guest us:%4.1f%%"
+			       " exact: %4.1f%% [", samples_per_sec,
+			       100.0 - (100.0 * ((samples_per_sec - ksamples_per_sec) /
+						 samples_per_sec)),
+			       100.0 - (100.0 * ((samples_per_sec - us_samples_per_sec) /
+						 samples_per_sec)),
+			       100.0 - (100.0 * ((samples_per_sec -
+						  guest_kernel_samples_per_sec) /
+						 samples_per_sec)),
+			       100.0 - (100.0 * ((samples_per_sec -
+						  guest_us_samples_per_sec) /
+						 samples_per_sec)),
+			       esamples_percent);
+	}
+
+	if (top->evlist->nr_entries == 1 || !top->display_weighted) {
+		struct perf_evsel *first;
+		first = list_entry(top->evlist->entries.next, struct perf_evsel, node);
+		ret += SNPRINTF(bf + ret, size - ret, "%" PRIu64 "%s ",
+				(uint64_t)first->attr.sample_period,
+				top->freq ? "Hz" : "");
+	}
+
+	if (!top->display_weighted) {
+		ret += SNPRINTF(bf + ret, size - ret, "%s",
+				event_name(top->sym_evsel));
+	} else {
+		/*
+		 * Don't let events eat all the space. Leaving 30 bytes
+		 * for the rest should be enough.
+		 */
+		size_t last_pos = size - 30;
+
+		list_for_each_entry(counter, &top->evlist->entries, node) {
+			ret += SNPRINTF(bf + ret, size - ret, "%s%s",
+					counter->idx ? "/" : "",
+					event_name(counter));
+			if (ret > last_pos) {
+				sprintf(bf + last_pos - 3, "..");
+				ret = last_pos - 1;
+				break;
+			}
+		}
+	}
+
+	ret += SNPRINTF(bf + ret, size - ret, "], ");
+
+	if (top->target_pid != -1)
+		ret += SNPRINTF(bf + ret, size - ret, " (target_pid: %d",
+				top->target_pid);
+	else if (top->target_tid != -1)
+		ret += SNPRINTF(bf + ret, size - ret, " (target_tid: %d",
+				top->target_tid);
+	else
+		ret += SNPRINTF(bf + ret, size - ret, " (all");
+
+	if (top->cpu_list)
+		ret += SNPRINTF(bf + ret, size - ret, ", CPU%s: %s)",
+				top->evlist->cpus->nr > 1 ? "s" : "", top->cpu_list);
+	else {
+		if (top->target_tid != -1)
+			ret += SNPRINTF(bf + ret, size - ret, ")");
+		else
+			ret += SNPRINTF(bf + ret, size - ret, ", %d CPU%s)",
+					top->evlist->cpus->nr,
+					top->evlist->cpus->nr > 1 ? "s" : "");
+	}
+
+	return ret;
+}
+
+void perf_top__reset_sample_counters(struct perf_top *top)
+{
+	top->samples = top->us_samples = top->kernel_samples =
+	top->exact_samples = top->guest_kernel_samples =
+	top->guest_us_samples = 0;
+}
+
+float perf_top__decay_samples(struct perf_top *top, struct rb_root *root)
+{
+	struct sym_entry *syme, *n;
+	float sum_ksamples = 0.0;
+	int snap = !top->display_weighted ? top->sym_evsel->idx : 0, j;
+
+	/* Sort the active symbols */
+	pthread_mutex_lock(&top->active_symbols_lock);
+	syme = list_entry(top->active_symbols.next, struct sym_entry, node);
+	pthread_mutex_unlock(&top->active_symbols_lock);
+
+	top->rb_entries = 0;
+	list_for_each_entry_safe_from(syme, n, &top->active_symbols, node) {
+		syme->snap_count = syme->count[snap];
+		if (syme->snap_count != 0) {
+
+			if ((top->hide_user_symbols &&
+			     syme->map->dso->kernel == DSO_TYPE_USER) ||
+			    (top->hide_kernel_symbols &&
+			     syme->map->dso->kernel == DSO_TYPE_KERNEL)) {
+				perf_top__remove_active_sym(top, syme);
+				continue;
+			}
+			syme->weight = sym_weight(syme, top);
+
+			if ((int)syme->snap_count >= top->count_filter) {
+				rb_insert_active_sym(root, syme);
+				++top->rb_entries;
+			}
+			sum_ksamples += syme->snap_count;
+
+			for (j = 0; j < top->evlist->nr_entries; j++)
+				syme->count[j] = top->zero ? 0 : syme->count[j] * 7 / 8;
+		} else
+			perf_top__remove_active_sym(top, syme);
+	}
+
+	return sum_ksamples;
+}
+
+/*
+ * Find the longest symbol name that will be displayed
+ */
+void perf_top__find_widths(struct perf_top *top, struct rb_root *root,
+			   int *dso_width, int *dso_short_width, int *sym_width)
+{
+	struct rb_node *nd;
+	int printed = 0;
+
+	*sym_width = *dso_width = *dso_short_width = 0;
+
+	for (nd = rb_first(root); nd; nd = rb_next(nd)) {
+		struct sym_entry *syme = rb_entry(nd, struct sym_entry, rb_node);
+		struct symbol *sym = sym_entry__symbol(syme);
+
+		if (++printed > top->print_entries ||
+		    (int)syme->snap_count < top->count_filter)
+			continue;
+
+		if (syme->map->dso->long_name_len > *dso_width)
+			*dso_width = syme->map->dso->long_name_len;
+
+		if (syme->map->dso->short_name_len > *dso_short_width)
+			*dso_short_width = syme->map->dso->short_name_len;
+
+		if (sym->namelen > *sym_width)
+			*sym_width = sym->namelen;
+	}
+}
diff --git a/tools/perf/util/top.h b/tools/perf/util/top.h
new file mode 100644
index 00000000..bfbf95bc
--- /dev/null
+++ b/tools/perf/util/top.h
@@ -0,0 +1,64 @@
+#ifndef __PERF_TOP_H
+#define __PERF_TOP_H 1
+
+#include "types.h"
+#include "../perf.h"
+#include <stddef.h>
+#include <pthread.h>
+#include <linux/list.h>
+#include <linux/rbtree.h>
+
+struct perf_evlist;
+struct perf_evsel;
+
+struct sym_entry {
+	struct rb_node		rb_node;
+	struct list_head	node;
+	unsigned long		snap_count;
+	double			weight;
+	struct map		*map;
+	unsigned long		count[0];
+};
+
+static inline struct symbol *sym_entry__symbol(struct sym_entry *self)
+{
+       return ((void *)self) + symbol_conf.priv_size;
+}
+
+struct perf_top {
+	struct perf_evlist *evlist;
+	/*
+	 * Symbols will be added here in perf_event__process_sample and will
+	 * get out after decayed.
+	 */
+	struct list_head   active_symbols;
+	pthread_mutex_t	   active_symbols_lock;
+	pthread_cond_t	   active_symbols_cond;
+	u64		   samples;
+	u64		   kernel_samples, us_samples;
+	u64		   exact_samples;
+	u64		   guest_us_samples, guest_kernel_samples;
+	int		   print_entries, count_filter, delay_secs;
+	int		   display_weighted, freq, rb_entries;
+	pid_t		   target_pid, target_tid;
+	bool		   hide_kernel_symbols, hide_user_symbols, zero;
+	const char	   *cpu_list;
+	struct sym_entry   *sym_filter_entry;
+	struct perf_evsel  *sym_evsel;
+};
+
+size_t perf_top__header_snprintf(struct perf_top *top, char *bf, size_t size);
+void perf_top__reset_sample_counters(struct perf_top *top);
+float perf_top__decay_samples(struct perf_top *top, struct rb_root *root);
+void perf_top__find_widths(struct perf_top *top, struct rb_root *root,
+			   int *dso_width, int *dso_short_width, int *sym_width);
+
+#ifdef NO_NEWT_SUPPORT
+static inline int perf_top__tui_browser(struct perf_top *top __used)
+{
+	return 0;
+}
+#else
+int perf_top__tui_browser(struct perf_top *top);
+#endif
+#endif /* __PERF_TOP_H */
diff --git a/tools/perf/util/trace-event-info.c b/tools/perf/util/trace-event-info.c
new file mode 100644
index 00000000..35729f4c
--- /dev/null
+++ b/tools/perf/util/trace-event-info.c
@@ -0,0 +1,565 @@
+/*
+ * Copyright (C) 2008,2009, Steven Rostedt <srostedt@redhat.com>
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License (not later!)
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ */
+#define _GNU_SOURCE
+#include <dirent.h>
+#include <mntent.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdarg.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/wait.h>
+#include <pthread.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include <ctype.h>
+#include <errno.h>
+#include <stdbool.h>
+#include <linux/list.h>
+#include <linux/kernel.h>
+
+#include "../perf.h"
+#include "trace-event.h"
+#include "debugfs.h"
+#include "evsel.h"
+
+#define VERSION "0.5"
+
+#define _STR(x) #x
+#define STR(x) _STR(x)
+#define MAX_PATH 256
+
+#define TRACE_CTRL	"tracing_on"
+#define TRACE		"trace"
+#define AVAILABLE	"available_tracers"
+#define CURRENT		"current_tracer"
+#define ITER_CTRL	"trace_options"
+#define MAX_LATENCY	"tracing_max_latency"
+
+unsigned int page_size;
+
+static const char *output_file = "trace.info";
+static int output_fd;
+
+struct event_list {
+	struct event_list *next;
+	const char *event;
+};
+
+struct events {
+	struct events *sibling;
+	struct events *children;
+	struct events *next;
+	char *name;
+};
+
+
+
+static void die(const char *fmt, ...)
+{
+	va_list ap;
+	int ret = errno;
+
+	if (errno)
+		perror("trace-cmd");
+	else
+		ret = -1;
+
+	va_start(ap, fmt);
+	fprintf(stderr, "  ");
+	vfprintf(stderr, fmt, ap);
+	va_end(ap);
+
+	fprintf(stderr, "\n");
+	exit(ret);
+}
+
+void *malloc_or_die(unsigned int size)
+{
+	void *data;
+
+	data = malloc(size);
+	if (!data)
+		die("malloc");
+	return data;
+}
+
+static const char *find_debugfs(void)
+{
+	const char *path = debugfs_mount(NULL);
+
+	if (!path)
+		die("Your kernel not support debugfs filesystem");
+
+	return path;
+}
+
+/*
+ * Finds the path to the debugfs/tracing
+ * Allocates the string and stores it.
+ */
+static const char *find_tracing_dir(void)
+{
+	static char *tracing;
+	static int tracing_found;
+	const char *debugfs;
+
+	if (tracing_found)
+		return tracing;
+
+	debugfs = find_debugfs();
+
+	tracing = malloc_or_die(strlen(debugfs) + 9);
+
+	sprintf(tracing, "%s/tracing", debugfs);
+
+	tracing_found = 1;
+	return tracing;
+}
+
+static char *get_tracing_file(const char *name)
+{
+	const char *tracing;
+	char *file;
+
+	tracing = find_tracing_dir();
+	if (!tracing)
+		return NULL;
+
+	file = malloc_or_die(strlen(tracing) + strlen(name) + 2);
+
+	sprintf(file, "%s/%s", tracing, name);
+	return file;
+}
+
+static void put_tracing_file(char *file)
+{
+	free(file);
+}
+
+static ssize_t calc_data_size;
+
+static ssize_t write_or_die(const void *buf, size_t len)
+{
+	int ret;
+
+	if (calc_data_size) {
+		calc_data_size += len;
+		return len;
+	}
+
+	ret = write(output_fd, buf, len);
+	if (ret < 0)
+		die("writing to '%s'", output_file);
+
+	return ret;
+}
+
+int bigendian(void)
+{
+	unsigned char str[] = { 0x1, 0x2, 0x3, 0x4, 0x0, 0x0, 0x0, 0x0};
+	unsigned int *ptr;
+
+	ptr = (unsigned int *)(void *)str;
+	return *ptr == 0x01020304;
+}
+
+static unsigned long long copy_file_fd(int fd)
+{
+	unsigned long long size = 0;
+	char buf[BUFSIZ];
+	int r;
+
+	do {
+		r = read(fd, buf, BUFSIZ);
+		if (r > 0) {
+			size += r;
+			write_or_die(buf, r);
+		}
+	} while (r > 0);
+
+	return size;
+}
+
+static unsigned long long copy_file(const char *file)
+{
+	unsigned long long size = 0;
+	int fd;
+
+	fd = open(file, O_RDONLY);
+	if (fd < 0)
+		die("Can't read '%s'", file);
+	size = copy_file_fd(fd);
+	close(fd);
+
+	return size;
+}
+
+static unsigned long get_size_fd(int fd)
+{
+	unsigned long long size = 0;
+	char buf[BUFSIZ];
+	int r;
+
+	do {
+		r = read(fd, buf, BUFSIZ);
+		if (r > 0)
+			size += r;
+	} while (r > 0);
+
+	lseek(fd, 0, SEEK_SET);
+
+	return size;
+}
+
+static unsigned long get_size(const char *file)
+{
+	unsigned long long size = 0;
+	int fd;
+
+	fd = open(file, O_RDONLY);
+	if (fd < 0)
+		die("Can't read '%s'", file);
+	size = get_size_fd(fd);
+	close(fd);
+
+	return size;
+}
+
+static void read_header_files(void)
+{
+	unsigned long long size, check_size;
+	char *path;
+	int fd;
+
+	path = get_tracing_file("events/header_page");
+	fd = open(path, O_RDONLY);
+	if (fd < 0)
+		die("can't read '%s'", path);
+
+	/* unfortunately, you can not stat debugfs files for size */
+	size = get_size_fd(fd);
+
+	write_or_die("header_page", 12);
+	write_or_die(&size, 8);
+	check_size = copy_file_fd(fd);
+	close(fd);
+
+	if (size != check_size)
+		die("wrong size for '%s' size=%lld read=%lld",
+		    path, size, check_size);
+	put_tracing_file(path);
+
+	path = get_tracing_file("events/header_event");
+	fd = open(path, O_RDONLY);
+	if (fd < 0)
+		die("can't read '%s'", path);
+
+	size = get_size_fd(fd);
+
+	write_or_die("header_event", 13);
+	write_or_die(&size, 8);
+	check_size = copy_file_fd(fd);
+	if (size != check_size)
+		die("wrong size for '%s'", path);
+	put_tracing_file(path);
+	close(fd);
+}
+
+static bool name_in_tp_list(char *sys, struct tracepoint_path *tps)
+{
+	while (tps) {
+		if (!strcmp(sys, tps->name))
+			return true;
+		tps = tps->next;
+	}
+
+	return false;
+}
+
+static void copy_event_system(const char *sys, struct tracepoint_path *tps)
+{
+	unsigned long long size, check_size;
+	struct dirent *dent;
+	struct stat st;
+	char *format;
+	DIR *dir;
+	int count = 0;
+	int ret;
+
+	dir = opendir(sys);
+	if (!dir)
+		die("can't read directory '%s'", sys);
+
+	while ((dent = readdir(dir))) {
+		if (dent->d_type != DT_DIR ||
+		    strcmp(dent->d_name, ".") == 0 ||
+		    strcmp(dent->d_name, "..") == 0 ||
+		    !name_in_tp_list(dent->d_name, tps))
+			continue;
+		format = malloc_or_die(strlen(sys) + strlen(dent->d_name) + 10);
+		sprintf(format, "%s/%s/format", sys, dent->d_name);
+		ret = stat(format, &st);
+		free(format);
+		if (ret < 0)
+			continue;
+		count++;
+	}
+
+	write_or_die(&count, 4);
+
+	rewinddir(dir);
+	while ((dent = readdir(dir))) {
+		if (dent->d_type != DT_DIR ||
+		    strcmp(dent->d_name, ".") == 0 ||
+		    strcmp(dent->d_name, "..") == 0 ||
+		    !name_in_tp_list(dent->d_name, tps))
+			continue;
+		format = malloc_or_die(strlen(sys) + strlen(dent->d_name) + 10);
+		sprintf(format, "%s/%s/format", sys, dent->d_name);
+		ret = stat(format, &st);
+
+		if (ret >= 0) {
+			/* unfortunately, you can not stat debugfs files for size */
+			size = get_size(format);
+			write_or_die(&size, 8);
+			check_size = copy_file(format);
+			if (size != check_size)
+				die("error in size of file '%s'", format);
+		}
+
+		free(format);
+	}
+	closedir(dir);
+}
+
+static void read_ftrace_files(struct tracepoint_path *tps)
+{
+	char *path;
+
+	path = get_tracing_file("events/ftrace");
+
+	copy_event_system(path, tps);
+
+	put_tracing_file(path);
+}
+
+static bool system_in_tp_list(char *sys, struct tracepoint_path *tps)
+{
+	while (tps) {
+		if (!strcmp(sys, tps->system))
+			return true;
+		tps = tps->next;
+	}
+
+	return false;
+}
+
+static void read_event_files(struct tracepoint_path *tps)
+{
+	struct dirent *dent;
+	struct stat st;
+	char *path;
+	char *sys;
+	DIR *dir;
+	int count = 0;
+	int ret;
+
+	path = get_tracing_file("events");
+
+	dir = opendir(path);
+	if (!dir)
+		die("can't read directory '%s'", path);
+
+	while ((dent = readdir(dir))) {
+		if (dent->d_type != DT_DIR ||
+		    strcmp(dent->d_name, ".") == 0 ||
+		    strcmp(dent->d_name, "..") == 0 ||
+		    strcmp(dent->d_name, "ftrace") == 0 ||
+		    !system_in_tp_list(dent->d_name, tps))
+			continue;
+		count++;
+	}
+
+	write_or_die(&count, 4);
+
+	rewinddir(dir);
+	while ((dent = readdir(dir))) {
+		if (dent->d_type != DT_DIR ||
+		    strcmp(dent->d_name, ".") == 0 ||
+		    strcmp(dent->d_name, "..") == 0 ||
+		    strcmp(dent->d_name, "ftrace") == 0 ||
+		    !system_in_tp_list(dent->d_name, tps))
+			continue;
+		sys = malloc_or_die(strlen(path) + strlen(dent->d_name) + 2);
+		sprintf(sys, "%s/%s", path, dent->d_name);
+		ret = stat(sys, &st);
+		if (ret >= 0) {
+			write_or_die(dent->d_name, strlen(dent->d_name) + 1);
+			copy_event_system(sys, tps);
+		}
+		free(sys);
+	}
+
+	closedir(dir);
+	put_tracing_file(path);
+}
+
+static void read_proc_kallsyms(void)
+{
+	unsigned int size, check_size;
+	const char *path = "/proc/kallsyms";
+	struct stat st;
+	int ret;
+
+	ret = stat(path, &st);
+	if (ret < 0) {
+		/* not found */
+		size = 0;
+		write_or_die(&size, 4);
+		return;
+	}
+	size = get_size(path);
+	write_or_die(&size, 4);
+	check_size = copy_file(path);
+	if (size != check_size)
+		die("error in size of file '%s'", path);
+
+}
+
+static void read_ftrace_printk(void)
+{
+	unsigned int size, check_size;
+	char *path;
+	struct stat st;
+	int ret;
+
+	path = get_tracing_file("printk_formats");
+	ret = stat(path, &st);
+	if (ret < 0) {
+		/* not found */
+		size = 0;
+		write_or_die(&size, 4);
+		goto out;
+	}
+	size = get_size(path);
+	write_or_die(&size, 4);
+	check_size = copy_file(path);
+	if (size != check_size)
+		die("error in size of file '%s'", path);
+out:
+	put_tracing_file(path);
+}
+
+static struct tracepoint_path *
+get_tracepoints_path(struct list_head *pattrs)
+{
+	struct tracepoint_path path, *ppath = &path;
+	struct perf_evsel *pos;
+	int nr_tracepoints = 0;
+
+	list_for_each_entry(pos, pattrs, node) {
+		if (pos->attr.type != PERF_TYPE_TRACEPOINT)
+			continue;
+		++nr_tracepoints;
+		ppath->next = tracepoint_id_to_path(pos->attr.config);
+		if (!ppath->next)
+			die("%s\n", "No memory to alloc tracepoints list");
+		ppath = ppath->next;
+	}
+
+	return nr_tracepoints > 0 ? path.next : NULL;
+}
+
+bool have_tracepoints(struct list_head *pattrs)
+{
+	struct perf_evsel *pos;
+
+	list_for_each_entry(pos, pattrs, node)
+		if (pos->attr.type == PERF_TYPE_TRACEPOINT)
+			return true;
+
+	return false;
+}
+
+int read_tracing_data(int fd, struct list_head *pattrs)
+{
+	char buf[BUFSIZ];
+	struct tracepoint_path *tps = get_tracepoints_path(pattrs);
+
+	/*
+	 * What? No tracepoints? No sense writing anything here, bail out.
+	 */
+	if (tps == NULL)
+		return -1;
+
+	output_fd = fd;
+
+	buf[0] = 23;
+	buf[1] = 8;
+	buf[2] = 68;
+	memcpy(buf + 3, "tracing", 7);
+
+	write_or_die(buf, 10);
+
+	write_or_die(VERSION, strlen(VERSION) + 1);
+
+	/* save endian */
+	if (bigendian())
+		buf[0] = 1;
+	else
+		buf[0] = 0;
+
+	write_or_die(buf, 1);
+
+	/* save size of long */
+	buf[0] = sizeof(long);
+	write_or_die(buf, 1);
+
+	/* save page_size */
+	page_size = sysconf(_SC_PAGESIZE);
+	write_or_die(&page_size, 4);
+
+	read_header_files();
+	read_ftrace_files(tps);
+	read_event_files(tps);
+	read_proc_kallsyms();
+	read_ftrace_printk();
+
+	return 0;
+}
+
+ssize_t read_tracing_data_size(int fd, struct list_head *pattrs)
+{
+	ssize_t size;
+	int err = 0;
+
+	calc_data_size = 1;
+	err = read_tracing_data(fd, pattrs);
+	size = calc_data_size - 1;
+	calc_data_size = 0;
+
+	if (err < 0)
+		return err;
+
+	return size;
+}
diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c
new file mode 100644
index 00000000..bf54c488
--- /dev/null
+++ b/tools/perf/util/trace-event-parse.c
@@ -0,0 +1,3147 @@
+/*
+ * Copyright (C) 2009, Steven Rostedt <srostedt@redhat.com>
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License (not later!)
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ *
+ *  The parts for function graph printing was taken and modified from the
+ *  Linux Kernel that were written by Frederic Weisbecker.
+ */
+#define _GNU_SOURCE
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+#include <errno.h>
+
+#undef _GNU_SOURCE
+#include "../perf.h"
+#include "util.h"
+#include "trace-event.h"
+
+int header_page_ts_offset;
+int header_page_ts_size;
+int header_page_size_offset;
+int header_page_size_size;
+int header_page_overwrite_offset;
+int header_page_overwrite_size;
+int header_page_data_offset;
+int header_page_data_size;
+
+bool latency_format;
+
+static char *input_buf;
+static unsigned long long input_buf_ptr;
+static unsigned long long input_buf_siz;
+
+static int cpus;
+static int long_size;
+static int is_flag_field;
+static int is_symbolic_field;
+
+static struct format_field *
+find_any_field(struct event *event, const char *name);
+
+static void init_input_buf(char *buf, unsigned long long size)
+{
+	input_buf = buf;
+	input_buf_siz = size;
+	input_buf_ptr = 0;
+}
+
+struct cmdline {
+	char *comm;
+	int pid;
+};
+
+static struct cmdline *cmdlines;
+static int cmdline_count;
+
+static int cmdline_cmp(const void *a, const void *b)
+{
+	const struct cmdline *ca = a;
+	const struct cmdline *cb = b;
+
+	if (ca->pid < cb->pid)
+		return -1;
+	if (ca->pid > cb->pid)
+		return 1;
+
+	return 0;
+}
+
+void parse_cmdlines(char *file, int size __unused)
+{
+	struct cmdline_list {
+		struct cmdline_list	*next;
+		char			*comm;
+		int			pid;
+	} *list = NULL, *item;
+	char *line;
+	char *next = NULL;
+	int i;
+
+	line = strtok_r(file, "\n", &next);
+	while (line) {
+		item = malloc_or_die(sizeof(*item));
+		sscanf(line, "%d %as", &item->pid,
+		       (float *)(void *)&item->comm); /* workaround gcc warning */
+		item->next = list;
+		list = item;
+		line = strtok_r(NULL, "\n", &next);
+		cmdline_count++;
+	}
+
+	cmdlines = malloc_or_die(sizeof(*cmdlines) * cmdline_count);
+
+	i = 0;
+	while (list) {
+		cmdlines[i].pid = list->pid;
+		cmdlines[i].comm = list->comm;
+		i++;
+		item = list;
+		list = list->next;
+		free(item);
+	}
+
+	qsort(cmdlines, cmdline_count, sizeof(*cmdlines), cmdline_cmp);
+}
+
+static struct func_map {
+	unsigned long long		addr;
+	char				*func;
+	char				*mod;
+} *func_list;
+static unsigned int func_count;
+
+static int func_cmp(const void *a, const void *b)
+{
+	const struct func_map *fa = a;
+	const struct func_map *fb = b;
+
+	if (fa->addr < fb->addr)
+		return -1;
+	if (fa->addr > fb->addr)
+		return 1;
+
+	return 0;
+}
+
+void parse_proc_kallsyms(char *file, unsigned int size __unused)
+{
+	struct func_list {
+		struct func_list	*next;
+		unsigned long long	addr;
+		char			*func;
+		char			*mod;
+	} *list = NULL, *item;
+	char *line;
+	char *next = NULL;
+	char *addr_str;
+	char ch;
+	int ret __used;
+	int i;
+
+	line = strtok_r(file, "\n", &next);
+	while (line) {
+		item = malloc_or_die(sizeof(*item));
+		item->mod = NULL;
+		ret = sscanf(line, "%as %c %as\t[%as",
+			     (float *)(void *)&addr_str, /* workaround gcc warning */
+			     &ch,
+			     (float *)(void *)&item->func,
+			     (float *)(void *)&item->mod);
+		item->addr = strtoull(addr_str, NULL, 16);
+		free(addr_str);
+
+		/* truncate the extra ']' */
+		if (item->mod)
+			item->mod[strlen(item->mod) - 1] = 0;
+
+
+		item->next = list;
+		list = item;
+		line = strtok_r(NULL, "\n", &next);
+		func_count++;
+	}
+
+	func_list = malloc_or_die(sizeof(*func_list) * (func_count + 1));
+
+	i = 0;
+	while (list) {
+		func_list[i].func = list->func;
+		func_list[i].addr = list->addr;
+		func_list[i].mod = list->mod;
+		i++;
+		item = list;
+		list = list->next;
+		free(item);
+	}
+
+	qsort(func_list, func_count, sizeof(*func_list), func_cmp);
+
+	/*
+	 * Add a special record at the end.
+	 */
+	func_list[func_count].func = NULL;
+	func_list[func_count].addr = 0;
+	func_list[func_count].mod = NULL;
+}
+
+/*
+ * We are searching for a record in between, not an exact
+ * match.
+ */
+static int func_bcmp(const void *a, const void *b)
+{
+	const struct func_map *fa = a;
+	const struct func_map *fb = b;
+
+	if ((fa->addr == fb->addr) ||
+
+	    (fa->addr > fb->addr &&
+	     fa->addr < (fb+1)->addr))
+		return 0;
+
+	if (fa->addr < fb->addr)
+		return -1;
+
+	return 1;
+}
+
+static struct func_map *find_func(unsigned long long addr)
+{
+	struct func_map *func;
+	struct func_map key;
+
+	key.addr = addr;
+
+	func = bsearch(&key, func_list, func_count, sizeof(*func_list),
+		       func_bcmp);
+
+	return func;
+}
+
+void print_funcs(void)
+{
+	int i;
+
+	for (i = 0; i < (int)func_count; i++) {
+		printf("%016llx %s",
+		       func_list[i].addr,
+		       func_list[i].func);
+		if (func_list[i].mod)
+			printf(" [%s]\n", func_list[i].mod);
+		else
+			printf("\n");
+	}
+}
+
+static struct printk_map {
+	unsigned long long		addr;
+	char				*printk;
+} *printk_list;
+static unsigned int printk_count;
+
+static int printk_cmp(const void *a, const void *b)
+{
+	const struct func_map *fa = a;
+	const struct func_map *fb = b;
+
+	if (fa->addr < fb->addr)
+		return -1;
+	if (fa->addr > fb->addr)
+		return 1;
+
+	return 0;
+}
+
+static struct printk_map *find_printk(unsigned long long addr)
+{
+	struct printk_map *printk;
+	struct printk_map key;
+
+	key.addr = addr;
+
+	printk = bsearch(&key, printk_list, printk_count, sizeof(*printk_list),
+			 printk_cmp);
+
+	return printk;
+}
+
+void parse_ftrace_printk(char *file, unsigned int size __unused)
+{
+	struct printk_list {
+		struct printk_list	*next;
+		unsigned long long	addr;
+		char			*printk;
+	} *list = NULL, *item;
+	char *line;
+	char *next = NULL;
+	char *addr_str;
+	int i;
+
+	line = strtok_r(file, "\n", &next);
+	while (line) {
+		addr_str = strsep(&line, ":");
+		if (!line) {
+			warning("error parsing print strings");
+			break;
+		}
+		item = malloc_or_die(sizeof(*item));
+		item->addr = strtoull(addr_str, NULL, 16);
+		/* fmt still has a space, skip it */
+		item->printk = strdup(line+1);
+		item->next = list;
+		list = item;
+		line = strtok_r(NULL, "\n", &next);
+		printk_count++;
+	}
+
+	printk_list = malloc_or_die(sizeof(*printk_list) * printk_count + 1);
+
+	i = 0;
+	while (list) {
+		printk_list[i].printk = list->printk;
+		printk_list[i].addr = list->addr;
+		i++;
+		item = list;
+		list = list->next;
+		free(item);
+	}
+
+	qsort(printk_list, printk_count, sizeof(*printk_list), printk_cmp);
+}
+
+void print_printk(void)
+{
+	int i;
+
+	for (i = 0; i < (int)printk_count; i++) {
+		printf("%016llx %s\n",
+		       printk_list[i].addr,
+		       printk_list[i].printk);
+	}
+}
+
+static struct event *alloc_event(void)
+{
+	struct event *event;
+
+	event = malloc_or_die(sizeof(*event));
+	memset(event, 0, sizeof(*event));
+
+	return event;
+}
+
+enum event_type {
+	EVENT_ERROR,
+	EVENT_NONE,
+	EVENT_SPACE,
+	EVENT_NEWLINE,
+	EVENT_OP,
+	EVENT_DELIM,
+	EVENT_ITEM,
+	EVENT_DQUOTE,
+	EVENT_SQUOTE,
+};
+
+static struct event *event_list;
+
+static void add_event(struct event *event)
+{
+	event->next = event_list;
+	event_list = event;
+}
+
+static int event_item_type(enum event_type type)
+{
+	switch (type) {
+	case EVENT_ITEM ... EVENT_SQUOTE:
+		return 1;
+	case EVENT_ERROR ... EVENT_DELIM:
+	default:
+		return 0;
+	}
+}
+
+static void free_arg(struct print_arg *arg)
+{
+	if (!arg)
+		return;
+
+	switch (arg->type) {
+	case PRINT_ATOM:
+		if (arg->atom.atom)
+			free(arg->atom.atom);
+		break;
+	case PRINT_NULL:
+	case PRINT_FIELD ... PRINT_OP:
+	default:
+		/* todo */
+		break;
+	}
+
+	free(arg);
+}
+
+static enum event_type get_type(int ch)
+{
+	if (ch == '\n')
+		return EVENT_NEWLINE;
+	if (isspace(ch))
+		return EVENT_SPACE;
+	if (isalnum(ch) || ch == '_')
+		return EVENT_ITEM;
+	if (ch == '\'')
+		return EVENT_SQUOTE;
+	if (ch == '"')
+		return EVENT_DQUOTE;
+	if (!isprint(ch))
+		return EVENT_NONE;
+	if (ch == '(' || ch == ')' || ch == ',')
+		return EVENT_DELIM;
+
+	return EVENT_OP;
+}
+
+static int __read_char(void)
+{
+	if (input_buf_ptr >= input_buf_siz)
+		return -1;
+
+	return input_buf[input_buf_ptr++];
+}
+
+static int __peek_char(void)
+{
+	if (input_buf_ptr >= input_buf_siz)
+		return -1;
+
+	return input_buf[input_buf_ptr];
+}
+
+static enum event_type __read_token(char **tok)
+{
+	char buf[BUFSIZ];
+	int ch, last_ch, quote_ch, next_ch;
+	int i = 0;
+	int tok_size = 0;
+	enum event_type type;
+
+	*tok = NULL;
+
+
+	ch = __read_char();
+	if (ch < 0)
+		return EVENT_NONE;
+
+	type = get_type(ch);
+	if (type == EVENT_NONE)
+		return type;
+
+	buf[i++] = ch;
+
+	switch (type) {
+	case EVENT_NEWLINE:
+	case EVENT_DELIM:
+		*tok = malloc_or_die(2);
+		(*tok)[0] = ch;
+		(*tok)[1] = 0;
+		return type;
+
+	case EVENT_OP:
+		switch (ch) {
+		case '-':
+			next_ch = __peek_char();
+			if (next_ch == '>') {
+				buf[i++] = __read_char();
+				break;
+			}
+			/* fall through */
+		case '+':
+		case '|':
+		case '&':
+		case '>':
+		case '<':
+			last_ch = ch;
+			ch = __peek_char();
+			if (ch != last_ch)
+				goto test_equal;
+			buf[i++] = __read_char();
+			switch (last_ch) {
+			case '>':
+			case '<':
+				goto test_equal;
+			default:
+				break;
+			}
+			break;
+		case '!':
+		case '=':
+			goto test_equal;
+		default: /* what should we do instead? */
+			break;
+		}
+		buf[i] = 0;
+		*tok = strdup(buf);
+		return type;
+
+ test_equal:
+		ch = __peek_char();
+		if (ch == '=')
+			buf[i++] = __read_char();
+		break;
+
+	case EVENT_DQUOTE:
+	case EVENT_SQUOTE:
+		/* don't keep quotes */
+		i--;
+		quote_ch = ch;
+		last_ch = 0;
+		do {
+			if (i == (BUFSIZ - 1)) {
+				buf[i] = 0;
+				if (*tok) {
+					*tok = realloc(*tok, tok_size + BUFSIZ);
+					if (!*tok)
+						return EVENT_NONE;
+					strcat(*tok, buf);
+				} else
+					*tok = strdup(buf);
+
+				if (!*tok)
+					return EVENT_NONE;
+				tok_size += BUFSIZ;
+				i = 0;
+			}
+			last_ch = ch;
+			ch = __read_char();
+			buf[i++] = ch;
+			/* the '\' '\' will cancel itself */
+			if (ch == '\\' && last_ch == '\\')
+				last_ch = 0;
+		} while (ch != quote_ch || last_ch == '\\');
+		/* remove the last quote */
+		i--;
+		goto out;
+
+	case EVENT_ERROR ... EVENT_SPACE:
+	case EVENT_ITEM:
+	default:
+		break;
+	}
+
+	while (get_type(__peek_char()) == type) {
+		if (i == (BUFSIZ - 1)) {
+			buf[i] = 0;
+			if (*tok) {
+				*tok = realloc(*tok, tok_size + BUFSIZ);
+				if (!*tok)
+					return EVENT_NONE;
+				strcat(*tok, buf);
+			} else
+				*tok = strdup(buf);
+
+			if (!*tok)
+				return EVENT_NONE;
+			tok_size += BUFSIZ;
+			i = 0;
+		}
+		ch = __read_char();
+		buf[i++] = ch;
+	}
+
+ out:
+	buf[i] = 0;
+	if (*tok) {
+		*tok = realloc(*tok, tok_size + i);
+		if (!*tok)
+			return EVENT_NONE;
+		strcat(*tok, buf);
+	} else
+		*tok = strdup(buf);
+	if (!*tok)
+		return EVENT_NONE;
+
+	return type;
+}
+
+static void free_token(char *tok)
+{
+	if (tok)
+		free(tok);
+}
+
+static enum event_type read_token(char **tok)
+{
+	enum event_type type;
+
+	for (;;) {
+		type = __read_token(tok);
+		if (type != EVENT_SPACE)
+			return type;
+
+		free_token(*tok);
+	}
+
+	/* not reached */
+	return EVENT_NONE;
+}
+
+/* no newline */
+static enum event_type read_token_item(char **tok)
+{
+	enum event_type type;
+
+	for (;;) {
+		type = __read_token(tok);
+		if (type != EVENT_SPACE && type != EVENT_NEWLINE)
+			return type;
+
+		free_token(*tok);
+	}
+
+	/* not reached */
+	return EVENT_NONE;
+}
+
+static int test_type(enum event_type type, enum event_type expect)
+{
+	if (type != expect) {
+		warning("Error: expected type %d but read %d",
+		    expect, type);
+		return -1;
+	}
+	return 0;
+}
+
+static int __test_type_token(enum event_type type, char *token,
+			     enum event_type expect, const char *expect_tok,
+			     bool warn)
+{
+	if (type != expect) {
+		if (warn)
+			warning("Error: expected type %d but read %d",
+				expect, type);
+		return -1;
+	}
+
+	if (strcmp(token, expect_tok) != 0) {
+		if (warn)
+			warning("Error: expected '%s' but read '%s'",
+				expect_tok, token);
+		return -1;
+	}
+	return 0;
+}
+
+static int test_type_token(enum event_type type, char *token,
+			   enum event_type expect, const char *expect_tok)
+{
+	return __test_type_token(type, token, expect, expect_tok, true);
+}
+
+static int __read_expect_type(enum event_type expect, char **tok, int newline_ok)
+{
+	enum event_type type;
+
+	if (newline_ok)
+		type = read_token(tok);
+	else
+		type = read_token_item(tok);
+	return test_type(type, expect);
+}
+
+static int read_expect_type(enum event_type expect, char **tok)
+{
+	return __read_expect_type(expect, tok, 1);
+}
+
+static int __read_expected(enum event_type expect, const char *str,
+			   int newline_ok, bool warn)
+{
+	enum event_type type;
+	char *token;
+	int ret;
+
+	if (newline_ok)
+		type = read_token(&token);
+	else
+		type = read_token_item(&token);
+
+	ret = __test_type_token(type, token, expect, str, warn);
+
+	free_token(token);
+
+	return ret;
+}
+
+static int read_expected(enum event_type expect, const char *str)
+{
+	return __read_expected(expect, str, 1, true);
+}
+
+static int read_expected_item(enum event_type expect, const char *str)
+{
+	return __read_expected(expect, str, 0, true);
+}
+
+static char *event_read_name(void)
+{
+	char *token;
+
+	if (read_expected(EVENT_ITEM, "name") < 0)
+		return NULL;
+
+	if (read_expected(EVENT_OP, ":") < 0)
+		return NULL;
+
+	if (read_expect_type(EVENT_ITEM, &token) < 0)
+		goto fail;
+
+	return token;
+
+ fail:
+	free_token(token);
+	return NULL;
+}
+
+static int event_read_id(void)
+{
+	char *token;
+	int id;
+
+	if (read_expected_item(EVENT_ITEM, "ID") < 0)
+		return -1;
+
+	if (read_expected(EVENT_OP, ":") < 0)
+		return -1;
+
+	if (read_expect_type(EVENT_ITEM, &token) < 0)
+		goto fail;
+
+	id = strtoul(token, NULL, 0);
+	free_token(token);
+	return id;
+
+ fail:
+	free_token(token);
+	return -1;
+}
+
+static int field_is_string(struct format_field *field)
+{
+	if ((field->flags & FIELD_IS_ARRAY) &&
+	    (!strstr(field->type, "char") || !strstr(field->type, "u8") ||
+	     !strstr(field->type, "s8")))
+		return 1;
+
+	return 0;
+}
+
+static int field_is_dynamic(struct format_field *field)
+{
+	if (!strncmp(field->type, "__data_loc", 10))
+		return 1;
+
+	return 0;
+}
+
+static int event_read_fields(struct event *event, struct format_field **fields)
+{
+	struct format_field *field = NULL;
+	enum event_type type;
+	char *token;
+	char *last_token;
+	int count = 0;
+
+	do {
+		type = read_token(&token);
+		if (type == EVENT_NEWLINE) {
+			free_token(token);
+			return count;
+		}
+
+		count++;
+
+		if (test_type_token(type, token, EVENT_ITEM, "field"))
+			goto fail;
+		free_token(token);
+
+		type = read_token(&token);
+		/*
+		 * The ftrace fields may still use the "special" name.
+		 * Just ignore it.
+		 */
+		if (event->flags & EVENT_FL_ISFTRACE &&
+		    type == EVENT_ITEM && strcmp(token, "special") == 0) {
+			free_token(token);
+			type = read_token(&token);
+		}
+
+		if (test_type_token(type, token, EVENT_OP, ":") < 0)
+			return -1;
+
+		if (read_expect_type(EVENT_ITEM, &token) < 0)
+			goto fail;
+
+		last_token = token;
+
+		field = malloc_or_die(sizeof(*field));
+		memset(field, 0, sizeof(*field));
+
+		/* read the rest of the type */
+		for (;;) {
+			type = read_token(&token);
+			if (type == EVENT_ITEM ||
+			    (type == EVENT_OP && strcmp(token, "*") == 0) ||
+			    /*
+			     * Some of the ftrace fields are broken and have
+			     * an illegal "." in them.
+			     */
+			    (event->flags & EVENT_FL_ISFTRACE &&
+			     type == EVENT_OP && strcmp(token, ".") == 0)) {
+
+				if (strcmp(token, "*") == 0)
+					field->flags |= FIELD_IS_POINTER;
+
+				if (field->type) {
+					field->type = realloc(field->type,
+							      strlen(field->type) +
+							      strlen(last_token) + 2);
+					strcat(field->type, " ");
+					strcat(field->type, last_token);
+				} else
+					field->type = last_token;
+				last_token = token;
+				continue;
+			}
+
+			break;
+		}
+
+		if (!field->type) {
+			die("no type found");
+			goto fail;
+		}
+		field->name = last_token;
+
+		if (test_type(type, EVENT_OP))
+			goto fail;
+
+		if (strcmp(token, "[") == 0) {
+			enum event_type last_type = type;
+			char *brackets = token;
+			int len;
+
+			field->flags |= FIELD_IS_ARRAY;
+
+			type = read_token(&token);
+		        while (strcmp(token, "]") != 0) {
+				if (last_type == EVENT_ITEM &&
+				    type == EVENT_ITEM)
+					len = 2;
+				else
+					len = 1;
+				last_type = type;
+
+				brackets = realloc(brackets,
+						   strlen(brackets) +
+						   strlen(token) + len);
+				if (len == 2)
+					strcat(brackets, " ");
+				strcat(brackets, token);
+				free_token(token);
+				type = read_token(&token);
+				if (type == EVENT_NONE) {
+					die("failed to find token");
+					goto fail;
+				}
+			}
+
+			free_token(token);
+
+			brackets = realloc(brackets, strlen(brackets) + 2);
+			strcat(brackets, "]");
+
+			/* add brackets to type */
+
+			type = read_token(&token);
+			/*
+			 * If the next token is not an OP, then it is of
+			 * the format: type [] item;
+			 */
+			if (type == EVENT_ITEM) {
+				field->type = realloc(field->type,
+						      strlen(field->type) +
+						      strlen(field->name) +
+						      strlen(brackets) + 2);
+				strcat(field->type, " ");
+				strcat(field->type, field->name);
+				free_token(field->name);
+				strcat(field->type, brackets);
+				field->name = token;
+				type = read_token(&token);
+			} else {
+				field->type = realloc(field->type,
+						      strlen(field->type) +
+						      strlen(brackets) + 1);
+				strcat(field->type, brackets);
+			}
+			free(brackets);
+		}
+
+		if (field_is_string(field)) {
+			field->flags |= FIELD_IS_STRING;
+			if (field_is_dynamic(field))
+				field->flags |= FIELD_IS_DYNAMIC;
+		}
+
+		if (test_type_token(type, token,  EVENT_OP, ";"))
+			goto fail;
+		free_token(token);
+
+		if (read_expected(EVENT_ITEM, "offset") < 0)
+			goto fail_expect;
+
+		if (read_expected(EVENT_OP, ":") < 0)
+			goto fail_expect;
+
+		if (read_expect_type(EVENT_ITEM, &token))
+			goto fail;
+		field->offset = strtoul(token, NULL, 0);
+		free_token(token);
+
+		if (read_expected(EVENT_OP, ";") < 0)
+			goto fail_expect;
+
+		if (read_expected(EVENT_ITEM, "size") < 0)
+			goto fail_expect;
+
+		if (read_expected(EVENT_OP, ":") < 0)
+			goto fail_expect;
+
+		if (read_expect_type(EVENT_ITEM, &token))
+			goto fail;
+		field->size = strtoul(token, NULL, 0);
+		free_token(token);
+
+		if (read_expected(EVENT_OP, ";") < 0)
+			goto fail_expect;
+
+		type = read_token(&token);
+		if (type != EVENT_NEWLINE) {
+			/* newer versions of the kernel have a "signed" type */
+			if (test_type_token(type, token, EVENT_ITEM, "signed"))
+				goto fail;
+
+			free_token(token);
+
+			if (read_expected(EVENT_OP, ":") < 0)
+				goto fail_expect;
+
+			if (read_expect_type(EVENT_ITEM, &token))
+				goto fail;
+
+			if (strtoul(token, NULL, 0))
+				field->flags |= FIELD_IS_SIGNED;
+
+			free_token(token);
+			if (read_expected(EVENT_OP, ";") < 0)
+				goto fail_expect;
+
+			if (read_expect_type(EVENT_NEWLINE, &token))
+				goto fail;
+		}
+
+		free_token(token);
+
+		*fields = field;
+		fields = &field->next;
+
+	} while (1);
+
+	return 0;
+
+fail:
+	free_token(token);
+fail_expect:
+	if (field)
+		free(field);
+	return -1;
+}
+
+static int event_read_format(struct event *event)
+{
+	char *token;
+	int ret;
+
+	if (read_expected_item(EVENT_ITEM, "format") < 0)
+		return -1;
+
+	if (read_expected(EVENT_OP, ":") < 0)
+		return -1;
+
+	if (read_expect_type(EVENT_NEWLINE, &token))
+		goto fail;
+	free_token(token);
+
+	ret = event_read_fields(event, &event->format.common_fields);
+	if (ret < 0)
+		return ret;
+	event->format.nr_common = ret;
+
+	ret = event_read_fields(event, &event->format.fields);
+	if (ret < 0)
+		return ret;
+	event->format.nr_fields = ret;
+
+	return 0;
+
+ fail:
+	free_token(token);
+	return -1;
+}
+
+enum event_type
+process_arg_token(struct event *event, struct print_arg *arg,
+		  char **tok, enum event_type type);
+
+static enum event_type
+process_arg(struct event *event, struct print_arg *arg, char **tok)
+{
+	enum event_type type;
+	char *token;
+
+	type = read_token(&token);
+	*tok = token;
+
+	return process_arg_token(event, arg, tok, type);
+}
+
+static enum event_type
+process_cond(struct event *event, struct print_arg *top, char **tok)
+{
+	struct print_arg *arg, *left, *right;
+	enum event_type type;
+	char *token = NULL;
+
+	arg = malloc_or_die(sizeof(*arg));
+	memset(arg, 0, sizeof(*arg));
+
+	left = malloc_or_die(sizeof(*left));
+
+	right = malloc_or_die(sizeof(*right));
+
+	arg->type = PRINT_OP;
+	arg->op.left = left;
+	arg->op.right = right;
+
+	*tok = NULL;
+	type = process_arg(event, left, &token);
+	if (test_type_token(type, token, EVENT_OP, ":"))
+		goto out_free;
+
+	arg->op.op = token;
+
+	type = process_arg(event, right, &token);
+
+	top->op.right = arg;
+
+	*tok = token;
+	return type;
+
+out_free:
+	free_token(*tok);
+	free(right);
+	free(left);
+	free_arg(arg);
+	return EVENT_ERROR;
+}
+
+static enum event_type
+process_array(struct event *event, struct print_arg *top, char **tok)
+{
+	struct print_arg *arg;
+	enum event_type type;
+	char *token = NULL;
+
+	arg = malloc_or_die(sizeof(*arg));
+	memset(arg, 0, sizeof(*arg));
+
+	*tok = NULL;
+	type = process_arg(event, arg, &token);
+	if (test_type_token(type, token, EVENT_OP, "]"))
+		goto out_free;
+
+	top->op.right = arg;
+
+	free_token(token);
+	type = read_token_item(&token);
+	*tok = token;
+
+	return type;
+
+out_free:
+	free_token(*tok);
+	free_arg(arg);
+	return EVENT_ERROR;
+}
+
+static int get_op_prio(char *op)
+{
+	if (!op[1]) {
+		switch (op[0]) {
+		case '*':
+		case '/':
+		case '%':
+			return 6;
+		case '+':
+		case '-':
+			return 7;
+			/* '>>' and '<<' are 8 */
+		case '<':
+		case '>':
+			return 9;
+			/* '==' and '!=' are 10 */
+		case '&':
+			return 11;
+		case '^':
+			return 12;
+		case '|':
+			return 13;
+		case '?':
+			return 16;
+		default:
+			die("unknown op '%c'", op[0]);
+			return -1;
+		}
+	} else {
+		if (strcmp(op, "++") == 0 ||
+		    strcmp(op, "--") == 0) {
+			return 3;
+		} else if (strcmp(op, ">>") == 0 ||
+			   strcmp(op, "<<") == 0) {
+			return 8;
+		} else if (strcmp(op, ">=") == 0 ||
+			   strcmp(op, "<=") == 0) {
+			return 9;
+		} else if (strcmp(op, "==") == 0 ||
+			   strcmp(op, "!=") == 0) {
+			return 10;
+		} else if (strcmp(op, "&&") == 0) {
+			return 14;
+		} else if (strcmp(op, "||") == 0) {
+			return 15;
+		} else {
+			die("unknown op '%s'", op);
+			return -1;
+		}
+	}
+}
+
+static void set_op_prio(struct print_arg *arg)
+{
+
+	/* single ops are the greatest */
+	if (!arg->op.left || arg->op.left->type == PRINT_NULL) {
+		arg->op.prio = 0;
+		return;
+	}
+
+	arg->op.prio = get_op_prio(arg->op.op);
+}
+
+static enum event_type
+process_op(struct event *event, struct print_arg *arg, char **tok)
+{
+	struct print_arg *left, *right = NULL;
+	enum event_type type;
+	char *token;
+
+	/* the op is passed in via tok */
+	token = *tok;
+
+	if (arg->type == PRINT_OP && !arg->op.left) {
+		/* handle single op */
+		if (token[1]) {
+			die("bad op token %s", token);
+			return EVENT_ERROR;
+		}
+		switch (token[0]) {
+		case '!':
+		case '+':
+		case '-':
+			break;
+		default:
+			die("bad op token %s", token);
+			return EVENT_ERROR;
+		}
+
+		/* make an empty left */
+		left = malloc_or_die(sizeof(*left));
+		left->type = PRINT_NULL;
+		arg->op.left = left;
+
+		right = malloc_or_die(sizeof(*right));
+		arg->op.right = right;
+
+		type = process_arg(event, right, tok);
+
+	} else if (strcmp(token, "?") == 0) {
+
+		left = malloc_or_die(sizeof(*left));
+		/* copy the top arg to the left */
+		*left = *arg;
+
+		arg->type = PRINT_OP;
+		arg->op.op = token;
+		arg->op.left = left;
+		arg->op.prio = 0;
+
+		type = process_cond(event, arg, tok);
+
+	} else if (strcmp(token, ">>") == 0 ||
+		   strcmp(token, "<<") == 0 ||
+		   strcmp(token, "&") == 0 ||
+		   strcmp(token, "|") == 0 ||
+		   strcmp(token, "&&") == 0 ||
+		   strcmp(token, "||") == 0 ||
+		   strcmp(token, "-") == 0 ||
+		   strcmp(token, "+") == 0 ||
+		   strcmp(token, "*") == 0 ||
+		   strcmp(token, "^") == 0 ||
+		   strcmp(token, "/") == 0 ||
+		   strcmp(token, "<") == 0 ||
+		   strcmp(token, ">") == 0 ||
+		   strcmp(token, "==") == 0 ||
+		   strcmp(token, "!=") == 0) {
+
+		left = malloc_or_die(sizeof(*left));
+
+		/* copy the top arg to the left */
+		*left = *arg;
+
+		arg->type = PRINT_OP;
+		arg->op.op = token;
+		arg->op.left = left;
+
+		set_op_prio(arg);
+
+		right = malloc_or_die(sizeof(*right));
+
+		type = read_token_item(&token);
+		*tok = token;
+
+		/* could just be a type pointer */
+		if ((strcmp(arg->op.op, "*") == 0) &&
+		    type == EVENT_DELIM && (strcmp(token, ")") == 0)) {
+			if (left->type != PRINT_ATOM)
+				die("bad pointer type");
+			left->atom.atom = realloc(left->atom.atom,
+					    sizeof(left->atom.atom) + 3);
+			strcat(left->atom.atom, " *");
+			*arg = *left;
+			free(arg);
+
+			return type;
+		}
+
+		type = process_arg_token(event, right, tok, type);
+
+		arg->op.right = right;
+
+	} else if (strcmp(token, "[") == 0) {
+
+		left = malloc_or_die(sizeof(*left));
+		*left = *arg;
+
+		arg->type = PRINT_OP;
+		arg->op.op = token;
+		arg->op.left = left;
+
+		arg->op.prio = 0;
+		type = process_array(event, arg, tok);
+
+	} else {
+		warning("unknown op '%s'", token);
+		event->flags |= EVENT_FL_FAILED;
+		/* the arg is now the left side */
+		return EVENT_NONE;
+	}
+
+	if (type == EVENT_OP) {
+		int prio;
+
+		/* higher prios need to be closer to the root */
+		prio = get_op_prio(*tok);
+
+		if (prio > arg->op.prio)
+			return process_op(event, arg, tok);
+
+		return process_op(event, right, tok);
+	}
+
+	return type;
+}
+
+static enum event_type
+process_entry(struct event *event __unused, struct print_arg *arg,
+	      char **tok)
+{
+	enum event_type type;
+	char *field;
+	char *token;
+
+	if (read_expected(EVENT_OP, "->") < 0)
+		return EVENT_ERROR;
+
+	if (read_expect_type(EVENT_ITEM, &token) < 0)
+		goto fail;
+	field = token;
+
+	arg->type = PRINT_FIELD;
+	arg->field.name = field;
+
+	if (is_flag_field) {
+		arg->field.field = find_any_field(event, arg->field.name);
+		arg->field.field->flags |= FIELD_IS_FLAG;
+		is_flag_field = 0;
+	} else if (is_symbolic_field) {
+		arg->field.field = find_any_field(event, arg->field.name);
+		arg->field.field->flags |= FIELD_IS_SYMBOLIC;
+		is_symbolic_field = 0;
+	}
+
+	type = read_token(&token);
+	*tok = token;
+
+	return type;
+
+fail:
+	free_token(token);
+	return EVENT_ERROR;
+}
+
+static char *arg_eval (struct print_arg *arg);
+
+static long long arg_num_eval(struct print_arg *arg)
+{
+	long long left, right;
+	long long val = 0;
+
+	switch (arg->type) {
+	case PRINT_ATOM:
+		val = strtoll(arg->atom.atom, NULL, 0);
+		break;
+	case PRINT_TYPE:
+		val = arg_num_eval(arg->typecast.item);
+		break;
+	case PRINT_OP:
+		switch (arg->op.op[0]) {
+		case '|':
+			left = arg_num_eval(arg->op.left);
+			right = arg_num_eval(arg->op.right);
+			if (arg->op.op[1])
+				val = left || right;
+			else
+				val = left | right;
+			break;
+		case '&':
+			left = arg_num_eval(arg->op.left);
+			right = arg_num_eval(arg->op.right);
+			if (arg->op.op[1])
+				val = left && right;
+			else
+				val = left & right;
+			break;
+		case '<':
+			left = arg_num_eval(arg->op.left);
+			right = arg_num_eval(arg->op.right);
+			switch (arg->op.op[1]) {
+			case 0:
+				val = left < right;
+				break;
+			case '<':
+				val = left << right;
+				break;
+			case '=':
+				val = left <= right;
+				break;
+			default:
+				die("unknown op '%s'", arg->op.op);
+			}
+			break;
+		case '>':
+			left = arg_num_eval(arg->op.left);
+			right = arg_num_eval(arg->op.right);
+			switch (arg->op.op[1]) {
+			case 0:
+				val = left > right;
+				break;
+			case '>':
+				val = left >> right;
+				break;
+			case '=':
+				val = left >= right;
+				break;
+			default:
+				die("unknown op '%s'", arg->op.op);
+			}
+			break;
+		case '=':
+			left = arg_num_eval(arg->op.left);
+			right = arg_num_eval(arg->op.right);
+
+			if (arg->op.op[1] != '=')
+				die("unknown op '%s'", arg->op.op);
+
+			val = left == right;
+			break;
+		case '!':
+			left = arg_num_eval(arg->op.left);
+			right = arg_num_eval(arg->op.right);
+
+			switch (arg->op.op[1]) {
+			case '=':
+				val = left != right;
+				break;
+			default:
+				die("unknown op '%s'", arg->op.op);
+			}
+			break;
+		default:
+			die("unknown op '%s'", arg->op.op);
+		}
+		break;
+
+	case PRINT_NULL:
+	case PRINT_FIELD ... PRINT_SYMBOL:
+	case PRINT_STRING:
+	default:
+		die("invalid eval type %d", arg->type);
+
+	}
+	return val;
+}
+
+static char *arg_eval (struct print_arg *arg)
+{
+	long long val;
+	static char buf[20];
+
+	switch (arg->type) {
+	case PRINT_ATOM:
+		return arg->atom.atom;
+	case PRINT_TYPE:
+		return arg_eval(arg->typecast.item);
+	case PRINT_OP:
+		val = arg_num_eval(arg);
+		sprintf(buf, "%lld", val);
+		return buf;
+
+	case PRINT_NULL:
+	case PRINT_FIELD ... PRINT_SYMBOL:
+	case PRINT_STRING:
+	default:
+		die("invalid eval type %d", arg->type);
+		break;
+	}
+
+	return NULL;
+}
+
+static enum event_type
+process_fields(struct event *event, struct print_flag_sym **list, char **tok)
+{
+	enum event_type type;
+	struct print_arg *arg = NULL;
+	struct print_flag_sym *field;
+	char *token = NULL;
+	char *value;
+
+	do {
+		free_token(token);
+		type = read_token_item(&token);
+		if (test_type_token(type, token, EVENT_OP, "{"))
+			break;
+
+		arg = malloc_or_die(sizeof(*arg));
+
+		free_token(token);
+		type = process_arg(event, arg, &token);
+		if (test_type_token(type, token, EVENT_DELIM, ","))
+			goto out_free;
+
+		field = malloc_or_die(sizeof(*field));
+		memset(field, 0, sizeof(*field));
+
+		value = arg_eval(arg);
+		field->value = strdup(value);
+
+		free_token(token);
+		type = process_arg(event, arg, &token);
+		if (test_type_token(type, token, EVENT_OP, "}"))
+			goto out_free;
+
+		value = arg_eval(arg);
+		field->str = strdup(value);
+		free_arg(arg);
+		arg = NULL;
+
+		*list = field;
+		list = &field->next;
+
+		free_token(token);
+		type = read_token_item(&token);
+	} while (type == EVENT_DELIM && strcmp(token, ",") == 0);
+
+	*tok = token;
+	return type;
+
+out_free:
+	free_arg(arg);
+	free_token(token);
+
+	return EVENT_ERROR;
+}
+
+static enum event_type
+process_flags(struct event *event, struct print_arg *arg, char **tok)
+{
+	struct print_arg *field;
+	enum event_type type;
+	char *token;
+
+	memset(arg, 0, sizeof(*arg));
+	arg->type = PRINT_FLAGS;
+
+	if (read_expected_item(EVENT_DELIM, "(") < 0)
+		return EVENT_ERROR;
+
+	field = malloc_or_die(sizeof(*field));
+
+	type = process_arg(event, field, &token);
+	while (type == EVENT_OP)
+		type = process_op(event, field, &token);
+	if (test_type_token(type, token, EVENT_DELIM, ","))
+		goto out_free;
+
+	arg->flags.field = field;
+
+	type = read_token_item(&token);
+	if (event_item_type(type)) {
+		arg->flags.delim = token;
+		type = read_token_item(&token);
+	}
+
+	if (test_type_token(type, token, EVENT_DELIM, ","))
+		goto out_free;
+
+	type = process_fields(event, &arg->flags.flags, &token);
+	if (test_type_token(type, token, EVENT_DELIM, ")"))
+		goto out_free;
+
+	free_token(token);
+	type = read_token_item(tok);
+	return type;
+
+out_free:
+	free_token(token);
+	return EVENT_ERROR;
+}
+
+static enum event_type
+process_symbols(struct event *event, struct print_arg *arg, char **tok)
+{
+	struct print_arg *field;
+	enum event_type type;
+	char *token;
+
+	memset(arg, 0, sizeof(*arg));
+	arg->type = PRINT_SYMBOL;
+
+	if (read_expected_item(EVENT_DELIM, "(") < 0)
+		return EVENT_ERROR;
+
+	field = malloc_or_die(sizeof(*field));
+
+	type = process_arg(event, field, &token);
+	while (type == EVENT_OP)
+		type = process_op(event, field, &token);
+	if (test_type_token(type, token, EVENT_DELIM, ","))
+		goto out_free;
+
+	arg->symbol.field = field;
+
+	type = process_fields(event, &arg->symbol.symbols, &token);
+	if (test_type_token(type, token, EVENT_DELIM, ")"))
+		goto out_free;
+
+	free_token(token);
+	type = read_token_item(tok);
+	return type;
+
+out_free:
+	free_token(token);
+	return EVENT_ERROR;
+}
+
+static enum event_type
+process_paren(struct event *event, struct print_arg *arg, char **tok)
+{
+	struct print_arg *item_arg;
+	enum event_type type;
+	char *token;
+
+	type = process_arg(event, arg, &token);
+
+	if (type == EVENT_ERROR)
+		return EVENT_ERROR;
+
+	if (type == EVENT_OP)
+		type = process_op(event, arg, &token);
+
+	if (type == EVENT_ERROR)
+		return EVENT_ERROR;
+
+	if (test_type_token(type, token, EVENT_DELIM, ")")) {
+		free_token(token);
+		return EVENT_ERROR;
+	}
+
+	free_token(token);
+	type = read_token_item(&token);
+
+	/*
+	 * If the next token is an item or another open paren, then
+	 * this was a typecast.
+	 */
+	if (event_item_type(type) ||
+	    (type == EVENT_DELIM && strcmp(token, "(") == 0)) {
+
+		/* make this a typecast and contine */
+
+		/* prevous must be an atom */
+		if (arg->type != PRINT_ATOM)
+			die("previous needed to be PRINT_ATOM");
+
+		item_arg = malloc_or_die(sizeof(*item_arg));
+
+		arg->type = PRINT_TYPE;
+		arg->typecast.type = arg->atom.atom;
+		arg->typecast.item = item_arg;
+		type = process_arg_token(event, item_arg, &token, type);
+
+	}
+
+	*tok = token;
+	return type;
+}
+
+
+static enum event_type
+process_str(struct event *event __unused, struct print_arg *arg, char **tok)
+{
+	enum event_type type;
+	char *token;
+
+	if (read_expected(EVENT_DELIM, "(") < 0)
+		return EVENT_ERROR;
+
+	if (read_expect_type(EVENT_ITEM, &token) < 0)
+		goto fail;
+
+	arg->type = PRINT_STRING;
+	arg->string.string = token;
+	arg->string.offset = -1;
+
+	if (read_expected(EVENT_DELIM, ")") < 0)
+		return EVENT_ERROR;
+
+	type = read_token(&token);
+	*tok = token;
+
+	return type;
+fail:
+	free_token(token);
+	return EVENT_ERROR;
+}
+
+enum event_type
+process_arg_token(struct event *event, struct print_arg *arg,
+		  char **tok, enum event_type type)
+{
+	char *token;
+	char *atom;
+
+	token = *tok;
+
+	switch (type) {
+	case EVENT_ITEM:
+		if (strcmp(token, "REC") == 0) {
+			free_token(token);
+			type = process_entry(event, arg, &token);
+		} else if (strcmp(token, "__print_flags") == 0) {
+			free_token(token);
+			is_flag_field = 1;
+			type = process_flags(event, arg, &token);
+		} else if (strcmp(token, "__print_symbolic") == 0) {
+			free_token(token);
+			is_symbolic_field = 1;
+			type = process_symbols(event, arg, &token);
+		} else if (strcmp(token, "__get_str") == 0) {
+			free_token(token);
+			type = process_str(event, arg, &token);
+		} else {
+			atom = token;
+			/* test the next token */
+			type = read_token_item(&token);
+
+			/* atoms can be more than one token long */
+			while (type == EVENT_ITEM) {
+				atom = realloc(atom, strlen(atom) + strlen(token) + 2);
+				strcat(atom, " ");
+				strcat(atom, token);
+				free_token(token);
+				type = read_token_item(&token);
+			}
+
+			/* todo, test for function */
+
+			arg->type = PRINT_ATOM;
+			arg->atom.atom = atom;
+		}
+		break;
+	case EVENT_DQUOTE:
+	case EVENT_SQUOTE:
+		arg->type = PRINT_ATOM;
+		arg->atom.atom = token;
+		type = read_token_item(&token);
+		break;
+	case EVENT_DELIM:
+		if (strcmp(token, "(") == 0) {
+			free_token(token);
+			type = process_paren(event, arg, &token);
+			break;
+		}
+	case EVENT_OP:
+		/* handle single ops */
+		arg->type = PRINT_OP;
+		arg->op.op = token;
+		arg->op.left = NULL;
+		type = process_op(event, arg, &token);
+
+		break;
+
+	case EVENT_ERROR ... EVENT_NEWLINE:
+	default:
+		die("unexpected type %d", type);
+	}
+	*tok = token;
+
+	return type;
+}
+
+static int event_read_print_args(struct event *event, struct print_arg **list)
+{
+	enum event_type type = EVENT_ERROR;
+	struct print_arg *arg;
+	char *token;
+	int args = 0;
+
+	do {
+		if (type == EVENT_NEWLINE) {
+			free_token(token);
+			type = read_token_item(&token);
+			continue;
+		}
+
+		arg = malloc_or_die(sizeof(*arg));
+		memset(arg, 0, sizeof(*arg));
+
+		type = process_arg(event, arg, &token);
+
+		if (type == EVENT_ERROR) {
+			free_arg(arg);
+			return -1;
+		}
+
+		*list = arg;
+		args++;
+
+		if (type == EVENT_OP) {
+			type = process_op(event, arg, &token);
+			list = &arg->next;
+			continue;
+		}
+
+		if (type == EVENT_DELIM && strcmp(token, ",") == 0) {
+			free_token(token);
+			*list = arg;
+			list = &arg->next;
+			continue;
+		}
+		break;
+	} while (type != EVENT_NONE);
+
+	if (type != EVENT_NONE)
+		free_token(token);
+
+	return args;
+}
+
+static int event_read_print(struct event *event)
+{
+	enum event_type type;
+	char *token;
+	int ret;
+
+	if (read_expected_item(EVENT_ITEM, "print") < 0)
+		return -1;
+
+	if (read_expected(EVENT_ITEM, "fmt") < 0)
+		return -1;
+
+	if (read_expected(EVENT_OP, ":") < 0)
+		return -1;
+
+	if (read_expect_type(EVENT_DQUOTE, &token) < 0)
+		goto fail;
+
+ concat:
+	event->print_fmt.format = token;
+	event->print_fmt.args = NULL;
+
+	/* ok to have no arg */
+	type = read_token_item(&token);
+
+	if (type == EVENT_NONE)
+		return 0;
+
+	/* Handle concatination of print lines */
+	if (type == EVENT_DQUOTE) {
+		char *cat;
+
+		cat = malloc_or_die(strlen(event->print_fmt.format) +
+				    strlen(token) + 1);
+		strcpy(cat, event->print_fmt.format);
+		strcat(cat, token);
+		free_token(token);
+		free_token(event->print_fmt.format);
+		event->print_fmt.format = NULL;
+		token = cat;
+		goto concat;
+	}
+
+	if (test_type_token(type, token, EVENT_DELIM, ","))
+		goto fail;
+
+	free_token(token);
+
+	ret = event_read_print_args(event, &event->print_fmt.args);
+	if (ret < 0)
+		return -1;
+
+	return ret;
+
+ fail:
+	free_token(token);
+	return -1;
+}
+
+static struct format_field *
+find_common_field(struct event *event, const char *name)
+{
+	struct format_field *format;
+
+	for (format = event->format.common_fields;
+	     format; format = format->next) {
+		if (strcmp(format->name, name) == 0)
+			break;
+	}
+
+	return format;
+}
+
+static struct format_field *
+find_field(struct event *event, const char *name)
+{
+	struct format_field *format;
+
+	for (format = event->format.fields;
+	     format; format = format->next) {
+		if (strcmp(format->name, name) == 0)
+			break;
+	}
+
+	return format;
+}
+
+static struct format_field *
+find_any_field(struct event *event, const char *name)
+{
+	struct format_field *format;
+
+	format = find_common_field(event, name);
+	if (format)
+		return format;
+	return find_field(event, name);
+}
+
+unsigned long long read_size(void *ptr, int size)
+{
+	switch (size) {
+	case 1:
+		return *(unsigned char *)ptr;
+	case 2:
+		return data2host2(ptr);
+	case 4:
+		return data2host4(ptr);
+	case 8:
+		return data2host8(ptr);
+	default:
+		/* BUG! */
+		return 0;
+	}
+}
+
+unsigned long long
+raw_field_value(struct event *event, const char *name, void *data)
+{
+	struct format_field *field;
+
+	field = find_any_field(event, name);
+	if (!field)
+		return 0ULL;
+
+	return read_size(data + field->offset, field->size);
+}
+
+void *raw_field_ptr(struct event *event, const char *name, void *data)
+{
+	struct format_field *field;
+
+	field = find_any_field(event, name);
+	if (!field)
+		return NULL;
+
+	if (field->flags & FIELD_IS_DYNAMIC) {
+		int offset;
+
+		offset = *(int *)(data + field->offset);
+		offset &= 0xffff;
+
+		return data + offset;
+	}
+
+	return data + field->offset;
+}
+
+static int get_common_info(const char *type, int *offset, int *size)
+{
+	struct event *event;
+	struct format_field *field;
+
+	/*
+	 * All events should have the same common elements.
+	 * Pick any event to find where the type is;
+	 */
+	if (!event_list)
+		die("no event_list!");
+
+	event = event_list;
+	field = find_common_field(event, type);
+	if (!field)
+		die("field '%s' not found", type);
+
+	*offset = field->offset;
+	*size = field->size;
+
+	return 0;
+}
+
+static int __parse_common(void *data, int *size, int *offset,
+			  const char *name)
+{
+	int ret;
+
+	if (!*size) {
+		ret = get_common_info(name, offset, size);
+		if (ret < 0)
+			return ret;
+	}
+	return read_size(data + *offset, *size);
+}
+
+int trace_parse_common_type(void *data)
+{
+	static int type_offset;
+	static int type_size;
+
+	return __parse_common(data, &type_size, &type_offset,
+			      "common_type");
+}
+
+int trace_parse_common_pid(void *data)
+{
+	static int pid_offset;
+	static int pid_size;
+
+	return __parse_common(data, &pid_size, &pid_offset,
+			      "common_pid");
+}
+
+int parse_common_pc(void *data)
+{
+	static int pc_offset;
+	static int pc_size;
+
+	return __parse_common(data, &pc_size, &pc_offset,
+			      "common_preempt_count");
+}
+
+int parse_common_flags(void *data)
+{
+	static int flags_offset;
+	static int flags_size;
+
+	return __parse_common(data, &flags_size, &flags_offset,
+			      "common_flags");
+}
+
+int parse_common_lock_depth(void *data)
+{
+	static int ld_offset;
+	static int ld_size;
+	int ret;
+
+	ret = __parse_common(data, &ld_size, &ld_offset,
+			     "common_lock_depth");
+	if (ret < 0)
+		return -1;
+
+	return ret;
+}
+
+struct event *trace_find_event(int id)
+{
+	struct event *event;
+
+	for (event = event_list; event; event = event->next) {
+		if (event->id == id)
+			break;
+	}
+	return event;
+}
+
+struct event *trace_find_next_event(struct event *event)
+{
+	if (!event)
+		return event_list;
+
+	return event->next;
+}
+
+static unsigned long long eval_num_arg(void *data, int size,
+				   struct event *event, struct print_arg *arg)
+{
+	unsigned long long val = 0;
+	unsigned long long left, right;
+	struct print_arg *larg;
+
+	switch (arg->type) {
+	case PRINT_NULL:
+		/* ?? */
+		return 0;
+	case PRINT_ATOM:
+		return strtoull(arg->atom.atom, NULL, 0);
+	case PRINT_FIELD:
+		if (!arg->field.field) {
+			arg->field.field = find_any_field(event, arg->field.name);
+			if (!arg->field.field)
+				die("field %s not found", arg->field.name);
+		}
+		/* must be a number */
+		val = read_size(data + arg->field.field->offset,
+				arg->field.field->size);
+		break;
+	case PRINT_FLAGS:
+	case PRINT_SYMBOL:
+		break;
+	case PRINT_TYPE:
+		return eval_num_arg(data, size, event, arg->typecast.item);
+	case PRINT_STRING:
+		return 0;
+		break;
+	case PRINT_OP:
+		if (strcmp(arg->op.op, "[") == 0) {
+			/*
+			 * Arrays are special, since we don't want
+			 * to read the arg as is.
+			 */
+			if (arg->op.left->type != PRINT_FIELD)
+				goto default_op; /* oops, all bets off */
+			larg = arg->op.left;
+			if (!larg->field.field) {
+				larg->field.field =
+					find_any_field(event, larg->field.name);
+				if (!larg->field.field)
+					die("field %s not found", larg->field.name);
+			}
+			right = eval_num_arg(data, size, event, arg->op.right);
+			val = read_size(data + larg->field.field->offset +
+					right * long_size, long_size);
+			break;
+		}
+ default_op:
+		left = eval_num_arg(data, size, event, arg->op.left);
+		right = eval_num_arg(data, size, event, arg->op.right);
+		switch (arg->op.op[0]) {
+		case '|':
+			if (arg->op.op[1])
+				val = left || right;
+			else
+				val = left | right;
+			break;
+		case '&':
+			if (arg->op.op[1])
+				val = left && right;
+			else
+				val = left & right;
+			break;
+		case '<':
+			switch (arg->op.op[1]) {
+			case 0:
+				val = left < right;
+				break;
+			case '<':
+				val = left << right;
+				break;
+			case '=':
+				val = left <= right;
+				break;
+			default:
+				die("unknown op '%s'", arg->op.op);
+			}
+			break;
+		case '>':
+			switch (arg->op.op[1]) {
+			case 0:
+				val = left > right;
+				break;
+			case '>':
+				val = left >> right;
+				break;
+			case '=':
+				val = left >= right;
+				break;
+			default:
+				die("unknown op '%s'", arg->op.op);
+			}
+			break;
+		case '=':
+			if (arg->op.op[1] != '=')
+				die("unknown op '%s'", arg->op.op);
+			val = left == right;
+			break;
+		case '-':
+			val = left - right;
+			break;
+		case '+':
+			val = left + right;
+			break;
+		default:
+			die("unknown op '%s'", arg->op.op);
+		}
+		break;
+	default: /* not sure what to do there */
+		return 0;
+	}
+	return val;
+}
+
+struct flag {
+	const char *name;
+	unsigned long long value;
+};
+
+static const struct flag flags[] = {
+	{ "HI_SOFTIRQ", 0 },
+	{ "TIMER_SOFTIRQ", 1 },
+	{ "NET_TX_SOFTIRQ", 2 },
+	{ "NET_RX_SOFTIRQ", 3 },
+	{ "BLOCK_SOFTIRQ", 4 },
+	{ "BLOCK_IOPOLL_SOFTIRQ", 5 },
+	{ "TASKLET_SOFTIRQ", 6 },
+	{ "SCHED_SOFTIRQ", 7 },
+	{ "HRTIMER_SOFTIRQ", 8 },
+	{ "RCU_SOFTIRQ", 9 },
+
+	{ "HRTIMER_NORESTART", 0 },
+	{ "HRTIMER_RESTART", 1 },
+};
+
+unsigned long long eval_flag(const char *flag)
+{
+	int i;
+
+	/*
+	 * Some flags in the format files do not get converted.
+	 * If the flag is not numeric, see if it is something that
+	 * we already know about.
+	 */
+	if (isdigit(flag[0]))
+		return strtoull(flag, NULL, 0);
+
+	for (i = 0; i < (int)(sizeof(flags)/sizeof(flags[0])); i++)
+		if (strcmp(flags[i].name, flag) == 0)
+			return flags[i].value;
+
+	return 0;
+}
+
+static void print_str_arg(void *data, int size,
+			  struct event *event, struct print_arg *arg)
+{
+	struct print_flag_sym *flag;
+	unsigned long long val, fval;
+	char *str;
+	int print;
+
+	switch (arg->type) {
+	case PRINT_NULL:
+		/* ?? */
+		return;
+	case PRINT_ATOM:
+		printf("%s", arg->atom.atom);
+		return;
+	case PRINT_FIELD:
+		if (!arg->field.field) {
+			arg->field.field = find_any_field(event, arg->field.name);
+			if (!arg->field.field)
+				die("field %s not found", arg->field.name);
+		}
+		str = malloc_or_die(arg->field.field->size + 1);
+		memcpy(str, data + arg->field.field->offset,
+		       arg->field.field->size);
+		str[arg->field.field->size] = 0;
+		printf("%s", str);
+		free(str);
+		break;
+	case PRINT_FLAGS:
+		val = eval_num_arg(data, size, event, arg->flags.field);
+		print = 0;
+		for (flag = arg->flags.flags; flag; flag = flag->next) {
+			fval = eval_flag(flag->value);
+			if (!val && !fval) {
+				printf("%s", flag->str);
+				break;
+			}
+			if (fval && (val & fval) == fval) {
+				if (print && arg->flags.delim)
+					printf("%s", arg->flags.delim);
+				printf("%s", flag->str);
+				print = 1;
+				val &= ~fval;
+			}
+		}
+		break;
+	case PRINT_SYMBOL:
+		val = eval_num_arg(data, size, event, arg->symbol.field);
+		for (flag = arg->symbol.symbols; flag; flag = flag->next) {
+			fval = eval_flag(flag->value);
+			if (val == fval) {
+				printf("%s", flag->str);
+				break;
+			}
+		}
+		break;
+
+	case PRINT_TYPE:
+		break;
+	case PRINT_STRING: {
+		int str_offset;
+
+		if (arg->string.offset == -1) {
+			struct format_field *f;
+
+			f = find_any_field(event, arg->string.string);
+			arg->string.offset = f->offset;
+		}
+		str_offset = *(int *)(data + arg->string.offset);
+		str_offset &= 0xffff;
+		printf("%s", ((char *)data) + str_offset);
+		break;
+	}
+	case PRINT_OP:
+		/*
+		 * The only op for string should be ? :
+		 */
+		if (arg->op.op[0] != '?')
+			return;
+		val = eval_num_arg(data, size, event, arg->op.left);
+		if (val)
+			print_str_arg(data, size, event, arg->op.right->op.left);
+		else
+			print_str_arg(data, size, event, arg->op.right->op.right);
+		break;
+	default:
+		/* well... */
+		break;
+	}
+}
+
+static struct print_arg *make_bprint_args(char *fmt, void *data, int size, struct event *event)
+{
+	static struct format_field *field, *ip_field;
+	struct print_arg *args, *arg, **next;
+	unsigned long long ip, val;
+	char *ptr;
+	void *bptr;
+
+	if (!field) {
+		field = find_field(event, "buf");
+		if (!field)
+			die("can't find buffer field for binary printk");
+		ip_field = find_field(event, "ip");
+		if (!ip_field)
+			die("can't find ip field for binary printk");
+	}
+
+	ip = read_size(data + ip_field->offset, ip_field->size);
+
+	/*
+	 * The first arg is the IP pointer.
+	 */
+	args = malloc_or_die(sizeof(*args));
+	arg = args;
+	arg->next = NULL;
+	next = &arg->next;
+
+	arg->type = PRINT_ATOM;
+	arg->atom.atom = malloc_or_die(32);
+	sprintf(arg->atom.atom, "%lld", ip);
+
+	/* skip the first "%pf : " */
+	for (ptr = fmt + 6, bptr = data + field->offset;
+	     bptr < data + size && *ptr; ptr++) {
+		int ls = 0;
+
+		if (*ptr == '%') {
+ process_again:
+			ptr++;
+			switch (*ptr) {
+			case '%':
+				break;
+			case 'l':
+				ls++;
+				goto process_again;
+			case 'L':
+				ls = 2;
+				goto process_again;
+			case '0' ... '9':
+				goto process_again;
+			case 'p':
+				ls = 1;
+				/* fall through */
+			case 'd':
+			case 'u':
+			case 'x':
+			case 'i':
+				/* the pointers are always 4 bytes aligned */
+				bptr = (void *)(((unsigned long)bptr + 3) &
+						~3);
+				switch (ls) {
+				case 0:
+				case 1:
+					ls = long_size;
+					break;
+				case 2:
+					ls = 8;
+				default:
+					break;
+				}
+				val = read_size(bptr, ls);
+				bptr += ls;
+				arg = malloc_or_die(sizeof(*arg));
+				arg->next = NULL;
+				arg->type = PRINT_ATOM;
+				arg->atom.atom = malloc_or_die(32);
+				sprintf(arg->atom.atom, "%lld", val);
+				*next = arg;
+				next = &arg->next;
+				break;
+			case 's':
+				arg = malloc_or_die(sizeof(*arg));
+				arg->next = NULL;
+				arg->type = PRINT_STRING;
+				arg->string.string = strdup(bptr);
+				bptr += strlen(bptr) + 1;
+				*next = arg;
+				next = &arg->next;
+			default:
+				break;
+			}
+		}
+	}
+
+	return args;
+}
+
+static void free_args(struct print_arg *args)
+{
+	struct print_arg *next;
+
+	while (args) {
+		next = args->next;
+
+		if (args->type == PRINT_ATOM)
+			free(args->atom.atom);
+		else
+			free(args->string.string);
+		free(args);
+		args = next;
+	}
+}
+
+static char *get_bprint_format(void *data, int size __unused, struct event *event)
+{
+	unsigned long long addr;
+	static struct format_field *field;
+	struct printk_map *printk;
+	char *format;
+	char *p;
+
+	if (!field) {
+		field = find_field(event, "fmt");
+		if (!field)
+			die("can't find format field for binary printk");
+		printf("field->offset = %d size=%d\n", field->offset, field->size);
+	}
+
+	addr = read_size(data + field->offset, field->size);
+
+	printk = find_printk(addr);
+	if (!printk) {
+		format = malloc_or_die(45);
+		sprintf(format, "%%pf : (NO FORMAT FOUND at %llx)\n",
+			addr);
+		return format;
+	}
+
+	p = printk->printk;
+	/* Remove any quotes. */
+	if (*p == '"')
+		p++;
+	format = malloc_or_die(strlen(p) + 10);
+	sprintf(format, "%s : %s", "%pf", p);
+	/* remove ending quotes and new line since we will add one too */
+	p = format + strlen(format) - 1;
+	if (*p == '"')
+		*p = 0;
+
+	p -= 2;
+	if (strcmp(p, "\\n") == 0)
+		*p = 0;
+
+	return format;
+}
+
+static void pretty_print(void *data, int size, struct event *event)
+{
+	struct print_fmt *print_fmt = &event->print_fmt;
+	struct print_arg *arg = print_fmt->args;
+	struct print_arg *args = NULL;
+	const char *ptr = print_fmt->format;
+	unsigned long long val;
+	struct func_map *func;
+	const char *saveptr;
+	char *bprint_fmt = NULL;
+	char format[32];
+	int show_func;
+	int len;
+	int ls;
+
+	if (event->flags & EVENT_FL_ISFUNC)
+		ptr = " %pF <-- %pF";
+
+	if (event->flags & EVENT_FL_ISBPRINT) {
+		bprint_fmt = get_bprint_format(data, size, event);
+		args = make_bprint_args(bprint_fmt, data, size, event);
+		arg = args;
+		ptr = bprint_fmt;
+	}
+
+	for (; *ptr; ptr++) {
+		ls = 0;
+		if (*ptr == '\\') {
+			ptr++;
+			switch (*ptr) {
+			case 'n':
+				printf("\n");
+				break;
+			case 't':
+				printf("\t");
+				break;
+			case 'r':
+				printf("\r");
+				break;
+			case '\\':
+				printf("\\");
+				break;
+			default:
+				printf("%c", *ptr);
+				break;
+			}
+
+		} else if (*ptr == '%') {
+			saveptr = ptr;
+			show_func = 0;
+ cont_process:
+			ptr++;
+			switch (*ptr) {
+			case '%':
+				printf("%%");
+				break;
+			case 'l':
+				ls++;
+				goto cont_process;
+			case 'L':
+				ls = 2;
+				goto cont_process;
+			case 'z':
+			case 'Z':
+			case '0' ... '9':
+				goto cont_process;
+			case 'p':
+				if (long_size == 4)
+					ls = 1;
+				else
+					ls = 2;
+
+				if (*(ptr+1) == 'F' ||
+				    *(ptr+1) == 'f') {
+					ptr++;
+					show_func = *ptr;
+				}
+
+				/* fall through */
+			case 'd':
+			case 'i':
+			case 'x':
+			case 'X':
+			case 'u':
+				if (!arg)
+					die("no argument match");
+
+				len = ((unsigned long)ptr + 1) -
+					(unsigned long)saveptr;
+
+				/* should never happen */
+				if (len > 32)
+					die("bad format!");
+
+				memcpy(format, saveptr, len);
+				format[len] = 0;
+
+				val = eval_num_arg(data, size, event, arg);
+				arg = arg->next;
+
+				if (show_func) {
+					func = find_func(val);
+					if (func) {
+						printf("%s", func->func);
+						if (show_func == 'F')
+							printf("+0x%llx",
+							       val - func->addr);
+						break;
+					}
+				}
+				switch (ls) {
+				case 0:
+					printf(format, (int)val);
+					break;
+				case 1:
+					printf(format, (long)val);
+					break;
+				case 2:
+					printf(format, (long long)val);
+					break;
+				default:
+					die("bad count (%d)", ls);
+				}
+				break;
+			case 's':
+				if (!arg)
+					die("no matching argument");
+
+				print_str_arg(data, size, event, arg);
+				arg = arg->next;
+				break;
+			default:
+				printf(">%c<", *ptr);
+
+			}
+		} else
+			printf("%c", *ptr);
+	}
+
+	if (args) {
+		free_args(args);
+		free(bprint_fmt);
+	}
+}
+
+static inline int log10_cpu(int nb)
+{
+	if (nb / 100)
+		return 3;
+	if (nb / 10)
+		return 2;
+	return 1;
+}
+
+static void print_lat_fmt(void *data, int size __unused)
+{
+	unsigned int lat_flags;
+	unsigned int pc;
+	int lock_depth;
+	int hardirq;
+	int softirq;
+
+	lat_flags = parse_common_flags(data);
+	pc = parse_common_pc(data);
+	lock_depth = parse_common_lock_depth(data);
+
+	hardirq = lat_flags & TRACE_FLAG_HARDIRQ;
+	softirq = lat_flags & TRACE_FLAG_SOFTIRQ;
+
+	printf("%c%c%c",
+	       (lat_flags & TRACE_FLAG_IRQS_OFF) ? 'd' :
+	       (lat_flags & TRACE_FLAG_IRQS_NOSUPPORT) ?
+	       'X' : '.',
+	       (lat_flags & TRACE_FLAG_NEED_RESCHED) ?
+	       'N' : '.',
+	       (hardirq && softirq) ? 'H' :
+	       hardirq ? 'h' : softirq ? 's' : '.');
+
+	if (pc)
+		printf("%x", pc);
+	else
+		printf(".");
+
+	if (lock_depth < 0)
+		printf(". ");
+	else
+		printf("%d ", lock_depth);
+}
+
+#define TRACE_GRAPH_INDENT	2
+
+static struct record *
+get_return_for_leaf(int cpu, int cur_pid, unsigned long long cur_func,
+		    struct record *next)
+{
+	struct format_field *field;
+	struct event *event;
+	unsigned long val;
+	int type;
+	int pid;
+
+	type = trace_parse_common_type(next->data);
+	event = trace_find_event(type);
+	if (!event)
+		return NULL;
+
+	if (!(event->flags & EVENT_FL_ISFUNCRET))
+		return NULL;
+
+	pid = trace_parse_common_pid(next->data);
+	field = find_field(event, "func");
+	if (!field)
+		die("function return does not have field func");
+
+	val = read_size(next->data + field->offset, field->size);
+
+	if (cur_pid != pid || cur_func != val)
+		return NULL;
+
+	/* this is a leaf, now advance the iterator */
+	return trace_read_data(cpu);
+}
+
+/* Signal a overhead of time execution to the output */
+static void print_graph_overhead(unsigned long long duration)
+{
+	/* Non nested entry or return */
+	if (duration == ~0ULL)
+		return (void)printf("  ");
+
+	/* Duration exceeded 100 msecs */
+	if (duration > 100000ULL)
+		return (void)printf("! ");
+
+	/* Duration exceeded 10 msecs */
+	if (duration > 10000ULL)
+		return (void)printf("+ ");
+
+	printf("  ");
+}
+
+static void print_graph_duration(unsigned long long duration)
+{
+	unsigned long usecs = duration / 1000;
+	unsigned long nsecs_rem = duration % 1000;
+	/* log10(ULONG_MAX) + '\0' */
+	char msecs_str[21];
+	char nsecs_str[5];
+	int len;
+	int i;
+
+	sprintf(msecs_str, "%lu", usecs);
+
+	/* Print msecs */
+	len = printf("%lu", usecs);
+
+	/* Print nsecs (we don't want to exceed 7 numbers) */
+	if (len < 7) {
+		snprintf(nsecs_str, 8 - len, "%03lu", nsecs_rem);
+		len += printf(".%s", nsecs_str);
+	}
+
+	printf(" us ");
+
+	/* Print remaining spaces to fit the row's width */
+	for (i = len; i < 7; i++)
+		printf(" ");
+
+	printf("|  ");
+}
+
+static void
+print_graph_entry_leaf(struct event *event, void *data, struct record *ret_rec)
+{
+	unsigned long long rettime, calltime;
+	unsigned long long duration, depth;
+	unsigned long long val;
+	struct format_field *field;
+	struct func_map *func;
+	struct event *ret_event;
+	int type;
+	int i;
+
+	type = trace_parse_common_type(ret_rec->data);
+	ret_event = trace_find_event(type);
+
+	field = find_field(ret_event, "rettime");
+	if (!field)
+		die("can't find rettime in return graph");
+	rettime = read_size(ret_rec->data + field->offset, field->size);
+
+	field = find_field(ret_event, "calltime");
+	if (!field)
+		die("can't find rettime in return graph");
+	calltime = read_size(ret_rec->data + field->offset, field->size);
+
+	duration = rettime - calltime;
+
+	/* Overhead */
+	print_graph_overhead(duration);
+
+	/* Duration */
+	print_graph_duration(duration);
+
+	field = find_field(event, "depth");
+	if (!field)
+		die("can't find depth in entry graph");
+	depth = read_size(data + field->offset, field->size);
+
+	/* Function */
+	for (i = 0; i < (int)(depth * TRACE_GRAPH_INDENT); i++)
+		printf(" ");
+
+	field = find_field(event, "func");
+	if (!field)
+		die("can't find func in entry graph");
+	val = read_size(data + field->offset, field->size);
+	func = find_func(val);
+
+	if (func)
+		printf("%s();", func->func);
+	else
+		printf("%llx();", val);
+}
+
+static void print_graph_nested(struct event *event, void *data)
+{
+	struct format_field *field;
+	unsigned long long depth;
+	unsigned long long val;
+	struct func_map *func;
+	int i;
+
+	/* No overhead */
+	print_graph_overhead(-1);
+
+	/* No time */
+	printf("           |  ");
+
+	field = find_field(event, "depth");
+	if (!field)
+		die("can't find depth in entry graph");
+	depth = read_size(data + field->offset, field->size);
+
+	/* Function */
+	for (i = 0; i < (int)(depth * TRACE_GRAPH_INDENT); i++)
+		printf(" ");
+
+	field = find_field(event, "func");
+	if (!field)
+		die("can't find func in entry graph");
+	val = read_size(data + field->offset, field->size);
+	func = find_func(val);
+
+	if (func)
+		printf("%s() {", func->func);
+	else
+		printf("%llx() {", val);
+}
+
+static void
+pretty_print_func_ent(void *data, int size, struct event *event,
+		      int cpu, int pid)
+{
+	struct format_field *field;
+	struct record *rec;
+	void *copy_data;
+	unsigned long val;
+
+	if (latency_format) {
+		print_lat_fmt(data, size);
+		printf(" | ");
+	}
+
+	field = find_field(event, "func");
+	if (!field)
+		die("function entry does not have func field");
+
+	val = read_size(data + field->offset, field->size);
+
+	/*
+	 * peek_data may unmap the data pointer. Copy it first.
+	 */
+	copy_data = malloc_or_die(size);
+	memcpy(copy_data, data, size);
+	data = copy_data;
+
+	rec = trace_peek_data(cpu);
+	if (rec) {
+		rec = get_return_for_leaf(cpu, pid, val, rec);
+		if (rec) {
+			print_graph_entry_leaf(event, data, rec);
+			goto out_free;
+		}
+	}
+	print_graph_nested(event, data);
+out_free:
+	free(data);
+}
+
+static void
+pretty_print_func_ret(void *data, int size __unused, struct event *event)
+{
+	unsigned long long rettime, calltime;
+	unsigned long long duration, depth;
+	struct format_field *field;
+	int i;
+
+	if (latency_format) {
+		print_lat_fmt(data, size);
+		printf(" | ");
+	}
+
+	field = find_field(event, "rettime");
+	if (!field)
+		die("can't find rettime in return graph");
+	rettime = read_size(data + field->offset, field->size);
+
+	field = find_field(event, "calltime");
+	if (!field)
+		die("can't find calltime in return graph");
+	calltime = read_size(data + field->offset, field->size);
+
+	duration = rettime - calltime;
+
+	/* Overhead */
+	print_graph_overhead(duration);
+
+	/* Duration */
+	print_graph_duration(duration);
+
+	field = find_field(event, "depth");
+	if (!field)
+		die("can't find depth in entry graph");
+	depth = read_size(data + field->offset, field->size);
+
+	/* Function */
+	for (i = 0; i < (int)(depth * TRACE_GRAPH_INDENT); i++)
+		printf(" ");
+
+	printf("}");
+}
+
+static void
+pretty_print_func_graph(void *data, int size, struct event *event,
+			int cpu, int pid)
+{
+	if (event->flags & EVENT_FL_ISFUNCENT)
+		pretty_print_func_ent(data, size, event, cpu, pid);
+	else if (event->flags & EVENT_FL_ISFUNCRET)
+		pretty_print_func_ret(data, size, event);
+	printf("\n");
+}
+
+void print_trace_event(int cpu, void *data, int size)
+{
+	struct event *event;
+	int type;
+	int pid;
+
+	type = trace_parse_common_type(data);
+
+	event = trace_find_event(type);
+	if (!event) {
+		warning("ug! no event found for type %d", type);
+		return;
+	}
+
+	pid = trace_parse_common_pid(data);
+
+	if (event->flags & (EVENT_FL_ISFUNCENT | EVENT_FL_ISFUNCRET))
+		return pretty_print_func_graph(data, size, event, cpu, pid);
+
+	if (latency_format)
+		print_lat_fmt(data, size);
+
+	if (event->flags & EVENT_FL_FAILED) {
+		printf("EVENT '%s' FAILED TO PARSE\n",
+		       event->name);
+		return;
+	}
+
+	pretty_print(data, size, event);
+}
+
+static void print_fields(struct print_flag_sym *field)
+{
+	printf("{ %s, %s }", field->value, field->str);
+	if (field->next) {
+		printf(", ");
+		print_fields(field->next);
+	}
+}
+
+static void print_args(struct print_arg *args)
+{
+	int print_paren = 1;
+
+	switch (args->type) {
+	case PRINT_NULL:
+		printf("null");
+		break;
+	case PRINT_ATOM:
+		printf("%s", args->atom.atom);
+		break;
+	case PRINT_FIELD:
+		printf("REC->%s", args->field.name);
+		break;
+	case PRINT_FLAGS:
+		printf("__print_flags(");
+		print_args(args->flags.field);
+		printf(", %s, ", args->flags.delim);
+		print_fields(args->flags.flags);
+		printf(")");
+		break;
+	case PRINT_SYMBOL:
+		printf("__print_symbolic(");
+		print_args(args->symbol.field);
+		printf(", ");
+		print_fields(args->symbol.symbols);
+		printf(")");
+		break;
+	case PRINT_STRING:
+		printf("__get_str(%s)", args->string.string);
+		break;
+	case PRINT_TYPE:
+		printf("(%s)", args->typecast.type);
+		print_args(args->typecast.item);
+		break;
+	case PRINT_OP:
+		if (strcmp(args->op.op, ":") == 0)
+			print_paren = 0;
+		if (print_paren)
+			printf("(");
+		print_args(args->op.left);
+		printf(" %s ", args->op.op);
+		print_args(args->op.right);
+		if (print_paren)
+			printf(")");
+		break;
+	default:
+		/* we should warn... */
+		return;
+	}
+	if (args->next) {
+		printf("\n");
+		print_args(args->next);
+	}
+}
+
+int parse_ftrace_file(char *buf, unsigned long size)
+{
+	struct format_field *field;
+	struct print_arg *arg, **list;
+	struct event *event;
+	int ret;
+
+	init_input_buf(buf, size);
+
+	event = alloc_event();
+	if (!event)
+		return -ENOMEM;
+
+	event->flags |= EVENT_FL_ISFTRACE;
+
+	event->name = event_read_name();
+	if (!event->name)
+		die("failed to read ftrace event name");
+
+	if (strcmp(event->name, "function") == 0)
+		event->flags |= EVENT_FL_ISFUNC;
+
+	else if (strcmp(event->name, "funcgraph_entry") == 0)
+		event->flags |= EVENT_FL_ISFUNCENT;
+
+	else if (strcmp(event->name, "funcgraph_exit") == 0)
+		event->flags |= EVENT_FL_ISFUNCRET;
+
+	else if (strcmp(event->name, "bprint") == 0)
+		event->flags |= EVENT_FL_ISBPRINT;
+
+	event->id = event_read_id();
+	if (event->id < 0)
+		die("failed to read ftrace event id");
+
+	add_event(event);
+
+	ret = event_read_format(event);
+	if (ret < 0)
+		die("failed to read ftrace event format");
+
+	ret = event_read_print(event);
+	if (ret < 0)
+		die("failed to read ftrace event print fmt");
+
+	/* New ftrace handles args */
+	if (ret > 0)
+		return 0;
+	/*
+	 * The arguments for ftrace files are parsed by the fields.
+	 * Set up the fields as their arguments.
+	 */
+	list = &event->print_fmt.args;
+	for (field = event->format.fields; field; field = field->next) {
+		arg = malloc_or_die(sizeof(*arg));
+		memset(arg, 0, sizeof(*arg));
+		*list = arg;
+		list = &arg->next;
+		arg->type = PRINT_FIELD;
+		arg->field.name = field->name;
+		arg->field.field = field;
+	}
+	return 0;
+}
+
+int parse_event_file(char *buf, unsigned long size, char *sys)
+{
+	struct event *event;
+	int ret;
+
+	init_input_buf(buf, size);
+
+	event = alloc_event();
+	if (!event)
+		return -ENOMEM;
+
+	event->name = event_read_name();
+	if (!event->name)
+		die("failed to read event name");
+
+	event->id = event_read_id();
+	if (event->id < 0)
+		die("failed to read event id");
+
+	ret = event_read_format(event);
+	if (ret < 0) {
+		warning("failed to read event format for %s", event->name);
+		goto event_failed;
+	}
+
+	ret = event_read_print(event);
+	if (ret < 0) {
+		warning("failed to read event print fmt for %s", event->name);
+		goto event_failed;
+	}
+
+	event->system = strdup(sys);
+
+#define PRINT_ARGS 0
+	if (PRINT_ARGS && event->print_fmt.args)
+		print_args(event->print_fmt.args);
+
+	add_event(event);
+	return 0;
+
+ event_failed:
+	event->flags |= EVENT_FL_FAILED;
+	/* still add it even if it failed */
+	add_event(event);
+	return -1;
+}
+
+void parse_set_info(int nr_cpus, int long_sz)
+{
+	cpus = nr_cpus;
+	long_size = long_sz;
+}
+
+int common_pc(struct scripting_context *context)
+{
+	return parse_common_pc(context->event_data);
+}
+
+int common_flags(struct scripting_context *context)
+{
+	return parse_common_flags(context->event_data);
+}
+
+int common_lock_depth(struct scripting_context *context)
+{
+	return parse_common_lock_depth(context->event_data);
+}
diff --git a/tools/perf/util/trace-event-read.c b/tools/perf/util/trace-event-read.c
new file mode 100644
index 00000000..f55cc3a7
--- /dev/null
+++ b/tools/perf/util/trace-event-read.c
@@ -0,0 +1,539 @@
+/*
+ * Copyright (C) 2009, Steven Rostedt <srostedt@redhat.com>
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License (not later!)
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ */
+#define _FILE_OFFSET_BITS 64
+
+#include <dirent.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <getopt.h>
+#include <stdarg.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/wait.h>
+#include <sys/mman.h>
+#include <pthread.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include <ctype.h>
+#include <errno.h>
+
+#include "../perf.h"
+#include "util.h"
+#include "trace-event.h"
+
+static int input_fd;
+
+static int read_page;
+
+int file_bigendian;
+int host_bigendian;
+static int long_size;
+
+static unsigned long	page_size;
+
+static ssize_t calc_data_size;
+static bool repipe;
+
+static int do_read(int fd, void *buf, int size)
+{
+	int rsize = size;
+
+	while (size) {
+		int ret = read(fd, buf, size);
+
+		if (ret <= 0)
+			return -1;
+
+		if (repipe) {
+			int retw = write(STDOUT_FILENO, buf, ret);
+
+			if (retw <= 0 || retw != ret)
+				die("repiping input file");
+		}
+
+		size -= ret;
+		buf += ret;
+	}
+
+	return rsize;
+}
+
+static int read_or_die(void *data, int size)
+{
+	int r;
+
+	r = do_read(input_fd, data, size);
+	if (r <= 0)
+		die("reading input file (size expected=%d received=%d)",
+		    size, r);
+
+	if (calc_data_size)
+		calc_data_size += r;
+
+	return r;
+}
+
+/* If it fails, the next read will report it */
+static void skip(int size)
+{
+	char buf[BUFSIZ];
+	int r;
+
+	while (size) {
+		r = size > BUFSIZ ? BUFSIZ : size;
+		read_or_die(buf, r);
+		size -= r;
+	};
+}
+
+static unsigned int read4(void)
+{
+	unsigned int data;
+
+	read_or_die(&data, 4);
+	return __data2host4(data);
+}
+
+static unsigned long long read8(void)
+{
+	unsigned long long data;
+
+	read_or_die(&data, 8);
+	return __data2host8(data);
+}
+
+static char *read_string(void)
+{
+	char buf[BUFSIZ];
+	char *str = NULL;
+	int size = 0;
+	off_t r;
+	char c;
+
+	for (;;) {
+		r = read(input_fd, &c, 1);
+		if (r < 0)
+			die("reading input file");
+
+		if (!r)
+			die("no data");
+
+		if (repipe) {
+			int retw = write(STDOUT_FILENO, &c, 1);
+
+			if (retw <= 0 || retw != r)
+				die("repiping input file string");
+		}
+
+		buf[size++] = c;
+
+		if (!c)
+			break;
+	}
+
+	if (calc_data_size)
+		calc_data_size += size;
+
+	str = malloc_or_die(size);
+	memcpy(str, buf, size);
+
+	return str;
+}
+
+static void read_proc_kallsyms(void)
+{
+	unsigned int size;
+	char *buf;
+
+	size = read4();
+	if (!size)
+		return;
+
+	buf = malloc_or_die(size + 1);
+	read_or_die(buf, size);
+	buf[size] = '\0';
+
+	parse_proc_kallsyms(buf, size);
+
+	free(buf);
+}
+
+static void read_ftrace_printk(void)
+{
+	unsigned int size;
+	char *buf;
+
+	size = read4();
+	if (!size)
+		return;
+
+	buf = malloc_or_die(size);
+	read_or_die(buf, size);
+
+	parse_ftrace_printk(buf, size);
+
+	free(buf);
+}
+
+static void read_header_files(void)
+{
+	unsigned long long size;
+	char *header_event;
+	char buf[BUFSIZ];
+
+	read_or_die(buf, 12);
+
+	if (memcmp(buf, "header_page", 12) != 0)
+		die("did not read header page");
+
+	size = read8();
+	skip(size);
+
+	/*
+	 * The size field in the page is of type long,
+	 * use that instead, since it represents the kernel.
+	 */
+	long_size = header_page_size_size;
+
+	read_or_die(buf, 13);
+	if (memcmp(buf, "header_event", 13) != 0)
+		die("did not read header event");
+
+	size = read8();
+	header_event = malloc_or_die(size);
+	read_or_die(header_event, size);
+	free(header_event);
+}
+
+static void read_ftrace_file(unsigned long long size)
+{
+	char *buf;
+
+	buf = malloc_or_die(size);
+	read_or_die(buf, size);
+	parse_ftrace_file(buf, size);
+	free(buf);
+}
+
+static void read_event_file(char *sys, unsigned long long size)
+{
+	char *buf;
+
+	buf = malloc_or_die(size);
+	read_or_die(buf, size);
+	parse_event_file(buf, size, sys);
+	free(buf);
+}
+
+static void read_ftrace_files(void)
+{
+	unsigned long long size;
+	int count;
+	int i;
+
+	count = read4();
+
+	for (i = 0; i < count; i++) {
+		size = read8();
+		read_ftrace_file(size);
+	}
+}
+
+static void read_event_files(void)
+{
+	unsigned long long size;
+	char *sys;
+	int systems;
+	int count;
+	int i,x;
+
+	systems = read4();
+
+	for (i = 0; i < systems; i++) {
+		sys = read_string();
+
+		count = read4();
+		for (x=0; x < count; x++) {
+			size = read8();
+			read_event_file(sys, size);
+		}
+	}
+}
+
+struct cpu_data {
+	unsigned long long	offset;
+	unsigned long long	size;
+	unsigned long long	timestamp;
+	struct record		*next;
+	char			*page;
+	int			cpu;
+	int			index;
+	int			page_size;
+};
+
+static struct cpu_data *cpu_data;
+
+static void update_cpu_data_index(int cpu)
+{
+	cpu_data[cpu].offset += page_size;
+	cpu_data[cpu].size -= page_size;
+	cpu_data[cpu].index = 0;
+}
+
+static void get_next_page(int cpu)
+{
+	off_t save_seek;
+	off_t ret;
+
+	if (!cpu_data[cpu].page)
+		return;
+
+	if (read_page) {
+		if (cpu_data[cpu].size <= page_size) {
+			free(cpu_data[cpu].page);
+			cpu_data[cpu].page = NULL;
+			return;
+		}
+
+		update_cpu_data_index(cpu);
+
+		/* other parts of the code may expect the pointer to not move */
+		save_seek = lseek(input_fd, 0, SEEK_CUR);
+
+		ret = lseek(input_fd, cpu_data[cpu].offset, SEEK_SET);
+		if (ret == (off_t)-1)
+			die("failed to lseek");
+		ret = read(input_fd, cpu_data[cpu].page, page_size);
+		if (ret < 0)
+			die("failed to read page");
+
+		/* reset the file pointer back */
+		lseek(input_fd, save_seek, SEEK_SET);
+
+		return;
+	}
+
+	munmap(cpu_data[cpu].page, page_size);
+	cpu_data[cpu].page = NULL;
+
+	if (cpu_data[cpu].size <= page_size)
+		return;
+
+	update_cpu_data_index(cpu);
+
+	cpu_data[cpu].page = mmap(NULL, page_size, PROT_READ, MAP_PRIVATE,
+				  input_fd, cpu_data[cpu].offset);
+	if (cpu_data[cpu].page == MAP_FAILED)
+		die("failed to mmap cpu %d at offset 0x%llx",
+		    cpu, cpu_data[cpu].offset);
+}
+
+static unsigned int type_len4host(unsigned int type_len_ts)
+{
+	if (file_bigendian)
+		return (type_len_ts >> 27) & ((1 << 5) - 1);
+	else
+		return type_len_ts & ((1 << 5) - 1);
+}
+
+static unsigned int ts4host(unsigned int type_len_ts)
+{
+	if (file_bigendian)
+		return type_len_ts & ((1 << 27) - 1);
+	else
+		return type_len_ts >> 5;
+}
+
+static int calc_index(void *ptr, int cpu)
+{
+	return (unsigned long)ptr - (unsigned long)cpu_data[cpu].page;
+}
+
+struct record *trace_peek_data(int cpu)
+{
+	struct record *data;
+	void *page = cpu_data[cpu].page;
+	int idx = cpu_data[cpu].index;
+	void *ptr = page + idx;
+	unsigned long long extend;
+	unsigned int type_len_ts;
+	unsigned int type_len;
+	unsigned int delta;
+	unsigned int length = 0;
+
+	if (cpu_data[cpu].next)
+		return cpu_data[cpu].next;
+
+	if (!page)
+		return NULL;
+
+	if (!idx) {
+		/* FIXME: handle header page */
+		if (header_page_ts_size != 8)
+			die("expected a long long type for timestamp");
+		cpu_data[cpu].timestamp = data2host8(ptr);
+		ptr += 8;
+		switch (header_page_size_size) {
+		case 4:
+			cpu_data[cpu].page_size = data2host4(ptr);
+			ptr += 4;
+			break;
+		case 8:
+			cpu_data[cpu].page_size = data2host8(ptr);
+			ptr += 8;
+			break;
+		default:
+			die("bad long size");
+		}
+		ptr = cpu_data[cpu].page + header_page_data_offset;
+	}
+
+read_again:
+	idx = calc_index(ptr, cpu);
+
+	if (idx >= cpu_data[cpu].page_size) {
+		get_next_page(cpu);
+		return trace_peek_data(cpu);
+	}
+
+	type_len_ts = data2host4(ptr);
+	ptr += 4;
+
+	type_len = type_len4host(type_len_ts);
+	delta = ts4host(type_len_ts);
+
+	switch (type_len) {
+	case RINGBUF_TYPE_PADDING:
+		if (!delta)
+			die("error, hit unexpected end of page");
+		length = data2host4(ptr);
+		ptr += 4;
+		length *= 4;
+		ptr += length;
+		goto read_again;
+
+	case RINGBUF_TYPE_TIME_EXTEND:
+		extend = data2host4(ptr);
+		ptr += 4;
+		extend <<= TS_SHIFT;
+		extend += delta;
+		cpu_data[cpu].timestamp += extend;
+		goto read_again;
+
+	case RINGBUF_TYPE_TIME_STAMP:
+		ptr += 12;
+		break;
+	case 0:
+		length = data2host4(ptr);
+		ptr += 4;
+		die("here! length=%d", length);
+		break;
+	default:
+		length = type_len * 4;
+		break;
+	}
+
+	cpu_data[cpu].timestamp += delta;
+
+	data = malloc_or_die(sizeof(*data));
+	memset(data, 0, sizeof(*data));
+
+	data->ts = cpu_data[cpu].timestamp;
+	data->size = length;
+	data->data = ptr;
+	ptr += length;
+
+	cpu_data[cpu].index = calc_index(ptr, cpu);
+	cpu_data[cpu].next = data;
+
+	return data;
+}
+
+struct record *trace_read_data(int cpu)
+{
+	struct record *data;
+
+	data = trace_peek_data(cpu);
+	cpu_data[cpu].next = NULL;
+
+	return data;
+}
+
+ssize_t trace_report(int fd, bool __repipe)
+{
+	char buf[BUFSIZ];
+	char test[] = { 23, 8, 68 };
+	char *version;
+	int show_version = 0;
+	int show_funcs = 0;
+	int show_printk = 0;
+	ssize_t size;
+
+	calc_data_size = 1;
+	repipe = __repipe;
+
+	input_fd = fd;
+
+	read_or_die(buf, 3);
+	if (memcmp(buf, test, 3) != 0)
+		die("no trace data in the file");
+
+	read_or_die(buf, 7);
+	if (memcmp(buf, "tracing", 7) != 0)
+		die("not a trace file (missing 'tracing' tag)");
+
+	version = read_string();
+	if (show_version)
+		printf("version = %s\n", version);
+	free(version);
+
+	read_or_die(buf, 1);
+	file_bigendian = buf[0];
+	host_bigendian = bigendian();
+
+	read_or_die(buf, 1);
+	long_size = buf[0];
+
+	page_size = read4();
+
+	read_header_files();
+
+	read_ftrace_files();
+	read_event_files();
+	read_proc_kallsyms();
+	read_ftrace_printk();
+
+	size = calc_data_size - 1;
+	calc_data_size = 0;
+	repipe = false;
+
+	if (show_funcs) {
+		print_funcs();
+		return size;
+	}
+	if (show_printk) {
+		print_printk();
+		return size;
+	}
+
+	return size;
+}
diff --git a/tools/perf/util/trace-event-scripting.c b/tools/perf/util/trace-event-scripting.c
new file mode 100644
index 00000000..c9dcbec7
--- /dev/null
+++ b/tools/perf/util/trace-event-scripting.c
@@ -0,0 +1,167 @@
+/*
+ * trace-event-scripting.  Scripting engine common and initialization code.
+ *
+ * Copyright (C) 2009-2010 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+#include <errno.h>
+
+#include "../perf.h"
+#include "util.h"
+#include "trace-event.h"
+
+struct scripting_context *scripting_context;
+
+static int stop_script_unsupported(void)
+{
+	return 0;
+}
+
+static void process_event_unsupported(union perf_event *event __unused,
+				      struct perf_sample *sample __unused,
+				      struct perf_evsel *evsel __unused,
+				      struct perf_session *session __unused,
+				      struct thread *thread __unused)
+{
+}
+
+static void print_python_unsupported_msg(void)
+{
+	fprintf(stderr, "Python scripting not supported."
+		"  Install libpython and rebuild perf to enable it.\n"
+		"For example:\n  # apt-get install python-dev (ubuntu)"
+		"\n  # yum install python-devel (Fedora)"
+		"\n  etc.\n");
+}
+
+static int python_start_script_unsupported(const char *script __unused,
+					   int argc __unused,
+					   const char **argv __unused)
+{
+	print_python_unsupported_msg();
+
+	return -1;
+}
+
+static int python_generate_script_unsupported(const char *outfile __unused)
+{
+	print_python_unsupported_msg();
+
+	return -1;
+}
+
+struct scripting_ops python_scripting_unsupported_ops = {
+	.name = "Python",
+	.start_script = python_start_script_unsupported,
+	.stop_script = stop_script_unsupported,
+	.process_event = process_event_unsupported,
+	.generate_script = python_generate_script_unsupported,
+};
+
+static void register_python_scripting(struct scripting_ops *scripting_ops)
+{
+	int err;
+	err = script_spec_register("Python", scripting_ops);
+	if (err)
+		die("error registering Python script extension");
+
+	err = script_spec_register("py", scripting_ops);
+	if (err)
+		die("error registering py script extension");
+
+	scripting_context = malloc(sizeof(struct scripting_context));
+}
+
+#ifdef NO_LIBPYTHON
+void setup_python_scripting(void)
+{
+	register_python_scripting(&python_scripting_unsupported_ops);
+}
+#else
+extern struct scripting_ops python_scripting_ops;
+
+void setup_python_scripting(void)
+{
+	register_python_scripting(&python_scripting_ops);
+}
+#endif
+
+static void print_perl_unsupported_msg(void)
+{
+	fprintf(stderr, "Perl scripting not supported."
+		"  Install libperl and rebuild perf to enable it.\n"
+		"For example:\n  # apt-get install libperl-dev (ubuntu)"
+		"\n  # yum install 'perl(ExtUtils::Embed)' (Fedora)"
+		"\n  etc.\n");
+}
+
+static int perl_start_script_unsupported(const char *script __unused,
+					 int argc __unused,
+					 const char **argv __unused)
+{
+	print_perl_unsupported_msg();
+
+	return -1;
+}
+
+static int perl_generate_script_unsupported(const char *outfile __unused)
+{
+	print_perl_unsupported_msg();
+
+	return -1;
+}
+
+struct scripting_ops perl_scripting_unsupported_ops = {
+	.name = "Perl",
+	.start_script = perl_start_script_unsupported,
+	.stop_script = stop_script_unsupported,
+	.process_event = process_event_unsupported,
+	.generate_script = perl_generate_script_unsupported,
+};
+
+static void register_perl_scripting(struct scripting_ops *scripting_ops)
+{
+	int err;
+	err = script_spec_register("Perl", scripting_ops);
+	if (err)
+		die("error registering Perl script extension");
+
+	err = script_spec_register("pl", scripting_ops);
+	if (err)
+		die("error registering pl script extension");
+
+	scripting_context = malloc(sizeof(struct scripting_context));
+}
+
+#ifdef NO_LIBPERL
+void setup_perl_scripting(void)
+{
+	register_perl_scripting(&perl_scripting_unsupported_ops);
+}
+#else
+extern struct scripting_ops perl_scripting_ops;
+
+void setup_perl_scripting(void)
+{
+	register_perl_scripting(&perl_scripting_ops);
+}
+#endif
diff --git a/tools/perf/util/trace-event.h b/tools/perf/util/trace-event.h
new file mode 100644
index 00000000..f674dda3
--- /dev/null
+++ b/tools/perf/util/trace-event.h
@@ -0,0 +1,302 @@
+#ifndef __PERF_TRACE_EVENTS_H
+#define __PERF_TRACE_EVENTS_H
+
+#include <stdbool.h>
+#include "parse-events.h"
+#include "session.h"
+
+#define __unused __attribute__((unused))
+
+
+#ifndef PAGE_MASK
+#define PAGE_MASK (page_size - 1)
+#endif
+
+enum {
+	RINGBUF_TYPE_PADDING		= 29,
+	RINGBUF_TYPE_TIME_EXTEND	= 30,
+	RINGBUF_TYPE_TIME_STAMP		= 31,
+};
+
+#ifndef TS_SHIFT
+#define TS_SHIFT		27
+#endif
+
+#define NSECS_PER_SEC		1000000000ULL
+#define NSECS_PER_USEC		1000ULL
+
+enum format_flags {
+	FIELD_IS_ARRAY		= 1,
+	FIELD_IS_POINTER	= 2,
+	FIELD_IS_SIGNED		= 4,
+	FIELD_IS_STRING		= 8,
+	FIELD_IS_DYNAMIC	= 16,
+	FIELD_IS_FLAG		= 32,
+	FIELD_IS_SYMBOLIC	= 64,
+};
+
+struct format_field {
+	struct format_field	*next;
+	char			*type;
+	char			*name;
+	int			offset;
+	int			size;
+	unsigned long		flags;
+};
+
+struct format {
+	int			nr_common;
+	int			nr_fields;
+	struct format_field	*common_fields;
+	struct format_field	*fields;
+};
+
+struct print_arg_atom {
+	char			*atom;
+};
+
+struct print_arg_string {
+	char			*string;
+	int			offset;
+};
+
+struct print_arg_field {
+	char			*name;
+	struct format_field	*field;
+};
+
+struct print_flag_sym {
+	struct print_flag_sym	*next;
+	char			*value;
+	char			*str;
+};
+
+struct print_arg_typecast {
+	char 			*type;
+	struct print_arg	*item;
+};
+
+struct print_arg_flags {
+	struct print_arg	*field;
+	char			*delim;
+	struct print_flag_sym	*flags;
+};
+
+struct print_arg_symbol {
+	struct print_arg	*field;
+	struct print_flag_sym	*symbols;
+};
+
+struct print_arg;
+
+struct print_arg_op {
+	char			*op;
+	int			prio;
+	struct print_arg	*left;
+	struct print_arg	*right;
+};
+
+struct print_arg_func {
+	char			*name;
+	struct print_arg	*args;
+};
+
+enum print_arg_type {
+	PRINT_NULL,
+	PRINT_ATOM,
+	PRINT_FIELD,
+	PRINT_FLAGS,
+	PRINT_SYMBOL,
+	PRINT_TYPE,
+	PRINT_STRING,
+	PRINT_OP,
+};
+
+struct print_arg {
+	struct print_arg		*next;
+	enum print_arg_type		type;
+	union {
+		struct print_arg_atom		atom;
+		struct print_arg_field		field;
+		struct print_arg_typecast	typecast;
+		struct print_arg_flags		flags;
+		struct print_arg_symbol		symbol;
+		struct print_arg_func		func;
+		struct print_arg_string		string;
+		struct print_arg_op		op;
+	};
+};
+
+struct print_fmt {
+	char			*format;
+	struct print_arg	*args;
+};
+
+struct event {
+	struct event		*next;
+	char			*name;
+	int			id;
+	int			flags;
+	struct format		format;
+	struct print_fmt	print_fmt;
+	char			*system;
+};
+
+enum {
+	EVENT_FL_ISFTRACE	= 0x01,
+	EVENT_FL_ISPRINT	= 0x02,
+	EVENT_FL_ISBPRINT	= 0x04,
+	EVENT_FL_ISFUNC		= 0x08,
+	EVENT_FL_ISFUNCENT	= 0x10,
+	EVENT_FL_ISFUNCRET	= 0x20,
+
+	EVENT_FL_FAILED		= 0x80000000
+};
+
+struct record {
+	unsigned long long ts;
+	int size;
+	void *data;
+};
+
+struct record *trace_peek_data(int cpu);
+struct record *trace_read_data(int cpu);
+
+void parse_set_info(int nr_cpus, int long_sz);
+
+ssize_t trace_report(int fd, bool repipe);
+
+void *malloc_or_die(unsigned int size);
+
+void parse_cmdlines(char *file, int size);
+void parse_proc_kallsyms(char *file, unsigned int size);
+void parse_ftrace_printk(char *file, unsigned int size);
+
+void print_funcs(void);
+void print_printk(void);
+
+int parse_ftrace_file(char *buf, unsigned long size);
+int parse_event_file(char *buf, unsigned long size, char *sys);
+void print_trace_event(int cpu, void *data, int size);
+
+extern int file_bigendian;
+extern int host_bigendian;
+
+int bigendian(void);
+
+static inline unsigned short __data2host2(unsigned short data)
+{
+	unsigned short swap;
+
+	if (host_bigendian == file_bigendian)
+		return data;
+
+	swap = ((data & 0xffULL) << 8) |
+		((data & (0xffULL << 8)) >> 8);
+
+	return swap;
+}
+
+static inline unsigned int __data2host4(unsigned int data)
+{
+	unsigned int swap;
+
+	if (host_bigendian == file_bigendian)
+		return data;
+
+	swap = ((data & 0xffULL) << 24) |
+		((data & (0xffULL << 8)) << 8) |
+		((data & (0xffULL << 16)) >> 8) |
+		((data & (0xffULL << 24)) >> 24);
+
+	return swap;
+}
+
+static inline unsigned long long __data2host8(unsigned long long data)
+{
+	unsigned long long swap;
+
+	if (host_bigendian == file_bigendian)
+		return data;
+
+	swap = ((data & 0xffULL) << 56) |
+		((data & (0xffULL << 8)) << 40) |
+		((data & (0xffULL << 16)) << 24) |
+		((data & (0xffULL << 24)) << 8) |
+		((data & (0xffULL << 32)) >> 8) |
+		((data & (0xffULL << 40)) >> 24) |
+		((data & (0xffULL << 48)) >> 40) |
+		((data & (0xffULL << 56)) >> 56);
+
+	return swap;
+}
+
+#define data2host2(ptr)		__data2host2(*(unsigned short *)ptr)
+#define data2host4(ptr)		__data2host4(*(unsigned int *)ptr)
+#define data2host8(ptr)		({				\
+	unsigned long long __val;				\
+								\
+	memcpy(&__val, (ptr), sizeof(unsigned long long));	\
+	__data2host8(__val);					\
+})
+
+extern int header_page_ts_offset;
+extern int header_page_ts_size;
+extern int header_page_size_offset;
+extern int header_page_size_size;
+extern int header_page_data_offset;
+extern int header_page_data_size;
+
+extern bool latency_format;
+
+int trace_parse_common_type(void *data);
+int trace_parse_common_pid(void *data);
+int parse_common_pc(void *data);
+int parse_common_flags(void *data);
+int parse_common_lock_depth(void *data);
+struct event *trace_find_event(int id);
+struct event *trace_find_next_event(struct event *event);
+unsigned long long read_size(void *ptr, int size);
+unsigned long long
+raw_field_value(struct event *event, const char *name, void *data);
+void *raw_field_ptr(struct event *event, const char *name, void *data);
+unsigned long long eval_flag(const char *flag);
+
+int read_tracing_data(int fd, struct list_head *pattrs);
+ssize_t read_tracing_data_size(int fd, struct list_head *pattrs);
+
+/* taken from kernel/trace/trace.h */
+enum trace_flag_type {
+	TRACE_FLAG_IRQS_OFF		= 0x01,
+	TRACE_FLAG_IRQS_NOSUPPORT	= 0x02,
+	TRACE_FLAG_NEED_RESCHED		= 0x04,
+	TRACE_FLAG_HARDIRQ		= 0x08,
+	TRACE_FLAG_SOFTIRQ		= 0x10,
+};
+
+struct scripting_ops {
+	const char *name;
+	int (*start_script) (const char *script, int argc, const char **argv);
+	int (*stop_script) (void);
+	void (*process_event) (union perf_event *event,
+			       struct perf_sample *sample,
+			       struct perf_evsel *evsel,
+			       struct perf_session *session,
+			       struct thread *thread);
+	int (*generate_script) (const char *outfile);
+};
+
+int script_spec_register(const char *spec, struct scripting_ops *ops);
+
+void setup_perl_scripting(void);
+void setup_python_scripting(void);
+
+struct scripting_context {
+	void *event_data;
+};
+
+int common_pc(struct scripting_context *context);
+int common_flags(struct scripting_context *context);
+int common_lock_depth(struct scripting_context *context);
+
+#endif /* __PERF_TRACE_EVENTS_H */
diff --git a/tools/perf/util/types.h b/tools/perf/util/types.h
new file mode 100644
index 00000000..5f3689a3
--- /dev/null
+++ b/tools/perf/util/types.h
@@ -0,0 +1,19 @@
+#ifndef __PERF_TYPES_H
+#define __PERF_TYPES_H
+
+#include <stdint.h>
+
+/*
+ * We define u64 as uint64_t for every architecture
+ * so that we can print it with "%"PRIx64 without getting warnings.
+ */
+typedef uint64_t	   u64;
+typedef int64_t		   s64;
+typedef unsigned int	   u32;
+typedef signed int	   s32;
+typedef unsigned short	   u16;
+typedef signed short	   s16;
+typedef unsigned char	   u8;
+typedef signed char	   s8;
+
+#endif /* __PERF_TYPES_H */
diff --git a/tools/perf/util/ui/browser.c b/tools/perf/util/ui/browser.c
new file mode 100644
index 00000000..611219f8
--- /dev/null
+++ b/tools/perf/util/ui/browser.c
@@ -0,0 +1,356 @@
+#include "libslang.h"
+#include "ui.h"
+#include <linux/compiler.h>
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include <stdlib.h>
+#include <sys/ttydefaults.h>
+#include "browser.h"
+#include "helpline.h"
+#include "../color.h"
+#include "../util.h"
+#include <stdio.h>
+
+static int ui_browser__percent_color(double percent, bool current)
+{
+	if (current)
+		return HE_COLORSET_SELECTED;
+	if (percent >= MIN_RED)
+		return HE_COLORSET_TOP;
+	if (percent >= MIN_GREEN)
+		return HE_COLORSET_MEDIUM;
+	return HE_COLORSET_NORMAL;
+}
+
+void ui_browser__set_color(struct ui_browser *self __used, int color)
+{
+	SLsmg_set_color(color);
+}
+
+void ui_browser__set_percent_color(struct ui_browser *self,
+				   double percent, bool current)
+{
+	 int color = ui_browser__percent_color(percent, current);
+	 ui_browser__set_color(self, color);
+}
+
+void ui_browser__gotorc(struct ui_browser *self, int y, int x)
+{
+	SLsmg_gotorc(self->y + y, self->x + x);
+}
+
+void ui_browser__list_head_seek(struct ui_browser *self, off_t offset, int whence)
+{
+	struct list_head *head = self->entries;
+	struct list_head *pos;
+
+	switch (whence) {
+	case SEEK_SET:
+		pos = head->next;
+		break;
+	case SEEK_CUR:
+		pos = self->top;
+		break;
+	case SEEK_END:
+		pos = head->prev;
+		break;
+	default:
+		return;
+	}
+
+	if (offset > 0) {
+		while (offset-- != 0)
+			pos = pos->next;
+	} else {
+		while (offset++ != 0)
+			pos = pos->prev;
+	}
+
+	self->top = pos;
+}
+
+void ui_browser__rb_tree_seek(struct ui_browser *self, off_t offset, int whence)
+{
+	struct rb_root *root = self->entries;
+	struct rb_node *nd;
+
+	switch (whence) {
+	case SEEK_SET:
+		nd = rb_first(root);
+		break;
+	case SEEK_CUR:
+		nd = self->top;
+		break;
+	case SEEK_END:
+		nd = rb_last(root);
+		break;
+	default:
+		return;
+	}
+
+	if (offset > 0) {
+		while (offset-- != 0)
+			nd = rb_next(nd);
+	} else {
+		while (offset++ != 0)
+			nd = rb_prev(nd);
+	}
+
+	self->top = nd;
+}
+
+unsigned int ui_browser__rb_tree_refresh(struct ui_browser *self)
+{
+	struct rb_node *nd;
+	int row = 0;
+
+	if (self->top == NULL)
+                self->top = rb_first(self->entries);
+
+	nd = self->top;
+
+	while (nd != NULL) {
+		ui_browser__gotorc(self, row, 0);
+		self->write(self, nd, row);
+		if (++row == self->height)
+			break;
+		nd = rb_next(nd);
+	}
+
+	return row;
+}
+
+bool ui_browser__is_current_entry(struct ui_browser *self, unsigned row)
+{
+	return self->top_idx + row == self->index;
+}
+
+void ui_browser__refresh_dimensions(struct ui_browser *self)
+{
+	int cols, rows;
+	newtGetScreenSize(&cols, &rows);
+
+	self->width = cols - 1;
+	self->height = rows - 2;
+	self->y = 1;
+	self->x = 0;
+}
+
+void ui_browser__reset_index(struct ui_browser *self)
+{
+	self->index = self->top_idx = 0;
+	self->seek(self, 0, SEEK_SET);
+}
+
+void ui_browser__add_exit_key(struct ui_browser *self, int key)
+{
+	newtFormAddHotKey(self->form, key);
+}
+
+void ui_browser__add_exit_keys(struct ui_browser *self, int keys[])
+{
+	int i = 0;
+
+	while (keys[i] && i < 64) {
+		ui_browser__add_exit_key(self, keys[i]);
+		++i;
+	}
+}
+
+void __ui_browser__show_title(struct ui_browser *browser, const char *title)
+{
+	SLsmg_gotorc(0, 0);
+	ui_browser__set_color(browser, NEWT_COLORSET_ROOT);
+	slsmg_write_nstring(title, browser->width);
+}
+
+void ui_browser__show_title(struct ui_browser *browser, const char *title)
+{
+	pthread_mutex_lock(&ui__lock);
+	__ui_browser__show_title(browser, title);
+	pthread_mutex_unlock(&ui__lock);
+}
+
+int ui_browser__show(struct ui_browser *self, const char *title,
+		     const char *helpline, ...)
+{
+	va_list ap;
+	int keys[] = { NEWT_KEY_UP, NEWT_KEY_DOWN, NEWT_KEY_PGUP,
+		       NEWT_KEY_PGDN, NEWT_KEY_HOME, NEWT_KEY_END, ' ',
+		       NEWT_KEY_LEFT, NEWT_KEY_ESCAPE, 'q', CTRL('c'), 0 };
+
+	if (self->form != NULL)
+		newtFormDestroy(self->form);
+
+	ui_browser__refresh_dimensions(self);
+	self->form = newtForm(NULL, NULL, 0);
+	if (self->form == NULL)
+		return -1;
+
+	self->sb = newtVerticalScrollbar(self->width, 1, self->height,
+					 HE_COLORSET_NORMAL,
+					 HE_COLORSET_SELECTED);
+	if (self->sb == NULL)
+		return -1;
+
+	pthread_mutex_lock(&ui__lock);
+	__ui_browser__show_title(self, title);
+
+	ui_browser__add_exit_keys(self, keys);
+	newtFormAddComponent(self->form, self->sb);
+
+	va_start(ap, helpline);
+	ui_helpline__vpush(helpline, ap);
+	va_end(ap);
+	pthread_mutex_unlock(&ui__lock);
+	return 0;
+}
+
+void ui_browser__hide(struct ui_browser *self)
+{
+	pthread_mutex_lock(&ui__lock);
+	newtFormDestroy(self->form);
+	self->form = NULL;
+	ui_helpline__pop();
+	pthread_mutex_unlock(&ui__lock);
+}
+
+int ui_browser__refresh(struct ui_browser *self)
+{
+	int row;
+
+	pthread_mutex_lock(&ui__lock);
+	newtScrollbarSet(self->sb, self->index, self->nr_entries - 1);
+	row = self->refresh(self);
+	ui_browser__set_color(self, HE_COLORSET_NORMAL);
+	SLsmg_fill_region(self->y + row, self->x,
+			  self->height - row, self->width, ' ');
+	pthread_mutex_unlock(&ui__lock);
+
+	return 0;
+}
+
+int ui_browser__run(struct ui_browser *self)
+{
+	struct newtExitStruct es;
+
+	if (ui_browser__refresh(self) < 0)
+		return -1;
+
+	while (1) {
+		off_t offset;
+
+		newtFormRun(self->form, &es);
+
+		if (es.reason != NEWT_EXIT_HOTKEY)
+			break;
+		switch (es.u.key) {
+		case NEWT_KEY_DOWN:
+			if (self->index == self->nr_entries - 1)
+				break;
+			++self->index;
+			if (self->index == self->top_idx + self->height) {
+				++self->top_idx;
+				self->seek(self, +1, SEEK_CUR);
+			}
+			break;
+		case NEWT_KEY_UP:
+			if (self->index == 0)
+				break;
+			--self->index;
+			if (self->index < self->top_idx) {
+				--self->top_idx;
+				self->seek(self, -1, SEEK_CUR);
+			}
+			break;
+		case NEWT_KEY_PGDN:
+		case ' ':
+			if (self->top_idx + self->height > self->nr_entries - 1)
+				break;
+
+			offset = self->height;
+			if (self->index + offset > self->nr_entries - 1)
+				offset = self->nr_entries - 1 - self->index;
+			self->index += offset;
+			self->top_idx += offset;
+			self->seek(self, +offset, SEEK_CUR);
+			break;
+		case NEWT_KEY_PGUP:
+			if (self->top_idx == 0)
+				break;
+
+			if (self->top_idx < self->height)
+				offset = self->top_idx;
+			else
+				offset = self->height;
+
+			self->index -= offset;
+			self->top_idx -= offset;
+			self->seek(self, -offset, SEEK_CUR);
+			break;
+		case NEWT_KEY_HOME:
+			ui_browser__reset_index(self);
+			break;
+		case NEWT_KEY_END:
+			offset = self->height - 1;
+			if (offset >= self->nr_entries)
+				offset = self->nr_entries - 1;
+
+			self->index = self->nr_entries - 1;
+			self->top_idx = self->index - offset;
+			self->seek(self, -offset, SEEK_END);
+			break;
+		default:
+			return es.u.key;
+		}
+		if (ui_browser__refresh(self) < 0)
+			return -1;
+	}
+	return -1;
+}
+
+unsigned int ui_browser__list_head_refresh(struct ui_browser *self)
+{
+	struct list_head *pos;
+	struct list_head *head = self->entries;
+	int row = 0;
+
+	if (self->top == NULL || self->top == self->entries)
+                self->top = head->next;
+
+	pos = self->top;
+
+	list_for_each_from(pos, head) {
+		ui_browser__gotorc(self, row, 0);
+		self->write(self, pos, row);
+		if (++row == self->height)
+			break;
+	}
+
+	return row;
+}
+
+static struct newtPercentTreeColors {
+	const char *topColorFg, *topColorBg;
+	const char *mediumColorFg, *mediumColorBg;
+	const char *normalColorFg, *normalColorBg;
+	const char *selColorFg, *selColorBg;
+	const char *codeColorFg, *codeColorBg;
+} defaultPercentTreeColors = {
+	"red",       "lightgray",
+	"green",     "lightgray",
+	"black",     "lightgray",
+	"lightgray", "magenta",
+	"blue",	     "lightgray",
+};
+
+void ui_browser__init(void)
+{
+	struct newtPercentTreeColors *c = &defaultPercentTreeColors;
+
+	sltt_set_color(HE_COLORSET_TOP, NULL, c->topColorFg, c->topColorBg);
+	sltt_set_color(HE_COLORSET_MEDIUM, NULL, c->mediumColorFg, c->mediumColorBg);
+	sltt_set_color(HE_COLORSET_NORMAL, NULL, c->normalColorFg, c->normalColorBg);
+	sltt_set_color(HE_COLORSET_SELECTED, NULL, c->selColorFg, c->selColorBg);
+	sltt_set_color(HE_COLORSET_CODE, NULL, c->codeColorFg, c->codeColorBg);
+}
diff --git a/tools/perf/util/ui/browser.h b/tools/perf/util/ui/browser.h
new file mode 100644
index 00000000..fc63dda1
--- /dev/null
+++ b/tools/perf/util/ui/browser.h
@@ -0,0 +1,52 @@
+#ifndef _PERF_UI_BROWSER_H_
+#define _PERF_UI_BROWSER_H_ 1
+
+#include <stdbool.h>
+#include <newt.h>
+#include <sys/types.h>
+#include "../types.h"
+
+#define HE_COLORSET_TOP		50
+#define HE_COLORSET_MEDIUM	51
+#define HE_COLORSET_NORMAL	52
+#define HE_COLORSET_SELECTED	53
+#define HE_COLORSET_CODE	54
+
+struct ui_browser {
+	newtComponent form, sb;
+	u64	      index, top_idx;
+	void	      *top, *entries;
+	u16	      y, x, width, height;
+	void	      *priv;
+	unsigned int  (*refresh)(struct ui_browser *self);
+	void	      (*write)(struct ui_browser *self, void *entry, int row);
+	void	      (*seek)(struct ui_browser *self, off_t offset, int whence);
+	u32	      nr_entries;
+};
+
+void ui_browser__set_color(struct ui_browser *self, int color);
+void ui_browser__set_percent_color(struct ui_browser *self,
+				   double percent, bool current);
+bool ui_browser__is_current_entry(struct ui_browser *self, unsigned row);
+void ui_browser__refresh_dimensions(struct ui_browser *self);
+void ui_browser__reset_index(struct ui_browser *self);
+
+void ui_browser__gotorc(struct ui_browser *self, int y, int x);
+void ui_browser__add_exit_key(struct ui_browser *self, int key);
+void ui_browser__add_exit_keys(struct ui_browser *self, int keys[]);
+void __ui_browser__show_title(struct ui_browser *browser, const char *title);
+void ui_browser__show_title(struct ui_browser *browser, const char *title);
+int ui_browser__show(struct ui_browser *self, const char *title,
+		     const char *helpline, ...);
+void ui_browser__hide(struct ui_browser *self);
+int ui_browser__refresh(struct ui_browser *self);
+int ui_browser__run(struct ui_browser *self);
+
+void ui_browser__rb_tree_seek(struct ui_browser *self, off_t offset, int whence);
+unsigned int ui_browser__rb_tree_refresh(struct ui_browser *self);
+
+void ui_browser__list_head_seek(struct ui_browser *self, off_t offset, int whence);
+unsigned int ui_browser__list_head_refresh(struct ui_browser *self);
+
+void ui_browser__init(void);
+#endif /* _PERF_UI_BROWSER_H_ */
diff --git a/tools/perf/util/ui/browsers/annotate.c b/tools/perf/util/ui/browsers/annotate.c
new file mode 100644
index 00000000..0229723a
--- /dev/null
+++ b/tools/perf/util/ui/browsers/annotate.c
@@ -0,0 +1,302 @@
+#include "../browser.h"
+#include "../helpline.h"
+#include "../libslang.h"
+#include "../../annotate.h"
+#include "../../hist.h"
+#include "../../sort.h"
+#include "../../symbol.h"
+#include <pthread.h>
+
+static void ui__error_window(const char *fmt, ...)
+{
+	va_list ap;
+
+	va_start(ap, fmt);
+	newtWinMessagev((char *)"Error", (char *)"Ok", (char *)fmt, ap);
+	va_end(ap);
+}
+
+struct annotate_browser {
+	struct ui_browser b;
+	struct rb_root	  entries;
+	struct rb_node	  *curr_hot;
+};
+
+struct objdump_line_rb_node {
+	struct rb_node	rb_node;
+	double		percent;
+	u32		idx;
+};
+
+static inline
+struct objdump_line_rb_node *objdump_line__rb(struct objdump_line *self)
+{
+	return (struct objdump_line_rb_node *)(self + 1);
+}
+
+static void annotate_browser__write(struct ui_browser *self, void *entry, int row)
+{
+	struct objdump_line *ol = rb_entry(entry, struct objdump_line, node);
+	bool current_entry = ui_browser__is_current_entry(self, row);
+	int width = self->width;
+
+	if (ol->offset != -1) {
+		struct objdump_line_rb_node *olrb = objdump_line__rb(ol);
+		ui_browser__set_percent_color(self, olrb->percent, current_entry);
+		slsmg_printf(" %7.2f ", olrb->percent);
+	} else {
+		ui_browser__set_percent_color(self, 0, current_entry);
+		slsmg_write_nstring(" ", 9);
+	}
+
+	SLsmg_write_char(':');
+	slsmg_write_nstring(" ", 8);
+	if (!*ol->line)
+		slsmg_write_nstring(" ", width - 18);
+	else
+		slsmg_write_nstring(ol->line, width - 18);
+
+	if (!current_entry)
+		ui_browser__set_color(self, HE_COLORSET_CODE);
+}
+
+static double objdump_line__calc_percent(struct objdump_line *self,
+					 struct symbol *sym, int evidx)
+{
+	double percent = 0.0;
+
+	if (self->offset != -1) {
+		int len = sym->end - sym->start;
+		unsigned int hits = 0;
+		struct annotation *notes = symbol__annotation(sym);
+		struct source_line *src_line = notes->src->lines;
+		struct sym_hist *h = annotation__histogram(notes, evidx);
+		s64 offset = self->offset;
+		struct objdump_line *next;
+
+		next = objdump__get_next_ip_line(&notes->src->source, self);
+		while (offset < (s64)len &&
+		       (next == NULL || offset < next->offset)) {
+			if (src_line) {
+				percent += src_line[offset].percent;
+			} else
+				hits += h->addr[offset];
+
+			++offset;
+		}
+		/*
+ 		 * If the percentage wasn't already calculated in
+ 		 * symbol__get_source_line, do it now:
+ 		 */
+		if (src_line == NULL && h->sum)
+			percent = 100.0 * hits / h->sum;
+	}
+
+	return percent;
+}
+
+static void objdump__insert_line(struct rb_root *self,
+				 struct objdump_line_rb_node *line)
+{
+	struct rb_node **p = &self->rb_node;
+	struct rb_node *parent = NULL;
+	struct objdump_line_rb_node *l;
+
+	while (*p != NULL) {
+		parent = *p;
+		l = rb_entry(parent, struct objdump_line_rb_node, rb_node);
+		if (line->percent < l->percent)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+	rb_link_node(&line->rb_node, parent, p);
+	rb_insert_color(&line->rb_node, self);
+}
+
+static void annotate_browser__set_top(struct annotate_browser *self,
+				      struct rb_node *nd)
+{
+	struct objdump_line_rb_node *rbpos;
+	struct objdump_line *pos;
+	unsigned back;
+
+	ui_browser__refresh_dimensions(&self->b);
+	back = self->b.height / 2;
+	rbpos = rb_entry(nd, struct objdump_line_rb_node, rb_node);
+	pos = ((struct objdump_line *)rbpos) - 1;
+	self->b.top_idx = self->b.index = rbpos->idx;
+
+	while (self->b.top_idx != 0 && back != 0) {
+		pos = list_entry(pos->node.prev, struct objdump_line, node);
+
+		--self->b.top_idx;
+		--back;
+	}
+
+	self->b.top = pos;
+	self->curr_hot = nd;
+}
+
+static void annotate_browser__calc_percent(struct annotate_browser *browser,
+					   int evidx)
+{
+	struct symbol *sym = browser->b.priv;
+	struct annotation *notes = symbol__annotation(sym);
+	struct objdump_line *pos;
+
+	browser->entries = RB_ROOT;
+
+	pthread_mutex_lock(&notes->lock);
+
+	list_for_each_entry(pos, &notes->src->source, node) {
+		struct objdump_line_rb_node *rbpos = objdump_line__rb(pos);
+		rbpos->percent = objdump_line__calc_percent(pos, sym, evidx);
+		if (rbpos->percent < 0.01) {
+			RB_CLEAR_NODE(&rbpos->rb_node);
+			continue;
+		}
+		objdump__insert_line(&browser->entries, rbpos);
+	}
+	pthread_mutex_unlock(&notes->lock);
+
+	browser->curr_hot = rb_last(&browser->entries);
+}
+
+static int annotate_browser__run(struct annotate_browser *self, int evidx,
+				 int refresh)
+{
+	struct rb_node *nd = NULL;
+	struct symbol *sym = self->b.priv;
+	/*
+	 * RIGHT To allow builtin-annotate to cycle thru multiple symbols by
+	 * examining the exit key for this function.
+	 */
+	int exit_keys[] = { 'H', NEWT_KEY_TAB, NEWT_KEY_UNTAB,
+			    NEWT_KEY_RIGHT, 0 };
+	int key;
+
+	if (ui_browser__show(&self->b, sym->name,
+			     "<-, -> or ESC: exit, TAB/shift+TAB: "
+			     "cycle hottest lines, H: Hottest") < 0)
+		return -1;
+
+	ui_browser__add_exit_keys(&self->b, exit_keys);
+	annotate_browser__calc_percent(self, evidx);
+
+	if (self->curr_hot)
+		annotate_browser__set_top(self, self->curr_hot);
+
+	nd = self->curr_hot;
+
+	if (refresh != 0)
+		newtFormSetTimer(self->b.form, refresh);
+
+	while (1) {
+		key = ui_browser__run(&self->b);
+
+		if (refresh != 0) {
+			annotate_browser__calc_percent(self, evidx);
+			/*
+			 * Current line focus got out of the list of most active
+			 * lines, NULL it so that if TAB|UNTAB is pressed, we
+			 * move to curr_hot (current hottest line).
+			 */
+			if (nd != NULL && RB_EMPTY_NODE(nd))
+				nd = NULL;
+		}
+
+		switch (key) {
+		case -1:
+			/*
+ 			 * FIXME we need to check if it was
+ 			 * es.reason == NEWT_EXIT_TIMER
+ 			 */
+			if (refresh != 0)
+				symbol__annotate_decay_histogram(sym, evidx);
+			continue;
+		case NEWT_KEY_TAB:
+			if (nd != NULL) {
+				nd = rb_prev(nd);
+				if (nd == NULL)
+					nd = rb_last(&self->entries);
+			} else
+				nd = self->curr_hot;
+			break;
+		case NEWT_KEY_UNTAB:
+			if (nd != NULL)
+				nd = rb_next(nd);
+				if (nd == NULL)
+					nd = rb_first(&self->entries);
+			else
+				nd = self->curr_hot;
+			break;
+		case 'H':
+			nd = self->curr_hot;
+			break;
+		default:
+			goto out;
+		}
+
+		if (nd != NULL)
+			annotate_browser__set_top(self, nd);
+	}
+out:
+	ui_browser__hide(&self->b);
+	return key;
+}
+
+int hist_entry__tui_annotate(struct hist_entry *he, int evidx)
+{
+	return symbol__tui_annotate(he->ms.sym, he->ms.map, evidx, 0);
+}
+
+int symbol__tui_annotate(struct symbol *sym, struct map *map, int evidx,
+			 int refresh)
+{
+	struct objdump_line *pos, *n;
+	struct annotation *notes;
+	struct annotate_browser browser = {
+		.b = {
+			.refresh = ui_browser__list_head_refresh,
+			.seek	 = ui_browser__list_head_seek,
+			.write	 = annotate_browser__write,
+			.priv	 = sym,
+		},
+	};
+	int ret;
+
+	if (sym == NULL)
+		return -1;
+
+	if (map->dso->annotate_warned)
+		return -1;
+
+	if (symbol__annotate(sym, map, sizeof(struct objdump_line_rb_node)) < 0) {
+		ui__error_window(ui_helpline__last_msg);
+		return -1;
+	}
+
+	ui_helpline__push("Press <- or ESC to exit");
+
+	notes = symbol__annotation(sym);
+
+	list_for_each_entry(pos, &notes->src->source, node) {
+		struct objdump_line_rb_node *rbpos;
+		size_t line_len = strlen(pos->line);
+
+		if (browser.b.width < line_len)
+			browser.b.width = line_len;
+		rbpos = objdump_line__rb(pos);
+		rbpos->idx = browser.b.nr_entries++;
+	}
+
+	browser.b.entries = &notes->src->source,
+	browser.b.width += 18; /* Percentage */
+	ret = annotate_browser__run(&browser, evidx, refresh);
+	list_for_each_entry_safe(pos, n, &notes->src->source, node) {
+		list_del(&pos->node);
+		objdump_line__free(pos);
+	}
+	return ret;
+}
diff --git a/tools/perf/util/ui/browsers/hists.c b/tools/perf/util/ui/browsers/hists.c
new file mode 100644
index 00000000..5d767c62
--- /dev/null
+++ b/tools/perf/util/ui/browsers/hists.c
@@ -0,0 +1,1138 @@
+#define _GNU_SOURCE
+#include <stdio.h>
+#undef _GNU_SOURCE
+#include "../libslang.h"
+#include <stdlib.h>
+#include <string.h>
+#include <newt.h>
+#include <linux/rbtree.h>
+
+#include "../../evsel.h"
+#include "../../evlist.h"
+#include "../../hist.h"
+#include "../../pstack.h"
+#include "../../sort.h"
+#include "../../util.h"
+
+#include "../browser.h"
+#include "../helpline.h"
+#include "../util.h"
+#include "map.h"
+
+struct hist_browser {
+	struct ui_browser   b;
+	struct hists	    *hists;
+	struct hist_entry   *he_selection;
+	struct map_symbol   *selection;
+};
+
+static void hist_browser__refresh_dimensions(struct hist_browser *self)
+{
+	/* 3 == +/- toggle symbol before actual hist_entry rendering */
+	self->b.width = 3 + (hists__sort_list_width(self->hists) +
+			     sizeof("[k]"));
+}
+
+static void hist_browser__reset(struct hist_browser *self)
+{
+	self->b.nr_entries = self->hists->nr_entries;
+	hist_browser__refresh_dimensions(self);
+	ui_browser__reset_index(&self->b);
+}
+
+static char tree__folded_sign(bool unfolded)
+{
+	return unfolded ? '-' : '+';
+}
+
+static char map_symbol__folded(const struct map_symbol *self)
+{
+	return self->has_children ? tree__folded_sign(self->unfolded) : ' ';
+}
+
+static char hist_entry__folded(const struct hist_entry *self)
+{
+	return map_symbol__folded(&self->ms);
+}
+
+static char callchain_list__folded(const struct callchain_list *self)
+{
+	return map_symbol__folded(&self->ms);
+}
+
+static void map_symbol__set_folding(struct map_symbol *self, bool unfold)
+{
+	self->unfolded = unfold ? self->has_children : false;
+}
+
+static int callchain_node__count_rows_rb_tree(struct callchain_node *self)
+{
+	int n = 0;
+	struct rb_node *nd;
+
+	for (nd = rb_first(&self->rb_root); nd; nd = rb_next(nd)) {
+		struct callchain_node *child = rb_entry(nd, struct callchain_node, rb_node);
+		struct callchain_list *chain;
+		char folded_sign = ' '; /* No children */
+
+		list_for_each_entry(chain, &child->val, list) {
+			++n;
+			/* We need this because we may not have children */
+			folded_sign = callchain_list__folded(chain);
+			if (folded_sign == '+')
+				break;
+		}
+
+		if (folded_sign == '-') /* Have children and they're unfolded */
+			n += callchain_node__count_rows_rb_tree(child);
+	}
+
+	return n;
+}
+
+static int callchain_node__count_rows(struct callchain_node *node)
+{
+	struct callchain_list *chain;
+	bool unfolded = false;
+	int n = 0;
+
+	list_for_each_entry(chain, &node->val, list) {
+		++n;
+		unfolded = chain->ms.unfolded;
+	}
+
+	if (unfolded)
+		n += callchain_node__count_rows_rb_tree(node);
+
+	return n;
+}
+
+static int callchain__count_rows(struct rb_root *chain)
+{
+	struct rb_node *nd;
+	int n = 0;
+
+	for (nd = rb_first(chain); nd; nd = rb_next(nd)) {
+		struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node);
+		n += callchain_node__count_rows(node);
+	}
+
+	return n;
+}
+
+static bool map_symbol__toggle_fold(struct map_symbol *self)
+{
+	if (!self->has_children)
+		return false;
+
+	self->unfolded = !self->unfolded;
+	return true;
+}
+
+static void callchain_node__init_have_children_rb_tree(struct callchain_node *self)
+{
+	struct rb_node *nd = rb_first(&self->rb_root);
+
+	for (nd = rb_first(&self->rb_root); nd; nd = rb_next(nd)) {
+		struct callchain_node *child = rb_entry(nd, struct callchain_node, rb_node);
+		struct callchain_list *chain;
+		bool first = true;
+
+		list_for_each_entry(chain, &child->val, list) {
+			if (first) {
+				first = false;
+				chain->ms.has_children = chain->list.next != &child->val ||
+							 !RB_EMPTY_ROOT(&child->rb_root);
+			} else
+				chain->ms.has_children = chain->list.next == &child->val &&
+							 !RB_EMPTY_ROOT(&child->rb_root);
+		}
+
+		callchain_node__init_have_children_rb_tree(child);
+	}
+}
+
+static void callchain_node__init_have_children(struct callchain_node *self)
+{
+	struct callchain_list *chain;
+
+	list_for_each_entry(chain, &self->val, list)
+		chain->ms.has_children = !RB_EMPTY_ROOT(&self->rb_root);
+
+	callchain_node__init_have_children_rb_tree(self);
+}
+
+static void callchain__init_have_children(struct rb_root *self)
+{
+	struct rb_node *nd;
+
+	for (nd = rb_first(self); nd; nd = rb_next(nd)) {
+		struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node);
+		callchain_node__init_have_children(node);
+	}
+}
+
+static void hist_entry__init_have_children(struct hist_entry *self)
+{
+	if (!self->init_have_children) {
+		self->ms.has_children = !RB_EMPTY_ROOT(&self->sorted_chain);
+		callchain__init_have_children(&self->sorted_chain);
+		self->init_have_children = true;
+	}
+}
+
+static bool hist_browser__toggle_fold(struct hist_browser *self)
+{
+	if (map_symbol__toggle_fold(self->selection)) {
+		struct hist_entry *he = self->he_selection;
+
+		hist_entry__init_have_children(he);
+		self->hists->nr_entries -= he->nr_rows;
+
+		if (he->ms.unfolded)
+			he->nr_rows = callchain__count_rows(&he->sorted_chain);
+		else
+			he->nr_rows = 0;
+		self->hists->nr_entries += he->nr_rows;
+		self->b.nr_entries = self->hists->nr_entries;
+
+		return true;
+	}
+
+	/* If it doesn't have children, no toggling performed */
+	return false;
+}
+
+static int callchain_node__set_folding_rb_tree(struct callchain_node *self, bool unfold)
+{
+	int n = 0;
+	struct rb_node *nd;
+
+	for (nd = rb_first(&self->rb_root); nd; nd = rb_next(nd)) {
+		struct callchain_node *child = rb_entry(nd, struct callchain_node, rb_node);
+		struct callchain_list *chain;
+		bool has_children = false;
+
+		list_for_each_entry(chain, &child->val, list) {
+			++n;
+			map_symbol__set_folding(&chain->ms, unfold);
+			has_children = chain->ms.has_children;
+		}
+
+		if (has_children)
+			n += callchain_node__set_folding_rb_tree(child, unfold);
+	}
+
+	return n;
+}
+
+static int callchain_node__set_folding(struct callchain_node *node, bool unfold)
+{
+	struct callchain_list *chain;
+	bool has_children = false;
+	int n = 0;
+
+	list_for_each_entry(chain, &node->val, list) {
+		++n;
+		map_symbol__set_folding(&chain->ms, unfold);
+		has_children = chain->ms.has_children;
+	}
+
+	if (has_children)
+		n += callchain_node__set_folding_rb_tree(node, unfold);
+
+	return n;
+}
+
+static int callchain__set_folding(struct rb_root *chain, bool unfold)
+{
+	struct rb_node *nd;
+	int n = 0;
+
+	for (nd = rb_first(chain); nd; nd = rb_next(nd)) {
+		struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node);
+		n += callchain_node__set_folding(node, unfold);
+	}
+
+	return n;
+}
+
+static void hist_entry__set_folding(struct hist_entry *self, bool unfold)
+{
+	hist_entry__init_have_children(self);
+	map_symbol__set_folding(&self->ms, unfold);
+
+	if (self->ms.has_children) {
+		int n = callchain__set_folding(&self->sorted_chain, unfold);
+		self->nr_rows = unfold ? n : 0;
+	} else
+		self->nr_rows = 0;
+}
+
+static void hists__set_folding(struct hists *self, bool unfold)
+{
+	struct rb_node *nd;
+
+	self->nr_entries = 0;
+
+	for (nd = rb_first(&self->entries); nd; nd = rb_next(nd)) {
+		struct hist_entry *he = rb_entry(nd, struct hist_entry, rb_node);
+		hist_entry__set_folding(he, unfold);
+		self->nr_entries += 1 + he->nr_rows;
+	}
+}
+
+static void hist_browser__set_folding(struct hist_browser *self, bool unfold)
+{
+	hists__set_folding(self->hists, unfold);
+	self->b.nr_entries = self->hists->nr_entries;
+	/* Go to the start, we may be way after valid entries after a collapse */
+	ui_browser__reset_index(&self->b);
+}
+
+static int hist_browser__run(struct hist_browser *self, const char *title)
+{
+	int key;
+	int exit_keys[] = { 'a', '?', 'h', 'C', 'd', 'D', 'E', 't',
+			    NEWT_KEY_ENTER, NEWT_KEY_RIGHT, NEWT_KEY_LEFT,
+			    NEWT_KEY_TAB, NEWT_KEY_UNTAB, 0, };
+
+	self->b.entries = &self->hists->entries;
+	self->b.nr_entries = self->hists->nr_entries;
+
+	hist_browser__refresh_dimensions(self);
+
+	if (ui_browser__show(&self->b, title,
+			     "Press '?' for help on key bindings") < 0)
+		return -1;
+
+	ui_browser__add_exit_keys(&self->b, exit_keys);
+
+	while (1) {
+		key = ui_browser__run(&self->b);
+
+		switch (key) {
+		case 'D': { /* Debug */
+			static int seq;
+			struct hist_entry *h = rb_entry(self->b.top,
+							struct hist_entry, rb_node);
+			ui_helpline__pop();
+			ui_helpline__fpush("%d: nr_ent=(%d,%d), height=%d, idx=%d, fve: idx=%d, row_off=%d, nrows=%d",
+					   seq++, self->b.nr_entries,
+					   self->hists->nr_entries,
+					   self->b.height,
+					   self->b.index,
+					   self->b.top_idx,
+					   h->row_offset, h->nr_rows);
+		}
+			break;
+		case 'C':
+			/* Collapse the whole world. */
+			hist_browser__set_folding(self, false);
+			break;
+		case 'E':
+			/* Expand the whole world. */
+			hist_browser__set_folding(self, true);
+			break;
+		case NEWT_KEY_ENTER:
+			if (hist_browser__toggle_fold(self))
+				break;
+			/* fall thru */
+		default:
+			goto out;
+		}
+	}
+out:
+	ui_browser__hide(&self->b);
+	return key;
+}
+
+static char *callchain_list__sym_name(struct callchain_list *self,
+				      char *bf, size_t bfsize)
+{
+	if (self->ms.sym)
+		return self->ms.sym->name;
+
+	snprintf(bf, bfsize, "%#" PRIx64, self->ip);
+	return bf;
+}
+
+#define LEVEL_OFFSET_STEP 3
+
+static int hist_browser__show_callchain_node_rb_tree(struct hist_browser *self,
+						     struct callchain_node *chain_node,
+						     u64 total, int level,
+						     unsigned short row,
+						     off_t *row_offset,
+						     bool *is_current_entry)
+{
+	struct rb_node *node;
+	int first_row = row, width, offset = level * LEVEL_OFFSET_STEP;
+	u64 new_total, remaining;
+
+	if (callchain_param.mode == CHAIN_GRAPH_REL)
+		new_total = chain_node->children_hit;
+	else
+		new_total = total;
+
+	remaining = new_total;
+	node = rb_first(&chain_node->rb_root);
+	while (node) {
+		struct callchain_node *child = rb_entry(node, struct callchain_node, rb_node);
+		struct rb_node *next = rb_next(node);
+		u64 cumul = callchain_cumul_hits(child);
+		struct callchain_list *chain;
+		char folded_sign = ' ';
+		int first = true;
+		int extra_offset = 0;
+
+		remaining -= cumul;
+
+		list_for_each_entry(chain, &child->val, list) {
+			char ipstr[BITS_PER_LONG / 4 + 1], *alloc_str;
+			const char *str;
+			int color;
+			bool was_first = first;
+
+			if (first)
+				first = false;
+			else
+				extra_offset = LEVEL_OFFSET_STEP;
+
+			folded_sign = callchain_list__folded(chain);
+			if (*row_offset != 0) {
+				--*row_offset;
+				goto do_next;
+			}
+
+			alloc_str = NULL;
+			str = callchain_list__sym_name(chain, ipstr, sizeof(ipstr));
+			if (was_first) {
+				double percent = cumul * 100.0 / new_total;
+
+				if (asprintf(&alloc_str, "%2.2f%% %s", percent, str) < 0)
+					str = "Not enough memory!";
+				else
+					str = alloc_str;
+			}
+
+			color = HE_COLORSET_NORMAL;
+			width = self->b.width - (offset + extra_offset + 2);
+			if (ui_browser__is_current_entry(&self->b, row)) {
+				self->selection = &chain->ms;
+				color = HE_COLORSET_SELECTED;
+				*is_current_entry = true;
+			}
+
+			ui_browser__set_color(&self->b, color);
+			ui_browser__gotorc(&self->b, row, 0);
+			slsmg_write_nstring(" ", offset + extra_offset);
+			slsmg_printf("%c ", folded_sign);
+			slsmg_write_nstring(str, width);
+			free(alloc_str);
+
+			if (++row == self->b.height)
+				goto out;
+do_next:
+			if (folded_sign == '+')
+				break;
+		}
+
+		if (folded_sign == '-') {
+			const int new_level = level + (extra_offset ? 2 : 1);
+			row += hist_browser__show_callchain_node_rb_tree(self, child, new_total,
+									 new_level, row, row_offset,
+									 is_current_entry);
+		}
+		if (row == self->b.height)
+			goto out;
+		node = next;
+	}
+out:
+	return row - first_row;
+}
+
+static int hist_browser__show_callchain_node(struct hist_browser *self,
+					     struct callchain_node *node,
+					     int level, unsigned short row,
+					     off_t *row_offset,
+					     bool *is_current_entry)
+{
+	struct callchain_list *chain;
+	int first_row = row,
+	     offset = level * LEVEL_OFFSET_STEP,
+	     width = self->b.width - offset;
+	char folded_sign = ' ';
+
+	list_for_each_entry(chain, &node->val, list) {
+		char ipstr[BITS_PER_LONG / 4 + 1], *s;
+		int color;
+
+		folded_sign = callchain_list__folded(chain);
+
+		if (*row_offset != 0) {
+			--*row_offset;
+			continue;
+		}
+
+		color = HE_COLORSET_NORMAL;
+		if (ui_browser__is_current_entry(&self->b, row)) {
+			self->selection = &chain->ms;
+			color = HE_COLORSET_SELECTED;
+			*is_current_entry = true;
+		}
+
+		s = callchain_list__sym_name(chain, ipstr, sizeof(ipstr));
+		ui_browser__gotorc(&self->b, row, 0);
+		ui_browser__set_color(&self->b, color);
+		slsmg_write_nstring(" ", offset);
+		slsmg_printf("%c ", folded_sign);
+		slsmg_write_nstring(s, width - 2);
+
+		if (++row == self->b.height)
+			goto out;
+	}
+
+	if (folded_sign == '-')
+		row += hist_browser__show_callchain_node_rb_tree(self, node,
+								 self->hists->stats.total_period,
+								 level + 1, row,
+								 row_offset,
+								 is_current_entry);
+out:
+	return row - first_row;
+}
+
+static int hist_browser__show_callchain(struct hist_browser *self,
+					struct rb_root *chain,
+					int level, unsigned short row,
+					off_t *row_offset,
+					bool *is_current_entry)
+{
+	struct rb_node *nd;
+	int first_row = row;
+
+	for (nd = rb_first(chain); nd; nd = rb_next(nd)) {
+		struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node);
+
+		row += hist_browser__show_callchain_node(self, node, level,
+							 row, row_offset,
+							 is_current_entry);
+		if (row == self->b.height)
+			break;
+	}
+
+	return row - first_row;
+}
+
+static int hist_browser__show_entry(struct hist_browser *self,
+				    struct hist_entry *entry,
+				    unsigned short row)
+{
+	char s[256];
+	double percent;
+	int printed = 0;
+	int color, width = self->b.width;
+	char folded_sign = ' ';
+	bool current_entry = ui_browser__is_current_entry(&self->b, row);
+	off_t row_offset = entry->row_offset;
+
+	if (current_entry) {
+		self->he_selection = entry;
+		self->selection = &entry->ms;
+	}
+
+	if (symbol_conf.use_callchain) {
+		hist_entry__init_have_children(entry);
+		folded_sign = hist_entry__folded(entry);
+	}
+
+	if (row_offset == 0) {
+		hist_entry__snprintf(entry, s, sizeof(s), self->hists, NULL, false,
+				     0, false, self->hists->stats.total_period);
+		percent = (entry->period * 100.0) / self->hists->stats.total_period;
+
+		color = HE_COLORSET_SELECTED;
+		if (!current_entry) {
+			if (percent >= MIN_RED)
+				color = HE_COLORSET_TOP;
+			else if (percent >= MIN_GREEN)
+				color = HE_COLORSET_MEDIUM;
+			else
+				color = HE_COLORSET_NORMAL;
+		}
+
+		ui_browser__set_color(&self->b, color);
+		ui_browser__gotorc(&self->b, row, 0);
+		if (symbol_conf.use_callchain) {
+			slsmg_printf("%c ", folded_sign);
+			width -= 2;
+		}
+		slsmg_write_nstring(s, width);
+		++row;
+		++printed;
+	} else
+		--row_offset;
+
+	if (folded_sign == '-' && row != self->b.height) {
+		printed += hist_browser__show_callchain(self, &entry->sorted_chain,
+							1, row, &row_offset,
+							&current_entry);
+		if (current_entry)
+			self->he_selection = entry;
+	}
+
+	return printed;
+}
+
+static unsigned int hist_browser__refresh(struct ui_browser *self)
+{
+	unsigned row = 0;
+	struct rb_node *nd;
+	struct hist_browser *hb = container_of(self, struct hist_browser, b);
+
+	if (self->top == NULL)
+		self->top = rb_first(&hb->hists->entries);
+
+	for (nd = self->top; nd; nd = rb_next(nd)) {
+		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+
+		if (h->filtered)
+			continue;
+
+		row += hist_browser__show_entry(hb, h, row);
+		if (row == self->height)
+			break;
+	}
+
+	return row;
+}
+
+static struct rb_node *hists__filter_entries(struct rb_node *nd)
+{
+	while (nd != NULL) {
+		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+		if (!h->filtered)
+			return nd;
+
+		nd = rb_next(nd);
+	}
+
+	return NULL;
+}
+
+static struct rb_node *hists__filter_prev_entries(struct rb_node *nd)
+{
+	while (nd != NULL) {
+		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+		if (!h->filtered)
+			return nd;
+
+		nd = rb_prev(nd);
+	}
+
+	return NULL;
+}
+
+static void ui_browser__hists_seek(struct ui_browser *self,
+				   off_t offset, int whence)
+{
+	struct hist_entry *h;
+	struct rb_node *nd;
+	bool first = true;
+
+	if (self->nr_entries == 0)
+		return;
+
+	switch (whence) {
+	case SEEK_SET:
+		nd = hists__filter_entries(rb_first(self->entries));
+		break;
+	case SEEK_CUR:
+		nd = self->top;
+		goto do_offset;
+	case SEEK_END:
+		nd = hists__filter_prev_entries(rb_last(self->entries));
+		first = false;
+		break;
+	default:
+		return;
+	}
+
+	/*
+	 * Moves not relative to the first visible entry invalidates its
+	 * row_offset:
+	 */
+	h = rb_entry(self->top, struct hist_entry, rb_node);
+	h->row_offset = 0;
+
+	/*
+	 * Here we have to check if nd is expanded (+), if it is we can't go
+	 * the next top level hist_entry, instead we must compute an offset of
+	 * what _not_ to show and not change the first visible entry.
+	 *
+	 * This offset increments when we are going from top to bottom and
+	 * decreases when we're going from bottom to top.
+	 *
+	 * As we don't have backpointers to the top level in the callchains
+	 * structure, we need to always print the whole hist_entry callchain,
+	 * skipping the first ones that are before the first visible entry
+	 * and stop when we printed enough lines to fill the screen.
+	 */
+do_offset:
+	if (offset > 0) {
+		do {
+			h = rb_entry(nd, struct hist_entry, rb_node);
+			if (h->ms.unfolded) {
+				u16 remaining = h->nr_rows - h->row_offset;
+				if (offset > remaining) {
+					offset -= remaining;
+					h->row_offset = 0;
+				} else {
+					h->row_offset += offset;
+					offset = 0;
+					self->top = nd;
+					break;
+				}
+			}
+			nd = hists__filter_entries(rb_next(nd));
+			if (nd == NULL)
+				break;
+			--offset;
+			self->top = nd;
+		} while (offset != 0);
+	} else if (offset < 0) {
+		while (1) {
+			h = rb_entry(nd, struct hist_entry, rb_node);
+			if (h->ms.unfolded) {
+				if (first) {
+					if (-offset > h->row_offset) {
+						offset += h->row_offset;
+						h->row_offset = 0;
+					} else {
+						h->row_offset += offset;
+						offset = 0;
+						self->top = nd;
+						break;
+					}
+				} else {
+					if (-offset > h->nr_rows) {
+						offset += h->nr_rows;
+						h->row_offset = 0;
+					} else {
+						h->row_offset = h->nr_rows + offset;
+						offset = 0;
+						self->top = nd;
+						break;
+					}
+				}
+			}
+
+			nd = hists__filter_prev_entries(rb_prev(nd));
+			if (nd == NULL)
+				break;
+			++offset;
+			self->top = nd;
+			if (offset == 0) {
+				/*
+				 * Last unfiltered hist_entry, check if it is
+				 * unfolded, if it is then we should have
+				 * row_offset at its last entry.
+				 */
+				h = rb_entry(nd, struct hist_entry, rb_node);
+				if (h->ms.unfolded)
+					h->row_offset = h->nr_rows;
+				break;
+			}
+			first = false;
+		}
+	} else {
+		self->top = nd;
+		h = rb_entry(nd, struct hist_entry, rb_node);
+		h->row_offset = 0;
+	}
+}
+
+static struct hist_browser *hist_browser__new(struct hists *hists)
+{
+	struct hist_browser *self = zalloc(sizeof(*self));
+
+	if (self) {
+		self->hists = hists;
+		self->b.refresh = hist_browser__refresh;
+		self->b.seek = ui_browser__hists_seek;
+	}
+
+	return self;
+}
+
+static void hist_browser__delete(struct hist_browser *self)
+{
+	free(self);
+}
+
+static struct hist_entry *hist_browser__selected_entry(struct hist_browser *self)
+{
+	return self->he_selection;
+}
+
+static struct thread *hist_browser__selected_thread(struct hist_browser *self)
+{
+	return self->he_selection->thread;
+}
+
+static int hists__browser_title(struct hists *self, char *bf, size_t size,
+				const char *ev_name, const struct dso *dso,
+				const struct thread *thread)
+{
+	char unit;
+	int printed;
+	unsigned long nr_events = self->stats.nr_events[PERF_RECORD_SAMPLE];
+
+	nr_events = convert_unit(nr_events, &unit);
+	printed = snprintf(bf, size, "Events: %lu%c %s", nr_events, unit, ev_name);
+
+	if (thread)
+		printed += snprintf(bf + printed, size - printed,
+				    ", Thread: %s(%d)",
+				    (thread->comm_set ? thread->comm : ""),
+				    thread->pid);
+	if (dso)
+		printed += snprintf(bf + printed, size - printed,
+				    ", DSO: %s", dso->short_name);
+	return printed;
+}
+
+static int perf_evsel__hists_browse(struct perf_evsel *evsel,
+				    const char *helpline, const char *ev_name,
+				    bool left_exits)
+{
+	struct hists *self = &evsel->hists;
+	struct hist_browser *browser = hist_browser__new(self);
+	struct pstack *fstack;
+	const struct thread *thread_filter = NULL;
+	const struct dso *dso_filter = NULL;
+	char msg[160];
+	int key = -1;
+
+	if (browser == NULL)
+		return -1;
+
+	fstack = pstack__new(2);
+	if (fstack == NULL)
+		goto out;
+
+	ui_helpline__push(helpline);
+
+	hists__browser_title(self, msg, sizeof(msg), ev_name,
+			     dso_filter, thread_filter);
+	while (1) {
+		const struct thread *thread = NULL;
+		const struct dso *dso = NULL;
+		char *options[16];
+		int nr_options = 0, choice = 0, i,
+		    annotate = -2, zoom_dso = -2, zoom_thread = -2,
+		    browse_map = -2;
+
+		key = hist_browser__run(browser, msg);
+
+		if (browser->he_selection != NULL) {
+			thread = hist_browser__selected_thread(browser);
+			dso = browser->selection->map ? browser->selection->map->dso : NULL;
+		}
+
+		switch (key) {
+		case NEWT_KEY_TAB:
+		case NEWT_KEY_UNTAB:
+			/*
+			 * Exit the browser, let hists__browser_tree
+			 * go to the next or previous
+			 */
+			goto out_free_stack;
+		case 'a':
+			if (browser->selection == NULL ||
+			    browser->selection->sym == NULL ||
+			    browser->selection->map->dso->annotate_warned)
+				continue;
+			goto do_annotate;
+		case 'd':
+			goto zoom_dso;
+		case 't':
+			goto zoom_thread;
+		case NEWT_KEY_F1:
+		case 'h':
+		case '?':
+			ui__help_window("->        Zoom into DSO/Threads & Annotate current symbol\n"
+					"<-        Zoom out\n"
+					"a         Annotate current symbol\n"
+					"h/?/F1    Show this window\n"
+					"C         Collapse all callchains\n"
+					"E         Expand all callchains\n"
+					"d         Zoom into current DSO\n"
+					"t         Zoom into current Thread\n"
+					"TAB/UNTAB Switch events\n"
+					"q/CTRL+C  Exit browser");
+			continue;
+		case NEWT_KEY_ENTER:
+		case NEWT_KEY_RIGHT:
+			/* menu */
+			break;
+		case NEWT_KEY_LEFT: {
+			const void *top;
+
+			if (pstack__empty(fstack)) {
+				/*
+				 * Go back to the perf_evsel_menu__run or other user
+				 */
+				if (left_exits)
+					goto out_free_stack;
+				continue;
+			}
+			top = pstack__pop(fstack);
+			if (top == &dso_filter)
+				goto zoom_out_dso;
+			if (top == &thread_filter)
+				goto zoom_out_thread;
+			continue;
+		}
+		case NEWT_KEY_ESCAPE:
+			if (!left_exits &&
+			    !ui__dialog_yesno("Do you really want to exit?"))
+				continue;
+			/* Fall thru */
+		default:
+			goto out_free_stack;
+		}
+
+		if (browser->selection != NULL &&
+		    browser->selection->sym != NULL &&
+		    !browser->selection->map->dso->annotate_warned &&
+		    asprintf(&options[nr_options], "Annotate %s",
+			     browser->selection->sym->name) > 0)
+			annotate = nr_options++;
+
+		if (thread != NULL &&
+		    asprintf(&options[nr_options], "Zoom %s %s(%d) thread",
+			     (thread_filter ? "out of" : "into"),
+			     (thread->comm_set ? thread->comm : ""),
+			     thread->pid) > 0)
+			zoom_thread = nr_options++;
+
+		if (dso != NULL &&
+		    asprintf(&options[nr_options], "Zoom %s %s DSO",
+			     (dso_filter ? "out of" : "into"),
+			     (dso->kernel ? "the Kernel" : dso->short_name)) > 0)
+			zoom_dso = nr_options++;
+
+		if (browser->selection != NULL &&
+		    browser->selection->map != NULL &&
+		    asprintf(&options[nr_options], "Browse map details") > 0)
+			browse_map = nr_options++;
+
+		options[nr_options++] = (char *)"Exit";
+
+		choice = ui__popup_menu(nr_options, options);
+
+		for (i = 0; i < nr_options - 1; ++i)
+			free(options[i]);
+
+		if (choice == nr_options - 1)
+			break;
+
+		if (choice == -1)
+			continue;
+
+		if (choice == annotate) {
+			struct hist_entry *he;
+do_annotate:
+			he = hist_browser__selected_entry(browser);
+			if (he == NULL)
+				continue;
+
+			hist_entry__tui_annotate(he, evsel->idx);
+		} else if (choice == browse_map)
+			map__browse(browser->selection->map);
+		else if (choice == zoom_dso) {
+zoom_dso:
+			if (dso_filter) {
+				pstack__remove(fstack, &dso_filter);
+zoom_out_dso:
+				ui_helpline__pop();
+				dso_filter = NULL;
+			} else {
+				if (dso == NULL)
+					continue;
+				ui_helpline__fpush("To zoom out press <- or -> + \"Zoom out of %s DSO\"",
+						   dso->kernel ? "the Kernel" : dso->short_name);
+				dso_filter = dso;
+				pstack__push(fstack, &dso_filter);
+			}
+			hists__filter_by_dso(self, dso_filter);
+			hists__browser_title(self, msg, sizeof(msg), ev_name,
+					     dso_filter, thread_filter);
+			hist_browser__reset(browser);
+		} else if (choice == zoom_thread) {
+zoom_thread:
+			if (thread_filter) {
+				pstack__remove(fstack, &thread_filter);
+zoom_out_thread:
+				ui_helpline__pop();
+				thread_filter = NULL;
+			} else {
+				ui_helpline__fpush("To zoom out press <- or -> + \"Zoom out of %s(%d) thread\"",
+						   thread->comm_set ? thread->comm : "",
+						   thread->pid);
+				thread_filter = thread;
+				pstack__push(fstack, &thread_filter);
+			}
+			hists__filter_by_thread(self, thread_filter);
+			hists__browser_title(self, msg, sizeof(msg), ev_name,
+					     dso_filter, thread_filter);
+			hist_browser__reset(browser);
+		}
+	}
+out_free_stack:
+	pstack__delete(fstack);
+out:
+	hist_browser__delete(browser);
+	return key;
+}
+
+struct perf_evsel_menu {
+	struct ui_browser b;
+	struct perf_evsel *selection;
+};
+
+static void perf_evsel_menu__write(struct ui_browser *browser,
+				   void *entry, int row)
+{
+	struct perf_evsel_menu *menu = container_of(browser,
+						    struct perf_evsel_menu, b);
+	struct perf_evsel *evsel = list_entry(entry, struct perf_evsel, node);
+	bool current_entry = ui_browser__is_current_entry(browser, row);
+	unsigned long nr_events = evsel->hists.stats.nr_events[PERF_RECORD_SAMPLE];
+	const char *ev_name = event_name(evsel);
+	char bf[256], unit;
+
+	ui_browser__set_color(browser, current_entry ? HE_COLORSET_SELECTED :
+						       HE_COLORSET_NORMAL);
+
+	nr_events = convert_unit(nr_events, &unit);
+	snprintf(bf, sizeof(bf), "%lu%c%s%s", nr_events,
+		 unit, unit == ' ' ? "" : " ", ev_name);
+	slsmg_write_nstring(bf, browser->width);
+
+	if (current_entry)
+		menu->selection = evsel;
+}
+
+static int perf_evsel_menu__run(struct perf_evsel_menu *menu, const char *help)
+{
+	int exit_keys[] = { NEWT_KEY_ENTER, NEWT_KEY_RIGHT, 0, };
+	struct perf_evlist *evlist = menu->b.priv;
+	struct perf_evsel *pos;
+	const char *ev_name, *title = "Available samples";
+	int key;
+
+	if (ui_browser__show(&menu->b, title,
+			     "ESC: exit, ENTER|->: Browse histograms") < 0)
+		return -1;
+
+	ui_browser__add_exit_keys(&menu->b, exit_keys);
+
+	while (1) {
+		key = ui_browser__run(&menu->b);
+
+		switch (key) {
+		case NEWT_KEY_RIGHT:
+		case NEWT_KEY_ENTER:
+			if (!menu->selection)
+				continue;
+			pos = menu->selection;
+browse_hists:
+			ev_name = event_name(pos);
+			key = perf_evsel__hists_browse(pos, help, ev_name, true);
+			ui_browser__show_title(&menu->b, title);
+			break;
+		case NEWT_KEY_LEFT:
+			continue;
+		case NEWT_KEY_ESCAPE:
+			if (!ui__dialog_yesno("Do you really want to exit?"))
+				continue;
+			/* Fall thru */
+		default:
+			goto out;
+		}
+
+		switch (key) {
+		case NEWT_KEY_TAB:
+			if (pos->node.next == &evlist->entries)
+				pos = list_entry(evlist->entries.next, struct perf_evsel, node);
+			else
+				pos = list_entry(pos->node.next, struct perf_evsel, node);
+			goto browse_hists;
+		case NEWT_KEY_UNTAB:
+			if (pos->node.prev == &evlist->entries)
+				pos = list_entry(evlist->entries.prev, struct perf_evsel, node);
+			else
+				pos = list_entry(pos->node.prev, struct perf_evsel, node);
+			goto browse_hists;
+		case 'q':
+		case CTRL('c'):
+			goto out;
+		default:
+			break;
+		}
+	}
+
+out:
+	ui_browser__hide(&menu->b);
+	return key;
+}
+
+static int __perf_evlist__tui_browse_hists(struct perf_evlist *evlist,
+					   const char *help)
+{
+	struct perf_evsel *pos;
+	struct perf_evsel_menu menu = {
+		.b = {
+			.entries    = &evlist->entries,
+			.refresh    = ui_browser__list_head_refresh,
+			.seek	    = ui_browser__list_head_seek,
+			.write	    = perf_evsel_menu__write,
+			.nr_entries = evlist->nr_entries,
+			.priv	    = evlist,
+		},
+	};
+
+	ui_helpline__push("Press ESC to exit");
+
+	list_for_each_entry(pos, &evlist->entries, node) {
+		const char *ev_name = event_name(pos);
+		size_t line_len = strlen(ev_name) + 7;
+
+		if (menu.b.width < line_len)
+			menu.b.width = line_len;
+		/*
+		 * Cache the evsel name, tracepoints have a _high_ cost per
+		 * event_name() call.
+		 */
+		if (pos->name == NULL)
+			pos->name = strdup(ev_name);
+	}
+
+	return perf_evsel_menu__run(&menu, help);
+}
+
+int perf_evlist__tui_browse_hists(struct perf_evlist *evlist, const char *help)
+{
+
+	if (evlist->nr_entries == 1) {
+		struct perf_evsel *first = list_entry(evlist->entries.next,
+						      struct perf_evsel, node);
+		const char *ev_name = event_name(first);
+		return perf_evsel__hists_browse(first, help, ev_name, false);
+	}
+
+	return __perf_evlist__tui_browse_hists(evlist, help);
+}
diff --git a/tools/perf/util/ui/browsers/map.c b/tools/perf/util/ui/browsers/map.c
new file mode 100644
index 00000000..8462bffe
--- /dev/null
+++ b/tools/perf/util/ui/browsers/map.c
@@ -0,0 +1,156 @@
+#include "../libslang.h"
+#include <elf.h>
+#include <inttypes.h>
+#include <sys/ttydefaults.h>
+#include <ctype.h>
+#include <string.h>
+#include <linux/bitops.h>
+#include "../../debug.h"
+#include "../../symbol.h"
+#include "../browser.h"
+#include "../helpline.h"
+#include "map.h"
+
+static int ui_entry__read(const char *title, char *bf, size_t size, int width)
+{
+	struct newtExitStruct es;
+	newtComponent form, entry;
+	const char *result;
+	int err = -1;
+
+	newtCenteredWindow(width, 1, title);
+	form = newtForm(NULL, NULL, 0);
+	if (form == NULL)
+		return -1;
+
+	entry = newtEntry(0, 0, "0x", width, &result, NEWT_FLAG_SCROLL);
+	if (entry == NULL)
+		goto out_free_form;
+
+	newtFormAddComponent(form, entry);
+	newtFormAddHotKey(form, NEWT_KEY_ENTER);
+	newtFormAddHotKey(form, NEWT_KEY_ESCAPE);
+	newtFormAddHotKey(form, NEWT_KEY_LEFT);
+	newtFormAddHotKey(form, CTRL('c'));
+	newtFormRun(form, &es);
+
+	if (result != NULL) {
+		strncpy(bf, result, size);
+		err = 0;
+	}
+out_free_form:
+	newtPopWindow();
+	newtFormDestroy(form);
+	return err;
+}
+
+struct map_browser {
+	struct ui_browser b;
+	struct map	  *map;
+	u8		  addrlen;
+};
+
+static void map_browser__write(struct ui_browser *self, void *nd, int row)
+{
+	struct symbol *sym = rb_entry(nd, struct symbol, rb_node);
+	struct map_browser *mb = container_of(self, struct map_browser, b);
+	bool current_entry = ui_browser__is_current_entry(self, row);
+	int width;
+
+	ui_browser__set_percent_color(self, 0, current_entry);
+	slsmg_printf("%*" PRIx64 " %*" PRIx64 " %c ",
+		     mb->addrlen, sym->start, mb->addrlen, sym->end,
+		     sym->binding == STB_GLOBAL ? 'g' :
+		     sym->binding == STB_LOCAL  ? 'l' : 'w');
+	width = self->width - ((mb->addrlen * 2) + 4);
+	if (width > 0)
+		slsmg_write_nstring(sym->name, width);
+}
+
+/* FIXME uber-kludgy, see comment on cmd_report... */
+static u32 *symbol__browser_index(struct symbol *self)
+{
+	return ((void *)self) - sizeof(struct rb_node) - sizeof(u32);
+}
+
+static int map_browser__search(struct map_browser *self)
+{
+	char target[512];
+	struct symbol *sym;
+	int err = ui_entry__read("Search by name/addr", target, sizeof(target), 40);
+
+	if (err)
+		return err;
+
+	if (target[0] == '0' && tolower(target[1]) == 'x') {
+		u64 addr = strtoull(target, NULL, 16);
+		sym = map__find_symbol(self->map, addr, NULL);
+	} else
+		sym = map__find_symbol_by_name(self->map, target, NULL);
+
+	if (sym != NULL) {
+		u32 *idx = symbol__browser_index(sym);
+
+		self->b.top = &sym->rb_node;
+		self->b.index = self->b.top_idx = *idx;
+	} else
+		ui_helpline__fpush("%s not found!", target);
+
+	return 0;
+}
+
+static int map_browser__run(struct map_browser *self)
+{
+	int key;
+
+	if (ui_browser__show(&self->b, self->map->dso->long_name,
+			     "Press <- or ESC to exit, %s / to search",
+			     verbose ? "" : "restart with -v to use") < 0)
+		return -1;
+
+	if (verbose)
+		ui_browser__add_exit_key(&self->b, '/');
+
+	while (1) {
+		key = ui_browser__run(&self->b);
+
+		if (verbose && key == '/')
+			map_browser__search(self);
+		else
+			break;
+	}
+
+	ui_browser__hide(&self->b);
+	return key;
+}
+
+int map__browse(struct map *self)
+{
+	struct map_browser mb = {
+		.b = {
+			.entries = &self->dso->symbols[self->type],
+			.refresh = ui_browser__rb_tree_refresh,
+			.seek	 = ui_browser__rb_tree_seek,
+			.write	 = map_browser__write,
+		},
+		.map = self,
+	};
+	struct rb_node *nd;
+	char tmp[BITS_PER_LONG / 4];
+	u64 maxaddr = 0;
+
+	for (nd = rb_first(mb.b.entries); nd; nd = rb_next(nd)) {
+		struct symbol *pos = rb_entry(nd, struct symbol, rb_node);
+
+		if (maxaddr < pos->end)
+			maxaddr = pos->end;
+		if (verbose) {
+			u32 *idx = symbol__browser_index(pos);
+			*idx = mb.b.nr_entries;
+		}
+		++mb.b.nr_entries;
+	}
+
+	mb.addrlen = snprintf(tmp, sizeof(tmp), "%" PRIx64, maxaddr);
+	return map_browser__run(&mb);
+}
diff --git a/tools/perf/util/ui/browsers/map.h b/tools/perf/util/ui/browsers/map.h
new file mode 100644
index 00000000..df8581a4
--- /dev/null
+++ b/tools/perf/util/ui/browsers/map.h
@@ -0,0 +1,6 @@
+#ifndef _PERF_UI_MAP_BROWSER_H_
+#define _PERF_UI_MAP_BROWSER_H_ 1
+struct map;
+
+int map__browse(struct map *self);
+#endif /* _PERF_UI_MAP_BROWSER_H_ */
diff --git a/tools/perf/util/ui/browsers/top.c b/tools/perf/util/ui/browsers/top.c
new file mode 100644
index 00000000..5a065385
--- /dev/null
+++ b/tools/perf/util/ui/browsers/top.c
@@ -0,0 +1,213 @@
+/*
+ * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Parts came from builtin-{top,stat,record}.c, see those files for further
+ * copyright notes.
+ *
+ * Released under the GPL v2. (and only v2, not any later version)
+ */
+#include "../browser.h"
+#include "../../annotate.h"
+#include "../helpline.h"
+#include "../libslang.h"
+#include "../util.h"
+#include "../../evlist.h"
+#include "../../hist.h"
+#include "../../sort.h"
+#include "../../symbol.h"
+#include "../../top.h"
+
+struct perf_top_browser {
+	struct ui_browser b;
+	struct rb_root	  root;
+	struct sym_entry  *selection;
+	float		  sum_ksamples;
+	int		  dso_width;
+	int		  dso_short_width;
+	int		  sym_width;
+};
+
+static void perf_top_browser__write(struct ui_browser *browser, void *entry, int row)
+{
+	struct perf_top_browser *top_browser = container_of(browser, struct perf_top_browser, b);
+	struct sym_entry *syme = rb_entry(entry, struct sym_entry, rb_node);
+	bool current_entry = ui_browser__is_current_entry(browser, row);
+	struct symbol *symbol = sym_entry__symbol(syme);
+	struct perf_top *top = browser->priv;
+	int width = browser->width;
+	double pcnt;
+
+	pcnt = 100.0 - (100.0 * ((top_browser->sum_ksamples - syme->snap_count) /
+				 top_browser->sum_ksamples));
+	ui_browser__set_percent_color(browser, pcnt, current_entry);
+
+	if (top->evlist->nr_entries == 1 || !top->display_weighted) {
+		slsmg_printf("%20.2f ", syme->weight);
+		width -= 24;
+	} else {
+		slsmg_printf("%9.1f %10ld ", syme->weight, syme->snap_count);
+		width -= 23;
+	}
+
+	slsmg_printf("%4.1f%%", pcnt);
+	width -= 7;
+
+	if (verbose) {
+		slsmg_printf(" %016" PRIx64, symbol->start);
+		width -= 17;
+	}
+
+	slsmg_printf(" %-*.*s ", top_browser->sym_width, top_browser->sym_width,
+		     symbol->name);
+	width -= top_browser->sym_width;
+	slsmg_write_nstring(width >= syme->map->dso->long_name_len ?
+				syme->map->dso->long_name :
+				syme->map->dso->short_name, width);
+
+	if (current_entry)
+		top_browser->selection = syme;
+}
+
+static void perf_top_browser__update_rb_tree(struct perf_top_browser *browser)
+{
+	struct perf_top *top = browser->b.priv;
+	u64 top_idx = browser->b.top_idx;
+
+	browser->root = RB_ROOT;
+	browser->b.top = NULL;
+	browser->sum_ksamples = perf_top__decay_samples(top, &browser->root);
+	/*
+ 	 * No active symbols
+ 	 */
+	if (top->rb_entries == 0)
+		return;
+
+	perf_top__find_widths(top, &browser->root, &browser->dso_width,
+			      &browser->dso_short_width,
+                              &browser->sym_width);
+	if (browser->sym_width + browser->dso_width > browser->b.width - 29) {
+		browser->dso_width = browser->dso_short_width;
+		if (browser->sym_width + browser->dso_width > browser->b.width - 29)
+			browser->sym_width = browser->b.width - browser->dso_width - 29;
+	}
+
+	/*
+	 * Adjust the ui_browser indexes since the entries in the browser->root
+	 * rb_tree may have changed, then seek it from start, so that we get a
+	 * possible new top of the screen.
+ 	 */
+	browser->b.nr_entries = top->rb_entries;
+
+	if (top_idx >= browser->b.nr_entries) {
+		if (browser->b.height >= browser->b.nr_entries)
+			top_idx = browser->b.nr_entries - browser->b.height;
+		else
+			top_idx = 0;
+	}
+
+	if (browser->b.index >= top_idx + browser->b.height)
+		browser->b.index = top_idx + browser->b.index - browser->b.top_idx;
+
+	if (browser->b.index >= browser->b.nr_entries)
+		browser->b.index = browser->b.nr_entries - 1;
+
+	browser->b.top_idx = top_idx;
+	browser->b.seek(&browser->b, top_idx, SEEK_SET);
+}
+
+static void perf_top_browser__annotate(struct perf_top_browser *browser)
+{
+	struct sym_entry *syme = browser->selection;
+	struct symbol *sym = sym_entry__symbol(syme);
+	struct annotation *notes = symbol__annotation(sym);
+	struct perf_top *top = browser->b.priv;
+
+	if (notes->src != NULL)
+		goto do_annotation;
+
+	pthread_mutex_lock(&notes->lock);
+
+	top->sym_filter_entry = NULL;
+
+	if (symbol__alloc_hist(sym, top->evlist->nr_entries) < 0) {
+		pr_err("Not enough memory for annotating '%s' symbol!\n",
+		       sym->name);
+		pthread_mutex_unlock(&notes->lock);
+		return;
+	}
+
+	top->sym_filter_entry = syme;
+
+	pthread_mutex_unlock(&notes->lock);
+do_annotation:
+	symbol__tui_annotate(sym, syme->map, 0, top->delay_secs * 1000);
+}
+
+static int perf_top_browser__run(struct perf_top_browser *browser)
+{
+	int key;
+	char title[160];
+	struct perf_top *top = browser->b.priv;
+	int delay_msecs = top->delay_secs * 1000;
+	int exit_keys[] = { 'a', NEWT_KEY_ENTER, NEWT_KEY_RIGHT, 0, };
+
+	perf_top_browser__update_rb_tree(browser);
+        perf_top__header_snprintf(top, title, sizeof(title));
+        perf_top__reset_sample_counters(top);
+
+	if (ui_browser__show(&browser->b, title,
+			     "ESC: exit, ENTER|->|a: Live Annotate") < 0)
+		return -1;
+
+	newtFormSetTimer(browser->b.form, delay_msecs);
+	ui_browser__add_exit_keys(&browser->b, exit_keys);
+
+	while (1) {
+		key = ui_browser__run(&browser->b);
+
+		switch (key) {
+		case -1:
+			/* FIXME we need to check if it was es.reason == NEWT_EXIT_TIMER */
+			perf_top_browser__update_rb_tree(browser);
+			perf_top__header_snprintf(top, title, sizeof(title));
+			perf_top__reset_sample_counters(top);
+			ui_browser__set_color(&browser->b, NEWT_COLORSET_ROOT);
+			SLsmg_gotorc(0, 0);
+			slsmg_write_nstring(title, browser->b.width);
+			break;
+		case 'a':
+		case NEWT_KEY_RIGHT:
+		case NEWT_KEY_ENTER:
+			if (browser->selection)
+				perf_top_browser__annotate(browser);
+			break;
+		case NEWT_KEY_LEFT:
+			continue;
+		case NEWT_KEY_ESCAPE:
+			if (!ui__dialog_yesno("Do you really want to exit?"))
+				continue;
+			/* Fall thru */
+		default:
+			goto out;
+		}
+	}
+out:
+	ui_browser__hide(&browser->b);
+	return key;
+}
+
+int perf_top__tui_browser(struct perf_top *top)
+{
+	struct perf_top_browser browser = {
+		.b = {
+			.entries = &browser.root,
+			.refresh = ui_browser__rb_tree_refresh,
+			.seek	 = ui_browser__rb_tree_seek,
+			.write	 = perf_top_browser__write,
+			.priv	 = top,
+		},
+	};
+
+	ui_helpline__push("Press <- or ESC to exit");
+	return perf_top_browser__run(&browser);
+}
diff --git a/tools/perf/util/ui/helpline.c b/tools/perf/util/ui/helpline.c
new file mode 100644
index 00000000..f36d2ff5
--- /dev/null
+++ b/tools/perf/util/ui/helpline.c
@@ -0,0 +1,72 @@
+#define _GNU_SOURCE
+#include <stdio.h>
+#include <stdlib.h>
+#include <newt.h>
+
+#include "../debug.h"
+#include "helpline.h"
+#include "ui.h"
+
+void ui_helpline__pop(void)
+{
+	newtPopHelpLine();
+}
+
+void ui_helpline__push(const char *msg)
+{
+	newtPushHelpLine(msg);
+}
+
+void ui_helpline__vpush(const char *fmt, va_list ap)
+{
+	char *s;
+
+	if (vasprintf(&s, fmt, ap) < 0)
+		vfprintf(stderr, fmt, ap);
+	else {
+		ui_helpline__push(s);
+		free(s);
+	}
+}
+
+void ui_helpline__fpush(const char *fmt, ...)
+{
+	va_list ap;
+
+	va_start(ap, fmt);
+	ui_helpline__vpush(fmt, ap);
+	va_end(ap);
+}
+
+void ui_helpline__puts(const char *msg)
+{
+	ui_helpline__pop();
+	ui_helpline__push(msg);
+}
+
+void ui_helpline__init(void)
+{
+	ui_helpline__puts(" ");
+}
+
+char ui_helpline__last_msg[1024];
+
+int ui_helpline__show_help(const char *format, va_list ap)
+{
+	int ret;
+	static int backlog;
+
+	pthread_mutex_lock(&ui__lock);
+	ret = vsnprintf(ui_helpline__last_msg + backlog,
+			sizeof(ui_helpline__last_msg) - backlog, format, ap);
+	backlog += ret;
+
+	if (ui_helpline__last_msg[backlog - 1] == '\n') {
+		ui_helpline__puts(ui_helpline__last_msg);
+		newtRefresh();
+		backlog = 0;
+	}
+	pthread_mutex_unlock(&ui__lock);
+
+	return ret;
+}
diff --git a/tools/perf/util/ui/helpline.h b/tools/perf/util/ui/helpline.h
new file mode 100644
index 00000000..ab6028d0
--- /dev/null
+++ b/tools/perf/util/ui/helpline.h
@@ -0,0 +1,11 @@
+#ifndef _PERF_UI_HELPLINE_H_
+#define _PERF_UI_HELPLINE_H_ 1
+
+void ui_helpline__init(void);
+void ui_helpline__pop(void);
+void ui_helpline__push(const char *msg);
+void ui_helpline__vpush(const char *fmt, va_list ap);
+void ui_helpline__fpush(const char *fmt, ...);
+void ui_helpline__puts(const char *msg);
+
+#endif /* _PERF_UI_HELPLINE_H_ */
diff --git a/tools/perf/util/ui/libslang.h b/tools/perf/util/ui/libslang.h
new file mode 100644
index 00000000..2b63e1c9
--- /dev/null
+++ b/tools/perf/util/ui/libslang.h
@@ -0,0 +1,27 @@
+#ifndef _PERF_UI_SLANG_H_
+#define _PERF_UI_SLANG_H_ 1
+/*
+ * slang versions <= 2.0.6 have a "#if HAVE_LONG_LONG" that breaks
+ * the build if it isn't defined. Use the equivalent one that glibc
+ * has on features.h.
+ */
+#include <features.h>
+#ifndef HAVE_LONG_LONG
+#define HAVE_LONG_LONG __GLIBC_HAVE_LONG_LONG
+#endif
+#include <slang.h>
+
+#if SLANG_VERSION < 20104
+#define slsmg_printf(msg, args...) \
+	SLsmg_printf((char *)(msg), ##args)
+#define slsmg_write_nstring(msg, len) \
+	SLsmg_write_nstring((char *)(msg), len)
+#define sltt_set_color(obj, name, fg, bg) \
+	SLtt_set_color(obj,(char *)(name), (char *)(fg), (char *)(bg))
+#else
+#define slsmg_printf SLsmg_printf
+#define slsmg_write_nstring SLsmg_write_nstring
+#define sltt_set_color SLtt_set_color
+#endif
+
+#endif /* _PERF_UI_SLANG_H_ */
diff --git a/tools/perf/util/ui/progress.c b/tools/perf/util/ui/progress.c
new file mode 100644
index 00000000..d7fc399d
--- /dev/null
+++ b/tools/perf/util/ui/progress.c
@@ -0,0 +1,60 @@
+#include <stdlib.h>
+#include <newt.h>
+#include "../cache.h"
+#include "progress.h"
+
+struct ui_progress {
+	newtComponent form, scale;
+};
+
+struct ui_progress *ui_progress__new(const char *title, u64 total)
+{
+	struct ui_progress *self = malloc(sizeof(*self));
+
+	if (self != NULL) {
+		int cols;
+
+		if (use_browser <= 0)
+			return self;
+		newtGetScreenSize(&cols, NULL);
+		cols -= 4;
+		newtCenteredWindow(cols, 1, title);
+		self->form  = newtForm(NULL, NULL, 0);
+		if (self->form == NULL)
+			goto out_free_self;
+		self->scale = newtScale(0, 0, cols, total);
+		if (self->scale == NULL)
+			goto out_free_form;
+		newtFormAddComponent(self->form, self->scale);
+		newtRefresh();
+	}
+
+	return self;
+
+out_free_form:
+	newtFormDestroy(self->form);
+out_free_self:
+	free(self);
+	return NULL;
+}
+
+void ui_progress__update(struct ui_progress *self, u64 curr)
+{
+	/*
+	 * FIXME: We should have a per UI backend way of showing progress,
+	 * stdio will just show a percentage as NN%, etc.
+	 */
+	if (use_browser <= 0)
+		return;
+	newtScaleSet(self->scale, curr);
+	newtRefresh();
+}
+
+void ui_progress__delete(struct ui_progress *self)
+{
+	if (use_browser > 0) {
+		newtFormDestroy(self->form);
+		newtPopWindow();
+	}
+	free(self);
+}
diff --git a/tools/perf/util/ui/progress.h b/tools/perf/util/ui/progress.h
new file mode 100644
index 00000000..a3820a0b
--- /dev/null
+++ b/tools/perf/util/ui/progress.h
@@ -0,0 +1,11 @@
+#ifndef _PERF_UI_PROGRESS_H_
+#define _PERF_UI_PROGRESS_H_ 1
+
+struct ui_progress;
+
+struct ui_progress *ui_progress__new(const char *title, u64 total);
+void ui_progress__delete(struct ui_progress *self);
+
+void ui_progress__update(struct ui_progress *self, u64 curr);
+
+#endif
diff --git a/tools/perf/util/ui/setup.c b/tools/perf/util/ui/setup.c
new file mode 100644
index 00000000..ee46d671
--- /dev/null
+++ b/tools/perf/util/ui/setup.c
@@ -0,0 +1,46 @@
+#include <newt.h>
+#include <signal.h>
+#include <stdbool.h>
+
+#include "../cache.h"
+#include "../debug.h"
+#include "browser.h"
+#include "helpline.h"
+#include "ui.h"
+
+pthread_mutex_t ui__lock = PTHREAD_MUTEX_INITIALIZER;
+
+static void newt_suspend(void *d __used)
+{
+	newtSuspend();
+	raise(SIGTSTP);
+	newtResume();
+}
+
+void setup_browser(bool fallback_to_pager)
+{
+	if (!isatty(1) || !use_browser || dump_trace) {
+		use_browser = 0;
+		if (fallback_to_pager)
+			setup_pager();
+		return;
+	}
+
+	use_browser = 1;
+	newtInit();
+	newtCls();
+	newtSetSuspendCallback(newt_suspend, NULL);
+	ui_helpline__init();
+	ui_browser__init();
+}
+
+void exit_browser(bool wait_for_ok)
+{
+	if (use_browser > 0) {
+		if (wait_for_ok) {
+			char title[] = "Fatal Error", ok[] = "Ok";
+			newtWinMessage(title, ok, ui_helpline__last_msg);
+		}
+		newtFinished();
+	}
+}
diff --git a/tools/perf/util/ui/ui.h b/tools/perf/util/ui/ui.h
new file mode 100644
index 00000000..d264e059
--- /dev/null
+++ b/tools/perf/util/ui/ui.h
@@ -0,0 +1,8 @@
+#ifndef _PERF_UI_H_
+#define _PERF_UI_H_ 1
+
+#include <pthread.h>
+
+extern pthread_mutex_t ui__lock;
+
+#endif /* _PERF_UI_H_ */
diff --git a/tools/perf/util/ui/util.c b/tools/perf/util/ui/util.c
new file mode 100644
index 00000000..fdf1fc8f
--- /dev/null
+++ b/tools/perf/util/ui/util.c
@@ -0,0 +1,130 @@
+#include <newt.h>
+#include <signal.h>
+#include <stdio.h>
+#include <stdbool.h>
+#include <string.h>
+#include <sys/ttydefaults.h>
+
+#include "../cache.h"
+#include "../debug.h"
+#include "browser.h"
+#include "helpline.h"
+#include "ui.h"
+#include "util.h"
+
+static void newt_form__set_exit_keys(newtComponent self)
+{
+	newtFormAddHotKey(self, NEWT_KEY_LEFT);
+	newtFormAddHotKey(self, NEWT_KEY_ESCAPE);
+	newtFormAddHotKey(self, 'Q');
+	newtFormAddHotKey(self, 'q');
+	newtFormAddHotKey(self, CTRL('c'));
+}
+
+static newtComponent newt_form__new(void)
+{
+	newtComponent self = newtForm(NULL, NULL, 0);
+	if (self)
+		newt_form__set_exit_keys(self);
+	return self;
+}
+
+int ui__popup_menu(int argc, char * const argv[])
+{
+	struct newtExitStruct es;
+	int i, rc = -1, max_len = 5;
+	newtComponent listbox, form = newt_form__new();
+
+	if (form == NULL)
+		return -1;
+
+	listbox = newtListbox(0, 0, argc, NEWT_FLAG_RETURNEXIT);
+	if (listbox == NULL)
+		goto out_destroy_form;
+
+	newtFormAddComponent(form, listbox);
+
+	for (i = 0; i < argc; ++i) {
+		int len = strlen(argv[i]);
+		if (len > max_len)
+			max_len = len;
+		if (newtListboxAddEntry(listbox, argv[i], (void *)(long)i))
+			goto out_destroy_form;
+	}
+
+	newtCenteredWindow(max_len, argc, NULL);
+	newtFormRun(form, &es);
+	rc = newtListboxGetCurrent(listbox) - NULL;
+	if (es.reason == NEWT_EXIT_HOTKEY)
+		rc = -1;
+	newtPopWindow();
+out_destroy_form:
+	newtFormDestroy(form);
+	return rc;
+}
+
+int ui__help_window(const char *text)
+{
+	struct newtExitStruct es;
+	newtComponent tb, form = newt_form__new();
+	int rc = -1;
+	int max_len = 0, nr_lines = 0;
+	const char *t;
+
+	if (form == NULL)
+		return -1;
+
+	t = text;
+	while (1) {
+		const char *sep = strchr(t, '\n');
+		int len;
+
+		if (sep == NULL)
+			sep = strchr(t, '\0');
+		len = sep - t;
+		if (max_len < len)
+			max_len = len;
+		++nr_lines;
+		if (*sep == '\0')
+			break;
+		t = sep + 1;
+	}
+
+	tb = newtTextbox(0, 0, max_len, nr_lines, 0);
+	if (tb == NULL)
+		goto out_destroy_form;
+
+	newtTextboxSetText(tb, text);
+	newtFormAddComponent(form, tb);
+	newtCenteredWindow(max_len, nr_lines, NULL);
+	newtFormRun(form, &es);
+	newtPopWindow();
+	rc = 0;
+out_destroy_form:
+	newtFormDestroy(form);
+	return rc;
+}
+
+static const char yes[] = "Yes", no[] = "No",
+		  warning_str[] = "Warning!", ok[] = "Ok";
+
+bool ui__dialog_yesno(const char *msg)
+{
+	/* newtWinChoice should really be accepting const char pointers... */
+	return newtWinChoice(NULL, (char *)yes, (char *)no, (char *)msg) == 1;
+}
+
+void ui__warning(const char *format, ...)
+{
+	va_list args;
+
+	va_start(args, format);
+	if (use_browser > 0) {
+		pthread_mutex_lock(&ui__lock);
+		newtWinMessagev((char *)warning_str, (char *)ok,
+				(char *)format, args);
+		pthread_mutex_unlock(&ui__lock);
+	} else
+		vfprintf(stderr, format, args);
+	va_end(args);
+}
diff --git a/tools/perf/util/ui/util.h b/tools/perf/util/ui/util.h
new file mode 100644
index 00000000..afcbc1d9
--- /dev/null
+++ b/tools/perf/util/ui/util.h
@@ -0,0 +1,10 @@
+#ifndef _PERF_UI_UTIL_H_
+#define _PERF_UI_UTIL_H_ 1
+
+#include <stdbool.h>
+
+int ui__popup_menu(int argc, char * const argv[]);
+int ui__help_window(const char *text);
+bool ui__dialog_yesno(const char *msg);
+
+#endif /* _PERF_UI_UTIL_H_ */
diff --git a/tools/perf/util/usage.c b/tools/perf/util/usage.c
new file mode 100644
index 00000000..e16bf9a7
--- /dev/null
+++ b/tools/perf/util/usage.c
@@ -0,0 +1,80 @@
+/*
+ * GIT - The information manager from hell
+ *
+ * Copyright (C) Linus Torvalds, 2005
+ */
+#include "util.h"
+
+static void report(const char *prefix, const char *err, va_list params)
+{
+	char msg[1024];
+	vsnprintf(msg, sizeof(msg), err, params);
+	fprintf(stderr, " %s%s\n", prefix, msg);
+}
+
+static NORETURN void usage_builtin(const char *err)
+{
+	fprintf(stderr, "\n Usage: %s\n", err);
+	exit(129);
+}
+
+static NORETURN void die_builtin(const char *err, va_list params)
+{
+	report(" Fatal: ", err, params);
+	exit(128);
+}
+
+static void error_builtin(const char *err, va_list params)
+{
+	report(" Error: ", err, params);
+}
+
+static void warn_builtin(const char *warn, va_list params)
+{
+	report(" Warning: ", warn, params);
+}
+
+/* If we are in a dlopen()ed .so write to a global variable would segfault
+ * (ugh), so keep things static. */
+static void (*usage_routine)(const char *err) NORETURN = usage_builtin;
+static void (*die_routine)(const char *err, va_list params) NORETURN = die_builtin;
+static void (*error_routine)(const char *err, va_list params) = error_builtin;
+static void (*warn_routine)(const char *err, va_list params) = warn_builtin;
+
+void set_die_routine(void (*routine)(const char *err, va_list params) NORETURN)
+{
+	die_routine = routine;
+}
+
+void usage(const char *err)
+{
+	usage_routine(err);
+}
+
+void die(const char *err, ...)
+{
+	va_list params;
+
+	va_start(params, err);
+	die_routine(err, params);
+	va_end(params);
+}
+
+int error(const char *err, ...)
+{
+	va_list params;
+
+	va_start(params, err);
+	error_routine(err, params);
+	va_end(params);
+	return -1;
+}
+
+void warning(const char *warn, ...)
+{
+	va_list params;
+
+	va_start(params, warn);
+	warn_routine(warn, params);
+	va_end(params);
+}
diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c
new file mode 100644
index 00000000..5b3ea49a
--- /dev/null
+++ b/tools/perf/util/util.c
@@ -0,0 +1,133 @@
+#include "util.h"
+#include <sys/mman.h>
+
+int mkdir_p(char *path, mode_t mode)
+{
+	struct stat st;
+	int err;
+	char *d = path;
+
+	if (*d != '/')
+		return -1;
+
+	if (stat(path, &st) == 0)
+		return 0;
+
+	while (*++d == '/');
+
+	while ((d = strchr(d, '/'))) {
+		*d = '\0';
+		err = stat(path, &st) && mkdir(path, mode);
+		*d++ = '/';
+		if (err)
+			return -1;
+		while (*d == '/')
+			++d;
+	}
+	return (stat(path, &st) && mkdir(path, mode)) ? -1 : 0;
+}
+
+static int slow_copyfile(const char *from, const char *to)
+{
+	int err = 0;
+	char *line = NULL;
+	size_t n;
+	FILE *from_fp = fopen(from, "r"), *to_fp;
+
+	if (from_fp == NULL)
+		goto out;
+
+	to_fp = fopen(to, "w");
+	if (to_fp == NULL)
+		goto out_fclose_from;
+
+	while (getline(&line, &n, from_fp) > 0)
+		if (fputs(line, to_fp) == EOF)
+			goto out_fclose_to;
+	err = 0;
+out_fclose_to:
+	fclose(to_fp);
+	free(line);
+out_fclose_from:
+	fclose(from_fp);
+out:
+	return err;
+}
+
+int copyfile(const char *from, const char *to)
+{
+	int fromfd, tofd;
+	struct stat st;
+	void *addr;
+	int err = -1;
+
+	if (stat(from, &st))
+		goto out;
+
+	if (st.st_size == 0) /* /proc? do it slowly... */
+		return slow_copyfile(from, to);
+
+	fromfd = open(from, O_RDONLY);
+	if (fromfd < 0)
+		goto out;
+
+	tofd = creat(to, 0755);
+	if (tofd < 0)
+		goto out_close_from;
+
+	addr = mmap(NULL, st.st_size, PROT_READ, MAP_PRIVATE, fromfd, 0);
+	if (addr == MAP_FAILED)
+		goto out_close_to;
+
+	if (write(tofd, addr, st.st_size) == st.st_size)
+		err = 0;
+
+	munmap(addr, st.st_size);
+out_close_to:
+	close(tofd);
+	if (err)
+		unlink(to);
+out_close_from:
+	close(fromfd);
+out:
+	return err;
+}
+
+unsigned long convert_unit(unsigned long value, char *unit)
+{
+	*unit = ' ';
+
+	if (value > 1000) {
+		value /= 1000;
+		*unit = 'K';
+	}
+
+	if (value > 1000) {
+		value /= 1000;
+		*unit = 'M';
+	}
+
+	if (value > 1000) {
+		value /= 1000;
+		*unit = 'G';
+	}
+
+	return value;
+}
+
+int readn(int fd, void *buf, size_t n)
+{
+	void *buf_start = buf;
+
+	while (n) {
+		int ret = read(fd, buf, n);
+
+		if (ret <= 0)
+			return ret;
+
+		n -= ret;
+		buf += ret;
+	}
+
+	return buf - buf_start;
+}
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
new file mode 100644
index 00000000..fc784284
--- /dev/null
+++ b/tools/perf/util/util.h
@@ -0,0 +1,247 @@
+#ifndef GIT_COMPAT_UTIL_H
+#define GIT_COMPAT_UTIL_H
+
+#define _FILE_OFFSET_BITS 64
+
+#ifndef FLEX_ARRAY
+/*
+ * See if our compiler is known to support flexible array members.
+ */
+#if defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L)
+# define FLEX_ARRAY /* empty */
+#elif defined(__GNUC__)
+# if (__GNUC__ >= 3)
+#  define FLEX_ARRAY /* empty */
+# else
+#  define FLEX_ARRAY 0 /* older GNU extension */
+# endif
+#endif
+
+/*
+ * Otherwise, default to safer but a bit wasteful traditional style
+ */
+#ifndef FLEX_ARRAY
+# define FLEX_ARRAY 1
+#endif
+#endif
+
+#define ARRAY_SIZE(x) (sizeof(x)/sizeof(x[0]))
+
+#ifdef __GNUC__
+#define TYPEOF(x) (__typeof__(x))
+#else
+#define TYPEOF(x)
+#endif
+
+#define MSB(x, bits) ((x) & TYPEOF(x)(~0ULL << (sizeof(x) * 8 - (bits))))
+#define HAS_MULTI_BITS(i)  ((i) & ((i) - 1))  /* checks if an integer has more than 1 bit set */
+
+/* Approximation of the length of the decimal representation of this type. */
+#define decimal_length(x)	((int)(sizeof(x) * 2.56 + 0.5) + 1)
+
+#define _ALL_SOURCE 1
+#define _GNU_SOURCE 1
+#define _BSD_SOURCE 1
+#define HAS_BOOL
+
+#include <unistd.h>
+#include <stdio.h>
+#include <sys/stat.h>
+#include <sys/statfs.h>
+#include <fcntl.h>
+#include <stdbool.h>
+#include <stddef.h>
+#include <stdlib.h>
+#include <stdarg.h>
+#include <string.h>
+#include <errno.h>
+#include <limits.h>
+#include <sys/param.h>
+#include <sys/types.h>
+#include <dirent.h>
+#include <sys/time.h>
+#include <time.h>
+#include <signal.h>
+#include <fnmatch.h>
+#include <assert.h>
+#include <regex.h>
+#include <utime.h>
+#include <sys/wait.h>
+#include <sys/poll.h>
+#include <sys/socket.h>
+#include <sys/ioctl.h>
+#include <sys/select.h>
+#include <netinet/in.h>
+#include <netinet/tcp.h>
+#include <arpa/inet.h>
+#include <netdb.h>
+#include <pwd.h>
+#include <inttypes.h>
+#include "../../../include/linux/magic.h"
+#include "types.h"
+#include <sys/ttydefaults.h>
+
+extern const char *graph_line;
+extern const char *graph_dotted_line;
+extern char buildid_dir[];
+
+/* On most systems <limits.h> would have given us this, but
+ * not on some systems (e.g. GNU/Hurd).
+ */
+#ifndef PATH_MAX
+#define PATH_MAX 4096
+#endif
+
+#ifndef PRIuMAX
+#define PRIuMAX "llu"
+#endif
+
+#ifndef PRIu32
+#define PRIu32 "u"
+#endif
+
+#ifndef PRIx32
+#define PRIx32 "x"
+#endif
+
+#ifndef PATH_SEP
+#define PATH_SEP ':'
+#endif
+
+#ifndef STRIP_EXTENSION
+#define STRIP_EXTENSION ""
+#endif
+
+#ifndef has_dos_drive_prefix
+#define has_dos_drive_prefix(path) 0
+#endif
+
+#ifndef is_dir_sep
+#define is_dir_sep(c) ((c) == '/')
+#endif
+
+#ifdef __GNUC__
+#define NORETURN __attribute__((__noreturn__))
+#else
+#define NORETURN
+#ifndef __attribute__
+#define __attribute__(x)
+#endif
+#endif
+
+/* General helper functions */
+extern void usage(const char *err) NORETURN;
+extern void die(const char *err, ...) NORETURN __attribute__((format (printf, 1, 2)));
+extern int error(const char *err, ...) __attribute__((format (printf, 1, 2)));
+extern void warning(const char *err, ...) __attribute__((format (printf, 1, 2)));
+
+#include "../../../include/linux/stringify.h"
+
+#define DIE_IF(cnd)	\
+	do { if (cnd)	\
+		die(" at (" __FILE__ ":" __stringify(__LINE__) "): "	\
+		    __stringify(cnd) "\n");				\
+	} while (0)
+
+
+extern void set_die_routine(void (*routine)(const char *err, va_list params) NORETURN);
+
+extern int prefixcmp(const char *str, const char *prefix);
+extern void set_buildid_dir(void);
+extern void disable_buildid_cache(void);
+
+static inline const char *skip_prefix(const char *str, const char *prefix)
+{
+	size_t len = strlen(prefix);
+	return strncmp(str, prefix, len) ? NULL : str + len;
+}
+
+#ifdef __GLIBC_PREREQ
+#if __GLIBC_PREREQ(2, 1)
+#define HAVE_STRCHRNUL
+#endif
+#endif
+
+#ifndef HAVE_STRCHRNUL
+#define strchrnul gitstrchrnul
+static inline char *gitstrchrnul(const char *s, int c)
+{
+	while (*s && *s != c)
+		s++;
+	return (char *)s;
+}
+#endif
+
+/*
+ * Wrappers:
+ */
+extern char *xstrdup(const char *str);
+extern void *xrealloc(void *ptr, size_t size) __attribute__((weak));
+
+
+static inline void *zalloc(size_t size)
+{
+	return calloc(1, size);
+}
+
+static inline int has_extension(const char *filename, const char *ext)
+{
+	size_t len = strlen(filename);
+	size_t extlen = strlen(ext);
+
+	return len > extlen && !memcmp(filename + len - extlen, ext, extlen);
+}
+
+/* Sane ctype - no locale, and works with signed chars */
+#undef isascii
+#undef isspace
+#undef isdigit
+#undef isxdigit
+#undef isalpha
+#undef isprint
+#undef isalnum
+#undef tolower
+#undef toupper
+
+extern unsigned char sane_ctype[256];
+#define GIT_SPACE		0x01
+#define GIT_DIGIT		0x02
+#define GIT_ALPHA		0x04
+#define GIT_GLOB_SPECIAL	0x08
+#define GIT_REGEX_SPECIAL	0x10
+#define GIT_PRINT_EXTRA		0x20
+#define GIT_PRINT		0x3E
+#define sane_istest(x,mask) ((sane_ctype[(unsigned char)(x)] & (mask)) != 0)
+#define isascii(x) (((x) & ~0x7f) == 0)
+#define isspace(x) sane_istest(x,GIT_SPACE)
+#define isdigit(x) sane_istest(x,GIT_DIGIT)
+#define isxdigit(x)	\
+	(sane_istest(toupper(x), GIT_ALPHA | GIT_DIGIT) && toupper(x) < 'G')
+#define isalpha(x) sane_istest(x,GIT_ALPHA)
+#define isalnum(x) sane_istest(x,GIT_ALPHA | GIT_DIGIT)
+#define isprint(x) sane_istest(x,GIT_PRINT)
+#define tolower(x) sane_case((unsigned char)(x), 0x20)
+#define toupper(x) sane_case((unsigned char)(x), 0)
+
+static inline int sane_case(int x, int high)
+{
+	if (sane_istest(x, GIT_ALPHA))
+		x = (x & ~0x20) | high;
+	return x;
+}
+
+int mkdir_p(char *path, mode_t mode);
+int copyfile(const char *from, const char *to);
+
+s64 perf_atoll(const char *str);
+char **argv_split(const char *str, int *argcp);
+void argv_free(char **argv);
+bool strglobmatch(const char *str, const char *pat);
+bool strlazymatch(const char *str, const char *pat);
+unsigned long convert_unit(unsigned long value, char *unit);
+int readn(int fd, void *buf, size_t size);
+
+#define _STR(x) #x
+#define STR(x) _STR(x)
+
+#endif
diff --git a/tools/perf/util/values.c b/tools/perf/util/values.c
new file mode 100644
index 00000000..bdd33470
--- /dev/null
+++ b/tools/perf/util/values.c
@@ -0,0 +1,231 @@
+#include <stdlib.h>
+
+#include "util.h"
+#include "values.h"
+
+void perf_read_values_init(struct perf_read_values *values)
+{
+	values->threads_max = 16;
+	values->pid = malloc(values->threads_max * sizeof(*values->pid));
+	values->tid = malloc(values->threads_max * sizeof(*values->tid));
+	values->value = malloc(values->threads_max * sizeof(*values->value));
+	if (!values->pid || !values->tid || !values->value)
+		die("failed to allocate read_values threads arrays");
+	values->threads = 0;
+
+	values->counters_max = 16;
+	values->counterrawid = malloc(values->counters_max
+				      * sizeof(*values->counterrawid));
+	values->countername = malloc(values->counters_max
+				     * sizeof(*values->countername));
+	if (!values->counterrawid || !values->countername)
+		die("failed to allocate read_values counters arrays");
+	values->counters = 0;
+}
+
+void perf_read_values_destroy(struct perf_read_values *values)
+{
+	int i;
+
+	if (!values->threads_max || !values->counters_max)
+		return;
+
+	for (i = 0; i < values->threads; i++)
+		free(values->value[i]);
+	free(values->pid);
+	free(values->tid);
+	free(values->counterrawid);
+	for (i = 0; i < values->counters; i++)
+		free(values->countername[i]);
+	free(values->countername);
+}
+
+static void perf_read_values__enlarge_threads(struct perf_read_values *values)
+{
+	values->threads_max *= 2;
+	values->pid = realloc(values->pid,
+			      values->threads_max * sizeof(*values->pid));
+	values->tid = realloc(values->tid,
+			      values->threads_max * sizeof(*values->tid));
+	values->value = realloc(values->value,
+				values->threads_max * sizeof(*values->value));
+	if (!values->pid || !values->tid || !values->value)
+		die("failed to enlarge read_values threads arrays");
+}
+
+static int perf_read_values__findnew_thread(struct perf_read_values *values,
+					    u32 pid, u32 tid)
+{
+	int i;
+
+	for (i = 0; i < values->threads; i++)
+		if (values->pid[i] == pid && values->tid[i] == tid)
+			return i;
+
+	if (values->threads == values->threads_max)
+		perf_read_values__enlarge_threads(values);
+
+	i = values->threads++;
+	values->pid[i] = pid;
+	values->tid[i] = tid;
+	values->value[i] = malloc(values->counters_max * sizeof(**values->value));
+	if (!values->value[i])
+		die("failed to allocate read_values counters array");
+
+	return i;
+}
+
+static void perf_read_values__enlarge_counters(struct perf_read_values *values)
+{
+	int i;
+
+	values->counters_max *= 2;
+	values->counterrawid = realloc(values->counterrawid,
+				       values->counters_max * sizeof(*values->counterrawid));
+	values->countername = realloc(values->countername,
+				      values->counters_max * sizeof(*values->countername));
+	if (!values->counterrawid || !values->countername)
+		die("failed to enlarge read_values counters arrays");
+
+	for (i = 0; i < values->threads; i++) {
+		values->value[i] = realloc(values->value[i],
+					   values->counters_max * sizeof(**values->value));
+		if (!values->value[i])
+			die("failed to enlarge read_values counters arrays");
+	}
+}
+
+static int perf_read_values__findnew_counter(struct perf_read_values *values,
+					     u64 rawid, const char *name)
+{
+	int i;
+
+	for (i = 0; i < values->counters; i++)
+		if (values->counterrawid[i] == rawid)
+			return i;
+
+	if (values->counters == values->counters_max)
+		perf_read_values__enlarge_counters(values);
+
+	i = values->counters++;
+	values->counterrawid[i] = rawid;
+	values->countername[i] = strdup(name);
+
+	return i;
+}
+
+void perf_read_values_add_value(struct perf_read_values *values,
+				u32 pid, u32 tid,
+				u64 rawid, const char *name, u64 value)
+{
+	int tindex, cindex;
+
+	tindex = perf_read_values__findnew_thread(values, pid, tid);
+	cindex = perf_read_values__findnew_counter(values, rawid, name);
+
+	values->value[tindex][cindex] = value;
+}
+
+static void perf_read_values__display_pretty(FILE *fp,
+					     struct perf_read_values *values)
+{
+	int i, j;
+	int pidwidth, tidwidth;
+	int *counterwidth;
+
+	counterwidth = malloc(values->counters * sizeof(*counterwidth));
+	if (!counterwidth)
+		die("failed to allocate counterwidth array");
+	tidwidth = 3;
+	pidwidth = 3;
+	for (j = 0; j < values->counters; j++)
+		counterwidth[j] = strlen(values->countername[j]);
+	for (i = 0; i < values->threads; i++) {
+		int width;
+
+		width = snprintf(NULL, 0, "%d", values->pid[i]);
+		if (width > pidwidth)
+			pidwidth = width;
+		width = snprintf(NULL, 0, "%d", values->tid[i]);
+		if (width > tidwidth)
+			tidwidth = width;
+		for (j = 0; j < values->counters; j++) {
+			width = snprintf(NULL, 0, "%" PRIu64, values->value[i][j]);
+			if (width > counterwidth[j])
+				counterwidth[j] = width;
+		}
+	}
+
+	fprintf(fp, "# %*s  %*s", pidwidth, "PID", tidwidth, "TID");
+	for (j = 0; j < values->counters; j++)
+		fprintf(fp, "  %*s", counterwidth[j], values->countername[j]);
+	fprintf(fp, "\n");
+
+	for (i = 0; i < values->threads; i++) {
+		fprintf(fp, "  %*d  %*d", pidwidth, values->pid[i],
+			tidwidth, values->tid[i]);
+		for (j = 0; j < values->counters; j++)
+			fprintf(fp, "  %*" PRIu64,
+				counterwidth[j], values->value[i][j]);
+		fprintf(fp, "\n");
+	}
+	free(counterwidth);
+}
+
+static void perf_read_values__display_raw(FILE *fp,
+					  struct perf_read_values *values)
+{
+	int width, pidwidth, tidwidth, namewidth, rawwidth, countwidth;
+	int i, j;
+
+	tidwidth = 3; /* TID */
+	pidwidth = 3; /* PID */
+	namewidth = 4; /* "Name" */
+	rawwidth = 3; /* "Raw" */
+	countwidth = 5; /* "Count" */
+
+	for (i = 0; i < values->threads; i++) {
+		width = snprintf(NULL, 0, "%d", values->pid[i]);
+		if (width > pidwidth)
+			pidwidth = width;
+		width = snprintf(NULL, 0, "%d", values->tid[i]);
+		if (width > tidwidth)
+			tidwidth = width;
+	}
+	for (j = 0; j < values->counters; j++) {
+		width = strlen(values->countername[j]);
+		if (width > namewidth)
+			namewidth = width;
+		width = snprintf(NULL, 0, "%" PRIx64, values->counterrawid[j]);
+		if (width > rawwidth)
+			rawwidth = width;
+	}
+	for (i = 0; i < values->threads; i++) {
+		for (j = 0; j < values->counters; j++) {
+			width = snprintf(NULL, 0, "%" PRIu64, values->value[i][j]);
+			if (width > countwidth)
+				countwidth = width;
+		}
+	}
+
+	fprintf(fp, "# %*s  %*s  %*s  %*s  %*s\n",
+		pidwidth, "PID", tidwidth, "TID",
+		namewidth, "Name", rawwidth, "Raw",
+		countwidth, "Count");
+	for (i = 0; i < values->threads; i++)
+		for (j = 0; j < values->counters; j++)
+			fprintf(fp, "  %*d  %*d  %*s  %*" PRIx64 "  %*" PRIu64,
+				pidwidth, values->pid[i],
+				tidwidth, values->tid[i],
+				namewidth, values->countername[j],
+				rawwidth, values->counterrawid[j],
+				countwidth, values->value[i][j]);
+}
+
+void perf_read_values_display(FILE *fp, struct perf_read_values *values, int raw)
+{
+	if (raw)
+		perf_read_values__display_raw(fp, values);
+	else
+		perf_read_values__display_pretty(fp, values);
+}
diff --git a/tools/perf/util/values.h b/tools/perf/util/values.h
new file mode 100644
index 00000000..2fa967e1
--- /dev/null
+++ b/tools/perf/util/values.h
@@ -0,0 +1,27 @@
+#ifndef __PERF_VALUES_H
+#define __PERF_VALUES_H
+
+#include "types.h"
+
+struct perf_read_values {
+	int threads;
+	int threads_max;
+	u32 *pid, *tid;
+	int counters;
+	int counters_max;
+	u64 *counterrawid;
+	char **countername;
+	u64 **value;
+};
+
+void perf_read_values_init(struct perf_read_values *values);
+void perf_read_values_destroy(struct perf_read_values *values);
+
+void perf_read_values_add_value(struct perf_read_values *values,
+				u32 pid, u32 tid,
+				u64 rawid, const char *name, u64 value);
+
+void perf_read_values_display(FILE *fp, struct perf_read_values *values,
+			      int raw);
+
+#endif /* __PERF_VALUES_H */
diff --git a/tools/perf/util/wrapper.c b/tools/perf/util/wrapper.c
new file mode 100644
index 00000000..73e900ed
--- /dev/null
+++ b/tools/perf/util/wrapper.c
@@ -0,0 +1,40 @@
+/*
+ * Various trivial helper wrappers around standard functions
+ */
+#include "cache.h"
+
+/*
+ * There's no pack memory to release - but stay close to the Git
+ * version so wrap this away:
+ */
+static inline void release_pack_memory(size_t size __used, int flag __used)
+{
+}
+
+char *xstrdup(const char *str)
+{
+	char *ret = strdup(str);
+	if (!ret) {
+		release_pack_memory(strlen(str) + 1, -1);
+		ret = strdup(str);
+		if (!ret)
+			die("Out of memory, strdup failed");
+	}
+	return ret;
+}
+
+void *xrealloc(void *ptr, size_t size)
+{
+	void *ret = realloc(ptr, size);
+	if (!ret && !size)
+		ret = realloc(ptr, 1);
+	if (!ret) {
+		release_pack_memory(size, -1);
+		ret = realloc(ptr, size);
+		if (!ret && !size)
+			ret = realloc(ptr, 1);
+		if (!ret)
+			die("Out of memory, realloc failed");
+	}
+	return ret;
+}
diff --git a/tools/perf/util/xyarray.c b/tools/perf/util/xyarray.c
new file mode 100644
index 00000000..22afbf6c
--- /dev/null
+++ b/tools/perf/util/xyarray.c
@@ -0,0 +1,20 @@
+#include "xyarray.h"
+#include "util.h"
+
+struct xyarray *xyarray__new(int xlen, int ylen, size_t entry_size)
+{
+	size_t row_size = ylen * entry_size;
+	struct xyarray *xy = zalloc(sizeof(*xy) + xlen * row_size);
+
+	if (xy != NULL) {
+		xy->entry_size = entry_size;
+		xy->row_size   = row_size;
+	}
+
+	return xy;
+}
+
+void xyarray__delete(struct xyarray *xy)
+{
+	free(xy);
+}
diff --git a/tools/perf/util/xyarray.h b/tools/perf/util/xyarray.h
new file mode 100644
index 00000000..c488a072
--- /dev/null
+++ b/tools/perf/util/xyarray.h
@@ -0,0 +1,20 @@
+#ifndef _PERF_XYARRAY_H_
+#define _PERF_XYARRAY_H_ 1
+
+#include <sys/types.h>
+
+struct xyarray {
+	size_t row_size;
+	size_t entry_size;
+	char contents[];
+};
+
+struct xyarray *xyarray__new(int xlen, int ylen, size_t entry_size);
+void xyarray__delete(struct xyarray *xy);
+
+static inline void *xyarray__entry(struct xyarray *xy, int x, int y)
+{
+	return &xy->contents[x * xy->row_size + y * xy->entry_size];
+}
+
+#endif /* _PERF_XYARRAY_H_ */