Merge branch 'linus' into tracing/hw-breakpoints

Conflicts:
	arch/x86/Kconfig
	arch/x86/kernel/traps.c
	arch/x86/power/cpu.c
	arch/x86/power/cpu_32.c
	kernel/Makefile

Semantic conflict:
	arch/x86/kernel/hw_breakpoint.c

Merge reason: Resolve the conflicts, move from put_cpu_no_sched() to
              put_cpu() in arch/x86/kernel/hw_breakpoint.c.

Signed-off-by: Ingo Molnar <mingo@elte.hu>

69 files changed, 15431 insertions, 0 deletions

diff --git a/tools/perf/.gitignore b/tools/perf/.gitignore
new file mode 100644
index 000000000000..d69a759a1046
--- /dev/null
+++ b/tools/perf/.gitignore
@@ -0,0 +1,16 @@
+PERF-BUILD-OPTIONS
+PERF-CFLAGS
+PERF-GUI-VARS
+PERF-VERSION-FILE
+perf
+perf-help
+perf-record
+perf-report
+perf-stat
+perf-top
+perf*.1
+perf*.xml
+common-cmds.h
+tags
+TAGS
+cscope*
diff --git a/tools/perf/Documentation/Makefile b/tools/perf/Documentation/Makefile
new file mode 100644
index 000000000000..5457192e1b41
--- /dev/null
+++ b/tools/perf/Documentation/Makefile
@@ -0,0 +1,300 @@
+MAN1_TXT= \
+	$(filter-out $(addsuffix .txt, $(ARTICLES) $(SP_ARTICLES)), \
+		$(wildcard perf-*.txt)) \
+	perf.txt
+MAN5_TXT=
+MAN7_TXT=
+
+MAN_TXT = $(MAN1_TXT) $(MAN5_TXT) $(MAN7_TXT)
+MAN_XML=$(patsubst %.txt,%.xml,$(MAN_TXT))
+MAN_HTML=$(patsubst %.txt,%.html,$(MAN_TXT))
+
+DOC_HTML=$(MAN_HTML)
+
+ARTICLES =
+# with their own formatting rules.
+SP_ARTICLES =
+API_DOCS = $(patsubst %.txt,%,$(filter-out technical/api-index-skel.txt technical/api-index.txt, $(wildcard technical/api-*.txt)))
+SP_ARTICLES += $(API_DOCS)
+SP_ARTICLES += technical/api-index
+
+DOC_HTML += $(patsubst %,%.html,$(ARTICLES) $(SP_ARTICLES))
+
+DOC_MAN1=$(patsubst %.txt,%.1,$(MAN1_TXT))
+DOC_MAN5=$(patsubst %.txt,%.5,$(MAN5_TXT))
+DOC_MAN7=$(patsubst %.txt,%.7,$(MAN7_TXT))
+
+prefix?=$(HOME)
+bindir?=$(prefix)/bin
+htmldir?=$(prefix)/share/doc/perf-doc
+pdfdir?=$(prefix)/share/doc/perf-doc
+mandir?=$(prefix)/share/man
+man1dir=$(mandir)/man1
+man5dir=$(mandir)/man5
+man7dir=$(mandir)/man7
+# DESTDIR=
+
+ASCIIDOC=asciidoc
+ASCIIDOC_EXTRA =
+MANPAGE_XSL = manpage-normal.xsl
+XMLTO_EXTRA =
+INSTALL?=install
+RM ?= rm -f
+DOC_REF = origin/man
+HTML_REF = origin/html
+
+infodir?=$(prefix)/share/info
+MAKEINFO=makeinfo
+INSTALL_INFO=install-info
+DOCBOOK2X_TEXI=docbook2x-texi
+DBLATEX=dblatex
+ifndef PERL_PATH
+	PERL_PATH = /usr/bin/perl
+endif
+
+-include ../config.mak.autogen
+-include ../config.mak
+
+#
+# For asciidoc ...
+#	-7.1.2,	no extra settings are needed.
+#	8.0-,	set ASCIIDOC8.
+#
+
+#
+# For docbook-xsl ...
+#	-1.68.1,	set ASCIIDOC_NO_ROFF? (based on changelog from 1.73.0)
+#	1.69.0,		no extra settings are needed?
+#	1.69.1-1.71.0,	set DOCBOOK_SUPPRESS_SP?
+#	1.71.1,		no extra settings are needed?
+#	1.72.0,		set DOCBOOK_XSL_172.
+#	1.73.0-,	set ASCIIDOC_NO_ROFF
+#
+
+#
+# If you had been using DOCBOOK_XSL_172 in an attempt to get rid
+# of 'the ".ft C" problem' in your generated manpages, and you
+# instead ended up with weird characters around callouts, try
+# using ASCIIDOC_NO_ROFF instead (it works fine with ASCIIDOC8).
+#
+
+ifdef ASCIIDOC8
+ASCIIDOC_EXTRA += -a asciidoc7compatible
+endif
+ifdef DOCBOOK_XSL_172
+ASCIIDOC_EXTRA += -a perf-asciidoc-no-roff
+MANPAGE_XSL = manpage-1.72.xsl
+else
+	ifdef ASCIIDOC_NO_ROFF
+	# docbook-xsl after 1.72 needs the regular XSL, but will not
+	# pass-thru raw roff codes from asciidoc.conf, so turn them off.
+	ASCIIDOC_EXTRA += -a perf-asciidoc-no-roff
+	endif
+endif
+ifdef MAN_BOLD_LITERAL
+XMLTO_EXTRA += -m manpage-bold-literal.xsl
+endif
+ifdef DOCBOOK_SUPPRESS_SP
+XMLTO_EXTRA += -m manpage-suppress-sp.xsl
+endif
+
+SHELL_PATH ?= $(SHELL)
+# Shell quote;
+SHELL_PATH_SQ = $(subst ','\'',$(SHELL_PATH))
+
+#
+# Please note that there is a minor bug in asciidoc.
+# The version after 6.0.3 _will_ include the patch found here:
+#   http://marc.theaimsgroup.com/?l=perf&m=111558757202243&w=2
+#
+# Until that version is released you may have to apply the patch
+# yourself - yes, all 6 characters of it!
+#
+
+QUIET_SUBDIR0  = +$(MAKE) -C # space to separate -C and subdir
+QUIET_SUBDIR1  =
+
+ifneq ($(findstring $(MAKEFLAGS),w),w)
+PRINT_DIR = --no-print-directory
+else # "make -w"
+NO_SUBDIR = :
+endif
+
+ifneq ($(findstring $(MAKEFLAGS),s),s)
+ifndef V
+	QUIET_ASCIIDOC	= @echo '   ' ASCIIDOC $@;
+	QUIET_XMLTO	= @echo '   ' XMLTO $@;
+	QUIET_DB2TEXI	= @echo '   ' DB2TEXI $@;
+	QUIET_MAKEINFO	= @echo '   ' MAKEINFO $@;
+	QUIET_DBLATEX	= @echo '   ' DBLATEX $@;
+	QUIET_XSLTPROC	= @echo '   ' XSLTPROC $@;
+	QUIET_GEN	= @echo '   ' GEN $@;
+	QUIET_STDERR	= 2> /dev/null
+	QUIET_SUBDIR0	= +@subdir=
+	QUIET_SUBDIR1	= ;$(NO_SUBDIR) echo '   ' SUBDIR $$subdir; \
+			  $(MAKE) $(PRINT_DIR) -C $$subdir
+	export V
+endif
+endif
+
+all: html man
+
+html: $(DOC_HTML)
+
+$(DOC_HTML) $(DOC_MAN1) $(DOC_MAN5) $(DOC_MAN7): asciidoc.conf
+
+man: man1 man5 man7
+man1: $(DOC_MAN1)
+man5: $(DOC_MAN5)
+man7: $(DOC_MAN7)
+
+info: perf.info perfman.info
+
+pdf: user-manual.pdf
+
+install: install-man
+
+install-man: man
+	$(INSTALL) -d -m 755 $(DESTDIR)$(man1dir)
+#	$(INSTALL) -d -m 755 $(DESTDIR)$(man5dir)
+#	$(INSTALL) -d -m 755 $(DESTDIR)$(man7dir)
+	$(INSTALL) -m 644 $(DOC_MAN1) $(DESTDIR)$(man1dir)
+#	$(INSTALL) -m 644 $(DOC_MAN5) $(DESTDIR)$(man5dir)
+#	$(INSTALL) -m 644 $(DOC_MAN7) $(DESTDIR)$(man7dir)
+
+install-info: info
+	$(INSTALL) -d -m 755 $(DESTDIR)$(infodir)
+	$(INSTALL) -m 644 perf.info perfman.info $(DESTDIR)$(infodir)
+	if test -r $(DESTDIR)$(infodir)/dir; then \
+	  $(INSTALL_INFO) --info-dir=$(DESTDIR)$(infodir) perf.info ;\
+	  $(INSTALL_INFO) --info-dir=$(DESTDIR)$(infodir) perfman.info ;\
+	else \
+	  echo "No directory found in $(DESTDIR)$(infodir)" >&2 ; \
+	fi
+
+install-pdf: pdf
+	$(INSTALL) -d -m 755 $(DESTDIR)$(pdfdir)
+	$(INSTALL) -m 644 user-manual.pdf $(DESTDIR)$(pdfdir)
+
+install-html: html
+	'$(SHELL_PATH_SQ)' ./install-webdoc.sh $(DESTDIR)$(htmldir)
+
+../PERF-VERSION-FILE: .FORCE-PERF-VERSION-FILE
+	$(QUIET_SUBDIR0)../ $(QUIET_SUBDIR1) PERF-VERSION-FILE
+
+-include ../PERF-VERSION-FILE
+
+#
+# Determine "include::" file references in asciidoc files.
+#
+doc.dep : $(wildcard *.txt) build-docdep.perl
+	$(QUIET_GEN)$(RM) $@+ $@ && \
+	$(PERL_PATH) ./build-docdep.perl >$@+ $(QUIET_STDERR) && \
+	mv $@+ $@
+
+-include doc.dep
+
+cmds_txt = cmds-ancillaryinterrogators.txt \
+	cmds-ancillarymanipulators.txt \
+	cmds-mainporcelain.txt \
+	cmds-plumbinginterrogators.txt \
+	cmds-plumbingmanipulators.txt \
+	cmds-synchingrepositories.txt \
+	cmds-synchelpers.txt \
+	cmds-purehelpers.txt \
+	cmds-foreignscminterface.txt
+
+$(cmds_txt): cmd-list.made
+
+cmd-list.made: cmd-list.perl ../command-list.txt $(MAN1_TXT)
+	$(QUIET_GEN)$(RM) $@ && \
+	$(PERL_PATH) ./cmd-list.perl ../command-list.txt $(QUIET_STDERR) && \
+	date >$@
+
+clean:
+	$(RM) *.xml *.xml+ *.html *.html+ *.1 *.5 *.7
+	$(RM) *.texi *.texi+ *.texi++ perf.info perfman.info
+	$(RM) howto-index.txt howto/*.html doc.dep
+	$(RM) technical/api-*.html technical/api-index.txt
+	$(RM) $(cmds_txt) *.made
+
+$(MAN_HTML): %.html : %.txt
+	$(QUIET_ASCIIDOC)$(RM) $@+ $@ && \
+	$(ASCIIDOC) -b xhtml11 -d manpage -f asciidoc.conf \
+		$(ASCIIDOC_EXTRA) -aperf_version=$(PERF_VERSION) -o $@+ $< && \
+	mv $@+ $@
+
+%.1 %.5 %.7 : %.xml
+	$(QUIET_XMLTO)$(RM) $@ && \
+	xmlto -m $(MANPAGE_XSL) $(XMLTO_EXTRA) man $<
+
+%.xml : %.txt
+	$(QUIET_ASCIIDOC)$(RM) $@+ $@ && \
+	$(ASCIIDOC) -b docbook -d manpage -f asciidoc.conf \
+		$(ASCIIDOC_EXTRA) -aperf_version=$(PERF_VERSION) -o $@+ $< && \
+	mv $@+ $@
+
+XSLT = docbook.xsl
+XSLTOPTS = --xinclude --stringparam html.stylesheet docbook-xsl.css
+
+user-manual.html: user-manual.xml
+	$(QUIET_XSLTPROC)xsltproc $(XSLTOPTS) -o $@ $(XSLT) $<
+
+perf.info: user-manual.texi
+	$(QUIET_MAKEINFO)$(MAKEINFO) --no-split -o $@ user-manual.texi
+
+user-manual.texi: user-manual.xml
+	$(QUIET_DB2TEXI)$(RM) $@+ $@ && \
+	$(DOCBOOK2X_TEXI) user-manual.xml --encoding=UTF-8 --to-stdout >$@++ && \
+	$(PERL_PATH) fix-texi.perl <$@++ >$@+ && \
+	rm $@++ && \
+	mv $@+ $@
+
+user-manual.pdf: user-manual.xml
+	$(QUIET_DBLATEX)$(RM) $@+ $@ && \
+	$(DBLATEX) -o $@+ -p /etc/asciidoc/dblatex/asciidoc-dblatex.xsl -s /etc/asciidoc/dblatex/asciidoc-dblatex.sty $< && \
+	mv $@+ $@
+
+perfman.texi: $(MAN_XML) cat-texi.perl
+	$(QUIET_DB2TEXI)$(RM) $@+ $@ && \
+	($(foreach xml,$(MAN_XML),$(DOCBOOK2X_TEXI) --encoding=UTF-8 \
+		--to-stdout $(xml) &&) true) > $@++ && \
+	$(PERL_PATH) cat-texi.perl $@ <$@++ >$@+ && \
+	rm $@++ && \
+	mv $@+ $@
+
+perfman.info: perfman.texi
+	$(QUIET_MAKEINFO)$(MAKEINFO) --no-split --no-validate $*.texi
+
+$(patsubst %.txt,%.texi,$(MAN_TXT)): %.texi : %.xml
+	$(QUIET_DB2TEXI)$(RM) $@+ $@ && \
+	$(DOCBOOK2X_TEXI) --to-stdout $*.xml >$@+ && \
+	mv $@+ $@
+
+howto-index.txt: howto-index.sh $(wildcard howto/*.txt)
+	$(QUIET_GEN)$(RM) $@+ $@ && \
+	'$(SHELL_PATH_SQ)' ./howto-index.sh $(wildcard howto/*.txt) >$@+ && \
+	mv $@+ $@
+
+$(patsubst %,%.html,$(ARTICLES)) : %.html : %.txt
+	$(QUIET_ASCIIDOC)$(ASCIIDOC) -b xhtml11 $*.txt
+
+WEBDOC_DEST = /pub/software/tools/perf/docs
+
+$(patsubst %.txt,%.html,$(wildcard howto/*.txt)): %.html : %.txt
+	$(QUIET_ASCIIDOC)$(RM) $@+ $@ && \
+	sed -e '1,/^$$/d' $< | $(ASCIIDOC) -b xhtml11 - >$@+ && \
+	mv $@+ $@
+
+install-webdoc : html
+	'$(SHELL_PATH_SQ)' ./install-webdoc.sh $(WEBDOC_DEST)
+
+quick-install: quick-install-man
+
+quick-install-man:
+	'$(SHELL_PATH_SQ)' ./install-doc-quick.sh $(DOC_REF) $(DESTDIR)$(mandir)
+
+quick-install-html:
+	'$(SHELL_PATH_SQ)' ./install-doc-quick.sh $(HTML_REF) $(DESTDIR)$(htmldir)
+
+.PHONY: .FORCE-PERF-VERSION-FILE
diff --git a/tools/perf/Documentation/asciidoc.conf b/tools/perf/Documentation/asciidoc.conf
new file mode 100644
index 000000000000..356b23a40339
--- /dev/null
+++ b/tools/perf/Documentation/asciidoc.conf
@@ -0,0 +1,91 @@
+## linkperf: macro
+#
+# Usage: linkperf:command[manpage-section]
+#
+# Note, {0} is the manpage section, while {target} is the command.
+#
+# Show PERF link as: <command>(<section>); if section is defined, else just show
+# the command.
+
+[macros]
+(?su)[\\]?(?P<name>linkperf):(?P<target>\S*?)\[(?P<attrlist>.*?)\]=
+
+[attributes]
+asterisk=&#42;
+plus=&#43;
+caret=&#94;
+startsb=&#91;
+endsb=&#93;
+tilde=&#126;
+
+ifdef::backend-docbook[]
+[linkperf-inlinemacro]
+{0%{target}}
+{0#<citerefentry>}
+{0#<refentrytitle>{target}</refentrytitle><manvolnum>{0}</manvolnum>}
+{0#</citerefentry>}
+endif::backend-docbook[]
+
+ifdef::backend-docbook[]
+ifndef::perf-asciidoc-no-roff[]
+# "unbreak" docbook-xsl v1.68 for manpages. v1.69 works with or without this.
+# v1.72 breaks with this because it replaces dots not in roff requests.
+[listingblock]
+<example><title>{title}</title>
+<literallayout>
+ifdef::doctype-manpage[]
+&#10;.ft C&#10;
+endif::doctype-manpage[]
+|
+ifdef::doctype-manpage[]
+&#10;.ft&#10;
+endif::doctype-manpage[]
+</literallayout>
+{title#}</example>
+endif::perf-asciidoc-no-roff[]
+
+ifdef::perf-asciidoc-no-roff[]
+ifdef::doctype-manpage[]
+# The following two small workarounds insert a simple paragraph after screen
+[listingblock]
+<example><title>{title}</title>
+<literallayout>
+|
+</literallayout><simpara></simpara>
+{title#}</example>
+
+[verseblock]
+<formalpara{id? id="{id}"}><title>{title}</title><para>
+{title%}<literallayout{id? id="{id}"}>
+{title#}<literallayout>
+|
+</literallayout>
+{title#}</para></formalpara>
+{title%}<simpara></simpara>
+endif::doctype-manpage[]
+endif::perf-asciidoc-no-roff[]
+endif::backend-docbook[]
+
+ifdef::doctype-manpage[]
+ifdef::backend-docbook[]
+[header]
+template::[header-declarations]
+<refentry>
+<refmeta>
+<refentrytitle>{mantitle}</refentrytitle>
+<manvolnum>{manvolnum}</manvolnum>
+<refmiscinfo class="source">perf</refmiscinfo>
+<refmiscinfo class="version">{perf_version}</refmiscinfo>
+<refmiscinfo class="manual">perf Manual</refmiscinfo>
+</refmeta>
+<refnamediv>
+  <refname>{manname}</refname>
+  <refpurpose>{manpurpose}</refpurpose>
+</refnamediv>
+endif::backend-docbook[]
+endif::doctype-manpage[]
+
+ifdef::backend-xhtml11[]
+[linkperf-inlinemacro]
+<a href="{target}.html">{target}{0?({0})}</a>
+endif::backend-xhtml11[]
diff --git a/tools/perf/Documentation/manpage-1.72.xsl b/tools/perf/Documentation/manpage-1.72.xsl
new file mode 100644
index 000000000000..b4d315cb8c47
--- /dev/null
+++ b/tools/perf/Documentation/manpage-1.72.xsl
@@ -0,0 +1,14 @@
+<!-- manpage-1.72.xsl:
+     special settings for manpages rendered from asciidoc+docbook
+     handles peculiarities in docbook-xsl 1.72.0 -->
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+		version="1.0">
+
+<xsl:import href="manpage-base.xsl"/>
+
+<!-- these are the special values for the roff control characters
+     needed for docbook-xsl 1.72.0 -->
+<xsl:param name="git.docbook.backslash">&#x2593;</xsl:param>
+<xsl:param name="git.docbook.dot"      >&#x2302;</xsl:param>
+
+</xsl:stylesheet>
diff --git a/tools/perf/Documentation/manpage-base.xsl b/tools/perf/Documentation/manpage-base.xsl
new file mode 100644
index 000000000000..a264fa616093
--- /dev/null
+++ b/tools/perf/Documentation/manpage-base.xsl
@@ -0,0 +1,35 @@
+<!-- manpage-base.xsl:
+     special formatting for manpages rendered from asciidoc+docbook -->
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+		version="1.0">
+
+<!-- these params silence some output from xmlto -->
+<xsl:param name="man.output.quietly" select="1"/>
+<xsl:param name="refentry.meta.get.quietly" select="1"/>
+
+<!-- convert asciidoc callouts to man page format;
+     git.docbook.backslash and git.docbook.dot params
+     must be supplied by another XSL file or other means -->
+<xsl:template match="co">
+	<xsl:value-of select="concat(
+			      $git.docbook.backslash,'fB(',
+			      substring-after(@id,'-'),')',
+			      $git.docbook.backslash,'fR')"/>
+</xsl:template>
+<xsl:template match="calloutlist">
+	<xsl:value-of select="$git.docbook.dot"/>
+	<xsl:text>sp&#10;</xsl:text>
+	<xsl:apply-templates/>
+	<xsl:text>&#10;</xsl:text>
+</xsl:template>
+<xsl:template match="callout">
+	<xsl:value-of select="concat(
+			      $git.docbook.backslash,'fB',
+			      substring-after(@arearefs,'-'),
+			      '. ',$git.docbook.backslash,'fR')"/>
+	<xsl:apply-templates/>
+	<xsl:value-of select="$git.docbook.dot"/>
+	<xsl:text>br&#10;</xsl:text>
+</xsl:template>
+
+</xsl:stylesheet>
diff --git a/tools/perf/Documentation/manpage-bold-literal.xsl b/tools/perf/Documentation/manpage-bold-literal.xsl
new file mode 100644
index 000000000000..608eb5df6281
--- /dev/null
+++ b/tools/perf/Documentation/manpage-bold-literal.xsl
@@ -0,0 +1,17 @@
+<!-- manpage-bold-literal.xsl:
+     special formatting for manpages rendered from asciidoc+docbook -->
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+		version="1.0">
+
+<!-- render literal text as bold (instead of plain or monospace);
+     this makes literal text easier to distinguish in manpages
+     viewed on a tty -->
+<xsl:template match="literal">
+	<xsl:value-of select="$git.docbook.backslash"/>
+	<xsl:text>fB</xsl:text>
+	<xsl:apply-templates/>
+	<xsl:value-of select="$git.docbook.backslash"/>
+	<xsl:text>fR</xsl:text>
+</xsl:template>
+
+</xsl:stylesheet>
diff --git a/tools/perf/Documentation/manpage-normal.xsl b/tools/perf/Documentation/manpage-normal.xsl
new file mode 100644
index 000000000000..a48f5b11f3dc
--- /dev/null
+++ b/tools/perf/Documentation/manpage-normal.xsl
@@ -0,0 +1,13 @@
+<!-- manpage-normal.xsl:
+     special settings for manpages rendered from asciidoc+docbook
+     handles anything we want to keep away from docbook-xsl 1.72.0 -->
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+		version="1.0">
+
+<xsl:import href="manpage-base.xsl"/>
+
+<!-- these are the normal values for the roff control characters -->
+<xsl:param name="git.docbook.backslash">\</xsl:param>
+<xsl:param name="git.docbook.dot"	>.</xsl:param>
+
+</xsl:stylesheet>
diff --git a/tools/perf/Documentation/manpage-suppress-sp.xsl b/tools/perf/Documentation/manpage-suppress-sp.xsl
new file mode 100644
index 000000000000..a63c7632a87d
--- /dev/null
+++ b/tools/perf/Documentation/manpage-suppress-sp.xsl
@@ -0,0 +1,21 @@
+<!-- manpage-suppress-sp.xsl:
+     special settings for manpages rendered from asciidoc+docbook
+     handles erroneous, inline .sp in manpage output of some
+     versions of docbook-xsl -->
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+		version="1.0">
+
+<!-- attempt to work around spurious .sp at the tail of the line
+     that some versions of docbook stylesheets seem to add -->
+<xsl:template match="simpara">
+  <xsl:variable name="content">
+    <xsl:apply-templates/>
+  </xsl:variable>
+  <xsl:value-of select="normalize-space($content)"/>
+  <xsl:if test="not(ancestor::authorblurb) and
+                not(ancestor::personblurb)">
+    <xsl:text>&#10;&#10;</xsl:text>
+  </xsl:if>
+</xsl:template>
+
+</xsl:stylesheet>
diff --git a/tools/perf/Documentation/perf-annotate.txt b/tools/perf/Documentation/perf-annotate.txt
new file mode 100644
index 000000000000..c9dcade06831
--- /dev/null
+++ b/tools/perf/Documentation/perf-annotate.txt
@@ -0,0 +1,29 @@
+perf-annotate(1)
+==============
+
+NAME
+----
+perf-annotate - Read perf.data (created by perf record) and display annotated code
+
+SYNOPSIS
+--------
+[verse]
+'perf annotate' [-i <file> | --input=file] symbol_name
+
+DESCRIPTION
+-----------
+This command reads the input file and displays an annotated version of the
+code. If the object file has debug symbols then the source code will be
+displayed alongside assembly code.
+
+If there is no debug info in the object, then annotated assembly is displayed.
+
+OPTIONS
+-------
+-i::
+--input=::
+        Input file name. (default: perf.data)
+
+SEE ALSO
+--------
+linkperf:perf-record[1]
diff --git a/tools/perf/Documentation/perf-help.txt b/tools/perf/Documentation/perf-help.txt
new file mode 100644
index 000000000000..514391818d1f
--- /dev/null
+++ b/tools/perf/Documentation/perf-help.txt
@@ -0,0 +1,38 @@
+perf-help(1)
+============
+
+NAME
+----
+perf-help - display help information about perf
+
+SYNOPSIS
+--------
+'perf help' [-a|--all] [COMMAND]
+
+DESCRIPTION
+-----------
+
+With no options and no COMMAND given, the synopsis of the 'perf'
+command and a list of the most commonly used perf commands are printed
+on the standard output.
+
+If the option '--all' or '-a' is given, then all available commands are
+printed on the standard output.
+
+If a perf command is named, a manual page for that command is brought
+up. The 'man' program is used by default for this purpose, but this
+can be overridden by other options or configuration variables.
+
+Note that `perf --help ...` is identical to `perf help ...` because the
+former is internally converted into the latter.
+
+OPTIONS
+-------
+-a::
+--all::
+	Prints all the available commands on the standard output. This
+	option supersedes any other option.
+
+PERF
+----
+Part of the linkperf:perf[1] suite
diff --git a/tools/perf/Documentation/perf-list.txt b/tools/perf/Documentation/perf-list.txt
new file mode 100644
index 000000000000..8290b9422668
--- /dev/null
+++ b/tools/perf/Documentation/perf-list.txt
@@ -0,0 +1,25 @@
+perf-list(1)
+============
+
+NAME
+----
+perf-list - List all symbolic event types
+
+SYNOPSIS
+--------
+[verse]
+'perf list'
+
+DESCRIPTION
+-----------
+This command displays the symbolic event types which can be selected in the
+various perf commands with the -e option.
+
+OPTIONS
+-------
+None
+
+SEE ALSO
+--------
+linkperf:perf-stat[1], linkperf:perf-top[1],
+linkperf:perf-record[1]
diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
new file mode 100644
index 000000000000..1dbc1eeb4c01
--- /dev/null
+++ b/tools/perf/Documentation/perf-record.txt
@@ -0,0 +1,42 @@
+perf-record(1)
+==============
+
+NAME
+----
+perf-record - Run a command and record its profile into perf.data
+
+SYNOPSIS
+--------
+[verse]
+'perf record' [-e <EVENT> | --event=EVENT] [-l] [-a] <command>
+'perf record' [-e <EVENT> | --event=EVENT] [-l] [-a] -- <command> [<options>]
+
+DESCRIPTION
+-----------
+This command runs a command and gathers a performance counter profile
+from it, into perf.data - without displaying anything.
+
+This file can then be inspected later on, using 'perf report'.
+
+
+OPTIONS
+-------
+<command>...::
+	Any command you can specify in a shell.
+
+-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.
+
+-a::
+        system-wide collection
+
+-l::
+        scale counter values
+
+SEE ALSO
+--------
+linkperf:perf-stat[1], linkperf:perf-list[1]
diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt
new file mode 100644
index 000000000000..52d3fc6846a9
--- /dev/null
+++ b/tools/perf/Documentation/perf-report.txt
@@ -0,0 +1,26 @@
+perf-report(1)
+==============
+
+NAME
+----
+perf-report - Read perf.data (created by perf record) and display the profile
+
+SYNOPSIS
+--------
+[verse]
+'perf report' [-i <file> | --input=file]
+
+DESCRIPTION
+-----------
+This command displays the performance counter profile information recorded
+via perf report.
+
+OPTIONS
+-------
+-i::
+--input=::
+        Input file name. (default: perf.data)
+
+SEE ALSO
+--------
+linkperf:perf-stat[1]
diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt
new file mode 100644
index 000000000000..c368a72721d7
--- /dev/null
+++ b/tools/perf/Documentation/perf-stat.txt
@@ -0,0 +1,66 @@
+perf-stat(1)
+============
+
+NAME
+----
+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>]
+
+DESCRIPTION
+-----------
+This command runs a command and gathers performance counter statistics
+from it.
+
+
+OPTIONS
+-------
+<command>...::
+	Any command you can specify in a shell.
+
+
+-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.
+
+-i::
+--inherit::
+        child tasks inherit counters
+-p::
+--pid=<pid>::
+        stat events on existing pid
+
+-a::
+        system-wide collection
+
+-l::
+        scale counter values
+
+EXAMPLES
+--------
+
+$ perf stat -- make -j
+
+ Performance counter stats for 'make -j':
+
+    8117.370256  task clock ticks     #      11.281 CPU utilization factor
+            678  context switches     #       0.000 M/sec
+            133  CPU migrations       #       0.000 M/sec
+         235724  pagefaults           #       0.029 M/sec
+    24821162526  CPU cycles           #    3057.784 M/sec
+    18687303457  instructions         #    2302.138 M/sec
+      172158895  cache references     #      21.209 M/sec
+       27075259  cache misses         #       3.335 M/sec
+
+ Wall-clock time elapsed:   719.554352 msecs
+
+SEE ALSO
+--------
+linkperf:perf-top[1], linkperf:perf-list[1]
diff --git a/tools/perf/Documentation/perf-top.txt b/tools/perf/Documentation/perf-top.txt
new file mode 100644
index 000000000000..539d01289725
--- /dev/null
+++ b/tools/perf/Documentation/perf-top.txt
@@ -0,0 +1,39 @@
+perf-top(1)
+===========
+
+NAME
+----
+perf-top - Run a command and profile it
+
+SYNOPSIS
+--------
+[verse]
+'perf top' [-e <EVENT> | --event=EVENT] [-l] [-a] <command>
+
+DESCRIPTION
+-----------
+This command runs a command and gathers a performance counter profile
+from it.
+
+
+OPTIONS
+-------
+<command>...::
+	Any command you can specify in a shell.
+
+-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.
+
+-a::
+        system-wide collection
+
+-l::
+        scale counter values
+
+SEE ALSO
+--------
+linkperf:perf-stat[1], linkperf:perf-list[1]
diff --git a/tools/perf/Documentation/perf.txt b/tools/perf/Documentation/perf.txt
new file mode 100644
index 000000000000..69c832557199
--- /dev/null
+++ b/tools/perf/Documentation/perf.txt
@@ -0,0 +1,24 @@
+perf(1)
+=======
+
+NAME
+----
+perf - Performance analysis tools for Linux
+
+SYNOPSIS
+--------
+[verse]
+'perf' [--version] [--help] COMMAND [ARGS]
+
+DESCRIPTION
+-----------
+Performance counters for Linux are 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.
+
+SEE ALSO
+--------
+linkperf:perf-stat[1], linkperf:perf-top[1],
+linkperf:perf-record[1], linkperf:perf-report[1],
+linkperf:perf-list[1]
diff --git a/tools/perf/Makefile b/tools/perf/Makefile
new file mode 100644
index 000000000000..0cbd5d6874ec
--- /dev/null
+++ b/tools/perf/Makefile
@@ -0,0 +1,929 @@
+# The default target of this Makefile is...
+all::
+
+# Define V=1 to have a 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
+# have been written to the final string if enough space had been available.
+#
+# Define FREAD_READS_DIRECTORIES if your are on a system which succeeds
+# when attempting to read from an fopen'ed directory.
+#
+# Define NO_OPENSSL environment variable if you do not have OpenSSL.
+# This also implies MOZILLA_SHA1.
+#
+# Define CURLDIR=/foo/bar if your curl header and library files are in
+# /foo/bar/include and /foo/bar/lib directories.
+#
+# Define EXPATDIR=/foo/bar if your expat header and library files are in
+# /foo/bar/include and /foo/bar/lib directories.
+#
+# Define NO_D_INO_IN_DIRENT if you don't have d_ino in your struct dirent.
+#
+# Define NO_D_TYPE_IN_DIRENT if your platform defines DT_UNKNOWN but lacks
+# d_type in struct dirent (latest Cygwin -- will be fixed soonish).
+#
+# Define NO_C99_FORMAT if your formatted IO functions (printf/scanf et.al.)
+# do not support the 'size specifiers' introduced by C99, namely ll, hh,
+# j, z, t. (representing long long int, char, intmax_t, size_t, ptrdiff_t).
+# some C compilers supported these specifiers prior to C99 as an extension.
+#
+# Define NO_STRCASESTR if you don't have strcasestr.
+#
+# Define NO_MEMMEM if you don't have memmem.
+#
+# Define NO_STRTOUMAX if you don't have strtoumax in the C library.
+# If your compiler also does not support long long or does not have
+# strtoull, define NO_STRTOULL.
+#
+# Define NO_SETENV if you don't have setenv in the C library.
+#
+# Define NO_UNSETENV if you don't have unsetenv in the C library.
+#
+# Define NO_MKDTEMP if you don't have mkdtemp in the C library.
+#
+# Define NO_SYS_SELECT_H if you don't have sys/select.h.
+#
+# Define NO_SYMLINK_HEAD if you never want .perf/HEAD to be a symbolic link.
+# Enable it on Windows.  By default, symrefs are still used.
+#
+# Define NO_SVN_TESTS if you want to skip time-consuming SVN interoperability
+# tests.  These tests take up a significant amount of the total test time
+# but are not needed unless you plan to talk to SVN repos.
+#
+# Define NO_FINK if you are building on Darwin/Mac OS X, have Fink
+# installed in /sw, but don't want PERF to link against any libraries
+# installed there.  If defined you may specify your own (or Fink's)
+# include directories and library directories by defining CFLAGS
+# and LDFLAGS appropriately.
+#
+# Define NO_DARWIN_PORTS if you are building on Darwin/Mac OS X,
+# have DarwinPorts installed in /opt/local, but don't want PERF to
+# link against any libraries installed there.  If defined you may
+# specify your own (or DarwinPort's) include directories and
+# library directories by defining CFLAGS and LDFLAGS appropriately.
+#
+# Define PPC_SHA1 environment variable when running make to make use of
+# a bundled SHA1 routine optimized for PowerPC.
+#
+# Define ARM_SHA1 environment variable when running make to make use of
+# a bundled SHA1 routine optimized for ARM.
+#
+# Define MOZILLA_SHA1 environment variable when running make to make use of
+# a bundled SHA1 routine coming from Mozilla. It is GPL'd and should be fast
+# on non-x86 architectures (e.g. PowerPC), while the OpenSSL version (default
+# choice) has very fast version optimized for i586.
+#
+# Define NEEDS_SSL_WITH_CRYPTO if you need -lcrypto with -lssl (Darwin).
+#
+# Define NEEDS_LIBICONV if linking with libc is not enough (Darwin).
+#
+# Define NEEDS_SOCKET if linking with libc is not enough (SunOS,
+# Patrick Mauritz).
+#
+# Define NO_MMAP if you want to avoid mmap.
+#
+# Define NO_PTHREADS if you do not have or do not want to use Pthreads.
+#
+# Define NO_PREAD if you have a problem with pread() system call (e.g.
+# cygwin.dll before v1.5.22).
+#
+# Define NO_FAST_WORKING_DIRECTORY if accessing objects in pack files is
+# generally faster on your platform than accessing the working directory.
+#
+# Define NO_TRUSTABLE_FILEMODE if your filesystem may claim to support
+# the executable mode bit, but doesn't really do so.
+#
+# Define NO_IPV6 if you lack IPv6 support and getaddrinfo().
+#
+# Define NO_SOCKADDR_STORAGE if your platform does not have struct
+# sockaddr_storage.
+#
+# Define NO_ICONV if your libc does not properly support iconv.
+#
+# Define OLD_ICONV if your library has an old iconv(), where the second
+# (input buffer pointer) parameter is declared with type (const char **).
+#
+# Define NO_DEFLATE_BOUND if your zlib does not have deflateBound.
+#
+# Define NO_R_TO_GCC_LINKER if your gcc does not like "-R/path/lib"
+# that tells runtime paths to dynamic libraries;
+# "-Wl,-rpath=/path/lib" is used instead.
+#
+# Define USE_NSEC below if you want perf to care about sub-second file mtimes
+# and ctimes. Note that you need recent glibc (at least 2.2.4) for this, and
+# it will BREAK YOUR LOCAL DIFFS! show-diff and anything using it will likely
+# randomly break unless your underlying filesystem supports those sub-second
+# times (my ext3 doesn't).
+#
+# Define USE_ST_TIMESPEC if your "struct stat" uses "st_ctimespec" instead of
+# "st_ctim"
+#
+# Define NO_NSEC if your "struct stat" does not have "st_ctim.tv_nsec"
+# available.  This automatically turns USE_NSEC off.
+#
+# Define USE_STDEV below if you want perf to care about the underlying device
+# change being considered an inode change from the update-index perspective.
+#
+# Define NO_ST_BLOCKS_IN_STRUCT_STAT if your platform does not have st_blocks
+# field that counts the on-disk footprint in 512-byte blocks.
+#
+# Define ASCIIDOC8 if you want to format documentation with AsciiDoc 8
+#
+# Define DOCBOOK_XSL_172 if you want to format man pages with DocBook XSL v1.72.
+#
+# Define NO_PERL_MAKEMAKER if you cannot use Makefiles generated by perl's
+# MakeMaker (e.g. using ActiveState under Cygwin).
+#
+# Define NO_PERL if you do not want Perl scripts or libraries at all.
+#
+# Define INTERNAL_QSORT to use Git's implementation of qsort(), which
+# is a simplified version of the merge sort used in glibc. This is
+# recommended if Git triggers O(n^2) behavior in your platform's qsort().
+#
+# 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).
+
+PERF-VERSION-FILE: .FORCE-PERF-VERSION-FILE
+	@$(SHELL_PATH) util/PERF-VERSION-GEN
+-include 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')
+uname_O := $(shell sh -c 'uname -o 2>/dev/null || echo not')
+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')
+
+# CFLAGS and LDFLAGS are for the users to override from the command line.
+
+CFLAGS = -ggdb3 -Wall -Werror -Wstrict-prototypes -Wmissing-declarations -Wmissing-prototypes -std=gnu99 -Wdeclaration-after-statement -O6
+LDFLAGS = -lpthread -lrt -lelf
+ALL_CFLAGS = $(CFLAGS)
+ALL_LDFLAGS = $(LDFLAGS)
+STRIP ?= strip
+
+# Among the variables below, these:
+#   perfexecdir
+#   template_dir
+#   mandir
+#   infodir
+#   htmldir
+#   ETC_PERFCONFIG (but not sysconfdir)
+# can be specified as a relative path some/where/else;
+# this is interpreted as relative to $(prefix) and "perf" at
+# runtime figures out where they are based on the path to the executable.
+# This can help installing the suite in a relocatable way.
+
+prefix = $(HOME)
+bindir_relative = bin
+bindir = $(prefix)/$(bindir_relative)
+mandir = share/man
+infodir = share/info
+perfexecdir = libexec/perf-core
+sharedir = $(prefix)/share
+template_dir = share/perf-core/templates
+htmldir = share/doc/perf-doc
+ifeq ($(prefix),/usr)
+sysconfdir = /etc
+ETC_PERFCONFIG = $(sysconfdir)/perfconfig
+else
+sysconfdir = $(prefix)/etc
+ETC_PERFCONFIG = etc/perfconfig
+endif
+lib = lib
+# DESTDIR=
+
+export prefix bindir sharedir sysconfdir
+
+CC = gcc
+AR = ar
+RM = rm -f
+TAR = tar
+FIND = find
+INSTALL = install
+RPMBUILD = rpmbuild
+PTHREAD_LIBS = -lpthread
+
+# sparse is architecture-neutral, which means that we need to tell it
+# explicitly what architecture to check for. Fix this up for yours..
+SPARSE_FLAGS = -D__BIG_ENDIAN__ -D__powerpc__
+
+
+
+### --- END CONFIGURATION SECTION ---
+
+# 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_LDFLAGS =
+
+# Guard against environment variables
+BUILTIN_OBJS =
+BUILT_INS =
+COMPAT_CFLAGS =
+COMPAT_OBJS =
+LIB_H =
+LIB_OBJS =
+SCRIPT_PERL =
+SCRIPT_SH =
+TEST_PROGRAMS =
+
+#
+# No scripts right now:
+#
+
+# SCRIPT_SH += perf-am.sh
+
+#
+# No Perl scripts right now:
+#
+
+# SCRIPT_PERL += perf-add--interactive.perl
+
+SCRIPTS = $(patsubst %.sh,%,$(SCRIPT_SH)) \
+	  $(patsubst %.perl,%,$(SCRIPT_PERL))
+
+# Empty...
+EXTRA_PROGRAMS =
+
+# ... and all the rest that could be moved out of bindir to perfexecdir
+PROGRAMS += $(EXTRA_PROGRAMS)
+
+#
+# Single 'perf' binary right now:
+#
+PROGRAMS += 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
+
+# Set paths to tools early so that they can be used for version tests.
+ifndef SHELL_PATH
+	SHELL_PATH = /bin/sh
+endif
+ifndef PERL_PATH
+	PERL_PATH = /usr/bin/perl
+endif
+
+export PERL_PATH
+
+LIB_FILE=libperf.a
+
+LIB_H += ../../include/linux/perf_counter.h
+LIB_H += perf.h
+LIB_H += util/list.h
+LIB_H += util/rbtree.h
+LIB_H += util/levenshtein.h
+LIB_H += util/parse-options.h
+LIB_H += util/parse-events.h
+LIB_H += util/quote.h
+LIB_H += util/util.h
+LIB_H += util/help.h
+LIB_H += util/strbuf.h
+LIB_H += util/string.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
+
+PERFLIBS = $(LIB_FILE)
+EXTLIBS =
+
+#
+# Platform specific tweaks
+#
+
+# We choose to avoid "if .. else if .. else .. endif endif"
+# because maintaining the nesting to match is a pain.  If
+# we had "elif" things would have been much nicer...
+
+-include config.mak.autogen
+-include config.mak
+
+ifeq ($(uname_S),Darwin)
+	ifndef NO_FINK
+		ifeq ($(shell test -d /sw/lib && echo y),y)
+			BASIC_CFLAGS += -I/sw/include
+			BASIC_LDFLAGS += -L/sw/lib
+		endif
+	endif
+	ifndef NO_DARWIN_PORTS
+		ifeq ($(shell test -d /opt/local/lib && echo y),y)
+			BASIC_CFLAGS += -I/opt/local/include
+			BASIC_LDFLAGS += -L/opt/local/lib
+		endif
+	endif
+	PTHREAD_LIBS =
+endif
+
+ifndef CC_LD_DYNPATH
+	ifdef NO_R_TO_GCC_LINKER
+		# Some gcc does not accept and pass -R to the linker to specify
+		# the runtime dynamic library path.
+		CC_LD_DYNPATH = -Wl,-rpath,
+	else
+		CC_LD_DYNPATH = -R
+	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
+ifdef NEEDS_NSL
+	EXTLIBS += -lnsl
+endif
+ifdef NO_D_TYPE_IN_DIRENT
+	BASIC_CFLAGS += -DNO_D_TYPE_IN_DIRENT
+endif
+ifdef NO_D_INO_IN_DIRENT
+	BASIC_CFLAGS += -DNO_D_INO_IN_DIRENT
+endif
+ifdef NO_ST_BLOCKS_IN_STRUCT_STAT
+	BASIC_CFLAGS += -DNO_ST_BLOCKS_IN_STRUCT_STAT
+endif
+ifdef USE_NSEC
+	BASIC_CFLAGS += -DUSE_NSEC
+endif
+ifdef USE_ST_TIMESPEC
+	BASIC_CFLAGS += -DUSE_ST_TIMESPEC
+endif
+ifdef NO_NSEC
+	BASIC_CFLAGS += -DNO_NSEC
+endif
+ifdef NO_C99_FORMAT
+	BASIC_CFLAGS += -DNO_C99_FORMAT
+endif
+ifdef SNPRINTF_RETURNS_BOGUS
+	COMPAT_CFLAGS += -DSNPRINTF_RETURNS_BOGUS
+	COMPAT_OBJS += compat/snprintf.o
+endif
+ifdef FREAD_READS_DIRECTORIES
+	COMPAT_CFLAGS += -DFREAD_READS_DIRECTORIES
+	COMPAT_OBJS += 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
+endif
+ifdef NO_STRTOUMAX
+	COMPAT_CFLAGS += -DNO_STRTOUMAX
+	COMPAT_OBJS += compat/strtoumax.o
+endif
+ifdef NO_STRTOULL
+	COMPAT_CFLAGS += -DNO_STRTOULL
+endif
+ifdef NO_SETENV
+	COMPAT_CFLAGS += -DNO_SETENV
+	COMPAT_OBJS += compat/setenv.o
+endif
+ifdef NO_MKDTEMP
+	COMPAT_CFLAGS += -DNO_MKDTEMP
+	COMPAT_OBJS += compat/mkdtemp.o
+endif
+ifdef NO_UNSETENV
+	COMPAT_CFLAGS += -DNO_UNSETENV
+	COMPAT_OBJS += 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
+else
+	ifdef USE_WIN32_MMAP
+		COMPAT_CFLAGS += -DUSE_WIN32_MMAP
+		COMPAT_OBJS += compat/win32mmap.o
+	endif
+endif
+ifdef NO_PREAD
+	COMPAT_CFLAGS += -DNO_PREAD
+	COMPAT_OBJS += compat/pread.o
+endif
+ifdef NO_FAST_WORKING_DIRECTORY
+	BASIC_CFLAGS += -DNO_FAST_WORKING_DIRECTORY
+endif
+ifdef NO_TRUSTABLE_FILEMODE
+	BASIC_CFLAGS += -DNO_TRUSTABLE_FILEMODE
+endif
+ifdef NO_IPV6
+	BASIC_CFLAGS += -DNO_IPV6
+endif
+ifdef NO_UINTMAX_T
+	BASIC_CFLAGS += -Duintmax_t=uint32_t
+endif
+ifdef NO_SOCKADDR_STORAGE
+ifdef NO_IPV6
+	BASIC_CFLAGS += -Dsockaddr_storage=sockaddr_in
+else
+	BASIC_CFLAGS += -Dsockaddr_storage=sockaddr_in6
+endif
+endif
+ifdef NO_INET_NTOP
+	LIB_OBJS += compat/inet_ntop.o
+endif
+ifdef NO_INET_PTON
+	LIB_OBJS += compat/inet_pton.o
+endif
+
+ifdef NO_ICONV
+	BASIC_CFLAGS += -DNO_ICONV
+endif
+
+ifdef OLD_ICONV
+	BASIC_CFLAGS += -DOLD_ICONV
+endif
+
+ifdef NO_DEFLATE_BOUND
+	BASIC_CFLAGS += -DNO_DEFLATE_BOUND
+endif
+
+ifdef PPC_SHA1
+	SHA1_HEADER = "ppc/sha1.h"
+	LIB_OBJS += ppc/sha1.o ppc/sha1ppc.o
+else
+ifdef ARM_SHA1
+	SHA1_HEADER = "arm/sha1.h"
+	LIB_OBJS += arm/sha1.o arm/sha1_arm.o
+else
+ifdef MOZILLA_SHA1
+	SHA1_HEADER = "mozilla-sha1/sha1.h"
+	LIB_OBJS += mozilla-sha1/sha1.o
+else
+	SHA1_HEADER = <openssl/sha.h>
+	EXTLIBS += $(LIB_4_CRYPTO)
+endif
+endif
+endif
+ifdef NO_PERL_MAKEMAKER
+	export NO_PERL_MAKEMAKER
+endif
+ifdef NO_HSTRERROR
+	COMPAT_CFLAGS += -DNO_HSTRERROR
+	COMPAT_OBJS += compat/hstrerror.o
+endif
+ifdef NO_MEMMEM
+	COMPAT_CFLAGS += -DNO_MEMMEM
+	COMPAT_OBJS += compat/memmem.o
+endif
+ifdef INTERNAL_QSORT
+	COMPAT_CFLAGS += -DINTERNAL_QSORT
+	COMPAT_OBJS += compat/qsort.o
+endif
+ifdef RUNTIME_PREFIX
+	COMPAT_CFLAGS += -DRUNTIME_PREFIX
+endif
+
+ifdef DIR_HAS_BSD_GROUP_SEMANTICS
+	COMPAT_CFLAGS += -DDIR_HAS_BSD_GROUP_SEMANTICS
+endif
+ifdef NO_EXTERNAL_GREP
+	BASIC_CFLAGS += -DNO_EXTERNAL_GREP
+endif
+
+ifeq ($(PERL_PATH),)
+NO_PERL=NoThanks
+endif
+
+QUIET_SUBDIR0  = +$(MAKE) -C # space to separate -C and subdir
+QUIET_SUBDIR1  =
+
+ifneq ($(findstring $(MAKEFLAGS),w),w)
+PRINT_DIR = --no-print-directory
+else # "make -w"
+NO_SUBDIR = :
+endif
+
+ifneq ($(findstring $(MAKEFLAGS),s),s)
+ifndef V
+	QUIET_CC       = @echo '   ' CC $@;
+	QUIET_AR       = @echo '   ' AR $@;
+	QUIET_LINK     = @echo '   ' LINK $@;
+	QUIET_BUILT_IN = @echo '   ' BUILTIN $@;
+	QUIET_GEN      = @echo '   ' GEN $@;
+	QUIET_SUBDIR0  = +@subdir=
+	QUIET_SUBDIR1  = ;$(NO_SUBDIR) echo '   ' SUBDIR $$subdir; \
+			 $(MAKE) $(PRINT_DIR) -C $$subdir
+	export V
+	export QUIET_GEN
+	export QUIET_BUILT_IN
+endif
+endif
+
+ifdef ASCIIDOC8
+	export ASCIIDOC8
+endif
+
+# Shell quote (do not use $(call) to accommodate ancient setups);
+
+SHA1_HEADER_SQ = $(subst ','\'',$(SHA1_HEADER))
+ETC_PERFCONFIG_SQ = $(subst ','\'',$(ETC_PERFCONFIG))
+
+DESTDIR_SQ = $(subst ','\'',$(DESTDIR))
+bindir_SQ = $(subst ','\'',$(bindir))
+bindir_relative_SQ = $(subst ','\'',$(bindir_relative))
+mandir_SQ = $(subst ','\'',$(mandir))
+infodir_SQ = $(subst ','\'',$(infodir))
+perfexecdir_SQ = $(subst ','\'',$(perfexecdir))
+template_dir_SQ = $(subst ','\'',$(template_dir))
+htmldir_SQ = $(subst ','\'',$(htmldir))
+prefix_SQ = $(subst ','\'',$(prefix))
+
+SHELL_PATH_SQ = $(subst ','\'',$(SHELL_PATH))
+PERL_PATH_SQ = $(subst ','\'',$(PERL_PATH))
+
+LIBS = $(PERFLIBS) $(EXTLIBS)
+
+BASIC_CFLAGS += -DSHA1_HEADER='$(SHA1_HEADER_SQ)' \
+	$(COMPAT_CFLAGS)
+LIB_OBJS += $(COMPAT_OBJS)
+
+ALL_CFLAGS += $(BASIC_CFLAGS)
+ALL_LDFLAGS += $(BASIC_LDFLAGS)
+
+export TAR INSTALL DESTDIR SHELL_PATH
+
+
+### Build rules
+
+SHELL = $(SHELL_PATH)
+
+all:: shell_compatibility_test $(ALL_PROGRAMS) $(BUILT_INS) $(OTHER_PROGRAMS) PERF-BUILD-OPTIONS
+ifneq (,$X)
+	$(foreach p,$(patsubst %$X,%,$(filter %$X,$(ALL_PROGRAMS) $(BUILT_INS) perf$X)), test '$p' -ef '$p$X' || $(RM) '$p';)
+endif
+
+all::
+
+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
+
+perf.o: perf.c common-cmds.h PERF-CFLAGS
+	$(QUIET_CC)$(CC) -DPERF_VERSION='"$(PERF_VERSION)"' \
+		'-DPERF_HTML_PATH="$(htmldir_SQ)"' \
+		$(ALL_CFLAGS) -c $(filter %.c,$^)
+
+perf$X: perf.o $(BUILTIN_OBJS) $(PERFLIBS)
+	$(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ perf.o \
+		$(BUILTIN_OBJS) $(ALL_LDFLAGS) $(LIBS)
+
+builtin-help.o: builtin-help.c common-cmds.h PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) \
+		'-DPERF_HTML_PATH="$(htmldir_SQ)"' \
+		'-DPERF_MAN_PATH="$(mandir_SQ)"' \
+		'-DPERF_INFO_PATH="$(infodir_SQ)"' $<
+
+$(BUILT_INS): perf$X
+	$(QUIET_BUILT_IN)$(RM) $@ && \
+	ln perf$X $@ 2>/dev/null || \
+	ln -s perf$X $@ 2>/dev/null || \
+	cp perf$X $@
+
+common-cmds.h: util/generate-cmdlist.sh command-list.txt
+
+common-cmds.h: $(wildcard Documentation/perf-*.txt)
+	$(QUIET_GEN)util/generate-cmdlist.sh > $@+ && mv $@+ $@
+
+$(patsubst %.sh,%,$(SCRIPT_SH)) : % : %.sh
+	$(QUIET_GEN)$(RM) $@ $@+ && \
+	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 $@+ $@
+
+configure: configure.ac
+	$(QUIET_GEN)$(RM) $@ $<+ && \
+	sed -e 's/@@PERF_VERSION@@/$(PERF_VERSION)/g' \
+	    $< > $<+ && \
+	autoconf -o $@ $<+ && \
+	$(RM) $<+
+
+# These can record PERF_VERSION
+perf.o perf.spec \
+	$(patsubst %.sh,%,$(SCRIPT_SH)) \
+	$(patsubst %.perl,%,$(SCRIPT_PERL)) \
+	: PERF-VERSION-FILE
+
+%.o: %.c PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) $<
+%.s: %.c PERF-CFLAGS
+	$(QUIET_CC)$(CC) -S $(ALL_CFLAGS) $<
+%.o: %.S
+	$(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) $<
+
+util/exec_cmd.o: util/exec_cmd.c PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $*.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)"' $<
+
+util/config.o: util/config.c PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $<
+
+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
+
+$(LIB_FILE): $(LIB_OBJS)
+	$(QUIET_AR)$(RM) $@ && $(AR) rcs $@ $(LIB_OBJS)
+
+doc:
+	$(MAKE) -C Documentation all
+
+man:
+	$(MAKE) -C Documentation man
+
+html:
+	$(MAKE) -C Documentation html
+
+info:
+	$(MAKE) -C Documentation info
+
+pdf:
+	$(MAKE) -C Documentation pdf
+
+TAGS:
+	$(RM) TAGS
+	$(FIND) . -name '*.[hcS]' -print | xargs etags -a
+
+tags:
+	$(RM) tags
+	$(FIND) . -name '*.[hcS]' -print | xargs ctags -a
+
+cscope:
+	$(RM) cscope*
+	$(FIND) . -name '*.[hcS]' -print | xargs cscope -b
+
+### Detect prefix changes
+TRACK_CFLAGS = $(subst ','\'',$(ALL_CFLAGS)):\
+             $(bindir_SQ):$(perfexecdir_SQ):$(template_dir_SQ):$(prefix_SQ)
+
+PERF-CFLAGS: .FORCE-PERF-CFLAGS
+	@FLAGS='$(TRACK_CFLAGS)'; \
+	    if test x"$$FLAGS" != x"`cat PERF-CFLAGS 2>/dev/null`" ; then \
+		echo 1>&2 "    * new build flags or prefix"; \
+		echo "$$FLAGS" >PERF-CFLAGS; \
+            fi
+
+# We need to apply sq twice, once to protect from the shell
+# that runs PERF-BUILD-OPTIONS, and then again to protect it
+# and the first level quoting from the shell that runs "echo".
+PERF-BUILD-OPTIONS: .FORCE-PERF-BUILD-OPTIONS
+	@echo SHELL_PATH=\''$(subst ','\'',$(SHELL_PATH_SQ))'\' >$@
+	@echo TAR=\''$(subst ','\'',$(subst ','\'',$(TAR)))'\' >>$@
+	@echo NO_CURL=\''$(subst ','\'',$(subst ','\'',$(NO_CURL)))'\' >>$@
+	@echo NO_PERL=\''$(subst ','\'',$(subst ','\'',$(NO_PERL)))'\' >>$@
+
+### Testing rules
+
+#
+# None right now:
+#
+# TEST_PROGRAMS += test-something$X
+
+all:: $(TEST_PROGRAMS)
+
+# GNU make supports exporting all variables by "export" without parameters.
+# However, the environment gets quite big, and some programs have problems
+# with that.
+
+export NO_SVN_TESTS
+
+check: common-cmds.h
+	if sparse; \
+	then \
+		for i in *.c */*.c; \
+		do \
+			sparse $(ALL_CFLAGS) $(SPARSE_FLAGS) $$i || exit; \
+		done; \
+	else \
+		echo 2>&1 "Did you mean 'make test'?"; \
+		exit 1; \
+	fi
+
+remove-dashes:
+	./fixup-builtins $(BUILT_INS) $(PROGRAMS) $(SCRIPTS)
+
+### Installation rules
+
+ifneq ($(filter /%,$(firstword $(template_dir))),)
+template_instdir = $(template_dir)
+else
+template_instdir = $(prefix)/$(template_dir)
+endif
+export template_instdir
+
+ifneq ($(filter /%,$(firstword $(perfexecdir))),)
+perfexec_instdir = $(perfexecdir)
+else
+perfexec_instdir = $(prefix)/$(perfexecdir)
+endif
+perfexec_instdir_SQ = $(subst ','\'',$(perfexec_instdir))
+export perfexec_instdir
+
+install: all
+	$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(bindir_SQ)'
+	$(INSTALL) perf$X '$(DESTDIR_SQ)$(bindir_SQ)'
+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';)
+endif
+endif
+
+install-doc:
+	$(MAKE) -C Documentation install
+
+install-man:
+	$(MAKE) -C Documentation install-man
+
+install-html:
+	$(MAKE) -C Documentation install-html
+
+install-info:
+	$(MAKE) -C Documentation install-info
+
+install-pdf:
+	$(MAKE) -C Documentation install-pdf
+
+quick-install-doc:
+	$(MAKE) -C Documentation quick-install
+
+quick-install-man:
+	$(MAKE) -C Documentation quick-install-man
+
+quick-install-html:
+	$(MAKE) -C Documentation quick-install-html
+
+
+### Maintainer's dist rules
+#
+# None right now
+#
+#
+# perf.spec: perf.spec.in
+#	sed -e 's/@@VERSION@@/$(PERF_VERSION)/g' < $< > $@+
+#	mv $@+ $@
+#
+# PERF_TARNAME=perf-$(PERF_VERSION)
+# dist: perf.spec perf-archive$(X) configure
+#	./perf-archive --format=tar \
+#		--prefix=$(PERF_TARNAME)/ HEAD^{tree} > $(PERF_TARNAME).tar
+#	@mkdir -p $(PERF_TARNAME)
+#	@cp perf.spec configure $(PERF_TARNAME)
+#	@echo $(PERF_VERSION) > $(PERF_TARNAME)/version
+#	$(TAR) rf $(PERF_TARNAME).tar \
+#		$(PERF_TARNAME)/perf.spec \
+#		$(PERF_TARNAME)/configure \
+#		$(PERF_TARNAME)/version
+#	@$(RM) -r $(PERF_TARNAME)
+#	gzip -f -9 $(PERF_TARNAME).tar
+#
+# htmldocs = perf-htmldocs-$(PERF_VERSION)
+# manpages = perf-manpages-$(PERF_VERSION)
+# dist-doc:
+#	$(RM) -r .doc-tmp-dir
+#	mkdir .doc-tmp-dir
+#	$(MAKE) -C Documentation WEBDOC_DEST=../.doc-tmp-dir install-webdoc
+#	cd .doc-tmp-dir && $(TAR) cf ../$(htmldocs).tar .
+#	gzip -n -9 -f $(htmldocs).tar
+#	:
+#	$(RM) -r .doc-tmp-dir
+#	mkdir -p .doc-tmp-dir/man1 .doc-tmp-dir/man5 .doc-tmp-dir/man7
+#	$(MAKE) -C Documentation DESTDIR=./ \
+#		man1dir=../.doc-tmp-dir/man1 \
+#		man5dir=../.doc-tmp-dir/man5 \
+#		man7dir=../.doc-tmp-dir/man7 \
+#		install
+#	cd .doc-tmp-dir && $(TAR) cf ../$(manpages).tar .
+#	gzip -n -9 -f $(manpages).tar
+#	$(RM) -r .doc-tmp-dir
+#
+# rpm: dist
+#	$(RPMBUILD) -ta $(PERF_TARNAME).tar.gz
+
+### Cleaning rules
+
+distclean: clean
+#	$(RM) configure
+
+clean:
+	$(RM) *.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) -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
+
+.PHONY: all install clean strip
+.PHONY: shell_compatibility_test please_set_SHELL_PATH_to_a_more_modern_shell
+.PHONY: .FORCE-PERF-VERSION-FILE TAGS tags cscope .FORCE-PERF-CFLAGS
+.PHONY: .FORCE-PERF-BUILD-OPTIONS
+
+### Make sure built-ins do not have dups and listed in perf.c
+#
+check-builtins::
+	./check-builtins.sh
+
+### Test suite coverage testing
+#
+# None right now
+#
+# .PHONY: coverage coverage-clean coverage-build coverage-report
+#
+# coverage:
+#	$(MAKE) coverage-build
+#	$(MAKE) coverage-report
+#
+# coverage-clean:
+#	rm -f *.gcda *.gcno
+#
+# COVERAGE_CFLAGS = $(CFLAGS) -O0 -ftest-coverage -fprofile-arcs
+# COVERAGE_LDFLAGS = $(CFLAGS)  -O0 -lgcov
+#
+# coverage-build: coverage-clean
+#	$(MAKE) CFLAGS="$(COVERAGE_CFLAGS)" LDFLAGS="$(COVERAGE_LDFLAGS)" all
+#	$(MAKE) CFLAGS="$(COVERAGE_CFLAGS)" LDFLAGS="$(COVERAGE_LDFLAGS)" \
+#		-j1 test
+#
+# coverage-report:
+#	gcov -b *.c */*.c
+#	grep '^function.*called 0 ' *.c.gcov */*.c.gcov \
+#		| sed -e 's/\([^:]*\)\.gcov: *function \([^ ]*\) called.*/\1: \2/' \
+#		| tee coverage-untested-functions
diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c
new file mode 100644
index 000000000000..b1ed5f766cb3
--- /dev/null
+++ b/tools/perf/builtin-annotate.c
@@ -0,0 +1,1356 @@
+/*
+ * builtin-annotate.c
+ *
+ * Builtin annotate command: Analyze the perf.data input file,
+ * look up and read DSOs and symbol information and display
+ * a histogram of results, along various sorting keys.
+ */
+#include "builtin.h"
+
+#include "util/util.h"
+
+#include "util/color.h"
+#include "util/list.h"
+#include "util/cache.h"
+#include "util/rbtree.h"
+#include "util/symbol.h"
+#include "util/string.h"
+
+#include "perf.h"
+
+#include "util/parse-options.h"
+#include "util/parse-events.h"
+
+#define SHOW_KERNEL	1
+#define SHOW_USER	2
+#define SHOW_HV		4
+
+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 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;
+
+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 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;
+
+	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;
+	__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 int64_t
+sort__comm_collapse(struct hist_entry *left, struct hist_entry *right)
+{
+	char *comm_l = left->thread->comm;
+	char *comm_r = right->thread->comm;
+
+	if (!comm_l || !comm_r) {
+		if (!comm_l && !comm_r)
+			return 0;
+		else if (!comm_l)
+			return -1;
+		else
+			return 1;
+	}
+
+	return strcmp(comm_l, comm_r);
+}
+
+static size_t
+sort__comm_print(FILE *fp, struct hist_entry *self)
+{
+	return fprintf(fp, "%16s", self->thread->comm);
+}
+
+static struct sort_entry sort_comm = {
+	.header		= "         Command",
+	.cmp		= sort__comm_cmp,
+	.collapse	= sort__comm_collapse,
+	.print		= sort__comm_print,
+};
+
+/* --sort dso */
+
+static int64_t
+sort__dso_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	struct dso *dso_l = left->dso;
+	struct dso *dso_r = right->dso;
+
+	if (!dso_l || !dso_r) {
+		if (!dso_l && !dso_r)
+			return 0;
+		else if (!dso_l)
+			return -1;
+		else
+			return 1;
+	}
+
+	return strcmp(dso_l->name, dso_r->name);
+}
+
+static size_t
+sort__dso_print(FILE *fp, struct hist_entry *self)
+{
+	if (self->dso)
+		return fprintf(fp, "%-25s", self->dso->name);
+
+	return fprintf(fp, "%016llx         ", (__u64)self->ip);
+}
+
+static struct sort_entry sort_dso = {
+	.header = "Shared Object            ",
+	.cmp	= sort__dso_cmp,
+	.print	= sort__dso_print,
+};
+
+/* --sort symbol */
+
+static int64_t
+sort__sym_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	__u64 ip_l, ip_r;
+
+	if (left->sym == right->sym)
+		return 0;
+
+	ip_l = left->sym ? left->sym->start : left->ip;
+	ip_r = right->sym ? right->sym->start : right->ip;
+
+	return (int64_t)(ip_r - ip_l);
+}
+
+static size_t
+sort__sym_print(FILE *fp, struct hist_entry *self)
+{
+	size_t ret = 0;
+
+	if (verbose)
+		ret += fprintf(fp, "%#018llx  ", (__u64)self->ip);
+
+	if (self->sym) {
+		ret += fprintf(fp, "[%c] %s",
+			self->dso == kernel_dso ? 'k' : '.', self->sym->name);
+	} else {
+		ret += fprintf(fp, "%#016llx", (__u64)self->ip);
+	}
+
+	return ret;
+}
+
+static struct sort_entry sort_sym = {
+	.header = "Symbol",
+	.cmp	= sort__sym_cmp,
+	.print	= sort__sym_print,
+};
+
+static int sort__need_collapse = 0;
+
+struct sort_dimension {
+	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 (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++;
+
+	return 0;
+}
+
+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);
+
+	dprintf("%p [%p]: PERF_EVENT_FORK: %d:%d\n",
+		(void *)(offset + head),
+		(void *)(long)(event->header.size),
+		event->fork.pid, event->fork.ppid);
+
+	if (!thread || !parent || thread__fork(thread, parent)) {
+		dprintf("problem processing PERF_EVENT_FORK, skipping event.\n");
+		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:
+		return -1;
+	}
+
+	return 0;
+}
+
+static int
+parse_line(FILE *file, struct symbol *sym, __u64 start, __u64 len)
+{
+	char *line = NULL, *tmp, *tmp2;
+	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) {
+		unsigned int hits = 0;
+		double percent = 0.0;
+		char *color = PERF_COLOR_NORMAL;
+
+		offset = line_ip - start;
+		if (offset < len)
+			hits = sym->hist[offset];
+
+		if (sym->hist_sum)
+			percent = 100.0 * hits / sym->hist_sum;
+
+		/*
+		 * 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;
+		}
+
+		color_fprintf(stdout, color, " %7.2f", percent);
+		printf(" :	");
+		color_fprintf(stdout, PERF_COLOR_BLUE, "%s\n", line);
+	} else {
+		if (!*line)
+			printf("         :\n");
+		else
+			printf("         :	%s\n", line);
+	}
+
+	return 0;
+}
+
+static void annotate_sym(struct dso *dso, struct symbol *sym)
+{
+	char *filename = dso->name;
+	__u64 start, end, len;
+	char command[PATH_MAX*2];
+	FILE *file;
+
+	if (!filename)
+		return;
+	if (dso == kernel_dso)
+		filename = vmlinux;
+
+	printf("\n------------------------------------------------\n");
+	printf(" Percent |	Source code & Disassembly of %s\n", filename);
+	printf("------------------------------------------------\n");
+
+	if (verbose >= 2)
+		printf("annotating [%p] %30s : [%p] %30s\n", dso, dso->name, sym, sym->name);
+
+	start = sym->obj_start;
+	if (!start)
+		start = sym->start;
+
+	end = start + sym->end - sym->start + 1;
+	len = sym->end - sym->start;
+
+	sprintf(command, "objdump --start-address=0x%016Lx --stop-address=0x%016Lx -dS %s", (__u64)start, (__u64)end, filename);
+
+	if (verbose >= 3)
+		printf("doing: %s\n", command);
+
+	file = popen(command, "r");
+	if (!file)
+		return;
+
+	while (!feof(file)) {
+		if (parse_line(file, sym, start, len) < 0)
+			break;
+	}
+
+	pclose(file);
+}
+
+static void find_annotations(void)
+{
+	struct rb_node *nd;
+	struct dso *dso;
+	int count = 0;
+
+	list_for_each_entry(dso, &dsos, node) {
+
+		for (nd = rb_first(&dso->syms); nd; nd = rb_next(nd)) {
+			struct symbol *sym = rb_entry(nd, struct symbol, rb_node);
+
+			if (sym->hist) {
+				annotate_sym(dso, sym);
+				count++;
+			}
+		}
+	}
+
+	if (!count)
+		printf(" Error: symbol '%s' not present amongst the samples.\n", sym_hist_filter);
+}
+
+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++;
+
+		/*
+		 * assume we lost track of the stream, check alignment, and
+		 * increment a single u64 in the hope to catch on again 'soon'.
+		 */
+
+		if (unlikely(head & 7))
+			head &= ~7ULL;
+
+		size = 8;
+	}
+
+	head += size;
+
+	if (offset + head < 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 >= 2)
+		dsos__fprintf(stdout);
+
+	collapse__resort();
+	output__resort();
+
+	find_annotations();
+
+	return rc;
+}
+
+static const char * const annotate_usage[] = {
+	"perf annotate [<options>] <command>",
+	NULL
+};
+
+static const struct option options[] = {
+	OPT_STRING('i', "input", &input_name, "file",
+		    "input file name"),
+	OPT_STRING('s', "symbol", &sym_hist_filter, "symbol",
+		    "symbol to annotate"),
+	OPT_BOOLEAN('v', "verbose", &verbose,
+		    "be more verbose (show symbol address, etc)"),
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+		    "dump raw trace in ASCII"),
+	OPT_STRING('k', "vmlinux", &vmlinux, "file", "vmlinux pathname"),
+	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) < 0) {
+			error("Unknown --sort key: `%s'", tok);
+			usage_with_options(annotate_usage, options);
+		}
+	}
+
+	free(str);
+}
+
+int cmd_annotate(int argc, const char **argv, const char *prefix)
+{
+	symbol__init();
+
+	page_size = getpagesize();
+
+	argc = parse_options(argc, argv, options, annotate_usage, 0);
+
+	setup_sorting();
+
+	if (argc) {
+		/*
+		 * Special case: if there's an argument left then assume tha
+		 * it's a symbol filter:
+		 */
+		if (argc > 1)
+			usage_with_options(annotate_usage, options);
+
+		sym_hist_filter = argv[0];
+	}
+
+	if (!sym_hist_filter)
+		usage_with_options(annotate_usage, options);
+
+	setup_pager();
+
+	return __cmd_annotate();
+}
diff --git a/tools/perf/builtin-help.c b/tools/perf/builtin-help.c
new file mode 100644
index 000000000000..0f32dc3f3c4c
--- /dev/null
+++ b/tools/perf/builtin-help.c
@@ -0,0 +1,461 @@
+/*
+ * builtin-help.c
+ *
+ * Builtin help command
+ */
+#include "util/cache.h"
+#include "builtin.h"
+#include "util/exec_cmd.h"
+#include "common-cmds.h"
+#include "util/parse-options.h"
+#include "util/run-command.h"
+#include "util/help.h"
+
+static struct man_viewer_list {
+	struct man_viewer_list *next;
+	char name[FLEX_ARRAY];
+} *man_viewer_list;
+
+static struct man_viewer_info_list {
+	struct man_viewer_info_list *next;
+	const char *info;
+	char name[FLEX_ARRAY];
+} *man_viewer_info_list;
+
+enum help_format {
+	HELP_FORMAT_MAN,
+	HELP_FORMAT_INFO,
+	HELP_FORMAT_WEB,
+};
+
+static int show_all = 0;
+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",
+			HELP_FORMAT_WEB),
+	OPT_SET_INT('i', "info", &help_format, "show info page",
+			HELP_FORMAT_INFO),
+	OPT_END(),
+};
+
+static const char * const builtin_help_usage[] = {
+	"perf help [--all] [--man|--web|--info] [command]",
+	NULL
+};
+
+static enum help_format parse_help_format(const char *format)
+{
+	if (!strcmp(format, "man"))
+		return HELP_FORMAT_MAN;
+	if (!strcmp(format, "info"))
+		return HELP_FORMAT_INFO;
+	if (!strcmp(format, "web") || !strcmp(format, "html"))
+		return HELP_FORMAT_WEB;
+	die("unrecognized help format '%s'", format);
+}
+
+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)
+	{
+		if (!strcasecmp(name, viewer->name))
+			return viewer->info;
+	}
+	return NULL;
+}
+
+static int check_emacsclient_version(void)
+{
+	struct strbuf buffer = STRBUF_INIT;
+	struct child_process ec_process;
+	const char *argv_ec[] = { "emacsclient", "--version", NULL };
+	int version;
+
+	/* emacsclient prints its version number on stderr */
+	memset(&ec_process, 0, sizeof(ec_process));
+	ec_process.argv = argv_ec;
+	ec_process.err = -1;
+	ec_process.stdout_to_stderr = 1;
+	if (start_command(&ec_process)) {
+		fprintf(stderr, "Failed to start emacsclient.\n");
+		return -1;
+	}
+	strbuf_read(&buffer, ec_process.err, 20);
+	close(ec_process.err);
+
+	/*
+	 * Don't bother checking return value, because "emacsclient --version"
+	 * seems to always exits with code 1.
+	 */
+	finish_command(&ec_process);
+
+	if (prefixcmp(buffer.buf, "emacsclient")) {
+		fprintf(stderr, "Failed to parse emacsclient version.\n");
+		strbuf_release(&buffer);
+		return -1;
+	}
+
+	strbuf_remove(&buffer, 0, strlen("emacsclient"));
+	version = atoi(buffer.buf);
+
+	if (version < 22) {
+		fprintf(stderr,
+			"emacsclient version '%d' too old (< 22).\n",
+			version);
+		strbuf_release(&buffer);
+		return -1;
+	}
+
+	strbuf_release(&buffer);
+	return 0;
+}
+
+static void exec_woman_emacs(const char* path, const char *page)
+{
+	if (!check_emacsclient_version()) {
+		/* This works only with emacsclient version >= 22. */
+		struct strbuf man_page = STRBUF_INIT;
+
+		if (!path)
+			path = "emacsclient";
+		strbuf_addf(&man_page, "(woman \"%s\")", page);
+		execlp(path, "emacsclient", "-e", man_page.buf, NULL);
+		warning("failed to exec '%s': %s", path, strerror(errno));
+	}
+}
+
+static void exec_man_konqueror(const char* path, const char *page)
+{
+	const char *display = getenv("DISPLAY");
+	if (display && *display) {
+		struct strbuf man_page = STRBUF_INIT;
+		const char *filename = "kfmclient";
+
+		/* It's simpler to launch konqueror using kfmclient. */
+		if (path) {
+			const char *file = strrchr(path, '/');
+			if (file && !strcmp(file + 1, "konqueror")) {
+				char *new = strdup(path);
+				char *dest = strrchr(new, '/');
+
+				/* strlen("konqueror") == strlen("kfmclient") */
+				strcpy(dest + 1, "kfmclient");
+				path = new;
+			}
+			if (file)
+				filename = file;
+		} else
+			path = "kfmclient";
+		strbuf_addf(&man_page, "man:%s(1)", page);
+		execlp(path, filename, "newTab", man_page.buf, NULL);
+		warning("failed to exec '%s': %s", path, strerror(errno));
+	}
+}
+
+static void exec_man_man(const char* path, const char *page)
+{
+	if (!path)
+		path = "man";
+	execlp(path, "man", page, NULL);
+	warning("failed to exec '%s': %s", path, strerror(errno));
+}
+
+static void exec_man_cmd(const char *cmd, const char *page)
+{
+	struct strbuf shell_cmd = STRBUF_INIT;
+	strbuf_addf(&shell_cmd, "%s %s", cmd, page);
+	execl("/bin/sh", "sh", "-c", shell_cmd.buf, NULL);
+	warning("failed to exec '%s': %s", cmd, strerror(errno));
+}
+
+static void add_man_viewer(const char *name)
+{
+	struct man_viewer_list **p = &man_viewer_list;
+	size_t len = strlen(name);
+
+	while (*p)
+		p = &((*p)->next);
+	*p = calloc(1, (sizeof(**p) + len + 1));
+	strncpy((*p)->name, name, len);
+}
+
+static int supported_man_viewer(const char *name, size_t len)
+{
+	return (!strncasecmp("man", name, len) ||
+		!strncasecmp("woman", name, len) ||
+		!strncasecmp("konqueror", name, len));
+}
+
+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);
+
+	strncpy(new->name, name, len);
+	new->info = strdup(value);
+	new->next = man_viewer_info_list;
+	man_viewer_info_list = new;
+}
+
+static int add_man_viewer_path(const char *name,
+			       size_t len,
+			       const char *value)
+{
+	if (supported_man_viewer(name, len))
+		do_add_man_viewer_info(name, len, value);
+	else
+		warning("'%s': path for unsupported man viewer.\n"
+			"Please consider using 'man.<tool>.cmd' instead.",
+			name);
+
+	return 0;
+}
+
+static int add_man_viewer_cmd(const char *name,
+			      size_t len,
+			      const char *value)
+{
+	if (supported_man_viewer(name, len))
+		warning("'%s': cmd for supported man viewer.\n"
+			"Please consider using 'man.<tool>.path' instead.",
+			name);
+	else
+		do_add_man_viewer_info(name, len, value);
+
+	return 0;
+}
+
+static int add_man_viewer_info(const char *var, const char *value)
+{
+	const char *name = var + 4;
+	const char *subkey = strrchr(name, '.');
+
+	if (!subkey)
+		return error("Config with no key for man viewer: %s", name);
+
+	if (!strcmp(subkey, ".path")) {
+		if (!value)
+			return config_error_nonbool(var);
+		return add_man_viewer_path(name, subkey - name, value);
+	}
+	if (!strcmp(subkey, ".cmd")) {
+		if (!value)
+			return config_error_nonbool(var);
+		return add_man_viewer_cmd(name, subkey - name, value);
+	}
+
+	warning("'%s': unsupported man viewer sub key.", subkey);
+	return 0;
+}
+
+static int perf_help_config(const char *var, const char *value, void *cb)
+{
+	if (!strcmp(var, "help.format")) {
+		if (!value)
+			return config_error_nonbool(var);
+		help_format = parse_help_format(value);
+		return 0;
+	}
+	if (!strcmp(var, "man.viewer")) {
+		if (!value)
+			return config_error_nonbool(var);
+		add_man_viewer(value);
+		return 0;
+	}
+	if (!prefixcmp(var, "man."))
+		return add_man_viewer_info(var, value);
+
+	return perf_default_config(var, value, cb);
+}
+
+static struct cmdnames main_cmds, other_cmds;
+
+void list_common_cmds_help(void)
+{
+	int i, longest = 0;
+
+	for (i = 0; i < ARRAY_SIZE(common_cmds); i++) {
+		if (longest < strlen(common_cmds[i].name))
+			longest = strlen(common_cmds[i].name);
+	}
+
+	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));
+		puts(common_cmds[i].help);
+	}
+}
+
+static int is_perf_command(const char *s)
+{
+	return is_in_cmdlist(&main_cmds, s) ||
+		is_in_cmdlist(&other_cmds, s);
+}
+
+static const char *prepend(const char *prefix, const char *cmd)
+{
+	size_t pre_len = strlen(prefix);
+	size_t cmd_len = strlen(cmd);
+	char *p = malloc(pre_len + cmd_len + 1);
+	memcpy(p, prefix, pre_len);
+	strcpy(p + pre_len, cmd);
+	return p;
+}
+
+static const char *cmd_to_page(const char *perf_cmd)
+{
+	if (!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);
+}
+
+static void setup_man_path(void)
+{
+	struct strbuf new_path = STRBUF_INIT;
+	const char *old_path = getenv("MANPATH");
+
+	/* We should always put ':' after our path. If there is no
+	 * old_path, the ':' at the end will let 'man' to try
+	 * system-wide paths after ours to find the manual page. If
+	 * there is old_path, we need ':' as delimiter. */
+	strbuf_addstr(&new_path, system_path(PERF_MAN_PATH));
+	strbuf_addch(&new_path, ':');
+	if (old_path)
+		strbuf_addstr(&new_path, old_path);
+
+	setenv("MANPATH", new_path.buf, 1);
+
+	strbuf_release(&new_path);
+}
+
+static void exec_viewer(const char *name, const char *page)
+{
+	const char *info = get_man_viewer_info(name);
+
+	if (!strcasecmp(name, "man"))
+		exec_man_man(info, page);
+	else if (!strcasecmp(name, "woman"))
+		exec_woman_emacs(info, page);
+	else if (!strcasecmp(name, "konqueror"))
+		exec_man_konqueror(info, page);
+	else if (info)
+		exec_man_cmd(info, page);
+	else
+		warning("'%s': unknown man viewer.", name);
+}
+
+static void show_man_page(const char *perf_cmd)
+{
+	struct man_viewer_list *viewer;
+	const char *page = cmd_to_page(perf_cmd);
+	const char *fallback = getenv("PERF_MAN_VIEWER");
+
+	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);
+	die("no man viewer handled the request");
+}
+
+static void show_info_page(const char *perf_cmd)
+{
+	const char *page = cmd_to_page(perf_cmd);
+	setenv("INFOPATH", system_path(PERF_INFO_PATH), 1);
+	execlp("info", "info", "perfman", page, NULL);
+}
+
+static void get_html_page_path(struct strbuf *page_path, const char *page)
+{
+	struct stat st;
+	const char *html_path = system_path(PERF_HTML_PATH);
+
+	/* Check that we have a perf documentation directory. */
+	if (stat(mkpath("%s/perf.html", html_path), &st)
+	    || !S_ISREG(st.st_mode))
+		die("'%s': not a documentation directory.", html_path);
+
+	strbuf_init(page_path, 0);
+	strbuf_addf(page_path, "%s/%s.html", html_path, page);
+}
+
+/*
+ * If open_html is not defined in a platform-specific way (see for
+ * example compat/mingw.h), we use the script web--browse to display
+ * HTML.
+ */
+#ifndef open_html
+static void open_html(const char *path)
+{
+	execl_perf_cmd("web--browse", "-c", "help.browser", path, NULL);
+}
+#endif
+
+static void show_html_page(const char *perf_cmd)
+{
+	const char *page = cmd_to_page(perf_cmd);
+	struct strbuf page_path; /* it leaks but we exec bellow */
+
+	get_html_page_path(&page_path, page);
+
+	open_html(page_path.buf);
+}
+
+int cmd_help(int argc, const char **argv, const char *prefix)
+{
+	const char *alias;
+	load_command_list("perf-", &main_cmds, &other_cmds);
+
+	perf_config(perf_help_config, NULL);
+
+	argc = parse_options(argc, argv, builtin_help_options,
+			builtin_help_usage, 0);
+
+	if (show_all) {
+		printf("\n usage: %s\n\n", perf_usage_string);
+		list_commands("perf commands", &main_cmds, &other_cmds);
+		printf(" %s\n\n", perf_more_info_string);
+		return 0;
+	}
+
+	if (!argv[0]) {
+		printf("\n usage: %s\n\n", perf_usage_string);
+		list_common_cmds_help();
+		printf("\n %s\n\n", perf_more_info_string);
+		return 0;
+	}
+
+	alias = alias_lookup(argv[0]);
+	if (alias && !is_perf_command(argv[0])) {
+		printf("`perf %s' is aliased to `%s'\n", argv[0], alias);
+		return 0;
+	}
+
+	switch (help_format) {
+	case HELP_FORMAT_MAN:
+		show_man_page(argv[0]);
+		break;
+	case HELP_FORMAT_INFO:
+		show_info_page(argv[0]);
+		break;
+	case HELP_FORMAT_WEB:
+		show_html_page(argv[0]);
+		break;
+	}
+
+	return 0;
+}
diff --git a/tools/perf/builtin-list.c b/tools/perf/builtin-list.c
new file mode 100644
index 000000000000..fe60e37c96ef
--- /dev/null
+++ b/tools/perf/builtin-list.c
@@ -0,0 +1,20 @@
+/*
+ * builtin-list.c
+ *
+ * Builtin list command: list all event types
+ *
+ * Copyright (C) 2009, Thomas Gleixner <tglx@linutronix.de>
+ * Copyright (C) 2008-2009, Red Hat Inc, Ingo Molnar <mingo@redhat.com>
+ */
+#include "builtin.h"
+
+#include "perf.h"
+
+#include "util/parse-options.h"
+#include "util/parse-events.h"
+
+int cmd_list(int argc, const char **argv, const char *prefix)
+{
+	print_events();
+	return 0;
+}
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
new file mode 100644
index 000000000000..0f5771f615da
--- /dev/null
+++ b/tools/perf/builtin-record.c
@@ -0,0 +1,585 @@
+/*
+ * builtin-record.c
+ *
+ * Builtin record command: Record the profile of a workload
+ * (or a CPU, or a PID) into the perf.data output file - for
+ * later analysis via perf report.
+ */
+#include "builtin.h"
+
+#include "perf.h"
+
+#include "util/util.h"
+#include "util/parse-options.h"
+#include "util/parse-events.h"
+#include "util/string.h"
+
+#include <unistd.h>
+#include <sched.h>
+
+#define ALIGN(x, a)		__ALIGN_MASK(x, (typeof(x))(a)-1)
+#define __ALIGN_MASK(x, mask)	(((x)+(mask))&~(mask))
+
+static int			fd[MAX_NR_CPUS][MAX_COUNTERS];
+
+static long			default_interval		= 100000;
+
+static int			nr_cpus				= 0;
+static unsigned int		page_size;
+static unsigned int		mmap_pages			= 128;
+static int			freq				= 0;
+static int			output;
+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			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;
+
+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];
+};
+
+
+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 unsigned int mmap_read_head(struct mmap_data *md)
+{
+	struct perf_counter_mmap_page *pc = md->base;
+	int head;
+
+	head = pc->data_head;
+	rmb();
+
+	return head;
+}
+
+static void mmap_read(struct mmap_data *md)
+{
+	unsigned int head = mmap_read_head(md);
+	unsigned int old = md->prev;
+	unsigned char *data = md->base + page_size;
+	unsigned long size;
+	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.
+	 *
+	 * If we somehow ended up ahead of the head, we got messed up.
+	 *
+	 * In either case, truncate and restart at head.
+	 */
+	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);
+
+		/*
+		 * head points to a known good entry, start there.
+		 */
+		old = head;
+	}
+
+	last_read = this_read;
+
+	if (old != head)
+		samples++;
+
+	size = head - old;
+
+	if ((old & md->mask) + size != (head & md->mask)) {
+		buf = &data[old & md->mask];
+		size = md->mask + 1 - (old & md->mask);
+		old += size;
+
+		while (size) {
+			int ret = write(output, buf, size);
+
+			if (ret < 0)
+				die("failed to write");
+
+			size -= ret;
+			buf += ret;
+
+			bytes_written += ret;
+		}
+	}
+
+	buf = &data[old & md->mask];
+	size = head - old;
+	old += size;
+
+	while (size) {
+		int ret = write(output, buf, size);
+
+		if (ret < 0)
+			die("failed to write");
+
+		size -= ret;
+		buf += ret;
+
+		bytes_written += ret;
+	}
+
+	md->prev = old;
+}
+
+static volatile int done = 0;
+static volatile int signr = -1;
+
+static void sig_handler(int sig)
+{
+	done = 1;
+	signr = sig;
+}
+
+static void sig_atexit(void)
+{
+	if (signr == -1)
+		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, ret;
+	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) {
+		fprintf(stderr, "couldn't open %s\n", filename);
+		exit(EXIT_FAILURE);
+	}
+	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;
+
+		ret = write(output, &comm_ev, comm_ev.header.size);
+		if (ret < 0) {
+			perror("failed to write");
+			exit(-1);
+		}
+		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;
+
+		comm_ev.tid = pid;
+
+		ret = write(output, &comm_ev, comm_ev.header.size);
+		if (ret < 0) {
+			perror("failed to write");
+			exit(-1);
+		}
+	}
+	closedir(tasks);
+	return;
+
+out_failure:
+	fprintf(stderr, "couldn't get COMM and pgid, malformed %s\n",
+		filename);
+	exit(EXIT_FAILURE);
+}
+
+static void pid_synthesize_mmap_samples(pid_t pid)
+{
+	char filename[PATH_MAX];
+	FILE *fp;
+
+	snprintf(filename, sizeof(filename), "/proc/%d/maps", pid);
+
+	fp = fopen(filename, "r");
+	if (fp == NULL) {
+		fprintf(stderr, "couldn't open %s\n", filename);
+		exit(EXIT_FAILURE);
+	}
+	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;
+
+			if (write(output, &mmap_ev, mmap_ev.header.size) < 0) {
+				perror("failed to write");
+				exit(-1);
+			}
+		}
+	}
+
+	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;
+
+		pid = strtol(dirent.d_name, &end, 10);
+		if (*end) /* only interested in proper numerical dirents */
+			continue;
+
+		pid_synthesize_comm_event(pid, 1);
+		pid_synthesize_mmap_samples(pid);
+	}
+
+	closedir(proc);
+}
+
+static int group_fd;
+
+static void create_counter(int counter, int cpu, pid_t pid)
+{
+	struct perf_counter_attr *attr = attrs + counter;
+	int track = 1;
+
+	attr->sample_type	= PERF_SAMPLE_IP | PERF_SAMPLE_TID;
+	if (freq) {
+		attr->sample_type	|= PERF_SAMPLE_PERIOD;
+		attr->freq		= 1;
+		attr->sample_freq	= freq;
+	}
+	attr->mmap		= track;
+	attr->comm		= track;
+	attr->inherit		= (cpu < 0) && inherit;
+	attr->disabled		= 1;
+
+	track = 0; /* only the first counter needs these */
+
+try_again:
+	fd[nr_cpu][counter] = sys_perf_counter_open(attr, pid, cpu, group_fd, 0);
+
+	if (fd[nr_cpu][counter] < 0) {
+		int err = errno;
+
+		if (err == EPERM)
+			die("Permission error - 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:
+		 */
+		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);
+	}
+
+	assert(fd[nr_cpu][counter] >= 0);
+	fcntl(fd[nr_cpu][counter], F_SETFL, O_NONBLOCK);
+
+	/*
+	 * 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, 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);
+	}
+
+	ioctl(fd[nr_cpu][counter], PERF_COUNTER_IOC_ENABLE);
+}
+
+static void open_counters(int cpu, pid_t pid)
+{
+	int counter;
+
+	if (pid > 0) {
+		pid_synthesize_comm_event(pid, 0);
+		pid_synthesize_mmap_samples(pid);
+	}
+
+	group_fd = -1;
+	for (counter = 0; counter < nr_counters; counter++)
+		create_counter(counter, cpu, pid);
+
+	nr_cpu++;
+}
+
+static int __cmd_record(int argc, const char **argv)
+{
+	int i, counter;
+	struct stat st;
+	pid_t pid;
+	int flags;
+	int ret;
+
+	page_size = sysconf(_SC_PAGE_SIZE);
+	nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
+	assert(nr_cpus <= MAX_NR_CPUS);
+	assert(nr_cpus >= 0);
+
+	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);
+		exit(-1);
+	}
+
+	flags = O_CREAT|O_RDWR;
+	if (append_file)
+		flags |= O_APPEND;
+	else
+		flags |= O_TRUNC;
+
+	output = open(output_name, flags, S_IRUSR|S_IWUSR);
+	if (output < 0) {
+		perror("failed to create output file");
+		exit(-1);
+	}
+
+	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);
+
+	atexit(sig_atexit);
+	signal(SIGCHLD, sig_handler);
+	signal(SIGINT, sig_handler);
+
+	if (target_pid == -1 && argc) {
+		pid = fork();
+		if (pid < 0)
+			perror("failed to fork");
+
+		if (!pid) {
+			if (execvp(argv[0], (char **)argv)) {
+				perror(argv[0]);
+				exit(-1);
+			}
+		}
+	}
+
+	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");
+			exit(-1);
+		}
+	}
+
+	if (system_wide)
+		synthesize_samples();
+
+	while (!done) {
+		int hits = samples;
+
+		for (i = 0; i < nr_cpu; i++) {
+			for (counter = 0; counter < nr_counters; counter++)
+				mmap_read(&mmap_array[i][counter]);
+		}
+
+		if (hits == samples)
+			ret = poll(event_array, nr_poll, 100);
+	}
+
+	/*
+	 * Approximate RIP event size: 24 bytes.
+	 */
+	fprintf(stderr,
+		"[ perf record: Captured and wrote %.3f MB %s (~%lld samples) ]\n",
+		(double)bytes_written / 1024.0 / 1024.0,
+		output_name,
+		bytes_written / 24);
+
+	return 0;
+}
+
+static const char * const record_usage[] = {
+	"perf record [<options>] [<command>]",
+	"perf record [<options>] -- <command> [<options>]",
+	NULL
+};
+
+static const struct option options[] = {
+	OPT_CALLBACK('e', "event", NULL, "event",
+		     "event selector. use 'perf list' to list available events",
+		     parse_events),
+	OPT_INTEGER('p', "pid", &target_pid,
+		    "record events on existing pid"),
+	OPT_INTEGER('r', "realtime", &realtime_prio,
+		    "collect data with this RT SCHED_FIFO priority"),
+	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_BOOLEAN('f', "force", &force,
+			"overwrite existing data file"),
+	OPT_LONG('c', "count", &default_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('v', "verbose", &verbose,
+		    "be more verbose (show counter open errors, etc)"),
+	OPT_END()
+};
+
+int cmd_record(int argc, const char **argv, const char *prefix)
+{
+	int counter;
+
+	argc = parse_options(argc, argv, options, record_usage, 0);
+	if (!argc && target_pid == -1 && !system_wide)
+		usage_with_options(record_usage, options);
+
+	if (!nr_counters) {
+		nr_counters	= 1;
+		attrs[0].type	= PERF_TYPE_HARDWARE;
+		attrs[0].config = PERF_COUNT_HW_CPU_CYCLES;
+	}
+
+	for (counter = 0; counter < nr_counters; counter++) {
+		if (attrs[counter].sample_period)
+			continue;
+
+		attrs[counter].sample_period = default_interval;
+	}
+
+	return __cmd_record(argc, argv);
+}
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
new file mode 100644
index 000000000000..82fa93b4db99
--- /dev/null
+++ b/tools/perf/builtin-report.c
@@ -0,0 +1,1316 @@
+/*
+ * builtin-report.c
+ *
+ * Builtin report command: Analyze the perf.data input file,
+ * look up and read DSOs and symbol information and display
+ * a histogram of results, along various sorting keys.
+ */
+#include "builtin.h"
+
+#include "util/util.h"
+
+#include "util/color.h"
+#include "util/list.h"
+#include "util/cache.h"
+#include "util/rbtree.h"
+#include "util/symbol.h"
+#include "util/string.h"
+
+#include "perf.h"
+
+#include "util/parse-options.h"
+#include "util/parse-events.h"
+
+#define SHOW_KERNEL	1
+#define SHOW_USER	2
+#define SHOW_HV		4
+
+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 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		full_paths;
+
+static unsigned long	page_size;
+static unsigned long	mmap_window = 32;
+
+struct ip_event {
+	struct perf_event_header header;
+	__u64 ip;
+	__u32 pid, tid;
+	__u64 period;
+};
+
+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;
+
+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;
+}
+
+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;
+		}
+
+		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 || anon)
+			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 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;
+
+	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;
+	__u64		 ip;
+	char		 level;
+
+	__u64		 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 int64_t
+sort__comm_collapse(struct hist_entry *left, struct hist_entry *right)
+{
+	char *comm_l = left->thread->comm;
+	char *comm_r = right->thread->comm;
+
+	if (!comm_l || !comm_r) {
+		if (!comm_l && !comm_r)
+			return 0;
+		else if (!comm_l)
+			return -1;
+		else
+			return 1;
+	}
+
+	return strcmp(comm_l, comm_r);
+}
+
+static size_t
+sort__comm_print(FILE *fp, struct hist_entry *self)
+{
+	return fprintf(fp, "%16s", self->thread->comm);
+}
+
+static struct sort_entry sort_comm = {
+	.header		= "         Command",
+	.cmp		= sort__comm_cmp,
+	.collapse	= sort__comm_collapse,
+	.print		= sort__comm_print,
+};
+
+/* --sort dso */
+
+static int64_t
+sort__dso_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	struct dso *dso_l = left->dso;
+	struct dso *dso_r = right->dso;
+
+	if (!dso_l || !dso_r) {
+		if (!dso_l && !dso_r)
+			return 0;
+		else if (!dso_l)
+			return -1;
+		else
+			return 1;
+	}
+
+	return strcmp(dso_l->name, dso_r->name);
+}
+
+static size_t
+sort__dso_print(FILE *fp, struct hist_entry *self)
+{
+	if (self->dso)
+		return fprintf(fp, "%-25s", self->dso->name);
+
+	return fprintf(fp, "%016llx         ", (__u64)self->ip);
+}
+
+static struct sort_entry sort_dso = {
+	.header = "Shared Object            ",
+	.cmp	= sort__dso_cmp,
+	.print	= sort__dso_print,
+};
+
+/* --sort symbol */
+
+static int64_t
+sort__sym_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	__u64 ip_l, ip_r;
+
+	if (left->sym == right->sym)
+		return 0;
+
+	ip_l = left->sym ? left->sym->start : left->ip;
+	ip_r = right->sym ? right->sym->start : right->ip;
+
+	return (int64_t)(ip_r - ip_l);
+}
+
+static size_t
+sort__sym_print(FILE *fp, struct hist_entry *self)
+{
+	size_t ret = 0;
+
+	if (verbose)
+		ret += fprintf(fp, "%#018llx  ", (__u64)self->ip);
+
+	if (self->sym) {
+		ret += fprintf(fp, "[%c] %s",
+			self->dso == kernel_dso ? 'k' : '.', self->sym->name);
+	} else {
+		ret += fprintf(fp, "%#016llx", (__u64)self->ip);
+	}
+
+	return ret;
+}
+
+static struct sort_entry sort_sym = {
+	.header = "Symbol",
+	.cmp	= sort__sym_cmp,
+	.print	= sort__sym_print,
+};
+
+static int sort__need_collapse = 0;
+
+struct sort_dimension {
+	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;
+}
+
+static size_t
+hist_entry__fprintf(FILE *fp, struct hist_entry *self, __u64 total_samples)
+{
+	struct sort_entry *se;
+	size_t ret;
+
+	if (total_samples) {
+		double percent = self->count * 100.0 / total_samples;
+		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 >= 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) {
+		fprintf(fp, "  ");
+		ret += se->print(fp, self);
+	}
+
+	ret += fprintf(fp, "\n");
+
+	return ret;
+}
+
+/*
+ * collect histogram counts
+ */
+
+static int
+hist_entry__add(struct thread *thread, struct map *map, struct dso *dso,
+		struct symbol *sym, __u64 ip, char level, __u64 count)
+{
+	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,
+	};
+	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->count += count;
+			return 0;
+		}
+
+		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 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)
+		fprintf(fp, "  %s", se->header);
+	fprintf(fp, "\n");
+
+	fprintf(fp, "# ........");
+	list_for_each_entry(se, &hist_entry__sort_list, list) {
+		int i;
+
+		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);
+	}
+
+	if (!strcmp(sort_order, default_sort_order)) {
+		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;
+
+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;
+
+	if (event->header.type & PERF_SAMPLE_PERIOD)
+		period = event->ip.period;
+
+	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);
+
+	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, period)) {
+			fprintf(stderr,
+		"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)
+{
+	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++;
+
+	return 0;
+}
+
+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);
+
+	dprintf("%p [%p]: PERF_EVENT_FORK: %d:%d\n",
+		(void *)(offset + head),
+		(void *)(long)(event->header.size),
+		event->fork.pid, event->fork.ppid);
+
+	if (!thread || !parent || thread__fork(thread, parent)) {
+		dprintf("problem processing PERF_EVENT_FORK, skipping event.\n");
+		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:
+		return -1;
+	}
+
+	return 0;
+}
+
+static int __cmd_report(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) {
+		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);
+	}
+
+	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;
+	}
+
+	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);
+	}
+
+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++;
+
+		/*
+		 * assume we lost track of the stream, check alignment, and
+		 * increment a single u64 in the hope to catch on again 'soon'.
+		 */
+
+		if (unlikely(head & 7))
+			head &= ~7ULL;
+
+		size = 8;
+	}
+
+	head += size;
+
+	if (offset + head < 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 >= 2)
+		dsos__fprintf(stdout);
+
+	collapse__resort();
+	output__resort();
+	output__fprintf(stdout, total);
+
+	return rc;
+}
+
+static const char * const report_usage[] = {
+	"perf report [<options>] <command>",
+	NULL
+};
+
+static const struct option options[] = {
+	OPT_STRING('i', "input", &input_name, "file",
+		    "input file name"),
+	OPT_BOOLEAN('v', "verbose", &verbose,
+		    "be more verbose (show symbol address, etc)"),
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+		    "dump raw trace in ASCII"),
+	OPT_STRING('k', "vmlinux", &vmlinux, "file", "vmlinux pathname"),
+	OPT_STRING('s', "sort", &sort_order, "key[,key2...]",
+		   "sort by key(s): pid, comm, dso, symbol. Default: pid,symbol"),
+	OPT_BOOLEAN('P', "full-paths", &full_paths,
+		    "Don't shorten the pathnames taking into account the cwd"),
+	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) < 0) {
+			error("Unknown --sort key: `%s'", tok);
+			usage_with_options(report_usage, options);
+		}
+	}
+
+	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);
+
+	setup_sorting();
+
+	/*
+	 * Any (unrecognized) arguments left?
+	 */
+	if (argc)
+		usage_with_options(report_usage, options);
+
+	setup_pager();
+
+	return __cmd_report();
+}
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
new file mode 100644
index 000000000000..c43e4a97dc42
--- /dev/null
+++ b/tools/perf/builtin-stat.c
@@ -0,0 +1,367 @@
+/*
+ * builtin-stat.c
+ *
+ * Builtin stat command: Give a precise performance counters summary
+ * overview about any workload, CPU or specific PID.
+ *
+ * Sample output:
+
+   $ perf stat ~/hackbench 10
+   Time: 0.104
+
+    Performance counter stats for '/home/mingo/hackbench':
+
+       1255.538611  task clock ticks     #      10.143 CPU utilization factor
+             54011  context switches     #       0.043 M/sec
+               385  CPU migrations       #       0.000 M/sec
+             17755  pagefaults           #       0.014 M/sec
+        3808323185  CPU cycles           #    3033.219 M/sec
+        1575111190  instructions         #    1254.530 M/sec
+          17367895  cache references     #      13.833 M/sec
+           7674421  cache misses         #       6.112 M/sec
+
+    Wall-clock time elapsed:   123.786620 msecs
+
+ *
+ * Copyright (C) 2008, Red Hat Inc, Ingo Molnar <mingo@redhat.com>
+ *
+ * Improvements and fixes by:
+ *
+ *   Arjan van de Ven <arjan@linux.intel.com>
+ *   Yanmin Zhang <yanmin.zhang@intel.com>
+ *   Wu Fengguang <fengguang.wu@intel.com>
+ *   Mike Galbraith <efault@gmx.de>
+ *   Paul Mackerras <paulus@samba.org>
+ *
+ * Released under the GPL v2. (and only v2, not any later version)
+ */
+
+#include "perf.h"
+#include "builtin.h"
+#include "util/util.h"
+#include "util/parse-options.h"
+#include "util/parse-events.h"
+
+#include <sys/prctl.h>
+
+static struct perf_counter_attr default_attrs[MAX_COUNTERS] = {
+
+  { .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	},
+
+  { .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	},
+
+};
+
+static int			system_wide			=  0;
+static int			inherit				=  1;
+static int			verbose				=  0;
+
+static int			fd[MAX_NR_CPUS][MAX_COUNTERS];
+
+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,
+};
+
+static __u64			event_res[MAX_COUNTERS][3];
+static __u64			event_scaled[MAX_COUNTERS];
+
+static __u64			runtime_nsecs;
+static __u64			walltime_nsecs;
+static __u64			runtime_cycles;
+
+static void create_perf_stat_counter(int counter)
+{
+	struct perf_counter_attr *attr = attrs + counter;
+
+	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;
+
+		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));
+		}
+	}
+}
+
+/*
+ * Does the counter have nsecs as a unit?
+ */
+static inline int nsec_counter(int counter)
+{
+	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)
+		return 1;
+
+	return 0;
+}
+
+/*
+ * Read out the results of a single counter:
+ */
+static void read_counter(int counter)
+{
+	__u64 *count, single_count[3];
+	ssize_t res;
+	int cpu, nv;
+	int scaled;
+
+	count = event_res[counter];
+
+	count[0] = count[1] = count[2] = 0;
+
+	nv = scale ? 3 : 1;
+	for (cpu = 0; cpu < nr_cpus; cpu ++) {
+		if (fd[cpu][counter] < 0)
+			continue;
+
+		res = read(fd[cpu][counter], single_count, nv * sizeof(__u64));
+		assert(res == nv * sizeof(__u64));
+
+		count[0] += single_count[0];
+		if (scale) {
+			count[1] += single_count[1];
+			count[2] += single_count[2];
+		}
+	}
+
+	scaled = 0;
+	if (scale) {
+		if (count[2] == 0) {
+			event_scaled[counter] = -1;
+			count[0] = 0;
+			return;
+		}
+
+		if (count[2] < count[1]) {
+			event_scaled[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 = count[0];
+	if (attrs[counter].type == PERF_TYPE_HARDWARE &&
+		attrs[counter].config == PERF_COUNT_HW_CPU_CYCLES)
+		runtime_cycles = count[0];
+}
+
+/*
+ * Print out the results of a single counter:
+ */
+static void print_counter(int counter)
+{
+	__u64 *count;
+	int scaled;
+
+	count = event_res[counter];
+	scaled = event_scaled[counter];
+
+	if (scaled == -1) {
+		fprintf(stderr, " %14s  %-20s\n",
+			"<not counted>", event_name(counter));
+		return;
+	}
+
+	if (nsec_counter(counter)) {
+		double msecs = (double)count[0] / 1000000;
+
+		fprintf(stderr, " %14.6f  %-20s",
+			msecs, event_name(counter));
+		if (attrs[counter].type == PERF_TYPE_SOFTWARE &&
+			attrs[counter].config == PERF_COUNT_SW_TASK_CLOCK) {
+
+			if (walltime_nsecs)
+				fprintf(stderr, " # %11.3f CPU utilization factor",
+					(double)count[0] / (double)walltime_nsecs);
+		}
+	} else {
+		fprintf(stderr, " %14Ld  %-20s",
+			count[0], event_name(counter));
+		if (runtime_nsecs)
+			fprintf(stderr, " # %11.3f M/sec",
+				(double)count[0]/runtime_nsecs*1000.0);
+		if (runtime_cycles &&
+			attrs[counter].type == PERF_TYPE_HARDWARE &&
+				attrs[counter].config == PERF_COUNT_HW_INSTRUCTIONS) {
+
+			fprintf(stderr, " # %1.3f per cycle",
+				(double)count[0] / (double)runtime_cycles);
+		}
+	}
+	if (scaled)
+		fprintf(stderr, "  (scaled from %.2f%%)",
+			(double) count[2] / count[1] * 100);
+	fprintf(stderr, "\n");
+}
+
+static int do_perf_stat(int argc, const char **argv)
+{
+	unsigned long long t0, t1;
+	int counter;
+	int status;
+	int pid;
+	int i;
+
+	if (!system_wide)
+		nr_cpus = 1;
+
+	for (counter = 0; counter < nr_counters; counter++)
+		create_perf_stat_counter(counter);
+
+	/*
+	 * Enable counters and exec the command:
+	 */
+	t0 = rdclock();
+	prctl(PR_TASK_PERF_COUNTERS_ENABLE);
+
+	if ((pid = fork()) < 0)
+		perror("failed to fork");
+
+	if (!pid) {
+		if (execvp(argv[0], (char **)argv)) {
+			perror(argv[0]);
+			exit(-1);
+		}
+	}
+
+	while (wait(&status) >= 0)
+		;
+
+	prctl(PR_TASK_PERF_COUNTERS_DISABLE);
+	t1 = rdclock();
+
+	walltime_nsecs = t1 - t0;
+
+	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, "\':\n");
+	fprintf(stderr, "\n");
+
+	for (counter = 0; counter < nr_counters; counter++)
+		read_counter(counter);
+
+	for (counter = 0; counter < nr_counters; counter++)
+		print_counter(counter);
+
+
+	fprintf(stderr, "\n");
+	fprintf(stderr, " Wall-clock time elapsed: %12.6f msecs\n",
+			(double)(t1-t0)/1e6);
+	fprintf(stderr, "\n");
+
+	return 0;
+}
+
+static volatile int signr = -1;
+
+static void skip_signal(int signo)
+{
+	signr = signo;
+}
+
+static void sig_atexit(void)
+{
+	if (signr == -1)
+		return;
+
+	signal(signr, SIG_DFL);
+	kill(getpid(), signr);
+}
+
+static const char * const stat_usage[] = {
+	"perf stat [<options>] <command>",
+	NULL
+};
+
+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_INTEGER('p', "pid", &target_pid,
+		    "stat events on existing pid"),
+	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,
+		    "be more verbose (show counter open errors, etc)"),
+	OPT_END()
+};
+
+int cmd_stat(int argc, const char **argv, const char *prefix)
+{
+	page_size = sysconf(_SC_PAGE_SIZE);
+
+	memcpy(attrs, default_attrs, sizeof(attrs));
+
+	argc = parse_options(argc, argv, options, stat_usage, 0);
+	if (!argc)
+		usage_with_options(stat_usage, options);
+
+	if (!nr_counters)
+		nr_counters = 8;
+
+	nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
+	assert(nr_cpus <= MAX_NR_CPUS);
+	assert(nr_cpus >= 0);
+
+	/*
+	 * We dont want to block the signals - that would cause
+	 * child tasks to inherit that and Ctrl-C would not work.
+	 * What we want is for Ctrl-C to work in the exec()-ed
+	 * task, but being ignored by perf stat itself:
+	 */
+	atexit(sig_atexit);
+	signal(SIGINT,  skip_signal);
+	signal(SIGALRM, skip_signal);
+	signal(SIGABRT, skip_signal);
+
+	return do_perf_stat(argc, argv);
+}
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
new file mode 100644
index 000000000000..fe338d3c5d7e
--- /dev/null
+++ b/tools/perf/builtin-top.c
@@ -0,0 +1,736 @@
+/*
+ * builtin-top.c
+ *
+ * Builtin top command: Display a continuously updated profile of
+ * any workload, CPU or specific PID.
+ *
+ * Copyright (C) 2008, Red Hat Inc, Ingo Molnar <mingo@redhat.com>
+ *
+ * Improvements and fixes by:
+ *
+ *   Arjan van de Ven <arjan@linux.intel.com>
+ *   Yanmin Zhang <yanmin.zhang@intel.com>
+ *   Wu Fengguang <fengguang.wu@intel.com>
+ *   Mike Galbraith <efault@gmx.de>
+ *   Paul Mackerras <paulus@samba.org>
+ *
+ * Released under the GPL v2. (and only v2, not any later version)
+ */
+#include "builtin.h"
+
+#include "perf.h"
+
+#include "util/symbol.h"
+#include "util/color.h"
+#include "util/util.h"
+#include "util/rbtree.h"
+#include "util/parse-options.h"
+#include "util/parse-events.h"
+
+#include <assert.h>
+#include <fcntl.h>
+
+#include <stdio.h>
+
+#include <errno.h>
+#include <time.h>
+#include <sched.h>
+#include <pthread.h>
+
+#include <sys/syscall.h>
+#include <sys/ioctl.h>
+#include <sys/poll.h>
+#include <sys/prctl.h>
+#include <sys/wait.h>
+#include <sys/uio.h>
+#include <sys/mman.h>
+
+#include <linux/unistd.h>
+#include <linux/types.h>
+
+static int			fd[MAX_NR_CPUS][MAX_COUNTERS];
+
+static int			system_wide			=  0;
+
+static int			default_interval		= 100000;
+
+static __u64			count_filter			=  5;
+static int			print_entries			= 15;
+
+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 unsigned int		page_size;
+static unsigned int		mmap_pages			= 16;
+static int			freq				=  0;
+static int			verbose				=  0;
+
+static char			*sym_filter;
+static unsigned long		filter_start;
+static unsigned long		filter_end;
+
+static int			delay_secs			=  2;
+static int			zero;
+static int			dump_symtab;
+
+/*
+ * Symbols
+ */
+
+static __u64			min_ip;
+static __u64			max_ip = -1ll;
+
+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;
+};
+
+struct sym_entry		*sym_filter_entry;
+
+struct dso			*kernel_dso;
+
+/*
+ * Symbols will be added here in record_ip and will get out
+ * after decayed.
+ */
+static LIST_HEAD(active_symbols);
+static pthread_mutex_t active_symbols_lock = PTHREAD_MUTEX_INITIALIZER;
+
+/*
+ * Ordering weight: count-1 * count-2 * ... / count-n
+ */
+static double sym_weight(const struct sym_entry *sym)
+{
+	double weight = sym->snap_count;
+	int counter;
+
+	for (counter = 1; counter < nr_counters-1; counter++)
+		weight *= sym->count[counter];
+
+	weight /= (sym->count[counter] + 1);
+
+	return weight;
+}
+
+static long			samples;
+static long			userspace_samples;
+static const char		CONSOLE_CLEAR[] = "";
+
+static void __list_insert_active_sym(struct sym_entry *syme)
+{
+	list_add(&syme->node, &active_symbols);
+}
+
+static void list_remove_active_sym(struct sym_entry *syme)
+{
+	pthread_mutex_lock(&active_symbols_lock);
+	list_del_init(&syme->node);
+	pthread_mutex_unlock(&active_symbols_lock);
+}
+
+static void rb_insert_active_sym(struct rb_root *tree, struct sym_entry *se)
+{
+	struct rb_node **p = &tree->rb_node;
+	struct rb_node *parent = NULL;
+	struct sym_entry *iter;
+
+	while (*p != NULL) {
+		parent = *p;
+		iter = rb_entry(parent, struct sym_entry, rb_node);
+
+		if (se->weight > iter->weight)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	rb_link_node(&se->rb_node, parent, p);
+	rb_insert_color(&se->rb_node, tree);
+}
+
+static void print_sym_table(void)
+{
+	int printed = 0, j;
+	int counter;
+	float samples_per_sec = samples/delay_secs;
+	float ksamples_per_sec = (samples-userspace_samples)/delay_secs;
+	float sum_ksamples = 0.0;
+	struct sym_entry *syme, *n;
+	struct rb_root tmp = RB_ROOT;
+	struct rb_node *nd;
+
+	samples = userspace_samples = 0;
+
+	/* Sort the active symbols */
+	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) {
+		syme->snap_count = syme->count[0];
+		if (syme->snap_count != 0) {
+			syme->weight = sym_weight(syme);
+			rb_insert_active_sym(&tmp, syme);
+			sum_ksamples += syme->snap_count;
+
+			for (j = 0; j < nr_counters; j++)
+				syme->count[j] = zero ? 0 : syme->count[j] * 7 / 8;
+		} else
+			list_remove_active_sym(syme);
+	}
+
+	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)));
+
+	if (nr_counters == 1) {
+		printf("%Ld", attrs[0].sample_period);
+		if (freq)
+			printf("Hz ");
+		else
+			printf(" ");
+	}
+
+	for (counter = 0; counter < nr_counters; counter++) {
+		if (counter)
+			printf("/");
+
+		printf("%s", event_name(counter));
+	}
+
+	printf( "], ");
+
+	if (target_pid != -1)
+		printf(" (target_pid: %d", target_pid);
+	else
+		printf(" (all");
+
+	if (profile_cpu != -1)
+		printf(", cpu: %d)\n", profile_cpu);
+	else {
+		if (target_pid != -1)
+			printf(")\n");
+		else
+			printf(", %d CPUs)\n", nr_cpus);
+	}
+
+	printf("------------------------------------------------------------------------------\n\n");
+
+	if (nr_counters == 1)
+		printf("             samples    pcnt");
+	else
+		printf("  weight     samples    pcnt");
+
+	printf("         RIP          kernel function\n"
+	       	       "  ______     _______   _____   ________________   _______________\n\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;
+		double pcnt;
+
+		if (++printed > print_entries || 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)
+			printf("%20.2f - ", syme->weight);
+		else
+			printf("%9.1f %10ld - ", syme->weight, syme->snap_count);
+
+		color_fprintf(stdout, color, "%4.1f%%", pcnt);
+		printf(" - %016llx : %s\n", sym->start, sym->name);
+	}
+}
+
+static void *display_thread(void *arg)
+{
+	struct pollfd stdin_poll = { .fd = 0, .events = POLLIN };
+	int delay_msecs = delay_secs * 1000;
+
+	printf("PerfTop refresh period: %d seconds\n", delay_secs);
+
+	do {
+		print_sym_table();
+	} while (!poll(&stdin_poll, 1, delay_msecs) == 1);
+
+	printf("key pressed - exiting.\n");
+	exit(0);
+
+	return NULL;
+}
+
+static int symbol_filter(struct dso *self, struct symbol *sym)
+{
+	static int filter_match;
+	struct sym_entry *syme;
+	const char *name = sym->name;
+
+	if (!strcmp(name, "_text") ||
+	    !strcmp(name, "_etext") ||
+	    !strcmp(name, "_sinittext") ||
+	    !strncmp("init_module", name, 11) ||
+	    !strncmp("cleanup_module", name, 14) ||
+	    strstr(name, "_text_start") ||
+	    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);
+		}
+	}
+
+	if (filter_match == 0 && sym_filter && !strcmp(name, sym_filter)) {
+		filter_match = 1;
+		filter_start = sym->start;
+	}
+
+
+	return 0;
+}
+
+static int parse_symbols(void)
+{
+	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;
+
+	node = rb_last(&kernel_dso->syms);
+	sym = rb_entry(node, struct symbol, rb_node);
+	max_ip = sym->end;
+
+	if (dump_symtab)
+		dso__fprintf(kernel_dso, stderr);
+
+	return 0;
+
+out_delete_dso:
+	dso__delete(kernel_dso);
+	kernel_dso = NULL;
+	return -1;
+}
+
+#define TRACE_COUNT     3
+
+/*
+ * 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 (sym != NULL) {
+		struct sym_entry *syme = dso__sym_priv(kernel_dso, sym);
+
+		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;
+		}
+	}
+
+	samples--;
+}
+
+static void process_event(__u64 ip, int counter)
+{
+	samples++;
+
+	if (ip < min_ip || ip > max_ip) {
+		userspace_samples++;
+		return;
+	}
+
+	record_ip(ip, counter);
+}
+
+struct mmap_data {
+	int			counter;
+	void			*base;
+	unsigned int		mask;
+	unsigned int		prev;
+};
+
+static unsigned int mmap_read_head(struct mmap_data *md)
+{
+	struct perf_counter_mmap_page *pc = md->base;
+	int head;
+
+	head = pc->data_head;
+	rmb();
+
+	return head;
+}
+
+struct timeval last_read, this_read;
+
+static void mmap_read_counter(struct mmap_data *md)
+{
+	unsigned int head = mmap_read_head(md);
+	unsigned int old = md->prev;
+	unsigned char *data = md->base + page_size;
+	int diff;
+
+	gettimeofday(&this_read, NULL);
+
+	/*
+	 * If we're further behind than half the buffer, there's a chance
+	 * the writer will bite our tail and mess up the samples under us.
+	 *
+	 * If we somehow ended up ahead of the head, we got messed up.
+	 *
+	 * In either case, truncate and restart at head.
+	 */
+	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);
+
+		/*
+		 * head points to a known good entry, start there.
+		 */
+		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;
+
+		size_t size = event->header.size;
+
+		/*
+		 * Event straddles the mmap boundary -- header should always
+		 * be inside due to u64 alignment of output.
+		 */
+		if ((old & md->mask) + size != ((old + size) & md->mask)) {
+			unsigned int offset = old;
+			unsigned int len = min(sizeof(*event), size), cpy;
+			void *dst = &event_copy;
+
+			do {
+				cpy = min(md->mask + 1 - (offset & md->mask), len);
+				memcpy(dst, &data[offset & md->mask], cpy);
+				offset += cpy;
+				dst += cpy;
+				len -= cpy;
+			} while (len);
+
+			event = &event_copy;
+		}
+
+		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 void mmap_read(void)
+{
+	int i, counter;
+
+	for (i = 0; i < nr_cpus; i++) {
+		for (counter = 0; counter < nr_counters; counter++)
+			mmap_read_counter(&mmap_array[i][counter]);
+	}
+}
+
+int nr_poll;
+int group_fd;
+
+static void start_counter(int i, int counter)
+{
+	struct perf_counter_attr *attr;
+	unsigned int cpu;
+
+	cpu = profile_cpu;
+	if (target_pid == -1 && profile_cpu == -1)
+		cpu = i;
+
+	attr = attrs + counter;
+
+	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 (fd[i][counter] < 0) {
+		int err = errno;
+
+		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:
+		 */
+		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);
+	}
+	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;
+
+	for (i = 0; i < nr_cpus; i++) {
+		group_fd = -1;
+		for (counter = 0; counter < nr_counters; counter++)
+			start_counter(i, counter);
+	}
+
+	/* Wait for a minimal set of events before starting the snapshot */
+	poll(event_array, nr_poll, 100);
+
+	mmap_read();
+
+	if (pthread_create(&thread, NULL, display_thread, NULL)) {
+		printf("Could not create display thread.\n");
+		exit(-1);
+	}
+
+	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");
+			exit(-1);
+		}
+	}
+
+	while (1) {
+		int hits = samples;
+
+		mmap_read();
+
+		if (hits == samples)
+			ret = poll(event_array, nr_poll, 100);
+	}
+
+	return 0;
+}
+
+static const char * const top_usage[] = {
+	"perf top [<options>]",
+	NULL
+};
+
+static const struct option options[] = {
+	OPT_CALLBACK('e', "event", NULL, "event",
+		     "event selector. use 'perf list' to list available events",
+		     parse_events),
+	OPT_INTEGER('c', "count", &default_interval,
+		    "event period to sample"),
+	OPT_INTEGER('p', "pid", &target_pid,
+		    "profile events on existing pid"),
+	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_INTEGER('r', "realtime", &realtime_prio,
+		    "collect data with this RT SCHED_FIFO priority"),
+	OPT_INTEGER('d', "delay", &delay_secs,
+		    "number of seconds to delay between refreshes"),
+	OPT_BOOLEAN('D', "dump-symtab", &dump_symtab,
+			    "dump the symbol table used for profiling"),
+	OPT_INTEGER('f', "count-filter", &count_filter,
+		    "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,
+		    "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,
+		    "be more verbose (show counter open errors, etc)"),
+	OPT_END()
+};
+
+int cmd_top(int argc, const char **argv, const char *prefix)
+{
+	int counter;
+
+	page_size = sysconf(_SC_PAGE_SIZE);
+
+	argc = parse_options(argc, argv, options, top_usage, 0);
+	if (argc)
+		usage_with_options(top_usage, options);
+
+	if (freq) {
+		default_interval = freq;
+		freq = 1;
+	}
+
+	/* CPU and PID are mutually exclusive */
+	if (target_pid != -1 && profile_cpu != -1) {
+		printf("WARNING: PID switch overriding CPU\n");
+		sleep(1);
+		profile_cpu = -1;
+	}
+
+	if (!nr_counters)
+		nr_counters = 1;
+
+	if (delay_secs < 1)
+		delay_secs = 1;
+
+	parse_symbols();
+
+	/*
+	 * Fill in the ones not specifically initialized via -c:
+	 */
+	for (counter = 0; counter < nr_counters; counter++) {
+		if (attrs[counter].sample_period)
+			continue;
+
+		attrs[counter].sample_period = default_interval;
+	}
+
+	nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
+	assert(nr_cpus <= MAX_NR_CPUS);
+	assert(nr_cpus >= 0);
+
+	if (target_pid != -1 || profile_cpu != -1)
+		nr_cpus = 1;
+
+	return __cmd_top();
+}
diff --git a/tools/perf/builtin.h b/tools/perf/builtin.h
new file mode 100644
index 000000000000..51d168230ee7
--- /dev/null
+++ b/tools/perf/builtin.h
@@ -0,0 +1,26 @@
+#ifndef BUILTIN_H
+#define BUILTIN_H
+
+#include "util/util.h"
+#include "util/strbuf.h"
+
+extern const char perf_version_string[];
+extern const char perf_usage_string[];
+extern const char perf_more_info_string[];
+
+extern void list_common_cmds_help(void);
+extern const char *help_unknown_cmd(const char *cmd);
+extern void prune_packed_objects(int);
+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_help(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_top(int argc, const char **argv, const char *prefix);
+extern int cmd_version(int argc, const char **argv, const char *prefix);
+extern int cmd_list(int argc, const char **argv, const char *prefix);
+
+#endif
diff --git a/tools/perf/command-list.txt b/tools/perf/command-list.txt
new file mode 100644
index 000000000000..eebce30afbc0
--- /dev/null
+++ b/tools/perf/command-list.txt
@@ -0,0 +1,10 @@
+#
+# List of known perf commands.
+# command name			category [deprecated] [common]
+#
+perf-annotate			mainporcelain common
+perf-list			mainporcelain common
+perf-record			mainporcelain common
+perf-report			mainporcelain common
+perf-stat			mainporcelain common
+perf-top			mainporcelain common
diff --git a/tools/perf/design.txt b/tools/perf/design.txt
new file mode 100644
index 000000000000..f71e0d245cba
--- /dev/null
+++ b/tools/perf/design.txt
@@ -0,0 +1,457 @@
+
+Performance Counters for Linux
+------------------------------
+
+Performance counters are special hardware registers available on most modern
+CPUs. These registers count the number of certain types of hw events: such
+as instructions executed, cachemisses suffered, or branches mis-predicted -
+without slowing down the kernel or applications. These registers can also
+trigger interrupts when a threshold number of events have passed - and can
+thus be used to profile the code that runs on that CPU.
+
+The Linux Performance Counter subsystem provides an abstraction of these
+hardware capabilities. It provides per task and per CPU counters, counter
+groups, and it provides event capabilities on top of those.  It
+provides "virtual" 64-bit counters, regardless of the width of the
+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()
+system call:
+
+   int sys_perf_counter_open(struct perf_counter_hw_event *hw_event_uptr,
+			     pid_t pid, int cpu, int group_fd,
+			     unsigned long flags);
+
+The syscall returns the new fd. The fd can be used via the normal
+VFS system calls: read() can be used to read the counter, fcntl()
+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:
+
+struct perf_counter_hw_event {
+        /*
+         * The MSB of the config word signifies if the rest contains cpu
+         * specific (raw) counter configuration data, if unset, the next
+         * 7 bits are an event type and the rest of the bits are the event
+         * identifier.
+         */
+        __u64                   config;
+
+        __u64                   irq_period;
+        __u32                   record_type;
+        __u32                   read_format;
+
+        __u64                   disabled       :  1, /* off by default        */
+                                inherit        :  1, /* children inherit it   */
+                                pinned         :  1, /* must always be on PMU */
+                                exclusive      :  1, /* only group on PMU     */
+                                exclude_user   :  1, /* don't count user      */
+                                exclude_kernel :  1, /* ditto kernel          */
+                                exclude_hv     :  1, /* ditto hypervisor      */
+                                exclude_idle   :  1, /* don't count when idle */
+                                mmap           :  1, /* include mmap data     */
+                                munmap         :  1, /* include munmap data   */
+                                comm           :  1, /* include comm data     */
+
+                                __reserved_1   : 52;
+
+        __u32                   extra_config_len;
+        __u32                   wakeup_events;  /* wakeup every n events */
+
+        __u64                   __reserved_2;
+        __u64                   __reserved_3;
+};
+
+The 'config' field specifies what the counter should count.  It
+is divided into 3 bit-fields:
+
+raw_type: 1 bit   (most significant bit)	0x8000_0000_0000_0000
+type:	  7 bits  (next most significant)	0x7f00_0000_0000_0000
+event_id: 56 bits (least significant)		0x00ff_ffff_ffff_ffff
+
+If 'raw_type' is 1, then the counter will count a hardware event
+specified by the remaining 63 bits of event_config.  The encoding is
+machine-specific.
+
+If 'raw_type' is 0, then the 'type' field says what kind of counter
+this is, with the following encoding:
+
+enum perf_event_types {
+	PERF_TYPE_HARDWARE		= 0,
+	PERF_TYPE_SOFTWARE		= 1,
+	PERF_TYPE_TRACEPOINT		= 2,
+};
+
+A counter of PERF_TYPE_HARDWARE will count the hardware event
+specified by 'event_id':
+
+/*
+ * Generalized performance counter event types, used by the hw_event.event_id
+ * parameter of the sys_perf_counter_open() syscall:
+ */
+enum hw_event_ids {
+	/*
+	 * Common hardware events, generalized by the kernel:
+	 */
+	PERF_COUNT_HW_CPU_CYCLES		= 0,
+	PERF_COUNT_HW_INSTRUCTIONS		= 1,
+	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_BUS_CYCLES		= 6,
+};
+
+These are standardized types of events that work relatively uniformly
+on all CPUs that implement Performance Counters support under Linux,
+although there may be variations (e.g., different CPUs might count
+cache references and misses at different levels of the cache hierarchy).
+If a CPU is not able to count the selected event, then the system call
+will return -EINVAL.
+
+More hw_event_types are supported as well, but they are CPU-specific
+and accessed as raw events.  For example, to count "External bus
+cycles while bus lock signal asserted" events on Intel Core CPUs, pass
+in a 0x4064 event_id value and set hw_event.raw_type to 1.
+
+A counter of type PERF_TYPE_SOFTWARE will count one of the available
+software events, selected by 'event_id':
+
+/*
+ * Special "software" counters provided by the kernel, even if the hardware
+ * does not support performance counters. These counters measure various
+ * physical and sw events of the kernel (and allow the profiling of them as
+ * well):
+ */
+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_CONTEXT_SWITCHES	= 3,
+	PERF_COUNT_SW_CPU_MIGRATIONS	= 4,
+	PERF_COUNT_SW_PAGE_FAULTS_MIN	= 5,
+	PERF_COUNT_SW_PAGE_FAULTS_MAJ	= 6,
+};
+
+Counters of the type PERF_TYPE_TRACEPOINT are available when the ftrace event
+tracer is available, and event_id values can be obtained from
+/debug/tracing/events/*/*/id
+
+
+Counters come in two flavours: counting counters and sampling
+counters.  A "counting" counter is one that is used for counting the
+number of events that occur, and is characterised by having
+irq_period = 0.
+
+
+A read() on a counter returns the current value of the counter and possible
+additional values as specified by 'read_format', each value is a u64 (8 bytes)
+in size.
+
+/*
+ * Bits that can be set in hw_event.read_format to request that
+ * reads on the counter should return the indicated quantities,
+ * in increasing order of bit value, after the counter value.
+ */
+enum perf_counter_read_format {
+        PERF_FORMAT_TOTAL_TIME_ENABLED  =  1,
+        PERF_FORMAT_TOTAL_TIME_RUNNING  =  2,
+};
+
+Using these additional values one can establish the overcommit ratio for a
+particular counter allowing one to take the round-robin scheduling effect
+into account.
+
+
+A "sampling" counter is one that is set up to generate an interrupt
+every N events, where N is given by 'irq_period'.  A sampling counter
+has irq_period > 0. The record_type controls what data is recorded on each
+interrupt:
+
+/*
+ * Bits that can be set in hw_event.record_type to request information
+ * in the overflow packets.
+ */
+enum perf_counter_record_format {
+        PERF_RECORD_IP          = 1U << 0,
+        PERF_RECORD_TID         = 1U << 1,
+        PERF_RECORD_TIME        = 1U << 2,
+        PERF_RECORD_ADDR        = 1U << 3,
+        PERF_RECORD_GROUP       = 1U << 4,
+        PERF_RECORD_CALLCHAIN   = 1U << 5,
+};
+
+Such (and other) events will be recorded in a ring-buffer, which is
+available to user-space using mmap() (see below).
+
+The 'disabled' bit specifies whether the counter starts out disabled
+or enabled.  If it is initially disabled, it can be enabled by ioctl
+or prctl (see below).
+
+The 'inherit' bit, if set, specifies that this counter should count
+events on descendant tasks as well as the task specified.  This only
+applies to new descendents, not to any existing descendents at the
+time the counter is created (nor to any new descendents of existing
+descendents).
+
+The 'pinned' bit, if set, specifies that the counter should always be
+on the CPU if at all possible.  It only applies to hardware counters
+and only to group leaders.  If a pinned counter cannot be put onto the
+CPU (e.g. because there are not enough hardware counters or because of
+a conflict with some other event), then the counter goes into an
+'error' state, where reads return end-of-file (i.e. read() returns 0)
+until the counter is subsequently enabled or disabled.
+
+The 'exclusive' bit, if set, specifies that when this counter's group
+is on the CPU, it should be the only group using the CPU's counters.
+In future, this will allow sophisticated monitoring programs to supply
+extra configuration information via 'extra_config_len' to exploit
+advanced features of the CPU's Performance Monitor Unit (PMU) that are
+not otherwise accessible and that might disrupt other hardware
+counters.
+
+The 'exclude_user', 'exclude_kernel' and 'exclude_hv' bits provide a
+way to request that counting of events be restricted to times when the
+CPU is in user, kernel and/or hypervisor mode.
+
+The 'mmap' and 'munmap' bits allow recording of PROT_EXEC mmap/munmap
+operations, these can be used to relate userspace IP addresses to actual
+code, even after the mapping (or even the whole process) is gone,
+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
+counter to be specific to a task:
+
+ pid == 0: if the pid parameter is zero, the counter is attached to the
+ current task.
+
+ pid > 0: the counter is attached to a specific task (if the current task
+ has sufficient privilege to do so)
+
+ pid < 0: all tasks are counted (per cpu counters)
+
+The 'cpu' parameter allows a counter to be made specific to a CPU:
+
+ cpu >= 0: the counter is restricted to a specific CPU
+ cpu == -1: the counter counts on all CPUs
+
+(Note: the combination of 'pid == -1' and 'cpu == -1' is not valid.)
+
+A 'pid > 0' and 'cpu == -1' counter is a per task counter that counts
+events of that task and 'follows' that task to whatever CPU the task
+gets schedule to. Per task counters can be created by any user, for
+their own tasks.
+
+A 'pid == -1' and 'cpu == x' counter is a per CPU counter that counts
+all events on CPU-x. Per CPU counters need CAP_SYS_ADMIN privilege.
+
+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
+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
+considered to be a group with only 1 member.)
+
+A counter group is scheduled onto the CPU as a unit, that is, it will
+only be put onto the CPU if all of the counters in the group can be
+put onto the CPU.  This means that the values of the member counters
+can be meaningfully compared, added, divided (to get ratios), etc.,
+with each other, since they have counted events for the same set of
+executed instructions.
+
+
+Like stated, asynchronous events, like counter overflow or PROT_EXEC mmap
+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
+as where the ring-buffer head is.
+
+/*
+ * Structure of the page that can be mapped via mmap
+ */
+struct perf_counter_mmap_page {
+        __u32   version;                /* version number of this structure */
+        __u32   compat_version;         /* lowest version this is compat with */
+
+        /*
+         * Bits needed to read the hw counters in user-space.
+         *
+         *   u32 seq;
+         *   s64 count;
+         *
+         *   do {
+         *     seq = pc->lock;
+         *
+         *     barrier()
+         *     if (pc->index) {
+         *       count = pmc_read(pc->index - 1);
+         *       count += pc->offset;
+         *     } else
+         *       goto regular_read;
+         *
+         *     barrier();
+         *   } while (pc->lock != seq);
+         *
+         * NOTE: for obvious reason this only works on self-monitoring
+         *       processes.
+         */
+        __u32   lock;                   /* seqlock for synchronization */
+        __u32   index;                  /* hardware counter identifier */
+        __s64   offset;                 /* add to hardware counter value */
+
+        /*
+         * 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().
+         */
+        __u32   data_head;              /* head in the data section */
+};
+
+NOTE: the hw-counter userspace bits are arch specific and are currently only
+      implemented on powerpc.
+
+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)
+
+struct perf_event_header {
+        __u32   type;
+        __u16   misc;
+        __u16   size;
+};
+
+enum perf_event_type {
+
+        /*
+         * The MMAP events record the PROT_EXEC mappings so that we can
+         * correlate userspace IPs to code. They have the following structure:
+         *
+         * struct {
+         *      struct perf_event_header        header;
+         *
+         *      u32                             pid, tid;
+         *      u64                             addr;
+         *      u64                             len;
+         *      u64                             pgoff;
+         *      char                            filename[];
+         * };
+         */
+        PERF_EVENT_MMAP                 = 1,
+        PERF_EVENT_MUNMAP               = 2,
+
+        /*
+         * struct {
+         *      struct perf_event_header        header;
+         *
+         *      u32                             pid, tid;
+         *      char                            comm[];
+         * };
+         */
+        PERF_EVENT_COMM                 = 3,
+
+        /*
+         * When header.misc & PERF_EVENT_MISC_OVERFLOW the event_type field
+         * will be PERF_RECORD_*
+         *
+         * struct {
+         *      struct perf_event_header        header;
+         *
+         *      { u64                   ip;       } && PERF_RECORD_IP
+         *      { u32                   pid, tid; } && PERF_RECORD_TID
+         *      { u64                   time;     } && PERF_RECORD_TIME
+         *      { u64                   addr;     } && PERF_RECORD_ADDR
+         *
+         *      { u64                   nr;
+         *        { u64 event, val; }   cnt[nr];  } && PERF_RECORD_GROUP
+         *
+         *      { u16                   nr,
+         *                              hv,
+         *                              kernel,
+         *                              user;
+         *        u64                   ips[nr];  } && PERF_RECORD_CALLCHAIN
+         * };
+         */
+};
+
+NOTE: PERF_RECORD_CALLCHAIN is arch specific and currently only implemented
+      on x86.
+
+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
+so many counter overflow events.
+
+Future work will include a splice() interface to the ring-buffer.
+
+
+Counters can be enabled and disabled in two ways: via ioctl and via
+prctl.  When a counter is disabled, it doesn't count or generate
+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);
+
+or disabled with
+
+	ioctl(fd, PERF_COUNTER_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
+counters in the group will count.  Enabling or disabling a member of a
+group other than the leader only affects that counter - disabling an
+non-leader stops that counter from counting but doesn't affect any
+other counter.
+
+Additionally, non-inherited overflow counters can use
+
+	ioctl(fd, PERF_COUNTER_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_COUNTERS_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
+this process has created on other processes.  It only enables or
+disables the group leaders, not any other members in the groups.
+
+
+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
+will need at least this:
+	- asm/perf_counter.h - a basic stub will suffice at first
+	- support for atomic64 types (and associated helper functions)
+	- set_perf_counter_pending() implemented
+
+If your architecture does have hardware capabilities, you can override the
+weak stub hw_perf_counter_init() to register hardware counters.
diff --git a/tools/perf/perf.c b/tools/perf/perf.c
new file mode 100644
index 000000000000..4eb725933703
--- /dev/null
+++ b/tools/perf/perf.c
@@ -0,0 +1,428 @@
+/*
+ * perf.c
+ *
+ * Performance analysis utility.
+ *
+ * This is the main hub from which the sub-commands (perf stat,
+ * perf top, perf record, perf report, etc.) are started.
+ */
+#include "builtin.h"
+
+#include "util/exec_cmd.h"
+#include "util/cache.h"
+#include "util/quote.h"
+#include "util/run-command.h"
+
+const char perf_usage_string[] =
+	"perf [--version] [--help] COMMAND [ARGS]";
+
+const char perf_more_info_string[] =
+	"See 'perf help COMMAND' for more information on a specific command.";
+
+static int use_pager = -1;
+struct pager_config {
+	const char *cmd;
+	int val;
+};
+
+static int pager_command_config(const char *var, const char *value, void *data)
+{
+	struct pager_config *c = data;
+	if (!prefixcmp(var, "pager.") && !strcmp(var + 6, c->cmd))
+		c->val = perf_config_bool(var, value);
+	return 0;
+}
+
+/* returns 0 for "no pager", 1 for "use pager", and -1 for "not specified" */
+int check_pager_config(const char *cmd)
+{
+	struct pager_config c;
+	c.cmd = cmd;
+	c.val = -1;
+	perf_config(pager_command_config, &c);
+	return c.val;
+}
+
+static void commit_pager_choice(void) {
+	switch (use_pager) {
+	case 0:
+		setenv("PERF_PAGER", "cat", 1);
+		break;
+	case 1:
+		/* setup_pager(); */
+		break;
+	default:
+		break;
+	}
+}
+
+static int handle_options(const char*** argv, int* argc, int* envchanged)
+{
+	int handled = 0;
+
+	while (*argc > 0) {
+		const char *cmd = (*argv)[0];
+		if (cmd[0] != '-')
+			break;
+
+		/*
+		 * For legacy reasons, the "version" and "help"
+		 * commands can be written with "--" prepended
+		 * to make them look like flags.
+		 */
+		if (!strcmp(cmd, "--help") || !strcmp(cmd, "--version"))
+			break;
+
+		/*
+		 * Check remaining flags.
+		 */
+		if (!prefixcmp(cmd, "--exec-path")) {
+			cmd += 11;
+			if (*cmd == '=')
+				perf_set_argv_exec_path(cmd + 1);
+			else {
+				puts(perf_exec_path());
+				exit(0);
+			}
+		} else if (!strcmp(cmd, "--html-path")) {
+			puts(system_path(PERF_HTML_PATH));
+			exit(0);
+		} else if (!strcmp(cmd, "-p") || !strcmp(cmd, "--paginate")) {
+			use_pager = 1;
+		} else if (!strcmp(cmd, "--no-pager")) {
+			use_pager = 0;
+			if (envchanged)
+				*envchanged = 1;
+		} else if (!strcmp(cmd, "--perf-dir")) {
+			if (*argc < 2) {
+				fprintf(stderr, "No directory given for --perf-dir.\n" );
+				usage(perf_usage_string);
+			}
+			setenv(PERF_DIR_ENVIRONMENT, (*argv)[1], 1);
+			if (envchanged)
+				*envchanged = 1;
+			(*argv)++;
+			(*argc)--;
+			handled++;
+		} else if (!prefixcmp(cmd, "--perf-dir=")) {
+			setenv(PERF_DIR_ENVIRONMENT, cmd + 10, 1);
+			if (envchanged)
+				*envchanged = 1;
+		} else if (!strcmp(cmd, "--work-tree")) {
+			if (*argc < 2) {
+				fprintf(stderr, "No directory given for --work-tree.\n" );
+				usage(perf_usage_string);
+			}
+			setenv(PERF_WORK_TREE_ENVIRONMENT, (*argv)[1], 1);
+			if (envchanged)
+				*envchanged = 1;
+			(*argv)++;
+			(*argc)--;
+		} else if (!prefixcmp(cmd, "--work-tree=")) {
+			setenv(PERF_WORK_TREE_ENVIRONMENT, cmd + 12, 1);
+			if (envchanged)
+				*envchanged = 1;
+		} else {
+			fprintf(stderr, "Unknown option: %s\n", cmd);
+			usage(perf_usage_string);
+		}
+
+		(*argv)++;
+		(*argc)--;
+		handled++;
+	}
+	return handled;
+}
+
+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 *alias_command;
+	char *alias_string;
+
+	alias_command = (*argv)[0];
+	alias_string = alias_lookup(alias_command);
+	if (alias_string) {
+		if (alias_string[0] == '!') {
+			if (*argcp > 1) {
+				struct strbuf buf;
+
+				strbuf_init(&buf, PATH_MAX);
+				strbuf_addstr(&buf, alias_string);
+				sq_quote_argv(&buf, (*argv) + 1, PATH_MAX);
+				free(alias_string);
+				alias_string = buf.buf;
+			}
+			ret = system(alias_string + 1);
+			if (ret >= 0 && WIFEXITED(ret) &&
+			    WEXITSTATUS(ret) != 127)
+				exit(WEXITSTATUS(ret));
+			die("Failed to run '%s' when expanding alias '%s'",
+			    alias_string + 1, alias_command);
+		}
+		count = split_cmdline(alias_string, &new_argv);
+		if (count < 0)
+			die("Bad alias.%s string", alias_command);
+		option_count = handle_options(&new_argv, &count, &envchanged);
+		if (envchanged)
+			die("alias '%s' changes environment variables\n"
+				 "You can use '!perf' in the alias to do this.",
+				 alias_command);
+		memmove(new_argv - option_count, new_argv,
+				count * sizeof(char *));
+		new_argv -= option_count;
+
+		if (count < 1)
+			die("empty alias for %s", alias_command);
+
+		if (!strcmp(alias_command, new_argv[0]))
+			die("recursive alias: %s", alias_command);
+
+		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;
+
+		*argv = new_argv;
+		*argcp += count - 1;
+
+		ret = 1;
+	}
+
+	errno = saved_errno;
+
+	return ret;
+}
+
+const char perf_version_string[] = PERF_VERSION;
+
+#define RUN_SETUP	(1<<0)
+#define USE_PAGER	(1<<1)
+/*
+ * require working tree to be present -- anything uses this needs
+ * RUN_SETUP for reading from the configuration file.
+ */
+#define NEED_WORK_TREE	(1<<2)
+
+struct cmd_struct {
+	const char *cmd;
+	int (*fn)(int, const char **, const char *);
+	int option;
+};
+
+static int run_builtin(struct cmd_struct *p, int argc, const char **argv)
+{
+	int status;
+	struct stat st;
+	const char *prefix;
+
+	prefix = NULL;
+	if (p->option & RUN_SETUP)
+		prefix = NULL; /* setup_perf_directory(); */
+
+	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() */;
+
+	status = p->fn(argc, argv, prefix);
+	if (status)
+		return status & 0xff;
+
+	/* Somebody closed stdout? */
+	if (fstat(fileno(stdout), &st))
+		return 0;
+	/* Ignore write errors for pipes and sockets.. */
+	if (S_ISFIFO(st.st_mode) || S_ISSOCK(st.st_mode))
+		return 0;
+
+	/* Check for ENOSPC and EIO errors.. */
+	if (fflush(stdout))
+		die("write failure on standard output: %s", strerror(errno));
+	if (ferror(stdout))
+		die("unknown write failure on standard output");
+	if (fclose(stdout))
+		die("close failed on standard output: %s", strerror(errno));
+	return 0;
+}
+
+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 },
+	};
+	int i;
+	static const char ext[] = STRIP_EXTENSION;
+
+	if (sizeof(ext) > 1) {
+		i = strlen(argv[0]) - strlen(ext);
+		if (i > 0 && !strcmp(argv[0] + i, ext)) {
+			char *argv0 = strdup(argv[0]);
+			argv[0] = cmd = argv0;
+			argv0[i] = '\0';
+		}
+	}
+
+	/* Turn "perf cmd --help" into "perf help cmd" */
+	if (argc > 1 && !strcmp(argv[1], "--help")) {
+		argv[1] = argv[0];
+		argv[0] = cmd = "help";
+	}
+
+	for (i = 0; i < ARRAY_SIZE(commands); i++) {
+		struct cmd_struct *p = commands+i;
+		if (strcmp(p->cmd, cmd))
+			continue;
+		exit(run_builtin(p, argc, argv));
+	}
+}
+
+static void execv_dashed_external(const char **argv)
+{
+	struct strbuf cmd = STRBUF_INIT;
+	const char *tmp;
+	int status;
+
+	strbuf_addf(&cmd, "perf-%s", argv[0]);
+
+	/*
+	 * argv[0] must be the perf command, but the argv array
+	 * belongs to the caller, and may be reused in
+	 * subsequent loop iterations. Save argv[0] and
+	 * restore it on error.
+	 */
+	tmp = argv[0];
+	argv[0] = cmd.buf;
+
+	/*
+	 * if we fail because the command is not found, it is
+	 * OK to return. Otherwise, we just pass along the status code.
+	 */
+	status = run_command_v_opt(argv, 0);
+	if (status != -ERR_RUN_COMMAND_EXEC) {
+		if (IS_RUN_COMMAND_ERR(status))
+			die("unable to run '%s'", argv[0]);
+		exit(-status);
+	}
+	errno = ENOENT; /* as if we called execvp */
+
+	argv[0] = tmp;
+
+	strbuf_release(&cmd);
+}
+
+static int run_argv(int *argcp, const char ***argv)
+{
+	int done_alias = 0;
+
+	while (1) {
+		/* See if it's an internal command */
+		handle_internal_command(*argcp, *argv);
+
+		/* .. then try the external ones */
+		execv_dashed_external(*argv);
+
+		/* It could be an alias -- this works around the insanity
+		 * of overriding "perf log" with "perf show" by having
+		 * alias.log = show
+		 */
+		if (done_alias || !handle_alias(argcp, argv))
+			break;
+		done_alias = 1;
+	}
+
+	return done_alias;
+}
+
+
+int main(int argc, const char **argv)
+{
+	const char *cmd;
+
+	cmd = perf_extract_argv0_path(argv[0]);
+	if (!cmd)
+		cmd = "perf-help";
+
+	/*
+	 * "perf-xxxx" is the same as "perf xxxx", but we obviously:
+	 *
+	 *  - cannot take flags in between the "perf" and the "xxxx".
+	 *  - cannot execute it externally (since it would just do
+	 *    the same thing over again)
+	 *
+	 * So we just directly call the internal command handler, and
+	 * die if that one cannot handle it.
+	 */
+	if (!prefixcmp(cmd, "perf-")) {
+		cmd += 5;
+		argv[0] = cmd;
+		handle_internal_command(argc, argv);
+		die("cannot handle %s internally", cmd);
+	}
+
+	/* Look for flags.. */
+	argv++;
+	argc--;
+	handle_options(&argv, &argc, NULL);
+	commit_pager_choice();
+	if (argc > 0) {
+		if (!prefixcmp(argv[0], "--"))
+			argv[0] += 2;
+	} else {
+		/* The user didn't specify a command; give them help */
+		printf("\n usage: %s\n\n", perf_usage_string);
+		list_common_cmds_help();
+		printf("\n %s\n\n", perf_more_info_string);
+		exit(1);
+	}
+	cmd = argv[0];
+
+	/*
+	 * We use PATH to find perf commands, but we prepend some higher
+	 * precidence 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;
+
+		was_alias = run_argv(&argc, &argv);
+		if (errno != ENOENT)
+			break;
+
+		if (was_alias) {
+			fprintf(stderr, "Expansion of alias '%s' failed; "
+				"'%s' is not a perf-command\n",
+				cmd, argv[0]);
+			exit(1);
+		}
+		if (!done_help) {
+			cmd = argv[0] = help_unknown_cmd(cmd);
+			done_help = 1;
+		} else
+			break;
+	}
+
+	fprintf(stderr, "Failed to run command '%s': %s\n",
+		cmd, strerror(errno));
+
+	return 1;
+}
diff --git a/tools/perf/perf.h b/tools/perf/perf.h
new file mode 100644
index 000000000000..87a1aca4a424
--- /dev/null
+++ b/tools/perf/perf.h
@@ -0,0 +1,68 @@
+#ifndef _PERF_PERF_H
+#define _PERF_PERF_H
+
+#if defined(__x86_64__) || defined(__i386__)
+#include "../../arch/x86/include/asm/unistd.h"
+#define rmb()		asm volatile("lfence" ::: "memory")
+#define cpu_relax()	asm volatile("rep; nop" ::: "memory");
+#endif
+
+#ifdef __powerpc__
+#include "../../arch/powerpc/include/asm/unistd.h"
+#define rmb()		asm volatile ("sync" ::: "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"
+
+/*
+ * prctl(PR_TASK_PERF_COUNTERS_DISABLE) will (cheaply) disable all
+ * counters in the current task.
+ */
+#define PR_TASK_PERF_COUNTERS_DISABLE   31
+#define PR_TASK_PERF_COUNTERS_ENABLE    32
+
+#ifndef NSEC_PER_SEC
+# define NSEC_PER_SEC			1000000000ULL
+#endif
+
+static inline unsigned long long rdclock(void)
+{
+	struct timespec ts;
+
+	clock_gettime(CLOCK_MONOTONIC, &ts);
+	return ts.tv_sec * 1000000000ULL + ts.tv_nsec;
+}
+
+/*
+ * Pick up some kernel type conventions:
+ */
+#define __user
+#define asmlinkage
+
+#define unlikely(x)	__builtin_expect(!!(x), 0)
+#define min(x, y) ({				\
+	typeof(x) _min1 = (x);			\
+	typeof(y) _min2 = (y);			\
+	(void) (&_min1 == &_min2);		\
+	_min1 < _min2 ? _min1 : _min2; })
+
+static inline int
+sys_perf_counter_open(struct perf_counter_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,
+		       group_fd, flags);
+}
+
+#define MAX_COUNTERS			256
+#define MAX_NR_CPUS			256
+
+#endif
diff --git a/tools/perf/util/PERF-VERSION-GEN b/tools/perf/util/PERF-VERSION-GEN
new file mode 100755
index 000000000000..c561d1538c03
--- /dev/null
+++ b/tools/perf/util/PERF-VERSION-GEN
@@ -0,0 +1,42 @@
+#!/bin/sh
+
+GVF=PERF-VERSION-FILE
+DEF_VER=v0.0.1.PERF
+
+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="$DEF_VER"
+elif test -d .git -o -f .git &&
+	VN=$(git describe --abbrev=4 HEAD 2>/dev/null) &&
+	case "$VN" in
+	*$LF*) (exit 1) ;;
+	v[0-9]*)
+		git update-index -q --refresh
+		test -z "$(git diff-index --name-only HEAD --)" ||
+		VN="$VN-dirty" ;;
+	esac
+then
+	VN=$(echo "$VN" | sed -e 's/-/./g');
+else
+	VN="$DEF_VER"
+fi
+
+VN=$(expr "$VN" : v*'\(.*\)')
+
+if test -r $GVF
+then
+	VC=$(sed -e 's/^PERF_VERSION = //' <$GVF)
+else
+	VC=unset
+fi
+test "$VN" = "$VC" || {
+	echo >&2 "PERF_VERSION = $VN"
+	echo "PERF_VERSION = $VN" >$GVF
+}
+
+
diff --git a/tools/perf/util/abspath.c b/tools/perf/util/abspath.c
new file mode 100644
index 000000000000..61d33b81fc97
--- /dev/null
+++ b/tools/perf/util/abspath.c
@@ -0,0 +1,117 @@
+#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];
+	char *pwd;
+	struct stat cwd_stat, pwd_stat;
+	if (getcwd(cwd, PATH_MAX) == NULL)
+		return NULL;
+	pwd = getenv("PWD");
+	if (pwd && strcmp(pwd, cwd)) {
+		stat(cwd, &cwd_stat);
+		if (!stat(pwd, &pwd_stat) &&
+		    pwd_stat.st_dev == cwd_stat.st_dev &&
+		    pwd_stat.st_ino == cwd_stat.st_ino) {
+			strlcpy(cwd, pwd, PATH_MAX);
+		}
+	}
+	return cwd;
+}
+
+const char *make_nonrelative_path(const char *path)
+{
+	static char buf[PATH_MAX + 1];
+
+	if (is_absolute_path(path)) {
+		if (strlcpy(buf, path, PATH_MAX) >= PATH_MAX)
+			die("Too long path: %.*s", 60, path);
+	} else {
+		const char *cwd = get_pwd_cwd();
+		if (!cwd)
+			die("Cannot determine the current working directory");
+		if (snprintf(buf, PATH_MAX, "%s/%s", cwd, path) >= PATH_MAX)
+			die("Too long path: %.*s", 60, path);
+	}
+	return buf;
+}
diff --git a/tools/perf/util/alias.c b/tools/perf/util/alias.c
new file mode 100644
index 000000000000..9b3dd2b428df
--- /dev/null
+++ b/tools/perf/util/alias.c
@@ -0,0 +1,77 @@
+#include "cache.h"
+
+static const char *alias_key;
+static char *alias_val;
+
+static int alias_lookup_cb(const char *k, const char *v, void *cb)
+{
+	if (!prefixcmp(k, "alias.") && !strcmp(k+6, alias_key)) {
+		if (!v)
+			return config_error_nonbool(k);
+		alias_val = strdup(v);
+		return 0;
+	}
+	return 0;
+}
+
+char *alias_lookup(const char *alias)
+{
+	alias_key = alias;
+	alias_val = NULL;
+	perf_config(alias_lookup_cb, NULL);
+	return alias_val;
+}
+
+int split_cmdline(char *cmdline, const char ***argv)
+{
+	int src, dst, count = 0, size = 16;
+	char quoted = 0;
+
+	*argv = malloc(sizeof(char*) * size);
+
+	/* split alias_string */
+	(*argv)[count++] = cmdline;
+	for (src = dst = 0; cmdline[src];) {
+		char c = cmdline[src];
+		if (!quoted && isspace(c)) {
+			cmdline[dst++] = 0;
+			while (cmdline[++src]
+					&& isspace(cmdline[src]))
+				; /* skip */
+			if (count >= size) {
+				size += 16;
+				*argv = realloc(*argv, sizeof(char*) * size);
+			}
+			(*argv)[count++] = cmdline + dst;
+		} else if (!quoted && (c == '\'' || c == '"')) {
+			quoted = c;
+			src++;
+		} else if (c == quoted) {
+			quoted = 0;
+			src++;
+		} else {
+			if (c == '\\' && quoted != '\'') {
+				src++;
+				c = cmdline[src];
+				if (!c) {
+					free(*argv);
+					*argv = NULL;
+					return error("cmdline ends with \\");
+				}
+			}
+			cmdline[dst++] = c;
+			src++;
+		}
+	}
+
+	cmdline[dst] = 0;
+
+	if (quoted) {
+		free(*argv);
+		*argv = NULL;
+		return error("unclosed quote");
+	}
+
+	return count;
+}
+
diff --git a/tools/perf/util/cache.h b/tools/perf/util/cache.h
new file mode 100644
index 000000000000..393d6146d13b
--- /dev/null
+++ b/tools/perf/util/cache.h
@@ -0,0 +1,119 @@
+#ifndef CACHE_H
+#define CACHE_H
+
+#include "util.h"
+#include "strbuf.h"
+
+#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]"
+
+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 *);
+
+/* pager.c */
+extern void setup_pager(void);
+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;
+
+char *alias_lookup(const char *alias);
+int split_cmdline(char *cmdline, const char ***argv);
+
+#define alloc_nr(x) (((x)+16)*3/2)
+
+/*
+ * Realloc the buffer pointed at by variable 'x' so that it can hold
+ * at least 'nr' entries; the number of entries currently allocated
+ * is 'alloc', using the standard growing factor alloc_nr() macro.
+ *
+ * DO NOT USE any expression with side-effect for 'x' or 'alloc'.
+ */
+#define ALLOC_GROW(x, nr, alloc) \
+	do { \
+		if ((nr) > alloc) { \
+			if (alloc_nr(alloc) < (nr)) \
+				alloc = (nr); \
+			else \
+				alloc = alloc_nr(alloc); \
+			x = xrealloc((x), alloc * sizeof(*(x))); \
+		} \
+	} while(0)
+
+
+static inline int is_absolute_path(const char *path)
+{
+	return path[0] == '/';
+}
+
+const char *make_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)));
+
+extern size_t strlcpy(char *dest, const char *src, size_t size);
+
+#endif /* CACHE_H */
diff --git a/tools/perf/util/color.c b/tools/perf/util/color.c
new file mode 100644
index 000000000000..9a8c20ccc53e
--- /dev/null
+++ b/tools/perf/util/color.c
@@ -0,0 +1,241 @@
+#include "cache.h"
+#include "color.h"
+
+int perf_use_color_default = -1;
+
+static int parse_color(const char *name, int len)
+{
+	static const char * const color_names[] = {
+		"normal", "black", "red", "green", "yellow",
+		"blue", "magenta", "cyan", "white"
+	};
+	char *end;
+	int i;
+	for (i = 0; i < ARRAY_SIZE(color_names); i++) {
+		const char *str = color_names[i];
+		if (!strncasecmp(name, str, len) && !str[len])
+			return i - 1;
+	}
+	i = strtol(name, &end, 10);
+	if (end - name == len && i >= -1 && i <= 255)
+		return i;
+	return -2;
+}
+
+static int parse_attr(const char *name, int len)
+{
+	static const int attr_values[] = { 1, 2, 4, 5, 7 };
+	static const char * const attr_names[] = {
+		"bold", "dim", "ul", "blink", "reverse"
+	};
+	int i;
+	for (i = 0; i < ARRAY_SIZE(attr_names); i++) {
+		const char *str = attr_names[i];
+		if (!strncasecmp(name, str, len) && !str[len])
+			return attr_values[i];
+	}
+	return -1;
+}
+
+void color_parse(const char *value, const char *var, char *dst)
+{
+	color_parse_mem(value, strlen(value), var, dst);
+}
+
+void color_parse_mem(const char *value, int value_len, const char *var,
+		char *dst)
+{
+	const char *ptr = value;
+	int len = value_len;
+	int attr = -1;
+	int fg = -2;
+	int bg = -2;
+
+	if (!strncasecmp(value, "reset", len)) {
+		strcpy(dst, PERF_COLOR_RESET);
+		return;
+	}
+
+	/* [fg [bg]] [attr] */
+	while (len > 0) {
+		const char *word = ptr;
+		int val, wordlen = 0;
+
+		while (len > 0 && !isspace(word[wordlen])) {
+			wordlen++;
+			len--;
+		}
+
+		ptr = word + wordlen;
+		while (len > 0 && isspace(*ptr)) {
+			ptr++;
+			len--;
+		}
+
+		val = parse_color(word, wordlen);
+		if (val >= -1) {
+			if (fg == -2) {
+				fg = val;
+				continue;
+			}
+			if (bg == -2) {
+				bg = val;
+				continue;
+			}
+			goto bad;
+		}
+		val = parse_attr(word, wordlen);
+		if (val < 0 || attr != -1)
+			goto bad;
+		attr = val;
+	}
+
+	if (attr >= 0 || fg >= 0 || bg >= 0) {
+		int sep = 0;
+
+		*dst++ = '\033';
+		*dst++ = '[';
+		if (attr >= 0) {
+			*dst++ = '0' + attr;
+			sep++;
+		}
+		if (fg >= 0) {
+			if (sep++)
+				*dst++ = ';';
+			if (fg < 8) {
+				*dst++ = '3';
+				*dst++ = '0' + fg;
+			} else {
+				dst += sprintf(dst, "38;5;%d", fg);
+			}
+		}
+		if (bg >= 0) {
+			if (sep++)
+				*dst++ = ';';
+			if (bg < 8) {
+				*dst++ = '4';
+				*dst++ = '0' + bg;
+			} else {
+				dst += sprintf(dst, "48;5;%d", bg);
+			}
+		}
+		*dst++ = 'm';
+	}
+	*dst = 0;
+	return;
+bad:
+	die("bad color value '%.*s' for variable '%s'", value_len, value, var);
+}
+
+int perf_config_colorbool(const char *var, const char *value, int stdout_is_tty)
+{
+	if (value) {
+		if (!strcasecmp(value, "never"))
+			return 0;
+		if (!strcasecmp(value, "always"))
+			return 1;
+		if (!strcasecmp(value, "auto"))
+			goto auto_color;
+	}
+
+	/* Missing or explicit false to turn off colorization */
+	if (!perf_config_bool(var, value))
+		return 0;
+
+	/* any normal truth value defaults to 'auto' */
+ auto_color:
+	if (stdout_is_tty < 0)
+		stdout_is_tty = isatty(1);
+	if (stdout_is_tty || (pager_in_use() && pager_use_color)) {
+		char *term = getenv("TERM");
+		if (term && strcmp(term, "dumb"))
+			return 1;
+	}
+	return 0;
+}
+
+int perf_color_default_config(const char *var, const char *value, void *cb)
+{
+	if (!strcmp(var, "color.ui")) {
+		perf_use_color_default = perf_config_colorbool(var, value, -1);
+		return 0;
+	}
+
+	return perf_default_config(var, value, cb);
+}
+
+static int color_vfprintf(FILE *fp, const char *color, const char *fmt,
+		va_list args, const char *trail)
+{
+	int r = 0;
+
+	/*
+	 * Auto-detect:
+	 */
+	if (perf_use_color_default < 0) {
+		if (isatty(1) || pager_in_use())
+			perf_use_color_default = 1;
+		else
+			perf_use_color_default = 0;
+	}
+
+	if (perf_use_color_default && *color)
+		r += fprintf(fp, "%s", color);
+	r += vfprintf(fp, fmt, args);
+	if (perf_use_color_default && *color)
+		r += fprintf(fp, "%s", PERF_COLOR_RESET);
+	if (trail)
+		r += fprintf(fp, "%s", trail);
+	return r;
+}
+
+
+
+int color_fprintf(FILE *fp, const char *color, const char *fmt, ...)
+{
+	va_list args;
+	int r;
+
+	va_start(args, fmt);
+	r = color_vfprintf(fp, color, fmt, args, NULL);
+	va_end(args);
+	return r;
+}
+
+int color_fprintf_ln(FILE *fp, const char *color, const char *fmt, ...)
+{
+	va_list args;
+	int r;
+	va_start(args, fmt);
+	r = color_vfprintf(fp, color, fmt, args, "\n");
+	va_end(args);
+	return r;
+}
+
+/*
+ * This function splits the buffer by newlines and colors the lines individually.
+ *
+ * Returns 0 on success.
+ */
+int color_fwrite_lines(FILE *fp, const char *color,
+		size_t count, const char *buf)
+{
+	if (!*color)
+		return fwrite(buf, count, 1, fp) != 1;
+	while (count) {
+		char *p = memchr(buf, '\n', count);
+		if (p != buf && (fputs(color, fp) < 0 ||
+				fwrite(buf, p ? p - buf : count, 1, fp) != 1 ||
+				fputs(PERF_COLOR_RESET, fp) < 0))
+			return -1;
+		if (!p)
+			return 0;
+		if (fputc('\n', fp) < 0)
+			return -1;
+		count -= p + 1 - buf;
+		buf = p + 1;
+	}
+	return 0;
+}
+
+
diff --git a/tools/perf/util/color.h b/tools/perf/util/color.h
new file mode 100644
index 000000000000..5abfd379582b
--- /dev/null
+++ b/tools/perf/util/color.h
@@ -0,0 +1,36 @@
+#ifndef COLOR_H
+#define COLOR_H
+
+/* "\033[1;38;5;2xx;48;5;2xxm\0" is 23 bytes */
+#define COLOR_MAXLEN 24
+
+#define PERF_COLOR_NORMAL	""
+#define PERF_COLOR_RESET	"\033[m"
+#define PERF_COLOR_BOLD		"\033[1m"
+#define PERF_COLOR_RED		"\033[31m"
+#define PERF_COLOR_GREEN	"\033[32m"
+#define PERF_COLOR_YELLOW	"\033[33m"
+#define PERF_COLOR_BLUE		"\033[34m"
+#define PERF_COLOR_MAGENTA	"\033[35m"
+#define PERF_COLOR_CYAN		"\033[36m"
+#define PERF_COLOR_BG_RED	"\033[41m"
+
+/*
+ * This variable stores the value of color.ui
+ */
+extern int perf_use_color_default;
+
+
+/*
+ * Use this instead of perf_default_config if you need the value of color.ui.
+ */
+int perf_color_default_config(const char *var, const char *value, void *cb);
+
+int perf_config_colorbool(const char *var, const char *value, int stdout_is_tty);
+void color_parse(const char *value, const char *var, char *dst);
+void color_parse_mem(const char *value, int len, const char *var, char *dst);
+int color_fprintf(FILE *fp, 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);
+
+#endif /* COLOR_H */
diff --git a/tools/perf/util/config.c b/tools/perf/util/config.c
new file mode 100644
index 000000000000..3dd13faa6a27
--- /dev/null
+++ b/tools/perf/util/config.c
@@ -0,0 +1,873 @@
+/*
+ * GIT - The information manager from hell
+ *
+ * Copyright (C) Linus Torvalds, 2005
+ * Copyright (C) Johannes Schindelin, 2005
+ *
+ */
+#include "util.h"
+#include "cache.h"
+#include "exec_cmd.h"
+
+#define MAXNAME (256)
+
+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 int get_next_char(void)
+{
+	int c;
+	FILE *f;
+
+	c = '\n';
+	if ((f = config_file) != NULL) {
+		c = fgetc(f);
+		if (c == '\r') {
+			/* DOS like systems */
+			c = fgetc(f);
+			if (c != '\n') {
+				ungetc(c, f);
+				c = '\r';
+			}
+		}
+		if (c == '\n')
+			config_linenr++;
+		if (c == EOF) {
+			config_file_eof = 1;
+			c = '\n';
+		}
+	}
+	return c;
+}
+
+static char *parse_value(void)
+{
+	static char value[1024];
+	int quote = 0, comment = 0, len = 0, space = 0;
+
+	for (;;) {
+		int c = get_next_char();
+		if (len >= sizeof(value) - 1)
+			return NULL;
+		if (c == '\n') {
+			if (quote)
+				return NULL;
+			value[len] = 0;
+			return value;
+		}
+		if (comment)
+			continue;
+		if (isspace(c) && !quote) {
+			space = 1;
+			continue;
+		}
+		if (!quote) {
+			if (c == ';' || c == '#') {
+				comment = 1;
+				continue;
+			}
+		}
+		if (space) {
+			if (len)
+				value[len++] = ' ';
+			space = 0;
+		}
+		if (c == '\\') {
+			c = get_next_char();
+			switch (c) {
+			case '\n':
+				continue;
+			case 't':
+				c = '\t';
+				break;
+			case 'b':
+				c = '\b';
+				break;
+			case 'n':
+				c = '\n';
+				break;
+			/* Some characters escape as themselves */
+			case '\\': case '"':
+				break;
+			/* Reject unknown escape sequences */
+			default:
+				return NULL;
+			}
+			value[len++] = c;
+			continue;
+		}
+		if (c == '"') {
+			quote = 1-quote;
+			continue;
+		}
+		value[len++] = c;
+	}
+}
+
+static inline int iskeychar(int c)
+{
+	return isalnum(c) || c == '-';
+}
+
+static int get_value(config_fn_t fn, void *data, char *name, unsigned int len)
+{
+	int c;
+	char *value;
+
+	/* Get the full name */
+	for (;;) {
+		c = get_next_char();
+		if (config_file_eof)
+			break;
+		if (!iskeychar(c))
+			break;
+		name[len++] = tolower(c);
+		if (len >= MAXNAME)
+			return -1;
+	}
+	name[len] = 0;
+	while (c == ' ' || c == '\t')
+		c = get_next_char();
+
+	value = NULL;
+	if (c != '\n') {
+		if (c != '=')
+			return -1;
+		value = parse_value();
+		if (!value)
+			return -1;
+	}
+	return fn(name, value, data);
+}
+
+static int get_extended_base_var(char *name, int baselen, int c)
+{
+	do {
+		if (c == '\n')
+			return -1;
+		c = get_next_char();
+	} while (isspace(c));
+
+	/* We require the format to be '[base "extension"]' */
+	if (c != '"')
+		return -1;
+	name[baselen++] = '.';
+
+	for (;;) {
+		int c = get_next_char();
+		if (c == '\n')
+			return -1;
+		if (c == '"')
+			break;
+		if (c == '\\') {
+			c = get_next_char();
+			if (c == '\n')
+				return -1;
+		}
+		name[baselen++] = c;
+		if (baselen > MAXNAME / 2)
+			return -1;
+	}
+
+	/* Final ']' */
+	if (get_next_char() != ']')
+		return -1;
+	return baselen;
+}
+
+static int get_base_var(char *name)
+{
+	int baselen = 0;
+
+	for (;;) {
+		int c = get_next_char();
+		if (config_file_eof)
+			return -1;
+		if (c == ']')
+			return baselen;
+		if (isspace(c))
+			return get_extended_base_var(name, baselen, c);
+		if (!iskeychar(c) && c != '.')
+			return -1;
+		if (baselen > MAXNAME / 2)
+			return -1;
+		name[baselen++] = tolower(c);
+	}
+}
+
+static int perf_parse_file(config_fn_t fn, void *data)
+{
+	int comment = 0;
+	int baselen = 0;
+	static char var[MAXNAME];
+
+	/* U+FEFF Byte Order Mark in UTF8 */
+	static const unsigned char *utf8_bom = (unsigned char *) "\xef\xbb\xbf";
+	const unsigned char *bomptr = utf8_bom;
+
+	for (;;) {
+		int c = get_next_char();
+		if (bomptr && *bomptr) {
+			/* We are at the file beginning; skip UTF8-encoded BOM
+			 * if present. Sane editors won't put this in on their
+			 * own, but e.g. Windows Notepad will do it happily. */
+			if ((unsigned char) c == *bomptr) {
+				bomptr++;
+				continue;
+			} else {
+				/* Do not tolerate partial BOM. */
+				if (bomptr != utf8_bom)
+					break;
+				/* No BOM at file beginning. Cool. */
+				bomptr = NULL;
+			}
+		}
+		if (c == '\n') {
+			if (config_file_eof)
+				return 0;
+			comment = 0;
+			continue;
+		}
+		if (comment || isspace(c))
+			continue;
+		if (c == '#' || c == ';') {
+			comment = 1;
+			continue;
+		}
+		if (c == '[') {
+			baselen = get_base_var(var);
+			if (baselen <= 0)
+				break;
+			var[baselen++] = '.';
+			var[baselen] = 0;
+			continue;
+		}
+		if (!isalpha(c))
+			break;
+		var[baselen] = tolower(c);
+		if (get_value(fn, data, var, baselen+1) < 0)
+			break;
+	}
+	die("bad config file line %d in %s", config_linenr, config_file_name);
+}
+
+static int parse_unit_factor(const char *end, unsigned long *val)
+{
+	if (!*end)
+		return 1;
+	else if (!strcasecmp(end, "k")) {
+		*val *= 1024;
+		return 1;
+	}
+	else if (!strcasecmp(end, "m")) {
+		*val *= 1024 * 1024;
+		return 1;
+	}
+	else if (!strcasecmp(end, "g")) {
+		*val *= 1024 * 1024 * 1024;
+		return 1;
+	}
+	return 0;
+}
+
+static int perf_parse_long(const char *value, long *ret)
+{
+	if (value && *value) {
+		char *end;
+		long val = strtol(value, &end, 0);
+		unsigned long factor = 1;
+		if (!parse_unit_factor(end, &factor))
+			return 0;
+		*ret = val * factor;
+		return 1;
+	}
+	return 0;
+}
+
+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)
+		die("bad config value for '%s' in %s", name, config_file_name);
+	die("bad config value for '%s'", name);
+}
+
+int perf_config_int(const char *name, const char *value)
+{
+	long ret = 0;
+	if (!perf_parse_long(value, &ret))
+		die_bad_config(name);
+	return ret;
+}
+
+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)
+{
+	*is_bool = 1;
+	if (!value)
+		return 1;
+	if (!*value)
+		return 0;
+	if (!strcasecmp(value, "true") || !strcasecmp(value, "yes") || !strcasecmp(value, "on"))
+		return 1;
+	if (!strcasecmp(value, "false") || !strcasecmp(value, "no") || !strcasecmp(value, "off"))
+		return 0;
+	*is_bool = 0;
+	return perf_config_int(name, value);
+}
+
+int perf_config_bool(const char *name, const char *value)
+{
+	int discard;
+	return !!perf_config_bool_or_int(name, value, &discard);
+}
+
+int perf_config_string(const char **dest, const char *var, const char *value)
+{
+	if (!value)
+		return config_error_nonbool(var);
+	*dest = strdup(value);
+	return 0;
+}
+
+static int perf_default_core_config(const char *var, const char *value)
+{
+	/* Add other config variables here and to Documentation/config.txt. */
+	return 0;
+}
+
+int perf_default_config(const char *var, const char *value, void *dummy)
+{
+	if (!prefixcmp(var, "core."))
+		return perf_default_core_config(var, value);
+
+	/* Add other config variables here and to Documentation/config.txt. */
+	return 0;
+}
+
+int perf_config_from_file(config_fn_t fn, const char *filename, void *data)
+{
+	int ret;
+	FILE *f = fopen(filename, "r");
+
+	ret = -1;
+	if (f) {
+		config_file = f;
+		config_file_name = filename;
+		config_linenr = 1;
+		config_file_eof = 0;
+		ret = perf_parse_file(fn, data);
+		fclose(f);
+		config_file_name = NULL;
+	}
+	return ret;
+}
+
+const char *perf_etc_perfconfig(void)
+{
+	static const char *system_wide;
+	if (!system_wide)
+		system_wide = system_path(ETC_PERFCONFIG);
+	return system_wide;
+}
+
+static int perf_env_bool(const char *k, int def)
+{
+	const char *v = getenv(k);
+	return v ? perf_config_bool(k, v) : def;
+}
+
+int perf_config_system(void)
+{
+	return !perf_env_bool("PERF_CONFIG_NOSYSTEM", 0);
+}
+
+int perf_config_global(void)
+{
+	return !perf_env_bool("PERF_CONFIG_NOGLOBAL", 0);
+}
+
+int perf_config(config_fn_t fn, void *data)
+{
+	int ret = 0, found = 0;
+	char *repo_config = NULL;
+	const char *home = NULL;
+
+	/* Setting $PERF_CONFIG makes perf read _only_ the given config file. */
+	if (config_exclusive_filename)
+		return perf_config_from_file(fn, config_exclusive_filename, data);
+	if (perf_config_system() && !access(perf_etc_perfconfig(), R_OK)) {
+		ret += perf_config_from_file(fn, perf_etc_perfconfig(),
+					    data);
+		found += 1;
+	}
+
+	home = getenv("HOME");
+	if (perf_config_global() && home) {
+		char *user_config = strdup(mkpath("%s/.perfconfig", home));
+		if (!access(user_config, R_OK)) {
+			ret += perf_config_from_file(fn, user_config, data);
+			found += 1;
+		}
+		free(user_config);
+	}
+
+	repo_config = perf_pathdup("config");
+	if (!access(repo_config, R_OK)) {
+		ret += perf_config_from_file(fn, repo_config, data);
+		found += 1;
+	}
+	free(repo_config);
+	if (found == 0)
+		return -1;
+	return ret;
+}
+
+/*
+ * Find all the stuff for perf_config_set() below.
+ */
+
+#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)
+{
+	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;
+}
+
+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;
+}
+
+static int store_write_pair(int fd, const char* key, const char* value)
+{
+	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);
+
+	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;
+	}
+	if (bracket_offset < equal_offset) {
+		*found_bracket = 1;
+		offset = bracket_offset+1;
+	} else
+		offset++;
+
+	return offset;
+}
+
+int perf_config_set(const char* key, const char* value)
+{
+	return perf_config_set_multivar(key, value, NULL, 0);
+}
+
+/*
+ * 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)
+{
+	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;
+		}
+
+		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;
+			}
+		}
+
+		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;
+		}
+
+		/* 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);
+	}
+
+	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);
+}
diff --git a/tools/perf/util/ctype.c b/tools/perf/util/ctype.c
new file mode 100644
index 000000000000..b90ec004f29c
--- /dev/null
+++ b/tools/perf/util/ctype.c
@@ -0,0 +1,26 @@
+/*
+ * Sane locale-independent, ASCII ctype.
+ *
+ * No surprises, and works with signed and unsigned chars.
+ */
+#include "cache.h"
+
+enum {
+	S = GIT_SPACE,
+	A = GIT_ALPHA,
+	D = GIT_DIGIT,
+	G = GIT_GLOB_SPECIAL,	/* *, ?, [, \\ */
+	R = GIT_REGEX_SPECIAL,	/* $, (, ), +, ., ^, {, | * */
+};
+
+unsigned char sane_ctype[256] = {
+	0, 0, 0, 0, 0, 0, 0, 0, 0, S, S, 0, 0, S, 0, 0,		/*   0.. 15 */
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,		/*  16.. 31 */
+	S, 0, 0, 0, R, 0, 0, 0, R, R, G, R, 0, 0, R, 0,		/*  32.. 47 */
+	D, D, D, D, D, D, D, D, D, D, 0, 0, 0, 0, 0, G,		/*  48.. 63 */
+	0, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,		/*  64.. 79 */
+	A, A, A, A, A, A, A, A, A, A, A, G, G, 0, R, 0,		/*  80.. 95 */
+	0, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,		/*  96..111 */
+	A, A, A, A, A, A, A, A, A, A, A, R, R, 0, 0, 0,		/* 112..127 */
+	/* Nothing in the 128.. range */
+};
diff --git a/tools/perf/util/environment.c b/tools/perf/util/environment.c
new file mode 100644
index 000000000000..275b0ee345f5
--- /dev/null
+++ b/tools/perf/util/environment.c
@@ -0,0 +1,9 @@
+/*
+ * We put all the perf config variables in this same object
+ * file, so that programs can link against the config parser
+ * without having to link against all the rest of perf.
+ */
+#include "cache.h"
+
+const char *pager_program;
+int pager_use_color = 1;
diff --git a/tools/perf/util/exec_cmd.c b/tools/perf/util/exec_cmd.c
new file mode 100644
index 000000000000..d39292263153
--- /dev/null
+++ b/tools/perf/util/exec_cmd.c
@@ -0,0 +1,165 @@
+#include "cache.h"
+#include "exec_cmd.h"
+#include "quote.h"
+#define MAX_ARGS	32
+
+extern char **environ;
+static const char *argv_exec_path;
+static const char *argv0_path;
+
+const char *system_path(const char *path)
+{
+#ifdef RUNTIME_PREFIX
+	static const char *prefix;
+#else
+	static const char *prefix = PREFIX;
+#endif
+	struct strbuf d = STRBUF_INIT;
+
+	if (is_absolute_path(path))
+		return path;
+
+#ifdef RUNTIME_PREFIX
+	assert(argv0_path);
+	assert(is_absolute_path(argv0_path));
+
+	if (!prefix &&
+	    !(prefix = strip_path_suffix(argv0_path, PERF_EXEC_PATH)) &&
+	    !(prefix = strip_path_suffix(argv0_path, BINDIR)) &&
+	    !(prefix = strip_path_suffix(argv0_path, "perf"))) {
+		prefix = PREFIX;
+		fprintf(stderr, "RUNTIME_PREFIX requested, "
+				"but prefix computation failed.  "
+				"Using static fallback '%s'.\n", prefix);
+	}
+#endif
+
+	strbuf_addf(&d, "%s/%s", prefix, path);
+	path = strbuf_detach(&d, NULL);
+	return path;
+}
+
+const char *perf_extract_argv0_path(const char *argv0)
+{
+	const char *slash;
+
+	if (!argv0 || !*argv0)
+		return NULL;
+	slash = argv0 + strlen(argv0);
+
+	while (argv0 <= slash && !is_dir_sep(*slash))
+		slash--;
+
+	if (slash >= argv0) {
+		argv0_path = strndup(argv0, slash - argv0);
+		return slash + 1;
+	}
+
+	return argv0;
+}
+
+void perf_set_argv_exec_path(const char *exec_path)
+{
+	argv_exec_path = exec_path;
+	/*
+	 * Propagate this setting to external programs.
+	 */
+	setenv(EXEC_PATH_ENVIRONMENT, exec_path, 1);
+}
+
+
+/* Returns the highest-priority, location to look for perf programs. */
+const char *perf_exec_path(void)
+{
+	const char *env;
+
+	if (argv_exec_path)
+		return argv_exec_path;
+
+	env = getenv(EXEC_PATH_ENVIRONMENT);
+	if (env && *env) {
+		return env;
+	}
+
+	return system_path(PERF_EXEC_PATH);
+}
+
+static void add_path(struct strbuf *out, const char *path)
+{
+	if (path && *path) {
+		if (is_absolute_path(path))
+			strbuf_addstr(out, path);
+		else
+			strbuf_addstr(out, make_nonrelative_path(path));
+
+		strbuf_addch(out, PATH_SEP);
+	}
+}
+
+void setup_path(void)
+{
+	const char *old_path = getenv("PATH");
+	struct strbuf new_path = STRBUF_INIT;
+
+	add_path(&new_path, perf_exec_path());
+	add_path(&new_path, argv0_path);
+
+	if (old_path)
+		strbuf_addstr(&new_path, old_path);
+	else
+		strbuf_addstr(&new_path, "/usr/local/bin:/usr/bin:/bin");
+
+	setenv("PATH", new_path.buf, 1);
+
+	strbuf_release(&new_path);
+}
+
+const char **prepare_perf_cmd(const char **argv)
+{
+	int argc;
+	const char **nargv;
+
+	for (argc = 0; argv[argc]; argc++)
+		; /* just counting */
+	nargv = malloc(sizeof(*nargv) * (argc + 2));
+
+	nargv[0] = "perf";
+	for (argc = 0; argv[argc]; argc++)
+		nargv[argc + 1] = argv[argc];
+	nargv[argc + 1] = NULL;
+	return nargv;
+}
+
+int execv_perf_cmd(const char **argv) {
+	const char **nargv = prepare_perf_cmd(argv);
+
+	/* execvp() can only ever return if it fails */
+	execvp("perf", (char **)nargv);
+
+	free(nargv);
+	return -1;
+}
+
+
+int execl_perf_cmd(const char *cmd,...)
+{
+	int argc;
+	const char *argv[MAX_ARGS + 1];
+	const char *arg;
+	va_list param;
+
+	va_start(param, cmd);
+	argv[0] = cmd;
+	argc = 1;
+	while (argc < MAX_ARGS) {
+		arg = argv[argc++] = va_arg(param, char *);
+		if (!arg)
+			break;
+	}
+	va_end(param);
+	if (MAX_ARGS <= argc)
+		return error("too many args to run %s", cmd);
+
+	argv[argc] = NULL;
+	return execv_perf_cmd(argv);
+}
diff --git a/tools/perf/util/exec_cmd.h b/tools/perf/util/exec_cmd.h
new file mode 100644
index 000000000000..effe25eb1545
--- /dev/null
+++ b/tools/perf/util/exec_cmd.h
@@ -0,0 +1,13 @@
+#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 */
diff --git a/tools/perf/util/generate-cmdlist.sh b/tools/perf/util/generate-cmdlist.sh
new file mode 100755
index 000000000000..f06f6fd148f8
--- /dev/null
+++ b/tools/perf/util/generate-cmdlist.sh
@@ -0,0 +1,24 @@
+#!/bin/sh
+
+echo "/* Automatically generated by $0 */
+struct cmdname_help
+{
+    char name[16];
+    char help[80];
+};
+
+static struct cmdname_help common_cmds[] = {"
+
+sed -n -e 's/^perf-\([^ 	]*\)[ 	].* common.*/\1/p' command-list.txt |
+sort |
+while read cmd
+do
+     sed -n '
+     /^NAME/,/perf-'"$cmd"'/H
+     ${
+            x
+            s/.*perf-'"$cmd"' - \(.*\)/  {"'"$cmd"'", "\1"},/
+	    p
+     }' "Documentation/perf-$cmd.txt"
+done
+echo "};"
diff --git a/tools/perf/util/help.c b/tools/perf/util/help.c
new file mode 100644
index 000000000000..6653f7dd1d78
--- /dev/null
+++ b/tools/perf/util/help.c
@@ -0,0 +1,367 @@
+#include "cache.h"
+#include "../builtin.h"
+#include "exec_cmd.h"
+#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)
+{
+	struct cmdname *ent = malloc(sizeof(*ent) + len + 1);
+
+	ent->len = len;
+	memcpy(ent->name, name, len);
+	ent->name[len] = 0;
+
+	ALLOC_GROW(cmds->names, cmds->cnt + 1, cmds->alloc);
+	cmds->names[cmds->cnt++] = ent;
+}
+
+static void clean_cmdnames(struct cmdnames *cmds)
+{
+	int i;
+	for (i = 0; i < cmds->cnt; ++i)
+		free(cmds->names[i]);
+	free(cmds->names);
+	cmds->cnt = 0;
+	cmds->alloc = 0;
+}
+
+static int cmdname_compare(const void *a_, const void *b_)
+{
+	struct cmdname *a = *(struct cmdname **)a_;
+	struct cmdname *b = *(struct cmdname **)b_;
+	return strcmp(a->name, b->name);
+}
+
+static void uniq(struct cmdnames *cmds)
+{
+	int i, j;
+
+	if (!cmds->cnt)
+		return;
+
+	for (i = j = 1; i < cmds->cnt; i++)
+		if (strcmp(cmds->names[i]->name, cmds->names[i-1]->name))
+			cmds->names[j++] = cmds->names[i];
+
+	cmds->cnt = j;
+}
+
+void exclude_cmds(struct cmdnames *cmds, struct cmdnames *excludes)
+{
+	int ci, cj, ei;
+	int cmp;
+
+	ci = cj = ei = 0;
+	while (ci < cmds->cnt && ei < excludes->cnt) {
+		cmp = strcmp(cmds->names[ci]->name, excludes->names[ei]->name);
+		if (cmp < 0)
+			cmds->names[cj++] = cmds->names[ci++];
+		else if (cmp == 0)
+			ci++, ei++;
+		else if (cmp > 0)
+			ei++;
+	}
+
+	while (ci < cmds->cnt)
+		cmds->names[cj++] = cmds->names[ci++];
+
+	cmds->cnt = cj;
+}
+
+static void pretty_print_string_list(struct cmdnames *cmds, int longest)
+{
+	int cols = 1, rows;
+	int space = longest + 1; /* min 1 SP between words */
+	int max_cols = term_columns() - 1; /* don't print *on* the edge */
+	int i, j;
+
+	if (space < max_cols)
+		cols = max_cols / space;
+	rows = (cmds->cnt + cols - 1) / cols;
+
+	for (i = 0; i < rows; i++) {
+		printf("  ");
+
+		for (j = 0; j < cols; j++) {
+			int n = j * rows + i;
+			int size = space;
+			if (n >= cmds->cnt)
+				break;
+			if (j == cols-1 || n + rows >= cmds->cnt)
+				size = 1;
+			printf("%-*s", size, cmds->names[n]->name);
+		}
+		putchar('\n');
+	}
+}
+
+static int is_executable(const char *name)
+{
+	struct stat st;
+
+	if (stat(name, &st) || /* stat, not lstat */
+	    !S_ISREG(st.st_mode))
+		return 0;
+
+#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;
+}
+
+static void list_commands_in_dir(struct cmdnames *cmds,
+					 const char *path,
+					 const char *prefix)
+{
+	int prefix_len;
+	DIR *dir = opendir(path);
+	struct dirent *de;
+	struct strbuf buf = STRBUF_INIT;
+	int len;
+
+	if (!dir)
+		return;
+	if (!prefix)
+		prefix = "perf-";
+	prefix_len = strlen(prefix);
+
+	strbuf_addf(&buf, "%s/", path);
+	len = buf.len;
+
+	while ((de = readdir(dir)) != NULL) {
+		int entlen;
+
+		if (prefixcmp(de->d_name, prefix))
+			continue;
+
+		strbuf_setlen(&buf, len);
+		strbuf_addstr(&buf, de->d_name);
+		if (!is_executable(buf.buf))
+			continue;
+
+		entlen = strlen(de->d_name) - prefix_len;
+		if (has_extension(de->d_name, ".exe"))
+			entlen -= 4;
+
+		add_cmdname(cmds, de->d_name + prefix_len, entlen);
+	}
+	closedir(dir);
+	strbuf_release(&buf);
+}
+
+void load_command_list(const char *prefix,
+		struct cmdnames *main_cmds,
+		struct cmdnames *other_cmds)
+{
+	const char *env_path = getenv("PATH");
+	const char *exec_path = perf_exec_path();
+
+	if (exec_path) {
+		list_commands_in_dir(main_cmds, exec_path, prefix);
+		qsort(main_cmds->names, main_cmds->cnt,
+		      sizeof(*main_cmds->names), cmdname_compare);
+		uniq(main_cmds);
+	}
+
+	if (env_path) {
+		char *paths, *path, *colon;
+		path = paths = strdup(env_path);
+		while (1) {
+			if ((colon = strchr(path, PATH_SEP)))
+				*colon = 0;
+			if (!exec_path || strcmp(path, exec_path))
+				list_commands_in_dir(other_cmds, path, prefix);
+
+			if (!colon)
+				break;
+			path = colon + 1;
+		}
+		free(paths);
+
+		qsort(other_cmds->names, other_cmds->cnt,
+		      sizeof(*other_cmds->names), cmdname_compare);
+		uniq(other_cmds);
+	}
+	exclude_cmds(other_cmds, main_cmds);
+}
+
+void list_commands(const char *title, struct cmdnames *main_cmds,
+		   struct cmdnames *other_cmds)
+{
+	int i, longest = 0;
+
+	for (i = 0; i < main_cmds->cnt; i++)
+		if (longest < main_cmds->names[i]->len)
+			longest = main_cmds->names[i]->len;
+	for (i = 0; i < other_cmds->cnt; i++)
+		if (longest < other_cmds->names[i]->len)
+			longest = other_cmds->names[i]->len;
+
+	if (main_cmds->cnt) {
+		const char *exec_path = perf_exec_path();
+		printf("available %s in '%s'\n", title, exec_path);
+		printf("----------------");
+		mput_char('-', strlen(title) + strlen(exec_path));
+		putchar('\n');
+		pretty_print_string_list(main_cmds, longest);
+		putchar('\n');
+	}
+
+	if (other_cmds->cnt) {
+		printf("%s available from elsewhere on your $PATH\n", title);
+		printf("---------------------------------------");
+		mput_char('-', strlen(title));
+		putchar('\n');
+		pretty_print_string_list(other_cmds, longest);
+		putchar('\n');
+	}
+}
+
+int is_in_cmdlist(struct cmdnames *c, const char *s)
+{
+	int i;
+	for (i = 0; i < c->cnt; i++)
+		if (!strcmp(s, c->names[i]->name))
+			return 1;
+	return 0;
+}
+
+static int autocorrect;
+static struct cmdnames aliases;
+
+static int perf_unknown_cmd_config(const char *var, const char *value, void *cb)
+{
+	if (!strcmp(var, "help.autocorrect"))
+		autocorrect = perf_config_int(var,value);
+	/* Also use aliases for command lookup */
+	if (!prefixcmp(var, "alias."))
+		add_cmdname(&aliases, var + 6, strlen(var + 6));
+
+	return perf_default_config(var, value, cb);
+}
+
+static int levenshtein_compare(const void *p1, const void *p2)
+{
+	const struct cmdname *const *c1 = p1, *const *c2 = p2;
+	const char *s1 = (*c1)->name, *s2 = (*c2)->name;
+	int l1 = (*c1)->len;
+	int l2 = (*c2)->len;
+	return l1 != l2 ? l1 - l2 : strcmp(s1, s2);
+}
+
+static void add_cmd_list(struct cmdnames *cmds, struct cmdnames *old)
+{
+	int i;
+	ALLOC_GROW(cmds->names, cmds->cnt + old->cnt, cmds->alloc);
+
+	for (i = 0; i < old->cnt; i++)
+		cmds->names[cmds->cnt++] = old->names[i];
+	free(old->names);
+	old->cnt = 0;
+	old->names = NULL;
+}
+
+const char *help_unknown_cmd(const char *cmd)
+{
+	int i, n = 0, best_similarity = 0;
+	struct cmdnames main_cmds, other_cmds;
+
+	memset(&main_cmds, 0, sizeof(main_cmds));
+	memset(&other_cmds, 0, sizeof(main_cmds));
+	memset(&aliases, 0, sizeof(aliases));
+
+	perf_config(perf_unknown_cmd_config, NULL);
+
+	load_command_list("perf-", &main_cmds, &other_cmds);
+
+	add_cmd_list(&main_cmds, &aliases);
+	add_cmd_list(&main_cmds, &other_cmds);
+	qsort(main_cmds.names, main_cmds.cnt,
+	      sizeof(main_cmds.names), cmdname_compare);
+	uniq(&main_cmds);
+
+	if (main_cmds.cnt) {
+		/* This reuses cmdname->len for similarity index */
+		for (i = 0; i < main_cmds.cnt; ++i)
+			main_cmds.names[i]->len =
+				levenshtein(cmd, main_cmds.names[i]->name, 0, 2, 1, 4);
+
+		qsort(main_cmds.names, main_cmds.cnt,
+		      sizeof(*main_cmds.names), levenshtein_compare);
+
+		best_similarity = main_cmds.names[0]->len;
+		n = 1;
+		while (n < main_cmds.cnt && best_similarity == main_cmds.names[n]->len)
+			++n;
+	}
+
+	if (autocorrect && n == 1) {
+		const char *assumed = main_cmds.names[0]->name;
+
+		main_cmds.names[0] = NULL;
+		clean_cmdnames(&main_cmds);
+		fprintf(stderr, "WARNING: You called a Git program named '%s', "
+			"which does not exist.\n"
+			"Continuing under the assumption that you meant '%s'\n",
+			cmd, assumed);
+		if (autocorrect > 0) {
+			fprintf(stderr, "in %0.1f seconds automatically...\n",
+				(float)autocorrect/10.0);
+			poll(NULL, 0, autocorrect * 100);
+		}
+		return assumed;
+	}
+
+	fprintf(stderr, "perf: '%s' is not a perf-command. See 'perf --help'.\n", cmd);
+
+	if (main_cmds.cnt && best_similarity < 6) {
+		fprintf(stderr, "\nDid you mean %s?\n",
+			n < 2 ? "this": "one of these");
+
+		for (i = 0; i < n; i++)
+			fprintf(stderr, "\t%s\n", main_cmds.names[i]->name);
+	}
+
+	exit(1);
+}
+
+int cmd_version(int argc, const char **argv, const char *prefix)
+{
+	printf("perf version %s\n", perf_version_string);
+	return 0;
+}
diff --git a/tools/perf/util/help.h b/tools/perf/util/help.h
new file mode 100644
index 000000000000..56bc15406ffc
--- /dev/null
+++ b/tools/perf/util/help.h
@@ -0,0 +1,29 @@
+#ifndef HELP_H
+#define HELP_H
+
+struct cmdnames {
+	int alloc;
+	int cnt;
+	struct cmdname {
+		size_t len; /* also used for similarity index in help.c */
+		char name[FLEX_ARRAY];
+	} **names;
+};
+
+static inline void mput_char(char c, unsigned int num)
+{
+	while(num--)
+		putchar(c);
+}
+
+void load_command_list(const char *prefix,
+		struct cmdnames *main_cmds,
+		struct cmdnames *other_cmds);
+void add_cmdname(struct cmdnames *cmds, const char *name, int 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 */
diff --git a/tools/perf/util/levenshtein.c b/tools/perf/util/levenshtein.c
new file mode 100644
index 000000000000..e521d1516df6
--- /dev/null
+++ b/tools/perf/util/levenshtein.c
@@ -0,0 +1,84 @@
+#include "cache.h"
+#include "levenshtein.h"
+
+/*
+ * This function implements the Damerau-Levenshtein algorithm to
+ * calculate a distance between strings.
+ *
+ * Basically, it says how many letters need to be swapped, substituted,
+ * deleted from, or added to string1, at least, to get string2.
+ *
+ * The idea is to build a distance matrix for the substrings of both
+ * strings.  To avoid a large space complexity, only the last three rows
+ * are kept in memory (if swaps had the same or higher cost as one deletion
+ * plus one insertion, only two rows would be needed).
+ *
+ * At any stage, "i + 1" denotes the length of the current substring of
+ * string1 that the distance is calculated for.
+ *
+ * row2 holds the current row, row1 the previous row (i.e. for the substring
+ * of string1 of length "i"), and row0 the row before that.
+ *
+ * In other words, at the start of the big loop, row2[j + 1] contains the
+ * Damerau-Levenshtein distance between the substring of string1 of length
+ * "i" and the substring of string2 of length "j + 1".
+ *
+ * All the big loop does is determine the partial minimum-cost paths.
+ *
+ * It does so by calculating the costs of the path ending in characters
+ * i (in string1) and j (in string2), respectively, given that the last
+ * operation is a substition, a swap, a deletion, or an insertion.
+ *
+ * This implementation allows the costs to be weighted:
+ *
+ * - w (as in "sWap")
+ * - s (as in "Substitution")
+ * - a (for insertion, AKA "Add")
+ * - d (as in "Deletion")
+ *
+ * Note that this algorithm calculates a distance _iff_ d == a.
+ */
+int levenshtein(const char *string1, const char *string2,
+		int w, int s, int a, int d)
+{
+	int len1 = strlen(string1), len2 = strlen(string2);
+	int *row0 = malloc(sizeof(int) * (len2 + 1));
+	int *row1 = malloc(sizeof(int) * (len2 + 1));
+	int *row2 = malloc(sizeof(int) * (len2 + 1));
+	int i, j;
+
+	for (j = 0; j <= len2; j++)
+		row1[j] = j * a;
+	for (i = 0; i < len1; i++) {
+		int *dummy;
+
+		row2[0] = (i + 1) * d;
+		for (j = 0; j < len2; j++) {
+			/* substitution */
+			row2[j + 1] = row1[j] + s * (string1[i] != string2[j]);
+			/* swap */
+			if (i > 0 && j > 0 && string1[i - 1] == string2[j] &&
+					string1[i] == string2[j - 1] &&
+					row2[j + 1] > row0[j - 1] + w)
+				row2[j + 1] = row0[j - 1] + w;
+			/* deletion */
+			if (row2[j + 1] > row1[j + 1] + d)
+				row2[j + 1] = row1[j + 1] + d;
+			/* insertion */
+			if (row2[j + 1] > row2[j] + a)
+				row2[j + 1] = row2[j] + a;
+		}
+
+		dummy = row0;
+		row0 = row1;
+		row1 = row2;
+		row2 = dummy;
+	}
+
+	i = row1[len2];
+	free(row0);
+	free(row1);
+	free(row2);
+
+	return i;
+}
diff --git a/tools/perf/util/levenshtein.h b/tools/perf/util/levenshtein.h
new file mode 100644
index 000000000000..0173abeef52c
--- /dev/null
+++ b/tools/perf/util/levenshtein.h
@@ -0,0 +1,8 @@
+#ifndef LEVENSHTEIN_H
+#define LEVENSHTEIN_H
+
+int levenshtein(const char *string1, const char *string2,
+	int swap_penalty, int substition_penalty,
+	int insertion_penalty, int deletion_penalty);
+
+#endif
diff --git a/tools/perf/util/list.h b/tools/perf/util/list.h
new file mode 100644
index 000000000000..e2548e8072cf
--- /dev/null
+++ b/tools/perf/util/list.h
@@ -0,0 +1,603 @@
+#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/pager.c b/tools/perf/util/pager.c
new file mode 100644
index 000000000000..a28bccae5458
--- /dev/null
+++ b/tools/perf/util/pager.c
@@ -0,0 +1,99 @@
+#include "cache.h"
+#include "run-command.h"
+#include "sigchain.h"
+
+/*
+ * This is split up from the rest of git so that we can do
+ * something different on Windows.
+ */
+
+static int spawned_pager;
+
+#ifndef __MINGW32__
+static void pager_preexec(void)
+{
+	/*
+	 * Work around bug in "less" by not starting it until we
+	 * have real input
+	 */
+	fd_set in;
+
+	FD_ZERO(&in);
+	FD_SET(0, &in);
+	select(1, &in, NULL, &in, NULL);
+
+	setenv("LESS", "FRSX", 0);
+}
+#endif
+
+static const char *pager_argv[] = { "sh", "-c", NULL, NULL };
+static struct child_process pager_process;
+
+static void wait_for_pager(void)
+{
+	fflush(stdout);
+	fflush(stderr);
+	/* signal EOF to pager */
+	close(1);
+	close(2);
+	finish_command(&pager_process);
+}
+
+static void wait_for_pager_signal(int signo)
+{
+	wait_for_pager();
+	sigchain_pop(signo);
+	raise(signo);
+}
+
+void setup_pager(void)
+{
+	const char *pager = getenv("PERF_PAGER");
+
+	if (!isatty(1))
+		return;
+	if (!pager) {
+		if (!pager_program)
+			perf_config(perf_default_config, NULL);
+		pager = pager_program;
+	}
+	if (!pager)
+		pager = getenv("PAGER");
+	if (!pager)
+		pager = "less";
+	else if (!*pager || !strcmp(pager, "cat"))
+		return;
+
+	spawned_pager = 1; /* means we are emitting to terminal */
+
+	/* spawn the pager */
+	pager_argv[2] = pager;
+	pager_process.argv = pager_argv;
+	pager_process.in = -1;
+#ifndef __MINGW32__
+	pager_process.preexec_cb = pager_preexec;
+#endif
+	if (start_command(&pager_process))
+		return;
+
+	/* original process continues, but writes to the pipe */
+	dup2(pager_process.in, 1);
+	if (isatty(2))
+		dup2(pager_process.in, 2);
+	close(pager_process.in);
+
+	/* this makes sure that the parent terminates after the pager */
+	sigchain_push_common(wait_for_pager_signal);
+	atexit(wait_for_pager);
+}
+
+int pager_in_use(void)
+{
+	const char *env;
+
+	if (spawned_pager)
+		return 1;
+
+	env = getenv("PERF_PAGER_IN_USE");
+	return env ? perf_config_bool("PERF_PAGER_IN_USE", env) : 0;
+}
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
new file mode 100644
index 000000000000..5a72586e1df0
--- /dev/null
+++ b/tools/perf/util/parse-events.c
@@ -0,0 +1,316 @@
+
+#include "../perf.h"
+#include "util.h"
+#include "parse-options.h"
+#include "parse-events.h"
+#include "exec_cmd.h"
+#include "string.h"
+
+extern char *strcasestr(const char *haystack, const char *needle);
+
+int					nr_counters;
+
+struct perf_counter_attr		attrs[MAX_COUNTERS];
+
+struct event_symbol {
+	__u8	type;
+	__u64	config;
+	char	*symbol;
+};
+
+#define C(x, y) .type = PERF_TYPE_##x, .config = PERF_COUNT_##y
+#define CR(x, y) .type = PERF_TYPE_##x, .config = y
+
+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",		},
+};
+
+#define __PERF_COUNTER_FIELD(config, name) \
+	((config & PERF_COUNTER_##name##_MASK) >> PERF_COUNTER_##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)
+
+static char *hw_event_names[] = {
+	"cycles",
+	"instructions",
+	"cache-references",
+	"cache-misses",
+	"branches",
+	"branch-misses",
+	"bus-cycles",
+};
+
+static char *sw_event_names[] = {
+	"cpu-clock-ticks",
+	"task-clock-ticks",
+	"page-faults",
+	"context-switches",
+	"CPU-migrations",
+	"minor-faults",
+	"major-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 char *hw_cache_op [][MAX_ALIASES] = {
+	{ "Load"		, "read"					},
+	{ "Store"		, "write"					},
+	{ "Prefetch"		, "speculative-read", "speculative-load"	},
+};
+
+static char *hw_cache_result [][MAX_ALIASES] = {
+	{ "Reference"		, "ops", "access"				},
+	{ "Miss"								},
+};
+
+char *event_name(int counter)
+{
+	__u64 config = attrs[counter].config;
+	int type = attrs[counter].type;
+	static char buf[32];
+
+	if (attrs[counter].type == PERF_TYPE_RAW) {
+		sprintf(buf, "raw 0x%llx", config);
+		return buf;
+	}
+
+	switch (type) {
+	case PERF_TYPE_HARDWARE:
+		if (config < PERF_COUNT_HW_MAX)
+			return hw_event_names[config];
+		return "unknown-hardware";
+
+	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)
+			return "unknown-ext-hardware-cache-type";
+
+		cache_op     = (config >>  8) & 0xff;
+		if (cache_op > PERF_COUNT_HW_CACHE_OP_MAX)
+			return "unknown-ext-hardware-cache-op";
+
+		cache_result = (config >> 16) & 0xff;
+		if (cache_result > PERF_COUNT_HW_CACHE_RESULT_MAX)
+			return "unknown-ext-hardware-cache-result";
+
+		sprintf(name, "%s-Cache-%s-%ses",
+			hw_cache[cache_type][0],
+			hw_cache_op[cache_op][0],
+			hw_cache_result[cache_result][0]);
+
+		return name;
+	}
+
+	case PERF_TYPE_SOFTWARE:
+		if (config < PERF_COUNT_SW_MAX)
+			return sw_event_names[config];
+		return "unknown-software";
+
+	default:
+		break;
+	}
+
+	return "unknown";
+}
+
+static int parse_aliases(const char *str, char *names[][MAX_ALIASES], int size)
+{
+	int i, j;
+
+	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;
+		}
+	}
+
+	return -1;
+}
+
+static int parse_generic_hw_symbols(const char *str, struct perf_counter_attr *attr)
+{
+	int cache_type = -1, cache_op = 0, cache_result = 0;
+
+	cache_type = parse_aliases(str, 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;
+
+	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:
+	 */
+	if (cache_result == -1)
+		cache_result = PERF_COUNT_HW_CACHE_RESULT_ACCESS;
+
+	attr->config = cache_type | (cache_op << 8) | (cache_result << 16);
+	attr->type = PERF_TYPE_HW_CACHE;
+
+	return 0;
+}
+
+/*
+ * 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)
+{
+	__u64 config, id;
+	int type;
+	unsigned int i;
+	const char *sep, *pstr;
+
+	if (str[0] == 'r' && hex2u64(str + 1, &config) > 0) {
+		attr->type = PERF_TYPE_RAW;
+		attr->config = config;
+
+		return 0;
+	}
+
+	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;
+
+		return 0;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(event_symbols); i++) {
+		if (!strncmp(str, event_symbols[i].symbol,
+			     strlen(event_symbols[i].symbol))) {
+
+			attr->type = event_symbols[i].type;
+			attr->config = event_symbols[i].config;
+
+			return 0;
+		}
+	}
+
+	return parse_generic_hw_symbols(str, attr);
+}
+
+int parse_events(const struct option *opt, const char *str, int unset)
+{
+	struct perf_counter_attr attr;
+	int ret;
+
+	memset(&attr, 0, sizeof(attr));
+again:
+	if (nr_counters == MAX_COUNTERS)
+		return -1;
+
+	ret = parse_event_symbols(str, &attr);
+	if (ret < 0)
+		return ret;
+
+	attrs[nr_counters] = attr;
+	nr_counters++;
+
+	str = strstr(str, ",");
+	if (str) {
+		str++;
+		goto again;
+	}
+
+	return 0;
+}
+
+static const char * const event_type_descriptors[] = {
+	"",
+	"Hardware event",
+	"Software event",
+	"Tracepoint event",
+	"Hardware cache event",
+};
+
+/*
+ * 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;
+
+	fprintf(stderr, "\n");
+	fprintf(stderr, "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;
+
+		if (type != prev_type)
+			fprintf(stderr, "\n");
+
+		fprintf(stderr, "  %-30s [%s]\n", syms->symbol,
+			event_type_descriptors[type]);
+
+		prev_type = type;
+	}
+
+	fprintf(stderr, "\n");
+	fprintf(stderr, "  %-30s [raw hardware event descriptor]\n",
+		"rNNN");
+	fprintf(stderr, "\n");
+
+	exit(129);
+}
diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h
new file mode 100644
index 000000000000..e3d552908e60
--- /dev/null
+++ b/tools/perf/util/parse-events.h
@@ -0,0 +1,17 @@
+
+/*
+ * Parse symbolic events/counts passed in as options:
+ */
+
+extern int			nr_counters;
+
+extern struct perf_counter_attr attrs[MAX_COUNTERS];
+
+extern char *event_name(int ctr);
+
+extern int parse_events(const struct option *opt, const char *str, int unset);
+
+#define EVENTS_HELP_MAX (128*1024)
+
+extern void print_events(void);
+
diff --git a/tools/perf/util/parse-options.c b/tools/perf/util/parse-options.c
new file mode 100644
index 000000000000..b3affb1658d2
--- /dev/null
+++ b/tools/perf/util/parse-options.c
@@ -0,0 +1,508 @@
+#include "util.h"
+#include "parse-options.h"
+#include "cache.h"
+
+#define OPT_SHORT 1
+#define OPT_UNSET 2
+
+static int opterror(const struct option *opt, const char *reason, int flags)
+{
+	if (flags & OPT_SHORT)
+		return error("switch `%c' %s", opt->short_name, reason);
+	if (flags & OPT_UNSET)
+		return error("option `no-%s' %s", opt->long_name, reason);
+	return error("option `%s' %s", opt->long_name, reason);
+}
+
+static int get_arg(struct parse_opt_ctx_t *p, const struct option *opt,
+		   int flags, const char **arg)
+{
+	if (p->opt) {
+		*arg = p->opt;
+		p->opt = NULL;
+	} else if (p->argc == 1 && (opt->flags & PARSE_OPT_LASTARG_DEFAULT)) {
+		*arg = (const char *)opt->defval;
+	} else if (p->argc > 1) {
+		p->argc--;
+		*arg = *++p->argv;
+	} else
+		return opterror(opt, "requires a value", flags);
+	return 0;
+}
+
+static int get_value(struct parse_opt_ctx_t *p,
+		     const struct option *opt, int flags)
+{
+	const char *s, *arg = NULL;
+	const int unset = flags & OPT_UNSET;
+
+	if (unset && p->opt)
+		return opterror(opt, "takes no value", flags);
+	if (unset && (opt->flags & PARSE_OPT_NONEG))
+		return opterror(opt, "isn't available", flags);
+
+	if (!(flags & OPT_SHORT) && p->opt) {
+		switch (opt->type) {
+		case OPTION_CALLBACK:
+			if (!(opt->flags & PARSE_OPT_NOARG))
+				break;
+			/* FALLTHROUGH */
+		case OPTION_BOOLEAN:
+		case OPTION_BIT:
+		case OPTION_SET_INT:
+		case OPTION_SET_PTR:
+			return opterror(opt, "takes no value", flags);
+		default:
+			break;
+		}
+	}
+
+	switch (opt->type) {
+	case OPTION_BIT:
+		if (unset)
+			*(int *)opt->value &= ~opt->defval;
+		else
+			*(int *)opt->value |= opt->defval;
+		return 0;
+
+	case OPTION_BOOLEAN:
+		*(int *)opt->value = unset ? 0 : *(int *)opt->value + 1;
+		return 0;
+
+	case OPTION_SET_INT:
+		*(int *)opt->value = unset ? 0 : opt->defval;
+		return 0;
+
+	case OPTION_SET_PTR:
+		*(void **)opt->value = unset ? NULL : (void *)opt->defval;
+		return 0;
+
+	case OPTION_STRING:
+		if (unset)
+			*(const char **)opt->value = NULL;
+		else if (opt->flags & PARSE_OPT_OPTARG && !p->opt)
+			*(const char **)opt->value = (const char *)opt->defval;
+		else
+			return get_arg(p, opt, flags, (const char **)opt->value);
+		return 0;
+
+	case OPTION_CALLBACK:
+		if (unset)
+			return (*opt->callback)(opt, NULL, 1) ? (-1) : 0;
+		if (opt->flags & PARSE_OPT_NOARG)
+			return (*opt->callback)(opt, NULL, 0) ? (-1) : 0;
+		if (opt->flags & PARSE_OPT_OPTARG && !p->opt)
+			return (*opt->callback)(opt, NULL, 0) ? (-1) : 0;
+		if (get_arg(p, opt, flags, &arg))
+			return -1;
+		return (*opt->callback)(opt, arg, 0) ? (-1) : 0;
+
+	case OPTION_INTEGER:
+		if (unset) {
+			*(int *)opt->value = 0;
+			return 0;
+		}
+		if (opt->flags & PARSE_OPT_OPTARG && !p->opt) {
+			*(int *)opt->value = opt->defval;
+			return 0;
+		}
+		if (get_arg(p, opt, flags, &arg))
+			return -1;
+		*(int *)opt->value = strtol(arg, (char **)&s, 10);
+		if (*s)
+			return opterror(opt, "expects a numerical value", flags);
+		return 0;
+
+	case OPTION_LONG:
+		if (unset) {
+			*(long *)opt->value = 0;
+			return 0;
+		}
+		if (opt->flags & PARSE_OPT_OPTARG && !p->opt) {
+			*(long *)opt->value = opt->defval;
+			return 0;
+		}
+		if (get_arg(p, opt, flags, &arg))
+			return -1;
+		*(long *)opt->value = strtol(arg, (char **)&s, 10);
+		if (*s)
+			return opterror(opt, "expects a numerical value", flags);
+		return 0;
+
+	default:
+		die("should not happen, someone must be hit on the forehead");
+	}
+}
+
+static int parse_short_opt(struct parse_opt_ctx_t *p, const struct option *options)
+{
+	for (; options->type != OPTION_END; options++) {
+		if (options->short_name == *p->opt) {
+			p->opt = p->opt[1] ? p->opt + 1 : NULL;
+			return get_value(p, options, OPT_SHORT);
+		}
+	}
+	return -2;
+}
+
+static int parse_long_opt(struct parse_opt_ctx_t *p, const char *arg,
+                          const struct option *options)
+{
+	const char *arg_end = strchr(arg, '=');
+	const struct option *abbrev_option = NULL, *ambiguous_option = NULL;
+	int abbrev_flags = 0, ambiguous_flags = 0;
+
+	if (!arg_end)
+		arg_end = arg + strlen(arg);
+
+	for (; options->type != OPTION_END; options++) {
+		const char *rest;
+		int flags = 0;
+
+		if (!options->long_name)
+			continue;
+
+		rest = skip_prefix(arg, options->long_name);
+		if (options->type == OPTION_ARGUMENT) {
+			if (!rest)
+				continue;
+			if (*rest == '=')
+				return opterror(options, "takes no value", flags);
+			if (*rest)
+				continue;
+			p->out[p->cpidx++] = arg - 2;
+			return 0;
+		}
+		if (!rest) {
+			/* abbreviated? */
+			if (!strncmp(options->long_name, arg, arg_end - arg)) {
+is_abbreviated:
+				if (abbrev_option) {
+					/*
+					 * If this is abbreviated, it is
+					 * ambiguous. So when there is no
+					 * exact match later, we need to
+					 * error out.
+					 */
+					ambiguous_option = abbrev_option;
+					ambiguous_flags = abbrev_flags;
+				}
+				if (!(flags & OPT_UNSET) && *arg_end)
+					p->opt = arg_end + 1;
+				abbrev_option = options;
+				abbrev_flags = flags;
+				continue;
+			}
+			/* negated and abbreviated very much? */
+			if (!prefixcmp("no-", arg)) {
+				flags |= OPT_UNSET;
+				goto is_abbreviated;
+			}
+			/* negated? */
+			if (strncmp(arg, "no-", 3))
+				continue;
+			flags |= OPT_UNSET;
+			rest = skip_prefix(arg + 3, options->long_name);
+			/* abbreviated and negated? */
+			if (!rest && !prefixcmp(options->long_name, arg + 3))
+				goto is_abbreviated;
+			if (!rest)
+				continue;
+		}
+		if (*rest) {
+			if (*rest != '=')
+				continue;
+			p->opt = rest + 1;
+		}
+		return get_value(p, options, flags);
+	}
+
+	if (ambiguous_option)
+		return error("Ambiguous option: %s "
+			"(could be --%s%s or --%s%s)",
+			arg,
+			(ambiguous_flags & OPT_UNSET) ?  "no-" : "",
+			ambiguous_option->long_name,
+			(abbrev_flags & OPT_UNSET) ?  "no-" : "",
+			abbrev_option->long_name);
+	if (abbrev_option)
+		return get_value(p, abbrev_option, abbrev_flags);
+	return -2;
+}
+
+static void check_typos(const char *arg, const struct option *options)
+{
+	if (strlen(arg) < 3)
+		return;
+
+	if (!prefixcmp(arg, "no-")) {
+		error ("did you mean `--%s` (with two dashes ?)", arg);
+		exit(129);
+	}
+
+	for (; options->type != OPTION_END; options++) {
+		if (!options->long_name)
+			continue;
+		if (!prefixcmp(options->long_name, arg)) {
+			error ("did you mean `--%s` (with two dashes ?)", arg);
+			exit(129);
+		}
+	}
+}
+
+void parse_options_start(struct parse_opt_ctx_t *ctx,
+			 int argc, const char **argv, int flags)
+{
+	memset(ctx, 0, sizeof(*ctx));
+	ctx->argc = argc - 1;
+	ctx->argv = argv + 1;
+	ctx->out  = argv;
+	ctx->cpidx = ((flags & PARSE_OPT_KEEP_ARGV0) != 0);
+	ctx->flags = flags;
+	if ((flags & PARSE_OPT_KEEP_UNKNOWN) &&
+	    (flags & PARSE_OPT_STOP_AT_NON_OPTION))
+		die("STOP_AT_NON_OPTION and KEEP_UNKNOWN don't go together");
+}
+
+static int usage_with_options_internal(const char * const *,
+				       const struct option *, int);
+
+int parse_options_step(struct parse_opt_ctx_t *ctx,
+		       const struct option *options,
+		       const char * const usagestr[])
+{
+	int internal_help = !(ctx->flags & PARSE_OPT_NO_INTERNAL_HELP);
+
+	/* we must reset ->opt, unknown short option leave it dangling */
+	ctx->opt = NULL;
+
+	for (; ctx->argc; ctx->argc--, ctx->argv++) {
+		const char *arg = ctx->argv[0];
+
+		if (*arg != '-' || !arg[1]) {
+			if (ctx->flags & PARSE_OPT_STOP_AT_NON_OPTION)
+				break;
+			ctx->out[ctx->cpidx++] = ctx->argv[0];
+			continue;
+		}
+
+		if (arg[1] != '-') {
+			ctx->opt = arg + 1;
+			if (internal_help && *ctx->opt == 'h')
+				return parse_options_usage(usagestr, options);
+			switch (parse_short_opt(ctx, options)) {
+			case -1:
+				return parse_options_usage(usagestr, options);
+			case -2:
+				goto unknown;
+			}
+			if (ctx->opt)
+				check_typos(arg + 1, options);
+			while (ctx->opt) {
+				if (internal_help && *ctx->opt == 'h')
+					return parse_options_usage(usagestr, options);
+				switch (parse_short_opt(ctx, options)) {
+				case -1:
+					return parse_options_usage(usagestr, options);
+				case -2:
+					/* fake a short option thing to hide the fact that we may have
+					 * started to parse aggregated stuff
+					 *
+					 * This is leaky, too bad.
+					 */
+					ctx->argv[0] = strdup(ctx->opt - 1);
+					*(char *)ctx->argv[0] = '-';
+					goto unknown;
+				}
+			}
+			continue;
+		}
+
+		if (!arg[2]) { /* "--" */
+			if (!(ctx->flags & PARSE_OPT_KEEP_DASHDASH)) {
+				ctx->argc--;
+				ctx->argv++;
+			}
+			break;
+		}
+
+		if (internal_help && !strcmp(arg + 2, "help-all"))
+			return usage_with_options_internal(usagestr, options, 1);
+		if (internal_help && !strcmp(arg + 2, "help"))
+			return parse_options_usage(usagestr, options);
+		switch (parse_long_opt(ctx, arg + 2, options)) {
+		case -1:
+			return parse_options_usage(usagestr, options);
+		case -2:
+			goto unknown;
+		}
+		continue;
+unknown:
+		if (!(ctx->flags & PARSE_OPT_KEEP_UNKNOWN))
+			return PARSE_OPT_UNKNOWN;
+		ctx->out[ctx->cpidx++] = ctx->argv[0];
+		ctx->opt = NULL;
+	}
+	return PARSE_OPT_DONE;
+}
+
+int parse_options_end(struct parse_opt_ctx_t *ctx)
+{
+	memmove(ctx->out + ctx->cpidx, ctx->argv, ctx->argc * sizeof(*ctx->out));
+	ctx->out[ctx->cpidx + ctx->argc] = NULL;
+	return ctx->cpidx + ctx->argc;
+}
+
+int parse_options(int argc, const char **argv, const struct option *options,
+		  const char * const usagestr[], int flags)
+{
+	struct parse_opt_ctx_t ctx;
+
+	parse_options_start(&ctx, argc, argv, flags);
+	switch (parse_options_step(&ctx, options, usagestr)) {
+	case PARSE_OPT_HELP:
+		exit(129);
+	case PARSE_OPT_DONE:
+		break;
+	default: /* PARSE_OPT_UNKNOWN */
+		if (ctx.argv[0][1] == '-') {
+			error("unknown option `%s'", ctx.argv[0] + 2);
+		} else {
+			error("unknown switch `%c'", *ctx.opt);
+		}
+		usage_with_options(usagestr, options);
+	}
+
+	return parse_options_end(&ctx);
+}
+
+#define USAGE_OPTS_WIDTH 24
+#define USAGE_GAP         2
+
+int usage_with_options_internal(const char * const *usagestr,
+				const struct option *opts, int full)
+{
+	if (!usagestr)
+		return PARSE_OPT_HELP;
+
+	fprintf(stderr, "\n usage: %s\n", *usagestr++);
+	while (*usagestr && **usagestr)
+		fprintf(stderr, "    or: %s\n", *usagestr++);
+	while (*usagestr) {
+		fprintf(stderr, "%s%s\n",
+				**usagestr ? "    " : "",
+				*usagestr);
+		usagestr++;
+	}
+
+	if (opts->type != OPTION_GROUP)
+		fputc('\n', stderr);
+
+	for (; opts->type != OPTION_END; opts++) {
+		size_t pos;
+		int pad;
+
+		if (opts->type == OPTION_GROUP) {
+			fputc('\n', stderr);
+			if (*opts->help)
+				fprintf(stderr, "%s\n", opts->help);
+			continue;
+		}
+		if (!full && (opts->flags & PARSE_OPT_HIDDEN))
+			continue;
+
+		pos = fprintf(stderr, "    ");
+		if (opts->short_name)
+			pos += fprintf(stderr, "-%c", opts->short_name);
+		if (opts->long_name && opts->short_name)
+			pos += fprintf(stderr, ", ");
+		if (opts->long_name)
+			pos += fprintf(stderr, "--%s", opts->long_name);
+
+		switch (opts->type) {
+		case OPTION_ARGUMENT:
+			break;
+		case OPTION_INTEGER:
+			if (opts->flags & PARSE_OPT_OPTARG)
+				if (opts->long_name)
+					pos += fprintf(stderr, "[=<n>]");
+				else
+					pos += fprintf(stderr, "[<n>]");
+			else
+				pos += fprintf(stderr, " <n>");
+			break;
+		case OPTION_CALLBACK:
+			if (opts->flags & PARSE_OPT_NOARG)
+				break;
+			/* FALLTHROUGH */
+		case OPTION_STRING:
+			if (opts->argh) {
+				if (opts->flags & PARSE_OPT_OPTARG)
+					if (opts->long_name)
+						pos += fprintf(stderr, "[=<%s>]", opts->argh);
+					else
+						pos += fprintf(stderr, "[<%s>]", opts->argh);
+				else
+					pos += fprintf(stderr, " <%s>", opts->argh);
+			} else {
+				if (opts->flags & PARSE_OPT_OPTARG)
+					if (opts->long_name)
+						pos += fprintf(stderr, "[=...]");
+					else
+						pos += fprintf(stderr, "[...]");
+				else
+					pos += fprintf(stderr, " ...");
+			}
+			break;
+		default: /* OPTION_{BIT,BOOLEAN,SET_INT,SET_PTR} */
+			break;
+		}
+
+		if (pos <= USAGE_OPTS_WIDTH)
+			pad = USAGE_OPTS_WIDTH - pos;
+		else {
+			fputc('\n', stderr);
+			pad = USAGE_OPTS_WIDTH;
+		}
+		fprintf(stderr, "%*s%s\n", pad + USAGE_GAP, "", opts->help);
+	}
+	fputc('\n', stderr);
+
+	return PARSE_OPT_HELP;
+}
+
+void usage_with_options(const char * const *usagestr,
+			const struct option *opts)
+{
+	usage_with_options_internal(usagestr, opts, 0);
+	exit(129);
+}
+
+int parse_options_usage(const char * const *usagestr,
+			const struct option *opts)
+{
+	return usage_with_options_internal(usagestr, opts, 0);
+}
+
+
+int parse_opt_verbosity_cb(const struct option *opt, const char *arg,
+			   int unset)
+{
+	int *target = opt->value;
+
+	if (unset)
+		/* --no-quiet, --no-verbose */
+		*target = 0;
+	else if (opt->short_name == 'v') {
+		if (*target >= 0)
+			(*target)++;
+		else
+			*target = 1;
+	} else {
+		if (*target <= 0)
+			(*target)--;
+		else
+			*target = -1;
+	}
+	return 0;
+}
diff --git a/tools/perf/util/parse-options.h b/tools/perf/util/parse-options.h
new file mode 100644
index 000000000000..a1039a6ce0eb
--- /dev/null
+++ b/tools/perf/util/parse-options.h
@@ -0,0 +1,174 @@
+#ifndef PARSE_OPTIONS_H
+#define PARSE_OPTIONS_H
+
+enum parse_opt_type {
+	/* special types */
+	OPTION_END,
+	OPTION_ARGUMENT,
+	OPTION_GROUP,
+	/* options with no arguments */
+	OPTION_BIT,
+	OPTION_BOOLEAN, /* _INCR would have been a better name */
+	OPTION_SET_INT,
+	OPTION_SET_PTR,
+	/* options with arguments (usually) */
+	OPTION_STRING,
+	OPTION_INTEGER,
+	OPTION_LONG,
+	OPTION_CALLBACK,
+};
+
+enum parse_opt_flags {
+	PARSE_OPT_KEEP_DASHDASH = 1,
+	PARSE_OPT_STOP_AT_NON_OPTION = 2,
+	PARSE_OPT_KEEP_ARGV0 = 4,
+	PARSE_OPT_KEEP_UNKNOWN = 8,
+	PARSE_OPT_NO_INTERNAL_HELP = 16,
+};
+
+enum parse_opt_option_flags {
+	PARSE_OPT_OPTARG  = 1,
+	PARSE_OPT_NOARG   = 2,
+	PARSE_OPT_NONEG   = 4,
+	PARSE_OPT_HIDDEN  = 8,
+	PARSE_OPT_LASTARG_DEFAULT = 16,
+};
+
+struct option;
+typedef int parse_opt_cb(const struct option *, const char *arg, int unset);
+
+/*
+ * `type`::
+ *   holds the type of the option, you must have an OPTION_END last in your
+ *   array.
+ *
+ * `short_name`::
+ *   the character to use as a short option name, '\0' if none.
+ *
+ * `long_name`::
+ *   the long option name, without the leading dashes, NULL if none.
+ *
+ * `value`::
+ *   stores pointers to the values to be filled.
+ *
+ * `argh`::
+ *   token to explain the kind of argument this option wants. Keep it
+ *   homogenous across the repository.
+ *
+ * `help`::
+ *   the short help associated to what the option does.
+ *   Must never be NULL (except for OPTION_END).
+ *   OPTION_GROUP uses this pointer to store the group header.
+ *
+ * `flags`::
+ *   mask of parse_opt_option_flags.
+ *   PARSE_OPT_OPTARG: says that the argument is optionnal (not for BOOLEANs)
+ *   PARSE_OPT_NOARG: says that this option takes no argument, for CALLBACKs
+ *   PARSE_OPT_NONEG: says that this option cannot be negated
+ *   PARSE_OPT_HIDDEN this option is skipped in the default usage, showed in
+ *                    the long one.
+ *
+ * `callback`::
+ *   pointer to the callback to use for OPTION_CALLBACK.
+ *
+ * `defval`::
+ *   default value to fill (*->value) with for PARSE_OPT_OPTARG.
+ *   OPTION_{BIT,SET_INT,SET_PTR} store the {mask,integer,pointer} to put in
+ *   the value when met.
+ *   CALLBACKS can use it like they want.
+ */
+struct option {
+	enum parse_opt_type type;
+	int short_name;
+	const char *long_name;
+	void *value;
+	const char *argh;
+	const char *help;
+
+	int flags;
+	parse_opt_cb *callback;
+	intptr_t defval;
+};
+
+#define 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 OPT_DATE(s, l, v, h) \
+	{ OPTION_CALLBACK, (s), (l), (v), "time",(h), 0, \
+	  parse_opt_approxidate_cb }
+#define OPT_CALLBACK(s, l, v, a, h, f) \
+	{ OPTION_CALLBACK, (s), (l), (v), (a), (h), 0, (f) }
+
+/* parse_options() will filter out the processed options and leave the
+ * non-option argments in argv[].
+ * Returns the number of arguments left in argv[].
+ */
+extern int parse_options(int argc, const char **argv,
+                         const struct option *options,
+                         const char * const usagestr[], int flags);
+
+extern NORETURN void usage_with_options(const char * const *usagestr,
+                                        const struct option *options);
+
+/*----- incremantal advanced APIs -----*/
+
+enum {
+	PARSE_OPT_HELP = -1,
+	PARSE_OPT_DONE,
+	PARSE_OPT_UNKNOWN,
+};
+
+/*
+ * It's okay for the caller to consume argv/argc in the usual way.
+ * Other fields of that structure are private to parse-options and should not
+ * be modified in any way.
+ */
+struct parse_opt_ctx_t {
+	const char **argv;
+	const char **out;
+	int argc, cpidx;
+	const char *opt;
+	int flags;
+};
+
+extern int parse_options_usage(const char * const *usagestr,
+			       const struct option *opts);
+
+extern void parse_options_start(struct parse_opt_ctx_t *ctx,
+				int argc, const char **argv, int flags);
+
+extern int parse_options_step(struct parse_opt_ctx_t *ctx,
+			      const struct option *options,
+			      const char * const usagestr[]);
+
+extern int parse_options_end(struct parse_opt_ctx_t *ctx);
+
+
+/*----- some often used options -----*/
+extern int parse_opt_abbrev_cb(const struct option *, const char *, int);
+extern int parse_opt_approxidate_cb(const struct option *, const char *, int);
+extern int parse_opt_verbosity_cb(const struct option *, const char *, int);
+
+#define OPT__VERBOSE(var)  OPT_BOOLEAN('v', "verbose", (var), "be verbose")
+#define OPT__QUIET(var)    OPT_BOOLEAN('q', "quiet",   (var), "be quiet")
+#define OPT__VERBOSITY(var) \
+	{ OPTION_CALLBACK, 'v', "verbose", (var), NULL, "be more verbose", \
+	  PARSE_OPT_NOARG, &parse_opt_verbosity_cb, 0 }, \
+	{ OPTION_CALLBACK, 'q', "quiet", (var), NULL, "be more quiet", \
+	  PARSE_OPT_NOARG, &parse_opt_verbosity_cb, 0 }
+#define OPT__DRY_RUN(var)  OPT_BOOLEAN('n', "dry-run", (var), "dry run")
+#define OPT__ABBREV(var)  \
+	{ OPTION_CALLBACK, 0, "abbrev", (var), "n", \
+	  "use <n> digits to display SHA-1s", \
+	  PARSE_OPT_OPTARG, &parse_opt_abbrev_cb, 0 }
+
+extern const char *parse_options_fix_filename(const char *prefix, const char *file);
+
+#endif
diff --git a/tools/perf/util/path.c b/tools/perf/util/path.c
new file mode 100644
index 000000000000..a501a40dd2cb
--- /dev/null
+++ b/tools/perf/util/path.c
@@ -0,0 +1,353 @@
+/*
+ * I'm tired of doing "vsnprintf()" etc just to open a
+ * file, so here's a "return static buffer with printf"
+ * interface for paths.
+ *
+ * It's obviously not thread-safe. Sue me. But it's quite
+ * useful for doing things like
+ *
+ *   f = open(mkpath("%s/%s.perf", base, name), O_RDONLY);
+ *
+ * which is what it's designed for.
+ */
+#include "cache.h"
+
+static char bad_path[] = "/bad-path/";
+/*
+ * Two hacks:
+ */
+
+static char *get_perf_dir(void)
+{
+	return ".";
+}
+
+size_t strlcpy(char *dest, const char *src, size_t size)
+{
+	size_t ret = strlen(src);
+
+	if (size) {
+		size_t len = (ret >= size) ? size - 1 : ret;
+		memcpy(dest, src, len);
+		dest[len] = '\0';
+	}
+	return ret;
+}
+
+
+static char *get_pathname(void)
+{
+	static char pathname_array[4][PATH_MAX];
+	static int index;
+	return pathname_array[3 & ++index];
+}
+
+static char *cleanup_path(char *path)
+{
+	/* Clean it up */
+	if (!memcmp(path, "./", 2)) {
+		path += 2;
+		while (*path == '/')
+			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();
+	size_t len;
+
+	len = strlen(perf_dir);
+	if (n < len + 1)
+		goto bad;
+	memcpy(buf, perf_dir, len);
+	if (len && !is_dir_sep(perf_dir[len-1]))
+		buf[len++] = '/';
+	len += vsnprintf(buf + len, n - len, fmt, args);
+	if (len >= n)
+		goto bad;
+	return cleanup_path(buf);
+bad:
+	strlcpy(buf, bad_path, n);
+	return buf;
+}
+
+char *perf_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];
+	va_list args;
+	va_start(args, fmt);
+	(void)perf_vsnpath(path, sizeof(path), fmt, args);
+	va_end(args);
+	return xstrdup(path);
+}
+
+char *mkpath(const char *fmt, ...)
+{
+	va_list args;
+	unsigned len;
+	char *pathname = get_pathname();
+
+	va_start(args, fmt);
+	len = vsnprintf(pathname, PATH_MAX, fmt, args);
+	va_end(args);
+	if (len >= PATH_MAX)
+		return bad_path;
+	return cleanup_path(pathname);
+}
+
+char *perf_path(const char *fmt, ...)
+{
+	const char *perf_dir = get_perf_dir();
+	char *pathname = get_pathname();
+	va_list args;
+	unsigned len;
+
+	len = strlen(perf_dir);
+	if (len > PATH_MAX-100)
+		return bad_path;
+	memcpy(pathname, perf_dir, len);
+	if (len && perf_dir[len-1] != '/')
+		pathname[len++] = '/';
+	va_start(args, fmt);
+	len += vsnprintf(pathname + len, PATH_MAX - len, fmt, args);
+	va_end(args);
+	if (len >= PATH_MAX)
+		return bad_path;
+	return cleanup_path(pathname);
+}
+
+
+/* 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)
+{
+	while (len && is_dir_sep(path[len - 1]))
+		len--;
+	return len;
+}
+
+/*
+ * If path ends with suffix (complete path components), returns the
+ * part before suffix (sans trailing directory separators).
+ * Otherwise returns NULL.
+ */
+char *strip_path_suffix(const char *path, const char *suffix)
+{
+	int path_len = strlen(path), suffix_len = strlen(suffix);
+
+	while (suffix_len) {
+		if (!path_len)
+			return NULL;
+
+		if (is_dir_sep(path[path_len - 1])) {
+			if (!is_dir_sep(suffix[suffix_len - 1]))
+				return NULL;
+			path_len = chomp_trailing_dir_sep(path, path_len);
+			suffix_len = chomp_trailing_dir_sep(suffix, suffix_len);
+		}
+		else if (path[--path_len] != suffix[--suffix_len])
+			return NULL;
+	}
+
+	if (path_len && !is_dir_sep(path[path_len - 1]))
+		return NULL;
+	return xstrndup(path, chomp_trailing_dir_sep(path, path_len));
+}
diff --git a/tools/perf/util/quote.c b/tools/perf/util/quote.c
new file mode 100644
index 000000000000..f18c5212bc92
--- /dev/null
+++ b/tools/perf/util/quote.c
@@ -0,0 +1,481 @@
+#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
+ *
+ * E.g.
+ *  original     sq_quote     result
+ *  name     ==> name      ==> 'name'
+ *  a b      ==> a b       ==> 'a b'
+ *  a'b      ==> a'\''b    ==> 'a'\''b'
+ *  a!b      ==> a'\!'b    ==> 'a'\!'b'
+ */
+static inline int need_bs_quote(char c)
+{
+	return (c == '\'' || c == '!');
+}
+
+void sq_quote_buf(struct strbuf *dst, const char *src)
+{
+	char *to_free = NULL;
+
+	if (dst->buf == src)
+		to_free = strbuf_detach(dst, NULL);
+
+	strbuf_addch(dst, '\'');
+	while (*src) {
+		size_t len = strcspn(src, "'!");
+		strbuf_add(dst, src, len);
+		src += len;
+		while (need_bs_quote(*src)) {
+			strbuf_addstr(dst, "'\\");
+			strbuf_addch(dst, *src++);
+			strbuf_addch(dst, '\'');
+		}
+	}
+	strbuf_addch(dst, '\'');
+	free(to_free);
+}
+
+void sq_quote_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;
+
+	/* Copy into destination buffer. */
+	strbuf_grow(dst, 255);
+	for (i = 0; argv[i]; ++i) {
+		strbuf_addch(dst, ' ');
+		sq_quote_buf(dst, argv[i]);
+		if (maxlen && dst->len > maxlen)
+			die("Too many or long arguments");
+	}
+}
+
+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
new file mode 100644
index 000000000000..5dfad89816db
--- /dev/null
+++ b/tools/perf/util/quote.h
@@ -0,0 +1,68 @@
+#ifndef QUOTE_H
+#define QUOTE_H
+
+#include <stddef.h>
+#include <stdio.h>
+
+/* Help to copy the thing properly quoted for the shell safety.
+ * any single quote is replaced with '\'', any exclamation point
+ * is replaced with '\!', and the whole thing is enclosed in a
+ * single quote pair.
+ *
+ * For example, if you are passing the result to system() as an
+ * argument:
+ *
+ * sprintf(cmd, "foobar %s %s", sq_quote(arg0), sq_quote(arg1))
+ *
+ * would be appropriate.  If the system() is going to call ssh to
+ * run the command on the other side:
+ *
+ * sprintf(cmd, "git-diff-tree %s %s", sq_quote(arg0), sq_quote(arg1));
+ * sprintf(rcmd, "ssh %s %s", sq_util/quote.host), sq_quote(cmd));
+ *
+ * Note that the above examples leak memory!  Remember to free result from
+ * sq_quote() in a real application.
+ *
+ * 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
diff --git a/tools/perf/util/rbtree.c b/tools/perf/util/rbtree.c
new file mode 100644
index 000000000000..b15ba9c7cb3f
--- /dev/null
+++ b/tools/perf/util/rbtree.c
@@ -0,0 +1,383 @@
+/*
+  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
new file mode 100644
index 000000000000..6bdc488a47fb
--- /dev/null
+++ b/tools/perf/util/rbtree.h
@@ -0,0 +1,171 @@
+/*
+  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
new file mode 100644
index 000000000000..b2f5e854f40a
--- /dev/null
+++ b/tools/perf/util/run-command.c
@@ -0,0 +1,395 @@
+#include "cache.h"
+#include "run-command.h"
+#include "exec_cmd.h"
+
+static inline void close_pair(int fd[2])
+{
+	close(fd[0]);
+	close(fd[1]);
+}
+
+static inline void dup_devnull(int to)
+{
+	int fd = open("/dev/null", O_RDWR);
+	dup2(fd, to);
+	close(fd);
+}
+
+int start_command(struct child_process *cmd)
+{
+	int need_in, need_out, need_err;
+	int fdin[2], fdout[2], fderr[2];
+
+	/*
+	 * In case of errors we must keep the promise to close FDs
+	 * that have been passed in via ->in and ->out.
+	 */
+
+	need_in = !cmd->no_stdin && cmd->in < 0;
+	if (need_in) {
+		if (pipe(fdin) < 0) {
+			if (cmd->out > 0)
+				close(cmd->out);
+			return -ERR_RUN_COMMAND_PIPE;
+		}
+		cmd->in = fdin[1];
+	}
+
+	need_out = !cmd->no_stdout
+		&& !cmd->stdout_to_stderr
+		&& cmd->out < 0;
+	if (need_out) {
+		if (pipe(fdout) < 0) {
+			if (need_in)
+				close_pair(fdin);
+			else if (cmd->in)
+				close(cmd->in);
+			return -ERR_RUN_COMMAND_PIPE;
+		}
+		cmd->out = fdout[0];
+	}
+
+	need_err = !cmd->no_stderr && cmd->err < 0;
+	if (need_err) {
+		if (pipe(fderr) < 0) {
+			if (need_in)
+				close_pair(fdin);
+			else if (cmd->in)
+				close(cmd->in);
+			if (need_out)
+				close_pair(fdout);
+			else if (cmd->out)
+				close(cmd->out);
+			return -ERR_RUN_COMMAND_PIPE;
+		}
+		cmd->err = fderr[0];
+	}
+
+#ifndef __MINGW32__
+	fflush(NULL);
+	cmd->pid = fork();
+	if (!cmd->pid) {
+		if (cmd->no_stdin)
+			dup_devnull(0);
+		else if (need_in) {
+			dup2(fdin[0], 0);
+			close_pair(fdin);
+		} else if (cmd->in) {
+			dup2(cmd->in, 0);
+			close(cmd->in);
+		}
+
+		if (cmd->no_stderr)
+			dup_devnull(2);
+		else if (need_err) {
+			dup2(fderr[1], 2);
+			close_pair(fderr);
+		}
+
+		if (cmd->no_stdout)
+			dup_devnull(1);
+		else if (cmd->stdout_to_stderr)
+			dup2(2, 1);
+		else if (need_out) {
+			dup2(fdout[1], 1);
+			close_pair(fdout);
+		} else if (cmd->out > 1) {
+			dup2(cmd->out, 1);
+			close(cmd->out);
+		}
+
+		if (cmd->dir && chdir(cmd->dir))
+			die("exec %s: cd to %s failed (%s)", cmd->argv[0],
+			    cmd->dir, strerror(errno));
+		if (cmd->env) {
+			for (; *cmd->env; cmd->env++) {
+				if (strchr(*cmd->env, '='))
+					putenv((char*)*cmd->env);
+				else
+					unsetenv(*cmd->env);
+			}
+		}
+		if (cmd->preexec_cb)
+			cmd->preexec_cb();
+		if (cmd->perf_cmd) {
+			execv_perf_cmd(cmd->argv);
+		} else {
+			execvp(cmd->argv[0], (char *const*) cmd->argv);
+		}
+		exit(127);
+	}
+#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;
+		if (need_in)
+			close_pair(fdin);
+		else if (cmd->in)
+			close(cmd->in);
+		if (need_out)
+			close_pair(fdout);
+		else if (cmd->out)
+			close(cmd->out);
+		if (need_err)
+			close_pair(fderr);
+		return err == ENOENT ?
+			-ERR_RUN_COMMAND_EXEC :
+			-ERR_RUN_COMMAND_FORK;
+	}
+
+	if (need_in)
+		close(fdin[0]);
+	else if (cmd->in)
+		close(cmd->in);
+
+	if (need_out)
+		close(fdout[1]);
+	else if (cmd->out)
+		close(cmd->out);
+
+	if (need_err)
+		close(fderr[1]);
+
+	return 0;
+}
+
+static int wait_or_whine(pid_t pid)
+{
+	for (;;) {
+		int status, code;
+		pid_t waiting = waitpid(pid, &status, 0);
+
+		if (waiting < 0) {
+			if (errno == EINTR)
+				continue;
+			error("waitpid failed (%s)", strerror(errno));
+			return -ERR_RUN_COMMAND_WAITPID;
+		}
+		if (waiting != pid)
+			return -ERR_RUN_COMMAND_WAITPID_WRONG_PID;
+		if (WIFSIGNALED(status))
+			return -ERR_RUN_COMMAND_WAITPID_SIGNAL;
+
+		if (!WIFEXITED(status))
+			return -ERR_RUN_COMMAND_WAITPID_NOEXIT;
+		code = WEXITSTATUS(status);
+		switch (code) {
+		case 127:
+			return -ERR_RUN_COMMAND_EXEC;
+		case 0:
+			return 0;
+		default:
+			return -code;
+		}
+	}
+}
+
+int finish_command(struct child_process *cmd)
+{
+	return wait_or_whine(cmd->pid);
+}
+
+int run_command(struct child_process *cmd)
+{
+	int code = start_command(cmd);
+	if (code)
+		return code;
+	return finish_command(cmd);
+}
+
+static void prepare_run_command_v_opt(struct child_process *cmd,
+				      const char **argv,
+				      int opt)
+{
+	memset(cmd, 0, sizeof(*cmd));
+	cmd->argv = argv;
+	cmd->no_stdin = opt & RUN_COMMAND_NO_STDIN ? 1 : 0;
+	cmd->perf_cmd = opt & RUN_PERF_CMD ? 1 : 0;
+	cmd->stdout_to_stderr = opt & RUN_COMMAND_STDOUT_TO_STDERR ? 1 : 0;
+}
+
+int run_command_v_opt(const char **argv, int opt)
+{
+	struct child_process cmd;
+	prepare_run_command_v_opt(&cmd, argv, opt);
+	return run_command(&cmd);
+}
+
+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
new file mode 100644
index 000000000000..328289f23669
--- /dev/null
+++ b/tools/perf/util/run-command.h
@@ -0,0 +1,93 @@
+#ifndef RUN_COMMAND_H
+#define RUN_COMMAND_H
+
+enum {
+	ERR_RUN_COMMAND_FORK = 10000,
+	ERR_RUN_COMMAND_EXEC,
+	ERR_RUN_COMMAND_PIPE,
+	ERR_RUN_COMMAND_WAITPID,
+	ERR_RUN_COMMAND_WAITPID_WRONG_PID,
+	ERR_RUN_COMMAND_WAITPID_SIGNAL,
+	ERR_RUN_COMMAND_WAITPID_NOEXIT,
+};
+#define IS_RUN_COMMAND_ERR(x) (-(x) >= ERR_RUN_COMMAND_FORK)
+
+struct child_process {
+	const char **argv;
+	pid_t pid;
+	/*
+	 * Using .in, .out, .err:
+	 * - Specify 0 for no redirections (child inherits stdin, stdout,
+	 *   stderr from parent).
+	 * - Specify -1 to have a pipe allocated as follows:
+	 *     .in: returns the writable pipe end; parent writes to it,
+	 *          the readable pipe end becomes child's stdin
+	 *     .out, .err: returns the readable pipe end; parent reads from
+	 *          it, the writable pipe end becomes child's stdout/stderr
+	 *   The caller of start_command() must close the returned FDs
+	 *   after it has completed reading from/writing to it!
+	 * - Specify > 0 to set a channel to a particular FD as follows:
+	 *     .in: a readable FD, becomes child's stdin
+	 *     .out: a writable FD, becomes child's stdout/stderr
+	 *     .err > 0 not supported
+	 *   The specified FD is closed by start_command(), even in case
+	 *   of errors!
+	 */
+	int in;
+	int out;
+	int err;
+	const char *dir;
+	const char *const *env;
+	unsigned no_stdin:1;
+	unsigned no_stdout:1;
+	unsigned no_stderr:1;
+	unsigned perf_cmd:1; /* if this is to be perf sub-command */
+	unsigned stdout_to_stderr:1;
+	void (*preexec_cb)(void);
+};
+
+int start_command(struct child_process *);
+int finish_command(struct child_process *);
+int run_command(struct child_process *);
+
+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
diff --git a/tools/perf/util/sigchain.c b/tools/perf/util/sigchain.c
new file mode 100644
index 000000000000..1118b99e57d3
--- /dev/null
+++ b/tools/perf/util/sigchain.c
@@ -0,0 +1,52 @@
+#include "sigchain.h"
+#include "cache.h"
+
+#define SIGCHAIN_MAX_SIGNALS 32
+
+struct sigchain_signal {
+	sigchain_fun *old;
+	int n;
+	int alloc;
+};
+static struct sigchain_signal signals[SIGCHAIN_MAX_SIGNALS];
+
+static void check_signum(int sig)
+{
+	if (sig < 1 || sig >= SIGCHAIN_MAX_SIGNALS)
+		die("BUG: signal out of range: %d", sig);
+}
+
+int sigchain_push(int sig, sigchain_fun f)
+{
+	struct sigchain_signal *s = signals + sig;
+	check_signum(sig);
+
+	ALLOC_GROW(s->old, s->n + 1, s->alloc);
+	s->old[s->n] = signal(sig, f);
+	if (s->old[s->n] == SIG_ERR)
+		return -1;
+	s->n++;
+	return 0;
+}
+
+int sigchain_pop(int sig)
+{
+	struct sigchain_signal *s = signals + sig;
+	check_signum(sig);
+	if (s->n < 1)
+		return 0;
+
+	if (signal(sig, s->old[s->n - 1]) == SIG_ERR)
+		return -1;
+	s->n--;
+	return 0;
+}
+
+void sigchain_push_common(sigchain_fun f)
+{
+	sigchain_push(SIGINT, f);
+	sigchain_push(SIGHUP, f);
+	sigchain_push(SIGTERM, f);
+	sigchain_push(SIGQUIT, f);
+	sigchain_push(SIGPIPE, f);
+}
diff --git a/tools/perf/util/sigchain.h b/tools/perf/util/sigchain.h
new file mode 100644
index 000000000000..618083bce0c6
--- /dev/null
+++ b/tools/perf/util/sigchain.h
@@ -0,0 +1,11 @@
+#ifndef SIGCHAIN_H
+#define 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 */
diff --git a/tools/perf/util/strbuf.c b/tools/perf/util/strbuf.c
new file mode 100644
index 000000000000..eaba09306802
--- /dev/null
+++ b/tools/perf/util/strbuf.c
@@ -0,0 +1,359 @@
+#include "cache.h"
+
+int prefixcmp(const char *str, const char *prefix)
+{
+	for (; ; str++, prefix++)
+		if (!*prefix)
+			return 0;
+		else if (*str != *prefix)
+			return (unsigned char)*prefix - (unsigned char)*str;
+}
+
+/*
+ * Used as the default ->buf value, so that people can always assume
+ * buf is non NULL and ->buf is NUL terminated even for a freshly
+ * initialized strbuf.
+ */
+char strbuf_slopbuf[1];
+
+void strbuf_init(struct strbuf *sb, size_t hint)
+{
+	sb->alloc = sb->len = 0;
+	sb->buf = strbuf_slopbuf;
+	if (hint)
+		strbuf_grow(sb, hint);
+}
+
+void strbuf_release(struct strbuf *sb)
+{
+	if (sb->alloc) {
+		free(sb->buf);
+		strbuf_init(sb, 0);
+	}
+}
+
+char *strbuf_detach(struct strbuf *sb, size_t *sz)
+{
+	char *res = sb->alloc ? sb->buf : NULL;
+	if (sz)
+		*sz = sb->len;
+	strbuf_init(sb, 0);
+	return res;
+}
+
+void strbuf_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)
+		die("you want to use way too much memory");
+	if (!sb->alloc)
+		sb->buf = NULL;
+	ALLOC_GROW(sb->buf, sb->len + extra + 1, sb->alloc);
+}
+
+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,
+				   const void *data, size_t dlen)
+{
+	if (pos + len < pos)
+		die("you want to use way too much memory");
+	if (pos > sb->len)
+		die("`pos' is too far after the end of the buffer");
+	if (pos + len > sb->len)
+		die("`pos + len' is too far after the end of the buffer");
+
+	if (dlen >= len)
+		strbuf_grow(sb, dlen - len);
+	memmove(sb->buf + pos + dlen,
+			sb->buf + pos + len,
+			sb->len - pos - len);
+	memcpy(sb->buf + pos, data, dlen);
+	strbuf_setlen(sb, sb->len + dlen - len);
+}
+
+void strbuf_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);
+}
+
+void strbuf_add(struct strbuf *sb, const void *data, size_t len)
+{
+	strbuf_grow(sb, len);
+	memcpy(sb->buf + sb->len, data, len);
+	strbuf_setlen(sb, sb->len + len);
+}
+
+void strbuf_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;
+	va_list ap;
+
+	if (!strbuf_avail(sb))
+		strbuf_grow(sb, 64);
+	va_start(ap, fmt);
+	len = vsnprintf(sb->buf + sb->len, sb->alloc - sb->len, fmt, ap);
+	va_end(ap);
+	if (len < 0)
+		die("your vsnprintf is broken");
+	if (len > strbuf_avail(sb)) {
+		strbuf_grow(sb, len);
+		va_start(ap, fmt);
+		len = vsnprintf(sb->buf + sb->len, sb->alloc - sb->len, fmt, ap);
+		va_end(ap);
+		if (len > strbuf_avail(sb)) {
+			die("this should not happen, your snprintf is broken");
+		}
+	}
+	strbuf_setlen(sb, sb->len + len);
+}
+
+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)
+{
+	size_t oldlen = sb->len;
+	size_t oldalloc = sb->alloc;
+
+	strbuf_grow(sb, hint ? hint : 8192);
+	for (;;) {
+		ssize_t cnt;
+
+		cnt = read(fd, sb->buf + sb->len, sb->alloc - sb->len - 1);
+		if (cnt < 0) {
+			if (oldalloc == 0)
+				strbuf_release(sb);
+			else
+				strbuf_setlen(sb, oldlen);
+			return -1;
+		}
+		if (!cnt)
+			break;
+		sb->len += cnt;
+		strbuf_grow(sb, 8192);
+	}
+
+	sb->buf[sb->len] = '\0';
+	return sb->len - oldlen;
+}
+
+#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
new file mode 100644
index 000000000000..9ee908a3ec5d
--- /dev/null
+++ b/tools/perf/util/strbuf.h
@@ -0,0 +1,137 @@
+#ifndef STRBUF_H
+#define STRBUF_H
+
+/*
+ * Strbuf's can be use in many ways: as a byte array, or to store arbitrary
+ * long, overflow safe strings.
+ *
+ * Strbufs has some invariants that are very important to keep in mind:
+ *
+ * 1. the ->buf member is always malloc-ed, hence strbuf's can be used to
+ *    build complex strings/buffers whose final size isn't easily known.
+ *
+ *    It is NOT legal to copy the ->buf pointer away.
+ *    `strbuf_detach' is the operation that detachs a buffer from its shell
+ *    while keeping the shell valid wrt its invariants.
+ *
+ * 2. the ->buf member is a byte array that has at least ->len + 1 bytes
+ *    allocated. The extra byte is used to store a '\0', allowing the ->buf
+ *    member to be a valid C-string. Every strbuf function ensure this
+ *    invariant is preserved.
+ *
+ *    Note that it is OK to "play" with the buffer directly if you work it
+ *    that way:
+ *
+ *    strbuf_grow(sb, SOME_SIZE);
+ *       ... Here, the memory array starting at sb->buf, and of length
+ *       ... strbuf_avail(sb) is all yours, and you are sure that
+ *       ... strbuf_avail(sb) is at least SOME_SIZE.
+ *    strbuf_setlen(sb, sb->len + SOME_OTHER_SIZE);
+ *
+ *    Of course, SOME_OTHER_SIZE must be smaller or equal to strbuf_avail(sb).
+ *
+ *    Doing so is safe, though if it has to be done in many places, adding the
+ *    missing API to the strbuf module is the way to go.
+ *
+ *    XXX: do _not_ assume that the area that is yours is of size ->alloc - 1
+ *         even if it's true in the current implementation. Alloc is somehow a
+ *         "private" member that should not be messed with.
+ */
+
+#include <assert.h>
+
+extern char strbuf_slopbuf[];
+struct strbuf {
+	size_t alloc;
+	size_t len;
+	char *buf;
+};
+
+#define STRBUF_INIT  { 0, 0, strbuf_slopbuf }
+
+/*----- strbuf life cycle -----*/
+extern void strbuf_init(struct strbuf *, size_t);
+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) {
+	return sb->alloc ? sb->alloc - sb->len - 1 : 0;
+}
+
+extern void strbuf_grow(struct strbuf *, size_t);
+
+static inline void strbuf_setlen(struct strbuf *sb, size_t len) {
+	if (!sb->alloc)
+		strbuf_grow(sb, 0);
+	assert(len < sb->alloc);
+	sb->len = len;
+	sb->buf[len] = '\0';
+}
+#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) {
+	strbuf_grow(sb, 1);
+	sb->buf[sb->len++] = 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);
+
+#endif /* STRBUF_H */
diff --git a/tools/perf/util/string.c b/tools/perf/util/string.c
new file mode 100644
index 000000000000..ec33c0c7f4e2
--- /dev/null
+++ b/tools/perf/util/string.c
@@ -0,0 +1,34 @@
+#include "string.h"
+
+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;
+}
diff --git a/tools/perf/util/string.h b/tools/perf/util/string.h
new file mode 100644
index 000000000000..72812c1c9a7a
--- /dev/null
+++ b/tools/perf/util/string.h
@@ -0,0 +1,8 @@
+#ifndef _PERF_STRING_H_
+#define _PERF_STRING_H_
+
+#include <linux/types.h>
+
+int hex2u64(const char *ptr, __u64 *val);
+
+#endif
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
new file mode 100644
index 000000000000..49a55f813712
--- /dev/null
+++ b/tools/perf/util/symbol.c
@@ -0,0 +1,641 @@
+#include "util.h"
+#include "../perf.h"
+#include "string.h"
+#include "symbol.h"
+
+#include <libelf.h>
+#include <gelf.h>
+#include <elf.h>
+
+const char *sym_hist_filter;
+
+static struct symbol *symbol__new(__u64 start, __u64 len,
+				  const char *name, unsigned int priv_size,
+				  __u64 obj_start, int verbose)
+{
+	size_t namelen = strlen(name) + 1;
+	struct symbol *self = calloc(1, priv_size + sizeof(*self) + namelen);
+
+	if (!self)
+		return NULL;
+
+	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);
+
+	self->obj_start= obj_start;
+	self->hist = NULL;
+	self->hist_sum = 0;
+
+	if (sym_hist_filter && !strcmp(name, sym_hist_filter))
+		self->hist = calloc(sizeof(__u64), len);
+
+	if (priv_size) {
+		memset(self, 0, priv_size);
+		self = ((void *)self) + priv_size;
+	}
+	self->start = start;
+	self->end   = start + len - 1;
+	memcpy(self->name, name, namelen);
+
+	return self;
+}
+
+static void symbol__delete(struct symbol *self, unsigned int priv_size)
+{
+	free(((void *)self) - 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);
+}
+
+struct dso *dso__new(const char *name, unsigned int sym_priv_size)
+{
+	struct dso *self = malloc(sizeof(*self) + strlen(name) + 1);
+
+	if (self != NULL) {
+		strcpy(self->name, name);
+		self->syms = RB_ROOT;
+		self->sym_priv_size = sym_priv_size;
+		self->find_symbol = dso__find_symbol;
+	}
+
+	return self;
+}
+
+static void dso__delete_symbols(struct dso *self)
+{
+	struct symbol *pos;
+	struct rb_node *next = rb_first(&self->syms);
+
+	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);
+	}
+}
+
+void dso__delete(struct dso *self)
+{
+	dso__delete_symbols(self);
+	free(self);
+}
+
+static void dso__insert_symbol(struct dso *self, struct symbol *sym)
+{
+	struct rb_node **p = &self->syms.rb_node;
+	struct rb_node *parent = NULL;
+	const __u64 ip = sym->start;
+	struct symbol *s;
+
+	while (*p != NULL) {
+		parent = *p;
+		s = rb_entry(parent, struct symbol, rb_node);
+		if (ip < s->start)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+	rb_link_node(&sym->rb_node, parent, p);
+	rb_insert_color(&sym->rb_node, &self->syms);
+}
+
+struct symbol *dso__find_symbol(struct dso *self, __u64 ip)
+{
+	struct rb_node *n;
+
+	if (self == NULL)
+		return NULL;
+
+	n = self->syms.rb_node;
+
+	while (n) {
+		struct symbol *s = rb_entry(n, struct symbol, rb_node);
+
+		if (ip < s->start)
+			n = n->rb_left;
+		else if (ip > s->end)
+			n = n->rb_right;
+		else
+			return s;
+	}
+
+	return NULL;
+}
+
+size_t dso__fprintf(struct dso *self, FILE *fp)
+{
+	size_t ret = fprintf(fp, "dso: %s\n", self->name);
+
+	struct rb_node *nd;
+	for (nd = rb_first(&self->syms); nd; nd = rb_next(nd)) {
+		struct symbol *pos = rb_entry(nd, struct symbol, rb_node);
+		ret += symbol__fprintf(pos, fp);
+	}
+
+	return ret;
+}
+
+static int dso__load_kallsyms(struct dso *self, symbol_filter_t filter, int verbose)
+{
+	struct rb_node *nd, *prevnd;
+	char *line = NULL;
+	size_t n;
+	FILE *file = fopen("/proc/kallsyms", "r");
+
+	if (file == NULL)
+		goto out_failure;
+
+	while (!feof(file)) {
+		__u64 start;
+		struct symbol *sym;
+		int line_len, len;
+		char symbol_type;
+
+		line_len = getline(&line, &n, file);
+		if (line_len < 0)
+			break;
+
+		if (!line)
+			goto out_failure;
+
+		line[--line_len] = '\0'; /* \n */
+
+		len = hex2u64(line, &start);
+
+		len++;
+		if (len + 2 >= line_len)
+			continue;
+
+		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);
+
+		if (sym == NULL)
+			goto out_delete_line;
+
+		if (filter && filter(self, sym))
+			symbol__delete(sym, self->sym_priv_size);
+		else
+			dso__insert_symbol(self, sym);
+	}
+
+	/*
+	 * Now that we have all sorted out, just set the ->end of all
+	 * symbols
+	 */
+	prevnd = rb_first(&self->syms);
+
+	if (prevnd == NULL)
+		goto out_delete_line;
+
+	for (nd = rb_next(prevnd); nd; nd = rb_next(nd)) {
+		struct symbol *prev = rb_entry(prevnd, struct symbol, rb_node),
+			      *curr = rb_entry(nd, struct symbol, rb_node);
+
+		prev->end = curr->start - 1;
+		prevnd = nd;
+	}
+
+	free(line);
+	fclose(file);
+
+	return 0;
+
+out_delete_line:
+	free(line);
+out_failure:
+	return -1;
+}
+
+static int dso__load_perf_map(struct dso *self, symbol_filter_t filter, int verbose)
+{
+	char *line = NULL;
+	size_t n;
+	FILE *file;
+	int nr_syms = 0;
+
+	file = fopen(self->name, "r");
+	if (file == NULL)
+		goto out_failure;
+
+	while (!feof(file)) {
+		__u64 start, size;
+		struct symbol *sym;
+		int line_len, len;
+
+		line_len = getline(&line, &n, file);
+		if (line_len < 0)
+			break;
+
+		if (!line)
+			goto out_failure;
+
+		line[--line_len] = '\0'; /* \n */
+
+		len = hex2u64(line, &start);
+
+		len++;
+		if (len + 2 >= line_len)
+			continue;
+
+		len += hex2u64(line + len, &size);
+
+		len++;
+		if (len + 2 >= line_len)
+			continue;
+
+		sym = symbol__new(start, size, line + len,
+				  self->sym_priv_size, start, verbose);
+
+		if (sym == NULL)
+			goto out_delete_line;
+
+		if (filter && filter(self, sym))
+			symbol__delete(sym, self->sym_priv_size);
+		else {
+			dso__insert_symbol(self, sym);
+			nr_syms++;
+		}
+	}
+
+	free(line);
+	fclose(file);
+
+	return nr_syms;
+
+out_delete_line:
+	free(line);
+out_failure:
+	return -1;
+}
+
+/**
+ * elf_symtab__for_each_symbol - iterate thru all the symbols
+ *
+ * @self: struct elf_symtab instance to iterate
+ * @index: uint32_t index
+ * @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))
+
+static inline uint8_t elf_sym__type(const GElf_Sym *sym)
+{
+	return GELF_ST_TYPE(sym->st_info);
+}
+
+static inline int elf_sym__is_function(const GElf_Sym *sym)
+{
+	return elf_sym__type(sym) == STT_FUNC &&
+	       sym->st_name != 0 &&
+	       sym->st_shndx != SHN_UNDEF &&
+	       sym->st_size != 0;
+}
+
+static inline const char *elf_sym__name(const GElf_Sym *sym,
+					const Elf_Data *symstrs)
+{
+	return symstrs->d_buf + sym->st_name;
+}
+
+static Elf_Scn *elf_section_by_name(Elf *elf, GElf_Ehdr *ep,
+				    GElf_Shdr *shp, const char *name,
+				    size_t *index)
+{
+	Elf_Scn *sec = NULL;
+	size_t cnt = 1;
+
+	while ((sec = elf_nextscn(elf, sec)) != NULL) {
+		char *str;
+
+		gelf_getshdr(sec, shp);
+		str = elf_strptr(elf, ep->e_shstrndx, shp->sh_name);
+		if (!strcmp(name, str)) {
+			if (index)
+				*index = cnt;
+			break;
+		}
+		++cnt;
+	}
+
+	return sec;
+}
+
+#define elf_section__for_each_rel(reldata, pos, pos_mem, idx, nr_entries) \
+	for (idx = 0, pos = gelf_getrel(reldata, 0, &pos_mem); \
+	     idx < nr_entries; \
+	     ++idx, pos = gelf_getrel(reldata, idx, &pos_mem))
+
+#define elf_section__for_each_rela(reldata, pos, pos_mem, idx, nr_entries) \
+	for (idx = 0, pos = gelf_getrela(reldata, 0, &pos_mem); \
+	     idx < nr_entries; \
+	     ++idx, pos = gelf_getrela(reldata, idx, &pos_mem))
+
+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)
+{
+	uint32_t nr_rel_entries, idx;
+	GElf_Sym sym;
+	__u64 plt_offset;
+	GElf_Shdr shdr_plt;
+	struct symbol *f;
+	GElf_Shdr shdr_rel_plt;
+	Elf_Data *reldata, *syms, *symstrs;
+	Elf_Scn *scn_plt_rel, *scn_symstrs;
+	char sympltname[1024];
+	int nr = 0, symidx;
+
+	scn_plt_rel = elf_section_by_name(elf, ehdr, &shdr_rel_plt,
+					  ".rela.plt", NULL);
+	if (scn_plt_rel == NULL) {
+		scn_plt_rel = elf_section_by_name(elf, ehdr, &shdr_rel_plt,
+						  ".rel.plt", NULL);
+		if (scn_plt_rel == NULL)
+			return 0;
+	}
+
+	if (shdr_rel_plt.sh_link != dynsym_idx)
+		return 0;
+
+	if (elf_section_by_name(elf, ehdr, &shdr_plt, ".plt", NULL) == NULL)
+		return 0;
+
+	/*
+	 * Fetch the relocation section to find the indexes to the GOT
+	 * and the symbols in the .dynsym they refer to.
+	 */
+	reldata = elf_getdata(scn_plt_rel, NULL);
+	if (reldata == NULL)
+		return -1;
+
+	syms = elf_getdata(scn_dynsym, NULL);
+	if (syms == NULL)
+		return -1;
+
+	scn_symstrs = elf_getscn(elf, shdr_dynsym->sh_link);
+	if (scn_symstrs == NULL)
+		return -1;
+
+	symstrs = elf_getdata(scn_symstrs, NULL);
+	if (symstrs == NULL)
+		return -1;
+
+	nr_rel_entries = shdr_rel_plt.sh_size / shdr_rel_plt.sh_entsize;
+	plt_offset = shdr_plt.sh_offset;
+
+	if (shdr_rel_plt.sh_type == SHT_RELA) {
+		GElf_Rela pos_mem, *pos;
+
+		elf_section__for_each_rela(reldata, pos, pos_mem, idx,
+					   nr_rel_entries) {
+			symidx = GELF_R_SYM(pos->r_info);
+			plt_offset += shdr_plt.sh_entsize;
+			gelf_getsym(syms, symidx, &sym);
+			snprintf(sympltname, sizeof(sympltname),
+				 "%s@plt", elf_sym__name(&sym, symstrs));
+
+			f = symbol__new(plt_offset, shdr_plt.sh_entsize,
+					sympltname, self->sym_priv_size, 0, verbose);
+			if (!f)
+				return -1;
+
+			dso__insert_symbol(self, f);
+			++nr;
+		}
+	} else if (shdr_rel_plt.sh_type == SHT_REL) {
+		GElf_Rel pos_mem, *pos;
+		elf_section__for_each_rel(reldata, pos, pos_mem, idx,
+					  nr_rel_entries) {
+			symidx = GELF_R_SYM(pos->r_info);
+			plt_offset += shdr_plt.sh_entsize;
+			gelf_getsym(syms, symidx, &sym);
+			snprintf(sympltname, sizeof(sympltname),
+				 "%s@plt", elf_sym__name(&sym, symstrs));
+
+			f = symbol__new(plt_offset, shdr_plt.sh_entsize,
+					sympltname, self->sym_priv_size, 0, verbose);
+			if (!f)
+				return -1;
+
+			dso__insert_symbol(self, 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;
+}
+
+static int dso__load_sym(struct dso *self, int fd, const char *name,
+			 symbol_filter_t filter, int verbose)
+{
+	Elf_Data *symstrs;
+	uint32_t nr_syms;
+	int err = -1;
+	uint32_t index;
+	GElf_Ehdr ehdr;
+	GElf_Shdr shdr;
+	Elf_Data *syms;
+	GElf_Sym sym;
+	Elf_Scn *sec, *sec_dynsym;
+	Elf *elf;
+	size_t dynsym_idx;
+	int nr = 0;
+
+	elf = elf_begin(fd, ELF_C_READ_MMAP, NULL);
+	if (elf == NULL) {
+		if (verbose)
+			fprintf(stderr, "%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__);
+		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)
+			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)
+			goto out_elf_end;
+
+		sec = sec_dynsym;
+		gelf_getshdr(sec, &shdr);
+	}
+
+	syms = elf_getdata(sec, NULL);
+	if (syms == NULL)
+		goto out_elf_end;
+
+	sec = elf_getscn(elf, shdr.sh_link);
+	if (sec == NULL)
+		goto out_elf_end;
+
+	symstrs = elf_getdata(sec, NULL);
+	if (symstrs == NULL)
+		goto out_elf_end;
+
+	nr_syms = shdr.sh_size / shdr.sh_entsize;
+
+	memset(&sym, 0, sizeof(sym));
+
+	elf_symtab__for_each_symbol(syms, nr_syms, index, sym) {
+		struct symbol *f;
+		__u64 obj_start;
+
+		if (!elf_sym__is_function(&sym))
+			continue;
+
+		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;
+
+		f = symbol__new(sym.st_value, sym.st_size,
+				elf_sym__name(&sym, symstrs),
+				self->sym_priv_size, obj_start, verbose);
+		if (!f)
+			goto out_elf_end;
+
+		if (filter && filter(self, f))
+			symbol__delete(f, self->sym_priv_size);
+		else {
+			dso__insert_symbol(self, f);
+			nr++;
+		}
+	}
+
+	err = nr;
+out_elf_end:
+	elf_end(elf);
+out_close:
+	return err;
+}
+
+int dso__load(struct dso *self, symbol_filter_t filter, int verbose)
+{
+	int size = strlen(self->name) + sizeof("/usr/lib/debug%s.debug");
+	char *name = malloc(size);
+	int variant = 0;
+	int ret = -1;
+	int fd;
+
+	if (!name)
+		return -1;
+
+	if (strncmp(self->name, "/tmp/perf-", 10) == 0)
+		return dso__load_perf_map(self, filter, verbose);
+
+more:
+	do {
+		switch (variant) {
+		case 0: /* Fedora */
+			snprintf(name, size, "/usr/lib/debug%s.debug", self->name);
+			break;
+		case 1: /* Ubuntu */
+			snprintf(name, size, "/usr/lib/debug%s", self->name);
+			break;
+		case 2: /* Sane people */
+			snprintf(name, size, "%s", self->name);
+			break;
+
+		default:
+			goto out;
+		}
+		variant++;
+
+		fd = open(name, O_RDONLY);
+	} while (fd < 0);
+
+	ret = dso__load_sym(self, fd, name, filter, verbose);
+	close(fd);
+
+	/*
+	 * Some people seem to have debuginfo files _WITHOUT_ debug info!?!?
+	 */
+	if (!ret)
+		goto more;
+
+out:
+	free(name);
+	return ret;
+}
+
+static int dso__load_vmlinux(struct dso *self, const char *vmlinux,
+			     symbol_filter_t filter, int verbose)
+{
+	int err, fd = open(vmlinux, O_RDONLY);
+
+	if (fd < 0)
+		return -1;
+
+	err = dso__load_sym(self, fd, vmlinux, filter, verbose);
+	close(fd);
+
+	return err;
+}
+
+int dso__load_kernel(struct dso *self, const char *vmlinux,
+		     symbol_filter_t filter, int verbose)
+{
+	int err = -1;
+
+	if (vmlinux)
+		err = dso__load_vmlinux(self, vmlinux, filter, verbose);
+
+	if (err)
+		err = dso__load_kallsyms(self, filter, verbose);
+
+	return err;
+}
+
+void symbol__init(void)
+{
+	elf_version(EV_CURRENT);
+}
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
new file mode 100644
index 000000000000..0d1292bd8270
--- /dev/null
+++ b/tools/perf/util/symbol.h
@@ -0,0 +1,47 @@
+#ifndef _PERF_SYMBOL_
+#define _PERF_SYMBOL_ 1
+
+#include <linux/types.h>
+#include "list.h"
+#include "rbtree.h"
+
+struct symbol {
+	struct rb_node	rb_node;
+	__u64		start;
+	__u64		end;
+	__u64		obj_start;
+	__u64		hist_sum;
+	__u64		*hist;
+	char		name[0];
+};
+
+struct dso {
+	struct list_head node;
+	struct rb_root	 syms;
+	unsigned int	 sym_priv_size;
+	struct symbol    *(*find_symbol)(struct dso *, __u64 ip);
+	char		 name[0];
+};
+
+const char *sym_hist_filter;
+
+typedef int (*symbol_filter_t)(struct dso *self, struct symbol *sym);
+
+struct dso *dso__new(const char *name, unsigned int sym_priv_size);
+void dso__delete(struct dso *self);
+
+static inline void *dso__sym_priv(struct dso *self, struct symbol *sym)
+{
+	return ((void *)sym) - self->sym_priv_size;
+}
+
+struct symbol *dso__find_symbol(struct dso *self, __u64 ip);
+
+int dso__load_kernel(struct dso *self, const char *vmlinux,
+		     symbol_filter_t filter, int verbose);
+int dso__load(struct dso *self, symbol_filter_t filter, int verbose);
+
+size_t dso__fprintf(struct dso *self, FILE *fp);
+
+void symbol__init(void);
+#endif /* _PERF_SYMBOL_ */
diff --git a/tools/perf/util/usage.c b/tools/perf/util/usage.c
new file mode 100644
index 000000000000..e16bf9a707e8
--- /dev/null
+++ b/tools/perf/util/usage.c
@@ -0,0 +1,80 @@
+/*
+ * GIT - The information manager from hell
+ *
+ * Copyright (C) Linus Torvalds, 2005
+ */
+#include "util.h"
+
+static void report(const char *prefix, const char *err, va_list params)
+{
+	char msg[1024];
+	vsnprintf(msg, sizeof(msg), err, params);
+	fprintf(stderr, " %s%s\n", prefix, msg);
+}
+
+static NORETURN void usage_builtin(const char *err)
+{
+	fprintf(stderr, "\n Usage: %s\n", err);
+	exit(129);
+}
+
+static NORETURN void die_builtin(const char *err, va_list params)
+{
+	report(" Fatal: ", err, params);
+	exit(128);
+}
+
+static void error_builtin(const char *err, va_list params)
+{
+	report(" Error: ", err, params);
+}
+
+static void warn_builtin(const char *warn, va_list params)
+{
+	report(" Warning: ", warn, params);
+}
+
+/* If we are in a dlopen()ed .so write to a global variable would segfault
+ * (ugh), so keep things static. */
+static void (*usage_routine)(const char *err) NORETURN = usage_builtin;
+static void (*die_routine)(const char *err, va_list params) NORETURN = die_builtin;
+static void (*error_routine)(const char *err, va_list params) = error_builtin;
+static void (*warn_routine)(const char *err, va_list params) = warn_builtin;
+
+void set_die_routine(void (*routine)(const char *err, va_list params) NORETURN)
+{
+	die_routine = routine;
+}
+
+void usage(const char *err)
+{
+	usage_routine(err);
+}
+
+void die(const char *err, ...)
+{
+	va_list params;
+
+	va_start(params, err);
+	die_routine(err, params);
+	va_end(params);
+}
+
+int error(const char *err, ...)
+{
+	va_list params;
+
+	va_start(params, err);
+	error_routine(err, params);
+	va_end(params);
+	return -1;
+}
+
+void warning(const char *warn, ...)
+{
+	va_list params;
+
+	va_start(params, warn);
+	warn_routine(warn, params);
+	va_end(params);
+}
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
new file mode 100644
index 000000000000..76590a16c271
--- /dev/null
+++ b/tools/perf/util/util.h
@@ -0,0 +1,410 @@
+#ifndef GIT_COMPAT_UTIL_H
+#define GIT_COMPAT_UTIL_H
+
+#define _FILE_OFFSET_BITS 64
+
+#ifndef FLEX_ARRAY
+/*
+ * See if our compiler is known to support flexible array members.
+ */
+#if defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L)
+# define FLEX_ARRAY /* empty */
+#elif defined(__GNUC__)
+# if (__GNUC__ >= 3)
+#  define FLEX_ARRAY /* empty */
+# else
+#  define FLEX_ARRAY 0 /* older GNU extension */
+# endif
+#endif
+
+/*
+ * Otherwise, default to safer but a bit wasteful traditional style
+ */
+#ifndef FLEX_ARRAY
+# define FLEX_ARRAY 1
+#endif
+#endif
+
+#define ARRAY_SIZE(x) (sizeof(x)/sizeof(x[0]))
+
+#ifdef __GNUC__
+#define TYPEOF(x) (__typeof__(x))
+#else
+#define TYPEOF(x)
+#endif
+
+#define MSB(x, bits) ((x) & TYPEOF(x)(~0ULL << (sizeof(x) * 8 - (bits))))
+#define HAS_MULTI_BITS(i)  ((i) & ((i) - 1))  /* checks if an integer has more than 1 bit set */
+
+/* Approximation of the length of the decimal representation of this type. */
+#define decimal_length(x)	((int)(sizeof(x) * 2.56 + 0.5) + 1)
+
+#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
+
+#include <unistd.h>
+#include <stdio.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <stddef.h>
+#include <stdlib.h>
+#include <stdarg.h>
+#include <string.h>
+#include <errno.h>
+#include <limits.h>
+#include <sys/param.h>
+#include <sys/types.h>
+#include <dirent.h>
+#include <sys/time.h>
+#include <time.h>
+#include <signal.h>
+#include <fnmatch.h>
+#include <assert.h>
+#include <regex.h>
+#include <utime.h>
+#ifndef __MINGW32__
+#include <sys/wait.h>
+#include <sys/poll.h>
+#include <sys/socket.h>
+#include <sys/ioctl.h>
+#ifndef NO_SYS_SELECT_H
+#include <sys/select.h>
+#endif
+#include <netinet/in.h>
+#include <netinet/tcp.h>
+#include <arpa/inet.h>
+#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__ */
+
+#ifndef NO_ICONV
+#include <iconv.h>
+#endif
+
+#ifndef NO_OPENSSL
+#include <openssl/ssl.h>
+#include <openssl/err.h>
+#endif
+
+/* On most systems <limits.h> would have given us this, but
+ * not on some systems (e.g. GNU/Hurd).
+ */
+#ifndef PATH_MAX
+#define PATH_MAX 4096
+#endif
+
+#ifndef PRIuMAX
+#define PRIuMAX "llu"
+#endif
+
+#ifndef PRIu32
+#define PRIu32 "u"
+#endif
+
+#ifndef PRIx32
+#define PRIx32 "x"
+#endif
+
+#ifndef PATH_SEP
+#define PATH_SEP ':'
+#endif
+
+#ifndef STRIP_EXTENSION
+#define STRIP_EXTENSION ""
+#endif
+
+#ifndef has_dos_drive_prefix
+#define has_dos_drive_prefix(path) 0
+#endif
+
+#ifndef is_dir_sep
+#define is_dir_sep(c) ((c) == '/')
+#endif
+
+#ifdef __GNUC__
+#define NORETURN __attribute__((__noreturn__))
+#else
+#define NORETURN
+#ifndef __attribute__
+#define __attribute__(x)
+#endif
+#endif
+
+/* General helper functions */
+extern void usage(const char *err) NORETURN;
+extern void die(const char *err, ...) NORETURN __attribute__((format (printf, 1, 2)));
+extern int error(const char *err, ...) __attribute__((format (printf, 1, 2)));
+extern void warning(const char *err, ...) __attribute__((format (printf, 1, 2)));
+
+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);
+
+static inline const char *skip_prefix(const char *str, const char *prefix)
+{
+	size_t len = strlen(prefix);
+	return strncmp(str, prefix, len) ? NULL : str + len;
+}
+
+#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
+#endif
+#endif
+
+#ifndef HAVE_STRCHRNUL
+#define strchrnul gitstrchrnul
+static inline char *gitstrchrnul(const char *s, int c)
+{
+	while (*s && *s != c)
+		s++;
+	return (char *)s;
+}
+#endif
+
+/*
+ * Wrappers:
+ */
+extern char *xstrdup(const char *str);
+extern void *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)
+{
+	return (size_t)len;
+}
+
+static inline int has_extension(const char *filename, const char *ext)
+{
+	size_t len = strlen(filename);
+	size_t extlen = strlen(ext);
+	return len > extlen && !memcmp(filename + len - extlen, ext, extlen);
+}
+
+/* Sane ctype - no locale, and works with signed chars */
+#undef isascii
+#undef isspace
+#undef isdigit
+#undef isalpha
+#undef isalnum
+#undef tolower
+#undef toupper
+extern unsigned char sane_ctype[256];
+#define GIT_SPACE 0x01
+#define GIT_DIGIT 0x02
+#define GIT_ALPHA 0x04
+#define GIT_GLOB_SPECIAL 0x08
+#define GIT_REGEX_SPECIAL 0x10
+#define sane_istest(x,mask) ((sane_ctype[(unsigned char)(x)] & (mask)) != 0)
+#define isascii(x) (((x) & ~0x7f) == 0)
+#define isspace(x) sane_istest(x,GIT_SPACE)
+#define isdigit(x) sane_istest(x,GIT_DIGIT)
+#define isalpha(x) sane_istest(x,GIT_ALPHA)
+#define isalnum(x) sane_istest(x,GIT_ALPHA | GIT_DIGIT)
+#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)
+
+static inline int sane_case(int x, int high)
+{
+	if (sane_istest(x, GIT_ALPHA))
+		x = (x & ~0x20) | 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
+# define FORCE_DIR_SET_GID 0
+#endif
+
+#ifdef NO_NSEC
+#undef USE_NSEC
+#define ST_CTIME_NSEC(st) 0
+#define ST_MTIME_NSEC(st) 0
+#else
+#ifdef USE_ST_TIMESPEC
+#define ST_CTIME_NSEC(st) ((unsigned int)((st).st_ctimespec.tv_nsec))
+#define ST_MTIME_NSEC(st) ((unsigned int)((st).st_mtimespec.tv_nsec))
+#else
+#define ST_CTIME_NSEC(st) ((unsigned int)((st).st_ctim.tv_nsec))
+#define ST_MTIME_NSEC(st) ((unsigned int)((st).st_mtim.tv_nsec))
+#endif
+#endif
+
+#endif
diff --git a/tools/perf/util/wrapper.c b/tools/perf/util/wrapper.c
new file mode 100644
index 000000000000..6350d65f6d9e
--- /dev/null
+++ b/tools/perf/util/wrapper.c
@@ -0,0 +1,206 @@
+/*
+ * Various trivial helper wrappers around standard functions
+ */
+#include "cache.h"
+
+/*
+ * There's no pack memory to release - but stay close to the Git
+ * version so wrap this away:
+ */
+static inline void release_pack_memory(size_t size, int flag)
+{
+}
+
+char *xstrdup(const char *str)
+{
+	char *ret = strdup(str);
+	if (!ret) {
+		release_pack_memory(strlen(str) + 1, -1);
+		ret = strdup(str);
+		if (!ret)
+			die("Out of memory, strdup failed");
+	}
+	return ret;
+}
+
+void *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);
+	if (!ret && !size)
+		ret = realloc(ptr, 1);
+	if (!ret) {
+		release_pack_memory(size, -1);
+		ret = realloc(ptr, size);
+		if (!ret && !size)
+			ret = realloc(ptr, 1);
+		if (!ret)
+			die("Out of memory, realloc failed");
+	}
+	return ret;
+}
+
+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;
+}