Merge branch 'origin' into devel-stable

Conflicts:
	arch/arm/mach-mx2/devices.c
	arch/arm/mach-mx2/devices.h
	sound/soc/pxa/pxa-ssp.c

38 files changed, 3318 insertions, 1351 deletions

diff --git a/tools/perf/util/build-id.c b/tools/perf/util/build-id.c
new file mode 100644
index 00000000000..04904b35ba8
--- /dev/null
+++ b/tools/perf/util/build-id.c
@@ -0,0 +1,39 @@
+/*
+ * build-id.c
+ *
+ * build-id support
+ *
+ * Copyright (C) 2009, 2010 Red Hat Inc.
+ * Copyright (C) 2009, 2010 Arnaldo Carvalho de Melo <acme@redhat.com>
+ */
+#include "build-id.h"
+#include "event.h"
+#include "symbol.h"
+#include <linux/kernel.h>
+
+static int build_id__mark_dso_hit(event_t *event, struct perf_session *session)
+{
+	struct addr_location al;
+	u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+	struct thread *thread = perf_session__findnew(session, event->ip.pid);
+
+	if (thread == NULL) {
+		pr_err("problem processing %d event, skipping it.\n",
+			event->header.type);
+		return -1;
+	}
+
+	thread__find_addr_map(thread, session, cpumode, MAP__FUNCTION,
+			      event->ip.ip, &al);
+
+	if (al.map != NULL)
+		al.map->dso->hit = 1;
+
+	return 0;
+}
+
+struct perf_event_ops build_id__mark_dso_hit_ops = {
+	.sample	= build_id__mark_dso_hit,
+	.mmap	= event__process_mmap,
+	.fork	= event__process_task,
+};
diff --git a/tools/perf/util/build-id.h b/tools/perf/util/build-id.h
new file mode 100644
index 00000000000..1d981d63cf9
--- /dev/null
+++ b/tools/perf/util/build-id.h
@@ -0,0 +1,8 @@
+#ifndef PERF_BUILD_ID_H_
+#define PERF_BUILD_ID_H_ 1
+
+#include "session.h"
+
+extern struct perf_event_ops build_id__mark_dso_hit_ops;
+
+#endif
diff --git a/tools/perf/util/data_map.c b/tools/perf/util/data_map.c
deleted file mode 100644
index b557b836de3..00000000000
--- a/tools/perf/util/data_map.c
+++ /dev/null
@@ -1,252 +0,0 @@
-#include "symbol.h"
-#include "util.h"
-#include "debug.h"
-#include "thread.h"
-#include "session.h"
-
-static int process_event_stub(event_t *event __used,
-			      struct perf_session *session __used)
-{
-	dump_printf(": unhandled!\n");
-	return 0;
-}
-
-static void perf_event_ops__fill_defaults(struct perf_event_ops *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;
-}
-
-static const char *event__name[] = {
-	[0]			 = "TOTAL",
-	[PERF_RECORD_MMAP]	 = "MMAP",
-	[PERF_RECORD_LOST]	 = "LOST",
-	[PERF_RECORD_COMM]	 = "COMM",
-	[PERF_RECORD_EXIT]	 = "EXIT",
-	[PERF_RECORD_THROTTLE]	 = "THROTTLE",
-	[PERF_RECORD_UNTHROTTLE] = "UNTHROTTLE",
-	[PERF_RECORD_FORK]	 = "FORK",
-	[PERF_RECORD_READ]	 = "READ",
-	[PERF_RECORD_SAMPLE]	 = "SAMPLE",
-};
-
-unsigned long event__total[PERF_RECORD_MAX];
-
-void event__print_totals(void)
-{
-	int i;
-	for (i = 0; i < PERF_RECORD_MAX; ++i)
-		pr_info("%10s events: %10ld\n",
-			event__name[i], event__total[i]);
-}
-
-static int process_event(event_t *event, struct perf_session *session,
-			 struct perf_event_ops *ops,
-			 unsigned long offset, unsigned long head)
-{
-	trace_event(event);
-
-	if (event->header.type < PERF_RECORD_MAX) {
-		dump_printf("%p [%p]: PERF_RECORD_%s",
-			    (void *)(offset + head),
-			    (void *)(long)(event->header.size),
-			    event__name[event->header.type]);
-		++event__total[0];
-		++event__total[event->header.type];
-	}
-
-	switch (event->header.type) {
-	case PERF_RECORD_SAMPLE:
-		return ops->process_sample_event(event, session);
-	case PERF_RECORD_MMAP:
-		return ops->process_mmap_event(event, session);
-	case PERF_RECORD_COMM:
-		return ops->process_comm_event(event, session);
-	case PERF_RECORD_FORK:
-		return ops->process_fork_event(event, session);
-	case PERF_RECORD_EXIT:
-		return ops->process_exit_event(event, session);
-	case PERF_RECORD_LOST:
-		return ops->process_lost_event(event, session);
-	case PERF_RECORD_READ:
-		return ops->process_read_event(event, session);
-	case PERF_RECORD_THROTTLE:
-		return ops->process_throttle_event(event, session);
-	case PERF_RECORD_UNTHROTTLE:
-		return ops->process_unthrottle_event(event, session);
-	default:
-		ops->total_unknown++;
-		return -1;
-	}
-}
-
-int perf_header__read_build_ids(int input, u64 offset, u64 size)
-{
-	struct build_id_event bev;
-	char filename[PATH_MAX];
-	u64 limit = offset + size;
-	int err = -1;
-
-	while (offset < limit) {
-		struct dso *dso;
-		ssize_t len;
-
-		if (read(input, &bev, sizeof(bev)) != sizeof(bev))
-			goto out;
-
-		len = bev.header.size - sizeof(bev);
-		if (read(input, filename, len) != len)
-			goto out;
-
-		dso = dsos__findnew(filename);
-		if (dso != NULL)
-			dso__set_build_id(dso, &bev.build_id);
-
-		offset += bev.header.size;
-	}
-	err = 0;
-out:
-	return err;
-}
-
-static struct thread *perf_session__register_idle_thread(struct perf_session *self)
-{
-	struct thread *thread = perf_session__findnew(self, 0);
-
-	if (!thread || thread__set_comm(thread, "swapper")) {
-		pr_err("problem inserting idle task.\n");
-		thread = NULL;
-	}
-
-	return thread;
-}
-
-int perf_session__process_events(struct perf_session *self,
-				 struct perf_event_ops *ops)
-{
-	int err;
-	unsigned long head, shift;
-	unsigned long offset = 0;
-	size_t	page_size;
-	event_t *event;
-	uint32_t size;
-	char *buf;
-
-	if (perf_session__register_idle_thread(self) == NULL)
-		return -ENOMEM;
-
-	perf_event_ops__fill_defaults(ops);
-
-	page_size = getpagesize();
-
-	head = self->header.data_offset;
-	self->sample_type = perf_header__sample_type(&self->header);
-
-	err = -EINVAL;
-	if (ops->sample_type_check && ops->sample_type_check(self) < 0)
-		goto out_err;
-
-	if (!ops->full_paths) {
-		char bf[PATH_MAX];
-
-		if (getcwd(bf, sizeof(bf)) == NULL) {
-			err = -errno;
-out_getcwd_err:
-			pr_err("failed to get the current directory\n");
-			goto out_err;
-		}
-		self->cwd = strdup(bf);
-		if (self->cwd == NULL) {
-			err = -ENOMEM;
-			goto out_getcwd_err;
-		}
-		self->cwdlen = strlen(self->cwd);
-	}
-
-	shift = page_size * (head / page_size);
-	offset += shift;
-	head -= shift;
-
-remap:
-	buf = mmap(NULL, page_size * self->mmap_window, PROT_READ,
-		   MAP_SHARED, self->fd, offset);
-	if (buf == MAP_FAILED) {
-		pr_err("failed to mmap file\n");
-		err = -errno;
-		goto out_err;
-	}
-
-more:
-	event = (event_t *)(buf + head);
-
-	size = event->header.size;
-	if (!size)
-		size = 8;
-
-	if (head + event->header.size >= page_size * self->mmap_window) {
-		int munmap_ret;
-
-		shift = page_size * (head / page_size);
-
-		munmap_ret = munmap(buf, page_size * self->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, self, ops, 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 >= self->header.data_offset + self->header.data_size)
-		goto done;
-
-	if (offset + head < self->size)
-		goto more;
-
-done:
-	err = 0;
-out_err:
-	return err;
-}
diff --git a/tools/perf/util/debug.c b/tools/perf/util/debug.c
index 28d520d5a1f..0905600c385 100644
--- a/tools/perf/util/debug.c
+++ b/tools/perf/util/debug.c
@@ -9,6 +9,7 @@
 #include "color.h"
 #include "event.h"
 #include "debug.h"
+#include "util.h"
 
 int verbose = 0;
 int dump_trace = 0;
diff --git a/tools/perf/util/debugfs.c b/tools/perf/util/debugfs.c
index 06b73ee02c4..a88fefc0cc0 100644
--- a/tools/perf/util/debugfs.c
+++ b/tools/perf/util/debugfs.c
@@ -106,16 +106,14 @@ int debugfs_valid_entry(const char *path)
 	return 0;
 }
 
-/* mount the debugfs somewhere */
+/* mount the debugfs somewhere if it's not mounted */
 
-int debugfs_mount(const char *mountpoint)
+char *debugfs_mount(const char *mountpoint)
 {
-	char mountcmd[128];
-
 	/* see if it's already mounted */
 	if (debugfs_find_mountpoint()) {
 		debugfs_premounted = 1;
-		return 0;
+		return debugfs_mountpoint;
 	}
 
 	/* if not mounted and no argument */
@@ -127,13 +125,14 @@ int debugfs_mount(const char *mountpoint)
 			mountpoint = "/sys/kernel/debug";
 	}
 
+	if (mount(NULL, mountpoint, "debugfs", 0, NULL) < 0)
+		return NULL;
+
 	/* save the mountpoint */
 	strncpy(debugfs_mountpoint, mountpoint, sizeof(debugfs_mountpoint));
+	debugfs_found = 1;
 
-	/* mount it */
-	snprintf(mountcmd, sizeof(mountcmd),
-		 "/bin/mount -t debugfs debugfs %s", mountpoint);
-	return system(mountcmd);
+	return debugfs_mountpoint;
 }
 
 /* umount the debugfs */
diff --git a/tools/perf/util/debugfs.h b/tools/perf/util/debugfs.h
index 3cd14f9ae78..83a02879745 100644
--- a/tools/perf/util/debugfs.h
+++ b/tools/perf/util/debugfs.h
@@ -15,7 +15,7 @@
 extern const char *debugfs_find_mountpoint(void);
 extern int debugfs_valid_mountpoint(const char *debugfs);
 extern int debugfs_valid_entry(const char *path);
-extern int debugfs_mount(const char *mountpoint);
+extern char *debugfs_mount(const char *mountpoint);
 extern int debugfs_umount(void);
 extern int debugfs_write(const char *entry, const char *value);
 extern int debugfs_read(const char *entry, char *buffer, size_t size);
diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c
index 8a9e6baa309..705ec63548b 100644
--- a/tools/perf/util/event.c
+++ b/tools/perf/util/event.c
@@ -8,8 +8,7 @@
 #include "thread.h"
 
 static pid_t event__synthesize_comm(pid_t pid, int full,
-				    int (*process)(event_t *event,
-						   struct perf_session *session),
+				    event__handler_t process,
 				    struct perf_session *session)
 {
 	event_t ev;
@@ -91,8 +90,7 @@ out_failure:
 }
 
 static int event__synthesize_mmap_events(pid_t pid, pid_t tgid,
-					 int (*process)(event_t *event,
-							struct perf_session *session),
+					 event__handler_t process,
 					 struct perf_session *session)
 {
 	char filename[PATH_MAX];
@@ -112,7 +110,10 @@ static int event__synthesize_mmap_events(pid_t pid, pid_t tgid,
 	while (1) {
 		char bf[BUFSIZ], *pbf = bf;
 		event_t ev = {
-			.header = { .type = PERF_RECORD_MMAP },
+			.header = {
+				.type = PERF_RECORD_MMAP,
+				.misc = 0, /* Just like the kernel, see kernel/perf_event.c __perf_event_mmap */
+			 },
 		};
 		int n;
 		size_t size;
@@ -156,9 +157,38 @@ static int event__synthesize_mmap_events(pid_t pid, pid_t tgid,
 	return 0;
 }
 
-int event__synthesize_thread(pid_t pid,
-			     int (*process)(event_t *event,
-					    struct perf_session *session),
+int event__synthesize_modules(event__handler_t process,
+			      struct perf_session *session)
+{
+	struct rb_node *nd;
+
+	for (nd = rb_first(&session->kmaps.maps[MAP__FUNCTION]);
+	     nd; nd = rb_next(nd)) {
+		event_t ev;
+		size_t size;
+		struct map *pos = rb_entry(nd, struct map, rb_node);
+
+		if (pos->dso->kernel)
+			continue;
+
+		size = ALIGN(pos->dso->long_name_len + 1, sizeof(u64));
+		memset(&ev, 0, sizeof(ev));
+		ev.mmap.header.misc = 1; /* kernel uses 0 for user space maps, see kernel/perf_event.c __perf_event_mmap */
+		ev.mmap.header.type = PERF_RECORD_MMAP;
+		ev.mmap.header.size = (sizeof(ev.mmap) -
+				        (sizeof(ev.mmap.filename) - size));
+		ev.mmap.start = pos->start;
+		ev.mmap.len   = pos->end - pos->start;
+
+		memcpy(ev.mmap.filename, pos->dso->long_name,
+		       pos->dso->long_name_len + 1);
+		process(&ev, session);
+	}
+
+	return 0;
+}
+
+int event__synthesize_thread(pid_t pid, event__handler_t process,
 			     struct perf_session *session)
 {
 	pid_t tgid = event__synthesize_comm(pid, 1, process, session);
@@ -167,8 +197,7 @@ int event__synthesize_thread(pid_t pid,
 	return event__synthesize_mmap_events(pid, tgid, process, session);
 }
 
-void event__synthesize_threads(int (*process)(event_t *event,
-					      struct perf_session *session),
+void event__synthesize_threads(event__handler_t process,
 			       struct perf_session *session)
 {
 	DIR *proc;
@@ -189,6 +218,59 @@ void event__synthesize_threads(int (*process)(event_t *event,
 	closedir(proc);
 }
 
+struct process_symbol_args {
+	const char *name;
+	u64	   start;
+};
+
+static int find_symbol_cb(void *arg, const char *name, char type, u64 start)
+{
+	struct process_symbol_args *args = arg;
+
+	/*
+	 * Must be a function or at least an alias, as in PARISC64, where "_text" is
+	 * an 'A' to the same address as "_stext".
+	 */
+	if (!(symbol_type__is_a(type, MAP__FUNCTION) ||
+	      type == 'A') || strcmp(name, args->name))
+		return 0;
+
+	args->start = start;
+	return 1;
+}
+
+int event__synthesize_kernel_mmap(event__handler_t process,
+				  struct perf_session *session,
+				  const char *symbol_name)
+{
+	size_t size;
+	event_t ev = {
+		.header = {
+			.type = PERF_RECORD_MMAP,
+			.misc = 1, /* kernel uses 0 for user space maps, see kernel/perf_event.c __perf_event_mmap */
+		},
+	};
+	/*
+	 * We should get this from /sys/kernel/sections/.text, but till that is
+	 * available use this, and after it is use this as a fallback for older
+	 * kernels.
+	 */
+	struct process_symbol_args args = { .name = symbol_name, };
+
+	if (kallsyms__parse("/proc/kallsyms", &args, find_symbol_cb) <= 0)
+		return -ENOENT;
+
+	size = snprintf(ev.mmap.filename, sizeof(ev.mmap.filename),
+			"[kernel.kallsyms.%s]", symbol_name) + 1;
+	size = ALIGN(size, sizeof(u64));
+	ev.mmap.header.size = (sizeof(ev.mmap) - (sizeof(ev.mmap.filename) - size));
+	ev.mmap.pgoff = args.start;
+	ev.mmap.start = session->vmlinux_maps[MAP__FUNCTION]->start;
+	ev.mmap.len   = session->vmlinux_maps[MAP__FUNCTION]->end - ev.mmap.start ;
+
+	return process(&ev, session);
+}
+
 static void thread__comm_adjust(struct thread *self)
 {
 	char *comm = self->comm;
@@ -240,22 +322,88 @@ int event__process_lost(event_t *self, struct perf_session *session)
 
 int event__process_mmap(event_t *self, struct perf_session *session)
 {
-	struct thread *thread = perf_session__findnew(session, self->mmap.pid);
-	struct map *map = map__new(&self->mmap, MAP__FUNCTION,
-				   session->cwd, session->cwdlen);
+	struct thread *thread;
+	struct map *map;
+
+	dump_printf(" %d/%d: [%#Lx(%#Lx) @ %#Lx]: %s\n",
+		    self->mmap.pid, self->mmap.tid, self->mmap.start,
+		    self->mmap.len, self->mmap.pgoff, self->mmap.filename);
+
+	if (self->mmap.pid == 0) {
+		static const char kmmap_prefix[] = "[kernel.kallsyms.";
+
+		if (self->mmap.filename[0] == '/') {
+			char short_module_name[1024];
+			char *name = strrchr(self->mmap.filename, '/'), *dot;
+
+			if (name == NULL)
+				goto out_problem;
+
+			++name; /* skip / */
+			dot = strrchr(name, '.');
+			if (dot == NULL)
+				goto out_problem;
+
+			snprintf(short_module_name, sizeof(short_module_name),
+				 "[%.*s]", (int)(dot - name), name);
+			strxfrchar(short_module_name, '-', '_');
+
+			map = perf_session__new_module_map(session,
+							   self->mmap.start,
+							   self->mmap.filename);
+			if (map == NULL)
+				goto out_problem;
+
+			name = strdup(short_module_name);
+			if (name == NULL)
+				goto out_problem;
+
+			map->dso->short_name = name;
+			map->end = map->start + self->mmap.len;
+		} else if (memcmp(self->mmap.filename, kmmap_prefix,
+				sizeof(kmmap_prefix) - 1) == 0) {
+			const char *symbol_name = (self->mmap.filename +
+						   sizeof(kmmap_prefix) - 1);
+			/*
+			 * Should be there already, from the build-id table in
+			 * the header.
+			 */
+			struct dso *kernel = __dsos__findnew(&dsos__kernel,
+							     "[kernel.kallsyms]");
+			if (kernel == NULL)
+				goto out_problem;
+
+			kernel->kernel = 1;
+			if (__perf_session__create_kernel_maps(session, kernel) < 0)
+				goto out_problem;
+
+			session->vmlinux_maps[MAP__FUNCTION]->start = self->mmap.start;
+			session->vmlinux_maps[MAP__FUNCTION]->end   = self->mmap.start + self->mmap.len;
+			/*
+			 * Be a bit paranoid here, some perf.data file came with
+			 * a zero sized synthesized MMAP event for the kernel.
+			 */
+			if (session->vmlinux_maps[MAP__FUNCTION]->end == 0)
+				session->vmlinux_maps[MAP__FUNCTION]->end = ~0UL;
+
+			perf_session__set_kallsyms_ref_reloc_sym(session, symbol_name,
+								 self->mmap.pgoff);
+		}
+		return 0;
+	}
 
-	dump_printf(" %d/%d: [%p(%p) @ %p]: %s\n",
-		    self->mmap.pid, self->mmap.tid,
-		    (void *)(long)self->mmap.start,
-		    (void *)(long)self->mmap.len,
-		    (void *)(long)self->mmap.pgoff,
-		    self->mmap.filename);
+	thread = perf_session__findnew(session, self->mmap.pid);
+	map = map__new(&self->mmap, MAP__FUNCTION,
+		       session->cwd, session->cwdlen);
 
 	if (thread == NULL || map == NULL)
-		dump_printf("problem processing PERF_RECORD_MMAP, skipping event.\n");
-	else
-		thread__insert_map(thread, map);
+		goto out_problem;
 
+	thread__insert_map(thread, map);
+	return 0;
+
+out_problem:
+	dump_printf("problem processing PERF_RECORD_MMAP, skipping event.\n");
 	return 0;
 }
 
@@ -284,11 +432,10 @@ int event__process_task(event_t *self, struct perf_session *session)
 	return 0;
 }
 
-void thread__find_addr_location(struct thread *self,
-				struct perf_session *session, u8 cpumode,
-				enum map_type type, u64 addr,
-				struct addr_location *al,
-				symbol_filter_t filter)
+void thread__find_addr_map(struct thread *self,
+			   struct perf_session *session, u8 cpumode,
+			   enum map_type type, u64 addr,
+			   struct addr_location *al)
 {
 	struct map_groups *mg = &self->mg;
 
@@ -303,7 +450,6 @@ void thread__find_addr_location(struct thread *self,
 	else {
 		al->level = 'H';
 		al->map = NULL;
-		al->sym = NULL;
 		return;
 	}
 try_again:
@@ -322,11 +468,21 @@ try_again:
 			mg = &session->kmaps;
 			goto try_again;
 		}
-		al->sym = NULL;
-	} else {
+	} else
 		al->addr = al->map->map_ip(al->map, al->addr);
-		al->sym = map__find_symbol(al->map, session, al->addr, filter);
-	}
+}
+
+void thread__find_addr_location(struct thread *self,
+				struct perf_session *session, u8 cpumode,
+				enum map_type type, u64 addr,
+				struct addr_location *al,
+				symbol_filter_t filter)
+{
+	thread__find_addr_map(self, session, cpumode, type, addr, al);
+	if (al->map != NULL)
+		al->sym = map__find_symbol(al->map, al->addr, filter);
+	else
+		al->sym = NULL;
 }
 
 static void dso__calc_col_width(struct dso *self)
diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h
index 690a96d0467..50a7132887f 100644
--- a/tools/perf/util/event.h
+++ b/tools/perf/util/event.h
@@ -1,10 +1,10 @@
 #ifndef __PERF_RECORD_H
 #define __PERF_RECORD_H
 
+#include <limits.h>
+
 #include "../perf.h"
-#include "util.h"
-#include <linux/list.h>
-#include <linux/rbtree.h>
+#include "map.h"
 
 /*
  * PERF_SAMPLE_IP | PERF_SAMPLE_TID | *
@@ -101,74 +101,19 @@ struct events_stats {
 
 void event__print_totals(void);
 
-enum map_type {
-	MAP__FUNCTION = 0,
-	MAP__VARIABLE,
-};
-
-#define MAP__NR_TYPES (MAP__VARIABLE + 1)
-
-struct map {
-	union {
-		struct rb_node	rb_node;
-		struct list_head node;
-	};
-	u64			start;
-	u64			end;
-	enum map_type		type;
-	u64			pgoff;
-	u64			(*map_ip)(struct map *, u64);
-	u64			(*unmap_ip)(struct map *, u64);
-	struct dso		*dso;
-};
-
-static inline u64 map__map_ip(struct map *map, u64 ip)
-{
-	return ip - map->start + map->pgoff;
-}
-
-static inline u64 map__unmap_ip(struct map *map, u64 ip)
-{
-	return ip + map->start - map->pgoff;
-}
-
-static inline u64 identity__map_ip(struct map *map __used, u64 ip)
-{
-	return ip;
-}
-
-struct symbol;
-
-typedef int (*symbol_filter_t)(struct map *map, struct symbol *sym);
-
-void map__init(struct map *self, enum map_type type,
-	       u64 start, u64 end, u64 pgoff, struct dso *dso);
-struct map *map__new(struct mmap_event *event, enum map_type,
-		     char *cwd, int cwdlen);
-void map__delete(struct map *self);
-struct map *map__clone(struct map *self);
-int map__overlap(struct map *l, struct map *r);
-size_t map__fprintf(struct map *self, FILE *fp);
-
 struct perf_session;
 
-int map__load(struct map *self, struct perf_session *session,
-	      symbol_filter_t filter);
-struct symbol *map__find_symbol(struct map *self, struct perf_session *session,
-				u64 addr, symbol_filter_t filter);
-struct symbol *map__find_symbol_by_name(struct map *self, const char *name,
-					struct perf_session *session,
-					symbol_filter_t filter);
-void map__fixup_start(struct map *self);
-void map__fixup_end(struct map *self);
-
-int event__synthesize_thread(pid_t pid,
-			     int (*process)(event_t *event,
-					    struct perf_session *session),
+typedef int (*event__handler_t)(event_t *event, struct perf_session *session);
+
+int event__synthesize_thread(pid_t pid, event__handler_t process,
 			     struct perf_session *session);
-void event__synthesize_threads(int (*process)(event_t *event,
-					      struct perf_session *session),
+void event__synthesize_threads(event__handler_t process,
 			       struct perf_session *session);
+int event__synthesize_kernel_mmap(event__handler_t process,
+				  struct perf_session *session,
+				  const char *symbol_name);
+int event__synthesize_modules(event__handler_t process,
+			      struct perf_session *session);
 
 int event__process_comm(event_t *self, struct perf_session *session);
 int event__process_lost(event_t *self, struct perf_session *session);
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index 8a0bca55106..6c9aa16ee51 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -1,8 +1,12 @@
+#define _FILE_OFFSET_BITS 64
+
 #include <sys/types.h>
+#include <byteswap.h>
 #include <unistd.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <linux/list.h>
+#include <linux/kernel.h>
 
 #include "util.h"
 #include "header.h"
@@ -105,24 +109,28 @@ struct perf_trace_event_type {
 static int event_count;
 static struct perf_trace_event_type *events;
 
-void perf_header__push_event(u64 id, const char *name)
+int perf_header__push_event(u64 id, const char *name)
 {
 	if (strlen(name) > MAX_EVENT_NAME)
 		pr_warning("Event %s will be truncated\n", name);
 
 	if (!events) {
 		events = malloc(sizeof(struct perf_trace_event_type));
-		if (!events)
-			die("nomem");
+		if (events == NULL)
+			return -ENOMEM;
 	} else {
-		events = realloc(events, (event_count + 1) * sizeof(struct perf_trace_event_type));
-		if (!events)
-			die("nomem");
+		struct perf_trace_event_type *nevents;
+
+		nevents = realloc(events, (event_count + 1) * sizeof(*events));
+		if (nevents == NULL)
+			return -ENOMEM;
+		events = nevents;
 	}
 	memset(&events[event_count], 0, sizeof(struct perf_trace_event_type));
 	events[event_count].event_id = id;
 	strncpy(events[event_count].name, name, MAX_EVENT_NAME - 1);
 	event_count++;
+	return 0;
 }
 
 char *perf_header__find_event(u64 id)
@@ -169,31 +177,48 @@ static int do_write(int fd, const void *buf, size_t size)
 	return 0;
 }
 
-static int __dsos__write_buildid_table(struct list_head *head, int fd)
+#define NAME_ALIGN 64
+
+static int write_padded(int fd, const void *bf, size_t count,
+			size_t count_aligned)
 {
-#define NAME_ALIGN	64
-	struct dso *pos;
 	static const char zero_buf[NAME_ALIGN];
+	int err = do_write(fd, bf, count);
+
+	if (!err)
+		err = do_write(fd, zero_buf, count_aligned - count);
+
+	return err;
+}
 
-	list_for_each_entry(pos, head, node) {
+#define dsos__for_each_with_build_id(pos, head)	\
+	list_for_each_entry(pos, head, node)	\
+		if (!pos->has_build_id)		\
+			continue;		\
+		else
+
+static int __dsos__write_buildid_table(struct list_head *head, u16 misc, int fd)
+{
+	struct dso *pos;
+
+	dsos__for_each_with_build_id(pos, head) {
 		int err;
 		struct build_id_event b;
 		size_t len;
 
-		if (!pos->has_build_id)
+		if (!pos->hit)
 			continue;
 		len = pos->long_name_len + 1;
 		len = ALIGN(len, NAME_ALIGN);
 		memset(&b, 0, sizeof(b));
 		memcpy(&b.build_id, pos->build_id, sizeof(pos->build_id));
+		b.header.misc = misc;
 		b.header.size = sizeof(b) + len;
 		err = do_write(fd, &b, sizeof(b));
 		if (err < 0)
 			return err;
-		err = do_write(fd, pos->long_name, pos->long_name_len + 1);
-		if (err < 0)
-			return err;
-		err = do_write(fd, zero_buf, len - pos->long_name_len - 1);
+		err = write_padded(fd, pos->long_name,
+				   pos->long_name_len + 1, len);
 		if (err < 0)
 			return err;
 	}
@@ -203,12 +228,143 @@ static int __dsos__write_buildid_table(struct list_head *head, int fd)
 
 static int dsos__write_buildid_table(int fd)
 {
-	int err = __dsos__write_buildid_table(&dsos__kernel, fd);
+	int err = __dsos__write_buildid_table(&dsos__kernel,
+					      PERF_RECORD_MISC_KERNEL, fd);
 	if (err == 0)
-		err = __dsos__write_buildid_table(&dsos__user, fd);
+		err = __dsos__write_buildid_table(&dsos__user,
+						  PERF_RECORD_MISC_USER, fd);
 	return err;
 }
 
+int build_id_cache__add_s(const char *sbuild_id, const char *debugdir,
+			  const char *name, bool is_kallsyms)
+{
+	const size_t size = PATH_MAX;
+	char *filename = malloc(size),
+	     *linkname = malloc(size), *targetname;
+	int len, err = -1;
+
+	if (filename == NULL || linkname == NULL)
+		goto out_free;
+
+	len = snprintf(filename, size, "%s%s%s",
+		       debugdir, is_kallsyms ? "/" : "", name);
+	if (mkdir_p(filename, 0755))
+		goto out_free;
+
+	snprintf(filename + len, sizeof(filename) - len, "/%s", sbuild_id);
+
+	if (access(filename, F_OK)) {
+		if (is_kallsyms) {
+			 if (copyfile("/proc/kallsyms", filename))
+				goto out_free;
+		} else if (link(name, filename) && copyfile(name, filename))
+			goto out_free;
+	}
+
+	len = snprintf(linkname, size, "%s/.build-id/%.2s",
+		       debugdir, sbuild_id);
+
+	if (access(linkname, X_OK) && mkdir_p(linkname, 0755))
+		goto out_free;
+
+	snprintf(linkname + len, size - len, "/%s", sbuild_id + 2);
+	targetname = filename + strlen(debugdir) - 5;
+	memcpy(targetname, "../..", 5);
+
+	if (symlink(targetname, linkname) == 0)
+		err = 0;
+out_free:
+	free(filename);
+	free(linkname);
+	return err;
+}
+
+static int build_id_cache__add_b(const u8 *build_id, size_t build_id_size,
+				 const char *name, const char *debugdir,
+				 bool is_kallsyms)
+{
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+	build_id__sprintf(build_id, build_id_size, sbuild_id);
+
+	return build_id_cache__add_s(sbuild_id, debugdir, name, is_kallsyms);
+}
+
+int build_id_cache__remove_s(const char *sbuild_id, const char *debugdir)
+{
+	const size_t size = PATH_MAX;
+	char *filename = malloc(size),
+	     *linkname = malloc(size);
+	int err = -1;
+
+	if (filename == NULL || linkname == NULL)
+		goto out_free;
+
+	snprintf(linkname, size, "%s/.build-id/%.2s/%s",
+		 debugdir, sbuild_id, sbuild_id + 2);
+
+	if (access(linkname, F_OK))
+		goto out_free;
+
+	if (readlink(linkname, filename, size) < 0)
+		goto out_free;
+
+	if (unlink(linkname))
+		goto out_free;
+
+	/*
+	 * Since the link is relative, we must make it absolute:
+	 */
+	snprintf(linkname, size, "%s/.build-id/%.2s/%s",
+		 debugdir, sbuild_id, filename);
+
+	if (unlink(linkname))
+		goto out_free;
+
+	err = 0;
+out_free:
+	free(filename);
+	free(linkname);
+	return err;
+}
+
+static int dso__cache_build_id(struct dso *self, const char *debugdir)
+{
+	bool is_kallsyms = self->kernel && self->long_name[0] != '/';
+
+	return build_id_cache__add_b(self->build_id, sizeof(self->build_id),
+				     self->long_name, debugdir, is_kallsyms);
+}
+
+static int __dsos__cache_build_ids(struct list_head *head, const char *debugdir)
+{
+	struct dso *pos;
+	int err = 0;
+
+	dsos__for_each_with_build_id(pos, head)
+		if (dso__cache_build_id(pos, debugdir))
+			err = -1;
+
+	return err;
+}
+
+static int dsos__cache_build_ids(void)
+{
+	int err_kernel, err_user;
+	char debugdir[PATH_MAX];
+
+	snprintf(debugdir, sizeof(debugdir), "%s/%s", getenv("HOME"),
+		 DEBUG_CACHE_DIR);
+
+	if (mkdir(debugdir, 0755) != 0 && errno != EEXIST)
+		return -1;
+
+	err_kernel = __dsos__cache_build_ids(&dsos__kernel, debugdir);
+	err_user   = __dsos__cache_build_ids(&dsos__user, debugdir);
+	return err_kernel || err_user ? -1 : 0;
+}
+
 static int perf_header__adds_write(struct perf_header *self, int fd)
 {
 	int nr_sections;
@@ -217,7 +373,7 @@ static int perf_header__adds_write(struct perf_header *self, int fd)
 	u64 sec_start;
 	int idx = 0, err;
 
-	if (dsos__read_build_ids())
+	if (dsos__read_build_ids(true))
 		perf_header__set_feat(self, HEADER_BUILD_ID);
 
 	nr_sections = bitmap_weight(self->adds_features, HEADER_FEAT_BITS);
@@ -257,7 +413,9 @@ static int perf_header__adds_write(struct perf_header *self, int fd)
 			pr_debug("failed to write buildid table\n");
 			goto out_free;
 		}
-		buildid_sec->size = lseek(fd, 0, SEEK_CUR) - buildid_sec->offset;
+		buildid_sec->size = lseek(fd, 0, SEEK_CUR) -
+					  buildid_sec->offset;
+		dsos__cache_build_ids();
 	}
 
 	lseek(fd, sec_start, SEEK_SET);
@@ -360,30 +518,43 @@ int perf_header__write(struct perf_header *self, int fd, bool at_exit)
 	return 0;
 }
 
-static void do_read(int fd, void *buf, size_t size)
+static int do_read(int fd, void *buf, size_t size)
 {
 	while (size) {
 		int ret = read(fd, buf, size);
 
-		if (ret < 0)
-			die("failed to read");
-		if (ret == 0)
-			die("failed to read: missing data");
+		if (ret <= 0)
+			return -1;
 
 		size -= ret;
 		buf += ret;
 	}
+
+	return 0;
+}
+
+static int perf_header__getbuffer64(struct perf_header *self,
+				    int fd, void *buf, size_t size)
+{
+	if (do_read(fd, buf, size))
+		return -1;
+
+	if (self->needs_swap)
+		mem_bswap_64(buf, size);
+
+	return 0;
 }
 
 int perf_header__process_sections(struct perf_header *self, int fd,
 				  int (*process)(struct perf_file_section *self,
+						 struct perf_header *ph,
 						 int feat, int fd))
 {
 	struct perf_file_section *feat_sec;
 	int nr_sections;
 	int sec_size;
 	int idx = 0;
-	int err = 0, feat = 1;
+	int err = -1, feat = 1;
 
 	nr_sections = bitmap_weight(self->adds_features, HEADER_FEAT_BITS);
 	if (!nr_sections)
@@ -397,33 +568,45 @@ int perf_header__process_sections(struct perf_header *self, int fd,
 
 	lseek(fd, self->data_offset + self->data_size, SEEK_SET);
 
-	do_read(fd, feat_sec, sec_size);
+	if (perf_header__getbuffer64(self, fd, feat_sec, sec_size))
+		goto out_free;
 
+	err = 0;
 	while (idx < nr_sections && feat < HEADER_LAST_FEATURE) {
 		if (perf_header__has_feat(self, feat)) {
 			struct perf_file_section *sec = &feat_sec[idx++];
 
-			err = process(sec, feat, fd);
+			err = process(sec, self, feat, fd);
 			if (err < 0)
 				break;
 		}
 		++feat;
 	}
-
+out_free:
 	free(feat_sec);
 	return err;
-};
+}
 
 int perf_file_header__read(struct perf_file_header *self,
 			   struct perf_header *ph, int fd)
 {
 	lseek(fd, 0, SEEK_SET);
-	do_read(fd, self, sizeof(*self));
 
-	if (self->magic     != PERF_MAGIC ||
-	    self->attr_size != sizeof(struct perf_file_attr))
+	if (do_read(fd, self, sizeof(*self)) ||
+	    memcmp(&self->magic, __perf_magic, sizeof(self->magic)))
 		return -1;
 
+	if (self->attr_size != sizeof(struct perf_file_attr)) {
+		u64 attr_size = bswap_64(self->attr_size);
+
+		if (attr_size != sizeof(struct perf_file_attr))
+			return -1;
+
+		mem_bswap_64(self, offsetof(struct perf_file_header,
+					    adds_features));
+		ph->needs_swap = true;
+	}
+
 	if (self->size != sizeof(*self)) {
 		/* Support the previous format */
 		if (self->size == offsetof(typeof(*self), adds_features))
@@ -433,19 +616,31 @@ int perf_file_header__read(struct perf_file_header *self,
 	}
 
 	memcpy(&ph->adds_features, &self->adds_features,
-	       sizeof(self->adds_features));
+	       sizeof(ph->adds_features));
+	/*
+	 * FIXME: hack that assumes that if we need swap the perf.data file
+	 * may be coming from an arch with a different word-size, ergo different
+	 * DEFINE_BITMAP format, investigate more later, but for now its mostly
+	 * safe to assume that we have a build-id section. Trace files probably
+	 * have several other issues in this realm anyway...
+	 */
+	if (ph->needs_swap) {
+		memset(&ph->adds_features, 0, sizeof(ph->adds_features));
+		perf_header__set_feat(ph, HEADER_BUILD_ID);
+	}
 
 	ph->event_offset = self->event_types.offset;
-	ph->event_size	 = self->event_types.size;
-	ph->data_offset	 = self->data.offset;
+	ph->event_size   = self->event_types.size;
+	ph->data_offset  = self->data.offset;
 	ph->data_size	 = self->data.size;
 	return 0;
 }
 
 static int perf_file_section__process(struct perf_file_section *self,
+				      struct perf_header *ph,
 				      int feat, int fd)
 {
-	if (lseek(fd, self->offset, SEEK_SET) < 0) {
+	if (lseek(fd, self->offset, SEEK_SET) == (off_t)-1) {
 		pr_debug("Failed to lseek to %Ld offset for feature %d, "
 			 "continuing...\n", self->offset, feat);
 		return 0;
@@ -457,7 +652,7 @@ static int perf_file_section__process(struct perf_file_section *self,
 		break;
 
 	case HEADER_BUILD_ID:
-		if (perf_header__read_build_ids(fd, self->offset, self->size))
+		if (perf_header__read_build_ids(ph, fd, self->offset, self->size))
 			pr_debug("Failed to read buildids, continuing...\n");
 		break;
 	default:
@@ -469,7 +664,7 @@ static int perf_file_section__process(struct perf_file_section *self,
 
 int perf_header__read(struct perf_header *self, int fd)
 {
-	struct perf_file_header f_header;
+	struct perf_file_header	f_header;
 	struct perf_file_attr	f_attr;
 	u64			f_id;
 	int nr_attrs, nr_ids, i, j;
@@ -486,7 +681,9 @@ int perf_header__read(struct perf_header *self, int fd)
 		struct perf_header_attr *attr;
 		off_t tmp;
 
-		do_read(fd, &f_attr, sizeof(f_attr));
+		if (perf_header__getbuffer64(self, fd, &f_attr, sizeof(f_attr)))
+			goto out_errno;
+
 		tmp = lseek(fd, 0, SEEK_CUR);
 
 		attr = perf_header_attr__new(&f_attr.attr);
@@ -497,7 +694,8 @@ int perf_header__read(struct perf_header *self, int fd)
 		lseek(fd, f_attr.ids.offset, SEEK_SET);
 
 		for (j = 0; j < nr_ids; j++) {
-			do_read(fd, &f_id, sizeof(f_id));
+			if (perf_header__getbuffer64(self, fd, &f_id, sizeof(f_id)))
+				goto out_errno;
 
 			if (perf_header_attr__add_id(attr, f_id) < 0) {
 				perf_header_attr__delete(attr);
@@ -517,7 +715,9 @@ int perf_header__read(struct perf_header *self, int fd)
 		events = malloc(f_header.event_types.size);
 		if (events == NULL)
 			return -ENOMEM;
-		do_read(fd, events, f_header.event_types.size);
+		if (perf_header__getbuffer64(self, fd, events,
+					     f_header.event_types.size))
+			goto out_errno;
 		event_count =  f_header.event_types.size / sizeof(struct perf_trace_event_type);
 	}
 
@@ -527,6 +727,8 @@ int perf_header__read(struct perf_header *self, int fd)
 
 	self->frozen = 1;
 	return 0;
+out_errno:
+	return -errno;
 }
 
 u64 perf_header__sample_type(struct perf_header *header)
diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h
index d118d05d3ab..82a6af72d4c 100644
--- a/tools/perf/util/header.h
+++ b/tools/perf/util/header.h
@@ -5,6 +5,7 @@
 #include <sys/types.h>
 #include <stdbool.h>
 #include "types.h"
+#include "event.h"
 
 #include <linux/bitmap.h>
 
@@ -52,6 +53,7 @@ struct perf_header {
 	u64			data_size;
 	u64			event_offset;
 	u64			event_size;
+	bool			needs_swap;
 	DECLARE_BITMAP(adds_features, HEADER_FEAT_BITS);
 };
 
@@ -64,7 +66,7 @@ int perf_header__write(struct perf_header *self, int fd, bool at_exit);
 int perf_header__add_attr(struct perf_header *self,
 			  struct perf_header_attr *attr);
 
-void perf_header__push_event(u64 id, const char *name);
+int 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_event_attr *attr);
@@ -80,6 +82,11 @@ bool perf_header__has_feat(const struct perf_header *self, int feat);
 
 int perf_header__process_sections(struct perf_header *self, int fd,
 				  int (*process)(struct perf_file_section *self,
+						 struct perf_header *ph,
 						 int feat, int fd));
 
+int build_id_cache__add_s(const char *sbuild_id, const char *debugdir,
+			  const char *name, bool is_kallsyms);
+int build_id_cache__remove_s(const char *sbuild_id, const char *debugdir);
+
 #endif /* __PERF_HEADER_H */
diff --git a/tools/perf/util/include/linux/hash.h b/tools/perf/util/include/linux/hash.h
new file mode 100644
index 00000000000..201f5739799
--- /dev/null
+++ b/tools/perf/util/include/linux/hash.h
@@ -0,0 +1,5 @@
+#include "../../../../include/linux/hash.h"
+
+#ifndef PERF_HASH_H
+#define PERF_HASH_H
+#endif
diff --git a/tools/perf/util/include/linux/kernel.h b/tools/perf/util/include/linux/kernel.h
index 21c0274c02f..f2611655ab5 100644
--- a/tools/perf/util/include/linux/kernel.h
+++ b/tools/perf/util/include/linux/kernel.h
@@ -101,5 +101,6 @@ simple_strtoul(const char *nptr, char **endptr, int base)
 	eprintf(n, pr_fmt(fmt), ##__VA_ARGS__)
 #define pr_debug2(fmt, ...) pr_debugN(2, pr_fmt(fmt), ##__VA_ARGS__)
 #define pr_debug3(fmt, ...) pr_debugN(3, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_debug4(fmt, ...) pr_debugN(4, pr_fmt(fmt), ##__VA_ARGS__)
 
 #endif
diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c
index c4d55a0da2e..e509cd59c67 100644
--- a/tools/perf/util/map.c
+++ b/tools/perf/util/map.c
@@ -5,6 +5,11 @@
 #include <stdio.h>
 #include "debug.h"
 
+const char *map_type__name[MAP__NR_TYPES] = {
+	[MAP__FUNCTION] = "Functions",
+	[MAP__VARIABLE] = "Variables",
+};
+
 static inline int is_anon_memory(const char *filename)
 {
 	return strcmp(filename, "//anon") == 0;
@@ -68,8 +73,13 @@ struct map *map__new(struct mmap_event *event, enum map_type type,
 		map__init(self, type, event->start, event->start + event->len,
 			  event->pgoff, dso);
 
-		if (self->dso == vdso || anon)
+		if (anon) {
+set_identity:
 			self->map_ip = self->unmap_ip = identity__map_ip;
+		} else if (strcmp(filename, "[vdso]") == 0) {
+			dso__set_loaded(dso, self->type);
+			goto set_identity;
+		}
 	}
 	return self;
 out_delete:
@@ -104,8 +114,7 @@ void map__fixup_end(struct map *self)
 
 #define DSO__DELETED "(deleted)"
 
-int map__load(struct map *self, struct perf_session *session,
-	      symbol_filter_t filter)
+int map__load(struct map *self, symbol_filter_t filter)
 {
 	const char *name = self->dso->long_name;
 	int nr;
@@ -113,7 +122,7 @@ int map__load(struct map *self, struct perf_session *session,
 	if (dso__loaded(self->dso, self->type))
 		return 0;
 
-	nr = dso__load(self->dso, self, session, filter);
+	nr = dso__load(self->dso, self, filter);
 	if (nr < 0) {
 		if (self->dso->has_build_id) {
 			char sbuild_id[BUILD_ID_SIZE * 2 + 1];
@@ -144,24 +153,29 @@ int map__load(struct map *self, struct perf_session *session,
 
 		return -1;
 	}
+	/*
+	 * Only applies to the kernel, as its symtabs aren't relative like the
+	 * module ones.
+	 */
+	if (self->dso->kernel)
+		map__reloc_vmlinux(self);
 
 	return 0;
 }
 
-struct symbol *map__find_symbol(struct map *self, struct perf_session *session,
-				u64 addr, symbol_filter_t filter)
+struct symbol *map__find_symbol(struct map *self, u64 addr,
+				symbol_filter_t filter)
 {
-	if (map__load(self, session, filter) < 0)
+	if (map__load(self, filter) < 0)
 		return NULL;
 
 	return dso__find_symbol(self->dso, self->type, addr);
 }
 
 struct symbol *map__find_symbol_by_name(struct map *self, const char *name,
-					struct perf_session *session,
 					symbol_filter_t filter)
 {
-	if (map__load(self, session, filter) < 0)
+	if (map__load(self, filter) < 0)
 		return NULL;
 
 	if (!dso__sorted_by_name(self->dso, self->type))
@@ -201,3 +215,23 @@ size_t map__fprintf(struct map *self, FILE *fp)
 	return fprintf(fp, " %Lx-%Lx %Lx %s\n",
 		       self->start, self->end, self->pgoff, self->dso->name);
 }
+
+/*
+ * objdump wants/reports absolute IPs for ET_EXEC, and RIPs for ET_DYN.
+ * map->dso->adjust_symbols==1 for ET_EXEC-like cases.
+ */
+u64 map__rip_2objdump(struct map *map, u64 rip)
+{
+	u64 addr = map->dso->adjust_symbols ?
+			map->unmap_ip(map, rip) :	/* RIP -> IP */
+			rip;
+	return addr;
+}
+
+u64 map__objdump_2ip(struct map *map, u64 addr)
+{
+	u64 ip = map->dso->adjust_symbols ?
+			addr :
+			map->unmap_ip(map, addr);	/* RIP -> IP */
+	return ip;
+}
diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h
new file mode 100644
index 00000000000..b756368076c
--- /dev/null
+++ b/tools/perf/util/map.h
@@ -0,0 +1,94 @@
+#ifndef __PERF_MAP_H
+#define __PERF_MAP_H
+
+#include <linux/compiler.h>
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include <linux/types.h>
+
+enum map_type {
+	MAP__FUNCTION = 0,
+	MAP__VARIABLE,
+};
+
+#define MAP__NR_TYPES (MAP__VARIABLE + 1)
+
+extern const char *map_type__name[MAP__NR_TYPES];
+
+struct dso;
+struct ref_reloc_sym;
+struct map_groups;
+
+struct map {
+	union {
+		struct rb_node	rb_node;
+		struct list_head node;
+	};
+	u64			start;
+	u64			end;
+	enum map_type		type;
+	u64			pgoff;
+
+	/* ip -> dso rip */
+	u64			(*map_ip)(struct map *, u64);
+	/* dso rip -> ip */
+	u64			(*unmap_ip)(struct map *, u64);
+
+	struct dso		*dso;
+};
+
+struct kmap {
+	struct ref_reloc_sym	*ref_reloc_sym;
+	struct map_groups	*kmaps;
+};
+
+static inline struct kmap *map__kmap(struct map *self)
+{
+	return (struct kmap *)(self + 1);
+}
+
+static inline u64 map__map_ip(struct map *map, u64 ip)
+{
+	return ip - map->start + map->pgoff;
+}
+
+static inline u64 map__unmap_ip(struct map *map, u64 ip)
+{
+	return ip + map->start - map->pgoff;
+}
+
+static inline u64 identity__map_ip(struct map *map __used, u64 ip)
+{
+	return ip;
+}
+
+
+/* rip/ip <-> addr suitable for passing to `objdump --start-address=` */
+u64 map__rip_2objdump(struct map *map, u64 rip);
+u64 map__objdump_2ip(struct map *map, u64 addr);
+
+struct symbol;
+struct mmap_event;
+
+typedef int (*symbol_filter_t)(struct map *map, struct symbol *sym);
+
+void map__init(struct map *self, enum map_type type,
+	       u64 start, u64 end, u64 pgoff, struct dso *dso);
+struct map *map__new(struct mmap_event *event, enum map_type,
+		     char *cwd, int cwdlen);
+void map__delete(struct map *self);
+struct map *map__clone(struct map *self);
+int map__overlap(struct map *l, struct map *r);
+size_t map__fprintf(struct map *self, FILE *fp);
+
+int map__load(struct map *self, symbol_filter_t filter);
+struct symbol *map__find_symbol(struct map *self,
+				u64 addr, symbol_filter_t filter);
+struct symbol *map__find_symbol_by_name(struct map *self, const char *name,
+					symbol_filter_t filter);
+void map__fixup_start(struct map *self);
+void map__fixup_end(struct map *self);
+
+void map__reloc_vmlinux(struct map *self);
+
+#endif /* __PERF_MAP_H */
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index e5bc0fb016b..05d0c5c2030 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -450,7 +450,8 @@ parse_single_tracepoint_event(char *sys_name,
 /* 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)
+parse_multiple_tracepoint_event(char *sys_name, const char *evt_exp,
+				char *flags)
 {
 	char evt_path[MAXPATHLEN];
 	struct dirent *evt_ent;
@@ -474,6 +475,9 @@ parse_subsystem_tracepoint_event(char *sys_name, char *flags)
 		    || !strcmp(evt_ent->d_name, "filter"))
 			continue;
 
+		if (!strglobmatch(evt_ent->d_name, evt_exp))
+			continue;
+
 		len = snprintf(event_opt, MAX_EVOPT_LEN, "%s:%s%s%s", sys_name,
 			       evt_ent->d_name, flags ? ":" : "",
 			       flags ?: "");
@@ -522,9 +526,10 @@ static enum event_result parse_tracepoint_event(const char **strp,
 	if (evt_length >= MAX_EVENT_LENGTH)
 		return EVT_FAILED;
 
-	if (!strcmp(evt_name, "*")) {
+	if (strpbrk(evt_name, "*?")) {
 		*strp = evt_name + evt_length;
-		return parse_subsystem_tracepoint_event(sys_name, flags);
+		return parse_multiple_tracepoint_event(sys_name, evt_name,
+						       flags);
 	} else
 		return parse_single_tracepoint_event(sys_name, evt_name,
 						     evt_length, flags,
@@ -753,11 +758,11 @@ modifier:
 	return ret;
 }
 
-static void store_event_type(const char *orgname)
+static int store_event_type(const char *orgname)
 {
 	char filename[PATH_MAX], *c;
 	FILE *file;
-	int id;
+	int id, n;
 
 	sprintf(filename, "%s/", debugfs_path);
 	strncat(filename, orgname, strlen(orgname));
@@ -769,11 +774,14 @@ static void store_event_type(const char *orgname)
 
 	file = fopen(filename, "r");
 	if (!file)
-		return;
-	if (fscanf(file, "%i", &id) < 1)
-		die("cannot store event ID");
+		return 0;
+	n = fscanf(file, "%i", &id);
 	fclose(file);
-	perf_header__push_event(id, orgname);
+	if (n < 1) {
+		pr_err("cannot store event ID\n");
+		return -EINVAL;
+	}
+	return perf_header__push_event(id, orgname);
 }
 
 int parse_events(const struct option *opt __used, const char *str, int unset __used)
@@ -782,7 +790,8 @@ int parse_events(const struct option *opt __used, const char *str, int unset __u
 	enum event_result ret;
 
 	if (strchr(str, ':'))
-		store_event_type(str);
+		if (store_event_type(str) < 0)
+			return -1;
 
 	for (;;) {
 		if (nr_counters == MAX_COUNTERS)
@@ -835,11 +844,12 @@ int parse_filter(const struct option *opt __used, const char *str,
 }
 
 static const char * const event_type_descriptors[] = {
-	"",
 	"Hardware event",
 	"Software event",
 	"Tracepoint event",
 	"Hardware cache event",
+	"Raw hardware event descriptor",
+	"Hardware breakpoint",
 };
 
 /*
@@ -872,7 +882,7 @@ static void print_tracepoint_events(void)
 			snprintf(evt_path, MAXPATHLEN, "%s:%s",
 				 sys_dirent.d_name, evt_dirent.d_name);
 			printf("  %-42s [%s]\n", evt_path,
-				event_type_descriptors[PERF_TYPE_TRACEPOINT+1]);
+				event_type_descriptors[PERF_TYPE_TRACEPOINT]);
 		}
 		closedir(evt_dir);
 	}
@@ -892,9 +902,7 @@ void print_events(void)
 	printf("List of pre-defined events (to be used in -e):\n");
 
 	for (i = 0; i < ARRAY_SIZE(event_symbols); i++, syms++) {
-		type = syms->type + 1;
-		if (type >= ARRAY_SIZE(event_type_descriptors))
-			type = 0;
+		type = syms->type;
 
 		if (type != prev_type)
 			printf("\n");
@@ -919,17 +927,19 @@ void print_events(void)
 			for (i = 0; i < PERF_COUNT_HW_CACHE_RESULT_MAX; i++) {
 				printf("  %-42s [%s]\n",
 					event_cache_name(type, op, i),
-					event_type_descriptors[4]);
+					event_type_descriptors[PERF_TYPE_HW_CACHE]);
 			}
 		}
 	}
 
 	printf("\n");
-	printf("  %-42s [raw hardware event descriptor]\n",
-		"rNNN");
+	printf("  %-42s [%s]\n",
+		"rNNN", event_type_descriptors[PERF_TYPE_RAW]);
 	printf("\n");
 
-	printf("  %-42s [hardware breakpoint]\n", "mem:<addr>[:access]");
+	printf("  %-42s [%s]\n",
+			"mem:<addr>[:access]",
+			event_type_descriptors[PERF_TYPE_BREAKPOINT]);
 	printf("\n");
 
 	print_tracepoint_events();
diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index 29465d44004..c971e81e9cb 100644
--- a/tools/perf/util/probe-event.c
+++ b/tools/perf/util/probe-event.c
@@ -37,6 +37,8 @@
 #include "string.h"
 #include "strlist.h"
 #include "debug.h"
+#include "cache.h"
+#include "color.h"
 #include "parse-events.h"  /* For debugfs_path */
 #include "probe-event.h"
 
@@ -62,6 +64,42 @@ static int e_snprintf(char *str, size_t size, const char *format, ...)
 	return ret;
 }
 
+void parse_line_range_desc(const char *arg, struct line_range *lr)
+{
+	const char *ptr;
+	char *tmp;
+	/*
+	 * <Syntax>
+	 * SRC:SLN[+NUM|-ELN]
+	 * FUNC[:SLN[+NUM|-ELN]]
+	 */
+	ptr = strchr(arg, ':');
+	if (ptr) {
+		lr->start = (unsigned int)strtoul(ptr + 1, &tmp, 0);
+		if (*tmp == '+')
+			lr->end = lr->start + (unsigned int)strtoul(tmp + 1,
+								    &tmp, 0);
+		else if (*tmp == '-')
+			lr->end = (unsigned int)strtoul(tmp + 1, &tmp, 0);
+		else
+			lr->end = 0;
+		pr_debug("Line range is %u to %u\n", lr->start, lr->end);
+		if (lr->end && lr->start > lr->end)
+			semantic_error("Start line must be smaller"
+				       " than end line.");
+		if (*tmp != '\0')
+			semantic_error("Tailing with invalid character '%d'.",
+				       *tmp);
+		tmp = strndup(arg, (ptr - arg));
+	} else
+		tmp = strdup(arg);
+
+	if (strchr(tmp, '.'))
+		lr->file = tmp;
+	else
+		lr->function = tmp;
+}
+
 /* Check the name is good for event/group */
 static bool check_event_name(const char *name)
 {
@@ -81,14 +119,14 @@ static void parse_perf_probe_probepoint(char *arg, struct probe_point *pp)
 	char c, nc = 0;
 	/*
 	 * <Syntax>
-	 * perf probe [EVENT=]SRC:LN
-	 * perf probe [EVENT=]FUNC[+OFFS|%return][@SRC]
+	 * perf probe [EVENT=]SRC[:LN|;PTN]
+	 * perf probe [EVENT=]FUNC[@SRC][+OFFS|%return|:LN|;PAT]
 	 *
 	 * TODO:Group name support
 	 */
 
-	ptr = strchr(arg, '=');
-	if (ptr) {	/* Event name */
+	ptr = strpbrk(arg, ";=@+%");
+	if (ptr && *ptr == '=') {	/* Event name */
 		*ptr = '\0';
 		tmp = ptr + 1;
 		ptr = strchr(arg, ':');
@@ -101,7 +139,7 @@ static void parse_perf_probe_probepoint(char *arg, struct probe_point *pp)
 		arg = tmp;
 	}
 
-	ptr = strpbrk(arg, ":+@%");
+	ptr = strpbrk(arg, ";:+@%");
 	if (ptr) {
 		nc = *ptr;
 		*ptr++ = '\0';
@@ -118,7 +156,11 @@ static void parse_perf_probe_probepoint(char *arg, struct probe_point *pp)
 	while (ptr) {
 		arg = ptr;
 		c = nc;
-		ptr = strpbrk(arg, ":+@%");
+		if (c == ';') {	/* Lazy pattern must be the last part */
+			pp->lazy_line = strdup(arg);
+			break;
+		}
+		ptr = strpbrk(arg, ";:+@%");
 		if (ptr) {
 			nc = *ptr;
 			*ptr++ = '\0';
@@ -127,13 +169,13 @@ static void parse_perf_probe_probepoint(char *arg, struct probe_point *pp)
 		case ':':	/* Line number */
 			pp->line = strtoul(arg, &tmp, 0);
 			if (*tmp != '\0')
-				semantic_error("There is non-digit charactor"
-						" in line number.");
+				semantic_error("There is non-digit char"
+					       " in line number.");
 			break;
 		case '+':	/* Byte offset from a symbol */
 			pp->offset = strtoul(arg, &tmp, 0);
 			if (*tmp != '\0')
-				semantic_error("There is non-digit charactor"
+				semantic_error("There is non-digit character"
 						" in offset.");
 			break;
 		case '@':	/* File name */
@@ -141,9 +183,6 @@ static void parse_perf_probe_probepoint(char *arg, struct probe_point *pp)
 				semantic_error("SRC@SRC is not allowed.");
 			pp->file = strdup(arg);
 			DIE_IF(pp->file == NULL);
-			if (ptr)
-				semantic_error("@SRC must be the last "
-					       "option.");
 			break;
 		case '%':	/* Probe places */
 			if (strcmp(arg, "return") == 0) {
@@ -158,11 +197,18 @@ static void parse_perf_probe_probepoint(char *arg, struct probe_point *pp)
 	}
 
 	/* Exclusion check */
+	if (pp->lazy_line && pp->line)
+		semantic_error("Lazy pattern can't be used with line number.");
+
+	if (pp->lazy_line && pp->offset)
+		semantic_error("Lazy pattern can't be used with offset.");
+
 	if (pp->line && pp->offset)
 		semantic_error("Offset can't be used with line number.");
 
-	if (!pp->line && pp->file && !pp->function)
-		semantic_error("File always requires line number.");
+	if (!pp->line && !pp->lazy_line && pp->file && !pp->function)
+		semantic_error("File always requires line number or "
+			       "lazy pattern.");
 
 	if (pp->offset && !pp->function)
 		semantic_error("Offset requires an entry function.");
@@ -170,11 +216,13 @@ static void parse_perf_probe_probepoint(char *arg, struct probe_point *pp)
 	if (pp->retprobe && !pp->function)
 		semantic_error("Return probe requires an entry function.");
 
-	if ((pp->offset || pp->line) && pp->retprobe)
-		semantic_error("Offset/Line can't be used with return probe.");
+	if ((pp->offset || pp->line || pp->lazy_line) && pp->retprobe)
+		semantic_error("Offset/Line/Lazy pattern can't be used with "
+			       "return probe.");
 
-	pr_debug("symbol:%s file:%s line:%d offset:%d, return:%d\n",
-		 pp->function, pp->file, pp->line, pp->offset, pp->retprobe);
+	pr_debug("symbol:%s file:%s line:%d offset:%d return:%d lazy:%s\n",
+		 pp->function, pp->file, pp->line, pp->offset, pp->retprobe,
+		 pp->lazy_line);
 }
 
 /* Parse perf-probe event definition */
@@ -272,6 +320,7 @@ int synthesize_perf_probe_point(struct probe_point *pp)
 	int ret;
 
 	pp->probes[0] = buf = zalloc(MAX_CMDLEN);
+	pp->found = 1;
 	if (!buf)
 		die("Failed to allocate memory by zalloc.");
 	if (pp->offset) {
@@ -294,6 +343,7 @@ int synthesize_perf_probe_point(struct probe_point *pp)
 error:
 		free(pp->probes[0]);
 		pp->probes[0] = NULL;
+		pp->found = 0;
 	}
 	return ret;
 }
@@ -368,7 +418,7 @@ static int open_kprobe_events(int flags, int mode)
 	if (ret < 0) {
 		if (errno == ENOENT)
 			die("kprobe_events file does not exist -"
-			    " please rebuild with CONFIG_KPROBE_TRACER.");
+			    " please rebuild with CONFIG_KPROBE_EVENT.");
 		else
 			die("Could not open kprobe_events file: %s",
 			    strerror(errno));
@@ -418,6 +468,8 @@ static void clear_probe_point(struct probe_point *pp)
 		free(pp->function);
 	if (pp->file)
 		free(pp->file);
+	if (pp->lazy_line)
+		free(pp->lazy_line);
 	for (i = 0; i < pp->nr_args; i++)
 		free(pp->args[i]);
 	if (pp->args)
@@ -455,6 +507,9 @@ void show_perf_probe_events(void)
 	struct strlist *rawlist;
 	struct str_node *ent;
 
+	setup_pager();
+
+	memset(&pp, 0, sizeof(pp));
 	fd = open_kprobe_events(O_RDONLY, 0);
 	rawlist = get_trace_kprobe_event_rawlist(fd);
 	close(fd);
@@ -675,3 +730,73 @@ void del_trace_kprobe_events(struct strlist *dellist)
 	close(fd);
 }
 
+#define LINEBUF_SIZE 256
+#define NR_ADDITIONAL_LINES 2
+
+static void show_one_line(FILE *fp, unsigned int l, bool skip, bool show_num)
+{
+	char buf[LINEBUF_SIZE];
+	const char *color = PERF_COLOR_BLUE;
+
+	if (fgets(buf, LINEBUF_SIZE, fp) == NULL)
+		goto error;
+	if (!skip) {
+		if (show_num)
+			fprintf(stdout, "%7u  %s", l, buf);
+		else
+			color_fprintf(stdout, color, "         %s", buf);
+	}
+
+	while (strlen(buf) == LINEBUF_SIZE - 1 &&
+	       buf[LINEBUF_SIZE - 2] != '\n') {
+		if (fgets(buf, LINEBUF_SIZE, fp) == NULL)
+			goto error;
+		if (!skip) {
+			if (show_num)
+				fprintf(stdout, "%s", buf);
+			else
+				color_fprintf(stdout, color, "%s", buf);
+		}
+	}
+	return;
+error:
+	if (feof(fp))
+		die("Source file is shorter than expected.");
+	else
+		die("File read error: %s", strerror(errno));
+}
+
+void show_line_range(struct line_range *lr)
+{
+	unsigned int l = 1;
+	struct line_node *ln;
+	FILE *fp;
+
+	setup_pager();
+
+	if (lr->function)
+		fprintf(stdout, "<%s:%d>\n", lr->function,
+			lr->start - lr->offset);
+	else
+		fprintf(stdout, "<%s:%d>\n", lr->file, lr->start);
+
+	fp = fopen(lr->path, "r");
+	if (fp == NULL)
+		die("Failed to open %s: %s", lr->path, strerror(errno));
+	/* Skip to starting line number */
+	while (l < lr->start)
+		show_one_line(fp, l++, true, false);
+
+	list_for_each_entry(ln, &lr->line_list, list) {
+		while (ln->line > l)
+			show_one_line(fp, (l++) - lr->offset, false, false);
+		show_one_line(fp, (l++) - lr->offset, false, true);
+	}
+
+	if (lr->end == INT_MAX)
+		lr->end = l + NR_ADDITIONAL_LINES;
+	while (l < lr->end && !feof(fp))
+		show_one_line(fp, (l++) - lr->offset, false, false);
+
+	fclose(fp);
+}
diff --git a/tools/perf/util/probe-event.h b/tools/perf/util/probe-event.h
index 7f1d499118c..711287d4bae 100644
--- a/tools/perf/util/probe-event.h
+++ b/tools/perf/util/probe-event.h
@@ -5,6 +5,7 @@
 #include "probe-finder.h"
 #include "strlist.h"
 
+extern void parse_line_range_desc(const char *arg, struct line_range *lr);
 extern void parse_perf_probe_event(const char *str, struct probe_point *pp,
 				   bool *need_dwarf);
 extern int synthesize_perf_probe_point(struct probe_point *pp);
@@ -15,6 +16,7 @@ extern void add_trace_kprobe_events(struct probe_point *probes, int nr_probes,
 				    bool force_add);
 extern void del_trace_kprobe_events(struct strlist *dellist);
 extern void show_perf_probe_events(void);
+extern void show_line_range(struct line_range *lr);
 
 /* Maximum index number of event-name postfix */
 #define MAX_EVENT_INDEX	1024
diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
index 4b852c0d16a..e77dc886760 100644
--- a/tools/perf/util/probe-finder.c
+++ b/tools/perf/util/probe-finder.c
@@ -32,21 +32,13 @@
 #include <stdarg.h>
 #include <ctype.h>
 
+#include "string.h"
 #include "event.h"
 #include "debug.h"
 #include "util.h"
 #include "probe-finder.h"
 
 
-/* Dwarf_Die Linkage to parent Die */
-struct die_link {
-	struct die_link *parent;	/* Parent die */
-	Dwarf_Die die;			/* Current die */
-};
-
-static Dwarf_Debug __dw_debug;
-static Dwarf_Error __dw_error;
-
 /*
  * Generic dwarf analysis helpers
  */
@@ -113,256 +105,190 @@ static int strtailcmp(const char *s1, const char *s2)
 	return 0;
 }
 
-/* Find the fileno of the target file. */
-static Dwarf_Unsigned cu_find_fileno(Dwarf_Die cu_die, const char *fname)
-{
-	Dwarf_Signed cnt, i;
-	Dwarf_Unsigned found = 0;
-	char **srcs;
-	int ret;
-
-	if (!fname)
-		return 0;
+/* Line number list operations */
 
-	ret = dwarf_srcfiles(cu_die, &srcs, &cnt, &__dw_error);
-	if (ret == DW_DLV_OK) {
-		for (i = 0; i < cnt && !found; i++) {
-			if (strtailcmp(srcs[i], fname) == 0)
-				found = i + 1;
-			dwarf_dealloc(__dw_debug, srcs[i], DW_DLA_STRING);
-		}
-		for (; i < cnt; i++)
-			dwarf_dealloc(__dw_debug, srcs[i], DW_DLA_STRING);
-		dwarf_dealloc(__dw_debug, srcs, DW_DLA_LIST);
+/* Add a line to line number list */
+static void line_list__add_line(struct list_head *head, unsigned int line)
+{
+	struct line_node *ln;
+	struct list_head *p;
+
+	/* Reverse search, because new line will be the last one */
+	list_for_each_entry_reverse(ln, head, list) {
+		if (ln->line < line) {
+			p = &ln->list;
+			goto found;
+		} else if (ln->line == line)	/* Already exist */
+			return ;
 	}
-	if (found)
-		pr_debug("found fno: %d\n", (int)found);
-	return found;
+	/* List is empty, or the smallest entry */
+	p = head;
+found:
+	pr_debug("line list: add a line %u\n", line);
+	ln = zalloc(sizeof(struct line_node));
+	DIE_IF(ln == NULL);
+	ln->line = line;
+	INIT_LIST_HEAD(&ln->list);
+	list_add(&ln->list, p);
 }
 
-/* Compare diename and tname */
-static int die_compare_name(Dwarf_Die dw_die, const char *tname)
+/* Check if the line in line number list */
+static int line_list__has_line(struct list_head *head, unsigned int line)
 {
-	char *name;
-	int ret;
-	ret = dwarf_diename(dw_die, &name, &__dw_error);
-	DIE_IF(ret == DW_DLV_ERROR);
-	if (ret == DW_DLV_OK) {
-		ret = strcmp(tname, name);
-		dwarf_dealloc(__dw_debug, name, DW_DLA_STRING);
-	} else
-		ret = -1;
-	return ret;
-}
+	struct line_node *ln;
 
-/* Check the address is in the subprogram(function). */
-static int die_within_subprogram(Dwarf_Die sp_die, Dwarf_Addr addr,
-				 Dwarf_Signed *offs)
-{
-	Dwarf_Addr lopc, hipc;
-	int ret;
+	/* Reverse search, because new line will be the last one */
+	list_for_each_entry(ln, head, list)
+		if (ln->line == line)
+			return 1;
 
-	/* TODO: check ranges */
-	ret = dwarf_lowpc(sp_die, &lopc, &__dw_error);
-	DIE_IF(ret == DW_DLV_ERROR);
-	if (ret == DW_DLV_NO_ENTRY)
-		return 0;
-	ret = dwarf_highpc(sp_die, &hipc, &__dw_error);
-	DIE_IF(ret != DW_DLV_OK);
-	if (lopc <= addr && addr < hipc) {
-		*offs = addr - lopc;
-		return 1;
-	} else
-		return 0;
+	return 0;
 }
 
-/* Check the die is inlined function */
-static Dwarf_Bool die_inlined_subprogram(Dwarf_Die dw_die)
+/* Init line number list */
+static void line_list__init(struct list_head *head)
 {
-	/* TODO: check strictly */
-	Dwarf_Bool inl;
-	int ret;
+	INIT_LIST_HEAD(head);
+}
 
-	ret = dwarf_hasattr(dw_die, DW_AT_inline, &inl, &__dw_error);
-	DIE_IF(ret == DW_DLV_ERROR);
-	return inl;
+/* Free line number list */
+static void line_list__free(struct list_head *head)
+{
+	struct line_node *ln;
+	while (!list_empty(head)) {
+		ln = list_first_entry(head, struct line_node, list);
+		list_del(&ln->list);
+		free(ln);
+	}
 }
 
-/* Get the offset of abstruct_origin */
-static Dwarf_Off die_get_abstract_origin(Dwarf_Die dw_die)
+/* Dwarf wrappers */
+
+/* Find the realpath of the target file. */
+static const char *cu_find_realpath(Dwarf_Die *cu_die, const char *fname)
 {
-	Dwarf_Attribute attr;
-	Dwarf_Off cu_offs;
+	Dwarf_Files *files;
+	size_t nfiles, i;
+	const char *src;
 	int ret;
 
-	ret = dwarf_attr(dw_die, DW_AT_abstract_origin, &attr, &__dw_error);
-	DIE_IF(ret != DW_DLV_OK);
-	ret = dwarf_formref(attr, &cu_offs, &__dw_error);
-	DIE_IF(ret != DW_DLV_OK);
-	dwarf_dealloc(__dw_debug, attr, DW_DLA_ATTR);
-	return cu_offs;
+	if (!fname)
+		return NULL;
+
+	ret = dwarf_getsrcfiles(cu_die, &files, &nfiles);
+	if (ret != 0)
+		return NULL;
+
+	for (i = 0; i < nfiles; i++) {
+		src = dwarf_filesrc(files, i, NULL, NULL);
+		if (strtailcmp(src, fname) == 0)
+			break;
+	}
+	return src;
 }
 
-/* Get entry pc(or low pc, 1st entry of ranges)  of the die */
-static Dwarf_Addr die_get_entrypc(Dwarf_Die dw_die)
+struct __addr_die_search_param {
+	Dwarf_Addr	addr;
+	Dwarf_Die	*die_mem;
+};
+
+static int __die_search_func_cb(Dwarf_Die *fn_die, void *data)
 {
-	Dwarf_Attribute attr;
-	Dwarf_Addr addr;
-	Dwarf_Off offs;
-	Dwarf_Ranges *ranges;
-	Dwarf_Signed cnt;
-	int ret;
+	struct __addr_die_search_param *ad = data;
 
-	/* Try to get entry pc */
-	ret = dwarf_attr(dw_die, DW_AT_entry_pc, &attr, &__dw_error);
-	DIE_IF(ret == DW_DLV_ERROR);
-	if (ret == DW_DLV_OK) {
-		ret = dwarf_formaddr(attr, &addr, &__dw_error);
-		DIE_IF(ret != DW_DLV_OK);
-		dwarf_dealloc(__dw_debug, attr, DW_DLA_ATTR);
-		return addr;
+	if (dwarf_tag(fn_die) == DW_TAG_subprogram &&
+	    dwarf_haspc(fn_die, ad->addr)) {
+		memcpy(ad->die_mem, fn_die, sizeof(Dwarf_Die));
+		return DWARF_CB_ABORT;
 	}
+	return DWARF_CB_OK;
+}
 
-	/* Try to get low pc */
-	ret = dwarf_lowpc(dw_die, &addr, &__dw_error);
-	DIE_IF(ret == DW_DLV_ERROR);
-	if (ret == DW_DLV_OK)
-		return addr;
-
-	/* Try to get ranges */
-	ret = dwarf_attr(dw_die, DW_AT_ranges, &attr, &__dw_error);
-	DIE_IF(ret != DW_DLV_OK);
-	ret = dwarf_formref(attr, &offs, &__dw_error);
-	DIE_IF(ret != DW_DLV_OK);
-	ret = dwarf_get_ranges(__dw_debug, offs, &ranges, &cnt, NULL,
-				&__dw_error);
-	DIE_IF(ret != DW_DLV_OK);
-	addr = ranges[0].dwr_addr1;
-	dwarf_ranges_dealloc(__dw_debug, ranges, cnt);
-	return addr;
+/* Search a real subprogram including this line, */
+static Dwarf_Die *die_get_real_subprogram(Dwarf_Die *cu_die, Dwarf_Addr addr,
+					  Dwarf_Die *die_mem)
+{
+	struct __addr_die_search_param ad;
+	ad.addr = addr;
+	ad.die_mem = die_mem;
+	/* dwarf_getscopes can't find subprogram. */
+	if (!dwarf_getfuncs(cu_die, __die_search_func_cb, &ad, 0))
+		return NULL;
+	else
+		return die_mem;
 }
 
-/*
- * Search a Die from Die tree.
- * Note: cur_link->die should be deallocated in this function.
- */
-static int __search_die_tree(struct die_link *cur_link,
-			     int (*die_cb)(struct die_link *, void *),
-			     void *data)
+/* Similar to dwarf_getfuncs, but returns inlined_subroutine if exists. */
+static Dwarf_Die *die_get_inlinefunc(Dwarf_Die *sp_die, Dwarf_Addr addr,
+				     Dwarf_Die *die_mem)
 {
-	Dwarf_Die new_die;
-	struct die_link new_link;
+	Dwarf_Die child_die;
 	int ret;
 
-	if (!die_cb)
-		return 0;
-
-	/* Check current die */
-	while (!(ret = die_cb(cur_link, data))) {
-		/* Check child die */
-		ret = dwarf_child(cur_link->die, &new_die, &__dw_error);
-		DIE_IF(ret == DW_DLV_ERROR);
-		if (ret == DW_DLV_OK) {
-			new_link.parent = cur_link;
-			new_link.die = new_die;
-			ret = __search_die_tree(&new_link, die_cb, data);
-			if (ret)
-				break;
-		}
+	ret = dwarf_child(sp_die, die_mem);
+	if (ret != 0)
+		return NULL;
 
-		/* Move to next sibling */
-		ret = dwarf_siblingof(__dw_debug, cur_link->die, &new_die,
-				      &__dw_error);
-		DIE_IF(ret == DW_DLV_ERROR);
-		dwarf_dealloc(__dw_debug, cur_link->die, DW_DLA_DIE);
-		cur_link->die = new_die;
-		if (ret == DW_DLV_NO_ENTRY)
-			return 0;
-	}
-	dwarf_dealloc(__dw_debug, cur_link->die, DW_DLA_DIE);
-	return ret;
-}
+	do {
+		if (dwarf_tag(die_mem) == DW_TAG_inlined_subroutine &&
+		    dwarf_haspc(die_mem, addr))
+			return die_mem;
 
-/* Search a die in its children's die tree */
-static int search_die_from_children(Dwarf_Die parent_die,
-				    int (*die_cb)(struct die_link *, void *),
-				    void *data)
-{
-	struct die_link new_link;
-	int ret;
+		if (die_get_inlinefunc(die_mem, addr, &child_die)) {
+			memcpy(die_mem, &child_die, sizeof(Dwarf_Die));
+			return die_mem;
+		}
+	} while (dwarf_siblingof(die_mem, die_mem) == 0);
 
-	new_link.parent = NULL;
-	ret = dwarf_child(parent_die, &new_link.die, &__dw_error);
-	DIE_IF(ret == DW_DLV_ERROR);
-	if (ret == DW_DLV_OK)
-		return __search_die_tree(&new_link, die_cb, data);
-	else
-		return 0;
+	return NULL;
 }
 
-/* Find a locdesc corresponding to the address */
-static int attr_get_locdesc(Dwarf_Attribute attr, Dwarf_Locdesc *desc,
-			    Dwarf_Addr addr)
+/* Compare diename and tname */
+static bool die_compare_name(Dwarf_Die *dw_die, const char *tname)
 {
-	Dwarf_Signed lcnt;
-	Dwarf_Locdesc **llbuf;
-	int ret, i;
-
-	ret = dwarf_loclist_n(attr, &llbuf, &lcnt, &__dw_error);
-	DIE_IF(ret != DW_DLV_OK);
-	ret = DW_DLV_NO_ENTRY;
-	for (i = 0; i < lcnt; ++i) {
-		if (llbuf[i]->ld_lopc <= addr &&
-		    llbuf[i]->ld_hipc > addr) {
-			memcpy(desc, llbuf[i], sizeof(Dwarf_Locdesc));
-			desc->ld_s =
-				malloc(sizeof(Dwarf_Loc) * llbuf[i]->ld_cents);
-			DIE_IF(desc->ld_s == NULL);
-			memcpy(desc->ld_s, llbuf[i]->ld_s,
-				sizeof(Dwarf_Loc) * llbuf[i]->ld_cents);
-			ret = DW_DLV_OK;
-			break;
-		}
-		dwarf_dealloc(__dw_debug, llbuf[i]->ld_s, DW_DLA_LOC_BLOCK);
-		dwarf_dealloc(__dw_debug, llbuf[i], DW_DLA_LOCDESC);
-	}
-	/* Releasing loop */
-	for (; i < lcnt; ++i) {
-		dwarf_dealloc(__dw_debug, llbuf[i]->ld_s, DW_DLA_LOC_BLOCK);
-		dwarf_dealloc(__dw_debug, llbuf[i], DW_DLA_LOCDESC);
-	}
-	dwarf_dealloc(__dw_debug, llbuf, DW_DLA_LIST);
-	return ret;
+	const char *name;
+	name = dwarf_diename(dw_die);
+	DIE_IF(name == NULL);
+	return strcmp(tname, name);
 }
 
-/* Get decl_file attribute value (file number) */
-static Dwarf_Unsigned die_get_decl_file(Dwarf_Die sp_die)
+/* Get entry pc(or low pc, 1st entry of ranges)  of the die */
+static Dwarf_Addr die_get_entrypc(Dwarf_Die *dw_die)
 {
-	Dwarf_Attribute attr;
-	Dwarf_Unsigned fno;
+	Dwarf_Addr epc;
 	int ret;
 
-	ret = dwarf_attr(sp_die, DW_AT_decl_file, &attr, &__dw_error);
-	DIE_IF(ret != DW_DLV_OK);
-	dwarf_formudata(attr, &fno, &__dw_error);
-	DIE_IF(ret != DW_DLV_OK);
-	dwarf_dealloc(__dw_debug, attr, DW_DLA_ATTR);
-	return fno;
+	ret = dwarf_entrypc(dw_die, &epc);
+	DIE_IF(ret == -1);
+	return epc;
 }
 
-/* Get decl_line attribute value (line number) */
-static Dwarf_Unsigned die_get_decl_line(Dwarf_Die sp_die)
+/* Get a variable die */
+static Dwarf_Die *die_find_variable(Dwarf_Die *sp_die, const char *name,
+				    Dwarf_Die *die_mem)
 {
-	Dwarf_Attribute attr;
-	Dwarf_Unsigned lno;
+	Dwarf_Die child_die;
+	int tag;
 	int ret;
 
-	ret = dwarf_attr(sp_die, DW_AT_decl_line, &attr, &__dw_error);
-	DIE_IF(ret != DW_DLV_OK);
-	dwarf_formudata(attr, &lno, &__dw_error);
-	DIE_IF(ret != DW_DLV_OK);
-	dwarf_dealloc(__dw_debug, attr, DW_DLA_ATTR);
-	return lno;
+	ret = dwarf_child(sp_die, die_mem);
+	if (ret != 0)
+		return NULL;
+
+	do {
+		tag = dwarf_tag(die_mem);
+		if ((tag == DW_TAG_formal_parameter ||
+		     tag == DW_TAG_variable) &&
+		    (die_compare_name(die_mem, name) == 0))
+			return die_mem;
+
+		if (die_find_variable(die_mem, name, &child_die)) {
+			memcpy(die_mem, &child_die, sizeof(Dwarf_Die));
+			return die_mem;
+		}
+	} while (dwarf_siblingof(die_mem, die_mem) == 0);
+
+	return NULL;
 }
 
 /*
@@ -370,47 +296,45 @@ static Dwarf_Unsigned die_get_decl_line(Dwarf_Die sp_die)
  */
 
 /* Show a location */
-static void show_location(Dwarf_Loc *loc, struct probe_finder *pf)
+static void show_location(Dwarf_Op *op, struct probe_finder *pf)
 {
-	Dwarf_Small op;
-	Dwarf_Unsigned regn;
-	Dwarf_Signed offs;
+	unsigned int regn;
+	Dwarf_Word offs = 0;
 	int deref = 0, ret;
 	const char *regs;
 
-	op = loc->lr_atom;
-
+	/* TODO: support CFA */
 	/* If this is based on frame buffer, set the offset */
-	if (op == DW_OP_fbreg) {
+	if (op->atom == DW_OP_fbreg) {
+		if (pf->fb_ops == NULL)
+			die("The attribute of frame base is not supported.\n");
 		deref = 1;
-		offs = (Dwarf_Signed)loc->lr_number;
-		op = pf->fbloc.ld_s[0].lr_atom;
-		loc = &pf->fbloc.ld_s[0];
-	} else
-		offs = 0;
+		offs = op->number;
+		op = &pf->fb_ops[0];
+	}
 
-	if (op >= DW_OP_breg0 && op <= DW_OP_breg31) {
-		regn = op - DW_OP_breg0;
-		offs += (Dwarf_Signed)loc->lr_number;
+	if (op->atom >= DW_OP_breg0 && op->atom <= DW_OP_breg31) {
+		regn = op->atom - DW_OP_breg0;
+		offs += op->number;
 		deref = 1;
-	} else if (op >= DW_OP_reg0 && op <= DW_OP_reg31) {
-		regn = op - DW_OP_reg0;
-	} else if (op == DW_OP_bregx) {
-		regn = loc->lr_number;
-		offs += (Dwarf_Signed)loc->lr_number2;
+	} else if (op->atom >= DW_OP_reg0 && op->atom <= DW_OP_reg31) {
+		regn = op->atom - DW_OP_reg0;
+	} else if (op->atom == DW_OP_bregx) {
+		regn = op->number;
+		offs += op->number2;
 		deref = 1;
-	} else if (op == DW_OP_regx) {
-		regn = loc->lr_number;
+	} else if (op->atom == DW_OP_regx) {
+		regn = op->number;
 	} else
-		die("Dwarf_OP %d is not supported.\n", op);
+		die("DW_OP %d is not supported.", op->atom);
 
 	regs = get_arch_regstr(regn);
 	if (!regs)
-		die("%lld exceeds max register number.\n", regn);
+		die("%u exceeds max register number.", regn);
 
 	if (deref)
-		ret = snprintf(pf->buf, pf->len,
-				 " %s=%+lld(%s)", pf->var, offs, regs);
+		ret = snprintf(pf->buf, pf->len, " %s=+%ju(%s)",
+			       pf->var, (uintmax_t)offs, regs);
 	else
 		ret = snprintf(pf->buf, pf->len, " %s=%s", pf->var, regs);
 	DIE_IF(ret < 0);
@@ -418,52 +342,37 @@ static void show_location(Dwarf_Loc *loc, struct probe_finder *pf)
 }
 
 /* Show a variables in kprobe event format */
-static void show_variable(Dwarf_Die vr_die, struct probe_finder *pf)
+static void show_variable(Dwarf_Die *vr_die, struct probe_finder *pf)
 {
 	Dwarf_Attribute attr;
-	Dwarf_Locdesc ld;
+	Dwarf_Op *expr;
+	size_t nexpr;
 	int ret;
 
-	ret = dwarf_attr(vr_die, DW_AT_location, &attr, &__dw_error);
-	if (ret != DW_DLV_OK)
+	if (dwarf_attr(vr_die, DW_AT_location, &attr) == NULL)
 		goto error;
-	ret = attr_get_locdesc(attr, &ld, (pf->addr - pf->cu_base));
-	if (ret != DW_DLV_OK)
+	/* TODO: handle more than 1 exprs */
+	ret = dwarf_getlocation_addr(&attr, (pf->addr - pf->cu_base),
+				     &expr, &nexpr, 1);
+	if (ret <= 0 || nexpr == 0)
 		goto error;
-	/* TODO? */
-	DIE_IF(ld.ld_cents != 1);
-	show_location(&ld.ld_s[0], pf);
-	free(ld.ld_s);
-	dwarf_dealloc(__dw_debug, attr, DW_DLA_ATTR);
+
+	show_location(expr, pf);
+	/* *expr will be cached in libdw. Don't free it. */
 	return ;
 error:
+	/* TODO: Support const_value */
 	die("Failed to find the location of %s at this address.\n"
-	    " Perhaps, it has been optimized out.\n", pf->var);
-}
-
-static int variable_callback(struct die_link *dlink, void *data)
-{
-	struct probe_finder *pf = (struct probe_finder *)data;
-	Dwarf_Half tag;
-	int ret;
-
-	ret = dwarf_tag(dlink->die, &tag, &__dw_error);
-	DIE_IF(ret == DW_DLV_ERROR);
-	if ((tag == DW_TAG_formal_parameter ||
-	     tag == DW_TAG_variable) &&
-	    (die_compare_name(dlink->die, pf->var) == 0)) {
-		show_variable(dlink->die, pf);
-		return 1;
-	}
-	/* TODO: Support struct members and arrays */
-	return 0;
+	    " Perhaps, it has been optimized out.", pf->var);
 }
 
 /* Find a variable in a subprogram die */
-static void find_variable(Dwarf_Die sp_die, struct probe_finder *pf)
+static void find_variable(Dwarf_Die *sp_die, struct probe_finder *pf)
 {
 	int ret;
+	Dwarf_Die vr_die;
 
+	/* TODO: Support struct members and arrays */
 	if (!is_c_varname(pf->var)) {
 		/* Output raw parameters */
 		ret = snprintf(pf->buf, pf->len, " %s", pf->var);
@@ -474,58 +383,51 @@ static void find_variable(Dwarf_Die sp_die, struct probe_finder *pf)
 
 	pr_debug("Searching '%s' variable in context.\n", pf->var);
 	/* Search child die for local variables and parameters. */
-	ret = search_die_from_children(sp_die, variable_callback, pf);
-	if (!ret)
-		die("Failed to find '%s' in this function.\n", pf->var);
-}
+	if (!die_find_variable(sp_die, pf->var, &vr_die))
+		die("Failed to find '%s' in this function.", pf->var);
 
-/* Get a frame base on the address */
-static void get_current_frame_base(Dwarf_Die sp_die, struct probe_finder *pf)
-{
-	Dwarf_Attribute attr;
-	int ret;
-
-	ret = dwarf_attr(sp_die, DW_AT_frame_base, &attr, &__dw_error);
-	DIE_IF(ret != DW_DLV_OK);
-	ret = attr_get_locdesc(attr, &pf->fbloc, (pf->addr - pf->cu_base));
-	DIE_IF(ret != DW_DLV_OK);
-	dwarf_dealloc(__dw_debug, attr, DW_DLA_ATTR);
-}
-
-static void free_current_frame_base(struct probe_finder *pf)
-{
-	free(pf->fbloc.ld_s);
-	memset(&pf->fbloc, 0, sizeof(Dwarf_Locdesc));
+	show_variable(&vr_die, pf);
 }
 
 /* Show a probe point to output buffer */
-static void show_probepoint(Dwarf_Die sp_die, Dwarf_Signed offs,
-			    struct probe_finder *pf)
+static void show_probe_point(Dwarf_Die *sp_die, struct probe_finder *pf)
 {
 	struct probe_point *pp = pf->pp;
-	char *name;
+	Dwarf_Addr eaddr;
+	Dwarf_Die die_mem;
+	const char *name;
 	char tmp[MAX_PROBE_BUFFER];
 	int ret, i, len;
+	Dwarf_Attribute fb_attr;
+	size_t nops;
+
+	/* If no real subprogram, find a real one */
+	if (!sp_die || dwarf_tag(sp_die) != DW_TAG_subprogram) {
+		sp_die = die_get_real_subprogram(&pf->cu_die,
+						 pf->addr, &die_mem);
+		if (!sp_die)
+			die("Probe point is not found in subprograms.");
+	}
 
 	/* Output name of probe point */
-	ret = dwarf_diename(sp_die, &name, &__dw_error);
-	DIE_IF(ret == DW_DLV_ERROR);
-	if (ret == DW_DLV_OK) {
-		ret = snprintf(tmp, MAX_PROBE_BUFFER, "%s+%u", name,
-				(unsigned int)offs);
+	name = dwarf_diename(sp_die);
+	if (name) {
+		dwarf_entrypc(sp_die, &eaddr);
+		ret = snprintf(tmp, MAX_PROBE_BUFFER, "%s+%lu", name,
+				(unsigned long)(pf->addr - eaddr));
 		/* Copy the function name if possible */
 		if (!pp->function) {
 			pp->function = strdup(name);
-			pp->offset = offs;
+			pp->offset = (size_t)(pf->addr - eaddr);
 		}
-		dwarf_dealloc(__dw_debug, name, DW_DLA_STRING);
 	} else {
 		/* This function has no name. */
-		ret = snprintf(tmp, MAX_PROBE_BUFFER, "0x%llx", pf->addr);
+		ret = snprintf(tmp, MAX_PROBE_BUFFER, "0x%jx",
+			       (uintmax_t)pf->addr);
 		if (!pp->function) {
 			/* TODO: Use _stext */
 			pp->function = strdup("");
-			pp->offset = (int)pf->addr;
+			pp->offset = (size_t)pf->addr;
 		}
 	}
 	DIE_IF(ret < 0);
@@ -533,8 +435,15 @@ static void show_probepoint(Dwarf_Die sp_die, Dwarf_Signed offs,
 	len = ret;
 	pr_debug("Probe point found: %s\n", tmp);
 
+	/* Get the frame base attribute/ops */
+	dwarf_attr(sp_die, DW_AT_frame_base, &fb_attr);
+	ret = dwarf_getlocation_addr(&fb_attr, (pf->addr - pf->cu_base),
+				     &pf->fb_ops, &nops, 1);
+	if (ret <= 0 || nops == 0)
+		pf->fb_ops = NULL;
+
 	/* Find each argument */
-	get_current_frame_base(sp_die, pf);
+	/* TODO: use dwarf_cfi_addrframe */
 	for (i = 0; i < pp->nr_args; i++) {
 		pf->var = pp->args[i];
 		pf->buf = &tmp[len];
@@ -542,189 +451,383 @@ static void show_probepoint(Dwarf_Die sp_die, Dwarf_Signed offs,
 		find_variable(sp_die, pf);
 		len += strlen(pf->buf);
 	}
-	free_current_frame_base(pf);
+
+	/* *pf->fb_ops will be cached in libdw. Don't free it. */
+	pf->fb_ops = NULL;
 
 	pp->probes[pp->found] = strdup(tmp);
 	pp->found++;
 }
 
-static int probeaddr_callback(struct die_link *dlink, void *data)
+/* Find probe point from its line number */
+static void find_probe_point_by_line(struct probe_finder *pf)
 {
-	struct probe_finder *pf = (struct probe_finder *)data;
-	Dwarf_Half tag;
-	Dwarf_Signed offs;
+	Dwarf_Lines *lines;
+	Dwarf_Line *line;
+	size_t nlines, i;
+	Dwarf_Addr addr;
+	int lineno;
 	int ret;
 
-	ret = dwarf_tag(dlink->die, &tag, &__dw_error);
-	DIE_IF(ret == DW_DLV_ERROR);
-	/* Check the address is in this subprogram */
-	if (tag == DW_TAG_subprogram &&
-	    die_within_subprogram(dlink->die, pf->addr, &offs)) {
-		show_probepoint(dlink->die, offs, pf);
-		return 1;
+	ret = dwarf_getsrclines(&pf->cu_die, &lines, &nlines);
+	DIE_IF(ret != 0);
+
+	for (i = 0; i < nlines; i++) {
+		line = dwarf_onesrcline(lines, i);
+		dwarf_lineno(line, &lineno);
+		if (lineno != pf->lno)
+			continue;
+
+		/* TODO: Get fileno from line, but how? */
+		if (strtailcmp(dwarf_linesrc(line, NULL, NULL), pf->fname) != 0)
+			continue;
+
+		ret = dwarf_lineaddr(line, &addr);
+		DIE_IF(ret != 0);
+		pr_debug("Probe line found: line[%d]:%d addr:0x%jx\n",
+			 (int)i, lineno, (uintmax_t)addr);
+		pf->addr = addr;
+
+		show_probe_point(NULL, pf);
+		/* Continuing, because target line might be inlined. */
 	}
-	return 0;
 }
 
-/* Find probe point from its line number */
-static void find_by_line(struct probe_finder *pf)
+/* Find lines which match lazy pattern */
+static int find_lazy_match_lines(struct list_head *head,
+				 const char *fname, const char *pat)
+{
+	char *fbuf, *p1, *p2;
+	int fd, line, nlines = 0;
+	struct stat st;
+
+	fd = open(fname, O_RDONLY);
+	if (fd < 0)
+		die("failed to open %s", fname);
+	DIE_IF(fstat(fd, &st) < 0);
+	fbuf = malloc(st.st_size + 2);
+	DIE_IF(fbuf == NULL);
+	DIE_IF(read(fd, fbuf, st.st_size) < 0);
+	close(fd);
+	fbuf[st.st_size] = '\n';	/* Dummy line */
+	fbuf[st.st_size + 1] = '\0';
+	p1 = fbuf;
+	line = 1;
+	while ((p2 = strchr(p1, '\n')) != NULL) {
+		*p2 = '\0';
+		if (strlazymatch(p1, pat)) {
+			line_list__add_line(head, line);
+			nlines++;
+		}
+		line++;
+		p1 = p2 + 1;
+	}
+	free(fbuf);
+	return nlines;
+}
+
+/* Find probe points from lazy pattern  */
+static void find_probe_point_lazy(Dwarf_Die *sp_die, struct probe_finder *pf)
 {
-	Dwarf_Signed cnt, i, clm;
-	Dwarf_Line *lines;
-	Dwarf_Unsigned lineno = 0;
+	Dwarf_Lines *lines;
+	Dwarf_Line *line;
+	size_t nlines, i;
 	Dwarf_Addr addr;
-	Dwarf_Unsigned fno;
+	Dwarf_Die die_mem;
+	int lineno;
 	int ret;
 
-	ret = dwarf_srclines(pf->cu_die, &lines, &cnt, &__dw_error);
-	DIE_IF(ret != DW_DLV_OK);
+	if (list_empty(&pf->lcache)) {
+		/* Matching lazy line pattern */
+		ret = find_lazy_match_lines(&pf->lcache, pf->fname,
+					    pf->pp->lazy_line);
+		if (ret <= 0)
+			die("No matched lines found in %s.", pf->fname);
+	}
+
+	ret = dwarf_getsrclines(&pf->cu_die, &lines, &nlines);
+	DIE_IF(ret != 0);
+	for (i = 0; i < nlines; i++) {
+		line = dwarf_onesrcline(lines, i);
 
-	for (i = 0; i < cnt; i++) {
-		ret = dwarf_line_srcfileno(lines[i], &fno, &__dw_error);
-		DIE_IF(ret != DW_DLV_OK);
-		if (fno != pf->fno)
+		dwarf_lineno(line, &lineno);
+		if (!line_list__has_line(&pf->lcache, lineno))
 			continue;
 
-		ret = dwarf_lineno(lines[i], &lineno, &__dw_error);
-		DIE_IF(ret != DW_DLV_OK);
-		if (lineno != pf->lno)
+		/* TODO: Get fileno from line, but how? */
+		if (strtailcmp(dwarf_linesrc(line, NULL, NULL), pf->fname) != 0)
 			continue;
 
-		ret = dwarf_lineoff(lines[i], &clm, &__dw_error);
-		DIE_IF(ret != DW_DLV_OK);
+		ret = dwarf_lineaddr(line, &addr);
+		DIE_IF(ret != 0);
+		if (sp_die) {
+			/* Address filtering 1: does sp_die include addr? */
+			if (!dwarf_haspc(sp_die, addr))
+				continue;
+			/* Address filtering 2: No child include addr? */
+			if (die_get_inlinefunc(sp_die, addr, &die_mem))
+				continue;
+		}
 
-		ret = dwarf_lineaddr(lines[i], &addr, &__dw_error);
-		DIE_IF(ret != DW_DLV_OK);
-		pr_debug("Probe line found: line[%d]:%u,%d addr:0x%llx\n",
-			 (int)i, (unsigned)lineno, (int)clm, addr);
+		pr_debug("Probe line found: line[%d]:%d addr:0x%llx\n",
+			 (int)i, lineno, (unsigned long long)addr);
 		pf->addr = addr;
-		/* Search a real subprogram including this line, */
-		ret = search_die_from_children(pf->cu_die,
-					       probeaddr_callback, pf);
-		if (ret == 0)
-			die("Probe point is not found in subprograms.\n");
+
+		show_probe_point(sp_die, pf);
 		/* Continuing, because target line might be inlined. */
 	}
-	dwarf_srclines_dealloc(__dw_debug, lines, cnt);
+	/* TODO: deallocate lines, but how? */
+}
+
+static int probe_point_inline_cb(Dwarf_Die *in_die, void *data)
+{
+	struct probe_finder *pf = (struct probe_finder *)data;
+	struct probe_point *pp = pf->pp;
+
+	if (pp->lazy_line)
+		find_probe_point_lazy(in_die, pf);
+	else {
+		/* Get probe address */
+		pf->addr = die_get_entrypc(in_die);
+		pf->addr += pp->offset;
+		pr_debug("found inline addr: 0x%jx\n",
+			 (uintmax_t)pf->addr);
+
+		show_probe_point(in_die, pf);
+	}
+
+	return DWARF_CB_OK;
 }
 
 /* Search function from function name */
-static int probefunc_callback(struct die_link *dlink, void *data)
+static int probe_point_search_cb(Dwarf_Die *sp_die, void *data)
 {
 	struct probe_finder *pf = (struct probe_finder *)data;
 	struct probe_point *pp = pf->pp;
-	struct die_link *lk;
-	Dwarf_Signed offs;
-	Dwarf_Half tag;
-	int ret;
 
-	ret = dwarf_tag(dlink->die, &tag, &__dw_error);
-	DIE_IF(ret == DW_DLV_ERROR);
-	if (tag == DW_TAG_subprogram) {
-		if (die_compare_name(dlink->die, pp->function) == 0) {
-			if (pp->line) {	/* Function relative line */
-				pf->fno = die_get_decl_file(dlink->die);
-				pf->lno = die_get_decl_line(dlink->die)
-					 + pp->line;
-				find_by_line(pf);
-				return 1;
-			}
-			if (die_inlined_subprogram(dlink->die)) {
-				/* Inlined function, save it. */
-				ret = dwarf_die_CU_offset(dlink->die,
-							  &pf->inl_offs,
-							  &__dw_error);
-				DIE_IF(ret != DW_DLV_OK);
-				pr_debug("inline definition offset %lld\n",
-					 pf->inl_offs);
-				return 0;	/* Continue to search */
-			}
-			/* Get probe address */
-			pf->addr = die_get_entrypc(dlink->die);
+	/* Check tag and diename */
+	if (dwarf_tag(sp_die) != DW_TAG_subprogram ||
+	    die_compare_name(sp_die, pp->function) != 0)
+		return 0;
+
+	pf->fname = dwarf_decl_file(sp_die);
+	if (pp->line) { /* Function relative line */
+		dwarf_decl_line(sp_die, &pf->lno);
+		pf->lno += pp->line;
+		find_probe_point_by_line(pf);
+	} else if (!dwarf_func_inline(sp_die)) {
+		/* Real function */
+		if (pp->lazy_line)
+			find_probe_point_lazy(sp_die, pf);
+		else {
+			pf->addr = die_get_entrypc(sp_die);
 			pf->addr += pp->offset;
 			/* TODO: Check the address in this function */
-			show_probepoint(dlink->die, pp->offset, pf);
-			return 1; /* Exit; no same symbol in this CU. */
-		}
-	} else if (tag == DW_TAG_inlined_subroutine && pf->inl_offs) {
-		if (die_get_abstract_origin(dlink->die) == pf->inl_offs) {
-			/* Get probe address */
-			pf->addr = die_get_entrypc(dlink->die);
-			pf->addr += pp->offset;
-			pr_debug("found inline addr: 0x%llx\n", pf->addr);
-			/* Inlined function. Get a real subprogram */
-			for (lk = dlink->parent; lk != NULL; lk = lk->parent) {
-				tag = 0;
-				dwarf_tag(lk->die, &tag, &__dw_error);
-				DIE_IF(ret == DW_DLV_ERROR);
-				if (tag == DW_TAG_subprogram &&
-				    !die_inlined_subprogram(lk->die))
-					goto found;
-			}
-			die("Failed to find real subprogram.\n");
-found:
-			/* Get offset from subprogram */
-			ret = die_within_subprogram(lk->die, pf->addr, &offs);
-			DIE_IF(!ret);
-			show_probepoint(lk->die, offs, pf);
-			/* Continue to search */
+			show_probe_point(sp_die, pf);
 		}
-	}
-	return 0;
+	} else
+		/* Inlined function: search instances */
+		dwarf_func_inline_instances(sp_die, probe_point_inline_cb, pf);
+
+	return 1; /* Exit; no same symbol in this CU. */
 }
 
-static void find_by_func(struct probe_finder *pf)
+static void find_probe_point_by_func(struct probe_finder *pf)
 {
-	search_die_from_children(pf->cu_die, probefunc_callback, pf);
+	dwarf_getfuncs(&pf->cu_die, probe_point_search_cb, pf, 0);
 }
 
 /* Find a probe point */
-int find_probepoint(int fd, struct probe_point *pp)
+int find_probe_point(int fd, struct probe_point *pp)
 {
-	Dwarf_Half addr_size = 0;
-	Dwarf_Unsigned next_cuh = 0;
-	int cu_number = 0, ret;
 	struct probe_finder pf = {.pp = pp};
+	int ret;
+	Dwarf_Off off, noff;
+	size_t cuhl;
+	Dwarf_Die *diep;
+	Dwarf *dbg;
 
-	ret = dwarf_init(fd, DW_DLC_READ, 0, 0, &__dw_debug, &__dw_error);
-	if (ret != DW_DLV_OK)
+	dbg = dwarf_begin(fd, DWARF_C_READ);
+	if (!dbg)
 		return -ENOENT;
 
 	pp->found = 0;
-	while (++cu_number) {
-		/* Search CU (Compilation Unit) */
-		ret = dwarf_next_cu_header(__dw_debug, NULL, NULL, NULL,
-			&addr_size, &next_cuh, &__dw_error);
-		DIE_IF(ret == DW_DLV_ERROR);
-		if (ret == DW_DLV_NO_ENTRY)
-			break;
-
+	off = 0;
+	line_list__init(&pf.lcache);
+	/* Loop on CUs (Compilation Unit) */
+	while (!dwarf_nextcu(dbg, off, &noff, &cuhl, NULL, NULL, NULL)) {
 		/* Get the DIE(Debugging Information Entry) of this CU */
-		ret = dwarf_siblingof(__dw_debug, 0, &pf.cu_die, &__dw_error);
-		DIE_IF(ret != DW_DLV_OK);
+		diep = dwarf_offdie(dbg, off + cuhl, &pf.cu_die);
+		if (!diep)
+			continue;
 
 		/* Check if target file is included. */
 		if (pp->file)
-			pf.fno = cu_find_fileno(pf.cu_die, pp->file);
+			pf.fname = cu_find_realpath(&pf.cu_die, pp->file);
+		else
+			pf.fname = NULL;
 
-		if (!pp->file || pf.fno) {
+		if (!pp->file || pf.fname) {
 			/* Save CU base address (for frame_base) */
-			ret = dwarf_lowpc(pf.cu_die, &pf.cu_base, &__dw_error);
-			DIE_IF(ret == DW_DLV_ERROR);
-			if (ret == DW_DLV_NO_ENTRY)
+			ret = dwarf_lowpc(&pf.cu_die, &pf.cu_base);
+			if (ret != 0)
 				pf.cu_base = 0;
 			if (pp->function)
-				find_by_func(&pf);
+				find_probe_point_by_func(&pf);
+			else if (pp->lazy_line)
+				find_probe_point_lazy(NULL, &pf);
 			else {
 				pf.lno = pp->line;
-				find_by_line(&pf);
+				find_probe_point_by_line(&pf);
 			}
 		}
-		dwarf_dealloc(__dw_debug, pf.cu_die, DW_DLA_DIE);
+		off = noff;
 	}
-	ret = dwarf_finish(__dw_debug, &__dw_error);
-	DIE_IF(ret != DW_DLV_OK);
+	line_list__free(&pf.lcache);
+	dwarf_end(dbg);
 
 	return pp->found;
 }
 
+/* Find line range from its line number */
+static void find_line_range_by_line(Dwarf_Die *sp_die, struct line_finder *lf)
+{
+	Dwarf_Lines *lines;
+	Dwarf_Line *line;
+	size_t nlines, i;
+	Dwarf_Addr addr;
+	int lineno;
+	int ret;
+	const char *src;
+	Dwarf_Die die_mem;
+
+	line_list__init(&lf->lr->line_list);
+	ret = dwarf_getsrclines(&lf->cu_die, &lines, &nlines);
+	DIE_IF(ret != 0);
+
+	for (i = 0; i < nlines; i++) {
+		line = dwarf_onesrcline(lines, i);
+		ret = dwarf_lineno(line, &lineno);
+		DIE_IF(ret != 0);
+		if (lf->lno_s > lineno || lf->lno_e < lineno)
+			continue;
+
+		if (sp_die) {
+			/* Address filtering 1: does sp_die include addr? */
+			ret = dwarf_lineaddr(line, &addr);
+			DIE_IF(ret != 0);
+			if (!dwarf_haspc(sp_die, addr))
+				continue;
+
+			/* Address filtering 2: No child include addr? */
+			if (die_get_inlinefunc(sp_die, addr, &die_mem))
+				continue;
+		}
+
+		/* TODO: Get fileno from line, but how? */
+		src = dwarf_linesrc(line, NULL, NULL);
+		if (strtailcmp(src, lf->fname) != 0)
+			continue;
+
+		/* Copy real path */
+		if (!lf->lr->path)
+			lf->lr->path = strdup(src);
+		line_list__add_line(&lf->lr->line_list, (unsigned int)lineno);
+	}
+	/* Update status */
+	if (!list_empty(&lf->lr->line_list))
+		lf->found = 1;
+	else {
+		free(lf->lr->path);
+		lf->lr->path = NULL;
+	}
+}
+
+static int line_range_inline_cb(Dwarf_Die *in_die, void *data)
+{
+	find_line_range_by_line(in_die, (struct line_finder *)data);
+	return DWARF_CB_ABORT;	/* No need to find other instances */
+}
+
+/* Search function from function name */
+static int line_range_search_cb(Dwarf_Die *sp_die, void *data)
+{
+	struct line_finder *lf = (struct line_finder *)data;
+	struct line_range *lr = lf->lr;
+
+	if (dwarf_tag(sp_die) == DW_TAG_subprogram &&
+	    die_compare_name(sp_die, lr->function) == 0) {
+		lf->fname = dwarf_decl_file(sp_die);
+		dwarf_decl_line(sp_die, &lr->offset);
+		pr_debug("fname: %s, lineno:%d\n", lf->fname, lr->offset);
+		lf->lno_s = lr->offset + lr->start;
+		if (!lr->end)
+			lf->lno_e = INT_MAX;
+		else
+			lf->lno_e = lr->offset + lr->end;
+		lr->start = lf->lno_s;
+		lr->end = lf->lno_e;
+		if (dwarf_func_inline(sp_die))
+			dwarf_func_inline_instances(sp_die,
+						    line_range_inline_cb, lf);
+		else
+			find_line_range_by_line(sp_die, lf);
+		return 1;
+	}
+	return 0;
+}
+
+static void find_line_range_by_func(struct line_finder *lf)
+{
+	dwarf_getfuncs(&lf->cu_die, line_range_search_cb, lf, 0);
+}
+
+int find_line_range(int fd, struct line_range *lr)
+{
+	struct line_finder lf = {.lr = lr, .found = 0};
+	int ret;
+	Dwarf_Off off = 0, noff;
+	size_t cuhl;
+	Dwarf_Die *diep;
+	Dwarf *dbg;
+
+	dbg = dwarf_begin(fd, DWARF_C_READ);
+	if (!dbg)
+		return -ENOENT;
+
+	/* Loop on CUs (Compilation Unit) */
+	while (!lf.found) {
+		ret = dwarf_nextcu(dbg, off, &noff, &cuhl, NULL, NULL, NULL);
+		if (ret != 0)
+			break;
+
+		/* Get the DIE(Debugging Information Entry) of this CU */
+		diep = dwarf_offdie(dbg, off + cuhl, &lf.cu_die);
+		if (!diep)
+			continue;
+
+		/* Check if target file is included. */
+		if (lr->file)
+			lf.fname = cu_find_realpath(&lf.cu_die, lr->file);
+		else
+			lf.fname = 0;
+
+		if (!lr->file || lf.fname) {
+			if (lr->function)
+				find_line_range_by_func(&lf);
+			else {
+				lf.lno_s = lr->start;
+				if (!lr->end)
+					lf.lno_e = INT_MAX;
+				else
+					lf.lno_e = lr->end;
+				find_line_range_by_line(NULL, &lf);
+			}
+		}
+		off = noff;
+	}
+	pr_debug("path: %lx\n", (unsigned long)lr->path);
+	dwarf_end(dbg);
+	return lf.found;
+}
+
diff --git a/tools/perf/util/probe-finder.h b/tools/perf/util/probe-finder.h
index a4086aaddb7..d1a651793ba 100644
--- a/tools/perf/util/probe-finder.h
+++ b/tools/perf/util/probe-finder.h
@@ -1,6 +1,9 @@
 #ifndef _PROBE_FINDER_H
 #define _PROBE_FINDER_H
 
+#include <stdbool.h>
+#include "util.h"
+
 #define MAX_PATH_LEN		 256
 #define MAX_PROBE_BUFFER	1024
 #define MAX_PROBES		 128
@@ -18,6 +21,7 @@ struct probe_point {
 	/* Inputs */
 	char			*file;			/* File name */
 	int			line;			/* Line number */
+	char			*lazy_line;		/* Lazy line pattern */
 
 	char			*function;		/* Function name */
 	int			offset;			/* Offset bytes */
@@ -32,34 +36,58 @@ struct probe_point {
 	char			*probes[MAX_PROBES];	/* Output buffers (will be allocated)*/
 };
 
-#ifndef NO_LIBDWARF
-extern int find_probepoint(int fd, struct probe_point *pp);
+/* Line number container */
+struct line_node {
+	struct list_head	list;
+	unsigned int		line;
+};
+
+/* Line range */
+struct line_range {
+	char			*file;			/* File name */
+	char			*function;		/* Function name */
+	unsigned int		start;			/* Start line number */
+	unsigned int		end;			/* End line number */
+	int			offset;			/* Start line offset */
+	char			*path;			/* Real path name */
+	struct list_head	line_list;		/* Visible lines */
+};
 
-/* Workaround for undefined _MIPS_SZLONG bug in libdwarf.h: */
-#ifndef _MIPS_SZLONG
-# define _MIPS_SZLONG		0
-#endif
+#ifndef NO_DWARF_SUPPORT
+extern int find_probe_point(int fd, struct probe_point *pp);
+extern int find_line_range(int fd, struct line_range *lr);
 
 #include <dwarf.h>
-#include <libdwarf.h>
+#include <libdw.h>
 
 struct probe_finder {
-	struct probe_point	*pp;			/* Target probe point */
+	struct probe_point	*pp;		/* Target probe point */
 
 	/* For function searching */
-	Dwarf_Addr		addr;			/* Address */
-	Dwarf_Unsigned		fno;			/* File number */
-	Dwarf_Unsigned		lno;			/* Line number */
-	Dwarf_Off		inl_offs;		/* Inline offset */
-	Dwarf_Die		cu_die;			/* Current CU */
+	Dwarf_Addr		addr;		/* Address */
+	const char		*fname;		/* File name */
+	int			lno;		/* Line number */
+	Dwarf_Die		cu_die;		/* Current CU */
 
 	/* For variable searching */
-	Dwarf_Addr		cu_base;		/* Current CU base address */
-	Dwarf_Locdesc		fbloc;			/* Location of Current Frame Base */
-	const char		*var;			/* Current variable name */
-	char			*buf;			/* Current output buffer */
-	int			len;			/* Length of output buffer */
+	Dwarf_Op		*fb_ops;	/* Frame base attribute */
+	Dwarf_Addr		cu_base;	/* Current CU base address */
+	const char		*var;		/* Current variable name */
+	char			*buf;		/* Current output buffer */
+	int			len;		/* Length of output buffer */
+	struct list_head	lcache;		/* Line cache for lazy match */
 };
-#endif /* NO_LIBDWARF */
+
+struct line_finder {
+	struct line_range	*lr;		/* Target line range */
+
+	const char		*fname;		/* File name */
+	int			lno_s;		/* Start line number */
+	int			lno_e;		/* End line number */
+	Dwarf_Die		cu_die;		/* Current CU */
+	int			found;
+};
+
+#endif /* NO_DWARF_SUPPORT */
 
 #endif /*_PROBE_FINDER_H */
diff --git a/tools/perf/util/trace-event-perl.c b/tools/perf/util/scripting-engines/trace-event-perl.c
index 6d6d76b8a21..5376378e0cf 100644
--- a/tools/perf/util/trace-event-perl.c
+++ b/tools/perf/util/scripting-engines/trace-event-perl.c
@@ -25,10 +25,16 @@
 #include <ctype.h>
 #include <errno.h>
 
-#include "../perf.h"
-#include "util.h"
-#include "trace-event.h"
-#include "trace-event-perl.h"
+#include "../../perf.h"
+#include "../util.h"
+#include "../trace-event.h"
+
+#include <EXTERN.h>
+#include <perl.h>
+
+void boot_Perf__Trace__Context(pTHX_ CV *cv);
+void boot_DynaLoader(pTHX_ CV *cv);
+typedef PerlInterpreter * INTERP;
 
 void xs_init(pTHX);
 
@@ -49,7 +55,7 @@ INTERP my_perl;
 
 struct event *events[FTRACE_MAX_EVENT];
 
-static struct scripting_context *scripting_context;
+extern struct scripting_context *scripting_context;
 
 static char *cur_field_name;
 static int zero_flag_atom;
@@ -239,33 +245,6 @@ static inline struct event *find_cache_event(int type)
 	return event;
 }
 
-int common_pc(struct scripting_context *context)
-{
-	int pc;
-
-	pc = parse_common_pc(context->event_data);
-
-	return pc;
-}
-
-int common_flags(struct scripting_context *context)
-{
-	int flags;
-
-	flags = parse_common_flags(context->event_data);
-
-	return flags;
-}
-
-int common_lock_depth(struct scripting_context *context)
-{
-	int lock_depth;
-
-	lock_depth = parse_common_lock_depth(context->event_data);
-
-	return lock_depth;
-}
-
 static void perl_process_event(int cpu, void *data,
 			       int size __unused,
 			       unsigned long long nsecs, char *comm)
@@ -587,75 +566,3 @@ struct scripting_ops perl_scripting_ops = {
 	.process_event = perl_process_event,
 	.generate_script = perl_generate_script,
 };
-
-static void print_unsupported_msg(void)
-{
-	fprintf(stderr, "Perl scripting not supported."
-		"  Install libperl and rebuild perf to enable it.\n"
-		"For example:\n  # apt-get install libperl-dev (ubuntu)"
-		"\n  # yum install perl-ExtUtils-Embed (Fedora)"
-		"\n  etc.\n");
-}
-
-static int perl_start_script_unsupported(const char *script __unused,
-					 int argc __unused,
-					 const char **argv __unused)
-{
-	print_unsupported_msg();
-
-	return -1;
-}
-
-static int perl_stop_script_unsupported(void)
-{
-	return 0;
-}
-
-static void perl_process_event_unsupported(int cpu __unused,
-					   void *data __unused,
-					   int size __unused,
-					   unsigned long long nsecs __unused,
-					   char *comm __unused)
-{
-}
-
-static int perl_generate_script_unsupported(const char *outfile __unused)
-{
-	print_unsupported_msg();
-
-	return -1;
-}
-
-struct scripting_ops perl_scripting_unsupported_ops = {
-	.name = "Perl",
-	.start_script = perl_start_script_unsupported,
-	.stop_script = perl_stop_script_unsupported,
-	.process_event = perl_process_event_unsupported,
-	.generate_script = perl_generate_script_unsupported,
-};
-
-static void register_perl_scripting(struct scripting_ops *scripting_ops)
-{
-	int err;
-	err = script_spec_register("Perl", scripting_ops);
-	if (err)
-		die("error registering Perl script extension");
-
-	err = script_spec_register("pl", scripting_ops);
-	if (err)
-		die("error registering pl script extension");
-
-	scripting_context = malloc(sizeof(struct scripting_context));
-}
-
-#ifdef NO_LIBPERL
-void setup_perl_scripting(void)
-{
-	register_perl_scripting(&perl_scripting_unsupported_ops);
-}
-#else
-void setup_perl_scripting(void)
-{
-	register_perl_scripting(&perl_scripting_ops);
-}
-#endif
diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c
new file mode 100644
index 00000000000..33a414bbba3
--- /dev/null
+++ b/tools/perf/util/scripting-engines/trace-event-python.c
@@ -0,0 +1,573 @@
+/*
+ * trace-event-python.  Feed trace events to an embedded Python interpreter.
+ *
+ * Copyright (C) 2010 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <Python.h>
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+#include <errno.h>
+
+#include "../../perf.h"
+#include "../util.h"
+#include "../trace-event.h"
+
+PyMODINIT_FUNC initperf_trace_context(void);
+
+#define FTRACE_MAX_EVENT				\
+	((1 << (sizeof(unsigned short) * 8)) - 1)
+
+struct event *events[FTRACE_MAX_EVENT];
+
+#define MAX_FIELDS	64
+#define N_COMMON_FIELDS	7
+
+extern struct scripting_context *scripting_context;
+
+static char *cur_field_name;
+static int zero_flag_atom;
+
+static PyObject *main_module, *main_dict;
+
+static void handler_call_die(const char *handler_name)
+{
+	PyErr_Print();
+	Py_FatalError("problem in Python trace event handler");
+}
+
+static void define_value(enum print_arg_type field_type,
+			 const char *ev_name,
+			 const char *field_name,
+			 const char *field_value,
+			 const char *field_str)
+{
+	const char *handler_name = "define_flag_value";
+	PyObject *handler, *t, *retval;
+	unsigned long long value;
+	unsigned n = 0;
+
+	if (field_type == PRINT_SYMBOL)
+		handler_name = "define_symbolic_value";
+
+	t = PyTuple_New(4);
+	if (!t)
+		Py_FatalError("couldn't create Python tuple");
+
+	value = eval_flag(field_value);
+
+	PyTuple_SetItem(t, n++, PyString_FromString(ev_name));
+	PyTuple_SetItem(t, n++, PyString_FromString(field_name));
+	PyTuple_SetItem(t, n++, PyInt_FromLong(value));
+	PyTuple_SetItem(t, n++, PyString_FromString(field_str));
+
+	handler = PyDict_GetItemString(main_dict, handler_name);
+	if (handler && PyCallable_Check(handler)) {
+		retval = PyObject_CallObject(handler, t);
+		if (retval == NULL)
+			handler_call_die(handler_name);
+	}
+
+	Py_DECREF(t);
+}
+
+static void define_values(enum print_arg_type field_type,
+			  struct print_flag_sym *field,
+			  const char *ev_name,
+			  const char *field_name)
+{
+	define_value(field_type, ev_name, field_name, field->value,
+		     field->str);
+
+	if (field->next)
+		define_values(field_type, field->next, ev_name, field_name);
+}
+
+static void define_field(enum print_arg_type field_type,
+			 const char *ev_name,
+			 const char *field_name,
+			 const char *delim)
+{
+	const char *handler_name = "define_flag_field";
+	PyObject *handler, *t, *retval;
+	unsigned n = 0;
+
+	if (field_type == PRINT_SYMBOL)
+		handler_name = "define_symbolic_field";
+
+	if (field_type == PRINT_FLAGS)
+		t = PyTuple_New(3);
+	else
+		t = PyTuple_New(2);
+	if (!t)
+		Py_FatalError("couldn't create Python tuple");
+
+	PyTuple_SetItem(t, n++, PyString_FromString(ev_name));
+	PyTuple_SetItem(t, n++, PyString_FromString(field_name));
+	if (field_type == PRINT_FLAGS)
+		PyTuple_SetItem(t, n++, PyString_FromString(delim));
+
+	handler = PyDict_GetItemString(main_dict, handler_name);
+	if (handler && PyCallable_Check(handler)) {
+		retval = PyObject_CallObject(handler, t);
+		if (retval == NULL)
+			handler_call_die(handler_name);
+	}
+
+	Py_DECREF(t);
+}
+
+static void define_event_symbols(struct event *event,
+				 const char *ev_name,
+				 struct print_arg *args)
+{
+	switch (args->type) {
+	case PRINT_NULL:
+		break;
+	case PRINT_ATOM:
+		define_value(PRINT_FLAGS, ev_name, cur_field_name, "0",
+			     args->atom.atom);
+		zero_flag_atom = 0;
+		break;
+	case PRINT_FIELD:
+		if (cur_field_name)
+			free(cur_field_name);
+		cur_field_name = strdup(args->field.name);
+		break;
+	case PRINT_FLAGS:
+		define_event_symbols(event, ev_name, args->flags.field);
+		define_field(PRINT_FLAGS, ev_name, cur_field_name,
+			     args->flags.delim);
+		define_values(PRINT_FLAGS, args->flags.flags, ev_name,
+			      cur_field_name);
+		break;
+	case PRINT_SYMBOL:
+		define_event_symbols(event, ev_name, args->symbol.field);
+		define_field(PRINT_SYMBOL, ev_name, cur_field_name, NULL);
+		define_values(PRINT_SYMBOL, args->symbol.symbols, ev_name,
+			      cur_field_name);
+		break;
+	case PRINT_STRING:
+		break;
+	case PRINT_TYPE:
+		define_event_symbols(event, ev_name, args->typecast.item);
+		break;
+	case PRINT_OP:
+		if (strcmp(args->op.op, ":") == 0)
+			zero_flag_atom = 1;
+		define_event_symbols(event, ev_name, args->op.left);
+		define_event_symbols(event, ev_name, args->op.right);
+		break;
+	default:
+		/* we should warn... */
+		return;
+	}
+
+	if (args->next)
+		define_event_symbols(event, ev_name, args->next);
+}
+
+static inline struct event *find_cache_event(int type)
+{
+	static char ev_name[256];
+	struct event *event;
+
+	if (events[type])
+		return events[type];
+
+	events[type] = event = trace_find_event(type);
+	if (!event)
+		return NULL;
+
+	sprintf(ev_name, "%s__%s", event->system, event->name);
+
+	define_event_symbols(event, ev_name, event->print_fmt.args);
+
+	return event;
+}
+
+static void python_process_event(int cpu, void *data,
+				 int size __unused,
+				 unsigned long long nsecs, char *comm)
+{
+	PyObject *handler, *retval, *context, *t;
+	static char handler_name[256];
+	struct format_field *field;
+	unsigned long long val;
+	unsigned long s, ns;
+	struct event *event;
+	unsigned n = 0;
+	int type;
+	int pid;
+
+	t = PyTuple_New(MAX_FIELDS);
+	if (!t)
+		Py_FatalError("couldn't create Python tuple");
+
+	type = trace_parse_common_type(data);
+
+	event = find_cache_event(type);
+	if (!event)
+		die("ug! no event found for type %d", type);
+
+	pid = trace_parse_common_pid(data);
+
+	sprintf(handler_name, "%s__%s", event->system, event->name);
+
+	s = nsecs / NSECS_PER_SEC;
+	ns = nsecs - s * NSECS_PER_SEC;
+
+	scripting_context->event_data = data;
+
+	context = PyCObject_FromVoidPtr(scripting_context, NULL);
+
+	PyTuple_SetItem(t, n++, PyString_FromString(handler_name));
+	PyTuple_SetItem(t, n++,
+			PyCObject_FromVoidPtr(scripting_context, NULL));
+	PyTuple_SetItem(t, n++, PyInt_FromLong(cpu));
+	PyTuple_SetItem(t, n++, PyInt_FromLong(s));
+	PyTuple_SetItem(t, n++, PyInt_FromLong(ns));
+	PyTuple_SetItem(t, n++, PyInt_FromLong(pid));
+	PyTuple_SetItem(t, n++, PyString_FromString(comm));
+
+	for (field = event->format.fields; field; field = field->next) {
+		if (field->flags & FIELD_IS_STRING) {
+			int offset;
+			if (field->flags & FIELD_IS_DYNAMIC) {
+				offset = *(int *)(data + field->offset);
+				offset &= 0xffff;
+			} else
+				offset = field->offset;
+			PyTuple_SetItem(t, n++,
+				PyString_FromString((char *)data + offset));
+		} else { /* FIELD_IS_NUMERIC */
+			val = read_size(data + field->offset, field->size);
+			if (field->flags & FIELD_IS_SIGNED) {
+				PyTuple_SetItem(t, n++, PyInt_FromLong(val));
+			} else {
+				PyTuple_SetItem(t, n++, PyInt_FromLong(val));
+			}
+		}
+	}
+
+	if (_PyTuple_Resize(&t, n) == -1)
+		Py_FatalError("error resizing Python tuple");
+
+	handler = PyDict_GetItemString(main_dict, handler_name);
+	if (handler && PyCallable_Check(handler)) {
+		retval = PyObject_CallObject(handler, t);
+		if (retval == NULL)
+			handler_call_die(handler_name);
+	} else {
+		handler = PyDict_GetItemString(main_dict, "trace_unhandled");
+		if (handler && PyCallable_Check(handler)) {
+			if (_PyTuple_Resize(&t, N_COMMON_FIELDS) == -1)
+				Py_FatalError("error resizing Python tuple");
+
+			retval = PyObject_CallObject(handler, t);
+			if (retval == NULL)
+				handler_call_die("trace_unhandled");
+		}
+	}
+
+	Py_DECREF(t);
+}
+
+static int run_start_sub(void)
+{
+	PyObject *handler, *retval;
+	int err = 0;
+
+	main_module = PyImport_AddModule("__main__");
+	if (main_module == NULL)
+		return -1;
+	Py_INCREF(main_module);
+
+	main_dict = PyModule_GetDict(main_module);
+	if (main_dict == NULL) {
+		err = -1;
+		goto error;
+	}
+	Py_INCREF(main_dict);
+
+	handler = PyDict_GetItemString(main_dict, "trace_begin");
+	if (handler == NULL || !PyCallable_Check(handler))
+		goto out;
+
+	retval = PyObject_CallObject(handler, NULL);
+	if (retval == NULL)
+		handler_call_die("trace_begin");
+
+	Py_DECREF(retval);
+	return err;
+error:
+	Py_XDECREF(main_dict);
+	Py_XDECREF(main_module);
+out:
+	return err;
+}
+
+/*
+ * Start trace script
+ */
+static int python_start_script(const char *script, int argc, const char **argv)
+{
+	const char **command_line;
+	char buf[PATH_MAX];
+	int i, err = 0;
+	FILE *fp;
+
+	command_line = malloc((argc + 1) * sizeof(const char *));
+	command_line[0] = script;
+	for (i = 1; i < argc + 1; i++)
+		command_line[i] = argv[i - 1];
+
+	Py_Initialize();
+
+	initperf_trace_context();
+
+	PySys_SetArgv(argc + 1, (char **)command_line);
+
+	fp = fopen(script, "r");
+	if (!fp) {
+		sprintf(buf, "Can't open python script \"%s\"", script);
+		perror(buf);
+		err = -1;
+		goto error;
+	}
+
+	err = PyRun_SimpleFile(fp, script);
+	if (err) {
+		fprintf(stderr, "Error running python script %s\n", script);
+		goto error;
+	}
+
+	err = run_start_sub();
+	if (err) {
+		fprintf(stderr, "Error starting python script %s\n", script);
+		goto error;
+	}
+
+	free(command_line);
+	fprintf(stderr, "perf trace started with Python script %s\n\n",
+		script);
+
+	return err;
+error:
+	Py_Finalize();
+	free(command_line);
+
+	return err;
+}
+
+/*
+ * Stop trace script
+ */
+static int python_stop_script(void)
+{
+	PyObject *handler, *retval;
+	int err = 0;
+
+	handler = PyDict_GetItemString(main_dict, "trace_end");
+	if (handler == NULL || !PyCallable_Check(handler))
+		goto out;
+
+	retval = PyObject_CallObject(handler, NULL);
+	if (retval == NULL)
+		handler_call_die("trace_end");
+	else
+		Py_DECREF(retval);
+out:
+	Py_XDECREF(main_dict);
+	Py_XDECREF(main_module);
+	Py_Finalize();
+
+	fprintf(stderr, "\nperf trace Python script stopped\n");
+
+	return err;
+}
+
+static int python_generate_script(const char *outfile)
+{
+	struct event *event = NULL;
+	struct format_field *f;
+	char fname[PATH_MAX];
+	int not_first, count;
+	FILE *ofp;
+
+	sprintf(fname, "%s.py", outfile);
+	ofp = fopen(fname, "w");
+	if (ofp == NULL) {
+		fprintf(stderr, "couldn't open %s\n", fname);
+		return -1;
+	}
+	fprintf(ofp, "# perf trace event handlers, "
+		"generated by perf trace -g python\n");
+
+	fprintf(ofp, "# Licensed under the terms of the GNU GPL"
+		" License version 2\n\n");
+
+	fprintf(ofp, "# The common_* event handler fields are the most useful "
+		"fields common to\n");
+
+	fprintf(ofp, "# all events.  They don't necessarily correspond to "
+		"the 'common_*' fields\n");
+
+	fprintf(ofp, "# in the format files.  Those fields not available as "
+		"handler params can\n");
+
+	fprintf(ofp, "# be retrieved using Python functions of the form "
+		"common_*(context).\n");
+
+	fprintf(ofp, "# See the perf-trace-python Documentation for the list "
+		"of available functions.\n\n");
+
+	fprintf(ofp, "import os\n");
+	fprintf(ofp, "import sys\n\n");
+
+	fprintf(ofp, "sys.path.append(os.environ['PERF_EXEC_PATH'] + \\\n");
+	fprintf(ofp, "\t'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')\n");
+	fprintf(ofp, "\nfrom perf_trace_context import *\n");
+	fprintf(ofp, "from Core import *\n\n\n");
+
+	fprintf(ofp, "def trace_begin():\n");
+	fprintf(ofp, "\tprint \"in trace_begin\"\n\n");
+
+	fprintf(ofp, "def trace_end():\n");
+	fprintf(ofp, "\tprint \"in trace_end\"\n\n");
+
+	while ((event = trace_find_next_event(event))) {
+		fprintf(ofp, "def %s__%s(", event->system, event->name);
+		fprintf(ofp, "event_name, ");
+		fprintf(ofp, "context, ");
+		fprintf(ofp, "common_cpu,\n");
+		fprintf(ofp, "\tcommon_secs, ");
+		fprintf(ofp, "common_nsecs, ");
+		fprintf(ofp, "common_pid, ");
+		fprintf(ofp, "common_comm,\n\t");
+
+		not_first = 0;
+		count = 0;
+
+		for (f = event->format.fields; f; f = f->next) {
+			if (not_first++)
+				fprintf(ofp, ", ");
+			if (++count % 5 == 0)
+				fprintf(ofp, "\n\t");
+
+			fprintf(ofp, "%s", f->name);
+		}
+		fprintf(ofp, "):\n");
+
+		fprintf(ofp, "\t\tprint_header(event_name, common_cpu, "
+			"common_secs, common_nsecs,\n\t\t\t"
+			"common_pid, common_comm)\n\n");
+
+		fprintf(ofp, "\t\tprint \"");
+
+		not_first = 0;
+		count = 0;
+
+		for (f = event->format.fields; f; f = f->next) {
+			if (not_first++)
+				fprintf(ofp, ", ");
+			if (count && count % 3 == 0) {
+				fprintf(ofp, "\" \\\n\t\t\"");
+			}
+			count++;
+
+			fprintf(ofp, "%s=", f->name);
+			if (f->flags & FIELD_IS_STRING ||
+			    f->flags & FIELD_IS_FLAG ||
+			    f->flags & FIELD_IS_SYMBOLIC)
+				fprintf(ofp, "%%s");
+			else if (f->flags & FIELD_IS_SIGNED)
+				fprintf(ofp, "%%d");
+			else
+				fprintf(ofp, "%%u");
+		}
+
+		fprintf(ofp, "\\n\" %% \\\n\t\t(");
+
+		not_first = 0;
+		count = 0;
+
+		for (f = event->format.fields; f; f = f->next) {
+			if (not_first++)
+				fprintf(ofp, ", ");
+
+			if (++count % 5 == 0)
+				fprintf(ofp, "\n\t\t");
+
+			if (f->flags & FIELD_IS_FLAG) {
+				if ((count - 1) % 5 != 0) {
+					fprintf(ofp, "\n\t\t");
+					count = 4;
+				}
+				fprintf(ofp, "flag_str(\"");
+				fprintf(ofp, "%s__%s\", ", event->system,
+					event->name);
+				fprintf(ofp, "\"%s\", %s)", f->name,
+					f->name);
+			} else if (f->flags & FIELD_IS_SYMBOLIC) {
+				if ((count - 1) % 5 != 0) {
+					fprintf(ofp, "\n\t\t");
+					count = 4;
+				}
+				fprintf(ofp, "symbol_str(\"");
+				fprintf(ofp, "%s__%s\", ", event->system,
+					event->name);
+				fprintf(ofp, "\"%s\", %s)", f->name,
+					f->name);
+			} else
+				fprintf(ofp, "%s", f->name);
+		}
+
+		fprintf(ofp, "),\n\n");
+	}
+
+	fprintf(ofp, "def trace_unhandled(event_name, context, "
+		"common_cpu, common_secs, common_nsecs,\n\t\t"
+		"common_pid, common_comm):\n");
+
+	fprintf(ofp, "\t\tprint_header(event_name, common_cpu, "
+		"common_secs, common_nsecs,\n\t\tcommon_pid, "
+		"common_comm)\n\n");
+
+	fprintf(ofp, "def print_header("
+		"event_name, cpu, secs, nsecs, pid, comm):\n"
+		"\tprint \"%%-20s %%5u %%05u.%%09u %%8u %%-20s \" %% \\\n\t"
+		"(event_name, cpu, secs, nsecs, pid, comm),\n");
+
+	fclose(ofp);
+
+	fprintf(stderr, "generated Python script: %s\n", fname);
+
+	return 0;
+}
+
+struct scripting_ops python_scripting_ops = {
+	.name = "Python",
+	.start_script = python_start_script,
+	.stop_script = python_stop_script,
+	.process_event = python_process_event,
+	.generate_script = python_generate_script,
+};
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index ce3a6c8abe7..0de7258e70a 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -1,5 +1,8 @@
+#define _FILE_OFFSET_BITS 64
+
 #include <linux/kernel.h>
 
+#include <byteswap.h>
 #include <unistd.h>
 #include <sys/types.h>
 
@@ -49,6 +52,11 @@ out_close:
 	return -1;
 }
 
+static inline int perf_session__create_kernel_maps(struct perf_session *self)
+{
+	return map_groups__create_kernel_maps(&self->kmaps, self->vmlinux_maps);
+}
+
 struct perf_session *perf_session__new(const char *filename, int mode, bool force)
 {
 	size_t len = filename ? strlen(filename) + 1 : 0;
@@ -66,13 +74,22 @@ struct perf_session *perf_session__new(const char *filename, int mode, bool forc
 	self->mmap_window = 32;
 	self->cwd = NULL;
 	self->cwdlen = 0;
+	self->unknown_events = 0;
 	map_groups__init(&self->kmaps);
 
-	if (perf_session__create_kernel_maps(self) < 0)
-		goto out_delete;
+	if (mode == O_RDONLY) {
+		if (perf_session__open(self, force) < 0)
+			goto out_delete;
+	} else if (mode == O_WRONLY) {
+		/*
+		 * In O_RDONLY mode this will be performed when reading the
+		 * kernel MMAP event, in event__process_mmap().
+		 */
+		if (perf_session__create_kernel_maps(self) < 0)
+			goto out_delete;
+	}
 
-	if (mode == O_RDONLY && perf_session__open(self, force) < 0)
-		goto out_delete;
+	self->sample_type = perf_header__sample_type(&self->header);
 out:
 	return self;
 out_free:
@@ -148,3 +165,409 @@ struct symbol **perf_session__resolve_callchain(struct perf_session *self,
 
 	return syms;
 }
+
+static int process_event_stub(event_t *event __used,
+			      struct perf_session *session __used)
+{
+	dump_printf(": unhandled!\n");
+	return 0;
+}
+
+static void perf_event_ops__fill_defaults(struct perf_event_ops *handler)
+{
+	if (handler->sample == NULL)
+		handler->sample = process_event_stub;
+	if (handler->mmap == NULL)
+		handler->mmap = process_event_stub;
+	if (handler->comm == NULL)
+		handler->comm = process_event_stub;
+	if (handler->fork == NULL)
+		handler->fork = process_event_stub;
+	if (handler->exit == NULL)
+		handler->exit = process_event_stub;
+	if (handler->lost == NULL)
+		handler->lost = process_event_stub;
+	if (handler->read == NULL)
+		handler->read = process_event_stub;
+	if (handler->throttle == NULL)
+		handler->throttle = process_event_stub;
+	if (handler->unthrottle == NULL)
+		handler->unthrottle = process_event_stub;
+}
+
+static const char *event__name[] = {
+	[0]			 = "TOTAL",
+	[PERF_RECORD_MMAP]	 = "MMAP",
+	[PERF_RECORD_LOST]	 = "LOST",
+	[PERF_RECORD_COMM]	 = "COMM",
+	[PERF_RECORD_EXIT]	 = "EXIT",
+	[PERF_RECORD_THROTTLE]	 = "THROTTLE",
+	[PERF_RECORD_UNTHROTTLE] = "UNTHROTTLE",
+	[PERF_RECORD_FORK]	 = "FORK",
+	[PERF_RECORD_READ]	 = "READ",
+	[PERF_RECORD_SAMPLE]	 = "SAMPLE",
+};
+
+unsigned long event__total[PERF_RECORD_MAX];
+
+void event__print_totals(void)
+{
+	int i;
+	for (i = 0; i < PERF_RECORD_MAX; ++i)
+		pr_info("%10s events: %10ld\n",
+			event__name[i], event__total[i]);
+}
+
+void mem_bswap_64(void *src, int byte_size)
+{
+	u64 *m = src;
+
+	while (byte_size > 0) {
+		*m = bswap_64(*m);
+		byte_size -= sizeof(u64);
+		++m;
+	}
+}
+
+static void event__all64_swap(event_t *self)
+{
+	struct perf_event_header *hdr = &self->header;
+	mem_bswap_64(hdr + 1, self->header.size - sizeof(*hdr));
+}
+
+static void event__comm_swap(event_t *self)
+{
+	self->comm.pid = bswap_32(self->comm.pid);
+	self->comm.tid = bswap_32(self->comm.tid);
+}
+
+static void event__mmap_swap(event_t *self)
+{
+	self->mmap.pid	 = bswap_32(self->mmap.pid);
+	self->mmap.tid	 = bswap_32(self->mmap.tid);
+	self->mmap.start = bswap_64(self->mmap.start);
+	self->mmap.len	 = bswap_64(self->mmap.len);
+	self->mmap.pgoff = bswap_64(self->mmap.pgoff);
+}
+
+static void event__task_swap(event_t *self)
+{
+	self->fork.pid	= bswap_32(self->fork.pid);
+	self->fork.tid	= bswap_32(self->fork.tid);
+	self->fork.ppid	= bswap_32(self->fork.ppid);
+	self->fork.ptid	= bswap_32(self->fork.ptid);
+	self->fork.time	= bswap_64(self->fork.time);
+}
+
+static void event__read_swap(event_t *self)
+{
+	self->read.pid		= bswap_32(self->read.pid);
+	self->read.tid		= bswap_32(self->read.tid);
+	self->read.value	= bswap_64(self->read.value);
+	self->read.time_enabled	= bswap_64(self->read.time_enabled);
+	self->read.time_running	= bswap_64(self->read.time_running);
+	self->read.id		= bswap_64(self->read.id);
+}
+
+typedef void (*event__swap_op)(event_t *self);
+
+static event__swap_op event__swap_ops[] = {
+	[PERF_RECORD_MMAP]   = event__mmap_swap,
+	[PERF_RECORD_COMM]   = event__comm_swap,
+	[PERF_RECORD_FORK]   = event__task_swap,
+	[PERF_RECORD_EXIT]   = event__task_swap,
+	[PERF_RECORD_LOST]   = event__all64_swap,
+	[PERF_RECORD_READ]   = event__read_swap,
+	[PERF_RECORD_SAMPLE] = event__all64_swap,
+	[PERF_RECORD_MAX]    = NULL,
+};
+
+static int perf_session__process_event(struct perf_session *self,
+				       event_t *event,
+				       struct perf_event_ops *ops,
+				       u64 offset, u64 head)
+{
+	trace_event(event);
+
+	if (event->header.type < PERF_RECORD_MAX) {
+		dump_printf("%#Lx [%#x]: PERF_RECORD_%s",
+			    offset + head, event->header.size,
+			    event__name[event->header.type]);
+		++event__total[0];
+		++event__total[event->header.type];
+	}
+
+	if (self->header.needs_swap && event__swap_ops[event->header.type])
+		event__swap_ops[event->header.type](event);
+
+	switch (event->header.type) {
+	case PERF_RECORD_SAMPLE:
+		return ops->sample(event, self);
+	case PERF_RECORD_MMAP:
+		return ops->mmap(event, self);
+	case PERF_RECORD_COMM:
+		return ops->comm(event, self);
+	case PERF_RECORD_FORK:
+		return ops->fork(event, self);
+	case PERF_RECORD_EXIT:
+		return ops->exit(event, self);
+	case PERF_RECORD_LOST:
+		return ops->lost(event, self);
+	case PERF_RECORD_READ:
+		return ops->read(event, self);
+	case PERF_RECORD_THROTTLE:
+		return ops->throttle(event, self);
+	case PERF_RECORD_UNTHROTTLE:
+		return ops->unthrottle(event, self);
+	default:
+		self->unknown_events++;
+		return -1;
+	}
+}
+
+void perf_event_header__bswap(struct perf_event_header *self)
+{
+	self->type = bswap_32(self->type);
+	self->misc = bswap_16(self->misc);
+	self->size = bswap_16(self->size);
+}
+
+int perf_header__read_build_ids(struct perf_header *self,
+				int input, u64 offset, u64 size)
+{
+	struct build_id_event bev;
+	char filename[PATH_MAX];
+	u64 limit = offset + size;
+	int err = -1;
+
+	while (offset < limit) {
+		struct dso *dso;
+		ssize_t len;
+		struct list_head *head = &dsos__user;
+
+		if (read(input, &bev, sizeof(bev)) != sizeof(bev))
+			goto out;
+
+		if (self->needs_swap)
+			perf_event_header__bswap(&bev.header);
+
+		len = bev.header.size - sizeof(bev);
+		if (read(input, filename, len) != len)
+			goto out;
+
+		if (bev.header.misc & PERF_RECORD_MISC_KERNEL)
+			head = &dsos__kernel;
+
+		dso = __dsos__findnew(head, filename);
+		if (dso != NULL) {
+			dso__set_build_id(dso, &bev.build_id);
+			if (head == &dsos__kernel && filename[0] == '[')
+				dso->kernel = 1;
+		}
+
+		offset += bev.header.size;
+	}
+	err = 0;
+out:
+	return err;
+}
+
+static struct thread *perf_session__register_idle_thread(struct perf_session *self)
+{
+	struct thread *thread = perf_session__findnew(self, 0);
+
+	if (thread == NULL || thread__set_comm(thread, "swapper")) {
+		pr_err("problem inserting idle task.\n");
+		thread = NULL;
+	}
+
+	return thread;
+}
+
+int __perf_session__process_events(struct perf_session *self,
+				   u64 data_offset, u64 data_size,
+				   u64 file_size, struct perf_event_ops *ops)
+{
+	int err, mmap_prot, mmap_flags;
+	u64 head, shift;
+	u64 offset = 0;
+	size_t	page_size;
+	event_t *event;
+	uint32_t size;
+	char *buf;
+
+	perf_event_ops__fill_defaults(ops);
+
+	page_size = sysconf(_SC_PAGESIZE);
+
+	head = data_offset;
+	shift = page_size * (head / page_size);
+	offset += shift;
+	head -= shift;
+
+	mmap_prot  = PROT_READ;
+	mmap_flags = MAP_SHARED;
+
+	if (self->header.needs_swap) {
+		mmap_prot  |= PROT_WRITE;
+		mmap_flags = MAP_PRIVATE;
+	}
+remap:
+	buf = mmap(NULL, page_size * self->mmap_window, mmap_prot,
+		   mmap_flags, self->fd, offset);
+	if (buf == MAP_FAILED) {
+		pr_err("failed to mmap file\n");
+		err = -errno;
+		goto out_err;
+	}
+
+more:
+	event = (event_t *)(buf + head);
+
+	if (self->header.needs_swap)
+		perf_event_header__bswap(&event->header);
+	size = event->header.size;
+	if (size == 0)
+		size = 8;
+
+	if (head + event->header.size >= page_size * self->mmap_window) {
+		int munmap_ret;
+
+		shift = page_size * (head / page_size);
+
+		munmap_ret = munmap(buf, page_size * self->mmap_window);
+		assert(munmap_ret == 0);
+
+		offset += shift;
+		head -= shift;
+		goto remap;
+	}
+
+	size = event->header.size;
+
+	dump_printf("\n%#Lx [%#x]: event: %d\n",
+		    offset + head, event->header.size, event->header.type);
+
+	if (size == 0 ||
+	    perf_session__process_event(self, event, ops, offset, head) < 0) {
+		dump_printf("%#Lx [%#x]: skipping unknown header type: %d\n",
+			    offset + head, event->header.size,
+			    event->header.type);
+		/*
+		 * assume we lost track of the stream, check alignment, and
+		 * increment a single u64 in the hope to catch on again 'soon'.
+		 */
+		if (unlikely(head & 7))
+			head &= ~7ULL;
+
+		size = 8;
+	}
+
+	head += size;
+
+	if (offset + head >= data_offset + data_size)
+		goto done;
+
+	if (offset + head < file_size)
+		goto more;
+done:
+	err = 0;
+out_err:
+	return err;
+}
+
+int perf_session__process_events(struct perf_session *self,
+				 struct perf_event_ops *ops)
+{
+	int err;
+
+	if (perf_session__register_idle_thread(self) == NULL)
+		return -ENOMEM;
+
+	if (!symbol_conf.full_paths) {
+		char bf[PATH_MAX];
+
+		if (getcwd(bf, sizeof(bf)) == NULL) {
+			err = -errno;
+out_getcwd_err:
+			pr_err("failed to get the current directory\n");
+			goto out_err;
+		}
+		self->cwd = strdup(bf);
+		if (self->cwd == NULL) {
+			err = -ENOMEM;
+			goto out_getcwd_err;
+		}
+		self->cwdlen = strlen(self->cwd);
+	}
+
+	err = __perf_session__process_events(self, self->header.data_offset,
+					     self->header.data_size,
+					     self->size, ops);
+out_err:
+	return err;
+}
+
+bool perf_session__has_traces(struct perf_session *self, const char *msg)
+{
+	if (!(self->sample_type & PERF_SAMPLE_RAW)) {
+		pr_err("No trace sample to read. Did you call 'perf %s'?\n", msg);
+		return false;
+	}
+
+	return true;
+}
+
+int perf_session__set_kallsyms_ref_reloc_sym(struct perf_session *self,
+					     const char *symbol_name,
+					     u64 addr)
+{
+	char *bracket;
+	enum map_type i;
+
+	self->ref_reloc_sym.name = strdup(symbol_name);
+	if (self->ref_reloc_sym.name == NULL)
+		return -ENOMEM;
+
+	bracket = strchr(self->ref_reloc_sym.name, ']');
+	if (bracket)
+		*bracket = '\0';
+
+	self->ref_reloc_sym.addr = addr;
+
+	for (i = 0; i < MAP__NR_TYPES; ++i) {
+		struct kmap *kmap = map__kmap(self->vmlinux_maps[i]);
+		kmap->ref_reloc_sym = &self->ref_reloc_sym;
+	}
+
+	return 0;
+}
+
+static u64 map__reloc_map_ip(struct map *map, u64 ip)
+{
+	return ip + (s64)map->pgoff;
+}
+
+static u64 map__reloc_unmap_ip(struct map *map, u64 ip)
+{
+	return ip - (s64)map->pgoff;
+}
+
+void map__reloc_vmlinux(struct map *self)
+{
+	struct kmap *kmap = map__kmap(self);
+	s64 reloc;
+
+	if (!kmap->ref_reloc_sym || !kmap->ref_reloc_sym->unrelocated_addr)
+		return;
+
+	reloc = (kmap->ref_reloc_sym->unrelocated_addr -
+		 kmap->ref_reloc_sym->addr);
+
+	if (!reloc)
+		return;
+
+	self->map_ip   = map__reloc_map_ip;
+	self->unmap_ip = map__reloc_unmap_ip;
+	self->pgoff    = reloc;
+}
diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h
index 32eaa1bada0..31950fcd8a4 100644
--- a/tools/perf/util/session.h
+++ b/tools/perf/util/session.h
@@ -3,13 +3,13 @@
 
 #include "event.h"
 #include "header.h"
+#include "symbol.h"
 #include "thread.h"
 #include <linux/rbtree.h>
 #include "../../../include/linux/perf_event.h"
 
 struct ip_callchain;
 struct thread;
-struct symbol;
 
 struct perf_session {
 	struct perf_header	header;
@@ -18,10 +18,13 @@ struct perf_session {
 	struct map_groups	kmaps;
 	struct rb_root		threads;
 	struct thread		*last_match;
+	struct map		*vmlinux_maps[MAP__NR_TYPES];
 	struct events_stats	events_stats;
 	unsigned long		event_total[PERF_RECORD_MAX];
+	unsigned long		unknown_events;
 	struct rb_root		hists;
 	u64			sample_type;
+	struct ref_reloc_sym	ref_reloc_sym;
 	int			fd;
 	int			cwdlen;
 	char			*cwd;
@@ -31,23 +34,25 @@ struct perf_session {
 typedef int (*event_op)(event_t *self, struct perf_session *session);
 
 struct perf_event_ops {
-	event_op	process_sample_event;
-	event_op	process_mmap_event;
-	event_op	process_comm_event;
-	event_op	process_fork_event;
-	event_op	process_exit_event;
-	event_op	process_lost_event;
-	event_op	process_read_event;
-	event_op	process_throttle_event;
-	event_op	process_unthrottle_event;
-	int		(*sample_type_check)(struct perf_session *session);
-	unsigned long	total_unknown;
-	bool		full_paths;
+	event_op sample,
+		 mmap,
+		 comm,
+		 fork,
+		 exit,
+		 lost,
+		 read,
+		 throttle,
+		 unthrottle;
 };
 
 struct perf_session *perf_session__new(const char *filename, int mode, bool force);
 void perf_session__delete(struct perf_session *self);
 
+void perf_event_header__bswap(struct perf_event_header *self);
+
+int __perf_session__process_events(struct perf_session *self,
+				   u64 data_offset, u64 data_size, u64 size,
+				   struct perf_event_ops *ops);
 int perf_session__process_events(struct perf_session *self,
 				 struct perf_event_ops *event_ops);
 
@@ -56,6 +61,28 @@ struct symbol **perf_session__resolve_callchain(struct perf_session *self,
 						struct ip_callchain *chain,
 						struct symbol **parent);
 
-int perf_header__read_build_ids(int input, u64 offset, u64 file_size);
+bool perf_session__has_traces(struct perf_session *self, const char *msg);
+
+int perf_header__read_build_ids(struct perf_header *self, int input,
+				u64 offset, u64 file_size);
+
+int perf_session__set_kallsyms_ref_reloc_sym(struct perf_session *self,
+					     const char *symbol_name,
+					     u64 addr);
+
+void mem_bswap_64(void *src, int byte_size);
+
+static inline int __perf_session__create_kernel_maps(struct perf_session *self,
+						struct dso *kernel)
+{
+	return __map_groups__create_kernel_maps(&self->kmaps,
+						self->vmlinux_maps, kernel);
+}
 
+static inline struct map *
+	perf_session__new_module_map(struct perf_session *self,
+				     u64 start, const char *filename)
+{
+	return map_groups__new_module(&self->kmaps, start, filename);
+}
 #endif /* __PERF_SESSION_H */
diff --git a/tools/perf/util/string.c b/tools/perf/util/string.c
index 5352d7dccc6..a175949ed21 100644
--- a/tools/perf/util/string.c
+++ b/tools/perf/util/string.c
@@ -227,16 +227,73 @@ fail:
 	return NULL;
 }
 
-/* Glob expression pattern matching */
-bool strglobmatch(const char *str, const char *pat)
+/* Character class matching */
+static bool __match_charclass(const char *pat, char c, const char **npat)
+{
+	bool complement = false, ret = true;
+
+	if (*pat == '!') {
+		complement = true;
+		pat++;
+	}
+	if (*pat++ == c)	/* First character is special */
+		goto end;
+
+	while (*pat && *pat != ']') {	/* Matching */
+		if (*pat == '-' && *(pat + 1) != ']') {	/* Range */
+			if (*(pat - 1) <= c && c <= *(pat + 1))
+				goto end;
+			if (*(pat - 1) > *(pat + 1))
+				goto error;
+			pat += 2;
+		} else if (*pat++ == c)
+			goto end;
+	}
+	if (!*pat)
+		goto error;
+	ret = false;
+
+end:
+	while (*pat && *pat != ']')	/* Searching closing */
+		pat++;
+	if (!*pat)
+		goto error;
+	*npat = pat + 1;
+	return complement ? !ret : ret;
+
+error:
+	return false;
+}
+
+/* Glob/lazy pattern matching */
+static bool __match_glob(const char *str, const char *pat, bool ignore_space)
 {
 	while (*str && *pat && *pat != '*') {
-		if (*pat == '?') {
+		if (ignore_space) {
+			/* Ignore spaces for lazy matching */
+			if (isspace(*str)) {
+				str++;
+				continue;
+			}
+			if (isspace(*pat)) {
+				pat++;
+				continue;
+			}
+		}
+		if (*pat == '?') {	/* Matches any single character */
 			str++;
 			pat++;
-		} else
-			if (*str++ != *pat++)
+			continue;
+		} else if (*pat == '[')	/* Character classes/Ranges */
+			if (__match_charclass(pat + 1, *str, &pat)) {
+				str++;
+				continue;
+			} else
 				return false;
+		else if (*pat == '\\') /* Escaped char match as normal char */
+			pat++;
+		if (*str++ != *pat++)
+			return false;
 	}
 	/* Check wild card */
 	if (*pat == '*') {
@@ -251,3 +308,32 @@ bool strglobmatch(const char *str, const char *pat)
 	return !*str && !*pat;
 }
 
+/**
+ * strglobmatch - glob expression pattern matching
+ * @str: the target string to match
+ * @pat: the pattern string to match
+ *
+ * This returns true if the @str matches @pat. @pat can includes wildcards
+ * ('*','?') and character classes ([CHARS], complementation and ranges are
+ * also supported). Also, this supports escape character ('\') to use special
+ * characters as normal character.
+ *
+ * Note: if @pat syntax is broken, this always returns false.
+ */
+bool strglobmatch(const char *str, const char *pat)
+{
+	return __match_glob(str, pat, false);
+}
+
+/**
+ * strlazymatch - matching pattern strings lazily with glob pattern
+ * @str: the target string to match
+ * @pat: the pattern string to match
+ *
+ * This is similar to strglobmatch, except this ignores spaces in
+ * the target string.
+ */
+bool strlazymatch(const char *str, const char *pat)
+{
+	return __match_glob(str, pat, true);
+}
diff --git a/tools/perf/util/string.h b/tools/perf/util/string.h
index 02ede58c54b..542e44de371 100644
--- a/tools/perf/util/string.h
+++ b/tools/perf/util/string.h
@@ -10,6 +10,7 @@ s64 perf_atoll(const char *str);
 char **argv_split(const char *str, int *argcp);
 void argv_free(char **argv);
 bool strglobmatch(const char *str, const char *pat);
+bool strlazymatch(const char *str, const char *pat);
 
 #define _STR(x) #x
 #define STR(x) _STR(x)
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index ab92763edb0..323c0aea0a9 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -1,6 +1,5 @@
 #include "util.h"
 #include "../perf.h"
-#include "session.h"
 #include "sort.h"
 #include "string.h"
 #include "symbol.h"
@@ -22,6 +21,7 @@
 enum dso_origin {
 	DSO__ORIG_KERNEL = 0,
 	DSO__ORIG_JAVA_JIT,
+	DSO__ORIG_BUILD_ID_CACHE,
 	DSO__ORIG_FEDORA,
 	DSO__ORIG_UBUNTU,
 	DSO__ORIG_BUILDID,
@@ -33,7 +33,7 @@ enum dso_origin {
 static void dsos__add(struct list_head *head, struct dso *dso);
 static struct map *map__new2(u64 start, struct dso *dso, enum map_type type);
 static int dso__load_kernel_sym(struct dso *self, struct map *map,
-				struct perf_session *session, symbol_filter_t filter);
+				symbol_filter_t filter);
 static int vmlinux_path__nr_entries;
 static char **vmlinux_path;
 
@@ -53,17 +53,12 @@ bool dso__sorted_by_name(const struct dso *self, enum map_type type)
 	return self->sorted_by_name & (1 << type);
 }
 
-static void dso__set_loaded(struct dso *self, enum map_type type)
-{
-	self->loaded |= (1 << type);
-}
-
 static void dso__set_sorted_by_name(struct dso *self, enum map_type type)
 {
 	self->sorted_by_name |= (1 << type);
 }
 
-static bool symbol_type__is_a(char symbol_type, enum map_type map_type)
+bool symbol_type__is_a(char symbol_type, enum map_type map_type)
 {
 	switch (map_type) {
 	case MAP__FUNCTION:
@@ -142,14 +137,14 @@ static struct symbol *symbol__new(u64 start, u64 len, const char *name)
 	self->start = start;
 	self->end   = len ? start + len - 1 : start;
 
-	pr_debug3("%s: %s %#Lx-%#Lx\n", __func__, name, start, self->end);
+	pr_debug4("%s: %s %#Lx-%#Lx\n", __func__, name, start, self->end);
 
 	memcpy(self->name, name, namelen);
 
 	return self;
 }
 
-static void symbol__delete(struct symbol *self)
+void symbol__delete(struct symbol *self)
 {
 	free(((void *)self) - symbol_conf.priv_size);
 }
@@ -160,7 +155,7 @@ static size_t symbol__fprintf(struct symbol *self, FILE *fp)
 		       self->start, self->end, self->name);
 }
 
-static void dso__set_long_name(struct dso *self, char *name)
+void dso__set_long_name(struct dso *self, char *name)
 {
 	if (name == NULL)
 		return;
@@ -175,7 +170,7 @@ static void dso__set_basename(struct dso *self)
 
 struct dso *dso__new(const char *name)
 {
-	struct dso *self = malloc(sizeof(*self) + strlen(name) + 1);
+	struct dso *self = zalloc(sizeof(*self) + strlen(name) + 1);
 
 	if (self != NULL) {
 		int i;
@@ -344,10 +339,10 @@ void dso__sort_by_name(struct dso *self, enum map_type type)
 				     &self->symbols[type]);
 }
 
-int build_id__sprintf(u8 *self, int len, char *bf)
+int build_id__sprintf(const u8 *self, int len, char *bf)
 {
 	char *bid = bf;
-	u8 *raw = self;
+	const u8 *raw = self;
 	int i;
 
 	for (i = 0; i < len; ++i) {
@@ -372,6 +367,10 @@ size_t dso__fprintf(struct dso *self, enum map_type type, FILE *fp)
 	struct rb_node *nd;
 	size_t ret = fprintf(fp, "dso: %s (", self->short_name);
 
+	if (self->short_name != self->long_name)
+		ret += fprintf(fp, "%s, ", self->long_name);
+	ret += fprintf(fp, "%s, %sloaded, ", map_type__name[type],
+		       self->loaded ? "" : "NOT ");
 	ret += dso__fprintf_buildid(self, fp);
 	ret += fprintf(fp, ")\n");
 	for (nd = rb_first(&self->symbols[type]); nd; nd = rb_next(nd)) {
@@ -382,24 +381,20 @@ size_t dso__fprintf(struct dso *self, enum map_type type, FILE *fp)
 	return ret;
 }
 
-/*
- * Loads the function entries in /proc/kallsyms into kernel_map->dso,
- * so that we can in the next step set the symbol ->end address and then
- * call kernel_maps__split_kallsyms.
- */
-static int dso__load_all_kallsyms(struct dso *self, struct map *map)
+int kallsyms__parse(const char *filename, void *arg,
+		    int (*process_symbol)(void *arg, const char *name,
+						     char type, u64 start))
 {
 	char *line = NULL;
 	size_t n;
-	struct rb_root *root = &self->symbols[map->type];
-	FILE *file = fopen("/proc/kallsyms", "r");
+	int err = 0;
+	FILE *file = fopen(filename, "r");
 
 	if (file == NULL)
 		goto out_failure;
 
 	while (!feof(file)) {
 		u64 start;
-		struct symbol *sym;
 		int line_len, len;
 		char symbol_type;
 		char *symbol_name;
@@ -420,43 +415,72 @@ static int dso__load_all_kallsyms(struct dso *self, struct map *map)
 			continue;
 
 		symbol_type = toupper(line[len]);
-		if (!symbol_type__is_a(symbol_type, map->type))
-			continue;
-
 		symbol_name = line + len + 2;
-		/*
-		 * Will fix up the end later, when we have all symbols sorted.
-		 */
-		sym = symbol__new(start, 0, symbol_name);
 
-		if (sym == NULL)
-			goto out_delete_line;
-		/*
-		 * We will pass the symbols to the filter later, in
-		 * map__split_kallsyms, when we have split the maps per module
-		 */
-		symbols__insert(root, sym);
+		err = process_symbol(arg, symbol_name, symbol_type, start);
+		if (err)
+			break;
 	}
 
 	free(line);
 	fclose(file);
+	return err;
 
-	return 0;
-
-out_delete_line:
-	free(line);
 out_failure:
 	return -1;
 }
 
+struct process_kallsyms_args {
+	struct map *map;
+	struct dso *dso;
+};
+
+static int map__process_kallsym_symbol(void *arg, const char *name,
+				       char type, u64 start)
+{
+	struct symbol *sym;
+	struct process_kallsyms_args *a = arg;
+	struct rb_root *root = &a->dso->symbols[a->map->type];
+
+	if (!symbol_type__is_a(type, a->map->type))
+		return 0;
+
+	/*
+	 * Will fix up the end later, when we have all symbols sorted.
+	 */
+	sym = symbol__new(start, 0, name);
+
+	if (sym == NULL)
+		return -ENOMEM;
+	/*
+	 * We will pass the symbols to the filter later, in
+	 * map__split_kallsyms, when we have split the maps per module
+	 */
+	symbols__insert(root, sym);
+	return 0;
+}
+
+/*
+ * Loads the function entries in /proc/kallsyms into kernel_map->dso,
+ * so that we can in the next step set the symbol ->end address and then
+ * call kernel_maps__split_kallsyms.
+ */
+static int dso__load_all_kallsyms(struct dso *self, const char *filename,
+				  struct map *map)
+{
+	struct process_kallsyms_args args = { .map = map, .dso = self, };
+	return kallsyms__parse(filename, &args, map__process_kallsym_symbol);
+}
+
 /*
  * Split the symbols into maps, making sure there are no overlaps, i.e. the
  * kernel range is broken in several maps, named [kernel].N, as we don't have
  * the original ELF section names vmlinux have.
  */
 static int dso__split_kallsyms(struct dso *self, struct map *map,
-			       struct perf_session *session, symbol_filter_t filter)
+			       symbol_filter_t filter)
 {
+	struct map_groups *kmaps = map__kmap(map)->kmaps;
 	struct map *curr_map = map;
 	struct symbol *pos;
 	int count = 0;
@@ -477,13 +501,17 @@ static int dso__split_kallsyms(struct dso *self, struct map *map,
 
 			*module++ = '\0';
 
-			if (strcmp(self->name, module)) {
-				curr_map = map_groups__find_by_name(&session->kmaps, map->type, module);
+			if (strcmp(curr_map->dso->short_name, module)) {
+				curr_map = map_groups__find_by_name(kmaps, map->type, module);
 				if (curr_map == NULL) {
 					pr_debug("/proc/{kallsyms,modules} "
-					         "inconsistency!\n");
+					         "inconsistency while looking "
+						 "for \"%s\" module!\n", module);
 					return -1;
 				}
+
+				if (curr_map->dso->loaded)
+					goto discard_symbol;
 			}
 			/*
 			 * So that we look just like we get from .ko files,
@@ -503,13 +531,13 @@ static int dso__split_kallsyms(struct dso *self, struct map *map,
 				return -1;
 
 			curr_map = map__new2(pos->start, dso, map->type);
-			if (map == NULL) {
+			if (curr_map == NULL) {
 				dso__delete(dso);
 				return -1;
 			}
 
 			curr_map->map_ip = curr_map->unmap_ip = identity__map_ip;
-			map_groups__insert(&session->kmaps, curr_map);
+			map_groups__insert(kmaps, curr_map);
 			++kernel_range;
 		}
 
@@ -528,17 +556,16 @@ discard_symbol:		rb_erase(&pos->rb_node, root);
 	return count;
 }
 
-
-static int dso__load_kallsyms(struct dso *self, struct map *map,
-			      struct perf_session *session, symbol_filter_t filter)
+int dso__load_kallsyms(struct dso *self, const char *filename,
+		       struct map *map, symbol_filter_t filter)
 {
-	if (dso__load_all_kallsyms(self, map) < 0)
+	if (dso__load_all_kallsyms(self, filename, map) < 0)
 		return -1;
 
 	symbols__fixup_end(&self->symbols[map->type]);
 	self->origin = DSO__ORIG_KERNEL;
 
-	return dso__split_kallsyms(self, map, session, filter);
+	return dso__split_kallsyms(self, map, filter);
 }
 
 static int dso__load_perf_map(struct dso *self, struct map *map,
@@ -864,10 +891,10 @@ static bool elf_sec__is_a(GElf_Shdr *self, Elf_Data *secstrs, enum map_type type
 	}
 }
 
-static int dso__load_sym(struct dso *self, struct map *map,
-			 struct perf_session *session, const char *name, int fd,
-			 symbol_filter_t filter, int kernel, int kmodule)
+static int dso__load_sym(struct dso *self, struct map *map, const char *name,
+			 int fd, symbol_filter_t filter, int kmodule)
 {
+	struct kmap *kmap = self->kernel ? map__kmap(map) : NULL;
 	struct map *curr_map = map;
 	struct dso *curr_dso = self;
 	size_t dso_name_len = strlen(self->short_name);
@@ -924,7 +951,7 @@ static int dso__load_sym(struct dso *self, struct map *map,
 	nr_syms = shdr.sh_size / shdr.sh_entsize;
 
 	memset(&sym, 0, sizeof(sym));
-	if (!kernel) {
+	if (!self->kernel) {
 		self->adjust_symbols = (ehdr.e_type == ET_EXEC ||
 				elf_section_by_name(elf, &ehdr, &shdr,
 						     ".gnu.prelink_undo",
@@ -933,11 +960,15 @@ static int dso__load_sym(struct dso *self, struct map *map,
 
 	elf_symtab__for_each_symbol(syms, nr_syms, idx, sym) {
 		struct symbol *f;
-		const char *elf_name;
+		const char *elf_name = elf_sym__name(&sym, symstrs);
 		char *demangled = NULL;
 		int is_label = elf_sym__is_label(&sym);
 		const char *section_name;
 
+		if (kmap && kmap->ref_reloc_sym && kmap->ref_reloc_sym->name &&
+		    strcmp(elf_name, kmap->ref_reloc_sym->name) == 0)
+			kmap->ref_reloc_sym->unrelocated_addr = sym.st_value;
+
 		if (!is_label && !elf_sym__is_a(&sym, map->type))
 			continue;
 
@@ -950,10 +981,9 @@ static int dso__load_sym(struct dso *self, struct map *map,
 		if (is_label && !elf_sec__is_a(&shdr, secstrs, map->type))
 			continue;
 
-		elf_name = elf_sym__name(&sym, symstrs);
 		section_name = elf_sec__name(&shdr, secstrs);
 
-		if (kernel || kmodule) {
+		if (self->kernel || kmodule) {
 			char dso_name[PATH_MAX];
 
 			if (strcmp(section_name,
@@ -969,7 +999,7 @@ static int dso__load_sym(struct dso *self, struct map *map,
 			snprintf(dso_name, sizeof(dso_name),
 				 "%s%s", self->short_name, section_name);
 
-			curr_map = map_groups__find_by_name(&session->kmaps, map->type, dso_name);
+			curr_map = map_groups__find_by_name(kmap->kmaps, map->type, dso_name);
 			if (curr_map == NULL) {
 				u64 start = sym.st_value;
 
@@ -980,7 +1010,7 @@ static int dso__load_sym(struct dso *self, struct map *map,
 				if (curr_dso == NULL)
 					goto out_elf_end;
 				curr_map = map__new2(start, curr_dso,
-						     MAP__FUNCTION);
+						     map->type);
 				if (curr_map == NULL) {
 					dso__delete(curr_dso);
 					goto out_elf_end;
@@ -988,8 +1018,9 @@ static int dso__load_sym(struct dso *self, struct map *map,
 				curr_map->map_ip = identity__map_ip;
 				curr_map->unmap_ip = identity__map_ip;
 				curr_dso->origin = DSO__ORIG_KERNEL;
-				map_groups__insert(&session->kmaps, curr_map);
+				map_groups__insert(kmap->kmaps, curr_map);
 				dsos__add(&dsos__kernel, curr_dso);
+				dso__set_loaded(curr_dso, map->type);
 			} else
 				curr_dso = curr_map->dso;
 
@@ -997,9 +1028,10 @@ static int dso__load_sym(struct dso *self, struct map *map,
 		}
 
 		if (curr_dso->adjust_symbols) {
-			pr_debug2("adjusting symbol: st_value: %Lx sh_addr: "
-				  "%Lx sh_offset: %Lx\n", (u64)sym.st_value,
-				  (u64)shdr.sh_addr, (u64)shdr.sh_offset);
+			pr_debug4("%s: adjusting symbol: st_value: %#Lx "
+				  "sh_addr: %#Lx sh_offset: %#Lx\n", __func__,
+				  (u64)sym.st_value, (u64)shdr.sh_addr,
+				  (u64)shdr.sh_offset);
 			sym.st_value -= shdr.sh_addr - shdr.sh_offset;
 		}
 		/*
@@ -1027,8 +1059,16 @@ new_symbol:
 	/*
 	 * For misannotated, zeroed, ASM function sizes.
 	 */
-	if (nr > 0)
+	if (nr > 0) {
 		symbols__fixup_end(&self->symbols[map->type]);
+		if (kmap) {
+			/*
+			 * We need to fixup this here too because we create new
+			 * maps here, for things like vsyscall sections.
+			 */
+			__map_groups__fixup_end(kmap->kmaps, map->type);
+		}
+	}
 	err = nr;
 out_elf_end:
 	elf_end(elf);
@@ -1041,25 +1081,28 @@ static bool dso__build_id_equal(const struct dso *self, u8 *build_id)
 	return memcmp(self->build_id, build_id, sizeof(self->build_id)) == 0;
 }
 
-static bool __dsos__read_build_ids(struct list_head *head)
+static bool __dsos__read_build_ids(struct list_head *head, bool with_hits)
 {
 	bool have_build_id = false;
 	struct dso *pos;
 
-	list_for_each_entry(pos, head, node)
+	list_for_each_entry(pos, head, node) {
+		if (with_hits && !pos->hit)
+			continue;
 		if (filename__read_build_id(pos->long_name, pos->build_id,
 					    sizeof(pos->build_id)) > 0) {
 			have_build_id	  = true;
 			pos->has_build_id = true;
 		}
+	}
 
 	return have_build_id;
 }
 
-bool dsos__read_build_ids(void)
+bool dsos__read_build_ids(bool with_hits)
 {
-	bool kbuildids = __dsos__read_build_ids(&dsos__kernel),
-	     ubuildids = __dsos__read_build_ids(&dsos__user);
+	bool kbuildids = __dsos__read_build_ids(&dsos__kernel, with_hits),
+	     ubuildids = __dsos__read_build_ids(&dsos__user, with_hits);
 	return kbuildids || ubuildids;
 }
 
@@ -1191,6 +1234,7 @@ char dso__symtab_origin(const struct dso *self)
 	static const char origin[] = {
 		[DSO__ORIG_KERNEL] =   'k',
 		[DSO__ORIG_JAVA_JIT] = 'j',
+		[DSO__ORIG_BUILD_ID_CACHE] = 'B',
 		[DSO__ORIG_FEDORA] =   'f',
 		[DSO__ORIG_UBUNTU] =   'u',
 		[DSO__ORIG_BUILDID] =  'b',
@@ -1203,19 +1247,19 @@ char dso__symtab_origin(const struct dso *self)
 	return origin[self->origin];
 }
 
-int dso__load(struct dso *self, struct map *map, struct perf_session *session,
-	      symbol_filter_t filter)
+int dso__load(struct dso *self, struct map *map, symbol_filter_t filter)
 {
 	int size = PATH_MAX;
 	char *name;
 	u8 build_id[BUILD_ID_SIZE];
+	char build_id_hex[BUILD_ID_SIZE * 2 + 1];
 	int ret = -1;
 	int fd;
 
 	dso__set_loaded(self, map->type);
 
 	if (self->kernel)
-		return dso__load_kernel_sym(self, map, session, filter);
+		return dso__load_kernel_sym(self, map, filter);
 
 	name = malloc(size);
 	if (!name)
@@ -1230,8 +1274,16 @@ int dso__load(struct dso *self, struct map *map, struct perf_session *session,
 		return ret;
 	}
 
-	self->origin = DSO__ORIG_FEDORA - 1;
+	self->origin = DSO__ORIG_BUILD_ID_CACHE;
 
+	if (self->has_build_id) {
+		build_id__sprintf(self->build_id, sizeof(self->build_id),
+				  build_id_hex);
+		snprintf(name, size, "%s/%s/.build-id/%.2s/%s",
+			 getenv("HOME"), DEBUG_CACHE_DIR,
+			 build_id_hex, build_id_hex + 2);
+		goto open_file;
+	}
 more:
 	do {
 		self->origin++;
@@ -1247,8 +1299,6 @@ more:
 		case DSO__ORIG_BUILDID:
 			if (filename__read_build_id(self->long_name, build_id,
 						    sizeof(build_id))) {
-				char build_id_hex[BUILD_ID_SIZE * 2 + 1];
-
 				build_id__sprintf(build_id, sizeof(build_id),
 						  build_id_hex);
 				snprintf(name, size,
@@ -1276,11 +1326,11 @@ compare_build_id:
 			if (!dso__build_id_equal(self, build_id))
 				goto more;
 		}
-
+open_file:
 		fd = open(name, O_RDONLY);
 	} while (fd < 0);
 
-	ret = dso__load_sym(self, map, NULL, name, fd, filter, 0, 0);
+	ret = dso__load_sym(self, map, name, fd, filter, 0);
 	close(fd);
 
 	/*
@@ -1309,14 +1359,34 @@ struct map *map_groups__find_by_name(struct map_groups *self,
 	for (nd = rb_first(&self->maps[type]); nd; nd = rb_next(nd)) {
 		struct map *map = rb_entry(nd, struct map, rb_node);
 
-		if (map->dso && strcmp(map->dso->name, name) == 0)
+		if (map->dso && strcmp(map->dso->short_name, name) == 0)
 			return map;
 	}
 
 	return NULL;
 }
 
-static int perf_session__set_modules_path_dir(struct perf_session *self, char *dirname)
+static int dso__kernel_module_get_build_id(struct dso *self)
+{
+	char filename[PATH_MAX];
+	/*
+	 * kernel module short names are of the form "[module]" and
+	 * we need just "module" here.
+	 */
+	const char *name = self->short_name + 1;
+
+	snprintf(filename, sizeof(filename),
+		 "/sys/module/%.*s/notes/.note.gnu.build-id",
+		 (int)strlen(name - 1), name);
+
+	if (sysfs__read_build_id(filename, self->build_id,
+				 sizeof(self->build_id)) == 0)
+		self->has_build_id = true;
+
+	return 0;
+}
+
+static int map_groups__set_modules_path_dir(struct map_groups *self, char *dirname)
 {
 	struct dirent *dent;
 	DIR *dir = opendir(dirname);
@@ -1336,7 +1406,7 @@ static int perf_session__set_modules_path_dir(struct perf_session *self, char *d
 
 			snprintf(path, sizeof(path), "%s/%s",
 				 dirname, dent->d_name);
-			if (perf_session__set_modules_path_dir(self, path) < 0)
+			if (map_groups__set_modules_path_dir(self, path) < 0)
 				goto failure;
 		} else {
 			char *dot = strrchr(dent->d_name, '.'),
@@ -1350,7 +1420,7 @@ static int perf_session__set_modules_path_dir(struct perf_session *self, char *d
 				 (int)(dot - dent->d_name), dent->d_name);
 
 			strxfrchar(dso_name, '-', '_');
-			map = map_groups__find_by_name(&self->kmaps, MAP__FUNCTION, dso_name);
+			map = map_groups__find_by_name(self, MAP__FUNCTION, dso_name);
 			if (map == NULL)
 				continue;
 
@@ -1361,6 +1431,7 @@ static int perf_session__set_modules_path_dir(struct perf_session *self, char *d
 			if (long_name == NULL)
 				goto failure;
 			dso__set_long_name(map->dso, long_name);
+			dso__kernel_module_get_build_id(map->dso);
 		}
 	}
 
@@ -1370,7 +1441,7 @@ failure:
 	return -1;
 }
 
-static int perf_session__set_modules_path(struct perf_session *self)
+static int map_groups__set_modules_path(struct map_groups *self)
 {
 	struct utsname uts;
 	char modules_path[PATH_MAX];
@@ -1381,7 +1452,7 @@ static int perf_session__set_modules_path(struct perf_session *self)
 	snprintf(modules_path, sizeof(modules_path), "/lib/modules/%s/kernel",
 		 uts.release);
 
-	return perf_session__set_modules_path_dir(self, modules_path);
+	return map_groups__set_modules_path_dir(self, modules_path);
 }
 
 /*
@@ -1391,8 +1462,8 @@ static int perf_session__set_modules_path(struct perf_session *self)
  */
 static struct map *map__new2(u64 start, struct dso *dso, enum map_type type)
 {
-	struct map *self = malloc(sizeof(*self));
-
+	struct map *self = zalloc(sizeof(*self) +
+				  (dso->kernel ? sizeof(struct kmap) : 0));
 	if (self != NULL) {
 		/*
 		 * ->end will be filled after we load all the symbols
@@ -1403,7 +1474,25 @@ static struct map *map__new2(u64 start, struct dso *dso, enum map_type type)
 	return self;
 }
 
-static int perf_session__create_module_maps(struct perf_session *self)
+struct map *map_groups__new_module(struct map_groups *self, u64 start,
+				   const char *filename)
+{
+	struct map *map;
+	struct dso *dso = __dsos__findnew(&dsos__kernel, filename);
+
+	if (dso == NULL)
+		return NULL;
+
+	map = map__new2(start, dso, MAP__FUNCTION);
+	if (map == NULL)
+		return NULL;
+
+	dso->origin = DSO__ORIG_KMODULE;
+	map_groups__insert(self, map);
+	return map;
+}
+
+static int map_groups__create_modules(struct map_groups *self)
 {
 	char *line = NULL;
 	size_t n;
@@ -1416,7 +1505,6 @@ static int perf_session__create_module_maps(struct perf_session *self)
 	while (!feof(file)) {
 		char name[PATH_MAX];
 		u64 start;
-		struct dso *dso;
 		char *sep;
 		int line_len;
 
@@ -1442,32 +1530,16 @@ static int perf_session__create_module_maps(struct perf_session *self)
 		*sep = '\0';
 
 		snprintf(name, sizeof(name), "[%s]", line);
-		dso = dso__new(name);
-
-		if (dso == NULL)
-			goto out_delete_line;
-
-		map = map__new2(start, dso, MAP__FUNCTION);
-		if (map == NULL) {
-			dso__delete(dso);
+		map = map_groups__new_module(self, start, name);
+		if (map == NULL)
 			goto out_delete_line;
-		}
-
-		snprintf(name, sizeof(name),
-			 "/sys/module/%s/notes/.note.gnu.build-id", line);
-		if (sysfs__read_build_id(name, dso->build_id,
-					 sizeof(dso->build_id)) == 0)
-			dso->has_build_id = true;
-
-		dso->origin = DSO__ORIG_KMODULE;
-		map_groups__insert(&self->kmaps, map);
-		dsos__add(&dsos__kernel, dso);
+		dso__kernel_module_get_build_id(map->dso);
 	}
 
 	free(line);
 	fclose(file);
 
-	return perf_session__set_modules_path(self);
+	return map_groups__set_modules_path(self);
 
 out_delete_line:
 	free(line);
@@ -1476,7 +1548,6 @@ out_failure:
 }
 
 static int dso__load_vmlinux(struct dso *self, struct map *map,
-			     struct perf_session *session,
 			     const char *vmlinux, symbol_filter_t filter)
 {
 	int err = -1, fd;
@@ -1510,51 +1581,124 @@ static int dso__load_vmlinux(struct dso *self, struct map *map,
 		return -1;
 
 	dso__set_loaded(self, map->type);
-	err = dso__load_sym(self, map, session, self->long_name, fd, filter, 1, 0);
+	err = dso__load_sym(self, map, vmlinux, fd, filter, 0);
 	close(fd);
 
+	if (err > 0)
+		pr_debug("Using %s for symbols\n", vmlinux);
+
+	return err;
+}
+
+int dso__load_vmlinux_path(struct dso *self, struct map *map,
+			   symbol_filter_t filter)
+{
+	int i, err = 0;
+
+	pr_debug("Looking at the vmlinux_path (%d entries long)\n",
+		 vmlinux_path__nr_entries);
+
+	for (i = 0; i < vmlinux_path__nr_entries; ++i) {
+		err = dso__load_vmlinux(self, map, vmlinux_path[i], filter);
+		if (err > 0) {
+			dso__set_long_name(self, strdup(vmlinux_path[i]));
+			break;
+		}
+	}
+
 	return err;
 }
 
 static int dso__load_kernel_sym(struct dso *self, struct map *map,
-				struct perf_session *session, symbol_filter_t filter)
+				symbol_filter_t filter)
 {
 	int err;
-	bool is_kallsyms;
+	const char *kallsyms_filename = NULL;
+	char *kallsyms_allocated_filename = NULL;
+	/*
+	 * Step 1: if the user specified a vmlinux filename, use it and only
+	 * it, reporting errors to the user if it cannot be used.
+	 *
+	 * For instance, try to analyse an ARM perf.data file _without_ a
+	 * build-id, or if the user specifies the wrong path to the right
+	 * vmlinux file, obviously we can't fallback to another vmlinux (a
+	 * x86_86 one, on the machine where analysis is being performed, say),
+	 * or worse, /proc/kallsyms.
+	 *
+	 * If the specified file _has_ a build-id and there is a build-id
+	 * section in the perf.data file, we will still do the expected
+	 * validation in dso__load_vmlinux and will bail out if they don't
+	 * match.
+	 */
+	if (symbol_conf.vmlinux_name != NULL) {
+		err = dso__load_vmlinux(self, map,
+					symbol_conf.vmlinux_name, filter);
+		goto out_try_fixup;
+	}
 
 	if (vmlinux_path != NULL) {
-		int i;
-		pr_debug("Looking at the vmlinux_path (%d entries long)\n",
-			 vmlinux_path__nr_entries);
-		for (i = 0; i < vmlinux_path__nr_entries; ++i) {
-			err = dso__load_vmlinux(self, map, session,
-						vmlinux_path[i], filter);
-			if (err > 0) {
-				pr_debug("Using %s for symbols\n",
-					 vmlinux_path[i]);
-				dso__set_long_name(self,
-						   strdup(vmlinux_path[i]));
-				goto out_fixup;
+		err = dso__load_vmlinux_path(self, map, filter);
+		if (err > 0)
+			goto out_fixup;
+	}
+
+	/*
+	 * Say the kernel DSO was created when processing the build-id header table,
+	 * we have a build-id, so check if it is the same as the running kernel,
+	 * using it if it is.
+	 */
+	if (self->has_build_id) {
+		u8 kallsyms_build_id[BUILD_ID_SIZE];
+		char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+		if (sysfs__read_build_id("/sys/kernel/notes", kallsyms_build_id,
+					 sizeof(kallsyms_build_id)) == 0) {
+			if (dso__build_id_equal(self, kallsyms_build_id)) {
+				kallsyms_filename = "/proc/kallsyms";
+				goto do_kallsyms;
 			}
 		}
-	}
+		/*
+		 * Now look if we have it on the build-id cache in
+		 * $HOME/.debug/[kernel.kallsyms].
+		 */
+		build_id__sprintf(self->build_id, sizeof(self->build_id),
+				  sbuild_id);
 
-	is_kallsyms = self->long_name[0] == '[';
-	if (is_kallsyms)
-		goto do_kallsyms;
+		if (asprintf(&kallsyms_allocated_filename,
+			     "%s/.debug/[kernel.kallsyms]/%s",
+			     getenv("HOME"), sbuild_id) == -1) {
+			pr_err("Not enough memory for kallsyms file lookup\n");
+			return -1;
+		}
 
-	err = dso__load_vmlinux(self, map, session, self->long_name, filter);
-	if (err <= 0) {
-		pr_info("The file %s cannot be used, "
-			"trying to use /proc/kallsyms...", self->long_name);
-do_kallsyms:
-		err = dso__load_kallsyms(self, map, session, filter);
-		if (err > 0 && !is_kallsyms)
-                        dso__set_long_name(self, strdup("[kernel.kallsyms]"));
+		kallsyms_filename = kallsyms_allocated_filename;
+
+		if (access(kallsyms_filename, F_OK)) {
+			pr_err("No kallsyms or vmlinux with build-id %s "
+			       "was found\n", sbuild_id);
+			free(kallsyms_allocated_filename);
+			return -1;
+		}
+	} else {
+		/*
+		 * Last resort, if we don't have a build-id and couldn't find
+		 * any vmlinux file, try the running kernel kallsyms table.
+		 */
+		kallsyms_filename = "/proc/kallsyms";
 	}
 
+do_kallsyms:
+	err = dso__load_kallsyms(self, kallsyms_filename, map, filter);
+	if (err > 0)
+		pr_debug("Using %s for symbols\n", kallsyms_filename);
+	free(kallsyms_allocated_filename);
+
+out_try_fixup:
 	if (err > 0) {
 out_fixup:
+		if (kallsyms_filename != NULL)
+			dso__set_long_name(self, strdup("[kernel.kallsyms]"));
 		map__fixup_start(map);
 		map__fixup_end(map);
 	}
@@ -1564,7 +1708,6 @@ out_fixup:
 
 LIST_HEAD(dsos__user);
 LIST_HEAD(dsos__kernel);
-struct dso *vdso;
 
 static void dsos__add(struct list_head *head, struct dso *dso)
 {
@@ -1576,19 +1719,19 @@ static struct dso *dsos__find(struct list_head *head, const char *name)
 	struct dso *pos;
 
 	list_for_each_entry(pos, head, node)
-		if (strcmp(pos->name, name) == 0)
+		if (strcmp(pos->long_name, name) == 0)
 			return pos;
 	return NULL;
 }
 
-struct dso *dsos__findnew(const char *name)
+struct dso *__dsos__findnew(struct list_head *head, const char *name)
 {
-	struct dso *dso = dsos__find(&dsos__user, name);
+	struct dso *dso = dsos__find(head, name);
 
 	if (!dso) {
 		dso = dso__new(name);
 		if (dso != NULL) {
-			dsos__add(&dsos__user, dso);
+			dsos__add(head, dso);
 			dso__set_basename(dso);
 		}
 	}
@@ -1613,75 +1756,78 @@ void dsos__fprintf(FILE *fp)
 	__dsos__fprintf(&dsos__user, fp);
 }
 
-static size_t __dsos__fprintf_buildid(struct list_head *head, FILE *fp)
+static size_t __dsos__fprintf_buildid(struct list_head *head, FILE *fp,
+				      bool with_hits)
 {
 	struct dso *pos;
 	size_t ret = 0;
 
 	list_for_each_entry(pos, head, node) {
+		if (with_hits && !pos->hit)
+			continue;
 		ret += dso__fprintf_buildid(pos, fp);
 		ret += fprintf(fp, " %s\n", pos->long_name);
 	}
 	return ret;
 }
 
-size_t dsos__fprintf_buildid(FILE *fp)
+size_t dsos__fprintf_buildid(FILE *fp, bool with_hits)
 {
-	return (__dsos__fprintf_buildid(&dsos__kernel, fp) +
-		__dsos__fprintf_buildid(&dsos__user, fp));
+	return (__dsos__fprintf_buildid(&dsos__kernel, fp, with_hits) +
+		__dsos__fprintf_buildid(&dsos__user, fp, with_hits));
 }
 
-static struct dso *dsos__create_kernel( const char *vmlinux)
+struct dso *dso__new_kernel(const char *name)
 {
-	struct dso *kernel = dso__new(vmlinux ?: "[kernel.kallsyms]");
+	struct dso *self = dso__new(name ?: "[kernel.kallsyms]");
 
-	if (kernel == NULL)
-		return NULL;
+	if (self != NULL) {
+		self->short_name = "[kernel]";
+		self->kernel	 = 1;
+	}
 
-	kernel->short_name = "[kernel]";
-	kernel->kernel	   = 1;
+	return self;
+}
 
-	vdso = dso__new("[vdso]");
-	if (vdso == NULL)
-		goto out_delete_kernel_dso;
-	dso__set_loaded(vdso, MAP__FUNCTION);
+void dso__read_running_kernel_build_id(struct dso *self)
+{
+	if (sysfs__read_build_id("/sys/kernel/notes", self->build_id,
+				 sizeof(self->build_id)) == 0)
+		self->has_build_id = true;
+}
 
-	if (sysfs__read_build_id("/sys/kernel/notes", kernel->build_id,
-				 sizeof(kernel->build_id)) == 0)
-		kernel->has_build_id = true;
+static struct dso *dsos__create_kernel(const char *vmlinux)
+{
+	struct dso *kernel = dso__new_kernel(vmlinux);
 
-	dsos__add(&dsos__kernel, kernel);
-	dsos__add(&dsos__user, vdso);
+	if (kernel != NULL) {
+		dso__read_running_kernel_build_id(kernel);
+		dsos__add(&dsos__kernel, kernel);
+	}
 
 	return kernel;
-
-out_delete_kernel_dso:
-	dso__delete(kernel);
-	return NULL;
 }
 
-static int map_groups__create_kernel_maps(struct map_groups *self, const char *vmlinux)
+int __map_groups__create_kernel_maps(struct map_groups *self,
+				     struct map *vmlinux_maps[MAP__NR_TYPES],
+				     struct dso *kernel)
 {
-	struct map *functions, *variables;
-	struct dso *kernel = dsos__create_kernel(vmlinux);
+	enum map_type type;
 
-	if (kernel == NULL)
-		return -1;
+	for (type = 0; type < MAP__NR_TYPES; ++type) {
+		struct kmap *kmap;
 
-	functions = map__new2(0, kernel, MAP__FUNCTION);
-	if (functions == NULL)
-		return -1;
+		vmlinux_maps[type] = map__new2(0, kernel, type);
+		if (vmlinux_maps[type] == NULL)
+			return -1;
 
-	variables = map__new2(0, kernel, MAP__VARIABLE);
-	if (variables == NULL) {
-		map__delete(functions);
-		return -1;
-	}
+		vmlinux_maps[type]->map_ip =
+			vmlinux_maps[type]->unmap_ip = identity__map_ip;
 
-	functions->map_ip = functions->unmap_ip =
-		variables->map_ip = variables->unmap_ip = identity__map_ip;
-	map_groups__insert(self, functions);
-	map_groups__insert(self, variables);
+		kmap = map__kmap(vmlinux_maps[type]);
+		kmap->kmaps = self;
+		map_groups__insert(self, vmlinux_maps[type]);
+	}
 
 	return 0;
 }
@@ -1791,19 +1937,22 @@ out_free_comm_list:
 	return -1;
 }
 
-int perf_session__create_kernel_maps(struct perf_session *self)
+int map_groups__create_kernel_maps(struct map_groups *self,
+				   struct map *vmlinux_maps[MAP__NR_TYPES])
 {
-	if (map_groups__create_kernel_maps(&self->kmaps,
-					   symbol_conf.vmlinux_name) < 0)
+	struct dso *kernel = dsos__create_kernel(symbol_conf.vmlinux_name);
+
+	if (kernel == NULL)
+		return -1;
+
+	if (__map_groups__create_kernel_maps(self, vmlinux_maps, kernel) < 0)
 		return -1;
 
-	if (symbol_conf.use_modules &&
-	    perf_session__create_module_maps(self) < 0)
-		pr_debug("Failed to load list of modules for session %s, "
-			 "continuing...\n", self->filename);
+	if (symbol_conf.use_modules && map_groups__create_modules(self) < 0)
+		pr_debug("Problems creating module maps, continuing anyway...\n");
 	/*
 	 * Now that we have all the maps created, just set the ->end of them:
 	 */
-	map_groups__fixup_end(&self->kmaps);
+	map_groups__fixup_end(self);
 	return 0;
 }
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index 8aded2356f7..280dadd32a0 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -8,6 +8,8 @@
 #include <linux/rbtree.h>
 #include "event.h"
 
+#define DEBUG_CACHE_DIR ".debug"
+
 #ifdef HAVE_CPLUS_DEMANGLE
 extern char *cplus_demangle(const char *, int);
 
@@ -49,6 +51,8 @@ struct symbol {
 	char		name[0];
 };
 
+void symbol__delete(struct symbol *self);
+
 struct strlist;
 
 struct symbol_conf {
@@ -58,7 +62,8 @@ struct symbol_conf {
 			sort_by_name,
 			show_nr_samples,
 			use_callchain,
-			exclude_other;
+			exclude_other,
+			full_paths;
 	const char	*vmlinux_name,
 			*field_sep;
 	char            *dso_list_str,
@@ -77,6 +82,12 @@ static inline void *symbol__priv(struct symbol *self)
 	return ((void *)self) - symbol_conf.priv_size;
 }
 
+struct ref_reloc_sym {
+	const char	*name;
+	u64		addr;
+	u64		unrelocated_addr;
+};
+
 struct addr_location {
 	struct thread *thread;
 	struct map    *map;
@@ -94,6 +105,7 @@ struct dso {
 	u8		 slen_calculated:1;
 	u8		 has_build_id:1;
 	u8		 kernel:1;
+	u8		 hit:1;
 	unsigned char	 origin;
 	u8		 sorted_by_name;
 	u8		 loaded;
@@ -105,37 +117,55 @@ struct dso {
 };
 
 struct dso *dso__new(const char *name);
+struct dso *dso__new_kernel(const char *name);
 void dso__delete(struct dso *self);
 
 bool dso__loaded(const struct dso *self, enum map_type type);
 bool dso__sorted_by_name(const struct dso *self, enum map_type type);
 
+static inline void dso__set_loaded(struct dso *self, enum map_type type)
+{
+	self->loaded |= (1 << type);
+}
+
 void dso__sort_by_name(struct dso *self, enum map_type type);
 
-struct perf_session;
+extern struct list_head dsos__user, dsos__kernel;
+
+struct dso *__dsos__findnew(struct list_head *head, const char *name);
+
+static inline struct dso *dsos__findnew(const char *name)
+{
+	return __dsos__findnew(&dsos__user, name);
+}
 
-struct dso *dsos__findnew(const char *name);
-int dso__load(struct dso *self, struct map *map, struct perf_session *session,
-	      symbol_filter_t filter);
+int dso__load(struct dso *self, struct map *map, symbol_filter_t filter);
+int dso__load_vmlinux_path(struct dso *self, struct map *map,
+			   symbol_filter_t filter);
+int dso__load_kallsyms(struct dso *self, const char *filename, struct map *map,
+		       symbol_filter_t filter);
 void dsos__fprintf(FILE *fp);
-size_t dsos__fprintf_buildid(FILE *fp);
+size_t dsos__fprintf_buildid(FILE *fp, bool with_hits);
 
 size_t dso__fprintf_buildid(struct dso *self, FILE *fp);
 size_t dso__fprintf(struct dso *self, enum map_type type, FILE *fp);
 char dso__symtab_origin(const struct dso *self);
+void dso__set_long_name(struct dso *self, char *name);
 void dso__set_build_id(struct dso *self, void *build_id);
+void dso__read_running_kernel_build_id(struct dso *self);
 struct symbol *dso__find_symbol(struct dso *self, enum map_type type, u64 addr);
 struct symbol *dso__find_symbol_by_name(struct dso *self, enum map_type type,
 					const char *name);
 
 int filename__read_build_id(const char *filename, void *bf, size_t size);
 int sysfs__read_build_id(const char *filename, void *bf, size_t size);
-bool dsos__read_build_ids(void);
-int build_id__sprintf(u8 *self, int len, char *bf);
+bool dsos__read_build_ids(bool with_hits);
+int build_id__sprintf(const u8 *self, int len, char *bf);
+int kallsyms__parse(const char *filename, void *arg,
+		    int (*process_symbol)(void *arg, const char *name,
+					  char type, u64 start));
 
 int symbol__init(void);
-int perf_session__create_kernel_maps(struct perf_session *self);
+bool symbol_type__is_a(char symbol_type, enum map_type map_type);
 
-extern struct list_head dsos__user, dsos__kernel;
-extern struct dso *vdso;
 #endif /* __PERF_SYMBOL */
diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c
index 4a08dcf50b6..21b92162282 100644
--- a/tools/perf/util/thread.c
+++ b/tools/perf/util/thread.c
@@ -31,12 +31,41 @@ static struct thread *thread__new(pid_t pid)
 	return self;
 }
 
+static void map_groups__flush(struct map_groups *self)
+{
+	int type;
+
+	for (type = 0; type < MAP__NR_TYPES; type++) {
+		struct rb_root *root = &self->maps[type];
+		struct rb_node *next = rb_first(root);
+
+		while (next) {
+			struct map *pos = rb_entry(next, struct map, rb_node);
+			next = rb_next(&pos->rb_node);
+			rb_erase(&pos->rb_node, root);
+			/*
+			 * We may have references to this map, for
+			 * instance in some hist_entry instances, so
+			 * just move them to a separate list.
+			 */
+			list_add_tail(&pos->node, &self->removed_maps[pos->type]);
+		}
+	}
+}
+
 int thread__set_comm(struct thread *self, const char *comm)
 {
+	int err;
+
 	if (self->comm)
 		free(self->comm);
 	self->comm = strdup(comm);
-	return self->comm ? 0 : -ENOMEM;
+	err = self->comm == NULL ? -ENOMEM : 0;
+	if (!err) {
+		self->comm_set = true;
+		map_groups__flush(&self->mg);
+	}
+	return err;
 }
 
 int thread__comm_len(struct thread *self)
@@ -50,11 +79,6 @@ int thread__comm_len(struct thread *self)
 	return self->comm_len;
 }
 
-static const char *map_type__name[MAP__NR_TYPES] = {
-	[MAP__FUNCTION] = "Functions",
-	[MAP__VARIABLE] = "Variables",
-};
-
 static size_t __map_groups__fprintf_maps(struct map_groups *self,
 					 enum map_type type, FILE *fp)
 {
@@ -255,11 +279,14 @@ int thread__fork(struct thread *self, struct thread *parent)
 {
 	int i;
 
-	if (self->comm)
-		free(self->comm);
-	self->comm = strdup(parent->comm);
-	if (!self->comm)
-		return -ENOMEM;
+	if (parent->comm_set) {
+		if (self->comm)
+			free(self->comm);
+		self->comm = strdup(parent->comm);
+		if (!self->comm)
+			return -ENOMEM;
+		self->comm_set = true;
+	}
 
 	for (i = 0; i < MAP__NR_TYPES; ++i)
 		if (map_groups__clone(&self->mg, &parent->mg, i) < 0)
@@ -282,14 +309,13 @@ size_t perf_session__fprintf(struct perf_session *self, FILE *fp)
 }
 
 struct symbol *map_groups__find_symbol(struct map_groups *self,
-				       struct perf_session *session,
 				       enum map_type type, u64 addr,
 				       symbol_filter_t filter)
 {
 	struct map *map = map_groups__find(self, type, addr);
 
 	if (map != NULL)
-		return map__find_symbol(map, session, map->map_ip(map, addr), filter);
+		return map__find_symbol(map, map->map_ip(map, addr), filter);
 
 	return NULL;
 }
diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h
index c206f72c888..0a28f39de54 100644
--- a/tools/perf/util/thread.h
+++ b/tools/perf/util/thread.h
@@ -15,6 +15,7 @@ struct thread {
 	struct map_groups	mg;
 	pid_t			pid;
 	char			shortname[3];
+	bool			comm_set;
 	char			*comm;
 	int			comm_len;
 };
@@ -48,23 +49,36 @@ static inline struct map *thread__find_map(struct thread *self,
 	return self ? map_groups__find(&self->mg, type, addr) : NULL;
 }
 
+void thread__find_addr_map(struct thread *self,
+			   struct perf_session *session, u8 cpumode,
+			   enum map_type type, u64 addr,
+			   struct addr_location *al);
+
 void thread__find_addr_location(struct thread *self,
 				struct perf_session *session, u8 cpumode,
 				enum map_type type, u64 addr,
 				struct addr_location *al,
 				symbol_filter_t filter);
 struct symbol *map_groups__find_symbol(struct map_groups *self,
-				       struct perf_session *session,
 				       enum map_type type, u64 addr,
 				       symbol_filter_t filter);
 
-static inline struct symbol *
-map_groups__find_function(struct map_groups *self, struct perf_session *session,
-			  u64 addr, symbol_filter_t filter)
+static inline struct symbol *map_groups__find_function(struct map_groups *self,
+						       u64 addr,
+						       symbol_filter_t filter)
 {
-	return map_groups__find_symbol(self, session, MAP__FUNCTION, addr, filter);
+	return map_groups__find_symbol(self, MAP__FUNCTION, addr, filter);
 }
 
 struct map *map_groups__find_by_name(struct map_groups *self,
 				     enum map_type type, const char *name);
+
+int __map_groups__create_kernel_maps(struct map_groups *self,
+				     struct map *vmlinux_maps[MAP__NR_TYPES],
+				     struct dso *kernel);
+int map_groups__create_kernel_maps(struct map_groups *self,
+				   struct map *vmlinux_maps[MAP__NR_TYPES]);
+
+struct map *map_groups__new_module(struct map_groups *self, u64 start,
+				   const char *filename);
 #endif	/* __PERF_THREAD_H */
diff --git a/tools/perf/util/trace-event-info.c b/tools/perf/util/trace-event-info.c
index cace3559553..5ea8973ad33 100644
--- a/tools/perf/util/trace-event-info.c
+++ b/tools/perf/util/trace-event-info.c
@@ -20,6 +20,7 @@
  */
 #define _GNU_SOURCE
 #include <dirent.h>
+#include <mntent.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
@@ -37,6 +38,7 @@
 
 #include "../perf.h"
 #include "trace-event.h"
+#include "debugfs.h"
 
 #define VERSION "0.5"
 
@@ -101,32 +103,12 @@ void *malloc_or_die(unsigned int size)
 
 static const char *find_debugfs(void)
 {
-	static char debugfs[MAX_PATH+1];
-	static int debugfs_found;
-	char type[100];
-	FILE *fp;
-
-	if (debugfs_found)
-		return debugfs;
-
-	if ((fp = fopen("/proc/mounts","r")) == NULL)
-		die("Can't open /proc/mounts for read");
-
-	while (fscanf(fp, "%*s %"
-		      STR(MAX_PATH)
-		      "s %99s %*s %*d %*d\n",
-		      debugfs, type) == 2) {
-		if (strcmp(type, "debugfs") == 0)
-			break;
-	}
-	fclose(fp);
-
-	if (strcmp(type, "debugfs") != 0)
-		die("debugfs not mounted, please mount");
+	const char *path = debugfs_mount(NULL);
 
-	debugfs_found = 1;
+	if (!path)
+		die("Your kernel not support debugfs filesystem");
 
-	return debugfs;
+	return path;
 }
 
 /*
@@ -271,6 +253,8 @@ static void read_header_files(void)
 	write_or_die("header_page", 12);
 	write_or_die(&size, 8);
 	check_size = copy_file_fd(fd);
+	close(fd);
+
 	if (size != check_size)
 		die("wrong size for '%s' size=%lld read=%lld",
 		    path, size, check_size);
@@ -289,6 +273,7 @@ static void read_header_files(void)
 	if (size != check_size)
 		die("wrong size for '%s'", path);
 	put_tracing_file(path);
+	close(fd);
 }
 
 static bool name_in_tp_list(char *sys, struct tracepoint_path *tps)
@@ -317,7 +302,8 @@ static void copy_event_system(const char *sys, struct tracepoint_path *tps)
 		die("can't read directory '%s'", sys);
 
 	while ((dent = readdir(dir))) {
-		if (strcmp(dent->d_name, ".") == 0 ||
+		if (dent->d_type != DT_DIR ||
+		    strcmp(dent->d_name, ".") == 0 ||
 		    strcmp(dent->d_name, "..") == 0 ||
 		    !name_in_tp_list(dent->d_name, tps))
 			continue;
@@ -334,7 +320,8 @@ static void copy_event_system(const char *sys, struct tracepoint_path *tps)
 
 	rewinddir(dir);
 	while ((dent = readdir(dir))) {
-		if (strcmp(dent->d_name, ".") == 0 ||
+		if (dent->d_type != DT_DIR ||
+		    strcmp(dent->d_name, ".") == 0 ||
 		    strcmp(dent->d_name, "..") == 0 ||
 		    !name_in_tp_list(dent->d_name, tps))
 			continue;
@@ -353,6 +340,7 @@ static void copy_event_system(const char *sys, struct tracepoint_path *tps)
 
 		free(format);
 	}
+	closedir(dir);
 }
 
 static void read_ftrace_files(struct tracepoint_path *tps)
@@ -394,26 +382,21 @@ static void read_event_files(struct tracepoint_path *tps)
 		die("can't read directory '%s'", path);
 
 	while ((dent = readdir(dir))) {
-		if (strcmp(dent->d_name, ".") == 0 ||
+		if (dent->d_type != DT_DIR ||
+		    strcmp(dent->d_name, ".") == 0 ||
 		    strcmp(dent->d_name, "..") == 0 ||
 		    strcmp(dent->d_name, "ftrace") == 0 ||
 		    !system_in_tp_list(dent->d_name, tps))
 			continue;
-		sys = malloc_or_die(strlen(path) + strlen(dent->d_name) + 2);
-		sprintf(sys, "%s/%s", path, dent->d_name);
-		ret = stat(sys, &st);
-		free(sys);
-		if (ret < 0)
-			continue;
-		if (S_ISDIR(st.st_mode))
-			count++;
+		count++;
 	}
 
 	write_or_die(&count, 4);
 
 	rewinddir(dir);
 	while ((dent = readdir(dir))) {
-		if (strcmp(dent->d_name, ".") == 0 ||
+		if (dent->d_type != DT_DIR ||
+		    strcmp(dent->d_name, ".") == 0 ||
 		    strcmp(dent->d_name, "..") == 0 ||
 		    strcmp(dent->d_name, "ftrace") == 0 ||
 		    !system_in_tp_list(dent->d_name, tps))
@@ -422,14 +405,13 @@ static void read_event_files(struct tracepoint_path *tps)
 		sprintf(sys, "%s/%s", path, dent->d_name);
 		ret = stat(sys, &st);
 		if (ret >= 0) {
-			if (S_ISDIR(st.st_mode)) {
-				write_or_die(dent->d_name, strlen(dent->d_name) + 1);
-				copy_event_system(sys, tps);
-			}
+			write_or_die(dent->d_name, strlen(dent->d_name) + 1);
+			copy_event_system(sys, tps);
 		}
 		free(sys);
 	}
 
+	closedir(dir);
 	put_tracing_file(path);
 }
 
@@ -533,7 +515,7 @@ int read_tracing_data(int fd, struct perf_event_attr *pattrs, int nb_events)
 	write_or_die(buf, 1);
 
 	/* save page_size */
-	page_size = getpagesize();
+	page_size = sysconf(_SC_PAGESIZE);
 	write_or_die(&page_size, 4);
 
 	read_header_files();
diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c
index c5c32be040b..9b3c20f42f9 100644
--- a/tools/perf/util/trace-event-parse.c
+++ b/tools/perf/util/trace-event-parse.c
@@ -1925,6 +1925,15 @@ void *raw_field_ptr(struct event *event, const char *name, void *data)
 	if (!field)
 		return NULL;
 
+	if (field->flags & FIELD_IS_STRING) {
+		int offset;
+
+		offset = *(int *)(data + field->offset);
+		offset &= 0xffff;
+
+		return data + offset;
+	}
+
 	return data + field->offset;
 }
 
@@ -3277,3 +3286,18 @@ void parse_set_info(int nr_cpus, int long_sz)
 	cpus = nr_cpus;
 	long_size = long_sz;
 }
+
+int common_pc(struct scripting_context *context)
+{
+	return parse_common_pc(context->event_data);
+}
+
+int common_flags(struct scripting_context *context)
+{
+	return parse_common_flags(context->event_data);
+}
+
+int common_lock_depth(struct scripting_context *context)
+{
+	return parse_common_lock_depth(context->event_data);
+}
diff --git a/tools/perf/util/trace-event-perl.h b/tools/perf/util/trace-event-perl.h
deleted file mode 100644
index e88fb26137b..00000000000
--- a/tools/perf/util/trace-event-perl.h
+++ /dev/null
@@ -1,55 +0,0 @@
-#ifndef __PERF_TRACE_EVENT_PERL_H
-#define __PERF_TRACE_EVENT_PERL_H
-#ifdef NO_LIBPERL
-typedef int INTERP;
-#define dSP
-#define ENTER
-#define SAVETMPS
-#define PUTBACK
-#define SPAGAIN
-#define FREETMPS
-#define LEAVE
-#define SP
-#define ERRSV
-#define G_SCALAR		(0)
-#define G_DISCARD		(0)
-#define G_NOARGS		(0)
-#define PUSHMARK(a)
-#define SvTRUE(a)		(0)
-#define XPUSHs(s)
-#define sv_2mortal(a)
-#define newSVpv(a,b)
-#define newSVuv(a)
-#define newSViv(a)
-#define get_cv(a,b)		(0)
-#define call_pv(a,b)		(0)
-#define perl_alloc()		(0)
-#define perl_construct(a)	(0)
-#define perl_parse(a,b,c,d,e)	(0)
-#define perl_run(a)		(0)
-#define perl_destruct(a)	(0)
-#define perl_free(a)		(0)
-#define pTHX			void
-#define CV			void
-#define dXSUB_SYS
-#define pTHX_
-static inline void newXS(const char *a, void *b, const char *c) {}
-static void boot_Perf__Trace__Context(pTHX_ CV *cv) {}
-static void boot_DynaLoader(pTHX_ CV *cv) {}
-#else
-#include <EXTERN.h>
-#include <perl.h>
-void boot_Perf__Trace__Context(pTHX_ CV *cv);
-void boot_DynaLoader(pTHX_ CV *cv);
-typedef PerlInterpreter * INTERP;
-#endif
-
-struct scripting_context {
-	void *event_data;
-};
-
-int common_pc(struct scripting_context *context);
-int common_flags(struct scripting_context *context);
-int common_lock_depth(struct scripting_context *context);
-
-#endif /* __PERF_TRACE_EVENT_PERL_H */
diff --git a/tools/perf/util/trace-event-read.c b/tools/perf/util/trace-event-read.c
index 1744422cafc..7cd1193918c 100644
--- a/tools/perf/util/trace-event-read.c
+++ b/tools/perf/util/trace-event-read.c
@@ -18,7 +18,7 @@
  *
  * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  */
-#define _LARGEFILE64_SOURCE
+#define _FILE_OFFSET_BITS 64
 
 #include <dirent.h>
 #include <stdio.h>
@@ -83,7 +83,7 @@ static char *read_string(void)
 	char *str = NULL;
 	int size = 0;
 	int i;
-	int r;
+	off_t r;
 
 	for (;;) {
 		r = read(input_fd, buf, BUFSIZ);
@@ -118,7 +118,7 @@ static char *read_string(void)
 
 	/* move the file descriptor to the end of the string */
 	r = lseek(input_fd, -(r - i), SEEK_CUR);
-	if (r < 0)
+	if (r == (off_t)-1)
 		die("lseek");
 
 	if (str) {
@@ -282,8 +282,8 @@ static void update_cpu_data_index(int cpu)
 
 static void get_next_page(int cpu)
 {
-	off64_t save_seek;
-	off64_t ret;
+	off_t save_seek;
+	off_t ret;
 
 	if (!cpu_data[cpu].page)
 		return;
@@ -298,17 +298,17 @@ static void get_next_page(int cpu)
 		update_cpu_data_index(cpu);
 
 		/* other parts of the code may expect the pointer to not move */
-		save_seek = lseek64(input_fd, 0, SEEK_CUR);
+		save_seek = lseek(input_fd, 0, SEEK_CUR);
 
-		ret = lseek64(input_fd, cpu_data[cpu].offset, SEEK_SET);
-		if (ret < 0)
+		ret = lseek(input_fd, cpu_data[cpu].offset, SEEK_SET);
+		if (ret == (off_t)-1)
 			die("failed to lseek");
 		ret = read(input_fd, cpu_data[cpu].page, page_size);
 		if (ret < 0)
 			die("failed to read page");
 
 		/* reset the file pointer back */
-		lseek64(input_fd, save_seek, SEEK_SET);
+		lseek(input_fd, save_seek, SEEK_SET);
 
 		return;
 	}
diff --git a/tools/perf/util/trace-event-scripting.c b/tools/perf/util/trace-event-scripting.c
new file mode 100644
index 00000000000..7ea983acfae
--- /dev/null
+++ b/tools/perf/util/trace-event-scripting.c
@@ -0,0 +1,167 @@
+/*
+ * trace-event-scripting.  Scripting engine common and initialization code.
+ *
+ * Copyright (C) 2009-2010 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+#include <errno.h>
+
+#include "../perf.h"
+#include "util.h"
+#include "trace-event.h"
+
+struct scripting_context *scripting_context;
+
+static int stop_script_unsupported(void)
+{
+	return 0;
+}
+
+static void process_event_unsupported(int cpu __unused,
+				      void *data __unused,
+				      int size __unused,
+				      unsigned long long nsecs __unused,
+				      char *comm __unused)
+{
+}
+
+static void print_python_unsupported_msg(void)
+{
+	fprintf(stderr, "Python scripting not supported."
+		"  Install libpython and rebuild perf to enable it.\n"
+		"For example:\n  # apt-get install python-dev (ubuntu)"
+		"\n  # yum install python-devel (Fedora)"
+		"\n  etc.\n");
+}
+
+static int python_start_script_unsupported(const char *script __unused,
+					   int argc __unused,
+					   const char **argv __unused)
+{
+	print_python_unsupported_msg();
+
+	return -1;
+}
+
+static int python_generate_script_unsupported(const char *outfile __unused)
+{
+	print_python_unsupported_msg();
+
+	return -1;
+}
+
+struct scripting_ops python_scripting_unsupported_ops = {
+	.name = "Python",
+	.start_script = python_start_script_unsupported,
+	.stop_script = stop_script_unsupported,
+	.process_event = process_event_unsupported,
+	.generate_script = python_generate_script_unsupported,
+};
+
+static void register_python_scripting(struct scripting_ops *scripting_ops)
+{
+	int err;
+	err = script_spec_register("Python", scripting_ops);
+	if (err)
+		die("error registering Python script extension");
+
+	err = script_spec_register("py", scripting_ops);
+	if (err)
+		die("error registering py script extension");
+
+	scripting_context = malloc(sizeof(struct scripting_context));
+}
+
+#ifdef NO_LIBPYTHON
+void setup_python_scripting(void)
+{
+	register_python_scripting(&python_scripting_unsupported_ops);
+}
+#else
+struct scripting_ops python_scripting_ops;
+
+void setup_python_scripting(void)
+{
+	register_python_scripting(&python_scripting_ops);
+}
+#endif
+
+static void print_perl_unsupported_msg(void)
+{
+	fprintf(stderr, "Perl scripting not supported."
+		"  Install libperl and rebuild perf to enable it.\n"
+		"For example:\n  # apt-get install libperl-dev (ubuntu)"
+		"\n  # yum install 'perl(ExtUtils::Embed)' (Fedora)"
+		"\n  etc.\n");
+}
+
+static int perl_start_script_unsupported(const char *script __unused,
+					 int argc __unused,
+					 const char **argv __unused)
+{
+	print_perl_unsupported_msg();
+
+	return -1;
+}
+
+static int perl_generate_script_unsupported(const char *outfile __unused)
+{
+	print_perl_unsupported_msg();
+
+	return -1;
+}
+
+struct scripting_ops perl_scripting_unsupported_ops = {
+	.name = "Perl",
+	.start_script = perl_start_script_unsupported,
+	.stop_script = stop_script_unsupported,
+	.process_event = process_event_unsupported,
+	.generate_script = perl_generate_script_unsupported,
+};
+
+static void register_perl_scripting(struct scripting_ops *scripting_ops)
+{
+	int err;
+	err = script_spec_register("Perl", scripting_ops);
+	if (err)
+		die("error registering Perl script extension");
+
+	err = script_spec_register("pl", scripting_ops);
+	if (err)
+		die("error registering pl script extension");
+
+	scripting_context = malloc(sizeof(struct scripting_context));
+}
+
+#ifdef NO_LIBPERL
+void setup_perl_scripting(void)
+{
+	register_perl_scripting(&perl_scripting_unsupported_ops);
+}
+#else
+struct scripting_ops perl_scripting_ops;
+
+void setup_perl_scripting(void)
+{
+	register_perl_scripting(&perl_scripting_ops);
+}
+#endif
diff --git a/tools/perf/util/trace-event.h b/tools/perf/util/trace-event.h
index 6ad405620c9..c3269b937db 100644
--- a/tools/perf/util/trace-event.h
+++ b/tools/perf/util/trace-event.h
@@ -279,7 +279,15 @@ struct scripting_ops {
 
 int script_spec_register(const char *spec, struct scripting_ops *ops);
 
-extern struct scripting_ops perl_scripting_ops;
 void setup_perl_scripting(void);
+void setup_python_scripting(void);
+
+struct scripting_context {
+	void *event_data;
+};
+
+int common_pc(struct scripting_context *context);
+int common_flags(struct scripting_context *context);
+int common_lock_depth(struct scripting_context *context);
 
 #endif /* __PERF_TRACE_EVENTS_H */
diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c
new file mode 100644
index 00000000000..f9b890fde68
--- /dev/null
+++ b/tools/perf/util/util.c
@@ -0,0 +1,94 @@
+#include "util.h"
+#include <sys/mman.h>
+
+int mkdir_p(char *path, mode_t mode)
+{
+	struct stat st;
+	int err;
+	char *d = path;
+
+	if (*d != '/')
+		return -1;
+
+	if (stat(path, &st) == 0)
+		return 0;
+
+	while (*++d == '/');
+
+	while ((d = strchr(d, '/'))) {
+		*d = '\0';
+		err = stat(path, &st) && mkdir(path, mode);
+		*d++ = '/';
+		if (err)
+			return -1;
+		while (*d == '/')
+			++d;
+	}
+	return (stat(path, &st) && mkdir(path, mode)) ? -1 : 0;
+}
+
+static int slow_copyfile(const char *from, const char *to)
+{
+	int err = 0;
+	char *line = NULL;
+	size_t n;
+	FILE *from_fp = fopen(from, "r"), *to_fp;
+
+	if (from_fp == NULL)
+		goto out;
+
+	to_fp = fopen(to, "w");
+	if (to_fp == NULL)
+		goto out_fclose_from;
+
+	while (getline(&line, &n, from_fp) > 0)
+		if (fputs(line, to_fp) == EOF)
+			goto out_fclose_to;
+	err = 0;
+out_fclose_to:
+	fclose(to_fp);
+	free(line);
+out_fclose_from:
+	fclose(from_fp);
+out:
+	return err;
+}
+
+int copyfile(const char *from, const char *to)
+{
+	int fromfd, tofd;
+	struct stat st;
+	void *addr;
+	int err = -1;
+
+	if (stat(from, &st))
+		goto out;
+
+	if (st.st_size == 0) /* /proc? do it slowly... */
+		return slow_copyfile(from, to);
+
+	fromfd = open(from, O_RDONLY);
+	if (fromfd < 0)
+		goto out;
+
+	tofd = creat(to, 0755);
+	if (tofd < 0)
+		goto out_close_from;
+
+	addr = mmap(NULL, st.st_size, PROT_READ, MAP_PRIVATE, fromfd, 0);
+	if (addr == MAP_FAILED)
+		goto out_close_to;
+
+	if (write(tofd, addr, st.st_size) == st.st_size)
+		err = 0;
+
+	munmap(addr, st.st_size);
+out_close_to:
+	close(tofd);
+	if (err)
+		unlink(to);
+out_close_from:
+	close(fromfd);
+out:
+	return err;
+}
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index c673d882588..0f5b2a6f108 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -403,4 +403,7 @@ void git_qsort(void *base, size_t nmemb, size_t size,
 #endif
 #endif
 
+int mkdir_p(char *path, mode_t mode);
+int copyfile(const char *from, const char *to);
+
 #endif
diff --git a/tools/perf/util/values.c b/tools/perf/util/values.c
index 1c15e39f99e..cfa55d686e3 100644
--- a/tools/perf/util/values.c
+++ b/tools/perf/util/values.c
@@ -169,6 +169,7 @@ static void perf_read_values__display_pretty(FILE *fp,
 				counterwidth[j], values->value[i][j]);
 		fprintf(fp, "\n");
 	}
+	free(counterwidth);
 }
 
 static void perf_read_values__display_raw(FILE *fp,