Merge branch 'master' of /home/davem/src/GIT/linux-2.6/

63 files changed, 3438 insertions, 1983 deletions

diff --git a/tools/perf/Documentation/perf-diff.txt b/tools/perf/Documentation/perf-diff.txt
new file mode 100644
index 00000000000..8974e208cba
--- /dev/null
+++ b/tools/perf/Documentation/perf-diff.txt
@@ -0,0 +1,55 @@
+perf-diff(1)
+==============
+
+NAME
+----
+perf-diff - Read two perf.data files and display the differential profile
+
+SYNOPSIS
+--------
+[verse]
+'perf diff' [oldfile] [newfile]
+
+DESCRIPTION
+-----------
+This command displays the performance difference amongst two perf.data files
+captured via perf record.
+
+If no parameters are passed it will assume perf.data.old and perf.data.
+
+OPTIONS
+-------
+-d::
+--dsos=::
+	Only consider symbols in these dsos. CSV that understands
+	file://filename entries.
+
+-C::
+--comms=::
+	Only consider symbols in these comms. CSV that understands
+	file://filename entries.
+
+-S::
+--symbols=::
+	Only consider these symbols. CSV that understands
+	file://filename entries.
+
+-s::
+--sort=::
+	Sort by key(s): pid, comm, dso, symbol.
+
+-t::
+--field-separator=::
+
+	Use a special separator character and don't pad with spaces, replacing
+	all occurances of this separator in symbol names (and other output)
+	with a '.' character, that thus it's the only non valid separator.
+
+-v::
+--verbose::
+	Be verbose, for instance, show the raw counts in addition to the
+	diff.
+
+SEE ALSO
+--------
+linkperf:perf-record[1]
diff --git a/tools/perf/Documentation/perf-kmem.txt b/tools/perf/Documentation/perf-kmem.txt
index 44b0ce35c28..eac4d852e7c 100644
--- a/tools/perf/Documentation/perf-kmem.txt
+++ b/tools/perf/Documentation/perf-kmem.txt
@@ -8,16 +8,16 @@ perf-kmem - Tool to trace/measure kernel memory(slab) properties
 SYNOPSIS
 --------
 [verse]
-'perf kmem' {record} [<options>]
+'perf kmem' {record|stat} [<options>]
 
 DESCRIPTION
 -----------
-There's two variants of perf kmem:
+There are two variants of perf kmem:
 
   'perf kmem record <command>' to record the kmem events
   of an arbitrary workload.
 
-  'perf kmem' to report kernel memory statistics.
+  'perf kmem stat' to report kernel memory statistics.
 
 OPTIONS
 -------
@@ -25,8 +25,11 @@ OPTIONS
 --input=<file>::
 	Select the input file (default: perf.data)
 
---stat=<caller|alloc>::
-	Select per callsite or per allocation statistics
+--caller::
+	Show per-callsite statistics
+
+--alloc::
+	Show per-allocation statistics
 
 -s <key[,key2...]>::
 --sort=<key[,key2...]>::
diff --git a/tools/perf/Documentation/perf-probe.txt b/tools/perf/Documentation/perf-probe.txt
index 9270594e6df..250e391b4bc 100644
--- a/tools/perf/Documentation/perf-probe.txt
+++ b/tools/perf/Documentation/perf-probe.txt
@@ -8,10 +8,13 @@ perf-probe - Define new dynamic tracepoints
 SYNOPSIS
 --------
 [verse]
-'perf probe' [options] --add 'PROBE' [--add 'PROBE' ...]
+'perf probe' [options] --add='PROBE' [...]
 or
-'perf probe' [options] 'PROBE' ['PROBE' ...]
-
+'perf probe' [options] PROBE
+or
+'perf probe' [options] --del='[GROUP:]EVENT' [...]
+or
+'perf probe' --list
 
 DESCRIPTION
 -----------
@@ -31,15 +34,24 @@ OPTIONS
         Be more verbose (show parsed arguments, etc).
 
 -a::
---add::
-	Define a probe point (see PROBE SYNTAX for detail)
+--add=::
+	Define a probe event (see PROBE SYNTAX for detail).
+
+-d::
+--del=::
+	Delete a probe event.
+
+-l::
+--list::
+	List up current probe events.
 
 PROBE SYNTAX
 ------------
 Probe points are defined by following syntax.
 
- "FUNC[+OFFS|:RLN|%return][@SRC]|SRC:ALN [ARG ...]"
+ "[EVENT=]FUNC[+OFFS|:RLN|%return][@SRC]|SRC:ALN [ARG ...]"
 
+'EVENT' specifies the name of new event, if omitted, it will be set the name of the probed function. Currently, event group name is set as 'probe'.
 'FUNC' specifies a probed function name, and it may have one of the following options; '+OFFS' is the offset from function entry address in bytes, 'RLN' is the relative-line number from function entry line, and '%return' means that it probes function return. In addition, 'SRC' specifies a source file which has that function.
 It is also possible to specify a probe point by the source line number by using 'SRC:ALN' syntax, where 'SRC' is the source file path and 'ALN' is the line number.
 'ARG' specifies the arguments of this probe point. You can use the name of local variable, or kprobe-tracer argument format (e.g. $retval, %ax, etc).
diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt
index 9dccb180b7a..abfabe9147a 100644
--- a/tools/perf/Documentation/perf-report.txt
+++ b/tools/perf/Documentation/perf-report.txt
@@ -39,6 +39,10 @@ OPTIONS
 	Only consider these symbols. CSV that understands
 	file://filename entries.
 
+-s::
+--sort=::
+	Sort by key(s): pid, comm, dso, symbol, parent.
+
 -w::
 --field-width=::
 	Force each column width to the provided list, for large terminal
diff --git a/tools/perf/Documentation/perf-trace.txt b/tools/perf/Documentation/perf-trace.txt
index 07065efa60e..60e5900da48 100644
--- a/tools/perf/Documentation/perf-trace.txt
+++ b/tools/perf/Documentation/perf-trace.txt
@@ -8,18 +8,43 @@ perf-trace - Read perf.data (created by perf record) and display trace output
 SYNOPSIS
 --------
 [verse]
-'perf trace' [-i <file> | --input=file] symbol_name
+'perf trace' {record <script> | report <script> [args] }
 
 DESCRIPTION
 -----------
 This command reads the input file and displays the trace recorded.
 
+There are several variants of perf trace:
+
+  'perf trace' to see a detailed trace of the workload that was
+  recorded.
+
+  'perf trace record <script>' to record the events required for 'perf
+  trace report'.  <script> is the name displayed in the output of
+  'perf trace --list' i.e. the actual script name minus any language
+  extension.
+
+  'perf trace report <script>' to run and display the results of
+  <script>.  <script> is the name displayed in the output of 'perf
+  trace --list' i.e. the actual script name minus any language
+  extension.  The perf.data output from a previous run of 'perf trace
+  record <script>' is used and should be present for this command to
+  succeed.
+
 OPTIONS
 -------
 -D::
 --dump-raw-trace=::
         Display verbose dump of the trace data.
 
+-L::
+--Latency=::
+        Show latency attributes (irqs/preemption disabled, etc).
+
+-l::
+--list=::
+        Display a list of available trace scripts.
+
 -s::
 --script=::
         Process trace data with the given script ([lang]:script[.ext]).
diff --git a/tools/perf/Makefile b/tools/perf/Makefile
index 23ec66098bd..2e7fa3a0680 100644
--- a/tools/perf/Makefile
+++ b/tools/perf/Makefile
@@ -237,8 +237,8 @@ lib = lib
 
 export prefix bindir sharedir sysconfdir
 
-CC = gcc
-AR = ar
+CC = $(CROSS_COMPILE)gcc
+AR = $(CROSS_COMPILE)ar
 RM = rm -f
 TAR = tar
 FIND = find
@@ -250,7 +250,19 @@ PTHREAD_LIBS = -lpthread
 # explicitly what architecture to check for. Fix this up for yours..
 SPARSE_FLAGS = -D__BIG_ENDIAN__ -D__powerpc__
 
-ifeq ($(shell sh -c "echo 'int foo(void) {char X[2]; return 3;}' | $(CC) -x c -c -Werror -fstack-protector-all - -o /dev/null "$(QUIET_STDERR)" && echo y"), y)
+ifeq ($(V), 2)
+	QUIET_STDERR = ">/dev/null"
+else
+	QUIET_STDERR = ">/dev/null 2>&1"
+endif
+
+BITBUCKET = "/dev/null"
+
+ifneq ($(shell sh -c "(echo '\#include <stdio.h>'; echo 'int main(void) { return puts(\"hi\"); }') | $(CC) -x c - $(ALL_CFLAGS) -o $(BITBUCKET) "$(QUIET_STDERR)" && echo y"), y)
+	BITBUCKET = .perf.dev.null
+endif
+
+ifeq ($(shell sh -c "echo 'int foo(void) {char X[2]; return 3;}' | $(CC) -x c -c -Werror -fstack-protector-all - -o $(BITBUCKET) "$(QUIET_STDERR)" && echo y"), y)
   CFLAGS := $(CFLAGS) -fstack-protector-all
 endif
 
@@ -343,23 +355,31 @@ LIB_H += util/include/linux/string.h
 LIB_H += util/include/linux/types.h
 LIB_H += util/include/asm/asm-offsets.h
 LIB_H += util/include/asm/bitops.h
+LIB_H += util/include/asm/bug.h
 LIB_H += util/include/asm/byteorder.h
 LIB_H += util/include/asm/swab.h
 LIB_H += util/include/asm/system.h
 LIB_H += util/include/asm/uaccess.h
 LIB_H += perf.h
+LIB_H += util/cache.h
+LIB_H += util/callchain.h
+LIB_H += util/debug.h
 LIB_H += util/debugfs.h
 LIB_H += util/event.h
+LIB_H += util/exec_cmd.h
 LIB_H += util/types.h
 LIB_H += util/levenshtein.h
 LIB_H += util/parse-options.h
 LIB_H += util/parse-events.h
 LIB_H += util/quote.h
 LIB_H += util/util.h
+LIB_H += util/header.h
 LIB_H += util/help.h
+LIB_H += util/session.h
 LIB_H += util/strbuf.h
 LIB_H += util/string.h
 LIB_H += util/strlist.h
+LIB_H += util/svghelper.h
 LIB_H += util/run-command.h
 LIB_H += util/sigchain.h
 LIB_H += util/symbol.h
@@ -368,7 +388,8 @@ LIB_H += util/values.h
 LIB_H += util/sort.h
 LIB_H += util/hist.h
 LIB_H += util/thread.h
-LIB_H += util/data_map.h
+LIB_H += util/trace-event.h
+LIB_H += util/trace-event-perl.h
 LIB_H += util/probe-finder.h
 LIB_H += util/probe-event.h
 
@@ -405,6 +426,7 @@ LIB_OBJS += util/callchain.o
 LIB_OBJS += util/values.o
 LIB_OBJS += util/debug.o
 LIB_OBJS += util/map.o
+LIB_OBJS += util/session.o
 LIB_OBJS += util/thread.o
 LIB_OBJS += util/trace-event-parse.o
 LIB_OBJS += util/trace-event-read.o
@@ -425,6 +447,7 @@ BUILTIN_OBJS += bench/sched-messaging.o
 BUILTIN_OBJS += bench/sched-pipe.o
 BUILTIN_OBJS += bench/mem-memcpy.o
 
+BUILTIN_OBJS += builtin-diff.o
 BUILTIN_OBJS += builtin-help.o
 BUILTIN_OBJS += builtin-sched.o
 BUILTIN_OBJS += builtin-buildid-list.o
@@ -440,11 +463,6 @@ BUILTIN_OBJS += builtin-kmem.o
 
 PERFLIBS = $(LIB_FILE)
 
-ifeq ($(V), 2)
-	QUIET_STDERR = ">/dev/null"
-else
-	QUIET_STDERR = ">/dev/null 2>&1"
-endif
 #
 # Platform specific tweaks
 #
@@ -472,30 +490,33 @@ ifeq ($(uname_S),Darwin)
 	PTHREAD_LIBS =
 endif
 
-ifeq ($(shell sh -c "(echo '\#include <libelf.h>'; echo 'int main(void) { Elf * elf = elf_begin(0, ELF_C_READ, 0); return (long)elf; }') | $(CC) -x c - $(ALL_CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -o /dev/null $(ALL_LDFLAGS) $(EXTLIBS) "$(QUIET_STDERR)" && echo y"), y)
-ifneq ($(shell sh -c "(echo '\#include <gnu/libc-version.h>'; echo 'int main(void) { const char * version = gnu_get_libc_version(); return (long)version; }') | $(CC) -x c - $(ALL_CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -o /dev/null $(ALL_LDFLAGS) $(EXTLIBS) "$(QUIET_STDERR)" && echo y"), y)
+ifeq ($(shell sh -c "(echo '\#include <libelf.h>'; echo 'int main(void) { Elf * elf = elf_begin(0, ELF_C_READ, 0); return (long)elf; }') | $(CC) -x c - $(ALL_CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -o $(BITBUCKET) $(ALL_LDFLAGS) $(EXTLIBS) "$(QUIET_STDERR)" && echo y"), y)
+ifneq ($(shell sh -c "(echo '\#include <gnu/libc-version.h>'; echo 'int main(void) { const char * version = gnu_get_libc_version(); return (long)version; }') | $(CC) -x c - $(ALL_CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -o $(BITBUCKET) $(ALL_LDFLAGS) $(EXTLIBS) "$(QUIET_STDERR)" && echo y"), y)
 	msg := $(error No gnu/libc-version.h found, please install glibc-dev[el]/glibc-static);
 endif
 
-	ifneq ($(shell sh -c "(echo '\#include <libelf.h>'; echo 'int main(void) { Elf * elf = elf_begin(0, ELF_C_READ_MMAP, 0); return (long)elf; }') | $(CC) -x c - $(ALL_CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -o /dev/null $(ALL_LDFLAGS) $(EXTLIBS) "$(QUIET_STDERR)" && echo y"), y)
+	ifneq ($(shell sh -c "(echo '\#include <libelf.h>'; echo 'int main(void) { Elf * elf = elf_begin(0, ELF_C_READ_MMAP, 0); return (long)elf; }') | $(CC) -x c - $(ALL_CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -o $(BITBUCKET) $(ALL_LDFLAGS) $(EXTLIBS) "$(QUIET_STDERR)" && echo y"), y)
 		BASIC_CFLAGS += -DLIBELF_NO_MMAP
 	endif
 else
 	msg := $(error No libelf.h/libelf found, please install libelf-dev/elfutils-libelf-devel and glibc-dev[el]);
 endif
 
-ifneq ($(shell sh -c "(echo '\#include <libdwarf/dwarf.h>'; echo '\#include <libdwarf/libdwarf.h>'; echo 'int main(void) { Dwarf_Debug dbg; Dwarf_Error err; Dwarf_Ranges *rng; dwarf_init(0, DW_DLC_READ, 0, 0, &dbg, &err); dwarf_get_ranges(dbg, 0, &rng, 0, 0, &err); return (long)dbg; }') | $(CC) -x c - $(ALL_CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -ldwarf -lelf -o /dev/null $(ALL_LDFLAGS) $(EXTLIBS) "$(QUIET_STDERR)" && echo y"), y)
+ifneq ($(shell sh -c "(echo '\#ifndef _MIPS_SZLONG'; echo '\#define _MIPS_SZLONG 0'; echo '\#endif'; echo '\#include <dwarf.h>'; echo '\#include <libdwarf.h>'; echo 'int main(void) { Dwarf_Debug dbg; Dwarf_Error err; Dwarf_Ranges *rng; dwarf_init(0, DW_DLC_READ, 0, 0, &dbg, &err); dwarf_get_ranges(dbg, 0, &rng, 0, 0, &err); return (long)dbg; }') | $(CC) -x c - $(ALL_CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -I/usr/include/libdwarf -ldwarf -lelf -o $(BITBUCKET) $(ALL_LDFLAGS) $(EXTLIBS) "$(QUIET_STDERR)" && echo y"), y)
 	msg := $(warning No libdwarf.h found or old libdwarf.h found, disables dwarf support. Please install libdwarf-dev/libdwarf-devel >= 20081231);
 	BASIC_CFLAGS += -DNO_LIBDWARF
 else
+	BASIC_CFLAGS += -I/usr/include/libdwarf
 	EXTLIBS += -lelf -ldwarf
 	LIB_OBJS += util/probe-finder.o
 endif
 
+ifndef NO_LIBPERL
 PERL_EMBED_LDOPTS = `perl -MExtUtils::Embed -e ldopts 2>/dev/null`
 PERL_EMBED_CCOPTS = `perl -MExtUtils::Embed -e ccopts 2>/dev/null`
+endif
 
-ifneq ($(shell sh -c "(echo '\#include <EXTERN.h>'; echo '\#include <perl.h>'; echo 'int main(void) { perl_alloc(); return 0; }') | $(CC) -x c - $(PERL_EMBED_CCOPTS) -o /dev/null $(PERL_EMBED_LDOPTS) > /dev/null 2>&1 && echo y"), y)
+ifneq ($(shell sh -c "(echo '\#include <EXTERN.h>'; echo '\#include <perl.h>'; echo 'int main(void) { perl_alloc(); return 0; }') | $(CC) -x c - $(PERL_EMBED_CCOPTS) -o $(BITBUCKET) $(PERL_EMBED_LDOPTS) > /dev/null 2>&1 && echo y"), y)
 	BASIC_CFLAGS += -DNO_LIBPERL
 else
 	ALL_LDFLAGS += $(PERL_EMBED_LDOPTS)
@@ -505,20 +526,20 @@ endif
 ifdef NO_DEMANGLE
 	BASIC_CFLAGS += -DNO_DEMANGLE
 else
-	has_bfd := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) $(EXTLIBS) -lbfd "$(QUIET_STDERR)" && echo y")
+	has_bfd := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o $(BITBUCKET) $(ALL_LDFLAGS) $(EXTLIBS) -lbfd "$(QUIET_STDERR)" && echo y")
 
 	ifeq ($(has_bfd),y)
 		EXTLIBS += -lbfd
 	else
-		has_bfd_iberty := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) $(EXTLIBS) -lbfd -liberty "$(QUIET_STDERR)" && echo y")
+		has_bfd_iberty := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o $(BITBUCKET) $(ALL_LDFLAGS) $(EXTLIBS) -lbfd -liberty "$(QUIET_STDERR)" && echo y")
 		ifeq ($(has_bfd_iberty),y)
 			EXTLIBS += -lbfd -liberty
 		else
-			has_bfd_iberty_z := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) $(EXTLIBS) -lbfd -liberty -lz "$(QUIET_STDERR)" && echo y")
+			has_bfd_iberty_z := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o $(BITBUCKET) $(ALL_LDFLAGS) $(EXTLIBS) -lbfd -liberty -lz "$(QUIET_STDERR)" && echo y")
 			ifeq ($(has_bfd_iberty_z),y)
 				EXTLIBS += -lbfd -liberty -lz
 			else
-				has_cplus_demangle := $(shell sh -c "(echo 'extern char *cplus_demangle(const char *, int);'; echo 'int main(void) { cplus_demangle(0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) $(EXTLIBS) -liberty "$(QUIET_STDERR)" && echo y")
+				has_cplus_demangle := $(shell sh -c "(echo 'extern char *cplus_demangle(const char *, int);'; echo 'int main(void) { cplus_demangle(0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o $(BITBUCKET) $(ALL_LDFLAGS) $(EXTLIBS) -liberty "$(QUIET_STDERR)" && echo y")
 				ifeq ($(has_cplus_demangle),y)
 					EXTLIBS += -liberty
 					BASIC_CFLAGS += -DHAVE_CPLUS_DEMANGLE
@@ -766,7 +787,7 @@ export TAR INSTALL DESTDIR SHELL_PATH
 
 SHELL = $(SHELL_PATH)
 
-all:: shell_compatibility_test $(ALL_PROGRAMS) $(BUILT_INS) $(OTHER_PROGRAMS) PERF-BUILD-OPTIONS
+all:: .perf.dev.null shell_compatibility_test $(ALL_PROGRAMS) $(BUILT_INS) $(OTHER_PROGRAMS) PERF-BUILD-OPTIONS
 ifneq (,$X)
 	$(foreach p,$(patsubst %$X,%,$(filter %$X,$(ALL_PROGRAMS) $(BUILT_INS) perf$X)), test '$p' -ef '$p$X' || $(RM) '$p';)
 endif
@@ -991,8 +1012,6 @@ install: all
 	$(INSTALL) scripts/perl/Perf-Trace-Util/lib/Perf/Trace/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util/lib/Perf/Trace'
 	$(INSTALL) scripts/perl/*.pl -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl'
 	$(INSTALL) scripts/perl/bin/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/bin'
-	$(INSTALL) scripts/perl/Perf-Trace-Util/Makefile.PL -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util'
-	$(INSTALL) scripts/perl/Perf-Trace-Util/README -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util'
 ifdef BUILT_INS
 	$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)'
 	$(INSTALL) $(BUILT_INS) '$(DESTDIR_SQ)$(perfexec_instdir_SQ)'
@@ -1095,6 +1114,11 @@ clean:
 .PHONY: .FORCE-PERF-VERSION-FILE TAGS tags cscope .FORCE-PERF-CFLAGS
 .PHONY: .FORCE-PERF-BUILD-OPTIONS
 
+.perf.dev.null:
+		touch .perf.dev.null
+
+.INTERMEDIATE:	.perf.dev.null
+
 ### Make sure built-ins do not have dups and listed in perf.c
 #
 check-builtins::
diff --git a/tools/perf/bench/sched-messaging.c b/tools/perf/bench/sched-messaging.c
index 605a2a959aa..81cee78181f 100644
--- a/tools/perf/bench/sched-messaging.c
+++ b/tools/perf/bench/sched-messaging.c
@@ -1,6 +1,6 @@
 /*
  *
- * builtin-bench-messaging.c
+ * sched-messaging.c
  *
  * messaging: Benchmark for scheduler and IPC mechanisms
  *
@@ -320,10 +320,12 @@ int bench_sched_messaging(int argc, const char **argv,
 		       num_groups, num_groups * 2 * num_fds,
 		       thread_mode ? "threads" : "processes");
 		printf(" %14s: %lu.%03lu [sec]\n", "Total time",
-		       diff.tv_sec, diff.tv_usec/1000);
+		       diff.tv_sec,
+		       (unsigned long) (diff.tv_usec/1000));
 		break;
 	case BENCH_FORMAT_SIMPLE:
-		printf("%lu.%03lu\n", diff.tv_sec, diff.tv_usec/1000);
+		printf("%lu.%03lu\n", diff.tv_sec,
+		       (unsigned long) (diff.tv_usec/1000));
 		break;
 	default:
 		/* reaching here is something disaster */
diff --git a/tools/perf/bench/sched-pipe.c b/tools/perf/bench/sched-pipe.c
index 238185f9797..4f77c7c2764 100644
--- a/tools/perf/bench/sched-pipe.c
+++ b/tools/perf/bench/sched-pipe.c
@@ -1,6 +1,6 @@
 /*
  *
- * builtin-bench-pipe.c
+ * sched-pipe.c
  *
  * pipe: Benchmark for pipe()
  *
@@ -87,7 +87,8 @@ int bench_sched_pipe(int argc, const char **argv,
 	if (pid) {
 		retpid = waitpid(pid, &wait_stat, 0);
 		assert((retpid == pid) && WIFEXITED(wait_stat));
-		return 0;
+	} else {
+		exit(0);
 	}
 
 	switch (bench_format) {
@@ -99,7 +100,8 @@ int bench_sched_pipe(int argc, const char **argv,
 		result_usec += diff.tv_usec;
 
 		printf(" %14s: %lu.%03lu [sec]\n\n", "Total time",
-		       diff.tv_sec, diff.tv_usec/1000);
+		       diff.tv_sec,
+		       (unsigned long) (diff.tv_usec/1000));
 
 		printf(" %14lf usecs/op\n",
 		       (double)result_usec / (double)loops);
@@ -110,7 +112,8 @@ int bench_sched_pipe(int argc, const char **argv,
 
 	case BENCH_FORMAT_SIMPLE:
 		printf("%lu.%03lu\n",
-		       diff.tv_sec, diff.tv_usec / 1000);
+		       diff.tv_sec,
+		       (unsigned long) (diff.tv_usec / 1000));
 		break;
 
 	default:
diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c
index 0bf2e8f9af5..593ff25006d 100644
--- a/tools/perf/builtin-annotate.c
+++ b/tools/perf/builtin-annotate.c
@@ -25,7 +25,7 @@
 #include "util/thread.h"
 #include "util/sort.h"
 #include "util/hist.h"
-#include "util/data_map.h"
+#include "util/session.h"
 
 static char		const *input_name = "perf.data";
 
@@ -51,11 +51,6 @@ struct sym_priv {
 	struct sym_ext	*ext;
 };
 
-static struct symbol_conf symbol_conf = {
-	.priv_size	  = sizeof(struct sym_priv),
-	.try_vmlinux_path = true,
-};
-
 static const char *sym_hist_filter;
 
 static int symbol_filter(struct map *map __used, struct symbol *sym)
@@ -121,30 +116,32 @@ static void hist_hit(struct hist_entry *he, u64 ip)
 			h->ip[offset]);
 }
 
-static int hist_entry__add(struct addr_location *al, u64 count)
+static int perf_session__add_hist_entry(struct perf_session *self,
+					struct addr_location *al, u64 count)
 {
 	bool hit;
-	struct hist_entry *he = __hist_entry__add(al, NULL, count, &hit);
+	struct hist_entry *he = __perf_session__add_hist_entry(self, al, NULL,
+							       count, &hit);
 	if (he == NULL)
 		return -ENOMEM;
 	hist_hit(he, al->addr);
 	return 0;
 }
 
-static int process_sample_event(event_t *event)
+static int process_sample_event(event_t *event, struct perf_session *session)
 {
 	struct addr_location al;
 
 	dump_printf("(IP, %d): %d: %p\n", event->header.misc,
 		    event->ip.pid, (void *)(long)event->ip.ip);
 
-	if (event__preprocess_sample(event, &al, symbol_filter) < 0) {
+	if (event__preprocess_sample(event, session, &al, symbol_filter) < 0) {
 		fprintf(stderr, "problem processing %d event, skipping it.\n",
 			event->header.type);
 		return -1;
 	}
 
-	if (hist_entry__add(&al, 1)) {
+	if (!al.filtered && perf_session__add_hist_entry(session, &al, 1)) {
 		fprintf(stderr, "problem incrementing symbol count, "
 				"skipping event\n");
 		return -1;
@@ -428,11 +425,11 @@ static void annotate_sym(struct hist_entry *he)
 		free_source_line(he, len);
 }
 
-static void find_annotations(void)
+static void perf_session__find_annotations(struct perf_session *self)
 {
 	struct rb_node *nd;
 
-	for (nd = rb_first(&output_hists); nd; nd = rb_next(nd)) {
+	for (nd = rb_first(&self->hists); nd; nd = rb_next(nd)) {
 		struct hist_entry *he = rb_entry(nd, struct hist_entry, rb_node);
 		struct sym_priv *priv;
 
@@ -453,7 +450,7 @@ static void find_annotations(void)
 	}
 }
 
-static struct perf_file_handler file_handler = {
+static struct perf_event_ops event_ops = {
 	.process_sample_event	= process_sample_event,
 	.process_mmap_event	= event__process_mmap,
 	.process_comm_event	= event__process_comm,
@@ -462,33 +459,33 @@ static struct perf_file_handler file_handler = {
 
 static int __cmd_annotate(void)
 {
-	struct perf_header *header;
-	struct thread *idle;
 	int ret;
+	struct perf_session *session;
 
-	idle = register_idle_thread();
-	register_perf_file_handler(&file_handler);
+	session = perf_session__new(input_name, O_RDONLY, force);
+	if (session == NULL)
+		return -ENOMEM;
 
-	ret = mmap_dispatch_perf_file(&header, input_name, 0, 0,
-				      &event__cwdlen, &event__cwd);
+	ret = perf_session__process_events(session, &event_ops);
 	if (ret)
-		return ret;
+		goto out_delete;
 
 	if (dump_trace) {
 		event__print_totals();
-		return 0;
+		goto out_delete;
 	}
 
 	if (verbose > 3)
-		threads__fprintf(stdout);
+		perf_session__fprintf(session, stdout);
 
 	if (verbose > 2)
 		dsos__fprintf(stdout);
 
-	collapse__resort();
-	output__resort(event__total[0]);
-
-	find_annotations();
+	perf_session__collapse_resort(session);
+	perf_session__output_resort(session, session->event_total[0]);
+	perf_session__find_annotations(session);
+out_delete:
+	perf_session__delete(session);
 
 	return ret;
 }
@@ -519,29 +516,17 @@ static const struct option options[] = {
 	OPT_END()
 };
 
-static void setup_sorting(void)
+int cmd_annotate(int argc, const char **argv, const char *prefix __used)
 {
-	char *tmp, *tok, *str = strdup(sort_order);
-
-	for (tok = strtok_r(str, ", ", &tmp);
-			tok; tok = strtok_r(NULL, ", ", &tmp)) {
-		if (sort_dimension__add(tok) < 0) {
-			error("Unknown --sort key: `%s'", tok);
-			usage_with_options(annotate_usage, options);
-		}
-	}
+	argc = parse_options(argc, argv, options, annotate_usage, 0);
 
-	free(str);
-}
+	symbol_conf.priv_size = sizeof(struct sym_priv);
+	symbol_conf.try_vmlinux_path = true;
 
-int cmd_annotate(int argc, const char **argv, const char *prefix __used)
-{
-	if (symbol__init(&symbol_conf) < 0)
+	if (symbol__init() < 0)
 		return -1;
 
-	argc = parse_options(argc, argv, options, annotate_usage, 0);
-
-	setup_sorting();
+	setup_sorting(annotate_usage, options);
 
 	if (argc) {
 		/*
diff --git a/tools/perf/builtin-bench.c b/tools/perf/builtin-bench.c
index e043eb83092..46996774e55 100644
--- a/tools/perf/builtin-bench.c
+++ b/tools/perf/builtin-bench.c
@@ -31,6 +31,9 @@ struct bench_suite {
 	const char *summary;
 	int (*fn)(int, const char **, const char *);
 };
+						\
+/* sentinel: easy for help */
+#define suite_all { "all", "test all suite (pseudo suite)", NULL }
 
 static struct bench_suite sched_suites[] = {
 	{ "messaging",
@@ -39,6 +42,7 @@ static struct bench_suite sched_suites[] = {
 	{ "pipe",
 	  "Flood of communication over pipe() between two processes",
 	  bench_sched_pipe      },
+	suite_all,
 	{ NULL,
 	  NULL,
 	  NULL                  }
@@ -48,6 +52,7 @@ static struct bench_suite mem_suites[] = {
 	{ "memcpy",
 	  "Simple memory copy in various ways",
 	  bench_mem_memcpy },
+	suite_all,
 	{ NULL,
 	  NULL,
 	  NULL             }
@@ -66,6 +71,9 @@ static struct bench_subsys subsystems[] = {
 	{ "mem",
 	  "memory access performance",
 	  mem_suites },
+	{ "all",		/* sentinel: easy for help */
+	  "test all subsystem (pseudo subsystem)",
+	  NULL },
 	{ NULL,
 	  NULL,
 	  NULL       }
@@ -75,11 +83,11 @@ static void dump_suites(int subsys_index)
 {
 	int i;
 
-	printf("List of available suites for %s...\n\n",
+	printf("# List of available suites for %s...\n\n",
 	       subsystems[subsys_index].name);
 
 	for (i = 0; subsystems[subsys_index].suites[i].name; i++)
-		printf("\t%s: %s\n",
+		printf("%14s: %s\n",
 		       subsystems[subsys_index].suites[i].name,
 		       subsystems[subsys_index].suites[i].summary);
 
@@ -110,10 +118,10 @@ static void print_usage(void)
 		printf("\t%s\n", bench_usage[i]);
 	printf("\n");
 
-	printf("List of available subsystems...\n\n");
+	printf("# List of available subsystems...\n\n");
 
 	for (i = 0; subsystems[i].name; i++)
-		printf("\t%s: %s\n",
+		printf("%14s: %s\n",
 		       subsystems[i].name, subsystems[i].summary);
 	printf("\n");
 }
@@ -131,6 +139,37 @@ static int bench_str2int(char *str)
 	return BENCH_FORMAT_UNKNOWN;
 }
 
+static void all_suite(struct bench_subsys *subsys)	  /* FROM HERE */
+{
+	int i;
+	const char *argv[2];
+	struct bench_suite *suites = subsys->suites;
+
+	argv[1] = NULL;
+	/*
+	 * TODO:
+	 * preparing preset parameters for
+	 * embedded, ordinary PC, HPC, etc...
+	 * will be helpful
+	 */
+	for (i = 0; suites[i].fn; i++) {
+		printf("# Running %s/%s benchmark...\n",
+		       subsys->name,
+		       suites[i].name);
+
+		argv[1] = suites[i].name;
+		suites[i].fn(1, argv, NULL);
+		printf("\n");
+	}
+}
+
+static void all_subsystem(void)
+{
+	int i;
+	for (i = 0; subsystems[i].suites; i++)
+		all_suite(&subsystems[i]);
+}
+
 int cmd_bench(int argc, const char **argv, const char *prefix __used)
 {
 	int i, j, status = 0;
@@ -155,6 +194,11 @@ int cmd_bench(int argc, const char **argv, const char *prefix __used)
 		goto end;
 	}
 
+	if (!strcmp(argv[0], "all")) {
+		all_subsystem();
+		goto end;
+	}
+
 	for (i = 0; subsystems[i].name; i++) {
 		if (strcmp(subsystems[i].name, argv[0]))
 			continue;
@@ -165,6 +209,11 @@ int cmd_bench(int argc, const char **argv, const char *prefix __used)
 			goto end;
 		}
 
+		if (!strcmp(argv[1], "all")) {
+			all_suite(&subsystems[i]);
+			goto end;
+		}
+
 		for (j = 0; subsystems[i].suites[j].name; j++) {
 			if (strcmp(subsystems[i].suites[j].name, argv[1]))
 				continue;
diff --git a/tools/perf/builtin-buildid-list.c b/tools/perf/builtin-buildid-list.c
index 7dee9d19ab7..1e99ac80691 100644
--- a/tools/perf/builtin-buildid-list.c
+++ b/tools/perf/builtin-buildid-list.c
@@ -9,17 +9,16 @@
 #include "builtin.h"
 #include "perf.h"
 #include "util/cache.h"
-#include "util/data_map.h"
 #include "util/debug.h"
-#include "util/header.h"
 #include "util/parse-options.h"
+#include "util/session.h"
 #include "util/symbol.h"
 
 static char const *input_name = "perf.data";
 static int force;
 
-static const char *const buildid_list_usage[] = {
-	"perf report [<options>]",
+static const char * const buildid_list_usage[] = {
+	"perf buildid-list [<options>]",
 	NULL
 };
 
@@ -55,56 +54,18 @@ static int perf_file_section__process_buildids(struct perf_file_section *self,
 static int __cmd_buildid_list(void)
 {
 	int err = -1;
-	struct perf_header *header;
-	struct perf_file_header f_header;
-	struct stat input_stat;
-	int input = open(input_name, O_RDONLY);
+	struct perf_session *session;
 
-	if (input < 0) {
-		pr_err("failed to open file: %s", input_name);
-		if (!strcmp(input_name, "perf.data"))
-			pr_err("  (try 'perf record' first)");
-		pr_err("\n");
-		goto out;
-	}
-
-	err = fstat(input, &input_stat);
-	if (err < 0) {
-		perror("failed to stat file");
-		goto out_close;
-	}
-
-	if (!force && input_stat.st_uid && (input_stat.st_uid != geteuid())) {
-		pr_err("file %s not owned by current user or root\n",
-		       input_name);
-		goto out_close;
-	}
-
-	if (!input_stat.st_size) {
-		pr_info("zero-sized file, nothing to do!\n");
-		goto out_close;
-	}
-
-	err = -1;
-	header = perf_header__new();
-	if (header == NULL)
-		goto out_close;
-
-	if (perf_file_header__read(&f_header, header, input) < 0) {
-		pr_warning("incompatible file format");
-		goto out_close;
-	}
+	session = perf_session__new(input_name, O_RDONLY, force);
+	if (session == NULL)
+		return -1;
 
-	err = perf_header__process_sections(header, input,
+	err = perf_header__process_sections(&session->header, session->fd,
 				         perf_file_section__process_buildids);
+	if (err >= 0)
+		dsos__fprintf_buildid(stdout);
 
-	if (err < 0)
-		goto out_close;
-
-	dsos__fprintf_buildid(stdout);
-out_close:
-	close(input);
-out:
+	perf_session__delete(session);
 	return err;
 }
 
diff --git a/tools/perf/builtin-diff.c b/tools/perf/builtin-diff.c
new file mode 100644
index 00000000000..bd71b8ceafb
--- /dev/null
+++ b/tools/perf/builtin-diff.c
@@ -0,0 +1,249 @@
+/*
+ * builtin-diff.c
+ *
+ * Builtin diff command: Analyze two perf.data input files, look up and read
+ * DSOs and symbol information, sort them and produce a diff.
+ */
+#include "builtin.h"
+
+#include "util/debug.h"
+#include "util/event.h"
+#include "util/hist.h"
+#include "util/session.h"
+#include "util/sort.h"
+#include "util/symbol.h"
+#include "util/util.h"
+
+#include <stdlib.h>
+
+static char const *input_old = "perf.data.old",
+		  *input_new = "perf.data";
+static char	  diff__default_sort_order[] = "dso,symbol";
+static int  force;
+static bool show_displacement;
+
+static int perf_session__add_hist_entry(struct perf_session *self,
+					struct addr_location *al, u64 count)
+{
+	bool hit;
+	struct hist_entry *he = __perf_session__add_hist_entry(self, al, NULL,
+							       count, &hit);
+	if (he == NULL)
+		return -ENOMEM;
+
+	if (hit)
+		he->count += count;
+
+	return 0;
+}
+
+static int diff__process_sample_event(event_t *event, struct perf_session *session)
+{
+	struct addr_location al;
+	struct sample_data data = { .period = 1, };
+
+	dump_printf("(IP, %d): %d: %p\n", event->header.misc,
+		    event->ip.pid, (void *)(long)event->ip.ip);
+
+	if (event__preprocess_sample(event, session, &al, NULL) < 0) {
+		pr_warning("problem processing %d event, skipping it.\n",
+			   event->header.type);
+		return -1;
+	}
+
+	if (al.filtered)
+		return 0;
+
+	event__parse_sample(event, session->sample_type, &data);
+
+	if (al.sym && perf_session__add_hist_entry(session, &al, data.period)) {
+		pr_warning("problem incrementing symbol count, skipping event\n");
+		return -1;
+	}
+
+	session->events_stats.total += data.period;
+	return 0;
+}
+
+static struct perf_event_ops event_ops = {
+	.process_sample_event = diff__process_sample_event,
+	.process_mmap_event   = event__process_mmap,
+	.process_comm_event   = event__process_comm,
+	.process_exit_event   = event__process_task,
+	.process_fork_event   = event__process_task,
+	.process_lost_event   = event__process_lost,
+};
+
+static void perf_session__insert_hist_entry_by_name(struct rb_root *root,
+						    struct hist_entry *he)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	struct hist_entry *iter;
+
+	while (*p != NULL) {
+		int cmp;
+		parent = *p;
+		iter = rb_entry(parent, struct hist_entry, rb_node);
+
+		cmp = strcmp(he->map->dso->name, iter->map->dso->name);
+		if (cmp > 0)
+			p = &(*p)->rb_left;
+		else if (cmp < 0)
+			p = &(*p)->rb_right;
+		else {
+			cmp = strcmp(he->sym->name, iter->sym->name);
+			if (cmp > 0)
+				p = &(*p)->rb_left;
+			else
+				p = &(*p)->rb_right;
+		}
+	}
+
+	rb_link_node(&he->rb_node, parent, p);
+	rb_insert_color(&he->rb_node, root);
+}
+
+static void perf_session__resort_by_name(struct perf_session *self)
+{
+	unsigned long position = 1;
+	struct rb_root tmp = RB_ROOT;
+	struct rb_node *next = rb_first(&self->hists);
+
+	while (next != NULL) {
+		struct hist_entry *n = rb_entry(next, struct hist_entry, rb_node);
+
+		next = rb_next(&n->rb_node);
+		rb_erase(&n->rb_node, &self->hists);
+		n->position = position++;
+		perf_session__insert_hist_entry_by_name(&tmp, n);
+	}
+
+	self->hists = tmp;
+}
+
+static struct hist_entry *
+perf_session__find_hist_entry_by_name(struct perf_session *self,
+				      struct hist_entry *he)
+{
+	struct rb_node *n = self->hists.rb_node;
+
+	while (n) {
+		struct hist_entry *iter = rb_entry(n, struct hist_entry, rb_node);
+		int cmp = strcmp(he->map->dso->name, iter->map->dso->name);
+
+		if (cmp > 0)
+			n = n->rb_left;
+		else if (cmp < 0)
+			n = n->rb_right;
+		else {
+			cmp = strcmp(he->sym->name, iter->sym->name);
+			if (cmp > 0)
+				n = n->rb_left;
+			else if (cmp < 0)
+				n = n->rb_right;
+			else
+				return iter;
+		}
+	}
+
+	return NULL;
+}
+
+static void perf_session__match_hists(struct perf_session *old_session,
+				      struct perf_session *new_session)
+{
+	struct rb_node *nd;
+
+	perf_session__resort_by_name(old_session);
+
+	for (nd = rb_first(&new_session->hists); nd; nd = rb_next(nd)) {
+		struct hist_entry *pos = rb_entry(nd, struct hist_entry, rb_node);
+		pos->pair = perf_session__find_hist_entry_by_name(old_session, pos);
+	}
+}
+
+static int __cmd_diff(void)
+{
+	int ret, i;
+	struct perf_session *session[2];
+
+	session[0] = perf_session__new(input_old, O_RDONLY, force);
+	session[1] = perf_session__new(input_new, O_RDONLY, force);
+	if (session[0] == NULL || session[1] == NULL)
+		return -ENOMEM;
+
+	for (i = 0; i < 2; ++i) {
+		ret = perf_session__process_events(session[i], &event_ops);
+		if (ret)
+			goto out_delete;
+		perf_session__output_resort(session[i], session[i]->events_stats.total);
+	}
+
+	perf_session__match_hists(session[0], session[1]);
+	perf_session__fprintf_hists(session[1], session[0],
+				    show_displacement, stdout);
+out_delete:
+	for (i = 0; i < 2; ++i)
+		perf_session__delete(session[i]);
+	return ret;
+}
+
+static const char * const diff_usage[] = {
+	"perf diff [<options>] [old_file] [new_file]",
+	NULL,
+};
+
+static const struct option options[] = {
+	OPT_BOOLEAN('v', "verbose", &verbose,
+		    "be more verbose (show symbol address, etc)"),
+	OPT_BOOLEAN('m', "displacement", &show_displacement,
+		    "Show position displacement relative to baseline"),
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+		    "dump raw trace in ASCII"),
+	OPT_BOOLEAN('f', "force", &force, "don't complain, do it"),
+	OPT_BOOLEAN('m', "modules", &symbol_conf.use_modules,
+		    "load module symbols - WARNING: use only with -k and LIVE kernel"),
+	OPT_BOOLEAN('P', "full-paths", &event_ops.full_paths,
+		    "Don't shorten the pathnames taking into account the cwd"),
+	OPT_STRING('d', "dsos", &symbol_conf.dso_list_str, "dso[,dso...]",
+		   "only consider symbols in these dsos"),
+	OPT_STRING('C', "comms", &symbol_conf.comm_list_str, "comm[,comm...]",
+		   "only consider symbols in these comms"),
+	OPT_STRING('S', "symbols", &symbol_conf.sym_list_str, "symbol[,symbol...]",
+		   "only consider these symbols"),
+	OPT_STRING('s', "sort", &sort_order, "key[,key2...]",
+		   "sort by key(s): pid, comm, dso, symbol, parent"),
+	OPT_STRING('t', "field-separator", &symbol_conf.field_sep, "separator",
+		   "separator for columns, no spaces will be added between "
+		   "columns '.' is reserved."),
+	OPT_END()
+};
+
+int cmd_diff(int argc, const char **argv, const char *prefix __used)
+{
+	sort_order = diff__default_sort_order;
+	argc = parse_options(argc, argv, options, diff_usage, 0);
+	if (argc) {
+		if (argc > 2)
+			usage_with_options(diff_usage, options);
+		if (argc == 2) {
+			input_old = argv[0];
+			input_new = argv[1];
+		} else
+			input_new = argv[0];
+	}
+
+	symbol_conf.exclude_other = false;
+	if (symbol__init() < 0)
+		return -1;
+
+	setup_sorting(diff_usage, options);
+	setup_pager();
+
+	sort_entry__setup_elide(&sort_dso, symbol_conf.dso_list, "dso", NULL);
+	sort_entry__setup_elide(&sort_comm, symbol_conf.comm_list, "comm", NULL);
+	sort_entry__setup_elide(&sort_sym, symbol_conf.sym_list, "symbol", NULL);
+
+	return __cmd_diff();
+}
diff --git a/tools/perf/builtin-kmem.c b/tools/perf/builtin-kmem.c
index 047fef74bd5..93c67bf53d2 100644
--- a/tools/perf/builtin-kmem.c
+++ b/tools/perf/builtin-kmem.c
@@ -6,12 +6,12 @@
 #include "util/symbol.h"
 #include "util/thread.h"
 #include "util/header.h"
+#include "util/session.h"
 
 #include "util/parse-options.h"
 #include "util/trace-event.h"
 
 #include "util/debug.h"
-#include "util/data_map.h"
 
 #include <linux/rbtree.h>
 
@@ -20,9 +20,6 @@ typedef int (*sort_fn_t)(struct alloc_stat *, struct alloc_stat *);
 
 static char const		*input_name = "perf.data";
 
-static struct perf_header	*header;
-static u64			sample_type;
-
 static int			alloc_flag;
 static int			caller_flag;
 
@@ -57,11 +54,6 @@ static struct rb_root root_caller_sorted;
 static unsigned long total_requested, total_allocated;
 static unsigned long nr_allocs, nr_cross_allocs;
 
-struct raw_event_sample {
-	u32 size;
-	char data[0];
-};
-
 #define PATH_SYS_NODE	"/sys/devices/system/node"
 
 static void init_cpunode_map(void)
@@ -145,7 +137,7 @@ static void insert_alloc_stat(unsigned long call_site, unsigned long ptr,
 	if (data && data->ptr == ptr) {
 		data->hit++;
 		data->bytes_req += bytes_req;
-		data->bytes_alloc += bytes_req;
+		data->bytes_alloc += bytes_alloc;
 	} else {
 		data = malloc(sizeof(*data));
 		if (!data)
@@ -185,7 +177,7 @@ static void insert_caller_stat(unsigned long call_site,
 	if (data && data->call_site == call_site) {
 		data->hit++;
 		data->bytes_req += bytes_req;
-		data->bytes_alloc += bytes_req;
+		data->bytes_alloc += bytes_alloc;
 	} else {
 		data = malloc(sizeof(*data));
 		if (!data)
@@ -201,7 +193,7 @@ static void insert_caller_stat(unsigned long call_site,
 	}
 }
 
-static void process_alloc_event(struct raw_event_sample *raw,
+static void process_alloc_event(void *data,
 				struct event *event,
 				int cpu,
 				u64 timestamp __used,
@@ -214,10 +206,10 @@ static void process_alloc_event(struct raw_event_sample *raw,
 	int bytes_alloc;
 	int node1, node2;
 
-	ptr = raw_field_value(event, "ptr", raw->data);
-	call_site = raw_field_value(event, "call_site", raw->data);
-	bytes_req = raw_field_value(event, "bytes_req", raw->data);
-	bytes_alloc = raw_field_value(event, "bytes_alloc", raw->data);
+	ptr = raw_field_value(event, "ptr", data);
+	call_site = raw_field_value(event, "call_site", data);
+	bytes_req = raw_field_value(event, "bytes_req", data);
+	bytes_alloc = raw_field_value(event, "bytes_alloc", data);
 
 	insert_alloc_stat(call_site, ptr, bytes_req, bytes_alloc, cpu);
 	insert_caller_stat(call_site, bytes_req, bytes_alloc);
@@ -227,7 +219,7 @@ static void process_alloc_event(struct raw_event_sample *raw,
 
 	if (node) {
 		node1 = cpunode_map[cpu];
-		node2 = raw_field_value(event, "node", raw->data);
+		node2 = raw_field_value(event, "node", data);
 		if (node1 != node2)
 			nr_cross_allocs++;
 	}
@@ -262,7 +254,7 @@ static struct alloc_stat *search_alloc_stat(unsigned long ptr,
 	return NULL;
 }
 
-static void process_free_event(struct raw_event_sample *raw,
+static void process_free_event(void *data,
 			       struct event *event,
 			       int cpu,
 			       u64 timestamp __used,
@@ -271,7 +263,7 @@ static void process_free_event(struct raw_event_sample *raw,
 	unsigned long ptr;
 	struct alloc_stat *s_alloc, *s_caller;
 
-	ptr = raw_field_value(event, "ptr", raw->data);
+	ptr = raw_field_value(event, "ptr", data);
 
 	s_alloc = search_alloc_stat(ptr, 0, &root_alloc_stat, ptr_cmp);
 	if (!s_alloc)
@@ -289,66 +281,53 @@ static void process_free_event(struct raw_event_sample *raw,
 }
 
 static void
-process_raw_event(event_t *raw_event __used, void *more_data,
+process_raw_event(event_t *raw_event __used, void *data,
 		  int cpu, u64 timestamp, struct thread *thread)
 {
-	struct raw_event_sample *raw = more_data;
 	struct event *event;
 	int type;
 
-	type = trace_parse_common_type(raw->data);
+	type = trace_parse_common_type(data);
 	event = trace_find_event(type);
 
 	if (!strcmp(event->name, "kmalloc") ||
 	    !strcmp(event->name, "kmem_cache_alloc")) {
-		process_alloc_event(raw, event, cpu, timestamp, thread, 0);
+		process_alloc_event(data, event, cpu, timestamp, thread, 0);
 		return;
 	}
 
 	if (!strcmp(event->name, "kmalloc_node") ||
 	    !strcmp(event->name, "kmem_cache_alloc_node")) {
-		process_alloc_event(raw, event, cpu, timestamp, thread, 1);
+		process_alloc_event(data, event, cpu, timestamp, thread, 1);
 		return;
 	}
 
 	if (!strcmp(event->name, "kfree") ||
 	    !strcmp(event->name, "kmem_cache_free")) {
-		process_free_event(raw, event, cpu, timestamp, thread);
+		process_free_event(data, event, cpu, timestamp, thread);
 		return;
 	}
 }
 
-static int process_sample_event(event_t *event)
+static int process_sample_event(event_t *event, struct perf_session *session)
 {
-	u64 ip = event->ip.ip;
-	u64 timestamp = -1;
-	u32 cpu = -1;
-	u64 period = 1;
-	void *more_data = event->ip.__more_data;
-	struct thread *thread = threads__findnew(event->ip.pid);
-
-	if (sample_type & PERF_SAMPLE_TIME) {
-		timestamp = *(u64 *)more_data;
-		more_data += sizeof(u64);
-	}
+	struct sample_data data;
+	struct thread *thread;
 
-	if (sample_type & PERF_SAMPLE_CPU) {
-		cpu = *(u32 *)more_data;
-		more_data += sizeof(u32);
-		more_data += sizeof(u32); /* reserved */
-	}
+	memset(&data, 0, sizeof(data));
+	data.time = -1;
+	data.cpu = -1;
+	data.period = 1;
 
-	if (sample_type & PERF_SAMPLE_PERIOD) {
-		period = *(u64 *)more_data;
-		more_data += sizeof(u64);
-	}
+	event__parse_sample(event, session->sample_type, &data);
 
 	dump_printf("(IP, %d): %d/%d: %p period: %Ld\n",
 		event->header.misc,
-		event->ip.pid, event->ip.tid,
-		(void *)(long)ip,
-		(long long)period);
+		data.pid, data.tid,
+		(void *)(long)data.ip,
+		(long long)data.period);
 
+	thread = perf_session__findnew(session, event->ip.pid);
 	if (thread == NULL) {
 		pr_debug("problem processing %d event, skipping it.\n",
 			 event->header.type);
@@ -357,16 +336,15 @@ static int process_sample_event(event_t *event)
 
 	dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid);
 
-	process_raw_event(event, more_data, cpu, timestamp, thread);
+	process_raw_event(event, data.raw_data, data.cpu,
+			  data.time, thread);
 
 	return 0;
 }
 
-static int sample_type_check(u64 type)
+static int sample_type_check(struct perf_session *session)
 {
-	sample_type = type;
-
-	if (!(sample_type & PERF_SAMPLE_RAW)) {
+	if (!(session->sample_type & PERF_SAMPLE_RAW)) {
 		fprintf(stderr,
 			"No trace sample to read. Did you call perf record "
 			"without -R?");
@@ -376,21 +354,12 @@ static int sample_type_check(u64 type)
 	return 0;
 }
 
-static struct perf_file_handler file_handler = {
+static struct perf_event_ops event_ops = {
 	.process_sample_event	= process_sample_event,
 	.process_comm_event	= event__process_comm,
 	.sample_type_check	= sample_type_check,
 };
 
-static int read_events(void)
-{
-	register_idle_thread();
-	register_perf_file_handler(&file_handler);
-
-	return mmap_dispatch_perf_file(&header, input_name, 0, 0,
-				       &event__cwdlen, &event__cwd);
-}
-
 static double fragmentation(unsigned long n_req, unsigned long n_alloc)
 {
 	if (n_alloc == 0)
@@ -399,13 +368,14 @@ static double fragmentation(unsigned long n_req, unsigned long n_alloc)
 		return 100.0 - (100.0 * n_req / n_alloc);
 }
 
-static void __print_result(struct rb_root *root, int n_lines, int is_caller)
+static void __print_result(struct rb_root *root, struct perf_session *session,
+			   int n_lines, int is_caller)
 {
 	struct rb_node *next;
 
 	printf("%.102s\n", graph_dotted_line);
 	printf(" %-34s |",  is_caller ? "Callsite": "Alloc Ptr");
-	printf(" Total_alloc/Per | Total_req/Per   | Hit   | Ping-pong | Frag\n");
+	printf(" Total_alloc/Per | Total_req/Per   | Hit      | Ping-pong | Frag\n");
 	printf("%.102s\n", graph_dotted_line);
 
 	next = rb_first(root);
@@ -420,7 +390,7 @@ static void __print_result(struct rb_root *root, int n_lines, int is_caller)
 		if (is_caller) {
 			addr = data->call_site;
 			if (!raw_ip)
-				sym = thread__find_function(kthread, addr, NULL);
+				sym = map_groups__find_function(&session->kmaps, session, addr, NULL);
 		} else
 			addr = data->ptr;
 
@@ -431,7 +401,7 @@ static void __print_result(struct rb_root *root, int n_lines, int is_caller)
 			snprintf(buf, sizeof(buf), "%#Lx", addr);
 		printf(" %-34s |", buf);
 
-		printf(" %9llu/%-5lu | %9llu/%-5lu | %6lu | %8lu | %6.3f%%\n",
+		printf(" %9llu/%-5lu | %9llu/%-5lu | %8lu | %8lu | %6.3f%%\n",
 		       (unsigned long long)data->bytes_alloc,
 		       (unsigned long)data->bytes_alloc / data->hit,
 		       (unsigned long long)data->bytes_req,
@@ -461,12 +431,12 @@ static void print_summary(void)
 	printf("Cross CPU allocations: %lu/%lu\n", nr_cross_allocs, nr_allocs);
 }
 
-static void print_result(void)
+static void print_result(struct perf_session *session)
 {
 	if (caller_flag)
-		__print_result(&root_caller_sorted, caller_lines, 1);
+		__print_result(&root_caller_sorted, session, caller_lines, 1);
 	if (alloc_flag)
-		__print_result(&root_alloc_sorted, alloc_lines, 0);
+		__print_result(&root_alloc_sorted, session, alloc_lines, 0);
 	print_summary();
 }
 
@@ -534,16 +504,24 @@ static void sort_result(void)
 
 static int __cmd_kmem(void)
 {
+	int err;
+	struct perf_session *session = perf_session__new(input_name, O_RDONLY, 0);
+	if (session == NULL)
+		return -ENOMEM;
+
 	setup_pager();
-	read_events();
+	err = perf_session__process_events(session, &event_ops);
+	if (err != 0)
+		goto out_delete;
 	sort_result();
-	print_result();
-
-	return 0;
+	print_result(session);
+out_delete:
+	perf_session__delete(session);
+	return err;
 }
 
 static const char * const kmem_usage[] = {
-	"perf kmem [<options>] {record}",
+	"perf kmem [<options>] {record|stat}",
 	NULL
 };
 
@@ -703,18 +681,17 @@ static int parse_sort_opt(const struct option *opt __used,
 	return 0;
 }
 
-static int parse_stat_opt(const struct option *opt __used,
-			  const char *arg, int unset __used)
+static int parse_caller_opt(const struct option *opt __used,
+			  const char *arg __used, int unset __used)
 {
-	if (!arg)
-		return -1;
+	caller_flag = (alloc_flag + 1);
+	return 0;
+}
 
-	if (strcmp(arg, "alloc") == 0)
-		alloc_flag = (caller_flag + 1);
-	else if (strcmp(arg, "caller") == 0)
-		caller_flag = (alloc_flag + 1);
-	else
-		return -1;
+static int parse_alloc_opt(const struct option *opt __used,
+			  const char *arg __used, int unset __used)
+{
+	alloc_flag = (caller_flag + 1);
 	return 0;
 }
 
@@ -739,14 +716,17 @@ static int parse_line_opt(const struct option *opt __used,
 static const struct option kmem_options[] = {
 	OPT_STRING('i', "input", &input_name, "file",
 		   "input file name"),
-	OPT_CALLBACK(0, "stat", NULL, "<alloc>|<caller>",
-		     "stat selector, Pass 'alloc' or 'caller'.",
-		     parse_stat_opt),
+	OPT_CALLBACK_NOOPT(0, "caller", NULL, NULL,
+			   "show per-callsite statistics",
+			   parse_caller_opt),
+	OPT_CALLBACK_NOOPT(0, "alloc", NULL, NULL,
+			   "show per-allocation statistics",
+			   parse_alloc_opt),
 	OPT_CALLBACK('s', "sort", NULL, "key[,key2...]",
 		     "sort by keys: ptr, call_site, bytes, hit, pingpong, frag",
 		     parse_sort_opt),
 	OPT_CALLBACK('l', "line", NULL, "num",
-		     "show n lins",
+		     "show n lines",
 		     parse_line_opt),
 	OPT_BOOLEAN(0, "raw-ip", &raw_ip, "show raw ip instead of symbol"),
 	OPT_END()
@@ -786,22 +766,27 @@ static int __cmd_record(int argc, const char **argv)
 
 int cmd_kmem(int argc, const char **argv, const char *prefix __used)
 {
-	symbol__init(0);
-
 	argc = parse_options(argc, argv, kmem_options, kmem_usage, 0);
 
-	if (argc && !strncmp(argv[0], "rec", 3))
-		return __cmd_record(argc, argv);
-	else if (argc)
+	if (!argc)
 		usage_with_options(kmem_usage, kmem_options);
 
-	if (list_empty(&caller_sort))
-		setup_sorting(&caller_sort, default_sort_order);
-	if (list_empty(&alloc_sort))
-		setup_sorting(&alloc_sort, default_sort_order);
+	symbol__init();
+
+	if (!strncmp(argv[0], "rec", 3)) {
+		return __cmd_record(argc, argv);
+	} else if (!strcmp(argv[0], "stat")) {
+		setup_cpunode_map();
 
-	setup_cpunode_map();
+		if (list_empty(&caller_sort))
+			setup_sorting(&caller_sort, default_sort_order);
+		if (list_empty(&alloc_sort))
+			setup_sorting(&alloc_sort, default_sort_order);
 
-	return __cmd_kmem();
+		return __cmd_kmem();
+	} else
+		usage_with_options(kmem_usage, kmem_options);
+
+	return 0;
 }
 
diff --git a/tools/perf/builtin-probe.c b/tools/perf/builtin-probe.c
index a58e11b7ea8..c1e6774fd3e 100644
--- a/tools/perf/builtin-probe.c
+++ b/tools/perf/builtin-probe.c
@@ -35,34 +35,33 @@
 #include "perf.h"
 #include "builtin.h"
 #include "util/util.h"
+#include "util/strlist.h"
 #include "util/event.h"
 #include "util/debug.h"
+#include "util/debugfs.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/session.h"
 #include "util/parse-options.h"
 #include "util/parse-events.h"	/* For debugfs_path */
 #include "util/probe-finder.h"
 #include "util/probe-event.h"
 
-/* Default vmlinux search paths */
-#define NR_SEARCH_PATH 3
-const char *default_search_path[NR_SEARCH_PATH] = {
-"/lib/modules/%s/build/vmlinux",		/* Custom build kernel */
-"/usr/lib/debug/lib/modules/%s/vmlinux",	/* Red Hat debuginfo */
-"/boot/vmlinux-debug-%s",			/* Ubuntu */
-};
-
 #define MAX_PATH_LEN 256
 #define MAX_PROBES 128
 
 /* Session management structure */
 static struct {
-	char *vmlinux;
-	char *release;
-	int need_dwarf;
+	bool need_dwarf;
+	bool list_events;
+	bool force_add;
 	int nr_probe;
 	struct probe_point probes[MAX_PROBES];
+	struct strlist *dellist;
+	struct perf_session *psession;
+	struct map *kmap;
 } session;
 
-static bool listing;
 
 /* Parse an event definition. Note that any error must die. */
 static void parse_probe_event(const char *str)
@@ -74,11 +73,30 @@ static void parse_probe_event(const char *str)
 		die("Too many probes (> %d) are specified.", MAX_PROBES);
 
 	/* Parse perf-probe event into probe_point */
-	session.need_dwarf = parse_perf_probe_event(str, pp);
+	parse_perf_probe_event(str, pp, &session.need_dwarf);
 
 	pr_debug("%d arguments\n", pp->nr_args);
 }
 
+static void parse_probe_event_argv(int argc, const char **argv)
+{
+	int i, len;
+	char *buf;
+
+	/* Bind up rest arguments */
+	len = 0;
+	for (i = 0; i < argc; i++)
+		len += strlen(argv[i]) + 1;
+	buf = zalloc(len + 1);
+	if (!buf)
+		die("Failed to allocate memory for binding arguments.");
+	len = 0;
+	for (i = 0; i < argc; i++)
+		len += sprintf(&buf[len], "%s ", argv[i]);
+	parse_probe_event(buf);
+	free(buf);
+}
+
 static int opt_add_probe_event(const struct option *opt __used,
 			      const char *str, int unset __used)
 {
@@ -87,40 +105,44 @@ static int opt_add_probe_event(const struct option *opt __used,
 	return 0;
 }
 
-#ifndef NO_LIBDWARF
-static int open_default_vmlinux(void)
+static int opt_del_probe_event(const struct option *opt __used,
+			       const char *str, int unset __used)
 {
-	struct utsname uts;
-	char fname[MAX_PATH_LEN];
-	int fd, ret, i;
-
-	ret = uname(&uts);
-	if (ret) {
-		pr_debug("uname() failed.\n");
-		return -errno;
+	if (str) {
+		if (!session.dellist)
+			session.dellist = strlist__new(true, NULL);
+		strlist__add(session.dellist, str);
 	}
-	session.release = uts.release;
-	for (i = 0; i < NR_SEARCH_PATH; i++) {
-		ret = snprintf(fname, MAX_PATH_LEN,
-			       default_search_path[i], session.release);
-		if (ret >= MAX_PATH_LEN || ret < 0) {
-			pr_debug("Filename(%d,%s) is too long.\n", i,
-				uts.release);
-			errno = E2BIG;
-			return -E2BIG;
-		}
-		pr_debug("try to open %s\n", fname);
-		fd = open(fname, O_RDONLY);
-		if (fd >= 0)
-			break;
+	return 0;
+}
+
+/* Currently just checking function name from symbol map */
+static void evaluate_probe_point(struct probe_point *pp)
+{
+	struct symbol *sym;
+	sym = map__find_symbol_by_name(session.kmap, pp->function,
+				       session.psession, NULL);
+	if (!sym)
+		die("Kernel symbol \'%s\' not found - probe not added.",
+		    pp->function);
+}
+
+#ifndef NO_LIBDWARF
+static int open_vmlinux(void)
+{
+	if (map__load(session.kmap, session.psession, NULL) < 0) {
+		pr_debug("Failed to load kernel map.\n");
+		return -EINVAL;
 	}
-	return fd;
+	pr_debug("Try to open %s\n", session.kmap->dso->long_name);
+	return open(session.kmap->dso->long_name, O_RDONLY);
 }
 #endif
 
 static const char * const probe_usage[] = {
 	"perf probe [<options>] 'PROBEDEF' ['PROBEDEF' ...]",
 	"perf probe [<options>] --add 'PROBEDEF' [--add 'PROBEDEF' ...]",
+	"perf probe [<options>] --del '[GROUP:]EVENT' ...",
 	"perf probe --list",
 	NULL
 };
@@ -129,19 +151,22 @@ static const struct option options[] = {
 	OPT_BOOLEAN('v', "verbose", &verbose,
 		    "be more verbose (show parsed arguments, etc)"),
 #ifndef NO_LIBDWARF
-	OPT_STRING('k', "vmlinux", &session.vmlinux, "file",
-		"vmlinux/module pathname"),
+	OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
+		   "file", "vmlinux pathname"),
 #endif
-	OPT_BOOLEAN('l', "list", &listing, "list up current probes"),
+	OPT_BOOLEAN('l', "list", &session.list_events,
+		    "list up current probe events"),
+	OPT_CALLBACK('d', "del", NULL, "[GROUP:]EVENT", "delete a probe event.",
+		opt_del_probe_event),
 	OPT_CALLBACK('a', "add", NULL,
 #ifdef NO_LIBDWARF
-		"FUNC[+OFFS|%return] [ARG ...]",
+		"[EVENT=]FUNC[+OFFS|%return] [ARG ...]",
 #else
-		"FUNC[+OFFS|%return|:RLN][@SRC]|SRC:ALN [ARG ...]",
+		"[EVENT=]FUNC[+OFFS|%return|:RLN][@SRC]|SRC:ALN [ARG ...]",
 #endif
 		"probe point definition, where\n"
-		"\t\tGRP:\tGroup name (optional)\n"
-		"\t\tNAME:\tEvent name\n"
+		"\t\tGROUP:\tGroup name (optional)\n"
+		"\t\tEVENT:\tEvent name\n"
 		"\t\tFUNC:\tFunction name\n"
 		"\t\tOFFS:\tOffset from function entry (in byte)\n"
 		"\t\t%return:\tPut the probe at function return\n"
@@ -155,12 +180,14 @@ static const struct option options[] = {
 #endif
 		"\t\t\tkprobe-tracer argument format.)\n",
 		opt_add_probe_event),
+	OPT_BOOLEAN('f', "force", &session.force_add, "forcibly add events"
+		    " with existing name"),
 	OPT_END()
 };
 
 int cmd_probe(int argc, const char **argv, const char *prefix __used)
 {
-	int i, j, ret;
+	int i, ret;
 #ifndef NO_LIBDWARF
 	int fd;
 #endif
@@ -168,54 +195,92 @@ int cmd_probe(int argc, const char **argv, const char *prefix __used)
 
 	argc = parse_options(argc, argv, options, probe_usage,
 			     PARSE_OPT_STOP_AT_NON_OPTION);
-	for (i = 0; i < argc; i++)
-		parse_probe_event(argv[i]);
+	if (argc > 0) {
+		if (strcmp(argv[0], "-") == 0) {
+			pr_warning("  Error: '-' is not supported.\n");
+			usage_with_options(probe_usage, options);
+		}
+		parse_probe_event_argv(argc, argv);
+	}
 
-	if ((session.nr_probe == 0 && !listing) ||
-	    (session.nr_probe != 0 && listing))
+	if ((!session.nr_probe && !session.dellist && !session.list_events))
 		usage_with_options(probe_usage, options);
 
-	if (listing) {
+	if (debugfs_valid_mountpoint(debugfs_path) < 0)
+		die("Failed to find debugfs path.");
+
+	if (session.list_events) {
+		if (session.nr_probe != 0 || session.dellist) {
+			pr_warning("  Error: Don't use --list with"
+				   " --add/--del.\n");
+			usage_with_options(probe_usage, options);
+		}
 		show_perf_probe_events();
 		return 0;
 	}
 
+	if (session.dellist) {
+		del_trace_kprobe_events(session.dellist);
+		strlist__delete(session.dellist);
+		if (session.nr_probe == 0)
+			return 0;
+	}
+
+	/* Initialize symbol maps for vmlinux */
+	symbol_conf.sort_by_name = true;
+	if (symbol_conf.vmlinux_name == NULL)
+		symbol_conf.try_vmlinux_path = true;
+	if (symbol__init() < 0)
+		die("Failed to init symbol map.");
+	session.psession = perf_session__new(NULL, O_WRONLY, false);
+	if (session.psession == NULL)
+		die("Failed to init perf_session.");
+	session.kmap = map_groups__find_by_name(&session.psession->kmaps,
+						MAP__FUNCTION,
+						"[kernel.kallsyms]");
+	if (!session.kmap)
+		die("Could not find kernel map.\n");
+
 	if (session.need_dwarf)
 #ifdef NO_LIBDWARF
 		die("Debuginfo-analysis is not supported");
 #else	/* !NO_LIBDWARF */
 		pr_debug("Some probes require debuginfo.\n");
 
-	if (session.vmlinux)
-		fd = open(session.vmlinux, O_RDONLY);
-	else
-		fd = open_default_vmlinux();
+	fd = open_vmlinux();
 	if (fd < 0) {
 		if (session.need_dwarf)
-			die("Could not open vmlinux/module file.");
+			die("Could not open debuginfo file.");
 
-		pr_warning("Could not open vmlinux/module file."
-			   " Try to use symbols.\n");
+		pr_debug("Could not open vmlinux/module file."
+			 " Try to use symbols.\n");
 		goto end_dwarf;
 	}
 
 	/* Searching probe points */
-	for (j = 0; j < session.nr_probe; j++) {
-		pp = &session.probes[j];
+	for (i = 0; i < session.nr_probe; i++) {
+		pp = &session.probes[i];
 		if (pp->found)
 			continue;
 
 		lseek(fd, SEEK_SET, 0);
 		ret = find_probepoint(fd, pp);
-		if (ret < 0) {
-			if (session.need_dwarf)
-				die("Could not analyze debuginfo.");
-
-			pr_warning("An error occurred in debuginfo analysis. Try to use symbols.\n");
-			break;
+		if (ret > 0)
+			continue;
+		if (ret == 0) {	/* No error but failed to find probe point. */
+			synthesize_perf_probe_point(pp);
+			die("Probe point '%s' not found. - probe not added.",
+			    pp->probes[0]);
+		}
+		/* Error path */
+		if (session.need_dwarf) {
+			if (ret == -ENOENT)
+				pr_warning("No dwarf info found in the vmlinux - please rebuild with CONFIG_DEBUG_INFO=y.\n");
+			die("Could not analyze debuginfo.");
 		}
-		if (ret == 0)	/* No error but failed to find probe point. */
-			die("No probe point found.");
+		pr_debug("An error occurred in debuginfo analysis."
+			 " Try to use symbols.\n");
+		break;
 	}
 	close(fd);
 
@@ -223,11 +288,12 @@ end_dwarf:
 #endif /* !NO_LIBDWARF */
 
 	/* Synthesize probes without dwarf */
-	for (j = 0; j < session.nr_probe; j++) {
-		pp = &session.probes[j];
+	for (i = 0; i < session.nr_probe; i++) {
+		pp = &session.probes[i];
 		if (pp->found)	/* This probe is already found. */
 			continue;
 
+		evaluate_probe_point(pp);
 		ret = synthesize_trace_kprobe_event(pp);
 		if (ret == -E2BIG)
 			die("probe point definition becomes too long.");
@@ -236,7 +302,8 @@ end_dwarf:
 	}
 
 	/* Settng up probe points */
-	add_trace_kprobe_events(session.probes, session.nr_probe);
+	add_trace_kprobe_events(session.probes, session.nr_probe,
+				session.force_add);
 	return 0;
 }
 
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 0e519c667e3..26542532273 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -17,6 +17,7 @@
 #include "util/header.h"
 #include "util/event.h"
 #include "util/debug.h"
+#include "util/session.h"
 #include "util/symbol.h"
 
 #include <unistd.h>
@@ -62,7 +63,7 @@ static int			nr_cpu				=      0;
 
 static int			file_new			=      1;
 
-struct perf_header		*header				=   NULL;
+static struct perf_session	*session;
 
 struct mmap_data {
 	int			counter;
@@ -122,7 +123,8 @@ static void write_event(event_t *buf, size_t size)
 	write_output(buf, size);
 }
 
-static int process_synthesized_event(event_t *event)
+static int process_synthesized_event(event_t *event,
+				     struct perf_session *self __used)
 {
 	write_event(event, event->header.size);
 	return 0;
@@ -216,12 +218,12 @@ static struct perf_header_attr *get_header_attr(struct perf_event_attr *a, int n
 {
 	struct perf_header_attr *h_attr;
 
-	if (nr < header->attrs) {
-		h_attr = header->attr[nr];
+	if (nr < session->header.attrs) {
+		h_attr = session->header.attr[nr];
 	} else {
 		h_attr = perf_header_attr__new(a);
 		if (h_attr != NULL)
-			if (perf_header__add_attr(header, h_attr) < 0) {
+			if (perf_header__add_attr(&session->header, h_attr) < 0) {
 				perf_header_attr__delete(h_attr);
 				h_attr = NULL;
 			}
@@ -276,7 +278,7 @@ static void create_counter(int counter, int cpu, pid_t pid)
 
 	attr->mmap		= track;
 	attr->comm		= track;
-	attr->inherit		= (cpu < 0) && inherit;
+	attr->inherit		= inherit;
 	attr->disabled		= 1;
 
 try_again:
@@ -395,9 +397,9 @@ static void open_counters(int cpu, pid_t pid)
 
 static void atexit_header(void)
 {
-	header->data_size += bytes_written;
+	session->header.data_size += bytes_written;
 
-	perf_header__write(header, output, true);
+	perf_header__write(&session->header, output, true);
 }
 
 static int __cmd_record(int argc, const char **argv)
@@ -408,6 +410,9 @@ static int __cmd_record(int argc, const char **argv)
 	int flags;
 	int err;
 	unsigned long waking = 0;
+	int child_ready_pipe[2], go_pipe[2];
+	const bool forks = target_pid == -1 && argc > 0;
+	char buf;
 
 	page_size = sysconf(_SC_PAGE_SIZE);
 	nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
@@ -418,11 +423,25 @@ static int __cmd_record(int argc, const char **argv)
 	signal(SIGCHLD, sig_handler);
 	signal(SIGINT, sig_handler);
 
+	if (forks && (pipe(child_ready_pipe) < 0 || pipe(go_pipe) < 0)) {
+		perror("failed to create pipes");
+		exit(-1);
+	}
+
 	if (!stat(output_name, &st) && st.st_size) {
-		if (!force && !append_file) {
-			fprintf(stderr, "Error, output file %s exists, use -A to append or -f to overwrite.\n",
-					output_name);
-			exit(-1);
+		if (!force) {
+			if (!append_file) {
+				pr_err("Error, output file %s exists, use -A "
+				       "to append or -f to overwrite.\n",
+				       output_name);
+				exit(-1);
+			}
+		} else {
+			char oldname[PATH_MAX];
+			snprintf(oldname, sizeof(oldname), "%s.old",
+				 output_name);
+			unlink(oldname);
+			rename(output_name, oldname);
 		}
 	} else {
 		append_file = 0;
@@ -440,24 +459,24 @@ static int __cmd_record(int argc, const char **argv)
 		exit(-1);
 	}
 
-	header = perf_header__new();
-	if (header == NULL) {
+	session = perf_session__new(output_name, O_WRONLY, force);
+	if (session == NULL) {
 		pr_err("Not enough memory for reading perf file header\n");
 		return -1;
 	}
 
 	if (!file_new) {
-		err = perf_header__read(header, output);
+		err = perf_header__read(&session->header, output);
 		if (err < 0)
 			return err;
 	}
 
 	if (raw_samples) {
-		perf_header__set_feat(header, HEADER_TRACE_INFO);
+		perf_header__set_feat(&session->header, HEADER_TRACE_INFO);
 	} else {
 		for (i = 0; i < nr_counters; i++) {
 			if (attrs[i].sample_type & PERF_SAMPLE_RAW) {
-				perf_header__set_feat(header, HEADER_TRACE_INFO);
+				perf_header__set_feat(&session->header, HEADER_TRACE_INFO);
 				break;
 			}
 		}
@@ -465,56 +484,79 @@ static int __cmd_record(int argc, const char **argv)
 
 	atexit(atexit_header);
 
-	if (!system_wide) {
-		pid = target_pid;
-		if (pid == -1)
-			pid = getpid();
-
-		open_counters(profile_cpu, pid);
-	} else {
-		if (profile_cpu != -1) {
-			open_counters(profile_cpu, target_pid);
-		} else {
-			for (i = 0; i < nr_cpus; i++)
-				open_counters(i, target_pid);
+	if (forks) {
+		pid = fork();
+		if (pid < 0) {
+			perror("failed to fork");
+			exit(-1);
 		}
-	}
 
-	if (file_new) {
-		err = perf_header__write(header, output, false);
-		if (err < 0)
-			return err;
-	}
+		if (!pid) {
+			close(child_ready_pipe[0]);
+			close(go_pipe[1]);
+			fcntl(go_pipe[0], F_SETFD, FD_CLOEXEC);
 
-	if (!system_wide)
-		event__synthesize_thread(pid, process_synthesized_event);
-	else
-		event__synthesize_threads(process_synthesized_event);
+			/*
+			 * Do a dummy execvp to get the PLT entry resolved,
+			 * so we avoid the resolver overhead on the real
+			 * execvp call.
+			 */
+			execvp("", (char **)argv);
 
-	if (target_pid == -1 && argc) {
-		pid = fork();
-		if (pid < 0)
-			die("failed to fork");
+			/*
+			 * Tell the parent we're ready to go
+			 */
+			close(child_ready_pipe[1]);
 
-		if (!pid) {
-			if (execvp(argv[0], (char **)argv)) {
-				perror(argv[0]);
-				exit(-1);
-			}
-		} else {
 			/*
-			 * Wait a bit for the execv'ed child to appear
-			 * and be updated in /proc
-			 * FIXME: Do you know a less heuristical solution?
+			 * Wait until the parent tells us to go.
 			 */
-			usleep(1000);
-			event__synthesize_thread(pid,
-						 process_synthesized_event);
+			if (read(go_pipe[0], &buf, 1) == -1)
+				perror("unable to read pipe");
+
+			execvp(argv[0], (char **)argv);
+
+			perror(argv[0]);
+			exit(-1);
 		}
 
 		child_pid = pid;
+
+		if (!system_wide)
+			target_pid = pid;
+
+		close(child_ready_pipe[1]);
+		close(go_pipe[0]);
+		/*
+		 * wait for child to settle
+		 */
+		if (read(child_ready_pipe[0], &buf, 1) == -1) {
+			perror("unable to read pipe");
+			exit(-1);
+		}
+		close(child_ready_pipe[0]);
+	}
+
+
+	if ((!system_wide && !inherit) || profile_cpu != -1) {
+		open_counters(profile_cpu, target_pid);
+	} else {
+		for (i = 0; i < nr_cpus; i++)
+			open_counters(i, target_pid);
 	}
 
+	if (file_new) {
+		err = perf_header__write(&session->header, output, false);
+		if (err < 0)
+			return err;
+	}
+
+	if (!system_wide && profile_cpu == -1)
+		event__synthesize_thread(pid, process_synthesized_event,
+					 session);
+	else
+		event__synthesize_threads(process_synthesized_event, session);
+
 	if (realtime_prio) {
 		struct sched_param param;
 
@@ -525,6 +567,12 @@ static int __cmd_record(int argc, const char **argv)
 		}
 	}
 
+	/*
+	 * Let the child rip
+	 */
+	if (forks)
+		close(go_pipe[1]);
+
 	for (;;) {
 		int hits = samples;
 
@@ -619,13 +667,13 @@ int cmd_record(int argc, const char **argv, const char *prefix __used)
 {
 	int counter;
 
-	symbol__init(0);
-
 	argc = parse_options(argc, argv, options, record_usage,
-		PARSE_OPT_STOP_AT_NON_OPTION);
-	if (!argc && target_pid == -1 && !system_wide)
+			    PARSE_OPT_STOP_AT_NON_OPTION);
+	if (!argc && target_pid == -1 && !system_wide && profile_cpu == -1)
 		usage_with_options(record_usage, options);
 
+	symbol__init();
+
 	if (!nr_counters) {
 		nr_counters	= 1;
 		attrs[0].type	= PERF_TYPE_HARDWARE;
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 383c4ab4f9a..db10c0e8eca 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -22,471 +22,46 @@
 #include "perf.h"
 #include "util/debug.h"
 #include "util/header.h"
+#include "util/session.h"
 
 #include "util/parse-options.h"
 #include "util/parse-events.h"
 
-#include "util/data_map.h"
 #include "util/thread.h"
 #include "util/sort.h"
 #include "util/hist.h"
 
 static char		const *input_name = "perf.data";
 
-static char		*dso_list_str, *comm_list_str, *sym_list_str,
-			*col_width_list_str;
-static struct strlist	*dso_list, *comm_list, *sym_list;
-
 static int		force;
 
-static int		full_paths;
-static int		show_nr_samples;
-
 static int		show_threads;
 static struct perf_read_values	show_threads_values;
 
 static char		default_pretty_printing_style[] = "normal";
 static char		*pretty_printing_style = default_pretty_printing_style;
 
-static int		exclude_other = 1;
-
 static char		callchain_default_opt[] = "fractal,0.5";
 
-static struct perf_header *header;
-
-static u64		sample_type;
-
-struct symbol_conf	symbol_conf;
-
-
-static size_t
-callchain__fprintf_left_margin(FILE *fp, int left_margin)
-{
-	int i;
-	int ret;
-
-	ret = fprintf(fp, "            ");
-
-	for (i = 0; i < left_margin; i++)
-		ret += fprintf(fp, " ");
-
-	return ret;
-}
-
-static size_t ipchain__fprintf_graph_line(FILE *fp, int depth, int depth_mask,
-					  int left_margin)
-{
-	int i;
-	size_t ret = 0;
-
-	ret += callchain__fprintf_left_margin(fp, left_margin);
-
-	for (i = 0; i < depth; i++)
-		if (depth_mask & (1 << i))
-			ret += fprintf(fp, "|          ");
-		else
-			ret += fprintf(fp, "           ");
-
-	ret += fprintf(fp, "\n");
-
-	return ret;
-}
-static size_t
-ipchain__fprintf_graph(FILE *fp, struct callchain_list *chain, int depth,
-		       int depth_mask, int count, u64 total_samples,
-		       int hits, int left_margin)
-{
-	int i;
-	size_t ret = 0;
-
-	ret += callchain__fprintf_left_margin(fp, left_margin);
-	for (i = 0; i < depth; i++) {
-		if (depth_mask & (1 << i))
-			ret += fprintf(fp, "|");
-		else
-			ret += fprintf(fp, " ");
-		if (!count && i == depth - 1) {
-			double percent;
-
-			percent = hits * 100.0 / total_samples;
-			ret += percent_color_fprintf(fp, "--%2.2f%%-- ", percent);
-		} else
-			ret += fprintf(fp, "%s", "          ");
-	}
-	if (chain->sym)
-		ret += fprintf(fp, "%s\n", chain->sym->name);
-	else
-		ret += fprintf(fp, "%p\n", (void *)(long)chain->ip);
-
-	return ret;
-}
-
-static struct symbol *rem_sq_bracket;
-static struct callchain_list rem_hits;
-
-static void init_rem_hits(void)
-{
-	rem_sq_bracket = malloc(sizeof(*rem_sq_bracket) + 6);
-	if (!rem_sq_bracket) {
-		fprintf(stderr, "Not enough memory to display remaining hits\n");
-		return;
-	}
-
-	strcpy(rem_sq_bracket->name, "[...]");
-	rem_hits.sym = rem_sq_bracket;
-}
-
-static size_t
-__callchain__fprintf_graph(FILE *fp, struct callchain_node *self,
-			   u64 total_samples, int depth, int depth_mask,
-			   int left_margin)
-{
-	struct rb_node *node, *next;
-	struct callchain_node *child;
-	struct callchain_list *chain;
-	int new_depth_mask = depth_mask;
-	u64 new_total;
-	u64 remaining;
-	size_t ret = 0;
-	int i;
-
-	if (callchain_param.mode == CHAIN_GRAPH_REL)
-		new_total = self->children_hit;
-	else
-		new_total = total_samples;
-
-	remaining = new_total;
-
-	node = rb_first(&self->rb_root);
-	while (node) {
-		u64 cumul;
-
-		child = rb_entry(node, struct callchain_node, rb_node);
-		cumul = cumul_hits(child);
-		remaining -= cumul;
-
-		/*
-		 * The depth mask manages the output of pipes that show
-		 * the depth. We don't want to keep the pipes of the current
-		 * level for the last child of this depth.
-		 * Except if we have remaining filtered hits. They will
-		 * supersede the last child
-		 */
-		next = rb_next(node);
-		if (!next && (callchain_param.mode != CHAIN_GRAPH_REL || !remaining))
-			new_depth_mask &= ~(1 << (depth - 1));
-
-		/*
-		 * But we keep the older depth mask for the line seperator
-		 * to keep the level link until we reach the last child
-		 */
-		ret += ipchain__fprintf_graph_line(fp, depth, depth_mask,
-						   left_margin);
-		i = 0;
-		list_for_each_entry(chain, &child->val, list) {
-			if (chain->ip >= PERF_CONTEXT_MAX)
-				continue;
-			ret += ipchain__fprintf_graph(fp, chain, depth,
-						      new_depth_mask, i++,
-						      new_total,
-						      cumul,
-						      left_margin);
-		}
-		ret += __callchain__fprintf_graph(fp, child, new_total,
-						  depth + 1,
-						  new_depth_mask | (1 << depth),
-						  left_margin);
-		node = next;
-	}
-
-	if (callchain_param.mode == CHAIN_GRAPH_REL &&
-		remaining && remaining != new_total) {
-
-		if (!rem_sq_bracket)
-			return ret;
-
-		new_depth_mask &= ~(1 << (depth - 1));
-
-		ret += ipchain__fprintf_graph(fp, &rem_hits, depth,
-					      new_depth_mask, 0, new_total,
-					      remaining, left_margin);
-	}
-
-	return ret;
-}
-
-
-static size_t
-callchain__fprintf_graph(FILE *fp, struct callchain_node *self,
-			 u64 total_samples, int left_margin)
-{
-	struct callchain_list *chain;
-	bool printed = false;
-	int i = 0;
-	int ret = 0;
-
-	list_for_each_entry(chain, &self->val, list) {
-		if (chain->ip >= PERF_CONTEXT_MAX)
-			continue;
-
-		if (!i++ && sort__first_dimension == SORT_SYM)
-			continue;
-
-		if (!printed) {
-			ret += callchain__fprintf_left_margin(fp, left_margin);
-			ret += fprintf(fp, "|\n");
-			ret += callchain__fprintf_left_margin(fp, left_margin);
-			ret += fprintf(fp, "---");
-
-			left_margin += 3;
-			printed = true;
-		} else
-			ret += callchain__fprintf_left_margin(fp, left_margin);
-
-		if (chain->sym)
-			ret += fprintf(fp, " %s\n", chain->sym->name);
-		else
-			ret += fprintf(fp, " %p\n", (void *)(long)chain->ip);
-	}
-
-	ret += __callchain__fprintf_graph(fp, self, total_samples, 1, 1, left_margin);
-
-	return ret;
-}
-
-static size_t
-callchain__fprintf_flat(FILE *fp, struct callchain_node *self,
-			u64 total_samples)
-{
-	struct callchain_list *chain;
-	size_t ret = 0;
-
-	if (!self)
-		return 0;
-
-	ret += callchain__fprintf_flat(fp, self->parent, total_samples);
-
-
-	list_for_each_entry(chain, &self->val, list) {
-		if (chain->ip >= PERF_CONTEXT_MAX)
-			continue;
-		if (chain->sym)
-			ret += fprintf(fp, "                %s\n", chain->sym->name);
-		else
-			ret += fprintf(fp, "                %p\n",
-					(void *)(long)chain->ip);
-	}
-
-	return ret;
-}
-
-static size_t
-hist_entry_callchain__fprintf(FILE *fp, struct hist_entry *self,
-			      u64 total_samples, int left_margin)
-{
-	struct rb_node *rb_node;
-	struct callchain_node *chain;
-	size_t ret = 0;
-
-	rb_node = rb_first(&self->sorted_chain);
-	while (rb_node) {
-		double percent;
-
-		chain = rb_entry(rb_node, struct callchain_node, rb_node);
-		percent = chain->hit * 100.0 / total_samples;
-		switch (callchain_param.mode) {
-		case CHAIN_FLAT:
-			ret += percent_color_fprintf(fp, "           %6.2f%%\n",
-						     percent);
-			ret += callchain__fprintf_flat(fp, chain, total_samples);
-			break;
-		case CHAIN_GRAPH_ABS: /* Falldown */
-		case CHAIN_GRAPH_REL:
-			ret += callchain__fprintf_graph(fp, chain, total_samples,
-							left_margin);
-		case CHAIN_NONE:
-		default:
-			break;
-		}
-		ret += fprintf(fp, "\n");
-		rb_node = rb_next(rb_node);
-	}
-
-	return ret;
-}
-
-static size_t
-hist_entry__fprintf(FILE *fp, struct hist_entry *self, u64 total_samples)
-{
-	struct sort_entry *se;
-	size_t ret;
-
-	if (exclude_other && !self->parent)
-		return 0;
-
-	if (total_samples)
-		ret = percent_color_fprintf(fp,
-					    field_sep ? "%.2f" : "   %6.2f%%",
-					(self->count * 100.0) / total_samples);
-	else
-		ret = fprintf(fp, field_sep ? "%lld" : "%12lld ", self->count);
-
-	if (show_nr_samples) {
-		if (field_sep)
-			fprintf(fp, "%c%lld", *field_sep, self->count);
-		else
-			fprintf(fp, "%11lld", self->count);
-	}
-
-	list_for_each_entry(se, &hist_entry__sort_list, list) {
-		if (se->elide)
-			continue;
-
-		fprintf(fp, "%s", field_sep ?: "  ");
-		ret += se->print(fp, self, se->width ? *se->width : 0);
-	}
-
-	ret += fprintf(fp, "\n");
-
-	if (callchain) {
-		int left_margin = 0;
-
-		if (sort__first_dimension == SORT_COMM) {
-			se = list_first_entry(&hist_entry__sort_list, typeof(*se),
-						list);
-			left_margin = se->width ? *se->width : 0;
-			left_margin -= thread__comm_len(self->thread);
-		}
-
-		hist_entry_callchain__fprintf(fp, self, total_samples,
-					      left_margin);
-	}
-
-	return ret;
-}
-
-/*
- *
- */
-
-static void dso__calc_col_width(struct dso *self)
-{
-	if (!col_width_list_str && !field_sep &&
-	    (!dso_list || strlist__has_entry(dso_list, self->name))) {
-		unsigned int slen = strlen(self->name);
-		if (slen > dsos__col_width)
-			dsos__col_width = slen;
-	}
-
-	self->slen_calculated = 1;
-}
-
-static void thread__comm_adjust(struct thread *self)
-{
-	char *comm = self->comm;
-
-	if (!col_width_list_str && !field_sep &&
-	    (!comm_list || strlist__has_entry(comm_list, comm))) {
-		unsigned int slen = strlen(comm);
-
-		if (slen > comms__col_width) {
-			comms__col_width = slen;
-			threads__col_width = slen + 6;
-		}
-	}
-}
-
-static int thread__set_comm_adjust(struct thread *self, const char *comm)
-{
-	int ret = thread__set_comm(self, comm);
-
-	if (ret)
-		return ret;
-
-	thread__comm_adjust(self);
-
-	return 0;
-}
-
-static int call__match(struct symbol *sym)
-{
-	if (sym->name && !regexec(&parent_regex, sym->name, 0, NULL, 0))
-		return 1;
-
-	return 0;
-}
-
-static struct symbol **resolve_callchain(struct thread *thread,
-					 struct ip_callchain *chain,
-					 struct symbol **parent)
-{
-	u8 cpumode = PERF_RECORD_MISC_USER;
-	struct symbol **syms = NULL;
-	unsigned int i;
-
-	if (callchain) {
-		syms = calloc(chain->nr, sizeof(*syms));
-		if (!syms) {
-			fprintf(stderr, "Can't allocate memory for symbols\n");
-			exit(-1);
-		}
-	}
-
-	for (i = 0; i < chain->nr; i++) {
-		u64 ip = chain->ips[i];
-		struct addr_location al;
-
-		if (ip >= PERF_CONTEXT_MAX) {
-			switch (ip) {
-			case PERF_CONTEXT_HV:
-				cpumode = PERF_RECORD_MISC_HYPERVISOR;	break;
-			case PERF_CONTEXT_KERNEL:
-				cpumode = PERF_RECORD_MISC_KERNEL;	break;
-			case PERF_CONTEXT_USER:
-				cpumode = PERF_RECORD_MISC_USER;	break;
-			default:
-				break;
-			}
-			continue;
-		}
-
-		thread__find_addr_location(thread, cpumode, MAP__FUNCTION,
-					   ip, &al, NULL);
-		if (al.sym != NULL) {
-			if (sort__has_parent && !*parent &&
-			    call__match(al.sym))
-				*parent = al.sym;
-			if (!callchain)
-				break;
-			syms[i] = al.sym;
-		}
-	}
-
-	return syms;
-}
-
-/*
- * collect histogram counts
- */
-
-static int hist_entry__add(struct addr_location *al,
-			   struct ip_callchain *chain, u64 count)
+static int perf_session__add_hist_entry(struct perf_session *self,
+					struct addr_location *al,
+					struct ip_callchain *chain, u64 count)
 {
 	struct symbol **syms = NULL, *parent = NULL;
 	bool hit;
 	struct hist_entry *he;
 
-	if ((sort__has_parent || callchain) && chain)
-		syms = resolve_callchain(al->thread, chain, &parent);
-
-	he = __hist_entry__add(al, parent, count, &hit);
+	if ((sort__has_parent || symbol_conf.use_callchain) && chain)
+		syms = perf_session__resolve_callchain(self, al->thread,
+						       chain, &parent);
+	he = __perf_session__add_hist_entry(self, al, parent, count, &hit);
 	if (he == NULL)
 		return -ENOMEM;
 
 	if (hit)
 		he->count += count;
 
-	if (callchain) {
+	if (symbol_conf.use_callchain) {
 		if (!hit)
 			callchain_init(&he->callchain);
 		append_chain(&he->callchain, chain, syms);
@@ -496,100 +71,6 @@ static int hist_entry__add(struct addr_location *al,
 	return 0;
 }
 
-static size_t output__fprintf(FILE *fp, u64 total_samples)
-{
-	struct hist_entry *pos;
-	struct sort_entry *se;
-	struct rb_node *nd;
-	size_t ret = 0;
-	unsigned int width;
-	char *col_width = col_width_list_str;
-	int raw_printing_style;
-
-	raw_printing_style = !strcmp(pretty_printing_style, "raw");
-
-	init_rem_hits();
-
-	fprintf(fp, "# Samples: %Ld\n", (u64)total_samples);
-	fprintf(fp, "#\n");
-
-	fprintf(fp, "# Overhead");
-	if (show_nr_samples) {
-		if (field_sep)
-			fprintf(fp, "%cSamples", *field_sep);
-		else
-			fputs("  Samples  ", fp);
-	}
-	list_for_each_entry(se, &hist_entry__sort_list, list) {
-		if (se->elide)
-			continue;
-		if (field_sep) {
-			fprintf(fp, "%c%s", *field_sep, se->header);
-			continue;
-		}
-		width = strlen(se->header);
-		if (se->width) {
-			if (col_width_list_str) {
-				if (col_width) {
-					*se->width = atoi(col_width);
-					col_width = strchr(col_width, ',');
-					if (col_width)
-						++col_width;
-				}
-			}
-			width = *se->width = max(*se->width, width);
-		}
-		fprintf(fp, "  %*s", width, se->header);
-	}
-	fprintf(fp, "\n");
-
-	if (field_sep)
-		goto print_entries;
-
-	fprintf(fp, "# ........");
-	if (show_nr_samples)
-		fprintf(fp, " ..........");
-	list_for_each_entry(se, &hist_entry__sort_list, list) {
-		unsigned int i;
-
-		if (se->elide)
-			continue;
-
-		fprintf(fp, "  ");
-		if (se->width)
-			width = *se->width;
-		else
-			width = strlen(se->header);
-		for (i = 0; i < width; i++)
-			fprintf(fp, ".");
-	}
-	fprintf(fp, "\n");
-
-	fprintf(fp, "#\n");
-
-print_entries:
-	for (nd = rb_first(&output_hists); nd; nd = rb_next(nd)) {
-		pos = rb_entry(nd, struct hist_entry, rb_node);
-		ret += hist_entry__fprintf(fp, pos, total_samples);
-	}
-
-	if (sort_order == default_sort_order &&
-			parent_pattern == default_parent_pattern) {
-		fprintf(fp, "#\n");
-		fprintf(fp, "# (For a higher level overview, try: perf report --sort comm,dso)\n");
-		fprintf(fp, "#\n");
-	}
-	fprintf(fp, "\n");
-
-	free(rem_sq_bracket);
-
-	if (show_threads)
-		perf_read_values_display(fp, &show_threads_values,
-					 raw_printing_style);
-
-	return ret;
-}
-
 static int validate_chain(struct ip_callchain *chain, event_t *event)
 {
 	unsigned int chain_size;
@@ -603,108 +84,60 @@ static int validate_chain(struct ip_callchain *chain, event_t *event)
 	return 0;
 }
 
-static int process_sample_event(event_t *event)
+static int process_sample_event(event_t *event, struct perf_session *session)
 {
-	u64 ip = event->ip.ip;
-	u64 period = 1;
-	void *more_data = event->ip.__more_data;
-	struct ip_callchain *chain = NULL;
-	int cpumode;
+	struct sample_data data = { .period = 1, };
 	struct addr_location al;
-	struct thread *thread = threads__findnew(event->ip.pid);
 
-	if (sample_type & PERF_SAMPLE_PERIOD) {
-		period = *(u64 *)more_data;
-		more_data += sizeof(u64);
-	}
+	event__parse_sample(event, session->sample_type, &data);
 
 	dump_printf("(IP, %d): %d/%d: %p period: %Ld\n",
 		event->header.misc,
-		event->ip.pid, event->ip.tid,
-		(void *)(long)ip,
-		(long long)period);
+		data.pid, data.tid,
+		(void *)(long)data.ip,
+		(long long)data.period);
 
-	if (sample_type & PERF_SAMPLE_CALLCHAIN) {
+	if (session->sample_type & PERF_SAMPLE_CALLCHAIN) {
 		unsigned int i;
 
-		chain = (void *)more_data;
+		dump_printf("... chain: nr:%Lu\n", data.callchain->nr);
 
-		dump_printf("... chain: nr:%Lu\n", chain->nr);
-
-		if (validate_chain(chain, event) < 0) {
+		if (validate_chain(data.callchain, event) < 0) {
 			pr_debug("call-chain problem with event, "
 				 "skipping it.\n");
 			return 0;
 		}
 
 		if (dump_trace) {
-			for (i = 0; i < chain->nr; i++)
-				dump_printf("..... %2d: %016Lx\n", i, chain->ips[i]);
+			for (i = 0; i < data.callchain->nr; i++)
+				dump_printf("..... %2d: %016Lx\n",
+					    i, data.callchain->ips[i]);
 		}
 	}
 
-	if (thread == NULL) {
-		pr_debug("problem processing %d event, skipping it.\n",
+	if (event__preprocess_sample(event, session, &al, NULL) < 0) {
+		fprintf(stderr, "problem processing %d event, skipping it.\n",
 			event->header.type);
 		return -1;
 	}
 
-	dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid);
-
-	if (comm_list && !strlist__has_entry(comm_list, thread->comm))
-		return 0;
-
-	cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
-
-	thread__find_addr_location(thread, cpumode,
-				   MAP__FUNCTION, ip, &al, NULL);
-	/*
-	 * We have to do this here as we may have a dso with no symbol hit that
-	 * has a name longer than the ones with symbols sampled.
-	 */
-	if (al.map && !sort_dso.elide && !al.map->dso->slen_calculated)
-		dso__calc_col_width(al.map->dso);
-
-	if (dso_list &&
-	    (!al.map || !al.map->dso ||
-	     !(strlist__has_entry(dso_list, al.map->dso->short_name) ||
-	       (al.map->dso->short_name != al.map->dso->long_name &&
-		strlist__has_entry(dso_list, al.map->dso->long_name)))))
+	if (al.filtered)
 		return 0;
 
-	if (sym_list && al.sym && !strlist__has_entry(sym_list, al.sym->name))
-		return 0;
-
-	if (hist_entry__add(&al, chain, period)) {
+	if (perf_session__add_hist_entry(session, &al, data.callchain, data.period)) {
 		pr_debug("problem incrementing symbol count, skipping event\n");
 		return -1;
 	}
 
-	event__stats.total += period;
-
+	session->events_stats.total += data.period;
 	return 0;
 }
 
-static int process_comm_event(event_t *event)
-{
-	struct thread *thread = threads__findnew(event->comm.pid);
-
-	dump_printf(": %s:%d\n", event->comm.comm, event->comm.pid);
-
-	if (thread == NULL ||
-	    thread__set_comm_adjust(thread, event->comm.comm)) {
-		dump_printf("problem processing PERF_RECORD_COMM, skipping event.\n");
-		return -1;
-	}
-
-	return 0;
-}
-
-static int process_read_event(event_t *event)
+static int process_read_event(event_t *event, struct perf_session *session __used)
 {
 	struct perf_event_attr *attr;
 
-	attr = perf_header__find_attr(event->read.id, header);
+	attr = perf_header__find_attr(event->read.id, &session->header);
 
 	if (show_threads) {
 		const char *name = attr ? __event_name(attr->type, attr->config)
@@ -723,25 +156,23 @@ static int process_read_event(event_t *event)
 	return 0;
 }
 
-static int sample_type_check(u64 type)
+static int sample_type_check(struct perf_session *session)
 {
-	sample_type = type;
-
-	if (!(sample_type & PERF_SAMPLE_CALLCHAIN)) {
+	if (!(session->sample_type & PERF_SAMPLE_CALLCHAIN)) {
 		if (sort__has_parent) {
 			fprintf(stderr, "selected --sort parent, but no"
 					" callchain data. Did you call"
 					" perf record without -g?\n");
 			return -1;
 		}
-		if (callchain) {
+		if (symbol_conf.use_callchain) {
 			fprintf(stderr, "selected -g but no callchain data."
 					" Did you call perf record without"
 					" -g?\n");
 			return -1;
 		}
-	} else if (callchain_param.mode != CHAIN_NONE && !callchain) {
-			callchain = 1;
+	} else if (callchain_param.mode != CHAIN_NONE && !symbol_conf.use_callchain) {
+			symbol_conf.use_callchain = true;
 			if (register_callchain_param(&callchain_param) < 0) {
 				fprintf(stderr, "Can't register callchain"
 						" params\n");
@@ -752,10 +183,10 @@ static int sample_type_check(u64 type)
 	return 0;
 }
 
-static struct perf_file_handler file_handler = {
+static struct perf_event_ops event_ops = {
 	.process_sample_event	= process_sample_event,
 	.process_mmap_event	= event__process_mmap,
-	.process_comm_event	= process_comm_event,
+	.process_comm_event	= event__process_comm,
 	.process_exit_event	= event__process_task,
 	.process_fork_event	= event__process_task,
 	.process_lost_event	= event__process_lost,
@@ -766,40 +197,47 @@ static struct perf_file_handler file_handler = {
 
 static int __cmd_report(void)
 {
-	struct thread *idle;
 	int ret;
+	struct perf_session *session;
 
-	idle = register_idle_thread();
-	thread__comm_adjust(idle);
+	session = perf_session__new(input_name, O_RDONLY, force);
+	if (session == NULL)
+		return -ENOMEM;
 
 	if (show_threads)
 		perf_read_values_init(&show_threads_values);
 
-	register_perf_file_handler(&file_handler);
-
-	ret = mmap_dispatch_perf_file(&header, input_name, force,
-				      full_paths, &event__cwdlen, &event__cwd);
+	ret = perf_session__process_events(session, &event_ops);
 	if (ret)
-		return ret;
+		goto out_delete;
 
 	if (dump_trace) {
 		event__print_totals();
-		return 0;
+		goto out_delete;
 	}
 
 	if (verbose > 3)
-		threads__fprintf(stdout);
+		perf_session__fprintf(session, stdout);
 
 	if (verbose > 2)
 		dsos__fprintf(stdout);
 
-	collapse__resort();
-	output__resort(event__stats.total);
-	output__fprintf(stdout, event__stats.total);
+	perf_session__collapse_resort(session);
+	perf_session__output_resort(session, session->events_stats.total);
+	fprintf(stdout, "# Samples: %Ld\n#\n", session->events_stats.total);
+	perf_session__fprintf_hists(session, NULL, false, stdout);
+	if (sort_order == default_sort_order &&
+	    parent_pattern == default_parent_pattern)
+		fprintf(stdout, "#\n# (For a higher level overview, try: perf report --sort comm,dso)\n#\n");
 
-	if (show_threads)
+	if (show_threads) {
+		bool raw_printing_style = !strcmp(pretty_printing_style, "raw");
+		perf_read_values_display(stdout, &show_threads_values,
+					 raw_printing_style);
 		perf_read_values_destroy(&show_threads_values);
-
+	}
+out_delete:
+	perf_session__delete(session);
 	return ret;
 }
 
@@ -810,7 +248,7 @@ parse_callchain_opt(const struct option *opt __used, const char *arg,
 	char *tok;
 	char *endptr;
 
-	callchain = 1;
+	symbol_conf.use_callchain = true;
 
 	if (!arg)
 		return 0;
@@ -831,7 +269,7 @@ parse_callchain_opt(const struct option *opt __used, const char *arg,
 
 	else if (!strncmp(tok, "none", strlen(arg))) {
 		callchain_param.mode = CHAIN_NONE;
-		callchain = 0;
+		symbol_conf.use_callchain = true;
 
 		return 0;
 	}
@@ -856,8 +294,7 @@ setup:
 	return 0;
 }
 
-//static const char * const report_usage[] = {
-const char * const report_usage[] = {
+static const char * const report_usage[] = {
 	"perf report [<options>] <command>",
 	NULL
 };
@@ -874,7 +311,7 @@ static const struct option options[] = {
 	OPT_BOOLEAN('f', "force", &force, "don't complain, do it"),
 	OPT_BOOLEAN('m', "modules", &symbol_conf.use_modules,
 		    "load module symbols - WARNING: use only with -k and LIVE kernel"),
-	OPT_BOOLEAN('n', "show-nr-samples", &show_nr_samples,
+	OPT_BOOLEAN('n', "show-nr-samples", &symbol_conf.show_nr_samples,
 		    "Show a column with the number of samples"),
 	OPT_BOOLEAN('T', "threads", &show_threads,
 		    "Show per-thread event counters"),
@@ -882,78 +319,46 @@ static const struct option options[] = {
 		   "pretty printing style key: normal raw"),
 	OPT_STRING('s', "sort", &sort_order, "key[,key2...]",
 		   "sort by key(s): pid, comm, dso, symbol, parent"),
-	OPT_BOOLEAN('P', "full-paths", &full_paths,
+	OPT_BOOLEAN('P', "full-paths", &event_ops.full_paths,
 		    "Don't shorten the pathnames taking into account the cwd"),
 	OPT_STRING('p', "parent", &parent_pattern, "regex",
 		   "regex filter to identify parent, see: '--sort parent'"),
-	OPT_BOOLEAN('x', "exclude-other", &exclude_other,
+	OPT_BOOLEAN('x', "exclude-other", &symbol_conf.exclude_other,
 		    "Only display entries with parent-match"),
 	OPT_CALLBACK_DEFAULT('g', "call-graph", NULL, "output_type,min_percent",
 		     "Display callchains using output_type and min percent threshold. "
 		     "Default: fractal,0.5", &parse_callchain_opt, callchain_default_opt),
-	OPT_STRING('d', "dsos", &dso_list_str, "dso[,dso...]",
+	OPT_STRING('d', "dsos", &symbol_conf.dso_list_str, "dso[,dso...]",
 		   "only consider symbols in these dsos"),
-	OPT_STRING('C', "comms", &comm_list_str, "comm[,comm...]",
+	OPT_STRING('C', "comms", &symbol_conf.comm_list_str, "comm[,comm...]",
 		   "only consider symbols in these comms"),
-	OPT_STRING('S', "symbols", &sym_list_str, "symbol[,symbol...]",
+	OPT_STRING('S', "symbols", &symbol_conf.sym_list_str, "symbol[,symbol...]",
 		   "only consider these symbols"),
-	OPT_STRING('w', "column-widths", &col_width_list_str,
+	OPT_STRING('w', "column-widths", &symbol_conf.col_width_list_str,
 		   "width[,width...]",
 		   "don't try to adjust column width, use these fixed values"),
-	OPT_STRING('t', "field-separator", &field_sep, "separator",
+	OPT_STRING('t', "field-separator", &symbol_conf.field_sep, "separator",
 		   "separator for columns, no spaces will be added between "
 		   "columns '.' is reserved."),
 	OPT_END()
 };
 
-static void setup_sorting(void)
+int cmd_report(int argc, const char **argv, const char *prefix __used)
 {
-	char *tmp, *tok, *str = strdup(sort_order);
-
-	for (tok = strtok_r(str, ", ", &tmp);
-			tok; tok = strtok_r(NULL, ", ", &tmp)) {
-		if (sort_dimension__add(tok) < 0) {
-			error("Unknown --sort key: `%s'", tok);
-			usage_with_options(report_usage, options);
-		}
-	}
-
-	free(str);
-}
+	argc = parse_options(argc, argv, options, report_usage, 0);
 
-static void setup_list(struct strlist **list, const char *list_str,
-		       struct sort_entry *se, const char *list_name,
-		       FILE *fp)
-{
-	if (list_str) {
-		*list = strlist__new(true, list_str);
-		if (!*list) {
-			fprintf(stderr, "problems parsing %s list\n",
-				list_name);
-			exit(129);
-		}
-		if (strlist__nr_entries(*list) == 1) {
-			fprintf(fp, "# %s: %s\n", list_name,
-				strlist__entry(*list, 0)->s);
-			se->elide = true;
-		}
-	}
-}
+	setup_pager();
 
-int cmd_report(int argc, const char **argv, const char *prefix __used)
-{
-	if (symbol__init(&symbol_conf) < 0)
+	if (symbol__init() < 0)
 		return -1;
 
-	argc = parse_options(argc, argv, options, report_usage, 0);
-
-	setup_sorting();
+	setup_sorting(report_usage, options);
 
 	if (parent_pattern != default_parent_pattern) {
 		sort_dimension__add("parent");
 		sort_parent.elide = 1;
 	} else
-		exclude_other = 0;
+		symbol_conf.exclude_other = false;
 
 	/*
 	 * Any (unrecognized) arguments left?
@@ -961,17 +366,9 @@ int cmd_report(int argc, const char **argv, const char *prefix __used)
 	if (argc)
 		usage_with_options(report_usage, options);
 
-	setup_pager();
-
-	setup_list(&dso_list, dso_list_str, &sort_dso, "dso", stdout);
-	setup_list(&comm_list, comm_list_str, &sort_comm, "comm", stdout);
-	setup_list(&sym_list, sym_list_str, &sort_sym, "symbol", stdout);
-
-	if (field_sep && *field_sep == '.') {
-		fputs("'.' is the only non valid --field-separator argument\n",
-		      stderr);
-		exit(129);
-	}
+	sort_entry__setup_elide(&sort_dso, symbol_conf.dso_list, "dso", stdout);
+	sort_entry__setup_elide(&sort_comm, symbol_conf.comm_list, "comm", stdout);
+	sort_entry__setup_elide(&sort_sym, symbol_conf.sym_list, "symbol", stdout);
 
 	return __cmd_report();
 }
diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c
index 26b782f26ee..80209df6cfe 100644
--- a/tools/perf/builtin-sched.c
+++ b/tools/perf/builtin-sched.c
@@ -6,14 +6,13 @@
 #include "util/symbol.h"
 #include "util/thread.h"
 #include "util/header.h"
+#include "util/session.h"
 
 #include "util/parse-options.h"
 #include "util/trace-event.h"
 
 #include "util/debug.h"
-#include "util/data_map.h"
 
-#include <sys/types.h>
 #include <sys/prctl.h>
 
 #include <semaphore.h>
@@ -22,9 +21,6 @@
 
 static char			const *input_name = "perf.data";
 
-static struct perf_header	*header;
-static u64			sample_type;
-
 static char			default_sort_order[] = "avg, max, switch, runtime";
 static char			*sort_order = default_sort_order;
 
@@ -141,6 +137,7 @@ struct work_atoms {
 	struct thread		*thread;
 	struct rb_node		node;
 	u64			max_lat;
+	u64			max_lat_at;
 	u64			total_lat;
 	u64			nb_atoms;
 	u64			total_runtime;
@@ -414,34 +411,33 @@ static u64 get_cpu_usage_nsec_parent(void)
 	return sum;
 }
 
-static u64 get_cpu_usage_nsec_self(void)
+static int self_open_counters(void)
 {
-	char filename [] = "/proc/1234567890/sched";
-	unsigned long msecs, nsecs;
-	char *line = NULL;
-	u64 total = 0;
-	size_t len = 0;
-	ssize_t chars;
-	FILE *file;
-	int ret;
+	struct perf_event_attr attr;
+	int fd;
 
-	sprintf(filename, "/proc/%d/sched", getpid());
-	file = fopen(filename, "r");
-	BUG_ON(!file);
+	memset(&attr, 0, sizeof(attr));
 
-	while ((chars = getline(&line, &len, file)) != -1) {
-		ret = sscanf(line, "se.sum_exec_runtime : %ld.%06ld\n",
-			&msecs, &nsecs);
-		if (ret == 2) {
-			total = msecs*1e6 + nsecs;
-			break;
-		}
-	}
-	if (line)
-		free(line);
-	fclose(file);
+	attr.type = PERF_TYPE_SOFTWARE;
+	attr.config = PERF_COUNT_SW_TASK_CLOCK;
+
+	fd = sys_perf_event_open(&attr, 0, -1, -1, 0);
+
+	if (fd < 0)
+		die("Error: sys_perf_event_open() syscall returned"
+		    "with %d (%s)\n", fd, strerror(errno));
+	return fd;
+}
+
+static u64 get_cpu_usage_nsec_self(int fd)
+{
+	u64 runtime;
+	int ret;
 
-	return total;
+	ret = read(fd, &runtime, sizeof(runtime));
+	BUG_ON(ret != sizeof(runtime));
+
+	return runtime;
 }
 
 static void *thread_func(void *ctx)
@@ -450,9 +446,11 @@ static void *thread_func(void *ctx)
 	u64 cpu_usage_0, cpu_usage_1;
 	unsigned long i, ret;
 	char comm2[22];
+	int fd;
 
 	sprintf(comm2, ":%s", this_task->comm);
 	prctl(PR_SET_NAME, comm2);
+	fd = self_open_counters();
 
 again:
 	ret = sem_post(&this_task->ready_for_work);
@@ -462,16 +460,15 @@ again:
 	ret = pthread_mutex_unlock(&start_work_mutex);
 	BUG_ON(ret);
 
-	cpu_usage_0 = get_cpu_usage_nsec_self();
+	cpu_usage_0 = get_cpu_usage_nsec_self(fd);
 
 	for (i = 0; i < this_task->nr_events; i++) {
 		this_task->curr_event = i;
 		process_sched_event(this_task, this_task->atoms[i]);
 	}
 
-	cpu_usage_1 = get_cpu_usage_nsec_self();
+	cpu_usage_1 = get_cpu_usage_nsec_self(fd);
 	this_task->cpu_usage = cpu_usage_1 - cpu_usage_0;
-
 	ret = sem_post(&this_task->work_done_sem);
 	BUG_ON(ret);
 
@@ -628,11 +625,6 @@ static void test_calibrations(void)
 	printf("the sleep test took %Ld nsecs\n", T1-T0);
 }
 
-struct raw_event_sample {
-	u32 size;
-	char data[0];
-};
-
 #define FILL_FIELD(ptr, field, event, data)	\
 	ptr.field = (typeof(ptr.field)) raw_field_value(event, #field, data)
 
@@ -736,18 +728,21 @@ struct trace_migrate_task_event {
 
 struct trace_sched_handler {
 	void (*switch_event)(struct trace_switch_event *,
+			     struct perf_session *,
 			     struct event *,
 			     int cpu,
 			     u64 timestamp,
 			     struct thread *thread);
 
 	void (*runtime_event)(struct trace_runtime_event *,
+			      struct perf_session *,
 			      struct event *,
 			      int cpu,
 			      u64 timestamp,
 			      struct thread *thread);
 
 	void (*wakeup_event)(struct trace_wakeup_event *,
+			     struct perf_session *,
 			     struct event *,
 			     int cpu,
 			     u64 timestamp,
@@ -760,6 +755,7 @@ struct trace_sched_handler {
 			   struct thread *thread);
 
 	void (*migrate_task_event)(struct trace_migrate_task_event *,
+			   struct perf_session *session,
 			   struct event *,
 			   int cpu,
 			   u64 timestamp,
@@ -769,6 +765,7 @@ struct trace_sched_handler {
 
 static void
 replay_wakeup_event(struct trace_wakeup_event *wakeup_event,
+		    struct perf_session *session __used,
 		    struct event *event,
 		    int cpu __used,
 		    u64 timestamp __used,
@@ -795,6 +792,7 @@ static u64 cpu_last_switched[MAX_CPUS];
 
 static void
 replay_switch_event(struct trace_switch_event *switch_event,
+		    struct perf_session *session __used,
 		    struct event *event,
 		    int cpu,
 		    u64 timestamp,
@@ -1019,13 +1017,16 @@ add_sched_in_event(struct work_atoms *atoms, u64 timestamp)
 
 	delta = atom->sched_in_time - atom->wake_up_time;
 	atoms->total_lat += delta;
-	if (delta > atoms->max_lat)
+	if (delta > atoms->max_lat) {
 		atoms->max_lat = delta;
+		atoms->max_lat_at = timestamp;
+	}
 	atoms->nb_atoms++;
 }
 
 static void
 latency_switch_event(struct trace_switch_event *switch_event,
+		     struct perf_session *session,
 		     struct event *event __used,
 		     int cpu,
 		     u64 timestamp,
@@ -1049,8 +1050,8 @@ latency_switch_event(struct trace_switch_event *switch_event,
 		die("hm, delta: %Ld < 0 ?\n", delta);
 
 
-	sched_out = threads__findnew(switch_event->prev_pid);
-	sched_in = threads__findnew(switch_event->next_pid);
+	sched_out = perf_session__findnew(session, switch_event->prev_pid);
+	sched_in = perf_session__findnew(session, switch_event->next_pid);
 
 	out_events = thread_atoms_search(&atom_root, sched_out, &cmp_pid);
 	if (!out_events) {
@@ -1078,12 +1079,13 @@ latency_switch_event(struct trace_switch_event *switch_event,
 
 static void
 latency_runtime_event(struct trace_runtime_event *runtime_event,
+		     struct perf_session *session,
 		     struct event *event __used,
 		     int cpu,
 		     u64 timestamp,
 		     struct thread *this_thread __used)
 {
-	struct thread *thread = threads__findnew(runtime_event->pid);
+	struct thread *thread = perf_session__findnew(session, runtime_event->pid);
 	struct work_atoms *atoms = thread_atoms_search(&atom_root, thread, &cmp_pid);
 
 	BUG_ON(cpu >= MAX_CPUS || cpu < 0);
@@ -1100,6 +1102,7 @@ latency_runtime_event(struct trace_runtime_event *runtime_event,
 
 static void
 latency_wakeup_event(struct trace_wakeup_event *wakeup_event,
+		     struct perf_session *session,
 		     struct event *__event __used,
 		     int cpu __used,
 		     u64 timestamp,
@@ -1113,7 +1116,7 @@ latency_wakeup_event(struct trace_wakeup_event *wakeup_event,
 	if (!wakeup_event->success)
 		return;
 
-	wakee = threads__findnew(wakeup_event->pid);
+	wakee = perf_session__findnew(session, wakeup_event->pid);
 	atoms = thread_atoms_search(&atom_root, wakee, &cmp_pid);
 	if (!atoms) {
 		thread_atoms_insert(wakee);
@@ -1147,6 +1150,7 @@ latency_wakeup_event(struct trace_wakeup_event *wakeup_event,
 
 static void
 latency_migrate_task_event(struct trace_migrate_task_event *migrate_task_event,
+		     struct perf_session *session,
 		     struct event *__event __used,
 		     int cpu __used,
 		     u64 timestamp,
@@ -1162,7 +1166,7 @@ latency_migrate_task_event(struct trace_migrate_task_event *migrate_task_event,
 	if (profile_cpu == -1)
 		return;
 
-	migrant = threads__findnew(migrate_task_event->pid);
+	migrant = perf_session__findnew(session, migrate_task_event->pid);
 	atoms = thread_atoms_search(&atom_root, migrant, &cmp_pid);
 	if (!atoms) {
 		thread_atoms_insert(migrant);
@@ -1216,10 +1220,11 @@ static void output_lat_thread(struct work_atoms *work_list)
 
 	avg = work_list->total_lat / work_list->nb_atoms;
 
-	printf("|%11.3f ms |%9llu | avg:%9.3f ms | max:%9.3f ms |\n",
+	printf("|%11.3f ms |%9llu | avg:%9.3f ms | max:%9.3f ms | max at: %9.6f s\n",
 	      (double)work_list->total_runtime / 1e6,
 		 work_list->nb_atoms, (double)avg / 1e6,
-		 (double)work_list->max_lat / 1e6);
+		 (double)work_list->max_lat / 1e6,
+		 (double)work_list->max_lat_at / 1e9);
 }
 
 static int pid_cmp(struct work_atoms *l, struct work_atoms *r)
@@ -1356,7 +1361,7 @@ static void sort_lat(void)
 static struct trace_sched_handler *trace_handler;
 
 static void
-process_sched_wakeup_event(struct raw_event_sample *raw,
+process_sched_wakeup_event(void *data, struct perf_session *session,
 			   struct event *event,
 			   int cpu __used,
 			   u64 timestamp __used,
@@ -1364,16 +1369,17 @@ process_sched_wakeup_event(struct raw_event_sample *raw,
 {
 	struct trace_wakeup_event wakeup_event;
 
-	FILL_COMMON_FIELDS(wakeup_event, event, raw->data);
+	FILL_COMMON_FIELDS(wakeup_event, event, data);
 
-	FILL_ARRAY(wakeup_event, comm, event, raw->data);
-	FILL_FIELD(wakeup_event, pid, event, raw->data);
-	FILL_FIELD(wakeup_event, prio, event, raw->data);
-	FILL_FIELD(wakeup_event, success, event, raw->data);
-	FILL_FIELD(wakeup_event, cpu, event, raw->data);
+	FILL_ARRAY(wakeup_event, comm, event, data);
+	FILL_FIELD(wakeup_event, pid, event, data);
+	FILL_FIELD(wakeup_event, prio, event, data);
+	FILL_FIELD(wakeup_event, success, event, data);
+	FILL_FIELD(wakeup_event, cpu, event, data);
 
 	if (trace_handler->wakeup_event)
-		trace_handler->wakeup_event(&wakeup_event, event, cpu, timestamp, thread);
+		trace_handler->wakeup_event(&wakeup_event, session, event,
+					    cpu, timestamp, thread);
 }
 
 /*
@@ -1391,6 +1397,7 @@ static char next_shortname2 = '0';
 
 static void
 map_switch_event(struct trace_switch_event *switch_event,
+		 struct perf_session *session,
 		 struct event *event __used,
 		 int this_cpu,
 		 u64 timestamp,
@@ -1418,8 +1425,8 @@ map_switch_event(struct trace_switch_event *switch_event,
 		die("hm, delta: %Ld < 0 ?\n", delta);
 
 
-	sched_out = threads__findnew(switch_event->prev_pid);
-	sched_in = threads__findnew(switch_event->next_pid);
+	sched_out = perf_session__findnew(session, switch_event->prev_pid);
+	sched_in = perf_session__findnew(session, switch_event->next_pid);
 
 	curr_thread[this_cpu] = sched_in;
 
@@ -1469,7 +1476,7 @@ map_switch_event(struct trace_switch_event *switch_event,
 
 
 static void
-process_sched_switch_event(struct raw_event_sample *raw,
+process_sched_switch_event(void *data, struct perf_session *session,
 			   struct event *event,
 			   int this_cpu,
 			   u64 timestamp __used,
@@ -1477,15 +1484,15 @@ process_sched_switch_event(struct raw_event_sample *raw,
 {
 	struct trace_switch_event switch_event;
 
-	FILL_COMMON_FIELDS(switch_event, event, raw->data);
+	FILL_COMMON_FIELDS(switch_event, event, data);
 
-	FILL_ARRAY(switch_event, prev_comm, event, raw->data);
-	FILL_FIELD(switch_event, prev_pid, event, raw->data);
-	FILL_FIELD(switch_event, prev_prio, event, raw->data);
-	FILL_FIELD(switch_event, prev_state, event, raw->data);
-	FILL_ARRAY(switch_event, next_comm, event, raw->data);
-	FILL_FIELD(switch_event, next_pid, event, raw->data);
-	FILL_FIELD(switch_event, next_prio, event, raw->data);
+	FILL_ARRAY(switch_event, prev_comm, event, data);
+	FILL_FIELD(switch_event, prev_pid, event, data);
+	FILL_FIELD(switch_event, prev_prio, event, data);
+	FILL_FIELD(switch_event, prev_state, event, data);
+	FILL_ARRAY(switch_event, next_comm, event, data);
+	FILL_FIELD(switch_event, next_pid, event, data);
+	FILL_FIELD(switch_event, next_prio, event, data);
 
 	if (curr_pid[this_cpu] != (u32)-1) {
 		/*
@@ -1496,13 +1503,14 @@ process_sched_switch_event(struct raw_event_sample *raw,
 			nr_context_switch_bugs++;
 	}
 	if (trace_handler->switch_event)
-		trace_handler->switch_event(&switch_event, event, this_cpu, timestamp, thread);
+		trace_handler->switch_event(&switch_event, session, event,
+					    this_cpu, timestamp, thread);
 
 	curr_pid[this_cpu] = switch_event.next_pid;
 }
 
 static void
-process_sched_runtime_event(struct raw_event_sample *raw,
+process_sched_runtime_event(void *data, struct perf_session *session,
 			   struct event *event,
 			   int cpu __used,
 			   u64 timestamp __used,
@@ -1510,17 +1518,17 @@ process_sched_runtime_event(struct raw_event_sample *raw,
 {
 	struct trace_runtime_event runtime_event;
 
-	FILL_ARRAY(runtime_event, comm, event, raw->data);
-	FILL_FIELD(runtime_event, pid, event, raw->data);
-	FILL_FIELD(runtime_event, runtime, event, raw->data);
-	FILL_FIELD(runtime_event, vruntime, event, raw->data);
+	FILL_ARRAY(runtime_event, comm, event, data);
+	FILL_FIELD(runtime_event, pid, event, data);
+	FILL_FIELD(runtime_event, runtime, event, data);
+	FILL_FIELD(runtime_event, vruntime, event, data);
 
 	if (trace_handler->runtime_event)
-		trace_handler->runtime_event(&runtime_event, event, cpu, timestamp, thread);
+		trace_handler->runtime_event(&runtime_event, session, event, cpu, timestamp, thread);
 }
 
 static void
-process_sched_fork_event(struct raw_event_sample *raw,
+process_sched_fork_event(void *data,
 			 struct event *event,
 			 int cpu __used,
 			 u64 timestamp __used,
@@ -1528,15 +1536,16 @@ process_sched_fork_event(struct raw_event_sample *raw,
 {
 	struct trace_fork_event fork_event;
 
-	FILL_COMMON_FIELDS(fork_event, event, raw->data);
+	FILL_COMMON_FIELDS(fork_event, event, data);
 
-	FILL_ARRAY(fork_event, parent_comm, event, raw->data);
-	FILL_FIELD(fork_event, parent_pid, event, raw->data);
-	FILL_ARRAY(fork_event, child_comm, event, raw->data);
-	FILL_FIELD(fork_event, child_pid, event, raw->data);
+	FILL_ARRAY(fork_event, parent_comm, event, data);
+	FILL_FIELD(fork_event, parent_pid, event, data);
+	FILL_ARRAY(fork_event, child_comm, event, data);
+	FILL_FIELD(fork_event, child_pid, event, data);
 
 	if (trace_handler->fork_event)
-		trace_handler->fork_event(&fork_event, event, cpu, timestamp, thread);
+		trace_handler->fork_event(&fork_event, event,
+					  cpu, timestamp, thread);
 }
 
 static void
@@ -1550,7 +1559,7 @@ process_sched_exit_event(struct event *event,
 }
 
 static void
-process_sched_migrate_task_event(struct raw_event_sample *raw,
+process_sched_migrate_task_event(void *data, struct perf_session *session,
 			   struct event *event,
 			   int cpu __used,
 			   u64 timestamp __used,
@@ -1558,80 +1567,67 @@ process_sched_migrate_task_event(struct raw_event_sample *raw,
 {
 	struct trace_migrate_task_event migrate_task_event;
 
-	FILL_COMMON_FIELDS(migrate_task_event, event, raw->data);
+	FILL_COMMON_FIELDS(migrate_task_event, event, data);
 
-	FILL_ARRAY(migrate_task_event, comm, event, raw->data);
-	FILL_FIELD(migrate_task_event, pid, event, raw->data);
-	FILL_FIELD(migrate_task_event, prio, event, raw->data);
-	FILL_FIELD(migrate_task_event, cpu, event, raw->data);
+	FILL_ARRAY(migrate_task_event, comm, event, data);
+	FILL_FIELD(migrate_task_event, pid, event, data);
+	FILL_FIELD(migrate_task_event, prio, event, data);
+	FILL_FIELD(migrate_task_event, cpu, event, data);
 
 	if (trace_handler->migrate_task_event)
-		trace_handler->migrate_task_event(&migrate_task_event, event, cpu, timestamp, thread);
+		trace_handler->migrate_task_event(&migrate_task_event, session,
+						 event, cpu, timestamp, thread);
 }
 
 static void
-process_raw_event(event_t *raw_event __used, void *more_data,
-		  int cpu, u64 timestamp, struct thread *thread)
+process_raw_event(event_t *raw_event __used, struct perf_session *session,
+		  void *data, int cpu, u64 timestamp, struct thread *thread)
 {
-	struct raw_event_sample *raw = more_data;
 	struct event *event;
 	int type;
 
-	type = trace_parse_common_type(raw->data);
+
+	type = trace_parse_common_type(data);
 	event = trace_find_event(type);
 
 	if (!strcmp(event->name, "sched_switch"))
-		process_sched_switch_event(raw, event, cpu, timestamp, thread);
+		process_sched_switch_event(data, session, event, cpu, timestamp, thread);
 	if (!strcmp(event->name, "sched_stat_runtime"))
-		process_sched_runtime_event(raw, event, cpu, timestamp, thread);
+		process_sched_runtime_event(data, session, event, cpu, timestamp, thread);
 	if (!strcmp(event->name, "sched_wakeup"))
-		process_sched_wakeup_event(raw, event, cpu, timestamp, thread);
+		process_sched_wakeup_event(data, session, event, cpu, timestamp, thread);
 	if (!strcmp(event->name, "sched_wakeup_new"))
-		process_sched_wakeup_event(raw, event, cpu, timestamp, thread);
+		process_sched_wakeup_event(data, session, event, cpu, timestamp, thread);
 	if (!strcmp(event->name, "sched_process_fork"))
-		process_sched_fork_event(raw, event, cpu, timestamp, thread);
+		process_sched_fork_event(data, event, cpu, timestamp, thread);
 	if (!strcmp(event->name, "sched_process_exit"))
 		process_sched_exit_event(event, cpu, timestamp, thread);
 	if (!strcmp(event->name, "sched_migrate_task"))
-		process_sched_migrate_task_event(raw, event, cpu, timestamp, thread);
+		process_sched_migrate_task_event(data, session, event, cpu, timestamp, thread);
 }
 
-static int process_sample_event(event_t *event)
+static int process_sample_event(event_t *event, struct perf_session *session)
 {
+	struct sample_data data;
 	struct thread *thread;
-	u64 ip = event->ip.ip;
-	u64 timestamp = -1;
-	u32 cpu = -1;
-	u64 period = 1;
-	void *more_data = event->ip.__more_data;
 
-	if (!(sample_type & PERF_SAMPLE_RAW))
+	if (!(session->sample_type & PERF_SAMPLE_RAW))
 		return 0;
 
-	thread = threads__findnew(event->ip.pid);
+	memset(&data, 0, sizeof(data));
+	data.time = -1;
+	data.cpu = -1;
+	data.period = -1;
 
-	if (sample_type & PERF_SAMPLE_TIME) {
-		timestamp = *(u64 *)more_data;
-		more_data += sizeof(u64);
-	}
-
-	if (sample_type & PERF_SAMPLE_CPU) {
-		cpu = *(u32 *)more_data;
-		more_data += sizeof(u32);
-		more_data += sizeof(u32); /* reserved */
-	}
-
-	if (sample_type & PERF_SAMPLE_PERIOD) {
-		period = *(u64 *)more_data;
-		more_data += sizeof(u64);
-	}
+	event__parse_sample(event, session->sample_type, &data);
 
 	dump_printf("(IP, %d): %d/%d: %p period: %Ld\n",
 		event->header.misc,
-		event->ip.pid, event->ip.tid,
-		(void *)(long)ip,
-		(long long)period);
+		data.pid, data.tid,
+		(void *)(long)data.ip,
+		(long long)data.period);
 
+	thread = perf_session__findnew(session, data.pid);
 	if (thread == NULL) {
 		pr_debug("problem processing %d event, skipping it.\n",
 			 event->header.type);
@@ -1640,15 +1636,16 @@ static int process_sample_event(event_t *event)
 
 	dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid);
 
-	if (profile_cpu != -1 && profile_cpu != (int) cpu)
+	if (profile_cpu != -1 && profile_cpu != (int)data.cpu)
 		return 0;
 
-	process_raw_event(event, more_data, cpu, timestamp, thread);
+	process_raw_event(event, session, data.raw_data, data.cpu, data.time, thread);
 
 	return 0;
 }
 
-static int process_lost_event(event_t *event __used)
+static int process_lost_event(event_t *event __used,
+			      struct perf_session *session __used)
 {
 	nr_lost_chunks++;
 	nr_lost_events += event->lost.lost;
@@ -1656,11 +1653,9 @@ static int process_lost_event(event_t *event __used)
 	return 0;
 }
 
-static int sample_type_check(u64 type)
+static int sample_type_check(struct perf_session *session __used)
 {
-	sample_type = type;
-
-	if (!(sample_type & PERF_SAMPLE_RAW)) {
+	if (!(session->sample_type & PERF_SAMPLE_RAW)) {
 		fprintf(stderr,
 			"No trace sample to read. Did you call perf record "
 			"without -R?");
@@ -1670,7 +1665,7 @@ static int sample_type_check(u64 type)
 	return 0;
 }
 
-static struct perf_file_handler file_handler = {
+static struct perf_event_ops event_ops = {
 	.process_sample_event	= process_sample_event,
 	.process_comm_event	= event__process_comm,
 	.process_lost_event	= process_lost_event,
@@ -1679,11 +1674,14 @@ static struct perf_file_handler file_handler = {
 
 static int read_events(void)
 {
-	register_idle_thread();
-	register_perf_file_handler(&file_handler);
+	int err;
+	struct perf_session *session = perf_session__new(input_name, O_RDONLY, 0);
+	if (session == NULL)
+		return -ENOMEM;
 
-	return mmap_dispatch_perf_file(&header, input_name, 0, 0,
-				       &event__cwdlen, &event__cwd);
+	err = perf_session__process_events(session, &event_ops);
+	perf_session__delete(session);
+	return err;
 }
 
 static void print_bad_events(void)
@@ -1724,9 +1722,9 @@ static void __cmd_lat(void)
 	read_events();
 	sort_lat();
 
-	printf("\n -----------------------------------------------------------------------------------------\n");
-	printf("  Task                  |   Runtime ms  | Switches | Average delay ms | Maximum delay ms |\n");
-	printf(" -----------------------------------------------------------------------------------------\n");
+	printf("\n ---------------------------------------------------------------------------------------------------------------\n");
+	printf("  Task                  |   Runtime ms  | Switches | Average delay ms | Maximum delay ms | Maximum delay at     |\n");
+	printf(" ---------------------------------------------------------------------------------------------------------------\n");
 
 	next = rb_first(&sorted_atom_root);
 
@@ -1902,13 +1900,18 @@ static int __cmd_record(int argc, const char **argv)
 
 int cmd_sched(int argc, const char **argv, const char *prefix __used)
 {
-	symbol__init(0);
-
 	argc = parse_options(argc, argv, sched_options, sched_usage,
 			     PARSE_OPT_STOP_AT_NON_OPTION);
 	if (!argc)
 		usage_with_options(sched_usage, sched_options);
 
+	/*
+	 * Aliased to 'perf trace' for now:
+	 */
+	if (!strcmp(argv[0], "trace"))
+		return cmd_trace(argc, argv, prefix);
+
+	symbol__init();
 	if (!strncmp(argv[0], "rec", 3)) {
 		return __cmd_record(argc, argv);
 	} else if (!strncmp(argv[0], "lat", 3)) {
@@ -1932,11 +1935,6 @@ int cmd_sched(int argc, const char **argv, const char *prefix __used)
 				usage_with_options(replay_usage, replay_options);
 		}
 		__cmd_replay();
-	} else if (!strcmp(argv[0], "trace")) {
-		/*
-		 * Aliased to 'perf trace' for now:
-		 */
-		return cmd_trace(argc, argv, prefix);
 	} else {
 		usage_with_options(sched_usage, sched_options);
 	}
diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c
index cb58b6605fc..3f8bbcfb1e9 100644
--- a/tools/perf/builtin-timechart.c
+++ b/tools/perf/builtin-timechart.c
@@ -30,15 +30,12 @@
 #include "util/parse-options.h"
 #include "util/parse-events.h"
 #include "util/event.h"
-#include "util/data_map.h"
+#include "util/session.h"
 #include "util/svghelper.h"
 
 static char		const *input_name = "perf.data";
 static char		const *output_name = "output.svg";
 
-
-static u64		sample_type;
-
 static unsigned int	numcpus;
 static u64		min_freq;	/* Lowest CPU frequency seen */
 static u64		max_freq;	/* Highest CPU frequency seen */
@@ -281,33 +278,30 @@ static int cpus_cstate_state[MAX_CPUS];
 static u64 cpus_pstate_start_times[MAX_CPUS];
 static u64 cpus_pstate_state[MAX_CPUS];
 
-static int
-process_comm_event(event_t *event)
+static int process_comm_event(event_t *event, struct perf_session *session __used)
 {
-	pid_set_comm(event->comm.pid, event->comm.comm);
+	pid_set_comm(event->comm.tid, event->comm.comm);
 	return 0;
 }
-static int
-process_fork_event(event_t *event)
+
+static int process_fork_event(event_t *event, struct perf_session *session __used)
 {
 	pid_fork(event->fork.pid, event->fork.ppid, event->fork.time);
 	return 0;
 }
 
-static int
-process_exit_event(event_t *event)
+static int process_exit_event(event_t *event, struct perf_session *session __used)
 {
 	pid_exit(event->fork.pid, event->fork.time);
 	return 0;
 }
 
 struct trace_entry {
-	u32			size;
 	unsigned short		type;
 	unsigned char		flags;
 	unsigned char		preempt_count;
 	int			pid;
-	int			tgid;
+	int			lock_depth;
 };
 
 struct power_entry {
@@ -481,46 +475,24 @@ static void sched_switch(int cpu, u64 timestamp, struct trace_entry *te)
 }
 
 
-static int
-process_sample_event(event_t *event)
+static int process_sample_event(event_t *event, struct perf_session *session)
 {
-	int cursor = 0;
-	u64 addr = 0;
-	u64 stamp = 0;
-	u32 cpu = 0;
-	u32 pid = 0;
+	struct sample_data data;
 	struct trace_entry *te;
 
-	if (sample_type & PERF_SAMPLE_IP)
-		cursor++;
-
-	if (sample_type & PERF_SAMPLE_TID) {
-		pid = event->sample.array[cursor]>>32;
-		cursor++;
-	}
-	if (sample_type & PERF_SAMPLE_TIME) {
-		stamp = event->sample.array[cursor++];
+	memset(&data, 0, sizeof(data));
 
-		if (!first_time || first_time > stamp)
-			first_time = stamp;
-		if (last_time < stamp)
-			last_time = stamp;
+	event__parse_sample(event, session->sample_type, &data);
 
+	if (session->sample_type & PERF_SAMPLE_TIME) {
+		if (!first_time || first_time > data.time)
+			first_time = data.time;
+		if (last_time < data.time)
+			last_time = data.time;
 	}
-	if (sample_type & PERF_SAMPLE_ADDR)
-		addr = event->sample.array[cursor++];
-	if (sample_type & PERF_SAMPLE_ID)
-		cursor++;
-	if (sample_type & PERF_SAMPLE_STREAM_ID)
-		cursor++;
-	if (sample_type & PERF_SAMPLE_CPU)
-		cpu = event->sample.array[cursor++] & 0xFFFFFFFF;
-	if (sample_type & PERF_SAMPLE_PERIOD)
-		cursor++;
-
-	te = (void *)&event->sample.array[cursor];
 
-	if (sample_type & PERF_SAMPLE_RAW && te->size > 0) {
+	te = (void *)data.raw_data;
+	if (session->sample_type & PERF_SAMPLE_RAW && data.raw_size > 0) {
 		char *event_str;
 		struct power_entry *pe;
 
@@ -532,19 +504,19 @@ process_sample_event(event_t *event)
 			return 0;
 
 		if (strcmp(event_str, "power:power_start") == 0)
-			c_state_start(cpu, stamp, pe->value);
+			c_state_start(data.cpu, data.time, pe->value);
 
 		if (strcmp(event_str, "power:power_end") == 0)
-			c_state_end(cpu, stamp);
+			c_state_end(data.cpu, data.time);
 
 		if (strcmp(event_str, "power:power_frequency") == 0)
-			p_state_change(cpu, stamp, pe->value);
+			p_state_change(data.cpu, data.time, pe->value);
 
 		if (strcmp(event_str, "sched:sched_wakeup") == 0)
-			sched_wakeup(cpu, stamp, pid, te);
+			sched_wakeup(data.cpu, data.time, data.pid, te);
 
 		if (strcmp(event_str, "sched:sched_switch") == 0)
-			sched_switch(cpu, stamp, te);
+			sched_switch(data.cpu, data.time, te);
 	}
 	return 0;
 }
@@ -597,16 +569,16 @@ static void end_sample_processing(void)
 	}
 }
 
-static u64 sample_time(event_t *event)
+static u64 sample_time(event_t *event, const struct perf_session *session)
 {
 	int cursor;
 
 	cursor = 0;
-	if (sample_type & PERF_SAMPLE_IP)
+	if (session->sample_type & PERF_SAMPLE_IP)
 		cursor++;
-	if (sample_type & PERF_SAMPLE_TID)
+	if (session->sample_type & PERF_SAMPLE_TID)
 		cursor++;
-	if (sample_type & PERF_SAMPLE_TIME)
+	if (session->sample_type & PERF_SAMPLE_TIME)
 		return event->sample.array[cursor];
 	return 0;
 }
@@ -616,8 +588,7 @@ static u64 sample_time(event_t *event)
  * We first queue all events, sorted backwards by insertion.
  * The order will get flipped later.
  */
-static int
-queue_sample_event(event_t *event)
+static int queue_sample_event(event_t *event, struct perf_session *session)
 {
 	struct sample_wrapper *copy, *prev;
 	int size;
@@ -631,7 +602,7 @@ queue_sample_event(event_t *event)
 	memset(copy, 0, size);
 
 	copy->next = NULL;
-	copy->timestamp = sample_time(event);
+	copy->timestamp = sample_time(event, session);
 
 	memcpy(&copy->data, event, event->sample.header.size);
 
@@ -1043,7 +1014,7 @@ static void write_svg_file(const char *filename)
 	svg_close();
 }
 
-static void process_samples(void)
+static void process_samples(struct perf_session *session)
 {
 	struct sample_wrapper *cursor;
 	event_t *event;
@@ -1054,15 +1025,13 @@ static void process_samples(void)
 	while (cursor) {
 		event = (void *)&cursor->data;
 		cursor = cursor->next;
-		process_sample_event(event);
+		process_sample_event(event, session);
 	}
 }
 
-static int sample_type_check(u64 type)
+static int sample_type_check(struct perf_session *session)
 {
-	sample_type = type;
-
-	if (!(sample_type & PERF_SAMPLE_RAW)) {
+	if (!(session->sample_type & PERF_SAMPLE_RAW)) {
 		fprintf(stderr, "No trace samples found in the file.\n"
 				"Have you used 'perf timechart record' to record it?\n");
 		return -1;
@@ -1071,7 +1040,7 @@ static int sample_type_check(u64 type)
 	return 0;
 }
 
-static struct perf_file_handler file_handler = {
+static struct perf_event_ops event_ops = {
 	.process_comm_event	= process_comm_event,
 	.process_fork_event	= process_fork_event,
 	.process_exit_event	= process_exit_event,
@@ -1081,17 +1050,17 @@ static struct perf_file_handler file_handler = {
 
 static int __cmd_timechart(void)
 {
-	struct perf_header *header;
+	struct perf_session *session = perf_session__new(input_name, O_RDONLY, 0);
 	int ret;
 
-	register_perf_file_handler(&file_handler);
+	if (session == NULL)
+		return -ENOMEM;
 
-	ret = mmap_dispatch_perf_file(&header, input_name, 0, 0,
-				      &event__cwdlen, &event__cwd);
+	ret = perf_session__process_events(session, &event_ops);
 	if (ret)
-		return EXIT_FAILURE;
+		goto out_delete;
 
-	process_samples();
+	process_samples(session);
 
 	end_sample_processing();
 
@@ -1101,8 +1070,9 @@ static int __cmd_timechart(void)
 
 	pr_info("Written %2.1f seconds of trace to %s.\n",
 		(last_time - first_time) / 1000000000.0, output_name);
-
-	return EXIT_SUCCESS;
+out_delete:
+	perf_session__delete(session);
+	return ret;
 }
 
 static const char * const timechart_usage[] = {
@@ -1167,11 +1137,11 @@ static const struct option options[] = {
 
 int cmd_timechart(int argc, const char **argv, const char *prefix __used)
 {
-	symbol__init(0);
-
 	argc = parse_options(argc, argv, options, timechart_usage,
 			PARSE_OPT_STOP_AT_NON_OPTION);
 
+	symbol__init();
+
 	if (argc && !strncmp(argv[0], "rec", 3))
 		return __cmd_record(argc, argv);
 	else if (argc)
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index e0a374d0e43..ddc584b6487 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -20,8 +20,9 @@
 
 #include "perf.h"
 
-#include "util/symbol.h"
 #include "util/color.h"
+#include "util/session.h"
+#include "util/symbol.h"
 #include "util/thread.h"
 #include "util/util.h"
 #include <linux/rbtree.h>
@@ -79,7 +80,6 @@ static int			dump_symtab                     =      0;
 static bool			hide_kernel_symbols		=  false;
 static bool			hide_user_symbols		=  false;
 static struct winsize		winsize;
-struct symbol_conf		symbol_conf;
 
 /*
  * Source
@@ -926,7 +926,8 @@ static int symbol_filter(struct map *map, struct symbol *sym)
 	return 0;
 }
 
-static void event__process_sample(const event_t *self, int counter)
+static void event__process_sample(const event_t *self,
+				 struct perf_session *session, int counter)
 {
 	u64 ip = self->ip.ip;
 	struct sym_entry *syme;
@@ -946,8 +947,8 @@ static void event__process_sample(const event_t *self, int counter)
 		return;
 	}
 
-	if (event__preprocess_sample(self, &al, symbol_filter) < 0 ||
-	    al.sym == NULL)
+	if (event__preprocess_sample(self, session, &al, symbol_filter) < 0 ||
+	    al.sym == NULL || al.filtered)
 		return;
 
 	syme = symbol__priv(al.sym);
@@ -965,14 +966,14 @@ static void event__process_sample(const event_t *self, int counter)
 	}
 }
 
-static int event__process(event_t *event)
+static int event__process(event_t *event, struct perf_session *session)
 {
 	switch (event->header.type) {
 	case PERF_RECORD_COMM:
-		event__process_comm(event);
+		event__process_comm(event, session);
 		break;
 	case PERF_RECORD_MMAP:
-		event__process_mmap(event);
+		event__process_mmap(event, session);
 		break;
 	default:
 		break;
@@ -999,7 +1000,8 @@ static unsigned int mmap_read_head(struct mmap_data *md)
 	return head;
 }
 
-static void mmap_read_counter(struct mmap_data *md)
+static void perf_session__mmap_read_counter(struct perf_session *self,
+					    struct mmap_data *md)
 {
 	unsigned int head = mmap_read_head(md);
 	unsigned int old = md->prev;
@@ -1052,9 +1054,9 @@ static void mmap_read_counter(struct mmap_data *md)
 		}
 
 		if (event->header.type == PERF_RECORD_SAMPLE)
-			event__process_sample(event, md->counter);
+			event__process_sample(event, self, md->counter);
 		else
-			event__process(event);
+			event__process(event, self);
 		old += size;
 	}
 
@@ -1064,13 +1066,13 @@ static void mmap_read_counter(struct mmap_data *md)
 static struct pollfd event_array[MAX_NR_CPUS * MAX_COUNTERS];
 static struct mmap_data mmap_array[MAX_NR_CPUS][MAX_COUNTERS];
 
-static void mmap_read(void)
+static void perf_session__mmap_read(struct perf_session *self)
 {
 	int i, counter;
 
 	for (i = 0; i < nr_cpus; i++) {
 		for (counter = 0; counter < nr_counters; counter++)
-			mmap_read_counter(&mmap_array[i][counter]);
+			perf_session__mmap_read_counter(self, &mmap_array[i][counter]);
 	}
 }
 
@@ -1155,11 +1157,18 @@ static int __cmd_top(void)
 	pthread_t thread;
 	int i, counter;
 	int ret;
+	/*
+	 * FIXME: perf_session__new should allow passing a O_MMAP, so that all this
+	 * mmap reading, etc is encapsulated in it. Use O_WRONLY for now.
+	 */
+	struct perf_session *session = perf_session__new(NULL, O_WRONLY, false);
+	if (session == NULL)
+		return -ENOMEM;
 
 	if (target_pid != -1)
-		event__synthesize_thread(target_pid, event__process);
+		event__synthesize_thread(target_pid, event__process, session);
 	else
-		event__synthesize_threads(event__process);
+		event__synthesize_threads(event__process, session);
 
 	for (i = 0; i < nr_cpus; i++) {
 		group_fd = -1;
@@ -1170,7 +1179,7 @@ static int __cmd_top(void)
 	/* Wait for a minimal set of events before starting the snapshot */
 	poll(event_array, nr_poll, 100);
 
-	mmap_read();
+	perf_session__mmap_read(session);
 
 	if (pthread_create(&thread, NULL, display_thread, NULL)) {
 		printf("Could not create display thread.\n");
@@ -1190,7 +1199,7 @@ static int __cmd_top(void)
 	while (1) {
 		int hits = samples;
 
-		mmap_read();
+		perf_session__mmap_read(session);
 
 		if (hits == samples)
 			ret = poll(event_array, nr_poll, 100);
@@ -1273,7 +1282,7 @@ int cmd_top(int argc, const char **argv, const char *prefix __used)
 				 (nr_counters + 1) * sizeof(unsigned long));
 	if (symbol_conf.vmlinux_name == NULL)
 		symbol_conf.try_vmlinux_path = true;
-	if (symbol__init(&symbol_conf) < 0)
+	if (symbol__init() < 0)
 		return -1;
 
 	if (delay_secs < 1)
diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index abb914aa7be..574a215e800 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -7,11 +7,14 @@
 #include "util/header.h"
 #include "util/exec_cmd.h"
 #include "util/trace-event.h"
+#include "util/session.h"
 
 static char const		*script_name;
 static char const		*generate_script_lang;
 
-static int default_start_script(const char *script __attribute((unused)))
+static int default_start_script(const char *script __unused,
+				int argc __unused,
+				const char **argv __unused)
 {
 	return 0;
 }
@@ -21,7 +24,7 @@ static int default_stop_script(void)
 	return 0;
 }
 
-static int default_generate_script(const char *outfile __attribute ((unused)))
+static int default_generate_script(const char *outfile __unused)
 {
 	return 0;
 }
@@ -56,77 +59,53 @@ static int cleanup_scripting(void)
 #include "util/debug.h"
 
 #include "util/trace-event.h"
-#include "util/data_map.h"
 #include "util/exec_cmd.h"
 
 static char const		*input_name = "perf.data";
 
-static struct perf_header	*header;
-static u64			sample_type;
-
-static int process_sample_event(event_t *event)
+static int process_sample_event(event_t *event, struct perf_session *session)
 {
-	u64 ip = event->ip.ip;
-	u64 timestamp = -1;
-	u32 cpu = -1;
-	u64 period = 1;
-	void *more_data = event->ip.__more_data;
-	struct thread *thread = threads__findnew(event->ip.pid);
-
-	if (sample_type & PERF_SAMPLE_TIME) {
-		timestamp = *(u64 *)more_data;
-		more_data += sizeof(u64);
-	}
+	struct sample_data data;
+	struct thread *thread;
 
-	if (sample_type & PERF_SAMPLE_CPU) {
-		cpu = *(u32 *)more_data;
-		more_data += sizeof(u32);
-		more_data += sizeof(u32); /* reserved */
-	}
+	memset(&data, 0, sizeof(data));
+	data.time = -1;
+	data.cpu = -1;
+	data.period = 1;
 
-	if (sample_type & PERF_SAMPLE_PERIOD) {
-		period = *(u64 *)more_data;
-		more_data += sizeof(u64);
-	}
+	event__parse_sample(event, session->sample_type, &data);
 
 	dump_printf("(IP, %d): %d/%d: %p period: %Ld\n",
 		event->header.misc,
-		event->ip.pid, event->ip.tid,
-		(void *)(long)ip,
-		(long long)period);
+		data.pid, data.tid,
+		(void *)(long)data.ip,
+		(long long)data.period);
 
+	thread = perf_session__findnew(session, event->ip.pid);
 	if (thread == NULL) {
 		pr_debug("problem processing %d event, skipping it.\n",
 			 event->header.type);
 		return -1;
 	}
 
-	dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid);
-
-	if (sample_type & PERF_SAMPLE_RAW) {
-		struct {
-			u32 size;
-			char data[0];
-		} *raw = more_data;
-
+	if (session->sample_type & PERF_SAMPLE_RAW) {
 		/*
 		 * FIXME: better resolve from pid from the struct trace_entry
 		 * field, although it should be the same than this perf
 		 * event pid
 		 */
-		scripting_ops->process_event(cpu, raw->data, raw->size,
-					     timestamp, thread->comm);
+		scripting_ops->process_event(data.cpu, data.raw_data,
+					     data.raw_size,
+					     data.time, thread->comm);
 	}
-	event__stats.total += period;
 
+	session->events_stats.total += data.period;
 	return 0;
 }
 
-static int sample_type_check(u64 type)
+static int sample_type_check(struct perf_session *session)
 {
-	sample_type = type;
-
-	if (!(sample_type & PERF_SAMPLE_RAW)) {
+	if (!(session->sample_type & PERF_SAMPLE_RAW)) {
 		fprintf(stderr,
 			"No trace sample to read. Did you call perf record "
 			"without -R?");
@@ -136,19 +115,15 @@ static int sample_type_check(u64 type)
 	return 0;
 }
 
-static struct perf_file_handler file_handler = {
+static struct perf_event_ops event_ops = {
 	.process_sample_event	= process_sample_event,
 	.process_comm_event	= event__process_comm,
 	.sample_type_check	= sample_type_check,
 };
 
-static int __cmd_trace(void)
+static int __cmd_trace(struct perf_session *session)
 {
-	register_idle_thread();
-	register_perf_file_handler(&file_handler);
-
-	return mmap_dispatch_perf_file(&header, input_name,
-				       0, 0, &event__cwdlen, &event__cwd);
+	return perf_session__process_events(session, &event_ops);
 }
 
 struct script_spec {
@@ -299,7 +274,245 @@ static int parse_scriptname(const struct option *opt __used,
 	return 0;
 }
 
-static const char * const annotate_usage[] = {
+#define for_each_lang(scripts_dir, lang_dirent, lang_next)		\
+	while (!readdir_r(scripts_dir, &lang_dirent, &lang_next) &&	\
+	       lang_next)						\
+		if (lang_dirent.d_type == DT_DIR &&			\
+		    (strcmp(lang_dirent.d_name, ".")) &&		\
+		    (strcmp(lang_dirent.d_name, "..")))
+
+#define for_each_script(lang_dir, script_dirent, script_next)		\
+	while (!readdir_r(lang_dir, &script_dirent, &script_next) &&	\
+	       script_next)						\
+		if (script_dirent.d_type != DT_DIR)
+
+
+#define RECORD_SUFFIX			"-record"
+#define REPORT_SUFFIX			"-report"
+
+struct script_desc {
+	struct list_head	node;
+	char			*name;
+	char			*half_liner;
+	char			*args;
+};
+
+LIST_HEAD(script_descs);
+
+static struct script_desc *script_desc__new(const char *name)
+{
+	struct script_desc *s = zalloc(sizeof(*s));
+
+	if (s != NULL)
+		s->name = strdup(name);
+
+	return s;
+}
+
+static void script_desc__delete(struct script_desc *s)
+{
+	free(s->name);
+	free(s);
+}
+
+static void script_desc__add(struct script_desc *s)
+{
+	list_add_tail(&s->node, &script_descs);
+}
+
+static struct script_desc *script_desc__find(const char *name)
+{
+	struct script_desc *s;
+
+	list_for_each_entry(s, &script_descs, node)
+		if (strcasecmp(s->name, name) == 0)
+			return s;
+	return NULL;
+}
+
+static struct script_desc *script_desc__findnew(const char *name)
+{
+	struct script_desc *s = script_desc__find(name);
+
+	if (s)
+		return s;
+
+	s = script_desc__new(name);
+	if (!s)
+		goto out_delete_desc;
+
+	script_desc__add(s);
+
+	return s;
+
+out_delete_desc:
+	script_desc__delete(s);
+
+	return NULL;
+}
+
+static char *ends_with(char *str, const char *suffix)
+{
+	size_t suffix_len = strlen(suffix);
+	char *p = str;
+
+	if (strlen(str) > suffix_len) {
+		p = str + strlen(str) - suffix_len;
+		if (!strncmp(p, suffix, suffix_len))
+			return p;
+	}
+
+	return NULL;
+}
+
+static char *ltrim(char *str)
+{
+	int len = strlen(str);
+
+	while (len && isspace(*str)) {
+		len--;
+		str++;
+	}
+
+	return str;
+}
+
+static int read_script_info(struct script_desc *desc, const char *filename)
+{
+	char line[BUFSIZ], *p;
+	FILE *fp;
+
+	fp = fopen(filename, "r");
+	if (!fp)
+		return -1;
+
+	while (fgets(line, sizeof(line), fp)) {
+		p = ltrim(line);
+		if (strlen(p) == 0)
+			continue;
+		if (*p != '#')
+			continue;
+		p++;
+		if (strlen(p) && *p == '!')
+			continue;
+
+		p = ltrim(p);
+		if (strlen(p) && p[strlen(p) - 1] == '\n')
+			p[strlen(p) - 1] = '\0';
+
+		if (!strncmp(p, "description:", strlen("description:"))) {
+			p += strlen("description:");
+			desc->half_liner = strdup(ltrim(p));
+			continue;
+		}
+
+		if (!strncmp(p, "args:", strlen("args:"))) {
+			p += strlen("args:");
+			desc->args = strdup(ltrim(p));
+			continue;
+		}
+	}
+
+	fclose(fp);
+
+	return 0;
+}
+
+static int list_available_scripts(const struct option *opt __used,
+				  const char *s __used, int unset __used)
+{
+	struct dirent *script_next, *lang_next, script_dirent, lang_dirent;
+	char scripts_path[MAXPATHLEN];
+	DIR *scripts_dir, *lang_dir;
+	char script_path[MAXPATHLEN];
+	char lang_path[MAXPATHLEN];
+	struct script_desc *desc;
+	char first_half[BUFSIZ];
+	char *script_root;
+	char *str;
+
+	snprintf(scripts_path, MAXPATHLEN, "%s/scripts", perf_exec_path());
+
+	scripts_dir = opendir(scripts_path);
+	if (!scripts_dir)
+		return -1;
+
+	for_each_lang(scripts_dir, lang_dirent, lang_next) {
+		snprintf(lang_path, MAXPATHLEN, "%s/%s/bin", scripts_path,
+			 lang_dirent.d_name);
+		lang_dir = opendir(lang_path);
+		if (!lang_dir)
+			continue;
+
+		for_each_script(lang_dir, script_dirent, script_next) {
+			script_root = strdup(script_dirent.d_name);
+			str = ends_with(script_root, REPORT_SUFFIX);
+			if (str) {
+				*str = '\0';
+				desc = script_desc__findnew(script_root);
+				snprintf(script_path, MAXPATHLEN, "%s/%s",
+					 lang_path, script_dirent.d_name);
+				read_script_info(desc, script_path);
+			}
+			free(script_root);
+		}
+	}
+
+	fprintf(stdout, "List of available trace scripts:\n");
+	list_for_each_entry(desc, &script_descs, node) {
+		sprintf(first_half, "%s %s", desc->name,
+			desc->args ? desc->args : "");
+		fprintf(stdout, "  %-36s %s\n", first_half,
+			desc->half_liner ? desc->half_liner : "");
+	}
+
+	exit(0);
+}
+
+static char *get_script_path(const char *script_root, const char *suffix)
+{
+	struct dirent *script_next, *lang_next, script_dirent, lang_dirent;
+	char scripts_path[MAXPATHLEN];
+	char script_path[MAXPATHLEN];
+	DIR *scripts_dir, *lang_dir;
+	char lang_path[MAXPATHLEN];
+	char *str, *__script_root;
+	char *path = NULL;
+
+	snprintf(scripts_path, MAXPATHLEN, "%s/scripts", perf_exec_path());
+
+	scripts_dir = opendir(scripts_path);
+	if (!scripts_dir)
+		return NULL;
+
+	for_each_lang(scripts_dir, lang_dirent, lang_next) {
+		snprintf(lang_path, MAXPATHLEN, "%s/%s/bin", scripts_path,
+			 lang_dirent.d_name);
+		lang_dir = opendir(lang_path);
+		if (!lang_dir)
+			continue;
+
+		for_each_script(lang_dir, script_dirent, script_next) {
+			__script_root = strdup(script_dirent.d_name);
+			str = ends_with(__script_root, suffix);
+			if (str) {
+				*str = '\0';
+				if (strcmp(__script_root, script_root))
+					continue;
+				snprintf(script_path, MAXPATHLEN, "%s/%s",
+					 lang_path, script_dirent.d_name);
+				path = strdup(script_path);
+				free(__script_root);
+				break;
+			}
+			free(__script_root);
+		}
+	}
+
+	return path;
+}
+
+static const char * const trace_usage[] = {
 	"perf trace [<options>] <command>",
 	NULL
 };
@@ -309,8 +522,10 @@ static const struct option options[] = {
 		    "dump raw trace in ASCII"),
 	OPT_BOOLEAN('v', "verbose", &verbose,
 		    "be more verbose (show symbol address, etc)"),
-	OPT_BOOLEAN('l', "latency", &latency_format,
+	OPT_BOOLEAN('L', "Latency", &latency_format,
 		    "show latency attributes (irqs/preemption disabled, etc)"),
+	OPT_CALLBACK_NOOPT('l', "list", NULL, NULL, "list available scripts",
+			   list_available_scripts),
 	OPT_CALLBACK('s', "script", NULL, "name",
 		     "script file name (lang:script name, script name, or *)",
 		     parse_scriptname),
@@ -322,24 +537,61 @@ static const struct option options[] = {
 
 int cmd_trace(int argc, const char **argv, const char *prefix __used)
 {
-	int err;
+	struct perf_session *session;
+	const char *suffix = NULL;
+	const char **__argv;
+	char *script_path;
+	int i, err;
+
+	if (argc >= 2 && strncmp(argv[1], "rec", strlen("rec")) == 0) {
+		if (argc < 3) {
+			fprintf(stderr,
+				"Please specify a record script\n");
+			return -1;
+		}
+		suffix = RECORD_SUFFIX;
+	}
+
+	if (argc >= 2 && strncmp(argv[1], "rep", strlen("rep")) == 0) {
+		if (argc < 3) {
+			fprintf(stderr,
+				"Please specify a report script\n");
+			return -1;
+		}
+		suffix = REPORT_SUFFIX;
+	}
 
-	symbol__init(0);
+	if (suffix) {
+		script_path = get_script_path(argv[2], suffix);
+		if (!script_path) {
+			fprintf(stderr, "script not found\n");
+			return -1;
+		}
 
-	setup_scripting();
+		__argv = malloc((argc + 1) * sizeof(const char *));
+		__argv[0] = "/bin/sh";
+		__argv[1] = script_path;
+		for (i = 3; i < argc; i++)
+			__argv[i - 1] = argv[i];
+		__argv[argc - 1] = NULL;
 
-	argc = parse_options(argc, argv, options, annotate_usage, 0);
-	if (argc) {
-		/*
-		 * Special case: if there's an argument left then assume tha
-		 * it's a symbol filter:
-		 */
-		if (argc > 1)
-			usage_with_options(annotate_usage, options);
+		execvp("/bin/sh", (char **)__argv);
+		exit(-1);
 	}
 
+	setup_scripting();
+
+	argc = parse_options(argc, argv, options, trace_usage,
+			     PARSE_OPT_STOP_AT_NON_OPTION);
+
+	if (symbol__init() < 0)
+		return -1;
 	setup_pager();
 
+	session = perf_session__new(input_name, O_RDONLY, 0);
+	if (session == NULL)
+		return -ENOMEM;
+
 	if (generate_script_lang) {
 		struct stat perf_stat;
 
@@ -366,23 +618,20 @@ int cmd_trace(int argc, const char **argv, const char *prefix __used)
 			return -1;
 		}
 
-		header = perf_header__new();
-		if (header == NULL)
-			return -1;
-
-		perf_header__read(header, input);
+		perf_header__read(&session->header, input);
 		err = scripting_ops->generate_script("perf-trace");
 		goto out;
 	}
 
 	if (script_name) {
-		err = scripting_ops->start_script(script_name);
+		err = scripting_ops->start_script(script_name, argc, argv);
 		if (err)
 			goto out;
 	}
 
-	err = __cmd_trace();
+	err = __cmd_trace(session);
 
+	perf_session__delete(session);
 	cleanup_scripting();
 out:
 	return err;
diff --git a/tools/perf/builtin.h b/tools/perf/builtin.h
index a3d8bf65f26..18035b1f16c 100644
--- a/tools/perf/builtin.h
+++ b/tools/perf/builtin.h
@@ -17,6 +17,7 @@ extern int check_pager_config(const char *cmd);
 extern int cmd_annotate(int argc, const char **argv, const char *prefix);
 extern int cmd_bench(int argc, const char **argv, const char *prefix);
 extern int cmd_buildid_list(int argc, const char **argv, const char *prefix);
+extern int cmd_diff(int argc, const char **argv, const char *prefix);
 extern int cmd_help(int argc, const char **argv, const char *prefix);
 extern int cmd_sched(int argc, const char **argv, const char *prefix);
 extern int cmd_list(int argc, const char **argv, const char *prefix);
diff --git a/tools/perf/command-list.txt b/tools/perf/command-list.txt
index 02b09ea17a3..71dc7c3fe7b 100644
--- a/tools/perf/command-list.txt
+++ b/tools/perf/command-list.txt
@@ -5,6 +5,7 @@
 perf-annotate			mainporcelain common
 perf-bench			mainporcelain common
 perf-buildid-list		mainporcelain common
+perf-diff			mainporcelain common
 perf-list			mainporcelain common
 perf-sched			mainporcelain common
 perf-record			mainporcelain common
diff --git a/tools/perf/design.txt b/tools/perf/design.txt
index f000c30877a..8d0de5130db 100644
--- a/tools/perf/design.txt
+++ b/tools/perf/design.txt
@@ -21,7 +21,7 @@ There's one file descriptor per virtual counter used.
 The special file descriptor is opened via the perf_event_open()
 system call:
 
-   int sys_perf_event_open(struct perf_event_hw_event *hw_event_uptr,
+   int sys_perf_event_open(struct perf_event_attr *hw_event_uptr,
 			     pid_t pid, int cpu, int group_fd,
 			     unsigned long flags);
 
@@ -32,9 +32,9 @@ can be used to set the blocking mode, etc.
 Multiple counters can be kept open at a time, and the counters
 can be poll()ed.
 
-When creating a new counter fd, 'perf_event_hw_event' is:
+When creating a new counter fd, 'perf_event_attr' is:
 
-struct perf_event_hw_event {
+struct perf_event_attr {
         /*
          * The MSB of the config word signifies if the rest contains cpu
          * specific (raw) counter configuration data, if unset, the next
@@ -399,7 +399,7 @@ Notification of new events is possible through poll()/select()/epoll() and
 fcntl() managing signals.
 
 Normally a notification is generated for every page filled, however one can
-additionally set perf_event_hw_event.wakeup_events to generate one every
+additionally set perf_event_attr.wakeup_events to generate one every
 so many counter overflow events.
 
 Future work will include a splice() interface to the ring-buffer.
diff --git a/tools/perf/perf.c b/tools/perf/perf.c
index cf64049bc9b..873e55fab37 100644
--- a/tools/perf/perf.c
+++ b/tools/perf/perf.c
@@ -286,6 +286,7 @@ static void handle_internal_command(int argc, const char **argv)
 	const char *cmd = argv[0];
 	static struct cmd_struct commands[] = {
 		{ "buildid-list", cmd_buildid_list, 0 },
+		{ "diff",	cmd_diff,	0 },
 		{ "help",	cmd_help,	0 },
 		{ "list",	cmd_list,	0 },
 		{ "record",	cmd_record,	0 },
diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index 454d5d55f32..75f941bfba9 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -59,6 +59,18 @@
 #define cpu_relax()	asm volatile ("hint @pause" ::: "memory")
 #endif
 
+#ifdef __arm__
+#include "../../arch/arm/include/asm/unistd.h"
+/*
+ * Use the __kuser_memory_barrier helper in the CPU helper page. See
+ * arch/arm/kernel/entry-armv.S in the kernel source for details.
+ */
+#define rmb()		asm volatile("mov r0, #0xffff0fff; mov lr, pc;" \
+				     "sub pc, r0, #95" ::: "r0", "lr", "cc", \
+				     "memory")
+#define cpu_relax()	asm volatile("":::"memory")
+#endif
+
 #include <time.h>
 #include <unistd.h>
 #include <sys/types.h>
diff --git a/tools/perf/scripts/perl/bin/check-perf-trace-report b/tools/perf/scripts/perl/bin/check-perf-trace-report
index 89948b01502..7fc4a033dd4 100644
--- a/tools/perf/scripts/perl/bin/check-perf-trace-report
+++ b/tools/perf/scripts/perl/bin/check-perf-trace-report
@@ -1,4 +1,5 @@
 #!/bin/bash
+# description: useless but exhaustive test script
 perf trace -s ~/libexec/perf-core/scripts/perl/check-perf-trace.pl
 
 
diff --git a/tools/perf/scripts/perl/bin/rw-by-file-report b/tools/perf/scripts/perl/bin/rw-by-file-report
index f5dcf9cb5bd..eddb9ccce6a 100644
--- a/tools/perf/scripts/perl/bin/rw-by-file-report
+++ b/tools/perf/scripts/perl/bin/rw-by-file-report
@@ -1,5 +1,7 @@
 #!/bin/bash
-perf trace -s ~/libexec/perf-core/scripts/perl/rw-by-file.pl
+# description: r/w activity for a program, by file
+# args: <comm>
+perf trace -s ~/libexec/perf-core/scripts/perl/rw-by-file.pl $1
 
 
 
diff --git a/tools/perf/scripts/perl/bin/rw-by-pid-report b/tools/perf/scripts/perl/bin/rw-by-pid-report
index cea16f78a3a..7f44c25cc85 100644
--- a/tools/perf/scripts/perl/bin/rw-by-pid-report
+++ b/tools/perf/scripts/perl/bin/rw-by-pid-report
@@ -1,4 +1,5 @@
 #!/bin/bash
+# description: system-wide r/w activity
 perf trace -s ~/libexec/perf-core/scripts/perl/rw-by-pid.pl
 
 
diff --git a/tools/perf/scripts/perl/bin/wakeup-latency-report b/tools/perf/scripts/perl/bin/wakeup-latency-report
index 85769dc456e..fce3adcb324 100644
--- a/tools/perf/scripts/perl/bin/wakeup-latency-report
+++ b/tools/perf/scripts/perl/bin/wakeup-latency-report
@@ -1,4 +1,5 @@
 #!/bin/bash
+# description: system-wide min/max/avg wakeup latency
 perf trace -s ~/libexec/perf-core/scripts/perl/wakeup-latency.pl
 
 
diff --git a/tools/perf/scripts/perl/bin/workqueue-stats-report b/tools/perf/scripts/perl/bin/workqueue-stats-report
index aa68435be92..71cfbd182fb 100644
--- a/tools/perf/scripts/perl/bin/workqueue-stats-report
+++ b/tools/perf/scripts/perl/bin/workqueue-stats-report
@@ -1,4 +1,5 @@
 #!/bin/bash
+# description: workqueue stats (ins/exe/create/destroy)
 perf trace -s ~/libexec/perf-core/scripts/perl/workqueue-stats.pl
 
 
diff --git a/tools/perf/scripts/perl/rw-by-file.pl b/tools/perf/scripts/perl/rw-by-file.pl
index 61f91561d84..2a39097687b 100644
--- a/tools/perf/scripts/perl/rw-by-file.pl
+++ b/tools/perf/scripts/perl/rw-by-file.pl
@@ -18,8 +18,9 @@ use lib "./Perf-Trace-Util/lib";
 use Perf::Trace::Core;
 use Perf::Trace::Util;
 
-# change this to the comm of the program you're interested in
-my $for_comm = "perf";
+my $usage = "perf trace -s rw-by-file.pl <comm>\n";
+
+my $for_comm = shift or die $usage;
 
 my %reads;
 my %writes;
diff --git a/tools/perf/util/data_map.c b/tools/perf/util/data_map.c
index ca0bedf637c..b557b836de3 100644
--- a/tools/perf/util/data_map.c
+++ b/tools/perf/util/data_map.c
@@ -1,20 +1,17 @@
-#include "data_map.h"
 #include "symbol.h"
 #include "util.h"
 #include "debug.h"
+#include "thread.h"
+#include "session.h"
 
-
-static struct perf_file_handler *curr_handler;
-static unsigned long	mmap_window = 32;
-static char		__cwd[PATH_MAX];
-
-static int process_event_stub(event_t *event __used)
+static int process_event_stub(event_t *event __used,
+			      struct perf_session *session __used)
 {
 	dump_printf(": unhandled!\n");
 	return 0;
 }
 
-void register_perf_file_handler(struct perf_file_handler *handler)
+static void perf_event_ops__fill_defaults(struct perf_event_ops *handler)
 {
 	if (!handler->process_sample_event)
 		handler->process_sample_event = process_event_stub;
@@ -34,8 +31,6 @@ void register_perf_file_handler(struct perf_file_handler *handler)
 		handler->process_throttle_event = process_event_stub;
 	if (!handler->process_unthrottle_event)
 		handler->process_unthrottle_event = process_event_stub;
-
-	curr_handler = handler;
 }
 
 static const char *event__name[] = {
@@ -61,8 +56,9 @@ void event__print_totals(void)
 			event__name[i], event__total[i]);
 }
 
-static int
-process_event(event_t *event, unsigned long offset, unsigned long head)
+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);
 
@@ -77,34 +73,34 @@ process_event(event_t *event, unsigned long offset, unsigned long head)
 
 	switch (event->header.type) {
 	case PERF_RECORD_SAMPLE:
-		return curr_handler->process_sample_event(event);
+		return ops->process_sample_event(event, session);
 	case PERF_RECORD_MMAP:
-		return curr_handler->process_mmap_event(event);
+		return ops->process_mmap_event(event, session);
 	case PERF_RECORD_COMM:
-		return curr_handler->process_comm_event(event);
+		return ops->process_comm_event(event, session);
 	case PERF_RECORD_FORK:
-		return curr_handler->process_fork_event(event);
+		return ops->process_fork_event(event, session);
 	case PERF_RECORD_EXIT:
-		return curr_handler->process_exit_event(event);
+		return ops->process_exit_event(event, session);
 	case PERF_RECORD_LOST:
-		return curr_handler->process_lost_event(event);
+		return ops->process_lost_event(event, session);
 	case PERF_RECORD_READ:
-		return curr_handler->process_read_event(event);
+		return ops->process_read_event(event, session);
 	case PERF_RECORD_THROTTLE:
-		return curr_handler->process_throttle_event(event);
+		return ops->process_throttle_event(event, session);
 	case PERF_RECORD_UNTHROTTLE:
-		return curr_handler->process_unthrottle_event(event);
+		return ops->process_unthrottle_event(event, session);
 	default:
-		curr_handler->total_unknown++;
+		ops->total_unknown++;
 		return -1;
 	}
 }
 
-int perf_header__read_build_ids(int input, off_t offset, off_t size)
+int perf_header__read_build_ids(int input, u64 offset, u64 size)
 {
 	struct build_id_event bev;
 	char filename[PATH_MAX];
-	off_t limit = offset + size;
+	u64 limit = offset + size;
 	int err = -1;
 
 	while (offset < limit) {
@@ -129,88 +125,58 @@ out:
 	return err;
 }
 
-int mmap_dispatch_perf_file(struct perf_header **pheader,
-			    const char *input_name,
-			    int force,
-			    int full_paths,
-			    int *cwdlen,
-			    char **cwd)
+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;
-	struct perf_header *header;
 	unsigned long head, shift;
 	unsigned long offset = 0;
-	struct stat input_stat;
 	size_t	page_size;
-	u64 sample_type;
 	event_t *event;
 	uint32_t size;
-	int input;
 	char *buf;
 
-	if (curr_handler == NULL) {
-		pr_debug("Forgot to register perf file handler\n");
-		return -EINVAL;
-	}
-
-	page_size = getpagesize();
-
-	input = open(input_name, O_RDONLY);
-	if (input < 0) {
-		pr_err("Failed to open file: %s", input_name);
-		if (!strcmp(input_name, "perf.data"))
-			pr_err("  (try 'perf record' first)");
-		pr_err("\n");
-		return -errno;
-	}
-
-	if (fstat(input, &input_stat) < 0) {
-		pr_err("failed to stat file");
-		err = -errno;
-		goto out_close;
-	}
-
-	err = -EACCES;
-	if (!force && input_stat.st_uid && (input_stat.st_uid != geteuid())) {
-		pr_err("file: %s not owned by current user or root\n",
-			input_name);
-		goto out_close;
-	}
-
-	if (input_stat.st_size == 0) {
-		pr_info("zero-sized file, nothing to do!\n");
-		goto done;
-	}
+	if (perf_session__register_idle_thread(self) == NULL)
+		return -ENOMEM;
 
-	err = -ENOMEM;
-	header = perf_header__new();
-	if (header == NULL)
-		goto out_close;
+	perf_event_ops__fill_defaults(ops);
 
-	err = perf_header__read(header, input);
-	if (err < 0)
-		goto out_delete;
-	*pheader = header;
-	head = header->data_offset;
+	page_size = getpagesize();
 
-	sample_type = perf_header__sample_type(header);
+	head = self->header.data_offset;
+	self->sample_type = perf_header__sample_type(&self->header);
 
 	err = -EINVAL;
-	if (curr_handler->sample_type_check &&
-	    curr_handler->sample_type_check(sample_type) < 0)
-		goto out_delete;
+	if (ops->sample_type_check && ops->sample_type_check(self) < 0)
+		goto out_err;
 
-	if (!full_paths) {
-		if (getcwd(__cwd, sizeof(__cwd)) == NULL) {
-			pr_err("failed to get the current directory\n");
+	if (!ops->full_paths) {
+		char bf[PATH_MAX];
+
+		if (getcwd(bf, sizeof(bf)) == NULL) {
 			err = -errno;
-			goto out_delete;
+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;
 		}
-		*cwd = __cwd;
-		*cwdlen = strlen(*cwd);
-	} else {
-		*cwd = NULL;
-		*cwdlen = 0;
+		self->cwdlen = strlen(self->cwd);
 	}
 
 	shift = page_size * (head / page_size);
@@ -218,12 +184,12 @@ int mmap_dispatch_perf_file(struct perf_header **pheader,
 	head -= shift;
 
 remap:
-	buf = mmap(NULL, page_size * mmap_window, PROT_READ,
-		   MAP_SHARED, input, offset);
+	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_delete;
+		goto out_err;
 	}
 
 more:
@@ -233,12 +199,12 @@ more:
 	if (!size)
 		size = 8;
 
-	if (head + event->header.size >= page_size * mmap_window) {
+	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 * mmap_window);
+		munmap_ret = munmap(buf, page_size * self->mmap_window);
 		assert(munmap_ret == 0);
 
 		offset += shift;
@@ -253,7 +219,7 @@ more:
 			(void *)(long)event->header.size,
 			event->header.type);
 
-	if (!size || process_event(event, offset, head) < 0) {
+	if (!size || process_event(event, self, ops, offset, head) < 0) {
 
 		dump_printf("%p [%p]: skipping unknown header type: %d\n",
 			(void *)(offset + head),
@@ -273,19 +239,14 @@ more:
 
 	head += size;
 
-	if (offset + head >= header->data_offset + header->data_size)
+	if (offset + head >= self->header.data_offset + self->header.data_size)
 		goto done;
 
-	if (offset + head < (unsigned long)input_stat.st_size)
+	if (offset + head < self->size)
 		goto more;
 
 done:
 	err = 0;
-out_close:
-	close(input);
-
+out_err:
 	return err;
-out_delete:
-	perf_header__delete(header);
-	goto out_close;
 }
diff --git a/tools/perf/util/data_map.h b/tools/perf/util/data_map.h
deleted file mode 100644
index 3180ff7e363..00000000000
--- a/tools/perf/util/data_map.h
+++ /dev/null
@@ -1,32 +0,0 @@
-#ifndef __PERF_DATAMAP_H
-#define __PERF_DATAMAP_H
-
-#include "event.h"
-#include "header.h"
-
-typedef int (*event_type_handler_t)(event_t *);
-
-struct perf_file_handler {
-	event_type_handler_t	process_sample_event;
-	event_type_handler_t	process_mmap_event;
-	event_type_handler_t	process_comm_event;
-	event_type_handler_t	process_fork_event;
-	event_type_handler_t	process_exit_event;
-	event_type_handler_t	process_lost_event;
-	event_type_handler_t	process_read_event;
-	event_type_handler_t	process_throttle_event;
-	event_type_handler_t	process_unthrottle_event;
-	int			(*sample_type_check)(u64 sample_type);
-	unsigned long		total_unknown;
-};
-
-void register_perf_file_handler(struct perf_file_handler *handler);
-int mmap_dispatch_perf_file(struct perf_header **pheader,
-			    const char *input_name,
-			    int force,
-			    int full_paths,
-			    int *cwdlen,
-			    char **cwd);
-int perf_header__read_build_ids(int input, off_t offset, off_t file_size);
-
-#endif
diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c
index 414b89d1bde..bb0fd6da2d5 100644
--- a/tools/perf/util/event.c
+++ b/tools/perf/util/event.c
@@ -1,11 +1,16 @@
 #include <linux/types.h>
 #include "event.h"
 #include "debug.h"
+#include "session.h"
+#include "sort.h"
 #include "string.h"
+#include "strlist.h"
 #include "thread.h"
 
 static pid_t event__synthesize_comm(pid_t pid, int full,
-				    int (*process)(event_t *event))
+				    int (*process)(event_t *event,
+						   struct perf_session *session),
+				    struct perf_session *session)
 {
 	event_t ev;
 	char filename[PATH_MAX];
@@ -54,7 +59,7 @@ out_race:
 	if (!full) {
 		ev.comm.tid = pid;
 
-		process(&ev);
+		process(&ev, session);
 		goto out_fclose;
 	}
 
@@ -72,7 +77,7 @@ out_race:
 
 		ev.comm.tid = pid;
 
-		process(&ev);
+		process(&ev, session);
 	}
 	closedir(tasks);
 
@@ -86,7 +91,9 @@ out_failure:
 }
 
 static int event__synthesize_mmap_events(pid_t pid, pid_t tgid,
-					 int (*process)(event_t *event))
+					 int (*process)(event_t *event,
+							struct perf_session *session),
+					 struct perf_session *session)
 {
 	char filename[PATH_MAX];
 	FILE *fp;
@@ -141,7 +148,7 @@ static int event__synthesize_mmap_events(pid_t pid, pid_t tgid,
 			ev.mmap.pid = tgid;
 			ev.mmap.tid = pid;
 
-			process(&ev);
+			process(&ev, session);
 		}
 	}
 
@@ -149,15 +156,20 @@ 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))
+int event__synthesize_thread(pid_t pid,
+			     int (*process)(event_t *event,
+					    struct perf_session *session),
+			     struct perf_session *session)
 {
-	pid_t tgid = event__synthesize_comm(pid, 1, process);
+	pid_t tgid = event__synthesize_comm(pid, 1, process, session);
 	if (tgid == -1)
 		return -1;
-	return event__synthesize_mmap_events(pid, tgid, process);
+	return event__synthesize_mmap_events(pid, tgid, process, session);
 }
 
-void event__synthesize_threads(int (*process)(event_t *event))
+void event__synthesize_threads(int (*process)(event_t *event,
+					      struct perf_session *session),
+			       struct perf_session *session)
 {
 	DIR *proc;
 	struct dirent dirent, *next;
@@ -171,24 +183,47 @@ void event__synthesize_threads(int (*process)(event_t *event))
 		if (*end) /* only interested in proper numerical dirents */
 			continue;
 
-		event__synthesize_thread(pid, process);
+		event__synthesize_thread(pid, process, session);
 	}
 
 	closedir(proc);
 }
 
-char *event__cwd;
-int  event__cwdlen;
+static void thread__comm_adjust(struct thread *self)
+{
+	char *comm = self->comm;
+
+	if (!symbol_conf.col_width_list_str && !symbol_conf.field_sep &&
+	    (!symbol_conf.comm_list ||
+	     strlist__has_entry(symbol_conf.comm_list, comm))) {
+		unsigned int slen = strlen(comm);
+
+		if (slen > comms__col_width) {
+			comms__col_width = slen;
+			threads__col_width = slen + 6;
+		}
+	}
+}
+
+static int thread__set_comm_adjust(struct thread *self, const char *comm)
+{
+	int ret = thread__set_comm(self, comm);
+
+	if (ret)
+		return ret;
 
-struct events_stats event__stats;
+	thread__comm_adjust(self);
 
-int event__process_comm(event_t *self)
+	return 0;
+}
+
+int event__process_comm(event_t *self, struct perf_session *session)
 {
-	struct thread *thread = threads__findnew(self->comm.pid);
+	struct thread *thread = perf_session__findnew(session, self->comm.pid);
 
 	dump_printf(": %s:%d\n", self->comm.comm, self->comm.pid);
 
-	if (thread == NULL || thread__set_comm(thread, self->comm.comm)) {
+	if (thread == NULL || thread__set_comm_adjust(thread, self->comm.comm)) {
 		dump_printf("problem processing PERF_RECORD_COMM, skipping event.\n");
 		return -1;
 	}
@@ -196,18 +231,18 @@ int event__process_comm(event_t *self)
 	return 0;
 }
 
-int event__process_lost(event_t *self)
+int event__process_lost(event_t *self, struct perf_session *session)
 {
 	dump_printf(": id:%Ld: lost:%Ld\n", self->lost.id, self->lost.lost);
-	event__stats.lost += self->lost.lost;
+	session->events_stats.lost += self->lost.lost;
 	return 0;
 }
 
-int event__process_mmap(event_t *self)
+int event__process_mmap(event_t *self, struct perf_session *session)
 {
-	struct thread *thread = threads__findnew(self->mmap.pid);
+	struct thread *thread = perf_session__findnew(session, self->mmap.pid);
 	struct map *map = map__new(&self->mmap, MAP__FUNCTION,
-				   event__cwd, event__cwdlen);
+				   session->cwd, session->cwdlen);
 
 	dump_printf(" %d/%d: [%p(%p) @ %p]: %s\n",
 		    self->mmap.pid, self->mmap.tid,
@@ -224,10 +259,10 @@ int event__process_mmap(event_t *self)
 	return 0;
 }
 
-int event__process_task(event_t *self)
+int event__process_task(event_t *self, struct perf_session *session)
 {
-	struct thread *thread = threads__findnew(self->fork.pid);
-	struct thread *parent = threads__findnew(self->fork.ppid);
+	struct thread *thread = perf_session__findnew(session, self->fork.pid);
+	struct thread *parent = perf_session__findnew(session, self->fork.ppid);
 
 	dump_printf("(%d:%d):(%d:%d)\n", self->fork.pid, self->fork.tid,
 		    self->fork.ppid, self->fork.ptid);
@@ -249,18 +284,20 @@ int event__process_task(event_t *self)
 	return 0;
 }
 
-void thread__find_addr_location(struct thread *self, u8 cpumode,
+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 thread *thread = al->thread = self;
+	struct map_groups *mg = &self->mg;
 
+	al->thread = self;
 	al->addr = addr;
 
 	if (cpumode & PERF_RECORD_MISC_KERNEL) {
 		al->level = 'k';
-		thread = kthread;
+		mg = &session->kmaps;
 	} else if (cpumode & PERF_RECORD_MISC_USER)
 		al->level = '.';
 	else {
@@ -270,7 +307,7 @@ void thread__find_addr_location(struct thread *self, u8 cpumode,
 		return;
 	}
 try_again:
-	al->map = thread__find_map(thread, type, al->addr);
+	al->map = map_groups__find(mg, type, al->addr);
 	if (al->map == NULL) {
 		/*
 		 * If this is outside of all known maps, and is a negative
@@ -281,32 +318,139 @@ try_again:
 		 * "[vdso]" dso, but for now lets use the old trick of looking
 		 * in the whole kernel symbol list.
 		 */
-		if ((long long)al->addr < 0 && thread != kthread) {
-			thread = kthread;
+		if ((long long)al->addr < 0 && mg != &session->kmaps) {
+			mg = &session->kmaps;
 			goto try_again;
 		}
 		al->sym = NULL;
 	} else {
 		al->addr = al->map->map_ip(al->map, al->addr);
-		al->sym = map__find_symbol(al->map, al->addr, filter);
+		al->sym = map__find_symbol(al->map, session, al->addr, filter);
 	}
 }
 
-int event__preprocess_sample(const event_t *self, struct addr_location *al,
-			     symbol_filter_t filter)
+static void dso__calc_col_width(struct dso *self)
+{
+	if (!symbol_conf.col_width_list_str && !symbol_conf.field_sep &&
+	    (!symbol_conf.dso_list ||
+	     strlist__has_entry(symbol_conf.dso_list, self->name))) {
+		unsigned int slen = strlen(self->name);
+		if (slen > dsos__col_width)
+			dsos__col_width = slen;
+	}
+
+	self->slen_calculated = 1;
+}
+
+int event__preprocess_sample(const event_t *self, struct perf_session *session,
+			     struct addr_location *al, symbol_filter_t filter)
 {
 	u8 cpumode = self->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
-	struct thread *thread = threads__findnew(self->ip.pid);
+	struct thread *thread = perf_session__findnew(session, self->ip.pid);
 
 	if (thread == NULL)
 		return -1;
 
+	if (symbol_conf.comm_list &&
+	    !strlist__has_entry(symbol_conf.comm_list, thread->comm))
+		goto out_filtered;
+
 	dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid);
 
-	thread__find_addr_location(thread, cpumode, MAP__FUNCTION,
+	thread__find_addr_location(thread, session, cpumode, MAP__FUNCTION,
 				   self->ip.ip, al, filter);
 	dump_printf(" ...... dso: %s\n",
 		    al->map ? al->map->dso->long_name :
 			al->level == 'H' ? "[hypervisor]" : "<not found>");
+	/*
+	 * We have to do this here as we may have a dso with no symbol hit that
+	 * has a name longer than the ones with symbols sampled.
+	 */
+	if (al->map && !sort_dso.elide && !al->map->dso->slen_calculated)
+		dso__calc_col_width(al->map->dso);
+
+	if (symbol_conf.dso_list &&
+	    (!al->map || !al->map->dso ||
+	     !(strlist__has_entry(symbol_conf.dso_list, al->map->dso->short_name) ||
+	       (al->map->dso->short_name != al->map->dso->long_name &&
+		strlist__has_entry(symbol_conf.dso_list, al->map->dso->long_name)))))
+		goto out_filtered;
+
+	if (symbol_conf.sym_list && al->sym &&
+	    !strlist__has_entry(symbol_conf.sym_list, al->sym->name))
+		goto out_filtered;
+
+	al->filtered = false;
+	return 0;
+
+out_filtered:
+	al->filtered = true;
+	return 0;
+}
+
+int event__parse_sample(event_t *event, u64 type, struct sample_data *data)
+{
+	u64 *array = event->sample.array;
+
+	if (type & PERF_SAMPLE_IP) {
+		data->ip = event->ip.ip;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_TID) {
+		u32 *p = (u32 *)array;
+		data->pid = p[0];
+		data->tid = p[1];
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_TIME) {
+		data->time = *array;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_ADDR) {
+		data->addr = *array;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_ID) {
+		data->id = *array;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_STREAM_ID) {
+		data->stream_id = *array;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_CPU) {
+		u32 *p = (u32 *)array;
+		data->cpu = *p;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_PERIOD) {
+		data->period = *array;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_READ) {
+		pr_debug("PERF_SAMPLE_READ is unsuported for now\n");
+		return -1;
+	}
+
+	if (type & PERF_SAMPLE_CALLCHAIN) {
+		data->callchain = (struct ip_callchain *)array;
+		array += 1 + data->callchain->nr;
+	}
+
+	if (type & PERF_SAMPLE_RAW) {
+		u32 *p = (u32 *)array;
+		data->raw_size = *p;
+		p++;
+		data->raw_data = p;
+	}
+
 	return 0;
 }
diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h
index a4cc8105cf6..690a96d0467 100644
--- a/tools/perf/util/event.h
+++ b/tools/perf/util/event.h
@@ -56,11 +56,25 @@ struct read_event {
 	u64 id;
 };
 
-struct sample_event{
+struct sample_event {
 	struct perf_event_header        header;
 	u64 array[];
 };
 
+struct sample_data {
+	u64 ip;
+	u32 pid, tid;
+	u64 time;
+	u64 addr;
+	u64 id;
+	u64 stream_id;
+	u32 cpu;
+	u64 period;
+	struct ip_callchain *callchain;
+	u32 raw_size;
+	void *raw_data;
+};
+
 #define BUILD_ID_SIZE 20
 
 struct build_id_event {
@@ -81,18 +95,19 @@ typedef union event_union {
 } event_t;
 
 struct events_stats {
-	unsigned long total;
-	unsigned long lost;
+	u64 total;
+	u64 lost;
 };
 
 void event__print_totals(void);
 
 enum map_type {
 	MAP__FUNCTION = 0,
-
-	MAP__NR_TYPES,
+	MAP__VARIABLE,
 };
 
+#define MAP__NR_TYPES (MAP__VARIABLE + 1)
+
 struct map {
 	union {
 		struct rb_node	rb_node;
@@ -134,26 +149,35 @@ 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 symbol *map__find_symbol(struct map *self, u64 addr,
-				symbol_filter_t filter);
+
+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));
-void event__synthesize_threads(int (*process)(event_t *event));
-
-extern char *event__cwd;
-extern int  event__cwdlen;
-extern struct events_stats event__stats;
-extern unsigned long event__total[PERF_RECORD_MAX];
+int event__synthesize_thread(pid_t pid,
+			     int (*process)(event_t *event,
+					    struct perf_session *session),
+			     struct perf_session *session);
+void event__synthesize_threads(int (*process)(event_t *event,
+					      struct perf_session *session),
+			       struct perf_session *session);
 
-int event__process_comm(event_t *self);
-int event__process_lost(event_t *self);
-int event__process_mmap(event_t *self);
-int event__process_task(event_t *self);
+int event__process_comm(event_t *self, struct perf_session *session);
+int event__process_lost(event_t *self, struct perf_session *session);
+int event__process_mmap(event_t *self, struct perf_session *session);
+int event__process_task(event_t *self, struct perf_session *session);
 
 struct addr_location;
-int event__preprocess_sample(const event_t *self, struct addr_location *al,
-			     symbol_filter_t filter);
+int event__preprocess_sample(const event_t *self, struct perf_session *session,
+			     struct addr_location *al, symbol_filter_t filter);
+int event__parse_sample(event_t *event, u64 type, struct sample_data *data);
 
 #endif /* __PERF_RECORD_H */
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index 4805e6dfd23..8a0bca55106 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -8,8 +8,8 @@
 #include "header.h"
 #include "../perf.h"
 #include "trace-event.h"
+#include "session.h"
 #include "symbol.h"
-#include "data_map.h"
 #include "debug.h"
 
 /*
@@ -58,35 +58,19 @@ int perf_header_attr__add_id(struct perf_header_attr *self, u64 id)
 	return 0;
 }
 
-/*
- * Create new perf.data header:
- */
-struct perf_header *perf_header__new(void)
+int perf_header__init(struct perf_header *self)
 {
-	struct perf_header *self = zalloc(sizeof(*self));
-
-	if (self != NULL) {
-		self->size = 1;
-		self->attr = malloc(sizeof(void *));
-
-		if (self->attr == NULL) {
-			free(self);
-			self = NULL;
-		}
-	}
-
-	return self;
+	self->size = 1;
+	self->attr = malloc(sizeof(void *));
+	return self->attr == NULL ? -ENOMEM : 0;
 }
 
-void perf_header__delete(struct perf_header *self)
+void perf_header__exit(struct perf_header *self)
 {
 	int i;
-
 	for (i = 0; i < self->attrs; ++i)
-		perf_header_attr__delete(self->attr[i]);
-
+                perf_header_attr__delete(self->attr[i]);
 	free(self->attr);
-	free(self);
 }
 
 int perf_header__add_attr(struct perf_header *self,
@@ -187,7 +171,9 @@ static int do_write(int fd, const void *buf, size_t size)
 
 static int __dsos__write_buildid_table(struct list_head *head, int fd)
 {
+#define NAME_ALIGN	64
 	struct dso *pos;
+	static const char zero_buf[NAME_ALIGN];
 
 	list_for_each_entry(pos, head, node) {
 		int err;
@@ -197,14 +183,17 @@ static int __dsos__write_buildid_table(struct list_head *head, int fd)
 		if (!pos->has_build_id)
 			continue;
 		len = pos->long_name_len + 1;
-		len = ALIGN(len, 64);
+		len = ALIGN(len, NAME_ALIGN);
 		memset(&b, 0, sizeof(b));
 		memcpy(&b.build_id, pos->build_id, sizeof(pos->build_id));
 		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, len);
+		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);
 		if (err < 0)
 			return err;
 	}
diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h
index d1dbe2b79c4..d118d05d3ab 100644
--- a/tools/perf/util/header.h
+++ b/tools/perf/util/header.h
@@ -55,8 +55,8 @@ struct perf_header {
 	DECLARE_BITMAP(adds_features, HEADER_FEAT_BITS);
 };
 
-struct perf_header *perf_header__new(void);
-void perf_header__delete(struct perf_header *self);
+int perf_header__init(struct perf_header *self);
+void perf_header__exit(struct perf_header *self);
 
 int perf_header__read(struct perf_header *self, int fd);
 int perf_header__write(struct perf_header *self, int fd, bool at_exit);
diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
index 0ebf6ee16ca..e8daf5ca6fd 100644
--- a/tools/perf/util/hist.c
+++ b/tools/perf/util/hist.c
@@ -1,9 +1,7 @@
 #include "hist.h"
-
-struct rb_root hist;
-struct rb_root collapse_hists;
-struct rb_root output_hists;
-int callchain;
+#include "session.h"
+#include "sort.h"
+#include <math.h>
 
 struct callchain_param	callchain_param = {
 	.mode	= CHAIN_GRAPH_REL,
@@ -14,11 +12,12 @@ struct callchain_param	callchain_param = {
  * histogram, sorted on item, collects counts
  */
 
-struct hist_entry *__hist_entry__add(struct addr_location *al,
-				     struct symbol *sym_parent,
-				     u64 count, bool *hit)
+struct hist_entry *__perf_session__add_hist_entry(struct perf_session *self,
+						  struct addr_location *al,
+						  struct symbol *sym_parent,
+						  u64 count, bool *hit)
 {
-	struct rb_node **p = &hist.rb_node;
+	struct rb_node **p = &self->hists.rb_node;
 	struct rb_node *parent = NULL;
 	struct hist_entry *he;
 	struct hist_entry entry = {
@@ -54,7 +53,7 @@ struct hist_entry *__hist_entry__add(struct addr_location *al,
 		return NULL;
 	*he = entry;
 	rb_link_node(&he->rb_node, parent, p);
-	rb_insert_color(&he->rb_node, &hist);
+	rb_insert_color(&he->rb_node, &self->hists);
 	*hit = false;
 	return he;
 }
@@ -102,9 +101,9 @@ void hist_entry__free(struct hist_entry *he)
  * collapse the histogram
  */
 
-void collapse__insert_entry(struct hist_entry *he)
+static void collapse__insert_entry(struct rb_root *root, struct hist_entry *he)
 {
-	struct rb_node **p = &collapse_hists.rb_node;
+	struct rb_node **p = &root->rb_node;
 	struct rb_node *parent = NULL;
 	struct hist_entry *iter;
 	int64_t cmp;
@@ -128,38 +127,45 @@ void collapse__insert_entry(struct hist_entry *he)
 	}
 
 	rb_link_node(&he->rb_node, parent, p);
-	rb_insert_color(&he->rb_node, &collapse_hists);
+	rb_insert_color(&he->rb_node, root);
 }
 
-void collapse__resort(void)
+void perf_session__collapse_resort(struct perf_session *self)
 {
+	struct rb_root tmp;
 	struct rb_node *next;
 	struct hist_entry *n;
 
 	if (!sort__need_collapse)
 		return;
 
-	next = rb_first(&hist);
+	tmp = RB_ROOT;
+	next = rb_first(&self->hists);
+
 	while (next) {
 		n = rb_entry(next, struct hist_entry, rb_node);
 		next = rb_next(&n->rb_node);
 
-		rb_erase(&n->rb_node, &hist);
-		collapse__insert_entry(n);
+		rb_erase(&n->rb_node, &self->hists);
+		collapse__insert_entry(&tmp, n);
 	}
+
+	self->hists = tmp;
 }
 
 /*
  * reverse the map, sort on count.
  */
 
-void output__insert_entry(struct hist_entry *he, u64 min_callchain_hits)
+static void perf_session__insert_output_hist_entry(struct rb_root *root,
+						   struct hist_entry *he,
+						   u64 min_callchain_hits)
 {
-	struct rb_node **p = &output_hists.rb_node;
+	struct rb_node **p = &root->rb_node;
 	struct rb_node *parent = NULL;
 	struct hist_entry *iter;
 
-	if (callchain)
+	if (symbol_conf.use_callchain)
 		callchain_param.sort(&he->sorted_chain, &he->callchain,
 				      min_callchain_hits, &callchain_param);
 
@@ -174,29 +180,483 @@ void output__insert_entry(struct hist_entry *he, u64 min_callchain_hits)
 	}
 
 	rb_link_node(&he->rb_node, parent, p);
-	rb_insert_color(&he->rb_node, &output_hists);
+	rb_insert_color(&he->rb_node, root);
 }
 
-void output__resort(u64 total_samples)
+void perf_session__output_resort(struct perf_session *self, u64 total_samples)
 {
+	struct rb_root tmp;
 	struct rb_node *next;
 	struct hist_entry *n;
-	struct rb_root *tree = &hist;
 	u64 min_callchain_hits;
 
 	min_callchain_hits =
 		total_samples * (callchain_param.min_percent / 100);
 
-	if (sort__need_collapse)
-		tree = &collapse_hists;
-
-	next = rb_first(tree);
+	tmp = RB_ROOT;
+	next = rb_first(&self->hists);
 
 	while (next) {
 		n = rb_entry(next, struct hist_entry, rb_node);
 		next = rb_next(&n->rb_node);
 
-		rb_erase(&n->rb_node, tree);
-		output__insert_entry(n, min_callchain_hits);
+		rb_erase(&n->rb_node, &self->hists);
+		perf_session__insert_output_hist_entry(&tmp, n,
+						       min_callchain_hits);
+	}
+
+	self->hists = tmp;
+}
+
+static size_t callchain__fprintf_left_margin(FILE *fp, int left_margin)
+{
+	int i;
+	int ret = fprintf(fp, "            ");
+
+	for (i = 0; i < left_margin; i++)
+		ret += fprintf(fp, " ");
+
+	return ret;
+}
+
+static size_t ipchain__fprintf_graph_line(FILE *fp, int depth, int depth_mask,
+					  int left_margin)
+{
+	int i;
+	size_t ret = callchain__fprintf_left_margin(fp, left_margin);
+
+	for (i = 0; i < depth; i++)
+		if (depth_mask & (1 << i))
+			ret += fprintf(fp, "|          ");
+		else
+			ret += fprintf(fp, "           ");
+
+	ret += fprintf(fp, "\n");
+
+	return ret;
+}
+
+static size_t ipchain__fprintf_graph(FILE *fp, struct callchain_list *chain,
+				     int depth, int depth_mask, int count,
+				     u64 total_samples, int hits,
+				     int left_margin)
+{
+	int i;
+	size_t ret = 0;
+
+	ret += callchain__fprintf_left_margin(fp, left_margin);
+	for (i = 0; i < depth; i++) {
+		if (depth_mask & (1 << i))
+			ret += fprintf(fp, "|");
+		else
+			ret += fprintf(fp, " ");
+		if (!count && i == depth - 1) {
+			double percent;
+
+			percent = hits * 100.0 / total_samples;
+			ret += percent_color_fprintf(fp, "--%2.2f%%-- ", percent);
+		} else
+			ret += fprintf(fp, "%s", "          ");
+	}
+	if (chain->sym)
+		ret += fprintf(fp, "%s\n", chain->sym->name);
+	else
+		ret += fprintf(fp, "%p\n", (void *)(long)chain->ip);
+
+	return ret;
+}
+
+static struct symbol *rem_sq_bracket;
+static struct callchain_list rem_hits;
+
+static void init_rem_hits(void)
+{
+	rem_sq_bracket = malloc(sizeof(*rem_sq_bracket) + 6);
+	if (!rem_sq_bracket) {
+		fprintf(stderr, "Not enough memory to display remaining hits\n");
+		return;
+	}
+
+	strcpy(rem_sq_bracket->name, "[...]");
+	rem_hits.sym = rem_sq_bracket;
+}
+
+static size_t __callchain__fprintf_graph(FILE *fp, struct callchain_node *self,
+					 u64 total_samples, int depth,
+					 int depth_mask, int left_margin)
+{
+	struct rb_node *node, *next;
+	struct callchain_node *child;
+	struct callchain_list *chain;
+	int new_depth_mask = depth_mask;
+	u64 new_total;
+	u64 remaining;
+	size_t ret = 0;
+	int i;
+
+	if (callchain_param.mode == CHAIN_GRAPH_REL)
+		new_total = self->children_hit;
+	else
+		new_total = total_samples;
+
+	remaining = new_total;
+
+	node = rb_first(&self->rb_root);
+	while (node) {
+		u64 cumul;
+
+		child = rb_entry(node, struct callchain_node, rb_node);
+		cumul = cumul_hits(child);
+		remaining -= cumul;
+
+		/*
+		 * The depth mask manages the output of pipes that show
+		 * the depth. We don't want to keep the pipes of the current
+		 * level for the last child of this depth.
+		 * Except if we have remaining filtered hits. They will
+		 * supersede the last child
+		 */
+		next = rb_next(node);
+		if (!next && (callchain_param.mode != CHAIN_GRAPH_REL || !remaining))
+			new_depth_mask &= ~(1 << (depth - 1));
+
+		/*
+		 * But we keep the older depth mask for the line seperator
+		 * to keep the level link until we reach the last child
+		 */
+		ret += ipchain__fprintf_graph_line(fp, depth, depth_mask,
+						   left_margin);
+		i = 0;
+		list_for_each_entry(chain, &child->val, list) {
+			if (chain->ip >= PERF_CONTEXT_MAX)
+				continue;
+			ret += ipchain__fprintf_graph(fp, chain, depth,
+						      new_depth_mask, i++,
+						      new_total,
+						      cumul,
+						      left_margin);
+		}
+		ret += __callchain__fprintf_graph(fp, child, new_total,
+						  depth + 1,
+						  new_depth_mask | (1 << depth),
+						  left_margin);
+		node = next;
+	}
+
+	if (callchain_param.mode == CHAIN_GRAPH_REL &&
+		remaining && remaining != new_total) {
+
+		if (!rem_sq_bracket)
+			return ret;
+
+		new_depth_mask &= ~(1 << (depth - 1));
+
+		ret += ipchain__fprintf_graph(fp, &rem_hits, depth,
+					      new_depth_mask, 0, new_total,
+					      remaining, left_margin);
+	}
+
+	return ret;
+}
+
+static size_t callchain__fprintf_graph(FILE *fp, struct callchain_node *self,
+				       u64 total_samples, int left_margin)
+{
+	struct callchain_list *chain;
+	bool printed = false;
+	int i = 0;
+	int ret = 0;
+
+	list_for_each_entry(chain, &self->val, list) {
+		if (chain->ip >= PERF_CONTEXT_MAX)
+			continue;
+
+		if (!i++ && sort__first_dimension == SORT_SYM)
+			continue;
+
+		if (!printed) {
+			ret += callchain__fprintf_left_margin(fp, left_margin);
+			ret += fprintf(fp, "|\n");
+			ret += callchain__fprintf_left_margin(fp, left_margin);
+			ret += fprintf(fp, "---");
+
+			left_margin += 3;
+			printed = true;
+		} else
+			ret += callchain__fprintf_left_margin(fp, left_margin);
+
+		if (chain->sym)
+			ret += fprintf(fp, " %s\n", chain->sym->name);
+		else
+			ret += fprintf(fp, " %p\n", (void *)(long)chain->ip);
+	}
+
+	ret += __callchain__fprintf_graph(fp, self, total_samples, 1, 1, left_margin);
+
+	return ret;
+}
+
+static size_t callchain__fprintf_flat(FILE *fp, struct callchain_node *self,
+				      u64 total_samples)
+{
+	struct callchain_list *chain;
+	size_t ret = 0;
+
+	if (!self)
+		return 0;
+
+	ret += callchain__fprintf_flat(fp, self->parent, total_samples);
+
+
+	list_for_each_entry(chain, &self->val, list) {
+		if (chain->ip >= PERF_CONTEXT_MAX)
+			continue;
+		if (chain->sym)
+			ret += fprintf(fp, "                %s\n", chain->sym->name);
+		else
+			ret += fprintf(fp, "                %p\n",
+					(void *)(long)chain->ip);
+	}
+
+	return ret;
+}
+
+static size_t hist_entry_callchain__fprintf(FILE *fp, struct hist_entry *self,
+					    u64 total_samples, int left_margin)
+{
+	struct rb_node *rb_node;
+	struct callchain_node *chain;
+	size_t ret = 0;
+
+	rb_node = rb_first(&self->sorted_chain);
+	while (rb_node) {
+		double percent;
+
+		chain = rb_entry(rb_node, struct callchain_node, rb_node);
+		percent = chain->hit * 100.0 / total_samples;
+		switch (callchain_param.mode) {
+		case CHAIN_FLAT:
+			ret += percent_color_fprintf(fp, "           %6.2f%%\n",
+						     percent);
+			ret += callchain__fprintf_flat(fp, chain, total_samples);
+			break;
+		case CHAIN_GRAPH_ABS: /* Falldown */
+		case CHAIN_GRAPH_REL:
+			ret += callchain__fprintf_graph(fp, chain, total_samples,
+							left_margin);
+		case CHAIN_NONE:
+		default:
+			break;
+		}
+		ret += fprintf(fp, "\n");
+		rb_node = rb_next(rb_node);
+	}
+
+	return ret;
+}
+
+static size_t hist_entry__fprintf(struct hist_entry *self,
+				  struct perf_session *session,
+				  struct perf_session *pair_session,
+				  bool show_displacement,
+				  long displacement, FILE *fp)
+{
+	struct sort_entry *se;
+	u64 count, total;
+	const char *sep = symbol_conf.field_sep;
+	size_t ret;
+
+	if (symbol_conf.exclude_other && !self->parent)
+		return 0;
+
+	if (pair_session) {
+		count = self->pair ? self->pair->count : 0;
+		total = pair_session->events_stats.total;
+	} else {
+		count = self->count;
+		total = session->events_stats.total;
+	}
+
+	if (total)
+		ret = percent_color_fprintf(fp, sep ? "%.2f" : "   %6.2f%%",
+					    (count * 100.0) / total);
+	else
+		ret = fprintf(fp, sep ? "%lld" : "%12lld ", count);
+
+	if (symbol_conf.show_nr_samples) {
+		if (sep)
+			fprintf(fp, "%c%lld", *sep, count);
+		else
+			fprintf(fp, "%11lld", count);
+	}
+
+	if (pair_session) {
+		char bf[32];
+		double old_percent = 0, new_percent = 0, diff;
+
+		if (total > 0)
+			old_percent = (count * 100.0) / total;
+		if (session->events_stats.total > 0)
+			new_percent = (self->count * 100.0) / session->events_stats.total;
+
+		diff = new_percent - old_percent;
+
+		if (fabs(diff) >= 0.01)
+			snprintf(bf, sizeof(bf), "%+4.2F%%", diff);
+		else
+			snprintf(bf, sizeof(bf), " ");
+
+		if (sep)
+			ret += fprintf(fp, "%c%s", *sep, bf);
+		else
+			ret += fprintf(fp, "%11.11s", bf);
+
+		if (show_displacement) {
+			if (displacement)
+				snprintf(bf, sizeof(bf), "%+4ld", displacement);
+			else
+				snprintf(bf, sizeof(bf), " ");
+
+			if (sep)
+				fprintf(fp, "%c%s", *sep, bf);
+			else
+				fprintf(fp, "%6.6s", bf);
+		}
+	}
+
+	list_for_each_entry(se, &hist_entry__sort_list, list) {
+		if (se->elide)
+			continue;
+
+		fprintf(fp, "%s", sep ?: "  ");
+		ret += se->print(fp, self, se->width ? *se->width : 0);
+	}
+
+	ret += fprintf(fp, "\n");
+
+	if (symbol_conf.use_callchain) {
+		int left_margin = 0;
+
+		if (sort__first_dimension == SORT_COMM) {
+			se = list_first_entry(&hist_entry__sort_list, typeof(*se),
+						list);
+			left_margin = se->width ? *se->width : 0;
+			left_margin -= thread__comm_len(self->thread);
+		}
+
+		hist_entry_callchain__fprintf(fp, self, session->events_stats.total,
+					      left_margin);
+	}
+
+	return ret;
+}
+
+size_t perf_session__fprintf_hists(struct perf_session *self,
+				   struct perf_session *pair,
+				   bool show_displacement, FILE *fp)
+{
+	struct sort_entry *se;
+	struct rb_node *nd;
+	size_t ret = 0;
+	unsigned long position = 1;
+	long displacement = 0;
+	unsigned int width;
+	const char *sep = symbol_conf.field_sep;
+	char *col_width = symbol_conf.col_width_list_str;
+
+	init_rem_hits();
+
+	fprintf(fp, "# %s", pair ? "Baseline" : "Overhead");
+
+	if (symbol_conf.show_nr_samples) {
+		if (sep)
+			fprintf(fp, "%cSamples", *sep);
+		else
+			fputs("  Samples  ", fp);
+	}
+
+	if (pair) {
+		if (sep)
+			ret += fprintf(fp, "%cDelta", *sep);
+		else
+			ret += fprintf(fp, "  Delta    ");
+
+		if (show_displacement) {
+			if (sep)
+				ret += fprintf(fp, "%cDisplacement", *sep);
+			else
+				ret += fprintf(fp, " Displ");
+		}
+	}
+
+	list_for_each_entry(se, &hist_entry__sort_list, list) {
+		if (se->elide)
+			continue;
+		if (sep) {
+			fprintf(fp, "%c%s", *sep, se->header);
+			continue;
+		}
+		width = strlen(se->header);
+		if (se->width) {
+			if (symbol_conf.col_width_list_str) {
+				if (col_width) {
+					*se->width = atoi(col_width);
+					col_width = strchr(col_width, ',');
+					if (col_width)
+						++col_width;
+				}
+			}
+			width = *se->width = max(*se->width, width);
+		}
+		fprintf(fp, "  %*s", width, se->header);
+	}
+	fprintf(fp, "\n");
+
+	if (sep)
+		goto print_entries;
+
+	fprintf(fp, "# ........");
+	if (symbol_conf.show_nr_samples)
+		fprintf(fp, " ..........");
+	if (pair) {
+		fprintf(fp, " ..........");
+		if (show_displacement)
+			fprintf(fp, " .....");
+	}
+	list_for_each_entry(se, &hist_entry__sort_list, list) {
+		unsigned int i;
+
+		if (se->elide)
+			continue;
+
+		fprintf(fp, "  ");
+		if (se->width)
+			width = *se->width;
+		else
+			width = strlen(se->header);
+		for (i = 0; i < width; i++)
+			fprintf(fp, ".");
+	}
+
+	fprintf(fp, "\n#\n");
+
+print_entries:
+	for (nd = rb_first(&self->hists); nd; nd = rb_next(nd)) {
+		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+
+		if (show_displacement) {
+			if (h->pair != NULL)
+				displacement = ((long)h->pair->position -
+					        (long)position);
+			else
+				displacement = 0;
+			++position;
+		}
+		ret += hist_entry__fprintf(h, self, pair, show_displacement,
+					   displacement, fp);
 	}
+
+	free(rem_sq_bracket);
+
+	return ret;
 }
diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h
index 3020db0c929..e5f99b24048 100644
--- a/tools/perf/util/hist.h
+++ b/tools/perf/util/hist.h
@@ -1,50 +1,27 @@
 #ifndef __PERF_HIST_H
 #define __PERF_HIST_H
-#include "../builtin.h"
 
-#include "util.h"
-
-#include "color.h"
-#include <linux/list.h>
-#include "cache.h"
-#include <linux/rbtree.h>
-#include "symbol.h"
-#include "string.h"
+#include <linux/types.h>
 #include "callchain.h"
-#include "strlist.h"
-#include "values.h"
-
-#include "../perf.h"
-#include "debug.h"
-#include "header.h"
-
-#include "parse-options.h"
-#include "parse-events.h"
 
-#include "thread.h"
-#include "sort.h"
-
-extern struct rb_root hist;
-extern struct rb_root collapse_hists;
-extern struct rb_root output_hists;
-extern int callchain;
 extern struct callchain_param callchain_param;
-extern unsigned long total;
-extern unsigned long total_mmap;
-extern unsigned long total_comm;
-extern unsigned long total_fork;
-extern unsigned long total_unknown;
-extern unsigned long total_lost;
 
-struct hist_entry *__hist_entry__add(struct addr_location *al,
-				     struct symbol *parent,
-				     u64 count, bool *hit);
+struct perf_session;
+struct hist_entry;
+struct addr_location;
+struct symbol;
+
+struct hist_entry *__perf_session__add_hist_entry(struct perf_session *self,
+						  struct addr_location *al,
+						  struct symbol *parent,
+						  u64 count, bool *hit);
 extern int64_t hist_entry__cmp(struct hist_entry *, struct hist_entry *);
 extern int64_t hist_entry__collapse(struct hist_entry *, struct hist_entry *);
-extern void hist_entry__free(struct hist_entry *);
-extern void collapse__insert_entry(struct hist_entry *);
-extern void collapse__resort(void);
-extern void output__insert_entry(struct hist_entry *, u64);
-extern void output__resort(u64);
+void hist_entry__free(struct hist_entry *);
 
+void perf_session__output_resort(struct perf_session *self, u64 total_samples);
+void perf_session__collapse_resort(struct perf_session *self);
+size_t perf_session__fprintf_hists(struct perf_session *self,
+				   struct perf_session *pair,
+				   bool show_displacement, FILE *fp);
 #endif	/* __PERF_HIST_H */
diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c
index 69f94fe9db2..c4d55a0da2e 100644
--- a/tools/perf/util/map.c
+++ b/tools/perf/util/map.c
@@ -104,43 +104,70 @@ void map__fixup_end(struct map *self)
 
 #define DSO__DELETED "(deleted)"
 
-struct symbol *map__find_symbol(struct map *self, u64 addr,
-				symbol_filter_t filter)
+int map__load(struct map *self, struct perf_session *session,
+	      symbol_filter_t filter)
 {
-	if (!dso__loaded(self->dso, self->type)) {
-		int nr = dso__load(self->dso, self, filter);
-
-		if (nr < 0) {
-			if (self->dso->has_build_id) {
-				char sbuild_id[BUILD_ID_SIZE * 2 + 1];
-
-				build_id__sprintf(self->dso->build_id,
-						  sizeof(self->dso->build_id),
-						  sbuild_id);
-				pr_warning("%s with build id %s not found",
-					   self->dso->long_name, sbuild_id);
-			} else
-				pr_warning("Failed to open %s",
-					   self->dso->long_name);
-			pr_warning(", continuing without symbols\n");
-			return NULL;
-		} else if (nr == 0) {
-			const char *name = self->dso->long_name;
-			const size_t len = strlen(name);
-			const size_t real_len = len - sizeof(DSO__DELETED);
-
-			if (len > sizeof(DSO__DELETED) &&
-			    strcmp(name + real_len + 1, DSO__DELETED) == 0) {
-				pr_warning("%.*s was updated, restart the long running apps that use it!\n",
-					   (int)real_len, name);
-			} else {
-				pr_warning("no symbols found in %s, maybe install a debug package?\n", name);
-			}
-			return NULL;
+	const char *name = self->dso->long_name;
+	int nr;
+
+	if (dso__loaded(self->dso, self->type))
+		return 0;
+
+	nr = dso__load(self->dso, self, session, filter);
+	if (nr < 0) {
+		if (self->dso->has_build_id) {
+			char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+			build_id__sprintf(self->dso->build_id,
+					  sizeof(self->dso->build_id),
+					  sbuild_id);
+			pr_warning("%s with build id %s not found",
+				   name, sbuild_id);
+		} else
+			pr_warning("Failed to open %s", name);
+
+		pr_warning(", continuing without symbols\n");
+		return -1;
+	} else if (nr == 0) {
+		const size_t len = strlen(name);
+		const size_t real_len = len - sizeof(DSO__DELETED);
+
+		if (len > sizeof(DSO__DELETED) &&
+		    strcmp(name + real_len + 1, DSO__DELETED) == 0) {
+			pr_warning("%.*s was updated, restart the long "
+				   "running apps that use it!\n",
+				   (int)real_len, name);
+		} else {
+			pr_warning("no symbols found in %s, maybe install "
+				   "a debug package?\n", name);
 		}
+
+		return -1;
 	}
 
-	return self->dso->find_symbol(self->dso, self->type, addr);
+	return 0;
+}
+
+struct symbol *map__find_symbol(struct map *self, struct perf_session *session,
+				u64 addr, symbol_filter_t filter)
+{
+	if (map__load(self, session, 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)
+		return NULL;
+
+	if (!dso__sorted_by_name(self->dso, self->type))
+		dso__sort_by_name(self->dso, self->type);
+
+	return dso__find_symbol_by_name(self->dso, self->type, name);
 }
 
 struct map *map__clone(struct map *self)
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 9e5dbd66d34..e5bc0fb016b 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -197,7 +197,7 @@ struct tracepoint_path *tracepoint_id_to_path(u64 config)
 			if (id == config) {
 				closedir(evt_dir);
 				closedir(sys_dir);
-				path = zalloc(sizeof(path));
+				path = zalloc(sizeof(*path));
 				path->system = malloc(MAX_EVENT_LENGTH);
 				if (!path->system) {
 					free(path);
@@ -467,7 +467,6 @@ parse_subsystem_tracepoint_event(char *sys_name, char *flags)
 	while ((evt_ent = readdir(evt_dir))) {
 		char event_opt[MAX_EVOPT_LEN + 1];
 		int len;
-		unsigned int rem = MAX_EVOPT_LEN;
 
 		if (!strcmp(evt_ent->d_name, ".")
 		    || !strcmp(evt_ent->d_name, "..")
@@ -475,20 +474,12 @@ parse_subsystem_tracepoint_event(char *sys_name, char *flags)
 		    || !strcmp(evt_ent->d_name, "filter"))
 			continue;
 
-		len = snprintf(event_opt, MAX_EVOPT_LEN, "%s:%s", sys_name,
-			       evt_ent->d_name);
+		len = snprintf(event_opt, MAX_EVOPT_LEN, "%s:%s%s%s", sys_name,
+			       evt_ent->d_name, flags ? ":" : "",
+			       flags ?: "");
 		if (len < 0)
 			return EVT_FAILED;
 
-		rem -= len;
-		if (flags) {
-			if (rem < strlen(flags) + 1)
-				return EVT_FAILED;
-
-			strcat(event_opt, ":");
-			strcat(event_opt, flags);
-		}
-
 		if (parse_events(NULL, event_opt, 0))
 			return EVT_FAILED;
 	}
diff --git a/tools/perf/util/parse-options.c b/tools/perf/util/parse-options.c
index 6d8af48c925..efebd5b476b 100644
--- a/tools/perf/util/parse-options.c
+++ b/tools/perf/util/parse-options.c
@@ -430,6 +430,9 @@ int usage_with_options_internal(const char * const *usagestr,
 		pos = fprintf(stderr, "    ");
 		if (opts->short_name)
 			pos += fprintf(stderr, "-%c", opts->short_name);
+		else
+			pos += fprintf(stderr, "    ");
+
 		if (opts->long_name && opts->short_name)
 			pos += fprintf(stderr, ", ");
 		if (opts->long_name)
diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index cd7fbda5e2a..29465d44004 100644
--- a/tools/perf/util/probe-event.c
+++ b/tools/perf/util/probe-event.c
@@ -48,6 +48,9 @@
 
 /* If there is no space to write, returns -E2BIG. */
 static int e_snprintf(char *str, size_t size, const char *format, ...)
+	__attribute__((format(printf, 3, 4)));
+
+static int e_snprintf(char *str, size_t size, const char *format, ...)
 {
 	int ret;
 	va_list ap;
@@ -59,6 +62,18 @@ static int e_snprintf(char *str, size_t size, const char *format, ...)
 	return ret;
 }
 
+/* Check the name is good for event/group */
+static bool check_event_name(const char *name)
+{
+	if (!isalpha(*name) && *name != '_')
+		return false;
+	while (*++name != '\0') {
+		if (!isalpha(*name) && !isdigit(*name) && *name != '_')
+			return false;
+	}
+	return true;
+}
+
 /* Parse probepoint definition. */
 static void parse_perf_probe_probepoint(char *arg, struct probe_point *pp)
 {
@@ -66,10 +81,26 @@ static void parse_perf_probe_probepoint(char *arg, struct probe_point *pp)
 	char c, nc = 0;
 	/*
 	 * <Syntax>
-	 * perf probe SRC:LN
-	 * perf probe FUNC[+OFFS|%return][@SRC]
+	 * perf probe [EVENT=]SRC:LN
+	 * perf probe [EVENT=]FUNC[+OFFS|%return][@SRC]
+	 *
+	 * TODO:Group name support
 	 */
 
+	ptr = strchr(arg, '=');
+	if (ptr) {	/* Event name */
+		*ptr = '\0';
+		tmp = ptr + 1;
+		ptr = strchr(arg, ':');
+		if (ptr)	/* Group name is not supported yet. */
+			semantic_error("Group name is not supported yet.");
+		if (!check_event_name(arg))
+			semantic_error("%s is bad for event name -it must "
+				       "follow C symbol-naming rule.", arg);
+		pp->event = strdup(arg);
+		arg = tmp;
+	}
+
 	ptr = strpbrk(arg, ":+@%");
 	if (ptr) {
 		nc = *ptr;
@@ -147,10 +178,13 @@ static void parse_perf_probe_probepoint(char *arg, struct probe_point *pp)
 }
 
 /* Parse perf-probe event definition */
-int parse_perf_probe_event(const char *str, struct probe_point *pp)
+void parse_perf_probe_event(const char *str, struct probe_point *pp,
+			    bool *need_dwarf)
 {
 	char **argv;
-	int argc, i, need_dwarf = 0;
+	int argc, i;
+
+	*need_dwarf = false;
 
 	argv = argv_split(str, &argc);
 	if (!argv)
@@ -161,7 +195,7 @@ int parse_perf_probe_event(const char *str, struct probe_point *pp)
 	/* Parse probe point */
 	parse_perf_probe_probepoint(argv[0], pp);
 	if (pp->file || pp->line)
-		need_dwarf = 1;
+		*need_dwarf = true;
 
 	/* Copy arguments and ensure return probe has no C argument */
 	pp->nr_args = argc - 1;
@@ -174,17 +208,15 @@ int parse_perf_probe_event(const char *str, struct probe_point *pp)
 			if (pp->retprobe)
 				semantic_error("You can't specify local"
 						" variable for kretprobe");
-			need_dwarf = 1;
+			*need_dwarf = true;
 		}
 	}
 
 	argv_free(argv);
-	return need_dwarf;
 }
 
 /* Parse kprobe_events event into struct probe_point */
-void parse_trace_kprobe_event(const char *str, char **group, char **event,
-			      struct probe_point *pp)
+void parse_trace_kprobe_event(const char *str, struct probe_point *pp)
 {
 	char pr;
 	char *p;
@@ -200,18 +232,17 @@ void parse_trace_kprobe_event(const char *str, char **group, char **event,
 
 	/* Scan event and group name. */
 	ret = sscanf(argv[0], "%c:%a[^/ \t]/%a[^ \t]",
-		     &pr, (float *)(void *)group, (float *)(void *)event);
+		     &pr, (float *)(void *)&pp->group,
+		     (float *)(void *)&pp->event);
 	if (ret != 3)
 		semantic_error("Failed to parse event name: %s", argv[0]);
-	pr_debug("Group:%s Event:%s probe:%c\n", *group, *event, pr);
-
-	if (!pp)
-		goto end;
+	pr_debug("Group:%s Event:%s probe:%c\n", pp->group, pp->event, pr);
 
 	pp->retprobe = (pr == 'r');
 
 	/* Scan function name and offset */
-	ret = sscanf(argv[1], "%a[^+]+%d", (float *)(void *)&pp->function, &pp->offset);
+	ret = sscanf(argv[1], "%a[^+]+%d", (float *)(void *)&pp->function,
+		     &pp->offset);
 	if (ret == 1)
 		pp->offset = 0;
 
@@ -230,15 +261,15 @@ void parse_trace_kprobe_event(const char *str, char **group, char **event,
 			die("Failed to copy argument.");
 	}
 
-end:
 	argv_free(argv);
 }
 
-int synthesize_perf_probe_event(struct probe_point *pp)
+/* Synthesize only probe point (not argument) */
+int synthesize_perf_probe_point(struct probe_point *pp)
 {
 	char *buf;
 	char offs[64] = "", line[64] = "";
-	int i, len, ret;
+	int ret;
 
 	pp->probes[0] = buf = zalloc(MAX_CMDLEN);
 	if (!buf)
@@ -258,11 +289,25 @@ int synthesize_perf_probe_event(struct probe_point *pp)
 		ret = e_snprintf(buf, MAX_CMDLEN, "%s%s%s%s", pp->function,
 				 offs, pp->retprobe ? "%return" : "", line);
 	else
-		ret = e_snprintf(buf, MAX_CMDLEN, "%s%s%s%s", pp->file, line);
-	if (ret <= 0)
-		goto error;
-	len = ret;
+		ret = e_snprintf(buf, MAX_CMDLEN, "%s%s", pp->file, line);
+	if (ret <= 0) {
+error:
+		free(pp->probes[0]);
+		pp->probes[0] = NULL;
+	}
+	return ret;
+}
+
+int synthesize_perf_probe_event(struct probe_point *pp)
+{
+	char *buf;
+	int i, len, ret;
+
+	len = synthesize_perf_probe_point(pp);
+	if (len < 0)
+		return 0;
 
+	buf = pp->probes[0];
 	for (i = 0; i < pp->nr_args; i++) {
 		ret = e_snprintf(&buf[len], MAX_CMDLEN - len, " %s",
 				 pp->args[i]);
@@ -275,6 +320,7 @@ int synthesize_perf_probe_event(struct probe_point *pp)
 	return pp->found;
 error:
 	free(pp->probes[0]);
+	pp->probes[0] = NULL;
 
 	return ret;
 }
@@ -304,6 +350,7 @@ int synthesize_trace_kprobe_event(struct probe_point *pp)
 	return pp->found;
 error:
 	free(pp->probes[0]);
+	pp->probes[0] = NULL;
 
 	return ret;
 }
@@ -363,6 +410,10 @@ static void clear_probe_point(struct probe_point *pp)
 {
 	int i;
 
+	if (pp->event)
+		free(pp->event);
+	if (pp->group)
+		free(pp->group);
 	if (pp->function)
 		free(pp->function);
 	if (pp->file)
@@ -373,15 +424,33 @@ static void clear_probe_point(struct probe_point *pp)
 		free(pp->args);
 	for (i = 0; i < pp->found; i++)
 		free(pp->probes[i]);
-	memset(pp, 0, sizeof(pp));
+	memset(pp, 0, sizeof(*pp));
+}
+
+/* Show an event */
+static void show_perf_probe_event(const char *event, const char *place,
+				  struct probe_point *pp)
+{
+	int i, ret;
+	char buf[128];
+
+	ret = e_snprintf(buf, 128, "%s:%s", pp->group, event);
+	if (ret < 0)
+		die("Failed to copy event: %s", strerror(-ret));
+	printf("  %-40s (on %s", buf, place);
+
+	if (pp->nr_args > 0) {
+		printf(" with");
+		for (i = 0; i < pp->nr_args; i++)
+			printf(" %s", pp->args[i]);
+	}
+	printf(")\n");
 }
 
 /* List up current perf-probe events */
 void show_perf_probe_events(void)
 {
-	unsigned int i;
 	int fd;
-	char *group, *event;
 	struct probe_point pp;
 	struct strlist *rawlist;
 	struct str_node *ent;
@@ -390,13 +459,12 @@ void show_perf_probe_events(void)
 	rawlist = get_trace_kprobe_event_rawlist(fd);
 	close(fd);
 
-	for (i = 0; i < strlist__nr_entries(rawlist); i++) {
-		ent = strlist__entry(rawlist, i);
-		parse_trace_kprobe_event(ent->s, &group, &event, &pp);
-		synthesize_perf_probe_event(&pp);
-		printf("[%s:%s]\t%s\n", group, event, pp.probes[0]);
-		free(group);
-		free(event);
+	strlist__for_each(ent, rawlist) {
+		parse_trace_kprobe_event(ent->s, &pp);
+		/* Synthesize only event probe point */
+		synthesize_perf_probe_point(&pp);
+		/* Show an event */
+		show_perf_probe_event(pp.event, pp.probes[0], &pp);
 		clear_probe_point(&pp);
 	}
 
@@ -404,21 +472,27 @@ void show_perf_probe_events(void)
 }
 
 /* Get current perf-probe event names */
-static struct strlist *get_perf_event_names(int fd)
+static struct strlist *get_perf_event_names(int fd, bool include_group)
 {
-	unsigned int i;
-	char *group, *event;
+	char buf[128];
 	struct strlist *sl, *rawlist;
 	struct str_node *ent;
+	struct probe_point pp;
 
+	memset(&pp, 0, sizeof(pp));
 	rawlist = get_trace_kprobe_event_rawlist(fd);
 
-	sl = strlist__new(false, NULL);
-	for (i = 0; i < strlist__nr_entries(rawlist); i++) {
-		ent = strlist__entry(rawlist, i);
-		parse_trace_kprobe_event(ent->s, &group, &event, NULL);
-		strlist__add(sl, event);
-		free(group);
+	sl = strlist__new(true, NULL);
+	strlist__for_each(ent, rawlist) {
+		parse_trace_kprobe_event(ent->s, &pp);
+		if (include_group) {
+			if (e_snprintf(buf, 128, "%s:%s", pp.group,
+				       pp.event) < 0)
+				die("Failed to copy group:event name.");
+			strlist__add(sl, buf);
+		} else
+			strlist__add(sl, pp.event);
+		clear_probe_point(&pp);
 	}
 
 	strlist__delete(rawlist);
@@ -426,24 +500,36 @@ static struct strlist *get_perf_event_names(int fd)
 	return sl;
 }
 
-static int write_trace_kprobe_event(int fd, const char *buf)
+static void write_trace_kprobe_event(int fd, const char *buf)
 {
 	int ret;
 
+	pr_debug("Writing event: %s\n", buf);
 	ret = write(fd, buf, strlen(buf));
 	if (ret <= 0)
-		die("Failed to create event.");
-	else
-		printf("Added new event: %s\n", buf);
-
-	return ret;
+		die("Failed to write event: %s", strerror(errno));
 }
 
 static void get_new_event_name(char *buf, size_t len, const char *base,
-			       struct strlist *namelist)
+			       struct strlist *namelist, bool allow_suffix)
 {
 	int i, ret;
-	for (i = 0; i < MAX_EVENT_INDEX; i++) {
+
+	/* Try no suffix */
+	ret = e_snprintf(buf, len, "%s", base);
+	if (ret < 0)
+		die("snprintf() failed: %s", strerror(-ret));
+	if (!strlist__has_entry(namelist, buf))
+		return;
+
+	if (!allow_suffix) {
+		pr_warning("Error: event \"%s\" already exists. "
+			   "(Use -f to force duplicates.)\n", base);
+		die("Can't add new event.");
+	}
+
+	/* Try to add suffix */
+	for (i = 1; i < MAX_EVENT_INDEX; i++) {
 		ret = e_snprintf(buf, len, "%s_%d", base, i);
 		if (ret < 0)
 			die("snprintf() failed: %s", strerror(-ret));
@@ -454,31 +540,138 @@ static void get_new_event_name(char *buf, size_t len, const char *base,
 		die("Too many events are on the same function.");
 }
 
-void add_trace_kprobe_events(struct probe_point *probes, int nr_probes)
+void add_trace_kprobe_events(struct probe_point *probes, int nr_probes,
+			     bool force_add)
 {
 	int i, j, fd;
 	struct probe_point *pp;
 	char buf[MAX_CMDLEN];
 	char event[64];
 	struct strlist *namelist;
+	bool allow_suffix;
 
 	fd = open_kprobe_events(O_RDWR, O_APPEND);
 	/* Get current event names */
-	namelist = get_perf_event_names(fd);
+	namelist = get_perf_event_names(fd, false);
 
 	for (j = 0; j < nr_probes; j++) {
 		pp = probes + j;
+		if (!pp->event)
+			pp->event = strdup(pp->function);
+		if (!pp->group)
+			pp->group = strdup(PERFPROBE_GROUP);
+		DIE_IF(!pp->event || !pp->group);
+		/* If force_add is true, suffix search is allowed */
+		allow_suffix = force_add;
 		for (i = 0; i < pp->found; i++) {
 			/* Get an unused new event name */
-			get_new_event_name(event, 64, pp->function, namelist);
+			get_new_event_name(event, 64, pp->event, namelist,
+					   allow_suffix);
 			snprintf(buf, MAX_CMDLEN, "%c:%s/%s %s\n",
 				 pp->retprobe ? 'r' : 'p',
-				 PERFPROBE_GROUP, event,
+				 pp->group, event,
 				 pp->probes[i]);
 			write_trace_kprobe_event(fd, buf);
+			printf("Added new event:\n");
+			/* Get the first parameter (probe-point) */
+			sscanf(pp->probes[i], "%s", buf);
+			show_perf_probe_event(event, buf, pp);
 			/* Add added event name to namelist */
 			strlist__add(namelist, event);
+			/*
+			 * Probes after the first probe which comes from same
+			 * user input are always allowed to add suffix, because
+			 * there might be several addresses corresponding to
+			 * one code line.
+			 */
+			allow_suffix = true;
+		}
+	}
+	/* Show how to use the event. */
+	printf("\nYou can now use it on all perf tools, such as:\n\n");
+	printf("\tperf record -e %s:%s -a sleep 1\n\n", PERFPROBE_GROUP, event);
+
+	strlist__delete(namelist);
+	close(fd);
+}
+
+static void __del_trace_kprobe_event(int fd, struct str_node *ent)
+{
+	char *p;
+	char buf[128];
+
+	/* Convert from perf-probe event to trace-kprobe event */
+	if (e_snprintf(buf, 128, "-:%s", ent->s) < 0)
+		die("Failed to copy event.");
+	p = strchr(buf + 2, ':');
+	if (!p)
+		die("Internal error: %s should have ':' but not.", ent->s);
+	*p = '/';
+
+	write_trace_kprobe_event(fd, buf);
+	printf("Remove event: %s\n", ent->s);
+}
+
+static void del_trace_kprobe_event(int fd, const char *group,
+				   const char *event, struct strlist *namelist)
+{
+	char buf[128];
+	struct str_node *ent, *n;
+	int found = 0;
+
+	if (e_snprintf(buf, 128, "%s:%s", group, event) < 0)
+		die("Failed to copy event.");
+
+	if (strpbrk(buf, "*?")) { /* Glob-exp */
+		strlist__for_each_safe(ent, n, namelist)
+			if (strglobmatch(ent->s, buf)) {
+				found++;
+				__del_trace_kprobe_event(fd, ent);
+				strlist__remove(namelist, ent);
+			}
+	} else {
+		ent = strlist__find(namelist, buf);
+		if (ent) {
+			found++;
+			__del_trace_kprobe_event(fd, ent);
+			strlist__remove(namelist, ent);
 		}
 	}
+	if (found == 0)
+		pr_info("Info: event \"%s\" does not exist, could not remove it.\n", buf);
+}
+
+void del_trace_kprobe_events(struct strlist *dellist)
+{
+	int fd;
+	const char *group, *event;
+	char *p, *str;
+	struct str_node *ent;
+	struct strlist *namelist;
+
+	fd = open_kprobe_events(O_RDWR, O_APPEND);
+	/* Get current event names */
+	namelist = get_perf_event_names(fd, true);
+
+	strlist__for_each(ent, dellist) {
+		str = strdup(ent->s);
+		if (!str)
+			die("Failed to copy event.");
+		pr_debug("Parsing: %s\n", str);
+		p = strchr(str, ':');
+		if (p) {
+			group = str;
+			*p = '\0';
+			event = p + 1;
+		} else {
+			group = "*";
+			event = str;
+		}
+		pr_debug("Group: %s, Event: %s\n", group, event);
+		del_trace_kprobe_event(fd, group, event, namelist);
+		free(str);
+	}
+	strlist__delete(namelist);
 	close(fd);
 }
+
diff --git a/tools/perf/util/probe-event.h b/tools/perf/util/probe-event.h
index 0c6fe56fe38..7f1d499118c 100644
--- a/tools/perf/util/probe-event.h
+++ b/tools/perf/util/probe-event.h
@@ -1,15 +1,19 @@
 #ifndef _PROBE_EVENT_H
 #define _PROBE_EVENT_H
 
+#include <stdbool.h>
 #include "probe-finder.h"
 #include "strlist.h"
 
-extern int parse_perf_probe_event(const char *str, struct probe_point *pp);
+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);
 extern int synthesize_perf_probe_event(struct probe_point *pp);
-extern void parse_trace_kprobe_event(const char *str, char **group,
-				     char **event, struct probe_point *pp);
+extern void parse_trace_kprobe_event(const char *str, struct probe_point *pp);
 extern int synthesize_trace_kprobe_event(struct probe_point *pp);
-extern void add_trace_kprobe_events(struct probe_point *probes, int nr_probes);
+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);
 
 /* Maximum index number of event-name postfix */
diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
index 293cdfc1b8c..4b852c0d16a 100644
--- a/tools/perf/util/probe-finder.c
+++ b/tools/perf/util/probe-finder.c
@@ -106,7 +106,7 @@ static int strtailcmp(const char *s1, const char *s2)
 {
 	int i1 = strlen(s1);
 	int i2 = strlen(s2);
-	while (--i1 > 0 && --i2 > 0) {
+	while (--i1 >= 0 && --i2 >= 0) {
 		if (s1[i1] != s2[i2])
 			return s1[i1] - s2[i2];
 	}
@@ -687,10 +687,8 @@ int find_probepoint(int fd, struct probe_point *pp)
 	struct probe_finder pf = {.pp = pp};
 
 	ret = dwarf_init(fd, DW_DLC_READ, 0, 0, &__dw_debug, &__dw_error);
-	if (ret != DW_DLV_OK) {
-		pr_warning("No dwarf info found in the vmlinux - please rebuild with CONFIG_DEBUG_INFO.\n");
+	if (ret != DW_DLV_OK)
 		return -ENOENT;
-	}
 
 	pp->found = 0;
 	while (++cu_number) {
diff --git a/tools/perf/util/probe-finder.h b/tools/perf/util/probe-finder.h
index bdebca6697d..a4086aaddb7 100644
--- a/tools/perf/util/probe-finder.h
+++ b/tools/perf/util/probe-finder.h
@@ -1,9 +1,9 @@
 #ifndef _PROBE_FINDER_H
 #define _PROBE_FINDER_H
 
-#define MAX_PATH_LEN 256
-#define MAX_PROBE_BUFFER 1024
-#define MAX_PROBES 128
+#define MAX_PATH_LEN		 256
+#define MAX_PROBE_BUFFER	1024
+#define MAX_PROBES		 128
 
 static inline int is_c_varname(const char *name)
 {
@@ -12,45 +12,53 @@ static inline int is_c_varname(const char *name)
 }
 
 struct probe_point {
+	char			*event;			/* Event name */
+	char			*group;			/* Event group */
+
 	/* Inputs */
-	char	*file;		/* File name */
-	int	line;		/* Line number */
+	char			*file;			/* File name */
+	int			line;			/* Line number */
 
-	char	*function;	/* Function name */
-	int	offset;		/* Offset bytes */
+	char			*function;		/* Function name */
+	int			offset;			/* Offset bytes */
 
-	int	nr_args;	/* Number of arguments */
-	char	**args;		/* Arguments */
+	int			nr_args;		/* Number of arguments */
+	char			**args;			/* Arguments */
 
-	int	retprobe;	/* Return probe */
+	int			retprobe;		/* Return probe */
 
 	/* Output */
-	int	found;		/* Number of found probe points */
-	char	*probes[MAX_PROBES];	/* Output buffers (will be allocated)*/
+	int			found;			/* Number of found probe points */
+	char			*probes[MAX_PROBES];	/* Output buffers (will be allocated)*/
 };
 
 #ifndef NO_LIBDWARF
 extern int find_probepoint(int fd, struct probe_point *pp);
 
-#include <libdwarf/dwarf.h>
-#include <libdwarf/libdwarf.h>
+/* Workaround for undefined _MIPS_SZLONG bug in libdwarf.h: */
+#ifndef _MIPS_SZLONG
+# define _MIPS_SZLONG		0
+#endif
+
+#include <dwarf.h>
+#include <libdwarf.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 */
+	Dwarf_Unsigned		fno;			/* File number */
+	Dwarf_Unsigned		lno;			/* Line number */
+	Dwarf_Off		inl_offs;		/* Inline offset */
+	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_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 */
 };
 #endif /* NO_LIBDWARF */
 
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
new file mode 100644
index 00000000000..ce3a6c8abe7
--- /dev/null
+++ b/tools/perf/util/session.c
@@ -0,0 +1,150 @@
+#include <linux/kernel.h>
+
+#include <unistd.h>
+#include <sys/types.h>
+
+#include "session.h"
+#include "sort.h"
+#include "util.h"
+
+static int perf_session__open(struct perf_session *self, bool force)
+{
+	struct stat input_stat;
+
+	self->fd = open(self->filename, O_RDONLY);
+	if (self->fd < 0) {
+		pr_err("failed to open file: %s", self->filename);
+		if (!strcmp(self->filename, "perf.data"))
+			pr_err("  (try 'perf record' first)");
+		pr_err("\n");
+		return -errno;
+	}
+
+	if (fstat(self->fd, &input_stat) < 0)
+		goto out_close;
+
+	if (!force && input_stat.st_uid && (input_stat.st_uid != geteuid())) {
+		pr_err("file %s not owned by current user or root\n",
+		       self->filename);
+		goto out_close;
+	}
+
+	if (!input_stat.st_size) {
+		pr_info("zero-sized file (%s), nothing to do!\n",
+			self->filename);
+		goto out_close;
+	}
+
+	if (perf_header__read(&self->header, self->fd) < 0) {
+		pr_err("incompatible file format");
+		goto out_close;
+	}
+
+	self->size = input_stat.st_size;
+	return 0;
+
+out_close:
+	close(self->fd);
+	self->fd = -1;
+	return -1;
+}
+
+struct perf_session *perf_session__new(const char *filename, int mode, bool force)
+{
+	size_t len = filename ? strlen(filename) + 1 : 0;
+	struct perf_session *self = zalloc(sizeof(*self) + len);
+
+	if (self == NULL)
+		goto out;
+
+	if (perf_header__init(&self->header) < 0)
+		goto out_free;
+
+	memcpy(self->filename, filename, len);
+	self->threads = RB_ROOT;
+	self->last_match = NULL;
+	self->mmap_window = 32;
+	self->cwd = NULL;
+	self->cwdlen = 0;
+	map_groups__init(&self->kmaps);
+
+	if (perf_session__create_kernel_maps(self) < 0)
+		goto out_delete;
+
+	if (mode == O_RDONLY && perf_session__open(self, force) < 0)
+		goto out_delete;
+out:
+	return self;
+out_free:
+	free(self);
+	return NULL;
+out_delete:
+	perf_session__delete(self);
+	return NULL;
+}
+
+void perf_session__delete(struct perf_session *self)
+{
+	perf_header__exit(&self->header);
+	close(self->fd);
+	free(self->cwd);
+	free(self);
+}
+
+static bool symbol__match_parent_regex(struct symbol *sym)
+{
+	if (sym->name && !regexec(&parent_regex, sym->name, 0, NULL, 0))
+		return 1;
+
+	return 0;
+}
+
+struct symbol **perf_session__resolve_callchain(struct perf_session *self,
+						struct thread *thread,
+						struct ip_callchain *chain,
+						struct symbol **parent)
+{
+	u8 cpumode = PERF_RECORD_MISC_USER;
+	struct symbol **syms = NULL;
+	unsigned int i;
+
+	if (symbol_conf.use_callchain) {
+		syms = calloc(chain->nr, sizeof(*syms));
+		if (!syms) {
+			fprintf(stderr, "Can't allocate memory for symbols\n");
+			exit(-1);
+		}
+	}
+
+	for (i = 0; i < chain->nr; i++) {
+		u64 ip = chain->ips[i];
+		struct addr_location al;
+
+		if (ip >= PERF_CONTEXT_MAX) {
+			switch (ip) {
+			case PERF_CONTEXT_HV:
+				cpumode = PERF_RECORD_MISC_HYPERVISOR;	break;
+			case PERF_CONTEXT_KERNEL:
+				cpumode = PERF_RECORD_MISC_KERNEL;	break;
+			case PERF_CONTEXT_USER:
+				cpumode = PERF_RECORD_MISC_USER;	break;
+			default:
+				break;
+			}
+			continue;
+		}
+
+		thread__find_addr_location(thread, self, cpumode,
+					   MAP__FUNCTION, ip, &al, NULL);
+		if (al.sym != NULL) {
+			if (sort__has_parent && !*parent &&
+			    symbol__match_parent_regex(al.sym))
+				*parent = al.sym;
+			if (!symbol_conf.use_callchain)
+				break;
+			syms[i] = al.sym;
+		}
+	}
+
+	return syms;
+}
diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h
new file mode 100644
index 00000000000..32eaa1bada0
--- /dev/null
+++ b/tools/perf/util/session.h
@@ -0,0 +1,61 @@
+#ifndef __PERF_SESSION_H
+#define __PERF_SESSION_H
+
+#include "event.h"
+#include "header.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;
+	unsigned long		size;
+	unsigned long		mmap_window;
+	struct map_groups	kmaps;
+	struct rb_root		threads;
+	struct thread		*last_match;
+	struct events_stats	events_stats;
+	unsigned long		event_total[PERF_RECORD_MAX];
+	struct rb_root		hists;
+	u64			sample_type;
+	int			fd;
+	int			cwdlen;
+	char			*cwd;
+	char filename[0];
+};
+
+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;
+};
+
+struct perf_session *perf_session__new(const char *filename, int mode, bool force);
+void perf_session__delete(struct perf_session *self);
+
+int perf_session__process_events(struct perf_session *self,
+				 struct perf_event_ops *event_ops);
+
+struct symbol **perf_session__resolve_callchain(struct perf_session *self,
+						struct thread *thread,
+						struct ip_callchain *chain,
+						struct symbol **parent);
+
+int perf_header__read_build_ids(int input, u64 offset, u64 file_size);
+
+#endif /* __PERF_SESSION_H */
diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c
index b490354d1b2..cb0f327de9e 100644
--- a/tools/perf/util/sort.c
+++ b/tools/perf/util/sort.c
@@ -288,3 +288,29 @@ int sort_dimension__add(const char *tok)
 
 	return -ESRCH;
 }
+
+void setup_sorting(const char * const usagestr[], const struct option *opts)
+{
+	char *tmp, *tok, *str = strdup(sort_order);
+
+	for (tok = strtok_r(str, ", ", &tmp);
+			tok; tok = strtok_r(NULL, ", ", &tmp)) {
+		if (sort_dimension__add(tok) < 0) {
+			error("Unknown --sort key: `%s'", tok);
+			usage_with_options(usagestr, opts);
+		}
+	}
+
+	free(str);
+}
+
+void sort_entry__setup_elide(struct sort_entry *self, struct strlist *list,
+			     const char *list_name, FILE *fp)
+{
+	if (list && strlist__nr_entries(list) == 1) {
+		if (fp != NULL)
+			fprintf(fp, "# %s: %s\n", list_name,
+				strlist__entry(list, 0)->s);
+		self->elide = true;
+	}
+}
diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h
index 333e664ff45..753f9ea99fb 100644
--- a/tools/perf/util/sort.h
+++ b/tools/perf/util/sort.h
@@ -49,9 +49,13 @@ struct hist_entry {
 	struct symbol		*sym;
 	u64			ip;
 	char			level;
-	struct symbol		*parent;
+	struct symbol	  *parent;
 	struct callchain_node	callchain;
-	struct rb_root		sorted_chain;
+	union {
+		unsigned long	  position;
+		struct hist_entry *pair;
+		struct rb_root	  sorted_chain;
+	};
 };
 
 enum sort_type {
@@ -81,6 +85,8 @@ struct sort_entry {
 extern struct sort_entry sort_thread;
 extern struct list_head hist_entry__sort_list;
 
+void setup_sorting(const char * const usagestr[], const struct option *opts);
+
 extern int repsep_fprintf(FILE *fp, const char *fmt, ...);
 extern size_t sort__thread_print(FILE *, struct hist_entry *, unsigned int);
 extern size_t sort__comm_print(FILE *, struct hist_entry *, unsigned int);
@@ -95,5 +101,7 @@ extern int64_t sort__sym_cmp(struct hist_entry *, struct hist_entry *);
 extern int64_t sort__parent_cmp(struct hist_entry *, struct hist_entry *);
 extern size_t sort__parent_print(FILE *, struct hist_entry *, unsigned int);
 extern int sort_dimension__add(const char *);
+void sort_entry__setup_elide(struct sort_entry *self, struct strlist *list,
+			     const char *list_name, FILE *fp);
 
 #endif	/* __PERF_SORT_H */
diff --git a/tools/perf/util/string.c b/tools/perf/util/string.c
index f24a8cc933d..5352d7dccc6 100644
--- a/tools/perf/util/string.c
+++ b/tools/perf/util/string.c
@@ -226,3 +226,28 @@ fail:
 	argv_free(argv);
 	return NULL;
 }
+
+/* Glob expression pattern matching */
+bool strglobmatch(const char *str, const char *pat)
+{
+	while (*str && *pat && *pat != '*') {
+		if (*pat == '?') {
+			str++;
+			pat++;
+		} else
+			if (*str++ != *pat++)
+				return false;
+	}
+	/* Check wild card */
+	if (*pat == '*') {
+		while (*pat == '*')
+			pat++;
+		if (!*pat)	/* Tail wild card matches all */
+			return true;
+		while (*str)
+			if (strglobmatch(str++, pat))
+				return true;
+	}
+	return !*str && !*pat;
+}
+
diff --git a/tools/perf/util/string.h b/tools/perf/util/string.h
index bfecec265a1..02ede58c54b 100644
--- a/tools/perf/util/string.h
+++ b/tools/perf/util/string.h
@@ -1,6 +1,7 @@
 #ifndef __PERF_STRING_H_
 #define __PERF_STRING_H_
 
+#include <stdbool.h>
 #include "types.h"
 
 int hex2u64(const char *ptr, u64 *val);
@@ -8,6 +9,7 @@ char *strxfrchar(char *s, char from, char to);
 s64 perf_atoll(const char *str);
 char **argv_split(const char *str, int *argcp);
 void argv_free(char **argv);
+bool strglobmatch(const char *str, const char *pat);
 
 #define _STR(x) #x
 #define STR(x) _STR(x)
diff --git a/tools/perf/util/strlist.c b/tools/perf/util/strlist.c
index 7ad38171dc2..6783a204355 100644
--- a/tools/perf/util/strlist.c
+++ b/tools/perf/util/strlist.c
@@ -102,7 +102,7 @@ void strlist__remove(struct strlist *self, struct str_node *sn)
 	str_node__delete(sn, self->dupstr);
 }
 
-bool strlist__has_entry(struct strlist *self, const char *entry)
+struct str_node *strlist__find(struct strlist *self, const char *entry)
 {
 	struct rb_node **p = &self->entries.rb_node;
 	struct rb_node *parent = NULL;
@@ -120,10 +120,10 @@ bool strlist__has_entry(struct strlist *self, const char *entry)
 		else if (rc < 0)
 			p = &(*p)->rb_right;
 		else
-			return true;
+			return sn;
 	}
 
-	return false;
+	return NULL;
 }
 
 static int strlist__parse_list_entry(struct strlist *self, const char *s)
diff --git a/tools/perf/util/strlist.h b/tools/perf/util/strlist.h
index cb4659306d7..3ba839007d2 100644
--- a/tools/perf/util/strlist.h
+++ b/tools/perf/util/strlist.h
@@ -23,7 +23,12 @@ int strlist__load(struct strlist *self, const char *filename);
 int strlist__add(struct strlist *self, const char *str);
 
 struct str_node *strlist__entry(const struct strlist *self, unsigned int idx);
-bool strlist__has_entry(struct strlist *self, const char *entry);
+struct str_node *strlist__find(struct strlist *self, const char *entry);
+
+static inline bool strlist__has_entry(struct strlist *self, const char *entry)
+{
+	return strlist__find(self, entry) != NULL;
+}
 
 static inline bool strlist__empty(const struct strlist *self)
 {
@@ -35,5 +40,39 @@ static inline unsigned int strlist__nr_entries(const struct strlist *self)
 	return self->nr_entries;
 }
 
+/* For strlist iteration */
+static inline struct str_node *strlist__first(struct strlist *self)
+{
+	struct rb_node *rn = rb_first(&self->entries);
+	return rn ? rb_entry(rn, struct str_node, rb_node) : NULL;
+}
+static inline struct str_node *strlist__next(struct str_node *sn)
+{
+	struct rb_node *rn;
+	if (!sn)
+		return NULL;
+	rn = rb_next(&sn->rb_node);
+	return rn ? rb_entry(rn, struct str_node, rb_node) : NULL;
+}
+
+/**
+ * strlist_for_each      - iterate over a strlist
+ * @pos:	the &struct str_node to use as a loop cursor.
+ * @self:	the &struct strlist for loop.
+ */
+#define strlist__for_each(pos, self)	\
+	for (pos = strlist__first(self); pos; pos = strlist__next(pos))
+
+/**
+ * strlist_for_each_safe - iterate over a strlist safe against removal of
+ *                         str_node
+ * @pos:	the &struct str_node to use as a loop cursor.
+ * @n:		another &struct str_node to use as temporary storage.
+ * @self:	the &struct strlist for loop.
+ */
+#define strlist__for_each_safe(pos, n, self)	\
+	for (pos = strlist__first(self), n = strlist__next(pos); pos;\
+	     pos = n, n = strlist__next(n))
+
 int strlist__parse_list(struct strlist *self, const char *s);
 #endif /* __PERF_STRLIST_H */
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index fffcb937cdc..ab92763edb0 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -1,5 +1,7 @@
 #include "util.h"
 #include "../perf.h"
+#include "session.h"
+#include "sort.h"
 #include "string.h"
 #include "symbol.h"
 #include "thread.h"
@@ -29,33 +31,50 @@ enum dso_origin {
 };
 
 static void dsos__add(struct list_head *head, struct dso *dso);
-static struct map *thread__find_map_by_name(struct thread *self, char *name);
 static struct map *map__new2(u64 start, struct dso *dso, enum map_type type);
-struct symbol *dso__find_symbol(struct dso *self, enum map_type type, u64 addr);
 static int dso__load_kernel_sym(struct dso *self, struct map *map,
-				struct thread *thread, symbol_filter_t filter);
-unsigned int symbol__priv_size;
+				struct perf_session *session, symbol_filter_t filter);
 static int vmlinux_path__nr_entries;
 static char **vmlinux_path;
 
-static struct symbol_conf symbol_conf__defaults = {
+struct symbol_conf symbol_conf = {
+	.exclude_other	  = true,
 	.use_modules	  = true,
 	.try_vmlinux_path = true,
 };
 
-static struct thread kthread_mem;
-struct thread *kthread = &kthread_mem;
-
 bool dso__loaded(const struct dso *self, enum map_type type)
 {
 	return self->loaded & (1 << type);
 }
 
+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)
+{
+	switch (map_type) {
+	case MAP__FUNCTION:
+		return symbol_type == 'T' || symbol_type == 'W';
+	case MAP__VARIABLE:
+		return symbol_type == 'D' || symbol_type == 'd';
+	default:
+		return false;
+	}
+}
+
 static void symbols__fixup_end(struct rb_root *self)
 {
 	struct rb_node *nd, *prevnd = rb_first(self);
@@ -79,7 +98,7 @@ static void symbols__fixup_end(struct rb_root *self)
 		curr->end = roundup(curr->start, 4096);
 }
 
-static void __thread__fixup_maps_end(struct thread *self, enum map_type type)
+static void __map_groups__fixup_end(struct map_groups *self, enum map_type type)
 {
 	struct map *prev, *curr;
 	struct rb_node *nd, *prevnd = rb_first(&self->maps[type]);
@@ -102,23 +121,23 @@ static void __thread__fixup_maps_end(struct thread *self, enum map_type type)
 	curr->end = ~0UL;
 }
 
-static void thread__fixup_maps_end(struct thread *self)
+static void map_groups__fixup_end(struct map_groups *self)
 {
 	int i;
 	for (i = 0; i < MAP__NR_TYPES; ++i)
-		__thread__fixup_maps_end(self, i);
+		__map_groups__fixup_end(self, i);
 }
 
 static struct symbol *symbol__new(u64 start, u64 len, const char *name)
 {
 	size_t namelen = strlen(name) + 1;
-	struct symbol *self = zalloc(symbol__priv_size +
+	struct symbol *self = zalloc(symbol_conf.priv_size +
 				     sizeof(*self) + namelen);
 	if (self == NULL)
 		return NULL;
 
-	if (symbol__priv_size)
-		self = ((void *)self) + symbol__priv_size;
+	if (symbol_conf.priv_size)
+		self = ((void *)self) + symbol_conf.priv_size;
 
 	self->start = start;
 	self->end   = len ? start + len - 1 : start;
@@ -132,7 +151,7 @@ static struct symbol *symbol__new(u64 start, u64 len, const char *name)
 
 static void symbol__delete(struct symbol *self)
 {
-	free(((void *)self) - symbol__priv_size);
+	free(((void *)self) - symbol_conf.priv_size);
 }
 
 static size_t symbol__fprintf(struct symbol *self, FILE *fp)
@@ -164,11 +183,11 @@ struct dso *dso__new(const char *name)
 		dso__set_long_name(self, self->name);
 		self->short_name = self->name;
 		for (i = 0; i < MAP__NR_TYPES; ++i)
-			self->symbols[i] = RB_ROOT;
-		self->find_symbol = dso__find_symbol;
+			self->symbols[i] = self->symbol_names[i] = RB_ROOT;
 		self->slen_calculated = 0;
 		self->origin = DSO__ORIG_NOT_FOUND;
 		self->loaded = 0;
+		self->sorted_by_name = 0;
 		self->has_build_id = 0;
 	}
 
@@ -246,11 +265,85 @@ static struct symbol *symbols__find(struct rb_root *self, u64 ip)
 	return NULL;
 }
 
-struct symbol *dso__find_symbol(struct dso *self, enum map_type type, u64 addr)
+struct symbol_name_rb_node {
+	struct rb_node	rb_node;
+	struct symbol	sym;
+};
+
+static void symbols__insert_by_name(struct rb_root *self, struct symbol *sym)
+{
+	struct rb_node **p = &self->rb_node;
+	struct rb_node *parent = NULL;
+	struct symbol_name_rb_node *symn = ((void *)sym) - sizeof(*parent), *s;
+
+	while (*p != NULL) {
+		parent = *p;
+		s = rb_entry(parent, struct symbol_name_rb_node, rb_node);
+		if (strcmp(sym->name, s->sym.name) < 0)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+	rb_link_node(&symn->rb_node, parent, p);
+	rb_insert_color(&symn->rb_node, self);
+}
+
+static void symbols__sort_by_name(struct rb_root *self, struct rb_root *source)
+{
+	struct rb_node *nd;
+
+	for (nd = rb_first(source); nd; nd = rb_next(nd)) {
+		struct symbol *pos = rb_entry(nd, struct symbol, rb_node);
+		symbols__insert_by_name(self, pos);
+	}
+}
+
+static struct symbol *symbols__find_by_name(struct rb_root *self, const char *name)
+{
+	struct rb_node *n;
+
+	if (self == NULL)
+		return NULL;
+
+	n = self->rb_node;
+
+	while (n) {
+		struct symbol_name_rb_node *s;
+		int cmp;
+
+		s = rb_entry(n, struct symbol_name_rb_node, rb_node);
+		cmp = strcmp(name, s->sym.name);
+
+		if (cmp < 0)
+			n = n->rb_left;
+		else if (cmp > 0)
+			n = n->rb_right;
+		else
+			return &s->sym;
+	}
+
+	return NULL;
+}
+
+struct symbol *dso__find_symbol(struct dso *self,
+				enum map_type type, u64 addr)
 {
 	return symbols__find(&self->symbols[type], addr);
 }
 
+struct symbol *dso__find_symbol_by_name(struct dso *self, enum map_type type,
+					const char *name)
+{
+	return symbols__find_by_name(&self->symbol_names[type], name);
+}
+
+void dso__sort_by_name(struct dso *self, enum map_type type)
+{
+	dso__set_sorted_by_name(self, type);
+	return symbols__sort_by_name(&self->symbol_names[type],
+				     &self->symbols[type]);
+}
+
 int build_id__sprintf(u8 *self, int len, char *bf)
 {
 	char *bid = bf;
@@ -327,10 +420,7 @@ static int dso__load_all_kallsyms(struct dso *self, struct map *map)
 			continue;
 
 		symbol_type = toupper(line[len]);
-		/*
-		 * We're interested only in code ('T'ext)
-		 */
-		if (symbol_type != 'T' && symbol_type != 'W')
+		if (!symbol_type__is_a(symbol_type, map->type))
 			continue;
 
 		symbol_name = line + len + 2;
@@ -364,8 +454,8 @@ out_failure:
  * 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 thread *thread,
-			       symbol_filter_t filter)
+static int dso__split_kallsyms(struct dso *self, struct map *map,
+			       struct perf_session *session, symbol_filter_t filter)
 {
 	struct map *curr_map = map;
 	struct symbol *pos;
@@ -382,13 +472,13 @@ static int dso__split_kallsyms(struct dso *self, struct map *map, struct thread
 
 		module = strchr(pos->name, '\t');
 		if (module) {
-			if (!thread->use_modules)
+			if (!symbol_conf.use_modules)
 				goto discard_symbol;
 
 			*module++ = '\0';
 
 			if (strcmp(self->name, module)) {
-				curr_map = thread__find_map_by_name(thread, module);
+				curr_map = map_groups__find_by_name(&session->kmaps, map->type, module);
 				if (curr_map == NULL) {
 					pr_debug("/proc/{kallsyms,modules} "
 					         "inconsistency!\n");
@@ -419,7 +509,7 @@ static int dso__split_kallsyms(struct dso *self, struct map *map, struct thread
 			}
 
 			curr_map->map_ip = curr_map->unmap_ip = identity__map_ip;
-			__thread__insert_map(thread, curr_map);
+			map_groups__insert(&session->kmaps, curr_map);
 			++kernel_range;
 		}
 
@@ -440,7 +530,7 @@ discard_symbol:		rb_erase(&pos->rb_node, root);
 
 
 static int dso__load_kallsyms(struct dso *self, struct map *map,
-			      struct thread *thread, symbol_filter_t filter)
+			      struct perf_session *session, symbol_filter_t filter)
 {
 	if (dso__load_all_kallsyms(self, map) < 0)
 		return -1;
@@ -448,14 +538,7 @@ static int dso__load_kallsyms(struct dso *self, struct map *map,
 	symbols__fixup_end(&self->symbols[map->type]);
 	self->origin = DSO__ORIG_KERNEL;
 
-	return dso__split_kallsyms(self, map, thread, filter);
-}
-
-size_t kernel_maps__fprintf(FILE *fp)
-{
-	size_t printed = fprintf(fp, "Kernel maps:\n");
-	printed += thread__fprintf_maps(kthread, fp);
-	return printed + fprintf(fp, "END kernel maps\n");
+	return dso__split_kallsyms(self, map, session, filter);
 }
 
 static int dso__load_perf_map(struct dso *self, struct map *map,
@@ -544,6 +627,13 @@ static inline int elf_sym__is_function(const GElf_Sym *sym)
 	       sym->st_shndx != SHN_UNDEF;
 }
 
+static inline bool elf_sym__is_object(const GElf_Sym *sym)
+{
+	return elf_sym__type(sym) == STT_OBJECT &&
+		sym->st_name != 0 &&
+		sym->st_shndx != SHN_UNDEF;
+}
+
 static inline int elf_sym__is_label(const GElf_Sym *sym)
 {
 	return elf_sym__type(sym) == STT_NOTYPE &&
@@ -564,6 +654,12 @@ static inline int elf_sec__is_text(const GElf_Shdr *shdr,
 	return strstr(elf_sec__name(shdr, secstrs), "text") != NULL;
 }
 
+static inline bool elf_sec__is_data(const GElf_Shdr *shdr,
+				    const Elf_Data *secstrs)
+{
+	return strstr(elf_sec__name(shdr, secstrs), "data") != NULL;
+}
+
 static inline const char *elf_sym__name(const GElf_Sym *sym,
 					const Elf_Data *symstrs)
 {
@@ -744,8 +840,32 @@ out:
 	return 0;
 }
 
+static bool elf_sym__is_a(GElf_Sym *self, enum map_type type)
+{
+	switch (type) {
+	case MAP__FUNCTION:
+		return elf_sym__is_function(self);
+	case MAP__VARIABLE:
+		return elf_sym__is_object(self);
+	default:
+		return false;
+	}
+}
+
+static bool elf_sec__is_a(GElf_Shdr *self, Elf_Data *secstrs, enum map_type type)
+{
+	switch (type) {
+	case MAP__FUNCTION:
+		return elf_sec__is_text(self, secstrs);
+	case MAP__VARIABLE:
+		return elf_sec__is_data(self, secstrs);
+	default:
+		return false;
+	}
+}
+
 static int dso__load_sym(struct dso *self, struct map *map,
-			 struct thread *thread, const char *name, int fd,
+			 struct perf_session *session, const char *name, int fd,
 			 symbol_filter_t filter, int kernel, int kmodule)
 {
 	struct map *curr_map = map;
@@ -818,7 +938,7 @@ static int dso__load_sym(struct dso *self, struct map *map,
 		int is_label = elf_sym__is_label(&sym);
 		const char *section_name;
 
-		if (!is_label && !elf_sym__is_function(&sym))
+		if (!is_label && !elf_sym__is_a(&sym, map->type))
 			continue;
 
 		sec = elf_getscn(elf, sym.st_shndx);
@@ -827,7 +947,7 @@ static int dso__load_sym(struct dso *self, struct map *map,
 
 		gelf_getshdr(sec, &shdr);
 
-		if (is_label && !elf_sec__is_text(&shdr, secstrs))
+		if (is_label && !elf_sec__is_a(&shdr, secstrs, map->type))
 			continue;
 
 		elf_name = elf_sym__name(&sym, symstrs);
@@ -849,7 +969,7 @@ 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 = thread__find_map_by_name(thread, dso_name);
+			curr_map = map_groups__find_by_name(&session->kmaps, map->type, dso_name);
 			if (curr_map == NULL) {
 				u64 start = sym.st_value;
 
@@ -868,7 +988,7 @@ 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;
-				__thread__insert_map(kthread, curr_map);
+				map_groups__insert(&session->kmaps, curr_map);
 				dsos__add(&dsos__kernel, curr_dso);
 			} else
 				curr_dso = curr_map->dso;
@@ -938,8 +1058,9 @@ static bool __dsos__read_build_ids(struct list_head *head)
 
 bool dsos__read_build_ids(void)
 {
-	return __dsos__read_build_ids(&dsos__kernel) ||
-	       __dsos__read_build_ids(&dsos__user);
+	bool kbuildids = __dsos__read_build_ids(&dsos__kernel),
+	     ubuildids = __dsos__read_build_ids(&dsos__user);
+	return kbuildids || ubuildids;
 }
 
 /*
@@ -1082,7 +1203,8 @@ char dso__symtab_origin(const struct dso *self)
 	return origin[self->origin];
 }
 
-int dso__load(struct dso *self, struct map *map, symbol_filter_t filter)
+int dso__load(struct dso *self, struct map *map, struct perf_session *session,
+	      symbol_filter_t filter)
 {
 	int size = PATH_MAX;
 	char *name;
@@ -1093,7 +1215,7 @@ int dso__load(struct dso *self, struct map *map, symbol_filter_t filter)
 	dso__set_loaded(self, map->type);
 
 	if (self->kernel)
-		return dso__load_kernel_sym(self, map, kthread, filter);
+		return dso__load_kernel_sym(self, map, session, filter);
 
 	name = malloc(size);
 	if (!name)
@@ -1179,11 +1301,12 @@ out:
 	return ret;
 }
 
-static struct map *thread__find_map_by_name(struct thread *self, char *name)
+struct map *map_groups__find_by_name(struct map_groups *self,
+				     enum map_type type, const char *name)
 {
 	struct rb_node *nd;
 
-	for (nd = rb_first(&self->maps[MAP__FUNCTION]); nd; nd = rb_next(nd)) {
+	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)
@@ -1193,7 +1316,7 @@ static struct map *thread__find_map_by_name(struct thread *self, char *name)
 	return NULL;
 }
 
-static int dsos__set_modules_path_dir(char *dirname)
+static int perf_session__set_modules_path_dir(struct perf_session *self, char *dirname)
 {
 	struct dirent *dent;
 	DIR *dir = opendir(dirname);
@@ -1213,7 +1336,7 @@ static int dsos__set_modules_path_dir(char *dirname)
 
 			snprintf(path, sizeof(path), "%s/%s",
 				 dirname, dent->d_name);
-			if (dsos__set_modules_path_dir(path) < 0)
+			if (perf_session__set_modules_path_dir(self, path) < 0)
 				goto failure;
 		} else {
 			char *dot = strrchr(dent->d_name, '.'),
@@ -1227,7 +1350,7 @@ static int dsos__set_modules_path_dir(char *dirname)
 				 (int)(dot - dent->d_name), dent->d_name);
 
 			strxfrchar(dso_name, '-', '_');
-			map = thread__find_map_by_name(kthread, dso_name);
+			map = map_groups__find_by_name(&self->kmaps, MAP__FUNCTION, dso_name);
 			if (map == NULL)
 				continue;
 
@@ -1247,7 +1370,7 @@ failure:
 	return -1;
 }
 
-static int dsos__set_modules_path(void)
+static int perf_session__set_modules_path(struct perf_session *self)
 {
 	struct utsname uts;
 	char modules_path[PATH_MAX];
@@ -1258,7 +1381,7 @@ static int dsos__set_modules_path(void)
 	snprintf(modules_path, sizeof(modules_path), "/lib/modules/%s/kernel",
 		 uts.release);
 
-	return dsos__set_modules_path_dir(modules_path);
+	return perf_session__set_modules_path_dir(self, modules_path);
 }
 
 /*
@@ -1280,7 +1403,7 @@ static struct map *map__new2(u64 start, struct dso *dso, enum map_type type)
 	return self;
 }
 
-static int thread__create_module_maps(struct thread *self)
+static int perf_session__create_module_maps(struct perf_session *self)
 {
 	char *line = NULL;
 	size_t n;
@@ -1337,14 +1460,14 @@ static int thread__create_module_maps(struct thread *self)
 			dso->has_build_id = true;
 
 		dso->origin = DSO__ORIG_KMODULE;
-		__thread__insert_map(self, map);
+		map_groups__insert(&self->kmaps, map);
 		dsos__add(&dsos__kernel, dso);
 	}
 
 	free(line);
 	fclose(file);
 
-	return dsos__set_modules_path();
+	return perf_session__set_modules_path(self);
 
 out_delete_line:
 	free(line);
@@ -1352,7 +1475,8 @@ out_failure:
 	return -1;
 }
 
-static int dso__load_vmlinux(struct dso *self, struct map *map, struct thread *thread,
+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;
@@ -1386,14 +1510,14 @@ static int dso__load_vmlinux(struct dso *self, struct map *map, struct thread *t
 		return -1;
 
 	dso__set_loaded(self, map->type);
-	err = dso__load_sym(self, map, thread, self->long_name, fd, filter, 1, 0);
+	err = dso__load_sym(self, map, session, self->long_name, fd, filter, 1, 0);
 	close(fd);
 
 	return err;
 }
 
 static int dso__load_kernel_sym(struct dso *self, struct map *map,
-				struct thread *thread, symbol_filter_t filter)
+				struct perf_session *session, symbol_filter_t filter)
 {
 	int err;
 	bool is_kallsyms;
@@ -1403,7 +1527,7 @@ static int dso__load_kernel_sym(struct dso *self, struct map *map,
 		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, thread,
+			err = dso__load_vmlinux(self, map, session,
 						vmlinux_path[i], filter);
 			if (err > 0) {
 				pr_debug("Using %s for symbols\n",
@@ -1419,12 +1543,12 @@ static int dso__load_kernel_sym(struct dso *self, struct map *map,
 	if (is_kallsyms)
 		goto do_kallsyms;
 
-	err = dso__load_vmlinux(self, map, thread, self->long_name, filter);
+	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, thread, filter);
+		err = dso__load_kallsyms(self, map, session, filter);
 		if (err > 0 && !is_kallsyms)
                         dso__set_long_name(self, strdup("[kernel.kallsyms]"));
 	}
@@ -1507,42 +1631,59 @@ size_t dsos__fprintf_buildid(FILE *fp)
 		__dsos__fprintf_buildid(&dsos__user, fp));
 }
 
-static int thread__create_kernel_map(struct thread *self, const char *vmlinux)
+static struct dso *dsos__create_kernel( const char *vmlinux)
 {
-	struct map *kmap;
 	struct dso *kernel = dso__new(vmlinux ?: "[kernel.kallsyms]");
 
 	if (kernel == NULL)
-		return -1;
-
-	kmap = map__new2(0, kernel, MAP__FUNCTION);
-	if (kmap == NULL)
-		goto out_delete_kernel_dso;
+		return NULL;
 
-	kmap->map_ip	   = kmap->unmap_ip = identity__map_ip;
 	kernel->short_name = "[kernel]";
 	kernel->kernel	   = 1;
 
 	vdso = dso__new("[vdso]");
 	if (vdso == NULL)
-		goto out_delete_kernel_map;
+		goto out_delete_kernel_dso;
 	dso__set_loaded(vdso, MAP__FUNCTION);
 
 	if (sysfs__read_build_id("/sys/kernel/notes", kernel->build_id,
 				 sizeof(kernel->build_id)) == 0)
 		kernel->has_build_id = true;
 
-	__thread__insert_map(self, kmap);
 	dsos__add(&dsos__kernel, kernel);
 	dsos__add(&dsos__user, vdso);
 
-	return 0;
+	return kernel;
 
-out_delete_kernel_map:
-	map__delete(kmap);
 out_delete_kernel_dso:
 	dso__delete(kernel);
-	return -1;
+	return NULL;
+}
+
+static int map_groups__create_kernel_maps(struct map_groups *self, const char *vmlinux)
+{
+	struct map *functions, *variables;
+	struct dso *kernel = dsos__create_kernel(vmlinux);
+
+	if (kernel == NULL)
+		return -1;
+
+	functions = map__new2(0, kernel, MAP__FUNCTION);
+	if (functions == NULL)
+		return -1;
+
+	variables = map__new2(0, kernel, MAP__VARIABLE);
+	if (variables == NULL) {
+		map__delete(functions);
+		return -1;
+	}
+
+	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);
+
+	return 0;
 }
 
 static void vmlinux_path__exit(void)
@@ -1600,29 +1741,69 @@ out_fail:
 	return -1;
 }
 
-int symbol__init(struct symbol_conf *conf)
+static int setup_list(struct strlist **list, const char *list_str,
+		      const char *list_name)
 {
-	const struct symbol_conf *pconf = conf ?: &symbol_conf__defaults;
+	if (list_str == NULL)
+		return 0;
 
+	*list = strlist__new(true, list_str);
+	if (!*list) {
+		pr_err("problems parsing %s list\n", list_name);
+		return -1;
+	}
+	return 0;
+}
+
+int symbol__init(void)
+{
 	elf_version(EV_CURRENT);
-	symbol__priv_size = pconf->priv_size;
-	thread__init(kthread, 0);
+	if (symbol_conf.sort_by_name)
+		symbol_conf.priv_size += (sizeof(struct symbol_name_rb_node) -
+					  sizeof(struct symbol));
 
-	if (pconf->try_vmlinux_path && vmlinux_path__init() < 0)
+	if (symbol_conf.try_vmlinux_path && vmlinux_path__init() < 0)
 		return -1;
 
-	if (thread__create_kernel_map(kthread, pconf->vmlinux_name) < 0) {
-		vmlinux_path__exit();
+	if (symbol_conf.field_sep && *symbol_conf.field_sep == '.') {
+		pr_err("'.' is the only non valid --field-separator argument\n");
 		return -1;
 	}
 
-	kthread->use_modules = pconf->use_modules;
-	if (pconf->use_modules && thread__create_module_maps(kthread) < 0)
-		pr_debug("Failed to load list of modules in use, "
-			 "continuing...\n");
+	if (setup_list(&symbol_conf.dso_list,
+		       symbol_conf.dso_list_str, "dso") < 0)
+		return -1;
+
+	if (setup_list(&symbol_conf.comm_list,
+		       symbol_conf.comm_list_str, "comm") < 0)
+		goto out_free_dso_list;
+
+	if (setup_list(&symbol_conf.sym_list,
+		       symbol_conf.sym_list_str, "symbol") < 0)
+		goto out_free_comm_list;
+
+	return 0;
+
+out_free_dso_list:
+	strlist__delete(symbol_conf.dso_list);
+out_free_comm_list:
+	strlist__delete(symbol_conf.comm_list);
+	return -1;
+}
+
+int perf_session__create_kernel_maps(struct perf_session *self)
+{
+	if (map_groups__create_kernel_maps(&self->kmaps,
+					   symbol_conf.vmlinux_name) < 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);
 	/*
 	 * Now that we have all the maps created, just set the ->end of them:
 	 */
-	thread__fixup_maps_end(kthread);
+	map_groups__fixup_end(&self->kmaps);
 	return 0;
 }
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index 17003efa0b3..8aded2356f7 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -49,18 +49,32 @@ struct symbol {
 	char		name[0];
 };
 
+struct strlist;
+
 struct symbol_conf {
 	unsigned short	priv_size;
 	bool		try_vmlinux_path,
-			use_modules;
-	const char	*vmlinux_name;
+			use_modules,
+			sort_by_name,
+			show_nr_samples,
+			use_callchain,
+			exclude_other;
+	const char	*vmlinux_name,
+			*field_sep;
+	char            *dso_list_str,
+			*comm_list_str,
+			*sym_list_str,
+			*col_width_list_str;
+       struct strlist	*dso_list,
+			*comm_list,
+			*sym_list;
 };
 
-extern unsigned int symbol__priv_size;
+extern struct symbol_conf symbol_conf;
 
 static inline void *symbol__priv(struct symbol *self)
 {
-	return ((void *)self) - symbol__priv_size;
+	return ((void *)self) - symbol_conf.priv_size;
 }
 
 struct addr_location {
@@ -69,18 +83,19 @@ struct addr_location {
 	struct symbol *sym;
 	u64	      addr;
 	char	      level;
+	bool	      filtered;
 };
 
 struct dso {
 	struct list_head node;
 	struct rb_root	 symbols[MAP__NR_TYPES];
-	struct symbol    *(*find_symbol)(struct dso *self,
-					 enum map_type type, u64 addr);
+	struct rb_root	 symbol_names[MAP__NR_TYPES];
 	u8		 adjust_symbols:1;
 	u8		 slen_calculated:1;
 	u8		 has_build_id:1;
 	u8		 kernel:1;
 	unsigned char	 origin;
+	u8		 sorted_by_name;
 	u8		 loaded;
 	u8		 build_id[BUILD_ID_SIZE];
 	u16		 long_name_len;
@@ -93,9 +108,15 @@ struct dso *dso__new(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);
+
+void dso__sort_by_name(struct dso *self, enum map_type type);
+
+struct perf_session;
 
 struct dso *dsos__findnew(const char *name);
-int dso__load(struct dso *self, struct map *map, symbol_filter_t filter);
+int dso__load(struct dso *self, struct map *map, struct perf_session *session,
+	      symbol_filter_t filter);
 void dsos__fprintf(FILE *fp);
 size_t dsos__fprintf_buildid(FILE *fp);
 
@@ -103,18 +124,18 @@ 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_build_id(struct dso *self, void *build_id);
+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);
 
-size_t kernel_maps__fprintf(FILE *fp);
-
-int symbol__init(struct symbol_conf *conf);
+int symbol__init(void);
+int perf_session__create_kernel_maps(struct perf_session *self);
 
-struct thread;
-struct thread *kthread;
 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 603f5610861..4a08dcf50b6 100644
--- a/tools/perf/util/thread.c
+++ b/tools/perf/util/thread.c
@@ -2,18 +2,14 @@
 #include <stdlib.h>
 #include <stdio.h>
 #include <string.h>
+#include "session.h"
 #include "thread.h"
 #include "util.h"
 #include "debug.h"
 
-static struct rb_root threads;
-static struct thread *last_match;
-
-void thread__init(struct thread *self, pid_t pid)
+void map_groups__init(struct map_groups *self)
 {
 	int i;
-	self->pid = pid;
-	self->comm = NULL;
 	for (i = 0; i < MAP__NR_TYPES; ++i) {
 		self->maps[i] = RB_ROOT;
 		INIT_LIST_HEAD(&self->removed_maps[i]);
@@ -25,7 +21,8 @@ static struct thread *thread__new(pid_t pid)
 	struct thread *self = zalloc(sizeof(*self));
 
 	if (self != NULL) {
-		thread__init(self, pid);
+		map_groups__init(&self->mg);
+		self->pid = pid;
 		self->comm = malloc(32);
 		if (self->comm)
 			snprintf(self->comm, 32, ":%d", self->pid);
@@ -55,10 +52,11 @@ int thread__comm_len(struct thread *self)
 
 static const char *map_type__name[MAP__NR_TYPES] = {
 	[MAP__FUNCTION] = "Functions",
+	[MAP__VARIABLE] = "Variables",
 };
 
-static size_t __thread__fprintf_maps(struct thread *self,
-				     enum map_type type, FILE *fp)
+static size_t __map_groups__fprintf_maps(struct map_groups *self,
+					 enum map_type type, FILE *fp)
 {
 	size_t printed = fprintf(fp, "%s:\n", map_type__name[type]);
 	struct rb_node *nd;
@@ -76,16 +74,16 @@ static size_t __thread__fprintf_maps(struct thread *self,
 	return printed;
 }
 
-size_t thread__fprintf_maps(struct thread *self, FILE *fp)
+size_t map_groups__fprintf_maps(struct map_groups *self, FILE *fp)
 {
 	size_t printed = 0, i;
 	for (i = 0; i < MAP__NR_TYPES; ++i)
-		printed += __thread__fprintf_maps(self, i, fp);
+		printed += __map_groups__fprintf_maps(self, i, fp);
 	return printed;
 }
 
-static size_t __thread__fprintf_removed_maps(struct thread *self,
-					     enum map_type type, FILE *fp)
+static size_t __map_groups__fprintf_removed_maps(struct map_groups *self,
+						 enum map_type type, FILE *fp)
 {
 	struct map *pos;
 	size_t printed = 0;
@@ -101,25 +99,30 @@ static size_t __thread__fprintf_removed_maps(struct thread *self,
 	return printed;
 }
 
-static size_t thread__fprintf_removed_maps(struct thread *self, FILE *fp)
+static size_t map_groups__fprintf_removed_maps(struct map_groups *self, FILE *fp)
 {
 	size_t printed = 0, i;
 	for (i = 0; i < MAP__NR_TYPES; ++i)
-		printed += __thread__fprintf_removed_maps(self, i, fp);
+		printed += __map_groups__fprintf_removed_maps(self, i, fp);
 	return printed;
 }
 
-static size_t thread__fprintf(struct thread *self, FILE *fp)
+static size_t map_groups__fprintf(struct map_groups *self, FILE *fp)
 {
-	size_t printed = fprintf(fp, "Thread %d %s\n", self->pid, self->comm);
-	printed += thread__fprintf_removed_maps(self, fp);
+	size_t printed = map_groups__fprintf_maps(self, fp);
 	printed += fprintf(fp, "Removed maps:\n");
-	return printed + thread__fprintf_removed_maps(self, fp);
+	return printed + map_groups__fprintf_removed_maps(self, fp);
+}
+
+static size_t thread__fprintf(struct thread *self, FILE *fp)
+{
+	return fprintf(fp, "Thread %d %s\n", self->pid, self->comm) +
+	       map_groups__fprintf(&self->mg, fp);
 }
 
-struct thread *threads__findnew(pid_t pid)
+struct thread *perf_session__findnew(struct perf_session *self, pid_t pid)
 {
-	struct rb_node **p = &threads.rb_node;
+	struct rb_node **p = &self->threads.rb_node;
 	struct rb_node *parent = NULL;
 	struct thread *th;
 
@@ -128,15 +131,15 @@ struct thread *threads__findnew(pid_t pid)
 	 * so most of the time we dont have to look up
 	 * the full rbtree:
 	 */
-	if (last_match && last_match->pid == pid)
-		return last_match;
+	if (self->last_match && self->last_match->pid == pid)
+		return self->last_match;
 
 	while (*p != NULL) {
 		parent = *p;
 		th = rb_entry(parent, struct thread, rb_node);
 
 		if (th->pid == pid) {
-			last_match = th;
+			self->last_match = th;
 			return th;
 		}
 
@@ -149,26 +152,15 @@ struct thread *threads__findnew(pid_t pid)
 	th = thread__new(pid);
 	if (th != NULL) {
 		rb_link_node(&th->rb_node, parent, p);
-		rb_insert_color(&th->rb_node, &threads);
-		last_match = th;
+		rb_insert_color(&th->rb_node, &self->threads);
+		self->last_match = th;
 	}
 
 	return th;
 }
 
-struct thread *register_idle_thread(void)
-{
-	struct thread *thread = threads__findnew(0);
-
-	if (!thread || thread__set_comm(thread, "swapper")) {
-		fprintf(stderr, "problem inserting idle task.\n");
-		exit(-1);
-	}
-
-	return thread;
-}
-
-static void thread__remove_overlappings(struct thread *self, struct map *map)
+static void map_groups__remove_overlappings(struct map_groups *self,
+					    struct map *map)
 {
 	struct rb_root *root = &self->maps[map->type];
 	struct rb_node *next = rb_first(root);
@@ -238,12 +230,15 @@ struct map *maps__find(struct rb_root *maps, u64 ip)
 
 void thread__insert_map(struct thread *self, struct map *map)
 {
-	thread__remove_overlappings(self, map);
-	maps__insert(&self->maps[map->type], map);
+	map_groups__remove_overlappings(&self->mg, map);
+	map_groups__insert(&self->mg, map);
 }
 
-static int thread__clone_maps(struct thread *self, struct thread *parent,
-			      enum map_type type)
+/*
+ * XXX This should not really _copy_ te maps, but refcount them.
+ */
+static int map_groups__clone(struct map_groups *self,
+			     struct map_groups *parent, enum map_type type)
 {
 	struct rb_node *nd;
 	for (nd = rb_first(&parent->maps[type]); nd; nd = rb_next(nd)) {
@@ -251,7 +246,7 @@ static int thread__clone_maps(struct thread *self, struct thread *parent,
 		struct map *new = map__clone(map);
 		if (new == NULL)
 			return -ENOMEM;
-		thread__insert_map(self, new);
+		map_groups__insert(self, new);
 	}
 	return 0;
 }
@@ -267,17 +262,17 @@ int thread__fork(struct thread *self, struct thread *parent)
 		return -ENOMEM;
 
 	for (i = 0; i < MAP__NR_TYPES; ++i)
-		if (thread__clone_maps(self, parent, i) < 0)
+		if (map_groups__clone(&self->mg, &parent->mg, i) < 0)
 			return -ENOMEM;
 	return 0;
 }
 
-size_t threads__fprintf(FILE *fp)
+size_t perf_session__fprintf(struct perf_session *self, FILE *fp)
 {
 	size_t ret = 0;
 	struct rb_node *nd;
 
-	for (nd = rb_first(&threads); nd; nd = rb_next(nd)) {
+	for (nd = rb_first(&self->threads); nd; nd = rb_next(nd)) {
 		struct thread *pos = rb_entry(nd, struct thread, rb_node);
 
 		ret += thread__fprintf(pos, fp);
@@ -286,14 +281,15 @@ size_t threads__fprintf(FILE *fp)
 	return ret;
 }
 
-struct symbol *thread__find_symbol(struct thread *self,
-				   enum map_type type, u64 addr,
-				   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)
 {
-	struct map *map = thread__find_map(self, type, addr);
+	struct map *map = map_groups__find(self, type, addr);
 
 	if (map != NULL)
-		return map__find_symbol(map, map->map_ip(map, addr), filter);
+		return map__find_symbol(map, session, map->map_ip(map, addr), filter);
 
 	return NULL;
 }
diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h
index 686d6e914d9..c206f72c888 100644
--- a/tools/perf/util/thread.h
+++ b/tools/perf/util/thread.h
@@ -5,52 +5,66 @@
 #include <unistd.h>
 #include "symbol.h"
 
-struct thread {
-	struct rb_node		rb_node;
+struct map_groups {
 	struct rb_root		maps[MAP__NR_TYPES];
 	struct list_head	removed_maps[MAP__NR_TYPES];
+};
+
+struct thread {
+	struct rb_node		rb_node;
+	struct map_groups	mg;
 	pid_t			pid;
-	bool			use_modules;
 	char			shortname[3];
 	char			*comm;
 	int			comm_len;
 };
 
-void thread__init(struct thread *self, pid_t pid);
+void map_groups__init(struct map_groups *self);
 int thread__set_comm(struct thread *self, const char *comm);
 int thread__comm_len(struct thread *self);
-struct thread *threads__findnew(pid_t pid);
-struct thread *register_idle_thread(void);
+struct thread *perf_session__findnew(struct perf_session *self, pid_t pid);
 void thread__insert_map(struct thread *self, struct map *map);
 int thread__fork(struct thread *self, struct thread *parent);
-size_t thread__fprintf_maps(struct thread *self, FILE *fp);
-size_t threads__fprintf(FILE *fp);
+size_t map_groups__fprintf_maps(struct map_groups *self, FILE *fp);
+size_t perf_session__fprintf(struct perf_session *self, FILE *fp);
 
 void maps__insert(struct rb_root *maps, struct map *map);
 struct map *maps__find(struct rb_root *maps, u64 addr);
 
-static inline struct map *thread__find_map(struct thread *self,
+static inline void map_groups__insert(struct map_groups *self, struct map *map)
+{
+	 maps__insert(&self->maps[map->type], map);
+}
+
+static inline struct map *map_groups__find(struct map_groups *self,
 					   enum map_type type, u64 addr)
 {
-	return self ? maps__find(&self->maps[type], addr) : NULL;
+	return maps__find(&self->maps[type], addr);
 }
 
-static inline void __thread__insert_map(struct thread *self, struct map *map)
+static inline struct map *thread__find_map(struct thread *self,
+					   enum map_type type, u64 addr)
 {
-	 maps__insert(&self->maps[map->type], map);
+	return self ? map_groups__find(&self->mg, type, addr) : NULL;
 }
 
-void thread__find_addr_location(struct thread *self, u8 cpumode,
+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 *thread__find_symbol(struct thread *self,
-				   enum map_type type, u64 addr,
-				   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 *
-thread__find_function(struct thread *self, u64 addr, symbol_filter_t filter)
+map_groups__find_function(struct map_groups *self, struct perf_session *session,
+			  u64 addr, symbol_filter_t filter)
 {
-	return thread__find_symbol(self, MAP__FUNCTION, addr, filter);
+	return map_groups__find_symbol(self, session, MAP__FUNCTION, addr, filter);
 }
+
+struct map *map_groups__find_by_name(struct map_groups *self,
+				     enum map_type type, const char *name);
 #endif	/* __PERF_THREAD_H */
diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c
index 0302405aa2c..c5c32be040b 100644
--- a/tools/perf/util/trace-event-parse.c
+++ b/tools/perf/util/trace-event-parse.c
@@ -177,7 +177,7 @@ void parse_proc_kallsyms(char *file, unsigned int size __unused)
 		func_count++;
 	}
 
-	func_list = malloc_or_die(sizeof(*func_list) * func_count + 1);
+	func_list = malloc_or_die(sizeof(*func_list) * (func_count + 1));
 
 	i = 0;
 	while (list) {
@@ -1477,7 +1477,7 @@ process_fields(struct event *event, struct print_flag_sym **list, char **tok)
 			goto out_free;
 
 		field = malloc_or_die(sizeof(*field));
-		memset(field, 0, sizeof(field));
+		memset(field, 0, sizeof(*field));
 
 		value = arg_eval(arg);
 		field->value = strdup(value);
diff --git a/tools/perf/util/trace-event-perl.c b/tools/perf/util/trace-event-perl.c
index 51e833fd58c..6d6d76b8a21 100644
--- a/tools/perf/util/trace-event-perl.c
+++ b/tools/perf/util/trace-event-perl.c
@@ -32,9 +32,6 @@
 
 void xs_init(pTHX);
 
-void boot_Perf__Trace__Context(pTHX_ CV *cv);
-void boot_DynaLoader(pTHX_ CV *cv);
-
 void xs_init(pTHX)
 {
 	const char *file = __FILE__;
@@ -270,7 +267,7 @@ int common_lock_depth(struct scripting_context *context)
 }
 
 static void perl_process_event(int cpu, void *data,
-			       int size __attribute((unused)),
+			       int size __unused,
 			       unsigned long long nsecs, char *comm)
 {
 	struct format_field *field;
@@ -362,28 +359,46 @@ static void run_start_sub(void)
 /*
  * Start trace script
  */
-static int perl_start_script(const char *script)
+static int perl_start_script(const char *script, int argc, const char **argv)
 {
-	const char *command_line[2] = { "", NULL };
+	const char **command_line;
+	int i, err = 0;
 
+	command_line = malloc((argc + 2) * sizeof(const char *));
+	command_line[0] = "";
 	command_line[1] = script;
+	for (i = 2; i < argc + 2; i++)
+		command_line[i] = argv[i - 2];
 
 	my_perl = perl_alloc();
 	perl_construct(my_perl);
 
-	if (perl_parse(my_perl, xs_init, 2, (char **)command_line,
-		       (char **)NULL))
-		return -1;
+	if (perl_parse(my_perl, xs_init, argc + 2, (char **)command_line,
+		       (char **)NULL)) {
+		err = -1;
+		goto error;
+	}
 
-	perl_run(my_perl);
-	if (SvTRUE(ERRSV))
-		return -1;
+	if (perl_run(my_perl)) {
+		err = -1;
+		goto error;
+	}
+
+	if (SvTRUE(ERRSV)) {
+		err = -1;
+		goto error;
+	}
 
 	run_start_sub();
 
+	free(command_line);
 	fprintf(stderr, "perf trace started with Perl script %s\n\n", script);
-
 	return 0;
+error:
+	perl_free(my_perl);
+	free(command_line);
+
+	return err;
 }
 
 /*
@@ -573,26 +588,74 @@ struct scripting_ops perl_scripting_ops = {
 	.generate_script = perl_generate_script,
 };
 
-#ifdef NO_LIBPERL
-void setup_perl_scripting(void)
+static void print_unsupported_msg(void)
 {
 	fprintf(stderr, "Perl scripting not supported."
-		"  Install libperl and rebuild perf to enable it.  e.g. "
-		"apt-get install libperl-dev (ubuntu), yum install "
-		"perl-ExtUtils-Embed (Fedora), etc.\n");
+		"  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");
 }
-#else
-void setup_perl_scripting(void)
+
+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", &perl_scripting_ops);
+	err = script_spec_register("Perl", scripting_ops);
 	if (err)
 		die("error registering Perl script extension");
 
-	err = script_spec_register("pl", &perl_scripting_ops);
+	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/trace-event-perl.h b/tools/perf/util/trace-event-perl.h
index 8fe0d866fe1..e88fb26137b 100644
--- a/tools/perf/util/trace-event-perl.h
+++ b/tools/perf/util/trace-event-perl.h
@@ -34,9 +34,13 @@ typedef int INTERP;
 #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
 
diff --git a/tools/perf/util/trace-event-read.c b/tools/perf/util/trace-event-read.c
index 342dfdd43f8..1744422cafc 100644
--- a/tools/perf/util/trace-event-read.c
+++ b/tools/perf/util/trace-event-read.c
@@ -145,8 +145,9 @@ static void read_proc_kallsyms(void)
 	if (!size)
 		return;
 
-	buf = malloc_or_die(size);
+	buf = malloc_or_die(size + 1);
 	read_or_die(buf, size);
+	buf[size] = '\0';
 
 	parse_proc_kallsyms(buf, size);
 
diff --git a/tools/perf/util/trace-event.h b/tools/perf/util/trace-event.h
index 81698d5e650..6ad405620c9 100644
--- a/tools/perf/util/trace-event.h
+++ b/tools/perf/util/trace-event.h
@@ -270,7 +270,7 @@ enum trace_flag_type {
 
 struct scripting_ops {
 	const char *name;
-	int (*start_script) (const char *);
+	int (*start_script) (const char *script, int argc, const char **argv);
 	int (*stop_script) (void);
 	void (*process_event) (int cpu, void *data, int size,
 			       unsigned long long nsecs, char *comm);