Merge commit 'perf/core' into perf/hw-breakpoint

Conflicts:
	kernel/Makefile
	kernel/trace/Makefile
	kernel/trace/trace.h
	samples/Makefile

Merge reason: We need to be uptodate with the perf events development
branch because we plan to rewrite the breakpoints API on top of
perf events.

58 files changed, 6810 insertions, 2817 deletions

diff --git a/tools/perf/.gitignore b/tools/perf/.gitignore
index d69a759a1046..0854f110bf7f 100644
--- a/tools/perf/.gitignore
+++ b/tools/perf/.gitignore
@@ -10,6 +10,7 @@ perf-stat
 perf-top
 perf*.1
 perf*.xml
+perf*.html
 common-cmds.h
 tags
 TAGS
diff --git a/tools/perf/Documentation/perf-sched.txt b/tools/perf/Documentation/perf-sched.txt
new file mode 100644
index 000000000000..1ce79198997b
--- /dev/null
+++ b/tools/perf/Documentation/perf-sched.txt
@@ -0,0 +1,41 @@
+perf-sched(1)
+==============
+
+NAME
+----
+perf-sched - Tool to trace/measure scheduler properties (latencies)
+
+SYNOPSIS
+--------
+[verse]
+'perf sched' {record|latency|replay|trace}
+
+DESCRIPTION
+-----------
+There's four variants of perf sched:
+
+  'perf sched record <command>' to record the scheduling events
+  of an arbitrary workload.
+
+  'perf sched latency' to report the per task scheduling latencies
+  and other scheduling properties of the workload.
+
+  'perf sched trace' to see a detailed trace of the workload that
+  was recorded.
+
+  'perf sched replay' to simulate the workload that was recorded
+  via perf sched record. (this is done by starting up mockup threads
+  that mimic the workload based on the events in the trace. These
+  threads can then replay the timings (CPU runtime and sleep patterns)
+  of the workload as it occured when it was recorded - and can repeat
+  it a number of times, measuring its performance.)
+
+OPTIONS
+-------
+-D::
+--dump-raw-trace=::
+        Display verbose dump of the sched data.
+
+SEE ALSO
+--------
+linkperf:perf-record[1]
diff --git a/tools/perf/Documentation/perf-timechart.txt b/tools/perf/Documentation/perf-timechart.txt
new file mode 100644
index 000000000000..a7910099d6fd
--- /dev/null
+++ b/tools/perf/Documentation/perf-timechart.txt
@@ -0,0 +1,41 @@
+perf-timechart(1)
+=================
+
+NAME
+----
+perf-timechart - Tool to visualize total system behavior during a workload
+
+SYNOPSIS
+--------
+[verse]
+'perf timechart' {record}
+
+DESCRIPTION
+-----------
+There are two variants of perf timechart:
+
+  'perf timechart record <command>' to record the system level events
+  of an arbitrary workload.
+
+  'perf timechart' to turn a trace into a Scalable Vector Graphics file,
+  that can be viewed with popular SVG viewers such as 'Inkscape'.
+
+OPTIONS
+-------
+-o::
+--output=::
+        Select the output file (default: output.svg)
+-i::
+--input=::
+        Select the input file (default: perf.data)
+-w::
+--width=::
+        Select the width of the SVG file (default: 1000)
+-p::
+--power-only::
+        Only output the CPU power section of the diagram
+
+
+SEE ALSO
+--------
+linkperf:perf-record[1]
diff --git a/tools/perf/Documentation/perf-trace.txt b/tools/perf/Documentation/perf-trace.txt
new file mode 100644
index 000000000000..41ed75398ca9
--- /dev/null
+++ b/tools/perf/Documentation/perf-trace.txt
@@ -0,0 +1,25 @@
+perf-trace(1)
+==============
+
+NAME
+----
+perf-trace - Read perf.data (created by perf record) and display trace output
+
+SYNOPSIS
+--------
+[verse]
+'perf trace' [-i <file> | --input=file] symbol_name
+
+DESCRIPTION
+-----------
+This command reads the input file and displays the trace recorded.
+
+OPTIONS
+-------
+-D::
+--dump-raw-trace=::
+        Display verbose dump of the trace data.
+
+SEE ALSO
+--------
+linkperf:perf-record[1]
diff --git a/tools/perf/Makefile b/tools/perf/Makefile
index 9f8d207a91bf..106c15055b50 100644
--- a/tools/perf/Makefile
+++ b/tools/perf/Makefile
@@ -157,11 +157,18 @@ uname_R := $(shell sh -c 'uname -r 2>/dev/null || echo not')
 uname_P := $(shell sh -c 'uname -p 2>/dev/null || echo not')
 uname_V := $(shell sh -c 'uname -v 2>/dev/null || echo not')
 
-# If we're on a 64-bit kernel, use -m64
-ifndef NO_64BIT
-	ifneq ($(patsubst %64,%,$(uname_M)),$(uname_M))
-	  M64 := -m64
-	endif
+#
+# Add -m32 for cross-builds:
+#
+ifdef NO_64BIT
+  MBITS := -m32
+else
+  #
+  # If we're on a 64-bit kernel, use -m64:
+  #
+  ifneq ($(patsubst %64,%,$(uname_M)),$(uname_M))
+    MBITS := -m64
+  endif
 endif
 
 # CFLAGS and LDFLAGS are for the users to override from the command line.
@@ -194,7 +201,7 @@ EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wold-style-definition
 EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wstrict-prototypes
 EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wdeclaration-after-statement
 
-CFLAGS = $(M64) -ggdb3 -Wall -Wextra -std=gnu99 -Werror -O6 -fstack-protector-all -D_FORTIFY_SOURCE=2 $(EXTRA_WARNINGS)
+CFLAGS = $(MBITS) -ggdb3 -Wall -Wextra -std=gnu99 -Werror -O6 -fstack-protector-all -D_FORTIFY_SOURCE=2 $(EXTRA_WARNINGS)
 LDFLAGS = -lpthread -lrt -lelf -lm
 ALL_CFLAGS = $(CFLAGS)
 ALL_LDFLAGS = $(LDFLAGS)
@@ -318,11 +325,12 @@ export PERL_PATH
 
 LIB_FILE=libperf.a
 
-LIB_H += ../../include/linux/perf_counter.h
+LIB_H += ../../include/linux/perf_event.h
 LIB_H += ../../include/linux/rbtree.h
 LIB_H += ../../include/linux/list.h
 LIB_H += util/include/linux/list.h
 LIB_H += perf.h
+LIB_H += util/event.h
 LIB_H += util/types.h
 LIB_H += util/levenshtein.h
 LIB_H += util/parse-options.h
@@ -336,9 +344,12 @@ LIB_H += util/strlist.h
 LIB_H += util/run-command.h
 LIB_H += util/sigchain.h
 LIB_H += util/symbol.h
-LIB_H += util/module.h
 LIB_H += util/color.h
 LIB_H += util/values.h
+LIB_H += util/sort.h
+LIB_H += util/hist.h
+LIB_H += util/thread.h
+LIB_H += util/data_map.h
 
 LIB_OBJS += util/abspath.o
 LIB_OBJS += util/alias.o
@@ -361,7 +372,6 @@ LIB_OBJS += util/usage.o
 LIB_OBJS += util/wrapper.o
 LIB_OBJS += util/sigchain.o
 LIB_OBJS += util/symbol.o
-LIB_OBJS += util/module.o
 LIB_OBJS += util/color.o
 LIB_OBJS += util/pager.o
 LIB_OBJS += util/header.o
@@ -373,13 +383,19 @@ LIB_OBJS += util/thread.o
 LIB_OBJS += util/trace-event-parse.o
 LIB_OBJS += util/trace-event-read.o
 LIB_OBJS += util/trace-event-info.o
+LIB_OBJS += util/svghelper.o
+LIB_OBJS += util/sort.o
+LIB_OBJS += util/hist.o
+LIB_OBJS += util/data_map.o
 
 BUILTIN_OBJS += builtin-annotate.o
 BUILTIN_OBJS += builtin-help.o
+BUILTIN_OBJS += builtin-sched.o
 BUILTIN_OBJS += builtin-list.o
 BUILTIN_OBJS += builtin-record.o
 BUILTIN_OBJS += builtin-report.o
 BUILTIN_OBJS += builtin-stat.o
+BUILTIN_OBJS += builtin-timechart.o
 BUILTIN_OBJS += builtin-top.o
 BUILTIN_OBJS += builtin-trace.o
 
@@ -413,7 +429,7 @@ ifeq ($(uname_S),Darwin)
 endif
 
 ifneq ($(shell sh -c "(echo '\#include <libelf.h>'; echo 'int main(void) { Elf * elf = elf_begin(0, ELF_C_READ_MMAP, 0); return (long)elf; }') | $(CC) -x c - $(ALL_CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -o /dev/null $(ALL_LDFLAGS) > /dev/null 2>&1 && echo y"), y)
-	msg := $(error No libelf.h/libelf found, please install libelf-dev/elfutils-libelf-devel);
+	msg := $(error No libelf.h/libelf found, please install libelf-dev/elfutils-libelf-devel and glibc-dev[el]);
 endif
 
 ifdef NO_DEMANGLE
@@ -710,6 +726,12 @@ builtin-help.o: builtin-help.c common-cmds.h PERF-CFLAGS
 		'-DPERF_MAN_PATH="$(mandir_SQ)"' \
 		'-DPERF_INFO_PATH="$(infodir_SQ)"' $<
 
+builtin-timechart.o: builtin-timechart.c common-cmds.h PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) \
+		'-DPERF_HTML_PATH="$(htmldir_SQ)"' \
+		'-DPERF_MAN_PATH="$(mandir_SQ)"' \
+		'-DPERF_INFO_PATH="$(infodir_SQ)"' $<
+
 $(BUILT_INS): perf$X
 	$(QUIET_BUILT_IN)$(RM) $@ && \
 	ln perf$X $@ 2>/dev/null || \
@@ -719,7 +741,7 @@ $(BUILT_INS): perf$X
 common-cmds.h: util/generate-cmdlist.sh command-list.txt
 
 common-cmds.h: $(wildcard Documentation/perf-*.txt)
-	$(QUIET_GEN)util/generate-cmdlist.sh > $@+ && mv $@+ $@
+	$(QUIET_GEN). util/generate-cmdlist.sh > $@+ && mv $@+ $@
 
 $(patsubst %.sh,%,$(SCRIPT_SH)) : % : %.sh
 	$(QUIET_GEN)$(RM) $@ $@+ && \
diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c
index 043d85b7e254..56ba71658d70 100644
--- a/tools/perf/builtin-annotate.c
+++ b/tools/perf/builtin-annotate.c
@@ -22,15 +22,13 @@
 #include "util/parse-options.h"
 #include "util/parse-events.h"
 #include "util/thread.h"
+#include "util/sort.h"
+#include "util/hist.h"
 
 static char		const *input_name = "perf.data";
 
-static char		default_sort_order[] = "comm,symbol";
-static char		*sort_order = default_sort_order;
-
 static int		force;
 static int		input;
-static int		show_mask = SHOW_KERNEL | SHOW_USER | SHOW_HV;
 
 static int		full_paths;
 
@@ -39,258 +37,12 @@ static int		print_line;
 static unsigned long	page_size;
 static unsigned long	mmap_window = 32;
 
-static struct rb_root	threads;
-static struct thread	*last_match;
-
-
 struct sym_ext {
 	struct rb_node	node;
 	double		percent;
 	char		*path;
 };
 
-/*
- * histogram, sorted on item, collects counts
- */
-
-static struct rb_root hist;
-
-struct hist_entry {
-	struct rb_node	 rb_node;
-
-	struct thread	 *thread;
-	struct map	 *map;
-	struct dso	 *dso;
-	struct symbol	 *sym;
-	u64	 ip;
-	char		 level;
-
-	uint32_t	 count;
-};
-
-/*
- * configurable sorting bits
- */
-
-struct sort_entry {
-	struct list_head list;
-
-	const char *header;
-
-	int64_t (*cmp)(struct hist_entry *, struct hist_entry *);
-	int64_t (*collapse)(struct hist_entry *, struct hist_entry *);
-	size_t	(*print)(FILE *fp, struct hist_entry *);
-};
-
-/* --sort pid */
-
-static int64_t
-sort__thread_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-	return right->thread->pid - left->thread->pid;
-}
-
-static size_t
-sort__thread_print(FILE *fp, struct hist_entry *self)
-{
-	return fprintf(fp, "%16s:%5d", self->thread->comm ?: "", self->thread->pid);
-}
-
-static struct sort_entry sort_thread = {
-	.header = "         Command:  Pid",
-	.cmp	= sort__thread_cmp,
-	.print	= sort__thread_print,
-};
-
-/* --sort comm */
-
-static int64_t
-sort__comm_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-	return right->thread->pid - left->thread->pid;
-}
-
-static 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) {
-		if (!comm_l && !comm_r)
-			return 0;
-		else if (!comm_l)
-			return -1;
-		else
-			return 1;
-	}
-
-	return strcmp(comm_l, comm_r);
-}
-
-static size_t
-sort__comm_print(FILE *fp, struct hist_entry *self)
-{
-	return fprintf(fp, "%16s", self->thread->comm);
-}
-
-static struct sort_entry sort_comm = {
-	.header		= "         Command",
-	.cmp		= sort__comm_cmp,
-	.collapse	= sort__comm_collapse,
-	.print		= sort__comm_print,
-};
-
-/* --sort dso */
-
-static int64_t
-sort__dso_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-	struct dso *dso_l = left->dso;
-	struct dso *dso_r = right->dso;
-
-	if (!dso_l || !dso_r) {
-		if (!dso_l && !dso_r)
-			return 0;
-		else if (!dso_l)
-			return -1;
-		else
-			return 1;
-	}
-
-	return strcmp(dso_l->name, dso_r->name);
-}
-
-static size_t
-sort__dso_print(FILE *fp, struct hist_entry *self)
-{
-	if (self->dso)
-		return fprintf(fp, "%-25s", self->dso->name);
-
-	return fprintf(fp, "%016llx         ", (u64)self->ip);
-}
-
-static struct sort_entry sort_dso = {
-	.header = "Shared Object            ",
-	.cmp	= sort__dso_cmp,
-	.print	= sort__dso_print,
-};
-
-/* --sort symbol */
-
-static int64_t
-sort__sym_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-	u64 ip_l, ip_r;
-
-	if (left->sym == right->sym)
-		return 0;
-
-	ip_l = left->sym ? left->sym->start : left->ip;
-	ip_r = right->sym ? right->sym->start : right->ip;
-
-	return (int64_t)(ip_r - ip_l);
-}
-
-static size_t
-sort__sym_print(FILE *fp, struct hist_entry *self)
-{
-	size_t ret = 0;
-
-	if (verbose)
-		ret += fprintf(fp, "%#018llx  ", (u64)self->ip);
-
-	if (self->sym) {
-		ret += fprintf(fp, "[%c] %s",
-			self->dso == kernel_dso ? 'k' : '.', self->sym->name);
-	} else {
-		ret += fprintf(fp, "%#016llx", (u64)self->ip);
-	}
-
-	return ret;
-}
-
-static struct sort_entry sort_sym = {
-	.header = "Symbol",
-	.cmp	= sort__sym_cmp,
-	.print	= sort__sym_print,
-};
-
-static int sort__need_collapse = 0;
-
-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,	},
-};
-
-static LIST_HEAD(hist_entry__sort_list);
-
-static int sort_dimension__add(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->collapse)
-			sort__need_collapse = 1;
-
-		list_add_tail(&sd->entry->list, &hist_entry__sort_list);
-		sd->taken = 1;
-
-		return 0;
-	}
-
-	return -ESRCH;
-}
-
-static 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->cmp(left, right);
-		if (cmp)
-			break;
-	}
-
-	return cmp;
-}
-
-static 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->collapse ?: se->cmp;
-
-		cmp = f(left, right);
-		if (cmp)
-			break;
-	}
-
-	return cmp;
-}
 
 /*
  * collect histogram counts
@@ -306,6 +58,7 @@ static void hist_hit(struct hist_entry *he, u64 ip)
 		return;
 
 	sym_size = sym->end - sym->start;
+	ip = he->map->map_ip(he->map, ip);
 	offset = ip - sym->start;
 
 	if (offset >= sym_size)
@@ -322,173 +75,27 @@ static void hist_hit(struct hist_entry *he, u64 ip)
 			sym->hist[offset]);
 }
 
-static int
-hist_entry__add(struct thread *thread, struct map *map, struct dso *dso,
-		struct symbol *sym, u64 ip, char level)
+static int hist_entry__add(struct thread *thread, struct map *map,
+			   struct symbol *sym, u64 ip, u64 count, char level)
 {
-	struct rb_node **p = &hist.rb_node;
-	struct rb_node *parent = NULL;
-	struct hist_entry *he;
-	struct hist_entry entry = {
-		.thread	= thread,
-		.map	= map,
-		.dso	= dso,
-		.sym	= sym,
-		.ip	= ip,
-		.level	= level,
-		.count	= 1,
-	};
-	int cmp;
-
-	while (*p != NULL) {
-		parent = *p;
-		he = rb_entry(parent, struct hist_entry, rb_node);
-
-		cmp = hist_entry__cmp(&entry, he);
-
-		if (!cmp) {
-			hist_hit(he, ip);
-
-			return 0;
-		}
-
-		if (cmp < 0)
-			p = &(*p)->rb_left;
-		else
-			p = &(*p)->rb_right;
-	}
-
-	he = malloc(sizeof(*he));
-	if (!he)
+	bool hit;
+	struct hist_entry *he = __hist_entry__add(thread, map, sym, NULL, ip,
+						  count, level, &hit);
+	if (he == NULL)
 		return -ENOMEM;
-	*he = entry;
-	rb_link_node(&he->rb_node, parent, p);
-	rb_insert_color(&he->rb_node, &hist);
-
+	if (hit)
+		hist_hit(he, ip);
 	return 0;
 }
 
-static void hist_entry__free(struct hist_entry *he)
-{
-	free(he);
-}
-
-/*
- * collapse the histogram
- */
-
-static struct rb_root collapse_hists;
-
-static void collapse__insert_entry(struct hist_entry *he)
-{
-	struct rb_node **p = &collapse_hists.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->count += he->count;
-			hist_entry__free(he);
-			return;
-		}
-
-		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, &collapse_hists);
-}
-
-static void collapse__resort(void)
-{
-	struct rb_node *next;
-	struct hist_entry *n;
-
-	if (!sort__need_collapse)
-		return;
-
-	next = rb_first(&hist);
-	while (next) {
-		n = rb_entry(next, struct hist_entry, rb_node);
-		next = rb_next(&n->rb_node);
-
-		rb_erase(&n->rb_node, &hist);
-		collapse__insert_entry(n);
-	}
-}
-
-/*
- * reverse the map, sort on count.
- */
-
-static struct rb_root output_hists;
-
-static void output__insert_entry(struct hist_entry *he)
-{
-	struct rb_node **p = &output_hists.rb_node;
-	struct rb_node *parent = NULL;
-	struct hist_entry *iter;
-
-	while (*p != NULL) {
-		parent = *p;
-		iter = rb_entry(parent, struct hist_entry, rb_node);
-
-		if (he->count > iter->count)
-			p = &(*p)->rb_left;
-		else
-			p = &(*p)->rb_right;
-	}
-
-	rb_link_node(&he->rb_node, parent, p);
-	rb_insert_color(&he->rb_node, &output_hists);
-}
-
-static void output__resort(void)
-{
-	struct rb_node *next;
-	struct hist_entry *n;
-	struct rb_root *tree = &hist;
-
-	if (sort__need_collapse)
-		tree = &collapse_hists;
-
-	next = rb_first(tree);
-
-	while (next) {
-		n = rb_entry(next, struct hist_entry, rb_node);
-		next = rb_next(&n->rb_node);
-
-		rb_erase(&n->rb_node, tree);
-		output__insert_entry(n);
-	}
-}
-
-static unsigned long total = 0,
-		     total_mmap = 0,
-		     total_comm = 0,
-		     total_fork = 0,
-		     total_unknown = 0;
-
 static int
 process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 {
 	char level;
-	int show = 0;
-	struct dso *dso = NULL;
-	struct thread *thread;
 	u64 ip = event->ip.ip;
 	struct map *map = NULL;
-
-	thread = threads__findnew(event->ip.pid, &threads, &last_match);
+	struct symbol *sym = NULL;
+	struct thread *thread = threads__findnew(event->ip.pid);
 
 	dump_printf("%p [%p]: PERF_EVENT (IP, %d): %d: %p\n",
 		(void *)(offset + head),
@@ -497,60 +104,53 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 		event->ip.pid,
 		(void *)(long)ip);
 
-	dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid);
-
 	if (thread == NULL) {
 		fprintf(stderr, "problem processing %d event, skipping it.\n",
 			event->header.type);
 		return -1;
 	}
 
-	if (event->header.misc & PERF_EVENT_MISC_KERNEL) {
-		show = SHOW_KERNEL;
-		level = 'k';
-
-		dso = kernel_dso;
-
-		dump_printf(" ...... dso: %s\n", dso->name);
-
-	} else if (event->header.misc & PERF_EVENT_MISC_USER) {
+	dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid);
 
-		show = SHOW_USER;
+	if (event->header.misc & PERF_RECORD_MISC_KERNEL) {
+		level = 'k';
+		sym = kernel_maps__find_symbol(ip, &map);
+		dump_printf(" ...... dso: %s\n",
+			    map ? map->dso->long_name : "<not found>");
+	} else if (event->header.misc & PERF_RECORD_MISC_USER) {
 		level = '.';
-
 		map = thread__find_map(thread, ip);
 		if (map != NULL) {
+got_map:
 			ip = map->map_ip(map, ip);
-			dso = map->dso;
+			sym = map->dso->find_symbol(map->dso, ip);
 		} else {
 			/*
 			 * 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 (which executes in user-mode):
+			 * 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)ip < 0)
-				dso = kernel_dso;
+			if ((long long)ip < 0) {
+				map = kernel_map;
+				goto got_map;
+			}
 		}
-		dump_printf(" ...... dso: %s\n", dso ? dso->name : "<not found>");
-
+		dump_printf(" ...... dso: %s\n",
+			    map ? map->dso->long_name : "<not found>");
 	} else {
-		show = SHOW_HV;
 		level = 'H';
 		dump_printf(" ...... dso: [hypervisor]\n");
 	}
 
-	if (show & show_mask) {
-		struct symbol *sym = NULL;
-
-		if (dso)
-			sym = dso->find_symbol(dso, ip);
-
-		if (hist_entry__add(thread, map, dso, sym, ip, level)) {
-			fprintf(stderr,
-		"problem incrementing symbol count, skipping event\n");
-			return -1;
-		}
+	if (hist_entry__add(thread, map, sym, ip, 1, level)) {
+		fprintf(stderr, "problem incrementing symbol count, "
+				"skipping event\n");
+		return -1;
 	}
 	total++;
 
@@ -560,12 +160,10 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 static int
 process_mmap_event(event_t *event, unsigned long offset, unsigned long head)
 {
-	struct thread *thread;
 	struct map *map = map__new(&event->mmap, NULL, 0);
+	struct thread *thread = threads__findnew(event->mmap.pid);
 
-	thread = threads__findnew(event->mmap.pid, &threads, &last_match);
-
-	dump_printf("%p [%p]: PERF_EVENT_MMAP %d: [%p(%p) @ %p]: %s\n",
+	dump_printf("%p [%p]: PERF_RECORD_MMAP %d: [%p(%p) @ %p]: %s\n",
 		(void *)(offset + head),
 		(void *)(long)(event->header.size),
 		event->mmap.pid,
@@ -575,7 +173,7 @@ process_mmap_event(event_t *event, unsigned long offset, unsigned long head)
 		event->mmap.filename);
 
 	if (thread == NULL || map == NULL) {
-		dump_printf("problem processing PERF_EVENT_MMAP, skipping event.\n");
+		dump_printf("problem processing PERF_RECORD_MMAP, skipping event.\n");
 		return 0;
 	}
 
@@ -588,17 +186,16 @@ process_mmap_event(event_t *event, unsigned long offset, unsigned long head)
 static int
 process_comm_event(event_t *event, unsigned long offset, unsigned long head)
 {
-	struct thread *thread;
+	struct thread *thread = threads__findnew(event->comm.pid);
 
-	thread = threads__findnew(event->comm.pid, &threads, &last_match);
-	dump_printf("%p [%p]: PERF_EVENT_COMM: %s:%d\n",
+	dump_printf("%p [%p]: PERF_RECORD_COMM: %s:%d\n",
 		(void *)(offset + head),
 		(void *)(long)(event->header.size),
 		event->comm.comm, event->comm.pid);
 
 	if (thread == NULL ||
 	    thread__set_comm(thread, event->comm.comm)) {
-		dump_printf("problem processing PERF_EVENT_COMM, skipping event.\n");
+		dump_printf("problem processing PERF_RECORD_COMM, skipping event.\n");
 		return -1;
 	}
 	total_comm++;
@@ -609,12 +206,10 @@ process_comm_event(event_t *event, unsigned long offset, unsigned long head)
 static int
 process_fork_event(event_t *event, unsigned long offset, unsigned long head)
 {
-	struct thread *thread;
-	struct thread *parent;
+	struct thread *thread = threads__findnew(event->fork.pid);
+	struct thread *parent = threads__findnew(event->fork.ppid);
 
-	thread = threads__findnew(event->fork.pid, &threads, &last_match);
-	parent = threads__findnew(event->fork.ppid, &threads, &last_match);
-	dump_printf("%p [%p]: PERF_EVENT_FORK: %d:%d\n",
+	dump_printf("%p [%p]: PERF_RECORD_FORK: %d:%d\n",
 		(void *)(offset + head),
 		(void *)(long)(event->header.size),
 		event->fork.pid, event->fork.ppid);
@@ -627,7 +222,7 @@ process_fork_event(event_t *event, unsigned long offset, unsigned long head)
 		return 0;
 
 	if (!thread || !parent || thread__fork(thread, parent)) {
-		dump_printf("problem processing PERF_EVENT_FORK, skipping event.\n");
+		dump_printf("problem processing PERF_RECORD_FORK, skipping event.\n");
 		return -1;
 	}
 	total_fork++;
@@ -639,23 +234,23 @@ static int
 process_event(event_t *event, unsigned long offset, unsigned long head)
 {
 	switch (event->header.type) {
-	case PERF_EVENT_SAMPLE:
+	case PERF_RECORD_SAMPLE:
 		return process_sample_event(event, offset, head);
 
-	case PERF_EVENT_MMAP:
+	case PERF_RECORD_MMAP:
 		return process_mmap_event(event, offset, head);
 
-	case PERF_EVENT_COMM:
+	case PERF_RECORD_COMM:
 		return process_comm_event(event, offset, head);
 
-	case PERF_EVENT_FORK:
+	case PERF_RECORD_FORK:
 		return process_fork_event(event, offset, head);
 	/*
 	 * We dont process them right now but they are fine:
 	 */
 
-	case PERF_EVENT_THROTTLE:
-	case PERF_EVENT_UNTHROTTLE:
+	case PERF_RECORD_THROTTLE:
+	case PERF_RECORD_UNTHROTTLE:
 		return 0;
 
 	default:
@@ -666,7 +261,7 @@ process_event(event_t *event, unsigned long offset, unsigned long head)
 }
 
 static int
-parse_line(FILE *file, struct symbol *sym, u64 start, u64 len)
+parse_line(FILE *file, struct symbol *sym, u64 len)
 {
 	char *line = NULL, *tmp, *tmp2;
 	static const char *prev_line;
@@ -716,7 +311,7 @@ parse_line(FILE *file, struct symbol *sym, u64 start, u64 len)
 		const char *color;
 		struct sym_ext *sym_ext = sym->priv;
 
-		offset = line_ip - start;
+		offset = line_ip - sym->start;
 		if (offset < len)
 			hits = sym->hist[offset];
 
@@ -795,7 +390,7 @@ static void free_source_line(struct symbol *sym, int len)
 
 /* Get the filename:line for the colored entries */
 static void
-get_source_line(struct symbol *sym, u64 start, int len, const char *filename)
+get_source_line(struct symbol *sym, int len, const char *filename)
 {
 	int i;
 	char cmd[PATH_MAX * 2];
@@ -820,7 +415,7 @@ get_source_line(struct symbol *sym, u64 start, int len, const char *filename)
 		if (sym_ext[i].percent <= 0.5)
 			continue;
 
-		offset = start + i;
+		offset = sym->start + i;
 		sprintf(cmd, "addr2line -e %s %016llx", filename, offset);
 		fp = popen(cmd, "r");
 		if (!fp)
@@ -872,31 +467,23 @@ static void print_summary(const char *filename)
 
 static void annotate_sym(struct dso *dso, struct symbol *sym)
 {
-	const char *filename = dso->name, *d_filename;
-	u64 start, end, len;
+	const char *filename = dso->long_name, *d_filename;
+	u64 len;
 	char command[PATH_MAX*2];
 	FILE *file;
 
 	if (!filename)
 		return;
-	if (sym->module)
-		filename = sym->module->path;
-	else if (dso == kernel_dso)
-		filename = vmlinux_name;
-
-	start = sym->obj_start;
-	if (!start)
-		start = sym->start;
+
 	if (full_paths)
 		d_filename = filename;
 	else
 		d_filename = basename(filename);
 
-	end = start + sym->end - sym->start + 1;
 	len = sym->end - sym->start;
 
 	if (print_line) {
-		get_source_line(sym, start, len, filename);
+		get_source_line(sym, len, filename);
 		print_summary(filename);
 	}
 
@@ -905,10 +492,11 @@ static void annotate_sym(struct dso *dso, struct symbol *sym)
 	printf("------------------------------------------------\n");
 
 	if (verbose >= 2)
-		printf("annotating [%p] %30s : [%p] %30s\n", dso, dso->name, sym, sym->name);
+		printf("annotating [%p] %30s : [%p] %30s\n",
+		       dso, dso->long_name, sym, sym->name);
 
 	sprintf(command, "objdump --start-address=0x%016Lx --stop-address=0x%016Lx -dS %s|grep -v %s",
-			(u64)start, (u64)end, filename, filename);
+		sym->start, sym->end, filename, filename);
 
 	if (verbose >= 3)
 		printf("doing: %s\n", command);
@@ -918,7 +506,7 @@ static void annotate_sym(struct dso *dso, struct symbol *sym)
 		return;
 
 	while (!feof(file)) {
-		if (parse_line(file, sym, start, len) < 0)
+		if (parse_line(file, sym, len) < 0)
 			break;
 	}
 
@@ -959,7 +547,7 @@ static int __cmd_annotate(void)
 	uint32_t size;
 	char *buf;
 
-	register_idle_thread(&threads, &last_match);
+	register_idle_thread();
 
 	input = open(input_name, O_RDONLY);
 	if (input < 0) {
@@ -1059,14 +647,14 @@ more:
 	if (dump_trace)
 		return 0;
 
-	if (verbose >= 3)
-		threads__fprintf(stdout, &threads);
+	if (verbose > 3)
+		threads__fprintf(stdout);
 
-	if (verbose >= 2)
+	if (verbose > 2)
 		dsos__fprintf(stdout);
 
 	collapse__resort();
-	output__resort();
+	output__resort(total);
 
 	find_annotations();
 
@@ -1139,5 +727,11 @@ int cmd_annotate(int argc, const char **argv, const char *prefix __used)
 
 	setup_pager();
 
+	if (field_sep && *field_sep == '.') {
+		fputs("'.' is the only non valid --field-separator argument\n",
+				stderr);
+		exit(129);
+	}
+
 	return __cmd_annotate();
 }
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 99a12fe86e9f..8b2c860c49a2 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -17,7 +17,6 @@
 #include "util/header.h"
 #include "util/event.h"
 #include "util/debug.h"
-#include "util/trace-event.h"
 
 #include <unistd.h>
 #include <sched.h>
@@ -27,42 +26,45 @@
 
 static int			fd[MAX_NR_CPUS][MAX_COUNTERS];
 
-static long			default_interval		= 100000;
+static long			default_interval		=      0;
 
-static int			nr_cpus				= 0;
+static int			nr_cpus				=      0;
 static unsigned int		page_size;
-static unsigned int		mmap_pages			= 128;
-static int			freq				= 0;
+static unsigned int		mmap_pages			=    128;
+static int			freq				=   1000;
 static int			output;
 static const char		*output_name			= "perf.data";
-static int			group				= 0;
-static unsigned int		realtime_prio			= 0;
-static int			raw_samples			= 0;
-static int			system_wide			= 0;
-static int			profile_cpu			= -1;
-static pid_t			target_pid			= -1;
-static int			inherit				= 1;
-static int			force				= 0;
-static int			append_file			= 0;
-static int			call_graph			= 0;
-static int			inherit_stat			= 0;
-static int			no_samples			= 0;
-static int			sample_address			= 0;
-
-static long			samples;
+static int			group				=      0;
+static unsigned int		realtime_prio			=      0;
+static int			raw_samples			=      0;
+static int			system_wide			=      0;
+static int			profile_cpu			=     -1;
+static pid_t			target_pid			=     -1;
+static pid_t			child_pid			=     -1;
+static int			inherit				=      1;
+static int			force				=      0;
+static int			append_file			=      0;
+static int			call_graph			=      0;
+static int			inherit_stat			=      0;
+static int			no_samples			=      0;
+static int			sample_address			=      0;
+static int			multiplex			=      0;
+static int			multiplex_fd			=     -1;
+
+static long			samples				=      0;
 static struct timeval		last_read;
 static struct timeval		this_read;
 
-static u64			bytes_written;
+static u64			bytes_written			=      0;
 
 static struct pollfd		event_array[MAX_NR_CPUS * MAX_COUNTERS];
 
-static int			nr_poll;
-static int			nr_cpu;
+static int			nr_poll				=      0;
+static int			nr_cpu				=      0;
 
-static int			file_new = 1;
+static int			file_new			=      1;
 
-struct perf_header		*header;
+struct perf_header		*header				=   NULL;
 
 struct mmap_data {
 	int			counter;
@@ -75,7 +77,7 @@ static struct mmap_data		mmap_array[MAX_NR_CPUS][MAX_COUNTERS];
 
 static unsigned long mmap_read_head(struct mmap_data *md)
 {
-	struct perf_counter_mmap_page *pc = md->base;
+	struct perf_event_mmap_page *pc = md->base;
 	long head;
 
 	head = pc->data_head;
@@ -86,7 +88,7 @@ static unsigned long mmap_read_head(struct mmap_data *md)
 
 static void mmap_write_tail(struct mmap_data *md, unsigned long tail)
 {
-	struct perf_counter_mmap_page *pc = md->base;
+	struct perf_event_mmap_page *pc = md->base;
 
 	/*
 	 * ensure all reads are done before we write the tail out.
@@ -182,6 +184,9 @@ static void sig_handler(int sig)
 
 static void sig_atexit(void)
 {
+	if (child_pid != -1)
+		kill(child_pid, SIGTERM);
+
 	if (signr == -1)
 		return;
 
@@ -231,7 +236,7 @@ static pid_t pid_synthesize_comm_event(pid_t pid, int full)
 		}
 	}
 
-	comm_ev.header.type = PERF_EVENT_COMM;
+	comm_ev.header.type = PERF_RECORD_COMM;
 	size = ALIGN(size, sizeof(u64));
 	comm_ev.header.size = sizeof(comm_ev) - (sizeof(comm_ev.comm) - size);
 
@@ -286,7 +291,7 @@ static void pid_synthesize_mmap_samples(pid_t pid, pid_t tgid)
 	while (1) {
 		char bf[BUFSIZ], *pbf = bf;
 		struct mmap_event mmap_ev = {
-			.header = { .type = PERF_EVENT_MMAP },
+			.header = { .type = PERF_RECORD_MMAP },
 		};
 		int n;
 		size_t size;
@@ -353,7 +358,7 @@ static void synthesize_all(void)
 
 static int group_fd;
 
-static struct perf_header_attr *get_header_attr(struct perf_counter_attr *a, int nr)
+static struct perf_header_attr *get_header_attr(struct perf_event_attr *a, int nr)
 {
 	struct perf_header_attr *h_attr;
 
@@ -369,9 +374,11 @@ static struct perf_header_attr *get_header_attr(struct perf_counter_attr *a, int
 
 static void create_counter(int counter, int cpu, pid_t pid)
 {
-	struct perf_counter_attr *attr = attrs + counter;
+	char *filter = filters[counter];
+	struct perf_event_attr *attr = attrs + counter;
 	struct perf_header_attr *h_attr;
 	int track = !counter; /* only the first counter needs these */
+	int ret;
 	struct {
 		u64 count;
 		u64 time_enabled;
@@ -415,7 +422,7 @@ static void create_counter(int counter, int cpu, pid_t pid)
 	attr->disabled		= 1;
 
 try_again:
-	fd[nr_cpu][counter] = sys_perf_counter_open(attr, pid, cpu, group_fd, 0);
+	fd[nr_cpu][counter] = sys_perf_event_open(attr, pid, cpu, group_fd, 0);
 
 	if (fd[nr_cpu][counter] < 0) {
 		int err = errno;
@@ -442,7 +449,7 @@ try_again:
 		printf("\n");
 		error("perfcounter syscall returned with %d (%s)\n",
 			fd[nr_cpu][counter], strerror(err));
-		die("No CONFIG_PERF_COUNTERS=y kernel support configured?\n");
+		die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
 		exit(-1);
 	}
 
@@ -470,22 +477,40 @@ try_again:
 	 */
 	if (group && group_fd == -1)
 		group_fd = fd[nr_cpu][counter];
+	if (multiplex && multiplex_fd == -1)
+		multiplex_fd = fd[nr_cpu][counter];
 
-	event_array[nr_poll].fd = fd[nr_cpu][counter];
-	event_array[nr_poll].events = POLLIN;
-	nr_poll++;
-
-	mmap_array[nr_cpu][counter].counter = counter;
-	mmap_array[nr_cpu][counter].prev = 0;
-	mmap_array[nr_cpu][counter].mask = mmap_pages*page_size - 1;
-	mmap_array[nr_cpu][counter].base = mmap(NULL, (mmap_pages+1)*page_size,
-			PROT_READ|PROT_WRITE, MAP_SHARED, fd[nr_cpu][counter], 0);
-	if (mmap_array[nr_cpu][counter].base == MAP_FAILED) {
-		error("failed to mmap with %d (%s)\n", errno, strerror(errno));
-		exit(-1);
+	if (multiplex && fd[nr_cpu][counter] != multiplex_fd) {
+
+		ret = ioctl(fd[nr_cpu][counter], PERF_EVENT_IOC_SET_OUTPUT, multiplex_fd);
+		assert(ret != -1);
+	} else {
+		event_array[nr_poll].fd = fd[nr_cpu][counter];
+		event_array[nr_poll].events = POLLIN;
+		nr_poll++;
+
+		mmap_array[nr_cpu][counter].counter = counter;
+		mmap_array[nr_cpu][counter].prev = 0;
+		mmap_array[nr_cpu][counter].mask = mmap_pages*page_size - 1;
+		mmap_array[nr_cpu][counter].base = mmap(NULL, (mmap_pages+1)*page_size,
+				PROT_READ|PROT_WRITE, MAP_SHARED, fd[nr_cpu][counter], 0);
+		if (mmap_array[nr_cpu][counter].base == MAP_FAILED) {
+			error("failed to mmap with %d (%s)\n", errno, strerror(errno));
+			exit(-1);
+		}
+	}
+
+	if (filter != NULL) {
+		ret = ioctl(fd[nr_cpu][counter],
+			    PERF_EVENT_IOC_SET_FILTER, filter);
+		if (ret) {
+			error("failed to set filter with %d (%s)\n", errno,
+			      strerror(errno));
+			exit(-1);
+		}
 	}
 
-	ioctl(fd[nr_cpu][counter], PERF_COUNTER_IOC_ENABLE);
+	ioctl(fd[nr_cpu][counter], PERF_EVENT_IOC_ENABLE);
 }
 
 static void open_counters(int cpu, pid_t pid)
@@ -513,6 +538,7 @@ static int __cmd_record(int argc, const char **argv)
 	pid_t pid = 0;
 	int flags;
 	int ret;
+	unsigned long waking = 0;
 
 	page_size = sysconf(_SC_PAGE_SIZE);
 	nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
@@ -550,17 +576,17 @@ static int __cmd_record(int argc, const char **argv)
 	else
 		header = perf_header__new();
 
-
 	if (raw_samples) {
-		read_tracing_data(attrs, nr_counters);
+		perf_header__set_trace_info();
 	} else {
 		for (i = 0; i < nr_counters; i++) {
 			if (attrs[i].sample_type & PERF_SAMPLE_RAW) {
-				read_tracing_data(attrs, nr_counters);
+				perf_header__set_trace_info();
 				break;
 			}
 		}
 	}
+
 	atexit(atexit_header);
 
 	if (!system_wide) {
@@ -598,6 +624,8 @@ static int __cmd_record(int argc, const char **argv)
 				exit(-1);
 			}
 		}
+
+		child_pid = pid;
 	}
 
 	if (realtime_prio) {
@@ -614,17 +642,29 @@ static int __cmd_record(int argc, const char **argv)
 		int hits = samples;
 
 		for (i = 0; i < nr_cpu; i++) {
-			for (counter = 0; counter < nr_counters; counter++)
-				mmap_read(&mmap_array[i][counter]);
+			for (counter = 0; counter < nr_counters; counter++) {
+				if (mmap_array[i][counter].base)
+					mmap_read(&mmap_array[i][counter]);
+			}
 		}
 
 		if (hits == samples) {
 			if (done)
 				break;
-			ret = poll(event_array, nr_poll, 100);
+			ret = poll(event_array, nr_poll, -1);
+			waking++;
+		}
+
+		if (done) {
+			for (i = 0; i < nr_cpu; i++) {
+				for (counter = 0; counter < nr_counters; counter++)
+					ioctl(fd[i][counter], PERF_EVENT_IOC_DISABLE);
+			}
 		}
 	}
 
+	fprintf(stderr, "[ perf record: Woken up %ld times to write data ]\n", waking);
+
 	/*
 	 * Approximate RIP event size: 24 bytes.
 	 */
@@ -647,6 +687,8 @@ static const struct option options[] = {
 	OPT_CALLBACK('e', "event", NULL, "event",
 		     "event selector. use 'perf list' to list available events",
 		     parse_events),
+	OPT_CALLBACK(0, "filter", NULL, "filter",
+		     "event filter", parse_filter),
 	OPT_INTEGER('p', "pid", &target_pid,
 		    "record events on existing pid"),
 	OPT_INTEGER('r', "realtime", &realtime_prio,
@@ -681,6 +723,8 @@ static const struct option options[] = {
 		    "Sample addresses"),
 	OPT_BOOLEAN('n', "no-samples", &no_samples,
 		    "don't sample"),
+	OPT_BOOLEAN('M', "multiplex", &multiplex,
+		    "multiplex counter output in a single channel"),
 	OPT_END()
 };
 
@@ -699,6 +743,18 @@ int cmd_record(int argc, const char **argv, const char *prefix __used)
 		attrs[0].config = PERF_COUNT_HW_CPU_CYCLES;
 	}
 
+	/*
+	 * User specified count overrides default frequency.
+	 */
+	if (default_interval)
+		freq = 0;
+	else if (freq) {
+		default_interval = freq;
+	} else {
+		fprintf(stderr, "frequency and count are zero, aborting\n");
+		exit(EXIT_FAILURE);
+	}
+
 	for (counter = 0; counter < nr_counters; counter++) {
 		if (attrs[counter].sample_period)
 			continue;
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index cdf9a8d27bb9..a4f8cc209151 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -26,20 +26,18 @@
 #include "util/parse-options.h"
 #include "util/parse-events.h"
 
+#include "util/data_map.h"
 #include "util/thread.h"
+#include "util/sort.h"
+#include "util/hist.h"
 
 static char		const *input_name = "perf.data";
 
-static char		default_sort_order[] = "comm,dso,symbol";
-static char		*sort_order = default_sort_order;
 static char		*dso_list_str, *comm_list_str, *sym_list_str,
 			*col_width_list_str;
 static struct strlist	*dso_list, *comm_list, *sym_list;
-static char		*field_sep;
 
 static int		force;
-static int		input;
-static int		show_mask = SHOW_KERNEL | SHOW_USER | SHOW_HV;
 
 static int		full_paths;
 static int		show_nr_samples;
@@ -50,368 +48,17 @@ static struct perf_read_values	show_threads_values;
 static char		default_pretty_printing_style[] = "normal";
 static char		*pretty_printing_style = default_pretty_printing_style;
 
-static unsigned long	page_size;
-static unsigned long	mmap_window = 32;
-
-static char		default_parent_pattern[] = "^sys_|^do_page_fault";
-static char		*parent_pattern = default_parent_pattern;
-static regex_t		parent_regex;
-
 static int		exclude_other = 1;
 
 static char		callchain_default_opt[] = "fractal,0.5";
 
-static int		callchain;
-
-static char		__cwd[PATH_MAX];
-static char		*cwd = __cwd;
+static char		*cwd;
 static int		cwdlen;
 
-static struct rb_root	threads;
-static struct thread	*last_match;
-
 static struct perf_header *header;
 
-static
-struct callchain_param	callchain_param = {
-	.mode	= CHAIN_GRAPH_REL,
-	.min_percent = 0.5
-};
-
 static u64		sample_type;
 
-static int repsep_fprintf(FILE *fp, const char *fmt, ...)
-{
-	int n;
-	va_list ap;
-
-	va_start(ap, fmt);
-	if (!field_sep)
-		n = vfprintf(fp, fmt, ap);
-	else {
-		char *bf = NULL;
-		n = vasprintf(&bf, fmt, ap);
-		if (n > 0) {
-			char *sep = bf;
-
-			while (1) {
-				sep = strchr(sep, *field_sep);
-				if (sep == NULL)
-					break;
-				*sep = '.';
-			}
-		}
-		fputs(bf, fp);
-		free(bf);
-	}
-	va_end(ap);
-	return n;
-}
-
-static unsigned int dsos__col_width,
-		    comms__col_width,
-		    threads__col_width;
-
-/*
- * histogram, sorted on item, collects counts
- */
-
-static struct rb_root hist;
-
-struct hist_entry {
-	struct rb_node		rb_node;
-
-	struct thread		*thread;
-	struct map		*map;
-	struct dso		*dso;
-	struct symbol		*sym;
-	struct symbol		*parent;
-	u64			ip;
-	char			level;
-	struct callchain_node	callchain;
-	struct rb_root		sorted_chain;
-
-	u64			count;
-};
-
-/*
- * configurable sorting bits
- */
-
-struct sort_entry {
-	struct list_head list;
-
-	const char *header;
-
-	int64_t (*cmp)(struct hist_entry *, struct hist_entry *);
-	int64_t (*collapse)(struct hist_entry *, struct hist_entry *);
-	size_t	(*print)(FILE *fp, struct hist_entry *, unsigned int width);
-	unsigned int *width;
-	bool	elide;
-};
-
-static int64_t cmp_null(void *l, void *r)
-{
-	if (!l && !r)
-		return 0;
-	else if (!l)
-		return -1;
-	else
-		return 1;
-}
-
-/* --sort pid */
-
-static int64_t
-sort__thread_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-	return right->thread->pid - left->thread->pid;
-}
-
-static size_t
-sort__thread_print(FILE *fp, struct hist_entry *self, unsigned int width)
-{
-	return repsep_fprintf(fp, "%*s:%5d", width - 6,
-			      self->thread->comm ?: "", self->thread->pid);
-}
-
-static struct sort_entry sort_thread = {
-	.header = "Command:  Pid",
-	.cmp	= sort__thread_cmp,
-	.print	= sort__thread_print,
-	.width	= &threads__col_width,
-};
-
-/* --sort comm */
-
-static int64_t
-sort__comm_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-	return right->thread->pid - left->thread->pid;
-}
-
-static 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);
-}
-
-static size_t
-sort__comm_print(FILE *fp, struct hist_entry *self, unsigned int width)
-{
-	return repsep_fprintf(fp, "%*s", width, self->thread->comm);
-}
-
-static struct sort_entry sort_comm = {
-	.header		= "Command",
-	.cmp		= sort__comm_cmp,
-	.collapse	= sort__comm_collapse,
-	.print		= sort__comm_print,
-	.width		= &comms__col_width,
-};
-
-/* --sort dso */
-
-static int64_t
-sort__dso_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-	struct dso *dso_l = left->dso;
-	struct dso *dso_r = right->dso;
-
-	if (!dso_l || !dso_r)
-		return cmp_null(dso_l, dso_r);
-
-	return strcmp(dso_l->name, dso_r->name);
-}
-
-static size_t
-sort__dso_print(FILE *fp, struct hist_entry *self, unsigned int width)
-{
-	if (self->dso)
-		return repsep_fprintf(fp, "%-*s", width, self->dso->name);
-
-	return repsep_fprintf(fp, "%*llx", width, (u64)self->ip);
-}
-
-static struct sort_entry sort_dso = {
-	.header = "Shared Object",
-	.cmp	= sort__dso_cmp,
-	.print	= sort__dso_print,
-	.width	= &dsos__col_width,
-};
-
-/* --sort symbol */
-
-static int64_t
-sort__sym_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-	u64 ip_l, ip_r;
-
-	if (left->sym == right->sym)
-		return 0;
-
-	ip_l = left->sym ? left->sym->start : left->ip;
-	ip_r = right->sym ? right->sym->start : right->ip;
-
-	return (int64_t)(ip_r - ip_l);
-}
-
-static size_t
-sort__sym_print(FILE *fp, struct hist_entry *self, unsigned int width __used)
-{
-	size_t ret = 0;
-
-	if (verbose)
-		ret += repsep_fprintf(fp, "%#018llx %c ", (u64)self->ip,
-				      dso__symtab_origin(self->dso));
-
-	ret += repsep_fprintf(fp, "[%c] ", self->level);
-	if (self->sym) {
-		ret += repsep_fprintf(fp, "%s", self->sym->name);
-
-		if (self->sym->module)
-			ret += repsep_fprintf(fp, "\t[%s]",
-					     self->sym->module->name);
-	} else {
-		ret += repsep_fprintf(fp, "%#016llx", (u64)self->ip);
-	}
-
-	return ret;
-}
-
-static struct sort_entry sort_sym = {
-	.header = "Symbol",
-	.cmp	= sort__sym_cmp,
-	.print	= sort__sym_print,
-};
-
-/* --sort parent */
-
-static 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 size_t
-sort__parent_print(FILE *fp, struct hist_entry *self, unsigned int width)
-{
-	return repsep_fprintf(fp, "%-*s", width,
-			      self->parent ? self->parent->name : "[other]");
-}
-
-static unsigned int parent_symbol__col_width;
-
-static struct sort_entry sort_parent = {
-	.header = "Parent symbol",
-	.cmp	= sort__parent_cmp,
-	.print	= sort__parent_print,
-	.width	= &parent_symbol__col_width,
-};
-
-static int sort__need_collapse = 0;
-static int sort__has_parent = 0;
-
-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,	},
-};
-
-static LIST_HEAD(hist_entry__sort_list);
-
-static 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->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));
-				fprintf(stderr, "Invalid regex: %s\n%s",
-					parent_pattern, err);
-				exit(-1);
-			}
-			sort__has_parent = 1;
-		}
-
-		list_add_tail(&sd->entry->list, &hist_entry__sort_list);
-		sd->taken = 1;
-
-		return 0;
-	}
-
-	return -ESRCH;
-}
-
-static 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->cmp(left, right);
-		if (cmp)
-			break;
-	}
-
-	return cmp;
-}
-
-static 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->collapse ?: se->cmp;
-
-		cmp = f(left, right);
-		if (cmp)
-			break;
-	}
-
-	return cmp;
-}
-
 static size_t ipchain__fprintf_graph_line(FILE *fp, int depth, int depth_mask)
 {
 	int i;
@@ -610,7 +257,6 @@ hist_entry_callchain__fprintf(FILE *fp, struct hist_entry *self,
 	return ret;
 }
 
-
 static size_t
 hist_entry__fprintf(FILE *fp, struct hist_entry *self, u64 total_samples)
 {
@@ -695,22 +341,17 @@ static int thread__set_comm_adjust(struct thread *self, const char *comm)
 
 
 static struct symbol *
-resolve_symbol(struct thread *thread, struct map **mapp,
-	       struct dso **dsop, u64 *ipp)
+resolve_symbol(struct thread *thread, struct map **mapp, u64 *ipp)
 {
-	struct dso *dso = dsop ? *dsop : NULL;
 	struct map *map = mapp ? *mapp : NULL;
 	u64 ip = *ipp;
 
-	if (!thread)
-		return NULL;
-
-	if (dso)
-		goto got_dso;
-
 	if (map)
 		goto got_map;
 
+	if (!thread)
+		return NULL;
+
 	map = thread__find_map(thread, ip);
 	if (map != NULL) {
 		/*
@@ -725,29 +366,26 @@ resolve_symbol(struct thread *thread, struct map **mapp,
 			*mapp = map;
 got_map:
 		ip = map->map_ip(map, ip);
-
-		dso = map->dso;
 	} else {
 		/*
 		 * 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 (which executes in user-mode):
+		 * 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)ip < 0)
-		dso = kernel_dso;
+			return kernel_maps__find_symbol(ip, mapp);
 	}
-	dump_printf(" ...... dso: %s\n", dso ? dso->name : "<not found>");
+	dump_printf(" ...... dso: %s\n",
+		    map ? map->dso->long_name : "<not found>");
 	dump_printf(" ...... map: %Lx -> %Lx\n", *ipp, ip);
 	*ipp  = ip;
 
-	if (dsop)
-		*dsop = dso;
-
-	if (!dso)
-		return NULL;
-got_dso:
-	return dso->find_symbol(dso, ip);
+	return map ? map->dso->find_symbol(map->dso, ip) : NULL;
 }
 
 static int call__match(struct symbol *sym)
@@ -758,9 +396,9 @@ static int call__match(struct symbol *sym)
 	return 0;
 }
 
-static struct symbol **
-resolve_callchain(struct thread *thread, struct map *map __used,
-		    struct ip_callchain *chain, struct hist_entry *entry)
+static struct symbol **resolve_callchain(struct thread *thread, struct map *map,
+					 struct ip_callchain *chain,
+					 struct symbol **parent)
 {
 	u64 context = PERF_CONTEXT_MAX;
 	struct symbol **syms = NULL;
@@ -776,8 +414,7 @@ resolve_callchain(struct thread *thread, struct map *map __used,
 
 	for (i = 0; i < chain->nr; i++) {
 		u64 ip = chain->ips[i];
-		struct dso *dso = NULL;
-		struct symbol *sym;
+		struct symbol *sym = NULL;
 
 		if (ip >= PERF_CONTEXT_MAX) {
 			context = ip;
@@ -786,21 +423,18 @@ resolve_callchain(struct thread *thread, struct map *map __used,
 
 		switch (context) {
 		case PERF_CONTEXT_HV:
-			dso = hypervisor_dso;
 			break;
 		case PERF_CONTEXT_KERNEL:
-			dso = kernel_dso;
+			sym = kernel_maps__find_symbol(ip, &map);
 			break;
 		default:
+			sym = resolve_symbol(thread, &map, &ip);
 			break;
 		}
 
-		sym = resolve_symbol(thread, NULL, &dso, &ip);
-
 		if (sym) {
-			if (sort__has_parent && call__match(sym) &&
-			    !entry->parent)
-				entry->parent = sym;
+			if (sort__has_parent && !*parent && call__match(sym))
+				*parent = sym;
 			if (!callchain)
 				break;
 			syms[i] = sym;
@@ -815,177 +449,35 @@ resolve_callchain(struct thread *thread, struct map *map __used,
  */
 
 static int
-hist_entry__add(struct thread *thread, struct map *map, struct dso *dso,
+hist_entry__add(struct thread *thread, struct map *map,
 		struct symbol *sym, u64 ip, struct ip_callchain *chain,
 		char level, u64 count)
 {
-	struct rb_node **p = &hist.rb_node;
-	struct rb_node *parent = NULL;
+	struct symbol **syms = NULL, *parent = NULL;
+	bool hit;
 	struct hist_entry *he;
-	struct symbol **syms = NULL;
-	struct hist_entry entry = {
-		.thread	= thread,
-		.map	= map,
-		.dso	= dso,
-		.sym	= sym,
-		.ip	= ip,
-		.level	= level,
-		.count	= count,
-		.parent = NULL,
-		.sorted_chain = RB_ROOT
-	};
-	int cmp;
 
 	if ((sort__has_parent || callchain) && chain)
-		syms = resolve_callchain(thread, map, chain, &entry);
-
-	while (*p != NULL) {
-		parent = *p;
-		he = rb_entry(parent, struct hist_entry, rb_node);
-
-		cmp = hist_entry__cmp(&entry, he);
+		syms = resolve_callchain(thread, map, chain, &parent);
 
-		if (!cmp) {
-			he->count += count;
-			if (callchain) {
-				append_chain(&he->callchain, chain, syms);
-				free(syms);
-			}
-			return 0;
-		}
+	he = __hist_entry__add(thread, map, sym, parent,
+			       ip, count, level, &hit);
+	if (he == NULL)
+		return -ENOMEM;
 
-		if (cmp < 0)
-			p = &(*p)->rb_left;
-		else
-			p = &(*p)->rb_right;
-	}
+	if (hit)
+		he->count += count;
 
-	he = malloc(sizeof(*he));
-	if (!he)
-		return -ENOMEM;
-	*he = entry;
 	if (callchain) {
-		callchain_init(&he->callchain);
+		if (!hit)
+			callchain_init(&he->callchain);
 		append_chain(&he->callchain, chain, syms);
 		free(syms);
 	}
-	rb_link_node(&he->rb_node, parent, p);
-	rb_insert_color(&he->rb_node, &hist);
 
 	return 0;
 }
 
-static void hist_entry__free(struct hist_entry *he)
-{
-	free(he);
-}
-
-/*
- * collapse the histogram
- */
-
-static struct rb_root collapse_hists;
-
-static void collapse__insert_entry(struct hist_entry *he)
-{
-	struct rb_node **p = &collapse_hists.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->count += he->count;
-			hist_entry__free(he);
-			return;
-		}
-
-		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, &collapse_hists);
-}
-
-static void collapse__resort(void)
-{
-	struct rb_node *next;
-	struct hist_entry *n;
-
-	if (!sort__need_collapse)
-		return;
-
-	next = rb_first(&hist);
-	while (next) {
-		n = rb_entry(next, struct hist_entry, rb_node);
-		next = rb_next(&n->rb_node);
-
-		rb_erase(&n->rb_node, &hist);
-		collapse__insert_entry(n);
-	}
-}
-
-/*
- * reverse the map, sort on count.
- */
-
-static struct rb_root output_hists;
-
-static void output__insert_entry(struct hist_entry *he, u64 min_callchain_hits)
-{
-	struct rb_node **p = &output_hists.rb_node;
-	struct rb_node *parent = NULL;
-	struct hist_entry *iter;
-
-	if (callchain)
-		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->count > iter->count)
-			p = &(*p)->rb_left;
-		else
-			p = &(*p)->rb_right;
-	}
-
-	rb_link_node(&he->rb_node, parent, p);
-	rb_insert_color(&he->rb_node, &output_hists);
-}
-
-static void output__resort(u64 total_samples)
-{
-	struct rb_node *next;
-	struct hist_entry *n;
-	struct rb_root *tree = &hist;
-	u64 min_callchain_hits;
-
-	min_callchain_hits = total_samples * (callchain_param.min_percent / 100);
-
-	if (sort__need_collapse)
-		tree = &collapse_hists;
-
-	next = rb_first(tree);
-
-	while (next) {
-		n = rb_entry(next, struct hist_entry, rb_node);
-		next = rb_next(&n->rb_node);
-
-		rb_erase(&n->rb_node, tree);
-		output__insert_entry(n, min_callchain_hits);
-	}
-}
-
 static size_t output__fprintf(FILE *fp, u64 total_samples)
 {
 	struct hist_entry *pos;
@@ -1080,13 +572,6 @@ print_entries:
 	return ret;
 }
 
-static unsigned long total = 0,
-		     total_mmap = 0,
-		     total_comm = 0,
-		     total_fork = 0,
-		     total_unknown = 0,
-		     total_lost = 0;
-
 static int validate_chain(struct ip_callchain *chain, event_t *event)
 {
 	unsigned int chain_size;
@@ -1104,24 +589,21 @@ static int
 process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 {
 	char level;
-	int show = 0;
-	struct dso *dso = NULL;
-	struct thread *thread;
+	struct symbol *sym = NULL;
 	u64 ip = event->ip.ip;
 	u64 period = 1;
 	struct map *map = NULL;
 	void *more_data = event->ip.__more_data;
 	struct ip_callchain *chain = NULL;
 	int cpumode;
-
-	thread = threads__findnew(event->ip.pid, &threads, &last_match);
+	struct thread *thread = threads__findnew(event->ip.pid);
 
 	if (sample_type & PERF_SAMPLE_PERIOD) {
 		period = *(u64 *)more_data;
 		more_data += sizeof(u64);
 	}
 
-	dump_printf("%p [%p]: PERF_EVENT_SAMPLE (IP, %d): %d/%d: %p period: %Ld\n",
+	dump_printf("%p [%p]: PERF_RECORD_SAMPLE (IP, %d): %d/%d: %p period: %Ld\n",
 		(void *)(offset + head),
 		(void *)(long)(event->header.size),
 		event->header.misc,
@@ -1147,56 +629,49 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 		}
 	}
 
-	dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid);
-
 	if (thread == NULL) {
 		eprintf("problem processing %d event, skipping it.\n",
 			event->header.type);
 		return -1;
 	}
 
+	dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid);
+
 	if (comm_list && !strlist__has_entry(comm_list, thread->comm))
 		return 0;
 
-	cpumode = event->header.misc & PERF_EVENT_MISC_CPUMODE_MASK;
+	cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
 
-	if (cpumode == PERF_EVENT_MISC_KERNEL) {
-		show = SHOW_KERNEL;
+	if (cpumode == PERF_RECORD_MISC_KERNEL) {
 		level = 'k';
-
-		dso = kernel_dso;
-
-		dump_printf(" ...... dso: %s\n", dso->name);
-
-	} else if (cpumode == PERF_EVENT_MISC_USER) {
-
-		show = SHOW_USER;
+		sym = kernel_maps__find_symbol(ip, &map);
+		dump_printf(" ...... dso: %s\n",
+			    map ? map->dso->long_name : "<not found>");
+	} else if (cpumode == PERF_RECORD_MISC_USER) {
 		level = '.';
+		sym = resolve_symbol(thread, &map, &ip);
 
 	} else {
-		show = SHOW_HV;
 		level = 'H';
-
-		dso = hypervisor_dso;
-
 		dump_printf(" ...... dso: [hypervisor]\n");
 	}
 
-	if (show & show_mask) {
-		struct symbol *sym = resolve_symbol(thread, &map, &dso, &ip);
-
-		if (dso_list && (!dso || !dso->name ||
-				 !strlist__has_entry(dso_list, dso->name)))
-			return 0;
+	if (dso_list &&
+	    (!map || !map->dso ||
+	     !(strlist__has_entry(dso_list, map->dso->short_name) ||
+	       (map->dso->short_name != map->dso->long_name &&
+		strlist__has_entry(dso_list, map->dso->long_name)))))
+		return 0;
 
-		if (sym_list && (!sym || !strlist__has_entry(sym_list, sym->name)))
-			return 0;
+	if (sym_list && sym && !strlist__has_entry(sym_list, sym->name))
+		return 0;
 
-		if (hist_entry__add(thread, map, dso, sym, ip, chain, level, period)) {
-			eprintf("problem incrementing symbol count, skipping event\n");
-			return -1;
-		}
+	if (hist_entry__add(thread, map, sym, ip,
+			    chain, level, period)) {
+		eprintf("problem incrementing symbol count, skipping event\n");
+		return -1;
 	}
+
 	total += period;
 
 	return 0;
@@ -1205,12 +680,10 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 static int
 process_mmap_event(event_t *event, unsigned long offset, unsigned long head)
 {
-	struct thread *thread;
 	struct map *map = map__new(&event->mmap, cwd, cwdlen);
+	struct thread *thread = threads__findnew(event->mmap.pid);
 
-	thread = threads__findnew(event->mmap.pid, &threads, &last_match);
-
-	dump_printf("%p [%p]: PERF_EVENT_MMAP %d/%d: [%p(%p) @ %p]: %s\n",
+	dump_printf("%p [%p]: PERF_RECORD_MMAP %d/%d: [%p(%p) @ %p]: %s\n",
 		(void *)(offset + head),
 		(void *)(long)(event->header.size),
 		event->mmap.pid,
@@ -1221,7 +694,7 @@ process_mmap_event(event_t *event, unsigned long offset, unsigned long head)
 		event->mmap.filename);
 
 	if (thread == NULL || map == NULL) {
-		dump_printf("problem processing PERF_EVENT_MMAP, skipping event.\n");
+		dump_printf("problem processing PERF_RECORD_MMAP, skipping event.\n");
 		return 0;
 	}
 
@@ -1234,18 +707,16 @@ process_mmap_event(event_t *event, unsigned long offset, unsigned long head)
 static int
 process_comm_event(event_t *event, unsigned long offset, unsigned long head)
 {
-	struct thread *thread;
+	struct thread *thread = threads__findnew(event->comm.pid);
 
-	thread = threads__findnew(event->comm.pid, &threads, &last_match);
-
-	dump_printf("%p [%p]: PERF_EVENT_COMM: %s:%d\n",
+	dump_printf("%p [%p]: PERF_RECORD_COMM: %s:%d\n",
 		(void *)(offset + head),
 		(void *)(long)(event->header.size),
 		event->comm.comm, event->comm.pid);
 
 	if (thread == NULL ||
 	    thread__set_comm_adjust(thread, event->comm.comm)) {
-		dump_printf("problem processing PERF_EVENT_COMM, skipping event.\n");
+		dump_printf("problem processing PERF_RECORD_COMM, skipping event.\n");
 		return -1;
 	}
 	total_comm++;
@@ -1256,16 +727,13 @@ process_comm_event(event_t *event, unsigned long offset, unsigned long head)
 static int
 process_task_event(event_t *event, unsigned long offset, unsigned long head)
 {
-	struct thread *thread;
-	struct thread *parent;
-
-	thread = threads__findnew(event->fork.pid, &threads, &last_match);
-	parent = threads__findnew(event->fork.ppid, &threads, &last_match);
+	struct thread *thread = threads__findnew(event->fork.pid);
+	struct thread *parent = threads__findnew(event->fork.ppid);
 
-	dump_printf("%p [%p]: PERF_EVENT_%s: (%d:%d):(%d:%d)\n",
+	dump_printf("%p [%p]: PERF_RECORD_%s: (%d:%d):(%d:%d)\n",
 		(void *)(offset + head),
 		(void *)(long)(event->header.size),
-		event->header.type == PERF_EVENT_FORK ? "FORK" : "EXIT",
+		event->header.type == PERF_RECORD_FORK ? "FORK" : "EXIT",
 		event->fork.pid, event->fork.tid,
 		event->fork.ppid, event->fork.ptid);
 
@@ -1276,11 +744,11 @@ process_task_event(event_t *event, unsigned long offset, unsigned long head)
 	if (thread == parent)
 		return 0;
 
-	if (event->header.type == PERF_EVENT_EXIT)
+	if (event->header.type == PERF_RECORD_EXIT)
 		return 0;
 
 	if (!thread || !parent || thread__fork(thread, parent)) {
-		dump_printf("problem processing PERF_EVENT_FORK, skipping event.\n");
+		dump_printf("problem processing PERF_RECORD_FORK, skipping event.\n");
 		return -1;
 	}
 	total_fork++;
@@ -1291,7 +759,7 @@ process_task_event(event_t *event, unsigned long offset, unsigned long head)
 static int
 process_lost_event(event_t *event, unsigned long offset, unsigned long head)
 {
-	dump_printf("%p [%p]: PERF_EVENT_LOST: id:%Ld: lost:%Ld\n",
+	dump_printf("%p [%p]: PERF_RECORD_LOST: id:%Ld: lost:%Ld\n",
 		(void *)(offset + head),
 		(void *)(long)(event->header.size),
 		event->lost.id,
@@ -1305,7 +773,7 @@ process_lost_event(event_t *event, unsigned long offset, unsigned long head)
 static int
 process_read_event(event_t *event, unsigned long offset, unsigned long head)
 {
-	struct perf_counter_attr *attr;
+	struct perf_event_attr *attr;
 
 	attr = perf_header__find_attr(event->read.id, header);
 
@@ -1319,7 +787,7 @@ process_read_event(event_t *event, unsigned long offset, unsigned long head)
 					   event->read.value);
 	}
 
-	dump_printf("%p [%p]: PERF_EVENT_READ: %d %d %s %Lu\n",
+	dump_printf("%p [%p]: PERF_RECORD_READ: %d %d %s %Lu\n",
 			(void *)(offset + head),
 			(void *)(long)(event->header.size),
 			event->read.pid,
@@ -1331,216 +799,79 @@ process_read_event(event_t *event, unsigned long offset, unsigned long head)
 	return 0;
 }
 
-static int
-process_event(event_t *event, unsigned long offset, unsigned long head)
+static int sample_type_check(u64 type)
 {
-	trace_event(event);
-
-	switch (event->header.type) {
-	case PERF_EVENT_SAMPLE:
-		return process_sample_event(event, offset, head);
-
-	case PERF_EVENT_MMAP:
-		return process_mmap_event(event, offset, head);
-
-	case PERF_EVENT_COMM:
-		return process_comm_event(event, offset, head);
-
-	case PERF_EVENT_FORK:
-	case PERF_EVENT_EXIT:
-		return process_task_event(event, offset, head);
-
-	case PERF_EVENT_LOST:
-		return process_lost_event(event, offset, head);
-
-	case PERF_EVENT_READ:
-		return process_read_event(event, offset, head);
-
-	/*
-	 * We dont process them right now but they are fine:
-	 */
-
-	case PERF_EVENT_THROTTLE:
-	case PERF_EVENT_UNTHROTTLE:
-		return 0;
-
-	default:
-		return -1;
-	}
-
-	return 0;
-}
-
-static int __cmd_report(void)
-{
-	int ret, rc = EXIT_FAILURE;
-	unsigned long offset = 0;
-	unsigned long head, shift;
-	struct stat input_stat;
-	struct thread *idle;
-	event_t *event;
-	uint32_t size;
-	char *buf;
-
-	idle = register_idle_thread(&threads, &last_match);
-	thread__comm_adjust(idle);
-
-	if (show_threads)
-		perf_read_values_init(&show_threads_values);
-
-	input = open(input_name, O_RDONLY);
-	if (input < 0) {
-		fprintf(stderr, " failed to open file: %s", input_name);
-		if (!strcmp(input_name, "perf.data"))
-			fprintf(stderr, "  (try 'perf record' first)");
-		fprintf(stderr, "\n");
-		exit(-1);
-	}
-
-	ret = fstat(input, &input_stat);
-	if (ret < 0) {
-		perror("failed to stat file");
-		exit(-1);
-	}
-
-	if (!force && input_stat.st_uid && (input_stat.st_uid != geteuid())) {
-		fprintf(stderr, "file: %s not owned by current user or root\n", input_name);
-		exit(-1);
-	}
-
-	if (!input_stat.st_size) {
-		fprintf(stderr, "zero-sized file, nothing to do!\n");
-		exit(0);
-	}
-
-	header = perf_header__read(input);
-	head = header->data_offset;
-
-	sample_type = perf_header__sample_type(header);
+	sample_type = type;
 
 	if (!(sample_type & PERF_SAMPLE_CALLCHAIN)) {
 		if (sort__has_parent) {
 			fprintf(stderr, "selected --sort parent, but no"
 					" callchain data. Did you call"
 					" perf record without -g?\n");
-			exit(-1);
+			return -1;
 		}
 		if (callchain) {
 			fprintf(stderr, "selected -g but no callchain data."
 					" Did you call perf record without"
 					" -g?\n");
-			exit(-1);
+			return -1;
 		}
 	} else if (callchain_param.mode != CHAIN_NONE && !callchain) {
 			callchain = 1;
 			if (register_callchain_param(&callchain_param) < 0) {
 				fprintf(stderr, "Can't register callchain"
 						" params\n");
-				exit(-1);
+				return -1;
 			}
 	}
 
-	if (load_kernel() < 0) {
-		perror("failed to load kernel symbols");
-		return EXIT_FAILURE;
-	}
-
-	if (!full_paths) {
-		if (getcwd(__cwd, sizeof(__cwd)) == NULL) {
-			perror("failed to get the current directory");
-			return EXIT_FAILURE;
-		}
-		cwdlen = strlen(cwd);
-	} else {
-		cwd = NULL;
-		cwdlen = 0;
-	}
-
-	shift = page_size * (head / page_size);
-	offset += shift;
-	head -= shift;
-
-remap:
-	buf = (char *)mmap(NULL, page_size * mmap_window, PROT_READ,
-			   MAP_SHARED, input, offset);
-	if (buf == MAP_FAILED) {
-		perror("failed to mmap file");
-		exit(-1);
-	}
-
-more:
-	event = (event_t *)(buf + head);
-
-	size = event->header.size;
-	if (!size)
-		size = 8;
-
-	if (head + event->header.size >= page_size * mmap_window) {
-		int munmap_ret;
-
-		shift = page_size * (head / page_size);
-
-		munmap_ret = munmap(buf, page_size * mmap_window);
-		assert(munmap_ret == 0);
-
-		offset += shift;
-		head -= shift;
-		goto remap;
-	}
-
-	size = event->header.size;
-
-	dump_printf("\n%p [%p]: event: %d\n",
-			(void *)(offset + head),
-			(void *)(long)event->header.size,
-			event->header.type);
-
-	if (!size || process_event(event, offset, head) < 0) {
-
-		dump_printf("%p [%p]: skipping unknown header type: %d\n",
-			(void *)(offset + head),
-			(void *)(long)(event->header.size),
-			event->header.type);
-
-		total_unknown++;
+	return 0;
+}
 
-		/*
-		 * assume we lost track of the stream, check alignment, and
-		 * increment a single u64 in the hope to catch on again 'soon'.
-		 */
+static struct perf_file_handler file_handler = {
+	.process_sample_event	= process_sample_event,
+	.process_mmap_event	= process_mmap_event,
+	.process_comm_event	= process_comm_event,
+	.process_exit_event	= process_task_event,
+	.process_fork_event	= process_task_event,
+	.process_lost_event	= process_lost_event,
+	.process_read_event	= process_read_event,
+	.sample_type_check	= sample_type_check,
+};
 
-		if (unlikely(head & 7))
-			head &= ~7ULL;
 
-		size = 8;
-	}
+static int __cmd_report(void)
+{
+	struct thread *idle;
+	int ret;
 
-	head += size;
+	idle = register_idle_thread();
+	thread__comm_adjust(idle);
 
-	if (offset + head >= header->data_offset + header->data_size)
-		goto done;
+	if (show_threads)
+		perf_read_values_init(&show_threads_values);
 
-	if (offset + head < (unsigned long)input_stat.st_size)
-		goto more;
+	register_perf_file_handler(&file_handler);
 
-done:
-	rc = EXIT_SUCCESS;
-	close(input);
+	ret = mmap_dispatch_perf_file(&header, input_name, force, full_paths,
+				      &cwdlen, &cwd);
+	if (ret)
+		return ret;
 
 	dump_printf("      IP events: %10ld\n", total);
 	dump_printf("    mmap events: %10ld\n", total_mmap);
 	dump_printf("    comm events: %10ld\n", total_comm);
 	dump_printf("    fork events: %10ld\n", total_fork);
 	dump_printf("    lost events: %10ld\n", total_lost);
-	dump_printf(" unknown events: %10ld\n", total_unknown);
+	dump_printf(" unknown events: %10ld\n", file_handler.total_unknown);
 
 	if (dump_trace)
 		return 0;
 
-	if (verbose >= 3)
-		threads__fprintf(stdout, &threads);
+	if (verbose > 3)
+		threads__fprintf(stdout);
 
-	if (verbose >= 2)
+	if (verbose > 2)
 		dsos__fprintf(stdout);
 
 	collapse__resort();
@@ -1550,7 +881,7 @@ done:
 	if (show_threads)
 		perf_read_values_destroy(&show_threads_values);
 
-	return rc;
+	return ret;
 }
 
 static int
@@ -1606,7 +937,8 @@ setup:
 	return 0;
 }
 
-static const char * const report_usage[] = {
+//static const char * const report_usage[] = {
+const char * const report_usage[] = {
 	"perf report [<options>] <command>",
 	NULL
 };
@@ -1692,8 +1024,6 @@ int cmd_report(int argc, const char **argv, const char *prefix __used)
 {
 	symbol__init();
 
-	page_size = getpagesize();
-
 	argc = parse_options(argc, argv, options, report_usage, 0);
 
 	setup_sorting();
diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c
new file mode 100644
index 000000000000..57ad3f458ef5
--- /dev/null
+++ b/tools/perf/builtin-sched.c
@@ -0,0 +1,1982 @@
+#include "builtin.h"
+#include "perf.h"
+
+#include "util/util.h"
+#include "util/cache.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/header.h"
+
+#include "util/parse-options.h"
+#include "util/trace-event.h"
+
+#include "util/debug.h"
+#include "util/data_map.h"
+
+#include <sys/types.h>
+#include <sys/prctl.h>
+
+#include <semaphore.h>
+#include <pthread.h>
+#include <math.h>
+
+static char			const *input_name = "perf.data";
+
+static unsigned long		total_comm = 0;
+
+static struct perf_header	*header;
+static u64			sample_type;
+
+static char			default_sort_order[] = "avg, max, switch, runtime";
+static char			*sort_order = default_sort_order;
+
+static int			profile_cpu = -1;
+
+static char			*cwd;
+static int			cwdlen;
+
+#define PR_SET_NAME		15               /* Set process name */
+#define MAX_CPUS		4096
+
+#define BUG_ON(x)		assert(!(x))
+
+static u64			run_measurement_overhead;
+static u64			sleep_measurement_overhead;
+
+#define COMM_LEN		20
+#define SYM_LEN			129
+
+#define MAX_PID			65536
+
+static unsigned long		nr_tasks;
+
+struct sched_atom;
+
+struct task_desc {
+	unsigned long		nr;
+	unsigned long		pid;
+	char			comm[COMM_LEN];
+
+	unsigned long		nr_events;
+	unsigned long		curr_event;
+	struct sched_atom	**atoms;
+
+	pthread_t		thread;
+	sem_t			sleep_sem;
+
+	sem_t			ready_for_work;
+	sem_t			work_done_sem;
+
+	u64			cpu_usage;
+};
+
+enum sched_event_type {
+	SCHED_EVENT_RUN,
+	SCHED_EVENT_SLEEP,
+	SCHED_EVENT_WAKEUP,
+	SCHED_EVENT_MIGRATION,
+};
+
+struct sched_atom {
+	enum sched_event_type	type;
+	u64			timestamp;
+	u64			duration;
+	unsigned long		nr;
+	int			specific_wait;
+	sem_t			*wait_sem;
+	struct task_desc	*wakee;
+};
+
+static struct task_desc		*pid_to_task[MAX_PID];
+
+static struct task_desc		**tasks;
+
+static pthread_mutex_t		start_work_mutex = PTHREAD_MUTEX_INITIALIZER;
+static u64			start_time;
+
+static pthread_mutex_t		work_done_wait_mutex = PTHREAD_MUTEX_INITIALIZER;
+
+static unsigned long		nr_run_events;
+static unsigned long		nr_sleep_events;
+static unsigned long		nr_wakeup_events;
+
+static unsigned long		nr_sleep_corrections;
+static unsigned long		nr_run_events_optimized;
+
+static unsigned long		targetless_wakeups;
+static unsigned long		multitarget_wakeups;
+
+static u64			cpu_usage;
+static u64			runavg_cpu_usage;
+static u64			parent_cpu_usage;
+static u64			runavg_parent_cpu_usage;
+
+static unsigned long		nr_runs;
+static u64			sum_runtime;
+static u64			sum_fluct;
+static u64			run_avg;
+
+static unsigned long		replay_repeat = 10;
+static unsigned long		nr_timestamps;
+static unsigned long		nr_unordered_timestamps;
+static unsigned long		nr_state_machine_bugs;
+static unsigned long		nr_context_switch_bugs;
+static unsigned long		nr_events;
+static unsigned long		nr_lost_chunks;
+static unsigned long		nr_lost_events;
+
+#define TASK_STATE_TO_CHAR_STR "RSDTtZX"
+
+enum thread_state {
+	THREAD_SLEEPING = 0,
+	THREAD_WAIT_CPU,
+	THREAD_SCHED_IN,
+	THREAD_IGNORE
+};
+
+struct work_atom {
+	struct list_head	list;
+	enum thread_state	state;
+	u64			sched_out_time;
+	u64			wake_up_time;
+	u64			sched_in_time;
+	u64			runtime;
+};
+
+struct work_atoms {
+	struct list_head	work_list;
+	struct thread		*thread;
+	struct rb_node		node;
+	u64			max_lat;
+	u64			total_lat;
+	u64			nb_atoms;
+	u64			total_runtime;
+};
+
+typedef int (*sort_fn_t)(struct work_atoms *, struct work_atoms *);
+
+static struct rb_root		atom_root, sorted_atom_root;
+
+static u64			all_runtime;
+static u64			all_count;
+
+
+static u64 get_nsecs(void)
+{
+	struct timespec ts;
+
+	clock_gettime(CLOCK_MONOTONIC, &ts);
+
+	return ts.tv_sec * 1000000000ULL + ts.tv_nsec;
+}
+
+static void burn_nsecs(u64 nsecs)
+{
+	u64 T0 = get_nsecs(), T1;
+
+	do {
+		T1 = get_nsecs();
+	} while (T1 + run_measurement_overhead < T0 + nsecs);
+}
+
+static void sleep_nsecs(u64 nsecs)
+{
+	struct timespec ts;
+
+	ts.tv_nsec = nsecs % 999999999;
+	ts.tv_sec = nsecs / 999999999;
+
+	nanosleep(&ts, NULL);
+}
+
+static void calibrate_run_measurement_overhead(void)
+{
+	u64 T0, T1, delta, min_delta = 1000000000ULL;
+	int i;
+
+	for (i = 0; i < 10; i++) {
+		T0 = get_nsecs();
+		burn_nsecs(0);
+		T1 = get_nsecs();
+		delta = T1-T0;
+		min_delta = min(min_delta, delta);
+	}
+	run_measurement_overhead = min_delta;
+
+	printf("run measurement overhead: %Ld nsecs\n", min_delta);
+}
+
+static void calibrate_sleep_measurement_overhead(void)
+{
+	u64 T0, T1, delta, min_delta = 1000000000ULL;
+	int i;
+
+	for (i = 0; i < 10; i++) {
+		T0 = get_nsecs();
+		sleep_nsecs(10000);
+		T1 = get_nsecs();
+		delta = T1-T0;
+		min_delta = min(min_delta, delta);
+	}
+	min_delta -= 10000;
+	sleep_measurement_overhead = min_delta;
+
+	printf("sleep measurement overhead: %Ld nsecs\n", min_delta);
+}
+
+static struct sched_atom *
+get_new_event(struct task_desc *task, u64 timestamp)
+{
+	struct sched_atom *event = calloc(1, sizeof(*event));
+	unsigned long idx = task->nr_events;
+	size_t size;
+
+	event->timestamp = timestamp;
+	event->nr = idx;
+
+	task->nr_events++;
+	size = sizeof(struct sched_atom *) * task->nr_events;
+	task->atoms = realloc(task->atoms, size);
+	BUG_ON(!task->atoms);
+
+	task->atoms[idx] = event;
+
+	return event;
+}
+
+static struct sched_atom *last_event(struct task_desc *task)
+{
+	if (!task->nr_events)
+		return NULL;
+
+	return task->atoms[task->nr_events - 1];
+}
+
+static void
+add_sched_event_run(struct task_desc *task, u64 timestamp, u64 duration)
+{
+	struct sched_atom *event, *curr_event = last_event(task);
+
+	/*
+	 * optimize an existing RUN event by merging this one
+	 * to it:
+	 */
+	if (curr_event && curr_event->type == SCHED_EVENT_RUN) {
+		nr_run_events_optimized++;
+		curr_event->duration += duration;
+		return;
+	}
+
+	event = get_new_event(task, timestamp);
+
+	event->type = SCHED_EVENT_RUN;
+	event->duration = duration;
+
+	nr_run_events++;
+}
+
+static void
+add_sched_event_wakeup(struct task_desc *task, u64 timestamp,
+		       struct task_desc *wakee)
+{
+	struct sched_atom *event, *wakee_event;
+
+	event = get_new_event(task, timestamp);
+	event->type = SCHED_EVENT_WAKEUP;
+	event->wakee = wakee;
+
+	wakee_event = last_event(wakee);
+	if (!wakee_event || wakee_event->type != SCHED_EVENT_SLEEP) {
+		targetless_wakeups++;
+		return;
+	}
+	if (wakee_event->wait_sem) {
+		multitarget_wakeups++;
+		return;
+	}
+
+	wakee_event->wait_sem = calloc(1, sizeof(*wakee_event->wait_sem));
+	sem_init(wakee_event->wait_sem, 0, 0);
+	wakee_event->specific_wait = 1;
+	event->wait_sem = wakee_event->wait_sem;
+
+	nr_wakeup_events++;
+}
+
+static void
+add_sched_event_sleep(struct task_desc *task, u64 timestamp,
+		      u64 task_state __used)
+{
+	struct sched_atom *event = get_new_event(task, timestamp);
+
+	event->type = SCHED_EVENT_SLEEP;
+
+	nr_sleep_events++;
+}
+
+static struct task_desc *register_pid(unsigned long pid, const char *comm)
+{
+	struct task_desc *task;
+
+	BUG_ON(pid >= MAX_PID);
+
+	task = pid_to_task[pid];
+
+	if (task)
+		return task;
+
+	task = calloc(1, sizeof(*task));
+	task->pid = pid;
+	task->nr = nr_tasks;
+	strcpy(task->comm, comm);
+	/*
+	 * every task starts in sleeping state - this gets ignored
+	 * if there's no wakeup pointing to this sleep state:
+	 */
+	add_sched_event_sleep(task, 0, 0);
+
+	pid_to_task[pid] = task;
+	nr_tasks++;
+	tasks = realloc(tasks, nr_tasks*sizeof(struct task_task *));
+	BUG_ON(!tasks);
+	tasks[task->nr] = task;
+
+	if (verbose)
+		printf("registered task #%ld, PID %ld (%s)\n", nr_tasks, pid, comm);
+
+	return task;
+}
+
+
+static void print_task_traces(void)
+{
+	struct task_desc *task;
+	unsigned long i;
+
+	for (i = 0; i < nr_tasks; i++) {
+		task = tasks[i];
+		printf("task %6ld (%20s:%10ld), nr_events: %ld\n",
+			task->nr, task->comm, task->pid, task->nr_events);
+	}
+}
+
+static void add_cross_task_wakeups(void)
+{
+	struct task_desc *task1, *task2;
+	unsigned long i, j;
+
+	for (i = 0; i < nr_tasks; i++) {
+		task1 = tasks[i];
+		j = i + 1;
+		if (j == nr_tasks)
+			j = 0;
+		task2 = tasks[j];
+		add_sched_event_wakeup(task1, 0, task2);
+	}
+}
+
+static void
+process_sched_event(struct task_desc *this_task __used, struct sched_atom *atom)
+{
+	int ret = 0;
+	u64 now;
+	long long delta;
+
+	now = get_nsecs();
+	delta = start_time + atom->timestamp - now;
+
+	switch (atom->type) {
+		case SCHED_EVENT_RUN:
+			burn_nsecs(atom->duration);
+			break;
+		case SCHED_EVENT_SLEEP:
+			if (atom->wait_sem)
+				ret = sem_wait(atom->wait_sem);
+			BUG_ON(ret);
+			break;
+		case SCHED_EVENT_WAKEUP:
+			if (atom->wait_sem)
+				ret = sem_post(atom->wait_sem);
+			BUG_ON(ret);
+			break;
+		case SCHED_EVENT_MIGRATION:
+			break;
+		default:
+			BUG_ON(1);
+	}
+}
+
+static u64 get_cpu_usage_nsec_parent(void)
+{
+	struct rusage ru;
+	u64 sum;
+	int err;
+
+	err = getrusage(RUSAGE_SELF, &ru);
+	BUG_ON(err);
+
+	sum =  ru.ru_utime.tv_sec*1e9 + ru.ru_utime.tv_usec*1e3;
+	sum += ru.ru_stime.tv_sec*1e9 + ru.ru_stime.tv_usec*1e3;
+
+	return sum;
+}
+
+static u64 get_cpu_usage_nsec_self(void)
+{
+	char filename [] = "/proc/1234567890/sched";
+	unsigned long msecs, nsecs;
+	char *line = NULL;
+	u64 total = 0;
+	size_t len = 0;
+	ssize_t chars;
+	FILE *file;
+	int ret;
+
+	sprintf(filename, "/proc/%d/sched", getpid());
+	file = fopen(filename, "r");
+	BUG_ON(!file);
+
+	while ((chars = getline(&line, &len, file)) != -1) {
+		ret = sscanf(line, "se.sum_exec_runtime : %ld.%06ld\n",
+			&msecs, &nsecs);
+		if (ret == 2) {
+			total = msecs*1e6 + nsecs;
+			break;
+		}
+	}
+	if (line)
+		free(line);
+	fclose(file);
+
+	return total;
+}
+
+static void *thread_func(void *ctx)
+{
+	struct task_desc *this_task = ctx;
+	u64 cpu_usage_0, cpu_usage_1;
+	unsigned long i, ret;
+	char comm2[22];
+
+	sprintf(comm2, ":%s", this_task->comm);
+	prctl(PR_SET_NAME, comm2);
+
+again:
+	ret = sem_post(&this_task->ready_for_work);
+	BUG_ON(ret);
+	ret = pthread_mutex_lock(&start_work_mutex);
+	BUG_ON(ret);
+	ret = pthread_mutex_unlock(&start_work_mutex);
+	BUG_ON(ret);
+
+	cpu_usage_0 = get_cpu_usage_nsec_self();
+
+	for (i = 0; i < this_task->nr_events; i++) {
+		this_task->curr_event = i;
+		process_sched_event(this_task, this_task->atoms[i]);
+	}
+
+	cpu_usage_1 = get_cpu_usage_nsec_self();
+	this_task->cpu_usage = cpu_usage_1 - cpu_usage_0;
+
+	ret = sem_post(&this_task->work_done_sem);
+	BUG_ON(ret);
+
+	ret = pthread_mutex_lock(&work_done_wait_mutex);
+	BUG_ON(ret);
+	ret = pthread_mutex_unlock(&work_done_wait_mutex);
+	BUG_ON(ret);
+
+	goto again;
+}
+
+static void create_tasks(void)
+{
+	struct task_desc *task;
+	pthread_attr_t attr;
+	unsigned long i;
+	int err;
+
+	err = pthread_attr_init(&attr);
+	BUG_ON(err);
+	err = pthread_attr_setstacksize(&attr, (size_t)(16*1024));
+	BUG_ON(err);
+	err = pthread_mutex_lock(&start_work_mutex);
+	BUG_ON(err);
+	err = pthread_mutex_lock(&work_done_wait_mutex);
+	BUG_ON(err);
+	for (i = 0; i < nr_tasks; i++) {
+		task = tasks[i];
+		sem_init(&task->sleep_sem, 0, 0);
+		sem_init(&task->ready_for_work, 0, 0);
+		sem_init(&task->work_done_sem, 0, 0);
+		task->curr_event = 0;
+		err = pthread_create(&task->thread, &attr, thread_func, task);
+		BUG_ON(err);
+	}
+}
+
+static void wait_for_tasks(void)
+{
+	u64 cpu_usage_0, cpu_usage_1;
+	struct task_desc *task;
+	unsigned long i, ret;
+
+	start_time = get_nsecs();
+	cpu_usage = 0;
+	pthread_mutex_unlock(&work_done_wait_mutex);
+
+	for (i = 0; i < nr_tasks; i++) {
+		task = tasks[i];
+		ret = sem_wait(&task->ready_for_work);
+		BUG_ON(ret);
+		sem_init(&task->ready_for_work, 0, 0);
+	}
+	ret = pthread_mutex_lock(&work_done_wait_mutex);
+	BUG_ON(ret);
+
+	cpu_usage_0 = get_cpu_usage_nsec_parent();
+
+	pthread_mutex_unlock(&start_work_mutex);
+
+	for (i = 0; i < nr_tasks; i++) {
+		task = tasks[i];
+		ret = sem_wait(&task->work_done_sem);
+		BUG_ON(ret);
+		sem_init(&task->work_done_sem, 0, 0);
+		cpu_usage += task->cpu_usage;
+		task->cpu_usage = 0;
+	}
+
+	cpu_usage_1 = get_cpu_usage_nsec_parent();
+	if (!runavg_cpu_usage)
+		runavg_cpu_usage = cpu_usage;
+	runavg_cpu_usage = (runavg_cpu_usage*9 + cpu_usage)/10;
+
+	parent_cpu_usage = cpu_usage_1 - cpu_usage_0;
+	if (!runavg_parent_cpu_usage)
+		runavg_parent_cpu_usage = parent_cpu_usage;
+	runavg_parent_cpu_usage = (runavg_parent_cpu_usage*9 +
+				   parent_cpu_usage)/10;
+
+	ret = pthread_mutex_lock(&start_work_mutex);
+	BUG_ON(ret);
+
+	for (i = 0; i < nr_tasks; i++) {
+		task = tasks[i];
+		sem_init(&task->sleep_sem, 0, 0);
+		task->curr_event = 0;
+	}
+}
+
+static void run_one_test(void)
+{
+	u64 T0, T1, delta, avg_delta, fluct, std_dev;
+
+	T0 = get_nsecs();
+	wait_for_tasks();
+	T1 = get_nsecs();
+
+	delta = T1 - T0;
+	sum_runtime += delta;
+	nr_runs++;
+
+	avg_delta = sum_runtime / nr_runs;
+	if (delta < avg_delta)
+		fluct = avg_delta - delta;
+	else
+		fluct = delta - avg_delta;
+	sum_fluct += fluct;
+	std_dev = sum_fluct / nr_runs / sqrt(nr_runs);
+	if (!run_avg)
+		run_avg = delta;
+	run_avg = (run_avg*9 + delta)/10;
+
+	printf("#%-3ld: %0.3f, ",
+		nr_runs, (double)delta/1000000.0);
+
+	printf("ravg: %0.2f, ",
+		(double)run_avg/1e6);
+
+	printf("cpu: %0.2f / %0.2f",
+		(double)cpu_usage/1e6, (double)runavg_cpu_usage/1e6);
+
+#if 0
+	/*
+	 * rusage statistics done by the parent, these are less
+	 * accurate than the sum_exec_runtime based statistics:
+	 */
+	printf(" [%0.2f / %0.2f]",
+		(double)parent_cpu_usage/1e6,
+		(double)runavg_parent_cpu_usage/1e6);
+#endif
+
+	printf("\n");
+
+	if (nr_sleep_corrections)
+		printf(" (%ld sleep corrections)\n", nr_sleep_corrections);
+	nr_sleep_corrections = 0;
+}
+
+static void test_calibrations(void)
+{
+	u64 T0, T1;
+
+	T0 = get_nsecs();
+	burn_nsecs(1e6);
+	T1 = get_nsecs();
+
+	printf("the run test took %Ld nsecs\n", T1-T0);
+
+	T0 = get_nsecs();
+	sleep_nsecs(1e6);
+	T1 = get_nsecs();
+
+	printf("the sleep test took %Ld nsecs\n", T1-T0);
+}
+
+static int
+process_comm_event(event_t *event, unsigned long offset, unsigned long head)
+{
+	struct thread *thread = threads__findnew(event->comm.tid);
+
+	dump_printf("%p [%p]: perf_event_comm: %s:%d\n",
+		(void *)(offset + head),
+		(void *)(long)(event->header.size),
+		event->comm.comm, event->comm.pid);
+
+	if (thread == NULL ||
+	    thread__set_comm(thread, event->comm.comm)) {
+		dump_printf("problem processing perf_event_comm, skipping event.\n");
+		return -1;
+	}
+	total_comm++;
+
+	return 0;
+}
+
+
+struct raw_event_sample {
+	u32 size;
+	char data[0];
+};
+
+#define FILL_FIELD(ptr, field, event, data)	\
+	ptr.field = (typeof(ptr.field)) raw_field_value(event, #field, data)
+
+#define FILL_ARRAY(ptr, array, event, data)			\
+do {								\
+	void *__array = raw_field_ptr(event, #array, data);	\
+	memcpy(ptr.array, __array, sizeof(ptr.array));	\
+} while(0)
+
+#define FILL_COMMON_FIELDS(ptr, event, data)			\
+do {								\
+	FILL_FIELD(ptr, common_type, event, data);		\
+	FILL_FIELD(ptr, common_flags, event, data);		\
+	FILL_FIELD(ptr, common_preempt_count, event, data);	\
+	FILL_FIELD(ptr, common_pid, event, data);		\
+	FILL_FIELD(ptr, common_tgid, event, data);		\
+} while (0)
+
+
+
+struct trace_switch_event {
+	u32 size;
+
+	u16 common_type;
+	u8 common_flags;
+	u8 common_preempt_count;
+	u32 common_pid;
+	u32 common_tgid;
+
+	char prev_comm[16];
+	u32 prev_pid;
+	u32 prev_prio;
+	u64 prev_state;
+	char next_comm[16];
+	u32 next_pid;
+	u32 next_prio;
+};
+
+struct trace_runtime_event {
+	u32 size;
+
+	u16 common_type;
+	u8 common_flags;
+	u8 common_preempt_count;
+	u32 common_pid;
+	u32 common_tgid;
+
+	char comm[16];
+	u32 pid;
+	u64 runtime;
+	u64 vruntime;
+};
+
+struct trace_wakeup_event {
+	u32 size;
+
+	u16 common_type;
+	u8 common_flags;
+	u8 common_preempt_count;
+	u32 common_pid;
+	u32 common_tgid;
+
+	char comm[16];
+	u32 pid;
+
+	u32 prio;
+	u32 success;
+	u32 cpu;
+};
+
+struct trace_fork_event {
+	u32 size;
+
+	u16 common_type;
+	u8 common_flags;
+	u8 common_preempt_count;
+	u32 common_pid;
+	u32 common_tgid;
+
+	char parent_comm[16];
+	u32 parent_pid;
+	char child_comm[16];
+	u32 child_pid;
+};
+
+struct trace_migrate_task_event {
+	u32 size;
+
+	u16 common_type;
+	u8 common_flags;
+	u8 common_preempt_count;
+	u32 common_pid;
+	u32 common_tgid;
+
+	char comm[16];
+	u32 pid;
+
+	u32 prio;
+	u32 cpu;
+};
+
+struct trace_sched_handler {
+	void (*switch_event)(struct trace_switch_event *,
+			     struct event *,
+			     int cpu,
+			     u64 timestamp,
+			     struct thread *thread);
+
+	void (*runtime_event)(struct trace_runtime_event *,
+			      struct event *,
+			      int cpu,
+			      u64 timestamp,
+			      struct thread *thread);
+
+	void (*wakeup_event)(struct trace_wakeup_event *,
+			     struct event *,
+			     int cpu,
+			     u64 timestamp,
+			     struct thread *thread);
+
+	void (*fork_event)(struct trace_fork_event *,
+			   struct event *,
+			   int cpu,
+			   u64 timestamp,
+			   struct thread *thread);
+
+	void (*migrate_task_event)(struct trace_migrate_task_event *,
+			   struct event *,
+			   int cpu,
+			   u64 timestamp,
+			   struct thread *thread);
+};
+
+
+static void
+replay_wakeup_event(struct trace_wakeup_event *wakeup_event,
+		    struct event *event,
+		    int cpu __used,
+		    u64 timestamp __used,
+		    struct thread *thread __used)
+{
+	struct task_desc *waker, *wakee;
+
+	if (verbose) {
+		printf("sched_wakeup event %p\n", event);
+
+		printf(" ... pid %d woke up %s/%d\n",
+			wakeup_event->common_pid,
+			wakeup_event->comm,
+			wakeup_event->pid);
+	}
+
+	waker = register_pid(wakeup_event->common_pid, "<unknown>");
+	wakee = register_pid(wakeup_event->pid, wakeup_event->comm);
+
+	add_sched_event_wakeup(waker, timestamp, wakee);
+}
+
+static u64 cpu_last_switched[MAX_CPUS];
+
+static void
+replay_switch_event(struct trace_switch_event *switch_event,
+		    struct event *event,
+		    int cpu,
+		    u64 timestamp,
+		    struct thread *thread __used)
+{
+	struct task_desc *prev, *next;
+	u64 timestamp0;
+	s64 delta;
+
+	if (verbose)
+		printf("sched_switch event %p\n", event);
+
+	if (cpu >= MAX_CPUS || cpu < 0)
+		return;
+
+	timestamp0 = cpu_last_switched[cpu];
+	if (timestamp0)
+		delta = timestamp - timestamp0;
+	else
+		delta = 0;
+
+	if (delta < 0)
+		die("hm, delta: %Ld < 0 ?\n", delta);
+
+	if (verbose) {
+		printf(" ... switch from %s/%d to %s/%d [ran %Ld nsecs]\n",
+			switch_event->prev_comm, switch_event->prev_pid,
+			switch_event->next_comm, switch_event->next_pid,
+			delta);
+	}
+
+	prev = register_pid(switch_event->prev_pid, switch_event->prev_comm);
+	next = register_pid(switch_event->next_pid, switch_event->next_comm);
+
+	cpu_last_switched[cpu] = timestamp;
+
+	add_sched_event_run(prev, timestamp, delta);
+	add_sched_event_sleep(prev, timestamp, switch_event->prev_state);
+}
+
+
+static void
+replay_fork_event(struct trace_fork_event *fork_event,
+		  struct event *event,
+		  int cpu __used,
+		  u64 timestamp __used,
+		  struct thread *thread __used)
+{
+	if (verbose) {
+		printf("sched_fork event %p\n", event);
+		printf("... parent: %s/%d\n", fork_event->parent_comm, fork_event->parent_pid);
+		printf("...  child: %s/%d\n", fork_event->child_comm, fork_event->child_pid);
+	}
+	register_pid(fork_event->parent_pid, fork_event->parent_comm);
+	register_pid(fork_event->child_pid, fork_event->child_comm);
+}
+
+static struct trace_sched_handler replay_ops  = {
+	.wakeup_event		= replay_wakeup_event,
+	.switch_event		= replay_switch_event,
+	.fork_event		= replay_fork_event,
+};
+
+struct sort_dimension {
+	const char		*name;
+	sort_fn_t		cmp;
+	struct list_head	list;
+};
+
+static LIST_HEAD(cmp_pid);
+
+static int
+thread_lat_cmp(struct list_head *list, struct work_atoms *l, struct work_atoms *r)
+{
+	struct sort_dimension *sort;
+	int ret = 0;
+
+	BUG_ON(list_empty(list));
+
+	list_for_each_entry(sort, list, list) {
+		ret = sort->cmp(l, r);
+		if (ret)
+			return ret;
+	}
+
+	return ret;
+}
+
+static struct work_atoms *
+thread_atoms_search(struct rb_root *root, struct thread *thread,
+			 struct list_head *sort_list)
+{
+	struct rb_node *node = root->rb_node;
+	struct work_atoms key = { .thread = thread };
+
+	while (node) {
+		struct work_atoms *atoms;
+		int cmp;
+
+		atoms = container_of(node, struct work_atoms, node);
+
+		cmp = thread_lat_cmp(sort_list, &key, atoms);
+		if (cmp > 0)
+			node = node->rb_left;
+		else if (cmp < 0)
+			node = node->rb_right;
+		else {
+			BUG_ON(thread != atoms->thread);
+			return atoms;
+		}
+	}
+	return NULL;
+}
+
+static void
+__thread_latency_insert(struct rb_root *root, struct work_atoms *data,
+			 struct list_head *sort_list)
+{
+	struct rb_node **new = &(root->rb_node), *parent = NULL;
+
+	while (*new) {
+		struct work_atoms *this;
+		int cmp;
+
+		this = container_of(*new, struct work_atoms, node);
+		parent = *new;
+
+		cmp = thread_lat_cmp(sort_list, data, this);
+
+		if (cmp > 0)
+			new = &((*new)->rb_left);
+		else
+			new = &((*new)->rb_right);
+	}
+
+	rb_link_node(&data->node, parent, new);
+	rb_insert_color(&data->node, root);
+}
+
+static void thread_atoms_insert(struct thread *thread)
+{
+	struct work_atoms *atoms;
+
+	atoms = calloc(sizeof(*atoms), 1);
+	if (!atoms)
+		die("No memory");
+
+	atoms->thread = thread;
+	INIT_LIST_HEAD(&atoms->work_list);
+	__thread_latency_insert(&atom_root, atoms, &cmp_pid);
+}
+
+static void
+latency_fork_event(struct trace_fork_event *fork_event __used,
+		   struct event *event __used,
+		   int cpu __used,
+		   u64 timestamp __used,
+		   struct thread *thread __used)
+{
+	/* should insert the newcomer */
+}
+
+__used
+static char sched_out_state(struct trace_switch_event *switch_event)
+{
+	const char *str = TASK_STATE_TO_CHAR_STR;
+
+	return str[switch_event->prev_state];
+}
+
+static void
+add_sched_out_event(struct work_atoms *atoms,
+		    char run_state,
+		    u64 timestamp)
+{
+	struct work_atom *atom;
+
+	atom = calloc(sizeof(*atom), 1);
+	if (!atom)
+		die("Non memory");
+
+	atom->sched_out_time = timestamp;
+
+	if (run_state == 'R') {
+		atom->state = THREAD_WAIT_CPU;
+		atom->wake_up_time = atom->sched_out_time;
+	}
+
+	list_add_tail(&atom->list, &atoms->work_list);
+}
+
+static void
+add_runtime_event(struct work_atoms *atoms, u64 delta, u64 timestamp __used)
+{
+	struct work_atom *atom;
+
+	BUG_ON(list_empty(&atoms->work_list));
+
+	atom = list_entry(atoms->work_list.prev, struct work_atom, list);
+
+	atom->runtime += delta;
+	atoms->total_runtime += delta;
+}
+
+static void
+add_sched_in_event(struct work_atoms *atoms, u64 timestamp)
+{
+	struct work_atom *atom;
+	u64 delta;
+
+	if (list_empty(&atoms->work_list))
+		return;
+
+	atom = list_entry(atoms->work_list.prev, struct work_atom, list);
+
+	if (atom->state != THREAD_WAIT_CPU)
+		return;
+
+	if (timestamp < atom->wake_up_time) {
+		atom->state = THREAD_IGNORE;
+		return;
+	}
+
+	atom->state = THREAD_SCHED_IN;
+	atom->sched_in_time = timestamp;
+
+	delta = atom->sched_in_time - atom->wake_up_time;
+	atoms->total_lat += delta;
+	if (delta > atoms->max_lat)
+		atoms->max_lat = delta;
+	atoms->nb_atoms++;
+}
+
+static void
+latency_switch_event(struct trace_switch_event *switch_event,
+		     struct event *event __used,
+		     int cpu,
+		     u64 timestamp,
+		     struct thread *thread __used)
+{
+	struct work_atoms *out_events, *in_events;
+	struct thread *sched_out, *sched_in;
+	u64 timestamp0;
+	s64 delta;
+
+	BUG_ON(cpu >= MAX_CPUS || cpu < 0);
+
+	timestamp0 = cpu_last_switched[cpu];
+	cpu_last_switched[cpu] = timestamp;
+	if (timestamp0)
+		delta = timestamp - timestamp0;
+	else
+		delta = 0;
+
+	if (delta < 0)
+		die("hm, delta: %Ld < 0 ?\n", delta);
+
+
+	sched_out = threads__findnew(switch_event->prev_pid);
+	sched_in = threads__findnew(switch_event->next_pid);
+
+	out_events = thread_atoms_search(&atom_root, sched_out, &cmp_pid);
+	if (!out_events) {
+		thread_atoms_insert(sched_out);
+		out_events = thread_atoms_search(&atom_root, sched_out, &cmp_pid);
+		if (!out_events)
+			die("out-event: Internal tree error");
+	}
+	add_sched_out_event(out_events, sched_out_state(switch_event), timestamp);
+
+	in_events = thread_atoms_search(&atom_root, sched_in, &cmp_pid);
+	if (!in_events) {
+		thread_atoms_insert(sched_in);
+		in_events = thread_atoms_search(&atom_root, sched_in, &cmp_pid);
+		if (!in_events)
+			die("in-event: Internal tree error");
+		/*
+		 * Take came in we have not heard about yet,
+		 * add in an initial atom in runnable state:
+		 */
+		add_sched_out_event(in_events, 'R', timestamp);
+	}
+	add_sched_in_event(in_events, timestamp);
+}
+
+static void
+latency_runtime_event(struct trace_runtime_event *runtime_event,
+		     struct event *event __used,
+		     int cpu,
+		     u64 timestamp,
+		     struct thread *this_thread __used)
+{
+	struct thread *thread = threads__findnew(runtime_event->pid);
+	struct work_atoms *atoms = thread_atoms_search(&atom_root, thread, &cmp_pid);
+
+	BUG_ON(cpu >= MAX_CPUS || cpu < 0);
+	if (!atoms) {
+		thread_atoms_insert(thread);
+		atoms = thread_atoms_search(&atom_root, thread, &cmp_pid);
+		if (!atoms)
+			die("in-event: Internal tree error");
+		add_sched_out_event(atoms, 'R', timestamp);
+	}
+
+	add_runtime_event(atoms, runtime_event->runtime, timestamp);
+}
+
+static void
+latency_wakeup_event(struct trace_wakeup_event *wakeup_event,
+		     struct event *__event __used,
+		     int cpu __used,
+		     u64 timestamp,
+		     struct thread *thread __used)
+{
+	struct work_atoms *atoms;
+	struct work_atom *atom;
+	struct thread *wakee;
+
+	/* Note for later, it may be interesting to observe the failing cases */
+	if (!wakeup_event->success)
+		return;
+
+	wakee = threads__findnew(wakeup_event->pid);
+	atoms = thread_atoms_search(&atom_root, wakee, &cmp_pid);
+	if (!atoms) {
+		thread_atoms_insert(wakee);
+		atoms = thread_atoms_search(&atom_root, wakee, &cmp_pid);
+		if (!atoms)
+			die("wakeup-event: Internal tree error");
+		add_sched_out_event(atoms, 'S', timestamp);
+	}
+
+	BUG_ON(list_empty(&atoms->work_list));
+
+	atom = list_entry(atoms->work_list.prev, struct work_atom, list);
+
+	/*
+	 * You WILL be missing events if you've recorded only
+	 * one CPU, or are only looking at only one, so don't
+	 * make useless noise.
+	 */
+	if (profile_cpu == -1 && atom->state != THREAD_SLEEPING)
+		nr_state_machine_bugs++;
+
+	nr_timestamps++;
+	if (atom->sched_out_time > timestamp) {
+		nr_unordered_timestamps++;
+		return;
+	}
+
+	atom->state = THREAD_WAIT_CPU;
+	atom->wake_up_time = timestamp;
+}
+
+static void
+latency_migrate_task_event(struct trace_migrate_task_event *migrate_task_event,
+		     struct event *__event __used,
+		     int cpu __used,
+		     u64 timestamp,
+		     struct thread *thread __used)
+{
+	struct work_atoms *atoms;
+	struct work_atom *atom;
+	struct thread *migrant;
+
+	/*
+	 * Only need to worry about migration when profiling one CPU.
+	 */
+	if (profile_cpu == -1)
+		return;
+
+	migrant = threads__findnew(migrate_task_event->pid);
+	atoms = thread_atoms_search(&atom_root, migrant, &cmp_pid);
+	if (!atoms) {
+		thread_atoms_insert(migrant);
+		register_pid(migrant->pid, migrant->comm);
+		atoms = thread_atoms_search(&atom_root, migrant, &cmp_pid);
+		if (!atoms)
+			die("migration-event: Internal tree error");
+		add_sched_out_event(atoms, 'R', timestamp);
+	}
+
+	BUG_ON(list_empty(&atoms->work_list));
+
+	atom = list_entry(atoms->work_list.prev, struct work_atom, list);
+	atom->sched_in_time = atom->sched_out_time = atom->wake_up_time = timestamp;
+
+	nr_timestamps++;
+
+	if (atom->sched_out_time > timestamp)
+		nr_unordered_timestamps++;
+}
+
+static struct trace_sched_handler lat_ops  = {
+	.wakeup_event		= latency_wakeup_event,
+	.switch_event		= latency_switch_event,
+	.runtime_event		= latency_runtime_event,
+	.fork_event		= latency_fork_event,
+	.migrate_task_event	= latency_migrate_task_event,
+};
+
+static void output_lat_thread(struct work_atoms *work_list)
+{
+	int i;
+	int ret;
+	u64 avg;
+
+	if (!work_list->nb_atoms)
+		return;
+	/*
+	 * Ignore idle threads:
+	 */
+	if (!strcmp(work_list->thread->comm, "swapper"))
+		return;
+
+	all_runtime += work_list->total_runtime;
+	all_count += work_list->nb_atoms;
+
+	ret = printf("  %s:%d ", work_list->thread->comm, work_list->thread->pid);
+
+	for (i = 0; i < 24 - ret; i++)
+		printf(" ");
+
+	avg = work_list->total_lat / work_list->nb_atoms;
+
+	printf("|%11.3f ms |%9llu | avg:%9.3f ms | max:%9.3f ms |\n",
+	      (double)work_list->total_runtime / 1e6,
+		 work_list->nb_atoms, (double)avg / 1e6,
+		 (double)work_list->max_lat / 1e6);
+}
+
+static int pid_cmp(struct work_atoms *l, struct work_atoms *r)
+{
+	if (l->thread->pid < r->thread->pid)
+		return -1;
+	if (l->thread->pid > r->thread->pid)
+		return 1;
+
+	return 0;
+}
+
+static struct sort_dimension pid_sort_dimension = {
+	.name			= "pid",
+	.cmp			= pid_cmp,
+};
+
+static int avg_cmp(struct work_atoms *l, struct work_atoms *r)
+{
+	u64 avgl, avgr;
+
+	if (!l->nb_atoms)
+		return -1;
+
+	if (!r->nb_atoms)
+		return 1;
+
+	avgl = l->total_lat / l->nb_atoms;
+	avgr = r->total_lat / r->nb_atoms;
+
+	if (avgl < avgr)
+		return -1;
+	if (avgl > avgr)
+		return 1;
+
+	return 0;
+}
+
+static struct sort_dimension avg_sort_dimension = {
+	.name			= "avg",
+	.cmp			= avg_cmp,
+};
+
+static int max_cmp(struct work_atoms *l, struct work_atoms *r)
+{
+	if (l->max_lat < r->max_lat)
+		return -1;
+	if (l->max_lat > r->max_lat)
+		return 1;
+
+	return 0;
+}
+
+static struct sort_dimension max_sort_dimension = {
+	.name			= "max",
+	.cmp			= max_cmp,
+};
+
+static int switch_cmp(struct work_atoms *l, struct work_atoms *r)
+{
+	if (l->nb_atoms < r->nb_atoms)
+		return -1;
+	if (l->nb_atoms > r->nb_atoms)
+		return 1;
+
+	return 0;
+}
+
+static struct sort_dimension switch_sort_dimension = {
+	.name			= "switch",
+	.cmp			= switch_cmp,
+};
+
+static int runtime_cmp(struct work_atoms *l, struct work_atoms *r)
+{
+	if (l->total_runtime < r->total_runtime)
+		return -1;
+	if (l->total_runtime > r->total_runtime)
+		return 1;
+
+	return 0;
+}
+
+static struct sort_dimension runtime_sort_dimension = {
+	.name			= "runtime",
+	.cmp			= runtime_cmp,
+};
+
+static struct sort_dimension *available_sorts[] = {
+	&pid_sort_dimension,
+	&avg_sort_dimension,
+	&max_sort_dimension,
+	&switch_sort_dimension,
+	&runtime_sort_dimension,
+};
+
+#define NB_AVAILABLE_SORTS	(int)(sizeof(available_sorts) / sizeof(struct sort_dimension *))
+
+static LIST_HEAD(sort_list);
+
+static int sort_dimension__add(const char *tok, struct list_head *list)
+{
+	int i;
+
+	for (i = 0; i < NB_AVAILABLE_SORTS; i++) {
+		if (!strcmp(available_sorts[i]->name, tok)) {
+			list_add_tail(&available_sorts[i]->list, list);
+
+			return 0;
+		}
+	}
+
+	return -1;
+}
+
+static void setup_sorting(void);
+
+static void sort_lat(void)
+{
+	struct rb_node *node;
+
+	for (;;) {
+		struct work_atoms *data;
+		node = rb_first(&atom_root);
+		if (!node)
+			break;
+
+		rb_erase(node, &atom_root);
+		data = rb_entry(node, struct work_atoms, node);
+		__thread_latency_insert(&sorted_atom_root, data, &sort_list);
+	}
+}
+
+static struct trace_sched_handler *trace_handler;
+
+static void
+process_sched_wakeup_event(struct raw_event_sample *raw,
+			   struct event *event,
+			   int cpu __used,
+			   u64 timestamp __used,
+			   struct thread *thread __used)
+{
+	struct trace_wakeup_event wakeup_event;
+
+	FILL_COMMON_FIELDS(wakeup_event, event, raw->data);
+
+	FILL_ARRAY(wakeup_event, comm, event, raw->data);
+	FILL_FIELD(wakeup_event, pid, event, raw->data);
+	FILL_FIELD(wakeup_event, prio, event, raw->data);
+	FILL_FIELD(wakeup_event, success, event, raw->data);
+	FILL_FIELD(wakeup_event, cpu, event, raw->data);
+
+	if (trace_handler->wakeup_event)
+		trace_handler->wakeup_event(&wakeup_event, event, cpu, timestamp, thread);
+}
+
+/*
+ * Track the current task - that way we can know whether there's any
+ * weird events, such as a task being switched away that is not current.
+ */
+static int max_cpu;
+
+static u32 curr_pid[MAX_CPUS] = { [0 ... MAX_CPUS-1] = -1 };
+
+static struct thread *curr_thread[MAX_CPUS];
+
+static char next_shortname1 = 'A';
+static char next_shortname2 = '0';
+
+static void
+map_switch_event(struct trace_switch_event *switch_event,
+		 struct event *event __used,
+		 int this_cpu,
+		 u64 timestamp,
+		 struct thread *thread __used)
+{
+	struct thread *sched_out, *sched_in;
+	int new_shortname;
+	u64 timestamp0;
+	s64 delta;
+	int cpu;
+
+	BUG_ON(this_cpu >= MAX_CPUS || this_cpu < 0);
+
+	if (this_cpu > max_cpu)
+		max_cpu = this_cpu;
+
+	timestamp0 = cpu_last_switched[this_cpu];
+	cpu_last_switched[this_cpu] = timestamp;
+	if (timestamp0)
+		delta = timestamp - timestamp0;
+	else
+		delta = 0;
+
+	if (delta < 0)
+		die("hm, delta: %Ld < 0 ?\n", delta);
+
+
+	sched_out = threads__findnew(switch_event->prev_pid);
+	sched_in = threads__findnew(switch_event->next_pid);
+
+	curr_thread[this_cpu] = sched_in;
+
+	printf("  ");
+
+	new_shortname = 0;
+	if (!sched_in->shortname[0]) {
+		sched_in->shortname[0] = next_shortname1;
+		sched_in->shortname[1] = next_shortname2;
+
+		if (next_shortname1 < 'Z') {
+			next_shortname1++;
+		} else {
+			next_shortname1='A';
+			if (next_shortname2 < '9') {
+				next_shortname2++;
+			} else {
+				next_shortname2='0';
+			}
+		}
+		new_shortname = 1;
+	}
+
+	for (cpu = 0; cpu <= max_cpu; cpu++) {
+		if (cpu != this_cpu)
+			printf(" ");
+		else
+			printf("*");
+
+		if (curr_thread[cpu]) {
+			if (curr_thread[cpu]->pid)
+				printf("%2s ", curr_thread[cpu]->shortname);
+			else
+				printf(".  ");
+		} else
+			printf("   ");
+	}
+
+	printf("  %12.6f secs ", (double)timestamp/1e9);
+	if (new_shortname) {
+		printf("%s => %s:%d\n",
+			sched_in->shortname, sched_in->comm, sched_in->pid);
+	} else {
+		printf("\n");
+	}
+}
+
+
+static void
+process_sched_switch_event(struct raw_event_sample *raw,
+			   struct event *event,
+			   int this_cpu,
+			   u64 timestamp __used,
+			   struct thread *thread __used)
+{
+	struct trace_switch_event switch_event;
+
+	FILL_COMMON_FIELDS(switch_event, event, raw->data);
+
+	FILL_ARRAY(switch_event, prev_comm, event, raw->data);
+	FILL_FIELD(switch_event, prev_pid, event, raw->data);
+	FILL_FIELD(switch_event, prev_prio, event, raw->data);
+	FILL_FIELD(switch_event, prev_state, event, raw->data);
+	FILL_ARRAY(switch_event, next_comm, event, raw->data);
+	FILL_FIELD(switch_event, next_pid, event, raw->data);
+	FILL_FIELD(switch_event, next_prio, event, raw->data);
+
+	if (curr_pid[this_cpu] != (u32)-1) {
+		/*
+		 * Are we trying to switch away a PID that is
+		 * not current?
+		 */
+		if (curr_pid[this_cpu] != switch_event.prev_pid)
+			nr_context_switch_bugs++;
+	}
+	if (trace_handler->switch_event)
+		trace_handler->switch_event(&switch_event, event, this_cpu, timestamp, thread);
+
+	curr_pid[this_cpu] = switch_event.next_pid;
+}
+
+static void
+process_sched_runtime_event(struct raw_event_sample *raw,
+			   struct event *event,
+			   int cpu __used,
+			   u64 timestamp __used,
+			   struct thread *thread __used)
+{
+	struct trace_runtime_event runtime_event;
+
+	FILL_ARRAY(runtime_event, comm, event, raw->data);
+	FILL_FIELD(runtime_event, pid, event, raw->data);
+	FILL_FIELD(runtime_event, runtime, event, raw->data);
+	FILL_FIELD(runtime_event, vruntime, event, raw->data);
+
+	if (trace_handler->runtime_event)
+		trace_handler->runtime_event(&runtime_event, event, cpu, timestamp, thread);
+}
+
+static void
+process_sched_fork_event(struct raw_event_sample *raw,
+			 struct event *event,
+			 int cpu __used,
+			 u64 timestamp __used,
+			 struct thread *thread __used)
+{
+	struct trace_fork_event fork_event;
+
+	FILL_COMMON_FIELDS(fork_event, event, raw->data);
+
+	FILL_ARRAY(fork_event, parent_comm, event, raw->data);
+	FILL_FIELD(fork_event, parent_pid, event, raw->data);
+	FILL_ARRAY(fork_event, child_comm, event, raw->data);
+	FILL_FIELD(fork_event, child_pid, event, raw->data);
+
+	if (trace_handler->fork_event)
+		trace_handler->fork_event(&fork_event, event, cpu, timestamp, thread);
+}
+
+static void
+process_sched_exit_event(struct event *event,
+			 int cpu __used,
+			 u64 timestamp __used,
+			 struct thread *thread __used)
+{
+	if (verbose)
+		printf("sched_exit event %p\n", event);
+}
+
+static void
+process_sched_migrate_task_event(struct raw_event_sample *raw,
+			   struct event *event,
+			   int cpu __used,
+			   u64 timestamp __used,
+			   struct thread *thread __used)
+{
+	struct trace_migrate_task_event migrate_task_event;
+
+	FILL_COMMON_FIELDS(migrate_task_event, event, raw->data);
+
+	FILL_ARRAY(migrate_task_event, comm, event, raw->data);
+	FILL_FIELD(migrate_task_event, pid, event, raw->data);
+	FILL_FIELD(migrate_task_event, prio, event, raw->data);
+	FILL_FIELD(migrate_task_event, cpu, event, raw->data);
+
+	if (trace_handler->migrate_task_event)
+		trace_handler->migrate_task_event(&migrate_task_event, event, cpu, timestamp, thread);
+}
+
+static void
+process_raw_event(event_t *raw_event __used, void *more_data,
+		  int cpu, u64 timestamp, struct thread *thread)
+{
+	struct raw_event_sample *raw = more_data;
+	struct event *event;
+	int type;
+
+	type = trace_parse_common_type(raw->data);
+	event = trace_find_event(type);
+
+	if (!strcmp(event->name, "sched_switch"))
+		process_sched_switch_event(raw, event, cpu, timestamp, thread);
+	if (!strcmp(event->name, "sched_stat_runtime"))
+		process_sched_runtime_event(raw, event, cpu, timestamp, thread);
+	if (!strcmp(event->name, "sched_wakeup"))
+		process_sched_wakeup_event(raw, event, cpu, timestamp, thread);
+	if (!strcmp(event->name, "sched_wakeup_new"))
+		process_sched_wakeup_event(raw, event, cpu, timestamp, thread);
+	if (!strcmp(event->name, "sched_process_fork"))
+		process_sched_fork_event(raw, event, cpu, timestamp, thread);
+	if (!strcmp(event->name, "sched_process_exit"))
+		process_sched_exit_event(event, cpu, timestamp, thread);
+	if (!strcmp(event->name, "sched_migrate_task"))
+		process_sched_migrate_task_event(raw, event, cpu, timestamp, thread);
+}
+
+static int
+process_sample_event(event_t *event, unsigned long offset, unsigned long head)
+{
+	struct thread *thread;
+	u64 ip = event->ip.ip;
+	u64 timestamp = -1;
+	u32 cpu = -1;
+	u64 period = 1;
+	void *more_data = event->ip.__more_data;
+
+	if (!(sample_type & PERF_SAMPLE_RAW))
+		return 0;
+
+	thread = threads__findnew(event->ip.pid);
+
+	if (sample_type & PERF_SAMPLE_TIME) {
+		timestamp = *(u64 *)more_data;
+		more_data += sizeof(u64);
+	}
+
+	if (sample_type & PERF_SAMPLE_CPU) {
+		cpu = *(u32 *)more_data;
+		more_data += sizeof(u32);
+		more_data += sizeof(u32); /* reserved */
+	}
+
+	if (sample_type & PERF_SAMPLE_PERIOD) {
+		period = *(u64 *)more_data;
+		more_data += sizeof(u64);
+	}
+
+	dump_printf("%p [%p]: PERF_RECORD_SAMPLE (IP, %d): %d/%d: %p period: %Ld\n",
+		(void *)(offset + head),
+		(void *)(long)(event->header.size),
+		event->header.misc,
+		event->ip.pid, event->ip.tid,
+		(void *)(long)ip,
+		(long long)period);
+
+	if (thread == NULL) {
+		eprintf("problem processing %d event, skipping it.\n",
+			event->header.type);
+		return -1;
+	}
+
+	dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid);
+
+	if (profile_cpu != -1 && profile_cpu != (int) cpu)
+		return 0;
+
+	process_raw_event(event, more_data, cpu, timestamp, thread);
+
+	return 0;
+}
+
+static int
+process_lost_event(event_t *event __used,
+		   unsigned long offset __used,
+		   unsigned long head __used)
+{
+	nr_lost_chunks++;
+	nr_lost_events += event->lost.lost;
+
+	return 0;
+}
+
+static int sample_type_check(u64 type)
+{
+	sample_type = type;
+
+	if (!(sample_type & PERF_SAMPLE_RAW)) {
+		fprintf(stderr,
+			"No trace sample to read. Did you call perf record "
+			"without -R?");
+		return -1;
+	}
+
+	return 0;
+}
+
+static struct perf_file_handler file_handler = {
+	.process_sample_event	= process_sample_event,
+	.process_comm_event	= process_comm_event,
+	.process_lost_event	= process_lost_event,
+	.sample_type_check	= sample_type_check,
+};
+
+static int read_events(void)
+{
+	register_idle_thread();
+	register_perf_file_handler(&file_handler);
+
+	return mmap_dispatch_perf_file(&header, input_name, 0, 0, &cwdlen, &cwd);
+}
+
+static void print_bad_events(void)
+{
+	if (nr_unordered_timestamps && nr_timestamps) {
+		printf("  INFO: %.3f%% unordered timestamps (%ld out of %ld)\n",
+			(double)nr_unordered_timestamps/(double)nr_timestamps*100.0,
+			nr_unordered_timestamps, nr_timestamps);
+	}
+	if (nr_lost_events && nr_events) {
+		printf("  INFO: %.3f%% lost events (%ld out of %ld, in %ld chunks)\n",
+			(double)nr_lost_events/(double)nr_events*100.0,
+			nr_lost_events, nr_events, nr_lost_chunks);
+	}
+	if (nr_state_machine_bugs && nr_timestamps) {
+		printf("  INFO: %.3f%% state machine bugs (%ld out of %ld)",
+			(double)nr_state_machine_bugs/(double)nr_timestamps*100.0,
+			nr_state_machine_bugs, nr_timestamps);
+		if (nr_lost_events)
+			printf(" (due to lost events?)");
+		printf("\n");
+	}
+	if (nr_context_switch_bugs && nr_timestamps) {
+		printf("  INFO: %.3f%% context switch bugs (%ld out of %ld)",
+			(double)nr_context_switch_bugs/(double)nr_timestamps*100.0,
+			nr_context_switch_bugs, nr_timestamps);
+		if (nr_lost_events)
+			printf(" (due to lost events?)");
+		printf("\n");
+	}
+}
+
+static void __cmd_lat(void)
+{
+	struct rb_node *next;
+
+	setup_pager();
+	read_events();
+	sort_lat();
+
+	printf("\n -----------------------------------------------------------------------------------------\n");
+	printf("  Task                  |   Runtime ms  | Switches | Average delay ms | Maximum delay ms |\n");
+	printf(" -----------------------------------------------------------------------------------------\n");
+
+	next = rb_first(&sorted_atom_root);
+
+	while (next) {
+		struct work_atoms *work_list;
+
+		work_list = rb_entry(next, struct work_atoms, node);
+		output_lat_thread(work_list);
+		next = rb_next(next);
+	}
+
+	printf(" -----------------------------------------------------------------------------------------\n");
+	printf("  TOTAL:                |%11.3f ms |%9Ld |\n",
+		(double)all_runtime/1e6, all_count);
+
+	printf(" ---------------------------------------------------\n");
+
+	print_bad_events();
+	printf("\n");
+
+}
+
+static struct trace_sched_handler map_ops  = {
+	.wakeup_event		= NULL,
+	.switch_event		= map_switch_event,
+	.runtime_event		= NULL,
+	.fork_event		= NULL,
+};
+
+static void __cmd_map(void)
+{
+	max_cpu = sysconf(_SC_NPROCESSORS_CONF);
+
+	setup_pager();
+	read_events();
+	print_bad_events();
+}
+
+static void __cmd_replay(void)
+{
+	unsigned long i;
+
+	calibrate_run_measurement_overhead();
+	calibrate_sleep_measurement_overhead();
+
+	test_calibrations();
+
+	read_events();
+
+	printf("nr_run_events:        %ld\n", nr_run_events);
+	printf("nr_sleep_events:      %ld\n", nr_sleep_events);
+	printf("nr_wakeup_events:     %ld\n", nr_wakeup_events);
+
+	if (targetless_wakeups)
+		printf("target-less wakeups:  %ld\n", targetless_wakeups);
+	if (multitarget_wakeups)
+		printf("multi-target wakeups: %ld\n", multitarget_wakeups);
+	if (nr_run_events_optimized)
+		printf("run atoms optimized: %ld\n",
+			nr_run_events_optimized);
+
+	print_task_traces();
+	add_cross_task_wakeups();
+
+	create_tasks();
+	printf("------------------------------------------------------------\n");
+	for (i = 0; i < replay_repeat; i++)
+		run_one_test();
+}
+
+
+static const char * const sched_usage[] = {
+	"perf sched [<options>] {record|latency|map|replay|trace}",
+	NULL
+};
+
+static const struct option sched_options[] = {
+	OPT_STRING('i', "input", &input_name, "file",
+		    "input file name"),
+	OPT_BOOLEAN('v', "verbose", &verbose,
+		    "be more verbose (show symbol address, etc)"),
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+		    "dump raw trace in ASCII"),
+	OPT_END()
+};
+
+static const char * const latency_usage[] = {
+	"perf sched latency [<options>]",
+	NULL
+};
+
+static const struct option latency_options[] = {
+	OPT_STRING('s', "sort", &sort_order, "key[,key2...]",
+		   "sort by key(s): runtime, switch, avg, max"),
+	OPT_BOOLEAN('v', "verbose", &verbose,
+		    "be more verbose (show symbol address, etc)"),
+	OPT_INTEGER('C', "CPU", &profile_cpu,
+		    "CPU to profile on"),
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+		    "dump raw trace in ASCII"),
+	OPT_END()
+};
+
+static const char * const replay_usage[] = {
+	"perf sched replay [<options>]",
+	NULL
+};
+
+static const struct option replay_options[] = {
+	OPT_INTEGER('r', "repeat", &replay_repeat,
+		    "repeat the workload replay N times (-1: infinite)"),
+	OPT_BOOLEAN('v', "verbose", &verbose,
+		    "be more verbose (show symbol address, etc)"),
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+		    "dump raw trace in ASCII"),
+	OPT_END()
+};
+
+static void setup_sorting(void)
+{
+	char *tmp, *tok, *str = strdup(sort_order);
+
+	for (tok = strtok_r(str, ", ", &tmp);
+			tok; tok = strtok_r(NULL, ", ", &tmp)) {
+		if (sort_dimension__add(tok, &sort_list) < 0) {
+			error("Unknown --sort key: `%s'", tok);
+			usage_with_options(latency_usage, latency_options);
+		}
+	}
+
+	free(str);
+
+	sort_dimension__add("pid", &cmp_pid);
+}
+
+static const char *record_args[] = {
+	"record",
+	"-a",
+	"-R",
+	"-M",
+	"-f",
+	"-m", "1024",
+	"-c", "1",
+	"-e", "sched:sched_switch:r",
+	"-e", "sched:sched_stat_wait:r",
+	"-e", "sched:sched_stat_sleep:r",
+	"-e", "sched:sched_stat_iowait:r",
+	"-e", "sched:sched_stat_runtime:r",
+	"-e", "sched:sched_process_exit:r",
+	"-e", "sched:sched_process_fork:r",
+	"-e", "sched:sched_wakeup:r",
+	"-e", "sched:sched_migrate_task:r",
+};
+
+static int __cmd_record(int argc, const char **argv)
+{
+	unsigned int rec_argc, i, j;
+	const char **rec_argv;
+
+	rec_argc = ARRAY_SIZE(record_args) + argc - 1;
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+
+	for (i = 0; i < ARRAY_SIZE(record_args); i++)
+		rec_argv[i] = strdup(record_args[i]);
+
+	for (j = 1; j < (unsigned int)argc; j++, i++)
+		rec_argv[i] = argv[j];
+
+	BUG_ON(i != rec_argc);
+
+	return cmd_record(i, rec_argv, NULL);
+}
+
+int cmd_sched(int argc, const char **argv, const char *prefix __used)
+{
+	symbol__init();
+
+	argc = parse_options(argc, argv, sched_options, sched_usage,
+			     PARSE_OPT_STOP_AT_NON_OPTION);
+	if (!argc)
+		usage_with_options(sched_usage, sched_options);
+
+	if (!strncmp(argv[0], "rec", 3)) {
+		return __cmd_record(argc, argv);
+	} else if (!strncmp(argv[0], "lat", 3)) {
+		trace_handler = &lat_ops;
+		if (argc > 1) {
+			argc = parse_options(argc, argv, latency_options, latency_usage, 0);
+			if (argc)
+				usage_with_options(latency_usage, latency_options);
+		}
+		setup_sorting();
+		__cmd_lat();
+	} else if (!strcmp(argv[0], "map")) {
+		trace_handler = &map_ops;
+		setup_sorting();
+		__cmd_map();
+	} else if (!strncmp(argv[0], "rep", 3)) {
+		trace_handler = &replay_ops;
+		if (argc) {
+			argc = parse_options(argc, argv, replay_options, replay_usage, 0);
+			if (argc)
+				usage_with_options(replay_usage, replay_options);
+		}
+		__cmd_replay();
+	} else if (!strcmp(argv[0], "trace")) {
+		/*
+		 * Aliased to 'perf trace' for now:
+		 */
+		return cmd_trace(argc, argv, prefix);
+	} else {
+		usage_with_options(sched_usage, sched_options);
+	}
+
+	return 0;
+}
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 61b828236c11..3db31e7bf173 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -48,7 +48,7 @@
 #include <sys/prctl.h>
 #include <math.h>
 
-static struct perf_counter_attr default_attrs[] = {
+static struct perf_event_attr default_attrs[] = {
 
   { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_TASK_CLOCK	},
   { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CONTEXT_SWITCHES},
@@ -69,7 +69,8 @@ static int			run_idx				=  0;
 static int			run_count			=  1;
 static int			inherit				=  1;
 static int			scale				=  1;
-static int			target_pid			= -1;
+static pid_t			target_pid			= -1;
+static pid_t			child_pid			= -1;
 static int			null_run			=  0;
 
 static int			fd[MAX_NR_CPUS][MAX_COUNTERS];
@@ -130,11 +131,11 @@ struct stats			runtime_cycles_stats;
 	 attrs[counter].config == PERF_COUNT_##c)
 
 #define ERR_PERF_OPEN \
-"Error: counter %d, sys_perf_counter_open() syscall returned with %d (%s)\n"
+"Error: counter %d, sys_perf_event_open() syscall returned with %d (%s)\n"
 
 static void create_perf_stat_counter(int counter, int pid)
 {
-	struct perf_counter_attr *attr = attrs + counter;
+	struct perf_event_attr *attr = attrs + counter;
 
 	if (scale)
 		attr->read_format = PERF_FORMAT_TOTAL_TIME_ENABLED |
@@ -144,7 +145,7 @@ static void create_perf_stat_counter(int counter, int pid)
 		unsigned int cpu;
 
 		for (cpu = 0; cpu < nr_cpus; cpu++) {
-			fd[cpu][counter] = sys_perf_counter_open(attr, -1, cpu, -1, 0);
+			fd[cpu][counter] = sys_perf_event_open(attr, -1, cpu, -1, 0);
 			if (fd[cpu][counter] < 0 && verbose)
 				fprintf(stderr, ERR_PERF_OPEN, counter,
 					fd[cpu][counter], strerror(errno));
@@ -154,7 +155,7 @@ static void create_perf_stat_counter(int counter, int pid)
 		attr->disabled	     = 1;
 		attr->enable_on_exec = 1;
 
-		fd[0][counter] = sys_perf_counter_open(attr, pid, -1, -1, 0);
+		fd[0][counter] = sys_perf_event_open(attr, pid, -1, -1, 0);
 		if (fd[0][counter] < 0 && verbose)
 			fprintf(stderr, ERR_PERF_OPEN, counter,
 				fd[0][counter], strerror(errno));
@@ -285,6 +286,8 @@ static int run_perf_stat(int argc __used, const char **argv)
 		exit(-1);
 	}
 
+	child_pid = pid;
+
 	/*
 	 * Wait for the child to be ready to exec.
 	 */
@@ -338,14 +341,24 @@ static void nsec_printout(int counter, double avg)
 
 static void abs_printout(int counter, double avg)
 {
+	double total, ratio = 0.0;
+
 	fprintf(stderr, " %14.0f  %-24s", avg, event_name(counter));
 
 	if (MATCH_EVENT(HARDWARE, HW_INSTRUCTIONS, counter)) {
-		fprintf(stderr, " # %10.3f IPC  ",
-				avg / avg_stats(&runtime_cycles_stats));
+		total = avg_stats(&runtime_cycles_stats);
+
+		if (total)
+			ratio = avg / total;
+
+		fprintf(stderr, " # %10.3f IPC  ", ratio);
 	} else {
-		fprintf(stderr, " # %10.3f M/sec",
-				1000.0 * avg / avg_stats(&runtime_nsecs_stats));
+		total = avg_stats(&runtime_nsecs_stats);
+
+		if (total)
+			ratio = 1000.0 * avg / total;
+
+		fprintf(stderr, " # %10.3f M/sec", ratio);
 	}
 }
 
@@ -423,6 +436,9 @@ static void skip_signal(int signo)
 
 static void sig_atexit(void)
 {
+	if (child_pid != -1)
+		kill(child_pid, SIGTERM);
+
 	if (signr == -1)
 		return;
 
diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c
new file mode 100644
index 000000000000..702d8fe58fbc
--- /dev/null
+++ b/tools/perf/builtin-timechart.c
@@ -0,0 +1,1162 @@
+/*
+ * builtin-timechart.c - make an svg timechart of system activity
+ *
+ * (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 "builtin.h"
+
+#include "util/util.h"
+
+#include "util/color.h"
+#include <linux/list.h>
+#include "util/cache.h"
+#include <linux/rbtree.h>
+#include "util/symbol.h"
+#include "util/string.h"
+#include "util/callchain.h"
+#include "util/strlist.h"
+
+#include "perf.h"
+#include "util/header.h"
+#include "util/parse-options.h"
+#include "util/parse-events.h"
+#include "util/svghelper.h"
+
+static char		const *input_name = "perf.data";
+static char		const *output_name = "output.svg";
+
+
+static unsigned long	page_size;
+static unsigned long	mmap_window = 32;
+static u64		sample_type;
+
+static unsigned int	numcpus;
+static u64		min_freq;	/* Lowest CPU frequency seen */
+static u64		max_freq;	/* Highest CPU frequency seen */
+static u64		turbo_frequency;
+
+static u64		first_time, last_time;
+
+static int		power_only;
+
+
+static struct perf_header	*header;
+
+struct per_pid;
+struct per_pidcomm;
+
+struct cpu_sample;
+struct power_event;
+struct wake_event;
+
+struct sample_wrapper;
+
+/*
+ * Datastructure layout:
+ * We keep an list of "pid"s, matching the kernels notion of a task struct.
+ * Each "pid" entry, has a list of "comm"s.
+ *	this is because we want to track different programs different, while
+ *	exec will reuse the original pid (by design).
+ * Each comm has a list of samples that will be used to draw
+ * final graph.
+ */
+
+struct per_pid {
+	struct per_pid *next;
+
+	int		pid;
+	int		ppid;
+
+	u64		start_time;
+	u64		end_time;
+	u64		total_time;
+	int		display;
+
+	struct per_pidcomm *all;
+	struct per_pidcomm *current;
+
+	int painted;
+};
+
+
+struct per_pidcomm {
+	struct per_pidcomm *next;
+
+	u64		start_time;
+	u64		end_time;
+	u64		total_time;
+
+	int		Y;
+	int		display;
+
+	long		state;
+	u64		state_since;
+
+	char		*comm;
+
+	struct cpu_sample *samples;
+};
+
+struct sample_wrapper {
+	struct sample_wrapper *next;
+
+	u64		timestamp;
+	unsigned char	data[0];
+};
+
+#define TYPE_NONE	0
+#define TYPE_RUNNING	1
+#define TYPE_WAITING	2
+#define TYPE_BLOCKED	3
+
+struct cpu_sample {
+	struct cpu_sample *next;
+
+	u64 start_time;
+	u64 end_time;
+	int type;
+	int cpu;
+};
+
+static struct per_pid *all_data;
+
+#define CSTATE 1
+#define PSTATE 2
+
+struct power_event {
+	struct power_event *next;
+	int type;
+	int state;
+	u64 start_time;
+	u64 end_time;
+	int cpu;
+};
+
+struct wake_event {
+	struct wake_event *next;
+	int waker;
+	int wakee;
+	u64 time;
+};
+
+static struct power_event    *power_events;
+static struct wake_event     *wake_events;
+
+struct sample_wrapper *all_samples;
+
+static struct per_pid *find_create_pid(int pid)
+{
+	struct per_pid *cursor = all_data;
+
+	while (cursor) {
+		if (cursor->pid == pid)
+			return cursor;
+		cursor = cursor->next;
+	}
+	cursor = malloc(sizeof(struct per_pid));
+	assert(cursor != NULL);
+	memset(cursor, 0, sizeof(struct per_pid));
+	cursor->pid = pid;
+	cursor->next = all_data;
+	all_data = cursor;
+	return cursor;
+}
+
+static void pid_set_comm(int pid, char *comm)
+{
+	struct per_pid *p;
+	struct per_pidcomm *c;
+	p = find_create_pid(pid);
+	c = p->all;
+	while (c) {
+		if (c->comm && strcmp(c->comm, comm) == 0) {
+			p->current = c;
+			return;
+		}
+		if (!c->comm) {
+			c->comm = strdup(comm);
+			p->current = c;
+			return;
+		}
+		c = c->next;
+	}
+	c = malloc(sizeof(struct per_pidcomm));
+	assert(c != NULL);
+	memset(c, 0, sizeof(struct per_pidcomm));
+	c->comm = strdup(comm);
+	p->current = c;
+	c->next = p->all;
+	p->all = c;
+}
+
+static void pid_fork(int pid, int ppid, u64 timestamp)
+{
+	struct per_pid *p, *pp;
+	p = find_create_pid(pid);
+	pp = find_create_pid(ppid);
+	p->ppid = ppid;
+	if (pp->current && pp->current->comm && !p->current)
+		pid_set_comm(pid, pp->current->comm);
+
+	p->start_time = timestamp;
+	if (p->current) {
+		p->current->start_time = timestamp;
+		p->current->state_since = timestamp;
+	}
+}
+
+static void pid_exit(int pid, u64 timestamp)
+{
+	struct per_pid *p;
+	p = find_create_pid(pid);
+	p->end_time = timestamp;
+	if (p->current)
+		p->current->end_time = timestamp;
+}
+
+static void
+pid_put_sample(int pid, int type, unsigned int cpu, u64 start, u64 end)
+{
+	struct per_pid *p;
+	struct per_pidcomm *c;
+	struct cpu_sample *sample;
+
+	p = find_create_pid(pid);
+	c = p->current;
+	if (!c) {
+		c = malloc(sizeof(struct per_pidcomm));
+		assert(c != NULL);
+		memset(c, 0, sizeof(struct per_pidcomm));
+		p->current = c;
+		c->next = p->all;
+		p->all = c;
+	}
+
+	sample = malloc(sizeof(struct cpu_sample));
+	assert(sample != NULL);
+	memset(sample, 0, sizeof(struct cpu_sample));
+	sample->start_time = start;
+	sample->end_time = end;
+	sample->type = type;
+	sample->next = c->samples;
+	sample->cpu = cpu;
+	c->samples = sample;
+
+	if (sample->type == TYPE_RUNNING && end > start && start > 0) {
+		c->total_time += (end-start);
+		p->total_time += (end-start);
+	}
+
+	if (c->start_time == 0 || c->start_time > start)
+		c->start_time = start;
+	if (p->start_time == 0 || p->start_time > start)
+		p->start_time = start;
+
+	if (cpu > numcpus)
+		numcpus = cpu;
+}
+
+#define MAX_CPUS 4096
+
+static u64 cpus_cstate_start_times[MAX_CPUS];
+static int cpus_cstate_state[MAX_CPUS];
+static u64 cpus_pstate_start_times[MAX_CPUS];
+static u64 cpus_pstate_state[MAX_CPUS];
+
+static int
+process_comm_event(event_t *event)
+{
+	pid_set_comm(event->comm.pid, event->comm.comm);
+	return 0;
+}
+static int
+process_fork_event(event_t *event)
+{
+	pid_fork(event->fork.pid, event->fork.ppid, event->fork.time);
+	return 0;
+}
+
+static int
+process_exit_event(event_t *event)
+{
+	pid_exit(event->fork.pid, event->fork.time);
+	return 0;
+}
+
+struct trace_entry {
+	u32			size;
+	unsigned short		type;
+	unsigned char		flags;
+	unsigned char		preempt_count;
+	int			pid;
+	int			tgid;
+};
+
+struct power_entry {
+	struct trace_entry te;
+	s64	type;
+	s64	value;
+};
+
+#define TASK_COMM_LEN 16
+struct wakeup_entry {
+	struct trace_entry te;
+	char comm[TASK_COMM_LEN];
+	int   pid;
+	int   prio;
+	int   success;
+};
+
+/*
+ * trace_flag_type is an enumeration that holds different
+ * states when a trace occurs. These are:
+ *  IRQS_OFF            - interrupts were disabled
+ *  IRQS_NOSUPPORT      - arch does not support irqs_disabled_flags
+ *  NEED_RESCED         - reschedule is requested
+ *  HARDIRQ             - inside an interrupt handler
+ *  SOFTIRQ             - inside a softirq handler
+ */
+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 sched_switch {
+	struct trace_entry te;
+	char prev_comm[TASK_COMM_LEN];
+	int  prev_pid;
+	int  prev_prio;
+	long prev_state; /* Arjan weeps. */
+	char next_comm[TASK_COMM_LEN];
+	int  next_pid;
+	int  next_prio;
+};
+
+static void c_state_start(int cpu, u64 timestamp, int state)
+{
+	cpus_cstate_start_times[cpu] = timestamp;
+	cpus_cstate_state[cpu] = state;
+}
+
+static void c_state_end(int cpu, u64 timestamp)
+{
+	struct power_event *pwr;
+	pwr = malloc(sizeof(struct power_event));
+	if (!pwr)
+		return;
+	memset(pwr, 0, sizeof(struct power_event));
+
+	pwr->state = cpus_cstate_state[cpu];
+	pwr->start_time = cpus_cstate_start_times[cpu];
+	pwr->end_time = timestamp;
+	pwr->cpu = cpu;
+	pwr->type = CSTATE;
+	pwr->next = power_events;
+
+	power_events = pwr;
+}
+
+static void p_state_change(int cpu, u64 timestamp, u64 new_freq)
+{
+	struct power_event *pwr;
+	pwr = malloc(sizeof(struct power_event));
+
+	if (new_freq > 8000000) /* detect invalid data */
+		return;
+
+	if (!pwr)
+		return;
+	memset(pwr, 0, sizeof(struct power_event));
+
+	pwr->state = cpus_pstate_state[cpu];
+	pwr->start_time = cpus_pstate_start_times[cpu];
+	pwr->end_time = timestamp;
+	pwr->cpu = cpu;
+	pwr->type = PSTATE;
+	pwr->next = power_events;
+
+	if (!pwr->start_time)
+		pwr->start_time = first_time;
+
+	power_events = pwr;
+
+	cpus_pstate_state[cpu] = new_freq;
+	cpus_pstate_start_times[cpu] = timestamp;
+
+	if ((u64)new_freq > max_freq)
+		max_freq = new_freq;
+
+	if (new_freq < min_freq || min_freq == 0)
+		min_freq = new_freq;
+
+	if (new_freq == max_freq - 1000)
+			turbo_frequency = max_freq;
+}
+
+static void
+sched_wakeup(int cpu, u64 timestamp, int pid, struct trace_entry *te)
+{
+	struct wake_event *we;
+	struct per_pid *p;
+	struct wakeup_entry *wake = (void *)te;
+
+	we = malloc(sizeof(struct wake_event));
+	if (!we)
+		return;
+
+	memset(we, 0, sizeof(struct wake_event));
+	we->time = timestamp;
+	we->waker = pid;
+
+	if ((te->flags & TRACE_FLAG_HARDIRQ) || (te->flags & TRACE_FLAG_SOFTIRQ))
+		we->waker = -1;
+
+	we->wakee = wake->pid;
+	we->next = wake_events;
+	wake_events = we;
+	p = find_create_pid(we->wakee);
+
+	if (p && p->current && p->current->state == TYPE_NONE) {
+		p->current->state_since = timestamp;
+		p->current->state = TYPE_WAITING;
+	}
+	if (p && p->current && p->current->state == TYPE_BLOCKED) {
+		pid_put_sample(p->pid, p->current->state, cpu, p->current->state_since, timestamp);
+		p->current->state_since = timestamp;
+		p->current->state = TYPE_WAITING;
+	}
+}
+
+static void sched_switch(int cpu, u64 timestamp, struct trace_entry *te)
+{
+	struct per_pid *p = NULL, *prev_p;
+	struct sched_switch *sw = (void *)te;
+
+
+	prev_p = find_create_pid(sw->prev_pid);
+
+	p = find_create_pid(sw->next_pid);
+
+	if (prev_p->current && prev_p->current->state != TYPE_NONE)
+		pid_put_sample(sw->prev_pid, TYPE_RUNNING, cpu, prev_p->current->state_since, timestamp);
+	if (p && p->current) {
+		if (p->current->state != TYPE_NONE)
+			pid_put_sample(sw->next_pid, p->current->state, cpu, p->current->state_since, timestamp);
+
+			p->current->state_since = timestamp;
+			p->current->state = TYPE_RUNNING;
+	}
+
+	if (prev_p->current) {
+		prev_p->current->state = TYPE_NONE;
+		prev_p->current->state_since = timestamp;
+		if (sw->prev_state & 2)
+			prev_p->current->state = TYPE_BLOCKED;
+		if (sw->prev_state == 0)
+			prev_p->current->state = TYPE_WAITING;
+	}
+}
+
+
+static int
+process_sample_event(event_t *event)
+{
+	int cursor = 0;
+	u64 addr = 0;
+	u64 stamp = 0;
+	u32 cpu = 0;
+	u32 pid = 0;
+	struct trace_entry *te;
+
+	if (sample_type & PERF_SAMPLE_IP)
+		cursor++;
+
+	if (sample_type & PERF_SAMPLE_TID) {
+		pid = event->sample.array[cursor]>>32;
+		cursor++;
+	}
+	if (sample_type & PERF_SAMPLE_TIME) {
+		stamp = event->sample.array[cursor++];
+
+		if (!first_time || first_time > stamp)
+			first_time = stamp;
+		if (last_time < stamp)
+			last_time = stamp;
+
+	}
+	if (sample_type & PERF_SAMPLE_ADDR)
+		addr = event->sample.array[cursor++];
+	if (sample_type & PERF_SAMPLE_ID)
+		cursor++;
+	if (sample_type & PERF_SAMPLE_STREAM_ID)
+		cursor++;
+	if (sample_type & PERF_SAMPLE_CPU)
+		cpu = event->sample.array[cursor++] & 0xFFFFFFFF;
+	if (sample_type & PERF_SAMPLE_PERIOD)
+		cursor++;
+
+	te = (void *)&event->sample.array[cursor];
+
+	if (sample_type & PERF_SAMPLE_RAW && te->size > 0) {
+		char *event_str;
+		struct power_entry *pe;
+
+		pe = (void *)te;
+
+		event_str = perf_header__find_event(te->type);
+
+		if (!event_str)
+			return 0;
+
+		if (strcmp(event_str, "power:power_start") == 0)
+			c_state_start(cpu, stamp, pe->value);
+
+		if (strcmp(event_str, "power:power_end") == 0)
+			c_state_end(cpu, stamp);
+
+		if (strcmp(event_str, "power:power_frequency") == 0)
+			p_state_change(cpu, stamp, pe->value);
+
+		if (strcmp(event_str, "sched:sched_wakeup") == 0)
+			sched_wakeup(cpu, stamp, pid, te);
+
+		if (strcmp(event_str, "sched:sched_switch") == 0)
+			sched_switch(cpu, stamp, te);
+	}
+	return 0;
+}
+
+/*
+ * After the last sample we need to wrap up the current C/P state
+ * and close out each CPU for these.
+ */
+static void end_sample_processing(void)
+{
+	u64 cpu;
+	struct power_event *pwr;
+
+	for (cpu = 0; cpu <= numcpus; cpu++) {
+		pwr = malloc(sizeof(struct power_event));
+		if (!pwr)
+			return;
+		memset(pwr, 0, sizeof(struct power_event));
+
+		/* C state */
+#if 0
+		pwr->state = cpus_cstate_state[cpu];
+		pwr->start_time = cpus_cstate_start_times[cpu];
+		pwr->end_time = last_time;
+		pwr->cpu = cpu;
+		pwr->type = CSTATE;
+		pwr->next = power_events;
+
+		power_events = pwr;
+#endif
+		/* P state */
+
+		pwr = malloc(sizeof(struct power_event));
+		if (!pwr)
+			return;
+		memset(pwr, 0, sizeof(struct power_event));
+
+		pwr->state = cpus_pstate_state[cpu];
+		pwr->start_time = cpus_pstate_start_times[cpu];
+		pwr->end_time = last_time;
+		pwr->cpu = cpu;
+		pwr->type = PSTATE;
+		pwr->next = power_events;
+
+		if (!pwr->start_time)
+			pwr->start_time = first_time;
+		if (!pwr->state)
+			pwr->state = min_freq;
+		power_events = pwr;
+	}
+}
+
+static u64 sample_time(event_t *event)
+{
+	int cursor;
+
+	cursor = 0;
+	if (sample_type & PERF_SAMPLE_IP)
+		cursor++;
+	if (sample_type & PERF_SAMPLE_TID)
+		cursor++;
+	if (sample_type & PERF_SAMPLE_TIME)
+		return event->sample.array[cursor];
+	return 0;
+}
+
+
+/*
+ * We first queue all events, sorted backwards by insertion.
+ * The order will get flipped later.
+ */
+static int
+queue_sample_event(event_t *event)
+{
+	struct sample_wrapper *copy, *prev;
+	int size;
+
+	size = event->sample.header.size + sizeof(struct sample_wrapper) + 8;
+
+	copy = malloc(size);
+	if (!copy)
+		return 1;
+
+	memset(copy, 0, size);
+
+	copy->next = NULL;
+	copy->timestamp = sample_time(event);
+
+	memcpy(&copy->data, event, event->sample.header.size);
+
+	/* insert in the right place in the list */
+
+	if (!all_samples) {
+		/* first sample ever */
+		all_samples = copy;
+		return 0;
+	}
+
+	if (all_samples->timestamp < copy->timestamp) {
+		/* insert at the head of the list */
+		copy->next = all_samples;
+		all_samples = copy;
+		return 0;
+	}
+
+	prev = all_samples;
+	while (prev->next) {
+		if (prev->next->timestamp < copy->timestamp) {
+			copy->next = prev->next;
+			prev->next = copy;
+			return 0;
+		}
+		prev = prev->next;
+	}
+	/* insert at the end of the list */
+	prev->next = copy;
+
+	return 0;
+}
+
+static void sort_queued_samples(void)
+{
+	struct sample_wrapper *cursor, *next;
+
+	cursor = all_samples;
+	all_samples = NULL;
+
+	while (cursor) {
+		next = cursor->next;
+		cursor->next = all_samples;
+		all_samples = cursor;
+		cursor = next;
+	}
+}
+
+/*
+ * Sort the pid datastructure
+ */
+static void sort_pids(void)
+{
+	struct per_pid *new_list, *p, *cursor, *prev;
+	/* sort by ppid first, then by pid, lowest to highest */
+
+	new_list = NULL;
+
+	while (all_data) {
+		p = all_data;
+		all_data = p->next;
+		p->next = NULL;
+
+		if (new_list == NULL) {
+			new_list = p;
+			p->next = NULL;
+			continue;
+		}
+		prev = NULL;
+		cursor = new_list;
+		while (cursor) {
+			if (cursor->ppid > p->ppid ||
+				(cursor->ppid == p->ppid && cursor->pid > p->pid)) {
+				/* must insert before */
+				if (prev) {
+					p->next = prev->next;
+					prev->next = p;
+					cursor = NULL;
+					continue;
+				} else {
+					p->next = new_list;
+					new_list = p;
+					cursor = NULL;
+					continue;
+				}
+			}
+
+			prev = cursor;
+			cursor = cursor->next;
+			if (!cursor)
+				prev->next = p;
+		}
+	}
+	all_data = new_list;
+}
+
+
+static void draw_c_p_states(void)
+{
+	struct power_event *pwr;
+	pwr = power_events;
+
+	/*
+	 * two pass drawing so that the P state bars are on top of the C state blocks
+	 */
+	while (pwr) {
+		if (pwr->type == CSTATE)
+			svg_cstate(pwr->cpu, pwr->start_time, pwr->end_time, pwr->state);
+		pwr = pwr->next;
+	}
+
+	pwr = power_events;
+	while (pwr) {
+		if (pwr->type == PSTATE) {
+			if (!pwr->state)
+				pwr->state = min_freq;
+			svg_pstate(pwr->cpu, pwr->start_time, pwr->end_time, pwr->state);
+		}
+		pwr = pwr->next;
+	}
+}
+
+static void draw_wakeups(void)
+{
+	struct wake_event *we;
+	struct per_pid *p;
+	struct per_pidcomm *c;
+
+	we = wake_events;
+	while (we) {
+		int from = 0, to = 0;
+		char *task_from = NULL, *task_to = NULL;
+
+		/* locate the column of the waker and wakee */
+		p = all_data;
+		while (p) {
+			if (p->pid == we->waker || p->pid == we->wakee) {
+				c = p->all;
+				while (c) {
+					if (c->Y && c->start_time <= we->time && c->end_time >= we->time) {
+						if (p->pid == we->waker) {
+							from = c->Y;
+							task_from = c->comm;
+						}
+						if (p->pid == we->wakee) {
+							to = c->Y;
+							task_to = c->comm;
+						}
+					}
+					c = c->next;
+				}
+			}
+			p = p->next;
+		}
+
+		if (we->waker == -1)
+			svg_interrupt(we->time, to);
+		else if (from && to && abs(from - to) == 1)
+			svg_wakeline(we->time, from, to);
+		else
+			svg_partial_wakeline(we->time, from, task_from, to, task_to);
+		we = we->next;
+	}
+}
+
+static void draw_cpu_usage(void)
+{
+	struct per_pid *p;
+	struct per_pidcomm *c;
+	struct cpu_sample *sample;
+	p = all_data;
+	while (p) {
+		c = p->all;
+		while (c) {
+			sample = c->samples;
+			while (sample) {
+				if (sample->type == TYPE_RUNNING)
+					svg_process(sample->cpu, sample->start_time, sample->end_time, "sample", c->comm);
+
+				sample = sample->next;
+			}
+			c = c->next;
+		}
+		p = p->next;
+	}
+}
+
+static void draw_process_bars(void)
+{
+	struct per_pid *p;
+	struct per_pidcomm *c;
+	struct cpu_sample *sample;
+	int Y = 0;
+
+	Y = 2 * numcpus + 2;
+
+	p = all_data;
+	while (p) {
+		c = p->all;
+		while (c) {
+			if (!c->display) {
+				c->Y = 0;
+				c = c->next;
+				continue;
+			}
+
+			svg_box(Y, c->start_time, c->end_time, "process");
+			sample = c->samples;
+			while (sample) {
+				if (sample->type == TYPE_RUNNING)
+					svg_sample(Y, sample->cpu, sample->start_time, sample->end_time);
+				if (sample->type == TYPE_BLOCKED)
+					svg_box(Y, sample->start_time, sample->end_time, "blocked");
+				if (sample->type == TYPE_WAITING)
+					svg_waiting(Y, sample->start_time, sample->end_time);
+				sample = sample->next;
+			}
+
+			if (c->comm) {
+				char comm[256];
+				if (c->total_time > 5000000000) /* 5 seconds */
+					sprintf(comm, "%s:%i (%2.2fs)", c->comm, p->pid, c->total_time / 1000000000.0);
+				else
+					sprintf(comm, "%s:%i (%3.1fms)", c->comm, p->pid, c->total_time / 1000000.0);
+
+				svg_text(Y, c->start_time, comm);
+			}
+			c->Y = Y;
+			Y++;
+			c = c->next;
+		}
+		p = p->next;
+	}
+}
+
+static int determine_display_tasks(u64 threshold)
+{
+	struct per_pid *p;
+	struct per_pidcomm *c;
+	int count = 0;
+
+	p = all_data;
+	while (p) {
+		p->display = 0;
+		if (p->start_time == 1)
+			p->start_time = first_time;
+
+		/* no exit marker, task kept running to the end */
+		if (p->end_time == 0)
+			p->end_time = last_time;
+		if (p->total_time >= threshold && !power_only)
+			p->display = 1;
+
+		c = p->all;
+
+		while (c) {
+			c->display = 0;
+
+			if (c->start_time == 1)
+				c->start_time = first_time;
+
+			if (c->total_time >= threshold && !power_only) {
+				c->display = 1;
+				count++;
+			}
+
+			if (c->end_time == 0)
+				c->end_time = last_time;
+
+			c = c->next;
+		}
+		p = p->next;
+	}
+	return count;
+}
+
+
+
+#define TIME_THRESH 10000000
+
+static void write_svg_file(const char *filename)
+{
+	u64 i;
+	int count;
+
+	numcpus++;
+
+
+	count = determine_display_tasks(TIME_THRESH);
+
+	/* We'd like to show at least 15 tasks; be less picky if we have fewer */
+	if (count < 15)
+		count = determine_display_tasks(TIME_THRESH / 10);
+
+	open_svg(filename, numcpus, count, first_time, last_time);
+
+	svg_time_grid();
+	svg_legenda();
+
+	for (i = 0; i < numcpus; i++)
+		svg_cpu_box(i, max_freq, turbo_frequency);
+
+	draw_cpu_usage();
+	draw_process_bars();
+	draw_c_p_states();
+	draw_wakeups();
+
+	svg_close();
+}
+
+static int
+process_event(event_t *event)
+{
+
+	switch (event->header.type) {
+
+	case PERF_RECORD_COMM:
+		return process_comm_event(event);
+	case PERF_RECORD_FORK:
+		return process_fork_event(event);
+	case PERF_RECORD_EXIT:
+		return process_exit_event(event);
+	case PERF_RECORD_SAMPLE:
+		return queue_sample_event(event);
+
+	/*
+	 * We dont process them right now but they are fine:
+	 */
+	case PERF_RECORD_MMAP:
+	case PERF_RECORD_THROTTLE:
+	case PERF_RECORD_UNTHROTTLE:
+		return 0;
+
+	default:
+		return -1;
+	}
+
+	return 0;
+}
+
+static void process_samples(void)
+{
+	struct sample_wrapper *cursor;
+	event_t *event;
+
+	sort_queued_samples();
+
+	cursor = all_samples;
+	while (cursor) {
+		event = (void *)&cursor->data;
+		cursor = cursor->next;
+		process_sample_event(event);
+	}
+}
+
+
+static int __cmd_timechart(void)
+{
+	int ret, rc = EXIT_FAILURE;
+	unsigned long offset = 0;
+	unsigned long head, shift;
+	struct stat statbuf;
+	event_t *event;
+	uint32_t size;
+	char *buf;
+	int input;
+
+	input = open(input_name, O_RDONLY);
+	if (input < 0) {
+		fprintf(stderr, " failed to open file: %s", input_name);
+		if (!strcmp(input_name, "perf.data"))
+			fprintf(stderr, "  (try 'perf record' first)");
+		fprintf(stderr, "\n");
+		exit(-1);
+	}
+
+	ret = fstat(input, &statbuf);
+	if (ret < 0) {
+		perror("failed to stat file");
+		exit(-1);
+	}
+
+	if (!statbuf.st_size) {
+		fprintf(stderr, "zero-sized file, nothing to do!\n");
+		exit(0);
+	}
+
+	header = perf_header__read(input);
+	head = header->data_offset;
+
+	sample_type = perf_header__sample_type(header);
+
+	shift = page_size * (head / page_size);
+	offset += shift;
+	head -= shift;
+
+remap:
+	buf = (char *)mmap(NULL, page_size * mmap_window, PROT_READ,
+			   MAP_SHARED, input, offset);
+	if (buf == MAP_FAILED) {
+		perror("failed to mmap file");
+		exit(-1);
+	}
+
+more:
+	event = (event_t *)(buf + head);
+
+	size = event->header.size;
+	if (!size)
+		size = 8;
+
+	if (head + event->header.size >= page_size * mmap_window) {
+		int ret2;
+
+		shift = page_size * (head / page_size);
+
+		ret2 = munmap(buf, page_size * mmap_window);
+		assert(ret2 == 0);
+
+		offset += shift;
+		head -= shift;
+		goto remap;
+	}
+
+	size = event->header.size;
+
+	if (!size || process_event(event) < 0) {
+
+		printf("%p [%p]: skipping unknown header type: %d\n",
+			(void *)(offset + head),
+			(void *)(long)(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 (offset + head >= header->data_offset + header->data_size)
+		goto done;
+
+	if (offset + head < (unsigned long)statbuf.st_size)
+		goto more;
+
+done:
+	rc = EXIT_SUCCESS;
+	close(input);
+
+
+	process_samples();
+
+	end_sample_processing();
+
+	sort_pids();
+
+	write_svg_file(output_name);
+
+	printf("Written %2.1f seconds of trace to %s.\n", (last_time - first_time) / 1000000000.0, output_name);
+
+	return rc;
+}
+
+static const char * const timechart_usage[] = {
+	"perf timechart [<options>] {record}",
+	NULL
+};
+
+static const char *record_args[] = {
+	"record",
+	"-a",
+	"-R",
+	"-M",
+	"-f",
+	"-c", "1",
+	"-e", "power:power_start",
+	"-e", "power:power_end",
+	"-e", "power:power_frequency",
+	"-e", "sched:sched_wakeup",
+	"-e", "sched:sched_switch",
+};
+
+static int __cmd_record(int argc, const char **argv)
+{
+	unsigned int rec_argc, i, j;
+	const char **rec_argv;
+
+	rec_argc = ARRAY_SIZE(record_args) + argc - 1;
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+
+	for (i = 0; i < ARRAY_SIZE(record_args); i++)
+		rec_argv[i] = strdup(record_args[i]);
+
+	for (j = 1; j < (unsigned int)argc; j++, i++)
+		rec_argv[i] = argv[j];
+
+	return cmd_record(i, rec_argv, NULL);
+}
+
+static const struct option options[] = {
+	OPT_STRING('i', "input", &input_name, "file",
+		    "input file name"),
+	OPT_STRING('o', "output", &output_name, "file",
+		    "output file name"),
+	OPT_INTEGER('w', "width", &svg_page_width,
+		    "page width"),
+	OPT_BOOLEAN('p', "power-only", &power_only,
+		    "output power data only"),
+	OPT_END()
+};
+
+
+int cmd_timechart(int argc, const char **argv, const char *prefix __used)
+{
+	symbol__init();
+
+	page_size = getpagesize();
+
+	argc = parse_options(argc, argv, options, timechart_usage,
+			PARSE_OPT_STOP_AT_NON_OPTION);
+
+	if (argc && !strncmp(argv[0], "rec", 3))
+		return __cmd_record(argc, argv);
+	else if (argc)
+		usage_with_options(timechart_usage, options);
+
+	setup_pager();
+
+	return __cmd_timechart();
+}
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index 4002ccb36750..2d8806bac258 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -22,6 +22,7 @@
 
 #include "util/symbol.h"
 #include "util/color.h"
+#include "util/thread.h"
 #include "util/util.h"
 #include <linux/rbtree.h>
 #include "util/parse-options.h"
@@ -54,26 +55,26 @@
 
 static int			fd[MAX_NR_CPUS][MAX_COUNTERS];
 
-static int			system_wide			=  0;
+static int			system_wide			=      0;
 
-static int			default_interval		= 100000;
+static int			default_interval		=      0;
 
-static int			count_filter			=  5;
-static int			print_entries			= 15;
+static int			count_filter			=      5;
+static int			print_entries			=     15;
 
-static int			target_pid			= -1;
-static int			inherit				=  0;
-static int			profile_cpu			= -1;
-static int			nr_cpus				=  0;
-static unsigned int		realtime_prio			=  0;
-static int			group				=  0;
+static int			target_pid			=     -1;
+static int			inherit				=      0;
+static int			profile_cpu			=     -1;
+static int			nr_cpus				=      0;
+static unsigned int		realtime_prio			=      0;
+static int			group				=      0;
 static unsigned int		page_size;
-static unsigned int		mmap_pages			= 16;
-static int			freq				=  0;
+static unsigned int		mmap_pages			=     16;
+static int			freq				=   1000; /* 1 KHz */
 
-static int			delay_secs			=  2;
-static int			zero;
-static int			dump_symtab;
+static int			delay_secs			=      2;
+static int			zero                            =      0;
+static int			dump_symtab                     =      0;
 
 /*
  * Source
@@ -86,19 +87,16 @@ struct source_line {
 	struct source_line	*next;
 };
 
-static char			*sym_filter			=  NULL;
-struct sym_entry		*sym_filter_entry		=  NULL;
-static int			sym_pcnt_filter			=  5;
-static int			sym_counter			=  0;
-static int			display_weighted		= -1;
+static char			*sym_filter			=   NULL;
+struct sym_entry		*sym_filter_entry		=   NULL;
+static int			sym_pcnt_filter			=      5;
+static int			sym_counter			=      0;
+static int			display_weighted		=     -1;
 
 /*
  * Symbols
  */
 
-static u64			min_ip;
-static u64			max_ip = -1ll;
-
 struct sym_entry {
 	struct rb_node		rb_node;
 	struct list_head	node;
@@ -106,6 +104,7 @@ struct sym_entry {
 	unsigned long		snap_count;
 	double			weight;
 	int			skip;
+	struct map		*map;
 	struct source_line	*source;
 	struct source_line	*lines;
 	struct source_line	**lines_tail;
@@ -119,12 +118,11 @@ struct sym_entry {
 static void parse_source(struct sym_entry *syme)
 {
 	struct symbol *sym;
-	struct module *module;
-	struct section *section = NULL;
+	struct map *map;
 	FILE *file;
 	char command[PATH_MAX*2];
-	const char *path = vmlinux_name;
-	u64 start, end, len;
+	const char *path;
+	u64 len;
 
 	if (!syme)
 		return;
@@ -135,27 +133,15 @@ static void parse_source(struct sym_entry *syme)
 	}
 
 	sym = (struct symbol *)(syme + 1);
-	module = sym->module;
-
-	if (module)
-		path = module->path;
-	if (!path)
-		return;
-
-	start = sym->obj_start;
-	if (!start)
-		start = sym->start;
+	map = syme->map;
+	path = map->dso->long_name;
 
-	if (module) {
-		section = module->sections->find_section(module->sections, ".text");
-		if (section)
-			start -= section->vma;
-	}
-
-	end = start + sym->end - sym->start + 1;
 	len = sym->end - sym->start;
 
-	sprintf(command, "objdump --start-address=0x%016Lx --stop-address=0x%016Lx -dS %s", start, end, path);
+	sprintf(command,
+		"objdump --start-address=0x%016Lx "
+			 "--stop-address=0x%016Lx -dS %s",
+		sym->start, sym->end, path);
 
 	file = popen(command, "r");
 	if (!file)
@@ -187,13 +173,11 @@ static void parse_source(struct sym_entry *syme)
 
 		if (strlen(src->line)>8 && src->line[8] == ':') {
 			src->eip = strtoull(src->line, NULL, 16);
-			if (section)
-				src->eip += section->vma;
+			src->eip += map->start;
 		}
 		if (strlen(src->line)>8 && src->line[16] == ':') {
 			src->eip = strtoull(src->line, NULL, 16);
-			if (section)
-				src->eip += section->vma;
+			src->eip += map->start;
 		}
 	}
 	pclose(file);
@@ -245,16 +229,9 @@ static void lookup_sym_source(struct sym_entry *syme)
 	struct symbol *symbol = (struct symbol *)(syme + 1);
 	struct source_line *line;
 	char pattern[PATH_MAX];
-	char *idx;
 
 	sprintf(pattern, "<%s>:", symbol->name);
 
-	if (symbol->module) {
-		idx = strstr(pattern, "\t");
-		if (idx)
-			*idx = 0;
-	}
-
 	pthread_mutex_lock(&syme->source_lock);
 	for (line = syme->lines; line; line = line->next) {
 		if (strstr(line->line, pattern)) {
@@ -516,8 +493,8 @@ static void print_sym_table(void)
 		if (verbose)
 			printf(" - %016llx", sym->start);
 		printf(" : %s", sym->name);
-		if (sym->module)
-			printf("\t[%s]", sym->module->name);
+		if (syme->map->dso->name[0] == '[')
+			printf(" \t%s", syme->map->dso->name);
 		printf("\n");
 	}
 }
@@ -782,12 +759,13 @@ static const char *skip_symbols[] = {
 	"exit_idle",
 	"mwait_idle",
 	"mwait_idle_with_hints",
+	"poll_idle",
 	"ppc64_runlatch_off",
 	"pseries_dedicated_idle_sleep",
 	NULL
 };
 
-static int symbol_filter(struct dso *self, struct symbol *sym)
+static int symbol_filter(struct map *map, struct symbol *sym)
 {
 	struct sym_entry *syme;
 	const char *name = sym->name;
@@ -809,7 +787,8 @@ static int symbol_filter(struct dso *self, struct symbol *sym)
 	    strstr(name, "_text_end"))
 		return 1;
 
-	syme = dso__sym_priv(self, sym);
+	syme = dso__sym_priv(map->dso, sym);
+	syme->map = map;
 	pthread_mutex_init(&syme->source_lock, NULL);
 	if (!sym_filter_entry && sym_filter && !strcmp(name, sym_filter))
 		sym_filter_entry = syme;
@@ -826,34 +805,14 @@ static int symbol_filter(struct dso *self, struct symbol *sym)
 
 static int parse_symbols(void)
 {
-	struct rb_node *node;
-	struct symbol  *sym;
-	int use_modules = vmlinux_name ? 1 : 0;
-
-	kernel_dso = dso__new("[kernel]", sizeof(struct sym_entry));
-	if (kernel_dso == NULL)
+	if (dsos__load_kernel(vmlinux_name, sizeof(struct sym_entry),
+			      symbol_filter, verbose, 1) <= 0)
 		return -1;
 
-	if (dso__load_kernel(kernel_dso, vmlinux_name, symbol_filter, verbose, use_modules) <= 0)
-		goto out_delete_dso;
-
-	node = rb_first(&kernel_dso->syms);
-	sym = rb_entry(node, struct symbol, rb_node);
-	min_ip = sym->start;
-
-	node = rb_last(&kernel_dso->syms);
-	sym = rb_entry(node, struct symbol, rb_node);
-	max_ip = sym->end;
-
 	if (dump_symtab)
-		dso__fprintf(kernel_dso, stderr);
+		dsos__fprintf(stderr);
 
 	return 0;
-
-out_delete_dso:
-	dso__delete(kernel_dso);
-	kernel_dso = NULL;
-	return -1;
 }
 
 /*
@@ -861,10 +820,11 @@ out_delete_dso:
  */
 static void record_ip(u64 ip, int counter)
 {
-	struct symbol *sym = dso__find_symbol(kernel_dso, ip);
+	struct map *map;
+	struct symbol *sym = kernel_maps__find_symbol(ip, &map);
 
 	if (sym != NULL) {
-		struct sym_entry *syme = dso__sym_priv(kernel_dso, sym);
+		struct sym_entry *syme = dso__sym_priv(map->dso, sym);
 
 		if (!syme->skip) {
 			syme->count[counter]++;
@@ -901,7 +861,7 @@ struct mmap_data {
 
 static unsigned int mmap_read_head(struct mmap_data *md)
 {
-	struct perf_counter_mmap_page *pc = md->base;
+	struct perf_event_mmap_page *pc = md->base;
 	int head;
 
 	head = pc->data_head;
@@ -910,8 +870,6 @@ static unsigned int mmap_read_head(struct mmap_data *md)
 	return head;
 }
 
-struct timeval last_read, this_read;
-
 static void mmap_read_counter(struct mmap_data *md)
 {
 	unsigned int head = mmap_read_head(md);
@@ -919,8 +877,6 @@ static void mmap_read_counter(struct mmap_data *md)
 	unsigned char *data = md->base + page_size;
 	int diff;
 
-	gettimeofday(&this_read, NULL);
-
 	/*
 	 * 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.
@@ -931,14 +887,7 @@ static void mmap_read_counter(struct mmap_data *md)
 	 */
 	diff = head - old;
 	if (diff > md->mask / 2 || diff < 0) {
-		struct timeval iv;
-		unsigned long msecs;
-
-		timersub(&this_read, &last_read, &iv);
-		msecs = iv.tv_sec*1000 + iv.tv_usec/1000;
-
-		fprintf(stderr, "WARNING: failed to keep up with mmap data."
-				"  Last read %lu msecs ago.\n", msecs);
+		fprintf(stderr, "WARNING: failed to keep up with mmap data.\n");
 
 		/*
 		 * head points to a known good entry, start there.
@@ -946,8 +895,6 @@ static void mmap_read_counter(struct mmap_data *md)
 		old = head;
 	}
 
-	last_read = this_read;
-
 	for (; old != head;) {
 		event_t *event = (event_t *)&data[old & md->mask];
 
@@ -977,9 +924,9 @@ static void mmap_read_counter(struct mmap_data *md)
 
 		old += size;
 
-		if (event->header.type == PERF_EVENT_SAMPLE) {
+		if (event->header.type == PERF_RECORD_SAMPLE) {
 			int user =
-	(event->header.misc & PERF_EVENT_MISC_CPUMODE_MASK) == PERF_EVENT_MISC_USER;
+	(event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK) == PERF_RECORD_MISC_USER;
 			process_event(event->ip.ip, md->counter, user);
 		}
 	}
@@ -1005,7 +952,7 @@ int group_fd;
 
 static void start_counter(int i, int counter)
 {
-	struct perf_counter_attr *attr;
+	struct perf_event_attr *attr;
 	int cpu;
 
 	cpu = profile_cpu;
@@ -1015,11 +962,17 @@ static void start_counter(int i, int counter)
 	attr = attrs + counter;
 
 	attr->sample_type	= PERF_SAMPLE_IP | PERF_SAMPLE_TID;
-	attr->freq		= freq;
+
+	if (freq) {
+		attr->sample_type	|= PERF_SAMPLE_PERIOD;
+		attr->freq		= 1;
+		attr->sample_freq	= freq;
+	}
+
 	attr->inherit		= (cpu < 0) && inherit;
 
 try_again:
-	fd[i][counter] = sys_perf_counter_open(attr, target_pid, cpu, group_fd, 0);
+	fd[i][counter] = sys_perf_event_open(attr, target_pid, cpu, group_fd, 0);
 
 	if (fd[i][counter] < 0) {
 		int err = errno;
@@ -1044,7 +997,7 @@ try_again:
 		printf("\n");
 		error("perfcounter syscall returned with %d (%s)\n",
 			fd[i][counter], strerror(err));
-		die("No CONFIG_PERF_COUNTERS=y kernel support configured?\n");
+		die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
 		exit(-1);
 	}
 	assert(fd[i][counter] >= 0);
@@ -1170,11 +1123,6 @@ int cmd_top(int argc, const char **argv, const char *prefix __used)
 	if (argc)
 		usage_with_options(top_usage, options);
 
-	if (freq) {
-		default_interval = freq;
-		freq = 1;
-	}
-
 	/* CPU and PID are mutually exclusive */
 	if (target_pid != -1 && profile_cpu != -1) {
 		printf("WARNING: PID switch overriding CPU\n");
@@ -1191,6 +1139,19 @@ int cmd_top(int argc, const char **argv, const char *prefix __used)
 	parse_symbols();
 	parse_source(sym_filter_entry);
 
+
+	/*
+	 * User specified count overrides default frequency.
+	 */
+	if (default_interval)
+		freq = 0;
+	else if (freq) {
+		default_interval = freq;
+	} else {
+		fprintf(stderr, "frequency and count are zero, aborting\n");
+		exit(EXIT_FAILURE);
+	}
+
 	/*
 	 * Fill in the ones not specifically initialized via -c:
 	 */
diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 914ab366e369..4c129ff0bb16 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -12,37 +12,33 @@
 #include "util/debug.h"
 
 #include "util/trace-event.h"
+#include "util/data_map.h"
 
 static char		const *input_name = "perf.data";
-static int		input;
-static unsigned long	page_size;
-static unsigned long	mmap_window = 32;
 
 static unsigned long	total = 0;
 static unsigned long	total_comm = 0;
 
-static struct rb_root	threads;
-static struct thread	*last_match;
-
 static struct perf_header *header;
 static u64		sample_type;
 
+static char		*cwd;
+static int		cwdlen;
+
 
 static int
 process_comm_event(event_t *event, unsigned long offset, unsigned long head)
 {
-	struct thread *thread;
+	struct thread *thread = threads__findnew(event->comm.pid);
 
-	thread = threads__findnew(event->comm.pid, &threads, &last_match);
-
-	dump_printf("%p [%p]: PERF_EVENT_COMM: %s:%d\n",
+	dump_printf("%p [%p]: PERF_RECORD_COMM: %s:%d\n",
 		(void *)(offset + head),
 		(void *)(long)(event->header.size),
 		event->comm.comm, event->comm.pid);
 
 	if (thread == NULL ||
 	    thread__set_comm(thread, event->comm.comm)) {
-		dump_printf("problem processing PERF_EVENT_COMM, skipping event.\n");
+		dump_printf("problem processing PERF_RECORD_COMM, skipping event.\n");
 		return -1;
 	}
 	total_comm++;
@@ -53,18 +49,12 @@ process_comm_event(event_t *event, unsigned long offset, unsigned long head)
 static int
 process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 {
-	char level;
-	int show = 0;
-	struct dso *dso = NULL;
-	struct thread *thread;
 	u64 ip = event->ip.ip;
 	u64 timestamp = -1;
 	u32 cpu = -1;
 	u64 period = 1;
 	void *more_data = event->ip.__more_data;
-	int cpumode;
-
-	thread = threads__findnew(event->ip.pid, &threads, &last_match);
+	struct thread *thread = threads__findnew(event->ip.pid);
 
 	if (sample_type & PERF_SAMPLE_TIME) {
 		timestamp = *(u64 *)more_data;
@@ -82,7 +72,7 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 		more_data += sizeof(u64);
 	}
 
-	dump_printf("%p [%p]: PERF_EVENT_SAMPLE (IP, %d): %d/%d: %p period: %Ld\n",
+	dump_printf("%p [%p]: PERF_RECORD_SAMPLE (IP, %d): %d/%d: %p period: %Ld\n",
 		(void *)(offset + head),
 		(void *)(long)(event->header.size),
 		event->header.misc,
@@ -90,37 +80,13 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 		(void *)(long)ip,
 		(long long)period);
 
-	dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid);
-
 	if (thread == NULL) {
 		eprintf("problem processing %d event, skipping it.\n",
 			event->header.type);
 		return -1;
 	}
 
-	cpumode = event->header.misc & PERF_EVENT_MISC_CPUMODE_MASK;
-
-	if (cpumode == PERF_EVENT_MISC_KERNEL) {
-		show = SHOW_KERNEL;
-		level = 'k';
-
-		dso = kernel_dso;
-
-		dump_printf(" ...... dso: %s\n", dso->name);
-
-	} else if (cpumode == PERF_EVENT_MISC_USER) {
-
-		show = SHOW_USER;
-		level = '.';
-
-	} else {
-		show = SHOW_HV;
-		level = 'H';
-
-		dso = hypervisor_dso;
-
-		dump_printf(" ...... dso: [hypervisor]\n");
-	}
+	dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid);
 
 	if (sample_type & PERF_SAMPLE_RAW) {
 		struct {
@@ -140,126 +106,32 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 	return 0;
 }
 
-static int
-process_event(event_t *event, unsigned long offset, unsigned long head)
+static int sample_type_check(u64 type)
 {
-	trace_event(event);
-
-	switch (event->header.type) {
-	case PERF_EVENT_MMAP ... PERF_EVENT_LOST:
-		return 0;
-
-	case PERF_EVENT_COMM:
-		return process_comm_event(event, offset, head);
-
-	case PERF_EVENT_EXIT ... PERF_EVENT_READ:
-		return 0;
+	sample_type = type;
 
-	case PERF_EVENT_SAMPLE:
-		return process_sample_event(event, offset, head);
-
-	case PERF_EVENT_MAX:
-	default:
+	if (!(sample_type & PERF_SAMPLE_RAW)) {
+		fprintf(stderr,
+			"No trace sample to read. Did you call perf record "
+			"without -R?");
 		return -1;
 	}
 
 	return 0;
 }
 
+static struct perf_file_handler file_handler = {
+	.process_sample_event	= process_sample_event,
+	.process_comm_event	= process_comm_event,
+	.sample_type_check	= sample_type_check,
+};
+
 static int __cmd_trace(void)
 {
-	int ret, rc = EXIT_FAILURE;
-	unsigned long offset = 0;
-	unsigned long head = 0;
-	struct stat perf_stat;
-	event_t *event;
-	uint32_t size;
-	char *buf;
-
-	trace_report();
-	register_idle_thread(&threads, &last_match);
-
-	input = open(input_name, O_RDONLY);
-	if (input < 0) {
-		perror("failed to open file");
-		exit(-1);
-	}
-
-	ret = fstat(input, &perf_stat);
-	if (ret < 0) {
-		perror("failed to stat file");
-		exit(-1);
-	}
+	register_idle_thread();
+	register_perf_file_handler(&file_handler);
 
-	if (!perf_stat.st_size) {
-		fprintf(stderr, "zero-sized file, nothing to do!\n");
-		exit(0);
-	}
-	header = perf_header__read(input);
-	head = header->data_offset;
-	sample_type = perf_header__sample_type(header);
-
-	if (!(sample_type & PERF_SAMPLE_RAW))
-		die("No trace sample to read. Did you call perf record "
-		    "without -R?");
-
-	if (load_kernel() < 0) {
-		perror("failed to load kernel symbols");
-		return EXIT_FAILURE;
-	}
-
-remap:
-	buf = (char *)mmap(NULL, page_size * mmap_window, PROT_READ,
-			   MAP_SHARED, input, offset);
-	if (buf == MAP_FAILED) {
-		perror("failed to mmap file");
-		exit(-1);
-	}
-
-more:
-	event = (event_t *)(buf + head);
-
-	size = event->header.size;
-	if (!size)
-		size = 8;
-
-	if (head + event->header.size >= page_size * mmap_window) {
-		unsigned long shift = page_size * (head / page_size);
-		int res;
-
-		res = munmap(buf, page_size * mmap_window);
-		assert(res == 0);
-
-		offset += shift;
-		head -= shift;
-		goto remap;
-	}
-
-	size = event->header.size;
-
-
-	if (!size || process_event(event, offset, head) < 0) {
-
-		/*
-		 * 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 (offset + head < (unsigned long)perf_stat.st_size)
-		goto more;
-
-	rc = EXIT_SUCCESS;
-	close(input);
-
-	return rc;
+	return mmap_dispatch_perf_file(&header, input_name, 0, 0, &cwdlen, &cwd);
 }
 
 static const char * const annotate_usage[] = {
@@ -272,13 +144,14 @@ static const struct option options[] = {
 		    "dump raw trace in ASCII"),
 	OPT_BOOLEAN('v', "verbose", &verbose,
 		    "be more verbose (show symbol address, etc)"),
+	OPT_BOOLEAN('l', "latency", &latency_format,
+		    "show latency attributes (irqs/preemption disabled, etc)"),
 	OPT_END()
 };
 
 int cmd_trace(int argc, const char **argv, const char *prefix __used)
 {
 	symbol__init();
-	page_size = getpagesize();
 
 	argc = parse_options(argc, argv, options, annotate_usage, 0);
 	if (argc) {
@@ -290,7 +163,6 @@ int cmd_trace(int argc, const char **argv, const char *prefix __used)
 			usage_with_options(annotate_usage, options);
 	}
 
-
 	setup_pager();
 
 	return __cmd_trace();
diff --git a/tools/perf/builtin.h b/tools/perf/builtin.h
index 3a63e41fb44e..e11d8d231c3b 100644
--- a/tools/perf/builtin.h
+++ b/tools/perf/builtin.h
@@ -16,12 +16,14 @@ extern int check_pager_config(const char *cmd);
 
 extern int cmd_annotate(int argc, const char **argv, const char *prefix);
 extern int cmd_help(int argc, const char **argv, const char *prefix);
+extern int cmd_sched(int argc, const char **argv, const char *prefix);
+extern int cmd_list(int argc, const char **argv, const char *prefix);
 extern int cmd_record(int argc, const char **argv, const char *prefix);
 extern int cmd_report(int argc, const char **argv, const char *prefix);
 extern int cmd_stat(int argc, const char **argv, const char *prefix);
+extern int cmd_timechart(int argc, const char **argv, const char *prefix);
 extern int cmd_top(int argc, const char **argv, const char *prefix);
-extern int cmd_version(int argc, const char **argv, const char *prefix);
-extern int cmd_list(int argc, const char **argv, const char *prefix);
 extern int cmd_trace(int argc, const char **argv, const char *prefix);
+extern int cmd_version(int argc, const char **argv, const char *prefix);
 
 #endif
diff --git a/tools/perf/command-list.txt b/tools/perf/command-list.txt
index eebce30afbc0..00326e230d87 100644
--- a/tools/perf/command-list.txt
+++ b/tools/perf/command-list.txt
@@ -4,7 +4,10 @@
 #
 perf-annotate			mainporcelain common
 perf-list			mainporcelain common
+perf-sched			mainporcelain common
 perf-record			mainporcelain common
 perf-report			mainporcelain common
 perf-stat			mainporcelain common
+perf-timechart			mainporcelain common
 perf-top			mainporcelain common
+perf-trace			mainporcelain common
diff --git a/tools/perf/design.txt b/tools/perf/design.txt
index f71e0d245cba..fdd42a824c98 100644
--- a/tools/perf/design.txt
+++ b/tools/perf/design.txt
@@ -18,10 +18,10 @@ underlying hardware counters.
 Performance counters are accessed via special file descriptors.
 There's one file descriptor per virtual counter used.
 
-The special file descriptor is opened via the perf_counter_open()
+The special file descriptor is opened via the perf_event_open()
 system call:
 
-   int sys_perf_counter_open(struct perf_counter_hw_event *hw_event_uptr,
+   int sys_perf_event_open(struct perf_event_hw_event *hw_event_uptr,
 			     pid_t pid, int cpu, int group_fd,
 			     unsigned long flags);
 
@@ -32,9 +32,9 @@ can be used to set the blocking mode, etc.
 Multiple counters can be kept open at a time, and the counters
 can be poll()ed.
 
-When creating a new counter fd, 'perf_counter_hw_event' is:
+When creating a new counter fd, 'perf_event_hw_event' is:
 
-struct perf_counter_hw_event {
+struct perf_event_hw_event {
         /*
          * The MSB of the config word signifies if the rest contains cpu
          * specific (raw) counter configuration data, if unset, the next
@@ -93,7 +93,7 @@ specified by 'event_id':
 
 /*
  * Generalized performance counter event types, used by the hw_event.event_id
- * parameter of the sys_perf_counter_open() syscall:
+ * parameter of the sys_perf_event_open() syscall:
  */
 enum hw_event_ids {
 	/*
@@ -159,7 +159,7 @@ in size.
  * reads on the counter should return the indicated quantities,
  * in increasing order of bit value, after the counter value.
  */
-enum perf_counter_read_format {
+enum perf_event_read_format {
         PERF_FORMAT_TOTAL_TIME_ENABLED  =  1,
         PERF_FORMAT_TOTAL_TIME_RUNNING  =  2,
 };
@@ -178,7 +178,7 @@ interrupt:
  * Bits that can be set in hw_event.record_type to request information
  * in the overflow packets.
  */
-enum perf_counter_record_format {
+enum perf_event_record_format {
         PERF_RECORD_IP          = 1U << 0,
         PERF_RECORD_TID         = 1U << 1,
         PERF_RECORD_TIME        = 1U << 2,
@@ -228,7 +228,7 @@ these events are recorded in the ring-buffer (see below).
 The 'comm' bit allows tracking of process comm data on process creation.
 This too is recorded in the ring-buffer (see below).
 
-The 'pid' parameter to the perf_counter_open() system call allows the
+The 'pid' parameter to the perf_event_open() system call allows the
 counter to be specific to a task:
 
  pid == 0: if the pid parameter is zero, the counter is attached to the
@@ -258,7 +258,7 @@ The 'flags' parameter is currently unused and must be zero.
 
 The 'group_fd' parameter allows counter "groups" to be set up.  A
 counter group has one counter which is the group "leader".  The leader
-is created first, with group_fd = -1 in the perf_counter_open call
+is created first, with group_fd = -1 in the perf_event_open call
 that creates it.  The rest of the group members are created
 subsequently, with group_fd giving the fd of the group leader.
 (A single counter on its own is created with group_fd = -1 and is
@@ -277,13 +277,13 @@ tracking are logged into a ring-buffer. This ring-buffer is created and
 accessed through mmap().
 
 The mmap size should be 1+2^n pages, where the first page is a meta-data page
-(struct perf_counter_mmap_page) that contains various bits of information such
+(struct perf_event_mmap_page) that contains various bits of information such
 as where the ring-buffer head is.
 
 /*
  * Structure of the page that can be mapped via mmap
  */
-struct perf_counter_mmap_page {
+struct perf_event_mmap_page {
         __u32   version;                /* version number of this structure */
         __u32   compat_version;         /* lowest version this is compat with */
 
@@ -317,7 +317,7 @@ struct perf_counter_mmap_page {
          * Control data for the mmap() data buffer.
          *
          * User-space reading this value should issue an rmb(), on SMP capable
-         * platforms, after reading this value -- see perf_counter_wakeup().
+         * platforms, after reading this value -- see perf_event_wakeup().
          */
         __u32   data_head;              /* head in the data section */
 };
@@ -327,9 +327,9 @@ NOTE: the hw-counter userspace bits are arch specific and are currently only
 
 The following 2^n pages are the ring-buffer which contains events of the form:
 
-#define PERF_EVENT_MISC_KERNEL          (1 << 0)
-#define PERF_EVENT_MISC_USER            (1 << 1)
-#define PERF_EVENT_MISC_OVERFLOW        (1 << 2)
+#define PERF_RECORD_MISC_KERNEL          (1 << 0)
+#define PERF_RECORD_MISC_USER            (1 << 1)
+#define PERF_RECORD_MISC_OVERFLOW        (1 << 2)
 
 struct perf_event_header {
         __u32   type;
@@ -353,8 +353,8 @@ enum perf_event_type {
          *      char                            filename[];
          * };
          */
-        PERF_EVENT_MMAP                 = 1,
-        PERF_EVENT_MUNMAP               = 2,
+        PERF_RECORD_MMAP                 = 1,
+        PERF_RECORD_MUNMAP               = 2,
 
         /*
          * struct {
@@ -364,10 +364,10 @@ enum perf_event_type {
          *      char                            comm[];
          * };
          */
-        PERF_EVENT_COMM                 = 3,
+        PERF_RECORD_COMM                 = 3,
 
         /*
-         * When header.misc & PERF_EVENT_MISC_OVERFLOW the event_type field
+         * When header.misc & PERF_RECORD_MISC_OVERFLOW the event_type field
          * will be PERF_RECORD_*
          *
          * struct {
@@ -397,7 +397,7 @@ Notification of new events is possible through poll()/select()/epoll() and
 fcntl() managing signals.
 
 Normally a notification is generated for every page filled, however one can
-additionally set perf_counter_hw_event.wakeup_events to generate one every
+additionally set perf_event_hw_event.wakeup_events to generate one every
 so many counter overflow events.
 
 Future work will include a splice() interface to the ring-buffer.
@@ -409,11 +409,11 @@ events but does continue to exist and maintain its count value.
 
 An individual counter or counter group can be enabled with
 
-	ioctl(fd, PERF_COUNTER_IOC_ENABLE);
+	ioctl(fd, PERF_EVENT_IOC_ENABLE);
 
 or disabled with
 
-	ioctl(fd, PERF_COUNTER_IOC_DISABLE);
+	ioctl(fd, PERF_EVENT_IOC_DISABLE);
 
 Enabling or disabling the leader of a group enables or disables the
 whole group; that is, while the group leader is disabled, none of the
@@ -424,16 +424,16 @@ other counter.
 
 Additionally, non-inherited overflow counters can use
 
-	ioctl(fd, PERF_COUNTER_IOC_REFRESH, nr);
+	ioctl(fd, PERF_EVENT_IOC_REFRESH, nr);
 
 to enable a counter for 'nr' events, after which it gets disabled again.
 
 A process can enable or disable all the counter groups that are
 attached to it, using prctl:
 
-	prctl(PR_TASK_PERF_COUNTERS_ENABLE);
+	prctl(PR_TASK_PERF_EVENTS_ENABLE);
 
-	prctl(PR_TASK_PERF_COUNTERS_DISABLE);
+	prctl(PR_TASK_PERF_EVENTS_DISABLE);
 
 This applies to all counters on the current process, whether created
 by this process or by another, and doesn't affect any counters that
@@ -447,11 +447,14 @@ Arch requirements
 If your architecture does not have hardware performance metrics, you can
 still use the generic software counters based on hrtimers for sampling.
 
-So to start with, in order to add HAVE_PERF_COUNTERS to your Kconfig, you
+So to start with, in order to add HAVE_PERF_EVENTS to your Kconfig, you
 will need at least this:
-	- asm/perf_counter.h - a basic stub will suffice at first
+	- asm/perf_event.h - a basic stub will suffice at first
 	- support for atomic64 types (and associated helper functions)
-	- set_perf_counter_pending() implemented
+	- set_perf_event_pending() implemented
 
 If your architecture does have hardware capabilities, you can override the
-weak stub hw_perf_counter_init() to register hardware counters.
+weak stub hw_perf_event_init() to register hardware counters.
+
+Architectures that have d-cache aliassing issues, such as Sparc and ARM,
+should select PERF_USE_VMALLOC in order to avoid these for perf mmap().
diff --git a/tools/perf/perf.c b/tools/perf/perf.c
index fe4589dde950..624e62d9d1e0 100644
--- a/tools/perf/perf.c
+++ b/tools/perf/perf.c
@@ -89,8 +89,8 @@ static int handle_options(const char*** argv, int* argc, int* envchanged)
 		/*
 		 * Check remaining flags.
 		 */
-		if (!prefixcmp(cmd, "--exec-path")) {
-			cmd += 11;
+		if (!prefixcmp(cmd, CMD_EXEC_PATH)) {
+			cmd += strlen(CMD_EXEC_PATH);
 			if (*cmd == '=')
 				perf_set_argv_exec_path(cmd + 1);
 			else {
@@ -117,8 +117,8 @@ static int handle_options(const char*** argv, int* argc, int* envchanged)
 			(*argv)++;
 			(*argc)--;
 			handled++;
-		} else if (!prefixcmp(cmd, "--perf-dir=")) {
-			setenv(PERF_DIR_ENVIRONMENT, cmd + 10, 1);
+		} else if (!prefixcmp(cmd, CMD_PERF_DIR)) {
+			setenv(PERF_DIR_ENVIRONMENT, cmd + strlen(CMD_PERF_DIR), 1);
 			if (envchanged)
 				*envchanged = 1;
 		} else if (!strcmp(cmd, "--work-tree")) {
@@ -131,8 +131,8 @@ static int handle_options(const char*** argv, int* argc, int* envchanged)
 				*envchanged = 1;
 			(*argv)++;
 			(*argc)--;
-		} else if (!prefixcmp(cmd, "--work-tree=")) {
-			setenv(PERF_WORK_TREE_ENVIRONMENT, cmd + 12, 1);
+		} else if (!prefixcmp(cmd, CMD_WORK_TREE)) {
+			setenv(PERF_WORK_TREE_ENVIRONMENT, cmd + strlen(CMD_WORK_TREE), 1);
 			if (envchanged)
 				*envchanged = 1;
 		} else if (!strcmp(cmd, "--debugfs-dir")) {
@@ -146,8 +146,8 @@ static int handle_options(const char*** argv, int* argc, int* envchanged)
 				*envchanged = 1;
 			(*argv)++;
 			(*argc)--;
-		} else if (!prefixcmp(cmd, "--debugfs-dir=")) {
-			strncpy(debugfs_mntpt, cmd + 14, MAXPATHLEN);
+		} else if (!prefixcmp(cmd, CMD_DEBUGFS_DIR)) {
+			strncpy(debugfs_mntpt, cmd + strlen(CMD_DEBUGFS_DIR), MAXPATHLEN);
 			debugfs_mntpt[MAXPATHLEN - 1] = '\0';
 			if (envchanged)
 				*envchanged = 1;
@@ -289,10 +289,12 @@ static void handle_internal_command(int argc, const char **argv)
 		{ "record", cmd_record, 0 },
 		{ "report", cmd_report, 0 },
 		{ "stat", cmd_stat, 0 },
+		{ "timechart", cmd_timechart, 0 },
 		{ "top", cmd_top, 0 },
 		{ "annotate", cmd_annotate, 0 },
 		{ "version", cmd_version, 0 },
 		{ "trace", cmd_trace, 0 },
+		{ "sched", cmd_sched, 0 },
 	};
 	unsigned int i;
 	static const char ext[] = STRIP_EXTENSION;
diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index 2abeb20d0bf3..8cc4623afd6f 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -52,15 +52,15 @@
 #include <sys/types.h>
 #include <sys/syscall.h>
 
-#include "../../include/linux/perf_counter.h"
+#include "../../include/linux/perf_event.h"
 #include "util/types.h"
 
 /*
- * prctl(PR_TASK_PERF_COUNTERS_DISABLE) will (cheaply) disable all
+ * prctl(PR_TASK_PERF_EVENTS_DISABLE) will (cheaply) disable all
  * counters in the current task.
  */
-#define PR_TASK_PERF_COUNTERS_DISABLE   31
-#define PR_TASK_PERF_COUNTERS_ENABLE    32
+#define PR_TASK_PERF_EVENTS_DISABLE   31
+#define PR_TASK_PERF_EVENTS_ENABLE    32
 
 #ifndef NSEC_PER_SEC
 # define NSEC_PER_SEC			1000000000ULL
@@ -90,12 +90,12 @@ static inline unsigned long long rdclock(void)
 	_min1 < _min2 ? _min1 : _min2; })
 
 static inline int
-sys_perf_counter_open(struct perf_counter_attr *attr,
+sys_perf_event_open(struct perf_event_attr *attr,
 		      pid_t pid, int cpu, int group_fd,
 		      unsigned long flags)
 {
 	attr->size = sizeof(*attr);
-	return syscall(__NR_perf_counter_open, attr, pid, cpu,
+	return syscall(__NR_perf_event_open, attr, pid, cpu,
 		       group_fd, flags);
 }
 
diff --git a/tools/perf/util/cache.h b/tools/perf/util/cache.h
index 6f8ea9d210b6..918eb376abe3 100644
--- a/tools/perf/util/cache.h
+++ b/tools/perf/util/cache.h
@@ -1,10 +1,15 @@
-#ifndef CACHE_H
-#define CACHE_H
+#ifndef __PERF_CACHE_H
+#define __PERF_CACHE_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 DEFAULT_PERF_DIR_ENVIRONMENT ".perf"
@@ -117,4 +122,4 @@ extern char *perf_pathdup(const char *fmt, ...)
 
 extern size_t strlcpy(char *dest, const char *src, size_t size);
 
-#endif /* CACHE_H */
+#endif /* __PERF_CACHE_H */
diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h
index 43cf3ea9e088..ad4626de4c2b 100644
--- a/tools/perf/util/callchain.h
+++ b/tools/perf/util/callchain.h
@@ -58,4 +58,4 @@ static inline u64 cumul_hits(struct callchain_node *node)
 int register_callchain_param(struct callchain_param *param);
 void append_chain(struct callchain_node *root, struct ip_callchain *chain,
 		  struct symbol **syms);
-#endif
+#endif	/* __PERF_CALLCHAIN_H */
diff --git a/tools/perf/util/color.h b/tools/perf/util/color.h
index 58d597564b99..24e8809210bb 100644
--- a/tools/perf/util/color.h
+++ b/tools/perf/util/color.h
@@ -1,5 +1,5 @@
-#ifndef COLOR_H
-#define COLOR_H
+#ifndef __PERF_COLOR_H
+#define __PERF_COLOR_H
 
 /* "\033[1;38;5;2xx;48;5;2xxm\0" is 23 bytes */
 #define COLOR_MAXLEN 24
@@ -39,4 +39,4 @@ int color_fwrite_lines(FILE *fp, const char *color, size_t count, const char *bu
 int percent_color_fprintf(FILE *fp, const char *fmt, double percent);
 const char *get_percent_color(double percent);
 
-#endif /* COLOR_H */
+#endif /* __PERF_COLOR_H */
diff --git a/tools/perf/util/data_map.c b/tools/perf/util/data_map.c
new file mode 100644
index 000000000000..242b0555ab91
--- /dev/null
+++ b/tools/perf/util/data_map.c
@@ -0,0 +1,222 @@
+#include "data_map.h"
+#include "symbol.h"
+#include "util.h"
+#include "debug.h"
+
+
+static struct perf_file_handler *curr_handler;
+static unsigned long	mmap_window = 32;
+static char		__cwd[PATH_MAX];
+
+static int
+process_event_stub(event_t *event __used,
+		   unsigned long offset __used,
+		   unsigned long head __used)
+{
+	return 0;
+}
+
+void register_perf_file_handler(struct perf_file_handler *handler)
+{
+	if (!handler->process_sample_event)
+		handler->process_sample_event = process_event_stub;
+	if (!handler->process_mmap_event)
+		handler->process_mmap_event = process_event_stub;
+	if (!handler->process_comm_event)
+		handler->process_comm_event = process_event_stub;
+	if (!handler->process_fork_event)
+		handler->process_fork_event = process_event_stub;
+	if (!handler->process_exit_event)
+		handler->process_exit_event = process_event_stub;
+	if (!handler->process_lost_event)
+		handler->process_lost_event = process_event_stub;
+	if (!handler->process_read_event)
+		handler->process_read_event = process_event_stub;
+	if (!handler->process_throttle_event)
+		handler->process_throttle_event = process_event_stub;
+	if (!handler->process_unthrottle_event)
+		handler->process_unthrottle_event = process_event_stub;
+
+	curr_handler = handler;
+}
+
+static int
+process_event(event_t *event, unsigned long offset, unsigned long head)
+{
+	trace_event(event);
+
+	switch (event->header.type) {
+	case PERF_RECORD_SAMPLE:
+		return curr_handler->process_sample_event(event, offset, head);
+	case PERF_RECORD_MMAP:
+		return curr_handler->process_mmap_event(event, offset, head);
+	case PERF_RECORD_COMM:
+		return curr_handler->process_comm_event(event, offset, head);
+	case PERF_RECORD_FORK:
+		return curr_handler->process_fork_event(event, offset, head);
+	case PERF_RECORD_EXIT:
+		return curr_handler->process_exit_event(event, offset, head);
+	case PERF_RECORD_LOST:
+		return curr_handler->process_lost_event(event, offset, head);
+	case PERF_RECORD_READ:
+		return curr_handler->process_read_event(event, offset, head);
+	case PERF_RECORD_THROTTLE:
+		return curr_handler->process_throttle_event(event, offset, head);
+	case PERF_RECORD_UNTHROTTLE:
+		return curr_handler->process_unthrottle_event(event, offset, head);
+	default:
+		curr_handler->total_unknown++;
+		return -1;
+	}
+}
+
+int mmap_dispatch_perf_file(struct perf_header **pheader,
+			    const char *input_name,
+			    int force,
+			    int full_paths,
+			    int *cwdlen,
+			    char **cwd)
+{
+	int ret, rc = EXIT_FAILURE;
+	struct perf_header *header;
+	unsigned long head, shift;
+	unsigned long offset = 0;
+	struct stat input_stat;
+	size_t	page_size;
+	u64 sample_type;
+	event_t *event;
+	uint32_t size;
+	int input;
+	char *buf;
+
+	if (!curr_handler)
+		die("Forgot to register perf file handler");
+
+	page_size = getpagesize();
+
+	input = open(input_name, O_RDONLY);
+	if (input < 0) {
+		fprintf(stderr, " failed to open file: %s", input_name);
+		if (!strcmp(input_name, "perf.data"))
+			fprintf(stderr, "  (try 'perf record' first)");
+		fprintf(stderr, "\n");
+		exit(-1);
+	}
+
+	ret = fstat(input, &input_stat);
+	if (ret < 0) {
+		perror("failed to stat file");
+		exit(-1);
+	}
+
+	if (!force && input_stat.st_uid && (input_stat.st_uid != geteuid())) {
+		fprintf(stderr, "file: %s not owned by current user or root\n",
+			input_name);
+		exit(-1);
+	}
+
+	if (!input_stat.st_size) {
+		fprintf(stderr, "zero-sized file, nothing to do!\n");
+		exit(0);
+	}
+
+	*pheader = perf_header__read(input);
+	header = *pheader;
+	head = header->data_offset;
+
+	sample_type = perf_header__sample_type(header);
+
+	if (curr_handler->sample_type_check)
+		if (curr_handler->sample_type_check(sample_type) < 0)
+			exit(-1);
+
+	if (load_kernel() < 0) {
+		perror("failed to load kernel symbols");
+		return EXIT_FAILURE;
+	}
+
+	if (!full_paths) {
+		if (getcwd(__cwd, sizeof(__cwd)) == NULL) {
+			perror("failed to get the current directory");
+			return EXIT_FAILURE;
+		}
+		*cwd = __cwd;
+		*cwdlen = strlen(*cwd);
+	} else {
+		*cwd = NULL;
+		*cwdlen = 0;
+	}
+
+	shift = page_size * (head / page_size);
+	offset += shift;
+	head -= shift;
+
+remap:
+	buf = (char *)mmap(NULL, page_size * mmap_window, PROT_READ,
+			   MAP_SHARED, input, offset);
+	if (buf == MAP_FAILED) {
+		perror("failed to mmap file");
+		exit(-1);
+	}
+
+more:
+	event = (event_t *)(buf + head);
+
+	size = event->header.size;
+	if (!size)
+		size = 8;
+
+	if (head + event->header.size >= page_size * mmap_window) {
+		int munmap_ret;
+
+		shift = page_size * (head / page_size);
+
+		munmap_ret = munmap(buf, page_size * mmap_window);
+		assert(munmap_ret == 0);
+
+		offset += shift;
+		head -= shift;
+		goto remap;
+	}
+
+	size = event->header.size;
+
+	dump_printf("\n%p [%p]: event: %d\n",
+			(void *)(offset + head),
+			(void *)(long)event->header.size,
+			event->header.type);
+
+	if (!size || process_event(event, offset, head) < 0) {
+
+		dump_printf("%p [%p]: skipping unknown header type: %d\n",
+			(void *)(offset + head),
+			(void *)(long)(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 (offset + head >= header->data_offset + header->data_size)
+		goto done;
+
+	if (offset + head < (unsigned long)input_stat.st_size)
+		goto more;
+
+done:
+	rc = EXIT_SUCCESS;
+	close(input);
+
+	return rc;
+}
+
+
diff --git a/tools/perf/util/data_map.h b/tools/perf/util/data_map.h
new file mode 100644
index 000000000000..716d1053b074
--- /dev/null
+++ b/tools/perf/util/data_map.h
@@ -0,0 +1,31 @@
+#ifndef __PERF_DATAMAP_H
+#define __PERF_DATAMAP_H
+
+#include "event.h"
+#include "header.h"
+
+typedef int (*event_type_handler_t)(event_t *, unsigned long, unsigned long);
+
+struct perf_file_handler {
+	event_type_handler_t	process_sample_event;
+	event_type_handler_t	process_mmap_event;
+	event_type_handler_t	process_comm_event;
+	event_type_handler_t	process_fork_event;
+	event_type_handler_t	process_exit_event;
+	event_type_handler_t	process_lost_event;
+	event_type_handler_t	process_read_event;
+	event_type_handler_t	process_throttle_event;
+	event_type_handler_t	process_unthrottle_event;
+	int			(*sample_type_check)(u64 sample_type);
+	unsigned long		total_unknown;
+};
+
+void register_perf_file_handler(struct perf_file_handler *handler);
+int mmap_dispatch_perf_file(struct perf_header **pheader,
+			    const char *input_name,
+			    int force,
+			    int full_paths,
+			    int *cwdlen,
+			    char **cwd);
+
+#endif
diff --git a/tools/perf/util/debug.h b/tools/perf/util/debug.h
index 437eea58ce40..02d1fa1c2465 100644
--- a/tools/perf/util/debug.h
+++ b/tools/perf/util/debug.h
@@ -1,4 +1,6 @@
 /* For debugging general purposes */
+#ifndef __PERF_DEBUG_H
+#define __PERF_DEBUG_H
 
 extern int verbose;
 extern int dump_trace;
@@ -6,3 +8,5 @@ extern int dump_trace;
 int eprintf(const char *fmt, ...) __attribute__((format(printf, 1, 2)));
 int dump_printf(const char *fmt, ...) __attribute__((format(printf, 1, 2)));
 void trace_event(event_t *event);
+
+#endif	/* __PERF_DEBUG_H */
diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h
index fa2d4e91d329..c2e62be62798 100644
--- a/tools/perf/util/event.h
+++ b/tools/perf/util/event.h
@@ -1,14 +1,10 @@
-#ifndef __PERF_EVENT_H
-#define __PERF_EVENT_H
+#ifndef __PERF_RECORD_H
+#define __PERF_RECORD_H
+
 #include "../perf.h"
 #include "util.h"
 #include <linux/list.h>
-
-enum {
-	SHOW_KERNEL	= 1,
-	SHOW_USER	= 2,
-	SHOW_HV		= 4,
-};
+#include <linux/rbtree.h>
 
 /*
  * PERF_SAMPLE_IP | PERF_SAMPLE_TID | *
@@ -39,6 +35,7 @@ struct fork_event {
 	struct perf_event_header header;
 	u32 pid, ppid;
 	u32 tid, ptid;
+	u64 time;
 };
 
 struct lost_event {
@@ -52,13 +49,19 @@ struct lost_event {
  */
 struct read_event {
 	struct perf_event_header header;
-	u32 pid,tid;
+	u32 pid, tid;
 	u64 value;
 	u64 time_enabled;
 	u64 time_running;
 	u64 id;
 };
 
+struct sample_event{
+	struct perf_event_header        header;
+	u64 array[];
+};
+
+
 typedef union event_union {
 	struct perf_event_header	header;
 	struct ip_event			ip;
@@ -67,10 +70,14 @@ typedef union event_union {
 	struct fork_event		fork;
 	struct lost_event		lost;
 	struct read_event		read;
+	struct sample_event		sample;
 } event_t;
 
 struct map {
-	struct list_head	node;
+	union {
+		struct rb_node	rb_node;
+		struct list_head node;
+	};
 	u64			start;
 	u64			end;
 	u64			pgoff;
@@ -93,4 +100,4 @@ 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);
 
-#endif
+#endif /* __PERF_RECORD_H */
diff --git a/tools/perf/util/exec_cmd.h b/tools/perf/util/exec_cmd.h
index effe25eb1545..31647ac92ed1 100644
--- a/tools/perf/util/exec_cmd.h
+++ b/tools/perf/util/exec_cmd.h
@@ -1,5 +1,5 @@
-#ifndef PERF_EXEC_CMD_H
-#define PERF_EXEC_CMD_H
+#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);
@@ -10,4 +10,4 @@ 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 */
+#endif /* __PERF_EXEC_CMD_H */
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index ec4d4c2f9522..9aae360c0f28 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -5,12 +5,13 @@
 
 #include "util.h"
 #include "header.h"
+#include "../perf.h"
+#include "trace-event.h"
 
 /*
- *
+ * Create new perf.data header attribute:
  */
-
-struct perf_header_attr *perf_header_attr__new(struct perf_counter_attr *attr)
+struct perf_header_attr *perf_header_attr__new(struct perf_event_attr *attr)
 {
 	struct perf_header_attr *self = malloc(sizeof(*self));
 
@@ -43,9 +44,8 @@ void perf_header_attr__add_id(struct perf_header_attr *self, u64 id)
 }
 
 /*
- *
+ * Create new perf.data header:
  */
-
 struct perf_header *perf_header__new(void)
 {
 	struct perf_header *self = malloc(sizeof(*self));
@@ -64,6 +64,8 @@ struct perf_header *perf_header__new(void)
 
 	self->data_offset = 0;
 	self->data_size = 0;
+	self->trace_info_offset = 0;
+	self->trace_info_size = 0;
 
 	return self;
 }
@@ -86,6 +88,46 @@ void perf_header__add_attr(struct perf_header *self,
 	self->attr[pos] = attr;
 }
 
+#define MAX_EVENT_NAME 64
+
+struct perf_trace_event_type {
+	u64	event_id;
+	char	name[MAX_EVENT_NAME];
+};
+
+static int event_count;
+static struct perf_trace_event_type *events;
+
+void perf_header__push_event(u64 id, const char *name)
+{
+	if (strlen(name) > MAX_EVENT_NAME)
+		printf("Event %s will be truncated\n", name);
+
+	if (!events) {
+		events = malloc(sizeof(struct perf_trace_event_type));
+		if (!events)
+			die("nomem");
+	} else {
+		events = realloc(events, (event_count + 1) * sizeof(struct perf_trace_event_type));
+		if (!events)
+			die("nomem");
+	}
+	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++;
+}
+
+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)
@@ -96,7 +138,7 @@ struct perf_file_section {
 };
 
 struct perf_file_attr {
-	struct perf_counter_attr	attr;
+	struct perf_event_attr	attr;
 	struct perf_file_section	ids;
 };
 
@@ -106,8 +148,17 @@ struct perf_file_header {
 	u64				attr_size;
 	struct perf_file_section	attrs;
 	struct perf_file_section	data;
+	struct perf_file_section	event_types;
+	struct perf_file_section	trace_info;
 };
 
+static int trace_info;
+
+void perf_header__set_trace_info(void)
+{
+	trace_info = 1;
+}
+
 static void do_write(int fd, void *buf, size_t size)
 {
 	while (size) {
@@ -154,6 +205,28 @@ void perf_header__write(struct perf_header *self, int fd)
 		do_write(fd, &f_attr, sizeof(f_attr));
 	}
 
+	self->event_offset = lseek(fd, 0, SEEK_CUR);
+	self->event_size = event_count * sizeof(struct perf_trace_event_type);
+	if (events)
+		do_write(fd, events, self->event_size);
+
+	if (trace_info) {
+		static int trace_info_written;
+
+		/*
+		 * Write it only once
+		 */
+		if (!trace_info_written) {
+			self->trace_info_offset = lseek(fd, 0, SEEK_CUR);
+			read_tracing_data(fd, attrs, nr_counters);
+			self->trace_info_size = lseek(fd, 0, SEEK_CUR) -
+						self->trace_info_offset;
+			trace_info_written = 1;
+		} else {
+			lseek(fd, self->trace_info_offset +
+				self->trace_info_size, SEEK_SET);
+		}
+	}
 
 	self->data_offset = lseek(fd, 0, SEEK_CUR);
 
@@ -169,6 +242,14 @@ void perf_header__write(struct perf_header *self, int fd)
 			.offset = self->data_offset,
 			.size	= self->data_size,
 		},
+		.event_types = {
+			.offset = self->event_offset,
+			.size	= self->event_size,
+		},
+		.trace_info = {
+			.offset = self->trace_info_offset,
+			.size = self->trace_info_size,
+		},
 	};
 
 	lseek(fd, 0, SEEK_SET);
@@ -206,10 +287,16 @@ struct perf_header *perf_header__read(int fd)
 	do_read(fd, &f_header, sizeof(f_header));
 
 	if (f_header.magic	!= PERF_MAGIC		||
-	    f_header.size	!= sizeof(f_header)	||
 	    f_header.attr_size	!= sizeof(f_attr))
 		die("incompatible file format");
 
+	if (f_header.size != sizeof(f_header)) {
+		/* Support the previous format */
+		if (f_header.size == offsetof(typeof(f_header), trace_info))
+			f_header.trace_info.size = 0;
+		else
+			die("incompatible file format");
+	}
 	nr_attrs = f_header.attrs.size / sizeof(f_attr);
 	lseek(fd, f_header.attrs.offset, SEEK_SET);
 
@@ -234,6 +321,26 @@ struct perf_header *perf_header__read(int fd)
 		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)
+			die("nomem");
+		do_read(fd, events, f_header.event_types.size);
+		event_count =  f_header.event_types.size / sizeof(struct perf_trace_event_type);
+	}
+
+	self->trace_info_offset = f_header.trace_info.offset;
+	self->trace_info_size = f_header.trace_info.size;
+
+	if (self->trace_info_size) {
+		lseek(fd, self->trace_info_offset, SEEK_SET);
+		trace_report(fd);
+	}
+
+	self->event_offset = f_header.event_types.offset;
+	self->event_size   = f_header.event_types.size;
+
 	self->data_offset = f_header.data.offset;
 	self->data_size   = f_header.data.size;
 
@@ -261,7 +368,7 @@ u64 perf_header__sample_type(struct perf_header *header)
 	return type;
 }
 
-struct perf_counter_attr *
+struct perf_event_attr *
 perf_header__find_attr(u64 id, struct perf_header *header)
 {
 	int i;
diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h
index 5d0a72ecc919..30aee5160dc0 100644
--- a/tools/perf/util/header.h
+++ b/tools/perf/util/header.h
@@ -1,12 +1,12 @@
-#ifndef _PERF_HEADER_H
-#define _PERF_HEADER_H
+#ifndef __PERF_HEADER_H
+#define __PERF_HEADER_H
 
-#include "../../../include/linux/perf_counter.h"
+#include "../../../include/linux/perf_event.h"
 #include <sys/types.h>
 #include "types.h"
 
 struct perf_header_attr {
-	struct perf_counter_attr attr;
+	struct perf_event_attr attr;
 	int ids, size;
 	u64 *id;
 	off_t id_offset;
@@ -19,6 +19,10 @@ struct perf_header {
 	s64 attr_offset;
 	u64 data_offset;
 	u64 data_size;
+	u64 event_offset;
+	u64 event_size;
+	u64 trace_info_offset;
+	u64 trace_info_size;
 };
 
 struct perf_header *perf_header__read(int fd);
@@ -27,15 +31,19 @@ void perf_header__write(struct perf_header *self, int fd);
 void perf_header__add_attr(struct perf_header *self,
 			   struct perf_header_attr *attr);
 
+void perf_header__push_event(u64 id, const char *name);
+char *perf_header__find_event(u64 id);
+
+
 struct perf_header_attr *
-perf_header_attr__new(struct perf_counter_attr *attr);
+perf_header_attr__new(struct perf_event_attr *attr);
 void perf_header_attr__add_id(struct perf_header_attr *self, u64 id);
 
 u64 perf_header__sample_type(struct perf_header *header);
-struct perf_counter_attr *
+struct perf_event_attr *
 perf_header__find_attr(u64 id, struct perf_header *header);
-
+void perf_header__set_trace_info(void);
 
 struct perf_header *perf_header__new(void);
 
-#endif /* _PERF_HEADER_H */
+#endif /* __PERF_HEADER_H */
diff --git a/tools/perf/util/help.h b/tools/perf/util/help.h
index 7128783637b4..7f5c6dedd714 100644
--- a/tools/perf/util/help.h
+++ b/tools/perf/util/help.h
@@ -1,5 +1,5 @@
-#ifndef HELP_H
-#define HELP_H
+#ifndef __PERF_HELP_H
+#define __PERF_HELP_H
 
 struct cmdnames {
 	size_t alloc;
@@ -26,4 +26,4 @@ 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 /* HELP_H */
+#endif /* __PERF_HELP_H */
diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
new file mode 100644
index 000000000000..7393a02fd8d4
--- /dev/null
+++ b/tools/perf/util/hist.c
@@ -0,0 +1,210 @@
+#include "hist.h"
+
+struct rb_root hist;
+struct rb_root collapse_hists;
+struct rb_root output_hists;
+int callchain;
+
+struct callchain_param	callchain_param = {
+	.mode	= CHAIN_GRAPH_REL,
+	.min_percent = 0.5
+};
+
+unsigned long total;
+unsigned long total_mmap;
+unsigned long total_comm;
+unsigned long total_fork;
+unsigned long total_unknown;
+unsigned long total_lost;
+
+/*
+ * histogram, sorted on item, collects counts
+ */
+
+struct hist_entry *__hist_entry__add(struct thread *thread, struct map *map,
+				     struct symbol *sym,
+				     struct symbol *sym_parent,
+				     u64 ip, u64 count, char level, bool *hit)
+{
+	struct rb_node **p = &hist.rb_node;
+	struct rb_node *parent = NULL;
+	struct hist_entry *he;
+	struct hist_entry entry = {
+		.thread	= thread,
+		.map	= map,
+		.sym	= sym,
+		.ip	= ip,
+		.level	= level,
+		.count	= count,
+		.parent = 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) {
+			*hit = true;
+			return he;
+		}
+
+		if (cmp < 0)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	he = malloc(sizeof(*he));
+	if (!he)
+		return NULL;
+	*he = entry;
+	rb_link_node(&he->rb_node, parent, p);
+	rb_insert_color(&he->rb_node, &hist);
+	*hit = false;
+	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->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->collapse ?: se->cmp;
+
+		cmp = f(left, right);
+		if (cmp)
+			break;
+	}
+
+	return cmp;
+}
+
+void hist_entry__free(struct hist_entry *he)
+{
+	free(he);
+}
+
+/*
+ * collapse the histogram
+ */
+
+void collapse__insert_entry(struct hist_entry *he)
+{
+	struct rb_node **p = &collapse_hists.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->count += he->count;
+			hist_entry__free(he);
+			return;
+		}
+
+		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, &collapse_hists);
+}
+
+void collapse__resort(void)
+{
+	struct rb_node *next;
+	struct hist_entry *n;
+
+	if (!sort__need_collapse)
+		return;
+
+	next = rb_first(&hist);
+	while (next) {
+		n = rb_entry(next, struct hist_entry, rb_node);
+		next = rb_next(&n->rb_node);
+
+		rb_erase(&n->rb_node, &hist);
+		collapse__insert_entry(n);
+	}
+}
+
+/*
+ * reverse the map, sort on count.
+ */
+
+void output__insert_entry(struct hist_entry *he, u64 min_callchain_hits)
+{
+	struct rb_node **p = &output_hists.rb_node;
+	struct rb_node *parent = NULL;
+	struct hist_entry *iter;
+
+	if (callchain)
+		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->count > iter->count)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	rb_link_node(&he->rb_node, parent, p);
+	rb_insert_color(&he->rb_node, &output_hists);
+}
+
+void output__resort(u64 total_samples)
+{
+	struct rb_node *next;
+	struct hist_entry *n;
+	struct rb_root *tree = &hist;
+	u64 min_callchain_hits;
+
+	min_callchain_hits =
+		total_samples * (callchain_param.min_percent / 100);
+
+	if (sort__need_collapse)
+		tree = &collapse_hists;
+
+	next = rb_first(tree);
+
+	while (next) {
+		n = rb_entry(next, struct hist_entry, rb_node);
+		next = rb_next(&n->rb_node);
+
+		rb_erase(&n->rb_node, tree);
+		output__insert_entry(n, min_callchain_hits);
+	}
+}
diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h
new file mode 100644
index 000000000000..ac2149c559b0
--- /dev/null
+++ b/tools/perf/util/hist.h
@@ -0,0 +1,50 @@
+#ifndef __PERF_HIST_H
+#define __PERF_HIST_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 struct rb_root hist;
+extern struct rb_root collapse_hists;
+extern struct rb_root output_hists;
+extern int callchain;
+extern struct callchain_param callchain_param;
+extern unsigned long total;
+extern unsigned long total_mmap;
+extern unsigned long total_comm;
+extern unsigned long total_fork;
+extern unsigned long total_unknown;
+extern unsigned long total_lost;
+
+struct hist_entry *__hist_entry__add(struct thread *thread, struct map *map,
+				     struct symbol *sym, struct symbol *parent,
+				     u64 ip, u64 count, char level, bool *hit);
+extern int64_t hist_entry__cmp(struct hist_entry *, struct hist_entry *);
+extern int64_t hist_entry__collapse(struct hist_entry *, struct hist_entry *);
+extern void hist_entry__free(struct hist_entry *);
+extern void collapse__insert_entry(struct hist_entry *);
+extern void collapse__resort(void);
+extern void output__insert_entry(struct hist_entry *, u64);
+extern void output__resort(u64);
+
+#endif	/* __PERF_HIST_H */
diff --git a/tools/perf/util/levenshtein.h b/tools/perf/util/levenshtein.h
index 0173abeef52c..b0fcb6d8a881 100644
--- a/tools/perf/util/levenshtein.h
+++ b/tools/perf/util/levenshtein.h
@@ -1,8 +1,8 @@
-#ifndef LEVENSHTEIN_H
-#define LEVENSHTEIN_H
+#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
+#endif /* __PERF_LEVENSHTEIN_H */
diff --git a/tools/perf/util/module.c b/tools/perf/util/module.c
deleted file mode 100644
index 3d567fe59c79..000000000000
--- a/tools/perf/util/module.c
+++ /dev/null
@@ -1,509 +0,0 @@
-#include "util.h"
-#include "../perf.h"
-#include "string.h"
-#include "module.h"
-
-#include <libelf.h>
-#include <gelf.h>
-#include <elf.h>
-#include <dirent.h>
-#include <sys/utsname.h>
-
-static unsigned int crc32(const char *p, unsigned int len)
-{
-	int i;
-	unsigned int crc = 0;
-
-	while (len--) {
-		crc ^= *p++;
-		for (i = 0; i < 8; i++)
-			crc = (crc >> 1) ^ ((crc & 1) ? 0xedb88320 : 0);
-	}
-	return crc;
-}
-
-/* module section methods */
-
-struct sec_dso *sec_dso__new_dso(const char *name)
-{
-	struct sec_dso *self = malloc(sizeof(*self) + strlen(name) + 1);
-
-	if (self != NULL) {
-		strcpy(self->name, name);
-		self->secs = RB_ROOT;
-		self->find_section = sec_dso__find_section;
-	}
-
-	return self;
-}
-
-static void sec_dso__delete_section(struct section *self)
-{
-	free(((void *)self));
-}
-
-void sec_dso__delete_sections(struct sec_dso *self)
-{
-	struct section *pos;
-	struct rb_node *next = rb_first(&self->secs);
-
-	while (next) {
-		pos = rb_entry(next, struct section, rb_node);
-		next = rb_next(&pos->rb_node);
-		rb_erase(&pos->rb_node, &self->secs);
-		sec_dso__delete_section(pos);
-	}
-}
-
-void sec_dso__delete_self(struct sec_dso *self)
-{
-	sec_dso__delete_sections(self);
-	free(self);
-}
-
-static void sec_dso__insert_section(struct sec_dso *self, struct section *sec)
-{
-	struct rb_node **p = &self->secs.rb_node;
-	struct rb_node *parent = NULL;
-	const u64 hash = sec->hash;
-	struct section *s;
-
-	while (*p != NULL) {
-		parent = *p;
-		s = rb_entry(parent, struct section, rb_node);
-		if (hash < s->hash)
-			p = &(*p)->rb_left;
-		else
-			p = &(*p)->rb_right;
-	}
-	rb_link_node(&sec->rb_node, parent, p);
-	rb_insert_color(&sec->rb_node, &self->secs);
-}
-
-struct section *sec_dso__find_section(struct sec_dso *self, const char *name)
-{
-	struct rb_node *n;
-	u64 hash;
-	int len;
-
-	if (self == NULL)
-		return NULL;
-
-	len = strlen(name);
-	hash = crc32(name, len);
-
-	n = self->secs.rb_node;
-
-	while (n) {
-		struct section *s = rb_entry(n, struct section, rb_node);
-
-		if (hash < s->hash)
-			n = n->rb_left;
-		else if (hash > s->hash)
-			n = n->rb_right;
-		else {
-			if (!strcmp(name, s->name))
-				return s;
-			else
-				n = rb_next(&s->rb_node);
-		}
-	}
-
-	return NULL;
-}
-
-static size_t sec_dso__fprintf_section(struct section *self, FILE *fp)
-{
-	return fprintf(fp, "name:%s vma:%llx path:%s\n",
-		       self->name, self->vma, self->path);
-}
-
-size_t sec_dso__fprintf(struct sec_dso *self, FILE *fp)
-{
-	size_t ret = fprintf(fp, "dso: %s\n", self->name);
-
-	struct rb_node *nd;
-	for (nd = rb_first(&self->secs); nd; nd = rb_next(nd)) {
-		struct section *pos = rb_entry(nd, struct section, rb_node);
-		ret += sec_dso__fprintf_section(pos, fp);
-	}
-
-	return ret;
-}
-
-static struct section *section__new(const char *name, const char *path)
-{
-	struct section *self = calloc(1, sizeof(*self));
-
-	if (!self)
-		goto out_failure;
-
-	self->name = calloc(1, strlen(name) + 1);
-	if (!self->name)
-		goto out_failure;
-
-	self->path = calloc(1, strlen(path) + 1);
-	if (!self->path)
-		goto out_failure;
-
-	strcpy(self->name, name);
-	strcpy(self->path, path);
-	self->hash = crc32(self->name, strlen(name));
-
-	return self;
-
-out_failure:
-	if (self) {
-		if (self->name)
-			free(self->name);
-		if (self->path)
-			free(self->path);
-		free(self);
-	}
-
-	return NULL;
-}
-
-/* module methods */
-
-struct mod_dso *mod_dso__new_dso(const char *name)
-{
-	struct mod_dso *self = malloc(sizeof(*self) + strlen(name) + 1);
-
-	if (self != NULL) {
-		strcpy(self->name, name);
-		self->mods = RB_ROOT;
-		self->find_module = mod_dso__find_module;
-	}
-
-	return self;
-}
-
-static void mod_dso__delete_module(struct module *self)
-{
-	free(((void *)self));
-}
-
-void mod_dso__delete_modules(struct mod_dso *self)
-{
-	struct module *pos;
-	struct rb_node *next = rb_first(&self->mods);
-
-	while (next) {
-		pos = rb_entry(next, struct module, rb_node);
-		next = rb_next(&pos->rb_node);
-		rb_erase(&pos->rb_node, &self->mods);
-		mod_dso__delete_module(pos);
-	}
-}
-
-void mod_dso__delete_self(struct mod_dso *self)
-{
-	mod_dso__delete_modules(self);
-	free(self);
-}
-
-static void mod_dso__insert_module(struct mod_dso *self, struct module *mod)
-{
-	struct rb_node **p = &self->mods.rb_node;
-	struct rb_node *parent = NULL;
-	const u64 hash = mod->hash;
-	struct module *m;
-
-	while (*p != NULL) {
-		parent = *p;
-		m = rb_entry(parent, struct module, rb_node);
-		if (hash < m->hash)
-			p = &(*p)->rb_left;
-		else
-			p = &(*p)->rb_right;
-	}
-	rb_link_node(&mod->rb_node, parent, p);
-	rb_insert_color(&mod->rb_node, &self->mods);
-}
-
-struct module *mod_dso__find_module(struct mod_dso *self, const char *name)
-{
-	struct rb_node *n;
-	u64 hash;
-	int len;
-
-	if (self == NULL)
-		return NULL;
-
-	len = strlen(name);
-	hash = crc32(name, len);
-
-	n = self->mods.rb_node;
-
-	while (n) {
-		struct module *m = rb_entry(n, struct module, rb_node);
-
-		if (hash < m->hash)
-			n = n->rb_left;
-		else if (hash > m->hash)
-			n = n->rb_right;
-		else {
-			if (!strcmp(name, m->name))
-				return m;
-			else
-				n = rb_next(&m->rb_node);
-		}
-	}
-
-	return NULL;
-}
-
-static size_t mod_dso__fprintf_module(struct module *self, FILE *fp)
-{
-	return fprintf(fp, "name:%s path:%s\n", self->name, self->path);
-}
-
-size_t mod_dso__fprintf(struct mod_dso *self, FILE *fp)
-{
-	struct rb_node *nd;
-	size_t ret;
-
-	ret = fprintf(fp, "dso: %s\n", self->name);
-
-	for (nd = rb_first(&self->mods); nd; nd = rb_next(nd)) {
-		struct module *pos = rb_entry(nd, struct module, rb_node);
-
-		ret += mod_dso__fprintf_module(pos, fp);
-	}
-
-	return ret;
-}
-
-static struct module *module__new(const char *name, const char *path)
-{
-	struct module *self = calloc(1, sizeof(*self));
-
-	if (!self)
-		goto out_failure;
-
-	self->name = calloc(1, strlen(name) + 1);
-	if (!self->name)
-		goto out_failure;
-
-	self->path = calloc(1, strlen(path) + 1);
-	if (!self->path)
-		goto out_failure;
-
-	strcpy(self->name, name);
-	strcpy(self->path, path);
-	self->hash = crc32(self->name, strlen(name));
-
-	return self;
-
-out_failure:
-	if (self) {
-		if (self->name)
-			free(self->name);
-		if (self->path)
-			free(self->path);
-		free(self);
-	}
-
-	return NULL;
-}
-
-static int mod_dso__load_sections(struct module *mod)
-{
-	int count = 0, path_len;
-	struct dirent *entry;
-	char *line = NULL;
-	char *dir_path;
-	DIR *dir;
-	size_t n;
-
-	path_len = strlen("/sys/module/");
-	path_len += strlen(mod->name);
-	path_len += strlen("/sections/");
-
-	dir_path = calloc(1, path_len + 1);
-	if (dir_path == NULL)
-		goto out_failure;
-
-	strcat(dir_path, "/sys/module/");
-	strcat(dir_path, mod->name);
-	strcat(dir_path, "/sections/");
-
-	dir = opendir(dir_path);
-	if (dir == NULL)
-		goto out_free;
-
-	while ((entry = readdir(dir))) {
-		struct section *section;
-		char *path, *vma;
-		int line_len;
-		FILE *file;
-
-		if (!strcmp(".", entry->d_name) || !strcmp("..", entry->d_name))
-			continue;
-
-		path = calloc(1, path_len + strlen(entry->d_name) + 1);
-		if (path == NULL)
-			break;
-		strcat(path, dir_path);
-		strcat(path, entry->d_name);
-
-		file = fopen(path, "r");
-		if (file == NULL) {
-			free(path);
-			break;
-		}
-
-		line_len = getline(&line, &n, file);
-		if (line_len < 0) {
-			free(path);
-			fclose(file);
-			break;
-		}
-
-		if (!line) {
-			free(path);
-			fclose(file);
-			break;
-		}
-
-		line[--line_len] = '\0'; /* \n */
-
-		vma = strstr(line, "0x");
-		if (!vma) {
-			free(path);
-			fclose(file);
-			break;
-		}
-		vma += 2;
-
-		section = section__new(entry->d_name, path);
-		if (!section) {
-			fprintf(stderr, "load_sections: allocation error\n");
-			free(path);
-			fclose(file);
-			break;
-		}
-
-		hex2u64(vma, &section->vma);
-		sec_dso__insert_section(mod->sections, section);
-
-		free(path);
-		fclose(file);
-		count++;
-	}
-
-	closedir(dir);
-	free(line);
-	free(dir_path);
-
-	return count;
-
-out_free:
-	free(dir_path);
-
-out_failure:
-	return count;
-}
-
-static int mod_dso__load_module_paths(struct mod_dso *self)
-{
-	struct utsname uts;
-	int count = 0, len;
-	char *line = NULL;
-	FILE *file;
-	char *path;
-	size_t n;
-
-	if (uname(&uts) < 0)
-		goto out_failure;
-
-	len = strlen("/lib/modules/");
-	len += strlen(uts.release);
-	len += strlen("/modules.dep");
-
-	path = calloc(1, len);
-	if (path == NULL)
-		goto out_failure;
-
-	strcat(path, "/lib/modules/");
-	strcat(path, uts.release);
-	strcat(path, "/modules.dep");
-
-	file = fopen(path, "r");
-	free(path);
-	if (file == NULL)
-		goto out_failure;
-
-	while (!feof(file)) {
-		char *name, *tmp;
-		struct module *module;
-		int line_len;
-
-		line_len = getline(&line, &n, file);
-		if (line_len < 0)
-			break;
-
-		if (!line)
-			goto out_failure;
-
-		line[--line_len] = '\0'; /* \n */
-
-		path = strtok(line, ":");
-		if (!path)
-			goto out_failure;
-
-		name = strdup(path);
-		name = strtok(name, "/");
-
-		tmp = name;
-
-		while (tmp) {
-			tmp = strtok(NULL, "/");
-			if (tmp)
-				name = tmp;
-		}
-		name = strsep(&name, ".");
-
-		/* Quirk: replace '-' with '_' in sound modules */
-		for (len = strlen(name); len; len--) {
-			if (*(name+len) == '-')
-				*(name+len) = '_';
-		}
-
-		module = module__new(name, path);
-		if (!module) {
-			fprintf(stderr, "load_module_paths: allocation error\n");
-			goto out_failure;
-		}
-		mod_dso__insert_module(self, module);
-
-		module->sections = sec_dso__new_dso("sections");
-		if (!module->sections) {
-			fprintf(stderr, "load_module_paths: allocation error\n");
-			goto out_failure;
-		}
-
-		module->active = mod_dso__load_sections(module);
-
-		if (module->active > 0)
-			count++;
-	}
-
-	free(line);
-	fclose(file);
-
-	return count;
-
-out_failure:
-	return -1;
-}
-
-int mod_dso__load_modules(struct mod_dso *dso)
-{
-	int err;
-
-	err = mod_dso__load_module_paths(dso);
-
-	return err;
-}
diff --git a/tools/perf/util/module.h b/tools/perf/util/module.h
deleted file mode 100644
index 8a592ef641ca..000000000000
--- a/tools/perf/util/module.h
+++ /dev/null
@@ -1,53 +0,0 @@
-#ifndef _PERF_MODULE_
-#define _PERF_MODULE_ 1
-
-#include <linux/types.h>
-#include "../types.h"
-#include <linux/list.h>
-#include <linux/rbtree.h>
-
-struct section {
-	struct rb_node	rb_node;
-	u64		hash;
-	u64		vma;
-	char		*name;
-	char		*path;
-};
-
-struct sec_dso {
-	struct list_head node;
-	struct rb_root	 secs;
-	struct section    *(*find_section)(struct sec_dso *, const char *name);
-	char		 name[0];
-};
-
-struct module {
-	struct rb_node	rb_node;
-	u64		hash;
-	char		*name;
-	char		*path;
-	struct sec_dso	*sections;
-	int		active;
-};
-
-struct mod_dso {
-	struct list_head node;
-	struct rb_root	 mods;
-	struct module    *(*find_module)(struct mod_dso *, const char *name);
-	char		 name[0];
-};
-
-struct sec_dso *sec_dso__new_dso(const char *name);
-void sec_dso__delete_sections(struct sec_dso *self);
-void sec_dso__delete_self(struct sec_dso *self);
-size_t sec_dso__fprintf(struct sec_dso *self, FILE *fp);
-struct section *sec_dso__find_section(struct sec_dso *self, const char *name);
-
-struct mod_dso *mod_dso__new_dso(const char *name);
-void mod_dso__delete_modules(struct mod_dso *self);
-void mod_dso__delete_self(struct mod_dso *self);
-size_t mod_dso__fprintf(struct mod_dso *self, FILE *fp);
-struct module *mod_dso__find_module(struct mod_dso *self, const char *name);
-int mod_dso__load_modules(struct mod_dso *dso);
-
-#endif /* _PERF_MODULE_ */
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index a587d41ae3c9..b097570e9623 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -6,10 +6,12 @@
 #include "exec_cmd.h"
 #include "string.h"
 #include "cache.h"
+#include "header.h"
 
-int					nr_counters;
+int				nr_counters;
 
-struct perf_counter_attr		attrs[MAX_COUNTERS];
+struct perf_event_attr		attrs[MAX_COUNTERS];
+char				*filters[MAX_COUNTERS];
 
 struct event_symbol {
 	u8		type;
@@ -18,6 +20,12 @@ struct event_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
@@ -41,13 +49,13 @@ static struct event_symbol event_symbols[] = {
   { CSW(CPU_MIGRATIONS),	"cpu-migrations",	"migrations"	},
 };
 
-#define __PERF_COUNTER_FIELD(config, name) \
-	((config & PERF_COUNTER_##name##_MASK) >> PERF_COUNTER_##name##_SHIFT)
+#define __PERF_EVENT_FIELD(config, name) \
+	((config & PERF_EVENT_##name##_MASK) >> PERF_EVENT_##name##_SHIFT)
 
-#define PERF_COUNTER_RAW(config)	__PERF_COUNTER_FIELD(config, RAW)
-#define PERF_COUNTER_CONFIG(config)	__PERF_COUNTER_FIELD(config, CONFIG)
-#define PERF_COUNTER_TYPE(config)	__PERF_COUNTER_FIELD(config, TYPE)
-#define PERF_COUNTER_ID(config)		__PERF_COUNTER_FIELD(config, EVENT)
+#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[] = {
 	"cycles",
@@ -139,7 +147,7 @@ static int tp_event_has_id(struct dirent *sys_dir, struct dirent *evt_dir)
 	   (strcmp(evt_dirent.d_name, "..")) &&				       \
 	   (!tp_event_has_id(&sys_dirent, &evt_dirent)))
 
-#define MAX_EVENT_LENGTH 30
+#define MAX_EVENT_LENGTH 512
 
 int valid_debugfs_mount(const char *debugfs)
 {
@@ -158,33 +166,31 @@ struct tracepoint_path *tracepoint_id_to_path(u64 config)
 	DIR *sys_dir, *evt_dir;
 	struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent;
 	char id_buf[4];
-	int sys_dir_fd, fd;
+	int fd;
 	u64 id;
 	char evt_path[MAXPATHLEN];
+	char dir_path[MAXPATHLEN];
 
 	if (valid_debugfs_mount(debugfs_path))
 		return NULL;
 
 	sys_dir = opendir(debugfs_path);
 	if (!sys_dir)
-		goto cleanup;
-	sys_dir_fd = dirfd(sys_dir);
+		return NULL;
 
 	for_each_subsystem(sys_dir, sys_dirent, sys_next) {
-		int dfd = openat(sys_dir_fd, sys_dirent.d_name,
-				 O_RDONLY|O_DIRECTORY), evt_dir_fd;
-		if (dfd == -1)
-			continue;
-		evt_dir = fdopendir(dfd);
-		if (!evt_dir) {
-			close(dfd);
+
+		snprintf(dir_path, MAXPATHLEN, "%s/%s", debugfs_path,
+			 sys_dirent.d_name);
+		evt_dir = opendir(dir_path);
+		if (!evt_dir)
 			continue;
-		}
-		evt_dir_fd = dirfd(evt_dir);
+
 		for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) {
-			snprintf(evt_path, MAXPATHLEN, "%s/id",
+
+			snprintf(evt_path, MAXPATHLEN, "%s/%s/id", dir_path,
 				 evt_dirent.d_name);
-			fd = openat(evt_dir_fd, evt_path, O_RDONLY);
+			fd = open(evt_path, O_RDONLY);
 			if (fd < 0)
 				continue;
 			if (read(fd, id_buf, sizeof(id_buf)) < 0) {
@@ -218,7 +224,6 @@ struct tracepoint_path *tracepoint_id_to_path(u64 config)
 		closedir(evt_dir);
 	}
 
-cleanup:
 	closedir(sys_dir);
 	return NULL;
 }
@@ -344,8 +349,8 @@ static int parse_aliases(const char **str, const char *names[][MAX_ALIASES], int
 	return -1;
 }
 
-static int
-parse_generic_hw_event(const char **str, struct perf_counter_attr *attr)
+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;
@@ -356,7 +361,7 @@ parse_generic_hw_event(const char **str, struct perf_counter_attr *attr)
 	 * then bail out:
 	 */
 	if (cache_type == -1)
-		return 0;
+		return EVT_FAILED;
 
 	while ((cache_op == -1 || cache_result == -1) && *s == '-') {
 		++s;
@@ -402,27 +407,115 @@ parse_generic_hw_event(const char **str, struct perf_counter_attr *attr)
 	attr->type = PERF_TYPE_HW_CACHE;
 
 	*str = s;
-	return 1;
+	return EVT_HANDLED;
+}
+
+static enum event_result
+parse_single_tracepoint_event(char *sys_name,
+			      const char *evt_name,
+			      unsigned int evt_length,
+			      char *flags,
+			      struct perf_event_attr *attr,
+			      const char **strp)
+{
+	char evt_path[MAXPATHLEN];
+	char id_buf[4];
+	u64 id;
+	int fd;
+
+	if (flags) {
+		if (!strncmp(flags, "record", strlen(flags))) {
+			attr->sample_type |= PERF_SAMPLE_RAW;
+			attr->sample_type |= PERF_SAMPLE_TIME;
+			attr->sample_type |= PERF_SAMPLE_CPU;
+		}
+	}
+
+	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 = evt_name + evt_length;
+
+	return EVT_HANDLED;
 }
 
-static int parse_tracepoint_event(const char **strp,
-				    struct perf_counter_attr *attr)
+/* sys + ':' + event + ':' + flags*/
+#define MAX_EVOPT_LEN	(MAX_EVENT_LENGTH * 2 + 2 + 128)
+static enum event_result
+parse_subsystem_tracepoint_event(char *sys_name, 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;
+		unsigned int rem = MAX_EVOPT_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;
+
+		len = snprintf(event_opt, MAX_EVOPT_LEN, "%s:%s", sys_name,
+			       evt_ent->d_name);
+		if (len < 0)
+			return EVT_FAILED;
+
+		rem -= len;
+		if (flags) {
+			if (rem < strlen(flags) + 1)
+				return EVT_FAILED;
+
+			strcat(event_opt, ":");
+			strcat(event_opt, flags);
+		}
+
+		if (parse_events(NULL, event_opt, 0))
+			return EVT_FAILED;
+	}
+
+	return EVT_HANDLED_ALL;
+}
+
+
+static enum event_result parse_tracepoint_event(const char **strp,
+				    struct perf_event_attr *attr)
 {
 	const char *evt_name;
 	char *flags;
 	char sys_name[MAX_EVENT_LENGTH];
-	char id_buf[4];
-	int fd;
 	unsigned int sys_length, evt_length;
-	u64 id;
-	char evt_path[MAXPATHLEN];
 
 	if (valid_debugfs_mount(debugfs_path))
 		return 0;
 
 	evt_name = strchr(*strp, ':');
 	if (!evt_name)
-		return 0;
+		return EVT_FAILED;
 
 	sys_length = evt_name - *strp;
 	if (sys_length >= MAX_EVENT_LENGTH)
@@ -434,32 +527,22 @@ static int parse_tracepoint_event(const char **strp,
 
 	flags = strchr(evt_name, ':');
 	if (flags) {
-		*flags = '\0';
+		/* split it out: */
+		evt_name = strndup(evt_name, flags - evt_name);
 		flags++;
-		if (!strncmp(flags, "record", strlen(flags)))
-			attr->sample_type |= PERF_SAMPLE_RAW;
 	}
 
 	evt_length = strlen(evt_name);
 	if (evt_length >= MAX_EVENT_LENGTH)
-		return 0;
+		return EVT_FAILED;
 
-	snprintf(evt_path, MAXPATHLEN, "%s/%s/%s/id", debugfs_path,
-		 sys_name, evt_name);
-	fd = open(evt_path, O_RDONLY);
-	if (fd < 0)
-		return 0;
-
-	if (read(fd, id_buf, sizeof(id_buf)) < 0) {
-		close(fd);
-		return 0;
-	}
-	close(fd);
-	id = atoll(id_buf);
-	attr->config = id;
-	attr->type = PERF_TYPE_TRACEPOINT;
-	*strp = evt_name + evt_length;
-	return 1;
+	if (!strcmp(evt_name, "*")) {
+		*strp = evt_name + evt_length;
+		return parse_subsystem_tracepoint_event(sys_name, flags);
+	} else
+		return parse_single_tracepoint_event(sys_name, evt_name,
+						     evt_length, flags,
+						     attr, strp);
 }
 
 static int check_events(const char *str, unsigned int i)
@@ -477,8 +560,8 @@ static int check_events(const char *str, unsigned int i)
 	return 0;
 }
 
-static int
-parse_symbolic_event(const char **strp, struct perf_counter_attr *attr)
+static enum event_result
+parse_symbolic_event(const char **strp, struct perf_event_attr *attr)
 {
 	const char *str = *strp;
 	unsigned int i;
@@ -490,32 +573,33 @@ parse_symbolic_event(const char **strp, struct perf_counter_attr *attr)
 			attr->type = event_symbols[i].type;
 			attr->config = event_symbols[i].config;
 			*strp = str + n;
-			return 1;
+			return EVT_HANDLED;
 		}
 	}
-	return 0;
+	return EVT_FAILED;
 }
 
-static int parse_raw_event(const char **strp, struct perf_counter_attr *attr)
+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 0;
+		return EVT_FAILED;
 	n = hex2u64(str + 1, &config);
 	if (n > 0) {
 		*strp = str + n + 1;
 		attr->type = PERF_TYPE_RAW;
 		attr->config = config;
-		return 1;
+		return EVT_HANDLED;
 	}
-	return 0;
+	return EVT_FAILED;
 }
 
-static int
-parse_numeric_event(const char **strp, struct perf_counter_attr *attr)
+static enum event_result
+parse_numeric_event(const char **strp, struct perf_event_attr *attr)
 {
 	const char *str = *strp;
 	char *endp;
@@ -530,14 +614,14 @@ parse_numeric_event(const char **strp, struct perf_counter_attr *attr)
 			attr->type = type;
 			attr->config = config;
 			*strp = endp;
-			return 1;
+			return EVT_HANDLED;
 		}
 	}
-	return 0;
+	return EVT_FAILED;
 }
 
-static int
-parse_event_modifier(const char **strp, struct perf_counter_attr *attr)
+static enum event_result
+parse_event_modifier(const char **strp, struct perf_event_attr *attr)
 {
 	const char *str = *strp;
 	int eu = 1, ek = 1, eh = 1;
@@ -569,37 +653,86 @@ parse_event_modifier(const char **strp, struct perf_counter_attr *attr)
  * Each event can have multiple symbolic names.
  * Symbolic names are (almost) exactly matched.
  */
-static int parse_event_symbols(const char **str, struct perf_counter_attr *attr)
+static enum event_result
+parse_event_symbols(const char **str, struct perf_event_attr *attr)
 {
-	if (!(parse_tracepoint_event(str, attr) ||
-	      parse_raw_event(str, attr) ||
-	      parse_numeric_event(str, attr) ||
-	      parse_symbolic_event(str, attr) ||
-	      parse_generic_hw_event(str, attr)))
-		return 0;
+	enum event_result ret;
+
+	ret = parse_tracepoint_event(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;
+
+	return EVT_FAILED;
+
+modifier:
 	parse_event_modifier(str, attr);
 
-	return 1;
+	return ret;
+}
+
+static void store_event_type(const char *orgname)
+{
+	char filename[PATH_MAX], *c;
+	FILE *file;
+	int id;
+
+	sprintf(filename, "%s/", debugfs_path);
+	strncat(filename, orgname, strlen(orgname));
+	strcat(filename, "/id");
+
+	c = strchr(filename, ':');
+	if (c)
+		*c = '/';
+
+	file = fopen(filename, "r");
+	if (!file)
+		return;
+	if (fscanf(file, "%i", &id) < 1)
+		die("cannot store event ID");
+	fclose(file);
+	perf_header__push_event(id, orgname);
 }
 
 int parse_events(const struct option *opt __used, const char *str, int unset __used)
 {
-	struct perf_counter_attr attr;
+	struct perf_event_attr attr;
+	enum event_result ret;
+
+	if (strchr(str, ':'))
+		store_event_type(str);
 
 	for (;;) {
 		if (nr_counters == MAX_COUNTERS)
 			return -1;
 
 		memset(&attr, 0, sizeof(attr));
-		if (!parse_event_symbols(&str, &attr))
+		ret = parse_event_symbols(&str, &attr);
+		if (ret == EVT_FAILED)
 			return -1;
 
 		if (!(*str == 0 || *str == ',' || isspace(*str)))
 			return -1;
 
-		attrs[nr_counters] = attr;
-		nr_counters++;
+		if (ret != EVT_HANDLED_ALL) {
+			attrs[nr_counters] = attr;
+			nr_counters++;
+		}
 
 		if (*str == 0)
 			break;
@@ -612,6 +745,28 @@ int parse_events(const struct option *opt __used, const char *str, int unset __u
 	return 0;
 }
 
+int parse_filter(const struct option *opt __used, const char *str,
+		 int unset __used)
+{
+	int i = nr_counters - 1;
+	int len = strlen(str);
+
+	if (i < 0 || attrs[i].type != PERF_TYPE_TRACEPOINT) {
+		fprintf(stderr,
+			"-F option should follow a -e tracepoint option\n");
+		return -1;
+	}
+
+	filters[i] = malloc(len + 1);
+	if (!filters[i]) {
+		fprintf(stderr, "not enough memory to hold filter string\n");
+		return -1;
+	}
+	strcpy(filters[i], str);
+
+	return 0;
+}
+
 static const char * const event_type_descriptors[] = {
 	"",
 	"Hardware event",
@@ -628,28 +783,24 @@ static void print_tracepoint_events(void)
 {
 	DIR *sys_dir, *evt_dir;
 	struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent;
-	int sys_dir_fd;
 	char evt_path[MAXPATHLEN];
+	char dir_path[MAXPATHLEN];
 
 	if (valid_debugfs_mount(debugfs_path))
 		return;
 
 	sys_dir = opendir(debugfs_path);
 	if (!sys_dir)
-		goto cleanup;
-	sys_dir_fd = dirfd(sys_dir);
+		return;
 
 	for_each_subsystem(sys_dir, sys_dirent, sys_next) {
-		int dfd = openat(sys_dir_fd, sys_dirent.d_name,
-				 O_RDONLY|O_DIRECTORY), evt_dir_fd;
-		if (dfd == -1)
-			continue;
-		evt_dir = fdopendir(dfd);
-		if (!evt_dir) {
-			close(dfd);
+
+		snprintf(dir_path, MAXPATHLEN, "%s/%s", debugfs_path,
+			 sys_dirent.d_name);
+		evt_dir = opendir(dir_path);
+		if (!evt_dir)
 			continue;
-		}
-		evt_dir_fd = dirfd(evt_dir);
+
 		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);
@@ -658,8 +809,6 @@ static void print_tracepoint_events(void)
 		}
 		closedir(evt_dir);
 	}
-
-cleanup:
 	closedir(sys_dir);
 }
 
diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h
index 60704c15961f..b8c1f64bc935 100644
--- a/tools/perf/util/parse-events.h
+++ b/tools/perf/util/parse-events.h
@@ -1,5 +1,5 @@
-#ifndef _PARSE_EVENTS_H
-#define _PARSE_EVENTS_H
+#ifndef __PERF_PARSE_EVENTS_H
+#define __PERF_PARSE_EVENTS_H
 /*
  * Parse symbolic events/counts passed in as options:
  */
@@ -16,12 +16,14 @@ extern struct tracepoint_path *tracepoint_id_to_path(u64 config);
 
 extern int			nr_counters;
 
-extern struct perf_counter_attr attrs[MAX_COUNTERS];
+extern struct perf_event_attr attrs[MAX_COUNTERS];
+extern char *filters[MAX_COUNTERS];
 
 extern const char *event_name(int ctr);
 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)
 
@@ -31,4 +33,4 @@ extern char debugfs_path[];
 extern int valid_debugfs_mount(const char *debugfs);
 
 
-#endif /* _PARSE_EVENTS_H */
+#endif /* __PERF_PARSE_EVENTS_H */
diff --git a/tools/perf/util/parse-options.h b/tools/perf/util/parse-options.h
index 8aa3464c7090..948805af43c2 100644
--- a/tools/perf/util/parse-options.h
+++ b/tools/perf/util/parse-options.h
@@ -1,5 +1,5 @@
-#ifndef PARSE_OPTIONS_H
-#define PARSE_OPTIONS_H
+#ifndef __PERF_PARSE_OPTIONS_H
+#define __PERF_PARSE_OPTIONS_H
 
 enum parse_opt_type {
 	/* special types */
@@ -104,6 +104,8 @@ struct option {
 	{ .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 }
 
@@ -172,4 +174,4 @@ extern int parse_opt_verbosity_cb(const struct option *, const char *, int);
 
 extern const char *parse_options_fix_filename(const char *prefix, const char *file);
 
-#endif
+#endif /* __PERF_PARSE_OPTIONS_H */
diff --git a/tools/perf/util/quote.h b/tools/perf/util/quote.h
index a5454a1d1c13..b6a019733919 100644
--- a/tools/perf/util/quote.h
+++ b/tools/perf/util/quote.h
@@ -1,5 +1,5 @@
-#ifndef QUOTE_H
-#define QUOTE_H
+#ifndef __PERF_QUOTE_H
+#define __PERF_QUOTE_H
 
 #include <stddef.h>
 #include <stdio.h>
@@ -65,4 +65,4 @@ extern void perl_quote_print(FILE *stream, const char *src);
 extern void python_quote_print(FILE *stream, const char *src);
 extern void tcl_quote_print(FILE *stream, const char *src);
 
-#endif
+#endif /* __PERF_QUOTE_H */
diff --git a/tools/perf/util/run-command.h b/tools/perf/util/run-command.h
index cc1837deba88..d79028727ce2 100644
--- a/tools/perf/util/run-command.h
+++ b/tools/perf/util/run-command.h
@@ -1,5 +1,5 @@
-#ifndef RUN_COMMAND_H
-#define RUN_COMMAND_H
+#ifndef __PERF_RUN_COMMAND_H
+#define __PERF_RUN_COMMAND_H
 
 enum {
 	ERR_RUN_COMMAND_FORK = 10000,
@@ -85,4 +85,4 @@ struct async {
 int start_async(struct async *async);
 int finish_async(struct async *async);
 
-#endif
+#endif /* __PERF_RUN_COMMAND_H */
diff --git a/tools/perf/util/sigchain.h b/tools/perf/util/sigchain.h
index 618083bce0c6..1a53c11265fd 100644
--- a/tools/perf/util/sigchain.h
+++ b/tools/perf/util/sigchain.h
@@ -1,5 +1,5 @@
-#ifndef SIGCHAIN_H
-#define SIGCHAIN_H
+#ifndef __PERF_SIGCHAIN_H
+#define __PERF_SIGCHAIN_H
 
 typedef void (*sigchain_fun)(int);
 
@@ -8,4 +8,4 @@ int sigchain_pop(int sig);
 
 void sigchain_push_common(sigchain_fun f);
 
-#endif /* SIGCHAIN_H */
+#endif /* __PERF_SIGCHAIN_H */
diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c
new file mode 100644
index 000000000000..40c9acd41cad
--- /dev/null
+++ b/tools/perf/util/sort.c
@@ -0,0 +1,276 @@
+#include "sort.h"
+
+regex_t		parent_regex;
+char		default_parent_pattern[] = "^sys_|^do_page_fault";
+char		*parent_pattern = default_parent_pattern;
+char		default_sort_order[] = "comm,dso,symbol";
+char		*sort_order = default_sort_order;
+int sort__need_collapse = 0;
+int sort__has_parent = 0;
+
+unsigned int dsos__col_width;
+unsigned int comms__col_width;
+unsigned int threads__col_width;
+static unsigned int parent_symbol__col_width;
+char * field_sep;
+
+LIST_HEAD(hist_entry__sort_list);
+
+struct sort_entry sort_thread = {
+	.header = "Command:  Pid",
+	.cmp	= sort__thread_cmp,
+	.print	= sort__thread_print,
+	.width	= &threads__col_width,
+};
+
+struct sort_entry sort_comm = {
+	.header		= "Command",
+	.cmp		= sort__comm_cmp,
+	.collapse	= sort__comm_collapse,
+	.print		= sort__comm_print,
+	.width		= &comms__col_width,
+};
+
+struct sort_entry sort_dso = {
+	.header = "Shared Object",
+	.cmp	= sort__dso_cmp,
+	.print	= sort__dso_print,
+	.width	= &dsos__col_width,
+};
+
+struct sort_entry sort_sym = {
+	.header = "Symbol",
+	.cmp	= sort__sym_cmp,
+	.print	= sort__sym_print,
+};
+
+struct sort_entry sort_parent = {
+	.header = "Parent symbol",
+	.cmp	= sort__parent_cmp,
+	.print	= sort__parent_print,
+	.width	= &parent_symbol__col_width,
+};
+
+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,	},
+};
+
+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;
+}
+
+int repsep_fprintf(FILE *fp, const char *fmt, ...)
+{
+	int n;
+	va_list ap;
+
+	va_start(ap, fmt);
+	if (!field_sep)
+		n = vfprintf(fp, fmt, ap);
+	else {
+		char *bf = NULL;
+		n = vasprintf(&bf, fmt, ap);
+		if (n > 0) {
+			char *sep = bf;
+
+			while (1) {
+				sep = strchr(sep, *field_sep);
+				if (sep == NULL)
+					break;
+				*sep = '.';
+			}
+		}
+		fputs(bf, fp);
+		free(bf);
+	}
+	va_end(ap);
+	return n;
+}
+
+size_t
+sort__thread_print(FILE *fp, struct hist_entry *self, unsigned int width)
+{
+	return repsep_fprintf(fp, "%*s:%5d", width - 6,
+			      self->thread->comm ?: "", self->thread->pid);
+}
+
+size_t
+sort__comm_print(FILE *fp, struct hist_entry *self, unsigned int width)
+{
+	return repsep_fprintf(fp, "%*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->map ? left->map->dso : NULL;
+	struct dso *dso_r = right->map ? right->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);
+}
+
+size_t
+sort__dso_print(FILE *fp, struct hist_entry *self, unsigned int width)
+{
+	if (self->map && self->map->dso) {
+		const char *dso_name = !verbose ? self->map->dso->short_name :
+						  self->map->dso->long_name;
+		return repsep_fprintf(fp, "%-*s", width, dso_name);
+	}
+
+	return repsep_fprintf(fp, "%*llx", width, (u64)self->ip);
+}
+
+/* --sort symbol */
+
+int64_t
+sort__sym_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	u64 ip_l, ip_r;
+
+	if (left->sym == right->sym)
+		return 0;
+
+	ip_l = left->sym ? left->sym->start : left->ip;
+	ip_r = right->sym ? right->sym->start : right->ip;
+
+	return (int64_t)(ip_r - ip_l);
+}
+
+
+size_t
+sort__sym_print(FILE *fp, struct hist_entry *self, unsigned int width __used)
+{
+	size_t ret = 0;
+
+	if (verbose) {
+		char o = self->map ? dso__symtab_origin(self->map->dso) : '!';
+		ret += repsep_fprintf(fp, "%#018llx %c ", (u64)self->ip, o);
+	}
+
+	ret += repsep_fprintf(fp, "[%c] ", self->level);
+	if (self->sym)
+		ret += repsep_fprintf(fp, "%s", self->sym->name);
+	else
+		ret += repsep_fprintf(fp, "%#016llx", (u64)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);
+}
+
+size_t
+sort__parent_print(FILE *fp, struct hist_entry *self, unsigned int width)
+{
+	return repsep_fprintf(fp, "%-*s", width,
+			      self->parent ? self->parent->name : "[other]");
+}
+
+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->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));
+				fprintf(stderr, "Invalid regex: %s\n%s",
+					parent_pattern, err);
+				exit(-1);
+			}
+			sort__has_parent = 1;
+		}
+
+		list_add_tail(&sd->entry->list, &hist_entry__sort_list);
+		sd->taken = 1;
+
+		return 0;
+	}
+
+	return -ESRCH;
+}
+
diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h
new file mode 100644
index 000000000000..13806d782af6
--- /dev/null
+++ b/tools/perf/util/sort.h
@@ -0,0 +1,90 @@
+#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 char *sort_order;
+extern char default_parent_pattern[];
+extern char *parent_pattern;
+extern 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 unsigned int dsos__col_width;
+extern unsigned int comms__col_width;
+extern unsigned int threads__col_width;
+
+struct hist_entry {
+	struct rb_node		rb_node;
+	u64			count;
+	struct thread		*thread;
+	struct map		*map;
+	struct symbol		*sym;
+	u64			ip;
+	char			level;
+	struct symbol		*parent;
+	struct callchain_node	callchain;
+	struct rb_root		sorted_chain;
+};
+
+/*
+ * configurable sorting bits
+ */
+
+struct sort_entry {
+	struct list_head list;
+
+	const char *header;
+
+	int64_t (*cmp)(struct hist_entry *, struct hist_entry *);
+	int64_t (*collapse)(struct hist_entry *, struct hist_entry *);
+	size_t	(*print)(FILE *fp, struct hist_entry *, unsigned int width);
+	unsigned int *width;
+	bool	elide;
+};
+
+extern struct sort_entry sort_thread;
+extern struct list_head hist_entry__sort_list;
+
+extern int repsep_fprintf(FILE *fp, const char *fmt, ...);
+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 *);
+extern size_t sort__parent_print(FILE *, struct hist_entry *, unsigned int);
+extern int sort_dimension__add(const char *);
+
+#endif	/* __PERF_SORT_H */
diff --git a/tools/perf/util/strbuf.h b/tools/perf/util/strbuf.h
index d2aa86c014c1..a3d121d6c83e 100644
--- a/tools/perf/util/strbuf.h
+++ b/tools/perf/util/strbuf.h
@@ -1,5 +1,5 @@
-#ifndef STRBUF_H
-#define STRBUF_H
+#ifndef __PERF_STRBUF_H
+#define __PERF_STRBUF_H
 
 /*
  * Strbuf's can be use in many ways: as a byte array, or to store arbitrary
@@ -134,4 +134,4 @@ extern int launch_editor(const char *path, struct strbuf *buffer, const char *co
 extern int strbuf_branchname(struct strbuf *sb, const char *name);
 extern int strbuf_check_branch_ref(struct strbuf *sb, const char *name);
 
-#endif /* STRBUF_H */
+#endif /* __PERF_STRBUF_H */
diff --git a/tools/perf/util/string.c b/tools/perf/util/string.c
index c93eca9a7be3..04743d3e9039 100644
--- a/tools/perf/util/string.c
+++ b/tools/perf/util/string.c
@@ -1,3 +1,4 @@
+#include <string.h>
 #include "string.h"
 
 static int hex(char ch)
@@ -32,3 +33,13 @@ int hex2u64(const char *ptr, u64 *long_val)
 
 	return p - ptr;
 }
+
+char *strxfrchar(char *s, char from, char to)
+{
+	char *p = s;
+
+	while ((p = strchr(p, from)) != NULL)
+		*p++ = to;
+
+	return s;
+}
diff --git a/tools/perf/util/string.h b/tools/perf/util/string.h
index bf39dfadfd24..2c84bf65ba0f 100644
--- a/tools/perf/util/string.h
+++ b/tools/perf/util/string.h
@@ -1,11 +1,12 @@
-#ifndef _PERF_STRING_H_
-#define _PERF_STRING_H_
+#ifndef __PERF_STRING_H_
+#define __PERF_STRING_H_
 
 #include "types.h"
 
 int hex2u64(const char *ptr, u64 *val);
+char *strxfrchar(char *s, char from, char to);
 
 #define _STR(x) #x
 #define STR(x) _STR(x)
 
-#endif
+#endif /* __PERF_STRING_H */
diff --git a/tools/perf/util/strlist.h b/tools/perf/util/strlist.h
index 921818e44a54..cb4659306d7b 100644
--- a/tools/perf/util/strlist.h
+++ b/tools/perf/util/strlist.h
@@ -1,5 +1,5 @@
-#ifndef STRLIST_H_
-#define STRLIST_H_
+#ifndef __PERF_STRLIST_H
+#define __PERF_STRLIST_H
 
 #include <linux/rbtree.h>
 #include <stdbool.h>
@@ -36,4 +36,4 @@ static inline unsigned int strlist__nr_entries(const struct strlist *self)
 }
 
 int strlist__parse_list(struct strlist *self, const char *s);
-#endif /* STRLIST_H_ */
+#endif /* __PERF_STRLIST_H */
diff --git a/tools/perf/util/svghelper.c b/tools/perf/util/svghelper.c
new file mode 100644
index 000000000000..856655d8b0b8
--- /dev/null
+++ b/tools/perf/util/svghelper.c
@@ -0,0 +1,500 @@
+/*
+ * 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 <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=\"%llu\" 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(214,214,  0); fill-opacity:0.3; 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","rect.c1");
+	svg_legenda_box(200,	"Deeper Idle", "rect.c3");
+	svg_legenda_box(350,	"Deepest Idle", "rect.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=\"%llu\" 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 000000000000..e0781989cc31
--- /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
index fd3d9c8e90fc..faa84f5d4f54 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -2,12 +2,14 @@
 #include "../perf.h"
 #include "string.h"
 #include "symbol.h"
+#include "thread.h"
 
 #include "debug.h"
 
 #include <libelf.h>
 #include <gelf.h>
 #include <elf.h>
+#include <sys/utsname.h>
 
 const char *sym_hist_filter;
 
@@ -18,12 +20,65 @@ enum dso_origin {
 	DSO__ORIG_UBUNTU,
 	DSO__ORIG_BUILDID,
 	DSO__ORIG_DSO,
+	DSO__ORIG_KMODULE,
 	DSO__ORIG_NOT_FOUND,
 };
 
-static struct symbol *symbol__new(u64 start, u64 len,
-				  const char *name, unsigned int priv_size,
-				  u64 obj_start, int v)
+static void dsos__add(struct dso *dso);
+static struct dso *dsos__find(const char *name);
+static struct map *map__new2(u64 start, struct dso *dso);
+static void kernel_maps__insert(struct map *map);
+
+static struct rb_root kernel_maps;
+
+static void dso__fixup_sym_end(struct dso *self)
+{
+	struct rb_node *nd, *prevnd = rb_first(&self->syms);
+	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 - 1;
+	}
+
+	/* Last entry */
+	if (curr->end == curr->start)
+		curr->end = roundup(curr->start, 4096);
+}
+
+static void kernel_maps__fixup_end(void)
+{
+	struct map *prev, *curr;
+	struct rb_node *nd, *prevnd = rb_first(&kernel_maps);
+
+	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;
+	}
+
+	nd = rb_last(&curr->dso->syms);
+	if (nd) {
+		struct symbol *sym = rb_entry(nd, struct symbol, rb_node);
+		curr->end = sym->end;
+	}
+}
+
+static struct symbol *symbol__new(u64 start, u64 len, const char *name,
+				  unsigned int priv_size, int v)
 {
 	size_t namelen = strlen(name) + 1;
 	struct symbol *self = calloc(1, priv_size + sizeof(*self) + namelen);
@@ -31,11 +86,10 @@ static struct symbol *symbol__new(u64 start, u64 len,
 	if (!self)
 		return NULL;
 
-	if (v >= 2)
-		printf("new symbol: %016Lx [%08lx]: %s, hist: %p, obj_start: %p\n",
-			(u64)start, (unsigned long)len, name, self->hist, (void *)(unsigned long)obj_start);
+	if (v > 2)
+		printf("new symbol: %016Lx [%08lx]: %s, hist: %p\n",
+			start, (unsigned long)len, name, self->hist);
 
-	self->obj_start= obj_start;
 	self->hist = NULL;
 	self->hist_sum = 0;
 
@@ -60,12 +114,8 @@ static void symbol__delete(struct symbol *self, unsigned int priv_size)
 
 static size_t symbol__fprintf(struct symbol *self, FILE *fp)
 {
-	if (!self->module)
-		return fprintf(fp, " %llx-%llx %s\n",
+	return fprintf(fp, " %llx-%llx %s\n",
 		       self->start, self->end, self->name);
-	else
-		return fprintf(fp, " %llx-%llx %s \t[%s]\n",
-		       self->start, self->end, self->name, self->module->name);
 }
 
 struct dso *dso__new(const char *name, unsigned int sym_priv_size)
@@ -74,6 +124,8 @@ struct dso *dso__new(const char *name, unsigned int sym_priv_size)
 
 	if (self != NULL) {
 		strcpy(self->name, name);
+		self->long_name = self->name;
+		self->short_name = self->name;
 		self->syms = RB_ROOT;
 		self->sym_priv_size = sym_priv_size;
 		self->find_symbol = dso__find_symbol;
@@ -100,6 +152,8 @@ static void dso__delete_symbols(struct dso *self)
 void dso__delete(struct dso *self)
 {
 	dso__delete_symbols(self);
+	if (self->long_name != self->name)
+		free(self->long_name);
 	free(self);
 }
 
@@ -147,7 +201,7 @@ struct symbol *dso__find_symbol(struct dso *self, u64 ip)
 
 size_t dso__fprintf(struct dso *self, FILE *fp)
 {
-	size_t ret = fprintf(fp, "dso: %s\n", self->name);
+	size_t ret = fprintf(fp, "dso: %s\n", self->short_name);
 
 	struct rb_node *nd;
 	for (nd = rb_first(&self->syms); nd; nd = rb_next(nd)) {
@@ -158,13 +212,16 @@ size_t dso__fprintf(struct dso *self, FILE *fp)
 	return ret;
 }
 
-static int dso__load_kallsyms(struct dso *self, symbol_filter_t filter, int v)
+/*
+ * 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 kernel_maps__load_all_kallsyms(int v)
 {
-	struct rb_node *nd, *prevnd;
 	char *line = NULL;
 	size_t n;
 	FILE *file = fopen("/proc/kallsyms", "r");
-	int count = 0;
 
 	if (file == NULL)
 		goto out_failure;
@@ -174,6 +231,7 @@ static int dso__load_kallsyms(struct dso *self, symbol_filter_t filter, int v)
 		struct symbol *sym;
 		int line_len, len;
 		char symbol_type;
+		char *symbol_name;
 
 		line_len = getline(&line, &n, file);
 		if (line_len < 0)
@@ -196,44 +254,24 @@ static int dso__load_kallsyms(struct dso *self, symbol_filter_t filter, int v)
 		 */
 		if (symbol_type != 'T' && symbol_type != 'W')
 			continue;
+
+		symbol_name = line + len + 2;
 		/*
-		 * Well fix up the end later, when we have all sorted.
+		 * Will fix up the end later, when we have all symbols sorted.
 		 */
-		sym = symbol__new(start, 0xdead, line + len + 2,
-				  self->sym_priv_size, 0, v);
+		sym = symbol__new(start, 0, symbol_name,
+				  kernel_map->dso->sym_priv_size, v);
 
 		if (sym == NULL)
 			goto out_delete_line;
 
-		if (filter && filter(self, sym))
-			symbol__delete(sym, self->sym_priv_size);
-		else {
-			dso__insert_symbol(self, sym);
-			count++;
-		}
-	}
-
-	/*
-	 * Now that we have all sorted out, just set the ->end of all
-	 * symbols
-	 */
-	prevnd = rb_first(&self->syms);
-
-	if (prevnd == NULL)
-		goto out_delete_line;
-
-	for (nd = rb_next(prevnd); nd; nd = rb_next(nd)) {
-		struct symbol *prev = rb_entry(prevnd, struct symbol, rb_node),
-			      *curr = rb_entry(nd, struct symbol, rb_node);
-
-		prev->end = curr->start - 1;
-		prevnd = nd;
+		dso__insert_symbol(kernel_map->dso, sym);
 	}
 
 	free(line);
 	fclose(file);
 
-	return count;
+	return 0;
 
 out_delete_line:
 	free(line);
@@ -241,14 +279,125 @@ out_failure:
 	return -1;
 }
 
-static int dso__load_perf_map(struct dso *self, symbol_filter_t filter, int v)
+/*
+ * 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 kernel_maps__split_kallsyms(symbol_filter_t filter, int use_modules)
+{
+	struct map *map = kernel_map;
+	struct symbol *pos;
+	int count = 0;
+	struct rb_node *next = rb_first(&kernel_map->dso->syms);
+	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 (!use_modules)
+				goto delete_symbol;
+
+			*module++ = '\0';
+
+			if (strcmp(map->dso->name, module)) {
+				map = kernel_maps__find_by_dso_name(module);
+				if (!map) {
+					fputs("/proc/{kallsyms,modules} "
+					      "inconsistency!\n", stderr);
+					return -1;
+				}
+			}
+			/*
+			 * So that we look just like we get from .ko files,
+			 * i.e. not prelinked, relative to map->start.
+			 */
+			pos->start = map->map_ip(map, pos->start);
+			pos->end   = map->map_ip(map, pos->end);
+		} else if (map != kernel_map) {
+			char dso_name[PATH_MAX];
+			struct dso *dso;
+
+			snprintf(dso_name, sizeof(dso_name), "[kernel].%d",
+				 kernel_range++);
+
+			dso = dso__new(dso_name,
+				       kernel_map->dso->sym_priv_size);
+			if (dso == NULL)
+				return -1;
+
+			map = map__new2(pos->start, dso);
+			if (map == NULL) {
+				dso__delete(dso);
+				return -1;
+			}
+
+			map->map_ip = vdso__map_ip;
+			kernel_maps__insert(map);
+			++kernel_range;
+		}
+
+		if (filter && filter(map, pos)) {
+delete_symbol:
+			rb_erase(&pos->rb_node, &kernel_map->dso->syms);
+			symbol__delete(pos, kernel_map->dso->sym_priv_size);
+		} else {
+			if (map != kernel_map) {
+				rb_erase(&pos->rb_node, &kernel_map->dso->syms);
+				dso__insert_symbol(map->dso, pos);
+			}
+			count++;
+		}
+	}
+
+	return count;
+}
+
+
+static int kernel_maps__load_kallsyms(symbol_filter_t filter,
+				      int use_modules, int v)
+{
+	if (kernel_maps__load_all_kallsyms(v))
+		return -1;
+
+	dso__fixup_sym_end(kernel_map->dso);
+
+	return kernel_maps__split_kallsyms(filter, use_modules);
+}
+
+static size_t kernel_maps__fprintf(FILE *fp, int v)
+{
+	size_t printed = fprintf(stderr, "Kernel maps:\n");
+	struct rb_node *nd;
+
+	for (nd = rb_first(&kernel_maps); 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 (v > 1) {
+			printed += dso__fprintf(pos->dso, fp);
+			printed += fprintf(fp, "--\n");
+		}
+	}
+
+	return printed + fprintf(stderr, "END kernel maps\n");
+}
+
+static int dso__load_perf_map(struct dso *self, struct map *map,
+			      symbol_filter_t filter, int v)
 {
 	char *line = NULL;
 	size_t n;
 	FILE *file;
 	int nr_syms = 0;
 
-	file = fopen(self->name, "r");
+	file = fopen(self->long_name, "r");
 	if (file == NULL)
 		goto out_failure;
 
@@ -279,12 +428,12 @@ static int dso__load_perf_map(struct dso *self, symbol_filter_t filter, int v)
 			continue;
 
 		sym = symbol__new(start, size, line + len,
-				  self->sym_priv_size, start, v);
+				  self->sym_priv_size, v);
 
 		if (sym == NULL)
 			goto out_delete_line;
 
-		if (filter && filter(self, sym))
+		if (filter && filter(map, sym))
 			symbol__delete(sym, self->sym_priv_size);
 		else {
 			dso__insert_symbol(self, sym);
@@ -324,8 +473,7 @@ 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 &&
-	       sym->st_size != 0;
+	       sym->st_shndx != SHN_UNDEF;
 }
 
 static inline int elf_sym__is_label(const GElf_Sym *sym)
@@ -410,7 +558,7 @@ static int dso__synthesize_plt_symbols(struct  dso *self, int v)
 	Elf *elf;
 	int nr = 0, symidx, fd, err = 0;
 
-	fd = open(self->name, O_RDONLY);
+	fd = open(self->long_name, O_RDONLY);
 	if (fd < 0)
 		goto out;
 
@@ -478,7 +626,7 @@ static int dso__synthesize_plt_symbols(struct  dso *self, int v)
 				 "%s@plt", elf_sym__name(&sym, symstrs));
 
 			f = symbol__new(plt_offset, shdr_plt.sh_entsize,
-					sympltname, self->sym_priv_size, 0, v);
+					sympltname, self->sym_priv_size, v);
 			if (!f)
 				goto out_elf_end;
 
@@ -496,7 +644,7 @@ static int dso__synthesize_plt_symbols(struct  dso *self, int v)
 				 "%s@plt", elf_sym__name(&sym, symstrs));
 
 			f = symbol__new(plt_offset, shdr_plt.sh_entsize,
-					sympltname, self->sym_priv_size, 0, v);
+					sympltname, self->sym_priv_size, v);
 			if (!f)
 				goto out_elf_end;
 
@@ -515,13 +663,17 @@ out_close:
 		return nr;
 out:
 	fprintf(stderr, "%s: problems reading %s PLT info.\n",
-		__func__, self->name);
+		__func__, self->long_name);
 	return 0;
 }
 
-static int dso__load_sym(struct dso *self, int fd, const char *name,
-			 symbol_filter_t filter, int v, struct module *mod)
+static int dso__load_sym(struct dso *self, struct map *map, const char *name,
+			 int fd, symbol_filter_t filter, int kernel,
+			 int kmodule, int v)
 {
+	struct map *curr_map = map;
+	struct dso *curr_dso = self;
+	size_t dso_name_len = strlen(self->short_name);
 	Elf_Data *symstrs, *secstrs;
 	uint32_t nr_syms;
 	int err = -1;
@@ -532,7 +684,7 @@ static int dso__load_sym(struct dso *self, int fd, const char *name,
 	GElf_Sym sym;
 	Elf_Scn *sec, *sec_strndx;
 	Elf *elf;
-	int nr = 0, kernel = !strcmp("[kernel]", self->name);
+	int nr = 0;
 
 	elf = elf_begin(fd, ELF_C_READ_MMAP, NULL);
 	if (elf == NULL) {
@@ -588,9 +740,7 @@ static int dso__load_sym(struct dso *self, int fd, const char *name,
 	elf_symtab__for_each_symbol(syms, nr_syms, idx, sym) {
 		struct symbol *f;
 		const char *elf_name;
-		char *demangled;
-		u64 obj_start;
-		struct section *section = NULL;
+		char *demangled = NULL;
 		int is_label = elf_sym__is_label(&sym);
 		const char *section_name;
 
@@ -606,52 +756,85 @@ static int dso__load_sym(struct dso *self, int fd, const char *name,
 		if (is_label && !elf_sec__is_text(&shdr, secstrs))
 			continue;
 
+		elf_name = elf_sym__name(&sym, symstrs);
 		section_name = elf_sec__name(&shdr, secstrs);
-		obj_start = sym.st_value;
 
-		if (self->adjust_symbols) {
-			if (v >= 2)
+		if (kernel || kmodule) {
+			char dso_name[PATH_MAX];
+
+			if (strcmp(section_name,
+				   curr_dso->short_name + dso_name_len) == 0)
+				goto new_symbol;
+
+			if (strcmp(section_name, ".text") == 0) {
+				curr_map = map;
+				curr_dso = self;
+				goto new_symbol;
+			}
+
+			snprintf(dso_name, sizeof(dso_name),
+				 "%s%s", self->short_name, section_name);
+
+			curr_map = kernel_maps__find_by_dso_name(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, self->sym_priv_size);
+				if (curr_dso == NULL)
+					goto out_elf_end;
+				curr_map = map__new2(start, curr_dso);
+				if (curr_map == NULL) {
+					dso__delete(curr_dso);
+					goto out_elf_end;
+				}
+				curr_map->map_ip = vdso__map_ip;
+				curr_dso->origin = DSO__ORIG_KERNEL;
+				kernel_maps__insert(curr_map);
+				dsos__add(curr_dso);
+			} else
+				curr_dso = curr_map->dso;
+
+			goto new_symbol;
+		}
+
+		if (curr_dso->adjust_symbols) {
+			if (v > 2)
 				printf("adjusting symbol: st_value: %Lx sh_addr: %Lx sh_offset: %Lx\n",
 					(u64)sym.st_value, (u64)shdr.sh_addr, (u64)shdr.sh_offset);
 
 			sym.st_value -= shdr.sh_addr - shdr.sh_offset;
 		}
-
-		if (mod) {
-			section = mod->sections->find_section(mod->sections, section_name);
-			if (section)
-				sym.st_value += section->vma;
-			else {
-				fprintf(stderr, "dso__load_sym() module %s lookup of %s failed\n",
-					mod->name, section_name);
-				goto out_elf_end;
-			}
-		}
 		/*
 		 * 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...
 		 */
-		elf_name = elf_sym__name(&sym, symstrs);
 		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, elf_name,
-				self->sym_priv_size, obj_start, v);
+				curr_dso->sym_priv_size, v);
 		free(demangled);
 		if (!f)
 			goto out_elf_end;
 
-		if (filter && filter(self, f))
-			symbol__delete(f, self->sym_priv_size);
+		if (filter && filter(curr_map, f))
+			symbol__delete(f, curr_dso->sym_priv_size);
 		else {
-			f->module = mod;
-			dso__insert_symbol(self, f);
+			dso__insert_symbol(curr_dso, f);
 			nr++;
 		}
 	}
 
+	/*
+	 * For misannotated, zeroed, ASM function sizes.
+	 */
+	if (nr > 0)
+		dso__fixup_sym_end(self);
 	err = nr;
 out_elf_end:
 	elf_end(elf);
@@ -671,7 +854,7 @@ static char *dso__read_build_id(struct dso *self, int v)
 	char *build_id = NULL, *bid;
 	unsigned char *raw;
 	Elf *elf;
-	int fd = open(self->name, O_RDONLY);
+	int fd = open(self->long_name, O_RDONLY);
 
 	if (fd < 0)
 		goto out;
@@ -680,7 +863,7 @@ static char *dso__read_build_id(struct dso *self, int v)
 	if (elf == NULL) {
 		if (v)
 			fprintf(stderr, "%s: cannot read %s ELF file.\n",
-				__func__, self->name);
+				__func__, self->long_name);
 		goto out_close;
 	}
 
@@ -709,7 +892,7 @@ static char *dso__read_build_id(struct dso *self, int v)
 		bid += 2;
 	}
 	if (v >= 2)
-		printf("%s(%s): %s\n", __func__, self->name, build_id);
+		printf("%s(%s): %s\n", __func__, self->long_name, build_id);
 out_elf_end:
 	elf_end(elf);
 out_close:
@@ -727,6 +910,7 @@ char dso__symtab_origin(const struct dso *self)
 		[DSO__ORIG_UBUNTU] =   'u',
 		[DSO__ORIG_BUILDID] =  'b',
 		[DSO__ORIG_DSO] =      'd',
+		[DSO__ORIG_KMODULE] =  'K',
 	};
 
 	if (self == NULL || self->origin == DSO__ORIG_NOT_FOUND)
@@ -734,7 +918,7 @@ char dso__symtab_origin(const struct dso *self)
 	return origin[self->origin];
 }
 
-int dso__load(struct dso *self, symbol_filter_t filter, int v)
+int dso__load(struct dso *self, struct map *map, symbol_filter_t filter, int v)
 {
 	int size = PATH_MAX;
 	char *name = malloc(size), *build_id = NULL;
@@ -747,7 +931,7 @@ int dso__load(struct dso *self, symbol_filter_t filter, int v)
 	self->adjust_symbols = 0;
 
 	if (strncmp(self->name, "/tmp/perf-", 10) == 0) {
-		ret = dso__load_perf_map(self, filter, v);
+		ret = dso__load_perf_map(self, map, filter, v);
 		self->origin = ret > 0 ? DSO__ORIG_JAVA_JIT :
 					 DSO__ORIG_NOT_FOUND;
 		return ret;
@@ -760,10 +944,12 @@ more:
 		self->origin++;
 		switch (self->origin) {
 		case DSO__ORIG_FEDORA:
-			snprintf(name, size, "/usr/lib/debug%s.debug", self->name);
+			snprintf(name, size, "/usr/lib/debug%s.debug",
+				 self->long_name);
 			break;
 		case DSO__ORIG_UBUNTU:
-			snprintf(name, size, "/usr/lib/debug%s", self->name);
+			snprintf(name, size, "/usr/lib/debug%s",
+				 self->long_name);
 			break;
 		case DSO__ORIG_BUILDID:
 			build_id = dso__read_build_id(self, v);
@@ -777,7 +963,7 @@ more:
 			self->origin++;
 			/* Fall thru */
 		case DSO__ORIG_DSO:
-			snprintf(name, size, "%s", self->name);
+			snprintf(name, size, "%s", self->long_name);
 			break;
 
 		default:
@@ -787,7 +973,7 @@ more:
 		fd = open(name, O_RDONLY);
 	} while (fd < 0);
 
-	ret = dso__load_sym(self, fd, name, filter, v, NULL);
+	ret = dso__load_sym(self, map, name, fd, filter, 0, 0, v);
 	close(fd);
 
 	/*
@@ -808,87 +994,243 @@ out:
 	return ret;
 }
 
-static int dso__load_module(struct dso *self, struct mod_dso *mods, const char *name,
-			     symbol_filter_t filter, int v)
+struct map *kernel_map;
+
+static void kernel_maps__insert(struct map *map)
+{
+	maps__insert(&kernel_maps, map);
+}
+
+struct symbol *kernel_maps__find_symbol(u64 ip, struct map **mapp)
 {
-	struct module *mod = mod_dso__find_module(mods, name);
-	int err = 0, fd;
+	struct map *map = maps__find(&kernel_maps, ip);
 
-	if (mod == NULL || !mod->active)
-		return err;
+	if (mapp)
+		*mapp = map;
 
-	fd = open(mod->path, O_RDONLY);
+	if (map) {
+		ip = map->map_ip(map, ip);
+		return map->dso->find_symbol(map->dso, ip);
+	}
 
-	if (fd < 0)
+	return NULL;
+}
+
+struct map *kernel_maps__find_by_dso_name(const char *name)
+{
+	struct rb_node *nd;
+
+	for (nd = rb_first(&kernel_maps); nd; nd = rb_next(nd)) {
+		struct map *map = rb_entry(nd, struct map, rb_node);
+
+		if (map->dso && strcmp(map->dso->name, name) == 0)
+			return map;
+	}
+
+	return NULL;
+}
+
+static int dso__load_module_sym(struct dso *self, struct map *map,
+				symbol_filter_t filter, int v)
+{
+	int err = 0, fd = open(self->long_name, O_RDONLY);
+
+	if (fd < 0) {
+		if (v)
+			fprintf(stderr, "%s: cannot open %s\n",
+				__func__, self->long_name);
 		return err;
+	}
 
-	err = dso__load_sym(self, fd, name, filter, v, mod);
+	err = dso__load_sym(self, map, self->long_name, fd, filter, 0, 1, v);
 	close(fd);
 
 	return err;
 }
 
-int dso__load_modules(struct dso *self, symbol_filter_t filter, int v)
+static int dsos__load_modules_sym_dir(char *dirname,
+				      symbol_filter_t filter, int v)
 {
-	struct mod_dso *mods = mod_dso__new_dso("modules");
-	struct module *pos;
-	struct rb_node *next;
-	int err;
+	struct dirent *dent;
+	int nr_symbols = 0, err;
+	DIR *dir = opendir(dirname);
 
-	err = mod_dso__load_modules(mods);
+	if (!dir) {
+		if (v)
+			fprintf(stderr, "%s: cannot open %s dir\n", __func__,
+				dirname);
+		return -1;
+	}
 
-	if (err <= 0)
-		return err;
+	while ((dent = readdir(dir)) != NULL) {
+		char path[PATH_MAX];
+
+		if (dent->d_type == DT_DIR) {
+			if (!strcmp(dent->d_name, ".") ||
+			    !strcmp(dent->d_name, ".."))
+				continue;
+
+			snprintf(path, sizeof(path), "%s/%s",
+				 dirname, dent->d_name);
+			err = dsos__load_modules_sym_dir(path, filter, v);
+			if (err < 0)
+				goto failure;
+		} else {
+			char *dot = strrchr(dent->d_name, '.'),
+			     dso_name[PATH_MAX];
+			struct map *map;
+			struct rb_node *last;
+
+			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 = kernel_maps__find_by_dso_name(dso_name);
+			if (map == NULL)
+				continue;
+
+			snprintf(path, sizeof(path), "%s/%s",
+				 dirname, dent->d_name);
+
+			map->dso->long_name = strdup(path);
+			if (map->dso->long_name == NULL)
+				goto failure;
+
+			err = dso__load_module_sym(map->dso, map, filter, v);
+			if (err < 0)
+				goto failure;
+			last = rb_last(&map->dso->syms);
+			if (last) {
+				struct symbol *sym;
+				/*
+				 * We do this here as well, even having the
+				 * symbol size found in the symtab because
+				 * misannotated ASM symbols may have the size
+				 * set to zero.
+				 */
+				dso__fixup_sym_end(map->dso);
+
+				sym = rb_entry(last, struct symbol, rb_node);
+				map->end = map->start + sym->end;
+			}
+		}
+		nr_symbols += err;
+	}
 
-	/*
-	 * Iterate over modules, and load active symbols.
-	 */
-	next = rb_first(&mods->mods);
-	while (next) {
-		pos = rb_entry(next, struct module, rb_node);
-		err = dso__load_module(self, mods, pos->name, filter, v);
+	return nr_symbols;
+failure:
+	closedir(dir);
+	return -1;
+}
 
-		if (err < 0)
-			break;
+static int dsos__load_modules_sym(symbol_filter_t filter, int v)
+{
+	struct utsname uts;
+	char modules_path[PATH_MAX];
 
-		next = rb_next(&pos->rb_node);
-	}
+	if (uname(&uts) < 0)
+		return -1;
 
-	if (err < 0) {
-		mod_dso__delete_modules(mods);
-		mod_dso__delete_self(mods);
-	}
+	snprintf(modules_path, sizeof(modules_path), "/lib/modules/%s/kernel",
+		 uts.release);
 
-	return err;
+	return dsos__load_modules_sym_dir(modules_path, filter, v);
 }
 
-static inline void dso__fill_symbol_holes(struct dso *self)
+/*
+ * 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)
 {
-	struct symbol *prev = NULL;
-	struct rb_node *nd;
+	struct map *self = malloc(sizeof(*self));
 
-	for (nd = rb_last(&self->syms); nd; nd = rb_prev(nd)) {
-		struct symbol *pos = rb_entry(nd, struct symbol, rb_node);
+	if (self != NULL) {
+		self->start = start;
+		/*
+		 * Will be filled after we load all the symbols
+		 */
+		self->end = 0;
+
+		self->pgoff = 0;
+		self->dso = dso;
+		self->map_ip = map__map_ip;
+		RB_CLEAR_NODE(&self->rb_node);
+	}
+	return self;
+}
 
-		if (prev) {
-			u64 hole = 0;
-			int alias = pos->start == prev->start;
+static int dsos__load_modules(unsigned int sym_priv_size)
+{
+	char *line = NULL;
+	size_t n;
+	FILE *file = fopen("/proc/modules", "r");
+	struct map *map;
+
+	if (file == NULL)
+		return -1;
 
-			if (!alias)
-				hole = prev->start - pos->end - 1;
+	while (!feof(file)) {
+		char name[PATH_MAX];
+		u64 start;
+		struct dso *dso;
+		char *sep;
+		int line_len;
 
-			if (hole || alias) {
-				if (alias)
-					pos->end = prev->end;
-				else if (hole)
-					pos->end = prev->start - 1;
-			}
+		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);
+		dso = dso__new(name, sym_priv_size);
+
+		if (dso == NULL)
+			goto out_delete_line;
+
+		map = map__new2(start, dso);
+		if (map == NULL) {
+			dso__delete(dso);
+			goto out_delete_line;
 		}
-		prev = pos;
+
+		dso->origin = DSO__ORIG_KMODULE;
+		kernel_maps__insert(map);
+		dsos__add(dso);
 	}
+
+	free(line);
+	fclose(file);
+
+	return 0;
+
+out_delete_line:
+	free(line);
+out_failure:
+	return -1;
 }
 
-static int dso__load_vmlinux(struct dso *self, const char *vmlinux,
+static int dso__load_vmlinux(struct dso *self, struct map *map,
+			     const char *vmlinux,
 			     symbol_filter_t filter, int v)
 {
 	int err, fd = open(vmlinux, O_RDONLY);
@@ -896,40 +1238,82 @@ static int dso__load_vmlinux(struct dso *self, const char *vmlinux,
 	if (fd < 0)
 		return -1;
 
-	err = dso__load_sym(self, fd, vmlinux, filter, v, NULL);
-
-	if (err > 0)
-		dso__fill_symbol_holes(self);
+	err = dso__load_sym(self, map, self->long_name, fd, filter, 1, 0, v);
 
 	close(fd);
 
 	return err;
 }
 
-int dso__load_kernel(struct dso *self, const char *vmlinux,
-		     symbol_filter_t filter, int v, int use_modules)
+int dsos__load_kernel(const char *vmlinux, unsigned int sym_priv_size,
+		      symbol_filter_t filter, int v, int use_modules)
 {
 	int err = -1;
+	struct dso *dso = dso__new(vmlinux, sym_priv_size);
+
+	if (dso == NULL)
+		return -1;
+
+	dso->short_name = "[kernel]";
+	kernel_map = map__new2(0, dso);
+	if (kernel_map == NULL)
+		goto out_delete_dso;
+
+	kernel_map->map_ip = vdso__map_ip;
+
+	if (use_modules && dsos__load_modules(sym_priv_size) < 0) {
+		fprintf(stderr, "Failed to load list of modules in use! "
+				"Continuing...\n");
+		use_modules = 0;
+	}
 
 	if (vmlinux) {
-		err = dso__load_vmlinux(self, vmlinux, filter, v);
-		if (err > 0 && use_modules)
-			err = dso__load_modules(self, filter, v);
+		err = dso__load_vmlinux(dso, kernel_map, vmlinux, filter, v);
+		if (err > 0 && use_modules) {
+			int syms = dsos__load_modules_sym(filter, v);
+
+			if (syms < 0)
+				fprintf(stderr, "Failed to read module symbols!"
+					" Continuing...\n");
+			else
+				err += syms;
+		}
 	}
 
 	if (err <= 0)
-		err = dso__load_kallsyms(self, filter, v);
+		err = kernel_maps__load_kallsyms(filter, use_modules, v);
 
-	if (err > 0)
-		self->origin = DSO__ORIG_KERNEL;
+	if (err > 0) {
+		struct rb_node *node = rb_first(&dso->syms);
+		struct symbol *sym = rb_entry(node, struct symbol, rb_node);
+
+		kernel_map->start = sym->start;
+		node = rb_last(&dso->syms);
+		sym = rb_entry(node, struct symbol, rb_node);
+		kernel_map->end = sym->end;
+
+		dso->origin = DSO__ORIG_KERNEL;
+		kernel_maps__insert(kernel_map);
+		/*
+		 * Now that we have all sorted out, just set the ->end of all
+		 * maps:
+		 */
+		kernel_maps__fixup_end();
+		dsos__add(dso);
+
+		if (v > 0)
+			kernel_maps__fprintf(stderr, v);
+	}
 
 	return err;
+
+out_delete_dso:
+	dso__delete(dso);
+	return -1;
 }
 
 LIST_HEAD(dsos);
-struct dso	*kernel_dso;
 struct dso	*vdso;
-struct dso	*hypervisor_dso;
 
 const char	*vmlinux_name = "vmlinux";
 int		modules;
@@ -961,7 +1345,7 @@ struct dso *dsos__findnew(const char *name)
 	if (!dso)
 		goto out_delete_dso;
 
-	nr = dso__load(dso, NULL, verbose);
+	nr = dso__load(dso, NULL, NULL, verbose);
 	if (nr < 0) {
 		eprintf("Failed to open: %s\n", name);
 		goto out_delete_dso;
@@ -986,43 +1370,20 @@ void dsos__fprintf(FILE *fp)
 		dso__fprintf(pos, fp);
 }
 
-static struct symbol *vdso__find_symbol(struct dso *dso, u64 ip)
-{
-	return dso__find_symbol(dso, ip);
-}
-
 int load_kernel(void)
 {
-	int err;
-
-	kernel_dso = dso__new("[kernel]", 0);
-	if (!kernel_dso)
+	if (dsos__load_kernel(vmlinux_name, 0, NULL, verbose, modules) <= 0)
 		return -1;
 
-	err = dso__load_kernel(kernel_dso, vmlinux_name, NULL, verbose, modules);
-	if (err <= 0) {
-		dso__delete(kernel_dso);
-		kernel_dso = NULL;
-	} else
-		dsos__add(kernel_dso);
-
 	vdso = dso__new("[vdso]", 0);
 	if (!vdso)
 		return -1;
 
-	vdso->find_symbol = vdso__find_symbol;
-
 	dsos__add(vdso);
 
-	hypervisor_dso = dso__new("[hypervisor]", 0);
-	if (!hypervisor_dso)
-		return -1;
-	dsos__add(hypervisor_dso);
-
-	return err;
+	return 0;
 }
 
-
 void symbol__init(void)
 {
 	elf_version(EV_CURRENT);
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index 6e8490716408..2e4522edeb07 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -1,11 +1,10 @@
-#ifndef _PERF_SYMBOL_
-#define _PERF_SYMBOL_ 1
+#ifndef __PERF_SYMBOL
+#define __PERF_SYMBOL 1
 
 #include <linux/types.h>
 #include "types.h"
 #include <linux/list.h>
 #include <linux/rbtree.h>
-#include "module.h"
 #include "event.h"
 
 #ifdef HAVE_CPLUS_DEMANGLE
@@ -36,10 +35,8 @@ struct symbol {
 	struct rb_node	rb_node;
 	u64		start;
 	u64		end;
-	u64		obj_start;
 	u64		hist_sum;
 	u64		*hist;
-	struct module	*module;
 	void		*priv;
 	char		name[0];
 };
@@ -52,12 +49,14 @@ struct dso {
 	unsigned char	 adjust_symbols;
 	unsigned char	 slen_calculated;
 	unsigned char	 origin;
+	const char	 *short_name;
+	char	 	 *long_name;
 	char		 name[0];
 };
 
 extern const char *sym_hist_filter;
 
-typedef int (*symbol_filter_t)(struct dso *self, struct symbol *sym);
+typedef int (*symbol_filter_t)(struct map *map, struct symbol *sym);
 
 struct dso *dso__new(const char *name, unsigned int sym_priv_size);
 void dso__delete(struct dso *self);
@@ -69,10 +68,10 @@ static inline void *dso__sym_priv(struct dso *self, struct symbol *sym)
 
 struct symbol *dso__find_symbol(struct dso *self, u64 ip);
 
-int dso__load_kernel(struct dso *self, const char *vmlinux,
-		     symbol_filter_t filter, int verbose, int modules);
-int dso__load_modules(struct dso *self, symbol_filter_t filter, int verbose);
-int dso__load(struct dso *self, symbol_filter_t filter, int verbose);
+int dsos__load_kernel(const char *vmlinux, unsigned int sym_priv_size,
+		      symbol_filter_t filter, int verbose, int modules);
+int dso__load(struct dso *self, struct map *map, symbol_filter_t filter,
+	      int verbose);
 struct dso *dsos__findnew(const char *name);
 void dsos__fprintf(FILE *fp);
 
@@ -84,9 +83,8 @@ int load_kernel(void);
 void symbol__init(void);
 
 extern struct list_head dsos;
-extern struct dso *kernel_dso;
+extern struct map *kernel_map;
 extern struct dso *vdso;
-extern struct dso *hypervisor_dso;
 extern const char *vmlinux_name;
 extern int   modules;
-#endif /* _PERF_SYMBOL_ */
+#endif /* __PERF_SYMBOL */
diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c
index 7635928ca278..f53fad7c0a8d 100644
--- a/tools/perf/util/thread.c
+++ b/tools/perf/util/thread.c
@@ -6,16 +6,20 @@
 #include "util.h"
 #include "debug.h"
 
+static struct rb_root threads;
+static struct thread *last_match;
+
 static struct thread *thread__new(pid_t pid)
 {
-	struct thread *self = malloc(sizeof(*self));
+	struct thread *self = calloc(1, sizeof(*self));
 
 	if (self != NULL) {
 		self->pid = pid;
 		self->comm = malloc(32);
 		if (self->comm)
 			snprintf(self->comm, 32, ":%d", self->pid);
-		INIT_LIST_HEAD(&self->maps);
+		self->maps = RB_ROOT;
+		INIT_LIST_HEAD(&self->removed_maps);
 	}
 
 	return self;
@@ -31,19 +35,27 @@ int thread__set_comm(struct thread *self, const char *comm)
 
 static size_t thread__fprintf(struct thread *self, FILE *fp)
 {
+	struct rb_node *nd;
 	struct map *pos;
-	size_t ret = fprintf(fp, "Thread %d %s\n", self->pid, self->comm);
+	size_t ret = fprintf(fp, "Thread %d %s\nCurrent maps:\n",
+			     self->pid, self->comm);
+
+	for (nd = rb_first(&self->maps); nd; nd = rb_next(nd)) {
+		pos = rb_entry(nd, struct map, rb_node);
+		ret += map__fprintf(pos, fp);
+	}
+
+	ret = fprintf(fp, "Removed maps:\n");
 
-	list_for_each_entry(pos, &self->maps, node)
+	list_for_each_entry(pos, &self->removed_maps, node)
 		ret += map__fprintf(pos, fp);
 
 	return ret;
 }
 
-struct thread *
-threads__findnew(pid_t pid, struct rb_root *threads, struct thread **last_match)
+struct thread *threads__findnew(pid_t pid)
 {
-	struct rb_node **p = &threads->rb_node;
+	struct rb_node **p = &threads.rb_node;
 	struct rb_node *parent = NULL;
 	struct thread *th;
 
@@ -52,15 +64,15 @@ threads__findnew(pid_t pid, struct rb_root *threads, struct thread **last_match)
 	 * so most of the time we dont have to look up
 	 * the full rbtree:
 	 */
-	if (*last_match && (*last_match)->pid == pid)
-		return *last_match;
+	if (last_match && last_match->pid == pid)
+		return last_match;
 
 	while (*p != NULL) {
 		parent = *p;
 		th = rb_entry(parent, struct thread, rb_node);
 
 		if (th->pid == pid) {
-			*last_match = th;
+			last_match = th;
 			return th;
 		}
 
@@ -73,19 +85,18 @@ threads__findnew(pid_t pid, struct rb_root *threads, struct thread **last_match)
 	th = thread__new(pid);
 	if (th != NULL) {
 		rb_link_node(&th->rb_node, parent, p);
-		rb_insert_color(&th->rb_node, threads);
-		*last_match = th;
+		rb_insert_color(&th->rb_node, &threads);
+		last_match = th;
 	}
 
 	return th;
 }
 
-struct thread *
-register_idle_thread(struct rb_root *threads, struct thread **last_match)
+struct thread *register_idle_thread(void)
 {
-	struct thread *thread = threads__findnew(0, threads, last_match);
+	struct thread *thread = threads__findnew(0);
 
-	if (!thread || thread__set_comm(thread, "[init]")) {
+	if (!thread || thread__set_comm(thread, "swapper")) {
 		fprintf(stderr, "problem inserting idle task.\n");
 		exit(-1);
 	}
@@ -93,42 +104,82 @@ register_idle_thread(struct rb_root *threads, struct thread **last_match)
 	return thread;
 }
 
-void thread__insert_map(struct thread *self, struct map *map)
+static void thread__remove_overlappings(struct thread *self, struct map *map)
 {
-	struct map *pos, *tmp;
-
-	list_for_each_entry_safe(pos, tmp, &self->maps, node) {
-		if (map__overlap(pos, map)) {
-			if (verbose >= 2) {
-				printf("overlapping maps:\n");
-				map__fprintf(map, stdout);
-				map__fprintf(pos, stdout);
-			}
-
-			if (map->start <= pos->start && map->end > pos->start)
-				pos->start = map->end;
-
-			if (map->end >= pos->end && map->start < pos->end)
-				pos->end = map->start;
-
-			if (verbose >= 2) {
-				printf("after collision:\n");
-				map__fprintf(pos, stdout);
-			}
-
-			if (pos->start >= pos->end) {
-				list_del_init(&pos->node);
-				free(pos);
-			}
+	struct rb_node *next = rb_first(&self->maps);
+
+	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) {
+			printf("overlapping maps:\n");
+			map__fprintf(map, stdout);
+			map__fprintf(pos, stdout);
 		}
+
+		rb_erase(&pos->rb_node, &self->maps);
+		/*
+		 * 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);
+	}
+}
+
+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);
+}
+
+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;
 	}
 
-	list_add_tail(&map->node, &self->maps);
+	return NULL;
+}
+
+void thread__insert_map(struct thread *self, struct map *map)
+{
+	thread__remove_overlappings(self, map);
+	maps__insert(&self->maps, map);
 }
 
 int thread__fork(struct thread *self, struct thread *parent)
 {
-	struct map *map;
+	struct rb_node *nd;
 
 	if (self->comm)
 		free(self->comm);
@@ -136,7 +187,8 @@ int thread__fork(struct thread *self, struct thread *parent)
 	if (!self->comm)
 		return -ENOMEM;
 
-	list_for_each_entry(map, &parent->maps, node) {
+	for (nd = rb_first(&parent->maps); nd; nd = rb_next(nd)) {
+		struct map *map = rb_entry(nd, struct map, rb_node);
 		struct map *new = map__clone(map);
 		if (!new)
 			return -ENOMEM;
@@ -146,26 +198,12 @@ int thread__fork(struct thread *self, struct thread *parent)
 	return 0;
 }
 
-struct map *thread__find_map(struct thread *self, u64 ip)
-{
-	struct map *pos;
-
-	if (self == NULL)
-		return NULL;
-
-	list_for_each_entry(pos, &self->maps, node)
-		if (ip >= pos->start && ip <= pos->end)
-			return pos;
-
-	return NULL;
-}
-
-size_t threads__fprintf(FILE *fp, struct rb_root *threads)
+size_t threads__fprintf(FILE *fp)
 {
 	size_t ret = 0;
 	struct rb_node *nd;
 
-	for (nd = rb_first(threads); nd; nd = rb_next(nd)) {
+	for (nd = rb_first(&threads); nd; nd = rb_next(nd)) {
 		struct thread *pos = rb_entry(nd, struct thread, rb_node);
 
 		ret += thread__fprintf(pos, fp);
diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h
index 634f2809a342..1abef3b7455d 100644
--- a/tools/perf/util/thread.h
+++ b/tools/perf/util/thread.h
@@ -1,21 +1,35 @@
+#ifndef __PERF_THREAD_H
+#define __PERF_THREAD_H
+
 #include <linux/rbtree.h>
-#include <linux/list.h>
 #include <unistd.h>
 #include "symbol.h"
 
 struct thread {
-	struct rb_node	 rb_node;
-	struct list_head maps;
-	pid_t		 pid;
-	char		 *comm;
+	struct rb_node		rb_node;
+	struct rb_root		maps;
+	struct list_head	removed_maps;
+	pid_t			pid;
+	char			shortname[3];
+	char			*comm;
 };
 
 int thread__set_comm(struct thread *self, const char *comm);
-struct thread *
-threads__findnew(pid_t pid, struct rb_root *threads, struct thread **last_match);
-struct thread *
-register_idle_thread(struct rb_root *threads, struct thread **last_match);
+struct thread *threads__findnew(pid_t pid);
+struct thread *register_idle_thread(void);
 void thread__insert_map(struct thread *self, struct map *map);
 int thread__fork(struct thread *self, struct thread *parent);
-struct map *thread__find_map(struct thread *self, u64 ip);
-size_t threads__fprintf(FILE *fp, struct rb_root *threads);
+size_t threads__fprintf(FILE *fp);
+
+void maps__insert(struct rb_root *maps, struct map *map);
+struct map *maps__find(struct rb_root *maps, u64 ip);
+
+struct symbol *kernel_maps__find_symbol(const u64 ip, struct map **mapp);
+struct map *kernel_maps__find_by_dso_name(const char *name);
+
+static inline struct map *thread__find_map(struct thread *self, u64 ip)
+{
+	return self ? maps__find(&self->maps, ip) : NULL;
+}
+
+#endif	/* __PERF_THREAD_H */
diff --git a/tools/perf/util/trace-event-info.c b/tools/perf/util/trace-event-info.c
index 6c9302a7274c..831052d4b4fb 100644
--- a/tools/perf/util/trace-event-info.c
+++ b/tools/perf/util/trace-event-info.c
@@ -458,7 +458,7 @@ static void read_proc_kallsyms(void)
 static void read_ftrace_printk(void)
 {
 	unsigned int size, check_size;
-	const char *path;
+	char *path;
 	struct stat st;
 	int ret;
 
@@ -468,23 +468,24 @@ static void read_ftrace_printk(void)
 		/* not found */
 		size = 0;
 		write_or_die(&size, 4);
-		return;
+		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 perf_counter_attr *pattrs, int nb_counters)
+get_tracepoints_path(struct perf_event_attr *pattrs, int nb_events)
 {
 	struct tracepoint_path path, *ppath = &path;
 	int i;
 
-	for (i = 0; i < nb_counters; i++) {
+	for (i = 0; i < nb_events; i++) {
 		if (pattrs[i].type != PERF_TYPE_TRACEPOINT)
 			continue;
 		ppath->next = tracepoint_id_to_path(pattrs[i].config);
@@ -495,14 +496,12 @@ get_tracepoints_path(struct perf_counter_attr *pattrs, int nb_counters)
 
 	return path.next;
 }
-void read_tracing_data(struct perf_counter_attr *pattrs, int nb_counters)
+void read_tracing_data(int fd, struct perf_event_attr *pattrs, int nb_events)
 {
 	char buf[BUFSIZ];
 	struct tracepoint_path *tps;
 
-	output_fd = open(output_file, O_WRONLY | O_CREAT | O_TRUNC | O_LARGEFILE, 0644);
-	if (output_fd < 0)
-		die("creating file '%s'", output_file);
+	output_fd = fd;
 
 	buf[0] = 23;
 	buf[1] = 8;
@@ -529,7 +528,7 @@ void read_tracing_data(struct perf_counter_attr *pattrs, int nb_counters)
 	page_size = getpagesize();
 	write_or_die(&page_size, 4);
 
-	tps = get_tracepoints_path(pattrs, nb_counters);
+	tps = get_tracepoints_path(pattrs, nb_events);
 
 	read_header_files();
 	read_ftrace_files(tps);
diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c
index 629e602d9405..4b61b497040e 100644
--- a/tools/perf/util/trace-event-parse.c
+++ b/tools/perf/util/trace-event-parse.c
@@ -40,6 +40,8 @@ int header_page_size_size;
 int header_page_data_offset;
 int header_page_data_size;
 
+int latency_format;
+
 static char *input_buf;
 static unsigned long long input_buf_ptr;
 static unsigned long long input_buf_siz;
@@ -284,18 +286,16 @@ void parse_ftrace_printk(char *file, unsigned int size __unused)
 	char *line;
 	char *next = NULL;
 	char *addr_str;
-	int ret;
+	char *fmt;
 	int i;
 
 	line = strtok_r(file, "\n", &next);
 	while (line) {
 		item = malloc_or_die(sizeof(*item));
-		ret = sscanf(line, "%as : %as",
-			     (float *)(void *)&addr_str, /* workaround gcc warning */
-			     (float *)(void *)&item->printk);
+		addr_str = strtok_r(line, ":", &fmt);
 		item->addr = strtoull(addr_str, NULL, 16);
-		free(addr_str);
-
+		/* fmt still has a space, skip it */
+		item->printk = strdup(fmt+1);
 		item->next = list;
 		list = item;
 		line = strtok_r(NULL, "\n", &next);
@@ -522,7 +522,10 @@ static enum event_type __read_token(char **tok)
 			last_ch = ch;
 			ch = __read_char();
 			buf[i++] = ch;
-		} while (ch != quote_ch && last_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;
@@ -610,7 +613,7 @@ static enum event_type read_token_item(char **tok)
 static int test_type(enum event_type type, enum event_type expect)
 {
 	if (type != expect) {
-		die("Error: expected type %d but read %d",
+		warning("Error: expected type %d but read %d",
 		    expect, type);
 		return -1;
 	}
@@ -618,16 +621,16 @@ static int test_type(enum event_type type, enum event_type expect)
 }
 
 static int test_type_token(enum event_type type, char *token,
-		    enum event_type expect, char *expect_tok)
+		    enum event_type expect, const char *expect_tok)
 {
 	if (type != expect) {
-		die("Error: expected type %d but read %d",
+		warning("Error: expected type %d but read %d",
 		    expect, type);
 		return -1;
 	}
 
 	if (strcmp(token, expect_tok) != 0) {
-		die("Error: expected '%s' but read '%s'",
+		warning("Error: expected '%s' but read '%s'",
 		    expect_tok, token);
 		return -1;
 	}
@@ -650,7 +653,7 @@ 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, char *str, int newline_ok)
+static int __read_expected(enum event_type expect, const char *str, int newline_ok)
 {
 	enum event_type type;
 	char *token;
@@ -665,15 +668,15 @@ static int __read_expected(enum event_type expect, char *str, int newline_ok)
 
 	free_token(token);
 
-	return 0;
+	return ret;
 }
 
-static int read_expected(enum event_type expect, char *str)
+static int read_expected(enum event_type expect, const char *str)
 {
 	return __read_expected(expect, str, 1);
 }
 
-static int read_expected_item(enum event_type expect, char *str)
+static int read_expected_item(enum event_type expect, const char *str)
 {
 	return __read_expected(expect, str, 0);
 }
@@ -682,10 +685,10 @@ static char *event_read_name(void)
 {
 	char *token;
 
-	if (read_expected(EVENT_ITEM, (char *)"name") < 0)
+	if (read_expected(EVENT_ITEM, "name") < 0)
 		return NULL;
 
-	if (read_expected(EVENT_OP, (char *)":") < 0)
+	if (read_expected(EVENT_OP, ":") < 0)
 		return NULL;
 
 	if (read_expect_type(EVENT_ITEM, &token) < 0)
@@ -703,10 +706,10 @@ static int event_read_id(void)
 	char *token;
 	int id;
 
-	if (read_expected_item(EVENT_ITEM, (char *)"ID") < 0)
+	if (read_expected_item(EVENT_ITEM, "ID") < 0)
 		return -1;
 
-	if (read_expected(EVENT_OP, (char *)":") < 0)
+	if (read_expected(EVENT_OP, ":") < 0)
 		return -1;
 
 	if (read_expect_type(EVENT_ITEM, &token) < 0)
@@ -721,6 +724,24 @@ static int event_read_id(void)
 	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 (!strcmp(field->type, "__data_loc"))
+		return 1;
+
+	return 0;
+}
+
 static int event_read_fields(struct event *event, struct format_field **fields)
 {
 	struct format_field *field = NULL;
@@ -738,7 +759,7 @@ static int event_read_fields(struct event *event, struct format_field **fields)
 
 		count++;
 
-		if (test_type_token(type, token, EVENT_ITEM, (char *)"field"))
+		if (test_type_token(type, token, EVENT_ITEM, "field"))
 			goto fail;
 		free_token(token);
 
@@ -753,7 +774,7 @@ static int event_read_fields(struct event *event, struct format_field **fields)
 			type = read_token(&token);
 		}
 
-		if (test_type_token(type, token, EVENT_OP, (char *)":") < 0)
+		if (test_type_token(type, token, EVENT_OP, ":") < 0)
 			return -1;
 
 		if (read_expect_type(EVENT_ITEM, &token) < 0)
@@ -865,14 +886,20 @@ static int event_read_fields(struct event *event, struct format_field **fields)
 			free(brackets);
 		}
 
-		if (test_type_token(type, token,  EVENT_OP, (char *)";"))
+		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, (char *)"offset") < 0)
+		if (read_expected(EVENT_ITEM, "offset") < 0)
 			goto fail_expect;
 
-		if (read_expected(EVENT_OP, (char *)":") < 0)
+		if (read_expected(EVENT_OP, ":") < 0)
 			goto fail_expect;
 
 		if (read_expect_type(EVENT_ITEM, &token))
@@ -880,13 +907,13 @@ static int event_read_fields(struct event *event, struct format_field **fields)
 		field->offset = strtoul(token, NULL, 0);
 		free_token(token);
 
-		if (read_expected(EVENT_OP, (char *)";") < 0)
+		if (read_expected(EVENT_OP, ";") < 0)
 			goto fail_expect;
 
-		if (read_expected(EVENT_ITEM, (char *)"size") < 0)
+		if (read_expected(EVENT_ITEM, "size") < 0)
 			goto fail_expect;
 
-		if (read_expected(EVENT_OP, (char *)":") < 0)
+		if (read_expected(EVENT_OP, ":") < 0)
 			goto fail_expect;
 
 		if (read_expect_type(EVENT_ITEM, &token))
@@ -894,11 +921,33 @@ static int event_read_fields(struct event *event, struct format_field **fields)
 		field->size = strtoul(token, NULL, 0);
 		free_token(token);
 
-		if (read_expected(EVENT_OP, (char *)";") < 0)
+		if (read_expected(EVENT_OP, ";") < 0)
 			goto fail_expect;
 
-		if (read_expect_type(EVENT_NEWLINE, &token) < 0)
-			goto fail;
+		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;
+
+			/* add signed type */
+
+			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;
@@ -921,10 +970,10 @@ static int event_read_format(struct event *event)
 	char *token;
 	int ret;
 
-	if (read_expected_item(EVENT_ITEM, (char *)"format") < 0)
+	if (read_expected_item(EVENT_ITEM, "format") < 0)
 		return -1;
 
-	if (read_expected(EVENT_OP, (char *)":") < 0)
+	if (read_expected(EVENT_OP, ":") < 0)
 		return -1;
 
 	if (read_expect_type(EVENT_NEWLINE, &token))
@@ -984,7 +1033,7 @@ process_cond(struct event *event, struct print_arg *top, char **tok)
 
 	*tok = NULL;
 	type = process_arg(event, left, &token);
-	if (test_type_token(type, token, EVENT_OP, (char *)":"))
+	if (test_type_token(type, token, EVENT_OP, ":"))
 		goto out_free;
 
 	arg->op.op = token;
@@ -1004,6 +1053,35 @@ out_free:
 	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]) {
@@ -1128,6 +1206,8 @@ process_op(struct event *event, struct print_arg *arg, char **tok)
 		   strcmp(token, "*") == 0 ||
 		   strcmp(token, "^") == 0 ||
 		   strcmp(token, "/") == 0 ||
+		   strcmp(token, "<") == 0 ||
+		   strcmp(token, ">") == 0 ||
 		   strcmp(token, "==") == 0 ||
 		   strcmp(token, "!=") == 0) {
 
@@ -1144,17 +1224,46 @@ process_op(struct event *event, struct print_arg *arg, char **tok)
 
 		right = malloc_or_die(sizeof(*right));
 
-		type = process_arg(event, right, tok);
+		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 {
-		die("unknown op '%s'", token);
+		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;
 
@@ -1178,7 +1287,7 @@ process_entry(struct event *event __unused, struct print_arg *arg,
 	char *field;
 	char *token;
 
-	if (read_expected(EVENT_OP, (char *)"->") < 0)
+	if (read_expected(EVENT_OP, "->") < 0)
 		return EVENT_ERROR;
 
 	if (read_expect_type(EVENT_ITEM, &token) < 0)
@@ -1338,14 +1447,14 @@ process_fields(struct event *event, struct print_flag_sym **list, char **tok)
 	do {
 		free_token(token);
 		type = read_token_item(&token);
-		if (test_type_token(type, token, EVENT_OP, (char *)"{"))
+		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, (char *)","))
+		if (test_type_token(type, token, EVENT_DELIM, ","))
 			goto out_free;
 
 		field = malloc_or_die(sizeof(*field));
@@ -1356,7 +1465,7 @@ process_fields(struct event *event, struct print_flag_sym **list, char **tok)
 
 		free_token(token);
 		type = process_arg(event, arg, &token);
-		if (test_type_token(type, token, EVENT_OP, (char *)"}"))
+		if (test_type_token(type, token, EVENT_OP, "}"))
 			goto out_free;
 
 		value = arg_eval(arg);
@@ -1391,13 +1500,13 @@ process_flags(struct event *event, struct print_arg *arg, char **tok)
 	memset(arg, 0, sizeof(*arg));
 	arg->type = PRINT_FLAGS;
 
-	if (read_expected_item(EVENT_DELIM, (char *)"(") < 0)
+	if (read_expected_item(EVENT_DELIM, "(") < 0)
 		return EVENT_ERROR;
 
 	field = malloc_or_die(sizeof(*field));
 
 	type = process_arg(event, field, &token);
-	if (test_type_token(type, token, EVENT_DELIM, (char *)","))
+	if (test_type_token(type, token, EVENT_DELIM, ","))
 		goto out_free;
 
 	arg->flags.field = field;
@@ -1408,11 +1517,11 @@ process_flags(struct event *event, struct print_arg *arg, char **tok)
 		type = read_token_item(&token);
 	}
 
-	if (test_type_token(type, token, EVENT_DELIM, (char *)","))
+	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, (char *)")"))
+	if (test_type_token(type, token, EVENT_DELIM, ")"))
 		goto out_free;
 
 	free_token(token);
@@ -1434,19 +1543,19 @@ process_symbols(struct event *event, struct print_arg *arg, char **tok)
 	memset(arg, 0, sizeof(*arg));
 	arg->type = PRINT_SYMBOL;
 
-	if (read_expected_item(EVENT_DELIM, (char *)"(") < 0)
+	if (read_expected_item(EVENT_DELIM, "(") < 0)
 		return EVENT_ERROR;
 
 	field = malloc_or_die(sizeof(*field));
 
 	type = process_arg(event, field, &token);
-	if (test_type_token(type, token, EVENT_DELIM, (char *)","))
+	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, (char *)")"))
+	if (test_type_token(type, token, EVENT_DELIM, ")"))
 		goto out_free;
 
 	free_token(token);
@@ -1463,7 +1572,6 @@ process_paren(struct event *event, struct print_arg *arg, char **tok)
 {
 	struct print_arg *item_arg;
 	enum event_type type;
-	int ptr_cast = 0;
 	char *token;
 
 	type = process_arg(event, arg, &token);
@@ -1471,28 +1579,13 @@ process_paren(struct event *event, struct print_arg *arg, char **tok)
 	if (type == EVENT_ERROR)
 		return EVENT_ERROR;
 
-	if (type == EVENT_OP) {
-		/* handle the ptr casts */
-		if (!strcmp(token, "*")) {
-			/*
-			 * FIXME: should we zapp whitespaces before ')' ?
-			 * (may require a peek_token_item())
-			 */
-			if (__peek_char() == ')') {
-				ptr_cast = 1;
-				free_token(token);
-				type = read_token_item(&token);
-			}
-		}
-		if (!ptr_cast) {
-			type = process_op(event, arg, &token);
+	if (type == EVENT_OP)
+		type = process_op(event, arg, &token);
 
-			if (type == EVENT_ERROR)
-				return EVENT_ERROR;
-		}
-	}
+	if (type == EVENT_ERROR)
+		return EVENT_ERROR;
 
-	if (test_type_token(type, token, EVENT_DELIM, (char *)")")) {
+	if (test_type_token(type, token, EVENT_DELIM, ")")) {
 		free_token(token);
 		return EVENT_ERROR;
 	}
@@ -1516,13 +1609,6 @@ process_paren(struct event *event, struct print_arg *arg, char **tok)
 		item_arg = malloc_or_die(sizeof(*item_arg));
 
 		arg->type = PRINT_TYPE;
-		if (ptr_cast) {
-			char *old = arg->atom.atom;
-
-			arg->atom.atom = malloc_or_die(strlen(old + 3));
-			sprintf(arg->atom.atom, "%s *", old);
-			free(old);
-		}
 		arg->typecast.type = arg->atom.atom;
 		arg->typecast.item = item_arg;
 		type = process_arg_token(event, item_arg, &token, type);
@@ -1540,7 +1626,7 @@ process_str(struct event *event __unused, struct print_arg *arg, char **tok)
 	enum event_type type;
 	char *token;
 
-	if (read_expected(EVENT_DELIM, (char *)"(") < 0)
+	if (read_expected(EVENT_DELIM, "(") < 0)
 		return EVENT_ERROR;
 
 	if (read_expect_type(EVENT_ITEM, &token) < 0)
@@ -1550,7 +1636,7 @@ process_str(struct event *event __unused, struct print_arg *arg, char **tok)
 	arg->string.string = token;
 	arg->string.offset = -1;
 
-	if (read_expected(EVENT_DELIM, (char *)")") < 0)
+	if (read_expected(EVENT_DELIM, ")") < 0)
 		return EVENT_ERROR;
 
 	type = read_token(&token);
@@ -1637,12 +1723,18 @@ process_arg_token(struct event *event, struct print_arg *arg,
 
 static int event_read_print_args(struct event *event, struct print_arg **list)
 {
-	enum event_type type;
+	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));
 
@@ -1683,18 +1775,19 @@ static int event_read_print(struct event *event)
 	char *token;
 	int ret;
 
-	if (read_expected_item(EVENT_ITEM, (char *)"print") < 0)
+	if (read_expected_item(EVENT_ITEM, "print") < 0)
 		return -1;
 
-	if (read_expected(EVENT_ITEM, (char *)"fmt") < 0)
+	if (read_expected(EVENT_ITEM, "fmt") < 0)
 		return -1;
 
-	if (read_expected(EVENT_OP, (char *)":") < 0)
+	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;
 
@@ -1704,7 +1797,22 @@ static int event_read_print(struct event *event)
 	if (type == EVENT_NONE)
 		return 0;
 
-	if (test_type_token(type, token, EVENT_DELIM, (char *)","))
+	/* 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);
@@ -1713,7 +1821,7 @@ static int event_read_print(struct event *event)
 	if (ret < 0)
 		return -1;
 
-	return 0;
+	return ret;
 
  fail:
 	free_token(token);
@@ -1776,6 +1884,29 @@ static unsigned long long read_size(void *ptr, int size)
 	}
 }
 
+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;
+
+	return data + field->offset;
+}
+
 static int get_common_info(const char *type, int *offset, int *size)
 {
 	struct event *event;
@@ -1799,40 +1930,70 @@ static int get_common_info(const char *type, int *offset, int *size)
 	return 0;
 }
 
-static int parse_common_type(void *data)
+static int __parse_common(void *data, int *size, int *offset,
+			  const char *name)
 {
-	static int type_offset;
-	static int type_size;
 	int ret;
 
-	if (!type_size) {
-		ret = get_common_info("common_type",
-				      &type_offset,
-				      &type_size);
+	if (!*size) {
+		ret = get_common_info(name, offset, size);
 		if (ret < 0)
 			return ret;
 	}
-	return read_size(data + type_offset, type_size);
+	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");
 }
 
 static int parse_common_pid(void *data)
 {
 	static int pid_offset;
 	static int pid_size;
+
+	return __parse_common(data, &pid_size, &pid_offset,
+			      "common_pid");
+}
+
+static 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");
+}
+
+static int parse_common_flags(void *data)
+{
+	static int flags_offset;
+	static int flags_size;
+
+	return __parse_common(data, &flags_size, &flags_offset,
+			      "common_flags");
+}
+
+static int parse_common_lock_depth(void *data)
+{
+	static int ld_offset;
+	static int ld_size;
 	int ret;
 
-	if (!pid_size) {
-		ret = get_common_info("common_pid",
-				      &pid_offset,
-				      &pid_size);
-		if (ret < 0)
-			return ret;
-	}
+	ret = __parse_common(data, &ld_size, &ld_offset,
+			     "common_lock_depth");
+	if (ret < 0)
+		return -1;
 
-	return read_size(data + pid_offset, pid_size);
+	return ret;
 }
 
-static struct event *find_event(int id)
+struct event *trace_find_event(int id)
 {
 	struct event *event;
 
@@ -1848,6 +2009,7 @@ static unsigned long long eval_num_arg(void *data, int size,
 {
 	unsigned long long val = 0;
 	unsigned long long left, right;
+	struct print_arg *larg;
 
 	switch (arg->type) {
 	case PRINT_NULL:
@@ -1874,6 +2036,26 @@ static unsigned long long eval_num_arg(void *data, int size,
 		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]) {
@@ -1924,6 +2106,12 @@ static unsigned long long eval_num_arg(void *data, int size,
 				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);
 		}
@@ -1945,10 +2133,11 @@ static const struct flag flags[] = {
 	{ "NET_TX_SOFTIRQ", 2 },
 	{ "NET_RX_SOFTIRQ", 3 },
 	{ "BLOCK_SOFTIRQ", 4 },
-	{ "TASKLET_SOFTIRQ", 5 },
-	{ "SCHED_SOFTIRQ", 6 },
-	{ "HRTIMER_SOFTIRQ", 7 },
-	{ "RCU_SOFTIRQ", 8 },
+	{ "BLOCK_IOPOLL_SOFTIRQ", 5 },
+	{ "TASKLET_SOFTIRQ", 6 },
+	{ "SCHED_SOFTIRQ", 7 },
+	{ "HRTIMER_SOFTIRQ", 8 },
+	{ "RCU_SOFTIRQ", 9 },
 
 	{ "HRTIMER_NORESTART", 0 },
 	{ "HRTIMER_RESTART", 1 },
@@ -2121,8 +2310,9 @@ static struct print_arg *make_bprint_args(char *fmt, void *data, int size, struc
 			case 'u':
 			case 'x':
 			case 'i':
-				bptr = (void *)(((unsigned long)bptr + (long_size - 1)) &
-						~(long_size - 1));
+				/* the pointers are always 4 bytes aligned */
+				bptr = (void *)(((unsigned long)bptr + 3) &
+						~3);
 				switch (ls) {
 				case 0:
 				case 1:
@@ -2246,7 +2436,27 @@ static void pretty_print(void *data, int size, struct event *event)
 
 	for (; *ptr; ptr++) {
 		ls = 0;
-		if (*ptr == '%') {
+		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:
@@ -2353,6 +2563,41 @@ static inline int log10_cpu(int nb)
 	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);
+}
+
 /* taken from Linux, written by Frederic Weisbecker */
 static void print_graph_cpu(int cpu)
 {
@@ -2420,8 +2665,8 @@ get_return_for_leaf(int cpu, int cur_pid, unsigned long long cur_func,
 	int type;
 	int pid;
 
-	type = parse_common_type(next->data);
-	event = find_event(type);
+	type = trace_parse_common_type(next->data);
+	event = trace_find_event(type);
 	if (!event)
 		return NULL;
 
@@ -2502,8 +2747,8 @@ print_graph_entry_leaf(struct event *event, void *data, struct record *ret_rec)
 	int type;
 	int i;
 
-	type = parse_common_type(ret_rec->data);
-	ret_event = find_event(type);
+	type = trace_parse_common_type(ret_rec->data);
+	ret_event = trace_find_event(type);
 
 	field = find_field(ret_event, "rettime");
 	if (!field)
@@ -2596,6 +2841,11 @@ pretty_print_func_ent(void *data, int size, struct event *event,
 
 	printf(" | ");
 
+	if (latency_format) {
+		print_lat_fmt(data, size);
+		printf(" | ");
+	}
+
 	field = find_field(event, "func");
 	if (!field)
 		die("function entry does not have func field");
@@ -2639,6 +2889,11 @@ pretty_print_func_ret(void *data, int size __unused, struct event *event,
 
 	printf(" | ");
 
+	if (latency_format) {
+		print_lat_fmt(data, size);
+		printf(" | ");
+	}
+
 	field = find_field(event, "rettime");
 	if (!field)
 		die("can't find rettime in return graph");
@@ -2696,11 +2951,13 @@ void print_event(int cpu, void *data, int size, unsigned long long nsecs,
 	nsecs -= secs * NSECS_PER_SEC;
 	usecs = nsecs / NSECS_PER_USEC;
 
-	type = parse_common_type(data);
+	type = trace_parse_common_type(data);
 
-	event = find_event(type);
-	if (!event)
-		die("ug! no event found for type %d", type);
+	event = trace_find_event(type);
+	if (!event) {
+		warning("ug! no event found for type %d", type);
+		return;
+	}
 
 	pid = parse_common_pid(data);
 
@@ -2708,9 +2965,20 @@ void print_event(int cpu, void *data, int size, unsigned long long nsecs,
 		return pretty_print_func_graph(data, size, event, cpu,
 					       pid, comm, secs, usecs);
 
-	printf("%16s-%-5d [%03d] %5lu.%09Lu: %s: ",
-	       comm, pid,  cpu,
-	       secs, nsecs, event->name);
+	if (latency_format) {
+		printf("%8.8s-%-5d %3d",
+		       comm, pid, cpu);
+		print_lat_fmt(data, size);
+	} else
+		printf("%16s-%-5d [%03d]", comm, pid,  cpu);
+
+	printf(" %5lu.%06lu: %s: ", secs, usecs, event->name);
+
+	if (event->flags & EVENT_FL_FAILED) {
+		printf("EVENT '%s' FAILED TO PARSE\n",
+		       event->name);
+		return;
+	}
 
 	pretty_print(data, size, event);
 	printf("\n");
@@ -2781,46 +3049,71 @@ static void print_args(struct print_arg *args)
 	}
 }
 
-static void parse_header_field(char *type,
+static void parse_header_field(const char *field,
 			       int *offset, int *size)
 {
 	char *token;
+	int type;
 
-	if (read_expected(EVENT_ITEM, (char *)"field") < 0)
+	if (read_expected(EVENT_ITEM, "field") < 0)
 		return;
-	if (read_expected(EVENT_OP, (char *)":") < 0)
+	if (read_expected(EVENT_OP, ":") < 0)
 		return;
+
 	/* type */
 	if (read_expect_type(EVENT_ITEM, &token) < 0)
-		return;
+		goto fail;
 	free_token(token);
 
-	if (read_expected(EVENT_ITEM, type) < 0)
+	if (read_expected(EVENT_ITEM, field) < 0)
 		return;
-	if (read_expected(EVENT_OP, (char *)";") < 0)
+	if (read_expected(EVENT_OP, ";") < 0)
 		return;
-	if (read_expected(EVENT_ITEM, (char *)"offset") < 0)
+	if (read_expected(EVENT_ITEM, "offset") < 0)
 		return;
-	if (read_expected(EVENT_OP, (char *)":") < 0)
+	if (read_expected(EVENT_OP, ":") < 0)
 		return;
 	if (read_expect_type(EVENT_ITEM, &token) < 0)
-		return;
+		goto fail;
 	*offset = atoi(token);
 	free_token(token);
-	if (read_expected(EVENT_OP, (char *)";") < 0)
+	if (read_expected(EVENT_OP, ";") < 0)
 		return;
-	if (read_expected(EVENT_ITEM, (char *)"size") < 0)
+	if (read_expected(EVENT_ITEM, "size") < 0)
 		return;
-	if (read_expected(EVENT_OP, (char *)":") < 0)
+	if (read_expected(EVENT_OP, ":") < 0)
 		return;
 	if (read_expect_type(EVENT_ITEM, &token) < 0)
-		return;
+		goto fail;
 	*size = atoi(token);
 	free_token(token);
-	if (read_expected(EVENT_OP, (char *)";") < 0)
-		return;
-	if (read_expect_type(EVENT_NEWLINE, &token) < 0)
+	if (read_expected(EVENT_OP, ";") < 0)
 		return;
+	type = read_token(&token);
+	if (type != EVENT_NEWLINE) {
+		/* newer versions of the kernel have a "signed" type */
+		if (type != EVENT_ITEM)
+			goto fail;
+
+		if (strcmp(token, "signed") != 0)
+			goto fail;
+
+		free_token(token);
+
+		if (read_expected(EVENT_OP, ":") < 0)
+			return;
+
+		if (read_expect_type(EVENT_ITEM, &token))
+			goto fail;
+
+		free_token(token);
+		if (read_expected(EVENT_OP, ";") < 0)
+			return;
+
+		if (read_expect_type(EVENT_NEWLINE, &token))
+			goto fail;
+	}
+ fail:
 	free_token(token);
 }
 
@@ -2828,11 +3121,11 @@ int parse_header_page(char *buf, unsigned long size)
 {
 	init_input_buf(buf, size);
 
-	parse_header_field((char *)"timestamp", &header_page_ts_offset,
+	parse_header_field("timestamp", &header_page_ts_offset,
 			   &header_page_ts_size);
-	parse_header_field((char *)"commit", &header_page_size_offset,
+	parse_header_field("commit", &header_page_size_offset,
 			   &header_page_size_size);
-	parse_header_field((char *)"data", &header_page_data_offset,
+	parse_header_field("data", &header_page_data_offset,
 			   &header_page_data_size);
 
 	return 0;
@@ -2883,6 +3176,9 @@ int parse_ftrace_file(char *buf, unsigned long size)
 	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.
@@ -2900,7 +3196,7 @@ int parse_ftrace_file(char *buf, unsigned long size)
 	return 0;
 }
 
-int parse_event_file(char *buf, unsigned long size, char *system__unused __unused)
+int parse_event_file(char *buf, unsigned long size, char *sys)
 {
 	struct event *event;
 	int ret;
@@ -2920,12 +3216,18 @@ int parse_event_file(char *buf, unsigned long size, char *system__unused __unuse
 		die("failed to read event id");
 
 	ret = event_read_format(event);
-	if (ret < 0)
-		die("failed to read event format");
+	if (ret < 0) {
+		warning("failed to read event format for %s", event->name);
+		goto event_failed;
+	}
 
 	ret = event_read_print(event);
-	if (ret < 0)
-		die("failed to read event print fmt");
+	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)
@@ -2933,6 +3235,12 @@ int parse_event_file(char *buf, unsigned long size, char *system__unused __unuse
 
 	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)
diff --git a/tools/perf/util/trace-event-read.c b/tools/perf/util/trace-event-read.c
index a1217a10632f..44292e06cca4 100644
--- a/tools/perf/util/trace-event-read.c
+++ b/tools/perf/util/trace-event-read.c
@@ -458,18 +458,16 @@ struct record *trace_read_data(int cpu)
 	return data;
 }
 
-void trace_report (void)
+void trace_report(int fd)
 {
-	const char *input_file = "trace.info";
 	char buf[BUFSIZ];
 	char test[] = { 23, 8, 68 };
 	char *version;
+	int show_version = 0;
 	int show_funcs = 0;
 	int show_printk = 0;
 
-	input_fd = open(input_file, O_RDONLY);
-	if (input_fd < 0)
-		die("opening '%s'\n", input_file);
+	input_fd = fd;
 
 	read_or_die(buf, 3);
 	if (memcmp(buf, test, 3) != 0)
@@ -480,7 +478,8 @@ void trace_report (void)
 		die("not a trace file (missing tracing)");
 
 	version = read_string();
-	printf("version = %s\n", version);
+	if (show_version)
+		printf("version = %s\n", version);
 	free(version);
 
 	read_or_die(buf, 1);
diff --git a/tools/perf/util/trace-event.h b/tools/perf/util/trace-event.h
index 420294a5773e..f6637c2fa1fe 100644
--- a/tools/perf/util/trace-event.h
+++ b/tools/perf/util/trace-event.h
@@ -1,5 +1,5 @@
-#ifndef _TRACE_EVENTS_H
-#define _TRACE_EVENTS_H
+#ifndef __PERF_TRACE_EVENTS_H
+#define __PERF_TRACE_EVENTS_H
 
 #include "parse-events.h"
 
@@ -26,6 +26,9 @@ enum {
 enum format_flags {
 	FIELD_IS_ARRAY		= 1,
 	FIELD_IS_POINTER	= 2,
+	FIELD_IS_SIGNED		= 4,
+	FIELD_IS_STRING		= 8,
+	FIELD_IS_DYNAMIC	= 16,
 };
 
 struct format_field {
@@ -132,15 +135,18 @@ struct event {
 	int			flags;
 	struct format		format;
 	struct print_fmt	print_fmt;
+	char			*system;
 };
 
 enum {
-	EVENT_FL_ISFTRACE	= 1,
-	EVENT_FL_ISPRINT	= 2,
-	EVENT_FL_ISBPRINT	= 4,
-	EVENT_FL_ISFUNC		= 8,
-	EVENT_FL_ISFUNCENT	= 16,
-	EVENT_FL_ISFUNCRET	= 32,
+	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 {
@@ -154,7 +160,7 @@ struct record *trace_read_data(int cpu);
 
 void parse_set_info(int nr_cpus, int long_sz);
 
-void trace_report(void);
+void trace_report(int fd);
 
 void *malloc_or_die(unsigned int size);
 
@@ -166,7 +172,7 @@ 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 *system);
+int parse_event_file(char *buf, unsigned long size, char *sys);
 void print_event(int cpu, void *data, int size, unsigned long long nsecs,
 		  char *comm);
 
@@ -233,8 +239,24 @@ extern int header_page_size_size;
 extern int header_page_data_offset;
 extern int header_page_data_size;
 
-int parse_header_page(char *buf, unsigned long size);
+extern int latency_format;
 
-void read_tracing_data(struct perf_counter_attr *pattrs, int nb_counters);
+int parse_header_page(char *buf, unsigned long size);
+int trace_parse_common_type(void *data);
+struct event *trace_find_event(int id);
+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);
+
+void read_tracing_data(int fd, struct perf_event_attr *pattrs, int nb_events);
+
+/* 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,
+};
 
-#endif /* _TRACE_EVENTS_H */
+#endif /* __PERF_TRACE_EVENTS_H */
diff --git a/tools/perf/util/types.h b/tools/perf/util/types.h
index 5e75f9005940..7d6b8331f898 100644
--- a/tools/perf/util/types.h
+++ b/tools/perf/util/types.h
@@ -1,5 +1,5 @@
-#ifndef _PERF_TYPES_H
-#define _PERF_TYPES_H
+#ifndef __PERF_TYPES_H
+#define __PERF_TYPES_H
 
 /*
  * We define u64 as unsigned long long for every architecture
@@ -14,4 +14,4 @@ typedef signed short	   s16;
 typedef unsigned char	   u8;
 typedef signed char	   s8;
 
-#endif /* _PERF_TYPES_H */
+#endif /* __PERF_TYPES_H */
diff --git a/tools/perf/util/values.h b/tools/perf/util/values.h
index cadf8cf2a590..2fa967e1a88a 100644
--- a/tools/perf/util/values.h
+++ b/tools/perf/util/values.h
@@ -1,5 +1,5 @@
-#ifndef _PERF_VALUES_H
-#define _PERF_VALUES_H
+#ifndef __PERF_VALUES_H
+#define __PERF_VALUES_H
 
 #include "types.h"
 
@@ -24,4 +24,4 @@ void perf_read_values_add_value(struct perf_read_values *values,
 void perf_read_values_display(FILE *fp, struct perf_read_values *values,
 			      int raw);
 
-#endif /* _PERF_VALUES_H */
+#endif /* __PERF_VALUES_H */