Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/torvalds/linux-2.6 into common/serial-rework

Conflicts:
	arch/sh/kernel/cpu/sh2/setup-sh7619.c
	arch/sh/kernel/cpu/sh2a/setup-mxg.c
	arch/sh/kernel/cpu/sh2a/setup-sh7201.c
	arch/sh/kernel/cpu/sh2a/setup-sh7203.c
	arch/sh/kernel/cpu/sh2a/setup-sh7206.c
	arch/sh/kernel/cpu/sh3/setup-sh7705.c
	arch/sh/kernel/cpu/sh3/setup-sh770x.c
	arch/sh/kernel/cpu/sh3/setup-sh7710.c
	arch/sh/kernel/cpu/sh3/setup-sh7720.c
	arch/sh/kernel/cpu/sh4/setup-sh4-202.c
	arch/sh/kernel/cpu/sh4/setup-sh7750.c
	arch/sh/kernel/cpu/sh4/setup-sh7760.c
	arch/sh/kernel/cpu/sh4a/setup-sh7343.c
	arch/sh/kernel/cpu/sh4a/setup-sh7366.c
	arch/sh/kernel/cpu/sh4a/setup-sh7722.c
	arch/sh/kernel/cpu/sh4a/setup-sh7723.c
	arch/sh/kernel/cpu/sh4a/setup-sh7724.c
	arch/sh/kernel/cpu/sh4a/setup-sh7763.c
	arch/sh/kernel/cpu/sh4a/setup-sh7770.c
	arch/sh/kernel/cpu/sh4a/setup-sh7780.c
	arch/sh/kernel/cpu/sh4a/setup-sh7785.c
	arch/sh/kernel/cpu/sh4a/setup-sh7786.c
	arch/sh/kernel/cpu/sh4a/setup-shx3.c
	arch/sh/kernel/cpu/sh5/setup-sh5.c
	drivers/serial/sh-sci.c
	drivers/serial/sh-sci.h
	include/linux/serial_sci.h

254 files changed, 44544 insertions, 7661 deletions

diff --git a/tools/perf/.gitignore b/tools/perf/.gitignore
index d69a759a104..cb43289e447 100644
--- a/tools/perf/.gitignore
+++ b/tools/perf/.gitignore
@@ -10,7 +10,13 @@ perf-stat
 perf-top
 perf*.1
 perf*.xml
+perf*.html
 common-cmds.h
+perf.data
+perf.data.old
+perf-archive
 tags
 TAGS
 cscope*
+config.mak
+config.mak.autogen
diff --git a/tools/perf/CREDITS b/tools/perf/CREDITS
new file mode 100644
index 00000000000..c2ddcb3acbd
--- /dev/null
+++ b/tools/perf/CREDITS
@@ -0,0 +1,30 @@
+Most of the infrastructure that 'perf' uses here has been reused
+from the Git project, as of version:
+
+    66996ec: Sync with 1.6.2.4
+
+Here is an (incomplete!) list of main contributors to those files
+in util/* and elsewhere:
+
+ Alex Riesen
+ Christian Couder
+ Dmitry Potapov
+ Jeff King
+ Johannes Schindelin
+ Johannes Sixt
+ Junio C Hamano
+ Linus Torvalds
+ Matthias Kestenholz
+ Michal Ostrowski
+ Miklos Vajna
+ Petr Baudis
+ Pierre Habouzit
+ René Scharfe
+ Samuel Tardieu
+ Shawn O. Pearce
+ Steffen Prohaska
+ Steve Haslam
+
+Thanks guys!
+
+The full history of the files can be found in the upstream Git commits.
diff --git a/tools/perf/Documentation/Makefile b/tools/perf/Documentation/Makefile
index 5457192e1b4..bd498d49695 100644
--- a/tools/perf/Documentation/Makefile
+++ b/tools/perf/Documentation/Makefile
@@ -24,7 +24,10 @@ DOC_MAN1=$(patsubst %.txt,%.1,$(MAN1_TXT))
 DOC_MAN5=$(patsubst %.txt,%.5,$(MAN5_TXT))
 DOC_MAN7=$(patsubst %.txt,%.7,$(MAN7_TXT))
 
+# Make the path relative to DESTDIR, not prefix
+ifndef DESTDIR
 prefix?=$(HOME)
+endif
 bindir?=$(prefix)/bin
 htmldir?=$(prefix)/share/doc/perf-doc
 pdfdir?=$(prefix)/share/doc/perf-doc
@@ -32,10 +35,9 @@ mandir?=$(prefix)/share/man
 man1dir=$(mandir)/man1
 man5dir=$(mandir)/man5
 man7dir=$(mandir)/man7
-# DESTDIR=
 
 ASCIIDOC=asciidoc
-ASCIIDOC_EXTRA =
+ASCIIDOC_EXTRA = --unsafe
 MANPAGE_XSL = manpage-normal.xsl
 XMLTO_EXTRA =
 INSTALL?=install
diff --git a/tools/perf/Documentation/examples.txt b/tools/perf/Documentation/examples.txt
new file mode 100644
index 00000000000..8eb6c489fb1
--- /dev/null
+++ b/tools/perf/Documentation/examples.txt
@@ -0,0 +1,225 @@
+
+                ------------------------------
+                ****** perf by examples ******
+                ------------------------------
+
+[ From an e-mail by Ingo Molnar, http://lkml.org/lkml/2009/8/4/346 ]
+
+
+First, discovery/enumeration of available counters can be done via
+'perf list':
+
+titan:~> perf list
+  [...]
+  kmem:kmalloc                             [Tracepoint event]
+  kmem:kmem_cache_alloc                    [Tracepoint event]
+  kmem:kmalloc_node                        [Tracepoint event]
+  kmem:kmem_cache_alloc_node               [Tracepoint event]
+  kmem:kfree                               [Tracepoint event]
+  kmem:kmem_cache_free                     [Tracepoint event]
+  kmem:mm_page_free_direct                 [Tracepoint event]
+  kmem:mm_pagevec_free                     [Tracepoint event]
+  kmem:mm_page_alloc                       [Tracepoint event]
+  kmem:mm_page_alloc_zone_locked           [Tracepoint event]
+  kmem:mm_page_pcpu_drain                  [Tracepoint event]
+  kmem:mm_page_alloc_extfrag               [Tracepoint event]
+
+Then any (or all) of the above event sources can be activated and
+measured. For example the page alloc/free properties of a 'hackbench
+run' are:
+
+ titan:~> perf stat -e kmem:mm_page_pcpu_drain -e kmem:mm_page_alloc
+ -e kmem:mm_pagevec_free -e kmem:mm_page_free_direct ./hackbench 10
+ Time: 0.575
+
+ Performance counter stats for './hackbench 10':
+
+          13857  kmem:mm_page_pcpu_drain
+          27576  kmem:mm_page_alloc
+           6025  kmem:mm_pagevec_free
+          20934  kmem:mm_page_free_direct
+
+    0.613972165  seconds time elapsed
+
+You can observe the statistical properties as well, by using the
+'repeat the workload N times' feature of perf stat:
+
+ titan:~> perf stat --repeat 5 -e kmem:mm_page_pcpu_drain -e
+   kmem:mm_page_alloc -e kmem:mm_pagevec_free -e
+   kmem:mm_page_free_direct ./hackbench 10
+ Time: 0.627
+ Time: 0.644
+ Time: 0.564
+ Time: 0.559
+ Time: 0.626
+
+ Performance counter stats for './hackbench 10' (5 runs):
+
+          12920  kmem:mm_page_pcpu_drain    ( +-   3.359% )
+          25035  kmem:mm_page_alloc         ( +-   3.783% )
+           6104  kmem:mm_pagevec_free       ( +-   0.934% )
+          18376  kmem:mm_page_free_direct   ( +-   4.941% )
+
+    0.643954516  seconds time elapsed   ( +-   2.363% )
+
+Furthermore, these tracepoints can be used to sample the workload as
+well. For example the page allocations done by a 'git gc' can be
+captured the following way:
+
+ titan:~/git> perf record -f -e kmem:mm_page_alloc -c 1 ./git gc
+ Counting objects: 1148, done.
+ Delta compression using up to 2 threads.
+ Compressing objects: 100% (450/450), done.
+ Writing objects: 100% (1148/1148), done.
+ Total 1148 (delta 690), reused 1148 (delta 690)
+ [ perf record: Captured and wrote 0.267 MB perf.data (~11679 samples) ]
+
+To check which functions generated page allocations:
+
+ titan:~/git> perf report
+ # Samples: 10646
+ #
+ # Overhead          Command               Shared Object
+ # ........  ...............  ..........................
+ #
+    23.57%       git-repack  /lib64/libc-2.5.so
+    21.81%              git  /lib64/libc-2.5.so
+    14.59%              git  ./git
+    11.79%       git-repack  ./git
+     7.12%              git  /lib64/ld-2.5.so
+     3.16%       git-repack  /lib64/libpthread-2.5.so
+     2.09%       git-repack  /bin/bash
+     1.97%               rm  /lib64/libc-2.5.so
+     1.39%               mv  /lib64/ld-2.5.so
+     1.37%               mv  /lib64/libc-2.5.so
+     1.12%       git-repack  /lib64/ld-2.5.so
+     0.95%               rm  /lib64/ld-2.5.so
+     0.90%  git-update-serv  /lib64/libc-2.5.so
+     0.73%  git-update-serv  /lib64/ld-2.5.so
+     0.68%             perf  /lib64/libpthread-2.5.so
+     0.64%       git-repack  /usr/lib64/libz.so.1.2.3
+
+Or to see it on a more finegrained level:
+
+titan:~/git> perf report --sort comm,dso,symbol
+# Samples: 10646
+#
+# Overhead          Command               Shared Object  Symbol
+# ........  ...............  ..........................  ......
+#
+     9.35%       git-repack  ./git                       [.] insert_obj_hash
+     9.12%              git  ./git                       [.] insert_obj_hash
+     7.31%              git  /lib64/libc-2.5.so          [.] memcpy
+     6.34%       git-repack  /lib64/libc-2.5.so          [.] _int_malloc
+     6.24%       git-repack  /lib64/libc-2.5.so          [.] memcpy
+     5.82%       git-repack  /lib64/libc-2.5.so          [.] __GI___fork
+     5.47%              git  /lib64/libc-2.5.so          [.] _int_malloc
+     2.99%              git  /lib64/libc-2.5.so          [.] memset
+
+Furthermore, call-graph sampling can be done too, of page
+allocations - to see precisely what kind of page allocations there
+are:
+
+ titan:~/git> perf record -f -g -e kmem:mm_page_alloc -c 1 ./git gc
+ Counting objects: 1148, done.
+ Delta compression using up to 2 threads.
+ Compressing objects: 100% (450/450), done.
+ Writing objects: 100% (1148/1148), done.
+ Total 1148 (delta 690), reused 1148 (delta 690)
+ [ perf record: Captured and wrote 0.963 MB perf.data (~42069 samples) ]
+
+ titan:~/git> perf report -g
+ # Samples: 10686
+ #
+ # Overhead          Command               Shared Object
+ # ........  ...............  ..........................
+ #
+    23.25%       git-repack  /lib64/libc-2.5.so
+                |
+                |--50.00%-- _int_free
+                |
+                |--37.50%-- __GI___fork
+                |          make_child
+                |
+                |--12.50%-- ptmalloc_unlock_all2
+                |          make_child
+                |
+                 --6.25%-- __GI_strcpy
+    21.61%              git  /lib64/libc-2.5.so
+                |
+                |--30.00%-- __GI_read
+                |          |
+                |           --83.33%-- git_config_from_file
+                |                     git_config
+                |                     |
+   [...]
+
+Or you can observe the whole system's page allocations for 10
+seconds:
+
+titan:~/git> perf stat -a -e kmem:mm_page_pcpu_drain -e
+kmem:mm_page_alloc -e kmem:mm_pagevec_free -e
+kmem:mm_page_free_direct sleep 10
+
+ Performance counter stats for 'sleep 10':
+
+         171585  kmem:mm_page_pcpu_drain
+         322114  kmem:mm_page_alloc
+          73623  kmem:mm_pagevec_free
+         254115  kmem:mm_page_free_direct
+
+   10.000591410  seconds time elapsed
+
+Or observe how fluctuating the page allocations are, via statistical
+analysis done over ten 1-second intervals:
+
+ titan:~/git> perf stat --repeat 10 -a -e kmem:mm_page_pcpu_drain -e
+   kmem:mm_page_alloc -e kmem:mm_pagevec_free -e
+   kmem:mm_page_free_direct sleep 1
+
+ Performance counter stats for 'sleep 1' (10 runs):
+
+          17254  kmem:mm_page_pcpu_drain    ( +-   3.709% )
+          34394  kmem:mm_page_alloc         ( +-   4.617% )
+           7509  kmem:mm_pagevec_free       ( +-   4.820% )
+          25653  kmem:mm_page_free_direct   ( +-   3.672% )
+
+    1.058135029  seconds time elapsed   ( +-   3.089% )
+
+Or you can annotate the recorded 'git gc' run on a per symbol basis
+and check which instructions/source-code generated page allocations:
+
+ titan:~/git> perf annotate __GI___fork
+ ------------------------------------------------
+  Percent |      Source code & Disassembly of libc-2.5.so
+ ------------------------------------------------
+          :
+          :
+          :      Disassembly of section .plt:
+          :      Disassembly of section .text:
+          :
+          :      00000031a2e95560 <__fork>:
+ [...]
+     0.00 :        31a2e95602:   b8 38 00 00 00          mov    $0x38,%eax
+     0.00 :        31a2e95607:   0f 05                   syscall
+    83.42 :        31a2e95609:   48 3d 00 f0 ff ff       cmp    $0xfffffffffffff000,%rax
+     0.00 :        31a2e9560f:   0f 87 4d 01 00 00       ja     31a2e95762 <__fork+0x202>
+     0.00 :        31a2e95615:   85 c0                   test   %eax,%eax
+
+( this shows that 83.42% of __GI___fork's page allocations come from
+  the 0x38 system call it performs. )
+
+etc. etc. - a lot more is possible. I could list a dozen of
+other different usecases straight away - neither of which is
+possible via /proc/vmstat.
+
+/proc/vmstat is not in the same league really, in terms of
+expressive power of system analysis and performance
+analysis.
+
+All that the above results needed were those new tracepoints
+in include/tracing/events/kmem.h.
+
+        Ingo
+
+
diff --git a/tools/perf/Documentation/perf-annotate.txt b/tools/perf/Documentation/perf-annotate.txt
index c9dcade0683..6f5a498608b 100644
--- a/tools/perf/Documentation/perf-annotate.txt
+++ b/tools/perf/Documentation/perf-annotate.txt
@@ -1,5 +1,5 @@
 perf-annotate(1)
-==============
+================
 
 NAME
 ----
@@ -8,7 +8,7 @@ perf-annotate - Read perf.data (created by perf record) and display annotated co
 SYNOPSIS
 --------
 [verse]
-'perf annotate' [-i <file> | --input=file] symbol_name
+'perf annotate' [-i <file> | --input=file] [symbol_name]
 
 DESCRIPTION
 -----------
@@ -24,6 +24,48 @@ OPTIONS
 --input=::
         Input file name. (default: perf.data)
 
+-d::
+--dsos=<dso[,dso...]>::
+        Only consider symbols in these dsos.
+-s::
+--symbol=<symbol>::
+        Symbol to annotate.
+
+-f::
+--force::
+        Don't complain, do it.
+
+-v::
+--verbose::
+        Be more verbose. (Show symbol address, etc)
+
+-D::
+--dump-raw-trace::
+        Dump raw trace in ASCII.
+
+-k::
+--vmlinux=<file>::
+        vmlinux pathname.
+
+-m::
+--modules::
+        Load module symbols. WARNING: use only with -k and LIVE kernel.
+
+-l::
+--print-line::
+        Print matching source lines (may be slow).
+
+-P::
+--full-paths::
+        Don't shorten the displayed pathnames.
+
+--stdio:: Use the stdio interface.
+
+--tui:: Use the TUI interface Use of --tui requires a tty, if one is not
+        present, as when piping to other commands, the stdio interface is
+        used. This interfaces starts by centering on the line with more
+        samples, TAB/UNTAB cycles through the lines with more samples.
+
 SEE ALSO
 --------
-linkperf:perf-record[1]
+linkperf:perf-record[1], linkperf:perf-report[1]
diff --git a/tools/perf/Documentation/perf-archive.txt b/tools/perf/Documentation/perf-archive.txt
new file mode 100644
index 00000000000..fae174dc7d0
--- /dev/null
+++ b/tools/perf/Documentation/perf-archive.txt
@@ -0,0 +1,22 @@
+perf-archive(1)
+===============
+
+NAME
+----
+perf-archive - Create archive with object files with build-ids found in perf.data file
+
+SYNOPSIS
+--------
+[verse]
+'perf archive' [file]
+
+DESCRIPTION
+-----------
+This command runs runs perf-buildid-list --with-hits, and collects the files
+with the buildids found so that analisys of perf.data contents can be possible
+on another machine.
+
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-buildid-list[1], linkperf:perf-report[1]
diff --git a/tools/perf/Documentation/perf-bench.txt b/tools/perf/Documentation/perf-bench.txt
new file mode 100644
index 00000000000..a3dbadb26ef
--- /dev/null
+++ b/tools/perf/Documentation/perf-bench.txt
@@ -0,0 +1,120 @@
+perf-bench(1)
+=============
+
+NAME
+----
+perf-bench - General framework for benchmark suites
+
+SYNOPSIS
+--------
+[verse]
+'perf bench' [<common options>] <subsystem> <suite> [<options>]
+
+DESCRIPTION
+-----------
+This 'perf bench' command is general framework for benchmark suites.
+
+COMMON OPTIONS
+--------------
+-f::
+--format=::
+Specify format style.
+Current available format styles are:
+
+'default'::
+Default style. This is mainly for human reading.
+---------------------
+% perf bench sched pipe                      # with no style specified
+(executing 1000000 pipe operations between two tasks)
+        Total time:5.855 sec
+                5.855061 usecs/op
+                170792 ops/sec
+---------------------
+
+'simple'::
+This simple style is friendly for automated
+processing by scripts.
+---------------------
+% perf bench --format=simple sched pipe      # specified simple
+5.988
+---------------------
+
+SUBSYSTEM
+---------
+
+'sched'::
+        Scheduler and IPC mechanisms.
+
+SUITES FOR 'sched'
+~~~~~~~~~~~~~~~~~~
+*messaging*::
+Suite for evaluating performance of scheduler and IPC mechanisms.
+Based on hackbench by Rusty Russell.
+
+Options of *pipe*
+^^^^^^^^^^^^^^^^^
+-p::
+--pipe::
+Use pipe() instead of socketpair()
+
+-t::
+--thread::
+Be multi thread instead of multi process
+
+-g::
+--group=::
+Specify number of groups
+
+-l::
+--loop=::
+Specify number of loops
+
+Example of *messaging*
+^^^^^^^^^^^^^^^^^^^^^^
+
+---------------------
+% perf bench sched messaging                 # run with default
+options (20 sender and receiver processes per group)
+(10 groups == 400 processes run)
+
+      Total time:0.308 sec
+
+% perf bench sched messaging -t -g 20        # be multi-thread, with 20 groups
+(20 sender and receiver threads per group)
+(20 groups == 800 threads run)
+
+      Total time:0.582 sec
+---------------------
+
+*pipe*::
+Suite for pipe() system call.
+Based on pipe-test-1m.c by Ingo Molnar.
+
+Options of *pipe*
+^^^^^^^^^^^^^^^^^
+-l::
+--loop=::
+Specify number of loops.
+
+Example of *pipe*
+^^^^^^^^^^^^^^^^^
+
+---------------------
+% perf bench sched pipe
+(executing 1000000 pipe operations between two tasks)
+
+        Total time:8.091 sec
+                8.091833 usecs/op
+                123581 ops/sec
+
+% perf bench sched pipe -l 1000              # loop 1000
+(executing 1000 pipe operations between two tasks)
+
+        Total time:0.016 sec
+                16.948000 usecs/op
+                59004 ops/sec
+---------------------
+
+SEE ALSO
+--------
+linkperf:perf[1]
diff --git a/tools/perf/Documentation/perf-buildid-cache.txt b/tools/perf/Documentation/perf-buildid-cache.txt
new file mode 100644
index 00000000000..c1057701a7d
--- /dev/null
+++ b/tools/perf/Documentation/perf-buildid-cache.txt
@@ -0,0 +1,33 @@
+perf-buildid-cache(1)
+=====================
+
+NAME
+----
+perf-buildid-cache - Manage build-id cache.
+
+SYNOPSIS
+--------
+[verse]
+'perf buildid-cache <options>'
+
+DESCRIPTION
+-----------
+This command manages the build-id cache. It can add and remove files to/from
+the cache. In the future it should as well purge older entries, set upper
+limits for the space used by the cache, etc.
+
+OPTIONS
+-------
+-a::
+--add=::
+        Add specified file to the cache.
+-r::
+--remove=::
+        Remove specified file from the cache.
+-v::
+--verbose::
+        Be more verbose.
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-report[1], linkperf:perf-buildid-list[1]
diff --git a/tools/perf/Documentation/perf-buildid-list.txt b/tools/perf/Documentation/perf-buildid-list.txt
new file mode 100644
index 00000000000..5eaac6f26d5
--- /dev/null
+++ b/tools/perf/Documentation/perf-buildid-list.txt
@@ -0,0 +1,37 @@
+perf-buildid-list(1)
+====================
+
+NAME
+----
+perf-buildid-list - List the buildids in a perf.data file
+
+SYNOPSIS
+--------
+[verse]
+'perf buildid-list <options>'
+
+DESCRIPTION
+-----------
+This command displays the buildids found in a perf.data file, so that other
+tools can be used to fetch packages with matching symbol tables for use by
+perf report.
+
+OPTIONS
+-------
+-H::
+--with-hits::
+        Show only DSOs with hits.
+-i::
+--input=::
+        Input file name. (default: perf.data)
+-f::
+--force::
+        Don't do ownership validation.
+-v::
+--verbose::
+        Be more verbose.
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-top[1],
+linkperf:perf-report[1]
diff --git a/tools/perf/Documentation/perf-diff.txt b/tools/perf/Documentation/perf-diff.txt
new file mode 100644
index 00000000000..74d7481ed7a
--- /dev/null
+++ b/tools/perf/Documentation/perf-diff.txt
@@ -0,0 +1,74 @@
+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
+-------
+-M::
+--displacement::
+        Show position displacement relative to baseline.
+
+-D::
+--dump-raw-trace::
+        Dump raw trace in ASCII.
+
+-m::
+--modules::
+        Load module symbols. WARNING: use only with -k and LIVE kernel
+
+-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 occurrences 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.
+
+-f::
+--force::
+       Don't complain, do it.
+
+--symfs=<directory>::
+        Look for files with symbols relative to this directory.
+
+SEE ALSO
+--------
+linkperf:perf-record[1]
diff --git a/tools/perf/Documentation/perf-inject.txt b/tools/perf/Documentation/perf-inject.txt
new file mode 100644
index 00000000000..025630d43cd
--- /dev/null
+++ b/tools/perf/Documentation/perf-inject.txt
@@ -0,0 +1,35 @@
+perf-inject(1)
+==============
+
+NAME
+----
+perf-inject - Filter to augment the events stream with additional information
+
+SYNOPSIS
+--------
+[verse]
+'perf inject <options>'
+
+DESCRIPTION
+-----------
+perf-inject reads a perf-record event stream and repipes it to stdout.  At any
+point the processing code can inject other events into the event stream - in
+this case build-ids (-b option) are read and injected as needed into the event
+stream.
+
+Build-ids are just the first user of perf-inject - potentially anything that
+needs userspace processing to augment the events stream with additional
+information could make use of this facility.
+
+OPTIONS
+-------
+-b::
+--build-ids=::
+        Inject build-ids into the output stream
+-v::
+--verbose::
+        Be more verbose.
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-report[1], linkperf:perf-archive[1]
diff --git a/tools/perf/Documentation/perf-kmem.txt b/tools/perf/Documentation/perf-kmem.txt
new file mode 100644
index 00000000000..a52fcde894c
--- /dev/null
+++ b/tools/perf/Documentation/perf-kmem.txt
@@ -0,0 +1,47 @@
+perf-kmem(1)
+============
+
+NAME
+----
+perf-kmem - Tool to trace/measure kernel memory(slab) properties
+
+SYNOPSIS
+--------
+[verse]
+'perf kmem' {record|stat} [<options>]
+
+DESCRIPTION
+-----------
+There are two variants of perf kmem:
+
+  'perf kmem record <command>' to record the kmem events
+  of an arbitrary workload.
+
+  'perf kmem stat' to report kernel memory statistics.
+
+OPTIONS
+-------
+-i <file>::
+--input=<file>::
+        Select the input file (default: perf.data)
+
+--caller::
+        Show per-callsite statistics
+
+--alloc::
+        Show per-allocation statistics
+
+-s <key[,key2...]>::
+--sort=<key[,key2...]>::
+        Sort the output (default: frag,hit,bytes)
+
+-l <num>::
+--line=<num>::
+        Print n lines only
+
+--raw-ip::
+        Print raw ip instead of symbol
+
+SEE ALSO
+--------
+linkperf:perf-record[1]
diff --git a/tools/perf/Documentation/perf-kvm.txt b/tools/perf/Documentation/perf-kvm.txt
new file mode 100644
index 00000000000..dd84cb2f0a8
--- /dev/null
+++ b/tools/perf/Documentation/perf-kvm.txt
@@ -0,0 +1,74 @@
+perf-kvm(1)
+===========
+
+NAME
+----
+perf-kvm - Tool to trace/measure kvm guest os
+
+SYNOPSIS
+--------
+[verse]
+'perf kvm' [--host] [--guest] [--guestmount=<path>
+        [--guestkallsyms=<path> --guestmodules=<path> | --guestvmlinux=<path>]]
+        {top|record|report|diff|buildid-list}
+'perf kvm' [--host] [--guest] [--guestkallsyms=<path> --guestmodules=<path>
+        | --guestvmlinux=<path>] {top|record|report|diff|buildid-list}
+
+DESCRIPTION
+-----------
+There are a couple of variants of perf kvm:
+
+  'perf kvm [options] top <command>' to generates and displays
+  a performance counter profile of guest os in realtime
+  of an arbitrary workload.
+
+  'perf kvm record <command>' to record the performance counter profile
+  of an arbitrary workload and save it into a perf data file. If both
+  --host and --guest are input, the perf data file name is perf.data.kvm.
+  If there is  no --host but --guest, the file name is perf.data.guest.
+  If there is no --guest but --host, the file name is perf.data.host.
+
+  'perf kvm report' to display the performance counter profile information
+  recorded via perf kvm record.
+
+  'perf kvm diff' to displays the performance difference amongst two perf.data
+  files captured via perf record.
+
+  'perf kvm buildid-list' to  display the buildids found in a perf data file,
+  so that other tools can be used to fetch packages with matching symbol tables
+  for use by perf report.
+
+OPTIONS
+-------
+-i::
+--input=::
+        Input file name.
+-o::
+--output::
+        Output file name.
+--host=::
+        Collect host side performance profile.
+--guest=::
+        Collect guest side performance profile.
+--guestmount=<path>::
+        Guest os root file system mount directory. Users mounts guest os
+        root directories under <path> by a specific filesystem access method,
+        typically, sshfs. For example, start 2 guest os. The one's pid is 8888
+        and the other's is 9999.
+        #mkdir ~/guestmount; cd ~/guestmount
+        #sshfs -o allow_other,direct_io -p 5551 localhost:/ 8888/
+        #sshfs -o allow_other,direct_io -p 5552 localhost:/ 9999/
+        #perf kvm --host --guest --guestmount=~/guestmount top
+--guestkallsyms=<path>::
+        Guest os /proc/kallsyms file copy. 'perf' kvm' reads it to get guest
+        kernel symbols. Users copy it out from guest os.
+--guestmodules=<path>::
+        Guest os /proc/modules file copy. 'perf' kvm' reads it to get guest
+        kernel module information. Users copy it out from guest os.
+--guestvmlinux=<path>::
+        Guest os kernel vmlinux.
+
+SEE ALSO
+--------
+linkperf:perf-top[1], linkperf:perf-record[1], linkperf:perf-report[1],
+linkperf:perf-diff[1], linkperf:perf-buildid-list[1]
diff --git a/tools/perf/Documentation/perf-list.txt b/tools/perf/Documentation/perf-list.txt
index 8290b942266..399751befee 100644
--- a/tools/perf/Documentation/perf-list.txt
+++ b/tools/perf/Documentation/perf-list.txt
@@ -15,6 +15,52 @@ DESCRIPTION
 This command displays the symbolic event types which can be selected in the
 various perf commands with the -e option.
 
+EVENT MODIFIERS
+---------------
+
+Events can optionally have a modifer by appending a colon and one or
+more modifiers.  Modifiers allow the user to restrict when events are
+counted with 'u' for user-space, 'k' for kernel, 'h' for hypervisor.
+
+The 'p' modifier can be used for specifying how precise the instruction
+address should be. The 'p' modifier is currently only implemented for
+Intel PEBS and can be specified multiple times:
+  0 - SAMPLE_IP can have arbitrary skid
+  1 - SAMPLE_IP must have constant skid
+  2 - SAMPLE_IP requested to have 0 skid
+  3 - SAMPLE_IP must have 0 skid
+
+The PEBS implementation now supports up to 2.
+
+RAW HARDWARE EVENT DESCRIPTOR
+-----------------------------
+Even when an event is not available in a symbolic form within perf right now,
+it can be encoded in a per processor specific way.
+
+For instance For x86 CPUs NNN represents the raw register encoding with the
+layout of IA32_PERFEVTSELx MSRs (see [Intel® 64 and IA-32 Architectures Software Developer's Manual Volume 3B: System Programming Guide] Figure 30-1 Layout
+of IA32_PERFEVTSELx MSRs) or AMD's PerfEvtSeln (see [AMD64 Architecture Programmer’s Manual Volume 2: System Programming], Page 344,
+Figure 13-7 Performance Event-Select Register (PerfEvtSeln)).
+
+Example:
+
+If the Intel docs for a QM720 Core i7 describe an event as:
+
+  Event  Umask  Event Mask
+  Num.   Value  Mnemonic    Description                        Comment
+
+  A8H      01H  LSD.UOPS    Counts the number of micro-ops     Use cmask=1 and
+                            delivered by loop stream detector  invert to count
+                                                               cycles
+
+raw encoding of 0x1A8 can be used:
+
+ perf stat -e r1a8 -a sleep 1
+ perf record -e r1a8 ...
+
+You should refer to the processor specific documentation for getting these
+details. Some of them are referenced in the SEE ALSO section below.
+
 OPTIONS
 -------
 None
@@ -22,4 +68,6 @@ None
 SEE ALSO
 --------
 linkperf:perf-stat[1], linkperf:perf-top[1],
-linkperf:perf-record[1]
+linkperf:perf-record[1],
+http://www.intel.com/Assets/PDF/manual/253669.pdf[Intel® 64 and IA-32 Architectures Software Developer's Manual Volume 3B: System Programming Guide],
+http://support.amd.com/us/Processor_TechDocs/24593.pdf[AMD64 Architecture Programmer’s Manual Volume 2: System Programming]
diff --git a/tools/perf/Documentation/perf-lock.txt b/tools/perf/Documentation/perf-lock.txt
new file mode 100644
index 00000000000..921de259ea1
--- /dev/null
+++ b/tools/perf/Documentation/perf-lock.txt
@@ -0,0 +1,44 @@
+perf-lock(1)
+============
+
+NAME
+----
+perf-lock - Analyze lock events
+
+SYNOPSIS
+--------
+[verse]
+'perf lock' {record|report|trace}
+
+DESCRIPTION
+-----------
+You can analyze various lock behaviours
+and statistics with this 'perf lock' command.
+
+  'perf lock record <command>' records lock events
+  between start and end <command>. And this command
+  produces the file "perf.data" which contains tracing
+  results of lock events.
+
+  'perf lock trace' shows raw lock events.
+
+  'perf lock report' reports statistical data.
+
+OPTIONS
+-------
+
+-i::
+--input=<file>::
+        Input file name.
+
+-v::
+--verbose::
+        Be more verbose (show symbol address, etc).
+
+-D::
+--dump-raw-trace::
+        Dump raw trace in ASCII.
+
+SEE ALSO
+--------
+linkperf:perf[1]
diff --git a/tools/perf/Documentation/perf-probe.txt b/tools/perf/Documentation/perf-probe.txt
new file mode 100644
index 00000000000..86b797a35aa
--- /dev/null
+++ b/tools/perf/Documentation/perf-probe.txt
@@ -0,0 +1,166 @@
+perf-probe(1)
+=============
+
+NAME
+----
+perf-probe - Define new dynamic tracepoints
+
+SYNOPSIS
+--------
+[verse]
+'perf probe' [options] --add='PROBE' [...]
+or
+'perf probe' [options] PROBE
+or
+'perf probe' [options] --del='[GROUP:]EVENT' [...]
+or
+'perf probe' --list
+or
+'perf probe' [options] --line='FUNC[:RLN[+NUM|:RLN2]]|SRC:ALN[+NUM|:ALN2]'
+or
+'perf probe' [options] --vars='PROBEPOINT'
+
+DESCRIPTION
+-----------
+This command defines dynamic tracepoint events, by symbol and registers
+without debuginfo, or by C expressions (C line numbers, C function names,
+and C local variables) with debuginfo.
+
+
+OPTIONS
+-------
+-k::
+--vmlinux=PATH::
+        Specify vmlinux path which has debuginfo (Dwarf binary).
+
+-m::
+--module=MODNAME::
+        Specify module name in which perf-probe searches probe points
+        or lines.
+
+-s::
+--source=PATH::
+        Specify path to kernel source.
+
+-v::
+--verbose::
+        Be more verbose (show parsed arguments, etc).
+
+-a::
+--add=::
+        Define a probe event (see PROBE SYNTAX for detail).
+
+-d::
+--del=::
+        Delete probe events. This accepts glob wildcards('*', '?') and character
+        classes(e.g. [a-z], [!A-Z]).
+
+-l::
+--list::
+        List up current probe events.
+
+-L::
+--line=::
+        Show source code lines which can be probed. This needs an argument
+        which specifies a range of the source code. (see LINE SYNTAX for detail)
+
+-V::
+--vars=::
+        Show available local variables at given probe point. The argument
+        syntax is same as PROBE SYNTAX, but NO ARGs.
+
+--externs::
+        (Only for --vars) Show external defined variables in addition to local
+        variables.
+
+-f::
+--force::
+        Forcibly add events with existing name.
+
+-n::
+--dry-run::
+        Dry run. With this option, --add and --del doesn't execute actual
+        adding and removal operations.
+
+--max-probes::
+        Set the maximum number of probe points for an event. Default is 128.
+
+PROBE SYNTAX
+------------
+Probe points are defined by following syntax.
+
+    1) Define event based on function name
+     [EVENT=]FUNC[@SRC][:RLN|+OFFS|%return|;PTN] [ARG ...]
+
+    2) Define event based on source file with line number
+     [EVENT=]SRC:ALN [ARG ...]
+
+    3) Define event based on source file with lazy pattern
+     [EVENT=]SRC;PTN [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. And ';PTN' means lazy matching pattern (see LAZY MATCHING). Note that ';PTN' must be the end of the probe point definition.  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 or lazy matching by using 'SRC:ALN' or 'SRC;PTN' syntax, where 'SRC' is the source file path, ':ALN' is the line number and ';PTN' is the lazy matching pattern.
+'ARG' specifies the arguments of this probe point, (see PROBE ARGUMENT).
+
+PROBE ARGUMENT
+--------------
+Each probe argument follows below syntax.
+
+ [NAME=]LOCALVAR|$retval|%REG|@SYMBOL[:TYPE]
+
+'NAME' specifies the name of this argument (optional). You can use the name of local variable, local data structure member (e.g. var->field, var.field2), local array with fixed index (e.g. array[1], var->array[0], var->pointer[2]), or kprobe-tracer argument format (e.g. $retval, %ax, etc). Note that the name of this argument will be set as the last member name if you specify a local data structure member (e.g. field2 for 'var->field1.field2'.)
+'TYPE' casts the type of this argument (optional). If omitted, perf probe automatically set the type based on debuginfo. You can specify 'string' type only for the local variable or structure member which is an array of or a pointer to 'char' or 'unsigned char' type.
+
+LINE SYNTAX
+-----------
+Line range is described by following syntax.
+
+ "FUNC[:RLN[+NUM|-RLN2]]|SRC[:ALN[+NUM|-ALN2]]"
+
+FUNC specifies the function name of showing lines. 'RLN' is the start line
+number from function entry line, and 'RLN2' is the end line number. As same as
+probe syntax, 'SRC' means the source file path, 'ALN' is start line number,
+and 'ALN2' is end line number in the file. It is also possible to specify how
+many lines to show by using 'NUM'.
+So, "source.c:100-120" shows lines between 100th to l20th in source.c file. And "func:10+20" shows 20 lines from 10th line of func function.
+
+LAZY MATCHING
+-------------
+ The lazy line matching is similar to glob matching but ignoring spaces in both of pattern and target. So this accepts wildcards('*', '?') and character classes(e.g. [a-z], [!A-Z]).
+
+e.g.
+ 'a=*' can matches 'a=b', 'a = b', 'a == b' and so on.
+
+This provides some sort of flexibility and robustness to probe point definitions against minor code changes. For example, actual 10th line of schedule() can be moved easily by modifying schedule(), but the same line matching 'rq=cpu_rq*' may still exist in the function.)
+
+
+EXAMPLES
+--------
+Display which lines in schedule() can be probed:
+
+ ./perf probe --line schedule
+
+Add a probe on schedule() function 12th line with recording cpu local variable:
+
+ ./perf probe schedule:12 cpu
+ or
+ ./perf probe --add='schedule:12 cpu'
+
+ this will add one or more probes which has the name start with "schedule".
+
+ Add probes on lines in schedule() function which calls update_rq_clock().
+
+ ./perf probe 'schedule;update_rq_clock*'
+ or
+ ./perf probe --add='schedule;update_rq_clock*'
+
+Delete all probes on schedule().
+
+ ./perf probe --del='schedule*'
+
+
+SEE ALSO
+--------
+linkperf:perf-trace[1], linkperf:perf-record[1]
diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
index 1dbc1eeb4c0..52462ae2645 100644
--- a/tools/perf/Documentation/perf-record.txt
+++ b/tools/perf/Documentation/perf-record.txt
@@ -26,16 +26,113 @@ OPTIONS
 
 -e::
 --event=::
-        Select the PMU event. Selection can be a symbolic event name
-        (use 'perf list' to list all events) or a raw PMU
-        event (eventsel+umask) in the form of rNNN where NNN is a
-         hexadecimal event descriptor.
+        Select the PMU event. Selection can be:
+
+        - a symbolic event name (use 'perf list' to list all events)
+
+        - a raw PMU event (eventsel+umask) in the form of rNNN where NNN is a
+          hexadecimal event descriptor.
+
+        - a hardware breakpoint event in the form of '\mem:addr[:access]'
+          where addr is the address in memory you want to break in.
+          Access is the memory access type (read, write, execute) it can
+          be passed as follows: '\mem:addr[:[r][w][x]]'.
+          If you want to profile read-write accesses in 0x1000, just set
+          'mem:0x1000:rw'.
+
+--filter=<filter>::
+        Event filter.
 
 -a::
-        system-wide collection
+--all-cpus::
+        System-wide collection from all CPUs.
 
 -l::
-        scale counter values
+        Scale counter values.
+
+-p::
+--pid=::
+        Record events on existing process ID.
+
+-t::
+--tid=::
+        Record events on existing thread ID.
+
+-r::
+--realtime=::
+        Collect data with this RT SCHED_FIFO priority.
+-A::
+--append::
+        Append to the output file to do incremental profiling.
+
+-f::
+--force::
+        Overwrite existing data file. (deprecated)
+
+-c::
+--count=::
+        Event period to sample.
+
+-o::
+--output=::
+        Output file name.
+
+-i::
+--no-inherit::
+        Child tasks do not inherit counters.
+-F::
+--freq=::
+        Profile at this frequency.
+
+-m::
+--mmap-pages=::
+        Number of mmap data pages.
+
+-g::
+--call-graph::
+        Do call-graph (stack chain/backtrace) recording.
+
+-q::
+--quiet::
+        Don't print any message, useful for scripting.
+
+-v::
+--verbose::
+        Be more verbose (show counter open errors, etc).
+
+-s::
+--stat::
+        Per thread counts.
+
+-d::
+--data::
+        Sample addresses.
+
+-T::
+--timestamp::
+        Sample timestamps. Use it with 'perf report -D' to see the timestamps,
+        for instance.
+
+-n::
+--no-samples::
+        Don't sample.
+
+-R::
+--raw-samples::
+Collect raw sample records from all opened counters (default for tracepoint counters).
+
+-C::
+--cpu::
+Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a
+comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
+In per-thread mode with inheritance mode on (default), samples are captured only when
+the thread executes on the designated CPUs. Default is to monitor all CPUs.
+
+-N::
+--no-buildid-cache::
+Do not update the builid cache. This saves some overhead in situations
+where the information in the perf.data file (which includes buildids)
+is sufficient.
 
 SEE ALSO
 --------
diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt
index 52d3fc6846a..8ba03d6e539 100644
--- a/tools/perf/Documentation/perf-report.txt
+++ b/tools/perf/Documentation/perf-report.txt
@@ -13,7 +13,7 @@ SYNOPSIS
 DESCRIPTION
 -----------
 This command displays the performance counter profile information recorded
-via perf report.
+via perf record.
 
 OPTIONS
 -------
@@ -21,6 +21,104 @@ OPTIONS
 --input=::
         Input file name. (default: perf.data)
 
+-v::
+--verbose::
+        Be more verbose. (show symbol address, etc)
+
+-d::
+--dsos=::
+        Only consider symbols in these dsos. CSV that understands
+        file://filename entries.
+-n::
+--show-nr-samples::
+        Show the number of samples for each symbol
+
+--showcpuutilization::
+        Show sample percentage for different cpu modes.
+
+-T::
+--threads::
+        Show per-thread event counters
+-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.
+
+-U::
+--hide-unresolved::
+        Only display entries resolved to a symbol.
+
+-s::
+--sort=::
+        Sort by key(s): pid, comm, dso, symbol, parent.
+
+-p::
+--parent=<regex>::
+        regex filter to identify parent, see: '--sort parent'
+
+-x::
+--exclude-other::
+        Only display entries with parent-match.
+
+-w::
+--column-widths=<width[,width...]>::
+        Force each column width to the provided list, for large terminal
+        readability.
+
+-t::
+--field-separator=::
+
+        Use a special separator character and don't pad with spaces, replacing
+        all occurrences of this separator in symbol names (and other output)
+        with a '.' character, that thus it's the only non valid separator.
+
+-D::
+--dump-raw-trace::
+        Dump raw trace in ASCII.
+
+-g [type,min]::
+--call-graph::
+        Display call chains using type and min percent threshold.
+        type can be either:
+        - flat: single column, linear exposure of call chains.
+        - graph: use a graph tree, displaying absolute overhead rates.
+        - fractal: like graph, but displays relative rates. Each branch of
+                 the tree is considered as a new profiled object. +
+        Default: fractal,0.5.
+
+--pretty=<key>::
+        Pretty printing style.  key: normal, raw
+
+--stdio:: Use the stdio interface.
+
+--tui:: Use the TUI interface, that is integrated with annotate and allows
+        zooming into DSOs or threads, among other features. Use of --tui
+        requires a tty, if one is not present, as when piping to other
+        commands, the stdio interface is used.
+
+-k::
+--vmlinux=<file>::
+        vmlinux pathname
+
+--kallsyms=<file>::
+        kallsyms pathname
+
+-m::
+--modules::
+        Load module symbols. WARNING: This should only be used with -k and
+        a LIVE kernel.
+
+-f::
+--force::
+        Don't complain, do it.
+
+--symfs=<directory>::
+        Look for files with symbols relative to this directory.
+
 SEE ALSO
 --------
 linkperf:perf-stat[1]
diff --git a/tools/perf/Documentation/perf-sched.txt b/tools/perf/Documentation/perf-sched.txt
new file mode 100644
index 00000000000..46822d5fde1
--- /dev/null
+++ b/tools/perf/Documentation/perf-sched.txt
@@ -0,0 +1,55 @@
+perf-sched(1)
+==============
+
+NAME
+----
+perf-sched - Tool to trace/measure scheduler properties (latencies)
+
+SYNOPSIS
+--------
+[verse]
+'perf sched' {record|latency|map|replay|trace}
+
+DESCRIPTION
+-----------
+There are five variants of perf sched:
+
+  'perf sched record <command>' to record the scheduling events
+  of an arbitrary workload.
+
+  'perf sched latency' to report the per task scheduling latencies
+  and other scheduling properties of the workload.
+
+  'perf sched trace' to see a detailed trace of the workload that
+  was recorded.
+
+  'perf sched replay' to simulate the workload that was recorded
+  via perf sched record. (this is done by starting up mockup threads
+  that mimic the workload based on the events in the trace. These
+  threads can then replay the timings (CPU runtime and sleep patterns)
+  of the workload as it occurred when it was recorded - and can repeat
+  it a number of times, measuring its performance.)
+
+  'perf sched map' to print a textual context-switching outline of
+  workload captured via perf sched record.  Columns stand for
+  individual CPUs, and the two-letter shortcuts stand for tasks that
+  are running on a CPU. A '*' denotes the CPU that had the event, and
+  a dot signals an idle CPU.
+
+OPTIONS
+-------
+-i::
+--input=<file>::
+        Input file name. (default: perf.data)
+
+-v::
+--verbose::
+        Be more verbose. (show symbol address, etc)
+
+-D::
+--dump-raw-trace=::
+        Display verbose dump of the sched data.
+
+SEE ALSO
+--------
+linkperf:perf-record[1]
diff --git a/tools/perf/Documentation/perf-script-perl.txt b/tools/perf/Documentation/perf-script-perl.txt
new file mode 100644
index 00000000000..5bb41e55a3a
--- /dev/null
+++ b/tools/perf/Documentation/perf-script-perl.txt
@@ -0,0 +1,217 @@
+perf-script-perl(1)
+==================
+
+NAME
+----
+perf-script-perl - Process trace data with a Perl script
+
+SYNOPSIS
+--------
+[verse]
+'perf script' [-s [Perl]:script[.pl] ]
+
+DESCRIPTION
+-----------
+
+This perf script option is used to process perf script data using perf's
+built-in Perl interpreter.  It reads and processes the input file and
+displays the results of the trace analysis implemented in the given
+Perl script, if any.
+
+STARTER SCRIPTS
+---------------
+
+You can avoid reading the rest of this document by running 'perf script
+-g perl' in the same directory as an existing perf.data trace file.
+That will generate a starter script containing a handler for each of
+the event types in the trace file; it simply prints every available
+field for each event in the trace file.
+
+You can also look at the existing scripts in
+~/libexec/perf-core/scripts/perl for typical examples showing how to
+do basic things like aggregate event data, print results, etc.  Also,
+the check-perf-script.pl script, while not interesting for its results,
+attempts to exercise all of the main scripting features.
+
+EVENT HANDLERS
+--------------
+
+When perf script is invoked using a trace script, a user-defined
+'handler function' is called for each event in the trace.  If there's
+no handler function defined for a given event type, the event is
+ignored (or passed to a 'trace_handled' function, see below) and the
+next event is processed.
+
+Most of the event's field values are passed as arguments to the
+handler function; some of the less common ones aren't - those are
+available as calls back into the perf executable (see below).
+
+As an example, the following perf record command can be used to record
+all sched_wakeup events in the system:
+
+ # perf record -a -e sched:sched_wakeup
+
+Traces meant to be processed using a script should be recorded with
+the above option: -a to enable system-wide collection.
+
+The format file for the sched_wakep event defines the following fields
+(see /sys/kernel/debug/tracing/events/sched/sched_wakeup/format):
+
+----
+ format:
+        field:unsigned short common_type;
+        field:unsigned char common_flags;
+        field:unsigned char common_preempt_count;
+        field:int common_pid;
+        field:int common_lock_depth;
+
+        field:char comm[TASK_COMM_LEN];
+        field:pid_t pid;
+        field:int prio;
+        field:int success;
+        field:int target_cpu;
+----
+
+The handler function for this event would be defined as:
+
+----
+sub sched::sched_wakeup
+{
+   my ($event_name, $context, $common_cpu, $common_secs,
+       $common_nsecs, $common_pid, $common_comm,
+       $comm, $pid, $prio, $success, $target_cpu) = @_;
+}
+----
+
+The handler function takes the form subsystem::event_name.
+
+The $common_* arguments in the handler's argument list are the set of
+arguments passed to all event handlers; some of the fields correspond
+to the common_* fields in the format file, but some are synthesized,
+and some of the common_* fields aren't common enough to to be passed
+to every event as arguments but are available as library functions.
+
+Here's a brief description of each of the invariant event args:
+
+ $event_name                the name of the event as text
+ $context                   an opaque 'cookie' used in calls back into perf
+ $common_cpu                the cpu the event occurred on
+ $common_secs               the secs portion of the event timestamp
+ $common_nsecs              the nsecs portion of the event timestamp
+ $common_pid                the pid of the current task
+ $common_comm               the name of the current process
+
+All of the remaining fields in the event's format file have
+counterparts as handler function arguments of the same name, as can be
+seen in the example above.
+
+The above provides the basics needed to directly access every field of
+every event in a trace, which covers 90% of what you need to know to
+write a useful trace script.  The sections below cover the rest.
+
+SCRIPT LAYOUT
+-------------
+
+Every perf script Perl script should start by setting up a Perl module
+search path and 'use'ing a few support modules (see module
+descriptions below):
+
+----
+ use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/perf-script-Util/lib";
+ use lib "./perf-script-Util/lib";
+ use Perf::Trace::Core;
+ use Perf::Trace::Context;
+ use Perf::Trace::Util;
+----
+
+The rest of the script can contain handler functions and support
+functions in any order.
+
+Aside from the event handler functions discussed above, every script
+can implement a set of optional functions:
+
+*trace_begin*, if defined, is called before any event is processed and
+gives scripts a chance to do setup tasks:
+
+----
+ sub trace_begin
+ {
+ }
+----
+
+*trace_end*, if defined, is called after all events have been
+ processed and gives scripts a chance to do end-of-script tasks, such
+ as display results:
+
+----
+sub trace_end
+{
+}
+----
+
+*trace_unhandled*, if defined, is called after for any event that
+ doesn't have a handler explicitly defined for it.  The standard set
+ of common arguments are passed into it:
+
+----
+sub trace_unhandled
+{
+    my ($event_name, $context, $common_cpu, $common_secs,
+        $common_nsecs, $common_pid, $common_comm) = @_;
+}
+----
+
+The remaining sections provide descriptions of each of the available
+built-in perf script Perl modules and their associated functions.
+
+AVAILABLE MODULES AND FUNCTIONS
+-------------------------------
+
+The following sections describe the functions and variables available
+via the various Perf::Trace::* Perl modules.  To use the functions and
+variables from the given module, add the corresponding 'use
+Perf::Trace::XXX' line to your perf script script.
+
+Perf::Trace::Core Module
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+These functions provide some essential functions to user scripts.
+
+The *flag_str* and *symbol_str* functions provide human-readable
+strings for flag and symbolic fields.  These correspond to the strings
+and values parsed from the 'print fmt' fields of the event format
+files:
+
+  flag_str($event_name, $field_name, $field_value) - returns the string represention corresponding to $field_value for the flag field $field_name of event $event_name
+  symbol_str($event_name, $field_name, $field_value) - returns the string represention corresponding to $field_value for the symbolic field $field_name of event $event_name
+
+Perf::Trace::Context Module
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Some of the 'common' fields in the event format file aren't all that
+common, but need to be made accessible to user scripts nonetheless.
+
+Perf::Trace::Context defines a set of functions that can be used to
+access this data in the context of the current event.  Each of these
+functions expects a $context variable, which is the same as the
+$context variable passed into every event handler as the second
+argument.
+
+ common_pc($context) - returns common_preempt count for the current event
+ common_flags($context) - returns common_flags for the current event
+ common_lock_depth($context) - returns common_lock_depth for the current event
+
+Perf::Trace::Util Module
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+Various utility functions for use with perf script:
+
+  nsecs($secs, $nsecs) - returns total nsecs given secs/nsecs pair
+  nsecs_secs($nsecs) - returns whole secs portion given nsecs
+  nsecs_nsecs($nsecs) - returns nsecs remainder given nsecs
+  nsecs_str($nsecs) - returns printable string in the form secs.nsecs
+  avg($total, $n) - returns average given a sum and a total number of values
+
+SEE ALSO
+--------
+linkperf:perf-script[1]
diff --git a/tools/perf/Documentation/perf-script-python.txt b/tools/perf/Documentation/perf-script-python.txt
new file mode 100644
index 00000000000..36b38277422
--- /dev/null
+++ b/tools/perf/Documentation/perf-script-python.txt
@@ -0,0 +1,623 @@
+perf-script-python(1)
+====================
+
+NAME
+----
+perf-script-python - Process trace data with a Python script
+
+SYNOPSIS
+--------
+[verse]
+'perf script' [-s [Python]:script[.py] ]
+
+DESCRIPTION
+-----------
+
+This perf script option is used to process perf script data using perf's
+built-in Python interpreter.  It reads and processes the input file and
+displays the results of the trace analysis implemented in the given
+Python script, if any.
+
+A QUICK EXAMPLE
+---------------
+
+This section shows the process, start to finish, of creating a working
+Python script that aggregates and extracts useful information from a
+raw perf script stream.  You can avoid reading the rest of this
+document if an example is enough for you; the rest of the document
+provides more details on each step and lists the library functions
+available to script writers.
+
+This example actually details the steps that were used to create the
+'syscall-counts' script you see when you list the available perf script
+scripts via 'perf script -l'.  As such, this script also shows how to
+integrate your script into the list of general-purpose 'perf script'
+scripts listed by that command.
+
+The syscall-counts script is a simple script, but demonstrates all the
+basic ideas necessary to create a useful script.  Here's an example
+of its output (syscall names are not yet supported, they will appear
+as numbers):
+
+----
+syscall events:
+
+event                                          count
+----------------------------------------  -----------
+sys_write                                     455067
+sys_getdents                                    4072
+sys_close                                       3037
+sys_swapoff                                     1769
+sys_read                                         923
+sys_sched_setparam                               826
+sys_open                                         331
+sys_newfstat                                     326
+sys_mmap                                         217
+sys_munmap                                       216
+sys_futex                                        141
+sys_select                                       102
+sys_poll                                          84
+sys_setitimer                                     12
+sys_writev                                         8
+15                                                 8
+sys_lseek                                          7
+sys_rt_sigprocmask                                 6
+sys_wait4                                          3
+sys_ioctl                                          3
+sys_set_robust_list                                1
+sys_exit                                           1
+56                                                 1
+sys_access                                         1
+----
+
+Basically our task is to keep a per-syscall tally that gets updated
+every time a system call occurs in the system.  Our script will do
+that, but first we need to record the data that will be processed by
+that script.  Theoretically, there are a couple of ways we could do
+that:
+
+- we could enable every event under the tracing/events/syscalls
+  directory, but this is over 600 syscalls, well beyond the number
+  allowable by perf.  These individual syscall events will however be
+  useful if we want to later use the guidance we get from the
+  general-purpose scripts to drill down and get more detail about
+  individual syscalls of interest.
+
+- we can enable the sys_enter and/or sys_exit syscalls found under
+  tracing/events/raw_syscalls.  These are called for all syscalls; the
+  'id' field can be used to distinguish between individual syscall
+  numbers.
+
+For this script, we only need to know that a syscall was entered; we
+don't care how it exited, so we'll use 'perf record' to record only
+the sys_enter events:
+
+----
+# perf record -a -e raw_syscalls:sys_enter
+
+^C[ perf record: Woken up 1 times to write data ]
+[ perf record: Captured and wrote 56.545 MB perf.data (~2470503 samples) ]
+----
+
+The options basically say to collect data for every syscall event
+system-wide and multiplex the per-cpu output into a single stream.
+That single stream will be recorded in a file in the current directory
+called perf.data.
+
+Once we have a perf.data file containing our data, we can use the -g
+'perf script' option to generate a Python script that will contain a
+callback handler for each event type found in the perf.data trace
+stream (for more details, see the STARTER SCRIPTS section).
+
+----
+# perf script -g python
+generated Python script: perf-script.py
+
+The output file created also in the current directory is named
+perf-script.py.  Here's the file in its entirety:
+
+# perf script event handlers, generated by perf script -g python
+# Licensed under the terms of the GNU GPL License version 2
+
+# The common_* event handler fields are the most useful fields common to
+# all events.  They don't necessarily correspond to the 'common_*' fields
+# in the format files.  Those fields not available as handler params can
+# be retrieved using Python functions of the form common_*(context).
+# See the perf-script-python Documentation for the list of available functions.
+
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+        '/scripts/python/perf-script-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+
+def trace_begin():
+        print "in trace_begin"
+
+def trace_end():
+        print "in trace_end"
+
+def raw_syscalls__sys_enter(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        id, args):
+                print_header(event_name, common_cpu, common_secs, common_nsecs,
+                        common_pid, common_comm)
+
+                print "id=%d, args=%s\n" % \
+                (id, args),
+
+def trace_unhandled(event_name, context, common_cpu, common_secs, common_nsecs,
+                common_pid, common_comm):
+                print_header(event_name, common_cpu, common_secs, common_nsecs,
+                common_pid, common_comm)
+
+def print_header(event_name, cpu, secs, nsecs, pid, comm):
+        print "%-20s %5u %05u.%09u %8u %-20s " % \
+        (event_name, cpu, secs, nsecs, pid, comm),
+----
+
+At the top is a comment block followed by some import statements and a
+path append which every perf script script should include.
+
+Following that are a couple generated functions, trace_begin() and
+trace_end(), which are called at the beginning and the end of the
+script respectively (for more details, see the SCRIPT_LAYOUT section
+below).
+
+Following those are the 'event handler' functions generated one for
+every event in the 'perf record' output.  The handler functions take
+the form subsystem__event_name, and contain named parameters, one for
+each field in the event; in this case, there's only one event,
+raw_syscalls__sys_enter().  (see the EVENT HANDLERS section below for
+more info on event handlers).
+
+The final couple of functions are, like the begin and end functions,
+generated for every script.  The first, trace_unhandled(), is called
+every time the script finds an event in the perf.data file that
+doesn't correspond to any event handler in the script.  This could
+mean either that the record step recorded event types that it wasn't
+really interested in, or the script was run against a trace file that
+doesn't correspond to the script.
+
+The script generated by -g option simply prints a line for each
+event found in the trace stream i.e. it basically just dumps the event
+and its parameter values to stdout.  The print_header() function is
+simply a utility function used for that purpose.  Let's rename the
+script and run it to see the default output:
+
+----
+# mv perf-script.py syscall-counts.py
+# perf script -s syscall-counts.py
+
+raw_syscalls__sys_enter     1 00840.847582083     7506 perf                  id=1, args=
+raw_syscalls__sys_enter     1 00840.847595764     7506 perf                  id=1, args=
+raw_syscalls__sys_enter     1 00840.847620860     7506 perf                  id=1, args=
+raw_syscalls__sys_enter     1 00840.847710478     6533 npviewer.bin          id=78, args=
+raw_syscalls__sys_enter     1 00840.847719204     6533 npviewer.bin          id=142, args=
+raw_syscalls__sys_enter     1 00840.847755445     6533 npviewer.bin          id=3, args=
+raw_syscalls__sys_enter     1 00840.847775601     6533 npviewer.bin          id=3, args=
+raw_syscalls__sys_enter     1 00840.847781820     6533 npviewer.bin          id=3, args=
+.
+.
+.
+----
+
+Of course, for this script, we're not interested in printing every
+trace event, but rather aggregating it in a useful way.  So we'll get
+rid of everything to do with printing as well as the trace_begin() and
+trace_unhandled() functions, which we won't be using.  That leaves us
+with this minimalistic skeleton:
+
+----
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+        '/scripts/python/perf-script-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+
+def trace_end():
+        print "in trace_end"
+
+def raw_syscalls__sys_enter(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        id, args):
+----
+
+In trace_end(), we'll simply print the results, but first we need to
+generate some results to print.  To do that we need to have our
+sys_enter() handler do the necessary tallying until all events have
+been counted.  A hash table indexed by syscall id is a good way to
+store that information; every time the sys_enter() handler is called,
+we simply increment a count associated with that hash entry indexed by
+that syscall id:
+
+----
+  syscalls = autodict()
+
+  try:
+    syscalls[id] += 1
+  except TypeError:
+    syscalls[id] = 1
+----
+
+The syscalls 'autodict' object is a special kind of Python dictionary
+(implemented in Core.py) that implements Perl's 'autovivifying' hashes
+in Python i.e. with autovivifying hashes, you can assign nested hash
+values without having to go to the trouble of creating intermediate
+levels if they don't exist e.g syscalls[comm][pid][id] = 1 will create
+the intermediate hash levels and finally assign the value 1 to the
+hash entry for 'id' (because the value being assigned isn't a hash
+object itself, the initial value is assigned in the TypeError
+exception.  Well, there may be a better way to do this in Python but
+that's what works for now).
+
+Putting that code into the raw_syscalls__sys_enter() handler, we
+effectively end up with a single-level dictionary keyed on syscall id
+and having the counts we've tallied as values.
+
+The print_syscall_totals() function iterates over the entries in the
+dictionary and displays a line for each entry containing the syscall
+name (the dictonary keys contain the syscall ids, which are passed to
+the Util function syscall_name(), which translates the raw syscall
+numbers to the corresponding syscall name strings).  The output is
+displayed after all the events in the trace have been processed, by
+calling the print_syscall_totals() function from the trace_end()
+handler called at the end of script processing.
+
+The final script producing the output shown above is shown in its
+entirety below (syscall_name() helper is not yet available, you can
+only deal with id's for now):
+
+----
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+        '/scripts/python/perf-script-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from Util import *
+
+syscalls = autodict()
+
+def trace_end():
+        print_syscall_totals()
+
+def raw_syscalls__sys_enter(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        id, args):
+        try:
+                syscalls[id] += 1
+        except TypeError:
+                syscalls[id] = 1
+
+def print_syscall_totals():
+    if for_comm is not None:
+            print "\nsyscall events for %s:\n\n" % (for_comm),
+    else:
+            print "\nsyscall events:\n\n",
+
+    print "%-40s  %10s\n" % ("event", "count"),
+    print "%-40s  %10s\n" % ("----------------------------------------", \
+                                 "-----------"),
+
+    for id, val in sorted(syscalls.iteritems(), key = lambda(k, v): (v, k), \
+                                  reverse = True):
+            print "%-40s  %10d\n" % (syscall_name(id), val),
+----
+
+The script can be run just as before:
+
+  # perf script -s syscall-counts.py
+
+So those are the essential steps in writing and running a script.  The
+process can be generalized to any tracepoint or set of tracepoints
+you're interested in - basically find the tracepoint(s) you're
+interested in by looking at the list of available events shown by
+'perf list' and/or look in /sys/kernel/debug/tracing events for
+detailed event and field info, record the corresponding trace data
+using 'perf record', passing it the list of interesting events,
+generate a skeleton script using 'perf script -g python' and modify the
+code to aggregate and display it for your particular needs.
+
+After you've done that you may end up with a general-purpose script
+that you want to keep around and have available for future use.  By
+writing a couple of very simple shell scripts and putting them in the
+right place, you can have your script listed alongside the other
+scripts listed by the 'perf script -l' command e.g.:
+
+----
+root@tropicana:~# perf script -l
+List of available trace scripts:
+  workqueue-stats                      workqueue stats (ins/exe/create/destroy)
+  wakeup-latency                       system-wide min/max/avg wakeup latency
+  rw-by-file <comm>                    r/w activity for a program, by file
+  rw-by-pid                            system-wide r/w activity
+----
+
+A nice side effect of doing this is that you also then capture the
+probably lengthy 'perf record' command needed to record the events for
+the script.
+
+To have the script appear as a 'built-in' script, you write two simple
+scripts, one for recording and one for 'reporting'.
+
+The 'record' script is a shell script with the same base name as your
+script, but with -record appended.  The shell script should be put
+into the perf/scripts/python/bin directory in the kernel source tree.
+In that script, you write the 'perf record' command-line needed for
+your script:
+
+----
+# cat kernel-source/tools/perf/scripts/python/bin/syscall-counts-record
+
+#!/bin/bash
+perf record -a -e raw_syscalls:sys_enter
+----
+
+The 'report' script is also a shell script with the same base name as
+your script, but with -report appended.  It should also be located in
+the perf/scripts/python/bin directory.  In that script, you write the
+'perf script -s' command-line needed for running your script:
+
+----
+# cat kernel-source/tools/perf/scripts/python/bin/syscall-counts-report
+
+#!/bin/bash
+# description: system-wide syscall counts
+perf script -s ~/libexec/perf-core/scripts/python/syscall-counts.py
+----
+
+Note that the location of the Python script given in the shell script
+is in the libexec/perf-core/scripts/python directory - this is where
+the script will be copied by 'make install' when you install perf.
+For the installation to install your script there, your script needs
+to be located in the perf/scripts/python directory in the kernel
+source tree:
+
+----
+# ls -al kernel-source/tools/perf/scripts/python
+
+root@tropicana:/home/trz/src/tip# ls -al tools/perf/scripts/python
+total 32
+drwxr-xr-x 4 trz trz 4096 2010-01-26 22:30 .
+drwxr-xr-x 4 trz trz 4096 2010-01-26 22:29 ..
+drwxr-xr-x 2 trz trz 4096 2010-01-26 22:29 bin
+-rw-r--r-- 1 trz trz 2548 2010-01-26 22:29 check-perf-script.py
+drwxr-xr-x 3 trz trz 4096 2010-01-26 22:49 perf-script-Util
+-rw-r--r-- 1 trz trz 1462 2010-01-26 22:30 syscall-counts.py
+----
+
+Once you've done that (don't forget to do a new 'make install',
+otherwise your script won't show up at run-time), 'perf script -l'
+should show a new entry for your script:
+
+----
+root@tropicana:~# perf script -l
+List of available trace scripts:
+  workqueue-stats                      workqueue stats (ins/exe/create/destroy)
+  wakeup-latency                       system-wide min/max/avg wakeup latency
+  rw-by-file <comm>                    r/w activity for a program, by file
+  rw-by-pid                            system-wide r/w activity
+  syscall-counts                       system-wide syscall counts
+----
+
+You can now perform the record step via 'perf script record':
+
+  # perf script record syscall-counts
+
+and display the output using 'perf script report':
+
+  # perf script report syscall-counts
+
+STARTER SCRIPTS
+---------------
+
+You can quickly get started writing a script for a particular set of
+trace data by generating a skeleton script using 'perf script -g
+python' in the same directory as an existing perf.data trace file.
+That will generate a starter script containing a handler for each of
+the event types in the trace file; it simply prints every available
+field for each event in the trace file.
+
+You can also look at the existing scripts in
+~/libexec/perf-core/scripts/python for typical examples showing how to
+do basic things like aggregate event data, print results, etc.  Also,
+the check-perf-script.py script, while not interesting for its results,
+attempts to exercise all of the main scripting features.
+
+EVENT HANDLERS
+--------------
+
+When perf script is invoked using a trace script, a user-defined
+'handler function' is called for each event in the trace.  If there's
+no handler function defined for a given event type, the event is
+ignored (or passed to a 'trace_handled' function, see below) and the
+next event is processed.
+
+Most of the event's field values are passed as arguments to the
+handler function; some of the less common ones aren't - those are
+available as calls back into the perf executable (see below).
+
+As an example, the following perf record command can be used to record
+all sched_wakeup events in the system:
+
+ # perf record -a -e sched:sched_wakeup
+
+Traces meant to be processed using a script should be recorded with
+the above option: -a to enable system-wide collection.
+
+The format file for the sched_wakep event defines the following fields
+(see /sys/kernel/debug/tracing/events/sched/sched_wakeup/format):
+
+----
+ format:
+        field:unsigned short common_type;
+        field:unsigned char common_flags;
+        field:unsigned char common_preempt_count;
+        field:int common_pid;
+        field:int common_lock_depth;
+
+        field:char comm[TASK_COMM_LEN];
+        field:pid_t pid;
+        field:int prio;
+        field:int success;
+        field:int target_cpu;
+----
+
+The handler function for this event would be defined as:
+
+----
+def sched__sched_wakeup(event_name, context, common_cpu, common_secs,
+       common_nsecs, common_pid, common_comm,
+       comm, pid, prio, success, target_cpu):
+       pass
+----
+
+The handler function takes the form subsystem__event_name.
+
+The common_* arguments in the handler's argument list are the set of
+arguments passed to all event handlers; some of the fields correspond
+to the common_* fields in the format file, but some are synthesized,
+and some of the common_* fields aren't common enough to to be passed
+to every event as arguments but are available as library functions.
+
+Here's a brief description of each of the invariant event args:
+
+ event_name                 the name of the event as text
+ context                    an opaque 'cookie' used in calls back into perf
+ common_cpu                 the cpu the event occurred on
+ common_secs                the secs portion of the event timestamp
+ common_nsecs               the nsecs portion of the event timestamp
+ common_pid                 the pid of the current task
+ common_comm                the name of the current process
+
+All of the remaining fields in the event's format file have
+counterparts as handler function arguments of the same name, as can be
+seen in the example above.
+
+The above provides the basics needed to directly access every field of
+every event in a trace, which covers 90% of what you need to know to
+write a useful trace script.  The sections below cover the rest.
+
+SCRIPT LAYOUT
+-------------
+
+Every perf script Python script should start by setting up a Python
+module search path and 'import'ing a few support modules (see module
+descriptions below):
+
+----
+ import os
+ import sys
+
+ sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+              '/scripts/python/perf-script-Util/lib/Perf/Trace')
+
+ from perf_trace_context import *
+ from Core import *
+----
+
+The rest of the script can contain handler functions and support
+functions in any order.
+
+Aside from the event handler functions discussed above, every script
+can implement a set of optional functions:
+
+*trace_begin*, if defined, is called before any event is processed and
+gives scripts a chance to do setup tasks:
+
+----
+def trace_begin:
+    pass
+----
+
+*trace_end*, if defined, is called after all events have been
+ processed and gives scripts a chance to do end-of-script tasks, such
+ as display results:
+
+----
+def trace_end:
+    pass
+----
+
+*trace_unhandled*, if defined, is called after for any event that
+ doesn't have a handler explicitly defined for it.  The standard set
+ of common arguments are passed into it:
+
+----
+def trace_unhandled(event_name, context, common_cpu, common_secs,
+        common_nsecs, common_pid, common_comm):
+    pass
+----
+
+The remaining sections provide descriptions of each of the available
+built-in perf script Python modules and their associated functions.
+
+AVAILABLE MODULES AND FUNCTIONS
+-------------------------------
+
+The following sections describe the functions and variables available
+via the various perf script Python modules.  To use the functions and
+variables from the given module, add the corresponding 'from XXXX
+import' line to your perf script script.
+
+Core.py Module
+~~~~~~~~~~~~~~
+
+These functions provide some essential functions to user scripts.
+
+The *flag_str* and *symbol_str* functions provide human-readable
+strings for flag and symbolic fields.  These correspond to the strings
+and values parsed from the 'print fmt' fields of the event format
+files:
+
+  flag_str(event_name, field_name, field_value) - returns the string represention corresponding to field_value for the flag field field_name of event event_name
+  symbol_str(event_name, field_name, field_value) - returns the string represention corresponding to field_value for the symbolic field field_name of event event_name
+
+The *autodict* function returns a special kind of Python
+dictionary that implements Perl's 'autovivifying' hashes in Python
+i.e. with autovivifying hashes, you can assign nested hash values
+without having to go to the trouble of creating intermediate levels if
+they don't exist.
+
+  autodict() - returns an autovivifying dictionary instance
+
+
+perf_trace_context Module
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Some of the 'common' fields in the event format file aren't all that
+common, but need to be made accessible to user scripts nonetheless.
+
+perf_trace_context defines a set of functions that can be used to
+access this data in the context of the current event.  Each of these
+functions expects a context variable, which is the same as the
+context variable passed into every event handler as the second
+argument.
+
+ common_pc(context) - returns common_preempt count for the current event
+ common_flags(context) - returns common_flags for the current event
+ common_lock_depth(context) - returns common_lock_depth for the current event
+
+Util.py Module
+~~~~~~~~~~~~~~
+
+Various utility functions for use with perf script:
+
+  nsecs(secs, nsecs) - returns total nsecs given secs/nsecs pair
+  nsecs_secs(nsecs) - returns whole secs portion given nsecs
+  nsecs_nsecs(nsecs) - returns nsecs remainder given nsecs
+  nsecs_str(nsecs) - returns printable string in the form secs.nsecs
+  avg(total, n) - returns average given a sum and a total number of values
+
+SEE ALSO
+--------
+linkperf:perf-script[1]
diff --git a/tools/perf/Documentation/perf-script.txt b/tools/perf/Documentation/perf-script.txt
new file mode 100644
index 00000000000..29ad94293cd
--- /dev/null
+++ b/tools/perf/Documentation/perf-script.txt
@@ -0,0 +1,118 @@
+perf-script(1)
+=============
+
+NAME
+----
+perf-script - Read perf.data (created by perf record) and display trace output
+
+SYNOPSIS
+--------
+[verse]
+'perf script' [<options>]
+'perf script' [<options>] record <script> [<record-options>] <command>
+'perf script' [<options>] report <script> [script-args]
+'perf script' [<options>] <script> <required-script-args> [<record-options>] <command>
+'perf script' [<options>] <top-script> [script-args]
+
+DESCRIPTION
+-----------
+This command reads the input file and displays the trace recorded.
+
+There are several variants of perf script:
+
+  'perf script' to see a detailed trace of the workload that was
+  recorded.
+
+  You can also run a set of pre-canned scripts that aggregate and
+  summarize the raw trace data in various ways (the list of scripts is
+  available via 'perf script -l').  The following variants allow you to
+  record and run those scripts:
+
+  'perf script record <script> <command>' to record the events required
+  for 'perf script report'.  <script> is the name displayed in the
+  output of 'perf script --list' i.e. the actual script name minus any
+  language extension.  If <command> is not specified, the events are
+  recorded using the -a (system-wide) 'perf record' option.
+
+  'perf script report <script> [args]' 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 script
+  record <script>' is used and should be present for this command to
+  succeed.  [args] refers to the (mainly optional) args expected by
+  the script.
+
+  'perf script <script> <required-script-args> <command>' to both
+  record the events required for <script> and to run the <script>
+  using 'live-mode' i.e. without writing anything to disk.  <script>
+  is the name displayed in the output of 'perf script --list' i.e. the
+  actual script name minus any language extension.  If <command> is
+  not specified, the events are recorded using the -a (system-wide)
+  'perf record' option.  If <script> has any required args, they
+  should be specified before <command>.  This mode doesn't allow for
+  optional script args to be specified; if optional script args are
+  desired, they can be specified using separate 'perf script record'
+  and 'perf script report' commands, with the stdout of the record step
+  piped to the stdin of the report script, using the '-o -' and '-i -'
+  options of the corresponding commands.
+
+  'perf script <top-script>' to both record the events required for
+  <top-script> and to run the <top-script> using 'live-mode'
+  i.e. without writing anything to disk.  <top-script> is the name
+  displayed in the output of 'perf script --list' i.e. the actual
+  script name minus any language extension; a <top-script> is defined
+  as any script name ending with the string 'top'.
+
+  [<record-options>] can be passed to the record steps of 'perf script
+  record' and 'live-mode' variants; this isn't possible however for
+  <top-script> 'live-mode' or 'perf script report' variants.
+
+  See the 'SEE ALSO' section for links to language-specific
+  information on how to write and run your own trace scripts.
+
+OPTIONS
+-------
+<command>...::
+        Any command you can specify in a shell.
+
+-D::
+--dump-raw-script=::
+        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 ['lang']::
+--script=::
+        Process trace data with the given script ([lang]:script[.ext]).
+        If the string 'lang' is specified in place of a script name, a
+        list of supported languages will be displayed instead.
+
+-g::
+--gen-script=::
+        Generate perf-script.[ext] starter script for given language,
+        using current perf.data.
+
+-a::
+        Force system-wide collection.  Scripts run without a <command>
+        normally use -a by default, while scripts run with a <command>
+        normally don't - this option allows the latter to be run in
+        system-wide mode.
+
+-i::
+--input=::
+        Input file name.
+
+-d::
+--debug-mode::
+        Do various checks like samples ordering and lost events.
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-script-perl[1],
+linkperf:perf-script-python[1]
diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt
index c368a72721d..b6da7affbbe 100644
--- a/tools/perf/Documentation/perf-stat.txt
+++ b/tools/perf/Documentation/perf-stat.txt
@@ -8,8 +8,8 @@ perf-stat - Run a command and gather performance counter statistics
 SYNOPSIS
 --------
 [verse]
-'perf stat' [-e <EVENT> | --event=EVENT] [-l] [-a] <command>
-'perf stat' [-e <EVENT> | --event=EVENT] [-l] [-a] -- <command> [<options>]
+'perf stat' [-e <EVENT> | --event=EVENT] [-a] <command>
+'perf stat' [-e <EVENT> | --event=EVENT] [-a] -- <command> [<options>]
 
 DESCRIPTION
 -----------
@@ -31,17 +31,57 @@ OPTIONS
          hexadecimal event descriptor.
 
 -i::
---inherit::
-        child tasks inherit counters
+--no-inherit::
+        child tasks do not inherit counters
 -p::
 --pid=<pid>::
-        stat events on existing pid
+        stat events on existing process id
+
+-t::
+--tid=<tid>::
+        stat events on existing thread id
 
--a::
-        system-wide collection
 
--l::
-        scale counter values
+-a::
+--all-cpus::
+        system-wide collection from all CPUs
+
+-c::
+--scale::
+        scale/normalize counter values
+
+-r::
+--repeat=<n>::
+        repeat command and print average + stddev (max: 100)
+
+-B::
+--big-num::
+        print large numbers with thousands' separators according to locale
+
+-C::
+--cpu=::
+Count only on the list of CPUs provided. Multiple CPUs can be provided as a
+comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
+In per-thread mode, this option is ignored. The -a option is still necessary
+to activate system-wide monitoring. Default is to count on all CPUs.
+
+-A::
+--no-aggr::
+Do not aggregate counts across all monitored CPUs in system-wide mode (-a).
+This option is only valid in system-wide mode.
+
+-n::
+--null::
+        null run - don't start any counters
+
+-v::
+--verbose::
+        be more verbose (show counter open errors, etc)
+
+-x SEP::
+--field-separator SEP::
+print counts using a CSV-style output to make it easy to import directly into
+spreadsheets. Columns are separated by the string specified in SEP.
 
 EXAMPLES
 --------
diff --git a/tools/perf/Documentation/perf-test.txt b/tools/perf/Documentation/perf-test.txt
new file mode 100644
index 00000000000..2c3b462f64b
--- /dev/null
+++ b/tools/perf/Documentation/perf-test.txt
@@ -0,0 +1,22 @@
+perf-test(1)
+============
+
+NAME
+----
+perf-test - Runs sanity tests.
+
+SYNOPSIS
+--------
+[verse]
+'perf test <options>'
+
+DESCRIPTION
+-----------
+This command does assorted sanity tests, initially through linked routines but
+also will look for a directory with more tests in the form of scripts.
+
+OPTIONS
+-------
+-v::
+--verbose::
+        Be more verbose.
diff --git a/tools/perf/Documentation/perf-timechart.txt b/tools/perf/Documentation/perf-timechart.txt
new file mode 100644
index 00000000000..d7b79e2ba2a
--- /dev/null
+++ b/tools/perf/Documentation/perf-timechart.txt
@@ -0,0 +1,46 @@
+perf-timechart(1)
+=================
+
+NAME
+----
+perf-timechart - Tool to visualize total system behavior during a workload
+
+SYNOPSIS
+--------
+[verse]
+'perf timechart' {record}
+
+DESCRIPTION
+-----------
+There are two variants of perf timechart:
+
+  'perf timechart record <command>' to record the system level events
+  of an arbitrary workload.
+
+  'perf timechart' to turn a trace into a Scalable Vector Graphics file,
+  that can be viewed with popular SVG viewers such as 'Inkscape'.
+
+OPTIONS
+-------
+-o::
+--output=::
+        Select the output file (default: output.svg)
+-i::
+--input=::
+        Select the input file (default: perf.data)
+-w::
+--width=::
+        Select the width of the SVG file (default: 1000)
+-P::
+--power-only::
+        Only output the CPU power section of the diagram
+-p::
+--process::
+        Select the processes to display, by name or PID
+
+--symfs=<directory>::
+        Look for files with symbols relative to this directory.
+
+SEE ALSO
+--------
+linkperf:perf-record[1]
diff --git a/tools/perf/Documentation/perf-top.txt b/tools/perf/Documentation/perf-top.txt
index 539d0128972..f6eb1cdafb7 100644
--- a/tools/perf/Documentation/perf-top.txt
+++ b/tools/perf/Documentation/perf-top.txt
@@ -3,36 +3,144 @@ perf-top(1)
 
 NAME
 ----
-perf-top - Run a command and profile it
+perf-top - System profiling tool.
 
 SYNOPSIS
 --------
 [verse]
-'perf top' [-e <EVENT> | --event=EVENT] [-l] [-a] <command>
+'perf top' [-e <EVENT> | --event=EVENT] [<options>]
 
 DESCRIPTION
 -----------
-This command runs a command and gathers a performance counter profile
-from it.
+This command generates and displays a performance counter profile in real time.
 
 
 OPTIONS
 -------
-<command>...::
-        Any command you can specify in a shell.
+-a::
+--all-cpus::
+        System-wide collection.  (default)
+
+-c <count>::
+--count=<count>::
+        Event period to sample.
+
+-C <cpu-list>::
+--cpu=<cpu>::
+Monitor only on the list of CPUs provided. Multiple CPUs can be provided as a
+comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
+Default is to monitor all CPUS.
+
+-d <seconds>::
+--delay=<seconds>::
+        Number of seconds to delay between refreshes.
 
--e::
---event=::
+-e <event>::
+--event=<event>::
         Select the PMU event. Selection can be a symbolic event name
         (use 'perf list' to list all events) or a raw PMU
         event (eventsel+umask) in the form of rNNN where NNN is a
-         hexadecimal event descriptor.
+        hexadecimal event descriptor.
 
--a::
-        system-wide collection
+-E <entries>::
+--entries=<entries>::
+        Display this many functions.
+
+-f <count>::
+--count-filter=<count>::
+        Only display functions with more events than this.
+
+-g::
+--group::
+        Put the counters into a counter group.
+
+-F <freq>::
+--freq=<freq>::
+        Profile at this frequency.
+
+-i::
+--inherit::
+        Child tasks inherit counters, only makes sens with -p option.
+
+-k <path>::
+--vmlinux=<path>::
+        Path to vmlinux.  Required for annotation functionality.
+
+-m <pages>::
+--mmap-pages=<pages>::
+        Number of mmapped data pages.
+
+-p <pid>::
+--pid=<pid>::
+        Profile events on existing Process ID.
+
+-t <tid>::
+--tid=<tid>::
+        Profile events on existing thread ID.
+
+-r <priority>::
+--realtime=<priority>::
+        Collect data with this RT SCHED_FIFO priority.
+
+-s <symbol>::
+--sym-annotate=<symbol>::
+        Annotate this symbol.
+
+-K::
+--hide_kernel_symbols::
+        Hide kernel symbols.
+
+-U::
+--hide_user_symbols::
+        Hide user symbols.
+
+-D::
+--dump-symtab::
+        Dump the symbol table used for profiling.
+
+-v::
+--verbose::
+        Be more verbose (show counter open errors, etc).
+
+-z::
+--zero::
+        Zero history across display updates.
+
+INTERACTIVE PROMPTING KEYS
+--------------------------
+
+[d]::
+        Display refresh delay.
+
+[e]::
+        Number of entries to display.
+
+[E]::
+        Event to display when multiple counters are active.
+
+[f]::
+        Profile display filter (>= hit count).
+
+[F]::
+        Annotation display filter (>= % of total).
+
+[s]::
+        Annotate symbol.
+
+[S]::
+        Stop annotation, return to full profile display.
+
+[w]::
+        Toggle between weighted sum and individual count[E]r profile.
+
+[z]::
+        Toggle event count zeroing across display updates.
+
+[qQ]::
+        Quit.
+
+Pressing any unmapped key displays a menu, and prompts for input.
 
--l::
-        scale counter values
 
 SEE ALSO
 --------
diff --git a/tools/perf/Documentation/perf.txt b/tools/perf/Documentation/perf.txt
index 69c83255719..0eeb247dc7d 100644
--- a/tools/perf/Documentation/perf.txt
+++ b/tools/perf/Documentation/perf.txt
@@ -12,7 +12,7 @@ SYNOPSIS
 
 DESCRIPTION
 -----------
-Performance counters for Linux are are a new kernel-based subsystem
+Performance counters for Linux are a new kernel-based subsystem
 that provide a framework for all things performance analysis. It
 covers hardware level (CPU/PMU, Performance Monitoring Unit) features
 and software features (software counters, tracepoints) as well.
diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST
new file mode 100644
index 00000000000..c12659d8cb2
--- /dev/null
+++ b/tools/perf/MANIFEST
@@ -0,0 +1,13 @@
+tools/perf
+include/linux/perf_event.h
+include/linux/rbtree.h
+include/linux/list.h
+include/linux/hash.h
+include/linux/stringify.h
+lib/rbtree.c
+include/linux/swab.h
+arch/*/include/asm/unistd*.h
+arch/*/lib/memcpy*.S
+include/linux/poison.h
+include/linux/magic.h
+include/linux/hw_breakpoint.h
diff --git a/tools/perf/Makefile b/tools/perf/Makefile
index 36d7eef4991..2b5387d53ba 100644
--- a/tools/perf/Makefile
+++ b/tools/perf/Makefile
@@ -1,7 +1,18 @@
+ifeq ("$(origin O)", "command line")
+        OUTPUT := $(O)/
+endif
+
 # The default target of this Makefile is...
 all::
 
+ifneq ($(OUTPUT),)
+# check that the output directory actually exists
+OUTDIR := $(shell cd $(OUTPUT) && /bin/pwd)
+$(if $(OUTDIR),, $(error output directory "$(OUTPUT)" does not exist))
+endif
+
 # Define V=1 to have a more verbose compile.
+# Define V=2 to have an even more verbose compile.
 #
 # Define SNPRINTF_RETURNS_BOGUS if your are on a system which snprintf()
 # or vsnprintf() return -1 instead of number of characters which would
@@ -145,10 +156,16 @@ all::
 # Define NO_EXTERNAL_GREP if you don't want "perf grep" to ever call
 # your external grep (e.g., if your system lacks grep, if its grep is
 # broken, or spawning external process is slower than built-in grep perf has).
+#
+# Define LDFLAGS=-static to build a static binary.
+#
+# Define EXTRA_CFLAGS=-m64 or EXTRA_CFLAGS=-m32 as appropriate for cross-builds.
+#
+# Define NO_DWARF if you do not want debug-info analysis feature at all.
 
-PERF-VERSION-FILE: .FORCE-PERF-VERSION-FILE
-        @$(SHELL_PATH) util/PERF-VERSION-GEN
--include PERF-VERSION-FILE
+$(OUTPUT)PERF-VERSION-FILE: .FORCE-PERF-VERSION-FILE
+        @$(SHELL_PATH) util/PERF-VERSION-GEN $(OUTPUT)
+-include $(OUTPUT)PERF-VERSION-FILE
 
 uname_S := $(shell sh -c 'uname -s 2>/dev/null || echo not')
 uname_M := $(shell sh -c 'uname -m 2>/dev/null || echo not')
@@ -157,16 +174,62 @@ uname_R := $(shell sh -c 'uname -r 2>/dev/null || echo not')
 uname_P := $(shell sh -c 'uname -p 2>/dev/null || echo not')
 uname_V := $(shell sh -c 'uname -v 2>/dev/null || echo not')
 
-# If we're on a 64-bit kernel, use -m64
-ifneq ($(patsubst %64,%,$(uname_M)),$(uname_M))
-  M64 := -m64
+ARCH ?= $(shell echo $(uname_M) | sed -e s/i.86/i386/ -e s/sun4u/sparc64/ \
+                                  -e s/arm.*/arm/ -e s/sa110/arm/ \
+                                  -e s/s390x/s390/ -e s/parisc64/parisc/ \
+                                  -e s/ppc.*/powerpc/ -e s/mips.*/mips/ \
+                                  -e s/sh[234].*/sh/ )
+
+# Additional ARCH settings for x86
+ifeq ($(ARCH),i386)
+        ARCH := x86
+endif
+ifeq ($(ARCH),x86_64)
+        RAW_ARCH := x86_64
+        ARCH := x86
+        ARCH_CFLAGS := -DARCH_X86_64
+        ARCH_INCLUDE = ../../arch/x86/lib/memcpy_64.S
 endif
 
 # CFLAGS and LDFLAGS are for the users to override from the command line.
 
-CFLAGS = $(M64) -ggdb3 -Wall -Wstrict-prototypes -Wmissing-declarations -Wmissing-prototypes -std=gnu99 -Wdeclaration-after-statement -Werror -O6
-LDFLAGS = -lpthread -lrt -lelf -lm
-ALL_CFLAGS = $(CFLAGS)
+#
+# Include saner warnings here, which can catch bugs:
+#
+
+EXTRA_WARNINGS := -Wformat
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wformat-security
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wformat-y2k
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wshadow
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Winit-self
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wpacked
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wredundant-decls
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wstack-protector
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wstrict-aliasing=3
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wswitch-default
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wswitch-enum
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wno-system-headers
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wundef
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wvolatile-register-var
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wwrite-strings
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wbad-function-cast
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wmissing-declarations
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wmissing-prototypes
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wnested-externs
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wold-style-definition
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wstrict-prototypes
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wdeclaration-after-statement
+
+ifeq ("$(origin DEBUG)", "command line")
+  PERF_DEBUG = $(DEBUG)
+endif
+ifndef PERF_DEBUG
+  CFLAGS_OPTIMIZE = -O6
+endif
+
+CFLAGS = -fno-omit-frame-pointer -ggdb3 -Wall -Wextra -std=gnu99 -Werror $(CFLAGS_OPTIMIZE) -D_FORTIFY_SOURCE=2 $(EXTRA_WARNINGS) $(EXTRA_CFLAGS)
+EXTLIBS = -lpthread -lrt -lelf -lm
+ALL_CFLAGS = $(CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
 ALL_LDFLAGS = $(LDFLAGS)
 STRIP ?= strip
 
@@ -182,7 +245,10 @@ STRIP ?= strip
 # runtime figures out where they are based on the path to the executable.
 # This can help installing the suite in a relocatable way.
 
+# Make the path relative to DESTDIR, not to prefix
+ifndef DESTDIR
 prefix = $(HOME)
+endif
 bindir_relative = bin
 bindir = $(prefix)/$(bindir_relative)
 mandir = share/man
@@ -199,13 +265,13 @@ sysconfdir = $(prefix)/etc
 ETC_PERFCONFIG = etc/perfconfig
 endif
 lib = lib
-# DESTDIR=
 
 export prefix bindir sharedir sysconfdir
 
-CC = gcc
-AR = ar
+CC = $(CROSS_COMPILE)gcc
+AR = $(CROSS_COMPILE)ar
 RM = rm -f
+MKDIR = mkdir
 TAR = tar
 FIND = find
 INSTALL = install
@@ -216,6 +282,17 @@ PTHREAD_LIBS = -lpthread
 # explicitly what architecture to check for. Fix this up for yours..
 SPARSE_FLAGS = -D__BIG_ENDIAN__ -D__powerpc__
 
+ifeq ($(V), 2)
+        QUIET_STDERR = ">/dev/null"
+else
+        QUIET_STDERR = ">/dev/null 2>&1"
+endif
+
+-include feature-tests.mak
+
+ifeq ($(call try-cc,$(SOURCE_HELLO),-Werror -fstack-protector-all),y)
+        CFLAGS := $(CFLAGS) -fstack-protector-all
+endif
 
 
 ### --- END CONFIGURATION SECTION ---
@@ -223,7 +300,7 @@ SPARSE_FLAGS = -D__BIG_ENDIAN__ -D__powerpc__
 # Those must not be GNU-specific; they are shared with perl/ which may
 # be built by a different compiler. (Note that this is an artifact now
 # but it still might be nice to keep that distinction.)
-BASIC_CFLAGS =
+BASIC_CFLAGS = -Iutil/include -Iarch/$(ARCH)/include
 BASIC_LDFLAGS =
 
 # Guard against environment variables
@@ -237,11 +314,10 @@ SCRIPT_PERL =
 SCRIPT_SH =
 TEST_PROGRAMS =
 
-#
-# No scripts right now:
-#
+SCRIPT_SH += perf-archive.sh
 
-# SCRIPT_SH += perf-am.sh
+grep-libs = $(filter -l%,$(1))
+strip-libs = $(filter-out -l%,$(1))
 
 #
 # No Perl scripts right now:
@@ -261,20 +337,17 @@ PROGRAMS += $(EXTRA_PROGRAMS)
 #
 # Single 'perf' binary right now:
 #
-PROGRAMS += perf
+PROGRAMS += $(OUTPUT)perf
 
 # List built-in command $C whose implementation cmd_$C() is not in
 # builtin-$C.o but is linked in as part of some other command.
 #
-# None right now:
-#
-# BUILT_INS += perf-init $X
 
 # what 'all' will build and 'install' will install, in perfexecdir
 ALL_PROGRAMS = $(PROGRAMS) $(SCRIPTS)
 
 # what 'all' will build but not install in perfexecdir
-OTHER_PROGRAMS = perf$X
+OTHER_PROGRAMS = $(OUTPUT)perf$X
 
 # Set paths to tools early so that they can be used for version tests.
 ifndef SHELL_PATH
@@ -286,59 +359,155 @@ endif
 
 export PERL_PATH
 
-LIB_FILE=libperf.a
-
-LIB_H += ../../include/linux/perf_counter.h
+LIB_FILE=$(OUTPUT)libperf.a
+
+LIB_H += ../../include/linux/perf_event.h
+LIB_H += ../../include/linux/rbtree.h
+LIB_H += ../../include/linux/list.h
+LIB_H += ../../include/linux/hash.h
+LIB_H += ../../include/linux/stringify.h
+LIB_H += util/include/linux/bitmap.h
+LIB_H += util/include/linux/bitops.h
+LIB_H += util/include/linux/compiler.h
+LIB_H += util/include/linux/ctype.h
+LIB_H += util/include/linux/kernel.h
+LIB_H += util/include/linux/list.h
+LIB_H += util/include/linux/module.h
+LIB_H += util/include/linux/poison.h
+LIB_H += util/include/linux/prefetch.h
+LIB_H += util/include/linux/rbtree.h
+LIB_H += util/include/linux/string.h
+LIB_H += util/include/linux/types.h
+LIB_H += util/include/linux/linkage.h
+LIB_H += util/include/asm/asm-offsets.h
+LIB_H += util/include/asm/bug.h
+LIB_H += util/include/asm/byteorder.h
+LIB_H += util/include/asm/hweight.h
+LIB_H += util/include/asm/swab.h
+LIB_H += util/include/asm/system.h
+LIB_H += util/include/asm/uaccess.h
+LIB_H += util/include/dwarf-regs.h
+LIB_H += util/include/asm/dwarf2.h
+LIB_H += util/include/asm/cpufeature.h
 LIB_H += perf.h
-LIB_H += types.h
-LIB_H += util/list.h
-LIB_H += util/rbtree.h
+LIB_H += util/cache.h
+LIB_H += util/callchain.h
+LIB_H += util/build-id.h
+LIB_H += util/debug.h
+LIB_H += util/debugfs.h
+LIB_H += util/event.h
+LIB_H += util/evsel.h
+LIB_H += util/exec_cmd.h
+LIB_H += util/types.h
 LIB_H += util/levenshtein.h
+LIB_H += util/map.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/xyarray.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
 LIB_H += util/color.h
-
-LIB_OBJS += util/abspath.o
-LIB_OBJS += util/alias.o
-LIB_OBJS += util/config.o
-LIB_OBJS += util/ctype.o
-LIB_OBJS += util/environment.o
-LIB_OBJS += util/exec_cmd.o
-LIB_OBJS += util/help.o
-LIB_OBJS += util/levenshtein.o
-LIB_OBJS += util/parse-options.o
-LIB_OBJS += util/parse-events.o
-LIB_OBJS += util/path.o
-LIB_OBJS += util/rbtree.o
-LIB_OBJS += util/run-command.o
-LIB_OBJS += util/quote.o
-LIB_OBJS += util/strbuf.o
-LIB_OBJS += util/string.o
-LIB_OBJS += util/usage.o
-LIB_OBJS += util/wrapper.o
-LIB_OBJS += util/sigchain.o
-LIB_OBJS += util/symbol.o
-LIB_OBJS += util/color.o
-LIB_OBJS += util/pager.o
-
-BUILTIN_OBJS += builtin-annotate.o
-BUILTIN_OBJS += builtin-help.o
-BUILTIN_OBJS += builtin-list.o
-BUILTIN_OBJS += builtin-record.o
-BUILTIN_OBJS += builtin-report.o
-BUILTIN_OBJS += builtin-stat.o
-BUILTIN_OBJS += builtin-top.o
+LIB_H += util/values.h
+LIB_H += util/sort.h
+LIB_H += util/hist.h
+LIB_H += util/thread.h
+LIB_H += util/trace-event.h
+LIB_H += util/probe-finder.h
+LIB_H += util/probe-event.h
+LIB_H += util/pstack.h
+LIB_H += util/cpumap.h
+LIB_H += $(ARCH_INCLUDE)
+
+LIB_OBJS += $(OUTPUT)util/abspath.o
+LIB_OBJS += $(OUTPUT)util/alias.o
+LIB_OBJS += $(OUTPUT)util/build-id.o
+LIB_OBJS += $(OUTPUT)util/config.o
+LIB_OBJS += $(OUTPUT)util/ctype.o
+LIB_OBJS += $(OUTPUT)util/debugfs.o
+LIB_OBJS += $(OUTPUT)util/environment.o
+LIB_OBJS += $(OUTPUT)util/event.o
+LIB_OBJS += $(OUTPUT)util/evsel.o
+LIB_OBJS += $(OUTPUT)util/exec_cmd.o
+LIB_OBJS += $(OUTPUT)util/help.o
+LIB_OBJS += $(OUTPUT)util/levenshtein.o
+LIB_OBJS += $(OUTPUT)util/parse-options.o
+LIB_OBJS += $(OUTPUT)util/parse-events.o
+LIB_OBJS += $(OUTPUT)util/path.o
+LIB_OBJS += $(OUTPUT)util/rbtree.o
+LIB_OBJS += $(OUTPUT)util/bitmap.o
+LIB_OBJS += $(OUTPUT)util/hweight.o
+LIB_OBJS += $(OUTPUT)util/run-command.o
+LIB_OBJS += $(OUTPUT)util/quote.o
+LIB_OBJS += $(OUTPUT)util/strbuf.o
+LIB_OBJS += $(OUTPUT)util/string.o
+LIB_OBJS += $(OUTPUT)util/strlist.o
+LIB_OBJS += $(OUTPUT)util/usage.o
+LIB_OBJS += $(OUTPUT)util/wrapper.o
+LIB_OBJS += $(OUTPUT)util/sigchain.o
+LIB_OBJS += $(OUTPUT)util/symbol.o
+LIB_OBJS += $(OUTPUT)util/color.o
+LIB_OBJS += $(OUTPUT)util/pager.o
+LIB_OBJS += $(OUTPUT)util/header.o
+LIB_OBJS += $(OUTPUT)util/callchain.o
+LIB_OBJS += $(OUTPUT)util/values.o
+LIB_OBJS += $(OUTPUT)util/debug.o
+LIB_OBJS += $(OUTPUT)util/map.o
+LIB_OBJS += $(OUTPUT)util/pstack.o
+LIB_OBJS += $(OUTPUT)util/session.o
+LIB_OBJS += $(OUTPUT)util/thread.o
+LIB_OBJS += $(OUTPUT)util/trace-event-parse.o
+LIB_OBJS += $(OUTPUT)util/trace-event-read.o
+LIB_OBJS += $(OUTPUT)util/trace-event-info.o
+LIB_OBJS += $(OUTPUT)util/trace-event-scripting.o
+LIB_OBJS += $(OUTPUT)util/svghelper.o
+LIB_OBJS += $(OUTPUT)util/sort.o
+LIB_OBJS += $(OUTPUT)util/hist.o
+LIB_OBJS += $(OUTPUT)util/probe-event.o
+LIB_OBJS += $(OUTPUT)util/util.o
+LIB_OBJS += $(OUTPUT)util/xyarray.o
+LIB_OBJS += $(OUTPUT)util/cpumap.o
+
+BUILTIN_OBJS += $(OUTPUT)builtin-annotate.o
+
+BUILTIN_OBJS += $(OUTPUT)builtin-bench.o
+
+# Benchmark modules
+BUILTIN_OBJS += $(OUTPUT)bench/sched-messaging.o
+BUILTIN_OBJS += $(OUTPUT)bench/sched-pipe.o
+ifeq ($(RAW_ARCH),x86_64)
+BUILTIN_OBJS += $(OUTPUT)bench/mem-memcpy-x86-64-asm.o
+endif
+BUILTIN_OBJS += $(OUTPUT)bench/mem-memcpy.o
+
+BUILTIN_OBJS += $(OUTPUT)builtin-diff.o
+BUILTIN_OBJS += $(OUTPUT)builtin-help.o
+BUILTIN_OBJS += $(OUTPUT)builtin-sched.o
+BUILTIN_OBJS += $(OUTPUT)builtin-buildid-list.o
+BUILTIN_OBJS += $(OUTPUT)builtin-buildid-cache.o
+BUILTIN_OBJS += $(OUTPUT)builtin-list.o
+BUILTIN_OBJS += $(OUTPUT)builtin-record.o
+BUILTIN_OBJS += $(OUTPUT)builtin-report.o
+BUILTIN_OBJS += $(OUTPUT)builtin-stat.o
+BUILTIN_OBJS += $(OUTPUT)builtin-timechart.o
+BUILTIN_OBJS += $(OUTPUT)builtin-top.o
+BUILTIN_OBJS += $(OUTPUT)builtin-script.o
+BUILTIN_OBJS += $(OUTPUT)builtin-probe.o
+BUILTIN_OBJS += $(OUTPUT)builtin-kmem.o
+BUILTIN_OBJS += $(OUTPUT)builtin-lock.o
+BUILTIN_OBJS += $(OUTPUT)builtin-kvm.o
+BUILTIN_OBJS += $(OUTPUT)builtin-test.o
+BUILTIN_OBJS += $(OUTPUT)builtin-inject.o
 
 PERFLIBS = $(LIB_FILE)
-EXTLIBS =
 
 #
 # Platform specific tweaks
@@ -351,6 +520,16 @@ EXTLIBS =
 -include config.mak.autogen
 -include config.mak
 
+ifndef NO_DWARF
+FLAGS_DWARF=$(ALL_CFLAGS) -ldw -lelf $(ALL_LDFLAGS) $(EXTLIBS)
+ifneq ($(call try-cc,$(SOURCE_DWARF),$(FLAGS_DWARF)),y)
+        msg := $(warning No libdw.h found or old libdw.h found or elfutils is older than 0.138, disables dwarf support. Please install new elfutils-devel/libdw-dev);
+        NO_DWARF := 1
+endif # Dwarf support
+endif # NO_DWARF
+
+-include arch/$(ARCH)/Makefile
+
 ifeq ($(uname_S),Darwin)
         ifndef NO_FINK
                 ifeq ($(shell test -d /sw/lib && echo y),y)
@@ -367,6 +546,145 @@ ifeq ($(uname_S),Darwin)
         PTHREAD_LIBS =
 endif
 
+ifneq ($(OUTPUT),)
+        BASIC_CFLAGS += -I$(OUTPUT)
+endif
+
+FLAGS_LIBELF=$(ALL_CFLAGS) $(ALL_LDFLAGS) $(EXTLIBS)
+ifneq ($(call try-cc,$(SOURCE_LIBELF),$(FLAGS_LIBELF)),y)
+        FLAGS_GLIBC=$(ALL_CFLAGS) $(ALL_LDFLAGS)
+        ifneq ($(call try-cc,$(SOURCE_GLIBC),$(FLAGS_GLIBC)),y)
+                msg := $(error No gnu/libc-version.h found, please install glibc-dev[el]/glibc-static);
+        else
+                msg := $(error No libelf.h/libelf found, please install libelf-dev/elfutils-libelf-devel);
+        endif
+endif
+
+ifneq ($(call try-cc,$(SOURCE_ELF_MMAP),$(FLAGS_COMMON)),y)
+        BASIC_CFLAGS += -DLIBELF_NO_MMAP
+endif
+
+ifndef NO_DWARF
+ifeq ($(origin PERF_HAVE_DWARF_REGS), undefined)
+        msg := $(warning DWARF register mappings have not been defined for architecture $(ARCH), DWARF support disabled);
+else
+        BASIC_CFLAGS += -DDWARF_SUPPORT
+        EXTLIBS += -lelf -ldw
+        LIB_OBJS += $(OUTPUT)util/probe-finder.o
+endif # PERF_HAVE_DWARF_REGS
+endif # NO_DWARF
+
+ifdef NO_NEWT
+        BASIC_CFLAGS += -DNO_NEWT_SUPPORT
+else
+        FLAGS_NEWT=$(ALL_CFLAGS) $(ALL_LDFLAGS) $(EXTLIBS) -lnewt
+        ifneq ($(call try-cc,$(SOURCE_NEWT),$(FLAGS_NEWT)),y)
+                msg := $(warning newt not found, disables TUI support. Please install newt-devel or libnewt-dev);
+                BASIC_CFLAGS += -DNO_NEWT_SUPPORT
+        else
+                # Fedora has /usr/include/slang/slang.h, but ubuntu /usr/include/slang.h
+                BASIC_CFLAGS += -I/usr/include/slang
+                EXTLIBS += -lnewt -lslang
+                LIB_OBJS += $(OUTPUT)util/ui/setup.o
+                LIB_OBJS += $(OUTPUT)util/ui/browser.o
+                LIB_OBJS += $(OUTPUT)util/ui/browsers/annotate.o
+                LIB_OBJS += $(OUTPUT)util/ui/browsers/hists.o
+                LIB_OBJS += $(OUTPUT)util/ui/browsers/map.o
+                LIB_OBJS += $(OUTPUT)util/ui/helpline.o
+                LIB_OBJS += $(OUTPUT)util/ui/progress.o
+                LIB_OBJS += $(OUTPUT)util/ui/util.o
+                LIB_H += util/ui/browser.h
+                LIB_H += util/ui/browsers/map.h
+                LIB_H += util/ui/helpline.h
+                LIB_H += util/ui/libslang.h
+                LIB_H += util/ui/progress.h
+                LIB_H += util/ui/util.h
+        endif
+endif
+
+ifdef NO_LIBPERL
+        BASIC_CFLAGS += -DNO_LIBPERL
+else
+       PERL_EMBED_LDOPTS = $(shell perl -MExtUtils::Embed -e ldopts 2>/dev/null)
+       PERL_EMBED_LDFLAGS = $(call strip-libs,$(PERL_EMBED_LDOPTS))
+       PERL_EMBED_LIBADD = $(call grep-libs,$(PERL_EMBED_LDOPTS))
+        PERL_EMBED_CCOPTS = `perl -MExtUtils::Embed -e ccopts 2>/dev/null`
+        FLAGS_PERL_EMBED=$(PERL_EMBED_CCOPTS) $(PERL_EMBED_LDOPTS)
+
+        ifneq ($(call try-cc,$(SOURCE_PERL_EMBED),$(FLAGS_PERL_EMBED)),y)
+                BASIC_CFLAGS += -DNO_LIBPERL
+        else
+               ALL_LDFLAGS += $(PERL_EMBED_LDFLAGS)
+               EXTLIBS += $(PERL_EMBED_LIBADD)
+                LIB_OBJS += $(OUTPUT)util/scripting-engines/trace-event-perl.o
+                LIB_OBJS += $(OUTPUT)scripts/perl/Perf-Trace-Util/Context.o
+        endif
+endif
+
+ifdef NO_LIBPYTHON
+        BASIC_CFLAGS += -DNO_LIBPYTHON
+else
+       PYTHON_EMBED_LDOPTS = $(shell python-config --ldflags 2>/dev/null)
+       PYTHON_EMBED_LDFLAGS = $(call strip-libs,$(PYTHON_EMBED_LDOPTS))
+       PYTHON_EMBED_LIBADD = $(call grep-libs,$(PYTHON_EMBED_LDOPTS))
+        PYTHON_EMBED_CCOPTS = `python-config --cflags 2>/dev/null`
+        FLAGS_PYTHON_EMBED=$(PYTHON_EMBED_CCOPTS) $(PYTHON_EMBED_LDOPTS)
+        ifneq ($(call try-cc,$(SOURCE_PYTHON_EMBED),$(FLAGS_PYTHON_EMBED)),y)
+                BASIC_CFLAGS += -DNO_LIBPYTHON
+        else
+               ALL_LDFLAGS += $(PYTHON_EMBED_LDFLAGS)
+               EXTLIBS += $(PYTHON_EMBED_LIBADD)
+                LIB_OBJS += $(OUTPUT)util/scripting-engines/trace-event-python.o
+                LIB_OBJS += $(OUTPUT)scripts/python/Perf-Trace-Util/Context.o
+        endif
+endif
+
+ifdef NO_DEMANGLE
+        BASIC_CFLAGS += -DNO_DEMANGLE
+else
+        ifdef HAVE_CPLUS_DEMANGLE
+                EXTLIBS += -liberty
+                BASIC_CFLAGS += -DHAVE_CPLUS_DEMANGLE
+        else
+                FLAGS_BFD=$(ALL_CFLAGS) $(ALL_LDFLAGS) $(EXTLIBS) -lbfd
+                has_bfd := $(call try-cc,$(SOURCE_BFD),$(FLAGS_BFD))
+                ifeq ($(has_bfd),y)
+                        EXTLIBS += -lbfd
+                else
+                        FLAGS_BFD_IBERTY=$(FLAGS_BFD) -liberty
+                        has_bfd_iberty := $(call try-cc,$(SOURCE_BFD),$(FLAGS_BFD_IBERTY))
+                        ifeq ($(has_bfd_iberty),y)
+                                EXTLIBS += -lbfd -liberty
+                        else
+                                FLAGS_BFD_IBERTY_Z=$(FLAGS_BFD_IBERTY) -lz
+                                has_bfd_iberty_z := $(call try-cc,$(SOURCE_BFD),$(FLAGS_BFD_IBERTY_Z))
+                                ifeq ($(has_bfd_iberty_z),y)
+                                        EXTLIBS += -lbfd -liberty -lz
+                                else
+                                        FLAGS_CPLUS_DEMANGLE=$(ALL_CFLAGS) $(ALL_LDFLAGS) $(EXTLIBS) -liberty
+                                        has_cplus_demangle := $(call try-cc,$(SOURCE_CPLUS_DEMANGLE),$(FLAGS_CPLUS_DEMANGLE))
+                                        ifeq ($(has_cplus_demangle),y)
+                                                EXTLIBS += -liberty
+                                                BASIC_CFLAGS += -DHAVE_CPLUS_DEMANGLE
+                                        else
+                                                msg := $(warning No bfd.h/libbfd found, install binutils-dev[el]/zlib-static to gain symbol demangling)
+                                                BASIC_CFLAGS += -DNO_DEMANGLE
+                                        endif
+                                endif
+                        endif
+                endif
+        endif
+endif
+
+
+ifdef NO_STRLCPY
+        BASIC_CFLAGS += -DNO_STRLCPY
+else
+        ifneq ($(call try-cc,$(SOURCE_STRLCPY),),y)
+                BASIC_CFLAGS += -DNO_STRLCPY
+        endif
+endif
+
 ifndef CC_LD_DYNPATH
         ifdef NO_R_TO_GCC_LINKER
                 # Some gcc does not accept and pass -R to the linker to specify
@@ -377,12 +695,6 @@ ifndef CC_LD_DYNPATH
         endif
 endif
 
-ifdef ZLIB_PATH
-        BASIC_CFLAGS += -I$(ZLIB_PATH)/include
-        EXTLIBS += -L$(ZLIB_PATH)/$(lib) $(CC_LD_DYNPATH)$(ZLIB_PATH)/$(lib)
-endif
-EXTLIBS += -lz
-
 ifdef NEEDS_SOCKET
         EXTLIBS += -lsocket
 endif
@@ -412,53 +724,53 @@ ifdef NO_C99_FORMAT
 endif
 ifdef SNPRINTF_RETURNS_BOGUS
         COMPAT_CFLAGS += -DSNPRINTF_RETURNS_BOGUS
-        COMPAT_OBJS += compat/snprintf.o
+        COMPAT_OBJS += $(OUTPUT)compat/snprintf.o
 endif
 ifdef FREAD_READS_DIRECTORIES
         COMPAT_CFLAGS += -DFREAD_READS_DIRECTORIES
-        COMPAT_OBJS += compat/fopen.o
+        COMPAT_OBJS += $(OUTPUT)compat/fopen.o
 endif
 ifdef NO_SYMLINK_HEAD
         BASIC_CFLAGS += -DNO_SYMLINK_HEAD
 endif
 ifdef NO_STRCASESTR
         COMPAT_CFLAGS += -DNO_STRCASESTR
-        COMPAT_OBJS += compat/strcasestr.o
+        COMPAT_OBJS += $(OUTPUT)compat/strcasestr.o
 endif
 ifdef NO_STRTOUMAX
         COMPAT_CFLAGS += -DNO_STRTOUMAX
-        COMPAT_OBJS += compat/strtoumax.o
+        COMPAT_OBJS += $(OUTPUT)compat/strtoumax.o
 endif
 ifdef NO_STRTOULL
         COMPAT_CFLAGS += -DNO_STRTOULL
 endif
 ifdef NO_SETENV
         COMPAT_CFLAGS += -DNO_SETENV
-        COMPAT_OBJS += compat/setenv.o
+        COMPAT_OBJS += $(OUTPUT)compat/setenv.o
 endif
 ifdef NO_MKDTEMP
         COMPAT_CFLAGS += -DNO_MKDTEMP
-        COMPAT_OBJS += compat/mkdtemp.o
+        COMPAT_OBJS += $(OUTPUT)compat/mkdtemp.o
 endif
 ifdef NO_UNSETENV
         COMPAT_CFLAGS += -DNO_UNSETENV
-        COMPAT_OBJS += compat/unsetenv.o
+        COMPAT_OBJS += $(OUTPUT)compat/unsetenv.o
 endif
 ifdef NO_SYS_SELECT_H
         BASIC_CFLAGS += -DNO_SYS_SELECT_H
 endif
 ifdef NO_MMAP
         COMPAT_CFLAGS += -DNO_MMAP
-        COMPAT_OBJS += compat/mmap.o
+        COMPAT_OBJS += $(OUTPUT)compat/mmap.o
 else
         ifdef USE_WIN32_MMAP
                 COMPAT_CFLAGS += -DUSE_WIN32_MMAP
-                COMPAT_OBJS += compat/win32mmap.o
+                COMPAT_OBJS += $(OUTPUT)compat/win32mmap.o
         endif
 endif
 ifdef NO_PREAD
         COMPAT_CFLAGS += -DNO_PREAD
-        COMPAT_OBJS += compat/pread.o
+        COMPAT_OBJS += $(OUTPUT)compat/pread.o
 endif
 ifdef NO_FAST_WORKING_DIRECTORY
         BASIC_CFLAGS += -DNO_FAST_WORKING_DIRECTORY
@@ -480,10 +792,10 @@ else
 endif
 endif
 ifdef NO_INET_NTOP
-        LIB_OBJS += compat/inet_ntop.o
+        LIB_OBJS += $(OUTPUT)compat/inet_ntop.o
 endif
 ifdef NO_INET_PTON
-        LIB_OBJS += compat/inet_pton.o
+        LIB_OBJS += $(OUTPUT)compat/inet_pton.o
 endif
 
 ifdef NO_ICONV
@@ -500,15 +812,15 @@ endif
 
 ifdef PPC_SHA1
         SHA1_HEADER = "ppc/sha1.h"
-        LIB_OBJS += ppc/sha1.o ppc/sha1ppc.o
+        LIB_OBJS += $(OUTPUT)ppc/sha1.o ppc/sha1ppc.o
 else
 ifdef ARM_SHA1
         SHA1_HEADER = "arm/sha1.h"
-        LIB_OBJS += arm/sha1.o arm/sha1_arm.o
+        LIB_OBJS += $(OUTPUT)arm/sha1.o $(OUTPUT)arm/sha1_arm.o
 else
 ifdef MOZILLA_SHA1
         SHA1_HEADER = "mozilla-sha1/sha1.h"
-        LIB_OBJS += mozilla-sha1/sha1.o
+        LIB_OBJS += $(OUTPUT)mozilla-sha1/sha1.o
 else
         SHA1_HEADER = <openssl/sha.h>
         EXTLIBS += $(LIB_4_CRYPTO)
@@ -520,15 +832,15 @@ ifdef NO_PERL_MAKEMAKER
 endif
 ifdef NO_HSTRERROR
         COMPAT_CFLAGS += -DNO_HSTRERROR
-        COMPAT_OBJS += compat/hstrerror.o
+        COMPAT_OBJS += $(OUTPUT)compat/hstrerror.o
 endif
 ifdef NO_MEMMEM
         COMPAT_CFLAGS += -DNO_MEMMEM
-        COMPAT_OBJS += compat/memmem.o
+        COMPAT_OBJS += $(OUTPUT)compat/memmem.o
 endif
 ifdef INTERNAL_QSORT
         COMPAT_CFLAGS += -DINTERNAL_QSORT
-        COMPAT_OBJS += compat/qsort.o
+        COMPAT_OBJS += $(OUTPUT)compat/qsort.o
 endif
 ifdef RUNTIME_PREFIX
         COMPAT_CFLAGS += -DRUNTIME_PREFIX
@@ -559,6 +871,7 @@ ifndef V
         QUIET_CC       = @echo '   ' CC $@;
         QUIET_AR       = @echo '   ' AR $@;
         QUIET_LINK     = @echo '   ' LINK $@;
+        QUIET_MKDIR    = @echo '   ' MKDIR $@;
         QUIET_BUILT_IN = @echo '   ' BUILTIN $@;
         QUIET_GEN      = @echo '   ' GEN $@;
         QUIET_SUBDIR0  = +@subdir=
@@ -592,13 +905,14 @@ prefix_SQ = $(subst ','\'',$(prefix))
 SHELL_PATH_SQ = $(subst ','\'',$(SHELL_PATH))
 PERL_PATH_SQ = $(subst ','\'',$(PERL_PATH))
 
-LIBS = $(PERFLIBS) $(EXTLIBS)
+LIBS = -Wl,--whole-archive $(PERFLIBS) -Wl,--no-whole-archive $(EXTLIBS)
 
 BASIC_CFLAGS += -DSHA1_HEADER='$(SHA1_HEADER_SQ)' \
         $(COMPAT_CFLAGS)
 LIB_OBJS += $(COMPAT_OBJS)
 
 ALL_CFLAGS += $(BASIC_CFLAGS)
+ALL_CFLAGS += $(ARCH_CFLAGS)
 ALL_LDFLAGS += $(BASIC_LDFLAGS)
 
 export TAR INSTALL DESTDIR SHELL_PATH
@@ -608,7 +922,7 @@ export TAR INSTALL DESTDIR SHELL_PATH
 
 SHELL = $(SHELL_PATH)
 
-all:: shell_compatibility_test $(ALL_PROGRAMS) $(BUILT_INS) $(OTHER_PROGRAMS) PERF-BUILD-OPTIONS
+all:: shell_compatibility_test $(ALL_PROGRAMS) $(BUILT_INS) $(OTHER_PROGRAMS) $(OUTPUT)PERF-BUILD-OPTIONS
 ifneq (,$X)
         $(foreach p,$(patsubst %$X,%,$(filter %$X,$(ALL_PROGRAMS) $(BUILT_INS) perf$X)), test '$p' -ef '$p$X' || $(RM) '$p';)
 endif
@@ -620,45 +934,51 @@ please_set_SHELL_PATH_to_a_more_modern_shell:
 
 shell_compatibility_test: please_set_SHELL_PATH_to_a_more_modern_shell
 
-strip: $(PROGRAMS) perf$X
-        $(STRIP) $(STRIP_OPTS) $(PROGRAMS) perf$X
+strip: $(PROGRAMS) $(OUTPUT)perf$X
+        $(STRIP) $(STRIP_OPTS) $(PROGRAMS) $(OUTPUT)perf$X
 
-perf.o: perf.c common-cmds.h PERF-CFLAGS
+$(OUTPUT)perf.o: perf.c $(OUTPUT)common-cmds.h $(OUTPUT)PERF-CFLAGS
         $(QUIET_CC)$(CC) -DPERF_VERSION='"$(PERF_VERSION)"' \
                 '-DPERF_HTML_PATH="$(htmldir_SQ)"' \
-                $(ALL_CFLAGS) -c $(filter %.c,$^)
+                $(ALL_CFLAGS) -c $(filter %.c,$^) -o $@
+
+$(OUTPUT)perf$X: $(OUTPUT)perf.o $(BUILTIN_OBJS) $(PERFLIBS)
+        $(QUIET_LINK)$(CC) $(ALL_CFLAGS) $(ALL_LDFLAGS) $(OUTPUT)perf.o \
+               $(BUILTIN_OBJS) $(LIBS) -o $@
 
-perf$X: perf.o $(BUILTIN_OBJS) $(PERFLIBS)
-        $(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ perf.o \
-                $(BUILTIN_OBJS) $(ALL_LDFLAGS) $(LIBS)
+$(OUTPUT)builtin-help.o: builtin-help.c $(OUTPUT)common-cmds.h $(OUTPUT)PERF-CFLAGS
+        $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) \
+                '-DPERF_HTML_PATH="$(htmldir_SQ)"' \
+                '-DPERF_MAN_PATH="$(mandir_SQ)"' \
+                '-DPERF_INFO_PATH="$(infodir_SQ)"' $<
 
-builtin-help.o: builtin-help.c common-cmds.h PERF-CFLAGS
-        $(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) \
+$(OUTPUT)builtin-timechart.o: builtin-timechart.c $(OUTPUT)common-cmds.h $(OUTPUT)PERF-CFLAGS
+        $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) \
                 '-DPERF_HTML_PATH="$(htmldir_SQ)"' \
                 '-DPERF_MAN_PATH="$(mandir_SQ)"' \
                 '-DPERF_INFO_PATH="$(infodir_SQ)"' $<
 
-$(BUILT_INS): perf$X
+$(BUILT_INS): $(OUTPUT)perf$X
         $(QUIET_BUILT_IN)$(RM) $@ && \
         ln perf$X $@ 2>/dev/null || \
         ln -s perf$X $@ 2>/dev/null || \
         cp perf$X $@
 
-common-cmds.h: util/generate-cmdlist.sh command-list.txt
+$(OUTPUT)common-cmds.h: util/generate-cmdlist.sh command-list.txt
 
-common-cmds.h: $(wildcard Documentation/perf-*.txt)
-        $(QUIET_GEN)util/generate-cmdlist.sh > $@+ && mv $@+ $@
+$(OUTPUT)common-cmds.h: $(wildcard Documentation/perf-*.txt)
+        $(QUIET_GEN). util/generate-cmdlist.sh > $@+ && mv $@+ $@
 
 $(patsubst %.sh,%,$(SCRIPT_SH)) : % : %.sh
-        $(QUIET_GEN)$(RM) $@ $@+ && \
+        $(QUIET_GEN)$(RM) $(OUTPUT)$@ $(OUTPUT)$@+ && \
         sed -e '1s|#!.*/sh|#!$(SHELL_PATH_SQ)|' \
             -e 's|@SHELL_PATH@|$(SHELL_PATH_SQ)|' \
             -e 's|@@PERL@@|$(PERL_PATH_SQ)|g' \
             -e 's/@@PERF_VERSION@@/$(PERF_VERSION)/g' \
             -e 's/@@NO_CURL@@/$(NO_CURL)/g' \
-            $@.sh >$@+ && \
-        chmod +x $@+ && \
-        mv $@+ $@
+            $@.sh > $(OUTPUT)$@+ && \
+        chmod +x $(OUTPUT)$@+ && \
+        mv $(OUTPUT)$@+ $(OUTPUT)$@
 
 configure: configure.ac
         $(QUIET_GEN)$(RM) $@ $<+ && \
@@ -668,38 +988,73 @@ configure: configure.ac
         $(RM) $<+
 
 # These can record PERF_VERSION
-perf.o perf.spec \
+$(OUTPUT)perf.o perf.spec \
         $(patsubst %.sh,%,$(SCRIPT_SH)) \
         $(patsubst %.perl,%,$(SCRIPT_PERL)) \
-        : PERF-VERSION-FILE
+        : $(OUTPUT)PERF-VERSION-FILE
 
-%.o: %.c PERF-CFLAGS
-        $(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) $<
-%.s: %.c PERF-CFLAGS
+$(OUTPUT)%.o: %.c $(OUTPUT)PERF-CFLAGS
+        $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $<
+$(OUTPUT)%.s: %.c $(OUTPUT)PERF-CFLAGS
         $(QUIET_CC)$(CC) -S $(ALL_CFLAGS) $<
-%.o: %.S
-        $(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) $<
+$(OUTPUT)%.o: %.S
+        $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $<
 
-util/exec_cmd.o: util/exec_cmd.c PERF-CFLAGS
-        $(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) \
+$(OUTPUT)util/exec_cmd.o: util/exec_cmd.c $(OUTPUT)PERF-CFLAGS
+        $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) \
                 '-DPERF_EXEC_PATH="$(perfexecdir_SQ)"' \
                 '-DBINDIR="$(bindir_relative_SQ)"' \
                 '-DPREFIX="$(prefix_SQ)"' \
                 $<
 
-builtin-init-db.o: builtin-init-db.c PERF-CFLAGS
-        $(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) -DDEFAULT_PERF_TEMPLATE_DIR='"$(template_dir_SQ)"' $<
+$(OUTPUT)builtin-init-db.o: builtin-init-db.c $(OUTPUT)PERF-CFLAGS
+        $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DDEFAULT_PERF_TEMPLATE_DIR='"$(template_dir_SQ)"' $<
+
+$(OUTPUT)util/config.o: util/config.c $(OUTPUT)PERF-CFLAGS
+        $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $<
+
+$(OUTPUT)util/ui/browser.o: util/ui/browser.c $(OUTPUT)PERF-CFLAGS
+        $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DENABLE_SLFUTURE_CONST $<
 
-util/config.o: util/config.c PERF-CFLAGS
-        $(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $<
+$(OUTPUT)util/ui/browsers/annotate.o: util/ui/browsers/annotate.c $(OUTPUT)PERF-CFLAGS
+        $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DENABLE_SLFUTURE_CONST $<
 
-perf-%$X: %.o $(PERFLIBS)
+$(OUTPUT)util/ui/browsers/hists.o: util/ui/browsers/hists.c $(OUTPUT)PERF-CFLAGS
+        $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DENABLE_SLFUTURE_CONST $<
+
+$(OUTPUT)util/ui/browsers/map.o: util/ui/browsers/map.c $(OUTPUT)PERF-CFLAGS
+        $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DENABLE_SLFUTURE_CONST $<
+
+$(OUTPUT)util/rbtree.o: ../../lib/rbtree.c $(OUTPUT)PERF-CFLAGS
+        $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $<
+
+$(OUTPUT)util/scripting-engines/trace-event-perl.o: util/scripting-engines/trace-event-perl.c $(OUTPUT)PERF-CFLAGS
+        $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $(PERL_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-shadow $<
+
+$(OUTPUT)scripts/perl/Perf-Trace-Util/Context.o: scripts/perl/Perf-Trace-Util/Context.c $(OUTPUT)PERF-CFLAGS
+        $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $(PERL_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-nested-externs $<
+
+$(OUTPUT)util/scripting-engines/trace-event-python.o: util/scripting-engines/trace-event-python.c $(OUTPUT)PERF-CFLAGS
+        $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $(PYTHON_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-shadow $<
+
+$(OUTPUT)scripts/python/Perf-Trace-Util/Context.o: scripts/python/Perf-Trace-Util/Context.c $(OUTPUT)PERF-CFLAGS
+        $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $(PYTHON_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-nested-externs $<
+
+$(OUTPUT)perf-%$X: %.o $(PERFLIBS)
         $(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ $(ALL_LDFLAGS) $(filter %.o,$^) $(LIBS)
 
 $(LIB_OBJS) $(BUILTIN_OBJS): $(LIB_H)
 $(patsubst perf-%$X,%.o,$(PROGRAMS)): $(LIB_H) $(wildcard */*.h)
 builtin-revert.o wt-status.o: wt-status.h
 
+# we compile into subdirectories. if the target directory is not the source directory, they might not exists. So
+# we depend the various files onto their directories.
+DIRECTORY_DEPS = $(LIB_OBJS) $(BUILTIN_OBJS) $(OUTPUT)PERF-VERSION-FILE $(OUTPUT)common-cmds.h
+$(DIRECTORY_DEPS): | $(sort $(dir $(DIRECTORY_DEPS)))
+# In the second step, we make a rule to actually create these directories
+$(sort $(dir $(DIRECTORY_DEPS))):
+        $(QUIET_MKDIR)$(MKDIR) -p $@ 2>/dev/null
+
 $(LIB_FILE): $(LIB_OBJS)
         $(QUIET_AR)$(RM) $@ && $(AR) rcs $@ $(LIB_OBJS)
 
@@ -734,17 +1089,17 @@ cscope:
 TRACK_CFLAGS = $(subst ','\'',$(ALL_CFLAGS)):\
              $(bindir_SQ):$(perfexecdir_SQ):$(template_dir_SQ):$(prefix_SQ)
 
-PERF-CFLAGS: .FORCE-PERF-CFLAGS
+$(OUTPUT)PERF-CFLAGS: .FORCE-PERF-CFLAGS
         @FLAGS='$(TRACK_CFLAGS)'; \
-            if test x"$$FLAGS" != x"`cat PERF-CFLAGS 2>/dev/null`" ; then \
+            if test x"$$FLAGS" != x"`cat $(OUTPUT)PERF-CFLAGS 2>/dev/null`" ; then \
                 echo 1>&2 "    * new build flags or prefix"; \
-                echo "$$FLAGS" >PERF-CFLAGS; \
+                echo "$$FLAGS" >$(OUTPUT)PERF-CFLAGS; \
             fi
 
 # We need to apply sq twice, once to protect from the shell
-# that runs PERF-BUILD-OPTIONS, and then again to protect it
+# that runs $(OUTPUT)PERF-BUILD-OPTIONS, and then again to protect it
 # and the first level quoting from the shell that runs "echo".
-PERF-BUILD-OPTIONS: .FORCE-PERF-BUILD-OPTIONS
+$(OUTPUT)PERF-BUILD-OPTIONS: .FORCE-PERF-BUILD-OPTIONS
         @echo SHELL_PATH=\''$(subst ','\'',$(SHELL_PATH_SQ))'\' >$@
         @echo TAR=\''$(subst ','\'',$(subst ','\'',$(TAR)))'\' >>$@
         @echo NO_CURL=\''$(subst ','\'',$(subst ','\'',$(NO_CURL)))'\' >>$@
@@ -765,7 +1120,7 @@ all:: $(TEST_PROGRAMS)
 
 export NO_SVN_TESTS
 
-check: common-cmds.h
+check: $(OUTPUT)common-cmds.h
         if sparse; \
         then \
                 for i in *.c */*.c; \
@@ -799,12 +1154,24 @@ export perfexec_instdir
 
 install: all
         $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(bindir_SQ)'
-        $(INSTALL) perf$X '$(DESTDIR_SQ)$(bindir_SQ)'
+        $(INSTALL) $(OUTPUT)perf$X '$(DESTDIR_SQ)$(bindir_SQ)'
+        $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util/lib/Perf/Trace'
+        $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/bin'
+        $(INSTALL) $(OUTPUT)perf-archive -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)'
+        $(INSTALL) scripts/perl/Perf-Trace-Util/lib/Perf/Trace/* -t '$(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'
+        $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/Perf-Trace-Util/lib/Perf/Trace'
+        $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/bin'
+        $(INSTALL) scripts/python/Perf-Trace-Util/lib/Perf/Trace/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/Perf-Trace-Util/lib/Perf/Trace'
+        $(INSTALL) scripts/python/*.py -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python'
+        $(INSTALL) scripts/python/bin/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/bin'
+
 ifdef BUILT_INS
         $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)'
         $(INSTALL) $(BUILT_INS) '$(DESTDIR_SQ)$(perfexec_instdir_SQ)'
 ifneq (,$X)
-        $(foreach p,$(patsubst %$X,%,$(filter %$X,$(ALL_PROGRAMS) $(BUILT_INS) perf$X)), $(RM) '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/$p';)
+        $(foreach p,$(patsubst %$X,%,$(filter %$X,$(ALL_PROGRAMS) $(BUILT_INS) $(OUTPUT)perf$X)), $(RM) '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/$p';)
 endif
 endif
 
@@ -885,17 +1252,17 @@ distclean: clean
 #       $(RM) configure
 
 clean:
-        $(RM) *.o */*.o $(LIB_FILE)
+        $(RM) *.o */*.o */*/*.o */*/*/*.o $(LIB_FILE)
         $(RM) $(ALL_PROGRAMS) $(BUILT_INS) perf$X
         $(RM) $(TEST_PROGRAMS)
-        $(RM) *.spec *.pyc *.pyo */*.pyc */*.pyo common-cmds.h TAGS tags cscope*
+        $(RM) *.spec *.pyc *.pyo */*.pyc */*.pyo $(OUTPUT)common-cmds.h TAGS tags cscope*
         $(RM) -r autom4te.cache
         $(RM) config.log config.mak.autogen config.mak.append config.status config.cache
         $(RM) -r $(PERF_TARNAME) .doc-tmp-dir
         $(RM) $(PERF_TARNAME).tar.gz perf-core_$(PERF_VERSION)-*.tar.gz
         $(RM) $(htmldocs).tar.gz $(manpages).tar.gz
         $(MAKE) -C Documentation/ clean
-        $(RM) PERF-VERSION-FILE PERF-CFLAGS PERF-BUILD-OPTIONS
+        $(RM) $(OUTPUT)PERF-VERSION-FILE $(OUTPUT)PERF-CFLAGS $(OUTPUT)PERF-BUILD-OPTIONS
 
 .PHONY: all install clean strip
 .PHONY: shell_compatibility_test please_set_SHELL_PATH_to_a_more_modern_shell
diff --git a/tools/perf/arch/arm/Makefile b/tools/perf/arch/arm/Makefile
new file mode 100644
index 00000000000..15130b50dfe
--- /dev/null
+++ b/tools/perf/arch/arm/Makefile
@@ -0,0 +1,4 @@
+ifndef NO_DWARF
+PERF_HAVE_DWARF_REGS := 1
+LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/dwarf-regs.o
+endif
diff --git a/tools/perf/arch/arm/util/dwarf-regs.c b/tools/perf/arch/arm/util/dwarf-regs.c
new file mode 100644
index 00000000000..fff6450c8c9
--- /dev/null
+++ b/tools/perf/arch/arm/util/dwarf-regs.c
@@ -0,0 +1,64 @@
+/*
+ * Mapping of DWARF debug register numbers into register names.
+ *
+ * Copyright (C) 2010 Will Deacon, ARM Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <libio.h>
+#include <dwarf-regs.h>
+
+struct pt_regs_dwarfnum {
+        const char *name;
+        unsigned int dwarfnum;
+};
+
+#define STR(s) #s
+#define REG_DWARFNUM_NAME(r, num) {.name = r, .dwarfnum = num}
+#define GPR_DWARFNUM_NAME(num) \
+        {.name = STR(%r##num), .dwarfnum = num}
+#define REG_DWARFNUM_END {.name = NULL, .dwarfnum = 0}
+
+/*
+ * Reference:
+ * http://infocenter.arm.com/help/topic/com.arm.doc.ihi0040a/IHI0040A_aadwarf.pdf
+ */
+static const struct pt_regs_dwarfnum regdwarfnum_table[] = {
+        GPR_DWARFNUM_NAME(0),
+        GPR_DWARFNUM_NAME(1),
+        GPR_DWARFNUM_NAME(2),
+        GPR_DWARFNUM_NAME(3),
+        GPR_DWARFNUM_NAME(4),
+        GPR_DWARFNUM_NAME(5),
+        GPR_DWARFNUM_NAME(6),
+        GPR_DWARFNUM_NAME(7),
+        GPR_DWARFNUM_NAME(8),
+        GPR_DWARFNUM_NAME(9),
+        GPR_DWARFNUM_NAME(10),
+        REG_DWARFNUM_NAME("%fp", 11),
+        REG_DWARFNUM_NAME("%ip", 12),
+        REG_DWARFNUM_NAME("%sp", 13),
+        REG_DWARFNUM_NAME("%lr", 14),
+        REG_DWARFNUM_NAME("%pc", 15),
+        REG_DWARFNUM_END,
+};
+
+/**
+ * get_arch_regstr() - lookup register name from it's DWARF register number
+ * @n:  the DWARF register number
+ *
+ * get_arch_regstr() returns the name of the register in struct
+ * regdwarfnum_table from it's DWARF register number. If the register is not
+ * found in the table, this returns NULL;
+ */
+const char *get_arch_regstr(unsigned int n)
+{
+        const struct pt_regs_dwarfnum *roff;
+        for (roff = regdwarfnum_table; roff->name != NULL; roff++)
+                if (roff->dwarfnum == n)
+                        return roff->name;
+        return NULL;
+}
diff --git a/tools/perf/arch/powerpc/Makefile b/tools/perf/arch/powerpc/Makefile
new file mode 100644
index 00000000000..15130b50dfe
--- /dev/null
+++ b/tools/perf/arch/powerpc/Makefile
@@ -0,0 +1,4 @@
+ifndef NO_DWARF
+PERF_HAVE_DWARF_REGS := 1
+LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/dwarf-regs.o
+endif
diff --git a/tools/perf/arch/powerpc/util/dwarf-regs.c b/tools/perf/arch/powerpc/util/dwarf-regs.c
new file mode 100644
index 00000000000..48ae0c5e3f7
--- /dev/null
+++ b/tools/perf/arch/powerpc/util/dwarf-regs.c
@@ -0,0 +1,88 @@
+/*
+ * Mapping of DWARF debug register numbers into register names.
+ *
+ * Copyright (C) 2010 Ian Munsie, IBM Corporation.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <libio.h>
+#include <dwarf-regs.h>
+
+
+struct pt_regs_dwarfnum {
+        const char *name;
+        unsigned int dwarfnum;
+};
+
+#define STR(s) #s
+#define REG_DWARFNUM_NAME(r, num) {.name = r, .dwarfnum = num}
+#define GPR_DWARFNUM_NAME(num)  \
+        {.name = STR(%gpr##num), .dwarfnum = num}
+#define REG_DWARFNUM_END {.name = NULL, .dwarfnum = 0}
+
+/*
+ * Reference:
+ * http://refspecs.linuxfoundation.org/ELF/ppc64/PPC-elf64abi-1.9.html
+ */
+static const struct pt_regs_dwarfnum regdwarfnum_table[] = {
+        GPR_DWARFNUM_NAME(0),
+        GPR_DWARFNUM_NAME(1),
+        GPR_DWARFNUM_NAME(2),
+        GPR_DWARFNUM_NAME(3),
+        GPR_DWARFNUM_NAME(4),
+        GPR_DWARFNUM_NAME(5),
+        GPR_DWARFNUM_NAME(6),
+        GPR_DWARFNUM_NAME(7),
+        GPR_DWARFNUM_NAME(8),
+        GPR_DWARFNUM_NAME(9),
+        GPR_DWARFNUM_NAME(10),
+        GPR_DWARFNUM_NAME(11),
+        GPR_DWARFNUM_NAME(12),
+        GPR_DWARFNUM_NAME(13),
+        GPR_DWARFNUM_NAME(14),
+        GPR_DWARFNUM_NAME(15),
+        GPR_DWARFNUM_NAME(16),
+        GPR_DWARFNUM_NAME(17),
+        GPR_DWARFNUM_NAME(18),
+        GPR_DWARFNUM_NAME(19),
+        GPR_DWARFNUM_NAME(20),
+        GPR_DWARFNUM_NAME(21),
+        GPR_DWARFNUM_NAME(22),
+        GPR_DWARFNUM_NAME(23),
+        GPR_DWARFNUM_NAME(24),
+        GPR_DWARFNUM_NAME(25),
+        GPR_DWARFNUM_NAME(26),
+        GPR_DWARFNUM_NAME(27),
+        GPR_DWARFNUM_NAME(28),
+        GPR_DWARFNUM_NAME(29),
+        GPR_DWARFNUM_NAME(30),
+        GPR_DWARFNUM_NAME(31),
+        REG_DWARFNUM_NAME("%msr",   66),
+        REG_DWARFNUM_NAME("%ctr",   109),
+        REG_DWARFNUM_NAME("%link",  108),
+        REG_DWARFNUM_NAME("%xer",   101),
+        REG_DWARFNUM_NAME("%dar",   119),
+        REG_DWARFNUM_NAME("%dsisr", 118),
+        REG_DWARFNUM_END,
+};
+
+/**
+ * get_arch_regstr() - lookup register name from it's DWARF register number
+ * @n:  the DWARF register number
+ *
+ * get_arch_regstr() returns the name of the register in struct
+ * regdwarfnum_table from it's DWARF register number. If the register is not
+ * found in the table, this returns NULL;
+ */
+const char *get_arch_regstr(unsigned int n)
+{
+        const struct pt_regs_dwarfnum *roff;
+        for (roff = regdwarfnum_table; roff->name != NULL; roff++)
+                if (roff->dwarfnum == n)
+                        return roff->name;
+        return NULL;
+}
diff --git a/tools/perf/arch/s390/Makefile b/tools/perf/arch/s390/Makefile
new file mode 100644
index 00000000000..15130b50dfe
--- /dev/null
+++ b/tools/perf/arch/s390/Makefile
@@ -0,0 +1,4 @@
+ifndef NO_DWARF
+PERF_HAVE_DWARF_REGS := 1
+LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/dwarf-regs.o
+endif
diff --git a/tools/perf/arch/s390/util/dwarf-regs.c b/tools/perf/arch/s390/util/dwarf-regs.c
new file mode 100644
index 00000000000..e19653e025f
--- /dev/null
+++ b/tools/perf/arch/s390/util/dwarf-regs.c
@@ -0,0 +1,22 @@
+/*
+ * Mapping of DWARF debug register numbers into register names.
+ *
+ *    Copyright IBM Corp. 2010
+ *    Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>,
+ *
+ */
+
+#include <libio.h>
+#include <dwarf-regs.h>
+
+#define NUM_GPRS 16
+
+static const char *gpr_names[NUM_GPRS] = {
+        "%r0", "%r1",  "%r2",  "%r3",  "%r4",  "%r5",  "%r6",  "%r7",
+        "%r8", "%r9", "%r10", "%r11", "%r12", "%r13", "%r14", "%r15",
+};
+
+const char *get_arch_regstr(unsigned int n)
+{
+        return (n >= NUM_GPRS) ? NULL : gpr_names[n];
+}
diff --git a/tools/perf/arch/sh/Makefile b/tools/perf/arch/sh/Makefile
new file mode 100644
index 00000000000..15130b50dfe
--- /dev/null
+++ b/tools/perf/arch/sh/Makefile
@@ -0,0 +1,4 @@
+ifndef NO_DWARF
+PERF_HAVE_DWARF_REGS := 1
+LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/dwarf-regs.o
+endif
diff --git a/tools/perf/arch/sh/util/dwarf-regs.c b/tools/perf/arch/sh/util/dwarf-regs.c
new file mode 100644
index 00000000000..a11edb007a6
--- /dev/null
+++ b/tools/perf/arch/sh/util/dwarf-regs.c
@@ -0,0 +1,55 @@
+/*
+ * Mapping of DWARF debug register numbers into register names.
+ *
+ * Copyright (C) 2010 Matt Fleming <matt@console-pimps.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+
+#include <libio.h>
+#include <dwarf-regs.h>
+
+/*
+ * Generic dwarf analysis helpers
+ */
+
+#define SH_MAX_REGS 18
+const char *sh_regs_table[SH_MAX_REGS] = {
+        "r0",
+        "r1",
+        "r2",
+        "r3",
+        "r4",
+        "r5",
+        "r6",
+        "r7",
+        "r8",
+        "r9",
+        "r10",
+        "r11",
+        "r12",
+        "r13",
+        "r14",
+        "r15",
+        "pc",
+        "pr",
+};
+
+/* Return architecture dependent register string (for kprobe-tracer) */
+const char *get_arch_regstr(unsigned int n)
+{
+        return (n <= SH_MAX_REGS) ? sh_regs_table[n] : NULL;
+}
diff --git a/tools/perf/arch/sparc/Makefile b/tools/perf/arch/sparc/Makefile
new file mode 100644
index 00000000000..15130b50dfe
--- /dev/null
+++ b/tools/perf/arch/sparc/Makefile
@@ -0,0 +1,4 @@
+ifndef NO_DWARF
+PERF_HAVE_DWARF_REGS := 1
+LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/dwarf-regs.o
+endif
diff --git a/tools/perf/arch/sparc/util/dwarf-regs.c b/tools/perf/arch/sparc/util/dwarf-regs.c
new file mode 100644
index 00000000000..0ab88483720
--- /dev/null
+++ b/tools/perf/arch/sparc/util/dwarf-regs.c
@@ -0,0 +1,43 @@
+/*
+ * Mapping of DWARF debug register numbers into register names.
+ *
+ * Copyright (C) 2010 David S. Miller <davem@davemloft.net>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <libio.h>
+#include <dwarf-regs.h>
+
+#define SPARC_MAX_REGS  96
+
+const char *sparc_regs_table[SPARC_MAX_REGS] = {
+        "%g0", "%g1", "%g2", "%g3", "%g4", "%g5", "%g6", "%g7",
+        "%o0", "%o1", "%o2", "%o3", "%o4", "%o5", "%sp", "%o7",
+        "%l0", "%l1", "%l2", "%l3", "%l4", "%l5", "%l6", "%l7",
+        "%i0", "%i1", "%i2", "%i3", "%i4", "%i5", "%fp", "%i7",
+        "%f0", "%f1", "%f2", "%f3", "%f4", "%f5", "%f6", "%f7",
+        "%f8", "%f9", "%f10", "%f11", "%f12", "%f13", "%f14", "%f15",
+        "%f16", "%f17", "%f18", "%f19", "%f20", "%f21", "%f22", "%f23",
+        "%f24", "%f25", "%f26", "%f27", "%f28", "%f29", "%f30", "%f31",
+        "%f32", "%f33", "%f34", "%f35", "%f36", "%f37", "%f38", "%f39",
+        "%f40", "%f41", "%f42", "%f43", "%f44", "%f45", "%f46", "%f47",
+        "%f48", "%f49", "%f50", "%f51", "%f52", "%f53", "%f54", "%f55",
+        "%f56", "%f57", "%f58", "%f59", "%f60", "%f61", "%f62", "%f63",
+};
+
+/**
+ * get_arch_regstr() - lookup register name from it's DWARF register number
+ * @n:  the DWARF register number
+ *
+ * get_arch_regstr() returns the name of the register in struct
+ * regdwarfnum_table from it's DWARF register number. If the register is not
+ * found in the table, this returns NULL;
+ */
+const char *get_arch_regstr(unsigned int n)
+{
+        return (n <= SPARC_MAX_REGS) ? sparc_regs_table[n] : NULL;
+}
diff --git a/tools/perf/arch/x86/Makefile b/tools/perf/arch/x86/Makefile
new file mode 100644
index 00000000000..15130b50dfe
--- /dev/null
+++ b/tools/perf/arch/x86/Makefile
@@ -0,0 +1,4 @@
+ifndef NO_DWARF
+PERF_HAVE_DWARF_REGS := 1
+LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/dwarf-regs.o
+endif
diff --git a/tools/perf/arch/x86/util/dwarf-regs.c b/tools/perf/arch/x86/util/dwarf-regs.c
new file mode 100644
index 00000000000..a794d308192
--- /dev/null
+++ b/tools/perf/arch/x86/util/dwarf-regs.c
@@ -0,0 +1,75 @@
+/*
+ * dwarf-regs.c : Mapping of DWARF debug register numbers into register names.
+ * Extracted from probe-finder.c
+ *
+ * Written by Masami Hiramatsu <mhiramat@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+
+#include <libio.h>
+#include <dwarf-regs.h>
+
+/*
+ * Generic dwarf analysis helpers
+ */
+
+#define X86_32_MAX_REGS 8
+const char *x86_32_regs_table[X86_32_MAX_REGS] = {
+        "%ax",
+        "%cx",
+        "%dx",
+        "%bx",
+        "$stack",       /* Stack address instead of %sp */
+        "%bp",
+        "%si",
+        "%di",
+};
+
+#define X86_64_MAX_REGS 16
+const char *x86_64_regs_table[X86_64_MAX_REGS] = {
+        "%ax",
+        "%dx",
+        "%cx",
+        "%bx",
+        "%si",
+        "%di",
+        "%bp",
+        "%sp",
+        "%r8",
+        "%r9",
+        "%r10",
+        "%r11",
+        "%r12",
+        "%r13",
+        "%r14",
+        "%r15",
+};
+
+/* TODO: switching by dwarf address size */
+#ifdef __x86_64__
+#define ARCH_MAX_REGS X86_64_MAX_REGS
+#define arch_regs_table x86_64_regs_table
+#else
+#define ARCH_MAX_REGS X86_32_MAX_REGS
+#define arch_regs_table x86_32_regs_table
+#endif
+
+/* Return architecture dependent register string (for kprobe-tracer) */
+const char *get_arch_regstr(unsigned int n)
+{
+        return (n <= ARCH_MAX_REGS) ? arch_regs_table[n] : NULL;
+}
diff --git a/tools/perf/bench/bench.h b/tools/perf/bench/bench.h
new file mode 100644
index 00000000000..f7781c6267c
--- /dev/null
+++ b/tools/perf/bench/bench.h
@@ -0,0 +1,17 @@
+#ifndef BENCH_H
+#define BENCH_H
+
+extern int bench_sched_messaging(int argc, const char **argv, const char *prefix);
+extern int bench_sched_pipe(int argc, const char **argv, const char *prefix);
+extern int bench_mem_memcpy(int argc, const char **argv, const char *prefix __used);
+
+#define BENCH_FORMAT_DEFAULT_STR        "default"
+#define BENCH_FORMAT_DEFAULT            0
+#define BENCH_FORMAT_SIMPLE_STR         "simple"
+#define BENCH_FORMAT_SIMPLE             1
+
+#define BENCH_FORMAT_UNKNOWN            -1
+
+extern int bench_format;
+
+#endif
diff --git a/tools/perf/bench/mem-memcpy-arch.h b/tools/perf/bench/mem-memcpy-arch.h
new file mode 100644
index 00000000000..a72e36cb539
--- /dev/null
+++ b/tools/perf/bench/mem-memcpy-arch.h
@@ -0,0 +1,12 @@
+
+#ifdef ARCH_X86_64
+
+#define MEMCPY_FN(fn, name, desc)               \
+        extern void *fn(void *, const void *, size_t);
+
+#include "mem-memcpy-x86-64-asm-def.h"
+
+#undef MEMCPY_FN
+
+#endif
+
diff --git a/tools/perf/bench/mem-memcpy-x86-64-asm-def.h b/tools/perf/bench/mem-memcpy-x86-64-asm-def.h
new file mode 100644
index 00000000000..d588b87696f
--- /dev/null
+++ b/tools/perf/bench/mem-memcpy-x86-64-asm-def.h
@@ -0,0 +1,4 @@
+
+MEMCPY_FN(__memcpy,
+        "x86-64-unrolled",
+        "unrolled memcpy() in arch/x86/lib/memcpy_64.S")
diff --git a/tools/perf/bench/mem-memcpy-x86-64-asm.S b/tools/perf/bench/mem-memcpy-x86-64-asm.S
new file mode 100644
index 00000000000..a57b66e853c
--- /dev/null
+++ b/tools/perf/bench/mem-memcpy-x86-64-asm.S
@@ -0,0 +1,2 @@
+
+#include "../../../arch/x86/lib/memcpy_64.S"
diff --git a/tools/perf/bench/mem-memcpy.c b/tools/perf/bench/mem-memcpy.c
new file mode 100644
index 00000000000..db82021f4b9
--- /dev/null
+++ b/tools/perf/bench/mem-memcpy.c
@@ -0,0 +1,297 @@
+/*
+ * mem-memcpy.c
+ *
+ * memcpy: Simple memory copy in various ways
+ *
+ * Written by Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
+ */
+#include <ctype.h>
+
+#include "../perf.h"
+#include "../util/util.h"
+#include "../util/parse-options.h"
+#include "../util/header.h"
+#include "bench.h"
+#include "mem-memcpy-arch.h"
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/time.h>
+#include <errno.h>
+
+#define K 1024
+
+static const char       *length_str     = "1MB";
+static const char       *routine        = "default";
+static bool             use_clock;
+static int              clock_fd;
+static bool             only_prefault;
+static bool             no_prefault;
+
+static const struct option options[] = {
+        OPT_STRING('l', "length", &length_str, "1MB",
+                    "Specify length of memory to copy. "
+                    "available unit: B, MB, GB (upper and lower)"),
+        OPT_STRING('r', "routine", &routine, "default",
+                    "Specify routine to copy"),
+        OPT_BOOLEAN('c', "clock", &use_clock,
+                    "Use CPU clock for measuring"),
+        OPT_BOOLEAN('o', "only-prefault", &only_prefault,
+                    "Show only the result with page faults before memcpy()"),
+        OPT_BOOLEAN('n', "no-prefault", &no_prefault,
+                    "Show only the result without page faults before memcpy()"),
+        OPT_END()
+};
+
+typedef void *(*memcpy_t)(void *, const void *, size_t);
+
+struct routine {
+        const char *name;
+        const char *desc;
+        memcpy_t fn;
+};
+
+struct routine routines[] = {
+        { "default",
+          "Default memcpy() provided by glibc",
+          memcpy },
+#ifdef ARCH_X86_64
+
+#define MEMCPY_FN(fn, name, desc) { name, desc, fn },
+#include "mem-memcpy-x86-64-asm-def.h"
+#undef MEMCPY_FN
+
+#endif
+
+        { NULL,
+          NULL,
+          NULL   }
+};
+
+static const char * const bench_mem_memcpy_usage[] = {
+        "perf bench mem memcpy <options>",
+        NULL
+};
+
+static struct perf_event_attr clock_attr = {
+        .type           = PERF_TYPE_HARDWARE,
+        .config         = PERF_COUNT_HW_CPU_CYCLES
+};
+
+static void init_clock(void)
+{
+        clock_fd = sys_perf_event_open(&clock_attr, getpid(), -1, -1, 0);
+
+        if (clock_fd < 0 && errno == ENOSYS)
+                die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
+        else
+                BUG_ON(clock_fd < 0);
+}
+
+static u64 get_clock(void)
+{
+        int ret;
+        u64 clk;
+
+        ret = read(clock_fd, &clk, sizeof(u64));
+        BUG_ON(ret != sizeof(u64));
+
+        return clk;
+}
+
+static double timeval2double(struct timeval *ts)
+{
+        return (double)ts->tv_sec +
+                (double)ts->tv_usec / (double)1000000;
+}
+
+static void alloc_mem(void **dst, void **src, size_t length)
+{
+        *dst = zalloc(length);
+        if (!dst)
+                die("memory allocation failed - maybe length is too large?\n");
+
+        *src = zalloc(length);
+        if (!src)
+                die("memory allocation failed - maybe length is too large?\n");
+}
+
+static u64 do_memcpy_clock(memcpy_t fn, size_t len, bool prefault)
+{
+        u64 clock_start = 0ULL, clock_end = 0ULL;
+        void *src = NULL, *dst = NULL;
+
+        alloc_mem(&src, &dst, len);
+
+        if (prefault)
+                fn(dst, src, len);
+
+        clock_start = get_clock();
+        fn(dst, src, len);
+        clock_end = get_clock();
+
+        free(src);
+        free(dst);
+        return clock_end - clock_start;
+}
+
+static double do_memcpy_gettimeofday(memcpy_t fn, size_t len, bool prefault)
+{
+        struct timeval tv_start, tv_end, tv_diff;
+        void *src = NULL, *dst = NULL;
+
+        alloc_mem(&src, &dst, len);
+
+        if (prefault)
+                fn(dst, src, len);
+
+        BUG_ON(gettimeofday(&tv_start, NULL));
+        fn(dst, src, len);
+        BUG_ON(gettimeofday(&tv_end, NULL));
+
+        timersub(&tv_end, &tv_start, &tv_diff);
+
+        free(src);
+        free(dst);
+        return (double)((double)len / timeval2double(&tv_diff));
+}
+
+#define pf (no_prefault ? 0 : 1)
+
+#define print_bps(x) do {                                       \
+                if (x < K)                                      \
+                        printf(" %14lf B/Sec", x);              \
+                else if (x < K * K)                             \
+                        printf(" %14lfd KB/Sec", x / K);        \
+                else if (x < K * K * K)                         \
+                        printf(" %14lf MB/Sec", x / K / K);     \
+                else                                            \
+                        printf(" %14lf GB/Sec", x / K / K / K); \
+        } while (0)
+
+int bench_mem_memcpy(int argc, const char **argv,
+                     const char *prefix __used)
+{
+        int i;
+        size_t len;
+        double result_bps[2];
+        u64 result_clock[2];
+
+        argc = parse_options(argc, argv, options,
+                             bench_mem_memcpy_usage, 0);
+
+        if (use_clock)
+                init_clock();
+
+        len = (size_t)perf_atoll((char *)length_str);
+
+        result_clock[0] = result_clock[1] = 0ULL;
+        result_bps[0] = result_bps[1] = 0.0;
+
+        if ((s64)len <= 0) {
+                fprintf(stderr, "Invalid length:%s\n", length_str);
+                return 1;
+        }
+
+        /* same to without specifying either of prefault and no-prefault */
+        if (only_prefault && no_prefault)
+                only_prefault = no_prefault = false;
+
+        for (i = 0; routines[i].name; i++) {
+                if (!strcmp(routines[i].name, routine))
+                        break;
+        }
+        if (!routines[i].name) {
+                printf("Unknown routine:%s\n", routine);
+                printf("Available routines...\n");
+                for (i = 0; routines[i].name; i++) {
+                        printf("\t%s ... %s\n",
+                               routines[i].name, routines[i].desc);
+                }
+                return 1;
+        }
+
+        if (bench_format == BENCH_FORMAT_DEFAULT)
+                printf("# Copying %s Bytes ...\n\n", length_str);
+
+        if (!only_prefault && !no_prefault) {
+                /* show both of results */
+                if (use_clock) {
+                        result_clock[0] =
+                                do_memcpy_clock(routines[i].fn, len, false);
+                        result_clock[1] =
+                                do_memcpy_clock(routines[i].fn, len, true);
+                } else {
+                        result_bps[0] =
+                                do_memcpy_gettimeofday(routines[i].fn,
+                                                len, false);
+                        result_bps[1] =
+                                do_memcpy_gettimeofday(routines[i].fn,
+                                                len, true);
+                }
+        } else {
+                if (use_clock) {
+                        result_clock[pf] =
+                                do_memcpy_clock(routines[i].fn,
+                                                len, only_prefault);
+                } else {
+                        result_bps[pf] =
+                                do_memcpy_gettimeofday(routines[i].fn,
+                                                len, only_prefault);
+                }
+        }
+
+        switch (bench_format) {
+        case BENCH_FORMAT_DEFAULT:
+                if (!only_prefault && !no_prefault) {
+                        if (use_clock) {
+                                printf(" %14lf Clock/Byte\n",
+                                        (double)result_clock[0]
+                                        / (double)len);
+                                printf(" %14lf Clock/Byte (with prefault)\n",
+                                        (double)result_clock[1]
+                                        / (double)len);
+                        } else {
+                                print_bps(result_bps[0]);
+                                printf("\n");
+                                print_bps(result_bps[1]);
+                                printf(" (with prefault)\n");
+                        }
+                } else {
+                        if (use_clock) {
+                                printf(" %14lf Clock/Byte",
+                                        (double)result_clock[pf]
+                                        / (double)len);
+                        } else
+                                print_bps(result_bps[pf]);
+
+                        printf("%s\n", only_prefault ? " (with prefault)" : "");
+                }
+                break;
+        case BENCH_FORMAT_SIMPLE:
+                if (!only_prefault && !no_prefault) {
+                        if (use_clock) {
+                                printf("%lf %lf\n",
+                                        (double)result_clock[0] / (double)len,
+                                        (double)result_clock[1] / (double)len);
+                        } else {
+                                printf("%lf %lf\n",
+                                        result_bps[0], result_bps[1]);
+                        }
+                } else {
+                        if (use_clock) {
+                                printf("%lf\n", (double)result_clock[pf]
+                                        / (double)len);
+                        } else
+                                printf("%lf\n", result_bps[pf]);
+                }
+                break;
+        default:
+                /* reaching this means there's some disaster: */
+                die("unknown format: %d\n", bench_format);
+                break;
+        }
+
+        return 0;
+}
diff --git a/tools/perf/bench/sched-messaging.c b/tools/perf/bench/sched-messaging.c
new file mode 100644
index 00000000000..d1d1b30f99c
--- /dev/null
+++ b/tools/perf/bench/sched-messaging.c
@@ -0,0 +1,336 @@
+/*
+ *
+ * sched-messaging.c
+ *
+ * messaging: Benchmark for scheduler and IPC mechanisms
+ *
+ * Based on hackbench by Rusty Russell <rusty@rustcorp.com.au>
+ * Ported to perf by Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
+ *
+ */
+
+#include "../perf.h"
+#include "../util/util.h"
+#include "../util/parse-options.h"
+#include "../builtin.h"
+#include "bench.h"
+
+/* Test groups of 20 processes spraying to 20 receivers */
+#include <pthread.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <errno.h>
+#include <unistd.h>
+#include <sys/types.h>
+#include <sys/socket.h>
+#include <sys/wait.h>
+#include <sys/time.h>
+#include <sys/poll.h>
+#include <limits.h>
+
+#define DATASIZE 100
+
+static bool use_pipes = false;
+static unsigned int loops = 100;
+static bool thread_mode = false;
+static unsigned int num_groups = 10;
+
+struct sender_context {
+        unsigned int num_fds;
+        int ready_out;
+        int wakefd;
+        int out_fds[0];
+};
+
+struct receiver_context {
+        unsigned int num_packets;
+        int in_fds[2];
+        int ready_out;
+        int wakefd;
+};
+
+static void barf(const char *msg)
+{
+        fprintf(stderr, "%s (error: %s)\n", msg, strerror(errno));
+        exit(1);
+}
+
+static void fdpair(int fds[2])
+{
+        if (use_pipes) {
+                if (pipe(fds) == 0)
+                        return;
+        } else {
+                if (socketpair(AF_UNIX, SOCK_STREAM, 0, fds) == 0)
+                        return;
+        }
+
+        barf(use_pipes ? "pipe()" : "socketpair()");
+}
+
+/* Block until we're ready to go */
+static void ready(int ready_out, int wakefd)
+{
+        char dummy;
+        struct pollfd pollfd = { .fd = wakefd, .events = POLLIN };
+
+        /* Tell them we're ready. */
+        if (write(ready_out, &dummy, 1) != 1)
+                barf("CLIENT: ready write");
+
+        /* Wait for "GO" signal */
+        if (poll(&pollfd, 1, -1) != 1)
+                barf("poll");
+}
+
+/* Sender sprays loops messages down each file descriptor */
+static void *sender(struct sender_context *ctx)
+{
+        char data[DATASIZE];
+        unsigned int i, j;
+
+        ready(ctx->ready_out, ctx->wakefd);
+
+        /* Now pump to every receiver. */
+        for (i = 0; i < loops; i++) {
+                for (j = 0; j < ctx->num_fds; j++) {
+                        int ret, done = 0;
+
+again:
+                        ret = write(ctx->out_fds[j], data + done,
+                                    sizeof(data)-done);
+                        if (ret < 0)
+                                barf("SENDER: write");
+                        done += ret;
+                        if (done < DATASIZE)
+                                goto again;
+                }
+        }
+
+        return NULL;
+}
+
+
+/* One receiver per fd */
+static void *receiver(struct receiver_context* ctx)
+{
+        unsigned int i;
+
+        if (!thread_mode)
+                close(ctx->in_fds[1]);
+
+        /* Wait for start... */
+        ready(ctx->ready_out, ctx->wakefd);
+
+        /* Receive them all */
+        for (i = 0; i < ctx->num_packets; i++) {
+                char data[DATASIZE];
+                int ret, done = 0;
+
+again:
+                ret = read(ctx->in_fds[0], data + done, DATASIZE - done);
+                if (ret < 0)
+                        barf("SERVER: read");
+                done += ret;
+                if (done < DATASIZE)
+                        goto again;
+        }
+
+        return NULL;
+}
+
+static pthread_t create_worker(void *ctx, void *(*func)(void *))
+{
+        pthread_attr_t attr;
+        pthread_t childid;
+        int err;
+
+        if (!thread_mode) {
+                /* process mode */
+                /* Fork the receiver. */
+                switch (fork()) {
+                case -1:
+                        barf("fork()");
+                        break;
+                case 0:
+                        (*func) (ctx);
+                        exit(0);
+                        break;
+                default:
+                        break;
+                }
+
+                return (pthread_t)0;
+        }
+
+        if (pthread_attr_init(&attr) != 0)
+                barf("pthread_attr_init:");
+
+#ifndef __ia64__
+        if (pthread_attr_setstacksize(&attr, PTHREAD_STACK_MIN) != 0)
+                barf("pthread_attr_setstacksize");
+#endif
+
+        err = pthread_create(&childid, &attr, func, ctx);
+        if (err != 0) {
+                fprintf(stderr, "pthread_create failed: %s (%d)\n",
+                        strerror(err), err);
+                exit(-1);
+        }
+        return childid;
+}
+
+static void reap_worker(pthread_t id)
+{
+        int proc_status;
+        void *thread_status;
+
+        if (!thread_mode) {
+                /* process mode */
+                wait(&proc_status);
+                if (!WIFEXITED(proc_status))
+                        exit(1);
+        } else {
+                pthread_join(id, &thread_status);
+        }
+}
+
+/* One group of senders and receivers */
+static unsigned int group(pthread_t *pth,
+                unsigned int num_fds,
+                int ready_out,
+                int wakefd)
+{
+        unsigned int i;
+        struct sender_context *snd_ctx = malloc(sizeof(struct sender_context)
+                        + num_fds * sizeof(int));
+
+        if (!snd_ctx)
+                barf("malloc()");
+
+        for (i = 0; i < num_fds; i++) {
+                int fds[2];
+                struct receiver_context *ctx = malloc(sizeof(*ctx));
+
+                if (!ctx)
+                        barf("malloc()");
+
+
+                /* Create the pipe between client and server */
+                fdpair(fds);
+
+                ctx->num_packets = num_fds * loops;
+                ctx->in_fds[0] = fds[0];
+                ctx->in_fds[1] = fds[1];
+                ctx->ready_out = ready_out;
+                ctx->wakefd = wakefd;
+
+                pth[i] = create_worker(ctx, (void *)receiver);
+
+                snd_ctx->out_fds[i] = fds[1];
+                if (!thread_mode)
+                        close(fds[0]);
+        }
+
+        /* Now we have all the fds, fork the senders */
+        for (i = 0; i < num_fds; i++) {
+                snd_ctx->ready_out = ready_out;
+                snd_ctx->wakefd = wakefd;
+                snd_ctx->num_fds = num_fds;
+
+                pth[num_fds+i] = create_worker(snd_ctx, (void *)sender);
+        }
+
+        /* Close the fds we have left */
+        if (!thread_mode)
+                for (i = 0; i < num_fds; i++)
+                        close(snd_ctx->out_fds[i]);
+
+        /* Return number of children to reap */
+        return num_fds * 2;
+}
+
+static const struct option options[] = {
+        OPT_BOOLEAN('p', "pipe", &use_pipes,
+                    "Use pipe() instead of socketpair()"),
+        OPT_BOOLEAN('t', "thread", &thread_mode,
+                    "Be multi thread instead of multi process"),
+        OPT_UINTEGER('g', "group", &num_groups, "Specify number of groups"),
+        OPT_UINTEGER('l', "loop", &loops, "Specify number of loops"),
+        OPT_END()
+};
+
+static const char * const bench_sched_message_usage[] = {
+        "perf bench sched messaging <options>",
+        NULL
+};
+
+int bench_sched_messaging(int argc, const char **argv,
+                    const char *prefix __used)
+{
+        unsigned int i, total_children;
+        struct timeval start, stop, diff;
+        unsigned int num_fds = 20;
+        int readyfds[2], wakefds[2];
+        char dummy;
+        pthread_t *pth_tab;
+
+        argc = parse_options(argc, argv, options,
+                             bench_sched_message_usage, 0);
+
+        pth_tab = malloc(num_fds * 2 * num_groups * sizeof(pthread_t));
+        if (!pth_tab)
+                barf("main:malloc()");
+
+        fdpair(readyfds);
+        fdpair(wakefds);
+
+        total_children = 0;
+        for (i = 0; i < num_groups; i++)
+                total_children += group(pth_tab+total_children, num_fds,
+                                        readyfds[1], wakefds[0]);
+
+        /* Wait for everyone to be ready */
+        for (i = 0; i < total_children; i++)
+                if (read(readyfds[0], &dummy, 1) != 1)
+                        barf("Reading for readyfds");
+
+        gettimeofday(&start, NULL);
+
+        /* Kick them off */
+        if (write(wakefds[1], &dummy, 1) != 1)
+                barf("Writing to start them");
+
+        /* Reap them all */
+        for (i = 0; i < total_children; i++)
+                reap_worker(pth_tab[i]);
+
+        gettimeofday(&stop, NULL);
+
+        timersub(&stop, &start, &diff);
+
+        switch (bench_format) {
+        case BENCH_FORMAT_DEFAULT:
+                printf("# %d sender and receiver %s per group\n",
+                       num_fds, thread_mode ? "threads" : "processes");
+                printf("# %d groups == %d %s run\n\n",
+                       num_groups, num_groups * 2 * num_fds,
+                       thread_mode ? "threads" : "processes");
+                printf(" %14s: %lu.%03lu [sec]\n", "Total time",
+                       diff.tv_sec,
+                       (unsigned long) (diff.tv_usec/1000));
+                break;
+        case BENCH_FORMAT_SIMPLE:
+                printf("%lu.%03lu\n", diff.tv_sec,
+                       (unsigned long) (diff.tv_usec/1000));
+                break;
+        default:
+                /* reaching here is something disaster */
+                fprintf(stderr, "Unknown format:%d\n", bench_format);
+                exit(1);
+                break;
+        }
+
+        return 0;
+}
diff --git a/tools/perf/bench/sched-pipe.c b/tools/perf/bench/sched-pipe.c
new file mode 100644
index 00000000000..d9ab3ce446a
--- /dev/null
+++ b/tools/perf/bench/sched-pipe.c
@@ -0,0 +1,127 @@
+/*
+ *
+ * sched-pipe.c
+ *
+ * pipe: Benchmark for pipe()
+ *
+ * Based on pipe-test-1m.c by Ingo Molnar <mingo@redhat.com>
+ *  http://people.redhat.com/mingo/cfs-scheduler/tools/pipe-test-1m.c
+ * Ported to perf by Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
+ *
+ */
+
+#include "../perf.h"
+#include "../util/util.h"
+#include "../util/parse-options.h"
+#include "../builtin.h"
+#include "bench.h"
+
+#include <unistd.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <signal.h>
+#include <sys/wait.h>
+#include <linux/unistd.h>
+#include <string.h>
+#include <errno.h>
+#include <assert.h>
+#include <sys/time.h>
+#include <sys/types.h>
+
+#define LOOPS_DEFAULT 1000000
+static int loops = LOOPS_DEFAULT;
+
+static const struct option options[] = {
+        OPT_INTEGER('l', "loop", &loops,
+                    "Specify number of loops"),
+        OPT_END()
+};
+
+static const char * const bench_sched_pipe_usage[] = {
+        "perf bench sched pipe <options>",
+        NULL
+};
+
+int bench_sched_pipe(int argc, const char **argv,
+                     const char *prefix __used)
+{
+        int pipe_1[2], pipe_2[2];
+        int m = 0, i;
+        struct timeval start, stop, diff;
+        unsigned long long result_usec = 0;
+
+        /*
+         * why does "ret" exist?
+         * discarding returned value of read(), write()
+         * causes error in building environment for perf
+         */
+        int ret, wait_stat;
+        pid_t pid, retpid;
+
+        argc = parse_options(argc, argv, options,
+                             bench_sched_pipe_usage, 0);
+
+        assert(!pipe(pipe_1));
+        assert(!pipe(pipe_2));
+
+        pid = fork();
+        assert(pid >= 0);
+
+        gettimeofday(&start, NULL);
+
+        if (!pid) {
+                for (i = 0; i < loops; i++) {
+                        ret = read(pipe_1[0], &m, sizeof(int));
+                        ret = write(pipe_2[1], &m, sizeof(int));
+                }
+        } else {
+                for (i = 0; i < loops; i++) {
+                        ret = write(pipe_1[1], &m, sizeof(int));
+                        ret = read(pipe_2[0], &m, sizeof(int));
+                }
+        }
+
+        gettimeofday(&stop, NULL);
+        timersub(&stop, &start, &diff);
+
+        if (pid) {
+                retpid = waitpid(pid, &wait_stat, 0);
+                assert((retpid == pid) && WIFEXITED(wait_stat));
+        } else {
+                exit(0);
+        }
+
+        switch (bench_format) {
+        case BENCH_FORMAT_DEFAULT:
+                printf("# Executed %d pipe operations between two tasks\n\n",
+                        loops);
+
+                result_usec = diff.tv_sec * 1000000;
+                result_usec += diff.tv_usec;
+
+                printf(" %14s: %lu.%03lu [sec]\n\n", "Total time",
+                       diff.tv_sec,
+                       (unsigned long) (diff.tv_usec/1000));
+
+                printf(" %14lf usecs/op\n",
+                       (double)result_usec / (double)loops);
+                printf(" %14d ops/sec\n",
+                       (int)((double)loops /
+                             ((double)result_usec / (double)1000000)));
+                break;
+
+        case BENCH_FORMAT_SIMPLE:
+                printf("%lu.%03lu\n",
+                       diff.tv_sec,
+                       (unsigned long) (diff.tv_usec / 1000));
+                break;
+
+        default:
+                /* reaching here is something disaster */
+                fprintf(stderr, "Unknown format:%d\n", bench_format);
+                exit(1);
+                break;
+        }
+
+        return 0;
+}
diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c
index 7e58e3ad150..c056cdc0691 100644
--- a/tools/perf/builtin-annotate.c
+++ b/tools/perf/builtin-annotate.c
@@ -10,1119 +10,109 @@
 #include "util/util.h"
 
 #include "util/color.h"
-#include "util/list.h"
+#include <linux/list.h>
 #include "util/cache.h"
-#include "util/rbtree.h"
+#include <linux/rbtree.h>
 #include "util/symbol.h"
-#include "util/string.h"
 
 #include "perf.h"
+#include "util/debug.h"
 
+#include "util/event.h"
 #include "util/parse-options.h"
 #include "util/parse-events.h"
-
-#define SHOW_KERNEL     1
-#define SHOW_USER       2
-#define SHOW_HV         4
-
-#define MIN_GREEN               0.5
-#define MIN_RED         5.0
-
+#include "util/thread.h"
+#include "util/sort.h"
+#include "util/hist.h"
+#include "util/session.h"
 
 static char             const *input_name = "perf.data";
-static char             *vmlinux = "vmlinux";
-
-static char             default_sort_order[] = "comm,symbol";
-static char             *sort_order = default_sort_order;
-
-static int              input;
-static int              show_mask = SHOW_KERNEL | SHOW_USER | SHOW_HV;
-
-static int              dump_trace = 0;
-#define dprintf(x...)   do { if (dump_trace) printf(x); } while (0)
-
-static int              verbose;
-
-static int              print_line;
-
-static unsigned long    page_size;
-static unsigned long    mmap_window = 32;
-
-struct ip_event {
-        struct perf_event_header header;
-        u64 ip;
-        u32 pid, tid;
-};
-
-struct mmap_event {
-        struct perf_event_header header;
-        u32 pid, tid;
-        u64 start;
-        u64 len;
-        u64 pgoff;
-        char filename[PATH_MAX];
-};
-
-struct comm_event {
-        struct perf_event_header header;
-        u32 pid, tid;
-        char comm[16];
-};
-
-struct fork_event {
-        struct perf_event_header header;
-        u32 pid, ppid;
-};
-
-struct period_event {
-        struct perf_event_header header;
-        u64 time;
-        u64 id;
-        u64 sample_period;
-};
-
-typedef union event_union {
-        struct perf_event_header        header;
-        struct ip_event                 ip;
-        struct mmap_event               mmap;
-        struct comm_event               comm;
-        struct fork_event               fork;
-        struct period_event             period;
-} event_t;
-
-
-struct sym_ext {
-        struct rb_node  node;
-        double          percent;
-        char            *path;
-};
-
-static LIST_HEAD(dsos);
-static struct dso *kernel_dso;
-static struct dso *vdso;
-
-
-static void dsos__add(struct dso *dso)
-{
-        list_add_tail(&dso->node, &dsos);
-}
-
-static struct dso *dsos__find(const char *name)
-{
-        struct dso *pos;
-
-        list_for_each_entry(pos, &dsos, node)
-                if (strcmp(pos->name, name) == 0)
-                        return pos;
-        return NULL;
-}
-
-static struct dso *dsos__findnew(const char *name)
-{
-        struct dso *dso = dsos__find(name);
-        int nr;
-
-        if (dso)
-                return dso;
-
-        dso = dso__new(name, 0);
-        if (!dso)
-                goto out_delete_dso;
-
-        nr = dso__load(dso, NULL, verbose);
-        if (nr < 0) {
-                if (verbose)
-                        fprintf(stderr, "Failed to open: %s\n", name);
-                goto out_delete_dso;
-        }
-        if (!nr && verbose) {
-                fprintf(stderr,
-                "No symbols found in: %s, maybe install a debug package?\n",
-                                name);
-        }
-
-        dsos__add(dso);
-
-        return dso;
-
-out_delete_dso:
-        dso__delete(dso);
-        return NULL;
-}
-
-static void dsos__fprintf(FILE *fp)
-{
-        struct dso *pos;
-
-        list_for_each_entry(pos, &dsos, node)
-                dso__fprintf(pos, fp);
-}
-
-static struct symbol *vdso__find_symbol(struct dso *dso, u64 ip)
-{
-        return dso__find_symbol(kernel_dso, ip);
-}
-
-static int load_kernel(void)
-{
-        int err;
-
-        kernel_dso = dso__new("[kernel]", 0);
-        if (!kernel_dso)
-                return -1;
-
-        err = dso__load_kernel(kernel_dso, vmlinux, NULL, verbose);
-        if (err) {
-                dso__delete(kernel_dso);
-                kernel_dso = NULL;
-        } else
-                dsos__add(kernel_dso);
-
-        vdso = dso__new("[vdso]", 0);
-        if (!vdso)
-                return -1;
-
-        vdso->find_symbol = vdso__find_symbol;
-
-        dsos__add(vdso);
-
-        return err;
-}
-
-struct map {
-        struct list_head node;
-        u64      start;
-        u64      end;
-        u64      pgoff;
-        u64      (*map_ip)(struct map *, u64);
-        struct dso       *dso;
-};
-
-static u64 map__map_ip(struct map *map, u64 ip)
-{
-        return ip - map->start + map->pgoff;
-}
-
-static u64 vdso__map_ip(struct map *map, u64 ip)
-{
-        return ip;
-}
-
-static struct map *map__new(struct mmap_event *event)
-{
-        struct map *self = malloc(sizeof(*self));
-
-        if (self != NULL) {
-                const char *filename = event->filename;
-
-                self->start = event->start;
-                self->end   = event->start + event->len;
-                self->pgoff = event->pgoff;
-
-                self->dso = dsos__findnew(filename);
-                if (self->dso == NULL)
-                        goto out_delete;
-
-                if (self->dso == vdso)
-                        self->map_ip = vdso__map_ip;
-                else
-                        self->map_ip = map__map_ip;
-        }
-        return self;
-out_delete:
-        free(self);
-        return NULL;
-}
-
-static struct map *map__clone(struct map *self)
-{
-        struct map *map = malloc(sizeof(*self));
-
-        if (!map)
-                return NULL;
-
-        memcpy(map, self, sizeof(*self));
-
-        return map;
-}
-
-static int map__overlap(struct map *l, struct map *r)
-{
-        if (l->start > r->start) {
-                struct map *t = l;
-                l = r;
-                r = t;
-        }
-
-        if (l->end > r->start)
-                return 1;
-
-        return 0;
-}
-
-static size_t map__fprintf(struct map *self, FILE *fp)
-{
-        return fprintf(fp, " %Lx-%Lx %Lx %s\n",
-                       self->start, self->end, self->pgoff, self->dso->name);
-}
-
-
-struct thread {
-        struct rb_node   rb_node;
-        struct list_head maps;
-        pid_t            pid;
-        char             *comm;
-};
-
-static struct thread *thread__new(pid_t pid)
-{
-        struct thread *self = malloc(sizeof(*self));
-
-        if (self != NULL) {
-                self->pid = pid;
-                self->comm = malloc(32);
-                if (self->comm)
-                        snprintf(self->comm, 32, ":%d", self->pid);
-                INIT_LIST_HEAD(&self->maps);
-        }
-
-        return self;
-}
-
-static int thread__set_comm(struct thread *self, const char *comm)
-{
-        if (self->comm)
-                free(self->comm);
-        self->comm = strdup(comm);
-        return self->comm ? 0 : -ENOMEM;
-}
-
-static size_t thread__fprintf(struct thread *self, FILE *fp)
-{
-        struct map *pos;
-        size_t ret = fprintf(fp, "Thread %d %s\n", self->pid, self->comm);
-
-        list_for_each_entry(pos, &self->maps, node)
-                ret += map__fprintf(pos, fp);
-
-        return ret;
-}
-
-
-static struct rb_root threads;
-static struct thread *last_match;
-
-static struct thread *threads__findnew(pid_t pid)
-{
-        struct rb_node **p = &threads.rb_node;
-        struct rb_node *parent = NULL;
-        struct thread *th;
-
-        /*
-         * Font-end cache - PID lookups come in blocks,
-         * so most of the time we dont have to look up
-         * the full rbtree:
-         */
-        if (last_match && last_match->pid == pid)
-                return last_match;
-
-        while (*p != NULL) {
-                parent = *p;
-                th = rb_entry(parent, struct thread, rb_node);
-
-                if (th->pid == pid) {
-                        last_match = th;
-                        return th;
-                }
 
-                if (pid < th->pid)
-                        p = &(*p)->rb_left;
-                else
-                        p = &(*p)->rb_right;
-        }
-
-        th = thread__new(pid);
-        if (th != NULL) {
-                rb_link_node(&th->rb_node, parent, p);
-                rb_insert_color(&th->rb_node, &threads);
-                last_match = th;
-        }
-
-        return th;
-}
-
-static void thread__insert_map(struct thread *self, struct map *map)
-{
-        struct map *pos, *tmp;
-
-        list_for_each_entry_safe(pos, tmp, &self->maps, node) {
-                if (map__overlap(pos, map)) {
-                        list_del_init(&pos->node);
-                        /* XXX leaks dsos */
-                        free(pos);
-                }
-        }
-
-        list_add_tail(&map->node, &self->maps);
-}
-
-static int thread__fork(struct thread *self, struct thread *parent)
-{
-        struct map *map;
-
-        if (self->comm)
-                free(self->comm);
-        self->comm = strdup(parent->comm);
-        if (!self->comm)
-                return -ENOMEM;
-
-        list_for_each_entry(map, &parent->maps, node) {
-                struct map *new = map__clone(map);
-                if (!new)
-                        return -ENOMEM;
-                thread__insert_map(self, new);
-        }
-
-        return 0;
-}
+static bool             force, use_tui, use_stdio;
 
-static struct map *thread__find_map(struct thread *self, u64 ip)
-{
-        struct map *pos;
-
-        if (self == NULL)
-                return NULL;
-
-        list_for_each_entry(pos, &self->maps, node)
-                if (ip >= pos->start && ip <= pos->end)
-                        return pos;
-
-        return NULL;
-}
-
-static size_t threads__fprintf(FILE *fp)
-{
-        size_t ret = 0;
-        struct rb_node *nd;
+static bool             full_paths;
 
-        for (nd = rb_first(&threads); nd; nd = rb_next(nd)) {
-                struct thread *pos = rb_entry(nd, struct thread, rb_node);
+static bool             print_line;
 
-                ret += thread__fprintf(pos, fp);
-        }
-
-        return ret;
-}
-
-/*
- * histogram, sorted on item, collects counts
- */
-
-static struct rb_root hist;
-
-struct hist_entry {
-        struct rb_node   rb_node;
-
-        struct thread    *thread;
-        struct map       *map;
-        struct dso       *dso;
-        struct symbol    *sym;
-        u64      ip;
-        char             level;
-
-        uint32_t         count;
-};
-
-/*
- * configurable sorting bits
- */
-
-struct sort_entry {
-        struct list_head list;
-
-        char *header;
-
-        int64_t (*cmp)(struct hist_entry *, struct hist_entry *);
-        int64_t (*collapse)(struct hist_entry *, struct hist_entry *);
-        size_t  (*print)(FILE *fp, struct hist_entry *);
-};
-
-/* --sort pid */
-
-static int64_t
-sort__thread_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-        return right->thread->pid - left->thread->pid;
-}
-
-static size_t
-sort__thread_print(FILE *fp, struct hist_entry *self)
-{
-        return fprintf(fp, "%16s:%5d", self->thread->comm ?: "", self->thread->pid);
-}
-
-static struct sort_entry sort_thread = {
-        .header = "         Command:  Pid",
-        .cmp    = sort__thread_cmp,
-        .print  = sort__thread_print,
-};
-
-/* --sort comm */
-
-static int64_t
-sort__comm_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-        return right->thread->pid - left->thread->pid;
-}
+static const char *sym_hist_filter;
 
-static int64_t
-sort__comm_collapse(struct hist_entry *left, struct hist_entry *right)
+static int hists__add_entry(struct hists *self, struct addr_location *al)
 {
-        char *comm_l = left->thread->comm;
-        char *comm_r = right->thread->comm;
-
-        if (!comm_l || !comm_r) {
-                if (!comm_l && !comm_r)
-                        return 0;
-                else if (!comm_l)
-                        return -1;
-                else
-                        return 1;
-        }
-
-        return strcmp(comm_l, comm_r);
-}
-
-static size_t
-sort__comm_print(FILE *fp, struct hist_entry *self)
-{
-        return fprintf(fp, "%16s", self->thread->comm);
-}
-
-static struct sort_entry sort_comm = {
-        .header         = "         Command",
-        .cmp            = sort__comm_cmp,
-        .collapse       = sort__comm_collapse,
-        .print          = sort__comm_print,
-};
-
-/* --sort dso */
-
-static int64_t
-sort__dso_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-        struct dso *dso_l = left->dso;
-        struct dso *dso_r = right->dso;
-
-        if (!dso_l || !dso_r) {
-                if (!dso_l && !dso_r)
-                        return 0;
-                else if (!dso_l)
-                        return -1;
-                else
-                        return 1;
-        }
-
-        return strcmp(dso_l->name, dso_r->name);
-}
-
-static size_t
-sort__dso_print(FILE *fp, struct hist_entry *self)
-{
-        if (self->dso)
-                return fprintf(fp, "%-25s", self->dso->name);
-
-        return fprintf(fp, "%016llx         ", (u64)self->ip);
-}
-
-static struct sort_entry sort_dso = {
-        .header = "Shared Object            ",
-        .cmp    = sort__dso_cmp,
-        .print  = sort__dso_print,
-};
-
-/* --sort symbol */
-
-static int64_t
-sort__sym_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-        u64 ip_l, ip_r;
-
-        if (left->sym == right->sym)
-                return 0;
-
-        ip_l = left->sym ? left->sym->start : left->ip;
-        ip_r = right->sym ? right->sym->start : right->ip;
-
-        return (int64_t)(ip_r - ip_l);
-}
-
-static size_t
-sort__sym_print(FILE *fp, struct hist_entry *self)
-{
-        size_t ret = 0;
-
-        if (verbose)
-                ret += fprintf(fp, "%#018llx  ", (u64)self->ip);
-
-        if (self->sym) {
-                ret += fprintf(fp, "[%c] %s",
-                        self->dso == kernel_dso ? 'k' : '.', self->sym->name);
-        } else {
-                ret += fprintf(fp, "%#016llx", (u64)self->ip);
-        }
-
-        return ret;
-}
-
-static struct sort_entry sort_sym = {
-        .header = "Symbol",
-        .cmp    = sort__sym_cmp,
-        .print  = sort__sym_print,
-};
-
-static int sort__need_collapse = 0;
-
-struct sort_dimension {
-        char                    *name;
-        struct sort_entry       *entry;
-        int                     taken;
-};
-
-static struct sort_dimension sort_dimensions[] = {
-        { .name = "pid",        .entry = &sort_thread,  },
-        { .name = "comm",       .entry = &sort_comm,    },
-        { .name = "dso",        .entry = &sort_dso,     },
-        { .name = "symbol",     .entry = &sort_sym,     },
-};
-
-static LIST_HEAD(hist_entry__sort_list);
-
-static int sort_dimension__add(char *tok)
-{
-        int i;
-
-        for (i = 0; i < ARRAY_SIZE(sort_dimensions); i++) {
-                struct sort_dimension *sd = &sort_dimensions[i];
-
-                if (sd->taken)
-                        continue;
-
-                if (strncasecmp(tok, sd->name, strlen(tok)))
-                        continue;
-
-                if (sd->entry->collapse)
-                        sort__need_collapse = 1;
-
-                list_add_tail(&sd->entry->list, &hist_entry__sort_list);
-                sd->taken = 1;
-
-                return 0;
-        }
-
-        return -ESRCH;
-}
-
-static int64_t
-hist_entry__cmp(struct hist_entry *left, struct hist_entry *right)
-{
-        struct sort_entry *se;
-        int64_t cmp = 0;
-
-        list_for_each_entry(se, &hist_entry__sort_list, list) {
-                cmp = se->cmp(left, right);
-                if (cmp)
-                        break;
-        }
-
-        return cmp;
-}
-
-static int64_t
-hist_entry__collapse(struct hist_entry *left, struct hist_entry *right)
-{
-        struct sort_entry *se;
-        int64_t cmp = 0;
-
-        list_for_each_entry(se, &hist_entry__sort_list, list) {
-                int64_t (*f)(struct hist_entry *, struct hist_entry *);
-
-                f = se->collapse ?: se->cmp;
-
-                cmp = f(left, right);
-                if (cmp)
-                        break;
-        }
-
-        return cmp;
-}
-
-/*
- * collect histogram counts
- */
-static void hist_hit(struct hist_entry *he, u64 ip)
-{
-        unsigned int sym_size, offset;
-        struct symbol *sym = he->sym;
-
-        he->count++;
-
-        if (!sym || !sym->hist)
-                return;
-
-        sym_size = sym->end - sym->start;
-        offset = ip - sym->start;
-
-        if (offset >= sym_size)
-                return;
-
-        sym->hist_sum++;
-        sym->hist[offset]++;
-
-        if (verbose >= 3)
-                printf("%p %s: count++ [ip: %p, %08Lx] => %Ld\n",
-                        (void *)(unsigned long)he->sym->start,
-                        he->sym->name,
-                        (void *)(unsigned long)ip, ip - he->sym->start,
-                        sym->hist[offset]);
-}
-
-static int
-hist_entry__add(struct thread *thread, struct map *map, struct dso *dso,
-                struct symbol *sym, u64 ip, char level)
-{
-        struct rb_node **p = &hist.rb_node;
-        struct rb_node *parent = NULL;
         struct hist_entry *he;
-        struct hist_entry entry = {
-                .thread = thread,
-                .map    = map,
-                .dso    = dso,
-                .sym    = sym,
-                .ip     = ip,
-                .level  = level,
-                .count  = 1,
-        };
-        int cmp;
 
-        while (*p != NULL) {
-                parent = *p;
-                he = rb_entry(parent, struct hist_entry, rb_node);
-
-                cmp = hist_entry__cmp(&entry, he);
-
-                if (!cmp) {
-                        hist_hit(he, ip);
-
-                        return 0;
+        if (sym_hist_filter != NULL &&
+            (al->sym == NULL || strcmp(sym_hist_filter, al->sym->name) != 0)) {
+                /* We're only interested in a symbol named sym_hist_filter */
+                if (al->sym != NULL) {
+                        rb_erase(&al->sym->rb_node,
+                                 &al->map->dso->symbols[al->map->type]);
+                        symbol__delete(al->sym);
                 }
-
-                if (cmp < 0)
-                        p = &(*p)->rb_left;
-                else
-                        p = &(*p)->rb_right;
-        }
-
-        he = malloc(sizeof(*he));
-        if (!he)
-                return -ENOMEM;
-        *he = entry;
-        rb_link_node(&he->rb_node, parent, p);
-        rb_insert_color(&he->rb_node, &hist);
-
-        return 0;
-}
-
-static void hist_entry__free(struct hist_entry *he)
-{
-        free(he);
-}
-
-/*
- * collapse the histogram
- */
-
-static struct rb_root collapse_hists;
-
-static void collapse__insert_entry(struct hist_entry *he)
-{
-        struct rb_node **p = &collapse_hists.rb_node;
-        struct rb_node *parent = NULL;
-        struct hist_entry *iter;
-        int64_t cmp;
-
-        while (*p != NULL) {
-                parent = *p;
-                iter = rb_entry(parent, struct hist_entry, rb_node);
-
-                cmp = hist_entry__collapse(iter, he);
-
-                if (!cmp) {
-                        iter->count += he->count;
-                        hist_entry__free(he);
-                        return;
-                }
-
-                if (cmp < 0)
-                        p = &(*p)->rb_left;
-                else
-                        p = &(*p)->rb_right;
-        }
-
-        rb_link_node(&he->rb_node, parent, p);
-        rb_insert_color(&he->rb_node, &collapse_hists);
-}
-
-static void collapse__resort(void)
-{
-        struct rb_node *next;
-        struct hist_entry *n;
-
-        if (!sort__need_collapse)
-                return;
-
-        next = rb_first(&hist);
-        while (next) {
-                n = rb_entry(next, struct hist_entry, rb_node);
-                next = rb_next(&n->rb_node);
-
-                rb_erase(&n->rb_node, &hist);
-                collapse__insert_entry(n);
-        }
-}
-
-/*
- * reverse the map, sort on count.
- */
-
-static struct rb_root output_hists;
-
-static void output__insert_entry(struct hist_entry *he)
-{
-        struct rb_node **p = &output_hists.rb_node;
-        struct rb_node *parent = NULL;
-        struct hist_entry *iter;
-
-        while (*p != NULL) {
-                parent = *p;
-                iter = rb_entry(parent, struct hist_entry, rb_node);
-
-                if (he->count > iter->count)
-                        p = &(*p)->rb_left;
-                else
-                        p = &(*p)->rb_right;
-        }
-
-        rb_link_node(&he->rb_node, parent, p);
-        rb_insert_color(&he->rb_node, &output_hists);
-}
-
-static void output__resort(void)
-{
-        struct rb_node *next;
-        struct hist_entry *n;
-        struct rb_root *tree = &hist;
-
-        if (sort__need_collapse)
-                tree = &collapse_hists;
-
-        next = rb_first(tree);
-
-        while (next) {
-                n = rb_entry(next, struct hist_entry, rb_node);
-                next = rb_next(&n->rb_node);
-
-                rb_erase(&n->rb_node, tree);
-                output__insert_entry(n);
-        }
-}
-
-static void register_idle_thread(void)
-{
-        struct thread *thread = threads__findnew(0);
-
-        if (thread == NULL ||
-                        thread__set_comm(thread, "[idle]")) {
-                fprintf(stderr, "problem inserting idle task.\n");
-                exit(-1);
-        }
-}
-
-static unsigned long total = 0,
-                     total_mmap = 0,
-                     total_comm = 0,
-                     total_fork = 0,
-                     total_unknown = 0;
-
-static int
-process_overflow_event(event_t *event, unsigned long offset, unsigned long head)
-{
-        char level;
-        int show = 0;
-        struct dso *dso = NULL;
-        struct thread *thread = threads__findnew(event->ip.pid);
-        u64 ip = event->ip.ip;
-        struct map *map = NULL;
-
-        dprintf("%p [%p]: PERF_EVENT (IP, %d): %d: %p\n",
-                (void *)(offset + head),
-                (void *)(long)(event->header.size),
-                event->header.misc,
-                event->ip.pid,
-                (void *)(long)ip);
-
-        dprintf(" ... thread: %s:%d\n", thread->comm, thread->pid);
-
-        if (thread == NULL) {
-                fprintf(stderr, "problem processing %d event, skipping it.\n",
-                        event->header.type);
-                return -1;
-        }
-
-        if (event->header.misc & PERF_EVENT_MISC_KERNEL) {
-                show = SHOW_KERNEL;
-                level = 'k';
-
-                dso = kernel_dso;
-
-                dprintf(" ...... dso: %s\n", dso->name);
-
-        } else if (event->header.misc & PERF_EVENT_MISC_USER) {
-
-                show = SHOW_USER;
-                level = '.';
-
-                map = thread__find_map(thread, ip);
-                if (map != NULL) {
-                        ip = map->map_ip(map, ip);
-                        dso = map->dso;
-                } else {
-                        /*
-                         * If this is outside of all known maps,
-                         * and is a negative address, try to look it
-                         * up in the kernel dso, as it might be a
-                         * vsyscall (which executes in user-mode):
-                         */
-                        if ((long long)ip < 0)
-                                dso = kernel_dso;
-                }
-                dprintf(" ...... dso: %s\n", dso ? dso->name : "<not found>");
-
-        } else {
-                show = SHOW_HV;
-                level = 'H';
-                dprintf(" ...... dso: [hypervisor]\n");
-        }
-
-        if (show & show_mask) {
-                struct symbol *sym = NULL;
-
-                if (dso)
-                        sym = dso->find_symbol(dso, ip);
-
-                if (hist_entry__add(thread, map, dso, sym, ip, level)) {
-                        fprintf(stderr,
-                "problem incrementing symbol count, skipping event\n");
-                        return -1;
-                }
-        }
-        total++;
-
-        return 0;
-}
-
-static int
-process_mmap_event(event_t *event, unsigned long offset, unsigned long head)
-{
-        struct thread *thread = threads__findnew(event->mmap.pid);
-        struct map *map = map__new(&event->mmap);
-
-        dprintf("%p [%p]: PERF_EVENT_MMAP %d: [%p(%p) @ %p]: %s\n",
-                (void *)(offset + head),
-                (void *)(long)(event->header.size),
-                event->mmap.pid,
-                (void *)(long)event->mmap.start,
-                (void *)(long)event->mmap.len,
-                (void *)(long)event->mmap.pgoff,
-                event->mmap.filename);
-
-        if (thread == NULL || map == NULL) {
-                dprintf("problem processing PERF_EVENT_MMAP, skipping event.\n");
                 return 0;
         }
 
-        thread__insert_map(thread, map);
-        total_mmap++;
-
-        return 0;
-}
-
-static int
-process_comm_event(event_t *event, unsigned long offset, unsigned long head)
-{
-        struct thread *thread = threads__findnew(event->comm.pid);
-
-        dprintf("%p [%p]: PERF_EVENT_COMM: %s:%d\n",
-                (void *)(offset + head),
-                (void *)(long)(event->header.size),
-                event->comm.comm, event->comm.pid);
-
-        if (thread == NULL ||
-            thread__set_comm(thread, event->comm.comm)) {
-                dprintf("problem processing PERF_EVENT_COMM, skipping event.\n");
-                return -1;
-        }
-        total_comm++;
+        he = __hists__add_entry(self, al, NULL, 1);
+        if (he == NULL)
+                return -ENOMEM;
 
-        return 0;
+        return hist_entry__inc_addr_samples(he, al->addr);
 }
 
-static int
-process_fork_event(event_t *event, unsigned long offset, unsigned long head)
+static int process_sample_event(event_t *event, struct sample_data *sample,
+                                struct perf_session *session)
 {
-        struct thread *thread = threads__findnew(event->fork.pid);
-        struct thread *parent = threads__findnew(event->fork.ppid);
-
-        dprintf("%p [%p]: PERF_EVENT_FORK: %d:%d\n",
-                (void *)(offset + head),
-                (void *)(long)(event->header.size),
-                event->fork.pid, event->fork.ppid);
+        struct addr_location al;
 
-        if (!thread || !parent || thread__fork(thread, parent)) {
-                dprintf("problem processing PERF_EVENT_FORK, skipping event.\n");
+        if (event__preprocess_sample(event, session, &al, sample, NULL) < 0) {
+                pr_warning("problem processing %d event, skipping it.\n",
+                           event->header.type);
                 return -1;
         }
-        total_fork++;
-
-        return 0;
-}
-
-static int
-process_period_event(event_t *event, unsigned long offset, unsigned long head)
-{
-        dprintf("%p [%p]: PERF_EVENT_PERIOD: time:%Ld, id:%Ld: period:%Ld\n",
-                (void *)(offset + head),
-                (void *)(long)(event->header.size),
-                event->period.time,
-                event->period.id,
-                event->period.sample_period);
-
-        return 0;
-}
-
-static int
-process_event(event_t *event, unsigned long offset, unsigned long head)
-{
-        if (event->header.misc & PERF_EVENT_MISC_OVERFLOW)
-                return process_overflow_event(event, offset, head);
-
-        switch (event->header.type) {
-        case PERF_EVENT_MMAP:
-                return process_mmap_event(event, offset, head);
-
-        case PERF_EVENT_COMM:
-                return process_comm_event(event, offset, head);
-
-        case PERF_EVENT_FORK:
-                return process_fork_event(event, offset, head);
 
-        case PERF_EVENT_PERIOD:
-                return process_period_event(event, offset, head);
-        /*
-         * We dont process them right now but they are fine:
-         */
-
-        case PERF_EVENT_THROTTLE:
-        case PERF_EVENT_UNTHROTTLE:
-                return 0;
-
-        default:
+        if (!al.filtered && hists__add_entry(&session->hists, &al)) {
+                pr_warning("problem incrementing symbol count, "
+                           "skipping event\n");
                 return -1;
         }
 
         return 0;
 }
 
-static char *get_color(double percent)
-{
-        char *color = PERF_COLOR_NORMAL;
-
-        /*
-         * We color high-overhead entries in red, mid-overhead
-         * entries in green - and keep the low overhead places
-         * normal:
-         */
-        if (percent >= MIN_RED)
-                color = PERF_COLOR_RED;
-        else {
-                if (percent > MIN_GREEN)
-                        color = PERF_COLOR_GREEN;
-        }
-        return color;
-}
-
-static int
-parse_line(FILE *file, struct symbol *sym, u64 start, u64 len)
+static int objdump_line__print(struct objdump_line *self,
+                               struct list_head *head,
+                               struct hist_entry *he, u64 len)
 {
-        char *line = NULL, *tmp, *tmp2;
+        struct symbol *sym = he->ms.sym;
         static const char *prev_line;
         static const char *prev_color;
-        unsigned int offset;
-        size_t line_len;
-        u64 line_ip;
-        int ret;
-        char *c;
 
-        if (getline(&line, &line_len, file) < 0)
-                return -1;
-        if (!line)
-                return -1;
-
-        c = strchr(line, '\n');
-        if (c)
-                *c = 0;
-
-        line_ip = -1;
-        offset = 0;
-        ret = -2;
-
-        /*
-         * Strip leading spaces:
-         */
-        tmp = line;
-        while (*tmp) {
-                if (*tmp != ' ')
-                        break;
-                tmp++;
-        }
-
-        if (*tmp) {
-                /*
-                 * Parse hexa addresses followed by ':'
-                 */
-                line_ip = strtoull(tmp, &tmp2, 16);
-                if (*tmp2 != ':')
-                        line_ip = -1;
-        }
-
-        if (line_ip != -1) {
+        if (self->offset != -1) {
                 const char *path = NULL;
                 unsigned int hits = 0;
                 double percent = 0.0;
-                char *color;
-                struct sym_ext *sym_ext = sym->priv;
-
-                offset = line_ip - start;
-                if (offset < len)
-                        hits = sym->hist[offset];
+                const char *color;
+                struct sym_priv *priv = symbol__priv(sym);
+                struct sym_ext *sym_ext = priv->ext;
+                struct sym_hist *h = priv->hist;
+                s64 offset = self->offset;
+                struct objdump_line *next = objdump__get_next_ip_line(head, self);
+
+                while (offset < (s64)len &&
+                       (next == NULL || offset < next->offset)) {
+                        if (sym_ext) {
+                                if (path == NULL)
+                                        path = sym_ext[offset].path;
+                                percent += sym_ext[offset].percent;
+                        } else
+                                hits += h->ip[offset];
+
+                        ++offset;
+                }
 
-                if (offset < len && sym_ext) {
-                        path = sym_ext[offset].path;
-                        percent = sym_ext[offset].percent;
-                } else if (sym->hist_sum)
-                        percent = 100.0 * hits / sym->hist_sum;
+                if (sym_ext == NULL && h->sum)
+                        percent = 100.0 * hits / h->sum;
 
-                color = get_color(percent);
+                color = get_percent_color(percent);
 
                 /*
                  * Also color the filename and line if needed, with
@@ -1140,12 +130,12 @@ parse_line(FILE *file, struct symbol *sym, u64 start, u64 len)
 
                 color_fprintf(stdout, color, " %7.2f", percent);
                 printf(" :      ");
-                color_fprintf(stdout, PERF_COLOR_BLUE, "%s\n", line);
+                color_fprintf(stdout, PERF_COLOR_BLUE, "%s\n", self->line);
         } else {
-                if (!*line)
+                if (!*self->line)
                         printf("         :\n");
                 else
-                        printf("         :      %s\n", line);
+                        printf("         :      %s\n", self->line);
         }
 
         return 0;
@@ -1173,9 +163,10 @@ static void insert_source_line(struct sym_ext *sym_ext)
         rb_insert_color(&sym_ext->node, &root_sym_ext);
 }
 
-static void free_source_line(struct symbol *sym, int len)
+static void free_source_line(struct hist_entry *he, int len)
 {
-        struct sym_ext *sym_ext = sym->priv;
+        struct sym_priv *priv = symbol__priv(he->ms.sym);
+        struct sym_ext *sym_ext = priv->ext;
         int i;
 
         if (!sym_ext)
@@ -1185,26 +176,30 @@ static void free_source_line(struct symbol *sym, int len)
                 free(sym_ext[i].path);
         free(sym_ext);
 
-        sym->priv = NULL;
+        priv->ext = NULL;
         root_sym_ext = RB_ROOT;
 }
 
 /* Get the filename:line for the colored entries */
 static void
-get_source_line(struct symbol *sym, u64 start, int len, char *filename)
+get_source_line(struct hist_entry *he, int len, const char *filename)
 {
+        struct symbol *sym = he->ms.sym;
+        u64 start;
         int i;
         char cmd[PATH_MAX * 2];
         struct sym_ext *sym_ext;
+        struct sym_priv *priv = symbol__priv(sym);
+        struct sym_hist *h = priv->hist;
 
-        if (!sym->hist_sum)
+        if (!h->sum)
                 return;
 
-        sym->priv = calloc(len, sizeof(struct sym_ext));
-        if (!sym->priv)
+        sym_ext = priv->ext = calloc(len, sizeof(struct sym_ext));
+        if (!priv->ext)
                 return;
 
-        sym_ext = sym->priv;
+        start = he->ms.map->unmap_ip(he->ms.map, sym->start);
 
         for (i = 0; i < len; i++) {
                 char *path = NULL;
@@ -1212,7 +207,7 @@ get_source_line(struct symbol *sym, u64 start, int len, char *filename)
                 u64 offset;
                 FILE *fp;
 
-                sym_ext[i].percent = 100.0 * sym->hist[i] / sym->hist_sum;
+                sym_ext[i].percent = 100.0 * h->ip[i] / h->sum;
                 if (sym_ext[i].percent <= 0.5)
                         continue;
 
@@ -1237,7 +232,7 @@ get_source_line(struct symbol *sym, u64 start, int len, char *filename)
         }
 }
 
-static void print_summary(char *filename)
+static void print_summary(const char *filename)
 {
         struct sym_ext *sym_ext;
         struct rb_node *node;
@@ -1253,12 +248,12 @@ static void print_summary(char *filename)
         node = rb_first(&root_sym_ext);
         while (node) {
                 double percent;
-                char *color;
+                const char *color;
                 char *path;
 
                 sym_ext = rb_entry(node, struct sym_ext, node);
                 percent = sym_ext->percent;
-                color = get_color(percent);
+                color = get_percent_color(percent);
                 path = sym_ext->path;
 
                 color_fprintf(stdout, color, " %7.2f %s", percent, path);
@@ -1266,195 +261,155 @@ static void print_summary(char *filename)
         }
 }
 
-static void annotate_sym(struct dso *dso, struct symbol *sym)
+static void hist_entry__print_hits(struct hist_entry *self)
 {
-        char *filename = dso->name;
-        u64 start, end, len;
-        char command[PATH_MAX*2];
-        FILE *file;
+        struct symbol *sym = self->ms.sym;
+        struct sym_priv *priv = symbol__priv(sym);
+        struct sym_hist *h = priv->hist;
+        u64 len = sym->end - sym->start, offset;
 
-        if (!filename)
-                return;
-        if (dso == kernel_dso)
-                filename = vmlinux;
+        for (offset = 0; offset < len; ++offset)
+                if (h->ip[offset] != 0)
+                        printf("%*Lx: %Lu\n", BITS_PER_LONG / 2,
+                               sym->start + offset, h->ip[offset]);
+        printf("%*s: %Lu\n", BITS_PER_LONG / 2, "h->sum", h->sum);
+}
+
+static int hist_entry__tty_annotate(struct hist_entry *he)
+{
+        struct map *map = he->ms.map;
+        struct dso *dso = map->dso;
+        struct symbol *sym = he->ms.sym;
+        const char *filename = dso->long_name, *d_filename;
+        u64 len;
+        LIST_HEAD(head);
+        struct objdump_line *pos, *n;
+
+        if (hist_entry__annotate(he, &head, 0) < 0)
+                return -1;
 
-        start = sym->obj_start;
-        if (!start)
-                start = sym->start;
+        if (full_paths)
+                d_filename = filename;
+        else
+                d_filename = basename(filename);
 
-        end = start + sym->end - sym->start + 1;
         len = sym->end - sym->start;
 
         if (print_line) {
-                get_source_line(sym, start, len, filename);
+                get_source_line(he, len, filename);
                 print_summary(filename);
         }
 
         printf("\n\n------------------------------------------------\n");
-        printf(" Percent |      Source code & Disassembly of %s\n", filename);
+        printf(" Percent |      Source code & Disassembly of %s\n", d_filename);
         printf("------------------------------------------------\n");
 
-        if (verbose >= 2)
-                printf("annotating [%p] %30s : [%p] %30s\n", dso, dso->name, sym, sym->name);
-
-        sprintf(command, "objdump --start-address=0x%016Lx --stop-address=0x%016Lx -dS %s", (u64)start, (u64)end, filename);
-
-        if (verbose >= 3)
-                printf("doing: %s\n", command);
-
-        file = popen(command, "r");
-        if (!file)
-                return;
+        if (verbose)
+                hist_entry__print_hits(he);
 
-        while (!feof(file)) {
-                if (parse_line(file, sym, start, len) < 0)
-                        break;
+        list_for_each_entry_safe(pos, n, &head, node) {
+                objdump_line__print(pos, &head, he, len);
+                list_del(&pos->node);
+                objdump_line__free(pos);
         }
 
-        pclose(file);
         if (print_line)
-                free_source_line(sym, len);
+                free_source_line(he, len);
+
+        return 0;
 }
 
-static void find_annotations(void)
+static void hists__find_annotations(struct hists *self)
 {
-        struct rb_node *nd;
-        struct dso *dso;
-        int count = 0;
+        struct rb_node *nd = rb_first(&self->entries), *next;
+        int key = KEY_RIGHT;
+
+        while (nd) {
+                struct hist_entry *he = rb_entry(nd, struct hist_entry, rb_node);
+                struct sym_priv *priv;
 
-        list_for_each_entry(dso, &dsos, node) {
+                if (he->ms.sym == NULL || he->ms.map->dso->annotate_warned)
+                        goto find_next;
 
-                for (nd = rb_first(&dso->syms); nd; nd = rb_next(nd)) {
-                        struct symbol *sym = rb_entry(nd, struct symbol, rb_node);
+                priv = symbol__priv(he->ms.sym);
+                if (priv->hist == NULL) {
+find_next:
+                        if (key == KEY_LEFT)
+                                nd = rb_prev(nd);
+                        else
+                                nd = rb_next(nd);
+                        continue;
+                }
 
-                        if (sym->hist) {
-                                annotate_sym(dso, sym);
-                                count++;
+                if (use_browser > 0) {
+                        key = hist_entry__tui_annotate(he);
+                        switch (key) {
+                        case KEY_RIGHT:
+                                next = rb_next(nd);
+                                break;
+                        case KEY_LEFT:
+                                next = rb_prev(nd);
+                                break;
+                        default:
+                                return;
                         }
+
+                        if (next != NULL)
+                                nd = next;
+                } else {
+                        hist_entry__tty_annotate(he);
+                        nd = rb_next(nd);
+                        /*
+                         * Since we have a hist_entry per IP for the same
+                         * symbol, free he->ms.sym->hist to signal we already
+                         * processed this symbol.
+                         */
+                        free(priv->hist);
+                        priv->hist = NULL;
                 }
         }
-
-        if (!count)
-                printf(" Error: symbol '%s' not present amongst the samples.\n", sym_hist_filter);
 }
 
+static struct perf_event_ops event_ops = {
+        .sample = process_sample_event,
+        .mmap   = event__process_mmap,
+        .comm   = event__process_comm,
+        .fork   = event__process_task,
+        .ordered_samples = true,
+        .ordering_requires_timestamps = true,
+};
+
 static int __cmd_annotate(void)
 {
-        int ret, rc = EXIT_FAILURE;
-        unsigned long offset = 0;
-        unsigned long head = 0;
-        struct stat stat;
-        event_t *event;
-        uint32_t size;
-        char *buf;
-
-        register_idle_thread();
-
-        input = open(input_name, O_RDONLY);
-        if (input < 0) {
-                perror("failed to open file");
-                exit(-1);
-        }
-
-        ret = fstat(input, &stat);
-        if (ret < 0) {
-                perror("failed to stat file");
-                exit(-1);
-        }
-
-        if (!stat.st_size) {
-                fprintf(stderr, "zero-sized file, nothing to do!\n");
-                exit(0);
-        }
-
-        if (load_kernel() < 0) {
-                perror("failed to load kernel symbols");
-                return EXIT_FAILURE;
-        }
-
-remap:
-        buf = (char *)mmap(NULL, page_size * mmap_window, PROT_READ,
-                           MAP_SHARED, input, offset);
-        if (buf == MAP_FAILED) {
-                perror("failed to mmap file");
-                exit(-1);
-        }
-
-more:
-        event = (event_t *)(buf + head);
-
-        size = event->header.size;
-        if (!size)
-                size = 8;
-
-        if (head + event->header.size >= page_size * mmap_window) {
-                unsigned long shift = page_size * (head / page_size);
-                int ret;
-
-                ret = munmap(buf, page_size * mmap_window);
-                assert(ret == 0);
-
-                offset += shift;
-                head -= shift;
-                goto remap;
-        }
-
-        size = event->header.size;
-
-        dprintf("%p [%p]: event: %d\n",
-                        (void *)(offset + head),
-                        (void *)(long)event->header.size,
-                        event->header.type);
-
-        if (!size || process_event(event, offset, head) < 0) {
-
-                dprintf("%p [%p]: skipping unknown header type: %d\n",
-                        (void *)(offset + head),
-                        (void *)(long)(event->header.size),
-                        event->header.type);
-
-                total_unknown++;
+        int ret;
+        struct perf_session *session;
 
-                /*
-                 * assume we lost track of the stream, check alignment, and
-                 * increment a single u64 in the hope to catch on again 'soon'.
-                 */
+        session = perf_session__new(input_name, O_RDONLY, force, false, &event_ops);
+        if (session == NULL)
+                return -ENOMEM;
 
-                if (unlikely(head & 7))
-                        head &= ~7ULL;
+        ret = perf_session__process_events(session, &event_ops);
+        if (ret)
+                goto out_delete;
 
-                size = 8;
+        if (dump_trace) {
+                perf_session__fprintf_nr_events(session, stdout);
+                goto out_delete;
         }
 
-        head += size;
-
-        if (offset + head < stat.st_size)
-                goto more;
-
-        rc = EXIT_SUCCESS;
-        close(input);
-
-        dprintf("      IP events: %10ld\n", total);
-        dprintf("    mmap events: %10ld\n", total_mmap);
-        dprintf("    comm events: %10ld\n", total_comm);
-        dprintf("    fork events: %10ld\n", total_fork);
-        dprintf(" unknown events: %10ld\n", total_unknown);
-
-        if (dump_trace)
-                return 0;
-
-        if (verbose >= 3)
-                threads__fprintf(stdout);
+        if (verbose > 3)
+                perf_session__fprintf(session, stdout);
 
-        if (verbose >= 2)
-                dsos__fprintf(stdout);
+        if (verbose > 2)
+                perf_session__fprintf_dsos(session, stdout);
 
-        collapse__resort();
-        output__resort();
-
-        find_annotations();
+        hists__collapse_resort(&session->hists);
+        hists__output_resort(&session->hists);
+        hists__find_annotations(&session->hists);
+out_delete:
+        perf_session__delete(session);
 
-        return rc;
+        return ret;
 }
 
 static const char * const annotate_usage[] = {
@@ -1465,42 +420,46 @@ static const char * const annotate_usage[] = {
 static const struct option options[] = {
         OPT_STRING('i', "input", &input_name, "file",
                     "input file name"),
+        OPT_STRING('d', "dsos", &symbol_conf.dso_list_str, "dso[,dso...]",
+                   "only consider symbols in these dsos"),
         OPT_STRING('s', "symbol", &sym_hist_filter, "symbol",
                     "symbol to annotate"),
-        OPT_BOOLEAN('v', "verbose", &verbose,
+        OPT_BOOLEAN('f', "force", &force, "don't complain, do it"),
+        OPT_INCR('v', "verbose", &verbose,
                     "be more verbose (show symbol address, etc)"),
         OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
                     "dump raw trace in ASCII"),
-        OPT_STRING('k', "vmlinux", &vmlinux, "file", "vmlinux pathname"),
+        OPT_BOOLEAN(0, "tui", &use_tui, "Use the TUI interface"),
+        OPT_BOOLEAN(0, "stdio", &use_stdio, "Use the stdio interface"),
+        OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
+                   "file", "vmlinux pathname"),
+        OPT_BOOLEAN('m', "modules", &symbol_conf.use_modules,
+                    "load module symbols - WARNING: use only with -k and LIVE kernel"),
         OPT_BOOLEAN('l', "print-line", &print_line,
                     "print matching source lines (may be slow)"),
+        OPT_BOOLEAN('P', "full-paths", &full_paths,
+                    "Don't shorten the displayed pathnames"),
         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);
+        argc = parse_options(argc, argv, options, annotate_usage, 0);
 
-        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);
-                }
-        }
+        if (use_stdio)
+                use_browser = 0;
+        else if (use_tui)
+                use_browser = 1;
 
-        free(str);
-}
+        setup_browser();
 
-int cmd_annotate(int argc, const char **argv, const char *prefix)
-{
-        symbol__init();
-
-        page_size = getpagesize();
+        symbol_conf.priv_size = sizeof(struct sym_priv);
+        symbol_conf.try_vmlinux_path = true;
 
-        argc = parse_options(argc, argv, options, annotate_usage, 0);
+        if (symbol__init() < 0)
+                return -1;
 
-        setup_sorting();
+        setup_sorting(annotate_usage, options);
 
         if (argc) {
                 /*
@@ -1513,10 +472,10 @@ int cmd_annotate(int argc, const char **argv, const char *prefix)
                 sym_hist_filter = argv[0];
         }
 
-        if (!sym_hist_filter)
-                usage_with_options(annotate_usage, options);
-
-        setup_pager();
+        if (field_sep && *field_sep == '.') {
+                pr_err("'.' is the only non valid --field-separator argument\n");
+                return -1;
+        }
 
         return __cmd_annotate();
 }
diff --git a/tools/perf/builtin-bench.c b/tools/perf/builtin-bench.c
new file mode 100644
index 00000000000..fcb96269852
--- /dev/null
+++ b/tools/perf/builtin-bench.c
@@ -0,0 +1,245 @@
+/*
+ *
+ * builtin-bench.c
+ *
+ * General benchmarking subsystem provided by perf
+ *
+ * Copyright (C) 2009, Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
+ *
+ */
+
+/*
+ *
+ * Available subsystem list:
+ *  sched ... scheduler and IPC mechanism
+ *  mem   ... memory access performance
+ *
+ */
+
+#include "perf.h"
+#include "util/util.h"
+#include "util/parse-options.h"
+#include "builtin.h"
+#include "bench/bench.h"
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+struct bench_suite {
+        const char *name;
+        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",
+          "Benchmark for scheduler and IPC mechanisms",
+          bench_sched_messaging },
+        { "pipe",
+          "Flood of communication over pipe() between two processes",
+          bench_sched_pipe      },
+        suite_all,
+        { NULL,
+          NULL,
+          NULL                  }
+};
+
+static struct bench_suite mem_suites[] = {
+        { "memcpy",
+          "Simple memory copy in various ways",
+          bench_mem_memcpy },
+        suite_all,
+        { NULL,
+          NULL,
+          NULL             }
+};
+
+struct bench_subsys {
+        const char *name;
+        const char *summary;
+        struct bench_suite *suites;
+};
+
+static struct bench_subsys subsystems[] = {
+        { "sched",
+          "scheduler and IPC mechanism",
+          sched_suites },
+        { "mem",
+          "memory access performance",
+          mem_suites },
+        { "all",                /* sentinel: easy for help */
+          "test all subsystem (pseudo subsystem)",
+          NULL },
+        { NULL,
+          NULL,
+          NULL       }
+};
+
+static void dump_suites(int subsys_index)
+{
+        int i;
+
+        printf("# List of available suites for %s...\n\n",
+               subsystems[subsys_index].name);
+
+        for (i = 0; subsystems[subsys_index].suites[i].name; i++)
+                printf("%14s: %s\n",
+                       subsystems[subsys_index].suites[i].name,
+                       subsystems[subsys_index].suites[i].summary);
+
+        printf("\n");
+        return;
+}
+
+static const char *bench_format_str;
+int bench_format = BENCH_FORMAT_DEFAULT;
+
+static const struct option bench_options[] = {
+        OPT_STRING('f', "format", &bench_format_str, "default",
+                    "Specify format style"),
+        OPT_END()
+};
+
+static const char * const bench_usage[] = {
+        "perf bench [<common options>] <subsystem> <suite> [<options>]",
+        NULL
+};
+
+static void print_usage(void)
+{
+        int i;
+
+        printf("Usage: \n");
+        for (i = 0; bench_usage[i]; i++)
+                printf("\t%s\n", bench_usage[i]);
+        printf("\n");
+
+        printf("# List of available subsystems...\n\n");
+
+        for (i = 0; subsystems[i].name; i++)
+                printf("%14s: %s\n",
+                       subsystems[i].name, subsystems[i].summary);
+        printf("\n");
+}
+
+static int bench_str2int(const char *str)
+{
+        if (!str)
+                return BENCH_FORMAT_DEFAULT;
+
+        if (!strcmp(str, BENCH_FORMAT_DEFAULT_STR))
+                return BENCH_FORMAT_DEFAULT;
+        else if (!strcmp(str, BENCH_FORMAT_SIMPLE_STR))
+                return BENCH_FORMAT_SIMPLE;
+
+        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;
+
+        if (argc < 2) {
+                /* No subsystem specified. */
+                print_usage();
+                goto end;
+        }
+
+        argc = parse_options(argc, argv, bench_options, bench_usage,
+                             PARSE_OPT_STOP_AT_NON_OPTION);
+
+        bench_format = bench_str2int(bench_format_str);
+        if (bench_format == BENCH_FORMAT_UNKNOWN) {
+                printf("Unknown format descriptor:%s\n", bench_format_str);
+                goto end;
+        }
+
+        if (argc < 1) {
+                print_usage();
+                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;
+
+                if (argc < 2) {
+                        /* No suite specified. */
+                        dump_suites(i);
+                        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;
+
+                        if (bench_format == BENCH_FORMAT_DEFAULT)
+                                printf("# Running %s/%s benchmark...\n",
+                                       subsystems[i].name,
+                                       subsystems[i].suites[j].name);
+                        status = subsystems[i].suites[j].fn(argc - 1,
+                                                            argv + 1, prefix);
+                        goto end;
+                }
+
+                if (!strcmp(argv[1], "-h") || !strcmp(argv[1], "--help")) {
+                        dump_suites(i);
+                        goto end;
+                }
+
+                printf("Unknown suite:%s for %s\n", argv[1], argv[0]);
+                status = 1;
+                goto end;
+        }
+
+        printf("Unknown subsystem:%s\n", argv[0]);
+        status = 1;
+
+end:
+        return status;
+}
diff --git a/tools/perf/builtin-buildid-cache.c b/tools/perf/builtin-buildid-cache.c
new file mode 100644
index 00000000000..29ad20e6791
--- /dev/null
+++ b/tools/perf/builtin-buildid-cache.c
@@ -0,0 +1,132 @@
+/*
+ * builtin-buildid-cache.c
+ *
+ * Builtin buildid-cache command: Manages build-id cache
+ *
+ * Copyright (C) 2010, Red Hat Inc.
+ * Copyright (C) 2010, Arnaldo Carvalho de Melo <acme@redhat.com>
+ */
+#include "builtin.h"
+#include "perf.h"
+#include "util/cache.h"
+#include "util/debug.h"
+#include "util/header.h"
+#include "util/parse-options.h"
+#include "util/strlist.h"
+#include "util/symbol.h"
+
+static char const *add_name_list_str, *remove_name_list_str;
+
+static const char * const buildid_cache_usage[] = {
+        "perf buildid-cache [<options>]",
+        NULL
+};
+
+static const struct option buildid_cache_options[] = {
+        OPT_STRING('a', "add", &add_name_list_str,
+                   "file list", "file(s) to add"),
+        OPT_STRING('r', "remove", &remove_name_list_str, "file list",
+                    "file(s) to remove"),
+        OPT_INCR('v', "verbose", &verbose, "be more verbose"),
+        OPT_END()
+};
+
+static int build_id_cache__add_file(const char *filename, const char *debugdir)
+{
+        char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+        u8 build_id[BUILD_ID_SIZE];
+        int err;
+
+        if (filename__read_build_id(filename, &build_id, sizeof(build_id)) < 0) {
+                pr_debug("Couldn't read a build-id in %s\n", filename);
+                return -1;
+        }
+
+        build_id__sprintf(build_id, sizeof(build_id), sbuild_id);
+        err = build_id_cache__add_s(sbuild_id, debugdir, filename, false);
+        if (verbose)
+                pr_info("Adding %s %s: %s\n", sbuild_id, filename,
+                        err ? "FAIL" : "Ok");
+        return err;
+}
+
+static int build_id_cache__remove_file(const char *filename __used,
+                                       const char *debugdir __used)
+{
+        u8 build_id[BUILD_ID_SIZE];
+        char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+        int err;
+
+        if (filename__read_build_id(filename, &build_id, sizeof(build_id)) < 0) {
+                pr_debug("Couldn't read a build-id in %s\n", filename);
+                return -1;
+        }
+
+        build_id__sprintf(build_id, sizeof(build_id), sbuild_id);
+        err = build_id_cache__remove_s(sbuild_id, debugdir);
+        if (verbose)
+                pr_info("Removing %s %s: %s\n", sbuild_id, filename,
+                        err ? "FAIL" : "Ok");
+
+        return err;
+}
+
+static int __cmd_buildid_cache(void)
+{
+        struct strlist *list;
+        struct str_node *pos;
+        char debugdir[PATH_MAX];
+
+        snprintf(debugdir, sizeof(debugdir), "%s", buildid_dir);
+
+        if (add_name_list_str) {
+                list = strlist__new(true, add_name_list_str);
+                if (list) {
+                        strlist__for_each(pos, list)
+                                if (build_id_cache__add_file(pos->s, debugdir)) {
+                                        if (errno == EEXIST) {
+                                                pr_debug("%s already in the cache\n",
+                                                         pos->s);
+                                                continue;
+                                        }
+                                        pr_warning("Couldn't add %s: %s\n",
+                                                   pos->s, strerror(errno));
+                                }
+
+                        strlist__delete(list);
+                }
+        }
+
+        if (remove_name_list_str) {
+                list = strlist__new(true, remove_name_list_str);
+                if (list) {
+                        strlist__for_each(pos, list)
+                                if (build_id_cache__remove_file(pos->s, debugdir)) {
+                                        if (errno == ENOENT) {
+                                                pr_debug("%s wasn't in the cache\n",
+                                                         pos->s);
+                                                continue;
+                                        }
+                                        pr_warning("Couldn't remove %s: %s\n",
+                                                   pos->s, strerror(errno));
+                                }
+
+                        strlist__delete(list);
+                }
+        }
+
+        return 0;
+}
+
+int cmd_buildid_cache(int argc, const char **argv, const char *prefix __used)
+{
+        argc = parse_options(argc, argv, buildid_cache_options,
+                             buildid_cache_usage, 0);
+
+        if (symbol__init() < 0)
+                return -1;
+
+        setup_pager();
+        return __cmd_buildid_cache();
+}
diff --git a/tools/perf/builtin-buildid-list.c b/tools/perf/builtin-buildid-list.c
new file mode 100644
index 00000000000..5af32ae9031
--- /dev/null
+++ b/tools/perf/builtin-buildid-list.c
@@ -0,0 +1,60 @@
+/*
+ * builtin-buildid-list.c
+ *
+ * Builtin buildid-list command: list buildids in perf.data
+ *
+ * Copyright (C) 2009, Red Hat Inc.
+ * Copyright (C) 2009, Arnaldo Carvalho de Melo <acme@redhat.com>
+ */
+#include "builtin.h"
+#include "perf.h"
+#include "util/build-id.h"
+#include "util/cache.h"
+#include "util/debug.h"
+#include "util/parse-options.h"
+#include "util/session.h"
+#include "util/symbol.h"
+
+static char const *input_name = "perf.data";
+static bool force;
+static bool with_hits;
+
+static const char * const buildid_list_usage[] = {
+        "perf buildid-list [<options>]",
+        NULL
+};
+
+static const struct option options[] = {
+        OPT_BOOLEAN('H', "with-hits", &with_hits, "Show only DSOs with hits"),
+        OPT_STRING('i', "input", &input_name, "file",
+                    "input file name"),
+        OPT_BOOLEAN('f', "force", &force, "don't complain, do it"),
+        OPT_INCR('v', "verbose", &verbose,
+                    "be more verbose"),
+        OPT_END()
+};
+
+static int __cmd_buildid_list(void)
+{
+        struct perf_session *session;
+
+        session = perf_session__new(input_name, O_RDONLY, force, false,
+                                    &build_id__mark_dso_hit_ops);
+        if (session == NULL)
+                return -1;
+
+        if (with_hits)
+                perf_session__process_events(session, &build_id__mark_dso_hit_ops);
+
+        perf_session__fprintf_dsos_buildid(session, stdout, with_hits);
+
+        perf_session__delete(session);
+        return 0;
+}
+
+int cmd_buildid_list(int argc, const char **argv, const char *prefix __used)
+{
+        argc = parse_options(argc, argv, options, buildid_list_usage, 0);
+        setup_pager();
+        return __cmd_buildid_list();
+}
diff --git a/tools/perf/builtin-diff.c b/tools/perf/builtin-diff.c
new file mode 100644
index 00000000000..3153e492dbc
--- /dev/null
+++ b/tools/perf/builtin-diff.c
@@ -0,0 +1,232 @@
+/*
+ * 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 bool  force;
+static bool show_displacement;
+
+static int hists__add_entry(struct hists *self,
+                            struct addr_location *al, u64 period)
+{
+        if (__hists__add_entry(self, al, NULL, period) != NULL)
+                return 0;
+        return -ENOMEM;
+}
+
+static int diff__process_sample_event(event_t *event,
+                                      struct sample_data *sample,
+                                      struct perf_session *session)
+{
+        struct addr_location al;
+
+        if (event__preprocess_sample(event, session, &al, sample, NULL) < 0) {
+                pr_warning("problem processing %d event, skipping it.\n",
+                           event->header.type);
+                return -1;
+        }
+
+        if (al.filtered || al.sym == NULL)
+                return 0;
+
+        if (hists__add_entry(&session->hists, &al, sample->period)) {
+                pr_warning("problem incrementing symbol period, skipping event\n");
+                return -1;
+        }
+
+        session->hists.stats.total_period += sample->period;
+        return 0;
+}
+
+static struct perf_event_ops event_ops = {
+        .sample = diff__process_sample_event,
+        .mmap   = event__process_mmap,
+        .comm   = event__process_comm,
+        .exit   = event__process_task,
+        .fork   = event__process_task,
+        .lost   = event__process_lost,
+        .ordered_samples = true,
+        .ordering_requires_timestamps = true,
+};
+
+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) {
+                parent = *p;
+                iter = rb_entry(parent, struct hist_entry, rb_node);
+                if (hist_entry__cmp(he, iter) < 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 hists__resort_entries(struct hists *self)
+{
+        unsigned long position = 1;
+        struct rb_root tmp = RB_ROOT;
+        struct rb_node *next = rb_first(&self->entries);
+
+        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->entries);
+                n->position = position++;
+                perf_session__insert_hist_entry_by_name(&tmp, n);
+        }
+
+        self->entries = tmp;
+}
+
+static void hists__set_positions(struct hists *self)
+{
+        hists__output_resort(self);
+        hists__resort_entries(self);
+}
+
+static struct hist_entry *hists__find_entry(struct hists *self,
+                                            struct hist_entry *he)
+{
+        struct rb_node *n = self->entries.rb_node;
+
+        while (n) {
+                struct hist_entry *iter = rb_entry(n, struct hist_entry, rb_node);
+                int64_t cmp = hist_entry__cmp(he, iter);
+
+                if (cmp < 0)
+                        n = n->rb_left;
+                else if (cmp > 0)
+                        n = n->rb_right;
+                else 
+                        return iter;
+        }
+
+        return NULL;
+}
+
+static void hists__match(struct hists *older, struct hists *newer)
+{
+        struct rb_node *nd;
+
+        for (nd = rb_first(&newer->entries); nd; nd = rb_next(nd)) {
+                struct hist_entry *pos = rb_entry(nd, struct hist_entry, rb_node);
+                pos->pair = hists__find_entry(older, pos);
+        }
+}
+
+static int __cmd_diff(void)
+{
+        int ret, i;
+        struct perf_session *session[2];
+
+        session[0] = perf_session__new(input_old, O_RDONLY, force, false, &event_ops);
+        session[1] = perf_session__new(input_new, O_RDONLY, force, false, &event_ops);
+        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;
+        }
+
+        hists__output_resort(&session[1]->hists);
+        if (show_displacement)
+                hists__set_positions(&session[0]->hists);
+
+        hists__match(&session[0]->hists, &session[1]->hists);
+        hists__fprintf(&session[1]->hists, &session[0]->hists,
+                       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_INCR('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_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_STRING(0, "symfs", &symbol_conf.symfs, "directory",
+                    "Look for files with symbols relative to this directory"),
+        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];
+        } else if (symbol_conf.default_guest_vmlinux_name ||
+                   symbol_conf.default_guest_kallsyms) {
+                input_old = "perf.data.host";
+                input_new = "perf.data.guest";
+        }
+
+        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-help.c b/tools/perf/builtin-help.c
index 0f32dc3f3c4..6d5a8a7faf4 100644
--- a/tools/perf/builtin-help.c
+++ b/tools/perf/builtin-help.c
@@ -3,6 +3,7 @@
  *
  * Builtin help command
  */
+#include "perf.h"
 #include "util/cache.h"
 #include "builtin.h"
 #include "util/exec_cmd.h"
@@ -28,14 +29,14 @@ enum help_format {
         HELP_FORMAT_WEB,
 };
 
-static int show_all = 0;
+static bool show_all = false;
 static enum help_format help_format = HELP_FORMAT_MAN;
 static struct option builtin_help_options[] = {
         OPT_BOOLEAN('a', "all", &show_all, "print all available commands"),
-        OPT_SET_INT('m', "man", &help_format, "show man page", HELP_FORMAT_MAN),
-        OPT_SET_INT('w', "web", &help_format, "show manual in web browser",
+        OPT_SET_UINT('m', "man", &help_format, "show man page", HELP_FORMAT_MAN),
+        OPT_SET_UINT('w', "web", &help_format, "show manual in web browser",
                         HELP_FORMAT_WEB),
-        OPT_SET_INT('i', "info", &help_format, "show info page",
+        OPT_SET_UINT('i', "info", &help_format, "show info page",
                         HELP_FORMAT_INFO),
         OPT_END(),
 };
@@ -60,8 +61,7 @@ static const char *get_man_viewer_info(const char *name)
 {
         struct man_viewer_info_list *viewer;
 
-        for (viewer = man_viewer_info_list; viewer; viewer = viewer->next)
-        {
+        for (viewer = man_viewer_info_list; viewer; viewer = viewer->next) {
                 if (!strcasecmp(name, viewer->name))
                         return viewer->info;
         }
@@ -114,7 +114,7 @@ static int check_emacsclient_version(void)
         return 0;
 }
 
-static void exec_woman_emacs(const char* path, const char *page)
+static void exec_woman_emacs(const char *path, const char *page)
 {
         if (!check_emacsclient_version()) {
                 /* This works only with emacsclient version >= 22. */
@@ -128,7 +128,7 @@ static void exec_woman_emacs(const char* path, const char *page)
         }
 }
 
-static void exec_man_konqueror(const char* path, const char *page)
+static void exec_man_konqueror(const char *path, const char *page)
 {
         const char *display = getenv("DISPLAY");
         if (display && *display) {
@@ -156,7 +156,7 @@ static void exec_man_konqueror(const char* path, const char *page)
         }
 }
 
-static void exec_man_man(const char* path, const char *page)
+static void exec_man_man(const char *path, const char *page)
 {
         if (!path)
                 path = "man";
@@ -179,7 +179,7 @@ static void add_man_viewer(const char *name)
 
         while (*p)
                 p = &((*p)->next);
-        *p = calloc(1, (sizeof(**p) + len + 1));
+        *p = zalloc(sizeof(**p) + len + 1);
         strncpy((*p)->name, name, len);
 }
 
@@ -194,7 +194,7 @@ static void do_add_man_viewer_info(const char *name,
                                    size_t len,
                                    const char *value)
 {
-        struct man_viewer_info_list *new = calloc(1, sizeof(*new) + len + 1);
+        struct man_viewer_info_list *new = zalloc(sizeof(*new) + len + 1);
 
         strncpy(new->name, name, len);
         new->info = strdup(value);
@@ -277,7 +277,7 @@ static struct cmdnames main_cmds, other_cmds;
 
 void list_common_cmds_help(void)
 {
-        int i, longest = 0;
+        unsigned int i, longest = 0;
 
         for (i = 0; i < ARRAY_SIZE(common_cmds); i++) {
                 if (longest < strlen(common_cmds[i].name))
@@ -286,8 +286,7 @@ void list_common_cmds_help(void)
 
         puts(" The most commonly used perf commands are:");
         for (i = 0; i < ARRAY_SIZE(common_cmds); i++) {
-                printf("   %s   ", common_cmds[i].name);
-                mput_char(' ', longest - strlen(common_cmds[i].name));
+                printf("   %-*s   ", longest, common_cmds[i].name);
                 puts(common_cmds[i].help);
         }
 }
@@ -314,8 +313,6 @@ static const char *cmd_to_page(const char *perf_cmd)
                 return "perf";
         else if (!prefixcmp(perf_cmd, "perf"))
                 return perf_cmd;
-        else if (is_perf_command(perf_cmd))
-                return prepend("perf-", perf_cmd);
         else
                 return prepend("perf-", perf_cmd);
 }
@@ -363,9 +360,8 @@ static void show_man_page(const char *perf_cmd)
 
         setup_man_path();
         for (viewer = man_viewer_list; viewer; viewer = viewer->next)
-        {
                 exec_viewer(viewer->name, page); /* will return when unable */
-        }
+
         if (fallback)
                 exec_viewer(fallback, page);
         exec_viewer("man", page);
@@ -415,9 +411,10 @@ static void show_html_page(const char *perf_cmd)
         open_html(page_path.buf);
 }
 
-int cmd_help(int argc, const char **argv, const char *prefix)
+int cmd_help(int argc, const char **argv, const char *prefix __used)
 {
         const char *alias;
+
         load_command_list("perf-", &main_cmds, &other_cmds);
 
         perf_config(perf_help_config, NULL);
@@ -454,6 +451,7 @@ int cmd_help(int argc, const char **argv, const char *prefix)
                 break;
         case HELP_FORMAT_WEB:
                 show_html_page(argv[0]);
+        default:
                 break;
         }
 
diff --git a/tools/perf/builtin-inject.c b/tools/perf/builtin-inject.c
new file mode 100644
index 00000000000..0c78ffa7bf6
--- /dev/null
+++ b/tools/perf/builtin-inject.c
@@ -0,0 +1,237 @@
+/*
+ * builtin-inject.c
+ *
+ * Builtin inject command: Examine the live mode (stdin) event stream
+ * and repipe it to stdout while optionally injecting additional
+ * events into it.
+ */
+#include "builtin.h"
+
+#include "perf.h"
+#include "util/session.h"
+#include "util/debug.h"
+
+#include "util/parse-options.h"
+
+static char             const *input_name = "-";
+static bool             inject_build_ids;
+
+static int event__repipe_synth(event_t *event,
+                               struct perf_session *session __used)
+{
+        uint32_t size;
+        void *buf = event;
+
+        size = event->header.size;
+
+        while (size) {
+                int ret = write(STDOUT_FILENO, buf, size);
+                if (ret < 0)
+                        return -errno;
+
+                size -= ret;
+                buf += ret;
+        }
+
+        return 0;
+}
+
+static int event__repipe(event_t *event, struct sample_data *sample __used,
+                         struct perf_session *session)
+{
+        return event__repipe_synth(event, session);
+}
+
+static int event__repipe_mmap(event_t *self, struct sample_data *sample,
+                              struct perf_session *session)
+{
+        int err;
+
+        err = event__process_mmap(self, sample, session);
+        event__repipe(self, sample, session);
+
+        return err;
+}
+
+static int event__repipe_task(event_t *self, struct sample_data *sample,
+                              struct perf_session *session)
+{
+        int err;
+
+        err = event__process_task(self, sample, session);
+        event__repipe(self, sample, session);
+
+        return err;
+}
+
+static int event__repipe_tracing_data(event_t *self,
+                                      struct perf_session *session)
+{
+        int err;
+
+        event__repipe_synth(self, session);
+        err = event__process_tracing_data(self, session);
+
+        return err;
+}
+
+static int dso__read_build_id(struct dso *self)
+{
+        if (self->has_build_id)
+                return 0;
+
+        if (filename__read_build_id(self->long_name, self->build_id,
+                                    sizeof(self->build_id)) > 0) {
+                self->has_build_id = true;
+                return 0;
+        }
+
+        return -1;
+}
+
+static int dso__inject_build_id(struct dso *self, struct perf_session *session)
+{
+        u16 misc = PERF_RECORD_MISC_USER;
+        struct machine *machine;
+        int err;
+
+        if (dso__read_build_id(self) < 0) {
+                pr_debug("no build_id found for %s\n", self->long_name);
+                return -1;
+        }
+
+        machine = perf_session__find_host_machine(session);
+        if (machine == NULL) {
+                pr_err("Can't find machine for session\n");
+                return -1;
+        }
+
+        if (self->kernel)
+                misc = PERF_RECORD_MISC_KERNEL;
+
+        err = event__synthesize_build_id(self, misc, event__repipe,
+                                         machine, session);
+        if (err) {
+                pr_err("Can't synthesize build_id event for %s\n", self->long_name);
+                return -1;
+        }
+
+        return 0;
+}
+
+static int event__inject_buildid(event_t *event, struct sample_data *sample,
+                                 struct perf_session *session)
+{
+        struct addr_location al;
+        struct thread *thread;
+        u8 cpumode;
+
+        cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+
+        thread = perf_session__findnew(session, event->ip.pid);
+        if (thread == NULL) {
+                pr_err("problem processing %d event, skipping it.\n",
+                       event->header.type);
+                goto repipe;
+        }
+
+        thread__find_addr_map(thread, session, cpumode, MAP__FUNCTION,
+                              event->ip.pid, event->ip.ip, &al);
+
+        if (al.map != NULL) {
+                if (!al.map->dso->hit) {
+                        al.map->dso->hit = 1;
+                        if (map__load(al.map, NULL) >= 0) {
+                                dso__inject_build_id(al.map->dso, session);
+                                /*
+                                 * If this fails, too bad, let the other side
+                                 * account this as unresolved.
+                                 */
+                        } else
+                                pr_warning("no symbols found in %s, maybe "
+                                           "install a debug package?\n",
+                                           al.map->dso->long_name);
+                }
+        }
+
+repipe:
+        event__repipe(event, sample, session);
+        return 0;
+}
+
+struct perf_event_ops inject_ops = {
+        .sample         = event__repipe,
+        .mmap           = event__repipe,
+        .comm           = event__repipe,
+        .fork           = event__repipe,
+        .exit           = event__repipe,
+        .lost           = event__repipe,
+        .read           = event__repipe,
+        .throttle       = event__repipe,
+        .unthrottle     = event__repipe,
+        .attr           = event__repipe_synth,
+        .event_type     = event__repipe_synth,
+        .tracing_data   = event__repipe_synth,
+        .build_id       = event__repipe_synth,
+};
+
+extern volatile int session_done;
+
+static void sig_handler(int sig __attribute__((__unused__)))
+{
+        session_done = 1;
+}
+
+static int __cmd_inject(void)
+{
+        struct perf_session *session;
+        int ret = -EINVAL;
+
+        signal(SIGINT, sig_handler);
+
+        if (inject_build_ids) {
+                inject_ops.sample       = event__inject_buildid;
+                inject_ops.mmap         = event__repipe_mmap;
+                inject_ops.fork         = event__repipe_task;
+                inject_ops.tracing_data = event__repipe_tracing_data;
+        }
+
+        session = perf_session__new(input_name, O_RDONLY, false, true, &inject_ops);
+        if (session == NULL)
+                return -ENOMEM;
+
+        ret = perf_session__process_events(session, &inject_ops);
+
+        perf_session__delete(session);
+
+        return ret;
+}
+
+static const char * const report_usage[] = {
+        "perf inject [<options>]",
+        NULL
+};
+
+static const struct option options[] = {
+        OPT_BOOLEAN('b', "build-ids", &inject_build_ids,
+                    "Inject build-ids into the output stream"),
+        OPT_INCR('v', "verbose", &verbose,
+                 "be more verbose (show build ids, etc)"),
+        OPT_END()
+};
+
+int cmd_inject(int argc, const char **argv, const char *prefix __used)
+{
+        argc = parse_options(argc, argv, options, report_usage, 0);
+
+        /*
+         * Any (unrecognized) arguments left?
+         */
+        if (argc)
+                usage_with_options(report_usage, options);
+
+        if (symbol__init() < 0)
+                return -1;
+
+        return __cmd_inject();
+}
diff --git a/tools/perf/builtin-kmem.c b/tools/perf/builtin-kmem.c
new file mode 100644
index 00000000000..def7ddc2fd4
--- /dev/null
+++ b/tools/perf/builtin-kmem.c
@@ -0,0 +1,777 @@
+#include "builtin.h"
+#include "perf.h"
+
+#include "util/util.h"
+#include "util/cache.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/header.h"
+#include "util/session.h"
+
+#include "util/parse-options.h"
+#include "util/trace-event.h"
+
+#include "util/debug.h"
+
+#include <linux/rbtree.h>
+
+struct alloc_stat;
+typedef int (*sort_fn_t)(struct alloc_stat *, struct alloc_stat *);
+
+static char const               *input_name = "perf.data";
+
+static int                      alloc_flag;
+static int                      caller_flag;
+
+static int                      alloc_lines = -1;
+static int                      caller_lines = -1;
+
+static bool                     raw_ip;
+
+static char                     default_sort_order[] = "frag,hit,bytes";
+
+static int                      *cpunode_map;
+static int                      max_cpu_num;
+
+struct alloc_stat {
+        u64     call_site;
+        u64     ptr;
+        u64     bytes_req;
+        u64     bytes_alloc;
+        u32     hit;
+        u32     pingpong;
+
+        short   alloc_cpu;
+
+        struct rb_node node;
+};
+
+static struct rb_root root_alloc_stat;
+static struct rb_root root_alloc_sorted;
+static struct rb_root root_caller_stat;
+static struct rb_root root_caller_sorted;
+
+static unsigned long total_requested, total_allocated;
+static unsigned long nr_allocs, nr_cross_allocs;
+
+#define PATH_SYS_NODE   "/sys/devices/system/node"
+
+static void init_cpunode_map(void)
+{
+        FILE *fp;
+        int i;
+
+        fp = fopen("/sys/devices/system/cpu/kernel_max", "r");
+        if (!fp) {
+                max_cpu_num = 4096;
+                return;
+        }
+
+        if (fscanf(fp, "%d", &max_cpu_num) < 1)
+                die("Failed to read 'kernel_max' from sysfs");
+        max_cpu_num++;
+
+        cpunode_map = calloc(max_cpu_num, sizeof(int));
+        if (!cpunode_map)
+                die("calloc");
+        for (i = 0; i < max_cpu_num; i++)
+                cpunode_map[i] = -1;
+        fclose(fp);
+}
+
+static void setup_cpunode_map(void)
+{
+        struct dirent *dent1, *dent2;
+        DIR *dir1, *dir2;
+        unsigned int cpu, mem;
+        char buf[PATH_MAX];
+
+        init_cpunode_map();
+
+        dir1 = opendir(PATH_SYS_NODE);
+        if (!dir1)
+                return;
+
+        while ((dent1 = readdir(dir1)) != NULL) {
+                if (dent1->d_type != DT_DIR ||
+                    sscanf(dent1->d_name, "node%u", &mem) < 1)
+                        continue;
+
+                snprintf(buf, PATH_MAX, "%s/%s", PATH_SYS_NODE, dent1->d_name);
+                dir2 = opendir(buf);
+                if (!dir2)
+                        continue;
+                while ((dent2 = readdir(dir2)) != NULL) {
+                        if (dent2->d_type != DT_LNK ||
+                            sscanf(dent2->d_name, "cpu%u", &cpu) < 1)
+                                continue;
+                        cpunode_map[cpu] = mem;
+                }
+        }
+}
+
+static void insert_alloc_stat(unsigned long call_site, unsigned long ptr,
+                              int bytes_req, int bytes_alloc, int cpu)
+{
+        struct rb_node **node = &root_alloc_stat.rb_node;
+        struct rb_node *parent = NULL;
+        struct alloc_stat *data = NULL;
+
+        while (*node) {
+                parent = *node;
+                data = rb_entry(*node, struct alloc_stat, node);
+
+                if (ptr > data->ptr)
+                        node = &(*node)->rb_right;
+                else if (ptr < data->ptr)
+                        node = &(*node)->rb_left;
+                else
+                        break;
+        }
+
+        if (data && data->ptr == ptr) {
+                data->hit++;
+                data->bytes_req += bytes_req;
+                data->bytes_alloc += bytes_alloc;
+        } else {
+                data = malloc(sizeof(*data));
+                if (!data)
+                        die("malloc");
+                data->ptr = ptr;
+                data->pingpong = 0;
+                data->hit = 1;
+                data->bytes_req = bytes_req;
+                data->bytes_alloc = bytes_alloc;
+
+                rb_link_node(&data->node, parent, node);
+                rb_insert_color(&data->node, &root_alloc_stat);
+        }
+        data->call_site = call_site;
+        data->alloc_cpu = cpu;
+}
+
+static void insert_caller_stat(unsigned long call_site,
+                              int bytes_req, int bytes_alloc)
+{
+        struct rb_node **node = &root_caller_stat.rb_node;
+        struct rb_node *parent = NULL;
+        struct alloc_stat *data = NULL;
+
+        while (*node) {
+                parent = *node;
+                data = rb_entry(*node, struct alloc_stat, node);
+
+                if (call_site > data->call_site)
+                        node = &(*node)->rb_right;
+                else if (call_site < data->call_site)
+                        node = &(*node)->rb_left;
+                else
+                        break;
+        }
+
+        if (data && data->call_site == call_site) {
+                data->hit++;
+                data->bytes_req += bytes_req;
+                data->bytes_alloc += bytes_alloc;
+        } else {
+                data = malloc(sizeof(*data));
+                if (!data)
+                        die("malloc");
+                data->call_site = call_site;
+                data->pingpong = 0;
+                data->hit = 1;
+                data->bytes_req = bytes_req;
+                data->bytes_alloc = bytes_alloc;
+
+                rb_link_node(&data->node, parent, node);
+                rb_insert_color(&data->node, &root_caller_stat);
+        }
+}
+
+static void process_alloc_event(void *data,
+                                struct event *event,
+                                int cpu,
+                                u64 timestamp __used,
+                                struct thread *thread __used,
+                                int node)
+{
+        unsigned long call_site;
+        unsigned long ptr;
+        int bytes_req;
+        int bytes_alloc;
+        int node1, node2;
+
+        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);
+
+        total_requested += bytes_req;
+        total_allocated += bytes_alloc;
+
+        if (node) {
+                node1 = cpunode_map[cpu];
+                node2 = raw_field_value(event, "node", data);
+                if (node1 != node2)
+                        nr_cross_allocs++;
+        }
+        nr_allocs++;
+}
+
+static int ptr_cmp(struct alloc_stat *, struct alloc_stat *);
+static int callsite_cmp(struct alloc_stat *, struct alloc_stat *);
+
+static struct alloc_stat *search_alloc_stat(unsigned long ptr,
+                                            unsigned long call_site,
+                                            struct rb_root *root,
+                                            sort_fn_t sort_fn)
+{
+        struct rb_node *node = root->rb_node;
+        struct alloc_stat key = { .ptr = ptr, .call_site = call_site };
+
+        while (node) {
+                struct alloc_stat *data;
+                int cmp;
+
+                data = rb_entry(node, struct alloc_stat, node);
+
+                cmp = sort_fn(&key, data);
+                if (cmp < 0)
+                        node = node->rb_left;
+                else if (cmp > 0)
+                        node = node->rb_right;
+                else
+                        return data;
+        }
+        return NULL;
+}
+
+static void process_free_event(void *data,
+                               struct event *event,
+                               int cpu,
+                               u64 timestamp __used,
+                               struct thread *thread __used)
+{
+        unsigned long ptr;
+        struct alloc_stat *s_alloc, *s_caller;
+
+        ptr = raw_field_value(event, "ptr", data);
+
+        s_alloc = search_alloc_stat(ptr, 0, &root_alloc_stat, ptr_cmp);
+        if (!s_alloc)
+                return;
+
+        if (cpu != s_alloc->alloc_cpu) {
+                s_alloc->pingpong++;
+
+                s_caller = search_alloc_stat(0, s_alloc->call_site,
+                                             &root_caller_stat, callsite_cmp);
+                assert(s_caller);
+                s_caller->pingpong++;
+        }
+        s_alloc->alloc_cpu = -1;
+}
+
+static void
+process_raw_event(event_t *raw_event __used, void *data,
+                  int cpu, u64 timestamp, struct thread *thread)
+{
+        struct event *event;
+        int type;
+
+        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(data, event, cpu, timestamp, thread, 0);
+                return;
+        }
+
+        if (!strcmp(event->name, "kmalloc_node") ||
+            !strcmp(event->name, "kmem_cache_alloc_node")) {
+                process_alloc_event(data, event, cpu, timestamp, thread, 1);
+                return;
+        }
+
+        if (!strcmp(event->name, "kfree") ||
+            !strcmp(event->name, "kmem_cache_free")) {
+                process_free_event(data, event, cpu, timestamp, thread);
+                return;
+        }
+}
+
+static int process_sample_event(event_t *event, struct sample_data *sample,
+                                struct perf_session *session)
+{
+        struct thread *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);
+
+        process_raw_event(event, sample->raw_data, sample->cpu,
+                          sample->time, thread);
+
+        return 0;
+}
+
+static struct perf_event_ops event_ops = {
+        .sample                 = process_sample_event,
+        .comm                   = event__process_comm,
+        .ordered_samples        = true,
+};
+
+static double fragmentation(unsigned long n_req, unsigned long n_alloc)
+{
+        if (n_alloc == 0)
+                return 0.0;
+        else
+                return 100.0 - (100.0 * n_req / n_alloc);
+}
+
+static void __print_result(struct rb_root *root, struct perf_session *session,
+                           int n_lines, int is_caller)
+{
+        struct rb_node *next;
+        struct machine *machine;
+
+        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("%.102s\n", graph_dotted_line);
+
+        next = rb_first(root);
+
+        machine = perf_session__find_host_machine(session);
+        if (!machine) {
+                pr_err("__print_result: couldn't find kernel information\n");
+                return;
+        }
+        while (next && n_lines--) {
+                struct alloc_stat *data = rb_entry(next, struct alloc_stat,
+                                                   node);
+                struct symbol *sym = NULL;
+                struct map *map;
+                char buf[BUFSIZ];
+                u64 addr;
+
+                if (is_caller) {
+                        addr = data->call_site;
+                        if (!raw_ip)
+                                sym = machine__find_kernel_function(machine, addr, &map, NULL);
+                } else
+                        addr = data->ptr;
+
+                if (sym != NULL)
+                        snprintf(buf, sizeof(buf), "%s+%Lx", sym->name,
+                                 addr - map->unmap_ip(map, sym->start));
+                else
+                        snprintf(buf, sizeof(buf), "%#Lx", addr);
+                printf(" %-34s |", buf);
+
+                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,
+                       (unsigned long)data->bytes_req / data->hit,
+                       (unsigned long)data->hit,
+                       (unsigned long)data->pingpong,
+                       fragmentation(data->bytes_req, data->bytes_alloc));
+
+                next = rb_next(next);
+        }
+
+        if (n_lines == -1)
+                printf(" ...                                | ...             | ...             | ...    | ...      | ...   \n");
+
+        printf("%.102s\n", graph_dotted_line);
+}
+
+static void print_summary(void)
+{
+        printf("\nSUMMARY\n=======\n");
+        printf("Total bytes requested: %lu\n", total_requested);
+        printf("Total bytes allocated: %lu\n", total_allocated);
+        printf("Total bytes wasted on internal fragmentation: %lu\n",
+               total_allocated - total_requested);
+        printf("Internal fragmentation: %f%%\n",
+               fragmentation(total_requested, total_allocated));
+        printf("Cross CPU allocations: %lu/%lu\n", nr_cross_allocs, nr_allocs);
+}
+
+static void print_result(struct perf_session *session)
+{
+        if (caller_flag)
+                __print_result(&root_caller_sorted, session, caller_lines, 1);
+        if (alloc_flag)
+                __print_result(&root_alloc_sorted, session, alloc_lines, 0);
+        print_summary();
+}
+
+struct sort_dimension {
+        const char              name[20];
+        sort_fn_t               cmp;
+        struct list_head        list;
+};
+
+static LIST_HEAD(caller_sort);
+static LIST_HEAD(alloc_sort);
+
+static void sort_insert(struct rb_root *root, struct alloc_stat *data,
+                        struct list_head *sort_list)
+{
+        struct rb_node **new = &(root->rb_node);
+        struct rb_node *parent = NULL;
+        struct sort_dimension *sort;
+
+        while (*new) {
+                struct alloc_stat *this;
+                int cmp = 0;
+
+                this = rb_entry(*new, struct alloc_stat, node);
+                parent = *new;
+
+                list_for_each_entry(sort, sort_list, list) {
+                        cmp = sort->cmp(data, this);
+                        if (cmp)
+                                break;
+                }
+
+                if (cmp > 0)
+                        new = &((*new)->rb_left);
+                else
+                        new = &((*new)->rb_right);
+        }
+
+        rb_link_node(&data->node, parent, new);
+        rb_insert_color(&data->node, root);
+}
+
+static void __sort_result(struct rb_root *root, struct rb_root *root_sorted,
+                          struct list_head *sort_list)
+{
+        struct rb_node *node;
+        struct alloc_stat *data;
+
+        for (;;) {
+                node = rb_first(root);
+                if (!node)
+                        break;
+
+                rb_erase(node, root);
+                data = rb_entry(node, struct alloc_stat, node);
+                sort_insert(root_sorted, data, sort_list);
+        }
+}
+
+static void sort_result(void)
+{
+        __sort_result(&root_alloc_stat, &root_alloc_sorted, &alloc_sort);
+        __sort_result(&root_caller_stat, &root_caller_sorted, &caller_sort);
+}
+
+static int __cmd_kmem(void)
+{
+        int err = -EINVAL;
+        struct perf_session *session = perf_session__new(input_name, O_RDONLY,
+                                                         0, false, &event_ops);
+        if (session == NULL)
+                return -ENOMEM;
+
+        if (perf_session__create_kernel_maps(session) < 0)
+                goto out_delete;
+
+        if (!perf_session__has_traces(session, "kmem record"))
+                goto out_delete;
+
+        setup_pager();
+        err = perf_session__process_events(session, &event_ops);
+        if (err != 0)
+                goto out_delete;
+        sort_result();
+        print_result(session);
+out_delete:
+        perf_session__delete(session);
+        return err;
+}
+
+static const char * const kmem_usage[] = {
+        "perf kmem [<options>] {record|stat}",
+        NULL
+};
+
+static int ptr_cmp(struct alloc_stat *l, struct alloc_stat *r)
+{
+        if (l->ptr < r->ptr)
+                return -1;
+        else if (l->ptr > r->ptr)
+                return 1;
+        return 0;
+}
+
+static struct sort_dimension ptr_sort_dimension = {
+        .name   = "ptr",
+        .cmp    = ptr_cmp,
+};
+
+static int callsite_cmp(struct alloc_stat *l, struct alloc_stat *r)
+{
+        if (l->call_site < r->call_site)
+                return -1;
+        else if (l->call_site > r->call_site)
+                return 1;
+        return 0;
+}
+
+static struct sort_dimension callsite_sort_dimension = {
+        .name   = "callsite",
+        .cmp    = callsite_cmp,
+};
+
+static int hit_cmp(struct alloc_stat *l, struct alloc_stat *r)
+{
+        if (l->hit < r->hit)
+                return -1;
+        else if (l->hit > r->hit)
+                return 1;
+        return 0;
+}
+
+static struct sort_dimension hit_sort_dimension = {
+        .name   = "hit",
+        .cmp    = hit_cmp,
+};
+
+static int bytes_cmp(struct alloc_stat *l, struct alloc_stat *r)
+{
+        if (l->bytes_alloc < r->bytes_alloc)
+                return -1;
+        else if (l->bytes_alloc > r->bytes_alloc)
+                return 1;
+        return 0;
+}
+
+static struct sort_dimension bytes_sort_dimension = {
+        .name   = "bytes",
+        .cmp    = bytes_cmp,
+};
+
+static int frag_cmp(struct alloc_stat *l, struct alloc_stat *r)
+{
+        double x, y;
+
+        x = fragmentation(l->bytes_req, l->bytes_alloc);
+        y = fragmentation(r->bytes_req, r->bytes_alloc);
+
+        if (x < y)
+                return -1;
+        else if (x > y)
+                return 1;
+        return 0;
+}
+
+static struct sort_dimension frag_sort_dimension = {
+        .name   = "frag",
+        .cmp    = frag_cmp,
+};
+
+static int pingpong_cmp(struct alloc_stat *l, struct alloc_stat *r)
+{
+        if (l->pingpong < r->pingpong)
+                return -1;
+        else if (l->pingpong > r->pingpong)
+                return 1;
+        return 0;
+}
+
+static struct sort_dimension pingpong_sort_dimension = {
+        .name   = "pingpong",
+        .cmp    = pingpong_cmp,
+};
+
+static struct sort_dimension *avail_sorts[] = {
+        &ptr_sort_dimension,
+        &callsite_sort_dimension,
+        &hit_sort_dimension,
+        &bytes_sort_dimension,
+        &frag_sort_dimension,
+        &pingpong_sort_dimension,
+};
+
+#define NUM_AVAIL_SORTS \
+        (int)(sizeof(avail_sorts) / sizeof(struct sort_dimension *))
+
+static int sort_dimension__add(const char *tok, struct list_head *list)
+{
+        struct sort_dimension *sort;
+        int i;
+
+        for (i = 0; i < NUM_AVAIL_SORTS; i++) {
+                if (!strcmp(avail_sorts[i]->name, tok)) {
+                        sort = malloc(sizeof(*sort));
+                        if (!sort)
+                                die("malloc");
+                        memcpy(sort, avail_sorts[i], sizeof(*sort));
+                        list_add_tail(&sort->list, list);
+                        return 0;
+                }
+        }
+
+        return -1;
+}
+
+static int setup_sorting(struct list_head *sort_list, const char *arg)
+{
+        char *tok;
+        char *str = strdup(arg);
+
+        if (!str)
+                die("strdup");
+
+        while (true) {
+                tok = strsep(&str, ",");
+                if (!tok)
+                        break;
+                if (sort_dimension__add(tok, sort_list) < 0) {
+                        error("Unknown --sort key: '%s'", tok);
+                        return -1;
+                }
+        }
+
+        free(str);
+        return 0;
+}
+
+static int parse_sort_opt(const struct option *opt __used,
+                          const char *arg, int unset __used)
+{
+        if (!arg)
+                return -1;
+
+        if (caller_flag > alloc_flag)
+                return setup_sorting(&caller_sort, arg);
+        else
+                return setup_sorting(&alloc_sort, arg);
+
+        return 0;
+}
+
+static int parse_caller_opt(const struct option *opt __used,
+                          const char *arg __used, int unset __used)
+{
+        caller_flag = (alloc_flag + 1);
+        return 0;
+}
+
+static int parse_alloc_opt(const struct option *opt __used,
+                          const char *arg __used, int unset __used)
+{
+        alloc_flag = (caller_flag + 1);
+        return 0;
+}
+
+static int parse_line_opt(const struct option *opt __used,
+                          const char *arg, int unset __used)
+{
+        int lines;
+
+        if (!arg)
+                return -1;
+
+        lines = strtoul(arg, NULL, 10);
+
+        if (caller_flag > alloc_flag)
+                caller_lines = lines;
+        else
+                alloc_lines = lines;
+
+        return 0;
+}
+
+static const struct option kmem_options[] = {
+        OPT_STRING('i', "input", &input_name, "file",
+                   "input file name"),
+        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 lines",
+                     parse_line_opt),
+        OPT_BOOLEAN(0, "raw-ip", &raw_ip, "show raw ip instead of symbol"),
+        OPT_END()
+};
+
+static const char *record_args[] = {
+        "record",
+        "-a",
+        "-R",
+        "-f",
+        "-c", "1",
+        "-e", "kmem:kmalloc",
+        "-e", "kmem:kmalloc_node",
+        "-e", "kmem:kfree",
+        "-e", "kmem:kmem_cache_alloc",
+        "-e", "kmem:kmem_cache_alloc_node",
+        "-e", "kmem:kmem_cache_free",
+};
+
+static int __cmd_record(int argc, const char **argv)
+{
+        unsigned int rec_argc, i, j;
+        const char **rec_argv;
+
+        rec_argc = ARRAY_SIZE(record_args) + argc - 1;
+        rec_argv = calloc(rec_argc + 1, sizeof(char *));
+
+        if (rec_argv == NULL)
+                return -ENOMEM;
+
+        for (i = 0; i < ARRAY_SIZE(record_args); i++)
+                rec_argv[i] = strdup(record_args[i]);
+
+        for (j = 1; j < (unsigned int)argc; j++, i++)
+                rec_argv[i] = argv[j];
+
+        return cmd_record(i, rec_argv, NULL);
+}
+
+int cmd_kmem(int argc, const char **argv, const char *prefix __used)
+{
+        argc = parse_options(argc, argv, kmem_options, kmem_usage, 0);
+
+        if (!argc)
+                usage_with_options(kmem_usage, kmem_options);
+
+        symbol__init();
+
+        if (!strncmp(argv[0], "rec", 3)) {
+                return __cmd_record(argc, argv);
+        } else if (!strcmp(argv[0], "stat")) {
+                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();
+        } else
+                usage_with_options(kmem_usage, kmem_options);
+
+        return 0;
+}
+
diff --git a/tools/perf/builtin-kvm.c b/tools/perf/builtin-kvm.c
new file mode 100644
index 00000000000..34d1e853829
--- /dev/null
+++ b/tools/perf/builtin-kvm.c
@@ -0,0 +1,144 @@
+#include "builtin.h"
+#include "perf.h"
+
+#include "util/util.h"
+#include "util/cache.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/header.h"
+#include "util/session.h"
+
+#include "util/parse-options.h"
+#include "util/trace-event.h"
+
+#include "util/debug.h"
+
+#include <sys/prctl.h>
+
+#include <semaphore.h>
+#include <pthread.h>
+#include <math.h>
+
+static const char               *file_name;
+static char                     name_buffer[256];
+
+bool                            perf_host = 1;
+bool                            perf_guest;
+
+static const char * const kvm_usage[] = {
+        "perf kvm [<options>] {top|record|report|diff|buildid-list}",
+        NULL
+};
+
+static const struct option kvm_options[] = {
+        OPT_STRING('i', "input", &file_name, "file",
+                   "Input file name"),
+        OPT_STRING('o', "output", &file_name, "file",
+                   "Output file name"),
+        OPT_BOOLEAN(0, "guest", &perf_guest,
+                    "Collect guest os data"),
+        OPT_BOOLEAN(0, "host", &perf_host,
+                    "Collect guest os data"),
+        OPT_STRING(0, "guestmount", &symbol_conf.guestmount, "directory",
+                   "guest mount directory under which every guest os"
+                   " instance has a subdir"),
+        OPT_STRING(0, "guestvmlinux", &symbol_conf.default_guest_vmlinux_name,
+                   "file", "file saving guest os vmlinux"),
+        OPT_STRING(0, "guestkallsyms", &symbol_conf.default_guest_kallsyms,
+                   "file", "file saving guest os /proc/kallsyms"),
+        OPT_STRING(0, "guestmodules", &symbol_conf.default_guest_modules,
+                   "file", "file saving guest os /proc/modules"),
+        OPT_END()
+};
+
+static int __cmd_record(int argc, const char **argv)
+{
+        int rec_argc, i = 0, j;
+        const char **rec_argv;
+
+        rec_argc = argc + 2;
+        rec_argv = calloc(rec_argc + 1, sizeof(char *));
+        rec_argv[i++] = strdup("record");
+        rec_argv[i++] = strdup("-o");
+        rec_argv[i++] = strdup(file_name);
+        for (j = 1; j < argc; j++, i++)
+                rec_argv[i] = argv[j];
+
+        BUG_ON(i != rec_argc);
+
+        return cmd_record(i, rec_argv, NULL);
+}
+
+static int __cmd_report(int argc, const char **argv)
+{
+        int rec_argc, i = 0, j;
+        const char **rec_argv;
+
+        rec_argc = argc + 2;
+        rec_argv = calloc(rec_argc + 1, sizeof(char *));
+        rec_argv[i++] = strdup("report");
+        rec_argv[i++] = strdup("-i");
+        rec_argv[i++] = strdup(file_name);
+        for (j = 1; j < argc; j++, i++)
+                rec_argv[i] = argv[j];
+
+        BUG_ON(i != rec_argc);
+
+        return cmd_report(i, rec_argv, NULL);
+}
+
+static int __cmd_buildid_list(int argc, const char **argv)
+{
+        int rec_argc, i = 0, j;
+        const char **rec_argv;
+
+        rec_argc = argc + 2;
+        rec_argv = calloc(rec_argc + 1, sizeof(char *));
+        rec_argv[i++] = strdup("buildid-list");
+        rec_argv[i++] = strdup("-i");
+        rec_argv[i++] = strdup(file_name);
+        for (j = 1; j < argc; j++, i++)
+                rec_argv[i] = argv[j];
+
+        BUG_ON(i != rec_argc);
+
+        return cmd_buildid_list(i, rec_argv, NULL);
+}
+
+int cmd_kvm(int argc, const char **argv, const char *prefix __used)
+{
+        perf_host = perf_guest = 0;
+
+        argc = parse_options(argc, argv, kvm_options, kvm_usage,
+                        PARSE_OPT_STOP_AT_NON_OPTION);
+        if (!argc)
+                usage_with_options(kvm_usage, kvm_options);
+
+        if (!perf_host)
+                perf_guest = 1;
+
+        if (!file_name) {
+                if (perf_host && !perf_guest)
+                        sprintf(name_buffer, "perf.data.host");
+                else if (!perf_host && perf_guest)
+                        sprintf(name_buffer, "perf.data.guest");
+                else
+                        sprintf(name_buffer, "perf.data.kvm");
+                file_name = name_buffer;
+        }
+
+        if (!strncmp(argv[0], "rec", 3))
+                return __cmd_record(argc, argv);
+        else if (!strncmp(argv[0], "rep", 3))
+                return __cmd_report(argc, argv);
+        else if (!strncmp(argv[0], "diff", 4))
+                return cmd_diff(argc, argv, NULL);
+        else if (!strncmp(argv[0], "top", 3))
+                return cmd_top(argc, argv, NULL);
+        else if (!strncmp(argv[0], "buildid-list", 12))
+                return __cmd_buildid_list(argc, argv);
+        else
+                usage_with_options(kvm_usage, kvm_options);
+
+        return 0;
+}
diff --git a/tools/perf/builtin-list.c b/tools/perf/builtin-list.c
index fe60e37c96e..d88c6961274 100644
--- a/tools/perf/builtin-list.c
+++ b/tools/perf/builtin-list.c
@@ -10,11 +10,12 @@
 
 #include "perf.h"
 
-#include "util/parse-options.h"
 #include "util/parse-events.h"
+#include "util/cache.h"
 
-int cmd_list(int argc, const char **argv, const char *prefix)
+int cmd_list(int argc __used, const char **argv __used, const char *prefix __used)
 {
+        setup_pager();
         print_events();
         return 0;
 }
diff --git a/tools/perf/builtin-lock.c b/tools/perf/builtin-lock.c
new file mode 100644
index 00000000000..b9c6e543297
--- /dev/null
+++ b/tools/perf/builtin-lock.c
@@ -0,0 +1,1004 @@
+#include "builtin.h"
+#include "perf.h"
+
+#include "util/util.h"
+#include "util/cache.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/header.h"
+
+#include "util/parse-options.h"
+#include "util/trace-event.h"
+
+#include "util/debug.h"
+#include "util/session.h"
+
+#include <sys/types.h>
+#include <sys/prctl.h>
+#include <semaphore.h>
+#include <pthread.h>
+#include <math.h>
+#include <limits.h>
+
+#include <linux/list.h>
+#include <linux/hash.h>
+
+static struct perf_session *session;
+
+/* based on kernel/lockdep.c */
+#define LOCKHASH_BITS           12
+#define LOCKHASH_SIZE           (1UL << LOCKHASH_BITS)
+
+static struct list_head lockhash_table[LOCKHASH_SIZE];
+
+#define __lockhashfn(key)       hash_long((unsigned long)key, LOCKHASH_BITS)
+#define lockhashentry(key)      (lockhash_table + __lockhashfn((key)))
+
+struct lock_stat {
+        struct list_head        hash_entry;
+        struct rb_node          rb;             /* used for sorting */
+
+        /*
+         * FIXME: raw_field_value() returns unsigned long long,
+         * so address of lockdep_map should be dealed as 64bit.
+         * Is there more better solution?
+         */
+        void                    *addr;          /* address of lockdep_map, used as ID */
+        char                    *name;          /* for strcpy(), we cannot use const */
+
+        unsigned int            nr_acquire;
+        unsigned int            nr_acquired;
+        unsigned int            nr_contended;
+        unsigned int            nr_release;
+
+        unsigned int            nr_readlock;
+        unsigned int            nr_trylock;
+        /* these times are in nano sec. */
+        u64                     wait_time_total;
+        u64                     wait_time_min;
+        u64                     wait_time_max;
+
+        int                     discard; /* flag of blacklist */
+};
+
+/*
+ * States of lock_seq_stat
+ *
+ * UNINITIALIZED is required for detecting first event of acquire.
+ * As the nature of lock events, there is no guarantee
+ * that the first event for the locks are acquire,
+ * it can be acquired, contended or release.
+ */
+#define SEQ_STATE_UNINITIALIZED      0         /* initial state */
+#define SEQ_STATE_RELEASED      1
+#define SEQ_STATE_ACQUIRING     2
+#define SEQ_STATE_ACQUIRED      3
+#define SEQ_STATE_READ_ACQUIRED 4
+#define SEQ_STATE_CONTENDED     5
+
+/*
+ * MAX_LOCK_DEPTH
+ * Imported from include/linux/sched.h.
+ * Should this be synchronized?
+ */
+#define MAX_LOCK_DEPTH 48
+
+/*
+ * struct lock_seq_stat:
+ * Place to put on state of one lock sequence
+ * 1) acquire -> acquired -> release
+ * 2) acquire -> contended -> acquired -> release
+ * 3) acquire (with read or try) -> release
+ * 4) Are there other patterns?
+ */
+struct lock_seq_stat {
+        struct list_head        list;
+        int                     state;
+        u64                     prev_event_time;
+        void                    *addr;
+
+        int                     read_count;
+};
+
+struct thread_stat {
+        struct rb_node          rb;
+
+        u32                     tid;
+        struct list_head        seq_list;
+};
+
+static struct rb_root           thread_stats;
+
+static struct thread_stat *thread_stat_find(u32 tid)
+{
+        struct rb_node *node;
+        struct thread_stat *st;
+
+        node = thread_stats.rb_node;
+        while (node) {
+                st = container_of(node, struct thread_stat, rb);
+                if (st->tid == tid)
+                        return st;
+                else if (tid < st->tid)
+                        node = node->rb_left;
+                else
+                        node = node->rb_right;
+        }
+
+        return NULL;
+}
+
+static void thread_stat_insert(struct thread_stat *new)
+{
+        struct rb_node **rb = &thread_stats.rb_node;
+        struct rb_node *parent = NULL;
+        struct thread_stat *p;
+
+        while (*rb) {
+                p = container_of(*rb, struct thread_stat, rb);
+                parent = *rb;
+
+                if (new->tid < p->tid)
+                        rb = &(*rb)->rb_left;
+                else if (new->tid > p->tid)
+                        rb = &(*rb)->rb_right;
+                else
+                        BUG_ON("inserting invalid thread_stat\n");
+        }
+
+        rb_link_node(&new->rb, parent, rb);
+        rb_insert_color(&new->rb, &thread_stats);
+}
+
+static struct thread_stat *thread_stat_findnew_after_first(u32 tid)
+{
+        struct thread_stat *st;
+
+        st = thread_stat_find(tid);
+        if (st)
+                return st;
+
+        st = zalloc(sizeof(struct thread_stat));
+        if (!st)
+                die("memory allocation failed\n");
+
+        st->tid = tid;
+        INIT_LIST_HEAD(&st->seq_list);
+
+        thread_stat_insert(st);
+
+        return st;
+}
+
+static struct thread_stat *thread_stat_findnew_first(u32 tid);
+static struct thread_stat *(*thread_stat_findnew)(u32 tid) =
+        thread_stat_findnew_first;
+
+static struct thread_stat *thread_stat_findnew_first(u32 tid)
+{
+        struct thread_stat *st;
+
+        st = zalloc(sizeof(struct thread_stat));
+        if (!st)
+                die("memory allocation failed\n");
+        st->tid = tid;
+        INIT_LIST_HEAD(&st->seq_list);
+
+        rb_link_node(&st->rb, NULL, &thread_stats.rb_node);
+        rb_insert_color(&st->rb, &thread_stats);
+
+        thread_stat_findnew = thread_stat_findnew_after_first;
+        return st;
+}
+
+/* build simple key function one is bigger than two */
+#define SINGLE_KEY(member)                                              \
+        static int lock_stat_key_ ## member(struct lock_stat *one,      \
+                                         struct lock_stat *two)         \
+        {                                                               \
+                return one->member > two->member;                       \
+        }
+
+SINGLE_KEY(nr_acquired)
+SINGLE_KEY(nr_contended)
+SINGLE_KEY(wait_time_total)
+SINGLE_KEY(wait_time_min)
+SINGLE_KEY(wait_time_max)
+
+struct lock_key {
+        /*
+         * name: the value for specify by user
+         * this should be simpler than raw name of member
+         * e.g. nr_acquired -> acquired, wait_time_total -> wait_total
+         */
+        const char              *name;
+        int                     (*key)(struct lock_stat*, struct lock_stat*);
+};
+
+static const char               *sort_key = "acquired";
+
+static int                      (*compare)(struct lock_stat *, struct lock_stat *);
+
+static struct rb_root           result; /* place to store sorted data */
+
+#define DEF_KEY_LOCK(name, fn_suffix)   \
+        { #name, lock_stat_key_ ## fn_suffix }
+struct lock_key keys[] = {
+        DEF_KEY_LOCK(acquired, nr_acquired),
+        DEF_KEY_LOCK(contended, nr_contended),
+        DEF_KEY_LOCK(wait_total, wait_time_total),
+        DEF_KEY_LOCK(wait_min, wait_time_min),
+        DEF_KEY_LOCK(wait_max, wait_time_max),
+
+        /* extra comparisons much complicated should be here */
+
+        { NULL, NULL }
+};
+
+static void select_key(void)
+{
+        int i;
+
+        for (i = 0; keys[i].name; i++) {
+                if (!strcmp(keys[i].name, sort_key)) {
+                        compare = keys[i].key;
+                        return;
+                }
+        }
+
+        die("Unknown compare key:%s\n", sort_key);
+}
+
+static void insert_to_result(struct lock_stat *st,
+                             int (*bigger)(struct lock_stat *, struct lock_stat *))
+{
+        struct rb_node **rb = &result.rb_node;
+        struct rb_node *parent = NULL;
+        struct lock_stat *p;
+
+        while (*rb) {
+                p = container_of(*rb, struct lock_stat, rb);
+                parent = *rb;
+
+                if (bigger(st, p))
+                        rb = &(*rb)->rb_left;
+                else
+                        rb = &(*rb)->rb_right;
+        }
+
+        rb_link_node(&st->rb, parent, rb);
+        rb_insert_color(&st->rb, &result);
+}
+
+/* returns left most element of result, and erase it */
+static struct lock_stat *pop_from_result(void)
+{
+        struct rb_node *node = result.rb_node;
+
+        if (!node)
+                return NULL;
+
+        while (node->rb_left)
+                node = node->rb_left;
+
+        rb_erase(node, &result);
+        return container_of(node, struct lock_stat, rb);
+}
+
+static struct lock_stat *lock_stat_findnew(void *addr, const char *name)
+{
+        struct list_head *entry = lockhashentry(addr);
+        struct lock_stat *ret, *new;
+
+        list_for_each_entry(ret, entry, hash_entry) {
+                if (ret->addr == addr)
+                        return ret;
+        }
+
+        new = zalloc(sizeof(struct lock_stat));
+        if (!new)
+                goto alloc_failed;
+
+        new->addr = addr;
+        new->name = zalloc(sizeof(char) * strlen(name) + 1);
+        if (!new->name)
+                goto alloc_failed;
+        strcpy(new->name, name);
+
+        new->wait_time_min = ULLONG_MAX;
+
+        list_add(&new->hash_entry, entry);
+        return new;
+
+alloc_failed:
+        die("memory allocation failed\n");
+}
+
+static char                     const *input_name = "perf.data";
+
+struct raw_event_sample {
+        u32                     size;
+        char                    data[0];
+};
+
+struct trace_acquire_event {
+        void                    *addr;
+        const char              *name;
+        int                     flag;
+};
+
+struct trace_acquired_event {
+        void                    *addr;
+        const char              *name;
+};
+
+struct trace_contended_event {
+        void                    *addr;
+        const char              *name;
+};
+
+struct trace_release_event {
+        void                    *addr;
+        const char              *name;
+};
+
+struct trace_lock_handler {
+        void (*acquire_event)(struct trace_acquire_event *,
+                              struct event *,
+                              int cpu,
+                              u64 timestamp,
+                              struct thread *thread);
+
+        void (*acquired_event)(struct trace_acquired_event *,
+                               struct event *,
+                               int cpu,
+                               u64 timestamp,
+                               struct thread *thread);
+
+        void (*contended_event)(struct trace_contended_event *,
+                                struct event *,
+                                int cpu,
+                                u64 timestamp,
+                                struct thread *thread);
+
+        void (*release_event)(struct trace_release_event *,
+                              struct event *,
+                              int cpu,
+                              u64 timestamp,
+                              struct thread *thread);
+};
+
+static struct lock_seq_stat *get_seq(struct thread_stat *ts, void *addr)
+{
+        struct lock_seq_stat *seq;
+
+        list_for_each_entry(seq, &ts->seq_list, list) {
+                if (seq->addr == addr)
+                        return seq;
+        }
+
+        seq = zalloc(sizeof(struct lock_seq_stat));
+        if (!seq)
+                die("Not enough memory\n");
+        seq->state = SEQ_STATE_UNINITIALIZED;
+        seq->addr = addr;
+
+        list_add(&seq->list, &ts->seq_list);
+        return seq;
+}
+
+enum broken_state {
+        BROKEN_ACQUIRE,
+        BROKEN_ACQUIRED,
+        BROKEN_CONTENDED,
+        BROKEN_RELEASE,
+        BROKEN_MAX,
+};
+
+static int bad_hist[BROKEN_MAX];
+
+enum acquire_flags {
+        TRY_LOCK = 1,
+        READ_LOCK = 2,
+};
+
+static void
+report_lock_acquire_event(struct trace_acquire_event *acquire_event,
+                        struct event *__event __used,
+                        int cpu __used,
+                        u64 timestamp __used,
+                        struct thread *thread __used)
+{
+        struct lock_stat *ls;
+        struct thread_stat *ts;
+        struct lock_seq_stat *seq;
+
+        ls = lock_stat_findnew(acquire_event->addr, acquire_event->name);
+        if (ls->discard)
+                return;
+
+        ts = thread_stat_findnew(thread->pid);
+        seq = get_seq(ts, acquire_event->addr);
+
+        switch (seq->state) {
+        case SEQ_STATE_UNINITIALIZED:
+        case SEQ_STATE_RELEASED:
+                if (!acquire_event->flag) {
+                        seq->state = SEQ_STATE_ACQUIRING;
+                } else {
+                        if (acquire_event->flag & TRY_LOCK)
+                                ls->nr_trylock++;
+                        if (acquire_event->flag & READ_LOCK)
+                                ls->nr_readlock++;
+                        seq->state = SEQ_STATE_READ_ACQUIRED;
+                        seq->read_count = 1;
+                        ls->nr_acquired++;
+                }
+                break;
+        case SEQ_STATE_READ_ACQUIRED:
+                if (acquire_event->flag & READ_LOCK) {
+                        seq->read_count++;
+                        ls->nr_acquired++;
+                        goto end;
+                } else {
+                        goto broken;
+                }
+                break;
+        case SEQ_STATE_ACQUIRED:
+        case SEQ_STATE_ACQUIRING:
+        case SEQ_STATE_CONTENDED:
+broken:
+                /* broken lock sequence, discard it */
+                ls->discard = 1;
+                bad_hist[BROKEN_ACQUIRE]++;
+                list_del(&seq->list);
+                free(seq);
+                goto end;
+                break;
+        default:
+                BUG_ON("Unknown state of lock sequence found!\n");
+                break;
+        }
+
+        ls->nr_acquire++;
+        seq->prev_event_time = timestamp;
+end:
+        return;
+}
+
+static void
+report_lock_acquired_event(struct trace_acquired_event *acquired_event,
+                         struct event *__event __used,
+                         int cpu __used,
+                         u64 timestamp __used,
+                         struct thread *thread __used)
+{
+        struct lock_stat *ls;
+        struct thread_stat *ts;
+        struct lock_seq_stat *seq;
+        u64 contended_term;
+
+        ls = lock_stat_findnew(acquired_event->addr, acquired_event->name);
+        if (ls->discard)
+                return;
+
+        ts = thread_stat_findnew(thread->pid);
+        seq = get_seq(ts, acquired_event->addr);
+
+        switch (seq->state) {
+        case SEQ_STATE_UNINITIALIZED:
+                /* orphan event, do nothing */
+                return;
+        case SEQ_STATE_ACQUIRING:
+                break;
+        case SEQ_STATE_CONTENDED:
+                contended_term = timestamp - seq->prev_event_time;
+                ls->wait_time_total += contended_term;
+                if (contended_term < ls->wait_time_min)
+                        ls->wait_time_min = contended_term;
+                if (ls->wait_time_max < contended_term)
+                        ls->wait_time_max = contended_term;
+                break;
+        case SEQ_STATE_RELEASED:
+        case SEQ_STATE_ACQUIRED:
+        case SEQ_STATE_READ_ACQUIRED:
+                /* broken lock sequence, discard it */
+                ls->discard = 1;
+                bad_hist[BROKEN_ACQUIRED]++;
+                list_del(&seq->list);
+                free(seq);
+                goto end;
+                break;
+
+        default:
+                BUG_ON("Unknown state of lock sequence found!\n");
+                break;
+        }
+
+        seq->state = SEQ_STATE_ACQUIRED;
+        ls->nr_acquired++;
+        seq->prev_event_time = timestamp;
+end:
+        return;
+}
+
+static void
+report_lock_contended_event(struct trace_contended_event *contended_event,
+                          struct event *__event __used,
+                          int cpu __used,
+                          u64 timestamp __used,
+                          struct thread *thread __used)
+{
+        struct lock_stat *ls;
+        struct thread_stat *ts;
+        struct lock_seq_stat *seq;
+
+        ls = lock_stat_findnew(contended_event->addr, contended_event->name);
+        if (ls->discard)
+                return;
+
+        ts = thread_stat_findnew(thread->pid);
+        seq = get_seq(ts, contended_event->addr);
+
+        switch (seq->state) {
+        case SEQ_STATE_UNINITIALIZED:
+                /* orphan event, do nothing */
+                return;
+        case SEQ_STATE_ACQUIRING:
+                break;
+        case SEQ_STATE_RELEASED:
+        case SEQ_STATE_ACQUIRED:
+        case SEQ_STATE_READ_ACQUIRED:
+        case SEQ_STATE_CONTENDED:
+                /* broken lock sequence, discard it */
+                ls->discard = 1;
+                bad_hist[BROKEN_CONTENDED]++;
+                list_del(&seq->list);
+                free(seq);
+                goto end;
+                break;
+        default:
+                BUG_ON("Unknown state of lock sequence found!\n");
+                break;
+        }
+
+        seq->state = SEQ_STATE_CONTENDED;
+        ls->nr_contended++;
+        seq->prev_event_time = timestamp;
+end:
+        return;
+}
+
+static void
+report_lock_release_event(struct trace_release_event *release_event,
+                        struct event *__event __used,
+                        int cpu __used,
+                        u64 timestamp __used,
+                        struct thread *thread __used)
+{
+        struct lock_stat *ls;
+        struct thread_stat *ts;
+        struct lock_seq_stat *seq;
+
+        ls = lock_stat_findnew(release_event->addr, release_event->name);
+        if (ls->discard)
+                return;
+
+        ts = thread_stat_findnew(thread->pid);
+        seq = get_seq(ts, release_event->addr);
+
+        switch (seq->state) {
+        case SEQ_STATE_UNINITIALIZED:
+                goto end;
+                break;
+        case SEQ_STATE_ACQUIRED:
+                break;
+        case SEQ_STATE_READ_ACQUIRED:
+                seq->read_count--;
+                BUG_ON(seq->read_count < 0);
+                if (!seq->read_count) {
+                        ls->nr_release++;
+                        goto end;
+                }
+                break;
+        case SEQ_STATE_ACQUIRING:
+        case SEQ_STATE_CONTENDED:
+        case SEQ_STATE_RELEASED:
+                /* broken lock sequence, discard it */
+                ls->discard = 1;
+                bad_hist[BROKEN_RELEASE]++;
+                goto free_seq;
+                break;
+        default:
+                BUG_ON("Unknown state of lock sequence found!\n");
+                break;
+        }
+
+        ls->nr_release++;
+free_seq:
+        list_del(&seq->list);
+        free(seq);
+end:
+        return;
+}
+
+/* lock oriented handlers */
+/* TODO: handlers for CPU oriented, thread oriented */
+static struct trace_lock_handler report_lock_ops  = {
+        .acquire_event          = report_lock_acquire_event,
+        .acquired_event         = report_lock_acquired_event,
+        .contended_event        = report_lock_contended_event,
+        .release_event          = report_lock_release_event,
+};
+
+static struct trace_lock_handler *trace_handler;
+
+static void
+process_lock_acquire_event(void *data,
+                           struct event *event __used,
+                           int cpu __used,
+                           u64 timestamp __used,
+                           struct thread *thread __used)
+{
+        struct trace_acquire_event acquire_event;
+        u64 tmp;                /* this is required for casting... */
+
+        tmp = raw_field_value(event, "lockdep_addr", data);
+        memcpy(&acquire_event.addr, &tmp, sizeof(void *));
+        acquire_event.name = (char *)raw_field_ptr(event, "name", data);
+        acquire_event.flag = (int)raw_field_value(event, "flag", data);
+
+        if (trace_handler->acquire_event)
+                trace_handler->acquire_event(&acquire_event, event, cpu, timestamp, thread);
+}
+
+static void
+process_lock_acquired_event(void *data,
+                            struct event *event __used,
+                            int cpu __used,
+                            u64 timestamp __used,
+                            struct thread *thread __used)
+{
+        struct trace_acquired_event acquired_event;
+        u64 tmp;                /* this is required for casting... */
+
+        tmp = raw_field_value(event, "lockdep_addr", data);
+        memcpy(&acquired_event.addr, &tmp, sizeof(void *));
+        acquired_event.name = (char *)raw_field_ptr(event, "name", data);
+
+        if (trace_handler->acquire_event)
+                trace_handler->acquired_event(&acquired_event, event, cpu, timestamp, thread);
+}
+
+static void
+process_lock_contended_event(void *data,
+                             struct event *event __used,
+                             int cpu __used,
+                             u64 timestamp __used,
+                             struct thread *thread __used)
+{
+        struct trace_contended_event contended_event;
+        u64 tmp;                /* this is required for casting... */
+
+        tmp = raw_field_value(event, "lockdep_addr", data);
+        memcpy(&contended_event.addr, &tmp, sizeof(void *));
+        contended_event.name = (char *)raw_field_ptr(event, "name", data);
+
+        if (trace_handler->acquire_event)
+                trace_handler->contended_event(&contended_event, event, cpu, timestamp, thread);
+}
+
+static void
+process_lock_release_event(void *data,
+                           struct event *event __used,
+                           int cpu __used,
+                           u64 timestamp __used,
+                           struct thread *thread __used)
+{
+        struct trace_release_event release_event;
+        u64 tmp;                /* this is required for casting... */
+
+        tmp = raw_field_value(event, "lockdep_addr", data);
+        memcpy(&release_event.addr, &tmp, sizeof(void *));
+        release_event.name = (char *)raw_field_ptr(event, "name", data);
+
+        if (trace_handler->acquire_event)
+                trace_handler->release_event(&release_event, event, cpu, timestamp, thread);
+}
+
+static void
+process_raw_event(void *data, int cpu, u64 timestamp, struct thread *thread)
+{
+        struct event *event;
+        int type;
+
+        type = trace_parse_common_type(data);
+        event = trace_find_event(type);
+
+        if (!strcmp(event->name, "lock_acquire"))
+                process_lock_acquire_event(data, event, cpu, timestamp, thread);
+        if (!strcmp(event->name, "lock_acquired"))
+                process_lock_acquired_event(data, event, cpu, timestamp, thread);
+        if (!strcmp(event->name, "lock_contended"))
+                process_lock_contended_event(data, event, cpu, timestamp, thread);
+        if (!strcmp(event->name, "lock_release"))
+                process_lock_release_event(data, event, cpu, timestamp, thread);
+}
+
+static void print_bad_events(int bad, int total)
+{
+        /* Output for debug, this have to be removed */
+        int i;
+        const char *name[4] =
+                { "acquire", "acquired", "contended", "release" };
+
+        pr_info("\n=== output for debug===\n\n");
+        pr_info("bad: %d, total: %d\n", bad, total);
+        pr_info("bad rate: %f %%\n", (double)bad / (double)total * 100);
+        pr_info("histogram of events caused bad sequence\n");
+        for (i = 0; i < BROKEN_MAX; i++)
+                pr_info(" %10s: %d\n", name[i], bad_hist[i]);
+}
+
+/* TODO: various way to print, coloring, nano or milli sec */
+static void print_result(void)
+{
+        struct lock_stat *st;
+        char cut_name[20];
+        int bad, total;
+
+        pr_info("%20s ", "Name");
+        pr_info("%10s ", "acquired");
+        pr_info("%10s ", "contended");
+
+        pr_info("%15s ", "total wait (ns)");
+        pr_info("%15s ", "max wait (ns)");
+        pr_info("%15s ", "min wait (ns)");
+
+        pr_info("\n\n");
+
+        bad = total = 0;
+        while ((st = pop_from_result())) {
+                total++;
+                if (st->discard) {
+                        bad++;
+                        continue;
+                }
+                bzero(cut_name, 20);
+
+                if (strlen(st->name) < 16) {
+                        /* output raw name */
+                        pr_info("%20s ", st->name);
+                } else {
+                        strncpy(cut_name, st->name, 16);
+                        cut_name[16] = '.';
+                        cut_name[17] = '.';
+                        cut_name[18] = '.';
+                        cut_name[19] = '\0';
+                        /* cut off name for saving output style */
+                        pr_info("%20s ", cut_name);
+                }
+
+                pr_info("%10u ", st->nr_acquired);
+                pr_info("%10u ", st->nr_contended);
+
+                pr_info("%15llu ", st->wait_time_total);
+                pr_info("%15llu ", st->wait_time_max);
+                pr_info("%15llu ", st->wait_time_min == ULLONG_MAX ?
+                       0 : st->wait_time_min);
+                pr_info("\n");
+        }
+
+        print_bad_events(bad, total);
+}
+
+static bool info_threads, info_map;
+
+static void dump_threads(void)
+{
+        struct thread_stat *st;
+        struct rb_node *node;
+        struct thread *t;
+
+        pr_info("%10s: comm\n", "Thread ID");
+
+        node = rb_first(&thread_stats);
+        while (node) {
+                st = container_of(node, struct thread_stat, rb);
+                t = perf_session__findnew(session, st->tid);
+                pr_info("%10d: %s\n", st->tid, t->comm);
+                node = rb_next(node);
+        };
+}
+
+static void dump_map(void)
+{
+        unsigned int i;
+        struct lock_stat *st;
+
+        pr_info("Address of instance: name of class\n");
+        for (i = 0; i < LOCKHASH_SIZE; i++) {
+                list_for_each_entry(st, &lockhash_table[i], hash_entry) {
+                        pr_info(" %p: %s\n", st->addr, st->name);
+                }
+        }
+}
+
+static void dump_info(void)
+{
+        if (info_threads)
+                dump_threads();
+        else if (info_map)
+                dump_map();
+        else
+                die("Unknown type of information\n");
+}
+
+static int process_sample_event(event_t *self, struct sample_data *sample,
+                                struct perf_session *s)
+{
+        struct thread *thread = perf_session__findnew(s, sample->tid);
+
+        if (thread == NULL) {
+                pr_debug("problem processing %d event, skipping it.\n",
+                        self->header.type);
+                return -1;
+        }
+
+        process_raw_event(sample->raw_data, sample->cpu, sample->time, thread);
+
+        return 0;
+}
+
+static struct perf_event_ops eops = {
+        .sample                 = process_sample_event,
+        .comm                   = event__process_comm,
+        .ordered_samples        = true,
+};
+
+static int read_events(void)
+{
+        session = perf_session__new(input_name, O_RDONLY, 0, false, &eops);
+        if (!session)
+                die("Initializing perf session failed\n");
+
+        return perf_session__process_events(session, &eops);
+}
+
+static void sort_result(void)
+{
+        unsigned int i;
+        struct lock_stat *st;
+
+        for (i = 0; i < LOCKHASH_SIZE; i++) {
+                list_for_each_entry(st, &lockhash_table[i], hash_entry) {
+                        insert_to_result(st, compare);
+                }
+        }
+}
+
+static void __cmd_report(void)
+{
+        setup_pager();
+        select_key();
+        read_events();
+        sort_result();
+        print_result();
+}
+
+static const char * const report_usage[] = {
+        "perf lock report [<options>]",
+        NULL
+};
+
+static const struct option report_options[] = {
+        OPT_STRING('k', "key", &sort_key, "acquired",
+                    "key for sorting"),
+        /* TODO: type */
+        OPT_END()
+};
+
+static const char * const info_usage[] = {
+        "perf lock info [<options>]",
+        NULL
+};
+
+static const struct option info_options[] = {
+        OPT_BOOLEAN('t', "threads", &info_threads,
+                    "dump thread list in perf.data"),
+        OPT_BOOLEAN('m', "map", &info_map,
+                    "map of lock instances (name:address table)"),
+        OPT_END()
+};
+
+static const char * const lock_usage[] = {
+        "perf lock [<options>] {record|trace|report}",
+        NULL
+};
+
+static const struct option lock_options[] = {
+        OPT_STRING('i', "input", &input_name, "file", "input file name"),
+        OPT_INCR('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"),
+        OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, "dump raw trace in ASCII"),
+        OPT_END()
+};
+
+static const char *record_args[] = {
+        "record",
+        "-R",
+        "-f",
+        "-m", "1024",
+        "-c", "1",
+        "-e", "lock:lock_acquire:r",
+        "-e", "lock:lock_acquired:r",
+        "-e", "lock:lock_contended:r",
+        "-e", "lock:lock_release:r",
+};
+
+static int __cmd_record(int argc, const char **argv)
+{
+        unsigned int rec_argc, i, j;
+        const char **rec_argv;
+
+        rec_argc = ARRAY_SIZE(record_args) + argc - 1;
+        rec_argv = calloc(rec_argc + 1, sizeof(char *));
+
+        if (rec_argv == NULL)
+                return -ENOMEM;
+
+        for (i = 0; i < ARRAY_SIZE(record_args); i++)
+                rec_argv[i] = strdup(record_args[i]);
+
+        for (j = 1; j < (unsigned int)argc; j++, i++)
+                rec_argv[i] = argv[j];
+
+        BUG_ON(i != rec_argc);
+
+        return cmd_record(i, rec_argv, NULL);
+}
+
+int cmd_lock(int argc, const char **argv, const char *prefix __used)
+{
+        unsigned int i;
+
+        symbol__init();
+        for (i = 0; i < LOCKHASH_SIZE; i++)
+                INIT_LIST_HEAD(lockhash_table + i);
+
+        argc = parse_options(argc, argv, lock_options, lock_usage,
+                             PARSE_OPT_STOP_AT_NON_OPTION);
+        if (!argc)
+                usage_with_options(lock_usage, lock_options);
+
+        if (!strncmp(argv[0], "rec", 3)) {
+                return __cmd_record(argc, argv);
+        } else if (!strncmp(argv[0], "report", 6)) {
+                trace_handler = &report_lock_ops;
+                if (argc) {
+                        argc = parse_options(argc, argv,
+                                             report_options, report_usage, 0);
+                        if (argc)
+                                usage_with_options(report_usage, report_options);
+                }
+                __cmd_report();
+        } else if (!strcmp(argv[0], "script")) {
+                /* Aliased to 'perf script' */
+                return cmd_script(argc, argv, prefix);
+        } else if (!strcmp(argv[0], "info")) {
+                if (argc) {
+                        argc = parse_options(argc, argv,
+                                             info_options, info_usage, 0);
+                        if (argc)
+                                usage_with_options(info_usage, info_options);
+                }
+                /* recycling report_lock_ops */
+                trace_handler = &report_lock_ops;
+                setup_pager();
+                read_events();
+                dump_info();
+        } else {
+                usage_with_options(lock_usage, lock_options);
+        }
+
+        return 0;
+}
diff --git a/tools/perf/builtin-probe.c b/tools/perf/builtin-probe.c
new file mode 100644
index 00000000000..add163c9f0e
--- /dev/null
+++ b/tools/perf/builtin-probe.c
@@ -0,0 +1,330 @@
+/*
+ * builtin-probe.c
+ *
+ * Builtin probe command: Set up probe events by C expression
+ *
+ * Written by Masami Hiramatsu <mhiramat@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+#define _GNU_SOURCE
+#include <sys/utsname.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <errno.h>
+#include <stdio.h>
+#include <unistd.h>
+#include <stdlib.h>
+#include <string.h>
+
+#undef _GNU_SOURCE
+#include "perf.h"
+#include "builtin.h"
+#include "util/util.h"
+#include "util/strlist.h"
+#include "util/symbol.h"
+#include "util/debug.h"
+#include "util/debugfs.h"
+#include "util/parse-options.h"
+#include "util/probe-finder.h"
+#include "util/probe-event.h"
+
+#define MAX_PATH_LEN 256
+
+/* Session management structure */
+static struct {
+        bool list_events;
+        bool force_add;
+        bool show_lines;
+        bool show_vars;
+        bool show_ext_vars;
+        bool mod_events;
+        int nevents;
+        struct perf_probe_event events[MAX_PROBES];
+        struct strlist *dellist;
+        struct line_range line_range;
+        const char *target_module;
+        int max_probe_points;
+} params;
+
+/* Parse an event definition. Note that any error must die. */
+static int parse_probe_event(const char *str)
+{
+        struct perf_probe_event *pev = &params.events[params.nevents];
+        int ret;
+
+        pr_debug("probe-definition(%d): %s\n", params.nevents, str);
+        if (++params.nevents == MAX_PROBES) {
+                pr_err("Too many probes (> %d) were specified.", MAX_PROBES);
+                return -1;
+        }
+
+        /* Parse a perf-probe command into event */
+        ret = parse_perf_probe_command(str, pev);
+        pr_debug("%d arguments\n", pev->nargs);
+
+        return ret;
+}
+
+static int parse_probe_event_argv(int argc, const char **argv)
+{
+        int i, len, ret;
+        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 == NULL)
+                return -ENOMEM;
+        len = 0;
+        for (i = 0; i < argc; i++)
+                len += sprintf(&buf[len], "%s ", argv[i]);
+        params.mod_events = true;
+        ret = parse_probe_event(buf);
+        free(buf);
+        return ret;
+}
+
+static int opt_add_probe_event(const struct option *opt __used,
+                              const char *str, int unset __used)
+{
+        if (str) {
+                params.mod_events = true;
+                return parse_probe_event(str);
+        } else
+                return 0;
+}
+
+static int opt_del_probe_event(const struct option *opt __used,
+                               const char *str, int unset __used)
+{
+        if (str) {
+                params.mod_events = true;
+                if (!params.dellist)
+                        params.dellist = strlist__new(true, NULL);
+                strlist__add(params.dellist, str);
+        }
+        return 0;
+}
+
+#ifdef DWARF_SUPPORT
+static int opt_show_lines(const struct option *opt __used,
+                          const char *str, int unset __used)
+{
+        int ret = 0;
+
+        if (str)
+                ret = parse_line_range_desc(str, &params.line_range);
+        INIT_LIST_HEAD(&params.line_range.line_list);
+        params.show_lines = true;
+
+        return ret;
+}
+
+static int opt_show_vars(const struct option *opt __used,
+                         const char *str, int unset __used)
+{
+        struct perf_probe_event *pev = &params.events[params.nevents];
+        int ret;
+
+        if (!str)
+                return 0;
+
+        ret = parse_probe_event(str);
+        if (!ret && pev->nargs != 0) {
+                pr_err("  Error: '--vars' doesn't accept arguments.\n");
+                return -EINVAL;
+        }
+        params.show_vars = true;
+
+        return ret;
+}
+#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",
+#ifdef DWARF_SUPPORT
+        "perf probe [<options>] --line 'LINEDESC'",
+        "perf probe [<options>] --vars 'PROBEPOINT'",
+#endif
+        NULL
+};
+
+static const struct option options[] = {
+        OPT_INCR('v', "verbose", &verbose,
+                    "be more verbose (show parsed arguments, etc)"),
+        OPT_BOOLEAN('l', "list", &params.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 DWARF_SUPPORT
+                "[EVENT=]FUNC[@SRC][+OFF|%return|:RL|;PT]|SRC:AL|SRC;PT"
+                " [[NAME=]ARG ...]",
+#else
+                "[EVENT=]FUNC[+OFF|%return] [[NAME=]ARG ...]",
+#endif
+                "probe point definition, where\n"
+                "\t\tGROUP:\tGroup name (optional)\n"
+                "\t\tEVENT:\tEvent name\n"
+                "\t\tFUNC:\tFunction name\n"
+                "\t\tOFF:\tOffset from function entry (in byte)\n"
+                "\t\t%return:\tPut the probe at function return\n"
+#ifdef DWARF_SUPPORT
+                "\t\tSRC:\tSource code path\n"
+                "\t\tRL:\tRelative line number from function entry.\n"
+                "\t\tAL:\tAbsolute line number in file.\n"
+                "\t\tPT:\tLazy expression of line code.\n"
+                "\t\tARG:\tProbe argument (local variable name or\n"
+                "\t\t\tkprobe-tracer argument format.)\n",
+#else
+                "\t\tARG:\tProbe argument (kprobe-tracer argument format.)\n",
+#endif
+                opt_add_probe_event),
+        OPT_BOOLEAN('f', "force", &params.force_add, "forcibly add events"
+                    " with existing name"),
+#ifdef DWARF_SUPPORT
+        OPT_CALLBACK('L', "line", NULL,
+                     "FUNC[:RLN[+NUM|-RLN2]]|SRC:ALN[+NUM|-ALN2]",
+                     "Show source code lines.", opt_show_lines),
+        OPT_CALLBACK('V', "vars", NULL,
+                     "FUNC[@SRC][+OFF|%return|:RL|;PT]|SRC:AL|SRC;PT",
+                     "Show accessible variables on PROBEDEF", opt_show_vars),
+        OPT_BOOLEAN('\0', "externs", &params.show_ext_vars,
+                    "Show external variables too (with --vars only)"),
+        OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
+                   "file", "vmlinux pathname"),
+        OPT_STRING('s', "source", &symbol_conf.source_prefix,
+                   "directory", "path to kernel source"),
+        OPT_STRING('m', "module", &params.target_module,
+                   "modname", "target module name"),
+#endif
+        OPT__DRY_RUN(&probe_event_dry_run),
+        OPT_INTEGER('\0', "max-probes", &params.max_probe_points,
+                 "Set how many probe points can be found for a probe."),
+        OPT_END()
+};
+
+int cmd_probe(int argc, const char **argv, const char *prefix __used)
+{
+        int ret;
+
+        argc = parse_options(argc, argv, options, probe_usage,
+                             PARSE_OPT_STOP_AT_NON_OPTION);
+        if (argc > 0) {
+                if (strcmp(argv[0], "-") == 0) {
+                        pr_warning("  Error: '-' is not supported.\n");
+                        usage_with_options(probe_usage, options);
+                }
+                ret = parse_probe_event_argv(argc, argv);
+                if (ret < 0) {
+                        pr_err("  Error: Parse Error.  (%d)\n", ret);
+                        return ret;
+                }
+        }
+
+        if (params.max_probe_points == 0)
+                params.max_probe_points = MAX_PROBES;
+
+        if ((!params.nevents && !params.dellist && !params.list_events &&
+             !params.show_lines))
+                usage_with_options(probe_usage, options);
+
+        /*
+         * Only consider the user's kernel image path if given.
+         */
+        symbol_conf.try_vmlinux_path = (symbol_conf.vmlinux_name == NULL);
+
+        if (params.list_events) {
+                if (params.mod_events) {
+                        pr_err("  Error: Don't use --list with --add/--del.\n");
+                        usage_with_options(probe_usage, options);
+                }
+                if (params.show_lines) {
+                        pr_err("  Error: Don't use --list with --line.\n");
+                        usage_with_options(probe_usage, options);
+                }
+                if (params.show_vars) {
+                        pr_err(" Error: Don't use --list with --vars.\n");
+                        usage_with_options(probe_usage, options);
+                }
+                ret = show_perf_probe_events();
+                if (ret < 0)
+                        pr_err("  Error: Failed to show event list. (%d)\n",
+                               ret);
+                return ret;
+        }
+
+#ifdef DWARF_SUPPORT
+        if (params.show_lines) {
+                if (params.mod_events) {
+                        pr_err("  Error: Don't use --line with"
+                               " --add/--del.\n");
+                        usage_with_options(probe_usage, options);
+                }
+                if (params.show_vars) {
+                        pr_err(" Error: Don't use --line with --vars.\n");
+                        usage_with_options(probe_usage, options);
+                }
+
+                ret = show_line_range(&params.line_range, params.target_module);
+                if (ret < 0)
+                        pr_err("  Error: Failed to show lines. (%d)\n", ret);
+                return ret;
+        }
+        if (params.show_vars) {
+                if (params.mod_events) {
+                        pr_err("  Error: Don't use --vars with"
+                               " --add/--del.\n");
+                        usage_with_options(probe_usage, options);
+                }
+                ret = show_available_vars(params.events, params.nevents,
+                                          params.max_probe_points,
+                                          params.target_module,
+                                          params.show_ext_vars);
+                if (ret < 0)
+                        pr_err("  Error: Failed to show vars. (%d)\n", ret);
+                return ret;
+        }
+#endif
+
+        if (params.dellist) {
+                ret = del_perf_probe_events(params.dellist);
+                strlist__delete(params.dellist);
+                if (ret < 0) {
+                        pr_err("  Error: Failed to delete events. (%d)\n", ret);
+                        return ret;
+                }
+        }
+
+        if (params.nevents) {
+                ret = add_perf_probe_events(params.events, params.nevents,
+                                            params.max_probe_points,
+                                            params.target_module,
+                                            params.force_add);
+                if (ret < 0) {
+                        pr_err("  Error: Failed to add events. (%d)\n", ret);
+                        return ret;
+                }
+        }
+        return 0;
+}
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index d7ebbd75754..7069bd3e90b 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -5,85 +5,92 @@
  * (or a CPU, or a PID) into the perf.data output file - for
  * later analysis via perf report.
  */
+#define _FILE_OFFSET_BITS 64
+
 #include "builtin.h"
 
 #include "perf.h"
 
+#include "util/build-id.h"
 #include "util/util.h"
 #include "util/parse-options.h"
 #include "util/parse-events.h"
-#include "util/string.h"
+
+#include "util/header.h"
+#include "util/event.h"
+#include "util/evsel.h"
+#include "util/debug.h"
+#include "util/session.h"
+#include "util/symbol.h"
+#include "util/cpumap.h"
 
 #include <unistd.h>
 #include <sched.h>
+#include <sys/mman.h>
 
-#define ALIGN(x, a)             __ALIGN_MASK(x, (typeof(x))(a)-1)
-#define __ALIGN_MASK(x, mask)   (((x)+(mask))&~(mask))
+#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y))
 
-static int                      fd[MAX_NR_CPUS][MAX_COUNTERS];
+enum write_mode_t {
+        WRITE_FORCE,
+        WRITE_APPEND
+};
 
-static long                     default_interval                = 100000;
+static u64                      user_interval                   = ULLONG_MAX;
+static u64                      default_interval                =      0;
+static u64                      sample_type;
 
-static int                      nr_cpus                         = 0;
+static struct cpu_map           *cpus;
 static unsigned int             page_size;
-static unsigned int             mmap_pages                      = 128;
-static int                      freq                            = 0;
+static unsigned int             mmap_pages                      =    128;
+static unsigned int             user_freq                       = UINT_MAX;
+static int                      freq                            =   1000;
 static int                      output;
+static int                      pipe_output                     =      0;
 static const char               *output_name                    = "perf.data";
-static int                      group                           = 0;
-static unsigned int             realtime_prio                   = 0;
-static int                      system_wide                     = 0;
-static pid_t                    target_pid                      = -1;
-static int                      inherit                         = 1;
-static int                      force                           = 0;
-static int                      append_file                     = 0;
-static int                      call_graph                      = 0;
-static int                      verbose                         = 0;
-
-static long                     samples;
-static struct timeval           last_read;
-static struct timeval           this_read;
-
-static u64                      bytes_written;
-
-static struct pollfd            event_array[MAX_NR_CPUS * MAX_COUNTERS];
-
-static int                      nr_poll;
-static int                      nr_cpu;
-
-static int                      file_new = 1;
-static struct perf_file_header  file_header;
-
-struct mmap_event {
-        struct perf_event_header        header;
-        u32                             pid;
-        u32                             tid;
-        u64                             start;
-        u64                             len;
-        u64                             pgoff;
-        char                            filename[PATH_MAX];
-};
-
-struct comm_event {
-        struct perf_event_header        header;
-        u32                             pid;
-        u32                             tid;
-        char                            comm[16];
-};
-
+static int                      group                           =      0;
+static int                      realtime_prio                   =      0;
+static bool                     raw_samples                     =  false;
+static bool                     sample_id_all_avail             =   true;
+static bool                     system_wide                     =  false;
+static pid_t                    target_pid                      =     -1;
+static pid_t                    target_tid                      =     -1;
+static struct thread_map        *threads;
+static pid_t                    child_pid                       =     -1;
+static bool                     no_inherit                      =  false;
+static enum write_mode_t        write_mode                      = WRITE_FORCE;
+static bool                     call_graph                      =  false;
+static bool                     inherit_stat                    =  false;
+static bool                     no_samples                      =  false;
+static bool                     sample_address                  =  false;
+static bool                     sample_time                     =  false;
+static bool                     no_buildid                      =  false;
+static bool                     no_buildid_cache                =  false;
+
+static long                     samples                         =      0;
+static u64                      bytes_written                   =      0;
+
+static struct pollfd            *event_array;
+
+static int                      nr_poll                         =      0;
+static int                      nr_cpu                          =      0;
+
+static int                      file_new                        =      1;
+static off_t                    post_processing_offset;
+
+static struct perf_session      *session;
+static const char               *cpu_list;
 
 struct mmap_data {
-        int                     counter;
         void                    *base;
         unsigned int            mask;
         unsigned int            prev;
 };
 
-static struct mmap_data         mmap_array[MAX_NR_CPUS][MAX_COUNTERS];
+static struct mmap_data         mmap_array[MAX_NR_CPUS];
 
 static unsigned long mmap_read_head(struct mmap_data *md)
 {
-        struct perf_counter_mmap_page *pc = md->base;
+        struct perf_event_mmap_page *pc = md->base;
         long head;
 
         head = pc->data_head;
@@ -94,7 +101,7 @@ static unsigned long mmap_read_head(struct mmap_data *md)
 
 static void mmap_write_tail(struct mmap_data *md, unsigned long tail)
 {
-        struct perf_counter_mmap_page *pc = md->base;
+        struct perf_event_mmap_page *pc = md->base;
 
         /*
          * ensure all reads are done before we write the tail out.
@@ -103,6 +110,11 @@ static void mmap_write_tail(struct mmap_data *md, unsigned long tail)
         pc->data_tail = tail;
 }
 
+static void advance_output(size_t size)
+{
+        bytes_written += size;
+}
+
 static void write_output(void *buf, size_t size)
 {
         while (size) {
@@ -118,6 +130,14 @@ static void write_output(void *buf, size_t size)
         }
 }
 
+static int process_synthesized_event(event_t *event,
+                                     struct sample_data *sample __used,
+                                     struct perf_session *self __used)
+{
+        write_output(event, event->header.size);
+        return 0;
+}
+
 static void mmap_read(struct mmap_data *md)
 {
         unsigned int head = mmap_read_head(md);
@@ -127,8 +147,6 @@ static void mmap_read(struct mmap_data *md)
         void *buf;
         int diff;
 
-        gettimeofday(&this_read, NULL);
-
         /*
          * If we're further behind than half the buffer, there's a chance
          * the writer will bite our tail and mess up the samples under us.
@@ -139,23 +157,13 @@ static void mmap_read(struct mmap_data *md)
          */
         diff = head - old;
         if (diff < 0) {
-                struct timeval iv;
-                unsigned long msecs;
-
-                timersub(&this_read, &last_read, &iv);
-                msecs = iv.tv_sec*1000 + iv.tv_usec/1000;
-
-                fprintf(stderr, "WARNING: failed to keep up with mmap data."
-                                "  Last read %lu msecs ago.\n", msecs);
-
+                fprintf(stderr, "WARNING: failed to keep up with mmap data\n");
                 /*
                  * head points to a known good entry, start there.
                  */
                 old = head;
         }
 
-        last_read = this_read;
-
         if (old != head)
                 samples++;
 
@@ -190,370 +198,617 @@ static void sig_handler(int sig)
 
 static void sig_atexit(void)
 {
-        if (signr == -1)
+        if (child_pid > 0)
+                kill(child_pid, SIGTERM);
+
+        if (signr == -1 || signr == SIGUSR1)
                 return;
 
         signal(signr, SIG_DFL);
         kill(getpid(), signr);
 }
 
-static void pid_synthesize_comm_event(pid_t pid, int full)
-{
-        struct comm_event comm_ev;
-        char filename[PATH_MAX];
-        char bf[BUFSIZ];
-        int fd;
-        size_t size;
-        char *field, *sep;
-        DIR *tasks;
-        struct dirent dirent, *next;
-
-        snprintf(filename, sizeof(filename), "/proc/%d/stat", pid);
-
-        fd = open(filename, O_RDONLY);
-        if (fd < 0) {
-                /*
-                 * We raced with a task exiting - just return:
-                 */
-                if (verbose)
-                        fprintf(stderr, "couldn't open %s\n", filename);
-                return;
-        }
-        if (read(fd, bf, sizeof(bf)) < 0) {
-                fprintf(stderr, "couldn't read %s\n", filename);
-                exit(EXIT_FAILURE);
-        }
-        close(fd);
-
-        /* 9027 (cat) R 6747 9027 6747 34816 9027 ... */
-        memset(&comm_ev, 0, sizeof(comm_ev));
-        field = strchr(bf, '(');
-        if (field == NULL)
-                goto out_failure;
-        sep = strchr(++field, ')');
-        if (sep == NULL)
-                goto out_failure;
-        size = sep - field;
-        memcpy(comm_ev.comm, field, size++);
-
-        comm_ev.pid = pid;
-        comm_ev.header.type = PERF_EVENT_COMM;
-        size = ALIGN(size, sizeof(u64));
-        comm_ev.header.size = sizeof(comm_ev) - (sizeof(comm_ev.comm) - size);
-
-        if (!full) {
-                comm_ev.tid = pid;
-
-                write_output(&comm_ev, comm_ev.header.size);
-                return;
-        }
-
-        snprintf(filename, sizeof(filename), "/proc/%d/task", pid);
-
-        tasks = opendir(filename);
-        while (!readdir_r(tasks, &dirent, &next) && next) {
-                char *end;
-                pid = strtol(dirent.d_name, &end, 10);
-                if (*end)
-                        continue;
+static int group_fd;
 
-                comm_ev.tid = pid;
+static struct perf_header_attr *get_header_attr(struct perf_event_attr *a, int nr)
+{
+        struct perf_header_attr *h_attr;
 
-                write_output(&comm_ev, comm_ev.header.size);
+        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(&session->header, h_attr) < 0) {
+                                perf_header_attr__delete(h_attr);
+                                h_attr = NULL;
+                        }
         }
-        closedir(tasks);
-        return;
 
-out_failure:
-        fprintf(stderr, "couldn't get COMM and pgid, malformed %s\n",
-                filename);
-        exit(EXIT_FAILURE);
+        return h_attr;
 }
 
-static void pid_synthesize_mmap_samples(pid_t pid)
+static void create_counter(struct perf_evsel *evsel, int cpu)
 {
-        char filename[PATH_MAX];
-        FILE *fp;
-
-        snprintf(filename, sizeof(filename), "/proc/%d/maps", pid);
+        char *filter = evsel->filter;
+        struct perf_event_attr *attr = &evsel->attr;
+        struct perf_header_attr *h_attr;
+        int track = !evsel->idx; /* only the first counter needs these */
+        int thread_index;
+        int ret;
+        struct {
+                u64 count;
+                u64 time_enabled;
+                u64 time_running;
+                u64 id;
+        } read_data;
+        /*
+         * Check if parse_single_tracepoint_event has already asked for
+         * PERF_SAMPLE_TIME.
+         *
+         * XXX this is kludgy but short term fix for problems introduced by
+         * eac23d1c that broke 'perf script' by having different sample_types
+         * when using multiple tracepoint events when we use a perf binary
+         * that tries to use sample_id_all on an older kernel.
+         *
+         * We need to move counter creation to perf_session, support
+         * different sample_types, etc.
+         */
+        bool time_needed = attr->sample_type & PERF_SAMPLE_TIME;
+
+        attr->read_format       = PERF_FORMAT_TOTAL_TIME_ENABLED |
+                                  PERF_FORMAT_TOTAL_TIME_RUNNING |
+                                  PERF_FORMAT_ID;
+
+        attr->sample_type       |= PERF_SAMPLE_IP | PERF_SAMPLE_TID;
+
+        if (nr_counters > 1)
+                attr->sample_type |= PERF_SAMPLE_ID;
 
-        fp = fopen(filename, "r");
-        if (fp == NULL) {
-                /*
-                 * We raced with a task exiting - just return:
-                 */
-                if (verbose)
-                        fprintf(stderr, "couldn't open %s\n", filename);
-                return;
-        }
-        while (1) {
-                char bf[BUFSIZ], *pbf = bf;
-                struct mmap_event mmap_ev = {
-                        .header.type = PERF_EVENT_MMAP,
-                };
-                int n;
-                size_t size;
-                if (fgets(bf, sizeof(bf), fp) == NULL)
-                        break;
-
-                /* 00400000-0040c000 r-xp 00000000 fd:01 41038  /bin/cat */
-                n = hex2u64(pbf, &mmap_ev.start);
-                if (n < 0)
-                        continue;
-                pbf += n + 1;
-                n = hex2u64(pbf, &mmap_ev.len);
-                if (n < 0)
-                        continue;
-                pbf += n + 3;
-                if (*pbf == 'x') { /* vm_exec */
-                        char *execname = strrchr(bf, ' ');
-
-                        if (execname == NULL || execname[1] != '/')
-                                continue;
-
-                        execname += 1;
-                        size = strlen(execname);
-                        execname[size - 1] = '\0'; /* Remove \n */
-                        memcpy(mmap_ev.filename, execname, size);
-                        size = ALIGN(size, sizeof(u64));
-                        mmap_ev.len -= mmap_ev.start;
-                        mmap_ev.header.size = (sizeof(mmap_ev) -
-                                               (sizeof(mmap_ev.filename) - size));
-                        mmap_ev.pid = pid;
-                        mmap_ev.tid = pid;
-
-                        write_output(&mmap_ev, mmap_ev.header.size);
+        /*
+         * We default some events to a 1 default interval. But keep
+         * it a weak assumption overridable by the user.
+         */
+        if (!attr->sample_period || (user_freq != UINT_MAX &&
+                                     user_interval != ULLONG_MAX)) {
+                if (freq) {
+                        attr->sample_type       |= PERF_SAMPLE_PERIOD;
+                        attr->freq              = 1;
+                        attr->sample_freq       = freq;
+                } else {
+                        attr->sample_period = default_interval;
                 }
         }
 
-        fclose(fp);
-}
-
-static void synthesize_samples(void)
-{
-        DIR *proc;
-        struct dirent dirent, *next;
-
-        proc = opendir("/proc");
-
-        while (!readdir_r(proc, &dirent, &next) && next) {
-                char *end;
-                pid_t pid;
+        if (no_samples)
+                attr->sample_freq = 0;
 
-                pid = strtol(dirent.d_name, &end, 10);
-                if (*end) /* only interested in proper numerical dirents */
-                        continue;
+        if (inherit_stat)
+                attr->inherit_stat = 1;
 
-                pid_synthesize_comm_event(pid, 1);
-                pid_synthesize_mmap_samples(pid);
+        if (sample_address) {
+                attr->sample_type       |= PERF_SAMPLE_ADDR;
+                attr->mmap_data = track;
         }
 
-        closedir(proc);
-}
+        if (call_graph)
+                attr->sample_type       |= PERF_SAMPLE_CALLCHAIN;
 
-static int group_fd;
+        if (system_wide)
+                attr->sample_type       |= PERF_SAMPLE_CPU;
 
-static void create_counter(int counter, int cpu, pid_t pid)
-{
-        struct perf_counter_attr *attr = attrs + counter;
-        int track = 1;
+        if (sample_id_all_avail &&
+            (sample_time || system_wide || !no_inherit || cpu_list))
+                attr->sample_type       |= PERF_SAMPLE_TIME;
 
-        attr->sample_type       = PERF_SAMPLE_IP | PERF_SAMPLE_TID;
+        if (raw_samples) {
+                attr->sample_type       |= PERF_SAMPLE_TIME;
+                attr->sample_type       |= PERF_SAMPLE_RAW;
+                attr->sample_type       |= PERF_SAMPLE_CPU;
+        }
 
-        if (freq) {
-                attr->sample_type       |= PERF_SAMPLE_PERIOD;
-                attr->freq              = 1;
-                attr->sample_freq       = freq;
+        attr->mmap              = track;
+        attr->comm              = track;
+        attr->inherit           = !no_inherit;
+        if (target_pid == -1 && target_tid == -1 && !system_wide) {
+                attr->disabled = 1;
+                attr->enable_on_exec = 1;
         }
+retry_sample_id:
+        attr->sample_id_all = sample_id_all_avail ? 1 : 0;
 
-        if (call_graph)
-                attr->sample_type       |= PERF_SAMPLE_CALLCHAIN;
+        for (thread_index = 0; thread_index < threads->nr; thread_index++) {
+try_again:
+                FD(evsel, nr_cpu, thread_index) = sys_perf_event_open(attr, threads->map[thread_index], cpu, group_fd, 0);
+
+                if (FD(evsel, nr_cpu, thread_index) < 0) {
+                        int err = errno;
+
+                        if (err == EPERM || err == EACCES)
+                                die("Permission error - are you root?\n"
+                                        "\t Consider tweaking"
+                                        " /proc/sys/kernel/perf_event_paranoid.\n");
+                        else if (err ==  ENODEV && cpu_list) {
+                                die("No such device - did you specify"
+                                        " an out-of-range profile CPU?\n");
+                        } else if (err == ENOENT) {
+                                die("%s event is not supported. ",
+                                     event_name(evsel));
+                        } else if (err == EINVAL && sample_id_all_avail) {
+                                /*
+                                 * Old kernel, no attr->sample_id_type_all field
+                                 */
+                                sample_id_all_avail = false;
+                                if (!sample_time && !raw_samples && !time_needed)
+                                        attr->sample_type &= ~PERF_SAMPLE_TIME;
+
+                                goto retry_sample_id;
+                        }
 
-        if (file_new) {
-                file_header.sample_type = attr->sample_type;
-        } else {
-                if (file_header.sample_type != attr->sample_type) {
-                        fprintf(stderr, "incompatible append\n");
+                        /*
+                         * If it's cycles then fall back to hrtimer
+                         * based cpu-clock-tick sw counter, which
+                         * is always available even if no PMU support:
+                         */
+                        if (attr->type == PERF_TYPE_HARDWARE
+                                        && attr->config == PERF_COUNT_HW_CPU_CYCLES) {
+
+                                if (verbose)
+                                        warning(" ... trying to fall back to cpu-clock-ticks\n");
+                                attr->type = PERF_TYPE_SOFTWARE;
+                                attr->config = PERF_COUNT_SW_CPU_CLOCK;
+                                goto try_again;
+                        }
+                        printf("\n");
+                        error("sys_perf_event_open() syscall returned with %d (%s).  /bin/dmesg may provide additional information.\n",
+                              FD(evsel, nr_cpu, thread_index), strerror(err));
+
+#if defined(__i386__) || defined(__x86_64__)
+                        if (attr->type == PERF_TYPE_HARDWARE && err == EOPNOTSUPP)
+                                die("No hardware sampling interrupt available."
+                                    " No APIC? If so then you can boot the kernel"
+                                    " with the \"lapic\" boot parameter to"
+                                    " force-enable it.\n");
+#endif
+
+                        die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
                         exit(-1);
                 }
-        }
 
-        attr->mmap              = track;
-        attr->comm              = track;
-        attr->inherit           = (cpu < 0) && inherit;
-        attr->disabled          = 1;
+                h_attr = get_header_attr(attr, evsel->idx);
+                if (h_attr == NULL)
+                        die("nomem\n");
 
-        track = 0; /* only the first counter needs these */
+                if (!file_new) {
+                        if (memcmp(&h_attr->attr, attr, sizeof(*attr))) {
+                                fprintf(stderr, "incompatible append\n");
+                                exit(-1);
+                        }
+                }
 
-try_again:
-        fd[nr_cpu][counter] = sys_perf_counter_open(attr, pid, cpu, group_fd, 0);
+                if (read(FD(evsel, nr_cpu, thread_index), &read_data, sizeof(read_data)) == -1) {
+                        perror("Unable to read perf file descriptor");
+                        exit(-1);
+                }
 
-        if (fd[nr_cpu][counter] < 0) {
-                int err = errno;
+                if (perf_header_attr__add_id(h_attr, read_data.id) < 0) {
+                        pr_warning("Not enough memory to add id\n");
+                        exit(-1);
+                }
 
-                if (err == EPERM)
-                        die("Permission error - are you root?\n");
+                assert(FD(evsel, nr_cpu, thread_index) >= 0);
+                fcntl(FD(evsel, nr_cpu, thread_index), F_SETFL, O_NONBLOCK);
 
                 /*
-                 * If it's cycles then fall back to hrtimer
-                 * based cpu-clock-tick sw counter, which
-                 * is always available even if no PMU support:
+                 * First counter acts as the group leader:
                  */
-                if (attr->type == PERF_TYPE_HARDWARE
-                        && attr->config == PERF_COUNT_HW_CPU_CYCLES) {
-
-                        if (verbose)
-                                warning(" ... trying to fall back to cpu-clock-ticks\n");
-                        attr->type = PERF_TYPE_SOFTWARE;
-                        attr->config = PERF_COUNT_SW_CPU_CLOCK;
-                        goto try_again;
-                }
-                printf("\n");
-                error("perfcounter syscall returned with %d (%s)\n",
-                        fd[nr_cpu][counter], strerror(err));
-                die("No CONFIG_PERF_COUNTERS=y kernel support configured?\n");
-                exit(-1);
-        }
+                if (group && group_fd == -1)
+                        group_fd = FD(evsel, nr_cpu, thread_index);
+
+                if (evsel->idx || thread_index) {
+                        struct perf_evsel *first;
+                        first = list_entry(evsel_list.next, struct perf_evsel, node);
+                        ret = ioctl(FD(evsel, nr_cpu, thread_index),
+                                    PERF_EVENT_IOC_SET_OUTPUT,
+                                    FD(first, nr_cpu, 0));
+                        if (ret) {
+                                error("failed to set output: %d (%s)\n", errno,
+                                                strerror(errno));
+                                exit(-1);
+                        }
+                } else {
+                        mmap_array[nr_cpu].prev = 0;
+                        mmap_array[nr_cpu].mask = mmap_pages*page_size - 1;
+                        mmap_array[nr_cpu].base = mmap(NULL, (mmap_pages+1)*page_size,
+                                PROT_READ | PROT_WRITE, MAP_SHARED, FD(evsel, nr_cpu, thread_index), 0);
+                        if (mmap_array[nr_cpu].base == MAP_FAILED) {
+                                error("failed to mmap with %d (%s)\n", errno, strerror(errno));
+                                exit(-1);
+                        }
 
-        assert(fd[nr_cpu][counter] >= 0);
-        fcntl(fd[nr_cpu][counter], F_SETFL, O_NONBLOCK);
+                        event_array[nr_poll].fd = FD(evsel, nr_cpu, thread_index);
+                        event_array[nr_poll].events = POLLIN;
+                        nr_poll++;
+                }
 
-        /*
-         * First counter acts as the group leader:
-         */
-        if (group && group_fd == -1)
-                group_fd = fd[nr_cpu][counter];
-
-        event_array[nr_poll].fd = fd[nr_cpu][counter];
-        event_array[nr_poll].events = POLLIN;
-        nr_poll++;
-
-        mmap_array[nr_cpu][counter].counter = counter;
-        mmap_array[nr_cpu][counter].prev = 0;
-        mmap_array[nr_cpu][counter].mask = mmap_pages*page_size - 1;
-        mmap_array[nr_cpu][counter].base = mmap(NULL, (mmap_pages+1)*page_size,
-                        PROT_READ|PROT_WRITE, MAP_SHARED, fd[nr_cpu][counter], 0);
-        if (mmap_array[nr_cpu][counter].base == MAP_FAILED) {
-                error("failed to mmap with %d (%s)\n", errno, strerror(errno));
-                exit(-1);
+                if (filter != NULL) {
+                        ret = ioctl(FD(evsel, nr_cpu, thread_index),
+                                    PERF_EVENT_IOC_SET_FILTER, filter);
+                        if (ret) {
+                                error("failed to set filter with %d (%s)\n", errno,
+                                                strerror(errno));
+                                exit(-1);
+                        }
+                }
         }
 
-        ioctl(fd[nr_cpu][counter], PERF_COUNTER_IOC_ENABLE);
+        if (!sample_type)
+                sample_type = attr->sample_type;
 }
 
-static void open_counters(int cpu, pid_t pid)
+static void open_counters(int cpu)
 {
-        int counter;
-
-        if (pid > 0) {
-                pid_synthesize_comm_event(pid, 0);
-                pid_synthesize_mmap_samples(pid);
-        }
+        struct perf_evsel *pos;
 
         group_fd = -1;
-        for (counter = 0; counter < nr_counters; counter++)
-                create_counter(counter, cpu, pid);
+
+        list_for_each_entry(pos, &evsel_list, node)
+                create_counter(pos, cpu);
 
         nr_cpu++;
 }
 
+static int process_buildids(void)
+{
+        u64 size = lseek(output, 0, SEEK_CUR);
+
+        if (size == 0)
+                return 0;
+
+        session->fd = output;
+        return __perf_session__process_events(session, post_processing_offset,
+                                              size - post_processing_offset,
+                                              size, &build_id__mark_dso_hit_ops);
+}
+
 static void atexit_header(void)
 {
-        file_header.data_size += bytes_written;
+        if (!pipe_output) {
+                session->header.data_size += bytes_written;
+
+                if (!no_buildid)
+                        process_buildids();
+                perf_header__write(&session->header, output, true);
+                perf_session__delete(session);
+                symbol__exit();
+        }
+}
+
+static void event__synthesize_guest_os(struct machine *machine, void *data)
+{
+        int err;
+        struct perf_session *psession = data;
+
+        if (machine__is_host(machine))
+                return;
 
-        if (pwrite(output, &file_header, sizeof(file_header), 0) == -1)
-                perror("failed to write on file headers");
+        /*
+         *As for guest kernel when processing subcommand record&report,
+         *we arrange module mmap prior to guest kernel mmap and trigger
+         *a preload dso because default guest module symbols are loaded
+         *from guest kallsyms instead of /lib/modules/XXX/XXX. This
+         *method is used to avoid symbol missing when the first addr is
+         *in module instead of in guest kernel.
+         */
+        err = event__synthesize_modules(process_synthesized_event,
+                                        psession, machine);
+        if (err < 0)
+                pr_err("Couldn't record guest kernel [%d]'s reference"
+                       " relocation symbol.\n", machine->pid);
+
+        /*
+         * We use _stext for guest kernel because guest kernel's /proc/kallsyms
+         * have no _text sometimes.
+         */
+        err = event__synthesize_kernel_mmap(process_synthesized_event,
+                                            psession, machine, "_text");
+        if (err < 0)
+                err = event__synthesize_kernel_mmap(process_synthesized_event,
+                                                    psession, machine, "_stext");
+        if (err < 0)
+                pr_err("Couldn't record guest kernel [%d]'s reference"
+                       " relocation symbol.\n", machine->pid);
+}
+
+static struct perf_event_header finished_round_event = {
+        .size = sizeof(struct perf_event_header),
+        .type = PERF_RECORD_FINISHED_ROUND,
+};
+
+static void mmap_read_all(void)
+{
+        int i;
+
+        for (i = 0; i < nr_cpu; i++) {
+                if (mmap_array[i].base)
+                        mmap_read(&mmap_array[i]);
+        }
+
+        if (perf_header__has_feat(&session->header, HEADER_TRACE_INFO))
+                write_output(&finished_round_event, sizeof(finished_round_event));
 }
 
 static int __cmd_record(int argc, const char **argv)
 {
-        int i, counter;
+        int i;
         struct stat st;
-        pid_t pid;
         int flags;
-        int ret;
+        int err;
+        unsigned long waking = 0;
+        int child_ready_pipe[2], go_pipe[2];
+        const bool forks = argc > 0;
+        char buf;
+        struct machine *machine;
 
         page_size = sysconf(_SC_PAGE_SIZE);
-        nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
-        assert(nr_cpus <= MAX_NR_CPUS);
-        assert(nr_cpus >= 0);
 
         atexit(sig_atexit);
         signal(SIGCHLD, sig_handler);
         signal(SIGINT, sig_handler);
+        signal(SIGUSR1, sig_handler);
 
-        if (!stat(output_name, &st) && !force && !append_file) {
-                fprintf(stderr, "Error, output file %s exists, use -A to append or -f to overwrite.\n",
-                                output_name);
+        if (forks && (pipe(child_ready_pipe) < 0 || pipe(go_pipe) < 0)) {
+                perror("failed to create pipes");
                 exit(-1);
         }
 
+        if (!strcmp(output_name, "-"))
+                pipe_output = 1;
+        else if (!stat(output_name, &st) && st.st_size) {
+                if (write_mode == WRITE_FORCE) {
+                        char oldname[PATH_MAX];
+                        snprintf(oldname, sizeof(oldname), "%s.old",
+                                 output_name);
+                        unlink(oldname);
+                        rename(output_name, oldname);
+                }
+        } else if (write_mode == WRITE_APPEND) {
+                write_mode = WRITE_FORCE;
+        }
+
         flags = O_CREAT|O_RDWR;
-        if (append_file)
+        if (write_mode == WRITE_APPEND)
                 file_new = 0;
         else
                 flags |= O_TRUNC;
 
-        output = open(output_name, flags, S_IRUSR|S_IWUSR);
+        if (pipe_output)
+                output = STDOUT_FILENO;
+        else
+                output = open(output_name, flags, S_IRUSR | S_IWUSR);
         if (output < 0) {
                 perror("failed to create output file");
                 exit(-1);
         }
 
+        session = perf_session__new(output_name, O_WRONLY,
+                                    write_mode == WRITE_FORCE, false, NULL);
+        if (session == NULL) {
+                pr_err("Not enough memory for reading perf file header\n");
+                return -1;
+        }
+
+        if (!no_buildid)
+                perf_header__set_feat(&session->header, HEADER_BUILD_ID);
+
         if (!file_new) {
-                if (read(output, &file_header, sizeof(file_header)) == -1) {
-                        perror("failed to read file headers");
+                err = perf_header__read(session, output);
+                if (err < 0)
+                        goto out_delete_session;
+        }
+
+        if (have_tracepoints(&evsel_list))
+                perf_header__set_feat(&session->header, HEADER_TRACE_INFO);
+
+        /*
+         * perf_session__delete(session) will be called at atexit_header()
+         */
+        atexit(atexit_header);
+
+        if (forks) {
+                child_pid = fork();
+                if (child_pid < 0) {
+                        perror("failed to fork");
                         exit(-1);
                 }
 
-                lseek(output, file_header.data_size, SEEK_CUR);
+                if (!child_pid) {
+                        if (pipe_output)
+                                dup2(2, 1);
+                        close(child_ready_pipe[0]);
+                        close(go_pipe[1]);
+                        fcntl(go_pipe[0], F_SETFD, FD_CLOEXEC);
+
+                        /*
+                         * Do a dummy execvp to get the PLT entry resolved,
+                         * so we avoid the resolver overhead on the real
+                         * execvp call.
+                         */
+                        execvp("", (char **)argv);
+
+                        /*
+                         * Tell the parent we're ready to go
+                         */
+                        close(child_ready_pipe[1]);
+
+                        /*
+                         * Wait until the parent tells us to go.
+                         */
+                        if (read(go_pipe[0], &buf, 1) == -1)
+                                perror("unable to read pipe");
+
+                        execvp(argv[0], (char **)argv);
+
+                        perror(argv[0]);
+                        kill(getppid(), SIGUSR1);
+                        exit(-1);
+                }
+
+                if (!system_wide && target_tid == -1 && target_pid == -1)
+                        threads->map[0] = child_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]);
         }
 
-        atexit(atexit_header);
+        if (!system_wide && no_inherit && !cpu_list) {
+                open_counters(-1);
+        } else {
+                for (i = 0; i < cpus->nr; i++)
+                        open_counters(cpus->map[i]);
+        }
 
-        if (!system_wide) {
-                open_counters(-1, target_pid != -1 ? target_pid : getpid());
-        } else for (i = 0; i < nr_cpus; i++)
-                open_counters(i, target_pid);
+        perf_session__set_sample_type(session, sample_type);
 
-        if (target_pid == -1 && argc) {
-                pid = fork();
-                if (pid < 0)
-                        perror("failed to fork");
+        if (pipe_output) {
+                err = perf_header__write_pipe(output);
+                if (err < 0)
+                        return err;
+        } else if (file_new) {
+                err = perf_header__write(&session->header, output, false);
+                if (err < 0)
+                        return err;
+        }
 
-                if (!pid) {
-                        if (execvp(argv[0], (char **)argv)) {
-                                perror(argv[0]);
-                                exit(-1);
+        post_processing_offset = lseek(output, 0, SEEK_CUR);
+
+        perf_session__set_sample_id_all(session, sample_id_all_avail);
+
+        if (pipe_output) {
+                err = event__synthesize_attrs(&session->header,
+                                              process_synthesized_event,
+                                              session);
+                if (err < 0) {
+                        pr_err("Couldn't synthesize attrs.\n");
+                        return err;
+                }
+
+                err = event__synthesize_event_types(process_synthesized_event,
+                                                    session);
+                if (err < 0) {
+                        pr_err("Couldn't synthesize event_types.\n");
+                        return err;
+                }
+
+                if (have_tracepoints(&evsel_list)) {
+                        /*
+                         * FIXME err <= 0 here actually means that
+                         * there were no tracepoints so its not really
+                         * an error, just that we don't need to
+                         * synthesize anything.  We really have to
+                         * return this more properly and also
+                         * propagate errors that now are calling die()
+                         */
+                        err = event__synthesize_tracing_data(output, &evsel_list,
+                                                             process_synthesized_event,
+                                                             session);
+                        if (err <= 0) {
+                                pr_err("Couldn't record tracing data.\n");
+                                return err;
                         }
+                        advance_output(err);
                 }
         }
 
+        machine = perf_session__find_host_machine(session);
+        if (!machine) {
+                pr_err("Couldn't find native kernel information.\n");
+                return -1;
+        }
+
+        err = event__synthesize_kernel_mmap(process_synthesized_event,
+                                            session, machine, "_text");
+        if (err < 0)
+                err = event__synthesize_kernel_mmap(process_synthesized_event,
+                                                    session, machine, "_stext");
+        if (err < 0)
+                pr_err("Couldn't record kernel reference relocation symbol\n"
+                       "Symbol resolution may be skewed if relocation was used (e.g. kexec).\n"
+                       "Check /proc/kallsyms permission or run as root.\n");
+
+        err = event__synthesize_modules(process_synthesized_event,
+                                        session, machine);
+        if (err < 0)
+                pr_err("Couldn't record kernel module information.\n"
+                       "Symbol resolution may be skewed if relocation was used (e.g. kexec).\n"
+                       "Check /proc/modules permission or run as root.\n");
+
+        if (perf_guest)
+                perf_session__process_machines(session, event__synthesize_guest_os);
+
+        if (!system_wide)
+                event__synthesize_thread(target_tid, process_synthesized_event,
+                                         session);
+        else
+                event__synthesize_threads(process_synthesized_event, session);
+
         if (realtime_prio) {
                 struct sched_param param;
 
                 param.sched_priority = realtime_prio;
                 if (sched_setscheduler(0, SCHED_FIFO, &param)) {
-                        printf("Could not set realtime priority.\n");
+                        pr_err("Could not set realtime priority.\n");
                         exit(-1);
                 }
         }
 
-        if (system_wide)
-                synthesize_samples();
+        /*
+         * Let the child rip
+         */
+        if (forks)
+                close(go_pipe[1]);
 
-        while (!done) {
+        for (;;) {
                 int hits = samples;
+                int thread;
+
+                mmap_read_all();
 
-                for (i = 0; i < nr_cpu; i++) {
-                        for (counter = 0; counter < nr_counters; counter++)
-                                mmap_read(&mmap_array[i][counter]);
+                if (hits == samples) {
+                        if (done)
+                                break;
+                        err = poll(event_array, nr_poll, -1);
+                        waking++;
                 }
 
-                if (hits == samples)
-                        ret = poll(event_array, nr_poll, 100);
+                if (done) {
+                        for (i = 0; i < nr_cpu; i++) {
+                                struct perf_evsel *pos;
+
+                                list_for_each_entry(pos, &evsel_list, node) {
+                                        for (thread = 0;
+                                                thread < threads->nr;
+                                                thread++)
+                                                ioctl(FD(pos, i, thread),
+                                                        PERF_EVENT_IOC_DISABLE);
+                                }
+                        }
+                }
         }
 
+        if (quiet || signr == SIGUSR1)
+                return 0;
+
+        fprintf(stderr, "[ perf record: Woken up %ld times to write data ]\n", waking);
+
         /*
          * Approximate RIP event size: 24 bytes.
          */
@@ -564,6 +819,10 @@ static int __cmd_record(int argc, const char **argv)
                 bytes_written / 24);
 
         return 0;
+
+out_delete_session:
+        perf_session__delete(session);
+        return err;
 }
 
 static const char * const record_usage[] = {
@@ -572,57 +831,137 @@ static const char * const record_usage[] = {
         NULL
 };
 
-static const struct option options[] = {
+static bool force, append_file;
+
+const struct option record_options[] = {
         OPT_CALLBACK('e', "event", NULL, "event",
                      "event selector. use 'perf list' to list available events",
                      parse_events),
+        OPT_CALLBACK(0, "filter", NULL, "filter",
+                     "event filter", parse_filter),
         OPT_INTEGER('p', "pid", &target_pid,
-                    "record events on existing pid"),
+                    "record events on existing process id"),
+        OPT_INTEGER('t', "tid", &target_tid,
+                    "record events on existing thread id"),
         OPT_INTEGER('r', "realtime", &realtime_prio,
                     "collect data with this RT SCHED_FIFO priority"),
+        OPT_BOOLEAN('R', "raw-samples", &raw_samples,
+                    "collect raw sample records from all opened counters"),
         OPT_BOOLEAN('a', "all-cpus", &system_wide,
                             "system-wide collection from all CPUs"),
         OPT_BOOLEAN('A', "append", &append_file,
                             "append to the output file to do incremental profiling"),
+        OPT_STRING('C', "cpu", &cpu_list, "cpu",
+                    "list of cpus to monitor"),
         OPT_BOOLEAN('f', "force", &force,
-                        "overwrite existing data file"),
-        OPT_LONG('c', "count", &default_interval,
-                    "event period to sample"),
+                        "overwrite existing data file (deprecated)"),
+        OPT_U64('c', "count", &user_interval, "event period to sample"),
         OPT_STRING('o', "output", &output_name, "file",
                     "output file name"),
-        OPT_BOOLEAN('i', "inherit", &inherit,
-                    "child tasks inherit counters"),
-        OPT_INTEGER('F', "freq", &freq,
-                    "profile at this frequency"),
-        OPT_INTEGER('m', "mmap-pages", &mmap_pages,
-                    "number of mmap data pages"),
+        OPT_BOOLEAN('i', "no-inherit", &no_inherit,
+                    "child tasks do not inherit counters"),
+        OPT_UINTEGER('F', "freq", &user_freq, "profile at this frequency"),
+        OPT_UINTEGER('m', "mmap-pages", &mmap_pages, "number of mmap data pages"),
         OPT_BOOLEAN('g', "call-graph", &call_graph,
                     "do call-graph (stack chain/backtrace) recording"),
-        OPT_BOOLEAN('v', "verbose", &verbose,
+        OPT_INCR('v', "verbose", &verbose,
                     "be more verbose (show counter open errors, etc)"),
+        OPT_BOOLEAN('q', "quiet", &quiet, "don't print any message"),
+        OPT_BOOLEAN('s', "stat", &inherit_stat,
+                    "per thread counts"),
+        OPT_BOOLEAN('d', "data", &sample_address,
+                    "Sample addresses"),
+        OPT_BOOLEAN('T', "timestamp", &sample_time, "Sample timestamps"),
+        OPT_BOOLEAN('n', "no-samples", &no_samples,
+                    "don't sample"),
+        OPT_BOOLEAN('N', "no-buildid-cache", &no_buildid_cache,
+                    "do not update the buildid cache"),
+        OPT_BOOLEAN('B', "no-buildid", &no_buildid,
+                    "do not collect buildids in perf.data"),
         OPT_END()
 };
 
-int cmd_record(int argc, const char **argv, const char *prefix)
+int cmd_record(int argc, const char **argv, const char *prefix __used)
 {
-        int counter;
+        int err = -ENOMEM;
+        struct perf_evsel *pos;
+
+        argc = parse_options(argc, argv, record_options, record_usage,
+                            PARSE_OPT_STOP_AT_NON_OPTION);
+        if (!argc && target_pid == -1 && target_tid == -1 &&
+                !system_wide && !cpu_list)
+                usage_with_options(record_usage, record_options);
+
+        if (force && append_file) {
+                fprintf(stderr, "Can't overwrite and append at the same time."
+                                " You need to choose between -f and -A");
+                usage_with_options(record_usage, record_options);
+        } else if (append_file) {
+                write_mode = WRITE_APPEND;
+        } else {
+                write_mode = WRITE_FORCE;
+        }
+
+        symbol__init();
 
-        argc = parse_options(argc, argv, options, record_usage, 0);
-        if (!argc && target_pid == -1 && !system_wide)
-                usage_with_options(record_usage, options);
+        if (no_buildid_cache || no_buildid)
+                disable_buildid_cache();
 
-        if (!nr_counters) {
-                nr_counters     = 1;
-                attrs[0].type   = PERF_TYPE_HARDWARE;
-                attrs[0].config = PERF_COUNT_HW_CPU_CYCLES;
+        if (list_empty(&evsel_list) && perf_evsel_list__create_default() < 0) {
+                pr_err("Not enough memory for event selector list\n");
+                goto out_symbol_exit;
         }
 
-        for (counter = 0; counter < nr_counters; counter++) {
-                if (attrs[counter].sample_period)
-                        continue;
+        if (target_pid != -1)
+                target_tid = target_pid;
 
-                attrs[counter].sample_period = default_interval;
+        threads = thread_map__new(target_pid, target_tid);
+        if (threads == NULL) {
+                pr_err("Problems finding threads of monitor\n");
+                usage_with_options(record_usage, record_options);
         }
 
-        return __cmd_record(argc, argv);
+        cpus = cpu_map__new(cpu_list);
+        if (cpus == NULL) {
+                perror("failed to parse CPUs map");
+                return -1;
+        }
+
+        list_for_each_entry(pos, &evsel_list, node) {
+                if (perf_evsel__alloc_fd(pos, cpus->nr, threads->nr) < 0)
+                        goto out_free_fd;
+        }
+        event_array = malloc((sizeof(struct pollfd) * MAX_NR_CPUS *
+                              MAX_COUNTERS * threads->nr));
+        if (!event_array)
+                goto out_free_fd;
+
+        if (user_interval != ULLONG_MAX)
+                default_interval = user_interval;
+        if (user_freq != UINT_MAX)
+                freq = user_freq;
+
+        /*
+         * User specified count overrides default frequency.
+         */
+        if (default_interval)
+                freq = 0;
+        else if (freq) {
+                default_interval = freq;
+        } else {
+                fprintf(stderr, "frequency and count are zero, aborting\n");
+                err = -EINVAL;
+                goto out_free_event_array;
+        }
+
+        err = __cmd_record(argc, argv);
+
+out_free_event_array:
+        free(event_array);
+out_free_fd:
+        thread_map__delete(threads);
+        threads = NULL;
+out_symbol_exit:
+        symbol__exit();
+        return err;
 }
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 5eb5566f0c9..75183a4518e 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -10,1509 +10,425 @@
 #include "util/util.h"
 
 #include "util/color.h"
-#include "util/list.h"
+#include <linux/list.h>
 #include "util/cache.h"
-#include "util/rbtree.h"
+#include <linux/rbtree.h>
 #include "util/symbol.h"
-#include "util/string.h"
+#include "util/callchain.h"
+#include "util/strlist.h"
+#include "util/values.h"
 
 #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"
 
-#define SHOW_KERNEL     1
-#define SHOW_USER       2
-#define SHOW_HV         4
+#include "util/thread.h"
+#include "util/sort.h"
+#include "util/hist.h"
 
 static char             const *input_name = "perf.data";
-static char             *vmlinux = NULL;
 
-static char             default_sort_order[] = "comm,dso";
-static char             *sort_order = default_sort_order;
+static bool             force, use_tui, use_stdio;
+static bool             hide_unresolved;
+static bool             dont_use_callchains;
 
-static int              input;
-static int              show_mask = SHOW_KERNEL | SHOW_USER | SHOW_HV;
+static bool             show_threads;
+static struct perf_read_values  show_threads_values;
 
-static int              dump_trace = 0;
-#define dprintf(x...)   do { if (dump_trace) printf(x); } while (0)
-#define cdprintf(x...)  do { if (dump_trace) color_fprintf(stdout, color, x); } while (0)
+static const char       default_pretty_printing_style[] = "normal";
+static const char       *pretty_printing_style = default_pretty_printing_style;
 
-static int              verbose;
-#define eprintf(x...)   do { if (verbose) fprintf(stderr, x); } while (0)
+static char             callchain_default_opt[] = "fractal,0.5";
 
-static int              full_paths;
-
-static unsigned long    page_size;
-static unsigned long    mmap_window = 32;
-
-static char             default_parent_pattern[] = "^sys_|^do_page_fault";
-static char             *parent_pattern = default_parent_pattern;
-static regex_t          parent_regex;
-
-static int              exclude_other = 1;
-
-struct ip_event {
-        struct perf_event_header header;
-        u64 ip;
-        u32 pid, tid;
-        unsigned char __more_data[];
-};
-
-struct ip_callchain {
-        u64 nr;
-        u64 ips[0];
-};
-
-struct mmap_event {
-        struct perf_event_header header;
-        u32 pid, tid;
-        u64 start;
-        u64 len;
-        u64 pgoff;
-        char filename[PATH_MAX];
-};
-
-struct comm_event {
-        struct perf_event_header header;
-        u32 pid, tid;
-        char comm[16];
-};
-
-struct fork_event {
-        struct perf_event_header header;
-        u32 pid, ppid;
-};
-
-struct period_event {
-        struct perf_event_header header;
-        u64 time;
-        u64 id;
-        u64 sample_period;
-};
-
-struct lost_event {
-        struct perf_event_header header;
-        u64 id;
-        u64 lost;
-};
-
-typedef union event_union {
-        struct perf_event_header        header;
-        struct ip_event                 ip;
-        struct mmap_event               mmap;
-        struct comm_event               comm;
-        struct fork_event               fork;
-        struct period_event             period;
-        struct lost_event               lost;
-} event_t;
-
-static LIST_HEAD(dsos);
-static struct dso *kernel_dso;
-static struct dso *vdso;
-
-static void dsos__add(struct dso *dso)
-{
-        list_add_tail(&dso->node, &dsos);
-}
-
-static struct dso *dsos__find(const char *name)
-{
-        struct dso *pos;
-
-        list_for_each_entry(pos, &dsos, node)
-                if (strcmp(pos->name, name) == 0)
-                        return pos;
-        return NULL;
-}
-
-static struct dso *dsos__findnew(const char *name)
-{
-        struct dso *dso = dsos__find(name);
-        int nr;
-
-        if (dso)
-                return dso;
-
-        dso = dso__new(name, 0);
-        if (!dso)
-                goto out_delete_dso;
-
-        nr = dso__load(dso, NULL, verbose);
-        if (nr < 0) {
-                eprintf("Failed to open: %s\n", name);
-                goto out_delete_dso;
-        }
-        if (!nr)
-                eprintf("No symbols found in: %s, maybe install a debug package?\n", name);
-
-        dsos__add(dso);
-
-        return dso;
-
-out_delete_dso:
-        dso__delete(dso);
-        return NULL;
-}
-
-static void dsos__fprintf(FILE *fp)
+static struct hists *perf_session__hists_findnew(struct perf_session *self,
+                                                 u64 event_stream, u32 type,
+                                                 u64 config)
 {
-        struct dso *pos;
-
-        list_for_each_entry(pos, &dsos, node)
-                dso__fprintf(pos, fp);
-}
-
-static struct symbol *vdso__find_symbol(struct dso *dso, u64 ip)
-{
-        return dso__find_symbol(kernel_dso, ip);
-}
-
-static int load_kernel(void)
-{
-        int err;
-
-        kernel_dso = dso__new("[kernel]", 0);
-        if (!kernel_dso)
-                return -1;
-
-        err = dso__load_kernel(kernel_dso, vmlinux, NULL, verbose);
-        if (err) {
-                dso__delete(kernel_dso);
-                kernel_dso = NULL;
-        } else
-                dsos__add(kernel_dso);
-
-        vdso = dso__new("[vdso]", 0);
-        if (!vdso)
-                return -1;
-
-        vdso->find_symbol = vdso__find_symbol;
-
-        dsos__add(vdso);
-
-        return err;
-}
-
-static char __cwd[PATH_MAX];
-static char *cwd = __cwd;
-static int cwdlen;
-
-static int strcommon(const char *pathname)
-{
-        int n = 0;
-
-        while (pathname[n] == cwd[n] && n < cwdlen)
-                ++n;
-
-        return n;
-}
-
-struct map {
-        struct list_head node;
-        u64      start;
-        u64      end;
-        u64      pgoff;
-        u64      (*map_ip)(struct map *, u64);
-        struct dso       *dso;
-};
-
-static u64 map__map_ip(struct map *map, u64 ip)
-{
-        return ip - map->start + map->pgoff;
-}
-
-static u64 vdso__map_ip(struct map *map, u64 ip)
-{
-        return ip;
-}
-
-static inline int is_anon_memory(const char *filename)
-{
-     return strcmp(filename, "//anon") == 0;
-}
-
-static struct map *map__new(struct mmap_event *event)
-{
-        struct map *self = malloc(sizeof(*self));
-
-        if (self != NULL) {
-                const char *filename = event->filename;
-                char newfilename[PATH_MAX];
-                int anon;
-
-                if (cwd) {
-                        int n = strcommon(filename);
-
-                        if (n == cwdlen) {
-                                snprintf(newfilename, sizeof(newfilename),
-                                         ".%s", filename + n);
-                                filename = newfilename;
-                        }
-                }
-
-                anon = is_anon_memory(filename);
-
-                if (anon) {
-                        snprintf(newfilename, sizeof(newfilename), "/tmp/perf-%d.map", event->pid);
-                        filename = newfilename;
-                }
+        struct rb_node **p = &self->hists_tree.rb_node;
+        struct rb_node *parent = NULL;
+        struct hists *iter, *new;
 
-                self->start = event->start;
-                self->end   = event->start + event->len;
-                self->pgoff = event->pgoff;
+        while (*p != NULL) {
+                parent = *p;
+                iter = rb_entry(parent, struct hists, rb_node);
+                if (iter->config == config)
+                        return iter;
 
-                self->dso = dsos__findnew(filename);
-                if (self->dso == NULL)
-                        goto out_delete;
 
-                if (self->dso == vdso || anon)
-                        self->map_ip = vdso__map_ip;
+                if (config > iter->config)
+                        p = &(*p)->rb_right;
                 else
-                        self->map_ip = map__map_ip;
+                        p = &(*p)->rb_left;
         }
-        return self;
-out_delete:
-        free(self);
-        return NULL;
-}
 
-static struct map *map__clone(struct map *self)
-{
-        struct map *map = malloc(sizeof(*self));
-
-        if (!map)
+        new = malloc(sizeof(struct hists));
+        if (new == NULL)
                 return NULL;
+        memset(new, 0, sizeof(struct hists));
+        new->event_stream = event_stream;
+        new->config = config;
+        new->type = type;
+        rb_link_node(&new->rb_node, parent, p);
+        rb_insert_color(&new->rb_node, &self->hists_tree);
+        return new;
+}
+
+static int perf_session__add_hist_entry(struct perf_session *self,
+                                        struct addr_location *al,
+                                        struct sample_data *data)
+{
+        struct map_symbol *syms = NULL;
+        struct symbol *parent = NULL;
+        int err = -ENOMEM;
+        struct hist_entry *he;
+        struct hists *hists;
+        struct perf_event_attr *attr;
 
-        memcpy(map, self, sizeof(*self));
-
-        return map;
-}
-
-static int map__overlap(struct map *l, struct map *r)
-{
-        if (l->start > r->start) {
-                struct map *t = l;
-                l = r;
-                r = t;
+        if ((sort__has_parent || symbol_conf.use_callchain) && data->callchain) {
+                syms = perf_session__resolve_callchain(self, al->thread,
+                                                       data->callchain, &parent);
+                if (syms == NULL)
+                        return -ENOMEM;
         }
 
-        if (l->end > r->start)
-                return 1;
-
-        return 0;
-}
-
-static size_t map__fprintf(struct map *self, FILE *fp)
-{
-        return fprintf(fp, " %Lx-%Lx %Lx %s\n",
-                       self->start, self->end, self->pgoff, self->dso->name);
-}
-
-
-struct thread {
-        struct rb_node   rb_node;
-        struct list_head maps;
-        pid_t            pid;
-        char             *comm;
-};
-
-static struct thread *thread__new(pid_t pid)
-{
-        struct thread *self = malloc(sizeof(*self));
-
-        if (self != NULL) {
-                self->pid = pid;
-                self->comm = malloc(32);
-                if (self->comm)
-                        snprintf(self->comm, 32, ":%d", self->pid);
-                INIT_LIST_HEAD(&self->maps);
+        attr = perf_header__find_attr(data->id, &self->header);
+        if (attr)
+                hists = perf_session__hists_findnew(self, data->id, attr->type, attr->config);
+        else
+                hists = perf_session__hists_findnew(self, data->id, 0, 0);
+        if (hists == NULL)
+                goto out_free_syms;
+        he = __hists__add_entry(hists, al, parent, data->period);
+        if (he == NULL)
+                goto out_free_syms;
+        err = 0;
+        if (symbol_conf.use_callchain) {
+                err = callchain_append(he->callchain, data->callchain, syms,
+                                       data->period);
+                if (err)
+                        goto out_free_syms;
         }
-
-        return self;
-}
-
-static int thread__set_comm(struct thread *self, const char *comm)
-{
-        if (self->comm)
-                free(self->comm);
-        self->comm = strdup(comm);
-        return self->comm ? 0 : -ENOMEM;
-}
-
-static size_t thread__fprintf(struct thread *self, FILE *fp)
-{
-        struct map *pos;
-        size_t ret = fprintf(fp, "Thread %d %s\n", self->pid, self->comm);
-
-        list_for_each_entry(pos, &self->maps, node)
-                ret += map__fprintf(pos, fp);
-
-        return ret;
-}
-
-
-static struct rb_root threads;
-static struct thread *last_match;
-
-static struct thread *threads__findnew(pid_t pid)
-{
-        struct rb_node **p = &threads.rb_node;
-        struct rb_node *parent = NULL;
-        struct thread *th;
-
         /*
-         * Font-end cache - PID lookups come in blocks,
-         * so most of the time we dont have to look up
-         * the full rbtree:
+         * Only in the newt browser we are doing integrated annotation,
+         * so we don't allocated the extra space needed because the stdio
+         * code will not use it.
          */
-        if (last_match && last_match->pid == pid)
-                return last_match;
-
-        while (*p != NULL) {
-                parent = *p;
-                th = rb_entry(parent, struct thread, rb_node);
-
-                if (th->pid == pid) {
-                        last_match = th;
-                        return th;
-                }
-
-                if (pid < th->pid)
-                        p = &(*p)->rb_left;
-                else
-                        p = &(*p)->rb_right;
-        }
-
-        th = thread__new(pid);
-        if (th != NULL) {
-                rb_link_node(&th->rb_node, parent, p);
-                rb_insert_color(&th->rb_node, &threads);
-                last_match = th;
-        }
-
-        return th;
+        if (use_browser > 0)
+                err = hist_entry__inc_addr_samples(he, al->addr);
+out_free_syms:
+        free(syms);
+        return err;
 }
 
-static void thread__insert_map(struct thread *self, struct map *map)
+static int add_event_total(struct perf_session *session,
+                           struct sample_data *data,
+                           struct perf_event_attr *attr)
 {
-        struct map *pos, *tmp;
+        struct hists *hists;
 
-        list_for_each_entry_safe(pos, tmp, &self->maps, node) {
-                if (map__overlap(pos, map)) {
-                        list_del_init(&pos->node);
-                        /* XXX leaks dsos */
-                        free(pos);
-                }
-        }
-
-        list_add_tail(&map->node, &self->maps);
-}
-
-static int thread__fork(struct thread *self, struct thread *parent)
-{
-        struct map *map;
+        if (attr)
+                hists = perf_session__hists_findnew(session, data->id,
+                                                    attr->type, attr->config);
+        else
+                hists = perf_session__hists_findnew(session, data->id, 0, 0);
 
-        if (self->comm)
-                free(self->comm);
-        self->comm = strdup(parent->comm);
-        if (!self->comm)
+        if (!hists)
                 return -ENOMEM;
 
-        list_for_each_entry(map, &parent->maps, node) {
-                struct map *new = map__clone(map);
-                if (!new)
-                        return -ENOMEM;
-                thread__insert_map(self, new);
-        }
-
+        hists->stats.total_period += data->period;
+        /*
+         * FIXME: add_event_total should be moved from here to
+         * perf_session__process_event so that the proper hist is passed to
+         * the event_op methods.
+         */
+        hists__inc_nr_events(hists, PERF_RECORD_SAMPLE);
+        session->hists.stats.total_period += data->period;
         return 0;
 }
 
-static struct map *thread__find_map(struct thread *self, u64 ip)
-{
-        struct map *pos;
-
-        if (self == NULL)
-                return NULL;
-
-        list_for_each_entry(pos, &self->maps, node)
-                if (ip >= pos->start && ip <= pos->end)
-                        return pos;
-
-        return NULL;
-}
-
-static size_t threads__fprintf(FILE *fp)
+static int process_sample_event(event_t *event, struct sample_data *sample,
+                                struct perf_session *session)
 {
-        size_t ret = 0;
-        struct rb_node *nd;
-
-        for (nd = rb_first(&threads); nd; nd = rb_next(nd)) {
-                struct thread *pos = rb_entry(nd, struct thread, rb_node);
-
-                ret += thread__fprintf(pos, fp);
-        }
-
-        return ret;
-}
-
-/*
- * histogram, sorted on item, collects counts
- */
-
-static struct rb_root hist;
-
-struct hist_entry {
-        struct rb_node   rb_node;
-
-        struct thread    *thread;
-        struct map       *map;
-        struct dso       *dso;
-        struct symbol    *sym;
-        struct symbol    *parent;
-        u64              ip;
-        char             level;
-
-        u64              count;
-};
-
-/*
- * configurable sorting bits
- */
+        struct addr_location al;
+        struct perf_event_attr *attr;
 
-struct sort_entry {
-        struct list_head list;
-
-        char *header;
-
-        int64_t (*cmp)(struct hist_entry *, struct hist_entry *);
-        int64_t (*collapse)(struct hist_entry *, struct hist_entry *);
-        size_t  (*print)(FILE *fp, struct hist_entry *);
-};
-
-static int64_t cmp_null(void *l, void *r)
-{
-        if (!l && !r)
-                return 0;
-        else if (!l)
+        if (event__preprocess_sample(event, session, &al, sample, NULL) < 0) {
+                fprintf(stderr, "problem processing %d event, skipping it.\n",
+                        event->header.type);
                 return -1;
-        else
-                return 1;
-}
-
-/* --sort pid */
-
-static int64_t
-sort__thread_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-        return right->thread->pid - left->thread->pid;
-}
-
-static size_t
-sort__thread_print(FILE *fp, struct hist_entry *self)
-{
-        return fprintf(fp, "%16s:%5d", self->thread->comm ?: "", self->thread->pid);
-}
-
-static struct sort_entry sort_thread = {
-        .header = "         Command:  Pid",
-        .cmp    = sort__thread_cmp,
-        .print  = sort__thread_print,
-};
-
-/* --sort comm */
-
-static int64_t
-sort__comm_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-        return right->thread->pid - left->thread->pid;
-}
-
-static int64_t
-sort__comm_collapse(struct hist_entry *left, struct hist_entry *right)
-{
-        char *comm_l = left->thread->comm;
-        char *comm_r = right->thread->comm;
-
-        if (!comm_l || !comm_r)
-                return cmp_null(comm_l, comm_r);
-
-        return strcmp(comm_l, comm_r);
-}
-
-static size_t
-sort__comm_print(FILE *fp, struct hist_entry *self)
-{
-        return fprintf(fp, "%16s", self->thread->comm);
-}
-
-static struct sort_entry sort_comm = {
-        .header         = "         Command",
-        .cmp            = sort__comm_cmp,
-        .collapse       = sort__comm_collapse,
-        .print          = sort__comm_print,
-};
-
-/* --sort dso */
-
-static int64_t
-sort__dso_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-        struct dso *dso_l = left->dso;
-        struct dso *dso_r = right->dso;
-
-        if (!dso_l || !dso_r)
-                return cmp_null(dso_l, dso_r);
-
-        return strcmp(dso_l->name, dso_r->name);
-}
-
-static size_t
-sort__dso_print(FILE *fp, struct hist_entry *self)
-{
-        if (self->dso)
-                return fprintf(fp, "%-25s", self->dso->name);
-
-        return fprintf(fp, "%016llx         ", (u64)self->ip);
-}
-
-static struct sort_entry sort_dso = {
-        .header = "Shared Object            ",
-        .cmp    = sort__dso_cmp,
-        .print  = sort__dso_print,
-};
-
-/* --sort symbol */
-
-static int64_t
-sort__sym_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-        u64 ip_l, ip_r;
-
-        if (left->sym == right->sym)
-                return 0;
-
-        ip_l = left->sym ? left->sym->start : left->ip;
-        ip_r = right->sym ? right->sym->start : right->ip;
-
-        return (int64_t)(ip_r - ip_l);
-}
-
-static size_t
-sort__sym_print(FILE *fp, struct hist_entry *self)
-{
-        size_t ret = 0;
-
-        if (verbose)
-                ret += fprintf(fp, "%#018llx  ", (u64)self->ip);
-
-        if (self->sym) {
-                ret += fprintf(fp, "[%c] %s",
-                        self->dso == kernel_dso ? 'k' : '.', self->sym->name);
-        } else {
-                ret += fprintf(fp, "%#016llx", (u64)self->ip);
         }
 
-        return ret;
-}
-
-static struct sort_entry sort_sym = {
-        .header = "Symbol",
-        .cmp    = sort__sym_cmp,
-        .print  = sort__sym_print,
-};
-
-/* --sort parent */
-
-static int64_t
-sort__parent_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-        struct symbol *sym_l = left->parent;
-        struct symbol *sym_r = right->parent;
-
-        if (!sym_l || !sym_r)
-                return cmp_null(sym_l, sym_r);
-
-        return strcmp(sym_l->name, sym_r->name);
-}
-
-static size_t
-sort__parent_print(FILE *fp, struct hist_entry *self)
-{
-        size_t ret = 0;
-
-        ret += fprintf(fp, "%-20s", self->parent ? self->parent->name : "[other]");
-
-        return ret;
-}
-
-static struct sort_entry sort_parent = {
-        .header = "Parent symbol       ",
-        .cmp    = sort__parent_cmp,
-        .print  = sort__parent_print,
-};
-
-static int sort__need_collapse = 0;
-static int sort__has_parent = 0;
-
-struct sort_dimension {
-        char                    *name;
-        struct sort_entry       *entry;
-        int                     taken;
-};
-
-static struct sort_dimension sort_dimensions[] = {
-        { .name = "pid",        .entry = &sort_thread,  },
-        { .name = "comm",       .entry = &sort_comm,    },
-        { .name = "dso",        .entry = &sort_dso,     },
-        { .name = "symbol",     .entry = &sort_sym,     },
-        { .name = "parent",     .entry = &sort_parent,  },
-};
-
-static LIST_HEAD(hist_entry__sort_list);
-
-static int sort_dimension__add(char *tok)
-{
-        int i;
-
-        for (i = 0; i < ARRAY_SIZE(sort_dimensions); i++) {
-                struct sort_dimension *sd = &sort_dimensions[i];
-
-                if (sd->taken)
-                        continue;
-
-                if (strncasecmp(tok, sd->name, strlen(tok)))
-                        continue;
-
-                if (sd->entry->collapse)
-                        sort__need_collapse = 1;
-
-                if (sd->entry == &sort_parent) {
-                        int ret = regcomp(&parent_regex, parent_pattern, REG_EXTENDED);
-                        if (ret) {
-                                char err[BUFSIZ];
-
-                                regerror(ret, &parent_regex, err, sizeof(err));
-                                fprintf(stderr, "Invalid regex: %s\n%s",
-                                        parent_pattern, err);
-                                exit(-1);
-                        }
-                        sort__has_parent = 1;
-                }
-
-                list_add_tail(&sd->entry->list, &hist_entry__sort_list);
-                sd->taken = 1;
-
+        if (al.filtered || (hide_unresolved && al.sym == NULL))
                 return 0;
-        }
-
-        return -ESRCH;
-}
-
-static int64_t
-hist_entry__cmp(struct hist_entry *left, struct hist_entry *right)
-{
-        struct sort_entry *se;
-        int64_t cmp = 0;
-
-        list_for_each_entry(se, &hist_entry__sort_list, list) {
-                cmp = se->cmp(left, right);
-                if (cmp)
-                        break;
-        }
-
-        return cmp;
-}
-
-static int64_t
-hist_entry__collapse(struct hist_entry *left, struct hist_entry *right)
-{
-        struct sort_entry *se;
-        int64_t cmp = 0;
 
-        list_for_each_entry(se, &hist_entry__sort_list, list) {
-                int64_t (*f)(struct hist_entry *, struct hist_entry *);
-
-                f = se->collapse ?: se->cmp;
-
-                cmp = f(left, right);
-                if (cmp)
-                        break;
+        if (perf_session__add_hist_entry(session, &al, sample)) {
+                pr_debug("problem incrementing symbol period, skipping event\n");
+                return -1;
         }
 
-        return cmp;
-}
-
-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) {
-                double percent = self->count * 100.0 / total_samples;
-                char *color = PERF_COLOR_NORMAL;
+        attr = perf_header__find_attr(sample->id, &session->header);
 
-                /*
-                 * We color high-overhead entries in red, mid-overhead
-                 * entries in green - and keep the low overhead places
-                 * normal:
-                 */
-                if (percent >= 5.0) {
-                        color = PERF_COLOR_RED;
-                } else {
-                        if (percent >= 0.5)
-                                color = PERF_COLOR_GREEN;
-                }
-
-                ret = color_fprintf(fp, color, "   %6.2f%%",
-                                (self->count * 100.0) / total_samples);
-        } else
-                ret = fprintf(fp, "%12Ld ", self->count);
-
-        list_for_each_entry(se, &hist_entry__sort_list, list) {
-                if (exclude_other && (se == &sort_parent))
-                        continue;
-
-                fprintf(fp, "  ");
-                ret += se->print(fp, self);
+        if (add_event_total(session, sample, attr)) {
+                pr_debug("problem adding event period\n");
+                return -1;
         }
 
-        ret += fprintf(fp, "\n");
-
-        return ret;
+        return 0;
 }
 
-/*
- *
- */
-
-static struct symbol *
-resolve_symbol(struct thread *thread, struct map **mapp,
-               struct dso **dsop, u64 *ipp)
+static int process_read_event(event_t *event, struct sample_data *sample __used,
+                              struct perf_session *session __used)
 {
-        struct dso *dso = dsop ? *dsop : NULL;
-        struct map *map = mapp ? *mapp : NULL;
-        uint64_t ip = *ipp;
-
-        if (!thread)
-                return NULL;
-
-        if (dso)
-                goto got_dso;
+        struct perf_event_attr *attr;
 
-        if (map)
-                goto got_map;
+        attr = perf_header__find_attr(event->read.id, &session->header);
 
-        map = thread__find_map(thread, ip);
-        if (map != NULL) {
-                if (mapp)
-                        *mapp = map;
-got_map:
-                ip = map->map_ip(map, ip);
-                *ipp  = ip;
-
-                dso = map->dso;
-        } else {
-                /*
-                 * If this is outside of all known maps,
-                 * and is a negative address, try to look it
-                 * up in the kernel dso, as it might be a
-                 * vsyscall (which executes in user-mode):
-                 */
-                if ((long long)ip < 0)
-                dso = kernel_dso;
+        if (show_threads) {
+                const char *name = attr ? __event_name(attr->type, attr->config)
+                                   : "unknown";
+                perf_read_values_add_value(&show_threads_values,
+                                           event->read.pid, event->read.tid,
+                                           event->read.id,
+                                           name,
+                                           event->read.value);
         }
-        dprintf(" ...... dso: %s\n", dso ? dso->name : "<not found>");
-
-        if (dsop)
-                *dsop = dso;
-
-        if (!dso)
-                return NULL;
-got_dso:
-        return dso->find_symbol(dso, ip);
-}
 
-static int call__match(struct symbol *sym)
-{
-        if (sym->name && !regexec(&parent_regex, sym->name, 0, NULL, 0))
-                return 1;
+        dump_printf(": %d %d %s %Lu\n", event->read.pid, event->read.tid,
+                    attr ? __event_name(attr->type, attr->config) : "FAIL",
+                    event->read.value);
 
         return 0;
 }
 
-/*
- * collect histogram counts
- */
-
-static int
-hist_entry__add(struct thread *thread, struct map *map, struct dso *dso,
-                struct symbol *sym, u64 ip, struct ip_callchain *chain,
-                char level, u64 count)
+static int perf_session__setup_sample_type(struct perf_session *self)
 {
-        struct rb_node **p = &hist.rb_node;
-        struct rb_node *parent = NULL;
-        struct hist_entry *he;
-        struct hist_entry entry = {
-                .thread = thread,
-                .map    = map,
-                .dso    = dso,
-                .sym    = sym,
-                .ip     = ip,
-                .level  = level,
-                .count  = count,
-                .parent = NULL,
-        };
-        int cmp;
-
-        if (sort__has_parent && chain) {
-                u64 context = PERF_CONTEXT_MAX;
-                int i;
-
-                for (i = 0; i < chain->nr; i++) {
-                        u64 ip = chain->ips[i];
-                        struct dso *dso = NULL;
-                        struct symbol *sym;
-
-                        if (ip >= PERF_CONTEXT_MAX) {
-                                context = ip;
-                                continue;
-                        }
-
-                        switch (context) {
-                        case PERF_CONTEXT_KERNEL:
-                                dso = kernel_dso;
-                                break;
-                        default:
-                                break;
-                        }
-
-                        sym = resolve_symbol(thread, NULL, &dso, &ip);
-
-                        if (sym && call__match(sym)) {
-                                entry.parent = sym;
-                                break;
-                        }
+        if (!(self->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 -EINVAL;
                 }
-        }
-
-        while (*p != NULL) {
-                parent = *p;
-                he = rb_entry(parent, struct hist_entry, rb_node);
-
-                cmp = hist_entry__cmp(&entry, he);
-
-                if (!cmp) {
-                        he->count += count;
-                        return 0;
+                if (symbol_conf.use_callchain) {
+                        fprintf(stderr, "selected -g but no callchain data."
+                                        " Did you call perf record without"
+                                        " -g?\n");
+                        return -1;
                 }
-
-                if (cmp < 0)
-                        p = &(*p)->rb_left;
-                else
-                        p = &(*p)->rb_right;
+        } else if (!dont_use_callchains && 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");
+                                return -EINVAL;
+                        }
         }
 
-        he = malloc(sizeof(*he));
-        if (!he)
-                return -ENOMEM;
-        *he = entry;
-        rb_link_node(&he->rb_node, parent, p);
-        rb_insert_color(&he->rb_node, &hist);
-
         return 0;
 }
 
-static void hist_entry__free(struct hist_entry *he)
-{
-        free(he);
-}
-
-/*
- * collapse the histogram
- */
-
-static struct rb_root collapse_hists;
-
-static void collapse__insert_entry(struct hist_entry *he)
-{
-        struct rb_node **p = &collapse_hists.rb_node;
-        struct rb_node *parent = NULL;
-        struct hist_entry *iter;
-        int64_t cmp;
-
-        while (*p != NULL) {
-                parent = *p;
-                iter = rb_entry(parent, struct hist_entry, rb_node);
-
-                cmp = hist_entry__collapse(iter, he);
-
-                if (!cmp) {
-                        iter->count += he->count;
-                        hist_entry__free(he);
-                        return;
-                }
-
-                if (cmp < 0)
-                        p = &(*p)->rb_left;
-                else
-                        p = &(*p)->rb_right;
-        }
+static struct perf_event_ops event_ops = {
+        .sample = process_sample_event,
+        .mmap   = event__process_mmap,
+        .comm   = event__process_comm,
+        .exit   = event__process_task,
+        .fork   = event__process_task,
+        .lost   = event__process_lost,
+        .read   = process_read_event,
+        .attr   = event__process_attr,
+        .event_type = event__process_event_type,
+        .tracing_data = event__process_tracing_data,
+        .build_id = event__process_build_id,
+        .ordered_samples = true,
+        .ordering_requires_timestamps = true,
+};
 
-        rb_link_node(&he->rb_node, parent, p);
-        rb_insert_color(&he->rb_node, &collapse_hists);
-}
+extern volatile int session_done;
 
-static void collapse__resort(void)
+static void sig_handler(int sig __used)
 {
-        struct rb_node *next;
-        struct hist_entry *n;
-
-        if (!sort__need_collapse)
-                return;
-
-        next = rb_first(&hist);
-        while (next) {
-                n = rb_entry(next, struct hist_entry, rb_node);
-                next = rb_next(&n->rb_node);
-
-                rb_erase(&n->rb_node, &hist);
-                collapse__insert_entry(n);
-        }
+        session_done = 1;
 }
 
-/*
- * reverse the map, sort on count.
- */
-
-static struct rb_root output_hists;
-
-static void output__insert_entry(struct hist_entry *he)
+static size_t hists__fprintf_nr_sample_events(struct hists *self,
+                                              const char *evname, FILE *fp)
 {
-        struct rb_node **p = &output_hists.rb_node;
-        struct rb_node *parent = NULL;
-        struct hist_entry *iter;
-
-        while (*p != NULL) {
-                parent = *p;
-                iter = rb_entry(parent, struct hist_entry, rb_node);
-
-                if (he->count > iter->count)
-                        p = &(*p)->rb_left;
-                else
-                        p = &(*p)->rb_right;
-        }
+        size_t ret;
+        char unit;
+        unsigned long nr_events = self->stats.nr_events[PERF_RECORD_SAMPLE];
 
-        rb_link_node(&he->rb_node, parent, p);
-        rb_insert_color(&he->rb_node, &output_hists);
+        nr_events = convert_unit(nr_events, &unit);
+        ret = fprintf(fp, "# Events: %lu%c", nr_events, unit);
+        if (evname != NULL)
+                ret += fprintf(fp, " %s", evname);
+        return ret + fprintf(fp, "\n#\n");
 }
 
-static void output__resort(void)
+static int hists__tty_browse_tree(struct rb_root *tree, const char *help)
 {
-        struct rb_node *next;
-        struct hist_entry *n;
-        struct rb_root *tree = &hist;
-
-        if (sort__need_collapse)
-                tree = &collapse_hists;
-
-        next = rb_first(tree);
+        struct rb_node *next = rb_first(tree);
 
         while (next) {
-                n = rb_entry(next, struct hist_entry, rb_node);
-                next = rb_next(&n->rb_node);
+                struct hists *hists = rb_entry(next, struct hists, rb_node);
+                const char *evname = NULL;
 
-                rb_erase(&n->rb_node, tree);
-                output__insert_entry(n);
-        }
-}
+                if (rb_first(&hists->entries) != rb_last(&hists->entries))
+                        evname = __event_name(hists->type, hists->config);
 
-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;
-
-        fprintf(fp, "\n");
-        fprintf(fp, "#\n");
-        fprintf(fp, "# (%Ld samples)\n", (u64)total_samples);
-        fprintf(fp, "#\n");
-
-        fprintf(fp, "# Overhead");
-        list_for_each_entry(se, &hist_entry__sort_list, list) {
-                if (exclude_other && (se == &sort_parent))
-                        continue;
-                fprintf(fp, "  %s", se->header);
-        }
-        fprintf(fp, "\n");
-
-        fprintf(fp, "# ........");
-        list_for_each_entry(se, &hist_entry__sort_list, list) {
-                int i;
-
-                if (exclude_other && (se == &sort_parent))
-                        continue;
-
-                fprintf(fp, "  ");
-                for (i = 0; i < strlen(se->header); i++)
-                        fprintf(fp, ".");
-        }
-        fprintf(fp, "\n");
-
-        fprintf(fp, "#\n");
-
-        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);
+                hists__fprintf_nr_sample_events(hists, evname, stdout);
+                hists__fprintf(hists, NULL, false, stdout);
+                fprintf(stdout, "\n\n");
+                next = rb_next(&hists->rb_node);
         }
 
         if (sort_order == default_sort_order &&
-                        parent_pattern == default_parent_pattern) {
-                fprintf(fp, "#\n");
-                fprintf(fp, "# (For more details, try: perf report --sort comm,dso,symbol)\n");
-                fprintf(fp, "#\n");
-        }
-        fprintf(fp, "\n");
-
-        return ret;
-}
-
-static void register_idle_thread(void)
-{
-        struct thread *thread = threads__findnew(0);
-
-        if (thread == NULL ||
-                        thread__set_comm(thread, "[idle]")) {
-                fprintf(stderr, "problem inserting idle task.\n");
-                exit(-1);
-        }
-}
-
-static unsigned long total = 0,
-                     total_mmap = 0,
-                     total_comm = 0,
-                     total_fork = 0,
-                     total_unknown = 0,
-                     total_lost = 0;
-
-static int validate_chain(struct ip_callchain *chain, event_t *event)
-{
-        unsigned int chain_size;
-
-        chain_size = event->header.size;
-        chain_size -= (unsigned long)&event->ip.__more_data - (unsigned long)event;
-
-        if (chain->nr*sizeof(u64) > chain_size)
-                return -1;
-
-        return 0;
-}
-
-static int
-process_overflow_event(event_t *event, unsigned long offset, unsigned long head)
-{
-        char level;
-        int show = 0;
-        struct dso *dso = NULL;
-        struct thread *thread = threads__findnew(event->ip.pid);
-        u64 ip = event->ip.ip;
-        u64 period = 1;
-        struct map *map = NULL;
-        void *more_data = event->ip.__more_data;
-        struct ip_callchain *chain = NULL;
-
-        if (event->header.type & PERF_SAMPLE_PERIOD) {
-                period = *(u64 *)more_data;
-                more_data += sizeof(u64);
-        }
-
-        dprintf("%p [%p]: PERF_EVENT (IP, %d): %d: %p period: %Ld\n",
-                (void *)(offset + head),
-                (void *)(long)(event->header.size),
-                event->header.misc,
-                event->ip.pid,
-                (void *)(long)ip,
-                (long long)period);
-
-        if (event->header.type & PERF_SAMPLE_CALLCHAIN) {
-                int i;
-
-                chain = (void *)more_data;
-
-                dprintf("... chain: nr:%Lu\n", chain->nr);
-
-                if (validate_chain(chain, event) < 0) {
-                        eprintf("call-chain problem with event, skipping it.\n");
-                        return 0;
-                }
-
-                if (dump_trace) {
-                        for (i = 0; i < chain->nr; i++)
-                                dprintf("..... %2d: %016Lx\n", i, chain->ips[i]);
+            parent_pattern == default_parent_pattern) {
+                fprintf(stdout, "#\n# (%s)\n#\n", help);
+
+                if (show_threads) {
+                        bool style = !strcmp(pretty_printing_style, "raw");
+                        perf_read_values_display(stdout, &show_threads_values,
+                                                 style);
+                        perf_read_values_destroy(&show_threads_values);
                 }
         }
 
-        dprintf(" ... thread: %s:%d\n", thread->comm, thread->pid);
-
-        if (thread == NULL) {
-                eprintf("problem processing %d event, skipping it.\n",
-                        event->header.type);
-                return -1;
-        }
-
-        if (event->header.misc & PERF_EVENT_MISC_KERNEL) {
-                show = SHOW_KERNEL;
-                level = 'k';
-
-                dso = kernel_dso;
-
-                dprintf(" ...... dso: %s\n", dso->name);
-
-        } else if (event->header.misc & PERF_EVENT_MISC_USER) {
-
-                show = SHOW_USER;
-                level = '.';
-
-        } else {
-                show = SHOW_HV;
-                level = 'H';
-                dprintf(" ...... dso: [hypervisor]\n");
-        }
-
-        if (show & show_mask) {
-                struct symbol *sym = resolve_symbol(thread, &map, &dso, &ip);
-
-                if (hist_entry__add(thread, map, dso, sym, ip, chain, level, period)) {
-                        eprintf("problem incrementing symbol count, skipping event\n");
-                        return -1;
-                }
-        }
-        total += period;
-
         return 0;
 }
 
-static int
-process_mmap_event(event_t *event, unsigned long offset, unsigned long head)
-{
-        struct thread *thread = threads__findnew(event->mmap.pid);
-        struct map *map = map__new(&event->mmap);
-
-        dprintf("%p [%p]: PERF_EVENT_MMAP %d: [%p(%p) @ %p]: %s\n",
-                (void *)(offset + head),
-                (void *)(long)(event->header.size),
-                event->mmap.pid,
-                (void *)(long)event->mmap.start,
-                (void *)(long)event->mmap.len,
-                (void *)(long)event->mmap.pgoff,
-                event->mmap.filename);
-
-        if (thread == NULL || map == NULL) {
-                dprintf("problem processing PERF_EVENT_MMAP, skipping event.\n");
-                return 0;
-        }
-
-        thread__insert_map(thread, map);
-        total_mmap++;
-
-        return 0;
-}
-
-static int
-process_comm_event(event_t *event, unsigned long offset, unsigned long head)
+static int __cmd_report(void)
 {
-        struct thread *thread = threads__findnew(event->comm.pid);
+        int ret = -EINVAL;
+        struct perf_session *session;
+        struct rb_node *next;
+        const char *help = "For a higher level overview, try: perf report --sort comm,dso";
 
-        dprintf("%p [%p]: PERF_EVENT_COMM: %s:%d\n",
-                (void *)(offset + head),
-                (void *)(long)(event->header.size),
-                event->comm.comm, event->comm.pid);
+        signal(SIGINT, sig_handler);
 
-        if (thread == NULL ||
-            thread__set_comm(thread, event->comm.comm)) {
-                dprintf("problem processing PERF_EVENT_COMM, skipping event.\n");
-                return -1;
-        }
-        total_comm++;
+        session = perf_session__new(input_name, O_RDONLY, force, false, &event_ops);
+        if (session == NULL)
+                return -ENOMEM;
 
-        return 0;
-}
+        if (show_threads)
+                perf_read_values_init(&show_threads_values);
 
-static int
-process_fork_event(event_t *event, unsigned long offset, unsigned long head)
-{
-        struct thread *thread = threads__findnew(event->fork.pid);
-        struct thread *parent = threads__findnew(event->fork.ppid);
+        ret = perf_session__setup_sample_type(session);
+        if (ret)
+                goto out_delete;
 
-        dprintf("%p [%p]: PERF_EVENT_FORK: %d:%d\n",
-                (void *)(offset + head),
-                (void *)(long)(event->header.size),
-                event->fork.pid, event->fork.ppid);
+        ret = perf_session__process_events(session, &event_ops);
+        if (ret)
+                goto out_delete;
 
-        if (!thread || !parent || thread__fork(thread, parent)) {
-                dprintf("problem processing PERF_EVENT_FORK, skipping event.\n");
-                return -1;
+        if (dump_trace) {
+                perf_session__fprintf_nr_events(session, stdout);
+                goto out_delete;
         }
-        total_fork++;
-
-        return 0;
-}
-
-static int
-process_period_event(event_t *event, unsigned long offset, unsigned long head)
-{
-        dprintf("%p [%p]: PERF_EVENT_PERIOD: time:%Ld, id:%Ld: period:%Ld\n",
-                (void *)(offset + head),
-                (void *)(long)(event->header.size),
-                event->period.time,
-                event->period.id,
-                event->period.sample_period);
-
-        return 0;
-}
-
-static int
-process_lost_event(event_t *event, unsigned long offset, unsigned long head)
-{
-        dprintf("%p [%p]: PERF_EVENT_LOST: id:%Ld: lost:%Ld\n",
-                (void *)(offset + head),
-                (void *)(long)(event->header.size),
-                event->lost.id,
-                event->lost.lost);
 
-        total_lost += event->lost.lost;
+        if (verbose > 3)
+                perf_session__fprintf(session, stdout);
 
-        return 0;
-}
+        if (verbose > 2)
+                perf_session__fprintf_dsos(session, stdout);
 
-static void trace_event(event_t *event)
-{
-        unsigned char *raw_event = (void *)event;
-        char *color = PERF_COLOR_BLUE;
-        int i, j;
-
-        if (!dump_trace)
-                return;
+        next = rb_first(&session->hists_tree);
+        while (next) {
+                struct hists *hists;
 
-        dprintf(".");
-        cdprintf("\n. ... raw event: size %d bytes\n", event->header.size);
+                hists = rb_entry(next, struct hists, rb_node);
+                hists__collapse_resort(hists);
+                hists__output_resort(hists);
+                next = rb_next(&hists->rb_node);
+        }
 
-        for (i = 0; i < event->header.size; i++) {
-                if ((i & 15) == 0) {
-                        dprintf(".");
-                        cdprintf("  %04x: ", i);
-                }
+        if (use_browser > 0)
+                hists__tui_browse_tree(&session->hists_tree, help);
+        else
+                hists__tty_browse_tree(&session->hists_tree, help);
 
-                cdprintf(" %02x", raw_event[i]);
-
-                if (((i & 15) == 15) || i == event->header.size-1) {
-                        cdprintf("  ");
-                        for (j = 0; j < 15-(i & 15); j++)
-                                cdprintf("   ");
-                        for (j = 0; j < (i & 15); j++) {
-                                if (isprint(raw_event[i-15+j]))
-                                        cdprintf("%c", raw_event[i-15+j]);
-                                else
-                                        cdprintf(".");
-                        }
-                        cdprintf("\n");
-                }
-        }
-        dprintf(".\n");
+out_delete:
+        /*
+         * Speed up the exit process, for large files this can
+         * take quite a while.
+         *
+         * XXX Enable this when using valgrind or if we ever
+         * librarize this command.
+         *
+         * Also experiment with obstacks to see how much speed
+         * up we'll get here.
+         *
+         * perf_session__delete(session);
+         */
+        return ret;
 }
 
 static int
-process_event(event_t *event, unsigned long offset, unsigned long head)
+parse_callchain_opt(const struct option *opt __used, const char *arg,
+                    int unset)
 {
-        trace_event(event);
-
-        if (event->header.misc & PERF_EVENT_MISC_OVERFLOW)
-                return process_overflow_event(event, offset, head);
-
-        switch (event->header.type) {
-        case PERF_EVENT_MMAP:
-                return process_mmap_event(event, offset, head);
-
-        case PERF_EVENT_COMM:
-                return process_comm_event(event, offset, head);
-
-        case PERF_EVENT_FORK:
-                return process_fork_event(event, offset, head);
-
-        case PERF_EVENT_PERIOD:
-                return process_period_event(event, offset, head);
-
-        case PERF_EVENT_LOST:
-                return process_lost_event(event, offset, head);
+        char *tok, *tok2;
+        char *endptr;
 
         /*
-         * We dont process them right now but they are fine:
+         * --no-call-graph
          */
-
-        case PERF_EVENT_THROTTLE:
-        case PERF_EVENT_UNTHROTTLE:
+        if (unset) {
+                dont_use_callchains = true;
                 return 0;
-
-        default:
-                return -1;
-        }
-
-        return 0;
-}
-
-static struct perf_file_header          file_header;
-
-static int __cmd_report(void)
-{
-        int ret, rc = EXIT_FAILURE;
-        unsigned long offset = 0;
-        unsigned long head = sizeof(file_header);
-        struct stat stat;
-        event_t *event;
-        uint32_t size;
-        char *buf;
-
-        register_idle_thread();
-
-        input = open(input_name, O_RDONLY);
-        if (input < 0) {
-                fprintf(stderr, " failed to open file: %s", input_name);
-                if (!strcmp(input_name, "perf.data"))
-                        fprintf(stderr, "  (try 'perf record' first)");
-                fprintf(stderr, "\n");
-                exit(-1);
         }
 
-        ret = fstat(input, &stat);
-        if (ret < 0) {
-                perror("failed to stat file");
-                exit(-1);
-        }
+        symbol_conf.use_callchain = true;
 
-        if (!stat.st_size) {
-                fprintf(stderr, "zero-sized file, nothing to do!\n");
-                exit(0);
-        }
-
-        if (read(input, &file_header, sizeof(file_header)) == -1) {
-                perror("failed to read file headers");
-                exit(-1);
-        }
-
-        if (sort__has_parent &&
-            !(file_header.sample_type & PERF_SAMPLE_CALLCHAIN)) {
-                fprintf(stderr, "selected --sort parent, but no callchain data\n");
-                exit(-1);
-        }
-
-        if (load_kernel() < 0) {
-                perror("failed to load kernel symbols");
-                return EXIT_FAILURE;
-        }
+        if (!arg)
+                return 0;
 
-        if (!full_paths) {
-                if (getcwd(__cwd, sizeof(__cwd)) == NULL) {
-                        perror("failed to get the current directory");
-                        return EXIT_FAILURE;
-                }
-                cwdlen = strlen(cwd);
-        } else {
-                cwd = NULL;
-                cwdlen = 0;
-        }
-remap:
-        buf = (char *)mmap(NULL, page_size * mmap_window, PROT_READ,
-                           MAP_SHARED, input, offset);
-        if (buf == MAP_FAILED) {
-                perror("failed to mmap file");
-                exit(-1);
-        }
+        tok = strtok((char *)arg, ",");
+        if (!tok)
+                return -1;
 
-more:
-        event = (event_t *)(buf + head);
+        /* get the output mode */
+        if (!strncmp(tok, "graph", strlen(arg)))
+                callchain_param.mode = CHAIN_GRAPH_ABS;
 
-        size = event->header.size;
-        if (!size)
-                size = 8;
+        else if (!strncmp(tok, "flat", strlen(arg)))
+                callchain_param.mode = CHAIN_FLAT;
 
-        if (head + event->header.size >= page_size * mmap_window) {
-                unsigned long shift = page_size * (head / page_size);
-                int ret;
+        else if (!strncmp(tok, "fractal", strlen(arg)))
+                callchain_param.mode = CHAIN_GRAPH_REL;
 
-                ret = munmap(buf, page_size * mmap_window);
-                assert(ret == 0);
+        else if (!strncmp(tok, "none", strlen(arg))) {
+                callchain_param.mode = CHAIN_NONE;
+                symbol_conf.use_callchain = false;
 
-                offset += shift;
-                head -= shift;
-                goto remap;
+                return 0;
         }
 
-        size = event->header.size;
-
-        dprintf("\n%p [%p]: event: %d\n",
-                        (void *)(offset + head),
-                        (void *)(long)event->header.size,
-                        event->header.type);
-
-        if (!size || process_event(event, offset, head) < 0) {
-
-                dprintf("%p [%p]: skipping unknown header type: %d\n",
-                        (void *)(offset + head),
-                        (void *)(long)(event->header.size),
-                        event->header.type);
+        else
+                return -1;
 
-                total_unknown++;
+        /* get the min percentage */
+        tok = strtok(NULL, ",");
+        if (!tok)
+                goto setup;
 
-                /*
-                 * assume we lost track of the stream, check alignment, and
-                 * increment a single u64 in the hope to catch on again 'soon'.
-                 */
-
-                if (unlikely(head & 7))
-                        head &= ~7ULL;
+        tok2 = strtok(NULL, ",");
+        callchain_param.min_percent = strtod(tok, &endptr);
+        if (tok == endptr)
+                return -1;
 
-                size = 8;
+        if (tok2)
+                callchain_param.print_limit = strtod(tok2, &endptr);
+setup:
+        if (register_callchain_param(&callchain_param) < 0) {
+                fprintf(stderr, "Can't register callchain params\n");
+                return -1;
         }
-
-        head += size;
-
-        if (offset + head >= sizeof(file_header) + file_header.data_size)
-                goto done;
-
-        if (offset + head < stat.st_size)
-                goto more;
-
-done:
-        rc = EXIT_SUCCESS;
-        close(input);
-
-        dprintf("      IP events: %10ld\n", total);
-        dprintf("    mmap events: %10ld\n", total_mmap);
-        dprintf("    comm events: %10ld\n", total_comm);
-        dprintf("    fork events: %10ld\n", total_fork);
-        dprintf("    lost events: %10ld\n", total_lost);
-        dprintf(" unknown events: %10ld\n", total_unknown);
-
-        if (dump_trace)
-                return 0;
-
-        if (verbose >= 3)
-                threads__fprintf(stdout);
-
-        if (verbose >= 2)
-                dsos__fprintf(stdout);
-
-        collapse__resort();
-        output__resort();
-        output__fprintf(stdout, total);
-
-        return rc;
+        return 0;
 }
 
 static const char * const report_usage[] = {
@@ -1523,51 +439,103 @@ static const char * const report_usage[] = {
 static const struct option options[] = {
         OPT_STRING('i', "input", &input_name, "file",
                     "input file name"),
-        OPT_BOOLEAN('v', "verbose", &verbose,
+        OPT_INCR('v', "verbose", &verbose,
                     "be more verbose (show symbol address, etc)"),
         OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
                     "dump raw trace in ASCII"),
-        OPT_STRING('k', "vmlinux", &vmlinux, "file", "vmlinux pathname"),
+        OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
+                   "file", "vmlinux pathname"),
+        OPT_STRING(0, "kallsyms", &symbol_conf.kallsyms_name,
+                   "file", "kallsyms pathname"),
+        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", &symbol_conf.show_nr_samples,
+                    "Show a column with the number of samples"),
+        OPT_BOOLEAN('T', "threads", &show_threads,
+                    "Show per-thread event counters"),
+        OPT_STRING(0, "pretty", &pretty_printing_style, "key",
+                   "pretty printing style key: normal raw"),
+        OPT_BOOLEAN(0, "tui", &use_tui, "Use the TUI interface"),
+        OPT_BOOLEAN(0, "stdio", &use_stdio, "Use the stdio interface"),
         OPT_STRING('s', "sort", &sort_order, "key[,key2...]",
                    "sort by key(s): pid, comm, dso, symbol, parent"),
-        OPT_BOOLEAN('P', "full-paths", &full_paths,
-                    "Don't shorten the pathnames taking into account the cwd"),
+        OPT_BOOLEAN(0, "showcpuutilization", &symbol_conf.show_cpu_utilization,
+                    "Show sample percentage for different cpu modes"),
         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 (graph, flat, fractal, or none) and min percent threshold. "
+                     "Default: fractal,0.5", &parse_callchain_opt, callchain_default_opt),
+        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('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", &symbol_conf.field_sep, "separator",
+                   "separator for columns, no spaces will be added between "
+                   "columns '.' is reserved."),
+        OPT_BOOLEAN('U', "hide-unresolved", &hide_unresolved,
+                    "Only display entries resolved to a symbol"),
+        OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory",
+                    "Look for files with symbols relative to this directory"),
         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);
+        argc = parse_options(argc, argv, options, report_usage, 0);
+
+        if (use_stdio)
+                use_browser = 0;
+        else if (use_tui)
+                use_browser = 1;
 
-        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);
+        if (strcmp(input_name, "-") != 0)
+                setup_browser();
+        else
+                use_browser = 0;
+        /*
+         * Only in the newt browser we are doing integrated annotation,
+         * so don't allocate extra space that won't be used in the stdio
+         * implementation.
+         */
+        if (use_browser > 0) {
+                symbol_conf.priv_size = sizeof(struct sym_priv);
+                /*
+                 * For searching by name on the "Browse map details".
+                 * providing it only in verbose mode not to bloat too
+                 * much struct symbol.
+                 */
+                if (verbose) {
+                        /*
+                         * XXX: Need to provide a less kludgy way to ask for
+                         * more space per symbol, the u32 is for the index on
+                         * the ui browser.
+                         * See symbol__browser_index.
+                         */
+                        symbol_conf.priv_size += sizeof(u32);
+                        symbol_conf.sort_by_name = true;
                 }
         }
 
-        free(str);
-}
-
-int cmd_report(int argc, const char **argv, const char *prefix)
-{
-        symbol__init();
-
-        page_size = getpagesize();
-
-        argc = parse_options(argc, argv, options, report_usage, 0);
+        if (symbol__init() < 0)
+                return -1;
 
-        setup_sorting();
+        setup_sorting(report_usage, options);
 
-        if (parent_pattern != default_parent_pattern)
-                sort_dimension__add("parent");
-        else
-                exclude_other = 0;
+        if (parent_pattern != default_parent_pattern) {
+                if (sort_dimension__add("parent") < 0)
+                        return -1;
+                sort_parent.elide = 1;
+        } else
+                symbol_conf.exclude_other = false;
 
         /*
          * Any (unrecognized) arguments left?
@@ -1575,7 +543,9 @@ int cmd_report(int argc, const char **argv, const char *prefix)
         if (argc)
                 usage_with_options(report_usage, options);
 
-        setup_pager();
+        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
new file mode 100644
index 00000000000..abd4b8497bc
--- /dev/null
+++ b/tools/perf/builtin-sched.c
@@ -0,0 +1,1921 @@
+#include "builtin.h"
+#include "perf.h"
+
+#include "util/util.h"
+#include "util/cache.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/header.h"
+#include "util/session.h"
+
+#include "util/parse-options.h"
+#include "util/trace-event.h"
+
+#include "util/debug.h"
+
+#include <sys/prctl.h>
+
+#include <semaphore.h>
+#include <pthread.h>
+#include <math.h>
+
+static char                     const *input_name = "perf.data";
+
+static char                     default_sort_order[] = "avg, max, switch, runtime";
+static const char               *sort_order = default_sort_order;
+
+static int                      profile_cpu = -1;
+
+#define PR_SET_NAME             15               /* Set process name */
+#define MAX_CPUS                4096
+
+static u64                      run_measurement_overhead;
+static u64                      sleep_measurement_overhead;
+
+#define COMM_LEN                20
+#define SYM_LEN                 129
+
+#define MAX_PID                 65536
+
+static unsigned long            nr_tasks;
+
+struct sched_atom;
+
+struct task_desc {
+        unsigned long           nr;
+        unsigned long           pid;
+        char                    comm[COMM_LEN];
+
+        unsigned long           nr_events;
+        unsigned long           curr_event;
+        struct sched_atom       **atoms;
+
+        pthread_t               thread;
+        sem_t                   sleep_sem;
+
+        sem_t                   ready_for_work;
+        sem_t                   work_done_sem;
+
+        u64                     cpu_usage;
+};
+
+enum sched_event_type {
+        SCHED_EVENT_RUN,
+        SCHED_EVENT_SLEEP,
+        SCHED_EVENT_WAKEUP,
+        SCHED_EVENT_MIGRATION,
+};
+
+struct sched_atom {
+        enum sched_event_type   type;
+        int                     specific_wait;
+        u64                     timestamp;
+        u64                     duration;
+        unsigned long           nr;
+        sem_t                   *wait_sem;
+        struct task_desc        *wakee;
+};
+
+static struct task_desc         *pid_to_task[MAX_PID];
+
+static struct task_desc         **tasks;
+
+static pthread_mutex_t          start_work_mutex = PTHREAD_MUTEX_INITIALIZER;
+static u64                      start_time;
+
+static pthread_mutex_t          work_done_wait_mutex = PTHREAD_MUTEX_INITIALIZER;
+
+static unsigned long            nr_run_events;
+static unsigned long            nr_sleep_events;
+static unsigned long            nr_wakeup_events;
+
+static unsigned long            nr_sleep_corrections;
+static unsigned long            nr_run_events_optimized;
+
+static unsigned long            targetless_wakeups;
+static unsigned long            multitarget_wakeups;
+
+static u64                      cpu_usage;
+static u64                      runavg_cpu_usage;
+static u64                      parent_cpu_usage;
+static u64                      runavg_parent_cpu_usage;
+
+static unsigned long            nr_runs;
+static u64                      sum_runtime;
+static u64                      sum_fluct;
+static u64                      run_avg;
+
+static unsigned int             replay_repeat = 10;
+static unsigned long            nr_timestamps;
+static unsigned long            nr_unordered_timestamps;
+static unsigned long            nr_state_machine_bugs;
+static unsigned long            nr_context_switch_bugs;
+static unsigned long            nr_events;
+static unsigned long            nr_lost_chunks;
+static unsigned long            nr_lost_events;
+
+#define TASK_STATE_TO_CHAR_STR "RSDTtZX"
+
+enum thread_state {
+        THREAD_SLEEPING = 0,
+        THREAD_WAIT_CPU,
+        THREAD_SCHED_IN,
+        THREAD_IGNORE
+};
+
+struct work_atom {
+        struct list_head        list;
+        enum thread_state       state;
+        u64                     sched_out_time;
+        u64                     wake_up_time;
+        u64                     sched_in_time;
+        u64                     runtime;
+};
+
+struct work_atoms {
+        struct list_head        work_list;
+        struct thread           *thread;
+        struct rb_node          node;
+        u64                     max_lat;
+        u64                     max_lat_at;
+        u64                     total_lat;
+        u64                     nb_atoms;
+        u64                     total_runtime;
+};
+
+typedef int (*sort_fn_t)(struct work_atoms *, struct work_atoms *);
+
+static struct rb_root           atom_root, sorted_atom_root;
+
+static u64                      all_runtime;
+static u64                      all_count;
+
+
+static u64 get_nsecs(void)
+{
+        struct timespec ts;
+
+        clock_gettime(CLOCK_MONOTONIC, &ts);
+
+        return ts.tv_sec * 1000000000ULL + ts.tv_nsec;
+}
+
+static void burn_nsecs(u64 nsecs)
+{
+        u64 T0 = get_nsecs(), T1;
+
+        do {
+                T1 = get_nsecs();
+        } while (T1 + run_measurement_overhead < T0 + nsecs);
+}
+
+static void sleep_nsecs(u64 nsecs)
+{
+        struct timespec ts;
+
+        ts.tv_nsec = nsecs % 999999999;
+        ts.tv_sec = nsecs / 999999999;
+
+        nanosleep(&ts, NULL);
+}
+
+static void calibrate_run_measurement_overhead(void)
+{
+        u64 T0, T1, delta, min_delta = 1000000000ULL;
+        int i;
+
+        for (i = 0; i < 10; i++) {
+                T0 = get_nsecs();
+                burn_nsecs(0);
+                T1 = get_nsecs();
+                delta = T1-T0;
+                min_delta = min(min_delta, delta);
+        }
+        run_measurement_overhead = min_delta;
+
+        printf("run measurement overhead: %Ld nsecs\n", min_delta);
+}
+
+static void calibrate_sleep_measurement_overhead(void)
+{
+        u64 T0, T1, delta, min_delta = 1000000000ULL;
+        int i;
+
+        for (i = 0; i < 10; i++) {
+                T0 = get_nsecs();
+                sleep_nsecs(10000);
+                T1 = get_nsecs();
+                delta = T1-T0;
+                min_delta = min(min_delta, delta);
+        }
+        min_delta -= 10000;
+        sleep_measurement_overhead = min_delta;
+
+        printf("sleep measurement overhead: %Ld nsecs\n", min_delta);
+}
+
+static struct sched_atom *
+get_new_event(struct task_desc *task, u64 timestamp)
+{
+        struct sched_atom *event = zalloc(sizeof(*event));
+        unsigned long idx = task->nr_events;
+        size_t size;
+
+        event->timestamp = timestamp;
+        event->nr = idx;
+
+        task->nr_events++;
+        size = sizeof(struct sched_atom *) * task->nr_events;
+        task->atoms = realloc(task->atoms, size);
+        BUG_ON(!task->atoms);
+
+        task->atoms[idx] = event;
+
+        return event;
+}
+
+static struct sched_atom *last_event(struct task_desc *task)
+{
+        if (!task->nr_events)
+                return NULL;
+
+        return task->atoms[task->nr_events - 1];
+}
+
+static void
+add_sched_event_run(struct task_desc *task, u64 timestamp, u64 duration)
+{
+        struct sched_atom *event, *curr_event = last_event(task);
+
+        /*
+         * optimize an existing RUN event by merging this one
+         * to it:
+         */
+        if (curr_event && curr_event->type == SCHED_EVENT_RUN) {
+                nr_run_events_optimized++;
+                curr_event->duration += duration;
+                return;
+        }
+
+        event = get_new_event(task, timestamp);
+
+        event->type = SCHED_EVENT_RUN;
+        event->duration = duration;
+
+        nr_run_events++;
+}
+
+static void
+add_sched_event_wakeup(struct task_desc *task, u64 timestamp,
+                       struct task_desc *wakee)
+{
+        struct sched_atom *event, *wakee_event;
+
+        event = get_new_event(task, timestamp);
+        event->type = SCHED_EVENT_WAKEUP;
+        event->wakee = wakee;
+
+        wakee_event = last_event(wakee);
+        if (!wakee_event || wakee_event->type != SCHED_EVENT_SLEEP) {
+                targetless_wakeups++;
+                return;
+        }
+        if (wakee_event->wait_sem) {
+                multitarget_wakeups++;
+                return;
+        }
+
+        wakee_event->wait_sem = zalloc(sizeof(*wakee_event->wait_sem));
+        sem_init(wakee_event->wait_sem, 0, 0);
+        wakee_event->specific_wait = 1;
+        event->wait_sem = wakee_event->wait_sem;
+
+        nr_wakeup_events++;
+}
+
+static void
+add_sched_event_sleep(struct task_desc *task, u64 timestamp,
+                      u64 task_state __used)
+{
+        struct sched_atom *event = get_new_event(task, timestamp);
+
+        event->type = SCHED_EVENT_SLEEP;
+
+        nr_sleep_events++;
+}
+
+static struct task_desc *register_pid(unsigned long pid, const char *comm)
+{
+        struct task_desc *task;
+
+        BUG_ON(pid >= MAX_PID);
+
+        task = pid_to_task[pid];
+
+        if (task)
+                return task;
+
+        task = zalloc(sizeof(*task));
+        task->pid = pid;
+        task->nr = nr_tasks;
+        strcpy(task->comm, comm);
+        /*
+         * every task starts in sleeping state - this gets ignored
+         * if there's no wakeup pointing to this sleep state:
+         */
+        add_sched_event_sleep(task, 0, 0);
+
+        pid_to_task[pid] = task;
+        nr_tasks++;
+        tasks = realloc(tasks, nr_tasks*sizeof(struct task_task *));
+        BUG_ON(!tasks);
+        tasks[task->nr] = task;
+
+        if (verbose)
+                printf("registered task #%ld, PID %ld (%s)\n", nr_tasks, pid, comm);
+
+        return task;
+}
+
+
+static void print_task_traces(void)
+{
+        struct task_desc *task;
+        unsigned long i;
+
+        for (i = 0; i < nr_tasks; i++) {
+                task = tasks[i];
+                printf("task %6ld (%20s:%10ld), nr_events: %ld\n",
+                        task->nr, task->comm, task->pid, task->nr_events);
+        }
+}
+
+static void add_cross_task_wakeups(void)
+{
+        struct task_desc *task1, *task2;
+        unsigned long i, j;
+
+        for (i = 0; i < nr_tasks; i++) {
+                task1 = tasks[i];
+                j = i + 1;
+                if (j == nr_tasks)
+                        j = 0;
+                task2 = tasks[j];
+                add_sched_event_wakeup(task1, 0, task2);
+        }
+}
+
+static void
+process_sched_event(struct task_desc *this_task __used, struct sched_atom *atom)
+{
+        int ret = 0;
+        u64 now;
+        long long delta;
+
+        now = get_nsecs();
+        delta = start_time + atom->timestamp - now;
+
+        switch (atom->type) {
+                case SCHED_EVENT_RUN:
+                        burn_nsecs(atom->duration);
+                        break;
+                case SCHED_EVENT_SLEEP:
+                        if (atom->wait_sem)
+                                ret = sem_wait(atom->wait_sem);
+                        BUG_ON(ret);
+                        break;
+                case SCHED_EVENT_WAKEUP:
+                        if (atom->wait_sem)
+                                ret = sem_post(atom->wait_sem);
+                        BUG_ON(ret);
+                        break;
+                case SCHED_EVENT_MIGRATION:
+                        break;
+                default:
+                        BUG_ON(1);
+        }
+}
+
+static u64 get_cpu_usage_nsec_parent(void)
+{
+        struct rusage ru;
+        u64 sum;
+        int err;
+
+        err = getrusage(RUSAGE_SELF, &ru);
+        BUG_ON(err);
+
+        sum =  ru.ru_utime.tv_sec*1e9 + ru.ru_utime.tv_usec*1e3;
+        sum += ru.ru_stime.tv_sec*1e9 + ru.ru_stime.tv_usec*1e3;
+
+        return sum;
+}
+
+static int self_open_counters(void)
+{
+        struct perf_event_attr attr;
+        int fd;
+
+        memset(&attr, 0, sizeof(attr));
+
+        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;
+
+        ret = read(fd, &runtime, sizeof(runtime));
+        BUG_ON(ret != sizeof(runtime));
+
+        return runtime;
+}
+
+static void *thread_func(void *ctx)
+{
+        struct task_desc *this_task = ctx;
+        u64 cpu_usage_0, cpu_usage_1;
+        unsigned long i, ret;
+        char comm2[22];
+        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);
+        BUG_ON(ret);
+        ret = pthread_mutex_lock(&start_work_mutex);
+        BUG_ON(ret);
+        ret = pthread_mutex_unlock(&start_work_mutex);
+        BUG_ON(ret);
+
+        cpu_usage_0 = get_cpu_usage_nsec_self(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(fd);
+        this_task->cpu_usage = cpu_usage_1 - cpu_usage_0;
+        ret = sem_post(&this_task->work_done_sem);
+        BUG_ON(ret);
+
+        ret = pthread_mutex_lock(&work_done_wait_mutex);
+        BUG_ON(ret);
+        ret = pthread_mutex_unlock(&work_done_wait_mutex);
+        BUG_ON(ret);
+
+        goto again;
+}
+
+static void create_tasks(void)
+{
+        struct task_desc *task;
+        pthread_attr_t attr;
+        unsigned long i;
+        int err;
+
+        err = pthread_attr_init(&attr);
+        BUG_ON(err);
+        err = pthread_attr_setstacksize(&attr,
+                        (size_t) max(16 * 1024, PTHREAD_STACK_MIN));
+        BUG_ON(err);
+        err = pthread_mutex_lock(&start_work_mutex);
+        BUG_ON(err);
+        err = pthread_mutex_lock(&work_done_wait_mutex);
+        BUG_ON(err);
+        for (i = 0; i < nr_tasks; i++) {
+                task = tasks[i];
+                sem_init(&task->sleep_sem, 0, 0);
+                sem_init(&task->ready_for_work, 0, 0);
+                sem_init(&task->work_done_sem, 0, 0);
+                task->curr_event = 0;
+                err = pthread_create(&task->thread, &attr, thread_func, task);
+                BUG_ON(err);
+        }
+}
+
+static void wait_for_tasks(void)
+{
+        u64 cpu_usage_0, cpu_usage_1;
+        struct task_desc *task;
+        unsigned long i, ret;
+
+        start_time = get_nsecs();
+        cpu_usage = 0;
+        pthread_mutex_unlock(&work_done_wait_mutex);
+
+        for (i = 0; i < nr_tasks; i++) {
+                task = tasks[i];
+                ret = sem_wait(&task->ready_for_work);
+                BUG_ON(ret);
+                sem_init(&task->ready_for_work, 0, 0);
+        }
+        ret = pthread_mutex_lock(&work_done_wait_mutex);
+        BUG_ON(ret);
+
+        cpu_usage_0 = get_cpu_usage_nsec_parent();
+
+        pthread_mutex_unlock(&start_work_mutex);
+
+        for (i = 0; i < nr_tasks; i++) {
+                task = tasks[i];
+                ret = sem_wait(&task->work_done_sem);
+                BUG_ON(ret);
+                sem_init(&task->work_done_sem, 0, 0);
+                cpu_usage += task->cpu_usage;
+                task->cpu_usage = 0;
+        }
+
+        cpu_usage_1 = get_cpu_usage_nsec_parent();
+        if (!runavg_cpu_usage)
+                runavg_cpu_usage = cpu_usage;
+        runavg_cpu_usage = (runavg_cpu_usage*9 + cpu_usage)/10;
+
+        parent_cpu_usage = cpu_usage_1 - cpu_usage_0;
+        if (!runavg_parent_cpu_usage)
+                runavg_parent_cpu_usage = parent_cpu_usage;
+        runavg_parent_cpu_usage = (runavg_parent_cpu_usage*9 +
+                                   parent_cpu_usage)/10;
+
+        ret = pthread_mutex_lock(&start_work_mutex);
+        BUG_ON(ret);
+
+        for (i = 0; i < nr_tasks; i++) {
+                task = tasks[i];
+                sem_init(&task->sleep_sem, 0, 0);
+                task->curr_event = 0;
+        }
+}
+
+static void run_one_test(void)
+{
+        u64 T0, T1, delta, avg_delta, fluct, std_dev;
+
+        T0 = get_nsecs();
+        wait_for_tasks();
+        T1 = get_nsecs();
+
+        delta = T1 - T0;
+        sum_runtime += delta;
+        nr_runs++;
+
+        avg_delta = sum_runtime / nr_runs;
+        if (delta < avg_delta)
+                fluct = avg_delta - delta;
+        else
+                fluct = delta - avg_delta;
+        sum_fluct += fluct;
+        std_dev = sum_fluct / nr_runs / sqrt(nr_runs);
+        if (!run_avg)
+                run_avg = delta;
+        run_avg = (run_avg*9 + delta)/10;
+
+        printf("#%-3ld: %0.3f, ",
+                nr_runs, (double)delta/1000000.0);
+
+        printf("ravg: %0.2f, ",
+                (double)run_avg/1e6);
+
+        printf("cpu: %0.2f / %0.2f",
+                (double)cpu_usage/1e6, (double)runavg_cpu_usage/1e6);
+
+#if 0
+        /*
+         * rusage statistics done by the parent, these are less
+         * accurate than the sum_exec_runtime based statistics:
+         */
+        printf(" [%0.2f / %0.2f]",
+                (double)parent_cpu_usage/1e6,
+                (double)runavg_parent_cpu_usage/1e6);
+#endif
+
+        printf("\n");
+
+        if (nr_sleep_corrections)
+                printf(" (%ld sleep corrections)\n", nr_sleep_corrections);
+        nr_sleep_corrections = 0;
+}
+
+static void test_calibrations(void)
+{
+        u64 T0, T1;
+
+        T0 = get_nsecs();
+        burn_nsecs(1e6);
+        T1 = get_nsecs();
+
+        printf("the run test took %Ld nsecs\n", T1-T0);
+
+        T0 = get_nsecs();
+        sleep_nsecs(1e6);
+        T1 = get_nsecs();
+
+        printf("the sleep test took %Ld nsecs\n", T1-T0);
+}
+
+#define FILL_FIELD(ptr, field, event, data)     \
+        ptr.field = (typeof(ptr.field)) raw_field_value(event, #field, data)
+
+#define FILL_ARRAY(ptr, array, event, data)                     \
+do {                                                            \
+        void *__array = raw_field_ptr(event, #array, data);     \
+        memcpy(ptr.array, __array, sizeof(ptr.array));  \
+} while(0)
+
+#define FILL_COMMON_FIELDS(ptr, event, data)                    \
+do {                                                            \
+        FILL_FIELD(ptr, common_type, event, data);              \
+        FILL_FIELD(ptr, common_flags, event, data);             \
+        FILL_FIELD(ptr, common_preempt_count, event, data);     \
+        FILL_FIELD(ptr, common_pid, event, data);               \
+        FILL_FIELD(ptr, common_tgid, event, data);              \
+} while (0)
+
+
+
+struct trace_switch_event {
+        u32 size;
+
+        u16 common_type;
+        u8 common_flags;
+        u8 common_preempt_count;
+        u32 common_pid;
+        u32 common_tgid;
+
+        char prev_comm[16];
+        u32 prev_pid;
+        u32 prev_prio;
+        u64 prev_state;
+        char next_comm[16];
+        u32 next_pid;
+        u32 next_prio;
+};
+
+struct trace_runtime_event {
+        u32 size;
+
+        u16 common_type;
+        u8 common_flags;
+        u8 common_preempt_count;
+        u32 common_pid;
+        u32 common_tgid;
+
+        char comm[16];
+        u32 pid;
+        u64 runtime;
+        u64 vruntime;
+};
+
+struct trace_wakeup_event {
+        u32 size;
+
+        u16 common_type;
+        u8 common_flags;
+        u8 common_preempt_count;
+        u32 common_pid;
+        u32 common_tgid;
+
+        char comm[16];
+        u32 pid;
+
+        u32 prio;
+        u32 success;
+        u32 cpu;
+};
+
+struct trace_fork_event {
+        u32 size;
+
+        u16 common_type;
+        u8 common_flags;
+        u8 common_preempt_count;
+        u32 common_pid;
+        u32 common_tgid;
+
+        char parent_comm[16];
+        u32 parent_pid;
+        char child_comm[16];
+        u32 child_pid;
+};
+
+struct trace_migrate_task_event {
+        u32 size;
+
+        u16 common_type;
+        u8 common_flags;
+        u8 common_preempt_count;
+        u32 common_pid;
+        u32 common_tgid;
+
+        char comm[16];
+        u32 pid;
+
+        u32 prio;
+        u32 cpu;
+};
+
+struct trace_sched_handler {
+        void (*switch_event)(struct trace_switch_event *,
+                             struct 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,
+                             struct thread *thread);
+
+        void (*fork_event)(struct trace_fork_event *,
+                           struct event *,
+                           int cpu,
+                           u64 timestamp,
+                           struct thread *thread);
+
+        void (*migrate_task_event)(struct trace_migrate_task_event *,
+                           struct perf_session *session,
+                           struct event *,
+                           int cpu,
+                           u64 timestamp,
+                           struct thread *thread);
+};
+
+
+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,
+                    struct thread *thread __used)
+{
+        struct task_desc *waker, *wakee;
+
+        if (verbose) {
+                printf("sched_wakeup event %p\n", event);
+
+                printf(" ... pid %d woke up %s/%d\n",
+                        wakeup_event->common_pid,
+                        wakeup_event->comm,
+                        wakeup_event->pid);
+        }
+
+        waker = register_pid(wakeup_event->common_pid, "<unknown>");
+        wakee = register_pid(wakeup_event->pid, wakeup_event->comm);
+
+        add_sched_event_wakeup(waker, timestamp, wakee);
+}
+
+static u64 cpu_last_switched[MAX_CPUS];
+
+static void
+replay_switch_event(struct trace_switch_event *switch_event,
+                    struct perf_session *session __used,
+                    struct event *event,
+                    int cpu,
+                    u64 timestamp,
+                    struct thread *thread __used)
+{
+        struct task_desc *prev, *next;
+        u64 timestamp0;
+        s64 delta;
+
+        if (verbose)
+                printf("sched_switch event %p\n", event);
+
+        if (cpu >= MAX_CPUS || cpu < 0)
+                return;
+
+        timestamp0 = cpu_last_switched[cpu];
+        if (timestamp0)
+                delta = timestamp - timestamp0;
+        else
+                delta = 0;
+
+        if (delta < 0)
+                die("hm, delta: %Ld < 0 ?\n", delta);
+
+        if (verbose) {
+                printf(" ... switch from %s/%d to %s/%d [ran %Ld nsecs]\n",
+                        switch_event->prev_comm, switch_event->prev_pid,
+                        switch_event->next_comm, switch_event->next_pid,
+                        delta);
+        }
+
+        prev = register_pid(switch_event->prev_pid, switch_event->prev_comm);
+        next = register_pid(switch_event->next_pid, switch_event->next_comm);
+
+        cpu_last_switched[cpu] = timestamp;
+
+        add_sched_event_run(prev, timestamp, delta);
+        add_sched_event_sleep(prev, timestamp, switch_event->prev_state);
+}
+
+
+static void
+replay_fork_event(struct trace_fork_event *fork_event,
+                  struct event *event,
+                  int cpu __used,
+                  u64 timestamp __used,
+                  struct thread *thread __used)
+{
+        if (verbose) {
+                printf("sched_fork event %p\n", event);
+                printf("... parent: %s/%d\n", fork_event->parent_comm, fork_event->parent_pid);
+                printf("...  child: %s/%d\n", fork_event->child_comm, fork_event->child_pid);
+        }
+        register_pid(fork_event->parent_pid, fork_event->parent_comm);
+        register_pid(fork_event->child_pid, fork_event->child_comm);
+}
+
+static struct trace_sched_handler replay_ops  = {
+        .wakeup_event           = replay_wakeup_event,
+        .switch_event           = replay_switch_event,
+        .fork_event             = replay_fork_event,
+};
+
+struct sort_dimension {
+        const char              *name;
+        sort_fn_t               cmp;
+        struct list_head        list;
+};
+
+static LIST_HEAD(cmp_pid);
+
+static int
+thread_lat_cmp(struct list_head *list, struct work_atoms *l, struct work_atoms *r)
+{
+        struct sort_dimension *sort;
+        int ret = 0;
+
+        BUG_ON(list_empty(list));
+
+        list_for_each_entry(sort, list, list) {
+                ret = sort->cmp(l, r);
+                if (ret)
+                        return ret;
+        }
+
+        return ret;
+}
+
+static struct work_atoms *
+thread_atoms_search(struct rb_root *root, struct thread *thread,
+                         struct list_head *sort_list)
+{
+        struct rb_node *node = root->rb_node;
+        struct work_atoms key = { .thread = thread };
+
+        while (node) {
+                struct work_atoms *atoms;
+                int cmp;
+
+                atoms = container_of(node, struct work_atoms, node);
+
+                cmp = thread_lat_cmp(sort_list, &key, atoms);
+                if (cmp > 0)
+                        node = node->rb_left;
+                else if (cmp < 0)
+                        node = node->rb_right;
+                else {
+                        BUG_ON(thread != atoms->thread);
+                        return atoms;
+                }
+        }
+        return NULL;
+}
+
+static void
+__thread_latency_insert(struct rb_root *root, struct work_atoms *data,
+                         struct list_head *sort_list)
+{
+        struct rb_node **new = &(root->rb_node), *parent = NULL;
+
+        while (*new) {
+                struct work_atoms *this;
+                int cmp;
+
+                this = container_of(*new, struct work_atoms, node);
+                parent = *new;
+
+                cmp = thread_lat_cmp(sort_list, data, this);
+
+                if (cmp > 0)
+                        new = &((*new)->rb_left);
+                else
+                        new = &((*new)->rb_right);
+        }
+
+        rb_link_node(&data->node, parent, new);
+        rb_insert_color(&data->node, root);
+}
+
+static void thread_atoms_insert(struct thread *thread)
+{
+        struct work_atoms *atoms = zalloc(sizeof(*atoms));
+        if (!atoms)
+                die("No memory");
+
+        atoms->thread = thread;
+        INIT_LIST_HEAD(&atoms->work_list);
+        __thread_latency_insert(&atom_root, atoms, &cmp_pid);
+}
+
+static void
+latency_fork_event(struct trace_fork_event *fork_event __used,
+                   struct event *event __used,
+                   int cpu __used,
+                   u64 timestamp __used,
+                   struct thread *thread __used)
+{
+        /* should insert the newcomer */
+}
+
+__used
+static char sched_out_state(struct trace_switch_event *switch_event)
+{
+        const char *str = TASK_STATE_TO_CHAR_STR;
+
+        return str[switch_event->prev_state];
+}
+
+static void
+add_sched_out_event(struct work_atoms *atoms,
+                    char run_state,
+                    u64 timestamp)
+{
+        struct work_atom *atom = zalloc(sizeof(*atom));
+        if (!atom)
+                die("Non memory");
+
+        atom->sched_out_time = timestamp;
+
+        if (run_state == 'R') {
+                atom->state = THREAD_WAIT_CPU;
+                atom->wake_up_time = atom->sched_out_time;
+        }
+
+        list_add_tail(&atom->list, &atoms->work_list);
+}
+
+static void
+add_runtime_event(struct work_atoms *atoms, u64 delta, u64 timestamp __used)
+{
+        struct work_atom *atom;
+
+        BUG_ON(list_empty(&atoms->work_list));
+
+        atom = list_entry(atoms->work_list.prev, struct work_atom, list);
+
+        atom->runtime += delta;
+        atoms->total_runtime += delta;
+}
+
+static void
+add_sched_in_event(struct work_atoms *atoms, u64 timestamp)
+{
+        struct work_atom *atom;
+        u64 delta;
+
+        if (list_empty(&atoms->work_list))
+                return;
+
+        atom = list_entry(atoms->work_list.prev, struct work_atom, list);
+
+        if (atom->state != THREAD_WAIT_CPU)
+                return;
+
+        if (timestamp < atom->wake_up_time) {
+                atom->state = THREAD_IGNORE;
+                return;
+        }
+
+        atom->state = THREAD_SCHED_IN;
+        atom->sched_in_time = timestamp;
+
+        delta = atom->sched_in_time - atom->wake_up_time;
+        atoms->total_lat += delta;
+        if (delta > atoms->max_lat) {
+                atoms->max_lat = delta;
+                atoms->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,
+                     struct thread *thread __used)
+{
+        struct work_atoms *out_events, *in_events;
+        struct thread *sched_out, *sched_in;
+        u64 timestamp0;
+        s64 delta;
+
+        BUG_ON(cpu >= MAX_CPUS || cpu < 0);
+
+        timestamp0 = cpu_last_switched[cpu];
+        cpu_last_switched[cpu] = timestamp;
+        if (timestamp0)
+                delta = timestamp - timestamp0;
+        else
+                delta = 0;
+
+        if (delta < 0)
+                die("hm, delta: %Ld < 0 ?\n", delta);
+
+
+        sched_out = 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) {
+                thread_atoms_insert(sched_out);
+                out_events = thread_atoms_search(&atom_root, sched_out, &cmp_pid);
+                if (!out_events)
+                        die("out-event: Internal tree error");
+        }
+        add_sched_out_event(out_events, sched_out_state(switch_event), timestamp);
+
+        in_events = thread_atoms_search(&atom_root, sched_in, &cmp_pid);
+        if (!in_events) {
+                thread_atoms_insert(sched_in);
+                in_events = thread_atoms_search(&atom_root, sched_in, &cmp_pid);
+                if (!in_events)
+                        die("in-event: Internal tree error");
+                /*
+                 * Take came in we have not heard about yet,
+                 * add in an initial atom in runnable state:
+                 */
+                add_sched_out_event(in_events, 'R', timestamp);
+        }
+        add_sched_in_event(in_events, timestamp);
+}
+
+static void
+latency_runtime_event(struct trace_runtime_event *runtime_event,
+                     struct perf_session *session,
+                     struct event *event __used,
+                     int cpu,
+                     u64 timestamp,
+                     struct thread *this_thread __used)
+{
+        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);
+        if (!atoms) {
+                thread_atoms_insert(thread);
+                atoms = thread_atoms_search(&atom_root, thread, &cmp_pid);
+                if (!atoms)
+                        die("in-event: Internal tree error");
+                add_sched_out_event(atoms, 'R', timestamp);
+        }
+
+        add_runtime_event(atoms, runtime_event->runtime, timestamp);
+}
+
+static void
+latency_wakeup_event(struct trace_wakeup_event *wakeup_event,
+                     struct perf_session *session,
+                     struct event *__event __used,
+                     int cpu __used,
+                     u64 timestamp,
+                     struct thread *thread __used)
+{
+        struct work_atoms *atoms;
+        struct work_atom *atom;
+        struct thread *wakee;
+
+        /* Note for later, it may be interesting to observe the failing cases */
+        if (!wakeup_event->success)
+                return;
+
+        wakee = perf_session__findnew(session, wakeup_event->pid);
+        atoms = thread_atoms_search(&atom_root, wakee, &cmp_pid);
+        if (!atoms) {
+                thread_atoms_insert(wakee);
+                atoms = thread_atoms_search(&atom_root, wakee, &cmp_pid);
+                if (!atoms)
+                        die("wakeup-event: Internal tree error");
+                add_sched_out_event(atoms, 'S', timestamp);
+        }
+
+        BUG_ON(list_empty(&atoms->work_list));
+
+        atom = list_entry(atoms->work_list.prev, struct work_atom, list);
+
+        /*
+         * You WILL be missing events if you've recorded only
+         * one CPU, or are only looking at only one, so don't
+         * make useless noise.
+         */
+        if (profile_cpu == -1 && atom->state != THREAD_SLEEPING)
+                nr_state_machine_bugs++;
+
+        nr_timestamps++;
+        if (atom->sched_out_time > timestamp) {
+                nr_unordered_timestamps++;
+                return;
+        }
+
+        atom->state = THREAD_WAIT_CPU;
+        atom->wake_up_time = timestamp;
+}
+
+static void
+latency_migrate_task_event(struct trace_migrate_task_event *migrate_task_event,
+                     struct perf_session *session,
+                     struct event *__event __used,
+                     int cpu __used,
+                     u64 timestamp,
+                     struct thread *thread __used)
+{
+        struct work_atoms *atoms;
+        struct work_atom *atom;
+        struct thread *migrant;
+
+        /*
+         * Only need to worry about migration when profiling one CPU.
+         */
+        if (profile_cpu == -1)
+                return;
+
+        migrant = perf_session__findnew(session, migrate_task_event->pid);
+        atoms = thread_atoms_search(&atom_root, migrant, &cmp_pid);
+        if (!atoms) {
+                thread_atoms_insert(migrant);
+                register_pid(migrant->pid, migrant->comm);
+                atoms = thread_atoms_search(&atom_root, migrant, &cmp_pid);
+                if (!atoms)
+                        die("migration-event: Internal tree error");
+                add_sched_out_event(atoms, 'R', timestamp);
+        }
+
+        BUG_ON(list_empty(&atoms->work_list));
+
+        atom = list_entry(atoms->work_list.prev, struct work_atom, list);
+        atom->sched_in_time = atom->sched_out_time = atom->wake_up_time = timestamp;
+
+        nr_timestamps++;
+
+        if (atom->sched_out_time > timestamp)
+                nr_unordered_timestamps++;
+}
+
+static struct trace_sched_handler lat_ops  = {
+        .wakeup_event           = latency_wakeup_event,
+        .switch_event           = latency_switch_event,
+        .runtime_event          = latency_runtime_event,
+        .fork_event             = latency_fork_event,
+        .migrate_task_event     = latency_migrate_task_event,
+};
+
+static void output_lat_thread(struct work_atoms *work_list)
+{
+        int i;
+        int ret;
+        u64 avg;
+
+        if (!work_list->nb_atoms)
+                return;
+        /*
+         * Ignore idle threads:
+         */
+        if (!strcmp(work_list->thread->comm, "swapper"))
+                return;
+
+        all_runtime += work_list->total_runtime;
+        all_count += work_list->nb_atoms;
+
+        ret = printf("  %s:%d ", work_list->thread->comm, work_list->thread->pid);
+
+        for (i = 0; i < 24 - ret; i++)
+                printf(" ");
+
+        avg = work_list->total_lat / work_list->nb_atoms;
+
+        printf("|%11.3f ms |%9llu | avg:%9.3f ms | max:%9.3f ms | 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_at / 1e9);
+}
+
+static int pid_cmp(struct work_atoms *l, struct work_atoms *r)
+{
+        if (l->thread->pid < r->thread->pid)
+                return -1;
+        if (l->thread->pid > r->thread->pid)
+                return 1;
+
+        return 0;
+}
+
+static struct sort_dimension pid_sort_dimension = {
+        .name                   = "pid",
+        .cmp                    = pid_cmp,
+};
+
+static int avg_cmp(struct work_atoms *l, struct work_atoms *r)
+{
+        u64 avgl, avgr;
+
+        if (!l->nb_atoms)
+                return -1;
+
+        if (!r->nb_atoms)
+                return 1;
+
+        avgl = l->total_lat / l->nb_atoms;
+        avgr = r->total_lat / r->nb_atoms;
+
+        if (avgl < avgr)
+                return -1;
+        if (avgl > avgr)
+                return 1;
+
+        return 0;
+}
+
+static struct sort_dimension avg_sort_dimension = {
+        .name                   = "avg",
+        .cmp                    = avg_cmp,
+};
+
+static int max_cmp(struct work_atoms *l, struct work_atoms *r)
+{
+        if (l->max_lat < r->max_lat)
+                return -1;
+        if (l->max_lat > r->max_lat)
+                return 1;
+
+        return 0;
+}
+
+static struct sort_dimension max_sort_dimension = {
+        .name                   = "max",
+        .cmp                    = max_cmp,
+};
+
+static int switch_cmp(struct work_atoms *l, struct work_atoms *r)
+{
+        if (l->nb_atoms < r->nb_atoms)
+                return -1;
+        if (l->nb_atoms > r->nb_atoms)
+                return 1;
+
+        return 0;
+}
+
+static struct sort_dimension switch_sort_dimension = {
+        .name                   = "switch",
+        .cmp                    = switch_cmp,
+};
+
+static int runtime_cmp(struct work_atoms *l, struct work_atoms *r)
+{
+        if (l->total_runtime < r->total_runtime)
+                return -1;
+        if (l->total_runtime > r->total_runtime)
+                return 1;
+
+        return 0;
+}
+
+static struct sort_dimension runtime_sort_dimension = {
+        .name                   = "runtime",
+        .cmp                    = runtime_cmp,
+};
+
+static struct sort_dimension *available_sorts[] = {
+        &pid_sort_dimension,
+        &avg_sort_dimension,
+        &max_sort_dimension,
+        &switch_sort_dimension,
+        &runtime_sort_dimension,
+};
+
+#define NB_AVAILABLE_SORTS      (int)(sizeof(available_sorts) / sizeof(struct sort_dimension *))
+
+static LIST_HEAD(sort_list);
+
+static int sort_dimension__add(const char *tok, struct list_head *list)
+{
+        int i;
+
+        for (i = 0; i < NB_AVAILABLE_SORTS; i++) {
+                if (!strcmp(available_sorts[i]->name, tok)) {
+                        list_add_tail(&available_sorts[i]->list, list);
+
+                        return 0;
+                }
+        }
+
+        return -1;
+}
+
+static void setup_sorting(void);
+
+static void sort_lat(void)
+{
+        struct rb_node *node;
+
+        for (;;) {
+                struct work_atoms *data;
+                node = rb_first(&atom_root);
+                if (!node)
+                        break;
+
+                rb_erase(node, &atom_root);
+                data = rb_entry(node, struct work_atoms, node);
+                __thread_latency_insert(&sorted_atom_root, data, &sort_list);
+        }
+}
+
+static struct trace_sched_handler *trace_handler;
+
+static void
+process_sched_wakeup_event(void *data, struct perf_session *session,
+                           struct event *event,
+                           int cpu __used,
+                           u64 timestamp __used,
+                           struct thread *thread __used)
+{
+        struct trace_wakeup_event wakeup_event;
+
+        FILL_COMMON_FIELDS(wakeup_event, event, 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, session, event,
+                                            cpu, timestamp, thread);
+}
+
+/*
+ * Track the current task - that way we can know whether there's any
+ * weird events, such as a task being switched away that is not current.
+ */
+static int max_cpu;
+
+static u32 curr_pid[MAX_CPUS] = { [0 ... MAX_CPUS-1] = -1 };
+
+static struct thread *curr_thread[MAX_CPUS];
+
+static char next_shortname1 = 'A';
+static char next_shortname2 = '0';
+
+static void
+map_switch_event(struct trace_switch_event *switch_event,
+                 struct perf_session *session,
+                 struct event *event __used,
+                 int this_cpu,
+                 u64 timestamp,
+                 struct thread *thread __used)
+{
+        struct thread *sched_out, *sched_in;
+        int new_shortname;
+        u64 timestamp0;
+        s64 delta;
+        int cpu;
+
+        BUG_ON(this_cpu >= MAX_CPUS || this_cpu < 0);
+
+        if (this_cpu > max_cpu)
+                max_cpu = this_cpu;
+
+        timestamp0 = cpu_last_switched[this_cpu];
+        cpu_last_switched[this_cpu] = timestamp;
+        if (timestamp0)
+                delta = timestamp - timestamp0;
+        else
+                delta = 0;
+
+        if (delta < 0)
+                die("hm, delta: %Ld < 0 ?\n", delta);
+
+
+        sched_out = perf_session__findnew(session, switch_event->prev_pid);
+        sched_in = perf_session__findnew(session, switch_event->next_pid);
+
+        curr_thread[this_cpu] = sched_in;
+
+        printf("  ");
+
+        new_shortname = 0;
+        if (!sched_in->shortname[0]) {
+                sched_in->shortname[0] = next_shortname1;
+                sched_in->shortname[1] = next_shortname2;
+
+                if (next_shortname1 < 'Z') {
+                        next_shortname1++;
+                } else {
+                        next_shortname1='A';
+                        if (next_shortname2 < '9') {
+                                next_shortname2++;
+                        } else {
+                                next_shortname2='0';
+                        }
+                }
+                new_shortname = 1;
+        }
+
+        for (cpu = 0; cpu <= max_cpu; cpu++) {
+                if (cpu != this_cpu)
+                        printf(" ");
+                else
+                        printf("*");
+
+                if (curr_thread[cpu]) {
+                        if (curr_thread[cpu]->pid)
+                                printf("%2s ", curr_thread[cpu]->shortname);
+                        else
+                                printf(".  ");
+                } else
+                        printf("   ");
+        }
+
+        printf("  %12.6f secs ", (double)timestamp/1e9);
+        if (new_shortname) {
+                printf("%s => %s:%d\n",
+                        sched_in->shortname, sched_in->comm, sched_in->pid);
+        } else {
+                printf("\n");
+        }
+}
+
+
+static void
+process_sched_switch_event(void *data, struct perf_session *session,
+                           struct event *event,
+                           int this_cpu,
+                           u64 timestamp __used,
+                           struct thread *thread __used)
+{
+        struct trace_switch_event switch_event;
+
+        FILL_COMMON_FIELDS(switch_event, event, 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) {
+                /*
+                 * Are we trying to switch away a PID that is
+                 * not current?
+                 */
+                if (curr_pid[this_cpu] != switch_event.prev_pid)
+                        nr_context_switch_bugs++;
+        }
+        if (trace_handler->switch_event)
+                trace_handler->switch_event(&switch_event, session, event,
+                                            this_cpu, timestamp, thread);
+
+        curr_pid[this_cpu] = switch_event.next_pid;
+}
+
+static void
+process_sched_runtime_event(void *data, struct perf_session *session,
+                           struct event *event,
+                           int cpu __used,
+                           u64 timestamp __used,
+                           struct thread *thread __used)
+{
+        struct trace_runtime_event runtime_event;
+
+        FILL_ARRAY(runtime_event, comm, event, 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, session, event, cpu, timestamp, thread);
+}
+
+static void
+process_sched_fork_event(void *data,
+                         struct event *event,
+                         int cpu __used,
+                         u64 timestamp __used,
+                         struct thread *thread __used)
+{
+        struct trace_fork_event fork_event;
+
+        FILL_COMMON_FIELDS(fork_event, event, 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);
+}
+
+static void
+process_sched_exit_event(struct event *event,
+                         int cpu __used,
+                         u64 timestamp __used,
+                         struct thread *thread __used)
+{
+        if (verbose)
+                printf("sched_exit event %p\n", event);
+}
+
+static void
+process_sched_migrate_task_event(void *data, struct perf_session *session,
+                           struct event *event,
+                           int cpu __used,
+                           u64 timestamp __used,
+                           struct thread *thread __used)
+{
+        struct trace_migrate_task_event migrate_task_event;
+
+        FILL_COMMON_FIELDS(migrate_task_event, event, 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, session,
+                                                 event, cpu, timestamp, thread);
+}
+
+static void
+process_raw_event(event_t *raw_event __used, struct perf_session *session,
+                  void *data, int cpu, u64 timestamp, struct thread *thread)
+{
+        struct event *event;
+        int type;
+
+
+        type = trace_parse_common_type(data);
+        event = trace_find_event(type);
+
+        if (!strcmp(event->name, "sched_switch"))
+                process_sched_switch_event(data, session, event, cpu, timestamp, thread);
+        if (!strcmp(event->name, "sched_stat_runtime"))
+                process_sched_runtime_event(data, session, event, cpu, timestamp, thread);
+        if (!strcmp(event->name, "sched_wakeup"))
+                process_sched_wakeup_event(data, session, event, cpu, timestamp, thread);
+        if (!strcmp(event->name, "sched_wakeup_new"))
+                process_sched_wakeup_event(data, session, event, cpu, timestamp, thread);
+        if (!strcmp(event->name, "sched_process_fork"))
+                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(data, session, event, cpu, timestamp, thread);
+}
+
+static int process_sample_event(event_t *event, struct sample_data *sample,
+                                struct perf_session *session)
+{
+        struct thread *thread;
+
+        if (!(session->sample_type & PERF_SAMPLE_RAW))
+                return 0;
+
+        thread = perf_session__findnew(session, sample->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 (profile_cpu != -1 && profile_cpu != (int)sample->cpu)
+                return 0;
+
+        process_raw_event(event, session, sample->raw_data, sample->cpu,
+                          sample->time, thread);
+
+        return 0;
+}
+
+static struct perf_event_ops event_ops = {
+        .sample                 = process_sample_event,
+        .comm                   = event__process_comm,
+        .lost                   = event__process_lost,
+        .fork                   = event__process_task,
+        .ordered_samples        = true,
+};
+
+static int read_events(void)
+{
+        int err = -EINVAL;
+        struct perf_session *session = perf_session__new(input_name, O_RDONLY,
+                                                         0, false, &event_ops);
+        if (session == NULL)
+                return -ENOMEM;
+
+        if (perf_session__has_traces(session, "record -R")) {
+                err = perf_session__process_events(session, &event_ops);
+                nr_events      = session->hists.stats.nr_events[0];
+                nr_lost_events = session->hists.stats.total_lost;
+                nr_lost_chunks = session->hists.stats.nr_events[PERF_RECORD_LOST];
+        }
+
+        perf_session__delete(session);
+        return err;
+}
+
+static void print_bad_events(void)
+{
+        if (nr_unordered_timestamps && nr_timestamps) {
+                printf("  INFO: %.3f%% unordered timestamps (%ld out of %ld)\n",
+                        (double)nr_unordered_timestamps/(double)nr_timestamps*100.0,
+                        nr_unordered_timestamps, nr_timestamps);
+        }
+        if (nr_lost_events && nr_events) {
+                printf("  INFO: %.3f%% lost events (%ld out of %ld, in %ld chunks)\n",
+                        (double)nr_lost_events/(double)nr_events*100.0,
+                        nr_lost_events, nr_events, nr_lost_chunks);
+        }
+        if (nr_state_machine_bugs && nr_timestamps) {
+                printf("  INFO: %.3f%% state machine bugs (%ld out of %ld)",
+                        (double)nr_state_machine_bugs/(double)nr_timestamps*100.0,
+                        nr_state_machine_bugs, nr_timestamps);
+                if (nr_lost_events)
+                        printf(" (due to lost events?)");
+                printf("\n");
+        }
+        if (nr_context_switch_bugs && nr_timestamps) {
+                printf("  INFO: %.3f%% context switch bugs (%ld out of %ld)",
+                        (double)nr_context_switch_bugs/(double)nr_timestamps*100.0,
+                        nr_context_switch_bugs, nr_timestamps);
+                if (nr_lost_events)
+                        printf(" (due to lost events?)");
+                printf("\n");
+        }
+}
+
+static void __cmd_lat(void)
+{
+        struct rb_node *next;
+
+        setup_pager();
+        read_events();
+        sort_lat();
+
+        printf("\n ---------------------------------------------------------------------------------------------------------------\n");
+        printf("  Task                  |   Runtime ms  | Switches | Average delay ms | Maximum delay ms | Maximum delay at     |\n");
+        printf(" ---------------------------------------------------------------------------------------------------------------\n");
+
+        next = rb_first(&sorted_atom_root);
+
+        while (next) {
+                struct work_atoms *work_list;
+
+                work_list = rb_entry(next, struct work_atoms, node);
+                output_lat_thread(work_list);
+                next = rb_next(next);
+        }
+
+        printf(" -----------------------------------------------------------------------------------------\n");
+        printf("  TOTAL:                |%11.3f ms |%9Ld |\n",
+                (double)all_runtime/1e6, all_count);
+
+        printf(" ---------------------------------------------------\n");
+
+        print_bad_events();
+        printf("\n");
+
+}
+
+static struct trace_sched_handler map_ops  = {
+        .wakeup_event           = NULL,
+        .switch_event           = map_switch_event,
+        .runtime_event          = NULL,
+        .fork_event             = NULL,
+};
+
+static void __cmd_map(void)
+{
+        max_cpu = sysconf(_SC_NPROCESSORS_CONF);
+
+        setup_pager();
+        read_events();
+        print_bad_events();
+}
+
+static void __cmd_replay(void)
+{
+        unsigned long i;
+
+        calibrate_run_measurement_overhead();
+        calibrate_sleep_measurement_overhead();
+
+        test_calibrations();
+
+        read_events();
+
+        printf("nr_run_events:        %ld\n", nr_run_events);
+        printf("nr_sleep_events:      %ld\n", nr_sleep_events);
+        printf("nr_wakeup_events:     %ld\n", nr_wakeup_events);
+
+        if (targetless_wakeups)
+                printf("target-less wakeups:  %ld\n", targetless_wakeups);
+        if (multitarget_wakeups)
+                printf("multi-target wakeups: %ld\n", multitarget_wakeups);
+        if (nr_run_events_optimized)
+                printf("run atoms optimized: %ld\n",
+                        nr_run_events_optimized);
+
+        print_task_traces();
+        add_cross_task_wakeups();
+
+        create_tasks();
+        printf("------------------------------------------------------------\n");
+        for (i = 0; i < replay_repeat; i++)
+                run_one_test();
+}
+
+
+static const char * const sched_usage[] = {
+        "perf sched [<options>] {record|latency|map|replay|trace}",
+        NULL
+};
+
+static const struct option sched_options[] = {
+        OPT_STRING('i', "input", &input_name, "file",
+                    "input file name"),
+        OPT_INCR('v', "verbose", &verbose,
+                    "be more verbose (show symbol address, etc)"),
+        OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+                    "dump raw trace in ASCII"),
+        OPT_END()
+};
+
+static const char * const latency_usage[] = {
+        "perf sched latency [<options>]",
+        NULL
+};
+
+static const struct option latency_options[] = {
+        OPT_STRING('s', "sort", &sort_order, "key[,key2...]",
+                   "sort by key(s): runtime, switch, avg, max"),
+        OPT_INCR('v', "verbose", &verbose,
+                    "be more verbose (show symbol address, etc)"),
+        OPT_INTEGER('C', "CPU", &profile_cpu,
+                    "CPU to profile on"),
+        OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+                    "dump raw trace in ASCII"),
+        OPT_END()
+};
+
+static const char * const replay_usage[] = {
+        "perf sched replay [<options>]",
+        NULL
+};
+
+static const struct option replay_options[] = {
+        OPT_UINTEGER('r', "repeat", &replay_repeat,
+                     "repeat the workload replay N times (-1: infinite)"),
+        OPT_INCR('v', "verbose", &verbose,
+                    "be more verbose (show symbol address, etc)"),
+        OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+                    "dump raw trace in ASCII"),
+        OPT_END()
+};
+
+static void setup_sorting(void)
+{
+        char *tmp, *tok, *str = strdup(sort_order);
+
+        for (tok = strtok_r(str, ", ", &tmp);
+                        tok; tok = strtok_r(NULL, ", ", &tmp)) {
+                if (sort_dimension__add(tok, &sort_list) < 0) {
+                        error("Unknown --sort key: `%s'", tok);
+                        usage_with_options(latency_usage, latency_options);
+                }
+        }
+
+        free(str);
+
+        sort_dimension__add("pid", &cmp_pid);
+}
+
+static const char *record_args[] = {
+        "record",
+        "-a",
+        "-R",
+        "-f",
+        "-m", "1024",
+        "-c", "1",
+        "-e", "sched:sched_switch:r",
+        "-e", "sched:sched_stat_wait:r",
+        "-e", "sched:sched_stat_sleep:r",
+        "-e", "sched:sched_stat_iowait:r",
+        "-e", "sched:sched_stat_runtime:r",
+        "-e", "sched:sched_process_exit:r",
+        "-e", "sched:sched_process_fork:r",
+        "-e", "sched:sched_wakeup:r",
+        "-e", "sched:sched_migrate_task:r",
+};
+
+static int __cmd_record(int argc, const char **argv)
+{
+        unsigned int rec_argc, i, j;
+        const char **rec_argv;
+
+        rec_argc = ARRAY_SIZE(record_args) + argc - 1;
+        rec_argv = calloc(rec_argc + 1, sizeof(char *));
+
+        if (rec_argv == NULL)
+                return -ENOMEM;
+
+        for (i = 0; i < ARRAY_SIZE(record_args); i++)
+                rec_argv[i] = strdup(record_args[i]);
+
+        for (j = 1; j < (unsigned int)argc; j++, i++)
+                rec_argv[i] = argv[j];
+
+        BUG_ON(i != rec_argc);
+
+        return cmd_record(i, rec_argv, NULL);
+}
+
+int cmd_sched(int argc, const char **argv, const char *prefix __used)
+{
+        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 script' for now:
+         */
+        if (!strcmp(argv[0], "script"))
+                return cmd_script(argc, argv, prefix);
+
+        symbol__init();
+        if (!strncmp(argv[0], "rec", 3)) {
+                return __cmd_record(argc, argv);
+        } else if (!strncmp(argv[0], "lat", 3)) {
+                trace_handler = &lat_ops;
+                if (argc > 1) {
+                        argc = parse_options(argc, argv, latency_options, latency_usage, 0);
+                        if (argc)
+                                usage_with_options(latency_usage, latency_options);
+                }
+                setup_sorting();
+                __cmd_lat();
+        } else if (!strcmp(argv[0], "map")) {
+                trace_handler = &map_ops;
+                setup_sorting();
+                __cmd_map();
+        } else if (!strncmp(argv[0], "rep", 3)) {
+                trace_handler = &replay_ops;
+                if (argc) {
+                        argc = parse_options(argc, argv, replay_options, replay_usage, 0);
+                        if (argc)
+                                usage_with_options(replay_usage, replay_options);
+                }
+                __cmd_replay();
+        } else {
+                usage_with_options(sched_usage, sched_options);
+        }
+
+        return 0;
+}
diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
new file mode 100644
index 00000000000..150a606002e
--- /dev/null
+++ b/tools/perf/builtin-script.c
@@ -0,0 +1,821 @@
+#include "builtin.h"
+
+#include "perf.h"
+#include "util/cache.h"
+#include "util/debug.h"
+#include "util/exec_cmd.h"
+#include "util/header.h"
+#include "util/parse-options.h"
+#include "util/session.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/trace-event.h"
+#include "util/parse-options.h"
+#include "util/util.h"
+
+static char const               *script_name;
+static char const               *generate_script_lang;
+static bool                     debug_mode;
+static u64                      last_timestamp;
+static u64                      nr_unordered;
+extern const struct option      record_options[];
+
+static int default_start_script(const char *script __unused,
+                                int argc __unused,
+                                const char **argv __unused)
+{
+        return 0;
+}
+
+static int default_stop_script(void)
+{
+        return 0;
+}
+
+static int default_generate_script(const char *outfile __unused)
+{
+        return 0;
+}
+
+static struct scripting_ops default_scripting_ops = {
+        .start_script           = default_start_script,
+        .stop_script            = default_stop_script,
+        .process_event          = print_event,
+        .generate_script        = default_generate_script,
+};
+
+static struct scripting_ops     *scripting_ops;
+
+static void setup_scripting(void)
+{
+        setup_perl_scripting();
+        setup_python_scripting();
+
+        scripting_ops = &default_scripting_ops;
+}
+
+static int cleanup_scripting(void)
+{
+        pr_debug("\nperf script stopped\n");
+
+        return scripting_ops->stop_script();
+}
+
+static char const               *input_name = "perf.data";
+
+static int process_sample_event(event_t *event, struct sample_data *sample,
+                                struct perf_session *session)
+{
+        struct thread *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;
+        }
+
+        if (session->sample_type & PERF_SAMPLE_RAW) {
+                if (debug_mode) {
+                        if (sample->time < last_timestamp) {
+                                pr_err("Samples misordered, previous: %llu "
+                                        "this: %llu\n", last_timestamp,
+                                        sample->time);
+                                nr_unordered++;
+                        }
+                        last_timestamp = sample->time;
+                        return 0;
+                }
+                /*
+                 * 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(sample->cpu, sample->raw_data,
+                                             sample->raw_size,
+                                             sample->time, thread->comm);
+        }
+
+        session->hists.stats.total_period += sample->period;
+        return 0;
+}
+
+static struct perf_event_ops event_ops = {
+        .sample = process_sample_event,
+        .comm   = event__process_comm,
+        .attr   = event__process_attr,
+        .event_type = event__process_event_type,
+        .tracing_data = event__process_tracing_data,
+        .build_id = event__process_build_id,
+        .ordering_requires_timestamps = true,
+        .ordered_samples = true,
+};
+
+extern volatile int session_done;
+
+static void sig_handler(int sig __unused)
+{
+        session_done = 1;
+}
+
+static int __cmd_script(struct perf_session *session)
+{
+        int ret;
+
+        signal(SIGINT, sig_handler);
+
+        ret = perf_session__process_events(session, &event_ops);
+
+        if (debug_mode)
+                pr_err("Misordered timestamps: %llu\n", nr_unordered);
+
+        return ret;
+}
+
+struct script_spec {
+        struct list_head        node;
+        struct scripting_ops    *ops;
+        char                    spec[0];
+};
+
+static LIST_HEAD(script_specs);
+
+static struct script_spec *script_spec__new(const char *spec,
+                                            struct scripting_ops *ops)
+{
+        struct script_spec *s = malloc(sizeof(*s) + strlen(spec) + 1);
+
+        if (s != NULL) {
+                strcpy(s->spec, spec);
+                s->ops = ops;
+        }
+
+        return s;
+}
+
+static void script_spec__delete(struct script_spec *s)
+{
+        free(s->spec);
+        free(s);
+}
+
+static void script_spec__add(struct script_spec *s)
+{
+        list_add_tail(&s->node, &script_specs);
+}
+
+static struct script_spec *script_spec__find(const char *spec)
+{
+        struct script_spec *s;
+
+        list_for_each_entry(s, &script_specs, node)
+                if (strcasecmp(s->spec, spec) == 0)
+                        return s;
+        return NULL;
+}
+
+static struct script_spec *script_spec__findnew(const char *spec,
+                                                struct scripting_ops *ops)
+{
+        struct script_spec *s = script_spec__find(spec);
+
+        if (s)
+                return s;
+
+        s = script_spec__new(spec, ops);
+        if (!s)
+                goto out_delete_spec;
+
+        script_spec__add(s);
+
+        return s;
+
+out_delete_spec:
+        script_spec__delete(s);
+
+        return NULL;
+}
+
+int script_spec_register(const char *spec, struct scripting_ops *ops)
+{
+        struct script_spec *s;
+
+        s = script_spec__find(spec);
+        if (s)
+                return -1;
+
+        s = script_spec__findnew(spec, ops);
+        if (!s)
+                return -1;
+
+        return 0;
+}
+
+static struct scripting_ops *script_spec__lookup(const char *spec)
+{
+        struct script_spec *s = script_spec__find(spec);
+        if (!s)
+                return NULL;
+
+        return s->ops;
+}
+
+static void list_available_languages(void)
+{
+        struct script_spec *s;
+
+        fprintf(stderr, "\n");
+        fprintf(stderr, "Scripting language extensions (used in "
+                "perf script -s [spec:]script.[spec]):\n\n");
+
+        list_for_each_entry(s, &script_specs, node)
+                fprintf(stderr, "  %-42s [%s]\n", s->spec, s->ops->name);
+
+        fprintf(stderr, "\n");
+}
+
+static int parse_scriptname(const struct option *opt __used,
+                            const char *str, int unset __used)
+{
+        char spec[PATH_MAX];
+        const char *script, *ext;
+        int len;
+
+        if (strcmp(str, "lang") == 0) {
+                list_available_languages();
+                exit(0);
+        }
+
+        script = strchr(str, ':');
+        if (script) {
+                len = script - str;
+                if (len >= PATH_MAX) {
+                        fprintf(stderr, "invalid language specifier");
+                        return -1;
+                }
+                strncpy(spec, str, len);
+                spec[len] = '\0';
+                scripting_ops = script_spec__lookup(spec);
+                if (!scripting_ops) {
+                        fprintf(stderr, "invalid language specifier");
+                        return -1;
+                }
+                script++;
+        } else {
+                script = str;
+                ext = strrchr(script, '.');
+                if (!ext) {
+                        fprintf(stderr, "invalid script extension");
+                        return -1;
+                }
+                scripting_ops = script_spec__lookup(++ext);
+                if (!scripting_ops) {
+                        fprintf(stderr, "invalid script extension");
+                        return -1;
+                }
+        }
+
+        script_name = strdup(script);
+
+        return 0;
+}
+
+/* Helper function for filesystems that return a dent->d_type DT_UNKNOWN */
+static int is_directory(const char *base_path, const struct dirent *dent)
+{
+        char path[PATH_MAX];
+        struct stat st;
+
+        sprintf(path, "%s/%s", base_path, dent->d_name);
+        if (stat(path, &st))
+                return 0;
+
+        return S_ISDIR(st.st_mode);
+}
+
+#define for_each_lang(scripts_path, scripts_dir, lang_dirent, lang_next)\
+        while (!readdir_r(scripts_dir, &lang_dirent, &lang_next) &&     \
+               lang_next)                                               \
+                if ((lang_dirent.d_type == DT_DIR ||                    \
+                     (lang_dirent.d_type == DT_UNKNOWN &&               \
+                      is_directory(scripts_path, &lang_dirent))) &&     \
+                    (strcmp(lang_dirent.d_name, ".")) &&                \
+                    (strcmp(lang_dirent.d_name, "..")))
+
+#define for_each_script(lang_path, lang_dir, script_dirent, script_next)\
+        while (!readdir_r(lang_dir, &script_dirent, &script_next) &&    \
+               script_next)                                             \
+                if (script_dirent.d_type != DT_DIR &&                   \
+                    (script_dirent.d_type != DT_UNKNOWN ||              \
+                     !is_directory(lang_path, &script_dirent)))
+
+
+#define RECORD_SUFFIX                   "-record"
+#define REPORT_SUFFIX                   "-report"
+
+struct script_desc {
+        struct list_head        node;
+        char                    *name;
+        char                    *half_liner;
+        char                    *args;
+};
+
+static LIST_HEAD(script_descs);
+
+static struct script_desc *script_desc__new(const char *name)
+{
+        struct script_desc *s = zalloc(sizeof(*s));
+
+        if (s != NULL && name)
+                s->name = strdup(name);
+
+        return s;
+}
+
+static void script_desc__delete(struct script_desc *s)
+{
+        free(s->name);
+        free(s->half_liner);
+        free(s->args);
+        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 const char *ends_with(const char *str, const char *suffix)
+{
+        size_t suffix_len = strlen(suffix);
+        const 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_path, 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_path, lang_dir, script_dirent, script_next) {
+                        script_root = strdup(script_dirent.d_name);
+                        str = (char *)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_path, 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_path, lang_dir, script_dirent, script_next) {
+                        __script_root = strdup(script_dirent.d_name);
+                        str = (char *)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 bool is_top_script(const char *script_path)
+{
+        return ends_with(script_path, "top") == NULL ? false : true;
+}
+
+static int has_required_arg(char *script_path)
+{
+        struct script_desc *desc;
+        int n_args = 0;
+        char *p;
+
+        desc = script_desc__new(NULL);
+
+        if (read_script_info(desc, script_path))
+                goto out;
+
+        if (!desc->args)
+                goto out;
+
+        for (p = desc->args; *p; p++)
+                if (*p == '<')
+                        n_args++;
+out:
+        script_desc__delete(desc);
+
+        return n_args;
+}
+
+static const char * const script_usage[] = {
+        "perf script [<options>]",
+        "perf script [<options>] record <script> [<record-options>] <command>",
+        "perf script [<options>] report <script> [script-args]",
+        "perf script [<options>] <script> [<record-options>] <command>",
+        "perf script [<options>] <top-script> [script-args]",
+        NULL
+};
+
+static const struct option options[] = {
+        OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+                    "dump raw trace in ASCII"),
+        OPT_INCR('v', "verbose", &verbose,
+                    "be more verbose (show symbol address, etc)"),
+        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),
+        OPT_STRING('g', "gen-script", &generate_script_lang, "lang",
+                   "generate perf-script.xx script in specified language"),
+        OPT_STRING('i', "input", &input_name, "file",
+                    "input file name"),
+        OPT_BOOLEAN('d', "debug-mode", &debug_mode,
+                   "do various checks like samples ordering and lost events"),
+
+        OPT_END()
+};
+
+static bool have_cmd(int argc, const char **argv)
+{
+        char **__argv = malloc(sizeof(const char *) * argc);
+
+        if (!__argv)
+                die("malloc");
+        memcpy(__argv, argv, sizeof(const char *) * argc);
+        argc = parse_options(argc, (const char **)__argv, record_options,
+                             NULL, PARSE_OPT_STOP_AT_NON_OPTION);
+        free(__argv);
+
+        return argc != 0;
+}
+
+int cmd_script(int argc, const char **argv, const char *prefix __used)
+{
+        char *rec_script_path = NULL;
+        char *rep_script_path = NULL;
+        struct perf_session *session;
+        char *script_path = NULL;
+        const char **__argv;
+        bool system_wide;
+        int i, j, err;
+
+        setup_scripting();
+
+        argc = parse_options(argc, argv, options, script_usage,
+                             PARSE_OPT_STOP_AT_NON_OPTION);
+
+        if (argc > 1 && !strncmp(argv[0], "rec", strlen("rec"))) {
+                rec_script_path = get_script_path(argv[1], RECORD_SUFFIX);
+                if (!rec_script_path)
+                        return cmd_record(argc, argv, NULL);
+        }
+
+        if (argc > 1 && !strncmp(argv[0], "rep", strlen("rep"))) {
+                rep_script_path = get_script_path(argv[1], REPORT_SUFFIX);
+                if (!rep_script_path) {
+                        fprintf(stderr,
+                                "Please specify a valid report script"
+                                "(see 'perf script -l' for listing)\n");
+                        return -1;
+                }
+        }
+
+        /* make sure PERF_EXEC_PATH is set for scripts */
+        perf_set_argv_exec_path(perf_exec_path());
+
+        if (argc && !script_name && !rec_script_path && !rep_script_path) {
+                int live_pipe[2];
+                int rep_args;
+                pid_t pid;
+
+                rec_script_path = get_script_path(argv[0], RECORD_SUFFIX);
+                rep_script_path = get_script_path(argv[0], REPORT_SUFFIX);
+
+                if (!rec_script_path && !rep_script_path) {
+                        fprintf(stderr, " Couldn't find script %s\n\n See perf"
+                                " script -l for available scripts.\n", argv[0]);
+                        usage_with_options(script_usage, options);
+                }
+
+                if (is_top_script(argv[0])) {
+                        rep_args = argc - 1;
+                } else {
+                        int rec_args;
+
+                        rep_args = has_required_arg(rep_script_path);
+                        rec_args = (argc - 1) - rep_args;
+                        if (rec_args < 0) {
+                                fprintf(stderr, " %s script requires options."
+                                        "\n\n See perf script -l for available "
+                                        "scripts and options.\n", argv[0]);
+                                usage_with_options(script_usage, options);
+                        }
+                }
+
+                if (pipe(live_pipe) < 0) {
+                        perror("failed to create pipe");
+                        exit(-1);
+                }
+
+                pid = fork();
+                if (pid < 0) {
+                        perror("failed to fork");
+                        exit(-1);
+                }
+
+                if (!pid) {
+                        system_wide = true;
+                        j = 0;
+
+                        dup2(live_pipe[1], 1);
+                        close(live_pipe[0]);
+
+                        if (!is_top_script(argv[0]))
+                                system_wide = !have_cmd(argc - rep_args,
+                                                        &argv[rep_args]);
+
+                        __argv = malloc((argc + 6) * sizeof(const char *));
+                        if (!__argv)
+                                die("malloc");
+
+                        __argv[j++] = "/bin/sh";
+                        __argv[j++] = rec_script_path;
+                        if (system_wide)
+                                __argv[j++] = "-a";
+                        __argv[j++] = "-q";
+                        __argv[j++] = "-o";
+                        __argv[j++] = "-";
+                        for (i = rep_args + 1; i < argc; i++)
+                                __argv[j++] = argv[i];
+                        __argv[j++] = NULL;
+
+                        execvp("/bin/sh", (char **)__argv);
+                        free(__argv);
+                        exit(-1);
+                }
+
+                dup2(live_pipe[0], 0);
+                close(live_pipe[1]);
+
+                __argv = malloc((argc + 4) * sizeof(const char *));
+                if (!__argv)
+                        die("malloc");
+                j = 0;
+                __argv[j++] = "/bin/sh";
+                __argv[j++] = rep_script_path;
+                for (i = 1; i < rep_args + 1; i++)
+                        __argv[j++] = argv[i];
+                __argv[j++] = "-i";
+                __argv[j++] = "-";
+                __argv[j++] = NULL;
+
+                execvp("/bin/sh", (char **)__argv);
+                free(__argv);
+                exit(-1);
+        }
+
+        if (rec_script_path)
+                script_path = rec_script_path;
+        if (rep_script_path)
+                script_path = rep_script_path;
+
+        if (script_path) {
+                system_wide = false;
+                j = 0;
+
+                if (rec_script_path)
+                        system_wide = !have_cmd(argc - 1, &argv[1]);
+
+                __argv = malloc((argc + 2) * sizeof(const char *));
+                if (!__argv)
+                        die("malloc");
+                __argv[j++] = "/bin/sh";
+                __argv[j++] = script_path;
+                if (system_wide)
+                        __argv[j++] = "-a";
+                for (i = 2; i < argc; i++)
+                        __argv[j++] = argv[i];
+                __argv[j++] = NULL;
+
+                execvp("/bin/sh", (char **)__argv);
+                free(__argv);
+                exit(-1);
+        }
+
+        if (symbol__init() < 0)
+                return -1;
+        if (!script_name)
+                setup_pager();
+
+        session = perf_session__new(input_name, O_RDONLY, 0, false, &event_ops);
+        if (session == NULL)
+                return -ENOMEM;
+
+        if (strcmp(input_name, "-") &&
+            !perf_session__has_traces(session, "record -R"))
+                return -EINVAL;
+
+        if (generate_script_lang) {
+                struct stat perf_stat;
+
+                int input = open(input_name, O_RDONLY);
+                if (input < 0) {
+                        perror("failed to open file");
+                        exit(-1);
+                }
+
+                err = fstat(input, &perf_stat);
+                if (err < 0) {
+                        perror("failed to stat file");
+                        exit(-1);
+                }
+
+                if (!perf_stat.st_size) {
+                        fprintf(stderr, "zero-sized file, nothing to do!\n");
+                        exit(0);
+                }
+
+                scripting_ops = script_spec__lookup(generate_script_lang);
+                if (!scripting_ops) {
+                        fprintf(stderr, "invalid language specifier");
+                        return -1;
+                }
+
+                err = scripting_ops->generate_script("perf-script");
+                goto out;
+        }
+
+        if (script_name) {
+                err = scripting_ops->start_script(script_name, argc, argv);
+                if (err)
+                        goto out;
+                pr_debug("perf script started with script %s\n\n", script_name);
+        }
+
+        err = __cmd_script(session);
+
+        perf_session__delete(session);
+        cleanup_scripting();
+out:
+        return err;
+}
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 6d3eeac1ea2..c385a63ebfd 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -32,6 +32,7 @@
  *   Wu Fengguang <fengguang.wu@intel.com>
  *   Mike Galbraith <efault@gmx.de>
  *   Paul Mackerras <paulus@samba.org>
+ *   Jaswinder Singh Rajput <jaswinder@kernel.org>
  *
  * Released under the GPL v2. (and only v2, not any later version)
  */
@@ -41,113 +42,148 @@
 #include "util/util.h"
 #include "util/parse-options.h"
 #include "util/parse-events.h"
+#include "util/event.h"
+#include "util/evsel.h"
+#include "util/debug.h"
+#include "util/header.h"
+#include "util/cpumap.h"
+#include "util/thread.h"
 
 #include <sys/prctl.h>
 #include <math.h>
+#include <locale.h>
 
-static struct perf_counter_attr default_attrs[MAX_COUNTERS] = {
+#define DEFAULT_SEPARATOR       " "
 
-  { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_TASK_CLOCK      },
-  { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CONTEXT_SWITCHES},
-  { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CPU_MIGRATIONS  },
-  { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_PAGE_FAULTS     },
+static struct perf_event_attr default_attrs[] = {
 
-  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CPU_CYCLES      },
-  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_INSTRUCTIONS    },
-  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_REFERENCES},
-  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_MISSES    },
+  { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_TASK_CLOCK              },
+  { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CONTEXT_SWITCHES        },
+  { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CPU_MIGRATIONS          },
+  { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_PAGE_FAULTS             },
 
-};
-
-static int                      system_wide                     =  0;
-static int                      inherit                         =  1;
-static int                      verbose                         =  0;
-
-static int                      fd[MAX_NR_CPUS][MAX_COUNTERS];
+  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CPU_CYCLES              },
+  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_INSTRUCTIONS            },
+  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_INSTRUCTIONS     },
+  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_MISSES           },
+  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_REFERENCES        },
+  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_MISSES            },
 
-static int                      target_pid                      = -1;
-static int                      nr_cpus                         =  0;
-static unsigned int             page_size;
-
-static int                      scale                           =  1;
-
-static const unsigned int default_count[] = {
-        1000000,
-        1000000,
-          10000,
-          10000,
-        1000000,
-          10000,
 };
 
-#define MAX_RUN 100
-
-static int                      run_count               =  1;
-static int                      run_idx                 =  0;
+static bool                     system_wide                     =  false;
+static struct cpu_map           *cpus;
+static int                      run_idx                         =  0;
+
+static int                      run_count                       =  1;
+static bool                     no_inherit                      = false;
+static bool                     scale                           =  true;
+static bool                     no_aggr                         = false;
+static pid_t                    target_pid                      = -1;
+static pid_t                    target_tid                      = -1;
+static struct thread_map        *threads;
+static pid_t                    child_pid                       = -1;
+static bool                     null_run                        =  false;
+static bool                     big_num                         =  true;
+static int                      big_num_opt                     =  -1;
+static const char               *cpu_list;
+static const char               *csv_sep                        = NULL;
+static bool                     csv_output                      = false;
+
+static volatile int done = 0;
+
+struct stats
+{
+        double n, mean, M2;
+};
 
-static u64                      event_res[MAX_RUN][MAX_COUNTERS][3];
-static u64                      event_scaled[MAX_RUN][MAX_COUNTERS];
+struct perf_stat {
+        struct stats      res_stats[3];
+};
 
-//static u64                    event_hist[MAX_RUN][MAX_COUNTERS][3];
+static int perf_evsel__alloc_stat_priv(struct perf_evsel *evsel)
+{
+        evsel->priv = zalloc(sizeof(struct perf_stat));
+        return evsel->priv == NULL ? -ENOMEM : 0;
+}
 
+static void perf_evsel__free_stat_priv(struct perf_evsel *evsel)
+{
+        free(evsel->priv);
+        evsel->priv = NULL;
+}
 
-static u64                      runtime_nsecs[MAX_RUN];
-static u64                      walltime_nsecs[MAX_RUN];
-static u64                      runtime_cycles[MAX_RUN];
+static void update_stats(struct stats *stats, u64 val)
+{
+        double delta;
 
-static u64                      event_res_avg[MAX_COUNTERS][3];
-static u64                      event_res_noise[MAX_COUNTERS][3];
+        stats->n++;
+        delta = val - stats->mean;
+        stats->mean += delta / stats->n;
+        stats->M2 += delta*(val - stats->mean);
+}
 
-static u64                      event_scaled_avg[MAX_COUNTERS];
+static double avg_stats(struct stats *stats)
+{
+        return stats->mean;
+}
 
-static u64                      runtime_nsecs_avg;
-static u64                      runtime_nsecs_noise;
+/*
+ * http://en.wikipedia.org/wiki/Algorithms_for_calculating_variance
+ *
+ *       (\Sum n_i^2) - ((\Sum n_i)^2)/n
+ * s^2 = -------------------------------
+ *                  n - 1
+ *
+ * http://en.wikipedia.org/wiki/Stddev
+ *
+ * The std dev of the mean is related to the std dev by:
+ *
+ *             s
+ * s_mean = -------
+ *          sqrt(n)
+ *
+ */
+static double stddev_stats(struct stats *stats)
+{
+        double variance = stats->M2 / (stats->n - 1);
+        double variance_mean = variance / stats->n;
 
-static u64                      walltime_nsecs_avg;
-static u64                      walltime_nsecs_noise;
+        return sqrt(variance_mean);
+}
 
-static u64                      runtime_cycles_avg;
-static u64                      runtime_cycles_noise;
+struct stats                    runtime_nsecs_stats[MAX_NR_CPUS];
+struct stats                    runtime_cycles_stats[MAX_NR_CPUS];
+struct stats                    runtime_branches_stats[MAX_NR_CPUS];
+struct stats                    walltime_nsecs_stats;
 
-static void create_perf_stat_counter(int counter)
+static int create_perf_stat_counter(struct perf_evsel *evsel)
 {
-        struct perf_counter_attr *attr = attrs + counter;
+        struct perf_event_attr *attr = &evsel->attr;
 
         if (scale)
                 attr->read_format = PERF_FORMAT_TOTAL_TIME_ENABLED |
                                     PERF_FORMAT_TOTAL_TIME_RUNNING;
 
-        if (system_wide) {
-                int cpu;
-                for (cpu = 0; cpu < nr_cpus; cpu ++) {
-                        fd[cpu][counter] = sys_perf_counter_open(attr, -1, cpu, -1, 0);
-                        if (fd[cpu][counter] < 0 && verbose) {
-                                printf("Error: counter %d, sys_perf_counter_open() syscall returned with %d (%s)\n", counter, fd[cpu][counter], strerror(errno));
-                        }
-                }
-        } else {
-                attr->inherit   = inherit;
-                attr->disabled  = 1;
+        if (system_wide)
+                return perf_evsel__open_per_cpu(evsel, cpus);
 
-                fd[0][counter] = sys_perf_counter_open(attr, 0, -1, -1, 0);
-                if (fd[0][counter] < 0 && verbose) {
-                        printf("Error: counter %d, sys_perf_counter_open() syscall returned with %d (%s)\n", counter, fd[0][counter], strerror(errno));
-                }
+        attr->inherit = !no_inherit;
+        if (target_pid == -1 && target_tid == -1) {
+                attr->disabled = 1;
+                attr->enable_on_exec = 1;
         }
+
+        return perf_evsel__open_per_thread(evsel, threads);
 }
 
 /*
  * Does the counter have nsecs as a unit?
  */
-static inline int nsec_counter(int counter)
+static inline int nsec_counter(struct perf_evsel *evsel)
 {
-        if (attrs[counter].type != PERF_TYPE_SOFTWARE)
-                return 0;
-
-        if (attrs[counter].config == PERF_COUNT_SW_CPU_CLOCK)
-                return 1;
-
-        if (attrs[counter].config == PERF_COUNT_SW_TASK_CLOCK)
+        if (perf_evsel__match(evsel, SOFTWARE, SW_CPU_CLOCK) ||
+            perf_evsel__match(evsel, SOFTWARE, SW_TASK_CLOCK))
                 return 1;
 
         return 0;
@@ -155,296 +191,405 @@ static inline int nsec_counter(int counter)
 
 /*
  * Read out the results of a single counter:
+ * aggregate counts across CPUs in system-wide mode
  */
-static void read_counter(int counter)
+static int read_counter_aggr(struct perf_evsel *counter)
 {
-        u64 *count, single_count[3];
-        ssize_t res;
-        int cpu, nv;
-        int scaled;
-
-        count = event_res[run_idx][counter];
+        struct perf_stat *ps = counter->priv;
+        u64 *count = counter->counts->aggr.values;
+        int i;
 
-        count[0] = count[1] = count[2] = 0;
+        if (__perf_evsel__read(counter, cpus->nr, threads->nr, scale) < 0)
+                return -1;
 
-        nv = scale ? 3 : 1;
-        for (cpu = 0; cpu < nr_cpus; cpu ++) {
-                if (fd[cpu][counter] < 0)
-                        continue;
+        for (i = 0; i < 3; i++)
+                update_stats(&ps->res_stats[i], count[i]);
 
-                res = read(fd[cpu][counter], single_count, nv * sizeof(u64));
-                assert(res == nv * sizeof(u64));
-                close(fd[cpu][counter]);
-                fd[cpu][counter] = -1;
-
-                count[0] += single_count[0];
-                if (scale) {
-                        count[1] += single_count[1];
-                        count[2] += single_count[2];
-                }
+        if (verbose) {
+                fprintf(stderr, "%s: %Ld %Ld %Ld\n", event_name(counter),
+                                count[0], count[1], count[2]);
         }
 
-        scaled = 0;
-        if (scale) {
-                if (count[2] == 0) {
-                        event_scaled[run_idx][counter] = -1;
-                        count[0] = 0;
-                        return;
-                }
-
-                if (count[2] < count[1]) {
-                        event_scaled[run_idx][counter] = 1;
-                        count[0] = (unsigned long long)
-                                ((double)count[0] * count[1] / count[2] + 0.5);
-                }
-        }
         /*
          * Save the full runtime - to allow normalization during printout:
          */
-        if (attrs[counter].type == PERF_TYPE_SOFTWARE &&
-                attrs[counter].config == PERF_COUNT_SW_TASK_CLOCK)
-                runtime_nsecs[run_idx] = count[0];
-        if (attrs[counter].type == PERF_TYPE_HARDWARE &&
-                attrs[counter].config == PERF_COUNT_HW_CPU_CYCLES)
-                runtime_cycles[run_idx] = count[0];
+        if (perf_evsel__match(counter, SOFTWARE, SW_TASK_CLOCK))
+                update_stats(&runtime_nsecs_stats[0], count[0]);
+        if (perf_evsel__match(counter, HARDWARE, HW_CPU_CYCLES))
+                update_stats(&runtime_cycles_stats[0], count[0]);
+        if (perf_evsel__match(counter, HARDWARE, HW_BRANCH_INSTRUCTIONS))
+                update_stats(&runtime_branches_stats[0], count[0]);
+
+        return 0;
 }
 
-static int run_perf_stat(int argc, const char **argv)
+/*
+ * Read out the results of a single counter:
+ * do not aggregate counts across CPUs in system-wide mode
+ */
+static int read_counter(struct perf_evsel *counter)
+{
+        u64 *count;
+        int cpu;
+
+        for (cpu = 0; cpu < cpus->nr; cpu++) {
+                if (__perf_evsel__read_on_cpu(counter, cpu, 0, scale) < 0)
+                        return -1;
+
+                count = counter->counts->cpu[cpu].values;
+
+                if (perf_evsel__match(counter, SOFTWARE, SW_TASK_CLOCK))
+                        update_stats(&runtime_nsecs_stats[cpu], count[0]);
+                if (perf_evsel__match(counter, HARDWARE, HW_CPU_CYCLES))
+                        update_stats(&runtime_cycles_stats[cpu], count[0]);
+                if (perf_evsel__match(counter, HARDWARE, HW_BRANCH_INSTRUCTIONS))
+                        update_stats(&runtime_branches_stats[cpu], count[0]);
+        }
+
+        return 0;
+}
+
+static int run_perf_stat(int argc __used, const char **argv)
 {
         unsigned long long t0, t1;
+        struct perf_evsel *counter;
         int status = 0;
-        int counter;
-        int pid;
+        int child_ready_pipe[2], go_pipe[2];
+        const bool forks = (argc > 0);
+        char buf;
 
-        if (!system_wide)
-                nr_cpus = 1;
+        if (forks && (pipe(child_ready_pipe) < 0 || pipe(go_pipe) < 0)) {
+                perror("failed to create pipes");
+                exit(1);
+        }
 
-        for (counter = 0; counter < nr_counters; counter++)
-                create_perf_stat_counter(counter);
+        if (forks) {
+                if ((child_pid = fork()) < 0)
+                        perror("failed to fork");
 
-        /*
-         * Enable counters and exec the command:
-         */
-        t0 = rdclock();
-        prctl(PR_TASK_PERF_COUNTERS_ENABLE);
+                if (!child_pid) {
+                        close(child_ready_pipe[0]);
+                        close(go_pipe[1]);
+                        fcntl(go_pipe[0], F_SETFD, FD_CLOEXEC);
+
+                        /*
+                         * Do a dummy execvp to get the PLT entry resolved,
+                         * so we avoid the resolver overhead on the real
+                         * execvp call.
+                         */
+                        execvp("", (char **)argv);
+
+                        /*
+                         * Tell the parent we're ready to go
+                         */
+                        close(child_ready_pipe[1]);
+
+                        /*
+                         * Wait until the parent tells us to go.
+                         */
+                        if (read(go_pipe[0], &buf, 1) == -1)
+                                perror("unable to read pipe");
 
-        if ((pid = fork()) < 0)
-                perror("failed to fork");
+                        execvp(argv[0], (char **)argv);
 
-        if (!pid) {
-                if (execvp(argv[0], (char **)argv)) {
                         perror(argv[0]);
                         exit(-1);
                 }
+
+                if (target_tid == -1 && target_pid == -1 && !system_wide)
+                        threads->map[0] = child_pid;
+
+                /*
+                 * Wait for the child to be ready to exec.
+                 */
+                close(child_ready_pipe[1]);
+                close(go_pipe[0]);
+                if (read(child_ready_pipe[0], &buf, 1) == -1)
+                        perror("unable to read pipe");
+                close(child_ready_pipe[0]);
         }
 
-        wait(&status);
+        list_for_each_entry(counter, &evsel_list, node) {
+                if (create_perf_stat_counter(counter) < 0) {
+                        if (errno == -EPERM || errno == -EACCES) {
+                                error("You may not have permission to collect %sstats.\n"
+                                      "\t Consider tweaking"
+                                      " /proc/sys/kernel/perf_event_paranoid or running as root.",
+                                      system_wide ? "system-wide " : "");
+                        } else if (errno == ENOENT) {
+                                error("%s event is not supported. ", event_name(counter));
+                        } else {
+                                error("open_counter returned with %d (%s). "
+                                      "/bin/dmesg may provide additional information.\n",
+                                       errno, strerror(errno));
+                        }
+                        if (child_pid != -1)
+                                kill(child_pid, SIGTERM);
+                        die("Not all events could be opened.\n");
+                        return -1;
+                }
+        }
+
+        /*
+         * Enable counters and exec the command:
+         */
+        t0 = rdclock();
+
+        if (forks) {
+                close(go_pipe[1]);
+                wait(&status);
+        } else {
+                while(!done) sleep(1);
+        }
 
-        prctl(PR_TASK_PERF_COUNTERS_DISABLE);
         t1 = rdclock();
 
-        walltime_nsecs[run_idx] = t1 - t0;
+        update_stats(&walltime_nsecs_stats, t1 - t0);
 
-        for (counter = 0; counter < nr_counters; counter++)
-                read_counter(counter);
+        if (no_aggr) {
+                list_for_each_entry(counter, &evsel_list, node) {
+                        read_counter(counter);
+                        perf_evsel__close_fd(counter, cpus->nr, 1);
+                }
+        } else {
+                list_for_each_entry(counter, &evsel_list, node) {
+                        read_counter_aggr(counter);
+                        perf_evsel__close_fd(counter, cpus->nr, threads->nr);
+                }
+        }
 
         return WEXITSTATUS(status);
 }
 
-static void print_noise(u64 *count, u64 *noise)
+static void print_noise(struct perf_evsel *evsel, double avg)
 {
-        if (run_count > 1)
-                fprintf(stderr, "   ( +- %7.3f%% )",
-                        (double)noise[0]/(count[0]+1)*100.0);
+        struct perf_stat *ps;
+
+        if (run_count == 1)
+                return;
+
+        ps = evsel->priv;
+        fprintf(stderr, "   ( +- %7.3f%% )",
+                        100 * stddev_stats(&ps->res_stats[0]) / avg);
 }
 
-static void nsec_printout(int counter, u64 *count, u64 *noise)
+static void nsec_printout(int cpu, struct perf_evsel *evsel, double avg)
 {
-        double msecs = (double)count[0] / 1000000;
+        double msecs = avg / 1e6;
+        char cpustr[16] = { '\0', };
+        const char *fmt = csv_output ? "%s%.6f%s%s" : "%s%18.6f%s%-24s";
 
-        fprintf(stderr, " %14.6f  %-20s", msecs, event_name(counter));
+        if (no_aggr)
+                sprintf(cpustr, "CPU%*d%s",
+                        csv_output ? 0 : -4,
+                        cpus->map[cpu], csv_sep);
 
-        if (attrs[counter].type == PERF_TYPE_SOFTWARE &&
-                attrs[counter].config == PERF_COUNT_SW_TASK_CLOCK) {
+        fprintf(stderr, fmt, cpustr, msecs, csv_sep, event_name(evsel));
 
-                if (walltime_nsecs_avg)
-                        fprintf(stderr, " # %10.3f CPUs ",
-                                (double)count[0] / (double)walltime_nsecs_avg);
-        }
-        print_noise(count, noise);
+        if (csv_output)
+                return;
+
+        if (perf_evsel__match(evsel, SOFTWARE, SW_TASK_CLOCK))
+                fprintf(stderr, " # %10.3f CPUs ",
+                                avg / avg_stats(&walltime_nsecs_stats));
 }
 
-static void abs_printout(int counter, u64 *count, u64 *noise)
+static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
 {
-        fprintf(stderr, " %14Ld  %-20s", count[0], event_name(counter));
+        double total, ratio = 0.0;
+        char cpustr[16] = { '\0', };
+        const char *fmt;
+
+        if (csv_output)
+                fmt = "%s%.0f%s%s";
+        else if (big_num)
+                fmt = "%s%'18.0f%s%-24s";
+        else
+                fmt = "%s%18.0f%s%-24s";
+
+        if (no_aggr)
+                sprintf(cpustr, "CPU%*d%s",
+                        csv_output ? 0 : -4,
+                        cpus->map[cpu], csv_sep);
+        else
+                cpu = 0;
 
-        if (runtime_cycles_avg &&
-                attrs[counter].type == PERF_TYPE_HARDWARE &&
-                        attrs[counter].config == PERF_COUNT_HW_INSTRUCTIONS) {
+        fprintf(stderr, fmt, cpustr, avg, csv_sep, event_name(evsel));
 
-                fprintf(stderr, " # %10.3f IPC  ",
-                        (double)count[0] / (double)runtime_cycles_avg);
-        } else {
-                if (runtime_nsecs_avg) {
-                        fprintf(stderr, " # %10.3f M/sec",
-                                (double)count[0]/runtime_nsecs_avg*1000.0);
-                }
+        if (csv_output)
+                return;
+
+        if (perf_evsel__match(evsel, HARDWARE, HW_INSTRUCTIONS)) {
+                total = avg_stats(&runtime_cycles_stats[cpu]);
+
+                if (total)
+                        ratio = avg / total;
+
+                fprintf(stderr, " # %10.3f IPC  ", ratio);
+        } else if (perf_evsel__match(evsel, HARDWARE, HW_BRANCH_MISSES) &&
+                        runtime_branches_stats[cpu].n != 0) {
+                total = avg_stats(&runtime_branches_stats[cpu]);
+
+                if (total)
+                        ratio = avg * 100 / total;
+
+                fprintf(stderr, " # %10.3f %%    ", ratio);
+
+        } else if (runtime_nsecs_stats[cpu].n != 0) {
+                total = avg_stats(&runtime_nsecs_stats[cpu]);
+
+                if (total)
+                        ratio = 1000.0 * avg / total;
+
+                fprintf(stderr, " # %10.3f M/sec", ratio);
         }
-        print_noise(count, noise);
 }
 
 /*
  * Print out the results of a single counter:
+ * aggregated counts in system-wide mode
  */
-static void print_counter(int counter)
+static void print_counter_aggr(struct perf_evsel *counter)
 {
-        u64 *count, *noise;
-        int scaled;
-
-        count = event_res_avg[counter];
-        noise = event_res_noise[counter];
-        scaled = event_scaled_avg[counter];
+        struct perf_stat *ps = counter->priv;
+        double avg = avg_stats(&ps->res_stats[0]);
+        int scaled = counter->counts->scaled;
 
         if (scaled == -1) {
-                fprintf(stderr, " %14s  %-20s\n",
-                        "<not counted>", event_name(counter));
+                fprintf(stderr, "%*s%s%-24s\n",
+                        csv_output ? 0 : 18,
+                        "<not counted>", csv_sep, event_name(counter));
                 return;
         }
 
         if (nsec_counter(counter))
-                nsec_printout(counter, count, noise);
+                nsec_printout(-1, counter, avg);
         else
-                abs_printout(counter, count, noise);
+                abs_printout(-1, counter, avg);
 
-        if (scaled)
-                fprintf(stderr, "  (scaled from %.2f%%)",
-                        (double) count[2] / count[1] * 100);
-
-        fprintf(stderr, "\n");
-}
+        if (csv_output) {
+                fputc('\n', stderr);
+                return;
+        }
 
-/*
- * normalize_noise noise values down to stddev:
- */
-static void normalize_noise(u64 *val)
-{
-        double res;
+        print_noise(counter, avg);
 
-        res = (double)*val / (run_count * sqrt((double)run_count));
+        if (scaled) {
+                double avg_enabled, avg_running;
 
-        *val = (u64)res;
-}
+                avg_enabled = avg_stats(&ps->res_stats[1]);
+                avg_running = avg_stats(&ps->res_stats[2]);
 
-static void update_avg(const char *name, int idx, u64 *avg, u64 *val)
-{
-        *avg += *val;
+                fprintf(stderr, "  (scaled from %.2f%%)",
+                                100 * avg_running / avg_enabled);
+        }
 
-        if (verbose > 1)
-                fprintf(stderr, "debug: %20s[%d]: %Ld\n", name, idx, *val);
+        fprintf(stderr, "\n");
 }
+
 /*
- * Calculate the averages and noises:
+ * Print out the results of a single counter:
+ * does not use aggregated count in system-wide
  */
-static void calc_avg(void)
+static void print_counter(struct perf_evsel *counter)
 {
-        int i, j;
-
-        if (verbose > 1)
-                fprintf(stderr, "\n");
-
-        for (i = 0; i < run_count; i++) {
-                update_avg("runtime", 0, &runtime_nsecs_avg, runtime_nsecs + i);
-                update_avg("walltime", 0, &walltime_nsecs_avg, walltime_nsecs + i);
-                update_avg("runtime_cycles", 0, &runtime_cycles_avg, runtime_cycles + i);
-
-                for (j = 0; j < nr_counters; j++) {
-                        update_avg("counter/0", j,
-                                event_res_avg[j]+0, event_res[i][j]+0);
-                        update_avg("counter/1", j,
-                                event_res_avg[j]+1, event_res[i][j]+1);
-                        update_avg("counter/2", j,
-                                event_res_avg[j]+2, event_res[i][j]+2);
-                        update_avg("scaled", j,
-                                event_scaled_avg + j, event_scaled[i]+j);
+        u64 ena, run, val;
+        int cpu;
+
+        for (cpu = 0; cpu < cpus->nr; cpu++) {
+                val = counter->counts->cpu[cpu].val;
+                ena = counter->counts->cpu[cpu].ena;
+                run = counter->counts->cpu[cpu].run;
+                if (run == 0 || ena == 0) {
+                        fprintf(stderr, "CPU%*d%s%*s%s%-24s",
+                                csv_output ? 0 : -4,
+                                cpus->map[cpu], csv_sep,
+                                csv_output ? 0 : 18,
+                                "<not counted>", csv_sep,
+                                event_name(counter));
+
+                        fprintf(stderr, "\n");
+                        continue;
                 }
-        }
-        runtime_nsecs_avg /= run_count;
-        walltime_nsecs_avg /= run_count;
-        runtime_cycles_avg /= run_count;
-
-        for (j = 0; j < nr_counters; j++) {
-                event_res_avg[j][0] /= run_count;
-                event_res_avg[j][1] /= run_count;
-                event_res_avg[j][2] /= run_count;
-        }
 
-        for (i = 0; i < run_count; i++) {
-                runtime_nsecs_noise +=
-                        abs((s64)(runtime_nsecs[i] - runtime_nsecs_avg));
-                walltime_nsecs_noise +=
-                        abs((s64)(walltime_nsecs[i] - walltime_nsecs_avg));
-                runtime_cycles_noise +=
-                        abs((s64)(runtime_cycles[i] - runtime_cycles_avg));
-
-                for (j = 0; j < nr_counters; j++) {
-                        event_res_noise[j][0] +=
-                                abs((s64)(event_res[i][j][0] - event_res_avg[j][0]));
-                        event_res_noise[j][1] +=
-                                abs((s64)(event_res[i][j][1] - event_res_avg[j][1]));
-                        event_res_noise[j][2] +=
-                                abs((s64)(event_res[i][j][2] - event_res_avg[j][2]));
-                }
-        }
+                if (nsec_counter(counter))
+                        nsec_printout(cpu, counter, val);
+                else
+                        abs_printout(cpu, counter, val);
 
-        normalize_noise(&runtime_nsecs_noise);
-        normalize_noise(&walltime_nsecs_noise);
-        normalize_noise(&runtime_cycles_noise);
+                if (!csv_output) {
+                        print_noise(counter, 1.0);
 
-        for (j = 0; j < nr_counters; j++) {
-                normalize_noise(&event_res_noise[j][0]);
-                normalize_noise(&event_res_noise[j][1]);
-                normalize_noise(&event_res_noise[j][2]);
+                        if (run != ena) {
+                                fprintf(stderr, "  (scaled from %.2f%%)",
+                                        100.0 * run / ena);
+                        }
+                }
+                fprintf(stderr, "\n");
         }
 }
 
 static void print_stat(int argc, const char **argv)
 {
-        int i, counter;
-
-        calc_avg();
+        struct perf_evsel *counter;
+        int i;
 
         fflush(stdout);
 
-        fprintf(stderr, "\n");
-        fprintf(stderr, " Performance counter stats for \'%s", argv[0]);
-
-        for (i = 1; i < argc; i++)
-                fprintf(stderr, " %s", argv[i]);
-
-        fprintf(stderr, "\'");
-        if (run_count > 1)
-                fprintf(stderr, " (%d runs)", run_count);
-        fprintf(stderr, ":\n\n");
-
-        for (counter = 0; counter < nr_counters; counter++)
-                print_counter(counter);
+        if (!csv_output) {
+                fprintf(stderr, "\n");
+                fprintf(stderr, " Performance counter stats for ");
+                if(target_pid == -1 && target_tid == -1) {
+                        fprintf(stderr, "\'%s", argv[0]);
+                        for (i = 1; i < argc; i++)
+                                fprintf(stderr, " %s", argv[i]);
+                } else if (target_pid != -1)
+                        fprintf(stderr, "process id \'%d", target_pid);
+                else
+                        fprintf(stderr, "thread id \'%d", target_tid);
+
+                fprintf(stderr, "\'");
+                if (run_count > 1)
+                        fprintf(stderr, " (%d runs)", run_count);
+                fprintf(stderr, ":\n\n");
+        }
 
+        if (no_aggr) {
+                list_for_each_entry(counter, &evsel_list, node)
+                        print_counter(counter);
+        } else {
+                list_for_each_entry(counter, &evsel_list, node)
+                        print_counter_aggr(counter);
+        }
 
-        fprintf(stderr, "\n");
-        fprintf(stderr, " %14.9f  seconds time elapsed.\n",
-                        (double)walltime_nsecs_avg/1e9);
-        fprintf(stderr, "\n");
+        if (!csv_output) {
+                fprintf(stderr, "\n");
+                fprintf(stderr, " %18.9f  seconds time elapsed",
+                                avg_stats(&walltime_nsecs_stats)/1e9);
+                if (run_count > 1) {
+                        fprintf(stderr, "   ( +- %7.3f%% )",
+                                100*stddev_stats(&walltime_nsecs_stats) /
+                                avg_stats(&walltime_nsecs_stats));
+                }
+                fprintf(stderr, "\n\n");
+        }
 }
 
 static volatile int signr = -1;
 
 static void skip_signal(int signo)
 {
+        if(child_pid == -1)
+                done = 1;
+
         signr = signo;
 }
 
 static void sig_atexit(void)
 {
+        if (child_pid != -1)
+                kill(child_pid, SIGTERM);
+
         if (signr == -1)
                 return;
 
@@ -453,49 +598,127 @@ static void sig_atexit(void)
 }
 
 static const char * const stat_usage[] = {
-        "perf stat [<options>] <command>",
+        "perf stat [<options>] [<command>]",
         NULL
 };
 
+static int stat__set_big_num(const struct option *opt __used,
+                             const char *s __used, int unset)
+{
+        big_num_opt = unset ? 0 : 1;
+        return 0;
+}
+
 static const struct option options[] = {
         OPT_CALLBACK('e', "event", NULL, "event",
                      "event selector. use 'perf list' to list available events",
                      parse_events),
-        OPT_BOOLEAN('i', "inherit", &inherit,
-                    "child tasks inherit counters"),
+        OPT_BOOLEAN('i', "no-inherit", &no_inherit,
+                    "child tasks do not inherit counters"),
         OPT_INTEGER('p', "pid", &target_pid,
-                    "stat events on existing pid"),
+                    "stat events on existing process id"),
+        OPT_INTEGER('t', "tid", &target_tid,
+                    "stat events on existing thread id"),
         OPT_BOOLEAN('a', "all-cpus", &system_wide,
-                            "system-wide collection from all CPUs"),
-        OPT_BOOLEAN('S', "scale", &scale,
-                            "scale/normalize counters"),
-        OPT_BOOLEAN('v', "verbose", &verbose,
+                    "system-wide collection from all CPUs"),
+        OPT_BOOLEAN('c', "scale", &scale,
+                    "scale/normalize counters"),
+        OPT_INCR('v', "verbose", &verbose,
                     "be more verbose (show counter open errors, etc)"),
         OPT_INTEGER('r', "repeat", &run_count,
                     "repeat command and print average + stddev (max: 100)"),
+        OPT_BOOLEAN('n', "null", &null_run,
+                    "null run - dont start any counters"),
+        OPT_CALLBACK_NOOPT('B', "big-num", NULL, NULL, 
+                           "print large numbers with thousands\' separators",
+                           stat__set_big_num),
+        OPT_STRING('C', "cpu", &cpu_list, "cpu",
+                    "list of cpus to monitor in system-wide"),
+        OPT_BOOLEAN('A', "no-aggr", &no_aggr,
+                    "disable CPU count aggregation"),
+        OPT_STRING('x', "field-separator", &csv_sep, "separator",
+                   "print counts with custom separator"),
         OPT_END()
 };
 
-int cmd_stat(int argc, const char **argv, const char *prefix)
+int cmd_stat(int argc, const char **argv, const char *prefix __used)
 {
-        int status;
+        struct perf_evsel *pos;
+        int status = -ENOMEM;
 
-        page_size = sysconf(_SC_PAGE_SIZE);
+        setlocale(LC_ALL, "");
 
-        memcpy(attrs, default_attrs, sizeof(attrs));
+        argc = parse_options(argc, argv, options, stat_usage,
+                PARSE_OPT_STOP_AT_NON_OPTION);
 
-        argc = parse_options(argc, argv, options, stat_usage, 0);
-        if (!argc)
+        if (csv_sep)
+                csv_output = true;
+        else
+                csv_sep = DEFAULT_SEPARATOR;
+
+        /*
+         * let the spreadsheet do the pretty-printing
+         */
+        if (csv_output) {
+                /* User explicitely passed -B? */
+                if (big_num_opt == 1) {
+                        fprintf(stderr, "-B option not supported with -x\n");
+                        usage_with_options(stat_usage, options);
+                } else /* Nope, so disable big number formatting */
+                        big_num = false;
+        } else if (big_num_opt == 0) /* User passed --no-big-num */
+                big_num = false;
+
+        if (!argc && target_pid == -1 && target_tid == -1)
+                usage_with_options(stat_usage, options);
+        if (run_count <= 0)
                 usage_with_options(stat_usage, options);
-        if (run_count <= 0 || run_count > MAX_RUN)
+
+        /* no_aggr is for system-wide only */
+        if (no_aggr && !system_wide)
                 usage_with_options(stat_usage, options);
 
-        if (!nr_counters)
-                nr_counters = 8;
+        /* Set attrs and nr_counters if no event is selected and !null_run */
+        if (!null_run && !nr_counters) {
+                size_t c;
 
-        nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
-        assert(nr_cpus <= MAX_NR_CPUS);
-        assert(nr_cpus >= 0);
+                nr_counters = ARRAY_SIZE(default_attrs);
+
+                for (c = 0; c < ARRAY_SIZE(default_attrs); ++c) {
+                        pos = perf_evsel__new(&default_attrs[c],
+                                              nr_counters);
+                        if (pos == NULL)
+                                goto out;
+                        list_add(&pos->node, &evsel_list);
+                }
+        }
+
+        if (target_pid != -1)
+                target_tid = target_pid;
+
+        threads = thread_map__new(target_pid, target_tid);
+        if (threads == NULL) {
+                pr_err("Problems finding threads of monitor\n");
+                usage_with_options(stat_usage, options);
+        }
+
+        if (system_wide)
+                cpus = cpu_map__new(cpu_list);
+        else
+                cpus = cpu_map__dummy_new();
+
+        if (cpus == NULL) {
+                perror("failed to parse CPUs map");
+                usage_with_options(stat_usage, options);
+                return -1;
+        }
+
+        list_for_each_entry(pos, &evsel_list, node) {
+                if (perf_evsel__alloc_stat_priv(pos) < 0 ||
+                    perf_evsel__alloc_counts(pos, cpus->nr) < 0 ||
+                    perf_evsel__alloc_fd(pos, cpus->nr, threads->nr) < 0)
+                        goto out_free_fd;
+        }
 
         /*
          * We dont want to block the signals - that would cause
@@ -511,11 +734,17 @@ int cmd_stat(int argc, const char **argv, const char *prefix)
         status = 0;
         for (run_idx = 0; run_idx < run_count; run_idx++) {
                 if (run_count != 1 && verbose)
-                        fprintf(stderr, "[ perf stat: executing run #%d ... ]\n", run_idx+1);
+                        fprintf(stderr, "[ perf stat: executing run #%d ... ]\n", run_idx + 1);
                 status = run_perf_stat(argc, argv);
         }
 
-        print_stat(argc, argv);
-
+        if (status != -1)
+                print_stat(argc, argv);
+out_free_fd:
+        list_for_each_entry(pos, &evsel_list, node)
+                perf_evsel__free_stat_priv(pos);
+out:
+        thread_map__delete(threads);
+        threads = NULL;
         return status;
 }
diff --git a/tools/perf/builtin-test.c b/tools/perf/builtin-test.c
new file mode 100644
index 00000000000..ed5696198d3
--- /dev/null
+++ b/tools/perf/builtin-test.c
@@ -0,0 +1,497 @@
+/*
+ * builtin-test.c
+ *
+ * Builtin regression testing command: ever growing number of sanity tests
+ */
+#include "builtin.h"
+
+#include "util/cache.h"
+#include "util/debug.h"
+#include "util/parse-options.h"
+#include "util/session.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+
+static long page_size;
+
+static int vmlinux_matches_kallsyms_filter(struct map *map __used, struct symbol *sym)
+{
+        bool *visited = symbol__priv(sym);
+        *visited = true;
+        return 0;
+}
+
+static int test__vmlinux_matches_kallsyms(void)
+{
+        int err = -1;
+        struct rb_node *nd;
+        struct symbol *sym;
+        struct map *kallsyms_map, *vmlinux_map;
+        struct machine kallsyms, vmlinux;
+        enum map_type type = MAP__FUNCTION;
+        struct ref_reloc_sym ref_reloc_sym = { .name = "_stext", };
+
+        /*
+         * Step 1:
+         *
+         * Init the machines that will hold kernel, modules obtained from
+         * both vmlinux + .ko files and from /proc/kallsyms split by modules.
+         */
+        machine__init(&kallsyms, "", HOST_KERNEL_ID);
+        machine__init(&vmlinux, "", HOST_KERNEL_ID);
+
+        /*
+         * Step 2:
+         *
+         * Create the kernel maps for kallsyms and the DSO where we will then
+         * load /proc/kallsyms. Also create the modules maps from /proc/modules
+         * and find the .ko files that match them in /lib/modules/`uname -r`/.
+         */
+        if (machine__create_kernel_maps(&kallsyms) < 0) {
+                pr_debug("machine__create_kernel_maps ");
+                return -1;
+        }
+
+        /*
+         * Step 3:
+         *
+         * Load and split /proc/kallsyms into multiple maps, one per module.
+         */
+        if (machine__load_kallsyms(&kallsyms, "/proc/kallsyms", type, NULL) <= 0) {
+                pr_debug("dso__load_kallsyms ");
+                goto out;
+        }
+
+        /*
+         * Step 4:
+         *
+         * kallsyms will be internally on demand sorted by name so that we can
+         * find the reference relocation * symbol, i.e. the symbol we will use
+         * to see if the running kernel was relocated by checking if it has the
+         * same value in the vmlinux file we load.
+         */
+        kallsyms_map = machine__kernel_map(&kallsyms, type);
+
+        sym = map__find_symbol_by_name(kallsyms_map, ref_reloc_sym.name, NULL);
+        if (sym == NULL) {
+                pr_debug("dso__find_symbol_by_name ");
+                goto out;
+        }
+
+        ref_reloc_sym.addr = sym->start;
+
+        /*
+         * Step 5:
+         *
+         * Now repeat step 2, this time for the vmlinux file we'll auto-locate.
+         */
+        if (machine__create_kernel_maps(&vmlinux) < 0) {
+                pr_debug("machine__create_kernel_maps ");
+                goto out;
+        }
+
+        vmlinux_map = machine__kernel_map(&vmlinux, type);
+        map__kmap(vmlinux_map)->ref_reloc_sym = &ref_reloc_sym;
+
+        /*
+         * Step 6:
+         *
+         * Locate a vmlinux file in the vmlinux path that has a buildid that
+         * matches the one of the running kernel.
+         *
+         * While doing that look if we find the ref reloc symbol, if we find it
+         * we'll have its ref_reloc_symbol.unrelocated_addr and then
+         * maps__reloc_vmlinux will notice and set proper ->[un]map_ip routines
+         * to fixup the symbols.
+         */
+        if (machine__load_vmlinux_path(&vmlinux, type,
+                                       vmlinux_matches_kallsyms_filter) <= 0) {
+                pr_debug("machine__load_vmlinux_path ");
+                goto out;
+        }
+
+        err = 0;
+        /*
+         * Step 7:
+         *
+         * Now look at the symbols in the vmlinux DSO and check if we find all of them
+         * in the kallsyms dso. For the ones that are in both, check its names and
+         * end addresses too.
+         */
+        for (nd = rb_first(&vmlinux_map->dso->symbols[type]); nd; nd = rb_next(nd)) {
+                struct symbol *pair, *first_pair;
+                bool backwards = true;
+
+                sym  = rb_entry(nd, struct symbol, rb_node);
+
+                if (sym->start == sym->end)
+                        continue;
+
+                first_pair = machine__find_kernel_symbol(&kallsyms, type, sym->start, NULL, NULL);
+                pair = first_pair;
+
+                if (pair && pair->start == sym->start) {
+next_pair:
+                        if (strcmp(sym->name, pair->name) == 0) {
+                                /*
+                                 * kallsyms don't have the symbol end, so we
+                                 * set that by using the next symbol start - 1,
+                                 * in some cases we get this up to a page
+                                 * wrong, trace_kmalloc when I was developing
+                                 * this code was one such example, 2106 bytes
+                                 * off the real size. More than that and we
+                                 * _really_ have a problem.
+                                 */
+                                s64 skew = sym->end - pair->end;
+                                if (llabs(skew) < page_size)
+                                        continue;
+
+                                pr_debug("%#Lx: diff end addr for %s v: %#Lx k: %#Lx\n",
+                                         sym->start, sym->name, sym->end, pair->end);
+                        } else {
+                                struct rb_node *nnd;
+detour:
+                                nnd = backwards ? rb_prev(&pair->rb_node) :
+                                                  rb_next(&pair->rb_node);
+                                if (nnd) {
+                                        struct symbol *next = rb_entry(nnd, struct symbol, rb_node);
+
+                                        if (next->start == sym->start) {
+                                                pair = next;
+                                                goto next_pair;
+                                        }
+                                }
+
+                                if (backwards) {
+                                        backwards = false;
+                                        pair = first_pair;
+                                        goto detour;
+                                }
+
+                                pr_debug("%#Lx: diff name v: %s k: %s\n",
+                                         sym->start, sym->name, pair->name);
+                        }
+                } else
+                        pr_debug("%#Lx: %s not on kallsyms\n", sym->start, sym->name);
+
+                err = -1;
+        }
+
+        if (!verbose)
+                goto out;
+
+        pr_info("Maps only in vmlinux:\n");
+
+        for (nd = rb_first(&vmlinux.kmaps.maps[type]); nd; nd = rb_next(nd)) {
+                struct map *pos = rb_entry(nd, struct map, rb_node), *pair;
+                /*
+                 * If it is the kernel, kallsyms is always "[kernel.kallsyms]", while
+                 * the kernel will have the path for the vmlinux file being used,
+                 * so use the short name, less descriptive but the same ("[kernel]" in
+                 * both cases.
+                 */
+                pair = map_groups__find_by_name(&kallsyms.kmaps, type,
+                                                (pos->dso->kernel ?
+                                                        pos->dso->short_name :
+                                                        pos->dso->name));
+                if (pair)
+                        pair->priv = 1;
+                else
+                        map__fprintf(pos, stderr);
+        }
+
+        pr_info("Maps in vmlinux with a different name in kallsyms:\n");
+
+        for (nd = rb_first(&vmlinux.kmaps.maps[type]); nd; nd = rb_next(nd)) {
+                struct map *pos = rb_entry(nd, struct map, rb_node), *pair;
+
+                pair = map_groups__find(&kallsyms.kmaps, type, pos->start);
+                if (pair == NULL || pair->priv)
+                        continue;
+
+                if (pair->start == pos->start) {
+                        pair->priv = 1;
+                        pr_info(" %Lx-%Lx %Lx %s in kallsyms as",
+                                pos->start, pos->end, pos->pgoff, pos->dso->name);
+                        if (pos->pgoff != pair->pgoff || pos->end != pair->end)
+                                pr_info(": \n*%Lx-%Lx %Lx",
+                                        pair->start, pair->end, pair->pgoff);
+                        pr_info(" %s\n", pair->dso->name);
+                        pair->priv = 1;
+                }
+        }
+
+        pr_info("Maps only in kallsyms:\n");
+
+        for (nd = rb_first(&kallsyms.kmaps.maps[type]);
+             nd; nd = rb_next(nd)) {
+                struct map *pos = rb_entry(nd, struct map, rb_node);
+
+                if (!pos->priv)
+                        map__fprintf(pos, stderr);
+        }
+out:
+        return err;
+}
+
+#include "util/cpumap.h"
+#include "util/evsel.h"
+#include <sys/types.h>
+
+static int trace_event__id(const char *event_name)
+{
+        char *filename;
+        int err = -1, fd;
+
+        if (asprintf(&filename,
+                     "/sys/kernel/debug/tracing/events/syscalls/%s/id",
+                     event_name) < 0)
+                return -1;
+
+        fd = open(filename, O_RDONLY);
+        if (fd >= 0) {
+                char id[16];
+                if (read(fd, id, sizeof(id)) > 0)
+                        err = atoi(id);
+                close(fd);
+        }
+
+        free(filename);
+        return err;
+}
+
+static int test__open_syscall_event(void)
+{
+        int err = -1, fd;
+        struct thread_map *threads;
+        struct perf_evsel *evsel;
+        struct perf_event_attr attr;
+        unsigned int nr_open_calls = 111, i;
+        int id = trace_event__id("sys_enter_open");
+
+        if (id < 0) {
+                pr_debug("is debugfs mounted on /sys/kernel/debug?\n");
+                return -1;
+        }
+
+        threads = thread_map__new(-1, getpid());
+        if (threads == NULL) {
+                pr_debug("thread_map__new\n");
+                return -1;
+        }
+
+        memset(&attr, 0, sizeof(attr));
+        attr.type = PERF_TYPE_TRACEPOINT;
+        attr.config = id;
+        evsel = perf_evsel__new(&attr, 0);
+        if (evsel == NULL) {
+                pr_debug("perf_evsel__new\n");
+                goto out_thread_map_delete;
+        }
+
+        if (perf_evsel__open_per_thread(evsel, threads) < 0) {
+                pr_debug("failed to open counter: %s, "
+                         "tweak /proc/sys/kernel/perf_event_paranoid?\n",
+                         strerror(errno));
+                goto out_evsel_delete;
+        }
+
+        for (i = 0; i < nr_open_calls; ++i) {
+                fd = open("/etc/passwd", O_RDONLY);
+                close(fd);
+        }
+
+        if (perf_evsel__read_on_cpu(evsel, 0, 0) < 0) {
+                pr_debug("perf_evsel__open_read_on_cpu\n");
+                goto out_close_fd;
+        }
+
+        if (evsel->counts->cpu[0].val != nr_open_calls) {
+                pr_debug("perf_evsel__read_on_cpu: expected to intercept %d calls, got %Ld\n",
+                         nr_open_calls, evsel->counts->cpu[0].val);
+                goto out_close_fd;
+        }
+        
+        err = 0;
+out_close_fd:
+        perf_evsel__close_fd(evsel, 1, threads->nr);
+out_evsel_delete:
+        perf_evsel__delete(evsel);
+out_thread_map_delete:
+        thread_map__delete(threads);
+        return err;
+}
+
+#include <sched.h>
+
+static int test__open_syscall_event_on_all_cpus(void)
+{
+        int err = -1, fd, cpu;
+        struct thread_map *threads;
+        struct cpu_map *cpus;
+        struct perf_evsel *evsel;
+        struct perf_event_attr attr;
+        unsigned int nr_open_calls = 111, i;
+        cpu_set_t *cpu_set;
+        size_t cpu_set_size;
+        int id = trace_event__id("sys_enter_open");
+
+        if (id < 0) {
+                pr_debug("is debugfs mounted on /sys/kernel/debug?\n");
+                return -1;
+        }
+
+        threads = thread_map__new(-1, getpid());
+        if (threads == NULL) {
+                pr_debug("thread_map__new\n");
+                return -1;
+        }
+
+        cpus = cpu_map__new(NULL);
+        if (threads == NULL) {
+                pr_debug("thread_map__new\n");
+                return -1;
+        }
+
+        cpu_set = CPU_ALLOC(cpus->nr);
+
+        if (cpu_set == NULL)
+                goto out_thread_map_delete;
+
+        cpu_set_size = CPU_ALLOC_SIZE(cpus->nr);
+        CPU_ZERO_S(cpu_set_size, cpu_set);
+
+        memset(&attr, 0, sizeof(attr));
+        attr.type = PERF_TYPE_TRACEPOINT;
+        attr.config = id;
+        evsel = perf_evsel__new(&attr, 0);
+        if (evsel == NULL) {
+                pr_debug("perf_evsel__new\n");
+                goto out_cpu_free;
+        }
+
+        if (perf_evsel__open(evsel, cpus, threads) < 0) {
+                pr_debug("failed to open counter: %s, "
+                         "tweak /proc/sys/kernel/perf_event_paranoid?\n",
+                         strerror(errno));
+                goto out_evsel_delete;
+        }
+
+        for (cpu = 0; cpu < cpus->nr; ++cpu) {
+                unsigned int ncalls = nr_open_calls + cpu;
+
+                CPU_SET(cpu, cpu_set);
+                sched_setaffinity(0, cpu_set_size, cpu_set);
+                for (i = 0; i < ncalls; ++i) {
+                        fd = open("/etc/passwd", O_RDONLY);
+                        close(fd);
+                }
+                CPU_CLR(cpu, cpu_set);
+        }
+
+        /*
+         * Here we need to explicitely preallocate the counts, as if
+         * we use the auto allocation it will allocate just for 1 cpu,
+         * as we start by cpu 0.
+         */
+        if (perf_evsel__alloc_counts(evsel, cpus->nr) < 0) {
+                pr_debug("perf_evsel__alloc_counts(ncpus=%d)\n", cpus->nr);
+                goto out_close_fd;
+        }
+
+        for (cpu = 0; cpu < cpus->nr; ++cpu) {
+                unsigned int expected;
+
+                if (perf_evsel__read_on_cpu(evsel, cpu, 0) < 0) {
+                        pr_debug("perf_evsel__open_read_on_cpu\n");
+                        goto out_close_fd;
+                }
+
+                expected = nr_open_calls + cpu;
+                if (evsel->counts->cpu[cpu].val != expected) {
+                        pr_debug("perf_evsel__read_on_cpu: expected to intercept %d calls on cpu %d, got %Ld\n",
+                                 expected, cpu, evsel->counts->cpu[cpu].val);
+                        goto out_close_fd;
+                }
+        }
+
+        err = 0;
+out_close_fd:
+        perf_evsel__close_fd(evsel, 1, threads->nr);
+out_evsel_delete:
+        perf_evsel__delete(evsel);
+out_cpu_free:
+        CPU_FREE(cpu_set);
+out_thread_map_delete:
+        thread_map__delete(threads);
+        return err;
+}
+
+static struct test {
+        const char *desc;
+        int (*func)(void);
+} tests[] = {
+        {
+                .desc = "vmlinux symtab matches kallsyms",
+                .func = test__vmlinux_matches_kallsyms,
+        },
+        {
+                .desc = "detect open syscall event",
+                .func = test__open_syscall_event,
+        },
+        {
+                .desc = "detect open syscall event on all cpus",
+                .func = test__open_syscall_event_on_all_cpus,
+        },
+        {
+                .func = NULL,
+        },
+};
+
+static int __cmd_test(void)
+{
+        int i = 0;
+
+        page_size = sysconf(_SC_PAGE_SIZE);
+
+        while (tests[i].func) {
+                int err;
+                pr_info("%2d: %s:", i + 1, tests[i].desc);
+                pr_debug("\n--- start ---\n");
+                err = tests[i].func();
+                pr_debug("---- end ----\n%s:", tests[i].desc);
+                pr_info(" %s\n", err ? "FAILED!\n" : "Ok");
+                ++i;
+        }
+
+        return 0;
+}
+
+static const char * const test_usage[] = {
+        "perf test [<options>]",
+        NULL,
+};
+
+static const struct option test_options[] = {
+        OPT_INTEGER('v', "verbose", &verbose,
+                    "be more verbose (show symbol address, etc)"),
+        OPT_END()
+};
+
+int cmd_test(int argc, const char **argv, const char *prefix __used)
+{
+        argc = parse_options(argc, argv, test_options, test_usage, 0);
+        if (argc)
+                usage_with_options(test_usage, test_options);
+
+        symbol_conf.priv_size = sizeof(int);
+        symbol_conf.sort_by_name = true;
+        symbol_conf.try_vmlinux_path = true;
+
+        if (symbol__init() < 0)
+                return -1;
+
+        setup_pager();
+
+        return __cmd_test();
+}
diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c
new file mode 100644
index 00000000000..746cf03cb05
--- /dev/null
+++ b/tools/perf/builtin-timechart.c
@@ -0,0 +1,1104 @@
+/*
+ * builtin-timechart.c - make an svg timechart of system activity
+ *
+ * (C) Copyright 2009 Intel Corporation
+ *
+ * Authors:
+ *     Arjan van de Ven <arjan@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include "builtin.h"
+
+#include "util/util.h"
+
+#include "util/color.h"
+#include <linux/list.h>
+#include "util/cache.h"
+#include <linux/rbtree.h>
+#include "util/symbol.h"
+#include "util/callchain.h"
+#include "util/strlist.h"
+
+#include "perf.h"
+#include "util/header.h"
+#include "util/parse-options.h"
+#include "util/parse-events.h"
+#include "util/event.h"
+#include "util/session.h"
+#include "util/svghelper.h"
+
+#define SUPPORT_OLD_POWER_EVENTS 1
+#define PWR_EVENT_EXIT -1
+
+
+static char             const *input_name = "perf.data";
+static char             const *output_name = "output.svg";
+
+static unsigned int     numcpus;
+static u64              min_freq;       /* Lowest CPU frequency seen */
+static u64              max_freq;       /* Highest CPU frequency seen */
+static u64              turbo_frequency;
+
+static u64              first_time, last_time;
+
+static bool             power_only;
+
+
+struct per_pid;
+struct per_pidcomm;
+
+struct cpu_sample;
+struct power_event;
+struct wake_event;
+
+struct sample_wrapper;
+
+/*
+ * Datastructure layout:
+ * We keep an list of "pid"s, matching the kernels notion of a task struct.
+ * Each "pid" entry, has a list of "comm"s.
+ *      this is because we want to track different programs different, while
+ *      exec will reuse the original pid (by design).
+ * Each comm has a list of samples that will be used to draw
+ * final graph.
+ */
+
+struct per_pid {
+        struct per_pid *next;
+
+        int             pid;
+        int             ppid;
+
+        u64             start_time;
+        u64             end_time;
+        u64             total_time;
+        int             display;
+
+        struct per_pidcomm *all;
+        struct per_pidcomm *current;
+};
+
+
+struct per_pidcomm {
+        struct per_pidcomm *next;
+
+        u64             start_time;
+        u64             end_time;
+        u64             total_time;
+
+        int             Y;
+        int             display;
+
+        long            state;
+        u64             state_since;
+
+        char            *comm;
+
+        struct cpu_sample *samples;
+};
+
+struct sample_wrapper {
+        struct sample_wrapper *next;
+
+        u64             timestamp;
+        unsigned char   data[0];
+};
+
+#define TYPE_NONE       0
+#define TYPE_RUNNING    1
+#define TYPE_WAITING    2
+#define TYPE_BLOCKED    3
+
+struct cpu_sample {
+        struct cpu_sample *next;
+
+        u64 start_time;
+        u64 end_time;
+        int type;
+        int cpu;
+};
+
+static struct per_pid *all_data;
+
+#define CSTATE 1
+#define PSTATE 2
+
+struct power_event {
+        struct power_event *next;
+        int type;
+        int state;
+        u64 start_time;
+        u64 end_time;
+        int cpu;
+};
+
+struct wake_event {
+        struct wake_event *next;
+        int waker;
+        int wakee;
+        u64 time;
+};
+
+static struct power_event    *power_events;
+static struct wake_event     *wake_events;
+
+struct process_filter;
+struct process_filter {
+        char                    *name;
+        int                     pid;
+        struct process_filter   *next;
+};
+
+static struct process_filter *process_filter;
+
+
+static struct per_pid *find_create_pid(int pid)
+{
+        struct per_pid *cursor = all_data;
+
+        while (cursor) {
+                if (cursor->pid == pid)
+                        return cursor;
+                cursor = cursor->next;
+        }
+        cursor = malloc(sizeof(struct per_pid));
+        assert(cursor != NULL);
+        memset(cursor, 0, sizeof(struct per_pid));
+        cursor->pid = pid;
+        cursor->next = all_data;
+        all_data = cursor;
+        return cursor;
+}
+
+static void pid_set_comm(int pid, char *comm)
+{
+        struct per_pid *p;
+        struct per_pidcomm *c;
+        p = find_create_pid(pid);
+        c = p->all;
+        while (c) {
+                if (c->comm && strcmp(c->comm, comm) == 0) {
+                        p->current = c;
+                        return;
+                }
+                if (!c->comm) {
+                        c->comm = strdup(comm);
+                        p->current = c;
+                        return;
+                }
+                c = c->next;
+        }
+        c = malloc(sizeof(struct per_pidcomm));
+        assert(c != NULL);
+        memset(c, 0, sizeof(struct per_pidcomm));
+        c->comm = strdup(comm);
+        p->current = c;
+        c->next = p->all;
+        p->all = c;
+}
+
+static void pid_fork(int pid, int ppid, u64 timestamp)
+{
+        struct per_pid *p, *pp;
+        p = find_create_pid(pid);
+        pp = find_create_pid(ppid);
+        p->ppid = ppid;
+        if (pp->current && pp->current->comm && !p->current)
+                pid_set_comm(pid, pp->current->comm);
+
+        p->start_time = timestamp;
+        if (p->current) {
+                p->current->start_time = timestamp;
+                p->current->state_since = timestamp;
+        }
+}
+
+static void pid_exit(int pid, u64 timestamp)
+{
+        struct per_pid *p;
+        p = find_create_pid(pid);
+        p->end_time = timestamp;
+        if (p->current)
+                p->current->end_time = timestamp;
+}
+
+static void
+pid_put_sample(int pid, int type, unsigned int cpu, u64 start, u64 end)
+{
+        struct per_pid *p;
+        struct per_pidcomm *c;
+        struct cpu_sample *sample;
+
+        p = find_create_pid(pid);
+        c = p->current;
+        if (!c) {
+                c = malloc(sizeof(struct per_pidcomm));
+                assert(c != NULL);
+                memset(c, 0, sizeof(struct per_pidcomm));
+                p->current = c;
+                c->next = p->all;
+                p->all = c;
+        }
+
+        sample = malloc(sizeof(struct cpu_sample));
+        assert(sample != NULL);
+        memset(sample, 0, sizeof(struct cpu_sample));
+        sample->start_time = start;
+        sample->end_time = end;
+        sample->type = type;
+        sample->next = c->samples;
+        sample->cpu = cpu;
+        c->samples = sample;
+
+        if (sample->type == TYPE_RUNNING && end > start && start > 0) {
+                c->total_time += (end-start);
+                p->total_time += (end-start);
+        }
+
+        if (c->start_time == 0 || c->start_time > start)
+                c->start_time = start;
+        if (p->start_time == 0 || p->start_time > start)
+                p->start_time = start;
+
+        if (cpu > numcpus)
+                numcpus = cpu;
+}
+
+#define MAX_CPUS 4096
+
+static u64 cpus_cstate_start_times[MAX_CPUS];
+static int cpus_cstate_state[MAX_CPUS];
+static u64 cpus_pstate_start_times[MAX_CPUS];
+static u64 cpus_pstate_state[MAX_CPUS];
+
+static int process_comm_event(event_t *event, struct sample_data *sample __used,
+                              struct perf_session *session __used)
+{
+        pid_set_comm(event->comm.tid, event->comm.comm);
+        return 0;
+}
+
+static int process_fork_event(event_t *event, struct sample_data *sample __used,
+                              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, struct sample_data *sample __used,
+                              struct perf_session *session __used)
+{
+        pid_exit(event->fork.pid, event->fork.time);
+        return 0;
+}
+
+struct trace_entry {
+        unsigned short          type;
+        unsigned char           flags;
+        unsigned char           preempt_count;
+        int                     pid;
+        int                     lock_depth;
+};
+
+#ifdef SUPPORT_OLD_POWER_EVENTS
+static int use_old_power_events;
+struct power_entry_old {
+        struct trace_entry te;
+        u64     type;
+        u64     value;
+        u64     cpu_id;
+};
+#endif
+
+struct power_processor_entry {
+        struct trace_entry te;
+        u32     state;
+        u32     cpu_id;
+};
+
+#define TASK_COMM_LEN 16
+struct wakeup_entry {
+        struct trace_entry te;
+        char comm[TASK_COMM_LEN];
+        int   pid;
+        int   prio;
+        int   success;
+};
+
+/*
+ * trace_flag_type is an enumeration that holds different
+ * states when a trace occurs. These are:
+ *  IRQS_OFF            - interrupts were disabled
+ *  IRQS_NOSUPPORT      - arch does not support irqs_disabled_flags
+ *  NEED_RESCED         - reschedule is requested
+ *  HARDIRQ             - inside an interrupt handler
+ *  SOFTIRQ             - inside a softirq handler
+ */
+enum trace_flag_type {
+        TRACE_FLAG_IRQS_OFF             = 0x01,
+        TRACE_FLAG_IRQS_NOSUPPORT       = 0x02,
+        TRACE_FLAG_NEED_RESCHED         = 0x04,
+        TRACE_FLAG_HARDIRQ              = 0x08,
+        TRACE_FLAG_SOFTIRQ              = 0x10,
+};
+
+
+
+struct sched_switch {
+        struct trace_entry te;
+        char prev_comm[TASK_COMM_LEN];
+        int  prev_pid;
+        int  prev_prio;
+        long prev_state; /* Arjan weeps. */
+        char next_comm[TASK_COMM_LEN];
+        int  next_pid;
+        int  next_prio;
+};
+
+static void c_state_start(int cpu, u64 timestamp, int state)
+{
+        cpus_cstate_start_times[cpu] = timestamp;
+        cpus_cstate_state[cpu] = state;
+}
+
+static void c_state_end(int cpu, u64 timestamp)
+{
+        struct power_event *pwr;
+        pwr = malloc(sizeof(struct power_event));
+        if (!pwr)
+                return;
+        memset(pwr, 0, sizeof(struct power_event));
+
+        pwr->state = cpus_cstate_state[cpu];
+        pwr->start_time = cpus_cstate_start_times[cpu];
+        pwr->end_time = timestamp;
+        pwr->cpu = cpu;
+        pwr->type = CSTATE;
+        pwr->next = power_events;
+
+        power_events = pwr;
+}
+
+static void p_state_change(int cpu, u64 timestamp, u64 new_freq)
+{
+        struct power_event *pwr;
+        pwr = malloc(sizeof(struct power_event));
+
+        if (new_freq > 8000000) /* detect invalid data */
+                return;
+
+        if (!pwr)
+                return;
+        memset(pwr, 0, sizeof(struct power_event));
+
+        pwr->state = cpus_pstate_state[cpu];
+        pwr->start_time = cpus_pstate_start_times[cpu];
+        pwr->end_time = timestamp;
+        pwr->cpu = cpu;
+        pwr->type = PSTATE;
+        pwr->next = power_events;
+
+        if (!pwr->start_time)
+                pwr->start_time = first_time;
+
+        power_events = pwr;
+
+        cpus_pstate_state[cpu] = new_freq;
+        cpus_pstate_start_times[cpu] = timestamp;
+
+        if ((u64)new_freq > max_freq)
+                max_freq = new_freq;
+
+        if (new_freq < min_freq || min_freq == 0)
+                min_freq = new_freq;
+
+        if (new_freq == max_freq - 1000)
+                        turbo_frequency = max_freq;
+}
+
+static void
+sched_wakeup(int cpu, u64 timestamp, int pid, struct trace_entry *te)
+{
+        struct wake_event *we;
+        struct per_pid *p;
+        struct wakeup_entry *wake = (void *)te;
+
+        we = malloc(sizeof(struct wake_event));
+        if (!we)
+                return;
+
+        memset(we, 0, sizeof(struct wake_event));
+        we->time = timestamp;
+        we->waker = pid;
+
+        if ((te->flags & TRACE_FLAG_HARDIRQ) || (te->flags & TRACE_FLAG_SOFTIRQ))
+                we->waker = -1;
+
+        we->wakee = wake->pid;
+        we->next = wake_events;
+        wake_events = we;
+        p = find_create_pid(we->wakee);
+
+        if (p && p->current && p->current->state == TYPE_NONE) {
+                p->current->state_since = timestamp;
+                p->current->state = TYPE_WAITING;
+        }
+        if (p && p->current && p->current->state == TYPE_BLOCKED) {
+                pid_put_sample(p->pid, p->current->state, cpu, p->current->state_since, timestamp);
+                p->current->state_since = timestamp;
+                p->current->state = TYPE_WAITING;
+        }
+}
+
+static void sched_switch(int cpu, u64 timestamp, struct trace_entry *te)
+{
+        struct per_pid *p = NULL, *prev_p;
+        struct sched_switch *sw = (void *)te;
+
+
+        prev_p = find_create_pid(sw->prev_pid);
+
+        p = find_create_pid(sw->next_pid);
+
+        if (prev_p->current && prev_p->current->state != TYPE_NONE)
+                pid_put_sample(sw->prev_pid, TYPE_RUNNING, cpu, prev_p->current->state_since, timestamp);
+        if (p && p->current) {
+                if (p->current->state != TYPE_NONE)
+                        pid_put_sample(sw->next_pid, p->current->state, cpu, p->current->state_since, timestamp);
+
+                p->current->state_since = timestamp;
+                p->current->state = TYPE_RUNNING;
+        }
+
+        if (prev_p->current) {
+                prev_p->current->state = TYPE_NONE;
+                prev_p->current->state_since = timestamp;
+                if (sw->prev_state & 2)
+                        prev_p->current->state = TYPE_BLOCKED;
+                if (sw->prev_state == 0)
+                        prev_p->current->state = TYPE_WAITING;
+        }
+}
+
+
+static int process_sample_event(event_t *event __used,
+                                struct sample_data *sample,
+                                struct perf_session *session)
+{
+        struct trace_entry *te;
+
+        if (session->sample_type & PERF_SAMPLE_TIME) {
+                if (!first_time || first_time > sample->time)
+                        first_time = sample->time;
+                if (last_time < sample->time)
+                        last_time = sample->time;
+        }
+
+        te = (void *)sample->raw_data;
+        if (session->sample_type & PERF_SAMPLE_RAW && sample->raw_size > 0) {
+                char *event_str;
+#ifdef SUPPORT_OLD_POWER_EVENTS
+                struct power_entry_old *peo;
+                peo = (void *)te;
+#endif
+                event_str = perf_header__find_event(te->type);
+
+                if (!event_str)
+                        return 0;
+
+                if (strcmp(event_str, "power:cpu_idle") == 0) {
+                        struct power_processor_entry *ppe = (void *)te;
+                        if (ppe->state == (u32)PWR_EVENT_EXIT)
+                                c_state_end(ppe->cpu_id, sample->time);
+                        else
+                                c_state_start(ppe->cpu_id, sample->time,
+                                              ppe->state);
+                }
+                else if (strcmp(event_str, "power:cpu_frequency") == 0) {
+                        struct power_processor_entry *ppe = (void *)te;
+                        p_state_change(ppe->cpu_id, sample->time, ppe->state);
+                }
+
+                else if (strcmp(event_str, "sched:sched_wakeup") == 0)
+                        sched_wakeup(sample->cpu, sample->time, sample->pid, te);
+
+                else if (strcmp(event_str, "sched:sched_switch") == 0)
+                        sched_switch(sample->cpu, sample->time, te);
+
+#ifdef SUPPORT_OLD_POWER_EVENTS
+                if (use_old_power_events) {
+                        if (strcmp(event_str, "power:power_start") == 0)
+                                c_state_start(peo->cpu_id, sample->time,
+                                              peo->value);
+
+                        else if (strcmp(event_str, "power:power_end") == 0)
+                                c_state_end(sample->cpu, sample->time);
+
+                        else if (strcmp(event_str,
+                                        "power:power_frequency") == 0)
+                                p_state_change(peo->cpu_id, sample->time,
+                                               peo->value);
+                }
+#endif
+        }
+        return 0;
+}
+
+/*
+ * After the last sample we need to wrap up the current C/P state
+ * and close out each CPU for these.
+ */
+static void end_sample_processing(void)
+{
+        u64 cpu;
+        struct power_event *pwr;
+
+        for (cpu = 0; cpu <= numcpus; cpu++) {
+                pwr = malloc(sizeof(struct power_event));
+                if (!pwr)
+                        return;
+                memset(pwr, 0, sizeof(struct power_event));
+
+                /* C state */
+#if 0
+                pwr->state = cpus_cstate_state[cpu];
+                pwr->start_time = cpus_cstate_start_times[cpu];
+                pwr->end_time = last_time;
+                pwr->cpu = cpu;
+                pwr->type = CSTATE;
+                pwr->next = power_events;
+
+                power_events = pwr;
+#endif
+                /* P state */
+
+                pwr = malloc(sizeof(struct power_event));
+                if (!pwr)
+                        return;
+                memset(pwr, 0, sizeof(struct power_event));
+
+                pwr->state = cpus_pstate_state[cpu];
+                pwr->start_time = cpus_pstate_start_times[cpu];
+                pwr->end_time = last_time;
+                pwr->cpu = cpu;
+                pwr->type = PSTATE;
+                pwr->next = power_events;
+
+                if (!pwr->start_time)
+                        pwr->start_time = first_time;
+                if (!pwr->state)
+                        pwr->state = min_freq;
+                power_events = pwr;
+        }
+}
+
+/*
+ * Sort the pid datastructure
+ */
+static void sort_pids(void)
+{
+        struct per_pid *new_list, *p, *cursor, *prev;
+        /* sort by ppid first, then by pid, lowest to highest */
+
+        new_list = NULL;
+
+        while (all_data) {
+                p = all_data;
+                all_data = p->next;
+                p->next = NULL;
+
+                if (new_list == NULL) {
+                        new_list = p;
+                        p->next = NULL;
+                        continue;
+                }
+                prev = NULL;
+                cursor = new_list;
+                while (cursor) {
+                        if (cursor->ppid > p->ppid ||
+                                (cursor->ppid == p->ppid && cursor->pid > p->pid)) {
+                                /* must insert before */
+                                if (prev) {
+                                        p->next = prev->next;
+                                        prev->next = p;
+                                        cursor = NULL;
+                                        continue;
+                                } else {
+                                        p->next = new_list;
+                                        new_list = p;
+                                        cursor = NULL;
+                                        continue;
+                                }
+                        }
+
+                        prev = cursor;
+                        cursor = cursor->next;
+                        if (!cursor)
+                                prev->next = p;
+                }
+        }
+        all_data = new_list;
+}
+
+
+static void draw_c_p_states(void)
+{
+        struct power_event *pwr;
+        pwr = power_events;
+
+        /*
+         * two pass drawing so that the P state bars are on top of the C state blocks
+         */
+        while (pwr) {
+                if (pwr->type == CSTATE)
+                        svg_cstate(pwr->cpu, pwr->start_time, pwr->end_time, pwr->state);
+                pwr = pwr->next;
+        }
+
+        pwr = power_events;
+        while (pwr) {
+                if (pwr->type == PSTATE) {
+                        if (!pwr->state)
+                                pwr->state = min_freq;
+                        svg_pstate(pwr->cpu, pwr->start_time, pwr->end_time, pwr->state);
+                }
+                pwr = pwr->next;
+        }
+}
+
+static void draw_wakeups(void)
+{
+        struct wake_event *we;
+        struct per_pid *p;
+        struct per_pidcomm *c;
+
+        we = wake_events;
+        while (we) {
+                int from = 0, to = 0;
+                char *task_from = NULL, *task_to = NULL;
+
+                /* locate the column of the waker and wakee */
+                p = all_data;
+                while (p) {
+                        if (p->pid == we->waker || p->pid == we->wakee) {
+                                c = p->all;
+                                while (c) {
+                                        if (c->Y && c->start_time <= we->time && c->end_time >= we->time) {
+                                                if (p->pid == we->waker && !from) {
+                                                        from = c->Y;
+                                                        task_from = strdup(c->comm);
+                                                }
+                                                if (p->pid == we->wakee && !to) {
+                                                        to = c->Y;
+                                                        task_to = strdup(c->comm);
+                                                }
+                                        }
+                                        c = c->next;
+                                }
+                                c = p->all;
+                                while (c) {
+                                        if (p->pid == we->waker && !from) {
+                                                from = c->Y;
+                                                task_from = strdup(c->comm);
+                                        }
+                                        if (p->pid == we->wakee && !to) {
+                                                to = c->Y;
+                                                task_to = strdup(c->comm);
+                                        }
+                                        c = c->next;
+                                }
+                        }
+                        p = p->next;
+                }
+
+                if (!task_from) {
+                        task_from = malloc(40);
+                        sprintf(task_from, "[%i]", we->waker);
+                }
+                if (!task_to) {
+                        task_to = malloc(40);
+                        sprintf(task_to, "[%i]", we->wakee);
+                }
+
+                if (we->waker == -1)
+                        svg_interrupt(we->time, to);
+                else if (from && to && abs(from - to) == 1)
+                        svg_wakeline(we->time, from, to);
+                else
+                        svg_partial_wakeline(we->time, from, task_from, to, task_to);
+                we = we->next;
+
+                free(task_from);
+                free(task_to);
+        }
+}
+
+static void draw_cpu_usage(void)
+{
+        struct per_pid *p;
+        struct per_pidcomm *c;
+        struct cpu_sample *sample;
+        p = all_data;
+        while (p) {
+                c = p->all;
+                while (c) {
+                        sample = c->samples;
+                        while (sample) {
+                                if (sample->type == TYPE_RUNNING)
+                                        svg_process(sample->cpu, sample->start_time, sample->end_time, "sample", c->comm);
+
+                                sample = sample->next;
+                        }
+                        c = c->next;
+                }
+                p = p->next;
+        }
+}
+
+static void draw_process_bars(void)
+{
+        struct per_pid *p;
+        struct per_pidcomm *c;
+        struct cpu_sample *sample;
+        int Y = 0;
+
+        Y = 2 * numcpus + 2;
+
+        p = all_data;
+        while (p) {
+                c = p->all;
+                while (c) {
+                        if (!c->display) {
+                                c->Y = 0;
+                                c = c->next;
+                                continue;
+                        }
+
+                        svg_box(Y, c->start_time, c->end_time, "process");
+                        sample = c->samples;
+                        while (sample) {
+                                if (sample->type == TYPE_RUNNING)
+                                        svg_sample(Y, sample->cpu, sample->start_time, sample->end_time);
+                                if (sample->type == TYPE_BLOCKED)
+                                        svg_box(Y, sample->start_time, sample->end_time, "blocked");
+                                if (sample->type == TYPE_WAITING)
+                                        svg_waiting(Y, sample->start_time, sample->end_time);
+                                sample = sample->next;
+                        }
+
+                        if (c->comm) {
+                                char comm[256];
+                                if (c->total_time > 5000000000) /* 5 seconds */
+                                        sprintf(comm, "%s:%i (%2.2fs)", c->comm, p->pid, c->total_time / 1000000000.0);
+                                else
+                                        sprintf(comm, "%s:%i (%3.1fms)", c->comm, p->pid, c->total_time / 1000000.0);
+
+                                svg_text(Y, c->start_time, comm);
+                        }
+                        c->Y = Y;
+                        Y++;
+                        c = c->next;
+                }
+                p = p->next;
+        }
+}
+
+static void add_process_filter(const char *string)
+{
+        struct process_filter *filt;
+        int pid;
+
+        pid = strtoull(string, NULL, 10);
+        filt = malloc(sizeof(struct process_filter));
+        if (!filt)
+                return;
+
+        filt->name = strdup(string);
+        filt->pid  = pid;
+        filt->next = process_filter;
+
+        process_filter = filt;
+}
+
+static int passes_filter(struct per_pid *p, struct per_pidcomm *c)
+{
+        struct process_filter *filt;
+        if (!process_filter)
+                return 1;
+
+        filt = process_filter;
+        while (filt) {
+                if (filt->pid && p->pid == filt->pid)
+                        return 1;
+                if (strcmp(filt->name, c->comm) == 0)
+                        return 1;
+                filt = filt->next;
+        }
+        return 0;
+}
+
+static int determine_display_tasks_filtered(void)
+{
+        struct per_pid *p;
+        struct per_pidcomm *c;
+        int count = 0;
+
+        p = all_data;
+        while (p) {
+                p->display = 0;
+                if (p->start_time == 1)
+                        p->start_time = first_time;
+
+                /* no exit marker, task kept running to the end */
+                if (p->end_time == 0)
+                        p->end_time = last_time;
+
+                c = p->all;
+
+                while (c) {
+                        c->display = 0;
+
+                        if (c->start_time == 1)
+                                c->start_time = first_time;
+
+                        if (passes_filter(p, c)) {
+                                c->display = 1;
+                                p->display = 1;
+                                count++;
+                        }
+
+                        if (c->end_time == 0)
+                                c->end_time = last_time;
+
+                        c = c->next;
+                }
+                p = p->next;
+        }
+        return count;
+}
+
+static int determine_display_tasks(u64 threshold)
+{
+        struct per_pid *p;
+        struct per_pidcomm *c;
+        int count = 0;
+
+        if (process_filter)
+                return determine_display_tasks_filtered();
+
+        p = all_data;
+        while (p) {
+                p->display = 0;
+                if (p->start_time == 1)
+                        p->start_time = first_time;
+
+                /* no exit marker, task kept running to the end */
+                if (p->end_time == 0)
+                        p->end_time = last_time;
+                if (p->total_time >= threshold && !power_only)
+                        p->display = 1;
+
+                c = p->all;
+
+                while (c) {
+                        c->display = 0;
+
+                        if (c->start_time == 1)
+                                c->start_time = first_time;
+
+                        if (c->total_time >= threshold && !power_only) {
+                                c->display = 1;
+                                count++;
+                        }
+
+                        if (c->end_time == 0)
+                                c->end_time = last_time;
+
+                        c = c->next;
+                }
+                p = p->next;
+        }
+        return count;
+}
+
+
+
+#define TIME_THRESH 10000000
+
+static void write_svg_file(const char *filename)
+{
+        u64 i;
+        int count;
+
+        numcpus++;
+
+
+        count = determine_display_tasks(TIME_THRESH);
+
+        /* We'd like to show at least 15 tasks; be less picky if we have fewer */
+        if (count < 15)
+                count = determine_display_tasks(TIME_THRESH / 10);
+
+        open_svg(filename, numcpus, count, first_time, last_time);
+
+        svg_time_grid();
+        svg_legenda();
+
+        for (i = 0; i < numcpus; i++)
+                svg_cpu_box(i, max_freq, turbo_frequency);
+
+        draw_cpu_usage();
+        draw_process_bars();
+        draw_c_p_states();
+        draw_wakeups();
+
+        svg_close();
+}
+
+static struct perf_event_ops event_ops = {
+        .comm                   = process_comm_event,
+        .fork                   = process_fork_event,
+        .exit                   = process_exit_event,
+        .sample                 = process_sample_event,
+        .ordered_samples        = true,
+};
+
+static int __cmd_timechart(void)
+{
+        struct perf_session *session = perf_session__new(input_name, O_RDONLY,
+                                                         0, false, &event_ops);
+        int ret = -EINVAL;
+
+        if (session == NULL)
+                return -ENOMEM;
+
+        if (!perf_session__has_traces(session, "timechart record"))
+                goto out_delete;
+
+        ret = perf_session__process_events(session, &event_ops);
+        if (ret)
+                goto out_delete;
+
+        end_sample_processing();
+
+        sort_pids();
+
+        write_svg_file(output_name);
+
+        pr_info("Written %2.1f seconds of trace to %s.\n",
+                (last_time - first_time) / 1000000000.0, output_name);
+out_delete:
+        perf_session__delete(session);
+        return ret;
+}
+
+static const char * const timechart_usage[] = {
+        "perf timechart [<options>] {record}",
+        NULL
+};
+
+#ifdef SUPPORT_OLD_POWER_EVENTS
+static const char * const record_old_args[] = {
+        "record",
+        "-a",
+        "-R",
+        "-f",
+        "-c", "1",
+        "-e", "power:power_start",
+        "-e", "power:power_end",
+        "-e", "power:power_frequency",
+        "-e", "sched:sched_wakeup",
+        "-e", "sched:sched_switch",
+};
+#endif
+
+static const char * const record_new_args[] = {
+        "record",
+        "-a",
+        "-R",
+        "-f",
+        "-c", "1",
+        "-e", "power:cpu_frequency",
+        "-e", "power:cpu_idle",
+        "-e", "sched:sched_wakeup",
+        "-e", "sched:sched_switch",
+};
+
+static int __cmd_record(int argc, const char **argv)
+{
+        unsigned int rec_argc, i, j;
+        const char **rec_argv;
+        const char * const *record_args = record_new_args;
+        unsigned int record_elems = ARRAY_SIZE(record_new_args);
+
+#ifdef SUPPORT_OLD_POWER_EVENTS
+        if (!is_valid_tracepoint("power:cpu_idle") &&
+            is_valid_tracepoint("power:power_start")) {
+                use_old_power_events = 1;
+                record_args = record_old_args;
+                record_elems = ARRAY_SIZE(record_old_args);
+        }
+#endif
+
+        rec_argc = record_elems + argc - 1;
+        rec_argv = calloc(rec_argc + 1, sizeof(char *));
+
+        if (rec_argv == NULL)
+                return -ENOMEM;
+
+        for (i = 0; i < record_elems; i++)
+                rec_argv[i] = strdup(record_args[i]);
+
+        for (j = 1; j < (unsigned int)argc; j++, i++)
+                rec_argv[i] = argv[j];
+
+        return cmd_record(i, rec_argv, NULL);
+}
+
+static int
+parse_process(const struct option *opt __used, const char *arg, int __used unset)
+{
+        if (arg)
+                add_process_filter(arg);
+        return 0;
+}
+
+static const struct option options[] = {
+        OPT_STRING('i', "input", &input_name, "file",
+                    "input file name"),
+        OPT_STRING('o', "output", &output_name, "file",
+                    "output file name"),
+        OPT_INTEGER('w', "width", &svg_page_width,
+                    "page width"),
+        OPT_BOOLEAN('P', "power-only", &power_only,
+                    "output power data only"),
+        OPT_CALLBACK('p', "process", NULL, "process",
+                      "process selector. Pass a pid or process name.",
+                       parse_process),
+        OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory",
+                    "Look for files with symbols relative to this directory"),
+        OPT_END()
+};
+
+
+int cmd_timechart(int argc, const char **argv, const char *prefix __used)
+{
+        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)
+                usage_with_options(timechart_usage, options);
+
+        setup_pager();
+
+        return __cmd_timechart();
+}
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index 5352b5e352e..6ce4042421b 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -20,17 +20,26 @@
 
 #include "perf.h"
 
-#include "util/symbol.h"
 #include "util/color.h"
+#include "util/evsel.h"
+#include "util/session.h"
+#include "util/symbol.h"
+#include "util/thread.h"
 #include "util/util.h"
-#include "util/rbtree.h"
+#include <linux/rbtree.h>
 #include "util/parse-options.h"
 #include "util/parse-events.h"
+#include "util/cpumap.h"
+#include "util/xyarray.h"
+
+#include "util/debug.h"
 
 #include <assert.h>
 #include <fcntl.h>
 
 #include <stdio.h>
+#include <termios.h>
+#include <unistd.h>
 
 #include <errno.h>
 #include <time.h>
@@ -48,55 +57,346 @@
 #include <linux/unistd.h>
 #include <linux/types.h>
 
-static int                      fd[MAX_NR_CPUS][MAX_COUNTERS];
+#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y))
 
-static int                      system_wide                     =  0;
+static bool                     system_wide                     =  false;
 
-static int                      default_interval                = 100000;
+static int                      default_interval                =      0;
 
-static u64                      count_filter                    =  5;
-static int                      print_entries                   = 15;
+static int                      count_filter                    =      5;
+static int                      print_entries;
 
-static int                      target_pid                      = -1;
-static int                      profile_cpu                     = -1;
-static int                      nr_cpus                         =  0;
-static unsigned int             realtime_prio                   =  0;
-static int                      group                           =  0;
+static int                      target_pid                      =     -1;
+static int                      target_tid                      =     -1;
+static struct thread_map        *threads;
+static bool                     inherit                         =  false;
+static struct cpu_map           *cpus;
+static int                      realtime_prio                   =      0;
+static bool                     group                           =  false;
 static unsigned int             page_size;
-static unsigned int             mmap_pages                      = 16;
-static int                      freq                            =  0;
-static int                      verbose                         =  0;
+static unsigned int             mmap_pages                      =     16;
+static int                      freq                            =   1000; /* 1 KHz */
+
+static int                      delay_secs                      =      2;
+static bool                     zero                            =  false;
+static bool                     dump_symtab                     =  false;
+
+static bool                     hide_kernel_symbols             =  false;
+static bool                     hide_user_symbols               =  false;
+static struct winsize           winsize;
 
-static char                     *sym_filter;
-static unsigned long            filter_start;
-static unsigned long            filter_end;
+/*
+ * Source
+ */
+
+struct source_line {
+        u64                     eip;
+        unsigned long           count[MAX_COUNTERS];
+        char                    *line;
+        struct source_line      *next;
+};
 
-static int                      delay_secs                      =  2;
-static int                      zero;
-static int                      dump_symtab;
+static const char               *sym_filter                     =   NULL;
+struct sym_entry                *sym_filter_entry               =   NULL;
+struct sym_entry                *sym_filter_entry_sched         =   NULL;
+static int                      sym_pcnt_filter                 =      5;
+static int                      sym_counter                     =      0;
+static struct perf_evsel        *sym_evsel                      =   NULL;
+static int                      display_weighted                =     -1;
+static const char               *cpu_list;
 
 /*
  * Symbols
  */
 
-static u64                      min_ip;
-static u64                      max_ip = -1ll;
+struct sym_entry_source {
+        struct source_line      *source;
+        struct source_line      *lines;
+        struct source_line      **lines_tail;
+        pthread_mutex_t         lock;
+};
 
 struct sym_entry {
         struct rb_node          rb_node;
         struct list_head        node;
-        unsigned long           count[MAX_COUNTERS];
         unsigned long           snap_count;
         double                  weight;
         int                     skip;
+        u16                     name_len;
+        u8                      origin;
+        struct map              *map;
+        struct sym_entry_source *src;
+        unsigned long           count[0];
 };
 
-struct sym_entry                *sym_filter_entry;
+/*
+ * Source functions
+ */
 
-struct dso                      *kernel_dso;
+static inline struct symbol *sym_entry__symbol(struct sym_entry *self)
+{
+       return ((void *)self) + symbol_conf.priv_size;
+}
+
+void get_term_dimensions(struct winsize *ws)
+{
+        char *s = getenv("LINES");
+
+        if (s != NULL) {
+                ws->ws_row = atoi(s);
+                s = getenv("COLUMNS");
+                if (s != NULL) {
+                        ws->ws_col = atoi(s);
+                        if (ws->ws_row && ws->ws_col)
+                                return;
+                }
+        }
+#ifdef TIOCGWINSZ
+        if (ioctl(1, TIOCGWINSZ, ws) == 0 &&
+            ws->ws_row && ws->ws_col)
+                return;
+#endif
+        ws->ws_row = 25;
+        ws->ws_col = 80;
+}
+
+static void update_print_entries(struct winsize *ws)
+{
+        print_entries = ws->ws_row;
+
+        if (print_entries > 9)
+                print_entries -= 9;
+}
+
+static void sig_winch_handler(int sig __used)
+{
+        get_term_dimensions(&winsize);
+        update_print_entries(&winsize);
+}
+
+static int parse_source(struct sym_entry *syme)
+{
+        struct symbol *sym;
+        struct sym_entry_source *source;
+        struct map *map;
+        FILE *file;
+        char command[PATH_MAX*2];
+        const char *path;
+        u64 len;
+
+        if (!syme)
+                return -1;
+
+        sym = sym_entry__symbol(syme);
+        map = syme->map;
+
+        /*
+         * We can't annotate with just /proc/kallsyms
+         */
+        if (map->dso->origin == DSO__ORIG_KERNEL)
+                return -1;
+
+        if (syme->src == NULL) {
+                syme->src = zalloc(sizeof(*source));
+                if (syme->src == NULL)
+                        return -1;
+                pthread_mutex_init(&syme->src->lock, NULL);
+        }
+
+        source = syme->src;
+
+        if (source->lines) {
+                pthread_mutex_lock(&source->lock);
+                goto out_assign;
+        }
+        path = map->dso->long_name;
+
+        len = sym->end - sym->start;
+
+        sprintf(command,
+                "objdump --start-address=%#0*Lx --stop-address=%#0*Lx -dS %s",
+                BITS_PER_LONG / 4, map__rip_2objdump(map, sym->start),
+                BITS_PER_LONG / 4, map__rip_2objdump(map, sym->end), path);
+
+        file = popen(command, "r");
+        if (!file)
+                return -1;
+
+        pthread_mutex_lock(&source->lock);
+        source->lines_tail = &source->lines;
+        while (!feof(file)) {
+                struct source_line *src;
+                size_t dummy = 0;
+                char *c, *sep;
+
+                src = malloc(sizeof(struct source_line));
+                assert(src != NULL);
+                memset(src, 0, sizeof(struct source_line));
+
+                if (getline(&src->line, &dummy, file) < 0)
+                        break;
+                if (!src->line)
+                        break;
+
+                c = strchr(src->line, '\n');
+                if (c)
+                        *c = 0;
+
+                src->next = NULL;
+                *source->lines_tail = src;
+                source->lines_tail = &src->next;
+
+                src->eip = strtoull(src->line, &sep, 16);
+                if (*sep == ':')
+                        src->eip = map__objdump_2ip(map, src->eip);
+                else /* this line has no ip info (e.g. source line) */
+                        src->eip = 0;
+        }
+        pclose(file);
+out_assign:
+        sym_filter_entry = syme;
+        pthread_mutex_unlock(&source->lock);
+        return 0;
+}
+
+static void __zero_source_counters(struct sym_entry *syme)
+{
+        int i;
+        struct source_line *line;
+
+        line = syme->src->lines;
+        while (line) {
+                for (i = 0; i < nr_counters; i++)
+                        line->count[i] = 0;
+                line = line->next;
+        }
+}
+
+static void record_precise_ip(struct sym_entry *syme, int counter, u64 ip)
+{
+        struct source_line *line;
+
+        if (syme != sym_filter_entry)
+                return;
+
+        if (pthread_mutex_trylock(&syme->src->lock))
+                return;
+
+        if (syme->src == NULL || syme->src->source == NULL)
+                goto out_unlock;
+
+        for (line = syme->src->lines; line; line = line->next) {
+                /* skip lines without IP info */
+                if (line->eip == 0)
+                        continue;
+                if (line->eip == ip) {
+                        line->count[counter]++;
+                        break;
+                }
+                if (line->eip > ip)
+                        break;
+        }
+out_unlock:
+        pthread_mutex_unlock(&syme->src->lock);
+}
+
+#define PATTERN_LEN             (BITS_PER_LONG / 4 + 2)
+
+static void lookup_sym_source(struct sym_entry *syme)
+{
+        struct symbol *symbol = sym_entry__symbol(syme);
+        struct source_line *line;
+        char pattern[PATTERN_LEN + 1];
+
+        sprintf(pattern, "%0*Lx <", BITS_PER_LONG / 4,
+                map__rip_2objdump(syme->map, symbol->start));
+
+        pthread_mutex_lock(&syme->src->lock);
+        for (line = syme->src->lines; line; line = line->next) {
+                if (memcmp(line->line, pattern, PATTERN_LEN) == 0) {
+                        syme->src->source = line;
+                        break;
+                }
+        }
+        pthread_mutex_unlock(&syme->src->lock);
+}
+
+static void show_lines(struct source_line *queue, int count, int total)
+{
+        int i;
+        struct source_line *line;
+
+        line = queue;
+        for (i = 0; i < count; i++) {
+                float pcnt = 100.0*(float)line->count[sym_counter]/(float)total;
+
+                printf("%8li %4.1f%%\t%s\n", line->count[sym_counter], pcnt, line->line);
+                line = line->next;
+        }
+}
+
+#define TRACE_COUNT     3
+
+static void show_details(struct sym_entry *syme)
+{
+        struct symbol *symbol;
+        struct source_line *line;
+        struct source_line *line_queue = NULL;
+        int displayed = 0;
+        int line_queue_count = 0, total = 0, more = 0;
+
+        if (!syme)
+                return;
+
+        if (!syme->src->source)
+                lookup_sym_source(syme);
+
+        if (!syme->src->source)
+                return;
+
+        symbol = sym_entry__symbol(syme);
+        printf("Showing %s for %s\n", event_name(sym_evsel), symbol->name);
+        printf("  Events  Pcnt (>=%d%%)\n", sym_pcnt_filter);
+
+        pthread_mutex_lock(&syme->src->lock);
+        line = syme->src->source;
+        while (line) {
+                total += line->count[sym_counter];
+                line = line->next;
+        }
+
+        line = syme->src->source;
+        while (line) {
+                float pcnt = 0.0;
+
+                if (!line_queue_count)
+                        line_queue = line;
+                line_queue_count++;
+
+                if (line->count[sym_counter])
+                        pcnt = 100.0 * line->count[sym_counter] / (float)total;
+                if (pcnt >= (float)sym_pcnt_filter) {
+                        if (displayed <= print_entries)
+                                show_lines(line_queue, line_queue_count, total);
+                        else more++;
+                        displayed += line_queue_count;
+                        line_queue_count = 0;
+                        line_queue = NULL;
+                } else if (line_queue_count > TRACE_COUNT) {
+                        line_queue = line_queue->next;
+                        line_queue_count--;
+                }
+
+                line->count[sym_counter] = zero ? 0 : line->count[sym_counter] * 7 / 8;
+                line = line->next;
+        }
+        pthread_mutex_unlock(&syme->src->lock);
+        if (more)
+                printf("%d lines not displayed, maybe increase display entries [e]\n", more);
+}
 
 /*
- * Symbols will be added here in record_ip and will get out
+ * Symbols will be added here in event__process_sample and will get out
  * after decayed.
  */
 static LIST_HEAD(active_symbols);
@@ -110,6 +410,9 @@ static double sym_weight(const struct sym_entry *sym)
         double weight = sym->snap_count;
         int counter;
 
+        if (!display_weighted)
+                return weight;
+
         for (counter = 1; counter < nr_counters-1; counter++)
                 weight *= sym->count[counter];
 
@@ -119,7 +422,9 @@ static double sym_weight(const struct sym_entry *sym)
 }
 
 static long                     samples;
-static long                     userspace_samples;
+static long                     kernel_samples, us_samples;
+static long                     exact_samples;
+static long                     guest_us_samples, guest_kernel_samples;
 static const char               CONSOLE_CLEAR[] = "";
 
 static void __list_insert_active_sym(struct sym_entry *syme)
@@ -157,15 +462,23 @@ static void rb_insert_active_sym(struct rb_root *tree, struct sym_entry *se)
 static void print_sym_table(void)
 {
         int printed = 0, j;
-        int counter;
+        struct perf_evsel *counter;
+        int snap = !display_weighted ? sym_counter : 0;
         float samples_per_sec = samples/delay_secs;
-        float ksamples_per_sec = (samples-userspace_samples)/delay_secs;
+        float ksamples_per_sec = kernel_samples/delay_secs;
+        float us_samples_per_sec = (us_samples)/delay_secs;
+        float guest_kernel_samples_per_sec = (guest_kernel_samples)/delay_secs;
+        float guest_us_samples_per_sec = (guest_us_samples)/delay_secs;
+        float esamples_percent = (100.0*exact_samples)/samples;
         float sum_ksamples = 0.0;
         struct sym_entry *syme, *n;
         struct rb_root tmp = RB_ROOT;
         struct rb_node *nd;
+        int sym_width = 0, dso_width = 0, dso_short_width = 0;
+        const int win_width = winsize.ws_col - 1;
 
-        samples = userspace_samples = 0;
+        samples = us_samples = kernel_samples = exact_samples = 0;
+        guest_kernel_samples = guest_us_samples = 0;
 
         /* Sort the active symbols */
         pthread_mutex_lock(&active_symbols_lock);
@@ -173,8 +486,16 @@ static void print_sym_table(void)
         pthread_mutex_unlock(&active_symbols_lock);
 
         list_for_each_entry_safe_from(syme, n, &active_symbols, node) {
-                syme->snap_count = syme->count[0];
+                syme->snap_count = syme->count[snap];
                 if (syme->snap_count != 0) {
+
+                        if ((hide_user_symbols &&
+                             syme->origin == PERF_RECORD_MISC_USER) ||
+                            (hide_kernel_symbols &&
+                             syme->origin == PERF_RECORD_MISC_KERNEL)) {
+                                list_remove_active_sym(syme);
+                                continue;
+                        }
                         syme->weight = sym_weight(syme);
                         rb_insert_active_sym(&tmp, syme);
                         sum_ksamples += syme->snap_count;
@@ -187,22 +508,46 @@ static void print_sym_table(void)
 
         puts(CONSOLE_CLEAR);
 
-        printf(
-"------------------------------------------------------------------------------\n");
-        printf( "   PerfTop:%8.0f irqs/sec  kernel:%4.1f%% [",
-                samples_per_sec,
-                100.0 - (100.0*((samples_per_sec-ksamples_per_sec)/samples_per_sec)));
+        printf("%-*.*s\n", win_width, win_width, graph_dotted_line);
+        if (!perf_guest) {
+                printf("   PerfTop:%8.0f irqs/sec  kernel:%4.1f%%"
+                        "  exact: %4.1f%% [",
+                        samples_per_sec,
+                        100.0 - (100.0 * ((samples_per_sec - ksamples_per_sec) /
+                                         samples_per_sec)),
+                        esamples_percent);
+        } else {
+                printf("   PerfTop:%8.0f irqs/sec  kernel:%4.1f%% us:%4.1f%%"
+                        " guest kernel:%4.1f%% guest us:%4.1f%%"
+                        " exact: %4.1f%% [",
+                        samples_per_sec,
+                        100.0 - (100.0 * ((samples_per_sec-ksamples_per_sec) /
+                                          samples_per_sec)),
+                        100.0 - (100.0 * ((samples_per_sec-us_samples_per_sec) /
+                                          samples_per_sec)),
+                        100.0 - (100.0 * ((samples_per_sec -
+                                                guest_kernel_samples_per_sec) /
+                                          samples_per_sec)),
+                        100.0 - (100.0 * ((samples_per_sec -
+                                           guest_us_samples_per_sec) /
+                                          samples_per_sec)),
+                        esamples_percent);
+        }
 
-        if (nr_counters == 1) {
-                printf("%Ld", (u64)attrs[0].sample_period);
+        if (nr_counters == 1 || !display_weighted) {
+                struct perf_evsel *first;
+                first = list_entry(evsel_list.next, struct perf_evsel, node);
+                printf("%Ld", first->attr.sample_period);
                 if (freq)
                         printf("Hz ");
                 else
                         printf(" ");
         }
 
-        for (counter = 0; counter < nr_counters; counter++) {
-                if (counter)
+        if (!display_weighted)
+                printf("%s", event_name(sym_evsel));
+        else list_for_each_entry(counter, &evsel_list, node) {
+                if (counter->idx)
                         printf("/");
 
                 printf("%s", event_name(counter));
@@ -212,85 +557,401 @@ static void print_sym_table(void)
 
         if (target_pid != -1)
                 printf(" (target_pid: %d", target_pid);
+        else if (target_tid != -1)
+                printf(" (target_tid: %d", target_tid);
         else
                 printf(" (all");
 
-        if (profile_cpu != -1)
-                printf(", cpu: %d)\n", profile_cpu);
+        if (cpu_list)
+                printf(", CPU%s: %s)\n", cpus->nr > 1 ? "s" : "", cpu_list);
         else {
-                if (target_pid != -1)
+                if (target_tid != -1)
                         printf(")\n");
                 else
-                        printf(", %d CPUs)\n", nr_cpus);
+                        printf(", %d CPU%s)\n", cpus->nr, cpus->nr > 1 ? "s" : "");
+        }
+
+        printf("%-*.*s\n", win_width, win_width, graph_dotted_line);
+
+        if (sym_filter_entry) {
+                show_details(sym_filter_entry);
+                return;
         }
 
-        printf("------------------------------------------------------------------------------\n\n");
+        /*
+         * Find the longest symbol name that will be displayed
+         */
+        for (nd = rb_first(&tmp); nd; nd = rb_next(nd)) {
+                syme = rb_entry(nd, struct sym_entry, rb_node);
+                if (++printed > print_entries ||
+                    (int)syme->snap_count < count_filter)
+                        continue;
+
+                if (syme->map->dso->long_name_len > dso_width)
+                        dso_width = syme->map->dso->long_name_len;
+
+                if (syme->map->dso->short_name_len > dso_short_width)
+                        dso_short_width = syme->map->dso->short_name_len;
+
+                if (syme->name_len > sym_width)
+                        sym_width = syme->name_len;
+        }
+
+        printed = 0;
 
+        if (sym_width + dso_width > winsize.ws_col - 29) {
+                dso_width = dso_short_width;
+                if (sym_width + dso_width > winsize.ws_col - 29)
+                        sym_width = winsize.ws_col - dso_width - 29;
+        }
+        putchar('\n');
         if (nr_counters == 1)
-                printf("             samples    pcnt");
+                printf("             samples  pcnt");
         else
-                printf("  weight     samples    pcnt");
-
-        printf("         RIP          kernel function\n"
-                       "  ______     _______   _____   ________________   _______________\n\n"
-        );
+                printf("   weight    samples  pcnt");
+
+        if (verbose)
+                printf("         RIP       ");
+        printf(" %-*.*s DSO\n", sym_width, sym_width, "function");
+        printf("   %s    _______ _____",
+               nr_counters == 1 ? "      " : "______");
+        if (verbose)
+                printf(" ________________");
+        printf(" %-*.*s", sym_width, sym_width, graph_line);
+        printf(" %-*.*s", dso_width, dso_width, graph_line);
+        puts("\n");
 
         for (nd = rb_first(&tmp); nd; nd = rb_next(nd)) {
-                struct sym_entry *syme = rb_entry(nd, struct sym_entry, rb_node);
-                struct symbol *sym = (struct symbol *)(syme + 1);
-                char *color = PERF_COLOR_NORMAL;
+                struct symbol *sym;
                 double pcnt;
 
-                if (++printed > print_entries || syme->snap_count < count_filter)
+                syme = rb_entry(nd, struct sym_entry, rb_node);
+                sym = sym_entry__symbol(syme);
+                if (++printed > print_entries || (int)syme->snap_count < count_filter)
                         continue;
 
                 pcnt = 100.0 - (100.0 * ((sum_ksamples - syme->snap_count) /
                                          sum_ksamples));
 
-                /*
-                 * We color high-overhead entries in red, mid-overhead
-                 * entries in green - and keep the low overhead places
-                 * normal:
-                 */
-                if (pcnt >= 5.0) {
-                        color = PERF_COLOR_RED;
-                } else {
-                        if (pcnt >= 0.5)
-                                color = PERF_COLOR_GREEN;
+                if (nr_counters == 1 || !display_weighted)
+                        printf("%20.2f ", syme->weight);
+                else
+                        printf("%9.1f %10ld ", syme->weight, syme->snap_count);
+
+                percent_color_fprintf(stdout, "%4.1f%%", pcnt);
+                if (verbose)
+                        printf(" %016llx", sym->start);
+                printf(" %-*.*s", sym_width, sym_width, sym->name);
+                printf(" %-*.*s\n", dso_width, dso_width,
+                       dso_width >= syme->map->dso->long_name_len ?
+                                        syme->map->dso->long_name :
+                                        syme->map->dso->short_name);
+        }
+}
+
+static void prompt_integer(int *target, const char *msg)
+{
+        char *buf = malloc(0), *p;
+        size_t dummy = 0;
+        int tmp;
+
+        fprintf(stdout, "\n%s: ", msg);
+        if (getline(&buf, &dummy, stdin) < 0)
+                return;
+
+        p = strchr(buf, '\n');
+        if (p)
+                *p = 0;
+
+        p = buf;
+        while(*p) {
+                if (!isdigit(*p))
+                        goto out_free;
+                p++;
+        }
+        tmp = strtoul(buf, NULL, 10);
+        *target = tmp;
+out_free:
+        free(buf);
+}
+
+static void prompt_percent(int *target, const char *msg)
+{
+        int tmp = 0;
+
+        prompt_integer(&tmp, msg);
+        if (tmp >= 0 && tmp <= 100)
+                *target = tmp;
+}
+
+static void prompt_symbol(struct sym_entry **target, const char *msg)
+{
+        char *buf = malloc(0), *p;
+        struct sym_entry *syme = *target, *n, *found = NULL;
+        size_t dummy = 0;
+
+        /* zero counters of active symbol */
+        if (syme) {
+                pthread_mutex_lock(&syme->src->lock);
+                __zero_source_counters(syme);
+                *target = NULL;
+                pthread_mutex_unlock(&syme->src->lock);
+        }
+
+        fprintf(stdout, "\n%s: ", msg);
+        if (getline(&buf, &dummy, stdin) < 0)
+                goto out_free;
+
+        p = strchr(buf, '\n');
+        if (p)
+                *p = 0;
+
+        pthread_mutex_lock(&active_symbols_lock);
+        syme = list_entry(active_symbols.next, struct sym_entry, node);
+        pthread_mutex_unlock(&active_symbols_lock);
+
+        list_for_each_entry_safe_from(syme, n, &active_symbols, node) {
+                struct symbol *sym = sym_entry__symbol(syme);
+
+                if (!strcmp(buf, sym->name)) {
+                        found = syme;
+                        break;
                 }
+        }
 
-                if (nr_counters == 1)
-                        printf("%20.2f - ", syme->weight);
-                else
-                        printf("%9.1f %10ld - ", syme->weight, syme->snap_count);
+        if (!found) {
+                fprintf(stderr, "Sorry, %s is not active.\n", buf);
+                sleep(1);
+                return;
+        } else
+                parse_source(found);
+
+out_free:
+        free(buf);
+}
+
+static void print_mapped_keys(void)
+{
+        char *name = NULL;
+
+        if (sym_filter_entry) {
+                struct symbol *sym = sym_entry__symbol(sym_filter_entry);
+                name = sym->name;
+        }
+
+        fprintf(stdout, "\nMapped keys:\n");
+        fprintf(stdout, "\t[d]     display refresh delay.             \t(%d)\n", delay_secs);
+        fprintf(stdout, "\t[e]     display entries (lines).           \t(%d)\n", print_entries);
 
-                color_fprintf(stdout, color, "%4.1f%%", pcnt);
-                printf(" - %016llx : %s\n", sym->start, sym->name);
+        if (nr_counters > 1)
+                fprintf(stdout, "\t[E]     active event counter.              \t(%s)\n", event_name(sym_evsel));
+
+        fprintf(stdout, "\t[f]     profile display filter (count).    \t(%d)\n", count_filter);
+
+        fprintf(stdout, "\t[F]     annotate display filter (percent). \t(%d%%)\n", sym_pcnt_filter);
+        fprintf(stdout, "\t[s]     annotate symbol.                   \t(%s)\n", name?: "NULL");
+        fprintf(stdout, "\t[S]     stop annotation.\n");
+
+        if (nr_counters > 1)
+                fprintf(stdout, "\t[w]     toggle display weighted/count[E]r. \t(%d)\n", display_weighted ? 1 : 0);
+
+        fprintf(stdout,
+                "\t[K]     hide kernel_symbols symbols.     \t(%s)\n",
+                hide_kernel_symbols ? "yes" : "no");
+        fprintf(stdout,
+                "\t[U]     hide user symbols.               \t(%s)\n",
+                hide_user_symbols ? "yes" : "no");
+        fprintf(stdout, "\t[z]     toggle sample zeroing.             \t(%d)\n", zero ? 1 : 0);
+        fprintf(stdout, "\t[qQ]    quit.\n");
+}
+
+static int key_mapped(int c)
+{
+        switch (c) {
+                case 'd':
+                case 'e':
+                case 'f':
+                case 'z':
+                case 'q':
+                case 'Q':
+                case 'K':
+                case 'U':
+                case 'F':
+                case 's':
+                case 'S':
+                        return 1;
+                case 'E':
+                case 'w':
+                        return nr_counters > 1 ? 1 : 0;
+                default:
+                        break;
         }
+
+        return 0;
 }
 
-static void *display_thread(void *arg)
+static void handle_keypress(struct perf_session *session, int c)
 {
-        struct pollfd stdin_poll = { .fd = 0, .events = POLLIN };
-        int delay_msecs = delay_secs * 1000;
+        if (!key_mapped(c)) {
+                struct pollfd stdin_poll = { .fd = 0, .events = POLLIN };
+                struct termios tc, save;
+
+                print_mapped_keys();
+                fprintf(stdout, "\nEnter selection, or unmapped key to continue: ");
+                fflush(stdout);
+
+                tcgetattr(0, &save);
+                tc = save;
+                tc.c_lflag &= ~(ICANON | ECHO);
+                tc.c_cc[VMIN] = 0;
+                tc.c_cc[VTIME] = 0;
+                tcsetattr(0, TCSANOW, &tc);
+
+                poll(&stdin_poll, 1, -1);
+                c = getc(stdin);
+
+                tcsetattr(0, TCSAFLUSH, &save);
+                if (!key_mapped(c))
+                        return;
+        }
+
+        switch (c) {
+                case 'd':
+                        prompt_integer(&delay_secs, "Enter display delay");
+                        if (delay_secs < 1)
+                                delay_secs = 1;
+                        break;
+                case 'e':
+                        prompt_integer(&print_entries, "Enter display entries (lines)");
+                        if (print_entries == 0) {
+                                sig_winch_handler(SIGWINCH);
+                                signal(SIGWINCH, sig_winch_handler);
+                        } else
+                                signal(SIGWINCH, SIG_DFL);
+                        break;
+                case 'E':
+                        if (nr_counters > 1) {
+                                fprintf(stderr, "\nAvailable events:");
+
+                                list_for_each_entry(sym_evsel, &evsel_list, node)
+                                        fprintf(stderr, "\n\t%d %s", sym_evsel->idx, event_name(sym_evsel));
+
+                                prompt_integer(&sym_counter, "Enter details event counter");
+
+                                if (sym_counter >= nr_counters) {
+                                        sym_evsel = list_entry(evsel_list.next, struct perf_evsel, node);
+                                        sym_counter = 0;
+                                        fprintf(stderr, "Sorry, no such event, using %s.\n", event_name(sym_evsel));
+                                        sleep(1);
+                                        break;
+                                }
+                                list_for_each_entry(sym_evsel, &evsel_list, node)
+                                        if (sym_evsel->idx == sym_counter)
+                                                break;
+                        } else sym_counter = 0;
+                        break;
+                case 'f':
+                        prompt_integer(&count_filter, "Enter display event count filter");
+                        break;
+                case 'F':
+                        prompt_percent(&sym_pcnt_filter, "Enter details display event filter (percent)");
+                        break;
+                case 'K':
+                        hide_kernel_symbols = !hide_kernel_symbols;
+                        break;
+                case 'q':
+                case 'Q':
+                        printf("exiting.\n");
+                        if (dump_symtab)
+                                perf_session__fprintf_dsos(session, stderr);
+                        exit(0);
+                case 's':
+                        prompt_symbol(&sym_filter_entry, "Enter details symbol");
+                        break;
+                case 'S':
+                        if (!sym_filter_entry)
+                                break;
+                        else {
+                                struct sym_entry *syme = sym_filter_entry;
+
+                                pthread_mutex_lock(&syme->src->lock);
+                                sym_filter_entry = NULL;
+                                __zero_source_counters(syme);
+                                pthread_mutex_unlock(&syme->src->lock);
+                        }
+                        break;
+                case 'U':
+                        hide_user_symbols = !hide_user_symbols;
+                        break;
+                case 'w':
+                        display_weighted = ~display_weighted;
+                        break;
+                case 'z':
+                        zero = !zero;
+                        break;
+                default:
+                        break;
+        }
+}
 
-        printf("PerfTop refresh period: %d seconds\n", delay_secs);
+static void *display_thread(void *arg __used)
+{
+        struct pollfd stdin_poll = { .fd = 0, .events = POLLIN };
+        struct termios tc, save;
+        int delay_msecs, c;
+        struct perf_session *session = (struct perf_session *) arg;
+
+        tcgetattr(0, &save);
+        tc = save;
+        tc.c_lflag &= ~(ICANON | ECHO);
+        tc.c_cc[VMIN] = 0;
+        tc.c_cc[VTIME] = 0;
+
+repeat:
+        delay_msecs = delay_secs * 1000;
+        tcsetattr(0, TCSANOW, &tc);
+        /* trash return*/
+        getc(stdin);
 
         do {
                 print_sym_table();
         } while (!poll(&stdin_poll, 1, delay_msecs) == 1);
 
-        printf("key pressed - exiting.\n");
-        exit(0);
+        c = getc(stdin);
+        tcsetattr(0, TCSAFLUSH, &save);
+
+        handle_keypress(session, c);
+        goto repeat;
 
         return NULL;
 }
 
-static int symbol_filter(struct dso *self, struct symbol *sym)
+/* Tag samples to be skipped. */
+static const char *skip_symbols[] = {
+        "default_idle",
+        "cpu_idle",
+        "enter_idle",
+        "exit_idle",
+        "mwait_idle",
+        "mwait_idle_with_hints",
+        "poll_idle",
+        "ppc64_runlatch_off",
+        "pseries_dedicated_idle_sleep",
+        NULL
+};
+
+static int symbol_filter(struct map *map, struct symbol *sym)
 {
-        static int filter_match;
         struct sym_entry *syme;
         const char *name = sym->name;
+        int i;
+
+        /*
+         * ppc64 uses function descriptors and appends a '.' to the
+         * start of every instruction address. Remove it.
+         */
+        if (name[0] == '.')
+                name++;
 
         if (!strcmp(name, "_text") ||
             !strcmp(name, "_etext") ||
@@ -301,119 +962,157 @@ static int symbol_filter(struct dso *self, struct symbol *sym)
             strstr(name, "_text_end"))
                 return 1;
 
-        syme = dso__sym_priv(self, sym);
-        /* Tag samples to be skipped. */
-        if (!strcmp("default_idle", name) ||
-            !strcmp("cpu_idle", name) ||
-            !strcmp("enter_idle", name) ||
-            !strcmp("exit_idle", name) ||
-            !strcmp("mwait_idle", name))
-                syme->skip = 1;
-
-        if (filter_match == 1) {
-                filter_end = sym->start;
-                filter_match = -1;
-                if (filter_end - filter_start > 10000) {
-                        fprintf(stderr,
-                                "hm, too large filter symbol <%s> - skipping.\n",
-                                sym_filter);
-                        fprintf(stderr, "symbol filter start: %016lx\n",
-                                filter_start);
-                        fprintf(stderr, "                end: %016lx\n",
-                                filter_end);
-                        filter_end = filter_start = 0;
-                        sym_filter = NULL;
-                        sleep(1);
-                }
+        syme = symbol__priv(sym);
+        syme->map = map;
+        syme->src = NULL;
+
+        if (!sym_filter_entry && sym_filter && !strcmp(name, sym_filter)) {
+                /* schedule initial sym_filter_entry setup */
+                sym_filter_entry_sched = syme;
+                sym_filter = NULL;
         }
 
-        if (filter_match == 0 && sym_filter && !strcmp(name, sym_filter)) {
-                filter_match = 1;
-                filter_start = sym->start;
+        for (i = 0; skip_symbols[i]; i++) {
+                if (!strcmp(skip_symbols[i], name)) {
+                        syme->skip = 1;
+                        break;
+                }
         }
 
+        if (!syme->skip)
+                syme->name_len = strlen(sym->name);
 
         return 0;
 }
 
-static int parse_symbols(void)
+static void event__process_sample(const event_t *self,
+                                  struct sample_data *sample,
+                                  struct perf_session *session,
+                                  struct perf_evsel *evsel)
 {
-        struct rb_node *node;
-        struct symbol  *sym;
-
-        kernel_dso = dso__new("[kernel]", sizeof(struct sym_entry));
-        if (kernel_dso == NULL)
-                return -1;
-
-        if (dso__load_kernel(kernel_dso, NULL, symbol_filter, 1) != 0)
-                goto out_delete_dso;
-
-        node = rb_first(&kernel_dso->syms);
-        sym = rb_entry(node, struct symbol, rb_node);
-        min_ip = sym->start;
+        u64 ip = self->ip.ip;
+        struct sym_entry *syme;
+        struct addr_location al;
+        struct machine *machine;
+        u8 origin = self->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
 
-        node = rb_last(&kernel_dso->syms);
-        sym = rb_entry(node, struct symbol, rb_node);
-        max_ip = sym->end;
+        ++samples;
 
-        if (dump_symtab)
-                dso__fprintf(kernel_dso, stderr);
+        switch (origin) {
+        case PERF_RECORD_MISC_USER:
+                ++us_samples;
+                if (hide_user_symbols)
+                        return;
+                machine = perf_session__find_host_machine(session);
+                break;
+        case PERF_RECORD_MISC_KERNEL:
+                ++kernel_samples;
+                if (hide_kernel_symbols)
+                        return;
+                machine = perf_session__find_host_machine(session);
+                break;
+        case PERF_RECORD_MISC_GUEST_KERNEL:
+                ++guest_kernel_samples;
+                machine = perf_session__find_machine(session, self->ip.pid);
+                break;
+        case PERF_RECORD_MISC_GUEST_USER:
+                ++guest_us_samples;
+                /*
+                 * TODO: we don't process guest user from host side
+                 * except simple counting.
+                 */
+                return;
+        default:
+                return;
+        }
 
-        return 0;
+        if (!machine && perf_guest) {
+                pr_err("Can't find guest [%d]'s kernel information\n",
+                        self->ip.pid);
+                return;
+        }
 
-out_delete_dso:
-        dso__delete(kernel_dso);
-        kernel_dso = NULL;
-        return -1;
-}
+        if (self->header.misc & PERF_RECORD_MISC_EXACT_IP)
+                exact_samples++;
 
-#define TRACE_COUNT     3
+        if (event__preprocess_sample(self, session, &al, sample,
+                                     symbol_filter) < 0 ||
+            al.filtered)
+                return;
 
-/*
- * Binary search in the histogram table and record the hit:
- */
-static void record_ip(u64 ip, int counter)
-{
-        struct symbol *sym = dso__find_symbol(kernel_dso, ip);
+        if (al.sym == NULL) {
+                /*
+                 * As we do lazy loading of symtabs we only will know if the
+                 * specified vmlinux file is invalid when we actually have a
+                 * hit in kernel space and then try to load it. So if we get
+                 * here and there are _no_ symbols in the DSO backing the
+                 * kernel map, bail out.
+                 *
+                 * We may never get here, for instance, if we use -K/
+                 * --hide-kernel-symbols, even if the user specifies an
+                 * invalid --vmlinux ;-)
+                 */
+                if (al.map == machine->vmlinux_maps[MAP__FUNCTION] &&
+                    RB_EMPTY_ROOT(&al.map->dso->symbols[MAP__FUNCTION])) {
+                        pr_err("The %s file can't be used\n",
+                               symbol_conf.vmlinux_name);
+                        exit(1);
+                }
 
-        if (sym != NULL) {
-                struct sym_entry *syme = dso__sym_priv(kernel_dso, sym);
+                return;
+        }
 
-                if (!syme->skip) {
-                        syme->count[counter]++;
-                        pthread_mutex_lock(&active_symbols_lock);
-                        if (list_empty(&syme->node) || !syme->node.next)
-                                __list_insert_active_sym(syme);
-                        pthread_mutex_unlock(&active_symbols_lock);
-                        return;
+        /* let's see, whether we need to install initial sym_filter_entry */
+        if (sym_filter_entry_sched) {
+                sym_filter_entry = sym_filter_entry_sched;
+                sym_filter_entry_sched = NULL;
+                if (parse_source(sym_filter_entry) < 0) {
+                        struct symbol *sym = sym_entry__symbol(sym_filter_entry);
+
+                        pr_err("Can't annotate %s", sym->name);
+                        if (sym_filter_entry->map->dso->origin == DSO__ORIG_KERNEL) {
+                                pr_err(": No vmlinux file was found in the path:\n");
+                                machine__fprintf_vmlinux_path(machine, stderr);
+                        } else
+                                pr_err(".\n");
+                        exit(1);
                 }
         }
 
-        samples--;
-}
-
-static void process_event(u64 ip, int counter)
-{
-        samples++;
-
-        if (ip < min_ip || ip > max_ip) {
-                userspace_samples++;
-                return;
+        syme = symbol__priv(al.sym);
+        if (!syme->skip) {
+                syme->count[evsel->idx]++;
+                syme->origin = origin;
+                record_precise_ip(syme, evsel->idx, ip);
+                pthread_mutex_lock(&active_symbols_lock);
+                if (list_empty(&syme->node) || !syme->node.next)
+                        __list_insert_active_sym(syme);
+                pthread_mutex_unlock(&active_symbols_lock);
         }
-
-        record_ip(ip, counter);
 }
 
 struct mmap_data {
-        int                     counter;
         void                    *base;
-        unsigned int            mask;
+        int                     mask;
         unsigned int            prev;
 };
 
+static int perf_evsel__alloc_mmap_per_thread(struct perf_evsel *evsel,
+                                             int ncpus, int nthreads)
+{
+        evsel->priv = xyarray__new(ncpus, nthreads, sizeof(struct mmap_data));
+        return evsel->priv != NULL ? 0 : -ENOMEM;
+}
+
+static void perf_evsel__free_mmap(struct perf_evsel *evsel)
+{
+        xyarray__delete(evsel->priv);
+        evsel->priv = NULL;
+}
+
 static unsigned int mmap_read_head(struct mmap_data *md)
 {
-        struct perf_counter_mmap_page *pc = md->base;
+        struct perf_event_mmap_page *pc = md->base;
         int head;
 
         head = pc->data_head;
@@ -422,17 +1121,18 @@ static unsigned int mmap_read_head(struct mmap_data *md)
         return head;
 }
 
-struct timeval last_read, this_read;
-
-static void mmap_read_counter(struct mmap_data *md)
+static void perf_session__mmap_read_counter(struct perf_session *self,
+                                            struct perf_evsel *evsel,
+                                            int cpu, int thread_idx)
 {
+        struct xyarray *mmap_array = evsel->priv;
+        struct mmap_data *md = xyarray__entry(mmap_array, cpu, thread_idx);
         unsigned int head = mmap_read_head(md);
         unsigned int old = md->prev;
         unsigned char *data = md->base + page_size;
+        struct sample_data sample;
         int diff;
 
-        gettimeofday(&this_read, NULL);
-
         /*
          * If we're further behind than half the buffer, there's a chance
          * the writer will bite our tail and mess up the samples under us.
@@ -443,14 +1143,7 @@ static void mmap_read_counter(struct mmap_data *md)
          */
         diff = head - old;
         if (diff > md->mask / 2 || diff < 0) {
-                struct timeval iv;
-                unsigned long msecs;
-
-                timersub(&this_read, &last_read, &iv);
-                msecs = iv.tv_sec*1000 + iv.tv_usec/1000;
-
-                fprintf(stderr, "WARNING: failed to keep up with mmap data."
-                                "  Last read %lu msecs ago.\n", msecs);
+                fprintf(stderr, "WARNING: failed to keep up with mmap data.\n");
 
                 /*
                  * head points to a known good entry, start there.
@@ -458,29 +1151,7 @@ static void mmap_read_counter(struct mmap_data *md)
                 old = head;
         }
 
-        last_read = this_read;
-
         for (; old != head;) {
-                struct ip_event {
-                        struct perf_event_header header;
-                        u64 ip;
-                        u32 pid, target_pid;
-                };
-                struct mmap_event {
-                        struct perf_event_header header;
-                        u32 pid, target_pid;
-                        u64 start;
-                        u64 len;
-                        u64 pgoff;
-                        char filename[PATH_MAX];
-                };
-
-                typedef union event_union {
-                        struct perf_event_header header;
-                        struct ip_event ip;
-                        struct mmap_event mmap;
-                } event_t;
-
                 event_t *event = (event_t *)&data[old & md->mask];
 
                 event_t event_copy;
@@ -507,116 +1178,151 @@ static void mmap_read_counter(struct mmap_data *md)
                         event = &event_copy;
                 }
 
+                event__parse_sample(event, self, &sample);
+                if (event->header.type == PERF_RECORD_SAMPLE)
+                        event__process_sample(event, &sample, self, evsel);
+                else
+                        event__process(event, &sample, self);
                 old += size;
-
-                if (event->header.misc & PERF_EVENT_MISC_OVERFLOW) {
-                        if (event->header.type & PERF_SAMPLE_IP)
-                                process_event(event->ip.ip, md->counter);
-                }
         }
 
         md->prev = old;
 }
 
-static struct pollfd event_array[MAX_NR_CPUS * MAX_COUNTERS];
-static struct mmap_data mmap_array[MAX_NR_CPUS][MAX_COUNTERS];
+static struct pollfd *event_array;
 
-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]);
+        struct perf_evsel *counter;
+        int i, thread_index;
+
+        for (i = 0; i < cpus->nr; i++) {
+                list_for_each_entry(counter, &evsel_list, node) {
+                        for (thread_index = 0;
+                                thread_index < threads->nr;
+                                thread_index++) {
+                                perf_session__mmap_read_counter(self,
+                                        counter, i, thread_index);
+                        }
+                }
         }
 }
 
 int nr_poll;
 int group_fd;
 
-static void start_counter(int i, int counter)
+static void start_counter(int i, struct perf_evsel *evsel)
 {
-        struct perf_counter_attr *attr;
-        unsigned int cpu;
+        struct xyarray *mmap_array = evsel->priv;
+        struct mmap_data *mm;
+        struct perf_event_attr *attr;
+        int cpu = -1;
+        int thread_index;
 
-        cpu = profile_cpu;
-        if (target_pid == -1 && profile_cpu == -1)
-                cpu = i;
+        if (target_tid == -1)
+                cpu = cpus->map[i];
 
-        attr = attrs + counter;
+        attr = &evsel->attr;
 
         attr->sample_type       = PERF_SAMPLE_IP | PERF_SAMPLE_TID;
-        attr->freq              = freq;
 
-try_again:
-        fd[i][counter] = sys_perf_counter_open(attr, target_pid, cpu, group_fd, 0);
+        if (freq) {
+                attr->sample_type       |= PERF_SAMPLE_PERIOD;
+                attr->freq              = 1;
+                attr->sample_freq       = freq;
+        }
+
+        attr->inherit           = (cpu < 0) && inherit;
+        attr->mmap              = 1;
 
-        if (fd[i][counter] < 0) {
-                int err = errno;
+        for (thread_index = 0; thread_index < threads->nr; thread_index++) {
+try_again:
+                FD(evsel, i, thread_index) = sys_perf_event_open(attr,
+                                threads->map[thread_index], cpu, group_fd, 0);
+
+                if (FD(evsel, i, thread_index) < 0) {
+                        int err = errno;
+
+                        if (err == EPERM || err == EACCES)
+                                die("Permission error - are you root?\n"
+                                        "\t Consider tweaking"
+                                        " /proc/sys/kernel/perf_event_paranoid.\n");
+                        if (err == ENOENT)
+                                die("%s event is not supported. ", event_name(evsel));
+                        /*
+                         * If it's cycles then fall back to hrtimer
+                         * based cpu-clock-tick sw counter, which
+                         * is always available even if no PMU support:
+                         */
+                        if (attr->type == PERF_TYPE_HARDWARE
+                                        && attr->config == PERF_COUNT_HW_CPU_CYCLES) {
+
+                                if (verbose)
+                                        warning(" ... trying to fall back to cpu-clock-ticks\n");
+
+                                attr->type = PERF_TYPE_SOFTWARE;
+                                attr->config = PERF_COUNT_SW_CPU_CLOCK;
+                                goto try_again;
+                        }
+                        printf("\n");
+                        error("sys_perf_event_open() syscall returned with %d (%s).  /bin/dmesg may provide additional information.\n",
+                                        FD(evsel, i, thread_index), strerror(err));
+                        die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
+                        exit(-1);
+                }
+                assert(FD(evsel, i, thread_index) >= 0);
+                fcntl(FD(evsel, i, thread_index), F_SETFL, O_NONBLOCK);
 
-                if (err == EPERM)
-                        die("No permission - are you root?\n");
                 /*
-                 * If it's cycles then fall back to hrtimer
-                 * based cpu-clock-tick sw counter, which
-                 * is always available even if no PMU support:
+                 * First counter acts as the group leader:
                  */
-                if (attr->type == PERF_TYPE_HARDWARE
-                        && attr->config == PERF_COUNT_HW_CPU_CYCLES) {
-
-                        if (verbose)
-                                warning(" ... trying to fall back to cpu-clock-ticks\n");
-
-                        attr->type = PERF_TYPE_SOFTWARE;
-                        attr->config = PERF_COUNT_SW_CPU_CLOCK;
-                        goto try_again;
-                }
-                printf("\n");
-                error("perfcounter syscall returned with %d (%s)\n",
-                        fd[i][counter], strerror(err));
-                die("No CONFIG_PERF_COUNTERS=y kernel support configured?\n");
-                exit(-1);
+                if (group && group_fd == -1)
+                        group_fd = FD(evsel, i, thread_index);
+
+                event_array[nr_poll].fd = FD(evsel, i, thread_index);
+                event_array[nr_poll].events = POLLIN;
+                nr_poll++;
+
+                mm = xyarray__entry(mmap_array, i, thread_index);
+                mm->prev = 0;
+                mm->mask = mmap_pages*page_size - 1;
+                mm->base = mmap(NULL, (mmap_pages+1)*page_size,
+                                PROT_READ, MAP_SHARED, FD(evsel, i, thread_index), 0);
+                if (mm->base == MAP_FAILED)
+                        die("failed to mmap with %d (%s)\n", errno, strerror(errno));
         }
-        assert(fd[i][counter] >= 0);
-        fcntl(fd[i][counter], F_SETFL, O_NONBLOCK);
-
-        /*
-         * First counter acts as the group leader:
-         */
-        if (group && group_fd == -1)
-                group_fd = fd[i][counter];
-
-        event_array[nr_poll].fd = fd[i][counter];
-        event_array[nr_poll].events = POLLIN;
-        nr_poll++;
-
-        mmap_array[i][counter].counter = counter;
-        mmap_array[i][counter].prev = 0;
-        mmap_array[i][counter].mask = mmap_pages*page_size - 1;
-        mmap_array[i][counter].base = mmap(NULL, (mmap_pages+1)*page_size,
-                        PROT_READ, MAP_SHARED, fd[i][counter], 0);
-        if (mmap_array[i][counter].base == MAP_FAILED)
-                die("failed to mmap with %d (%s)\n", errno, strerror(errno));
 }
 
 static int __cmd_top(void)
 {
         pthread_t thread;
-        int i, counter;
-        int ret;
+        struct perf_evsel *counter;
+        int i, 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, false, NULL);
+        if (session == NULL)
+                return -ENOMEM;
+
+        if (target_tid != -1)
+                event__synthesize_thread(target_tid, event__process, session);
+        else
+                event__synthesize_threads(event__process, session);
 
-        for (i = 0; i < nr_cpus; i++) {
+        for (i = 0; i < cpus->nr; i++) {
                 group_fd = -1;
-                for (counter = 0; counter < nr_counters; counter++)
+                list_for_each_entry(counter, &evsel_list, node)
                         start_counter(i, counter);
         }
 
         /* Wait for a minimal set of events before starting the snapshot */
-        poll(event_array, nr_poll, 100);
+        poll(&event_array[0], nr_poll, 100);
 
-        mmap_read();
+        perf_session__mmap_read(session);
 
-        if (pthread_create(&thread, NULL, display_thread, NULL)) {
+        if (pthread_create(&thread, NULL, display_thread, session)) {
                 printf("Could not create display thread.\n");
                 exit(-1);
         }
@@ -634,7 +1340,7 @@ 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);
@@ -655,13 +1361,18 @@ static const struct option options[] = {
         OPT_INTEGER('c', "count", &default_interval,
                     "event period to sample"),
         OPT_INTEGER('p', "pid", &target_pid,
-                    "profile events on existing pid"),
+                    "profile events on existing process id"),
+        OPT_INTEGER('t', "tid", &target_tid,
+                    "profile events on existing thread id"),
         OPT_BOOLEAN('a', "all-cpus", &system_wide,
                             "system-wide collection from all CPUs"),
-        OPT_INTEGER('C', "CPU", &profile_cpu,
-                    "CPU to profile on"),
-        OPT_INTEGER('m', "mmap-pages", &mmap_pages,
-                    "number of mmap data pages"),
+        OPT_STRING('C', "cpu", &cpu_list, "cpu",
+                    "list of cpus to monitor"),
+        OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
+                   "file", "vmlinux pathname"),
+        OPT_BOOLEAN('K', "hide_kernel_symbols", &hide_kernel_symbols,
+                    "hide kernel symbols"),
+        OPT_UINTEGER('m', "mmap-pages", &mmap_pages, "number of mmap data pages"),
         OPT_INTEGER('r', "realtime", &realtime_prio,
                     "collect data with this RT SCHED_FIFO priority"),
         OPT_INTEGER('d', "delay", &delay_secs,
@@ -672,22 +1383,27 @@ static const struct option options[] = {
                     "only display functions with more events than this"),
         OPT_BOOLEAN('g', "group", &group,
                             "put the counters into a counter group"),
-        OPT_STRING('s', "sym-filter", &sym_filter, "pattern",
-                    "only display symbols matchig this pattern"),
-        OPT_BOOLEAN('z', "zero", &group,
+        OPT_BOOLEAN('i', "inherit", &inherit,
+                    "child tasks inherit counters"),
+        OPT_STRING('s', "sym-annotate", &sym_filter, "symbol name",
+                    "symbol to annotate"),
+        OPT_BOOLEAN('z', "zero", &zero,
                     "zero history across updates"),
         OPT_INTEGER('F', "freq", &freq,
                     "profile at this frequency"),
         OPT_INTEGER('E', "entries", &print_entries,
                     "display this many functions"),
-        OPT_BOOLEAN('v', "verbose", &verbose,
+        OPT_BOOLEAN('U', "hide_user_symbols", &hide_user_symbols,
+                    "hide user symbols"),
+        OPT_INCR('v', "verbose", &verbose,
                     "be more verbose (show counter open errors, etc)"),
         OPT_END()
 };
 
-int cmd_top(int argc, const char **argv, const char *prefix)
+int cmd_top(int argc, const char **argv, const char *prefix __used)
 {
-        int counter;
+        struct perf_evsel *pos;
+        int status = -ENOMEM;
 
         page_size = sysconf(_SC_PAGE_SIZE);
 
@@ -695,42 +1411,85 @@ int cmd_top(int argc, const char **argv, const char *prefix)
         if (argc)
                 usage_with_options(top_usage, options);
 
-        if (freq) {
-                default_interval = freq;
-                freq = 1;
+        if (target_pid != -1)
+                target_tid = target_pid;
+
+        threads = thread_map__new(target_pid, target_tid);
+        if (threads == NULL) {
+                pr_err("Problems finding threads of monitor\n");
+                usage_with_options(top_usage, options);
         }
 
+        event_array = malloc((sizeof(struct pollfd) *
+                              MAX_NR_CPUS * MAX_COUNTERS * threads->nr));
+        if (!event_array)
+                return -ENOMEM;
+
         /* CPU and PID are mutually exclusive */
-        if (target_pid != -1 && profile_cpu != -1) {
+        if (target_tid > 0 && cpu_list) {
                 printf("WARNING: PID switch overriding CPU\n");
                 sleep(1);
-                profile_cpu = -1;
+                cpu_list = NULL;
         }
 
-        if (!nr_counters)
-                nr_counters = 1;
+        if (!nr_counters && perf_evsel_list__create_default() < 0) {
+                pr_err("Not enough memory for event selector list\n");
+                return -ENOMEM;
+        }
 
         if (delay_secs < 1)
                 delay_secs = 1;
 
-        parse_symbols();
-
         /*
-         * Fill in the ones not specifically initialized via -c:
+         * User specified count overrides default frequency.
          */
-        for (counter = 0; counter < nr_counters; counter++) {
-                if (attrs[counter].sample_period)
+        if (default_interval)
+                freq = 0;
+        else if (freq) {
+                default_interval = freq;
+        } else {
+                fprintf(stderr, "frequency and count are zero, aborting\n");
+                exit(EXIT_FAILURE);
+        }
+
+        if (target_tid != -1)
+                cpus = cpu_map__dummy_new();
+        else
+                cpus = cpu_map__new(cpu_list);
+
+        if (cpus == NULL)
+                usage_with_options(top_usage, options);
+
+        list_for_each_entry(pos, &evsel_list, node) {
+                if (perf_evsel__alloc_mmap_per_thread(pos, cpus->nr, threads->nr) < 0 ||
+                    perf_evsel__alloc_fd(pos, cpus->nr, threads->nr) < 0)
+                        goto out_free_fd;
+                /*
+                 * Fill in the ones not specifically initialized via -c:
+                 */
+                if (pos->attr.sample_period)
                         continue;
 
-                attrs[counter].sample_period = default_interval;
+                pos->attr.sample_period = default_interval;
         }
 
-        nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
-        assert(nr_cpus <= MAX_NR_CPUS);
-        assert(nr_cpus >= 0);
+        symbol_conf.priv_size = (sizeof(struct sym_entry) +
+                                 (nr_counters + 1) * sizeof(unsigned long));
+
+        symbol_conf.try_vmlinux_path = (symbol_conf.vmlinux_name == NULL);
+        if (symbol__init() < 0)
+                return -1;
+
+        get_term_dimensions(&winsize);
+        if (print_entries == 0) {
+                update_print_entries(&winsize);
+                signal(SIGWINCH, sig_winch_handler);
+        }
 
-        if (target_pid != -1 || profile_cpu != -1)
-                nr_cpus = 1;
+        status = __cmd_top();
+out_free_fd:
+        list_for_each_entry(pos, &evsel_list, node)
+                perf_evsel__free_mmap(pos);
 
-        return __cmd_top();
+        return status;
 }
diff --git a/tools/perf/builtin.h b/tools/perf/builtin.h
index 51d168230ee..c7798c7f24e 100644
--- a/tools/perf/builtin.h
+++ b/tools/perf/builtin.h
@@ -15,12 +15,25 @@ extern int read_line_with_nul(char *buf, int size, FILE *file);
 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_cache(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);
 extern int cmd_record(int argc, const char **argv, const char *prefix);
 extern int cmd_report(int argc, const char **argv, const char *prefix);
 extern int cmd_stat(int argc, const char **argv, const char *prefix);
+extern int cmd_timechart(int argc, const char **argv, const char *prefix);
 extern int cmd_top(int argc, const char **argv, const char *prefix);
+extern int cmd_script(int argc, const char **argv, const char *prefix);
 extern int cmd_version(int argc, const char **argv, const char *prefix);
-extern int cmd_list(int argc, const char **argv, const char *prefix);
+extern int cmd_probe(int argc, const char **argv, const char *prefix);
+extern int cmd_kmem(int argc, const char **argv, const char *prefix);
+extern int cmd_lock(int argc, const char **argv, const char *prefix);
+extern int cmd_kvm(int argc, const char **argv, const char *prefix);
+extern int cmd_test(int argc, const char **argv, const char *prefix);
+extern int cmd_inject(int argc, const char **argv, const char *prefix);
 
 #endif
diff --git a/tools/perf/command-list.txt b/tools/perf/command-list.txt
index eebce30afbc..16b5088cf8f 100644
--- a/tools/perf/command-list.txt
+++ b/tools/perf/command-list.txt
@@ -3,8 +3,22 @@
 # command name                  category [deprecated] [common]
 #
 perf-annotate                   mainporcelain common
+perf-archive                    mainporcelain common
+perf-bench                      mainporcelain common
+perf-buildid-cache              mainporcelain common
+perf-buildid-list               mainporcelain common
+perf-diff                       mainporcelain common
+perf-inject                     mainporcelain common
 perf-list                       mainporcelain common
+perf-sched                      mainporcelain common
 perf-record                     mainporcelain common
 perf-report                     mainporcelain common
 perf-stat                       mainporcelain common
+perf-timechart                  mainporcelain common
 perf-top                        mainporcelain common
+perf-script                     mainporcelain common
+perf-probe                      mainporcelain common
+perf-kmem                       mainporcelain common
+perf-lock                       mainporcelain common
+perf-kvm                        mainporcelain common
+perf-test                       mainporcelain common
diff --git a/tools/perf/design.txt b/tools/perf/design.txt
index f71e0d245cb..bd0bb1b1279 100644
--- a/tools/perf/design.txt
+++ b/tools/perf/design.txt
@@ -18,10 +18,10 @@ underlying hardware counters.
 Performance counters are accessed via special file descriptors.
 There's one file descriptor per virtual counter used.
 
-The special file descriptor is opened via the perf_counter_open()
+The special file descriptor is opened via the perf_event_open()
 system call:
 
-   int sys_perf_counter_open(struct perf_counter_hw_event *hw_event_uptr,
+   int sys_perf_event_open(struct perf_event_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_counter_hw_event' is:
+When creating a new counter fd, 'perf_event_attr' is:
 
-struct perf_counter_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
@@ -93,7 +93,7 @@ specified by 'event_id':
 
 /*
  * Generalized performance counter event types, used by the hw_event.event_id
- * parameter of the sys_perf_counter_open() syscall:
+ * parameter of the sys_perf_event_open() syscall:
  */
 enum hw_event_ids {
         /*
@@ -101,10 +101,10 @@ enum hw_event_ids {
          */
         PERF_COUNT_HW_CPU_CYCLES                = 0,
         PERF_COUNT_HW_INSTRUCTIONS              = 1,
-        PERF_COUNT_HW_CACHE_REFERENCES  = 2,
+        PERF_COUNT_HW_CACHE_REFERENCES          = 2,
         PERF_COUNT_HW_CACHE_MISSES              = 3,
         PERF_COUNT_HW_BRANCH_INSTRUCTIONS       = 4,
-        PERF_COUNT_HW_BRANCH_MISSES     = 5,
+        PERF_COUNT_HW_BRANCH_MISSES             = 5,
         PERF_COUNT_HW_BUS_CYCLES                = 6,
 };
 
@@ -131,12 +131,14 @@ software events, selected by 'event_id':
  */
 enum sw_event_ids {
         PERF_COUNT_SW_CPU_CLOCK         = 0,
-        PERF_COUNT_SW_TASK_CLOCK                = 1,
-        PERF_COUNT_SW_PAGE_FAULTS               = 2,
+        PERF_COUNT_SW_TASK_CLOCK        = 1,
+        PERF_COUNT_SW_PAGE_FAULTS       = 2,
         PERF_COUNT_SW_CONTEXT_SWITCHES  = 3,
         PERF_COUNT_SW_CPU_MIGRATIONS    = 4,
         PERF_COUNT_SW_PAGE_FAULTS_MIN   = 5,
         PERF_COUNT_SW_PAGE_FAULTS_MAJ   = 6,
+        PERF_COUNT_SW_ALIGNMENT_FAULTS  = 7,
+        PERF_COUNT_SW_EMULATION_FAULTS  = 8,
 };
 
 Counters of the type PERF_TYPE_TRACEPOINT are available when the ftrace event
@@ -159,7 +161,7 @@ in size.
  * reads on the counter should return the indicated quantities,
  * in increasing order of bit value, after the counter value.
  */
-enum perf_counter_read_format {
+enum perf_event_read_format {
         PERF_FORMAT_TOTAL_TIME_ENABLED  =  1,
         PERF_FORMAT_TOTAL_TIME_RUNNING  =  2,
 };
@@ -178,7 +180,7 @@ interrupt:
  * Bits that can be set in hw_event.record_type to request information
  * in the overflow packets.
  */
-enum perf_counter_record_format {
+enum perf_event_record_format {
         PERF_RECORD_IP          = 1U << 0,
         PERF_RECORD_TID         = 1U << 1,
         PERF_RECORD_TIME        = 1U << 2,
@@ -228,7 +230,7 @@ these events are recorded in the ring-buffer (see below).
 The 'comm' bit allows tracking of process comm data on process creation.
 This too is recorded in the ring-buffer (see below).
 
-The 'pid' parameter to the perf_counter_open() system call allows the
+The 'pid' parameter to the perf_event_open() system call allows the
 counter to be specific to a task:
 
  pid == 0: if the pid parameter is zero, the counter is attached to the
@@ -258,7 +260,7 @@ The 'flags' parameter is currently unused and must be zero.
 
 The 'group_fd' parameter allows counter "groups" to be set up.  A
 counter group has one counter which is the group "leader".  The leader
-is created first, with group_fd = -1 in the perf_counter_open call
+is created first, with group_fd = -1 in the perf_event_open call
 that creates it.  The rest of the group members are created
 subsequently, with group_fd giving the fd of the group leader.
 (A single counter on its own is created with group_fd = -1 and is
@@ -277,13 +279,13 @@ tracking are logged into a ring-buffer. This ring-buffer is created and
 accessed through mmap().
 
 The mmap size should be 1+2^n pages, where the first page is a meta-data page
-(struct perf_counter_mmap_page) that contains various bits of information such
+(struct perf_event_mmap_page) that contains various bits of information such
 as where the ring-buffer head is.
 
 /*
  * Structure of the page that can be mapped via mmap
  */
-struct perf_counter_mmap_page {
+struct perf_event_mmap_page {
         __u32   version;                /* version number of this structure */
         __u32   compat_version;         /* lowest version this is compat with */
 
@@ -317,7 +319,7 @@ struct perf_counter_mmap_page {
          * Control data for the mmap() data buffer.
          *
          * User-space reading this value should issue an rmb(), on SMP capable
-         * platforms, after reading this value -- see perf_counter_wakeup().
+         * platforms, after reading this value -- see perf_event_wakeup().
          */
         __u32   data_head;              /* head in the data section */
 };
@@ -327,9 +329,9 @@ NOTE: the hw-counter userspace bits are arch specific and are currently only
 
 The following 2^n pages are the ring-buffer which contains events of the form:
 
-#define PERF_EVENT_MISC_KERNEL          (1 << 0)
-#define PERF_EVENT_MISC_USER            (1 << 1)
-#define PERF_EVENT_MISC_OVERFLOW        (1 << 2)
+#define PERF_RECORD_MISC_KERNEL          (1 << 0)
+#define PERF_RECORD_MISC_USER            (1 << 1)
+#define PERF_RECORD_MISC_OVERFLOW        (1 << 2)
 
 struct perf_event_header {
         __u32   type;
@@ -353,8 +355,8 @@ enum perf_event_type {
          *      char                            filename[];
          * };
          */
-        PERF_EVENT_MMAP                 = 1,
-        PERF_EVENT_MUNMAP               = 2,
+        PERF_RECORD_MMAP                 = 1,
+        PERF_RECORD_MUNMAP               = 2,
 
         /*
          * struct {
@@ -364,10 +366,10 @@ enum perf_event_type {
          *      char                            comm[];
          * };
          */
-        PERF_EVENT_COMM                 = 3,
+        PERF_RECORD_COMM                 = 3,
 
         /*
-         * When header.misc & PERF_EVENT_MISC_OVERFLOW the event_type field
+         * When header.misc & PERF_RECORD_MISC_OVERFLOW the event_type field
          * will be PERF_RECORD_*
          *
          * struct {
@@ -397,7 +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_counter_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.
@@ -409,11 +411,11 @@ events but does continue to exist and maintain its count value.
 
 An individual counter or counter group can be enabled with
 
-        ioctl(fd, PERF_COUNTER_IOC_ENABLE);
+        ioctl(fd, PERF_EVENT_IOC_ENABLE);
 
 or disabled with
 
-        ioctl(fd, PERF_COUNTER_IOC_DISABLE);
+        ioctl(fd, PERF_EVENT_IOC_DISABLE);
 
 Enabling or disabling the leader of a group enables or disables the
 whole group; that is, while the group leader is disabled, none of the
@@ -424,16 +426,16 @@ other counter.
 
 Additionally, non-inherited overflow counters can use
 
-        ioctl(fd, PERF_COUNTER_IOC_REFRESH, nr);
+        ioctl(fd, PERF_EVENT_IOC_REFRESH, nr);
 
 to enable a counter for 'nr' events, after which it gets disabled again.
 
 A process can enable or disable all the counter groups that are
 attached to it, using prctl:
 
-        prctl(PR_TASK_PERF_COUNTERS_ENABLE);
+        prctl(PR_TASK_PERF_EVENTS_ENABLE);
 
-        prctl(PR_TASK_PERF_COUNTERS_DISABLE);
+        prctl(PR_TASK_PERF_EVENTS_DISABLE);
 
 This applies to all counters on the current process, whether created
 by this process or by another, and doesn't affect any counters that
@@ -447,11 +449,14 @@ Arch requirements
 If your architecture does not have hardware performance metrics, you can
 still use the generic software counters based on hrtimers for sampling.
 
-So to start with, in order to add HAVE_PERF_COUNTERS to your Kconfig, you
+So to start with, in order to add HAVE_PERF_EVENTS to your Kconfig, you
 will need at least this:
-        - asm/perf_counter.h - a basic stub will suffice at first
+        - asm/perf_event.h - a basic stub will suffice at first
         - support for atomic64 types (and associated helper functions)
-        - set_perf_counter_pending() implemented
+        - set_perf_event_pending() implemented
 
 If your architecture does have hardware capabilities, you can override the
-weak stub hw_perf_counter_init() to register hardware counters.
+weak stub hw_perf_event_init() to register hardware counters.
+
+Architectures that have d-cache aliassing issues, such as Sparc and ARM,
+should select PERF_USE_VMALLOC in order to avoid these for perf mmap().
diff --git a/tools/perf/feature-tests.mak b/tools/perf/feature-tests.mak
new file mode 100644
index 00000000000..b041ca67a2c
--- /dev/null
+++ b/tools/perf/feature-tests.mak
@@ -0,0 +1,130 @@
+define SOURCE_HELLO
+#include <stdio.h>
+int main(void)
+{
+        return puts(\"hi\");
+}
+endef
+
+ifndef NO_DWARF
+define SOURCE_DWARF
+#include <dwarf.h>
+#include <elfutils/libdw.h>
+#include <elfutils/version.h>
+#ifndef _ELFUTILS_PREREQ
+#error
+#endif
+
+int main(void)
+{
+        Dwarf *dbg = dwarf_begin(0, DWARF_C_READ);
+        return (long)dbg;
+}
+endef
+endif
+
+define SOURCE_LIBELF
+#include <libelf.h>
+
+int main(void)
+{
+        Elf *elf = elf_begin(0, ELF_C_READ, 0);
+        return (long)elf;
+}
+endef
+
+define SOURCE_GLIBC
+#include <gnu/libc-version.h>
+
+int main(void)
+{
+        const char *version = gnu_get_libc_version();
+        return (long)version;
+}
+endef
+
+define SOURCE_ELF_MMAP
+#include <libelf.h>
+int main(void)
+{
+        Elf *elf = elf_begin(0, ELF_C_READ_MMAP, 0);
+        return (long)elf;
+}
+endef
+
+ifndef NO_NEWT
+define SOURCE_NEWT
+#include <newt.h>
+
+int main(void)
+{
+        newtInit();
+        newtCls();
+        return newtFinished();
+}
+endef
+endif
+
+ifndef NO_LIBPERL
+define SOURCE_PERL_EMBED
+#include <EXTERN.h>
+#include <perl.h>
+
+int main(void)
+{
+perl_alloc();
+return 0;
+}
+endef
+endif
+
+ifndef NO_LIBPYTHON
+define SOURCE_PYTHON_EMBED
+#include <Python.h>
+
+int main(void)
+{
+        Py_Initialize();
+        return 0;
+}
+endef
+endif
+
+define SOURCE_BFD
+#include <bfd.h>
+
+int main(void)
+{
+        bfd_demangle(0, 0, 0);
+        return 0;
+}
+endef
+
+define SOURCE_CPLUS_DEMANGLE
+extern char *cplus_demangle(const char *, int);
+
+int main(void)
+{
+        cplus_demangle(0, 0);
+        return 0;
+}
+endef
+
+define SOURCE_STRLCPY
+#include <stdlib.h>
+extern size_t strlcpy(char *dest, const char *src, size_t size);
+
+int main(void)
+{
+        strlcpy(NULL, NULL, 0);
+        return 0;
+}
+endef
+
+# try-cc
+# Usage: option = $(call try-cc, source-to-build, cc-options)
+try-cc = $(shell sh -c                                            \
+        'TMP="$(OUTPUT)$(TMPOUT).$$$$";                           \
+         echo "$(1)" |                                            \
+         $(CC) -x c - $(2) -o "$$TMP" > /dev/null 2>&1 && echo y; \
+         rm -f "$$TMP"')
diff --git a/tools/perf/perf-archive.sh b/tools/perf/perf-archive.sh
new file mode 100644
index 00000000000..677e59d62a8
--- /dev/null
+++ b/tools/perf/perf-archive.sh
@@ -0,0 +1,46 @@
+#!/bin/bash
+# perf archive
+# Arnaldo Carvalho de Melo <acme@redhat.com>
+
+PERF_DATA=perf.data
+if [ $# -ne 0 ] ; then
+        PERF_DATA=$1
+fi
+
+#
+# PERF_BUILDID_DIR environment variable set by perf
+# path to buildid directory, default to $HOME/.debug
+#
+if [ -z $PERF_BUILDID_DIR ]; then
+        PERF_BUILDID_DIR=~/.debug/
+else
+        # append / to make substitutions work
+        PERF_BUILDID_DIR=$PERF_BUILDID_DIR/
+fi
+
+BUILDIDS=$(mktemp /tmp/perf-archive-buildids.XXXXXX)
+NOBUILDID=0000000000000000000000000000000000000000
+
+perf buildid-list -i $PERF_DATA --with-hits | grep -v "^$NOBUILDID " > $BUILDIDS
+if [ ! -s $BUILDIDS ] ; then
+        echo "perf archive: no build-ids found"
+        rm -f $BUILDIDS
+        exit 1
+fi
+
+MANIFEST=$(mktemp /tmp/perf-archive-manifest.XXXXXX)
+
+cut -d ' ' -f 1 $BUILDIDS | \
+while read build_id ; do
+        linkname=$PERF_BUILDID_DIR.build-id/${build_id:0:2}/${build_id:2}
+        filename=$(readlink -f $linkname)
+        echo ${linkname#$PERF_BUILDID_DIR} >> $MANIFEST
+        echo ${filename#$PERF_BUILDID_DIR} >> $MANIFEST
+done
+
+tar cfj $PERF_DATA.tar.bz2 -C $PERF_BUILDID_DIR -T $MANIFEST
+rm -f $MANIFEST $BUILDIDS
+echo -e "Now please run:\n"
+echo -e "$ tar xvf $PERF_DATA.tar.bz2 -C ~/.debug\n"
+echo "wherever you need to run 'perf report' on."
+exit 0
diff --git a/tools/perf/perf.c b/tools/perf/perf.c
index 4eb72593370..5b1ecd66bb3 100644
--- a/tools/perf/perf.c
+++ b/tools/perf/perf.c
@@ -12,6 +12,8 @@
 #include "util/cache.h"
 #include "util/quote.h"
 #include "util/run-command.h"
+#include "util/parse-events.h"
+#include "util/debugfs.h"
 
 const char perf_usage_string[] =
         "perf [--version] [--help] COMMAND [ARGS]";
@@ -19,12 +21,16 @@ const char perf_usage_string[] =
 const char perf_more_info_string[] =
         "See 'perf help COMMAND' for more information on a specific command.";
 
+int use_browser = -1;
 static int use_pager = -1;
+
 struct pager_config {
         const char *cmd;
         int val;
 };
 
+static char debugfs_mntpt[MAXPATHLEN];
+
 static int pager_command_config(const char *var, const char *value, void *data)
 {
         struct pager_config *c = data;
@@ -43,7 +49,26 @@ int check_pager_config(const char *cmd)
         return c.val;
 }
 
-static void commit_pager_choice(void) {
+static int tui_command_config(const char *var, const char *value, void *data)
+{
+        struct pager_config *c = data;
+        if (!prefixcmp(var, "tui.") && !strcmp(var + 4, c->cmd))
+                c->val = perf_config_bool(var, value);
+        return 0;
+}
+
+/* returns 0 for "no tui", 1 for "use tui", and -1 for "not specified" */
+static int check_tui_config(const char *cmd)
+{
+        struct pager_config c;
+        c.cmd = cmd;
+        c.val = -1;
+        perf_config(tui_command_config, &c);
+        return c.val;
+}
+
+static void commit_pager_choice(void)
+{
         switch (use_pager) {
         case 0:
                 setenv("PERF_PAGER", "cat", 1);
@@ -56,7 +81,16 @@ static void commit_pager_choice(void) {
         }
 }
 
-static int handle_options(const char*** argv, int* argc, int* envchanged)
+static void set_debugfs_path(void)
+{
+        char *path;
+
+        path = getenv(PERF_DEBUGFS_ENVIRONMENT);
+        snprintf(debugfs_path, MAXPATHLEN, "%s/%s", path ?: debugfs_mntpt,
+                 "tracing/events");
+}
+
+static int handle_options(const char ***argv, int *argc, int *envchanged)
 {
         int handled = 0;
 
@@ -76,8 +110,8 @@ static int handle_options(const char*** argv, int* argc, int* envchanged)
                 /*
                  * Check remaining flags.
                  */
-                if (!prefixcmp(cmd, "--exec-path")) {
-                        cmd += 11;
+                if (!prefixcmp(cmd, CMD_EXEC_PATH)) {
+                        cmd += strlen(CMD_EXEC_PATH);
                         if (*cmd == '=')
                                 perf_set_argv_exec_path(cmd + 1);
                         else {
@@ -95,7 +129,7 @@ static int handle_options(const char*** argv, int* argc, int* envchanged)
                                 *envchanged = 1;
                 } else if (!strcmp(cmd, "--perf-dir")) {
                         if (*argc < 2) {
-                                fprintf(stderr, "No directory given for --perf-dir.\n" );
+                                fprintf(stderr, "No directory given for --perf-dir.\n");
                                 usage(perf_usage_string);
                         }
                         setenv(PERF_DIR_ENVIRONMENT, (*argv)[1], 1);
@@ -104,13 +138,13 @@ static int handle_options(const char*** argv, int* argc, int* envchanged)
                         (*argv)++;
                         (*argc)--;
                         handled++;
-                } else if (!prefixcmp(cmd, "--perf-dir=")) {
-                        setenv(PERF_DIR_ENVIRONMENT, cmd + 10, 1);
+                } else if (!prefixcmp(cmd, CMD_PERF_DIR)) {
+                        setenv(PERF_DIR_ENVIRONMENT, cmd + strlen(CMD_PERF_DIR), 1);
                         if (envchanged)
                                 *envchanged = 1;
                 } else if (!strcmp(cmd, "--work-tree")) {
                         if (*argc < 2) {
-                                fprintf(stderr, "No directory given for --work-tree.\n" );
+                                fprintf(stderr, "No directory given for --work-tree.\n");
                                 usage(perf_usage_string);
                         }
                         setenv(PERF_WORK_TREE_ENVIRONMENT, (*argv)[1], 1);
@@ -118,8 +152,24 @@ static int handle_options(const char*** argv, int* argc, int* envchanged)
                                 *envchanged = 1;
                         (*argv)++;
                         (*argc)--;
-                } else if (!prefixcmp(cmd, "--work-tree=")) {
-                        setenv(PERF_WORK_TREE_ENVIRONMENT, cmd + 12, 1);
+                } else if (!prefixcmp(cmd, CMD_WORK_TREE)) {
+                        setenv(PERF_WORK_TREE_ENVIRONMENT, cmd + strlen(CMD_WORK_TREE), 1);
+                        if (envchanged)
+                                *envchanged = 1;
+                } else if (!strcmp(cmd, "--debugfs-dir")) {
+                        if (*argc < 2) {
+                                fprintf(stderr, "No directory given for --debugfs-dir.\n");
+                                usage(perf_usage_string);
+                        }
+                        strncpy(debugfs_mntpt, (*argv)[1], MAXPATHLEN);
+                        debugfs_mntpt[MAXPATHLEN - 1] = '\0';
+                        if (envchanged)
+                                *envchanged = 1;
+                        (*argv)++;
+                        (*argc)--;
+                } else if (!prefixcmp(cmd, CMD_DEBUGFS_DIR)) {
+                        strncpy(debugfs_mntpt, cmd + strlen(CMD_DEBUGFS_DIR), MAXPATHLEN);
+                        debugfs_mntpt[MAXPATHLEN - 1] = '\0';
                         if (envchanged)
                                 *envchanged = 1;
                 } else {
@@ -138,7 +188,7 @@ static int handle_alias(int *argcp, const char ***argv)
 {
         int envchanged = 0, ret = 0, saved_errno = errno;
         int count, option_count;
-        const char** new_argv;
+        const char **new_argv;
         const char *alias_command;
         char *alias_string;
 
@@ -180,11 +230,11 @@ static int handle_alias(int *argcp, const char ***argv)
                 if (!strcmp(alias_command, new_argv[0]))
                         die("recursive alias: %s", alias_command);
 
-                new_argv = realloc(new_argv, sizeof(char*) *
+                new_argv = realloc(new_argv, sizeof(char *) *
                                     (count + *argcp + 1));
                 /* insert after command name */
-                memcpy(new_argv + count, *argv + 1, sizeof(char*) * *argcp);
-                new_argv[count+*argcp] = NULL;
+                memcpy(new_argv + count, *argv + 1, sizeof(char *) * *argcp);
+                new_argv[count + *argcp] = NULL;
 
                 *argv = new_argv;
                 *argcp += count - 1;
@@ -223,16 +273,21 @@ static int run_builtin(struct cmd_struct *p, int argc, const char **argv)
         if (p->option & RUN_SETUP)
                 prefix = NULL; /* setup_perf_directory(); */
 
+        if (use_browser == -1)
+                use_browser = check_tui_config(p->cmd);
+
         if (use_pager == -1 && p->option & RUN_SETUP)
                 use_pager = check_pager_config(p->cmd);
         if (use_pager == -1 && p->option & USE_PAGER)
                 use_pager = 1;
         commit_pager_choice();
-
-        if (p->option & NEED_WORK_TREE)
-                /* setup_work_tree() */;
+        set_debugfs_path();
 
         status = p->fn(argc, argv, prefix);
+        exit_browser(status);
+
+        perf_evsel_list__delete();
+
         if (status)
                 return status & 0xff;
 
@@ -257,16 +312,29 @@ static void handle_internal_command(int argc, const char **argv)
 {
         const char *cmd = argv[0];
         static struct cmd_struct commands[] = {
-                { "help", cmd_help, 0 },
-                { "list", cmd_list, 0 },
-                { "record", cmd_record, 0 },
-                { "report", cmd_report, 0 },
-                { "stat", cmd_stat, 0 },
-                { "top", cmd_top, 0 },
-                { "annotate", cmd_annotate, 0 },
-                { "version", cmd_version, 0 },
+                { "buildid-cache", cmd_buildid_cache, 0 },
+                { "buildid-list", cmd_buildid_list, 0 },
+                { "diff",       cmd_diff,       0 },
+                { "help",       cmd_help,       0 },
+                { "list",       cmd_list,       0 },
+                { "record",     cmd_record,     0 },
+                { "report",     cmd_report,     0 },
+                { "bench",      cmd_bench,      0 },
+                { "stat",       cmd_stat,       0 },
+                { "timechart",  cmd_timechart,  0 },
+                { "top",        cmd_top,        0 },
+                { "annotate",   cmd_annotate,   0 },
+                { "version",    cmd_version,    0 },
+                { "script",     cmd_script,     0 },
+                { "sched",      cmd_sched,      0 },
+                { "probe",      cmd_probe,      0 },
+                { "kmem",       cmd_kmem,       0 },
+                { "lock",       cmd_lock,       0 },
+                { "kvm",        cmd_kvm,        0 },
+                { "test",       cmd_test,       0 },
+                { "inject",     cmd_inject,     0 },
         };
-        int i;
+        unsigned int i;
         static const char ext[] = STRIP_EXTENSION;
 
         if (sizeof(ext) > 1) {
@@ -349,6 +417,16 @@ static int run_argv(int *argcp, const char ***argv)
         return done_alias;
 }
 
+/* mini /proc/mounts parser: searching for "^blah /mount/point debugfs" */
+static void get_debugfs_mntpt(void)
+{
+        const char *path = debugfs_mount(NULL);
+
+        if (path)
+                strncpy(debugfs_mntpt, path, sizeof(debugfs_mntpt));
+        else
+                debugfs_mntpt[0] = '\0';
+}
 
 int main(int argc, const char **argv)
 {
@@ -357,7 +435,8 @@ int main(int argc, const char **argv)
         cmd = perf_extract_argv0_path(argv[0]);
         if (!cmd)
                 cmd = "perf-help";
-
+        /* get debugfs mount point from /proc/mounts */
+        get_debugfs_mntpt();
         /*
          * "perf-xxxx" is the same as "perf xxxx", but we obviously:
          *
@@ -380,6 +459,9 @@ int main(int argc, const char **argv)
         argc--;
         handle_options(&argv, &argc, NULL);
         commit_pager_choice();
+        set_debugfs_path();
+        set_buildid_dir();
+
         if (argc > 0) {
                 if (!prefixcmp(argv[0], "--"))
                         argv[0] += 2;
@@ -394,15 +476,15 @@ int main(int argc, const char **argv)
 
         /*
          * We use PATH to find perf commands, but we prepend some higher
-         * precidence paths: the "--exec-path" option, the PERF_EXEC_PATH
+         * precedence paths: the "--exec-path" option, the PERF_EXEC_PATH
          * environment, and the $(perfexecdir) from the Makefile at build
          * time.
          */
         setup_path();
 
         while (1) {
-                static int done_help = 0;
-                static int was_alias = 0;
+                static int done_help;
+                static int was_alias;
 
                 was_alias = run_argv(&argc, &argv);
                 if (errno != ENOENT)
diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index ceb68aa51f7..95aaf565c70 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -1,7 +1,17 @@
 #ifndef _PERF_PERF_H
 #define _PERF_PERF_H
 
-#if defined(__x86_64__) || defined(__i386__)
+struct winsize;
+
+void get_term_dimensions(struct winsize *ws);
+
+#if defined(__i386__)
+#include "../../arch/x86/include/asm/unistd.h"
+#define rmb()           asm volatile("lock; addl $0,0(%%esp)" ::: "memory")
+#define cpu_relax()     asm volatile("rep; nop" ::: "memory");
+#endif
+
+#if defined(__x86_64__)
 #include "../../arch/x86/include/asm/unistd.h"
 #define rmb()           asm volatile("lfence" ::: "memory")
 #define cpu_relax()     asm volatile("rep; nop" ::: "memory");
@@ -19,20 +29,77 @@
 #define cpu_relax()     asm volatile("" ::: "memory");
 #endif
 
+#ifdef __sh__
+#include "../../arch/sh/include/asm/unistd.h"
+#if defined(__SH4A__) || defined(__SH5__)
+# define rmb()          asm volatile("synco" ::: "memory")
+#else
+# define rmb()          asm volatile("" ::: "memory")
+#endif
+#define cpu_relax()     asm volatile("" ::: "memory")
+#endif
+
+#ifdef __hppa__
+#include "../../arch/parisc/include/asm/unistd.h"
+#define rmb()           asm volatile("" ::: "memory")
+#define cpu_relax()     asm volatile("" ::: "memory");
+#endif
+
+#ifdef __sparc__
+#include "../../arch/sparc/include/asm/unistd.h"
+#define rmb()           asm volatile("":::"memory")
+#define cpu_relax()     asm volatile("":::"memory")
+#endif
+
+#ifdef __alpha__
+#include "../../arch/alpha/include/asm/unistd.h"
+#define rmb()           asm volatile("mb" ::: "memory")
+#define cpu_relax()     asm volatile("" ::: "memory")
+#endif
+
+#ifdef __ia64__
+#include "../../arch/ia64/include/asm/unistd.h"
+#define rmb()           asm volatile ("mf" ::: "memory")
+#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()           ((void(*)(void))0xffff0fa0)()
+#define cpu_relax()     asm volatile("":::"memory")
+#endif
+
+#ifdef __mips__
+#include "../../arch/mips/include/asm/unistd.h"
+#define rmb()           asm volatile(                                   \
+                                ".set   mips2\n\t"                      \
+                                "sync\n\t"                              \
+                                ".set   mips0"                          \
+                                : /* no output */                       \
+                                : /* no input */                        \
+                                : "memory")
+#define cpu_relax()     asm volatile("" ::: "memory")
+#endif
+
 #include <time.h>
 #include <unistd.h>
 #include <sys/types.h>
 #include <sys/syscall.h>
 
-#include "../../include/linux/perf_counter.h"
-#include "types.h"
+#include "../../include/linux/perf_event.h"
+#include "util/types.h"
+#include <stdbool.h>
 
 /*
- * prctl(PR_TASK_PERF_COUNTERS_DISABLE) will (cheaply) disable all
+ * prctl(PR_TASK_PERF_EVENTS_DISABLE) will (cheaply) disable all
  * counters in the current task.
  */
-#define PR_TASK_PERF_COUNTERS_DISABLE   31
-#define PR_TASK_PERF_COUNTERS_ENABLE    32
+#define PR_TASK_PERF_EVENTS_DISABLE   31
+#define PR_TASK_PERF_EVENTS_ENABLE    32
 
 #ifndef NSEC_PER_SEC
 # define NSEC_PER_SEC                   1000000000ULL
@@ -60,22 +127,23 @@ static inline unsigned long long rdclock(void)
         _min1 < _min2 ? _min1 : _min2; })
 
 static inline int
-sys_perf_counter_open(struct perf_counter_attr *attr,
+sys_perf_event_open(struct perf_event_attr *attr,
                       pid_t pid, int cpu, int group_fd,
                       unsigned long flags)
 {
         attr->size = sizeof(*attr);
-        return syscall(__NR_perf_counter_open, attr, pid, cpu,
+        return syscall(__NR_perf_event_open, attr, pid, cpu,
                        group_fd, flags);
 }
 
 #define MAX_COUNTERS                    256
 #define MAX_NR_CPUS                     256
 
-struct perf_file_header {
-        u64     version;
-        u64     sample_type;
-        u64     data_size;
+struct ip_callchain {
+        u64 nr;
+        u64 ips[0];
 };
 
+extern bool perf_host, perf_guest;
+
 #endif
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/Context.c b/tools/perf/scripts/perl/Perf-Trace-Util/Context.c
new file mode 100644
index 00000000000..790ceba6ad3
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/Context.c
@@ -0,0 +1,135 @@
+/*
+ * This file was generated automatically by ExtUtils::ParseXS version 2.18_02 from the
+ * contents of Context.xs. Do not edit this file, edit Context.xs instead.
+ *
+ *      ANY CHANGES MADE HERE WILL BE LOST! 
+ *
+ */
+
+#line 1 "Context.xs"
+/*
+ * Context.xs.  XS interfaces for perf script.
+ *
+ * Copyright (C) 2009 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include "EXTERN.h"
+#include "perl.h"
+#include "XSUB.h"
+#include "../../../perf.h"
+#include "../../../util/trace-event.h"
+
+#ifndef PERL_UNUSED_VAR
+#  define PERL_UNUSED_VAR(var) if (0) var = var
+#endif
+
+#line 42 "Context.c"
+
+XS(XS_Perf__Trace__Context_common_pc); /* prototype to pass -Wmissing-prototypes */
+XS(XS_Perf__Trace__Context_common_pc)
+{
+#ifdef dVAR
+    dVAR; dXSARGS;
+#else
+    dXSARGS;
+#endif
+    if (items != 1)
+       Perl_croak(aTHX_ "Usage: %s(%s)", "Perf::Trace::Context::common_pc", "context");
+    PERL_UNUSED_VAR(cv); /* -W */
+    {
+        struct scripting_context *      context = INT2PTR(struct scripting_context *,SvIV(ST(0)));
+        int     RETVAL;
+        dXSTARG;
+
+        RETVAL = common_pc(context);
+        XSprePUSH; PUSHi((IV)RETVAL);
+    }
+    XSRETURN(1);
+}
+
+
+XS(XS_Perf__Trace__Context_common_flags); /* prototype to pass -Wmissing-prototypes */
+XS(XS_Perf__Trace__Context_common_flags)
+{
+#ifdef dVAR
+    dVAR; dXSARGS;
+#else
+    dXSARGS;
+#endif
+    if (items != 1)
+       Perl_croak(aTHX_ "Usage: %s(%s)", "Perf::Trace::Context::common_flags", "context");
+    PERL_UNUSED_VAR(cv); /* -W */
+    {
+        struct scripting_context *      context = INT2PTR(struct scripting_context *,SvIV(ST(0)));
+        int     RETVAL;
+        dXSTARG;
+
+        RETVAL = common_flags(context);
+        XSprePUSH; PUSHi((IV)RETVAL);
+    }
+    XSRETURN(1);
+}
+
+
+XS(XS_Perf__Trace__Context_common_lock_depth); /* prototype to pass -Wmissing-prototypes */
+XS(XS_Perf__Trace__Context_common_lock_depth)
+{
+#ifdef dVAR
+    dVAR; dXSARGS;
+#else
+    dXSARGS;
+#endif
+    if (items != 1)
+       Perl_croak(aTHX_ "Usage: %s(%s)", "Perf::Trace::Context::common_lock_depth", "context");
+    PERL_UNUSED_VAR(cv); /* -W */
+    {
+        struct scripting_context *      context = INT2PTR(struct scripting_context *,SvIV(ST(0)));
+        int     RETVAL;
+        dXSTARG;
+
+        RETVAL = common_lock_depth(context);
+        XSprePUSH; PUSHi((IV)RETVAL);
+    }
+    XSRETURN(1);
+}
+
+#ifdef __cplusplus
+extern "C"
+#endif
+XS(boot_Perf__Trace__Context); /* prototype to pass -Wmissing-prototypes */
+XS(boot_Perf__Trace__Context)
+{
+#ifdef dVAR
+    dVAR; dXSARGS;
+#else
+    dXSARGS;
+#endif
+    const char* file = __FILE__;
+
+    PERL_UNUSED_VAR(cv); /* -W */
+    PERL_UNUSED_VAR(items); /* -W */
+    XS_VERSION_BOOTCHECK ;
+
+        newXSproto("Perf::Trace::Context::common_pc", XS_Perf__Trace__Context_common_pc, file, "$");
+        newXSproto("Perf::Trace::Context::common_flags", XS_Perf__Trace__Context_common_flags, file, "$");
+        newXSproto("Perf::Trace::Context::common_lock_depth", XS_Perf__Trace__Context_common_lock_depth, file, "$");
+    if (PL_unitcheckav)
+         call_list(PL_scopestack_ix, PL_unitcheckav);
+    XSRETURN_YES;
+}
+
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/Context.xs b/tools/perf/scripts/perl/Perf-Trace-Util/Context.xs
new file mode 100644
index 00000000000..c1e2ed1ed34
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/Context.xs
@@ -0,0 +1,42 @@
+/*
+ * Context.xs.  XS interfaces for perf script.
+ *
+ * Copyright (C) 2009 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include "EXTERN.h"
+#include "perl.h"
+#include "XSUB.h"
+#include "../../../perf.h"
+#include "../../../util/script-event.h"
+
+MODULE = Perf::Trace::Context           PACKAGE = Perf::Trace::Context
+PROTOTYPES: ENABLE
+
+int
+common_pc(context)
+        struct scripting_context * context
+
+int
+common_flags(context)
+        struct scripting_context * context
+
+int
+common_lock_depth(context)
+        struct scripting_context * context
+
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/Makefile.PL b/tools/perf/scripts/perl/Perf-Trace-Util/Makefile.PL
new file mode 100644
index 00000000000..decdeb0f678
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/Makefile.PL
@@ -0,0 +1,17 @@
+use 5.010000;
+use ExtUtils::MakeMaker;
+# See lib/ExtUtils/MakeMaker.pm for details of how to influence
+# the contents of the Makefile that is written.
+WriteMakefile(
+    NAME              => 'Perf::Trace::Context',
+    VERSION_FROM      => 'lib/Perf/Trace/Context.pm', # finds $VERSION
+    PREREQ_PM         => {}, # e.g., Module::Name => 1.1
+    ($] >= 5.005 ?     ## Add these new keywords supported since 5.005
+      (ABSTRACT_FROM  => 'lib/Perf/Trace/Context.pm', # retrieve abstract from module
+       AUTHOR         => 'Tom Zanussi <tzanussi@gmail.com>') : ()),
+    LIBS              => [''], # e.g., '-lm'
+    DEFINE            => '-I ../..', # e.g., '-DHAVE_SOMETHING'
+    INC               => '-I.', # e.g., '-I. -I/usr/include/other'
+        # Un-comment this if you add C files to link with later:
+    OBJECT            => 'Context.o', # link all the C files too
+);
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/README b/tools/perf/scripts/perl/Perf-Trace-Util/README
new file mode 100644
index 00000000000..2f0c7f3043e
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/README
@@ -0,0 +1,59 @@
+Perf-Trace-Util version 0.01
+============================
+
+This module contains utility functions for use with perf script.
+
+Core.pm and Util.pm are pure Perl modules; Core.pm contains routines
+that the core perf support for Perl calls on and should always be
+'used', while Util.pm contains useful but optional utility functions
+that scripts may want to use.  Context.pm contains the Perl->C
+interface that allows scripts to access data in the embedding perf
+executable; scripts wishing to do that should 'use Context.pm'.
+
+The Perl->C perf interface is completely driven by Context.xs.  If you
+want to add new Perl functions that end up accessing C data in the
+perf executable, you add desciptions of the new functions here.
+scripting_context is a pointer to the perf data in the perf executable
+that you want to access - it's passed as the second parameter,
+$context, to all handler functions.
+
+After you do that:
+
+  perl Makefile.PL   # to create a Makefile for the next step
+  make               # to create Context.c
+
+  edit Context.c to add const to the char* file = __FILE__ line in
+  XS(boot_Perf__Trace__Context) to silence a warning/error.
+
+  You can delete the Makefile, object files and anything else that was
+  generated e.g. blib and shared library, etc, except for of course
+  Context.c
+
+  You should then be able to run the normal perf make as usual.
+
+INSTALLATION
+
+Building perf with perf script Perl scripting should install this
+module in the right place.
+
+You should make sure libperl and ExtUtils/Embed.pm are installed first
+e.g. apt-get install libperl-dev or yum install perl-ExtUtils-Embed.
+
+DEPENDENCIES
+
+This module requires these other modules and libraries:
+
+  None
+
+COPYRIGHT AND LICENCE
+
+Copyright (C) 2009 by Tom Zanussi <tzanussi@gmail.com>
+
+This library is free software; you can redistribute it and/or modify
+it under the same terms as Perl itself, either Perl version 5.10.0 or,
+at your option, any later version of Perl 5 you may have available.
+
+Alternatively, this software may be distributed under the terms of the
+GNU General Public License ("GPL") version 2 as published by the Free
+Software Foundation.
+
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Context.pm b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Context.pm
new file mode 100644
index 00000000000..4e2f6039ac9
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Context.pm
@@ -0,0 +1,55 @@
+package Perf::Trace::Context;
+
+use 5.010000;
+use strict;
+use warnings;
+
+require Exporter;
+
+our @ISA = qw(Exporter);
+
+our %EXPORT_TAGS = ( 'all' => [ qw(
+) ] );
+
+our @EXPORT_OK = ( @{ $EXPORT_TAGS{'all'} } );
+
+our @EXPORT = qw(
+        common_pc common_flags common_lock_depth
+);
+
+our $VERSION = '0.01';
+
+require XSLoader;
+XSLoader::load('Perf::Trace::Context', $VERSION);
+
+1;
+__END__
+=head1 NAME
+
+Perf::Trace::Context - Perl extension for accessing functions in perf.
+
+=head1 SYNOPSIS
+
+  use Perf::Trace::Context;
+
+=head1 SEE ALSO
+
+Perf (script) documentation
+
+=head1 AUTHOR
+
+Tom Zanussi, E<lt>tzanussi@gmail.com<gt>
+
+=head1 COPYRIGHT AND LICENSE
+
+Copyright (C) 2009 by Tom Zanussi
+
+This library is free software; you can redistribute it and/or modify
+it under the same terms as Perl itself, either Perl version 5.10.0 or,
+at your option, any later version of Perl 5 you may have available.
+
+Alternatively, this software may be distributed under the terms of the
+GNU General Public License ("GPL") version 2 as published by the Free
+Software Foundation.
+
+=cut
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Core.pm b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Core.pm
new file mode 100644
index 00000000000..9158458d3ee
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Core.pm
@@ -0,0 +1,192 @@
+package Perf::Trace::Core;
+
+use 5.010000;
+use strict;
+use warnings;
+
+require Exporter;
+
+our @ISA = qw(Exporter);
+
+our %EXPORT_TAGS = ( 'all' => [ qw(
+) ] );
+
+our @EXPORT_OK = ( @{ $EXPORT_TAGS{'all'} } );
+
+our @EXPORT = qw(
+define_flag_field define_flag_value flag_str dump_flag_fields
+define_symbolic_field define_symbolic_value symbol_str dump_symbolic_fields
+trace_flag_str
+);
+
+our $VERSION = '0.01';
+
+my %trace_flags = (0x00 => "NONE",
+                   0x01 => "IRQS_OFF",
+                   0x02 => "IRQS_NOSUPPORT",
+                   0x04 => "NEED_RESCHED",
+                   0x08 => "HARDIRQ",
+                   0x10 => "SOFTIRQ");
+
+sub trace_flag_str
+{
+    my ($value) = @_;
+
+    my $string;
+
+    my $print_delim = 0;
+
+    foreach my $idx (sort {$a <=> $b} keys %trace_flags) {
+        if (!$value && !$idx) {
+            $string .= "NONE";
+            last;
+        }
+
+        if ($idx && ($value & $idx) == $idx) {
+            if ($print_delim) {
+                $string .= " | ";
+            }
+            $string .= "$trace_flags{$idx}";
+            $print_delim = 1;
+            $value &= ~$idx;
+        }
+    }
+
+    return $string;
+}
+
+my %flag_fields;
+my %symbolic_fields;
+
+sub flag_str
+{
+    my ($event_name, $field_name, $value) = @_;
+
+    my $string;
+
+    if ($flag_fields{$event_name}{$field_name}) {
+        my $print_delim = 0;
+        foreach my $idx (sort {$a <=> $b} keys %{$flag_fields{$event_name}{$field_name}{"values"}}) {
+            if (!$value && !$idx) {
+                $string .= "$flag_fields{$event_name}{$field_name}{'values'}{$idx}";
+                last;
+            }
+            if ($idx && ($value & $idx) == $idx) {
+                if ($print_delim && $flag_fields{$event_name}{$field_name}{'delim'}) {
+                    $string .= " $flag_fields{$event_name}{$field_name}{'delim'} ";
+                }
+                $string .= "$flag_fields{$event_name}{$field_name}{'values'}{$idx}";
+                $print_delim = 1;
+                $value &= ~$idx;
+            }
+        }
+    }
+
+    return $string;
+}
+
+sub define_flag_field
+{
+    my ($event_name, $field_name, $delim) = @_;
+
+    $flag_fields{$event_name}{$field_name}{"delim"} = $delim;
+}
+
+sub define_flag_value
+{
+    my ($event_name, $field_name, $value, $field_str) = @_;
+
+    $flag_fields{$event_name}{$field_name}{"values"}{$value} = $field_str;
+}
+
+sub dump_flag_fields
+{
+    for my $event (keys %flag_fields) {
+        print "event $event:\n";
+        for my $field (keys %{$flag_fields{$event}}) {
+            print "    field: $field:\n";
+            print "        delim: $flag_fields{$event}{$field}{'delim'}\n";
+            foreach my $idx (sort {$a <=> $b} keys %{$flag_fields{$event}{$field}{"values"}}) {
+                print "        value $idx: $flag_fields{$event}{$field}{'values'}{$idx}\n";
+            }
+        }
+    }
+}
+
+sub symbol_str
+{
+    my ($event_name, $field_name, $value) = @_;
+
+    if ($symbolic_fields{$event_name}{$field_name}) {
+        foreach my $idx (sort {$a <=> $b} keys %{$symbolic_fields{$event_name}{$field_name}{"values"}}) {
+            if (!$value && !$idx) {
+                return "$symbolic_fields{$event_name}{$field_name}{'values'}{$idx}";
+                last;
+            }
+            if ($value == $idx) {
+                return "$symbolic_fields{$event_name}{$field_name}{'values'}{$idx}";
+            }
+        }
+    }
+
+    return undef;
+}
+
+sub define_symbolic_field
+{
+    my ($event_name, $field_name) = @_;
+
+    # nothing to do, really
+}
+
+sub define_symbolic_value
+{
+    my ($event_name, $field_name, $value, $field_str) = @_;
+
+    $symbolic_fields{$event_name}{$field_name}{"values"}{$value} = $field_str;
+}
+
+sub dump_symbolic_fields
+{
+    for my $event (keys %symbolic_fields) {
+        print "event $event:\n";
+        for my $field (keys %{$symbolic_fields{$event}}) {
+            print "    field: $field:\n";
+            foreach my $idx (sort {$a <=> $b} keys %{$symbolic_fields{$event}{$field}{"values"}}) {
+                print "        value $idx: $symbolic_fields{$event}{$field}{'values'}{$idx}\n";
+            }
+        }
+    }
+}
+
+1;
+__END__
+=head1 NAME
+
+Perf::Trace::Core - Perl extension for perf script
+
+=head1 SYNOPSIS
+
+  use Perf::Trace::Core
+
+=head1 SEE ALSO
+
+Perf (script) documentation
+
+=head1 AUTHOR
+
+Tom Zanussi, E<lt>tzanussi@gmail.com<gt>
+
+=head1 COPYRIGHT AND LICENSE
+
+Copyright (C) 2009 by Tom Zanussi
+
+This library is free software; you can redistribute it and/or modify
+it under the same terms as Perl itself, either Perl version 5.10.0 or,
+at your option, any later version of Perl 5 you may have available.
+
+Alternatively, this software may be distributed under the terms of the
+GNU General Public License ("GPL") version 2 as published by the Free
+Software Foundation.
+
+=cut
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm
new file mode 100644
index 00000000000..05350011462
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm
@@ -0,0 +1,94 @@
+package Perf::Trace::Util;
+
+use 5.010000;
+use strict;
+use warnings;
+
+require Exporter;
+
+our @ISA = qw(Exporter);
+
+our %EXPORT_TAGS = ( 'all' => [ qw(
+) ] );
+
+our @EXPORT_OK = ( @{ $EXPORT_TAGS{'all'} } );
+
+our @EXPORT = qw(
+avg nsecs nsecs_secs nsecs_nsecs nsecs_usecs print_nsecs
+clear_term
+);
+
+our $VERSION = '0.01';
+
+sub avg
+{
+    my ($total, $n) = @_;
+
+    return $total / $n;
+}
+
+my $NSECS_PER_SEC    = 1000000000;
+
+sub nsecs
+{
+    my ($secs, $nsecs) = @_;
+
+    return $secs * $NSECS_PER_SEC + $nsecs;
+}
+
+sub nsecs_secs {
+    my ($nsecs) = @_;
+
+    return $nsecs / $NSECS_PER_SEC;
+}
+
+sub nsecs_nsecs {
+    my ($nsecs) = @_;
+
+    return $nsecs % $NSECS_PER_SEC;
+}
+
+sub nsecs_str {
+    my ($nsecs) = @_;
+
+    my $str = sprintf("%5u.%09u", nsecs_secs($nsecs), nsecs_nsecs($nsecs));
+
+    return $str;
+}
+
+sub clear_term
+{
+    print "\x1b[H\x1b[2J";
+}
+
+1;
+__END__
+=head1 NAME
+
+Perf::Trace::Util - Perl extension for perf script
+
+=head1 SYNOPSIS
+
+  use Perf::Trace::Util;
+
+=head1 SEE ALSO
+
+Perf (script) documentation
+
+=head1 AUTHOR
+
+Tom Zanussi, E<lt>tzanussi@gmail.com<gt>
+
+=head1 COPYRIGHT AND LICENSE
+
+Copyright (C) 2009 by Tom Zanussi
+
+This library is free software; you can redistribute it and/or modify
+it under the same terms as Perl itself, either Perl version 5.10.0 or,
+at your option, any later version of Perl 5 you may have available.
+
+Alternatively, this software may be distributed under the terms of the
+GNU General Public License ("GPL") version 2 as published by the Free
+Software Foundation.
+
+=cut
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/typemap b/tools/perf/scripts/perl/Perf-Trace-Util/typemap
new file mode 100644
index 00000000000..840836804aa
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/typemap
@@ -0,0 +1 @@
+struct scripting_context * T_PTR
diff --git a/tools/perf/scripts/perl/bin/check-perf-trace-record b/tools/perf/scripts/perl/bin/check-perf-trace-record
new file mode 100644
index 00000000000..423ad6aed05
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/check-perf-trace-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -a -e kmem:kmalloc -e irq:softirq_entry -e kmem:kfree
diff --git a/tools/perf/scripts/perl/bin/failed-syscalls-record b/tools/perf/scripts/perl/bin/failed-syscalls-record
new file mode 100644
index 00000000000..8104895a7b6
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/failed-syscalls-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -e raw_syscalls:sys_exit $@
diff --git a/tools/perf/scripts/perl/bin/failed-syscalls-report b/tools/perf/scripts/perl/bin/failed-syscalls-report
new file mode 100644
index 00000000000..9f83cc1ad8b
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/failed-syscalls-report
@@ -0,0 +1,10 @@
+#!/bin/bash
+# description: system-wide failed syscalls
+# args: [comm]
+if [ $# -gt 0 ] ; then
+    if ! expr match "$1" "-" > /dev/null ; then
+        comm=$1
+        shift
+    fi
+fi
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/failed-syscalls.pl $comm
diff --git a/tools/perf/scripts/perl/bin/rw-by-file-record b/tools/perf/scripts/perl/bin/rw-by-file-record
new file mode 100644
index 00000000000..33efc8673aa
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/rw-by-file-record
@@ -0,0 +1,3 @@
+#!/bin/bash
+perf record -e syscalls:sys_enter_read -e syscalls:sys_enter_write $@
+
diff --git a/tools/perf/scripts/perl/bin/rw-by-file-report b/tools/perf/scripts/perl/bin/rw-by-file-report
new file mode 100644
index 00000000000..77200b3f310
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/rw-by-file-report
@@ -0,0 +1,10 @@
+#!/bin/bash
+# description: r/w activity for a program, by file
+# args: <comm>
+if [ $# -lt 1 ] ; then
+    echo "usage: rw-by-file <comm>"
+    exit
+fi
+comm=$1
+shift
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/rw-by-file.pl $comm
diff --git a/tools/perf/scripts/perl/bin/rw-by-pid-record b/tools/perf/scripts/perl/bin/rw-by-pid-record
new file mode 100644
index 00000000000..7cb9db23044
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/rw-by-pid-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -e syscalls:sys_enter_read -e syscalls:sys_exit_read -e syscalls:sys_enter_write -e syscalls:sys_exit_write $@
diff --git a/tools/perf/scripts/perl/bin/rw-by-pid-report b/tools/perf/scripts/perl/bin/rw-by-pid-report
new file mode 100644
index 00000000000..a27b9f311f9
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/rw-by-pid-report
@@ -0,0 +1,3 @@
+#!/bin/bash
+# description: system-wide r/w activity
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/rw-by-pid.pl
diff --git a/tools/perf/scripts/perl/bin/rwtop-record b/tools/perf/scripts/perl/bin/rwtop-record
new file mode 100644
index 00000000000..7cb9db23044
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/rwtop-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -e syscalls:sys_enter_read -e syscalls:sys_exit_read -e syscalls:sys_enter_write -e syscalls:sys_exit_write $@
diff --git a/tools/perf/scripts/perl/bin/rwtop-report b/tools/perf/scripts/perl/bin/rwtop-report
new file mode 100644
index 00000000000..83e11ec2e19
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/rwtop-report
@@ -0,0 +1,20 @@
+#!/bin/bash
+# description: system-wide r/w top
+# args: [interval]
+n_args=0
+for i in "$@"
+do
+    if expr match "$i" "-" > /dev/null ; then
+        break
+    fi
+    n_args=$(( $n_args + 1 ))
+done
+if [ "$n_args" -gt 1 ] ; then
+    echo "usage: rwtop-report [interval]"
+    exit
+fi
+if [ "$n_args" -gt 0 ] ; then
+    interval=$1
+    shift
+fi
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/rwtop.pl $interval
diff --git a/tools/perf/scripts/perl/bin/wakeup-latency-record b/tools/perf/scripts/perl/bin/wakeup-latency-record
new file mode 100644
index 00000000000..464251a1bd7
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/wakeup-latency-record
@@ -0,0 +1,6 @@
+#!/bin/bash
+perf record -e sched:sched_switch -e sched:sched_wakeup $@
+
+
+
+
diff --git a/tools/perf/scripts/perl/bin/wakeup-latency-report b/tools/perf/scripts/perl/bin/wakeup-latency-report
new file mode 100644
index 00000000000..889e8130cca
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/wakeup-latency-report
@@ -0,0 +1,3 @@
+#!/bin/bash
+# description: system-wide min/max/avg wakeup latency
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/wakeup-latency.pl
diff --git a/tools/perf/scripts/perl/bin/workqueue-stats-record b/tools/perf/scripts/perl/bin/workqueue-stats-record
new file mode 100644
index 00000000000..8edda9078d5
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/workqueue-stats-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -e workqueue:workqueue_creation -e workqueue:workqueue_destruction -e workqueue:workqueue_execution -e workqueue:workqueue_insertion $@
diff --git a/tools/perf/scripts/perl/bin/workqueue-stats-report b/tools/perf/scripts/perl/bin/workqueue-stats-report
new file mode 100644
index 00000000000..6d91411d248
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/workqueue-stats-report
@@ -0,0 +1,3 @@
+#!/bin/bash
+# description: workqueue stats (ins/exe/create/destroy)
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/workqueue-stats.pl
diff --git a/tools/perf/scripts/perl/check-perf-trace.pl b/tools/perf/scripts/perl/check-perf-trace.pl
new file mode 100644
index 00000000000..4e7076c2061
--- /dev/null
+++ b/tools/perf/scripts/perl/check-perf-trace.pl
@@ -0,0 +1,106 @@
+# perf script event handlers, generated by perf script -g perl
+# (c) 2009, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+
+# This script tests basic functionality such as flag and symbol
+# strings, common_xxx() calls back into perf, begin, end, unhandled
+# events, etc.  Basically, if this script runs successfully and
+# displays expected results, perl scripting support should be ok.
+
+use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib";
+use lib "./Perf-Trace-Util/lib";
+use Perf::Trace::Core;
+use Perf::Trace::Context;
+use Perf::Trace::Util;
+
+sub trace_begin
+{
+    print "trace_begin\n";
+}
+
+sub trace_end
+{
+    print "trace_end\n";
+
+    print_unhandled();
+}
+
+sub irq::softirq_entry
+{
+        my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+            $common_pid, $common_comm,
+            $vec) = @_;
+
+        print_header($event_name, $common_cpu, $common_secs, $common_nsecs,
+                     $common_pid, $common_comm);
+
+        print_uncommon($context);
+
+        printf("vec=%s\n",
+               symbol_str("irq::softirq_entry", "vec", $vec));
+}
+
+sub kmem::kmalloc
+{
+        my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+            $common_pid, $common_comm,
+            $call_site, $ptr, $bytes_req, $bytes_alloc,
+            $gfp_flags) = @_;
+
+        print_header($event_name, $common_cpu, $common_secs, $common_nsecs,
+                     $common_pid, $common_comm);
+
+        print_uncommon($context);
+
+        printf("call_site=%p, ptr=%p, bytes_req=%u, bytes_alloc=%u, ".
+               "gfp_flags=%s\n",
+               $call_site, $ptr, $bytes_req, $bytes_alloc,
+
+               flag_str("kmem::kmalloc", "gfp_flags", $gfp_flags));
+}
+
+# print trace fields not included in handler args
+sub print_uncommon
+{
+    my ($context) = @_;
+
+    printf("common_preempt_count=%d, common_flags=%s, common_lock_depth=%d, ",
+           common_pc($context), trace_flag_str(common_flags($context)),
+           common_lock_depth($context));
+
+}
+
+my %unhandled;
+
+sub print_unhandled
+{
+    if ((scalar keys %unhandled) == 0) {
+        return;
+    }
+
+    print "\nunhandled events:\n\n";
+
+    printf("%-40s  %10s\n", "event", "count");
+    printf("%-40s  %10s\n", "----------------------------------------",
+           "-----------");
+
+    foreach my $event_name (keys %unhandled) {
+        printf("%-40s  %10d\n", $event_name, $unhandled{$event_name});
+    }
+}
+
+sub trace_unhandled
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm) = @_;
+
+    $unhandled{$event_name}++;
+}
+
+sub print_header
+{
+        my ($event_name, $cpu, $secs, $nsecs, $pid, $comm) = @_;
+
+        printf("%-20s %5u %05u.%09u %8u %-20s ",
+               $event_name, $cpu, $secs, $nsecs, $pid, $comm);
+}
diff --git a/tools/perf/scripts/perl/failed-syscalls.pl b/tools/perf/scripts/perl/failed-syscalls.pl
new file mode 100644
index 00000000000..94bc25a347e
--- /dev/null
+++ b/tools/perf/scripts/perl/failed-syscalls.pl
@@ -0,0 +1,42 @@
+# failed system call counts
+# (c) 2010, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+#
+# Displays system-wide failed system call totals
+# If a [comm] arg is specified, only syscalls called by [comm] are displayed.
+
+use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib";
+use lib "./Perf-Trace-Util/lib";
+use Perf::Trace::Core;
+use Perf::Trace::Context;
+use Perf::Trace::Util;
+
+my $for_comm = shift;
+
+my %failed_syscalls;
+
+sub raw_syscalls::sys_exit
+{
+        my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+            $common_pid, $common_comm,
+            $id, $ret) = @_;
+
+        if ($ret < 0) {
+            $failed_syscalls{$common_comm}++;
+        }
+}
+
+sub trace_end
+{
+    printf("\nfailed syscalls by comm:\n\n");
+
+    printf("%-20s  %10s\n", "comm", "# errors");
+    printf("%-20s  %6s  %10s\n", "--------------------", "----------");
+
+    foreach my $comm (sort {$failed_syscalls{$b} <=> $failed_syscalls{$a}}
+                      keys %failed_syscalls) {
+        next if ($for_comm && $comm ne $for_comm);
+
+        printf("%-20s  %10s\n", $comm, $failed_syscalls{$comm});
+    }
+}
diff --git a/tools/perf/scripts/perl/rw-by-file.pl b/tools/perf/scripts/perl/rw-by-file.pl
new file mode 100644
index 00000000000..74844ee2be3
--- /dev/null
+++ b/tools/perf/scripts/perl/rw-by-file.pl
@@ -0,0 +1,106 @@
+#!/usr/bin/perl -w
+# (c) 2009, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+
+# Display r/w activity for files read/written to for a given program
+
+# The common_* event handler fields are the most useful fields common to
+# all events.  They don't necessarily correspond to the 'common_*' fields
+# in the status files.  Those fields not available as handler params can
+# be retrieved via script functions of the form get_common_*().
+
+use 5.010000;
+use strict;
+use warnings;
+
+use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib";
+use lib "./Perf-Trace-Util/lib";
+use Perf::Trace::Core;
+use Perf::Trace::Util;
+
+my $usage = "perf script -s rw-by-file.pl <comm>\n";
+
+my $for_comm = shift or die $usage;
+
+my %reads;
+my %writes;
+
+sub syscalls::sys_enter_read
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm, $nr, $fd, $buf, $count) = @_;
+
+    if ($common_comm eq $for_comm) {
+        $reads{$fd}{bytes_requested} += $count;
+        $reads{$fd}{total_reads}++;
+    }
+}
+
+sub syscalls::sys_enter_write
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm, $nr, $fd, $buf, $count) = @_;
+
+    if ($common_comm eq $for_comm) {
+        $writes{$fd}{bytes_written} += $count;
+        $writes{$fd}{total_writes}++;
+    }
+}
+
+sub trace_end
+{
+    printf("file read counts for $for_comm:\n\n");
+
+    printf("%6s  %10s  %10s\n", "fd", "# reads", "bytes_requested");
+    printf("%6s  %10s  %10s\n", "------", "----------", "-----------");
+
+    foreach my $fd (sort {$reads{$b}{bytes_requested} <=>
+                              $reads{$a}{bytes_requested}} keys %reads) {
+        my $total_reads = $reads{$fd}{total_reads};
+        my $bytes_requested = $reads{$fd}{bytes_requested};
+        printf("%6u  %10u  %10u\n", $fd, $total_reads, $bytes_requested);
+    }
+
+    printf("\nfile write counts for $for_comm:\n\n");
+
+    printf("%6s  %10s  %10s\n", "fd", "# writes", "bytes_written");
+    printf("%6s  %10s  %10s\n", "------", "----------", "-----------");
+
+    foreach my $fd (sort {$writes{$b}{bytes_written} <=>
+                              $writes{$a}{bytes_written}} keys %writes) {
+        my $total_writes = $writes{$fd}{total_writes};
+        my $bytes_written = $writes{$fd}{bytes_written};
+        printf("%6u  %10u  %10u\n", $fd, $total_writes, $bytes_written);
+    }
+
+    print_unhandled();
+}
+
+my %unhandled;
+
+sub print_unhandled
+{
+    if ((scalar keys %unhandled) == 0) {
+        return;
+    }
+
+    print "\nunhandled events:\n\n";
+
+    printf("%-40s  %10s\n", "event", "count");
+    printf("%-40s  %10s\n", "----------------------------------------",
+           "-----------");
+
+    foreach my $event_name (keys %unhandled) {
+        printf("%-40s  %10d\n", $event_name, $unhandled{$event_name});
+    }
+}
+
+sub trace_unhandled
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm) = @_;
+
+    $unhandled{$event_name}++;
+}
+
+
diff --git a/tools/perf/scripts/perl/rw-by-pid.pl b/tools/perf/scripts/perl/rw-by-pid.pl
new file mode 100644
index 00000000000..9db23c9daf5
--- /dev/null
+++ b/tools/perf/scripts/perl/rw-by-pid.pl
@@ -0,0 +1,184 @@
+#!/usr/bin/perl -w
+# (c) 2009, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+
+# Display r/w activity for all processes
+
+# The common_* event handler fields are the most useful fields common to
+# all events.  They don't necessarily correspond to the 'common_*' fields
+# in the status files.  Those fields not available as handler params can
+# be retrieved via script functions of the form get_common_*().
+
+use 5.010000;
+use strict;
+use warnings;
+
+use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib";
+use lib "./Perf-Trace-Util/lib";
+use Perf::Trace::Core;
+use Perf::Trace::Util;
+
+my %reads;
+my %writes;
+
+sub syscalls::sys_exit_read
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm,
+        $nr, $ret) = @_;
+
+    if ($ret > 0) {
+        $reads{$common_pid}{bytes_read} += $ret;
+    } else {
+        if (!defined ($reads{$common_pid}{bytes_read})) {
+            $reads{$common_pid}{bytes_read} = 0;
+        }
+        $reads{$common_pid}{errors}{$ret}++;
+    }
+}
+
+sub syscalls::sys_enter_read
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm,
+        $nr, $fd, $buf, $count) = @_;
+
+    $reads{$common_pid}{bytes_requested} += $count;
+    $reads{$common_pid}{total_reads}++;
+    $reads{$common_pid}{comm} = $common_comm;
+}
+
+sub syscalls::sys_exit_write
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm,
+        $nr, $ret) = @_;
+
+    if ($ret <= 0) {
+        $writes{$common_pid}{errors}{$ret}++;
+    }
+}
+
+sub syscalls::sys_enter_write
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm,
+        $nr, $fd, $buf, $count) = @_;
+
+    $writes{$common_pid}{bytes_written} += $count;
+    $writes{$common_pid}{total_writes}++;
+    $writes{$common_pid}{comm} = $common_comm;
+}
+
+sub trace_end
+{
+    printf("read counts by pid:\n\n");
+
+    printf("%6s  %20s  %10s  %10s  %10s\n", "pid", "comm",
+           "# reads", "bytes_requested", "bytes_read");
+    printf("%6s  %-20s  %10s  %10s  %10s\n", "------", "--------------------",
+           "-----------", "----------", "----------");
+
+    foreach my $pid (sort { ($reads{$b}{bytes_read} || 0) <=>
+                                ($reads{$a}{bytes_read} || 0) } keys %reads) {
+        my $comm = $reads{$pid}{comm} || "";
+        my $total_reads = $reads{$pid}{total_reads} || 0;
+        my $bytes_requested = $reads{$pid}{bytes_requested} || 0;
+        my $bytes_read = $reads{$pid}{bytes_read} || 0;
+
+        printf("%6s  %-20s  %10s  %10s  %10s\n", $pid, $comm,
+               $total_reads, $bytes_requested, $bytes_read);
+    }
+
+    printf("\nfailed reads by pid:\n\n");
+
+    printf("%6s  %20s  %6s  %10s\n", "pid", "comm", "error #", "# errors");
+    printf("%6s  %20s  %6s  %10s\n", "------", "--------------------",
+           "------", "----------");
+
+    my @errcounts = ();
+
+    foreach my $pid (keys %reads) {
+        foreach my $error (keys %{$reads{$pid}{errors}}) {
+            my $comm = $reads{$pid}{comm} || "";
+            my $errcount = $reads{$pid}{errors}{$error} || 0;
+            push @errcounts, [$pid, $comm, $error, $errcount];
+        }
+    }
+
+    @errcounts = sort { $b->[3] <=> $a->[3] } @errcounts;
+
+    for my $i (0 .. $#errcounts) {
+        printf("%6d  %-20s  %6d  %10s\n", $errcounts[$i][0],
+               $errcounts[$i][1], $errcounts[$i][2], $errcounts[$i][3]);
+    }
+
+    printf("\nwrite counts by pid:\n\n");
+
+    printf("%6s  %20s  %10s  %10s\n", "pid", "comm",
+           "# writes", "bytes_written");
+    printf("%6s  %-20s  %10s  %10s\n", "------", "--------------------",
+           "-----------", "----------");
+
+    foreach my $pid (sort { ($writes{$b}{bytes_written} || 0) <=>
+                        ($writes{$a}{bytes_written} || 0)} keys %writes) {
+        my $comm = $writes{$pid}{comm} || "";
+        my $total_writes = $writes{$pid}{total_writes} || 0;
+        my $bytes_written = $writes{$pid}{bytes_written} || 0;
+
+        printf("%6s  %-20s  %10s  %10s\n", $pid, $comm,
+               $total_writes, $bytes_written);
+    }
+
+    printf("\nfailed writes by pid:\n\n");
+
+    printf("%6s  %20s  %6s  %10s\n", "pid", "comm", "error #", "# errors");
+    printf("%6s  %20s  %6s  %10s\n", "------", "--------------------",
+           "------", "----------");
+
+    @errcounts = ();
+
+    foreach my $pid (keys %writes) {
+        foreach my $error (keys %{$writes{$pid}{errors}}) {
+            my $comm = $writes{$pid}{comm} || "";
+            my $errcount = $writes{$pid}{errors}{$error} || 0;
+            push @errcounts, [$pid, $comm, $error, $errcount];
+        }
+    }
+
+    @errcounts = sort { $b->[3] <=> $a->[3] } @errcounts;
+
+    for my $i (0 .. $#errcounts) {
+        printf("%6d  %-20s  %6d  %10s\n", $errcounts[$i][0],
+               $errcounts[$i][1], $errcounts[$i][2], $errcounts[$i][3]);
+    }
+
+    print_unhandled();
+}
+
+my %unhandled;
+
+sub print_unhandled
+{
+    if ((scalar keys %unhandled) == 0) {
+        return;
+    }
+
+    print "\nunhandled events:\n\n";
+
+    printf("%-40s  %10s\n", "event", "count");
+    printf("%-40s  %10s\n", "----------------------------------------",
+           "-----------");
+
+    foreach my $event_name (keys %unhandled) {
+        printf("%-40s  %10d\n", $event_name, $unhandled{$event_name});
+    }
+}
+
+sub trace_unhandled
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm) = @_;
+
+    $unhandled{$event_name}++;
+}
diff --git a/tools/perf/scripts/perl/rwtop.pl b/tools/perf/scripts/perl/rwtop.pl
new file mode 100644
index 00000000000..4bb3ecd3347
--- /dev/null
+++ b/tools/perf/scripts/perl/rwtop.pl
@@ -0,0 +1,199 @@
+#!/usr/bin/perl -w
+# (c) 2010, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+
+# read/write top
+#
+# Periodically displays system-wide r/w call activity, broken down by
+# pid.  If an [interval] arg is specified, the display will be
+# refreshed every [interval] seconds.  The default interval is 3
+# seconds.
+
+use 5.010000;
+use strict;
+use warnings;
+
+use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib";
+use lib "./Perf-Trace-Util/lib";
+use Perf::Trace::Core;
+use Perf::Trace::Util;
+
+my $default_interval = 3;
+my $nlines = 20;
+my $print_thread;
+my $print_pending = 0;
+
+my %reads;
+my %writes;
+
+my $interval = shift;
+if (!$interval) {
+    $interval = $default_interval;
+}
+
+sub syscalls::sys_exit_read
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm,
+        $nr, $ret) = @_;
+
+    print_check();
+
+    if ($ret > 0) {
+        $reads{$common_pid}{bytes_read} += $ret;
+    } else {
+        if (!defined ($reads{$common_pid}{bytes_read})) {
+            $reads{$common_pid}{bytes_read} = 0;
+        }
+        $reads{$common_pid}{errors}{$ret}++;
+    }
+}
+
+sub syscalls::sys_enter_read
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm,
+        $nr, $fd, $buf, $count) = @_;
+
+    print_check();
+
+    $reads{$common_pid}{bytes_requested} += $count;
+    $reads{$common_pid}{total_reads}++;
+    $reads{$common_pid}{comm} = $common_comm;
+}
+
+sub syscalls::sys_exit_write
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm,
+        $nr, $ret) = @_;
+
+    print_check();
+
+    if ($ret <= 0) {
+        $writes{$common_pid}{errors}{$ret}++;
+    }
+}
+
+sub syscalls::sys_enter_write
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm,
+        $nr, $fd, $buf, $count) = @_;
+
+    print_check();
+
+    $writes{$common_pid}{bytes_written} += $count;
+    $writes{$common_pid}{total_writes}++;
+    $writes{$common_pid}{comm} = $common_comm;
+}
+
+sub trace_begin
+{
+    $SIG{ALRM} = \&set_print_pending;
+    alarm 1;
+}
+
+sub trace_end
+{
+    print_unhandled();
+    print_totals();
+}
+
+sub print_check()
+{
+    if ($print_pending == 1) {
+        $print_pending = 0;
+        print_totals();
+    }
+}
+
+sub set_print_pending()
+{
+    $print_pending = 1;
+    alarm $interval;
+}
+
+sub print_totals
+{
+    my $count;
+
+    $count = 0;
+
+    clear_term();
+
+    printf("\nread counts by pid:\n\n");
+
+    printf("%6s  %20s  %10s  %10s  %10s\n", "pid", "comm",
+           "# reads", "bytes_req", "bytes_read");
+    printf("%6s  %-20s  %10s  %10s  %10s\n", "------", "--------------------",
+           "----------", "----------", "----------");
+
+    foreach my $pid (sort { ($reads{$b}{bytes_read} || 0) <=>
+                               ($reads{$a}{bytes_read} || 0) } keys %reads) {
+        my $comm = $reads{$pid}{comm} || "";
+        my $total_reads = $reads{$pid}{total_reads} || 0;
+        my $bytes_requested = $reads{$pid}{bytes_requested} || 0;
+        my $bytes_read = $reads{$pid}{bytes_read} || 0;
+
+        printf("%6s  %-20s  %10s  %10s  %10s\n", $pid, $comm,
+               $total_reads, $bytes_requested, $bytes_read);
+
+        if (++$count == $nlines) {
+            last;
+        }
+    }
+
+    $count = 0;
+
+    printf("\nwrite counts by pid:\n\n");
+
+    printf("%6s  %20s  %10s  %13s\n", "pid", "comm",
+           "# writes", "bytes_written");
+    printf("%6s  %-20s  %10s  %13s\n", "------", "--------------------",
+           "----------", "-------------");
+
+    foreach my $pid (sort { ($writes{$b}{bytes_written} || 0) <=>
+                        ($writes{$a}{bytes_written} || 0)} keys %writes) {
+        my $comm = $writes{$pid}{comm} || "";
+        my $total_writes = $writes{$pid}{total_writes} || 0;
+        my $bytes_written = $writes{$pid}{bytes_written} || 0;
+
+        printf("%6s  %-20s  %10s  %13s\n", $pid, $comm,
+               $total_writes, $bytes_written);
+
+        if (++$count == $nlines) {
+            last;
+        }
+    }
+
+    %reads = ();
+    %writes = ();
+}
+
+my %unhandled;
+
+sub print_unhandled
+{
+    if ((scalar keys %unhandled) == 0) {
+        return;
+    }
+
+    print "\nunhandled events:\n\n";
+
+    printf("%-40s  %10s\n", "event", "count");
+    printf("%-40s  %10s\n", "----------------------------------------",
+           "-----------");
+
+    foreach my $event_name (keys %unhandled) {
+        printf("%-40s  %10d\n", $event_name, $unhandled{$event_name});
+    }
+}
+
+sub trace_unhandled
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm) = @_;
+
+    $unhandled{$event_name}++;
+}
diff --git a/tools/perf/scripts/perl/wakeup-latency.pl b/tools/perf/scripts/perl/wakeup-latency.pl
new file mode 100644
index 00000000000..d9143dcec6c
--- /dev/null
+++ b/tools/perf/scripts/perl/wakeup-latency.pl
@@ -0,0 +1,107 @@
+#!/usr/bin/perl -w
+# (c) 2009, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+
+# Display avg/min/max wakeup latency
+
+# The common_* event handler fields are the most useful fields common to
+# all events.  They don't necessarily correspond to the 'common_*' fields
+# in the status files.  Those fields not available as handler params can
+# be retrieved via script functions of the form get_common_*().
+
+use 5.010000;
+use strict;
+use warnings;
+
+use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib";
+use lib "./Perf-Trace-Util/lib";
+use Perf::Trace::Core;
+use Perf::Trace::Util;
+
+my %last_wakeup;
+
+my $max_wakeup_latency;
+my $min_wakeup_latency;
+my $total_wakeup_latency = 0;
+my $total_wakeups = 0;
+
+sub sched::sched_switch
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm,
+        $prev_comm, $prev_pid, $prev_prio, $prev_state, $next_comm, $next_pid,
+        $next_prio) = @_;
+
+    my $wakeup_ts = $last_wakeup{$common_cpu}{ts};
+    if ($wakeup_ts) {
+        my $switch_ts = nsecs($common_secs, $common_nsecs);
+        my $wakeup_latency = $switch_ts - $wakeup_ts;
+        if ($wakeup_latency > $max_wakeup_latency) {
+            $max_wakeup_latency = $wakeup_latency;
+        }
+        if ($wakeup_latency < $min_wakeup_latency) {
+            $min_wakeup_latency = $wakeup_latency;
+        }
+        $total_wakeup_latency += $wakeup_latency;
+        $total_wakeups++;
+    }
+    $last_wakeup{$common_cpu}{ts} = 0;
+}
+
+sub sched::sched_wakeup
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm,
+        $comm, $pid, $prio, $success, $target_cpu) = @_;
+
+    $last_wakeup{$target_cpu}{ts} = nsecs($common_secs, $common_nsecs);
+}
+
+sub trace_begin
+{
+    $min_wakeup_latency = 1000000000;
+    $max_wakeup_latency = 0;
+}
+
+sub trace_end
+{
+    printf("wakeup_latency stats:\n\n");
+    print "total_wakeups: $total_wakeups\n";
+    if ($total_wakeups) {
+        printf("avg_wakeup_latency (ns): %u\n",
+               avg($total_wakeup_latency, $total_wakeups));
+    } else {
+        printf("avg_wakeup_latency (ns): N/A\n");
+    }
+    printf("min_wakeup_latency (ns): %u\n", $min_wakeup_latency);
+    printf("max_wakeup_latency (ns): %u\n", $max_wakeup_latency);
+
+    print_unhandled();
+}
+
+my %unhandled;
+
+sub print_unhandled
+{
+    if ((scalar keys %unhandled) == 0) {
+        return;
+    }
+
+    print "\nunhandled events:\n\n";
+
+    printf("%-40s  %10s\n", "event", "count");
+    printf("%-40s  %10s\n", "----------------------------------------",
+           "-----------");
+
+    foreach my $event_name (keys %unhandled) {
+        printf("%-40s  %10d\n", $event_name, $unhandled{$event_name});
+    }
+}
+
+sub trace_unhandled
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm) = @_;
+
+    $unhandled{$event_name}++;
+}
diff --git a/tools/perf/scripts/perl/workqueue-stats.pl b/tools/perf/scripts/perl/workqueue-stats.pl
new file mode 100644
index 00000000000..a8eaff5119e
--- /dev/null
+++ b/tools/perf/scripts/perl/workqueue-stats.pl
@@ -0,0 +1,129 @@
+#!/usr/bin/perl -w
+# (c) 2009, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+
+# Displays workqueue stats
+#
+# Usage:
+#
+#   perf record -c 1 -f -a -R -e workqueue:workqueue_creation -e
+#     workqueue:workqueue_destruction -e workqueue:workqueue_execution
+#     -e workqueue:workqueue_insertion
+#
+#   perf script -p -s tools/perf/scripts/perl/workqueue-stats.pl
+
+use 5.010000;
+use strict;
+use warnings;
+
+use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib";
+use lib "./Perf-Trace-Util/lib";
+use Perf::Trace::Core;
+use Perf::Trace::Util;
+
+my @cpus;
+
+sub workqueue::workqueue_destruction
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm,
+        $thread_comm, $thread_pid) = @_;
+
+    $cpus[$common_cpu]{$thread_pid}{destroyed}++;
+    $cpus[$common_cpu]{$thread_pid}{comm} = $thread_comm;
+}
+
+sub workqueue::workqueue_creation
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm,
+        $thread_comm, $thread_pid, $cpu) = @_;
+
+    $cpus[$common_cpu]{$thread_pid}{created}++;
+    $cpus[$common_cpu]{$thread_pid}{comm} = $thread_comm;
+}
+
+sub workqueue::workqueue_execution
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm,
+        $thread_comm, $thread_pid, $func) = @_;
+
+    $cpus[$common_cpu]{$thread_pid}{executed}++;
+    $cpus[$common_cpu]{$thread_pid}{comm} = $thread_comm;
+}
+
+sub workqueue::workqueue_insertion
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm,
+        $thread_comm, $thread_pid, $func) = @_;
+
+    $cpus[$common_cpu]{$thread_pid}{inserted}++;
+    $cpus[$common_cpu]{$thread_pid}{comm} = $thread_comm;
+}
+
+sub trace_end
+{
+    print "workqueue work stats:\n\n";
+    my $cpu = 0;
+    printf("%3s %6s %6s\t%-20s\n", "cpu", "ins", "exec", "name");
+    printf("%3s %6s %6s\t%-20s\n", "---", "---", "----", "----");
+    foreach my $pidhash (@cpus) {
+        while ((my $pid, my $wqhash) = each %$pidhash) {
+            my $ins = $$wqhash{'inserted'} || 0;
+            my $exe = $$wqhash{'executed'} || 0;
+            my $comm = $$wqhash{'comm'} || "";
+            if ($ins || $exe) {
+                printf("%3u %6u %6u\t%-20s\n", $cpu, $ins, $exe, $comm);
+            }
+        }
+        $cpu++;
+    }
+
+    $cpu = 0;
+    print "\nworkqueue lifecycle stats:\n\n";
+    printf("%3s %6s %6s\t%-20s\n", "cpu", "created", "destroyed", "name");
+    printf("%3s %6s %6s\t%-20s\n", "---", "-------", "---------", "----");
+    foreach my $pidhash (@cpus) {
+        while ((my $pid, my $wqhash) = each %$pidhash) {
+            my $created = $$wqhash{'created'} || 0;
+            my $destroyed = $$wqhash{'destroyed'} || 0;
+            my $comm = $$wqhash{'comm'} || "";
+            if ($created || $destroyed) {
+                printf("%3u %6u %6u\t%-20s\n", $cpu, $created, $destroyed,
+                       $comm);
+            }
+        }
+        $cpu++;
+    }
+
+    print_unhandled();
+}
+
+my %unhandled;
+
+sub print_unhandled
+{
+    if ((scalar keys %unhandled) == 0) {
+        return;
+    }
+
+    print "\nunhandled events:\n\n";
+
+    printf("%-40s  %10s\n", "event", "count");
+    printf("%-40s  %10s\n", "----------------------------------------",
+           "-----------");
+
+    foreach my $event_name (keys %unhandled) {
+        printf("%-40s  %10d\n", $event_name, $unhandled{$event_name});
+    }
+}
+
+sub trace_unhandled
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+        $common_pid, $common_comm) = @_;
+
+    $unhandled{$event_name}++;
+}
diff --git a/tools/perf/scripts/python/Perf-Trace-Util/Context.c b/tools/perf/scripts/python/Perf-Trace-Util/Context.c
new file mode 100644
index 00000000000..315067b8f55
--- /dev/null
+++ b/tools/perf/scripts/python/Perf-Trace-Util/Context.c
@@ -0,0 +1,88 @@
+/*
+ * Context.c.  Python interfaces for perf script.
+ *
+ * Copyright (C) 2010 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <Python.h>
+#include "../../../perf.h"
+#include "../../../util/trace-event.h"
+
+PyMODINIT_FUNC initperf_trace_context(void);
+
+static PyObject *perf_trace_context_common_pc(PyObject *self, PyObject *args)
+{
+        static struct scripting_context *scripting_context;
+        PyObject *context;
+        int retval;
+
+        if (!PyArg_ParseTuple(args, "O", &context))
+                return NULL;
+
+        scripting_context = PyCObject_AsVoidPtr(context);
+        retval = common_pc(scripting_context);
+
+        return Py_BuildValue("i", retval);
+}
+
+static PyObject *perf_trace_context_common_flags(PyObject *self,
+                                                 PyObject *args)
+{
+        static struct scripting_context *scripting_context;
+        PyObject *context;
+        int retval;
+
+        if (!PyArg_ParseTuple(args, "O", &context))
+                return NULL;
+
+        scripting_context = PyCObject_AsVoidPtr(context);
+        retval = common_flags(scripting_context);
+
+        return Py_BuildValue("i", retval);
+}
+
+static PyObject *perf_trace_context_common_lock_depth(PyObject *self,
+                                                      PyObject *args)
+{
+        static struct scripting_context *scripting_context;
+        PyObject *context;
+        int retval;
+
+        if (!PyArg_ParseTuple(args, "O", &context))
+                return NULL;
+
+        scripting_context = PyCObject_AsVoidPtr(context);
+        retval = common_lock_depth(scripting_context);
+
+        return Py_BuildValue("i", retval);
+}
+
+static PyMethodDef ContextMethods[] = {
+        { "common_pc", perf_trace_context_common_pc, METH_VARARGS,
+          "Get the common preempt count event field value."},
+        { "common_flags", perf_trace_context_common_flags, METH_VARARGS,
+          "Get the common flags event field value."},
+        { "common_lock_depth", perf_trace_context_common_lock_depth,
+          METH_VARARGS, "Get the common lock depth event field value."},
+        { NULL, NULL, 0, NULL}
+};
+
+PyMODINIT_FUNC initperf_trace_context(void)
+{
+        (void) Py_InitModule("perf_trace_context", ContextMethods);
+}
diff --git a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Core.py b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Core.py
new file mode 100644
index 00000000000..de7211e4fa4
--- /dev/null
+++ b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Core.py
@@ -0,0 +1,121 @@
+# Core.py - Python extension for perf script, core functions
+#
+# Copyright (C) 2010 by Tom Zanussi <tzanussi@gmail.com>
+#
+# This software may be distributed under the terms of the GNU General
+# Public License ("GPL") version 2 as published by the Free Software
+# Foundation.
+
+from collections import defaultdict
+
+def autodict():
+    return defaultdict(autodict)
+
+flag_fields = autodict()
+symbolic_fields = autodict()
+
+def define_flag_field(event_name, field_name, delim):
+    flag_fields[event_name][field_name]['delim'] = delim
+
+def define_flag_value(event_name, field_name, value, field_str):
+    flag_fields[event_name][field_name]['values'][value] = field_str
+
+def define_symbolic_field(event_name, field_name):
+    # nothing to do, really
+    pass
+
+def define_symbolic_value(event_name, field_name, value, field_str):
+    symbolic_fields[event_name][field_name]['values'][value] = field_str
+
+def flag_str(event_name, field_name, value):
+    string = ""
+
+    if flag_fields[event_name][field_name]:
+        print_delim = 0
+        keys = flag_fields[event_name][field_name]['values'].keys()
+        keys.sort()
+        for idx in keys:
+            if not value and not idx:
+                string += flag_fields[event_name][field_name]['values'][idx]
+                break
+            if idx and (value & idx) == idx:
+                if print_delim and flag_fields[event_name][field_name]['delim']:
+                    string += " " + flag_fields[event_name][field_name]['delim'] + " "
+                string += flag_fields[event_name][field_name]['values'][idx]
+                print_delim = 1
+                value &= ~idx
+
+    return string
+
+def symbol_str(event_name, field_name, value):
+    string = ""
+
+    if symbolic_fields[event_name][field_name]:
+        keys = symbolic_fields[event_name][field_name]['values'].keys()
+        keys.sort()
+        for idx in keys:
+            if not value and not idx:
+                string = symbolic_fields[event_name][field_name]['values'][idx]
+                break
+            if (value == idx):
+                string = symbolic_fields[event_name][field_name]['values'][idx]
+                break
+
+    return string
+
+trace_flags = { 0x00: "NONE", \
+                    0x01: "IRQS_OFF", \
+                    0x02: "IRQS_NOSUPPORT", \
+                    0x04: "NEED_RESCHED", \
+                    0x08: "HARDIRQ", \
+                    0x10: "SOFTIRQ" }
+
+def trace_flag_str(value):
+    string = ""
+    print_delim = 0
+
+    keys = trace_flags.keys()
+
+    for idx in keys:
+        if not value and not idx:
+            string += "NONE"
+            break
+
+        if idx and (value & idx) == idx:
+            if print_delim:
+                string += " | ";
+            string += trace_flags[idx]
+            print_delim = 1
+            value &= ~idx
+
+    return string
+
+
+def taskState(state):
+        states = {
+                0 : "R",
+                1 : "S",
+                2 : "D",
+                64: "DEAD"
+        }
+
+        if state not in states:
+                return "Unknown"
+
+        return states[state]
+
+
+class EventHeaders:
+        def __init__(self, common_cpu, common_secs, common_nsecs,
+                     common_pid, common_comm):
+                self.cpu = common_cpu
+                self.secs = common_secs
+                self.nsecs = common_nsecs
+                self.pid = common_pid
+                self.comm = common_comm
+
+        def ts(self):
+                return (self.secs * (10 ** 9)) + self.nsecs
+
+        def ts_format(self):
+                return "%d.%d" % (self.secs, int(self.nsecs / 1000))
diff --git a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/SchedGui.py b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/SchedGui.py
new file mode 100644
index 00000000000..fdd92f69905
--- /dev/null
+++ b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/SchedGui.py
@@ -0,0 +1,184 @@
+# SchedGui.py - Python extension for perf script, basic GUI code for
+#               traces drawing and overview.
+#
+# Copyright (C) 2010 by Frederic Weisbecker <fweisbec@gmail.com>
+#
+# This software is distributed under the terms of the GNU General
+# Public License ("GPL") version 2 as published by the Free Software
+# Foundation.
+
+
+try:
+        import wx
+except ImportError:
+        raise ImportError, "You need to install the wxpython lib for this script"
+
+
+class RootFrame(wx.Frame):
+        Y_OFFSET = 100
+        RECT_HEIGHT = 100
+        RECT_SPACE = 50
+        EVENT_MARKING_WIDTH = 5
+
+        def __init__(self, sched_tracer, title, parent = None, id = -1):
+                wx.Frame.__init__(self, parent, id, title)
+
+                (self.screen_width, self.screen_height) = wx.GetDisplaySize()
+                self.screen_width -= 10
+                self.screen_height -= 10
+                self.zoom = 0.5
+                self.scroll_scale = 20
+                self.sched_tracer = sched_tracer
+                self.sched_tracer.set_root_win(self)
+                (self.ts_start, self.ts_end) = sched_tracer.interval()
+                self.update_width_virtual()
+                self.nr_rects = sched_tracer.nr_rectangles() + 1
+                self.height_virtual = RootFrame.Y_OFFSET + (self.nr_rects * (RootFrame.RECT_HEIGHT + RootFrame.RECT_SPACE))
+
+                # whole window panel
+                self.panel = wx.Panel(self, size=(self.screen_width, self.screen_height))
+
+                # scrollable container
+                self.scroll = wx.ScrolledWindow(self.panel)
+                self.scroll.SetScrollbars(self.scroll_scale, self.scroll_scale, self.width_virtual / self.scroll_scale, self.height_virtual / self.scroll_scale)
+                self.scroll.EnableScrolling(True, True)
+                self.scroll.SetFocus()
+
+                # scrollable drawing area
+                self.scroll_panel = wx.Panel(self.scroll, size=(self.screen_width - 15, self.screen_height / 2))
+                self.scroll_panel.Bind(wx.EVT_PAINT, self.on_paint)
+                self.scroll_panel.Bind(wx.EVT_KEY_DOWN, self.on_key_press)
+                self.scroll_panel.Bind(wx.EVT_LEFT_DOWN, self.on_mouse_down)
+                self.scroll.Bind(wx.EVT_PAINT, self.on_paint)
+                self.scroll.Bind(wx.EVT_KEY_DOWN, self.on_key_press)
+                self.scroll.Bind(wx.EVT_LEFT_DOWN, self.on_mouse_down)
+
+                self.scroll.Fit()
+                self.Fit()
+
+                self.scroll_panel.SetDimensions(-1, -1, self.width_virtual, self.height_virtual, wx.SIZE_USE_EXISTING)
+
+                self.txt = None
+
+                self.Show(True)
+
+        def us_to_px(self, val):
+                return val / (10 ** 3) * self.zoom
+
+        def px_to_us(self, val):
+                return (val / self.zoom) * (10 ** 3)
+
+        def scroll_start(self):
+                (x, y) = self.scroll.GetViewStart()
+                return (x * self.scroll_scale, y * self.scroll_scale)
+
+        def scroll_start_us(self):
+                (x, y) = self.scroll_start()
+                return self.px_to_us(x)
+
+        def paint_rectangle_zone(self, nr, color, top_color, start, end):
+                offset_px = self.us_to_px(start - self.ts_start)
+                width_px = self.us_to_px(end - self.ts_start)
+
+                offset_py = RootFrame.Y_OFFSET + (nr * (RootFrame.RECT_HEIGHT + RootFrame.RECT_SPACE))
+                width_py = RootFrame.RECT_HEIGHT
+
+                dc = self.dc
+
+                if top_color is not None:
+                        (r, g, b) = top_color
+                        top_color = wx.Colour(r, g, b)
+                        brush = wx.Brush(top_color, wx.SOLID)
+                        dc.SetBrush(brush)
+                        dc.DrawRectangle(offset_px, offset_py, width_px, RootFrame.EVENT_MARKING_WIDTH)
+                        width_py -= RootFrame.EVENT_MARKING_WIDTH
+                        offset_py += RootFrame.EVENT_MARKING_WIDTH
+
+                (r ,g, b) = color
+                color = wx.Colour(r, g, b)
+                brush = wx.Brush(color, wx.SOLID)
+                dc.SetBrush(brush)
+                dc.DrawRectangle(offset_px, offset_py, width_px, width_py)
+
+        def update_rectangles(self, dc, start, end):
+                start += self.ts_start
+                end += self.ts_start
+                self.sched_tracer.fill_zone(start, end)
+
+        def on_paint(self, event):
+                dc = wx.PaintDC(self.scroll_panel)
+                self.dc = dc
+
+                width = min(self.width_virtual, self.screen_width)
+                (x, y) = self.scroll_start()
+                start = self.px_to_us(x)
+                end = self.px_to_us(x + width)
+                self.update_rectangles(dc, start, end)
+
+        def rect_from_ypixel(self, y):
+                y -= RootFrame.Y_OFFSET
+                rect = y / (RootFrame.RECT_HEIGHT + RootFrame.RECT_SPACE)
+                height = y % (RootFrame.RECT_HEIGHT + RootFrame.RECT_SPACE)
+
+                if rect < 0 or rect > self.nr_rects - 1 or height > RootFrame.RECT_HEIGHT:
+                        return -1
+
+                return rect
+
+        def update_summary(self, txt):
+                if self.txt:
+                        self.txt.Destroy()
+                self.txt = wx.StaticText(self.panel, -1, txt, (0, (self.screen_height / 2) + 50))
+
+
+        def on_mouse_down(self, event):
+                (x, y) = event.GetPositionTuple()
+                rect = self.rect_from_ypixel(y)
+                if rect == -1:
+                        return
+
+                t = self.px_to_us(x) + self.ts_start
+
+                self.sched_tracer.mouse_down(rect, t)
+
+
+        def update_width_virtual(self):
+                self.width_virtual = self.us_to_px(self.ts_end - self.ts_start)
+
+        def __zoom(self, x):
+                self.update_width_virtual()
+                (xpos, ypos) = self.scroll.GetViewStart()
+                xpos = self.us_to_px(x) / self.scroll_scale
+                self.scroll.SetScrollbars(self.scroll_scale, self.scroll_scale, self.width_virtual / self.scroll_scale, self.height_virtual / self.scroll_scale, xpos, ypos)
+                self.Refresh()
+
+        def zoom_in(self):
+                x = self.scroll_start_us()
+                self.zoom *= 2
+                self.__zoom(x)
+
+        def zoom_out(self):
+                x = self.scroll_start_us()
+                self.zoom /= 2
+                self.__zoom(x)
+
+
+        def on_key_press(self, event):
+                key = event.GetRawKeyCode()
+                if key == ord("+"):
+                        self.zoom_in()
+                        return
+                if key == ord("-"):
+                        self.zoom_out()
+                        return
+
+                key = event.GetKeyCode()
+                (x, y) = self.scroll.GetViewStart()
+                if key == wx.WXK_RIGHT:
+                        self.scroll.Scroll(x + 1, y)
+                elif key == wx.WXK_LEFT:
+                        self.scroll.Scroll(x - 1, y)
+                elif key == wx.WXK_DOWN:
+                        self.scroll.Scroll(x, y + 1)
+                elif key == wx.WXK_UP:
+                        self.scroll.Scroll(x, y - 1)
diff --git a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py
new file mode 100644
index 00000000000..15c8400240f
--- /dev/null
+++ b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py
@@ -0,0 +1,86 @@
+# Util.py - Python extension for perf script, miscellaneous utility code
+#
+# Copyright (C) 2010 by Tom Zanussi <tzanussi@gmail.com>
+#
+# This software may be distributed under the terms of the GNU General
+# Public License ("GPL") version 2 as published by the Free Software
+# Foundation.
+
+import errno, os
+
+FUTEX_WAIT = 0
+FUTEX_WAKE = 1
+FUTEX_PRIVATE_FLAG = 128
+FUTEX_CLOCK_REALTIME = 256
+FUTEX_CMD_MASK = ~(FUTEX_PRIVATE_FLAG | FUTEX_CLOCK_REALTIME)
+
+NSECS_PER_SEC    = 1000000000
+
+def avg(total, n):
+    return total / n
+
+def nsecs(secs, nsecs):
+    return secs * NSECS_PER_SEC + nsecs
+
+def nsecs_secs(nsecs):
+    return nsecs / NSECS_PER_SEC
+
+def nsecs_nsecs(nsecs):
+    return nsecs % NSECS_PER_SEC
+
+def nsecs_str(nsecs):
+    str = "%5u.%09u" % (nsecs_secs(nsecs), nsecs_nsecs(nsecs)),
+    return str
+
+def add_stats(dict, key, value):
+        if not dict.has_key(key):
+                dict[key] = (value, value, value, 1)
+        else:
+                min, max, avg, count = dict[key]
+                if value < min:
+                        min = value
+                if value > max:
+                        max = value
+                avg = (avg + value) / 2
+                dict[key] = (min, max, avg, count + 1)
+
+def clear_term():
+    print("\x1b[H\x1b[2J")
+
+audit_package_warned = False
+
+try:
+        import audit
+        machine_to_id = {
+                'x86_64': audit.MACH_86_64,
+                'alpha' : audit.MACH_ALPHA,
+                'ia64'  : audit.MACH_IA64,
+                'ppc'   : audit.MACH_PPC,
+                'ppc64' : audit.MACH_PPC64,
+                's390'  : audit.MACH_S390,
+                's390x' : audit.MACH_S390X,
+                'i386'  : audit.MACH_X86,
+                'i586'  : audit.MACH_X86,
+                'i686'  : audit.MACH_X86,
+        }
+        try:
+                machine_to_id['armeb'] = audit.MACH_ARMEB
+        except:
+                pass
+        machine_id = machine_to_id[os.uname()[4]]
+except:
+        if not audit_package_warned:
+                audit_package_warned = True
+                print "Install the audit-libs-python package to get syscall names"
+
+def syscall_name(id):
+        try:
+                return audit.audit_syscall_to_name(id, machine_id)
+        except:
+                return str(id)
+
+def strerror(nr):
+        try:
+                return errno.errorcode[abs(nr)]
+        except:
+                return "Unknown %d errno" % nr
diff --git a/tools/perf/scripts/python/bin/failed-syscalls-by-pid-record b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-record
new file mode 100644
index 00000000000..8104895a7b6
--- /dev/null
+++ b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -e raw_syscalls:sys_exit $@
diff --git a/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report
new file mode 100644
index 00000000000..fda5096d0cb
--- /dev/null
+++ b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report
@@ -0,0 +1,10 @@
+#!/bin/bash
+# description: system-wide failed syscalls, by pid
+# args: [comm]
+if [ $# -gt 0 ] ; then
+    if ! expr match "$1" "-" > /dev/null ; then
+        comm=$1
+        shift
+    fi
+fi
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/failed-syscalls-by-pid.py $comm
diff --git a/tools/perf/scripts/python/bin/futex-contention-record b/tools/perf/scripts/python/bin/futex-contention-record
new file mode 100644
index 00000000000..b1495c9a9b2
--- /dev/null
+++ b/tools/perf/scripts/python/bin/futex-contention-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -e syscalls:sys_enter_futex -e syscalls:sys_exit_futex $@
diff --git a/tools/perf/scripts/python/bin/futex-contention-report b/tools/perf/scripts/python/bin/futex-contention-report
new file mode 100644
index 00000000000..6c44271091a
--- /dev/null
+++ b/tools/perf/scripts/python/bin/futex-contention-report
@@ -0,0 +1,4 @@
+#!/bin/bash
+# description: futext contention measurement
+
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/futex-contention.py
diff --git a/tools/perf/scripts/python/bin/netdev-times-record b/tools/perf/scripts/python/bin/netdev-times-record
new file mode 100644
index 00000000000..558754b840a
--- /dev/null
+++ b/tools/perf/scripts/python/bin/netdev-times-record
@@ -0,0 +1,8 @@
+#!/bin/bash
+perf record -e net:net_dev_xmit -e net:net_dev_queue            \
+                -e net:netif_receive_skb -e net:netif_rx                \
+                -e skb:consume_skb -e skb:kfree_skb                     \
+                -e skb:skb_copy_datagram_iovec -e napi:napi_poll        \
+                -e irq:irq_handler_entry -e irq:irq_handler_exit        \
+                -e irq:softirq_entry -e irq:softirq_exit                \
+                -e irq:softirq_raise $@
diff --git a/tools/perf/scripts/python/bin/netdev-times-report b/tools/perf/scripts/python/bin/netdev-times-report
new file mode 100644
index 00000000000..8f759291da8
--- /dev/null
+++ b/tools/perf/scripts/python/bin/netdev-times-report
@@ -0,0 +1,5 @@
+#!/bin/bash
+# description: display a process of packet and processing time
+# args: [tx] [rx] [dev=] [debug]
+
+perf script -s "$PERF_EXEC_PATH"/scripts/python/netdev-times.py $@
diff --git a/tools/perf/scripts/python/bin/sched-migration-record b/tools/perf/scripts/python/bin/sched-migration-record
new file mode 100644
index 00000000000..7493fddbe99
--- /dev/null
+++ b/tools/perf/scripts/python/bin/sched-migration-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -m 16384 -e sched:sched_wakeup -e sched:sched_wakeup_new -e sched:sched_switch -e sched:sched_migrate_task $@
diff --git a/tools/perf/scripts/python/bin/sched-migration-report b/tools/perf/scripts/python/bin/sched-migration-report
new file mode 100644
index 00000000000..68b037a1849
--- /dev/null
+++ b/tools/perf/scripts/python/bin/sched-migration-report
@@ -0,0 +1,3 @@
+#!/bin/bash
+# description: sched migration overview
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/sched-migration.py
diff --git a/tools/perf/scripts/python/bin/sctop-record b/tools/perf/scripts/python/bin/sctop-record
new file mode 100644
index 00000000000..4efbfaa7f6a
--- /dev/null
+++ b/tools/perf/scripts/python/bin/sctop-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -e raw_syscalls:sys_enter $@
diff --git a/tools/perf/scripts/python/bin/sctop-report b/tools/perf/scripts/python/bin/sctop-report
new file mode 100644
index 00000000000..c32db294124
--- /dev/null
+++ b/tools/perf/scripts/python/bin/sctop-report
@@ -0,0 +1,24 @@
+#!/bin/bash
+# description: syscall top
+# args: [comm] [interval]
+n_args=0
+for i in "$@"
+do
+    if expr match "$i" "-" > /dev/null ; then
+        break
+    fi
+    n_args=$(( $n_args + 1 ))
+done
+if [ "$n_args" -gt 2 ] ; then
+    echo "usage: sctop-report [comm] [interval]"
+    exit
+fi
+if [ "$n_args" -gt 1 ] ; then
+    comm=$1
+    interval=$2
+    shift 2
+elif [ "$n_args" -gt 0 ] ; then
+    interval=$1
+    shift
+fi
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/sctop.py $comm $interval
diff --git a/tools/perf/scripts/python/bin/syscall-counts-by-pid-record b/tools/perf/scripts/python/bin/syscall-counts-by-pid-record
new file mode 100644
index 00000000000..4efbfaa7f6a
--- /dev/null
+++ b/tools/perf/scripts/python/bin/syscall-counts-by-pid-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -e raw_syscalls:sys_enter $@
diff --git a/tools/perf/scripts/python/bin/syscall-counts-by-pid-report b/tools/perf/scripts/python/bin/syscall-counts-by-pid-report
new file mode 100644
index 00000000000..16eb8d65c54
--- /dev/null
+++ b/tools/perf/scripts/python/bin/syscall-counts-by-pid-report
@@ -0,0 +1,10 @@
+#!/bin/bash
+# description: system-wide syscall counts, by pid
+# args: [comm]
+if [ $# -gt 0 ] ; then
+    if ! expr match "$1" "-" > /dev/null ; then
+        comm=$1
+        shift
+    fi
+fi
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/syscall-counts-by-pid.py $comm
diff --git a/tools/perf/scripts/python/bin/syscall-counts-record b/tools/perf/scripts/python/bin/syscall-counts-record
new file mode 100644
index 00000000000..4efbfaa7f6a
--- /dev/null
+++ b/tools/perf/scripts/python/bin/syscall-counts-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -e raw_syscalls:sys_enter $@
diff --git a/tools/perf/scripts/python/bin/syscall-counts-report b/tools/perf/scripts/python/bin/syscall-counts-report
new file mode 100644
index 00000000000..0f0e9d453bb
--- /dev/null
+++ b/tools/perf/scripts/python/bin/syscall-counts-report
@@ -0,0 +1,10 @@
+#!/bin/bash
+# description: system-wide syscall counts
+# args: [comm]
+if [ $# -gt 0 ] ; then
+    if ! expr match "$1" "-" > /dev/null ; then
+        comm=$1
+        shift
+    fi
+fi
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/syscall-counts.py $comm
diff --git a/tools/perf/scripts/python/check-perf-trace.py b/tools/perf/scripts/python/check-perf-trace.py
new file mode 100644
index 00000000000..4647a7694cf
--- /dev/null
+++ b/tools/perf/scripts/python/check-perf-trace.py
@@ -0,0 +1,82 @@
+# perf script event handlers, generated by perf script -g python
+# (c) 2010, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+#
+# This script tests basic functionality such as flag and symbol
+# strings, common_xxx() calls back into perf, begin, end, unhandled
+# events, etc.  Basically, if this script runs successfully and
+# displays expected results, Python scripting support should be ok.
+
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+        '/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from Core import *
+from perf_trace_context import *
+
+unhandled = autodict()
+
+def trace_begin():
+        print "trace_begin"
+        pass
+
+def trace_end():
+        print_unhandled()
+
+def irq__softirq_entry(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        vec):
+                print_header(event_name, common_cpu, common_secs, common_nsecs,
+                        common_pid, common_comm)
+
+                print_uncommon(context)
+
+                print "vec=%s\n" % \
+                (symbol_str("irq__softirq_entry", "vec", vec)),
+
+def kmem__kmalloc(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        call_site, ptr, bytes_req, bytes_alloc,
+        gfp_flags):
+                print_header(event_name, common_cpu, common_secs, common_nsecs,
+                        common_pid, common_comm)
+
+                print_uncommon(context)
+
+                print "call_site=%u, ptr=%u, bytes_req=%u, " \
+                "bytes_alloc=%u, gfp_flags=%s\n" % \
+                (call_site, ptr, bytes_req, bytes_alloc,
+
+                flag_str("kmem__kmalloc", "gfp_flags", gfp_flags)),
+
+def trace_unhandled(event_name, context, event_fields_dict):
+    try:
+        unhandled[event_name] += 1
+    except TypeError:
+        unhandled[event_name] = 1
+
+def print_header(event_name, cpu, secs, nsecs, pid, comm):
+        print "%-20s %5u %05u.%09u %8u %-20s " % \
+        (event_name, cpu, secs, nsecs, pid, comm),
+
+# print trace fields not included in handler args
+def print_uncommon(context):
+    print "common_preempt_count=%d, common_flags=%s, common_lock_depth=%d, " \
+        % (common_pc(context), trace_flag_str(common_flags(context)), \
+               common_lock_depth(context))
+
+def print_unhandled():
+    keys = unhandled.keys()
+    if not keys:
+        return
+
+    print "\nunhandled events:\n\n",
+
+    print "%-40s  %10s\n" % ("event", "count"),
+    print "%-40s  %10s\n" % ("----------------------------------------", \
+                                 "-----------"),
+
+    for event_name in keys:
+        print "%-40s  %10d\n" % (event_name, unhandled[event_name])
diff --git a/tools/perf/scripts/python/failed-syscalls-by-pid.py b/tools/perf/scripts/python/failed-syscalls-by-pid.py
new file mode 100644
index 00000000000..85805fac411
--- /dev/null
+++ b/tools/perf/scripts/python/failed-syscalls-by-pid.py
@@ -0,0 +1,73 @@
+# failed system call counts, by pid
+# (c) 2010, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+#
+# Displays system-wide failed system call totals, broken down by pid.
+# If a [comm] arg is specified, only syscalls called by [comm] are displayed.
+
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+        '/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from Util import *
+
+usage = "perf script -s syscall-counts-by-pid.py [comm|pid]\n";
+
+for_comm = None
+for_pid = None
+
+if len(sys.argv) > 2:
+        sys.exit(usage)
+
+if len(sys.argv) > 1:
+        try:
+                for_pid = int(sys.argv[1])
+        except:
+                for_comm = sys.argv[1]
+
+syscalls = autodict()
+
+def trace_begin():
+        print "Press control+C to stop and show the summary"
+
+def trace_end():
+        print_error_totals()
+
+def raw_syscalls__sys_exit(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        id, ret):
+        if (for_comm and common_comm != for_comm) or \
+           (for_pid  and common_pid  != for_pid ):
+                return
+
+        if ret < 0:
+                try:
+                        syscalls[common_comm][common_pid][id][ret] += 1
+                except TypeError:
+                        syscalls[common_comm][common_pid][id][ret] = 1
+
+def print_error_totals():
+    if for_comm is not None:
+            print "\nsyscall errors for %s:\n\n" % (for_comm),
+    else:
+            print "\nsyscall errors:\n\n",
+
+    print "%-30s  %10s\n" % ("comm [pid]", "count"),
+    print "%-30s  %10s\n" % ("------------------------------", \
+                                 "----------"),
+
+    comm_keys = syscalls.keys()
+    for comm in comm_keys:
+            pid_keys = syscalls[comm].keys()
+            for pid in pid_keys:
+                    print "\n%s [%d]\n" % (comm, pid),
+                    id_keys = syscalls[comm][pid].keys()
+                    for id in id_keys:
+                            print "  syscall: %-16s\n" % syscall_name(id),
+                            ret_keys = syscalls[comm][pid][id].keys()
+                            for ret, val in sorted(syscalls[comm][pid][id].iteritems(), key = lambda(k, v): (v, k),  reverse = True):
+                                    print "    err = %-20s  %10d\n" % (strerror(ret), val),
diff --git a/tools/perf/scripts/python/futex-contention.py b/tools/perf/scripts/python/futex-contention.py
new file mode 100644
index 00000000000..11e70a388d4
--- /dev/null
+++ b/tools/perf/scripts/python/futex-contention.py
@@ -0,0 +1,50 @@
+# futex contention
+# (c) 2010, Arnaldo Carvalho de Melo <acme@redhat.com>
+# Licensed under the terms of the GNU GPL License version 2
+#
+# Translation of:
+#
+# http://sourceware.org/systemtap/wiki/WSFutexContention
+#
+# to perf python scripting.
+#
+# Measures futex contention
+
+import os, sys
+sys.path.append(os.environ['PERF_EXEC_PATH'] + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+from Util import *
+
+process_names = {}
+thread_thislock = {}
+thread_blocktime = {}
+
+lock_waits = {} # long-lived stats on (tid,lock) blockage elapsed time
+process_names = {} # long-lived pid-to-execname mapping
+
+def syscalls__sys_enter_futex(event, ctxt, cpu, s, ns, tid, comm,
+                              nr, uaddr, op, val, utime, uaddr2, val3):
+        cmd = op & FUTEX_CMD_MASK
+        if cmd != FUTEX_WAIT:
+                return # we don't care about originators of WAKE events
+
+        process_names[tid] = comm
+        thread_thislock[tid] = uaddr
+        thread_blocktime[tid] = nsecs(s, ns)
+
+def syscalls__sys_exit_futex(event, ctxt, cpu, s, ns, tid, comm,
+                             nr, ret):
+        if thread_blocktime.has_key(tid):
+                elapsed = nsecs(s, ns) - thread_blocktime[tid]
+                add_stats(lock_waits, (tid, thread_thislock[tid]), elapsed)
+                del thread_blocktime[tid]
+                del thread_thislock[tid]
+
+def trace_begin():
+        print "Press control+C to stop and show the summary"
+
+def trace_end():
+        for (tid, lock) in lock_waits:
+                min, max, avg, count = lock_waits[tid, lock]
+                print "%s[%d] lock %x contended %d times, %d avg ns" % \
+                      (process_names[tid], tid, lock, count, avg)
+
diff --git a/tools/perf/scripts/python/netdev-times.py b/tools/perf/scripts/python/netdev-times.py
new file mode 100644
index 00000000000..9aa0a32972e
--- /dev/null
+++ b/tools/perf/scripts/python/netdev-times.py
@@ -0,0 +1,464 @@
+# Display a process of packets and processed time.
+# It helps us to investigate networking or network device.
+#
+# options
+# tx: show only tx chart
+# rx: show only rx chart
+# dev=: show only thing related to specified device
+# debug: work with debug mode. It shows buffer status.
+
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+        '/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from Util import *
+
+all_event_list = []; # insert all tracepoint event related with this script
+irq_dic = {}; # key is cpu and value is a list which stacks irqs
+              # which raise NET_RX softirq
+net_rx_dic = {}; # key is cpu and value include time of NET_RX softirq-entry
+                 # and a list which stacks receive
+receive_hunk_list = []; # a list which include a sequence of receive events
+rx_skb_list = []; # received packet list for matching
+                       # skb_copy_datagram_iovec
+
+buffer_budget = 65536; # the budget of rx_skb_list, tx_queue_list and
+                       # tx_xmit_list
+of_count_rx_skb_list = 0; # overflow count
+
+tx_queue_list = []; # list of packets which pass through dev_queue_xmit
+of_count_tx_queue_list = 0; # overflow count
+
+tx_xmit_list = [];  # list of packets which pass through dev_hard_start_xmit
+of_count_tx_xmit_list = 0; # overflow count
+
+tx_free_list = [];  # list of packets which is freed
+
+# options
+show_tx = 0;
+show_rx = 0;
+dev = 0; # store a name of device specified by option "dev="
+debug = 0;
+
+# indices of event_info tuple
+EINFO_IDX_NAME=   0
+EINFO_IDX_CONTEXT=1
+EINFO_IDX_CPU=    2
+EINFO_IDX_TIME=   3
+EINFO_IDX_PID=    4
+EINFO_IDX_COMM=   5
+
+# Calculate a time interval(msec) from src(nsec) to dst(nsec)
+def diff_msec(src, dst):
+        return (dst - src) / 1000000.0
+
+# Display a process of transmitting a packet
+def print_transmit(hunk):
+        if dev != 0 and hunk['dev'].find(dev) < 0:
+                return
+        print "%7s %5d %6d.%06dsec %12.3fmsec      %12.3fmsec" % \
+                (hunk['dev'], hunk['len'],
+                nsecs_secs(hunk['queue_t']),
+                nsecs_nsecs(hunk['queue_t'])/1000,
+                diff_msec(hunk['queue_t'], hunk['xmit_t']),
+                diff_msec(hunk['xmit_t'], hunk['free_t']))
+
+# Format for displaying rx packet processing
+PF_IRQ_ENTRY= "  irq_entry(+%.3fmsec irq=%d:%s)"
+PF_SOFT_ENTRY="  softirq_entry(+%.3fmsec)"
+PF_NAPI_POLL= "  napi_poll_exit(+%.3fmsec %s)"
+PF_JOINT=     "         |"
+PF_WJOINT=    "         |            |"
+PF_NET_RECV=  "         |---netif_receive_skb(+%.3fmsec skb=%x len=%d)"
+PF_NET_RX=    "         |---netif_rx(+%.3fmsec skb=%x)"
+PF_CPY_DGRAM= "         |      skb_copy_datagram_iovec(+%.3fmsec %d:%s)"
+PF_KFREE_SKB= "         |      kfree_skb(+%.3fmsec location=%x)"
+PF_CONS_SKB=  "         |      consume_skb(+%.3fmsec)"
+
+# Display a process of received packets and interrputs associated with
+# a NET_RX softirq
+def print_receive(hunk):
+        show_hunk = 0
+        irq_list = hunk['irq_list']
+        cpu = irq_list[0]['cpu']
+        base_t = irq_list[0]['irq_ent_t']
+        # check if this hunk should be showed
+        if dev != 0:
+                for i in range(len(irq_list)):
+                        if irq_list[i]['name'].find(dev) >= 0:
+                                show_hunk = 1
+                                break
+        else:
+                show_hunk = 1
+        if show_hunk == 0:
+                return
+
+        print "%d.%06dsec cpu=%d" % \
+                (nsecs_secs(base_t), nsecs_nsecs(base_t)/1000, cpu)
+        for i in range(len(irq_list)):
+                print PF_IRQ_ENTRY % \
+                        (diff_msec(base_t, irq_list[i]['irq_ent_t']),
+                        irq_list[i]['irq'], irq_list[i]['name'])
+                print PF_JOINT
+                irq_event_list = irq_list[i]['event_list']
+                for j in range(len(irq_event_list)):
+                        irq_event = irq_event_list[j]
+                        if irq_event['event'] == 'netif_rx':
+                                print PF_NET_RX % \
+                                        (diff_msec(base_t, irq_event['time']),
+                                        irq_event['skbaddr'])
+                                print PF_JOINT
+        print PF_SOFT_ENTRY % \
+                diff_msec(base_t, hunk['sirq_ent_t'])
+        print PF_JOINT
+        event_list = hunk['event_list']
+        for i in range(len(event_list)):
+                event = event_list[i]
+                if event['event_name'] == 'napi_poll':
+                        print PF_NAPI_POLL % \
+                            (diff_msec(base_t, event['event_t']), event['dev'])
+                        if i == len(event_list) - 1:
+                                print ""
+                        else:
+                                print PF_JOINT
+                else:
+                        print PF_NET_RECV % \
+                            (diff_msec(base_t, event['event_t']), event['skbaddr'],
+                                event['len'])
+                        if 'comm' in event.keys():
+                                print PF_WJOINT
+                                print PF_CPY_DGRAM % \
+                                        (diff_msec(base_t, event['comm_t']),
+                                        event['pid'], event['comm'])
+                        elif 'handle' in event.keys():
+                                print PF_WJOINT
+                                if event['handle'] == "kfree_skb":
+                                        print PF_KFREE_SKB % \
+                                                (diff_msec(base_t,
+                                                event['comm_t']),
+                                                event['location'])
+                                elif event['handle'] == "consume_skb":
+                                        print PF_CONS_SKB % \
+                                                diff_msec(base_t,
+                                                        event['comm_t'])
+                        print PF_JOINT
+
+def trace_begin():
+        global show_tx
+        global show_rx
+        global dev
+        global debug
+
+        for i in range(len(sys.argv)):
+                if i == 0:
+                        continue
+                arg = sys.argv[i]
+                if arg == 'tx':
+                        show_tx = 1
+                elif arg =='rx':
+                        show_rx = 1
+                elif arg.find('dev=',0, 4) >= 0:
+                        dev = arg[4:]
+                elif arg == 'debug':
+                        debug = 1
+        if show_tx == 0  and show_rx == 0:
+                show_tx = 1
+                show_rx = 1
+
+def trace_end():
+        # order all events in time
+        all_event_list.sort(lambda a,b :cmp(a[EINFO_IDX_TIME],
+                                            b[EINFO_IDX_TIME]))
+        # process all events
+        for i in range(len(all_event_list)):
+                event_info = all_event_list[i]
+                name = event_info[EINFO_IDX_NAME]
+                if name == 'irq__softirq_exit':
+                        handle_irq_softirq_exit(event_info)
+                elif name == 'irq__softirq_entry':
+                        handle_irq_softirq_entry(event_info)
+                elif name == 'irq__softirq_raise':
+                        handle_irq_softirq_raise(event_info)
+                elif name == 'irq__irq_handler_entry':
+                        handle_irq_handler_entry(event_info)
+                elif name == 'irq__irq_handler_exit':
+                        handle_irq_handler_exit(event_info)
+                elif name == 'napi__napi_poll':
+                        handle_napi_poll(event_info)
+                elif name == 'net__netif_receive_skb':
+                        handle_netif_receive_skb(event_info)
+                elif name == 'net__netif_rx':
+                        handle_netif_rx(event_info)
+                elif name == 'skb__skb_copy_datagram_iovec':
+                        handle_skb_copy_datagram_iovec(event_info)
+                elif name == 'net__net_dev_queue':
+                        handle_net_dev_queue(event_info)
+                elif name == 'net__net_dev_xmit':
+                        handle_net_dev_xmit(event_info)
+                elif name == 'skb__kfree_skb':
+                        handle_kfree_skb(event_info)
+                elif name == 'skb__consume_skb':
+                        handle_consume_skb(event_info)
+        # display receive hunks
+        if show_rx:
+                for i in range(len(receive_hunk_list)):
+                        print_receive(receive_hunk_list[i])
+        # display transmit hunks
+        if show_tx:
+                print "   dev    len      Qdisc        " \
+                        "       netdevice             free"
+                for i in range(len(tx_free_list)):
+                        print_transmit(tx_free_list[i])
+        if debug:
+                print "debug buffer status"
+                print "----------------------------"
+                print "xmit Qdisc:remain:%d overflow:%d" % \
+                        (len(tx_queue_list), of_count_tx_queue_list)
+                print "xmit netdevice:remain:%d overflow:%d" % \
+                        (len(tx_xmit_list), of_count_tx_xmit_list)
+                print "receive:remain:%d overflow:%d" % \
+                        (len(rx_skb_list), of_count_rx_skb_list)
+
+# called from perf, when it finds a correspoinding event
+def irq__softirq_entry(name, context, cpu, sec, nsec, pid, comm, vec):
+        if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX":
+                return
+        event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec)
+        all_event_list.append(event_info)
+
+def irq__softirq_exit(name, context, cpu, sec, nsec, pid, comm, vec):
+        if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX":
+                return
+        event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec)
+        all_event_list.append(event_info)
+
+def irq__softirq_raise(name, context, cpu, sec, nsec, pid, comm, vec):
+        if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX":
+                return
+        event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec)
+        all_event_list.append(event_info)
+
+def irq__irq_handler_entry(name, context, cpu, sec, nsec, pid, comm,
+                        irq, irq_name):
+        event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+                        irq, irq_name)
+        all_event_list.append(event_info)
+
+def irq__irq_handler_exit(name, context, cpu, sec, nsec, pid, comm, irq, ret):
+        event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, irq, ret)
+        all_event_list.append(event_info)
+
+def napi__napi_poll(name, context, cpu, sec, nsec, pid, comm, napi, dev_name):
+        event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+                        napi, dev_name)
+        all_event_list.append(event_info)
+
+def net__netif_receive_skb(name, context, cpu, sec, nsec, pid, comm, skbaddr,
+                        skblen, dev_name):
+        event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+                        skbaddr, skblen, dev_name)
+        all_event_list.append(event_info)
+
+def net__netif_rx(name, context, cpu, sec, nsec, pid, comm, skbaddr,
+                        skblen, dev_name):
+        event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+                        skbaddr, skblen, dev_name)
+        all_event_list.append(event_info)
+
+def net__net_dev_queue(name, context, cpu, sec, nsec, pid, comm,
+                        skbaddr, skblen, dev_name):
+        event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+                        skbaddr, skblen, dev_name)
+        all_event_list.append(event_info)
+
+def net__net_dev_xmit(name, context, cpu, sec, nsec, pid, comm,
+                        skbaddr, skblen, rc, dev_name):
+        event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+                        skbaddr, skblen, rc ,dev_name)
+        all_event_list.append(event_info)
+
+def skb__kfree_skb(name, context, cpu, sec, nsec, pid, comm,
+                        skbaddr, protocol, location):
+        event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+                        skbaddr, protocol, location)
+        all_event_list.append(event_info)
+
+def skb__consume_skb(name, context, cpu, sec, nsec, pid, comm, skbaddr):
+        event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+                        skbaddr)
+        all_event_list.append(event_info)
+
+def skb__skb_copy_datagram_iovec(name, context, cpu, sec, nsec, pid, comm,
+        skbaddr, skblen):
+        event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+                        skbaddr, skblen)
+        all_event_list.append(event_info)
+
+def handle_irq_handler_entry(event_info):
+        (name, context, cpu, time, pid, comm, irq, irq_name) = event_info
+        if cpu not in irq_dic.keys():
+                irq_dic[cpu] = []
+        irq_record = {'irq':irq, 'name':irq_name, 'cpu':cpu, 'irq_ent_t':time}
+        irq_dic[cpu].append(irq_record)
+
+def handle_irq_handler_exit(event_info):
+        (name, context, cpu, time, pid, comm, irq, ret) = event_info
+        if cpu not in irq_dic.keys():
+                return
+        irq_record = irq_dic[cpu].pop()
+        if irq != irq_record['irq']:
+                return
+        irq_record.update({'irq_ext_t':time})
+        # if an irq doesn't include NET_RX softirq, drop.
+        if 'event_list' in irq_record.keys():
+                irq_dic[cpu].append(irq_record)
+
+def handle_irq_softirq_raise(event_info):
+        (name, context, cpu, time, pid, comm, vec) = event_info
+        if cpu not in irq_dic.keys() \
+        or len(irq_dic[cpu]) == 0:
+                return
+        irq_record = irq_dic[cpu].pop()
+        if 'event_list' in irq_record.keys():
+                irq_event_list = irq_record['event_list']
+        else:
+                irq_event_list = []
+        irq_event_list.append({'time':time, 'event':'sirq_raise'})
+        irq_record.update({'event_list':irq_event_list})
+        irq_dic[cpu].append(irq_record)
+
+def handle_irq_softirq_entry(event_info):
+        (name, context, cpu, time, pid, comm, vec) = event_info
+        net_rx_dic[cpu] = {'sirq_ent_t':time, 'event_list':[]}
+
+def handle_irq_softirq_exit(event_info):
+        (name, context, cpu, time, pid, comm, vec) = event_info
+        irq_list = []
+        event_list = 0
+        if cpu in irq_dic.keys():
+                irq_list = irq_dic[cpu]
+                del irq_dic[cpu]
+        if cpu in net_rx_dic.keys():
+                sirq_ent_t = net_rx_dic[cpu]['sirq_ent_t']
+                event_list = net_rx_dic[cpu]['event_list']
+                del net_rx_dic[cpu]
+        if irq_list == [] or event_list == 0:
+                return
+        rec_data = {'sirq_ent_t':sirq_ent_t, 'sirq_ext_t':time,
+                    'irq_list':irq_list, 'event_list':event_list}
+        # merge information realted to a NET_RX softirq
+        receive_hunk_list.append(rec_data)
+
+def handle_napi_poll(event_info):
+        (name, context, cpu, time, pid, comm, napi, dev_name) = event_info
+        if cpu in net_rx_dic.keys():
+                event_list = net_rx_dic[cpu]['event_list']
+                rec_data = {'event_name':'napi_poll',
+                                'dev':dev_name, 'event_t':time}
+                event_list.append(rec_data)
+
+def handle_netif_rx(event_info):
+        (name, context, cpu, time, pid, comm,
+                skbaddr, skblen, dev_name) = event_info
+        if cpu not in irq_dic.keys() \
+        or len(irq_dic[cpu]) == 0:
+                return
+        irq_record = irq_dic[cpu].pop()
+        if 'event_list' in irq_record.keys():
+                irq_event_list = irq_record['event_list']
+        else:
+                irq_event_list = []
+        irq_event_list.append({'time':time, 'event':'netif_rx',
+                'skbaddr':skbaddr, 'skblen':skblen, 'dev_name':dev_name})
+        irq_record.update({'event_list':irq_event_list})
+        irq_dic[cpu].append(irq_record)
+
+def handle_netif_receive_skb(event_info):
+        global of_count_rx_skb_list
+
+        (name, context, cpu, time, pid, comm,
+                skbaddr, skblen, dev_name) = event_info
+        if cpu in net_rx_dic.keys():
+                rec_data = {'event_name':'netif_receive_skb',
+                            'event_t':time, 'skbaddr':skbaddr, 'len':skblen}
+                event_list = net_rx_dic[cpu]['event_list']
+                event_list.append(rec_data)
+                rx_skb_list.insert(0, rec_data)
+                if len(rx_skb_list) > buffer_budget:
+                        rx_skb_list.pop()
+                        of_count_rx_skb_list += 1
+
+def handle_net_dev_queue(event_info):
+        global of_count_tx_queue_list
+
+        (name, context, cpu, time, pid, comm,
+                skbaddr, skblen, dev_name) = event_info
+        skb = {'dev':dev_name, 'skbaddr':skbaddr, 'len':skblen, 'queue_t':time}
+        tx_queue_list.insert(0, skb)
+        if len(tx_queue_list) > buffer_budget:
+                tx_queue_list.pop()
+                of_count_tx_queue_list += 1
+
+def handle_net_dev_xmit(event_info):
+        global of_count_tx_xmit_list
+
+        (name, context, cpu, time, pid, comm,
+                skbaddr, skblen, rc, dev_name) = event_info
+        if rc == 0: # NETDEV_TX_OK
+                for i in range(len(tx_queue_list)):
+                        skb = tx_queue_list[i]
+                        if skb['skbaddr'] == skbaddr:
+                                skb['xmit_t'] = time
+                                tx_xmit_list.insert(0, skb)
+                                del tx_queue_list[i]
+                                if len(tx_xmit_list) > buffer_budget:
+                                        tx_xmit_list.pop()
+                                        of_count_tx_xmit_list += 1
+                                return
+
+def handle_kfree_skb(event_info):
+        (name, context, cpu, time, pid, comm,
+                skbaddr, protocol, location) = event_info
+        for i in range(len(tx_queue_list)):
+                skb = tx_queue_list[i]
+                if skb['skbaddr'] == skbaddr:
+                        del tx_queue_list[i]
+                        return
+        for i in range(len(tx_xmit_list)):
+                skb = tx_xmit_list[i]
+                if skb['skbaddr'] == skbaddr:
+                        skb['free_t'] = time
+                        tx_free_list.append(skb)
+                        del tx_xmit_list[i]
+                        return
+        for i in range(len(rx_skb_list)):
+                rec_data = rx_skb_list[i]
+                if rec_data['skbaddr'] == skbaddr:
+                        rec_data.update({'handle':"kfree_skb",
+                                        'comm':comm, 'pid':pid, 'comm_t':time})
+                        del rx_skb_list[i]
+                        return
+
+def handle_consume_skb(event_info):
+        (name, context, cpu, time, pid, comm, skbaddr) = event_info
+        for i in range(len(tx_xmit_list)):
+                skb = tx_xmit_list[i]
+                if skb['skbaddr'] == skbaddr:
+                        skb['free_t'] = time
+                        tx_free_list.append(skb)
+                        del tx_xmit_list[i]
+                        return
+
+def handle_skb_copy_datagram_iovec(event_info):
+        (name, context, cpu, time, pid, comm, skbaddr, skblen) = event_info
+        for i in range(len(rx_skb_list)):
+                rec_data = rx_skb_list[i]
+                if skbaddr == rec_data['skbaddr']:
+                        rec_data.update({'handle':"skb_copy_datagram_iovec",
+                                        'comm':comm, 'pid':pid, 'comm_t':time})
+                        del rx_skb_list[i]
+                        return
diff --git a/tools/perf/scripts/python/sched-migration.py b/tools/perf/scripts/python/sched-migration.py
new file mode 100644
index 00000000000..74d55ec08ae
--- /dev/null
+++ b/tools/perf/scripts/python/sched-migration.py
@@ -0,0 +1,461 @@
+#!/usr/bin/python
+#
+# Cpu task migration overview toy
+#
+# Copyright (C) 2010 Frederic Weisbecker <fweisbec@gmail.com>
+#
+# perf script event handlers have been generated by perf script -g python
+#
+# This software is distributed under the terms of the GNU General
+# Public License ("GPL") version 2 as published by the Free Software
+# Foundation.
+
+
+import os
+import sys
+
+from collections import defaultdict
+from UserList import UserList
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+        '/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+sys.path.append('scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from SchedGui import *
+
+
+threads = { 0 : "idle"}
+
+def thread_name(pid):
+        return "%s:%d" % (threads[pid], pid)
+
+class RunqueueEventUnknown:
+        @staticmethod
+        def color():
+                return None
+
+        def __repr__(self):
+                return "unknown"
+
+class RunqueueEventSleep:
+        @staticmethod
+        def color():
+                return (0, 0, 0xff)
+
+        def __init__(self, sleeper):
+                self.sleeper = sleeper
+
+        def __repr__(self):
+                return "%s gone to sleep" % thread_name(self.sleeper)
+
+class RunqueueEventWakeup:
+        @staticmethod
+        def color():
+                return (0xff, 0xff, 0)
+
+        def __init__(self, wakee):
+                self.wakee = wakee
+
+        def __repr__(self):
+                return "%s woke up" % thread_name(self.wakee)
+
+class RunqueueEventFork:
+        @staticmethod
+        def color():
+                return (0, 0xff, 0)
+
+        def __init__(self, child):
+                self.child = child
+
+        def __repr__(self):
+                return "new forked task %s" % thread_name(self.child)
+
+class RunqueueMigrateIn:
+        @staticmethod
+        def color():
+                return (0, 0xf0, 0xff)
+
+        def __init__(self, new):
+                self.new = new
+
+        def __repr__(self):
+                return "task migrated in %s" % thread_name(self.new)
+
+class RunqueueMigrateOut:
+        @staticmethod
+        def color():
+                return (0xff, 0, 0xff)
+
+        def __init__(self, old):
+                self.old = old
+
+        def __repr__(self):
+                return "task migrated out %s" % thread_name(self.old)
+
+class RunqueueSnapshot:
+        def __init__(self, tasks = [0], event = RunqueueEventUnknown()):
+                self.tasks = tuple(tasks)
+                self.event = event
+
+        def sched_switch(self, prev, prev_state, next):
+                event = RunqueueEventUnknown()
+
+                if taskState(prev_state) == "R" and next in self.tasks \
+                        and prev in self.tasks:
+                        return self
+
+                if taskState(prev_state) != "R":
+                        event = RunqueueEventSleep(prev)
+
+                next_tasks = list(self.tasks[:])
+                if prev in self.tasks:
+                        if taskState(prev_state) != "R":
+                                next_tasks.remove(prev)
+                elif taskState(prev_state) == "R":
+                        next_tasks.append(prev)
+
+                if next not in next_tasks:
+                        next_tasks.append(next)
+
+                return RunqueueSnapshot(next_tasks, event)
+
+        def migrate_out(self, old):
+                if old not in self.tasks:
+                        return self
+                next_tasks = [task for task in self.tasks if task != old]
+
+                return RunqueueSnapshot(next_tasks, RunqueueMigrateOut(old))
+
+        def __migrate_in(self, new, event):
+                if new in self.tasks:
+                        self.event = event
+                        return self
+                next_tasks = self.tasks[:] + tuple([new])
+
+                return RunqueueSnapshot(next_tasks, event)
+
+        def migrate_in(self, new):
+                return self.__migrate_in(new, RunqueueMigrateIn(new))
+
+        def wake_up(self, new):
+                return self.__migrate_in(new, RunqueueEventWakeup(new))
+
+        def wake_up_new(self, new):
+                return self.__migrate_in(new, RunqueueEventFork(new))
+
+        def load(self):
+                """ Provide the number of tasks on the runqueue.
+                    Don't count idle"""
+                return len(self.tasks) - 1
+
+        def __repr__(self):
+                ret = self.tasks.__repr__()
+                ret += self.origin_tostring()
+
+                return ret
+
+class TimeSlice:
+        def __init__(self, start, prev):
+                self.start = start
+                self.prev = prev
+                self.end = start
+                # cpus that triggered the event
+                self.event_cpus = []
+                if prev is not None:
+                        self.total_load = prev.total_load
+                        self.rqs = prev.rqs.copy()
+                else:
+                        self.rqs = defaultdict(RunqueueSnapshot)
+                        self.total_load = 0
+
+        def __update_total_load(self, old_rq, new_rq):
+                diff = new_rq.load() - old_rq.load()
+                self.total_load += diff
+
+        def sched_switch(self, ts_list, prev, prev_state, next, cpu):
+                old_rq = self.prev.rqs[cpu]
+                new_rq = old_rq.sched_switch(prev, prev_state, next)
+
+                if old_rq is new_rq:
+                        return
+
+                self.rqs[cpu] = new_rq
+                self.__update_total_load(old_rq, new_rq)
+                ts_list.append(self)
+                self.event_cpus = [cpu]
+
+        def migrate(self, ts_list, new, old_cpu, new_cpu):
+                if old_cpu == new_cpu:
+                        return
+                old_rq = self.prev.rqs[old_cpu]
+                out_rq = old_rq.migrate_out(new)
+                self.rqs[old_cpu] = out_rq
+                self.__update_total_load(old_rq, out_rq)
+
+                new_rq = self.prev.rqs[new_cpu]
+                in_rq = new_rq.migrate_in(new)
+                self.rqs[new_cpu] = in_rq
+                self.__update_total_load(new_rq, in_rq)
+
+                ts_list.append(self)
+
+                if old_rq is not out_rq:
+                        self.event_cpus.append(old_cpu)
+                self.event_cpus.append(new_cpu)
+
+        def wake_up(self, ts_list, pid, cpu, fork):
+                old_rq = self.prev.rqs[cpu]
+                if fork:
+                        new_rq = old_rq.wake_up_new(pid)
+                else:
+                        new_rq = old_rq.wake_up(pid)
+
+                if new_rq is old_rq:
+                        return
+                self.rqs[cpu] = new_rq
+                self.__update_total_load(old_rq, new_rq)
+                ts_list.append(self)
+                self.event_cpus = [cpu]
+
+        def next(self, t):
+                self.end = t
+                return TimeSlice(t, self)
+
+class TimeSliceList(UserList):
+        def __init__(self, arg = []):
+                self.data = arg
+
+        def get_time_slice(self, ts):
+                if len(self.data) == 0:
+                        slice = TimeSlice(ts, TimeSlice(-1, None))
+                else:
+                        slice = self.data[-1].next(ts)
+                return slice
+
+        def find_time_slice(self, ts):
+                start = 0
+                end = len(self.data)
+                found = -1
+                searching = True
+                while searching:
+                        if start == end or start == end - 1:
+                                searching = False
+
+                        i = (end + start) / 2
+                        if self.data[i].start <= ts and self.data[i].end >= ts:
+                                found = i
+                                end = i
+                                continue
+
+                        if self.data[i].end < ts:
+                                start = i
+
+                        elif self.data[i].start > ts:
+                                end = i
+
+                return found
+
+        def set_root_win(self, win):
+                self.root_win = win
+
+        def mouse_down(self, cpu, t):
+                idx = self.find_time_slice(t)
+                if idx == -1:
+                        return
+
+                ts = self[idx]
+                rq = ts.rqs[cpu]
+                raw = "CPU: %d\n" % cpu
+                raw += "Last event : %s\n" % rq.event.__repr__()
+                raw += "Timestamp : %d.%06d\n" % (ts.start / (10 ** 9), (ts.start % (10 ** 9)) / 1000)
+                raw += "Duration : %6d us\n" % ((ts.end - ts.start) / (10 ** 6))
+                raw += "Load = %d\n" % rq.load()
+                for t in rq.tasks:
+                        raw += "%s \n" % thread_name(t)
+
+                self.root_win.update_summary(raw)
+
+        def update_rectangle_cpu(self, slice, cpu):
+                rq = slice.rqs[cpu]
+
+                if slice.total_load != 0:
+                        load_rate = rq.load() / float(slice.total_load)
+                else:
+                        load_rate = 0
+
+                red_power = int(0xff - (0xff * load_rate))
+                color = (0xff, red_power, red_power)
+
+                top_color = None
+
+                if cpu in slice.event_cpus:
+                        top_color = rq.event.color()
+
+                self.root_win.paint_rectangle_zone(cpu, color, top_color, slice.start, slice.end)
+
+        def fill_zone(self, start, end):
+                i = self.find_time_slice(start)
+                if i == -1:
+                        return
+
+                for i in xrange(i, len(self.data)):
+                        timeslice = self.data[i]
+                        if timeslice.start > end:
+                                return
+
+                        for cpu in timeslice.rqs:
+                                self.update_rectangle_cpu(timeslice, cpu)
+
+        def interval(self):
+                if len(self.data) == 0:
+                        return (0, 0)
+
+                return (self.data[0].start, self.data[-1].end)
+
+        def nr_rectangles(self):
+                last_ts = self.data[-1]
+                max_cpu = 0
+                for cpu in last_ts.rqs:
+                        if cpu > max_cpu:
+                                max_cpu = cpu
+                return max_cpu
+
+
+class SchedEventProxy:
+        def __init__(self):
+                self.current_tsk = defaultdict(lambda : -1)
+                self.timeslices = TimeSliceList()
+
+        def sched_switch(self, headers, prev_comm, prev_pid, prev_prio, prev_state,
+                         next_comm, next_pid, next_prio):
+                """ Ensure the task we sched out this cpu is really the one
+                    we logged. Otherwise we may have missed traces """
+
+                on_cpu_task = self.current_tsk[headers.cpu]
+
+                if on_cpu_task != -1 and on_cpu_task != prev_pid:
+                        print "Sched switch event rejected ts: %s cpu: %d prev: %s(%d) next: %s(%d)" % \
+                                (headers.ts_format(), headers.cpu, prev_comm, prev_pid, next_comm, next_pid)
+
+                threads[prev_pid] = prev_comm
+                threads[next_pid] = next_comm
+                self.current_tsk[headers.cpu] = next_pid
+
+                ts = self.timeslices.get_time_slice(headers.ts())
+                ts.sched_switch(self.timeslices, prev_pid, prev_state, next_pid, headers.cpu)
+
+        def migrate(self, headers, pid, prio, orig_cpu, dest_cpu):
+                ts = self.timeslices.get_time_slice(headers.ts())
+                ts.migrate(self.timeslices, pid, orig_cpu, dest_cpu)
+
+        def wake_up(self, headers, comm, pid, success, target_cpu, fork):
+                if success == 0:
+                        return
+                ts = self.timeslices.get_time_slice(headers.ts())
+                ts.wake_up(self.timeslices, pid, target_cpu, fork)
+
+
+def trace_begin():
+        global parser
+        parser = SchedEventProxy()
+
+def trace_end():
+        app = wx.App(False)
+        timeslices = parser.timeslices
+        frame = RootFrame(timeslices, "Migration")
+        app.MainLoop()
+
+def sched__sched_stat_runtime(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        comm, pid, runtime, vruntime):
+        pass
+
+def sched__sched_stat_iowait(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        comm, pid, delay):
+        pass
+
+def sched__sched_stat_sleep(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        comm, pid, delay):
+        pass
+
+def sched__sched_stat_wait(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        comm, pid, delay):
+        pass
+
+def sched__sched_process_fork(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        parent_comm, parent_pid, child_comm, child_pid):
+        pass
+
+def sched__sched_process_wait(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        comm, pid, prio):
+        pass
+
+def sched__sched_process_exit(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        comm, pid, prio):
+        pass
+
+def sched__sched_process_free(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        comm, pid, prio):
+        pass
+
+def sched__sched_migrate_task(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        comm, pid, prio, orig_cpu,
+        dest_cpu):
+        headers = EventHeaders(common_cpu, common_secs, common_nsecs,
+                                common_pid, common_comm)
+        parser.migrate(headers, pid, prio, orig_cpu, dest_cpu)
+
+def sched__sched_switch(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        prev_comm, prev_pid, prev_prio, prev_state,
+        next_comm, next_pid, next_prio):
+
+        headers = EventHeaders(common_cpu, common_secs, common_nsecs,
+                                common_pid, common_comm)
+        parser.sched_switch(headers, prev_comm, prev_pid, prev_prio, prev_state,
+                         next_comm, next_pid, next_prio)
+
+def sched__sched_wakeup_new(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        comm, pid, prio, success,
+        target_cpu):
+        headers = EventHeaders(common_cpu, common_secs, common_nsecs,
+                                common_pid, common_comm)
+        parser.wake_up(headers, comm, pid, success, target_cpu, 1)
+
+def sched__sched_wakeup(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        comm, pid, prio, success,
+        target_cpu):
+        headers = EventHeaders(common_cpu, common_secs, common_nsecs,
+                                common_pid, common_comm)
+        parser.wake_up(headers, comm, pid, success, target_cpu, 0)
+
+def sched__sched_wait_task(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        comm, pid, prio):
+        pass
+
+def sched__sched_kthread_stop_ret(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        ret):
+        pass
+
+def sched__sched_kthread_stop(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        comm, pid):
+        pass
+
+def trace_unhandled(event_name, context, common_cpu, common_secs, common_nsecs,
+                common_pid, common_comm):
+        pass
diff --git a/tools/perf/scripts/python/sctop.py b/tools/perf/scripts/python/sctop.py
new file mode 100644
index 00000000000..42c267e292f
--- /dev/null
+++ b/tools/perf/scripts/python/sctop.py
@@ -0,0 +1,75 @@
+# system call top
+# (c) 2010, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+#
+# Periodically displays system-wide system call totals, broken down by
+# syscall.  If a [comm] arg is specified, only syscalls called by
+# [comm] are displayed. If an [interval] arg is specified, the display
+# will be refreshed every [interval] seconds.  The default interval is
+# 3 seconds.
+
+import os, sys, thread, time
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+        '/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from Util import *
+
+usage = "perf script -s sctop.py [comm] [interval]\n";
+
+for_comm = None
+default_interval = 3
+interval = default_interval
+
+if len(sys.argv) > 3:
+        sys.exit(usage)
+
+if len(sys.argv) > 2:
+        for_comm = sys.argv[1]
+        interval = int(sys.argv[2])
+elif len(sys.argv) > 1:
+        try:
+                interval = int(sys.argv[1])
+        except ValueError:
+                for_comm = sys.argv[1]
+                interval = default_interval
+
+syscalls = autodict()
+
+def trace_begin():
+        thread.start_new_thread(print_syscall_totals, (interval,))
+        pass
+
+def raw_syscalls__sys_enter(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        id, args):
+        if for_comm is not None:
+                if common_comm != for_comm:
+                        return
+        try:
+                syscalls[id] += 1
+        except TypeError:
+                syscalls[id] = 1
+
+def print_syscall_totals(interval):
+        while 1:
+                clear_term()
+                if for_comm is not None:
+                        print "\nsyscall events for %s:\n\n" % (for_comm),
+                else:
+                        print "\nsyscall events:\n\n",
+
+                print "%-40s  %10s\n" % ("event", "count"),
+                print "%-40s  %10s\n" % ("----------------------------------------", \
+                                                 "----------"),
+
+                for id, val in sorted(syscalls.iteritems(), key = lambda(k, v): (v, k), \
+                                              reverse = True):
+                        try:
+                                print "%-40s  %10d\n" % (syscall_name(id), val),
+                        except TypeError:
+                                pass
+                syscalls.clear()
+                time.sleep(interval)
diff --git a/tools/perf/scripts/python/syscall-counts-by-pid.py b/tools/perf/scripts/python/syscall-counts-by-pid.py
new file mode 100644
index 00000000000..c64d1c55d74
--- /dev/null
+++ b/tools/perf/scripts/python/syscall-counts-by-pid.py
@@ -0,0 +1,69 @@
+# system call counts, by pid
+# (c) 2010, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+#
+# Displays system-wide system call totals, broken down by syscall.
+# If a [comm] arg is specified, only syscalls called by [comm] are displayed.
+
+import os, sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+        '/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from Util import syscall_name
+
+usage = "perf script -s syscall-counts-by-pid.py [comm]\n";
+
+for_comm = None
+for_pid = None
+
+if len(sys.argv) > 2:
+        sys.exit(usage)
+
+if len(sys.argv) > 1:
+        try:
+                for_pid = int(sys.argv[1])
+        except:
+                for_comm = sys.argv[1]
+
+syscalls = autodict()
+
+def trace_begin():
+        print "Press control+C to stop and show the summary"
+
+def trace_end():
+        print_syscall_totals()
+
+def raw_syscalls__sys_enter(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        id, args):
+
+        if (for_comm and common_comm != for_comm) or \
+           (for_pid  and common_pid  != for_pid ):
+                return
+        try:
+                syscalls[common_comm][common_pid][id] += 1
+        except TypeError:
+                syscalls[common_comm][common_pid][id] = 1
+
+def print_syscall_totals():
+    if for_comm is not None:
+            print "\nsyscall events for %s:\n\n" % (for_comm),
+    else:
+            print "\nsyscall events by comm/pid:\n\n",
+
+    print "%-40s  %10s\n" % ("comm [pid]/syscalls", "count"),
+    print "%-40s  %10s\n" % ("----------------------------------------", \
+                                 "----------"),
+
+    comm_keys = syscalls.keys()
+    for comm in comm_keys:
+            pid_keys = syscalls[comm].keys()
+            for pid in pid_keys:
+                    print "\n%s [%d]\n" % (comm, pid),
+                    id_keys = syscalls[comm][pid].keys()
+                    for id, val in sorted(syscalls[comm][pid].iteritems(), \
+                                  key = lambda(k, v): (v, k),  reverse = True):
+                            print "  %-38s  %10d\n" % (syscall_name(id), val),
diff --git a/tools/perf/scripts/python/syscall-counts.py b/tools/perf/scripts/python/syscall-counts.py
new file mode 100644
index 00000000000..b435d3f188e
--- /dev/null
+++ b/tools/perf/scripts/python/syscall-counts.py
@@ -0,0 +1,59 @@
+# system call counts
+# (c) 2010, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+#
+# Displays system-wide system call totals, broken down by syscall.
+# If a [comm] arg is specified, only syscalls called by [comm] are displayed.
+
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+        '/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from Util import syscall_name
+
+usage = "perf script -s syscall-counts.py [comm]\n";
+
+for_comm = None
+
+if len(sys.argv) > 2:
+        sys.exit(usage)
+
+if len(sys.argv) > 1:
+        for_comm = sys.argv[1]
+
+syscalls = autodict()
+
+def trace_begin():
+        print "Press control+C to stop and show the summary"
+
+def trace_end():
+        print_syscall_totals()
+
+def raw_syscalls__sys_enter(event_name, context, common_cpu,
+        common_secs, common_nsecs, common_pid, common_comm,
+        id, args):
+        if for_comm is not None:
+                if common_comm != for_comm:
+                        return
+        try:
+                syscalls[id] += 1
+        except TypeError:
+                syscalls[id] = 1
+
+def print_syscall_totals():
+    if for_comm is not None:
+            print "\nsyscall events for %s:\n\n" % (for_comm),
+    else:
+            print "\nsyscall events:\n\n",
+
+    print "%-40s  %10s\n" % ("event", "count"),
+    print "%-40s  %10s\n" % ("----------------------------------------", \
+                                 "-----------"),
+
+    for id, val in sorted(syscalls.iteritems(), key = lambda(k, v): (v, k), \
+                                  reverse = True):
+            print "%-40s  %10d\n" % (syscall_name(id), val),
diff --git a/tools/perf/util/PERF-VERSION-GEN b/tools/perf/util/PERF-VERSION-GEN
index c561d1538c0..97d76562a1a 100755
--- a/tools/perf/util/PERF-VERSION-GEN
+++ b/tools/perf/util/PERF-VERSION-GEN
@@ -1,17 +1,17 @@
 #!/bin/sh
 
-GVF=PERF-VERSION-FILE
-DEF_VER=v0.0.1.PERF
+if [ $# -eq 1 ]  ; then
+        OUTPUT=$1
+fi
+
+GVF=${OUTPUT}PERF-VERSION-FILE
 
 LF='
 '
 
-# First see if there is a version file (included in release tarballs),
-# then try git-describe, then default.
-if test -f version
-then
-        VN=$(cat version) || VN="$DEF_VER"
-elif test -d .git -o -f .git &&
+# First check if there is a .git to get the version from git describe
+# otherwise try to get the version from the kernel makefile
+if test -d ../../.git -o -f ../../.git &&
         VN=$(git describe --abbrev=4 HEAD 2>/dev/null) &&
         case "$VN" in
         *$LF*) (exit 1) ;;
@@ -23,7 +23,12 @@ elif test -d .git -o -f .git &&
 then
         VN=$(echo "$VN" | sed -e 's/-/./g');
 else
-        VN="$DEF_VER"
+        eval `grep '^VERSION\s*=' ../../Makefile|tr -d ' '`
+        eval `grep '^PATCHLEVEL\s*=' ../../Makefile|tr -d ' '`
+        eval `grep '^SUBLEVEL\s*=' ../../Makefile|tr -d ' '`
+        eval `grep '^EXTRAVERSION\s*=' ../../Makefile|tr -d ' '`
+
+        VN="${VERSION}.${PATCHLEVEL}.${SUBLEVEL}${EXTRAVERSION}"
 fi
 
 VN=$(expr "$VN" : v*'\(.*\)')
diff --git a/tools/perf/util/abspath.c b/tools/perf/util/abspath.c
index 61d33b81fc9..0e76affe9c3 100644
--- a/tools/perf/util/abspath.c
+++ b/tools/perf/util/abspath.c
@@ -1,85 +1,5 @@
 #include "cache.h"
 
-/*
- * Do not use this for inspecting *tracked* content.  When path is a
- * symlink to a directory, we do not want to say it is a directory when
- * dealing with tracked content in the working tree.
- */
-static int is_directory(const char *path)
-{
-        struct stat st;
-        return (!stat(path, &st) && S_ISDIR(st.st_mode));
-}
-
-/* We allow "recursive" symbolic links. Only within reason, though. */
-#define MAXDEPTH 5
-
-const char *make_absolute_path(const char *path)
-{
-        static char bufs[2][PATH_MAX + 1], *buf = bufs[0], *next_buf = bufs[1];
-        char cwd[1024] = "";
-        int buf_index = 1, len;
-
-        int depth = MAXDEPTH;
-        char *last_elem = NULL;
-        struct stat st;
-
-        if (strlcpy(buf, path, PATH_MAX) >= PATH_MAX)
-                die ("Too long path: %.*s", 60, path);
-
-        while (depth--) {
-                if (!is_directory(buf)) {
-                        char *last_slash = strrchr(buf, '/');
-                        if (last_slash) {
-                                *last_slash = '\0';
-                                last_elem = xstrdup(last_slash + 1);
-                        } else {
-                                last_elem = xstrdup(buf);
-                                *buf = '\0';
-                        }
-                }
-
-                if (*buf) {
-                        if (!*cwd && !getcwd(cwd, sizeof(cwd)))
-                                die ("Could not get current working directory");
-
-                        if (chdir(buf))
-                                die ("Could not switch to '%s'", buf);
-                }
-                if (!getcwd(buf, PATH_MAX))
-                        die ("Could not get current working directory");
-
-                if (last_elem) {
-                        int len = strlen(buf);
-                        if (len + strlen(last_elem) + 2 > PATH_MAX)
-                                die ("Too long path name: '%s/%s'",
-                                                buf, last_elem);
-                        buf[len] = '/';
-                        strcpy(buf + len + 1, last_elem);
-                        free(last_elem);
-                        last_elem = NULL;
-                }
-
-                if (!lstat(buf, &st) && S_ISLNK(st.st_mode)) {
-                        len = readlink(buf, next_buf, PATH_MAX);
-                        if (len < 0)
-                                die ("Invalid symlink: %s", buf);
-                        if (PATH_MAX <= len)
-                                die("symbolic link too long: %s", buf);
-                        next_buf[len] = '\0';
-                        buf = next_buf;
-                        buf_index = 1 - buf_index;
-                        next_buf = bufs[buf_index];
-                } else
-                        break;
-        }
-
-        if (*cwd && chdir(cwd))
-                die ("Could not change back to '%s'", cwd);
-
-        return buf;
-}
-
 static const char *get_pwd_cwd(void)
 {
         static char cwd[PATH_MAX + 1];
diff --git a/tools/perf/util/alias.c b/tools/perf/util/alias.c
index 9b3dd2b428d..b8144e80bb1 100644
--- a/tools/perf/util/alias.c
+++ b/tools/perf/util/alias.c
@@ -3,7 +3,7 @@
 static const char *alias_key;
 static char *alias_val;
 
-static int alias_lookup_cb(const char *k, const char *v, void *cb)
+static int alias_lookup_cb(const char *k, const char *v, void *cb __used)
 {
         if (!prefixcmp(k, "alias.") && !strcmp(k+6, alias_key)) {
                 if (!v)
diff --git a/tools/perf/util/bitmap.c b/tools/perf/util/bitmap.c
new file mode 100644
index 00000000000..5e230acae1e
--- /dev/null
+++ b/tools/perf/util/bitmap.c
@@ -0,0 +1,21 @@
+/*
+ * From lib/bitmap.c
+ * Helper functions for bitmap.h.
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2.  See the file COPYING for more details.
+ */
+#include <linux/bitmap.h>
+
+int __bitmap_weight(const unsigned long *bitmap, int bits)
+{
+        int k, w = 0, lim = bits/BITS_PER_LONG;
+
+        for (k = 0; k < lim; k++)
+                w += hweight_long(bitmap[k]);
+
+        if (bits % BITS_PER_LONG)
+                w += hweight_long(bitmap[k] & BITMAP_LAST_WORD_MASK(bits));
+
+        return w;
+}
diff --git a/tools/perf/util/build-id.c b/tools/perf/util/build-id.c
new file mode 100644
index 00000000000..deffb8c9607
--- /dev/null
+++ b/tools/perf/util/build-id.c
@@ -0,0 +1,80 @@
+/*
+ * build-id.c
+ *
+ * build-id support
+ *
+ * Copyright (C) 2009, 2010 Red Hat Inc.
+ * Copyright (C) 2009, 2010 Arnaldo Carvalho de Melo <acme@redhat.com>
+ */
+#include "util.h"
+#include <stdio.h>
+#include "build-id.h"
+#include "event.h"
+#include "symbol.h"
+#include <linux/kernel.h>
+#include "debug.h"
+
+static int build_id__mark_dso_hit(event_t *event,
+                                  struct sample_data *sample __used,
+                                  struct perf_session *session)
+{
+        struct addr_location al;
+        u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+        struct thread *thread = perf_session__findnew(session, event->ip.pid);
+
+        if (thread == NULL) {
+                pr_err("problem processing %d event, skipping it.\n",
+                        event->header.type);
+                return -1;
+        }
+
+        thread__find_addr_map(thread, session, cpumode, MAP__FUNCTION,
+                              event->ip.pid, event->ip.ip, &al);
+
+        if (al.map != NULL)
+                al.map->dso->hit = 1;
+
+        return 0;
+}
+
+static int event__exit_del_thread(event_t *self, struct sample_data *sample __used,
+                                  struct perf_session *session)
+{
+        struct thread *thread = perf_session__findnew(session, self->fork.tid);
+
+        dump_printf("(%d:%d):(%d:%d)\n", self->fork.pid, self->fork.tid,
+                    self->fork.ppid, self->fork.ptid);
+
+        if (thread) {
+                rb_erase(&thread->rb_node, &session->threads);
+                session->last_match = NULL;
+                thread__delete(thread);
+        }
+
+        return 0;
+}
+
+struct perf_event_ops build_id__mark_dso_hit_ops = {
+        .sample = build_id__mark_dso_hit,
+        .mmap   = event__process_mmap,
+        .fork   = event__process_task,
+        .exit   = event__exit_del_thread,
+};
+
+char *dso__build_id_filename(struct dso *self, char *bf, size_t size)
+{
+        char build_id_hex[BUILD_ID_SIZE * 2 + 1];
+
+        if (!self->has_build_id)
+                return NULL;
+
+        build_id__sprintf(self->build_id, sizeof(self->build_id), build_id_hex);
+        if (bf == NULL) {
+                if (asprintf(&bf, "%s/.build-id/%.2s/%s", buildid_dir,
+                             build_id_hex, build_id_hex + 2) < 0)
+                        return NULL;
+        } else
+                snprintf(bf, size, "%s/.build-id/%.2s/%s", buildid_dir,
+                         build_id_hex, build_id_hex + 2);
+        return bf;
+}
diff --git a/tools/perf/util/build-id.h b/tools/perf/util/build-id.h
new file mode 100644
index 00000000000..5dafb00eaa0
--- /dev/null
+++ b/tools/perf/util/build-id.h
@@ -0,0 +1,10 @@
+#ifndef PERF_BUILD_ID_H_
+#define PERF_BUILD_ID_H_ 1
+
+#include "session.h"
+
+extern struct perf_event_ops build_id__mark_dso_hit_ops;
+
+char *dso__build_id_filename(struct dso *self, char *bf, size_t size);
+
+#endif
diff --git a/tools/perf/util/cache.h b/tools/perf/util/cache.h
index 393d6146d13..a7729797fd9 100644
--- a/tools/perf/util/cache.h
+++ b/tools/perf/util/cache.h
@@ -1,61 +1,29 @@
-#ifndef CACHE_H
-#define CACHE_H
+#ifndef __PERF_CACHE_H
+#define __PERF_CACHE_H
 
+#include <stdbool.h>
 #include "util.h"
 #include "strbuf.h"
+#include "../perf.h"
+
+#define CMD_EXEC_PATH "--exec-path"
+#define CMD_PERF_DIR "--perf-dir="
+#define CMD_WORK_TREE "--work-tree="
+#define CMD_DEBUGFS_DIR "--debugfs-dir="
 
 #define PERF_DIR_ENVIRONMENT "PERF_DIR"
 #define PERF_WORK_TREE_ENVIRONMENT "PERF_WORK_TREE"
-#define DEFAULT_PERF_DIR_ENVIRONMENT ".perf"
-#define DB_ENVIRONMENT "PERF_OBJECT_DIRECTORY"
-#define INDEX_ENVIRONMENT "PERF_INDEX_FILE"
-#define GRAFT_ENVIRONMENT "PERF_GRAFT_FILE"
-#define TEMPLATE_DIR_ENVIRONMENT "PERF_TEMPLATE_DIR"
-#define CONFIG_ENVIRONMENT "PERF_CONFIG"
 #define EXEC_PATH_ENVIRONMENT "PERF_EXEC_PATH"
-#define CEILING_DIRECTORIES_ENVIRONMENT "PERF_CEILING_DIRECTORIES"
-#define PERFATTRIBUTES_FILE ".perfattributes"
-#define INFOATTRIBUTES_FILE "info/attributes"
-#define ATTRIBUTE_MACRO_PREFIX "[attr]"
+#define DEFAULT_PERF_DIR_ENVIRONMENT ".perf"
+#define PERF_DEBUGFS_ENVIRONMENT "PERF_DEBUGFS_DIR"
 
 typedef int (*config_fn_t)(const char *, const char *, void *);
 extern int perf_default_config(const char *, const char *, void *);
-extern int perf_config_from_file(config_fn_t fn, const char *, void *);
 extern int perf_config(config_fn_t fn, void *);
-extern int perf_parse_ulong(const char *, unsigned long *);
 extern int perf_config_int(const char *, const char *);
-extern unsigned long perf_config_ulong(const char *, const char *);
-extern int perf_config_bool_or_int(const char *, const char *, int *);
 extern int perf_config_bool(const char *, const char *);
-extern int perf_config_string(const char **, const char *, const char *);
-extern int perf_config_set(const char *, const char *);
-extern int perf_config_set_multivar(const char *, const char *, const char *, int);
-extern int perf_config_rename_section(const char *, const char *);
-extern const char *perf_etc_perfconfig(void);
-extern int check_repository_format_version(const char *var, const char *value, void *cb);
-extern int perf_config_system(void);
-extern int perf_config_global(void);
 extern int config_error_nonbool(const char *);
-extern const char *config_exclusive_filename;
-
-#define MAX_PERFNAME (1000)
-extern char perf_default_email[MAX_PERFNAME];
-extern char perf_default_name[MAX_PERFNAME];
-extern int user_ident_explicitly_given;
-
-extern const char *perf_log_output_encoding;
-extern const char *perf_mailmap_file;
-
-/* IO helper functions */
-extern void maybe_flush_or_die(FILE *, const char *);
-extern int copy_fd(int ifd, int ofd);
-extern int copy_file(const char *dst, const char *src, int mode);
-extern ssize_t read_in_full(int fd, void *buf, size_t count);
-extern ssize_t write_in_full(int fd, const void *buf, size_t count);
-extern void write_or_die(int fd, const void *buf, size_t count);
-extern int write_or_whine(int fd, const void *buf, size_t count, const char *msg);
-extern int write_or_whine_pipe(int fd, const void *buf, size_t count, const char *msg);
-extern void fsync_or_die(int fd, const char *);
+extern const char *perf_config_dirname(const char *, const char *);
 
 /* pager.c */
 extern void setup_pager(void);
@@ -63,8 +31,18 @@ extern const char *pager_program;
 extern int pager_in_use(void);
 extern int pager_use_color;
 
-extern const char *editor_program;
-extern const char *excludes_file;
+extern int use_browser;
+
+#ifdef NO_NEWT_SUPPORT
+static inline void setup_browser(void)
+{
+        setup_pager();
+}
+static inline void exit_browser(bool wait_for_ok __used) {}
+#else
+void setup_browser(void);
+void exit_browser(bool wait_for_ok);
+#endif
 
 char *alias_lookup(const char *alias);
 int split_cmdline(char *cmdline, const char ***argv);
@@ -95,25 +73,17 @@ static inline int is_absolute_path(const char *path)
         return path[0] == '/';
 }
 
-const char *make_absolute_path(const char *path);
 const char *make_nonrelative_path(const char *path);
-const char *make_relative_path(const char *abs, const char *base);
-int normalize_path_copy(char *dst, const char *src);
-int longest_ancestor_length(const char *path, const char *prefix_list);
 char *strip_path_suffix(const char *path, const char *suffix);
 
 extern char *mkpath(const char *fmt, ...) __attribute__((format (printf, 1, 2)));
 extern char *perf_path(const char *fmt, ...) __attribute__((format (printf, 1, 2)));
-/* perf_mkstemp() - create tmp file honoring TMPDIR variable */
-extern int perf_mkstemp(char *path, size_t len, const char *template);
 
-extern char *mksnpath(char *buf, size_t n, const char *fmt, ...)
-        __attribute__((format (printf, 3, 4)));
-extern char *perf_snpath(char *buf, size_t n, const char *fmt, ...)
-        __attribute__((format (printf, 3, 4)));
 extern char *perf_pathdup(const char *fmt, ...)
         __attribute__((format (printf, 1, 2)));
 
+#ifdef NO_STRLCPY
 extern size_t strlcpy(char *dest, const char *src, size_t size);
+#endif
 
-#endif /* CACHE_H */
+#endif /* __PERF_CACHE_H */
diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c
new file mode 100644
index 00000000000..e12d539417b
--- /dev/null
+++ b/tools/perf/util/callchain.c
@@ -0,0 +1,464 @@
+/*
+ * Copyright (C) 2009-2010, Frederic Weisbecker <fweisbec@gmail.com>
+ *
+ * Handle the callchains from the stream in an ad-hoc radix tree and then
+ * sort them in an rbtree.
+ *
+ * Using a radix for code path provides a fast retrieval and factorizes
+ * memory use. Also that lets us use the paths in a hierarchical graph view.
+ *
+ */
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <stdbool.h>
+#include <errno.h>
+#include <math.h>
+
+#include "util.h"
+#include "callchain.h"
+
+bool ip_callchain__valid(struct ip_callchain *chain, const event_t *event)
+{
+        unsigned int chain_size = event->header.size;
+        chain_size -= (unsigned long)&event->ip.__more_data - (unsigned long)event;
+        return chain->nr * sizeof(u64) <= chain_size;
+}
+
+#define chain_for_each_child(child, parent)     \
+        list_for_each_entry(child, &parent->children, brothers)
+
+#define chain_for_each_child_safe(child, next, parent)  \
+        list_for_each_entry_safe(child, next, &parent->children, brothers)
+
+static void
+rb_insert_callchain(struct rb_root *root, struct callchain_node *chain,
+                    enum chain_mode mode)
+{
+        struct rb_node **p = &root->rb_node;
+        struct rb_node *parent = NULL;
+        struct callchain_node *rnode;
+        u64 chain_cumul = cumul_hits(chain);
+
+        while (*p) {
+                u64 rnode_cumul;
+
+                parent = *p;
+                rnode = rb_entry(parent, struct callchain_node, rb_node);
+                rnode_cumul = cumul_hits(rnode);
+
+                switch (mode) {
+                case CHAIN_FLAT:
+                        if (rnode->hit < chain->hit)
+                                p = &(*p)->rb_left;
+                        else
+                                p = &(*p)->rb_right;
+                        break;
+                case CHAIN_GRAPH_ABS: /* Falldown */
+                case CHAIN_GRAPH_REL:
+                        if (rnode_cumul < chain_cumul)
+                                p = &(*p)->rb_left;
+                        else
+                                p = &(*p)->rb_right;
+                        break;
+                case CHAIN_NONE:
+                default:
+                        break;
+                }
+        }
+
+        rb_link_node(&chain->rb_node, parent, p);
+        rb_insert_color(&chain->rb_node, root);
+}
+
+static void
+__sort_chain_flat(struct rb_root *rb_root, struct callchain_node *node,
+                  u64 min_hit)
+{
+        struct callchain_node *child;
+
+        chain_for_each_child(child, node)
+                __sort_chain_flat(rb_root, child, min_hit);
+
+        if (node->hit && node->hit >= min_hit)
+                rb_insert_callchain(rb_root, node, CHAIN_FLAT);
+}
+
+/*
+ * Once we get every callchains from the stream, we can now
+ * sort them by hit
+ */
+static void
+sort_chain_flat(struct rb_root *rb_root, struct callchain_root *root,
+                u64 min_hit, struct callchain_param *param __used)
+{
+        __sort_chain_flat(rb_root, &root->node, min_hit);
+}
+
+static void __sort_chain_graph_abs(struct callchain_node *node,
+                                   u64 min_hit)
+{
+        struct callchain_node *child;
+
+        node->rb_root = RB_ROOT;
+
+        chain_for_each_child(child, node) {
+                __sort_chain_graph_abs(child, min_hit);
+                if (cumul_hits(child) >= min_hit)
+                        rb_insert_callchain(&node->rb_root, child,
+                                            CHAIN_GRAPH_ABS);
+        }
+}
+
+static void
+sort_chain_graph_abs(struct rb_root *rb_root, struct callchain_root *chain_root,
+                     u64 min_hit, struct callchain_param *param __used)
+{
+        __sort_chain_graph_abs(&chain_root->node, min_hit);
+        rb_root->rb_node = chain_root->node.rb_root.rb_node;
+}
+
+static void __sort_chain_graph_rel(struct callchain_node *node,
+                                   double min_percent)
+{
+        struct callchain_node *child;
+        u64 min_hit;
+
+        node->rb_root = RB_ROOT;
+        min_hit = ceil(node->children_hit * min_percent);
+
+        chain_for_each_child(child, node) {
+                __sort_chain_graph_rel(child, min_percent);
+                if (cumul_hits(child) >= min_hit)
+                        rb_insert_callchain(&node->rb_root, child,
+                                            CHAIN_GRAPH_REL);
+        }
+}
+
+static void
+sort_chain_graph_rel(struct rb_root *rb_root, struct callchain_root *chain_root,
+                     u64 min_hit __used, struct callchain_param *param)
+{
+        __sort_chain_graph_rel(&chain_root->node, param->min_percent / 100.0);
+        rb_root->rb_node = chain_root->node.rb_root.rb_node;
+}
+
+int register_callchain_param(struct callchain_param *param)
+{
+        switch (param->mode) {
+        case CHAIN_GRAPH_ABS:
+                param->sort = sort_chain_graph_abs;
+                break;
+        case CHAIN_GRAPH_REL:
+                param->sort = sort_chain_graph_rel;
+                break;
+        case CHAIN_FLAT:
+                param->sort = sort_chain_flat;
+                break;
+        case CHAIN_NONE:
+        default:
+                return -1;
+        }
+        return 0;
+}
+
+/*
+ * Create a child for a parent. If inherit_children, then the new child
+ * will become the new parent of it's parent children
+ */
+static struct callchain_node *
+create_child(struct callchain_node *parent, bool inherit_children)
+{
+        struct callchain_node *new;
+
+        new = zalloc(sizeof(*new));
+        if (!new) {
+                perror("not enough memory to create child for code path tree");
+                return NULL;
+        }
+        new->parent = parent;
+        INIT_LIST_HEAD(&new->children);
+        INIT_LIST_HEAD(&new->val);
+
+        if (inherit_children) {
+                struct callchain_node *next;
+
+                list_splice(&parent->children, &new->children);
+                INIT_LIST_HEAD(&parent->children);
+
+                chain_for_each_child(next, new)
+                        next->parent = new;
+        }
+        list_add_tail(&new->brothers, &parent->children);
+
+        return new;
+}
+
+
+struct resolved_ip {
+        u64               ip;
+        struct map_symbol ms;
+};
+
+struct resolved_chain {
+        u64                     nr;
+        struct resolved_ip      ips[0];
+};
+
+
+/*
+ * Fill the node with callchain values
+ */
+static void
+fill_node(struct callchain_node *node, struct resolved_chain *chain, int start)
+{
+        unsigned int i;
+
+        for (i = start; i < chain->nr; i++) {
+                struct callchain_list *call;
+
+                call = zalloc(sizeof(*call));
+                if (!call) {
+                        perror("not enough memory for the code path tree");
+                        return;
+                }
+                call->ip = chain->ips[i].ip;
+                call->ms = chain->ips[i].ms;
+                list_add_tail(&call->list, &node->val);
+        }
+        node->val_nr = chain->nr - start;
+        if (!node->val_nr)
+                pr_warning("Warning: empty node in callchain tree\n");
+}
+
+static void
+add_child(struct callchain_node *parent, struct resolved_chain *chain,
+          int start, u64 period)
+{
+        struct callchain_node *new;
+
+        new = create_child(parent, false);
+        fill_node(new, chain, start);
+
+        new->children_hit = 0;
+        new->hit = period;
+}
+
+/*
+ * Split the parent in two parts (a new child is created) and
+ * give a part of its callchain to the created child.
+ * Then create another child to host the given callchain of new branch
+ */
+static void
+split_add_child(struct callchain_node *parent, struct resolved_chain *chain,
+                struct callchain_list *to_split, int idx_parents, int idx_local,
+                u64 period)
+{
+        struct callchain_node *new;
+        struct list_head *old_tail;
+        unsigned int idx_total = idx_parents + idx_local;
+
+        /* split */
+        new = create_child(parent, true);
+
+        /* split the callchain and move a part to the new child */
+        old_tail = parent->val.prev;
+        list_del_range(&to_split->list, old_tail);
+        new->val.next = &to_split->list;
+        new->val.prev = old_tail;
+        to_split->list.prev = &new->val;
+        old_tail->next = &new->val;
+
+        /* split the hits */
+        new->hit = parent->hit;
+        new->children_hit = parent->children_hit;
+        parent->children_hit = cumul_hits(new);
+        new->val_nr = parent->val_nr - idx_local;
+        parent->val_nr = idx_local;
+
+        /* create a new child for the new branch if any */
+        if (idx_total < chain->nr) {
+                parent->hit = 0;
+                add_child(parent, chain, idx_total, period);
+                parent->children_hit += period;
+        } else {
+                parent->hit = period;
+        }
+}
+
+static int
+append_chain(struct callchain_node *root, struct resolved_chain *chain,
+             unsigned int start, u64 period);
+
+static void
+append_chain_children(struct callchain_node *root, struct resolved_chain *chain,
+                      unsigned int start, u64 period)
+{
+        struct callchain_node *rnode;
+
+        /* lookup in childrens */
+        chain_for_each_child(rnode, root) {
+                unsigned int ret = append_chain(rnode, chain, start, period);
+
+                if (!ret)
+                        goto inc_children_hit;
+        }
+        /* nothing in children, add to the current node */
+        add_child(root, chain, start, period);
+
+inc_children_hit:
+        root->children_hit += period;
+}
+
+static int
+append_chain(struct callchain_node *root, struct resolved_chain *chain,
+             unsigned int start, u64 period)
+{
+        struct callchain_list *cnode;
+        unsigned int i = start;
+        bool found = false;
+
+        /*
+         * Lookup in the current node
+         * If we have a symbol, then compare the start to match
+         * anywhere inside a function.
+         */
+        list_for_each_entry(cnode, &root->val, list) {
+                struct symbol *sym;
+
+                if (i == chain->nr)
+                        break;
+
+                sym = chain->ips[i].ms.sym;
+
+                if (cnode->ms.sym && sym) {
+                        if (cnode->ms.sym->start != sym->start)
+                                break;
+                } else if (cnode->ip != chain->ips[i].ip)
+                        break;
+
+                if (!found)
+                        found = true;
+                i++;
+        }
+
+        /* matches not, relay on the parent */
+        if (!found)
+                return -1;
+
+        /* we match only a part of the node. Split it and add the new chain */
+        if (i - start < root->val_nr) {
+                split_add_child(root, chain, cnode, start, i - start, period);
+                return 0;
+        }
+
+        /* we match 100% of the path, increment the hit */
+        if (i - start == root->val_nr && i == chain->nr) {
+                root->hit += period;
+                return 0;
+        }
+
+        /* We match the node and still have a part remaining */
+        append_chain_children(root, chain, i, period);
+
+        return 0;
+}
+
+static void filter_context(struct ip_callchain *old, struct resolved_chain *new,
+                           struct map_symbol *syms)
+{
+        int i, j = 0;
+
+        for (i = 0; i < (int)old->nr; i++) {
+                if (old->ips[i] >= PERF_CONTEXT_MAX)
+                        continue;
+
+                new->ips[j].ip = old->ips[i];
+                new->ips[j].ms = syms[i];
+                j++;
+        }
+
+        new->nr = j;
+}
+
+
+int callchain_append(struct callchain_root *root, struct ip_callchain *chain,
+                     struct map_symbol *syms, u64 period)
+{
+        struct resolved_chain *filtered;
+
+        if (!chain->nr)
+                return 0;
+
+        filtered = zalloc(sizeof(*filtered) +
+                          chain->nr * sizeof(struct resolved_ip));
+        if (!filtered)
+                return -ENOMEM;
+
+        filter_context(chain, filtered, syms);
+
+        if (!filtered->nr)
+                goto end;
+
+        append_chain_children(&root->node, filtered, 0, period);
+
+        if (filtered->nr > root->max_depth)
+                root->max_depth = filtered->nr;
+end:
+        free(filtered);
+
+        return 0;
+}
+
+static int
+merge_chain_branch(struct callchain_node *dst, struct callchain_node *src,
+                   struct resolved_chain *chain)
+{
+        struct callchain_node *child, *next_child;
+        struct callchain_list *list, *next_list;
+        int old_pos = chain->nr;
+        int err = 0;
+
+        list_for_each_entry_safe(list, next_list, &src->val, list) {
+                chain->ips[chain->nr].ip = list->ip;
+                chain->ips[chain->nr].ms = list->ms;
+                chain->nr++;
+                list_del(&list->list);
+                free(list);
+        }
+
+        if (src->hit)
+                append_chain_children(dst, chain, 0, src->hit);
+
+        chain_for_each_child_safe(child, next_child, src) {
+                err = merge_chain_branch(dst, child, chain);
+                if (err)
+                        break;
+
+                list_del(&child->brothers);
+                free(child);
+        }
+
+        chain->nr = old_pos;
+
+        return err;
+}
+
+int callchain_merge(struct callchain_root *dst, struct callchain_root *src)
+{
+        struct resolved_chain *chain;
+        int err;
+
+        chain = malloc(sizeof(*chain) +
+                       src->max_depth * sizeof(struct resolved_ip));
+        if (!chain)
+                return -ENOMEM;
+
+        chain->nr = 0;
+
+        err = merge_chain_branch(&dst->node, &src->node, chain);
+
+        free(chain);
+
+        return err;
+}
diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h
new file mode 100644
index 00000000000..c15fb8c24ad
--- /dev/null
+++ b/tools/perf/util/callchain.h
@@ -0,0 +1,75 @@
+#ifndef __PERF_CALLCHAIN_H
+#define __PERF_CALLCHAIN_H
+
+#include "../perf.h"
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include "event.h"
+#include "symbol.h"
+
+enum chain_mode {
+        CHAIN_NONE,
+        CHAIN_FLAT,
+        CHAIN_GRAPH_ABS,
+        CHAIN_GRAPH_REL
+};
+
+struct callchain_node {
+        struct callchain_node   *parent;
+        struct list_head        brothers;
+        struct list_head        children;
+        struct list_head        val;
+        struct rb_node          rb_node; /* to sort nodes in an rbtree */
+        struct rb_root          rb_root; /* sorted tree of children */
+        unsigned int            val_nr;
+        u64                     hit;
+        u64                     children_hit;
+};
+
+struct callchain_root {
+        u64                     max_depth;
+        struct callchain_node   node;
+};
+
+struct callchain_param;
+
+typedef void (*sort_chain_func_t)(struct rb_root *, struct callchain_root *,
+                                 u64, struct callchain_param *);
+
+struct callchain_param {
+        enum chain_mode         mode;
+        u32                     print_limit;
+        double                  min_percent;
+        sort_chain_func_t       sort;
+};
+
+struct callchain_list {
+        u64                     ip;
+        struct map_symbol       ms;
+        struct list_head        list;
+};
+
+static inline void callchain_init(struct callchain_root *root)
+{
+        INIT_LIST_HEAD(&root->node.brothers);
+        INIT_LIST_HEAD(&root->node.children);
+        INIT_LIST_HEAD(&root->node.val);
+
+        root->node.parent = NULL;
+        root->node.hit = 0;
+        root->node.children_hit = 0;
+        root->max_depth = 0;
+}
+
+static inline u64 cumul_hits(struct callchain_node *node)
+{
+        return node->hit + node->children_hit;
+}
+
+int register_callchain_param(struct callchain_param *param);
+int callchain_append(struct callchain_root *root, struct ip_callchain *chain,
+                     struct map_symbol *syms, u64 period);
+int callchain_merge(struct callchain_root *dst, struct callchain_root *src);
+
+bool ip_callchain__valid(struct ip_callchain *chain, const event_t *event);
+#endif  /* __PERF_CALLCHAIN_H */
diff --git a/tools/perf/util/color.c b/tools/perf/util/color.c
index 9a8c20ccc53..e191eb9a667 100644
--- a/tools/perf/util/color.c
+++ b/tools/perf/util/color.c
@@ -11,7 +11,8 @@ static int parse_color(const char *name, int len)
         };
         char *end;
         int i;
-        for (i = 0; i < ARRAY_SIZE(color_names); i++) {
+
+        for (i = 0; i < (int)ARRAY_SIZE(color_names); i++) {
                 const char *str = color_names[i];
                 if (!strncasecmp(name, str, len) && !str[len])
                         return i - 1;
@@ -28,7 +29,8 @@ static int parse_attr(const char *name, int len)
         static const char * const attr_names[] = {
                 "bold", "dim", "ul", "blink", "reverse"
         };
-        int i;
+        unsigned int i;
+
         for (i = 0; i < ARRAY_SIZE(attr_names); i++) {
                 const char *str = attr_names[i];
                 if (!strncasecmp(name, str, len) && !str[len])
@@ -164,7 +166,32 @@ int perf_color_default_config(const char *var, const char *value, void *cb)
         return perf_default_config(var, value, cb);
 }
 
-static int color_vfprintf(FILE *fp, const char *color, const char *fmt,
+static int __color_vsnprintf(char *bf, size_t size, const char *color,
+                             const char *fmt, va_list args, const char *trail)
+{
+        int r = 0;
+
+        /*
+         * Auto-detect:
+         */
+        if (perf_use_color_default < 0) {
+                if (isatty(1) || pager_in_use())
+                        perf_use_color_default = 1;
+                else
+                        perf_use_color_default = 0;
+        }
+
+        if (perf_use_color_default && *color)
+                r += snprintf(bf, size, "%s", color);
+        r += vsnprintf(bf + r, size - r, fmt, args);
+        if (perf_use_color_default && *color)
+                r += snprintf(bf + r, size - r, "%s", PERF_COLOR_RESET);
+        if (trail)
+                r += snprintf(bf + r, size - r, "%s", trail);
+        return r;
+}
+
+static int __color_vfprintf(FILE *fp, const char *color, const char *fmt,
                 va_list args, const char *trail)
 {
         int r = 0;
@@ -189,7 +216,28 @@ static int color_vfprintf(FILE *fp, const char *color, const char *fmt,
         return r;
 }
 
+int color_vsnprintf(char *bf, size_t size, const char *color,
+                    const char *fmt, va_list args)
+{
+        return __color_vsnprintf(bf, size, color, fmt, args, NULL);
+}
+
+int color_vfprintf(FILE *fp, const char *color, const char *fmt, va_list args)
+{
+        return __color_vfprintf(fp, color, fmt, args, NULL);
+}
+
+int color_snprintf(char *bf, size_t size, const char *color,
+                   const char *fmt, ...)
+{
+        va_list args;
+        int r;
 
+        va_start(args, fmt);
+        r = color_vsnprintf(bf, size, color, fmt, args);
+        va_end(args);
+        return r;
+}
 
 int color_fprintf(FILE *fp, const char *color, const char *fmt, ...)
 {
@@ -197,7 +245,7 @@ int color_fprintf(FILE *fp, const char *color, const char *fmt, ...)
         int r;
 
         va_start(args, fmt);
-        r = color_vfprintf(fp, color, fmt, args, NULL);
+        r = color_vfprintf(fp, color, fmt, args);
         va_end(args);
         return r;
 }
@@ -207,7 +255,7 @@ int color_fprintf_ln(FILE *fp, const char *color, const char *fmt, ...)
         va_list args;
         int r;
         va_start(args, fmt);
-        r = color_vfprintf(fp, color, fmt, args, "\n");
+        r = __color_vfprintf(fp, color, fmt, args, "\n");
         va_end(args);
         return r;
 }
@@ -222,10 +270,12 @@ int color_fwrite_lines(FILE *fp, const char *color,
 {
         if (!*color)
                 return fwrite(buf, count, 1, fp) != 1;
+
         while (count) {
                 char *p = memchr(buf, '\n', count);
+
                 if (p != buf && (fputs(color, fp) < 0 ||
-                                fwrite(buf, p ? p - buf : count, 1, fp) != 1 ||
+                                fwrite(buf, p ? (size_t)(p - buf) : count, 1, fp) != 1 ||
                                 fputs(PERF_COLOR_RESET, fp) < 0))
                         return -1;
                 if (!p)
@@ -238,4 +288,37 @@ int color_fwrite_lines(FILE *fp, const char *color,
         return 0;
 }
 
+const char *get_percent_color(double percent)
+{
+        const char *color = PERF_COLOR_NORMAL;
+
+        /*
+         * We color high-overhead entries in red, mid-overhead
+         * entries in green - and keep the low overhead places
+         * normal:
+         */
+        if (percent >= MIN_RED)
+                color = PERF_COLOR_RED;
+        else {
+                if (percent > MIN_GREEN)
+                        color = PERF_COLOR_GREEN;
+        }
+        return color;
+}
+
+int percent_color_fprintf(FILE *fp, const char *fmt, double percent)
+{
+        int r;
+        const char *color;
 
+        color = get_percent_color(percent);
+        r = color_fprintf(fp, color, fmt, percent);
+
+        return r;
+}
+
+int percent_color_snprintf(char *bf, size_t size, const char *fmt, double percent)
+{
+        const char *color = get_percent_color(percent);
+        return color_snprintf(bf, size, color, fmt, percent);
+}
diff --git a/tools/perf/util/color.h b/tools/perf/util/color.h
index 5abfd379582..dea082b7960 100644
--- a/tools/perf/util/color.h
+++ b/tools/perf/util/color.h
@@ -1,5 +1,5 @@
-#ifndef COLOR_H
-#define COLOR_H
+#ifndef __PERF_COLOR_H
+#define __PERF_COLOR_H
 
 /* "\033[1;38;5;2xx;48;5;2xxm\0" is 23 bytes */
 #define COLOR_MAXLEN 24
@@ -15,6 +15,9 @@
 #define PERF_COLOR_CYAN         "\033[36m"
 #define PERF_COLOR_BG_RED       "\033[41m"
 
+#define MIN_GREEN       0.5
+#define MIN_RED         5.0
+
 /*
  * This variable stores the value of color.ui
  */
@@ -29,8 +32,15 @@ int perf_color_default_config(const char *var, const char *value, void *cb);
 int perf_config_colorbool(const char *var, const char *value, int stdout_is_tty);
 void color_parse(const char *value, const char *var, char *dst);
 void color_parse_mem(const char *value, int len, const char *var, char *dst);
+int color_vsnprintf(char *bf, size_t size, const char *color,
+                    const char *fmt, va_list args);
+int color_vfprintf(FILE *fp, const char *color, const char *fmt, va_list args);
 int color_fprintf(FILE *fp, const char *color, const char *fmt, ...);
+int color_snprintf(char *bf, size_t size, const char *color, const char *fmt, ...);
 int color_fprintf_ln(FILE *fp, const char *color, const char *fmt, ...);
 int color_fwrite_lines(FILE *fp, const char *color, size_t count, const char *buf);
+int percent_color_snprintf(char *bf, size_t size, const char *fmt, double percent);
+int percent_color_fprintf(FILE *fp, const char *fmt, double percent);
+const char *get_percent_color(double percent);
 
-#endif /* COLOR_H */
+#endif /* __PERF_COLOR_H */
diff --git a/tools/perf/util/config.c b/tools/perf/util/config.c
index 3dd13faa6a2..e02d78cae70 100644
--- a/tools/perf/util/config.c
+++ b/tools/perf/util/config.c
@@ -11,12 +11,17 @@
 
 #define MAXNAME (256)
 
+#define DEBUG_CACHE_DIR ".debug"
+
+
+char buildid_dir[MAXPATHLEN]; /* root dir for buildid, binary cache */
+
 static FILE *config_file;
 static const char *config_file_name;
 static int config_linenr;
 static int config_file_eof;
 
-const char *config_exclusive_filename = NULL;
+static const char *config_exclusive_filename;
 
 static int get_next_char(void)
 {
@@ -47,10 +52,12 @@ static int get_next_char(void)
 static char *parse_value(void)
 {
         static char value[1024];
-        int quote = 0, comment = 0, len = 0, space = 0;
+        int quote = 0, comment = 0, space = 0;
+        size_t len = 0;
 
         for (;;) {
                 int c = get_next_char();
+
                 if (len >= sizeof(value) - 1)
                         return NULL;
                 if (c == '\n') {
@@ -125,7 +132,7 @@ static int get_value(config_fn_t fn, void *data, char *name, unsigned int len)
                         break;
                 if (!iskeychar(c))
                         break;
-                name[len++] = tolower(c);
+                name[len++] = c;
                 if (len >= MAXNAME)
                         return -1;
         }
@@ -158,17 +165,18 @@ static int get_extended_base_var(char *name, int baselen, int c)
         name[baselen++] = '.';
 
         for (;;) {
-                int c = get_next_char();
-                if (c == '\n')
+                int ch = get_next_char();
+
+                if (ch == '\n')
                         return -1;
-                if (c == '"')
+                if (ch == '"')
                         break;
-                if (c == '\\') {
-                        c = get_next_char();
-                        if (c == '\n')
+                if (ch == '\\') {
+                        ch = get_next_char();
+                        if (ch == '\n')
                                 return -1;
                 }
-                name[baselen++] = c;
+                name[baselen++] = ch;
                 if (baselen > MAXNAME / 2)
                         return -1;
         }
@@ -288,19 +296,6 @@ static int perf_parse_long(const char *value, long *ret)
         return 0;
 }
 
-int perf_parse_ulong(const char *value, unsigned long *ret)
-{
-        if (value && *value) {
-                char *end;
-                unsigned long val = strtoul(value, &end, 0);
-                if (!parse_unit_factor(end, &val))
-                        return 0;
-                *ret = val;
-                return 1;
-        }
-        return 0;
-}
-
 static void die_bad_config(const char *name)
 {
         if (config_file_name)
@@ -316,15 +311,7 @@ int perf_config_int(const char *name, const char *value)
         return ret;
 }
 
-unsigned long perf_config_ulong(const char *name, const char *value)
-{
-        unsigned long ret;
-        if (!perf_parse_ulong(value, &ret))
-                die_bad_config(name);
-        return ret;
-}
-
-int perf_config_bool_or_int(const char *name, const char *value, int *is_bool)
+static int perf_config_bool_or_int(const char *name, const char *value, int *is_bool)
 {
         *is_bool = 1;
         if (!value)
@@ -345,21 +332,20 @@ int perf_config_bool(const char *name, const char *value)
         return !!perf_config_bool_or_int(name, value, &discard);
 }
 
-int perf_config_string(const char **dest, const char *var, const char *value)
+const char *perf_config_dirname(const char *name, const char *value)
 {
-        if (!value)
-                return config_error_nonbool(var);
-        *dest = strdup(value);
-        return 0;
+        if (!name)
+                return NULL;
+        return value;
 }
 
-static int perf_default_core_config(const char *var, const char *value)
+static int perf_default_core_config(const char *var __used, const char *value __used)
 {
         /* Add other config variables here and to Documentation/config.txt. */
         return 0;
 }
 
-int perf_default_config(const char *var, const char *value, void *dummy)
+int perf_default_config(const char *var, const char *value, void *dummy __used)
 {
         if (!prefixcmp(var, "core."))
                 return perf_default_core_config(var, value);
@@ -368,7 +354,7 @@ int perf_default_config(const char *var, const char *value, void *dummy)
         return 0;
 }
 
-int perf_config_from_file(config_fn_t fn, const char *filename, void *data)
+static int perf_config_from_file(config_fn_t fn, const char *filename, void *data)
 {
         int ret;
         FILE *f = fopen(filename, "r");
@@ -386,7 +372,7 @@ int perf_config_from_file(config_fn_t fn, const char *filename, void *data)
         return ret;
 }
 
-const char *perf_etc_perfconfig(void)
+static const char *perf_etc_perfconfig(void)
 {
         static const char *system_wide;
         if (!system_wide)
@@ -400,12 +386,12 @@ static int perf_env_bool(const char *k, int def)
         return v ? perf_config_bool(k, v) : def;
 }
 
-int perf_config_system(void)
+static int perf_config_system(void)
 {
         return !perf_env_bool("PERF_CONFIG_NOSYSTEM", 0);
 }
 
-int perf_config_global(void)
+static int perf_config_global(void)
 {
         return !perf_env_bool("PERF_CONFIG_NOGLOBAL", 0);
 }
@@ -447,427 +433,60 @@ int perf_config(config_fn_t fn, void *data)
 }
 
 /*
- * Find all the stuff for perf_config_set() below.
+ * Call this to report error for your variable that should not
+ * get a boolean value (i.e. "[my] var" means "true").
  */
-
-#define MAX_MATCHES 512
-
-static struct {
-        int baselen;
-        char* key;
-        int do_not_match;
-        regex_t* value_regex;
-        int multi_replace;
-        size_t offset[MAX_MATCHES];
-        enum { START, SECTION_SEEN, SECTION_END_SEEN, KEY_SEEN } state;
-        int seen;
-} store;
-
-static int matches(const char* key, const char* value)
-{
-        return !strcmp(key, store.key) &&
-                (store.value_regex == NULL ||
-                 (store.do_not_match ^
-                  !regexec(store.value_regex, value, 0, NULL, 0)));
-}
-
-static int store_aux(const char* key, const char* value, void *cb)
+int config_error_nonbool(const char *var)
 {
-        const char *ep;
-        size_t section_len;
-
-        switch (store.state) {
-        case KEY_SEEN:
-                if (matches(key, value)) {
-                        if (store.seen == 1 && store.multi_replace == 0) {
-                                warning("%s has multiple values", key);
-                        } else if (store.seen >= MAX_MATCHES) {
-                                error("too many matches for %s", key);
-                                return 1;
-                        }
-
-                        store.offset[store.seen] = ftell(config_file);
-                        store.seen++;
-                }
-                break;
-        case SECTION_SEEN:
-                /*
-                 * What we are looking for is in store.key (both
-                 * section and var), and its section part is baselen
-                 * long.  We found key (again, both section and var).
-                 * We would want to know if this key is in the same
-                 * section as what we are looking for.  We already
-                 * know we are in the same section as what should
-                 * hold store.key.
-                 */
-                ep = strrchr(key, '.');
-                section_len = ep - key;
-
-                if ((section_len != store.baselen) ||
-                    memcmp(key, store.key, section_len+1)) {
-                        store.state = SECTION_END_SEEN;
-                        break;
-                }
-
-                /*
-                 * Do not increment matches: this is no match, but we
-                 * just made sure we are in the desired section.
-                 */
-                store.offset[store.seen] = ftell(config_file);
-                /* fallthru */
-        case SECTION_END_SEEN:
-        case START:
-                if (matches(key, value)) {
-                        store.offset[store.seen] = ftell(config_file);
-                        store.state = KEY_SEEN;
-                        store.seen++;
-                } else {
-                        if (strrchr(key, '.') - key == store.baselen &&
-                              !strncmp(key, store.key, store.baselen)) {
-                                        store.state = SECTION_SEEN;
-                                        store.offset[store.seen] = ftell(config_file);
-                        }
-                }
-        }
-        return 0;
+        return error("Missing value for '%s'", var);
 }
 
-static int store_write_section(int fd, const char* key)
-{
-        const char *dot;
-        int i, success;
-        struct strbuf sb = STRBUF_INIT;
-
-        dot = memchr(key, '.', store.baselen);
-        if (dot) {
-                strbuf_addf(&sb, "[%.*s \"", (int)(dot - key), key);
-                for (i = dot - key + 1; i < store.baselen; i++) {
-                        if (key[i] == '"' || key[i] == '\\')
-                                strbuf_addch(&sb, '\\');
-                        strbuf_addch(&sb, key[i]);
-                }
-                strbuf_addstr(&sb, "\"]\n");
-        } else {
-                strbuf_addf(&sb, "[%.*s]\n", store.baselen, key);
-        }
-
-        success = write_in_full(fd, sb.buf, sb.len) == sb.len;
-        strbuf_release(&sb);
-
-        return success;
-}
+struct buildid_dir_config {
+        char *dir;
+};
 
-static int store_write_pair(int fd, const char* key, const char* value)
+static int buildid_dir_command_config(const char *var, const char *value,
+                                      void *data)
 {
-        int i, success;
-        int length = strlen(key + store.baselen + 1);
-        const char *quote = "";
-        struct strbuf sb = STRBUF_INIT;
-
-        /*
-         * Check to see if the value needs to be surrounded with a dq pair.
-         * Note that problematic characters are always backslash-quoted; this
-         * check is about not losing leading or trailing SP and strings that
-         * follow beginning-of-comment characters (i.e. ';' and '#') by the
-         * configuration parser.
-         */
-        if (value[0] == ' ')
-                quote = "\"";
-        for (i = 0; value[i]; i++)
-                if (value[i] == ';' || value[i] == '#')
-                        quote = "\"";
-        if (i && value[i - 1] == ' ')
-                quote = "\"";
-
-        strbuf_addf(&sb, "\t%.*s = %s",
-                    length, key + store.baselen + 1, quote);
-
-        for (i = 0; value[i]; i++)
-                switch (value[i]) {
-                case '\n':
-                        strbuf_addstr(&sb, "\\n");
-                        break;
-                case '\t':
-                        strbuf_addstr(&sb, "\\t");
-                        break;
-                case '"':
-                case '\\':
-                        strbuf_addch(&sb, '\\');
-                default:
-                        strbuf_addch(&sb, value[i]);
-                        break;
-                }
-        strbuf_addf(&sb, "%s\n", quote);
+        struct buildid_dir_config *c = data;
+        const char *v;
 
-        success = write_in_full(fd, sb.buf, sb.len) == sb.len;
-        strbuf_release(&sb);
-
-        return success;
-}
-
-static ssize_t find_beginning_of_line(const char* contents, size_t size,
-        size_t offset_, int* found_bracket)
-{
-        size_t equal_offset = size, bracket_offset = size;
-        ssize_t offset;
-
-contline:
-        for (offset = offset_-2; offset > 0
-                        && contents[offset] != '\n'; offset--)
-                switch (contents[offset]) {
-                        case '=': equal_offset = offset; break;
-                        case ']': bracket_offset = offset; break;
-                }
-        if (offset > 0 && contents[offset-1] == '\\') {
-                offset_ = offset;
-                goto contline;
+        /* same dir for all commands */
+        if (!prefixcmp(var, "buildid.") && !strcmp(var + 8, "dir")) {
+                v = perf_config_dirname(var, value);
+                if (!v)
+                        return -1;
+                strncpy(c->dir, v, MAXPATHLEN-1);
+                c->dir[MAXPATHLEN-1] = '\0';
         }
-        if (bracket_offset < equal_offset) {
-                *found_bracket = 1;
-                offset = bracket_offset+1;
-        } else
-                offset++;
-
-        return offset;
+        return 0;
 }
 
-int perf_config_set(const char* key, const char* value)
+static void check_buildid_dir_config(void)
 {
-        return perf_config_set_multivar(key, value, NULL, 0);
+        struct buildid_dir_config c;
+        c.dir = buildid_dir;
+        perf_config(buildid_dir_command_config, &c);
 }
 
-/*
- * If value==NULL, unset in (remove from) config,
- * if value_regex!=NULL, disregard key/value pairs where value does not match.
- * if multi_replace==0, nothing, or only one matching key/value is replaced,
- *     else all matching key/values (regardless how many) are removed,
- *     before the new pair is written.
- *
- * Returns 0 on success.
- *
- * This function does this:
- *
- * - it locks the config file by creating ".perf/config.lock"
- *
- * - it then parses the config using store_aux() as validator to find
- *   the position on the key/value pair to replace. If it is to be unset,
- *   it must be found exactly once.
- *
- * - the config file is mmap()ed and the part before the match (if any) is
- *   written to the lock file, then the changed part and the rest.
- *
- * - the config file is removed and the lock file rename()d to it.
- *
- */
-int perf_config_set_multivar(const char* key, const char* value,
-        const char* value_regex, int multi_replace)
+void set_buildid_dir(void)
 {
-        int i, dot;
-        int fd = -1, in_fd;
-        int ret = 0;
-        char* config_filename;
-        const char* last_dot = strrchr(key, '.');
-
-        if (config_exclusive_filename)
-                config_filename = strdup(config_exclusive_filename);
-        else
-                config_filename = perf_pathdup("config");
-
-        /*
-         * Since "key" actually contains the section name and the real
-         * key name separated by a dot, we have to know where the dot is.
-         */
-
-        if (last_dot == NULL) {
-                error("key does not contain a section: %s", key);
-                ret = 2;
-                goto out_free;
-        }
-        store.baselen = last_dot - key;
-
-        store.multi_replace = multi_replace;
-
-        /*
-         * Validate the key and while at it, lower case it for matching.
-         */
-        store.key = malloc(strlen(key) + 1);
-        dot = 0;
-        for (i = 0; key[i]; i++) {
-                unsigned char c = key[i];
-                if (c == '.')
-                        dot = 1;
-                /* Leave the extended basename untouched.. */
-                if (!dot || i > store.baselen) {
-                        if (!iskeychar(c) || (i == store.baselen+1 && !isalpha(c))) {
-                                error("invalid key: %s", key);
-                                free(store.key);
-                                ret = 1;
-                                goto out_free;
-                        }
-                        c = tolower(c);
-                } else if (c == '\n') {
-                        error("invalid key (newline): %s", key);
-                        free(store.key);
-                        ret = 1;
-                        goto out_free;
-                }
-                store.key[i] = c;
-        }
-        store.key[i] = 0;
-
-        /*
-         * If .perf/config does not exist yet, write a minimal version.
-         */
-        in_fd = open(config_filename, O_RDONLY);
-        if ( in_fd < 0 ) {
-                free(store.key);
-
-                if ( ENOENT != errno ) {
-                        error("opening %s: %s", config_filename,
-                              strerror(errno));
-                        ret = 3; /* same as "invalid config file" */
-                        goto out_free;
-                }
-                /* if nothing to unset, error out */
-                if (value == NULL) {
-                        ret = 5;
-                        goto out_free;
-                }
+        buildid_dir[0] = '\0';
 
-                store.key = (char*)key;
-                if (!store_write_section(fd, key) ||
-                    !store_write_pair(fd, key, value))
-                        goto write_err_out;
-        } else {
-                struct stat st;
-                char* contents;
-                size_t contents_sz, copy_begin, copy_end;
-                int i, new_line = 0;
-
-                if (value_regex == NULL)
-                        store.value_regex = NULL;
-                else {
-                        if (value_regex[0] == '!') {
-                                store.do_not_match = 1;
-                                value_regex++;
-                        } else
-                                store.do_not_match = 0;
-
-                        store.value_regex = (regex_t*)malloc(sizeof(regex_t));
-                        if (regcomp(store.value_regex, value_regex,
-                                        REG_EXTENDED)) {
-                                error("invalid pattern: %s", value_regex);
-                                free(store.value_regex);
-                                ret = 6;
-                                goto out_free;
-                        }
-                }
+        /* try config file */
+        check_buildid_dir_config();
 
-                store.offset[0] = 0;
-                store.state = START;
-                store.seen = 0;
-
-                /*
-                 * After this, store.offset will contain the *end* offset
-                 * of the last match, or remain at 0 if no match was found.
-                 * As a side effect, we make sure to transform only a valid
-                 * existing config file.
-                 */
-                if (perf_config_from_file(store_aux, config_filename, NULL)) {
-                        error("invalid config file %s", config_filename);
-                        free(store.key);
-                        if (store.value_regex != NULL) {
-                                regfree(store.value_regex);
-                                free(store.value_regex);
-                        }
-                        ret = 3;
-                        goto out_free;
-                }
-
-                free(store.key);
-                if (store.value_regex != NULL) {
-                        regfree(store.value_regex);
-                        free(store.value_regex);
-                }
-
-                /* if nothing to unset, or too many matches, error out */
-                if ((store.seen == 0 && value == NULL) ||
-                                (store.seen > 1 && multi_replace == 0)) {
-                        ret = 5;
-                        goto out_free;
-                }
-
-                fstat(in_fd, &st);
-                contents_sz = xsize_t(st.st_size);
-                contents = mmap(NULL, contents_sz, PROT_READ,
-                        MAP_PRIVATE, in_fd, 0);
-                close(in_fd);
-
-                if (store.seen == 0)
-                        store.seen = 1;
-
-                for (i = 0, copy_begin = 0; i < store.seen; i++) {
-                        if (store.offset[i] == 0) {
-                                store.offset[i] = copy_end = contents_sz;
-                        } else if (store.state != KEY_SEEN) {
-                                copy_end = store.offset[i];
-                        } else
-                                copy_end = find_beginning_of_line(
-                                        contents, contents_sz,
-                                        store.offset[i]-2, &new_line);
-
-                        if (copy_end > 0 && contents[copy_end-1] != '\n')
-                                new_line = 1;
-
-                        /* write the first part of the config */
-                        if (copy_end > copy_begin) {
-                                if (write_in_full(fd, contents + copy_begin,
-                                                  copy_end - copy_begin) <
-                                    copy_end - copy_begin)
-                                        goto write_err_out;
-                                if (new_line &&
-                                    write_in_full(fd, "\n", 1) != 1)
-                                        goto write_err_out;
-                        }
-                        copy_begin = store.offset[i];
-                }
-
-                /* write the pair (value == NULL means unset) */
-                if (value != NULL) {
-                        if (store.state == START) {
-                                if (!store_write_section(fd, key))
-                                        goto write_err_out;
-                        }
-                        if (!store_write_pair(fd, key, value))
-                                goto write_err_out;
+        /* default to $HOME/.debug */
+        if (buildid_dir[0] == '\0') {
+                char *v = getenv("HOME");
+                if (v) {
+                        snprintf(buildid_dir, MAXPATHLEN-1, "%s/%s",
+                                 v, DEBUG_CACHE_DIR);
+                } else {
+                        strncpy(buildid_dir, DEBUG_CACHE_DIR, MAXPATHLEN-1);
                 }
-
-                /* write the rest of the config */
-                if (copy_begin < contents_sz)
-                        if (write_in_full(fd, contents + copy_begin,
-                                          contents_sz - copy_begin) <
-                            contents_sz - copy_begin)
-                                goto write_err_out;
-
-                munmap(contents, contents_sz);
+                buildid_dir[MAXPATHLEN-1] = '\0';
         }
-
-        ret = 0;
-
-out_free:
-        free(config_filename);
-        return ret;
-
-write_err_out:
-        goto out_free;
-
-}
-
-/*
- * Call this to report error for your variable that should not
- * get a boolean value (i.e. "[my] var" means "true").
- */
-int config_error_nonbool(const char *var)
-{
-        return error("Missing value for '%s'", var);
+        /* for communicating with external commands */
+        setenv("PERF_BUILDID_DIR", buildid_dir, 1);
 }
diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c
new file mode 100644
index 00000000000..3ccaa104338
--- /dev/null
+++ b/tools/perf/util/cpumap.c
@@ -0,0 +1,179 @@
+#include "util.h"
+#include "../perf.h"
+#include "cpumap.h"
+#include <assert.h>
+#include <stdio.h>
+
+static struct cpu_map *cpu_map__default_new(void)
+{
+        struct cpu_map *cpus;
+        int nr_cpus;
+
+        nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
+        if (nr_cpus < 0)
+                return NULL;
+
+        cpus = malloc(sizeof(*cpus) + nr_cpus * sizeof(int));
+        if (cpus != NULL) {
+                int i;
+                for (i = 0; i < nr_cpus; ++i)
+                        cpus->map[i] = i;
+
+                cpus->nr = nr_cpus;
+        }
+
+        return cpus;
+}
+
+static struct cpu_map *cpu_map__trim_new(int nr_cpus, int *tmp_cpus)
+{
+        size_t payload_size = nr_cpus * sizeof(int);
+        struct cpu_map *cpus = malloc(sizeof(*cpus) + payload_size);
+
+        if (cpus != NULL) {
+                cpus->nr = nr_cpus;
+                memcpy(cpus->map, tmp_cpus, payload_size);
+        }
+
+        return cpus;
+}
+
+static struct cpu_map *cpu_map__read_all_cpu_map(void)
+{
+        struct cpu_map *cpus = NULL;
+        FILE *onlnf;
+        int nr_cpus = 0;
+        int *tmp_cpus = NULL, *tmp;
+        int max_entries = 0;
+        int n, cpu, prev;
+        char sep;
+
+        onlnf = fopen("/sys/devices/system/cpu/online", "r");
+        if (!onlnf)
+                return cpu_map__default_new();
+
+        sep = 0;
+        prev = -1;
+        for (;;) {
+                n = fscanf(onlnf, "%u%c", &cpu, &sep);
+                if (n <= 0)
+                        break;
+                if (prev >= 0) {
+                        int new_max = nr_cpus + cpu - prev - 1;
+
+                        if (new_max >= max_entries) {
+                                max_entries = new_max + MAX_NR_CPUS / 2;
+                                tmp = realloc(tmp_cpus, max_entries * sizeof(int));
+                                if (tmp == NULL)
+                                        goto out_free_tmp;
+                                tmp_cpus = tmp;
+                        }
+
+                        while (++prev < cpu)
+                                tmp_cpus[nr_cpus++] = prev;
+                }
+                if (nr_cpus == max_entries) {
+                        max_entries += MAX_NR_CPUS;
+                        tmp = realloc(tmp_cpus, max_entries * sizeof(int));
+                        if (tmp == NULL)
+                                goto out_free_tmp;
+                        tmp_cpus = tmp;
+                }
+
+                tmp_cpus[nr_cpus++] = cpu;
+                if (n == 2 && sep == '-')
+                        prev = cpu;
+                else
+                        prev = -1;
+                if (n == 1 || sep == '\n')
+                        break;
+        }
+
+        if (nr_cpus > 0)
+                cpus = cpu_map__trim_new(nr_cpus, tmp_cpus);
+        else
+                cpus = cpu_map__default_new();
+out_free_tmp:
+        free(tmp_cpus);
+        fclose(onlnf);
+        return cpus;
+}
+
+struct cpu_map *cpu_map__new(const char *cpu_list)
+{
+        struct cpu_map *cpus = NULL;
+        unsigned long start_cpu, end_cpu = 0;
+        char *p = NULL;
+        int i, nr_cpus = 0;
+        int *tmp_cpus = NULL, *tmp;
+        int max_entries = 0;
+
+        if (!cpu_list)
+                return cpu_map__read_all_cpu_map();
+
+        if (!isdigit(*cpu_list))
+                goto out;
+
+        while (isdigit(*cpu_list)) {
+                p = NULL;
+                start_cpu = strtoul(cpu_list, &p, 0);
+                if (start_cpu >= INT_MAX
+                    || (*p != '\0' && *p != ',' && *p != '-'))
+                        goto invalid;
+
+                if (*p == '-') {
+                        cpu_list = ++p;
+                        p = NULL;
+                        end_cpu = strtoul(cpu_list, &p, 0);
+
+                        if (end_cpu >= INT_MAX || (*p != '\0' && *p != ','))
+                                goto invalid;
+
+                        if (end_cpu < start_cpu)
+                                goto invalid;
+                } else {
+                        end_cpu = start_cpu;
+                }
+
+                for (; start_cpu <= end_cpu; start_cpu++) {
+                        /* check for duplicates */
+                        for (i = 0; i < nr_cpus; i++)
+                                if (tmp_cpus[i] == (int)start_cpu)
+                                        goto invalid;
+
+                        if (nr_cpus == max_entries) {
+                                max_entries += MAX_NR_CPUS;
+                                tmp = realloc(tmp_cpus, max_entries * sizeof(int));
+                                if (tmp == NULL)
+                                        goto invalid;
+                                tmp_cpus = tmp;
+                        }
+                        tmp_cpus[nr_cpus++] = (int)start_cpu;
+                }
+                if (*p)
+                        ++p;
+
+                cpu_list = p;
+        }
+
+        if (nr_cpus > 0)
+                cpus = cpu_map__trim_new(nr_cpus, tmp_cpus);
+        else
+                cpus = cpu_map__default_new();
+invalid:
+        free(tmp_cpus);
+out:
+        return cpus;
+}
+
+struct cpu_map *cpu_map__dummy_new(void)
+{
+        struct cpu_map *cpus = malloc(sizeof(*cpus) + sizeof(int));
+
+        if (cpus != NULL) {
+                cpus->nr = 1;
+                cpus->map[0] = -1;
+        }
+
+        return cpus;
+}
diff --git a/tools/perf/util/cpumap.h b/tools/perf/util/cpumap.h
new file mode 100644
index 00000000000..f7a4f42f630
--- /dev/null
+++ b/tools/perf/util/cpumap.h
@@ -0,0 +1,13 @@
+#ifndef __PERF_CPUMAP_H
+#define __PERF_CPUMAP_H
+
+struct cpu_map {
+        int nr;
+        int map[];
+};
+
+struct cpu_map *cpu_map__new(const char *cpu_list);
+struct cpu_map *cpu_map__dummy_new(void);
+void *cpu_map__delete(struct cpu_map *map);
+
+#endif /* __PERF_CPUMAP_H */
diff --git a/tools/perf/util/ctype.c b/tools/perf/util/ctype.c
index 0b791bd346b..35073621e5d 100644
--- a/tools/perf/util/ctype.c
+++ b/tools/perf/util/ctype.c
@@ -29,3 +29,11 @@ unsigned char sane_ctype[256] = {
         A, A, A, A, A, A, A, A, A, A, A, R, R, P, P, 0,         /* 112..127 */
         /* Nothing in the 128.. range */
 };
+
+const char *graph_line =
+        "_____________________________________________________________________"
+        "_____________________________________________________________________";
+const char *graph_dotted_line =
+        "---------------------------------------------------------------------"
+        "---------------------------------------------------------------------"
+        "---------------------------------------------------------------------";
diff --git a/tools/perf/util/debug.c b/tools/perf/util/debug.c
new file mode 100644
index 00000000000..01bbe8ecec3
--- /dev/null
+++ b/tools/perf/util/debug.c
@@ -0,0 +1,94 @@
+/* For general debugging purposes */
+
+#include "../perf.h"
+
+#include <string.h>
+#include <stdarg.h>
+#include <stdio.h>
+
+#include "cache.h"
+#include "color.h"
+#include "event.h"
+#include "debug.h"
+#include "util.h"
+
+int verbose;
+bool dump_trace = false, quiet = false;
+
+int eprintf(int level, const char *fmt, ...)
+{
+        va_list args;
+        int ret = 0;
+
+        if (verbose >= level) {
+                va_start(args, fmt);
+                if (use_browser > 0)
+                        ret = ui_helpline__show_help(fmt, args);
+                else
+                        ret = vfprintf(stderr, fmt, args);
+                va_end(args);
+        }
+
+        return ret;
+}
+
+int dump_printf(const char *fmt, ...)
+{
+        va_list args;
+        int ret = 0;
+
+        if (dump_trace) {
+                va_start(args, fmt);
+                ret = vprintf(fmt, args);
+                va_end(args);
+        }
+
+        return ret;
+}
+
+#ifdef NO_NEWT_SUPPORT
+void ui__warning(const char *format, ...)
+{
+        va_list args;
+
+        va_start(args, format);
+        vfprintf(stderr, format, args);
+        va_end(args);
+}
+#endif
+
+void trace_event(event_t *event)
+{
+        unsigned char *raw_event = (void *)event;
+        const char *color = PERF_COLOR_BLUE;
+        int i, j;
+
+        if (!dump_trace)
+                return;
+
+        printf(".");
+        color_fprintf(stdout, color, "\n. ... raw event: size %d bytes\n",
+                      event->header.size);
+
+        for (i = 0; i < event->header.size; i++) {
+                if ((i & 15) == 0) {
+                        printf(".");
+                        color_fprintf(stdout, color, "  %04x: ", i);
+                }
+
+                color_fprintf(stdout, color, " %02x", raw_event[i]);
+
+                if (((i & 15) == 15) || i == event->header.size-1) {
+                        color_fprintf(stdout, color, "  ");
+                        for (j = 0; j < 15-(i & 15); j++)
+                                color_fprintf(stdout, color, "   ");
+                        for (j = i & ~15; j <= i; j++) {
+                                color_fprintf(stdout, color, "%c",
+                                              isprint(raw_event[j]) ?
+                                              raw_event[j] : '.');
+                        }
+                        color_fprintf(stdout, color, "\n");
+                }
+        }
+        printf(".\n");
+}
diff --git a/tools/perf/util/debug.h b/tools/perf/util/debug.h
new file mode 100644
index 00000000000..ca35fd66b5d
--- /dev/null
+++ b/tools/perf/util/debug.h
@@ -0,0 +1,40 @@
+/* For debugging general purposes */
+#ifndef __PERF_DEBUG_H
+#define __PERF_DEBUG_H
+
+#include <stdbool.h>
+#include "event.h"
+
+extern int verbose;
+extern bool quiet, dump_trace;
+
+int dump_printf(const char *fmt, ...) __attribute__((format(printf, 1, 2)));
+void trace_event(event_t *event);
+
+struct ui_progress;
+
+#ifdef NO_NEWT_SUPPORT
+static inline int ui_helpline__show_help(const char *format __used, va_list ap __used)
+{
+        return 0;
+}
+
+static inline struct ui_progress *ui_progress__new(const char *title __used,
+                                                   u64 total __used)
+{
+        return (struct ui_progress *)1;
+}
+
+static inline void ui_progress__update(struct ui_progress *self __used,
+                                       u64 curr __used) {}
+
+static inline void ui_progress__delete(struct ui_progress *self __used) {}
+#else
+extern char ui_helpline__last_msg[];
+int ui_helpline__show_help(const char *format, va_list ap);
+#include "ui/progress.h"
+#endif
+
+void ui__warning(const char *format, ...) __attribute__((format(printf, 1, 2)));
+
+#endif  /* __PERF_DEBUG_H */
diff --git a/tools/perf/util/debugfs.c b/tools/perf/util/debugfs.c
new file mode 100644
index 00000000000..a88fefc0cc0
--- /dev/null
+++ b/tools/perf/util/debugfs.c
@@ -0,0 +1,240 @@
+#include "util.h"
+#include "debugfs.h"
+#include "cache.h"
+
+static int debugfs_premounted;
+static char debugfs_mountpoint[MAX_PATH+1];
+
+static const char *debugfs_known_mountpoints[] = {
+        "/sys/kernel/debug/",
+        "/debug/",
+        0,
+};
+
+/* use this to force a umount */
+void debugfs_force_cleanup(void)
+{
+        debugfs_find_mountpoint();
+        debugfs_premounted = 0;
+        debugfs_umount();
+}
+
+/* construct a full path to a debugfs element */
+int debugfs_make_path(const char *element, char *buffer, int size)
+{
+        int len;
+
+        if (strlen(debugfs_mountpoint) == 0) {
+                buffer[0] = '\0';
+                return -1;
+        }
+
+        len = strlen(debugfs_mountpoint) + strlen(element) + 1;
+        if (len >= size)
+                return len+1;
+
+        snprintf(buffer, size-1, "%s/%s", debugfs_mountpoint, element);
+        return 0;
+}
+
+static int debugfs_found;
+
+/* find the path to the mounted debugfs */
+const char *debugfs_find_mountpoint(void)
+{
+        const char **ptr;
+        char type[100];
+        FILE *fp;
+
+        if (debugfs_found)
+                return (const char *) debugfs_mountpoint;
+
+        ptr = debugfs_known_mountpoints;
+        while (*ptr) {
+                if (debugfs_valid_mountpoint(*ptr) == 0) {
+                        debugfs_found = 1;
+                        strcpy(debugfs_mountpoint, *ptr);
+                        return debugfs_mountpoint;
+                }
+                ptr++;
+        }
+
+        /* give up and parse /proc/mounts */
+        fp = fopen("/proc/mounts", "r");
+        if (fp == NULL)
+                die("Can't open /proc/mounts for read");
+
+        while (fscanf(fp, "%*s %"
+                      STR(MAX_PATH)
+                      "s %99s %*s %*d %*d\n",
+                      debugfs_mountpoint, type) == 2) {
+                if (strcmp(type, "debugfs") == 0)
+                        break;
+        }
+        fclose(fp);
+
+        if (strcmp(type, "debugfs") != 0)
+                return NULL;
+
+        debugfs_found = 1;
+
+        return debugfs_mountpoint;
+}
+
+/* verify that a mountpoint is actually a debugfs instance */
+
+int debugfs_valid_mountpoint(const char *debugfs)
+{
+        struct statfs st_fs;
+
+        if (statfs(debugfs, &st_fs) < 0)
+                return -ENOENT;
+        else if (st_fs.f_type != (long) DEBUGFS_MAGIC)
+                return -ENOENT;
+
+        return 0;
+}
+
+
+int debugfs_valid_entry(const char *path)
+{
+        struct stat st;
+
+        if (stat(path, &st))
+                return -errno;
+
+        return 0;
+}
+
+/* mount the debugfs somewhere if it's not mounted */
+
+char *debugfs_mount(const char *mountpoint)
+{
+        /* see if it's already mounted */
+        if (debugfs_find_mountpoint()) {
+                debugfs_premounted = 1;
+                return debugfs_mountpoint;
+        }
+
+        /* if not mounted and no argument */
+        if (mountpoint == NULL) {
+                /* see if environment variable set */
+                mountpoint = getenv(PERF_DEBUGFS_ENVIRONMENT);
+                /* if no environment variable, use default */
+                if (mountpoint == NULL)
+                        mountpoint = "/sys/kernel/debug";
+        }
+
+        if (mount(NULL, mountpoint, "debugfs", 0, NULL) < 0)
+                return NULL;
+
+        /* save the mountpoint */
+        strncpy(debugfs_mountpoint, mountpoint, sizeof(debugfs_mountpoint));
+        debugfs_found = 1;
+
+        return debugfs_mountpoint;
+}
+
+/* umount the debugfs */
+
+int debugfs_umount(void)
+{
+        char umountcmd[128];
+        int ret;
+
+        /* if it was already mounted, leave it */
+        if (debugfs_premounted)
+                return 0;
+
+        /* make sure it's a valid mount point */
+        ret = debugfs_valid_mountpoint(debugfs_mountpoint);
+        if (ret)
+                return ret;
+
+        snprintf(umountcmd, sizeof(umountcmd),
+                 "/bin/umount %s", debugfs_mountpoint);
+        return system(umountcmd);
+}
+
+int debugfs_write(const char *entry, const char *value)
+{
+        char path[MAX_PATH+1];
+        int ret, count;
+        int fd;
+
+        /* construct the path */
+        snprintf(path, sizeof(path), "%s/%s", debugfs_mountpoint, entry);
+
+        /* verify that it exists */
+        ret = debugfs_valid_entry(path);
+        if (ret)
+                return ret;
+
+        /* get how many chars we're going to write */
+        count = strlen(value);
+
+        /* open the debugfs entry */
+        fd = open(path, O_RDWR);
+        if (fd < 0)
+                return -errno;
+
+        while (count > 0) {
+                /* write it */
+                ret = write(fd, value, count);
+                if (ret <= 0) {
+                        if (ret == EAGAIN)
+                                continue;
+                        close(fd);
+                        return -errno;
+                }
+                count -= ret;
+        }
+
+        /* close it */
+        close(fd);
+
+        /* return success */
+        return 0;
+}
+
+/*
+ * read a debugfs entry
+ * returns the number of chars read or a negative errno
+ */
+int debugfs_read(const char *entry, char *buffer, size_t size)
+{
+        char path[MAX_PATH+1];
+        int ret;
+        int fd;
+
+        /* construct the path */
+        snprintf(path, sizeof(path), "%s/%s", debugfs_mountpoint, entry);
+
+        /* verify that it exists */
+        ret = debugfs_valid_entry(path);
+        if (ret)
+                return ret;
+
+        /* open the debugfs entry */
+        fd = open(path, O_RDONLY);
+        if (fd < 0)
+                return -errno;
+
+        do {
+                /* read it */
+                ret = read(fd, buffer, size);
+                if (ret == 0) {
+                        close(fd);
+                        return EOF;
+                }
+        } while (ret < 0 && errno == EAGAIN);
+
+        /* close it */
+        close(fd);
+
+        /* make *sure* there's a null character at the end */
+        buffer[ret] = '\0';
+
+        /* return the number of chars read */
+        return ret;
+}
diff --git a/tools/perf/util/debugfs.h b/tools/perf/util/debugfs.h
new file mode 100644
index 00000000000..83a02879745
--- /dev/null
+++ b/tools/perf/util/debugfs.h
@@ -0,0 +1,25 @@
+#ifndef __DEBUGFS_H__
+#define __DEBUGFS_H__
+
+#include <sys/mount.h>
+
+#ifndef MAX_PATH
+# define MAX_PATH 256
+#endif
+
+#ifndef STR
+# define _STR(x) #x
+# define STR(x) _STR(x)
+#endif
+
+extern const char *debugfs_find_mountpoint(void);
+extern int debugfs_valid_mountpoint(const char *debugfs);
+extern int debugfs_valid_entry(const char *path);
+extern char *debugfs_mount(const char *mountpoint);
+extern int debugfs_umount(void);
+extern int debugfs_write(const char *entry, const char *value);
+extern int debugfs_read(const char *entry, char *buffer, size_t size);
+extern void debugfs_force_cleanup(void);
+extern int debugfs_make_path(const char *element, char *buffer, int size);
+
+#endif /* __DEBUGFS_H__ */
diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c
new file mode 100644
index 00000000000..2302ec051bb
--- /dev/null
+++ b/tools/perf/util/event.c
@@ -0,0 +1,952 @@
+#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 const char *event__name[] = {
+        [0]                      = "TOTAL",
+        [PERF_RECORD_MMAP]       = "MMAP",
+        [PERF_RECORD_LOST]       = "LOST",
+        [PERF_RECORD_COMM]       = "COMM",
+        [PERF_RECORD_EXIT]       = "EXIT",
+        [PERF_RECORD_THROTTLE]   = "THROTTLE",
+        [PERF_RECORD_UNTHROTTLE] = "UNTHROTTLE",
+        [PERF_RECORD_FORK]       = "FORK",
+        [PERF_RECORD_READ]       = "READ",
+        [PERF_RECORD_SAMPLE]     = "SAMPLE",
+        [PERF_RECORD_HEADER_ATTR]        = "ATTR",
+        [PERF_RECORD_HEADER_EVENT_TYPE]  = "EVENT_TYPE",
+        [PERF_RECORD_HEADER_TRACING_DATA]        = "TRACING_DATA",
+        [PERF_RECORD_HEADER_BUILD_ID]    = "BUILD_ID",
+        [PERF_RECORD_FINISHED_ROUND]     = "FINISHED_ROUND",
+};
+
+const char *event__get_event_name(unsigned int id)
+{
+        if (id >= ARRAY_SIZE(event__name))
+                return "INVALID";
+        if (!event__name[id])
+                return "UNKNOWN";
+        return event__name[id];
+}
+
+static struct sample_data synth_sample = {
+        .pid       = -1,
+        .tid       = -1,
+        .time      = -1,
+        .stream_id = -1,
+        .cpu       = -1,
+        .period    = 1,
+};
+
+static pid_t event__synthesize_comm(event_t *event, pid_t pid, int full,
+                                    event__handler_t process,
+                                    struct perf_session *session)
+{
+        char filename[PATH_MAX];
+        char bf[BUFSIZ];
+        FILE *fp;
+        size_t size = 0;
+        DIR *tasks;
+        struct dirent dirent, *next;
+        pid_t tgid = 0;
+
+        snprintf(filename, sizeof(filename), "/proc/%d/status", pid);
+
+        fp = fopen(filename, "r");
+        if (fp == NULL) {
+out_race:
+                /*
+                 * We raced with a task exiting - just return:
+                 */
+                pr_debug("couldn't open %s\n", filename);
+                return 0;
+        }
+
+        memset(&event->comm, 0, sizeof(event->comm));
+
+        while (!event->comm.comm[0] || !event->comm.pid) {
+                if (fgets(bf, sizeof(bf), fp) == NULL) {
+                        pr_warning("couldn't get COMM and pgid, malformed %s\n", filename);
+                        goto out;
+                }
+
+                if (memcmp(bf, "Name:", 5) == 0) {
+                        char *name = bf + 5;
+                        while (*name && isspace(*name))
+                                ++name;
+                        size = strlen(name) - 1;
+                        memcpy(event->comm.comm, name, size++);
+                } else if (memcmp(bf, "Tgid:", 5) == 0) {
+                        char *tgids = bf + 5;
+                        while (*tgids && isspace(*tgids))
+                                ++tgids;
+                        tgid = event->comm.pid = atoi(tgids);
+                }
+        }
+
+        event->comm.header.type = PERF_RECORD_COMM;
+        size = ALIGN(size, sizeof(u64));
+        memset(event->comm.comm + size, 0, session->id_hdr_size);
+        event->comm.header.size = (sizeof(event->comm) -
+                                (sizeof(event->comm.comm) - size) +
+                                session->id_hdr_size);
+        if (!full) {
+                event->comm.tid = pid;
+
+                process(event, &synth_sample, session);
+                goto out;
+        }
+
+        snprintf(filename, sizeof(filename), "/proc/%d/task", pid);
+
+        tasks = opendir(filename);
+        if (tasks == NULL)
+                goto out_race;
+
+        while (!readdir_r(tasks, &dirent, &next) && next) {
+                char *end;
+                pid = strtol(dirent.d_name, &end, 10);
+                if (*end)
+                        continue;
+
+                event->comm.tid = pid;
+
+                process(event, &synth_sample, session);
+        }
+
+        closedir(tasks);
+out:
+        fclose(fp);
+
+        return tgid;
+}
+
+static int event__synthesize_mmap_events(event_t *event, pid_t pid, pid_t tgid,
+                                         event__handler_t process,
+                                         struct perf_session *session)
+{
+        char filename[PATH_MAX];
+        FILE *fp;
+
+        snprintf(filename, sizeof(filename), "/proc/%d/maps", pid);
+
+        fp = fopen(filename, "r");
+        if (fp == NULL) {
+                /*
+                 * We raced with a task exiting - just return:
+                 */
+                pr_debug("couldn't open %s\n", filename);
+                return -1;
+        }
+
+        event->header.type = PERF_RECORD_MMAP;
+        /*
+         * Just like the kernel, see __perf_event_mmap in kernel/perf_event.c
+         */
+        event->header.misc = PERF_RECORD_MISC_USER;
+
+        while (1) {
+                char bf[BUFSIZ], *pbf = bf;
+                int n;
+                size_t size;
+                if (fgets(bf, sizeof(bf), fp) == NULL)
+                        break;
+
+                /* 00400000-0040c000 r-xp 00000000 fd:01 41038  /bin/cat */
+                n = hex2u64(pbf, &event->mmap.start);
+                if (n < 0)
+                        continue;
+                pbf += n + 1;
+                n = hex2u64(pbf, &event->mmap.len);
+                if (n < 0)
+                        continue;
+                pbf += n + 3;
+                if (*pbf == 'x') { /* vm_exec */
+                        char *execname = strchr(bf, '/');
+
+                        /* Catch VDSO */
+                        if (execname == NULL)
+                                execname = strstr(bf, "[vdso]");
+
+                        if (execname == NULL)
+                                continue;
+
+                        pbf += 3;
+                        n = hex2u64(pbf, &event->mmap.pgoff);
+
+                        size = strlen(execname);
+                        execname[size - 1] = '\0'; /* Remove \n */
+                        memcpy(event->mmap.filename, execname, size);
+                        size = ALIGN(size, sizeof(u64));
+                        event->mmap.len -= event->mmap.start;
+                        event->mmap.header.size = (sizeof(event->mmap) -
+                                                (sizeof(event->mmap.filename) - size));
+                        memset(event->mmap.filename + size, 0, session->id_hdr_size);
+                        event->mmap.header.size += session->id_hdr_size;
+                        event->mmap.pid = tgid;
+                        event->mmap.tid = pid;
+
+                        process(event, &synth_sample, session);
+                }
+        }
+
+        fclose(fp);
+        return 0;
+}
+
+int event__synthesize_modules(event__handler_t process,
+                              struct perf_session *session,
+                              struct machine *machine)
+{
+        struct rb_node *nd;
+        struct map_groups *kmaps = &machine->kmaps;
+        event_t *event = zalloc(sizeof(event->mmap) + session->id_hdr_size);
+
+        if (event == NULL) {
+                pr_debug("Not enough memory synthesizing mmap event "
+                         "for kernel modules\n");
+                return -1;
+        }
+
+        event->header.type = PERF_RECORD_MMAP;
+
+        /*
+         * kernel uses 0 for user space maps, see kernel/perf_event.c
+         * __perf_event_mmap
+         */
+        if (machine__is_host(machine))
+                event->header.misc = PERF_RECORD_MISC_KERNEL;
+        else
+                event->header.misc = PERF_RECORD_MISC_GUEST_KERNEL;
+
+        for (nd = rb_first(&kmaps->maps[MAP__FUNCTION]);
+             nd; nd = rb_next(nd)) {
+                size_t size;
+                struct map *pos = rb_entry(nd, struct map, rb_node);
+
+                if (pos->dso->kernel)
+                        continue;
+
+                size = ALIGN(pos->dso->long_name_len + 1, sizeof(u64));
+                event->mmap.header.type = PERF_RECORD_MMAP;
+                event->mmap.header.size = (sizeof(event->mmap) -
+                                        (sizeof(event->mmap.filename) - size));
+                memset(event->mmap.filename + size, 0, session->id_hdr_size);
+                event->mmap.header.size += session->id_hdr_size;
+                event->mmap.start = pos->start;
+                event->mmap.len   = pos->end - pos->start;
+                event->mmap.pid   = machine->pid;
+
+                memcpy(event->mmap.filename, pos->dso->long_name,
+                       pos->dso->long_name_len + 1);
+                process(event, &synth_sample, session);
+        }
+
+        free(event);
+        return 0;
+}
+
+static int __event__synthesize_thread(event_t *comm_event, event_t *mmap_event,
+                                      pid_t pid, event__handler_t process,
+                                      struct perf_session *session)
+{
+        pid_t tgid = event__synthesize_comm(comm_event, pid, 1, process,
+                                            session);
+        if (tgid == -1)
+                return -1;
+        return event__synthesize_mmap_events(mmap_event, pid, tgid,
+                                             process, session);
+}
+
+int event__synthesize_thread(pid_t pid, event__handler_t process,
+                             struct perf_session *session)
+{
+        event_t *comm_event, *mmap_event;
+        int err = -1;
+
+        comm_event = malloc(sizeof(comm_event->comm) + session->id_hdr_size);
+        if (comm_event == NULL)
+                goto out;
+
+        mmap_event = malloc(sizeof(mmap_event->mmap) + session->id_hdr_size);
+        if (mmap_event == NULL)
+                goto out_free_comm;
+
+        err = __event__synthesize_thread(comm_event, mmap_event, pid,
+                                         process, session);
+        free(mmap_event);
+out_free_comm:
+        free(comm_event);
+out:
+        return err;
+}
+
+int event__synthesize_threads(event__handler_t process,
+                              struct perf_session *session)
+{
+        DIR *proc;
+        struct dirent dirent, *next;
+        event_t *comm_event, *mmap_event;
+        int err = -1;
+
+        comm_event = malloc(sizeof(comm_event->comm) + session->id_hdr_size);
+        if (comm_event == NULL)
+                goto out;
+
+        mmap_event = malloc(sizeof(mmap_event->mmap) + session->id_hdr_size);
+        if (mmap_event == NULL)
+                goto out_free_comm;
+
+        proc = opendir("/proc");
+        if (proc == NULL)
+                goto out_free_mmap;
+
+        while (!readdir_r(proc, &dirent, &next) && next) {
+                char *end;
+                pid_t pid = strtol(dirent.d_name, &end, 10);
+
+                if (*end) /* only interested in proper numerical dirents */
+                        continue;
+
+                __event__synthesize_thread(comm_event, mmap_event, pid,
+                                           process, session);
+        }
+
+        closedir(proc);
+        err = 0;
+out_free_mmap:
+        free(mmap_event);
+out_free_comm:
+        free(comm_event);
+out:
+        return err;
+}
+
+struct process_symbol_args {
+        const char *name;
+        u64        start;
+};
+
+static int find_symbol_cb(void *arg, const char *name, char type,
+                          u64 start, u64 end __used)
+{
+        struct process_symbol_args *args = arg;
+
+        /*
+         * Must be a function or at least an alias, as in PARISC64, where "_text" is
+         * an 'A' to the same address as "_stext".
+         */
+        if (!(symbol_type__is_a(type, MAP__FUNCTION) ||
+              type == 'A') || strcmp(name, args->name))
+                return 0;
+
+        args->start = start;
+        return 1;
+}
+
+int event__synthesize_kernel_mmap(event__handler_t process,
+                                  struct perf_session *session,
+                                  struct machine *machine,
+                                  const char *symbol_name)
+{
+        size_t size;
+        const char *filename, *mmap_name;
+        char path[PATH_MAX];
+        char name_buff[PATH_MAX];
+        struct map *map;
+        int err;
+        /*
+         * We should get this from /sys/kernel/sections/.text, but till that is
+         * available use this, and after it is use this as a fallback for older
+         * kernels.
+         */
+        struct process_symbol_args args = { .name = symbol_name, };
+        event_t *event = zalloc(sizeof(event->mmap) + session->id_hdr_size);
+
+        if (event == NULL) {
+                pr_debug("Not enough memory synthesizing mmap event "
+                         "for kernel modules\n");
+                return -1;
+        }
+
+        mmap_name = machine__mmap_name(machine, name_buff, sizeof(name_buff));
+        if (machine__is_host(machine)) {
+                /*
+                 * kernel uses PERF_RECORD_MISC_USER for user space maps,
+                 * see kernel/perf_event.c __perf_event_mmap
+                 */
+                event->header.misc = PERF_RECORD_MISC_KERNEL;
+                filename = "/proc/kallsyms";
+        } else {
+                event->header.misc = PERF_RECORD_MISC_GUEST_KERNEL;
+                if (machine__is_default_guest(machine))
+                        filename = (char *) symbol_conf.default_guest_kallsyms;
+                else {
+                        sprintf(path, "%s/proc/kallsyms", machine->root_dir);
+                        filename = path;
+                }
+        }
+
+        if (kallsyms__parse(filename, &args, find_symbol_cb) <= 0)
+                return -ENOENT;
+
+        map = machine->vmlinux_maps[MAP__FUNCTION];
+        size = snprintf(event->mmap.filename, sizeof(event->mmap.filename),
+                        "%s%s", mmap_name, symbol_name) + 1;
+        size = ALIGN(size, sizeof(u64));
+        event->mmap.header.type = PERF_RECORD_MMAP;
+        event->mmap.header.size = (sizeof(event->mmap) -
+                        (sizeof(event->mmap.filename) - size) + session->id_hdr_size);
+        event->mmap.pgoff = args.start;
+        event->mmap.start = map->start;
+        event->mmap.len   = map->end - event->mmap.start;
+        event->mmap.pid   = machine->pid;
+
+        err = process(event, &synth_sample, session);
+        free(event);
+
+        return err;
+}
+
+static void thread__comm_adjust(struct thread *self, struct hists *hists)
+{
+        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))) {
+                u16 slen = strlen(comm);
+
+                if (hists__new_col_len(hists, HISTC_COMM, slen))
+                        hists__set_col_len(hists, HISTC_THREAD, slen + 6);
+        }
+}
+
+static int thread__set_comm_adjust(struct thread *self, const char *comm,
+                                   struct hists *hists)
+{
+        int ret = thread__set_comm(self, comm);
+
+        if (ret)
+                return ret;
+
+        thread__comm_adjust(self, hists);
+
+        return 0;
+}
+
+int event__process_comm(event_t *self, struct sample_data *sample __used,
+                        struct perf_session *session)
+{
+        struct thread *thread = perf_session__findnew(session, self->comm.tid);
+
+        dump_printf(": %s:%d\n", self->comm.comm, self->comm.tid);
+
+        if (thread == NULL || thread__set_comm_adjust(thread, self->comm.comm,
+                                                      &session->hists)) {
+                dump_printf("problem processing PERF_RECORD_COMM, skipping event.\n");
+                return -1;
+        }
+
+        return 0;
+}
+
+int event__process_lost(event_t *self, struct sample_data *sample __used,
+                        struct perf_session *session)
+{
+        dump_printf(": id:%Ld: lost:%Ld\n", self->lost.id, self->lost.lost);
+        session->hists.stats.total_lost += self->lost.lost;
+        return 0;
+}
+
+static void event_set_kernel_mmap_len(struct map **maps, event_t *self)
+{
+        maps[MAP__FUNCTION]->start = self->mmap.start;
+        maps[MAP__FUNCTION]->end   = self->mmap.start + self->mmap.len;
+        /*
+         * Be a bit paranoid here, some perf.data file came with
+         * a zero sized synthesized MMAP event for the kernel.
+         */
+        if (maps[MAP__FUNCTION]->end == 0)
+                maps[MAP__FUNCTION]->end = ~0ULL;
+}
+
+static int event__process_kernel_mmap(event_t *self,
+                        struct perf_session *session)
+{
+        struct map *map;
+        char kmmap_prefix[PATH_MAX];
+        struct machine *machine;
+        enum dso_kernel_type kernel_type;
+        bool is_kernel_mmap;
+
+        machine = perf_session__findnew_machine(session, self->mmap.pid);
+        if (!machine) {
+                pr_err("Can't find id %d's machine\n", self->mmap.pid);
+                goto out_problem;
+        }
+
+        machine__mmap_name(machine, kmmap_prefix, sizeof(kmmap_prefix));
+        if (machine__is_host(machine))
+                kernel_type = DSO_TYPE_KERNEL;
+        else
+                kernel_type = DSO_TYPE_GUEST_KERNEL;
+
+        is_kernel_mmap = memcmp(self->mmap.filename,
+                                kmmap_prefix,
+                                strlen(kmmap_prefix)) == 0;
+        if (self->mmap.filename[0] == '/' ||
+            (!is_kernel_mmap && self->mmap.filename[0] == '[')) {
+
+                char short_module_name[1024];
+                char *name, *dot;
+
+                if (self->mmap.filename[0] == '/') {
+                        name = strrchr(self->mmap.filename, '/');
+                        if (name == NULL)
+                                goto out_problem;
+
+                        ++name; /* skip / */
+                        dot = strrchr(name, '.');
+                        if (dot == NULL)
+                                goto out_problem;
+                        snprintf(short_module_name, sizeof(short_module_name),
+                                        "[%.*s]", (int)(dot - name), name);
+                        strxfrchar(short_module_name, '-', '_');
+                } else
+                        strcpy(short_module_name, self->mmap.filename);
+
+                map = machine__new_module(machine, self->mmap.start,
+                                          self->mmap.filename);
+                if (map == NULL)
+                        goto out_problem;
+
+                name = strdup(short_module_name);
+                if (name == NULL)
+                        goto out_problem;
+
+                map->dso->short_name = name;
+                map->dso->sname_alloc = 1;
+                map->end = map->start + self->mmap.len;
+        } else if (is_kernel_mmap) {
+                const char *symbol_name = (self->mmap.filename +
+                                strlen(kmmap_prefix));
+                /*
+                 * Should be there already, from the build-id table in
+                 * the header.
+                 */
+                struct dso *kernel = __dsos__findnew(&machine->kernel_dsos,
+                                                     kmmap_prefix);
+                if (kernel == NULL)
+                        goto out_problem;
+
+                kernel->kernel = kernel_type;
+                if (__machine__create_kernel_maps(machine, kernel) < 0)
+                        goto out_problem;
+
+                event_set_kernel_mmap_len(machine->vmlinux_maps, self);
+                perf_session__set_kallsyms_ref_reloc_sym(machine->vmlinux_maps,
+                                                         symbol_name,
+                                                         self->mmap.pgoff);
+                if (machine__is_default_guest(machine)) {
+                        /*
+                         * preload dso of guest kernel and modules
+                         */
+                        dso__load(kernel, machine->vmlinux_maps[MAP__FUNCTION],
+                                  NULL);
+                }
+        }
+        return 0;
+out_problem:
+        return -1;
+}
+
+int event__process_mmap(event_t *self, struct sample_data *sample __used,
+                        struct perf_session *session)
+{
+        struct machine *machine;
+        struct thread *thread;
+        struct map *map;
+        u8 cpumode = self->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+        int ret = 0;
+
+        dump_printf(" %d/%d: [%#Lx(%#Lx) @ %#Lx]: %s\n",
+                        self->mmap.pid, self->mmap.tid, self->mmap.start,
+                        self->mmap.len, self->mmap.pgoff, self->mmap.filename);
+
+        if (cpumode == PERF_RECORD_MISC_GUEST_KERNEL ||
+            cpumode == PERF_RECORD_MISC_KERNEL) {
+                ret = event__process_kernel_mmap(self, session);
+                if (ret < 0)
+                        goto out_problem;
+                return 0;
+        }
+
+        machine = perf_session__find_host_machine(session);
+        if (machine == NULL)
+                goto out_problem;
+        thread = perf_session__findnew(session, self->mmap.pid);
+        if (thread == NULL)
+                goto out_problem;
+        map = map__new(&machine->user_dsos, self->mmap.start,
+                        self->mmap.len, self->mmap.pgoff,
+                        self->mmap.pid, self->mmap.filename,
+                        MAP__FUNCTION);
+        if (map == NULL)
+                goto out_problem;
+
+        thread__insert_map(thread, map);
+        return 0;
+
+out_problem:
+        dump_printf("problem processing PERF_RECORD_MMAP, skipping event.\n");
+        return 0;
+}
+
+int event__process_task(event_t *self, struct sample_data *sample __used,
+                        struct perf_session *session)
+{
+        struct thread *thread = perf_session__findnew(session, self->fork.tid);
+        struct thread *parent = perf_session__findnew(session, self->fork.ptid);
+
+        dump_printf("(%d:%d):(%d:%d)\n", self->fork.pid, self->fork.tid,
+                    self->fork.ppid, self->fork.ptid);
+
+        if (self->header.type == PERF_RECORD_EXIT) {
+                perf_session__remove_thread(session, thread);
+                return 0;
+        }
+
+        if (thread == NULL || parent == NULL ||
+            thread__fork(thread, parent) < 0) {
+                dump_printf("problem processing PERF_RECORD_FORK, skipping event.\n");
+                return -1;
+        }
+
+        return 0;
+}
+
+int event__process(event_t *event, struct sample_data *sample,
+                   struct perf_session *session)
+{
+        switch (event->header.type) {
+        case PERF_RECORD_COMM:
+                event__process_comm(event, sample, session);
+                break;
+        case PERF_RECORD_MMAP:
+                event__process_mmap(event, sample, session);
+                break;
+        case PERF_RECORD_FORK:
+        case PERF_RECORD_EXIT:
+                event__process_task(event, sample, session);
+                break;
+        default:
+                break;
+        }
+
+        return 0;
+}
+
+void thread__find_addr_map(struct thread *self,
+                           struct perf_session *session, u8 cpumode,
+                           enum map_type type, pid_t pid, u64 addr,
+                           struct addr_location *al)
+{
+        struct map_groups *mg = &self->mg;
+        struct machine *machine = NULL;
+
+        al->thread = self;
+        al->addr = addr;
+        al->cpumode = cpumode;
+        al->filtered = false;
+
+        if (cpumode == PERF_RECORD_MISC_KERNEL && perf_host) {
+                al->level = 'k';
+                machine = perf_session__find_host_machine(session);
+                if (machine == NULL) {
+                        al->map = NULL;
+                        return;
+                }
+                mg = &machine->kmaps;
+        } else if (cpumode == PERF_RECORD_MISC_USER && perf_host) {
+                al->level = '.';
+                machine = perf_session__find_host_machine(session);
+        } else if (cpumode == PERF_RECORD_MISC_GUEST_KERNEL && perf_guest) {
+                al->level = 'g';
+                machine = perf_session__find_machine(session, pid);
+                if (machine == NULL) {
+                        al->map = NULL;
+                        return;
+                }
+                mg = &machine->kmaps;
+        } else {
+                /*
+                 * 'u' means guest os user space.
+                 * TODO: We don't support guest user space. Might support late.
+                 */
+                if (cpumode == PERF_RECORD_MISC_GUEST_USER && perf_guest)
+                        al->level = 'u';
+                else
+                        al->level = 'H';
+                al->map = NULL;
+
+                if ((cpumode == PERF_RECORD_MISC_GUEST_USER ||
+                        cpumode == PERF_RECORD_MISC_GUEST_KERNEL) &&
+                        !perf_guest)
+                        al->filtered = true;
+                if ((cpumode == PERF_RECORD_MISC_USER ||
+                        cpumode == PERF_RECORD_MISC_KERNEL) &&
+                        !perf_host)
+                        al->filtered = true;
+
+                return;
+        }
+try_again:
+        al->map = map_groups__find(mg, type, al->addr);
+        if (al->map == NULL) {
+                /*
+                 * If this is outside of all known maps, and is a negative
+                 * address, try to look it up in the kernel dso, as it might be
+                 * a vsyscall or vdso (which executes in user-mode).
+                 *
+                 * XXX This is nasty, we should have a symbol list in the
+                 * "[vdso]" dso, but for now lets use the old trick of looking
+                 * in the whole kernel symbol list.
+                 */
+                if ((long long)al->addr < 0 &&
+                    cpumode == PERF_RECORD_MISC_KERNEL &&
+                    machine && mg != &machine->kmaps) {
+                        mg = &machine->kmaps;
+                        goto try_again;
+                }
+        } else
+                al->addr = al->map->map_ip(al->map, al->addr);
+}
+
+void thread__find_addr_location(struct thread *self,
+                                struct perf_session *session, u8 cpumode,
+                                enum map_type type, pid_t pid, u64 addr,
+                                struct addr_location *al,
+                                symbol_filter_t filter)
+{
+        thread__find_addr_map(self, session, cpumode, type, pid, addr, al);
+        if (al->map != NULL)
+                al->sym = map__find_symbol(al->map, al->addr, filter);
+        else
+                al->sym = NULL;
+}
+
+static void dso__calc_col_width(struct dso *self, struct hists *hists)
+{
+        if (!symbol_conf.col_width_list_str && !symbol_conf.field_sep &&
+            (!symbol_conf.dso_list ||
+             strlist__has_entry(symbol_conf.dso_list, self->name))) {
+                u16 slen = dso__name_len(self);
+                hists__new_col_len(hists, HISTC_DSO, slen);
+        }
+
+        self->slen_calculated = 1;
+}
+
+int event__preprocess_sample(const event_t *self, struct perf_session *session,
+                             struct addr_location *al, struct sample_data *data,
+                             symbol_filter_t filter)
+{
+        u8 cpumode = self->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+        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);
+        /*
+         * Have we already created the kernel maps for the host machine?
+         *
+         * This should have happened earlier, when we processed the kernel MMAP
+         * events, but for older perf.data files there was no such thing, so do
+         * it now.
+         */
+        if (cpumode == PERF_RECORD_MISC_KERNEL &&
+            session->host_machine.vmlinux_maps[MAP__FUNCTION] == NULL)
+                machine__create_kernel_maps(&session->host_machine);
+
+        thread__find_addr_map(thread, session, cpumode, MAP__FUNCTION,
+                              self->ip.pid, self->ip.ip, al);
+        dump_printf(" ...... dso: %s\n",
+                    al->map ? al->map->dso->long_name :
+                        al->level == 'H' ? "[hypervisor]" : "<not found>");
+        al->sym = NULL;
+        al->cpu = data->cpu;
+
+        if (al->map) {
+                if (symbol_conf.dso_list &&
+                    (!al->map || !al->map->dso ||
+                     !(strlist__has_entry(symbol_conf.dso_list,
+                                          al->map->dso->short_name) ||
+                       (al->map->dso->short_name != al->map->dso->long_name &&
+                        strlist__has_entry(symbol_conf.dso_list,
+                                           al->map->dso->long_name)))))
+                        goto out_filtered;
+                /*
+                 * 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 (!sort_dso.elide && !al->map->dso->slen_calculated)
+                        dso__calc_col_width(al->map->dso, &session->hists);
+
+                al->sym = map__find_symbol(al->map, al->addr, filter);
+        } else {
+                const unsigned int unresolved_col_width = BITS_PER_LONG / 4;
+
+                if (hists__col_len(&session->hists, HISTC_DSO) < unresolved_col_width &&
+                    !symbol_conf.col_width_list_str && !symbol_conf.field_sep &&
+                    !symbol_conf.dso_list)
+                        hists__set_col_len(&session->hists, HISTC_DSO,
+                                           unresolved_col_width);
+        }
+
+        if (symbol_conf.sym_list && al->sym &&
+            !strlist__has_entry(symbol_conf.sym_list, al->sym->name))
+                goto out_filtered;
+
+        return 0;
+
+out_filtered:
+        al->filtered = true;
+        return 0;
+}
+
+static int event__parse_id_sample(const event_t *event,
+                                  struct perf_session *session,
+                                  struct sample_data *sample)
+{
+        const u64 *array;
+        u64 type;
+
+        sample->cpu = sample->pid = sample->tid = -1;
+        sample->stream_id = sample->id = sample->time = -1ULL;
+
+        if (!session->sample_id_all)
+                return 0;
+
+        array = event->sample.array;
+        array += ((event->header.size -
+                   sizeof(event->header)) / sizeof(u64)) - 1;
+        type = session->sample_type;
+
+        if (type & PERF_SAMPLE_CPU) {
+                u32 *p = (u32 *)array;
+                sample->cpu = *p;
+                array--;
+        }
+
+        if (type & PERF_SAMPLE_STREAM_ID) {
+                sample->stream_id = *array;
+                array--;
+        }
+
+        if (type & PERF_SAMPLE_ID) {
+                sample->id = *array;
+                array--;
+        }
+
+        if (type & PERF_SAMPLE_TIME) {
+                sample->time = *array;
+                array--;
+        }
+
+        if (type & PERF_SAMPLE_TID) {
+                u32 *p = (u32 *)array;
+                sample->pid = p[0];
+                sample->tid = p[1];
+        }
+
+        return 0;
+}
+
+int event__parse_sample(const event_t *event, struct perf_session *session,
+                        struct sample_data *data)
+{
+        const u64 *array;
+        u64 type;
+
+        if (event->header.type != PERF_RECORD_SAMPLE)
+                return event__parse_id_sample(event, session, data);
+
+        array = event->sample.array;
+        type = session->sample_type;
+
+        if (type & PERF_SAMPLE_IP) {
+                data->ip = event->ip.ip;
+                array++;
+        }
+
+        if (type & PERF_SAMPLE_TID) {
+                u32 *p = (u32 *)array;
+                data->pid = p[0];
+                data->tid = p[1];
+                array++;
+        }
+
+        if (type & PERF_SAMPLE_TIME) {
+                data->time = *array;
+                array++;
+        }
+
+        if (type & PERF_SAMPLE_ADDR) {
+                data->addr = *array;
+                array++;
+        }
+
+        data->id = -1ULL;
+        if (type & PERF_SAMPLE_ID) {
+                data->id = *array;
+                array++;
+        }
+
+        if (type & PERF_SAMPLE_STREAM_ID) {
+                data->stream_id = *array;
+                array++;
+        }
+
+        if (type & PERF_SAMPLE_CPU) {
+                u32 *p = (u32 *)array;
+                data->cpu = *p;
+                array++;
+        } else
+                data->cpu = -1;
+
+        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
new file mode 100644
index 00000000000..2b7e91902f1
--- /dev/null
+++ b/tools/perf/util/event.h
@@ -0,0 +1,177 @@
+#ifndef __PERF_RECORD_H
+#define __PERF_RECORD_H
+
+#include <limits.h>
+
+#include "../perf.h"
+#include "map.h"
+
+/*
+ * PERF_SAMPLE_IP | PERF_SAMPLE_TID | *
+ */
+struct ip_event {
+        struct perf_event_header header;
+        u64 ip;
+        u32 pid, tid;
+        unsigned char __more_data[];
+};
+
+struct mmap_event {
+        struct perf_event_header header;
+        u32 pid, tid;
+        u64 start;
+        u64 len;
+        u64 pgoff;
+        char filename[PATH_MAX];
+};
+
+struct comm_event {
+        struct perf_event_header header;
+        u32 pid, tid;
+        char comm[16];
+};
+
+struct fork_event {
+        struct perf_event_header header;
+        u32 pid, ppid;
+        u32 tid, ptid;
+        u64 time;
+};
+
+struct lost_event {
+        struct perf_event_header header;
+        u64 id;
+        u64 lost;
+};
+
+/*
+ * PERF_FORMAT_ENABLED | PERF_FORMAT_RUNNING | PERF_FORMAT_ID
+ */
+struct read_event {
+        struct perf_event_header header;
+        u32 pid, tid;
+        u64 value;
+        u64 time_enabled;
+        u64 time_running;
+        u64 id;
+};
+
+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;
+        u64 period;
+        u32 cpu;
+        u32 raw_size;
+        void *raw_data;
+        struct ip_callchain *callchain;
+};
+
+#define BUILD_ID_SIZE 20
+
+struct build_id_event {
+        struct perf_event_header header;
+        pid_t                    pid;
+        u8                       build_id[ALIGN(BUILD_ID_SIZE, sizeof(u64))];
+        char                     filename[];
+};
+
+enum perf_user_event_type { /* above any possible kernel type */
+        PERF_RECORD_USER_TYPE_START             = 64,
+        PERF_RECORD_HEADER_ATTR                 = 64,
+        PERF_RECORD_HEADER_EVENT_TYPE           = 65,
+        PERF_RECORD_HEADER_TRACING_DATA         = 66,
+        PERF_RECORD_HEADER_BUILD_ID             = 67,
+        PERF_RECORD_FINISHED_ROUND              = 68,
+        PERF_RECORD_HEADER_MAX
+};
+
+struct attr_event {
+        struct perf_event_header header;
+        struct perf_event_attr attr;
+        u64 id[];
+};
+
+#define MAX_EVENT_NAME 64
+
+struct perf_trace_event_type {
+        u64     event_id;
+        char    name[MAX_EVENT_NAME];
+};
+
+struct event_type_event {
+        struct perf_event_header header;
+        struct perf_trace_event_type event_type;
+};
+
+struct tracing_data_event {
+        struct perf_event_header header;
+        u32 size;
+};
+
+typedef union event_union {
+        struct perf_event_header        header;
+        struct ip_event                 ip;
+        struct mmap_event               mmap;
+        struct comm_event               comm;
+        struct fork_event               fork;
+        struct lost_event               lost;
+        struct read_event               read;
+        struct sample_event             sample;
+        struct attr_event               attr;
+        struct event_type_event         event_type;
+        struct tracing_data_event       tracing_data;
+        struct build_id_event           build_id;
+} event_t;
+
+void event__print_totals(void);
+
+struct perf_session;
+
+typedef int (*event__handler_synth_t)(event_t *event, 
+                                      struct perf_session *session);
+typedef int (*event__handler_t)(event_t *event, struct sample_data *sample,
+                                struct perf_session *session);
+
+int event__synthesize_thread(pid_t pid, event__handler_t process,
+                             struct perf_session *session);
+int event__synthesize_threads(event__handler_t process,
+                              struct perf_session *session);
+int event__synthesize_kernel_mmap(event__handler_t process,
+                                struct perf_session *session,
+                                struct machine *machine,
+                                const char *symbol_name);
+
+int event__synthesize_modules(event__handler_t process,
+                              struct perf_session *session,
+                              struct machine *machine);
+
+int event__process_comm(event_t *self, struct sample_data *sample,
+                        struct perf_session *session);
+int event__process_lost(event_t *self, struct sample_data *sample,
+                        struct perf_session *session);
+int event__process_mmap(event_t *self, struct sample_data *sample,
+                        struct perf_session *session);
+int event__process_task(event_t *self, struct sample_data *sample,
+                        struct perf_session *session);
+int event__process(event_t *event, struct sample_data *sample,
+                   struct perf_session *session);
+
+struct addr_location;
+int event__preprocess_sample(const event_t *self, struct perf_session *session,
+                             struct addr_location *al, struct sample_data *data,
+                             symbol_filter_t filter);
+int event__parse_sample(const event_t *event, struct perf_session *session,
+                        struct sample_data *sample);
+
+const char *event__get_event_name(unsigned int id);
+
+#endif /* __PERF_RECORD_H */
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
new file mode 100644
index 00000000000..f5cfed60af9
--- /dev/null
+++ b/tools/perf/util/evsel.c
@@ -0,0 +1,201 @@
+#include "evsel.h"
+#include "../perf.h"
+#include "util.h"
+#include "cpumap.h"
+#include "thread.h"
+
+#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y))
+
+struct perf_evsel *perf_evsel__new(struct perf_event_attr *attr, int idx)
+{
+        struct perf_evsel *evsel = zalloc(sizeof(*evsel));
+
+        if (evsel != NULL) {
+                evsel->idx         = idx;
+                evsel->attr        = *attr;
+                INIT_LIST_HEAD(&evsel->node);
+        }
+
+        return evsel;
+}
+
+int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
+{
+        evsel->fd = xyarray__new(ncpus, nthreads, sizeof(int));
+        return evsel->fd != NULL ? 0 : -ENOMEM;
+}
+
+int perf_evsel__alloc_counts(struct perf_evsel *evsel, int ncpus)
+{
+        evsel->counts = zalloc((sizeof(*evsel->counts) +
+                                (ncpus * sizeof(struct perf_counts_values))));
+        return evsel->counts != NULL ? 0 : -ENOMEM;
+}
+
+void perf_evsel__free_fd(struct perf_evsel *evsel)
+{
+        xyarray__delete(evsel->fd);
+        evsel->fd = NULL;
+}
+
+void perf_evsel__close_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
+{
+        int cpu, thread;
+
+        for (cpu = 0; cpu < ncpus; cpu++)
+                for (thread = 0; thread < nthreads; ++thread) {
+                        close(FD(evsel, cpu, thread));
+                        FD(evsel, cpu, thread) = -1;
+                }
+}
+
+void perf_evsel__delete(struct perf_evsel *evsel)
+{
+        assert(list_empty(&evsel->node));
+        xyarray__delete(evsel->fd);
+        free(evsel);
+}
+
+int __perf_evsel__read_on_cpu(struct perf_evsel *evsel,
+                              int cpu, int thread, bool scale)
+{
+        struct perf_counts_values count;
+        size_t nv = scale ? 3 : 1;
+
+        if (FD(evsel, cpu, thread) < 0)
+                return -EINVAL;
+
+        if (evsel->counts == NULL && perf_evsel__alloc_counts(evsel, cpu + 1) < 0)
+                return -ENOMEM;
+
+        if (readn(FD(evsel, cpu, thread), &count, nv * sizeof(u64)) < 0)
+                return -errno;
+
+        if (scale) {
+                if (count.run == 0)
+                        count.val = 0;
+                else if (count.run < count.ena)
+                        count.val = (u64)((double)count.val * count.ena / count.run + 0.5);
+        } else
+                count.ena = count.run = 0;
+
+        evsel->counts->cpu[cpu] = count;
+        return 0;
+}
+
+int __perf_evsel__read(struct perf_evsel *evsel,
+                       int ncpus, int nthreads, bool scale)
+{
+        size_t nv = scale ? 3 : 1;
+        int cpu, thread;
+        struct perf_counts_values *aggr = &evsel->counts->aggr, count;
+
+        aggr->val = 0;
+
+        for (cpu = 0; cpu < ncpus; cpu++) {
+                for (thread = 0; thread < nthreads; thread++) {
+                        if (FD(evsel, cpu, thread) < 0)
+                                continue;
+
+                        if (readn(FD(evsel, cpu, thread),
+                                  &count, nv * sizeof(u64)) < 0)
+                                return -errno;
+
+                        aggr->val += count.val;
+                        if (scale) {
+                                aggr->ena += count.ena;
+                                aggr->run += count.run;
+                        }
+                }
+        }
+
+        evsel->counts->scaled = 0;
+        if (scale) {
+                if (aggr->run == 0) {
+                        evsel->counts->scaled = -1;
+                        aggr->val = 0;
+                        return 0;
+                }
+
+                if (aggr->run < aggr->ena) {
+                        evsel->counts->scaled = 1;
+                        aggr->val = (u64)((double)aggr->val * aggr->ena / aggr->run + 0.5);
+                }
+        } else
+                aggr->ena = aggr->run = 0;
+
+        return 0;
+}
+
+static int __perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
+                              struct thread_map *threads)
+{
+        int cpu, thread;
+
+        if (evsel->fd == NULL &&
+            perf_evsel__alloc_fd(evsel, cpus->nr, threads->nr) < 0)
+                return -1;
+
+        for (cpu = 0; cpu < cpus->nr; cpu++) {
+                for (thread = 0; thread < threads->nr; thread++) {
+                        FD(evsel, cpu, thread) = sys_perf_event_open(&evsel->attr,
+                                                                     threads->map[thread],
+                                                                     cpus->map[cpu], -1, 0);
+                        if (FD(evsel, cpu, thread) < 0)
+                                goto out_close;
+                }
+        }
+
+        return 0;
+
+out_close:
+        do {
+                while (--thread >= 0) {
+                        close(FD(evsel, cpu, thread));
+                        FD(evsel, cpu, thread) = -1;
+                }
+                thread = threads->nr;
+        } while (--cpu >= 0);
+        return -1;
+}
+
+static struct {
+        struct cpu_map map;
+        int cpus[1];
+} empty_cpu_map = {
+        .map.nr = 1,
+        .cpus   = { -1, },
+};
+
+static struct {
+        struct thread_map map;
+        int threads[1];
+} empty_thread_map = {
+        .map.nr  = 1,
+        .threads = { -1, },
+};
+
+int perf_evsel__open(struct perf_evsel *evsel,
+                     struct cpu_map *cpus, struct thread_map *threads)
+{
+
+        if (cpus == NULL) {
+                /* Work around old compiler warnings about strict aliasing */
+                cpus = &empty_cpu_map.map;
+        }
+
+        if (threads == NULL)
+                threads = &empty_thread_map.map;
+
+        return __perf_evsel__open(evsel, cpus, threads);
+}
+
+int perf_evsel__open_per_cpu(struct perf_evsel *evsel, struct cpu_map *cpus)
+{
+        return __perf_evsel__open(evsel, cpus, &empty_thread_map.map);
+}
+
+int perf_evsel__open_per_thread(struct perf_evsel *evsel, struct thread_map *threads)
+{
+        return __perf_evsel__open(evsel, &empty_cpu_map.map, threads);
+}
diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
new file mode 100644
index 00000000000..b2d755fe88a
--- /dev/null
+++ b/tools/perf/util/evsel.h
@@ -0,0 +1,115 @@
+#ifndef __PERF_EVSEL_H
+#define __PERF_EVSEL_H 1
+
+#include <linux/list.h>
+#include <stdbool.h>
+#include "../../../include/linux/perf_event.h"
+#include "types.h"
+#include "xyarray.h"
+ 
+struct perf_counts_values {
+        union {
+                struct {
+                        u64 val;
+                        u64 ena;
+                        u64 run;
+                };
+                u64 values[3];
+        };
+};
+
+struct perf_counts {
+        s8                        scaled;
+        struct perf_counts_values aggr;
+        struct perf_counts_values cpu[];
+};
+
+struct perf_evsel {
+        struct list_head        node;
+        struct perf_event_attr  attr;
+        char                    *filter;
+        struct xyarray          *fd;
+        struct perf_counts      *counts;
+        int                     idx;
+        void                    *priv;
+};
+
+struct cpu_map;
+struct thread_map;
+
+struct perf_evsel *perf_evsel__new(struct perf_event_attr *attr, int idx);
+void perf_evsel__delete(struct perf_evsel *evsel);
+
+int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads);
+int perf_evsel__alloc_counts(struct perf_evsel *evsel, int ncpus);
+void perf_evsel__free_fd(struct perf_evsel *evsel);
+void perf_evsel__close_fd(struct perf_evsel *evsel, int ncpus, int nthreads);
+
+int perf_evsel__open_per_cpu(struct perf_evsel *evsel, struct cpu_map *cpus);
+int perf_evsel__open_per_thread(struct perf_evsel *evsel, struct thread_map *threads);
+int perf_evsel__open(struct perf_evsel *evsel, 
+                     struct cpu_map *cpus, struct thread_map *threads);
+
+#define perf_evsel__match(evsel, t, c)          \
+        (evsel->attr.type == PERF_TYPE_##t &&   \
+         evsel->attr.config == PERF_COUNT_##c)
+
+int __perf_evsel__read_on_cpu(struct perf_evsel *evsel,
+                              int cpu, int thread, bool scale);
+
+/**
+ * perf_evsel__read_on_cpu - Read out the results on a CPU and thread
+ *
+ * @evsel - event selector to read value
+ * @cpu - CPU of interest
+ * @thread - thread of interest
+ */
+static inline int perf_evsel__read_on_cpu(struct perf_evsel *evsel,
+                                          int cpu, int thread)
+{
+        return __perf_evsel__read_on_cpu(evsel, cpu, thread, false);
+}
+
+/**
+ * perf_evsel__read_on_cpu_scaled - Read out the results on a CPU and thread, scaled
+ *
+ * @evsel - event selector to read value
+ * @cpu - CPU of interest
+ * @thread - thread of interest
+ */
+static inline int perf_evsel__read_on_cpu_scaled(struct perf_evsel *evsel,
+                                                 int cpu, int thread)
+{
+        return __perf_evsel__read_on_cpu(evsel, cpu, thread, true);
+}
+
+int __perf_evsel__read(struct perf_evsel *evsel, int ncpus, int nthreads,
+                       bool scale);
+
+/**
+ * perf_evsel__read - Read the aggregate results on all CPUs
+ *
+ * @evsel - event selector to read value
+ * @ncpus - Number of cpus affected, from zero
+ * @nthreads - Number of threads affected, from zero
+ */
+static inline int perf_evsel__read(struct perf_evsel *evsel,
+                                    int ncpus, int nthreads)
+{
+        return __perf_evsel__read(evsel, ncpus, nthreads, false);
+}
+
+/**
+ * perf_evsel__read_scaled - Read the aggregate results on all CPUs, scaled
+ *
+ * @evsel - event selector to read value
+ * @ncpus - Number of cpus affected, from zero
+ * @nthreads - Number of threads affected, from zero
+ */
+static inline int perf_evsel__read_scaled(struct perf_evsel *evsel,
+                                          int ncpus, int nthreads)
+{
+        return __perf_evsel__read(evsel, ncpus, nthreads, true);
+}
+
+#endif /* __PERF_EVSEL_H */
diff --git a/tools/perf/util/exec_cmd.c b/tools/perf/util/exec_cmd.c
index d3929226315..67eeff57156 100644
--- a/tools/perf/util/exec_cmd.c
+++ b/tools/perf/util/exec_cmd.c
@@ -1,9 +1,11 @@
 #include "cache.h"
 #include "exec_cmd.h"
 #include "quote.h"
+
+#include <string.h>
+
 #define MAX_ARGS        32
 
-extern char **environ;
 static const char *argv_exec_path;
 static const char *argv0_path;
 
@@ -52,7 +54,7 @@ const char *perf_extract_argv0_path(const char *argv0)
 
         if (slash >= argv0) {
                 argv0_path = strndup(argv0, slash - argv0);
-                return slash + 1;
+                return argv0_path ? slash + 1 : NULL;
         }
 
         return argv0;
@@ -114,7 +116,7 @@ void setup_path(void)
         strbuf_release(&new_path);
 }
 
-const char **prepare_perf_cmd(const char **argv)
+static const char **prepare_perf_cmd(const char **argv)
 {
         int argc;
         const char **nargv;
diff --git a/tools/perf/util/exec_cmd.h b/tools/perf/util/exec_cmd.h
index effe25eb154..bc4b915963f 100644
--- a/tools/perf/util/exec_cmd.h
+++ b/tools/perf/util/exec_cmd.h
@@ -1,13 +1,12 @@
-#ifndef PERF_EXEC_CMD_H
-#define PERF_EXEC_CMD_H
+#ifndef __PERF_EXEC_CMD_H
+#define __PERF_EXEC_CMD_H
 
 extern void perf_set_argv_exec_path(const char *exec_path);
 extern const char *perf_extract_argv0_path(const char *path);
 extern const char *perf_exec_path(void);
 extern void setup_path(void);
-extern const char **prepare_perf_cmd(const char **argv);
 extern int execv_perf_cmd(const char **argv); /* NULL terminated */
 extern int execl_perf_cmd(const char *cmd, ...);
 extern const char *system_path(const char *path);
 
-#endif /* PERF_EXEC_CMD_H */
+#endif /* __PERF_EXEC_CMD_H */
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
new file mode 100644
index 00000000000..989fa2dee2f
--- /dev/null
+++ b/tools/perf/util/header.c
@@ -0,0 +1,1232 @@
+#define _FILE_OFFSET_BITS 64
+
+#include <sys/types.h>
+#include <byteswap.h>
+#include <unistd.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <linux/list.h>
+#include <linux/kernel.h>
+
+#include "util.h"
+#include "header.h"
+#include "../perf.h"
+#include "trace-event.h"
+#include "session.h"
+#include "symbol.h"
+#include "debug.h"
+
+static bool no_buildid_cache = false;
+
+/*
+ * Create new perf.data header attribute:
+ */
+struct perf_header_attr *perf_header_attr__new(struct perf_event_attr *attr)
+{
+        struct perf_header_attr *self = malloc(sizeof(*self));
+
+        if (self != NULL) {
+                self->attr = *attr;
+                self->ids  = 0;
+                self->size = 1;
+                self->id   = malloc(sizeof(u64));
+                if (self->id == NULL) {
+                        free(self);
+                        self = NULL;
+                }
+        }
+
+        return self;
+}
+
+void perf_header_attr__delete(struct perf_header_attr *self)
+{
+        free(self->id);
+        free(self);
+}
+
+int perf_header_attr__add_id(struct perf_header_attr *self, u64 id)
+{
+        int pos = self->ids;
+
+        self->ids++;
+        if (self->ids > self->size) {
+                int nsize = self->size * 2;
+                u64 *nid = realloc(self->id, nsize * sizeof(u64));
+
+                if (nid == NULL)
+                        return -1;
+
+                self->size = nsize;
+                self->id = nid;
+        }
+        self->id[pos] = id;
+        return 0;
+}
+
+int perf_header__init(struct perf_header *self)
+{
+        self->size = 1;
+        self->attr = malloc(sizeof(void *));
+        return self->attr == NULL ? -ENOMEM : 0;
+}
+
+void perf_header__exit(struct perf_header *self)
+{
+        int i;
+        for (i = 0; i < self->attrs; ++i)
+                perf_header_attr__delete(self->attr[i]);
+        free(self->attr);
+}
+
+int perf_header__add_attr(struct perf_header *self,
+                          struct perf_header_attr *attr)
+{
+        if (self->frozen)
+                return -1;
+
+        if (self->attrs == self->size) {
+                int nsize = self->size * 2;
+                struct perf_header_attr **nattr;
+
+                nattr = realloc(self->attr, nsize * sizeof(void *));
+                if (nattr == NULL)
+                        return -1;
+
+                self->size = nsize;
+                self->attr = nattr;
+        }
+
+        self->attr[self->attrs++] = attr;
+        return 0;
+}
+
+static int event_count;
+static struct perf_trace_event_type *events;
+
+int perf_header__push_event(u64 id, const char *name)
+{
+        if (strlen(name) > MAX_EVENT_NAME)
+                pr_warning("Event %s will be truncated\n", name);
+
+        if (!events) {
+                events = malloc(sizeof(struct perf_trace_event_type));
+                if (events == NULL)
+                        return -ENOMEM;
+        } else {
+                struct perf_trace_event_type *nevents;
+
+                nevents = realloc(events, (event_count + 1) * sizeof(*events));
+                if (nevents == NULL)
+                        return -ENOMEM;
+                events = nevents;
+        }
+        memset(&events[event_count], 0, sizeof(struct perf_trace_event_type));
+        events[event_count].event_id = id;
+        strncpy(events[event_count].name, name, MAX_EVENT_NAME - 1);
+        event_count++;
+        return 0;
+}
+
+char *perf_header__find_event(u64 id)
+{
+        int i;
+        for (i = 0 ; i < event_count; i++) {
+                if (events[i].event_id == id)
+                        return events[i].name;
+        }
+        return NULL;
+}
+
+static const char *__perf_magic = "PERFFILE";
+
+#define PERF_MAGIC      (*(u64 *)__perf_magic)
+
+struct perf_file_attr {
+        struct perf_event_attr  attr;
+        struct perf_file_section        ids;
+};
+
+void perf_header__set_feat(struct perf_header *self, int feat)
+{
+        set_bit(feat, self->adds_features);
+}
+
+void perf_header__clear_feat(struct perf_header *self, int feat)
+{
+        clear_bit(feat, self->adds_features);
+}
+
+bool perf_header__has_feat(const struct perf_header *self, int feat)
+{
+        return test_bit(feat, self->adds_features);
+}
+
+static int do_write(int fd, const void *buf, size_t size)
+{
+        while (size) {
+                int ret = write(fd, buf, size);
+
+                if (ret < 0)
+                        return -errno;
+
+                size -= ret;
+                buf += ret;
+        }
+
+        return 0;
+}
+
+#define NAME_ALIGN 64
+
+static int write_padded(int fd, const void *bf, size_t count,
+                        size_t count_aligned)
+{
+        static const char zero_buf[NAME_ALIGN];
+        int err = do_write(fd, bf, count);
+
+        if (!err)
+                err = do_write(fd, zero_buf, count_aligned - count);
+
+        return err;
+}
+
+#define dsos__for_each_with_build_id(pos, head) \
+        list_for_each_entry(pos, head, node)    \
+                if (!pos->has_build_id)         \
+                        continue;               \
+                else
+
+static int __dsos__write_buildid_table(struct list_head *head, pid_t pid,
+                                u16 misc, int fd)
+{
+        struct dso *pos;
+
+        dsos__for_each_with_build_id(pos, head) {
+                int err;
+                struct build_id_event b;
+                size_t len;
+
+                if (!pos->hit)
+                        continue;
+                len = pos->long_name_len + 1;
+                len = ALIGN(len, NAME_ALIGN);
+                memset(&b, 0, sizeof(b));
+                memcpy(&b.build_id, pos->build_id, sizeof(pos->build_id));
+                b.pid = pid;
+                b.header.misc = misc;
+                b.header.size = sizeof(b) + len;
+                err = do_write(fd, &b, sizeof(b));
+                if (err < 0)
+                        return err;
+                err = write_padded(fd, pos->long_name,
+                                   pos->long_name_len + 1, len);
+                if (err < 0)
+                        return err;
+        }
+
+        return 0;
+}
+
+static int machine__write_buildid_table(struct machine *self, int fd)
+{
+        int err;
+        u16 kmisc = PERF_RECORD_MISC_KERNEL,
+            umisc = PERF_RECORD_MISC_USER;
+
+        if (!machine__is_host(self)) {
+                kmisc = PERF_RECORD_MISC_GUEST_KERNEL;
+                umisc = PERF_RECORD_MISC_GUEST_USER;
+        }
+
+        err = __dsos__write_buildid_table(&self->kernel_dsos, self->pid,
+                                          kmisc, fd);
+        if (err == 0)
+                err = __dsos__write_buildid_table(&self->user_dsos,
+                                                  self->pid, umisc, fd);
+        return err;
+}
+
+static int dsos__write_buildid_table(struct perf_header *header, int fd)
+{
+        struct perf_session *session = container_of(header,
+                        struct perf_session, header);
+        struct rb_node *nd;
+        int err = machine__write_buildid_table(&session->host_machine, fd);
+
+        if (err)
+                return err;
+
+        for (nd = rb_first(&session->machines); nd; nd = rb_next(nd)) {
+                struct machine *pos = rb_entry(nd, struct machine, rb_node);
+                err = machine__write_buildid_table(pos, fd);
+                if (err)
+                        break;
+        }
+        return err;
+}
+
+int build_id_cache__add_s(const char *sbuild_id, const char *debugdir,
+                          const char *name, bool is_kallsyms)
+{
+        const size_t size = PATH_MAX;
+        char *realname = realpath(name, NULL),
+             *filename = malloc(size),
+             *linkname = malloc(size), *targetname;
+        int len, err = -1;
+
+        if (realname == NULL || filename == NULL || linkname == NULL)
+                goto out_free;
+
+        len = snprintf(filename, size, "%s%s%s",
+                       debugdir, is_kallsyms ? "/" : "", realname);
+        if (mkdir_p(filename, 0755))
+                goto out_free;
+
+        snprintf(filename + len, sizeof(filename) - len, "/%s", sbuild_id);
+
+        if (access(filename, F_OK)) {
+                if (is_kallsyms) {
+                         if (copyfile("/proc/kallsyms", filename))
+                                goto out_free;
+                } else if (link(realname, filename) && copyfile(name, filename))
+                        goto out_free;
+        }
+
+        len = snprintf(linkname, size, "%s/.build-id/%.2s",
+                       debugdir, sbuild_id);
+
+        if (access(linkname, X_OK) && mkdir_p(linkname, 0755))
+                goto out_free;
+
+        snprintf(linkname + len, size - len, "/%s", sbuild_id + 2);
+        targetname = filename + strlen(debugdir) - 5;
+        memcpy(targetname, "../..", 5);
+
+        if (symlink(targetname, linkname) == 0)
+                err = 0;
+out_free:
+        free(realname);
+        free(filename);
+        free(linkname);
+        return err;
+}
+
+static int build_id_cache__add_b(const u8 *build_id, size_t build_id_size,
+                                 const char *name, const char *debugdir,
+                                 bool is_kallsyms)
+{
+        char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+        build_id__sprintf(build_id, build_id_size, sbuild_id);
+
+        return build_id_cache__add_s(sbuild_id, debugdir, name, is_kallsyms);
+}
+
+int build_id_cache__remove_s(const char *sbuild_id, const char *debugdir)
+{
+        const size_t size = PATH_MAX;
+        char *filename = malloc(size),
+             *linkname = malloc(size);
+        int err = -1;
+
+        if (filename == NULL || linkname == NULL)
+                goto out_free;
+
+        snprintf(linkname, size, "%s/.build-id/%.2s/%s",
+                 debugdir, sbuild_id, sbuild_id + 2);
+
+        if (access(linkname, F_OK))
+                goto out_free;
+
+        if (readlink(linkname, filename, size) < 0)
+                goto out_free;
+
+        if (unlink(linkname))
+                goto out_free;
+
+        /*
+         * Since the link is relative, we must make it absolute:
+         */
+        snprintf(linkname, size, "%s/.build-id/%.2s/%s",
+                 debugdir, sbuild_id, filename);
+
+        if (unlink(linkname))
+                goto out_free;
+
+        err = 0;
+out_free:
+        free(filename);
+        free(linkname);
+        return err;
+}
+
+static int dso__cache_build_id(struct dso *self, const char *debugdir)
+{
+        bool is_kallsyms = self->kernel && self->long_name[0] != '/';
+
+        return build_id_cache__add_b(self->build_id, sizeof(self->build_id),
+                                     self->long_name, debugdir, is_kallsyms);
+}
+
+static int __dsos__cache_build_ids(struct list_head *head, const char *debugdir)
+{
+        struct dso *pos;
+        int err = 0;
+
+        dsos__for_each_with_build_id(pos, head)
+                if (dso__cache_build_id(pos, debugdir))
+                        err = -1;
+
+        return err;
+}
+
+static int machine__cache_build_ids(struct machine *self, const char *debugdir)
+{
+        int ret = __dsos__cache_build_ids(&self->kernel_dsos, debugdir);
+        ret |= __dsos__cache_build_ids(&self->user_dsos, debugdir);
+        return ret;
+}
+
+static int perf_session__cache_build_ids(struct perf_session *self)
+{
+        struct rb_node *nd;
+        int ret;
+        char debugdir[PATH_MAX];
+
+        snprintf(debugdir, sizeof(debugdir), "%s", buildid_dir);
+
+        if (mkdir(debugdir, 0755) != 0 && errno != EEXIST)
+                return -1;
+
+        ret = machine__cache_build_ids(&self->host_machine, debugdir);
+
+        for (nd = rb_first(&self->machines); nd; nd = rb_next(nd)) {
+                struct machine *pos = rb_entry(nd, struct machine, rb_node);
+                ret |= machine__cache_build_ids(pos, debugdir);
+        }
+        return ret ? -1 : 0;
+}
+
+static bool machine__read_build_ids(struct machine *self, bool with_hits)
+{
+        bool ret = __dsos__read_build_ids(&self->kernel_dsos, with_hits);
+        ret |= __dsos__read_build_ids(&self->user_dsos, with_hits);
+        return ret;
+}
+
+static bool perf_session__read_build_ids(struct perf_session *self, bool with_hits)
+{
+        struct rb_node *nd;
+        bool ret = machine__read_build_ids(&self->host_machine, with_hits);
+
+        for (nd = rb_first(&self->machines); nd; nd = rb_next(nd)) {
+                struct machine *pos = rb_entry(nd, struct machine, rb_node);
+                ret |= machine__read_build_ids(pos, with_hits);
+        }
+
+        return ret;
+}
+
+static int perf_header__adds_write(struct perf_header *self, int fd)
+{
+        int nr_sections;
+        struct perf_session *session;
+        struct perf_file_section *feat_sec;
+        int sec_size;
+        u64 sec_start;
+        int idx = 0, err;
+
+        session = container_of(self, struct perf_session, header);
+
+        if (perf_header__has_feat(self, HEADER_BUILD_ID &&
+            !perf_session__read_build_ids(session, true)))
+                perf_header__clear_feat(self, HEADER_BUILD_ID);
+
+        nr_sections = bitmap_weight(self->adds_features, HEADER_FEAT_BITS);
+        if (!nr_sections)
+                return 0;
+
+        feat_sec = calloc(sizeof(*feat_sec), nr_sections);
+        if (feat_sec == NULL)
+                return -ENOMEM;
+
+        sec_size = sizeof(*feat_sec) * nr_sections;
+
+        sec_start = self->data_offset + self->data_size;
+        lseek(fd, sec_start + sec_size, SEEK_SET);
+
+        if (perf_header__has_feat(self, HEADER_TRACE_INFO)) {
+                struct perf_file_section *trace_sec;
+
+                trace_sec = &feat_sec[idx++];
+
+                /* Write trace info */
+                trace_sec->offset = lseek(fd, 0, SEEK_CUR);
+                read_tracing_data(fd, &evsel_list);
+                trace_sec->size = lseek(fd, 0, SEEK_CUR) - trace_sec->offset;
+        }
+
+        if (perf_header__has_feat(self, HEADER_BUILD_ID)) {
+                struct perf_file_section *buildid_sec;
+
+                buildid_sec = &feat_sec[idx++];
+
+                /* Write build-ids */
+                buildid_sec->offset = lseek(fd, 0, SEEK_CUR);
+                err = dsos__write_buildid_table(self, fd);
+                if (err < 0) {
+                        pr_debug("failed to write buildid table\n");
+                        goto out_free;
+                }
+                buildid_sec->size = lseek(fd, 0, SEEK_CUR) -
+                                          buildid_sec->offset;
+                if (!no_buildid_cache)
+                        perf_session__cache_build_ids(session);
+        }
+
+        lseek(fd, sec_start, SEEK_SET);
+        err = do_write(fd, feat_sec, sec_size);
+        if (err < 0)
+                pr_debug("failed to write feature section\n");
+out_free:
+        free(feat_sec);
+        return err;
+}
+
+int perf_header__write_pipe(int fd)
+{
+        struct perf_pipe_file_header f_header;
+        int err;
+
+        f_header = (struct perf_pipe_file_header){
+                .magic     = PERF_MAGIC,
+                .size      = sizeof(f_header),
+        };
+
+        err = do_write(fd, &f_header, sizeof(f_header));
+        if (err < 0) {
+                pr_debug("failed to write perf pipe header\n");
+                return err;
+        }
+
+        return 0;
+}
+
+int perf_header__write(struct perf_header *self, int fd, bool at_exit)
+{
+        struct perf_file_header f_header;
+        struct perf_file_attr   f_attr;
+        struct perf_header_attr *attr;
+        int i, err;
+
+        lseek(fd, sizeof(f_header), SEEK_SET);
+
+        for (i = 0; i < self->attrs; i++) {
+                attr = self->attr[i];
+
+                attr->id_offset = lseek(fd, 0, SEEK_CUR);
+                err = do_write(fd, attr->id, attr->ids * sizeof(u64));
+                if (err < 0) {
+                        pr_debug("failed to write perf header\n");
+                        return err;
+                }
+        }
+
+
+        self->attr_offset = lseek(fd, 0, SEEK_CUR);
+
+        for (i = 0; i < self->attrs; i++) {
+                attr = self->attr[i];
+
+                f_attr = (struct perf_file_attr){
+                        .attr = attr->attr,
+                        .ids  = {
+                                .offset = attr->id_offset,
+                                .size   = attr->ids * sizeof(u64),
+                        }
+                };
+                err = do_write(fd, &f_attr, sizeof(f_attr));
+                if (err < 0) {
+                        pr_debug("failed to write perf header attribute\n");
+                        return err;
+                }
+        }
+
+        self->event_offset = lseek(fd, 0, SEEK_CUR);
+        self->event_size = event_count * sizeof(struct perf_trace_event_type);
+        if (events) {
+                err = do_write(fd, events, self->event_size);
+                if (err < 0) {
+                        pr_debug("failed to write perf header events\n");
+                        return err;
+                }
+        }
+
+        self->data_offset = lseek(fd, 0, SEEK_CUR);
+
+        if (at_exit) {
+                err = perf_header__adds_write(self, fd);
+                if (err < 0)
+                        return err;
+        }
+
+        f_header = (struct perf_file_header){
+                .magic     = PERF_MAGIC,
+                .size      = sizeof(f_header),
+                .attr_size = sizeof(f_attr),
+                .attrs = {
+                        .offset = self->attr_offset,
+                        .size   = self->attrs * sizeof(f_attr),
+                },
+                .data = {
+                        .offset = self->data_offset,
+                        .size   = self->data_size,
+                },
+                .event_types = {
+                        .offset = self->event_offset,
+                        .size   = self->event_size,
+                },
+        };
+
+        memcpy(&f_header.adds_features, &self->adds_features, sizeof(self->adds_features));
+
+        lseek(fd, 0, SEEK_SET);
+        err = do_write(fd, &f_header, sizeof(f_header));
+        if (err < 0) {
+                pr_debug("failed to write perf header\n");
+                return err;
+        }
+        lseek(fd, self->data_offset + self->data_size, SEEK_SET);
+
+        self->frozen = 1;
+        return 0;
+}
+
+static int perf_header__getbuffer64(struct perf_header *self,
+                                    int fd, void *buf, size_t size)
+{
+        if (readn(fd, buf, size) <= 0)
+                return -1;
+
+        if (self->needs_swap)
+                mem_bswap_64(buf, size);
+
+        return 0;
+}
+
+int perf_header__process_sections(struct perf_header *self, int fd,
+                                  int (*process)(struct perf_file_section *self,
+                                                 struct perf_header *ph,
+                                                 int feat, int fd))
+{
+        struct perf_file_section *feat_sec;
+        int nr_sections;
+        int sec_size;
+        int idx = 0;
+        int err = -1, feat = 1;
+
+        nr_sections = bitmap_weight(self->adds_features, HEADER_FEAT_BITS);
+        if (!nr_sections)
+                return 0;
+
+        feat_sec = calloc(sizeof(*feat_sec), nr_sections);
+        if (!feat_sec)
+                return -1;
+
+        sec_size = sizeof(*feat_sec) * nr_sections;
+
+        lseek(fd, self->data_offset + self->data_size, SEEK_SET);
+
+        if (perf_header__getbuffer64(self, fd, feat_sec, sec_size))
+                goto out_free;
+
+        err = 0;
+        while (idx < nr_sections && feat < HEADER_LAST_FEATURE) {
+                if (perf_header__has_feat(self, feat)) {
+                        struct perf_file_section *sec = &feat_sec[idx++];
+
+                        err = process(sec, self, feat, fd);
+                        if (err < 0)
+                                break;
+                }
+                ++feat;
+        }
+out_free:
+        free(feat_sec);
+        return err;
+}
+
+int perf_file_header__read(struct perf_file_header *self,
+                           struct perf_header *ph, int fd)
+{
+        lseek(fd, 0, SEEK_SET);
+
+        if (readn(fd, self, sizeof(*self)) <= 0 ||
+            memcmp(&self->magic, __perf_magic, sizeof(self->magic)))
+                return -1;
+
+        if (self->attr_size != sizeof(struct perf_file_attr)) {
+                u64 attr_size = bswap_64(self->attr_size);
+
+                if (attr_size != sizeof(struct perf_file_attr))
+                        return -1;
+
+                mem_bswap_64(self, offsetof(struct perf_file_header,
+                                            adds_features));
+                ph->needs_swap = true;
+        }
+
+        if (self->size != sizeof(*self)) {
+                /* Support the previous format */
+                if (self->size == offsetof(typeof(*self), adds_features))
+                        bitmap_zero(self->adds_features, HEADER_FEAT_BITS);
+                else
+                        return -1;
+        }
+
+        memcpy(&ph->adds_features, &self->adds_features,
+               sizeof(ph->adds_features));
+        /*
+         * FIXME: hack that assumes that if we need swap the perf.data file
+         * may be coming from an arch with a different word-size, ergo different
+         * DEFINE_BITMAP format, investigate more later, but for now its mostly
+         * safe to assume that we have a build-id section. Trace files probably
+         * have several other issues in this realm anyway...
+         */
+        if (ph->needs_swap) {
+                memset(&ph->adds_features, 0, sizeof(ph->adds_features));
+                perf_header__set_feat(ph, HEADER_BUILD_ID);
+        }
+
+        ph->event_offset = self->event_types.offset;
+        ph->event_size   = self->event_types.size;
+        ph->data_offset  = self->data.offset;
+        ph->data_size    = self->data.size;
+        return 0;
+}
+
+static int __event_process_build_id(struct build_id_event *bev,
+                                    char *filename,
+                                    struct perf_session *session)
+{
+        int err = -1;
+        struct list_head *head;
+        struct machine *machine;
+        u16 misc;
+        struct dso *dso;
+        enum dso_kernel_type dso_type;
+
+        machine = perf_session__findnew_machine(session, bev->pid);
+        if (!machine)
+                goto out;
+
+        misc = bev->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+
+        switch (misc) {
+        case PERF_RECORD_MISC_KERNEL:
+                dso_type = DSO_TYPE_KERNEL;
+                head = &machine->kernel_dsos;
+                break;
+        case PERF_RECORD_MISC_GUEST_KERNEL:
+                dso_type = DSO_TYPE_GUEST_KERNEL;
+                head = &machine->kernel_dsos;
+                break;
+        case PERF_RECORD_MISC_USER:
+        case PERF_RECORD_MISC_GUEST_USER:
+                dso_type = DSO_TYPE_USER;
+                head = &machine->user_dsos;
+                break;
+        default:
+                goto out;
+        }
+
+        dso = __dsos__findnew(head, filename);
+        if (dso != NULL) {
+                char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+                dso__set_build_id(dso, &bev->build_id);
+
+                if (filename[0] == '[')
+                        dso->kernel = dso_type;
+
+                build_id__sprintf(dso->build_id, sizeof(dso->build_id),
+                                  sbuild_id);
+                pr_debug("build id event received for %s: %s\n",
+                         dso->long_name, sbuild_id);
+        }
+
+        err = 0;
+out:
+        return err;
+}
+
+static int perf_header__read_build_ids(struct perf_header *self,
+                        int input, u64 offset, u64 size)
+{
+        struct perf_session *session = container_of(self,
+                        struct perf_session, header);
+        struct build_id_event bev;
+        char filename[PATH_MAX];
+        u64 limit = offset + size;
+        int err = -1;
+
+        while (offset < limit) {
+                ssize_t len;
+
+                if (read(input, &bev, sizeof(bev)) != sizeof(bev))
+                        goto out;
+
+                if (self->needs_swap)
+                        perf_event_header__bswap(&bev.header);
+
+                len = bev.header.size - sizeof(bev);
+                if (read(input, filename, len) != len)
+                        goto out;
+
+                __event_process_build_id(&bev, filename, session);
+
+                offset += bev.header.size;
+        }
+        err = 0;
+out:
+        return err;
+}
+
+static int perf_file_section__process(struct perf_file_section *self,
+                                      struct perf_header *ph,
+                                      int feat, int fd)
+{
+        if (lseek(fd, self->offset, SEEK_SET) == (off_t)-1) {
+                pr_debug("Failed to lseek to %Ld offset for feature %d, "
+                         "continuing...\n", self->offset, feat);
+                return 0;
+        }
+
+        switch (feat) {
+        case HEADER_TRACE_INFO:
+                trace_report(fd, false);
+                break;
+
+        case HEADER_BUILD_ID:
+                if (perf_header__read_build_ids(ph, fd, self->offset, self->size))
+                        pr_debug("Failed to read buildids, continuing...\n");
+                break;
+        default:
+                pr_debug("unknown feature %d, continuing...\n", feat);
+        }
+
+        return 0;
+}
+
+static int perf_file_header__read_pipe(struct perf_pipe_file_header *self,
+                                       struct perf_header *ph, int fd,
+                                       bool repipe)
+{
+        if (readn(fd, self, sizeof(*self)) <= 0 ||
+            memcmp(&self->magic, __perf_magic, sizeof(self->magic)))
+                return -1;
+
+        if (repipe && do_write(STDOUT_FILENO, self, sizeof(*self)) < 0)
+                return -1;
+
+        if (self->size != sizeof(*self)) {
+                u64 size = bswap_64(self->size);
+
+                if (size != sizeof(*self))
+                        return -1;
+
+                ph->needs_swap = true;
+        }
+
+        return 0;
+}
+
+static int perf_header__read_pipe(struct perf_session *session, int fd)
+{
+        struct perf_header *self = &session->header;
+        struct perf_pipe_file_header f_header;
+
+        if (perf_file_header__read_pipe(&f_header, self, fd,
+                                        session->repipe) < 0) {
+                pr_debug("incompatible file format\n");
+                return -EINVAL;
+        }
+
+        session->fd = fd;
+
+        return 0;
+}
+
+int perf_header__read(struct perf_session *session, int fd)
+{
+        struct perf_header *self = &session->header;
+        struct perf_file_header f_header;
+        struct perf_file_attr   f_attr;
+        u64                     f_id;
+        int nr_attrs, nr_ids, i, j;
+
+        if (session->fd_pipe)
+                return perf_header__read_pipe(session, fd);
+
+        if (perf_file_header__read(&f_header, self, fd) < 0) {
+                pr_debug("incompatible file format\n");
+                return -EINVAL;
+        }
+
+        nr_attrs = f_header.attrs.size / sizeof(f_attr);
+        lseek(fd, f_header.attrs.offset, SEEK_SET);
+
+        for (i = 0; i < nr_attrs; i++) {
+                struct perf_header_attr *attr;
+                off_t tmp;
+
+                if (perf_header__getbuffer64(self, fd, &f_attr, sizeof(f_attr)))
+                        goto out_errno;
+
+                tmp = lseek(fd, 0, SEEK_CUR);
+
+                attr = perf_header_attr__new(&f_attr.attr);
+                if (attr == NULL)
+                         return -ENOMEM;
+
+                nr_ids = f_attr.ids.size / sizeof(u64);
+                lseek(fd, f_attr.ids.offset, SEEK_SET);
+
+                for (j = 0; j < nr_ids; j++) {
+                        if (perf_header__getbuffer64(self, fd, &f_id, sizeof(f_id)))
+                                goto out_errno;
+
+                        if (perf_header_attr__add_id(attr, f_id) < 0) {
+                                perf_header_attr__delete(attr);
+                                return -ENOMEM;
+                        }
+                }
+                if (perf_header__add_attr(self, attr) < 0) {
+                        perf_header_attr__delete(attr);
+                        return -ENOMEM;
+                }
+
+                lseek(fd, tmp, SEEK_SET);
+        }
+
+        if (f_header.event_types.size) {
+                lseek(fd, f_header.event_types.offset, SEEK_SET);
+                events = malloc(f_header.event_types.size);
+                if (events == NULL)
+                        return -ENOMEM;
+                if (perf_header__getbuffer64(self, fd, events,
+                                             f_header.event_types.size))
+                        goto out_errno;
+                event_count =  f_header.event_types.size / sizeof(struct perf_trace_event_type);
+        }
+
+        perf_header__process_sections(self, fd, perf_file_section__process);
+
+        lseek(fd, self->data_offset, SEEK_SET);
+
+        self->frozen = 1;
+        return 0;
+out_errno:
+        return -errno;
+}
+
+u64 perf_header__sample_type(struct perf_header *header)
+{
+        u64 type = 0;
+        int i;
+
+        for (i = 0; i < header->attrs; i++) {
+                struct perf_header_attr *attr = header->attr[i];
+
+                if (!type)
+                        type = attr->attr.sample_type;
+                else if (type != attr->attr.sample_type)
+                        die("non matching sample_type");
+        }
+
+        return type;
+}
+
+bool perf_header__sample_id_all(const struct perf_header *header)
+{
+        bool value = false, first = true;
+        int i;
+
+        for (i = 0; i < header->attrs; i++) {
+                struct perf_header_attr *attr = header->attr[i];
+
+                if (first) {
+                        value = attr->attr.sample_id_all;
+                        first = false;
+                } else if (value != attr->attr.sample_id_all)
+                        die("non matching sample_id_all");
+        }
+
+        return value;
+}
+
+struct perf_event_attr *
+perf_header__find_attr(u64 id, struct perf_header *header)
+{
+        int i;
+
+        /*
+         * We set id to -1 if the data file doesn't contain sample
+         * ids. This can happen when the data file contains one type
+         * of event and in that case, the header can still store the
+         * event attribute information. Check for this and avoid
+         * walking through the entire list of ids which may be large.
+         */
+        if (id == -1ULL) {
+                if (header->attrs > 0)
+                        return &header->attr[0]->attr;
+                return NULL;
+        }
+
+        for (i = 0; i < header->attrs; i++) {
+                struct perf_header_attr *attr = header->attr[i];
+                int j;
+
+                for (j = 0; j < attr->ids; j++) {
+                        if (attr->id[j] == id)
+                                return &attr->attr;
+                }
+        }
+
+        return NULL;
+}
+
+int event__synthesize_attr(struct perf_event_attr *attr, u16 ids, u64 *id,
+                           event__handler_t process,
+                           struct perf_session *session)
+{
+        event_t *ev;
+        size_t size;
+        int err;
+
+        size = sizeof(struct perf_event_attr);
+        size = ALIGN(size, sizeof(u64));
+        size += sizeof(struct perf_event_header);
+        size += ids * sizeof(u64);
+
+        ev = malloc(size);
+
+        if (ev == NULL)
+                return -ENOMEM;
+
+        ev->attr.attr = *attr;
+        memcpy(ev->attr.id, id, ids * sizeof(u64));
+
+        ev->attr.header.type = PERF_RECORD_HEADER_ATTR;
+        ev->attr.header.size = size;
+
+        err = process(ev, NULL, session);
+
+        free(ev);
+
+        return err;
+}
+
+int event__synthesize_attrs(struct perf_header *self, event__handler_t process,
+                            struct perf_session *session)
+{
+        struct perf_header_attr *attr;
+        int i, err = 0;
+
+        for (i = 0; i < self->attrs; i++) {
+                attr = self->attr[i];
+
+                err = event__synthesize_attr(&attr->attr, attr->ids, attr->id,
+                                             process, session);
+                if (err) {
+                        pr_debug("failed to create perf header attribute\n");
+                        return err;
+                }
+        }
+
+        return err;
+}
+
+int event__process_attr(event_t *self, struct perf_session *session)
+{
+        struct perf_header_attr *attr;
+        unsigned int i, ids, n_ids;
+
+        attr = perf_header_attr__new(&self->attr.attr);
+        if (attr == NULL)
+                return -ENOMEM;
+
+        ids = self->header.size;
+        ids -= (void *)&self->attr.id - (void *)self;
+        n_ids = ids / sizeof(u64);
+
+        for (i = 0; i < n_ids; i++) {
+                if (perf_header_attr__add_id(attr, self->attr.id[i]) < 0) {
+                        perf_header_attr__delete(attr);
+                        return -ENOMEM;
+                }
+        }
+
+        if (perf_header__add_attr(&session->header, attr) < 0) {
+                perf_header_attr__delete(attr);
+                return -ENOMEM;
+        }
+
+        perf_session__update_sample_type(session);
+
+        return 0;
+}
+
+int event__synthesize_event_type(u64 event_id, char *name,
+                                 event__handler_t process,
+                                 struct perf_session *session)
+{
+        event_t ev;
+        size_t size = 0;
+        int err = 0;
+
+        memset(&ev, 0, sizeof(ev));
+
+        ev.event_type.event_type.event_id = event_id;
+        memset(ev.event_type.event_type.name, 0, MAX_EVENT_NAME);
+        strncpy(ev.event_type.event_type.name, name, MAX_EVENT_NAME - 1);
+
+        ev.event_type.header.type = PERF_RECORD_HEADER_EVENT_TYPE;
+        size = strlen(name);
+        size = ALIGN(size, sizeof(u64));
+        ev.event_type.header.size = sizeof(ev.event_type) -
+                (sizeof(ev.event_type.event_type.name) - size);
+
+        err = process(&ev, NULL, session);
+
+        return err;
+}
+
+int event__synthesize_event_types(event__handler_t process,
+                                  struct perf_session *session)
+{
+        struct perf_trace_event_type *type;
+        int i, err = 0;
+
+        for (i = 0; i < event_count; i++) {
+                type = &events[i];
+
+                err = event__synthesize_event_type(type->event_id, type->name,
+                                                   process, session);
+                if (err) {
+                        pr_debug("failed to create perf header event type\n");
+                        return err;
+                }
+        }
+
+        return err;
+}
+
+int event__process_event_type(event_t *self,
+                              struct perf_session *session __unused)
+{
+        if (perf_header__push_event(self->event_type.event_type.event_id,
+                                    self->event_type.event_type.name) < 0)
+                return -ENOMEM;
+
+        return 0;
+}
+
+int event__synthesize_tracing_data(int fd, struct list_head *pattrs,
+                                   event__handler_t process,
+                                   struct perf_session *session __unused)
+{
+        event_t ev;
+        ssize_t size = 0, aligned_size = 0, padding;
+        int err = 0;
+
+        memset(&ev, 0, sizeof(ev));
+
+        ev.tracing_data.header.type = PERF_RECORD_HEADER_TRACING_DATA;
+        size = read_tracing_data_size(fd, pattrs);
+        if (size <= 0)
+                return size;
+        aligned_size = ALIGN(size, sizeof(u64));
+        padding = aligned_size - size;
+        ev.tracing_data.header.size = sizeof(ev.tracing_data);
+        ev.tracing_data.size = aligned_size;
+
+        process(&ev, NULL, session);
+
+        err = read_tracing_data(fd, pattrs);
+        write_padded(fd, NULL, 0, padding);
+
+        return aligned_size;
+}
+
+int event__process_tracing_data(event_t *self,
+                                struct perf_session *session)
+{
+        ssize_t size_read, padding, size = self->tracing_data.size;
+        off_t offset = lseek(session->fd, 0, SEEK_CUR);
+        char buf[BUFSIZ];
+
+        /* setup for reading amidst mmap */
+        lseek(session->fd, offset + sizeof(struct tracing_data_event),
+              SEEK_SET);
+
+        size_read = trace_report(session->fd, session->repipe);
+
+        padding = ALIGN(size_read, sizeof(u64)) - size_read;
+
+        if (read(session->fd, buf, padding) < 0)
+                die("reading input file");
+        if (session->repipe) {
+                int retw = write(STDOUT_FILENO, buf, padding);
+                if (retw <= 0 || retw != padding)
+                        die("repiping tracing data padding");
+        }
+
+        if (size_read + padding != size)
+                die("tracing data size mismatch");
+
+        return size_read + padding;
+}
+
+int event__synthesize_build_id(struct dso *pos, u16 misc,
+                               event__handler_t process,
+                               struct machine *machine,
+                               struct perf_session *session)
+{
+        event_t ev;
+        size_t len;
+        int err = 0;
+
+        if (!pos->hit)
+                return err;
+
+        memset(&ev, 0, sizeof(ev));
+
+        len = pos->long_name_len + 1;
+        len = ALIGN(len, NAME_ALIGN);
+        memcpy(&ev.build_id.build_id, pos->build_id, sizeof(pos->build_id));
+        ev.build_id.header.type = PERF_RECORD_HEADER_BUILD_ID;
+        ev.build_id.header.misc = misc;
+        ev.build_id.pid = machine->pid;
+        ev.build_id.header.size = sizeof(ev.build_id) + len;
+        memcpy(&ev.build_id.filename, pos->long_name, pos->long_name_len);
+
+        err = process(&ev, NULL, session);
+
+        return err;
+}
+
+int event__process_build_id(event_t *self,
+                            struct perf_session *session)
+{
+        __event_process_build_id(&self->build_id,
+                                 self->build_id.filename,
+                                 session);
+        return 0;
+}
+
+void disable_buildid_cache(void)
+{
+        no_buildid_cache = true;
+}
diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h
new file mode 100644
index 00000000000..33f16be7b72
--- /dev/null
+++ b/tools/perf/util/header.h
@@ -0,0 +1,128 @@
+#ifndef __PERF_HEADER_H
+#define __PERF_HEADER_H
+
+#include "../../../include/linux/perf_event.h"
+#include <sys/types.h>
+#include <stdbool.h>
+#include "types.h"
+#include "event.h"
+
+#include <linux/bitmap.h>
+
+struct perf_header_attr {
+        struct perf_event_attr attr;
+        int ids, size;
+        u64 *id;
+        off_t id_offset;
+};
+
+enum {
+        HEADER_TRACE_INFO = 1,
+        HEADER_BUILD_ID,
+        HEADER_LAST_FEATURE,
+};
+
+#define HEADER_FEAT_BITS                        256
+
+struct perf_file_section {
+        u64 offset;
+        u64 size;
+};
+
+struct perf_file_header {
+        u64                             magic;
+        u64                             size;
+        u64                             attr_size;
+        struct perf_file_section        attrs;
+        struct perf_file_section        data;
+        struct perf_file_section        event_types;
+        DECLARE_BITMAP(adds_features, HEADER_FEAT_BITS);
+};
+
+struct perf_pipe_file_header {
+        u64                             magic;
+        u64                             size;
+};
+
+struct perf_header;
+
+int perf_file_header__read(struct perf_file_header *self,
+                           struct perf_header *ph, int fd);
+
+struct perf_header {
+        int                     frozen;
+        int                     attrs, size;
+        bool                    needs_swap;
+        struct perf_header_attr **attr;
+        s64                     attr_offset;
+        u64                     data_offset;
+        u64                     data_size;
+        u64                     event_offset;
+        u64                     event_size;
+        DECLARE_BITMAP(adds_features, HEADER_FEAT_BITS);
+};
+
+int perf_header__init(struct perf_header *self);
+void perf_header__exit(struct perf_header *self);
+
+int perf_header__read(struct perf_session *session, int fd);
+int perf_header__write(struct perf_header *self, int fd, bool at_exit);
+int perf_header__write_pipe(int fd);
+
+int perf_header__add_attr(struct perf_header *self,
+                          struct perf_header_attr *attr);
+
+int perf_header__push_event(u64 id, const char *name);
+char *perf_header__find_event(u64 id);
+
+struct perf_header_attr *perf_header_attr__new(struct perf_event_attr *attr);
+void perf_header_attr__delete(struct perf_header_attr *self);
+
+int perf_header_attr__add_id(struct perf_header_attr *self, u64 id);
+
+u64 perf_header__sample_type(struct perf_header *header);
+bool perf_header__sample_id_all(const struct perf_header *header);
+struct perf_event_attr *
+perf_header__find_attr(u64 id, struct perf_header *header);
+void perf_header__set_feat(struct perf_header *self, int feat);
+void perf_header__clear_feat(struct perf_header *self, int feat);
+bool perf_header__has_feat(const struct perf_header *self, int feat);
+
+int perf_header__process_sections(struct perf_header *self, int fd,
+                                  int (*process)(struct perf_file_section *self,
+                                                 struct perf_header *ph,
+                                                 int feat, int fd));
+
+int build_id_cache__add_s(const char *sbuild_id, const char *debugdir,
+                          const char *name, bool is_kallsyms);
+int build_id_cache__remove_s(const char *sbuild_id, const char *debugdir);
+
+int event__synthesize_attr(struct perf_event_attr *attr, u16 ids, u64 *id,
+                           event__handler_t process,
+                           struct perf_session *session);
+int event__synthesize_attrs(struct perf_header *self,
+                            event__handler_t process,
+                            struct perf_session *session);
+int event__process_attr(event_t *self, struct perf_session *session);
+
+int event__synthesize_event_type(u64 event_id, char *name,
+                                 event__handler_t process,
+                                 struct perf_session *session);
+int event__synthesize_event_types(event__handler_t process,
+                                  struct perf_session *session);
+int event__process_event_type(event_t *self,
+                              struct perf_session *session);
+
+int event__synthesize_tracing_data(int fd, struct list_head *pattrs,
+                                   event__handler_t process,
+                                   struct perf_session *session);
+int event__process_tracing_data(event_t *self,
+                                struct perf_session *session);
+
+int event__synthesize_build_id(struct dso *pos, u16 misc,
+                               event__handler_t process,
+                               struct machine *machine,
+                               struct perf_session *session);
+int event__process_build_id(event_t *self, struct perf_session *session);
+
+#endif /* __PERF_HEADER_H */
diff --git a/tools/perf/util/help.c b/tools/perf/util/help.c
index 6653f7dd1d7..6f2975a0035 100644
--- a/tools/perf/util/help.c
+++ b/tools/perf/util/help.c
@@ -4,29 +4,7 @@
 #include "levenshtein.h"
 #include "help.h"
 
-/* most GUI terminals set COLUMNS (although some don't export it) */
-static int term_columns(void)
-{
-        char *col_string = getenv("COLUMNS");
-        int n_cols;
-
-        if (col_string && (n_cols = atoi(col_string)) > 0)
-                return n_cols;
-
-#ifdef TIOCGWINSZ
-        {
-                struct winsize ws;
-                if (!ioctl(1, TIOCGWINSZ, &ws)) {
-                        if (ws.ws_col)
-                                return ws.ws_col;
-                }
-        }
-#endif
-
-        return 80;
-}
-
-void add_cmdname(struct cmdnames *cmds, const char *name, int len)
+void add_cmdname(struct cmdnames *cmds, const char *name, size_t len)
 {
         struct cmdname *ent = malloc(sizeof(*ent) + len + 1);
 
@@ -40,7 +18,8 @@ void add_cmdname(struct cmdnames *cmds, const char *name, int len)
 
 static void clean_cmdnames(struct cmdnames *cmds)
 {
-        int i;
+        unsigned int i;
+
         for (i = 0; i < cmds->cnt; ++i)
                 free(cmds->names[i]);
         free(cmds->names);
@@ -57,7 +36,7 @@ static int cmdname_compare(const void *a_, const void *b_)
 
 static void uniq(struct cmdnames *cmds)
 {
-        int i, j;
+        unsigned int i, j;
 
         if (!cmds->cnt)
                 return;
@@ -71,7 +50,7 @@ static void uniq(struct cmdnames *cmds)
 
 void exclude_cmds(struct cmdnames *cmds, struct cmdnames *excludes)
 {
-        int ci, cj, ei;
+        size_t ci, cj, ei;
         int cmp;
 
         ci = cj = ei = 0;
@@ -95,9 +74,13 @@ static void pretty_print_string_list(struct cmdnames *cmds, int longest)
 {
         int cols = 1, rows;
         int space = longest + 1; /* min 1 SP between words */
-        int max_cols = term_columns() - 1; /* don't print *on* the edge */
+        struct winsize win;
+        int max_cols;
         int i, j;
 
+        get_term_dimensions(&win);
+        max_cols = win.ws_col - 1; /* don't print *on* the edge */
+
         if (space < max_cols)
                 cols = max_cols / space;
         rows = (cmds->cnt + cols - 1) / cols;
@@ -106,8 +89,9 @@ static void pretty_print_string_list(struct cmdnames *cmds, int longest)
                 printf("  ");
 
                 for (j = 0; j < cols; j++) {
-                        int n = j * rows + i;
-                        int size = space;
+                        unsigned int n = j * rows + i;
+                        unsigned int size = space;
+
                         if (n >= cmds->cnt)
                                 break;
                         if (j == cols-1 || n + rows >= cmds->cnt)
@@ -126,21 +110,6 @@ static int is_executable(const char *name)
             !S_ISREG(st.st_mode))
                 return 0;
 
-#ifdef __MINGW32__
-        /* cannot trust the executable bit, peek into the file instead */
-        char buf[3] = { 0 };
-        int n;
-        int fd = open(name, O_RDONLY);
-        st.st_mode &= ~S_IXUSR;
-        if (fd >= 0) {
-                n = read(fd, buf, 2);
-                if (n == 2)
-                        /* DOS executables start with "MZ" */
-                        if (!strcmp(buf, "#!") || !strcmp(buf, "MZ"))
-                                st.st_mode |= S_IXUSR;
-                close(fd);
-        }
-#endif
         return st.st_mode & S_IXUSR;
 }
 
@@ -223,7 +192,7 @@ void load_command_list(const char *prefix,
 void list_commands(const char *title, struct cmdnames *main_cmds,
                    struct cmdnames *other_cmds)
 {
-        int i, longest = 0;
+        unsigned int i, longest = 0;
 
         for (i = 0; i < main_cmds->cnt; i++)
                 if (longest < main_cmds->names[i]->len)
@@ -254,7 +223,8 @@ void list_commands(const char *title, struct cmdnames *main_cmds,
 
 int is_in_cmdlist(struct cmdnames *c, const char *s)
 {
-        int i;
+        unsigned int i;
+
         for (i = 0; i < c->cnt; i++)
                 if (!strcmp(s, c->names[i]->name))
                         return 1;
@@ -286,7 +256,8 @@ static int levenshtein_compare(const void *p1, const void *p2)
 
 static void add_cmd_list(struct cmdnames *cmds, struct cmdnames *old)
 {
-        int i;
+        unsigned int i;
+
         ALLOC_GROW(cmds->names, cmds->cnt + old->cnt, cmds->alloc);
 
         for (i = 0; i < old->cnt; i++)
@@ -298,7 +269,7 @@ static void add_cmd_list(struct cmdnames *cmds, struct cmdnames *old)
 
 const char *help_unknown_cmd(const char *cmd)
 {
-        int i, n = 0, best_similarity = 0;
+        unsigned int i, n = 0, best_similarity = 0;
         struct cmdnames main_cmds, other_cmds;
 
         memset(&main_cmds, 0, sizeof(main_cmds));
@@ -335,7 +306,7 @@ const char *help_unknown_cmd(const char *cmd)
 
                 main_cmds.names[0] = NULL;
                 clean_cmdnames(&main_cmds);
-                fprintf(stderr, "WARNING: You called a Git program named '%s', "
+                fprintf(stderr, "WARNING: You called a perf program named '%s', "
                         "which does not exist.\n"
                         "Continuing under the assumption that you meant '%s'\n",
                         cmd, assumed);
@@ -360,7 +331,7 @@ const char *help_unknown_cmd(const char *cmd)
         exit(1);
 }
 
-int cmd_version(int argc, const char **argv, const char *prefix)
+int cmd_version(int argc __used, const char **argv __used, const char *prefix __used)
 {
         printf("perf version %s\n", perf_version_string);
         return 0;
diff --git a/tools/perf/util/help.h b/tools/perf/util/help.h
index 56bc15406ff..7f5c6dedd71 100644
--- a/tools/perf/util/help.h
+++ b/tools/perf/util/help.h
@@ -1,9 +1,9 @@
-#ifndef HELP_H
-#define HELP_H
+#ifndef __PERF_HELP_H
+#define __PERF_HELP_H
 
 struct cmdnames {
-        int alloc;
-        int cnt;
+        size_t alloc;
+        size_t cnt;
         struct cmdname {
                 size_t len; /* also used for similarity index in help.c */
                 char name[FLEX_ARRAY];
@@ -19,11 +19,11 @@ static inline void mput_char(char c, unsigned int num)
 void load_command_list(const char *prefix,
                 struct cmdnames *main_cmds,
                 struct cmdnames *other_cmds);
-void add_cmdname(struct cmdnames *cmds, const char *name, int len);
+void add_cmdname(struct cmdnames *cmds, const char *name, size_t len);
 /* Here we require that excludes is a sorted list. */
 void exclude_cmds(struct cmdnames *cmds, struct cmdnames *excludes);
 int is_in_cmdlist(struct cmdnames *c, const char *s);
 void list_commands(const char *title, struct cmdnames *main_cmds,
                    struct cmdnames *other_cmds);
 
-#endif /* HELP_H */
+#endif /* __PERF_HELP_H */
diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
new file mode 100644
index 00000000000..c749ba6136a
--- /dev/null
+++ b/tools/perf/util/hist.c
@@ -0,0 +1,1189 @@
+#include "util.h"
+#include "build-id.h"
+#include "hist.h"
+#include "session.h"
+#include "sort.h"
+#include <math.h>
+
+enum hist_filter {
+        HIST_FILTER__DSO,
+        HIST_FILTER__THREAD,
+        HIST_FILTER__PARENT,
+};
+
+struct callchain_param  callchain_param = {
+        .mode   = CHAIN_GRAPH_REL,
+        .min_percent = 0.5
+};
+
+u16 hists__col_len(struct hists *self, enum hist_column col)
+{
+        return self->col_len[col];
+}
+
+void hists__set_col_len(struct hists *self, enum hist_column col, u16 len)
+{
+        self->col_len[col] = len;
+}
+
+bool hists__new_col_len(struct hists *self, enum hist_column col, u16 len)
+{
+        if (len > hists__col_len(self, col)) {
+                hists__set_col_len(self, col, len);
+                return true;
+        }
+        return false;
+}
+
+static void hists__reset_col_len(struct hists *self)
+{
+        enum hist_column col;
+
+        for (col = 0; col < HISTC_NR_COLS; ++col)
+                hists__set_col_len(self, col, 0);
+}
+
+static void hists__calc_col_len(struct hists *self, struct hist_entry *h)
+{
+        u16 len;
+
+        if (h->ms.sym)
+                hists__new_col_len(self, HISTC_SYMBOL, h->ms.sym->namelen);
+
+        len = thread__comm_len(h->thread);
+        if (hists__new_col_len(self, HISTC_COMM, len))
+                hists__set_col_len(self, HISTC_THREAD, len + 6);
+
+        if (h->ms.map) {
+                len = dso__name_len(h->ms.map->dso);
+                hists__new_col_len(self, HISTC_DSO, len);
+        }
+}
+
+static void hist_entry__add_cpumode_period(struct hist_entry *self,
+                                           unsigned int cpumode, u64 period)
+{
+        switch (cpumode) {
+        case PERF_RECORD_MISC_KERNEL:
+                self->period_sys += period;
+                break;
+        case PERF_RECORD_MISC_USER:
+                self->period_us += period;
+                break;
+        case PERF_RECORD_MISC_GUEST_KERNEL:
+                self->period_guest_sys += period;
+                break;
+        case PERF_RECORD_MISC_GUEST_USER:
+                self->period_guest_us += period;
+                break;
+        default:
+                break;
+        }
+}
+
+/*
+ * histogram, sorted on item, collects periods
+ */
+
+static struct hist_entry *hist_entry__new(struct hist_entry *template)
+{
+        size_t callchain_size = symbol_conf.use_callchain ? sizeof(struct callchain_root) : 0;
+        struct hist_entry *self = malloc(sizeof(*self) + callchain_size);
+
+        if (self != NULL) {
+                *self = *template;
+                self->nr_events = 1;
+                if (self->ms.map)
+                        self->ms.map->referenced = true;
+                if (symbol_conf.use_callchain)
+                        callchain_init(self->callchain);
+        }
+
+        return self;
+}
+
+static void hists__inc_nr_entries(struct hists *self, struct hist_entry *h)
+{
+        if (!h->filtered) {
+                hists__calc_col_len(self, h);
+                ++self->nr_entries;
+        }
+}
+
+static u8 symbol__parent_filter(const struct symbol *parent)
+{
+        if (symbol_conf.exclude_other && parent == NULL)
+                return 1 << HIST_FILTER__PARENT;
+        return 0;
+}
+
+struct hist_entry *__hists__add_entry(struct hists *self,
+                                      struct addr_location *al,
+                                      struct symbol *sym_parent, u64 period)
+{
+        struct rb_node **p = &self->entries.rb_node;
+        struct rb_node *parent = NULL;
+        struct hist_entry *he;
+        struct hist_entry entry = {
+                .thread = al->thread,
+                .ms = {
+                        .map    = al->map,
+                        .sym    = al->sym,
+                },
+                .cpu    = al->cpu,
+                .ip     = al->addr,
+                .level  = al->level,
+                .period = period,
+                .parent = sym_parent,
+                .filtered = symbol__parent_filter(sym_parent),
+        };
+        int cmp;
+
+        while (*p != NULL) {
+                parent = *p;
+                he = rb_entry(parent, struct hist_entry, rb_node);
+
+                cmp = hist_entry__cmp(&entry, he);
+
+                if (!cmp) {
+                        he->period += period;
+                        ++he->nr_events;
+                        goto out;
+                }
+
+                if (cmp < 0)
+                        p = &(*p)->rb_left;
+                else
+                        p = &(*p)->rb_right;
+        }
+
+        he = hist_entry__new(&entry);
+        if (!he)
+                return NULL;
+        rb_link_node(&he->rb_node, parent, p);
+        rb_insert_color(&he->rb_node, &self->entries);
+        hists__inc_nr_entries(self, he);
+out:
+        hist_entry__add_cpumode_period(he, al->cpumode, period);
+        return he;
+}
+
+int64_t
+hist_entry__cmp(struct hist_entry *left, struct hist_entry *right)
+{
+        struct sort_entry *se;
+        int64_t cmp = 0;
+
+        list_for_each_entry(se, &hist_entry__sort_list, list) {
+                cmp = se->se_cmp(left, right);
+                if (cmp)
+                        break;
+        }
+
+        return cmp;
+}
+
+int64_t
+hist_entry__collapse(struct hist_entry *left, struct hist_entry *right)
+{
+        struct sort_entry *se;
+        int64_t cmp = 0;
+
+        list_for_each_entry(se, &hist_entry__sort_list, list) {
+                int64_t (*f)(struct hist_entry *, struct hist_entry *);
+
+                f = se->se_collapse ?: se->se_cmp;
+
+                cmp = f(left, right);
+                if (cmp)
+                        break;
+        }
+
+        return cmp;
+}
+
+void hist_entry__free(struct hist_entry *he)
+{
+        free(he);
+}
+
+/*
+ * collapse the histogram
+ */
+
+static bool collapse__insert_entry(struct rb_root *root, struct hist_entry *he)
+{
+        struct rb_node **p = &root->rb_node;
+        struct rb_node *parent = NULL;
+        struct hist_entry *iter;
+        int64_t cmp;
+
+        while (*p != NULL) {
+                parent = *p;
+                iter = rb_entry(parent, struct hist_entry, rb_node);
+
+                cmp = hist_entry__collapse(iter, he);
+
+                if (!cmp) {
+                        iter->period += he->period;
+                        if (symbol_conf.use_callchain)
+                                callchain_merge(iter->callchain, he->callchain);
+                        hist_entry__free(he);
+                        return false;
+                }
+
+                if (cmp < 0)
+                        p = &(*p)->rb_left;
+                else
+                        p = &(*p)->rb_right;
+        }
+
+        rb_link_node(&he->rb_node, parent, p);
+        rb_insert_color(&he->rb_node, root);
+        return true;
+}
+
+void hists__collapse_resort(struct hists *self)
+{
+        struct rb_root tmp;
+        struct rb_node *next;
+        struct hist_entry *n;
+
+        if (!sort__need_collapse)
+                return;
+
+        tmp = RB_ROOT;
+        next = rb_first(&self->entries);
+        self->nr_entries = 0;
+        hists__reset_col_len(self);
+
+        while (next) {
+                n = rb_entry(next, struct hist_entry, rb_node);
+                next = rb_next(&n->rb_node);
+
+                rb_erase(&n->rb_node, &self->entries);
+                if (collapse__insert_entry(&tmp, n))
+                        hists__inc_nr_entries(self, n);
+        }
+
+        self->entries = tmp;
+}
+
+/*
+ * reverse the map, sort on period.
+ */
+
+static void __hists__insert_output_entry(struct rb_root *entries,
+                                         struct hist_entry *he,
+                                         u64 min_callchain_hits)
+{
+        struct rb_node **p = &entries->rb_node;
+        struct rb_node *parent = NULL;
+        struct hist_entry *iter;
+
+        if (symbol_conf.use_callchain)
+                callchain_param.sort(&he->sorted_chain, he->callchain,
+                                      min_callchain_hits, &callchain_param);
+
+        while (*p != NULL) {
+                parent = *p;
+                iter = rb_entry(parent, struct hist_entry, rb_node);
+
+                if (he->period > iter->period)
+                        p = &(*p)->rb_left;
+                else
+                        p = &(*p)->rb_right;
+        }
+
+        rb_link_node(&he->rb_node, parent, p);
+        rb_insert_color(&he->rb_node, entries);
+}
+
+void hists__output_resort(struct hists *self)
+{
+        struct rb_root tmp;
+        struct rb_node *next;
+        struct hist_entry *n;
+        u64 min_callchain_hits;
+
+        min_callchain_hits = self->stats.total_period * (callchain_param.min_percent / 100);
+
+        tmp = RB_ROOT;
+        next = rb_first(&self->entries);
+
+        self->nr_entries = 0;
+        hists__reset_col_len(self);
+
+        while (next) {
+                n = rb_entry(next, struct hist_entry, rb_node);
+                next = rb_next(&n->rb_node);
+
+                rb_erase(&n->rb_node, &self->entries);
+                __hists__insert_output_entry(&tmp, n, min_callchain_hits);
+                hists__inc_nr_entries(self, n);
+        }
+
+        self->entries = tmp;
+}
+
+static size_t callchain__fprintf_left_margin(FILE *fp, int left_margin)
+{
+        int i;
+        int ret = fprintf(fp, "            ");
+
+        for (i = 0; i < left_margin; i++)
+                ret += fprintf(fp, " ");
+
+        return ret;
+}
+
+static size_t ipchain__fprintf_graph_line(FILE *fp, int depth, int depth_mask,
+                                          int left_margin)
+{
+        int i;
+        size_t ret = callchain__fprintf_left_margin(fp, left_margin);
+
+        for (i = 0; i < depth; i++)
+                if (depth_mask & (1 << i))
+                        ret += fprintf(fp, "|          ");
+                else
+                        ret += fprintf(fp, "           ");
+
+        ret += fprintf(fp, "\n");
+
+        return ret;
+}
+
+static size_t ipchain__fprintf_graph(FILE *fp, struct callchain_list *chain,
+                                     int depth, int depth_mask, int period,
+                                     u64 total_samples, u64 hits,
+                                     int left_margin)
+{
+        int i;
+        size_t ret = 0;
+
+        ret += callchain__fprintf_left_margin(fp, left_margin);
+        for (i = 0; i < depth; i++) {
+                if (depth_mask & (1 << i))
+                        ret += fprintf(fp, "|");
+                else
+                        ret += fprintf(fp, " ");
+                if (!period && i == depth - 1) {
+                        double percent;
+
+                        percent = hits * 100.0 / total_samples;
+                        ret += percent_color_fprintf(fp, "--%2.2f%%-- ", percent);
+                } else
+                        ret += fprintf(fp, "%s", "          ");
+        }
+        if (chain->ms.sym)
+                ret += fprintf(fp, "%s\n", chain->ms.sym->name);
+        else
+                ret += fprintf(fp, "%p\n", (void *)(long)chain->ip);
+
+        return ret;
+}
+
+static struct symbol *rem_sq_bracket;
+static struct callchain_list rem_hits;
+
+static void init_rem_hits(void)
+{
+        rem_sq_bracket = malloc(sizeof(*rem_sq_bracket) + 6);
+        if (!rem_sq_bracket) {
+                fprintf(stderr, "Not enough memory to display remaining hits\n");
+                return;
+        }
+
+        strcpy(rem_sq_bracket->name, "[...]");
+        rem_hits.ms.sym = rem_sq_bracket;
+}
+
+static size_t __callchain__fprintf_graph(FILE *fp, struct callchain_node *self,
+                                         u64 total_samples, int depth,
+                                         int depth_mask, int left_margin)
+{
+        struct rb_node *node, *next;
+        struct callchain_node *child;
+        struct callchain_list *chain;
+        int new_depth_mask = depth_mask;
+        u64 new_total;
+        u64 remaining;
+        size_t ret = 0;
+        int i;
+        uint entries_printed = 0;
+
+        if (callchain_param.mode == CHAIN_GRAPH_REL)
+                new_total = self->children_hit;
+        else
+                new_total = total_samples;
+
+        remaining = new_total;
+
+        node = rb_first(&self->rb_root);
+        while (node) {
+                u64 cumul;
+
+                child = rb_entry(node, struct callchain_node, rb_node);
+                cumul = cumul_hits(child);
+                remaining -= cumul;
+
+                /*
+                 * The depth mask manages the output of pipes that show
+                 * the depth. We don't want to keep the pipes of the current
+                 * level for the last child of this depth.
+                 * Except if we have remaining filtered hits. They will
+                 * supersede the last child
+                 */
+                next = rb_next(node);
+                if (!next && (callchain_param.mode != CHAIN_GRAPH_REL || !remaining))
+                        new_depth_mask &= ~(1 << (depth - 1));
+
+                /*
+                 * But we keep the older depth mask for the line separator
+                 * to keep the level link until we reach the last child
+                 */
+                ret += ipchain__fprintf_graph_line(fp, depth, depth_mask,
+                                                   left_margin);
+                i = 0;
+                list_for_each_entry(chain, &child->val, list) {
+                        ret += ipchain__fprintf_graph(fp, chain, depth,
+                                                      new_depth_mask, i++,
+                                                      new_total,
+                                                      cumul,
+                                                      left_margin);
+                }
+                ret += __callchain__fprintf_graph(fp, child, new_total,
+                                                  depth + 1,
+                                                  new_depth_mask | (1 << depth),
+                                                  left_margin);
+                node = next;
+                if (++entries_printed == callchain_param.print_limit)
+                        break;
+        }
+
+        if (callchain_param.mode == CHAIN_GRAPH_REL &&
+                remaining && remaining != new_total) {
+
+                if (!rem_sq_bracket)
+                        return ret;
+
+                new_depth_mask &= ~(1 << (depth - 1));
+
+                ret += ipchain__fprintf_graph(fp, &rem_hits, depth,
+                                              new_depth_mask, 0, new_total,
+                                              remaining, left_margin);
+        }
+
+        return ret;
+}
+
+static size_t callchain__fprintf_graph(FILE *fp, struct callchain_node *self,
+                                       u64 total_samples, int left_margin)
+{
+        struct callchain_list *chain;
+        bool printed = false;
+        int i = 0;
+        int ret = 0;
+        u32 entries_printed = 0;
+
+        list_for_each_entry(chain, &self->val, list) {
+                if (!i++ && sort__first_dimension == SORT_SYM)
+                        continue;
+
+                if (!printed) {
+                        ret += callchain__fprintf_left_margin(fp, left_margin);
+                        ret += fprintf(fp, "|\n");
+                        ret += callchain__fprintf_left_margin(fp, left_margin);
+                        ret += fprintf(fp, "---");
+
+                        left_margin += 3;
+                        printed = true;
+                } else
+                        ret += callchain__fprintf_left_margin(fp, left_margin);
+
+                if (chain->ms.sym)
+                        ret += fprintf(fp, " %s\n", chain->ms.sym->name);
+                else
+                        ret += fprintf(fp, " %p\n", (void *)(long)chain->ip);
+
+                if (++entries_printed == callchain_param.print_limit)
+                        break;
+        }
+
+        ret += __callchain__fprintf_graph(fp, self, total_samples, 1, 1, left_margin);
+
+        return ret;
+}
+
+static size_t callchain__fprintf_flat(FILE *fp, struct callchain_node *self,
+                                      u64 total_samples)
+{
+        struct callchain_list *chain;
+        size_t ret = 0;
+
+        if (!self)
+                return 0;
+
+        ret += callchain__fprintf_flat(fp, self->parent, total_samples);
+
+
+        list_for_each_entry(chain, &self->val, list) {
+                if (chain->ip >= PERF_CONTEXT_MAX)
+                        continue;
+                if (chain->ms.sym)
+                        ret += fprintf(fp, "                %s\n", chain->ms.sym->name);
+                else
+                        ret += fprintf(fp, "                %p\n",
+                                        (void *)(long)chain->ip);
+        }
+
+        return ret;
+}
+
+static size_t hist_entry_callchain__fprintf(FILE *fp, struct hist_entry *self,
+                                            u64 total_samples, int left_margin)
+{
+        struct rb_node *rb_node;
+        struct callchain_node *chain;
+        size_t ret = 0;
+        u32 entries_printed = 0;
+
+        rb_node = rb_first(&self->sorted_chain);
+        while (rb_node) {
+                double percent;
+
+                chain = rb_entry(rb_node, struct callchain_node, rb_node);
+                percent = chain->hit * 100.0 / total_samples;
+                switch (callchain_param.mode) {
+                case CHAIN_FLAT:
+                        ret += percent_color_fprintf(fp, "           %6.2f%%\n",
+                                                     percent);
+                        ret += callchain__fprintf_flat(fp, chain, total_samples);
+                        break;
+                case CHAIN_GRAPH_ABS: /* Falldown */
+                case CHAIN_GRAPH_REL:
+                        ret += callchain__fprintf_graph(fp, chain, total_samples,
+                                                        left_margin);
+                case CHAIN_NONE:
+                default:
+                        break;
+                }
+                ret += fprintf(fp, "\n");
+                if (++entries_printed == callchain_param.print_limit)
+                        break;
+                rb_node = rb_next(rb_node);
+        }
+
+        return ret;
+}
+
+int hist_entry__snprintf(struct hist_entry *self, char *s, size_t size,
+                         struct hists *hists, struct hists *pair_hists,
+                         bool show_displacement, long displacement,
+                         bool color, u64 session_total)
+{
+        struct sort_entry *se;
+        u64 period, total, period_sys, period_us, period_guest_sys, period_guest_us;
+        const char *sep = symbol_conf.field_sep;
+        int ret;
+
+        if (symbol_conf.exclude_other && !self->parent)
+                return 0;
+
+        if (pair_hists) {
+                period = self->pair ? self->pair->period : 0;
+                total = pair_hists->stats.total_period;
+                period_sys = self->pair ? self->pair->period_sys : 0;
+                period_us = self->pair ? self->pair->period_us : 0;
+                period_guest_sys = self->pair ? self->pair->period_guest_sys : 0;
+                period_guest_us = self->pair ? self->pair->period_guest_us : 0;
+        } else {
+                period = self->period;
+                total = session_total;
+                period_sys = self->period_sys;
+                period_us = self->period_us;
+                period_guest_sys = self->period_guest_sys;
+                period_guest_us = self->period_guest_us;
+        }
+
+        if (total) {
+                if (color)
+                        ret = percent_color_snprintf(s, size,
+                                                     sep ? "%.2f" : "   %6.2f%%",
+                                                     (period * 100.0) / total);
+                else
+                        ret = snprintf(s, size, sep ? "%.2f" : "   %6.2f%%",
+                                       (period * 100.0) / total);
+                if (symbol_conf.show_cpu_utilization) {
+                        ret += percent_color_snprintf(s + ret, size - ret,
+                                        sep ? "%.2f" : "   %6.2f%%",
+                                        (period_sys * 100.0) / total);
+                        ret += percent_color_snprintf(s + ret, size - ret,
+                                        sep ? "%.2f" : "   %6.2f%%",
+                                        (period_us * 100.0) / total);
+                        if (perf_guest) {
+                                ret += percent_color_snprintf(s + ret,
+                                                size - ret,
+                                                sep ? "%.2f" : "   %6.2f%%",
+                                                (period_guest_sys * 100.0) /
+                                                                total);
+                                ret += percent_color_snprintf(s + ret,
+                                                size - ret,
+                                                sep ? "%.2f" : "   %6.2f%%",
+                                                (period_guest_us * 100.0) /
+                                                                total);
+                        }
+                }
+        } else
+                ret = snprintf(s, size, sep ? "%lld" : "%12lld ", period);
+
+        if (symbol_conf.show_nr_samples) {
+                if (sep)
+                        ret += snprintf(s + ret, size - ret, "%c%lld", *sep, period);
+                else
+                        ret += snprintf(s + ret, size - ret, "%11lld", period);
+        }
+
+        if (pair_hists) {
+                char bf[32];
+                double old_percent = 0, new_percent = 0, diff;
+
+                if (total > 0)
+                        old_percent = (period * 100.0) / total;
+                if (session_total > 0)
+                        new_percent = (self->period * 100.0) / session_total;
+
+                diff = new_percent - old_percent;
+
+                if (fabs(diff) >= 0.01)
+                        snprintf(bf, sizeof(bf), "%+4.2F%%", diff);
+                else
+                        snprintf(bf, sizeof(bf), " ");
+
+                if (sep)
+                        ret += snprintf(s + ret, size - ret, "%c%s", *sep, bf);
+                else
+                        ret += snprintf(s + ret, size - ret, "%11.11s", bf);
+
+                if (show_displacement) {
+                        if (displacement)
+                                snprintf(bf, sizeof(bf), "%+4ld", displacement);
+                        else
+                                snprintf(bf, sizeof(bf), " ");
+
+                        if (sep)
+                                ret += snprintf(s + ret, size - ret, "%c%s", *sep, bf);
+                        else
+                                ret += snprintf(s + ret, size - ret, "%6.6s", bf);
+                }
+        }
+
+        list_for_each_entry(se, &hist_entry__sort_list, list) {
+                if (se->elide)
+                        continue;
+
+                ret += snprintf(s + ret, size - ret, "%s", sep ?: "  ");
+                ret += se->se_snprintf(self, s + ret, size - ret,
+                                       hists__col_len(hists, se->se_width_idx));
+        }
+
+        return ret;
+}
+
+int hist_entry__fprintf(struct hist_entry *self, struct hists *hists,
+                        struct hists *pair_hists, bool show_displacement,
+                        long displacement, FILE *fp, u64 session_total)
+{
+        char bf[512];
+        hist_entry__snprintf(self, bf, sizeof(bf), hists, pair_hists,
+                             show_displacement, displacement,
+                             true, session_total);
+        return fprintf(fp, "%s\n", bf);
+}
+
+static size_t hist_entry__fprintf_callchain(struct hist_entry *self,
+                                            struct hists *hists, FILE *fp,
+                                            u64 session_total)
+{
+        int left_margin = 0;
+
+        if (sort__first_dimension == SORT_COMM) {
+                struct sort_entry *se = list_first_entry(&hist_entry__sort_list,
+                                                         typeof(*se), list);
+                left_margin = hists__col_len(hists, se->se_width_idx);
+                left_margin -= thread__comm_len(self->thread);
+        }
+
+        return hist_entry_callchain__fprintf(fp, self, session_total,
+                                             left_margin);
+}
+
+size_t hists__fprintf(struct hists *self, struct hists *pair,
+                      bool show_displacement, FILE *fp)
+{
+        struct sort_entry *se;
+        struct rb_node *nd;
+        size_t ret = 0;
+        unsigned long position = 1;
+        long displacement = 0;
+        unsigned int width;
+        const char *sep = symbol_conf.field_sep;
+        const char *col_width = symbol_conf.col_width_list_str;
+
+        init_rem_hits();
+
+        fprintf(fp, "# %s", pair ? "Baseline" : "Overhead");
+
+        if (symbol_conf.show_nr_samples) {
+                if (sep)
+                        fprintf(fp, "%cSamples", *sep);
+                else
+                        fputs("  Samples  ", fp);
+        }
+
+        if (symbol_conf.show_cpu_utilization) {
+                if (sep) {
+                        ret += fprintf(fp, "%csys", *sep);
+                        ret += fprintf(fp, "%cus", *sep);
+                        if (perf_guest) {
+                                ret += fprintf(fp, "%cguest sys", *sep);
+                                ret += fprintf(fp, "%cguest us", *sep);
+                        }
+                } else {
+                        ret += fprintf(fp, "  sys  ");
+                        ret += fprintf(fp, "  us  ");
+                        if (perf_guest) {
+                                ret += fprintf(fp, "  guest sys  ");
+                                ret += fprintf(fp, "  guest us  ");
+                        }
+                }
+        }
+
+        if (pair) {
+                if (sep)
+                        ret += fprintf(fp, "%cDelta", *sep);
+                else
+                        ret += fprintf(fp, "  Delta    ");
+
+                if (show_displacement) {
+                        if (sep)
+                                ret += fprintf(fp, "%cDisplacement", *sep);
+                        else
+                                ret += fprintf(fp, " Displ");
+                }
+        }
+
+        list_for_each_entry(se, &hist_entry__sort_list, list) {
+                if (se->elide)
+                        continue;
+                if (sep) {
+                        fprintf(fp, "%c%s", *sep, se->se_header);
+                        continue;
+                }
+                width = strlen(se->se_header);
+                if (symbol_conf.col_width_list_str) {
+                        if (col_width) {
+                                hists__set_col_len(self, se->se_width_idx,
+                                                   atoi(col_width));
+                                col_width = strchr(col_width, ',');
+                                if (col_width)
+                                        ++col_width;
+                        }
+                }
+                if (!hists__new_col_len(self, se->se_width_idx, width))
+                        width = hists__col_len(self, se->se_width_idx);
+                fprintf(fp, "  %*s", width, se->se_header);
+        }
+        fprintf(fp, "\n");
+
+        if (sep)
+                goto print_entries;
+
+        fprintf(fp, "# ........");
+        if (symbol_conf.show_nr_samples)
+                fprintf(fp, " ..........");
+        if (pair) {
+                fprintf(fp, " ..........");
+                if (show_displacement)
+                        fprintf(fp, " .....");
+        }
+        list_for_each_entry(se, &hist_entry__sort_list, list) {
+                unsigned int i;
+
+                if (se->elide)
+                        continue;
+
+                fprintf(fp, "  ");
+                width = hists__col_len(self, se->se_width_idx);
+                if (width == 0)
+                        width = strlen(se->se_header);
+                for (i = 0; i < width; i++)
+                        fprintf(fp, ".");
+        }
+
+        fprintf(fp, "\n#\n");
+
+print_entries:
+        for (nd = rb_first(&self->entries); nd; nd = rb_next(nd)) {
+                struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+
+                if (show_displacement) {
+                        if (h->pair != NULL)
+                                displacement = ((long)h->pair->position -
+                                                (long)position);
+                        else
+                                displacement = 0;
+                        ++position;
+                }
+                ret += hist_entry__fprintf(h, self, pair, show_displacement,
+                                           displacement, fp, self->stats.total_period);
+
+                if (symbol_conf.use_callchain)
+                        ret += hist_entry__fprintf_callchain(h, self, fp,
+                                                             self->stats.total_period);
+                if (h->ms.map == NULL && verbose > 1) {
+                        __map_groups__fprintf_maps(&h->thread->mg,
+                                                   MAP__FUNCTION, verbose, fp);
+                        fprintf(fp, "%.10s end\n", graph_dotted_line);
+                }
+        }
+
+        free(rem_sq_bracket);
+
+        return ret;
+}
+
+/*
+ * See hists__fprintf to match the column widths
+ */
+unsigned int hists__sort_list_width(struct hists *self)
+{
+        struct sort_entry *se;
+        int ret = 9; /* total % */
+
+        if (symbol_conf.show_cpu_utilization) {
+                ret += 7; /* count_sys % */
+                ret += 6; /* count_us % */
+                if (perf_guest) {
+                        ret += 13; /* count_guest_sys % */
+                        ret += 12; /* count_guest_us % */
+                }
+        }
+
+        if (symbol_conf.show_nr_samples)
+                ret += 11;
+
+        list_for_each_entry(se, &hist_entry__sort_list, list)
+                if (!se->elide)
+                        ret += 2 + hists__col_len(self, se->se_width_idx);
+
+        if (verbose) /* Addr + origin */
+                ret += 3 + BITS_PER_LONG / 4;
+
+        return ret;
+}
+
+static void hists__remove_entry_filter(struct hists *self, struct hist_entry *h,
+                                       enum hist_filter filter)
+{
+        h->filtered &= ~(1 << filter);
+        if (h->filtered)
+                return;
+
+        ++self->nr_entries;
+        if (h->ms.unfolded)
+                self->nr_entries += h->nr_rows;
+        h->row_offset = 0;
+        self->stats.total_period += h->period;
+        self->stats.nr_events[PERF_RECORD_SAMPLE] += h->nr_events;
+
+        hists__calc_col_len(self, h);
+}
+
+void hists__filter_by_dso(struct hists *self, const struct dso *dso)
+{
+        struct rb_node *nd;
+
+        self->nr_entries = self->stats.total_period = 0;
+        self->stats.nr_events[PERF_RECORD_SAMPLE] = 0;
+        hists__reset_col_len(self);
+
+        for (nd = rb_first(&self->entries); nd; nd = rb_next(nd)) {
+                struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+
+                if (symbol_conf.exclude_other && !h->parent)
+                        continue;
+
+                if (dso != NULL && (h->ms.map == NULL || h->ms.map->dso != dso)) {
+                        h->filtered |= (1 << HIST_FILTER__DSO);
+                        continue;
+                }
+
+                hists__remove_entry_filter(self, h, HIST_FILTER__DSO);
+        }
+}
+
+void hists__filter_by_thread(struct hists *self, const struct thread *thread)
+{
+        struct rb_node *nd;
+
+        self->nr_entries = self->stats.total_period = 0;
+        self->stats.nr_events[PERF_RECORD_SAMPLE] = 0;
+        hists__reset_col_len(self);
+
+        for (nd = rb_first(&self->entries); nd; nd = rb_next(nd)) {
+                struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+
+                if (thread != NULL && h->thread != thread) {
+                        h->filtered |= (1 << HIST_FILTER__THREAD);
+                        continue;
+                }
+
+                hists__remove_entry_filter(self, h, HIST_FILTER__THREAD);
+        }
+}
+
+static int symbol__alloc_hist(struct symbol *self)
+{
+        struct sym_priv *priv = symbol__priv(self);
+        const int size = (sizeof(*priv->hist) +
+                          (self->end - self->start) * sizeof(u64));
+
+        priv->hist = zalloc(size);
+        return priv->hist == NULL ? -1 : 0;
+}
+
+int hist_entry__inc_addr_samples(struct hist_entry *self, u64 ip)
+{
+        unsigned int sym_size, offset;
+        struct symbol *sym = self->ms.sym;
+        struct sym_priv *priv;
+        struct sym_hist *h;
+
+        if (!sym || !self->ms.map)
+                return 0;
+
+        priv = symbol__priv(sym);
+        if (priv->hist == NULL && symbol__alloc_hist(sym) < 0)
+                return -ENOMEM;
+
+        sym_size = sym->end - sym->start;
+        offset = ip - sym->start;
+
+        pr_debug3("%s: ip=%#Lx\n", __func__, self->ms.map->unmap_ip(self->ms.map, ip));
+
+        if (offset >= sym_size)
+                return 0;
+
+        h = priv->hist;
+        h->sum++;
+        h->ip[offset]++;
+
+        pr_debug3("%#Lx %s: period++ [ip: %#Lx, %#Lx] => %Ld\n", self->ms.sym->start,
+                  self->ms.sym->name, ip, ip - self->ms.sym->start, h->ip[offset]);
+        return 0;
+}
+
+static struct objdump_line *objdump_line__new(s64 offset, char *line, size_t privsize)
+{
+        struct objdump_line *self = malloc(sizeof(*self) + privsize);
+
+        if (self != NULL) {
+                self->offset = offset;
+                self->line = line;
+        }
+
+        return self;
+}
+
+void objdump_line__free(struct objdump_line *self)
+{
+        free(self->line);
+        free(self);
+}
+
+static void objdump__add_line(struct list_head *head, struct objdump_line *line)
+{
+        list_add_tail(&line->node, head);
+}
+
+struct objdump_line *objdump__get_next_ip_line(struct list_head *head,
+                                               struct objdump_line *pos)
+{
+        list_for_each_entry_continue(pos, head, node)
+                if (pos->offset >= 0)
+                        return pos;
+
+        return NULL;
+}
+
+static int hist_entry__parse_objdump_line(struct hist_entry *self, FILE *file,
+                                          struct list_head *head, size_t privsize)
+{
+        struct symbol *sym = self->ms.sym;
+        struct objdump_line *objdump_line;
+        char *line = NULL, *tmp, *tmp2, *c;
+        size_t line_len;
+        s64 line_ip, offset = -1;
+
+        if (getline(&line, &line_len, file) < 0)
+                return -1;
+
+        if (!line)
+                return -1;
+
+        while (line_len != 0 && isspace(line[line_len - 1]))
+                line[--line_len] = '\0';
+
+        c = strchr(line, '\n');
+        if (c)
+                *c = 0;
+
+        line_ip = -1;
+
+        /*
+         * Strip leading spaces:
+         */
+        tmp = line;
+        while (*tmp) {
+                if (*tmp != ' ')
+                        break;
+                tmp++;
+        }
+
+        if (*tmp) {
+                /*
+                 * Parse hexa addresses followed by ':'
+                 */
+                line_ip = strtoull(tmp, &tmp2, 16);
+                if (*tmp2 != ':' || tmp == tmp2 || tmp2[1] == '\0')
+                        line_ip = -1;
+        }
+
+        if (line_ip != -1) {
+                u64 start = map__rip_2objdump(self->ms.map, sym->start),
+                    end = map__rip_2objdump(self->ms.map, sym->end);
+
+                offset = line_ip - start;
+                if (offset < 0 || (u64)line_ip > end)
+                        offset = -1;
+        }
+
+        objdump_line = objdump_line__new(offset, line, privsize);
+        if (objdump_line == NULL) {
+                free(line);
+                return -1;
+        }
+        objdump__add_line(head, objdump_line);
+
+        return 0;
+}
+
+int hist_entry__annotate(struct hist_entry *self, struct list_head *head,
+                         size_t privsize)
+{
+        struct symbol *sym = self->ms.sym;
+        struct map *map = self->ms.map;
+        struct dso *dso = map->dso;
+        char *filename = dso__build_id_filename(dso, NULL, 0);
+        bool free_filename = true;
+        char command[PATH_MAX * 2];
+        FILE *file;
+        int err = 0;
+        u64 len;
+        char symfs_filename[PATH_MAX];
+
+        if (filename) {
+                snprintf(symfs_filename, sizeof(symfs_filename), "%s%s",
+                         symbol_conf.symfs, filename);
+        }
+
+        if (filename == NULL) {
+                if (dso->has_build_id) {
+                        pr_err("Can't annotate %s: not enough memory\n",
+                               sym->name);
+                        return -ENOMEM;
+                }
+                goto fallback;
+        } else if (readlink(symfs_filename, command, sizeof(command)) < 0 ||
+                   strstr(command, "[kernel.kallsyms]") ||
+                   access(symfs_filename, R_OK)) {
+                free(filename);
+fallback:
+                /*
+                 * If we don't have build-ids or the build-id file isn't in the
+                 * cache, or is just a kallsyms file, well, lets hope that this
+                 * DSO is the same as when 'perf record' ran.
+                 */
+                filename = dso->long_name;
+                snprintf(symfs_filename, sizeof(symfs_filename), "%s%s",
+                         symbol_conf.symfs, filename);
+                free_filename = false;
+        }
+
+        if (dso->origin == DSO__ORIG_KERNEL) {
+                if (dso->annotate_warned)
+                        goto out_free_filename;
+                err = -ENOENT;
+                dso->annotate_warned = 1;
+                pr_err("Can't annotate %s: No vmlinux file was found in the "
+                       "path\n", sym->name);
+                goto out_free_filename;
+        }
+
+        pr_debug("%s: filename=%s, sym=%s, start=%#Lx, end=%#Lx\n", __func__,
+                 filename, sym->name, map->unmap_ip(map, sym->start),
+                 map->unmap_ip(map, sym->end));
+
+        len = sym->end - sym->start;
+
+        pr_debug("annotating [%p] %30s : [%p] %30s\n",
+                 dso, dso->long_name, sym, sym->name);
+
+        snprintf(command, sizeof(command),
+                 "objdump --start-address=0x%016Lx --stop-address=0x%016Lx -dS -C %s|grep -v %s|expand",
+                 map__rip_2objdump(map, sym->start),
+                 map__rip_2objdump(map, sym->end),
+                 symfs_filename, filename);
+
+        pr_debug("Executing: %s\n", command);
+
+        file = popen(command, "r");
+        if (!file)
+                goto out_free_filename;
+
+        while (!feof(file))
+                if (hist_entry__parse_objdump_line(self, file, head, privsize) < 0)
+                        break;
+
+        pclose(file);
+out_free_filename:
+        if (free_filename)
+                free(filename);
+        return err;
+}
+
+void hists__inc_nr_events(struct hists *self, u32 type)
+{
+        ++self->stats.nr_events[0];
+        ++self->stats.nr_events[type];
+}
+
+size_t hists__fprintf_nr_events(struct hists *self, FILE *fp)
+{
+        int i;
+        size_t ret = 0;
+
+        for (i = 0; i < PERF_RECORD_HEADER_MAX; ++i) {
+                const char *name = event__get_event_name(i);
+
+                if (!strcmp(name, "UNKNOWN"))
+                        continue;
+
+                ret += fprintf(fp, "%16s events: %10d\n", name,
+                               self->stats.nr_events[i]);
+        }
+
+        return ret;
+}
diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h
new file mode 100644
index 00000000000..ee789856a8c
--- /dev/null
+++ b/tools/perf/util/hist.h
@@ -0,0 +1,150 @@
+#ifndef __PERF_HIST_H
+#define __PERF_HIST_H
+
+#include <linux/types.h>
+#include "callchain.h"
+
+extern struct callchain_param callchain_param;
+
+struct hist_entry;
+struct addr_location;
+struct symbol;
+struct rb_root;
+
+struct objdump_line {
+        struct list_head node;
+        s64              offset;
+        char             *line;
+};
+
+void objdump_line__free(struct objdump_line *self);
+struct objdump_line *objdump__get_next_ip_line(struct list_head *head,
+                                               struct objdump_line *pos);
+
+struct sym_hist {
+        u64             sum;
+        u64             ip[0];
+};
+
+struct sym_ext {
+        struct rb_node  node;
+        double          percent;
+        char            *path;
+};
+
+struct sym_priv {
+        struct sym_hist *hist;
+        struct sym_ext  *ext;
+};
+
+/*
+ * The kernel collects the number of events it couldn't send in a stretch and
+ * when possible sends this number in a PERF_RECORD_LOST event. The number of
+ * such "chunks" of lost events is stored in .nr_events[PERF_EVENT_LOST] while
+ * total_lost tells exactly how many events the kernel in fact lost, i.e. it is
+ * the sum of all struct lost_event.lost fields reported.
+ *
+ * The total_period is needed because by default auto-freq is used, so
+ * multipling nr_events[PERF_EVENT_SAMPLE] by a frequency isn't possible to get
+ * the total number of low level events, it is necessary to to sum all struct
+ * sample_event.period and stash the result in total_period.
+ */
+struct events_stats {
+        u64 total_period;
+        u64 total_lost;
+        u64 total_invalid_chains;
+        u32 nr_events[PERF_RECORD_HEADER_MAX];
+        u32 nr_unknown_events;
+        u32 nr_invalid_chains;
+};
+
+enum hist_column {
+        HISTC_SYMBOL,
+        HISTC_DSO,
+        HISTC_THREAD,
+        HISTC_COMM,
+        HISTC_PARENT,
+        HISTC_CPU,
+        HISTC_NR_COLS, /* Last entry */
+};
+
+struct hists {
+        struct rb_node          rb_node;
+        struct rb_root          entries;
+        u64                     nr_entries;
+        struct events_stats     stats;
+        u64                     config;
+        u64                     event_stream;
+        u32                     type;
+        u16                     col_len[HISTC_NR_COLS];
+};
+
+struct hist_entry *__hists__add_entry(struct hists *self,
+                                      struct addr_location *al,
+                                      struct symbol *parent, u64 period);
+extern int64_t hist_entry__cmp(struct hist_entry *, struct hist_entry *);
+extern int64_t hist_entry__collapse(struct hist_entry *, struct hist_entry *);
+int hist_entry__fprintf(struct hist_entry *self, struct hists *hists,
+                        struct hists *pair_hists, bool show_displacement,
+                        long displacement, FILE *fp, u64 total);
+int hist_entry__snprintf(struct hist_entry *self, char *bf, size_t size,
+                         struct hists *hists, struct hists *pair_hists,
+                         bool show_displacement, long displacement,
+                         bool color, u64 total);
+void hist_entry__free(struct hist_entry *);
+
+void hists__output_resort(struct hists *self);
+void hists__collapse_resort(struct hists *self);
+
+void hists__inc_nr_events(struct hists *self, u32 type);
+size_t hists__fprintf_nr_events(struct hists *self, FILE *fp);
+
+size_t hists__fprintf(struct hists *self, struct hists *pair,
+                      bool show_displacement, FILE *fp);
+
+int hist_entry__inc_addr_samples(struct hist_entry *self, u64 ip);
+int hist_entry__annotate(struct hist_entry *self, struct list_head *head,
+                         size_t privsize);
+
+void hists__filter_by_dso(struct hists *self, const struct dso *dso);
+void hists__filter_by_thread(struct hists *self, const struct thread *thread);
+
+u16 hists__col_len(struct hists *self, enum hist_column col);
+void hists__set_col_len(struct hists *self, enum hist_column col, u16 len);
+bool hists__new_col_len(struct hists *self, enum hist_column col, u16 len);
+
+#ifdef NO_NEWT_SUPPORT
+static inline int hists__browse(struct hists *self __used,
+                                const char *helpline __used,
+                                const char *ev_name __used)
+{
+        return 0;
+}
+
+static inline int hists__tui_browse_tree(struct rb_root *self __used,
+                                         const char *help __used)
+{
+        return 0;
+}
+
+static inline int hist_entry__tui_annotate(struct hist_entry *self __used)
+{
+        return 0;
+}
+#define KEY_LEFT -1
+#define KEY_RIGHT -2
+#else
+#include <newt.h>
+int hists__browse(struct hists *self, const char *helpline,
+                  const char *ev_name);
+int hist_entry__tui_annotate(struct hist_entry *self);
+
+#define KEY_LEFT NEWT_KEY_LEFT
+#define KEY_RIGHT NEWT_KEY_RIGHT
+
+int hists__tui_browse_tree(struct rb_root *self, const char *help);
+#endif
+
+unsigned int hists__sort_list_width(struct hists *self);
+
+#endif  /* __PERF_HIST_H */
diff --git a/tools/perf/util/hweight.c b/tools/perf/util/hweight.c
new file mode 100644
index 00000000000..5c1d0d099f0
--- /dev/null
+++ b/tools/perf/util/hweight.c
@@ -0,0 +1,31 @@
+#include <linux/bitops.h>
+
+/**
+ * hweightN - returns the hamming weight of a N-bit word
+ * @x: the word to weigh
+ *
+ * The Hamming Weight of a number is the total number of bits set in it.
+ */
+
+unsigned int hweight32(unsigned int w)
+{
+        unsigned int res = w - ((w >> 1) & 0x55555555);
+        res = (res & 0x33333333) + ((res >> 2) & 0x33333333);
+        res = (res + (res >> 4)) & 0x0F0F0F0F;
+        res = res + (res >> 8);
+        return (res + (res >> 16)) & 0x000000FF;
+}
+
+unsigned long hweight64(__u64 w)
+{
+#if BITS_PER_LONG == 32
+        return hweight32((unsigned int)(w >> 32)) + hweight32((unsigned int)w);
+#elif BITS_PER_LONG == 64
+        __u64 res = w - ((w >> 1) & 0x5555555555555555ul);
+        res = (res & 0x3333333333333333ul) + ((res >> 2) & 0x3333333333333333ul);
+        res = (res + (res >> 4)) & 0x0F0F0F0F0F0F0F0Ful;
+        res = res + (res >> 8);
+        res = res + (res >> 16);
+        return (res + (res >> 32)) & 0x00000000000000FFul;
+#endif
+}
diff --git a/tools/perf/util/include/asm/asm-offsets.h b/tools/perf/util/include/asm/asm-offsets.h
new file mode 100644
index 00000000000..ed538942523
--- /dev/null
+++ b/tools/perf/util/include/asm/asm-offsets.h
@@ -0,0 +1 @@
+/* stub */
diff --git a/tools/perf/util/include/asm/bug.h b/tools/perf/util/include/asm/bug.h
new file mode 100644
index 00000000000..7fcc6810adc
--- /dev/null
+++ b/tools/perf/util/include/asm/bug.h
@@ -0,0 +1,22 @@
+#ifndef _PERF_ASM_GENERIC_BUG_H
+#define _PERF_ASM_GENERIC_BUG_H
+
+#define __WARN_printf(arg...)   do { fprintf(stderr, arg); } while (0)
+
+#define WARN(condition, format...) ({           \
+        int __ret_warn_on = !!(condition);      \
+        if (unlikely(__ret_warn_on))            \
+                __WARN_printf(format);          \
+        unlikely(__ret_warn_on);                \
+})
+
+#define WARN_ONCE(condition, format...) ({      \
+        static int __warned;                    \
+        int __ret_warn_once = !!(condition);    \
+                                                \
+        if (unlikely(__ret_warn_once))          \
+                if (WARN(!__warned, format))    \
+                        __warned = 1;           \
+        unlikely(__ret_warn_once);              \
+})
+#endif
diff --git a/tools/perf/util/include/asm/byteorder.h b/tools/perf/util/include/asm/byteorder.h
new file mode 100644
index 00000000000..b722abe3a62
--- /dev/null
+++ b/tools/perf/util/include/asm/byteorder.h
@@ -0,0 +1,2 @@
+#include <asm/types.h>
+#include "../../../../include/linux/swab.h"
diff --git a/tools/perf/util/include/asm/cpufeature.h b/tools/perf/util/include/asm/cpufeature.h
new file mode 100644
index 00000000000..acffd5e4d1d
--- /dev/null
+++ b/tools/perf/util/include/asm/cpufeature.h
@@ -0,0 +1,9 @@
+
+#ifndef PERF_CPUFEATURE_H
+#define PERF_CPUFEATURE_H
+
+/* cpufeature.h ... dummy header file for including arch/x86/lib/memcpy_64.S */
+
+#define X86_FEATURE_REP_GOOD 0
+
+#endif  /* PERF_CPUFEATURE_H */
diff --git a/tools/perf/util/include/asm/dwarf2.h b/tools/perf/util/include/asm/dwarf2.h
new file mode 100644
index 00000000000..bb4198e7837
--- /dev/null
+++ b/tools/perf/util/include/asm/dwarf2.h
@@ -0,0 +1,11 @@
+
+#ifndef PERF_DWARF2_H
+#define PERF_DWARF2_H
+
+/* dwarf2.h ... dummy header file for including arch/x86/lib/memcpy_64.S */
+
+#define CFI_STARTPROC
+#define CFI_ENDPROC
+
+#endif  /* PERF_DWARF2_H */
+
diff --git a/tools/perf/util/include/asm/hweight.h b/tools/perf/util/include/asm/hweight.h
new file mode 100644
index 00000000000..36cf26d434a
--- /dev/null
+++ b/tools/perf/util/include/asm/hweight.h
@@ -0,0 +1,8 @@
+#ifndef PERF_HWEIGHT_H
+#define PERF_HWEIGHT_H
+
+#include <linux/types.h>
+unsigned int hweight32(unsigned int w);
+unsigned long hweight64(__u64 w);
+
+#endif /* PERF_HWEIGHT_H */
diff --git a/tools/perf/util/include/asm/swab.h b/tools/perf/util/include/asm/swab.h
new file mode 100644
index 00000000000..ed538942523
--- /dev/null
+++ b/tools/perf/util/include/asm/swab.h
@@ -0,0 +1 @@
+/* stub */
diff --git a/tools/perf/util/include/asm/system.h b/tools/perf/util/include/asm/system.h
new file mode 100644
index 00000000000..710cecca972
--- /dev/null
+++ b/tools/perf/util/include/asm/system.h
@@ -0,0 +1 @@
+/* Empty */
diff --git a/tools/perf/util/include/asm/uaccess.h b/tools/perf/util/include/asm/uaccess.h
new file mode 100644
index 00000000000..d0f72b8fcc3
--- /dev/null
+++ b/tools/perf/util/include/asm/uaccess.h
@@ -0,0 +1,14 @@
+#ifndef _PERF_ASM_UACCESS_H_
+#define _PERF_ASM_UACCESS_H_
+
+#define __get_user(src, dest)                                           \
+({                                                                      \
+        (src) = *dest;                                                  \
+        0;                                                              \
+})
+
+#define get_user        __get_user
+
+#define access_ok(type, addr, size)     1
+
+#endif
diff --git a/tools/perf/util/include/dwarf-regs.h b/tools/perf/util/include/dwarf-regs.h
new file mode 100644
index 00000000000..cf6727e99c4
--- /dev/null
+++ b/tools/perf/util/include/dwarf-regs.h
@@ -0,0 +1,8 @@
+#ifndef _PERF_DWARF_REGS_H_
+#define _PERF_DWARF_REGS_H_
+
+#ifdef DWARF_SUPPORT
+const char *get_arch_regstr(unsigned int n);
+#endif
+
+#endif
diff --git a/tools/perf/util/include/linux/bitmap.h b/tools/perf/util/include/linux/bitmap.h
new file mode 100644
index 00000000000..eda4416efa0
--- /dev/null
+++ b/tools/perf/util/include/linux/bitmap.h
@@ -0,0 +1,35 @@
+#ifndef _PERF_BITOPS_H
+#define _PERF_BITOPS_H
+
+#include <string.h>
+#include <linux/bitops.h>
+
+int __bitmap_weight(const unsigned long *bitmap, int bits);
+
+#define BITMAP_LAST_WORD_MASK(nbits)                                    \
+(                                                                       \
+        ((nbits) % BITS_PER_LONG) ?                                     \
+                (1UL<<((nbits) % BITS_PER_LONG))-1 : ~0UL               \
+)
+
+#define small_const_nbits(nbits) \
+        (__builtin_constant_p(nbits) && (nbits) <= BITS_PER_LONG)
+
+static inline void bitmap_zero(unsigned long *dst, int nbits)
+{
+        if (small_const_nbits(nbits))
+                *dst = 0UL;
+        else {
+                int len = BITS_TO_LONGS(nbits) * sizeof(unsigned long);
+                memset(dst, 0, len);
+        }
+}
+
+static inline int bitmap_weight(const unsigned long *src, int nbits)
+{
+        if (small_const_nbits(nbits))
+                return hweight_long(*src & BITMAP_LAST_WORD_MASK(nbits));
+        return __bitmap_weight(src, nbits);
+}
+
+#endif /* _PERF_BITOPS_H */
diff --git a/tools/perf/util/include/linux/bitops.h b/tools/perf/util/include/linux/bitops.h
new file mode 100644
index 00000000000..8be0b968ca0
--- /dev/null
+++ b/tools/perf/util/include/linux/bitops.h
@@ -0,0 +1,32 @@
+#ifndef _PERF_LINUX_BITOPS_H_
+#define _PERF_LINUX_BITOPS_H_
+
+#include <linux/kernel.h>
+#include <asm/hweight.h>
+
+#define BITS_PER_LONG __WORDSIZE
+#define BITS_PER_BYTE           8
+#define BITS_TO_LONGS(nr)       DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(long))
+
+static inline void set_bit(int nr, unsigned long *addr)
+{
+        addr[nr / BITS_PER_LONG] |= 1UL << (nr % BITS_PER_LONG);
+}
+
+static inline void clear_bit(int nr, unsigned long *addr)
+{
+        addr[nr / BITS_PER_LONG] &= ~(1UL << (nr % BITS_PER_LONG));
+}
+
+static __always_inline int test_bit(unsigned int nr, const unsigned long *addr)
+{
+        return ((1UL << (nr % BITS_PER_LONG)) &
+                (((unsigned long *)addr)[nr / BITS_PER_LONG])) != 0;
+}
+
+static inline unsigned long hweight_long(unsigned long w)
+{
+        return sizeof(w) == 4 ? hweight32(w) : hweight64(w);
+}
+
+#endif
diff --git a/tools/perf/util/include/linux/compiler.h b/tools/perf/util/include/linux/compiler.h
new file mode 100644
index 00000000000..791f9dd27eb
--- /dev/null
+++ b/tools/perf/util/include/linux/compiler.h
@@ -0,0 +1,12 @@
+#ifndef _PERF_LINUX_COMPILER_H_
+#define _PERF_LINUX_COMPILER_H_
+
+#ifndef __always_inline
+#define __always_inline inline
+#endif
+#define __user
+#define __attribute_const__
+
+#define __used          __attribute__((__unused__))
+
+#endif
diff --git a/tools/perf/util/include/linux/ctype.h b/tools/perf/util/include/linux/ctype.h
new file mode 100644
index 00000000000..a53d4ee1e0b
--- /dev/null
+++ b/tools/perf/util/include/linux/ctype.h
@@ -0,0 +1 @@
+#include "../util.h"
diff --git a/tools/perf/util/include/linux/hash.h b/tools/perf/util/include/linux/hash.h
new file mode 100644
index 00000000000..201f5739799
--- /dev/null
+++ b/tools/perf/util/include/linux/hash.h
@@ -0,0 +1,5 @@
+#include "../../../../include/linux/hash.h"
+
+#ifndef PERF_HASH_H
+#define PERF_HASH_H
+#endif
diff --git a/tools/perf/util/include/linux/kernel.h b/tools/perf/util/include/linux/kernel.h
new file mode 100644
index 00000000000..1eb804fd3fb
--- /dev/null
+++ b/tools/perf/util/include/linux/kernel.h
@@ -0,0 +1,111 @@
+#ifndef PERF_LINUX_KERNEL_H_
+#define PERF_LINUX_KERNEL_H_
+
+#include <stdarg.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <assert.h>
+
+#define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
+
+#define ALIGN(x,a)              __ALIGN_MASK(x,(typeof(x))(a)-1)
+#define __ALIGN_MASK(x,mask)    (((x)+(mask))&~(mask))
+
+#ifndef offsetof
+#define offsetof(TYPE, MEMBER) ((size_t) &((TYPE *)0)->MEMBER)
+#endif
+
+#ifndef container_of
+/**
+ * container_of - cast a member of a structure out to the containing structure
+ * @ptr:        the pointer to the member.
+ * @type:       the type of the container struct this is embedded in.
+ * @member:     the name of the member within the struct.
+ *
+ */
+#define container_of(ptr, type, member) ({                      \
+        const typeof(((type *)0)->member) * __mptr = (ptr);     \
+        (type *)((char *)__mptr - offsetof(type, member)); })
+#endif
+
+#define BUILD_BUG_ON_ZERO(e) (sizeof(struct { int:-!!(e); }))
+
+#ifndef max
+#define max(x, y) ({                            \
+        typeof(x) _max1 = (x);                  \
+        typeof(y) _max2 = (y);                  \
+        (void) (&_max1 == &_max2);              \
+        _max1 > _max2 ? _max1 : _max2; })
+#endif
+
+#ifndef min
+#define min(x, y) ({                            \
+        typeof(x) _min1 = (x);                  \
+        typeof(y) _min2 = (y);                  \
+        (void) (&_min1 == &_min2);              \
+        _min1 < _min2 ? _min1 : _min2; })
+#endif
+
+#ifndef BUG_ON
+#define BUG_ON(cond) assert(!(cond))
+#endif
+
+/*
+ * Both need more care to handle endianness
+ * (Don't use bitmap_copy_le() for now)
+ */
+#define cpu_to_le64(x)  (x)
+#define cpu_to_le32(x)  (x)
+
+static inline int
+vscnprintf(char *buf, size_t size, const char *fmt, va_list args)
+{
+        int i;
+        ssize_t ssize = size;
+
+        i = vsnprintf(buf, size, fmt, args);
+
+        return (i >= ssize) ? (ssize - 1) : i;
+}
+
+static inline int scnprintf(char * buf, size_t size, const char * fmt, ...)
+{
+        va_list args;
+        ssize_t ssize = size;
+        int i;
+
+        va_start(args, fmt);
+        i = vsnprintf(buf, size, fmt, args);
+        va_end(args);
+
+        return (i >= ssize) ? (ssize - 1) : i;
+}
+
+static inline unsigned long
+simple_strtoul(const char *nptr, char **endptr, int base)
+{
+        return strtoul(nptr, endptr, base);
+}
+
+int eprintf(int level,
+            const char *fmt, ...) __attribute__((format(printf, 2, 3)));
+
+#ifndef pr_fmt
+#define pr_fmt(fmt) fmt
+#endif
+
+#define pr_err(fmt, ...) \
+        eprintf(0, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_warning(fmt, ...) \
+        eprintf(0, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_info(fmt, ...) \
+        eprintf(0, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_debug(fmt, ...) \
+        eprintf(1, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_debugN(n, fmt, ...) \
+        eprintf(n, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_debug2(fmt, ...) pr_debugN(2, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_debug3(fmt, ...) pr_debugN(3, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_debug4(fmt, ...) pr_debugN(4, pr_fmt(fmt), ##__VA_ARGS__)
+
+#endif
diff --git a/tools/perf/util/include/linux/linkage.h b/tools/perf/util/include/linux/linkage.h
new file mode 100644
index 00000000000..06387cffe12
--- /dev/null
+++ b/tools/perf/util/include/linux/linkage.h
@@ -0,0 +1,13 @@
+
+#ifndef PERF_LINUX_LINKAGE_H_
+#define PERF_LINUX_LINKAGE_H_
+
+/* linkage.h ... for including arch/x86/lib/memcpy_64.S */
+
+#define ENTRY(name)                             \
+        .globl name;                            \
+        name:
+
+#define ENDPROC(name)
+
+#endif  /* PERF_LINUX_LINKAGE_H_ */
diff --git a/tools/perf/util/include/linux/list.h b/tools/perf/util/include/linux/list.h
new file mode 100644
index 00000000000..f5ca26e53fb
--- /dev/null
+++ b/tools/perf/util/include/linux/list.h
@@ -0,0 +1,26 @@
+#include "../../../../include/linux/list.h"
+
+#ifndef PERF_LIST_H
+#define PERF_LIST_H
+/**
+ * list_del_range - deletes range of entries from list.
+ * @begin: first element in the range to delete from the list.
+ * @end: last element in the range to delete from the list.
+ * Note: list_empty on the range of entries does not return true after this,
+ * the entries is in an undefined state.
+ */
+static inline void list_del_range(struct list_head *begin,
+                                  struct list_head *end)
+{
+        begin->prev->next = end->next;
+        end->next->prev = begin->prev;
+}
+
+/**
+ * list_for_each_from   -       iterate over a list from one of its nodes
+ * @pos:  the &struct list_head to use as a loop cursor, from where to start
+ * @head: the head for your list.
+ */
+#define list_for_each_from(pos, head) \
+        for (; prefetch(pos->next), pos != (head); pos = pos->next)
+#endif
diff --git a/tools/perf/util/include/linux/module.h b/tools/perf/util/include/linux/module.h
new file mode 100644
index 00000000000..b43e2dc21e0
--- /dev/null
+++ b/tools/perf/util/include/linux/module.h
@@ -0,0 +1,6 @@
+#ifndef PERF_LINUX_MODULE_H
+#define PERF_LINUX_MODULE_H
+
+#define EXPORT_SYMBOL(name)
+
+#endif
diff --git a/tools/perf/util/include/linux/poison.h b/tools/perf/util/include/linux/poison.h
new file mode 100644
index 00000000000..fef6dbc9ce1
--- /dev/null
+++ b/tools/perf/util/include/linux/poison.h
@@ -0,0 +1 @@
+#include "../../../../include/linux/poison.h"
diff --git a/tools/perf/util/include/linux/prefetch.h b/tools/perf/util/include/linux/prefetch.h
new file mode 100644
index 00000000000..7841e485d8c
--- /dev/null
+++ b/tools/perf/util/include/linux/prefetch.h
@@ -0,0 +1,6 @@
+#ifndef PERF_LINUX_PREFETCH_H
+#define PERF_LINUX_PREFETCH_H
+
+static inline void prefetch(void *a __attribute__((unused))) { }
+
+#endif
diff --git a/tools/perf/util/include/linux/rbtree.h b/tools/perf/util/include/linux/rbtree.h
new file mode 100644
index 00000000000..7a243a14303
--- /dev/null
+++ b/tools/perf/util/include/linux/rbtree.h
@@ -0,0 +1 @@
+#include "../../../../include/linux/rbtree.h"
diff --git a/tools/perf/util/include/linux/string.h b/tools/perf/util/include/linux/string.h
new file mode 100644
index 00000000000..3b2f5900276
--- /dev/null
+++ b/tools/perf/util/include/linux/string.h
@@ -0,0 +1 @@
+#include <string.h>
diff --git a/tools/perf/util/include/linux/types.h b/tools/perf/util/include/linux/types.h
new file mode 100644
index 00000000000..12de3b8112f
--- /dev/null
+++ b/tools/perf/util/include/linux/types.h
@@ -0,0 +1,21 @@
+#ifndef _PERF_LINUX_TYPES_H_
+#define _PERF_LINUX_TYPES_H_
+
+#include <asm/types.h>
+
+#define DECLARE_BITMAP(name,bits) \
+        unsigned long name[BITS_TO_LONGS(bits)]
+
+struct list_head {
+        struct list_head *next, *prev;
+};
+
+struct hlist_head {
+        struct hlist_node *first;
+};
+
+struct hlist_node {
+        struct hlist_node *next, **pprev;
+};
+
+#endif
diff --git a/tools/perf/util/levenshtein.h b/tools/perf/util/levenshtein.h
index 0173abeef52..b0fcb6d8a88 100644
--- a/tools/perf/util/levenshtein.h
+++ b/tools/perf/util/levenshtein.h
@@ -1,8 +1,8 @@
-#ifndef LEVENSHTEIN_H
-#define LEVENSHTEIN_H
+#ifndef __PERF_LEVENSHTEIN_H
+#define __PERF_LEVENSHTEIN_H
 
 int levenshtein(const char *string1, const char *string2,
         int swap_penalty, int substition_penalty,
         int insertion_penalty, int deletion_penalty);
 
-#endif
+#endif /* __PERF_LEVENSHTEIN_H */
diff --git a/tools/perf/util/list.h b/tools/perf/util/list.h
deleted file mode 100644
index e2548e8072c..00000000000
--- a/tools/perf/util/list.h
+++ /dev/null
@@ -1,603 +0,0 @@
-#ifndef _LINUX_LIST_H
-#define _LINUX_LIST_H
-/*
-  Copyright (C) Cast of dozens, comes from the Linux kernel
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms of version 2 of the GNU General Public License as
-  published by the Free Software Foundation.
-*/
-
-#include <stddef.h>
-
-/*
- * These are non-NULL pointers that will result in page faults
- * under normal circumstances, used to verify that nobody uses
- * non-initialized list entries.
- */
-#define LIST_POISON1 ((void *)0x00100100)
-#define LIST_POISON2 ((void *)0x00200200)
-
-/**
- * container_of - cast a member of a structure out to the containing structure
- * @ptr:        the pointer to the member.
- * @type:       the type of the container struct this is embedded in.
- * @member:     the name of the member within the struct.
- *
- */
-#define container_of(ptr, type, member) ({                      \
-        const typeof( ((type *)0)->member ) *__mptr = (ptr);    \
-        (type *)( (char *)__mptr - offsetof(type,member) );})
-
-/*
- * Simple doubly linked list implementation.
- *
- * Some of the internal functions ("__xxx") are useful when
- * manipulating whole lists rather than single entries, as
- * sometimes we already know the next/prev entries and we can
- * generate better code by using them directly rather than
- * using the generic single-entry routines.
- */
-
-struct list_head {
-        struct list_head *next, *prev;
-};
-
-#define LIST_HEAD_INIT(name) { &(name), &(name) }
-
-#define LIST_HEAD(name) \
-        struct list_head name = LIST_HEAD_INIT(name)
-
-static inline void INIT_LIST_HEAD(struct list_head *list)
-{
-        list->next = list;
-        list->prev = list;
-}
-
-/*
- * Insert a new entry between two known consecutive entries.
- *
- * This is only for internal list manipulation where we know
- * the prev/next entries already!
- */
-static inline void __list_add(struct list_head *new,
-                              struct list_head *prev,
-                              struct list_head *next)
-{
-        next->prev = new;
-        new->next = next;
-        new->prev = prev;
-        prev->next = new;
-}
-
-/**
- * list_add - add a new entry
- * @new: new entry to be added
- * @head: list head to add it after
- *
- * Insert a new entry after the specified head.
- * This is good for implementing stacks.
- */
-static inline void list_add(struct list_head *new, struct list_head *head)
-{
-        __list_add(new, head, head->next);
-}
-
-/**
- * list_add_tail - add a new entry
- * @new: new entry to be added
- * @head: list head to add it before
- *
- * Insert a new entry before the specified head.
- * This is useful for implementing queues.
- */
-static inline void list_add_tail(struct list_head *new, struct list_head *head)
-{
-        __list_add(new, head->prev, head);
-}
-
-/*
- * Delete a list entry by making the prev/next entries
- * point to each other.
- *
- * This is only for internal list manipulation where we know
- * the prev/next entries already!
- */
-static inline void __list_del(struct list_head * prev, struct list_head * next)
-{
-        next->prev = prev;
-        prev->next = next;
-}
-
-/**
- * list_del - deletes entry from list.
- * @entry: the element to delete from the list.
- * Note: list_empty on entry does not return true after this, the entry is
- * in an undefined state.
- */
-static inline void list_del(struct list_head *entry)
-{
-        __list_del(entry->prev, entry->next);
-        entry->next = LIST_POISON1;
-        entry->prev = LIST_POISON2;
-}
-
-/**
- * list_del_range - deletes range of entries from list.
- * @beging: first element in the range to delete from the list.
- * @beging: first element in the range to delete from the list.
- * Note: list_empty on the range of entries does not return true after this,
- * the entries is in an undefined state.
- */
-static inline void list_del_range(struct list_head *begin,
-                                  struct list_head *end)
-{
-        begin->prev->next = end->next;
-        end->next->prev = begin->prev;
-}
-
-/**
- * list_replace - replace old entry by new one
- * @old : the element to be replaced
- * @new : the new element to insert
- * Note: if 'old' was empty, it will be overwritten.
- */
-static inline void list_replace(struct list_head *old,
-                                struct list_head *new)
-{
-        new->next = old->next;
-        new->next->prev = new;
-        new->prev = old->prev;
-        new->prev->next = new;
-}
-
-static inline void list_replace_init(struct list_head *old,
-                                        struct list_head *new)
-{
-        list_replace(old, new);
-        INIT_LIST_HEAD(old);
-}
-
-/**
- * list_del_init - deletes entry from list and reinitialize it.
- * @entry: the element to delete from the list.
- */
-static inline void list_del_init(struct list_head *entry)
-{
-        __list_del(entry->prev, entry->next);
-        INIT_LIST_HEAD(entry);
-}
-
-/**
- * list_move - delete from one list and add as another's head
- * @list: the entry to move
- * @head: the head that will precede our entry
- */
-static inline void list_move(struct list_head *list, struct list_head *head)
-{
-        __list_del(list->prev, list->next);
-        list_add(list, head);
-}
-
-/**
- * list_move_tail - delete from one list and add as another's tail
- * @list: the entry to move
- * @head: the head that will follow our entry
- */
-static inline void list_move_tail(struct list_head *list,
-                                  struct list_head *head)
-{
-        __list_del(list->prev, list->next);
-        list_add_tail(list, head);
-}
-
-/**
- * list_is_last - tests whether @list is the last entry in list @head
- * @list: the entry to test
- * @head: the head of the list
- */
-static inline int list_is_last(const struct list_head *list,
-                                const struct list_head *head)
-{
-        return list->next == head;
-}
-
-/**
- * list_empty - tests whether a list is empty
- * @head: the list to test.
- */
-static inline int list_empty(const struct list_head *head)
-{
-        return head->next == head;
-}
-
-/**
- * list_empty_careful - tests whether a list is empty and not being modified
- * @head: the list to test
- *
- * Description:
- * tests whether a list is empty _and_ checks that no other CPU might be
- * in the process of modifying either member (next or prev)
- *
- * NOTE: using list_empty_careful() without synchronization
- * can only be safe if the only activity that can happen
- * to the list entry is list_del_init(). Eg. it cannot be used
- * if another CPU could re-list_add() it.
- */
-static inline int list_empty_careful(const struct list_head *head)
-{
-        struct list_head *next = head->next;
-        return (next == head) && (next == head->prev);
-}
-
-static inline void __list_splice(struct list_head *list,
-                                 struct list_head *head)
-{
-        struct list_head *first = list->next;
-        struct list_head *last = list->prev;
-        struct list_head *at = head->next;
-
-        first->prev = head;
-        head->next = first;
-
-        last->next = at;
-        at->prev = last;
-}
-
-/**
- * list_splice - join two lists
- * @list: the new list to add.
- * @head: the place to add it in the first list.
- */
-static inline void list_splice(struct list_head *list, struct list_head *head)
-{
-        if (!list_empty(list))
-                __list_splice(list, head);
-}
-
-/**
- * list_splice_init - join two lists and reinitialise the emptied list.
- * @list: the new list to add.
- * @head: the place to add it in the first list.
- *
- * The list at @list is reinitialised
- */
-static inline void list_splice_init(struct list_head *list,
-                                    struct list_head *head)
-{
-        if (!list_empty(list)) {
-                __list_splice(list, head);
-                INIT_LIST_HEAD(list);
-        }
-}
-
-/**
- * list_entry - get the struct for this entry
- * @ptr:        the &struct list_head pointer.
- * @type:       the type of the struct this is embedded in.
- * @member:     the name of the list_struct within the struct.
- */
-#define list_entry(ptr, type, member) \
-        container_of(ptr, type, member)
-
-/**
- * list_first_entry - get the first element from a list
- * @ptr:       the list head to take the element from.
- * @type:      the type of the struct this is embedded in.
- * @member:    the name of the list_struct within the struct.
- *
- * Note, that list is expected to be not empty.
- */
-#define list_first_entry(ptr, type, member) \
-        list_entry((ptr)->next, type, member)
-
-/**
- * list_for_each        -       iterate over a list
- * @pos:        the &struct list_head to use as a loop cursor.
- * @head:       the head for your list.
- */
-#define list_for_each(pos, head) \
-        for (pos = (head)->next; pos != (head); \
-                pos = pos->next)
-
-/**
- * __list_for_each      -       iterate over a list
- * @pos:        the &struct list_head to use as a loop cursor.
- * @head:       the head for your list.
- *
- * This variant differs from list_for_each() in that it's the
- * simplest possible list iteration code, no prefetching is done.
- * Use this for code that knows the list to be very short (empty
- * or 1 entry) most of the time.
- */
-#define __list_for_each(pos, head) \
-        for (pos = (head)->next; pos != (head); pos = pos->next)
-
-/**
- * list_for_each_prev   -       iterate over a list backwards
- * @pos:        the &struct list_head to use as a loop cursor.
- * @head:       the head for your list.
- */
-#define list_for_each_prev(pos, head) \
-        for (pos = (head)->prev; pos != (head); \
-                pos = pos->prev)
-
-/**
- * list_for_each_safe - iterate over a list safe against removal of list entry
- * @pos:        the &struct list_head to use as a loop cursor.
- * @n:          another &struct list_head to use as temporary storage
- * @head:       the head for your list.
- */
-#define list_for_each_safe(pos, n, head) \
-        for (pos = (head)->next, n = pos->next; pos != (head); \
-                pos = n, n = pos->next)
-
-/**
- * list_for_each_entry  -       iterate over list of given type
- * @pos:        the type * to use as a loop cursor.
- * @head:       the head for your list.
- * @member:     the name of the list_struct within the struct.
- */
-#define list_for_each_entry(pos, head, member)                          \
-        for (pos = list_entry((head)->next, typeof(*pos), member);      \
-             &pos->member != (head);    \
-             pos = list_entry(pos->member.next, typeof(*pos), member))
-
-/**
- * list_for_each_entry_reverse - iterate backwards over list of given type.
- * @pos:        the type * to use as a loop cursor.
- * @head:       the head for your list.
- * @member:     the name of the list_struct within the struct.
- */
-#define list_for_each_entry_reverse(pos, head, member)                  \
-        for (pos = list_entry((head)->prev, typeof(*pos), member);      \
-             &pos->member != (head);    \
-             pos = list_entry(pos->member.prev, typeof(*pos), member))
-
-/**
- * list_prepare_entry - prepare a pos entry for use in list_for_each_entry_continue
- * @pos:        the type * to use as a start point
- * @head:       the head of the list
- * @member:     the name of the list_struct within the struct.
- *
- * Prepares a pos entry for use as a start point in list_for_each_entry_continue.
- */
-#define list_prepare_entry(pos, head, member) \
-        ((pos) ? : list_entry(head, typeof(*pos), member))
-
-/**
- * list_for_each_entry_continue - continue iteration over list of given type
- * @pos:        the type * to use as a loop cursor.
- * @head:       the head for your list.
- * @member:     the name of the list_struct within the struct.
- *
- * Continue to iterate over list of given type, continuing after
- * the current position.
- */
-#define list_for_each_entry_continue(pos, head, member)                 \
-        for (pos = list_entry(pos->member.next, typeof(*pos), member);  \
-             &pos->member != (head);    \
-             pos = list_entry(pos->member.next, typeof(*pos), member))
-
-/**
- * list_for_each_entry_from - iterate over list of given type from the current point
- * @pos:        the type * to use as a loop cursor.
- * @head:       the head for your list.
- * @member:     the name of the list_struct within the struct.
- *
- * Iterate over list of given type, continuing from current position.
- */
-#define list_for_each_entry_from(pos, head, member)                     \
-        for (; &pos->member != (head);  \
-             pos = list_entry(pos->member.next, typeof(*pos), member))
-
-/**
- * list_for_each_entry_safe - iterate over list of given type safe against removal of list entry
- * @pos:        the type * to use as a loop cursor.
- * @n:          another type * to use as temporary storage
- * @head:       the head for your list.
- * @member:     the name of the list_struct within the struct.
- */
-#define list_for_each_entry_safe(pos, n, head, member)                  \
-        for (pos = list_entry((head)->next, typeof(*pos), member),      \
-                n = list_entry(pos->member.next, typeof(*pos), member); \
-             &pos->member != (head);                                    \
-             pos = n, n = list_entry(n->member.next, typeof(*n), member))
-
-/**
- * list_for_each_entry_safe_continue
- * @pos:        the type * to use as a loop cursor.
- * @n:          another type * to use as temporary storage
- * @head:       the head for your list.
- * @member:     the name of the list_struct within the struct.
- *
- * Iterate over list of given type, continuing after current point,
- * safe against removal of list entry.
- */
-#define list_for_each_entry_safe_continue(pos, n, head, member)                 \
-        for (pos = list_entry(pos->member.next, typeof(*pos), member),          \
-                n = list_entry(pos->member.next, typeof(*pos), member);         \
-             &pos->member != (head);                                            \
-             pos = n, n = list_entry(n->member.next, typeof(*n), member))
-
-/**
- * list_for_each_entry_safe_from
- * @pos:        the type * to use as a loop cursor.
- * @n:          another type * to use as temporary storage
- * @head:       the head for your list.
- * @member:     the name of the list_struct within the struct.
- *
- * Iterate over list of given type from current point, safe against
- * removal of list entry.
- */
-#define list_for_each_entry_safe_from(pos, n, head, member)                     \
-        for (n = list_entry(pos->member.next, typeof(*pos), member);            \
-             &pos->member != (head);                                            \
-             pos = n, n = list_entry(n->member.next, typeof(*n), member))
-
-/**
- * list_for_each_entry_safe_reverse
- * @pos:        the type * to use as a loop cursor.
- * @n:          another type * to use as temporary storage
- * @head:       the head for your list.
- * @member:     the name of the list_struct within the struct.
- *
- * Iterate backwards over list of given type, safe against removal
- * of list entry.
- */
-#define list_for_each_entry_safe_reverse(pos, n, head, member)          \
-        for (pos = list_entry((head)->prev, typeof(*pos), member),      \
-                n = list_entry(pos->member.prev, typeof(*pos), member); \
-             &pos->member != (head);                                    \
-             pos = n, n = list_entry(n->member.prev, typeof(*n), member))
-
-/*
- * Double linked lists with a single pointer list head.
- * Mostly useful for hash tables where the two pointer list head is
- * too wasteful.
- * You lose the ability to access the tail in O(1).
- */
-
-struct hlist_head {
-        struct hlist_node *first;
-};
-
-struct hlist_node {
-        struct hlist_node *next, **pprev;
-};
-
-#define HLIST_HEAD_INIT { .first = NULL }
-#define HLIST_HEAD(name) struct hlist_head name = {  .first = NULL }
-#define INIT_HLIST_HEAD(ptr) ((ptr)->first = NULL)
-static inline void INIT_HLIST_NODE(struct hlist_node *h)
-{
-        h->next = NULL;
-        h->pprev = NULL;
-}
-
-static inline int hlist_unhashed(const struct hlist_node *h)
-{
-        return !h->pprev;
-}
-
-static inline int hlist_empty(const struct hlist_head *h)
-{
-        return !h->first;
-}
-
-static inline void __hlist_del(struct hlist_node *n)
-{
-        struct hlist_node *next = n->next;
-        struct hlist_node **pprev = n->pprev;
-        *pprev = next;
-        if (next)
-                next->pprev = pprev;
-}
-
-static inline void hlist_del(struct hlist_node *n)
-{
-        __hlist_del(n);
-        n->next = LIST_POISON1;
-        n->pprev = LIST_POISON2;
-}
-
-static inline void hlist_del_init(struct hlist_node *n)
-{
-        if (!hlist_unhashed(n)) {
-                __hlist_del(n);
-                INIT_HLIST_NODE(n);
-        }
-}
-
-static inline void hlist_add_head(struct hlist_node *n, struct hlist_head *h)
-{
-        struct hlist_node *first = h->first;
-        n->next = first;
-        if (first)
-                first->pprev = &n->next;
-        h->first = n;
-        n->pprev = &h->first;
-}
-
-/* next must be != NULL */
-static inline void hlist_add_before(struct hlist_node *n,
-                                        struct hlist_node *next)
-{
-        n->pprev = next->pprev;
-        n->next = next;
-        next->pprev = &n->next;
-        *(n->pprev) = n;
-}
-
-static inline void hlist_add_after(struct hlist_node *n,
-                                        struct hlist_node *next)
-{
-        next->next = n->next;
-        n->next = next;
-        next->pprev = &n->next;
-
-        if(next->next)
-                next->next->pprev  = &next->next;
-}
-
-#define hlist_entry(ptr, type, member) container_of(ptr,type,member)
-
-#define hlist_for_each(pos, head) \
-        for (pos = (head)->first; pos; \
-             pos = pos->next)
-
-#define hlist_for_each_safe(pos, n, head) \
-        for (pos = (head)->first; pos && ({ n = pos->next; 1; }); \
-             pos = n)
-
-/**
- * hlist_for_each_entry - iterate over list of given type
- * @tpos:       the type * to use as a loop cursor.
- * @pos:        the &struct hlist_node to use as a loop cursor.
- * @head:       the head for your list.
- * @member:     the name of the hlist_node within the struct.
- */
-#define hlist_for_each_entry(tpos, pos, head, member)                    \
-        for (pos = (head)->first;                                        \
-             pos &&                      \
-                ({ tpos = hlist_entry(pos, typeof(*tpos), member); 1;}); \
-             pos = pos->next)
-
-/**
- * hlist_for_each_entry_continue - iterate over a hlist continuing after current point
- * @tpos:       the type * to use as a loop cursor.
- * @pos:        the &struct hlist_node to use as a loop cursor.
- * @member:     the name of the hlist_node within the struct.
- */
-#define hlist_for_each_entry_continue(tpos, pos, member)                 \
-        for (pos = (pos)->next;                                          \
-             pos &&                      \
-                ({ tpos = hlist_entry(pos, typeof(*tpos), member); 1;}); \
-             pos = pos->next)
-
-/**
- * hlist_for_each_entry_from - iterate over a hlist continuing from current point
- * @tpos:       the type * to use as a loop cursor.
- * @pos:        the &struct hlist_node to use as a loop cursor.
- * @member:     the name of the hlist_node within the struct.
- */
-#define hlist_for_each_entry_from(tpos, pos, member)                     \
-        for (; pos &&                    \
-                ({ tpos = hlist_entry(pos, typeof(*tpos), member); 1;}); \
-             pos = pos->next)
-
-/**
- * hlist_for_each_entry_safe - iterate over list of given type safe against removal of list entry
- * @tpos:       the type * to use as a loop cursor.
- * @pos:        the &struct hlist_node to use as a loop cursor.
- * @n:          another &struct hlist_node to use as temporary storage
- * @head:       the head for your list.
- * @member:     the name of the hlist_node within the struct.
- */
-#define hlist_for_each_entry_safe(tpos, pos, n, head, member)            \
-        for (pos = (head)->first;                                        \
-             pos && ({ n = pos->next; 1; }) &&                           \
-                ({ tpos = hlist_entry(pos, typeof(*tpos), member); 1;}); \
-             pos = n)
-
-#endif
diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c
new file mode 100644
index 00000000000..3a7eb6ec0ee
--- /dev/null
+++ b/tools/perf/util/map.c
@@ -0,0 +1,682 @@
+#include "symbol.h"
+#include <errno.h>
+#include <limits.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdio.h>
+#include <unistd.h>
+#include "map.h"
+
+const char *map_type__name[MAP__NR_TYPES] = {
+        [MAP__FUNCTION] = "Functions",
+        [MAP__VARIABLE] = "Variables",
+};
+
+static inline int is_anon_memory(const char *filename)
+{
+        return strcmp(filename, "//anon") == 0;
+}
+
+void map__init(struct map *self, enum map_type type,
+               u64 start, u64 end, u64 pgoff, struct dso *dso)
+{
+        self->type     = type;
+        self->start    = start;
+        self->end      = end;
+        self->pgoff    = pgoff;
+        self->dso      = dso;
+        self->map_ip   = map__map_ip;
+        self->unmap_ip = map__unmap_ip;
+        RB_CLEAR_NODE(&self->rb_node);
+        self->groups   = NULL;
+        self->referenced = false;
+}
+
+struct map *map__new(struct list_head *dsos__list, u64 start, u64 len,
+                     u64 pgoff, u32 pid, char *filename,
+                     enum map_type type)
+{
+        struct map *self = malloc(sizeof(*self));
+
+        if (self != NULL) {
+                char newfilename[PATH_MAX];
+                struct dso *dso;
+                int anon;
+
+                anon = is_anon_memory(filename);
+
+                if (anon) {
+                        snprintf(newfilename, sizeof(newfilename), "/tmp/perf-%d.map", pid);
+                        filename = newfilename;
+                }
+
+                dso = __dsos__findnew(dsos__list, filename);
+                if (dso == NULL)
+                        goto out_delete;
+
+                map__init(self, type, start, start + len, pgoff, dso);
+
+                if (anon) {
+set_identity:
+                        self->map_ip = self->unmap_ip = identity__map_ip;
+                } else if (strcmp(filename, "[vdso]") == 0) {
+                        dso__set_loaded(dso, self->type);
+                        goto set_identity;
+                }
+        }
+        return self;
+out_delete:
+        free(self);
+        return NULL;
+}
+
+void map__delete(struct map *self)
+{
+        free(self);
+}
+
+void map__fixup_start(struct map *self)
+{
+        struct rb_root *symbols = &self->dso->symbols[self->type];
+        struct rb_node *nd = rb_first(symbols);
+        if (nd != NULL) {
+                struct symbol *sym = rb_entry(nd, struct symbol, rb_node);
+                self->start = sym->start;
+        }
+}
+
+void map__fixup_end(struct map *self)
+{
+        struct rb_root *symbols = &self->dso->symbols[self->type];
+        struct rb_node *nd = rb_last(symbols);
+        if (nd != NULL) {
+                struct symbol *sym = rb_entry(nd, struct symbol, rb_node);
+                self->end = sym->end;
+        }
+}
+
+#define DSO__DELETED "(deleted)"
+
+int map__load(struct map *self, symbol_filter_t filter)
+{
+        const char *name = self->dso->long_name;
+        int nr;
+
+        if (dso__loaded(self->dso, self->type))
+                return 0;
+
+        nr = dso__load(self->dso, self, filter);
+        if (nr < 0) {
+                if (self->dso->has_build_id) {
+                        char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+                        build_id__sprintf(self->dso->build_id,
+                                          sizeof(self->dso->build_id),
+                                          sbuild_id);
+                        pr_warning("%s with build id %s not found",
+                                   name, sbuild_id);
+                } else
+                        pr_warning("Failed to open %s", name);
+
+                pr_warning(", continuing without symbols\n");
+                return -1;
+        } else if (nr == 0) {
+                const size_t len = strlen(name);
+                const size_t real_len = len - sizeof(DSO__DELETED);
+
+                if (len > sizeof(DSO__DELETED) &&
+                    strcmp(name + real_len + 1, DSO__DELETED) == 0) {
+                        pr_warning("%.*s was updated, restart the long "
+                                   "running apps that use it!\n",
+                                   (int)real_len, name);
+                } else {
+                        pr_warning("no symbols found in %s, maybe install "
+                                   "a debug package?\n", name);
+                }
+
+                return -1;
+        }
+        /*
+         * Only applies to the kernel, as its symtabs aren't relative like the
+         * module ones.
+         */
+        if (self->dso->kernel)
+                map__reloc_vmlinux(self);
+
+        return 0;
+}
+
+struct symbol *map__find_symbol(struct map *self, u64 addr,
+                                symbol_filter_t filter)
+{
+        if (map__load(self, filter) < 0)
+                return NULL;
+
+        return dso__find_symbol(self->dso, self->type, addr);
+}
+
+struct symbol *map__find_symbol_by_name(struct map *self, const char *name,
+                                        symbol_filter_t filter)
+{
+        if (map__load(self, filter) < 0)
+                return NULL;
+
+        if (!dso__sorted_by_name(self->dso, self->type))
+                dso__sort_by_name(self->dso, self->type);
+
+        return dso__find_symbol_by_name(self->dso, self->type, name);
+}
+
+struct map *map__clone(struct map *self)
+{
+        struct map *map = malloc(sizeof(*self));
+
+        if (!map)
+                return NULL;
+
+        memcpy(map, self, sizeof(*self));
+
+        return map;
+}
+
+int map__overlap(struct map *l, struct map *r)
+{
+        if (l->start > r->start) {
+                struct map *t = l;
+                l = r;
+                r = t;
+        }
+
+        if (l->end > r->start)
+                return 1;
+
+        return 0;
+}
+
+size_t map__fprintf(struct map *self, FILE *fp)
+{
+        return fprintf(fp, " %Lx-%Lx %Lx %s\n",
+                       self->start, self->end, self->pgoff, self->dso->name);
+}
+
+/*
+ * objdump wants/reports absolute IPs for ET_EXEC, and RIPs for ET_DYN.
+ * map->dso->adjust_symbols==1 for ET_EXEC-like cases.
+ */
+u64 map__rip_2objdump(struct map *map, u64 rip)
+{
+        u64 addr = map->dso->adjust_symbols ?
+                        map->unmap_ip(map, rip) :       /* RIP -> IP */
+                        rip;
+        return addr;
+}
+
+u64 map__objdump_2ip(struct map *map, u64 addr)
+{
+        u64 ip = map->dso->adjust_symbols ?
+                        addr :
+                        map->unmap_ip(map, addr);       /* RIP -> IP */
+        return ip;
+}
+
+void map_groups__init(struct map_groups *self)
+{
+        int i;
+        for (i = 0; i < MAP__NR_TYPES; ++i) {
+                self->maps[i] = RB_ROOT;
+                INIT_LIST_HEAD(&self->removed_maps[i]);
+        }
+        self->machine = NULL;
+}
+
+static void maps__delete(struct rb_root *self)
+{
+        struct rb_node *next = rb_first(self);
+
+        while (next) {
+                struct map *pos = rb_entry(next, struct map, rb_node);
+
+                next = rb_next(&pos->rb_node);
+                rb_erase(&pos->rb_node, self);
+                map__delete(pos);
+        }
+}
+
+static void maps__delete_removed(struct list_head *self)
+{
+        struct map *pos, *n;
+
+        list_for_each_entry_safe(pos, n, self, node) {
+                list_del(&pos->node);
+                map__delete(pos);
+        }
+}
+
+void map_groups__exit(struct map_groups *self)
+{
+        int i;
+
+        for (i = 0; i < MAP__NR_TYPES; ++i) {
+                maps__delete(&self->maps[i]);
+                maps__delete_removed(&self->removed_maps[i]);
+        }
+}
+
+void map_groups__flush(struct map_groups *self)
+{
+        int type;
+
+        for (type = 0; type < MAP__NR_TYPES; type++) {
+                struct rb_root *root = &self->maps[type];
+                struct rb_node *next = rb_first(root);
+
+                while (next) {
+                        struct map *pos = rb_entry(next, struct map, rb_node);
+                        next = rb_next(&pos->rb_node);
+                        rb_erase(&pos->rb_node, root);
+                        /*
+                         * We may have references to this map, for
+                         * instance in some hist_entry instances, so
+                         * just move them to a separate list.
+                         */
+                        list_add_tail(&pos->node, &self->removed_maps[pos->type]);
+                }
+        }
+}
+
+struct symbol *map_groups__find_symbol(struct map_groups *self,
+                                       enum map_type type, u64 addr,
+                                       struct map **mapp,
+                                       symbol_filter_t filter)
+{
+        struct map *map = map_groups__find(self, type, addr);
+
+        if (map != NULL) {
+                if (mapp != NULL)
+                        *mapp = map;
+                return map__find_symbol(map, map->map_ip(map, addr), filter);
+        }
+
+        return NULL;
+}
+
+struct symbol *map_groups__find_symbol_by_name(struct map_groups *self,
+                                               enum map_type type,
+                                               const char *name,
+                                               struct map **mapp,
+                                               symbol_filter_t filter)
+{
+        struct rb_node *nd;
+
+        for (nd = rb_first(&self->maps[type]); nd; nd = rb_next(nd)) {
+                struct map *pos = rb_entry(nd, struct map, rb_node);
+                struct symbol *sym = map__find_symbol_by_name(pos, name, filter);
+
+                if (sym == NULL)
+                        continue;
+                if (mapp != NULL)
+                        *mapp = pos;
+                return sym;
+        }
+
+        return NULL;
+}
+
+size_t __map_groups__fprintf_maps(struct map_groups *self,
+                                  enum map_type type, int verbose, FILE *fp)
+{
+        size_t printed = fprintf(fp, "%s:\n", map_type__name[type]);
+        struct rb_node *nd;
+
+        for (nd = rb_first(&self->maps[type]); nd; nd = rb_next(nd)) {
+                struct map *pos = rb_entry(nd, struct map, rb_node);
+                printed += fprintf(fp, "Map:");
+                printed += map__fprintf(pos, fp);
+                if (verbose > 2) {
+                        printed += dso__fprintf(pos->dso, type, fp);
+                        printed += fprintf(fp, "--\n");
+                }
+        }
+
+        return printed;
+}
+
+size_t map_groups__fprintf_maps(struct map_groups *self, int verbose, FILE *fp)
+{
+        size_t printed = 0, i;
+        for (i = 0; i < MAP__NR_TYPES; ++i)
+                printed += __map_groups__fprintf_maps(self, i, verbose, fp);
+        return printed;
+}
+
+static size_t __map_groups__fprintf_removed_maps(struct map_groups *self,
+                                                 enum map_type type,
+                                                 int verbose, FILE *fp)
+{
+        struct map *pos;
+        size_t printed = 0;
+
+        list_for_each_entry(pos, &self->removed_maps[type], node) {
+                printed += fprintf(fp, "Map:");
+                printed += map__fprintf(pos, fp);
+                if (verbose > 1) {
+                        printed += dso__fprintf(pos->dso, type, fp);
+                        printed += fprintf(fp, "--\n");
+                }
+        }
+        return printed;
+}
+
+static size_t map_groups__fprintf_removed_maps(struct map_groups *self,
+                                               int verbose, FILE *fp)
+{
+        size_t printed = 0, i;
+        for (i = 0; i < MAP__NR_TYPES; ++i)
+                printed += __map_groups__fprintf_removed_maps(self, i, verbose, fp);
+        return printed;
+}
+
+size_t map_groups__fprintf(struct map_groups *self, int verbose, FILE *fp)
+{
+        size_t printed = map_groups__fprintf_maps(self, verbose, fp);
+        printed += fprintf(fp, "Removed maps:\n");
+        return printed + map_groups__fprintf_removed_maps(self, verbose, fp);
+}
+
+int map_groups__fixup_overlappings(struct map_groups *self, struct map *map,
+                                   int verbose, FILE *fp)
+{
+        struct rb_root *root = &self->maps[map->type];
+        struct rb_node *next = rb_first(root);
+        int err = 0;
+
+        while (next) {
+                struct map *pos = rb_entry(next, struct map, rb_node);
+                next = rb_next(&pos->rb_node);
+
+                if (!map__overlap(pos, map))
+                        continue;
+
+                if (verbose >= 2) {
+                        fputs("overlapping maps:\n", fp);
+                        map__fprintf(map, fp);
+                        map__fprintf(pos, fp);
+                }
+
+                rb_erase(&pos->rb_node, root);
+                /*
+                 * Now check if we need to create new maps for areas not
+                 * overlapped by the new map:
+                 */
+                if (map->start > pos->start) {
+                        struct map *before = map__clone(pos);
+
+                        if (before == NULL) {
+                                err = -ENOMEM;
+                                goto move_map;
+                        }
+
+                        before->end = map->start - 1;
+                        map_groups__insert(self, before);
+                        if (verbose >= 2)
+                                map__fprintf(before, fp);
+                }
+
+                if (map->end < pos->end) {
+                        struct map *after = map__clone(pos);
+
+                        if (after == NULL) {
+                                err = -ENOMEM;
+                                goto move_map;
+                        }
+
+                        after->start = map->end + 1;
+                        map_groups__insert(self, after);
+                        if (verbose >= 2)
+                                map__fprintf(after, fp);
+                }
+move_map:
+                /*
+                 * If we have references, just move them to a separate list.
+                 */
+                if (pos->referenced)
+                        list_add_tail(&pos->node, &self->removed_maps[map->type]);
+                else
+                        map__delete(pos);
+
+                if (err)
+                        return err;
+        }
+
+        return 0;
+}
+
+/*
+ * XXX This should not really _copy_ te maps, but refcount them.
+ */
+int map_groups__clone(struct map_groups *self,
+                      struct map_groups *parent, enum map_type type)
+{
+        struct rb_node *nd;
+        for (nd = rb_first(&parent->maps[type]); nd; nd = rb_next(nd)) {
+                struct map *map = rb_entry(nd, struct map, rb_node);
+                struct map *new = map__clone(map);
+                if (new == NULL)
+                        return -ENOMEM;
+                map_groups__insert(self, new);
+        }
+        return 0;
+}
+
+static u64 map__reloc_map_ip(struct map *map, u64 ip)
+{
+        return ip + (s64)map->pgoff;
+}
+
+static u64 map__reloc_unmap_ip(struct map *map, u64 ip)
+{
+        return ip - (s64)map->pgoff;
+}
+
+void map__reloc_vmlinux(struct map *self)
+{
+        struct kmap *kmap = map__kmap(self);
+        s64 reloc;
+
+        if (!kmap->ref_reloc_sym || !kmap->ref_reloc_sym->unrelocated_addr)
+                return;
+
+        reloc = (kmap->ref_reloc_sym->unrelocated_addr -
+                 kmap->ref_reloc_sym->addr);
+
+        if (!reloc)
+                return;
+
+        self->map_ip   = map__reloc_map_ip;
+        self->unmap_ip = map__reloc_unmap_ip;
+        self->pgoff    = reloc;
+}
+
+void maps__insert(struct rb_root *maps, struct map *map)
+{
+        struct rb_node **p = &maps->rb_node;
+        struct rb_node *parent = NULL;
+        const u64 ip = map->start;
+        struct map *m;
+
+        while (*p != NULL) {
+                parent = *p;
+                m = rb_entry(parent, struct map, rb_node);
+                if (ip < m->start)
+                        p = &(*p)->rb_left;
+                else
+                        p = &(*p)->rb_right;
+        }
+
+        rb_link_node(&map->rb_node, parent, p);
+        rb_insert_color(&map->rb_node, maps);
+}
+
+void maps__remove(struct rb_root *self, struct map *map)
+{
+        rb_erase(&map->rb_node, self);
+}
+
+struct map *maps__find(struct rb_root *maps, u64 ip)
+{
+        struct rb_node **p = &maps->rb_node;
+        struct rb_node *parent = NULL;
+        struct map *m;
+
+        while (*p != NULL) {
+                parent = *p;
+                m = rb_entry(parent, struct map, rb_node);
+                if (ip < m->start)
+                        p = &(*p)->rb_left;
+                else if (ip > m->end)
+                        p = &(*p)->rb_right;
+                else
+                        return m;
+        }
+
+        return NULL;
+}
+
+int machine__init(struct machine *self, const char *root_dir, pid_t pid)
+{
+        map_groups__init(&self->kmaps);
+        RB_CLEAR_NODE(&self->rb_node);
+        INIT_LIST_HEAD(&self->user_dsos);
+        INIT_LIST_HEAD(&self->kernel_dsos);
+
+        self->kmaps.machine = self;
+        self->pid           = pid;
+        self->root_dir      = strdup(root_dir);
+        return self->root_dir == NULL ? -ENOMEM : 0;
+}
+
+static void dsos__delete(struct list_head *self)
+{
+        struct dso *pos, *n;
+
+        list_for_each_entry_safe(pos, n, self, node) {
+                list_del(&pos->node);
+                dso__delete(pos);
+        }
+}
+
+void machine__exit(struct machine *self)
+{
+        map_groups__exit(&self->kmaps);
+        dsos__delete(&self->user_dsos);
+        dsos__delete(&self->kernel_dsos);
+        free(self->root_dir);
+        self->root_dir = NULL;
+}
+
+void machine__delete(struct machine *self)
+{
+        machine__exit(self);
+        free(self);
+}
+
+struct machine *machines__add(struct rb_root *self, pid_t pid,
+                              const char *root_dir)
+{
+        struct rb_node **p = &self->rb_node;
+        struct rb_node *parent = NULL;
+        struct machine *pos, *machine = malloc(sizeof(*machine));
+
+        if (!machine)
+                return NULL;
+
+        if (machine__init(machine, root_dir, pid) != 0) {
+                free(machine);
+                return NULL;
+        }
+
+        while (*p != NULL) {
+                parent = *p;
+                pos = rb_entry(parent, struct machine, rb_node);
+                if (pid < pos->pid)
+                        p = &(*p)->rb_left;
+                else
+                        p = &(*p)->rb_right;
+        }
+
+        rb_link_node(&machine->rb_node, parent, p);
+        rb_insert_color(&machine->rb_node, self);
+
+        return machine;
+}
+
+struct machine *machines__find(struct rb_root *self, pid_t pid)
+{
+        struct rb_node **p = &self->rb_node;
+        struct rb_node *parent = NULL;
+        struct machine *machine;
+        struct machine *default_machine = NULL;
+
+        while (*p != NULL) {
+                parent = *p;
+                machine = rb_entry(parent, struct machine, rb_node);
+                if (pid < machine->pid)
+                        p = &(*p)->rb_left;
+                else if (pid > machine->pid)
+                        p = &(*p)->rb_right;
+                else
+                        return machine;
+                if (!machine->pid)
+                        default_machine = machine;
+        }
+
+        return default_machine;
+}
+
+struct machine *machines__findnew(struct rb_root *self, pid_t pid)
+{
+        char path[PATH_MAX];
+        const char *root_dir;
+        struct machine *machine = machines__find(self, pid);
+
+        if (!machine || machine->pid != pid) {
+                if (pid == HOST_KERNEL_ID || pid == DEFAULT_GUEST_KERNEL_ID)
+                        root_dir = "";
+                else {
+                        if (!symbol_conf.guestmount)
+                                goto out;
+                        sprintf(path, "%s/%d", symbol_conf.guestmount, pid);
+                        if (access(path, R_OK)) {
+                                pr_err("Can't access file %s\n", path);
+                                goto out;
+                        }
+                        root_dir = path;
+                }
+                machine = machines__add(self, pid, root_dir);
+        }
+
+out:
+        return machine;
+}
+
+void machines__process(struct rb_root *self, machine__process_t process, void *data)
+{
+        struct rb_node *nd;
+
+        for (nd = rb_first(self); nd; nd = rb_next(nd)) {
+                struct machine *pos = rb_entry(nd, struct machine, rb_node);
+                process(pos, data);
+        }
+}
+
+char *machine__mmap_name(struct machine *self, char *bf, size_t size)
+{
+        if (machine__is_host(self))
+                snprintf(bf, size, "[%s]", "kernel.kallsyms");
+        else if (machine__is_default_guest(self))
+                snprintf(bf, size, "[%s]", "guest.kernel.kallsyms");
+        else
+                snprintf(bf, size, "[%s.%d]", "guest.kernel.kallsyms", self->pid);
+
+        return bf;
+}
diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h
new file mode 100644
index 00000000000..b397c038372
--- /dev/null
+++ b/tools/perf/util/map.h
@@ -0,0 +1,237 @@
+#ifndef __PERF_MAP_H
+#define __PERF_MAP_H
+
+#include <linux/compiler.h>
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include <stdio.h>
+#include <stdbool.h>
+#include "types.h"
+
+enum map_type {
+        MAP__FUNCTION = 0,
+        MAP__VARIABLE,
+};
+
+#define MAP__NR_TYPES (MAP__VARIABLE + 1)
+
+extern const char *map_type__name[MAP__NR_TYPES];
+
+struct dso;
+struct ref_reloc_sym;
+struct map_groups;
+struct machine;
+
+struct map {
+        union {
+                struct rb_node  rb_node;
+                struct list_head node;
+        };
+        u64                     start;
+        u64                     end;
+        u8 /* enum map_type */  type;
+        bool                    referenced;
+        u32                     priv;
+        u64                     pgoff;
+
+        /* ip -> dso rip */
+        u64                     (*map_ip)(struct map *, u64);
+        /* dso rip -> ip */
+        u64                     (*unmap_ip)(struct map *, u64);
+
+        struct dso              *dso;
+        struct map_groups       *groups;
+};
+
+struct kmap {
+        struct ref_reloc_sym    *ref_reloc_sym;
+        struct map_groups       *kmaps;
+};
+
+struct map_groups {
+        struct rb_root   maps[MAP__NR_TYPES];
+        struct list_head removed_maps[MAP__NR_TYPES];
+        struct machine   *machine;
+};
+
+/* Native host kernel uses -1 as pid index in machine */
+#define HOST_KERNEL_ID                  (-1)
+#define DEFAULT_GUEST_KERNEL_ID         (0)
+
+struct machine {
+        struct rb_node    rb_node;
+        pid_t             pid;
+        char              *root_dir;
+        struct list_head  user_dsos;
+        struct list_head  kernel_dsos;
+        struct map_groups kmaps;
+        struct map        *vmlinux_maps[MAP__NR_TYPES];
+};
+
+static inline
+struct map *machine__kernel_map(struct machine *self, enum map_type type)
+{
+        return self->vmlinux_maps[type];
+}
+
+static inline struct kmap *map__kmap(struct map *self)
+{
+        return (struct kmap *)(self + 1);
+}
+
+static inline u64 map__map_ip(struct map *map, u64 ip)
+{
+        return ip - map->start + map->pgoff;
+}
+
+static inline u64 map__unmap_ip(struct map *map, u64 ip)
+{
+        return ip + map->start - map->pgoff;
+}
+
+static inline u64 identity__map_ip(struct map *map __used, u64 ip)
+{
+        return ip;
+}
+
+
+/* rip/ip <-> addr suitable for passing to `objdump --start-address=` */
+u64 map__rip_2objdump(struct map *map, u64 rip);
+u64 map__objdump_2ip(struct map *map, u64 addr);
+
+struct symbol;
+
+typedef int (*symbol_filter_t)(struct map *map, struct symbol *sym);
+
+void map__init(struct map *self, enum map_type type,
+               u64 start, u64 end, u64 pgoff, struct dso *dso);
+struct map *map__new(struct list_head *dsos__list, u64 start, u64 len,
+                     u64 pgoff, u32 pid, char *filename,
+                     enum map_type type);
+void map__delete(struct map *self);
+struct map *map__clone(struct map *self);
+int map__overlap(struct map *l, struct map *r);
+size_t map__fprintf(struct map *self, FILE *fp);
+
+int map__load(struct map *self, symbol_filter_t filter);
+struct symbol *map__find_symbol(struct map *self,
+                                u64 addr, symbol_filter_t filter);
+struct symbol *map__find_symbol_by_name(struct map *self, const char *name,
+                                        symbol_filter_t filter);
+void map__fixup_start(struct map *self);
+void map__fixup_end(struct map *self);
+
+void map__reloc_vmlinux(struct map *self);
+
+size_t __map_groups__fprintf_maps(struct map_groups *self,
+                                  enum map_type type, int verbose, FILE *fp);
+void maps__insert(struct rb_root *maps, struct map *map);
+void maps__remove(struct rb_root *self, struct map *map);
+struct map *maps__find(struct rb_root *maps, u64 addr);
+void map_groups__init(struct map_groups *self);
+void map_groups__exit(struct map_groups *self);
+int map_groups__clone(struct map_groups *self,
+                      struct map_groups *parent, enum map_type type);
+size_t map_groups__fprintf(struct map_groups *self, int verbose, FILE *fp);
+size_t map_groups__fprintf_maps(struct map_groups *self, int verbose, FILE *fp);
+
+typedef void (*machine__process_t)(struct machine *self, void *data);
+
+void machines__process(struct rb_root *self, machine__process_t process, void *data);
+struct machine *machines__add(struct rb_root *self, pid_t pid,
+                              const char *root_dir);
+struct machine *machines__find_host(struct rb_root *self);
+struct machine *machines__find(struct rb_root *self, pid_t pid);
+struct machine *machines__findnew(struct rb_root *self, pid_t pid);
+char *machine__mmap_name(struct machine *self, char *bf, size_t size);
+int machine__init(struct machine *self, const char *root_dir, pid_t pid);
+void machine__exit(struct machine *self);
+void machine__delete(struct machine *self);
+
+/*
+ * Default guest kernel is defined by parameter --guestkallsyms
+ * and --guestmodules
+ */
+static inline bool machine__is_default_guest(struct machine *self)
+{
+        return self ? self->pid == DEFAULT_GUEST_KERNEL_ID : false;
+}
+
+static inline bool machine__is_host(struct machine *self)
+{
+        return self ? self->pid == HOST_KERNEL_ID : false;
+}
+
+static inline void map_groups__insert(struct map_groups *self, struct map *map)
+{
+        maps__insert(&self->maps[map->type], map);
+        map->groups = self;
+}
+
+static inline void map_groups__remove(struct map_groups *self, struct map *map)
+{
+        maps__remove(&self->maps[map->type], map);
+}
+
+static inline struct map *map_groups__find(struct map_groups *self,
+                                           enum map_type type, u64 addr)
+{
+        return maps__find(&self->maps[type], addr);
+}
+
+struct symbol *map_groups__find_symbol(struct map_groups *self,
+                                       enum map_type type, u64 addr,
+                                       struct map **mapp,
+                                       symbol_filter_t filter);
+
+struct symbol *map_groups__find_symbol_by_name(struct map_groups *self,
+                                               enum map_type type,
+                                               const char *name,
+                                               struct map **mapp,
+                                               symbol_filter_t filter);
+
+static inline
+struct symbol *machine__find_kernel_symbol(struct machine *self,
+                                           enum map_type type, u64 addr,
+                                           struct map **mapp,
+                                           symbol_filter_t filter)
+{
+        return map_groups__find_symbol(&self->kmaps, type, addr, mapp, filter);
+}
+
+static inline
+struct symbol *machine__find_kernel_function(struct machine *self, u64 addr,
+                                             struct map **mapp,
+                                             symbol_filter_t filter)
+{
+        return machine__find_kernel_symbol(self, MAP__FUNCTION, addr, mapp, filter);
+}
+
+static inline
+struct symbol *map_groups__find_function_by_name(struct map_groups *self,
+                                                 const char *name, struct map **mapp,
+                                                 symbol_filter_t filter)
+{
+        return map_groups__find_symbol_by_name(self, MAP__FUNCTION, name, mapp, filter);
+}
+
+static inline
+struct symbol *machine__find_kernel_function_by_name(struct machine *self,
+                                                     const char *name,
+                                                     struct map **mapp,
+                                                     symbol_filter_t filter)
+{
+        return map_groups__find_function_by_name(&self->kmaps, name, mapp,
+                                                 filter);
+}
+
+int map_groups__fixup_overlappings(struct map_groups *self, struct map *map,
+                                   int verbose, FILE *fp);
+
+struct map *map_groups__find_by_name(struct map_groups *self,
+                                     enum map_type type, const char *name);
+struct map *machine__new_module(struct machine *self, u64 start, const char *filename);
+
+void map_groups__flush(struct map_groups *self);
+
+#endif /* __PERF_MAP_H */
diff --git a/tools/perf/util/pager.c b/tools/perf/util/pager.c
index a28bccae545..1915de20dca 100644
--- a/tools/perf/util/pager.c
+++ b/tools/perf/util/pager.c
@@ -9,7 +9,6 @@
 
 static int spawned_pager;
 
-#ifndef __MINGW32__
 static void pager_preexec(void)
 {
         /*
@@ -24,7 +23,6 @@ static void pager_preexec(void)
 
         setenv("LESS", "FRSX", 0);
 }
-#endif
 
 static const char *pager_argv[] = { "sh", "-c", NULL, NULL };
 static struct child_process pager_process;
@@ -70,9 +68,8 @@ void setup_pager(void)
         pager_argv[2] = pager;
         pager_process.argv = pager_argv;
         pager_process.in = -1;
-#ifndef __MINGW32__
         pager_process.preexec_cb = pager_preexec;
-#endif
+
         if (start_command(&pager_process))
                 return;
 
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 35d04da38d6..5cb6f4bde90 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -1,58 +1,67 @@
-
-#include "../perf.h"
+#include "../../../include/linux/hw_breakpoint.h"
 #include "util.h"
+#include "../perf.h"
+#include "evsel.h"
 #include "parse-options.h"
 #include "parse-events.h"
 #include "exec_cmd.h"
 #include "string.h"
+#include "symbol.h"
+#include "cache.h"
+#include "header.h"
+#include "debugfs.h"
 
-extern char *strcasestr(const char *haystack, const char *needle);
-
-int                                     nr_counters;
+int                             nr_counters;
 
-struct perf_counter_attr                attrs[MAX_COUNTERS];
+LIST_HEAD(evsel_list);
 
 struct event_symbol {
-        u8      type;
-        u64     config;
-        char    *symbol;
+        u8              type;
+        u64             config;
+        const char      *symbol;
+        const char      *alias;
 };
 
-#define C(x, y) .type = PERF_TYPE_##x, .config = PERF_COUNT_##y
-#define CR(x, y) .type = PERF_TYPE_##x, .config = y
+enum event_result {
+        EVT_FAILED,
+        EVT_HANDLED,
+        EVT_HANDLED_ALL
+};
+
+char debugfs_path[MAXPATHLEN];
+
+#define CHW(x) .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_##x
+#define CSW(x) .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_##x
 
 static struct event_symbol event_symbols[] = {
-  { C(HARDWARE, HW_CPU_CYCLES),         "cpu-cycles",           },
-  { C(HARDWARE, HW_CPU_CYCLES),         "cycles",               },
-  { C(HARDWARE, HW_INSTRUCTIONS),       "instructions",         },
-  { C(HARDWARE, HW_CACHE_REFERENCES),   "cache-references",     },
-  { C(HARDWARE, HW_CACHE_MISSES),       "cache-misses",         },
-  { C(HARDWARE, HW_BRANCH_INSTRUCTIONS),"branch-instructions",  },
-  { C(HARDWARE, HW_BRANCH_INSTRUCTIONS),"branches",             },
-  { C(HARDWARE, HW_BRANCH_MISSES),      "branch-misses",        },
-  { C(HARDWARE, HW_BUS_CYCLES),         "bus-cycles",           },
-
-  { C(SOFTWARE, SW_CPU_CLOCK),          "cpu-clock",            },
-  { C(SOFTWARE, SW_TASK_CLOCK),         "task-clock",           },
-  { C(SOFTWARE, SW_PAGE_FAULTS),        "page-faults",          },
-  { C(SOFTWARE, SW_PAGE_FAULTS),        "faults",               },
-  { C(SOFTWARE, SW_PAGE_FAULTS_MIN),    "minor-faults",         },
-  { C(SOFTWARE, SW_PAGE_FAULTS_MAJ),    "major-faults",         },
-  { C(SOFTWARE, SW_CONTEXT_SWITCHES),   "context-switches",     },
-  { C(SOFTWARE, SW_CONTEXT_SWITCHES),   "cs",                   },
-  { C(SOFTWARE, SW_CPU_MIGRATIONS),     "cpu-migrations",       },
-  { C(SOFTWARE, SW_CPU_MIGRATIONS),     "migrations",           },
+  { CHW(CPU_CYCLES),            "cpu-cycles",           "cycles"        },
+  { CHW(INSTRUCTIONS),          "instructions",         ""              },
+  { CHW(CACHE_REFERENCES),      "cache-references",     ""              },
+  { CHW(CACHE_MISSES),          "cache-misses",         ""              },
+  { CHW(BRANCH_INSTRUCTIONS),   "branch-instructions",  "branches"      },
+  { CHW(BRANCH_MISSES),         "branch-misses",        ""              },
+  { CHW(BUS_CYCLES),            "bus-cycles",           ""              },
+
+  { CSW(CPU_CLOCK),             "cpu-clock",            ""              },
+  { CSW(TASK_CLOCK),            "task-clock",           ""              },
+  { CSW(PAGE_FAULTS),           "page-faults",          "faults"        },
+  { CSW(PAGE_FAULTS_MIN),       "minor-faults",         ""              },
+  { CSW(PAGE_FAULTS_MAJ),       "major-faults",         ""              },
+  { CSW(CONTEXT_SWITCHES),      "context-switches",     "cs"            },
+  { CSW(CPU_MIGRATIONS),        "cpu-migrations",       "migrations"    },
+  { CSW(ALIGNMENT_FAULTS),      "alignment-faults",     ""              },
+  { CSW(EMULATION_FAULTS),      "emulation-faults",     ""              },
 };
 
-#define __PERF_COUNTER_FIELD(config, name) \
-        ((config & PERF_COUNTER_##name##_MASK) >> PERF_COUNTER_##name##_SHIFT)
+#define __PERF_EVENT_FIELD(config, name) \
+        ((config & PERF_EVENT_##name##_MASK) >> PERF_EVENT_##name##_SHIFT)
 
-#define PERF_COUNTER_RAW(config)        __PERF_COUNTER_FIELD(config, RAW)
-#define PERF_COUNTER_CONFIG(config)     __PERF_COUNTER_FIELD(config, CONFIG)
-#define PERF_COUNTER_TYPE(config)       __PERF_COUNTER_FIELD(config, TYPE)
-#define PERF_COUNTER_ID(config)         __PERF_COUNTER_FIELD(config, EVENT)
+#define PERF_EVENT_RAW(config)  __PERF_EVENT_FIELD(config, RAW)
+#define PERF_EVENT_CONFIG(config)       __PERF_EVENT_FIELD(config, CONFIG)
+#define PERF_EVENT_TYPE(config) __PERF_EVENT_FIELD(config, TYPE)
+#define PERF_EVENT_ID(config)           __PERF_EVENT_FIELD(config, EVENT)
 
-static char *hw_event_names[] = {
+static const char *hw_event_names[] = {
         "cycles",
         "instructions",
         "cache-references",
@@ -62,7 +71,7 @@ static char *hw_event_names[] = {
         "bus-cycles",
 };
 
-static char *sw_event_names[] = {
+static const char *sw_event_names[] = {
         "cpu-clock-msecs",
         "task-clock-msecs",
         "page-faults",
@@ -70,37 +79,206 @@ static char *sw_event_names[] = {
         "CPU-migrations",
         "minor-faults",
         "major-faults",
+        "alignment-faults",
+        "emulation-faults",
 };
 
 #define MAX_ALIASES 8
 
-static char *hw_cache [][MAX_ALIASES] = {
-        { "L1-data"             , "l1-d", "l1d"                                 },
-        { "L1-instruction"      , "l1-i", "l1i"                                 },
-        { "L2"                  , "l2"                                          },
-        { "Data-TLB"            , "dtlb", "d-tlb"                               },
-        { "Instruction-TLB"     , "itlb", "i-tlb"                               },
-        { "Branch"              , "bpu" , "btb", "bpc"                          },
+static const char *hw_cache[][MAX_ALIASES] = {
+ { "L1-dcache", "l1-d",         "l1d",          "L1-data",              },
+ { "L1-icache", "l1-i",         "l1i",          "L1-instruction",       },
+ { "LLC",       "L2"                                                    },
+ { "dTLB",      "d-tlb",        "Data-TLB",                             },
+ { "iTLB",      "i-tlb",        "Instruction-TLB",                      },
+ { "branch",    "branches",     "bpu",          "btb",          "bpc",  },
 };
 
-static char *hw_cache_op [][MAX_ALIASES] = {
-        { "Load"                , "read"                                        },
-        { "Store"               , "write"                                       },
-        { "Prefetch"            , "speculative-read", "speculative-load"        },
+static const char *hw_cache_op[][MAX_ALIASES] = {
+ { "load",      "loads",        "read",                                 },
+ { "store",     "stores",       "write",                                },
+ { "prefetch",  "prefetches",   "speculative-read", "speculative-load", },
 };
 
-static char *hw_cache_result [][MAX_ALIASES] = {
-        { "Reference"           , "ops", "access"                               },
-        { "Miss"                                                                },
+static const char *hw_cache_result[][MAX_ALIASES] = {
+ { "refs",      "Reference",    "ops",          "access",               },
+ { "misses",    "miss",                                                 },
 };
 
-char *event_name(int counter)
+#define C(x)            PERF_COUNT_HW_CACHE_##x
+#define CACHE_READ      (1 << C(OP_READ))
+#define CACHE_WRITE     (1 << C(OP_WRITE))
+#define CACHE_PREFETCH  (1 << C(OP_PREFETCH))
+#define COP(x)          (1 << x)
+
+/*
+ * cache operartion stat
+ * L1I : Read and prefetch only
+ * ITLB and BPU : Read-only
+ */
+static unsigned long hw_cache_stat[C(MAX)] = {
+ [C(L1D)]       = (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
+ [C(L1I)]       = (CACHE_READ | CACHE_PREFETCH),
+ [C(LL)]        = (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
+ [C(DTLB)]      = (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
+ [C(ITLB)]      = (CACHE_READ),
+ [C(BPU)]       = (CACHE_READ),
+};
+
+#define for_each_subsystem(sys_dir, sys_dirent, sys_next)              \
+        while (!readdir_r(sys_dir, &sys_dirent, &sys_next) && sys_next)        \
+        if (sys_dirent.d_type == DT_DIR &&                                     \
+           (strcmp(sys_dirent.d_name, ".")) &&                                 \
+           (strcmp(sys_dirent.d_name, "..")))
+
+static int tp_event_has_id(struct dirent *sys_dir, struct dirent *evt_dir)
+{
+        char evt_path[MAXPATHLEN];
+        int fd;
+
+        snprintf(evt_path, MAXPATHLEN, "%s/%s/%s/id", debugfs_path,
+                        sys_dir->d_name, evt_dir->d_name);
+        fd = open(evt_path, O_RDONLY);
+        if (fd < 0)
+                return -EINVAL;
+        close(fd);
+
+        return 0;
+}
+
+#define for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next)              \
+        while (!readdir_r(evt_dir, &evt_dirent, &evt_next) && evt_next)        \
+        if (evt_dirent.d_type == DT_DIR &&                                     \
+           (strcmp(evt_dirent.d_name, ".")) &&                                 \
+           (strcmp(evt_dirent.d_name, "..")) &&                                \
+           (!tp_event_has_id(&sys_dirent, &evt_dirent)))
+
+#define MAX_EVENT_LENGTH 512
+
+
+struct tracepoint_path *tracepoint_id_to_path(u64 config)
+{
+        struct tracepoint_path *path = NULL;
+        DIR *sys_dir, *evt_dir;
+        struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent;
+        char id_buf[4];
+        int fd;
+        u64 id;
+        char evt_path[MAXPATHLEN];
+        char dir_path[MAXPATHLEN];
+
+        if (debugfs_valid_mountpoint(debugfs_path))
+                return NULL;
+
+        sys_dir = opendir(debugfs_path);
+        if (!sys_dir)
+                return NULL;
+
+        for_each_subsystem(sys_dir, sys_dirent, sys_next) {
+
+                snprintf(dir_path, MAXPATHLEN, "%s/%s", debugfs_path,
+                         sys_dirent.d_name);
+                evt_dir = opendir(dir_path);
+                if (!evt_dir)
+                        continue;
+
+                for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) {
+
+                        snprintf(evt_path, MAXPATHLEN, "%s/%s/id", dir_path,
+                                 evt_dirent.d_name);
+                        fd = open(evt_path, O_RDONLY);
+                        if (fd < 0)
+                                continue;
+                        if (read(fd, id_buf, sizeof(id_buf)) < 0) {
+                                close(fd);
+                                continue;
+                        }
+                        close(fd);
+                        id = atoll(id_buf);
+                        if (id == config) {
+                                closedir(evt_dir);
+                                closedir(sys_dir);
+                                path = zalloc(sizeof(*path));
+                                path->system = malloc(MAX_EVENT_LENGTH);
+                                if (!path->system) {
+                                        free(path);
+                                        return NULL;
+                                }
+                                path->name = malloc(MAX_EVENT_LENGTH);
+                                if (!path->name) {
+                                        free(path->system);
+                                        free(path);
+                                        return NULL;
+                                }
+                                strncpy(path->system, sys_dirent.d_name,
+                                        MAX_EVENT_LENGTH);
+                                strncpy(path->name, evt_dirent.d_name,
+                                        MAX_EVENT_LENGTH);
+                                return path;
+                        }
+                }
+                closedir(evt_dir);
+        }
+
+        closedir(sys_dir);
+        return NULL;
+}
+
+#define TP_PATH_LEN (MAX_EVENT_LENGTH * 2 + 1)
+static const char *tracepoint_id_to_name(u64 config)
+{
+        static char buf[TP_PATH_LEN];
+        struct tracepoint_path *path;
+
+        path = tracepoint_id_to_path(config);
+        if (path) {
+                snprintf(buf, TP_PATH_LEN, "%s:%s", path->system, path->name);
+                free(path->name);
+                free(path->system);
+                free(path);
+        } else
+                snprintf(buf, TP_PATH_LEN, "%s:%s", "unknown", "unknown");
+
+        return buf;
+}
+
+static int is_cache_op_valid(u8 cache_type, u8 cache_op)
+{
+        if (hw_cache_stat[cache_type] & COP(cache_op))
+                return 1;       /* valid */
+        else
+                return 0;       /* invalid */
+}
+
+static char *event_cache_name(u8 cache_type, u8 cache_op, u8 cache_result)
+{
+        static char name[50];
+
+        if (cache_result) {
+                sprintf(name, "%s-%s-%s", hw_cache[cache_type][0],
+                        hw_cache_op[cache_op][0],
+                        hw_cache_result[cache_result][0]);
+        } else {
+                sprintf(name, "%s-%s", hw_cache[cache_type][0],
+                        hw_cache_op[cache_op][1]);
+        }
+
+        return name;
+}
+
+const char *event_name(struct perf_evsel *evsel)
+{
+        u64 config = evsel->attr.config;
+        int type = evsel->attr.type;
+
+        return __event_name(type, config);
+}
+
+const char *__event_name(int type, u64 config)
 {
-        u64 config = attrs[counter].config;
-        int type = attrs[counter].type;
         static char buf[32];
 
-        if (attrs[counter].type == PERF_TYPE_RAW) {
+        if (type == PERF_TYPE_RAW) {
                 sprintf(buf, "raw 0x%llx", config);
                 return buf;
         }
@@ -113,7 +291,6 @@ char *event_name(int counter)
 
         case PERF_TYPE_HW_CACHE: {
                 u8 cache_type, cache_op, cache_result;
-                static char name[100];
 
                 cache_type   = (config >>  0) & 0xff;
                 if (cache_type > PERF_COUNT_HW_CACHE_MAX)
@@ -127,12 +304,10 @@ char *event_name(int counter)
                 if (cache_result > PERF_COUNT_HW_CACHE_RESULT_MAX)
                         return "unknown-ext-hardware-cache-result";
 
-                sprintf(name, "%s-Cache-%s-%ses",
-                        hw_cache[cache_type][0],
-                        hw_cache_op[cache_op][0],
-                        hw_cache_result[cache_result][0]);
+                if (!is_cache_op_valid(cache_type, cache_op))
+                        return "invalid-cache";
 
-                return name;
+                return event_cache_name(cache_type, cache_op, cache_result);
         }
 
         case PERF_TYPE_SOFTWARE:
@@ -140,6 +315,9 @@ char *event_name(int counter)
                         return sw_event_names[config];
                 return "unknown-software";
 
+        case PERF_TYPE_TRACEPOINT:
+                return tracepoint_id_to_name(config);
+
         default:
                 break;
         }
@@ -147,43 +325,74 @@ char *event_name(int counter)
         return "unknown";
 }
 
-static int parse_aliases(const char *str, char *names[][MAX_ALIASES], int size)
+static int parse_aliases(const char **str, const char *names[][MAX_ALIASES], int size)
 {
         int i, j;
+        int n, longest = -1;
 
         for (i = 0; i < size; i++) {
-                for (j = 0; j < MAX_ALIASES; j++) {
-                        if (!names[i][j])
-                                break;
-                        if (strcasestr(str, names[i][j]))
-                                return i;
+                for (j = 0; j < MAX_ALIASES && names[i][j]; j++) {
+                        n = strlen(names[i][j]);
+                        if (n > longest && !strncasecmp(*str, names[i][j], n))
+                                longest = n;
+                }
+                if (longest > 0) {
+                        *str += longest;
+                        return i;
                 }
         }
 
         return -1;
 }
 
-static int parse_generic_hw_symbols(const char *str, struct perf_counter_attr *attr)
+static enum event_result
+parse_generic_hw_event(const char **str, struct perf_event_attr *attr)
 {
-        int cache_type = -1, cache_op = 0, cache_result = 0;
+        const char *s = *str;
+        int cache_type = -1, cache_op = -1, cache_result = -1;
 
-        cache_type = parse_aliases(str, hw_cache, PERF_COUNT_HW_CACHE_MAX);
+        cache_type = parse_aliases(&s, hw_cache, PERF_COUNT_HW_CACHE_MAX);
         /*
          * No fallback - if we cannot get a clear cache type
          * then bail out:
          */
         if (cache_type == -1)
-                return -EINVAL;
+                return EVT_FAILED;
+
+        while ((cache_op == -1 || cache_result == -1) && *s == '-') {
+                ++s;
+
+                if (cache_op == -1) {
+                        cache_op = parse_aliases(&s, hw_cache_op,
+                                                PERF_COUNT_HW_CACHE_OP_MAX);
+                        if (cache_op >= 0) {
+                                if (!is_cache_op_valid(cache_type, cache_op))
+                                        return 0;
+                                continue;
+                        }
+                }
+
+                if (cache_result == -1) {
+                        cache_result = parse_aliases(&s, hw_cache_result,
+                                                PERF_COUNT_HW_CACHE_RESULT_MAX);
+                        if (cache_result >= 0)
+                                continue;
+                }
+
+                /*
+                 * Can't parse this as a cache op or result, so back up
+                 * to the '-'.
+                 */
+                --s;
+                break;
+        }
 
-        cache_op = parse_aliases(str, hw_cache_op, PERF_COUNT_HW_CACHE_OP_MAX);
         /*
          * Fall back to reads:
          */
         if (cache_op == -1)
                 cache_op = PERF_COUNT_HW_CACHE_OP_READ;
 
-        cache_result = parse_aliases(str, hw_cache_result,
-                                        PERF_COUNT_HW_CACHE_RESULT_MAX);
         /*
          * Fall back to accesses:
          */
@@ -193,124 +402,641 @@ static int parse_generic_hw_symbols(const char *str, struct perf_counter_attr *a
         attr->config = cache_type | (cache_op << 8) | (cache_result << 16);
         attr->type = PERF_TYPE_HW_CACHE;
 
-        return 0;
+        *str = s;
+        return EVT_HANDLED;
 }
 
-/*
- * Each event can have multiple symbolic names.
- * Symbolic names are (almost) exactly matched.
- */
-static int parse_event_symbols(const char *str, struct perf_counter_attr *attr)
+static enum event_result
+parse_single_tracepoint_event(char *sys_name,
+                              const char *evt_name,
+                              unsigned int evt_length,
+                              struct perf_event_attr *attr,
+                              const char **strp)
 {
-        u64 config, id;
-        int type;
-        unsigned int i;
-        const char *sep, *pstr;
+        char evt_path[MAXPATHLEN];
+        char id_buf[4];
+        u64 id;
+        int fd;
 
-        if (str[0] == 'r' && hex2u64(str + 1, &config) > 0) {
-                attr->type = PERF_TYPE_RAW;
-                attr->config = config;
+        snprintf(evt_path, MAXPATHLEN, "%s/%s/%s/id", debugfs_path,
+                 sys_name, evt_name);
+
+        fd = open(evt_path, O_RDONLY);
+        if (fd < 0)
+                return EVT_FAILED;
+
+        if (read(fd, id_buf, sizeof(id_buf)) < 0) {
+                close(fd);
+                return EVT_FAILED;
+        }
+
+        close(fd);
+        id = atoll(id_buf);
+        attr->config = id;
+        attr->type = PERF_TYPE_TRACEPOINT;
+        *strp += strlen(sys_name) + evt_length + 1; /* + 1 for the ':' */
+
+        attr->sample_type |= PERF_SAMPLE_RAW;
+        attr->sample_type |= PERF_SAMPLE_TIME;
+        attr->sample_type |= PERF_SAMPLE_CPU;
+
+        attr->sample_period = 1;
+
+
+        return EVT_HANDLED;
+}
+
+/* sys + ':' + event + ':' + flags*/
+#define MAX_EVOPT_LEN   (MAX_EVENT_LENGTH * 2 + 2 + 128)
+static enum event_result
+parse_multiple_tracepoint_event(char *sys_name, const char *evt_exp,
+                                char *flags)
+{
+        char evt_path[MAXPATHLEN];
+        struct dirent *evt_ent;
+        DIR *evt_dir;
+
+        snprintf(evt_path, MAXPATHLEN, "%s/%s", debugfs_path, sys_name);
+        evt_dir = opendir(evt_path);
+
+        if (!evt_dir) {
+                perror("Can't open event dir");
+                return EVT_FAILED;
+        }
+
+        while ((evt_ent = readdir(evt_dir))) {
+                char event_opt[MAX_EVOPT_LEN + 1];
+                int len;
+
+                if (!strcmp(evt_ent->d_name, ".")
+                    || !strcmp(evt_ent->d_name, "..")
+                    || !strcmp(evt_ent->d_name, "enable")
+                    || !strcmp(evt_ent->d_name, "filter"))
+                        continue;
+
+                if (!strglobmatch(evt_ent->d_name, evt_exp))
+                        continue;
+
+                len = snprintf(event_opt, MAX_EVOPT_LEN, "%s:%s%s%s", sys_name,
+                               evt_ent->d_name, flags ? ":" : "",
+                               flags ?: "");
+                if (len < 0)
+                        return EVT_FAILED;
 
+                if (parse_events(NULL, event_opt, 0))
+                        return EVT_FAILED;
+        }
+
+        return EVT_HANDLED_ALL;
+}
+
+static int store_event_type(const char *orgname)
+{
+        char filename[PATH_MAX], *c;
+        FILE *file;
+        int id, n;
+
+        sprintf(filename, "%s/", debugfs_path);
+        strncat(filename, orgname, strlen(orgname));
+        strcat(filename, "/id");
+
+        c = strchr(filename, ':');
+        if (c)
+                *c = '/';
+
+        file = fopen(filename, "r");
+        if (!file)
                 return 0;
+        n = fscanf(file, "%i", &id);
+        fclose(file);
+        if (n < 1) {
+                pr_err("cannot store event ID\n");
+                return -EINVAL;
         }
+        return perf_header__push_event(id, orgname);
+}
 
-        pstr = str;
-        sep = strchr(pstr, ':');
-        if (sep) {
-                type = atoi(pstr);
-                pstr = sep + 1;
-                id = atoi(pstr);
-                sep = strchr(pstr, ':');
-                if (sep) {
-                        pstr = sep + 1;
-                        if (strchr(pstr, 'k'))
-                                attr->exclude_user = 1;
-                        if (strchr(pstr, 'u'))
-                                attr->exclude_kernel = 1;
-                }
-                attr->type = type;
-                attr->config = id;
+static enum event_result parse_tracepoint_event(const char **strp,
+                                    struct perf_event_attr *attr)
+{
+        const char *evt_name;
+        char *flags = NULL, *comma_loc;
+        char sys_name[MAX_EVENT_LENGTH];
+        unsigned int sys_length, evt_length;
 
+        if (debugfs_valid_mountpoint(debugfs_path))
                 return 0;
+
+        evt_name = strchr(*strp, ':');
+        if (!evt_name)
+                return EVT_FAILED;
+
+        sys_length = evt_name - *strp;
+        if (sys_length >= MAX_EVENT_LENGTH)
+                return 0;
+
+        strncpy(sys_name, *strp, sys_length);
+        sys_name[sys_length] = '\0';
+        evt_name = evt_name + 1;
+
+        comma_loc = strchr(evt_name, ',');
+        if (comma_loc) {
+                /* take the event name up to the comma */
+                evt_name = strndup(evt_name, comma_loc - evt_name);
+        }
+        flags = strchr(evt_name, ':');
+        if (flags) {
+                /* split it out: */
+                evt_name = strndup(evt_name, flags - evt_name);
+                flags++;
         }
 
-        for (i = 0; i < ARRAY_SIZE(event_symbols); i++) {
-                if (!strncmp(str, event_symbols[i].symbol,
-                             strlen(event_symbols[i].symbol))) {
+        evt_length = strlen(evt_name);
+        if (evt_length >= MAX_EVENT_LENGTH)
+                return EVT_FAILED;
+        if (strpbrk(evt_name, "*?")) {
+                *strp += strlen(sys_name) + evt_length;
+                return parse_multiple_tracepoint_event(sys_name, evt_name,
+                                                       flags);
+        } else {
+                if (store_event_type(evt_name) < 0)
+                        return EVT_FAILED;
+
+                return parse_single_tracepoint_event(sys_name, evt_name,
+                                                     evt_length, attr, strp);
+        }
+}
+
+static enum event_result
+parse_breakpoint_type(const char *type, const char **strp,
+                      struct perf_event_attr *attr)
+{
+        int i;
+
+        for (i = 0; i < 3; i++) {
+                if (!type[i])
+                        break;
+
+                switch (type[i]) {
+                case 'r':
+                        attr->bp_type |= HW_BREAKPOINT_R;
+                        break;
+                case 'w':
+                        attr->bp_type |= HW_BREAKPOINT_W;
+                        break;
+                case 'x':
+                        attr->bp_type |= HW_BREAKPOINT_X;
+                        break;
+                default:
+                        return EVT_FAILED;
+                }
+        }
+        if (!attr->bp_type) /* Default */
+                attr->bp_type = HW_BREAKPOINT_R | HW_BREAKPOINT_W;
+
+        *strp = type + i;
+
+        return EVT_HANDLED;
+}
 
+static enum event_result
+parse_breakpoint_event(const char **strp, struct perf_event_attr *attr)
+{
+        const char *target;
+        const char *type;
+        char *endaddr;
+        u64 addr;
+        enum event_result err;
+
+        target = strchr(*strp, ':');
+        if (!target)
+                return EVT_FAILED;
+
+        if (strncmp(*strp, "mem", target - *strp) != 0)
+                return EVT_FAILED;
+
+        target++;
+
+        addr = strtoull(target, &endaddr, 0);
+        if (target == endaddr)
+                return EVT_FAILED;
+
+        attr->bp_addr = addr;
+        *strp = endaddr;
+
+        type = strchr(target, ':');
+
+        /* If no type is defined, just rw as default */
+        if (!type) {
+                attr->bp_type = HW_BREAKPOINT_R | HW_BREAKPOINT_W;
+        } else {
+                err = parse_breakpoint_type(++type, strp, attr);
+                if (err == EVT_FAILED)
+                        return EVT_FAILED;
+        }
+
+        /*
+         * We should find a nice way to override the access length
+         * Provide some defaults for now
+         */
+        if (attr->bp_type == HW_BREAKPOINT_X)
+                attr->bp_len = sizeof(long);
+        else
+                attr->bp_len = HW_BREAKPOINT_LEN_4;
+
+        attr->type = PERF_TYPE_BREAKPOINT;
+
+        return EVT_HANDLED;
+}
+
+static int check_events(const char *str, unsigned int i)
+{
+        int n;
+
+        n = strlen(event_symbols[i].symbol);
+        if (!strncmp(str, event_symbols[i].symbol, n))
+                return n;
+
+        n = strlen(event_symbols[i].alias);
+        if (n)
+                if (!strncmp(str, event_symbols[i].alias, n))
+                        return n;
+        return 0;
+}
+
+static enum event_result
+parse_symbolic_event(const char **strp, struct perf_event_attr *attr)
+{
+        const char *str = *strp;
+        unsigned int i;
+        int n;
+
+        for (i = 0; i < ARRAY_SIZE(event_symbols); i++) {
+                n = check_events(str, i);
+                if (n > 0) {
                         attr->type = event_symbols[i].type;
                         attr->config = event_symbols[i].config;
+                        *strp = str + n;
+                        return EVT_HANDLED;
+                }
+        }
+        return EVT_FAILED;
+}
+
+static enum event_result
+parse_raw_event(const char **strp, struct perf_event_attr *attr)
+{
+        const char *str = *strp;
+        u64 config;
+        int n;
 
-                        return 0;
+        if (*str != 'r')
+                return EVT_FAILED;
+        n = hex2u64(str + 1, &config);
+        if (n > 0) {
+                *strp = str + n + 1;
+                attr->type = PERF_TYPE_RAW;
+                attr->config = config;
+                return EVT_HANDLED;
+        }
+        return EVT_FAILED;
+}
+
+static enum event_result
+parse_numeric_event(const char **strp, struct perf_event_attr *attr)
+{
+        const char *str = *strp;
+        char *endp;
+        unsigned long type;
+        u64 config;
+
+        type = strtoul(str, &endp, 0);
+        if (endp > str && type < PERF_TYPE_MAX && *endp == ':') {
+                str = endp + 1;
+                config = strtoul(str, &endp, 0);
+                if (endp > str) {
+                        attr->type = type;
+                        attr->config = config;
+                        *strp = endp;
+                        return EVT_HANDLED;
                 }
         }
+        return EVT_FAILED;
+}
+
+static enum event_result
+parse_event_modifier(const char **strp, struct perf_event_attr *attr)
+{
+        const char *str = *strp;
+        int exclude = 0;
+        int eu = 0, ek = 0, eh = 0, precise = 0;
+
+        if (*str++ != ':')
+                return 0;
+        while (*str) {
+                if (*str == 'u') {
+                        if (!exclude)
+                                exclude = eu = ek = eh = 1;
+                        eu = 0;
+                } else if (*str == 'k') {
+                        if (!exclude)
+                                exclude = eu = ek = eh = 1;
+                        ek = 0;
+                } else if (*str == 'h') {
+                        if (!exclude)
+                                exclude = eu = ek = eh = 1;
+                        eh = 0;
+                } else if (*str == 'p') {
+                        precise++;
+                } else
+                        break;
 
-        return parse_generic_hw_symbols(str, attr);
+                ++str;
+        }
+        if (str >= *strp + 2) {
+                *strp = str;
+                attr->exclude_user   = eu;
+                attr->exclude_kernel = ek;
+                attr->exclude_hv     = eh;
+                attr->precise_ip     = precise;
+                return 1;
+        }
+        return 0;
 }
 
-int parse_events(const struct option *opt, const char *str, int unset)
+/*
+ * Each event can have multiple symbolic names.
+ * Symbolic names are (almost) exactly matched.
+ */
+static enum event_result
+parse_event_symbols(const char **str, struct perf_event_attr *attr)
 {
-        struct perf_counter_attr attr;
-        int ret;
+        enum event_result ret;
 
-        memset(&attr, 0, sizeof(attr));
-again:
-        if (nr_counters == MAX_COUNTERS)
-                return -1;
+        ret = parse_tracepoint_event(str, attr);
+        if (ret != EVT_FAILED)
+                goto modifier;
+
+        ret = parse_raw_event(str, attr);
+        if (ret != EVT_FAILED)
+                goto modifier;
+
+        ret = parse_numeric_event(str, attr);
+        if (ret != EVT_FAILED)
+                goto modifier;
+
+        ret = parse_symbolic_event(str, attr);
+        if (ret != EVT_FAILED)
+                goto modifier;
+
+        ret = parse_generic_hw_event(str, attr);
+        if (ret != EVT_FAILED)
+                goto modifier;
+
+        ret = parse_breakpoint_event(str, attr);
+        if (ret != EVT_FAILED)
+                goto modifier;
+
+        fprintf(stderr, "invalid or unsupported event: '%s'\n", *str);
+        fprintf(stderr, "Run 'perf list' for a list of valid events\n");
+        return EVT_FAILED;
+
+modifier:
+        parse_event_modifier(str, attr);
+
+        return ret;
+}
+
+int parse_events(const struct option *opt __used, const char *str, int unset __used)
+{
+        struct perf_event_attr attr;
+        enum event_result ret;
 
-        ret = parse_event_symbols(str, &attr);
-        if (ret < 0)
-                return ret;
+        for (;;) {
+                memset(&attr, 0, sizeof(attr));
+                ret = parse_event_symbols(&str, &attr);
+                if (ret == EVT_FAILED)
+                        return -1;
 
-        attrs[nr_counters] = attr;
-        nr_counters++;
+                if (!(*str == 0 || *str == ',' || isspace(*str)))
+                        return -1;
 
-        str = strstr(str, ",");
-        if (str) {
-                str++;
-                goto again;
+                if (ret != EVT_HANDLED_ALL) {
+                        struct perf_evsel *evsel;
+                        evsel = perf_evsel__new(&attr,
+                                                nr_counters);
+                        if (evsel == NULL)
+                                return -1;
+                        list_add_tail(&evsel->node, &evsel_list);
+                        ++nr_counters;
+                }
+
+                if (*str == 0)
+                        break;
+                if (*str == ',')
+                        ++str;
+                while (isspace(*str))
+                        ++str;
+        }
+
+        return 0;
+}
+
+int parse_filter(const struct option *opt __used, const char *str,
+                 int unset __used)
+{
+        struct perf_evsel *last = NULL;
+
+        if (!list_empty(&evsel_list))
+                last = list_entry(evsel_list.prev, struct perf_evsel, node);
+
+        if (last == NULL || last->attr.type != PERF_TYPE_TRACEPOINT) {
+                fprintf(stderr,
+                        "-F option should follow a -e tracepoint option\n");
+                return -1;
+        }
+
+        last->filter = strdup(str);
+        if (last->filter == NULL) {
+                fprintf(stderr, "not enough memory to hold filter string\n");
+                return -1;
         }
 
         return 0;
 }
 
 static const char * const event_type_descriptors[] = {
-        "",
         "Hardware event",
         "Software event",
         "Tracepoint event",
         "Hardware cache event",
+        "Raw hardware event descriptor",
+        "Hardware breakpoint",
 };
 
 /*
+ * Print the events from <debugfs_mount_point>/tracing/events
+ */
+
+static void print_tracepoint_events(void)
+{
+        DIR *sys_dir, *evt_dir;
+        struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent;
+        char evt_path[MAXPATHLEN];
+        char dir_path[MAXPATHLEN];
+
+        if (debugfs_valid_mountpoint(debugfs_path))
+                return;
+
+        sys_dir = opendir(debugfs_path);
+        if (!sys_dir)
+                return;
+
+        for_each_subsystem(sys_dir, sys_dirent, sys_next) {
+
+                snprintf(dir_path, MAXPATHLEN, "%s/%s", debugfs_path,
+                         sys_dirent.d_name);
+                evt_dir = opendir(dir_path);
+                if (!evt_dir)
+                        continue;
+
+                for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) {
+                        snprintf(evt_path, MAXPATHLEN, "%s:%s",
+                                 sys_dirent.d_name, evt_dirent.d_name);
+                        printf("  %-42s [%s]\n", evt_path,
+                                event_type_descriptors[PERF_TYPE_TRACEPOINT]);
+                }
+                closedir(evt_dir);
+        }
+        closedir(sys_dir);
+}
+
+/*
+ * Check whether event is in <debugfs_mount_point>/tracing/events
+ */
+
+int is_valid_tracepoint(const char *event_string)
+{
+        DIR *sys_dir, *evt_dir;
+        struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent;
+        char evt_path[MAXPATHLEN];
+        char dir_path[MAXPATHLEN];
+
+        if (debugfs_valid_mountpoint(debugfs_path))
+                return 0;
+
+        sys_dir = opendir(debugfs_path);
+        if (!sys_dir)
+                return 0;
+
+        for_each_subsystem(sys_dir, sys_dirent, sys_next) {
+
+                snprintf(dir_path, MAXPATHLEN, "%s/%s", debugfs_path,
+                         sys_dirent.d_name);
+                evt_dir = opendir(dir_path);
+                if (!evt_dir)
+                        continue;
+
+                for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) {
+                        snprintf(evt_path, MAXPATHLEN, "%s:%s",
+                                 sys_dirent.d_name, evt_dirent.d_name);
+                        if (!strcmp(evt_path, event_string)) {
+                                closedir(evt_dir);
+                                closedir(sys_dir);
+                                return 1;
+                        }
+                }
+                closedir(evt_dir);
+        }
+        closedir(sys_dir);
+        return 0;
+}
+
+/*
  * Print the help text for the event symbols:
  */
 void print_events(void)
 {
         struct event_symbol *syms = event_symbols;
-        unsigned int i, type, prev_type = -1;
+        unsigned int i, type, op, prev_type = -1;
+        char name[40];
 
-        fprintf(stderr, "\n");
-        fprintf(stderr, "List of pre-defined events (to be used in -e):\n");
+        printf("\n");
+        printf("List of pre-defined events (to be used in -e):\n");
 
         for (i = 0; i < ARRAY_SIZE(event_symbols); i++, syms++) {
-                type = syms->type + 1;
-                if (type > ARRAY_SIZE(event_type_descriptors))
-                        type = 0;
+                type = syms->type;
 
                 if (type != prev_type)
-                        fprintf(stderr, "\n");
+                        printf("\n");
 
-                fprintf(stderr, "  %-30s [%s]\n", syms->symbol,
+                if (strlen(syms->alias))
+                        sprintf(name, "%s OR %s", syms->symbol, syms->alias);
+                else
+                        strcpy(name, syms->symbol);
+                printf("  %-42s [%s]\n", name,
                         event_type_descriptors[type]);
 
                 prev_type = type;
         }
 
-        fprintf(stderr, "\n");
-        fprintf(stderr, "  %-30s [raw hardware event descriptor]\n",
-                "rNNN");
-        fprintf(stderr, "\n");
+        printf("\n");
+        for (type = 0; type < PERF_COUNT_HW_CACHE_MAX; type++) {
+                for (op = 0; op < PERF_COUNT_HW_CACHE_OP_MAX; op++) {
+                        /* skip invalid cache type */
+                        if (!is_cache_op_valid(type, op))
+                                continue;
+
+                        for (i = 0; i < PERF_COUNT_HW_CACHE_RESULT_MAX; i++) {
+                                printf("  %-42s [%s]\n",
+                                        event_cache_name(type, op, i),
+                                        event_type_descriptors[PERF_TYPE_HW_CACHE]);
+                        }
+                }
+        }
+
+        printf("\n");
+        printf("  %-42s [%s]\n",
+                "rNNN (see 'perf list --help' on how to encode it)",
+               event_type_descriptors[PERF_TYPE_RAW]);
+        printf("\n");
+
+        printf("  %-42s [%s]\n",
+                        "mem:<addr>[:access]",
+                        event_type_descriptors[PERF_TYPE_BREAKPOINT]);
+        printf("\n");
+
+        print_tracepoint_events();
 
         exit(129);
 }
+
+int perf_evsel_list__create_default(void)
+{
+        struct perf_evsel *evsel;
+        struct perf_event_attr attr;
+
+        memset(&attr, 0, sizeof(attr));
+        attr.type = PERF_TYPE_HARDWARE;
+        attr.config = PERF_COUNT_HW_CPU_CYCLES;
+
+        evsel = perf_evsel__new(&attr, 0);
+
+        if (evsel == NULL)
+                return -ENOMEM;
+
+        list_add(&evsel->node, &evsel_list);
+        ++nr_counters;
+        return 0;
+}
+
+void perf_evsel_list__delete(void)
+{
+        struct perf_evsel *pos, *n;
+
+        list_for_each_entry_safe(pos, n, &evsel_list, node) {
+                list_del_init(&pos->node);
+                perf_evsel__delete(pos);
+        }
+        nr_counters = 0;
+}
diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h
index e3d552908e6..b82cafb8377 100644
--- a/tools/perf/util/parse-events.h
+++ b/tools/perf/util/parse-events.h
@@ -1,17 +1,44 @@
-
+#ifndef __PERF_PARSE_EVENTS_H
+#define __PERF_PARSE_EVENTS_H
 /*
  * Parse symbolic events/counts passed in as options:
  */
 
-extern int                      nr_counters;
+#include "../../../include/linux/perf_event.h"
+
+struct list_head;
+struct perf_evsel;
+
+extern struct list_head evsel_list;
+
+int perf_evsel_list__create_default(void);
+void perf_evsel_list__delete(void);
 
-extern struct perf_counter_attr attrs[MAX_COUNTERS];
+struct option;
 
-extern char *event_name(int ctr);
+struct tracepoint_path {
+        char *system;
+        char *name;
+        struct tracepoint_path *next;
+};
+
+extern struct tracepoint_path *tracepoint_id_to_path(u64 config);
+extern bool have_tracepoints(struct list_head *evsel_list);
+
+extern int                      nr_counters;
+
+const char *event_name(struct perf_evsel *event);
+extern const char *__event_name(int type, u64 config);
 
 extern int parse_events(const struct option *opt, const char *str, int unset);
+extern int parse_filter(const struct option *opt, const char *str, int unset);
 
 #define EVENTS_HELP_MAX (128*1024)
 
 extern void print_events(void);
+extern int is_valid_tracepoint(const char *event_string);
+
+extern char debugfs_path[];
+extern int valid_debugfs_mount(const char *debugfs);
 
+#endif /* __PERF_PARSE_EVENTS_H */
diff --git a/tools/perf/util/parse-options.c b/tools/perf/util/parse-options.c
index b3affb1658d..99d02aa57db 100644
--- a/tools/perf/util/parse-options.c
+++ b/tools/perf/util/parse-options.c
@@ -20,7 +20,8 @@ static int get_arg(struct parse_opt_ctx_t *p, const struct option *opt,
         if (p->opt) {
                 *arg = p->opt;
                 p->opt = NULL;
-        } else if (p->argc == 1 && (opt->flags & PARSE_OPT_LASTARG_DEFAULT)) {
+        } else if ((opt->flags & PARSE_OPT_LASTARG_DEFAULT) && (p->argc == 1 ||
+                    **(p->argv + 1) == '-')) {
                 *arg = (const char *)opt->defval;
         } else if (p->argc > 1) {
                 p->argc--;
@@ -48,10 +49,19 @@ static int get_value(struct parse_opt_ctx_t *p,
                                 break;
                         /* FALLTHROUGH */
                 case OPTION_BOOLEAN:
+                case OPTION_INCR:
                 case OPTION_BIT:
-                case OPTION_SET_INT:
+                case OPTION_SET_UINT:
                 case OPTION_SET_PTR:
                         return opterror(opt, "takes no value", flags);
+                case OPTION_END:
+                case OPTION_ARGUMENT:
+                case OPTION_GROUP:
+                case OPTION_STRING:
+                case OPTION_INTEGER:
+                case OPTION_UINTEGER:
+                case OPTION_LONG:
+                case OPTION_U64:
                 default:
                         break;
                 }
@@ -66,11 +76,15 @@ static int get_value(struct parse_opt_ctx_t *p,
                 return 0;
 
         case OPTION_BOOLEAN:
+                *(bool *)opt->value = unset ? false : true;
+                return 0;
+
+        case OPTION_INCR:
                 *(int *)opt->value = unset ? 0 : *(int *)opt->value + 1;
                 return 0;
 
-        case OPTION_SET_INT:
-                *(int *)opt->value = unset ? 0 : opt->defval;
+        case OPTION_SET_UINT:
+                *(unsigned int *)opt->value = unset ? 0 : opt->defval;
                 return 0;
 
         case OPTION_SET_PTR:
@@ -113,6 +127,22 @@ static int get_value(struct parse_opt_ctx_t *p,
                         return opterror(opt, "expects a numerical value", flags);
                 return 0;
 
+        case OPTION_UINTEGER:
+                if (unset) {
+                        *(unsigned int *)opt->value = 0;
+                        return 0;
+                }
+                if (opt->flags & PARSE_OPT_OPTARG && !p->opt) {
+                        *(unsigned int *)opt->value = opt->defval;
+                        return 0;
+                }
+                if (get_arg(p, opt, flags, &arg))
+                        return -1;
+                *(unsigned int *)opt->value = strtol(arg, (char **)&s, 10);
+                if (*s)
+                        return opterror(opt, "expects a numerical value", flags);
+                return 0;
+
         case OPTION_LONG:
                 if (unset) {
                         *(long *)opt->value = 0;
@@ -129,6 +159,25 @@ static int get_value(struct parse_opt_ctx_t *p,
                         return opterror(opt, "expects a numerical value", flags);
                 return 0;
 
+        case OPTION_U64:
+                if (unset) {
+                        *(u64 *)opt->value = 0;
+                        return 0;
+                }
+                if (opt->flags & PARSE_OPT_OPTARG && !p->opt) {
+                        *(u64 *)opt->value = opt->defval;
+                        return 0;
+                }
+                if (get_arg(p, opt, flags, &arg))
+                        return -1;
+                *(u64 *)opt->value = strtoull(arg, (char **)&s, 10);
+                if (*s)
+                        return opterror(opt, "expects a numerical value", flags);
+                return 0;
+
+        case OPTION_END:
+        case OPTION_ARGUMENT:
+        case OPTION_GROUP:
         default:
                 die("should not happen, someone must be hit on the forehead");
         }
@@ -295,6 +344,8 @@ int parse_options_step(struct parse_opt_ctx_t *ctx,
                                 return parse_options_usage(usagestr, options);
                         case -2:
                                 goto unknown;
+                        default:
+                                break;
                         }
                         if (ctx->opt)
                                 check_typos(arg + 1, options);
@@ -313,6 +364,8 @@ int parse_options_step(struct parse_opt_ctx_t *ctx,
                                         ctx->argv[0] = strdup(ctx->opt - 1);
                                         *(char *)ctx->argv[0] = '-';
                                         goto unknown;
+                                default:
+                                        break;
                                 }
                         }
                         continue;
@@ -335,6 +388,8 @@ int parse_options_step(struct parse_opt_ctx_t *ctx,
                         return parse_options_usage(usagestr, options);
                 case -2:
                         goto unknown;
+                default:
+                        break;
                 }
                 continue;
 unknown:
@@ -414,6 +469,9 @@ 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)
@@ -422,7 +480,10 @@ int usage_with_options_internal(const char * const *usagestr,
                 switch (opts->type) {
                 case OPTION_ARGUMENT:
                         break;
+                case OPTION_LONG:
+                case OPTION_U64:
                 case OPTION_INTEGER:
+                case OPTION_UINTEGER:
                         if (opts->flags & PARSE_OPT_OPTARG)
                                 if (opts->long_name)
                                         pos += fprintf(stderr, "[=<n>]");
@@ -454,7 +515,14 @@ int usage_with_options_internal(const char * const *usagestr,
                                         pos += fprintf(stderr, " ...");
                         }
                         break;
-                default: /* OPTION_{BIT,BOOLEAN,SET_INT,SET_PTR} */
+                default: /* OPTION_{BIT,BOOLEAN,SET_UINT,SET_PTR} */
+                case OPTION_END:
+                case OPTION_GROUP:
+                case OPTION_BIT:
+                case OPTION_BOOLEAN:
+                case OPTION_INCR:
+                case OPTION_SET_UINT:
+                case OPTION_SET_PTR:
                         break;
                 }
 
@@ -474,6 +542,7 @@ int usage_with_options_internal(const char * const *usagestr,
 void usage_with_options(const char * const *usagestr,
                         const struct option *opts)
 {
+        exit_browser(false);
         usage_with_options_internal(usagestr, opts, 0);
         exit(129);
 }
@@ -485,7 +554,7 @@ int parse_options_usage(const char * const *usagestr,
 }
 
 
-int parse_opt_verbosity_cb(const struct option *opt, const char *arg,
+int parse_opt_verbosity_cb(const struct option *opt, const char *arg __used,
                            int unset)
 {
         int *target = opt->value;
diff --git a/tools/perf/util/parse-options.h b/tools/perf/util/parse-options.h
index a1039a6ce0e..abc31a1dac1 100644
--- a/tools/perf/util/parse-options.h
+++ b/tools/perf/util/parse-options.h
@@ -1,5 +1,8 @@
-#ifndef PARSE_OPTIONS_H
-#define PARSE_OPTIONS_H
+#ifndef __PERF_PARSE_OPTIONS_H
+#define __PERF_PARSE_OPTIONS_H
+
+#include <linux/kernel.h>
+#include <stdbool.h>
 
 enum parse_opt_type {
         /* special types */
@@ -8,14 +11,17 @@ enum parse_opt_type {
         OPTION_GROUP,
         /* options with no arguments */
         OPTION_BIT,
-        OPTION_BOOLEAN, /* _INCR would have been a better name */
-        OPTION_SET_INT,
+        OPTION_BOOLEAN,
+        OPTION_INCR,
+        OPTION_SET_UINT,
         OPTION_SET_PTR,
         /* options with arguments (usually) */
         OPTION_STRING,
         OPTION_INTEGER,
         OPTION_LONG,
         OPTION_CALLBACK,
+        OPTION_U64,
+        OPTION_UINTEGER,
 };
 
 enum parse_opt_flags {
@@ -73,7 +79,7 @@ typedef int parse_opt_cb(const struct option *, const char *arg, int unset);
  *
  * `defval`::
  *   default value to fill (*->value) with for PARSE_OPT_OPTARG.
- *   OPTION_{BIT,SET_INT,SET_PTR} store the {mask,integer,pointer} to put in
+ *   OPTION_{BIT,SET_UINT,SET_PTR} store the {mask,integer,pointer} to put in
  *   the value when met.
  *   CALLBACKS can use it like they want.
  */
@@ -90,21 +96,33 @@ struct option {
         intptr_t defval;
 };
 
-#define OPT_END()                   { OPTION_END }
-#define OPT_ARGUMENT(l, h)          { OPTION_ARGUMENT, 0, (l), NULL, NULL, (h) }
-#define OPT_GROUP(h)                { OPTION_GROUP, 0, NULL, NULL, NULL, (h) }
-#define OPT_BIT(s, l, v, h, b)      { OPTION_BIT, (s), (l), (v), NULL, (h), 0, NULL, (b) }
-#define OPT_BOOLEAN(s, l, v, h)     { OPTION_BOOLEAN, (s), (l), (v), NULL, (h) }
-#define OPT_SET_INT(s, l, v, h, i)  { OPTION_SET_INT, (s), (l), (v), NULL, (h), 0, NULL, (i) }
-#define OPT_SET_PTR(s, l, v, h, p)  { OPTION_SET_PTR, (s), (l), (v), NULL, (h), 0, NULL, (p) }
-#define OPT_INTEGER(s, l, v, h)     { OPTION_INTEGER, (s), (l), (v), NULL, (h) }
-#define OPT_LONG(s, l, v, h)        { OPTION_LONG, (s), (l), (v), NULL, (h) }
-#define OPT_STRING(s, l, v, a, h)   { OPTION_STRING,  (s), (l), (v), (a), (h) }
+#define check_vtype(v, type) ( BUILD_BUG_ON_ZERO(!__builtin_types_compatible_p(typeof(v), type)) + v )
+
+#define OPT_END()                   { .type = OPTION_END }
+#define OPT_ARGUMENT(l, h)          { .type = OPTION_ARGUMENT, .long_name = (l), .help = (h) }
+#define OPT_GROUP(h)                { .type = OPTION_GROUP, .help = (h) }
+#define OPT_BIT(s, l, v, h, b)      { .type = OPTION_BIT, .short_name = (s), .long_name = (l), .value = check_vtype(v, int *), .help = (h), .defval = (b) }
+#define OPT_BOOLEAN(s, l, v, h)     { .type = OPTION_BOOLEAN, .short_name = (s), .long_name = (l), .value = check_vtype(v, bool *), .help = (h) }
+#define OPT_INCR(s, l, v, h)        { .type = OPTION_INCR, .short_name = (s), .long_name = (l), .value = check_vtype(v, int *), .help = (h) }
+#define OPT_SET_UINT(s, l, v, h, i)  { .type = OPTION_SET_UINT, .short_name = (s), .long_name = (l), .value = check_vtype(v, unsigned int *), .help = (h), .defval = (i) }
+#define OPT_SET_PTR(s, l, v, h, p)  { .type = OPTION_SET_PTR, .short_name = (s), .long_name = (l), .value = (v), .help = (h), .defval = (p) }
+#define OPT_INTEGER(s, l, v, h)     { .type = OPTION_INTEGER, .short_name = (s), .long_name = (l), .value = check_vtype(v, int *), .help = (h) }
+#define OPT_UINTEGER(s, l, v, h)    { .type = OPTION_UINTEGER, .short_name = (s), .long_name = (l), .value = check_vtype(v, unsigned int *), .help = (h) }
+#define OPT_LONG(s, l, v, h)        { .type = OPTION_LONG, .short_name = (s), .long_name = (l), .value = check_vtype(v, long *), .help = (h) }
+#define OPT_U64(s, l, v, h)         { .type = OPTION_U64, .short_name = (s), .long_name = (l), .value = check_vtype(v, u64 *), .help = (h) }
+#define OPT_STRING(s, l, v, a, h)   { .type = OPTION_STRING,  .short_name = (s), .long_name = (l), .value = check_vtype(v, const char **), (a), .help = (h) }
 #define OPT_DATE(s, l, v, h) \
-        { OPTION_CALLBACK, (s), (l), (v), "time",(h), 0, \
-          parse_opt_approxidate_cb }
+        { .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), .argh = "time", .help = (h), .callback = parse_opt_approxidate_cb }
 #define OPT_CALLBACK(s, l, v, a, h, f) \
-        { OPTION_CALLBACK, (s), (l), (v), (a), (h), 0, (f) }
+        { .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), (a), .help = (h), .callback = (f) }
+#define OPT_CALLBACK_NOOPT(s, l, v, a, h, f) \
+        { .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), (a), .help = (h), .callback = (f), .flags = PARSE_OPT_NOARG }
+#define OPT_CALLBACK_DEFAULT(s, l, v, a, h, f, d) \
+        { .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), (a), .help = (h), .callback = (f), .defval = (intptr_t)d, .flags = PARSE_OPT_LASTARG_DEFAULT }
+#define OPT_CALLBACK_DEFAULT_NOOPT(s, l, v, a, h, f, d) \
+        { .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l),\
+        .value = (v), (a), .help = (h), .callback = (f), .defval = (intptr_t)d,\
+        .flags = PARSE_OPT_LASTARG_DEFAULT | PARSE_OPT_NOARG}
 
 /* parse_options() will filter out the processed options and leave the
  * non-option argments in argv[].
@@ -171,4 +189,4 @@ extern int parse_opt_verbosity_cb(const struct option *, const char *, int);
 
 extern const char *parse_options_fix_filename(const char *prefix, const char *file);
 
-#endif
+#endif /* __PERF_PARSE_OPTIONS_H */
diff --git a/tools/perf/util/path.c b/tools/perf/util/path.c
index a501a40dd2c..bd749771142 100644
--- a/tools/perf/util/path.c
+++ b/tools/perf/util/path.c
@@ -17,11 +17,12 @@ static char bad_path[] = "/bad-path/";
  * Two hacks:
  */
 
-static char *get_perf_dir(void)
+static const char *get_perf_dir(void)
 {
         return ".";
 }
 
+#ifdef NO_STRLCPY
 size_t strlcpy(char *dest, const char *src, size_t size)
 {
         size_t ret = strlen(src);
@@ -33,13 +34,14 @@ size_t strlcpy(char *dest, const char *src, size_t size)
         }
         return ret;
 }
-
+#endif
 
 static char *get_pathname(void)
 {
         static char pathname_array[4][PATH_MAX];
-        static int index;
-        return pathname_array[3 & ++index];
+        static int idx;
+
+        return pathname_array[3 & ++idx];
 }
 
 static char *cleanup_path(char *path)
@@ -53,21 +55,6 @@ static char *cleanup_path(char *path)
         return path;
 }
 
-char *mksnpath(char *buf, size_t n, const char *fmt, ...)
-{
-        va_list args;
-        unsigned len;
-
-        va_start(args, fmt);
-        len = vsnprintf(buf, n, fmt, args);
-        va_end(args);
-        if (len >= n) {
-                strlcpy(buf, bad_path, n);
-                return buf;
-        }
-        return cleanup_path(buf);
-}
-
 static char *perf_vsnpath(char *buf, size_t n, const char *fmt, va_list args)
 {
         const char *perf_dir = get_perf_dir();
@@ -88,15 +75,6 @@ bad:
         return buf;
 }
 
-char *perf_snpath(char *buf, size_t n, const char *fmt, ...)
-{
-        va_list args;
-        va_start(args, fmt);
-        (void)perf_vsnpath(buf, n, fmt, args);
-        va_end(args);
-        return buf;
-}
-
 char *perf_pathdup(const char *fmt, ...)
 {
         char path[PATH_MAX];
@@ -142,180 +120,6 @@ char *perf_path(const char *fmt, ...)
         return cleanup_path(pathname);
 }
 
-
-/* perf_mkstemp() - create tmp file honoring TMPDIR variable */
-int perf_mkstemp(char *path, size_t len, const char *template)
-{
-        const char *tmp;
-        size_t n;
-
-        tmp = getenv("TMPDIR");
-        if (!tmp)
-                tmp = "/tmp";
-        n = snprintf(path, len, "%s/%s", tmp, template);
-        if (len <= n) {
-                errno = ENAMETOOLONG;
-                return -1;
-        }
-        return mkstemp(path);
-}
-
-
-const char *make_relative_path(const char *abs, const char *base)
-{
-        static char buf[PATH_MAX + 1];
-        int baselen;
-        if (!base)
-                return abs;
-        baselen = strlen(base);
-        if (prefixcmp(abs, base))
-                return abs;
-        if (abs[baselen] == '/')
-                baselen++;
-        else if (base[baselen - 1] != '/')
-                return abs;
-        strcpy(buf, abs + baselen);
-        return buf;
-}
-
-/*
- * It is okay if dst == src, but they should not overlap otherwise.
- *
- * Performs the following normalizations on src, storing the result in dst:
- * - Ensures that components are separated by '/' (Windows only)
- * - Squashes sequences of '/'.
- * - Removes "." components.
- * - Removes ".." components, and the components the precede them.
- * Returns failure (non-zero) if a ".." component appears as first path
- * component anytime during the normalization. Otherwise, returns success (0).
- *
- * Note that this function is purely textual.  It does not follow symlinks,
- * verify the existence of the path, or make any system calls.
- */
-int normalize_path_copy(char *dst, const char *src)
-{
-        char *dst0;
-
-        if (has_dos_drive_prefix(src)) {
-                *dst++ = *src++;
-                *dst++ = *src++;
-        }
-        dst0 = dst;
-
-        if (is_dir_sep(*src)) {
-                *dst++ = '/';
-                while (is_dir_sep(*src))
-                        src++;
-        }
-
-        for (;;) {
-                char c = *src;
-
-                /*
-                 * A path component that begins with . could be
-                 * special:
-                 * (1) "." and ends   -- ignore and terminate.
-                 * (2) "./"           -- ignore them, eat slash and continue.
-                 * (3) ".." and ends  -- strip one and terminate.
-                 * (4) "../"          -- strip one, eat slash and continue.
-                 */
-                if (c == '.') {
-                        if (!src[1]) {
-                                /* (1) */
-                                src++;
-                        } else if (is_dir_sep(src[1])) {
-                                /* (2) */
-                                src += 2;
-                                while (is_dir_sep(*src))
-                                        src++;
-                                continue;
-                        } else if (src[1] == '.') {
-                                if (!src[2]) {
-                                        /* (3) */
-                                        src += 2;
-                                        goto up_one;
-                                } else if (is_dir_sep(src[2])) {
-                                        /* (4) */
-                                        src += 3;
-                                        while (is_dir_sep(*src))
-                                                src++;
-                                        goto up_one;
-                                }
-                        }
-                }
-
-                /* copy up to the next '/', and eat all '/' */
-                while ((c = *src++) != '\0' && !is_dir_sep(c))
-                        *dst++ = c;
-                if (is_dir_sep(c)) {
-                        *dst++ = '/';
-                        while (is_dir_sep(c))
-                                c = *src++;
-                        src--;
-                } else if (!c)
-                        break;
-                continue;
-
-        up_one:
-                /*
-                 * dst0..dst is prefix portion, and dst[-1] is '/';
-                 * go up one level.
-                 */
-                dst--;  /* go to trailing '/' */
-                if (dst <= dst0)
-                        return -1;
-                /* Windows: dst[-1] cannot be backslash anymore */
-                while (dst0 < dst && dst[-1] != '/')
-                        dst--;
-        }
-        *dst = '\0';
-        return 0;
-}
-
-/*
- * path = Canonical absolute path
- * prefix_list = Colon-separated list of absolute paths
- *
- * Determines, for each path in prefix_list, whether the "prefix" really
- * is an ancestor directory of path.  Returns the length of the longest
- * ancestor directory, excluding any trailing slashes, or -1 if no prefix
- * is an ancestor.  (Note that this means 0 is returned if prefix_list is
- * "/".) "/foo" is not considered an ancestor of "/foobar".  Directories
- * are not considered to be their own ancestors.  path must be in a
- * canonical form: empty components, or "." or ".." components are not
- * allowed.  prefix_list may be null, which is like "".
- */
-int longest_ancestor_length(const char *path, const char *prefix_list)
-{
-        char buf[PATH_MAX+1];
-        const char *ceil, *colon;
-        int len, max_len = -1;
-
-        if (prefix_list == NULL || !strcmp(path, "/"))
-                return -1;
-
-        for (colon = ceil = prefix_list; *colon; ceil = colon+1) {
-                for (colon = ceil; *colon && *colon != PATH_SEP; colon++);
-                len = colon - ceil;
-                if (len == 0 || len > PATH_MAX || !is_absolute_path(ceil))
-                        continue;
-                strlcpy(buf, ceil, len+1);
-                if (normalize_path_copy(buf, buf) < 0)
-                        continue;
-                len = strlen(buf);
-                if (len > 0 && buf[len-1] == '/')
-                        buf[--len] = '\0';
-
-                if (!strncmp(path, buf, len) &&
-                    path[len] == '/' &&
-                    len > max_len) {
-                        max_len = len;
-                }
-        }
-
-        return max_len;
-}
-
 /* strip arbitrary amount of directory separators at end of path */
 static inline int chomp_trailing_dir_sep(const char *path, int len)
 {
@@ -349,5 +153,5 @@ char *strip_path_suffix(const char *path, const char *suffix)
 
         if (path_len && !is_dir_sep(path[path_len - 1]))
                 return NULL;
-        return xstrndup(path, chomp_trailing_dir_sep(path, path_len));
+        return strndup(path, chomp_trailing_dir_sep(path, path_len));
 }
diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
new file mode 100644
index 00000000000..128aaab0aed
--- /dev/null
+++ b/tools/perf/util/probe-event.c
@@ -0,0 +1,1915 @@
+/*
+ * probe-event.c : perf-probe definition to probe_events format converter
+ *
+ * Written by Masami Hiramatsu <mhiramat@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+
+#define _GNU_SOURCE
+#include <sys/utsname.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <errno.h>
+#include <stdio.h>
+#include <unistd.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdarg.h>
+#include <limits.h>
+
+#undef _GNU_SOURCE
+#include "util.h"
+#include "event.h"
+#include "string.h"
+#include "strlist.h"
+#include "debug.h"
+#include "cache.h"
+#include "color.h"
+#include "symbol.h"
+#include "thread.h"
+#include "debugfs.h"
+#include "trace-event.h"        /* For __unused */
+#include "probe-event.h"
+#include "probe-finder.h"
+
+#define MAX_CMDLEN 256
+#define MAX_PROBE_ARGS 128
+#define PERFPROBE_GROUP "probe"
+
+bool probe_event_dry_run;       /* Dry run flag */
+
+#define semantic_error(msg ...) pr_err("Semantic error :" msg)
+
+/* If there is no space to write, returns -E2BIG. */
+static int e_snprintf(char *str, size_t size, const char *format, ...)
+        __attribute__((format(printf, 3, 4)));
+
+static int e_snprintf(char *str, size_t size, const char *format, ...)
+{
+        int ret;
+        va_list ap;
+        va_start(ap, format);
+        ret = vsnprintf(str, size, format, ap);
+        va_end(ap);
+        if (ret >= (int)size)
+                ret = -E2BIG;
+        return ret;
+}
+
+static char *synthesize_perf_probe_point(struct perf_probe_point *pp);
+static struct machine machine;
+
+/* Initialize symbol maps and path of vmlinux/modules */
+static int init_vmlinux(void)
+{
+        int ret;
+
+        symbol_conf.sort_by_name = true;
+        if (symbol_conf.vmlinux_name == NULL)
+                symbol_conf.try_vmlinux_path = true;
+        else
+                pr_debug("Use vmlinux: %s\n", symbol_conf.vmlinux_name);
+        ret = symbol__init();
+        if (ret < 0) {
+                pr_debug("Failed to init symbol map.\n");
+                goto out;
+        }
+
+        ret = machine__init(&machine, "", HOST_KERNEL_ID);
+        if (ret < 0)
+                goto out;
+
+        if (machine__create_kernel_maps(&machine) < 0) {
+                pr_debug("machine__create_kernel_maps() failed.\n");
+                goto out;
+        }
+out:
+        if (ret < 0)
+                pr_warning("Failed to init vmlinux path.\n");
+        return ret;
+}
+
+static struct symbol *__find_kernel_function_by_name(const char *name,
+                                                     struct map **mapp)
+{
+        return machine__find_kernel_function_by_name(&machine, name, mapp,
+                                                     NULL);
+}
+
+const char *kernel_get_module_path(const char *module)
+{
+        struct dso *dso;
+        struct map *map;
+        const char *vmlinux_name;
+
+        if (module) {
+                list_for_each_entry(dso, &machine.kernel_dsos, node) {
+                        if (strncmp(dso->short_name + 1, module,
+                                    dso->short_name_len - 2) == 0)
+                                goto found;
+                }
+                pr_debug("Failed to find module %s.\n", module);
+                return NULL;
+        }
+
+        map = machine.vmlinux_maps[MAP__FUNCTION];
+        dso = map->dso;
+
+        vmlinux_name = symbol_conf.vmlinux_name;
+        if (vmlinux_name) {
+                if (dso__load_vmlinux(dso, map, vmlinux_name, NULL) <= 0)
+                        return NULL;
+        } else {
+                if (dso__load_vmlinux_path(dso, map, NULL) <= 0) {
+                        pr_debug("Failed to load kernel map.\n");
+                        return NULL;
+                }
+        }
+found:
+        return dso->long_name;
+}
+
+#ifdef DWARF_SUPPORT
+static int open_vmlinux(const char *module)
+{
+        const char *path = kernel_get_module_path(module);
+        if (!path) {
+                pr_err("Failed to find path of %s module.\n",
+                       module ?: "kernel");
+                return -ENOENT;
+        }
+        pr_debug("Try to open %s\n", path);
+        return open(path, O_RDONLY);
+}
+
+/*
+ * Convert trace point to probe point with debuginfo
+ * Currently only handles kprobes.
+ */
+static int kprobe_convert_to_perf_probe(struct probe_trace_point *tp,
+                                        struct perf_probe_point *pp)
+{
+        struct symbol *sym;
+        struct map *map;
+        u64 addr;
+        int ret = -ENOENT;
+
+        sym = __find_kernel_function_by_name(tp->symbol, &map);
+        if (sym) {
+                addr = map->unmap_ip(map, sym->start + tp->offset);
+                pr_debug("try to find %s+%ld@%llx\n", tp->symbol,
+                         tp->offset, addr);
+                ret = find_perf_probe_point((unsigned long)addr, pp);
+        }
+        if (ret <= 0) {
+                pr_debug("Failed to find corresponding probes from "
+                         "debuginfo. Use kprobe event information.\n");
+                pp->function = strdup(tp->symbol);
+                if (pp->function == NULL)
+                        return -ENOMEM;
+                pp->offset = tp->offset;
+        }
+        pp->retprobe = tp->retprobe;
+
+        return 0;
+}
+
+/* Try to find perf_probe_event with debuginfo */
+static int try_to_find_probe_trace_events(struct perf_probe_event *pev,
+                                           struct probe_trace_event **tevs,
+                                           int max_tevs, const char *module)
+{
+        bool need_dwarf = perf_probe_event_need_dwarf(pev);
+        int fd, ntevs;
+
+        fd = open_vmlinux(module);
+        if (fd < 0) {
+                if (need_dwarf) {
+                        pr_warning("Failed to open debuginfo file.\n");
+                        return fd;
+                }
+                pr_debug("Could not open vmlinux. Try to use symbols.\n");
+                return 0;
+        }
+
+        /* Searching trace events corresponding to probe event */
+        ntevs = find_probe_trace_events(fd, pev, tevs, max_tevs);
+        close(fd);
+
+        if (ntevs > 0) {        /* Succeeded to find trace events */
+                pr_debug("find %d probe_trace_events.\n", ntevs);
+                return ntevs;
+        }
+
+        if (ntevs == 0) {       /* No error but failed to find probe point. */
+                pr_warning("Probe point '%s' not found.\n",
+                           synthesize_perf_probe_point(&pev->point));
+                return -ENOENT;
+        }
+        /* Error path : ntevs < 0 */
+        pr_debug("An error occurred in debuginfo analysis (%d).\n", ntevs);
+        if (ntevs == -EBADF) {
+                pr_warning("Warning: No dwarf info found in the vmlinux - "
+                        "please rebuild kernel with CONFIG_DEBUG_INFO=y.\n");
+                if (!need_dwarf) {
+                        pr_debug("Trying to use symbols.\n");
+                        return 0;
+                }
+        }
+        return ntevs;
+}
+
+/*
+ * Find a src file from a DWARF tag path. Prepend optional source path prefix
+ * and chop off leading directories that do not exist. Result is passed back as
+ * a newly allocated path on success.
+ * Return 0 if file was found and readable, -errno otherwise.
+ */
+static int get_real_path(const char *raw_path, const char *comp_dir,
+                         char **new_path)
+{
+        const char *prefix = symbol_conf.source_prefix;
+
+        if (!prefix) {
+                if (raw_path[0] != '/' && comp_dir)
+                        /* If not an absolute path, try to use comp_dir */
+                        prefix = comp_dir;
+                else {
+                        if (access(raw_path, R_OK) == 0) {
+                                *new_path = strdup(raw_path);
+                                return 0;
+                        } else
+                                return -errno;
+                }
+        }
+
+        *new_path = malloc((strlen(prefix) + strlen(raw_path) + 2));
+        if (!*new_path)
+                return -ENOMEM;
+
+        for (;;) {
+                sprintf(*new_path, "%s/%s", prefix, raw_path);
+
+                if (access(*new_path, R_OK) == 0)
+                        return 0;
+
+                if (!symbol_conf.source_prefix)
+                        /* In case of searching comp_dir, don't retry */
+                        return -errno;
+
+                switch (errno) {
+                case ENAMETOOLONG:
+                case ENOENT:
+                case EROFS:
+                case EFAULT:
+                        raw_path = strchr(++raw_path, '/');
+                        if (!raw_path) {
+                                free(*new_path);
+                                *new_path = NULL;
+                                return -ENOENT;
+                        }
+                        continue;
+
+                default:
+                        free(*new_path);
+                        *new_path = NULL;
+                        return -errno;
+                }
+        }
+}
+
+#define LINEBUF_SIZE 256
+#define NR_ADDITIONAL_LINES 2
+
+static int __show_one_line(FILE *fp, int l, bool skip, bool show_num)
+{
+        char buf[LINEBUF_SIZE];
+        const char *color = show_num ? "" : PERF_COLOR_BLUE;
+        const char *prefix = NULL;
+
+        do {
+                if (fgets(buf, LINEBUF_SIZE, fp) == NULL)
+                        goto error;
+                if (skip)
+                        continue;
+                if (!prefix) {
+                        prefix = show_num ? "%7d  " : "         ";
+                        color_fprintf(stdout, color, prefix, l);
+                }
+                color_fprintf(stdout, color, "%s", buf);
+
+        } while (strchr(buf, '\n') == NULL);
+
+        return 1;
+error:
+        if (ferror(fp)) {
+                pr_warning("File read error: %s\n", strerror(errno));
+                return -1;
+        }
+        return 0;
+}
+
+static int _show_one_line(FILE *fp, int l, bool skip, bool show_num)
+{
+        int rv = __show_one_line(fp, l, skip, show_num);
+        if (rv == 0) {
+                pr_warning("Source file is shorter than expected.\n");
+                rv = -1;
+        }
+        return rv;
+}
+
+#define show_one_line_with_num(f,l)     _show_one_line(f,l,false,true)
+#define show_one_line(f,l)              _show_one_line(f,l,false,false)
+#define skip_one_line(f,l)              _show_one_line(f,l,true,false)
+#define show_one_line_or_eof(f,l)       __show_one_line(f,l,false,false)
+
+/*
+ * Show line-range always requires debuginfo to find source file and
+ * line number.
+ */
+int show_line_range(struct line_range *lr, const char *module)
+{
+        int l = 1;
+        struct line_node *ln;
+        FILE *fp;
+        int fd, ret;
+        char *tmp;
+
+        /* Search a line range */
+        ret = init_vmlinux();
+        if (ret < 0)
+                return ret;
+
+        fd = open_vmlinux(module);
+        if (fd < 0) {
+                pr_warning("Failed to open debuginfo file.\n");
+                return fd;
+        }
+
+        ret = find_line_range(fd, lr);
+        close(fd);
+        if (ret == 0) {
+                pr_warning("Specified source line is not found.\n");
+                return -ENOENT;
+        } else if (ret < 0) {
+                pr_warning("Debuginfo analysis failed. (%d)\n", ret);
+                return ret;
+        }
+
+        /* Convert source file path */
+        tmp = lr->path;
+        ret = get_real_path(tmp, lr->comp_dir, &lr->path);
+        free(tmp);      /* Free old path */
+        if (ret < 0) {
+                pr_warning("Failed to find source file. (%d)\n", ret);
+                return ret;
+        }
+
+        setup_pager();
+
+        if (lr->function)
+                fprintf(stdout, "<%s:%d>\n", lr->function,
+                        lr->start - lr->offset);
+        else
+                fprintf(stdout, "<%s:%d>\n", lr->path, lr->start);
+
+        fp = fopen(lr->path, "r");
+        if (fp == NULL) {
+                pr_warning("Failed to open %s: %s\n", lr->path,
+                           strerror(errno));
+                return -errno;
+        }
+        /* Skip to starting line number */
+        while (l < lr->start) {
+                ret = skip_one_line(fp, l++);
+                if (ret < 0)
+                        goto end;
+        }
+
+        list_for_each_entry(ln, &lr->line_list, list) {
+                for (; ln->line > l; l++) {
+                        ret = show_one_line(fp, l - lr->offset);
+                        if (ret < 0)
+                                goto end;
+                }
+                ret = show_one_line_with_num(fp, l++ - lr->offset);
+                if (ret < 0)
+                        goto end;
+        }
+
+        if (lr->end == INT_MAX)
+                lr->end = l + NR_ADDITIONAL_LINES;
+        while (l <= lr->end) {
+                ret = show_one_line_or_eof(fp, l++ - lr->offset);
+                if (ret <= 0)
+                        break;
+        }
+end:
+        fclose(fp);
+        return ret;
+}
+
+static int show_available_vars_at(int fd, struct perf_probe_event *pev,
+                                  int max_vls, bool externs)
+{
+        char *buf;
+        int ret, i;
+        struct str_node *node;
+        struct variable_list *vls = NULL, *vl;
+
+        buf = synthesize_perf_probe_point(&pev->point);
+        if (!buf)
+                return -EINVAL;
+        pr_debug("Searching variables at %s\n", buf);
+
+        ret = find_available_vars_at(fd, pev, &vls, max_vls, externs);
+        if (ret > 0) {
+                /* Some variables were found */
+                fprintf(stdout, "Available variables at %s\n", buf);
+                for (i = 0; i < ret; i++) {
+                        vl = &vls[i];
+                        /*
+                         * A probe point might be converted to
+                         * several trace points.
+                         */
+                        fprintf(stdout, "\t@<%s+%lu>\n", vl->point.symbol,
+                                vl->point.offset);
+                        free(vl->point.symbol);
+                        if (vl->vars) {
+                                strlist__for_each(node, vl->vars)
+                                        fprintf(stdout, "\t\t%s\n", node->s);
+                                strlist__delete(vl->vars);
+                        } else
+                                fprintf(stdout, "(No variables)\n");
+                }
+                free(vls);
+        } else
+                pr_err("Failed to find variables at %s (%d)\n", buf, ret);
+
+        free(buf);
+        return ret;
+}
+
+/* Show available variables on given probe point */
+int show_available_vars(struct perf_probe_event *pevs, int npevs,
+                        int max_vls, const char *module, bool externs)
+{
+        int i, fd, ret = 0;
+
+        ret = init_vmlinux();
+        if (ret < 0)
+                return ret;
+
+        fd = open_vmlinux(module);
+        if (fd < 0) {
+                pr_warning("Failed to open debug information file.\n");
+                return fd;
+        }
+
+        setup_pager();
+
+        for (i = 0; i < npevs && ret >= 0; i++)
+                ret = show_available_vars_at(fd, &pevs[i], max_vls, externs);
+
+        close(fd);
+        return ret;
+}
+
+#else   /* !DWARF_SUPPORT */
+
+static int kprobe_convert_to_perf_probe(struct probe_trace_point *tp,
+                                        struct perf_probe_point *pp)
+{
+        struct symbol *sym;
+
+        sym = __find_kernel_function_by_name(tp->symbol, NULL);
+        if (!sym) {
+                pr_err("Failed to find symbol %s in kernel.\n", tp->symbol);
+                return -ENOENT;
+        }
+        pp->function = strdup(tp->symbol);
+        if (pp->function == NULL)
+                return -ENOMEM;
+        pp->offset = tp->offset;
+        pp->retprobe = tp->retprobe;
+
+        return 0;
+}
+
+static int try_to_find_probe_trace_events(struct perf_probe_event *pev,
+                                struct probe_trace_event **tevs __unused,
+                                int max_tevs __unused, const char *mod __unused)
+{
+        if (perf_probe_event_need_dwarf(pev)) {
+                pr_warning("Debuginfo-analysis is not supported.\n");
+                return -ENOSYS;
+        }
+        return 0;
+}
+
+int show_line_range(struct line_range *lr __unused, const char *module __unused)
+{
+        pr_warning("Debuginfo-analysis is not supported.\n");
+        return -ENOSYS;
+}
+
+int show_available_vars(struct perf_probe_event *pevs __unused,
+                        int npevs __unused, int max_vls __unused,
+                        const char *module __unused, bool externs __unused)
+{
+        pr_warning("Debuginfo-analysis is not supported.\n");
+        return -ENOSYS;
+}
+#endif
+
+static int parse_line_num(char **ptr, int *val, const char *what)
+{
+        const char *start = *ptr;
+
+        errno = 0;
+        *val = strtol(*ptr, ptr, 0);
+        if (errno || *ptr == start) {
+                semantic_error("'%s' is not a valid number.\n", what);
+                return -EINVAL;
+        }
+        return 0;
+}
+
+/*
+ * Stuff 'lr' according to the line range described by 'arg'.
+ * The line range syntax is described by:
+ *
+ *         SRC[:SLN[+NUM|-ELN]]
+ *         FNC[:SLN[+NUM|-ELN]]
+ */
+int parse_line_range_desc(const char *arg, struct line_range *lr)
+{
+        char *range, *name = strdup(arg);
+        int err;
+
+        if (!name)
+                return -ENOMEM;
+
+        lr->start = 0;
+        lr->end = INT_MAX;
+
+        range = strchr(name, ':');
+        if (range) {
+                *range++ = '\0';
+
+                err = parse_line_num(&range, &lr->start, "start line");
+                if (err)
+                        goto err;
+
+                if (*range == '+' || *range == '-') {
+                        const char c = *range++;
+
+                        err = parse_line_num(&range, &lr->end, "end line");
+                        if (err)
+                                goto err;
+
+                        if (c == '+') {
+                                lr->end += lr->start;
+                                /*
+                                 * Adjust the number of lines here.
+                                 * If the number of lines == 1, the
+                                 * the end of line should be equal to
+                                 * the start of line.
+                                 */
+                                lr->end--;
+                        }
+                }
+
+                pr_debug("Line range is %d to %d\n", lr->start, lr->end);
+
+                err = -EINVAL;
+                if (lr->start > lr->end) {
+                        semantic_error("Start line must be smaller"
+                                       " than end line.\n");
+                        goto err;
+                }
+                if (*range != '\0') {
+                        semantic_error("Tailing with invalid str '%s'.\n", range);
+                        goto err;
+                }
+        }
+
+        if (strchr(name, '.'))
+                lr->file = name;
+        else
+                lr->function = name;
+
+        return 0;
+err:
+        free(name);
+        return err;
+}
+
+/* Check the name is good for event/group */
+static bool check_event_name(const char *name)
+{
+        if (!isalpha(*name) && *name != '_')
+                return false;
+        while (*++name != '\0') {
+                if (!isalpha(*name) && !isdigit(*name) && *name != '_')
+                        return false;
+        }
+        return true;
+}
+
+/* Parse probepoint definition. */
+static int parse_perf_probe_point(char *arg, struct perf_probe_event *pev)
+{
+        struct perf_probe_point *pp = &pev->point;
+        char *ptr, *tmp;
+        char c, nc = 0;
+        /*
+         * <Syntax>
+         * perf probe [EVENT=]SRC[:LN|;PTN]
+         * perf probe [EVENT=]FUNC[@SRC][+OFFS|%return|:LN|;PAT]
+         *
+         * TODO:Group name support
+         */
+
+        ptr = strpbrk(arg, ";=@+%");
+        if (ptr && *ptr == '=') {       /* Event name */
+                *ptr = '\0';
+                tmp = ptr + 1;
+                if (strchr(arg, ':')) {
+                        semantic_error("Group name is not supported yet.\n");
+                        return -ENOTSUP;
+                }
+                if (!check_event_name(arg)) {
+                        semantic_error("%s is bad for event name -it must "
+                                       "follow C symbol-naming rule.\n", arg);
+                        return -EINVAL;
+                }
+                pev->event = strdup(arg);
+                if (pev->event == NULL)
+                        return -ENOMEM;
+                pev->group = NULL;
+                arg = tmp;
+        }
+
+        ptr = strpbrk(arg, ";:+@%");
+        if (ptr) {
+                nc = *ptr;
+                *ptr++ = '\0';
+        }
+
+        tmp = strdup(arg);
+        if (tmp == NULL)
+                return -ENOMEM;
+
+        /* Check arg is function or file and copy it */
+        if (strchr(tmp, '.'))   /* File */
+                pp->file = tmp;
+        else                    /* Function */
+                pp->function = tmp;
+
+        /* Parse other options */
+        while (ptr) {
+                arg = ptr;
+                c = nc;
+                if (c == ';') { /* Lazy pattern must be the last part */
+                        pp->lazy_line = strdup(arg);
+                        if (pp->lazy_line == NULL)
+                                return -ENOMEM;
+                        break;
+                }
+                ptr = strpbrk(arg, ";:+@%");
+                if (ptr) {
+                        nc = *ptr;
+                        *ptr++ = '\0';
+                }
+                switch (c) {
+                case ':':       /* Line number */
+                        pp->line = strtoul(arg, &tmp, 0);
+                        if (*tmp != '\0') {
+                                semantic_error("There is non-digit char"
+                                               " in line number.\n");
+                                return -EINVAL;
+                        }
+                        break;
+                case '+':       /* Byte offset from a symbol */
+                        pp->offset = strtoul(arg, &tmp, 0);
+                        if (*tmp != '\0') {
+                                semantic_error("There is non-digit character"
+                                                " in offset.\n");
+                                return -EINVAL;
+                        }
+                        break;
+                case '@':       /* File name */
+                        if (pp->file) {
+                                semantic_error("SRC@SRC is not allowed.\n");
+                                return -EINVAL;
+                        }
+                        pp->file = strdup(arg);
+                        if (pp->file == NULL)
+                                return -ENOMEM;
+                        break;
+                case '%':       /* Probe places */
+                        if (strcmp(arg, "return") == 0) {
+                                pp->retprobe = 1;
+                        } else {        /* Others not supported yet */
+                                semantic_error("%%%s is not supported.\n", arg);
+                                return -ENOTSUP;
+                        }
+                        break;
+                default:        /* Buggy case */
+                        pr_err("This program has a bug at %s:%d.\n",
+                                __FILE__, __LINE__);
+                        return -ENOTSUP;
+                        break;
+                }
+        }
+
+        /* Exclusion check */
+        if (pp->lazy_line && pp->line) {
+                semantic_error("Lazy pattern can't be used with"
+                               " line number.\n");
+                return -EINVAL;
+        }
+
+        if (pp->lazy_line && pp->offset) {
+                semantic_error("Lazy pattern can't be used with offset.\n");
+                return -EINVAL;
+        }
+
+        if (pp->line && pp->offset) {
+                semantic_error("Offset can't be used with line number.\n");
+                return -EINVAL;
+        }
+
+        if (!pp->line && !pp->lazy_line && pp->file && !pp->function) {
+                semantic_error("File always requires line number or "
+                               "lazy pattern.\n");
+                return -EINVAL;
+        }
+
+        if (pp->offset && !pp->function) {
+                semantic_error("Offset requires an entry function.\n");
+                return -EINVAL;
+        }
+
+        if (pp->retprobe && !pp->function) {
+                semantic_error("Return probe requires an entry function.\n");
+                return -EINVAL;
+        }
+
+        if ((pp->offset || pp->line || pp->lazy_line) && pp->retprobe) {
+                semantic_error("Offset/Line/Lazy pattern can't be used with "
+                               "return probe.\n");
+                return -EINVAL;
+        }
+
+        pr_debug("symbol:%s file:%s line:%d offset:%lu return:%d lazy:%s\n",
+                 pp->function, pp->file, pp->line, pp->offset, pp->retprobe,
+                 pp->lazy_line);
+        return 0;
+}
+
+/* Parse perf-probe event argument */
+static int parse_perf_probe_arg(char *str, struct perf_probe_arg *arg)
+{
+        char *tmp, *goodname;
+        struct perf_probe_arg_field **fieldp;
+
+        pr_debug("parsing arg: %s into ", str);
+
+        tmp = strchr(str, '=');
+        if (tmp) {
+                arg->name = strndup(str, tmp - str);
+                if (arg->name == NULL)
+                        return -ENOMEM;
+                pr_debug("name:%s ", arg->name);
+                str = tmp + 1;
+        }
+
+        tmp = strchr(str, ':');
+        if (tmp) {      /* Type setting */
+                *tmp = '\0';
+                arg->type = strdup(tmp + 1);
+                if (arg->type == NULL)
+                        return -ENOMEM;
+                pr_debug("type:%s ", arg->type);
+        }
+
+        tmp = strpbrk(str, "-.[");
+        if (!is_c_varname(str) || !tmp) {
+                /* A variable, register, symbol or special value */
+                arg->var = strdup(str);
+                if (arg->var == NULL)
+                        return -ENOMEM;
+                pr_debug("%s\n", arg->var);
+                return 0;
+        }
+
+        /* Structure fields or array element */
+        arg->var = strndup(str, tmp - str);
+        if (arg->var == NULL)
+                return -ENOMEM;
+        goodname = arg->var;
+        pr_debug("%s, ", arg->var);
+        fieldp = &arg->field;
+
+        do {
+                *fieldp = zalloc(sizeof(struct perf_probe_arg_field));
+                if (*fieldp == NULL)
+                        return -ENOMEM;
+                if (*tmp == '[') {      /* Array */
+                        str = tmp;
+                        (*fieldp)->index = strtol(str + 1, &tmp, 0);
+                        (*fieldp)->ref = true;
+                        if (*tmp != ']' || tmp == str + 1) {
+                                semantic_error("Array index must be a"
+                                                " number.\n");
+                                return -EINVAL;
+                        }
+                        tmp++;
+                        if (*tmp == '\0')
+                                tmp = NULL;
+                } else {                /* Structure */
+                        if (*tmp == '.') {
+                                str = tmp + 1;
+                                (*fieldp)->ref = false;
+                        } else if (tmp[1] == '>') {
+                                str = tmp + 2;
+                                (*fieldp)->ref = true;
+                        } else {
+                                semantic_error("Argument parse error: %s\n",
+                                               str);
+                                return -EINVAL;
+                        }
+                        tmp = strpbrk(str, "-.[");
+                }
+                if (tmp) {
+                        (*fieldp)->name = strndup(str, tmp - str);
+                        if ((*fieldp)->name == NULL)
+                                return -ENOMEM;
+                        if (*str != '[')
+                                goodname = (*fieldp)->name;
+                        pr_debug("%s(%d), ", (*fieldp)->name, (*fieldp)->ref);
+                        fieldp = &(*fieldp)->next;
+                }
+        } while (tmp);
+        (*fieldp)->name = strdup(str);
+        if ((*fieldp)->name == NULL)
+                return -ENOMEM;
+        if (*str != '[')
+                goodname = (*fieldp)->name;
+        pr_debug("%s(%d)\n", (*fieldp)->name, (*fieldp)->ref);
+
+        /* If no name is specified, set the last field name (not array index)*/
+        if (!arg->name) {
+                arg->name = strdup(goodname);
+                if (arg->name == NULL)
+                        return -ENOMEM;
+        }
+        return 0;
+}
+
+/* Parse perf-probe event command */
+int parse_perf_probe_command(const char *cmd, struct perf_probe_event *pev)
+{
+        char **argv;
+        int argc, i, ret = 0;
+
+        argv = argv_split(cmd, &argc);
+        if (!argv) {
+                pr_debug("Failed to split arguments.\n");
+                return -ENOMEM;
+        }
+        if (argc - 1 > MAX_PROBE_ARGS) {
+                semantic_error("Too many probe arguments (%d).\n", argc - 1);
+                ret = -ERANGE;
+                goto out;
+        }
+        /* Parse probe point */
+        ret = parse_perf_probe_point(argv[0], pev);
+        if (ret < 0)
+                goto out;
+
+        /* Copy arguments and ensure return probe has no C argument */
+        pev->nargs = argc - 1;
+        pev->args = zalloc(sizeof(struct perf_probe_arg) * pev->nargs);
+        if (pev->args == NULL) {
+                ret = -ENOMEM;
+                goto out;
+        }
+        for (i = 0; i < pev->nargs && ret >= 0; i++) {
+                ret = parse_perf_probe_arg(argv[i + 1], &pev->args[i]);
+                if (ret >= 0 &&
+                    is_c_varname(pev->args[i].var) && pev->point.retprobe) {
+                        semantic_error("You can't specify local variable for"
+                                       " kretprobe.\n");
+                        ret = -EINVAL;
+                }
+        }
+out:
+        argv_free(argv);
+
+        return ret;
+}
+
+/* Return true if this perf_probe_event requires debuginfo */
+bool perf_probe_event_need_dwarf(struct perf_probe_event *pev)
+{
+        int i;
+
+        if (pev->point.file || pev->point.line || pev->point.lazy_line)
+                return true;
+
+        for (i = 0; i < pev->nargs; i++)
+                if (is_c_varname(pev->args[i].var))
+                        return true;
+
+        return false;
+}
+
+/* Parse probe_events event into struct probe_point */
+static int parse_probe_trace_command(const char *cmd,
+                                        struct probe_trace_event *tev)
+{
+        struct probe_trace_point *tp = &tev->point;
+        char pr;
+        char *p;
+        int ret, i, argc;
+        char **argv;
+
+        pr_debug("Parsing probe_events: %s\n", cmd);
+        argv = argv_split(cmd, &argc);
+        if (!argv) {
+                pr_debug("Failed to split arguments.\n");
+                return -ENOMEM;
+        }
+        if (argc < 2) {
+                semantic_error("Too few probe arguments.\n");
+                ret = -ERANGE;
+                goto out;
+        }
+
+        /* Scan event and group name. */
+        ret = sscanf(argv[0], "%c:%a[^/ \t]/%a[^ \t]",
+                     &pr, (float *)(void *)&tev->group,
+                     (float *)(void *)&tev->event);
+        if (ret != 3) {
+                semantic_error("Failed to parse event name: %s\n", argv[0]);
+                ret = -EINVAL;
+                goto out;
+        }
+        pr_debug("Group:%s Event:%s probe:%c\n", tev->group, tev->event, pr);
+
+        tp->retprobe = (pr == 'r');
+
+        /* Scan function name and offset */
+        ret = sscanf(argv[1], "%a[^+]+%lu", (float *)(void *)&tp->symbol,
+                     &tp->offset);
+        if (ret == 1)
+                tp->offset = 0;
+
+        tev->nargs = argc - 2;
+        tev->args = zalloc(sizeof(struct probe_trace_arg) * tev->nargs);
+        if (tev->args == NULL) {
+                ret = -ENOMEM;
+                goto out;
+        }
+        for (i = 0; i < tev->nargs; i++) {
+                p = strchr(argv[i + 2], '=');
+                if (p)  /* We don't need which register is assigned. */
+                        *p++ = '\0';
+                else
+                        p = argv[i + 2];
+                tev->args[i].name = strdup(argv[i + 2]);
+                /* TODO: parse regs and offset */
+                tev->args[i].value = strdup(p);
+                if (tev->args[i].name == NULL || tev->args[i].value == NULL) {
+                        ret = -ENOMEM;
+                        goto out;
+                }
+        }
+        ret = 0;
+out:
+        argv_free(argv);
+        return ret;
+}
+
+/* Compose only probe arg */
+int synthesize_perf_probe_arg(struct perf_probe_arg *pa, char *buf, size_t len)
+{
+        struct perf_probe_arg_field *field = pa->field;
+        int ret;
+        char *tmp = buf;
+
+        if (pa->name && pa->var)
+                ret = e_snprintf(tmp, len, "%s=%s", pa->name, pa->var);
+        else
+                ret = e_snprintf(tmp, len, "%s", pa->name ? pa->name : pa->var);
+        if (ret <= 0)
+                goto error;
+        tmp += ret;
+        len -= ret;
+
+        while (field) {
+                if (field->name[0] == '[')
+                        ret = e_snprintf(tmp, len, "%s", field->name);
+                else
+                        ret = e_snprintf(tmp, len, "%s%s",
+                                         field->ref ? "->" : ".", field->name);
+                if (ret <= 0)
+                        goto error;
+                tmp += ret;
+                len -= ret;
+                field = field->next;
+        }
+
+        if (pa->type) {
+                ret = e_snprintf(tmp, len, ":%s", pa->type);
+                if (ret <= 0)
+                        goto error;
+                tmp += ret;
+                len -= ret;
+        }
+
+        return tmp - buf;
+error:
+        pr_debug("Failed to synthesize perf probe argument: %s\n",
+                 strerror(-ret));
+        return ret;
+}
+
+/* Compose only probe point (not argument) */
+static char *synthesize_perf_probe_point(struct perf_probe_point *pp)
+{
+        char *buf, *tmp;
+        char offs[32] = "", line[32] = "", file[32] = "";
+        int ret, len;
+
+        buf = zalloc(MAX_CMDLEN);
+        if (buf == NULL) {
+                ret = -ENOMEM;
+                goto error;
+        }
+        if (pp->offset) {
+                ret = e_snprintf(offs, 32, "+%lu", pp->offset);
+                if (ret <= 0)
+                        goto error;
+        }
+        if (pp->line) {
+                ret = e_snprintf(line, 32, ":%d", pp->line);
+                if (ret <= 0)
+                        goto error;
+        }
+        if (pp->file) {
+                tmp = pp->file;
+                len = strlen(tmp);
+                if (len > 30) {
+                        tmp = strchr(pp->file + len - 30, '/');
+                        tmp = tmp ? tmp + 1 : pp->file + len - 30;
+                }
+                ret = e_snprintf(file, 32, "@%s", tmp);
+                if (ret <= 0)
+                        goto error;
+        }
+
+        if (pp->function)
+                ret = e_snprintf(buf, MAX_CMDLEN, "%s%s%s%s%s", pp->function,
+                                 offs, pp->retprobe ? "%return" : "", line,
+                                 file);
+        else
+                ret = e_snprintf(buf, MAX_CMDLEN, "%s%s", file, line);
+        if (ret <= 0)
+                goto error;
+
+        return buf;
+error:
+        pr_debug("Failed to synthesize perf probe point: %s\n",
+                 strerror(-ret));
+        if (buf)
+                free(buf);
+        return NULL;
+}
+
+#if 0
+char *synthesize_perf_probe_command(struct perf_probe_event *pev)
+{
+        char *buf;
+        int i, len, ret;
+
+        buf = synthesize_perf_probe_point(&pev->point);
+        if (!buf)
+                return NULL;
+
+        len = strlen(buf);
+        for (i = 0; i < pev->nargs; i++) {
+                ret = e_snprintf(&buf[len], MAX_CMDLEN - len, " %s",
+                                 pev->args[i].name);
+                if (ret <= 0) {
+                        free(buf);
+                        return NULL;
+                }
+                len += ret;
+        }
+
+        return buf;
+}
+#endif
+
+static int __synthesize_probe_trace_arg_ref(struct probe_trace_arg_ref *ref,
+                                             char **buf, size_t *buflen,
+                                             int depth)
+{
+        int ret;
+        if (ref->next) {
+                depth = __synthesize_probe_trace_arg_ref(ref->next, buf,
+                                                         buflen, depth + 1);
+                if (depth < 0)
+                        goto out;
+        }
+
+        ret = e_snprintf(*buf, *buflen, "%+ld(", ref->offset);
+        if (ret < 0)
+                depth = ret;
+        else {
+                *buf += ret;
+                *buflen -= ret;
+        }
+out:
+        return depth;
+
+}
+
+static int synthesize_probe_trace_arg(struct probe_trace_arg *arg,
+                                       char *buf, size_t buflen)
+{
+        struct probe_trace_arg_ref *ref = arg->ref;
+        int ret, depth = 0;
+        char *tmp = buf;
+
+        /* Argument name or separator */
+        if (arg->name)
+                ret = e_snprintf(buf, buflen, " %s=", arg->name);
+        else
+                ret = e_snprintf(buf, buflen, " ");
+        if (ret < 0)
+                return ret;
+        buf += ret;
+        buflen -= ret;
+
+        /* Special case: @XXX */
+        if (arg->value[0] == '@' && arg->ref)
+                        ref = ref->next;
+
+        /* Dereferencing arguments */
+        if (ref) {
+                depth = __synthesize_probe_trace_arg_ref(ref, &buf,
+                                                          &buflen, 1);
+                if (depth < 0)
+                        return depth;
+        }
+
+        /* Print argument value */
+        if (arg->value[0] == '@' && arg->ref)
+                ret = e_snprintf(buf, buflen, "%s%+ld", arg->value,
+                                 arg->ref->offset);
+        else
+                ret = e_snprintf(buf, buflen, "%s", arg->value);
+        if (ret < 0)
+                return ret;
+        buf += ret;
+        buflen -= ret;
+
+        /* Closing */
+        while (depth--) {
+                ret = e_snprintf(buf, buflen, ")");
+                if (ret < 0)
+                        return ret;
+                buf += ret;
+                buflen -= ret;
+        }
+        /* Print argument type */
+        if (arg->type) {
+                ret = e_snprintf(buf, buflen, ":%s", arg->type);
+                if (ret <= 0)
+                        return ret;
+                buf += ret;
+        }
+
+        return buf - tmp;
+}
+
+char *synthesize_probe_trace_command(struct probe_trace_event *tev)
+{
+        struct probe_trace_point *tp = &tev->point;
+        char *buf;
+        int i, len, ret;
+
+        buf = zalloc(MAX_CMDLEN);
+        if (buf == NULL)
+                return NULL;
+
+        len = e_snprintf(buf, MAX_CMDLEN, "%c:%s/%s %s+%lu",
+                         tp->retprobe ? 'r' : 'p',
+                         tev->group, tev->event,
+                         tp->symbol, tp->offset);
+        if (len <= 0)
+                goto error;
+
+        for (i = 0; i < tev->nargs; i++) {
+                ret = synthesize_probe_trace_arg(&tev->args[i], buf + len,
+                                                  MAX_CMDLEN - len);
+                if (ret <= 0)
+                        goto error;
+                len += ret;
+        }
+
+        return buf;
+error:
+        free(buf);
+        return NULL;
+}
+
+static int convert_to_perf_probe_event(struct probe_trace_event *tev,
+                                       struct perf_probe_event *pev)
+{
+        char buf[64] = "";
+        int i, ret;
+
+        /* Convert event/group name */
+        pev->event = strdup(tev->event);
+        pev->group = strdup(tev->group);
+        if (pev->event == NULL || pev->group == NULL)
+                return -ENOMEM;
+
+        /* Convert trace_point to probe_point */
+        ret = kprobe_convert_to_perf_probe(&tev->point, &pev->point);
+        if (ret < 0)
+                return ret;
+
+        /* Convert trace_arg to probe_arg */
+        pev->nargs = tev->nargs;
+        pev->args = zalloc(sizeof(struct perf_probe_arg) * pev->nargs);
+        if (pev->args == NULL)
+                return -ENOMEM;
+        for (i = 0; i < tev->nargs && ret >= 0; i++) {
+                if (tev->args[i].name)
+                        pev->args[i].name = strdup(tev->args[i].name);
+                else {
+                        ret = synthesize_probe_trace_arg(&tev->args[i],
+                                                          buf, 64);
+                        pev->args[i].name = strdup(buf);
+                }
+                if (pev->args[i].name == NULL && ret >= 0)
+                        ret = -ENOMEM;
+        }
+
+        if (ret < 0)
+                clear_perf_probe_event(pev);
+
+        return ret;
+}
+
+void clear_perf_probe_event(struct perf_probe_event *pev)
+{
+        struct perf_probe_point *pp = &pev->point;
+        struct perf_probe_arg_field *field, *next;
+        int i;
+
+        if (pev->event)
+                free(pev->event);
+        if (pev->group)
+                free(pev->group);
+        if (pp->file)
+                free(pp->file);
+        if (pp->function)
+                free(pp->function);
+        if (pp->lazy_line)
+                free(pp->lazy_line);
+        for (i = 0; i < pev->nargs; i++) {
+                if (pev->args[i].name)
+                        free(pev->args[i].name);
+                if (pev->args[i].var)
+                        free(pev->args[i].var);
+                if (pev->args[i].type)
+                        free(pev->args[i].type);
+                field = pev->args[i].field;
+                while (field) {
+                        next = field->next;
+                        if (field->name)
+                                free(field->name);
+                        free(field);
+                        field = next;
+                }
+        }
+        if (pev->args)
+                free(pev->args);
+        memset(pev, 0, sizeof(*pev));
+}
+
+static void clear_probe_trace_event(struct probe_trace_event *tev)
+{
+        struct probe_trace_arg_ref *ref, *next;
+        int i;
+
+        if (tev->event)
+                free(tev->event);
+        if (tev->group)
+                free(tev->group);
+        if (tev->point.symbol)
+                free(tev->point.symbol);
+        for (i = 0; i < tev->nargs; i++) {
+                if (tev->args[i].name)
+                        free(tev->args[i].name);
+                if (tev->args[i].value)
+                        free(tev->args[i].value);
+                if (tev->args[i].type)
+                        free(tev->args[i].type);
+                ref = tev->args[i].ref;
+                while (ref) {
+                        next = ref->next;
+                        free(ref);
+                        ref = next;
+                }
+        }
+        if (tev->args)
+                free(tev->args);
+        memset(tev, 0, sizeof(*tev));
+}
+
+static int open_kprobe_events(bool readwrite)
+{
+        char buf[PATH_MAX];
+        const char *__debugfs;
+        int ret;
+
+        __debugfs = debugfs_find_mountpoint();
+        if (__debugfs == NULL) {
+                pr_warning("Debugfs is not mounted.\n");
+                return -ENOENT;
+        }
+
+        ret = e_snprintf(buf, PATH_MAX, "%stracing/kprobe_events", __debugfs);
+        if (ret >= 0) {
+                pr_debug("Opening %s write=%d\n", buf, readwrite);
+                if (readwrite && !probe_event_dry_run)
+                        ret = open(buf, O_RDWR, O_APPEND);
+                else
+                        ret = open(buf, O_RDONLY, 0);
+        }
+
+        if (ret < 0) {
+                if (errno == ENOENT)
+                        pr_warning("kprobe_events file does not exist - please"
+                                 " rebuild kernel with CONFIG_KPROBE_EVENT.\n");
+                else
+                        pr_warning("Failed to open kprobe_events file: %s\n",
+                                   strerror(errno));
+        }
+        return ret;
+}
+
+/* Get raw string list of current kprobe_events */
+static struct strlist *get_probe_trace_command_rawlist(int fd)
+{
+        int ret, idx;
+        FILE *fp;
+        char buf[MAX_CMDLEN];
+        char *p;
+        struct strlist *sl;
+
+        sl = strlist__new(true, NULL);
+
+        fp = fdopen(dup(fd), "r");
+        while (!feof(fp)) {
+                p = fgets(buf, MAX_CMDLEN, fp);
+                if (!p)
+                        break;
+
+                idx = strlen(p) - 1;
+                if (p[idx] == '\n')
+                        p[idx] = '\0';
+                ret = strlist__add(sl, buf);
+                if (ret < 0) {
+                        pr_debug("strlist__add failed: %s\n", strerror(-ret));
+                        strlist__delete(sl);
+                        return NULL;
+                }
+        }
+        fclose(fp);
+
+        return sl;
+}
+
+/* Show an event */
+static int show_perf_probe_event(struct perf_probe_event *pev)
+{
+        int i, ret;
+        char buf[128];
+        char *place;
+
+        /* Synthesize only event probe point */
+        place = synthesize_perf_probe_point(&pev->point);
+        if (!place)
+                return -EINVAL;
+
+        ret = e_snprintf(buf, 128, "%s:%s", pev->group, pev->event);
+        if (ret < 0)
+                return ret;
+
+        printf("  %-20s (on %s", buf, place);
+
+        if (pev->nargs > 0) {
+                printf(" with");
+                for (i = 0; i < pev->nargs; i++) {
+                        ret = synthesize_perf_probe_arg(&pev->args[i],
+                                                        buf, 128);
+                        if (ret < 0)
+                                break;
+                        printf(" %s", buf);
+                }
+        }
+        printf(")\n");
+        free(place);
+        return ret;
+}
+
+/* List up current perf-probe events */
+int show_perf_probe_events(void)
+{
+        int fd, ret;
+        struct probe_trace_event tev;
+        struct perf_probe_event pev;
+        struct strlist *rawlist;
+        struct str_node *ent;
+
+        setup_pager();
+        ret = init_vmlinux();
+        if (ret < 0)
+                return ret;
+
+        memset(&tev, 0, sizeof(tev));
+        memset(&pev, 0, sizeof(pev));
+
+        fd = open_kprobe_events(false);
+        if (fd < 0)
+                return fd;
+
+        rawlist = get_probe_trace_command_rawlist(fd);
+        close(fd);
+        if (!rawlist)
+                return -ENOENT;
+
+        strlist__for_each(ent, rawlist) {
+                ret = parse_probe_trace_command(ent->s, &tev);
+                if (ret >= 0) {
+                        ret = convert_to_perf_probe_event(&tev, &pev);
+                        if (ret >= 0)
+                                ret = show_perf_probe_event(&pev);
+                }
+                clear_perf_probe_event(&pev);
+                clear_probe_trace_event(&tev);
+                if (ret < 0)
+                        break;
+        }
+        strlist__delete(rawlist);
+
+        return ret;
+}
+
+/* Get current perf-probe event names */
+static struct strlist *get_probe_trace_event_names(int fd, bool include_group)
+{
+        char buf[128];
+        struct strlist *sl, *rawlist;
+        struct str_node *ent;
+        struct probe_trace_event tev;
+        int ret = 0;
+
+        memset(&tev, 0, sizeof(tev));
+        rawlist = get_probe_trace_command_rawlist(fd);
+        sl = strlist__new(true, NULL);
+        strlist__for_each(ent, rawlist) {
+                ret = parse_probe_trace_command(ent->s, &tev);
+                if (ret < 0)
+                        break;
+                if (include_group) {
+                        ret = e_snprintf(buf, 128, "%s:%s", tev.group,
+                                        tev.event);
+                        if (ret >= 0)
+                                ret = strlist__add(sl, buf);
+                } else
+                        ret = strlist__add(sl, tev.event);
+                clear_probe_trace_event(&tev);
+                if (ret < 0)
+                        break;
+        }
+        strlist__delete(rawlist);
+
+        if (ret < 0) {
+                strlist__delete(sl);
+                return NULL;
+        }
+        return sl;
+}
+
+static int write_probe_trace_event(int fd, struct probe_trace_event *tev)
+{
+        int ret = 0;
+        char *buf = synthesize_probe_trace_command(tev);
+
+        if (!buf) {
+                pr_debug("Failed to synthesize probe trace event.\n");
+                return -EINVAL;
+        }
+
+        pr_debug("Writing event: %s\n", buf);
+        if (!probe_event_dry_run) {
+                ret = write(fd, buf, strlen(buf));
+                if (ret <= 0)
+                        pr_warning("Failed to write event: %s\n",
+                                   strerror(errno));
+        }
+        free(buf);
+        return ret;
+}
+
+static int get_new_event_name(char *buf, size_t len, const char *base,
+                              struct strlist *namelist, bool allow_suffix)
+{
+        int i, ret;
+
+        /* Try no suffix */
+        ret = e_snprintf(buf, len, "%s", base);
+        if (ret < 0) {
+                pr_debug("snprintf() failed: %s\n", strerror(-ret));
+                return ret;
+        }
+        if (!strlist__has_entry(namelist, buf))
+                return 0;
+
+        if (!allow_suffix) {
+                pr_warning("Error: event \"%s\" already exists. "
+                           "(Use -f to force duplicates.)\n", base);
+                return -EEXIST;
+        }
+
+        /* Try to add suffix */
+        for (i = 1; i < MAX_EVENT_INDEX; i++) {
+                ret = e_snprintf(buf, len, "%s_%d", base, i);
+                if (ret < 0) {
+                        pr_debug("snprintf() failed: %s\n", strerror(-ret));
+                        return ret;
+                }
+                if (!strlist__has_entry(namelist, buf))
+                        break;
+        }
+        if (i == MAX_EVENT_INDEX) {
+                pr_warning("Too many events are on the same function.\n");
+                ret = -ERANGE;
+        }
+
+        return ret;
+}
+
+static int __add_probe_trace_events(struct perf_probe_event *pev,
+                                     struct probe_trace_event *tevs,
+                                     int ntevs, bool allow_suffix)
+{
+        int i, fd, ret;
+        struct probe_trace_event *tev = NULL;
+        char buf[64];
+        const char *event, *group;
+        struct strlist *namelist;
+
+        fd = open_kprobe_events(true);
+        if (fd < 0)
+                return fd;
+        /* Get current event names */
+        namelist = get_probe_trace_event_names(fd, false);
+        if (!namelist) {
+                pr_debug("Failed to get current event list.\n");
+                return -EIO;
+        }
+
+        ret = 0;
+        printf("Add new event%s\n", (ntevs > 1) ? "s:" : ":");
+        for (i = 0; i < ntevs; i++) {
+                tev = &tevs[i];
+                if (pev->event)
+                        event = pev->event;
+                else
+                        if (pev->point.function)
+                                event = pev->point.function;
+                        else
+                                event = tev->point.symbol;
+                if (pev->group)
+                        group = pev->group;
+                else
+                        group = PERFPROBE_GROUP;
+
+                /* Get an unused new event name */
+                ret = get_new_event_name(buf, 64, event,
+                                         namelist, allow_suffix);
+                if (ret < 0)
+                        break;
+                event = buf;
+
+                tev->event = strdup(event);
+                tev->group = strdup(group);
+                if (tev->event == NULL || tev->group == NULL) {
+                        ret = -ENOMEM;
+                        break;
+                }
+                ret = write_probe_trace_event(fd, tev);
+                if (ret < 0)
+                        break;
+                /* Add added event name to namelist */
+                strlist__add(namelist, event);
+
+                /* Trick here - save current event/group */
+                event = pev->event;
+                group = pev->group;
+                pev->event = tev->event;
+                pev->group = tev->group;
+                show_perf_probe_event(pev);
+                /* Trick here - restore current event/group */
+                pev->event = (char *)event;
+                pev->group = (char *)group;
+
+                /*
+                 * Probes after the first probe which comes from same
+                 * user input are always allowed to add suffix, because
+                 * there might be several addresses corresponding to
+                 * one code line.
+                 */
+                allow_suffix = true;
+        }
+
+        if (ret >= 0) {
+                /* Show how to use the event. */
+                printf("\nYou can now use it on all perf tools, such as:\n\n");
+                printf("\tperf record -e %s:%s -aR sleep 1\n\n", tev->group,
+                         tev->event);
+        }
+
+        strlist__delete(namelist);
+        close(fd);
+        return ret;
+}
+
+static int convert_to_probe_trace_events(struct perf_probe_event *pev,
+                                          struct probe_trace_event **tevs,
+                                          int max_tevs, const char *module)
+{
+        struct symbol *sym;
+        int ret = 0, i;
+        struct probe_trace_event *tev;
+
+        /* Convert perf_probe_event with debuginfo */
+        ret = try_to_find_probe_trace_events(pev, tevs, max_tevs, module);
+        if (ret != 0)
+                return ret;
+
+        /* Allocate trace event buffer */
+        tev = *tevs = zalloc(sizeof(struct probe_trace_event));
+        if (tev == NULL)
+                return -ENOMEM;
+
+        /* Copy parameters */
+        tev->point.symbol = strdup(pev->point.function);
+        if (tev->point.symbol == NULL) {
+                ret = -ENOMEM;
+                goto error;
+        }
+        tev->point.offset = pev->point.offset;
+        tev->point.retprobe = pev->point.retprobe;
+        tev->nargs = pev->nargs;
+        if (tev->nargs) {
+                tev->args = zalloc(sizeof(struct probe_trace_arg)
+                                   * tev->nargs);
+                if (tev->args == NULL) {
+                        ret = -ENOMEM;
+                        goto error;
+                }
+                for (i = 0; i < tev->nargs; i++) {
+                        if (pev->args[i].name) {
+                                tev->args[i].name = strdup(pev->args[i].name);
+                                if (tev->args[i].name == NULL) {
+                                        ret = -ENOMEM;
+                                        goto error;
+                                }
+                        }
+                        tev->args[i].value = strdup(pev->args[i].var);
+                        if (tev->args[i].value == NULL) {
+                                ret = -ENOMEM;
+                                goto error;
+                        }
+                        if (pev->args[i].type) {
+                                tev->args[i].type = strdup(pev->args[i].type);
+                                if (tev->args[i].type == NULL) {
+                                        ret = -ENOMEM;
+                                        goto error;
+                                }
+                        }
+                }
+        }
+
+        /* Currently just checking function name from symbol map */
+        sym = __find_kernel_function_by_name(tev->point.symbol, NULL);
+        if (!sym) {
+                pr_warning("Kernel symbol \'%s\' not found.\n",
+                           tev->point.symbol);
+                ret = -ENOENT;
+                goto error;
+        }
+
+        return 1;
+error:
+        clear_probe_trace_event(tev);
+        free(tev);
+        *tevs = NULL;
+        return ret;
+}
+
+struct __event_package {
+        struct perf_probe_event         *pev;
+        struct probe_trace_event        *tevs;
+        int                             ntevs;
+};
+
+int add_perf_probe_events(struct perf_probe_event *pevs, int npevs,
+                          int max_tevs, const char *module, bool force_add)
+{
+        int i, j, ret;
+        struct __event_package *pkgs;
+
+        pkgs = zalloc(sizeof(struct __event_package) * npevs);
+        if (pkgs == NULL)
+                return -ENOMEM;
+
+        /* Init vmlinux path */
+        ret = init_vmlinux();
+        if (ret < 0) {
+                free(pkgs);
+                return ret;
+        }
+
+        /* Loop 1: convert all events */
+        for (i = 0; i < npevs; i++) {
+                pkgs[i].pev = &pevs[i];
+                /* Convert with or without debuginfo */
+                ret  = convert_to_probe_trace_events(pkgs[i].pev,
+                                                     &pkgs[i].tevs,
+                                                     max_tevs,
+                                                     module);
+                if (ret < 0)
+                        goto end;
+                pkgs[i].ntevs = ret;
+        }
+
+        /* Loop 2: add all events */
+        for (i = 0; i < npevs && ret >= 0; i++)
+                ret = __add_probe_trace_events(pkgs[i].pev, pkgs[i].tevs,
+                                                pkgs[i].ntevs, force_add);
+end:
+        /* Loop 3: cleanup and free trace events  */
+        for (i = 0; i < npevs; i++) {
+                for (j = 0; j < pkgs[i].ntevs; j++)
+                        clear_probe_trace_event(&pkgs[i].tevs[j]);
+                free(pkgs[i].tevs);
+        }
+        free(pkgs);
+
+        return ret;
+}
+
+static int __del_trace_probe_event(int fd, struct str_node *ent)
+{
+        char *p;
+        char buf[128];
+        int ret;
+
+        /* Convert from perf-probe event to trace-probe event */
+        ret = e_snprintf(buf, 128, "-:%s", ent->s);
+        if (ret < 0)
+                goto error;
+
+        p = strchr(buf + 2, ':');
+        if (!p) {
+                pr_debug("Internal error: %s should have ':' but not.\n",
+                         ent->s);
+                ret = -ENOTSUP;
+                goto error;
+        }
+        *p = '/';
+
+        pr_debug("Writing event: %s\n", buf);
+        ret = write(fd, buf, strlen(buf));
+        if (ret < 0)
+                goto error;
+
+        printf("Remove event: %s\n", ent->s);
+        return 0;
+error:
+        pr_warning("Failed to delete event: %s\n", strerror(-ret));
+        return ret;
+}
+
+static int del_trace_probe_event(int fd, const char *group,
+                                  const char *event, struct strlist *namelist)
+{
+        char buf[128];
+        struct str_node *ent, *n;
+        int found = 0, ret = 0;
+
+        ret = e_snprintf(buf, 128, "%s:%s", group, event);
+        if (ret < 0) {
+                pr_err("Failed to copy event.\n");
+                return ret;
+        }
+
+        if (strpbrk(buf, "*?")) { /* Glob-exp */
+                strlist__for_each_safe(ent, n, namelist)
+                        if (strglobmatch(ent->s, buf)) {
+                                found++;
+                                ret = __del_trace_probe_event(fd, ent);
+                                if (ret < 0)
+                                        break;
+                                strlist__remove(namelist, ent);
+                        }
+        } else {
+                ent = strlist__find(namelist, buf);
+                if (ent) {
+                        found++;
+                        ret = __del_trace_probe_event(fd, ent);
+                        if (ret >= 0)
+                                strlist__remove(namelist, ent);
+                }
+        }
+        if (found == 0 && ret >= 0)
+                pr_info("Info: Event \"%s\" does not exist.\n", buf);
+
+        return ret;
+}
+
+int del_perf_probe_events(struct strlist *dellist)
+{
+        int fd, ret = 0;
+        const char *group, *event;
+        char *p, *str;
+        struct str_node *ent;
+        struct strlist *namelist;
+
+        fd = open_kprobe_events(true);
+        if (fd < 0)
+                return fd;
+
+        /* Get current event names */
+        namelist = get_probe_trace_event_names(fd, true);
+        if (namelist == NULL)
+                return -EINVAL;
+
+        strlist__for_each(ent, dellist) {
+                str = strdup(ent->s);
+                if (str == NULL) {
+                        ret = -ENOMEM;
+                        break;
+                }
+                pr_debug("Parsing: %s\n", str);
+                p = strchr(str, ':');
+                if (p) {
+                        group = str;
+                        *p = '\0';
+                        event = p + 1;
+                } else {
+                        group = "*";
+                        event = str;
+                }
+                pr_debug("Group: %s, Event: %s\n", group, event);
+                ret = del_trace_probe_event(fd, group, event, namelist);
+                free(str);
+                if (ret < 0)
+                        break;
+        }
+        strlist__delete(namelist);
+        close(fd);
+
+        return ret;
+}
+
diff --git a/tools/perf/util/probe-event.h b/tools/perf/util/probe-event.h
new file mode 100644
index 00000000000..5accbedfea3
--- /dev/null
+++ b/tools/perf/util/probe-event.h
@@ -0,0 +1,135 @@
+#ifndef _PROBE_EVENT_H
+#define _PROBE_EVENT_H
+
+#include <stdbool.h>
+#include "strlist.h"
+
+extern bool probe_event_dry_run;
+
+/* kprobe-tracer tracing point */
+struct probe_trace_point {
+        char            *symbol;        /* Base symbol */
+        unsigned long   offset;         /* Offset from symbol */
+        bool            retprobe;       /* Return probe flag */
+};
+
+/* probe-tracer tracing argument referencing offset */
+struct probe_trace_arg_ref {
+        struct probe_trace_arg_ref      *next;  /* Next reference */
+        long                            offset; /* Offset value */
+};
+
+/* kprobe-tracer tracing argument */
+struct probe_trace_arg {
+        char                            *name;  /* Argument name */
+        char                            *value; /* Base value */
+        char                            *type;  /* Type name */
+        struct probe_trace_arg_ref      *ref;   /* Referencing offset */
+};
+
+/* kprobe-tracer tracing event (point + arg) */
+struct probe_trace_event {
+        char                            *event; /* Event name */
+        char                            *group; /* Group name */
+        struct probe_trace_point        point;  /* Trace point */
+        int                             nargs;  /* Number of args */
+        struct probe_trace_arg          *args;  /* Arguments */
+};
+
+/* Perf probe probing point */
+struct perf_probe_point {
+        char            *file;          /* File path */
+        char            *function;      /* Function name */
+        int             line;           /* Line number */
+        bool            retprobe;       /* Return probe flag */
+        char            *lazy_line;     /* Lazy matching pattern */
+        unsigned long   offset;         /* Offset from function entry */
+};
+
+/* Perf probe probing argument field chain */
+struct perf_probe_arg_field {
+        struct perf_probe_arg_field     *next;  /* Next field */
+        char                            *name;  /* Name of the field */
+        long                            index;  /* Array index number */
+        bool                            ref;    /* Referencing flag */
+};
+
+/* Perf probe probing argument */
+struct perf_probe_arg {
+        char                            *name;  /* Argument name */
+        char                            *var;   /* Variable name */
+        char                            *type;  /* Type name */
+        struct perf_probe_arg_field     *field; /* Structure fields */
+};
+
+/* Perf probe probing event (point + arg) */
+struct perf_probe_event {
+        char                    *event; /* Event name */
+        char                    *group; /* Group name */
+        struct perf_probe_point point;  /* Probe point */
+        int                     nargs;  /* Number of arguments */
+        struct perf_probe_arg   *args;  /* Arguments */
+};
+
+
+/* Line number container */
+struct line_node {
+        struct list_head        list;
+        int                     line;
+};
+
+/* Line range */
+struct line_range {
+        char                    *file;          /* File name */
+        char                    *function;      /* Function name */
+        int                     start;          /* Start line number */
+        int                     end;            /* End line number */
+        int                     offset;         /* Start line offset */
+        char                    *path;          /* Real path name */
+        char                    *comp_dir;      /* Compile directory */
+        struct list_head        line_list;      /* Visible lines */
+};
+
+/* List of variables */
+struct variable_list {
+        struct probe_trace_point        point;  /* Actual probepoint */
+        struct strlist                  *vars;  /* Available variables */
+};
+
+/* Command string to events */
+extern int parse_perf_probe_command(const char *cmd,
+                                    struct perf_probe_event *pev);
+
+/* Events to command string */
+extern char *synthesize_perf_probe_command(struct perf_probe_event *pev);
+extern char *synthesize_probe_trace_command(struct probe_trace_event *tev);
+extern int synthesize_perf_probe_arg(struct perf_probe_arg *pa, char *buf,
+                                     size_t len);
+
+/* Check the perf_probe_event needs debuginfo */
+extern bool perf_probe_event_need_dwarf(struct perf_probe_event *pev);
+
+/* Release event contents */
+extern void clear_perf_probe_event(struct perf_probe_event *pev);
+
+/* Command string to line-range */
+extern int parse_line_range_desc(const char *cmd, struct line_range *lr);
+
+/* Internal use: Return kernel/module path */
+extern const char *kernel_get_module_path(const char *module);
+
+extern int add_perf_probe_events(struct perf_probe_event *pevs, int npevs,
+                                 int max_probe_points, const char *module,
+                                 bool force_add);
+extern int del_perf_probe_events(struct strlist *dellist);
+extern int show_perf_probe_events(void);
+extern int show_line_range(struct line_range *lr, const char *module);
+extern int show_available_vars(struct perf_probe_event *pevs, int npevs,
+                               int max_probe_points, const char *module,
+                               bool externs);
+
+
+/* Maximum index number of event-name postfix */
+#define MAX_EVENT_INDEX 1024
+
+#endif /*_PROBE_EVENT_H */
diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
new file mode 100644
index 00000000000..ab83b6ac5d6
--- /dev/null
+++ b/tools/perf/util/probe-finder.c
@@ -0,0 +1,1861 @@
+/*
+ * probe-finder.c : C expression to kprobe event converter
+ *
+ * Written by Masami Hiramatsu <mhiramat@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+
+#include <sys/utsname.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <errno.h>
+#include <stdio.h>
+#include <unistd.h>
+#include <getopt.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdarg.h>
+#include <ctype.h>
+#include <dwarf-regs.h>
+
+#include "event.h"
+#include "debug.h"
+#include "util.h"
+#include "symbol.h"
+#include "probe-finder.h"
+
+/* Kprobe tracer basic type is up to u64 */
+#define MAX_BASIC_TYPE_BITS     64
+
+/*
+ * Compare the tail of two strings.
+ * Return 0 if whole of either string is same as another's tail part.
+ */
+static int strtailcmp(const char *s1, const char *s2)
+{
+        int i1 = strlen(s1);
+        int i2 = strlen(s2);
+        while (--i1 >= 0 && --i2 >= 0) {
+                if (s1[i1] != s2[i2])
+                        return s1[i1] - s2[i2];
+        }
+        return 0;
+}
+
+/* Line number list operations */
+
+/* Add a line to line number list */
+static int line_list__add_line(struct list_head *head, int line)
+{
+        struct line_node *ln;
+        struct list_head *p;
+
+        /* Reverse search, because new line will be the last one */
+        list_for_each_entry_reverse(ln, head, list) {
+                if (ln->line < line) {
+                        p = &ln->list;
+                        goto found;
+                } else if (ln->line == line)    /* Already exist */
+                        return 1;
+        }
+        /* List is empty, or the smallest entry */
+        p = head;
+found:
+        pr_debug("line list: add a line %u\n", line);
+        ln = zalloc(sizeof(struct line_node));
+        if (ln == NULL)
+                return -ENOMEM;
+        ln->line = line;
+        INIT_LIST_HEAD(&ln->list);
+        list_add(&ln->list, p);
+        return 0;
+}
+
+/* Check if the line in line number list */
+static int line_list__has_line(struct list_head *head, int line)
+{
+        struct line_node *ln;
+
+        /* Reverse search, because new line will be the last one */
+        list_for_each_entry(ln, head, list)
+                if (ln->line == line)
+                        return 1;
+
+        return 0;
+}
+
+/* Init line number list */
+static void line_list__init(struct list_head *head)
+{
+        INIT_LIST_HEAD(head);
+}
+
+/* Free line number list */
+static void line_list__free(struct list_head *head)
+{
+        struct line_node *ln;
+        while (!list_empty(head)) {
+                ln = list_first_entry(head, struct line_node, list);
+                list_del(&ln->list);
+                free(ln);
+        }
+}
+
+/* Dwarf FL wrappers */
+static char *debuginfo_path;    /* Currently dummy */
+
+static const Dwfl_Callbacks offline_callbacks = {
+        .find_debuginfo = dwfl_standard_find_debuginfo,
+        .debuginfo_path = &debuginfo_path,
+
+        .section_address = dwfl_offline_section_address,
+
+        /* We use this table for core files too.  */
+        .find_elf = dwfl_build_id_find_elf,
+};
+
+/* Get a Dwarf from offline image */
+static Dwarf *dwfl_init_offline_dwarf(int fd, Dwfl **dwflp, Dwarf_Addr *bias)
+{
+        Dwfl_Module *mod;
+        Dwarf *dbg = NULL;
+
+        if (!dwflp)
+                return NULL;
+
+        *dwflp = dwfl_begin(&offline_callbacks);
+        if (!*dwflp)
+                return NULL;
+
+        mod = dwfl_report_offline(*dwflp, "", "", fd);
+        if (!mod)
+                goto error;
+
+        dbg = dwfl_module_getdwarf(mod, bias);
+        if (!dbg) {
+error:
+                dwfl_end(*dwflp);
+                *dwflp = NULL;
+        }
+        return dbg;
+}
+
+#if _ELFUTILS_PREREQ(0, 148)
+/* This method is buggy if elfutils is older than 0.148 */
+static int __linux_kernel_find_elf(Dwfl_Module *mod,
+                                   void **userdata,
+                                   const char *module_name,
+                                   Dwarf_Addr base,
+                                   char **file_name, Elf **elfp)
+{
+        int fd;
+        const char *path = kernel_get_module_path(module_name);
+
+        pr_debug2("Use file %s for %s\n", path, module_name);
+        if (path) {
+                fd = open(path, O_RDONLY);
+                if (fd >= 0) {
+                        *file_name = strdup(path);
+                        return fd;
+                }
+        }
+        /* If failed, try to call standard method */
+        return dwfl_linux_kernel_find_elf(mod, userdata, module_name, base,
+                                          file_name, elfp);
+}
+
+static const Dwfl_Callbacks kernel_callbacks = {
+        .find_debuginfo = dwfl_standard_find_debuginfo,
+        .debuginfo_path = &debuginfo_path,
+
+        .find_elf = __linux_kernel_find_elf,
+        .section_address = dwfl_linux_kernel_module_section_address,
+};
+
+/* Get a Dwarf from live kernel image */
+static Dwarf *dwfl_init_live_kernel_dwarf(Dwarf_Addr addr, Dwfl **dwflp,
+                                          Dwarf_Addr *bias)
+{
+        Dwarf *dbg;
+
+        if (!dwflp)
+                return NULL;
+
+        *dwflp = dwfl_begin(&kernel_callbacks);
+        if (!*dwflp)
+                return NULL;
+
+        /* Load the kernel dwarves: Don't care the result here */
+        dwfl_linux_kernel_report_kernel(*dwflp);
+        dwfl_linux_kernel_report_modules(*dwflp);
+
+        dbg = dwfl_addrdwarf(*dwflp, addr, bias);
+        /* Here, check whether we could get a real dwarf */
+        if (!dbg) {
+                pr_debug("Failed to find kernel dwarf at %lx\n",
+                         (unsigned long)addr);
+                dwfl_end(*dwflp);
+                *dwflp = NULL;
+        }
+        return dbg;
+}
+#else
+/* With older elfutils, this just support kernel module... */
+static Dwarf *dwfl_init_live_kernel_dwarf(Dwarf_Addr addr __used, Dwfl **dwflp,
+                                          Dwarf_Addr *bias)
+{
+        int fd;
+        const char *path = kernel_get_module_path("kernel");
+
+        if (!path) {
+                pr_err("Failed to find vmlinux path\n");
+                return NULL;
+        }
+
+        pr_debug2("Use file %s for debuginfo\n", path);
+        fd = open(path, O_RDONLY);
+        if (fd < 0)
+                return NULL;
+
+        return dwfl_init_offline_dwarf(fd, dwflp, bias);
+}
+#endif
+
+/* Dwarf wrappers */
+
+/* Find the realpath of the target file. */
+static const char *cu_find_realpath(Dwarf_Die *cu_die, const char *fname)
+{
+        Dwarf_Files *files;
+        size_t nfiles, i;
+        const char *src = NULL;
+        int ret;
+
+        if (!fname)
+                return NULL;
+
+        ret = dwarf_getsrcfiles(cu_die, &files, &nfiles);
+        if (ret != 0)
+                return NULL;
+
+        for (i = 0; i < nfiles; i++) {
+                src = dwarf_filesrc(files, i, NULL, NULL);
+                if (strtailcmp(src, fname) == 0)
+                        break;
+        }
+        if (i == nfiles)
+                return NULL;
+        return src;
+}
+
+/* Get DW_AT_comp_dir (should be NULL with older gcc) */
+static const char *cu_get_comp_dir(Dwarf_Die *cu_die)
+{
+        Dwarf_Attribute attr;
+        if (dwarf_attr(cu_die, DW_AT_comp_dir, &attr) == NULL)
+                return NULL;
+        return dwarf_formstring(&attr);
+}
+
+/* Compare diename and tname */
+static bool die_compare_name(Dwarf_Die *dw_die, const char *tname)
+{
+        const char *name;
+        name = dwarf_diename(dw_die);
+        return name ? (strcmp(tname, name) == 0) : false;
+}
+
+/* Get type die */
+static Dwarf_Die *die_get_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem)
+{
+        Dwarf_Attribute attr;
+
+        if (dwarf_attr_integrate(vr_die, DW_AT_type, &attr) &&
+            dwarf_formref_die(&attr, die_mem))
+                return die_mem;
+        else
+                return NULL;
+}
+
+/* Get a type die, but skip qualifiers */
+static Dwarf_Die *__die_get_real_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem)
+{
+        int tag;
+
+        do {
+                vr_die = die_get_type(vr_die, die_mem);
+                if (!vr_die)
+                        break;
+                tag = dwarf_tag(vr_die);
+        } while (tag == DW_TAG_const_type ||
+                 tag == DW_TAG_restrict_type ||
+                 tag == DW_TAG_volatile_type ||
+                 tag == DW_TAG_shared_type);
+
+        return vr_die;
+}
+
+/* Get a type die, but skip qualifiers and typedef */
+static Dwarf_Die *die_get_real_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem)
+{
+        do {
+                vr_die = __die_get_real_type(vr_die, die_mem);
+        } while (vr_die && dwarf_tag(vr_die) == DW_TAG_typedef);
+
+        return vr_die;
+}
+
+static bool die_is_signed_type(Dwarf_Die *tp_die)
+{
+        Dwarf_Attribute attr;
+        Dwarf_Word ret;
+
+        if (dwarf_attr(tp_die, DW_AT_encoding, &attr) == NULL ||
+            dwarf_formudata(&attr, &ret) != 0)
+                return false;
+
+        return (ret == DW_ATE_signed_char || ret == DW_ATE_signed ||
+                ret == DW_ATE_signed_fixed);
+}
+
+static int die_get_byte_size(Dwarf_Die *tp_die)
+{
+        Dwarf_Attribute attr;
+        Dwarf_Word ret;
+
+        if (dwarf_attr(tp_die, DW_AT_byte_size, &attr) == NULL ||
+            dwarf_formudata(&attr, &ret) != 0)
+                return 0;
+
+        return (int)ret;
+}
+
+/* Get data_member_location offset */
+static int die_get_data_member_location(Dwarf_Die *mb_die, Dwarf_Word *offs)
+{
+        Dwarf_Attribute attr;
+        Dwarf_Op *expr;
+        size_t nexpr;
+        int ret;
+
+        if (dwarf_attr(mb_die, DW_AT_data_member_location, &attr) == NULL)
+                return -ENOENT;
+
+        if (dwarf_formudata(&attr, offs) != 0) {
+                /* DW_AT_data_member_location should be DW_OP_plus_uconst */
+                ret = dwarf_getlocation(&attr, &expr, &nexpr);
+                if (ret < 0 || nexpr == 0)
+                        return -ENOENT;
+
+                if (expr[0].atom != DW_OP_plus_uconst || nexpr != 1) {
+                        pr_debug("Unable to get offset:Unexpected OP %x (%zd)\n",
+                                 expr[0].atom, nexpr);
+                        return -ENOTSUP;
+                }
+                *offs = (Dwarf_Word)expr[0].number;
+        }
+        return 0;
+}
+
+/* Return values for die_find callbacks */
+enum {
+        DIE_FIND_CB_FOUND = 0,          /* End of Search */
+        DIE_FIND_CB_CHILD = 1,          /* Search only children */
+        DIE_FIND_CB_SIBLING = 2,        /* Search only siblings */
+        DIE_FIND_CB_CONTINUE = 3,       /* Search children and siblings */
+};
+
+/* Search a child die */
+static Dwarf_Die *die_find_child(Dwarf_Die *rt_die,
+                                 int (*callback)(Dwarf_Die *, void *),
+                                 void *data, Dwarf_Die *die_mem)
+{
+        Dwarf_Die child_die;
+        int ret;
+
+        ret = dwarf_child(rt_die, die_mem);
+        if (ret != 0)
+                return NULL;
+
+        do {
+                ret = callback(die_mem, data);
+                if (ret == DIE_FIND_CB_FOUND)
+                        return die_mem;
+
+                if ((ret & DIE_FIND_CB_CHILD) &&
+                    die_find_child(die_mem, callback, data, &child_die)) {
+                        memcpy(die_mem, &child_die, sizeof(Dwarf_Die));
+                        return die_mem;
+                }
+        } while ((ret & DIE_FIND_CB_SIBLING) &&
+                 dwarf_siblingof(die_mem, die_mem) == 0);
+
+        return NULL;
+}
+
+struct __addr_die_search_param {
+        Dwarf_Addr      addr;
+        Dwarf_Die       *die_mem;
+};
+
+static int __die_search_func_cb(Dwarf_Die *fn_die, void *data)
+{
+        struct __addr_die_search_param *ad = data;
+
+        if (dwarf_tag(fn_die) == DW_TAG_subprogram &&
+            dwarf_haspc(fn_die, ad->addr)) {
+                memcpy(ad->die_mem, fn_die, sizeof(Dwarf_Die));
+                return DWARF_CB_ABORT;
+        }
+        return DWARF_CB_OK;
+}
+
+/* Search a real subprogram including this line, */
+static Dwarf_Die *die_find_real_subprogram(Dwarf_Die *cu_die, Dwarf_Addr addr,
+                                           Dwarf_Die *die_mem)
+{
+        struct __addr_die_search_param ad;
+        ad.addr = addr;
+        ad.die_mem = die_mem;
+        /* dwarf_getscopes can't find subprogram. */
+        if (!dwarf_getfuncs(cu_die, __die_search_func_cb, &ad, 0))
+                return NULL;
+        else
+                return die_mem;
+}
+
+/* die_find callback for inline function search */
+static int __die_find_inline_cb(Dwarf_Die *die_mem, void *data)
+{
+        Dwarf_Addr *addr = data;
+
+        if (dwarf_tag(die_mem) == DW_TAG_inlined_subroutine &&
+            dwarf_haspc(die_mem, *addr))
+                return DIE_FIND_CB_FOUND;
+
+        return DIE_FIND_CB_CONTINUE;
+}
+
+/* Similar to dwarf_getfuncs, but returns inlined_subroutine if exists. */
+static Dwarf_Die *die_find_inlinefunc(Dwarf_Die *sp_die, Dwarf_Addr addr,
+                                      Dwarf_Die *die_mem)
+{
+        return die_find_child(sp_die, __die_find_inline_cb, &addr, die_mem);
+}
+
+struct __find_variable_param {
+        const char *name;
+        Dwarf_Addr addr;
+};
+
+static int __die_find_variable_cb(Dwarf_Die *die_mem, void *data)
+{
+        struct __find_variable_param *fvp = data;
+        int tag;
+
+        tag = dwarf_tag(die_mem);
+        if ((tag == DW_TAG_formal_parameter ||
+             tag == DW_TAG_variable) &&
+            die_compare_name(die_mem, fvp->name))
+                return DIE_FIND_CB_FOUND;
+
+        if (dwarf_haspc(die_mem, fvp->addr))
+                return DIE_FIND_CB_CONTINUE;
+        else
+                return DIE_FIND_CB_SIBLING;
+}
+
+/* Find a variable called 'name' at given address */
+static Dwarf_Die *die_find_variable_at(Dwarf_Die *sp_die, const char *name,
+                                       Dwarf_Addr addr, Dwarf_Die *die_mem)
+{
+        struct __find_variable_param fvp = { .name = name, .addr = addr};
+
+        return die_find_child(sp_die, __die_find_variable_cb, (void *)&fvp,
+                              die_mem);
+}
+
+static int __die_find_member_cb(Dwarf_Die *die_mem, void *data)
+{
+        const char *name = data;
+
+        if ((dwarf_tag(die_mem) == DW_TAG_member) &&
+            die_compare_name(die_mem, name))
+                return DIE_FIND_CB_FOUND;
+
+        return DIE_FIND_CB_SIBLING;
+}
+
+/* Find a member called 'name' */
+static Dwarf_Die *die_find_member(Dwarf_Die *st_die, const char *name,
+                                  Dwarf_Die *die_mem)
+{
+        return die_find_child(st_die, __die_find_member_cb, (void *)name,
+                              die_mem);
+}
+
+/* Get the name of given variable DIE */
+static int die_get_typename(Dwarf_Die *vr_die, char *buf, int len)
+{
+        Dwarf_Die type;
+        int tag, ret, ret2;
+        const char *tmp = "";
+
+        if (__die_get_real_type(vr_die, &type) == NULL)
+                return -ENOENT;
+
+        tag = dwarf_tag(&type);
+        if (tag == DW_TAG_array_type || tag == DW_TAG_pointer_type)
+                tmp = "*";
+        else if (tag == DW_TAG_subroutine_type) {
+                /* Function pointer */
+                ret = snprintf(buf, len, "(function_type)");
+                return (ret >= len) ? -E2BIG : ret;
+        } else {
+                if (!dwarf_diename(&type))
+                        return -ENOENT;
+                if (tag == DW_TAG_union_type)
+                        tmp = "union ";
+                else if (tag == DW_TAG_structure_type)
+                        tmp = "struct ";
+                /* Write a base name */
+                ret = snprintf(buf, len, "%s%s", tmp, dwarf_diename(&type));
+                return (ret >= len) ? -E2BIG : ret;
+        }
+        ret = die_get_typename(&type, buf, len);
+        if (ret > 0) {
+                ret2 = snprintf(buf + ret, len - ret, "%s", tmp);
+                ret = (ret2 >= len - ret) ? -E2BIG : ret2 + ret;
+        }
+        return ret;
+}
+
+/* Get the name and type of given variable DIE, stored as "type\tname" */
+static int die_get_varname(Dwarf_Die *vr_die, char *buf, int len)
+{
+        int ret, ret2;
+
+        ret = die_get_typename(vr_die, buf, len);
+        if (ret < 0) {
+                pr_debug("Failed to get type, make it unknown.\n");
+                ret = snprintf(buf, len, "(unknown_type)");
+        }
+        if (ret > 0) {
+                ret2 = snprintf(buf + ret, len - ret, "\t%s",
+                                dwarf_diename(vr_die));
+                ret = (ret2 >= len - ret) ? -E2BIG : ret2 + ret;
+        }
+        return ret;
+}
+
+/*
+ * Probe finder related functions
+ */
+
+static struct probe_trace_arg_ref *alloc_trace_arg_ref(long offs)
+{
+        struct probe_trace_arg_ref *ref;
+        ref = zalloc(sizeof(struct probe_trace_arg_ref));
+        if (ref != NULL)
+                ref->offset = offs;
+        return ref;
+}
+
+/*
+ * Convert a location into trace_arg.
+ * If tvar == NULL, this just checks variable can be converted.
+ */
+static int convert_variable_location(Dwarf_Die *vr_die, Dwarf_Addr addr,
+                                     Dwarf_Op *fb_ops,
+                                     struct probe_trace_arg *tvar)
+{
+        Dwarf_Attribute attr;
+        Dwarf_Op *op;
+        size_t nops;
+        unsigned int regn;
+        Dwarf_Word offs = 0;
+        bool ref = false;
+        const char *regs;
+        int ret;
+
+        if (dwarf_attr(vr_die, DW_AT_external, &attr) != NULL)
+                goto static_var;
+
+        /* TODO: handle more than 1 exprs */
+        if (dwarf_attr(vr_die, DW_AT_location, &attr) == NULL ||
+            dwarf_getlocation_addr(&attr, addr, &op, &nops, 1) <= 0 ||
+            nops == 0) {
+                /* TODO: Support const_value */
+                return -ENOENT;
+        }
+
+        if (op->atom == DW_OP_addr) {
+static_var:
+                if (!tvar)
+                        return 0;
+                /* Static variables on memory (not stack), make @varname */
+                ret = strlen(dwarf_diename(vr_die));
+                tvar->value = zalloc(ret + 2);
+                if (tvar->value == NULL)
+                        return -ENOMEM;
+                snprintf(tvar->value, ret + 2, "@%s", dwarf_diename(vr_die));
+                tvar->ref = alloc_trace_arg_ref((long)offs);
+                if (tvar->ref == NULL)
+                        return -ENOMEM;
+                return 0;
+        }
+
+        /* If this is based on frame buffer, set the offset */
+        if (op->atom == DW_OP_fbreg) {
+                if (fb_ops == NULL)
+                        return -ENOTSUP;
+                ref = true;
+                offs = op->number;
+                op = &fb_ops[0];
+        }
+
+        if (op->atom >= DW_OP_breg0 && op->atom <= DW_OP_breg31) {
+                regn = op->atom - DW_OP_breg0;
+                offs += op->number;
+                ref = true;
+        } else if (op->atom >= DW_OP_reg0 && op->atom <= DW_OP_reg31) {
+                regn = op->atom - DW_OP_reg0;
+        } else if (op->atom == DW_OP_bregx) {
+                regn = op->number;
+                offs += op->number2;
+                ref = true;
+        } else if (op->atom == DW_OP_regx) {
+                regn = op->number;
+        } else {
+                pr_debug("DW_OP %x is not supported.\n", op->atom);
+                return -ENOTSUP;
+        }
+
+        if (!tvar)
+                return 0;
+
+        regs = get_arch_regstr(regn);
+        if (!regs) {
+                /* This should be a bug in DWARF or this tool */
+                pr_warning("Mapping for the register number %u "
+                           "missing on this architecture.\n", regn);
+                return -ERANGE;
+        }
+
+        tvar->value = strdup(regs);
+        if (tvar->value == NULL)
+                return -ENOMEM;
+
+        if (ref) {
+                tvar->ref = alloc_trace_arg_ref((long)offs);
+                if (tvar->ref == NULL)
+                        return -ENOMEM;
+        }
+        return 0;
+}
+
+static int convert_variable_type(Dwarf_Die *vr_die,
+                                 struct probe_trace_arg *tvar,
+                                 const char *cast)
+{
+        struct probe_trace_arg_ref **ref_ptr = &tvar->ref;
+        Dwarf_Die type;
+        char buf[16];
+        int ret;
+
+        /* TODO: check all types */
+        if (cast && strcmp(cast, "string") != 0) {
+                /* Non string type is OK */
+                tvar->type = strdup(cast);
+                return (tvar->type == NULL) ? -ENOMEM : 0;
+        }
+
+        if (die_get_real_type(vr_die, &type) == NULL) {
+                pr_warning("Failed to get a type information of %s.\n",
+                           dwarf_diename(vr_die));
+                return -ENOENT;
+        }
+
+        pr_debug("%s type is %s.\n",
+                 dwarf_diename(vr_die), dwarf_diename(&type));
+
+        if (cast && strcmp(cast, "string") == 0) {      /* String type */
+                ret = dwarf_tag(&type);
+                if (ret != DW_TAG_pointer_type &&
+                    ret != DW_TAG_array_type) {
+                        pr_warning("Failed to cast into string: "
+                                   "%s(%s) is not a pointer nor array.\n",
+                                   dwarf_diename(vr_die), dwarf_diename(&type));
+                        return -EINVAL;
+                }
+                if (ret == DW_TAG_pointer_type) {
+                        if (die_get_real_type(&type, &type) == NULL) {
+                                pr_warning("Failed to get a type"
+                                           " information.\n");
+                                return -ENOENT;
+                        }
+                        while (*ref_ptr)
+                                ref_ptr = &(*ref_ptr)->next;
+                        /* Add new reference with offset +0 */
+                        *ref_ptr = zalloc(sizeof(struct probe_trace_arg_ref));
+                        if (*ref_ptr == NULL) {
+                                pr_warning("Out of memory error\n");
+                                return -ENOMEM;
+                        }
+                }
+                if (!die_compare_name(&type, "char") &&
+                    !die_compare_name(&type, "unsigned char")) {
+                        pr_warning("Failed to cast into string: "
+                                   "%s is not (unsigned) char *.\n",
+                                   dwarf_diename(vr_die));
+                        return -EINVAL;
+                }
+                tvar->type = strdup(cast);
+                return (tvar->type == NULL) ? -ENOMEM : 0;
+        }
+
+        ret = die_get_byte_size(&type) * 8;
+        if (ret) {
+                /* Check the bitwidth */
+                if (ret > MAX_BASIC_TYPE_BITS) {
+                        pr_info("%s exceeds max-bitwidth."
+                                " Cut down to %d bits.\n",
+                                dwarf_diename(&type), MAX_BASIC_TYPE_BITS);
+                        ret = MAX_BASIC_TYPE_BITS;
+                }
+
+                ret = snprintf(buf, 16, "%c%d",
+                               die_is_signed_type(&type) ? 's' : 'u', ret);
+                if (ret < 0 || ret >= 16) {
+                        if (ret >= 16)
+                                ret = -E2BIG;
+                        pr_warning("Failed to convert variable type: %s\n",
+                                   strerror(-ret));
+                        return ret;
+                }
+                tvar->type = strdup(buf);
+                if (tvar->type == NULL)
+                        return -ENOMEM;
+        }
+        return 0;
+}
+
+static int convert_variable_fields(Dwarf_Die *vr_die, const char *varname,
+                                    struct perf_probe_arg_field *field,
+                                    struct probe_trace_arg_ref **ref_ptr,
+                                    Dwarf_Die *die_mem)
+{
+        struct probe_trace_arg_ref *ref = *ref_ptr;
+        Dwarf_Die type;
+        Dwarf_Word offs;
+        int ret, tag;
+
+        pr_debug("converting %s in %s\n", field->name, varname);
+        if (die_get_real_type(vr_die, &type) == NULL) {
+                pr_warning("Failed to get the type of %s.\n", varname);
+                return -ENOENT;
+        }
+        pr_debug2("Var real type: (%x)\n", (unsigned)dwarf_dieoffset(&type));
+        tag = dwarf_tag(&type);
+
+        if (field->name[0] == '[' &&
+            (tag == DW_TAG_array_type || tag == DW_TAG_pointer_type)) {
+                if (field->next)
+                        /* Save original type for next field */
+                        memcpy(die_mem, &type, sizeof(*die_mem));
+                /* Get the type of this array */
+                if (die_get_real_type(&type, &type) == NULL) {
+                        pr_warning("Failed to get the type of %s.\n", varname);
+                        return -ENOENT;
+                }
+                pr_debug2("Array real type: (%x)\n",
+                         (unsigned)dwarf_dieoffset(&type));
+                if (tag == DW_TAG_pointer_type) {
+                        ref = zalloc(sizeof(struct probe_trace_arg_ref));
+                        if (ref == NULL)
+                                return -ENOMEM;
+                        if (*ref_ptr)
+                                (*ref_ptr)->next = ref;
+                        else
+                                *ref_ptr = ref;
+                }
+                ref->offset += die_get_byte_size(&type) * field->index;
+                if (!field->next)
+                        /* Save vr_die for converting types */
+                        memcpy(die_mem, vr_die, sizeof(*die_mem));
+                goto next;
+        } else if (tag == DW_TAG_pointer_type) {
+                /* Check the pointer and dereference */
+                if (!field->ref) {
+                        pr_err("Semantic error: %s must be referred by '->'\n",
+                               field->name);
+                        return -EINVAL;
+                }
+                /* Get the type pointed by this pointer */
+                if (die_get_real_type(&type, &type) == NULL) {
+                        pr_warning("Failed to get the type of %s.\n", varname);
+                        return -ENOENT;
+                }
+                /* Verify it is a data structure  */
+                if (dwarf_tag(&type) != DW_TAG_structure_type) {
+                        pr_warning("%s is not a data structure.\n", varname);
+                        return -EINVAL;
+                }
+
+                ref = zalloc(sizeof(struct probe_trace_arg_ref));
+                if (ref == NULL)
+                        return -ENOMEM;
+                if (*ref_ptr)
+                        (*ref_ptr)->next = ref;
+                else
+                        *ref_ptr = ref;
+        } else {
+                /* Verify it is a data structure  */
+                if (tag != DW_TAG_structure_type) {
+                        pr_warning("%s is not a data structure.\n", varname);
+                        return -EINVAL;
+                }
+                if (field->name[0] == '[') {
+                        pr_err("Semantic error: %s is not a pointor"
+                               " nor array.\n", varname);
+                        return -EINVAL;
+                }
+                if (field->ref) {
+                        pr_err("Semantic error: %s must be referred by '.'\n",
+                               field->name);
+                        return -EINVAL;
+                }
+                if (!ref) {
+                        pr_warning("Structure on a register is not "
+                                   "supported yet.\n");
+                        return -ENOTSUP;
+                }
+        }
+
+        if (die_find_member(&type, field->name, die_mem) == NULL) {
+                pr_warning("%s(tyep:%s) has no member %s.\n", varname,
+                           dwarf_diename(&type), field->name);
+                return -EINVAL;
+        }
+
+        /* Get the offset of the field */
+        ret = die_get_data_member_location(die_mem, &offs);
+        if (ret < 0) {
+                pr_warning("Failed to get the offset of %s.\n", field->name);
+                return ret;
+        }
+        ref->offset += (long)offs;
+
+next:
+        /* Converting next field */
+        if (field->next)
+                return convert_variable_fields(die_mem, field->name,
+                                        field->next, &ref, die_mem);
+        else
+                return 0;
+}
+
+/* Show a variables in kprobe event format */
+static int convert_variable(Dwarf_Die *vr_die, struct probe_finder *pf)
+{
+        Dwarf_Die die_mem;
+        int ret;
+
+        pr_debug("Converting variable %s into trace event.\n",
+                 dwarf_diename(vr_die));
+
+        ret = convert_variable_location(vr_die, pf->addr, pf->fb_ops,
+                                        pf->tvar);
+        if (ret == -ENOENT)
+                pr_err("Failed to find the location of %s at this address.\n"
+                       " Perhaps, it has been optimized out.\n", pf->pvar->var);
+        else if (ret == -ENOTSUP)
+                pr_err("Sorry, we don't support this variable location yet.\n");
+        else if (pf->pvar->field) {
+                ret = convert_variable_fields(vr_die, pf->pvar->var,
+                                              pf->pvar->field, &pf->tvar->ref,
+                                              &die_mem);
+                vr_die = &die_mem;
+        }
+        if (ret == 0)
+                ret = convert_variable_type(vr_die, pf->tvar, pf->pvar->type);
+        /* *expr will be cached in libdw. Don't free it. */
+        return ret;
+}
+
+/* Find a variable in a subprogram die */
+static int find_variable(Dwarf_Die *sp_die, struct probe_finder *pf)
+{
+        Dwarf_Die vr_die, *scopes;
+        char buf[32], *ptr;
+        int ret, nscopes;
+
+        if (!is_c_varname(pf->pvar->var)) {
+                /* Copy raw parameters */
+                pf->tvar->value = strdup(pf->pvar->var);
+                if (pf->tvar->value == NULL)
+                        return -ENOMEM;
+                if (pf->pvar->type) {
+                        pf->tvar->type = strdup(pf->pvar->type);
+                        if (pf->tvar->type == NULL)
+                                return -ENOMEM;
+                }
+                if (pf->pvar->name) {
+                        pf->tvar->name = strdup(pf->pvar->name);
+                        if (pf->tvar->name == NULL)
+                                return -ENOMEM;
+                } else
+                        pf->tvar->name = NULL;
+                return 0;
+        }
+
+        if (pf->pvar->name)
+                pf->tvar->name = strdup(pf->pvar->name);
+        else {
+                ret = synthesize_perf_probe_arg(pf->pvar, buf, 32);
+                if (ret < 0)
+                        return ret;
+                ptr = strchr(buf, ':'); /* Change type separator to _ */
+                if (ptr)
+                        *ptr = '_';
+                pf->tvar->name = strdup(buf);
+        }
+        if (pf->tvar->name == NULL)
+                return -ENOMEM;
+
+        pr_debug("Searching '%s' variable in context.\n",
+                 pf->pvar->var);
+        /* Search child die for local variables and parameters. */
+        if (die_find_variable_at(sp_die, pf->pvar->var, pf->addr, &vr_die))
+                ret = convert_variable(&vr_die, pf);
+        else {
+                /* Search upper class */
+                nscopes = dwarf_getscopes_die(sp_die, &scopes);
+                while (nscopes-- > 1) {
+                        pr_debug("Searching variables in %s\n",
+                                 dwarf_diename(&scopes[nscopes]));
+                        /* We should check this scope, so give dummy address */
+                        if (die_find_variable_at(&scopes[nscopes],
+                                                 pf->pvar->var, 0,
+                                                 &vr_die)) {
+                                ret = convert_variable(&vr_die, pf);
+                                goto found;
+                        }
+                }
+                if (scopes)
+                        free(scopes);
+                ret = -ENOENT;
+        }
+found:
+        if (ret < 0)
+                pr_warning("Failed to find '%s' in this function.\n",
+                           pf->pvar->var);
+        return ret;
+}
+
+/* Convert subprogram DIE to trace point */
+static int convert_to_trace_point(Dwarf_Die *sp_die, Dwarf_Addr paddr,
+                                  bool retprobe, struct probe_trace_point *tp)
+{
+        Dwarf_Addr eaddr;
+        const char *name;
+
+        /* Copy the name of probe point */
+        name = dwarf_diename(sp_die);
+        if (name) {
+                if (dwarf_entrypc(sp_die, &eaddr) != 0) {
+                        pr_warning("Failed to get entry address of %s\n",
+                                   dwarf_diename(sp_die));
+                        return -ENOENT;
+                }
+                tp->symbol = strdup(name);
+                if (tp->symbol == NULL)
+                        return -ENOMEM;
+                tp->offset = (unsigned long)(paddr - eaddr);
+        } else
+                /* This function has no name. */
+                tp->offset = (unsigned long)paddr;
+
+        /* Return probe must be on the head of a subprogram */
+        if (retprobe) {
+                if (eaddr != paddr) {
+                        pr_warning("Return probe must be on the head of"
+                                   " a real function.\n");
+                        return -EINVAL;
+                }
+                tp->retprobe = true;
+        }
+
+        return 0;
+}
+
+/* Call probe_finder callback with real subprogram DIE */
+static int call_probe_finder(Dwarf_Die *sp_die, struct probe_finder *pf)
+{
+        Dwarf_Die die_mem;
+        Dwarf_Attribute fb_attr;
+        size_t nops;
+        int ret;
+
+        /* If no real subprogram, find a real one */
+        if (!sp_die || dwarf_tag(sp_die) != DW_TAG_subprogram) {
+                sp_die = die_find_real_subprogram(&pf->cu_die,
+                                                  pf->addr, &die_mem);
+                if (!sp_die) {
+                        pr_warning("Failed to find probe point in any "
+                                   "functions.\n");
+                        return -ENOENT;
+                }
+        }
+
+        /* Get the frame base attribute/ops */
+        dwarf_attr(sp_die, DW_AT_frame_base, &fb_attr);
+        ret = dwarf_getlocation_addr(&fb_attr, pf->addr, &pf->fb_ops, &nops, 1);
+        if (ret <= 0 || nops == 0) {
+                pf->fb_ops = NULL;
+#if _ELFUTILS_PREREQ(0, 142)
+        } else if (nops == 1 && pf->fb_ops[0].atom == DW_OP_call_frame_cfa &&
+                   pf->cfi != NULL) {
+                Dwarf_Frame *frame;
+                if (dwarf_cfi_addrframe(pf->cfi, pf->addr, &frame) != 0 ||
+                    dwarf_frame_cfa(frame, &pf->fb_ops, &nops) != 0) {
+                        pr_warning("Failed to get call frame on 0x%jx\n",
+                                   (uintmax_t)pf->addr);
+                        return -ENOENT;
+                }
+#endif
+        }
+
+        /* Call finder's callback handler */
+        ret = pf->callback(sp_die, pf);
+
+        /* *pf->fb_ops will be cached in libdw. Don't free it. */
+        pf->fb_ops = NULL;
+
+        return ret;
+}
+
+/* Find probe point from its line number */
+static int find_probe_point_by_line(struct probe_finder *pf)
+{
+        Dwarf_Lines *lines;
+        Dwarf_Line *line;
+        size_t nlines, i;
+        Dwarf_Addr addr;
+        int lineno;
+        int ret = 0;
+
+        if (dwarf_getsrclines(&pf->cu_die, &lines, &nlines) != 0) {
+                pr_warning("No source lines found.\n");
+                return -ENOENT;
+        }
+
+        for (i = 0; i < nlines && ret == 0; i++) {
+                line = dwarf_onesrcline(lines, i);
+                if (dwarf_lineno(line, &lineno) != 0 ||
+                    lineno != pf->lno)
+                        continue;
+
+                /* TODO: Get fileno from line, but how? */
+                if (strtailcmp(dwarf_linesrc(line, NULL, NULL), pf->fname) != 0)
+                        continue;
+
+                if (dwarf_lineaddr(line, &addr) != 0) {
+                        pr_warning("Failed to get the address of the line.\n");
+                        return -ENOENT;
+                }
+                pr_debug("Probe line found: line[%d]:%d addr:0x%jx\n",
+                         (int)i, lineno, (uintmax_t)addr);
+                pf->addr = addr;
+
+                ret = call_probe_finder(NULL, pf);
+                /* Continuing, because target line might be inlined. */
+        }
+        return ret;
+}
+
+/* Find lines which match lazy pattern */
+static int find_lazy_match_lines(struct list_head *head,
+                                 const char *fname, const char *pat)
+{
+        char *fbuf, *p1, *p2;
+        int fd, line, nlines = -1;
+        struct stat st;
+
+        fd = open(fname, O_RDONLY);
+        if (fd < 0) {
+                pr_warning("Failed to open %s: %s\n", fname, strerror(-fd));
+                return -errno;
+        }
+
+        if (fstat(fd, &st) < 0) {
+                pr_warning("Failed to get the size of %s: %s\n",
+                           fname, strerror(errno));
+                nlines = -errno;
+                goto out_close;
+        }
+
+        nlines = -ENOMEM;
+        fbuf = malloc(st.st_size + 2);
+        if (fbuf == NULL)
+                goto out_close;
+        if (read(fd, fbuf, st.st_size) < 0) {
+                pr_warning("Failed to read %s: %s\n", fname, strerror(errno));
+                nlines = -errno;
+                goto out_free_fbuf;
+        }
+        fbuf[st.st_size] = '\n';        /* Dummy line */
+        fbuf[st.st_size + 1] = '\0';
+        p1 = fbuf;
+        line = 1;
+        nlines = 0;
+        while ((p2 = strchr(p1, '\n')) != NULL) {
+                *p2 = '\0';
+                if (strlazymatch(p1, pat)) {
+                        line_list__add_line(head, line);
+                        nlines++;
+                }
+                line++;
+                p1 = p2 + 1;
+        }
+out_free_fbuf:
+        free(fbuf);
+out_close:
+        close(fd);
+        return nlines;
+}
+
+/* Find probe points from lazy pattern  */
+static int find_probe_point_lazy(Dwarf_Die *sp_die, struct probe_finder *pf)
+{
+        Dwarf_Lines *lines;
+        Dwarf_Line *line;
+        size_t nlines, i;
+        Dwarf_Addr addr;
+        Dwarf_Die die_mem;
+        int lineno;
+        int ret = 0;
+
+        if (list_empty(&pf->lcache)) {
+                /* Matching lazy line pattern */
+                ret = find_lazy_match_lines(&pf->lcache, pf->fname,
+                                            pf->pev->point.lazy_line);
+                if (ret == 0) {
+                        pr_debug("No matched lines found in %s.\n", pf->fname);
+                        return 0;
+                } else if (ret < 0)
+                        return ret;
+        }
+
+        if (dwarf_getsrclines(&pf->cu_die, &lines, &nlines) != 0) {
+                pr_warning("No source lines found.\n");
+                return -ENOENT;
+        }
+
+        for (i = 0; i < nlines && ret >= 0; i++) {
+                line = dwarf_onesrcline(lines, i);
+
+                if (dwarf_lineno(line, &lineno) != 0 ||
+                    !line_list__has_line(&pf->lcache, lineno))
+                        continue;
+
+                /* TODO: Get fileno from line, but how? */
+                if (strtailcmp(dwarf_linesrc(line, NULL, NULL), pf->fname) != 0)
+                        continue;
+
+                if (dwarf_lineaddr(line, &addr) != 0) {
+                        pr_debug("Failed to get the address of line %d.\n",
+                                 lineno);
+                        continue;
+                }
+                if (sp_die) {
+                        /* Address filtering 1: does sp_die include addr? */
+                        if (!dwarf_haspc(sp_die, addr))
+                                continue;
+                        /* Address filtering 2: No child include addr? */
+                        if (die_find_inlinefunc(sp_die, addr, &die_mem))
+                                continue;
+                }
+
+                pr_debug("Probe line found: line[%d]:%d addr:0x%llx\n",
+                         (int)i, lineno, (unsigned long long)addr);
+                pf->addr = addr;
+
+                ret = call_probe_finder(sp_die, pf);
+                /* Continuing, because target line might be inlined. */
+        }
+        /* TODO: deallocate lines, but how? */
+        return ret;
+}
+
+/* Callback parameter with return value */
+struct dwarf_callback_param {
+        void *data;
+        int retval;
+};
+
+static int probe_point_inline_cb(Dwarf_Die *in_die, void *data)
+{
+        struct dwarf_callback_param *param = data;
+        struct probe_finder *pf = param->data;
+        struct perf_probe_point *pp = &pf->pev->point;
+        Dwarf_Addr addr;
+
+        if (pp->lazy_line)
+                param->retval = find_probe_point_lazy(in_die, pf);
+        else {
+                /* Get probe address */
+                if (dwarf_entrypc(in_die, &addr) != 0) {
+                        pr_warning("Failed to get entry address of %s.\n",
+                                   dwarf_diename(in_die));
+                        param->retval = -ENOENT;
+                        return DWARF_CB_ABORT;
+                }
+                pf->addr = addr;
+                pf->addr += pp->offset;
+                pr_debug("found inline addr: 0x%jx\n",
+                         (uintmax_t)pf->addr);
+
+                param->retval = call_probe_finder(in_die, pf);
+                if (param->retval < 0)
+                        return DWARF_CB_ABORT;
+        }
+
+        return DWARF_CB_OK;
+}
+
+/* Search function from function name */
+static int probe_point_search_cb(Dwarf_Die *sp_die, void *data)
+{
+        struct dwarf_callback_param *param = data;
+        struct probe_finder *pf = param->data;
+        struct perf_probe_point *pp = &pf->pev->point;
+
+        /* Check tag and diename */
+        if (dwarf_tag(sp_die) != DW_TAG_subprogram ||
+            !die_compare_name(sp_die, pp->function))
+                return DWARF_CB_OK;
+
+        pf->fname = dwarf_decl_file(sp_die);
+        if (pp->line) { /* Function relative line */
+                dwarf_decl_line(sp_die, &pf->lno);
+                pf->lno += pp->line;
+                param->retval = find_probe_point_by_line(pf);
+        } else if (!dwarf_func_inline(sp_die)) {
+                /* Real function */
+                if (pp->lazy_line)
+                        param->retval = find_probe_point_lazy(sp_die, pf);
+                else {
+                        if (dwarf_entrypc(sp_die, &pf->addr) != 0) {
+                                pr_warning("Failed to get entry address of "
+                                           "%s.\n", dwarf_diename(sp_die));
+                                param->retval = -ENOENT;
+                                return DWARF_CB_ABORT;
+                        }
+                        pf->addr += pp->offset;
+                        /* TODO: Check the address in this function */
+                        param->retval = call_probe_finder(sp_die, pf);
+                }
+        } else {
+                struct dwarf_callback_param _param = {.data = (void *)pf,
+                                                      .retval = 0};
+                /* Inlined function: search instances */
+                dwarf_func_inline_instances(sp_die, probe_point_inline_cb,
+                                            &_param);
+                param->retval = _param.retval;
+        }
+
+        return DWARF_CB_ABORT; /* Exit; no same symbol in this CU. */
+}
+
+static int find_probe_point_by_func(struct probe_finder *pf)
+{
+        struct dwarf_callback_param _param = {.data = (void *)pf,
+                                              .retval = 0};
+        dwarf_getfuncs(&pf->cu_die, probe_point_search_cb, &_param, 0);
+        return _param.retval;
+}
+
+/* Find probe points from debuginfo */
+static int find_probes(int fd, struct probe_finder *pf)
+{
+        struct perf_probe_point *pp = &pf->pev->point;
+        Dwarf_Off off, noff;
+        size_t cuhl;
+        Dwarf_Die *diep;
+        Dwarf *dbg = NULL;
+        Dwfl *dwfl;
+        Dwarf_Addr bias;        /* Currently ignored */
+        int ret = 0;
+
+        dbg = dwfl_init_offline_dwarf(fd, &dwfl, &bias);
+        if (!dbg) {
+                pr_warning("No debug information found in the vmlinux - "
+                        "please rebuild with CONFIG_DEBUG_INFO=y.\n");
+                return -EBADF;
+        }
+
+#if _ELFUTILS_PREREQ(0, 142)
+        /* Get the call frame information from this dwarf */
+        pf->cfi = dwarf_getcfi(dbg);
+#endif
+
+        off = 0;
+        line_list__init(&pf->lcache);
+        /* Loop on CUs (Compilation Unit) */
+        while (!dwarf_nextcu(dbg, off, &noff, &cuhl, NULL, NULL, NULL) &&
+               ret >= 0) {
+                /* Get the DIE(Debugging Information Entry) of this CU */
+                diep = dwarf_offdie(dbg, off + cuhl, &pf->cu_die);
+                if (!diep)
+                        continue;
+
+                /* Check if target file is included. */
+                if (pp->file)
+                        pf->fname = cu_find_realpath(&pf->cu_die, pp->file);
+                else
+                        pf->fname = NULL;
+
+                if (!pp->file || pf->fname) {
+                        if (pp->function)
+                                ret = find_probe_point_by_func(pf);
+                        else if (pp->lazy_line)
+                                ret = find_probe_point_lazy(NULL, pf);
+                        else {
+                                pf->lno = pp->line;
+                                ret = find_probe_point_by_line(pf);
+                        }
+                }
+                off = noff;
+        }
+        line_list__free(&pf->lcache);
+        if (dwfl)
+                dwfl_end(dwfl);
+
+        return ret;
+}
+
+/* Add a found probe point into trace event list */
+static int add_probe_trace_event(Dwarf_Die *sp_die, struct probe_finder *pf)
+{
+        struct trace_event_finder *tf =
+                        container_of(pf, struct trace_event_finder, pf);
+        struct probe_trace_event *tev;
+        int ret, i;
+
+        /* Check number of tevs */
+        if (tf->ntevs == tf->max_tevs) {
+                pr_warning("Too many( > %d) probe point found.\n",
+                           tf->max_tevs);
+                return -ERANGE;
+        }
+        tev = &tf->tevs[tf->ntevs++];
+
+        ret = convert_to_trace_point(sp_die, pf->addr, pf->pev->point.retprobe,
+                                     &tev->point);
+        if (ret < 0)
+                return ret;
+
+        pr_debug("Probe point found: %s+%lu\n", tev->point.symbol,
+                 tev->point.offset);
+
+        /* Find each argument */
+        tev->nargs = pf->pev->nargs;
+        tev->args = zalloc(sizeof(struct probe_trace_arg) * tev->nargs);
+        if (tev->args == NULL)
+                return -ENOMEM;
+        for (i = 0; i < pf->pev->nargs; i++) {
+                pf->pvar = &pf->pev->args[i];
+                pf->tvar = &tev->args[i];
+                ret = find_variable(sp_die, pf);
+                if (ret != 0)
+                        return ret;
+        }
+
+        return 0;
+}
+
+/* Find probe_trace_events specified by perf_probe_event from debuginfo */
+int find_probe_trace_events(int fd, struct perf_probe_event *pev,
+                            struct probe_trace_event **tevs, int max_tevs)
+{
+        struct trace_event_finder tf = {
+                        .pf = {.pev = pev, .callback = add_probe_trace_event},
+                        .max_tevs = max_tevs};
+        int ret;
+
+        /* Allocate result tevs array */
+        *tevs = zalloc(sizeof(struct probe_trace_event) * max_tevs);
+        if (*tevs == NULL)
+                return -ENOMEM;
+
+        tf.tevs = *tevs;
+        tf.ntevs = 0;
+
+        ret = find_probes(fd, &tf.pf);
+        if (ret < 0) {
+                free(*tevs);
+                *tevs = NULL;
+                return ret;
+        }
+
+        return (ret < 0) ? ret : tf.ntevs;
+}
+
+#define MAX_VAR_LEN 64
+
+/* Collect available variables in this scope */
+static int collect_variables_cb(Dwarf_Die *die_mem, void *data)
+{
+        struct available_var_finder *af = data;
+        struct variable_list *vl;
+        char buf[MAX_VAR_LEN];
+        int tag, ret;
+
+        vl = &af->vls[af->nvls - 1];
+
+        tag = dwarf_tag(die_mem);
+        if (tag == DW_TAG_formal_parameter ||
+            tag == DW_TAG_variable) {
+                ret = convert_variable_location(die_mem, af->pf.addr,
+                                                af->pf.fb_ops, NULL);
+                if (ret == 0) {
+                        ret = die_get_varname(die_mem, buf, MAX_VAR_LEN);
+                        pr_debug2("Add new var: %s\n", buf);
+                        if (ret > 0)
+                                strlist__add(vl->vars, buf);
+                }
+        }
+
+        if (af->child && dwarf_haspc(die_mem, af->pf.addr))
+                return DIE_FIND_CB_CONTINUE;
+        else
+                return DIE_FIND_CB_SIBLING;
+}
+
+/* Add a found vars into available variables list */
+static int add_available_vars(Dwarf_Die *sp_die, struct probe_finder *pf)
+{
+        struct available_var_finder *af =
+                        container_of(pf, struct available_var_finder, pf);
+        struct variable_list *vl;
+        Dwarf_Die die_mem, *scopes = NULL;
+        int ret, nscopes;
+
+        /* Check number of tevs */
+        if (af->nvls == af->max_vls) {
+                pr_warning("Too many( > %d) probe point found.\n", af->max_vls);
+                return -ERANGE;
+        }
+        vl = &af->vls[af->nvls++];
+
+        ret = convert_to_trace_point(sp_die, pf->addr, pf->pev->point.retprobe,
+                                     &vl->point);
+        if (ret < 0)
+                return ret;
+
+        pr_debug("Probe point found: %s+%lu\n", vl->point.symbol,
+                 vl->point.offset);
+
+        /* Find local variables */
+        vl->vars = strlist__new(true, NULL);
+        if (vl->vars == NULL)
+                return -ENOMEM;
+        af->child = true;
+        die_find_child(sp_die, collect_variables_cb, (void *)af, &die_mem);
+
+        /* Find external variables */
+        if (!af->externs)
+                goto out;
+        /* Don't need to search child DIE for externs. */
+        af->child = false;
+        nscopes = dwarf_getscopes_die(sp_die, &scopes);
+        while (nscopes-- > 1)
+                die_find_child(&scopes[nscopes], collect_variables_cb,
+                               (void *)af, &die_mem);
+        if (scopes)
+                free(scopes);
+
+out:
+        if (strlist__empty(vl->vars)) {
+                strlist__delete(vl->vars);
+                vl->vars = NULL;
+        }
+
+        return ret;
+}
+
+/* Find available variables at given probe point */
+int find_available_vars_at(int fd, struct perf_probe_event *pev,
+                           struct variable_list **vls, int max_vls,
+                           bool externs)
+{
+        struct available_var_finder af = {
+                        .pf = {.pev = pev, .callback = add_available_vars},
+                        .max_vls = max_vls, .externs = externs};
+        int ret;
+
+        /* Allocate result vls array */
+        *vls = zalloc(sizeof(struct variable_list) * max_vls);
+        if (*vls == NULL)
+                return -ENOMEM;
+
+        af.vls = *vls;
+        af.nvls = 0;
+
+        ret = find_probes(fd, &af.pf);
+        if (ret < 0) {
+                /* Free vlist for error */
+                while (af.nvls--) {
+                        if (af.vls[af.nvls].point.symbol)
+                                free(af.vls[af.nvls].point.symbol);
+                        if (af.vls[af.nvls].vars)
+                                strlist__delete(af.vls[af.nvls].vars);
+                }
+                free(af.vls);
+                *vls = NULL;
+                return ret;
+        }
+
+        return (ret < 0) ? ret : af.nvls;
+}
+
+/* Reverse search */
+int find_perf_probe_point(unsigned long addr, struct perf_probe_point *ppt)
+{
+        Dwarf_Die cudie, spdie, indie;
+        Dwarf *dbg = NULL;
+        Dwfl *dwfl = NULL;
+        Dwarf_Line *line;
+        Dwarf_Addr laddr, eaddr, bias = 0;
+        const char *tmp;
+        int lineno, ret = 0;
+        bool found = false;
+
+        /* Open the live linux kernel */
+        dbg = dwfl_init_live_kernel_dwarf(addr, &dwfl, &bias);
+        if (!dbg) {
+                pr_warning("No debug information found in the vmlinux - "
+                        "please rebuild with CONFIG_DEBUG_INFO=y.\n");
+                ret = -EINVAL;
+                goto end;
+        }
+
+        /* Adjust address with bias */
+        addr += bias;
+        /* Find cu die */
+        if (!dwarf_addrdie(dbg, (Dwarf_Addr)addr - bias, &cudie)) {
+                pr_warning("Failed to find debug information for address %lx\n",
+                           addr);
+                ret = -EINVAL;
+                goto end;
+        }
+
+        /* Find a corresponding line */
+        line = dwarf_getsrc_die(&cudie, (Dwarf_Addr)addr);
+        if (line) {
+                if (dwarf_lineaddr(line, &laddr) == 0 &&
+                    (Dwarf_Addr)addr == laddr &&
+                    dwarf_lineno(line, &lineno) == 0) {
+                        tmp = dwarf_linesrc(line, NULL, NULL);
+                        if (tmp) {
+                                ppt->line = lineno;
+                                ppt->file = strdup(tmp);
+                                if (ppt->file == NULL) {
+                                        ret = -ENOMEM;
+                                        goto end;
+                                }
+                                found = true;
+                        }
+                }
+        }
+
+        /* Find a corresponding function */
+        if (die_find_real_subprogram(&cudie, (Dwarf_Addr)addr, &spdie)) {
+                tmp = dwarf_diename(&spdie);
+                if (!tmp || dwarf_entrypc(&spdie, &eaddr) != 0)
+                        goto end;
+
+                if (ppt->line) {
+                        if (die_find_inlinefunc(&spdie, (Dwarf_Addr)addr,
+                                                &indie)) {
+                                /* addr in an inline function */
+                                tmp = dwarf_diename(&indie);
+                                if (!tmp)
+                                        goto end;
+                                ret = dwarf_decl_line(&indie, &lineno);
+                        } else {
+                                if (eaddr == addr) {    /* Function entry */
+                                        lineno = ppt->line;
+                                        ret = 0;
+                                } else
+                                        ret = dwarf_decl_line(&spdie, &lineno);
+                        }
+                        if (ret == 0) {
+                                /* Make a relative line number */
+                                ppt->line -= lineno;
+                                goto found;
+                        }
+                }
+                /* We don't have a line number, let's use offset */
+                ppt->offset = addr - (unsigned long)eaddr;
+found:
+                ppt->function = strdup(tmp);
+                if (ppt->function == NULL) {
+                        ret = -ENOMEM;
+                        goto end;
+                }
+                found = true;
+        }
+
+end:
+        if (dwfl)
+                dwfl_end(dwfl);
+        if (ret >= 0)
+                ret = found ? 1 : 0;
+        return ret;
+}
+
+/* Add a line and store the src path */
+static int line_range_add_line(const char *src, unsigned int lineno,
+                               struct line_range *lr)
+{
+        /* Copy source path */
+        if (!lr->path) {
+                lr->path = strdup(src);
+                if (lr->path == NULL)
+                        return -ENOMEM;
+        }
+        return line_list__add_line(&lr->line_list, lineno);
+}
+
+/* Search function declaration lines */
+static int line_range_funcdecl_cb(Dwarf_Die *sp_die, void *data)
+{
+        struct dwarf_callback_param *param = data;
+        struct line_finder *lf = param->data;
+        const char *src;
+        int lineno;
+
+        src = dwarf_decl_file(sp_die);
+        if (src && strtailcmp(src, lf->fname) != 0)
+                return DWARF_CB_OK;
+
+        if (dwarf_decl_line(sp_die, &lineno) != 0 ||
+            (lf->lno_s > lineno || lf->lno_e < lineno))
+                return DWARF_CB_OK;
+
+        param->retval = line_range_add_line(src, lineno, lf->lr);
+        if (param->retval < 0)
+                return DWARF_CB_ABORT;
+        return DWARF_CB_OK;
+}
+
+static int find_line_range_func_decl_lines(struct line_finder *lf)
+{
+        struct dwarf_callback_param param = {.data = (void *)lf, .retval = 0};
+        dwarf_getfuncs(&lf->cu_die, line_range_funcdecl_cb, &param, 0);
+        return param.retval;
+}
+
+/* Find line range from its line number */
+static int find_line_range_by_line(Dwarf_Die *sp_die, struct line_finder *lf)
+{
+        Dwarf_Lines *lines;
+        Dwarf_Line *line;
+        size_t nlines, i;
+        Dwarf_Addr addr;
+        int lineno, ret = 0;
+        const char *src;
+        Dwarf_Die die_mem;
+
+        line_list__init(&lf->lr->line_list);
+        if (dwarf_getsrclines(&lf->cu_die, &lines, &nlines) != 0) {
+                pr_warning("No source lines found.\n");
+                return -ENOENT;
+        }
+
+        /* Search probable lines on lines list */
+        for (i = 0; i < nlines; i++) {
+                line = dwarf_onesrcline(lines, i);
+                if (dwarf_lineno(line, &lineno) != 0 ||
+                    (lf->lno_s > lineno || lf->lno_e < lineno))
+                        continue;
+
+                if (sp_die) {
+                        /* Address filtering 1: does sp_die include addr? */
+                        if (dwarf_lineaddr(line, &addr) != 0 ||
+                            !dwarf_haspc(sp_die, addr))
+                                continue;
+
+                        /* Address filtering 2: No child include addr? */
+                        if (die_find_inlinefunc(sp_die, addr, &die_mem))
+                                continue;
+                }
+
+                /* TODO: Get fileno from line, but how? */
+                src = dwarf_linesrc(line, NULL, NULL);
+                if (strtailcmp(src, lf->fname) != 0)
+                        continue;
+
+                ret = line_range_add_line(src, lineno, lf->lr);
+                if (ret < 0)
+                        return ret;
+        }
+
+        /*
+         * Dwarf lines doesn't include function declarations. We have to
+         * check functions list or given function.
+         */
+        if (sp_die) {
+                src = dwarf_decl_file(sp_die);
+                if (src && dwarf_decl_line(sp_die, &lineno) == 0 &&
+                    (lf->lno_s <= lineno && lf->lno_e >= lineno))
+                        ret = line_range_add_line(src, lineno, lf->lr);
+        } else
+                ret = find_line_range_func_decl_lines(lf);
+
+        /* Update status */
+        if (ret >= 0)
+                if (!list_empty(&lf->lr->line_list))
+                        ret = lf->found = 1;
+                else
+                        ret = 0;        /* Lines are not found */
+        else {
+                free(lf->lr->path);
+                lf->lr->path = NULL;
+        }
+        return ret;
+}
+
+static int line_range_inline_cb(Dwarf_Die *in_die, void *data)
+{
+        struct dwarf_callback_param *param = data;
+
+        param->retval = find_line_range_by_line(in_die, param->data);
+        return DWARF_CB_ABORT;  /* No need to find other instances */
+}
+
+/* Search function from function name */
+static int line_range_search_cb(Dwarf_Die *sp_die, void *data)
+{
+        struct dwarf_callback_param *param = data;
+        struct line_finder *lf = param->data;
+        struct line_range *lr = lf->lr;
+
+        pr_debug("find (%llx) %s\n",
+                 (unsigned long long)dwarf_dieoffset(sp_die),
+                 dwarf_diename(sp_die));
+        if (dwarf_tag(sp_die) == DW_TAG_subprogram &&
+            die_compare_name(sp_die, lr->function)) {
+                lf->fname = dwarf_decl_file(sp_die);
+                dwarf_decl_line(sp_die, &lr->offset);
+                pr_debug("fname: %s, lineno:%d\n", lf->fname, lr->offset);
+                lf->lno_s = lr->offset + lr->start;
+                if (lf->lno_s < 0)      /* Overflow */
+                        lf->lno_s = INT_MAX;
+                lf->lno_e = lr->offset + lr->end;
+                if (lf->lno_e < 0)      /* Overflow */
+                        lf->lno_e = INT_MAX;
+                pr_debug("New line range: %d to %d\n", lf->lno_s, lf->lno_e);
+                lr->start = lf->lno_s;
+                lr->end = lf->lno_e;
+                if (dwarf_func_inline(sp_die)) {
+                        struct dwarf_callback_param _param;
+                        _param.data = (void *)lf;
+                        _param.retval = 0;
+                        dwarf_func_inline_instances(sp_die,
+                                                    line_range_inline_cb,
+                                                    &_param);
+                        param->retval = _param.retval;
+                } else
+                        param->retval = find_line_range_by_line(sp_die, lf);
+                return DWARF_CB_ABORT;
+        }
+        return DWARF_CB_OK;
+}
+
+static int find_line_range_by_func(struct line_finder *lf)
+{
+        struct dwarf_callback_param param = {.data = (void *)lf, .retval = 0};
+        dwarf_getfuncs(&lf->cu_die, line_range_search_cb, &param, 0);
+        return param.retval;
+}
+
+int find_line_range(int fd, struct line_range *lr)
+{
+        struct line_finder lf = {.lr = lr, .found = 0};
+        int ret = 0;
+        Dwarf_Off off = 0, noff;
+        size_t cuhl;
+        Dwarf_Die *diep;
+        Dwarf *dbg = NULL;
+        Dwfl *dwfl;
+        Dwarf_Addr bias;        /* Currently ignored */
+        const char *comp_dir;
+
+        dbg = dwfl_init_offline_dwarf(fd, &dwfl, &bias);
+        if (!dbg) {
+                pr_warning("No debug information found in the vmlinux - "
+                        "please rebuild with CONFIG_DEBUG_INFO=y.\n");
+                return -EBADF;
+        }
+
+        /* Loop on CUs (Compilation Unit) */
+        while (!lf.found && ret >= 0) {
+                if (dwarf_nextcu(dbg, off, &noff, &cuhl, NULL, NULL, NULL) != 0)
+                        break;
+
+                /* Get the DIE(Debugging Information Entry) of this CU */
+                diep = dwarf_offdie(dbg, off + cuhl, &lf.cu_die);
+                if (!diep)
+                        continue;
+
+                /* Check if target file is included. */
+                if (lr->file)
+                        lf.fname = cu_find_realpath(&lf.cu_die, lr->file);
+                else
+                        lf.fname = 0;
+
+                if (!lr->file || lf.fname) {
+                        if (lr->function)
+                                ret = find_line_range_by_func(&lf);
+                        else {
+                                lf.lno_s = lr->start;
+                                lf.lno_e = lr->end;
+                                ret = find_line_range_by_line(NULL, &lf);
+                        }
+                }
+                off = noff;
+        }
+
+        /* Store comp_dir */
+        if (lf.found) {
+                comp_dir = cu_get_comp_dir(&lf.cu_die);
+                if (comp_dir) {
+                        lr->comp_dir = strdup(comp_dir);
+                        if (!lr->comp_dir)
+                                ret = -ENOMEM;
+                }
+        }
+
+        pr_debug("path: %s\n", lr->path);
+        dwfl_end(dwfl);
+        return (ret < 0) ? ret : lf.found;
+}
+
diff --git a/tools/perf/util/probe-finder.h b/tools/perf/util/probe-finder.h
new file mode 100644
index 00000000000..beaefc3c122
--- /dev/null
+++ b/tools/perf/util/probe-finder.h
@@ -0,0 +1,91 @@
+#ifndef _PROBE_FINDER_H
+#define _PROBE_FINDER_H
+
+#include <stdbool.h>
+#include "util.h"
+#include "probe-event.h"
+
+#define MAX_PATH_LEN             256
+#define MAX_PROBE_BUFFER        1024
+#define MAX_PROBES               128
+
+static inline int is_c_varname(const char *name)
+{
+        /* TODO */
+        return isalpha(name[0]) || name[0] == '_';
+}
+
+#ifdef DWARF_SUPPORT
+/* Find probe_trace_events specified by perf_probe_event from debuginfo */
+extern int find_probe_trace_events(int fd, struct perf_probe_event *pev,
+                                    struct probe_trace_event **tevs,
+                                    int max_tevs);
+
+/* Find a perf_probe_point from debuginfo */
+extern int find_perf_probe_point(unsigned long addr,
+                                 struct perf_probe_point *ppt);
+
+/* Find a line range */
+extern int find_line_range(int fd, struct line_range *lr);
+
+/* Find available variables */
+extern int find_available_vars_at(int fd, struct perf_probe_event *pev,
+                                  struct variable_list **vls, int max_points,
+                                  bool externs);
+
+#include <dwarf.h>
+#include <elfutils/libdw.h>
+#include <elfutils/libdwfl.h>
+#include <elfutils/version.h>
+
+struct probe_finder {
+        struct perf_probe_event *pev;           /* Target probe event */
+
+        /* Callback when a probe point is found */
+        int (*callback)(Dwarf_Die *sp_die, struct probe_finder *pf);
+
+        /* For function searching */
+        int                     lno;            /* Line number */
+        Dwarf_Addr              addr;           /* Address */
+        const char              *fname;         /* Real file name */
+        Dwarf_Die               cu_die;         /* Current CU */
+        struct list_head        lcache;         /* Line cache for lazy match */
+
+        /* For variable searching */
+#if _ELFUTILS_PREREQ(0, 142)
+        Dwarf_CFI               *cfi;           /* Call Frame Information */
+#endif
+        Dwarf_Op                *fb_ops;        /* Frame base attribute */
+        struct perf_probe_arg   *pvar;          /* Current target variable */
+        struct probe_trace_arg  *tvar;          /* Current result variable */
+};
+
+struct trace_event_finder {
+        struct probe_finder     pf;
+        struct probe_trace_event *tevs;         /* Found trace events */
+        int                     ntevs;          /* Number of trace events */
+        int                     max_tevs;       /* Max number of trace events */
+};
+
+struct available_var_finder {
+        struct probe_finder     pf;
+        struct variable_list    *vls;           /* Found variable lists */
+        int                     nvls;           /* Number of variable lists */
+        int                     max_vls;        /* Max no. of variable lists */
+        bool                    externs;        /* Find external vars too */
+        bool                    child;          /* Search child scopes */
+};
+
+struct line_finder {
+        struct line_range       *lr;            /* Target line range */
+
+        const char              *fname;         /* File name */
+        int                     lno_s;          /* Start line number */
+        int                     lno_e;          /* End line number */
+        Dwarf_Die               cu_die;         /* Current CU */
+        int                     found;
+};
+
+#endif /* DWARF_SUPPORT */
+
+#endif /*_PROBE_FINDER_H */
diff --git a/tools/perf/util/pstack.c b/tools/perf/util/pstack.c
new file mode 100644
index 00000000000..13d36faf64e
--- /dev/null
+++ b/tools/perf/util/pstack.c
@@ -0,0 +1,75 @@
+/*
+ * Simple pointer stack
+ *
+ * (c) 2010 Arnaldo Carvalho de Melo <acme@redhat.com>
+ */
+
+#include "util.h"
+#include "pstack.h"
+#include <linux/kernel.h>
+#include <stdlib.h>
+
+struct pstack {
+        unsigned short  top;
+        unsigned short  max_nr_entries;
+        void            *entries[0];
+};
+
+struct pstack *pstack__new(unsigned short max_nr_entries)
+{
+        struct pstack *self = zalloc((sizeof(*self) +
+                                     max_nr_entries * sizeof(void *)));
+        if (self != NULL)
+                self->max_nr_entries = max_nr_entries;
+        return self;
+}
+
+void pstack__delete(struct pstack *self)
+{
+        free(self);
+}
+
+bool pstack__empty(const struct pstack *self)
+{
+        return self->top == 0;
+}
+
+void pstack__remove(struct pstack *self, void *key)
+{
+        unsigned short i = self->top, last_index = self->top - 1;
+
+        while (i-- != 0) {
+                if (self->entries[i] == key) {
+                        if (i < last_index)
+                                memmove(self->entries + i,
+                                        self->entries + i + 1,
+                                        (last_index - i) * sizeof(void *));
+                        --self->top;
+                        return;
+                }
+        }
+        pr_err("%s: %p not on the pstack!\n", __func__, key);
+}
+
+void pstack__push(struct pstack *self, void *key)
+{
+        if (self->top == self->max_nr_entries) {
+                pr_err("%s: top=%d, overflow!\n", __func__, self->top);
+                return;
+        }
+        self->entries[self->top++] = key;
+}
+
+void *pstack__pop(struct pstack *self)
+{
+        void *ret;
+
+        if (self->top == 0) {
+                pr_err("%s: underflow!\n", __func__);
+                return NULL;
+        }
+
+        ret = self->entries[--self->top];
+        self->entries[self->top] = NULL;
+        return ret;
+}
diff --git a/tools/perf/util/pstack.h b/tools/perf/util/pstack.h
new file mode 100644
index 00000000000..4cedea59f51
--- /dev/null
+++ b/tools/perf/util/pstack.h
@@ -0,0 +1,14 @@
+#ifndef _PERF_PSTACK_
+#define _PERF_PSTACK_
+
+#include <stdbool.h>
+
+struct pstack;
+struct pstack *pstack__new(unsigned short max_nr_entries);
+void pstack__delete(struct pstack *self);
+bool pstack__empty(const struct pstack *self);
+void pstack__remove(struct pstack *self, void *key);
+void pstack__push(struct pstack *self, void *key);
+void *pstack__pop(struct pstack *self);
+
+#endif /* _PERF_PSTACK_ */
diff --git a/tools/perf/util/quote.c b/tools/perf/util/quote.c
index f18c5212bc9..01f03242b86 100644
--- a/tools/perf/util/quote.c
+++ b/tools/perf/util/quote.c
@@ -1,8 +1,6 @@
 #include "cache.h"
 #include "quote.h"
 
-int quote_path_fully = 1;
-
 /* Help to copy the thing properly quoted for the shell safety.
  * any single quote is replaced with '\'', any exclamation point
  * is replaced with '\!', and the whole thing is enclosed in a
@@ -19,7 +17,7 @@ static inline int need_bs_quote(char c)
         return (c == '\'' || c == '!');
 }
 
-void sq_quote_buf(struct strbuf *dst, const char *src)
+static void sq_quote_buf(struct strbuf *dst, const char *src)
 {
         char *to_free = NULL;
 
@@ -41,23 +39,6 @@ void sq_quote_buf(struct strbuf *dst, const char *src)
         free(to_free);
 }
 
-void sq_quote_print(FILE *stream, const char *src)
-{
-        char c;
-
-        fputc('\'', stream);
-        while ((c = *src++)) {
-                if (need_bs_quote(c)) {
-                        fputs("'\\", stream);
-                        fputc(c, stream);
-                        fputc('\'', stream);
-                } else {
-                        fputc(c, stream);
-                }
-        }
-        fputc('\'', stream);
-}
-
 void sq_quote_argv(struct strbuf *dst, const char** argv, size_t maxlen)
 {
         int i;
@@ -71,411 +52,3 @@ void sq_quote_argv(struct strbuf *dst, const char** argv, size_t maxlen)
                         die("Too many or long arguments");
         }
 }
-
-char *sq_dequote_step(char *arg, char **next)
-{
-        char *dst = arg;
-        char *src = arg;
-        char c;
-
-        if (*src != '\'')
-                return NULL;
-        for (;;) {
-                c = *++src;
-                if (!c)
-                        return NULL;
-                if (c != '\'') {
-                        *dst++ = c;
-                        continue;
-                }
-                /* We stepped out of sq */
-                switch (*++src) {
-                case '\0':
-                        *dst = 0;
-                        if (next)
-                                *next = NULL;
-                        return arg;
-                case '\\':
-                        c = *++src;
-                        if (need_bs_quote(c) && *++src == '\'') {
-                                *dst++ = c;
-                                continue;
-                        }
-                /* Fallthrough */
-                default:
-                        if (!next || !isspace(*src))
-                                return NULL;
-                        do {
-                                c = *++src;
-                        } while (isspace(c));
-                        *dst = 0;
-                        *next = src;
-                        return arg;
-                }
-        }
-}
-
-char *sq_dequote(char *arg)
-{
-        return sq_dequote_step(arg, NULL);
-}
-
-int sq_dequote_to_argv(char *arg, const char ***argv, int *nr, int *alloc)
-{
-        char *next = arg;
-
-        if (!*arg)
-                return 0;
-        do {
-                char *dequoted = sq_dequote_step(next, &next);
-                if (!dequoted)
-                        return -1;
-                ALLOC_GROW(*argv, *nr + 1, *alloc);
-                (*argv)[(*nr)++] = dequoted;
-        } while (next);
-
-        return 0;
-}
-
-/* 1 means: quote as octal
- * 0 means: quote as octal if (quote_path_fully)
- * -1 means: never quote
- * c: quote as "\\c"
- */
-#define X8(x)   x, x, x, x, x, x, x, x
-#define X16(x)  X8(x), X8(x)
-static signed char const sq_lookup[256] = {
-        /*           0    1    2    3    4    5    6    7 */
-        /* 0x00 */   1,   1,   1,   1,   1,   1,   1, 'a',
-        /* 0x08 */ 'b', 't', 'n', 'v', 'f', 'r',   1,   1,
-        /* 0x10 */ X16(1),
-        /* 0x20 */  -1,  -1, '"',  -1,  -1,  -1,  -1,  -1,
-        /* 0x28 */ X16(-1), X16(-1), X16(-1),
-        /* 0x58 */  -1,  -1,  -1,  -1,'\\',  -1,  -1,  -1,
-        /* 0x60 */ X16(-1), X8(-1),
-        /* 0x78 */  -1,  -1,  -1,  -1,  -1,  -1,  -1,   1,
-        /* 0x80 */ /* set to 0 */
-};
-
-static inline int sq_must_quote(char c)
-{
-        return sq_lookup[(unsigned char)c] + quote_path_fully > 0;
-}
-
-/* returns the longest prefix not needing a quote up to maxlen if positive.
-   This stops at the first \0 because it's marked as a character needing an
-   escape */
-static size_t next_quote_pos(const char *s, ssize_t maxlen)
-{
-        size_t len;
-        if (maxlen < 0) {
-                for (len = 0; !sq_must_quote(s[len]); len++);
-        } else {
-                for (len = 0; len < maxlen && !sq_must_quote(s[len]); len++);
-        }
-        return len;
-}
-
-/*
- * C-style name quoting.
- *
- * (1) if sb and fp are both NULL, inspect the input name and counts the
- *     number of bytes that are needed to hold c_style quoted version of name,
- *     counting the double quotes around it but not terminating NUL, and
- *     returns it.
- *     However, if name does not need c_style quoting, it returns 0.
- *
- * (2) if sb or fp are not NULL, it emits the c_style quoted version
- *     of name, enclosed with double quotes if asked and needed only.
- *     Return value is the same as in (1).
- */
-static size_t quote_c_style_counted(const char *name, ssize_t maxlen,
-                                    struct strbuf *sb, FILE *fp, int no_dq)
-{
-#undef EMIT
-#define EMIT(c)                                 \
-        do {                                        \
-                if (sb) strbuf_addch(sb, (c));          \
-                if (fp) fputc((c), fp);                 \
-                count++;                                \
-        } while (0)
-#define EMITBUF(s, l)                           \
-        do {                                        \
-                int __ret;                              \
-                if (sb) strbuf_add(sb, (s), (l));       \
-                if (fp) __ret = fwrite((s), (l), 1, fp);        \
-                count += (l);                           \
-        } while (0)
-
-        size_t len, count = 0;
-        const char *p = name;
-
-        for (;;) {
-                int ch;
-
-                len = next_quote_pos(p, maxlen);
-                if (len == maxlen || !p[len])
-                        break;
-
-                if (!no_dq && p == name)
-                        EMIT('"');
-
-                EMITBUF(p, len);
-                EMIT('\\');
-                p += len;
-                ch = (unsigned char)*p++;
-                if (sq_lookup[ch] >= ' ') {
-                        EMIT(sq_lookup[ch]);
-                } else {
-                        EMIT(((ch >> 6) & 03) + '0');
-                        EMIT(((ch >> 3) & 07) + '0');
-                        EMIT(((ch >> 0) & 07) + '0');
-                }
-        }
-
-        EMITBUF(p, len);
-        if (p == name)   /* no ending quote needed */
-                return 0;
-
-        if (!no_dq)
-                EMIT('"');
-        return count;
-}
-
-size_t quote_c_style(const char *name, struct strbuf *sb, FILE *fp, int nodq)
-{
-        return quote_c_style_counted(name, -1, sb, fp, nodq);
-}
-
-void quote_two_c_style(struct strbuf *sb, const char *prefix, const char *path, int nodq)
-{
-        if (quote_c_style(prefix, NULL, NULL, 0) ||
-            quote_c_style(path, NULL, NULL, 0)) {
-                if (!nodq)
-                        strbuf_addch(sb, '"');
-                quote_c_style(prefix, sb, NULL, 1);
-                quote_c_style(path, sb, NULL, 1);
-                if (!nodq)
-                        strbuf_addch(sb, '"');
-        } else {
-                strbuf_addstr(sb, prefix);
-                strbuf_addstr(sb, path);
-        }
-}
-
-void write_name_quoted(const char *name, FILE *fp, int terminator)
-{
-        if (terminator) {
-                quote_c_style(name, NULL, fp, 0);
-        } else {
-                fputs(name, fp);
-        }
-        fputc(terminator, fp);
-}
-
-extern void write_name_quotedpfx(const char *pfx, size_t pfxlen,
-                                 const char *name, FILE *fp, int terminator)
-{
-        int needquote = 0;
-
-        if (terminator) {
-                needquote = next_quote_pos(pfx, pfxlen) < pfxlen
-                        || name[next_quote_pos(name, -1)];
-        }
-        if (needquote) {
-                fputc('"', fp);
-                quote_c_style_counted(pfx, pfxlen, NULL, fp, 1);
-                quote_c_style(name, NULL, fp, 1);
-                fputc('"', fp);
-        } else {
-                int ret;
-
-                ret = fwrite(pfx, pfxlen, 1, fp);
-                fputs(name, fp);
-        }
-        fputc(terminator, fp);
-}
-
-/* quote path as relative to the given prefix */
-char *quote_path_relative(const char *in, int len,
-                          struct strbuf *out, const char *prefix)
-{
-        int needquote;
-
-        if (len < 0)
-                len = strlen(in);
-
-        /* "../" prefix itself does not need quoting, but "in" might. */
-        needquote = next_quote_pos(in, len) < len;
-        strbuf_setlen(out, 0);
-        strbuf_grow(out, len);
-
-        if (needquote)
-                strbuf_addch(out, '"');
-        if (prefix) {
-                int off = 0;
-                while (prefix[off] && off < len && prefix[off] == in[off])
-                        if (prefix[off] == '/') {
-                                prefix += off + 1;
-                                in += off + 1;
-                                len -= off + 1;
-                                off = 0;
-                        } else
-                                off++;
-
-                for (; *prefix; prefix++)
-                        if (*prefix == '/')
-                                strbuf_addstr(out, "../");
-        }
-
-        quote_c_style_counted (in, len, out, NULL, 1);
-
-        if (needquote)
-                strbuf_addch(out, '"');
-        if (!out->len)
-                strbuf_addstr(out, "./");
-
-        return out->buf;
-}
-
-/*
- * C-style name unquoting.
- *
- * Quoted should point at the opening double quote.
- * + Returns 0 if it was able to unquote the string properly, and appends the
- *   result in the strbuf `sb'.
- * + Returns -1 in case of error, and doesn't touch the strbuf. Though note
- *   that this function will allocate memory in the strbuf, so calling
- *   strbuf_release is mandatory whichever result unquote_c_style returns.
- *
- * Updates endp pointer to point at one past the ending double quote if given.
- */
-int unquote_c_style(struct strbuf *sb, const char *quoted, const char **endp)
-{
-        size_t oldlen = sb->len, len;
-        int ch, ac;
-
-        if (*quoted++ != '"')
-                return -1;
-
-        for (;;) {
-                len = strcspn(quoted, "\"\\");
-                strbuf_add(sb, quoted, len);
-                quoted += len;
-
-                switch (*quoted++) {
-                  case '"':
-                        if (endp)
-                                *endp = quoted;
-                        return 0;
-                  case '\\':
-                        break;
-                  default:
-                        goto error;
-                }
-
-                switch ((ch = *quoted++)) {
-                case 'a': ch = '\a'; break;
-                case 'b': ch = '\b'; break;
-                case 'f': ch = '\f'; break;
-                case 'n': ch = '\n'; break;
-                case 'r': ch = '\r'; break;
-                case 't': ch = '\t'; break;
-                case 'v': ch = '\v'; break;
-
-                case '\\': case '"':
-                        break; /* verbatim */
-
-                /* octal values with first digit over 4 overflow */
-                case '0': case '1': case '2': case '3':
-                                        ac = ((ch - '0') << 6);
-                        if ((ch = *quoted++) < '0' || '7' < ch)
-                                goto error;
-                                        ac |= ((ch - '0') << 3);
-                        if ((ch = *quoted++) < '0' || '7' < ch)
-                                goto error;
-                                        ac |= (ch - '0');
-                                        ch = ac;
-                                        break;
-                                default:
-                        goto error;
-                        }
-                strbuf_addch(sb, ch);
-                }
-
-  error:
-        strbuf_setlen(sb, oldlen);
-        return -1;
-}
-
-/* quoting as a string literal for other languages */
-
-void perl_quote_print(FILE *stream, const char *src)
-{
-        const char sq = '\'';
-        const char bq = '\\';
-        char c;
-
-        fputc(sq, stream);
-        while ((c = *src++)) {
-                if (c == sq || c == bq)
-                        fputc(bq, stream);
-                fputc(c, stream);
-        }
-        fputc(sq, stream);
-}
-
-void python_quote_print(FILE *stream, const char *src)
-{
-        const char sq = '\'';
-        const char bq = '\\';
-        const char nl = '\n';
-        char c;
-
-        fputc(sq, stream);
-        while ((c = *src++)) {
-                if (c == nl) {
-                        fputc(bq, stream);
-                        fputc('n', stream);
-                        continue;
-                }
-                if (c == sq || c == bq)
-                        fputc(bq, stream);
-                fputc(c, stream);
-        }
-        fputc(sq, stream);
-}
-
-void tcl_quote_print(FILE *stream, const char *src)
-{
-        char c;
-
-        fputc('"', stream);
-        while ((c = *src++)) {
-                switch (c) {
-                case '[': case ']':
-                case '{': case '}':
-                case '$': case '\\': case '"':
-                        fputc('\\', stream);
-                default:
-                        fputc(c, stream);
-                        break;
-                case '\f':
-                        fputs("\\f", stream);
-                        break;
-                case '\r':
-                        fputs("\\r", stream);
-                        break;
-                case '\n':
-                        fputs("\\n", stream);
-                        break;
-                case '\t':
-                        fputs("\\t", stream);
-                        break;
-                case '\v':
-                        fputs("\\v", stream);
-                        break;
-                }
-        }
-        fputc('"', stream);
-}
diff --git a/tools/perf/util/quote.h b/tools/perf/util/quote.h
index 5dfad89816d..172889ea234 100644
--- a/tools/perf/util/quote.h
+++ b/tools/perf/util/quote.h
@@ -1,5 +1,5 @@
-#ifndef QUOTE_H
-#define QUOTE_H
+#ifndef __PERF_QUOTE_H
+#define __PERF_QUOTE_H
 
 #include <stddef.h>
 #include <stdio.h>
@@ -22,47 +22,8 @@
  *
  * Note that the above examples leak memory!  Remember to free result from
  * sq_quote() in a real application.
- *
- * sq_quote_buf() writes to an existing buffer of specified size; it
- * will return the number of characters that would have been written
- * excluding the final null regardless of the buffer size.
  */
 
-extern void sq_quote_print(FILE *stream, const char *src);
-
-extern void sq_quote_buf(struct strbuf *, const char *src);
 extern void sq_quote_argv(struct strbuf *, const char **argv, size_t maxlen);
 
-/* This unwraps what sq_quote() produces in place, but returns
- * NULL if the input does not look like what sq_quote would have
- * produced.
- */
-extern char *sq_dequote(char *);
-
-/*
- * Same as the above, but can be used to unwrap many arguments in the
- * same string separated by space. "next" is changed to point to the
- * next argument that should be passed as first parameter. When there
- * is no more argument to be dequoted, "next" is updated to point to NULL.
- */
-extern char *sq_dequote_step(char *arg, char **next);
-extern int sq_dequote_to_argv(char *arg, const char ***argv, int *nr, int *alloc);
-
-extern int unquote_c_style(struct strbuf *, const char *quoted, const char **endp);
-extern size_t quote_c_style(const char *name, struct strbuf *, FILE *, int no_dq);
-extern void quote_two_c_style(struct strbuf *, const char *, const char *, int);
-
-extern void write_name_quoted(const char *name, FILE *, int terminator);
-extern void write_name_quotedpfx(const char *pfx, size_t pfxlen,
-                                 const char *name, FILE *, int terminator);
-
-/* quote path as relative to the given prefix */
-char *quote_path_relative(const char *in, int len,
-                          struct strbuf *out, const char *prefix);
-
-/* quoting as a string literal for other languages */
-extern void perl_quote_print(FILE *stream, const char *src);
-extern void python_quote_print(FILE *stream, const char *src);
-extern void tcl_quote_print(FILE *stream, const char *src);
-
-#endif
+#endif /* __PERF_QUOTE_H */
diff --git a/tools/perf/util/rbtree.c b/tools/perf/util/rbtree.c
deleted file mode 100644
index b15ba9c7cb3..00000000000
--- a/tools/perf/util/rbtree.c
+++ /dev/null
@@ -1,383 +0,0 @@
-/*
-  Red Black Trees
-  (C) 1999  Andrea Arcangeli <andrea@suse.de>
-  (C) 2002  David Woodhouse <dwmw2@infradead.org>
-  
-  This program is free software; you can redistribute it and/or modify
-  it under the terms of the GNU General Public License as published by
-  the Free Software Foundation; either version 2 of the License, or
-  (at your option) any later version.
-
-  This program is distributed in the hope that it will be useful,
-  but WITHOUT ANY WARRANTY; without even the implied warranty of
-  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-  GNU General Public License for more details.
-
-  You should have received a copy of the GNU General Public License
-  along with this program; if not, write to the Free Software
-  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
-
-  linux/lib/rbtree.c
-*/
-
-#include "rbtree.h"
-
-static void __rb_rotate_left(struct rb_node *node, struct rb_root *root)
-{
-        struct rb_node *right = node->rb_right;
-        struct rb_node *parent = rb_parent(node);
-
-        if ((node->rb_right = right->rb_left))
-                rb_set_parent(right->rb_left, node);
-        right->rb_left = node;
-
-        rb_set_parent(right, parent);
-
-        if (parent)
-        {
-                if (node == parent->rb_left)
-                        parent->rb_left = right;
-                else
-                        parent->rb_right = right;
-        }
-        else
-                root->rb_node = right;
-        rb_set_parent(node, right);
-}
-
-static void __rb_rotate_right(struct rb_node *node, struct rb_root *root)
-{
-        struct rb_node *left = node->rb_left;
-        struct rb_node *parent = rb_parent(node);
-
-        if ((node->rb_left = left->rb_right))
-                rb_set_parent(left->rb_right, node);
-        left->rb_right = node;
-
-        rb_set_parent(left, parent);
-
-        if (parent)
-        {
-                if (node == parent->rb_right)
-                        parent->rb_right = left;
-                else
-                        parent->rb_left = left;
-        }
-        else
-                root->rb_node = left;
-        rb_set_parent(node, left);
-}
-
-void rb_insert_color(struct rb_node *node, struct rb_root *root)
-{
-        struct rb_node *parent, *gparent;
-
-        while ((parent = rb_parent(node)) && rb_is_red(parent))
-        {
-                gparent = rb_parent(parent);
-
-                if (parent == gparent->rb_left)
-                {
-                        {
-                                register struct rb_node *uncle = gparent->rb_right;
-                                if (uncle && rb_is_red(uncle))
-                                {
-                                        rb_set_black(uncle);
-                                        rb_set_black(parent);
-                                        rb_set_red(gparent);
-                                        node = gparent;
-                                        continue;
-                                }
-                        }
-
-                        if (parent->rb_right == node)
-                        {
-                                register struct rb_node *tmp;
-                                __rb_rotate_left(parent, root);
-                                tmp = parent;
-                                parent = node;
-                                node = tmp;
-                        }
-
-                        rb_set_black(parent);
-                        rb_set_red(gparent);
-                        __rb_rotate_right(gparent, root);
-                } else {
-                        {
-                                register struct rb_node *uncle = gparent->rb_left;
-                                if (uncle && rb_is_red(uncle))
-                                {
-                                        rb_set_black(uncle);
-                                        rb_set_black(parent);
-                                        rb_set_red(gparent);
-                                        node = gparent;
-                                        continue;
-                                }
-                        }
-
-                        if (parent->rb_left == node)
-                        {
-                                register struct rb_node *tmp;
-                                __rb_rotate_right(parent, root);
-                                tmp = parent;
-                                parent = node;
-                                node = tmp;
-                        }
-
-                        rb_set_black(parent);
-                        rb_set_red(gparent);
-                        __rb_rotate_left(gparent, root);
-                }
-        }
-
-        rb_set_black(root->rb_node);
-}
-
-static void __rb_erase_color(struct rb_node *node, struct rb_node *parent,
-                             struct rb_root *root)
-{
-        struct rb_node *other;
-
-        while ((!node || rb_is_black(node)) && node != root->rb_node)
-        {
-                if (parent->rb_left == node)
-                {
-                        other = parent->rb_right;
-                        if (rb_is_red(other))
-                        {
-                                rb_set_black(other);
-                                rb_set_red(parent);
-                                __rb_rotate_left(parent, root);
-                                other = parent->rb_right;
-                        }
-                        if ((!other->rb_left || rb_is_black(other->rb_left)) &&
-                            (!other->rb_right || rb_is_black(other->rb_right)))
-                        {
-                                rb_set_red(other);
-                                node = parent;
-                                parent = rb_parent(node);
-                        }
-                        else
-                        {
-                                if (!other->rb_right || rb_is_black(other->rb_right))
-                                {
-                                        rb_set_black(other->rb_left);
-                                        rb_set_red(other);
-                                        __rb_rotate_right(other, root);
-                                        other = parent->rb_right;
-                                }
-                                rb_set_color(other, rb_color(parent));
-                                rb_set_black(parent);
-                                rb_set_black(other->rb_right);
-                                __rb_rotate_left(parent, root);
-                                node = root->rb_node;
-                                break;
-                        }
-                }
-                else
-                {
-                        other = parent->rb_left;
-                        if (rb_is_red(other))
-                        {
-                                rb_set_black(other);
-                                rb_set_red(parent);
-                                __rb_rotate_right(parent, root);
-                                other = parent->rb_left;
-                        }
-                        if ((!other->rb_left || rb_is_black(other->rb_left)) &&
-                            (!other->rb_right || rb_is_black(other->rb_right)))
-                        {
-                                rb_set_red(other);
-                                node = parent;
-                                parent = rb_parent(node);
-                        }
-                        else
-                        {
-                                if (!other->rb_left || rb_is_black(other->rb_left))
-                                {
-                                        rb_set_black(other->rb_right);
-                                        rb_set_red(other);
-                                        __rb_rotate_left(other, root);
-                                        other = parent->rb_left;
-                                }
-                                rb_set_color(other, rb_color(parent));
-                                rb_set_black(parent);
-                                rb_set_black(other->rb_left);
-                                __rb_rotate_right(parent, root);
-                                node = root->rb_node;
-                                break;
-                        }
-                }
-        }
-        if (node)
-                rb_set_black(node);
-}
-
-void rb_erase(struct rb_node *node, struct rb_root *root)
-{
-        struct rb_node *child, *parent;
-        int color;
-
-        if (!node->rb_left)
-                child = node->rb_right;
-        else if (!node->rb_right)
-                child = node->rb_left;
-        else
-        {
-                struct rb_node *old = node, *left;
-
-                node = node->rb_right;
-                while ((left = node->rb_left) != NULL)
-                        node = left;
-                child = node->rb_right;
-                parent = rb_parent(node);
-                color = rb_color(node);
-
-                if (child)
-                        rb_set_parent(child, parent);
-                if (parent == old) {
-                        parent->rb_right = child;
-                        parent = node;
-                } else
-                        parent->rb_left = child;
-
-                node->rb_parent_color = old->rb_parent_color;
-                node->rb_right = old->rb_right;
-                node->rb_left = old->rb_left;
-
-                if (rb_parent(old))
-                {
-                        if (rb_parent(old)->rb_left == old)
-                                rb_parent(old)->rb_left = node;
-                        else
-                                rb_parent(old)->rb_right = node;
-                } else
-                        root->rb_node = node;
-
-                rb_set_parent(old->rb_left, node);
-                if (old->rb_right)
-                        rb_set_parent(old->rb_right, node);
-                goto color;
-        }
-
-        parent = rb_parent(node);
-        color = rb_color(node);
-
-        if (child)
-                rb_set_parent(child, parent);
-        if (parent)
-        {
-                if (parent->rb_left == node)
-                        parent->rb_left = child;
-                else
-                        parent->rb_right = child;
-        }
-        else
-                root->rb_node = child;
-
- color:
-        if (color == RB_BLACK)
-                __rb_erase_color(child, parent, root);
-}
-
-/*
- * This function returns the first node (in sort order) of the tree.
- */
-struct rb_node *rb_first(const struct rb_root *root)
-{
-        struct rb_node  *n;
-
-        n = root->rb_node;
-        if (!n)
-                return NULL;
-        while (n->rb_left)
-                n = n->rb_left;
-        return n;
-}
-
-struct rb_node *rb_last(const struct rb_root *root)
-{
-        struct rb_node  *n;
-
-        n = root->rb_node;
-        if (!n)
-                return NULL;
-        while (n->rb_right)
-                n = n->rb_right;
-        return n;
-}
-
-struct rb_node *rb_next(const struct rb_node *node)
-{
-        struct rb_node *parent;
-
-        if (rb_parent(node) == node)
-                return NULL;
-
-        /* If we have a right-hand child, go down and then left as far
-           as we can. */
-        if (node->rb_right) {
-                node = node->rb_right; 
-                while (node->rb_left)
-                        node=node->rb_left;
-                return (struct rb_node *)node;
-        }
-
-        /* No right-hand children.  Everything down and left is
-           smaller than us, so any 'next' node must be in the general
-           direction of our parent. Go up the tree; any time the
-           ancestor is a right-hand child of its parent, keep going
-           up. First time it's a left-hand child of its parent, said
-           parent is our 'next' node. */
-        while ((parent = rb_parent(node)) && node == parent->rb_right)
-                node = parent;
-
-        return parent;
-}
-
-struct rb_node *rb_prev(const struct rb_node *node)
-{
-        struct rb_node *parent;
-
-        if (rb_parent(node) == node)
-                return NULL;
-
-        /* If we have a left-hand child, go down and then right as far
-           as we can. */
-        if (node->rb_left) {
-                node = node->rb_left; 
-                while (node->rb_right)
-                        node=node->rb_right;
-                return (struct rb_node *)node;
-        }
-
-        /* No left-hand children. Go up till we find an ancestor which
-           is a right-hand child of its parent */
-        while ((parent = rb_parent(node)) && node == parent->rb_left)
-                node = parent;
-
-        return parent;
-}
-
-void rb_replace_node(struct rb_node *victim, struct rb_node *new,
-                     struct rb_root *root)
-{
-        struct rb_node *parent = rb_parent(victim);
-
-        /* Set the surrounding nodes to point to the replacement */
-        if (parent) {
-                if (victim == parent->rb_left)
-                        parent->rb_left = new;
-                else
-                        parent->rb_right = new;
-        } else {
-                root->rb_node = new;
-        }
-        if (victim->rb_left)
-                rb_set_parent(victim->rb_left, new);
-        if (victim->rb_right)
-                rb_set_parent(victim->rb_right, new);
-
-        /* Copy the pointers/colour from the victim to the replacement */
-        *new = *victim;
-}
diff --git a/tools/perf/util/rbtree.h b/tools/perf/util/rbtree.h
deleted file mode 100644
index 6bdc488a47f..00000000000
--- a/tools/perf/util/rbtree.h
+++ /dev/null
@@ -1,171 +0,0 @@
-/*
-  Red Black Trees
-  (C) 1999  Andrea Arcangeli <andrea@suse.de>
-  
-  This program is free software; you can redistribute it and/or modify
-  it under the terms of the GNU General Public License as published by
-  the Free Software Foundation; either version 2 of the License, or
-  (at your option) any later version.
-
-  This program is distributed in the hope that it will be useful,
-  but WITHOUT ANY WARRANTY; without even the implied warranty of
-  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-  GNU General Public License for more details.
-
-  You should have received a copy of the GNU General Public License
-  along with this program; if not, write to the Free Software
-  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
-
-  linux/include/linux/rbtree.h
-
-  To use rbtrees you'll have to implement your own insert and search cores.
-  This will avoid us to use callbacks and to drop drammatically performances.
-  I know it's not the cleaner way,  but in C (not in C++) to get
-  performances and genericity...
-
-  Some example of insert and search follows here. The search is a plain
-  normal search over an ordered tree. The insert instead must be implemented
-  int two steps: as first thing the code must insert the element in
-  order as a red leaf in the tree, then the support library function
-  rb_insert_color() must be called. Such function will do the
-  not trivial work to rebalance the rbtree if necessary.
-
------------------------------------------------------------------------
-static inline struct page * rb_search_page_cache(struct inode * inode,
-                                                 unsigned long offset)
-{
-        struct rb_node * n = inode->i_rb_page_cache.rb_node;
-        struct page * page;
-
-        while (n)
-        {
-                page = rb_entry(n, struct page, rb_page_cache);
-
-                if (offset < page->offset)
-                        n = n->rb_left;
-                else if (offset > page->offset)
-                        n = n->rb_right;
-                else
-                        return page;
-        }
-        return NULL;
-}
-
-static inline struct page * __rb_insert_page_cache(struct inode * inode,
-                                                   unsigned long offset,
-                                                   struct rb_node * node)
-{
-        struct rb_node ** p = &inode->i_rb_page_cache.rb_node;
-        struct rb_node * parent = NULL;
-        struct page * page;
-
-        while (*p)
-        {
-                parent = *p;
-                page = rb_entry(parent, struct page, rb_page_cache);
-
-                if (offset < page->offset)
-                        p = &(*p)->rb_left;
-                else if (offset > page->offset)
-                        p = &(*p)->rb_right;
-                else
-                        return page;
-        }
-
-        rb_link_node(node, parent, p);
-
-        return NULL;
-}
-
-static inline struct page * rb_insert_page_cache(struct inode * inode,
-                                                 unsigned long offset,
-                                                 struct rb_node * node)
-{
-        struct page * ret;
-        if ((ret = __rb_insert_page_cache(inode, offset, node)))
-                goto out;
-        rb_insert_color(node, &inode->i_rb_page_cache);
- out:
-        return ret;
-}
------------------------------------------------------------------------
-*/
-
-#ifndef _LINUX_RBTREE_H
-#define _LINUX_RBTREE_H
-
-#include <stddef.h>
-
-/**
- * container_of - cast a member of a structure out to the containing structure
- * @ptr:        the pointer to the member.
- * @type:       the type of the container struct this is embedded in.
- * @member:     the name of the member within the struct.
- *
- */
-#define container_of(ptr, type, member) ({                      \
-        const typeof( ((type *)0)->member ) *__mptr = (ptr);    \
-        (type *)( (char *)__mptr - offsetof(type,member) );})
-
-struct rb_node
-{
-        unsigned long  rb_parent_color;
-#define RB_RED          0
-#define RB_BLACK        1
-        struct rb_node *rb_right;
-        struct rb_node *rb_left;
-} __attribute__((aligned(sizeof(long))));
-    /* The alignment might seem pointless, but allegedly CRIS needs it */
-
-struct rb_root
-{
-        struct rb_node *rb_node;
-};
-
-
-#define rb_parent(r)   ((struct rb_node *)((r)->rb_parent_color & ~3))
-#define rb_color(r)   ((r)->rb_parent_color & 1)
-#define rb_is_red(r)   (!rb_color(r))
-#define rb_is_black(r) rb_color(r)
-#define rb_set_red(r)  do { (r)->rb_parent_color &= ~1; } while (0)
-#define rb_set_black(r)  do { (r)->rb_parent_color |= 1; } while (0)
-
-static inline void rb_set_parent(struct rb_node *rb, struct rb_node *p)
-{
-        rb->rb_parent_color = (rb->rb_parent_color & 3) | (unsigned long)p;
-}
-static inline void rb_set_color(struct rb_node *rb, int color)
-{
-        rb->rb_parent_color = (rb->rb_parent_color & ~1) | color;
-}
-
-#define RB_ROOT (struct rb_root) { NULL, }
-#define rb_entry(ptr, type, member) container_of(ptr, type, member)
-
-#define RB_EMPTY_ROOT(root)     ((root)->rb_node == NULL)
-#define RB_EMPTY_NODE(node)     (rb_parent(node) == node)
-#define RB_CLEAR_NODE(node)     (rb_set_parent(node, node))
-
-extern void rb_insert_color(struct rb_node *, struct rb_root *);
-extern void rb_erase(struct rb_node *, struct rb_root *);
-
-/* Find logical next and previous nodes in a tree */
-extern struct rb_node *rb_next(const struct rb_node *);
-extern struct rb_node *rb_prev(const struct rb_node *);
-extern struct rb_node *rb_first(const struct rb_root *);
-extern struct rb_node *rb_last(const struct rb_root *);
-
-/* Fast replacement of a single node without remove/rebalance/add/rebalance */
-extern void rb_replace_node(struct rb_node *victim, struct rb_node *new, 
-                            struct rb_root *root);
-
-static inline void rb_link_node(struct rb_node * node, struct rb_node * parent,
-                                struct rb_node ** rb_link)
-{
-        node->rb_parent_color = (unsigned long )parent;
-        node->rb_left = node->rb_right = NULL;
-
-        *rb_link = node;
-}
-
-#endif  /* _LINUX_RBTREE_H */
diff --git a/tools/perf/util/run-command.c b/tools/perf/util/run-command.c
index b2f5e854f40..da8e9b285f5 100644
--- a/tools/perf/util/run-command.c
+++ b/tools/perf/util/run-command.c
@@ -65,7 +65,6 @@ int start_command(struct child_process *cmd)
                 cmd->err = fderr[0];
         }
 
-#ifndef __MINGW32__
         fflush(NULL);
         cmd->pid = fork();
         if (!cmd->pid) {
@@ -118,71 +117,6 @@ int start_command(struct child_process *cmd)
                 }
                 exit(127);
         }
-#else
-        int s0 = -1, s1 = -1, s2 = -1;  /* backups of stdin, stdout, stderr */
-        const char **sargv = cmd->argv;
-        char **env = environ;
-
-        if (cmd->no_stdin) {
-                s0 = dup(0);
-                dup_devnull(0);
-        } else if (need_in) {
-                s0 = dup(0);
-                dup2(fdin[0], 0);
-        } else if (cmd->in) {
-                s0 = dup(0);
-                dup2(cmd->in, 0);
-        }
-
-        if (cmd->no_stderr) {
-                s2 = dup(2);
-                dup_devnull(2);
-        } else if (need_err) {
-                s2 = dup(2);
-                dup2(fderr[1], 2);
-        }
-
-        if (cmd->no_stdout) {
-                s1 = dup(1);
-                dup_devnull(1);
-        } else if (cmd->stdout_to_stderr) {
-                s1 = dup(1);
-                dup2(2, 1);
-        } else if (need_out) {
-                s1 = dup(1);
-                dup2(fdout[1], 1);
-        } else if (cmd->out > 1) {
-                s1 = dup(1);
-                dup2(cmd->out, 1);
-        }
-
-        if (cmd->dir)
-                die("chdir in start_command() not implemented");
-        if (cmd->env) {
-                env = copy_environ();
-                for (; *cmd->env; cmd->env++)
-                        env = env_setenv(env, *cmd->env);
-        }
-
-        if (cmd->perf_cmd) {
-                cmd->argv = prepare_perf_cmd(cmd->argv);
-        }
-
-        cmd->pid = mingw_spawnvpe(cmd->argv[0], cmd->argv, env);
-
-        if (cmd->env)
-                free_environ(env);
-        if (cmd->perf_cmd)
-                free(cmd->argv);
-
-        cmd->argv = sargv;
-        if (s0 >= 0)
-                dup2(s0, 0), close(s0);
-        if (s1 >= 0)
-                dup2(s1, 1), close(s1);
-        if (s2 >= 0)
-                dup2(s2, 2), close(s2);
-#endif
 
         if (cmd->pid < 0) {
                 int err = errno;
@@ -278,118 +212,3 @@ int run_command_v_opt(const char **argv, int opt)
         prepare_run_command_v_opt(&cmd, argv, opt);
         return run_command(&cmd);
 }
-
-int run_command_v_opt_cd_env(const char **argv, int opt, const char *dir, const char *const *env)
-{
-        struct child_process cmd;
-        prepare_run_command_v_opt(&cmd, argv, opt);
-        cmd.dir = dir;
-        cmd.env = env;
-        return run_command(&cmd);
-}
-
-#ifdef __MINGW32__
-static __stdcall unsigned run_thread(void *data)
-{
-        struct async *async = data;
-        return async->proc(async->fd_for_proc, async->data);
-}
-#endif
-
-int start_async(struct async *async)
-{
-        int pipe_out[2];
-
-        if (pipe(pipe_out) < 0)
-                return error("cannot create pipe: %s", strerror(errno));
-        async->out = pipe_out[0];
-
-#ifndef __MINGW32__
-        /* Flush stdio before fork() to avoid cloning buffers */
-        fflush(NULL);
-
-        async->pid = fork();
-        if (async->pid < 0) {
-                error("fork (async) failed: %s", strerror(errno));
-                close_pair(pipe_out);
-                return -1;
-        }
-        if (!async->pid) {
-                close(pipe_out[0]);
-                exit(!!async->proc(pipe_out[1], async->data));
-        }
-        close(pipe_out[1]);
-#else
-        async->fd_for_proc = pipe_out[1];
-        async->tid = (HANDLE) _beginthreadex(NULL, 0, run_thread, async, 0, NULL);
-        if (!async->tid) {
-                error("cannot create thread: %s", strerror(errno));
-                close_pair(pipe_out);
-                return -1;
-        }
-#endif
-        return 0;
-}
-
-int finish_async(struct async *async)
-{
-#ifndef __MINGW32__
-        int ret = 0;
-
-        if (wait_or_whine(async->pid))
-                ret = error("waitpid (async) failed");
-#else
-        DWORD ret = 0;
-        if (WaitForSingleObject(async->tid, INFINITE) != WAIT_OBJECT_0)
-                ret = error("waiting for thread failed: %lu", GetLastError());
-        else if (!GetExitCodeThread(async->tid, &ret))
-                ret = error("cannot get thread exit code: %lu", GetLastError());
-        CloseHandle(async->tid);
-#endif
-        return ret;
-}
-
-int run_hook(const char *index_file, const char *name, ...)
-{
-        struct child_process hook;
-        const char **argv = NULL, *env[2];
-        char index[PATH_MAX];
-        va_list args;
-        int ret;
-        size_t i = 0, alloc = 0;
-
-        if (access(perf_path("hooks/%s", name), X_OK) < 0)
-                return 0;
-
-        va_start(args, name);
-        ALLOC_GROW(argv, i + 1, alloc);
-        argv[i++] = perf_path("hooks/%s", name);
-        while (argv[i-1]) {
-                ALLOC_GROW(argv, i + 1, alloc);
-                argv[i++] = va_arg(args, const char *);
-        }
-        va_end(args);
-
-        memset(&hook, 0, sizeof(hook));
-        hook.argv = argv;
-        hook.no_stdin = 1;
-        hook.stdout_to_stderr = 1;
-        if (index_file) {
-                snprintf(index, sizeof(index), "PERF_INDEX_FILE=%s", index_file);
-                env[0] = index;
-                env[1] = NULL;
-                hook.env = env;
-        }
-
-        ret = start_command(&hook);
-        free(argv);
-        if (ret) {
-                warning("Could not spawn %s", argv[0]);
-                return ret;
-        }
-        ret = finish_command(&hook);
-        if (ret == -ERR_RUN_COMMAND_WAITPID_SIGNAL)
-                warning("%s exited due to uncaught signal", argv[0]);
-
-        return ret;
-}
diff --git a/tools/perf/util/run-command.h b/tools/perf/util/run-command.h
index 328289f2366..1ef264d5069 100644
--- a/tools/perf/util/run-command.h
+++ b/tools/perf/util/run-command.h
@@ -1,5 +1,5 @@
-#ifndef RUN_COMMAND_H
-#define RUN_COMMAND_H
+#ifndef __PERF_RUN_COMMAND_H
+#define __PERF_RUN_COMMAND_H
 
 enum {
         ERR_RUN_COMMAND_FORK = 10000,
@@ -50,44 +50,9 @@ int start_command(struct child_process *);
 int finish_command(struct child_process *);
 int run_command(struct child_process *);
 
-extern int run_hook(const char *index_file, const char *name, ...);
-
 #define RUN_COMMAND_NO_STDIN 1
 #define RUN_PERF_CMD         2  /*If this is to be perf sub-command */
 #define RUN_COMMAND_STDOUT_TO_STDERR 4
 int run_command_v_opt(const char **argv, int opt);
 
-/*
- * env (the environment) is to be formatted like environ: "VAR=VALUE".
- * To unset an environment variable use just "VAR".
- */
-int run_command_v_opt_cd_env(const char **argv, int opt, const char *dir, const char *const *env);
-
-/*
- * The purpose of the following functions is to feed a pipe by running
- * a function asynchronously and providing output that the caller reads.
- *
- * It is expected that no synchronization and mutual exclusion between
- * the caller and the feed function is necessary so that the function
- * can run in a thread without interfering with the caller.
- */
-struct async {
-        /*
-         * proc writes to fd and closes it;
-         * returns 0 on success, non-zero on failure
-         */
-        int (*proc)(int fd, void *data);
-        void *data;
-        int out;        /* caller reads from here and closes it */
-#ifndef __MINGW32__
-        pid_t pid;
-#else
-        HANDLE tid;
-        int fd_for_proc;
-#endif
-};
-
-int start_async(struct async *async);
-int finish_async(struct async *async);
-
-#endif
+#endif /* __PERF_RUN_COMMAND_H */
diff --git a/tools/perf/util/scripting-engines/trace-event-perl.c b/tools/perf/util/scripting-engines/trace-event-perl.c
new file mode 100644
index 00000000000..93680818e24
--- /dev/null
+++ b/tools/perf/util/scripting-engines/trace-event-perl.c
@@ -0,0 +1,565 @@
+/*
+ * trace-event-perl.  Feed perf script events to an embedded Perl interpreter.
+ *
+ * Copyright (C) 2009 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+#include <errno.h>
+
+#include "../../perf.h"
+#include "../util.h"
+#include "../trace-event.h"
+
+#include <EXTERN.h>
+#include <perl.h>
+
+void boot_Perf__Trace__Context(pTHX_ CV *cv);
+void boot_DynaLoader(pTHX_ CV *cv);
+typedef PerlInterpreter * INTERP;
+
+void xs_init(pTHX);
+
+void xs_init(pTHX)
+{
+        const char *file = __FILE__;
+        dXSUB_SYS;
+
+        newXS("Perf::Trace::Context::bootstrap", boot_Perf__Trace__Context,
+              file);
+        newXS("DynaLoader::boot_DynaLoader", boot_DynaLoader, file);
+}
+
+INTERP my_perl;
+
+#define FTRACE_MAX_EVENT                                \
+        ((1 << (sizeof(unsigned short) * 8)) - 1)
+
+struct event *events[FTRACE_MAX_EVENT];
+
+extern struct scripting_context *scripting_context;
+
+static char *cur_field_name;
+static int zero_flag_atom;
+
+static void define_symbolic_value(const char *ev_name,
+                                  const char *field_name,
+                                  const char *field_value,
+                                  const char *field_str)
+{
+        unsigned long long value;
+        dSP;
+
+        value = eval_flag(field_value);
+
+        ENTER;
+        SAVETMPS;
+        PUSHMARK(SP);
+
+        XPUSHs(sv_2mortal(newSVpv(ev_name, 0)));
+        XPUSHs(sv_2mortal(newSVpv(field_name, 0)));
+        XPUSHs(sv_2mortal(newSVuv(value)));
+        XPUSHs(sv_2mortal(newSVpv(field_str, 0)));
+
+        PUTBACK;
+        if (get_cv("main::define_symbolic_value", 0))
+                call_pv("main::define_symbolic_value", G_SCALAR);
+        SPAGAIN;
+        PUTBACK;
+        FREETMPS;
+        LEAVE;
+}
+
+static void define_symbolic_values(struct print_flag_sym *field,
+                                   const char *ev_name,
+                                   const char *field_name)
+{
+        define_symbolic_value(ev_name, field_name, field->value, field->str);
+        if (field->next)
+                define_symbolic_values(field->next, ev_name, field_name);
+}
+
+static void define_symbolic_field(const char *ev_name,
+                                  const char *field_name)
+{
+        dSP;
+
+        ENTER;
+        SAVETMPS;
+        PUSHMARK(SP);
+
+        XPUSHs(sv_2mortal(newSVpv(ev_name, 0)));
+        XPUSHs(sv_2mortal(newSVpv(field_name, 0)));
+
+        PUTBACK;
+        if (get_cv("main::define_symbolic_field", 0))
+                call_pv("main::define_symbolic_field", G_SCALAR);
+        SPAGAIN;
+        PUTBACK;
+        FREETMPS;
+        LEAVE;
+}
+
+static void define_flag_value(const char *ev_name,
+                              const char *field_name,
+                              const char *field_value,
+                              const char *field_str)
+{
+        unsigned long long value;
+        dSP;
+
+        value = eval_flag(field_value);
+
+        ENTER;
+        SAVETMPS;
+        PUSHMARK(SP);
+
+        XPUSHs(sv_2mortal(newSVpv(ev_name, 0)));
+        XPUSHs(sv_2mortal(newSVpv(field_name, 0)));
+        XPUSHs(sv_2mortal(newSVuv(value)));
+        XPUSHs(sv_2mortal(newSVpv(field_str, 0)));
+
+        PUTBACK;
+        if (get_cv("main::define_flag_value", 0))
+                call_pv("main::define_flag_value", G_SCALAR);
+        SPAGAIN;
+        PUTBACK;
+        FREETMPS;
+        LEAVE;
+}
+
+static void define_flag_values(struct print_flag_sym *field,
+                               const char *ev_name,
+                               const char *field_name)
+{
+        define_flag_value(ev_name, field_name, field->value, field->str);
+        if (field->next)
+                define_flag_values(field->next, ev_name, field_name);
+}
+
+static void define_flag_field(const char *ev_name,
+                              const char *field_name,
+                              const char *delim)
+{
+        dSP;
+
+        ENTER;
+        SAVETMPS;
+        PUSHMARK(SP);
+
+        XPUSHs(sv_2mortal(newSVpv(ev_name, 0)));
+        XPUSHs(sv_2mortal(newSVpv(field_name, 0)));
+        XPUSHs(sv_2mortal(newSVpv(delim, 0)));
+
+        PUTBACK;
+        if (get_cv("main::define_flag_field", 0))
+                call_pv("main::define_flag_field", G_SCALAR);
+        SPAGAIN;
+        PUTBACK;
+        FREETMPS;
+        LEAVE;
+}
+
+static void define_event_symbols(struct event *event,
+                                 const char *ev_name,
+                                 struct print_arg *args)
+{
+        switch (args->type) {
+        case PRINT_NULL:
+                break;
+        case PRINT_ATOM:
+                define_flag_value(ev_name, cur_field_name, "0",
+                                  args->atom.atom);
+                zero_flag_atom = 0;
+                break;
+        case PRINT_FIELD:
+                if (cur_field_name)
+                        free(cur_field_name);
+                cur_field_name = strdup(args->field.name);
+                break;
+        case PRINT_FLAGS:
+                define_event_symbols(event, ev_name, args->flags.field);
+                define_flag_field(ev_name, cur_field_name, args->flags.delim);
+                define_flag_values(args->flags.flags, ev_name, cur_field_name);
+                break;
+        case PRINT_SYMBOL:
+                define_event_symbols(event, ev_name, args->symbol.field);
+                define_symbolic_field(ev_name, cur_field_name);
+                define_symbolic_values(args->symbol.symbols, ev_name,
+                                       cur_field_name);
+                break;
+        case PRINT_STRING:
+                break;
+        case PRINT_TYPE:
+                define_event_symbols(event, ev_name, args->typecast.item);
+                break;
+        case PRINT_OP:
+                if (strcmp(args->op.op, ":") == 0)
+                        zero_flag_atom = 1;
+                define_event_symbols(event, ev_name, args->op.left);
+                define_event_symbols(event, ev_name, args->op.right);
+                break;
+        default:
+                /* we should warn... */
+                return;
+        }
+
+        if (args->next)
+                define_event_symbols(event, ev_name, args->next);
+}
+
+static inline struct event *find_cache_event(int type)
+{
+        static char ev_name[256];
+        struct event *event;
+
+        if (events[type])
+                return events[type];
+
+        events[type] = event = trace_find_event(type);
+        if (!event)
+                return NULL;
+
+        sprintf(ev_name, "%s::%s", event->system, event->name);
+
+        define_event_symbols(event, ev_name, event->print_fmt.args);
+
+        return event;
+}
+
+static void perl_process_event(int cpu, void *data,
+                               int size __unused,
+                               unsigned long long nsecs, char *comm)
+{
+        struct format_field *field;
+        static char handler[256];
+        unsigned long long val;
+        unsigned long s, ns;
+        struct event *event;
+        int type;
+        int pid;
+
+        dSP;
+
+        type = trace_parse_common_type(data);
+
+        event = find_cache_event(type);
+        if (!event)
+                die("ug! no event found for type %d", type);
+
+        pid = trace_parse_common_pid(data);
+
+        sprintf(handler, "%s::%s", event->system, event->name);
+
+        s = nsecs / NSECS_PER_SEC;
+        ns = nsecs - s * NSECS_PER_SEC;
+
+        scripting_context->event_data = data;
+
+        ENTER;
+        SAVETMPS;
+        PUSHMARK(SP);
+
+        XPUSHs(sv_2mortal(newSVpv(handler, 0)));
+        XPUSHs(sv_2mortal(newSViv(PTR2IV(scripting_context))));
+        XPUSHs(sv_2mortal(newSVuv(cpu)));
+        XPUSHs(sv_2mortal(newSVuv(s)));
+        XPUSHs(sv_2mortal(newSVuv(ns)));
+        XPUSHs(sv_2mortal(newSViv(pid)));
+        XPUSHs(sv_2mortal(newSVpv(comm, 0)));
+
+        /* common fields other than pid can be accessed via xsub fns */
+
+        for (field = event->format.fields; field; field = field->next) {
+                if (field->flags & FIELD_IS_STRING) {
+                        int offset;
+                        if (field->flags & FIELD_IS_DYNAMIC) {
+                                offset = *(int *)(data + field->offset);
+                                offset &= 0xffff;
+                        } else
+                                offset = field->offset;
+                        XPUSHs(sv_2mortal(newSVpv((char *)data + offset, 0)));
+                } else { /* FIELD_IS_NUMERIC */
+                        val = read_size(data + field->offset, field->size);
+                        if (field->flags & FIELD_IS_SIGNED) {
+                                XPUSHs(sv_2mortal(newSViv(val)));
+                        } else {
+                                XPUSHs(sv_2mortal(newSVuv(val)));
+                        }
+                }
+        }
+
+        PUTBACK;
+
+        if (get_cv(handler, 0))
+                call_pv(handler, G_SCALAR);
+        else if (get_cv("main::trace_unhandled", 0)) {
+                XPUSHs(sv_2mortal(newSVpv(handler, 0)));
+                XPUSHs(sv_2mortal(newSViv(PTR2IV(scripting_context))));
+                XPUSHs(sv_2mortal(newSVuv(cpu)));
+                XPUSHs(sv_2mortal(newSVuv(nsecs)));
+                XPUSHs(sv_2mortal(newSViv(pid)));
+                XPUSHs(sv_2mortal(newSVpv(comm, 0)));
+                call_pv("main::trace_unhandled", G_SCALAR);
+        }
+        SPAGAIN;
+        PUTBACK;
+        FREETMPS;
+        LEAVE;
+}
+
+static void run_start_sub(void)
+{
+        dSP; /* access to Perl stack */
+        PUSHMARK(SP);
+
+        if (get_cv("main::trace_begin", 0))
+                call_pv("main::trace_begin", G_DISCARD | G_NOARGS);
+}
+
+/*
+ * Start trace script
+ */
+static int perl_start_script(const char *script, int argc, const char **argv)
+{
+        const char **command_line;
+        int i, err = 0;
+
+        command_line = malloc((argc + 2) * sizeof(const char *));
+        command_line[0] = "";
+        command_line[1] = script;
+        for (i = 2; i < argc + 2; i++)
+                command_line[i] = argv[i - 2];
+
+        my_perl = perl_alloc();
+        perl_construct(my_perl);
+
+        if (perl_parse(my_perl, xs_init, argc + 2, (char **)command_line,
+                       (char **)NULL)) {
+                err = -1;
+                goto error;
+        }
+
+        if (perl_run(my_perl)) {
+                err = -1;
+                goto error;
+        }
+
+        if (SvTRUE(ERRSV)) {
+                err = -1;
+                goto error;
+        }
+
+        run_start_sub();
+
+        free(command_line);
+        return 0;
+error:
+        perl_free(my_perl);
+        free(command_line);
+
+        return err;
+}
+
+/*
+ * Stop trace script
+ */
+static int perl_stop_script(void)
+{
+        dSP; /* access to Perl stack */
+        PUSHMARK(SP);
+
+        if (get_cv("main::trace_end", 0))
+                call_pv("main::trace_end", G_DISCARD | G_NOARGS);
+
+        perl_destruct(my_perl);
+        perl_free(my_perl);
+
+        return 0;
+}
+
+static int perl_generate_script(const char *outfile)
+{
+        struct event *event = NULL;
+        struct format_field *f;
+        char fname[PATH_MAX];
+        int not_first, count;
+        FILE *ofp;
+
+        sprintf(fname, "%s.pl", outfile);
+        ofp = fopen(fname, "w");
+        if (ofp == NULL) {
+                fprintf(stderr, "couldn't open %s\n", fname);
+                return -1;
+        }
+
+        fprintf(ofp, "# perf script event handlers, "
+                "generated by perf script -g perl\n");
+
+        fprintf(ofp, "# Licensed under the terms of the GNU GPL"
+                " License version 2\n\n");
+
+        fprintf(ofp, "# The common_* event handler fields are the most useful "
+                "fields common to\n");
+
+        fprintf(ofp, "# all events.  They don't necessarily correspond to "
+                "the 'common_*' fields\n");
+
+        fprintf(ofp, "# in the format files.  Those fields not available as "
+                "handler params can\n");
+
+        fprintf(ofp, "# be retrieved using Perl functions of the form "
+                "common_*($context).\n");
+
+        fprintf(ofp, "# See Context.pm for the list of available "
+                "functions.\n\n");
+
+        fprintf(ofp, "use lib \"$ENV{'PERF_EXEC_PATH'}/scripts/perl/"
+                "Perf-Trace-Util/lib\";\n");
+
+        fprintf(ofp, "use lib \"./Perf-Trace-Util/lib\";\n");
+        fprintf(ofp, "use Perf::Trace::Core;\n");
+        fprintf(ofp, "use Perf::Trace::Context;\n");
+        fprintf(ofp, "use Perf::Trace::Util;\n\n");
+
+        fprintf(ofp, "sub trace_begin\n{\n\t# optional\n}\n\n");
+        fprintf(ofp, "sub trace_end\n{\n\t# optional\n}\n\n");
+
+        while ((event = trace_find_next_event(event))) {
+                fprintf(ofp, "sub %s::%s\n{\n", event->system, event->name);
+                fprintf(ofp, "\tmy (");
+
+                fprintf(ofp, "$event_name, ");
+                fprintf(ofp, "$context, ");
+                fprintf(ofp, "$common_cpu, ");
+                fprintf(ofp, "$common_secs, ");
+                fprintf(ofp, "$common_nsecs,\n");
+                fprintf(ofp, "\t    $common_pid, ");
+                fprintf(ofp, "$common_comm,\n\t    ");
+
+                not_first = 0;
+                count = 0;
+
+                for (f = event->format.fields; f; f = f->next) {
+                        if (not_first++)
+                                fprintf(ofp, ", ");
+                        if (++count % 5 == 0)
+                                fprintf(ofp, "\n\t    ");
+
+                        fprintf(ofp, "$%s", f->name);
+                }
+                fprintf(ofp, ") = @_;\n\n");
+
+                fprintf(ofp, "\tprint_header($event_name, $common_cpu, "
+                        "$common_secs, $common_nsecs,\n\t             "
+                        "$common_pid, $common_comm);\n\n");
+
+                fprintf(ofp, "\tprintf(\"");
+
+                not_first = 0;
+                count = 0;
+
+                for (f = event->format.fields; f; f = f->next) {
+                        if (not_first++)
+                                fprintf(ofp, ", ");
+                        if (count && count % 4 == 0) {
+                                fprintf(ofp, "\".\n\t       \"");
+                        }
+                        count++;
+
+                        fprintf(ofp, "%s=", f->name);
+                        if (f->flags & FIELD_IS_STRING ||
+                            f->flags & FIELD_IS_FLAG ||
+                            f->flags & FIELD_IS_SYMBOLIC)
+                                fprintf(ofp, "%%s");
+                        else if (f->flags & FIELD_IS_SIGNED)
+                                fprintf(ofp, "%%d");
+                        else
+                                fprintf(ofp, "%%u");
+                }
+
+                fprintf(ofp, "\\n\",\n\t       ");
+
+                not_first = 0;
+                count = 0;
+
+                for (f = event->format.fields; f; f = f->next) {
+                        if (not_first++)
+                                fprintf(ofp, ", ");
+
+                        if (++count % 5 == 0)
+                                fprintf(ofp, "\n\t       ");
+
+                        if (f->flags & FIELD_IS_FLAG) {
+                                if ((count - 1) % 5 != 0) {
+                                        fprintf(ofp, "\n\t       ");
+                                        count = 4;
+                                }
+                                fprintf(ofp, "flag_str(\"");
+                                fprintf(ofp, "%s::%s\", ", event->system,
+                                        event->name);
+                                fprintf(ofp, "\"%s\", $%s)", f->name,
+                                        f->name);
+                        } else if (f->flags & FIELD_IS_SYMBOLIC) {
+                                if ((count - 1) % 5 != 0) {
+                                        fprintf(ofp, "\n\t       ");
+                                        count = 4;
+                                }
+                                fprintf(ofp, "symbol_str(\"");
+                                fprintf(ofp, "%s::%s\", ", event->system,
+                                        event->name);
+                                fprintf(ofp, "\"%s\", $%s)", f->name,
+                                        f->name);
+                        } else
+                                fprintf(ofp, "$%s", f->name);
+                }
+
+                fprintf(ofp, ");\n");
+                fprintf(ofp, "}\n\n");
+        }
+
+        fprintf(ofp, "sub trace_unhandled\n{\n\tmy ($event_name, $context, "
+                "$common_cpu, $common_secs, $common_nsecs,\n\t    "
+                "$common_pid, $common_comm) = @_;\n\n");
+
+        fprintf(ofp, "\tprint_header($event_name, $common_cpu, "
+                "$common_secs, $common_nsecs,\n\t             $common_pid, "
+                "$common_comm);\n}\n\n");
+
+        fprintf(ofp, "sub print_header\n{\n"
+                "\tmy ($event_name, $cpu, $secs, $nsecs, $pid, $comm) = @_;\n\n"
+                "\tprintf(\"%%-20s %%5u %%05u.%%09u %%8u %%-20s \",\n\t       "
+                "$event_name, $cpu, $secs, $nsecs, $pid, $comm);\n}");
+
+        fclose(ofp);
+
+        fprintf(stderr, "generated Perl script: %s\n", fname);
+
+        return 0;
+}
+
+struct scripting_ops perl_scripting_ops = {
+        .name = "Perl",
+        .start_script = perl_start_script,
+        .stop_script = perl_stop_script,
+        .process_event = perl_process_event,
+        .generate_script = perl_generate_script,
+};
diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c
new file mode 100644
index 00000000000..c6d99334bdf
--- /dev/null
+++ b/tools/perf/util/scripting-engines/trace-event-python.c
@@ -0,0 +1,594 @@
+/*
+ * trace-event-python.  Feed trace events to an embedded Python interpreter.
+ *
+ * Copyright (C) 2010 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <Python.h>
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+#include <errno.h>
+
+#include "../../perf.h"
+#include "../util.h"
+#include "../trace-event.h"
+
+PyMODINIT_FUNC initperf_trace_context(void);
+
+#define FTRACE_MAX_EVENT                                \
+        ((1 << (sizeof(unsigned short) * 8)) - 1)
+
+struct event *events[FTRACE_MAX_EVENT];
+
+#define MAX_FIELDS      64
+#define N_COMMON_FIELDS 7
+
+extern struct scripting_context *scripting_context;
+
+static char *cur_field_name;
+static int zero_flag_atom;
+
+static PyObject *main_module, *main_dict;
+
+static void handler_call_die(const char *handler_name)
+{
+        PyErr_Print();
+        Py_FatalError("problem in Python trace event handler");
+}
+
+static void define_value(enum print_arg_type field_type,
+                         const char *ev_name,
+                         const char *field_name,
+                         const char *field_value,
+                         const char *field_str)
+{
+        const char *handler_name = "define_flag_value";
+        PyObject *handler, *t, *retval;
+        unsigned long long value;
+        unsigned n = 0;
+
+        if (field_type == PRINT_SYMBOL)
+                handler_name = "define_symbolic_value";
+
+        t = PyTuple_New(4);
+        if (!t)
+                Py_FatalError("couldn't create Python tuple");
+
+        value = eval_flag(field_value);
+
+        PyTuple_SetItem(t, n++, PyString_FromString(ev_name));
+        PyTuple_SetItem(t, n++, PyString_FromString(field_name));
+        PyTuple_SetItem(t, n++, PyInt_FromLong(value));
+        PyTuple_SetItem(t, n++, PyString_FromString(field_str));
+
+        handler = PyDict_GetItemString(main_dict, handler_name);
+        if (handler && PyCallable_Check(handler)) {
+                retval = PyObject_CallObject(handler, t);
+                if (retval == NULL)
+                        handler_call_die(handler_name);
+        }
+
+        Py_DECREF(t);
+}
+
+static void define_values(enum print_arg_type field_type,
+                          struct print_flag_sym *field,
+                          const char *ev_name,
+                          const char *field_name)
+{
+        define_value(field_type, ev_name, field_name, field->value,
+                     field->str);
+
+        if (field->next)
+                define_values(field_type, field->next, ev_name, field_name);
+}
+
+static void define_field(enum print_arg_type field_type,
+                         const char *ev_name,
+                         const char *field_name,
+                         const char *delim)
+{
+        const char *handler_name = "define_flag_field";
+        PyObject *handler, *t, *retval;
+        unsigned n = 0;
+
+        if (field_type == PRINT_SYMBOL)
+                handler_name = "define_symbolic_field";
+
+        if (field_type == PRINT_FLAGS)
+                t = PyTuple_New(3);
+        else
+                t = PyTuple_New(2);
+        if (!t)
+                Py_FatalError("couldn't create Python tuple");
+
+        PyTuple_SetItem(t, n++, PyString_FromString(ev_name));
+        PyTuple_SetItem(t, n++, PyString_FromString(field_name));
+        if (field_type == PRINT_FLAGS)
+                PyTuple_SetItem(t, n++, PyString_FromString(delim));
+
+        handler = PyDict_GetItemString(main_dict, handler_name);
+        if (handler && PyCallable_Check(handler)) {
+                retval = PyObject_CallObject(handler, t);
+                if (retval == NULL)
+                        handler_call_die(handler_name);
+        }
+
+        Py_DECREF(t);
+}
+
+static void define_event_symbols(struct event *event,
+                                 const char *ev_name,
+                                 struct print_arg *args)
+{
+        switch (args->type) {
+        case PRINT_NULL:
+                break;
+        case PRINT_ATOM:
+                define_value(PRINT_FLAGS, ev_name, cur_field_name, "0",
+                             args->atom.atom);
+                zero_flag_atom = 0;
+                break;
+        case PRINT_FIELD:
+                if (cur_field_name)
+                        free(cur_field_name);
+                cur_field_name = strdup(args->field.name);
+                break;
+        case PRINT_FLAGS:
+                define_event_symbols(event, ev_name, args->flags.field);
+                define_field(PRINT_FLAGS, ev_name, cur_field_name,
+                             args->flags.delim);
+                define_values(PRINT_FLAGS, args->flags.flags, ev_name,
+                              cur_field_name);
+                break;
+        case PRINT_SYMBOL:
+                define_event_symbols(event, ev_name, args->symbol.field);
+                define_field(PRINT_SYMBOL, ev_name, cur_field_name, NULL);
+                define_values(PRINT_SYMBOL, args->symbol.symbols, ev_name,
+                              cur_field_name);
+                break;
+        case PRINT_STRING:
+                break;
+        case PRINT_TYPE:
+                define_event_symbols(event, ev_name, args->typecast.item);
+                break;
+        case PRINT_OP:
+                if (strcmp(args->op.op, ":") == 0)
+                        zero_flag_atom = 1;
+                define_event_symbols(event, ev_name, args->op.left);
+                define_event_symbols(event, ev_name, args->op.right);
+                break;
+        default:
+                /* we should warn... */
+                return;
+        }
+
+        if (args->next)
+                define_event_symbols(event, ev_name, args->next);
+}
+
+static inline struct event *find_cache_event(int type)
+{
+        static char ev_name[256];
+        struct event *event;
+
+        if (events[type])
+                return events[type];
+
+        events[type] = event = trace_find_event(type);
+        if (!event)
+                return NULL;
+
+        sprintf(ev_name, "%s__%s", event->system, event->name);
+
+        define_event_symbols(event, ev_name, event->print_fmt.args);
+
+        return event;
+}
+
+static void python_process_event(int cpu, void *data,
+                                 int size __unused,
+                                 unsigned long long nsecs, char *comm)
+{
+        PyObject *handler, *retval, *context, *t, *obj, *dict = NULL;
+        static char handler_name[256];
+        struct format_field *field;
+        unsigned long long val;
+        unsigned long s, ns;
+        struct event *event;
+        unsigned n = 0;
+        int type;
+        int pid;
+
+        t = PyTuple_New(MAX_FIELDS);
+        if (!t)
+                Py_FatalError("couldn't create Python tuple");
+
+        type = trace_parse_common_type(data);
+
+        event = find_cache_event(type);
+        if (!event)
+                die("ug! no event found for type %d", type);
+
+        pid = trace_parse_common_pid(data);
+
+        sprintf(handler_name, "%s__%s", event->system, event->name);
+
+        handler = PyDict_GetItemString(main_dict, handler_name);
+        if (handler && !PyCallable_Check(handler))
+                handler = NULL;
+        if (!handler) {
+                dict = PyDict_New();
+                if (!dict)
+                        Py_FatalError("couldn't create Python dict");
+        }
+        s = nsecs / NSECS_PER_SEC;
+        ns = nsecs - s * NSECS_PER_SEC;
+
+        scripting_context->event_data = data;
+
+        context = PyCObject_FromVoidPtr(scripting_context, NULL);
+
+        PyTuple_SetItem(t, n++, PyString_FromString(handler_name));
+        PyTuple_SetItem(t, n++,
+                        PyCObject_FromVoidPtr(scripting_context, NULL));
+
+        if (handler) {
+                PyTuple_SetItem(t, n++, PyInt_FromLong(cpu));
+                PyTuple_SetItem(t, n++, PyInt_FromLong(s));
+                PyTuple_SetItem(t, n++, PyInt_FromLong(ns));
+                PyTuple_SetItem(t, n++, PyInt_FromLong(pid));
+                PyTuple_SetItem(t, n++, PyString_FromString(comm));
+        } else {
+                PyDict_SetItemString(dict, "common_cpu", PyInt_FromLong(cpu));
+                PyDict_SetItemString(dict, "common_s", PyInt_FromLong(s));
+                PyDict_SetItemString(dict, "common_ns", PyInt_FromLong(ns));
+                PyDict_SetItemString(dict, "common_pid", PyInt_FromLong(pid));
+                PyDict_SetItemString(dict, "common_comm", PyString_FromString(comm));
+        }
+        for (field = event->format.fields; field; field = field->next) {
+                if (field->flags & FIELD_IS_STRING) {
+                        int offset;
+                        if (field->flags & FIELD_IS_DYNAMIC) {
+                                offset = *(int *)(data + field->offset);
+                                offset &= 0xffff;
+                        } else
+                                offset = field->offset;
+                        obj = PyString_FromString((char *)data + offset);
+                } else { /* FIELD_IS_NUMERIC */
+                        val = read_size(data + field->offset, field->size);
+                        if (field->flags & FIELD_IS_SIGNED) {
+                                if ((long long)val >= LONG_MIN &&
+                                    (long long)val <= LONG_MAX)
+                                        obj = PyInt_FromLong(val);
+                                else
+                                        obj = PyLong_FromLongLong(val);
+                        } else {
+                                if (val <= LONG_MAX)
+                                        obj = PyInt_FromLong(val);
+                                else
+                                        obj = PyLong_FromUnsignedLongLong(val);
+                        }
+                }
+                if (handler)
+                        PyTuple_SetItem(t, n++, obj);
+                else
+                        PyDict_SetItemString(dict, field->name, obj);
+
+        }
+        if (!handler)
+                PyTuple_SetItem(t, n++, dict);
+
+        if (_PyTuple_Resize(&t, n) == -1)
+                Py_FatalError("error resizing Python tuple");
+
+        if (handler) {
+                retval = PyObject_CallObject(handler, t);
+                if (retval == NULL)
+                        handler_call_die(handler_name);
+        } else {
+                handler = PyDict_GetItemString(main_dict, "trace_unhandled");
+                if (handler && PyCallable_Check(handler)) {
+
+                        retval = PyObject_CallObject(handler, t);
+                        if (retval == NULL)
+                                handler_call_die("trace_unhandled");
+                }
+                Py_DECREF(dict);
+        }
+
+        Py_DECREF(t);
+}
+
+static int run_start_sub(void)
+{
+        PyObject *handler, *retval;
+        int err = 0;
+
+        main_module = PyImport_AddModule("__main__");
+        if (main_module == NULL)
+                return -1;
+        Py_INCREF(main_module);
+
+        main_dict = PyModule_GetDict(main_module);
+        if (main_dict == NULL) {
+                err = -1;
+                goto error;
+        }
+        Py_INCREF(main_dict);
+
+        handler = PyDict_GetItemString(main_dict, "trace_begin");
+        if (handler == NULL || !PyCallable_Check(handler))
+                goto out;
+
+        retval = PyObject_CallObject(handler, NULL);
+        if (retval == NULL)
+                handler_call_die("trace_begin");
+
+        Py_DECREF(retval);
+        return err;
+error:
+        Py_XDECREF(main_dict);
+        Py_XDECREF(main_module);
+out:
+        return err;
+}
+
+/*
+ * Start trace script
+ */
+static int python_start_script(const char *script, int argc, const char **argv)
+{
+        const char **command_line;
+        char buf[PATH_MAX];
+        int i, err = 0;
+        FILE *fp;
+
+        command_line = malloc((argc + 1) * sizeof(const char *));
+        command_line[0] = script;
+        for (i = 1; i < argc + 1; i++)
+                command_line[i] = argv[i - 1];
+
+        Py_Initialize();
+
+        initperf_trace_context();
+
+        PySys_SetArgv(argc + 1, (char **)command_line);
+
+        fp = fopen(script, "r");
+        if (!fp) {
+                sprintf(buf, "Can't open python script \"%s\"", script);
+                perror(buf);
+                err = -1;
+                goto error;
+        }
+
+        err = PyRun_SimpleFile(fp, script);
+        if (err) {
+                fprintf(stderr, "Error running python script %s\n", script);
+                goto error;
+        }
+
+        err = run_start_sub();
+        if (err) {
+                fprintf(stderr, "Error starting python script %s\n", script);
+                goto error;
+        }
+
+        free(command_line);
+
+        return err;
+error:
+        Py_Finalize();
+        free(command_line);
+
+        return err;
+}
+
+/*
+ * Stop trace script
+ */
+static int python_stop_script(void)
+{
+        PyObject *handler, *retval;
+        int err = 0;
+
+        handler = PyDict_GetItemString(main_dict, "trace_end");
+        if (handler == NULL || !PyCallable_Check(handler))
+                goto out;
+
+        retval = PyObject_CallObject(handler, NULL);
+        if (retval == NULL)
+                handler_call_die("trace_end");
+        else
+                Py_DECREF(retval);
+out:
+        Py_XDECREF(main_dict);
+        Py_XDECREF(main_module);
+        Py_Finalize();
+
+        return err;
+}
+
+static int python_generate_script(const char *outfile)
+{
+        struct event *event = NULL;
+        struct format_field *f;
+        char fname[PATH_MAX];
+        int not_first, count;
+        FILE *ofp;
+
+        sprintf(fname, "%s.py", outfile);
+        ofp = fopen(fname, "w");
+        if (ofp == NULL) {
+                fprintf(stderr, "couldn't open %s\n", fname);
+                return -1;
+        }
+        fprintf(ofp, "# perf script event handlers, "
+                "generated by perf script -g python\n");
+
+        fprintf(ofp, "# Licensed under the terms of the GNU GPL"
+                " License version 2\n\n");
+
+        fprintf(ofp, "# The common_* event handler fields are the most useful "
+                "fields common to\n");
+
+        fprintf(ofp, "# all events.  They don't necessarily correspond to "
+                "the 'common_*' fields\n");
+
+        fprintf(ofp, "# in the format files.  Those fields not available as "
+                "handler params can\n");
+
+        fprintf(ofp, "# be retrieved using Python functions of the form "
+                "common_*(context).\n");
+
+        fprintf(ofp, "# See the perf-trace-python Documentation for the list "
+                "of available functions.\n\n");
+
+        fprintf(ofp, "import os\n");
+        fprintf(ofp, "import sys\n\n");
+
+        fprintf(ofp, "sys.path.append(os.environ['PERF_EXEC_PATH'] + \\\n");
+        fprintf(ofp, "\t'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')\n");
+        fprintf(ofp, "\nfrom perf_trace_context import *\n");
+        fprintf(ofp, "from Core import *\n\n\n");
+
+        fprintf(ofp, "def trace_begin():\n");
+        fprintf(ofp, "\tprint \"in trace_begin\"\n\n");
+
+        fprintf(ofp, "def trace_end():\n");
+        fprintf(ofp, "\tprint \"in trace_end\"\n\n");
+
+        while ((event = trace_find_next_event(event))) {
+                fprintf(ofp, "def %s__%s(", event->system, event->name);
+                fprintf(ofp, "event_name, ");
+                fprintf(ofp, "context, ");
+                fprintf(ofp, "common_cpu,\n");
+                fprintf(ofp, "\tcommon_secs, ");
+                fprintf(ofp, "common_nsecs, ");
+                fprintf(ofp, "common_pid, ");
+                fprintf(ofp, "common_comm,\n\t");
+
+                not_first = 0;
+                count = 0;
+
+                for (f = event->format.fields; f; f = f->next) {
+                        if (not_first++)
+                                fprintf(ofp, ", ");
+                        if (++count % 5 == 0)
+                                fprintf(ofp, "\n\t");
+
+                        fprintf(ofp, "%s", f->name);
+                }
+                fprintf(ofp, "):\n");
+
+                fprintf(ofp, "\t\tprint_header(event_name, common_cpu, "
+                        "common_secs, common_nsecs,\n\t\t\t"
+                        "common_pid, common_comm)\n\n");
+
+                fprintf(ofp, "\t\tprint \"");
+
+                not_first = 0;
+                count = 0;
+
+                for (f = event->format.fields; f; f = f->next) {
+                        if (not_first++)
+                                fprintf(ofp, ", ");
+                        if (count && count % 3 == 0) {
+                                fprintf(ofp, "\" \\\n\t\t\"");
+                        }
+                        count++;
+
+                        fprintf(ofp, "%s=", f->name);
+                        if (f->flags & FIELD_IS_STRING ||
+                            f->flags & FIELD_IS_FLAG ||
+                            f->flags & FIELD_IS_SYMBOLIC)
+                                fprintf(ofp, "%%s");
+                        else if (f->flags & FIELD_IS_SIGNED)
+                                fprintf(ofp, "%%d");
+                        else
+                                fprintf(ofp, "%%u");
+                }
+
+                fprintf(ofp, "\\n\" %% \\\n\t\t(");
+
+                not_first = 0;
+                count = 0;
+
+                for (f = event->format.fields; f; f = f->next) {
+                        if (not_first++)
+                                fprintf(ofp, ", ");
+
+                        if (++count % 5 == 0)
+                                fprintf(ofp, "\n\t\t");
+
+                        if (f->flags & FIELD_IS_FLAG) {
+                                if ((count - 1) % 5 != 0) {
+                                        fprintf(ofp, "\n\t\t");
+                                        count = 4;
+                                }
+                                fprintf(ofp, "flag_str(\"");
+                                fprintf(ofp, "%s__%s\", ", event->system,
+                                        event->name);
+                                fprintf(ofp, "\"%s\", %s)", f->name,
+                                        f->name);
+                        } else if (f->flags & FIELD_IS_SYMBOLIC) {
+                                if ((count - 1) % 5 != 0) {
+                                        fprintf(ofp, "\n\t\t");
+                                        count = 4;
+                                }
+                                fprintf(ofp, "symbol_str(\"");
+                                fprintf(ofp, "%s__%s\", ", event->system,
+                                        event->name);
+                                fprintf(ofp, "\"%s\", %s)", f->name,
+                                        f->name);
+                        } else
+                                fprintf(ofp, "%s", f->name);
+                }
+
+                fprintf(ofp, "),\n\n");
+        }
+
+        fprintf(ofp, "def trace_unhandled(event_name, context, "
+                "event_fields_dict):\n");
+
+        fprintf(ofp, "\t\tprint ' '.join(['%%s=%%s'%%(k,str(v))"
+                "for k,v in sorted(event_fields_dict.items())])\n\n");
+
+        fprintf(ofp, "def print_header("
+                "event_name, cpu, secs, nsecs, pid, comm):\n"
+                "\tprint \"%%-20s %%5u %%05u.%%09u %%8u %%-20s \" %% \\\n\t"
+                "(event_name, cpu, secs, nsecs, pid, comm),\n");
+
+        fclose(ofp);
+
+        fprintf(stderr, "generated Python script: %s\n", fname);
+
+        return 0;
+}
+
+struct scripting_ops python_scripting_ops = {
+        .name = "Python",
+        .start_script = python_start_script,
+        .stop_script = python_stop_script,
+        .process_event = python_process_event,
+        .generate_script = python_generate_script,
+};
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
new file mode 100644
index 00000000000..313dac2d94c
--- /dev/null
+++ b/tools/perf/util/session.c
@@ -0,0 +1,1134 @@
+#define _FILE_OFFSET_BITS 64
+
+#include <linux/kernel.h>
+
+#include <byteswap.h>
+#include <unistd.h>
+#include <sys/types.h>
+#include <sys/mman.h>
+
+#include "session.h"
+#include "sort.h"
+#include "util.h"
+
+static int perf_session__open(struct perf_session *self, bool force)
+{
+        struct stat input_stat;
+
+        if (!strcmp(self->filename, "-")) {
+                self->fd_pipe = true;
+                self->fd = STDIN_FILENO;
+
+                if (perf_header__read(self, self->fd) < 0)
+                        pr_err("incompatible file format");
+
+                return 0;
+        }
+
+        self->fd = open(self->filename, O_RDONLY);
+        if (self->fd < 0) {
+                int err = errno;
+
+                pr_err("failed to open %s: %s", self->filename, strerror(err));
+                if (err == ENOENT && !strcmp(self->filename, "perf.data"))
+                        pr_err("  (try 'perf record' first)");
+                pr_err("\n");
+                return -errno;
+        }
+
+        if (fstat(self->fd, &input_stat) < 0)
+                goto out_close;
+
+        if (!force && input_stat.st_uid && (input_stat.st_uid != geteuid())) {
+                pr_err("file %s not owned by current user or root\n",
+                       self->filename);
+                goto out_close;
+        }
+
+        if (!input_stat.st_size) {
+                pr_info("zero-sized file (%s), nothing to do!\n",
+                        self->filename);
+                goto out_close;
+        }
+
+        if (perf_header__read(self, 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;
+}
+
+static void perf_session__id_header_size(struct perf_session *session)
+{
+       struct sample_data *data;
+       u64 sample_type = session->sample_type;
+       u16 size = 0;
+
+        if (!session->sample_id_all)
+                goto out;
+
+       if (sample_type & PERF_SAMPLE_TID)
+               size += sizeof(data->tid) * 2;
+
+       if (sample_type & PERF_SAMPLE_TIME)
+               size += sizeof(data->time);
+
+       if (sample_type & PERF_SAMPLE_ID)
+               size += sizeof(data->id);
+
+       if (sample_type & PERF_SAMPLE_STREAM_ID)
+               size += sizeof(data->stream_id);
+
+       if (sample_type & PERF_SAMPLE_CPU)
+               size += sizeof(data->cpu) * 2;
+out:
+       session->id_hdr_size = size;
+}
+
+void perf_session__set_sample_id_all(struct perf_session *session, bool value)
+{
+        session->sample_id_all = value;
+        perf_session__id_header_size(session);
+}
+
+void perf_session__set_sample_type(struct perf_session *session, u64 type)
+{
+        session->sample_type = type;
+}
+
+void perf_session__update_sample_type(struct perf_session *self)
+{
+        self->sample_type = perf_header__sample_type(&self->header);
+        self->sample_id_all = perf_header__sample_id_all(&self->header);
+        perf_session__id_header_size(self);
+}
+
+int perf_session__create_kernel_maps(struct perf_session *self)
+{
+        int ret = machine__create_kernel_maps(&self->host_machine);
+
+        if (ret >= 0)
+                ret = machines__create_guest_kernel_maps(&self->machines);
+        return ret;
+}
+
+static void perf_session__destroy_kernel_maps(struct perf_session *self)
+{
+        machine__destroy_kernel_maps(&self->host_machine);
+        machines__destroy_guest_kernel_maps(&self->machines);
+}
+
+struct perf_session *perf_session__new(const char *filename, int mode,
+                                       bool force, bool repipe,
+                                       struct perf_event_ops *ops)
+{
+        size_t len = filename ? strlen(filename) + 1 : 0;
+        struct perf_session *self = zalloc(sizeof(*self) + len);
+
+        if (self == NULL)
+                goto out;
+
+        if (perf_header__init(&self->header) < 0)
+                goto out_free;
+
+        memcpy(self->filename, filename, len);
+        self->threads = RB_ROOT;
+        INIT_LIST_HEAD(&self->dead_threads);
+        self->hists_tree = RB_ROOT;
+        self->last_match = NULL;
+        /*
+         * On 64bit we can mmap the data file in one go. No need for tiny mmap
+         * slices. On 32bit we use 32MB.
+         */
+#if BITS_PER_LONG == 64
+        self->mmap_window = ULLONG_MAX;
+#else
+        self->mmap_window = 32 * 1024 * 1024ULL;
+#endif
+        self->machines = RB_ROOT;
+        self->repipe = repipe;
+        INIT_LIST_HEAD(&self->ordered_samples.samples);
+        INIT_LIST_HEAD(&self->ordered_samples.sample_cache);
+        INIT_LIST_HEAD(&self->ordered_samples.to_free);
+        machine__init(&self->host_machine, "", HOST_KERNEL_ID);
+
+        if (mode == O_RDONLY) {
+                if (perf_session__open(self, force) < 0)
+                        goto out_delete;
+        } else if (mode == O_WRONLY) {
+                /*
+                 * In O_RDONLY mode this will be performed when reading the
+                 * kernel MMAP event, in event__process_mmap().
+                 */
+                if (perf_session__create_kernel_maps(self) < 0)
+                        goto out_delete;
+        }
+
+        perf_session__update_sample_type(self);
+
+        if (ops && ops->ordering_requires_timestamps &&
+            ops->ordered_samples && !self->sample_id_all) {
+                dump_printf("WARNING: No sample_id_all support, falling back to unordered processing\n");
+                ops->ordered_samples = false;
+        }
+
+out:
+        return self;
+out_free:
+        free(self);
+        return NULL;
+out_delete:
+        perf_session__delete(self);
+        return NULL;
+}
+
+static void perf_session__delete_dead_threads(struct perf_session *self)
+{
+        struct thread *n, *t;
+
+        list_for_each_entry_safe(t, n, &self->dead_threads, node) {
+                list_del(&t->node);
+                thread__delete(t);
+        }
+}
+
+static void perf_session__delete_threads(struct perf_session *self)
+{
+        struct rb_node *nd = rb_first(&self->threads);
+
+        while (nd) {
+                struct thread *t = rb_entry(nd, struct thread, rb_node);
+
+                rb_erase(&t->rb_node, &self->threads);
+                nd = rb_next(nd);
+                thread__delete(t);
+        }
+}
+
+void perf_session__delete(struct perf_session *self)
+{
+        perf_header__exit(&self->header);
+        perf_session__destroy_kernel_maps(self);
+        perf_session__delete_dead_threads(self);
+        perf_session__delete_threads(self);
+        machine__exit(&self->host_machine);
+        close(self->fd);
+        free(self);
+}
+
+void perf_session__remove_thread(struct perf_session *self, struct thread *th)
+{
+        self->last_match = NULL;
+        rb_erase(&th->rb_node, &self->threads);
+        /*
+         * We may have references to this thread, for instance in some hist_entry
+         * instances, so just move them to a separate list.
+         */
+        list_add_tail(&th->node, &self->dead_threads);
+}
+
+static bool symbol__match_parent_regex(struct symbol *sym)
+{
+        if (sym->name && !regexec(&parent_regex, sym->name, 0, NULL, 0))
+                return 1;
+
+        return 0;
+}
+
+struct map_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;
+        unsigned int i;
+        struct map_symbol *syms = calloc(chain->nr, sizeof(*syms));
+
+        if (!syms)
+                return NULL;
+
+        for (i = 0; i < chain->nr; i++) {
+                u64 ip = chain->ips[i];
+                struct addr_location al;
+
+                if (ip >= PERF_CONTEXT_MAX) {
+                        switch (ip) {
+                        case PERF_CONTEXT_HV:
+                                cpumode = PERF_RECORD_MISC_HYPERVISOR;  break;
+                        case PERF_CONTEXT_KERNEL:
+                                cpumode = PERF_RECORD_MISC_KERNEL;      break;
+                        case PERF_CONTEXT_USER:
+                                cpumode = PERF_RECORD_MISC_USER;        break;
+                        default:
+                                break;
+                        }
+                        continue;
+                }
+
+                al.filtered = false;
+                thread__find_addr_location(thread, self, cpumode,
+                                MAP__FUNCTION, thread->pid, ip, &al, NULL);
+                if (al.sym != NULL) {
+                        if (sort__has_parent && !*parent &&
+                            symbol__match_parent_regex(al.sym))
+                                *parent = al.sym;
+                        if (!symbol_conf.use_callchain)
+                                break;
+                        syms[i].map = al.map;
+                        syms[i].sym = al.sym;
+                }
+        }
+
+        return syms;
+}
+
+static int process_event_synth_stub(event_t *event __used,
+                                    struct perf_session *session __used)
+{
+        dump_printf(": unhandled!\n");
+        return 0;
+}
+
+static int process_event_stub(event_t *event __used,
+                              struct sample_data *sample __used,
+                              struct perf_session *session __used)
+{
+        dump_printf(": unhandled!\n");
+        return 0;
+}
+
+static int process_finished_round_stub(event_t *event __used,
+                                       struct perf_session *session __used,
+                                       struct perf_event_ops *ops __used)
+{
+        dump_printf(": unhandled!\n");
+        return 0;
+}
+
+static int process_finished_round(event_t *event,
+                                  struct perf_session *session,
+                                  struct perf_event_ops *ops);
+
+static void perf_event_ops__fill_defaults(struct perf_event_ops *handler)
+{
+        if (handler->sample == NULL)
+                handler->sample = process_event_stub;
+        if (handler->mmap == NULL)
+                handler->mmap = process_event_stub;
+        if (handler->comm == NULL)
+                handler->comm = process_event_stub;
+        if (handler->fork == NULL)
+                handler->fork = process_event_stub;
+        if (handler->exit == NULL)
+                handler->exit = process_event_stub;
+        if (handler->lost == NULL)
+                handler->lost = event__process_lost;
+        if (handler->read == NULL)
+                handler->read = process_event_stub;
+        if (handler->throttle == NULL)
+                handler->throttle = process_event_stub;
+        if (handler->unthrottle == NULL)
+                handler->unthrottle = process_event_stub;
+        if (handler->attr == NULL)
+                handler->attr = process_event_synth_stub;
+        if (handler->event_type == NULL)
+                handler->event_type = process_event_synth_stub;
+        if (handler->tracing_data == NULL)
+                handler->tracing_data = process_event_synth_stub;
+        if (handler->build_id == NULL)
+                handler->build_id = process_event_synth_stub;
+        if (handler->finished_round == NULL) {
+                if (handler->ordered_samples)
+                        handler->finished_round = process_finished_round;
+                else
+                        handler->finished_round = process_finished_round_stub;
+        }
+}
+
+void mem_bswap_64(void *src, int byte_size)
+{
+        u64 *m = src;
+
+        while (byte_size > 0) {
+                *m = bswap_64(*m);
+                byte_size -= sizeof(u64);
+                ++m;
+        }
+}
+
+static void event__all64_swap(event_t *self)
+{
+        struct perf_event_header *hdr = &self->header;
+        mem_bswap_64(hdr + 1, self->header.size - sizeof(*hdr));
+}
+
+static void event__comm_swap(event_t *self)
+{
+        self->comm.pid = bswap_32(self->comm.pid);
+        self->comm.tid = bswap_32(self->comm.tid);
+}
+
+static void event__mmap_swap(event_t *self)
+{
+        self->mmap.pid   = bswap_32(self->mmap.pid);
+        self->mmap.tid   = bswap_32(self->mmap.tid);
+        self->mmap.start = bswap_64(self->mmap.start);
+        self->mmap.len   = bswap_64(self->mmap.len);
+        self->mmap.pgoff = bswap_64(self->mmap.pgoff);
+}
+
+static void event__task_swap(event_t *self)
+{
+        self->fork.pid  = bswap_32(self->fork.pid);
+        self->fork.tid  = bswap_32(self->fork.tid);
+        self->fork.ppid = bswap_32(self->fork.ppid);
+        self->fork.ptid = bswap_32(self->fork.ptid);
+        self->fork.time = bswap_64(self->fork.time);
+}
+
+static void event__read_swap(event_t *self)
+{
+        self->read.pid          = bswap_32(self->read.pid);
+        self->read.tid          = bswap_32(self->read.tid);
+        self->read.value        = bswap_64(self->read.value);
+        self->read.time_enabled = bswap_64(self->read.time_enabled);
+        self->read.time_running = bswap_64(self->read.time_running);
+        self->read.id           = bswap_64(self->read.id);
+}
+
+static void event__attr_swap(event_t *self)
+{
+        size_t size;
+
+        self->attr.attr.type            = bswap_32(self->attr.attr.type);
+        self->attr.attr.size            = bswap_32(self->attr.attr.size);
+        self->attr.attr.config          = bswap_64(self->attr.attr.config);
+        self->attr.attr.sample_period   = bswap_64(self->attr.attr.sample_period);
+        self->attr.attr.sample_type     = bswap_64(self->attr.attr.sample_type);
+        self->attr.attr.read_format     = bswap_64(self->attr.attr.read_format);
+        self->attr.attr.wakeup_events   = bswap_32(self->attr.attr.wakeup_events);
+        self->attr.attr.bp_type         = bswap_32(self->attr.attr.bp_type);
+        self->attr.attr.bp_addr         = bswap_64(self->attr.attr.bp_addr);
+        self->attr.attr.bp_len          = bswap_64(self->attr.attr.bp_len);
+
+        size = self->header.size;
+        size -= (void *)&self->attr.id - (void *)self;
+        mem_bswap_64(self->attr.id, size);
+}
+
+static void event__event_type_swap(event_t *self)
+{
+        self->event_type.event_type.event_id =
+                bswap_64(self->event_type.event_type.event_id);
+}
+
+static void event__tracing_data_swap(event_t *self)
+{
+        self->tracing_data.size = bswap_32(self->tracing_data.size);
+}
+
+typedef void (*event__swap_op)(event_t *self);
+
+static event__swap_op event__swap_ops[] = {
+        [PERF_RECORD_MMAP]   = event__mmap_swap,
+        [PERF_RECORD_COMM]   = event__comm_swap,
+        [PERF_RECORD_FORK]   = event__task_swap,
+        [PERF_RECORD_EXIT]   = event__task_swap,
+        [PERF_RECORD_LOST]   = event__all64_swap,
+        [PERF_RECORD_READ]   = event__read_swap,
+        [PERF_RECORD_SAMPLE] = event__all64_swap,
+        [PERF_RECORD_HEADER_ATTR]   = event__attr_swap,
+        [PERF_RECORD_HEADER_EVENT_TYPE]   = event__event_type_swap,
+        [PERF_RECORD_HEADER_TRACING_DATA]   = event__tracing_data_swap,
+        [PERF_RECORD_HEADER_BUILD_ID]   = NULL,
+        [PERF_RECORD_HEADER_MAX]    = NULL,
+};
+
+struct sample_queue {
+        u64                     timestamp;
+        u64                     file_offset;
+        event_t                 *event;
+        struct list_head        list;
+};
+
+static void perf_session_free_sample_buffers(struct perf_session *session)
+{
+        struct ordered_samples *os = &session->ordered_samples;
+
+        while (!list_empty(&os->to_free)) {
+                struct sample_queue *sq;
+
+                sq = list_entry(os->to_free.next, struct sample_queue, list);
+                list_del(&sq->list);
+                free(sq);
+        }
+}
+
+static int perf_session_deliver_event(struct perf_session *session,
+                                      event_t *event,
+                                      struct sample_data *sample,
+                                      struct perf_event_ops *ops,
+                                      u64 file_offset);
+
+static void flush_sample_queue(struct perf_session *s,
+                               struct perf_event_ops *ops)
+{
+        struct ordered_samples *os = &s->ordered_samples;
+        struct list_head *head = &os->samples;
+        struct sample_queue *tmp, *iter;
+        struct sample_data sample;
+        u64 limit = os->next_flush;
+        u64 last_ts = os->last_sample ? os->last_sample->timestamp : 0ULL;
+
+        if (!ops->ordered_samples || !limit)
+                return;
+
+        list_for_each_entry_safe(iter, tmp, head, list) {
+                if (iter->timestamp > limit)
+                        break;
+
+                event__parse_sample(iter->event, s, &sample);
+                perf_session_deliver_event(s, iter->event, &sample, ops,
+                                           iter->file_offset);
+
+                os->last_flush = iter->timestamp;
+                list_del(&iter->list);
+                list_add(&iter->list, &os->sample_cache);
+        }
+
+        if (list_empty(head)) {
+                os->last_sample = NULL;
+        } else if (last_ts <= limit) {
+                os->last_sample =
+                        list_entry(head->prev, struct sample_queue, list);
+        }
+}
+
+/*
+ * When perf record finishes a pass on every buffers, it records this pseudo
+ * event.
+ * We record the max timestamp t found in the pass n.
+ * Assuming these timestamps are monotonic across cpus, we know that if
+ * a buffer still has events with timestamps below t, they will be all
+ * available and then read in the pass n + 1.
+ * Hence when we start to read the pass n + 2, we can safely flush every
+ * events with timestamps below t.
+ *
+ *    ============ PASS n =================
+ *       CPU 0         |   CPU 1
+ *                     |
+ *    cnt1 timestamps  |   cnt2 timestamps
+ *          1          |         2
+ *          2          |         3
+ *          -          |         4  <--- max recorded
+ *
+ *    ============ PASS n + 1 ==============
+ *       CPU 0         |   CPU 1
+ *                     |
+ *    cnt1 timestamps  |   cnt2 timestamps
+ *          3          |         5
+ *          4          |         6
+ *          5          |         7 <---- max recorded
+ *
+ *      Flush every events below timestamp 4
+ *
+ *    ============ PASS n + 2 ==============
+ *       CPU 0         |   CPU 1
+ *                     |
+ *    cnt1 timestamps  |   cnt2 timestamps
+ *          6          |         8
+ *          7          |         9
+ *          -          |         10
+ *
+ *      Flush every events below timestamp 7
+ *      etc...
+ */
+static int process_finished_round(event_t *event __used,
+                                  struct perf_session *session,
+                                  struct perf_event_ops *ops)
+{
+        flush_sample_queue(session, ops);
+        session->ordered_samples.next_flush = session->ordered_samples.max_timestamp;
+
+        return 0;
+}
+
+/* The queue is ordered by time */
+static void __queue_event(struct sample_queue *new, struct perf_session *s)
+{
+        struct ordered_samples *os = &s->ordered_samples;
+        struct sample_queue *sample = os->last_sample;
+        u64 timestamp = new->timestamp;
+        struct list_head *p;
+
+        os->last_sample = new;
+
+        if (!sample) {
+                list_add(&new->list, &os->samples);
+                os->max_timestamp = timestamp;
+                return;
+        }
+
+        /*
+         * last_sample might point to some random place in the list as it's
+         * the last queued event. We expect that the new event is close to
+         * this.
+         */
+        if (sample->timestamp <= timestamp) {
+                while (sample->timestamp <= timestamp) {
+                        p = sample->list.next;
+                        if (p == &os->samples) {
+                                list_add_tail(&new->list, &os->samples);
+                                os->max_timestamp = timestamp;
+                                return;
+                        }
+                        sample = list_entry(p, struct sample_queue, list);
+                }
+                list_add_tail(&new->list, &sample->list);
+        } else {
+                while (sample->timestamp > timestamp) {
+                        p = sample->list.prev;
+                        if (p == &os->samples) {
+                                list_add(&new->list, &os->samples);
+                                return;
+                        }
+                        sample = list_entry(p, struct sample_queue, list);
+                }
+                list_add(&new->list, &sample->list);
+        }
+}
+
+#define MAX_SAMPLE_BUFFER       (64 * 1024 / sizeof(struct sample_queue))
+
+static int perf_session_queue_event(struct perf_session *s, event_t *event,
+                                    struct sample_data *data, u64 file_offset)
+{
+        struct ordered_samples *os = &s->ordered_samples;
+        struct list_head *sc = &os->sample_cache;
+        u64 timestamp = data->time;
+        struct sample_queue *new;
+
+        if (!timestamp || timestamp == ~0ULL)
+                return -ETIME;
+
+        if (timestamp < s->ordered_samples.last_flush) {
+                printf("Warning: Timestamp below last timeslice flush\n");
+                return -EINVAL;
+        }
+
+        if (!list_empty(sc)) {
+                new = list_entry(sc->next, struct sample_queue, list);
+                list_del(&new->list);
+        } else if (os->sample_buffer) {
+                new = os->sample_buffer + os->sample_buffer_idx;
+                if (++os->sample_buffer_idx == MAX_SAMPLE_BUFFER)
+                        os->sample_buffer = NULL;
+        } else {
+                os->sample_buffer = malloc(MAX_SAMPLE_BUFFER * sizeof(*new));
+                if (!os->sample_buffer)
+                        return -ENOMEM;
+                list_add(&os->sample_buffer->list, &os->to_free);
+                os->sample_buffer_idx = 2;
+                new = os->sample_buffer + 1;
+        }
+
+        new->timestamp = timestamp;
+        new->file_offset = file_offset;
+        new->event = event;
+
+        __queue_event(new, s);
+
+        return 0;
+}
+
+static void callchain__printf(struct sample_data *sample)
+{
+        unsigned int i;
+
+        printf("... chain: nr:%Lu\n", sample->callchain->nr);
+
+        for (i = 0; i < sample->callchain->nr; i++)
+                printf("..... %2d: %016Lx\n", i, sample->callchain->ips[i]);
+}
+
+static void perf_session__print_tstamp(struct perf_session *session,
+                                       event_t *event,
+                                       struct sample_data *sample)
+{
+        if (event->header.type != PERF_RECORD_SAMPLE &&
+            !session->sample_id_all) {
+                fputs("-1 -1 ", stdout);
+                return;
+        }
+
+        if ((session->sample_type & PERF_SAMPLE_CPU))
+                printf("%u ", sample->cpu);
+
+        if (session->sample_type & PERF_SAMPLE_TIME)
+                printf("%Lu ", sample->time);
+}
+
+static void dump_event(struct perf_session *session, event_t *event,
+                       u64 file_offset, struct sample_data *sample)
+{
+        if (!dump_trace)
+                return;
+
+        printf("\n%#Lx [%#x]: event: %d\n", file_offset, event->header.size,
+               event->header.type);
+
+        trace_event(event);
+
+        if (sample)
+                perf_session__print_tstamp(session, event, sample);
+
+        printf("%#Lx [%#x]: PERF_RECORD_%s", file_offset, event->header.size,
+               event__get_event_name(event->header.type));
+}
+
+static void dump_sample(struct perf_session *session, event_t *event,
+                        struct sample_data *sample)
+{
+        if (!dump_trace)
+                return;
+
+        printf("(IP, %d): %d/%d: %#Lx period: %Ld\n", event->header.misc,
+               sample->pid, sample->tid, sample->ip, sample->period);
+
+        if (session->sample_type & PERF_SAMPLE_CALLCHAIN)
+                callchain__printf(sample);
+}
+
+static int perf_session_deliver_event(struct perf_session *session,
+                                      event_t *event,
+                                      struct sample_data *sample,
+                                      struct perf_event_ops *ops,
+                                      u64 file_offset)
+{
+        dump_event(session, event, file_offset, sample);
+
+        switch (event->header.type) {
+        case PERF_RECORD_SAMPLE:
+                dump_sample(session, event, sample);
+                return ops->sample(event, sample, session);
+        case PERF_RECORD_MMAP:
+                return ops->mmap(event, sample, session);
+        case PERF_RECORD_COMM:
+                return ops->comm(event, sample, session);
+        case PERF_RECORD_FORK:
+                return ops->fork(event, sample, session);
+        case PERF_RECORD_EXIT:
+                return ops->exit(event, sample, session);
+        case PERF_RECORD_LOST:
+                return ops->lost(event, sample, session);
+        case PERF_RECORD_READ:
+                return ops->read(event, sample, session);
+        case PERF_RECORD_THROTTLE:
+                return ops->throttle(event, sample, session);
+        case PERF_RECORD_UNTHROTTLE:
+                return ops->unthrottle(event, sample, session);
+        default:
+                ++session->hists.stats.nr_unknown_events;
+                return -1;
+        }
+}
+
+static int perf_session__preprocess_sample(struct perf_session *session,
+                                           event_t *event, struct sample_data *sample)
+{
+        if (event->header.type != PERF_RECORD_SAMPLE ||
+            !(session->sample_type & PERF_SAMPLE_CALLCHAIN))
+                return 0;
+
+        if (!ip_callchain__valid(sample->callchain, event)) {
+                pr_debug("call-chain problem with event, skipping it.\n");
+                ++session->hists.stats.nr_invalid_chains;
+                session->hists.stats.total_invalid_chains += sample->period;
+                return -EINVAL;
+        }
+        return 0;
+}
+
+static int perf_session__process_user_event(struct perf_session *session, event_t *event,
+                                            struct perf_event_ops *ops, u64 file_offset)
+{
+        dump_event(session, event, file_offset, NULL);
+
+        /* These events are processed right away */
+        switch (event->header.type) {
+        case PERF_RECORD_HEADER_ATTR:
+                return ops->attr(event, session);
+        case PERF_RECORD_HEADER_EVENT_TYPE:
+                return ops->event_type(event, session);
+        case PERF_RECORD_HEADER_TRACING_DATA:
+                /* setup for reading amidst mmap */
+                lseek(session->fd, file_offset, SEEK_SET);
+                return ops->tracing_data(event, session);
+        case PERF_RECORD_HEADER_BUILD_ID:
+                return ops->build_id(event, session);
+        case PERF_RECORD_FINISHED_ROUND:
+                return ops->finished_round(event, session, ops);
+        default:
+                return -EINVAL;
+        }
+}
+
+static int perf_session__process_event(struct perf_session *session,
+                                       event_t *event,
+                                       struct perf_event_ops *ops,
+                                       u64 file_offset)
+{
+        struct sample_data sample;
+        int ret;
+
+        if (session->header.needs_swap && event__swap_ops[event->header.type])
+                event__swap_ops[event->header.type](event);
+
+        if (event->header.type >= PERF_RECORD_HEADER_MAX)
+                return -EINVAL;
+
+        hists__inc_nr_events(&session->hists, event->header.type);
+
+        if (event->header.type >= PERF_RECORD_USER_TYPE_START)
+                return perf_session__process_user_event(session, event, ops, file_offset);
+
+        /*
+         * For all kernel events we get the sample data
+         */
+        event__parse_sample(event, session, &sample);
+
+        /* Preprocess sample records - precheck callchains */
+        if (perf_session__preprocess_sample(session, event, &sample))
+                return 0;
+
+        if (ops->ordered_samples) {
+                ret = perf_session_queue_event(session, event, &sample,
+                                               file_offset);
+                if (ret != -ETIME)
+                        return ret;
+        }
+
+        return perf_session_deliver_event(session, event, &sample, ops,
+                                          file_offset);
+}
+
+void perf_event_header__bswap(struct perf_event_header *self)
+{
+        self->type = bswap_32(self->type);
+        self->misc = bswap_16(self->misc);
+        self->size = bswap_16(self->size);
+}
+
+static struct thread *perf_session__register_idle_thread(struct perf_session *self)
+{
+        struct thread *thread = perf_session__findnew(self, 0);
+
+        if (thread == NULL || thread__set_comm(thread, "swapper")) {
+                pr_err("problem inserting idle task.\n");
+                thread = NULL;
+        }
+
+        return thread;
+}
+
+static void perf_session__warn_about_errors(const struct perf_session *session,
+                                            const struct perf_event_ops *ops)
+{
+        if (ops->lost == event__process_lost &&
+            session->hists.stats.total_lost != 0) {
+                ui__warning("Processed %Lu events and LOST %Lu!\n\n"
+                            "Check IO/CPU overload!\n\n",
+                            session->hists.stats.total_period,
+                            session->hists.stats.total_lost);
+        }
+
+        if (session->hists.stats.nr_unknown_events != 0) {
+                ui__warning("Found %u unknown events!\n\n"
+                            "Is this an older tool processing a perf.data "
+                            "file generated by a more recent tool?\n\n"
+                            "If that is not the case, consider "
+                            "reporting to linux-kernel@vger.kernel.org.\n\n",
+                            session->hists.stats.nr_unknown_events);
+        }
+
+        if (session->hists.stats.nr_invalid_chains != 0) {
+                ui__warning("Found invalid callchains!\n\n"
+                            "%u out of %u events were discarded for this reason.\n\n"
+                            "Consider reporting to linux-kernel@vger.kernel.org.\n\n",
+                            session->hists.stats.nr_invalid_chains,
+                            session->hists.stats.nr_events[PERF_RECORD_SAMPLE]);
+        }
+}
+
+#define session_done()  (*(volatile int *)(&session_done))
+volatile int session_done;
+
+static int __perf_session__process_pipe_events(struct perf_session *self,
+                                               struct perf_event_ops *ops)
+{
+        event_t event;
+        uint32_t size;
+        int skip = 0;
+        u64 head;
+        int err;
+        void *p;
+
+        perf_event_ops__fill_defaults(ops);
+
+        head = 0;
+more:
+        err = readn(self->fd, &event, sizeof(struct perf_event_header));
+        if (err <= 0) {
+                if (err == 0)
+                        goto done;
+
+                pr_err("failed to read event header\n");
+                goto out_err;
+        }
+
+        if (self->header.needs_swap)
+                perf_event_header__bswap(&event.header);
+
+        size = event.header.size;
+        if (size == 0)
+                size = 8;
+
+        p = &event;
+        p += sizeof(struct perf_event_header);
+
+        if (size - sizeof(struct perf_event_header)) {
+                err = readn(self->fd, p, size - sizeof(struct perf_event_header));
+                if (err <= 0) {
+                        if (err == 0) {
+                                pr_err("unexpected end of event stream\n");
+                                goto done;
+                        }
+
+                        pr_err("failed to read event data\n");
+                        goto out_err;
+                }
+        }
+
+        if (size == 0 ||
+            (skip = perf_session__process_event(self, &event, ops, head)) < 0) {
+                dump_printf("%#Lx [%#x]: skipping unknown header type: %d\n",
+                            head, event.header.size, event.header.type);
+                /*
+                 * assume we lost track of the stream, check alignment, and
+                 * increment a single u64 in the hope to catch on again 'soon'.
+                 */
+                if (unlikely(head & 7))
+                        head &= ~7ULL;
+
+                size = 8;
+        }
+
+        head += size;
+
+        if (skip > 0)
+                head += skip;
+
+        if (!session_done())
+                goto more;
+done:
+        err = 0;
+out_err:
+        perf_session__warn_about_errors(self, ops);
+        perf_session_free_sample_buffers(self);
+        return err;
+}
+
+int __perf_session__process_events(struct perf_session *session,
+                                   u64 data_offset, u64 data_size,
+                                   u64 file_size, struct perf_event_ops *ops)
+{
+        u64 head, page_offset, file_offset, file_pos, progress_next;
+        int err, mmap_prot, mmap_flags, map_idx = 0;
+        struct ui_progress *progress;
+        size_t  page_size, mmap_size;
+        char *buf, *mmaps[8];
+        event_t *event;
+        uint32_t size;
+
+        perf_event_ops__fill_defaults(ops);
+
+        page_size = sysconf(_SC_PAGESIZE);
+
+        page_offset = page_size * (data_offset / page_size);
+        file_offset = page_offset;
+        head = data_offset - page_offset;
+
+        if (data_offset + data_size < file_size)
+                file_size = data_offset + data_size;
+
+        progress_next = file_size / 16;
+        progress = ui_progress__new("Processing events...", file_size);
+        if (progress == NULL)
+                return -1;
+
+        mmap_size = session->mmap_window;
+        if (mmap_size > file_size)
+                mmap_size = file_size;
+
+        memset(mmaps, 0, sizeof(mmaps));
+
+        mmap_prot  = PROT_READ;
+        mmap_flags = MAP_SHARED;
+
+        if (session->header.needs_swap) {
+                mmap_prot  |= PROT_WRITE;
+                mmap_flags = MAP_PRIVATE;
+        }
+remap:
+        buf = mmap(NULL, mmap_size, mmap_prot, mmap_flags, session->fd,
+                   file_offset);
+        if (buf == MAP_FAILED) {
+                pr_err("failed to mmap file\n");
+                err = -errno;
+                goto out_err;
+        }
+        mmaps[map_idx] = buf;
+        map_idx = (map_idx + 1) & (ARRAY_SIZE(mmaps) - 1);
+        file_pos = file_offset + head;
+
+more:
+        event = (event_t *)(buf + head);
+
+        if (session->header.needs_swap)
+                perf_event_header__bswap(&event->header);
+        size = event->header.size;
+        if (size == 0)
+                size = 8;
+
+        if (head + event->header.size > mmap_size) {
+                if (mmaps[map_idx]) {
+                        munmap(mmaps[map_idx], mmap_size);
+                        mmaps[map_idx] = NULL;
+                }
+
+                page_offset = page_size * (head / page_size);
+                file_offset += page_offset;
+                head -= page_offset;
+                goto remap;
+        }
+
+        size = event->header.size;
+
+        if (size == 0 ||
+            perf_session__process_event(session, event, ops, file_pos) < 0) {
+                dump_printf("%#Lx [%#x]: skipping unknown header type: %d\n",
+                            file_offset + head, event->header.size,
+                            event->header.type);
+                /*
+                 * assume we lost track of the stream, check alignment, and
+                 * increment a single u64 in the hope to catch on again 'soon'.
+                 */
+                if (unlikely(head & 7))
+                        head &= ~7ULL;
+
+                size = 8;
+        }
+
+        head += size;
+        file_pos += size;
+
+        if (file_pos >= progress_next) {
+                progress_next += file_size / 16;
+                ui_progress__update(progress, file_pos);
+        }
+
+        if (file_pos < file_size)
+                goto more;
+
+        err = 0;
+        /* do the final flush for ordered samples */
+        session->ordered_samples.next_flush = ULLONG_MAX;
+        flush_sample_queue(session, ops);
+out_err:
+        ui_progress__delete(progress);
+        perf_session__warn_about_errors(session, ops);
+        perf_session_free_sample_buffers(session);
+        return err;
+}
+
+int perf_session__process_events(struct perf_session *self,
+                                 struct perf_event_ops *ops)
+{
+        int err;
+
+        if (perf_session__register_idle_thread(self) == NULL)
+                return -ENOMEM;
+
+        if (!self->fd_pipe)
+                err = __perf_session__process_events(self,
+                                                     self->header.data_offset,
+                                                     self->header.data_size,
+                                                     self->size, ops);
+        else
+                err = __perf_session__process_pipe_events(self, ops);
+
+        return err;
+}
+
+bool perf_session__has_traces(struct perf_session *self, const char *msg)
+{
+        if (!(self->sample_type & PERF_SAMPLE_RAW)) {
+                pr_err("No trace sample to read. Did you call 'perf %s'?\n", msg);
+                return false;
+        }
+
+        return true;
+}
+
+int perf_session__set_kallsyms_ref_reloc_sym(struct map **maps,
+                                             const char *symbol_name,
+                                             u64 addr)
+{
+        char *bracket;
+        enum map_type i;
+        struct ref_reloc_sym *ref;
+
+        ref = zalloc(sizeof(struct ref_reloc_sym));
+        if (ref == NULL)
+                return -ENOMEM;
+
+        ref->name = strdup(symbol_name);
+        if (ref->name == NULL) {
+                free(ref);
+                return -ENOMEM;
+        }
+
+        bracket = strchr(ref->name, ']');
+        if (bracket)
+                *bracket = '\0';
+
+        ref->addr = addr;
+
+        for (i = 0; i < MAP__NR_TYPES; ++i) {
+                struct kmap *kmap = map__kmap(maps[i]);
+                kmap->ref_reloc_sym = ref;
+        }
+
+        return 0;
+}
+
+size_t perf_session__fprintf_dsos(struct perf_session *self, FILE *fp)
+{
+        return __dsos__fprintf(&self->host_machine.kernel_dsos, fp) +
+               __dsos__fprintf(&self->host_machine.user_dsos, fp) +
+               machines__fprintf_dsos(&self->machines, fp);
+}
+
+size_t perf_session__fprintf_dsos_buildid(struct perf_session *self, FILE *fp,
+                                          bool with_hits)
+{
+        size_t ret = machine__fprintf_dsos_buildid(&self->host_machine, fp, with_hits);
+        return ret + machines__fprintf_dsos_buildid(&self->machines, fp, with_hits);
+}
diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h
new file mode 100644
index 00000000000..decd83f274f
--- /dev/null
+++ b/tools/perf/util/session.h
@@ -0,0 +1,157 @@
+#ifndef __PERF_SESSION_H
+#define __PERF_SESSION_H
+
+#include "hist.h"
+#include "event.h"
+#include "header.h"
+#include "symbol.h"
+#include "thread.h"
+#include <linux/rbtree.h>
+#include "../../../include/linux/perf_event.h"
+
+struct sample_queue;
+struct ip_callchain;
+struct thread;
+
+struct ordered_samples {
+        u64                     last_flush;
+        u64                     next_flush;
+        u64                     max_timestamp;
+        struct list_head        samples;
+        struct list_head        sample_cache;
+        struct list_head        to_free;
+        struct sample_queue     *sample_buffer;
+        struct sample_queue     *last_sample;
+        int                     sample_buffer_idx;
+};
+
+struct perf_session {
+        struct perf_header      header;
+        unsigned long           size;
+        unsigned long           mmap_window;
+        struct rb_root          threads;
+        struct list_head        dead_threads;
+        struct thread           *last_match;
+        struct machine          host_machine;
+        struct rb_root          machines;
+        struct rb_root          hists_tree;
+        /*
+         * FIXME: should point to the first entry in hists_tree and
+         *        be a hists instance. Right now its only 'report'
+         *        that is using ->hists_tree while all the rest use
+         *        ->hists.
+         */
+        struct hists            hists;
+        u64                     sample_type;
+        int                     fd;
+        bool                    fd_pipe;
+        bool                    repipe;
+        bool                    sample_id_all;
+        u16                     id_hdr_size;
+        int                     cwdlen;
+        char                    *cwd;
+        struct ordered_samples  ordered_samples;
+        char filename[0];
+};
+
+struct perf_event_ops;
+
+typedef int (*event_op)(event_t *self, struct sample_data *sample,
+                        struct perf_session *session);
+typedef int (*event_synth_op)(event_t *self, struct perf_session *session);
+typedef int (*event_op2)(event_t *self, struct perf_session *session,
+                         struct perf_event_ops *ops);
+
+struct perf_event_ops {
+        event_op        sample,
+                        mmap,
+                        comm,
+                        fork,
+                        exit,
+                        lost,
+                        read,
+                        throttle,
+                        unthrottle;
+        event_synth_op  attr,
+                        event_type,
+                        tracing_data,
+                        build_id;
+        event_op2       finished_round;
+        bool            ordered_samples;
+        bool            ordering_requires_timestamps;
+};
+
+struct perf_session *perf_session__new(const char *filename, int mode,
+                                       bool force, bool repipe,
+                                       struct perf_event_ops *ops);
+void perf_session__delete(struct perf_session *self);
+
+void perf_event_header__bswap(struct perf_event_header *self);
+
+int __perf_session__process_events(struct perf_session *self,
+                                   u64 data_offset, u64 data_size, u64 size,
+                                   struct perf_event_ops *ops);
+int perf_session__process_events(struct perf_session *self,
+                                 struct perf_event_ops *event_ops);
+
+struct map_symbol *perf_session__resolve_callchain(struct perf_session *self,
+                                                   struct thread *thread,
+                                                   struct ip_callchain *chain,
+                                                   struct symbol **parent);
+
+bool perf_session__has_traces(struct perf_session *self, const char *msg);
+
+int perf_session__set_kallsyms_ref_reloc_sym(struct map **maps,
+                                             const char *symbol_name,
+                                             u64 addr);
+
+void mem_bswap_64(void *src, int byte_size);
+
+int perf_session__create_kernel_maps(struct perf_session *self);
+
+void perf_session__update_sample_type(struct perf_session *self);
+void perf_session__set_sample_id_all(struct perf_session *session, bool value);
+void perf_session__set_sample_type(struct perf_session *session, u64 type);
+void perf_session__remove_thread(struct perf_session *self, struct thread *th);
+
+static inline
+struct machine *perf_session__find_host_machine(struct perf_session *self)
+{
+        return &self->host_machine;
+}
+
+static inline
+struct machine *perf_session__find_machine(struct perf_session *self, pid_t pid)
+{
+        if (pid == HOST_KERNEL_ID)
+                return &self->host_machine;
+        return machines__find(&self->machines, pid);
+}
+
+static inline
+struct machine *perf_session__findnew_machine(struct perf_session *self, pid_t pid)
+{
+        if (pid == HOST_KERNEL_ID)
+                return &self->host_machine;
+        return machines__findnew(&self->machines, pid);
+}
+
+static inline
+void perf_session__process_machines(struct perf_session *self,
+                                    machine__process_t process)
+{
+        process(&self->host_machine, self);
+        return machines__process(&self->machines, process, self);
+}
+
+size_t perf_session__fprintf_dsos(struct perf_session *self, FILE *fp);
+
+size_t perf_session__fprintf_dsos_buildid(struct perf_session *self,
+                                          FILE *fp, bool with_hits);
+
+static inline
+size_t perf_session__fprintf_nr_events(struct perf_session *self, FILE *fp)
+{
+        return hists__fprintf_nr_events(&self->hists, fp);
+}
+#endif /* __PERF_SESSION_H */
diff --git a/tools/perf/util/sigchain.c b/tools/perf/util/sigchain.c
index 1118b99e57d..ba785e9b184 100644
--- a/tools/perf/util/sigchain.c
+++ b/tools/perf/util/sigchain.c
@@ -16,7 +16,7 @@ static void check_signum(int sig)
                 die("BUG: signal out of range: %d", sig);
 }
 
-int sigchain_push(int sig, sigchain_fun f)
+static int sigchain_push(int sig, sigchain_fun f)
 {
         struct sigchain_signal *s = signals + sig;
         check_signum(sig);
diff --git a/tools/perf/util/sigchain.h b/tools/perf/util/sigchain.h
index 618083bce0c..959d64eb555 100644
--- a/tools/perf/util/sigchain.h
+++ b/tools/perf/util/sigchain.h
@@ -1,11 +1,10 @@
-#ifndef SIGCHAIN_H
-#define SIGCHAIN_H
+#ifndef __PERF_SIGCHAIN_H
+#define __PERF_SIGCHAIN_H
 
 typedef void (*sigchain_fun)(int);
 
-int sigchain_push(int sig, sigchain_fun f);
 int sigchain_pop(int sig);
 
 void sigchain_push_common(sigchain_fun f);
 
-#endif /* SIGCHAIN_H */
+#endif /* __PERF_SIGCHAIN_H */
diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c
new file mode 100644
index 00000000000..f44fa541d56
--- /dev/null
+++ b/tools/perf/util/sort.c
@@ -0,0 +1,345 @@
+#include "sort.h"
+#include "hist.h"
+
+regex_t         parent_regex;
+const char      default_parent_pattern[] = "^sys_|^do_page_fault";
+const char      *parent_pattern = default_parent_pattern;
+const char      default_sort_order[] = "comm,dso,symbol";
+const char      *sort_order = default_sort_order;
+int             sort__need_collapse = 0;
+int             sort__has_parent = 0;
+
+enum sort_type  sort__first_dimension;
+
+char * field_sep;
+
+LIST_HEAD(hist_entry__sort_list);
+
+static int hist_entry__thread_snprintf(struct hist_entry *self, char *bf,
+                                       size_t size, unsigned int width);
+static int hist_entry__comm_snprintf(struct hist_entry *self, char *bf,
+                                     size_t size, unsigned int width);
+static int hist_entry__dso_snprintf(struct hist_entry *self, char *bf,
+                                    size_t size, unsigned int width);
+static int hist_entry__sym_snprintf(struct hist_entry *self, char *bf,
+                                    size_t size, unsigned int width);
+static int hist_entry__parent_snprintf(struct hist_entry *self, char *bf,
+                                       size_t size, unsigned int width);
+static int hist_entry__cpu_snprintf(struct hist_entry *self, char *bf,
+                                    size_t size, unsigned int width);
+
+struct sort_entry sort_thread = {
+        .se_header      = "Command:  Pid",
+        .se_cmp         = sort__thread_cmp,
+        .se_snprintf    = hist_entry__thread_snprintf,
+        .se_width_idx   = HISTC_THREAD,
+};
+
+struct sort_entry sort_comm = {
+        .se_header      = "Command",
+        .se_cmp         = sort__comm_cmp,
+        .se_collapse    = sort__comm_collapse,
+        .se_snprintf    = hist_entry__comm_snprintf,
+        .se_width_idx   = HISTC_COMM,
+};
+
+struct sort_entry sort_dso = {
+        .se_header      = "Shared Object",
+        .se_cmp         = sort__dso_cmp,
+        .se_snprintf    = hist_entry__dso_snprintf,
+        .se_width_idx   = HISTC_DSO,
+};
+
+struct sort_entry sort_sym = {
+        .se_header      = "Symbol",
+        .se_cmp         = sort__sym_cmp,
+        .se_snprintf    = hist_entry__sym_snprintf,
+        .se_width_idx   = HISTC_SYMBOL,
+};
+
+struct sort_entry sort_parent = {
+        .se_header      = "Parent symbol",
+        .se_cmp         = sort__parent_cmp,
+        .se_snprintf    = hist_entry__parent_snprintf,
+        .se_width_idx   = HISTC_PARENT,
+};
+ 
+struct sort_entry sort_cpu = {
+        .se_header      = "CPU",
+        .se_cmp         = sort__cpu_cmp,
+        .se_snprintf    = hist_entry__cpu_snprintf,
+        .se_width_idx   = HISTC_CPU,
+};
+
+struct sort_dimension {
+        const char              *name;
+        struct sort_entry       *entry;
+        int                     taken;
+};
+
+static struct sort_dimension sort_dimensions[] = {
+        { .name = "pid",        .entry = &sort_thread,  },
+        { .name = "comm",       .entry = &sort_comm,    },
+        { .name = "dso",        .entry = &sort_dso,     },
+        { .name = "symbol",     .entry = &sort_sym,     },
+        { .name = "parent",     .entry = &sort_parent,  },
+        { .name = "cpu",        .entry = &sort_cpu,     },
+};
+
+int64_t cmp_null(void *l, void *r)
+{
+        if (!l && !r)
+                return 0;
+        else if (!l)
+                return -1;
+        else
+                return 1;
+}
+
+/* --sort pid */
+
+int64_t
+sort__thread_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+        return right->thread->pid - left->thread->pid;
+}
+
+static int repsep_snprintf(char *bf, size_t size, const char *fmt, ...)
+{
+        int n;
+        va_list ap;
+
+        va_start(ap, fmt);
+        n = vsnprintf(bf, size, fmt, ap);
+        if (field_sep && n > 0) {
+                char *sep = bf;
+
+                while (1) {
+                        sep = strchr(sep, *field_sep);
+                        if (sep == NULL)
+                                break;
+                        *sep = '.';
+                }
+        }
+        va_end(ap);
+        return n;
+}
+
+static int hist_entry__thread_snprintf(struct hist_entry *self, char *bf,
+                                       size_t size, unsigned int width)
+{
+        return repsep_snprintf(bf, size, "%*s:%5d", width,
+                              self->thread->comm ?: "", self->thread->pid);
+}
+
+static int hist_entry__comm_snprintf(struct hist_entry *self, char *bf,
+                                     size_t size, unsigned int width)
+{
+        return repsep_snprintf(bf, size, "%*s", width, self->thread->comm);
+}
+
+/* --sort dso */
+
+int64_t
+sort__dso_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+        struct dso *dso_l = left->ms.map ? left->ms.map->dso : NULL;
+        struct dso *dso_r = right->ms.map ? right->ms.map->dso : NULL;
+        const char *dso_name_l, *dso_name_r;
+
+        if (!dso_l || !dso_r)
+                return cmp_null(dso_l, dso_r);
+
+        if (verbose) {
+                dso_name_l = dso_l->long_name;
+                dso_name_r = dso_r->long_name;
+        } else {
+                dso_name_l = dso_l->short_name;
+                dso_name_r = dso_r->short_name;
+        }
+
+        return strcmp(dso_name_l, dso_name_r);
+}
+
+static int hist_entry__dso_snprintf(struct hist_entry *self, char *bf,
+                                    size_t size, unsigned int width)
+{
+        if (self->ms.map && self->ms.map->dso) {
+                const char *dso_name = !verbose ? self->ms.map->dso->short_name :
+                                                  self->ms.map->dso->long_name;
+                return repsep_snprintf(bf, size, "%-*s", width, dso_name);
+        }
+
+        return repsep_snprintf(bf, size, "%-*s", width, "[unknown]");
+}
+
+/* --sort symbol */
+
+int64_t
+sort__sym_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+        u64 ip_l, ip_r;
+
+        if (left->ms.sym == right->ms.sym)
+                return 0;
+
+        ip_l = left->ms.sym ? left->ms.sym->start : left->ip;
+        ip_r = right->ms.sym ? right->ms.sym->start : right->ip;
+
+        return (int64_t)(ip_r - ip_l);
+}
+
+static int hist_entry__sym_snprintf(struct hist_entry *self, char *bf,
+                                    size_t size, unsigned int width __used)
+{
+        size_t ret = 0;
+
+        if (verbose) {
+                char o = self->ms.map ? dso__symtab_origin(self->ms.map->dso) : '!';
+                ret += repsep_snprintf(bf, size, "%-#*llx %c ",
+                                       BITS_PER_LONG / 4, self->ip, o);
+        }
+
+        ret += repsep_snprintf(bf + ret, size - ret, "[%c] ", self->level);
+        if (self->ms.sym)
+                ret += repsep_snprintf(bf + ret, size - ret, "%s",
+                                       self->ms.sym->name);
+        else
+                ret += repsep_snprintf(bf + ret, size - ret, "%-#*llx",
+                                       BITS_PER_LONG / 4, self->ip);
+
+        return ret;
+}
+
+/* --sort comm */
+
+int64_t
+sort__comm_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+        return right->thread->pid - left->thread->pid;
+}
+
+int64_t
+sort__comm_collapse(struct hist_entry *left, struct hist_entry *right)
+{
+        char *comm_l = left->thread->comm;
+        char *comm_r = right->thread->comm;
+
+        if (!comm_l || !comm_r)
+                return cmp_null(comm_l, comm_r);
+
+        return strcmp(comm_l, comm_r);
+}
+
+/* --sort parent */
+
+int64_t
+sort__parent_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+        struct symbol *sym_l = left->parent;
+        struct symbol *sym_r = right->parent;
+
+        if (!sym_l || !sym_r)
+                return cmp_null(sym_l, sym_r);
+
+        return strcmp(sym_l->name, sym_r->name);
+}
+
+static int hist_entry__parent_snprintf(struct hist_entry *self, char *bf,
+                                       size_t size, unsigned int width)
+{
+        return repsep_snprintf(bf, size, "%-*s", width,
+                              self->parent ? self->parent->name : "[other]");
+}
+
+/* --sort cpu */
+
+int64_t
+sort__cpu_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+        return right->cpu - left->cpu;
+}
+
+static int hist_entry__cpu_snprintf(struct hist_entry *self, char *bf,
+                                       size_t size, unsigned int width)
+{
+        return repsep_snprintf(bf, size, "%-*d", width, self->cpu);
+}
+
+int sort_dimension__add(const char *tok)
+{
+        unsigned int i;
+
+        for (i = 0; i < ARRAY_SIZE(sort_dimensions); i++) {
+                struct sort_dimension *sd = &sort_dimensions[i];
+
+                if (sd->taken)
+                        continue;
+
+                if (strncasecmp(tok, sd->name, strlen(tok)))
+                        continue;
+
+                if (sd->entry->se_collapse)
+                        sort__need_collapse = 1;
+
+                if (sd->entry == &sort_parent) {
+                        int ret = regcomp(&parent_regex, parent_pattern, REG_EXTENDED);
+                        if (ret) {
+                                char err[BUFSIZ];
+
+                                regerror(ret, &parent_regex, err, sizeof(err));
+                                pr_err("Invalid regex: %s\n%s", parent_pattern, err);
+                                return -EINVAL;
+                        }
+                        sort__has_parent = 1;
+                }
+
+                if (list_empty(&hist_entry__sort_list)) {
+                        if (!strcmp(sd->name, "pid"))
+                                sort__first_dimension = SORT_PID;
+                        else if (!strcmp(sd->name, "comm"))
+                                sort__first_dimension = SORT_COMM;
+                        else if (!strcmp(sd->name, "dso"))
+                                sort__first_dimension = SORT_DSO;
+                        else if (!strcmp(sd->name, "symbol"))
+                                sort__first_dimension = SORT_SYM;
+                        else if (!strcmp(sd->name, "parent"))
+                                sort__first_dimension = SORT_PARENT;
+                        else if (!strcmp(sd->name, "cpu"))
+                                sort__first_dimension = SORT_CPU;
+                }
+
+                list_add_tail(&sd->entry->list, &hist_entry__sort_list);
+                sd->taken = 1;
+
+                return 0;
+        }
+
+        return -ESRCH;
+}
+
+void setup_sorting(const char * const usagestr[], const struct option *opts)
+{
+        char *tmp, *tok, *str = strdup(sort_order);
+
+        for (tok = strtok_r(str, ", ", &tmp);
+                        tok; tok = strtok_r(NULL, ", ", &tmp)) {
+                if (sort_dimension__add(tok) < 0) {
+                        error("Unknown --sort key: `%s'", tok);
+                        usage_with_options(usagestr, opts);
+                }
+        }
+
+        free(str);
+}
+
+void sort_entry__setup_elide(struct sort_entry *self, struct strlist *list,
+                             const char *list_name, FILE *fp)
+{
+        if (list && strlist__nr_entries(list) == 1) {
+                if (fp != NULL)
+                        fprintf(fp, "# %s: %s\n", list_name,
+                                strlist__entry(list, 0)->s);
+                self->elide = true;
+        }
+}
diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h
new file mode 100644
index 00000000000..0b91053a7d1
--- /dev/null
+++ b/tools/perf/util/sort.h
@@ -0,0 +1,124 @@
+#ifndef __PERF_SORT_H
+#define __PERF_SORT_H
+#include "../builtin.h"
+
+#include "util.h"
+
+#include "color.h"
+#include <linux/list.h>
+#include "cache.h"
+#include <linux/rbtree.h>
+#include "symbol.h"
+#include "string.h"
+#include "callchain.h"
+#include "strlist.h"
+#include "values.h"
+
+#include "../perf.h"
+#include "debug.h"
+#include "header.h"
+
+#include "parse-options.h"
+#include "parse-events.h"
+
+#include "thread.h"
+#include "sort.h"
+
+extern regex_t parent_regex;
+extern const char *sort_order;
+extern const char default_parent_pattern[];
+extern const char *parent_pattern;
+extern const char default_sort_order[];
+extern int sort__need_collapse;
+extern int sort__has_parent;
+extern char *field_sep;
+extern struct sort_entry sort_comm;
+extern struct sort_entry sort_dso;
+extern struct sort_entry sort_sym;
+extern struct sort_entry sort_parent;
+extern enum sort_type sort__first_dimension;
+
+/**
+ * struct hist_entry - histogram entry
+ *
+ * @row_offset - offset from the first callchain expanded to appear on screen
+ * @nr_rows - rows expanded in callchain, recalculated on folding/unfolding
+ */
+struct hist_entry {
+        struct rb_node          rb_node;
+        u64                     period;
+        u64                     period_sys;
+        u64                     period_us;
+        u64                     period_guest_sys;
+        u64                     period_guest_us;
+        struct map_symbol       ms;
+        struct thread           *thread;
+        u64                     ip;
+        s32                     cpu;
+        u32                     nr_events;
+
+        /* XXX These two should move to some tree widget lib */
+        u16                     row_offset;
+        u16                     nr_rows;
+
+        bool                    init_have_children;
+        char                    level;
+        u8                      filtered;
+        struct symbol           *parent;
+        union {
+                unsigned long     position;
+                struct hist_entry *pair;
+                struct rb_root    sorted_chain;
+        };
+        struct callchain_root   callchain[0];
+};
+
+enum sort_type {
+        SORT_PID,
+        SORT_COMM,
+        SORT_DSO,
+        SORT_SYM,
+        SORT_PARENT,
+        SORT_CPU,
+};
+
+/*
+ * configurable sorting bits
+ */
+
+struct sort_entry {
+        struct list_head list;
+
+        const char *se_header;
+
+        int64_t (*se_cmp)(struct hist_entry *, struct hist_entry *);
+        int64_t (*se_collapse)(struct hist_entry *, struct hist_entry *);
+        int     (*se_snprintf)(struct hist_entry *self, char *bf, size_t size,
+                               unsigned int width);
+        u8      se_width_idx;
+        bool    elide;
+};
+
+extern struct sort_entry sort_thread;
+extern struct list_head hist_entry__sort_list;
+
+void setup_sorting(const char * const usagestr[], const struct option *opts);
+
+extern size_t sort__thread_print(FILE *, struct hist_entry *, unsigned int);
+extern size_t sort__comm_print(FILE *, struct hist_entry *, unsigned int);
+extern size_t sort__dso_print(FILE *, struct hist_entry *, unsigned int);
+extern size_t sort__sym_print(FILE *, struct hist_entry *, unsigned int __used);
+extern int64_t cmp_null(void *, void *);
+extern int64_t sort__thread_cmp(struct hist_entry *, struct hist_entry *);
+extern int64_t sort__comm_cmp(struct hist_entry *, struct hist_entry *);
+extern int64_t sort__comm_collapse(struct hist_entry *, struct hist_entry *);
+extern int64_t sort__dso_cmp(struct hist_entry *, struct hist_entry *);
+extern int64_t sort__sym_cmp(struct hist_entry *, struct hist_entry *);
+extern int64_t sort__parent_cmp(struct hist_entry *, struct hist_entry *);
+int64_t sort__cpu_cmp(struct hist_entry *left, struct hist_entry *right);
+extern size_t sort__parent_print(FILE *, struct hist_entry *, unsigned int);
+extern int sort_dimension__add(const char *);
+void sort_entry__setup_elide(struct sort_entry *self, struct strlist *list,
+                             const char *list_name, FILE *fp);
+
+#endif  /* __PERF_SORT_H */
diff --git a/tools/perf/util/strbuf.c b/tools/perf/util/strbuf.c
index eaba0930680..92e068517c1 100644
--- a/tools/perf/util/strbuf.c
+++ b/tools/perf/util/strbuf.c
@@ -16,7 +16,7 @@ int prefixcmp(const char *str, const char *prefix)
  */
 char strbuf_slopbuf[1];
 
-void strbuf_init(struct strbuf *sb, size_t hint)
+void strbuf_init(struct strbuf *sb, ssize_t hint)
 {
         sb->alloc = sb->len = 0;
         sb->buf = strbuf_slopbuf;
@@ -41,16 +41,6 @@ char *strbuf_detach(struct strbuf *sb, size_t *sz)
         return res;
 }
 
-void strbuf_attach(struct strbuf *sb, void *buf, size_t len, size_t alloc)
-{
-        strbuf_release(sb);
-        sb->buf   = buf;
-        sb->len   = len;
-        sb->alloc = alloc;
-        strbuf_grow(sb, 0);
-        sb->buf[sb->len] = '\0';
-}
-
 void strbuf_grow(struct strbuf *sb, size_t extra)
 {
         if (sb->len + extra + 1 <= sb->len)
@@ -60,93 +50,7 @@ void strbuf_grow(struct strbuf *sb, size_t extra)
         ALLOC_GROW(sb->buf, sb->len + extra + 1, sb->alloc);
 }
 
-void strbuf_trim(struct strbuf *sb)
-{
-        char *b = sb->buf;
-        while (sb->len > 0 && isspace((unsigned char)sb->buf[sb->len - 1]))
-                sb->len--;
-        while (sb->len > 0 && isspace(*b)) {
-                b++;
-                sb->len--;
-        }
-        memmove(sb->buf, b, sb->len);
-        sb->buf[sb->len] = '\0';
-}
-void strbuf_rtrim(struct strbuf *sb)
-{
-        while (sb->len > 0 && isspace((unsigned char)sb->buf[sb->len - 1]))
-                sb->len--;
-        sb->buf[sb->len] = '\0';
-}
-
-void strbuf_ltrim(struct strbuf *sb)
-{
-        char *b = sb->buf;
-        while (sb->len > 0 && isspace(*b)) {
-                b++;
-                sb->len--;
-        }
-        memmove(sb->buf, b, sb->len);
-        sb->buf[sb->len] = '\0';
-}
-
-void strbuf_tolower(struct strbuf *sb)
-{
-        int i;
-        for (i = 0; i < sb->len; i++)
-                sb->buf[i] = tolower(sb->buf[i]);
-}
-
-struct strbuf **strbuf_split(const struct strbuf *sb, int delim)
-{
-        int alloc = 2, pos = 0;
-        char *n, *p;
-        struct strbuf **ret;
-        struct strbuf *t;
-
-        ret = calloc(alloc, sizeof(struct strbuf *));
-        p = n = sb->buf;
-        while (n < sb->buf + sb->len) {
-                int len;
-                n = memchr(n, delim, sb->len - (n - sb->buf));
-                if (pos + 1 >= alloc) {
-                        alloc = alloc * 2;
-                        ret = realloc(ret, sizeof(struct strbuf *) * alloc);
-                }
-                if (!n)
-                        n = sb->buf + sb->len - 1;
-                len = n - p + 1;
-                t = malloc(sizeof(struct strbuf));
-                strbuf_init(t, len);
-                strbuf_add(t, p, len);
-                ret[pos] = t;
-                ret[++pos] = NULL;
-                p = ++n;
-        }
-        return ret;
-}
-
-void strbuf_list_free(struct strbuf **sbs)
-{
-        struct strbuf **s = sbs;
-
-        while (*s) {
-                strbuf_release(*s);
-                free(*s++);
-        }
-        free(sbs);
-}
-
-int strbuf_cmp(const struct strbuf *a, const struct strbuf *b)
-{
-        int len = a->len < b->len ? a->len: b->len;
-        int cmp = memcmp(a->buf, b->buf, len);
-        if (cmp)
-                return cmp;
-        return a->len < b->len ? -1: a->len != b->len;
-}
-
-void strbuf_splice(struct strbuf *sb, size_t pos, size_t len,
+static void strbuf_splice(struct strbuf *sb, size_t pos, size_t len,
                                    const void *data, size_t dlen)
 {
         if (pos + len < pos)
@@ -165,11 +69,6 @@ void strbuf_splice(struct strbuf *sb, size_t pos, size_t len,
         strbuf_setlen(sb, sb->len + dlen - len);
 }
 
-void strbuf_insert(struct strbuf *sb, size_t pos, const void *data, size_t len)
-{
-        strbuf_splice(sb, pos, 0, data, len);
-}
-
 void strbuf_remove(struct strbuf *sb, size_t pos, size_t len)
 {
         strbuf_splice(sb, pos, len, NULL, 0);
@@ -182,13 +81,6 @@ void strbuf_add(struct strbuf *sb, const void *data, size_t len)
         strbuf_setlen(sb, sb->len + len);
 }
 
-void strbuf_adddup(struct strbuf *sb, size_t pos, size_t len)
-{
-        strbuf_grow(sb, len);
-        memcpy(sb->buf + sb->len, sb->buf + pos, len);
-        strbuf_setlen(sb, sb->len + len);
-}
-
 void strbuf_addf(struct strbuf *sb, const char *fmt, ...)
 {
         int len;
@@ -213,58 +105,7 @@ void strbuf_addf(struct strbuf *sb, const char *fmt, ...)
         strbuf_setlen(sb, sb->len + len);
 }
 
-void strbuf_expand(struct strbuf *sb, const char *format, expand_fn_t fn,
-                   void *context)
-{
-        for (;;) {
-                const char *percent;
-                size_t consumed;
-
-                percent = strchrnul(format, '%');
-                strbuf_add(sb, format, percent - format);
-                if (!*percent)
-                        break;
-                format = percent + 1;
-
-                consumed = fn(sb, format, context);
-                if (consumed)
-                        format += consumed;
-                else
-                        strbuf_addch(sb, '%');
-        }
-}
-
-size_t strbuf_expand_dict_cb(struct strbuf *sb, const char *placeholder,
-                void *context)
-{
-        struct strbuf_expand_dict_entry *e = context;
-        size_t len;
-
-        for (; e->placeholder && (len = strlen(e->placeholder)); e++) {
-                if (!strncmp(placeholder, e->placeholder, len)) {
-                        if (e->value)
-                                strbuf_addstr(sb, e->value);
-                        return len;
-                }
-        }
-        return 0;
-}
-
-size_t strbuf_fread(struct strbuf *sb, size_t size, FILE *f)
-{
-        size_t res;
-        size_t oldalloc = sb->alloc;
-
-        strbuf_grow(sb, size);
-        res = fread(sb->buf + sb->len, 1, size, f);
-        if (res > 0)
-                strbuf_setlen(sb, sb->len + res);
-        else if (res < 0 && oldalloc == 0)
-                strbuf_release(sb);
-        return res;
-}
-
-ssize_t strbuf_read(struct strbuf *sb, int fd, size_t hint)
+ssize_t strbuf_read(struct strbuf *sb, int fd, ssize_t hint)
 {
         size_t oldlen = sb->len;
         size_t oldalloc = sb->alloc;
@@ -290,70 +131,3 @@ ssize_t strbuf_read(struct strbuf *sb, int fd, size_t hint)
         sb->buf[sb->len] = '\0';
         return sb->len - oldlen;
 }
-
-#define STRBUF_MAXLINK (2*PATH_MAX)
-
-int strbuf_readlink(struct strbuf *sb, const char *path, size_t hint)
-{
-        size_t oldalloc = sb->alloc;
-
-        if (hint < 32)
-                hint = 32;
-
-        while (hint < STRBUF_MAXLINK) {
-                int len;
-
-                strbuf_grow(sb, hint);
-                len = readlink(path, sb->buf, hint);
-                if (len < 0) {
-                        if (errno != ERANGE)
-                                break;
-                } else if (len < hint) {
-                        strbuf_setlen(sb, len);
-                        return 0;
-                }
-
-                /* .. the buffer was too small - try again */
-                hint *= 2;
-        }
-        if (oldalloc == 0)
-                strbuf_release(sb);
-        return -1;
-}
-
-int strbuf_getline(struct strbuf *sb, FILE *fp, int term)
-{
-        int ch;
-
-        strbuf_grow(sb, 0);
-        if (feof(fp))
-                return EOF;
-
-        strbuf_reset(sb);
-        while ((ch = fgetc(fp)) != EOF) {
-                if (ch == term)
-                        break;
-                strbuf_grow(sb, 1);
-                sb->buf[sb->len++] = ch;
-        }
-        if (ch == EOF && sb->len == 0)
-                return EOF;
-
-        sb->buf[sb->len] = '\0';
-        return 0;
-}
-
-int strbuf_read_file(struct strbuf *sb, const char *path, size_t hint)
-{
-        int fd, len;
-
-        fd = open(path, O_RDONLY);
-        if (fd < 0)
-                return -1;
-        len = strbuf_read(sb, fd, hint);
-        close(fd);
-        if (len < 0)
-                return -1;
-
-        return len;
-}
diff --git a/tools/perf/util/strbuf.h b/tools/perf/util/strbuf.h
index 9ee908a3ec5..436ac319f6c 100644
--- a/tools/perf/util/strbuf.h
+++ b/tools/perf/util/strbuf.h
@@ -1,5 +1,5 @@
-#ifndef STRBUF_H
-#define STRBUF_H
+#ifndef __PERF_STRBUF_H
+#define __PERF_STRBUF_H
 
 /*
  * Strbuf's can be use in many ways: as a byte array, or to store arbitrary
@@ -50,18 +50,12 @@ struct strbuf {
 #define STRBUF_INIT  { 0, 0, strbuf_slopbuf }
 
 /*----- strbuf life cycle -----*/
-extern void strbuf_init(struct strbuf *, size_t);
+extern void strbuf_init(struct strbuf *buf, ssize_t hint);
 extern void strbuf_release(struct strbuf *);
 extern char *strbuf_detach(struct strbuf *, size_t *);
-extern void strbuf_attach(struct strbuf *, void *, size_t, size_t);
-static inline void strbuf_swap(struct strbuf *a, struct strbuf *b) {
-        struct strbuf tmp = *a;
-        *a = *b;
-        *b = tmp;
-}
 
 /*----- strbuf size related -----*/
-static inline size_t strbuf_avail(const struct strbuf *sb) {
+static inline ssize_t strbuf_avail(const struct strbuf *sb) {
         return sb->alloc ? sb->alloc - sb->len - 1 : 0;
 }
 
@@ -74,17 +68,6 @@ static inline void strbuf_setlen(struct strbuf *sb, size_t len) {
         sb->len = len;
         sb->buf[len] = '\0';
 }
-#define strbuf_reset(sb)  strbuf_setlen(sb, 0)
-
-/*----- content related -----*/
-extern void strbuf_trim(struct strbuf *);
-extern void strbuf_rtrim(struct strbuf *);
-extern void strbuf_ltrim(struct strbuf *);
-extern int strbuf_cmp(const struct strbuf *, const struct strbuf *);
-extern void strbuf_tolower(struct strbuf *);
-
-extern struct strbuf **strbuf_split(const struct strbuf *, int delim);
-extern void strbuf_list_free(struct strbuf **);
 
 /*----- add data in your buffer -----*/
 static inline void strbuf_addch(struct strbuf *sb, int c) {
@@ -93,45 +76,17 @@ static inline void strbuf_addch(struct strbuf *sb, int c) {
         sb->buf[sb->len] = '\0';
 }
 
-extern void strbuf_insert(struct strbuf *, size_t pos, const void *, size_t);
 extern void strbuf_remove(struct strbuf *, size_t pos, size_t len);
 
-/* splice pos..pos+len with given data */
-extern void strbuf_splice(struct strbuf *, size_t pos, size_t len,
-                          const void *, size_t);
-
 extern void strbuf_add(struct strbuf *, const void *, size_t);
 static inline void strbuf_addstr(struct strbuf *sb, const char *s) {
         strbuf_add(sb, s, strlen(s));
 }
-static inline void strbuf_addbuf(struct strbuf *sb, const struct strbuf *sb2) {
-        strbuf_add(sb, sb2->buf, sb2->len);
-}
-extern void strbuf_adddup(struct strbuf *sb, size_t pos, size_t len);
-
-typedef size_t (*expand_fn_t) (struct strbuf *sb, const char *placeholder, void *context);
-extern void strbuf_expand(struct strbuf *sb, const char *format, expand_fn_t fn, void *context);
-struct strbuf_expand_dict_entry {
-        const char *placeholder;
-        const char *value;
-};
-extern size_t strbuf_expand_dict_cb(struct strbuf *sb, const char *placeholder, void *context);
 
 __attribute__((format(printf,2,3)))
 extern void strbuf_addf(struct strbuf *sb, const char *fmt, ...);
 
-extern size_t strbuf_fread(struct strbuf *, size_t, FILE *);
 /* XXX: if read fails, any partial read is undone */
-extern ssize_t strbuf_read(struct strbuf *, int fd, size_t hint);
-extern int strbuf_read_file(struct strbuf *sb, const char *path, size_t hint);
-extern int strbuf_readlink(struct strbuf *sb, const char *path, size_t hint);
-
-extern int strbuf_getline(struct strbuf *, FILE *, int);
-
-extern void stripspace(struct strbuf *buf, int skip_comments);
-extern int launch_editor(const char *path, struct strbuf *buffer, const char *const *env);
-
-extern int strbuf_branchname(struct strbuf *sb, const char *name);
-extern int strbuf_check_branch_ref(struct strbuf *sb, const char *name);
+extern ssize_t strbuf_read(struct strbuf *, int fd, ssize_t hint);
 
-#endif /* STRBUF_H */
+#endif /* __PERF_STRBUF_H */
diff --git a/tools/perf/util/string.c b/tools/perf/util/string.c
index c93eca9a7be..8fc0bd3a3a4 100644
--- a/tools/perf/util/string.c
+++ b/tools/perf/util/string.c
@@ -1,34 +1,296 @@
+#include "util.h"
 #include "string.h"
 
-static int hex(char ch)
+#define K 1024LL
+/*
+ * perf_atoll()
+ * Parse (\d+)(b|B|kb|KB|mb|MB|gb|GB|tb|TB) (e.g. "256MB")
+ * and return its numeric value
+ */
+s64 perf_atoll(const char *str)
 {
-        if ((ch >= '0') && (ch <= '9'))
-                return ch - '0';
-        if ((ch >= 'a') && (ch <= 'f'))
-                return ch - 'a' + 10;
-        if ((ch >= 'A') && (ch <= 'F'))
-                return ch - 'A' + 10;
-        return -1;
+        unsigned int i;
+        s64 length = -1, unit = 1;
+
+        if (!isdigit(str[0]))
+                goto out_err;
+
+        for (i = 1; i < strlen(str); i++) {
+                switch (str[i]) {
+                case 'B':
+                case 'b':
+                        break;
+                case 'K':
+                        if (str[i + 1] != 'B')
+                                goto out_err;
+                        else
+                                goto kilo;
+                case 'k':
+                        if (str[i + 1] != 'b')
+                                goto out_err;
+kilo:
+                        unit = K;
+                        break;
+                case 'M':
+                        if (str[i + 1] != 'B')
+                                goto out_err;
+                        else
+                                goto mega;
+                case 'm':
+                        if (str[i + 1] != 'b')
+                                goto out_err;
+mega:
+                        unit = K * K;
+                        break;
+                case 'G':
+                        if (str[i + 1] != 'B')
+                                goto out_err;
+                        else
+                                goto giga;
+                case 'g':
+                        if (str[i + 1] != 'b')
+                                goto out_err;
+giga:
+                        unit = K * K * K;
+                        break;
+                case 'T':
+                        if (str[i + 1] != 'B')
+                                goto out_err;
+                        else
+                                goto tera;
+                case 't':
+                        if (str[i + 1] != 'b')
+                                goto out_err;
+tera:
+                        unit = K * K * K * K;
+                        break;
+                case '\0':      /* only specified figures */
+                        unit = 1;
+                        break;
+                default:
+                        if (!isdigit(str[i]))
+                                goto out_err;
+                        break;
+                }
+        }
+
+        length = atoll(str) * unit;
+        goto out;
+
+out_err:
+        length = -1;
+out:
+        return length;
 }
 
 /*
- * While we find nice hex chars, build a long_val.
- * Return number of chars processed.
+ * Helper function for splitting a string into an argv-like array.
+ * originaly copied from lib/argv_split.c
  */
-int hex2u64(const char *ptr, u64 *long_val)
+static const char *skip_sep(const char *cp)
 {
-        const char *p = ptr;
-        *long_val = 0;
+        while (*cp && isspace(*cp))
+                cp++;
 
-        while (*p) {
-                const int hex_val = hex(*p);
+        return cp;
+}
 
-                if (hex_val < 0)
-                        break;
+static const char *skip_arg(const char *cp)
+{
+        while (*cp && !isspace(*cp))
+                cp++;
+
+        return cp;
+}
+
+static int count_argc(const char *str)
+{
+        int count = 0;
+
+        while (*str) {
+                str = skip_sep(str);
+                if (*str) {
+                        count++;
+                        str = skip_arg(str);
+                }
+        }
+
+        return count;
+}
+
+/**
+ * argv_free - free an argv
+ * @argv - the argument vector to be freed
+ *
+ * Frees an argv and the strings it points to.
+ */
+void argv_free(char **argv)
+{
+        char **p;
+        for (p = argv; *p; p++)
+                free(*p);
+
+        free(argv);
+}
+
+/**
+ * argv_split - split a string at whitespace, returning an argv
+ * @str: the string to be split
+ * @argcp: returned argument count
+ *
+ * Returns an array of pointers to strings which are split out from
+ * @str.  This is performed by strictly splitting on white-space; no
+ * quote processing is performed.  Multiple whitespace characters are
+ * considered to be a single argument separator.  The returned array
+ * is always NULL-terminated.  Returns NULL on memory allocation
+ * failure.
+ */
+char **argv_split(const char *str, int *argcp)
+{
+        int argc = count_argc(str);
+        char **argv = zalloc(sizeof(*argv) * (argc+1));
+        char **argvp;
+
+        if (argv == NULL)
+                goto out;
+
+        if (argcp)
+                *argcp = argc;
+
+        argvp = argv;
+
+        while (*str) {
+                str = skip_sep(str);
+
+                if (*str) {
+                        const char *p = str;
+                        char *t;
+
+                        str = skip_arg(str);
 
-                *long_val = (*long_val << 4) | hex_val;
-                p++;
+                        t = strndup(p, str-p);
+                        if (t == NULL)
+                                goto fail;
+                        *argvp++ = t;
+                }
         }
+        *argvp = NULL;
 
-        return p - ptr;
+out:
+        return argv;
+
+fail:
+        argv_free(argv);
+        return NULL;
+}
+
+/* Character class matching */
+static bool __match_charclass(const char *pat, char c, const char **npat)
+{
+        bool complement = false, ret = true;
+
+        if (*pat == '!') {
+                complement = true;
+                pat++;
+        }
+        if (*pat++ == c)        /* First character is special */
+                goto end;
+
+        while (*pat && *pat != ']') {   /* Matching */
+                if (*pat == '-' && *(pat + 1) != ']') { /* Range */
+                        if (*(pat - 1) <= c && c <= *(pat + 1))
+                                goto end;
+                        if (*(pat - 1) > *(pat + 1))
+                                goto error;
+                        pat += 2;
+                } else if (*pat++ == c)
+                        goto end;
+        }
+        if (!*pat)
+                goto error;
+        ret = false;
+
+end:
+        while (*pat && *pat != ']')     /* Searching closing */
+                pat++;
+        if (!*pat)
+                goto error;
+        *npat = pat + 1;
+        return complement ? !ret : ret;
+
+error:
+        return false;
+}
+
+/* Glob/lazy pattern matching */
+static bool __match_glob(const char *str, const char *pat, bool ignore_space)
+{
+        while (*str && *pat && *pat != '*') {
+                if (ignore_space) {
+                        /* Ignore spaces for lazy matching */
+                        if (isspace(*str)) {
+                                str++;
+                                continue;
+                        }
+                        if (isspace(*pat)) {
+                                pat++;
+                                continue;
+                        }
+                }
+                if (*pat == '?') {      /* Matches any single character */
+                        str++;
+                        pat++;
+                        continue;
+                } else if (*pat == '[') /* Character classes/Ranges */
+                        if (__match_charclass(pat + 1, *str, &pat)) {
+                                str++;
+                                continue;
+                        } else
+                                return false;
+                else if (*pat == '\\') /* Escaped char match as normal char */
+                        pat++;
+                if (*str++ != *pat++)
+                        return false;
+        }
+        /* Check wild card */
+        if (*pat == '*') {
+                while (*pat == '*')
+                        pat++;
+                if (!*pat)      /* Tail wild card matches all */
+                        return true;
+                while (*str)
+                        if (__match_glob(str++, pat, ignore_space))
+                                return true;
+        }
+        return !*str && !*pat;
+}
+
+/**
+ * strglobmatch - glob expression pattern matching
+ * @str: the target string to match
+ * @pat: the pattern string to match
+ *
+ * This returns true if the @str matches @pat. @pat can includes wildcards
+ * ('*','?') and character classes ([CHARS], complementation and ranges are
+ * also supported). Also, this supports escape character ('\') to use special
+ * characters as normal character.
+ *
+ * Note: if @pat syntax is broken, this always returns false.
+ */
+bool strglobmatch(const char *str, const char *pat)
+{
+        return __match_glob(str, pat, false);
+}
+
+/**
+ * strlazymatch - matching pattern strings lazily with glob pattern
+ * @str: the target string to match
+ * @pat: the pattern string to match
+ *
+ * This is similar to strglobmatch, except this ignores spaces in
+ * the target string.
+ */
+bool strlazymatch(const char *str, const char *pat)
+{
+        return __match_glob(str, pat, true);
 }
diff --git a/tools/perf/util/string.h b/tools/perf/util/string.h
deleted file mode 100644
index 37b03255b42..00000000000
--- a/tools/perf/util/string.h
+++ /dev/null
@@ -1,8 +0,0 @@
-#ifndef _PERF_STRING_H_
-#define _PERF_STRING_H_
-
-#include "../types.h"
-
-int hex2u64(const char *ptr, u64 *val);
-
-#endif
diff --git a/tools/perf/util/strlist.c b/tools/perf/util/strlist.c
new file mode 100644
index 00000000000..6783a204355
--- /dev/null
+++ b/tools/perf/util/strlist.c
@@ -0,0 +1,200 @@
+/*
+ * (c) 2009 Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Licensed under the GPLv2.
+ */
+
+#include "strlist.h"
+#include <errno.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+static struct str_node *str_node__new(const char *s, bool dupstr)
+{
+        struct str_node *self = malloc(sizeof(*self));
+
+        if (self != NULL) {
+                if (dupstr) {
+                        s = strdup(s);
+                        if (s == NULL)
+                                goto out_delete;
+                }
+                self->s = s;
+        }
+
+        return self;
+
+out_delete:
+        free(self);
+        return NULL;
+}
+
+static void str_node__delete(struct str_node *self, bool dupstr)
+{
+        if (dupstr)
+                free((void *)self->s);
+        free(self);
+}
+
+int strlist__add(struct strlist *self, const char *new_entry)
+{
+        struct rb_node **p = &self->entries.rb_node;
+        struct rb_node *parent = NULL;
+        struct str_node *sn;
+
+        while (*p != NULL) {
+                int rc;
+
+                parent = *p;
+                sn = rb_entry(parent, struct str_node, rb_node);
+                rc = strcmp(sn->s, new_entry);
+
+                if (rc > 0)
+                        p = &(*p)->rb_left;
+                else if (rc < 0)
+                        p = &(*p)->rb_right;
+                else
+                        return -EEXIST;
+        }
+
+        sn = str_node__new(new_entry, self->dupstr);
+        if (sn == NULL)
+                return -ENOMEM;
+
+        rb_link_node(&sn->rb_node, parent, p);
+        rb_insert_color(&sn->rb_node, &self->entries);
+        ++self->nr_entries;
+
+        return 0;
+}
+
+int strlist__load(struct strlist *self, const char *filename)
+{
+        char entry[1024];
+        int err;
+        FILE *fp = fopen(filename, "r");
+
+        if (fp == NULL)
+                return errno;
+
+        while (fgets(entry, sizeof(entry), fp) != NULL) {
+                const size_t len = strlen(entry);
+
+                if (len == 0)
+                        continue;
+                entry[len - 1] = '\0';
+
+                err = strlist__add(self, entry);
+                if (err != 0)
+                        goto out;
+        }
+
+        err = 0;
+out:
+        fclose(fp);
+        return err;
+}
+
+void strlist__remove(struct strlist *self, struct str_node *sn)
+{
+        rb_erase(&sn->rb_node, &self->entries);
+        str_node__delete(sn, self->dupstr);
+}
+
+struct str_node *strlist__find(struct strlist *self, const char *entry)
+{
+        struct rb_node **p = &self->entries.rb_node;
+        struct rb_node *parent = NULL;
+
+        while (*p != NULL) {
+                struct str_node *sn;
+                int rc;
+
+                parent = *p;
+                sn = rb_entry(parent, struct str_node, rb_node);
+                rc = strcmp(sn->s, entry);
+
+                if (rc > 0)
+                        p = &(*p)->rb_left;
+                else if (rc < 0)
+                        p = &(*p)->rb_right;
+                else
+                        return sn;
+        }
+
+        return NULL;
+}
+
+static int strlist__parse_list_entry(struct strlist *self, const char *s)
+{
+        if (strncmp(s, "file://", 7) == 0)
+                return strlist__load(self, s + 7);
+
+        return strlist__add(self, s);
+}
+
+int strlist__parse_list(struct strlist *self, const char *s)
+{
+        char *sep;
+        int err;
+
+        while ((sep = strchr(s, ',')) != NULL) {
+                *sep = '\0';
+                err = strlist__parse_list_entry(self, s);
+                *sep = ',';
+                if (err != 0)
+                        return err;
+                s = sep + 1;
+        }
+
+        return *s ? strlist__parse_list_entry(self, s) : 0;
+}
+
+struct strlist *strlist__new(bool dupstr, const char *slist)
+{
+        struct strlist *self = malloc(sizeof(*self));
+
+        if (self != NULL) {
+                self->entries    = RB_ROOT;
+                self->dupstr     = dupstr;
+                self->nr_entries = 0;
+                if (slist && strlist__parse_list(self, slist) != 0)
+                        goto out_error;
+        }
+
+        return self;
+out_error:
+        free(self);
+        return NULL;
+}
+
+void strlist__delete(struct strlist *self)
+{
+        if (self != NULL) {
+                struct str_node *pos;
+                struct rb_node *next = rb_first(&self->entries);
+
+                while (next) {
+                        pos = rb_entry(next, struct str_node, rb_node);
+                        next = rb_next(&pos->rb_node);
+                        strlist__remove(self, pos);
+                }
+                self->entries = RB_ROOT;
+                free(self);
+        }
+}
+
+struct str_node *strlist__entry(const struct strlist *self, unsigned int idx)
+{
+        struct rb_node *nd;
+
+        for (nd = rb_first(&self->entries); nd; nd = rb_next(nd)) {
+                struct str_node *pos = rb_entry(nd, struct str_node, rb_node);
+
+                if (!idx--)
+                        return pos;
+        }
+
+        return NULL;
+}
diff --git a/tools/perf/util/strlist.h b/tools/perf/util/strlist.h
new file mode 100644
index 00000000000..3ba839007d2
--- /dev/null
+++ b/tools/perf/util/strlist.h
@@ -0,0 +1,78 @@
+#ifndef __PERF_STRLIST_H
+#define __PERF_STRLIST_H
+
+#include <linux/rbtree.h>
+#include <stdbool.h>
+
+struct str_node {
+        struct rb_node rb_node;
+        const char     *s;
+};
+
+struct strlist {
+        struct rb_root entries;
+        unsigned int   nr_entries;
+        bool           dupstr;
+};
+
+struct strlist *strlist__new(bool dupstr, const char *slist);
+void strlist__delete(struct strlist *self);
+
+void strlist__remove(struct strlist *self, struct str_node *sn);
+int strlist__load(struct strlist *self, const char *filename);
+int strlist__add(struct strlist *self, const char *str);
+
+struct str_node *strlist__entry(const struct strlist *self, unsigned int idx);
+struct str_node *strlist__find(struct strlist *self, const char *entry);
+
+static inline bool strlist__has_entry(struct strlist *self, const char *entry)
+{
+        return strlist__find(self, entry) != NULL;
+}
+
+static inline bool strlist__empty(const struct strlist *self)
+{
+        return self->nr_entries == 0;
+}
+
+static inline unsigned int strlist__nr_entries(const struct strlist *self)
+{
+        return self->nr_entries;
+}
+
+/* For strlist iteration */
+static inline struct str_node *strlist__first(struct strlist *self)
+{
+        struct rb_node *rn = rb_first(&self->entries);
+        return rn ? rb_entry(rn, struct str_node, rb_node) : NULL;
+}
+static inline struct str_node *strlist__next(struct str_node *sn)
+{
+        struct rb_node *rn;
+        if (!sn)
+                return NULL;
+        rn = rb_next(&sn->rb_node);
+        return rn ? rb_entry(rn, struct str_node, rb_node) : NULL;
+}
+
+/**
+ * strlist_for_each      - iterate over a strlist
+ * @pos:        the &struct str_node to use as a loop cursor.
+ * @self:       the &struct strlist for loop.
+ */
+#define strlist__for_each(pos, self)    \
+        for (pos = strlist__first(self); pos; pos = strlist__next(pos))
+
+/**
+ * strlist_for_each_safe - iterate over a strlist safe against removal of
+ *                         str_node
+ * @pos:        the &struct str_node to use as a loop cursor.
+ * @n:          another &struct str_node to use as temporary storage.
+ * @self:       the &struct strlist for loop.
+ */
+#define strlist__for_each_safe(pos, n, self)    \
+        for (pos = strlist__first(self), n = strlist__next(pos); pos;\
+             pos = n, n = strlist__next(n))
+
+int strlist__parse_list(struct strlist *self, const char *s);
+#endif /* __PERF_STRLIST_H */
diff --git a/tools/perf/util/svghelper.c b/tools/perf/util/svghelper.c
new file mode 100644
index 00000000000..b3637db025a
--- /dev/null
+++ b/tools/perf/util/svghelper.c
@@ -0,0 +1,500 @@
+/*
+ * svghelper.c - helper functions for outputting svg
+ *
+ * (C) Copyright 2009 Intel Corporation
+ *
+ * Authors:
+ *     Arjan van de Ven <arjan@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <string.h>
+
+#include "svghelper.h"
+
+static u64 first_time, last_time;
+static u64 turbo_frequency, max_freq;
+
+
+#define SLOT_MULT 30.0
+#define SLOT_HEIGHT 25.0
+
+int svg_page_width = 1000;
+
+#define MIN_TEXT_SIZE 0.01
+
+static u64 total_height;
+static FILE *svgfile;
+
+static double cpu2slot(int cpu)
+{
+        return 2 * cpu + 1;
+}
+
+static double cpu2y(int cpu)
+{
+        return cpu2slot(cpu) * SLOT_MULT;
+}
+
+static double time2pixels(u64 time)
+{
+        double X;
+
+        X = 1.0 * svg_page_width * (time - first_time) / (last_time - first_time);
+        return X;
+}
+
+/*
+ * Round text sizes so that the svg viewer only needs a discrete
+ * number of renderings of the font
+ */
+static double round_text_size(double size)
+{
+        int loop = 100;
+        double target = 10.0;
+
+        if (size >= 10.0)
+                return size;
+        while (loop--) {
+                if (size >= target)
+                        return target;
+                target = target / 2.0;
+        }
+        return size;
+}
+
+void open_svg(const char *filename, int cpus, int rows, u64 start, u64 end)
+{
+        int new_width;
+
+        svgfile = fopen(filename, "w");
+        if (!svgfile) {
+                fprintf(stderr, "Cannot open %s for output\n", filename);
+                return;
+        }
+        first_time = start;
+        first_time = first_time / 100000000 * 100000000;
+        last_time = end;
+
+        /*
+         * if the recording is short, we default to a width of 1000, but
+         * for longer recordings we want at least 200 units of width per second
+         */
+        new_width = (last_time - first_time) / 5000000;
+
+        if (new_width > svg_page_width)
+                svg_page_width = new_width;
+
+        total_height = (1 + rows + cpu2slot(cpus)) * SLOT_MULT;
+        fprintf(svgfile, "<?xml version=\"1.0\" standalone=\"no\"?> \n");
+        fprintf(svgfile, "<svg width=\"%i\" height=\"%llu\" version=\"1.1\" xmlns=\"http://www.w3.org/2000/svg\">\n", svg_page_width, total_height);
+
+        fprintf(svgfile, "<defs>\n  <style type=\"text/css\">\n    <![CDATA[\n");
+
+        fprintf(svgfile, "      rect          { stroke-width: 1; }\n");
+        fprintf(svgfile, "      rect.process  { fill:rgb(180,180,180); fill-opacity:0.9; stroke-width:1;   stroke:rgb(  0,  0,  0); } \n");
+        fprintf(svgfile, "      rect.process2 { fill:rgb(180,180,180); fill-opacity:0.9; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+        fprintf(svgfile, "      rect.sample   { fill:rgb(  0,  0,255); fill-opacity:0.8; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+        fprintf(svgfile, "      rect.blocked  { fill:rgb(255,  0,  0); fill-opacity:0.5; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+        fprintf(svgfile, "      rect.waiting  { fill:rgb(224,214,  0); fill-opacity:0.8; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+        fprintf(svgfile, "      rect.WAITING  { fill:rgb(255,214, 48); fill-opacity:0.6; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+        fprintf(svgfile, "      rect.cpu      { fill:rgb(192,192,192); fill-opacity:0.2; stroke-width:0.5; stroke:rgb(128,128,128); } \n");
+        fprintf(svgfile, "      rect.pstate   { fill:rgb(128,128,128); fill-opacity:0.8; stroke-width:0; } \n");
+        fprintf(svgfile, "      rect.c1       { fill:rgb(255,214,214); fill-opacity:0.5; stroke-width:0; } \n");
+        fprintf(svgfile, "      rect.c2       { fill:rgb(255,172,172); fill-opacity:0.5; stroke-width:0; } \n");
+        fprintf(svgfile, "      rect.c3       { fill:rgb(255,130,130); fill-opacity:0.5; stroke-width:0; } \n");
+        fprintf(svgfile, "      rect.c4       { fill:rgb(255, 88, 88); fill-opacity:0.5; stroke-width:0; } \n");
+        fprintf(svgfile, "      rect.c5       { fill:rgb(255, 44, 44); fill-opacity:0.5; stroke-width:0; } \n");
+        fprintf(svgfile, "      rect.c6       { fill:rgb(255,  0,  0); fill-opacity:0.5; stroke-width:0; } \n");
+        fprintf(svgfile, "      line.pstate   { stroke:rgb(255,255,  0); stroke-opacity:0.8; stroke-width:2; } \n");
+
+        fprintf(svgfile, "    ]]>\n   </style>\n</defs>\n");
+}
+
+void svg_box(int Yslot, u64 start, u64 end, const char *type)
+{
+        if (!svgfile)
+                return;
+
+        fprintf(svgfile, "<rect x=\"%4.8f\" width=\"%4.8f\" y=\"%4.1f\" height=\"%4.1f\" class=\"%s\"/>\n",
+                time2pixels(start), time2pixels(end)-time2pixels(start), Yslot * SLOT_MULT, SLOT_HEIGHT, type);
+}
+
+void svg_sample(int Yslot, int cpu, u64 start, u64 end)
+{
+        double text_size;
+        if (!svgfile)
+                return;
+
+        fprintf(svgfile, "<rect x=\"%4.8f\" width=\"%4.8f\" y=\"%4.1f\" height=\"%4.1f\" class=\"sample\"/>\n",
+                time2pixels(start), time2pixels(end)-time2pixels(start), Yslot * SLOT_MULT, SLOT_HEIGHT);
+
+        text_size = (time2pixels(end)-time2pixels(start));
+        if (cpu > 9)
+                text_size = text_size/2;
+        if (text_size > 1.25)
+                text_size = 1.25;
+        text_size = round_text_size(text_size);
+
+        if (text_size > MIN_TEXT_SIZE)
+                fprintf(svgfile, "<text x=\"%1.8f\" y=\"%1.8f\" font-size=\"%1.8fpt\">%i</text>\n",
+                        time2pixels(start), Yslot *  SLOT_MULT + SLOT_HEIGHT - 1, text_size,  cpu + 1);
+
+}
+
+static char *time_to_string(u64 duration)
+{
+        static char text[80];
+
+        text[0] = 0;
+
+        if (duration < 1000) /* less than 1 usec */
+                return text;
+
+        if (duration < 1000 * 1000) { /* less than 1 msec */
+                sprintf(text, "%4.1f us", duration / 1000.0);
+                return text;
+        }
+        sprintf(text, "%4.1f ms", duration / 1000.0 / 1000);
+
+        return text;
+}
+
+void svg_waiting(int Yslot, u64 start, u64 end)
+{
+        char *text;
+        const char *style;
+        double font_size;
+
+        if (!svgfile)
+                return;
+
+        style = "waiting";
+
+        if (end-start > 10 * 1000000) /* 10 msec */
+                style = "WAITING";
+
+        text = time_to_string(end-start);
+
+        font_size = 1.0 * (time2pixels(end)-time2pixels(start));
+
+        if (font_size > 3)
+                font_size = 3;
+
+        font_size = round_text_size(font_size);
+
+        fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\">\n", time2pixels(start), Yslot * SLOT_MULT);
+        fprintf(svgfile, "<rect x=\"0\" width=\"%4.8f\" y=\"0\" height=\"%4.1f\" class=\"%s\"/>\n",
+                time2pixels(end)-time2pixels(start), SLOT_HEIGHT, style);
+        if (font_size > MIN_TEXT_SIZE)
+                fprintf(svgfile, "<text transform=\"rotate(90)\" font-size=\"%1.8fpt\"> %s</text>\n",
+                        font_size, text);
+        fprintf(svgfile, "</g>\n");
+}
+
+static char *cpu_model(void)
+{
+        static char cpu_m[255];
+        char buf[256];
+        FILE *file;
+
+        cpu_m[0] = 0;
+        /* CPU type */
+        file = fopen("/proc/cpuinfo", "r");
+        if (file) {
+                while (fgets(buf, 255, file)) {
+                        if (strstr(buf, "model name")) {
+                                strncpy(cpu_m, &buf[13], 255);
+                                break;
+                        }
+                }
+                fclose(file);
+        }
+
+        /* CPU type */
+        file = fopen("/sys/devices/system/cpu/cpu0/cpufreq/scaling_available_frequencies", "r");
+        if (file) {
+                while (fgets(buf, 255, file)) {
+                        unsigned int freq;
+                        freq = strtoull(buf, NULL, 10);
+                        if (freq > max_freq)
+                                max_freq = freq;
+                }
+                fclose(file);
+        }
+        return cpu_m;
+}
+
+void svg_cpu_box(int cpu, u64 __max_freq, u64 __turbo_freq)
+{
+        char cpu_string[80];
+        if (!svgfile)
+                return;
+
+        max_freq = __max_freq;
+        turbo_frequency = __turbo_freq;
+
+        fprintf(svgfile, "<rect x=\"%4.8f\" width=\"%4.8f\" y=\"%4.1f\" height=\"%4.1f\" class=\"cpu\"/>\n",
+                time2pixels(first_time),
+                time2pixels(last_time)-time2pixels(first_time),
+                cpu2y(cpu), SLOT_MULT+SLOT_HEIGHT);
+
+        sprintf(cpu_string, "CPU %i", (int)cpu+1);
+        fprintf(svgfile, "<text x=\"%4.8f\" y=\"%4.8f\">%s</text>\n",
+                10+time2pixels(first_time), cpu2y(cpu) + SLOT_HEIGHT/2, cpu_string);
+
+        fprintf(svgfile, "<text transform=\"translate(%4.8f,%4.8f)\" font-size=\"1.25pt\">%s</text>\n",
+                10+time2pixels(first_time), cpu2y(cpu) + SLOT_MULT + SLOT_HEIGHT - 4, cpu_model());
+}
+
+void svg_process(int cpu, u64 start, u64 end, const char *type, const char *name)
+{
+        double width;
+
+        if (!svgfile)
+                return;
+
+
+        fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\">\n", time2pixels(start), cpu2y(cpu));
+        fprintf(svgfile, "<rect x=\"0\" width=\"%4.8f\" y=\"0\" height=\"%4.1f\" class=\"%s\"/>\n",
+                time2pixels(end)-time2pixels(start), SLOT_MULT+SLOT_HEIGHT, type);
+        width = time2pixels(end)-time2pixels(start);
+        if (width > 6)
+                width = 6;
+
+        width = round_text_size(width);
+
+        if (width > MIN_TEXT_SIZE)
+                fprintf(svgfile, "<text transform=\"rotate(90)\" font-size=\"%3.8fpt\">%s</text>\n",
+                        width, name);
+
+        fprintf(svgfile, "</g>\n");
+}
+
+void svg_cstate(int cpu, u64 start, u64 end, int type)
+{
+        double width;
+        char style[128];
+
+        if (!svgfile)
+                return;
+
+
+        if (type > 6)
+                type = 6;
+        sprintf(style, "c%i", type);
+
+        fprintf(svgfile, "<rect class=\"%s\" x=\"%4.8f\" width=\"%4.8f\" y=\"%4.1f\" height=\"%4.1f\"/>\n",
+                style,
+                time2pixels(start), time2pixels(end)-time2pixels(start),
+                cpu2y(cpu), SLOT_MULT+SLOT_HEIGHT);
+
+        width = (time2pixels(end)-time2pixels(start))/2.0;
+        if (width > 6)
+                width = 6;
+
+        width = round_text_size(width);
+
+        if (width > MIN_TEXT_SIZE)
+                fprintf(svgfile, "<text x=\"%4.8f\" y=\"%4.8f\" font-size=\"%3.8fpt\">C%i</text>\n",
+                        time2pixels(start), cpu2y(cpu)+width, width, type);
+}
+
+static char *HzToHuman(unsigned long hz)
+{
+        static char buffer[1024];
+        unsigned long long Hz;
+
+        memset(buffer, 0, 1024);
+
+        Hz = hz;
+
+        /* default: just put the Number in */
+        sprintf(buffer, "%9lli", Hz);
+
+        if (Hz > 1000)
+                sprintf(buffer, " %6lli Mhz", (Hz+500)/1000);
+
+        if (Hz > 1500000)
+                sprintf(buffer, " %6.2f Ghz", (Hz+5000.0)/1000000);
+
+        if (Hz == turbo_frequency)
+                sprintf(buffer, "Turbo");
+
+        return buffer;
+}
+
+void svg_pstate(int cpu, u64 start, u64 end, u64 freq)
+{
+        double height = 0;
+
+        if (!svgfile)
+                return;
+
+        if (max_freq)
+                height = freq * 1.0 / max_freq * (SLOT_HEIGHT + SLOT_MULT);
+        height = 1 + cpu2y(cpu) + SLOT_MULT + SLOT_HEIGHT - height;
+        fprintf(svgfile, "<line x1=\"%4.8f\" x2=\"%4.8f\" y1=\"%4.1f\" y2=\"%4.1f\" class=\"pstate\"/>\n",
+                time2pixels(start), time2pixels(end), height, height);
+        fprintf(svgfile, "<text x=\"%4.8f\" y=\"%4.8f\" font-size=\"0.25pt\">%s</text>\n",
+                time2pixels(start), height+0.9, HzToHuman(freq));
+
+}
+
+
+void svg_partial_wakeline(u64 start, int row1, char *desc1, int row2, char *desc2)
+{
+        double height;
+
+        if (!svgfile)
+                return;
+
+
+        if (row1 < row2) {
+                if (row1) {
+                        fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+                                time2pixels(start), row1 * SLOT_MULT + SLOT_HEIGHT,  time2pixels(start), row1 * SLOT_MULT + SLOT_HEIGHT + SLOT_MULT/32);
+                        if (desc2)
+                                fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\"><text transform=\"rotate(90)\" font-size=\"0.02pt\">%s &gt;</text></g>\n",
+                                        time2pixels(start), row1 * SLOT_MULT + SLOT_HEIGHT + SLOT_HEIGHT/48, desc2);
+                }
+                if (row2) {
+                        fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+                                time2pixels(start), row2 * SLOT_MULT - SLOT_MULT/32,  time2pixels(start), row2 * SLOT_MULT);
+                        if (desc1)
+                                fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\"><text transform=\"rotate(90)\" font-size=\"0.02pt\">%s &gt;</text></g>\n",
+                                        time2pixels(start), row2 * SLOT_MULT - SLOT_MULT/32, desc1);
+                }
+        } else {
+                if (row2) {
+                        fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+                                time2pixels(start), row2 * SLOT_MULT + SLOT_HEIGHT,  time2pixels(start), row2 * SLOT_MULT + SLOT_HEIGHT + SLOT_MULT/32);
+                        if (desc1)
+                                fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\"><text transform=\"rotate(90)\" font-size=\"0.02pt\">%s &lt;</text></g>\n",
+                                        time2pixels(start), row2 * SLOT_MULT + SLOT_HEIGHT + SLOT_MULT/48, desc1);
+                }
+                if (row1) {
+                        fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+                                time2pixels(start), row1 * SLOT_MULT - SLOT_MULT/32,  time2pixels(start), row1 * SLOT_MULT);
+                        if (desc2)
+                                fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\"><text transform=\"rotate(90)\" font-size=\"0.02pt\">%s &lt;</text></g>\n",
+                                        time2pixels(start), row1 * SLOT_MULT - SLOT_HEIGHT/32, desc2);
+                }
+        }
+        height = row1 * SLOT_MULT;
+        if (row2 > row1)
+                height += SLOT_HEIGHT;
+        if (row1)
+                fprintf(svgfile, "<circle  cx=\"%4.8f\" cy=\"%4.2f\" r = \"0.01\"  style=\"fill:rgb(32,255,32)\"/>\n",
+                        time2pixels(start), height);
+}
+
+void svg_wakeline(u64 start, int row1, int row2)
+{
+        double height;
+
+        if (!svgfile)
+                return;
+
+
+        if (row1 < row2)
+                fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+                        time2pixels(start), row1 * SLOT_MULT + SLOT_HEIGHT,  time2pixels(start), row2 * SLOT_MULT);
+        else
+                fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+                        time2pixels(start), row2 * SLOT_MULT + SLOT_HEIGHT,  time2pixels(start), row1 * SLOT_MULT);
+
+        height = row1 * SLOT_MULT;
+        if (row2 > row1)
+                height += SLOT_HEIGHT;
+        fprintf(svgfile, "<circle  cx=\"%4.8f\" cy=\"%4.2f\" r = \"0.01\"  style=\"fill:rgb(32,255,32)\"/>\n",
+                        time2pixels(start), height);
+}
+
+void svg_interrupt(u64 start, int row)
+{
+        if (!svgfile)
+                return;
+
+        fprintf(svgfile, "<circle  cx=\"%4.8f\" cy=\"%4.2f\" r = \"0.01\"  style=\"fill:rgb(255,128,128)\"/>\n",
+                        time2pixels(start), row * SLOT_MULT);
+        fprintf(svgfile, "<circle  cx=\"%4.8f\" cy=\"%4.2f\" r = \"0.01\"  style=\"fill:rgb(255,128,128)\"/>\n",
+                        time2pixels(start), row * SLOT_MULT + SLOT_HEIGHT);
+}
+
+void svg_text(int Yslot, u64 start, const char *text)
+{
+        if (!svgfile)
+                return;
+
+        fprintf(svgfile, "<text x=\"%4.8f\" y=\"%4.8f\">%s</text>\n",
+                time2pixels(start), Yslot * SLOT_MULT+SLOT_HEIGHT/2, text);
+}
+
+static void svg_legenda_box(int X, const char *text, const char *style)
+{
+        double boxsize;
+        boxsize = SLOT_HEIGHT / 2;
+
+        fprintf(svgfile, "<rect x=\"%i\" width=\"%4.8f\" y=\"0\" height=\"%4.1f\" class=\"%s\"/>\n",
+                X, boxsize, boxsize, style);
+        fprintf(svgfile, "<text transform=\"translate(%4.8f, %4.8f)\" font-size=\"%4.8fpt\">%s</text>\n",
+                X + boxsize + 5, boxsize, 0.8 * boxsize, text);
+}
+
+void svg_legenda(void)
+{
+        if (!svgfile)
+                return;
+
+        svg_legenda_box(0,      "Running", "sample");
+        svg_legenda_box(100,    "Idle","rect.c1");
+        svg_legenda_box(200,    "Deeper Idle", "rect.c3");
+        svg_legenda_box(350,    "Deepest Idle", "rect.c6");
+        svg_legenda_box(550,    "Sleeping", "process2");
+        svg_legenda_box(650,    "Waiting for cpu", "waiting");
+        svg_legenda_box(800,    "Blocked on IO", "blocked");
+}
+
+void svg_time_grid(void)
+{
+        u64 i;
+
+        if (!svgfile)
+                return;
+
+        i = first_time;
+        while (i < last_time) {
+                int color = 220;
+                double thickness = 0.075;
+                if ((i % 100000000) == 0) {
+                        thickness = 0.5;
+                        color = 192;
+                }
+                if ((i % 1000000000) == 0) {
+                        thickness = 2.0;
+                        color = 128;
+                }
+
+                fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%llu\" style=\"stroke:rgb(%i,%i,%i);stroke-width:%1.3f\"/>\n",
+                        time2pixels(i), SLOT_MULT/2, time2pixels(i), total_height, color, color, color, thickness);
+
+                i += 10000000;
+        }
+}
+
+void svg_close(void)
+{
+        if (svgfile) {
+                fprintf(svgfile, "</svg>\n");
+                fclose(svgfile);
+                svgfile = NULL;
+        }
+}
diff --git a/tools/perf/util/svghelper.h b/tools/perf/util/svghelper.h
new file mode 100644
index 00000000000..e0781989cc3
--- /dev/null
+++ b/tools/perf/util/svghelper.h
@@ -0,0 +1,28 @@
+#ifndef __PERF_SVGHELPER_H
+#define __PERF_SVGHELPER_H
+
+#include "types.h"
+
+extern void open_svg(const char *filename, int cpus, int rows, u64 start, u64 end);
+extern void svg_box(int Yslot, u64 start, u64 end, const char *type);
+extern void svg_sample(int Yslot, int cpu, u64 start, u64 end);
+extern void svg_waiting(int Yslot, u64 start, u64 end);
+extern void svg_cpu_box(int cpu, u64 max_frequency, u64 turbo_frequency);
+
+
+extern void svg_process(int cpu, u64 start, u64 end, const char *type, const char *name);
+extern void svg_cstate(int cpu, u64 start, u64 end, int type);
+extern void svg_pstate(int cpu, u64 start, u64 end, u64 freq);
+
+
+extern void svg_time_grid(void);
+extern void svg_legenda(void);
+extern void svg_wakeline(u64 start, int row1, int row2);
+extern void svg_partial_wakeline(u64 start, int row1, char *desc1, int row2, char *desc2);
+extern void svg_interrupt(u64 start, int row);
+extern void svg_text(int Yslot, u64 start, const char *text);
+extern void svg_close(void);
+
+extern int svg_page_width;
+
+#endif /* __PERF_SVGHELPER_H */
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index 86e14375e74..15ccfba8cdf 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -1,93 +1,257 @@
-#include "util.h"
-#include "../perf.h"
-#include "string.h"
+#define _GNU_SOURCE
+#include <ctype.h>
+#include <dirent.h>
+#include <errno.h>
+#include <libgen.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include <string.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/param.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include "build-id.h"
+#include "debug.h"
 #include "symbol.h"
+#include "strlist.h"
 
 #include <libelf.h>
 #include <gelf.h>
 #include <elf.h>
+#include <limits.h>
+#include <sys/utsname.h>
+
+#ifndef KSYM_NAME_LEN
+#define KSYM_NAME_LEN 128
+#endif
+
+#ifndef NT_GNU_BUILD_ID
+#define NT_GNU_BUILD_ID 3
+#endif
+
+static bool dso__build_id_equal(const struct dso *self, u8 *build_id);
+static int elf_read_build_id(Elf *elf, void *bf, size_t size);
+static void dsos__add(struct list_head *head, struct dso *dso);
+static struct map *map__new2(u64 start, struct dso *dso, enum map_type type);
+static int dso__load_kernel_sym(struct dso *self, struct map *map,
+                                symbol_filter_t filter);
+static int dso__load_guest_kernel_sym(struct dso *self, struct map *map,
+                        symbol_filter_t filter);
+static int vmlinux_path__nr_entries;
+static char **vmlinux_path;
+
+struct symbol_conf symbol_conf = {
+        .exclude_other    = true,
+        .use_modules      = true,
+        .try_vmlinux_path = true,
+        .symfs            = "",
+};
+
+int dso__name_len(const struct dso *self)
+{
+        if (verbose)
+                return self->long_name_len;
 
-const char *sym_hist_filter;
+        return self->short_name_len;
+}
 
-static struct symbol *symbol__new(u64 start, u64 len,
-                                  const char *name, unsigned int priv_size,
-                                  u64 obj_start, int verbose)
+bool dso__loaded(const struct dso *self, enum map_type type)
 {
-        size_t namelen = strlen(name) + 1;
-        struct symbol *self = calloc(1, priv_size + sizeof(*self) + namelen);
+        return self->loaded & (1 << type);
+}
 
-        if (!self)
-                return NULL;
+bool dso__sorted_by_name(const struct dso *self, enum map_type type)
+{
+        return self->sorted_by_name & (1 << type);
+}
+
+static void dso__set_sorted_by_name(struct dso *self, enum map_type type)
+{
+        self->sorted_by_name |= (1 << type);
+}
+
+bool symbol_type__is_a(char symbol_type, enum map_type map_type)
+{
+        switch (map_type) {
+        case MAP__FUNCTION:
+                return symbol_type == 'T' || symbol_type == 'W';
+        case MAP__VARIABLE:
+                return symbol_type == 'D' || symbol_type == 'd';
+        default:
+                return false;
+        }
+}
 
-        if (verbose >= 2)
-                printf("new symbol: %016Lx [%08lx]: %s, hist: %p, obj_start: %p\n",
-                        (u64)start, (unsigned long)len, name, self->hist, (void *)(unsigned long)obj_start);
+static void symbols__fixup_end(struct rb_root *self)
+{
+        struct rb_node *nd, *prevnd = rb_first(self);
+        struct symbol *curr, *prev;
 
-        self->obj_start= obj_start;
-        self->hist = NULL;
-        self->hist_sum = 0;
+        if (prevnd == NULL)
+                return;
 
-        if (sym_hist_filter && !strcmp(name, sym_hist_filter))
-                self->hist = calloc(sizeof(u64), len);
+        curr = rb_entry(prevnd, struct symbol, rb_node);
 
-        if (priv_size) {
-                memset(self, 0, priv_size);
-                self = ((void *)self) + priv_size;
+        for (nd = rb_next(prevnd); nd; nd = rb_next(nd)) {
+                prev = curr;
+                curr = rb_entry(nd, struct symbol, rb_node);
+
+                if (prev->end == prev->start && prev->end != curr->start)
+                        prev->end = curr->start - 1;
         }
-        self->start = start;
-        self->end   = start + len - 1;
+
+        /* Last entry */
+        if (curr->end == curr->start)
+                curr->end = roundup(curr->start, 4096);
+}
+
+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]);
+
+        if (prevnd == NULL)
+                return;
+
+        curr = rb_entry(prevnd, struct map, rb_node);
+
+        for (nd = rb_next(prevnd); nd; nd = rb_next(nd)) {
+                prev = curr;
+                curr = rb_entry(nd, struct map, rb_node);
+                prev->end = curr->start - 1;
+        }
+
+        /*
+         * We still haven't the actual symbols, so guess the
+         * last map final address.
+         */
+        curr->end = ~0ULL;
+}
+
+static void map_groups__fixup_end(struct map_groups *self)
+{
+        int i;
+        for (i = 0; i < MAP__NR_TYPES; ++i)
+                __map_groups__fixup_end(self, i);
+}
+
+static struct symbol *symbol__new(u64 start, u64 len, u8 binding,
+                                  const char *name)
+{
+        size_t namelen = strlen(name) + 1;
+        struct symbol *self = calloc(1, (symbol_conf.priv_size +
+                                         sizeof(*self) + namelen));
+        if (self == NULL)
+                return NULL;
+
+        if (symbol_conf.priv_size)
+                self = ((void *)self) + symbol_conf.priv_size;
+
+        self->start   = start;
+        self->end     = len ? start + len - 1 : start;
+        self->binding = binding;
+        self->namelen = namelen - 1;
+
+        pr_debug4("%s: %s %#Lx-%#Lx\n", __func__, name, start, self->end);
+
         memcpy(self->name, name, namelen);
 
         return self;
 }
 
-static void symbol__delete(struct symbol *self, unsigned int priv_size)
+void symbol__delete(struct symbol *self)
 {
-        free(((void *)self) - priv_size);
+        free(((void *)self) - symbol_conf.priv_size);
 }
 
 static size_t symbol__fprintf(struct symbol *self, FILE *fp)
 {
-        return fprintf(fp, " %llx-%llx %s\n",
-                       self->start, self->end, self->name);
+        return fprintf(fp, " %llx-%llx %c %s\n",
+                       self->start, self->end,
+                       self->binding == STB_GLOBAL ? 'g' :
+                       self->binding == STB_LOCAL  ? 'l' : 'w',
+                       self->name);
 }
 
-struct dso *dso__new(const char *name, unsigned int sym_priv_size)
+void dso__set_long_name(struct dso *self, char *name)
 {
-        struct dso *self = malloc(sizeof(*self) + strlen(name) + 1);
+        if (name == NULL)
+                return;
+        self->long_name = name;
+        self->long_name_len = strlen(name);
+}
+
+static void dso__set_short_name(struct dso *self, const char *name)
+{
+        if (name == NULL)
+                return;
+        self->short_name = name;
+        self->short_name_len = strlen(name);
+}
+
+static void dso__set_basename(struct dso *self)
+{
+        dso__set_short_name(self, basename(self->long_name));
+}
+
+struct dso *dso__new(const char *name)
+{
+        struct dso *self = calloc(1, sizeof(*self) + strlen(name) + 1);
 
         if (self != NULL) {
+                int i;
                 strcpy(self->name, name);
-                self->syms = RB_ROOT;
-                self->sym_priv_size = sym_priv_size;
-                self->find_symbol = dso__find_symbol;
+                dso__set_long_name(self, self->name);
+                dso__set_short_name(self, self->name);
+                for (i = 0; i < MAP__NR_TYPES; ++i)
+                        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;
+                self->kernel = DSO_TYPE_USER;
+                INIT_LIST_HEAD(&self->node);
         }
 
         return self;
 }
 
-static void dso__delete_symbols(struct dso *self)
+static void symbols__delete(struct rb_root *self)
 {
         struct symbol *pos;
-        struct rb_node *next = rb_first(&self->syms);
+        struct rb_node *next = rb_first(self);
 
         while (next) {
                 pos = rb_entry(next, struct symbol, rb_node);
                 next = rb_next(&pos->rb_node);
-                rb_erase(&pos->rb_node, &self->syms);
-                symbol__delete(pos, self->sym_priv_size);
+                rb_erase(&pos->rb_node, self);
+                symbol__delete(pos);
         }
 }
 
 void dso__delete(struct dso *self)
 {
-        dso__delete_symbols(self);
+        int i;
+        for (i = 0; i < MAP__NR_TYPES; ++i)
+                symbols__delete(&self->symbols[i]);
+        if (self->sname_alloc)
+                free((char *)self->short_name);
+        if (self->lname_alloc)
+                free(self->long_name);
         free(self);
 }
 
-static void dso__insert_symbol(struct dso *self, struct symbol *sym)
+void dso__set_build_id(struct dso *self, void *build_id)
 {
-        struct rb_node **p = &self->syms.rb_node;
+        memcpy(self->build_id, build_id, sizeof(self->build_id));
+        self->has_build_id = 1;
+}
+
+static void symbols__insert(struct rb_root *self, struct symbol *sym)
+{
+        struct rb_node **p = &self->rb_node;
         struct rb_node *parent = NULL;
         const u64 ip = sym->start;
         struct symbol *s;
@@ -101,17 +265,17 @@ static void dso__insert_symbol(struct dso *self, struct symbol *sym)
                         p = &(*p)->rb_right;
         }
         rb_link_node(&sym->rb_node, parent, p);
-        rb_insert_color(&sym->rb_node, &self->syms);
+        rb_insert_color(&sym->rb_node, self);
 }
 
-struct symbol *dso__find_symbol(struct dso *self, u64 ip)
+static struct symbol *symbols__find(struct rb_root *self, u64 ip)
 {
         struct rb_node *n;
 
         if (self == NULL)
                 return NULL;
 
-        n = self->syms.rb_node;
+        n = self->rb_node;
 
         while (n) {
                 struct symbol *s = rb_entry(n, struct symbol, rb_node);
@@ -127,12 +291,136 @@ struct symbol *dso__find_symbol(struct dso *self, u64 ip)
         return NULL;
 }
 
-size_t dso__fprintf(struct dso *self, FILE *fp)
+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)
 {
-        size_t ret = fprintf(fp, "dso: %s\n", self->name);
+        struct rb_node **p = &self->rb_node;
+        struct rb_node *parent = NULL;
+        struct symbol_name_rb_node *symn, *s;
+
+        symn = container_of(sym, struct symbol_name_rb_node, sym);
 
+        while (*p != NULL) {
+                parent = *p;
+                s = rb_entry(parent, struct symbol_name_rb_node, rb_node);
+                if (strcmp(sym->name, s->sym.name) < 0)
+                        p = &(*p)->rb_left;
+                else
+                        p = &(*p)->rb_right;
+        }
+        rb_link_node(&symn->rb_node, parent, p);
+        rb_insert_color(&symn->rb_node, self);
+}
+
+static void symbols__sort_by_name(struct rb_root *self, struct rb_root *source)
+{
         struct rb_node *nd;
-        for (nd = rb_first(&self->syms); nd; nd = rb_next(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(const u8 *self, int len, char *bf)
+{
+        char *bid = bf;
+        const u8 *raw = self;
+        int i;
+
+        for (i = 0; i < len; ++i) {
+                sprintf(bid, "%02x", *raw);
+                ++raw;
+                bid += 2;
+        }
+
+        return raw - self;
+}
+
+size_t dso__fprintf_buildid(struct dso *self, FILE *fp)
+{
+        char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+        build_id__sprintf(self->build_id, sizeof(self->build_id), sbuild_id);
+        return fprintf(fp, "%s", sbuild_id);
+}
+
+size_t dso__fprintf_symbols_by_name(struct dso *self, enum map_type type, FILE *fp)
+{
+        size_t ret = 0;
+        struct rb_node *nd;
+        struct symbol_name_rb_node *pos;
+
+        for (nd = rb_first(&self->symbol_names[type]); nd; nd = rb_next(nd)) {
+                pos = rb_entry(nd, struct symbol_name_rb_node, rb_node);
+                fprintf(fp, "%s\n", pos->sym.name);
+        }
+
+        return ret;
+}
+
+size_t dso__fprintf(struct dso *self, enum map_type type, FILE *fp)
+{
+        struct rb_node *nd;
+        size_t ret = fprintf(fp, "dso: %s (", self->short_name);
+
+        if (self->short_name != self->long_name)
+                ret += fprintf(fp, "%s, ", self->long_name);
+        ret += fprintf(fp, "%s, %sloaded, ", map_type__name[type],
+                       self->loaded ? "" : "NOT ");
+        ret += dso__fprintf_buildid(self, fp);
+        ret += fprintf(fp, ")\n");
+        for (nd = rb_first(&self->symbols[type]); nd; nd = rb_next(nd)) {
                 struct symbol *pos = rb_entry(nd, struct symbol, rb_node);
                 ret += symbol__fprintf(pos, fp);
         }
@@ -140,29 +428,37 @@ size_t dso__fprintf(struct dso *self, FILE *fp)
         return ret;
 }
 
-static int dso__load_kallsyms(struct dso *self, symbol_filter_t filter, int verbose)
+int kallsyms__parse(const char *filename, void *arg,
+                    int (*process_symbol)(void *arg, const char *name,
+                                          char type, u64 start, u64 end))
 {
-        struct rb_node *nd, *prevnd;
         char *line = NULL;
         size_t n;
-        FILE *file = fopen("/proc/kallsyms", "r");
+        int err = -1;
+        u64 prev_start = 0;
+        char prev_symbol_type = 0;
+        char *prev_symbol_name;
+        FILE *file = fopen(filename, "r");
 
         if (file == NULL)
                 goto out_failure;
 
+        prev_symbol_name = malloc(KSYM_NAME_LEN);
+        if (prev_symbol_name == NULL)
+                goto out_close;
+
+        err = 0;
+
         while (!feof(file)) {
                 u64 start;
-                struct symbol *sym;
                 int line_len, len;
                 char symbol_type;
+                char *symbol_name;
 
                 line_len = getline(&line, &n, file);
-                if (line_len < 0)
+                if (line_len < 0 || !line)
                         break;
 
-                if (!line)
-                        goto out_failure;
-
                 line[--line_len] = '\0'; /* \n */
 
                 len = hex2u64(line, &start);
@@ -172,62 +468,234 @@ static int dso__load_kallsyms(struct dso *self, symbol_filter_t filter, int verb
                         continue;
 
                 symbol_type = toupper(line[len]);
-                /*
-                 * We're interested only in code ('T'ext)
-                 */
-                if (symbol_type != 'T' && symbol_type != 'W')
-                        continue;
-                /*
-                 * Well fix up the end later, when we have all sorted.
-                 */
-                sym = symbol__new(start, 0xdead, line + len + 2,
-                                  self->sym_priv_size, 0, verbose);
+                len += 2;
+                symbol_name = line + len;
+                len = line_len - len;
 
-                if (sym == NULL)
-                        goto out_delete_line;
+                if (len >= KSYM_NAME_LEN) {
+                        err = -1;
+                        break;
+                }
 
-                if (filter && filter(self, sym))
-                        symbol__delete(sym, self->sym_priv_size);
-                else
-                        dso__insert_symbol(self, sym);
+                if (prev_symbol_type) {
+                        u64 end = start;
+                        if (end != prev_start)
+                                --end;
+                        err = process_symbol(arg, prev_symbol_name,
+                                             prev_symbol_type, prev_start, end);
+                        if (err)
+                                break;
+                }
+
+                memcpy(prev_symbol_name, symbol_name, len + 1);
+                prev_symbol_type = symbol_type;
+                prev_start = start;
         }
 
+        free(prev_symbol_name);
+        free(line);
+out_close:
+        fclose(file);
+        return err;
+
+out_failure:
+        return -1;
+}
+
+struct process_kallsyms_args {
+        struct map *map;
+        struct dso *dso;
+};
+
+static u8 kallsyms2elf_type(char type)
+{
+        if (type == 'W')
+                return STB_WEAK;
+
+        return isupper(type) ? STB_GLOBAL : STB_LOCAL;
+}
+
+static int map__process_kallsym_symbol(void *arg, const char *name,
+                                       char type, u64 start, u64 end)
+{
+        struct symbol *sym;
+        struct process_kallsyms_args *a = arg;
+        struct rb_root *root = &a->dso->symbols[a->map->type];
+
+        if (!symbol_type__is_a(type, a->map->type))
+                return 0;
+
+        sym = symbol__new(start, end - start + 1,
+                          kallsyms2elf_type(type), name);
+        if (sym == NULL)
+                return -ENOMEM;
         /*
-         * Now that we have all sorted out, just set the ->end of all
-         * symbols
+         * We will pass the symbols to the filter later, in
+         * map__split_kallsyms, when we have split the maps per module
          */
-        prevnd = rb_first(&self->syms);
+        symbols__insert(root, sym);
 
-        if (prevnd == NULL)
-                goto out_delete_line;
+        return 0;
+}
 
-        for (nd = rb_next(prevnd); nd; nd = rb_next(nd)) {
-                struct symbol *prev = rb_entry(prevnd, struct symbol, rb_node),
-                              *curr = rb_entry(nd, struct symbol, rb_node);
+/*
+ * Loads the function entries in /proc/kallsyms into kernel_map->dso,
+ * so that we can in the next step set the symbol ->end address and then
+ * call kernel_maps__split_kallsyms.
+ */
+static int dso__load_all_kallsyms(struct dso *self, const char *filename,
+                                  struct map *map)
+{
+        struct process_kallsyms_args args = { .map = map, .dso = self, };
+        return kallsyms__parse(filename, &args, map__process_kallsym_symbol);
+}
 
-                prev->end = curr->start - 1;
-                prevnd = nd;
+/*
+ * Split the symbols into maps, making sure there are no overlaps, i.e. the
+ * kernel range is broken in several maps, named [kernel].N, as we don't have
+ * the original ELF section names vmlinux have.
+ */
+static int dso__split_kallsyms(struct dso *self, struct map *map,
+                               symbol_filter_t filter)
+{
+        struct map_groups *kmaps = map__kmap(map)->kmaps;
+        struct machine *machine = kmaps->machine;
+        struct map *curr_map = map;
+        struct symbol *pos;
+        int count = 0, moved = 0;       
+        struct rb_root *root = &self->symbols[map->type];
+        struct rb_node *next = rb_first(root);
+        int kernel_range = 0;
+
+        while (next) {
+                char *module;
+
+                pos = rb_entry(next, struct symbol, rb_node);
+                next = rb_next(&pos->rb_node);
+
+                module = strchr(pos->name, '\t');
+                if (module) {
+                        if (!symbol_conf.use_modules)
+                                goto discard_symbol;
+
+                        *module++ = '\0';
+
+                        if (strcmp(curr_map->dso->short_name, module)) {
+                                if (curr_map != map &&
+                                    self->kernel == DSO_TYPE_GUEST_KERNEL &&
+                                    machine__is_default_guest(machine)) {
+                                        /*
+                                         * We assume all symbols of a module are
+                                         * continuous in * kallsyms, so curr_map
+                                         * points to a module and all its
+                                         * symbols are in its kmap. Mark it as
+                                         * loaded.
+                                         */
+                                        dso__set_loaded(curr_map->dso,
+                                                        curr_map->type);
+                                }
+
+                                curr_map = map_groups__find_by_name(kmaps,
+                                                        map->type, module);
+                                if (curr_map == NULL) {
+                                        pr_debug("%s/proc/{kallsyms,modules} "
+                                                 "inconsistency while looking "
+                                                 "for \"%s\" module!\n",
+                                                 machine->root_dir, module);
+                                        curr_map = map;
+                                        goto discard_symbol;
+                                }
+
+                                if (curr_map->dso->loaded &&
+                                    !machine__is_default_guest(machine))
+                                        goto discard_symbol;
+                        }
+                        /*
+                         * So that we look just like we get from .ko files,
+                         * i.e. not prelinked, relative to map->start.
+                         */
+                        pos->start = curr_map->map_ip(curr_map, pos->start);
+                        pos->end   = curr_map->map_ip(curr_map, pos->end);
+                } else if (curr_map != map) {
+                        char dso_name[PATH_MAX];
+                        struct dso *dso;
+
+                        if (count == 0) {
+                                curr_map = map;
+                                goto filter_symbol;
+                        }
+
+                        if (self->kernel == DSO_TYPE_GUEST_KERNEL)
+                                snprintf(dso_name, sizeof(dso_name),
+                                        "[guest.kernel].%d",
+                                        kernel_range++);
+                        else
+                                snprintf(dso_name, sizeof(dso_name),
+                                        "[kernel].%d",
+                                        kernel_range++);
+
+                        dso = dso__new(dso_name);
+                        if (dso == NULL)
+                                return -1;
+
+                        dso->kernel = self->kernel;
+
+                        curr_map = map__new2(pos->start, dso, map->type);
+                        if (curr_map == NULL) {
+                                dso__delete(dso);
+                                return -1;
+                        }
+
+                        curr_map->map_ip = curr_map->unmap_ip = identity__map_ip;
+                        map_groups__insert(kmaps, curr_map);
+                        ++kernel_range;
+                }
+filter_symbol:
+                if (filter && filter(curr_map, pos)) {
+discard_symbol:         rb_erase(&pos->rb_node, root);
+                        symbol__delete(pos);
+                } else {
+                        if (curr_map != map) {
+                                rb_erase(&pos->rb_node, root);
+                                symbols__insert(&curr_map->dso->symbols[curr_map->type], pos);
+                                ++moved;
+                        } else
+                                ++count;
+                }
         }
 
-        free(line);
-        fclose(file);
+        if (curr_map != map &&
+            self->kernel == DSO_TYPE_GUEST_KERNEL &&
+            machine__is_default_guest(kmaps->machine)) {
+                dso__set_loaded(curr_map->dso, curr_map->type);
+        }
 
-        return 0;
+        return count + moved;
+}
 
-out_delete_line:
-        free(line);
-out_failure:
-        return -1;
+int dso__load_kallsyms(struct dso *self, const char *filename,
+                       struct map *map, symbol_filter_t filter)
+{
+        if (dso__load_all_kallsyms(self, filename, map) < 0)
+                return -1;
+
+        if (self->kernel == DSO_TYPE_GUEST_KERNEL)
+                self->origin = DSO__ORIG_GUEST_KERNEL;
+        else
+                self->origin = DSO__ORIG_KERNEL;
+
+        return dso__split_kallsyms(self, map, filter);
 }
 
-static int dso__load_perf_map(struct dso *self, symbol_filter_t filter, int verbose)
+static int dso__load_perf_map(struct dso *self, struct map *map,
+                              symbol_filter_t filter)
 {
         char *line = NULL;
         size_t n;
         FILE *file;
         int nr_syms = 0;
 
-        file = fopen(self->name, "r");
+        file = fopen(self->long_name, "r");
         if (file == NULL)
                 goto out_failure;
 
@@ -257,16 +725,15 @@ static int dso__load_perf_map(struct dso *self, symbol_filter_t filter, int verb
                 if (len + 2 >= line_len)
                         continue;
 
-                sym = symbol__new(start, size, line + len,
-                                  self->sym_priv_size, start, verbose);
+                sym = symbol__new(start, size, STB_GLOBAL, line + len);
 
                 if (sym == NULL)
                         goto out_delete_line;
 
-                if (filter && filter(self, sym))
-                        symbol__delete(sym, self->sym_priv_size);
+                if (filter && filter(map, sym))
+                        symbol__delete(sym);
                 else {
-                        dso__insert_symbol(self, sym);
+                        symbols__insert(&self->symbols[map->type], sym);
                         nr_syms++;
                 }
         }
@@ -286,13 +753,13 @@ out_failure:
  * elf_symtab__for_each_symbol - iterate thru all the symbols
  *
  * @self: struct elf_symtab instance to iterate
- * @index: uint32_t index
+ * @idx: uint32_t idx
  * @sym: GElf_Sym iterator
  */
-#define elf_symtab__for_each_symbol(syms, nr_syms, index, sym) \
-        for (index = 0, gelf_getsym(syms, index, &sym);\
-             index < nr_syms; \
-             index++, gelf_getsym(syms, index, &sym))
+#define elf_symtab__for_each_symbol(syms, nr_syms, idx, sym) \
+        for (idx = 0, gelf_getsym(syms, idx, &sym);\
+             idx < nr_syms; \
+             idx++, gelf_getsym(syms, idx, &sym))
 
 static inline uint8_t elf_sym__type(const GElf_Sym *sym)
 {
@@ -303,8 +770,40 @@ static inline int elf_sym__is_function(const GElf_Sym *sym)
 {
         return elf_sym__type(sym) == STT_FUNC &&
                sym->st_name != 0 &&
-               sym->st_shndx != SHN_UNDEF &&
-               sym->st_size != 0;
+               sym->st_shndx != SHN_UNDEF;
+}
+
+static inline bool elf_sym__is_object(const GElf_Sym *sym)
+{
+        return elf_sym__type(sym) == STT_OBJECT &&
+                sym->st_name != 0 &&
+                sym->st_shndx != SHN_UNDEF;
+}
+
+static inline int elf_sym__is_label(const GElf_Sym *sym)
+{
+        return elf_sym__type(sym) == STT_NOTYPE &&
+                sym->st_name != 0 &&
+                sym->st_shndx != SHN_UNDEF &&
+                sym->st_shndx != SHN_ABS;
+}
+
+static inline const char *elf_sec__name(const GElf_Shdr *shdr,
+                                        const Elf_Data *secstrs)
+{
+        return secstrs->d_buf + shdr->sh_name;
+}
+
+static inline int elf_sec__is_text(const GElf_Shdr *shdr,
+                                        const Elf_Data *secstrs)
+{
+        return strstr(elf_sec__name(shdr, secstrs), "text") != NULL;
+}
+
+static inline bool elf_sec__is_data(const GElf_Shdr *shdr,
+                                    const Elf_Data *secstrs)
+{
+        return strstr(elf_sec__name(shdr, secstrs), "data") != NULL;
 }
 
 static inline const char *elf_sym__name(const GElf_Sym *sym,
@@ -315,7 +814,7 @@ static inline const char *elf_sym__name(const GElf_Sym *sym,
 
 static Elf_Scn *elf_section_by_name(Elf *elf, GElf_Ehdr *ep,
                                     GElf_Shdr *shp, const char *name,
-                                    size_t *index)
+                                    size_t *idx)
 {
         Elf_Scn *sec = NULL;
         size_t cnt = 1;
@@ -326,8 +825,8 @@ static Elf_Scn *elf_section_by_name(Elf *elf, GElf_Ehdr *ep,
                 gelf_getshdr(sec, shp);
                 str = elf_strptr(elf, ep->e_shstrndx, shp->sh_name);
                 if (!strcmp(name, str)) {
-                        if (index)
-                                *index = cnt;
+                        if (idx)
+                                *idx = cnt;
                         break;
                 }
                 ++cnt;
@@ -346,56 +845,85 @@ static Elf_Scn *elf_section_by_name(Elf *elf, GElf_Ehdr *ep,
              idx < nr_entries; \
              ++idx, pos = gelf_getrela(reldata, idx, &pos_mem))
 
-static int dso__synthesize_plt_symbols(struct  dso *self, Elf *elf,
-                                       GElf_Ehdr *ehdr, Elf_Scn *scn_dynsym,
-                                       GElf_Shdr *shdr_dynsym,
-                                       size_t dynsym_idx, int verbose)
+/*
+ * We need to check if we have a .dynsym, so that we can handle the
+ * .plt, synthesizing its symbols, that aren't on the symtabs (be it
+ * .dynsym or .symtab).
+ * And always look at the original dso, not at debuginfo packages, that
+ * have the PLT data stripped out (shdr_rel_plt.sh_type == SHT_NOBITS).
+ */
+static int dso__synthesize_plt_symbols(struct  dso *self, struct map *map,
+                                       symbol_filter_t filter)
 {
         uint32_t nr_rel_entries, idx;
         GElf_Sym sym;
         u64 plt_offset;
         GElf_Shdr shdr_plt;
         struct symbol *f;
-        GElf_Shdr shdr_rel_plt;
+        GElf_Shdr shdr_rel_plt, shdr_dynsym;
         Elf_Data *reldata, *syms, *symstrs;
-        Elf_Scn *scn_plt_rel, *scn_symstrs;
+        Elf_Scn *scn_plt_rel, *scn_symstrs, *scn_dynsym;
+        size_t dynsym_idx;
+        GElf_Ehdr ehdr;
         char sympltname[1024];
-        int nr = 0, symidx;
+        Elf *elf;
+        int nr = 0, symidx, fd, err = 0;
+        char name[PATH_MAX];
+
+        snprintf(name, sizeof(name), "%s%s",
+                 symbol_conf.symfs, self->long_name);
+        fd = open(name, O_RDONLY);
+        if (fd < 0)
+                goto out;
+
+        elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL);
+        if (elf == NULL)
+                goto out_close;
+
+        if (gelf_getehdr(elf, &ehdr) == NULL)
+                goto out_elf_end;
 
-        scn_plt_rel = elf_section_by_name(elf, ehdr, &shdr_rel_plt,
+        scn_dynsym = elf_section_by_name(elf, &ehdr, &shdr_dynsym,
+                                         ".dynsym", &dynsym_idx);
+        if (scn_dynsym == NULL)
+                goto out_elf_end;
+
+        scn_plt_rel = elf_section_by_name(elf, &ehdr, &shdr_rel_plt,
                                           ".rela.plt", NULL);
         if (scn_plt_rel == NULL) {
-                scn_plt_rel = elf_section_by_name(elf, ehdr, &shdr_rel_plt,
+                scn_plt_rel = elf_section_by_name(elf, &ehdr, &shdr_rel_plt,
                                                   ".rel.plt", NULL);
                 if (scn_plt_rel == NULL)
-                        return 0;
+                        goto out_elf_end;
         }
 
+        err = -1;
+
         if (shdr_rel_plt.sh_link != dynsym_idx)
-                return 0;
+                goto out_elf_end;
 
-        if (elf_section_by_name(elf, ehdr, &shdr_plt, ".plt", NULL) == NULL)
-                return 0;
+        if (elf_section_by_name(elf, &ehdr, &shdr_plt, ".plt", NULL) == NULL)
+                goto out_elf_end;
 
         /*
-         * Fetch the relocation section to find the indexes to the GOT
+         * Fetch the relocation section to find the idxes to the GOT
          * and the symbols in the .dynsym they refer to.
          */
         reldata = elf_getdata(scn_plt_rel, NULL);
         if (reldata == NULL)
-                return -1;
+                goto out_elf_end;
 
         syms = elf_getdata(scn_dynsym, NULL);
         if (syms == NULL)
-                return -1;
+                goto out_elf_end;
 
-        scn_symstrs = elf_getscn(elf, shdr_dynsym->sh_link);
+        scn_symstrs = elf_getscn(elf, shdr_dynsym.sh_link);
         if (scn_symstrs == NULL)
-                return -1;
+                goto out_elf_end;
 
         symstrs = elf_getdata(scn_symstrs, NULL);
         if (symstrs == NULL)
-                return -1;
+                goto out_elf_end;
 
         nr_rel_entries = shdr_rel_plt.sh_size / shdr_rel_plt.sh_entsize;
         plt_offset = shdr_plt.sh_offset;
@@ -412,12 +940,16 @@ static int dso__synthesize_plt_symbols(struct  dso *self, Elf *elf,
                                  "%s@plt", elf_sym__name(&sym, symstrs));
 
                         f = symbol__new(plt_offset, shdr_plt.sh_entsize,
-                                        sympltname, self->sym_priv_size, 0, verbose);
+                                        STB_GLOBAL, sympltname);
                         if (!f)
-                                return -1;
-
-                        dso__insert_symbol(self, f);
-                        ++nr;
+                                goto out_elf_end;
+
+                        if (filter && filter(map, f))
+                                symbol__delete(f);
+                        else {
+                                symbols__insert(&self->symbols[map->type], f);
+                                ++nr;
+                        }
                 }
         } else if (shdr_rel_plt.sh_type == SHT_REL) {
                 GElf_Rel pos_mem, *pos;
@@ -430,81 +962,133 @@ static int dso__synthesize_plt_symbols(struct  dso *self, Elf *elf,
                                  "%s@plt", elf_sym__name(&sym, symstrs));
 
                         f = symbol__new(plt_offset, shdr_plt.sh_entsize,
-                                        sympltname, self->sym_priv_size, 0, verbose);
+                                        STB_GLOBAL, sympltname);
                         if (!f)
-                                return -1;
-
-                        dso__insert_symbol(self, f);
-                        ++nr;
+                                goto out_elf_end;
+
+                        if (filter && filter(map, f))
+                                symbol__delete(f);
+                        else {
+                                symbols__insert(&self->symbols[map->type], f);
+                                ++nr;
+                        }
                 }
-        } else {
-                /*
-                 * TODO: There are still one more shdr_rel_plt.sh_type
-                 * I have to investigate, but probably should be ignored.
-                 */
         }
 
-        return nr;
+        err = 0;
+out_elf_end:
+        elf_end(elf);
+out_close:
+        close(fd);
+
+        if (err == 0)
+                return nr;
+out:
+        pr_debug("%s: problems reading %s PLT info.\n",
+                 __func__, self->long_name);
+        return 0;
 }
 
-static int dso__load_sym(struct dso *self, int fd, const char *name,
-                         symbol_filter_t filter, int verbose)
+static bool elf_sym__is_a(GElf_Sym *self, enum map_type type)
 {
-        Elf_Data *symstrs;
+        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 size_t elf_addr_to_index(Elf *elf, GElf_Addr addr)
+{
+        Elf_Scn *sec = NULL;
+        GElf_Shdr shdr;
+        size_t cnt = 1;
+
+        while ((sec = elf_nextscn(elf, sec)) != NULL) {
+                gelf_getshdr(sec, &shdr);
+
+                if ((addr >= shdr.sh_addr) &&
+                    (addr < (shdr.sh_addr + shdr.sh_size)))
+                        return cnt;
+
+                ++cnt;
+        }
+
+        return -1;
+}
+
+static int dso__load_sym(struct dso *self, struct map *map, const char *name,
+                         int fd, symbol_filter_t filter, int kmodule,
+                         int want_symtab)
+{
+        struct kmap *kmap = self->kernel ? map__kmap(map) : NULL;
+        struct map *curr_map = map;
+        struct dso *curr_dso = self;
+        Elf_Data *symstrs, *secstrs;
         uint32_t nr_syms;
         int err = -1;
-        uint32_t index;
+        uint32_t idx;
         GElf_Ehdr ehdr;
-        GElf_Shdr shdr;
-        Elf_Data *syms;
+        GElf_Shdr shdr, opdshdr;
+        Elf_Data *syms, *opddata = NULL;
         GElf_Sym sym;
-        Elf_Scn *sec, *sec_dynsym;
+        Elf_Scn *sec, *sec_strndx, *opdsec;
         Elf *elf;
-        size_t dynsym_idx;
         int nr = 0;
+        size_t opdidx = 0;
 
-        elf = elf_begin(fd, ELF_C_READ_MMAP, NULL);
+        elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL);
         if (elf == NULL) {
-                if (verbose)
-                        fprintf(stderr, "%s: cannot read %s ELF file.\n",
-                                __func__, name);
+                pr_debug("%s: cannot read %s ELF file.\n", __func__, name);
                 goto out_close;
         }
 
         if (gelf_getehdr(elf, &ehdr) == NULL) {
-                if (verbose)
-                        fprintf(stderr, "%s: cannot get elf header.\n", __func__);
+                pr_debug("%s: cannot get elf header.\n", __func__);
                 goto out_elf_end;
         }
 
-        /*
-         * We need to check if we have a .dynsym, so that we can handle the
-         * .plt, synthesizing its symbols, that aren't on the symtabs (be it
-         * .dynsym or .symtab)
-         */
-        sec_dynsym = elf_section_by_name(elf, &ehdr, &shdr,
-                                         ".dynsym", &dynsym_idx);
-        if (sec_dynsym != NULL) {
-                nr = dso__synthesize_plt_symbols(self, elf, &ehdr,
-                                                 sec_dynsym, &shdr,
-                                                 dynsym_idx, verbose);
-                if (nr < 0)
+        /* Always reject images with a mismatched build-id: */
+        if (self->has_build_id) {
+                u8 build_id[BUILD_ID_SIZE];
+
+                if (elf_read_build_id(elf, build_id,
+                                      BUILD_ID_SIZE) != BUILD_ID_SIZE)
+                        goto out_elf_end;
+
+                if (!dso__build_id_equal(self, build_id))
                         goto out_elf_end;
         }
 
-        /*
-         * But if we have a full .symtab (that is a superset of .dynsym) we
-         * should add the symbols not in the .dynsyn
-         */
         sec = elf_section_by_name(elf, &ehdr, &shdr, ".symtab", NULL);
         if (sec == NULL) {
-                if (sec_dynsym == NULL)
+                if (want_symtab)
                         goto out_elf_end;
 
-                sec = sec_dynsym;
-                gelf_getshdr(sec, &shdr);
+                sec = elf_section_by_name(elf, &ehdr, &shdr, ".dynsym", NULL);
+                if (sec == NULL)
+                        goto out_elf_end;
         }
 
+        opdsec = elf_section_by_name(elf, &ehdr, &opdshdr, ".opd", &opdidx);
+        if (opdsec)
+                opddata = elf_rawdata(opdsec, NULL);
+
         syms = elf_getdata(sec, NULL);
         if (syms == NULL)
                 goto out_elf_end;
@@ -517,40 +1101,155 @@ static int dso__load_sym(struct dso *self, int fd, const char *name,
         if (symstrs == NULL)
                 goto out_elf_end;
 
+        sec_strndx = elf_getscn(elf, ehdr.e_shstrndx);
+        if (sec_strndx == NULL)
+                goto out_elf_end;
+
+        secstrs = elf_getdata(sec_strndx, NULL);
+        if (secstrs == NULL)
+                goto out_elf_end;
+
         nr_syms = shdr.sh_size / shdr.sh_entsize;
 
         memset(&sym, 0, sizeof(sym));
-
-        elf_symtab__for_each_symbol(syms, nr_syms, index, sym) {
+        if (self->kernel == DSO_TYPE_USER) {
+                self->adjust_symbols = (ehdr.e_type == ET_EXEC ||
+                                elf_section_by_name(elf, &ehdr, &shdr,
+                                                     ".gnu.prelink_undo",
+                                                     NULL) != NULL);
+        } else self->adjust_symbols = 0;
+
+        elf_symtab__for_each_symbol(syms, nr_syms, idx, sym) {
                 struct symbol *f;
-                u64 obj_start;
+                const char *elf_name = elf_sym__name(&sym, symstrs);
+                char *demangled = NULL;
+                int is_label = elf_sym__is_label(&sym);
+                const char *section_name;
 
-                if (!elf_sym__is_function(&sym))
+                if (kmap && kmap->ref_reloc_sym && kmap->ref_reloc_sym->name &&
+                    strcmp(elf_name, kmap->ref_reloc_sym->name) == 0)
+                        kmap->ref_reloc_sym->unrelocated_addr = sym.st_value;
+
+                if (!is_label && !elf_sym__is_a(&sym, map->type))
                         continue;
 
+                /* Reject ARM ELF "mapping symbols": these aren't unique and
+                 * don't identify functions, so will confuse the profile
+                 * output: */
+                if (ehdr.e_machine == EM_ARM) {
+                        if (!strcmp(elf_name, "$a") ||
+                            !strcmp(elf_name, "$d") ||
+                            !strcmp(elf_name, "$t"))
+                                continue;
+                }
+
+                if (opdsec && sym.st_shndx == opdidx) {
+                        u32 offset = sym.st_value - opdshdr.sh_addr;
+                        u64 *opd = opddata->d_buf + offset;
+                        sym.st_value = *opd;
+                        sym.st_shndx = elf_addr_to_index(elf, sym.st_value);
+                }
+
                 sec = elf_getscn(elf, sym.st_shndx);
                 if (!sec)
                         goto out_elf_end;
 
                 gelf_getshdr(sec, &shdr);
-                obj_start = sym.st_value;
 
-                sym.st_value -= shdr.sh_addr - shdr.sh_offset;
+                if (is_label && !elf_sec__is_a(&shdr, secstrs, map->type))
+                        continue;
+
+                section_name = elf_sec__name(&shdr, secstrs);
+
+                if (self->kernel != DSO_TYPE_USER || kmodule) {
+                        char dso_name[PATH_MAX];
+
+                        if (strcmp(section_name,
+                                   (curr_dso->short_name +
+                                    self->short_name_len)) == 0)
+                                goto new_symbol;
+
+                        if (strcmp(section_name, ".text") == 0) {
+                                curr_map = map;
+                                curr_dso = self;
+                                goto new_symbol;
+                        }
+
+                        snprintf(dso_name, sizeof(dso_name),
+                                 "%s%s", self->short_name, section_name);
+
+                        curr_map = map_groups__find_by_name(kmap->kmaps, map->type, dso_name);
+                        if (curr_map == NULL) {
+                                u64 start = sym.st_value;
+
+                                if (kmodule)
+                                        start += map->start + shdr.sh_offset;
+
+                                curr_dso = dso__new(dso_name);
+                                if (curr_dso == NULL)
+                                        goto out_elf_end;
+                                curr_dso->kernel = self->kernel;
+                                curr_map = map__new2(start, curr_dso,
+                                                     map->type);
+                                if (curr_map == NULL) {
+                                        dso__delete(curr_dso);
+                                        goto out_elf_end;
+                                }
+                                curr_map->map_ip = identity__map_ip;
+                                curr_map->unmap_ip = identity__map_ip;
+                                curr_dso->origin = self->origin;
+                                map_groups__insert(kmap->kmaps, curr_map);
+                                dsos__add(&self->node, curr_dso);
+                                dso__set_loaded(curr_dso, map->type);
+                        } else
+                                curr_dso = curr_map->dso;
+
+                        goto new_symbol;
+                }
 
+                if (curr_dso->adjust_symbols) {
+                        pr_debug4("%s: adjusting symbol: st_value: %#Lx "
+                                  "sh_addr: %#Lx sh_offset: %#Lx\n", __func__,
+                                  (u64)sym.st_value, (u64)shdr.sh_addr,
+                                  (u64)shdr.sh_offset);
+                        sym.st_value -= shdr.sh_addr - shdr.sh_offset;
+                }
+                /*
+                 * We need to figure out if the object was created from C++ sources
+                 * DWARF DW_compile_unit has this, but we don't always have access
+                 * to it...
+                 */
+                demangled = bfd_demangle(NULL, elf_name, DMGL_PARAMS | DMGL_ANSI);
+                if (demangled != NULL)
+                        elf_name = demangled;
+new_symbol:
                 f = symbol__new(sym.st_value, sym.st_size,
-                                elf_sym__name(&sym, symstrs),
-                                self->sym_priv_size, obj_start, verbose);
+                                GELF_ST_BIND(sym.st_info), elf_name);
+                free(demangled);
                 if (!f)
                         goto out_elf_end;
 
-                if (filter && filter(self, f))
-                        symbol__delete(f, self->sym_priv_size);
+                if (filter && filter(curr_map, f))
+                        symbol__delete(f);
                 else {
-                        dso__insert_symbol(self, f);
+                        symbols__insert(&curr_dso->symbols[curr_map->type], f);
                         nr++;
                 }
         }
 
+        /*
+         * For misannotated, zeroed, ASM function sizes.
+         */
+        if (nr > 0) {
+                symbols__fixup_end(&self->symbols[map->type]);
+                if (kmap) {
+                        /*
+                         * We need to fixup this here too because we create new
+                         * maps here, for things like vsyscall sections.
+                         */
+                        __map_groups__fixup_end(kmap->kmaps, map->type);
+                }
+        }
         err = nr;
 out_elf_end:
         elf_end(elf);
@@ -558,84 +1257,1339 @@ out_close:
         return err;
 }
 
-int dso__load(struct dso *self, symbol_filter_t filter, int verbose)
+static bool dso__build_id_equal(const struct dso *self, u8 *build_id)
+{
+        return memcmp(self->build_id, build_id, sizeof(self->build_id)) == 0;
+}
+
+bool __dsos__read_build_ids(struct list_head *head, bool with_hits)
+{
+        bool have_build_id = false;
+        struct dso *pos;
+
+        list_for_each_entry(pos, head, node) {
+                if (with_hits && !pos->hit)
+                        continue;
+                if (pos->has_build_id) {
+                        have_build_id = true;
+                        continue;
+                }
+                if (filename__read_build_id(pos->long_name, pos->build_id,
+                                            sizeof(pos->build_id)) > 0) {
+                        have_build_id     = true;
+                        pos->has_build_id = true;
+                }
+        }
+
+        return have_build_id;
+}
+
+/*
+ * Align offset to 4 bytes as needed for note name and descriptor data.
+ */
+#define NOTE_ALIGN(n) (((n) + 3) & -4U)
+
+static int elf_read_build_id(Elf *elf, void *bf, size_t size)
+{
+        int err = -1;
+        GElf_Ehdr ehdr;
+        GElf_Shdr shdr;
+        Elf_Data *data;
+        Elf_Scn *sec;
+        Elf_Kind ek;
+        void *ptr;
+
+        if (size < BUILD_ID_SIZE)
+                goto out;
+
+        ek = elf_kind(elf);
+        if (ek != ELF_K_ELF)
+                goto out;
+
+        if (gelf_getehdr(elf, &ehdr) == NULL) {
+                pr_err("%s: cannot get elf header.\n", __func__);
+                goto out;
+        }
+
+        sec = elf_section_by_name(elf, &ehdr, &shdr,
+                                  ".note.gnu.build-id", NULL);
+        if (sec == NULL) {
+                sec = elf_section_by_name(elf, &ehdr, &shdr,
+                                          ".notes", NULL);
+                if (sec == NULL)
+                        goto out;
+        }
+
+        data = elf_getdata(sec, NULL);
+        if (data == NULL)
+                goto out;
+
+        ptr = data->d_buf;
+        while (ptr < (data->d_buf + data->d_size)) {
+                GElf_Nhdr *nhdr = ptr;
+                int namesz = NOTE_ALIGN(nhdr->n_namesz),
+                    descsz = NOTE_ALIGN(nhdr->n_descsz);
+                const char *name;
+
+                ptr += sizeof(*nhdr);
+                name = ptr;
+                ptr += namesz;
+                if (nhdr->n_type == NT_GNU_BUILD_ID &&
+                    nhdr->n_namesz == sizeof("GNU")) {
+                        if (memcmp(name, "GNU", sizeof("GNU")) == 0) {
+                                memcpy(bf, ptr, BUILD_ID_SIZE);
+                                err = BUILD_ID_SIZE;
+                                break;
+                        }
+                }
+                ptr += descsz;
+        }
+
+out:
+        return err;
+}
+
+int filename__read_build_id(const char *filename, void *bf, size_t size)
+{
+        int fd, err = -1;
+        Elf *elf;
+
+        if (size < BUILD_ID_SIZE)
+                goto out;
+
+        fd = open(filename, O_RDONLY);
+        if (fd < 0)
+                goto out;
+
+        elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL);
+        if (elf == NULL) {
+                pr_debug2("%s: cannot read %s ELF file.\n", __func__, filename);
+                goto out_close;
+        }
+
+        err = elf_read_build_id(elf, bf, size);
+
+        elf_end(elf);
+out_close:
+        close(fd);
+out:
+        return err;
+}
+
+int sysfs__read_build_id(const char *filename, void *build_id, size_t size)
+{
+        int fd, err = -1;
+
+        if (size < BUILD_ID_SIZE)
+                goto out;
+
+        fd = open(filename, O_RDONLY);
+        if (fd < 0)
+                goto out;
+
+        while (1) {
+                char bf[BUFSIZ];
+                GElf_Nhdr nhdr;
+                int namesz, descsz;
+
+                if (read(fd, &nhdr, sizeof(nhdr)) != sizeof(nhdr))
+                        break;
+
+                namesz = NOTE_ALIGN(nhdr.n_namesz);
+                descsz = NOTE_ALIGN(nhdr.n_descsz);
+                if (nhdr.n_type == NT_GNU_BUILD_ID &&
+                    nhdr.n_namesz == sizeof("GNU")) {
+                        if (read(fd, bf, namesz) != namesz)
+                                break;
+                        if (memcmp(bf, "GNU", sizeof("GNU")) == 0) {
+                                if (read(fd, build_id,
+                                    BUILD_ID_SIZE) == BUILD_ID_SIZE) {
+                                        err = 0;
+                                        break;
+                                }
+                        } else if (read(fd, bf, descsz) != descsz)
+                                break;
+                } else {
+                        int n = namesz + descsz;
+                        if (read(fd, bf, n) != n)
+                                break;
+                }
+        }
+        close(fd);
+out:
+        return err;
+}
+
+char dso__symtab_origin(const struct dso *self)
 {
-        int size = strlen(self->name) + sizeof("/usr/lib/debug%s.debug");
-        char *name = malloc(size);
-        int variant = 0;
+        static const char origin[] = {
+                [DSO__ORIG_KERNEL] =   'k',
+                [DSO__ORIG_JAVA_JIT] = 'j',
+                [DSO__ORIG_BUILD_ID_CACHE] = 'B',
+                [DSO__ORIG_FEDORA] =   'f',
+                [DSO__ORIG_UBUNTU] =   'u',
+                [DSO__ORIG_BUILDID] =  'b',
+                [DSO__ORIG_DSO] =      'd',
+                [DSO__ORIG_KMODULE] =  'K',
+                [DSO__ORIG_GUEST_KERNEL] =  'g',
+                [DSO__ORIG_GUEST_KMODULE] =  'G',
+        };
+
+        if (self == NULL || self->origin == DSO__ORIG_NOT_FOUND)
+                return '!';
+        return origin[self->origin];
+}
+
+int dso__load(struct dso *self, struct map *map, symbol_filter_t filter)
+{
+        int size = PATH_MAX;
+        char *name;
         int ret = -1;
         int fd;
+        struct machine *machine;
+        const char *root_dir;
+        int want_symtab;
+
+        dso__set_loaded(self, map->type);
 
+        if (self->kernel == DSO_TYPE_KERNEL)
+                return dso__load_kernel_sym(self, map, filter);
+        else if (self->kernel == DSO_TYPE_GUEST_KERNEL)
+                return dso__load_guest_kernel_sym(self, map, filter);
+
+        if (map->groups && map->groups->machine)
+                machine = map->groups->machine;
+        else
+                machine = NULL;
+
+        name = malloc(size);
         if (!name)
                 return -1;
 
-        if (strncmp(self->name, "/tmp/perf-", 10) == 0)
-                return dso__load_perf_map(self, filter, verbose);
+        self->adjust_symbols = 0;
 
-more:
-        do {
-                switch (variant) {
-                case 0: /* Fedora */
-                        snprintf(name, size, "/usr/lib/debug%s.debug", self->name);
+        if (strncmp(self->name, "/tmp/perf-", 10) == 0) {
+                ret = dso__load_perf_map(self, map, filter);
+                self->origin = ret > 0 ? DSO__ORIG_JAVA_JIT :
+                                         DSO__ORIG_NOT_FOUND;
+                return ret;
+        }
+
+        /* Iterate over candidate debug images.
+         * On the first pass, only load images if they have a full symtab.
+         * Failing that, do a second pass where we accept .dynsym also
+         */
+        for (self->origin = DSO__ORIG_BUILD_ID_CACHE, want_symtab = 1;
+             self->origin != DSO__ORIG_NOT_FOUND;
+             self->origin++) {
+                switch (self->origin) {
+                case DSO__ORIG_BUILD_ID_CACHE:
+                        /* skip the locally configured cache if a symfs is given */
+                        if (symbol_conf.symfs[0] ||
+                            (dso__build_id_filename(self, name, size) == NULL)) {
+                                continue;
+                        }
+                        break;
+                case DSO__ORIG_FEDORA:
+                        snprintf(name, size, "%s/usr/lib/debug%s.debug",
+                                 symbol_conf.symfs, self->long_name);
                         break;
-                case 1: /* Ubuntu */
-                        snprintf(name, size, "/usr/lib/debug%s", self->name);
+                case DSO__ORIG_UBUNTU:
+                        snprintf(name, size, "%s/usr/lib/debug%s",
+                                 symbol_conf.symfs, self->long_name);
                         break;
-                case 2: /* Sane people */
-                        snprintf(name, size, "%s", self->name);
+                case DSO__ORIG_BUILDID: {
+                        char build_id_hex[BUILD_ID_SIZE * 2 + 1];
+
+                        if (!self->has_build_id)
+                                continue;
+
+                        build_id__sprintf(self->build_id,
+                                          sizeof(self->build_id),
+                                          build_id_hex);
+                        snprintf(name, size,
+                                 "%s/usr/lib/debug/.build-id/%.2s/%s.debug",
+                                 symbol_conf.symfs, build_id_hex, build_id_hex + 2);
+                        }
+                        break;
+                case DSO__ORIG_DSO:
+                        snprintf(name, size, "%s%s",
+                             symbol_conf.symfs, self->long_name);
+                        break;
+                case DSO__ORIG_GUEST_KMODULE:
+                        if (map->groups && map->groups->machine)
+                                root_dir = map->groups->machine->root_dir;
+                        else
+                                root_dir = "";
+                        snprintf(name, size, "%s%s%s", symbol_conf.symfs,
+                                 root_dir, self->long_name);
+                        break;
+
+                case DSO__ORIG_KMODULE:
+                        snprintf(name, size, "%s%s", symbol_conf.symfs,
+                                 self->long_name);
                         break;
 
                 default:
-                        goto out;
+                        /*
+                         * If we wanted a full symtab but no image had one,
+                         * relax our requirements and repeat the search.
+                         */
+                        if (want_symtab) {
+                                want_symtab = 0;
+                                self->origin = DSO__ORIG_BUILD_ID_CACHE;
+                        } else
+                                continue;
                 }
-                variant++;
 
+                /* Name is now the name of the next image to try */
                 fd = open(name, O_RDONLY);
-        } while (fd < 0);
+                if (fd < 0)
+                        continue;
 
-        ret = dso__load_sym(self, fd, name, filter, verbose);
-        close(fd);
+                ret = dso__load_sym(self, map, name, fd, filter, 0,
+                                    want_symtab);
+                close(fd);
+
+                /*
+                 * Some people seem to have debuginfo files _WITHOUT_ debug
+                 * info!?!?
+                 */
+                if (!ret)
+                        continue;
+
+                if (ret > 0) {
+                        int nr_plt = dso__synthesize_plt_symbols(self, map, filter);
+                        if (nr_plt > 0)
+                                ret += nr_plt;
+                        break;
+                }
+        }
 
+        free(name);
+        if (ret < 0 && strstr(self->name, " (deleted)") != NULL)
+                return 0;
+        return ret;
+}
+
+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[type]); nd; nd = rb_next(nd)) {
+                struct map *map = rb_entry(nd, struct map, rb_node);
+
+                if (map->dso && strcmp(map->dso->short_name, name) == 0)
+                        return map;
+        }
+
+        return NULL;
+}
+
+static int dso__kernel_module_get_build_id(struct dso *self,
+                                const char *root_dir)
+{
+        char filename[PATH_MAX];
         /*
-         * Some people seem to have debuginfo files _WITHOUT_ debug info!?!?
+         * kernel module short names are of the form "[module]" and
+         * we need just "module" here.
          */
-        if (!ret)
-                goto more;
+        const char *name = self->short_name + 1;
+
+        snprintf(filename, sizeof(filename),
+                 "%s/sys/module/%.*s/notes/.note.gnu.build-id",
+                 root_dir, (int)strlen(name) - 1, name);
+
+        if (sysfs__read_build_id(filename, self->build_id,
+                                 sizeof(self->build_id)) == 0)
+                self->has_build_id = true;
+
+        return 0;
+}
+
+static int map_groups__set_modules_path_dir(struct map_groups *self,
+                                const char *dir_name)
+{
+        struct dirent *dent;
+        DIR *dir = opendir(dir_name);
+        int ret = 0;
+
+        if (!dir) {
+                pr_debug("%s: cannot open %s dir\n", __func__, dir_name);
+                return -1;
+        }
+
+        while ((dent = readdir(dir)) != NULL) {
+                char path[PATH_MAX];
+                struct stat st;
+
+                /*sshfs might return bad dent->d_type, so we have to stat*/
+                sprintf(path, "%s/%s", dir_name, dent->d_name);
+                if (stat(path, &st))
+                        continue;
+
+                if (S_ISDIR(st.st_mode)) {
+                        if (!strcmp(dent->d_name, ".") ||
+                            !strcmp(dent->d_name, ".."))
+                                continue;
+
+                        snprintf(path, sizeof(path), "%s/%s",
+                                 dir_name, dent->d_name);
+                        ret = map_groups__set_modules_path_dir(self, path);
+                        if (ret < 0)
+                                goto out;
+                } else {
+                        char *dot = strrchr(dent->d_name, '.'),
+                             dso_name[PATH_MAX];
+                        struct map *map;
+                        char *long_name;
+
+                        if (dot == NULL || strcmp(dot, ".ko"))
+                                continue;
+                        snprintf(dso_name, sizeof(dso_name), "[%.*s]",
+                                 (int)(dot - dent->d_name), dent->d_name);
+
+                        strxfrchar(dso_name, '-', '_');
+                        map = map_groups__find_by_name(self, MAP__FUNCTION, dso_name);
+                        if (map == NULL)
+                                continue;
+
+                        snprintf(path, sizeof(path), "%s/%s",
+                                 dir_name, dent->d_name);
+
+                        long_name = strdup(path);
+                        if (long_name == NULL) {
+                                ret = -1;
+                                goto out;
+                        }
+                        dso__set_long_name(map->dso, long_name);
+                        map->dso->lname_alloc = 1;
+                        dso__kernel_module_get_build_id(map->dso, "");
+                }
+        }
 
 out:
-        free(name);
+        closedir(dir);
         return ret;
 }
 
-static int dso__load_vmlinux(struct dso *self, const char *vmlinux,
-                             symbol_filter_t filter, int verbose)
+static char *get_kernel_version(const char *root_dir)
 {
-        int err, fd = open(vmlinux, O_RDONLY);
+        char version[PATH_MAX];
+        FILE *file;
+        char *name, *tmp;
+        const char *prefix = "Linux version ";
+
+        sprintf(version, "%s/proc/version", root_dir);
+        file = fopen(version, "r");
+        if (!file)
+                return NULL;
+
+        version[0] = '\0';
+        tmp = fgets(version, sizeof(version), file);
+        fclose(file);
+
+        name = strstr(version, prefix);
+        if (!name)
+                return NULL;
+        name += strlen(prefix);
+        tmp = strchr(name, ' ');
+        if (tmp)
+                *tmp = '\0';
+
+        return strdup(name);
+}
+
+static int machine__set_modules_path(struct machine *self)
+{
+        char *version;
+        char modules_path[PATH_MAX];
+
+        version = get_kernel_version(self->root_dir);
+        if (!version)
+                return -1;
+
+        snprintf(modules_path, sizeof(modules_path), "%s/lib/modules/%s/kernel",
+                 self->root_dir, version);
+        free(version);
+
+        return map_groups__set_modules_path_dir(&self->kmaps, modules_path);
+}
+
+/*
+ * Constructor variant for modules (where we know from /proc/modules where
+ * they are loaded) and for vmlinux, where only after we load all the
+ * symbols we'll know where it starts and ends.
+ */
+static struct map *map__new2(u64 start, struct dso *dso, enum map_type type)
+{
+        struct map *self = calloc(1, (sizeof(*self) +
+                                      (dso->kernel ? sizeof(struct kmap) : 0)));
+        if (self != NULL) {
+                /*
+                 * ->end will be filled after we load all the symbols
+                 */
+                map__init(self, type, start, 0, 0, dso);
+        }
+
+        return self;
+}
+
+struct map *machine__new_module(struct machine *self, u64 start,
+                                const char *filename)
+{
+        struct map *map;
+        struct dso *dso = __dsos__findnew(&self->kernel_dsos, filename);
+
+        if (dso == NULL)
+                return NULL;
+
+        map = map__new2(start, dso, MAP__FUNCTION);
+        if (map == NULL)
+                return NULL;
+
+        if (machine__is_host(self))
+                dso->origin = DSO__ORIG_KMODULE;
+        else
+                dso->origin = DSO__ORIG_GUEST_KMODULE;
+        map_groups__insert(&self->kmaps, map);
+        return map;
+}
+
+static int machine__create_modules(struct machine *self)
+{
+        char *line = NULL;
+        size_t n;
+        FILE *file;
+        struct map *map;
+        const char *modules;
+        char path[PATH_MAX];
+
+        if (machine__is_default_guest(self))
+                modules = symbol_conf.default_guest_modules;
+        else {
+                sprintf(path, "%s/proc/modules", self->root_dir);
+                modules = path;
+        }
+
+        file = fopen(modules, "r");
+        if (file == NULL)
+                return -1;
+
+        while (!feof(file)) {
+                char name[PATH_MAX];
+                u64 start;
+                char *sep;
+                int line_len;
+
+                line_len = getline(&line, &n, file);
+                if (line_len < 0)
+                        break;
+
+                if (!line)
+                        goto out_failure;
+
+                line[--line_len] = '\0'; /* \n */
+
+                sep = strrchr(line, 'x');
+                if (sep == NULL)
+                        continue;
+
+                hex2u64(sep + 1, &start);
+
+                sep = strchr(line, ' ');
+                if (sep == NULL)
+                        continue;
+
+                *sep = '\0';
+
+                snprintf(name, sizeof(name), "[%s]", line);
+                map = machine__new_module(self, start, name);
+                if (map == NULL)
+                        goto out_delete_line;
+                dso__kernel_module_get_build_id(map->dso, self->root_dir);
+        }
+
+        free(line);
+        fclose(file);
+
+        return machine__set_modules_path(self);
 
+out_delete_line:
+        free(line);
+out_failure:
+        return -1;
+}
+
+int dso__load_vmlinux(struct dso *self, struct map *map,
+                      const char *vmlinux, symbol_filter_t filter)
+{
+        int err = -1, fd;
+        char symfs_vmlinux[PATH_MAX];
+
+        snprintf(symfs_vmlinux, sizeof(symfs_vmlinux), "%s/%s",
+                 symbol_conf.symfs, vmlinux);
+        fd = open(symfs_vmlinux, O_RDONLY);
         if (fd < 0)
                 return -1;
 
-        err = dso__load_sym(self, fd, vmlinux, filter, verbose);
+        dso__set_loaded(self, map->type);
+        err = dso__load_sym(self, map, symfs_vmlinux, fd, filter, 0, 0);
         close(fd);
 
+        if (err > 0)
+                pr_debug("Using %s for symbols\n", symfs_vmlinux);
+
         return err;
 }
 
-int dso__load_kernel(struct dso *self, const char *vmlinux,
-                     symbol_filter_t filter, int verbose)
+int dso__load_vmlinux_path(struct dso *self, struct map *map,
+                           symbol_filter_t filter)
 {
-        int err = -1;
+        int i, err = 0;
+        char *filename;
+
+        pr_debug("Looking at the vmlinux_path (%d entries long)\n",
+                 vmlinux_path__nr_entries + 1);
+
+        filename = dso__build_id_filename(self, NULL, 0);
+        if (filename != NULL) {
+                err = dso__load_vmlinux(self, map, filename, filter);
+                if (err > 0) {
+                        dso__set_long_name(self, filename);
+                        goto out;
+                }
+                free(filename);
+        }
+
+        for (i = 0; i < vmlinux_path__nr_entries; ++i) {
+                err = dso__load_vmlinux(self, map, vmlinux_path[i], filter);
+                if (err > 0) {
+                        dso__set_long_name(self, strdup(vmlinux_path[i]));
+                        break;
+                }
+        }
+out:
+        return err;
+}
+
+static int dso__load_kernel_sym(struct dso *self, struct map *map,
+                                symbol_filter_t filter)
+{
+        int err;
+        const char *kallsyms_filename = NULL;
+        char *kallsyms_allocated_filename = NULL;
+        /*
+         * Step 1: if the user specified a kallsyms or vmlinux filename, use
+         * it and only it, reporting errors to the user if it cannot be used.
+         *
+         * For instance, try to analyse an ARM perf.data file _without_ a
+         * build-id, or if the user specifies the wrong path to the right
+         * vmlinux file, obviously we can't fallback to another vmlinux (a
+         * x86_86 one, on the machine where analysis is being performed, say),
+         * or worse, /proc/kallsyms.
+         *
+         * If the specified file _has_ a build-id and there is a build-id
+         * section in the perf.data file, we will still do the expected
+         * validation in dso__load_vmlinux and will bail out if they don't
+         * match.
+         */
+        if (symbol_conf.kallsyms_name != NULL) {
+                kallsyms_filename = symbol_conf.kallsyms_name;
+                goto do_kallsyms;
+        }
+
+        if (symbol_conf.vmlinux_name != NULL) {
+                err = dso__load_vmlinux(self, map,
+                                        symbol_conf.vmlinux_name, filter);
+                if (err > 0) {
+                        dso__set_long_name(self,
+                                           strdup(symbol_conf.vmlinux_name));
+                        goto out_fixup;
+                }
+                return err;
+        }
+
+        if (vmlinux_path != NULL) {
+                err = dso__load_vmlinux_path(self, map, filter);
+                if (err > 0)
+                        goto out_fixup;
+        }
+
+        /* do not try local files if a symfs was given */
+        if (symbol_conf.symfs[0] != 0)
+                return -1;
+
+        /*
+         * Say the kernel DSO was created when processing the build-id header table,
+         * we have a build-id, so check if it is the same as the running kernel,
+         * using it if it is.
+         */
+        if (self->has_build_id) {
+                u8 kallsyms_build_id[BUILD_ID_SIZE];
+                char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+                if (sysfs__read_build_id("/sys/kernel/notes", kallsyms_build_id,
+                                         sizeof(kallsyms_build_id)) == 0) {
+                        if (dso__build_id_equal(self, kallsyms_build_id)) {
+                                kallsyms_filename = "/proc/kallsyms";
+                                goto do_kallsyms;
+                        }
+                }
+                /*
+                 * Now look if we have it on the build-id cache in
+                 * $HOME/.debug/[kernel.kallsyms].
+                 */
+                build_id__sprintf(self->build_id, sizeof(self->build_id),
+                                  sbuild_id);
+
+                if (asprintf(&kallsyms_allocated_filename,
+                             "%s/.debug/[kernel.kallsyms]/%s",
+                             getenv("HOME"), sbuild_id) == -1) {
+                        pr_err("Not enough memory for kallsyms file lookup\n");
+                        return -1;
+                }
+
+                kallsyms_filename = kallsyms_allocated_filename;
+
+                if (access(kallsyms_filename, F_OK)) {
+                        pr_err("No kallsyms or vmlinux with build-id %s "
+                               "was found\n", sbuild_id);
+                        free(kallsyms_allocated_filename);
+                        return -1;
+                }
+        } else {
+                /*
+                 * Last resort, if we don't have a build-id and couldn't find
+                 * any vmlinux file, try the running kernel kallsyms table.
+                 */
+                kallsyms_filename = "/proc/kallsyms";
+        }
+
+do_kallsyms:
+        err = dso__load_kallsyms(self, kallsyms_filename, map, filter);
+        if (err > 0)
+                pr_debug("Using %s for symbols\n", kallsyms_filename);
+        free(kallsyms_allocated_filename);
+
+        if (err > 0) {
+out_fixup:
+                if (kallsyms_filename != NULL)
+                        dso__set_long_name(self, strdup("[kernel.kallsyms]"));
+                map__fixup_start(map);
+                map__fixup_end(map);
+        }
 
-        if (vmlinux)
-                err = dso__load_vmlinux(self, vmlinux, filter, verbose);
+        return err;
+}
+
+static int dso__load_guest_kernel_sym(struct dso *self, struct map *map,
+                                symbol_filter_t filter)
+{
+        int err;
+        const char *kallsyms_filename = NULL;
+        struct machine *machine;
+        char path[PATH_MAX];
+
+        if (!map->groups) {
+                pr_debug("Guest kernel map hasn't the point to groups\n");
+                return -1;
+        }
+        machine = map->groups->machine;
 
-        if (err)
-                err = dso__load_kallsyms(self, filter, verbose);
+        if (machine__is_default_guest(machine)) {
+                /*
+                 * if the user specified a vmlinux filename, use it and only
+                 * it, reporting errors to the user if it cannot be used.
+                 * Or use file guest_kallsyms inputted by user on commandline
+                 */
+                if (symbol_conf.default_guest_vmlinux_name != NULL) {
+                        err = dso__load_vmlinux(self, map,
+                                symbol_conf.default_guest_vmlinux_name, filter);
+                        goto out_try_fixup;
+                }
+
+                kallsyms_filename = symbol_conf.default_guest_kallsyms;
+                if (!kallsyms_filename)
+                        return -1;
+        } else {
+                sprintf(path, "%s/proc/kallsyms", machine->root_dir);
+                kallsyms_filename = path;
+        }
+
+        err = dso__load_kallsyms(self, kallsyms_filename, map, filter);
+        if (err > 0)
+                pr_debug("Using %s for symbols\n", kallsyms_filename);
+
+out_try_fixup:
+        if (err > 0) {
+                if (kallsyms_filename != NULL) {
+                        machine__mmap_name(machine, path, sizeof(path));
+                        dso__set_long_name(self, strdup(path));
+                }
+                map__fixup_start(map);
+                map__fixup_end(map);
+        }
 
         return err;
 }
 
-void symbol__init(void)
+static void dsos__add(struct list_head *head, struct dso *dso)
+{
+        list_add_tail(&dso->node, head);
+}
+
+static struct dso *dsos__find(struct list_head *head, const char *name)
+{
+        struct dso *pos;
+
+        list_for_each_entry(pos, head, node)
+                if (strcmp(pos->long_name, name) == 0)
+                        return pos;
+        return NULL;
+}
+
+struct dso *__dsos__findnew(struct list_head *head, const char *name)
 {
+        struct dso *dso = dsos__find(head, name);
+
+        if (!dso) {
+                dso = dso__new(name);
+                if (dso != NULL) {
+                        dsos__add(head, dso);
+                        dso__set_basename(dso);
+                }
+        }
+
+        return dso;
+}
+
+size_t __dsos__fprintf(struct list_head *head, FILE *fp)
+{
+        struct dso *pos;
+        size_t ret = 0;
+
+        list_for_each_entry(pos, head, node) {
+                int i;
+                for (i = 0; i < MAP__NR_TYPES; ++i)
+                        ret += dso__fprintf(pos, i, fp);
+        }
+
+        return ret;
+}
+
+size_t machines__fprintf_dsos(struct rb_root *self, FILE *fp)
+{
+        struct rb_node *nd;
+        size_t ret = 0;
+
+        for (nd = rb_first(self); nd; nd = rb_next(nd)) {
+                struct machine *pos = rb_entry(nd, struct machine, rb_node);
+                ret += __dsos__fprintf(&pos->kernel_dsos, fp);
+                ret += __dsos__fprintf(&pos->user_dsos, fp);
+        }
+
+        return ret;
+}
+
+static size_t __dsos__fprintf_buildid(struct list_head *head, FILE *fp,
+                                      bool with_hits)
+{
+        struct dso *pos;
+        size_t ret = 0;
+
+        list_for_each_entry(pos, head, node) {
+                if (with_hits && !pos->hit)
+                        continue;
+                ret += dso__fprintf_buildid(pos, fp);
+                ret += fprintf(fp, " %s\n", pos->long_name);
+        }
+        return ret;
+}
+
+size_t machine__fprintf_dsos_buildid(struct machine *self, FILE *fp, bool with_hits)
+{
+        return __dsos__fprintf_buildid(&self->kernel_dsos, fp, with_hits) +
+               __dsos__fprintf_buildid(&self->user_dsos, fp, with_hits);
+}
+
+size_t machines__fprintf_dsos_buildid(struct rb_root *self, FILE *fp, bool with_hits)
+{
+        struct rb_node *nd;
+        size_t ret = 0;
+
+        for (nd = rb_first(self); nd; nd = rb_next(nd)) {
+                struct machine *pos = rb_entry(nd, struct machine, rb_node);
+                ret += machine__fprintf_dsos_buildid(pos, fp, with_hits);
+        }
+        return ret;
+}
+
+struct dso *dso__new_kernel(const char *name)
+{
+        struct dso *self = dso__new(name ?: "[kernel.kallsyms]");
+
+        if (self != NULL) {
+                dso__set_short_name(self, "[kernel]");
+                self->kernel = DSO_TYPE_KERNEL;
+        }
+
+        return self;
+}
+
+static struct dso *dso__new_guest_kernel(struct machine *machine,
+                                        const char *name)
+{
+        char bf[PATH_MAX];
+        struct dso *self = dso__new(name ?: machine__mmap_name(machine, bf, sizeof(bf)));
+
+        if (self != NULL) {
+                dso__set_short_name(self, "[guest.kernel]");
+                self->kernel = DSO_TYPE_GUEST_KERNEL;
+        }
+
+        return self;
+}
+
+void dso__read_running_kernel_build_id(struct dso *self, struct machine *machine)
+{
+        char path[PATH_MAX];
+
+        if (machine__is_default_guest(machine))
+                return;
+        sprintf(path, "%s/sys/kernel/notes", machine->root_dir);
+        if (sysfs__read_build_id(path, self->build_id,
+                                 sizeof(self->build_id)) == 0)
+                self->has_build_id = true;
+}
+
+static struct dso *machine__create_kernel(struct machine *self)
+{
+        const char *vmlinux_name = NULL;
+        struct dso *kernel;
+
+        if (machine__is_host(self)) {
+                vmlinux_name = symbol_conf.vmlinux_name;
+                kernel = dso__new_kernel(vmlinux_name);
+        } else {
+                if (machine__is_default_guest(self))
+                        vmlinux_name = symbol_conf.default_guest_vmlinux_name;
+                kernel = dso__new_guest_kernel(self, vmlinux_name);
+        }
+
+        if (kernel != NULL) {
+                dso__read_running_kernel_build_id(kernel, self);
+                dsos__add(&self->kernel_dsos, kernel);
+        }
+        return kernel;
+}
+
+struct process_args {
+        u64 start;
+};
+
+static int symbol__in_kernel(void *arg, const char *name,
+                             char type __used, u64 start, u64 end __used)
+{
+        struct process_args *args = arg;
+
+        if (strchr(name, '['))
+                return 0;
+
+        args->start = start;
+        return 1;
+}
+
+/* Figure out the start address of kernel map from /proc/kallsyms */
+static u64 machine__get_kernel_start_addr(struct machine *machine)
+{
+        const char *filename;
+        char path[PATH_MAX];
+        struct process_args args;
+
+        if (machine__is_host(machine)) {
+                filename = "/proc/kallsyms";
+        } else {
+                if (machine__is_default_guest(machine))
+                        filename = (char *)symbol_conf.default_guest_kallsyms;
+                else {
+                        sprintf(path, "%s/proc/kallsyms", machine->root_dir);
+                        filename = path;
+                }
+        }
+
+        if (kallsyms__parse(filename, &args, symbol__in_kernel) <= 0)
+                return 0;
+
+        return args.start;
+}
+
+int __machine__create_kernel_maps(struct machine *self, struct dso *kernel)
+{
+        enum map_type type;
+        u64 start = machine__get_kernel_start_addr(self);
+
+        for (type = 0; type < MAP__NR_TYPES; ++type) {
+                struct kmap *kmap;
+
+                self->vmlinux_maps[type] = map__new2(start, kernel, type);
+                if (self->vmlinux_maps[type] == NULL)
+                        return -1;
+
+                self->vmlinux_maps[type]->map_ip =
+                        self->vmlinux_maps[type]->unmap_ip = identity__map_ip;
+
+                kmap = map__kmap(self->vmlinux_maps[type]);
+                kmap->kmaps = &self->kmaps;
+                map_groups__insert(&self->kmaps, self->vmlinux_maps[type]);
+        }
+
+        return 0;
+}
+
+void machine__destroy_kernel_maps(struct machine *self)
+{
+        enum map_type type;
+
+        for (type = 0; type < MAP__NR_TYPES; ++type) {
+                struct kmap *kmap;
+
+                if (self->vmlinux_maps[type] == NULL)
+                        continue;
+
+                kmap = map__kmap(self->vmlinux_maps[type]);
+                map_groups__remove(&self->kmaps, self->vmlinux_maps[type]);
+                if (kmap->ref_reloc_sym) {
+                        /*
+                         * ref_reloc_sym is shared among all maps, so free just
+                         * on one of them.
+                         */
+                        if (type == MAP__FUNCTION) {
+                                free((char *)kmap->ref_reloc_sym->name);
+                                kmap->ref_reloc_sym->name = NULL;
+                                free(kmap->ref_reloc_sym);
+                        }
+                        kmap->ref_reloc_sym = NULL;
+                }
+
+                map__delete(self->vmlinux_maps[type]);
+                self->vmlinux_maps[type] = NULL;
+        }
+}
+
+int machine__create_kernel_maps(struct machine *self)
+{
+        struct dso *kernel = machine__create_kernel(self);
+
+        if (kernel == NULL ||
+            __machine__create_kernel_maps(self, kernel) < 0)
+                return -1;
+
+        if (symbol_conf.use_modules && machine__create_modules(self) < 0)
+                pr_debug("Problems creating module maps, continuing anyway...\n");
+        /*
+         * Now that we have all the maps created, just set the ->end of them:
+         */
+        map_groups__fixup_end(&self->kmaps);
+        return 0;
+}
+
+static void vmlinux_path__exit(void)
+{
+        while (--vmlinux_path__nr_entries >= 0) {
+                free(vmlinux_path[vmlinux_path__nr_entries]);
+                vmlinux_path[vmlinux_path__nr_entries] = NULL;
+        }
+
+        free(vmlinux_path);
+        vmlinux_path = NULL;
+}
+
+static int vmlinux_path__init(void)
+{
+        struct utsname uts;
+        char bf[PATH_MAX];
+
+        vmlinux_path = malloc(sizeof(char *) * 5);
+        if (vmlinux_path == NULL)
+                return -1;
+
+        vmlinux_path[vmlinux_path__nr_entries] = strdup("vmlinux");
+        if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+                goto out_fail;
+        ++vmlinux_path__nr_entries;
+        vmlinux_path[vmlinux_path__nr_entries] = strdup("/boot/vmlinux");
+        if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+                goto out_fail;
+        ++vmlinux_path__nr_entries;
+
+        /* only try running kernel version if no symfs was given */
+        if (symbol_conf.symfs[0] != 0)
+                return 0;
+
+        if (uname(&uts) < 0)
+                return -1;
+
+        snprintf(bf, sizeof(bf), "/boot/vmlinux-%s", uts.release);
+        vmlinux_path[vmlinux_path__nr_entries] = strdup(bf);
+        if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+                goto out_fail;
+        ++vmlinux_path__nr_entries;
+        snprintf(bf, sizeof(bf), "/lib/modules/%s/build/vmlinux", uts.release);
+        vmlinux_path[vmlinux_path__nr_entries] = strdup(bf);
+        if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+                goto out_fail;
+        ++vmlinux_path__nr_entries;
+        snprintf(bf, sizeof(bf), "/usr/lib/debug/lib/modules/%s/vmlinux",
+                 uts.release);
+        vmlinux_path[vmlinux_path__nr_entries] = strdup(bf);
+        if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+                goto out_fail;
+        ++vmlinux_path__nr_entries;
+
+        return 0;
+
+out_fail:
+        vmlinux_path__exit();
+        return -1;
+}
+
+size_t machine__fprintf_vmlinux_path(struct machine *self, FILE *fp)
+{
+        int i;
+        size_t printed = 0;
+        struct dso *kdso = self->vmlinux_maps[MAP__FUNCTION]->dso;
+
+        if (kdso->has_build_id) {
+                char filename[PATH_MAX];
+                if (dso__build_id_filename(kdso, filename, sizeof(filename)))
+                        printed += fprintf(fp, "[0] %s\n", filename);
+        }
+
+        for (i = 0; i < vmlinux_path__nr_entries; ++i)
+                printed += fprintf(fp, "[%d] %s\n",
+                                   i + kdso->has_build_id, vmlinux_path[i]);
+
+        return printed;
+}
+
+static int setup_list(struct strlist **list, const char *list_str,
+                      const char *list_name)
+{
+        if (list_str == NULL)
+                return 0;
+
+        *list = strlist__new(true, list_str);
+        if (!*list) {
+                pr_err("problems parsing %s list\n", list_name);
+                return -1;
+        }
+        return 0;
+}
+
+int symbol__init(void)
+{
+        const char *symfs;
+
+        if (symbol_conf.initialized)
+                return 0;
+
         elf_version(EV_CURRENT);
+        if (symbol_conf.sort_by_name)
+                symbol_conf.priv_size += (sizeof(struct symbol_name_rb_node) -
+                                          sizeof(struct symbol));
+
+        if (symbol_conf.try_vmlinux_path && vmlinux_path__init() < 0)
+                return -1;
+
+        if (symbol_conf.field_sep && *symbol_conf.field_sep == '.') {
+                pr_err("'.' is the only non valid --field-separator argument\n");
+                return -1;
+        }
+
+        if (setup_list(&symbol_conf.dso_list,
+                       symbol_conf.dso_list_str, "dso") < 0)
+                return -1;
+
+        if (setup_list(&symbol_conf.comm_list,
+                       symbol_conf.comm_list_str, "comm") < 0)
+                goto out_free_dso_list;
+
+        if (setup_list(&symbol_conf.sym_list,
+                       symbol_conf.sym_list_str, "symbol") < 0)
+                goto out_free_comm_list;
+
+        /*
+         * A path to symbols of "/" is identical to ""
+         * reset here for simplicity.
+         */
+        symfs = realpath(symbol_conf.symfs, NULL);
+        if (symfs == NULL)
+                symfs = symbol_conf.symfs;
+        if (strcmp(symfs, "/") == 0)
+                symbol_conf.symfs = "";
+        if (symfs != symbol_conf.symfs)
+                free((void *)symfs);
+
+        symbol_conf.initialized = true;
+        return 0;
+
+out_free_dso_list:
+        strlist__delete(symbol_conf.dso_list);
+out_free_comm_list:
+        strlist__delete(symbol_conf.comm_list);
+        return -1;
+}
+
+void symbol__exit(void)
+{
+        if (!symbol_conf.initialized)
+                return;
+        strlist__delete(symbol_conf.sym_list);
+        strlist__delete(symbol_conf.dso_list);
+        strlist__delete(symbol_conf.comm_list);
+        vmlinux_path__exit();
+        symbol_conf.sym_list = symbol_conf.dso_list = symbol_conf.comm_list = NULL;
+        symbol_conf.initialized = false;
+}
+
+int machines__create_kernel_maps(struct rb_root *self, pid_t pid)
+{
+        struct machine *machine = machines__findnew(self, pid);
+
+        if (machine == NULL)
+                return -1;
+
+        return machine__create_kernel_maps(machine);
+}
+
+static int hex(char ch)
+{
+        if ((ch >= '0') && (ch <= '9'))
+                return ch - '0';
+        if ((ch >= 'a') && (ch <= 'f'))
+                return ch - 'a' + 10;
+        if ((ch >= 'A') && (ch <= 'F'))
+                return ch - 'A' + 10;
+        return -1;
+}
+
+/*
+ * While we find nice hex chars, build a long_val.
+ * Return number of chars processed.
+ */
+int hex2u64(const char *ptr, u64 *long_val)
+{
+        const char *p = ptr;
+        *long_val = 0;
+
+        while (*p) {
+                const int hex_val = hex(*p);
+
+                if (hex_val < 0)
+                        break;
+
+                *long_val = (*long_val << 4) | hex_val;
+                p++;
+        }
+
+        return p - ptr;
+}
+
+char *strxfrchar(char *s, char from, char to)
+{
+        char *p = s;
+
+        while ((p = strchr(p, from)) != NULL)
+                *p++ = to;
+
+        return s;
+}
+
+int machines__create_guest_kernel_maps(struct rb_root *self)
+{
+        int ret = 0;
+        struct dirent **namelist = NULL;
+        int i, items = 0;
+        char path[PATH_MAX];
+        pid_t pid;
+
+        if (symbol_conf.default_guest_vmlinux_name ||
+            symbol_conf.default_guest_modules ||
+            symbol_conf.default_guest_kallsyms) {
+                machines__create_kernel_maps(self, DEFAULT_GUEST_KERNEL_ID);
+        }
+
+        if (symbol_conf.guestmount) {
+                items = scandir(symbol_conf.guestmount, &namelist, NULL, NULL);
+                if (items <= 0)
+                        return -ENOENT;
+                for (i = 0; i < items; i++) {
+                        if (!isdigit(namelist[i]->d_name[0])) {
+                                /* Filter out . and .. */
+                                continue;
+                        }
+                        pid = atoi(namelist[i]->d_name);
+                        sprintf(path, "%s/%s/proc/kallsyms",
+                                symbol_conf.guestmount,
+                                namelist[i]->d_name);
+                        ret = access(path, R_OK);
+                        if (ret) {
+                                pr_debug("Can't access file %s\n", path);
+                                goto failure;
+                        }
+                        machines__create_kernel_maps(self, pid);
+                }
+failure:
+                free(namelist);
+        }
+
+        return ret;
+}
+
+void machines__destroy_guest_kernel_maps(struct rb_root *self)
+{
+        struct rb_node *next = rb_first(self);
+
+        while (next) {
+                struct machine *pos = rb_entry(next, struct machine, rb_node);
+
+                next = rb_next(&pos->rb_node);
+                rb_erase(&pos->rb_node, self);
+                machine__delete(pos);
+        }
+}
+
+int machine__load_kallsyms(struct machine *self, const char *filename,
+                           enum map_type type, symbol_filter_t filter)
+{
+        struct map *map = self->vmlinux_maps[type];
+        int ret = dso__load_kallsyms(map->dso, filename, map, filter);
+
+        if (ret > 0) {
+                dso__set_loaded(map->dso, type);
+                /*
+                 * Since /proc/kallsyms will have multiple sessions for the
+                 * kernel, with modules between them, fixup the end of all
+                 * sections.
+                 */
+                __map_groups__fixup_end(&self->kmaps, type);
+        }
+
+        return ret;
+}
+
+int machine__load_vmlinux_path(struct machine *self, enum map_type type,
+                               symbol_filter_t filter)
+{
+        struct map *map = self->vmlinux_maps[type];
+        int ret = dso__load_vmlinux_path(map->dso, map, filter);
+
+        if (ret > 0) {
+                dso__set_loaded(map->dso, type);
+                map__reloc_vmlinux(map);
+        }
+
+        return ret;
 }
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index ea332e56e45..670cd1c88f5 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -1,49 +1,236 @@
-#ifndef _PERF_SYMBOL_
-#define _PERF_SYMBOL_ 1
+#ifndef __PERF_SYMBOL
+#define __PERF_SYMBOL 1
 
 #include <linux/types.h>
-#include "../types.h"
-#include "list.h"
-#include "rbtree.h"
+#include <stdbool.h>
+#include <stdint.h>
+#include "map.h"
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include <stdio.h>
+
+#ifdef HAVE_CPLUS_DEMANGLE
+extern char *cplus_demangle(const char *, int);
+
+static inline char *bfd_demangle(void __used *v, const char *c, int i)
+{
+        return cplus_demangle(c, i);
+}
+#else
+#ifdef NO_DEMANGLE
+static inline char *bfd_demangle(void __used *v, const char __used *c,
+                                 int __used i)
+{
+        return NULL;
+}
+#else
+#include <bfd.h>
+#endif
+#endif
+
+int hex2u64(const char *ptr, u64 *val);
+char *strxfrchar(char *s, char from, char to);
+
+/*
+ * libelf 0.8.x and earlier do not support ELF_C_READ_MMAP;
+ * for newer versions we can use mmap to reduce memory usage:
+ */
+#ifdef LIBELF_NO_MMAP
+# define PERF_ELF_C_READ_MMAP ELF_C_READ
+#else
+# define PERF_ELF_C_READ_MMAP ELF_C_READ_MMAP
+#endif
+
+#ifndef DMGL_PARAMS
+#define DMGL_PARAMS      (1 << 0)       /* Include function args */
+#define DMGL_ANSI        (1 << 1)       /* Include const, volatile, etc */
+#endif
+
+#define BUILD_ID_SIZE 20
 
 struct symbol {
         struct rb_node  rb_node;
         u64             start;
         u64             end;
-        u64             obj_start;
-        u64             hist_sum;
-        u64             *hist;
-        void            *priv;
+        u16             namelen;
+        u8              binding;
         char            name[0];
 };
 
+void symbol__delete(struct symbol *self);
+
+struct strlist;
+
+struct symbol_conf {
+        unsigned short  priv_size;
+        bool            try_vmlinux_path,
+                        use_modules,
+                        sort_by_name,
+                        show_nr_samples,
+                        use_callchain,
+                        exclude_other,
+                        show_cpu_utilization,
+                        initialized;
+        const char      *vmlinux_name,
+                        *kallsyms_name,
+                        *source_prefix,
+                        *field_sep;
+        const char      *default_guest_vmlinux_name,
+                        *default_guest_kallsyms,
+                        *default_guest_modules;
+        const char      *guestmount;
+        const char      *dso_list_str,
+                        *comm_list_str,
+                        *sym_list_str,
+                        *col_width_list_str;
+       struct strlist   *dso_list,
+                        *comm_list,
+                        *sym_list;
+        const char      *symfs;
+};
+
+extern struct symbol_conf symbol_conf;
+
+static inline void *symbol__priv(struct symbol *self)
+{
+        return ((void *)self) - symbol_conf.priv_size;
+}
+
+struct ref_reloc_sym {
+        const char      *name;
+        u64             addr;
+        u64             unrelocated_addr;
+};
+
+struct map_symbol {
+        struct map    *map;
+        struct symbol *sym;
+        bool          unfolded;
+        bool          has_children;
+};
+
+struct addr_location {
+        struct thread *thread;
+        struct map    *map;
+        struct symbol *sym;
+        u64           addr;
+        char          level;
+        bool          filtered;
+        u8            cpumode;
+        s32           cpu;
+};
+
+enum dso_kernel_type {
+        DSO_TYPE_USER = 0,
+        DSO_TYPE_KERNEL,
+        DSO_TYPE_GUEST_KERNEL
+};
+
 struct dso {
         struct list_head node;
-        struct rb_root   syms;
-        unsigned int     sym_priv_size;
-        struct symbol    *(*find_symbol)(struct dso *, u64 ip);
+        struct rb_root   symbols[MAP__NR_TYPES];
+        struct rb_root   symbol_names[MAP__NR_TYPES];
+        enum dso_kernel_type    kernel;
+        u8               adjust_symbols:1;
+        u8               slen_calculated:1;
+        u8               has_build_id:1;
+        u8               hit:1;
+        u8               annotate_warned:1;
+        u8               sname_alloc:1;
+        u8               lname_alloc:1;
+        unsigned char    origin;
+        u8               sorted_by_name;
+        u8               loaded;
+        u8               build_id[BUILD_ID_SIZE];
+        const char       *short_name;
+        char             *long_name;
+        u16              long_name_len;
+        u16              short_name_len;
         char             name[0];
 };
 
-const char *sym_hist_filter;
+struct dso *dso__new(const char *name);
+struct dso *dso__new_kernel(const char *name);
+void dso__delete(struct dso *self);
 
-typedef int (*symbol_filter_t)(struct dso *self, struct symbol *sym);
+int dso__name_len(const struct dso *self);
 
-struct dso *dso__new(const char *name, unsigned int sym_priv_size);
-void dso__delete(struct dso *self);
+bool dso__loaded(const struct dso *self, enum map_type type);
+bool dso__sorted_by_name(const struct dso *self, enum map_type type);
 
-static inline void *dso__sym_priv(struct dso *self, struct symbol *sym)
+static inline void dso__set_loaded(struct dso *self, enum map_type type)
 {
-        return ((void *)sym) - self->sym_priv_size;
+        self->loaded |= (1 << type);
 }
 
-struct symbol *dso__find_symbol(struct dso *self, u64 ip);
+void dso__sort_by_name(struct dso *self, enum map_type type);
+
+struct dso *__dsos__findnew(struct list_head *head, const char *name);
+
+int dso__load(struct dso *self, struct map *map, symbol_filter_t filter);
+int dso__load_vmlinux(struct dso *self, struct map *map,
+                      const char *vmlinux, symbol_filter_t filter);
+int dso__load_vmlinux_path(struct dso *self, struct map *map,
+                           symbol_filter_t filter);
+int dso__load_kallsyms(struct dso *self, const char *filename, struct map *map,
+                       symbol_filter_t filter);
+int machine__load_kallsyms(struct machine *self, const char *filename,
+                           enum map_type type, symbol_filter_t filter);
+int machine__load_vmlinux_path(struct machine *self, enum map_type type,
+                               symbol_filter_t filter);
+
+size_t __dsos__fprintf(struct list_head *head, FILE *fp);
+
+size_t machine__fprintf_dsos_buildid(struct machine *self, FILE *fp, bool with_hits);
+size_t machines__fprintf_dsos(struct rb_root *self, FILE *fp);
+size_t machines__fprintf_dsos_buildid(struct rb_root *self, FILE *fp, bool with_hits);
+
+size_t dso__fprintf_buildid(struct dso *self, FILE *fp);
+size_t dso__fprintf_symbols_by_name(struct dso *self, enum map_type type, FILE *fp);
+size_t dso__fprintf(struct dso *self, enum map_type type, FILE *fp);
+
+enum dso_origin {
+        DSO__ORIG_KERNEL = 0,
+        DSO__ORIG_GUEST_KERNEL,
+        DSO__ORIG_JAVA_JIT,
+        DSO__ORIG_BUILD_ID_CACHE,
+        DSO__ORIG_FEDORA,
+        DSO__ORIG_UBUNTU,
+        DSO__ORIG_BUILDID,
+        DSO__ORIG_DSO,
+        DSO__ORIG_GUEST_KMODULE,
+        DSO__ORIG_KMODULE,
+        DSO__ORIG_NOT_FOUND,
+};
+
+char dso__symtab_origin(const struct dso *self);
+void dso__set_long_name(struct dso *self, char *name);
+void dso__set_build_id(struct dso *self, void *build_id);
+void dso__read_running_kernel_build_id(struct dso *self, struct machine *machine);
+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(struct list_head *head, bool with_hits);
+int build_id__sprintf(const u8 *self, int len, char *bf);
+int kallsyms__parse(const char *filename, void *arg,
+                    int (*process_symbol)(void *arg, const char *name,
+                                          char type, u64 start, u64 end));
+
+void machine__destroy_kernel_maps(struct machine *self);
+int __machine__create_kernel_maps(struct machine *self, struct dso *kernel);
+int machine__create_kernel_maps(struct machine *self);
+
+int machines__create_kernel_maps(struct rb_root *self, pid_t pid);
+int machines__create_guest_kernel_maps(struct rb_root *self);
+void machines__destroy_guest_kernel_maps(struct rb_root *self);
 
-int dso__load_kernel(struct dso *self, const char *vmlinux,
-                     symbol_filter_t filter, int verbose);
-int dso__load(struct dso *self, symbol_filter_t filter, int verbose);
+int symbol__init(void);
+void symbol__exit(void);
+bool symbol_type__is_a(char symbol_type, enum map_type map_type);
 
-size_t dso__fprintf(struct dso *self, FILE *fp);
+size_t machine__fprintf_vmlinux_path(struct machine *self, FILE *fp);
 
-void symbol__init(void);
-#endif /* _PERF_SYMBOL_ */
+#endif /* __PERF_SYMBOL */
diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c
new file mode 100644
index 00000000000..00f4eade2e3
--- /dev/null
+++ b/tools/perf/util/thread.c
@@ -0,0 +1,195 @@
+#include "../perf.h"
+#include <stdlib.h>
+#include <stdio.h>
+#include <string.h>
+#include "session.h"
+#include "thread.h"
+#include "util.h"
+#include "debug.h"
+
+/* Skip "." and ".." directories */
+static int filter(const struct dirent *dir)
+{
+        if (dir->d_name[0] == '.')
+                return 0;
+        else
+                return 1;
+}
+
+struct thread_map *thread_map__new_by_pid(pid_t pid)
+{
+        struct thread_map *threads;
+        char name[256];
+        int items;
+        struct dirent **namelist = NULL;
+        int i;
+
+        sprintf(name, "/proc/%d/task", pid);
+        items = scandir(name, &namelist, filter, NULL);
+        if (items <= 0)
+                return NULL;
+
+        threads = malloc(sizeof(*threads) + sizeof(pid_t) * items);
+        if (threads != NULL) {
+                for (i = 0; i < items; i++)
+                        threads->map[i] = atoi(namelist[i]->d_name);
+                threads->nr = items;
+        }
+
+        for (i=0; i<items; i++)
+                free(namelist[i]);
+        free(namelist);
+
+        return threads;
+}
+
+struct thread_map *thread_map__new_by_tid(pid_t tid)
+{
+        struct thread_map *threads = malloc(sizeof(*threads) + sizeof(pid_t));
+
+        if (threads != NULL) {
+                threads->map[0] = tid;
+                threads->nr     = 1;
+        }
+
+        return threads;
+}
+
+struct thread_map *thread_map__new(pid_t pid, pid_t tid)
+{
+        if (pid != -1)
+                return thread_map__new_by_pid(pid);
+        return thread_map__new_by_tid(tid);
+}
+
+static struct thread *thread__new(pid_t pid)
+{
+        struct thread *self = zalloc(sizeof(*self));
+
+        if (self != NULL) {
+                map_groups__init(&self->mg);
+                self->pid = pid;
+                self->comm = malloc(32);
+                if (self->comm)
+                        snprintf(self->comm, 32, ":%d", self->pid);
+        }
+
+        return self;
+}
+
+void thread__delete(struct thread *self)
+{
+        map_groups__exit(&self->mg);
+        free(self->comm);
+        free(self);
+}
+
+int thread__set_comm(struct thread *self, const char *comm)
+{
+        int err;
+
+        if (self->comm)
+                free(self->comm);
+        self->comm = strdup(comm);
+        err = self->comm == NULL ? -ENOMEM : 0;
+        if (!err) {
+                self->comm_set = true;
+                map_groups__flush(&self->mg);
+        }
+        return err;
+}
+
+int thread__comm_len(struct thread *self)
+{
+        if (!self->comm_len) {
+                if (!self->comm)
+                        return 0;
+                self->comm_len = strlen(self->comm);
+        }
+
+        return self->comm_len;
+}
+
+static size_t thread__fprintf(struct thread *self, FILE *fp)
+{
+        return fprintf(fp, "Thread %d %s\n", self->pid, self->comm) +
+               map_groups__fprintf(&self->mg, verbose, fp);
+}
+
+struct thread *perf_session__findnew(struct perf_session *self, pid_t pid)
+{
+        struct rb_node **p = &self->threads.rb_node;
+        struct rb_node *parent = NULL;
+        struct thread *th;
+
+        /*
+         * Font-end cache - PID lookups come in blocks,
+         * so most of the time we dont have to look up
+         * the full rbtree:
+         */
+        if (self->last_match && self->last_match->pid == pid)
+                return self->last_match;
+
+        while (*p != NULL) {
+                parent = *p;
+                th = rb_entry(parent, struct thread, rb_node);
+
+                if (th->pid == pid) {
+                        self->last_match = th;
+                        return th;
+                }
+
+                if (pid < th->pid)
+                        p = &(*p)->rb_left;
+                else
+                        p = &(*p)->rb_right;
+        }
+
+        th = thread__new(pid);
+        if (th != NULL) {
+                rb_link_node(&th->rb_node, parent, p);
+                rb_insert_color(&th->rb_node, &self->threads);
+                self->last_match = th;
+        }
+
+        return th;
+}
+
+void thread__insert_map(struct thread *self, struct map *map)
+{
+        map_groups__fixup_overlappings(&self->mg, map, verbose, stderr);
+        map_groups__insert(&self->mg, map);
+}
+
+int thread__fork(struct thread *self, struct thread *parent)
+{
+        int i;
+
+        if (parent->comm_set) {
+                if (self->comm)
+                        free(self->comm);
+                self->comm = strdup(parent->comm);
+                if (!self->comm)
+                        return -ENOMEM;
+                self->comm_set = true;
+        }
+
+        for (i = 0; i < MAP__NR_TYPES; ++i)
+                if (map_groups__clone(&self->mg, &parent->mg, i) < 0)
+                        return -ENOMEM;
+        return 0;
+}
+
+size_t perf_session__fprintf(struct perf_session *self, FILE *fp)
+{
+        size_t ret = 0;
+        struct rb_node *nd;
+
+        for (nd = rb_first(&self->threads); nd; nd = rb_next(nd)) {
+                struct thread *pos = rb_entry(nd, struct thread, rb_node);
+
+                ret += thread__fprintf(pos, fp);
+        }
+
+        return ret;
+}
diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h
new file mode 100644
index 00000000000..d7574101054
--- /dev/null
+++ b/tools/perf/util/thread.h
@@ -0,0 +1,62 @@
+#ifndef __PERF_THREAD_H
+#define __PERF_THREAD_H
+
+#include <linux/rbtree.h>
+#include <unistd.h>
+#include "symbol.h"
+
+struct thread {
+        union {
+                struct rb_node   rb_node;
+                struct list_head node;
+        };
+        struct map_groups       mg;
+        pid_t                   pid;
+        char                    shortname[3];
+        bool                    comm_set;
+        char                    *comm;
+        int                     comm_len;
+};
+
+struct thread_map {
+        int nr;
+        int map[];
+};
+
+struct perf_session;
+
+void thread__delete(struct thread *self);
+
+struct thread_map *thread_map__new_by_pid(pid_t pid);
+struct thread_map *thread_map__new_by_tid(pid_t tid);
+struct thread_map *thread_map__new(pid_t pid, pid_t tid);
+
+static inline void thread_map__delete(struct thread_map *threads)
+{
+        free(threads);
+}
+
+int thread__set_comm(struct thread *self, const char *comm);
+int thread__comm_len(struct thread *self);
+struct thread *perf_session__findnew(struct perf_session *self, pid_t pid);
+void thread__insert_map(struct thread *self, struct map *map);
+int thread__fork(struct thread *self, struct thread *parent);
+size_t perf_session__fprintf(struct perf_session *self, FILE *fp);
+
+static inline struct map *thread__find_map(struct thread *self,
+                                           enum map_type type, u64 addr)
+{
+        return self ? map_groups__find(&self->mg, type, addr) : NULL;
+}
+
+void thread__find_addr_map(struct thread *self,
+                           struct perf_session *session, u8 cpumode,
+                           enum map_type type, pid_t pid, u64 addr,
+                           struct addr_location *al);
+
+void thread__find_addr_location(struct thread *self,
+                                struct perf_session *session, u8 cpumode,
+                                enum map_type type, pid_t pid, u64 addr,
+                                struct addr_location *al,
+                                symbol_filter_t filter);
+#endif  /* __PERF_THREAD_H */
diff --git a/tools/perf/util/trace-event-info.c b/tools/perf/util/trace-event-info.c
new file mode 100644
index 00000000000..35729f4c40c
--- /dev/null
+++ b/tools/perf/util/trace-event-info.c
@@ -0,0 +1,565 @@
+/*
+ * Copyright (C) 2008,2009, Steven Rostedt <srostedt@redhat.com>
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License (not later!)
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ */
+#define _GNU_SOURCE
+#include <dirent.h>
+#include <mntent.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdarg.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/wait.h>
+#include <pthread.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include <ctype.h>
+#include <errno.h>
+#include <stdbool.h>
+#include <linux/list.h>
+#include <linux/kernel.h>
+
+#include "../perf.h"
+#include "trace-event.h"
+#include "debugfs.h"
+#include "evsel.h"
+
+#define VERSION "0.5"
+
+#define _STR(x) #x
+#define STR(x) _STR(x)
+#define MAX_PATH 256
+
+#define TRACE_CTRL      "tracing_on"
+#define TRACE           "trace"
+#define AVAILABLE       "available_tracers"
+#define CURRENT         "current_tracer"
+#define ITER_CTRL       "trace_options"
+#define MAX_LATENCY     "tracing_max_latency"
+
+unsigned int page_size;
+
+static const char *output_file = "trace.info";
+static int output_fd;
+
+struct event_list {
+        struct event_list *next;
+        const char *event;
+};
+
+struct events {
+        struct events *sibling;
+        struct events *children;
+        struct events *next;
+        char *name;
+};
+
+
+
+static void die(const char *fmt, ...)
+{
+        va_list ap;
+        int ret = errno;
+
+        if (errno)
+                perror("trace-cmd");
+        else
+                ret = -1;
+
+        va_start(ap, fmt);
+        fprintf(stderr, "  ");
+        vfprintf(stderr, fmt, ap);
+        va_end(ap);
+
+        fprintf(stderr, "\n");
+        exit(ret);
+}
+
+void *malloc_or_die(unsigned int size)
+{
+        void *data;
+
+        data = malloc(size);
+        if (!data)
+                die("malloc");
+        return data;
+}
+
+static const char *find_debugfs(void)
+{
+        const char *path = debugfs_mount(NULL);
+
+        if (!path)
+                die("Your kernel not support debugfs filesystem");
+
+        return path;
+}
+
+/*
+ * Finds the path to the debugfs/tracing
+ * Allocates the string and stores it.
+ */
+static const char *find_tracing_dir(void)
+{
+        static char *tracing;
+        static int tracing_found;
+        const char *debugfs;
+
+        if (tracing_found)
+                return tracing;
+
+        debugfs = find_debugfs();
+
+        tracing = malloc_or_die(strlen(debugfs) + 9);
+
+        sprintf(tracing, "%s/tracing", debugfs);
+
+        tracing_found = 1;
+        return tracing;
+}
+
+static char *get_tracing_file(const char *name)
+{
+        const char *tracing;
+        char *file;
+
+        tracing = find_tracing_dir();
+        if (!tracing)
+                return NULL;
+
+        file = malloc_or_die(strlen(tracing) + strlen(name) + 2);
+
+        sprintf(file, "%s/%s", tracing, name);
+        return file;
+}
+
+static void put_tracing_file(char *file)
+{
+        free(file);
+}
+
+static ssize_t calc_data_size;
+
+static ssize_t write_or_die(const void *buf, size_t len)
+{
+        int ret;
+
+        if (calc_data_size) {
+                calc_data_size += len;
+                return len;
+        }
+
+        ret = write(output_fd, buf, len);
+        if (ret < 0)
+                die("writing to '%s'", output_file);
+
+        return ret;
+}
+
+int bigendian(void)
+{
+        unsigned char str[] = { 0x1, 0x2, 0x3, 0x4, 0x0, 0x0, 0x0, 0x0};
+        unsigned int *ptr;
+
+        ptr = (unsigned int *)(void *)str;
+        return *ptr == 0x01020304;
+}
+
+static unsigned long long copy_file_fd(int fd)
+{
+        unsigned long long size = 0;
+        char buf[BUFSIZ];
+        int r;
+
+        do {
+                r = read(fd, buf, BUFSIZ);
+                if (r > 0) {
+                        size += r;
+                        write_or_die(buf, r);
+                }
+        } while (r > 0);
+
+        return size;
+}
+
+static unsigned long long copy_file(const char *file)
+{
+        unsigned long long size = 0;
+        int fd;
+
+        fd = open(file, O_RDONLY);
+        if (fd < 0)
+                die("Can't read '%s'", file);
+        size = copy_file_fd(fd);
+        close(fd);
+
+        return size;
+}
+
+static unsigned long get_size_fd(int fd)
+{
+        unsigned long long size = 0;
+        char buf[BUFSIZ];
+        int r;
+
+        do {
+                r = read(fd, buf, BUFSIZ);
+                if (r > 0)
+                        size += r;
+        } while (r > 0);
+
+        lseek(fd, 0, SEEK_SET);
+
+        return size;
+}
+
+static unsigned long get_size(const char *file)
+{
+        unsigned long long size = 0;
+        int fd;
+
+        fd = open(file, O_RDONLY);
+        if (fd < 0)
+                die("Can't read '%s'", file);
+        size = get_size_fd(fd);
+        close(fd);
+
+        return size;
+}
+
+static void read_header_files(void)
+{
+        unsigned long long size, check_size;
+        char *path;
+        int fd;
+
+        path = get_tracing_file("events/header_page");
+        fd = open(path, O_RDONLY);
+        if (fd < 0)
+                die("can't read '%s'", path);
+
+        /* unfortunately, you can not stat debugfs files for size */
+        size = get_size_fd(fd);
+
+        write_or_die("header_page", 12);
+        write_or_die(&size, 8);
+        check_size = copy_file_fd(fd);
+        close(fd);
+
+        if (size != check_size)
+                die("wrong size for '%s' size=%lld read=%lld",
+                    path, size, check_size);
+        put_tracing_file(path);
+
+        path = get_tracing_file("events/header_event");
+        fd = open(path, O_RDONLY);
+        if (fd < 0)
+                die("can't read '%s'", path);
+
+        size = get_size_fd(fd);
+
+        write_or_die("header_event", 13);
+        write_or_die(&size, 8);
+        check_size = copy_file_fd(fd);
+        if (size != check_size)
+                die("wrong size for '%s'", path);
+        put_tracing_file(path);
+        close(fd);
+}
+
+static bool name_in_tp_list(char *sys, struct tracepoint_path *tps)
+{
+        while (tps) {
+                if (!strcmp(sys, tps->name))
+                        return true;
+                tps = tps->next;
+        }
+
+        return false;
+}
+
+static void copy_event_system(const char *sys, struct tracepoint_path *tps)
+{
+        unsigned long long size, check_size;
+        struct dirent *dent;
+        struct stat st;
+        char *format;
+        DIR *dir;
+        int count = 0;
+        int ret;
+
+        dir = opendir(sys);
+        if (!dir)
+                die("can't read directory '%s'", sys);
+
+        while ((dent = readdir(dir))) {
+                if (dent->d_type != DT_DIR ||
+                    strcmp(dent->d_name, ".") == 0 ||
+                    strcmp(dent->d_name, "..") == 0 ||
+                    !name_in_tp_list(dent->d_name, tps))
+                        continue;
+                format = malloc_or_die(strlen(sys) + strlen(dent->d_name) + 10);
+                sprintf(format, "%s/%s/format", sys, dent->d_name);
+                ret = stat(format, &st);
+                free(format);
+                if (ret < 0)
+                        continue;
+                count++;
+        }
+
+        write_or_die(&count, 4);
+
+        rewinddir(dir);
+        while ((dent = readdir(dir))) {
+                if (dent->d_type != DT_DIR ||
+                    strcmp(dent->d_name, ".") == 0 ||
+                    strcmp(dent->d_name, "..") == 0 ||
+                    !name_in_tp_list(dent->d_name, tps))
+                        continue;
+                format = malloc_or_die(strlen(sys) + strlen(dent->d_name) + 10);
+                sprintf(format, "%s/%s/format", sys, dent->d_name);
+                ret = stat(format, &st);
+
+                if (ret >= 0) {
+                        /* unfortunately, you can not stat debugfs files for size */
+                        size = get_size(format);
+                        write_or_die(&size, 8);
+                        check_size = copy_file(format);
+                        if (size != check_size)
+                                die("error in size of file '%s'", format);
+                }
+
+                free(format);
+        }
+        closedir(dir);
+}
+
+static void read_ftrace_files(struct tracepoint_path *tps)
+{
+        char *path;
+
+        path = get_tracing_file("events/ftrace");
+
+        copy_event_system(path, tps);
+
+        put_tracing_file(path);
+}
+
+static bool system_in_tp_list(char *sys, struct tracepoint_path *tps)
+{
+        while (tps) {
+                if (!strcmp(sys, tps->system))
+                        return true;
+                tps = tps->next;
+        }
+
+        return false;
+}
+
+static void read_event_files(struct tracepoint_path *tps)
+{
+        struct dirent *dent;
+        struct stat st;
+        char *path;
+        char *sys;
+        DIR *dir;
+        int count = 0;
+        int ret;
+
+        path = get_tracing_file("events");
+
+        dir = opendir(path);
+        if (!dir)
+                die("can't read directory '%s'", path);
+
+        while ((dent = readdir(dir))) {
+                if (dent->d_type != DT_DIR ||
+                    strcmp(dent->d_name, ".") == 0 ||
+                    strcmp(dent->d_name, "..") == 0 ||
+                    strcmp(dent->d_name, "ftrace") == 0 ||
+                    !system_in_tp_list(dent->d_name, tps))
+                        continue;
+                count++;
+        }
+
+        write_or_die(&count, 4);
+
+        rewinddir(dir);
+        while ((dent = readdir(dir))) {
+                if (dent->d_type != DT_DIR ||
+                    strcmp(dent->d_name, ".") == 0 ||
+                    strcmp(dent->d_name, "..") == 0 ||
+                    strcmp(dent->d_name, "ftrace") == 0 ||
+                    !system_in_tp_list(dent->d_name, tps))
+                        continue;
+                sys = malloc_or_die(strlen(path) + strlen(dent->d_name) + 2);
+                sprintf(sys, "%s/%s", path, dent->d_name);
+                ret = stat(sys, &st);
+                if (ret >= 0) {
+                        write_or_die(dent->d_name, strlen(dent->d_name) + 1);
+                        copy_event_system(sys, tps);
+                }
+                free(sys);
+        }
+
+        closedir(dir);
+        put_tracing_file(path);
+}
+
+static void read_proc_kallsyms(void)
+{
+        unsigned int size, check_size;
+        const char *path = "/proc/kallsyms";
+        struct stat st;
+        int ret;
+
+        ret = stat(path, &st);
+        if (ret < 0) {
+                /* not found */
+                size = 0;
+                write_or_die(&size, 4);
+                return;
+        }
+        size = get_size(path);
+        write_or_die(&size, 4);
+        check_size = copy_file(path);
+        if (size != check_size)
+                die("error in size of file '%s'", path);
+
+}
+
+static void read_ftrace_printk(void)
+{
+        unsigned int size, check_size;
+        char *path;
+        struct stat st;
+        int ret;
+
+        path = get_tracing_file("printk_formats");
+        ret = stat(path, &st);
+        if (ret < 0) {
+                /* not found */
+                size = 0;
+                write_or_die(&size, 4);
+                goto out;
+        }
+        size = get_size(path);
+        write_or_die(&size, 4);
+        check_size = copy_file(path);
+        if (size != check_size)
+                die("error in size of file '%s'", path);
+out:
+        put_tracing_file(path);
+}
+
+static struct tracepoint_path *
+get_tracepoints_path(struct list_head *pattrs)
+{
+        struct tracepoint_path path, *ppath = &path;
+        struct perf_evsel *pos;
+        int nr_tracepoints = 0;
+
+        list_for_each_entry(pos, pattrs, node) {
+                if (pos->attr.type != PERF_TYPE_TRACEPOINT)
+                        continue;
+                ++nr_tracepoints;
+                ppath->next = tracepoint_id_to_path(pos->attr.config);
+                if (!ppath->next)
+                        die("%s\n", "No memory to alloc tracepoints list");
+                ppath = ppath->next;
+        }
+
+        return nr_tracepoints > 0 ? path.next : NULL;
+}
+
+bool have_tracepoints(struct list_head *pattrs)
+{
+        struct perf_evsel *pos;
+
+        list_for_each_entry(pos, pattrs, node)
+                if (pos->attr.type == PERF_TYPE_TRACEPOINT)
+                        return true;
+
+        return false;
+}
+
+int read_tracing_data(int fd, struct list_head *pattrs)
+{
+        char buf[BUFSIZ];
+        struct tracepoint_path *tps = get_tracepoints_path(pattrs);
+
+        /*
+         * What? No tracepoints? No sense writing anything here, bail out.
+         */
+        if (tps == NULL)
+                return -1;
+
+        output_fd = fd;
+
+        buf[0] = 23;
+        buf[1] = 8;
+        buf[2] = 68;
+        memcpy(buf + 3, "tracing", 7);
+
+        write_or_die(buf, 10);
+
+        write_or_die(VERSION, strlen(VERSION) + 1);
+
+        /* save endian */
+        if (bigendian())
+                buf[0] = 1;
+        else
+                buf[0] = 0;
+
+        write_or_die(buf, 1);
+
+        /* save size of long */
+        buf[0] = sizeof(long);
+        write_or_die(buf, 1);
+
+        /* save page_size */
+        page_size = sysconf(_SC_PAGESIZE);
+        write_or_die(&page_size, 4);
+
+        read_header_files();
+        read_ftrace_files(tps);
+        read_event_files(tps);
+        read_proc_kallsyms();
+        read_ftrace_printk();
+
+        return 0;
+}
+
+ssize_t read_tracing_data_size(int fd, struct list_head *pattrs)
+{
+        ssize_t size;
+        int err = 0;
+
+        calc_data_size = 1;
+        err = read_tracing_data(fd, pattrs);
+        size = calc_data_size - 1;
+        calc_data_size = 0;
+
+        if (err < 0)
+                return err;
+
+        return size;
+}
diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c
new file mode 100644
index 00000000000..73a02223c62
--- /dev/null
+++ b/tools/perf/util/trace-event-parse.c
@@ -0,0 +1,3233 @@
+/*
+ * Copyright (C) 2009, Steven Rostedt <srostedt@redhat.com>
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License (not later!)
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ *
+ *  The parts for function graph printing was taken and modified from the
+ *  Linux Kernel that were written by Frederic Weisbecker.
+ */
+#define _GNU_SOURCE
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+#include <errno.h>
+
+#undef _GNU_SOURCE
+#include "../perf.h"
+#include "util.h"
+#include "trace-event.h"
+
+int header_page_ts_offset;
+int header_page_ts_size;
+int header_page_size_offset;
+int header_page_size_size;
+int header_page_overwrite_offset;
+int header_page_overwrite_size;
+int header_page_data_offset;
+int header_page_data_size;
+
+bool latency_format;
+
+static char *input_buf;
+static unsigned long long input_buf_ptr;
+static unsigned long long input_buf_siz;
+
+static int cpus;
+static int long_size;
+static int is_flag_field;
+static int is_symbolic_field;
+
+static struct format_field *
+find_any_field(struct event *event, const char *name);
+
+static void init_input_buf(char *buf, unsigned long long size)
+{
+        input_buf = buf;
+        input_buf_siz = size;
+        input_buf_ptr = 0;
+}
+
+struct cmdline {
+        char *comm;
+        int pid;
+};
+
+static struct cmdline *cmdlines;
+static int cmdline_count;
+
+static int cmdline_cmp(const void *a, const void *b)
+{
+        const struct cmdline *ca = a;
+        const struct cmdline *cb = b;
+
+        if (ca->pid < cb->pid)
+                return -1;
+        if (ca->pid > cb->pid)
+                return 1;
+
+        return 0;
+}
+
+void parse_cmdlines(char *file, int size __unused)
+{
+        struct cmdline_list {
+                struct cmdline_list     *next;
+                char                    *comm;
+                int                     pid;
+        } *list = NULL, *item;
+        char *line;
+        char *next = NULL;
+        int i;
+
+        line = strtok_r(file, "\n", &next);
+        while (line) {
+                item = malloc_or_die(sizeof(*item));
+                sscanf(line, "%d %as", &item->pid,
+                       (float *)(void *)&item->comm); /* workaround gcc warning */
+                item->next = list;
+                list = item;
+                line = strtok_r(NULL, "\n", &next);
+                cmdline_count++;
+        }
+
+        cmdlines = malloc_or_die(sizeof(*cmdlines) * cmdline_count);
+
+        i = 0;
+        while (list) {
+                cmdlines[i].pid = list->pid;
+                cmdlines[i].comm = list->comm;
+                i++;
+                item = list;
+                list = list->next;
+                free(item);
+        }
+
+        qsort(cmdlines, cmdline_count, sizeof(*cmdlines), cmdline_cmp);
+}
+
+static struct func_map {
+        unsigned long long              addr;
+        char                            *func;
+        char                            *mod;
+} *func_list;
+static unsigned int func_count;
+
+static int func_cmp(const void *a, const void *b)
+{
+        const struct func_map *fa = a;
+        const struct func_map *fb = b;
+
+        if (fa->addr < fb->addr)
+                return -1;
+        if (fa->addr > fb->addr)
+                return 1;
+
+        return 0;
+}
+
+void parse_proc_kallsyms(char *file, unsigned int size __unused)
+{
+        struct func_list {
+                struct func_list        *next;
+                unsigned long long      addr;
+                char                    *func;
+                char                    *mod;
+        } *list = NULL, *item;
+        char *line;
+        char *next = NULL;
+        char *addr_str;
+        char ch;
+        int ret;
+        int i;
+
+        line = strtok_r(file, "\n", &next);
+        while (line) {
+                item = malloc_or_die(sizeof(*item));
+                item->mod = NULL;
+                ret = sscanf(line, "%as %c %as\t[%as",
+                             (float *)(void *)&addr_str, /* workaround gcc warning */
+                             &ch,
+                             (float *)(void *)&item->func,
+                             (float *)(void *)&item->mod);
+                item->addr = strtoull(addr_str, NULL, 16);
+                free(addr_str);
+
+                /* truncate the extra ']' */
+                if (item->mod)
+                        item->mod[strlen(item->mod) - 1] = 0;
+
+
+                item->next = list;
+                list = item;
+                line = strtok_r(NULL, "\n", &next);
+                func_count++;
+        }
+
+        func_list = malloc_or_die(sizeof(*func_list) * (func_count + 1));
+
+        i = 0;
+        while (list) {
+                func_list[i].func = list->func;
+                func_list[i].addr = list->addr;
+                func_list[i].mod = list->mod;
+                i++;
+                item = list;
+                list = list->next;
+                free(item);
+        }
+
+        qsort(func_list, func_count, sizeof(*func_list), func_cmp);
+
+        /*
+         * Add a special record at the end.
+         */
+        func_list[func_count].func = NULL;
+        func_list[func_count].addr = 0;
+        func_list[func_count].mod = NULL;
+}
+
+/*
+ * We are searching for a record in between, not an exact
+ * match.
+ */
+static int func_bcmp(const void *a, const void *b)
+{
+        const struct func_map *fa = a;
+        const struct func_map *fb = b;
+
+        if ((fa->addr == fb->addr) ||
+
+            (fa->addr > fb->addr &&
+             fa->addr < (fb+1)->addr))
+                return 0;
+
+        if (fa->addr < fb->addr)
+                return -1;
+
+        return 1;
+}
+
+static struct func_map *find_func(unsigned long long addr)
+{
+        struct func_map *func;
+        struct func_map key;
+
+        key.addr = addr;
+
+        func = bsearch(&key, func_list, func_count, sizeof(*func_list),
+                       func_bcmp);
+
+        return func;
+}
+
+void print_funcs(void)
+{
+        int i;
+
+        for (i = 0; i < (int)func_count; i++) {
+                printf("%016llx %s",
+                       func_list[i].addr,
+                       func_list[i].func);
+                if (func_list[i].mod)
+                        printf(" [%s]\n", func_list[i].mod);
+                else
+                        printf("\n");
+        }
+}
+
+static struct printk_map {
+        unsigned long long              addr;
+        char                            *printk;
+} *printk_list;
+static unsigned int printk_count;
+
+static int printk_cmp(const void *a, const void *b)
+{
+        const struct func_map *fa = a;
+        const struct func_map *fb = b;
+
+        if (fa->addr < fb->addr)
+                return -1;
+        if (fa->addr > fb->addr)
+                return 1;
+
+        return 0;
+}
+
+static struct printk_map *find_printk(unsigned long long addr)
+{
+        struct printk_map *printk;
+        struct printk_map key;
+
+        key.addr = addr;
+
+        printk = bsearch(&key, printk_list, printk_count, sizeof(*printk_list),
+                         printk_cmp);
+
+        return printk;
+}
+
+void parse_ftrace_printk(char *file, unsigned int size __unused)
+{
+        struct printk_list {
+                struct printk_list      *next;
+                unsigned long long      addr;
+                char                    *printk;
+        } *list = NULL, *item;
+        char *line;
+        char *next = NULL;
+        char *addr_str;
+        int i;
+
+        line = strtok_r(file, "\n", &next);
+        while (line) {
+                addr_str = strsep(&line, ":");
+                if (!line) {
+                        warning("error parsing print strings");
+                        break;
+                }
+                item = malloc_or_die(sizeof(*item));
+                item->addr = strtoull(addr_str, NULL, 16);
+                /* fmt still has a space, skip it */
+                item->printk = strdup(line+1);
+                item->next = list;
+                list = item;
+                line = strtok_r(NULL, "\n", &next);
+                printk_count++;
+        }
+
+        printk_list = malloc_or_die(sizeof(*printk_list) * printk_count + 1);
+
+        i = 0;
+        while (list) {
+                printk_list[i].printk = list->printk;
+                printk_list[i].addr = list->addr;
+                i++;
+                item = list;
+                list = list->next;
+                free(item);
+        }
+
+        qsort(printk_list, printk_count, sizeof(*printk_list), printk_cmp);
+}
+
+void print_printk(void)
+{
+        int i;
+
+        for (i = 0; i < (int)printk_count; i++) {
+                printf("%016llx %s\n",
+                       printk_list[i].addr,
+                       printk_list[i].printk);
+        }
+}
+
+static struct event *alloc_event(void)
+{
+        struct event *event;
+
+        event = malloc_or_die(sizeof(*event));
+        memset(event, 0, sizeof(*event));
+
+        return event;
+}
+
+enum event_type {
+        EVENT_ERROR,
+        EVENT_NONE,
+        EVENT_SPACE,
+        EVENT_NEWLINE,
+        EVENT_OP,
+        EVENT_DELIM,
+        EVENT_ITEM,
+        EVENT_DQUOTE,
+        EVENT_SQUOTE,
+};
+
+static struct event *event_list;
+
+static void add_event(struct event *event)
+{
+        event->next = event_list;
+        event_list = event;
+}
+
+static int event_item_type(enum event_type type)
+{
+        switch (type) {
+        case EVENT_ITEM ... EVENT_SQUOTE:
+                return 1;
+        case EVENT_ERROR ... EVENT_DELIM:
+        default:
+                return 0;
+        }
+}
+
+static void free_arg(struct print_arg *arg)
+{
+        if (!arg)
+                return;
+
+        switch (arg->type) {
+        case PRINT_ATOM:
+                if (arg->atom.atom)
+                        free(arg->atom.atom);
+                break;
+        case PRINT_NULL:
+        case PRINT_FIELD ... PRINT_OP:
+        default:
+                /* todo */
+                break;
+        }
+
+        free(arg);
+}
+
+static enum event_type get_type(int ch)
+{
+        if (ch == '\n')
+                return EVENT_NEWLINE;
+        if (isspace(ch))
+                return EVENT_SPACE;
+        if (isalnum(ch) || ch == '_')
+                return EVENT_ITEM;
+        if (ch == '\'')
+                return EVENT_SQUOTE;
+        if (ch == '"')
+                return EVENT_DQUOTE;
+        if (!isprint(ch))
+                return EVENT_NONE;
+        if (ch == '(' || ch == ')' || ch == ',')
+                return EVENT_DELIM;
+
+        return EVENT_OP;
+}
+
+static int __read_char(void)
+{
+        if (input_buf_ptr >= input_buf_siz)
+                return -1;
+
+        return input_buf[input_buf_ptr++];
+}
+
+static int __peek_char(void)
+{
+        if (input_buf_ptr >= input_buf_siz)
+                return -1;
+
+        return input_buf[input_buf_ptr];
+}
+
+static enum event_type __read_token(char **tok)
+{
+        char buf[BUFSIZ];
+        int ch, last_ch, quote_ch, next_ch;
+        int i = 0;
+        int tok_size = 0;
+        enum event_type type;
+
+        *tok = NULL;
+
+
+        ch = __read_char();
+        if (ch < 0)
+                return EVENT_NONE;
+
+        type = get_type(ch);
+        if (type == EVENT_NONE)
+                return type;
+
+        buf[i++] = ch;
+
+        switch (type) {
+        case EVENT_NEWLINE:
+        case EVENT_DELIM:
+                *tok = malloc_or_die(2);
+                (*tok)[0] = ch;
+                (*tok)[1] = 0;
+                return type;
+
+        case EVENT_OP:
+                switch (ch) {
+                case '-':
+                        next_ch = __peek_char();
+                        if (next_ch == '>') {
+                                buf[i++] = __read_char();
+                                break;
+                        }
+                        /* fall through */
+                case '+':
+                case '|':
+                case '&':
+                case '>':
+                case '<':
+                        last_ch = ch;
+                        ch = __peek_char();
+                        if (ch != last_ch)
+                                goto test_equal;
+                        buf[i++] = __read_char();
+                        switch (last_ch) {
+                        case '>':
+                        case '<':
+                                goto test_equal;
+                        default:
+                                break;
+                        }
+                        break;
+                case '!':
+                case '=':
+                        goto test_equal;
+                default: /* what should we do instead? */
+                        break;
+                }
+                buf[i] = 0;
+                *tok = strdup(buf);
+                return type;
+
+ test_equal:
+                ch = __peek_char();
+                if (ch == '=')
+                        buf[i++] = __read_char();
+                break;
+
+        case EVENT_DQUOTE:
+        case EVENT_SQUOTE:
+                /* don't keep quotes */
+                i--;
+                quote_ch = ch;
+                last_ch = 0;
+                do {
+                        if (i == (BUFSIZ - 1)) {
+                                buf[i] = 0;
+                                if (*tok) {
+                                        *tok = realloc(*tok, tok_size + BUFSIZ);
+                                        if (!*tok)
+                                                return EVENT_NONE;
+                                        strcat(*tok, buf);
+                                } else
+                                        *tok = strdup(buf);
+
+                                if (!*tok)
+                                        return EVENT_NONE;
+                                tok_size += BUFSIZ;
+                                i = 0;
+                        }
+                        last_ch = ch;
+                        ch = __read_char();
+                        buf[i++] = ch;
+                        /* the '\' '\' will cancel itself */
+                        if (ch == '\\' && last_ch == '\\')
+                                last_ch = 0;
+                } while (ch != quote_ch || last_ch == '\\');
+                /* remove the last quote */
+                i--;
+                goto out;
+
+        case EVENT_ERROR ... EVENT_SPACE:
+        case EVENT_ITEM:
+        default:
+                break;
+        }
+
+        while (get_type(__peek_char()) == type) {
+                if (i == (BUFSIZ - 1)) {
+                        buf[i] = 0;
+                        if (*tok) {
+                                *tok = realloc(*tok, tok_size + BUFSIZ);
+                                if (!*tok)
+                                        return EVENT_NONE;
+                                strcat(*tok, buf);
+                        } else
+                                *tok = strdup(buf);
+
+                        if (!*tok)
+                                return EVENT_NONE;
+                        tok_size += BUFSIZ;
+                        i = 0;
+                }
+                ch = __read_char();
+                buf[i++] = ch;
+        }
+
+ out:
+        buf[i] = 0;
+        if (*tok) {
+                *tok = realloc(*tok, tok_size + i);
+                if (!*tok)
+                        return EVENT_NONE;
+                strcat(*tok, buf);
+        } else
+                *tok = strdup(buf);
+        if (!*tok)
+                return EVENT_NONE;
+
+        return type;
+}
+
+static void free_token(char *tok)
+{
+        if (tok)
+                free(tok);
+}
+
+static enum event_type read_token(char **tok)
+{
+        enum event_type type;
+
+        for (;;) {
+                type = __read_token(tok);
+                if (type != EVENT_SPACE)
+                        return type;
+
+                free_token(*tok);
+        }
+
+        /* not reached */
+        return EVENT_NONE;
+}
+
+/* no newline */
+static enum event_type read_token_item(char **tok)
+{
+        enum event_type type;
+
+        for (;;) {
+                type = __read_token(tok);
+                if (type != EVENT_SPACE && type != EVENT_NEWLINE)
+                        return type;
+
+                free_token(*tok);
+        }
+
+        /* not reached */
+        return EVENT_NONE;
+}
+
+static int test_type(enum event_type type, enum event_type expect)
+{
+        if (type != expect) {
+                warning("Error: expected type %d but read %d",
+                    expect, type);
+                return -1;
+        }
+        return 0;
+}
+
+static int __test_type_token(enum event_type type, char *token,
+                             enum event_type expect, const char *expect_tok,
+                             bool warn)
+{
+        if (type != expect) {
+                if (warn)
+                        warning("Error: expected type %d but read %d",
+                                expect, type);
+                return -1;
+        }
+
+        if (strcmp(token, expect_tok) != 0) {
+                if (warn)
+                        warning("Error: expected '%s' but read '%s'",
+                                expect_tok, token);
+                return -1;
+        }
+        return 0;
+}
+
+static int test_type_token(enum event_type type, char *token,
+                           enum event_type expect, const char *expect_tok)
+{
+        return __test_type_token(type, token, expect, expect_tok, true);
+}
+
+static int __read_expect_type(enum event_type expect, char **tok, int newline_ok)
+{
+        enum event_type type;
+
+        if (newline_ok)
+                type = read_token(tok);
+        else
+                type = read_token_item(tok);
+        return test_type(type, expect);
+}
+
+static int read_expect_type(enum event_type expect, char **tok)
+{
+        return __read_expect_type(expect, tok, 1);
+}
+
+static int __read_expected(enum event_type expect, const char *str,
+                           int newline_ok, bool warn)
+{
+        enum event_type type;
+        char *token;
+        int ret;
+
+        if (newline_ok)
+                type = read_token(&token);
+        else
+                type = read_token_item(&token);
+
+        ret = __test_type_token(type, token, expect, str, warn);
+
+        free_token(token);
+
+        return ret;
+}
+
+static int read_expected(enum event_type expect, const char *str)
+{
+        return __read_expected(expect, str, 1, true);
+}
+
+static int read_expected_item(enum event_type expect, const char *str)
+{
+        return __read_expected(expect, str, 0, true);
+}
+
+static char *event_read_name(void)
+{
+        char *token;
+
+        if (read_expected(EVENT_ITEM, "name") < 0)
+                return NULL;
+
+        if (read_expected(EVENT_OP, ":") < 0)
+                return NULL;
+
+        if (read_expect_type(EVENT_ITEM, &token) < 0)
+                goto fail;
+
+        return token;
+
+ fail:
+        free_token(token);
+        return NULL;
+}
+
+static int event_read_id(void)
+{
+        char *token;
+        int id;
+
+        if (read_expected_item(EVENT_ITEM, "ID") < 0)
+                return -1;
+
+        if (read_expected(EVENT_OP, ":") < 0)
+                return -1;
+
+        if (read_expect_type(EVENT_ITEM, &token) < 0)
+                goto fail;
+
+        id = strtoul(token, NULL, 0);
+        free_token(token);
+        return id;
+
+ fail:
+        free_token(token);
+        return -1;
+}
+
+static int field_is_string(struct format_field *field)
+{
+        if ((field->flags & FIELD_IS_ARRAY) &&
+            (!strstr(field->type, "char") || !strstr(field->type, "u8") ||
+             !strstr(field->type, "s8")))
+                return 1;
+
+        return 0;
+}
+
+static int field_is_dynamic(struct format_field *field)
+{
+        if (!strncmp(field->type, "__data_loc", 10))
+                return 1;
+
+        return 0;
+}
+
+static int event_read_fields(struct event *event, struct format_field **fields)
+{
+        struct format_field *field = NULL;
+        enum event_type type;
+        char *token;
+        char *last_token;
+        int count = 0;
+
+        do {
+                type = read_token(&token);
+                if (type == EVENT_NEWLINE) {
+                        free_token(token);
+                        return count;
+                }
+
+                count++;
+
+                if (test_type_token(type, token, EVENT_ITEM, "field"))
+                        goto fail;
+                free_token(token);
+
+                type = read_token(&token);
+                /*
+                 * The ftrace fields may still use the "special" name.
+                 * Just ignore it.
+                 */
+                if (event->flags & EVENT_FL_ISFTRACE &&
+                    type == EVENT_ITEM && strcmp(token, "special") == 0) {
+                        free_token(token);
+                        type = read_token(&token);
+                }
+
+                if (test_type_token(type, token, EVENT_OP, ":") < 0)
+                        return -1;
+
+                if (read_expect_type(EVENT_ITEM, &token) < 0)
+                        goto fail;
+
+                last_token = token;
+
+                field = malloc_or_die(sizeof(*field));
+                memset(field, 0, sizeof(*field));
+
+                /* read the rest of the type */
+                for (;;) {
+                        type = read_token(&token);
+                        if (type == EVENT_ITEM ||
+                            (type == EVENT_OP && strcmp(token, "*") == 0) ||
+                            /*
+                             * Some of the ftrace fields are broken and have
+                             * an illegal "." in them.
+                             */
+                            (event->flags & EVENT_FL_ISFTRACE &&
+                             type == EVENT_OP && strcmp(token, ".") == 0)) {
+
+                                if (strcmp(token, "*") == 0)
+                                        field->flags |= FIELD_IS_POINTER;
+
+                                if (field->type) {
+                                        field->type = realloc(field->type,
+                                                              strlen(field->type) +
+                                                              strlen(last_token) + 2);
+                                        strcat(field->type, " ");
+                                        strcat(field->type, last_token);
+                                } else
+                                        field->type = last_token;
+                                last_token = token;
+                                continue;
+                        }
+
+                        break;
+                }
+
+                if (!field->type) {
+                        die("no type found");
+                        goto fail;
+                }
+                field->name = last_token;
+
+                if (test_type(type, EVENT_OP))
+                        goto fail;
+
+                if (strcmp(token, "[") == 0) {
+                        enum event_type last_type = type;
+                        char *brackets = token;
+                        int len;
+
+                        field->flags |= FIELD_IS_ARRAY;
+
+                        type = read_token(&token);
+                        while (strcmp(token, "]") != 0) {
+                                if (last_type == EVENT_ITEM &&
+                                    type == EVENT_ITEM)
+                                        len = 2;
+                                else
+                                        len = 1;
+                                last_type = type;
+
+                                brackets = realloc(brackets,
+                                                   strlen(brackets) +
+                                                   strlen(token) + len);
+                                if (len == 2)
+                                        strcat(brackets, " ");
+                                strcat(brackets, token);
+                                free_token(token);
+                                type = read_token(&token);
+                                if (type == EVENT_NONE) {
+                                        die("failed to find token");
+                                        goto fail;
+                                }
+                        }
+
+                        free_token(token);
+
+                        brackets = realloc(brackets, strlen(brackets) + 2);
+                        strcat(brackets, "]");
+
+                        /* add brackets to type */
+
+                        type = read_token(&token);
+                        /*
+                         * If the next token is not an OP, then it is of
+                         * the format: type [] item;
+                         */
+                        if (type == EVENT_ITEM) {
+                                field->type = realloc(field->type,
+                                                      strlen(field->type) +
+                                                      strlen(field->name) +
+                                                      strlen(brackets) + 2);
+                                strcat(field->type, " ");
+                                strcat(field->type, field->name);
+                                free_token(field->name);
+                                strcat(field->type, brackets);
+                                field->name = token;
+                                type = read_token(&token);
+                        } else {
+                                field->type = realloc(field->type,
+                                                      strlen(field->type) +
+                                                      strlen(brackets) + 1);
+                                strcat(field->type, brackets);
+                        }
+                        free(brackets);
+                }
+
+                if (field_is_string(field)) {
+                        field->flags |= FIELD_IS_STRING;
+                        if (field_is_dynamic(field))
+                                field->flags |= FIELD_IS_DYNAMIC;
+                }
+
+                if (test_type_token(type, token,  EVENT_OP, ";"))
+                        goto fail;
+                free_token(token);
+
+                if (read_expected(EVENT_ITEM, "offset") < 0)
+                        goto fail_expect;
+
+                if (read_expected(EVENT_OP, ":") < 0)
+                        goto fail_expect;
+
+                if (read_expect_type(EVENT_ITEM, &token))
+                        goto fail;
+                field->offset = strtoul(token, NULL, 0);
+                free_token(token);
+
+                if (read_expected(EVENT_OP, ";") < 0)
+                        goto fail_expect;
+
+                if (read_expected(EVENT_ITEM, "size") < 0)
+                        goto fail_expect;
+
+                if (read_expected(EVENT_OP, ":") < 0)
+                        goto fail_expect;
+
+                if (read_expect_type(EVENT_ITEM, &token))
+                        goto fail;
+                field->size = strtoul(token, NULL, 0);
+                free_token(token);
+
+                if (read_expected(EVENT_OP, ";") < 0)
+                        goto fail_expect;
+
+                type = read_token(&token);
+                if (type != EVENT_NEWLINE) {
+                        /* newer versions of the kernel have a "signed" type */
+                        if (test_type_token(type, token, EVENT_ITEM, "signed"))
+                                goto fail;
+
+                        free_token(token);
+
+                        if (read_expected(EVENT_OP, ":") < 0)
+                                goto fail_expect;
+
+                        if (read_expect_type(EVENT_ITEM, &token))
+                                goto fail;
+
+                        if (strtoul(token, NULL, 0))
+                                field->flags |= FIELD_IS_SIGNED;
+
+                        free_token(token);
+                        if (read_expected(EVENT_OP, ";") < 0)
+                                goto fail_expect;
+
+                        if (read_expect_type(EVENT_NEWLINE, &token))
+                                goto fail;
+                }
+
+                free_token(token);
+
+                *fields = field;
+                fields = &field->next;
+
+        } while (1);
+
+        return 0;
+
+fail:
+        free_token(token);
+fail_expect:
+        if (field)
+                free(field);
+        return -1;
+}
+
+static int event_read_format(struct event *event)
+{
+        char *token;
+        int ret;
+
+        if (read_expected_item(EVENT_ITEM, "format") < 0)
+                return -1;
+
+        if (read_expected(EVENT_OP, ":") < 0)
+                return -1;
+
+        if (read_expect_type(EVENT_NEWLINE, &token))
+                goto fail;
+        free_token(token);
+
+        ret = event_read_fields(event, &event->format.common_fields);
+        if (ret < 0)
+                return ret;
+        event->format.nr_common = ret;
+
+        ret = event_read_fields(event, &event->format.fields);
+        if (ret < 0)
+                return ret;
+        event->format.nr_fields = ret;
+
+        return 0;
+
+ fail:
+        free_token(token);
+        return -1;
+}
+
+enum event_type
+process_arg_token(struct event *event, struct print_arg *arg,
+                  char **tok, enum event_type type);
+
+static enum event_type
+process_arg(struct event *event, struct print_arg *arg, char **tok)
+{
+        enum event_type type;
+        char *token;
+
+        type = read_token(&token);
+        *tok = token;
+
+        return process_arg_token(event, arg, tok, type);
+}
+
+static enum event_type
+process_cond(struct event *event, struct print_arg *top, char **tok)
+{
+        struct print_arg *arg, *left, *right;
+        enum event_type type;
+        char *token = NULL;
+
+        arg = malloc_or_die(sizeof(*arg));
+        memset(arg, 0, sizeof(*arg));
+
+        left = malloc_or_die(sizeof(*left));
+
+        right = malloc_or_die(sizeof(*right));
+
+        arg->type = PRINT_OP;
+        arg->op.left = left;
+        arg->op.right = right;
+
+        *tok = NULL;
+        type = process_arg(event, left, &token);
+        if (test_type_token(type, token, EVENT_OP, ":"))
+                goto out_free;
+
+        arg->op.op = token;
+
+        type = process_arg(event, right, &token);
+
+        top->op.right = arg;
+
+        *tok = token;
+        return type;
+
+out_free:
+        free_token(*tok);
+        free(right);
+        free(left);
+        free_arg(arg);
+        return EVENT_ERROR;
+}
+
+static enum event_type
+process_array(struct event *event, struct print_arg *top, char **tok)
+{
+        struct print_arg *arg;
+        enum event_type type;
+        char *token = NULL;
+
+        arg = malloc_or_die(sizeof(*arg));
+        memset(arg, 0, sizeof(*arg));
+
+        *tok = NULL;
+        type = process_arg(event, arg, &token);
+        if (test_type_token(type, token, EVENT_OP, "]"))
+                goto out_free;
+
+        top->op.right = arg;
+
+        free_token(token);
+        type = read_token_item(&token);
+        *tok = token;
+
+        return type;
+
+out_free:
+        free_token(*tok);
+        free_arg(arg);
+        return EVENT_ERROR;
+}
+
+static int get_op_prio(char *op)
+{
+        if (!op[1]) {
+                switch (op[0]) {
+                case '*':
+                case '/':
+                case '%':
+                        return 6;
+                case '+':
+                case '-':
+                        return 7;
+                        /* '>>' and '<<' are 8 */
+                case '<':
+                case '>':
+                        return 9;
+                        /* '==' and '!=' are 10 */
+                case '&':
+                        return 11;
+                case '^':
+                        return 12;
+                case '|':
+                        return 13;
+                case '?':
+                        return 16;
+                default:
+                        die("unknown op '%c'", op[0]);
+                        return -1;
+                }
+        } else {
+                if (strcmp(op, "++") == 0 ||
+                    strcmp(op, "--") == 0) {
+                        return 3;
+                } else if (strcmp(op, ">>") == 0 ||
+                           strcmp(op, "<<") == 0) {
+                        return 8;
+                } else if (strcmp(op, ">=") == 0 ||
+                           strcmp(op, "<=") == 0) {
+                        return 9;
+                } else if (strcmp(op, "==") == 0 ||
+                           strcmp(op, "!=") == 0) {
+                        return 10;
+                } else if (strcmp(op, "&&") == 0) {
+                        return 14;
+                } else if (strcmp(op, "||") == 0) {
+                        return 15;
+                } else {
+                        die("unknown op '%s'", op);
+                        return -1;
+                }
+        }
+}
+
+static void set_op_prio(struct print_arg *arg)
+{
+
+        /* single ops are the greatest */
+        if (!arg->op.left || arg->op.left->type == PRINT_NULL) {
+                arg->op.prio = 0;
+                return;
+        }
+
+        arg->op.prio = get_op_prio(arg->op.op);
+}
+
+static enum event_type
+process_op(struct event *event, struct print_arg *arg, char **tok)
+{
+        struct print_arg *left, *right = NULL;
+        enum event_type type;
+        char *token;
+
+        /* the op is passed in via tok */
+        token = *tok;
+
+        if (arg->type == PRINT_OP && !arg->op.left) {
+                /* handle single op */
+                if (token[1]) {
+                        die("bad op token %s", token);
+                        return EVENT_ERROR;
+                }
+                switch (token[0]) {
+                case '!':
+                case '+':
+                case '-':
+                        break;
+                default:
+                        die("bad op token %s", token);
+                        return EVENT_ERROR;
+                }
+
+                /* make an empty left */
+                left = malloc_or_die(sizeof(*left));
+                left->type = PRINT_NULL;
+                arg->op.left = left;
+
+                right = malloc_or_die(sizeof(*right));
+                arg->op.right = right;
+
+                type = process_arg(event, right, tok);
+
+        } else if (strcmp(token, "?") == 0) {
+
+                left = malloc_or_die(sizeof(*left));
+                /* copy the top arg to the left */
+                *left = *arg;
+
+                arg->type = PRINT_OP;
+                arg->op.op = token;
+                arg->op.left = left;
+                arg->op.prio = 0;
+
+                type = process_cond(event, arg, tok);
+
+        } else if (strcmp(token, ">>") == 0 ||
+                   strcmp(token, "<<") == 0 ||
+                   strcmp(token, "&") == 0 ||
+                   strcmp(token, "|") == 0 ||
+                   strcmp(token, "&&") == 0 ||
+                   strcmp(token, "||") == 0 ||
+                   strcmp(token, "-") == 0 ||
+                   strcmp(token, "+") == 0 ||
+                   strcmp(token, "*") == 0 ||
+                   strcmp(token, "^") == 0 ||
+                   strcmp(token, "/") == 0 ||
+                   strcmp(token, "<") == 0 ||
+                   strcmp(token, ">") == 0 ||
+                   strcmp(token, "==") == 0 ||
+                   strcmp(token, "!=") == 0) {
+
+                left = malloc_or_die(sizeof(*left));
+
+                /* copy the top arg to the left */
+                *left = *arg;
+
+                arg->type = PRINT_OP;
+                arg->op.op = token;
+                arg->op.left = left;
+
+                set_op_prio(arg);
+
+                right = malloc_or_die(sizeof(*right));
+
+                type = read_token_item(&token);
+                *tok = token;
+
+                /* could just be a type pointer */
+                if ((strcmp(arg->op.op, "*") == 0) &&
+                    type == EVENT_DELIM && (strcmp(token, ")") == 0)) {
+                        if (left->type != PRINT_ATOM)
+                                die("bad pointer type");
+                        left->atom.atom = realloc(left->atom.atom,
+                                            sizeof(left->atom.atom) + 3);
+                        strcat(left->atom.atom, " *");
+                        *arg = *left;
+                        free(arg);
+
+                        return type;
+                }
+
+                type = process_arg_token(event, right, tok, type);
+
+                arg->op.right = right;
+
+        } else if (strcmp(token, "[") == 0) {
+
+                left = malloc_or_die(sizeof(*left));
+                *left = *arg;
+
+                arg->type = PRINT_OP;
+                arg->op.op = token;
+                arg->op.left = left;
+
+                arg->op.prio = 0;
+                type = process_array(event, arg, tok);
+
+        } else {
+                warning("unknown op '%s'", token);
+                event->flags |= EVENT_FL_FAILED;
+                /* the arg is now the left side */
+                return EVENT_NONE;
+        }
+
+        if (type == EVENT_OP) {
+                int prio;
+
+                /* higher prios need to be closer to the root */
+                prio = get_op_prio(*tok);
+
+                if (prio > arg->op.prio)
+                        return process_op(event, arg, tok);
+
+                return process_op(event, right, tok);
+        }
+
+        return type;
+}
+
+static enum event_type
+process_entry(struct event *event __unused, struct print_arg *arg,
+              char **tok)
+{
+        enum event_type type;
+        char *field;
+        char *token;
+
+        if (read_expected(EVENT_OP, "->") < 0)
+                return EVENT_ERROR;
+
+        if (read_expect_type(EVENT_ITEM, &token) < 0)
+                goto fail;
+        field = token;
+
+        arg->type = PRINT_FIELD;
+        arg->field.name = field;
+
+        if (is_flag_field) {
+                arg->field.field = find_any_field(event, arg->field.name);
+                arg->field.field->flags |= FIELD_IS_FLAG;
+                is_flag_field = 0;
+        } else if (is_symbolic_field) {
+                arg->field.field = find_any_field(event, arg->field.name);
+                arg->field.field->flags |= FIELD_IS_SYMBOLIC;
+                is_symbolic_field = 0;
+        }
+
+        type = read_token(&token);
+        *tok = token;
+
+        return type;
+
+fail:
+        free_token(token);
+        return EVENT_ERROR;
+}
+
+static char *arg_eval (struct print_arg *arg);
+
+static long long arg_num_eval(struct print_arg *arg)
+{
+        long long left, right;
+        long long val = 0;
+
+        switch (arg->type) {
+        case PRINT_ATOM:
+                val = strtoll(arg->atom.atom, NULL, 0);
+                break;
+        case PRINT_TYPE:
+                val = arg_num_eval(arg->typecast.item);
+                break;
+        case PRINT_OP:
+                switch (arg->op.op[0]) {
+                case '|':
+                        left = arg_num_eval(arg->op.left);
+                        right = arg_num_eval(arg->op.right);
+                        if (arg->op.op[1])
+                                val = left || right;
+                        else
+                                val = left | right;
+                        break;
+                case '&':
+                        left = arg_num_eval(arg->op.left);
+                        right = arg_num_eval(arg->op.right);
+                        if (arg->op.op[1])
+                                val = left && right;
+                        else
+                                val = left & right;
+                        break;
+                case '<':
+                        left = arg_num_eval(arg->op.left);
+                        right = arg_num_eval(arg->op.right);
+                        switch (arg->op.op[1]) {
+                        case 0:
+                                val = left < right;
+                                break;
+                        case '<':
+                                val = left << right;
+                                break;
+                        case '=':
+                                val = left <= right;
+                                break;
+                        default:
+                                die("unknown op '%s'", arg->op.op);
+                        }
+                        break;
+                case '>':
+                        left = arg_num_eval(arg->op.left);
+                        right = arg_num_eval(arg->op.right);
+                        switch (arg->op.op[1]) {
+                        case 0:
+                                val = left > right;
+                                break;
+                        case '>':
+                                val = left >> right;
+                                break;
+                        case '=':
+                                val = left >= right;
+                                break;
+                        default:
+                                die("unknown op '%s'", arg->op.op);
+                        }
+                        break;
+                case '=':
+                        left = arg_num_eval(arg->op.left);
+                        right = arg_num_eval(arg->op.right);
+
+                        if (arg->op.op[1] != '=')
+                                die("unknown op '%s'", arg->op.op);
+
+                        val = left == right;
+                        break;
+                case '!':
+                        left = arg_num_eval(arg->op.left);
+                        right = arg_num_eval(arg->op.right);
+
+                        switch (arg->op.op[1]) {
+                        case '=':
+                                val = left != right;
+                                break;
+                        default:
+                                die("unknown op '%s'", arg->op.op);
+                        }
+                        break;
+                default:
+                        die("unknown op '%s'", arg->op.op);
+                }
+                break;
+
+        case PRINT_NULL:
+        case PRINT_FIELD ... PRINT_SYMBOL:
+        case PRINT_STRING:
+        default:
+                die("invalid eval type %d", arg->type);
+
+        }
+        return val;
+}
+
+static char *arg_eval (struct print_arg *arg)
+{
+        long long val;
+        static char buf[20];
+
+        switch (arg->type) {
+        case PRINT_ATOM:
+                return arg->atom.atom;
+        case PRINT_TYPE:
+                return arg_eval(arg->typecast.item);
+        case PRINT_OP:
+                val = arg_num_eval(arg);
+                sprintf(buf, "%lld", val);
+                return buf;
+
+        case PRINT_NULL:
+        case PRINT_FIELD ... PRINT_SYMBOL:
+        case PRINT_STRING:
+        default:
+                die("invalid eval type %d", arg->type);
+                break;
+        }
+
+        return NULL;
+}
+
+static enum event_type
+process_fields(struct event *event, struct print_flag_sym **list, char **tok)
+{
+        enum event_type type;
+        struct print_arg *arg = NULL;
+        struct print_flag_sym *field;
+        char *token = NULL;
+        char *value;
+
+        do {
+                free_token(token);
+                type = read_token_item(&token);
+                if (test_type_token(type, token, EVENT_OP, "{"))
+                        break;
+
+                arg = malloc_or_die(sizeof(*arg));
+
+                free_token(token);
+                type = process_arg(event, arg, &token);
+                if (test_type_token(type, token, EVENT_DELIM, ","))
+                        goto out_free;
+
+                field = malloc_or_die(sizeof(*field));
+                memset(field, 0, sizeof(*field));
+
+                value = arg_eval(arg);
+                field->value = strdup(value);
+
+                free_token(token);
+                type = process_arg(event, arg, &token);
+                if (test_type_token(type, token, EVENT_OP, "}"))
+                        goto out_free;
+
+                value = arg_eval(arg);
+                field->str = strdup(value);
+                free_arg(arg);
+                arg = NULL;
+
+                *list = field;
+                list = &field->next;
+
+                free_token(token);
+                type = read_token_item(&token);
+        } while (type == EVENT_DELIM && strcmp(token, ",") == 0);
+
+        *tok = token;
+        return type;
+
+out_free:
+        free_arg(arg);
+        free_token(token);
+
+        return EVENT_ERROR;
+}
+
+static enum event_type
+process_flags(struct event *event, struct print_arg *arg, char **tok)
+{
+        struct print_arg *field;
+        enum event_type type;
+        char *token;
+
+        memset(arg, 0, sizeof(*arg));
+        arg->type = PRINT_FLAGS;
+
+        if (read_expected_item(EVENT_DELIM, "(") < 0)
+                return EVENT_ERROR;
+
+        field = malloc_or_die(sizeof(*field));
+
+        type = process_arg(event, field, &token);
+        if (test_type_token(type, token, EVENT_DELIM, ","))
+                goto out_free;
+
+        arg->flags.field = field;
+
+        type = read_token_item(&token);
+        if (event_item_type(type)) {
+                arg->flags.delim = token;
+                type = read_token_item(&token);
+        }
+
+        if (test_type_token(type, token, EVENT_DELIM, ","))
+                goto out_free;
+
+        type = process_fields(event, &arg->flags.flags, &token);
+        if (test_type_token(type, token, EVENT_DELIM, ")"))
+                goto out_free;
+
+        free_token(token);
+        type = read_token_item(tok);
+        return type;
+
+out_free:
+        free_token(token);
+        return EVENT_ERROR;
+}
+
+static enum event_type
+process_symbols(struct event *event, struct print_arg *arg, char **tok)
+{
+        struct print_arg *field;
+        enum event_type type;
+        char *token;
+
+        memset(arg, 0, sizeof(*arg));
+        arg->type = PRINT_SYMBOL;
+
+        if (read_expected_item(EVENT_DELIM, "(") < 0)
+                return EVENT_ERROR;
+
+        field = malloc_or_die(sizeof(*field));
+
+        type = process_arg(event, field, &token);
+        if (test_type_token(type, token, EVENT_DELIM, ","))
+                goto out_free;
+
+        arg->symbol.field = field;
+
+        type = process_fields(event, &arg->symbol.symbols, &token);
+        if (test_type_token(type, token, EVENT_DELIM, ")"))
+                goto out_free;
+
+        free_token(token);
+        type = read_token_item(tok);
+        return type;
+
+out_free:
+        free_token(token);
+        return EVENT_ERROR;
+}
+
+static enum event_type
+process_paren(struct event *event, struct print_arg *arg, char **tok)
+{
+        struct print_arg *item_arg;
+        enum event_type type;
+        char *token;
+
+        type = process_arg(event, arg, &token);
+
+        if (type == EVENT_ERROR)
+                return EVENT_ERROR;
+
+        if (type == EVENT_OP)
+                type = process_op(event, arg, &token);
+
+        if (type == EVENT_ERROR)
+                return EVENT_ERROR;
+
+        if (test_type_token(type, token, EVENT_DELIM, ")")) {
+                free_token(token);
+                return EVENT_ERROR;
+        }
+
+        free_token(token);
+        type = read_token_item(&token);
+
+        /*
+         * If the next token is an item or another open paren, then
+         * this was a typecast.
+         */
+        if (event_item_type(type) ||
+            (type == EVENT_DELIM && strcmp(token, "(") == 0)) {
+
+                /* make this a typecast and contine */
+
+                /* prevous must be an atom */
+                if (arg->type != PRINT_ATOM)
+                        die("previous needed to be PRINT_ATOM");
+
+                item_arg = malloc_or_die(sizeof(*item_arg));
+
+                arg->type = PRINT_TYPE;
+                arg->typecast.type = arg->atom.atom;
+                arg->typecast.item = item_arg;
+                type = process_arg_token(event, item_arg, &token, type);
+
+        }
+
+        *tok = token;
+        return type;
+}
+
+
+static enum event_type
+process_str(struct event *event __unused, struct print_arg *arg, char **tok)
+{
+        enum event_type type;
+        char *token;
+
+        if (read_expected(EVENT_DELIM, "(") < 0)
+                return EVENT_ERROR;
+
+        if (read_expect_type(EVENT_ITEM, &token) < 0)
+                goto fail;
+
+        arg->type = PRINT_STRING;
+        arg->string.string = token;
+        arg->string.offset = -1;
+
+        if (read_expected(EVENT_DELIM, ")") < 0)
+                return EVENT_ERROR;
+
+        type = read_token(&token);
+        *tok = token;
+
+        return type;
+fail:
+        free_token(token);
+        return EVENT_ERROR;
+}
+
+enum event_type
+process_arg_token(struct event *event, struct print_arg *arg,
+                  char **tok, enum event_type type)
+{
+        char *token;
+        char *atom;
+
+        token = *tok;
+
+        switch (type) {
+        case EVENT_ITEM:
+                if (strcmp(token, "REC") == 0) {
+                        free_token(token);
+                        type = process_entry(event, arg, &token);
+                } else if (strcmp(token, "__print_flags") == 0) {
+                        free_token(token);
+                        is_flag_field = 1;
+                        type = process_flags(event, arg, &token);
+                } else if (strcmp(token, "__print_symbolic") == 0) {
+                        free_token(token);
+                        is_symbolic_field = 1;
+                        type = process_symbols(event, arg, &token);
+                } else if (strcmp(token, "__get_str") == 0) {
+                        free_token(token);
+                        type = process_str(event, arg, &token);
+                } else {
+                        atom = token;
+                        /* test the next token */
+                        type = read_token_item(&token);
+
+                        /* atoms can be more than one token long */
+                        while (type == EVENT_ITEM) {
+                                atom = realloc(atom, strlen(atom) + strlen(token) + 2);
+                                strcat(atom, " ");
+                                strcat(atom, token);
+                                free_token(token);
+                                type = read_token_item(&token);
+                        }
+
+                        /* todo, test for function */
+
+                        arg->type = PRINT_ATOM;
+                        arg->atom.atom = atom;
+                }
+                break;
+        case EVENT_DQUOTE:
+        case EVENT_SQUOTE:
+                arg->type = PRINT_ATOM;
+                arg->atom.atom = token;
+                type = read_token_item(&token);
+                break;
+        case EVENT_DELIM:
+                if (strcmp(token, "(") == 0) {
+                        free_token(token);
+                        type = process_paren(event, arg, &token);
+                        break;
+                }
+        case EVENT_OP:
+                /* handle single ops */
+                arg->type = PRINT_OP;
+                arg->op.op = token;
+                arg->op.left = NULL;
+                type = process_op(event, arg, &token);
+
+                break;
+
+        case EVENT_ERROR ... EVENT_NEWLINE:
+        default:
+                die("unexpected type %d", type);
+        }
+        *tok = token;
+
+        return type;
+}
+
+static int event_read_print_args(struct event *event, struct print_arg **list)
+{
+        enum event_type type = EVENT_ERROR;
+        struct print_arg *arg;
+        char *token;
+        int args = 0;
+
+        do {
+                if (type == EVENT_NEWLINE) {
+                        free_token(token);
+                        type = read_token_item(&token);
+                        continue;
+                }
+
+                arg = malloc_or_die(sizeof(*arg));
+                memset(arg, 0, sizeof(*arg));
+
+                type = process_arg(event, arg, &token);
+
+                if (type == EVENT_ERROR) {
+                        free_arg(arg);
+                        return -1;
+                }
+
+                *list = arg;
+                args++;
+
+                if (type == EVENT_OP) {
+                        type = process_op(event, arg, &token);
+                        list = &arg->next;
+                        continue;
+                }
+
+                if (type == EVENT_DELIM && strcmp(token, ",") == 0) {
+                        free_token(token);
+                        *list = arg;
+                        list = &arg->next;
+                        continue;
+                }
+                break;
+        } while (type != EVENT_NONE);
+
+        if (type != EVENT_NONE)
+                free_token(token);
+
+        return args;
+}
+
+static int event_read_print(struct event *event)
+{
+        enum event_type type;
+        char *token;
+        int ret;
+
+        if (read_expected_item(EVENT_ITEM, "print") < 0)
+                return -1;
+
+        if (read_expected(EVENT_ITEM, "fmt") < 0)
+                return -1;
+
+        if (read_expected(EVENT_OP, ":") < 0)
+                return -1;
+
+        if (read_expect_type(EVENT_DQUOTE, &token) < 0)
+                goto fail;
+
+ concat:
+        event->print_fmt.format = token;
+        event->print_fmt.args = NULL;
+
+        /* ok to have no arg */
+        type = read_token_item(&token);
+
+        if (type == EVENT_NONE)
+                return 0;
+
+        /* Handle concatination of print lines */
+        if (type == EVENT_DQUOTE) {
+                char *cat;
+
+                cat = malloc_or_die(strlen(event->print_fmt.format) +
+                                    strlen(token) + 1);
+                strcpy(cat, event->print_fmt.format);
+                strcat(cat, token);
+                free_token(token);
+                free_token(event->print_fmt.format);
+                event->print_fmt.format = NULL;
+                token = cat;
+                goto concat;
+        }
+
+        if (test_type_token(type, token, EVENT_DELIM, ","))
+                goto fail;
+
+        free_token(token);
+
+        ret = event_read_print_args(event, &event->print_fmt.args);
+        if (ret < 0)
+                return -1;
+
+        return ret;
+
+ fail:
+        free_token(token);
+        return -1;
+}
+
+static struct format_field *
+find_common_field(struct event *event, const char *name)
+{
+        struct format_field *format;
+
+        for (format = event->format.common_fields;
+             format; format = format->next) {
+                if (strcmp(format->name, name) == 0)
+                        break;
+        }
+
+        return format;
+}
+
+static struct format_field *
+find_field(struct event *event, const char *name)
+{
+        struct format_field *format;
+
+        for (format = event->format.fields;
+             format; format = format->next) {
+                if (strcmp(format->name, name) == 0)
+                        break;
+        }
+
+        return format;
+}
+
+static struct format_field *
+find_any_field(struct event *event, const char *name)
+{
+        struct format_field *format;
+
+        format = find_common_field(event, name);
+        if (format)
+                return format;
+        return find_field(event, name);
+}
+
+unsigned long long read_size(void *ptr, int size)
+{
+        switch (size) {
+        case 1:
+                return *(unsigned char *)ptr;
+        case 2:
+                return data2host2(ptr);
+        case 4:
+                return data2host4(ptr);
+        case 8:
+                return data2host8(ptr);
+        default:
+                /* BUG! */
+                return 0;
+        }
+}
+
+unsigned long long
+raw_field_value(struct event *event, const char *name, void *data)
+{
+        struct format_field *field;
+
+        field = find_any_field(event, name);
+        if (!field)
+                return 0ULL;
+
+        return read_size(data + field->offset, field->size);
+}
+
+void *raw_field_ptr(struct event *event, const char *name, void *data)
+{
+        struct format_field *field;
+
+        field = find_any_field(event, name);
+        if (!field)
+                return NULL;
+
+        if (field->flags & FIELD_IS_DYNAMIC) {
+                int offset;
+
+                offset = *(int *)(data + field->offset);
+                offset &= 0xffff;
+
+                return data + offset;
+        }
+
+        return data + field->offset;
+}
+
+static int get_common_info(const char *type, int *offset, int *size)
+{
+        struct event *event;
+        struct format_field *field;
+
+        /*
+         * All events should have the same common elements.
+         * Pick any event to find where the type is;
+         */
+        if (!event_list)
+                die("no event_list!");
+
+        event = event_list;
+        field = find_common_field(event, type);
+        if (!field)
+                die("field '%s' not found", type);
+
+        *offset = field->offset;
+        *size = field->size;
+
+        return 0;
+}
+
+static int __parse_common(void *data, int *size, int *offset,
+                          const char *name)
+{
+        int ret;
+
+        if (!*size) {
+                ret = get_common_info(name, offset, size);
+                if (ret < 0)
+                        return ret;
+        }
+        return read_size(data + *offset, *size);
+}
+
+int trace_parse_common_type(void *data)
+{
+        static int type_offset;
+        static int type_size;
+
+        return __parse_common(data, &type_size, &type_offset,
+                              "common_type");
+}
+
+int trace_parse_common_pid(void *data)
+{
+        static int pid_offset;
+        static int pid_size;
+
+        return __parse_common(data, &pid_size, &pid_offset,
+                              "common_pid");
+}
+
+int parse_common_pc(void *data)
+{
+        static int pc_offset;
+        static int pc_size;
+
+        return __parse_common(data, &pc_size, &pc_offset,
+                              "common_preempt_count");
+}
+
+int parse_common_flags(void *data)
+{
+        static int flags_offset;
+        static int flags_size;
+
+        return __parse_common(data, &flags_size, &flags_offset,
+                              "common_flags");
+}
+
+int parse_common_lock_depth(void *data)
+{
+        static int ld_offset;
+        static int ld_size;
+        int ret;
+
+        ret = __parse_common(data, &ld_size, &ld_offset,
+                             "common_lock_depth");
+        if (ret < 0)
+                return -1;
+
+        return ret;
+}
+
+struct event *trace_find_event(int id)
+{
+        struct event *event;
+
+        for (event = event_list; event; event = event->next) {
+                if (event->id == id)
+                        break;
+        }
+        return event;
+}
+
+struct event *trace_find_next_event(struct event *event)
+{
+        if (!event)
+                return event_list;
+
+        return event->next;
+}
+
+static unsigned long long eval_num_arg(void *data, int size,
+                                   struct event *event, struct print_arg *arg)
+{
+        unsigned long long val = 0;
+        unsigned long long left, right;
+        struct print_arg *larg;
+
+        switch (arg->type) {
+        case PRINT_NULL:
+                /* ?? */
+                return 0;
+        case PRINT_ATOM:
+                return strtoull(arg->atom.atom, NULL, 0);
+        case PRINT_FIELD:
+                if (!arg->field.field) {
+                        arg->field.field = find_any_field(event, arg->field.name);
+                        if (!arg->field.field)
+                                die("field %s not found", arg->field.name);
+                }
+                /* must be a number */
+                val = read_size(data + arg->field.field->offset,
+                                arg->field.field->size);
+                break;
+        case PRINT_FLAGS:
+        case PRINT_SYMBOL:
+                break;
+        case PRINT_TYPE:
+                return eval_num_arg(data, size, event, arg->typecast.item);
+        case PRINT_STRING:
+                return 0;
+                break;
+        case PRINT_OP:
+                if (strcmp(arg->op.op, "[") == 0) {
+                        /*
+                         * Arrays are special, since we don't want
+                         * to read the arg as is.
+                         */
+                        if (arg->op.left->type != PRINT_FIELD)
+                                goto default_op; /* oops, all bets off */
+                        larg = arg->op.left;
+                        if (!larg->field.field) {
+                                larg->field.field =
+                                        find_any_field(event, larg->field.name);
+                                if (!larg->field.field)
+                                        die("field %s not found", larg->field.name);
+                        }
+                        right = eval_num_arg(data, size, event, arg->op.right);
+                        val = read_size(data + larg->field.field->offset +
+                                        right * long_size, long_size);
+                        break;
+                }
+ default_op:
+                left = eval_num_arg(data, size, event, arg->op.left);
+                right = eval_num_arg(data, size, event, arg->op.right);
+                switch (arg->op.op[0]) {
+                case '|':
+                        if (arg->op.op[1])
+                                val = left || right;
+                        else
+                                val = left | right;
+                        break;
+                case '&':
+                        if (arg->op.op[1])
+                                val = left && right;
+                        else
+                                val = left & right;
+                        break;
+                case '<':
+                        switch (arg->op.op[1]) {
+                        case 0:
+                                val = left < right;
+                                break;
+                        case '<':
+                                val = left << right;
+                                break;
+                        case '=':
+                                val = left <= right;
+                                break;
+                        default:
+                                die("unknown op '%s'", arg->op.op);
+                        }
+                        break;
+                case '>':
+                        switch (arg->op.op[1]) {
+                        case 0:
+                                val = left > right;
+                                break;
+                        case '>':
+                                val = left >> right;
+                                break;
+                        case '=':
+                                val = left >= right;
+                                break;
+                        default:
+                                die("unknown op '%s'", arg->op.op);
+                        }
+                        break;
+                case '=':
+                        if (arg->op.op[1] != '=')
+                                die("unknown op '%s'", arg->op.op);
+                        val = left == right;
+                        break;
+                case '-':
+                        val = left - right;
+                        break;
+                case '+':
+                        val = left + right;
+                        break;
+                default:
+                        die("unknown op '%s'", arg->op.op);
+                }
+                break;
+        default: /* not sure what to do there */
+                return 0;
+        }
+        return val;
+}
+
+struct flag {
+        const char *name;
+        unsigned long long value;
+};
+
+static const struct flag flags[] = {
+        { "HI_SOFTIRQ", 0 },
+        { "TIMER_SOFTIRQ", 1 },
+        { "NET_TX_SOFTIRQ", 2 },
+        { "NET_RX_SOFTIRQ", 3 },
+        { "BLOCK_SOFTIRQ", 4 },
+        { "BLOCK_IOPOLL_SOFTIRQ", 5 },
+        { "TASKLET_SOFTIRQ", 6 },
+        { "SCHED_SOFTIRQ", 7 },
+        { "HRTIMER_SOFTIRQ", 8 },
+        { "RCU_SOFTIRQ", 9 },
+
+        { "HRTIMER_NORESTART", 0 },
+        { "HRTIMER_RESTART", 1 },
+};
+
+unsigned long long eval_flag(const char *flag)
+{
+        int i;
+
+        /*
+         * Some flags in the format files do not get converted.
+         * If the flag is not numeric, see if it is something that
+         * we already know about.
+         */
+        if (isdigit(flag[0]))
+                return strtoull(flag, NULL, 0);
+
+        for (i = 0; i < (int)(sizeof(flags)/sizeof(flags[0])); i++)
+                if (strcmp(flags[i].name, flag) == 0)
+                        return flags[i].value;
+
+        return 0;
+}
+
+static void print_str_arg(void *data, int size,
+                          struct event *event, struct print_arg *arg)
+{
+        struct print_flag_sym *flag;
+        unsigned long long val, fval;
+        char *str;
+        int print;
+
+        switch (arg->type) {
+        case PRINT_NULL:
+                /* ?? */
+                return;
+        case PRINT_ATOM:
+                printf("%s", arg->atom.atom);
+                return;
+        case PRINT_FIELD:
+                if (!arg->field.field) {
+                        arg->field.field = find_any_field(event, arg->field.name);
+                        if (!arg->field.field)
+                                die("field %s not found", arg->field.name);
+                }
+                str = malloc_or_die(arg->field.field->size + 1);
+                memcpy(str, data + arg->field.field->offset,
+                       arg->field.field->size);
+                str[arg->field.field->size] = 0;
+                printf("%s", str);
+                free(str);
+                break;
+        case PRINT_FLAGS:
+                val = eval_num_arg(data, size, event, arg->flags.field);
+                print = 0;
+                for (flag = arg->flags.flags; flag; flag = flag->next) {
+                        fval = eval_flag(flag->value);
+                        if (!val && !fval) {
+                                printf("%s", flag->str);
+                                break;
+                        }
+                        if (fval && (val & fval) == fval) {
+                                if (print && arg->flags.delim)
+                                        printf("%s", arg->flags.delim);
+                                printf("%s", flag->str);
+                                print = 1;
+                                val &= ~fval;
+                        }
+                }
+                break;
+        case PRINT_SYMBOL:
+                val = eval_num_arg(data, size, event, arg->symbol.field);
+                for (flag = arg->symbol.symbols; flag; flag = flag->next) {
+                        fval = eval_flag(flag->value);
+                        if (val == fval) {
+                                printf("%s", flag->str);
+                                break;
+                        }
+                }
+                break;
+
+        case PRINT_TYPE:
+                break;
+        case PRINT_STRING: {
+                int str_offset;
+
+                if (arg->string.offset == -1) {
+                        struct format_field *f;
+
+                        f = find_any_field(event, arg->string.string);
+                        arg->string.offset = f->offset;
+                }
+                str_offset = *(int *)(data + arg->string.offset);
+                str_offset &= 0xffff;
+                printf("%s", ((char *)data) + str_offset);
+                break;
+        }
+        case PRINT_OP:
+                /*
+                 * The only op for string should be ? :
+                 */
+                if (arg->op.op[0] != '?')
+                        return;
+                val = eval_num_arg(data, size, event, arg->op.left);
+                if (val)
+                        print_str_arg(data, size, event, arg->op.right->op.left);
+                else
+                        print_str_arg(data, size, event, arg->op.right->op.right);
+                break;
+        default:
+                /* well... */
+                break;
+        }
+}
+
+static struct print_arg *make_bprint_args(char *fmt, void *data, int size, struct event *event)
+{
+        static struct format_field *field, *ip_field;
+        struct print_arg *args, *arg, **next;
+        unsigned long long ip, val;
+        char *ptr;
+        void *bptr;
+
+        if (!field) {
+                field = find_field(event, "buf");
+                if (!field)
+                        die("can't find buffer field for binary printk");
+                ip_field = find_field(event, "ip");
+                if (!ip_field)
+                        die("can't find ip field for binary printk");
+        }
+
+        ip = read_size(data + ip_field->offset, ip_field->size);
+
+        /*
+         * The first arg is the IP pointer.
+         */
+        args = malloc_or_die(sizeof(*args));
+        arg = args;
+        arg->next = NULL;
+        next = &arg->next;
+
+        arg->type = PRINT_ATOM;
+        arg->atom.atom = malloc_or_die(32);
+        sprintf(arg->atom.atom, "%lld", ip);
+
+        /* skip the first "%pf : " */
+        for (ptr = fmt + 6, bptr = data + field->offset;
+             bptr < data + size && *ptr; ptr++) {
+                int ls = 0;
+
+                if (*ptr == '%') {
+ process_again:
+                        ptr++;
+                        switch (*ptr) {
+                        case '%':
+                                break;
+                        case 'l':
+                                ls++;
+                                goto process_again;
+                        case 'L':
+                                ls = 2;
+                                goto process_again;
+                        case '0' ... '9':
+                                goto process_again;
+                        case 'p':
+                                ls = 1;
+                                /* fall through */
+                        case 'd':
+                        case 'u':
+                        case 'x':
+                        case 'i':
+                                /* the pointers are always 4 bytes aligned */
+                                bptr = (void *)(((unsigned long)bptr + 3) &
+                                                ~3);
+                                switch (ls) {
+                                case 0:
+                                case 1:
+                                        ls = long_size;
+                                        break;
+                                case 2:
+                                        ls = 8;
+                                default:
+                                        break;
+                                }
+                                val = read_size(bptr, ls);
+                                bptr += ls;
+                                arg = malloc_or_die(sizeof(*arg));
+                                arg->next = NULL;
+                                arg->type = PRINT_ATOM;
+                                arg->atom.atom = malloc_or_die(32);
+                                sprintf(arg->atom.atom, "%lld", val);
+                                *next = arg;
+                                next = &arg->next;
+                                break;
+                        case 's':
+                                arg = malloc_or_die(sizeof(*arg));
+                                arg->next = NULL;
+                                arg->type = PRINT_STRING;
+                                arg->string.string = strdup(bptr);
+                                bptr += strlen(bptr) + 1;
+                                *next = arg;
+                                next = &arg->next;
+                        default:
+                                break;
+                        }
+                }
+        }
+
+        return args;
+}
+
+static void free_args(struct print_arg *args)
+{
+        struct print_arg *next;
+
+        while (args) {
+                next = args->next;
+
+                if (args->type == PRINT_ATOM)
+                        free(args->atom.atom);
+                else
+                        free(args->string.string);
+                free(args);
+                args = next;
+        }
+}
+
+static char *get_bprint_format(void *data, int size __unused, struct event *event)
+{
+        unsigned long long addr;
+        static struct format_field *field;
+        struct printk_map *printk;
+        char *format;
+        char *p;
+
+        if (!field) {
+                field = find_field(event, "fmt");
+                if (!field)
+                        die("can't find format field for binary printk");
+                printf("field->offset = %d size=%d\n", field->offset, field->size);
+        }
+
+        addr = read_size(data + field->offset, field->size);
+
+        printk = find_printk(addr);
+        if (!printk) {
+                format = malloc_or_die(45);
+                sprintf(format, "%%pf : (NO FORMAT FOUND at %llx)\n",
+                        addr);
+                return format;
+        }
+
+        p = printk->printk;
+        /* Remove any quotes. */
+        if (*p == '"')
+                p++;
+        format = malloc_or_die(strlen(p) + 10);
+        sprintf(format, "%s : %s", "%pf", p);
+        /* remove ending quotes and new line since we will add one too */
+        p = format + strlen(format) - 1;
+        if (*p == '"')
+                *p = 0;
+
+        p -= 2;
+        if (strcmp(p, "\\n") == 0)
+                *p = 0;
+
+        return format;
+}
+
+static void pretty_print(void *data, int size, struct event *event)
+{
+        struct print_fmt *print_fmt = &event->print_fmt;
+        struct print_arg *arg = print_fmt->args;
+        struct print_arg *args = NULL;
+        const char *ptr = print_fmt->format;
+        unsigned long long val;
+        struct func_map *func;
+        const char *saveptr;
+        char *bprint_fmt = NULL;
+        char format[32];
+        int show_func;
+        int len;
+        int ls;
+
+        if (event->flags & EVENT_FL_ISFUNC)
+                ptr = " %pF <-- %pF";
+
+        if (event->flags & EVENT_FL_ISBPRINT) {
+                bprint_fmt = get_bprint_format(data, size, event);
+                args = make_bprint_args(bprint_fmt, data, size, event);
+                arg = args;
+                ptr = bprint_fmt;
+        }
+
+        for (; *ptr; ptr++) {
+                ls = 0;
+                if (*ptr == '\\') {
+                        ptr++;
+                        switch (*ptr) {
+                        case 'n':
+                                printf("\n");
+                                break;
+                        case 't':
+                                printf("\t");
+                                break;
+                        case 'r':
+                                printf("\r");
+                                break;
+                        case '\\':
+                                printf("\\");
+                                break;
+                        default:
+                                printf("%c", *ptr);
+                                break;
+                        }
+
+                } else if (*ptr == '%') {
+                        saveptr = ptr;
+                        show_func = 0;
+ cont_process:
+                        ptr++;
+                        switch (*ptr) {
+                        case '%':
+                                printf("%%");
+                                break;
+                        case 'l':
+                                ls++;
+                                goto cont_process;
+                        case 'L':
+                                ls = 2;
+                                goto cont_process;
+                        case 'z':
+                        case 'Z':
+                        case '0' ... '9':
+                                goto cont_process;
+                        case 'p':
+                                if (long_size == 4)
+                                        ls = 1;
+                                else
+                                        ls = 2;
+
+                                if (*(ptr+1) == 'F' ||
+                                    *(ptr+1) == 'f') {
+                                        ptr++;
+                                        show_func = *ptr;
+                                }
+
+                                /* fall through */
+                        case 'd':
+                        case 'i':
+                        case 'x':
+                        case 'X':
+                        case 'u':
+                                if (!arg)
+                                        die("no argument match");
+
+                                len = ((unsigned long)ptr + 1) -
+                                        (unsigned long)saveptr;
+
+                                /* should never happen */
+                                if (len > 32)
+                                        die("bad format!");
+
+                                memcpy(format, saveptr, len);
+                                format[len] = 0;
+
+                                val = eval_num_arg(data, size, event, arg);
+                                arg = arg->next;
+
+                                if (show_func) {
+                                        func = find_func(val);
+                                        if (func) {
+                                                printf("%s", func->func);
+                                                if (show_func == 'F')
+                                                        printf("+0x%llx",
+                                                               val - func->addr);
+                                                break;
+                                        }
+                                }
+                                switch (ls) {
+                                case 0:
+                                        printf(format, (int)val);
+                                        break;
+                                case 1:
+                                        printf(format, (long)val);
+                                        break;
+                                case 2:
+                                        printf(format, (long long)val);
+                                        break;
+                                default:
+                                        die("bad count (%d)", ls);
+                                }
+                                break;
+                        case 's':
+                                if (!arg)
+                                        die("no matching argument");
+
+                                print_str_arg(data, size, event, arg);
+                                arg = arg->next;
+                                break;
+                        default:
+                                printf(">%c<", *ptr);
+
+                        }
+                } else
+                        printf("%c", *ptr);
+        }
+
+        if (args) {
+                free_args(args);
+                free(bprint_fmt);
+        }
+}
+
+static inline int log10_cpu(int nb)
+{
+        if (nb / 100)
+                return 3;
+        if (nb / 10)
+                return 2;
+        return 1;
+}
+
+static void print_lat_fmt(void *data, int size __unused)
+{
+        unsigned int lat_flags;
+        unsigned int pc;
+        int lock_depth;
+        int hardirq;
+        int softirq;
+
+        lat_flags = parse_common_flags(data);
+        pc = parse_common_pc(data);
+        lock_depth = parse_common_lock_depth(data);
+
+        hardirq = lat_flags & TRACE_FLAG_HARDIRQ;
+        softirq = lat_flags & TRACE_FLAG_SOFTIRQ;
+
+        printf("%c%c%c",
+               (lat_flags & TRACE_FLAG_IRQS_OFF) ? 'd' :
+               (lat_flags & TRACE_FLAG_IRQS_NOSUPPORT) ?
+               'X' : '.',
+               (lat_flags & TRACE_FLAG_NEED_RESCHED) ?
+               'N' : '.',
+               (hardirq && softirq) ? 'H' :
+               hardirq ? 'h' : softirq ? 's' : '.');
+
+        if (pc)
+                printf("%x", pc);
+        else
+                printf(".");
+
+        if (lock_depth < 0)
+                printf(".");
+        else
+                printf("%d", lock_depth);
+}
+
+/* taken from Linux, written by Frederic Weisbecker */
+static void print_graph_cpu(int cpu)
+{
+        int i;
+        int log10_this = log10_cpu(cpu);
+        int log10_all = log10_cpu(cpus);
+
+
+        /*
+         * Start with a space character - to make it stand out
+         * to the right a bit when trace output is pasted into
+         * email:
+         */
+        printf(" ");
+
+        /*
+         * Tricky - we space the CPU field according to the max
+         * number of online CPUs. On a 2-cpu system it would take
+         * a maximum of 1 digit - on a 128 cpu system it would
+         * take up to 3 digits:
+         */
+        for (i = 0; i < log10_all - log10_this; i++)
+                printf(" ");
+
+        printf("%d) ", cpu);
+}
+
+#define TRACE_GRAPH_PROCINFO_LENGTH     14
+#define TRACE_GRAPH_INDENT      2
+
+static void print_graph_proc(int pid, const char *comm)
+{
+        /* sign + log10(MAX_INT) + '\0' */
+        char pid_str[11];
+        int spaces = 0;
+        int len;
+        int i;
+
+        sprintf(pid_str, "%d", pid);
+
+        /* 1 stands for the "-" character */
+        len = strlen(comm) + strlen(pid_str) + 1;
+
+        if (len < TRACE_GRAPH_PROCINFO_LENGTH)
+                spaces = TRACE_GRAPH_PROCINFO_LENGTH - len;
+
+        /* First spaces to align center */
+        for (i = 0; i < spaces / 2; i++)
+                printf(" ");
+
+        printf("%s-%s", comm, pid_str);
+
+        /* Last spaces to align center */
+        for (i = 0; i < spaces - (spaces / 2); i++)
+                printf(" ");
+}
+
+static struct record *
+get_return_for_leaf(int cpu, int cur_pid, unsigned long long cur_func,
+                    struct record *next)
+{
+        struct format_field *field;
+        struct event *event;
+        unsigned long val;
+        int type;
+        int pid;
+
+        type = trace_parse_common_type(next->data);
+        event = trace_find_event(type);
+        if (!event)
+                return NULL;
+
+        if (!(event->flags & EVENT_FL_ISFUNCRET))
+                return NULL;
+
+        pid = trace_parse_common_pid(next->data);
+        field = find_field(event, "func");
+        if (!field)
+                die("function return does not have field func");
+
+        val = read_size(next->data + field->offset, field->size);
+
+        if (cur_pid != pid || cur_func != val)
+                return NULL;
+
+        /* this is a leaf, now advance the iterator */
+        return trace_read_data(cpu);
+}
+
+/* Signal a overhead of time execution to the output */
+static void print_graph_overhead(unsigned long long duration)
+{
+        /* Non nested entry or return */
+        if (duration == ~0ULL)
+                return (void)printf("  ");
+
+        /* Duration exceeded 100 msecs */
+        if (duration > 100000ULL)
+                return (void)printf("! ");
+
+        /* Duration exceeded 10 msecs */
+        if (duration > 10000ULL)
+                return (void)printf("+ ");
+
+        printf("  ");
+}
+
+static void print_graph_duration(unsigned long long duration)
+{
+        unsigned long usecs = duration / 1000;
+        unsigned long nsecs_rem = duration % 1000;
+        /* log10(ULONG_MAX) + '\0' */
+        char msecs_str[21];
+        char nsecs_str[5];
+        int len;
+        int i;
+
+        sprintf(msecs_str, "%lu", usecs);
+
+        /* Print msecs */
+        len = printf("%lu", usecs);
+
+        /* Print nsecs (we don't want to exceed 7 numbers) */
+        if (len < 7) {
+                snprintf(nsecs_str, 8 - len, "%03lu", nsecs_rem);
+                len += printf(".%s", nsecs_str);
+        }
+
+        printf(" us ");
+
+        /* Print remaining spaces to fit the row's width */
+        for (i = len; i < 7; i++)
+                printf(" ");
+
+        printf("|  ");
+}
+
+static void
+print_graph_entry_leaf(struct event *event, void *data, struct record *ret_rec)
+{
+        unsigned long long rettime, calltime;
+        unsigned long long duration, depth;
+        unsigned long long val;
+        struct format_field *field;
+        struct func_map *func;
+        struct event *ret_event;
+        int type;
+        int i;
+
+        type = trace_parse_common_type(ret_rec->data);
+        ret_event = trace_find_event(type);
+
+        field = find_field(ret_event, "rettime");
+        if (!field)
+                die("can't find rettime in return graph");
+        rettime = read_size(ret_rec->data + field->offset, field->size);
+
+        field = find_field(ret_event, "calltime");
+        if (!field)
+                die("can't find rettime in return graph");
+        calltime = read_size(ret_rec->data + field->offset, field->size);
+
+        duration = rettime - calltime;
+
+        /* Overhead */
+        print_graph_overhead(duration);
+
+        /* Duration */
+        print_graph_duration(duration);
+
+        field = find_field(event, "depth");
+        if (!field)
+                die("can't find depth in entry graph");
+        depth = read_size(data + field->offset, field->size);
+
+        /* Function */
+        for (i = 0; i < (int)(depth * TRACE_GRAPH_INDENT); i++)
+                printf(" ");
+
+        field = find_field(event, "func");
+        if (!field)
+                die("can't find func in entry graph");
+        val = read_size(data + field->offset, field->size);
+        func = find_func(val);
+
+        if (func)
+                printf("%s();", func->func);
+        else
+                printf("%llx();", val);
+}
+
+static void print_graph_nested(struct event *event, void *data)
+{
+        struct format_field *field;
+        unsigned long long depth;
+        unsigned long long val;
+        struct func_map *func;
+        int i;
+
+        /* No overhead */
+        print_graph_overhead(-1);
+
+        /* No time */
+        printf("           |  ");
+
+        field = find_field(event, "depth");
+        if (!field)
+                die("can't find depth in entry graph");
+        depth = read_size(data + field->offset, field->size);
+
+        /* Function */
+        for (i = 0; i < (int)(depth * TRACE_GRAPH_INDENT); i++)
+                printf(" ");
+
+        field = find_field(event, "func");
+        if (!field)
+                die("can't find func in entry graph");
+        val = read_size(data + field->offset, field->size);
+        func = find_func(val);
+
+        if (func)
+                printf("%s() {", func->func);
+        else
+                printf("%llx() {", val);
+}
+
+static void
+pretty_print_func_ent(void *data, int size, struct event *event,
+                      int cpu, int pid, const char *comm,
+                      unsigned long secs, unsigned long usecs)
+{
+        struct format_field *field;
+        struct record *rec;
+        void *copy_data;
+        unsigned long val;
+
+        printf("%5lu.%06lu |  ", secs, usecs);
+
+        print_graph_cpu(cpu);
+        print_graph_proc(pid, comm);
+
+        printf(" | ");
+
+        if (latency_format) {
+                print_lat_fmt(data, size);
+                printf(" | ");
+        }
+
+        field = find_field(event, "func");
+        if (!field)
+                die("function entry does not have func field");
+
+        val = read_size(data + field->offset, field->size);
+
+        /*
+         * peek_data may unmap the data pointer. Copy it first.
+         */
+        copy_data = malloc_or_die(size);
+        memcpy(copy_data, data, size);
+        data = copy_data;
+
+        rec = trace_peek_data(cpu);
+        if (rec) {
+                rec = get_return_for_leaf(cpu, pid, val, rec);
+                if (rec) {
+                        print_graph_entry_leaf(event, data, rec);
+                        goto out_free;
+                }
+        }
+        print_graph_nested(event, data);
+out_free:
+        free(data);
+}
+
+static void
+pretty_print_func_ret(void *data, int size __unused, struct event *event,
+                      int cpu, int pid, const char *comm,
+                      unsigned long secs, unsigned long usecs)
+{
+        unsigned long long rettime, calltime;
+        unsigned long long duration, depth;
+        struct format_field *field;
+        int i;
+
+        printf("%5lu.%06lu |  ", secs, usecs);
+
+        print_graph_cpu(cpu);
+        print_graph_proc(pid, comm);
+
+        printf(" | ");
+
+        if (latency_format) {
+                print_lat_fmt(data, size);
+                printf(" | ");
+        }
+
+        field = find_field(event, "rettime");
+        if (!field)
+                die("can't find rettime in return graph");
+        rettime = read_size(data + field->offset, field->size);
+
+        field = find_field(event, "calltime");
+        if (!field)
+                die("can't find calltime in return graph");
+        calltime = read_size(data + field->offset, field->size);
+
+        duration = rettime - calltime;
+
+        /* Overhead */
+        print_graph_overhead(duration);
+
+        /* Duration */
+        print_graph_duration(duration);
+
+        field = find_field(event, "depth");
+        if (!field)
+                die("can't find depth in entry graph");
+        depth = read_size(data + field->offset, field->size);
+
+        /* Function */
+        for (i = 0; i < (int)(depth * TRACE_GRAPH_INDENT); i++)
+                printf(" ");
+
+        printf("}");
+}
+
+static void
+pretty_print_func_graph(void *data, int size, struct event *event,
+                        int cpu, int pid, const char *comm,
+                        unsigned long secs, unsigned long usecs)
+{
+        if (event->flags & EVENT_FL_ISFUNCENT)
+                pretty_print_func_ent(data, size, event,
+                                      cpu, pid, comm, secs, usecs);
+        else if (event->flags & EVENT_FL_ISFUNCRET)
+                pretty_print_func_ret(data, size, event,
+                                      cpu, pid, comm, secs, usecs);
+        printf("\n");
+}
+
+void print_event(int cpu, void *data, int size, unsigned long long nsecs,
+                  char *comm)
+{
+        struct event *event;
+        unsigned long secs;
+        unsigned long usecs;
+        int type;
+        int pid;
+
+        secs = nsecs / NSECS_PER_SEC;
+        nsecs -= secs * NSECS_PER_SEC;
+        usecs = nsecs / NSECS_PER_USEC;
+
+        type = trace_parse_common_type(data);
+
+        event = trace_find_event(type);
+        if (!event) {
+                warning("ug! no event found for type %d", type);
+                return;
+        }
+
+        pid = trace_parse_common_pid(data);
+
+        if (event->flags & (EVENT_FL_ISFUNCENT | EVENT_FL_ISFUNCRET))
+                return pretty_print_func_graph(data, size, event, cpu,
+                                               pid, comm, secs, usecs);
+
+        if (latency_format) {
+                printf("%8.8s-%-5d %3d",
+                       comm, pid, cpu);
+                print_lat_fmt(data, size);
+        } else
+                printf("%16s-%-5d [%03d]", comm, pid,  cpu);
+
+        printf(" %5lu.%06lu: %s: ", secs, usecs, event->name);
+
+        if (event->flags & EVENT_FL_FAILED) {
+                printf("EVENT '%s' FAILED TO PARSE\n",
+                       event->name);
+                return;
+        }
+
+        pretty_print(data, size, event);
+        printf("\n");
+}
+
+static void print_fields(struct print_flag_sym *field)
+{
+        printf("{ %s, %s }", field->value, field->str);
+        if (field->next) {
+                printf(", ");
+                print_fields(field->next);
+        }
+}
+
+static void print_args(struct print_arg *args)
+{
+        int print_paren = 1;
+
+        switch (args->type) {
+        case PRINT_NULL:
+                printf("null");
+                break;
+        case PRINT_ATOM:
+                printf("%s", args->atom.atom);
+                break;
+        case PRINT_FIELD:
+                printf("REC->%s", args->field.name);
+                break;
+        case PRINT_FLAGS:
+                printf("__print_flags(");
+                print_args(args->flags.field);
+                printf(", %s, ", args->flags.delim);
+                print_fields(args->flags.flags);
+                printf(")");
+                break;
+        case PRINT_SYMBOL:
+                printf("__print_symbolic(");
+                print_args(args->symbol.field);
+                printf(", ");
+                print_fields(args->symbol.symbols);
+                printf(")");
+                break;
+        case PRINT_STRING:
+                printf("__get_str(%s)", args->string.string);
+                break;
+        case PRINT_TYPE:
+                printf("(%s)", args->typecast.type);
+                print_args(args->typecast.item);
+                break;
+        case PRINT_OP:
+                if (strcmp(args->op.op, ":") == 0)
+                        print_paren = 0;
+                if (print_paren)
+                        printf("(");
+                print_args(args->op.left);
+                printf(" %s ", args->op.op);
+                print_args(args->op.right);
+                if (print_paren)
+                        printf(")");
+                break;
+        default:
+                /* we should warn... */
+                return;
+        }
+        if (args->next) {
+                printf("\n");
+                print_args(args->next);
+        }
+}
+
+int parse_ftrace_file(char *buf, unsigned long size)
+{
+        struct format_field *field;
+        struct print_arg *arg, **list;
+        struct event *event;
+        int ret;
+
+        init_input_buf(buf, size);
+
+        event = alloc_event();
+        if (!event)
+                return -ENOMEM;
+
+        event->flags |= EVENT_FL_ISFTRACE;
+
+        event->name = event_read_name();
+        if (!event->name)
+                die("failed to read ftrace event name");
+
+        if (strcmp(event->name, "function") == 0)
+                event->flags |= EVENT_FL_ISFUNC;
+
+        else if (strcmp(event->name, "funcgraph_entry") == 0)
+                event->flags |= EVENT_FL_ISFUNCENT;
+
+        else if (strcmp(event->name, "funcgraph_exit") == 0)
+                event->flags |= EVENT_FL_ISFUNCRET;
+
+        else if (strcmp(event->name, "bprint") == 0)
+                event->flags |= EVENT_FL_ISBPRINT;
+
+        event->id = event_read_id();
+        if (event->id < 0)
+                die("failed to read ftrace event id");
+
+        add_event(event);
+
+        ret = event_read_format(event);
+        if (ret < 0)
+                die("failed to read ftrace event format");
+
+        ret = event_read_print(event);
+        if (ret < 0)
+                die("failed to read ftrace event print fmt");
+
+        /* New ftrace handles args */
+        if (ret > 0)
+                return 0;
+        /*
+         * The arguments for ftrace files are parsed by the fields.
+         * Set up the fields as their arguments.
+         */
+        list = &event->print_fmt.args;
+        for (field = event->format.fields; field; field = field->next) {
+                arg = malloc_or_die(sizeof(*arg));
+                memset(arg, 0, sizeof(*arg));
+                *list = arg;
+                list = &arg->next;
+                arg->type = PRINT_FIELD;
+                arg->field.name = field->name;
+                arg->field.field = field;
+        }
+        return 0;
+}
+
+int parse_event_file(char *buf, unsigned long size, char *sys)
+{
+        struct event *event;
+        int ret;
+
+        init_input_buf(buf, size);
+
+        event = alloc_event();
+        if (!event)
+                return -ENOMEM;
+
+        event->name = event_read_name();
+        if (!event->name)
+                die("failed to read event name");
+
+        event->id = event_read_id();
+        if (event->id < 0)
+                die("failed to read event id");
+
+        ret = event_read_format(event);
+        if (ret < 0) {
+                warning("failed to read event format for %s", event->name);
+                goto event_failed;
+        }
+
+        ret = event_read_print(event);
+        if (ret < 0) {
+                warning("failed to read event print fmt for %s", event->name);
+                goto event_failed;
+        }
+
+        event->system = strdup(sys);
+
+#define PRINT_ARGS 0
+        if (PRINT_ARGS && event->print_fmt.args)
+                print_args(event->print_fmt.args);
+
+        add_event(event);
+        return 0;
+
+ event_failed:
+        event->flags |= EVENT_FL_FAILED;
+        /* still add it even if it failed */
+        add_event(event);
+        return -1;
+}
+
+void parse_set_info(int nr_cpus, int long_sz)
+{
+        cpus = nr_cpus;
+        long_size = long_sz;
+}
+
+int common_pc(struct scripting_context *context)
+{
+        return parse_common_pc(context->event_data);
+}
+
+int common_flags(struct scripting_context *context)
+{
+        return parse_common_flags(context->event_data);
+}
+
+int common_lock_depth(struct scripting_context *context)
+{
+        return parse_common_lock_depth(context->event_data);
+}
diff --git a/tools/perf/util/trace-event-read.c b/tools/perf/util/trace-event-read.c
new file mode 100644
index 00000000000..f55cc3a765a
--- /dev/null
+++ b/tools/perf/util/trace-event-read.c
@@ -0,0 +1,539 @@
+/*
+ * Copyright (C) 2009, Steven Rostedt <srostedt@redhat.com>
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License (not later!)
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ */
+#define _FILE_OFFSET_BITS 64
+
+#include <dirent.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <getopt.h>
+#include <stdarg.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/wait.h>
+#include <sys/mman.h>
+#include <pthread.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include <ctype.h>
+#include <errno.h>
+
+#include "../perf.h"
+#include "util.h"
+#include "trace-event.h"
+
+static int input_fd;
+
+static int read_page;
+
+int file_bigendian;
+int host_bigendian;
+static int long_size;
+
+static unsigned long    page_size;
+
+static ssize_t calc_data_size;
+static bool repipe;
+
+static int do_read(int fd, void *buf, int size)
+{
+        int rsize = size;
+
+        while (size) {
+                int ret = read(fd, buf, size);
+
+                if (ret <= 0)
+                        return -1;
+
+                if (repipe) {
+                        int retw = write(STDOUT_FILENO, buf, ret);
+
+                        if (retw <= 0 || retw != ret)
+                                die("repiping input file");
+                }
+
+                size -= ret;
+                buf += ret;
+        }
+
+        return rsize;
+}
+
+static int read_or_die(void *data, int size)
+{
+        int r;
+
+        r = do_read(input_fd, data, size);
+        if (r <= 0)
+                die("reading input file (size expected=%d received=%d)",
+                    size, r);
+
+        if (calc_data_size)
+                calc_data_size += r;
+
+        return r;
+}
+
+/* If it fails, the next read will report it */
+static void skip(int size)
+{
+        char buf[BUFSIZ];
+        int r;
+
+        while (size) {
+                r = size > BUFSIZ ? BUFSIZ : size;
+                read_or_die(buf, r);
+                size -= r;
+        };
+}
+
+static unsigned int read4(void)
+{
+        unsigned int data;
+
+        read_or_die(&data, 4);
+        return __data2host4(data);
+}
+
+static unsigned long long read8(void)
+{
+        unsigned long long data;
+
+        read_or_die(&data, 8);
+        return __data2host8(data);
+}
+
+static char *read_string(void)
+{
+        char buf[BUFSIZ];
+        char *str = NULL;
+        int size = 0;
+        off_t r;
+        char c;
+
+        for (;;) {
+                r = read(input_fd, &c, 1);
+                if (r < 0)
+                        die("reading input file");
+
+                if (!r)
+                        die("no data");
+
+                if (repipe) {
+                        int retw = write(STDOUT_FILENO, &c, 1);
+
+                        if (retw <= 0 || retw != r)
+                                die("repiping input file string");
+                }
+
+                buf[size++] = c;
+
+                if (!c)
+                        break;
+        }
+
+        if (calc_data_size)
+                calc_data_size += size;
+
+        str = malloc_or_die(size);
+        memcpy(str, buf, size);
+
+        return str;
+}
+
+static void read_proc_kallsyms(void)
+{
+        unsigned int size;
+        char *buf;
+
+        size = read4();
+        if (!size)
+                return;
+
+        buf = malloc_or_die(size + 1);
+        read_or_die(buf, size);
+        buf[size] = '\0';
+
+        parse_proc_kallsyms(buf, size);
+
+        free(buf);
+}
+
+static void read_ftrace_printk(void)
+{
+        unsigned int size;
+        char *buf;
+
+        size = read4();
+        if (!size)
+                return;
+
+        buf = malloc_or_die(size);
+        read_or_die(buf, size);
+
+        parse_ftrace_printk(buf, size);
+
+        free(buf);
+}
+
+static void read_header_files(void)
+{
+        unsigned long long size;
+        char *header_event;
+        char buf[BUFSIZ];
+
+        read_or_die(buf, 12);
+
+        if (memcmp(buf, "header_page", 12) != 0)
+                die("did not read header page");
+
+        size = read8();
+        skip(size);
+
+        /*
+         * The size field in the page is of type long,
+         * use that instead, since it represents the kernel.
+         */
+        long_size = header_page_size_size;
+
+        read_or_die(buf, 13);
+        if (memcmp(buf, "header_event", 13) != 0)
+                die("did not read header event");
+
+        size = read8();
+        header_event = malloc_or_die(size);
+        read_or_die(header_event, size);
+        free(header_event);
+}
+
+static void read_ftrace_file(unsigned long long size)
+{
+        char *buf;
+
+        buf = malloc_or_die(size);
+        read_or_die(buf, size);
+        parse_ftrace_file(buf, size);
+        free(buf);
+}
+
+static void read_event_file(char *sys, unsigned long long size)
+{
+        char *buf;
+
+        buf = malloc_or_die(size);
+        read_or_die(buf, size);
+        parse_event_file(buf, size, sys);
+        free(buf);
+}
+
+static void read_ftrace_files(void)
+{
+        unsigned long long size;
+        int count;
+        int i;
+
+        count = read4();
+
+        for (i = 0; i < count; i++) {
+                size = read8();
+                read_ftrace_file(size);
+        }
+}
+
+static void read_event_files(void)
+{
+        unsigned long long size;
+        char *sys;
+        int systems;
+        int count;
+        int i,x;
+
+        systems = read4();
+
+        for (i = 0; i < systems; i++) {
+                sys = read_string();
+
+                count = read4();
+                for (x=0; x < count; x++) {
+                        size = read8();
+                        read_event_file(sys, size);
+                }
+        }
+}
+
+struct cpu_data {
+        unsigned long long      offset;
+        unsigned long long      size;
+        unsigned long long      timestamp;
+        struct record           *next;
+        char                    *page;
+        int                     cpu;
+        int                     index;
+        int                     page_size;
+};
+
+static struct cpu_data *cpu_data;
+
+static void update_cpu_data_index(int cpu)
+{
+        cpu_data[cpu].offset += page_size;
+        cpu_data[cpu].size -= page_size;
+        cpu_data[cpu].index = 0;
+}
+
+static void get_next_page(int cpu)
+{
+        off_t save_seek;
+        off_t ret;
+
+        if (!cpu_data[cpu].page)
+                return;
+
+        if (read_page) {
+                if (cpu_data[cpu].size <= page_size) {
+                        free(cpu_data[cpu].page);
+                        cpu_data[cpu].page = NULL;
+                        return;
+                }
+
+                update_cpu_data_index(cpu);
+
+                /* other parts of the code may expect the pointer to not move */
+                save_seek = lseek(input_fd, 0, SEEK_CUR);
+
+                ret = lseek(input_fd, cpu_data[cpu].offset, SEEK_SET);
+                if (ret == (off_t)-1)
+                        die("failed to lseek");
+                ret = read(input_fd, cpu_data[cpu].page, page_size);
+                if (ret < 0)
+                        die("failed to read page");
+
+                /* reset the file pointer back */
+                lseek(input_fd, save_seek, SEEK_SET);
+
+                return;
+        }
+
+        munmap(cpu_data[cpu].page, page_size);
+        cpu_data[cpu].page = NULL;
+
+        if (cpu_data[cpu].size <= page_size)
+                return;
+
+        update_cpu_data_index(cpu);
+
+        cpu_data[cpu].page = mmap(NULL, page_size, PROT_READ, MAP_PRIVATE,
+                                  input_fd, cpu_data[cpu].offset);
+        if (cpu_data[cpu].page == MAP_FAILED)
+                die("failed to mmap cpu %d at offset 0x%llx",
+                    cpu, cpu_data[cpu].offset);
+}
+
+static unsigned int type_len4host(unsigned int type_len_ts)
+{
+        if (file_bigendian)
+                return (type_len_ts >> 27) & ((1 << 5) - 1);
+        else
+                return type_len_ts & ((1 << 5) - 1);
+}
+
+static unsigned int ts4host(unsigned int type_len_ts)
+{
+        if (file_bigendian)
+                return type_len_ts & ((1 << 27) - 1);
+        else
+                return type_len_ts >> 5;
+}
+
+static int calc_index(void *ptr, int cpu)
+{
+        return (unsigned long)ptr - (unsigned long)cpu_data[cpu].page;
+}
+
+struct record *trace_peek_data(int cpu)
+{
+        struct record *data;
+        void *page = cpu_data[cpu].page;
+        int idx = cpu_data[cpu].index;
+        void *ptr = page + idx;
+        unsigned long long extend;
+        unsigned int type_len_ts;
+        unsigned int type_len;
+        unsigned int delta;
+        unsigned int length = 0;
+
+        if (cpu_data[cpu].next)
+                return cpu_data[cpu].next;
+
+        if (!page)
+                return NULL;
+
+        if (!idx) {
+                /* FIXME: handle header page */
+                if (header_page_ts_size != 8)
+                        die("expected a long long type for timestamp");
+                cpu_data[cpu].timestamp = data2host8(ptr);
+                ptr += 8;
+                switch (header_page_size_size) {
+                case 4:
+                        cpu_data[cpu].page_size = data2host4(ptr);
+                        ptr += 4;
+                        break;
+                case 8:
+                        cpu_data[cpu].page_size = data2host8(ptr);
+                        ptr += 8;
+                        break;
+                default:
+                        die("bad long size");
+                }
+                ptr = cpu_data[cpu].page + header_page_data_offset;
+        }
+
+read_again:
+        idx = calc_index(ptr, cpu);
+
+        if (idx >= cpu_data[cpu].page_size) {
+                get_next_page(cpu);
+                return trace_peek_data(cpu);
+        }
+
+        type_len_ts = data2host4(ptr);
+        ptr += 4;
+
+        type_len = type_len4host(type_len_ts);
+        delta = ts4host(type_len_ts);
+
+        switch (type_len) {
+        case RINGBUF_TYPE_PADDING:
+                if (!delta)
+                        die("error, hit unexpected end of page");
+                length = data2host4(ptr);
+                ptr += 4;
+                length *= 4;
+                ptr += length;
+                goto read_again;
+
+        case RINGBUF_TYPE_TIME_EXTEND:
+                extend = data2host4(ptr);
+                ptr += 4;
+                extend <<= TS_SHIFT;
+                extend += delta;
+                cpu_data[cpu].timestamp += extend;
+                goto read_again;
+
+        case RINGBUF_TYPE_TIME_STAMP:
+                ptr += 12;
+                break;
+        case 0:
+                length = data2host4(ptr);
+                ptr += 4;
+                die("here! length=%d", length);
+                break;
+        default:
+                length = type_len * 4;
+                break;
+        }
+
+        cpu_data[cpu].timestamp += delta;
+
+        data = malloc_or_die(sizeof(*data));
+        memset(data, 0, sizeof(*data));
+
+        data->ts = cpu_data[cpu].timestamp;
+        data->size = length;
+        data->data = ptr;
+        ptr += length;
+
+        cpu_data[cpu].index = calc_index(ptr, cpu);
+        cpu_data[cpu].next = data;
+
+        return data;
+}
+
+struct record *trace_read_data(int cpu)
+{
+        struct record *data;
+
+        data = trace_peek_data(cpu);
+        cpu_data[cpu].next = NULL;
+
+        return data;
+}
+
+ssize_t trace_report(int fd, bool __repipe)
+{
+        char buf[BUFSIZ];
+        char test[] = { 23, 8, 68 };
+        char *version;
+        int show_version = 0;
+        int show_funcs = 0;
+        int show_printk = 0;
+        ssize_t size;
+
+        calc_data_size = 1;
+        repipe = __repipe;
+
+        input_fd = fd;
+
+        read_or_die(buf, 3);
+        if (memcmp(buf, test, 3) != 0)
+                die("no trace data in the file");
+
+        read_or_die(buf, 7);
+        if (memcmp(buf, "tracing", 7) != 0)
+                die("not a trace file (missing 'tracing' tag)");
+
+        version = read_string();
+        if (show_version)
+                printf("version = %s\n", version);
+        free(version);
+
+        read_or_die(buf, 1);
+        file_bigendian = buf[0];
+        host_bigendian = bigendian();
+
+        read_or_die(buf, 1);
+        long_size = buf[0];
+
+        page_size = read4();
+
+        read_header_files();
+
+        read_ftrace_files();
+        read_event_files();
+        read_proc_kallsyms();
+        read_ftrace_printk();
+
+        size = calc_data_size - 1;
+        calc_data_size = 0;
+        repipe = false;
+
+        if (show_funcs) {
+                print_funcs();
+                return size;
+        }
+        if (show_printk) {
+                print_printk();
+                return size;
+        }
+
+        return size;
+}
diff --git a/tools/perf/util/trace-event-scripting.c b/tools/perf/util/trace-event-scripting.c
new file mode 100644
index 00000000000..f7af2fca965
--- /dev/null
+++ b/tools/perf/util/trace-event-scripting.c
@@ -0,0 +1,167 @@
+/*
+ * trace-event-scripting.  Scripting engine common and initialization code.
+ *
+ * Copyright (C) 2009-2010 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+#include <errno.h>
+
+#include "../perf.h"
+#include "util.h"
+#include "trace-event.h"
+
+struct scripting_context *scripting_context;
+
+static int stop_script_unsupported(void)
+{
+        return 0;
+}
+
+static void process_event_unsupported(int cpu __unused,
+                                      void *data __unused,
+                                      int size __unused,
+                                      unsigned long long nsecs __unused,
+                                      char *comm __unused)
+{
+}
+
+static void print_python_unsupported_msg(void)
+{
+        fprintf(stderr, "Python scripting not supported."
+                "  Install libpython and rebuild perf to enable it.\n"
+                "For example:\n  # apt-get install python-dev (ubuntu)"
+                "\n  # yum install python-devel (Fedora)"
+                "\n  etc.\n");
+}
+
+static int python_start_script_unsupported(const char *script __unused,
+                                           int argc __unused,
+                                           const char **argv __unused)
+{
+        print_python_unsupported_msg();
+
+        return -1;
+}
+
+static int python_generate_script_unsupported(const char *outfile __unused)
+{
+        print_python_unsupported_msg();
+
+        return -1;
+}
+
+struct scripting_ops python_scripting_unsupported_ops = {
+        .name = "Python",
+        .start_script = python_start_script_unsupported,
+        .stop_script = stop_script_unsupported,
+        .process_event = process_event_unsupported,
+        .generate_script = python_generate_script_unsupported,
+};
+
+static void register_python_scripting(struct scripting_ops *scripting_ops)
+{
+        int err;
+        err = script_spec_register("Python", scripting_ops);
+        if (err)
+                die("error registering Python script extension");
+
+        err = script_spec_register("py", scripting_ops);
+        if (err)
+                die("error registering py script extension");
+
+        scripting_context = malloc(sizeof(struct scripting_context));
+}
+
+#ifdef NO_LIBPYTHON
+void setup_python_scripting(void)
+{
+        register_python_scripting(&python_scripting_unsupported_ops);
+}
+#else
+extern struct scripting_ops python_scripting_ops;
+
+void setup_python_scripting(void)
+{
+        register_python_scripting(&python_scripting_ops);
+}
+#endif
+
+static void print_perl_unsupported_msg(void)
+{
+        fprintf(stderr, "Perl scripting not supported."
+                "  Install libperl and rebuild perf to enable it.\n"
+                "For example:\n  # apt-get install libperl-dev (ubuntu)"
+                "\n  # yum install 'perl(ExtUtils::Embed)' (Fedora)"
+                "\n  etc.\n");
+}
+
+static int perl_start_script_unsupported(const char *script __unused,
+                                         int argc __unused,
+                                         const char **argv __unused)
+{
+        print_perl_unsupported_msg();
+
+        return -1;
+}
+
+static int perl_generate_script_unsupported(const char *outfile __unused)
+{
+        print_perl_unsupported_msg();
+
+        return -1;
+}
+
+struct scripting_ops perl_scripting_unsupported_ops = {
+        .name = "Perl",
+        .start_script = perl_start_script_unsupported,
+        .stop_script = stop_script_unsupported,
+        .process_event = process_event_unsupported,
+        .generate_script = perl_generate_script_unsupported,
+};
+
+static void register_perl_scripting(struct scripting_ops *scripting_ops)
+{
+        int err;
+        err = script_spec_register("Perl", scripting_ops);
+        if (err)
+                die("error registering Perl script extension");
+
+        err = script_spec_register("pl", scripting_ops);
+        if (err)
+                die("error registering pl script extension");
+
+        scripting_context = malloc(sizeof(struct scripting_context));
+}
+
+#ifdef NO_LIBPERL
+void setup_perl_scripting(void)
+{
+        register_perl_scripting(&perl_scripting_unsupported_ops);
+}
+#else
+extern struct scripting_ops perl_scripting_ops;
+
+void setup_perl_scripting(void)
+{
+        register_perl_scripting(&perl_scripting_ops);
+}
+#endif
diff --git a/tools/perf/util/trace-event.h b/tools/perf/util/trace-event.h
new file mode 100644
index 00000000000..b5f12ca24d9
--- /dev/null
+++ b/tools/perf/util/trace-event.h
@@ -0,0 +1,299 @@
+#ifndef __PERF_TRACE_EVENTS_H
+#define __PERF_TRACE_EVENTS_H
+
+#include <stdbool.h>
+#include "parse-events.h"
+
+#define __unused __attribute__((unused))
+
+
+#ifndef PAGE_MASK
+#define PAGE_MASK (page_size - 1)
+#endif
+
+enum {
+        RINGBUF_TYPE_PADDING            = 29,
+        RINGBUF_TYPE_TIME_EXTEND        = 30,
+        RINGBUF_TYPE_TIME_STAMP         = 31,
+};
+
+#ifndef TS_SHIFT
+#define TS_SHIFT                27
+#endif
+
+#define NSECS_PER_SEC           1000000000ULL
+#define NSECS_PER_USEC          1000ULL
+
+enum format_flags {
+        FIELD_IS_ARRAY          = 1,
+        FIELD_IS_POINTER        = 2,
+        FIELD_IS_SIGNED         = 4,
+        FIELD_IS_STRING         = 8,
+        FIELD_IS_DYNAMIC        = 16,
+        FIELD_IS_FLAG           = 32,
+        FIELD_IS_SYMBOLIC       = 64,
+};
+
+struct format_field {
+        struct format_field     *next;
+        char                    *type;
+        char                    *name;
+        int                     offset;
+        int                     size;
+        unsigned long           flags;
+};
+
+struct format {
+        int                     nr_common;
+        int                     nr_fields;
+        struct format_field     *common_fields;
+        struct format_field     *fields;
+};
+
+struct print_arg_atom {
+        char                    *atom;
+};
+
+struct print_arg_string {
+        char                    *string;
+        int                     offset;
+};
+
+struct print_arg_field {
+        char                    *name;
+        struct format_field     *field;
+};
+
+struct print_flag_sym {
+        struct print_flag_sym   *next;
+        char                    *value;
+        char                    *str;
+};
+
+struct print_arg_typecast {
+        char                    *type;
+        struct print_arg        *item;
+};
+
+struct print_arg_flags {
+        struct print_arg        *field;
+        char                    *delim;
+        struct print_flag_sym   *flags;
+};
+
+struct print_arg_symbol {
+        struct print_arg        *field;
+        struct print_flag_sym   *symbols;
+};
+
+struct print_arg;
+
+struct print_arg_op {
+        char                    *op;
+        int                     prio;
+        struct print_arg        *left;
+        struct print_arg        *right;
+};
+
+struct print_arg_func {
+        char                    *name;
+        struct print_arg        *args;
+};
+
+enum print_arg_type {
+        PRINT_NULL,
+        PRINT_ATOM,
+        PRINT_FIELD,
+        PRINT_FLAGS,
+        PRINT_SYMBOL,
+        PRINT_TYPE,
+        PRINT_STRING,
+        PRINT_OP,
+};
+
+struct print_arg {
+        struct print_arg                *next;
+        enum print_arg_type             type;
+        union {
+                struct print_arg_atom           atom;
+                struct print_arg_field          field;
+                struct print_arg_typecast       typecast;
+                struct print_arg_flags          flags;
+                struct print_arg_symbol         symbol;
+                struct print_arg_func           func;
+                struct print_arg_string         string;
+                struct print_arg_op             op;
+        };
+};
+
+struct print_fmt {
+        char                    *format;
+        struct print_arg        *args;
+};
+
+struct event {
+        struct event            *next;
+        char                    *name;
+        int                     id;
+        int                     flags;
+        struct format           format;
+        struct print_fmt        print_fmt;
+        char                    *system;
+};
+
+enum {
+        EVENT_FL_ISFTRACE       = 0x01,
+        EVENT_FL_ISPRINT        = 0x02,
+        EVENT_FL_ISBPRINT       = 0x04,
+        EVENT_FL_ISFUNC         = 0x08,
+        EVENT_FL_ISFUNCENT      = 0x10,
+        EVENT_FL_ISFUNCRET      = 0x20,
+
+        EVENT_FL_FAILED         = 0x80000000
+};
+
+struct record {
+        unsigned long long ts;
+        int size;
+        void *data;
+};
+
+struct record *trace_peek_data(int cpu);
+struct record *trace_read_data(int cpu);
+
+void parse_set_info(int nr_cpus, int long_sz);
+
+ssize_t trace_report(int fd, bool repipe);
+
+void *malloc_or_die(unsigned int size);
+
+void parse_cmdlines(char *file, int size);
+void parse_proc_kallsyms(char *file, unsigned int size);
+void parse_ftrace_printk(char *file, unsigned int size);
+
+void print_funcs(void);
+void print_printk(void);
+
+int parse_ftrace_file(char *buf, unsigned long size);
+int parse_event_file(char *buf, unsigned long size, char *sys);
+void print_event(int cpu, void *data, int size, unsigned long long nsecs,
+                  char *comm);
+
+extern int file_bigendian;
+extern int host_bigendian;
+
+int bigendian(void);
+
+static inline unsigned short __data2host2(unsigned short data)
+{
+        unsigned short swap;
+
+        if (host_bigendian == file_bigendian)
+                return data;
+
+        swap = ((data & 0xffULL) << 8) |
+                ((data & (0xffULL << 8)) >> 8);
+
+        return swap;
+}
+
+static inline unsigned int __data2host4(unsigned int data)
+{
+        unsigned int swap;
+
+        if (host_bigendian == file_bigendian)
+                return data;
+
+        swap = ((data & 0xffULL) << 24) |
+                ((data & (0xffULL << 8)) << 8) |
+                ((data & (0xffULL << 16)) >> 8) |
+                ((data & (0xffULL << 24)) >> 24);
+
+        return swap;
+}
+
+static inline unsigned long long __data2host8(unsigned long long data)
+{
+        unsigned long long swap;
+
+        if (host_bigendian == file_bigendian)
+                return data;
+
+        swap = ((data & 0xffULL) << 56) |
+                ((data & (0xffULL << 8)) << 40) |
+                ((data & (0xffULL << 16)) << 24) |
+                ((data & (0xffULL << 24)) << 8) |
+                ((data & (0xffULL << 32)) >> 8) |
+                ((data & (0xffULL << 40)) >> 24) |
+                ((data & (0xffULL << 48)) >> 40) |
+                ((data & (0xffULL << 56)) >> 56);
+
+        return swap;
+}
+
+#define data2host2(ptr)         __data2host2(*(unsigned short *)ptr)
+#define data2host4(ptr)         __data2host4(*(unsigned int *)ptr)
+#define data2host8(ptr)         ({                              \
+        unsigned long long __val;                               \
+                                                                \
+        memcpy(&__val, (ptr), sizeof(unsigned long long));      \
+        __data2host8(__val);                                    \
+})
+
+extern int header_page_ts_offset;
+extern int header_page_ts_size;
+extern int header_page_size_offset;
+extern int header_page_size_size;
+extern int header_page_data_offset;
+extern int header_page_data_size;
+
+extern bool latency_format;
+
+int trace_parse_common_type(void *data);
+int trace_parse_common_pid(void *data);
+int parse_common_pc(void *data);
+int parse_common_flags(void *data);
+int parse_common_lock_depth(void *data);
+struct event *trace_find_event(int id);
+struct event *trace_find_next_event(struct event *event);
+unsigned long long read_size(void *ptr, int size);
+unsigned long long
+raw_field_value(struct event *event, const char *name, void *data);
+void *raw_field_ptr(struct event *event, const char *name, void *data);
+unsigned long long eval_flag(const char *flag);
+
+int read_tracing_data(int fd, struct list_head *pattrs);
+ssize_t read_tracing_data_size(int fd, struct list_head *pattrs);
+
+/* taken from kernel/trace/trace.h */
+enum trace_flag_type {
+        TRACE_FLAG_IRQS_OFF             = 0x01,
+        TRACE_FLAG_IRQS_NOSUPPORT       = 0x02,
+        TRACE_FLAG_NEED_RESCHED         = 0x04,
+        TRACE_FLAG_HARDIRQ              = 0x08,
+        TRACE_FLAG_SOFTIRQ              = 0x10,
+};
+
+struct scripting_ops {
+        const char *name;
+        int (*start_script) (const char *script, int argc, const char **argv);
+        int (*stop_script) (void);
+        void (*process_event) (int cpu, void *data, int size,
+                               unsigned long long nsecs, char *comm);
+        int (*generate_script) (const char *outfile);
+};
+
+int script_spec_register(const char *spec, struct scripting_ops *ops);
+
+void setup_perl_scripting(void);
+void setup_python_scripting(void);
+
+struct scripting_context {
+        void *event_data;
+};
+
+int common_pc(struct scripting_context *context);
+int common_flags(struct scripting_context *context);
+int common_lock_depth(struct scripting_context *context);
+
+#endif /* __PERF_TRACE_EVENTS_H */
diff --git a/tools/perf/types.h b/tools/perf/util/types.h
index 5e75f900594..7d6b8331f89 100644
--- a/tools/perf/types.h
+++ b/tools/perf/util/types.h
@@ -1,5 +1,5 @@
-#ifndef _PERF_TYPES_H
-#define _PERF_TYPES_H
+#ifndef __PERF_TYPES_H
+#define __PERF_TYPES_H
 
 /*
  * We define u64 as unsigned long long for every architecture
@@ -14,4 +14,4 @@ typedef signed short	   s16;
 typedef unsigned char      u8;
 typedef signed char        s8;
 
-#endif /* _PERF_TYPES_H */
+#endif /* __PERF_TYPES_H */
diff --git a/tools/perf/util/ui/browser.c b/tools/perf/util/ui/browser.c
new file mode 100644
index 00000000000..8bc010edca2
--- /dev/null
+++ b/tools/perf/util/ui/browser.c
@@ -0,0 +1,337 @@
+#include "libslang.h"
+#include <linux/compiler.h>
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include <stdlib.h>
+#include <sys/ttydefaults.h>
+#include "browser.h"
+#include "helpline.h"
+#include "../color.h"
+#include "../util.h"
+#include <stdio.h>
+
+static int ui_browser__percent_color(double percent, bool current)
+{
+        if (current)
+                return HE_COLORSET_SELECTED;
+        if (percent >= MIN_RED)
+                return HE_COLORSET_TOP;
+        if (percent >= MIN_GREEN)
+                return HE_COLORSET_MEDIUM;
+        return HE_COLORSET_NORMAL;
+}
+
+void ui_browser__set_color(struct ui_browser *self __used, int color)
+{
+        SLsmg_set_color(color);
+}
+
+void ui_browser__set_percent_color(struct ui_browser *self,
+                                   double percent, bool current)
+{
+         int color = ui_browser__percent_color(percent, current);
+         ui_browser__set_color(self, color);
+}
+
+void ui_browser__gotorc(struct ui_browser *self, int y, int x)
+{
+        SLsmg_gotorc(self->y + y, self->x + x);
+}
+
+void ui_browser__list_head_seek(struct ui_browser *self, off_t offset, int whence)
+{
+        struct list_head *head = self->entries;
+        struct list_head *pos;
+
+        switch (whence) {
+        case SEEK_SET:
+                pos = head->next;
+                break;
+        case SEEK_CUR:
+                pos = self->top;
+                break;
+        case SEEK_END:
+                pos = head->prev;
+                break;
+        default:
+                return;
+        }
+
+        if (offset > 0) {
+                while (offset-- != 0)
+                        pos = pos->next;
+        } else {
+                while (offset++ != 0)
+                        pos = pos->prev;
+        }
+
+        self->top = pos;
+}
+
+void ui_browser__rb_tree_seek(struct ui_browser *self, off_t offset, int whence)
+{
+        struct rb_root *root = self->entries;
+        struct rb_node *nd;
+
+        switch (whence) {
+        case SEEK_SET:
+                nd = rb_first(root);
+                break;
+        case SEEK_CUR:
+                nd = self->top;
+                break;
+        case SEEK_END:
+                nd = rb_last(root);
+                break;
+        default:
+                return;
+        }
+
+        if (offset > 0) {
+                while (offset-- != 0)
+                        nd = rb_next(nd);
+        } else {
+                while (offset++ != 0)
+                        nd = rb_prev(nd);
+        }
+
+        self->top = nd;
+}
+
+unsigned int ui_browser__rb_tree_refresh(struct ui_browser *self)
+{
+        struct rb_node *nd;
+        int row = 0;
+
+        if (self->top == NULL)
+                self->top = rb_first(self->entries);
+
+        nd = self->top;
+
+        while (nd != NULL) {
+                ui_browser__gotorc(self, row, 0);
+                self->write(self, nd, row);
+                if (++row == self->height)
+                        break;
+                nd = rb_next(nd);
+        }
+
+        return row;
+}
+
+bool ui_browser__is_current_entry(struct ui_browser *self, unsigned row)
+{
+        return self->top_idx + row == self->index;
+}
+
+void ui_browser__refresh_dimensions(struct ui_browser *self)
+{
+        int cols, rows;
+        newtGetScreenSize(&cols, &rows);
+
+        self->width = cols - 1;
+        self->height = rows - 2;
+        self->y = 1;
+        self->x = 0;
+}
+
+void ui_browser__reset_index(struct ui_browser *self)
+{
+        self->index = self->top_idx = 0;
+        self->seek(self, 0, SEEK_SET);
+}
+
+void ui_browser__add_exit_key(struct ui_browser *self, int key)
+{
+        newtFormAddHotKey(self->form, key);
+}
+
+void ui_browser__add_exit_keys(struct ui_browser *self, int keys[])
+{
+        int i = 0;
+
+        while (keys[i] && i < 64) {
+                ui_browser__add_exit_key(self, keys[i]);
+                ++i;
+        }
+}
+
+int ui_browser__show(struct ui_browser *self, const char *title,
+                     const char *helpline, ...)
+{
+        va_list ap;
+        int keys[] = { NEWT_KEY_UP, NEWT_KEY_DOWN, NEWT_KEY_PGUP,
+                       NEWT_KEY_PGDN, NEWT_KEY_HOME, NEWT_KEY_END, ' ',
+                       NEWT_KEY_LEFT, NEWT_KEY_ESCAPE, 'q', CTRL('c'), 0 };
+
+        if (self->form != NULL)
+                newtFormDestroy(self->form);
+
+        ui_browser__refresh_dimensions(self);
+        self->form = newtForm(NULL, NULL, 0);
+        if (self->form == NULL)
+                return -1;
+
+        self->sb = newtVerticalScrollbar(self->width, 1, self->height,
+                                         HE_COLORSET_NORMAL,
+                                         HE_COLORSET_SELECTED);
+        if (self->sb == NULL)
+                return -1;
+
+        SLsmg_gotorc(0, 0);
+        ui_browser__set_color(self, NEWT_COLORSET_ROOT);
+        slsmg_write_nstring(title, self->width);
+
+        ui_browser__add_exit_keys(self, keys);
+        newtFormAddComponent(self->form, self->sb);
+
+        va_start(ap, helpline);
+        ui_helpline__vpush(helpline, ap);
+        va_end(ap);
+        return 0;
+}
+
+void ui_browser__hide(struct ui_browser *self)
+{
+        newtFormDestroy(self->form);
+        self->form = NULL;
+        ui_helpline__pop();
+}
+
+int ui_browser__refresh(struct ui_browser *self)
+{
+        int row;
+
+        newtScrollbarSet(self->sb, self->index, self->nr_entries - 1);
+        row = self->refresh(self);
+        ui_browser__set_color(self, HE_COLORSET_NORMAL);
+        SLsmg_fill_region(self->y + row, self->x,
+                          self->height - row, self->width, ' ');
+
+        return 0;
+}
+
+int ui_browser__run(struct ui_browser *self)
+{
+        struct newtExitStruct es;
+
+        if (ui_browser__refresh(self) < 0)
+                return -1;
+
+        while (1) {
+                off_t offset;
+
+                newtFormRun(self->form, &es);
+
+                if (es.reason != NEWT_EXIT_HOTKEY)
+                        break;
+                switch (es.u.key) {
+                case NEWT_KEY_DOWN:
+                        if (self->index == self->nr_entries - 1)
+                                break;
+                        ++self->index;
+                        if (self->index == self->top_idx + self->height) {
+                                ++self->top_idx;
+                                self->seek(self, +1, SEEK_CUR);
+                        }
+                        break;
+                case NEWT_KEY_UP:
+                        if (self->index == 0)
+                                break;
+                        --self->index;
+                        if (self->index < self->top_idx) {
+                                --self->top_idx;
+                                self->seek(self, -1, SEEK_CUR);
+                        }
+                        break;
+                case NEWT_KEY_PGDN:
+                case ' ':
+                        if (self->top_idx + self->height > self->nr_entries - 1)
+                                break;
+
+                        offset = self->height;
+                        if (self->index + offset > self->nr_entries - 1)
+                                offset = self->nr_entries - 1 - self->index;
+                        self->index += offset;
+                        self->top_idx += offset;
+                        self->seek(self, +offset, SEEK_CUR);
+                        break;
+                case NEWT_KEY_PGUP:
+                        if (self->top_idx == 0)
+                                break;
+
+                        if (self->top_idx < self->height)
+                                offset = self->top_idx;
+                        else
+                                offset = self->height;
+
+                        self->index -= offset;
+                        self->top_idx -= offset;
+                        self->seek(self, -offset, SEEK_CUR);
+                        break;
+                case NEWT_KEY_HOME:
+                        ui_browser__reset_index(self);
+                        break;
+                case NEWT_KEY_END:
+                        offset = self->height - 1;
+                        if (offset >= self->nr_entries)
+                                offset = self->nr_entries - 1;
+
+                        self->index = self->nr_entries - 1;
+                        self->top_idx = self->index - offset;
+                        self->seek(self, -offset, SEEK_END);
+                        break;
+                default:
+                        return es.u.key;
+                }
+                if (ui_browser__refresh(self) < 0)
+                        return -1;
+        }
+        return -1;
+}
+
+unsigned int ui_browser__list_head_refresh(struct ui_browser *self)
+{
+        struct list_head *pos;
+        struct list_head *head = self->entries;
+        int row = 0;
+
+        if (self->top == NULL || self->top == self->entries)
+                self->top = head->next;
+
+        pos = self->top;
+
+        list_for_each_from(pos, head) {
+                ui_browser__gotorc(self, row, 0);
+                self->write(self, pos, row);
+                if (++row == self->height)
+                        break;
+        }
+
+        return row;
+}
+
+static struct newtPercentTreeColors {
+        const char *topColorFg, *topColorBg;
+        const char *mediumColorFg, *mediumColorBg;
+        const char *normalColorFg, *normalColorBg;
+        const char *selColorFg, *selColorBg;
+        const char *codeColorFg, *codeColorBg;
+} defaultPercentTreeColors = {
+        "red",       "lightgray",
+        "green",     "lightgray",
+        "black",     "lightgray",
+        "lightgray", "magenta",
+        "blue",      "lightgray",
+};
+
+void ui_browser__init(void)
+{
+        struct newtPercentTreeColors *c = &defaultPercentTreeColors;
+
+        sltt_set_color(HE_COLORSET_TOP, NULL, c->topColorFg, c->topColorBg);
+        sltt_set_color(HE_COLORSET_MEDIUM, NULL, c->mediumColorFg, c->mediumColorBg);
+        sltt_set_color(HE_COLORSET_NORMAL, NULL, c->normalColorFg, c->normalColorBg);
+        sltt_set_color(HE_COLORSET_SELECTED, NULL, c->selColorFg, c->selColorBg);
+        sltt_set_color(HE_COLORSET_CODE, NULL, c->codeColorFg, c->codeColorBg);
+}
diff --git a/tools/perf/util/ui/browser.h b/tools/perf/util/ui/browser.h
new file mode 100644
index 00000000000..0dc7e4da36f
--- /dev/null
+++ b/tools/perf/util/ui/browser.h
@@ -0,0 +1,51 @@
+#ifndef _PERF_UI_BROWSER_H_
+#define _PERF_UI_BROWSER_H_ 1
+
+#include <stdbool.h>
+#include <newt.h>
+#include <sys/types.h>
+#include "../types.h"
+
+#define HE_COLORSET_TOP         50
+#define HE_COLORSET_MEDIUM      51
+#define HE_COLORSET_NORMAL      52
+#define HE_COLORSET_SELECTED    53
+#define HE_COLORSET_CODE        54
+
+struct ui_browser {
+        newtComponent form, sb;
+        u64           index, top_idx;
+        void          *top, *entries;
+        u16           y, x, width, height;
+        void          *priv;
+        unsigned int  (*refresh)(struct ui_browser *self);
+        void          (*write)(struct ui_browser *self, void *entry, int row);
+        void          (*seek)(struct ui_browser *self, off_t offset, int whence);
+        u32           nr_entries;
+};
+
+
+void ui_browser__set_color(struct ui_browser *self, int color);
+void ui_browser__set_percent_color(struct ui_browser *self,
+                                   double percent, bool current);
+bool ui_browser__is_current_entry(struct ui_browser *self, unsigned row);
+void ui_browser__refresh_dimensions(struct ui_browser *self);
+void ui_browser__reset_index(struct ui_browser *self);
+
+void ui_browser__gotorc(struct ui_browser *self, int y, int x);
+void ui_browser__add_exit_key(struct ui_browser *self, int key);
+void ui_browser__add_exit_keys(struct ui_browser *self, int keys[]);
+int ui_browser__show(struct ui_browser *self, const char *title,
+                     const char *helpline, ...);
+void ui_browser__hide(struct ui_browser *self);
+int ui_browser__refresh(struct ui_browser *self);
+int ui_browser__run(struct ui_browser *self);
+
+void ui_browser__rb_tree_seek(struct ui_browser *self, off_t offset, int whence);
+unsigned int ui_browser__rb_tree_refresh(struct ui_browser *self);
+
+void ui_browser__list_head_seek(struct ui_browser *self, off_t offset, int whence);
+unsigned int ui_browser__list_head_refresh(struct ui_browser *self);
+
+void ui_browser__init(void);
+#endif /* _PERF_UI_BROWSER_H_ */
diff --git a/tools/perf/util/ui/browsers/annotate.c b/tools/perf/util/ui/browsers/annotate.c
new file mode 100644
index 00000000000..82b78f99251
--- /dev/null
+++ b/tools/perf/util/ui/browsers/annotate.c
@@ -0,0 +1,237 @@
+#include "../browser.h"
+#include "../helpline.h"
+#include "../libslang.h"
+#include "../../hist.h"
+#include "../../sort.h"
+#include "../../symbol.h"
+
+static void ui__error_window(const char *fmt, ...)
+{
+        va_list ap;
+
+        va_start(ap, fmt);
+        newtWinMessagev((char *)"Error", (char *)"Ok", (char *)fmt, ap);
+        va_end(ap);
+}
+
+struct annotate_browser {
+        struct ui_browser b;
+        struct rb_root    entries;
+        struct rb_node    *curr_hot;
+};
+
+struct objdump_line_rb_node {
+        struct rb_node  rb_node;
+        double          percent;
+        u32             idx;
+};
+
+static inline
+struct objdump_line_rb_node *objdump_line__rb(struct objdump_line *self)
+{
+        return (struct objdump_line_rb_node *)(self + 1);
+}
+
+static void annotate_browser__write(struct ui_browser *self, void *entry, int row)
+{
+        struct objdump_line *ol = rb_entry(entry, struct objdump_line, node);
+        bool current_entry = ui_browser__is_current_entry(self, row);
+        int width = self->width;
+
+        if (ol->offset != -1) {
+                struct objdump_line_rb_node *olrb = objdump_line__rb(ol);
+                ui_browser__set_percent_color(self, olrb->percent, current_entry);
+                slsmg_printf(" %7.2f ", olrb->percent);
+                if (!current_entry)
+                        ui_browser__set_color(self, HE_COLORSET_CODE);
+        } else {
+                ui_browser__set_percent_color(self, 0, current_entry);
+                slsmg_write_nstring(" ", 9);
+        }
+
+        SLsmg_write_char(':');
+        slsmg_write_nstring(" ", 8);
+        if (!*ol->line)
+                slsmg_write_nstring(" ", width - 18);
+        else
+                slsmg_write_nstring(ol->line, width - 18);
+}
+
+static double objdump_line__calc_percent(struct objdump_line *self,
+                                         struct list_head *head,
+                                         struct symbol *sym)
+{
+        double percent = 0.0;
+
+        if (self->offset != -1) {
+                int len = sym->end - sym->start;
+                unsigned int hits = 0;
+                struct sym_priv *priv = symbol__priv(sym);
+                struct sym_ext *sym_ext = priv->ext;
+                struct sym_hist *h = priv->hist;
+                s64 offset = self->offset;
+                struct objdump_line *next = objdump__get_next_ip_line(head, self);
+
+
+                while (offset < (s64)len &&
+                       (next == NULL || offset < next->offset)) {
+                        if (sym_ext) {
+                                percent += sym_ext[offset].percent;
+                        } else
+                                hits += h->ip[offset];
+
+                        ++offset;
+                }
+
+                if (sym_ext == NULL && h->sum)
+                        percent = 100.0 * hits / h->sum;
+        }
+
+        return percent;
+}
+
+static void objdump__insert_line(struct rb_root *self,
+                                 struct objdump_line_rb_node *line)
+{
+        struct rb_node **p = &self->rb_node;
+        struct rb_node *parent = NULL;
+        struct objdump_line_rb_node *l;
+
+        while (*p != NULL) {
+                parent = *p;
+                l = rb_entry(parent, struct objdump_line_rb_node, rb_node);
+                if (line->percent < l->percent)
+                        p = &(*p)->rb_left;
+                else
+                        p = &(*p)->rb_right;
+        }
+        rb_link_node(&line->rb_node, parent, p);
+        rb_insert_color(&line->rb_node, self);
+}
+
+static void annotate_browser__set_top(struct annotate_browser *self,
+                                      struct rb_node *nd)
+{
+        struct objdump_line_rb_node *rbpos;
+        struct objdump_line *pos;
+        unsigned back;
+
+        ui_browser__refresh_dimensions(&self->b);
+        back = self->b.height / 2;
+        rbpos = rb_entry(nd, struct objdump_line_rb_node, rb_node);
+        pos = ((struct objdump_line *)rbpos) - 1;
+        self->b.top_idx = self->b.index = rbpos->idx;
+
+        while (self->b.top_idx != 0 && back != 0) {
+                pos = list_entry(pos->node.prev, struct objdump_line, node);
+
+                --self->b.top_idx;
+                --back;
+        }
+
+        self->b.top = pos;
+        self->curr_hot = nd;
+}
+
+static int annotate_browser__run(struct annotate_browser *self)
+{
+        struct rb_node *nd;
+        struct hist_entry *he = self->b.priv;
+        int key;
+
+        if (ui_browser__show(&self->b, he->ms.sym->name,
+                             "<-, -> or ESC: exit, TAB/shift+TAB: cycle thru samples") < 0)
+                return -1;
+        /*
+         * To allow builtin-annotate to cycle thru multiple symbols by
+         * examining the exit key for this function.
+         */
+        ui_browser__add_exit_key(&self->b, NEWT_KEY_RIGHT);
+
+        nd = self->curr_hot;
+        if (nd) {
+                int tabs[] = { NEWT_KEY_TAB, NEWT_KEY_UNTAB, 0 };
+                ui_browser__add_exit_keys(&self->b, tabs);
+        }
+
+        while (1) {
+                key = ui_browser__run(&self->b);
+
+                switch (key) {
+                case NEWT_KEY_TAB:
+                        nd = rb_prev(nd);
+                        if (nd == NULL)
+                                nd = rb_last(&self->entries);
+                        annotate_browser__set_top(self, nd);
+                        break;
+                case NEWT_KEY_UNTAB:
+                        nd = rb_next(nd);
+                        if (nd == NULL)
+                                nd = rb_first(&self->entries);
+                        annotate_browser__set_top(self, nd);
+                        break;
+                default:
+                        goto out;
+                }
+        }
+out:
+        ui_browser__hide(&self->b);
+        return key;
+}
+
+int hist_entry__tui_annotate(struct hist_entry *self)
+{
+        struct objdump_line *pos, *n;
+        struct objdump_line_rb_node *rbpos;
+        LIST_HEAD(head);
+        struct annotate_browser browser = {
+                .b = {
+                        .entries = &head,
+                        .refresh = ui_browser__list_head_refresh,
+                        .seek    = ui_browser__list_head_seek,
+                        .write   = annotate_browser__write,
+                        .priv    = self,
+                },
+        };
+        int ret;
+
+        if (self->ms.sym == NULL)
+                return -1;
+
+        if (self->ms.map->dso->annotate_warned)
+                return -1;
+
+        if (hist_entry__annotate(self, &head, sizeof(*rbpos)) < 0) {
+                ui__error_window(ui_helpline__last_msg);
+                return -1;
+        }
+
+        ui_helpline__push("Press <- or ESC to exit");
+
+        list_for_each_entry(pos, &head, node) {
+                size_t line_len = strlen(pos->line);
+                if (browser.b.width < line_len)
+                        browser.b.width = line_len;
+                rbpos = objdump_line__rb(pos);
+                rbpos->idx = browser.b.nr_entries++;
+                rbpos->percent = objdump_line__calc_percent(pos, &head, self->ms.sym);
+                if (rbpos->percent < 0.01)
+                        continue;
+                objdump__insert_line(&browser.entries, rbpos);
+        }
+
+        /*
+         * Position the browser at the hottest line.
+         */
+        browser.curr_hot = rb_last(&browser.entries);
+        if (browser.curr_hot)
+                annotate_browser__set_top(&browser, browser.curr_hot);
+
+        browser.b.width += 18; /* Percentage */
+        ret = annotate_browser__run(&browser);
+        list_for_each_entry_safe(pos, n, &head, node) {
+                list_del(&pos->node);
+                objdump_line__free(pos);
+        }
+        return ret;
+}
diff --git a/tools/perf/util/ui/browsers/hists.c b/tools/perf/util/ui/browsers/hists.c
new file mode 100644
index 00000000000..ebda8c3fde9
--- /dev/null
+++ b/tools/perf/util/ui/browsers/hists.c
@@ -0,0 +1,1013 @@
+#define _GNU_SOURCE
+#include <stdio.h>
+#undef _GNU_SOURCE
+#include "../libslang.h"
+#include <stdlib.h>
+#include <string.h>
+#include <newt.h>
+#include <linux/rbtree.h>
+
+#include "../../hist.h"
+#include "../../pstack.h"
+#include "../../sort.h"
+#include "../../util.h"
+
+#include "../browser.h"
+#include "../helpline.h"
+#include "../util.h"
+#include "map.h"
+
+struct hist_browser {
+        struct ui_browser   b;
+        struct hists        *hists;
+        struct hist_entry   *he_selection;
+        struct map_symbol   *selection;
+};
+
+static void hist_browser__refresh_dimensions(struct hist_browser *self)
+{
+        /* 3 == +/- toggle symbol before actual hist_entry rendering */
+        self->b.width = 3 + (hists__sort_list_width(self->hists) +
+                             sizeof("[k]"));
+}
+
+static void hist_browser__reset(struct hist_browser *self)
+{
+        self->b.nr_entries = self->hists->nr_entries;
+        hist_browser__refresh_dimensions(self);
+        ui_browser__reset_index(&self->b);
+}
+
+static char tree__folded_sign(bool unfolded)
+{
+        return unfolded ? '-' : '+';
+}
+
+static char map_symbol__folded(const struct map_symbol *self)
+{
+        return self->has_children ? tree__folded_sign(self->unfolded) : ' ';
+}
+
+static char hist_entry__folded(const struct hist_entry *self)
+{
+        return map_symbol__folded(&self->ms);
+}
+
+static char callchain_list__folded(const struct callchain_list *self)
+{
+        return map_symbol__folded(&self->ms);
+}
+
+static void map_symbol__set_folding(struct map_symbol *self, bool unfold)
+{
+        self->unfolded = unfold ? self->has_children : false;
+}
+
+static int callchain_node__count_rows_rb_tree(struct callchain_node *self)
+{
+        int n = 0;
+        struct rb_node *nd;
+
+        for (nd = rb_first(&self->rb_root); nd; nd = rb_next(nd)) {
+                struct callchain_node *child = rb_entry(nd, struct callchain_node, rb_node);
+                struct callchain_list *chain;
+                char folded_sign = ' '; /* No children */
+
+                list_for_each_entry(chain, &child->val, list) {
+                        ++n;
+                        /* We need this because we may not have children */
+                        folded_sign = callchain_list__folded(chain);
+                        if (folded_sign == '+')
+                                break;
+                }
+
+                if (folded_sign == '-') /* Have children and they're unfolded */
+                        n += callchain_node__count_rows_rb_tree(child);
+        }
+
+        return n;
+}
+
+static int callchain_node__count_rows(struct callchain_node *node)
+{
+        struct callchain_list *chain;
+        bool unfolded = false;
+        int n = 0;
+
+        list_for_each_entry(chain, &node->val, list) {
+                ++n;
+                unfolded = chain->ms.unfolded;
+        }
+
+        if (unfolded)
+                n += callchain_node__count_rows_rb_tree(node);
+
+        return n;
+}
+
+static int callchain__count_rows(struct rb_root *chain)
+{
+        struct rb_node *nd;
+        int n = 0;
+
+        for (nd = rb_first(chain); nd; nd = rb_next(nd)) {
+                struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node);
+                n += callchain_node__count_rows(node);
+        }
+
+        return n;
+}
+
+static bool map_symbol__toggle_fold(struct map_symbol *self)
+{
+        if (!self->has_children)
+                return false;
+
+        self->unfolded = !self->unfolded;
+        return true;
+}
+
+static void callchain_node__init_have_children_rb_tree(struct callchain_node *self)
+{
+        struct rb_node *nd = rb_first(&self->rb_root);
+
+        for (nd = rb_first(&self->rb_root); nd; nd = rb_next(nd)) {
+                struct callchain_node *child = rb_entry(nd, struct callchain_node, rb_node);
+                struct callchain_list *chain;
+                bool first = true;
+
+                list_for_each_entry(chain, &child->val, list) {
+                        if (first) {
+                                first = false;
+                                chain->ms.has_children = chain->list.next != &child->val ||
+                                                         !RB_EMPTY_ROOT(&child->rb_root);
+                        } else
+                                chain->ms.has_children = chain->list.next == &child->val &&
+                                                         !RB_EMPTY_ROOT(&child->rb_root);
+                }
+
+                callchain_node__init_have_children_rb_tree(child);
+        }
+}
+
+static void callchain_node__init_have_children(struct callchain_node *self)
+{
+        struct callchain_list *chain;
+
+        list_for_each_entry(chain, &self->val, list)
+                chain->ms.has_children = !RB_EMPTY_ROOT(&self->rb_root);
+
+        callchain_node__init_have_children_rb_tree(self);
+}
+
+static void callchain__init_have_children(struct rb_root *self)
+{
+        struct rb_node *nd;
+
+        for (nd = rb_first(self); nd; nd = rb_next(nd)) {
+                struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node);
+                callchain_node__init_have_children(node);
+        }
+}
+
+static void hist_entry__init_have_children(struct hist_entry *self)
+{
+        if (!self->init_have_children) {
+                self->ms.has_children = !RB_EMPTY_ROOT(&self->sorted_chain);
+                callchain__init_have_children(&self->sorted_chain);
+                self->init_have_children = true;
+        }
+}
+
+static bool hist_browser__toggle_fold(struct hist_browser *self)
+{
+        if (map_symbol__toggle_fold(self->selection)) {
+                struct hist_entry *he = self->he_selection;
+
+                hist_entry__init_have_children(he);
+                self->hists->nr_entries -= he->nr_rows;
+
+                if (he->ms.unfolded)
+                        he->nr_rows = callchain__count_rows(&he->sorted_chain);
+                else
+                        he->nr_rows = 0;
+                self->hists->nr_entries += he->nr_rows;
+                self->b.nr_entries = self->hists->nr_entries;
+
+                return true;
+        }
+
+        /* If it doesn't have children, no toggling performed */
+        return false;
+}
+
+static int callchain_node__set_folding_rb_tree(struct callchain_node *self, bool unfold)
+{
+        int n = 0;
+        struct rb_node *nd;
+
+        for (nd = rb_first(&self->rb_root); nd; nd = rb_next(nd)) {
+                struct callchain_node *child = rb_entry(nd, struct callchain_node, rb_node);
+                struct callchain_list *chain;
+                bool has_children = false;
+
+                list_for_each_entry(chain, &child->val, list) {
+                        ++n;
+                        map_symbol__set_folding(&chain->ms, unfold);
+                        has_children = chain->ms.has_children;
+                }
+
+                if (has_children)
+                        n += callchain_node__set_folding_rb_tree(child, unfold);
+        }
+
+        return n;
+}
+
+static int callchain_node__set_folding(struct callchain_node *node, bool unfold)
+{
+        struct callchain_list *chain;
+        bool has_children = false;
+        int n = 0;
+
+        list_for_each_entry(chain, &node->val, list) {
+                ++n;
+                map_symbol__set_folding(&chain->ms, unfold);
+                has_children = chain->ms.has_children;
+        }
+
+        if (has_children)
+                n += callchain_node__set_folding_rb_tree(node, unfold);
+
+        return n;
+}
+
+static int callchain__set_folding(struct rb_root *chain, bool unfold)
+{
+        struct rb_node *nd;
+        int n = 0;
+
+        for (nd = rb_first(chain); nd; nd = rb_next(nd)) {
+                struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node);
+                n += callchain_node__set_folding(node, unfold);
+        }
+
+        return n;
+}
+
+static void hist_entry__set_folding(struct hist_entry *self, bool unfold)
+{
+        hist_entry__init_have_children(self);
+        map_symbol__set_folding(&self->ms, unfold);
+
+        if (self->ms.has_children) {
+                int n = callchain__set_folding(&self->sorted_chain, unfold);
+                self->nr_rows = unfold ? n : 0;
+        } else
+                self->nr_rows = 0;
+}
+
+static void hists__set_folding(struct hists *self, bool unfold)
+{
+        struct rb_node *nd;
+
+        self->nr_entries = 0;
+
+        for (nd = rb_first(&self->entries); nd; nd = rb_next(nd)) {
+                struct hist_entry *he = rb_entry(nd, struct hist_entry, rb_node);
+                hist_entry__set_folding(he, unfold);
+                self->nr_entries += 1 + he->nr_rows;
+        }
+}
+
+static void hist_browser__set_folding(struct hist_browser *self, bool unfold)
+{
+        hists__set_folding(self->hists, unfold);
+        self->b.nr_entries = self->hists->nr_entries;
+        /* Go to the start, we may be way after valid entries after a collapse */
+        ui_browser__reset_index(&self->b);
+}
+
+static int hist_browser__run(struct hist_browser *self, const char *title)
+{
+        int key;
+        int exit_keys[] = { 'a', '?', 'h', 'C', 'd', 'D', 'E', 't',
+                            NEWT_KEY_ENTER, NEWT_KEY_RIGHT, NEWT_KEY_LEFT, 0, };
+
+        self->b.entries = &self->hists->entries;
+        self->b.nr_entries = self->hists->nr_entries;
+
+        hist_browser__refresh_dimensions(self);
+
+        if (ui_browser__show(&self->b, title,
+                             "Press '?' for help on key bindings") < 0)
+                return -1;
+
+        ui_browser__add_exit_keys(&self->b, exit_keys);
+
+        while (1) {
+                key = ui_browser__run(&self->b);
+
+                switch (key) {
+                case 'D': { /* Debug */
+                        static int seq;
+                        struct hist_entry *h = rb_entry(self->b.top,
+                                                        struct hist_entry, rb_node);
+                        ui_helpline__pop();
+                        ui_helpline__fpush("%d: nr_ent=(%d,%d), height=%d, idx=%d, fve: idx=%d, row_off=%d, nrows=%d",
+                                           seq++, self->b.nr_entries,
+                                           self->hists->nr_entries,
+                                           self->b.height,
+                                           self->b.index,
+                                           self->b.top_idx,
+                                           h->row_offset, h->nr_rows);
+                }
+                        break;
+                case 'C':
+                        /* Collapse the whole world. */
+                        hist_browser__set_folding(self, false);
+                        break;
+                case 'E':
+                        /* Expand the whole world. */
+                        hist_browser__set_folding(self, true);
+                        break;
+                case NEWT_KEY_ENTER:
+                        if (hist_browser__toggle_fold(self))
+                                break;
+                        /* fall thru */
+                default:
+                        goto out;
+                }
+        }
+out:
+        ui_browser__hide(&self->b);
+        return key;
+}
+
+static char *callchain_list__sym_name(struct callchain_list *self,
+                                      char *bf, size_t bfsize)
+{
+        if (self->ms.sym)
+                return self->ms.sym->name;
+
+        snprintf(bf, bfsize, "%#Lx", self->ip);
+        return bf;
+}
+
+#define LEVEL_OFFSET_STEP 3
+
+static int hist_browser__show_callchain_node_rb_tree(struct hist_browser *self,
+                                                     struct callchain_node *chain_node,
+                                                     u64 total, int level,
+                                                     unsigned short row,
+                                                     off_t *row_offset,
+                                                     bool *is_current_entry)
+{
+        struct rb_node *node;
+        int first_row = row, width, offset = level * LEVEL_OFFSET_STEP;
+        u64 new_total, remaining;
+
+        if (callchain_param.mode == CHAIN_GRAPH_REL)
+                new_total = chain_node->children_hit;
+        else
+                new_total = total;
+
+        remaining = new_total;
+        node = rb_first(&chain_node->rb_root);
+        while (node) {
+                struct callchain_node *child = rb_entry(node, struct callchain_node, rb_node);
+                struct rb_node *next = rb_next(node);
+                u64 cumul = cumul_hits(child);
+                struct callchain_list *chain;
+                char folded_sign = ' ';
+                int first = true;
+                int extra_offset = 0;
+
+                remaining -= cumul;
+
+                list_for_each_entry(chain, &child->val, list) {
+                        char ipstr[BITS_PER_LONG / 4 + 1], *alloc_str;
+                        const char *str;
+                        int color;
+                        bool was_first = first;
+
+                        if (first)
+                                first = false;
+                        else
+                                extra_offset = LEVEL_OFFSET_STEP;
+
+                        folded_sign = callchain_list__folded(chain);
+                        if (*row_offset != 0) {
+                                --*row_offset;
+                                goto do_next;
+                        }
+
+                        alloc_str = NULL;
+                        str = callchain_list__sym_name(chain, ipstr, sizeof(ipstr));
+                        if (was_first) {
+                                double percent = cumul * 100.0 / new_total;
+
+                                if (asprintf(&alloc_str, "%2.2f%% %s", percent, str) < 0)
+                                        str = "Not enough memory!";
+                                else
+                                        str = alloc_str;
+                        }
+
+                        color = HE_COLORSET_NORMAL;
+                        width = self->b.width - (offset + extra_offset + 2);
+                        if (ui_browser__is_current_entry(&self->b, row)) {
+                                self->selection = &chain->ms;
+                                color = HE_COLORSET_SELECTED;
+                                *is_current_entry = true;
+                        }
+
+                        ui_browser__set_color(&self->b, color);
+                        ui_browser__gotorc(&self->b, row, 0);
+                        slsmg_write_nstring(" ", offset + extra_offset);
+                        slsmg_printf("%c ", folded_sign);
+                        slsmg_write_nstring(str, width);
+                        free(alloc_str);
+
+                        if (++row == self->b.height)
+                                goto out;
+do_next:
+                        if (folded_sign == '+')
+                                break;
+                }
+
+                if (folded_sign == '-') {
+                        const int new_level = level + (extra_offset ? 2 : 1);
+                        row += hist_browser__show_callchain_node_rb_tree(self, child, new_total,
+                                                                         new_level, row, row_offset,
+                                                                         is_current_entry);
+                }
+                if (row == self->b.height)
+                        goto out;
+                node = next;
+        }
+out:
+        return row - first_row;
+}
+
+static int hist_browser__show_callchain_node(struct hist_browser *self,
+                                             struct callchain_node *node,
+                                             int level, unsigned short row,
+                                             off_t *row_offset,
+                                             bool *is_current_entry)
+{
+        struct callchain_list *chain;
+        int first_row = row,
+             offset = level * LEVEL_OFFSET_STEP,
+             width = self->b.width - offset;
+        char folded_sign = ' ';
+
+        list_for_each_entry(chain, &node->val, list) {
+                char ipstr[BITS_PER_LONG / 4 + 1], *s;
+                int color;
+
+                folded_sign = callchain_list__folded(chain);
+
+                if (*row_offset != 0) {
+                        --*row_offset;
+                        continue;
+                }
+
+                color = HE_COLORSET_NORMAL;
+                if (ui_browser__is_current_entry(&self->b, row)) {
+                        self->selection = &chain->ms;
+                        color = HE_COLORSET_SELECTED;
+                        *is_current_entry = true;
+                }
+
+                s = callchain_list__sym_name(chain, ipstr, sizeof(ipstr));
+                ui_browser__gotorc(&self->b, row, 0);
+                ui_browser__set_color(&self->b, color);
+                slsmg_write_nstring(" ", offset);
+                slsmg_printf("%c ", folded_sign);
+                slsmg_write_nstring(s, width - 2);
+
+                if (++row == self->b.height)
+                        goto out;
+        }
+
+        if (folded_sign == '-')
+                row += hist_browser__show_callchain_node_rb_tree(self, node,
+                                                                 self->hists->stats.total_period,
+                                                                 level + 1, row,
+                                                                 row_offset,
+                                                                 is_current_entry);
+out:
+        return row - first_row;
+}
+
+static int hist_browser__show_callchain(struct hist_browser *self,
+                                        struct rb_root *chain,
+                                        int level, unsigned short row,
+                                        off_t *row_offset,
+                                        bool *is_current_entry)
+{
+        struct rb_node *nd;
+        int first_row = row;
+
+        for (nd = rb_first(chain); nd; nd = rb_next(nd)) {
+                struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node);
+
+                row += hist_browser__show_callchain_node(self, node, level,
+                                                         row, row_offset,
+                                                         is_current_entry);
+                if (row == self->b.height)
+                        break;
+        }
+
+        return row - first_row;
+}
+
+static int hist_browser__show_entry(struct hist_browser *self,
+                                    struct hist_entry *entry,
+                                    unsigned short row)
+{
+        char s[256];
+        double percent;
+        int printed = 0;
+        int color, width = self->b.width;
+        char folded_sign = ' ';
+        bool current_entry = ui_browser__is_current_entry(&self->b, row);
+        off_t row_offset = entry->row_offset;
+
+        if (current_entry) {
+                self->he_selection = entry;
+                self->selection = &entry->ms;
+        }
+
+        if (symbol_conf.use_callchain) {
+                hist_entry__init_have_children(entry);
+                folded_sign = hist_entry__folded(entry);
+        }
+
+        if (row_offset == 0) {
+                hist_entry__snprintf(entry, s, sizeof(s), self->hists, NULL, false,
+                                     0, false, self->hists->stats.total_period);
+                percent = (entry->period * 100.0) / self->hists->stats.total_period;
+
+                color = HE_COLORSET_SELECTED;
+                if (!current_entry) {
+                        if (percent >= MIN_RED)
+                                color = HE_COLORSET_TOP;
+                        else if (percent >= MIN_GREEN)
+                                color = HE_COLORSET_MEDIUM;
+                        else
+                                color = HE_COLORSET_NORMAL;
+                }
+
+                ui_browser__set_color(&self->b, color);
+                ui_browser__gotorc(&self->b, row, 0);
+                if (symbol_conf.use_callchain) {
+                        slsmg_printf("%c ", folded_sign);
+                        width -= 2;
+                }
+                slsmg_write_nstring(s, width);
+                ++row;
+                ++printed;
+        } else
+                --row_offset;
+
+        if (folded_sign == '-' && row != self->b.height) {
+                printed += hist_browser__show_callchain(self, &entry->sorted_chain,
+                                                        1, row, &row_offset,
+                                                        &current_entry);
+                if (current_entry)
+                        self->he_selection = entry;
+        }
+
+        return printed;
+}
+
+static unsigned int hist_browser__refresh(struct ui_browser *self)
+{
+        unsigned row = 0;
+        struct rb_node *nd;
+        struct hist_browser *hb = container_of(self, struct hist_browser, b);
+
+        if (self->top == NULL)
+                self->top = rb_first(&hb->hists->entries);
+
+        for (nd = self->top; nd; nd = rb_next(nd)) {
+                struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+
+                if (h->filtered)
+                        continue;
+
+                row += hist_browser__show_entry(hb, h, row);
+                if (row == self->height)
+                        break;
+        }
+
+        return row;
+}
+
+static struct rb_node *hists__filter_entries(struct rb_node *nd)
+{
+        while (nd != NULL) {
+                struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+                if (!h->filtered)
+                        return nd;
+
+                nd = rb_next(nd);
+        }
+
+        return NULL;
+}
+
+static struct rb_node *hists__filter_prev_entries(struct rb_node *nd)
+{
+        while (nd != NULL) {
+                struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+                if (!h->filtered)
+                        return nd;
+
+                nd = rb_prev(nd);
+        }
+
+        return NULL;
+}
+
+static void ui_browser__hists_seek(struct ui_browser *self,
+                                   off_t offset, int whence)
+{
+        struct hist_entry *h;
+        struct rb_node *nd;
+        bool first = true;
+
+        switch (whence) {
+        case SEEK_SET:
+                nd = hists__filter_entries(rb_first(self->entries));
+                break;
+        case SEEK_CUR:
+                nd = self->top;
+                goto do_offset;
+        case SEEK_END:
+                nd = hists__filter_prev_entries(rb_last(self->entries));
+                first = false;
+                break;
+        default:
+                return;
+        }
+
+        /*
+         * Moves not relative to the first visible entry invalidates its
+         * row_offset:
+         */
+        h = rb_entry(self->top, struct hist_entry, rb_node);
+        h->row_offset = 0;
+
+        /*
+         * Here we have to check if nd is expanded (+), if it is we can't go
+         * the next top level hist_entry, instead we must compute an offset of
+         * what _not_ to show and not change the first visible entry.
+         *
+         * This offset increments when we are going from top to bottom and
+         * decreases when we're going from bottom to top.
+         *
+         * As we don't have backpointers to the top level in the callchains
+         * structure, we need to always print the whole hist_entry callchain,
+         * skipping the first ones that are before the first visible entry
+         * and stop when we printed enough lines to fill the screen.
+         */
+do_offset:
+        if (offset > 0) {
+                do {
+                        h = rb_entry(nd, struct hist_entry, rb_node);
+                        if (h->ms.unfolded) {
+                                u16 remaining = h->nr_rows - h->row_offset;
+                                if (offset > remaining) {
+                                        offset -= remaining;
+                                        h->row_offset = 0;
+                                } else {
+                                        h->row_offset += offset;
+                                        offset = 0;
+                                        self->top = nd;
+                                        break;
+                                }
+                        }
+                        nd = hists__filter_entries(rb_next(nd));
+                        if (nd == NULL)
+                                break;
+                        --offset;
+                        self->top = nd;
+                } while (offset != 0);
+        } else if (offset < 0) {
+                while (1) {
+                        h = rb_entry(nd, struct hist_entry, rb_node);
+                        if (h->ms.unfolded) {
+                                if (first) {
+                                        if (-offset > h->row_offset) {
+                                                offset += h->row_offset;
+                                                h->row_offset = 0;
+                                        } else {
+                                                h->row_offset += offset;
+                                                offset = 0;
+                                                self->top = nd;
+                                                break;
+                                        }
+                                } else {
+                                        if (-offset > h->nr_rows) {
+                                                offset += h->nr_rows;
+                                                h->row_offset = 0;
+                                        } else {
+                                                h->row_offset = h->nr_rows + offset;
+                                                offset = 0;
+                                                self->top = nd;
+                                                break;
+                                        }
+                                }
+                        }
+
+                        nd = hists__filter_prev_entries(rb_prev(nd));
+                        if (nd == NULL)
+                                break;
+                        ++offset;
+                        self->top = nd;
+                        if (offset == 0) {
+                                /*
+                                 * Last unfiltered hist_entry, check if it is
+                                 * unfolded, if it is then we should have
+                                 * row_offset at its last entry.
+                                 */
+                                h = rb_entry(nd, struct hist_entry, rb_node);
+                                if (h->ms.unfolded)
+                                        h->row_offset = h->nr_rows;
+                                break;
+                        }
+                        first = false;
+                }
+        } else {
+                self->top = nd;
+                h = rb_entry(nd, struct hist_entry, rb_node);
+                h->row_offset = 0;
+        }
+}
+
+static struct hist_browser *hist_browser__new(struct hists *hists)
+{
+        struct hist_browser *self = zalloc(sizeof(*self));
+
+        if (self) {
+                self->hists = hists;
+                self->b.refresh = hist_browser__refresh;
+                self->b.seek = ui_browser__hists_seek;
+        }
+
+        return self;
+}
+
+static void hist_browser__delete(struct hist_browser *self)
+{
+        free(self);
+}
+
+static struct hist_entry *hist_browser__selected_entry(struct hist_browser *self)
+{
+        return self->he_selection;
+}
+
+static struct thread *hist_browser__selected_thread(struct hist_browser *self)
+{
+        return self->he_selection->thread;
+}
+
+static int hists__browser_title(struct hists *self, char *bf, size_t size,
+                                const char *ev_name, const struct dso *dso,
+                                const struct thread *thread)
+{
+        char unit;
+        int printed;
+        unsigned long nr_events = self->stats.nr_events[PERF_RECORD_SAMPLE];
+
+        nr_events = convert_unit(nr_events, &unit);
+        printed = snprintf(bf, size, "Events: %lu%c %s", nr_events, unit, ev_name);
+
+        if (thread)
+                printed += snprintf(bf + printed, size - printed,
+                                    ", Thread: %s(%d)",
+                                    (thread->comm_set ? thread->comm : ""),
+                                    thread->pid);
+        if (dso)
+                printed += snprintf(bf + printed, size - printed,
+                                    ", DSO: %s", dso->short_name);
+        return printed;
+}
+
+int hists__browse(struct hists *self, const char *helpline, const char *ev_name)
+{
+        struct hist_browser *browser = hist_browser__new(self);
+        struct pstack *fstack;
+        const struct thread *thread_filter = NULL;
+        const struct dso *dso_filter = NULL;
+        char msg[160];
+        int key = -1;
+
+        if (browser == NULL)
+                return -1;
+
+        fstack = pstack__new(2);
+        if (fstack == NULL)
+                goto out;
+
+        ui_helpline__push(helpline);
+
+        hists__browser_title(self, msg, sizeof(msg), ev_name,
+                             dso_filter, thread_filter);
+        while (1) {
+                const struct thread *thread;
+                const struct dso *dso;
+                char *options[16];
+                int nr_options = 0, choice = 0, i,
+                    annotate = -2, zoom_dso = -2, zoom_thread = -2,
+                    browse_map = -2;
+
+                key = hist_browser__run(browser, msg);
+
+                thread = hist_browser__selected_thread(browser);
+                dso = browser->selection->map ? browser->selection->map->dso : NULL;
+
+                switch (key) {
+                case NEWT_KEY_TAB:
+                case NEWT_KEY_UNTAB:
+                        /*
+                         * Exit the browser, let hists__browser_tree
+                         * go to the next or previous
+                         */
+                        goto out_free_stack;
+                case 'a':
+                        if (browser->selection->map == NULL &&
+                            browser->selection->map->dso->annotate_warned)
+                                continue;
+                        goto do_annotate;
+                case 'd':
+                        goto zoom_dso;
+                case 't':
+                        goto zoom_thread;
+                case NEWT_KEY_F1:
+                case 'h':
+                case '?':
+                        ui__help_window("->        Zoom into DSO/Threads & Annotate current symbol\n"
+                                        "<-        Zoom out\n"
+                                        "a         Annotate current symbol\n"
+                                        "h/?/F1    Show this window\n"
+                                        "C         Collapse all callchains\n"
+                                        "E         Expand all callchains\n"
+                                        "d         Zoom into current DSO\n"
+                                        "t         Zoom into current Thread\n"
+                                        "q/CTRL+C  Exit browser");
+                        continue;
+                case NEWT_KEY_ENTER:
+                case NEWT_KEY_RIGHT:
+                        /* menu */
+                        break;
+                case NEWT_KEY_LEFT: {
+                        const void *top;
+
+                        if (pstack__empty(fstack))
+                                continue;
+                        top = pstack__pop(fstack);
+                        if (top == &dso_filter)
+                                goto zoom_out_dso;
+                        if (top == &thread_filter)
+                                goto zoom_out_thread;
+                        continue;
+                }
+                case NEWT_KEY_ESCAPE:
+                        if (!ui__dialog_yesno("Do you really want to exit?"))
+                                continue;
+                        /* Fall thru */
+                default:
+                        goto out_free_stack;
+                }
+
+                if (browser->selection->sym != NULL &&
+                    !browser->selection->map->dso->annotate_warned &&
+                    asprintf(&options[nr_options], "Annotate %s",
+                             browser->selection->sym->name) > 0)
+                        annotate = nr_options++;
+
+                if (thread != NULL &&
+                    asprintf(&options[nr_options], "Zoom %s %s(%d) thread",
+                             (thread_filter ? "out of" : "into"),
+                             (thread->comm_set ? thread->comm : ""),
+                             thread->pid) > 0)
+                        zoom_thread = nr_options++;
+
+                if (dso != NULL &&
+                    asprintf(&options[nr_options], "Zoom %s %s DSO",
+                             (dso_filter ? "out of" : "into"),
+                             (dso->kernel ? "the Kernel" : dso->short_name)) > 0)
+                        zoom_dso = nr_options++;
+
+                if (browser->selection->map != NULL &&
+                    asprintf(&options[nr_options], "Browse map details") > 0)
+                        browse_map = nr_options++;
+
+                options[nr_options++] = (char *)"Exit";
+
+                choice = ui__popup_menu(nr_options, options);
+
+                for (i = 0; i < nr_options - 1; ++i)
+                        free(options[i]);
+
+                if (choice == nr_options - 1)
+                        break;
+
+                if (choice == -1)
+                        continue;
+
+                if (choice == annotate) {
+                        struct hist_entry *he;
+do_annotate:
+                        if (browser->selection->map->dso->origin == DSO__ORIG_KERNEL) {
+                                browser->selection->map->dso->annotate_warned = 1;
+                                ui_helpline__puts("No vmlinux file found, can't "
+                                                 "annotate with just a "
+                                                 "kallsyms file");
+                                continue;
+                        }
+
+                        he = hist_browser__selected_entry(browser);
+                        if (he == NULL)
+                                continue;
+
+                        hist_entry__tui_annotate(he);
+                } else if (choice == browse_map)
+                        map__browse(browser->selection->map);
+                else if (choice == zoom_dso) {
+zoom_dso:
+                        if (dso_filter) {
+                                pstack__remove(fstack, &dso_filter);
+zoom_out_dso:
+                                ui_helpline__pop();
+                                dso_filter = NULL;
+                        } else {
+                                if (dso == NULL)
+                                        continue;
+                                ui_helpline__fpush("To zoom out press <- or -> + \"Zoom out of %s DSO\"",
+                                                   dso->kernel ? "the Kernel" : dso->short_name);
+                                dso_filter = dso;
+                                pstack__push(fstack, &dso_filter);
+                        }
+                        hists__filter_by_dso(self, dso_filter);
+                        hists__browser_title(self, msg, sizeof(msg), ev_name,
+                                             dso_filter, thread_filter);
+                        hist_browser__reset(browser);
+                } else if (choice == zoom_thread) {
+zoom_thread:
+                        if (thread_filter) {
+                                pstack__remove(fstack, &thread_filter);
+zoom_out_thread:
+                                ui_helpline__pop();
+                                thread_filter = NULL;
+                        } else {
+                                ui_helpline__fpush("To zoom out press <- or -> + \"Zoom out of %s(%d) thread\"",
+                                                   thread->comm_set ? thread->comm : "",
+                                                   thread->pid);
+                                thread_filter = thread;
+                                pstack__push(fstack, &thread_filter);
+                        }
+                        hists__filter_by_thread(self, thread_filter);
+                        hists__browser_title(self, msg, sizeof(msg), ev_name,
+                                             dso_filter, thread_filter);
+                        hist_browser__reset(browser);
+                }
+        }
+out_free_stack:
+        pstack__delete(fstack);
+out:
+        hist_browser__delete(browser);
+        return key;
+}
+
+int hists__tui_browse_tree(struct rb_root *self, const char *help)
+{
+        struct rb_node *first = rb_first(self), *nd = first, *next;
+        int key = 0;
+
+        while (nd) {
+                struct hists *hists = rb_entry(nd, struct hists, rb_node);
+                const char *ev_name = __event_name(hists->type, hists->config);
+
+                key = hists__browse(hists, help, ev_name);
+                switch (key) {
+                case NEWT_KEY_TAB:
+                        next = rb_next(nd);
+                        if (next)
+                                nd = next;
+                        break;
+                case NEWT_KEY_UNTAB:
+                        if (nd == first)
+                                continue;
+                        nd = rb_prev(nd);
+                default:
+                        return key;
+                }
+        }
+
+        return key;
+}
diff --git a/tools/perf/util/ui/browsers/map.c b/tools/perf/util/ui/browsers/map.c
new file mode 100644
index 00000000000..e35437dfa5b
--- /dev/null
+++ b/tools/perf/util/ui/browsers/map.c
@@ -0,0 +1,155 @@
+#include "../libslang.h"
+#include <elf.h>
+#include <sys/ttydefaults.h>
+#include <ctype.h>
+#include <string.h>
+#include <linux/bitops.h>
+#include "../../debug.h"
+#include "../../symbol.h"
+#include "../browser.h"
+#include "../helpline.h"
+#include "map.h"
+
+static int ui_entry__read(const char *title, char *bf, size_t size, int width)
+{
+        struct newtExitStruct es;
+        newtComponent form, entry;
+        const char *result;
+        int err = -1;
+
+        newtCenteredWindow(width, 1, title);
+        form = newtForm(NULL, NULL, 0);
+        if (form == NULL)
+                return -1;
+
+        entry = newtEntry(0, 0, "0x", width, &result, NEWT_FLAG_SCROLL);
+        if (entry == NULL)
+                goto out_free_form;
+
+        newtFormAddComponent(form, entry);
+        newtFormAddHotKey(form, NEWT_KEY_ENTER);
+        newtFormAddHotKey(form, NEWT_KEY_ESCAPE);
+        newtFormAddHotKey(form, NEWT_KEY_LEFT);
+        newtFormAddHotKey(form, CTRL('c'));
+        newtFormRun(form, &es);
+
+        if (result != NULL) {
+                strncpy(bf, result, size);
+                err = 0;
+        }
+out_free_form:
+        newtPopWindow();
+        newtFormDestroy(form);
+        return 0;
+}
+
+struct map_browser {
+        struct ui_browser b;
+        struct map        *map;
+        u8                addrlen;
+};
+
+static void map_browser__write(struct ui_browser *self, void *nd, int row)
+{
+        struct symbol *sym = rb_entry(nd, struct symbol, rb_node);
+        struct map_browser *mb = container_of(self, struct map_browser, b);
+        bool current_entry = ui_browser__is_current_entry(self, row);
+        int width;
+
+        ui_browser__set_percent_color(self, 0, current_entry);
+        slsmg_printf("%*llx %*llx %c ",
+                     mb->addrlen, sym->start, mb->addrlen, sym->end,
+                     sym->binding == STB_GLOBAL ? 'g' :
+                     sym->binding == STB_LOCAL  ? 'l' : 'w');
+        width = self->width - ((mb->addrlen * 2) + 4);
+        if (width > 0)
+                slsmg_write_nstring(sym->name, width);
+}
+
+/* FIXME uber-kludgy, see comment on cmd_report... */
+static u32 *symbol__browser_index(struct symbol *self)
+{
+        return ((void *)self) - sizeof(struct rb_node) - sizeof(u32);
+}
+
+static int map_browser__search(struct map_browser *self)
+{
+        char target[512];
+        struct symbol *sym;
+        int err = ui_entry__read("Search by name/addr", target, sizeof(target), 40);
+
+        if (err)
+                return err;
+
+        if (target[0] == '0' && tolower(target[1]) == 'x') {
+                u64 addr = strtoull(target, NULL, 16);
+                sym = map__find_symbol(self->map, addr, NULL);
+        } else
+                sym = map__find_symbol_by_name(self->map, target, NULL);
+
+        if (sym != NULL) {
+                u32 *idx = symbol__browser_index(sym);
+
+                self->b.top = &sym->rb_node;
+                self->b.index = self->b.top_idx = *idx;
+        } else
+                ui_helpline__fpush("%s not found!", target);
+
+        return 0;
+}
+
+static int map_browser__run(struct map_browser *self)
+{
+        int key;
+
+        if (ui_browser__show(&self->b, self->map->dso->long_name,
+                             "Press <- or ESC to exit, %s / to search",
+                             verbose ? "" : "restart with -v to use") < 0)
+                return -1;
+
+        if (verbose)
+                ui_browser__add_exit_key(&self->b, '/');
+
+        while (1) {
+                key = ui_browser__run(&self->b);
+
+                if (verbose && key == '/')
+                        map_browser__search(self);
+                else
+                        break;
+        }
+
+        ui_browser__hide(&self->b);
+        return key;
+}
+
+int map__browse(struct map *self)
+{
+        struct map_browser mb = {
+                .b = {
+                        .entries = &self->dso->symbols[self->type],
+                        .refresh = ui_browser__rb_tree_refresh,
+                        .seek    = ui_browser__rb_tree_seek,
+                        .write   = map_browser__write,
+                },
+                .map = self,
+        };
+        struct rb_node *nd;
+        char tmp[BITS_PER_LONG / 4];
+        u64 maxaddr = 0;
+
+        for (nd = rb_first(mb.b.entries); nd; nd = rb_next(nd)) {
+                struct symbol *pos = rb_entry(nd, struct symbol, rb_node);
+
+                if (maxaddr < pos->end)
+                        maxaddr = pos->end;
+                if (verbose) {
+                        u32 *idx = symbol__browser_index(pos);
+                        *idx = mb.b.nr_entries;
+                }
+                ++mb.b.nr_entries;
+        }
+
+        mb.addrlen = snprintf(tmp, sizeof(tmp), "%llx", maxaddr);
+        return map_browser__run(&mb);
+}
diff --git a/tools/perf/util/ui/browsers/map.h b/tools/perf/util/ui/browsers/map.h
new file mode 100644
index 00000000000..df8581a43e1
--- /dev/null
+++ b/tools/perf/util/ui/browsers/map.h
@@ -0,0 +1,6 @@
+#ifndef _PERF_UI_MAP_BROWSER_H_
+#define _PERF_UI_MAP_BROWSER_H_ 1
+struct map;
+
+int map__browse(struct map *self);
+#endif /* _PERF_UI_MAP_BROWSER_H_ */
diff --git a/tools/perf/util/ui/helpline.c b/tools/perf/util/ui/helpline.c
new file mode 100644
index 00000000000..8d79daa4458
--- /dev/null
+++ b/tools/perf/util/ui/helpline.c
@@ -0,0 +1,69 @@
+#define _GNU_SOURCE
+#include <stdio.h>
+#include <stdlib.h>
+#include <newt.h>
+
+#include "../debug.h"
+#include "helpline.h"
+
+void ui_helpline__pop(void)
+{
+        newtPopHelpLine();
+}
+
+void ui_helpline__push(const char *msg)
+{
+        newtPushHelpLine(msg);
+}
+
+void ui_helpline__vpush(const char *fmt, va_list ap)
+{
+        char *s;
+
+        if (vasprintf(&s, fmt, ap) < 0)
+                vfprintf(stderr, fmt, ap);
+        else {
+                ui_helpline__push(s);
+                free(s);
+        }
+}
+
+void ui_helpline__fpush(const char *fmt, ...)
+{
+        va_list ap;
+
+        va_start(ap, fmt);
+        ui_helpline__vpush(fmt, ap);
+        va_end(ap);
+}
+
+void ui_helpline__puts(const char *msg)
+{
+        ui_helpline__pop();
+        ui_helpline__push(msg);
+}
+
+void ui_helpline__init(void)
+{
+        ui_helpline__puts(" ");
+}
+
+char ui_helpline__last_msg[1024];
+
+int ui_helpline__show_help(const char *format, va_list ap)
+{
+        int ret;
+        static int backlog;
+
+        ret = vsnprintf(ui_helpline__last_msg + backlog,
+                        sizeof(ui_helpline__last_msg) - backlog, format, ap);
+        backlog += ret;
+
+        if (ui_helpline__last_msg[backlog - 1] == '\n') {
+                ui_helpline__puts(ui_helpline__last_msg);
+                newtRefresh();
+                backlog = 0;
+        }
+
+        return ret;
+}
diff --git a/tools/perf/util/ui/helpline.h b/tools/perf/util/ui/helpline.h
new file mode 100644
index 00000000000..ab6028d0c40
--- /dev/null
+++ b/tools/perf/util/ui/helpline.h
@@ -0,0 +1,11 @@
+#ifndef _PERF_UI_HELPLINE_H_
+#define _PERF_UI_HELPLINE_H_ 1
+
+void ui_helpline__init(void);
+void ui_helpline__pop(void);
+void ui_helpline__push(const char *msg);
+void ui_helpline__vpush(const char *fmt, va_list ap);
+void ui_helpline__fpush(const char *fmt, ...);
+void ui_helpline__puts(const char *msg);
+
+#endif /* _PERF_UI_HELPLINE_H_ */
diff --git a/tools/perf/util/ui/libslang.h b/tools/perf/util/ui/libslang.h
new file mode 100644
index 00000000000..5623da8e808
--- /dev/null
+++ b/tools/perf/util/ui/libslang.h
@@ -0,0 +1,27 @@
+#ifndef _PERF_UI_SLANG_H_
+#define _PERF_UI_SLANG_H_ 1
+/*
+ * slang versions <= 2.0.6 have a "#if HAVE_LONG_LONG" that breaks
+ * the build if it isn't defined. Use the equivalent one that glibc
+ * has on features.h.
+ */
+#include <features.h>
+#ifndef HAVE_LONG_LONG
+#define HAVE_LONG_LONG __GLIBC_HAVE_LONG_LONG
+#endif
+#include <slang.h>
+
+#if SLANG_VERSION < 20104
+#define slsmg_printf(msg, args...) \
+        SLsmg_printf((char *)msg, ##args)
+#define slsmg_write_nstring(msg, len) \
+        SLsmg_write_nstring((char *)msg, len)
+#define sltt_set_color(obj, name, fg, bg) \
+        SLtt_set_color(obj,(char *)name, (char *)fg, (char *)bg)
+#else
+#define slsmg_printf SLsmg_printf
+#define slsmg_write_nstring SLsmg_write_nstring
+#define sltt_set_color SLtt_set_color
+#endif
+
+#endif /* _PERF_UI_SLANG_H_ */
diff --git a/tools/perf/util/ui/progress.c b/tools/perf/util/ui/progress.c
new file mode 100644
index 00000000000..d7fc399d36b
--- /dev/null
+++ b/tools/perf/util/ui/progress.c
@@ -0,0 +1,60 @@
+#include <stdlib.h>
+#include <newt.h>
+#include "../cache.h"
+#include "progress.h"
+
+struct ui_progress {
+        newtComponent form, scale;
+};
+
+struct ui_progress *ui_progress__new(const char *title, u64 total)
+{
+        struct ui_progress *self = malloc(sizeof(*self));
+
+        if (self != NULL) {
+                int cols;
+
+                if (use_browser <= 0)
+                        return self;
+                newtGetScreenSize(&cols, NULL);
+                cols -= 4;
+                newtCenteredWindow(cols, 1, title);
+                self->form  = newtForm(NULL, NULL, 0);
+                if (self->form == NULL)
+                        goto out_free_self;
+                self->scale = newtScale(0, 0, cols, total);
+                if (self->scale == NULL)
+                        goto out_free_form;
+                newtFormAddComponent(self->form, self->scale);
+                newtRefresh();
+        }
+
+        return self;
+
+out_free_form:
+        newtFormDestroy(self->form);
+out_free_self:
+        free(self);
+        return NULL;
+}
+
+void ui_progress__update(struct ui_progress *self, u64 curr)
+{
+        /*
+         * FIXME: We should have a per UI backend way of showing progress,
+         * stdio will just show a percentage as NN%, etc.
+         */
+        if (use_browser <= 0)
+                return;
+        newtScaleSet(self->scale, curr);
+        newtRefresh();
+}
+
+void ui_progress__delete(struct ui_progress *self)
+{
+        if (use_browser > 0) {
+                newtFormDestroy(self->form);
+                newtPopWindow();
+        }
+        free(self);
+}
diff --git a/tools/perf/util/ui/progress.h b/tools/perf/util/ui/progress.h
new file mode 100644
index 00000000000..a3820a0beb5
--- /dev/null
+++ b/tools/perf/util/ui/progress.h
@@ -0,0 +1,11 @@
+#ifndef _PERF_UI_PROGRESS_H_
+#define _PERF_UI_PROGRESS_H_ 1
+
+struct ui_progress;
+
+struct ui_progress *ui_progress__new(const char *title, u64 total);
+void ui_progress__delete(struct ui_progress *self);
+
+void ui_progress__update(struct ui_progress *self, u64 curr);
+
+#endif
diff --git a/tools/perf/util/ui/setup.c b/tools/perf/util/ui/setup.c
new file mode 100644
index 00000000000..662085032eb
--- /dev/null
+++ b/tools/perf/util/ui/setup.c
@@ -0,0 +1,42 @@
+#include <newt.h>
+#include <signal.h>
+#include <stdbool.h>
+
+#include "../cache.h"
+#include "../debug.h"
+#include "browser.h"
+#include "helpline.h"
+
+static void newt_suspend(void *d __used)
+{
+        newtSuspend();
+        raise(SIGTSTP);
+        newtResume();
+}
+
+void setup_browser(void)
+{
+        if (!isatty(1) || !use_browser || dump_trace) {
+                use_browser = 0;
+                setup_pager();
+                return;
+        }
+
+        use_browser = 1;
+        newtInit();
+        newtCls();
+        newtSetSuspendCallback(newt_suspend, NULL);
+        ui_helpline__init();
+        ui_browser__init();
+}
+
+void exit_browser(bool wait_for_ok)
+{
+        if (use_browser > 0) {
+                if (wait_for_ok) {
+                        char title[] = "Fatal Error", ok[] = "Ok";
+                        newtWinMessage(title, ok, ui_helpline__last_msg);
+                }
+                newtFinished();
+        }
+}
diff --git a/tools/perf/util/ui/util.c b/tools/perf/util/ui/util.c
new file mode 100644
index 00000000000..7b5a8926624
--- /dev/null
+++ b/tools/perf/util/ui/util.c
@@ -0,0 +1,127 @@
+#include <newt.h>
+#include <signal.h>
+#include <stdio.h>
+#include <stdbool.h>
+#include <string.h>
+#include <sys/ttydefaults.h>
+
+#include "../cache.h"
+#include "../debug.h"
+#include "browser.h"
+#include "helpline.h"
+#include "util.h"
+
+static void newt_form__set_exit_keys(newtComponent self)
+{
+        newtFormAddHotKey(self, NEWT_KEY_LEFT);
+        newtFormAddHotKey(self, NEWT_KEY_ESCAPE);
+        newtFormAddHotKey(self, 'Q');
+        newtFormAddHotKey(self, 'q');
+        newtFormAddHotKey(self, CTRL('c'));
+}
+
+static newtComponent newt_form__new(void)
+{
+        newtComponent self = newtForm(NULL, NULL, 0);
+        if (self)
+                newt_form__set_exit_keys(self);
+        return self;
+}
+
+int ui__popup_menu(int argc, char * const argv[])
+{
+        struct newtExitStruct es;
+        int i, rc = -1, max_len = 5;
+        newtComponent listbox, form = newt_form__new();
+
+        if (form == NULL)
+                return -1;
+
+        listbox = newtListbox(0, 0, argc, NEWT_FLAG_RETURNEXIT);
+        if (listbox == NULL)
+                goto out_destroy_form;
+
+        newtFormAddComponent(form, listbox);
+
+        for (i = 0; i < argc; ++i) {
+                int len = strlen(argv[i]);
+                if (len > max_len)
+                        max_len = len;
+                if (newtListboxAddEntry(listbox, argv[i], (void *)(long)i))
+                        goto out_destroy_form;
+        }
+
+        newtCenteredWindow(max_len, argc, NULL);
+        newtFormRun(form, &es);
+        rc = newtListboxGetCurrent(listbox) - NULL;
+        if (es.reason == NEWT_EXIT_HOTKEY)
+                rc = -1;
+        newtPopWindow();
+out_destroy_form:
+        newtFormDestroy(form);
+        return rc;
+}
+
+int ui__help_window(const char *text)
+{
+        struct newtExitStruct es;
+        newtComponent tb, form = newt_form__new();
+        int rc = -1;
+        int max_len = 0, nr_lines = 0;
+        const char *t;
+
+        if (form == NULL)
+                return -1;
+
+        t = text;
+        while (1) {
+                const char *sep = strchr(t, '\n');
+                int len;
+
+                if (sep == NULL)
+                        sep = strchr(t, '\0');
+                len = sep - t;
+                if (max_len < len)
+                        max_len = len;
+                ++nr_lines;
+                if (*sep == '\0')
+                        break;
+                t = sep + 1;
+        }
+
+        tb = newtTextbox(0, 0, max_len, nr_lines, 0);
+        if (tb == NULL)
+                goto out_destroy_form;
+
+        newtTextboxSetText(tb, text);
+        newtFormAddComponent(form, tb);
+        newtCenteredWindow(max_len, nr_lines, NULL);
+        newtFormRun(form, &es);
+        newtPopWindow();
+        rc = 0;
+out_destroy_form:
+        newtFormDestroy(form);
+        return rc;
+}
+
+static const char yes[] = "Yes", no[] = "No",
+                  warning_str[] = "Warning!", ok[] = "Ok";
+
+bool ui__dialog_yesno(const char *msg)
+{
+        /* newtWinChoice should really be accepting const char pointers... */
+        return newtWinChoice(NULL, (char *)yes, (char *)no, (char *)msg) == 1;
+}
+
+void ui__warning(const char *format, ...)
+{
+        va_list args;
+
+        va_start(args, format);
+        if (use_browser > 0)
+                newtWinMessagev((char *)warning_str, (char *)ok,
+                                (char *)format, args);
+        else
+                vfprintf(stderr, format, args);
+        va_end(args);
+}
diff --git a/tools/perf/util/ui/util.h b/tools/perf/util/ui/util.h
new file mode 100644
index 00000000000..afcbc1d9953
--- /dev/null
+++ b/tools/perf/util/ui/util.h
@@ -0,0 +1,10 @@
+#ifndef _PERF_UI_UTIL_H_
+#define _PERF_UI_UTIL_H_ 1
+
+#include <stdbool.h>
+
+int ui__popup_menu(int argc, char * const argv[]);
+int ui__help_window(const char *text);
+bool ui__dialog_yesno(const char *msg);
+
+#endif /* _PERF_UI_UTIL_H_ */
diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c
new file mode 100644
index 00000000000..5b3ea49aa63
--- /dev/null
+++ b/tools/perf/util/util.c
@@ -0,0 +1,133 @@
+#include "util.h"
+#include <sys/mman.h>
+
+int mkdir_p(char *path, mode_t mode)
+{
+        struct stat st;
+        int err;
+        char *d = path;
+
+        if (*d != '/')
+                return -1;
+
+        if (stat(path, &st) == 0)
+                return 0;
+
+        while (*++d == '/');
+
+        while ((d = strchr(d, '/'))) {
+                *d = '\0';
+                err = stat(path, &st) && mkdir(path, mode);
+                *d++ = '/';
+                if (err)
+                        return -1;
+                while (*d == '/')
+                        ++d;
+        }
+        return (stat(path, &st) && mkdir(path, mode)) ? -1 : 0;
+}
+
+static int slow_copyfile(const char *from, const char *to)
+{
+        int err = 0;
+        char *line = NULL;
+        size_t n;
+        FILE *from_fp = fopen(from, "r"), *to_fp;
+
+        if (from_fp == NULL)
+                goto out;
+
+        to_fp = fopen(to, "w");
+        if (to_fp == NULL)
+                goto out_fclose_from;
+
+        while (getline(&line, &n, from_fp) > 0)
+                if (fputs(line, to_fp) == EOF)
+                        goto out_fclose_to;
+        err = 0;
+out_fclose_to:
+        fclose(to_fp);
+        free(line);
+out_fclose_from:
+        fclose(from_fp);
+out:
+        return err;
+}
+
+int copyfile(const char *from, const char *to)
+{
+        int fromfd, tofd;
+        struct stat st;
+        void *addr;
+        int err = -1;
+
+        if (stat(from, &st))
+                goto out;
+
+        if (st.st_size == 0) /* /proc? do it slowly... */
+                return slow_copyfile(from, to);
+
+        fromfd = open(from, O_RDONLY);
+        if (fromfd < 0)
+                goto out;
+
+        tofd = creat(to, 0755);
+        if (tofd < 0)
+                goto out_close_from;
+
+        addr = mmap(NULL, st.st_size, PROT_READ, MAP_PRIVATE, fromfd, 0);
+        if (addr == MAP_FAILED)
+                goto out_close_to;
+
+        if (write(tofd, addr, st.st_size) == st.st_size)
+                err = 0;
+
+        munmap(addr, st.st_size);
+out_close_to:
+        close(tofd);
+        if (err)
+                unlink(to);
+out_close_from:
+        close(fromfd);
+out:
+        return err;
+}
+
+unsigned long convert_unit(unsigned long value, char *unit)
+{
+        *unit = ' ';
+
+        if (value > 1000) {
+                value /= 1000;
+                *unit = 'K';
+        }
+
+        if (value > 1000) {
+                value /= 1000;
+                *unit = 'M';
+        }
+
+        if (value > 1000) {
+                value /= 1000;
+                *unit = 'G';
+        }
+
+        return value;
+}
+
+int readn(int fd, void *buf, size_t n)
+{
+        void *buf_start = buf;
+
+        while (n) {
+                int ret = read(fd, buf, n);
+
+                if (ret <= 0)
+                        return ret;
+
+                n -= ret;
+                buf += ret;
+        }
+
+        return buf - buf_start;
+}
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index b8cfed776d8..e833f26f3bf 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -39,18 +39,17 @@
 /* Approximation of the length of the decimal representation of this type. */
 #define decimal_length(x)       ((int)(sizeof(x) * 2.56 + 0.5) + 1)
 
-#if !defined(__APPLE__) && !defined(__FreeBSD__)  && !defined(__USLC__) && !defined(_M_UNIX)
-#define _XOPEN_SOURCE 600 /* glibc2 and AIX 5.3L need 500, OpenBSD needs 600 for S_ISLNK() */
-#define _XOPEN_SOURCE_EXTENDED 1 /* AIX 5.3L needs this */
-#endif
 #define _ALL_SOURCE 1
 #define _GNU_SOURCE 1
 #define _BSD_SOURCE 1
+#define HAS_BOOL
 
 #include <unistd.h>
 #include <stdio.h>
 #include <sys/stat.h>
+#include <sys/statfs.h>
 #include <fcntl.h>
+#include <stdbool.h>
 #include <stddef.h>
 #include <stdlib.h>
 #include <stdarg.h>
@@ -67,7 +66,6 @@
 #include <assert.h>
 #include <regex.h>
 #include <utime.h>
-#ifndef __MINGW32__
 #include <sys/wait.h>
 #include <sys/poll.h>
 #include <sys/socket.h>
@@ -81,25 +79,18 @@
 #include <netdb.h>
 #include <pwd.h>
 #include <inttypes.h>
-#if defined(__CYGWIN__)
-#undef _XOPEN_SOURCE
-#include <grp.h>
-#define _XOPEN_SOURCE 600
-#include "compat/cygwin.h"
-#else
-#undef _ALL_SOURCE /* AIX 5.3L defines a struct list with _ALL_SOURCE. */
-#include <grp.h>
-#define _ALL_SOURCE 1
-#endif
-#else   /* __MINGW32__ */
-/* pull in Windows compatibility stuff */
-#include "compat/mingw.h"
-#endif  /* __MINGW32__ */
+#include "../../../include/linux/magic.h"
+#include "types.h"
+#include <sys/ttydefaults.h>
 
 #ifndef NO_ICONV
 #include <iconv.h>
 #endif
 
+extern const char *graph_line;
+extern const char *graph_dotted_line;
+extern char buildid_dir[];
+
 /* On most systems <limits.h> would have given us this, but
  * not on some systems (e.g. GNU/Hurd).
  */
@@ -150,10 +141,20 @@ extern void die(const char *err, ...) NORETURN __attribute__((format (printf, 1,
 extern int error(const char *err, ...) __attribute__((format (printf, 1, 2)));
 extern void warning(const char *err, ...) __attribute__((format (printf, 1, 2)));
 
+#include "../../../include/linux/stringify.h"
+
+#define DIE_IF(cnd)     \
+        do { if (cnd)   \
+                die(" at (" __FILE__ ":" __stringify(__LINE__) "): "    \
+                    __stringify(cnd) "\n");                             \
+        } while (0)
+
+
 extern void set_die_routine(void (*routine)(const char *err, va_list params) NORETURN);
 
 extern int prefixcmp(const char *str, const char *prefix);
-extern time_t tm_to_time_t(const struct tm *tm);
+extern void set_buildid_dir(void);
+extern void disable_buildid_cache(void);
 
 static inline const char *skip_prefix(const char *str, const char *prefix)
 {
@@ -161,119 +162,6 @@ static inline const char *skip_prefix(const char *str, const char *prefix)
         return strncmp(str, prefix, len) ? NULL : str + len;
 }
 
-#if defined(NO_MMAP) || defined(USE_WIN32_MMAP)
-
-#ifndef PROT_READ
-#define PROT_READ 1
-#define PROT_WRITE 2
-#define MAP_PRIVATE 1
-#define MAP_FAILED ((void*)-1)
-#endif
-
-#define mmap git_mmap
-#define munmap git_munmap
-extern void *git_mmap(void *start, size_t length, int prot, int flags, int fd, off_t offset);
-extern int git_munmap(void *start, size_t length);
-
-#else /* NO_MMAP || USE_WIN32_MMAP */
-
-#include <sys/mman.h>
-
-#endif /* NO_MMAP || USE_WIN32_MMAP */
-
-#ifdef NO_MMAP
-
-/* This value must be multiple of (pagesize * 2) */
-#define DEFAULT_PACKED_GIT_WINDOW_SIZE (1 * 1024 * 1024)
-
-#else /* NO_MMAP */
-
-/* This value must be multiple of (pagesize * 2) */
-#define DEFAULT_PACKED_GIT_WINDOW_SIZE \
-        (sizeof(void*) >= 8 \
-                ?  1 * 1024 * 1024 * 1024 \
-                : 32 * 1024 * 1024)
-
-#endif /* NO_MMAP */
-
-#ifdef NO_ST_BLOCKS_IN_STRUCT_STAT
-#define on_disk_bytes(st) ((st).st_size)
-#else
-#define on_disk_bytes(st) ((st).st_blocks * 512)
-#endif
-
-#define DEFAULT_PACKED_GIT_LIMIT \
-        ((1024L * 1024L) * (sizeof(void*) >= 8 ? 8192 : 256))
-
-#ifdef NO_PREAD
-#define pread git_pread
-extern ssize_t git_pread(int fd, void *buf, size_t count, off_t offset);
-#endif
-/*
- * Forward decl that will remind us if its twin in cache.h changes.
- * This function is used in compat/pread.c.  But we can't include
- * cache.h there.
- */
-extern ssize_t read_in_full(int fd, void *buf, size_t count);
-
-#ifdef NO_SETENV
-#define setenv gitsetenv
-extern int gitsetenv(const char *, const char *, int);
-#endif
-
-#ifdef NO_MKDTEMP
-#define mkdtemp gitmkdtemp
-extern char *gitmkdtemp(char *);
-#endif
-
-#ifdef NO_UNSETENV
-#define unsetenv gitunsetenv
-extern void gitunsetenv(const char *);
-#endif
-
-#ifdef NO_STRCASESTR
-#define strcasestr gitstrcasestr
-extern char *gitstrcasestr(const char *haystack, const char *needle);
-#endif
-
-#ifdef NO_STRLCPY
-#define strlcpy gitstrlcpy
-extern size_t gitstrlcpy(char *, const char *, size_t);
-#endif
-
-#ifdef NO_STRTOUMAX
-#define strtoumax gitstrtoumax
-extern uintmax_t gitstrtoumax(const char *, char **, int);
-#endif
-
-#ifdef NO_HSTRERROR
-#define hstrerror githstrerror
-extern const char *githstrerror(int herror);
-#endif
-
-#ifdef NO_MEMMEM
-#define memmem gitmemmem
-void *gitmemmem(const void *haystack, size_t haystacklen,
-                const void *needle, size_t needlelen);
-#endif
-
-#ifdef FREAD_READS_DIRECTORIES
-#ifdef fopen
-#undef fopen
-#endif
-#define fopen(a,b) git_fopen(a,b)
-extern FILE *git_fopen(const char*, const char*);
-#endif
-
-#ifdef SNPRINTF_RETURNS_BOGUS
-#define snprintf git_snprintf
-extern int git_snprintf(char *str, size_t maxsize,
-                        const char *format, ...);
-#define vsnprintf git_vsnprintf
-extern int git_vsnprintf(char *str, size_t maxsize,
-                         const char *format, va_list ap);
-#endif
-
 #ifdef __GLIBC_PREREQ
 #if __GLIBC_PREREQ(2, 1)
 #define HAVE_STRCHRNUL
@@ -294,27 +182,19 @@ static inline char *gitstrchrnul(const char *s, int c)
  * Wrappers:
  */
 extern char *xstrdup(const char *str);
-extern void *xmalloc(size_t size);
-extern void *xmemdupz(const void *data, size_t len);
-extern char *xstrndup(const char *str, size_t len);
-extern void *xrealloc(void *ptr, size_t size);
-extern void *xcalloc(size_t nmemb, size_t size);
-extern void *xmmap(void *start, size_t length, int prot, int flags, int fd, off_t offset);
-extern ssize_t xread(int fd, void *buf, size_t len);
-extern ssize_t xwrite(int fd, const void *buf, size_t len);
-extern int xdup(int fd);
-extern FILE *xfdopen(int fd, const char *mode);
-extern int xmkstemp(char *template);
-
-static inline size_t xsize_t(off_t len)
+extern void *xrealloc(void *ptr, size_t size) __attribute__((weak));
+
+
+static inline void *zalloc(size_t size)
 {
-        return (size_t)len;
+        return calloc(1, size);
 }
 
 static inline int has_extension(const char *filename, const char *ext)
 {
         size_t len = strlen(filename);
         size_t extlen = strlen(ext);
+
         return len > extlen && !memcmp(filename + len - extlen, ext, extlen);
 }
 
@@ -322,10 +202,13 @@ static inline int has_extension(const char *filename, const char *ext)
 #undef isascii
 #undef isspace
 #undef isdigit
+#undef isxdigit
 #undef isalpha
+#undef isprint
 #undef isalnum
 #undef tolower
 #undef toupper
+
 extern unsigned char sane_ctype[256];
 #define GIT_SPACE               0x01
 #define GIT_DIGIT               0x02
@@ -338,11 +221,11 @@ extern unsigned char sane_ctype[256];
 #define isascii(x) (((x) & ~0x7f) == 0)
 #define isspace(x) sane_istest(x,GIT_SPACE)
 #define isdigit(x) sane_istest(x,GIT_DIGIT)
+#define isxdigit(x)     \
+        (sane_istest(toupper(x), GIT_ALPHA | GIT_DIGIT) && toupper(x) < 'G')
 #define isalpha(x) sane_istest(x,GIT_ALPHA)
 #define isalnum(x) sane_istest(x,GIT_ALPHA | GIT_DIGIT)
 #define isprint(x) sane_istest(x,GIT_PRINT)
-#define is_glob_special(x) sane_istest(x,GIT_GLOB_SPECIAL)
-#define is_regex_special(x) sane_istest(x,GIT_GLOB_SPECIAL | GIT_REGEX_SPECIAL)
 #define tolower(x) sane_case((unsigned char)(x), 0x20)
 #define toupper(x) sane_case((unsigned char)(x), 0)
 
@@ -353,38 +236,6 @@ static inline int sane_case(int x, int high)
         return x;
 }
 
-static inline int strtoul_ui(char const *s, int base, unsigned int *result)
-{
-        unsigned long ul;
-        char *p;
-
-        errno = 0;
-        ul = strtoul(s, &p, base);
-        if (errno || *p || p == s || (unsigned int) ul != ul)
-                return -1;
-        *result = ul;
-        return 0;
-}
-
-static inline int strtol_i(char const *s, int base, int *result)
-{
-        long ul;
-        char *p;
-
-        errno = 0;
-        ul = strtol(s, &p, base);
-        if (errno || *p || p == s || (int) ul != ul)
-                return -1;
-        *result = ul;
-        return 0;
-}
-
-#ifdef INTERNAL_QSORT
-void git_qsort(void *base, size_t nmemb, size_t size,
-               int(*compar)(const void *, const void *));
-#define qsort git_qsort
-#endif
-
 #ifndef DIR_HAS_BSD_GROUP_SEMANTICS
 # define FORCE_DIR_SET_GID S_ISGID
 #else
@@ -405,4 +256,18 @@ void git_qsort(void *base, size_t nmemb, size_t size,
 #endif
 #endif
 
+int mkdir_p(char *path, mode_t mode);
+int copyfile(const char *from, const char *to);
+
+s64 perf_atoll(const char *str);
+char **argv_split(const char *str, int *argcp);
+void argv_free(char **argv);
+bool strglobmatch(const char *str, const char *pat);
+bool strlazymatch(const char *str, const char *pat);
+unsigned long convert_unit(unsigned long value, char *unit);
+int readn(int fd, void *buf, size_t size);
+
+#define _STR(x) #x
+#define STR(x) _STR(x)
+
 #endif
diff --git a/tools/perf/util/values.c b/tools/perf/util/values.c
new file mode 100644
index 00000000000..cfa55d686e3
--- /dev/null
+++ b/tools/perf/util/values.c
@@ -0,0 +1,231 @@
+#include <stdlib.h>
+
+#include "util.h"
+#include "values.h"
+
+void perf_read_values_init(struct perf_read_values *values)
+{
+        values->threads_max = 16;
+        values->pid = malloc(values->threads_max * sizeof(*values->pid));
+        values->tid = malloc(values->threads_max * sizeof(*values->tid));
+        values->value = malloc(values->threads_max * sizeof(*values->value));
+        if (!values->pid || !values->tid || !values->value)
+                die("failed to allocate read_values threads arrays");
+        values->threads = 0;
+
+        values->counters_max = 16;
+        values->counterrawid = malloc(values->counters_max
+                                      * sizeof(*values->counterrawid));
+        values->countername = malloc(values->counters_max
+                                     * sizeof(*values->countername));
+        if (!values->counterrawid || !values->countername)
+                die("failed to allocate read_values counters arrays");
+        values->counters = 0;
+}
+
+void perf_read_values_destroy(struct perf_read_values *values)
+{
+        int i;
+
+        if (!values->threads_max || !values->counters_max)
+                return;
+
+        for (i = 0; i < values->threads; i++)
+                free(values->value[i]);
+        free(values->pid);
+        free(values->tid);
+        free(values->counterrawid);
+        for (i = 0; i < values->counters; i++)
+                free(values->countername[i]);
+        free(values->countername);
+}
+
+static void perf_read_values__enlarge_threads(struct perf_read_values *values)
+{
+        values->threads_max *= 2;
+        values->pid = realloc(values->pid,
+                              values->threads_max * sizeof(*values->pid));
+        values->tid = realloc(values->tid,
+                              values->threads_max * sizeof(*values->tid));
+        values->value = realloc(values->value,
+                                values->threads_max * sizeof(*values->value));
+        if (!values->pid || !values->tid || !values->value)
+                die("failed to enlarge read_values threads arrays");
+}
+
+static int perf_read_values__findnew_thread(struct perf_read_values *values,
+                                            u32 pid, u32 tid)
+{
+        int i;
+
+        for (i = 0; i < values->threads; i++)
+                if (values->pid[i] == pid && values->tid[i] == tid)
+                        return i;
+
+        if (values->threads == values->threads_max)
+                perf_read_values__enlarge_threads(values);
+
+        i = values->threads++;
+        values->pid[i] = pid;
+        values->tid[i] = tid;
+        values->value[i] = malloc(values->counters_max * sizeof(**values->value));
+        if (!values->value[i])
+                die("failed to allocate read_values counters array");
+
+        return i;
+}
+
+static void perf_read_values__enlarge_counters(struct perf_read_values *values)
+{
+        int i;
+
+        values->counters_max *= 2;
+        values->counterrawid = realloc(values->counterrawid,
+                                       values->counters_max * sizeof(*values->counterrawid));
+        values->countername = realloc(values->countername,
+                                      values->counters_max * sizeof(*values->countername));
+        if (!values->counterrawid || !values->countername)
+                die("failed to enlarge read_values counters arrays");
+
+        for (i = 0; i < values->threads; i++) {
+                values->value[i] = realloc(values->value[i],
+                                           values->counters_max * sizeof(**values->value));
+                if (!values->value[i])
+                        die("failed to enlarge read_values counters arrays");
+        }
+}
+
+static int perf_read_values__findnew_counter(struct perf_read_values *values,
+                                             u64 rawid, const char *name)
+{
+        int i;
+
+        for (i = 0; i < values->counters; i++)
+                if (values->counterrawid[i] == rawid)
+                        return i;
+
+        if (values->counters == values->counters_max)
+                perf_read_values__enlarge_counters(values);
+
+        i = values->counters++;
+        values->counterrawid[i] = rawid;
+        values->countername[i] = strdup(name);
+
+        return i;
+}
+
+void perf_read_values_add_value(struct perf_read_values *values,
+                                u32 pid, u32 tid,
+                                u64 rawid, const char *name, u64 value)
+{
+        int tindex, cindex;
+
+        tindex = perf_read_values__findnew_thread(values, pid, tid);
+        cindex = perf_read_values__findnew_counter(values, rawid, name);
+
+        values->value[tindex][cindex] = value;
+}
+
+static void perf_read_values__display_pretty(FILE *fp,
+                                             struct perf_read_values *values)
+{
+        int i, j;
+        int pidwidth, tidwidth;
+        int *counterwidth;
+
+        counterwidth = malloc(values->counters * sizeof(*counterwidth));
+        if (!counterwidth)
+                die("failed to allocate counterwidth array");
+        tidwidth = 3;
+        pidwidth = 3;
+        for (j = 0; j < values->counters; j++)
+                counterwidth[j] = strlen(values->countername[j]);
+        for (i = 0; i < values->threads; i++) {
+                int width;
+
+                width = snprintf(NULL, 0, "%d", values->pid[i]);
+                if (width > pidwidth)
+                        pidwidth = width;
+                width = snprintf(NULL, 0, "%d", values->tid[i]);
+                if (width > tidwidth)
+                        tidwidth = width;
+                for (j = 0; j < values->counters; j++) {
+                        width = snprintf(NULL, 0, "%Lu", values->value[i][j]);
+                        if (width > counterwidth[j])
+                                counterwidth[j] = width;
+                }
+        }
+
+        fprintf(fp, "# %*s  %*s", pidwidth, "PID", tidwidth, "TID");
+        for (j = 0; j < values->counters; j++)
+                fprintf(fp, "  %*s", counterwidth[j], values->countername[j]);
+        fprintf(fp, "\n");
+
+        for (i = 0; i < values->threads; i++) {
+                fprintf(fp, "  %*d  %*d", pidwidth, values->pid[i],
+                        tidwidth, values->tid[i]);
+                for (j = 0; j < values->counters; j++)
+                        fprintf(fp, "  %*Lu",
+                                counterwidth[j], values->value[i][j]);
+                fprintf(fp, "\n");
+        }
+        free(counterwidth);
+}
+
+static void perf_read_values__display_raw(FILE *fp,
+                                          struct perf_read_values *values)
+{
+        int width, pidwidth, tidwidth, namewidth, rawwidth, countwidth;
+        int i, j;
+
+        tidwidth = 3; /* TID */
+        pidwidth = 3; /* PID */
+        namewidth = 4; /* "Name" */
+        rawwidth = 3; /* "Raw" */
+        countwidth = 5; /* "Count" */
+
+        for (i = 0; i < values->threads; i++) {
+                width = snprintf(NULL, 0, "%d", values->pid[i]);
+                if (width > pidwidth)
+                        pidwidth = width;
+                width = snprintf(NULL, 0, "%d", values->tid[i]);
+                if (width > tidwidth)
+                        tidwidth = width;
+        }
+        for (j = 0; j < values->counters; j++) {
+                width = strlen(values->countername[j]);
+                if (width > namewidth)
+                        namewidth = width;
+                width = snprintf(NULL, 0, "%llx", values->counterrawid[j]);
+                if (width > rawwidth)
+                        rawwidth = width;
+        }
+        for (i = 0; i < values->threads; i++) {
+                for (j = 0; j < values->counters; j++) {
+                        width = snprintf(NULL, 0, "%Lu", values->value[i][j]);
+                        if (width > countwidth)
+                                countwidth = width;
+                }
+        }
+
+        fprintf(fp, "# %*s  %*s  %*s  %*s  %*s\n",
+                pidwidth, "PID", tidwidth, "TID",
+                namewidth, "Name", rawwidth, "Raw",
+                countwidth, "Count");
+        for (i = 0; i < values->threads; i++)
+                for (j = 0; j < values->counters; j++)
+                        fprintf(fp, "  %*d  %*d  %*s  %*llx  %*Lu\n",
+                                pidwidth, values->pid[i],
+                                tidwidth, values->tid[i],
+                                namewidth, values->countername[j],
+                                rawwidth, values->counterrawid[j],
+                                countwidth, values->value[i][j]);
+}
+
+void perf_read_values_display(FILE *fp, struct perf_read_values *values, int raw)
+{
+        if (raw)
+                perf_read_values__display_raw(fp, values);
+        else
+                perf_read_values__display_pretty(fp, values);
+}
diff --git a/tools/perf/util/values.h b/tools/perf/util/values.h
new file mode 100644
index 00000000000..2fa967e1a88
--- /dev/null
+++ b/tools/perf/util/values.h
@@ -0,0 +1,27 @@
+#ifndef __PERF_VALUES_H
+#define __PERF_VALUES_H
+
+#include "types.h"
+
+struct perf_read_values {
+        int threads;
+        int threads_max;
+        u32 *pid, *tid;
+        int counters;
+        int counters_max;
+        u64 *counterrawid;
+        char **countername;
+        u64 **value;
+};
+
+void perf_read_values_init(struct perf_read_values *values);
+void perf_read_values_destroy(struct perf_read_values *values);
+
+void perf_read_values_add_value(struct perf_read_values *values,
+                                u32 pid, u32 tid,
+                                u64 rawid, const char *name, u64 value);
+
+void perf_read_values_display(FILE *fp, struct perf_read_values *values,
+                              int raw);
+
+#endif /* __PERF_VALUES_H */
diff --git a/tools/perf/util/wrapper.c b/tools/perf/util/wrapper.c
index 6350d65f6d9..73e900edb5a 100644
--- a/tools/perf/util/wrapper.c
+++ b/tools/perf/util/wrapper.c
@@ -7,7 +7,7 @@
  * There's no pack memory to release - but stay close to the Git
  * version so wrap this away:
  */
-static inline void release_pack_memory(size_t size, int flag)
+static inline void release_pack_memory(size_t size __used, int flag __used)
 {
 }
 
@@ -23,45 +23,6 @@ char *xstrdup(const char *str)
         return ret;
 }
 
-void *xmalloc(size_t size)
-{
-        void *ret = malloc(size);
-        if (!ret && !size)
-                ret = malloc(1);
-        if (!ret) {
-                release_pack_memory(size, -1);
-                ret = malloc(size);
-                if (!ret && !size)
-                        ret = malloc(1);
-                if (!ret)
-                        die("Out of memory, malloc failed");
-        }
-#ifdef XMALLOC_POISON
-        memset(ret, 0xA5, size);
-#endif
-        return ret;
-}
-
-/*
- * xmemdupz() allocates (len + 1) bytes of memory, duplicates "len" bytes of
- * "data" to the allocated memory, zero terminates the allocated memory,
- * and returns a pointer to the allocated memory. If the allocation fails,
- * the program dies.
- */
-void *xmemdupz(const void *data, size_t len)
-{
-        char *p = xmalloc(len + 1);
-        memcpy(p, data, len);
-        p[len] = '\0';
-        return p;
-}
-
-char *xstrndup(const char *str, size_t len)
-{
-        char *p = memchr(str, '\0', len);
-        return xmemdupz(str, p ? p - str : len);
-}
-
 void *xrealloc(void *ptr, size_t size)
 {
         void *ret = realloc(ptr, size);
@@ -77,130 +38,3 @@ void *xrealloc(void *ptr, size_t size)
         }
         return ret;
 }
-
-void *xcalloc(size_t nmemb, size_t size)
-{
-        void *ret = calloc(nmemb, size);
-        if (!ret && (!nmemb || !size))
-                ret = calloc(1, 1);
-        if (!ret) {
-                release_pack_memory(nmemb * size, -1);
-                ret = calloc(nmemb, size);
-                if (!ret && (!nmemb || !size))
-                        ret = calloc(1, 1);
-                if (!ret)
-                        die("Out of memory, calloc failed");
-        }
-        return ret;
-}
-
-void *xmmap(void *start, size_t length,
-        int prot, int flags, int fd, off_t offset)
-{
-        void *ret = mmap(start, length, prot, flags, fd, offset);
-        if (ret == MAP_FAILED) {
-                if (!length)
-                        return NULL;
-                release_pack_memory(length, fd);
-                ret = mmap(start, length, prot, flags, fd, offset);
-                if (ret == MAP_FAILED)
-                        die("Out of memory? mmap failed: %s", strerror(errno));
-        }
-        return ret;
-}
-
-/*
- * xread() is the same a read(), but it automatically restarts read()
- * operations with a recoverable error (EAGAIN and EINTR). xread()
- * DOES NOT GUARANTEE that "len" bytes is read even if the data is available.
- */
-ssize_t xread(int fd, void *buf, size_t len)
-{
-        ssize_t nr;
-        while (1) {
-                nr = read(fd, buf, len);
-                if ((nr < 0) && (errno == EAGAIN || errno == EINTR))
-                        continue;
-                return nr;
-        }
-}
-
-/*
- * xwrite() is the same a write(), but it automatically restarts write()
- * operations with a recoverable error (EAGAIN and EINTR). xwrite() DOES NOT
- * GUARANTEE that "len" bytes is written even if the operation is successful.
- */
-ssize_t xwrite(int fd, const void *buf, size_t len)
-{
-        ssize_t nr;
-        while (1) {
-                nr = write(fd, buf, len);
-                if ((nr < 0) && (errno == EAGAIN || errno == EINTR))
-                        continue;
-                return nr;
-        }
-}
-
-ssize_t read_in_full(int fd, void *buf, size_t count)
-{
-        char *p = buf;
-        ssize_t total = 0;
-
-        while (count > 0) {
-                ssize_t loaded = xread(fd, p, count);
-                if (loaded <= 0)
-                        return total ? total : loaded;
-                count -= loaded;
-                p += loaded;
-                total += loaded;
-        }
-
-        return total;
-}
-
-ssize_t write_in_full(int fd, const void *buf, size_t count)
-{
-        const char *p = buf;
-        ssize_t total = 0;
-
-        while (count > 0) {
-                ssize_t written = xwrite(fd, p, count);
-                if (written < 0)
-                        return -1;
-                if (!written) {
-                        errno = ENOSPC;
-                        return -1;
-                }
-                count -= written;
-                p += written;
-                total += written;
-        }
-
-        return total;
-}
-
-int xdup(int fd)
-{
-        int ret = dup(fd);
-        if (ret < 0)
-                die("dup failed: %s", strerror(errno));
-        return ret;
-}
-
-FILE *xfdopen(int fd, const char *mode)
-{
-        FILE *stream = fdopen(fd, mode);
-        if (stream == NULL)
-                die("Out of memory? fdopen failed: %s", strerror(errno));
-        return stream;
-}
-
-int xmkstemp(char *template)
-{
-        int fd;
-
-        fd = mkstemp(template);
-        if (fd < 0)
-                die("Unable to create temporary file: %s", strerror(errno));
-        return fd;
-}
diff --git a/tools/perf/util/xyarray.c b/tools/perf/util/xyarray.c
new file mode 100644
index 00000000000..22afbf6c536
--- /dev/null
+++ b/tools/perf/util/xyarray.c
@@ -0,0 +1,20 @@
+#include "xyarray.h"
+#include "util.h"
+
+struct xyarray *xyarray__new(int xlen, int ylen, size_t entry_size)
+{
+        size_t row_size = ylen * entry_size;
+        struct xyarray *xy = zalloc(sizeof(*xy) + xlen * row_size);
+
+        if (xy != NULL) {
+                xy->entry_size = entry_size;
+                xy->row_size   = row_size;
+        }
+
+        return xy;
+}
+
+void xyarray__delete(struct xyarray *xy)
+{
+        free(xy);
+}
diff --git a/tools/perf/util/xyarray.h b/tools/perf/util/xyarray.h
new file mode 100644
index 00000000000..c488a07275d
--- /dev/null
+++ b/tools/perf/util/xyarray.h
@@ -0,0 +1,20 @@
+#ifndef _PERF_XYARRAY_H_
+#define _PERF_XYARRAY_H_ 1
+
+#include <sys/types.h>
+
+struct xyarray {
+        size_t row_size;
+        size_t entry_size;
+        char contents[];
+};
+
+struct xyarray *xyarray__new(int xlen, int ylen, size_t entry_size);
+void xyarray__delete(struct xyarray *xy);
+
+static inline void *xyarray__entry(struct xyarray *xy, int x, int y)
+{
+        return &xy->contents[x * xy->row_size + y * xy->entry_size];
+}
+
+#endif /* _PERF_XYARRAY_H_ */